ISEC 88 INTERNATIONAL SOLVENT EXTRACTION CONFERENCE

Moscow, USSR July 18 24,1988

Conference Papers Volume IV

USSR Academy of Sciences International Committee f o r Solvent Extraction Chemistry and Technology

ISEC '88 INTERNATIONAL SOLVENT EXTRACTION CONFERENCE Mosccw,USSR July 11-24,1988

Conference Papers Volume IV

Moscow- Nauka -1988

ШЮ 5*2.61

жеишткшь soLrmr ЕХТЯАСТЮТ CORFERBICE (звнз'вв). CODferenoe Papers:

Volume I Plenary Lectures

1. Fundamentals 2. Extraotants

Volume II

Volume III

Volume IV

3. Interphase Phenomena, Mass Transfer and Kinetics

4­. Equipment 5. Modelling and Control

6. Liquid Membranes 7. Metal Extraction 8. Organic and Bioorganic Processes

9. Analytical Chemistry 10. Nuclear Processes 11. Hydrometallurgy

ra 5­oe­oo7ao»­4

© Vernadsky Institute of Geochemistry and Analytical Chemistry of the USSR Academy of Sciences , 198B

ANALYTICAL CHEMISTRY

NBC EXTRACTANTS AMD EXTRACTION SYSTEMS IB ANALYTICAL j CHEMISTRY I — — — B.Ya.Spivakov and O.M.PetruKhln, Vernadeky Institute of Geochemist­ry and Analytical Chemistry, Academy of Sciences of the USSR, Mendeleev Institute of Chemical Technology, Moscow, USSR Efficiency of liquid­liquid extraction as a method of separation

and pxeconcentration ­ selectivity and extraction percentage ­ is de­termined by the seleotion of the extractant and the extraction system as a whole.

In principle there can not be universal, i.e. multipurpose extroe­tants or systems. Therefore discussing the properties of extractants, it is reasonable to consider them taking into account what type of com­' pounds they extract. The extractable compounds can be devided into two, large groups: neutral species (coordinatively unsolvated compounds, metal chelates and coordinatively solvated or mixed compounds) and ion pairs (coordinatively unsolvated ion pairs, mineral and complex metal­containing acids) £\J. Besides, different compounds for some reason not included into the above groups may form a separate one.

Metal chelates have been well studied and widely used. Chelate­forming extractants containing a mobile proton are commonly used in analytical chemistry and technology.

The following notions were formulated Just due to the generaliza­tion of the experimental data obtained in the study of thie class of extractants. These were: the functional group (FG) as a group of hete­roatoms responsible for the analytical effect of the reaction, and electron­active group (BAG) or substitutuer.t affecting the PG proper­ties. It is worth emphasizing that up to now the notion of the extrac­tion reagent assumes the possibility to distinguish independent PG and EAG in the extxactent organic matrix. The Btudy of the extraction be­havior of metals in systems with chelate­forming agents has made it possible for the first time to formulate the theoretical dependencies of the extraction constants on the parameters of metal under extracti­on, extractant and solvent. This in its turn provoked the discussion of the extraction power (EP) of the extractante.

Among the offered BXtxactants we shall diBcuss the following ones. The reagents [(?hO)2ES]2im (X=0,S) are shown in / V to act as effec­tive chelate­forming extractants permitting the rare­earth and acti­niae elements or Ag +

, Hg and Au 3 + to be extracted depending oa the nature of the donor atoms.

Sodium diethyldithiocarbamate is one oJ the most widely used rea­gents for heavy metal preconcentration. A recently offered extractant called tris­dithiocarbamic acid is of interest. It allows the uranium to be preconcentrated in the analysis of sea water d].

PG of many analytical chelate­forming reagents are known to have л

analogues among technological extractante. There now exist some examples of the reverse motion of ideas and extractante. 4­Methyl­N­

Table 1. Donor atoms and functional groups of monodentate neutral extractantв

0 s В Р

­ 0 ­

$P=0

?As=0

>я­с=о

>Я­Р,<0

­s­

JC=S

3»P=S

o=sc

1 SE

>N­C=S

P

(ЪикгЗ)

а

"©я

Е Ш ­ ^ н

J h 4

Ph­P ­ 0 ­

$P=0

?As=0

>я­с=о

>Я­Р,<0

­s­

JC=S

3»P=S

o=sc

1 SE

>N­C=S

P

(ЪикгЗ)

а

"©я

Е Ш ­ ^ н

0

­ 0 ­

$P=0

?As=0

>я­с=о

>Я­Р,<0

­s­

JC=S

3»P=S

o=sc

1 SE

>N­C=S

P

(ЪикгЗ)

а

"©я

Е Ш ­ ^ н

l ­Pd­

Ан O H 3 ­ ^ C ­ O H 3

0 0

B­hydroxyquinolinylbenzene­sulphonamide, being a low molecular methyl analogue of the technological extractant LIX­84 has been offered and studied as an extraction­photometric reagent for copper /V» A number of other attractive ohelate­forming reagents was offered in recent years.

Another large class of the extracted neutral compounds consists of coordinatively solvated mixed compounds. These are electron do­nor­acceptor ones with a neutral complex of extracted metal as elect­ron­acceptor molecule) and on organic base (extractant) as electron­donor molecule. Therefore the theory of the extraction of this type of compounds is based to a great extent on that of the electron do­nor­acceptor complex formation. Extraction by use of neutral extrac­tants have rather long been developing mainly for monodentate oxygen­containing extractants, ethers and ketones above alKT&ble 1). It was in studying monodentate extractanta that the EP theory of the related extractants has been developed. This field of solvent extrac­tion has been developing by etudying the extractants with new hetero­atome as donor atoms and through PO complication.

The examples of neutral extraotants containing almost all hetero­atoms of interest are presently available. The PC themselves have been complicating alongside. We would like now to concentrate on the following possible conclusions. Plrst, fox complex FG containing se­veral heteroatoms there may be cases where various atoms act as donor ones. Second, one can hardly single out the two separate groups, i.e. PG and BAG. Thus, the possibility of application of the developed ap­proaches to predict the behavior of elements in extraction systems | with such extractants becomes hardly discussible. Finally, the avai­

5

lability of the extractants with complex different PG lead to the synthesis, stud; and use of polydentate reagents practically designed only from PG, The open chain extractants were followed by cyclic ones and then by the reagents containing both open and cyclic fragments (neutral and acidic). Above all these are organophoaphorua reagents containing phosphoryl (I) and carbamoyl groups (II), amidosulphides (III), aoldoketosulphides (IT), cyclic polyethers (V). Many of these extractants have been used to solve definite problems.

I CO

_/ R' j( (CH2)2 (CH2)2 C° J­J­0

N

C­0 И К 0 f« / \

I /

H

N II III IV Among macrocyclic reagents containing other heteroatome besides

oxygen we can mention dibenzo­15­crown­5 analogues containing various combinations of oxygen, nitrogen and sulfur as donor atoms /57. The study of this group of reagents allowed the methods of silver deter­mination to be developed. The seleotivity of these methods is higher than that of the methods based on the use of dialkyl sulphides.

Presently, those specialized in the organic synthesis can synthe­size reagents with random number of PG. Here the most acute problem arises concerning the development of the extractant Б? theory, allow­ing the selectivity relative to any element selected at random to be predicted. Up to now the reagents used in the extraction of any one type of compounds were diecussed. The so­called "different compounds" not included into the above mentioned groups due to whatever reasons, j should be considered. Among reagents for such compounds organotin ex­ i tractents H2SH(N0,)2 £^J , for Instance, are rather interesting. These ' reagents possess a unique extraction power in the relation to multi­charged phosphate, arsenate and other oxo­anions. These extractants are of interest from the theoretical point of view as well. Formal description of the extraction with such reagents is completely analo­gous to that applied to conventional anion­exchange extraction proces­ses. On the other hand, the extracted anion enters the inner coordina­tion ephere of the tin atom. And the extractable compounds are essen­tially the coordinatively solvated ones.

Another example concerns the possibility of the use of crown­ethers (I>) to extract metal halide anions in the form of complexes of ion pbirs (Kat+

)„ « m

~ where Kat+ denotes Hl+ or HL+ (M­alkali metal m n+m

Xshalid* or some other anion). In thie case the crown ethers seem to act as solvatlng agent, the alkali metal ­ crown ether complex pre­senting the cation part of the ion pair /П7­

The compounds discussed here together with the majority separated by liquid­liquid extraction are usually extracted in a two­phase sys­

tem of water ­ water­immiscible organic solvent. Yet the possibiliti­es of the extraction as a method of separation and preconcentration of substances in analytical chemistry and not only there, may be ex­tended due to the so­called non­traditional systems.

Systems with water­soluble organic solvents or without any organic solvents will be attributed to such ones. With these systems it is possible to extract various hydrated compounds using, e.g., water­so­luble organic reagentsi to extract substances in absence of any orga­nic solvent; to obtain water­soluble electroconductive extracts; to increase the extraction rate; to achieve high enrichment factors and to solve some other problems not always solvable through the develop­ment of new reagents.

The application of two­phase systems of water ­ water­soluble or­• ganic solvent (ffOS) is conditioned by the method of subsequent deter­mination of elements in the extract. Japanese chemists /Й/ used sys­tems with acetonitrile or propylene carbonate and chlorides or sulfa­tes to extract ion pairs, including anionic halide, thiocyanate and chelate complexes, in order to develop extraetlon­polarographic me­thods. The obtained extractants are electroconductive and may be di­rectly used in polarography. Such systems with acetone quarantee a high sensitivity of the flame atomic absorption analysis. WOs contain­ing considerable amounts of water or as they are sometimes refered to as "hydrophylic extractante" were used to extract water­soluble orga­nic and high molecular substances such as albumins, amino acids and other highly hydrated substanoee /97.

The WOS­based systems may also be used in so­called "homogenous ex­traction" /10/. In this technique an extractable compound is formed in a homogenous phase on the addition of WOS (for instance, propilencarbo­

/ nate) to the water solution of metals for faster equilibrium attain­ment. Then a salting­out agent is added or the preheated solution is cooled to cause phase separation. This technique permits to shorten the time for the equilibrium attainment from several hours to several minutes against the traditional extraction system. The extraction of Pe(III) with thenoyltrifluoroacetone is Just the case.

As all of us are specialized in liquid extraction, we work in such a field where the majority of problems is solved by means of organic

: solvents. One cannot but admit, however, that organic solvents have some unfavourable properties in terms of their practical application. Many solvents are flammable or explosive, toxic, volatile. The latter drawback may be partially eliminated by using fusible extractants.

Two­phase aqueous systems free of organic solvent are now known to exist. Two­phase aqueous polymer systems, applied in biochemistry :for the extraction of proteins, cells and other biological Bubstances /117, rank among then. Distributions occur between immiscible solu­tions of two polymers or between a polymer solution and an inorganic

*~ 7

salt solution, i.e. among phases providing similar conditions for hyd­ration. This makes it possible to extract substances which concentra­te in the aqueous phase when traditional organo­aqueous systems are applied. Besides the hydration properties, the polymeric phase posses­ses another essential feature. That is the value of its relative hyd­rophobiclty ­ the value of free energy of hypothetic ­CHg group trans­fer from the given system to n­octanol. Polymeric phases according to their hydrophobicity rank between water and organic solvents.

The study of the two­phase systems based on water­soluble polymers allowed them to be applied in extraction of inorganic compounds as well, water­soluble organic reagent of different classes being invol­ved ШИ. Such reagents can form complexes with metal cations which 1, are transfered from aqueous solution of an inorganic salt (sulfate, к carbonate, phosphate, etc.) to aqueous polytethylene glycol) (PEG) jj solution with high distribution coefficients. The number of polar and В dissociated groups in the reagent molaculee and coordinative unsatura­j| tion of complexes showed no marked negative influence on the extrac­tion of metal complexes.

PSG itself containing donor oxygen etonm may serve as an effective extractant for complexes of metals with inorganic Uganda, for in­stance, halide or thiocyanate ions. Thus, the distribution coeffici­ent of copper(II) extracted from sulfate­thiocyanate solution with 406 PEG is 10­fold higher than with tributyl phosphate and 104

­fold ', higher than with diisopropyl ether from the same aqueous solution. J It is of interest that anionic dicharged cupric complex, containing two water molecules is extracted by PEG. The extraction of such a complex in a traditional system would be hindered. I

Another type of two­phase aqueous systems was offered recently £Пл Phase separation occurring in ternary systems of water­pyrazolone de­rivative ­ organic acid is due to chemical interaction between two solid organic substances dissolved in water, for instance, antipyrine (Ant) and monochloroBcetic acid (MCAA). The heavy phase separated from the aqueous solution is the solution of Ant monochloroacetate with a high concentration of MCAA. The excess of both Ant and the acid in this phase provides conditions for the extraction of metal complexes.

Оле of interesting methods of extraction and enrichment of hydrat­ed metal complexes is baaed on the use of ternary systems with two aqueous phases £147• Such systems include water, ether, mineral acid and its salt. Hydrated and solvated mineral acid is the main compo­nent of the third phase. Due to a high percentage of water in this phase the equilibrium between two aqueous phases with nearly the sami composition is provided. This determines the same character of hydra­tation and the identity of forms of metal complexes in the mentioned &.. phaees. Methods of the extraction of halide complexes of platinum me­"

8

tale including tricharged anions into the third phase have been deve­loped. The extraction of [IrBrg] complex not extracted in traditior­al systems may just serve as an example.

Recently interesting studies devoted to three­phase systems with two organic phases were made. These investigations were carried out at Kiev and Perm Universities (USSR). Such systems allow the effecti­ve extraction of different metals to be carried out in a number of cases. Their use has made it possible to achieve extremely high en­richment factors due to a great difference between the volumes of wa­ter phase and the second organic phase.

The extraction in the above mentioned two­ and three­phase systems goes well together with subsequent determination techniques, thus ma­king them interesting both from the theoretical and practical view­point.

Finally, it is desirable to answer the question discussed at ISEC­30 in Liege: what is better ­ to search for new extractants and extrac­tion systems or try to obtain more complete and reliable information about the available systems. It seems that both are песевеагу. More­over, we believe that either of the directions will be in progress so far as the liquid­liquid extraction will help solving practical problems.

References 1. Zolotov Yu.A., Kua'min H.H., Petruknin O.M., Spivakov B.Ya.//

Anal.Chin.Acta. 1986. Vol. 180. F.I 37. 2. Havratil 0., Ectehard R.//6th. Inter.Conf.Adv.and Appl.Anal.Chem.

Praot.­Trace Anal. Racova Dolina,1986. S.l.,S.a. P.48. 3. Tabushi J. .Kolcube Y., Yoshaawa A.//J.Am.Chem.Soc.1985.Vol.107.P.4585. 4. Caetresana J.M. at al.//Quimioa Analytlca, 1987. Vol.5.P.409.

(b. Zolotov Yu.A., Poddubnykfc L.P., Daltrlenko S.G. at al.//Zh.Anal. Khlmii. 1986. Vol.41. P.1046.

|б. Snkinev V.M., Spivakov B.Ya., Vorob'eva O.A., Zolotov Yu.A.// Anal.Chlm.Acta. 1985. Vol.167. P. 145.

7. КовЫиа H., Oniehi H.//Analyst. 1986.Vol.111(11). P.1261. 8. Hagaoaa Y., Yamada Т., Maegawa Y., Puwa T.//Hikrochim. Acta.

198*. Vol.2. P.107. 9. Kofanov V.I., Kevinnaya L.V.//Papers Respubl.Conf.po Analit.Khlmll.

Kiev: «ацкота Duaka, 1985. P­58. •0, Tine Ch., William» B.T.//ISBCBO. Abstracts.Li'eg», 1980. P.234. H­ Alb«i­t«»on P. A. Partition of Cell Partielee and Maoroaoleculee.

Stookholmt Alaquist and Wikaell, 1971. ?1 Spivakov B.Ya., Shkinev V.M., Zrarova T.I. et al.//I3BC'86.

Preprint!. Munehen, 1986. Vol.2. P.547. ­ rB Petrov B.I., Afendikova O.Yu.//Zh.Prikl.KMuii. 1985.Vol.58.P.2194.

Щ Kbalonin A.3., •ovakovskaya B.G., и.и»«». L.B.//YII Veeeoyuin. Conf. po Butraktaii. Mbeoowt Bauka, 1984. P.47.

9

APPLICATION OF A DOUBLE LIQUID MEHBRAHB FOR ACTINIDE DETERMINATION INI 9­2

BIOLOGICAL SAMPLES *

R.Chiar iz ia (ENEA­Casaccia,Rome,Italy),P.R.Danesi (IAEA­Seibersdorf,Vienna,

Austr ia) ,E.P.Horwitz and P.K.Tse, Argonne National Laboratory,Argonne,I I ,

USA

In previous work L*»2j a supported l i q u i d membrane (SLM 1) cons i s t i ng of a

0.25 M CMPO and 0.75 M TBP so lu t i on in decal in (CHPO s tands for octy l (phenyl )

N,N­diisobutylcarbamoylmethylphosphine ox ide , the n e u t r a l bi func t iona l organo—

phosphoruB bx t r ac t an t developed a t Argonne National Laboratory for the TRUFX

procese Гз J ,and TBP s tands for t r ib i ' ty lphospha te ) has been used for removing

a c t i n i d e s from syn the t i c l i q u i d nuclear was tes ,concen t ra t ing them in a s u i t a ­

b l e s t r i p s o l u t i o n . The s e l e c t i v e t r a n s p o r t of a c t i m d e s through the d e s c r i ­

bed membrane takes place because the c a r r i e r CMPO r e a c t s with t r i ­ , t e t r a ­ and

hexavalent a c t i n i d e s in HNO s o l u t i o n s . The ex t r ac t i on and s t r i p p i n g e q u i l i ­

b r i a can be represented for a t r i v a l e n t ca t ion as [_3J :

M + 31KL + 35(HNO ) < = = = = > H(N0 ) E (HNO ) +(3n­m)HN0 (1) 3 Э п 3 J о 3 m J

where E represents CMPO and the bar indicates membrane species.

The driving force for the uphill transport of the actinides is generated

by the concentration difference of nitrate ions :n the feed and strip eolu­

t?.ons. As eq.(l) shows,HNO is also transported to some extent through the

membrane. Its accumulation into the strip solution eventually stops the acti­

nide transport. To obviate this problem m ref. |2J a second supported liquid

membrane ISLM 2) was used in series with the first one. It contained the car­

rier P ...i JM­T (a primary aliphatic amine) which removed only HNO, from

the ч ;

н > tion,neutralizing it in a third aqueous compartment containing

an NaOH. In this way the actinide transport could proceed to com­

pl<.tic... The вате flat­sheet double liquid membrane system,schematically

ehown in Fig.1,has been usee, in this work,where the transport of selected ac­

tinideB from HNO solutions or acidified urine samples has been studied with

the objective of demostrating the applicability of the SLM technique to the

separation of actinides from bioasaay samples.

Work performed under the auspices of the Office of Basi~ Energy Sciences,

Division of Chemical Sciences,U.S.Department of Energy.undtr contract number

W­31­109­RNG­38.

(0

FEED , 4V4SLM0 ,ч

MJ.25NbCMP0,

ч0.75М><ГВРл

<1есаИп\ \ч

\ ­

\ ч \ \ , ч Л Лч ч

> HNO­

polypropylene;

. s u p p o r t s

STRIP P^IMENE

4

",^ 4 \4M­T^V Леса I in , ч \ ; > 4

4 * , ч \ \ 4 ^

NaOH 2.5 M

An's

OTHER — 7 Ь ол

У >ч

' \; ;

Fig.l. Schematic description of the double SLM system used to remove actinides from dilute solutions such as nuclear waste solutions or bioassay samples

In a first sen ее of experiments the transport of HNO through the double SLM system was investigated in order to measure the influence of the HNO concentre tion in the feed and of the Primene JM­T concentration in the SLM 2 membrane on the capability of the system to maintain good stripping conditions [or

long timee ( [НПО 1 <0.05 И for about 24 hours). It was found in this way L 3

J

strip that,by using water as atrip solution,the HN03 concentration in the feed solu­tion had not to exceed 2 M.and that of the Primene JM­T in SLM 2 had to be in the range 0.75­1 M. For higher concentrations of h7K>3 in the feed solution, and for Primene JM­T concentrations above or below the indicated range,the re­moval of HNO from the strip solution was not efficient enough to allow a

3 fast and quantitative recovery of actinides in the intermediate compartment {strip solution).

In another series of experiments different strip solutions were tested. Good results were obtained with the following stno solutions: all M formic acid • 0.05 M hydroxylammonium formate (HAF), Ы the вате as before * 0.05 M oxalic acid, c) 2 M ammonium chloride t 0.05 И oxalic acid (an ammonium chlo­ride oxalate Bolution IB often used for actinides electrodepoeltion [A] ).

The results obtained for amencium transport through the double SLM system with the above mentioned strip solutions are reported in Fig.2. In all cases Am(III) was effectively removed from the feed with 98­99 % removal reached after about Bix hours. The permeability coefficient of A>(III),calculated

­3 ­1 from the straight lines of Fig.2,was in the range (1.0­1.3)xl0 cmxs , in good agreement with literature values [_lj .

The transport of uranyl ions through the same membrane system was investi­gated in similar experiments. Also in this case a 99 % metal separation from

11

N«4C ,

­H

2C

20

4

HCOOH­HAF

HcooH­HftF­H^Q,

the feed solut ion was achie­

ved in s i x hours. So»e of

the i n i t i a l V(Yl) (5­20SJ,

however,was round at the end

of the experiment in the

NaOH coapartaent. rtiis fact

i s probably due to the for­

mation in the s tr ip solut ion

of алюшс formate and oxa­

late complexes capable of

reacting with the l iqu id a­

nion exchanger Priaene JS­T.

Fig­3 shows the reHulta 0 2 4 6 8 10 t,hr

F i g . 2 . Aa(III) feed a c t i v i t y ve t (hr) for various of Aa(III) transport expe­

s t r i p so lut ions . Feed=2 II HHO ; SLM 2=1 II Priaene riments performed by using " 2

Jtt­T; aeabranea area=i .? l cm i volume of feed and ac id i f ied urine as feed s c

NaOH s o l u t i o n s ^ ca ; volume of s t r i p solut ion 3

lu t ion . This solut ion was

10 caT; s t i r r i n g speed of f eed , s t r ip and ЙаОН s o ­ prepared by adding concen­

lut ion = 200 rvm trated n i t r i c acid to raw

< V

H. * . • „ UNFILTERED ' о

cpnj^L m

w» a •

«8

• FILTERED

trat ion of 2 M. The a c i d i ­

f ied urine so lut ion was then

digested at room temperature

for some hours before use

in a double l iquid aeabrane

experiment. I t was spiked 241

with AH j u s t before the exp*­iaent was s tar ted .

6 8 10 t,hr r ig+3. Aa(I£I) feed a c t i v i t y ve t<hr) for i m f i l ­

tered and f i l t e r e d ac id i f i ed (2 Я НПО J urine as

feed so lu t ion . Strip so lut ion = 1 Ж НС00Н + 0.05

The data obtained vith

unf i l tered urine show a

awch slower and part ia l t r a ­

nsport, indicat ing a Beabra­

ne "fouling" by the organic

• a t e r i a l s t i l l present i n

the feed so lu t ion . Much bet ­

t e r r e s u l t s «ere obtained by p r e f i l t e r i n g acid digested urine on a s a a l l со ­

i u

* n f i l l e d with poiraue beads of polyester resinffi] .In t h i s case ,as shown in

in Fig.2

12

Fig .3 ,a 96 * recovery of the i n i t i a l Am(UI) ­activity was achieved in about

7 hours with a rat io of membrane area to feed volune of 0.43 a» and a per­

­4 ­1

meability value of Aa(III) equal to 9.6x10 cmxa . In the same experiment

a pract ica l ly quantitative recovery of A»(III) (>99.9 X) in the s t r i p so lu ­

tion was reached after 24 hours (not shown in the f igure ) .

A much fas ter separation process i s expected by using hollow f iber modu­

l e s with a much larger area/volume r a t i o .

More experimental work i s needed to fu l ly assess the appl icabi l i ty of sup­

ported l iquid membrane systems to the separation of act in ides from biological

sanples. The areas requiring more invest igat ion are:

U the use of hollow f ibers with a high membrane area to feed volume ra t io

and a high feed volume to вtrip volume r a t i o .

2) the research on more powerful stripping agents capable of stripping the

act inide elements from the QtPO containing шеаЬгапе even in the presence

of r e l a t i v e l y high HNO concentrations, in t h i s case the use of a second

SLM could be avoided and the whole experimental system would be simpler.

Also,the risk of loosing a fract ion of the actinidee as anionic complexes

in the NaOH compartment would be eliminated.

References

1. P.R,Danesi,R.Chiarizia.P.G.Rickert, K.P.Horwitz / / S o l v e n t Extr.Ion Exch.

J3(1 ,2) .1U (1985).

2. R.Chiarizia and P.R.Danesi / /Separation Science and Technol. 22(2 ,3) .641

(1987).

3. E.P.Horwitz and W.w.Schulz//Proc.Intern.Solv.Extr.Conf. ISEC 'вб.Мипспвр,

Sept .11­16,1966.Vol .I ,p .81 and references there in ,

4 . J.C.Veselsky //Mikrochimica Acta, 1(1­2) ,79 (1978).

5. E.P.Horwitz and M.L.Dietz. Method for the Concentration and Separation

of Actinides from Biological md Environmental Samples . AHL Patent Саве

S­65,064 , Oct.21,1986,

13

THE MEMBRAUE ECTRACTIOIT ­ A STTITABLE METHOD POR THE |

V.ivUkulaJ, P.Rajec, R.Kopunec, A,Svec, F.Macasek; Department of Nuclear Che­mistry Comenius University, Bratislava, CZECHOSLOVAKIA

In recent yeirs membranes have been widely used in studies of Ion transport. The development of these systems can be triced back to experiments by Nerst and RIesenfeld (1902) [1]. Not* we have several methods to utilize liquid membra­nes for separation processes. An emulsion t/pe consisting of water­oil­water emulsion [2]; a supported type consisting of an oily phase held by capillary forces within the pores of «ieroporous membranes (3); a liquid Ml» pertraction £4) where the feed and stripping solution flow down along vertical solid porous supports and between the supports there is a liquid membrane.

A theory of the process has been developed for various models of simple end activated permeation processes and different engineering condition Г5­73 ­ This paper describes liquid emulsion extraction for preconcentration of metals. Me have treated membrane extraction as ar, analogy of solvent extraction.

Considering в chemical equilibrium in three phases system­ «re can obtain a formula similar to solvent extraction In batch condition [B].

<p­i>Dr V Vv

i > ' i i ' V u i ft

MX = ­ ­ • i П ) v ­ feed volume; v ­ stripping volume;

p = 1+ * (2) D ­ distribution ratio for inner boundary; 1] II p ­ pertraction factor, i.e. ratio of the

amount of metal in emulsion to that in organic phase. Rk).should be considered as the maximum efficiency parameter. From the practical point of view it is difficult to reach the teoretical value besause of emulsion breaking, feed emulsification, sttippant permeation, mutual transport of matrix etc. Dun to this, chemical potential difference between the feed and strippant phase diminishes and the permeation efficiency decreases too. A large pertraction factor p enables to мяв carriers with low solubility in membrane, with low 0. or too expensive but selective reagent in small amounts. For large volume samples we have performed emulsion membrane extraction in dynamic conditions [9]. Me contacted the feed volume (v.) with a fixed amount of emulsion continuously in an apparatus with working volume smaller than the feed (V <V ). It was no solute concentration at the begining of praconcentratian and it has been assumed that the system is in equilibrium in the working pert of the column. Changes of the emulsion volume were neglected. Then the yield of the differential preconcentration ensuing from a general equation is :

M . ч „ , ., „ Hi „ „ ч ,X\ * г,с ­ concentration of solute in dv (c ­c) ­ dc< v*v =*­ +V 0 ) (3) F'

1 *• u И feeding or rafinate, reap.,

r­VH/V is the working phase ratio in the apparatus

14

(p­UO­e v /v •^"•••1 /_•! Application of aeabrane extraction

I I as a preconcentration aethod Is connected with high value or the pertractlon factor because :

High value of the preconcentration - ^ . (p­i)vTt ­* ­ ( P ­ I > V , I <

5 ) f

"C t

" " M C h C

'n ^ C q U > I t 0 *'"»

C

I l l C

I l " in lielting case, Is preference

or of aeabrane extraction even at low С

И < p

"1 J O

ir

i V

I ,,. 0 which is in the case of lo* C

I0 p 1

Г

1+ II extractant concentration reached

in both batch and dynaaic conditions. Separation factor of two aetals is fro» the «eabxane extraction point of view :

Kinetic factor far three phases ( c

H/ c

I3 (p­1)

(p­1) systeas with the saae composition and aonotype netal series can be used to reach S * 1.

APPLICATIONS A nodified turbine type iapeller with stabilisation of rotation has been used

for preparation of emulsions [10]. A pertraction column apparatus for Mow through Dreconcentration is constructed [11].

URA.Jli'4 KELEX 100 and OEHPA were used for effective separation of uranium fro»

slightly acidic solution [12,13,14]. As • most perspective carrier in n­alkane membrane with inner (stripping) solution 1­Z.5H H PO or н so DEHPA was chosen. This system was used for preconcentration of uranium from 20­501 in flow through column apparatus. The preconcentration factof was higher than 350 and recovery was of 9B%.

CEPtUM A pxeconcentration factor of 100 was reached with nenbrane containing TOPO

carrier and in high nitrate concentration. Citrate and EDTA as inner solution were *mol/l (fig.1).

Kambrane extraction of carlua. Haabrane:0.01N TOPO in n­dodecan«,4X jPAN 60,outer solution:!.10"*Н Ce 3

*0,2,3) t2H NeNQ3<1,2,3,4,3,6),1.10"5

M Св3

*(4),1.10~6

M Ca3

* (3),

U 1

Ca only (6), inner iolution:0.01M E0TA.1M NaNO (1>, 1M Hi ,Cit(2,3,4,5,6),

15

COBALT, MANGANESE Batch pertraetlon of Co end Mn with aroaatic beta­hydroxyoxiae (ulX 6*N, LIX 65,

ASF) industrial extractants «as studied. Kinetic and hydrodynaaic characteristics of emulsion with «inlaw* of retention of Co in the aeabrane were optiaalised. For Co it can be reached a preconcentratiDn factor of 100­170 in batch end 300 in flow throuhg pertraction with high recovery using SPAN 80 detergent. we did

* ^j*—^

Na •

a* •

<w \

w V • * ­ * ­ ^ f e

с, K»> —\ ­r­ UN *m Ha

li 1

/

0*

OU06

0*4 ^ U _ li

1 /

0,02 i **

S" t * (SKMI) 0,2

i /W ч/М Fig.2 Meabrane extraction of cobalt and Manganese. Menbrane: IX LIX 64N in n­dadekane,

­5 2+ ­5 2 + 3% ECA «360, outer solution: borate buffer pH 7.9, 7.70 H Co , 7,10 H Mn , 0.1M NaNO,, inner solution: 0.1M H_S0., r = 0.1, r = 1­0

** 2 A I II Fig.3 Flow membrane extraction of cobalt and aanganese. Membrane: IX L I X 64N in n­dodecane, 3X ECA 4J60, outer solution: borate buffer pH 7.9, 1.10~5

M Co *, 1.10* И Hn *,D.1H NaND , inner solution: 0.1M HC1, r = 0.1, r = 1.0, ve= 6 da*h~ ,eaulsian 0.2 da*, recovery (inner solution): 98X Co, 28% Mn

not succeeded in separation of cobalt and Manganese because SPAN 80 served as carrier for Co U 8 X ) and for Hn (30X) a: veil [15,16]. Polyaaine ECA 4360 causes slaver transport and enables us to enhance the separation factor of Co/Hn (fig.2}. Polyaalne influence has been shown In the flow through pertraction too. The recovery of Hn drops fro* 9*« to 27* but that of Co rests on the value of 95% (fig.3).

STRONTIUM, CALCIUM Eayision aeabrane extraction of ST with crown end picric acid as carriers

shows that preconcentratian of Sr and its separation Ггоа Се can be compared with the ratio of U

» C

/ D

I C

f o r toluene aeabrane and 18C6 crown at appropriate concen­tration. Transport of Sr is cauaed by picric anion gradlend in solution I and II which are controlled by HC1 in the Inner solution (0.3M) [17]. The Influence cf Ca on the recovery and the transport rate of «trontjy» j« shown 1л rig».4 and 5. Ca 1* practically not pertractad and this can ba used for separation of Sr fro* highly concentrated (­0.2И) Ca solution with high recovery. Presence of Ca in Iso­lated inner solution is caused by secondary eaulgetlon during pettractlon. in this way the separation factor Is lower but it can be laproved by changing of detergent

16

Recovery of strontium in aeabrane extraction. Membrane: toluene, AX SPAN 80, ­5 2+ ­3 ­5

outer solution: 1 4 0 И Sr , xM CaCl , 4.5 ­10 H picric acid, 5­10 H 18C6, 0.1.

*H" 1.0, pertraction time 10 mln inner solution: 0.5H HC1, FlQ.5 Me«brane extraction of strontiuau Conditions see fig.4, concentration of CaCl in outer solution 0.07 (1), 0.15 (2), 0.3 (3), 0.6 ool.dn" 3 t> .'

(polyaalne Instead of SPAN 80), by using the appropriate «enbrane solution (chlo­

rinated hydrocarbon instaed of toluene) and by changing of the ratio V__/V . Toluene eaulslon stabilised by SPAN 80 is generaly not suitable for a long term pertraction because of loss of its hydrodynanic properties ­ it turns to paste. Application of polyaaine ECA 4360 is good fron the point of secondary eeulgatian.

2* It decreases the К value fro* 1.6­1.8 to 1.0­;.1 (0.1­0.2H CJ In outer solution) and keeps the eaulsion in liquid for*. The tenside effect an the pertraction of Sr is shown In fig.6. The pertraction of Sr and Ca with toluene (fig.7) and tri­

» , 30 УГН

Heabrane extraction of strontlua. Heabrane: toluene, 4% ECA 4360, outer solution: 1­10 И SrCl,, «.5­ 10 M picric «cid, 5 4 0 M 1BC6, 0.15M Ca(N0 ) (2), the S I M •< (2) but «Knout Ca

2

* (1), inner sol. 0.15H HC1, r ^ 0.1, r . 1.0 flg.7 Meafcrane attraction or strcntlu» and calciua. Meabrane: toluene, 4Я ECA 4760,

5­10 н 1«C« ( , ), 5 4 0 и 1BC6 < , ), Inner solution: 0.5H HC1,

17

• ' V M " • / [ « * ] Hembrane extraction of strontium and calcium. Membrane: l,2,4­trichlor­

benzene, 4% ECA 436D. conditions see fig.7 Flg.9 Membrane extraction of strnntlum. Heabrane: 1,2,4­trichlorbenzene, 4* ECA «360 (1), 4« SPAN SO (2), outer solution: 1 10~ 5

H SrCl ,

_­'. _­*. 5 to" к picric acid, i to" и 18C6, inner solution: о.5и нзд.г^о.ог, г = 1.0

chlorbenzene (fig.В) membrane is shown. Trichlorbenzene emulsion stabilized by polyamlne Is suitable for 50rold preconcentration of strontium (fig.9).

PERTECHNETATE The llculd aeabrane extraction tilth Aliquat J36 as a carrier has been studied (IB]

The recovery of pertechnate In the inner solution In experiment with liquid mem­brane «as estimated from the results of solvent extraction according to formula (l). A goad agreement between the calculated and experimental results has confirmed the three phase distribution model.

REFERENCES 1. ».Nernst.E.H.Rlesenfeld //Ann.Physic. £, 600 (1902). 2. N.N.L! //AICHE Journal. Г7 (1971).459. 3 <E.L.Cusler //Д1СНЕ Journal.Ц (1971). 130O. 4. LBoyadzhiev /'proceed.1SEC 83,Denver,Co USA (1983). 5. L.Boyadzhiev,T.Sapundzhlsv,E.eezenskek //Seper.Sci. J£ (1978). 541. 6. H.S.Ho,r.a.H«ttOn,CN.i.ightroot,N.N.<_l //AICHE Journal..28. ( 1982). «62. 7. M.Teramoto,T.sakai,K.Yanagava,Y.Mlyake//separ sci.Technol. 1_8 (1983). 985. 8. F.Nacase*,P.RaJec,R.i<opunec,V.Hlkulaj//Solv.Extr.Ion Exch._2_ (1984). 227. 9. V.MlkulaJ,P.Ra.lec,«.Svec//rChem.llsty.8p_ (1906). 545. K).A.Svec,P.RaJec//C

h

«"­l­lsty.Z7 (1983). 206. 11. V.HlkulaJ,P.Ra.Jec,A.Svec//Chem.Llsty.ep_ (19851.545. ' 12. F.Hacasek,P.RaJec,V.Rehacek t Vu Nhoc Ann, T .Popovnakova / / J .Rad loana l .Nuc l .

L e t t . Л (1985) . 529.

13. F.Nacasek.P.RaJec.Vu Nhoc 4 b n , V.Rehacek / / P r o c . ISCC 86, Nunchen J9B6.

14. F.Nacasek.P.RaJJc.Vu Nhoc Atm//Proc CHISA, Praha, 1987.

15. V.Hlku laJ ,N .H .Cuong/ / ) .R«d loana l N u c l . Cham,Ar t ic les . 101_ (19861.59 .

16. V.HlkulaJ ,P.RaJac,A.Svec,N.H.Cuong,V.N.Ahry/J .Radioanal Nucl.Chem.Art . 101(1986).

17. v . M l k u l a J , J . H l a t k y , L . v e s a k o v a / / J ­ R 4 d l a n a l . N u c l . C h e m . A r t . 101 ( 1 9 8 6 ) . 5 1 . \

18. P.RaJec,F .Hacasek,O.Belan/ /J .Radloanal .Nucl .cham.Art . 101_ ( 1 9 8 6 ) . 7 1 . r.

18

AHPHIMETRIC DETERMINATION OF IONIC AND SONIONIC SURFACTANTS

N. Buschaann, Anorganisch­Chenisches Institut der Universitat, Wilhelm­Klenm­StraBe 8, D­WOU Miinster, FRG

1. Introduction. This paper proposes a new titration technique for the determina­tion of surfactants. The "Amphimetry" is a distribution titration of ion pairs con­tinously carried out in emulsions. Dichloro^thane ("DCE") is used as organic solvent for all titrations» For indication it is necessary to add an ion pair forming redox indicator to the solution. The redox potential of the free indicator in the aqueous phase is monitored by a platinum electrode and shuws a significant change at the endpoint. Reference electrode is a conventional Ag/AgCl/LiCl(EtOH)­electrode or a special all­solid­state reference electrode.

The titration procedure differs for anionic, catirnic and nonionic surfactants. The ionic surfactants are titrated directly by a bulky counter ion. The indicator must be of the same charge but less extractable than the analyte. The nonionics are determined by displacement titration of the associate formed by K +

, nonionic and an anionic surfactant. The cationic indicator which must be better extractable than the associate of potassium and the ionionic.

2. Titration of Anionic Surfactants. [i"j Anionic surfactants are titrated in a water/DCE­emulsion with a standardized solution of the cationic surfactant Hyamine 1622 at a pH of 8. MnO^~ or HCrO^~ can be used as indicators. To avoid an indicator­error rh*> ions are added in the form of their ion associates with the titrant. After addition to the test solution rhe titrant changes in a first step to the aqueous phase and an equivalent amount of anionic is extracted, cf. figure 1. During titration of the well­stirred solution the ion associate of anionic surfac­tant and cationic titrant is formed and extracted into DCE. At the endpoint also the indicator is extracted. The indication curve (figure 2) is made up by the redox potential of the free indicator in the aqueous phase and an adsorption potential of the cationic titrant.

The calibration graphs show linearity from 0.5 umol to 400 umol per АО ml emul­sion. The standard deviation is s=0.9% at 10 umol (n=7). The following concentra­

1. Addition of Indicator: £Г ;Т*1о • * i ­ { Д " : Г ) в * Ь

2.Titration:

*w ' T w ' i**i T

*)«

Э. Endpoint: —. <r : Tl 0

l i b . l . T i t ra t i on procedure Eor anionics, I=Indicator , T=Titrant, A=Anionic. o=organic, w=aqueous phase

, mi BMPAC t r

Fig. 2. T i t ra t ion of 10 umol of bis­(2­ethylhexyl)­sul fosuccinute (=•Aerosol 0T ) Indicator: Permanganate

19

tions (mg/1) of salts do not interfere with the determination [1,2]: KaNO^ (150), NaH 2P0 6 (1500), NH^Cl (350), K 2S0 4 (700). NaCl (350), MgCl 2 (350). CaCl 2 (350), FeS0 u (70), nonionics (50­100). Humic acid shows a strong interference even at very low concentrations.

3. Titration of Cationic Surfactants. [3] Figure 3 shows the titration procedure for cationic surfactants, which is ­ besides the opposite charges ­ analogous to that for the anionics. The cationic complexes of F e 2 + with 1,10­Phenanthroline, 2,2'­Bipyridyl or 2,4,6­Tri­(2­pyridyl}­s­triazine can be used as indicators. A typical titration curve is shown in figure 4. The linear range of the calibration graph and the standard deviation are the same as for anionic surfactants. Even large amounts of salts do not interfere with the determination (concentrations in mgjl): NaH 2P0 4 (2000), K 2S0^ (2000), CaCl 2 (2000), MgCl 2 (1200), FeCl 3 (500).

1 Addition of Indicator: 0*.T­)0.A*w .f**;T­] 0. ll

•(AM"},,

3.Endpoint: 1*w * Tw 0*; T"] 0 mlOHS Fig. 3. Titration procedure Fig. 4. Titration of 10 umol of for cationics. I­Indicator, Hyamine 1622 T=Titrant, A­Caticnic, Indicator: Fe­phenanthroline o*organic, v>equeous phase

Even if the cationic surfactants are in a lipophilic matrix they can be deter­mined amphinetrically without preceding isolation. As an example for those problems the titration of a diesel oil additive is examined. ;

4, Titration of Honionic Surfactants, [4J Nonionic surfactants ("NIOs") con­ 1 taining ethylene oxide groups can trap cations like K + or Ba**" and behave as quasi­ ' cationic. It is possible to extract the associate of K +

, N10 and an anionic surfac­tant into an organic solvent and determine the N10 in the organic phase by means of an amphinetrical displacement titration. Figure 5 shows the procedure, figure 6 a typical titration curve. The indicator must be cationic and better extractable than the associate containing the N10 so that it is displaced later than the former.

The calibration functions are linear in the range from 0.5 ag to 20 ag НТО in the organic phase after pre­extraction. The standard deviation is s­l.lX (n»6) for a content of 5 ag Triton X 305. Concentrations up to 1500 mg/1 of the following salts do not interfere with the determinationt NaH 2F0 4, CaCl 2, ffcCl2, FeCl 3 or NaN0 3. Humic acid shows a strong interference.

The slope of the calibration function depends on the average number of ethylene oxide groups in the N10, cf. figure 7. This indicates that one potassium cation is trapped by an average number of 20 ethylene oxide groups. Two ways for the quanti­fication are possible: individual calibration functions for each nonionc (with respect to the individual slopes) or a general calibration function for one standard i substance.

20

1. Pre­ Extraction of Nonionics; K* . N 1 0 * A" . {K N10* . * ­ } 0

Sto«rm« qraanic pha««

2. Displacement Titration : (K­NIO*; A " ) 0 . T * • {T*. A * ) 0 . K*. N10

3. Endpoin! I Displacement o* Indicator): { Г ; А ' ) о * Г ' . f l * , A ­ ] 0 . 1^

F J R . 5 . T i t r a t i o n procedure for nonion ics . 1=Indica tor , T=Ti t ran t , A=Anionic. o=organic, w=aqueous phase

F i g . 6. T i t r a t i o n of 5 nig of BftU 56 (9 EO­groups) I n d i c a t o r : Fe­phenanthrol ine

150 Q

/

100

«/

SO

A ш ­ 0.499 • П ­

50 100 150 200 2S0 300

OTfilOn

H O ­ [ E . ­ O ] ­ < Q > ­ I ,

no­{Et­o] ­ n

HO­[EI­O] _ f CH­CH ­o l ­ [ E I ­ ol ­H

"°­[«­°H­"

Fig. 7. Plot of the slopes of the individual calibration functions for several different nonionics vs. average number of ethylene oxide groups

Iе

. Conclusions. The electrocheaically indicated aaphiaetry is a siaple, fast» in­expensive, sensitive and precise aethode for the determination of ionic and nonionic surfactants. The detection liait is worse than that of the extraction photoaetric aethods so that the surfactants Must be enriched froa water or waste water samples before determination. As the titration is carried out continously the aaphiaetry car. be run on automated titration systeas. This is an advantage over the extraction photoaetric Methods.

References 1. Buvchaaon N.vIfcland F,/JFresenius Z, Anal,Chea,f 1965) 322.583­566.

2. Katz D. Dlploaa thesis (1966) HUnster. Tenside Deterg ­ to be published

3. Buschaann ДОг«*сп1ив г.АпеЬСп«в к(19в7) 326.123­126. 4. Buschaann N,/#reaenlus Z.Anal.Chea., in press.

21

SOME ASPECTS OP THE THEORY OP METAL­HAblDE COMPLEX EXTRACTIOH WITH BASIC DYES P.P.Rleh, I.S.Balog, Ya.R.Bazel, V.V.Bagreev, Uagorod State University, Uzgorod, USSR, Vernadsky Institute of Geochemistry and Analytical Chemistry, Academy of Sciences of the USSR, Moscow, USSR

Extraction of metal­hallde complexes with basic dyes is widely used in extraction­photometric and extraction­fluorometrie methods. These methods display high selectivity and low detection limit. Howe­ver, the theory of extraction of ion associates (IA), formed by metal­halide complexes with basic dyes is not sufficiently developed.

We have found some characteristics of dyes1 extraction ability as a function of their basicity, size, charge, substituent nature and stru­cture of the molecule as a whole.

In order to figure out the influence of dye nature on its' extrac­tion properties we have carried out systematical study of extraction of halogen complexes of Cd, Hg, Sa, In, Tl, Pb and Sb with different triphenyl­ and triaryImethane, rhodamine, cysnine, styrene, thiazane and some other dyes.

It is found that the size of the dye molecule does not affect the­ir extraction properties. Thus, refractometric radii of brilliant green.BG (3.64 X), and crystalviolet, CV (3.66 I) and also malakhite green.MG (3.46 i) and fuchsin.FH (3.36 1) does not differ significan­tly, as the extraction ability of BG is much more higher than of CV­and that of HG is higher than of FN: ion associates formed by metal­hallde complexes and FN does not extracted at all with sol­vents of low polarity.

The general basicity of the dyes also does not determine their ex­traction properties. Thus, total basicity increases in the raw MG­BG­WF­uV, but extraction of Ga, In and Sb decreases in the sequence BG­MG­CV­MV.

For the dyes of the same type and class in most cases there is a correlation between total positive charge on association stage and the extraction efficiency. The charge magnitude ou the boundary gro­ups carrying the maximal positive charge increases in a sequence KV (3.80) ­ HO (3.92) ­W(4.40) ­ BG (4.68) and in the same sequence increases their extraction ability for Sb, Cd and In.

The behaviour of the rhodamine dyes was found to be the similar. The hydrophilloity­lipophilicity balance of the dyes influence

extraction properties markedly. Dyes having hydrophilic groups (­MHg>>liH,­H»H­, ОН", СО0Н", etc) strongly interacting with water molecules could faintly extract metal­ halide complexes in solvents ot low polarity.

9­5

22

A very faint extraction power of fuchsin compared to other tri­phenylmethane dyes (TPMD) with benzene as a solvent could be explaned by very high hydxophilicity of FN which molecule has three NHg­ gro­ups and strongly interacts with water molecules. Th? similar relati­onship has been found for triarylmethane and thiazene dyes. With the increase of hydrophilicity in the raws victoria blue 4P­victoria blue B­basic blue К and methylene blue­azur II­azur I­thionene extrac­tion efficiency falls down. The same influence has hydrophilic COOH­group for rhodamine dyes; their extraction power raises in the raw rhodamine C­rhodamine GG­butilrhodamine.

It is shown for astrafloxine derivatives R­CH»CH­CH=R aa an example that the substitution of methyl group, attached to ternary nitrogen, for oxyethyl one leads to the decrease of extraction of Hgl~ . HgBr, and HgCl^ from 99.98 and 7856 to 40,20 and 256 accordingly.

Host of halogen complexes (HgXt, InX.4, GaCl^, etc.) are poorly ex­tracted with solvents of low polarity but the extraction is increased to a great extent when using solvent mixtures containing electron­­donor solvents (ketones, ethers, etc.) because of H­bonds forming ( >MH...o; ; ­C00H...0; j ­ОН...ОС) and solvatation of the dyes.

We have also investigated the influence of stereochemical factors on the extraction efficiency of basic dyes. The extraction of halogen complexes of Hg, Od, In and Tl with hinoline dyes of different struc­ture has been studied. It is found that the dyes which are o­substitu­ted to nitrogen atom carrying positive charge and taking part in asso­ciation process with halogen complex hae extraction ability less than m­ and p­subatituted ones which haB no stereochemical interference.

The extraction efficiency of the dyes increases with the increase in electrophilicity of radicals in amino­groupe. Since, dyes having p­di­ethylarainophenyl group extract metal­halide complexes better that analogous ones with p­dinethylaminophenyl group. The similar results were obtained for triphenylaethaae, rhodamine, cyanine­CR.­OH" CH­R 2), hinoline­CR.­CHg­N­Rj) and azo­dyes(R1­»«N­R2).

An increase of total charge of the dyes decreases extraction effi­ciency sharply. Replacement of one charge dye cation by two charge one leads to draatic decrease in the distribution coefficient of ion asso­ciate. Two charge reagents­methyl green (Het.G) and iodic green (IG) and their complexes with 3b, Hg, Tl are weakly extracted with benzene and itB derivativee| for quantitative extraction the use of more polar solvents such as chloroform, dichloroethane and mixture of benzene and nitrobenzene is necessary.

• Two charge cations Met.G and IG does not extract "coordlnatlonally saturated" metal­halide complexes as those of cadmium and lead.

23

There are two characteristic cases for the extraction process with two charge cyanine dyes:

a) dye molecule has nitrogen atoms carrying positive charges and divided by common conjunction chain )NwC­(CH) ­Ik ;

b) these atoms are isolated by hydrocarbon group *H­CH2­(CH2)n­If?. In the first case extraction of metal­halide complexes is not observed practically, in the second case extraction proceeds as effictively aa with the analogous one charge cyanine dyes.

The general rule is shown for the extraction of metal­halide comple­

xes with cyanine dyes of К1­СН»(0Нп)­0Нж112 structure. Extraction of InClT, HgBrl, GaCl7,PeClV, CuClg increases with the enlargement of hyd­rocarbon chain.

The extraction efficiency of dyes with halogen complexes of metals which cations could be treated as "soft" acids (according tc Pyrson) raises with the increase in the acidocomplex stability from chloride to iodide. Thus for IA halogen complexes of Tl(III) with 2­[l/5 diaet­hylaminothienyl­2/­vinyl­2]­1,3,3­trimethyl­3N­indol (DTVTI) the ex­tent of extraction falls in the raw Til" ­ TlBr^ ­ TlClT (see Table).

Sxtraction of 'jalogen complexes of Tl(III) with DTVTI

Complex L S KTIX" •''­max 4 Hfpj^ %

R*T1I^ H+IlBr^ I^TICI^

31.8 23.9 18.3

580 575 570

9 9 . 8 79 .1 62.5

A batochrome shift of absorbance maxima of IA extracts appears to be raised from different polarization of dye chromophore system in the field of Tl(III) halogenide complexes.

Some metals (Ga, In, Pe(III)) being "hard" Pyrson's acids was found to be easily extracted as IA with basic dyes from salting­out solutions such as alkali and alkali earth chlorides and HH.C1 or A1C1,. Homally GaCl" , InCl^ and feOl^ oonplexes with basic dyes couldn't be extracted by solrents of low polarity from the solutions of low «old oontant. If other eonditioaa being ваша tin Baltlng­ont «•«*• oaald k« axraneed In to* aeouanoeai

bi+ > Ha + > K + > Rb + > SHj > Cs*

and M g2 +

> C a2 +

> !

24

The distribution coefficients of metals are in linear correlation with salting­out cation radii. Such a dependence is shown on И*. for the extraction system Fe(III)­MeCl(MeCl2)­astrazone purple 2G­toluene.

The steady increase in the extraction of IA is observed depending on the charge of salting­out cation in the raw Ma < Mg* < A1J which is in accordance with the increase of cation hydratation extend. LiCl displays the highest salting­out effect compared to other metal chlo­rides. Those effect obeys Sechenov's law for the extreotlon of chlori­de complexes of Ga, In and Pe(Il'I) with different bafj­C dyes ( tri­phenylmethane, cyanine and azo­dyes).

The Influence of wa­ter structure on the ex­traction of IA should be mentioned. Рог ехалг"1 ­if one change ordinary wa­ter in the systm Fe ­

LiOl­HgO­ethylviolet­tolu­

tjr treated with magnetic

1.0 15„Л The dependence of lg D p versus radius of salting­out metal

field the extraction of FelA increases by 5­10Й. The same effect is observed when adding some hundredth percent of polyethyleneglicol (PEG­5000) or polyecrylamide which macromolecules distort water structure in some way. The increase in the extraction of IA when changing ordi­narly water seems to be explaned by modification of water structure and decrease of its' activity. For the same reason the solubility of IA in water phase is decreased.

In the extraction process of metal­chloride complexes with basic dyes in the presence of salting­out agents tha most effective solvents are aromatic hydrocarbons and some acetic esters (DK<6). These solvents are convenient because they practically do not extract metal­chloride complexes in the absence of dyes and extract dye chlorides not signifi­cantly.

Their extractio.. with aromatic hydrocarbons decreases with the de­crease of Hyldebrand solubility parameter in the raw benzene> toluene> etnylbenzene> propylbenzene >butylbenzene. The linear correlation is observed between Ig D» and alkyl chain length in benzene derivatives. The extraction efficiency of IA decreases also with the increase of sub­stut'ition extent in benzene ring in the raw benzene > toluene> o,p,ra­xylol>mesetylen. A quantitative characterstic of SA extraction effici­ency in dependence of sol ;ent basicity and molecular mass has been found.

25

EXTRACTION AS A SOORCB 0? ADDITIONAL INFORMATION J — WHEN CONCENTRATIONS IN HUMICOHPONBHT STSTEMS ARE I

9

" 'HTOLTAN30USLY DBTBRMHED I.G.Perkov, Kharkov State University, Kharkov, USSR

Information from measurement of analytical slgnala when they ace upon each other under certain assumptionfljmay be given by the following equa­tion Pj^f = nELd Sin гц' r^ ­ 1 (1) when %­ signals permiting factor, their selectivity, dash above ­a sign of value constancy within signal measurement interval from r u to r^;? ­the number of graduations on n signals intensity I, which is concerned with the interval of measurement results similarity,i: S = I/A

When different methods of analysis are applied extraction influence upon analytical signal has to do with an increase of its intensity in the process of extractions! concentration, i.e. with an increase of S grada­tions number.

Extractions! concentration is usually combined with group or individu­al separation, i.e. with decrease of the number of compounds defined in extracts obtained. As seen from equation (1) at high level analytical sig­nals intensity measurement standardization overlapping at interval r ­r,, decrease of their number in information growth from each of them and henck in similarity improvement of concentration definition results.

In most method of analysis with broad non­selective signals it's dif­ficult to use equation (1) for strict apriori comparison of results beca­ : use of permittance factors value uncertainty.

The choice and comparison of optimum information concerning components concentration from additive overlapping signals measurement in the metho­ds is advisable to do by minimization of errors in the results of concen­tration determination with the help of certain means and criteri»f2'J. Er­ror minimization ­ is the choice of information when all factors influen­cing it are included: the number of overlapping signals, their permittance and intensity. However it(s convenient to use equation (1) for causal re­lationship of error minimization results with information from signals measurement.

So, simultaneous photometric N d and S determination as being extracted into butylacetate of heteroligand complexes with pyridineazorezorcine (FAS and tetraphynilborate (TPB) after initial measurement optimization results in statistically uncertain results (see Table 1, N 1),

Fig.1 shows the absence of information about concentration of each si­multaneously defined components in signals measured. N.and S complexes lightabaorbance spectra are of mener difference in X , i.e. pei­Jiting factors R 0 end hence P = 0. These non­informative signals may be modi­fied with the keep of extraction in order to obtain their better permit­tance, к

26

LrFAB TPB complexes extraction at various pH may be measured on lightabsorbance of extracts. A quotient of lightabsorbance division upon lanthanide general concentration induced into a systea aay be ex­pressed as graduated coefficient provided that extraction in a system is constant when concentrations are measured in the range of their de­termination* Such coefficient spectra for Nd and Sm complexes extracts depending on pH are given in Plg.1, validity of such treatment of analy­sis graduation with excess of completing reagents in a system is shown inf3]. From Pig.1 it follows that maxima in extraction spectra £­pH are shifted,i.e. signal permiting factor is considerable and equations sys­tem for Hd and Sm simultaneous determination, which is composed using initial measurement optimization criteria 2 , gives satisfactory results (Table 1, N 2 ) as compared to a previous variant of analysis.

Table 1. Besults of simultaneous (Hd and Sm) determination with PAH and TPB Cm­ 1CKmol/l ­ concentrations defined. It's given: C Mj 1,56<10~5

mol/l standard error at n = 6 C

S«= 1,47­10"5

mol/l, S0­105

mol/l N pH Л .«™ C Hd ^ d c Sm a Sm 1 2

6,8 6,3; 6,6; 6,6

5000­550 510­550

1.7 1.5

2,2 0,14

1.4 1,5

1,1 0.09

1

In method molecular absorbance spetroscopy some additional in­formation about concentrations of components defined simultane­ously may be obtained when they are extracted into various sol­vents, i.e. at tae expense of sol­

£­10­* 2 \s\

500 X,nm 550 _ _ ?,0 7,5 pH .»..,,­,».... ­­ ""~ *"",,­, _. Fig.1. Extraction (£­pH) and absor­vatochromatics effects/47. Change — ° . , ,

J „ \ ~ ^ bance (£­Л) spectra for Nd(1), Sm(2) of medium may influence absorbance

. with PAR and TPB complexes spectra of each component great­ly and in several ways. Liniar equation system of Buger's law may be com­posed by choosing optimum positions from spectra (f-J.- and "­solvent for simultaneous photometric concentrations determination. Log wave bands of absorbance spectra of Mn(11) and Zn with PAN in CCl^ and CH,C00G 5H 1 1

extraction complexes show considerable solvatochromatics effects (see ­M3.2). The results of simultaneous determination of Mn and Zn concen­tration obtained by equation system resolution which are composed on ad­ditive lightabsorbance measurement in optimum positions in summary spect­rum of each extract (N 1,2) and in the number of positions from absor­bance spectre in two extracts (N 3) are given in Table 2. In the last case standard deviations in the results of concentrations determination are much less, since equation system there is more information about

27

Table 2. Result of simultaneous determination of Mn and Zn with PIN CjlO^mol/l, Sc'10°mol/1. Are given: C^» 1,1?.10"

5

mol/l, °lln

= °> 9 2 ­10"5

mol/l, n = 6

N SOLVENT , As nm C l l n Sun C Z n S Z n 1 2 3

CCl^; 5*5; 550; 555; 570, 575 0 ^ 4 . 0 ^ ; 530,550,555,565,570 CC1 3 ; 550,560. С ^ ^ . ­ Э Э О , 565 , 570

0,91 0,92

0,92

0 ,12 0,10 0 , 0 8

1,18 1,17 1,17

0 ,09 0 ,10 0 ,07

concentrations of each component because of using a greater number of signals, the number of components determined being equal. Cases of using solvatochromatics effect for growth of multicomponent

extraction photometric analysis results similarity may have a wide ap­plication since their realization does not increase duration of deter­mination: necessary repeated extraction by one solvent inchanged for ex­traction in several solvents.

Much additional information may be obtained using interligand exchange in metal chelate ex­tracts. Numerous variants of exchange realization combined give an opportunity to use all factors of information growth about concentrations of signal components ­ H, S and n according to equa­tion (1).

So, similarity of Cu, Fe and Ni simultaneous determination results in waste water on measure­ment of summary lightabsorbence of these metals pyridinazonaphtolate extracts in CH01,is much worse their results obtained by procedure modi­fied by indurlng oxlquinoline into extract (Table 3). Oxiquinoline in metathetical reactions only with cooper pyridinazonaphtolate yields cooper oxinate in extract not absorbing light in the range of spectrum, which is used when signal is

measured. The results of the determination on Gu lightabsorbance extract difference before and after oxlne introduction are of less standard di­viatlon, as results of Fe and Hi simultaneous determination on lightab­sorbance of complexes with FIN which were not changed after oxine additi­on. When waste waters are analysed the task of choosing analytical rea­gent for Cd and Zn determination is for instance, what is better for de­terming these elements: PAN or ditizone? Prediction, as inf2l, relative standard deviation Sr, of these elements simultaneous determination with PAN is about 10S(for the middle of determined concentrations fange), for ditlione ­ ie about 5*. When Hn is present extraction of only Cd and Zn *

550 580 X, nm Fig.2. Absorbance spectra of complexes Zn(1,2) and Hn(3,4) with PAN in CCl^and CH 3C00C 5H 1 1(1,3)

28

with PAH is impossible: Mn Table j. Results of sicaultaneous deter­is also extracted[5], and Sr m i a a t i c m of Cu, Fe and Cu with PAH

with (N2) and without use (1) of inter ligand exchange, C­IO^mol/1; S^ 1 0

mol/i ­ standard derivation, n = 4. N C

Fe ^ e C

Cu S

Cu C

Zn S

Zn 1 2

1,18 1,17

0,02 0,02

0,44 0,42

0,03 0,01

0,82 0,84

0,03 0,01

prediction when Cd, Mn дрД Zn with PAN are determined simultaneously fails: Sr=30% Extractions of Cd and Zn complexes with ditizone in CHC1, also inteferes bin. In this case interligand ex­change in extracts allows realization of the following scheme of analysis which provides determi­nation of all components with certain validity Cd and Zn with ditiaone being included. Extraction and concentration of Cu, Cd, Fe, Mn and Zn with PAN i.n CHCl,complexes at pH 10 ­» measurement of extract light­absorbance, А.^-г reextraction Cd, Mn and Zn in buffer solution with pH= 4 ­» measurement of extract lightabsorbance A,­» oxiquinoline introducti­on into extract and Cu, Fe and Ni determination as menrioned above * pH growth of reextract up to 11 and Cd, lin and Zn extraction with oxine solution in CHClz­vintroduction of ditizone into a solution subjected to the metathetical reaction with oxinates only Cd and Zn measurement ex­tract lightabsorbance and calculations •* concentrations resolution of Cd and Zn in (A^­Ag) with partial graduation coefficient for their pyri­dinazonaphtolates and Mn concentration calculation. The above example shows that interligand­ exchange in extracts combin­

ed with simultaneous concentrations determination may be used as a simple means of information growtt about cincentrations of components determined. Neutral chelates extraction into non­polar organic solvents may give essential information when complexing in solvents is studied, to use more exact when metrological unity of equilibrium conctants de­termination results is provided. It has to do which possibility to use the neutral chelate state in organic phase as a standard one since when salt composition of water phase is changed.

References 1. Vainfordner J. Numerical representations in analysis of substances.

In Spectroscopic methods of elements traces determination. U.: Mir, 1979. 49* p.

Drosd A.V. and Ar;zebashev G.V.//J. of analyt.chemistry. N 1. p.62. Chan Van Gkhan an J Drozd A.V. 113, H 4. P.606. Drozd A.V. and Shitchanin N.S. // J N 12. P. 2145. Dro«d A . V . / / J . rr analyt. chemistry.1982.V.37.H 10

Perkov I .U., 1987. V. *2, Perkov I .G., 1986. V. 41, Perkov I .G., 1986. V. 41, Perkov I .G., P. 1800.

of analyt. chemistry,

of analyt.chemistry.

29

PROSPECTS FOR THE APPLICATION OP THRBB­PHASB EXTRACTION Г~^~~j SYSTEMS TO AlfAiYTICAb PRACTICE I > A.P.Balonin, Polytechnik Institute, Leningrad, USSR The moet of the separating methods are based on the formation of

a new phase in whose volume or on surface analyte is concentrated. Only extraction allows to obtain an equilibrium between several dif­ferent in their nature and practically easily separated liquid phases. This ensures the principal possibility to distribute each of analytes in a separate phase.

Traditionally the extraction was considered as a process of inte­raction of an analyte with Just an organic compound , i.e. solvent. Now a concept is getting confirmed that a process of extraction by the phase, i.e. a complex extraetant with properties different from a so­lvent, takes place. In multi­phase extraction systems the new phases being formed are al­ways Intermediate in their properties between the aqueus and the or­ganic phases have different extraction ability.

The most videly spread and studied type of multy­phase extraction systems 1в the three­phase systems (IPS). There are more than 200 pu­blications on TPS in literature but in most of them the formation of the third phase OP) is considered as a drawback. In most of papers no difference is made between TPS of different types.

The lack of systematic arrangement of TPS made difficult to desc­ribe them and to find out the most prospective ones for their practi­cal use.

We have suggested the following classification of TPS on the basis of the nature of equilibrium phases (Table 1).

TPS­1,1 are formed while extraction from acid solutions by mixtures of amines or TBP with diluents takes place.The 3F of this type of TPS was used aa on extractant for effective extraction and concentration of micro­eleaents for the first time [l].

TPS­1.2 are formed by lamination of an aqueous solutions of the po­lar organic solvent in the presence of inorganic salt and non­polar diluent. For the first time in analytical chemistry such system was used for simultaneous quantitative separation of three.elemente by ex­traction each of them Into one of the equilibrium phases [2] . It cou­ld be assumed that the existenoe of significant number of polar solve­nts combined with different complex forming agents produces the 3P with essentially different properties and makes these systems useful for realization of multi­phase separation methods.

TPS­2 are the first ones which were prcticelly used. In the most studied system WeClj­HCl­E^O­DIPE the 3P is formed by lamination of organic phase into ether phase and phase of hydrate­solvate CH3) of BTaClj,. Senesis, composition and properties of the ЭР confirm that it

30

Tuble 1. Classification of TPS

Equilibrium phases I water* org.1+ org.'г

Three­phase extraction systems T

3P is formed by

T solvate of inorg, compound

T Г

water* water2+org

polar organic solvent

HS of complex metal­ha­

llde acid

I RS of

aoid

HS of the strong heteropo­

ly acid

organic phase |non­polar diluent

Type of TPS

Examples

D С extractant

I TFS­1.2 T

TPS­2 TPS­3.1

KC1­

­HgO­

­OHjCS­

г

6н

Н

FeCl3

­ЯС1­

(i­ijR^O

X I

ips­з.а Me" 2­­HCl­

(с гн 5) 2о

мо 3­н 2о­

3 P O 4 ­

(с 2н 5) 2о

is the second organic phase in its nature. Extraction ability of the 3P depends on free HC1 content in it. For example, extraction of cop­per (D^0,3) could result in its significant loss. But it is possible to optimize the conditions of extraction to make TPS­2 effective for group separation of micro­elements byn

casting of the matrix" and makes method compatible with a lot of methods of final determination becau­se the separated impurities are left in the aqueous phase, thus re­extraction is not necessary.

The most unusual type of TPS was discovered and studied by us [3j. TPS­3 are formed in systems HeX^­Hx­H^O­I^O as a result of lamination of the aqueous phase at high acid and salt concentrations (Fig.1). Here X­Cl'.Br'.ClO^; RjO­DBK.DPE.DIPE. The 3P is produced by HS of HX with various ratio water/ether. The contents of the 3P are determined mainly by the nature of the acid and less influenced by cation of the salt.Thus for all chloride systems investigated (HgO+DBEVCHCl+JIeCljj) »5.5(S«0,2). The confirmation of high water and acid content with con­siderable concentration of ether in the ЭР (table 2) makes it unique extractant comprising two equilibrium aqueous phases with similar con­ditions of hydration and complex formation, being unusual for conven­tional extraction. The JF has intermediate values of dielectric cons­tant between aqueous and organic phases. This determines the same pri­mary hydration (solvation) of ions and identity of complex anions' fo­rms in the aqueous and the 3P. The distribution is mainly influenced by the proportion of charge and radius of ion at this conditions. The investigation of distribution of more then twenty coplex ions proved

3t

that 1л empirical equation 0~p«K"c5Li*. I o T dependence of micro­com­ponents' concentration in the 3P on their initial concentration in so­lution parameters К and m are influenced by the charge and radius of complex ion (Pig.2).

three phases

lahie 2. The composition of the ЭР

20 40 60 SOQmkM Pig.1.Phase diagram of TPS­3

m

UPS Content.» mol. % UPS nzo ether HC1 salt

TPS­2 TPS­3

f7­35­

65­37 55­80 10­15

4­12 17­20

" 3­4 0.2­0.5

1,0

0,5

IrClWtys

IrCt?

0,5 iO %/r ~m7f%

Fig.2.Pependancy of parameters m & К on radius г & charge Z of ions

The large si*e of ion rge ions (table J ) .

The th i rd phase of TPS< Being fast and simple, t e r i s t i c s .

Table 3.Extraction of metals by the

Complex [osirj

2

^ [ptCl2(SnCl3)J

2

" flrCl(SnCl3)^P" jpaCKSnCl­j)^

4

" RuCl2(SnCl3)j3­

provides effective extraction even for high­cha­

~3 was used for extraction of platinum metals[4J, the procedures have good metrological charac­

The use of the 3P as a high­effec­platinum tive extractant has advantage of eaay 3P of IPS re­extraction of extracted components.

To vary the extractive ability of the 3P it is necessary to charge its contents by using different RgO and / or anions. It is possible to predict new TPS­3 on the basis of discovered criteria using data on equilibria in МеХд­HgO and HX­HgO­HgO systems.To form TPS­3 it is necessary that: (1) an organic solvent (ether) should be dissolved sufficiently in the aqueous phase of the three­coaponent system

60 100

7 300 260

7­150 ?­200

in the range of complete hydration of the acid (Pig.3)J (­4 f»i one mol of HeCl equivalents should be h< 6 mol of water in saturated so­lution (12 mol for perchloratee) (table 4).

Table 4. Solubility of salts and the TPS­3 formation

ether •A 20

10

FJ6­.3,

MPE

40 20 IM\ Solubility of ether in EC1 acid

Salt h TPS Salt h res

HH 4CI 3.0 _ A1C1 3 5.4 + HH+Br 7.2 ­ CrClj 10.4 ­

MH 4I 4.7 + CoCl2 6.8 ­NaCl 9.0 ­ HiCla 5.6 + HaBr *T? + Hal ,4t7_ +

Prospects for use of TPS are based on the tiro aspects of their no­velty: ­ realization of multi­phase extraction method different from always two­phase conventional extraction methodsJ ­ intermediate properties of the ЭР as unusual extractant. It should be mentioned that this unusual extractant could be also used for the two­phase extraction equilibrium.

The requirements to method of separation and concentration has be­en changed because of development of modern analytical methods. Very, often now there is no need in quantitative separating of interfering­components. It is sufficient to diminish their contents to 100­1000­fold excess over the concentration of an analyte. While using high­selective methods of final determination it is reasonably to perform group concentration of analytes. Thus non­selective extractant should be put in practice. TPS of all types could be effectively used to sol­ve these problems. On the other hand, TPS­1.2 and TPS­3 could provide high­selective separation especially by application of non­traditional water­soluble reagents.

The results of investigation and practical application of TPS prove the possibility to enricl analytical practice by non­traditional ext­ractants and extraction methods. Puther investigations of TPS will co­ntribute essentially extensive application of extraction In analytical chemistry.

References 1. Zhivopistsev У.Р., Ponosov I.

Vol.16.P.1432. 2. Pyatnitsky I.V.

1983. Vol.51.P.

N. »t al.//Z. neorgan.Khimli.1963.

Prankovsky V.A. et al. //Ukrainsky Khim.Zhumal, .газ.

Balonin A.S., Kalinins L.B., Deniaov Yu.P.// Izvestiya VUZ.Kaimiya i Khim.Technol.1979.Vol22.P.47. Kalinins 1.В. Novakovekaya E.CHalonin A.S./V Zavodskay» labor. 1962. Vol.4B.P.6.

33

ОН НЕ» EFFECTIVE TECHNIQUES FOR HCTRACTIOH SEPAHATIOH AHD i DETBRltnrATIOir OF STJBSTAJTCES Hf TWO­ Л1ТО THREB­PHASB LIQUID I SYBTaiS I.V.Pyatnltaky, L.b.Xo^oaiet» and 7.a.»rankov»ky, Kiev Stat* University, Kiev, UBSK The oaw techniques proposed give «Ida poaalbilitlaa for toe extrac­

tion separation and concentration of organic and non­organic substan­

ces in three­phase systems and the selective extraction­photometric determination of elements in two­phase systems.

Three­phase systems. The new type of three­phase extraction sys­

tems (TES), v i z . , an aqueous solution of mineral acid sal ts , a polar i and a non­polar solvent, has proved to be effective for the concentra­ • tion and the separation of non­organic aa well as organic substances present in various environmentals [ 1 ] . The poss ibi l i ty was shown and the conditions were established for the cation extraction of Zn, Cd, 7, Cr, Al, Hi, Ce, Cu, Fe, Ug and rare earths in TES by means of pe­

largonic, caprlc and lauric acids la combination with a l ly l ­ , pentyl­, ; decyl­ or benzylamine. The obtained experimental data on the extract!­

on distribution of metals in TBS led to the following conclusions con­

cerning the possibi l i ty of the separation of elements by means of a single three­phase extraction.

1. Metals, which form oxygen­containing anions ( e .g . , vanadium, chromium, manganese, bismuth, molybdenum, tungsten e tc . ) are practi­ { cally retained completely into the aqueous phase in the alkaline me­ j dium. I

2. iletals, which form stable amine complexes ( e . g . , copper, oi :.S '1, j cobalt, zinc, cadmium e t c . ) , are quantitatively extracted into the i middle phase under the same conditions. '

3. lietals, which do not give compounds mentioned ( e . g . , iron, alu­

minium and rare earths), are extracted quantitatively into the upper phase.

The techniques were elaborated for the separation of triad metals, e .g . , iron­copper­vanadium, iron­copper­chromium, iron­nickel­chromiun e tc . , by means of a single three­phase extraction.

The study was performed on the extraction behaviour over 100 orga­

nic compounds of different classes: non­ionic, cationic and anionic ' surfactants, dyes, mineral o i l s , one­, two­ and three­atom phenols, amines, oonocarboxylic acids, fats and aromatic hydrocarbons. New tecl niques were worked out for the single extraction separation of three groups of organic substances ­ high molecular weight amines, surfac­

tants and three­atom phenols; mineral o i l s or fats , one­atom phenols and low molecular weight amines; aromatic hydrocarbons, surfactants, dyes etc. We proposed a new technique for the concentration of metalsv or organic substances which allows not only concentrate but at the Bar­

me time separate these substances tranafering them into different or­

ganic phases of the three­phase system. Пек techniques «ere elaborated for separating and determining the mixture of 4­5 organic substanceв ­

two dyes, trioctylamine, fat and oi l ­ by means of performing two pa­

ral le l three­phase extractions over pH ranges 0­4 and 9^V3. The table presents experimental data on the extraction behaviour of organic sub­

stances in TES: 3 H aqueous solution of sodium chloride ­ acetonitrile ­ hexane.

The distribution of organic substances between the three phases of the system: 3 H aqueous solution HaCL ­ CH,CH ­ CVH,..

Phases Substances ^ e x

Phase 111 (hexane)

Amines < > C 1 0 ) Mineral o i l s Fats (vege tab le , aminal) Lubricant mater ia l s (cable l u b r i c a n t s e t c . ) Uonocarboxylic ac ids (>C,| ,­) Aromatic hydrocarbons (benzene, t o l y e n e , xy lene)

8 ­ 1 4 0 ­ 1 4 0 ­ 1 4 0 ­ 1 4 0 ­ 6

0 ­ 1 4

Phase 11 ( a c e t o n i t ­

r i l e )

One­atom phenols (phenol, c r e z o l s , 1­naphthol) Surfactants (an ion ic , c a t i o n i c , n o n ­ i o n i c ) Amines ( C J ­ C ^ Q ) Dyes (ac id grsen anthraquinone 1Г­2С, acid br ight green anthraquinone H­4G, d i r e c t black 20 , c r y s t a l v i o l e t ) Bengal pink В

0 ­ 9 0 ­ 1 4

10 ­ 14

3 ­ 1 2 0 ­ 6

Phase 1 (3» NaCl)

Amines (С,,­*^) Three­atom phenols Soupe i lonoearboxylic ac ids (C_.­C,) Dyeв ( c a l c i o n , xy leno l orange, d i r e c t black 2C, r e a d i l y cleaned red , srsenazo 1 , acid s c a r l e t 2G, acid chromium dark blue) Cbromazurol S

0 ­ 8 8 ­ 1 4 8 ­ 1 4 0 ­ 1 4

3 ­ 1 2 9 ­ 1 3

i This table shows the possibility of separating quite different com­binations of three organic compounds in a single extraction.

The calculations made for the interaction energies of substances with aqueous end organic phases allowed us to explain and predict the reasons of compound distribution into different phases of the three­лЬаве extraction system. ! Two­Tflme systems. In order to increase selectivity, sensitivity Contrast of the reactions between metallochrome reagents and metals

35

in their1 extraction­photometric determination we proposed to use sized extractants, composed of chloroform, concurrent acid reagents, mono­carboxylic acids ia particular and neutral solvating organic compo­unds, e.g., alcohols, Л study has been made of the extraction of coup­lers e of gallium, indiua, aluminium and other metals with Pyrocatechol Violet (PCT), Briochroae Black T (EBT), Bromopyrogallol Bed (BIB) and other metallochrome reagents into the mixtures of chloroform, capronic (propionic) acid and n­butyl (n­amyl) alcohol.

Monocarboxylic acids and alcohols provide the extraction of hydro­philic coloured complexes into the organ!; phase by пиале of solvation

of these coapounds. iionocarboxylic acid anions influence concurrently the complexation between metals and dyes and thus promote the extractioi selectivity of coloured compounds. The composition of complexes extrac­ted into the organic phase can he shown on the example of gallium com­pound with BOB and capronic acid Ъу the given formula.

The expression for extraction constant of gallium complex with PCT "into the mixture of chloroform, capronic (ВЯ) and propionic (Hi) acid/ is as follows:

*ex [Sa(BH4H3Ind)RA­6HHj"0 [ H + J

?

w

l o7

The concentration of coloured complex of gallium in organic phase was estimated by use of optical density of the extract, taking into account the degree of extraction of metal into organic phase. The con­centration of H + was determined experimentally. The concentration of free gallium lone in aqueous phase was calculated taking into account the) hydrolysis and the interaction with propionic and capronio acids by the equation!

• F * « „ ­"Ga(w)

where C(j a / W \ ­ i s overall galliua concentration in aqueous phase after the extractioni

a = • ' • * • • • — — • — t

. 1 0 ­ 2 p H

where Kj[ and к£ ­ are first and second hydrolysis constants of galliu b(d) ­ is a coefficient, presenting the concentration of anions of propionic (capronic) add in aqueous phase Л~(Н~), calculated by use of the distribution constant P, dlmarization constant K4 and dissooi­i»:

36

ation constant Kg, (Egg) by the equation: 0^­],= ­ IBUA W( V ^ U ) 2 / ^HA ,

where Kg. + fH*J • F H A fH*J •°Д4 = i

8 n and. 6 - are s t a b i l i t y constants of gallium complexes with propio­

n ic and capronic acids respectively, ^ J ^ J ­ i s s t a b i l i t y constant of gallium complex with POT, fH­Ind~3­ i s equilibrium concentration of PCT in aqueous phase:

, < cprv " [Ga(HBLH,Ind)HA­6HHl ) [В*],. [H ?Ind ] = — S a =­* — ,

( Ю, + [H +

J w ) ( D + 1 ) where C p ^ ­ overall concentration of PCV, Kg­ diasociation constant of the phenol group of PCT substituted by metal, D ­ distribution co­efficient of PCV, The concentration of nononeric form of propionic (capronic) acid in organic phase was calculated by the equation:

[нл1 ~3a Л / ( *** >2 + ­^­

* KD(HA)

V * VHA) 2 K

D(HA) The calculation was made with the computer 11­6000 using programs

in TOBTEAH. The value Kex=(1,<4o,5)­10^ was found for this system. The use of the mixed extractants

promotes the optimisation of spect­rophotometry! с characteristics of со­ ' loured complexes of metals. Fig. shows that in the case of gallium 0,3­

complei with ECB there is a counter­directed shift of absorption spect­ 0,0| ral lines of extracts of dye and 480 540 600 660 \,nm complex compared with those in aque­ The absorption spectra of ous solutions. It provides the Inc­ BCB(1­3) and its complexes with re as la reaction contrast from 30 Qa(4­6) in aqueous solutions to 90 nm. At: the saas time it incre­ (5Л), chloroform solutions of aaes the absorption intensityC2J. EE(1,5)and mixed extractants(2,6)

The use of mixed extractants made it possible to elaborate exclusi­vely selective, sensitive and contrast methods for gallium and indium determining.

Reference» 1. Pyatnitsky i . T . , Prankoreky V.A./ /Isvest ia Vuxov. Chemistry and

chem. technology. 1987­ Vol.30. P.3 . 2 . Pyatnlteliy I.V.,KolomietB L . W / J . i n i i . C b » * . 1985. Vol.40. P.115

37

THREE PHASE EXTRACTION IE THE TRACE AHALYSIS S.Arpad;jan,E.Vasileva,St.Alexandrov Faculty of Chemistry, 1126 Sofia,Bulgaria

9­9

The determination of traces in insoluble lead compounds after their transformation to PbfDDTC), and extraction of the DDTC­comp­lexes of Cd»Co,Cu,i^,Si,Zn thus formed offers another way for sepa­ration and preconcentration [1] .These three phase extraction tech­nique could solve some difficult analytical problems connected with the trace analysis in various pure substances.The possibili­ties for quantitative separation of traces of motal dithiocarbama­tes from a dithiocarbamate as residue are not investigated [2,3]. The precipitation of the macrocomponent as dithiocarbamate at si­multaneous extraction of the dithiocarbamate complexes of the trace elements in small organic phase were applied to analysis of pure TeC2,SiS0. and CoSO^.

Table 1. Three phase extrac tion for trace analysis in Te0 2 tion for trace analysis in Te0 2

Ele­

ment 2g TeO, dissolved in NaOH 1g TeO, transformed to Te(DDTC)1 Ele­

ment Recovery(SG) ь/ь 0

х RecoveryW b/b 0*

pH 9 pH 10 pH 12 pH 10 pH 6,5 pH 10 pH io

Co 98±2 98f3 90+3 0,99 98+2 98+2 0,99 Cu 99+1 99+1 94+4 0,99 96+2 98±2 0,97 Pe 97±2 96+3 ­ 3+1 0,96 90+4 95+3 0,88 Si 98+3 98+2 85+3 0,99 94+2 98+2 0,90 Pb 98±3 98+2 0,98 6B+5 94+3 0,92

The dependence of the three phase extraction on pH for trace analysis of Te0 2 (Table 1) indicates that the best results are achieved at pH 10.The TeO, samples were dissolved in HaOH and then the pH­value adjusted with HC1(1:1).Quantitative concentration of the trace impurities as dithiocarbamate complexes in the organic phase by direct transformation of the TeOg to Te(DDTC). in pre­sence of borax buffer(pH 10) is possible only for 1g sample.The Standard addition method ni'ist be applied for calibration.For 2g sample the transformation from one solid phase(TeOj) to another (Te(DDT'"),) is not complete.

The three phase extraction technique permits the determination of traces in 1g HiS0.(Table 2).

38

Tab!» 2. Three phase extraction for trace analysis in HiSO

Sle-nent

Becovery(X)

*/V

Sle-nent o' w

*/V

Sle-nent

T i l " 1 1:2 1:3 1:4 1:5 */V 3d 82±5 90±3 >99 >99 >99 0,99 3o 90±4 98±2 >99 >99 >99 0,99 Cu - - - >99 >99 0,99 Pe - - - >99 >99 0,98 Pb 80±4 90±4 98±2 >99 >99 0,97 2n 80*6 88±5 92±4 >99 >99 0,97

«b - slope of the calibration curve after ethree phase extraction from matrix.

V slope of the calibration curve after extraction from water standard solutions.

Table 3. Statistical characteristic of the three phase extraction procedure

Ele-

neat

Concentration

jU«/g

I e 0 2 HiSO, 4 Ele-

neat

Concentration

jU«/g sr(*> Dlfytg/g) y*> M.*(ii.g/g)

Cd Co Cu Pe Si Pb Zn

0,8 1.6 0,8 1,6 1.6 1.6 0,8

6 4 3 5 4

0,08 0,2 0,05 0,3 0,2 0,3 0,08

5 3 2 4

4 5

0,1 0,3 0,07 0,5 '

0,5 0,1

ij . . - detection limit

Th» data represented on Fig. show the effect of extracting time on trace recovery in the extraction from HiSO. after dissol­ving one gramm of the sample in 10 mlH.O and precipitating the macrocomponent as Hi(DDTC)2.Lower recoveries of the trace elements in the organic phase at higher extraction time are presumably due to their adsorption on the solid phase during the extraction step.

39

Extracting time .minutes Effect of extracting time on trace recovery

The three phase extraction is not appropriate for analysis of cobalt compounds.The cobalt diethyl dithiocarbamate residue does not permit the formation and separation of throe phases and the atomic absorption measurement of the organic phase is impossible >

References 1. Arpadjan S.,Ko;|narslca A. .Djingova R.//Presenius Z.Anal.Chem.

1986.Vol.325.P.278. 2. Zolotov Yu.A. ,Kuzmin H.M. Eonzentrirovanie mikroelementov.

Hoskwa :Himia, 1982. 3« Reeves R.D.,Brooks R.R.Trace element analysis of geological

materials. Hew York - Toronto:.John Wiley and Sons,1978.

4»

ON SOME HEW POSSIBILITIES OP THE EXTRACTION CHROMATOGRAPHY 9­10

V.A.Luginin, G.S.Katykhin, N.E.Kondratieva, L.T.Dubrovina, Leningrad State University, USSR

Versatility, simplisity and great metodioal flexibility of extrac­tion in comparison with ion exchange accountedfor widespread us of the extraction chromatography in 60­s and 70»s. However the nearly complete instrumentation, especially after the appearance of ion chromatography blocked the further development. The main disadvan­tage of the extraction chromatography ­ the unatubility of the stati­onary phase on an inert support creates results in some serious difficulties in using the extraction chromatography columns in HPLC.

The.t is тгпу the extraction chromatography now moves in the follo­wing directions:

1) the use of bond higly effective sorbents with a further fixa­tion of a stationary phase on their sufface togetherwith up to date chromatographic equipment, e.g. ion­pair chromatography; 2) the use of solid phase polymeric materials with an extractant in the sorbent matrix (solid extractants); 3) the use for semipreparative separations of a drop counter flux chromatography as a kind of "column less" chromatography; 4) a modification of the classic extraction chromatography with the aim of enhancing the selectivity of separation,the proper choice of stabilization conditions for the stationary phase on a support and the automation of bath the separation process and determination. The selectivity of separation process can be enhanced by a combi­

nation of several separation mechanisms, as has been shown in P­3J . We investigated chromatographic ion behaviour under simultaneous extraction and electromigration. The process was carried out by app­lying to the extraction chromatography column a DC voltage of vari­ous manitude and polarity. If the direction of ion electromigration in the electric field coincides withthe direction of the eluent flux, the potential is considered to be positive (i.e. accelerating). The change of polarity is considered to be equivalent to a "negative" potential (i.e. deccelarating potential). An outlay of the used equi­pment based on a typical liquid chromatography is given in Pig.1.

The main units are: 1 ­ a DC power supply; 2 ­ a special chroma­tography column with an electrode compartment and cooling; 3 ­ a •feeding pump; 4 ­ a photometric detector; 5 ­ a recorder.

The behaviour of Ag(I), Cu(II) and Ce(III) ions has been studed in lystems HDEHP ­ aquous asidic solutions. The potential qradient va­

41

л 1 2

I 3

FigЛ. A diagram of an experimental set up with an automatic regi­stration for the use potentials applied to the column

ried in the range of 10 ­ 30 v/cra. For comparison the chroraatograms were obtained without no potential applied to the column and also the ions' migration in the electric field for no movement of the unsta­tionary phase. Pig.2 gives the elution curves of Ag(I) ions with no potential and with a potential of different polarity. The eluent flux may kept constant at 0.2 ml/min»cm .

Fig.2, i—J—5—<ПЙГ Elution curves of Ag(I) ions: 1 no potential applied;

2 ­ accelerating potential applied; 3 ­ decelerating potential applied

In Table 1 data are given on chromatographic behaviour of Ag(I), Cu(II) and Ce(IlI) ions in case accelerating (+) and decelerating {­) potential applied to the column.

Table 1. Maxima of the chromatographic peak values (V ) and half­height' Widths (WA _) in units of free column as a function of ex­perimental conditions Voltage

gradient, v/cm

Chromatographic parameters of the ions Voltage gradient,

v/cm Ag(I) Cu(II) Ce(lII)

Voltage gradient,

v/cm V mr W

0.5 mr W

0.5 V mr «0.5

0 + 10 +20 ­10 ­20

2.0*0.2 1.5*0.2 1.3*0.2 2.3*0.2 3.5*0.3

1.4*0.1 0.9*0.1 0.6±0.1 2.3*0.1 3.4*0.2

2.8*0.3 2.2*0.3 1.5*0.2 2.9*0.3 3.2*0.3

1.5*0.1 1.0*0.1 0.7*0.1 2.2*0.1 2.6*0.1

2.5*0.2 2.0*0.2

2.5*0.2

2.9*0.;,

1.0*0./

4.7*0.

All of the ions on application of the positive potentials are eluted earlies and the chromatographic peaks shorten. If the sign is reversed the picture is inverted, but the effect of the ion charge is difficult to evaluate. The results show that with an applied poteij^

42

tial one can get the double effect; the change of selectivity because of the field influence upon the ce.pacity factor and the increase in efficiency because of the peak width change.

On the base of the obtained results separation of V(IV) and V(V) and also Oe(III) and Pr(III) has Ъелп carried out in the system HDEKP ­ mineral acid under of acidity wh .ch is in "normal" chromatographic condition excludes the separation ]4].

The used extractant ­ HDEHP has a low rate of mass transition for therfore one can expect an greater rate of mass transition and influ­ence for the other sistems *ith more faster kinetics.

An important characteristic of he stationary phase in the extrac­tion chromatography is the stabili у of the phase On a support. The loss of the extractant into eluent may cause a change of the detector responce because the extractant in the solution under certain condi­tional markedly shifts lowers the ixtinction of coloured complexes of some ion metals, as been shown n [5] .

Two ways to overcome the drawb&oks are possible: first ­ removal of dissolved extractants from the solution; second ­ the search of synergetic combinations of extraction sistems with a higher stability of the phase on a support.

Experimentally it was found that upon passing the eluate through a layer "dry" inert support (e.g. eflon) excludes the extractant from the solution and excludes the associeted systematic errors. That is why the experimental set up included an auxiliary column with the inert support put in after the separation column. ТБР ­ HCL sistem for the separation of an ion mixture of Fe(III), Hi(II) and Mo(VI), chosen as a model sistem. The mon ;oring of elements contents was carried out using known nhotometr. ; methods of Pe(III), Ni(II) and Mo(VI) determination. Under thes< conditions the analysises were ma­de with and without the auxilliar; column for the removal of the ext­ractant. The results are given in Table 2.

Table 2. The results of quntitt­tive determination of metals by extraction­chromato­photoroetri; method Metal Taken,

ppm DelTmind metal, ppm Metal Taken,

ppm extrac;ant not re noved

extractant removed

Pe(III) Ni(II) feu(VI)

100 20 30

•х,п=ю i ;

m­% s r,% X,n=10 «V* sT,%

Pe(III) Ni(II) feu(VI)

100 20 30

80 20 22

20,0

26.7

20.5 4.2 13.6

101 20 29

1.0

3.3

4.0 3.1 4.5

43

Another way the increase of selectivity and efficiency can be ob­taned by the use of senergetic combinations of extractants. We have studied in details a system ­f twoji­diketones: HDBH and HAA, which substantially betterthan others in kinetic parameters, thus providing high extraction of metalls under large flux rates Щ . Besides the зуг»ет is higly stable on the support.

The synergetic effect here is shown on the example of extraction study of Ni, Co, Cu and U. Thecorressponding curves are given in Pig. 3 . The composition of the formed complexes has been determined and

the extraction constants has been evaluated. Prom the temperatyre depen­dence of the constants the thermody­namic characteristics of the proper synergetic extraction reaction have been evaluated. The analysis of the result makes it possible to suggest a mechanism of formation of synerge­tic adduct upon extraction of the ions.

Presented material shows that the possibilities of extraction chroma­tography are far from exhausted. In 86­87s the new models of liquid chro­matographies tecniques with the inert fluid flow line were appeared. Their gidravlic line were made from titan

or polimer materials for the biotechnology (LC 410 BIO ­ Perkin­Elmer, Bio LCC ­ Dionex, etc.). It can be relied upon succeccful using them in inorganic chromatography analysis with the strong acids or the complex reagents used as mobile phase.

References 1. Luginin V.A., Tserkovnitskaya I.A. //The Problems of Analytical

Chemistry. Leningrad Univ., 1976. Nurab.1. P.155. 2. Luginin V.A., Tserkovnitskaya l.kJ/Tbld. P. 162. 3. Luginin V.A., Kazachenko S.Yu., Tserkovnitskaya I.A.//Zh. Analit.

Khim. 1977.Vol.32.N12. P.2324. 4. Luginin V.A., Dubrovina L,T.//Abstr. of Reports solvent extraction

conf. USSR.atga, 1982. Vol.3. P­73. 5. Luginin V.A., Yakushkina M.S., Tserkovnitskaya I.A.//The Problems

of Analytical Chemistry. Leningrad Univ., 1981. Uumb.3. P­3 6. Luginin V.A., Korooeynikova E.G., Tserkovnitskaya I.A.//Vestnik

Leningrad Univ. Piz. Kfcim. 1981. H10. P. 111.

The curves of ay­nergetic extraction metals in the system ЩФМ­HAA­CHCL, 1­ U; 2­ Co; 3­ Cu; 4­ Ni

44

SOLVENT EXTRACTION OF PRECIOUS METALS BY DERIVATIVES OF DEHYDRODITHIZONE 9­11

A.Kettrup, M.Grote, G.Pickert, Angewandte Chemie, University of Paderborn, D­4790 Paderborn, FRG Several sulphur and nitrogen containing ligands have been al­

ready utilized for the selectivp separation of precious metals. However, chelating solid and liquid extractants of this type coor­dinate platinum metals very strongly, so that the release will be incomplete and the ligand may decompose. Evidence is given that the binding of anionic chlorocomplexes of metals predominantly via an anion­exchange mechanism makes the reversion of the extraction easier.

Sone time ago we observed such effects by comparing two differ­ent types of functionalized polymers with incorporated sulfur bon­ded dithizone (P­D) or dehydrodithizone (P­TD) as functional groups [1,2].

Obviously, *­.he properties of the an ion­exchanger "P­TD" were more profitable in comparison to the chelating resin P­D. Although the sorbent P­TD removed gold, osmium and the primary pla­tinum group metals effectively and reversibly, some special problems remained. For example it was difficult to adsorb and to elute Ir(IV) completely from the resin. Furthermore, the desorption yields of the primary platinum group metals were also affected by co­extracted iri­

dium. Therefore we were keenly interested in the extraction behav­iour of liquid analogues of P­TD resin. For this purpose various substituted derivatives of dehydrodithizone were prepared as demon­strated in Figure 1.

°^ JTD ­­ n л .N—N „ e

n Б a 1 Phenyl (p­substituted)

2 ­ с 9 н 1 9

3 "

C

11H

23 4

" C

13H

27 5 ­ C

17H

35 Flg.1. Formulaes and symbols of tetrazolium salts

t 45

These pale yellow, crystalline compounds are soluble In alcohols, halogenated alkanes and aromatlcs.

In order to investigate their basic properties, extraction procedures were carried out with 5 • 1<Г3

то1аг solutions o£ the appropriate extractant, dissolved in chloroform. Equal volumes of a tenfold molar excess of the anion exchanger and aqueous metal salt solutions were shaken together for the appropriate time and then analyzed.

The aryl-substituted compound T-D1 extracts some metal ions in their most common higher oxidation states very easily. ftu(III). Os(IV) and Ir(IV) are separated from the aqueous phase to more than 90 % as soon as they are shaken, while the extraction yield of Pt(IV) is remarkable lower and in the case of Fe(III) negligible. However, it is noteworthy that Pd(II)- and Pt(II)- are precipitated immediately in the organic phase by the aryl-substituted extrac-tants.

in contrast to this behaviour, the alkyl-derivatives form ion-pairs with chloro-complex anions of PGM's without precipitation. As demonstrated by Fig.2 individual solutions of gold and precious metals can be treated successfully with TD-2 in the whole range of hydrochloric acid concentration. The results obtained leads to the following order of extractability. Au(III) > Os(IV) > Ir(IV) > Pt(IV) >> Pd(II) .

As expected, the extractability of iridium drops down nearly completely after reduction to the trivalent state. However, the analogues behaviour of the Pt(II) species is somehow surprising, as in common ion-exchange systems chlorocomplexes of both Pt(IV> and Pt(II) are effectively extracted.

60

40 •

го -

5 с HCl lul/l)

Fig.2. Effect of hydrochloric acid concentration on the extraction of individual metal ions by TD-2

46

Similar extraction yields, obtained for the decyl-derivative TD-2, resulted also by use of the other long-chain alkyl compounds. However, the tetradecyl - and octadecyl - derivatives exhibited an increasing tendency to form emulsions during the extraction step., so that further work was restricted to TD-2,

For a final investigation of the basic extraction properties of this tetrazolium salt mixtures of gold, platinum group metals and base metals in their common higher oxidation states were applied at increasing hydrochloric acid molarities. The data obtained are pre­sented in a diagram (Pig.3).

Fig.. 3. Simultaneous Extraction of precious und base Metals at different HCl-concentrations (shaking time 1h)

The result of this study meets our expectation from the results presented before. Thus Rh and Ru as well as Cu and Ni are extracted only negligible in every case, whereas the increasing concentration of hydrochloric acid furthers the formation of extractable chloro complexes of Fe(III). However, it is noteworthy that the base metal extra-;rH can be removed simply by action of dilute hydrochloric acid.

The ability of a metal loaded extractant to be stripped is a very important property. In the case of the palladium-loaded tetra­zolium salts, dilute perchloric acid acts satisfactorily as strip­ping medium.

A solution of thiourea is also effectiv in contrast to the thiocyanate ligands. The choice of a stripping agent depends on the actual separation problem as the presence of coextracted metal ions influences the stripping characteristics of the individual metals strongly.

47

For example, admixed Ir(IV) leads to coprecipitation of Pd and Pt during the thiourea procedure. Only gold is stripped completely and osmium remains unattached in the organic phase. Repeating this experiment with a different feed solution (Fig.4) which did not contain Ir results, however, a quantitative release of Pt and Pd together with Au into the aqueous solution.

Fig.4. Simultaneous stripping of metal ions from TD-2 with thiourea (Ir not present) -extractant freshly loaded-

It is interesting to note that a three days storage of the loaded TD-2 causes a retarded release of Pt from the organic phase by action of thiourea. This effect observed may be a hint at a more complicated mechanism of extraction, involving not only simple ion-pair formation.

Nevertheless, the class of sulphur containing tetrazolium salts is still under investigation and additional experiments may verify some advantageous properties.

References 1. M. Grote, A. Kettrup //Anal.Chim.Acta. 172 (1985), 223. 2. M. Grote, A. Kettrup//Anal.Chlm.Acta. 175 (1985). 239.

48

9­12 EXTRACTION О? METALS WITH CROWH­COMPOUNDS AND ORGANIC ACIDS ANIONS

V.V.Sukhan, A.Yu.Hazarenko, Kiev State Univers i ty , Kiev, USSR

Numerous papers report on metal ion e x t r a c t i o n with crown compounds and d i f f e r e n t counter­Ions , mostly p l c r a t e s ( eg the recent r e v i e w [ Y | ) . In previous p u b l i c a t i o n s we have s tud ied the in f luence of organic dye anion and the nature of di luent on metal e x t r a c t i o n w i t h dibenzo­18­

crown­6, 18­crown­6 and 15­crown­5 (eg £2 , j j ) . This communication dea l s with the new r e s u l t s i n the study of ternary complexes: metal ion ­ crown compound ­ organic anion and with the a n a l y t i c a l use of such complexes.

There are three c l a s s e s of ternary complex compounds i n the organic phase: 1 . contact ion p a i r s H(crown)* A n i o n

­

, 2 . so lvent separated ion p a i r s (outer­sphere complexes) , 3 . d i s s o c i a t e d ions i n s o l u t i o n . C l a s ­

s e s 1 , 2 and 3 can be di s t ingu i shed from e x t r a c t i o n e q u i l i b r i a data: the r e a c t i o n s i n t h e s e cases are descr ibed by d i f f e r e n t equat ions . When the metal concentrat ion i n the organic phase was higher than 10~^ mol/dm­>, we never observed d i s s o c i a t i o n i n CHC1­,. For a l l phenolate i ons the contact and so lvent separated ion p a i r s have d i f f e r e n t UT s p e c t r a : the di f f erence becomes more s i g n i f i c a n t i n derivated s p e c t r a . For example, in the spectrum of the second der iva t ion of T118C6 C

6H

2 ^S 0

3C F

3 ' 3 ° ~ 't n e r e a r e *"° bands at 3*6 nm (contact ion p a i r ) and

at 361 nm ( so lven t separated i on p a i r ) . A s t r u c t u r e , i n which one a n i ­

on i s d i r e c t l y coordinated by metal i o n whi le another i s separated , was proposed f o r s e v e r a l M(crown) i o n s .

Extract ion e q u i l i b r i a are described by the equat ions:

L « L Q P L = [ L ] / [ L ] , (1) M + L «г ML B L = [MbJ/[MVJL], (2) M + L Q + п А ^ Ш Л ц К е х =[MLA n\ 0/[M][L'i o[A]

n (3 ) i n some cases (eg CC1,C00H) a l s o

Н 4 Ч РНА 'ЩЫ> U )

HAQ + L Q = HA­L0 K a M . [НА­^уВДЗД, . (5)

In (1)­(5) L denotes crown compound, A counter­ion and the subscript "o" oiganic phase. Thermodynamic exiraction constants were calculated from concentration constants using Devies equation for activity coef­ficients in the aqueous phase.

4. Зак. 382 49

In contra st to wall known extraction of picrates, the use of new counter­ion carboxylates and trifluorsulfonylpnenolates aakes poaelbla extraction from strong acidic madia.

Crown ethers as well as acyclic polyethers iaprove the axtxaction of lead carboxylatas. The synergetlc effect increases for halogencar­boxyllc acids, the extraction of carboxylates with crown ethers was also effected for a number of alkali and alkaline earth metals, silver and thallium (Table 1).

Table 1 Extract ion constants of II 18C6 (СС1,С0О)п

complexes at 290 K. Concentration constants f o r 1 U LiCCljCOO

lf+ l e K

e x <

­C O X i C

^ l g K ^ ( therm.)

Ag* 1.23 + 0.05 1.8 Р ш = 0.07 T l * 2 . 5 4 + 0 . 0 3 3 . 0 Na* 0 .34 + 0.02 °­9 Kass = 8 0

K* 2.31 + 0.01 2 . 9 Kb* 1.53 + 0 .03 2 .0 "Cs(18C6) 2CCl,C00 Cs* 1.54 + 0 . 0 3 * 2 . 4 * а*г* 0 .94 + 0 . 0 2 2 .2 S r

2

* 2.98 + 0.01 4 . 2 Ba^* 2 .40 + 0 .02 3.6 * > * * 8.02 B i > 6.8

We have not found any communication on B i ( I I I ) e x t r a c t i o n by crown e thers i n c a t i o n i c form. The reason of t h i s may be the formation of Bi(OH), hydroxide In s l i g h t l y a c i d i c media. The e x t r a c t i o n system 18­crown­6 ­ CC1,C00H g i v e s us the p o s s i b i l i t y t o complete Bl e x t r a c ­

t i o n from strong a c i d i c media (pH 1­2) i n t o CHC1, or CHgClg. The f o r ­

mation of unstable complexes BiCCl.COO2

* (B = 10) i n aqueous phase and B118C6 (CCl,CO0), i n organic phase has been found.

The s u b s t i t u t i o n of H0 2 group i n p i o r i c ac id molecule by SOgCFj group increases the d i s s o c i a t i o n constant of phenole molecule ( f o r CgH^CToSOgKOH, pK = ­ 2 . 6 ) and the e x t r a c t i o n constants of c o r r e s ­

ponding complexes (Table 2 ) .

Table 2 . Extract ion constants of И Ц , complexes ( l g K B X )

A "~—­­^Щ£ CaEEG­1500 TUBG­1500 PbEEG­1500 Cs18C6 T118C6 рывсе C 6 B 2(E 0 2 ) 3 0 2

6 . 3 4 .1 7 . 6

8 . 4 11 .5

4 . 4 7 .8

6­3 9.8

11.7 13

50

Acyclic polyethylene glycolaa (JPEG) are another class of extrac­

tanta for metal Ion extract ion. The highest K f i x values were obtained for PBG­1000 and EEG­15O0 (the number of CHgCHgO groups i s 23 and 35)

The beat extraction was observed for T l + and Pb + (Table 3) . The

Table 3. Thermodynamic extraction constants of MLAg complexes at 290 К (A = 1­picrate, 2­methanyl yellow)

ь C a 2 + S r 3 + B a 2 + P b 2 +

ь 1 ] 2 1 | 2 1 | 2 1 I 2

EBG­400 FBG­600 PBG­1025 FBS­1000 PEb­1500 PEG­3000 iFBG­4000 ,BB18C6 J18C6 |Ш24С8

6 .4 5 И 6 .5

6.7

5 6.04 3.06

7 . 4 6.35 7 .6

6.65 6.7 1.1

4 . 9 6 .2 9.7 1 0 . 4 3 .98

7.8 9.2

7.95

9 .7 10 .7 6.85

7.7 9.4 a .25 9.7 6 .4 8.35 8.35 9 . 8 8 .0 8.0 8.5 6 .6 8.0 11 .8 13­2 6,31

subst i tut ion of OH groups in FBG by quinoline increases the extrac­

t i on constants of lead complexes, but in contrast t o PEG such a com­

pound extracts 3d metal ions . The extraction constants of polypropy­

lene ь­1усо1 complexes are comparatively small. The values of К for acyclic polyethers depend on dilutent much

stronger than those for aaorocyclic crown ethers . The parameters Ijj, or HP* permit one t o predict the beat di lutents i CHOI,, OHgClg, СЛЦШр. But in t h i s group of solvents wa have not found any corre­

la t ion between Sp 1 HP* and lgK e 3 c (Table 4 ) . Table 4 . Extraction constants of FbPBS­1500 Ag complexes in dif­

ferent eolvents (lgK y )

Solvent BF* «r Hethanyl yel low Ficrate

0HC1, 4 . 5 39 .1 9 . 8 8 .35 ChgC^ 4 , 0 4 1 . 1 8.9 CgH^C^ 3.5 4 1 . 9 8 .5 7.7

<W°2 4 . 3 ca 9.7 C 6 H 5 O I 2 . 4 37 .5 < 6 < 5 ­I

51

The ЛЯ values f o r PbFBS­1500(picrate) 2 and PbPBG­3000(picrate) 2

are near ly t h e same: ­ 2 1 . 1 + 1 . 0 and ­ 2 0 . 6 + 0 . 5 kcal /mol , while f o r SrPBG­150Xpicrate) 2 ДН = ­18 + 1 kca l /mol .

The e x t r a c t i o n constants depend not only on the c a v i t y s i z e of crown ethtar molecule . The complete d e s c r i p t i o n of e x t r a c t i o n process nedds an account of a l l l igand groups i n coordinat ion spheres . The correspondence between the geometric and e l e c t r o n i c s t r u c t u r e s of me­

t a l i on and i t s l i gand attachment i n ternary complex i s a reason of p e r f e c t t ctract ion i n s t e a d of "macrocyclic e f f e c t " , which cannot be used f o r .odands. I t i s c l e a r that the r o l e of counter­ ion and s o l ­

vent i n e . ­ t rac tao i l i ty and s e l e c t i v i t y may be s i g n i f i c a n t and they may change the e x t r a c t i o n p r o p e r t i e s of metal­crown ether system.

A mode J. based on analogy between f i l t e r and crown­ether molecule i n a two­phf зе organic so lvent ­water system i s proposed f o r potassium channel i l e c t i v e f i l t e r . The s e l e c t i v i t y of K

+ channel can be quali­

t a t i v e l y described by t h i s model. A s e l e c t i v e photometric method f o r potassium determination using

t r o p e o l l a 000­11 and 18­crown­6 i s proposed f o r a number of p o l y i o ­

n i c s o l u t i o n s for i n j e c t i o n s containing Ca, Na, CO?­ e t c . Tor t h e

highest s e l e c t i v i t y of photometric sodium determination w i t h 15­crown­

5 was ob+ ^iaed using to luene and xylene as d i l u t e n t s . A numt зг of photometric and atomic­absorption methods f o r l e a d ( I I )

determin cion i n water, food , a l l o y s have been developed, beau can be extractec from s l i g h t a c i d i c s o l u t i o n s i n t o CHC1­. with 18­crown­6 using llei lanyl Yellow as counter­ ion ( q = 5 * 10 ) . We can determine l e a d a f t e r e x t r a c t i o n with 18­erown­6 and CC1,C0QH d i r e c t l y i n orga­

n i c phase with FAS or a f t e r r e e x t r a c t i o n with aul foarsazen . A l l t h e s e methods were used for l ead determination i n d i f f a r e n t a l l o y s ( 0 . 0 1 ­

1 * Pb) . 3ismuth can be separated by e x t r a c t i o n i n CHC1, with 1 8 ­

crown­6 tnd CC1­,COOH from Co, Cu, Zn, Cd, I , Al e t c . Photometric and atomic­absorption methods f o r Bi determination a f t e r e x t r a c t i o n are proposed.

A num or of photometric e x t r a c t i o n methods f o r determining crown ether ( > 0 .1 ppm) and a c y c l i c polyethylene oxides (> lppm) i n v a r i ­

ous s o l v e n t s have been developed. The d i s t r i b u t i o n constants of crown compound' i n severa l systems with some d i l u t e n t s and d i f f e r e n t i o n i c background i n aqueous phase were determined by t h e s e methods.

References

1 . Taked Y. / / Top. Curr. Chem. Vol .121 , p . 1­38. 2 . AlexyJc N.P . , P y a t n i t s k i i I . V . , Nazarenko A.Y. / / ZhAKh. 1983.

V o l . 3 8 , p. 2176­2180. 3 . Nazarenko A.Y. / / Biopbysica.1984. V o l . 2 9 . P. 607.

52

THE EXTRACTION OP CARBOXYLATED COHHEXES OP METALS i У­13

1И CHKMICAL ANALYSIS I 1 А.К Charykov, S.N. Osipov, A.V. Kovenia, Lenin/jrad State Uni­versity, USSR Carbonic acids (CA) of various types ­ fatty monobasic, dicar­

bonlc (such as sebacynic), aromatic, fatty­aromatic, &­oxyfalty­aromatic and their halogenderivatives are effective for extraction of some metals, including CM , C O , M . Mn ,Col ,Zn , Ho , Si ,Sn ,nl , Ge ,8e, Pe and many others. In comparison with extragents, appertained to another classes CA distinguish one self by accesibility, low toxic properties, low^olubility in the water (for caprylic ­ C 8 , and pelargonic acids ­ C 9 , it has va­lue less then 0,2 and C,1 gramme (litre respectively), good abi­lity for regeneration (ability to act in the exclusive circles), high distrubtion coefficients for carboxylated complexes of metals (CCM), wide possibilities for variation of CA initial concentrati­ons in the dussolvents and values of pH under extraction. These properties advance CA as perspective analytical reagents for iso­lation, concentration and determination of some groups and separa­te metals from various materials and media.

CCM forming in the heterogeneous extraction systpms, have vari­ous chemical composition (normal carboxylates, "oversolvated" car­boxylates, variouametal and variousligand mixed and polymeric car­boxylates), extraction constants values К е д which beeing dependent on basic properties of metal cations, ionic strength, pH , concen­trations conditions and polarity of organic phase

We have investigated many CCU's for their subsequent using in the chemical analysis [ 1­6 ] . For example some of them are pre ­sented in the Table 1, where symbol в ЛГУ and H 2 R 2 designate mai­nly monomeric or dimeric state of CA molecules of proper monobasic acids in the dissolvent respectively C 1

­6] • Under using of undiluted liquid CA or their concentrated solu­

tions in the dissolvents mixed cation­exchange­solvate mechanism of extraction takes place, and interaction may be expressed by one of three common shemes:

M*+ m НД, = V[% ( m ­ Z/ОНД • Z H* . (1) rf\ Q, Ak% м ПХ — A6 t MRZf<v(m ­ г/г ­ »/* И KsfaVH­te)

Неге " symbol of г­charged cation and line over symbol means that it belongs to organic phase.

53

Table 1. Some characteristics of CCK, forming under extraction by benzillc (diphenyloxyacetic), caprylic ( C 5 ) and pelargo­nic ( C-. ) acids solutions

Ion Composition of aq.phase

Composition of org.phase

CCM com­position И К ex

CU2 + 0.25M UaC104

benzilic acid, iHR. ) heptanol CuR2 ­3.80+0.15

H i2 + 0.25M HaG104

benzilic acid, CHK ) heptanol HiRg ­6.40+0.25

Mn2 + 0.25 И НаОЮ 4

caprylic acid (H 2R 2)­CC1 4 MnR 2.3H 2R 2 ­11.22+0.11

Pb2 + 0.5 M №71+

0.05 M t­iSO. pelargonfc" acTc (H2R2)­CC1. Ta2PbR4. 4H 2R 2 ­13.20+0.21

Zn2 + 0.5 M HaCl +

0. О5И !%S0 4

peisrgomc acid (H 2H 2)­CCI 4 ZnR2 2H 2R 2 г­5.65+0.25

Information about stoichiometry and ^e< values allow to cho­ose conditions of the quantitative isolation of metals from aque ous phase. Thus, values of Р " 0 Э 9 , corresponding extraction de­gree E=99# under эгэрег ratio between arganic and aqueous phase volumes t * V/t/ a id at proper CA concentration may be computed as (sheme 1):

р н 0 9 9 = Н2 +

^ ­ ^К е

* ­т

¥н л ] + ^ м

) U )

d, H&- from total analytical coticen­

Here ^ - so called "side reaction" coefficient £ 1] , which ta­kes into account the whole complex formation processes of M ion in the aqueous phase, whose value is inverse regarding the part of noncomplexed on tration C M in the aqueous phase.

According obtained information about He.* and literary data for ' ^ values, we have developed some methodes tozr group separation of heavy and rare metals ( Fe, ?l, Zn, Cd, "Мл, Co, i/if <>n 1 from nature waters, including sea waters. With using of nondiluted liq id fatty acids, such as caprylic and pe­largonic vie have reached degree of concentration "t =100 at guara­nteed efficiency of extraction E^99% by single extraction, with metrex components (ions of Ax). Mq , Ca , (%• > S0 4 , HCOj )

54

regaining in the aqueous phase. This fact provide for elaboration of some straight electrothermic methodes ЛАЗ determination of Vb , if\, Cd , iWn ,Sn directly at the organic phase, free from aatrex effect and influence effects, due to other components. Detection limit (DL) for all mentioned methodes does'nt exeed 0,1 mcG/1, it's absolute value beeing near 0,1 nanoG. Relative standard devi­ation ( Sc ) for initial solutions, containing 1­10 mcG/1 variates in the range 0,07­0,15.

Another way of utili ation of CCM for chemical analysis purpo­ses me have reali ed by extraction of "various­ligand" mixed com­plexes, containing exept anions and solvating molecules of CAf so­me hydrophobic chromophoric ligands С О ­phenantraline= Phj py­

ridilazonaphtol= HFAM f morine= HMox. , <<­nitrozo­p­naphtol^HMV) On this base we have developed some selective and sensitive com­bined methodes of extraction­photometric determination of Ft*­, Си * Co » & e i n t h e nature waters. Some characteristics of dyed mixed CCM's are presented in the Table 2.

Table 2.Some characteristics of dyed various­ligand CCM's iMetal Kxtragent Cromoph.

ligand Chem.compoa. of CCM X PH 0„ (ИМ)

5H peler­ _b Pe gonic i 3­10 J

M ' acid, j Fh Pe(PhJ 3R 2­6H 2H 2 50 4.3 512 1.1 5M pelar»­. acid | H P A H

CuPANR­H2R2 25 3.2­4.6 550 2.5 ; 3M pelar­] I a c i d j ram Co(HN)2R 10 4.0 415 I 2.1

4­

These methodea are characteri ed with concentration DL, not exe­eding 1,5 mcG/1: St beeing in the range 0,05­0,1 for initial con­centrations of elements 3­10 mcG/1,

Exhaustive extraction of CCIJ's with guaranteed Б г­ Э9% affords an opportunity for preparation of the standard solutions of metals in the nonaqueous media, which have chemical composition closely imitating composition of cruide oils, oil products and other non­aqueous liquids. Such solutions have great significance for deve­loping of quantitative spectroscopic methodes. By double extrac­tion from standard aqueous solutions at proper pH values and fol­

55

lowing dilution of extracts with appropriate nonaqueous liquids w we were able to prepare the whole range of nonaqueous standards, some examples of which are presented In the Table 3.

Table 3­Examples of nonaqueous standard solutions of metals, obtained by extraction with CA

netal С m C

/ Extragent composition PH О «9

Composition of ССИ

CuZ + 0.100 2M caproic ac.

­heptanol 4.8­5.5 CuR2­ 2HR

Cu2 + 5.00

undiluted pe­larg.ac. 4.2­5.5 Cu 2R . 2HR

Fe* 2.00 2H heptanoic ac.­heptanol

3.5­4.3 NaPeR. 4

Pb2 +

0.03 2И pelargonic ac. ­ f Cfv 4.2­5.0 Ha2PbR4­ 4H 2R ?

M n2 +

0.50 undiluted cap­rylic ac. 5.5­6.0 * Д 4 . 3H 2R 2

Thus, it is possible to obtain nonaqueous solutions, containing 10 - t0'!y of different metals in the various organic matrex (hydrocarbons aromatic compounds higher alcohols and so on), which are stable at keeping and homogenous in cimposition. The stabili­ty of such samples and reproducibility of their analytical proper­ties (with coefficient of variation less, than 3% under testing during one year) have been proven by continioua control.

References 1. Djalalova P.M., Charykov Л.К./7 Vestn.LGU.1979. N22. P. 93. 2. Charykov A.K., Osipov N.N. // Jurn. Obsch. Chlm. 1982. Vol.52.

P. 2393. 3. Charykov A.K., Osipov N.N. , Panichev N.A. // Izv.vuzov. Khim,

Khlm. Techn. 1982. Vol. 25. P. 830. 4. Osipov N.N., Charykov A.K., Panichev N.A.// Jurn.Analit.Chim.

19B3. Vol. 38. P.635. 5. Charykov A.K., Alexandrova E.A.// Vestn.LCU. 1984. P. 75­6. Charykov A.K., Qsipov N.N., NemetB S.M..Koolnja A.V.// Jurn.

Analit. Chim. 1987. Vol. 42. P. 820. 7. Rlngbom A. Complexation In analytical Chemistry. New York,1963,

322 p.

56

­OuOUHdD J.IQUID IOfl SXGHANGSFT Ы ANALYTICAL СЩЫБТНГ . I I.A.Shevchuk, Donetsk Stjte University, Donetsk, USSH"

The coloured liquid ion exchangers are characterized by simpli­city of their production, selectivity and the low limit of detec­tion of some ions and biologically active organic substances. The solvent which presents the second phase is reusable without remov­al from the reaction zone. Optical density of the solvent is mea­sured both in water and organic phases depending on the substance which is being defined. The reagents widely available and cheap such as crystal violet, brilliant green, xylene orange and other compounds are used for liquid ion exchangers production. Getting mixed with complex ions these coloured substances produce compo­unds being little dissolved in water. These solid associates in the form of residue can be preserved for a long time. They achieve the colour of that dye stuff in which they are in this solution depending on the latter acidity. Thus from 6­9 m solution of hyd­rochloric acid produced by crystal violet (CVC1) and chlorstibiat­ion (A ­

) is separated in the form of an orange residue /17. The associate orange colour is conditioned by two charge form of crystal violet being formed at the high acidity of the solution. To get a coloured liquid ion exchanger 30­50 ml of water are pour­ed into the separating funnel, 0,05 g of solid ion associate is put in, 30 ml of toluene or some other solvent are added. Being mixed the solvent gets the violet colour as a result of transiti­on of two charge form of the dye stuff into one charge form. Then it is filtered through the dry paper and dissolved with the sol­vent to the necessary optic density which is not changed if pre­served in a dark place* It should be noted that the dry solid associate without water is not dissolved in toluene. The associ­ate may undergo the exchange reactions as well as acidity­reduc­tion ones producing crystal violet without application of any or­ganic solvents; the optic density of water solution permits to judge «bout the concentration of the substance which is being de­fined. The latter is introduced into one of the phases depending зл its solubility. The choice of the phase for measuring the op­tic density depends on the existence of some other coloured sub­stances; in caae of the absence of those mentioned above the optic density measurement can be carried out in any of the phases. When the water phase is either dyed or turbid the optic density measu­rement is carried out in the organic phase.

The coloured ion exchangers nay undergo the reactions of exchan­ge and acidity­reduction. In the exchange reactions cation of the

57

organic dye stuff ia able to be substituted for colourless cations or coloured organic cations of various nature. The substitution degree depends on the nature of organic cation, its structure, ba­sicity, hydrophobic, resistance and solubility of the compounds which are being formed. These and other factors define selectivity of pointing out various organic cations with the help of coloured liquid ion exchangers.

The acidity­reduction reaction of coloured liquid ion exchangers with various substances has been studied by the author of the paper together with Oekhtyarenko L.I. and Yankovskaya E.V. The detection limit depends on the substance nature.

Table 1. The organic substances detection As far as sensiti­limit in the reaction with OVA vity is concerned

the new method has some advantages comparing with those already existing. Thus while defining cysteine with СТА the detection limit is one order of magnitude below (lower) than by the

modified nitroferricyanide and phenantroline methods and much lower than by ninhydric method. Sergeyev et al. have studied the selective extraction­photometry defining some phenols, amines, aminoacids and heterocyclic with nitrogen compounds by means of CVA/y.

Coloured liquid ion exchangers are produced as a result of the interaction of colourless organic cations and coloured complex compounds of the type Fe(SCN)J, Co(SCN)|~ and the like as well. When (R,lffi)2Co(SCH)^ coapounds witb phosphate ions are added to associate solution the decrease of optic density of solution is observed. Sulphates, chlorides, fluorides also affect the colour associate (S,NH)2Co(SCM)^ but to the lesser degree than the phoephateB.

The coloured liquid ion exchangers have been studied on the ba­sis of the associates with alkylammonium cation and intra­complex compound of cation of the metal with xylene orange [S. Carrying out these reactions in the solutions of bicarbonate in presence of alkylamineu salts allows to reduce the limit of detection of me­talls ions and to increase reaction contrastivity due to reextrac­tion of surplus amount of xylene orange when exchanging it to Ы ­

The substance which is being defined

Detection limit mkg/ml

cysteine 0,16 glutatione 0,25 pyrocatechol 0,06 hydroquinone 0,15 resorcinol 0,5*

58

karbonate­ions. The coloured associates resistance depends on the nature of colourleae organic cation. More resistant coloured asso­ciates react to the presence of mineral salts ­nich as chlorides and sulphates of sodium to a lesser degree (Table 2).

Table 2. Dependence of optical density of associates produced by yttrium (III) and xylene orange with

organic cations [KHCOJ] ­0,2 mol 0,1 И chloride salt of reagent in the Optical density

J mixture CCl^ + propanole in salts in salts absence presence

cetylpyridine 0,99 0,38 alkyltrioethylammonium (

C

TQ­^TQ) 0,90 0,08 trialkylbensilamnonium (0»­C„) 0,97 0,05 dode cylammonium 0,98 0,90 the primary alkylainines fraction ( C

JQ­CJ 2) 0,9? 0,90 dimethyldodecylanBODium 0,96 0,63 dinethylalkylammoniua (Cj^­Cjg) 0,99 0,92

The coloured associates undergo the reactions of exchange and acidity­reduction with ions and organic compounds at different phases: liquid опев not mixing, the solid phase ­ water, ae well as in the nicellar solutions where the micelles act as microphase.

References 1. Shevchuk I.A., Louis J.^Heports thesis at the Hldconference

on extraction, Donetsk, 1987. P. 32. 2. Sergeyev G.I., Korenmann I.M., Tikhomirova S.V.//Journal of

Analytical Chemistry. 1985. Vol. #0. P. 154. 3. Shevchuk I.A., Konovalenko L.I., Sinonova Т.К.//

XII Hendeleyev Congress in General and Applied Chemistry. Re­ports and informations, n 1. Analytical Chemistry, 1981. P. 329.

59

BIFUlfCTIOlTAL AMMONIUM COMPOUHDS AS РНОМЕПГС BXHUGHTIS j 9­15

OP METALS I B.k.Hakhmanjko,S.V.Polishchux,G.A.TsvIrko,G.L.Starobinets,S.M.Leshchev Research Institute of Physical­Chemical Problems, Byelorussian State University, Minsk, USSR

It is known that bibasic extragents of anions, namely, dialkyl tin salts [l], bis­quaternary phosphonium salts [2], exhibit pronounced bivalent anion affinity. We have investigated as metal eztragents the salts of dinonylaminoethyl­^ ­trinonylammonium (ДЫАЬ­ji -ISA) and methy pentadecylethylenediammonium (biPUEDA). „

[|j>­CH4­CHe­feg] I " [!­>W­CHa­CH8­W(R J2I" ШАЬ-fi ­THA iodide BPDEDA diiodide

The above compounds have been synthesized by stepwise alkylation of ethylenediamine and purified in the octane­dimethylformamide extractic system. Extraction of cadmium bromide and iodide complexes by MPDEDA picrate in toluene and its binary mixtures with chloroform, amylaseta­te and octane by the intermediate exchange method has been investiga­ted. For comparison with monofunctlonal QAS, cadmium extraction by tri nonyloctadecylammonium (MJODA) has been studied under similar conditio

The cher oal analysis of extracts, the analysis of bilogarithmic de< pendences of cadmium distribution coefficients on a reverse coefficien of picric acid distribution as well as the comparison of the dependen­ces of side reactions coefficients, calculated from tabular instabili­ I ty constants of cadmium iodides and bromides complexes, and tentative " exchange constants on concentration of iodide and bromide ions hare re | vealad that mono­QAS and Ыв­QAS extract cadmium as a bivalent parti­ ' cle Сс1Г ч

а

­ . In this case equilibria for HPDEDA in the absence of self­associa­

tion in the organic phase may be described by the following reactions: (1) (2)

:

< * ( 3 )

Analysis of a great bulls: of data (about 100 experimental points) at concentrations ranged from 10"' to 10"^ for bis­QAS and from 10" to У 10 m/1 for cadmium has shown that with an anionexchanger being fil­led with cadmium up to 30%, the extraction equilibrium is described by Eq.(2), in the case of cadmium exoeeding 50% Eq.(3) is employed, while :

Bq.U) is not applicable. For cadmium percentage being lese 50%, the value of K. (Table 1) is estimated. Constant values of K, are indica­tive of the extraction process described by Eqs.(2,3). From these data it follows that the tentative exchange constants both for iodide and .j,

t

60

bromide complexes axe higher by 2­3 order of a magnitude than those for aono­vAS which implies the increased synthesized anion­exchanger affi­nity of cadmium halogeniae complexes.

.,» Pic" Table 1. Tentative exchange constants Kcdlq lor МРП6ЙА

and THODA der ived fo r d i f f e r e n t cadmium c o n c e n t r a t i o n s

U P D E D A I H 0 D A Cum KPIC­ CuUI fi.Pit Cc*[U can) Jip;c 2 . 1 0 " ^ 6 ­ 7 ­ 1 0 2 3 ­ 1 0 ­ 4

0.73 2 . 1 0 ­ * 3.56 8 . 1 0 ­ 3 1 . 0 . 1 0 " 3

4 ­ 1 0 " 5 7 . 3 Ч 0 2 1 . 1 0 ­ 3 0.68 6 ­ 1 0 ­ 4

3.05 4 ­ 1 0 ­ 3 1 . 1 . Ю ­ 3

2 ­ 1 0 " 4 2 . 5 ­ 1 0 3 3 . 1 0 ­ 3

0.64 8 . 1 0 ~ 4 2.9 1­10"* 1 . 1 ­ Ю ­ 3

4 ­ 1 0 " 4 3 . 2 . 1 0 3 1 . 1 0 ­ 2

0.59 1 .10" 3 2.67 3 ­ i o " ^ 1 . 1 . Ю ­ 3

2 ­ 1 0 " 3 2 . 1 ­ 1 0 3 5 ­ 1 0 " 2 0.36 4 ­ 1 0 ­ 2

4 .6 5 ­ Ю ­ 2 1 .1 ­10" 3

R

\ + /

Л'/ Чне­ОК' N

Rj 54* te

At first it has been expected that the elevated MPDEDA affinity of Ufq is att­ributed to the formation of ionic chelate likely to that described by Eq.(1): " However, the fact that extraction is better described by bqs.(2,3) su­ggests a more complicated process of the extraction. It is most likely that extrac­ ­. ­tion strengthening is a result of forming '1С a particle:

With the ItPDEDA extraction proceeding in an organic phase, the following reactions may also take place: rihR'sNaCdr*'^r[&»i#sCt/rj'Jn u ) SfaKs&PicbCc/r,*-у­й//^"'^r ^ A r " / f o u ' ^ ' # / y

?

7 w i W i ^ < 5 )

4[RiRs^h^'^^Cd/^-^[^^2^lk(R,lik)i(Cdr4)fl + 6 Pic' (6) Ionic cyclic oligomers, products of the reaction presented by Eq.(4), and linear oligomers Eqs.(5,6) are likely to exist at high extragent concentrations that is undireotly evidenced by an abrupt viscosity in­crease of 0.05 m/1 bie­QAS in toluene with I"­ions substituted forCJT4

observed by the authors. The K­ values of cadmium iodide complexes ex­ceed those of bromide complexes by 2­3 oxderr of an magnitude. Fig.1 presents a bilogarithmic dependence of tentative exchange constants on concentration of iodide ions. A similar dependence with a mnTlnnim at the 0.2 m/1 concentration of bromide ions is also obtaned for bromide complexes. Maxima on the curves axe explained by the fact that at the given ligand concentrations a maximum number of the complex forms

our which possesses the highest extractant affinity. A study of the orga­nic solvent effect on the liquid anion­exohanger affinity of the cad­mium halogenide oomplexes has revealed that the cadmium extraotability with transition fxoa toluene to the mixtures toluene­octane, increases. This may be result of stronger solvation of Cic~­anions as compared to

61

Сс1Гц ­anions with decreasing a toluene content in the mixture that mat;s their anion­exchanger affinity different. Chloroform, amylaceta­te a­.ded to toluene affects only slightly the extractibility of cad­mium halogenide complexes by AiPUbDA salts.

Dinonylaminoethyl­£ ­trinonylam­monium iodide iB an intermediate sub­stance in the liPDEDA synthesis. A specific feature of DHAE­JJ ­THA, ач compared to monofucctional QAS, is their high solubility in inert ali­phatic hydrocarbons as well aa pos­sibility to control the number of io­nogenic sites in an extractant mole­cule by pH variation.

The extraction of halogenide com­plexes of mercury (II) from alkali media by m«'i­j! ­THA salts has been examined by the intermediate exchan­ge method. A study has also been ma­de of the extraction by HJODA salts. In both cases the extraction haloge­nide complexes of mercury (II) pro­ceeds as the anion exchange reaction:

where Й ­ CgH 1 g. ^ ^ for ШАЬ­_р ­THA .

The nature of substituents at the tetrasubstituted nitrogen atom only slightly influences the extractiuj.Iity of halogenide complexes mercu­ry (II) by ТН0ДА and ШАЗ-f-WA salts, with polar solvents (alcohol, chloroform, toluene) being used. However, stronger branching in the DHAE­.fi ­THA molecule and a greater size of its radical initiate an abrupt increase of there solubility in inert aliphatic hydrocarbons as compared to monofunctional QAS to yield a pronounced analytical effect. Table 2 summarises the values of к' „. obtained for

Fig.1. The bilogarithmic de­pendence of tentative К е х с Ь

on iodide ion concentration for HPDEDA in toluene

A1

for THODA and R* ­ (CgH,^

Table 2. summarises the values of K e z o i l 4[!*

E

'xchCt* values obtained for the salts of HiODA in toluene and ШАЕ-fi ­TNA in decane

A n i o n I U 0 С А D В A E­ p ­T 5 A A n i o n toluene dacan» toluene decane

HgCt»

Нч1\

1.2.10 5

3.2.10 7

9.8.10 9

prac t ica l ly undissolved

5.0.10 5

9.8.10 7

1.0.10 1 0

1.0.10 b

3.4«10 8

1.8.10 1 2

62

the salts of ШГСША and ОЫАЬ­JJ ­ТКЛ in toluene and decane. л use of deeane solutions of DHAt­^­ТИА salts instead of toluene solutions of TNOtfA salts causes a pronounced increase of the extractibility of the mercury (11) halogenide complexes that may be due to decreased ex­penditures of energy in resolvation during the exchange process. But the expenditures of energy decrease with strengthening of the hydropho­bic nature of Hjfj ­ion that gives rise to different anion­exchanger affinities (Table 2).

The extractibility of halogenide complexes of mercury (II) is essen­tially affected by the nature and concentration of a ligand. Fig. 2 de­picts a bilogarithmic dependence of tentative exchange constants of K T on concentration of iodide ions. Similar dependences with maxima at chloride and bromide ions concentre­ ,H j. tions equal to 10"1 and 2.10"2 m/1 re­ Ц К р' spectively, have been obtaned in the case of extracted chloride and bromi­de complexes of mercury (II?­ Maxima on the curves are attributed to the fact that at these ligand concentra­tions mercury (II) is in the form of Ц9Г5 ­anions which exhibits the hig­ ' } | 3 у pj hest extractant affinity. A maximum Fig.2. The bilogarithmic de­shift towards highest ligand concent­ pendence of tentative K

e x o n

M

l £ rations in the series I"< Br <• Ct" is on iodide ion concentration due to weakening of the strength of for DNAE­jJ ­TNA in decane extracted №%Г£ ­complexes in this se­ries. The obtained data are correlated well with tabular K^^^ values of the corresponding halogenide complexes of mercury (II). An increase in the K'($­3 value, with transition from chloride to iodide complexes of mercury (II) is explained by increasing the size and strengthening hydrophobic nature of a metal complex anion, as a ligand size increases.

Thus, application of new bifunctional ammonium extragents, i.e. KPDbDA and ШАЕ­£ ­TNA improve essentially the selectivity of extrac­tion of halogenide complexes of mercury and cadmium that may be employ­ed for developing the highly effective methods of extraction, concen­trating and determination of the given metals. It may be presumed that the synthesized extragents will possess the raised affinity to other bivalent anions. The authors possese data wich are indicative of eleva­ted affinity of SO4" andWOq'­ions to bis­QAS.

nelerences 1. Spivakov B.Y., Shklnev V.M., Vorcb'e\a G.A., Zolotov Xa.A. // Ext­

raction Chemistry, Hovosiblrsk:Hauka publ., 1984. P.132­142. 2. Akira 0., Kunihiko 13., Makoto T. //Bunseky kagaki. 1984. Vol. 33,

I 6. P. 187­194.

63

THE SOLVEHT EXTRA.CTIOH OF DIAIXICLDITHIOPHOSPHATBS g _ l 6

IN АЯЛЬТИСАЪ СННПЗКГС

I .P .Al imar in , T.V.Rodionovn, V.H.Ivanov and L.V.Karavaeva, Moscow S t a t e U n i v e r s i t y , Moscow, USSR

Dialkyldl th iophospheric a d d s (HO) 2PSSH are widely used as rea ­

gents for the solvent extraction of Inorganic metal sa l t s . The valu­

es , characterizing the process of extraction ( constants of extrac­

tion, partition constants and stabi l i ty constants of complexes) give a possibil i ty to use these reagents for the separation and concent­

ration of many metal ions. For example di­2­ethylhexyldithiophospho­

ric acid D2EHDTPH has been applied for the separation of Bi(III ) , Cu ( I I ) . T1(I), Ag(I), As(III) f l . 2 ] .

However this i s not a good selective reagent since i t forms very stable complexes with many metal ions that are extracted from aqueo­

us solutions under equal conditions. Therefore the procedure of se­

paration of metal ions by D2EHDTPH i s too complex. So separate the metals more suitable to apply derivatives of dialkyldithiophosphoric acids with a small hydrocarbon chain, for example DEDPTH of DBDTPH.

The comparative extraction of Hg(II) by DEDTPH and DBDTPH has be­ ;

en studied using radiometric method. Carbon tetrachloride was used as organic solvent. The results of the extraction of Hg(II) as a fun­

ction of hydrochloric and sulfuric acids concentration indicate that DEDTPH i s much more effective extractant for Hg(II). This reagent qu­

antitatively extracts Hg(II) in more wide region of concentration of ; HC1 and HgSO. (Table ) . The different behavior in distribution of Hg (

(II) dialkyldithiophosphates in the range from 0 to 7 i s probably ex­ I plained by different stabi l i ty of these complexes and the fonaation of negatively charged complexes HgL ~ , where L ­ anion of diallcyl­

dlthiophosphate. It has been established that the molar ratio Ue:L in the extractable species i s 1:2 when both extractants are used.

The extraction conditions of Hg(II), Cd(II) and Zn(II)

The metal ion (II) DBDTPH DBDTPH

Hg 8 M H 2 S 0 4 ­ pH 7 6 H HgSC^ ­ pH 2

5 И Н01 ­ pH 7 3 И НС1 ­ pH 2 3 , 0 ­ 0 , 3 И HOI Г 3 ]

Cd pH 2 ­ 6 pH 1 ­ 7 Zn no extract pH 2 ­ 4

0 , 0 1 ­ 0 , 1 2 И НС1 £ 3 ]

64

The d i s t r i b u t i o n r a t i o of Hg(II) was measured as a funct ion of Hg ( I I ) concentrat ion i n aqueous phase at constant concentrat ion of r e ­

agents and HOI or HgSC^ . The e f f e c t of Hg(II) concentrat ion was i n ­

v e s t i g a t e d by varying the concentrat ion of metal from 1 .10~7 to

2 . 1 0 " ' m o l / 1 . The d i s t r i b u t i o n c o e f f i c i e n t s of Hg(II) using as e x t r a ­

ctant are i n dependent on the concentrat ion of Hg(II) from S U Ho SO. (6 II HC1) to pH 7 . «hen Hg(H) i s extracted by DBDTPH from hydrochlo­

r i c and sulphuric a c i d s o l u t i o n s ( 6 , 0 ­ 1 , 0 ) the d i s t r i b u t i o n r a t i o i n ­

creases and i s constant i f the metal i s extracted from the aqueous phaee i n t h e pH r e g i o n from 0 t o 4 . Probably the e x t r a c t a b l e compound a s s o c i a t e s i n organic phase.

The degree of a s s o c i a t i o n of Hg(II) dibuthyldithiophcsphate i n o r ­

ganic phase has been determined using measurement of d i s t r i b u t i o n ra­

t i o as funct ion of Hg(II) concentrat ion i n aqueous phase. The s lope of the dependence l o g D J ^ ^ J ­ l og

c

j j g ( 0 ) i s about 4 corresponding to the complex (Hgl^). when Hg(II) i s extracted by DBDTPH from sulphuric a c i d ( 6 , 0 ­ 0 ,5 N ) . The a s s o c i a t i o n of Hg(II) dibuthyldithiophosphate s t a r t s when the HgClI) concentrat ion i n aqueous phase i s 1.10** mol/1 ь

The a s s o c i a t i o n of Hg(II) complexes i n organic phase i n the range pH from 0 to 4 or by DBDTPH from 8 M HgSO. to pH 7.

In reg ion of high a c i d i t i e s the e x t r a c t i o n of Hg(II) may formally be described by the heterogeneous r e a c t i o n

H R 2 + + z m ^ HgLg + г н* (DEDTPH)

4 H g2 + + 8 H L ^ ( H g I 2 ) 4 +8 H

+ (DBDTPH)

The d i s t r i b u t i o n of Od(l l ) by the reagents under i n v e s t i g a t i o n i s a funct ion of the hydrogen i o n s and reagents concentrat ion . For the quant i ta t ive e x t r a c t i o n of Cd(II) a large excess of reagents i s requ­

r e d . The complexation of Cd(II) with DBDTPH and DBDTPH has been s t u ­

died us ing the d i s t r i b u t i o n method. The r e s u l t s of the i n v e s t i g a t i o n are g iven i n the t a b l e . In parenthes i s the data for DEDTPH are g iven: l o g K e x ­ 5 , 3 ( 4 , 1 ) , l o g J ^ 2 ­ 10,6 ( 5 , 3 ) , l o g ^ ­ 3 , 6 2 , l o g ^ o ­

9 , 7 . In the pH range of 0 ­ 4 the dibuthyldithiophosphate of cadmium i s extrac ted corresponding to the equation ;

C d2 + + 2 H X ^

CdL

Z(0) + 2 H +

The data f o r the e x t r a c t i o n of Zn(II) by the DBDTPH correspond to those report by \_4~\­ These authors give (the present r e s u l t s are g i ­

ven « i th in brackets ) s l og K e x ­ 1,67 ( 1 , 9 0 ) , l o g A 2 e 2 " 6 . 0 5 ( 6 , 1 0 ) ,

Ю в Л г ­ 5,37 ( 4 , 0 3 ) , 1 ° в 4 г " 2 , 2 ? (

2

'3 8

> ­

On the basis of the experimental d»ta too methods of the separati­on of He(II), Cd(II) and Zn(II) by DBDTPH were developed (Figure ).

65

M e t h o d 1

HB, Zn, Cd . extract ion from 1 И HgSO.

~ ^ 1 e x t r a c t i o n froa pH 2 - 4

aqueous phase \ ex trac t

Zn Cd

M e t h o d

Hg, Zn, Cd

extra c t i o n from pH 2 - 4

~~~--^^ aqueous phase 0 rganlc phase ^ ^ - ^ ^

c t i o n from pH 2 - 4

~~~--^^ aqueous phase

Hg, Cd Zn

-e e x t r a c t i o n by 1 H HC1

organic : pat N sequeous phase

^ Cd

Methods of the separation of Hg(II), Zn(XI) and Cd(II) by dietbyldithiophosphoric acid

The method of the separation of Hg(II) trom Cd(II) and Zn(II) was exanined radiometrically. The results are satisfactory, s p does not exceed 0,01 •

References 1. Levin I.S., Sergeeva V.V.,.Tiachenko L.Y.// Zhurn.analyt.Chom.

1973. V.28, H 5. P. 962. 2. Levin I.S., Sergeeva V.V. et al. // Int.3olv.Bxtr.Conf.London.

1974. P. 2137. 3 . Handley Т.Н., Pean J.4. , / / Analyt.Chem. 1962. V. 3 4 , H 10. P.1312. 4 . Wingefors S . , Rydberg J . / / Acta chem.Scand. 1980. V. 3 4 4 , N 5 .

P. 3 1 3 .

66

9­17 USB OP АШОЯПЩ DI­(2­BTHYIfiBirL)DITHIOPHOSPHATE FOR BDRACTIO» 0? ALKALHTB BARTH ELEMENTS DURING DETERIUNATIOH OP Li. Rb, Cs IH BRUTES Yu.H.SaBollov, S.S.Shatskaya, Institute of Solid State Chemistry and Processing of Mineral Raw Materials, Siberian Branch of the USSR Academy of Sciences, Novosibirsk, USSR To decrease the detection limit of flame photometry and atomic ab­

sorption determination of Li, Rb, Cs in calcium chloride brines and products of their treatment one should separate the general interfe­rence components Ca and Sr, e.g. by extraction.

An extractant and a diluent for the given systems are chosen such that high distribution ratios (D) for Ca and Sr are obtained at the lowest D's for Li, Rb and Св.

Tho study of extraction of alkali and alkaline earth elements by di­(2­ethylhexyl) phosphoric (DEHP) and d4­(2­ethyli.exyl) dithiophos­phorlc (DEBWP) acids as well as by ammonium salts showed that the use of ammonium di­^­ethylhexyDdithiophosphatedtH.R) with tributyl phosphate (TBP) provides the best separation of Ca, Sr and Mg from solutions [l.3]. Fig.1 shows the maximum extraction of alkaline earth metals at pH'5 obtained by neutralization of the aqueous pha­se with ammonium hydrate, the extractant being partially or complete­ly in the organic phase as a univalent cation.

Ammonium salt is strongly associ­ated in saturated hydrocarbons in contrast to polar solvents [1 J. The­refore, the extraction of metals as a function of solvent may be descri­bed as follows: Por saturated hydrocarbons

a 100

l l en + + (UH 4 R) p — M e T O 4 ( p _ n ) H p +nHH 4

+ , Я

1 M M 3+MeJ ЛЗ^ Г-*

where p is a degree of extractant as­sociation, 1 is a number of reaction centres occupied by one molecule of metal. Por polar solvents (alcohols, TBP)

pH

,1"),

Me" Jorg.

"«•МеВд + nHH 4 i

[т/У Ц»* [HH4Rf "SS?^

(2)

S 10 Fig.1. Extraction of Sr(1,1 Ca(2,2',2») and Li(3,3',3") in 1 1Ш.С1 with 1 К БВНШР into

4 ootanol (1­3) and heptane (1*­3') as a function of pH. 0.5 M (DEHDTP­TBP) ­ heptane (1"­3") . C,, a, g / 1 : Ca ­ 7.4, Sr ­ 1.0, Li ­ 0.1

67

The effective extraction constants (Table 1) have been calculated according to the equation (1) or (2).

Table 1. Effective extraction constants of metals • 1

System

1 о в К

в 1 • l o e K e x 2

• 1

System Li ' Hb Os Mg Ca Sr

DEHP­NH.­4

­ heptane

0 .07 2.22 -

- 0.06 • 1.46 i i

- 0.03 i DEHDTP­HH4­

­ heptane ­ 0 . 7 2 -

- 0.10 ­0.22 * o.ia i i cos i o.o8

0.4 ­ 0.64 -

. - 0.05 - 0.10 • 0.67 i j

i 0.04 D5HKTP­HH.­

­ d c t a n o l ­ 0 . 4 8 i-

* 0.02

DEHDTF­HH.­

­ TBP i .6 i

\- 0.06 ­ 1 — 1 i

The extraction of metals by heptane solutions of HH.R decreases in the sequence Cs >Rb >Li ­Ha ~K.

The use of ammonium salts has some advantages as compared to the acidic form. Salts are, as a rule, more stab­le, require no pH regulation, provide good lamination arid practically do not i change phase volume. The cation ex­change of ИН. for alkaline earth ele­ments with a further HH.Cl sublimation allows the salt background of a brine to be decreased 'by 95­97$ in one stage of extraction.

Fig.2. Distribution ratio of CaO), Sr(2), Mg(3), Li(4), Rb(5), Cs(6) as a function of the БНШТР­НН. concentra­tion in the mixture with TBP in heptane (the broken lines are the same but in the ab­sence of TBP). The total con­centration of JJH й and IBP is equal to 1 M

Heutral additives of TBP result in i higher capacity of extractant in rela­

2+ tion to Me , alightly increase Xi ex­traction and sharply suppress Eb and C. extraction. Fig.2 shows the influence of extractant composition (mixture DEHDTP­m4­TBP) upon extraction of me­tals. „

Factors of synergism S •*°е­Т^Щ versus the nature of extracted metal3 and HH.RiTBP ratio are given in Table 2.

68

Table 2. Factors of synergism

S " 1 о

в Dmix ­ 1 о

в D1>

D2 ** °

\^H.H:TBP H e

n + \ ^ 1:4 | 2:3 ! 3:2 ; 4:1

СаУ +

Li+

Rb+

=s+

0.98 0.05 ­0.38 ­0.57

0.89 0.06

­0.24 ­0.30

0.51 0.04 ­0.13 ­0.15

0.18 0 ­0.05 ­0.05

Synergetlc effect has the maximum for the extractant composition HH R:TBP ш 1:4. The absolute value S decreases in the sequence Ca^Sr^ Mg>L±.

The obtained data allowed the optimum conditions to be chosen to separate the CaCl;, base from Id, ЛЬ, Сз with a further flame photometry determination of Li and atomic absorption detection of Eb and Cs in the aqueous phase.

The 0.5 II mixture of ammonium dialkyl dithiophosphate with TBP has been used for matrix breaking during lithium determination. In the given systems, one can observe the minimum distribution ratios inde­pendent on its concentration within 0.1­10 ppm and inversely propor­tional to the ammonia ion activity in the aqueous phase as well as the maximum separation coefficients j*He

2+

/i,i+ (e

­S­ 1°8 £ca/Li i a

equal to 1.8, 1.4 and 1.15 for calcium, strontium and magnesium, res­pectively) .

The use of equimolar (0.5 M) NH.H mixture with TBP in heptane dur­ing one stage of extraction at v

o r K

!

,v

a q c 3:1 allows the contents of

Ca C7.6 g/1) and Sr (0.32 g/1) in the Li­bearing brine to be decre­ased to 0.2 g/1 of Ca and 0.02 g/1 of Sr which do not prevent from determination of Li. Besides that, in the presence of Ca and Sr the value D L i is constant and equal to 0.19 in a wide concentration in­terval of Li in the aqueous phase, which Indicates the absence of co­: extraction in the given systems.

Ammonia does not wash calcium from an organic phase up to 5.4 M icontent (pH 11.4). A calcium reextraction can be observed at HH.C1 .< concentration more than C 4 It. The optimum conditions for washing J the rest of lithium in organic phase are as follows: 0.4 M NH.C1 and J1 It HH,, v

o r ~ : Y a a " 1=2. The extraction cleaning decreases the de­

tection limit of lithium in calcium chloride brines by an order and .provides detection of Li in strontium compounds. Relative standard |deviations of lithium determination do not exceed 0.05.

j 69

The use of HH.R mixture (0.5 M) and TBP (2 П) in heptane at У : :V • 2:1 decreases Ca concentration in brine from 8 g/1 to 0.4 g/1, aq

D R b and Dp being equal to 0.16 and 0.2 and constant as metal concent­ration being changed from 10 to 10 g/1.

The substitution of ННдС1 for the CaCLg base with a further HH.C1 sublimation allows alkali elements to be concentrated and alkaline earth element effect to be removed.

Л technique of extraction­atom­absorption determination of Rb and Cs in calcium chloride brines with general salt background equal to 400 g/1 for Hb and Cs contents (0.5 ­ 3).10~3 g/1 with a relative standard deviation less than 0.07 has been developed.

References 1. Samoilov Yu.ll., Yukhin Yu.H., Shatskaya S.S. // Zh. Neorg. KMm.

1985. Vol.30. P.2053. 2. Shatskaya S.S., Samoilov Yu.M. // Zh. Anal. Khim. 1987. Vol.43.

P.272.

70

*

CYCLIC THIOUREA DERIVATIVES ­ 1TOVEL REAGEBTS TOR THE CELECTIVE ISOLATIO» AID DETBRHUtATIOH OF BISMUTH 9­18

V.A.Hazarenko, I.S.Preanyak, E.I.Shellkhlna, V.P.Antonorloh, Bogmtrty Phjrwioo­Cheelcal Institute, Acadeay of Sciences of the Ukrainian SSR, Odessa, USSR To develop the selective procedures for the isolation and extrac­

tion ­phot ome trie determination of the JUS quantities of bismuth, the extraction of bismuth complexes with the cyclic thiourea derivatives (CTD) and halogenide ions from the sulphuric acid media was studied. In the presence of the excess of the completing agents the formation of the extractable complexes Bl­CID­A" was observed, where A~«C1" Br", I". The following CTD were studied: / V yCHj

HN NH S

Ethylene thiourea (EIU)

HN L s

Trimethy­

lene thio­urea (TTU)

HjC­N. .N­CH, 8

H N 4 . N H

N.ltf­dimethyl­

ethylene thio­urea (DMETtO

Bethyltri­

methylene thiourea (MTTU)

thylene thiourea (TUTU)

The optimal conditions for the complexation and extraction were found using the method of the experimental design of Box­Wilson.

Under these conditions the spec­ j) trophotometric characteristics of the extractable complexes were de­termined. The data obtained are diven in Table 1.

The absorption spectra of the complexes don't practically depend on the CTD nature, but depend con­siderably on the nature of a mono­basic acid anion (Fig.). The de­termination of the basicity cons­tants of the studied compounds (Ta­ble 2) and their distribution coef­ficients between the aqueous and

20 V 1000cm" 400 450 500 X nm

The absorption spectra of organic phased (Table 3) allowed to the bismuth complexes with tet­calculate the conditional constants ramethylene thiourea extracted of the bismuth extraction (Table 40» by chloroform in the presence

The extraction constants decre­ase regularly in the series I> Br> Cl~ which proves the empirically found phenomenon of the preferenti­

of the anions. Aqueous phase: 1 • 10 —

TI Bi; 0.005 M TJITO; 0,5 M HgSO^ 1. 0,15 И Cl"j 2. 0,01 M B­*" 3. 0.001 M I 1 » 1 cm

71

Table 1. Conditions of the Formation photometric Characteristics Cyclic Thiourea Derivatives

and Extraction and Spectro­of Bismuth Complexes with

System tH 2 so 4 ] , H [CTD] ,И [A"]»M A.nm f ­10" 4

S o l v e n t , e x t r a c t i o n Г Я cm Anion

tH 2 so 4 ] , H [CTD] ,И [A"]»M A.nm f ­10" 4

S o l v e n t , e x t r a c t i o n Г Я

ЕГО CI" Br" I "

3 . 0 ­ 3 . 5 5.25­3.0

0.6

0.05 0.08 0.05

0.15 0.08 0.004

395 422 480 1.0

MIBK, 0 0 % MIBK, -x 85? СНС1^,=90%

ITU Br" I "

0 .5 1.5­4 .0

0.04 0.02

0.15 0.005

428 485

'1.3 1.4

CHC13,>95!& CHC13,>95%

или CI" Br" I "

0.5 0.5 0.5

0.03 0.006 0.006

0.15 0.01 0.001

405 430 480

1.2 0 .8 1.1

СНС13,>95Й CHC13,>95% СНС13,>95?°

DMETI Br" I " I "

3 .3 1.5­5 .0

1.5

0.05 0.003 0.014

0.02 0.003 0.004

430 485 492

1.3 0.75 0.9

to luene ,?95Й to luene ,>95S CHC1 3, >9555

итти Cl~ Br" I "

7.4 1.5

0 . 3 ­ 3 . 5

0.03 0.02 0.01

0.23 0.04 0.002

403 430 482

1.7 1.5 •T.3

CHC1 3,>95# CHC13>>95% CHC13;>95%

lahle 2. The Basicity Constants (Кщ+) of Some Cyclic Thioureas

^4 [CTDH

+

] CTD ETU ITU TMTU DMETU MTTU

K B H + 117 13.5 5.6 676 8.9

Table 3. Coefficients of CTD Distribution (E) between the Aqueous and Organic Phases (E •» 0 /С )

^ ^ O r g a n i c

0 T I ) < D h a . s e Toluene c c i 4

1 , 2 ­ d l ­c h l o r o ­e thane

CHClj Ethy l a c e t a t e

Butyl a l c o h o l ­1

MIBK

E»U 0.03 0.0 0.17 Q.17 0.30 0.83 0.20 ITU 0.21 0.13 0.21 0.45 0.50 0.80 0.29 M1TU 0.33 0.0 1.72 3.54 1.43 3.12 1.12 DMHTO 40 25 320 550 ­ 48 61 WTTU 0.07 0.05 0.66 Г.Э5 0.40 1.57 0.49

72

Table 4. Logariphms of the extraction constants (lg Kfi) of bismuth complexes with the cyclic thioureas and halogenide­ions with chloroform

(1 « 1 ­У / К

вн­] ^ °

C O m (°В1­С

со Ш»С

А­ ­3 C

com^ <C

CTD­20

com>*

4 \. ETU rra TMTU TMETU HTTU

CI" * 7.9 _ 7.6 _ 7.5

Br" 8.9* 9.9 12.5 10.2** 10.9 I" I~

13.1 14.3 14.9 13.5 **

15.1 15.7

•ft

extraction with М1БК; extraction with toluene.

al extraction of iodide­thiouxea complexes and corresponds to the greater strength of the bismuth iodide complexes.

The maxim] values of the extraction constants were obtained for СТО possessing the largest cycle size (seven­membered for THTU) or the most carbon atoms in the radical at thioamide fragment in the studied series which may ue due to the increase of the ligand hydro­phoby and, hence, of the bismuth complex. Introducing the methyl groups at the carbon atoms, and, especially, at the nitrogen atoms, results in the increase of the CTD molecule hydrophoby, and in the corresponding increase of the extraction constants values. In case of H,N­dimethyl derivatives of DMETU the ligand hydrophoby increases so, that the complex extraction by the low­polar solvent toluene becomes possible with the high value of the extraction constant.

It should be noted that for the bismuth extraction CTD are the most suitable as well as the solvents where the moderate distribution of CTD between the aqueous and organic phases (E • 1*50) is realized.

: This is due to the fact that the formation of the complex Bi­CTD­A~ takes place in the aqueous phasa but its stability in the extract de­

: mands the certain excess of the ligand in the organic phase. The obtained data on the constants of the complexes extraction and

i their spectrophotometry characteristics allowed to choose the opti­: mal systems lor the analytical application: iodide and bromide bis­i; muth complexes with Bix­ and seven­membered CTD.

73

table 5. The Results of Bianuth Determination

Object Biemith content, % Object certified found

tungsten ore 2039-81 Sulphide ore 793-76 sulphide ore 792-76 Sickel alloy 11 44 Standard piece of rare metal netaaoaatite DBR-1 3tandard piece of chalcophilic netasomatite DBZ-1 Standard piece of meymechite DEM-1 Standard piece of dunite CD'I-1 №

2 ° 5

2.3.10"2

Э.4-10"3

4.5-10"3

(4.2+0.2)-70"'J

-

5-10- 4

(2.2+0.1).10-2

(3.3+0.1)-10~3

(4.2+0.1).10"3

(4.26+0.13)'Ю"3

(4.4+0.4)"10"3

(4.6*0.5). 10" 4

(2.2+0.2)-Ю-4

(1.1+0.2)-Ю-4

(4.8+0.2)-Ю*4

It was found that the extraction-photometric determination of 2-50 ug of bismuth as the complex Bi-TMTU-I- in not interefered with 0.5-1.0 g of Ni, Fe, Co, Al, Mn, Zn, Hb, Sb, Sn; 100-200 mg of U, II, Mo, W, as well as 1 g of ascorbic and tartaric acids, sulphates. Oxidants should be reduced before the extraction.

The extraction-photometric methods for the determination up to 10~

4

56 of bismuth in nickel alloys, tungsten ore, polymetallic ores, niobium pent oxide, rocks, metallic antimony were worked out. The results of bismuth determination in some objects are given in Table 5.

74

Щ

A COMPARATIVE STUDY EXTRACTIOH OF METAL COHFOUltDS WITH I | CHROMOPHORB REAGENTS AHD LOHG-CHAIN QUATERHARY AMMOHIUMI Z 1 SALTS AM) THBIR SOLUBILIZATION WITH MICELLES OP HON-IONIC SURPACTAHTS

M.H.Tananaiko, L.A.Gorenstein, T.I.Viaotskaya, Kiev State University, Kiev, USSR

USxe usage of the surfactants of different nature as additional components of analytical reactions opens a l o t of new poss ibi l i tes for their ut i l izat ion. Under such conditions we can see changes in spectrophotometric characteristics! oxidizing potentials of substan­ces and there appears the possibi l i ty of r.hnnging the concentrational conditions of reactions,solubility of products. I t allowed wide used of surfactants (SURF) in photometry and extractional separation,in

chromatography and a number of other cases [1J • In this paper we have looked through the usage of SURF in extrac­

tion, where they are additional components of metal ion reactionals with hydrophobic reagents,as well ss using of solubilization produ­

cts of such reactions by micellar forms of SURF. Both methods have big practical importance, Ihey permit stsight forward execution of analytical reactions in hooogenic and two-phase environments.

One can single out three groups compounds,formed in the metal--coroioophore reagents-SURF system:

1. imycSUHtf)^ ; 2. MRx(aSURF)y I 3. K( nSUKP)^ ,

where cSUBF-cationlc, aSUUF-anionic, nSUKF non-ionic SURF.

Qhey interact by association mechanism, mixed ligand complex formation and formation of complex compounds of metal ion with non-- ionic SURF molecules [ 1 , 2] .

At the extraction of auch compounds we usually use just a l i t t l e quantity of SURF, which does not exceed their threshold of micelles formation in water solutions. Under such conditions we extract more simple associates of reagents H - SURF, that increases contrast of photometric reactions .For

the fullfilment of similar reactions in water solutions we use redundant quantity of SURF.In using charged SURF i t i s possible the concurrent influence on the central ion and the reduction of painting product output. Bsides there can be formed intensively pain­ted associates of reagents,which make worse the photometric conditions.

. Such influence i s especially developed in alkaline environment. By our experiments we have shown that i t i s possible to use non-

75

ionic SUKF for solubilization of metal compounds with chromophoric reagents and cSBEF. For solubilization on* can use non­ionic SURF. taiceilar forms of such SUSP stabi l ise the compounds in water so lut i ­

ons, at the same time providing nigh contrast of reactions. Such ef­

fects i l lustrated Wgure 1 (a,b),on the example of molybdenum сов­

рои tie with brompyrogallol red ( ВДО ) and cetylpyridine chloride ( &=). I t i s seen from the f i g . 1,that in solutions (pH 6) the exe­

cution of reaction i s not effective due to the disturbing action of associates reagents painting (Pig.1,a,curve3). Solubilization of compounds with ВЯВ and CF by micelles of non­ionic SUKF QP­10(poly­

oxoethylatea. ether) hypsocnromically shifts the absorption band rea­

gents associates' (M.g.1 ,b,eurve3)> litre we have the increase of complex output (Pig.1 ,b,ourve4). Thit, effects are due to the fact that non­ionic SUW? suppressed dissociation of hydroxide group of BPB.

ffig.1. Electron spectrum of reagents and complexes of molybdenum

(in absence­ a and in presence­ b of OP­10).1,1'­BPH«

2,г'­1к>­ВЯЦ5,3'­ВШ­СР! 4,4' ­ Ко­ BFB ­СР.

°lto= °ВРК =°СГ S­'O"5

"» °ов­10 2 ­ 1 0 ­ 4 pH 6

Cf joins to the most dissociated sulfo­groups of reagent, that i s not in the chain of the cosjugate.lt increases the contrast of the reaction. »juong interesting paculiatit ies of the reaction one should pointout the broadening of optimal conditions of interaction in al­

kaline f i e ld , that i s also concerned withthe influence of non­ionic SUHF. At the extraction of compounds lio­BOt­CP by chloroform a l l the spectrophotouetric characteristics of the compounds are retained. С The data are i l lustrrated in Table Л j .

It i s known from the l i teraturefj l , that micelles of SUE? or

74

cueii pieiaicexiar a&bTegai­es inf luence on the centra l i ons ,prov ia ing their аепу d m nation. Under such condi t ions the r e a c t i o n a l a b i l i t y of metal i ons i s increased. ,a lso the number of uni ted l i gauds i s i n c r e a ­

sed as wel l as the compound at a b i l i t y grows. Such dependence was shown on the example of r e a c t i o n with brom­

pyroga l lo le red and hexyldiantipyri lmethane (HDAk). I t i s knoun that

the HDAJiis a large hydrophobic b a s e . To use r e a c t i o n s with t h i s r e a ­

gent t y p i c a l l y tace the e x t r a c t i o n by organic s o l v e n t s . Our e x p e r i ­

ments i n d i c a t e , tha i t h i s r e a c t i o n could be f u l l f i e l e d i n water phase a t s o l u b i l i z a t i o n the compound by m i c e l l e s of OP­lo.Under such condi­

t i ons the r e a c t i o n p a s s e s i n more wide range of a c i d i t y , t h e jo ined groups B­fli incceases (Table 1)* for optimal formation of compounds i s considerably l e s s HDAb( 4­6 t imes , i n s t e a d of 800 times at e x t r a c t i o n ) I t proves the grawth of s t a b i l i t y bond i n compounds.

Table 1 . 'ine chemica l ­ana ly t i ca l c h a r a c t e r i s t i c s of the complexes

Keactants Stoicliioiuelry pH •Ушах, £ шах Medium

b:E:SUHF opt . niu х Ю­ 4

BPB­OF 1:1 6 ,5 62C ­ aqueous BPK­CP­ 1:1 6 ,5 570 ­ m i c e l l e s OP­10 ВРВ­СР 1:1 1 , 5 520 ­ e x t r . CHCL,

m i c e l l e s 0P­10 Ьо­В£Ь­0Р 1:2:1 1 , 5 ­ 7 , 0 620 5 , * e x t r . CHCL, m i c e l l e s 0P­10

Jto­Bi­Ts­CF 1:2:2 1 . 2 ­ 1 , 6 620 5 , 3 e x t r . CHCL, m i c e l l e s OP­10 ЙО­ВРЬ­ША*. 1:2 2 ,^­3 ,7 620 * , 5 e x t r . CHCL, m i c e l l e s OP­10

т-вт-шлм 1:1:2 0 , 5 ­ 0 , 6 620 3,8 e x t r . CHCL, m i c e l l e s OP­10 Cd­PAB­OP­10 1:2 10 500 10 e x t r . CHCL, m i c e l l e s OP­10

Cd­PAH­OF 1:2:2 9 , 2 520 1 1 , 8 aqueous;extr.

l'he studied, r e a c t i o n prevents general chemical i n t e r e s t as i t snows the p r i n c i p a l p o s s i b i l i t y of s o l u b i l i z a t i o n by m i c e l l e s of nSUEP compour ­. Including d i f f e r e n t organic Ъаае.

lhe e f f e c t s under cons iderat ion are t y p i c a l f o r the other chro­

uophoric reagent s . So, the s imi lar dependence axe made f o r brom­

phenole blue , puluo rones , oxiazo­compounds. As an example i n the

Table 1 are compared the proper t i e s of e x t r a c t i o n oadmium compounds » i t h PiB and CP and s o l u b i l i z a t i o n products by mice l lar forms of SUBP. One e f f e e t i v n e s a of both metods makes the r e a c t i o n u n i ­

77

vereale applicable for extractional oetermination and photometry in water solution [4] •

Thus, the solubilization of metal compounds with chroaophoric reagents and organic base by micelles of non­ionic SUBJ allows to increase the contrast of reaction, aa wall aa stabilized the painting In time. I t Is similar to the extraction»! ayatems. At the sane time the application of nSUB? micelles widen the f i e ld

of Interaction in дИгиМпа environments, increases the reaction ca­

pability of central ions, promote the formation of more Intensively painted and coordlnately saturated compounds»

'lb» alove mentioned reactions have wide application at separa­

tion and determinating of metal or SUSP of different nature. Some of develops methods we have ahoun in the Table 2. Obey come up to standards methods. Table 2. The analytical research of the reactions Jt­B­cSUBF­nSUBF

Ueactanta Object Determi­nable component

The i n t e r v a l of fo l lowing Ber's law mkg/ml

Sr

Со­F*. £F­CP­0f­1 0* blood serum Go 0,004­0 ,006 0,035

Й>­ВН(­СР­ОР­10 s t e e l , ( s o l ) Uo(Cf) 0 , 1 ­ 0 , 4 ( 0 , 6 ­ 3 , 5 )

0 , 0 3 ;0 ,04)

to­BPE­HDAk­OP­l о s t e e l , ( s o l ) •o(Of­10) 0 , 3 ­ 0 , 8 (10­50)

0 ,022 ( 0 , 0 4 )

Cd­PAB­OP­IO sewage Cd 0 , 1 ­ 1 , 0 0,026

* S3.PB­ phenylphluoronr

Heferencee 1. Pelisaett i £.,.Pramanro X. / / Anal.Chim.Aota. 1985. Vol.

169. P. 1. 2. Nasarenko A.Xu., Tananaiko H.H.,Todradie G.A.// Dokl.Akad.Nank

UkrSSB, Зег В. 1983. *11. f>50. 3. Ohernova U.K.// Zh.Analyt. Ш.Ш. 1977­ Vol .32* 8.

P.1477. 4. Tananaiko M.H..Vysotakaya T.I . / / VTkr. Ehla. Zh. 1982. Vol.

4 8 * 6 . P. 629 ( lav. VUZ. 3SSB . 1984. Tol. 27, ш 5. P. 544.

78

OiRECT COMBINATION OF SOLVENT EXTRACTION AND ANODIC STRIPPING) „ 2 Q

VdtTAMMETRr OF SOME METALS I — — ­

J.Labuda and M.Vanfckove, Department of Analytical Chemistry, Slovak Technical University, 812 37 Bratislava, Czechoslovakia

Stripping voltammetry is the most sensitive electroanalytical me­thod for the determination of trace levels of substances which may be preconcentrated onto the working electrode prior to the actual quanti­tation. A markedly increased selectivity of analysis can be achieved by the employment jf effective separation technique. The use of solvent extraction in stripping voltamnetry is usually associated with reex­traction of metal into aqueous phase or involves mineralization of the organic phase. From the point of view of working steps number re­duction a direct determination in non­aqueous medium seems to be pros­pective.

In our previous study [l] we showed the possibility of electrolytic accumulation of metals on hanging mercury drop electrode (HMDE) ter extraction of their diethyldithiocarbamate (DOC) complexes to be ze-

ne. Analytical data for cadmium determination in 1:1 benzene­me nanol medium and separation of cooper as an interfering element from the mixture with cadmium, lead and zinc have been obtained. The aim of this work is to verify the possibility of deposition potenti? . de­crease in the presence of mercuric ions which undergo to the rae

J

M ex­

change reaction with the central atom of dithiocarbamate с nplexes. The question of selectivity of analysis of cadmium, lead, tnallium and indium giving overlapping stripping peaks has been air > studied.

Experimental. For the extraction NaDDC (cupral) as wf 1 ai more selective and stable Zn(DDC>2 were used. Experimental cunditiDn* are given in Table 1. Polarographic analyser of the PA 2 tjpe with the set of Rotating Disc Electrode (Laboratorni prlstroje,Praha) were employ­ed. A HMDE of the Kemula E­69b type (Radiometer, Copenhagen) and mer­cury film electrode MFE prepared "in situ" by the me cury deposition on a glassy carbon disc were used as indicating ele trodes. The refe­rence electrode was calomel electrode with 4 M LiC and a salt bridge filled with the supporting electrolyte. The auxiliary electrode was platinum electrode.

Results and Discusssion. The dependence of с Jmium, lead and thallium stripping peak current on deposition potential has been in­vestigated. From the time independent reductir i potential values it follows that in the presence as well as in Xb absence of mercuric Ions the DDC complexes are reduced during th deposition step on both

79

Table 1. Experimental conditions of solvent ex traction voltammetric determination

Metal Aqueous Organic phase/ Supporting electrolyte/ ion phase agent deposition potential

Cd2 < 0.05­0.3 M HC10, benzene/Zn(DOC) 2 benzene­methanol 1:1

or pH 5 ­ 6 or NaDDC 0.1 M NaCluu/­0.9 V Pb

2

* 0.05 ­ 1 M HC10 4 benzene/Zn(DDC) 2 benzene­methanol 1:1 01 pH 5 ­ 6 or NaDDC D.l M NaC10 4/ ­0.9 V

Tl +

pH 5 ­ 6 benzene/NaODC benzene­methanol 1:1 0.1 M NaC10 A/ ­0.65 V

In3

* 2 M HC10 4 chloroform/ chloroform­ethanol­water Zn(DDC) 2 1:4:1; 0.005 M CHjCOuNa,

0.006 M HC1, 0.D6 M KBr,­­1.1 V

HMDE and MFE. In the case of thallium, however, the difference between the reduction potential of Tl and T1DDC is due to law complex stabi­lity, only small. Addition of HCIO. lead to the shift of the deposi­tion potential towards the more positive values.­ This is obviously caused by the release of metal from DDC complex prior to the electro­lysis. The condition of the exchange reaction evidently lies in the presence of Hg(II) reactive form. The reaction equilibrium is conside­rably shifted to the tight and the exchange reaction proceeds quickly.

The Hg(II) concentration was maintained in the 10" ­ M region in order to ensure the excess of Hg(II) against the extracted complexes as well as its usual value for the mercury film formation. No signi­ficant peak current changes were observed in the range 3x10 to 6 х Ю М HgClI) and 5X10" 3 to 5xlD ­ 2 M HCIO^. At the piesence of mercuric salt the electrochemical reveresibility of the redox systems increases.

Under the conditions securing metal release from DOC complex the calibration curve was obtained for cadmium. The dependence of peak current an cadmium concentration in the original aqueous solution in the range 10 M is linear. The reproducibility is characterized by the relative standard deviation cca 7 U at the level lxlO ­ 7 M, the limit of the determination is lxlO M. It was also tested the possibility to make the extraction as well as the voltarwnetric measu­rement in the sane vessel.

The difference of the values of stability and extraction constants of the individual DDC complexes [2, Э] gives the assumption of the separation of these also at the great concentration ratio of metals.

80

The ascertained possibility of determination is summarized in Table 2. The ratio between the metal amount which was determined and that taken for the analysis was always near one. The determination may be realized also on the principle Df one vessel.

Table 2. Separation conditions for tm.­tal determination in the presence of interfering metal

Molar ratio of metals Aqueous phase

Agent/ Organic phase

Extraction time

Cd ­ Tl 0.3 M HC10 4 Zn(DDC) 2/ benzene 3 min 1 : 1000 Tl ­ Cd pH 5 ­ 6 NaOOC/ benzene 3 min 1 : 200 In ­ Cd 2 f. HCIO, Zn(DDC)2/chloroforr.' 10 mi^ 1 : 1000 Cd гее, •' In ­ Tl г м нею. Zn(00C)2/chloroforn 10 m . 1 : 1000 1 The obtained results confirm the possibility to use the formation

of DDC complexes in the determination of metal traces in more compli­cated systems by the combinated rr.3thod : solvent extraction stripping voltammetry.The possibility of influencing the extraction selectivity [3] allows 'o consider the determination of other metals, too. The efficiency of separation is increased by back extraction of interfering metal from the organic to aqueous phase. Potential possibilities of stripping voltammetry allow also the determination at lower concentra­tion level than at the verified one. It is, however, necessary to keep the principles of trace analysis inclusive the work in one vessel. Preliminary separation in the extraction step is here both perspective as well as tedious and risky procedure.

References 1 . Labuda J . , DanclkDva' R . / / Chem.Ptjp. 1983, V o l . 37. P.733. 2. Byrko V . M . / / D i t i o k a r b a m a t y ( D i t h i o c a r b a m a t e s ) . Moscow : ­ ^uka ,

1984. 3. Bajo S . / / J.Radioanal.fttueJUChem. ; 8 5 . V o l . 9 0 . P . 5 1 .

6.3uc. 382 81

EXTRACTIVE Х­ОЮ­МЯОНВЗСПГСВ HBTHODS ОТ DBTHMI»ATIO" О» I ВЕИОТЯ АП) CBALCOGBI КЬЯШИЗ * М Н LOW­MLTDTO KXTRACTAHTS ' F.I.Lobanov, S.S.iiossagaabetova, V.A.Leonov, A.K.Hurtaeva, V.V.Yakehin, Kazakh State Uni~ereity, Alma­Ata, USSR

The applying of commercial X­ray­spectral apparature for toe analysis of technological solutions, orea and concentrates has been limited by low­sensitivity apparature, difflcultieb In the concent­rating procedure and preparation of radiating patterns. There is the task of determination of bismuth and chalcogens In complicated ob­

jects. Bismuth extraction from the solutions with different anionic com­

position by low­melting extractants has been studied. Technical mo­nocarboxylic acida (fractions C­.­Cg., stearic acid, 8­hydroxyquino­llne, mixture of tributylphosphate (TBP) with paraffin hydrocarbons served an the low­melting extractants.

The optimal conditions of bismuth­ion extraction from halogenide, nitrate and sulphate solution have been found for different reagents. The composition and behaviour of the extracting compounds ­ere estab­lished by the methods of IH­spectroscopy, spectrophotometry , X­ray phase analysis. The increase of hydrogen ions concentration ?aad to change of composition of extracting compounds from BiXj.jTBP to BiX,.2TBP.

The extraction constants in the system bismuth ­ stearic acid ­paraffin hydrocarbons ­ water at 80°C have been calculated:

X w = 0,87 i 0,12 (1,0 U stearic acid in paraffin); Kgx = 0,85 ± 0,07 (0,1 И stearic acid in paraffin). There was established on the basis of achieved results that the

most perspective ;.iw­melting extractants for the extraction of bis­muth in analytical practice are high carboxylic acids (HCA).

extraction of chalco^en elements ( sulphur(VI), selenium(IV), ^ellurium(IV) that accompany bismuth in ores and minerals was stu­

died. Extraction of these elements was carried out by the melts of ЛСА (technical fractions with the number of carbon atoms > 15) from neutral and bromide solutions.

The influence of different factors on the distribution of above­mentioned anions in the system the melt of HCA ­ wate>­: acidity of aqueous phase, the contact time, the phase volume ratio, the concent­ration of elements, ionic strenght and salt composition of aqueous phase was studied. The op> ша1 conditions of the maximal extraction of sulphur(VI), selenlum(IV) and tellurium(IV) in the absence and

i

82

in the presence of the chlorides and nitrates of soiie metals have been found.

There was established, that selenium and tellurium are extracted from acidic chloride and bromide solutions. The aaxiaal extraction was observed In the case of eelenium(IY), which has been extracted quantitatively In the wide interval of acidity. The composition of extracted species was proved: H,SeCI6.2Hfi (for the 2­4 M HCI). The difference in extractive capacity of Se(IV) and Te(IT) allows to carry out efficient separation of these elements.

The considerable influence of salt composition of the aqueous phc­ee on the extractive capacity of chalcogen elements was established. There was found the dependence between the concentration of different metal cations in aqueous phase and extractive properties of chalco­gens.

The series of the Influence of different ions on the extraction of the anions with the melts of ЧСА. have been found. The obtained re ­suits pex­mlt to propose the express methods of determination of chal­cogen elements in the solutions with complex salt composition. Some objects such as semiconductor materials, sewage were analysed by pro­posed methods.

83

ЛРГЫСАТГОН OP ЕХГ.АСТ10Н AHD KXTRACTIOH CHROLiATOGHAPHIC | ~ 9­22

PR^COHCaiTRATIOr О? ТМИШТ1Т5 Ш THE /JJALYSIS 07 COrtPOUED 1 1 THI1URIDES S.S.Grazhulene, V.K.Xarandaehev, Yu.I.Popandopulo, li.I.Chaplygina, L.V.Zadneprui, E.A.Borsova, Institute of РгоЫешз of Microelectronics Technology and Superpure Katerials, Academy of Sciences of the USSR, 142432 Chernogolovka, USSR

To eliminate matrix effects in different multielement methods of the determination of micro impurities in 2n, Cd, Hg, Sn and РЪ tpllurides it is essential that matrix elements should be isolated. To do this, extraction and extraction chromatography can be used to advantage.

The technique of sorption of a number of matrix elements on the column, which has been insufficiently studied, ia of great interest from the viewpoint of a single­stage separation of varly^a elements for multielement methods of analysis. An optisial choice of '­he txtrac­tion system enables the method to be unified with геа'пь­•'. ­о a wide range of materials to be analyzed as well as different is;a of detec­tion.

Practically all the matrix elements comprising semiconductor tellu­rites are readily extracted either with tri­n­butilphosphate (TBP) or tri­n­octylamine (TOA) from 3­6 M IiCl solutions L 1

»2

!' W e b a T e found

that a TOA+TBP mixture efficiently extracts the above microcomponents both separately and in different combinations. The mixture was found to exhibit a synergistic action (Fig.,11)1

.

Pig. 1. Extract ion of microamounts of Zn, Cd,, Hg, Sn, Pb and Те with TOA and TBP mixtures. Curve 1 ­ Те (Cj =0.2 M); 2 ­ Hg(0.2 M); 3 ­ Cd(0.2 M)j 4 ­ Sn(0.07 M); 5 ­ Pb (10~3 M); 6 ­ Zn(0.2 M). Curves 1­5 ­3 M HC1; 6 ­ 0.1 M HC1

0 20 40 60 80 100^ TBP 100S6 TOA 60 40 20 0

Yet, under these conditions many impurities are poorly extracted (Ka, Mg, Al, K, Ca, V, Cr, Kl, Ito, Rb, Sr, Y, Cs, Ba, REE, Bi), if at all (the distribution coefficient < 10~2

) <Fi$.2).

i g u *

. ^ 1 .

• « i ^ L ^ * ~"

* . ^ 1 .

­ ^ a 5 % * = * % 4 ^ 5 ^ч>= б " 5 * ^ 3N

­

84

lgE Thus, the possibility exists of separating the iapuritles from the matrix elements. Besides, the latter were found to sup­press extraction of ifflpurity elements. By way of example, Fig.3 shows the data on the influence of zinc microamounts on extraction of impurities. However, in extraction separa­tion the relative concentration coefficients (F) are small in­spite of low distribution coef­ficients of impurities and sup­pression of their extraction by !1&IL- Extraction of Hi (curve 1), macrocomponents. This indicates z

r(2), Sc(3), Cu(4), Co(5), As(o), that in the cases in question 'W> * "

ь а TBP+TOA mixture (1:1) extraction preconcentration can­ i n accordance with HOI conoentra­not afford efficient separation *xon of matrix and impurity elements. Yet, in neutron activation analysis of the matrices the values of P should not be smaller that 10 [з]. Such values can be achieved in extraction chromatographic preconcentration by means of multiple ext­raction­reextraction cycles.

2 4 60 H C 1,M

Pig.3. Extraction of microamounts (~10 M) of elements in the absence (1) and presence (2) of 0.2 U Zn

It has been previously Bhown [•] that the use of extraction chrom­atography in the system of TOA+TBP mix.ure­HCl solutions allows an efficient single­stage separation of the following impurity elements: Ha, Mg, Al, K, Ca, So, V, Cr, Mn, Co, Hi, Cu, As, Rb, Sr, I, Zr, Ag, Cs, Ba, REE, Hf and Bi. One of the disadvantages of extraction chrom­

es

atography lies in a relatively large volume of impurity containing eluate which increases the time of separation and blank test values.

To reduce the volume of the eluate, we have studied the effect of column feed by the matrix elements. The eluate volume was quantitative­ly shotvn to decrease with increasing weight of the sample (Pig.4).

0.10 •

0.05 ­

C, rel. unit 1 2 Pig.4. 2inc elution curves. The

elution was performed with 9 П КС1 from a column filled with 15 g of a PTF3+TBP mixture (H=17 em; D=1.2cm; V,,=8 ml; V =5 nl) in the presence (curve 1) rnd absence (curve 2) of 100 mg of Те

To optimize the height of the sorbent layer (and to minimize the eluete volume) at a certain sample weight the distribution of macrocomponents along the column height was studied (Pig.5).

Pig. 5. Distribution of the matrix elements along the height of the sorbent layer upon separation of impurities from 100 mg of CujBe^le; the column filled with 15 g of PTFE with a TOA+TBP mixture (B>19.5 cm; D=1.05 oms 70=10 ml; V s=5 ml). Eluent ­ 3 Ы НС1

1, cm

Assuming that the least sorbed matrix element is distributed in the chromatographic column In accordance with the normal law, the sorbent layer height can be calculated for a given value P A/B' where A is the

10'

least sorbed matrix element and В is the most sorbed impurity. Pig.6 shows the results of the calculation for the case of separating impu­rities from 100 mg of Cd_Hg.,_xTe.

P Pig.6. Coefficient of relative oon­

10° L ^ - centration of impurities as a func­1Q4 . tion of the sorbent layer height

upon separation of impurities from 100 mg of 0* H*i_rTe­ Column with

"Yl "l.om D«1.05 om. Eluent ­ 3 M HC1 4(­r 86

Pig. 7. aiutlon curveo of Ha, Cr, La, Hf (curve 1), As (2) and Cu (3); column with a high sorbent layer (A) and an optimal height (B). fluent ­3 E НС1. Column A; H­19.5 am; D­1.05 em; V„=10 ml; V e­5 ml. Column B: 11*12 cm; D»1.05 cm; V„­7 ml; V =3.1 ml

C, rel. units

0.10 ­

0.05

о.го 0.15 к в

0.10 I A 0.05 \k\ . ­ ! . , ­ 1

30 50

The use of the foregoing approach tor optimisation of the aorbent layer height Is illustrated in Fig. 7 by impurity elution curves for an optimal size column (calculation­ made for P=10 ) and a column with a sorbent layer height exceeding its optimal value. ?гош these curves it is evident that optimization enables the eluate volume to be reduced (in this case by 30S6).

The investigations performed made it possible to develop a universal technique of single­stage extraction­chromatographic separation of more than 20 impurity elements fru all semiconductor tellurides. The tech­nique is suitable for neutron­activation and atomic spectroscopy meth­ods of analysis and affords determination of impurities with detection limits from 10"4 to 10"%.

References Llshimori Т., Hakamura B. // Japan At. Energy Comm. Rept. JAERI ­1047, 1963.

2.Ishimori Т., Akatsu E., Tsukuechi K., Kobune Т., Osuba J., Kimura K., Onawa G., Uchiyama H. // Japan At. Energy Comm. Rept. JAERI ­ 1106, 1966.

3.Karandashev V.K., Yakovlev YU.v., Grazhulene S.S., Kolotov V.P. // J. Radioanal. Nucl. Chem., letters. 1985. Vol.95, N 6. P.367.

4.Grazhulene S.S., Popandopulo Yu.I., Karandaehev V.K., Zolotaryova N.I. // Analyst. 1987. Vol.112, H 4. P.455.

87

9­23 EXTRACTION AAS DETERMINATION OF TRACES OF ELEMENTS IN PURE PLATINUM Z.Aneva,Refinery and Petrochenical Research Institute,8104 Bourgas, Buigaria S.Arpadjan.S,Alex&nrov,Chemical Faculty of Sofia University,1l26 Sofia, Bulgaria Emission spectroscopy[1,2] and atomic absorpt ion spectrometry[3,4 J

with preliminary separation of the matrix element by solvent extraction have been used for microimpurities determination in high purity plati­num. However , the ex tr agent us&d (trioctylamin and isoarryl alcohol­methyl­isobutylketone) are not selective with respect to gold,iron,molybdenum, antimony and tin.

In *his work solvent extraction of the chloride complexes of Au(III), Fe(III),Sb( V),Sn(IV) and Mo(VI) form diluted hydrochloric acid into methyuisobutylketone (MIBK)[5,6] in combination with flam \AS determina­• n of these elements in high purity platinum, is proposed.

The group extraction of 1­40 mg­l" Au,Fe,Mo and Sb, and 20­100 mg­Г

1 Sn,in the presence of 10­100 g­l" 1 Pt from 6 M HC1 at V W : V Q ^ 3 : 1 , 2 : 1 , 1:1 and 1:2,and 300 s phases contr :t has been investigated.The effect of platinum on thp extraction d atomization of each element.under in­vestigation,has been considered [7].It has been found that, the extract­ion of microelements is not influenced quantitatively by the matrix ele­ment,which is partially extracted (D=0.05),but the latter suppresses their AAS determination.This problem has been overcome by completely stripping the platinum from the organic phase by 6 M HC1.

The behaviour of other low extractable microimpurity elements (Al,Ca, Mg,Ni,Cr,Pb,Cu,Mn) attending platinum, have also been studied.A depress­ing influence of the matrix element on their extraction and nore effect­ive stripping in its presence,have been found out.

The mutual interferences of the extracted elements in the flame at different quantitative ratios, have been investigared [7].Availability of interfences in condensed phase at interferent to analyte ratios of more than 10:1, has been found.These interferences have been very pro­nounced at ratios of 50:1 and resulted in 25% of modifications of the absorbance signals.

One of the simplest ways to eliminate interfering effects is to use additions with respective action.Considering the extraction mechanism of chloride complexes into MIBK.the influence of long­chain quaternary ammonium salts, has been studied [7].The latter are not suitable for selective elements extraction, but when added to MIBK,after its separa­tion from the aqueous phase,.eliminate cowpletely all the interfering effects.Their protective action has been explained by for.ming stable ion pairs with oversized alkylammonium cation.

On the basis of these investigations, a simple and reasonably ртасс i­cal procedure has been proposed.After sample decomposition by aqua regia, the analytes have been extracted from 6 M HC1 solution of Pt(100 g*l" ) jnto equal volume of MIBK.The aqueous phase has been removed and the pla­tinum which has remained in the organic solution has been stripped twice by the equal volumes of 6 M HC1.A protective agent,trioctylmethylammoni­um chloride, (II against V ) has been added.The calibrating solutions ha­ve been treated in the same manner.

The absorbances have been measured at resonance absorption lines in C2H2/air flame for Fe,Au and Sb.and in C 2H 2/N 20 flame for Mo and Sn at 1.6/19 1­min and 3.6/14 1­min" flow rates^respectively,

Data on accuracy,precision(RSD) and detection limits(DL) are presented in Table.

Characteristics of the method proposed,t(P=0.95,n=5)=2.78; t (P=0.95,n=4)=3.18

Analysis by flame AAS for NBS standard reference mate­rial No.681 has been in good agreement with the certified values (in ppm): Element AAS SD Recom. Report­

values ed values

Ele­ment Added

mKl_

Jj Found mil"

1

Error t

n RSD exp DL(I)

Fe .5 .45*.05 10.0 4 .10 2.0 2.10"5

1.0 1.004.05 0 .05 0 Au 2.5 2.48s.05 .8 .02 .80 i.10"5

5.0 4.90*.15 2.0 .03 1.49 Sb 2.0 1.901.09 5.0 .05 2.48 4.10 ­ 4

2.5 2.501.08 0 .03 0 Mo 2.5 2:571.10 2.8 .04 1.40 4.10"5

5.0 5.1С!.27 2.0 .05 .83 .'in 2.0 2.010 0 0 0 1.10

­3

4.0 3.91.10 2.5 4 .03 2.0C

5.1 0 S 5.0 0.1 9

3.5­12 5­10

References Shuvaeva O.V., Ivanova S.N., Yudelevich I.G., Chanlsheva T.A.// lsvestija SO AN SSSR. Ser.chim.n. 1984. Vol.4,N 11. P.64. Shuvaeva O.V., Yudelevich I.G., Ivanova S.N., Chanisheva T.A.// Isvestija SO AN SSSR. Ser.chim.n. 1984. Vol.6, N 17. P.101. Aneva Z., Arpadjan S., Alexandrov S.M Kovatcheva K. // Miltrochlm. Acta [Wien]. 1987. ?.341. Aneva Z., Pankova M., Arpadjan S., Alexandrov Anal.Chem. 1987. Vol.326. P.561. Boswell C.R., Brooks R.R. // Mikrochim.Acta [Wien

// Fresenius 2.

P.814. 1965. 6. Zolotov Yu.A., Iofa B.Z., Chuchalin L.K. Extraction of halo­com­

plexes of metals [Russian]. MoskowrNauka, 1973. 7. Aneva Z., Arpadjan S. // J.Analyt. Atomic Spectrometry [in press].

89

ктлстате савитнлпог и так AMLYSIS OP HIGH гонт i „ 2 4 i И Л Т И Ш АШ) Я Н Ш И Н L 1 О.Т. Shuvayeva, S.1* Хтапота, I.G. Yudelevieh, Т.A. Chanyaheva, Institute of Inorganic Chemistry of the Siberian Branch of the ЛсаДецг of Science* of the U33H, Xovosibirak, USSB

Determination of the Impurity composition of pure platinum metals is a very Important problem of analytical chemistry. A considerable effect of some impurity elements on the operational characteristics of pre­cious metal wares and the absence of strict correlations between the physico­chemical properties of a substance and its Impurity composi­tion requires control over a wide range of impurities in noble me­tals. At the present time there ere no simple readily available me­thods for analysis of high purity platinum and rhodium with a determi­nation of the microcomponents at a level of 10"' to 10"' %. Concent ­ration of the impurities by an extractive separation Л the base com­ponent makes possible to decrease considerably the detection 3 jilte of the known instrumental methods.

The present work is devoted to the study of the possl^ _jity of an extractive separation of the base in the analysis of high purity pla­tinum and rhodium with subsequent spectral determination of the Im­purities in a highly sensitive variant using the graphite collector with 0.5* HaCl developed at the Institute of Inorganic Chemistry of the Siberian Branch of th» USSR Academy of Sciences £\J. The method places rigid requirements on the quality of the matrix component se­paration ( the purification factor must be not less than 10 ) which is due to the interfering effect oi' the Pt and Bh spectra as well as to the effect of the base residues on the spectral line intensities of other elements. The extractant for the matrix separation in the analysis of high purity substances must have a high capacity, effi­ciency, and selectivity with respect to the base element, it must be easy to purify and be practically insoluble In water.

The most effective extractants for platinum are nitrogen­containing e­tractants: amines, salts of quaternary ammonium bases, N­oxides of triootylamine and tk ­ nonylpyridine /27.

Рог the quaternary salts of ammonium bases the distribution coef­ficients of Pt(IV) are considerably affected by the counter­ion : the higher is the ion hydration energy the hljdier is the probability for him to go to the aqueous phase. Prom this point of view НС0Г and riC,0T forms are more preferable /37. In the extraction with free amines the extraction can be expected to be still more effective since here there will be no competing effect of the exchanging ion at all. Pig. 1 shows 1­ extraction isotherme with nitrogen­containing extractants from 0.1 mol/1 hydrochloric acid solutions Bince It is in

90

such nearly neutral solutions where the extraction with the anion­exchange type extractanta can be expected to be most selective due to unfavorable conditions for the existence of anionic chlorocomplexes of impurity elements. A calculation of the number of extraction stag­es required for e quantitati­ve recovery of Pt(IV) from solutions with Cp.= 100 mg/ral using the working line у * = Ж х ­ x Q) ( x, у are Pt con­centrations in organic and aqueous phases , \ is the ratio of the volumes of the aqueous and organic phases ) shows that only for trioctyl­emine an effective single­batch extraction of Pt(IV) can be expected. In the saturation part of the extraction curve of Pt(IV) with trioctylamiiie the mole ratio RJl : Pt is equal to 2.04 which is possible both in the case of formation of an ionic associate by the reaction :

Pig. 1. Pt<IV) extraction isotherms from 0.1 mol/1 HC1; С = 0.5 mol/1 1 ­R 3FHHC 20 4; 3 ­ 2­NP0

2 ­ E 3 Nj

E 3U + H 2PtCl £ i 6 *=• (R^gPtClg (1) and in the case of insertion of an amine into the inner coordination sphere of platinum :

2 R 3 U R 3 HHPC1 5 HR 3 +2HC1

(2) P t C l

A^)2 + г HOI

Stripping of Pt(IV) frcm. freshly prepared organic nhaae with nitric acid (1)1) results in going of more than 99»Oj6 of platinum into the aqueous phase , an observt Mon In f avou;.' o* the reaction taking route (1).

After separation of the base the concentration of Pt in the aqueous solution is 10 to 25 ^Vml which amounte to 0.1 to 0.25* of the weight of the graphite collector containing the concentrated Impurity ele­ments. However for Al, Hn, La, Gr, Ti there is a depressing effect of Pt on the spectral line intensities at Pt content from 0.01 Я to 0,5 % . Therefore to eliminate the interfering effect of the matrix 0.25% Pt was introduce! into the referenoe samples.

The developed method for analysis of high purity platinum makes

91

possible to determine 22 impurity elements ( Be ­ at the level of 7 10" 8 %; Cd, in, Hn, Cu, Ag, Tl ­ (1­8)­Ю ­ 7 * ; Ba, Co, La, Hi, Pb, Ti ­ (1­7)­10~6 % i Al, Ca, Mg, Rh, Ru, Cr, Zn ­ ( 1 ­ 2 M O ­ 5 i; Ir, Sr at the level of 1.10~* * ). In general, the presence of Pt in the extraction system practically does not affect the behaviour of the impurities. For Ag, In, and Tl the extraction was found to be depressed by the base presence and for Pe, Sn, V, Bi, Те coext­

raction is observed. The choice of a reagent for removal of the base in the analysis of

high purity rhodium is a very complex problem which is associated with the kinetic inertness and a relatively high charge of the Rh complex anions. It can be expected that the extractability of slight­ly hydrated polynuclear complex anions will be increased when using anion­exchsnge type reagents which was found in the extraction of Pt with the salts of quaternary ammonium bases /4?.

Evaporation and subsequent drying of the complex halogenides of Rh(III) result in formation of polynuclear species in which the ha­logenide­ions act as bridges. In the aqueous solution the polynucle­ar complexes depolymerize , with the equilibrium state corresponding to the anion RhXg С~Ъ .?. In the extractive dissolution of the poly­nuclear complexes of Rh there occurs their rapid transfer into the organic phase after which they slowly go into the aqueous phase under depolymerization» In order to retain as much as possible of rhodium ic the organic phase it was necessary to find suitable con­ditions preventing the depolymerization of the polymeric species. It turned out that when (RhBr,)n is dissolved directly in the ext­ractant ( TOA) in the subsequent contact with HC1 solutions 99.97 56 of Rh is retained in the organic phase , with some of the impurities being concentrated in the aqueous phase. Since in the presence of Rh ita contents in the graphite collector of ^ 1.0i6 were not found to decrease the spectral line intensities of the impurities the ato­mic ­ emission determination of the impurities from the graphite

uo *m­* MO

IR­ape­trum of compound I m­K­ 3/?Ч Z.H5

HJR­spsctrum of compound I P i g . 2

92

collector can be carried out using a single set of the reference sam­ples on the besia of graphit powder. The method allows determination ol' Be, In, ,Mn at the level of (1­5). ю

­6 ­7 *; Ba, Co, Hi, Ti, Cr ­

A3., Ca, Mg, P at the level (1­5) • 10~ 5 %. The be­­ (1­2) • 10" haviour of the impurities is independent of the presence of Bh and is determined only by the concentration of the aoid in the aqueoua phase. The Eh compound which is formed in the organic phase when using thia method of the base separation was studied by various physical methods. In Pig. 2 are shown IE and Й Ш spectra of compound I which was isolated from the organic phase. The signal ­ 7.5 m.p. in the PMR spectrum indicates the presence of the R 3 m

+ group in the mo­lecule of the compound under Btudy. The polymeric nature of this compound is responsible for the presence of the absorption bands at 133 and 255 cm" corresponding to bridging and terminal Br atoms. There are no spectroscopic data indicative of the coordination of H to Rh. Therefore we can suggest existence in the organic phase of a compound corresponding to the formula

к Й 3 Ю 2 J Br Br ,Br . Br — H h ­ B r — Rh­Br Br Br ­Br

­Rh­Br • ^ B r

The elemental analysis data

С H Rh Br

found ialw.lated

50.1 51.С

9.2

9.1 9.5 9.1

30.9 28.3

and mass­spectral methods.

in which the cationic part is solvated. for I is presented in the table.

The developed method? for ana­lysis of high purity Pt and Rh have better sensitivity than the direct spectral methods now in use in industry and are practi ­cally not inferior to the hard­ly available neutron activation

References 1. Yudelevich I.G., Buyanova L.JI., Shalpalcova I.R.

Khimiko­spektralnyj analte veshchestv vysoJcoj chistoty. NovosloirsI:: Nauka, 1963.

2. Shuvayeva O.V., Ivanovo S.5., Yudelevlch I.G., Chanysheva Т.к.// Izv. Sib, Otd. AH SSSR. 1984. Vol. 4, H 11. P, 64.

3. Ivanov I.M., Gindin L.H., Chichagova G.Wlzv. Sib. Otd. AN SSSR. 1967. Vol. 3. Sex. Khim. Каик, N 7. P. 100.

4. Ivanova S.H., Druahinina I.A,, Belayev L.V./Uhurn. Seorg. Khim, 1983. Vol. 28, N 5. P. 1261. Belayev L.y., VenedActov A.B., Khranenko 8.P.i0koord. KM». 1983. Vol. 9, H 1. E. 120.

5.

»3

SObVBJTE EXTRACTICar OF ВОВЬЕ MBTAbS »ITH AZAABAMMS I | 0? I>IBEIZ0­18­CR0n­6 ' ' •..K.Beklemlshsv, B.M.Kus'min, S.B.Panfllove and Yu.A.Zolotov, Vernadsky Institute of Geochemistry and Analytical Chemistry, Academy of Sciences of the OSSB, Moscow State University, Moscow, USSR

The noble metals extraction with aacroeyolic compounds has been only very poorly studied, though maorocyoles axe promisable as se lec­

tive extxactanta. We have studied a aexies of 14 nitrogen­containing macrocodes in respect to 30 metals

j ^ ^ Y ^ N *•*­ CH2,CH2iO,CH2| CH2,0| ,Ш. « Ц ^ р ^ 0.0« HTa,CH2) CHg.HTs) HTs.BTsi

Т С у m

« i ° | 0,BTs| HH.CHjj CHg.HH) '0 O ^ V / HH,HH| BH,0| O.BH

^ ^X

^ J (Та « p­CH3­C6H4­S02­) Quantitatively ox pooxly ertxacted exe only Pd, Os(IV), Pt(XV),

Au(III), and Hg(II)| Co, Hi, Cu, Ag, and T1(I) demand special counter­

ions (dipicxylaminate) for their extraction, and the following metals axe not ertxacted (pH 1 ­ 10t 1­10

­3

M macxocyclas in CHClj): Ha, Ca, Sc, Ti(IV), Cx(III), Hn(II), Fe(III) , Zn, Oa, Se(IV), Zr(IV), Bb(V), Ru(III), cd, Sn(II), Sb(III), Те(IV), Ba, Ce(III), Ix(IT). This makes up a possibil i ty fox aeleotive extraction of osmium(IV) ox palladium. Their extxactlon has been studied in detai l .

Osmium i s extracted wit i maoxocyole X»0, I­ЯН in 1,2­diohloroetha­

ne (Pig .I ) , other macxocycles and diluents axe Хаев effeotive. Tha dixsot and the reverse extraction i s fast (not more then 20 sek)| the i. 'ertxon spw­tra show the presence of OsClg in the organic phase,

eleotxophoxesls data for fresh­prepared extracts also point at metal presence in the anionic form. Aftex more than 2 h storage

. ­ ­' migration towards the anode changes for i ta migration to the de, that can ba interpreted as the formation of cationle Os(II)

• .III) complexes. The competitive process^ auoh as the extraction of anions of the

aqueous phase have taken staaied. Using the radioactive tracers method the chloride ion distribution coefficienta vs. 0 C 1 and 0 L (L: 1.0, Y­BH) have been obtained. The chloride ion i s pooxly extracted! i t s total concentration in the organic phase i s of the order of 10

_ 5

« (иС 1>1'10~* ­ 2<1C~3). Bevertheleaa i t can seriously effaot the osmi­

um distribution. Possible equilibria of chloride extxaction with L have been sugge­

sted, and the system of equations has been ooapiled, inoluding the matter balance, deotioneutxality, equilibria constants, and D c l »x­

pxesslons. The nuebox of variables has >een one more than the number of equations, so two unknown constants (extraction and dlssosiatloc

94

l o g Do,

I L 1<«1>о.

г

0

3 7 pH KUt„I. Extract ion of 0»(IT) (5­

• I O­ ,

l t ) « i t h L I O­ 3

! ! manrooycls 1­0 , Y»«H, fro» 0.04 И »ЬС1; buffer or НЯО,; I ­ 1 , 2 ­ d l c a l o r o s t b s a » , 2­CBCL,

­5 ­« ­3 l o g C^

ПяЛ. Extract ion of Os(IV) (5­ю""

7

*) e i t h aacrooyole X=0, Y­ЯН i n I .S ­d loh loros thane , pH 3 . 8 ; I ­ 0.04 M, 2 ­ 0 . * M «aCl

XogDy,,

fC ­ I I l o g 0 ( Д

? l g . 3 . Bxtraotlon of Oe(IT) with naorocyele x ­ 0 , Y­HH (Ь) In I r 2 ­ d i c h l o r o e t h a n e , pH 3 . 8 ; I ­ I ­ I 0

­ 3

M X, 2 ­ 3 ­ I 0­ 4

K L On Flg .2*3 оигтвв are c a l c u l a ­

ted «1th the found constants

3 7 pH F i g , 4 . Extract ion of palladium

( I . I 0 ~ * M , I­I0­ : 5

M macrooyeles: I ­

X­y=CHp i n to luene , 0 .1 M NaC10 4, 2 ,3 ­ X«KH, Y­CH2 i n CH01 3; 1,3 ­

n i t r a t e s o l u t i o n , 2 .5 h, 2 ­ o h l o ­

r lde s o l u t i o n , IS mln

of the complex) bare been optimised s imultaneous ly . The fo l l owing scbeoe i s the nost adequate t o tbe experiment:

K

C1 L ( e ) + H

+

+ C l " LRC1 (o) ( K

C 1= 7 . 6 ­ Ю * ) ,

LHC1 м,а (o) = b H

( o )+ a

( „ ) ». d,Cl 8 . O . I 0­ 5

) ,

( I )

(2)

ь

( 0 )+ н

* з ^ Ь Н * (Kj ' 5 . I ­ I 03

, speotrophotometrlcal ly determined) (3)

The model i s true for C L < (1­3) °10_ 3

M and C C 1 < 4 ' 1 0 ~2

H at oonstant pH 3 . 8 .

Further tbe d e e o r i ­ t i o n of osmium d i s t r i b u t i o n has baen f u l f i l s d as shown abora, using the obtained oonstants ( 1 ) ­ ( 3 ) . Tbe best c o i n c i ­

dence with the experiment I s g i r e n by the aches* i

2 I ( o ) + 0 S C l | " + 2 H+

*4 ,0а

*0s iH0scaj ( o ) + ьн+о) <

K

0s* + .S ­

I 0I 0

> <*)

LBOaCL. 6 ( 0 ) ' b H

( o )+ 0 , C 1

i 6(0) v ' d . o . ­ 1 .6 .Ю­*) (5)

The eurres ca lcu la ted with constants ( l ) ­ ( 5 ) ( P i g . 2 , 3 ) general ly

95

fit the experimental points | worse fit is obtained for D ^ ­ C Q 8 de­pendences.

To estimate the selectivity the distribution of other elements in the osmium extraction oonditiona пае been studied (1>10~^> L, pH 3.8, 0.03 • HaCl, 20 век):

IS 0 sI V

K u1 1 1

R uI V a »

1 1 1 Pd I r

1 1 1 I r " P t

1 7 A »1 1 1 H g

1 1

loglfc 1.5 ­1 о —If..*

­ 2 . 1 •c­1.3 ­ 1 . 2 < ­ 2 . 5 ­ 1 . 9 0 .9 0. 1.3 The nethod for extractiTe­oatslymetric osmium determination ba­

sed on the o­aminophenol oxidation with hydrogen peroxide. The reacti­on goes in the mixture: extract(1,2­dichloroet*aane) ­ ethanol ­ \ ter solution (1.1:1.0:0.14), in the presence of sodium acetate (acti­v ­tor). From 3*10 to 0.02 Mg/ml osmium can be selectively determined (the detection limit ia 1.6.10~*ywg/nl). The determination is not hindered by 1­10* ­ 1'105

­fold amounts of Al, Cr(ITI), Mn(II), ГвИХ), Со, ГЛ, Cu, Zn, Hh(III), Ag, 3e(IV), Te(IT), Pb| (1­2) M O 3 ­ of Fu(III)*(IV , Pd, IrCIV); 600 ­ Pt(IV)| 200 ­ Hg, and 100 ­ Au(III).

Palladium (rig.4) is slowly extracted)' the equilibrium is not at­tained during several hours. Proa chloride solution» palladium is ex­tracted more rapidly than from nitrate ones. Electrophoresis data (aaorooycle: X»HH, Y=.CH2) show that the metal la in the oationic part of the complex. The protons are shown not to" be released during the extraction. The determination of CI and Pd (by means of CI and

glTes the composition Pd:L:Cl«1:1:2 for a solid complex and Pd.'Cl » U 2 tox an extract (J­HH.CHg). That la, Pdbclg (or its oligo­mer) ie formed.

The extraction is aeleotivei the dietrlbution ratios of other me­tals are given below (logD for nitrate solutions, I­10' of phase contact)

,­3 M L, 2.5 h

I , cond. Pd Oat IV) Pt(IV) Au(III3, Hg Ag Cu 11 Co

HH.CHg i n CHClj, pH 4 CH 2 ,0H 2 i n t o l u e n e , pHI

1.35

0.95

­ 0 . 4

­ 1 . 2

­ 1 . 4 «

­ 1 . 7

­ 0 . 4 »

­ 0 . 9

­ 1 . 3 "

­ 1 . 0

­ 2 . 5 ­ 1 . 6

4­2.4

­ 2 . 8

«­Э.8

<­2.B

<­2.B

* In the presence of 1 .3 * HaCl

The method of extract iva­atomio­aoaorpt ion (wi th e lectrothermal atomlzation) determination of palladium. With the use of e x t r a c t i o n (macrooyole: X«IH, T­CHg) 0.006 ­ 0.075 /*в/"1 of palladium i n a s o l u ­

t i o n can he determined (the detec t ion l i m i t lm 0 . 0 0 3 / i g / m l ) . Tb* b a l e * g iven amounts of other elements do not i n t e r f e r e the determination of 4>I0"

7

« of palladium: Co, Ag (2.5'^0*'i)} I I , Cu, Pb ( 2 . 5 ­ 1 03

) ! Hg ( i ­ I O

3

) , Zn (250) | B u ( I H ) , Ku(IV), Pt(IT)', Au(III) (25) I Rh(IIT) ( 5 ) | T! Os(IV), I r ( I V ) , Pe(III ) ( 2 . 5 ) . Г:

96

9­26 EXTRACTION OP DITHTOCARBOXYLATES OP PLATINUM METALS IH ANALYTICAL CHSMISTRY G.P.Mezharaup, E.J.Janson, I.S.Kunitskaya and Z.A.Garnak, Peter Stucka Latvia State University, Riga, USSR We have studied the extraction of benzene­, naphthalene­, quinoli­

ne­, indole­ and phenylmethanedithioearboxylates of palladium(II), pla­tinum(II,IV), rhodium(II), iridium(III,IV), osmium(III,IV> and ruthe­nium(III)[I]. The obtained chelate complex compounds are follow or yellow­brown and can be readily extracted with such organic solvents as chloroform, carbon tetrachloride, benzene, toluene. Buthylacetate and iBopentanol are the best extragenta for the hydroxylic substituted dithiocarboxylates.

The light absorbtion of the extract containing dithiocarboxylate of platinum metal is usually measured in the relation with an extract ob­tained in an blank experiment, i.e. in absence of complex­forming sub­stance [2]. Depending on the aqueous phase pH dithiocarboxylic acids can aleo pass into the extract partially or completely. ^

Absorbtion of the extract of the blank experiment is: *ex=

^HL*v " b

* Therefore, when meanning absorbtion of the extract containing e x

dithiocarboxylate in the relation with the extract of the blank experi­ment the following constant is really measured:

A e x=(fe M e, " "'MeLj * & H L ) fMe.D. A=eonst, pH­const.

This constant dosB not depend on the reagent surplus and thus is very important for the extraction­photometric determination of plati­num metals. The given relationship enables to calculate not the real molar coefficient of the absorbtion of complex dithiocarboxylate but only the difference в ц в т ­ "^eLJ^HL > functionally depending on the wave­length of the absorbed light and the aqueous phase pH. When describing the absorbtion of the extract by means of this difference the graphic representation is a three­dimensional diagram, having axes of the wave­length of the absorbed light, aqueous phase pH and the dif­ference mentioned above (fig. I).

The absorption of extracts la studied fro» separate sections of the diagram. If the section is examined with constant aqueous phase pH, spectral characteristics of the extract are obtained, if the weve­­length of the light is constant, the pH interval of the aqueous phase where dithiocarboxylate is formed and extracted is determined. Fig. represents two sections of the tree­diaensional diagran with the depen­dence ^ц вт >­'

, х

и 6­ * *m, o n t b e wave­length end the aqueous phase pH. It has been found that with the exception of palladium dithiocer­

boxylates, preliminary heating Ingredients st 95°C for 5­50 minutes Is necessary in order to get maximal yield oomplex formation. It has been

7. 3«it. 382 97

Two sections of the three dimensional diagram showing the dependence of e ^ ­ H * ^ e m

on the wave­length and pH of the aqueous phase

is found to be: PdL~, PtLg, PtL.,

determined that the ligand nature influences the speed of the in­teraction, but the increase of its surplus accelerates the pro­cess of the formation of comple­xes negligably. Among the nume­rous studied additions formic acid was found to be the only one

Л to accelerate slightly the com­plex formation. Interesting pro­perties are demonstrated by pal­ladiumOII), whos9 dithiocarboxy­lalea can be extracted even from the concentrated hydrochloric acid.

Composition of the formed com­plex compounds is determined by the mole­ratio, continuous­varia­tion methods, the shift of equi­librium and by means of the Bjer­rum function and Asmus method and

RhL,, IrL», OsLj, OsL., RuL, (L dithiocarooxylate­ion)

It has been found that aqueous solutions of the aquahydrogenethyle­nediaminetetraaoetstoruthenium(lll) containing ions £ВлдУ(НгО)] " react with dithiocarboxylic acids (HL) at room temperature according to the following scheme [з]: O ^ H j O ) ] " + HL = [RUYL] 2

" + HjO+

, ­>he ratio Hu:L = 1:1 is determined in aqueous solutions with the mole­ratio and the shift of equilibrium methods.

As follows from the equation of complexforming reaction, the formed mixed­ligand lone are negatively charged therefore they do not undergo extraction by organic solvents. The extraction ta possible in the pre­sence of hydrophobous cations! benzenethiouronlum (BTO), diphenylgua­nidinium (DFG), triphenylguanidinium (TFG), tetraphenylphosphonium (TFP), tetraphenylarsonlum (TPA), tetraphanylpyridlniua (5PP). The possibility of using various organic solvents for the extraction has been tested (table 1).

Polar solventa possess higher extracting ability than non­polar solvents. With the increase of the dielectric permittivity of alipha­tic alcohols their ability of extracting ionic associates rises. The ionic associates are extracted from the aqueous phase most completely by all the investigated solvents in the presence of TPP, TPA and TPP. The extraction degree of mixej­ligand complexes of ruthenlum(III) in the presence of BTU, DPG and TPG considerably depends on the nature of the solvent. The reaction takes piece best of all with the mixture

95

/

r

Table I. Extraction degree (56) of the ruthenium(III) ionic associates containing 4­dimethylaminobenzenedithlocarboxylate­ions

( t = 20±I°C,­ n = 5; s « I* )

i Hydrophobous cations (Organic Solvent BTV DFG TFG TFF TFA TFP Isopentanol : chloroform (3:7) 90 100 99 100 100 100 Isopentanol 78 82 88 99 98 99 Butanol 85 88 87 92 95 95 Octanol 10 74 75 90 85 96 Chloroform 16 34 25 92 87 94

! Nitromethane 1 ­ 22 15 84 84 92

(3"7> of isooentanol and chloroform and the mixture 1.1:1) of isobuta­nol and nitromethane.

It is better to иве EFG as a hyUrophoboue cation for practical pur­poses Ъесаиве of its availability and easy purification. It has been found that quantitative extraction of ionic associates of ruthenium (IIT) requires 1000­fold surplus of DFG irrespective of the ty):>e of substituted benzenedithiocarboxylate­ion.

Table 2 represents the data of spectroohotometric determination of platinum metals using several dithiocarboxylic acids.

Table 2. Spectrophotometric determination of platimun metals

i Reagents Me ^«f/rnl ' fe.IO­4 . . , ....

Anm ; Acidity

(H,C)*N­­(0)­CSSH K u d u ) Os(III)

0,02­10,0 0,02­8,5

3,7 7,0

520 480

pH 2,5­6,5

CSSH

i Ф j OCH,

Ru(III) ,' 0,1­35,0 Pd(II) 0,02­9,0 Ir(TII) j 0,05­25,0 Pt(II) j 0,02­10,0

0,8 4,8 2,8 7,0

498 425 420 488

pH 3,0­7,0

pH 6,0­7,9 6JJ HCI ­ pH 3,0

i References ';• I. Mezharaup G.P.//Izv. AN Latv.SSR. Ser. Chem. 1987. N 2. P.192. ;• 2. Janson S.J.//Dithiocarboxylates in Analytical Chemistry. Riga'.

Zinetne, 1978. 136 p. 3. Mezharaup G.P., Kunitskaya I.S., Janaon E.J.//Zh,Anal.Chem. 1986.

; Vol.41, Я I. P.99.

i 9»

THE USB О? АЫПГЬАНШНЕ IH ESTRACTION­IHSTRDMENTAI, METHODS I 1 FOR ИАТППШ METAL DETERHmATIOH I

9"

2 7 I A.A. Vasilieva, V.G. Torgov, H.K. Kalish, L.V. Zelentsova, Т.К. Korda, H.V. Uakslmova and I.G. Yudelevich, Institute of Inorganic Chemistry of the Siberian Branch of the USSR Academy of Sciences, Novosibirsk, USSR The possibilities of highly sensitive atomic­absorption (AAM) and

atomic­emission(AEtl) methods for group determination of platinum me­tals (PM) can be realized in the analysis of complex objects by elimi­nation of the interfering effects of the macrobase elements by means of extractive concentration. The most effective extractants for group separation of PM are the primary aromatic amines, in particular, n­octylaniline (0A) and its easily available commercial analog ­ n­al­kylanillne (AA)[1J. In contrast to S­ and S­H­containing compounds [2,3] these extractants are able to extract quantitatively all PM in­dependently of the oxidation state and do not require addition of la­bilizing additives to accelerate the extraction.

Chemistry of PM extraction wl.n n­octvlaniline. The extraction of РИ with weakly basic 0A is accompanied by formation in the organic phase of different species depending on C ™ . Extraction of anionic acido forms of РИ is observed at HC1 concentrations sufficient to con i vert 0A to 0A­HC1. Such concentrations are equal to ЗИ Н01 for Ir.Rh, and Ru [A] and 5­*M НС1 for Pt and Pd. At lower acidities the extrac contain coordination or mixed compounds. Thus for the extraction of P and Pt from 3M'HC1 , the extracts were shown, by IR and electron spec troscopy and preparatively, to contain [Pd(0A)2ClJ and[0AHPt(OA)Cl5"j A quantitative description of the Pli extraction is complicated by association of OA'HOl in toluene ( which is the most convenient sol­vent for the instrumental analysis of an extract), polymerization of the extracted complexes at high PH concentrations or their dissociat­ion at low PIC contents. These processes are less pronounced in a cap­rylic acid media where w« were able to determine the anion exchange extraction constants for Pd and Pt. But in this case the distribution coefficients (D) are by two orders of magnitude lower than for toluen solutions of 0A.

Absolute concentration of PH. the optimum acidity for group extrac/ tion of PM is 2­ЭМНС1. A quantitative extraction of PM is observed over a wide range of their concentrations ( Table 1) and at PM eonoen tration» from 10 to 10 Tl the values of D are almost constant (25­33 for Pd and Rh, 60­90 for Pt, 100­120 for Ir and Ru, OA HC1 cor centra tion ­ 0.5H). At a two orders of magnitude lower initial PM cor centratioa the value of D increases to 200­250 which ellowe a 5 to 10­fold absolute concentration in the analysis of depleted and ^

100

highly depleted ore» «bile retaining the extraction (R,ft) at tie le­vel 91­98 $ up to the ratio of the aqueous to organic phase volumes of Uq:0) ш 10:1 (Table 1). Ir and Ru in the oxidation state (III) and (IV) show practically the same values of D. From the point of view completeness of the ext­

Table 1. The extraction of РИ with AA so­lutions in toluene from ЗН НС1 at Aq:0=1:1 ration and the physical

properties of the extract it is expedient to use in the subeequebt analysis 0.5U toluene solution of AA. Relative concentrat­ion of PH. The study of the extraction of the macrobase elements in the analized objects in the range of their concent­rations from 0.01 to 111 and at 1 to 3M HOI showed that the values of D are mostly at the level n­10" (for 0.5 U AA), except for Zn, Sn(II), Pb (D« =1­10). The value of D is independent of the me­tal concentrations for Co, Ca, Hi, increases with it for Fe, Hg, Al, Sn, Cu and Cr(III) and decreases for Hn and Zn. The distribution coefficients of Pt and non­noble metals are equal to 103

­10 . The extraction of microquantities of PM against the background of the non­noble metal macroquantities studied on model mixtures simulating copper­nickel production products and on technological products is practically the same as that presen­ted in Table 1. A fall of R to 65­70% «as observed only in the case of Pd. Therefore, in the analysis use was made of a mixture of AA (0.411) with petroleum sulfides (PS) (0.15 H) which ensures a ten­fold concen­

v tration of Pd at B­98­100Jt. To avoid possible effects of non­noble me­..­­ tale on the extraot analysis and the physico­cheaioal properties of "' the extract it ш washed with 211 HC1 and 1M НЖ> Э ( to remove Pb).Ap­proximately in accord with 0 of the non­noble metals, eaoh washing pro­duces a 102

­times Increase of the PM concentration. After two weening» ?tb* metal ooctent» amount to less than 10 to 20 p.p.m. The '•'PI losses do not exceed 2­3K, the overall separation coefficients of $Pt and non­noble mettle increase to 10 6 ­ 10 8

.

C

AA'» C

PM' K

Extraction, 36 C

AA'» C

PM' K

та. К Rh Ir Ru 0.2 0.5 1.0

2­Ю~£

2­10"2

2­10

92 97 98

94 98 99

83 97 97

94 99 99

­

0.2 0.5 1.0

5­10"J

5­10"3

5­Ю"3

93 97 97

98 98 99

92 96 98

99 99 99

­

0.5 1­10­

* 100 100 95 95 96 0.2 0.5

6­Ю­ 5

6­10"5

100 100

100 100

93 98

91 95

96 98

0.5 0.5* Э.5** Э.5***

V10~" 1­10"

6

1­10"6

1­10­ 6

100 99 98 92

99 98 91 91

98 97 92 75

­

97 96 91 86

*Aq:0= 5:1, 10 1 20:1

101

Interferences In the Instrumental analysis of the extract.The mu­tual effect of PH and '.be effect of non­noble metals on PIC In the AA and AE methods is considered.

In the flam* atonization in AAH there are both the mutual effects of PH and the effects of Fe,Cu,Bi,Pb and other metals when they are present in concentrations equal to those of PH. Although these effect! are of different nature they all can be eliminated by adding to the e: tract a buffer solution La(HO,), in ethyl alcohol with up to Vf> of La The measurements are carried out using reference solutions containing all PH and La. In the case of Pd the reference solutions are used without La since other PH and the non­noble metals do not affect the Pd analytical signal up to their contents of 1 rag/ml. In all cases th< extractant concentrations in the reference solution and in the extrac being analieed are kept constant and equal to each other.

In the electro­thermal atomization there are no mutual effects of PH for Pd, Pt, and Eh up to a 1000­fold excess of each metal and no effects of Ir and Ku up to their contents in a 200­fold excess. In th determination of Ir and Ru a decrease in the analytical signal of the metals was established at Pt:Ru, Ru:Ir, and Hhslr ratios of 200:1» 10:1, and 40:1, respectively. It was also found that a 150­fold exces of Pd and Pt and a 600­fold excess of fih.Pd, and Ir do not affect thei.; atomic absorption of Ir and Ru, respectively. For products with an \ unfavourable PH ratio methods have been developed for their extractic • separation. The PH analytical signal Is practically unaffected by the . presence of non­noble metals in amounts up to a 400­fold excess of A3 j Cr.Ug, a 2000­fold excess of Au and Pb, a 8000­fold excess of Fe,Cu, | Hi and other metals. Washing of the extract is needed in this case tc stabilize the extract composition and the analysis conditions. For tl same reasons the extractant concentration is kept constant in the ex­tract and in the reference solution.

Th<) atomic­emission method of PH determination in an extract invol ves ashing of the extract, dilution with a spectral buffer and int­roduction into the arc plasma by a " blow­in ­ spilling1

' method. In this method no mutual effects of PH are observed up to their 100­foli excess. Xh* effeot of non­noble metals on the FH spectral line inten­sities and the background is completely removed by washing of the ex/ ract. Comparison of the calibration plots obtained by extraction for each PH concentration and by dilution of extracts with known PH con­tents Indicated their coincidence. She second technique is less lab ­consuming and allows preparation of samples for several measurements She linearity region ot the calibration plots is, *­10 3 : Ir and Hu 0.3 ­ 7; Pd and Pt ­ 0.3 ­ 14 ( for Pd to 50) ; Hh ­ 0.15 ­ 7.

txtraotlon­lastra—ntal methods of analysi». xhe analysis proce ••* dure* involve decomposition of the products with а НС1+ШГО. mixture^'

102

Table g. Hetrological chaxecterlstics of the

; samples with high Si and S contents are first treated with HP or mnealed). The insoluble residue is then fused with Ha­O, and the melt Ls leached with HC1 after which both solutions are added together. The «tractive separation of РИ is preceded by removal of HNO, by alter ­lately treating the resulting solution with HC1 and water ( the presen­ce of 0.5И HHO, in the solution decreases the value of D of Rh and Ru to 2.7 and 0.3, respectively). To increase the reliability of the ana­Lyeis for produots with very low РИ contents ( 10""­10" 36) double­concentration methods have been developed ( the first stage ­ to pro­luoe а РЬ alloy or a Hi matte). In the analysis of the Pb alloy the ?H are separated from the Pb macroquantities by thermal dissociation rith nitric acid and subsequent leaching of ?b nitrate with water, rhe insoluble residue containing Pit and not nore than 0.2% of Pb Is dissolved in HC1 md then the extraction Ls carried out ( an maloguous procedure is ised in the analysis 3f products containing nore than 30* Ag). The developed methods ( their characteris­tics are given in Table 2) find applicat­ion in the analysis of geochemical samples, various ores and their is eztraoted quantitatively concentration products, liquid and solid technological products of hydro­ and pyrometallurgical stages in copper­nickel production in­dicating a universal character of the method for group separation ox F*.

References ­ 1. Vasilieva A.A., (Sindln L.H., Shul'aan R.S. et al.// Izv. SO AH

SS3R. Ser. Shim. Sauk. 1983. » 4. P. 65. Z. Seregina I.P., Petrukhln O.K., Poraanovskii A.A., Zolotov SU.A.//

Dokl. AH SS3R. 1984. Vol. 276, Я 2. P. 385. ,3. Dlaaerta­tae A., Verbeok A.A.// Analyt. Cblm. Acta. 1977. Vol. 91.

P. 287, j'.4. Uastafayev H.P. , Shul' man R.3., Gindln l.U.// Izv. SO AH SSSR. jf Ser. Shim. Hauk. 1978. H 2. P. 54.

FH determination methods H Determination

method Ooncentration ranee . %

S

r Ir Rh Ku Pd Pt

The atomic­emission me­thod la the "blow­in ­spilling" variant

3­10"Y

­1­10~s

6­10"7

­1­10~5

3­10"6

­V10"5

1 ­10"6

­1­10"5

3­10"6

­T10"5

B­10"7

­3­10"6

0.25­0.19 0.20­0.09 0.25­0.10 0.20­0.07 0.20­0.05 0.25­0.10

Pd Pt Bu Rh

Atomic­ab­sorption me­thod, air­acetylene flame

1'10"3

­5­10"2

1­10~3

­1­10~1

0.09­0.02

0.06­0.01 Ir Pt Bh Ru

Atomic­absorption method, graphite furnace

4­1O"5

­2­10~3

4­Ю"3

­1­Ю"2

5­10"6

­1­10"3

1­10"4

­7­10"2

0.15­0.02 0.09­0.01 0.15­0.02 0.15­0.01

+"ln the case of AA­r­FS. mixture all gold

103

THE APPLICATIOK OF EXTRACTION IN ANALYTICAL CHEMISTRY OP BORON

J.M.Schwartz, Institute of Inorganic Chemistry, Academy of Sciences of the Latvian SSR, Salaspils, USSR

Extraction is used in analytical chemistry of boron for: I.sepa­ration of boron from interfering elements, 2.concentration of trace amounts of boron, 3.extraction­spectrophotometric­, 4.extraction­titrimetric determination of boron, 5.standartization of sample for spectral­, atom­abeorbtion and flame­spectrophotometric determina­tion of boron.

Recently, quite a large number of methods for extraction­spectro­photometric determination of minute quantities of boron has been de­veloped. They are based on the formation of anionic chelate complexes of boron with ligands containing гЧ­diols or cX ­hydroxycarboxylic groups, which permits to extract boron from a weak acidic or weak alkaline solution in the form of ion pairs together with basic dyes. Some examples see In Table J, Those methods are rather sensitive but not selective and require the separation of boron from most of ele­ments of periodic system.

Table I. The extraction­spectrophotometric methods of determine­ ' tlon of boron based on the formation of ion pairs

Chelating ligand

The Ьав­lo dye

The organic solvent pH nm

p JBefe­

'rence

3 ,5 ­d i ­ t e r t . ­

buthylpyroka­

techol

ethyl violet toluene 6,3­8,7 610 Ю 5 i ft7

2­methyl­2­hy­

droxybutiric acid

malachite green

chlorben­

zene 2,5­4,3 629 " : C27

2,3­dihydroxy­

naphtalene c r i s t a l violet

" 3,0­7,0 595 I

til

mandellc acid malachite gxeen

benzene 2,2­3,0 633 6.25­10 4 M 2,3­dihydroxy­

naphtalene chrompyrazol I! i ^

The universal method of separation and concentration of boron iB the extraction by aliphatic J5­diols in organic solvent (for example СНС1 3)Л7. We have studied more than 20JJ­diols C­5 ­ C­14, including some isomers C­9, C­10. The extraction is based on the formation of cyclic esters, which are stable towards hydrolysis and soluble in or­ganic solvents. The extractants are effective in acidic, neutral and £

i 104

9­28

Fig.1. The depundence of boric acid extraction by p> ­diols in CHClj (0,5M) on pH: 1.­3­methyl­1,3­butanediol, 2.­1,3­nonanediol,3.­2,6­dimethyl­4,6­octanediol,4.­2,2­dipropyl­1,3­propanediol,5,­2­isopro­

pyl­5­methyl­1,3­hexanediol Pig.2. The dependence of boric acid extraction on the concentration of ­diol in CHClj: I ­ 1,3­butanediol, 2 ­ 3­methyl­1,3­butanediolt

3 ­ 1,3­nonanediol| 4 ­ 2­propyl­1,3­heptanediol| 5 ­ 2­buthyl­1,3­octanediol) 6 ­ 2,4­dimethyl­2,4­octanediol| 7 ­ 2,4­dlmethyl­2,4­nonanediol

weak alkaline media (Pig.1). The reextraotion la performed by the so­lution of alkali (2­356). The partition coefficient usually increases with the growth of the concentration of jl­diol till saturation (Pig. 2). In воте solvents (CHC1,) the synergetic effect is observed owing to diminishing of the self association of J!>­diols. The extraction equilibrium is reached afer 1 min. or some hours and depends on the structure ofp­diol and the acidity of solution. In acidic medium it requires less tl­ "he rate and completeness of «extraction is also dependent on 1 <­ture of jj ­diol.

The miners? ­. especially KaCl and MgClo, have the salting­out* effect, whioh ­ jsted with the concentration of salt, but in some cases the tiae *­­ sat ting­in of equilibrium is increased.

The effeotivltj of extraction is not connected with tbt concentra­

tion of H,BO, in the solution, hut the excess of JJ­dlol (10x in mol) la neoeaeary.

The type of ester (1:1,2:1 or 3=2), which is formed depends on the struoture and solubility ofJi­diol. In the oase of more soluble and liophylic Ji­dlols, fox example n­1,3­diole, 3­methyl­1,3­butanediol,­it Is a cyclic «stor 1:1, in the case of leas soluble in water and cryetallin dlols ­ lees «table towards hydrolysis but nora extrae­• table eater 2:1. The latter JJ­diols are more effective as extraotante. nOnly the thxeecoordinated boron is found in the extraet by IB­apaotro­Fecopy. The existenoe of cyclic esters R<JJ>BOH and B<^>B­OROH in ex­

105

Table 2. Extraction properties of some J!­diols in the system: j>­diol/CHCl3 (0,5M) ­ HJBOJ ­ H20,25°C

­diol The compound formed

Limiting D

H 3 E 0 3

Solubility ,«,25°C diol in H.o: H,0 in

d , ^diol n­3,5­oktanediol 2,2,4­trimet;.yl­

1,3­pentanediol n­1,3­nonanediol 2,2,5­trimethyl­

1,3­hexanediol 2­propyl­1,3­hep­

tanediol 2­ieopropyl­5­methyl­

1,3­hexanediol 2,2­dipropyl­1,3­

propanediol 2,4­dimethyl­2,4­

oktanediol 2,6­dimethyl­4,6­

oktanediol

3:2

2:1 1:1

Г 1

3:2

2:1

2И 3:2 2:1

1:1

26

30 9

31

4?

335

23

63

14

­ [Insoluble 1,22 J 23,83

0,52 I insoluble

' 8,51 1,02

I insoluble

' 8,51

0,62 2,56

0,28 insoluble

1,22

2,13

1,95

0,91

tracts is confirmed by the PMB­spectra upon the example of the reac­tion of 3­methyl­1,3­butanediol with 1ЦВ0, in water ­ organic medium,

J?or all of J}­diols investigated, condition may be selected, when Dy jjO­i * 20. This allows to extract boron quantitatively by two cycles when macroamounts, and by one cycle when microamounts of boron are del termined. The distinctions of diols ae to solubility, rate of extrac­tion and reextraction and so on make it necessary to use in each con­crete case various diols.

Thus, one of the more appropriate JJ­ciols for the concentration of boron is 2­isopropyX­5­methyl­1,3­hexanediol, which allows to concent rate boron in extract for 3 ­ 5 times.

Por extraction­titrimetrle determination of boron, which is based on the separation of boron from interfering elements by extraction in CHClj end alkalimetrio titration the extract by addition of mannitol with visual or potentiometric indication of equivalence point, the aoet appropriate are 3­««tbyl­1 ,>butan«diol,2­ethyl­1,3­nexan«dlol, 2­methyl,2,4­pentanediol,2,2­dipropjl­1^­propanediol, whiche allow easy «extraction of boron. 2,6­и1т^пу1­4,6­оМапва1о1,г­ргору1­1,3­heptanedlol are used for titrieetrio deteir.ination of boron onlv af­ter leaxtxaction.

For epeotrophotoeetrlc determination of alcroanounte of boron wit» oaroin or 1,l'­dlantrial4 onlj le«a soluble p­dlole are useful, for a

106

Table 3. Ihe maximum possible excess of interfering ion by extrac­tion­titrimetric determination of boron (M:B weight)

1 •

и MBD HD DMOD DDPD

Al+3 | 170 10 interf 10 G a + 3 i 20 75 10 I n + 3 75 75 110 L a + 3 , 150 150 70 SiO ~ 2 | i n t e r f . Ti(IV) !

350 SiO ~ 2 | i n t e r f . Ti(IV) ! 110 200 GeUV) i 60 50 i n t e r f . S n + 2 ; 1C0 160 200 Р Ъ + 2 20 100 300 PO:

3

4 , As+ 3

100 70 110 280 PO:3

4 , As+ 3 10 100 90 S b + 3

5 5 65 B i + 3 100 100 300 С Г + 3 60 900 MoO"2 300 150 50 1000 W 0 4 , 100 200 V0+2

200 200 20 50 F e + 2

125 1600 C O + 2 220 1500 H i + 2 200 3650

MBD ­ 3­methyl­1,3­burtaned i o l , HD •­ 1 ,3 ­nonaned io l , DMOD ­ 2 , 6 ­ d i ­

m e t h y l ­ 4 . 6 ­ o k t a n e d i o l , DEP D ­ 2 , 2 ­ d i p r o p y 1 ­ 1 , 3 ­ p r o p a n e d i o l .

example 2,2­dipropyl­1,3­propanediol,1,3­nonoanediol,2­isopropyl­5­methyl­1,3­hexanediol, Allji­diols investigated can be used when atom­apeorbtion or flame­spectrophotometry is applied, not rtquiring preli­minary separation of boron from the extract.

The extraction of boron with JJ­diols is highly selective (Table 3 ) . On this the methods of determination of boron in steel and ni­kelous alloys, mineral salts, inhibitors of corrosion, conservants of wood and other technical objects are baaed.

These methods warrant an economy of the effortB and time of the chemist ­ analyst.

References 1 . Oshima M. e t a l . / / A n a l . C h e m . 1984. V o l . 5 6 . P . 9 4 8 . 2 . Toei K. e t a l . / / A n a l y s t , 1981. Vol .106 . P .7B1 . 3 . Sato S. et a l . / / B u n s e k l Kagaku. 1984. V o l . 3 3 . P . 8 7 . 4 . Sato S. e t a l . / / T a l a n t a , 1985. V o l . 3 2 . P .447 .

[ 5. Zhivopiscev B . P . e t a l . / / Z a v o d . l a b o » . 1985 .Vo l .51 . P . 1 0 . ! 6. Schwartz J . M . / / I z v e s t i j a AH L a t v . S S R . S e r . c h i m . i g s i . P . 5 5 7 .

' 107

EXTRACTIVE CHROMATOGRAPHY AS AH EXPRESS METHOD OP PHENOL A N D 1 9-29 OIL-PRODUCTS POLLUTION CONTROL OVER FRESH AND SEA WATERS I— I Ya.I.Korenman, A.T.Alymova, T.N.Ermolayeva, S.P.Kalinkina, P.T.Su-chanov, Technologic Institute, Voronezh, USSR, Polytechnical Ins­titute, Lipetsk, USSR

Extractive chromatography relates to a new research method of sepa­ration and analysis of substances. The first full review in that fi­eld was published in 1978 [1] . Sufficient results have been achieved in theory and practice of extractive chromatography applied for sepa­ration of inorganic compounds. In organic analysis extractive chroma­tography is not practically used although first publications on the subject appeared in the 60-th. So far there have been no systematic studies of the influence of different factors such as the supporter properties of solvent extractive agents and extracted organic com­pounds to the effectiveness of extractive processes, separation and concentration of compounds. However- extractive chromatography is ra­ther perspective for such important ecological problems as the deve­lopment of non-reagent methods for sewage treatment, preservation of the environment from high toxic compounds, and can increase selecti­vity in the analysis of organic compounds.

The idea of the technique is that poromeric material with high p

specific surface (200-400 m /g) is covered by organic solvent (effec­tive extractive agent for individual or complex compound separation by static extraction) and the process of extraction and re-extraction runs dynamically. Thin layer of organic solvent increases the rate of extraction, provides high degree of concentration, as extractive com­pounds for narrow zone in column analogous to eorbtion zona in chro­matography. The influence of different factors on the effectiveness of extractive chromatographical column with poromeric copolymers as a bed of extractive agent for detection of organic substances has not so far been studied.

We have studied the influence of chemical-physical properties of extractive agents, sorbtion ability of supporters as well as the qua­lity of extractive agents, column parametres and the size of sorbent particles to the effectiveness of extractive chromatographical column for phenol compounds and oil products extraction from sea and fresh ' «raters. As fresh and sea waters have principal value for human life it seems reasonable to use extractive chromatography for solving the problems connected with water protection from phenol and oil products pollution.

We have developed a number of proceeures for industrial toxical substances extraction from potable, mineral and sea waters. The rele- ••

tions between different extractive agents and their supporters was studied with n­amilaoetate, toluene, tetrachlormethane, 3­n­butil­phosphate (TBP) and octane TBP mixture. Poro, з copolymers of styrene and divinylbenzene were used as the supporters as they contain speci­fic groups in matrix. Copolymers of styrene and DVB swallowing results in keeping larger organic solvent than non­swallowing sorbents. Extrac­ted compound as well as a solvent penetrate the polymer matrix retard­i ing re­extraction. However stronger connection between extractive agent and the supporter decrease its solubility in effluent comparing with static extraction condition. These assumptions were verified ex­perimentally with porous neutral adsorber, с under studying extractive chroraatographical systems used for phenol compound analysis. Correla­tion has been established between the quantity of organic solvents (amylacetate, geptanol, tetrachlormethane, toluene and TBP) and their surface retention, toughness, dielectrical penetration. It s­hows that extractive agent molecules are kept on hydrophobic surface of sorbents based on styrene and DVB due to their adsorbtion forces of different nature (dispersional, orientational and dipole­dipole interconnection). The increasing polarity of specific groups results in better keeping of polar solvents and becomes, for instance, for amilacetate and gep­tanol of a considerable value (1.8­1.6 g/g). To use toluene and tetra­chlorraethane as extractive agents is not reasonable because these sol­vents are ready washed.

By varying extractive agents a great number of sorbtion qualities of polymer can be produced. Thus, for example, covering the surface of neutral adsorbent polysorb­I by octane TBP mixture we observe the in­version of the alkaliphenol sorbtion row (Table 1). Treatment of the polysorb­I by pure TBP results in decreasing of phenol sorbtion in phenol­cresol ­xylenol row. Delution of TBP by octane results ia in­version of the row, i.e. sorbtion in the row is increasing. The treat­ment of the polysorb­I by octane decreases the sorbtion of the polar components. This phenomenon may oe used for separation of non­polar

Table 1. Phenol sorbtion by polyaorb­I, treated by octane and TBP alternative composition (n«3; P»0,95)

Phenols Content of TBP in mixture with octane, % mass Phenols

0 10 20 30 4P 50 100 Phenol 2­Hethylphenol 4­Methylphenol 2.5­Dimethylphenol

5 29 33 41

25 38 47 61

36 64 74 83

46 69 78 87

54 75 81 90

74 86 90 90

72 64 57 60

109

substances in that ease. Sorbents treated by the organic solution ac­quire new properties compared with non-treated ones. Such sorbents we call "combined".

The influence of specific groups to supporters ability to keep or­ganic solvents - effective extractive agents for phenol and oil pro­ducts separation - have been studied on copolymers of styrene and DVB containing nitronitryl and amid groups, and also acetyleted copolymer.

Sorbtion ability of the untreated sorbents depends upon the nature of specific groups and sorbited molecuias. This phenomenon hampers the effective prediction of chromatographical processes while using sucfc sorbents. Treatment of the surface of specific sorbenta scans their own adsorbtlon band, that is why the sorbtion of phenol compounds in­crease in phenol-cresol-xylenol row. The established dependence coin-cede with the character of phenol distribution between water and ami-lacete in liquid extraction. The significance of extractive agent in phenol sorbtion by "combined" sorbents decreases while sorbent polari­ty increases (Table 2).

Table 2. Phenol sorbtion by specific sorbents before CD and after (II) treatment by amyleacetate

Sorbents Polarity

according to Sorbited phenols, %

Sorbents Polarity

according to Phenol 4-Methyl- 3.^-I>imethyl-Kohrscheider Phenol Phe:iol Kohrscheider

I II I II I II Polysorb-H 1255 75 80 60 86 50 90 Nltropolysorb 1078 30 58 44 77 44 88 Hitrilpolyeorb 727 45 65 50 84 68 97 Acetylpolyeorb - 14 48 24 59 65 92 Polysorb-I 206 10 37 17 53 40 57

The capacity of the oolumns was evaluated according to the value HETP (Height of Equivalent Theoretical Plate). In practice the most Important is the character of dependence HETP from the quantity of ex­tractive agent covered the supporter as it is possible to vary this factor for optimization purposes. The value of HETP for the column with 5, 10, 20 am layer (1 sm diameter) was found by means of front curves. It was established that ratio HETP quantity of extractive agent is shown by the curve with minimum correlative to a relation­ship between extractive agent and the supporter 1:1. Such relation results in maximum separation of the extractive compounds from water (95-985S) with the highest concentration coefficient (50-100%).

110

The present research permitted us to work out general recommenda­tions for separation of mieroquantity of phenols and oil products for Express­evaluation of the level of water pollution [2­4% To detect phenols it is recommended to fill the column by neutval porous adsor­bent on the base copolymer of styrene and DVB with araylaoetste (1:1). The most effective effluent is the aqueous solution of chloride alka­li metals alkalined by sodium hydroxide pH 12­13­ The presence of phe­nol was found spectrophotometrieally or gaschromatogrsphically. The last way was optimized by special procedure consisting of microextrac­tive concentration of phenols in effluent. This procedure provides the detection of phenols at 0.5­10~J mg/1 concentration. The time of ana­lysis is not more then 0.5 h. Combined with photometrieal detection of phenols extractive­chromatographical method gives us reliable and time­saving technique of evaluation of the level of water pollution in field. It may be used for all sources of water supply including sea and ocean. These methods are used to detect phenols in the water of the Baltic Sea, the Don and Volga rivers and the Baikal.

To detect oil products the surface of the supporter is covered by N­gexan. The column is also filled by gexan[5J. The contents of oil products in effluent is detected spectrofluometrioally or refractomet­cically. The advantage of the method in question consists in time­sav­ing and precision of analysis.

The adduced examples demonstrate broad possibilities of extractive chromatography in solving analytical and ecological problems connect­ed with different measures directed to water protection from organic toxical substances.

References 1. Extractive Chromatography / Ed. by Brown Т., Gersini G. M.: Myr

Publishing House, 1978. 627 p. 2. Korenman Ya.I., Alymova А.Т., Kobeleva N.S. // Journal of Analyti­

cal Chemistry. V. 39, И1. P. 162­171. 3. Korenman Ya.I., Medvedeva E.I., Alymova А.Т., Zhilinakajja K.I.,

Lorents X.B. // Chemistry and Water Technology, 1986. V. В, Н5, P. 60­61.

4. A.o. N 1064968 / Korenman Ya.I., Alymova А.Т., Kobeleva H.S. Bull. Izobret. N1, 1984.

5. A.c. H 1176239 / Korenman Ya.I., Alymova А.Т., Medvedeva E.I., Zhilinakaja K.I. Bull. Izobret. МЭ2, 1985.

1H

SEPARATION OP GOLD, PALLADIUM, PLATINUM, IEIDIUM AND HHODIUM BY EXTRACTION WITH TOPO АИГ DOSO 9­30

H.HoJski, Department of Analytical Chemistry, Wares* Technical Uniwersity, Warsaw, Poland Solvent extraction methods, especially the ones involving neut­

ral solvating reagents are particularly useful for the separation of noble metals. This is due to the well­known advantages of extrac­tion as separation technique and also to such properties of noble metals as the ability to form complexes with various types of ligands in aqueous solutions, occurrence in various oxidation states and dif­ferences of ligand exchange reaction rates.

A new method of separation of gold, platinum, iridium and rhodi­um, based on the systematic examination of the extraction of halide complexes of noble ootals with trioctylphosphine oxide (TOPO) and di­octyl sulphoxide (DOSO) in different solvents, has been worked out.

The scheme of separation of Au, P4, Ft, Ir and Rh by solvent extraction with TOPO and DOSO

| A u ( I l I ) , P d ( I I ) , Pt(IV) , B h ( I I I ) , I r ( I I I ) ­ 2 К НС1 [ 1

| A u ( H I ) j — 0,1 H TOPO (chloroforms) 1

| P u ( I I ) , Pt ( IV) , RhCHI), I r ( I I I ) ­ 2 II HOI | 1

| Pd(II) | | Pd(II) |

r P t t t V h RhCTITh 3>(III> ­ 2 К НС1 |

| | Pt(IV) | — 0,1 И TOPO (diohloroethaoe)

J Hh(III), Ir(III) ­ 2 Ы НС1 J

HBr, B r 2

| « h ( I I I ) , Ir ( IV) ­ 2 11 HBr |

| | l r ( I 7 ) | •m o . l И TOPO (d lchlorosthane)

1 Rh(III) ­ 2 И HBrj

The initial solution in 2 • EC1 oontslns Au(III), Pd(II), Pt(IV), Ir(III) and Bh(III) In the for» of chloride complexes. Extraction with 0.1 I TOPO solution in оhiorofor» enables the separation of

112

gold. Chloroform, as solvent, diminishes the extractive power of TOPO to such an extent that the extraction of Pd(II) and Pt(IV) is impossi­ble.

Palladium(Il) is extracted from the aqueous pnase after WJC sepu­u­tion of gold using 0.1 M DOSO solution in dichloroethane. In contrast to other platinum metals, only palladium(II) can be extracted efficien­

tly from i. M HC1 medium by means of this extractant. After the separation of palladium, the aqueous phase contains ple­

tinum(IV), iridium(III) and rhodium(III) in 2 И hydrochloric acid. Platinum is separated from this solution by the extraction with 0.1 M TOPO solution in dlehloroethans.

Before the separation of iridium from rhodium, the aqueous phase medium is changed from 2 H Hoi to 2 11 hydrobromic acid with a simul­taneous transition of Ir(IIl) chloride complex into Ir(IV) bromide complex. In order to carry out this operation, evaporation with hydro­bromlc acid in the presence of Br is necessary. Iridium(IV) bromide complex is extracted with 0.1 V TOPO solution in dichloroethane.

Results of separation of Au, Pd, Pt, Ir and Rh by solvent extraction with TOPO and DOSO

I Added (ug) Determined after separation (ug) Au Pd Pt Ir Rh Au Pd Pt I* Rh 200 20 100 100 50 195 1B 93 92 54 200 100 20 50 100 197 99 22 46 92 20 100 20 100 50 19 93 20 92 52 20 50 50 100 100 20 51 48 98 96 50 20 20 200 50 46 21 19 191 48 50 50 200 50 200 49 46 195 44 207

Tbe separation method* presented In this paper has been verified on a nunbt.­ of synthetio «IxtureB of different noble metals rations. The effectiveness of the separation baa bean checked by Beans of epectrcphotoaetr methods and AAS. The results, shown In Table demon­

strate the correctness of the applied solution.

а.зж.382 113

RELATIOHSHIFS BUST ISC BBTWBEN HOLAR EXTRACTIOB PERCENT AUDI STABILITY AHD STRUCTURE OP 2­PROraim"IDAZOLE COMPLEXES ЛГ I——— AQBEODS SOLUTIONS B.Lenareik, R.Cxopek, K.Kurdziel, Institute of Chemistry, Pedagogical University, 25­020 Klelce, Poland

FitTaction of natal complexes by авале of organic solvents from aqueous solutions depends on complex stability, structure of its co­ordination sphere, ligand ­ metal bond type, and on other parameter».

We had demonstrated before that elkyi euhetltuent attached to 2­poeition of 1,3­dia.zole ring made a sterio hindrance. It does not шаек, however, ­the donor sites of 1,3­diazoles so effectively an in case of ­ substituted pyridine derivatives. The effect of 2­alkyliml­dazole alkyl substituent on complexes formation depends on central ion properties.

Hence, we assumed that the differences in stability and structure of the complexes, resulted from the steric effect, can be utilized for the selective extraction of metal cations.

All the measurements /potentiometric, optic, and extraction/ were carried out at 298 К and at the same, 0.5, ionic strength maintained by means of КН0­.

Co­ordination process of 2­n­pxopyllmidaaola with Co(II), Hl(II), Cu(II), Cd(II) and Zn(II) was analysed also by taking into account the foraation curves, i.e. the relationships between mean ligand num­ber value and ligand concentration, /У, at equilibrium. The sets of Д 7 variables, together with the corresponding n values,were calcula­ted from pB values determination for the solutions containing cations of the aetale investigated, and from determined dissociation constant, K a. Precipitates had befti appeared in the solutions containing Zn(II) and 2­n­propylimidaiole, and this fact preoluded the realization of planned investigations for Zn(II).

The determined 5 and DJ values ware used for the formation cur­ves, В • f/pL/, drawing, and for calculation of stability constants for the complexes formed in the solutions investigated. The obtained values of stability constants are listed in Table 1.

It results from the analysis of Table 1 data that the sterlc effect, caused by the presence of n­propyl group In 2­ position of imidazole ring, differentiates the processes of formation of the oomplexes of all five metals investigated by oausing changes in their stability and, In some cases, in the symmetry of their oo­ordlnation spheres.

Than, we studied extraction process of the oomplesee for several organic extraction solvents used. Extraction pxoosss was investigated

114

Tabl» I . Stability constant* of the* »etal coaplexee of 2­n­propy­

lialdaiola in aqueous solution at 296 К

Сantral ion l o g 4 l o g г

l e g 3 l o g ч

H+

C d2 +

C o2 +

1 12 +

C u2 +

8 , 0 2 2.02 i 0.07 0 .80 i 0.03 1.37 i 0.05 3­71 i 0.04

2.60 ± 0.06 2 . 5 0 ± 0.07 2.61 i 0 .02 7 .09 ± 0.03

5.10 £ 0,07 4 . 6 0 i 0.05

9.51 ± 0.09 12.53 ± 0.03 ­

through analysis of a relationship between the aetal extraction coef­

ficient and 2­n­propyllaldalola concentration, L, at equlllbriua In aqueous pbaaa.

In order to elucidate exactly «blob complex froa the aeries of one­core complexes of a given aetal undergoes axtraotlon, we deter­

mined indlridual partition coefficients for each of them. The values of the ooefflclenta, found graphically from the D « z

р

„л

п depen­

dence, are l isted In Table 2. Table 2

Metal Extract ion so lvent "г P

3 h Probable ooapo­a l t l o n of the ooa­p l e x e s undergoing e x t r a c t i o n

Cu(U) 2­Butanol

Bensyl a lcoho l

5.1 i 0 .6

2 0 . 0 ± 2 . 5

/ & o . 4 s 2 72 +

Z6uL4S278+

Co(II)

­о S H

O

mo

o см

Z"i

Э.75 i 0 .08 ( . 0 3 ± 0 .54

/cob 3 s7z +

£6oL3sJ2

+

ca(ii) 2­Butanol

Banayl a lcoho l

!.4 i 0.3

1.68 i 0.54

11.6 ± 1.5

Даь 357г +

/саь­зЛ8

* /0db|si

2 +

It results froa the Table that for Co(II) the third encoeelve ooa­

plex Is extraoted as a sols ooaplei froa the formed ones. The oomple­

xas with four co­ordinated г­n­propylialdasole «oleoulee waa extracted fro» Cut II) solutiona. Cd(II) ana Zn(II) coordination ooapounds acqui­

red hydrophobic properties aoat quickly beoaus* the «xtraotion begins for the second successive ooaplax.

We showed In the last column of Table 2 the probable ooaposltion of the complexes passing to the organic phase, under aasuaptlon that

115

the salts containing complex cation, aolvatad with one or two alcohol molecules, passed into that phase. We could aeeuae that complex ca­tions of cobalt possessed the co»oidination number « 4 and paeudotet-ra^edral symmetry, since absorption bands at ca 17 500 cm were met in spectra of both aqueous and organic phases.

Cd(Il) complexes are much less sensitive to the steric effect than the Bi(ll) or octahedral Co(II) oooplexee. The coordination num­ber was equal to 4 both in aqueous and organic phase.

The complexes investigated differ explicitly with susceptibility to the extraction, especially if bensyl alcohol was used. If the ex­traction with benzyl alcohol and 2-n-propylimidazole were accompli­shed in ?H region 6.6^-7.5, Co(II) could be easily separated from Cu(II), 11(11), Zn(Il) and Cd(II), whereas Ou(II) could be separated at pH <6.6.

116

NUCLEAR PROCESSES

10­1 EXTRACTANTS TOR THE SKPARATIOff OF TRAHSFLOTOKIUH ELEMENTS B.P.Myasoedov, Vernadsky Institute of Geochemistry and Analytical Chemistry, Academy of Soiencea of the USSR, Moscow, USSR The achievements in the development of solvent extraction are olo­

sely connected with the study of the chemical properties of the radi­oactive elements. Back in 1805 Buoholz Pi ] established, that uranyl nitrate was very soluble in diethyl ether, and forty years later Tre­nch researcher Peligot [2] used ether ^o purify uranyl nitrate by re­crystallization. A century later in 1942­44 this extractant was used in the Chicago Metallurgical Laboratory in the first experiments on Plutonium separation, and then for the separation and purification of large amounts of uranium and plutonium in the course of the Manhattan Projeot. The disoovery by Worf ГзД of the capability of tributylphos­phate (TBP) to extract uranium from nitrate solutions was not less important. It formed the basis of the well­known "Purex" process. The further development of extraction applied to radioaotive elements can conventionally be divided into two large periods.

During the first period, which was up to the mid­sixties, the ma­jor efforts of the scientists of many oountries were first of all di­rected to the development of the plant­soale solvent extraction pro­cesses for recovering Plutonium and uranium from nuclear reactor fu­els. TBP proved to be the most convenient due to its perfaot extrac­tion characteristics and commercial availability. However, it has some disadvantages with respect to radiation stability and aqueous­­phase solubility. Nevertheless, up to these days TBP has been wide­ly applied in the industrial technology. The first three volumes of the monography about the properties and numerous fields of applicati­on of this unique extractant have already been published [4J.

The same period is characterized by extensive researoh into the ex­traction behaviour of practically all the radioaotive elements, both natural and. artificial. Various olasses of extraotantst neutral ex­traotants, which form ooordlnationally non­solvatised or solvatlaed compounds 1 cation­ and anion­exohange extraotanta with different ac­tive atoms) mixtures of extractants (synergistic extraction) were used to study the problems of the coordination»! chemistry of the ex­traction prooessea, the composition of the extracted oompounde, the / effect of the properties of the aqueous and organic phases, kinetics of extraction and many other factors. In toe USSR this work has acti­vely been oonduoted In the Veruadsky Institute of Geochemistry and Analytioal Chemistry, the All­Oolon Research Institutes of Inorganlo Materials and Chemioal Technology, the Radium Institute, the Kuroha­tov Institute of the Atomlo Power, the Institute of Physical Ohemis­try and in some other Institutions. •,

118

In the seventies extraction was intensively developed. In that pe­riod the main efforts of scientists, along with the further .develop­ment of the different aspect of the nuclear fuel cycle, were also di­rected to the solution of a difficult task to eztraot and separate trivalent actinides and lanthanides. New extraction systems and met­hods of substance transfer between phases (membrane extraction) were studied; more efficient and selective extraotants (mono­ and polyden­tate ones) were synthesized and studied; synergistic and sorption ex­traction systems were developed. Mathematical methods ware used to op­timize and modelize extraction processes of the radioactive elements separation.

In the recent years great success has been achieved in the study of the extraction of transplutonium elements (TFB) from the highly acidic solutions with the partioular purpose to separata tho long­lived arae­rlcium and curium isotopes from the waste radioactive solutions. Since the atomic stations are developed at a great extend, the volume of such waste solutions is estimated to reach <­ 55000 tons by 1995, and it is supposed to reach up to 125000 tone by 2000. The isolation of the loi^­lived araerioium and curium isotopes from the waste radioacti­ve solutions subjected to disposal is a complicated radiochemical task. Most of the known extraotants do not fit the task, except for the neutral bi­ and polyfunctions!, organophosphorus and phosphorus­nitrogen­containing compounds f5j. The most intensive work rtth the­se extractants is conducted in the USSR (Vernadsky Institute of Qeo­ohemistry and Analytical Chemistry, the All­Union Research Institute of Inorganic Materials) and in the "0& (Argon National Laboratory), several papers have lately appeared in the Japan, Chechoslovakia, Franoe, India.

More than 60 phosphorus and phcsphorus­nitrogen­oontaining neutral bi­ and polydentate extraotants have been studied in the Vernadsky Institute. These reagents were synthesized by the laboratory of acade­mician M.I.Kabaehnik in the Institute of Blementoorganic Compounds of the USSR Academy of Sciences. We studied the effsc of their struc­ture, i.e. the conformation of the moleoule, nature of the substitu­ents at the atoms of phosphorus, nitrogen, in the methylene bridge; bridge composition, ­ upon extraction capability, selectivity and solubility of these extraotants Гб,71. The first group of extraotante inoludae the phosphorus containing bi­ and trifunctional extraotants with the linear bridge between the P=0 groups and their derivatives, in whioh the hydrogen atom in the bridge is substituted by another atom or radical. The exxtraotanta of this group have the highest ex­traction capability and a high stability to radiation. With the use of these reagents the methods of the extraction isolation of TPE fro» highly aoldi" salt­containing solution» with simultaneous con­

119

centration to 100 times have been developed. The extractants have a typical group character and they are not тегу selective Гб_). The se­cond group of extractants includes Ы­, tri­ and tetrafunctional phosphorus­containing eztractants with bridge of ortho­ and metazyle­ne, mesithylsne, durrole and other methyl­substituted aromatic compo­unds bonded the functional groups. The undoubtful priority of these extractants is the high selectivity, revealed л relation to Differ­ent elements depending on the location of the functional groups in the aromatic compounds of the bridge [V[. The phoephor'is­aitrogen­oon­taining extractants, (dialkyl(diaryl)£dialkylcarbamoylmethylphosphl­nej oxidee have a high extraction capability. Their priority lies in their higher solubility iu organic) solvents and in a relative simpli­city of synthesis. These extractants are not selective as the alkyl­enediphosphine dioxides | 7"|.

Along with the numerous priorities of dXyxidea and earbamoyls they have essential drawbacksi they cannot practically be solved in alipha­tic solvents; their oomplexes with metals cannot sufficiently be sol­ved in the avail. ,bla solvents. One of the ways to eliminate these drawbacks is to add monodentate reagents of the ТВ? type to the solu­tions of bidentate extractants. It was proved that TBP eliminates the formation of the third phase during the extraction of macroatnounts of elements, besides, it leads to the non­additive inorease in the TPB distribution ooeffioienta (the synergistic effect). We have also pro­ved that IBP addition makes it possible to usn aliphatic diluents as well»

Per bidentate organophosphorua extractants, a known "aryl effect" should be mentioned, which oonsists in the increasing of extraction capability of reagents when the alkyl radicals at phosphorus are sub­stituted by the leas electronegative ones, i.e. Cheryl radicals, for the first time the "aryl effect" was only observed during the TFI and lanthanidea extraction by alkylenediphoaphine dioxide»| ther it was discovered in the course of extraction by dlaryl(dialkyl)£dlalkyloar­baBoylswtbyl­phoiphineJoxidee f<f). The "aryl effect" was believed to oocur durJ ..g the extraction by bidentate extractants with a methylene (or vinylene) bridge between the functional groups, and to disappear when the bridge is prolonged to ethylene., fe have reoently discovered the "aryl effect" in the oourse of TPI tau europium extraction fcy the extractions with meta­xylene used as bridge binding too functional P­0 groups (Tif.D.

tl Valine solutions also eeea effioient for aotinide isolation and concentration. The extraction of metals from euoh solutions has not praotioally been stalled yet. Our investigations have proved the capa­bility of aotlnldes and «any other eloenta to be extraoted from alka­

1»

line solutions Is the presence of the complexing agents by various classes of extractants: Aliquat 336, amines, alkyl­pyrocatechols, CDOP), alkyl­derivatives of amino­alcohols (ЛА), andJi­diketones £1CQ. Several studied extractsnts, especially alkylpyrocatechol are characterized by a high extrac­tion capability in relation to TFE. They are efficient for the extraction of these elements from solutions with the concen­tration of sodium hydroxide end carbonates of alkaline metals of being about 5­6 mol/1 (Pig.2). The efficiency of the extraction isolation of TUB is affected by the nature of completing Uganda, oapable to keep elements in alka­line solutions in the soluble form. The role of completing age­nts is sometimes reduced just to keep the elements in the soluble form in alkaline media (иС­оху­carbonic acids, carbonate­ions). In other cases the oomplexing agents (aminopolyphosphonio acids) are transfered to the organic pha­se together with the metal. There­fore depending on the conditions,

a y* t ЗД / У Л a / 7 9

ГЛ

f/ / 0 V Я/ J

•1 /f. ­3 ­2 log[K] ,M

,g.1. "Aryl effect" in extrac­tion of Am(III) from 3 M HNOj

M&. 0,b . 2.

1,5 " 4 W a 0 H

] Iй

Extraction of Am(III) from alkaline solutions by dif­ferent extractants

elements oan be extracted from alkaline and carbonate solutions in the form of ionio associates, the anion part of whioh oontaine either the hydroxooomplezea of the corresponding metale or their compounds with the oomplexlag agent. They oan also have the form of chelates satura­ted and hydrated compounds. The methods of extraction and extraction chromatography [_ 11J in alkaline and oarbonate solutions are used to solve a number of practical tasks, firstly, under oertain conditions quantitative separation and concentration of elements oan be conduc­ted with the help of all the tested extractants. Seoondly, the con­centration of trivalent aotinldes with their simultaneous separation from the elements in other oxidation states and iron can b:. aohieved in these systems. The selectivity of the extraction separation oan

121

be Increased In two wayet by varying the alkaline concentration of the media and the contact time of phases. The best method for the trlvalent emerlclum and ourium separation in carbonate asdis Is ex­traction chromatographyi since separation la improved due to the dif­ferences both in the values of the mass transfer coefficients and In the equilibrium process»

The membrane extraction and the use of ldnetlo factors open up new opportunltes in the separation of radioactive elements. There are seve­ral ways to organize the process i.e. extraction with membranes, impre­gnated membranes and the «nuleion membrane extraction. The main priori­ties of the membrane extraction as compared to the liquid extraction consist in the low expenditure of expansive extractants. as well as very high degrees of isolation and ooncentratlon. In some oases the membrane extraction processes are more seleotive. The drawbacks of the method consist in the slow rate as oompared with the liquid extraction and besides in a number of cases it is only a periodical process. It is also more technically sophisticated. The electric field considerably acoelerates the mass transfer of the material through the membranes. Membrane extraction is most efficient for the processing of extremely diluted solutions;, inoluding the radiochemical wastes. The most effi­cient method of isolation and separation of elements is the combina­tion of membrane extraction with the use of redox reaotione. A method of emulsion membrane extraction with HDBHP was developed in our labo­ratory for the aotlnlde separation from ooncentxated ИГРА solutions. Up 96* of actinide are transtered to emulsion in the course of the extraction, while not more than 1st of americium can be isolated by one contact, when liquid extraction is applied.

The methods of separation using various oxidation states of the se­parated elements are the most seleotive Pt2~]* Thus the conditions of the tetravalent anerioium extraction by secondary amines from nitric and sulphuric aoids solutions in the presenoe of phosphotungstate ions ware found in our laboratory for the first time Г13]. к high de­gree of amerieium (both mioro and macroamounts) separation from ourl­um and other trivalent elements is achieved. Tetravalent berkelium can be quantitatively extracted from mineral acid solutions by many extract ant si organophosphorus ones and^­diketones; and in the pres­enoe of heteropolyanions they oan be extracted by primary, secondary, tertiary amines and quaternary ammonium bases.

The pedtavalent amerioium can be isolated from solutions of various composition by extraction with ТТЛ, 1­phenyl­3­methyl­4­bensoylpyrazo­lone­5 (PHHP), ШЯНС and ammonium pyrrolidlnedithlocarbamate. With the use of HBP and HXEHB the maoroamounts of pentavalent aaerioiun were extracted for the first time, its absorption spectra in the organio media were taken, and the data on its stability were obtained. The "i

122

separation factor of americium and ourium in the course of the pen­tavalent americium extraction by ИИВР and HDBHP makes (>­j­6)­103

. Using the mixtures of eztractants(Fl(BF or ТТЛ with ТОЮ, picrolonic acid with eulphoxides from low acidity solutions) the trivalent acti­nides are extracted with high distribution coefficients, while the pentavalent americium remains in the aqueous phase.

Thus a conclusion can be drawn, that extraction has played and is still playing an essential role in the development of the chemistry of radioactive elements, particularly actinldes. On its basis highly efficient methods of isolation of radioactive elements from the mate­rial of various nature, both technological and analytical, have been developed by this time. The extensive experimental data accumulated in this field has made it possible to draw important conclusions on various aspect? of the theory of the extraction processes. It can be stated that the use of new efficient and selective extractants and systems, new ways of conducting extraction prooessee and knowledge about their kinetics will give important information about chemical properties of radioactive elements and prompt efficient ways of their isolation, separation and concentration.

References 1. Bucholz C.F.//J. Chem. von A.F.Gehlen. 1805. Vol,4(17). P.134. 2. Peligot В.//АПП. Chlm. Phys. 3rd Series. 1842. Vol.5. P.5­47. 3. Worf J.C.//J. Am. Chem. Soc. 1949. Vol.71. P.3257­3258. 4. Schulz W.W., j.D.Mavratil (Eds.).Science and Technology of Tri­

butyl Phosphate. Floridai ORG Press. Inc. Roca Raton, 1984. 5. Schulz W.W.,Havratil ,T.D. // Recent Development in Separation

Science /Ed. Mormon N. Vol.7. CRO Press. Inc. Roca Raton, 1983. P.31.

6. Ohmutova U.K. et a W / J . Inorg. Nucl. Chem. 1980. Vol.42. P.897. 7. Myaaoedov B.F. et al.//Solv. Ext. Ion Bxoh. 1986. Vol.4. P.61. 8. Myasoedov B.P. et al.//3olv. Ext. Ion Exoh 1983. Vol.1. P.689. 9. Medved1 T.Ya. et al./Azv. Akad. Hauk SSSR. 3er. Khim. 1981.

Vol.9. P.2121. 10. Karalova Z.K. et al.//Zh. Analyt. Khim. 1984. Vol.39. 1?. 119­11. Karalova Z.K. et al.//Radlochim. 19B5. Vol.27. P.751. 12. Mikheev H.B.,Myasoedov B.P.// Handbook on the Physios and Chemistry

of the Actinides/Ed. A. J.Priman, C.Keller. New York:Elsevier, 1985. P.347.

13. Myasoedov B.P. et al.//J. Less­Common Met. 1986. Vol.122. P.195.

123

HBW ACTIHIDBS EXTRACTAHTS POR THE TOOLBAR TOBLS REPROCBSSIHG 10­2

C.Musikas, H.Germain, Commissariat a l'Energie Atomique ­IRDI/DERDCA/DGR/SEP/SCPR BP N6 ­ 92265 Fontenay­Aux­Roses, ftranoe To decrease the coat of the irradiated nuclear fuels reprocessing)

it is necessary to simplify the process and to limit the amounts of radioactive wastes. The use of new extractants could fulfill these requirements. Completely incinerable extractants and nonbasic solvent regeneration are favourable to wastes limitations. Highly selective actinides extractants towards fission products are necessary to sim­plify the chemical processes. The caboxylic acids amides are good can­didates for such programm Zl ­ 37. We will survey the extraction che» mistry of H, N­dialkylamides which are a promising alternative to TBP and of N,N'­tetraaUiymalonamidea Z4> 5.7. They cje good extractants of the actinides ions including trivalent ones and could be an alter­native to the polyfunctionaa extractants.

N,N­dialkylamides properties. The selection of N,N­dialkylamides as a function of the solubility of nitrato uranyl adducts in alipha­tic hydrocarbons has been carried out. It has been found that the amides containing the radical 2­ethyl hexyl as the nitrogen substi­tuents form adducts which ar. highly soluble in hydrocarbons like normal or ramified dodecan (ГРН). Nitric acid forms the adducts HJJO­jLg, HU0,L and (HN0,)2L There L represents an ami die extractant. IR spectroscopy indicates that L and Щ 0 , are linked by hydrogen bond.

Depending of the aqueous acidity, the extraction of uranium (VI) occurs via two mechanisms /Г4_7 Uo|

+ + 2KCC + 2E ­»TO.ДнбТТЗо from low acidity media and ио| + + 3N0j + H + + I ­» UO^lHOjI^HX fr ., hig­her acidic solutions. These two cases can be distinguished ry the U­V spectroscopy of the organic phase. The spectra of the two U(VI) species are shown in Pig.1.

Plutonium (IV) extraction is sensitive to the nature of the C»0 substituent. The difference between linear (DOHA),

с 2 н 5

с 5н 1 1с­о­к(сн г­сн­с„. э) г, and branched (ЮТА),

CH 3 J^g CH.J­C­CH2­C­O­H(CH2 TR-O^)2,

hexanamides is illustrated b. the

124

Fig.I • U(VI) U.V­via. spectra (1M amide into TPH. Acidic and neutral eolutions)

LHN0 3l a q.mol/l

Pu(IV) DOHA о DOTA о into TPH and aqueous HNO3 solutions

PuCIVJ distribution ratios plotted as a function of HNOj concentra­tion in figure 2. U(VI) distribution ratios do not vary so much by changing the C=0 substituent; so the branched substituent on the carbonyl offer possibilities of (U<VI)­Pu(IV) coextraction from highly concentrated nitric acid and U(VI)­PuCTV) separation at lower acidities (Figure 3).

This branched hexanamides give better fission product separations as shown by Figure 4 extraction curves.

Radiolysis and hydrolysis are important for practical applications in the field of nuclear fuels. We investigated the stability of 1M DOHA in И И mixed twice a day with one volume of aqueous 4И HKO, or C.5H HNO,. It can be seen Figure 5 that the disappearance of DOHA is correlated to the hexanoic acid production which io the main degrada­tion product. The hydrolysis of DOHA and TBP are compared in Table 1, it can be seen that the two extractants are rather stable. Small dif­

10 Utvll.OOlA

UlVH.DOMA

1 1­1

1­Э

/// VPvUVl.OOTA

J 0.1 /y\ ••

/ VfuIIVl.OOMA

,o­3 _/­, ,_ • ' ' ' ' F^U

Fig. 3 . Distribution of Pu(IV) and U(VI) between DOTA or DOHA *nd aqueous HNO,

0.1 1 [HW0 3] a ( ;, mol/1

­1

ь

DOTA

­2 Hu Z r / /

­3 —Ль

Ftg.4 . D between 1M DOHA (a) or 1M DOTA (b) in TPH and aqueous nitric solutions

125

Fig.5 • Of X 1H DOHA radiolysed and bexanotc acid yields dose rate in the ргеяепсе of 4N and 0.5N mo,

ferences In the figures arc probably due to analytical incertainties and a better comparison necessitates improvements in these technique.

Table 1 Hexanolc acid (HA) and N,H­di(2­ethyl)nexylamine (DBA) yielda in the hydrolysis of DOHA at 110°­115°C in the presence of aqueous HBO, (stirred solutions) and comparison with TBP hydrolysis

HHO,, mol­1

1

Time, houra

DOHA, mol­1"

1

HA, mol­l"

1

DBA, mol­1"

1

IBP, mol­1~

1

HDBP, mol­1"

1

0.5 4 0.95 0 0

4 4

1 4

0.95 0.94

0 0.015

0 0

0.5 5

4 4

1.08 1.08

0.0008 0.0047

A mixer­settler test of a first reprocessing cycle has been carried out using 1H DOHA la TPH as extractant Although DOHA is not the best choice for U(VI)­l?u(IV), partition at low acidity has been tempted in order to precise the problems linked to U­Pu non reductive partition. The flow­sheet for this experiment ie shown schematically in figure 6 and the main results are oontalned in Table 2. The partition battery was not 100% efficient beoause of the slow Pu(lV) baek­extraotion kinetics which was too low for the liquids flux used. All the other operations were satisfactory and show that N,X­dialkylanid«s are promising extractante for the future.

Polyfunctionnal amides properties. The use of N,N­dialkylamid«s as an extraetant for the nuclear fuels reprocessing suggests to In­vestigate the extractive properties of polyaunides in order to extract also the trlvalent actinldes lone contained in the radioactive wastes solutions. The pioneer work of Slddall £ b j showed that Oe(III) was ,$

я " Dose^

6 0

0o (H Rad)

126

Solvent ООН*

ntaai h*' m*m TM

Sttv«nt:0.24lrr' ООН* Н into ГРК

г r1

MHO» C.1H ucvi) ai*»i­' Pu[N) M t « 1­1 Zr(W) tract* AmdlDlrocM

Scrat­.OOMlrr1

HNOi »N HNOj 11H 0.00*81 h"

1 HHO, aiN O.Wl­h­1

нно»аот

g'ig­fe • Fl°w sheet for a f i r s t reprocessing cycle using IM DOHA in ГРН

Table 2 Results of a mixer settler first cycle reprocessing separation using Ш DOHA in TPH as solvent

Operation Element Feed Raffinate (g.l'

1

) (% of feed in eac» step)

Extraction Aqueous phase Aqueous phase U(VI) 251.8 0.015

Pu(IT) 2.48 0.14 Zr(IY) Traces 99.5 Am(III) Traces 99.5

Partition Organic phase Aqueous phase U(VI) 81.7 0.37 Pu(IV) 0.805 95.5

Back­Extraction Organic phase Aqueous piiase U(VI) 44.7 99.9 Pu(IV) 0.0198 95.2

poorly extracted by H.N'­tetrabutyloxalamide, N.H'­tetrabutylgluta­ramide and H.N'­tetrabutyleueoinamide. In a first step we selected malonamides proper to extract Am(III) from aqueous nitric acid £(>, 7 J. It has been found that tr.H'­dimethyldialkylmalonamideB are the beat Am(III) «xtractant because of their chelating properties. It was found further that the substitution of a methyl hydrogen by an alkyl ot oxyalkyl radical inproves the extractive behavior of dlaad.­

127

Fin.7 . Am(IIt) dis tr ibut ion (g ) . rat ios between aqueous HNO3 and various 0.5M diamide (CH.C H N С 0) CHR in ГЛ with R ­

1) H, also c J L , insteaS of C,Hr

2) C 6 H I 3 3) C 2 H 4 0C 2 H 5 ' 4) C 2 H 4 0C 2 H 5 ­OC 6 H 1 3

1 2 3 4 5 6

[HN0,]„ q mol l"

des, This feature is illustrated by Figure 7 plots. Dianides show good prospects as aotinidea extractant from radioctive wastes. Others metal ion extraction are given at this conference Ей J*

Conclusion. Hew non­organophosphorus extractants are appealing in the nuclear fuels reprocessing field because of their oomplete incinerability and the nature of their radiolytlo products which do not alter the process. Their extraction cbamisiry presented here is encouraging to pursue the research and development since several advantages over the PURE! process can be sees immediately. Among them are non reductrive U­Pu partition; better fission product decontami­nation, better 11(71} back­extraction ... Solvent regeneration which is not yet known will be the next important step to corroborate the interest of amides,

Keferencea 1. Siddall T.H.III//J.Phya.Chem. I960. Vol.64. P.1863. 2. Gaaparini G.H.. Groesi G.//Solv.Ertr.Ion Bxoh. 1986.Vol.+.P.1233. 3. Uusikas C.//Tc appear in Sep.Sci.Teehnol. 4. Descoule H., Musikas C.//J.bess.Coan.]fetals. 1986. Vol.122.P.256. 5. Siddall T.H.III//J.Inorj.Wucl.Chen. 1963. Vol.25. P.883. 6. Muaikee 0., Hubert H.//ISBC83.Proceedings. Denver,1983. P.449. 7. Husikae C., Hubert H.//3olv.Bxtr.Ion Exch.(in press). 8. Cuillerdier C., Hubert H., Hoel P., Hueikaa 0.//I3BC'SB.Papers.

Koscow, 1988. Vol.1.P.291*

128

i

BEHAVIDUR DF URANIUM AND PLUTONIUM ON CHROMATOGRAPHIC COLUMNS I I 10­3

CONTAINING AMIDES AS EXTRACTAN1S * ' W.Smulek, H.Al­Rawi*, L.Fuks and M.Borkowski Department of Radiochemistry, Institute of Nuclear Chemistry and Technology, Warsaw, Poland

•Nuclear Research Centre, Baghdad­Tuwaitha, Iraq N.N­disubstitute aliphatic amides are promising extractants for

such actinides as uranium, neptunium and plutonium. Dn the other hand these amides are poor extractants for fission products. These valuable properties indicate that amides could find application in the nuclear industry. From this point of view amides can be considered as alter­native extractants to tri­n­butylphosrnate (TBP).

Advantageous properties of amides and extraction experiments with amides have been described in the literature Q.,2]. Good extraction properties of amides imply that they could b­ used in extraction chromatography for separation purposes [2].

The present work aimed at examination of the behaviour of some actinides on chromatographic columns containing newly synthesized N,N­dihexylcyclohexylbutyramide (denoted as DHCHBA) and N,N­di­sec­butylcyclohexylbutyramide (DSBCHBA) as extractants. Special attention was focussed on the separation of the various oxidation states af actinides by these amides. In acidic aqueous solutions light actini­des attain the following four oxidation states: An(III), An(IV), An(V), An(VI) . From among these oxidation states Np(lII) 50 and Pu(V) are not present in ordinary aqueous solutions since they are un­stable and easily oxidized by the air.

The existing literature extraction data for plutonium and neptunium are, in general, at variance. This is caused by the difficulties in removing Np(V), Pu(III) and colloidal plutonium from aqueous solutions. The comparatively large experimental errors associated with the extraction data obtained, e.g., for hexavalent plutonium have been 2 4MB* indicated by McKay [з] . ° ­

5 M H N D

3 It is to be emphasized that column ex­ Fjg.l. Separation of fission traction chromatography applied in this products and uranium on the work enables to obtain true extraction amide columns

129

data for actinide elemen's because each oxidation state of these elements can be separated as an individual chromatographic peak.

Experimental The chromatographic columns used throughout the work were prepa

red applying the procedure elaborated earlier in our Department [4,5]. Silicon treated kieselguhr "Hyflo­Super Cell" was mixed with an ap­propriate amount of the amide extractant (20X w/и), diluted with a volatile diluent, e.g. hexane, the latter being carefully evaporated. The length of the column bed was about 10 cm and the column diameter was ­0.3 cm. The volume of the samples introduced into the column

­2 3 was about 10 um . The eluting agent was 0.5M HNO,. The flow rate of

­3 3 2 ­ 1 the eluant was about 9*10 uro cm»min . All the experiments were car­ried out at room temperature,

Z i 0.5M HNOj

Fig.2. Separation of Am(III) and uranium on the amide columns

The alpha activity of the studied ac­

tinides was measured using a liquid scintillator consisting of a mixture of 5 g dm~ 3

PP0 and 25% (v/v) of HDEHP in toluene. Identification of the particular }f­ and o^­energies was performed with the help of Y~­ and o(­ray spectrometry using a multichannel analyser (0RTEC­

Model 7150). Uranium was determined spect­

rophotometrically with the Arsenazo­III method.

Oxidation of neptunium and plutonium to the hexavalent state was performed by means of NaBr0 3 in.l.OM HNOj. Nh^OH­HCl was used as the reducing agent for Pu(III) No catalytic reagents were used. The reduction process lasted for about 24 hrs at room temperature. It should be mentione that U(VI), Np(V), Pu(IV) and Am(III) are the most stable valence states of the actinide ions in the aqueous solutions use

The amides used in this work were synthesized in the NRC Baghdad. The amides have the following characteristics:

n, 293 0 d2 9 8

( k g m­ 3

) b.p.(K)/Pa

DHCHBA 1.4700 944 DSBCHBA 1.4735 952

the purity of the amides was better than 96*.

429­435/400 407­409/800

1Э0

The distribution ratio, 0, of the actinides were calculated from the elutian curves according to the formula [б].

0.5M HNO

Fig.3. Separation of plutonium in different oxidation states on the amide columns

where: V ­ total retention mr

volume. V ­ mobile phase volume, ' m

V ­ retention volume, s ­ statio­nary phase volume. The reproduci­bility of the results was about 5% Results and discussion

From the previous studies [l,2] i

+ is known that the N,N­disubsti­tuted alkyl amides are not ex­tracting fission products from nitric acid solutions. This fact enables clean separation of fis­sion products from uranium, as can be seen in Fig.l., where 1 3 7

C s and 1 4 4

C e were separated from 0.1 mg of uranium. Similarly to the tervalent lanthanides (Ce, Eu) also AmClII) can be separated from uranium, and an example of such a separation is presented in Fig.2. The sepa­ration of the various valence states, of plutonium and nep­tunium on the amide columns can be seen in Figs.3 and 4, respectively. Pu(III) is not retained on the column and is eluted in the free volume of the column.

The calculated distribu­

tion ratios of the particu­lar valence states of the ac­tinides are listed in Table. It is seen that di­sec­butyl­cyclohexylbutyramide extracts the hexavalent actinides much F i

9 ­4

/ Separation of neptunium in dif­better than does the OHCHBA f e

" n t oxidation states on the amide columns

131

extractactant, which on its turn, is more effecient for Pu(IV). The better extraction of the hexavalent actinides by DSBCHBA, as compared with OHCHBA may be due tD the increased electron density an the carhonyl oxygen caused by the sec-butyl radicals. Since the extraction mechanisms of metals by amides and TBP are similar therefore, far a comparison, also column extraction data, under compa­rable experimental conditions, for TBP are given in this Table.

It is worthy of notice that the amide columns demonstrated a good characteristics and stability. The height equivalent to the theoretical plate CHETP) ranged usually from 0.2 to 0.5 mm. The amide columns could be used for several runs.

Distribution ratios for actinides in different oxidation states calculated from column experiments. 0.5M HNO, was used as the eluant

Extractent U(VI) Np(VI) Pu(VI) Pu(IV) AmCIII) 1 5

QHCHBA UIBCHBA

9.62 20.3

8.55 16.2

1.43 5.16

14.3 6.80

0.005 0.016

TBP 9.92 9.91 1.65 12.7 1.27

1) From liquid-liquid extraction with undilute amides and TBP from concentrated nitric acid.

References 1. Gasparini G.M., Grossi G.//Sep.Science and Technology. 1980. Vol. 15.

P.825 CIncluding all references mentioned there). 2. Smuiek W., Salman S.M.//Proceedings of Conference "Studies on repro­

cessing Df irradiated nuclear fuel and disposal of radioactive wastes", Piestany, 22-25 April, 1985. Czechoslovak Commission for Atomic Energy. 19B5. Vol. III. P.161.

3. Magon L., McKay H.A.C., Wain A.G//J.Inorg.Nucl.Chem. 1974. Vol.36. P.5B49.

4. Fidells I., Siekierski S.//J.Chromatog, 1965. Vol. 17. P.542. ' 5. Smulek W .//Nukleonika , 1969. Vol. 14. P.521. 6. Markl P., Schmid E.R.// Extraction Chromatography /Edited by

Brown T. and Ghersini G. Akademiai KiadD Budapest, 1975. p.65.

132

*

BI­ AND P0LYDEN2ATE ORGANOPHOSPHOROUS COMPOUNDS AS Г EXTRACTANTS FOR ACTINIDES PROM LIQUID WASTE (USE OP EFFECT | _ 1 0 " 4

OP ANOMALOUS ARTE, STABILITY INCRBASE

A.M.Rozen, Z . I . N i k o l o t o v a , N.A.Kar tasheva , All­Unlon Research I n s t i ­

t u t e of I n o r g a n i c M a t e r i a l s , Moscow, USSR

At p r e s e n t i n t h e VST. ' . USA and o t h e r c o u n t r i e s b i ­ and p o l y d e n t a t e organophosphorous e x t r a c t a n t s have aroused i n t e r e s t due t o t h e i r a b i l i ­

t y t o e x t r a c t r e s i d u e s of a c t i n i d e s from Purex l i q u i d waste withou t i t s s p e c i a l c o n d i t i o n i n g [i—5j . We have s t u d i e d a a e r i e s of b i ­ , t r i ­ and p o l y d e n t a t e e x t r a c t a n t s : diphosph ine d i o x i d e s Ph 2P(0)CH,(0)PPh^(abbrev. 4Ph); PhgPloiCHatOPoct CaPhZoctbctjPlOjCHa^Poct ^Dtt toCPWSCH^WPbej iflf); P^PtOjtCHJ^PP^MPh^Hiy^^W^tH^OjPPhj^PblCH^^^jPIDK^^HWPh^­cU (4Ph(C.4=CH)cLs);carbamoylphosphine oxides OC.t2P(0)CM2(0)CMBll2 ( 0 c t 2 / B t 4 2 ) ;

РКгР(0)СНг(0)СМ8иг(РЬ2/8иг);­ЬЕ2Р(о;сНг(0)СМ8иг (toE 2/e« 2);

PhjPjjNttij (Рп2/РЬ/ви2); "here Ph = C 6H S 'Ы = CHjCaH/,; Oct = C 6Hi 7

tridentate extractants (0]_ p _Ю) (TDE­R)' w h

ere R="CHs and Ph

as well as an acid polydantate extractant о ОН ' where n=2­4;R=C^H9CH(CjH5)CHeis poly­2 ethylhexylphosphonitrile acid (P2EHPNA). With these extractants extraction can be further improved through the use of the effect of anomalous aryl stability increase (AASI) found by us [1]. The essence of the effect is the following.Our experiments showed, e.g. [6,7j, that when extracting with neutral mono­dentate extractants the extraction ability (EA) was reduoed with an in­crease of the substituent electronegativity (X) in an extractant mole­cule. This is explained by the fact that complex form by an acceptor­donor mechanism (the donor is the oxygen extractantj­ as X grows, the substituents draw off the electron density from the oxygen, its donor ability decrease. Correspondingly, the complex stability and, hence,EA are reduced, e.g. in the phosphine oxides R­jPO­phosphate (80),P0 seri­es, since XR=2.0 and Хд 0 =2.5.

However, on extraction with diphosphine dioxides and carbamoylphos­phine oxides it turned out tnat the replacement of alkyl substituents with more electronegative phenyl (X=2,3) or tolyl ones the Am distri­bution coefficient (^д™) did not only reduce Dut, on the contrary, im­proved it by two orders (fig.la , 0.01 mole/j. solutions of dioxides in dichloroethane (DCE) fl­3]. The same is observed for the extraction with 0.05 mole/ solutions of carbamoylphouphine oxides in dichloro­ethane (fig.1b). The similar effect is observed in the extraction of uranium (VI) ar1 Pu(IV) with solutions of dioxides and carbamoylphos­

133

phine oxides in diohloroethane (fig.2). However, one should distingu­ish between the true AASI (when both the distribution coefficients and 1 0 a o i A m

12. " ^ CHh/a,], mo («/I

equilibrium constants grow, i.e., the com­plex stability in­creases which is ob­served in the ext­raction of An(III) and Ln(III)) and the visible AASI (when the distribution co­

CHNO,I , w

Fig,1 oiflm­ acidity relation efficients grow through the acid extraction weakening and the growth of the free extraotant concentration while the equilibrium constants de­

crease) which is observed on the U and Pu (¥1 and IV) extraction. The aame effect is observed when

e other aryl groups, Flg.a.Otu.oCPu.­ acidity relation e

­£

" t 0 l y l OTeS ,

were introduced (fig.1,2). Studies were carried out to find out the reasons of this anomaly. It turned out that in accordance with the ex­pectations the donor ability of phosphorvl oxygen of an isolated ligand decreased with the introduction of aryl groups into an extractant mole­ j cule which is confirmed by the growth of the PO IR frequency and a dec­rease In the extraction of monodentatically coordinated nitric acid when going from 4 Oct to 4 Ph. The stability increases the moment a eixmember cycle closes and bidentatically coordinated metal complex forms. This effect is associated with the chemical nature of aryl groups: when the latter are replaced by groups of the same electronegativity but of a different nature, e.g., С1(СН 2) г, it disappears. When aryl groups are separated from phosphorus with a CHj­group the effect disappears;the same takes place when the methylene bridge is replaced by the ethylene and propylene ones and the effact is restored when the vinylene bridge is introduced (fig.1c). This means that the electron density is delo­calized from the aryl groups to the cycle. It can be also assumed that s system of conjugated bonds forms, the cycle is "aromatized". Whate­ver the causes are, the effect of AASI is very useful for the extract­ion of actinides from liquid waste.

The main difficulty in using the above compounds results from the poor compatibility of those extractants and particularly their oomple­xes (with lanthanides) with hydrocarbon diluente. One of the ways is to introduce alkyl substituents into benzene nuclei, i.e., to use di­ ^ phosphine dioxide 4 tol [в] and carbamoylphosphine oxide tolg/bu2 [9] .

134

tHNO­sl.m PiguofcMq relation

^Cau. mole/1 Pig.4. Eu extraction ieoterma

oL ftm

6 D»№A"*Wi Pig. 5. TDB extraction (flm, curve 1­Bu)

However, these compounds are also poorly soluble in paraffin diluents. To extrac, from solutions close in their composition to liquid waste (e.g., 3 mole/1 HMO, and 10g/l Eu) the required extractant concentrat­ion is 0.3­0.5 mole/1. To provide the compatibility isoparaffin with 20$ octyl alcohol and 1 mole/1 TBP are needed. Due to the alcohol d. of Am greatly decreases. To increase e<Am trichlorobenzene (TCB) as a diluent was used. The isotherms of the micro Am extraction in the pre­sence of 10 g/1 Eu with 0.3 mole/1 solutions of 4 tol and tolg/bUo in TCB were taken (Fig.3). The distribution coefficients remain rather hihg, i.e., at the phase flow ratio 1:1 one­two stages are sufficient for the full Am extraction.

Extraction of europium nitrate maoroamounts with the indicated bi­dentate axtraotants was also studied (Pig.4). It is established that along with the known tri­ and disolvates found by us earlier in the ex­traction of Am and Eu microamounts monoeolvate is also formed (y equil exceeds 0.15; i.e., Z = y/L =»0.5 whioh points to the presence of some solvate 1:1), i.e., solvates Eu(N0,),­L where q=1,2,3 are formed.

The Am and Eu extraction with two tridentate compounds has been stu­died (Pig.5,a, 0.05 mole/1 in DCE). The replacement of methyl radioal by phenyl one results in a significant improvement of the extraction; the true AAST is observed. As compared to the bidentate analogue Fh/ Fh/bu, a considerable extraction improvement is observed whioh points to a tridentate coordination; this is also confirmed by Ш­spectra. A strong effect of a diflaent on extraction with TDE­Ph ie noted (Pig.5b).

135

"he advantage of the extractants discussed consists in the extract­ion of actinides (III) at the level of An (IV,VI) [l] as well as in the possible simultaneous suppression of Ш Ю , extraction and improve­ment of actinides extraction through the introduction of aryl substitu­ents. Apart from the difficulties associated with a diluent the dis­advantage is a lov; selectivity; following the co­extraction of An and Ln special operations are required to separate them.

Of the class of polydentate extractants poly­2 ethylhexylphospho­niirile acid (Р2БНРМЛ) suggested in pcf| was studied. As it can he se­en from fig.6 Р2ЕНРНЛ as distinct from D2EHPA effectively extracts ame­ricium from a highly acidic medium: =^­^=7 at the P2EHPNA=2.5$ (in do­decane) and ­Х­яши^ mole/1. The composition of the resultant complexes

was established; the P2EHPNA extraction cons­tant for Am is 10 times higher than D2EKPA one. 5?his is likely a result of the linear cha­racter of P2EHPNA polymers (no dimerisation that fcindd P0 and OH centres), transformation to the imide form, containing only OH groups and the possible hi­ and polydentate coordinat­ion of metals.

л EjCwIOj,! Pig.6. P2EHPNA extraction

Pief erenc es 1. Rozen A.M., Nikolotova Z.I., Kartasheva H.A.//DAN SSSR. 1975. Vol.

222. P.1151. 2. Kozen A.M. et al.//Investigation on nuclear fuel reprocessing:

IV Symp.SEV. Prague, 7977. Vol.2. P.22. 3. Rozen A.M. et al.//Radiokhimiya. 1986. Vol.28. P.407. 565 4. Horwitz E., Schulz W..//ISEC86. Miinohen, 1966. Vol.1. P. 181. 5. Myasoedov B.P. et al.,//Solv.Extr.Ion Exch. 1986. V0I.4.P.61. 6. Roaen A.M.//Atomic Energy Rev. 1968. Vol.6. P.100. 7. Rozen A.M. et al.//J.Inorg. Nucl. Ohem. Suppl. 1976. Р.ггЭ. 8. Roaan A.M., Nikolotova Z.I., Kartaaheva Я.А. // Auth.Cert.

USSR H 601971. B.I. 1979. К 35. P.257. 9. Roaen A.M., Nlklforov A.S. et al.//DAIf SSSR. 1985. Vol.285. P.165. 10. Laskorin B.W. et al.//Auth.Cert.USSR. И 181812. В.I. 1966. M 10.

P. 80.

136

EXTRACTION OP TRANS PLUTONIUM ELEMENTS PROM ACID MEDIA BY MIXTURE OP NEUTRAL ORGANOPHOSPHORUS EXTRACTANTS

10­5

H.K.Chmutova, G.A.Pribylova, H.P.Nesterova, B.F.Myasoedov and H.I.Kabachnik, Vemadsky Institute of Geocbealstry and Analytical Chemistry, Academy of Sciences, of the USSR, Moscow, USSR DialkylCdiarylMdialkylcarbamoylmethyl] phosphine oxides (CHPO) is

known can be used successfully for concentrating of transplutonium ele­ments (ТРБ). The disadvantages of these extractants are a low solubili­ty of metal complexes in organic diluents and an incompatibility with diluents.

It was Bhown [l,2] that these disadvantages could be overcame by ad­dition of tributylphosphate (TBP), which had good compatibility with many diluents. The objective of this study was to determine regulariti­es in changes of TPE distribution ratios (D Tp E) and values of their de­viation from additivities (S), concentration of all the system compo­unds and both the nature of CHFO and diluent.

We used extractants of general formula: R'R',

P(0)CH2C(0)jre2'" where R'­R'^Ph, Tol, Bu, BuO or R'­Ph, R'^BuO; R'"»Bt or Bu. Extractants are given in accordance to substituents at phosphorus and nitrogen atoms, respectively*

It was shown that in case of Am(III) extraction from 3M HNO, by 0.02511 solutions of all СЫРО the addition of TBP gave nonadditive en­

­1

[ UNO, I , Fig.l. Acid dependency for Am(III) extraction by 0.05» PhgEtg (1), PhgBug (3), TolgBtg (5) and their mixture with 1.5M TBP (2,4,6) in DCS; (7)­1­5M TBP in DCE

7

[вдо 3] ,Ы Pig.2. Acid dependency for Am(III) extraction by 0.05M PhgEt2 (1) and a mixture of 0.05" PhgBtg and 0.O5M TBP (2), 0.5M TBP (3), 1.5И TBP (4),­ (5)­0.5M TBP» (6)­1.51l TBP. Solvent­DCE

137

hancement ot D ^ (S>j). The study of Jm(III) extraction отег a wide ra­nge of nitric acid concentrations shorn both synergistic (S») and an­tagonistic (S<1) effect could be taken place in the systems (Fig­1). Therefore the degree of D. increasing, S and its manifestation region

ratio and [HBO­] depend on CMPO individual. TBP J on In extraction using a mixture of extrac­

tantc froB НЯО, solutions of various concentration is shown in Fig.2. From Kg.2 it is obvious that dependency of D. and 3 on [THPJls the higher the stronger f HIO,j .

(e.g. [TBP] : [(ЯРО The influence of

Fig­3 gives [СИРО ] influes e upon Am extraction in the presence of TBP. It is shown that increas ng of D. takes place only when the ra­tio of [lBP] to [CMPo] is hig 4, the antagonistic effect is obsnrved un­der smaller ratio and higher CHPO] and [ HH0J<2 that can be seen from ?ig.1 (curves 3­6).

Fig.4 shows that solvent nature is of great importance in am extra­ction by a mixture of extractantn. Under the sane conditions [HTOJ<2H we observed the synergistic effect both with o­xylene (curves 5,6) and 1,2,4­trichlorobenaeBe (TOB) ^curves 3,4) and antagonistic effect with dichloroethane (ВСЕ) (curves 1,2).

Fig­5 shows that addition of TBP also results in Dj^ increasing in the presence of macroamounts of Eu(III). It was Bhown that upon TBP

lDg(Pi,Etg],ll

Plg.3­ Distribution ratio о Am(III) ve. molarity of Ph 2

from 3K (a) and 7м (Ъ) ННО. oy a mixture of 1.5M TBP (1) aid 0.025M TBP (2) with Ph gBt g. (3)­Ph2Bt2 in DCB

Acid dependency for im(III) extraction by 0.05M Tol 2Et 2 (1,3,5) end 11 mixtu­re of 0.05H Tol 2Et 2 and 1.5M TBP (2,4,6) in DCS (1,2); 1,2,4­TCB (3,4) and o­xylene (5,6)

138

addition the sediment didn't form that took place in the system with alone СИРО and [афо.ЗМ.

It will be expressed some views about the nature of synergistic effect in these systems. It is known that with increasing of [HBO,J extraction mechanism of Am with СИРО is changed: in weakly acid solutions im(NO,).n СИРО is extracted, in highly acid ­H*Am(B03)".nCHPO. On changing [HH0,] and complexes composition the role of TBF and nature of sy­nergistic effect are changed­

0.1

0.05 ­p b V

"rr lo, •Ы aq ,M Pig.5. Dependency of Eu(III) amo­unt in organic phase (a) and D­v on extracting by 0.1M Tol­EtgO), 21t TBP(2) and a mixture of 0.1H TolgBtg and IH TBP (3) on Eu(III) concentration from 3¥ HBO­

Conclusion In case of TPE extraction by a mixture of CMPO and TBP D ^ deviati­

on from additivity is observed, which values are connected with СИРО nature and solvent. Synergistic effect increases with enhancement of [TBP] : [СИРО] ratio and [HNO,]. The appearance of antagonistic effect takes place under large [CHPO] and low [HBO,].

Differences in synergistic effect values in case of extraction by a mixture of extractants from low and high nitric acid concentration de­pend on changing of extracted complexes composition with СИРО (Aa(H0,),.nCHPO and Н+

*п(Н0,Г.пСИР0, respectively) and the role of TBP in deviation from additivity.

References Hyasoedov B.F., Chmutova H.K., Kochetkova H.E. et al. // Solv. Extr. and Ion Exch. 1986. Vol.4, H1.P.61. Horwiti B.P., Kalina D.G. // Solv. Extr. and Ion Bxch. 1984. Vol.2, H 2. P.179­

13?

SOLVENT EXTRACTION OF PROTACTDIIUM(V) BY TRIPHENrLPHDSPHHE i , I 10­b

AND TRIPHSSTLARSUfE OXIDES FROM NITRIC ACID MEDIUM ' H. B. Maghrawy, S.A. El­Reefy and H. F. Aly,

Hot Lab. Center, Atomic Energy Establishment, Cairo, Egypt

Introduction

The extraction of actinides with tertiary arsine oxides has been studied in detail in relation to the partition of U, Th, Hp and Pu in different oxidation states between aqueous nitric acid solutions and solutions of aliphatic and aromatic arsine oxides (1). It has been ahown Wat all the tertiary arsine oxides investigated are powerful extractants of trans­urpnium elements and have extraction constant values ranging from 10 ­

"01 . These values exceed by more than three orders of magnitude the

constants for the extraction of the actinides by the isomeric phosphine oxides and are among some of the highest of all those known at the pre­sent time (2). In fact, no data are available concerning the pentavalent oxidation state of the actinides. As has been known, the element Pa(V) has the most stable oxidation state and therefore, is chosen for this study, where the extraction behaviour of Pa(V) with B),F0 and Pfc,AsO in chloroform from nitric acid solutions is investigated.

Experimental Ph,P0 and Ph,AaO were of purum grade and supplied by Pluka. Chloroforn and all other chemicals were of A.R. purity grade and obtained from BDH (england). The radiotracer Pa (K7.4 d) was separated from neutron irradiated ThOTO,), samples by diisobutyl ketone (3). Extraction procedure The details of the extraction procedure were described elsewhere (4). The distribution ratio, 3>, was determined at 25 ± 1°C, and defined by

D counting rate (ем) per ml of the organic phase " counting rate (con; per ml or tne aqueous pnase *

Results and discussion

In the present study, the main two factors affecting the distribution ratio, B, i.e. nitric acid and extractant concentrations are followed at constant conditions of the other factor. As shown in Fig. 1, at constant PhjPO concentration, Dp /v\ Increases with the increase in HKO, concen­tration from 1 ­ 9И by three orders of magnitude. At constant concentra­tion of PhjAeO, Dp_(y\ Increases with HHO* concentration by one order

140

u

Pa(V)

° Oi.00111 Fh,AsO A 0.0O25M

J

• 0.005H О. «0.01M • 0.025M

Ц cone.

1 ! ' 5 ЬшоЛ, M P i g . 2 . Effect of HNO, cone.

З Г С

Р 1 ( Т ) J

1 з 5 LmroJ, H . g . 1 . Effect of НТГО,

of magnitude, J i g . 2 , The aforementioned extract ion behaviour for both extractants i s repeatedly obtained at di f ferent extractant concentrations which indicates that the extract ion stoichiometry for a s p e c i f i c extrac­

tant i s the sane i r re spec t ive of the concentration used. From P i g . 3 and 4 , the logarithmic r e l a t i o n between Dp ,„•. and [extractant] i s l i n e a r with slope one. This was carried oat at 3 di f f erent ШНи,], indicat ing that the extracted complex epecies accomodate only one extractant mole­

cule mainly a l l ­ o v e r the studied HNO, concentration range.

To get deeper i n s i g h t about the extrac t ion mechanism and the nature of extracted s p e c i e s , the logarithmic r e l a t i o n between

в

р а ( у ) ал

^ а+ gj j 0

(a^jjjjQ ­ mean molar a c t i v i t y of HHO,) i s presented i n F i g . 5 and 6 for Phlpo and Ph,AsO, r e s p e c t i v e l y . In case of Ph­PO, two s tra ight l i n e s of s lopes 2 and 1 are obtained, suggest ing the extrac t ion atoichiometry

HHOK

­ 3 ­ 2 log&iuPOj

ях.з. Btfeot of aio OODO.

­ 2 loglph^AeO]

141

> 1 л

а 9 п

' 1 • 1

* .­*~ *2*"^~"^ ~ " " " * " " " ^ ­ ^ • * " "

J~^~~~ <­"*^"г­»^'*^* л~**~~''

L^" ­ ­ е ­ . о ­ " * " " ' ^ о ­ " * " " ' ^

S-* К 11­001»

. 1

• О.ММ • 0.011 • 0.013Я

. 1 . . — — 1 . .

О 1 l o g а^да) ° 1 l o g а

йВН0 3

«­S­.5 Effect of а ± ш 1 0 on D p a ( v ) P i g . 6 , Effec t of а щ , , 0 on B f t ( T )

proceeds v i a : i . at lower HNO, a c t i v i t y , log a ^ g 4 0.7

*2+ P a № 0 3 ) ' г йог И1 Х И) PaflP^g . ?h 3 H

Н . at higher НПО, a c t i v i t y , l og t%m0 > 0 . 7

Pa(H03)J HO: Ph,SO Ы Г О 3 ) 5 ' 5ЦД)

(1)

(2)

In case of PhjAsO, Fig; 6, a £roup of almost p a r a l l e l s tra ight l^nes of slope at 1 , al lover the atudied HNOj a c t i v i t y range are obtained. This c l e a r l y i n d i c a t e s that the extrac t ion stoichiometry proceeds v i a

Pa(K03)J so: m,Asu (3)

The formation constants of the proposed extraction were not evaluated since the interaction of HNOj with the two extractants ie not clear at high acid and extractant concentrations (5). However, it is found that an empirical equation can satisfactorily represents the extraction data. This is achieved by considering the effect oi' both nitric acid and extractant concentrations on the distribution ratio. A general expression for this dependency la given as follows :

D

Pa(V) "1 fextractant] (4)

l og Dp /уч « log It, + k 2 log textractant l + k~ log tHHOj] . . .

The values of the constants k 1 t kg and k, are calculated by f i t t i n g equation ( • ) with the experimentally obtained D­values; Pig. 1 and 2 and the best f i t for both PluPO and Ph~AeO i s obtained, with the fo l low­

ing two equations : l o g ^ a O O " ° '

2 7 + 1°S t?h3PO] + 3 l og nfflOji . . . ( 6 ) ; with % « 0;18, and

142

l og D p a ( V ) « 2.05 + l o g CPnylaOl + l og fflHOjI . . . . ( 7 ) ;

with S « 0 . 0 1 , where S i s the error ecuare sum, defined by :

The s o l i d l i n e » drawn i n Fig; 1 and 3 are obtained .from equation ( 6 ) ,

whereby the corresponding Xinee in Pig . 2 and 4 are obtained from equar­

t ion ( 7 ) . The error square sum of the data i n d i c a t e s the good v a l i d i t y of the empirical equations deduced with respect to the experimental r e ­

s u l t s obtained. The dif ference i n the values of the constants obtained between Jh­PO and Ph»AeO systeme can be re la ted to the di f f eren t i n t e r ­

act ions between each extractant , n i t r i c ac id , the di luent and the Pa(V)­

a p e c i e s .

Comparing equations (6) and ( 7 ) , the value of ^(ИцАаО) ia much higher than that of k.,(Ph,P0). This i s r e f l e c t e d by the high extrac t ion of Pa(V) with 7h­AsO than '*ith the same concentration of Ph»P0 up to 7.5M НПО,. The high capabi l i ty of PhjAsC over Ph.PO can be attr ibuted to the high e lec tron donating power of oxygen In ЙиЛаО than the o.iygen In Ph,P0 based on the low e l e c t r o n e g a t i v i t y of As than P. However, the dif ference i n the e l e c t r o n e g a t i v i t y of Аэ(2;18) than P(2.19) i s rather small (6) and cannot account for such pronounced difference i n the ex*Tactabl l i ty between Ph­AeO and РЦРО. S t r u c t r a l l y , arsenyl gxoup involves cooperat­

ive in terac t ion between the arsenic and the oxygen i n the у As*— 0" o" ­bond and the p ­ d„ conjucation of the unshared e lectron pairs of oxygen with the highes t vacant Ай o r b i t a l s of arsen ic . As a r e s u l t , the •7 As — ­ 0 bond aqulres p a r t i a l double bond character . This can form­

a l l y explain the high extract ion power of IfcjAeO over PbjPO.

References

1. B.H.Laskorin, V.V.Takshln, E.A.Fi l lppov, K.A.IoTibosvetova, V.A.Belov and G.G.Arkhipova//Proceedings of the IVth Symposium on"Stud ie s on the Processing of Irradiated Fuel . Karlovy Vary, 1977. P . 2 2 .

2. B.H.Iiaskorin, V.V.Yakshin, E.A.Fil ippov, H.A.byubosvetova, V.A;Belov and S.G.Arkhlpova//Radiokhimlya. J20. 702 (1978) .

3 . A.G.Goble and A.G.Maadock//J. Inorg. Kucl. Chem. _7. 9

* (1958) .

4. H.B.Haghrawy, C.Girgis and A JS.Abdel­Gawad//J. Radioanal. Duel. Chem. 8 1 Л . 5 (1984) .

5. A.H.baskorin, T.V.rakehin, and H.A.ftruboev6tova//uspekhi Xhlmii. 50. 860 (1981) . ~~

6 . H.Zakareia, S.M;Khalifa, J.AiDaoud, and H.F;Aly//Radloohira. Acta. 39, 89 (1986) .

143

EELECTIOB OF EXTEACTIOH SYSTEMS FOB S3PARATI0H OF SOKE LAHTHAHIDES

V.Jedin&lcOTa and Z.Dvorak, Prague Institute of Chemical Technology, Czechoslovakia The selection ot conditions for the extraction separation of metals

ie associated with the determination of the effect of molecular struc­ture of a extragent on the extraction equilibrium. One of the contri­buting factors ma; be the characteristics of the substituente, i.e. their nature, size, mass and symmetry of these radicals. Studies of the effect of the composition of amines on their extraction ability concentrate in the first place on an aseeement of the induction ef­fects of the suDstituents which determine the nucleophllity of nitro­gen atom.

The effect of the extragent structure may be described quantitati­vely with sufficient accuracy by means of one­ or several­parameter equations which express the important characteristics of the system affecting the extraction properties within the series under investi­gation. A correlation of experimental data with the effect of the in­dividual factors shows that extraction series may be described by me­ans of two­parameter equatlor of the following type

log К = log K Q + QZ6"+ SE E s , (1)

where К is the extraction constant of given system, К is a constant accepted ая в standard, Q and S are oonstants for all the systems of the series which differ only In the nature of the aubstituents I bonded to the functional group, 6 and Eg are parameters characteriz­ ' ing the induction and steric effects of the subetituents in the in­vestigated type of an extraction series. Values Of and E_ (taken over from literature) were mostly obtained from measurements of organic substrates. A number of equations of type (1) have been determined for each series with given experimental values of distribution ratios and sums Z6" and £ Eg which have been evaluated from tabular data for amine in question. Application of this method with special reference to the extraotion of lanthanides by benzyldialkylaminea is illustrated in Table.

A one­parameter correlation equation Including only one of these parameter* was always used to assess the contribution of the Inducti­on and steric effects involved In a particular series. Data show that the one­parameter correlation log Djje ­ZS Is valid only for benzyl­dialkylamines with the number of carbons equal to 4 ­ 8, where the linear 'ependencs of log Dga onSCwms found for Ce(IXI) and Eu(III).

10­7

144

literature does not provide sufficient data on E„ set constants which would allow to express the sterlc effects peculiar to amines. The values of E g constants for the individual substituents were chosen on the basis of eteric constants published in literature[l­4)and modi­fied values which approximate those of Hansen sets. Eg values for sub­stituents which have not been published in literature were added on the basis of the established rules.

Induction and Steric Constants and Theoretic Values of D

Ce' DBu

a n d a

Ce/£u C a l c u l a t e d io

* Tertiary Benzyldial­kylamines and Dibenzylalkylaminee

Amine 2 6 ­ £ E

S D

e

4> u

Ce D

cf »5? 4? i •

,v c a l ""Ce/Eu

BDEH 0,02 4.8I 1,6 2 , 8 BDBuK ­ 0 . 0 4 5.07 9.7 10,10 4 . 5 4.75 2.13 BDHeH ­ 0 . 0 6 5.17 6.4 8 .24 3.7 3 .93 2 .10 BDOH ­ 0 . 0 8 5.41 4 .2 4 .63 3 .0 2.46 1.88 ВШоН ­ 0 . 1 0 5.57 3.6 3.21 1.9 1.81 1.77 BDLH ­ 0 . 1 3 5.71 2.2 2.41 1.2 1.39 1.73 DBBuH 0.31 5.25 2.2 2.51 2.7 2.71 0.93 DBON 0.29 5.42 1.4 1.69 DBHeH 0.30 5.30 1.9 2.26

As amine extraction follows the pattern of an equilibrium reaction where induction effects are always involved to a lesser or greater extent, as corroborated by the correlation the authors have found for £6" and £ E S > expressed by means of a two­parameter equation* As re­gards the extraction of Ce(III) by tertiary benzylalkylamines, with log D 0 « 6.7 the following constants have been evaluated :<!,= ­1.1 and S C e = 6.7. In the case of Eu<III) extraction, with log D 0 =» 5.1, the constants are: Q E u ­ ­0.2 and<T C e « 0.87. A comparison of values D^ Q obtained by calculation with experimental data is given in Table, As it 1э evident from Table,the two­paraseter correlation facilitates description of the extraction experimental data obtained for alkylami­nes, and it may also вегте as a basis for the analysis of the indivi­dual factors peculiar to the structure of the extragent and their se­lectivity. This description, however, only applies to a speoiflc class of ertregents and it is not universal.

As regards the extraction of two elements, the log of the separa­

10. 3ax. 382 145

tlon factor ( logOC, j 2 ) , eipreeeed by means of equation (l), gives the following expression

i o g o ( 1 > 2 = log D O 1 ­ log D 0 2 + д>, ­ qgzs* ff\ . «г^гЕд .

= log « ° > 2 + <?1>2гб" + a ' 1 t 2 Z B s , ( 2 )

It is therefore possible to draw the conclusion that the effect of the extragent structure on the separation factor can be described by means of these correlation equations. The values of<p1 ­ and & „ c o

~ efficients and the corresponding values of the separation factors can serve as an indication of the influence of the induction and atoxic effects of substituents on the separation prooesa selectivity.

References 1. Fain V.A. // Osnovy kolicestvennoj teoriji organiceekioh reaxcij,

ed.Chimida,1967. P. 359. 2. Exnei 0.// Korelacni vitahy v organicкё chemii, SHTL, Alfa,

Prague, 1981. Chap. 2,3,6. 3. Taft R. // Sterio Effects In Organic Chemistry / Ed. Newman M.S.

Hew York: Wiley. Chap, 2. 4. Unger S.H., Hansen C. // Quantitative feodels of Static Effected

ProgressingPhysical Organic Chemistry / ed. R.W. Taft. Dew York: Wiley and Sons, 1976. Vol.12. P. 91.

146

USE OP TWO­PHASE AQUEOUS SYSTEMS BASED OH POLYETHYLENE GLYCOL FOR ISOLATION AND SEPARATION OP AMERICIUM IN I

1 0

~8

DIFFERENT OXIDATION STATES N.P.ltolochui'KOva, V.Ya.Frenkel, V.M.Shkinev > B.F.Hyasoedov, Vernadsky Institute of Geochemistry end Analytical Chemistry, Academy of Sciences of tbe USSR, Moscow, USSR Solvent e: traction of americium in higher oxidation statea is wide­

ly used for i лег1с!ит separation from other actinides and is especial­ly promissim for americium separation from curium [1] . When using traditional olvent extraction systems americium in higher oxidation states is un table in organic phase due to reduction in the presence of organic e­:tractens and solvents* In this respect it was of interest to use two­t ase systems based on polyethylene glycol free of organic solvents for americium extraction in higher oxidation states*

Bxtractio i of trivalent actinides has been studied previously in the followii ; systems: polyethylene glycol (PEG) ­ ammonium sulfate ­water ­ arsenazo III and PEG ­ potassium carbonate ­ water ­ alizarine complexone [2­3] . Extraction of americium in higher oxidation states by PEG has r;en investigated from concentrated carbonate solutions in the preaenci of alizarine complexone (AC). Americium has been establi­shed to qua titatively pass into the PEG phase at AC concentrations more or equt1 0*02 M. In this case the efficiency of americium isola­tion is con ­ant for different oxidation states when taking trace amo­unts of aaericium (10 ­ 10 M); and distribution coefficients of amerieium (IV)-, (V) and (71) differ significantly from those of other actinides in the same oxidation states (see Table • ) • It apparently

Table 1• Distribution coefficients of actiniums in case of extraction into the PEG phase from K„C0, (..•.' mass Я>) in the presence of 0.02 M AC

. totals Ami ­ I D Cm(III) Am(V)» Np(V) Pa(V) Am(VT)» U(VI) Am(IV)* pudv: D 43­7 127­3 7­3 0.15 0.12 33 .2 0.05 40 .8 13­4

' I n i t i a l oxidat ion s t a t e s of Am prior t< e x t r a c t i o n

results from trace amounts ot americium (IV), (V) and (VI) to be un­stable in ais system and to reduce to americium (III)•

Behavio г of macroamounts of americium ( * 10 M) in higher oxida­tion state in carbonate solutions was studied using spectrophotometry. Americium as electrochemically oxidized to a hexavalent state . Pig. 1 shows the absorption upectra jf ^mc^icium in potegsium carbona­te solution recorded after oxidation and addition of PEG and AC. The data obtained indicate a quantitative aaericium oxidation to americium (VI) and no change in valence state of americium caused by PEG. After complexone addition to this system absorption oanda appear in solution

147

A ад

0,6

ОЛ

0Л[

Л*

о Pig­I­

•**г­ 5» 600 А,пт _ Absorption spectra of

Am carbonate solution: 1 ­ af­ter the oxidation, 2 ­ after the contact with PEG, 3 ­ af­ter the addition of AC, 4 ­after the contact of Am with AC during 26 hours

absorption spectrum of Am(VT) which are characteristic of Am(III)l (maximum at 508 nm) and Am(V) (maximum at 518 nm). These spectra show that macroamounts of Am(VI) are reduced by AC giving ma­inly Am(V). Under a prolonged contact with the extractant the part of Am(III) does increase* Fig. 1, curve 4 shows that the spectrum recorded in 26 hours corresponds mainly to Am(HI) absorp­tion spectrum in carbonate solutions. Macroamounts of Am(V) were obtained by electrochemical reduction of Am(VI). Spectrophotometry has shown that Am(V) not move into the PEG phase containing 0.02 M AC nor reduce to Am(III). Thie enabled us to develope a separation method of macroamounts of Am(V) from Cm in carbonate solutions. oC­ Spectra

of an initial solution and that of a saline solution recorded af­ter curium extraction by PBG are shown in Pig. 2.CC­ Spectrum shows practically no curium in the sali­ne phase. Separation factor of Am and Cm is more than 100 for one extraction cycle by PEG containing AC.

Americium extraction in the PES ­ ammonium sulfate ­ water ­ arse­nazo III system has been studied­Conditions have been found in this system for quantitative extraction of Am(V) by P3i». Americium extrac­tion as a function of equilibrium salt phase pH is given in Table 2­A quantitative aaericlum extracti­

on is observed at pH 4­5 in the presence uf 10"^ IE arsenazo III. A ape­ctrophotome trie study has shown Am(V) to form a complex with arsenazo III of a 1:1 composition. Three absorption bands are observed in the absorption spectrum of a sulfate solution containing arseuasu III (1.1 x 10" 4 H) and Am(V) (5­4 r. 10~ 4 Ю which have maxima at 550, 580 and 670 nm (see Pig. 3). These absorption bande remain in the soluti­on absorption spectrum of PBG after Aia(V) extraction, but unlike sall­

630 700 Channel

960 Number

Pig.2. oC­ Spectra of the initial solution of Am(V) and Cm (A) and of the salt phase after the ex­traction by PEG in the presence of AC (B)

148

Table 2­ Distribution coefficients of actinides as a function of pH in case of extraction by PEG in the presence of 10"^ M areenazo III

pH Am(III) Am(IV}» Pu(IV) Am(V)* 2.1 0.30 ­ 16.6 ­2.6 0.31 1.04 ­ ­2.8 0.95 ­ 36.6 ­3.0 1.50 5.20 ­ 0.23 4.0 46.20 57.51 59 7 52.72 •Initial oxidation states of Am prioj to extraction

ft6

<K*

o»

о «о «о т Л > Ш 1 Pig.3» Absorption spectra of Am(V) complexes with ar­senazo III in (HH.)230. so­lution (1) and PEG (2)

ne solutions a small shift is observed in characteristic absorption bands to a long­wave region: 580, 620 and 660 nm. Optical densities of Am(V) solutions with areenazo III in PEG are much higher than tho­se in saline solutions. The molar extinction coefficients of Am(V) complex in the PEG solution are 1x10* at 620 nm and 1x10 l.mol . cm at 6S0 nm. One can clearly see in the spectra two absorption bands at 517 and 718 am belonging to Am(V) together with those cha­racteristic of americium complex with агвепако III. There are no ab­sorption bands characteristic of Am(III) in the spectrum of the PEG phase. This indicates no reduction of Am(Y) upon extraction by PSG containing areenazo III. A spectrophotometry study of Am(VI) extrac­tion has shown Am(VT) to reduce to Am(V) under the contact with arse­nazo III, Am(V) being subsequently extracted by PEG.

It was interesting to study the behaviour of Am(IV) in the system under consideration since arsunazo III complexes with four­charged me­tal cations are more stable than those with triply­charged cations. The PEG solution has been found previously to quantitatively extract tetravalent plutonium and thorium at lower pH values of saline solu­tions than trivalent actinides­ However, at low pH values of a saline solution distribution coefficients of Am(IV) are not much higher than those of Am(III) (see Table 2)•

Thus, two­phase aqueous systems based on polyethylene glycol can be widely used for extraction isolation and separation of americium in different oxidation states.

References 1. Hyasoedov B.P., Lebedev I.A.//4adiochim.Acta. 1983­ Vol.32. P.55­2. Shkinev V.M., Molochnikova И.Р. , Zvarova T.I. et al.//J.Hadioanal.

Hucl.Cham., Articles. 1985­ Vol.88. P.115­3. Molochnikova H.P., Prenkel V.Ya , Myaeoedov B.P. et al.//Hadlokhi­

miya. 1987­ Vol.29. P.32. 149

SEPARATION OF TPE PROII ЗОНЕ OTHBH EtEMBNTS ИЮИ CARBONATE i SOLUTIONS Bi" EXTRACTION AND EXTRACTION CHROMATOGRAPHY I­

1° ~

9

Z.K.Karalova, T.I.Bukina, E.A.Lavrinovich, B.F.Hyasoedov, Vemadsky Institute of Geochemistry and Analytical Chemistry, Academy of Sciences of the USSR, Мое cow, USSR

Actinides in carbonate solutions form various complex compounds, the content and the stability of which being affected by the concen­tration and relation of the components, the pH value and other fac­tors. The formation of the anion complexes of the following composi­tion is typical for TPE: Am(C0 3) 2 " (lg/ = 11.44); Am(C03)2(OH)

2

" (lg/=15.56); Am(C0 3),

3" (lg/­13); Am(C0 3) 3 t>H)4

~; Am(C0 3) 4' •(OH)

6

"; Cm(C03)«(0H) 6­ [ij. Taking into consideration the variable composition of carbonate complexes of actinides in the aqueous pha­se, one can expect that extractants of different classes will es­sentially affect the content of the compounds in the organic phase. Primary amines (N­decylamine (ПА)) extract TI'E and BEE from carbo­nate solutions in the form of ion aecoOiite', of (RNHJJJ. [M(C0 3) 3J composition including metal tricarbonate complex as anion [2­3J. Chelating reagente (fi ­diketones (РЫЮР, HTTA); alkylpyrocatecholes ' (HOP, TAP) extract trivalent TPE and REE in the form of intracom­ . plex coordinationally saturated and hydrolized compounds of the type of M(PhHBP).nH20( H(TTA)3.aH20 £ 4­5] J Na2[HD0P2(0H)(H20)J [6­7].

In that case carbonate­ion is not the necessary component of the extracted compounds, but only a complexing agent, which con­fines the hydralizing elements in the low alkaline media in the so­luble form. Since the extraction process of the intracomplex compo­unds and the disintegration of the complex compounds in the aqueous phi зе take some time, the role of the kinetic factors, especially in metals with alkylpyrocatechole, is essentially increased [ в]• As the experimental data show, the rate of mass transfer is determi­ned ty slow chemical reactions, which take place both in the volume of tt? aqueous phase and at the phase boundary. Although the incre­ase in the pH of the initial solution facilitates these reactions, > the possibility of the formation of the interphase film grows. To avoid the effect of the interphase film on the mass transfer rate, the extraction ahould be conducted with high content of the alka­line metal oarbonate. Firstly, the increase in the content of pota­ssium carbonate and alkali increases the concentration of the inter­mediate hydroxocarbonate complexes, which are responsible for the velocity of the beginning of the reaction, ana secondly, it allows ^ to avoid the accumulation of the products of hydrolytic poliraeri­ b

150

zation in the zone of the reaction. Therefore, by varying the con­centration of caustic potassium, one can achieve quantitative iso­lation of amerieium even from a saturated solution of KgCO­j £7J (Pig.1). The most part of the studied extractante rather highly ex­tract both macro­ and micro­amounts of elements from car­bonate solutions. As the same time the extracts, saturated with metals, based on DA, PhMBF, TAP, with the concen­tration of the order of n. 10 Л,

can only be obtained by con­centration, since TPE carbona­te compounds, as well as REE, are comparatively poorly so­luble in carbonate eolu+tons.

Extraction from carbonate solutions can be used not on­ly for TPE isolation, but al­so their separation. Like in the case with alkaline solu­tions, the sequence of acti­nide isolation from carbona­te solutions in decreased in the series M(III)> H(IV)> H(V)> } M(VI), according to the de­gree of oxidation. It is cle­arly manifested on Pig.1­2 where the comparison is made between the metal extraction with alkylpyroeatecholes [ 7 J and fi ­dilcetones £4­5}. Sin­ce hexavalent and tetravalent elements are extracted worse than trivalent elements, in these systems rather high se­paration factors (n.10*) for pairs Am(III)­U(VIh Cm(III)­U(VI)i Am(III)­Pu(IV)» Cm(III)­Pu(IV).

"СкгС0з.М

Р1Я.1. Extraction of metals from КЛЮ­ЛМ') solutions and the mixture of K 2C0 3 and 2 M KOH(M") by 0.04 M DOP in toluene

*4 5

1°8<4"> S.M

P i g . 2 . Extract ion of metals from 2 B o , so lu t ions with 0.1 И PhltBP

i n JOBK

151

The selectivity of the extraction separation can be increased by two ways: by varying the alkalinity of the media and the time period of the phase contact. Thus, adding small amounts of caustic potassi­um into the carbonate solutions not only enhance the extraction of many elements, but also increases the factors of americium separation from berkelium,cerium and other elements,expert europium ( Е ^ / Б ^ г ) .

Among the agents tested for the extraction separation of Am­Cm, TAP is practically the only agent, which extracts curium and cali­fornium from carbonate solutions better than americium (Dpга/0д_= 2.2) [8­9J. Additional possibility of separating these elements occurs during the re­extraction by potassium carbonate mixtures and salts of arainopolycarbonic acids. The comparison of the kinetic ac­tivity of the studied re­extracts has shown that DTPA content es­sentially effect the rate of the re­extraction of the elements un­der study, and the increase in its concentration at constant K 2C0, content leads to a sharper increase in the rate of curium re­extra­ction against americium. Under the optimal conditions the re­extra­ction rate constants of these elements differ in more than an order.

The efficiency of the separation of trivalent elements grows du­ring the transfer from a single static extraction to extraction chromatography due to the differences both In the values of equi­librium constants and the values of the mass transfer coefficients [8}. By varying the content and the rate of the penetration of the solution through the column with fluoroplaet­4, imprecated with 5%­TAP in toluene,the fractions were obtained at the stages of sorption, washing and desorption, containing ~905t of curium and 1* of americium.

References 1. Bidoglio G.//Radiochem.and Radioanal.lett. 1982. Vol.53. P.45. 2. Karalova Z.K., Myasoedov B.P. et al.//Badiokhiinia. 1983.Vol.25.

P. 595. 3. Karalova Z.K. .Myasoedov B.P. et al.//Ibid.1985. Vol.27. P.47. 4. Karalova Z.K..Hyasoedov B.P., Devirths Б.А. et al.//Ibld. 1936.

Vol.28. P.47. 5. Karalova Z.K., Devirths E.A., Myasoedov B.P.//Ibid. 1987.Vol.29.

P. 33. 6. Myasoedov B.?., Karalova Z.K., Kuznetsova V.S. et al.//Ibid.

1980. Vol.22. P.347. 7. Karalova Z.K., Hyaooedov B.P. et al.//Ibid. 1983. Vol.25. P.187. 8. Hovikov A.P., Bukina T.I., Karalova Z.K., Myasoedov B.P.//Ibid.

1987. Vol.29. P.184. 9. Karalova Ъ.\., Bukina T.I., Hyaeoedov B.P. et al.//Ibid. P.335.

152

10­10 MFLUEKCE OP THE SOLVENT EXTRACTIOfl OH THE REDOX EQUILIBRIA, OH THE EXAMPLE OP TRANSURANIUM ELEMENTS V.N.Kosyakov, N.G.Yakovlev and M.M.Vlasov, Kurchatov Atomic Energy Institute, Moscow, USSR The main part of the most efficient techniques for recovery and pu­

rification of the transuranium elements are based on a combination ol solvent extraction with redox cycles* Dealing with the trace quanti­ties of an element,solvent extraction technique is widely used for studying redox properties, complex formation and other processes ta­king place in aqueous solution» The results obtained and their inter­pretation in all cases depends very much on the experimental condi­tions and the way of the data treatment.

Any redox equilibrium supposes existence of at least two oxidation states in a ystem. Por a simplicity we shall confine ourselves to consideration of two forms, Me(Ox) and Me(Red), being in a reversible equilibrium. Here we shall operate by such a practical parameters as: relation of analytical concentration of oxidized­ and reduced forms in an aqueous phase, a r.'z"^ji''and in a solvent extraction system as a whole, A ­ Ц * у № ' ' | jdistribution coefficients of the both

UA'MU + lMelMJe t0™ D

ox= Й Й " * D« a

= ШШ *** "Р«*»".11у deter­mined distribution coefficient of the element under consideration:

As shown in /17 , the relations connecting all these parameters with each other can be obtained by simple transformation of the above expressions in to:

Я= £2* (2); А = й­£2 (3); D = — 2 2 Ш. (4) D

ox ­ D D

red + ' '

+ a

Speaking about the influence of solvent extraction on the redox equilibria, we should always compare the concentration relations of oxidized and reduced forms in an initial aqueous solution, " CI »t

and in a final solvent extraction system as a whole, "A". In the paper we shall consider three practical aspects, resulting from the mutual consideration of solvent extraction­ and redox equilibria.

I. Solvent extraction technique for stydying redox equilibria.There are two cases, which should be considered here:

­ when the solvent extraction process is not accompanied by any redox reaction and result in a simple distribution of existing Ox­and Red­forms between aqueous and organic phases;

­ when the solvent extraction process is accompanied by a re­dox reaction. In the first case t.v j concentration relation of oxi­

153

dized and reduced forms in the initial aqueous solution is equal to the corresponding relation in the solvent extraction system as a whole i.e. a° * A f a. The second case is characterized by the conditions, «hen the concentration relation of the oxidized and re­duced forms in the aqueous solution before the start of extraction process and after the attainment of an extraction equilibrium, re­mains unchanged, i.e. a° » a f A. This case is a rather frequent one, when in an aqueous solution a redox potential, "E", is supported con­tinuously by an introduction of a redox agent or by an electrochemical device.

The results of the analytical consideration of the equantions(2,3, for both cases and their variations are summarized In Table .

The equations for the calculation of "a°" from the solvent extraction data

System General form D

o x ' D

r e d At the condi t ions when. System General form

D

o x ' D

r e d В

о ^ > ^ r e d * * 1

D

o x> D

>D

r e d ^ ^ r e d » ' D

o x< D < I

W

Type I a ­ «

D

­D

r e d ^I >

o x+ 1

>

<D

o x ­D

>( D

r e d+ 1

> a° ­ A = D »" ­ A = : I

Type I I a" D ­

Г

~ 4 „ „ , , A r. D a ­ С D ~

ox ox

Q „ ^D

r e d z

D

r e d A D

Type I I D

o x "D

„ „ , , A r. D a ­ С D ~

ox ox

Q „ ^D

r e d z

D

r e d A D

The conditions of the applicability of a solvent extraction techni­que and the corresponding equations for the calculation of a°­values are represented in TaSle I as well, (then all the distribution coeffi­cients are known, the solvent extraction technique is applicable for both types of system, but the equations for the calculation of "a 0

" are different, depending on the type of the system. In case II even for approximate calculations it is necessary to know not only appa­rent distribution coefficient but the distribution coefficient of the most extractable form as well.

The possibilities of application of solvent extraction techni­que for evaluation of formal redox potential of Bk(IV)/Bk(III) couple ' at the condition when DTJV(TV) i e n o t ^own exactly and possible mis­takes in data interpretation are considered in detail in work ftj .

II.The condition selection for stabilization of an element at the desirable oxidation «tat». As follows from eq.O), relation "A" in an extraction system can differ from relation "a 0

" by a factor of several hundreds or thousands. For example, an oxidant continuously acting during extraction process provides 1SS oxidation of a metal ion in "

154

aqueous phase, i.e. a°=0,01. At D Q I » 10* and D r e d * 0,01 oxida­tion degree in the whole solvent extraction system according to eq.O) will be 99%. Generally in the systems with D o x » 1 and И­ед^ 1 en oxidant, which can not be predicted on the basic of Its properties in monofase aqueous solution, nay be very efficient [z] . On the other hand, in case of reduction back­extraction procedure even relatively strong reductants (usually active In aqueous solution) nay be unsffi­cient for quantitative transfer an element into aqueous phase. For example, at electrochemical oxidation of Bk(III) directly in extrae­tion­chromatography column (with 10% HDEHP on carbonfibre) complete 'oxidation and sorbtion of Bk take place at potential +1,4 V [3J (which is 130 mV lower than the formal potential of Bk(IV)/Bk(III) in corresponding medium). Berkellum stripping by electrochemical reduc­tion takes place at 1,0 V and less. When using TBP as an extractast, which under similar conditions has much lower distribution coeffici­ent for Bk(IT), the corresponding potentials increase up to 1,7 V and 1,3 V.

Ill.Extraction radlopolarography for determination of redox poten­tials of Tranaur llum ions. The main advantage of this technique is an opportunity to determine formal potentials of the elements, which are not available in weighale quantities.

Suppose "ES" be a formal redox potential of ae(Ox)/He(Red) in a solution and "E" be a potential, which could be supported in the solution by an electrochemical device. Let us introduce and extrac­tant predominantly extracting the element in a definite oxidation state. When solvent extraction process is equilibrated 2quatiuns(2­4) are valid. On the other hand the concentration relation of oxidized and reduced forms, "a", will be determined by the Hernat equations Substituting eq.(2) into the Jfernst equation, we obtain:

(5).

Khot/ing the potential E and the corresponding distribution coef­ficients, the value of E| can be determined on the basis of eq.(5). In practice a dependence of the apparent distribution coefficient, "D", upon the solution potential la interesting to draw. To obtain this dependence let us uolve eq.(5) relatively D and represent the resulting expression in a log.­form:

^*a^+$V­**i/ ( 6 ) The plot of lgD vex­cua E represents an S­shaped ourve. The values

of "lgD" change within the limits of D „ u and D . The infleotlon point of the ourve, E 1 / 2 , is determined by double differeatiation

1S5

of eq.6:

ь

1/2 2,3 5 p l g ^ a (7)

and corresponds to

IgD. 1/2 Ц°вх + l g D

red (8)

The determination of the inflection point coordinates from eq.(8) and the relation ox. from the experimental data permits us to cal­

^ d „ culate the value of E­ using eq.(7).

The best way to control ihe potential of the solution is to use an electrochemical sell with a working electrode of large surface, which simultaneously serves for supporting the organic phase. The technique was succesfully tested 4 in the determination of the formal redox potential of Bk(IV)/Bk(III) couple.

The theoretical depen­dence of the apparent distri­bution coefficient "D" upon the oxidation potential, "E",­at the conditions: 1) 2)

D

red D

red • 0,1, 0,1,

D

ox D

ox » 1,0 ­ 10j

3) 4)

D

red D

red a 0,1,

10, D

ox D

o x ­

= 10* 0,1

In all cases a » 1

0,24 E­Ej, V

1 . H.G.Yakovlev, V.H.Koeyakov, G.M.KaHikova / / J.Hadloanal.Chem. 1982. Vol . 75 , Я 1 ­ 2 . P. 7.

2 . V.H.Xoayakov, r.G.Davyaov / / Radioohimiya. 19B3. Vol. 25, я г . p. 207.

3 . ir.H.&wyakov, N.G.Yakovlev, H.M.Vlasov / / Badioohlmiya. 1986. Vol. 28 , н 4 , P. 547.

4 . V.N.Koeyakov, IJ.G.TukoTlev, M.M.Vlasov / / Radiochlmiya. 1986. Vol.,28, К 4 . P. 543,

156

POSSIBILITIES 07 AHERICIOM­COHJUH­SKPARATION IN CONNECTION VJ;TH REDOX PROCBSSES P. FUchtner, A. Hermann and D. Nebel, Academy of Sciences of the GDR, Central Institute of Nuclear Research, Roaeendorf, GDR Trivalent amerloium and curium are known to behave very similar

to each other in chemical reactions. For this reaeon up to now the separation of Am and Cm пав been a ta&k not easily tc be accom­plished. Attempts have been made to use the higher oxidation states of a­iericium (Ara­V, Am­VI) for its aeparation from curium which under appropriate conditions remains trivalent or tetravalent in so­lution /Л7. There are a lot of two­phase ауstems with different che­mical behaviour of Am(VI) and Cm(III) or Ara(V) and Cm(III) and, con­sequently, high separation factors have been achieved. If the oxidi­zing conditions have been chosen to establish pentavalent amerioium in solution, a precipitate of К.АтО.ДСО­)­ оап be used to separate Аш from Cm which is remaining in the liquid phase. In extraction systems Cm(IH) can be transferred into an organic phase containing di(2ethylhexyl)phosphoric acid (HDBHP), 1­phenyl­3­methy1­A­benzoyl­

pyrasolone­5 or 2­thenoyltrifluoroacetone and Am(V)will be left in the aqueous solution &J. After oxidizing araerioium to the hexava­lent state and establishing the corresponding conditions in solution Am(VI) can be extracted by HDEHP and Cm(III) is retained in the aqueous phase CiJ. In all the separation procedures mentioned the problem con­centrates on how to hold americium in the higher valence during the time of separation.

In connection with arnericiv^­curium­separation by lig.uid­liq.uid extraction problems of oxidizing Ai~ to the higher valence states by using ammonlumperoxidlsulphate in thu presence of silver ions ав а catalyst have been investigated. The different oxidation states of Am have been Identified spectrophotometrically by the help of characte­ristic absorption peaks of the species present.

The results presented In Table 1 deaonstrate a slow and Incomplete oxidation of An to the hexavalent state by (KH^pSpOc, (in the presence of Ag ) at room temperature. By performing the oxidation at elevated temperatures (60 C) one oan reach the ooaplete oxidation to Am(VI) within 10 minutes, but the hexavalent state will he stable only for a few minutes. Another condition consists in a rather low aoidity of the solution. Already in 0.5 N НЯ0. only 85 % of Am(Vl) have bean found. Therefore oondltione of better stability of hexavalent aaeriolum should be elabsrated. In this context the influence of potassium phosphorwolfra­nate has been thoroughly investigated.

Potassium phospnorwolframate (К. Г^О^А^&А> K p v

) has beoone well known as a strong inorganic" oomplexlng agent capable of stabilizing

157

Table 1. Experimental result» on oxidation of amerlcium (Condltlone; 0.001 И Am, 0.1 M (HH^SgOg, 0.01 M л«М0у

я Concentration of HN0,, II

Tempera­

ture, °C Time to maxi­mum of oxida­tion, min

Result of oxidation я Concentration of HN0,, II

Tempera­

ture, °C Time to maxi­mum of oxida­tion, min Valence

Btate Degree

1 2 3

0.1 0.1 0.5

25 80 80

?60 10 8

Am­VI Am­VI Am­VI

80 100 85

the higher valence states of aatinoides in solution £PJ. The reeulte collected in Table 2 give an impreeeive picture of the enormoue in­fluence of KPW on the oxidation of Am by means of (HH.) pS_0 f i in the presence of ell* r ions. Obviously, the ratio of KVW to Am ie of high importance and determines th« final valence state of Am under the given conditions. Khen comparing the data of the first three lines of Taole 2

Table 2. Influence of potassium phosphorvolfraroate CKPW) on *he oxidation of americium

(Conditions: 0.001 M Am, 0­1 M ( N H ^ ^ O g , 0.01 M AgNO^

H Concentra­

tion of НПО,, M

Tempe­

rature KPW/Am ratio

Time to maxi­mum of oxida­tion, ffiin

Result of oxidation H Concentra­

tion of НПО,, M

Tempe­

rature KPW/Am ratio

Time to maxi­mum of oxida­tion, ffiin Valenoe

state Degree

1 0.1 25 5.0 imediately Am­IV 100 2 0.1 25 0.2 15 Am­VI 100 3 1.0 25 0.2 *0 Am­VI 100 It 2.0 25 0.2 90 Аш­VI 35 5 2.0 25 0.6 15 Am­VI 100 6 2.5 25 0.6 20 Am­VI 100 7 2.5 50 0.6 15 Am­VI 75 8 2.5 50 1 .0 15 Am­VI 65 9 3.0 25 0.2 170 Am­VI 15

10 3.0 25 0.6 120 Am­VI 75 11 3.0 50 0.6 45 Am­VI 16 12 3­0 70 0.6 20 Am­VI 8 13 k.O 70 0.6 no oxid. itlon

158

with thoee of Table 1 one should be convinced of the possibility of oxi­dising Am completely to Am­VI at room temperature and up to a concentra­tion of HHO­ of 1 M by the aid of KPW. But the amount of KPW should be subetoichiometric in relation to Am» otherwise Am­IV would be stabilized rather than Am­VI (see run No. 1 of Table 2 ) . It should be mentioned that Am­VI is stable for л long time Cup to two days) in the presence of the quantity of KPW following from the data of Table 2. Even at higher concentrations of HNO­ (up to 2,5 H> the quantitative oxidation of Am to Am­VI by (Nh\) 2S Og/Ag + can be accomplished, provided that the mole ratio KPV/Am has been increased to 0.6. Somewhat astoniehing, but a rise in temperature so useful in oxidizing Am without KPW has been ehown to lover the degree of Am­VI in the preaenoe of KPW (see runs No. 6­8 and 10­12 of Table 2). At higher concentrations of HN0„ (more than 2.5 M ) , no conditions for oxidizing Am to Am­VI have been found.

In reel separation problems one ahould expect Am­Cm to be accompanied by other ions, e. g. by aaltlng­out agents (Mg ) or by elements of the rare earth group (Ce ). Therefore the influence of other ions on the oxidation of Am under the conditions outlined above has been investigated. It follows from the data of Table 3 that foreign ions introduce additio­nal complications into the oxidation process because it will be more difficult to establish the r^ght concentration of KPW. Under the condi­tions of 100 % oxidation to Aro­VI (see Table 2) the yield of Am­VI de­creases drastically in the ргевепсе of maeroquantitiee of Mg(N0„)g. Fortunately, thie influence can be overcome up to a certain concentration of the possible salting­out agent by an increase in the amount of KPW. Also, if cerium is present in a concentration tenfold that of Am, condi­tions of the complete oxidation to Am­VI have been found lying within a rather narrow range of the KPW concentrations. Thus, the amount of £PW

Table 3. Influence of other ionB on the^oxidatlon of An in the pre­sence of KPW (Cond.: 1M H N 0 v 23 C, 1<T J

H Am, 0.1M S^* , 0.01K Ag ) N Ion and Its

concentration КРИ . Нх10

ч Time to maxi­mum of oxida­tion, rain

Result Valence

of oxidation state, degree (%)

1 1,0 M M g + + 2 180 Am­VI (75) 2 1.0 M M g + + h 90 Am­VI (95) 3 1.0 H H g + +

6 60 Am­VI (100) It 2.0 M M g + +

6 200 Am­VI (15) 5 2.0 H M g + + 10 150 Am­VI (80) / Am­IV (30) 6 0.01 M C e

3 + 6 360 Am­VI ( 5) / Ce­IV (100) 7 0.01 M C e

3 + 100 180 Am­VI (25) / Ce­IV (100) Э 0.01 И Се

3

* •­.о 180 Am­VI (100)/ Ce­IV (100) 9 0.01 M C e

3 + 200 5 Am­IV (100)/ Ce­IV (100)

15?

ehouId be at least 150 % of the concentration of Ce, but already an In­crease to 200 % will reault in Am­IV instead of Am­VI.

Attempts have been made to use the reeulta outlined above for the se­paration of A T and Cm from solutions of spent nuclear fuel in an analy­tical scale. A special method of solvent extraction ­ extraction chroma­tography ­ with tributylphoephate (TBP) or solutions of HDEHP in n­hep­tane as the stationary phasee and HNO solutions as the mobile phases пав been applied. In the system TBP/1M HNO 0.3 H Mg(N0 ) the parti­tion coefficient (Kp) of Am­VI is 20 whereas all the other species of Am (Am­V, Am­IV, Am­III) are ecarcely extracted with Kp<0... An even much stronger extracting power for Am­VI was measured with HDEHP, an 0.2 M so­lution of which should take up Am­VI from an aqueous phaae 0.05 H in HH0_ with Kp^250 /"57. In the solvent extraction systems mentioned above and in the oxidizing conditions Cm remains trivalant (or tetravalent with KPW) and is not extracted. In Pig. an example of the extraction chromato­graphic separation of tracer concentrations of Am and Cm from a rare earths fraction of fission products is shown. A Cellte column has been impregnated with 0.2 H HDEHP in n­heptane. The oxidation at 80°C has been performed with 0.01 H AgNO­, 0.1 H (HH^gSgOg, 0.1 M HNO without KPW. After 15 win. and chilling i­­ an ice­bail, the solution has been loaded on to the column and quickly eluted by the aid of overpressure.

When collecting the whole 1H­HN0­­

Ara fraction some percent of Cm ac­tivity is still present. The laet portions of the eluate are free of Cm. With tracers Am and Cm no enhancement of the separation by KPW has been found. Our explana­tion ia the strong dependence of the resulting valence state of Am from the mole ratio KPW/Am and up to now attempts have failed to eatabish the desired ratio by ta­lcing into acoount the alpha acti­vity of the initial solution.

References 1. Koayakov, V,K. et al./' Sadiokhimiya 1977. Vol.19. P.571. 2. Myaaoedov B.F. et al.//Zhurn. analit. khim. 1971. Vol 26. P.1984. 3­ Hulet E.K.//J. Inorg. Nucl. Chero. 1964. Vol.26. P.1721, 4. Hilyukova M.S. et al .//Radiochem. Hadioanal. Lett.1930. Vol.44. P.259. 5. Zangen M.//J. Inorg. Nucl. Chem. 1966. Vol.26. P.1693,

Extraction chromatographic Am­Cm separation with 0.2 H HCBHP. Column: 4 mm i.d, 100 mm length. 1­ oxidizing solution, 2­ 1И HN0_

160

EXTRACTION TECHNIGUES FOP ISOLATING RADIONUCLIDES FROM TARGETS,1 l Q ­ 1 IRRADIATED BY NUCLEAR PARTICLES. E.s.Y'of , Yu.G.Sevostyanov, A.A.Abramov, M.C.­A.Gighirhanov, A.G. Mackiachcov, V.P.Ovcharenko. Moscow State University, Obninsk Physicr Energetics Institute. USSR

Extraction technigues are most suitable in isolating pure radioacti­ve isotopes from targets irradiatad by nuclear particles in cyclotron or nuclear reactor.

In isolating radionuclides the irradiated target is dissolved,, the extraction of the radionuclide from the solution is carried out. Two extraction technigues are mostly used: a) the aimed isotope is selectively extracted; the organic phase being washed, it is then extracted back into a sustable agueous solution, or b) the extraction removal of the target material and some radionuclides (impurities) is carried on to result in the isotope being produced in agueous phase. After used are also some other technigues for additional purification of the radionuclides obtained. Ir. using solid extraction chromatography the first method is preferable, i.e. to extract the aimed radionuclide separating it from the target material and impuri­ties. For producing pure radioactive isotopes we used liquid extrac­tants in suitable diluents a.nd solid one based on tributyiphosphate. Tributylphosphate and other oxygen ­ containing organic solvents, amines and salts of gurternary ammonia bases, crown ethers and other reagents in diluents were used as liquid extractants. Irradiation of targets on high­energy cyclotrons of /> , d and c£­paxticles cojnmonly results in long­lived radionuclides of two are more elements. Isolation of two or three radionuclides from one irradated target is more reaso­nable and less expensive.

As an example, let us consider the technique for isolating sodium­22 and aluminium­26 from cyclotron irradiated targets made of magnesium or its alloys. Isolation of sodium­22 was performed by 15­crown­5 solvent extraction, that of aluminium­26 by 10 vol.% triocty­lamine of oxalic acid [(C8H­7),NH] ­,C?0. in solution of chleroform in benrene or toluene.

Sodium extraction by 15­crown­5 solution in diluent proceeds as fol­lows:

* ­ "ex + Na

(«) * Att)

+ 15

­kr

­5

<DJ = = N a + lS­4r­5Aw, (1)

where K g x is an extraction concentration constant, A ­ is usually an organic acid anion. Of common use is sufficiently strong picric acid.

п.э«к.зв: 141

As was shown by our investigations the following dependence for hydro­carbons and their chloro derivatives is observed

K

ex * a + b /

$ , l 2 i

where 8 ­ is dielectric of diluent of crown ether. The use of chloro^ form as a crown ­ ether doluent results in rather a big value of cons­tant of 15­crown­5 sodium picrate extraction (lg к » 4,00). Greater extraction constant values are also observed for triphenyl methane dyes for bromthymol blue in particular (lg K e x « 4,9). Values of sodium distribution coefficients may be doubled provided that lithium hydro­xide solution extraction is applied. Lithium is very weakly extracted by 15­crowu­5.

Selective extraction of aluminium (III) from weakly acidic solutions by crioctylamine of oxalic acid is caused by formation of its stable oxalate complexes. Only iron (III) is also well extracted by this extractant. The magnesium target being dissolved in 2,5 M sulphuric acid (excess in relation to magnesium 1,02­1,03) and рн of 1,0*1,5 be­ing established by ammonia. Aluminium­26 is extracted by 0,02­0,03 H solution (TOAH)2C20. in benzene or toluene with 10% of chloroform, the organic phase is washed by 0,02 M solution of oxalic acid. Back extrac­tion of aluminium­26 is made by 6 M HCl.Sodium­22 is then extracted by chloroform solution where the concentration of 15­croun­5 is 0,2 M and that of picric acid is 0,1 M; the organic phase being washed, it ia then extracted back by water. The isotope yield is over 95%, the puri­fication coefficient from radioactive and non­radioactive impurities beins 10 5

­106

.

Let us now consider the production of vanadium­41,<'? a.. scandium­­46 from cyclotron irradiated tatanium targets by means of ^Taction and tributylpbosphate solid extraction chromotography. Values of scan­dium (III) and titanium (IV) distribution coefficients were determined at their different concentrations from solutions of xydrochloric acid of different concentration on salid tributylphosphate reagents in sta­tic and dynamic conditions. At any concentration of hydrochloric acid scandium (III) values of distribution coefficients are much larger than those of titanium, on the basis of these and model experiments the fol­lowing techniques for producing vanadium­48,49 and scandium­46 were proposed. Titanium target is dissolved in hydrochloric ?.*:id with chlo­rine and titanium (IV) is obtained in the solution. On addition of po­tassium chlorate vanadium­48,49 is extracted by dipropyIketone and extracted back by diluted hydrochloric acid, th*2 aqueous solution with hydrochloric acid concentration of about 8,5 м being passed through Uie column 1,2 cm in diameter and 20 cm long filled with soliC tribu­

162

&5#Hl зммс

1 0 20 60 100 150 200 300 Eluent Volume

Elution of scandium­46 and titanium (IV) from solid tributyl­phosphate reagents

tylphosphate at the rate of 0,5­0,6 ml/min. Titanium (IV) is totally washed by 150­200 ml 8,5 M hydrochloric acid and then scandium­46 is eluted in 30­40 ml 3 M hydrochloric acid. The isotope yield is over 95% radiochemical isotope purity being ower 99% the amount of in­active impurities of many elements being less than 0,05­0,5 mg/ml solution. Typical elution curves are given in Fig,

Let us consider the techniques for isolating technetium­95 m and technetium­96 and molybdenum­93 from «^­particle irradiated niobium targets. Irradiated niobium is dissolved in a teflon vessel with hyd­roxide melt at 240­280°C. The melt is dissolved in water with hydrogen perrxide, then technetium isotopes are extracted by 5­10 M isolution of nitrate (chloride, bromide) of ethyltrioctylammonium in toluene. Technetium back extraction is carried on by 1 M ammonium carbonate solution after 70% dolution of the organic phase by amyl or hexyl al­cohol. And then on reducing cesium hydroxide concentration molybdenum­­93 is also extracted by QAB salt.

Thallium isotopes, thallium­201,202 are produced by proton and deut­ron irradiation of mercury. Using isotopeenriched mercury it is possib­le to produce thallium­201 containing negligible amounts of other thallium isotopes. Irradiated mercury is dissolved in nitric acid. On

163

addition of hydrochloric acid and sodium chloride thallium-201 and oold isotopes are extracted by an ether (chlorex, di-n-butyl ether, etc.). The organic phase being washed by chlorine saturated hydrochlo­ric acid thallium-201 and gold isotopes are extracted back by a weak solution of sodium hydroxide. Gold isotopes are quantitatively removed from a weakly acidic solution by metallic copper cementation. We have developed the similar technigue to produce thallium-201 from thallium via lead-201. The irradiated thallium target is dissolved in 5 M sup-huric acid and on addition of sodium chloride the target material -thallium is quantitatively removed by multiple extraction by chlorex or another ether. Thallium-201 after it's accumulation from lead-201 is extracted by the same ether and extracted back by water or a weak alkali solution. Thallium-201 yield over 90% may be expected.

In coclusion, the techniques for producing silver-105 from oC-particles irradiated rhodium targets may be given.

Cyclotron irradiated rhodium is melted with tin in a graphite crucible. The cooled melt is treated by hydrochloric acid to result in total dissolution of tin, and then amorphous rhodium is dissolved in sulphuric acid, silver-105 is extracted from 2-3N, H-SO. by 0,05% dithizone solution in carbon tetrachloride, 97-948 of silver-705 being extracted at one extraction with rhodium separation coefficient 4 (3-4)«10 . Silver-105 yield is no less than 85% with isotope purrty 95% (5% of silver-106 impurity).

We also developed extraction techniques for isolating the radio­nuclides of tungsten- 181, tringsten-178, osmium-135, rhodium-102 and others.

164

GROSS AND RADIATION YIELDS OP PHENOL DERIVATIVES IN I .__„ I NITROBENZENE SYSTEMS WITH AQUEOUS PHASE AND ITS USE 1 ­ 1 FOR EXTRACTION OP NIOBIUM S.6echova, F.Macasek, R.Cech, Department of Nuclear Chemistry, Comenius University, Bratislava, Czechoslovakia RadiolysiB of nitrobenzene is of interest because nitrobenzene is

used as an extraction solvent for radioceBium recovery P\]. It was indicated that the radiolysis of extractants should be investigated in the presence of an aqueous phase /2,37, e d the theory of the ra­diolysis in two­phase systeme was elaborated by ue ^4i57­ As a con­sequence of the theory an aprlori "two­рпаве additivity rule" for radiation yields in emulsions was formulated 1УЯ'. Its importance was demonstrated on the nitrobenzene­.»ater systems. Partial radiation yields in the phases being different, the autoradiolysis in actual systems depends also on the phase ratio and specific activity of aqueous and organic phases and it cannot be predicted accurately from the data obtained from irradiation of separated or even statically contacted phases A3. In the present work we investigated the forma­tion of nitrophenols in irradiated emulsions of nitrobenzene, estab­lished partial and gross radiation yields of phenols and nitroanlline.

Table 1. Gross radiation yields of the products. The phase volume ratio is varied from 1:9 to 5:5

r e t aqueous phase Product н2о 1.5 M­H 2S0 4 зм­шо3

p ­ n l t r o p h e n o l 0 . 2 5 ­ 0 . 5 3 0 .36­0 .45 0 .05 ­0 .14 ra­nitrophenol 0 .41­0 .95 0.61­1„37 immeasurable p ­ n i t r o s o p h e n o l ­ шгше&одгаЫе 0 .04 ­0 .07 2 , 4 ­ d i n i t r o p h e n o l ­ ­ 0 .26 ­0 .29 2 , 5 ­ d l n i t r o p h e n o l ­ 0 .27 ­0 ,42 o­aminophenol 0 .14­0 .58 0 . 2 3 ­ 0 . 5 0 ­m­aminophenol 0 .12­0 .37 0 .17­0 .40 ­p­aminophenol 3 .15­0 .48 0 .19 ­0 .47 ­o ­ n i t r o a n i l i n e ­ ­ 0 .27­0 .47

Nitrobenzene phase was found to be several times more reactive than the aqueous one In respect of the phenols and aniline derivatives formation. The yields for organic phase (GJJO indicate that the latter

;

* one ia a more reactive phase, i.e. the products are formed preferably

165

i n nitrobenzene saturated by water. However, a contr ibut ion of the aqueous phase t o the gross rad ia t ion y i e l d s i s far from being n e g l i g i b ­

l e for a l l the productB, except perhaps the aminophenols i n pure water. In 3 H­HHO, s o l u t i o n about 30­50% of dini trophenols and o ­ n i t r o a c i l i n e i s produced i n aqueous phase.

F i n a l l y , products of r a d i o l y s i s of nitrobenzene­water systems were

Table 2 . P a r t i a l rad ia t ion y i e l d s of the products . Gj­aqueous phase, Gjj­organic phase

1 Product

Aqueous phase I 1 Product н2о 1.5 M­H 2S0 4 зи­шо3

p­nitrophenol

m­nit r ophe nol

p­nitrosophenol

2 ,4 ­d in i tropheno l

2 ,5­d in i trophenol

o ­ n i t r o a n i l i n e

o­aminophenol

m­aminophenol

p­aminophenol

Gj 0.16+0.02 G

I I 0.97+0.07 Gj 0.39+0.05 G n 1.63+0.15 Gj G

I I G

I G

I I G

I G

I I G

I G

I I Gj. 0.02+O.OOB aT1 1.02+0.030 Gj 0.06+0.002 GJJ 0.60+0.010 Gj 0.07+0.004 GJJ 0.80+0.10

0.33+0.02 0.55+0.08 0.40+0.01 2.15+0.33 unmeasurable unmeasurable

0.16+0.003 0.74+0.010 0.11+0.004 0.61+.0.010 0.12+0.004 0.75+0.010

0.04+.0.01 0.22+0.02 unmeasurablc ^ immeasurable 0.04+0.01 j 0.09±0.02 0.26+0.01 0.29+0.01 0.25+0.02 0.55+0.03 0.23+0.01 0.63+0.01

inves t i ga ted as an ertractant for separat ion of niobium. The best resul tB were obtained with o­aminophenol, p­aminophenol and o ­ n i t r o ­

a n i l i n e . In Pig . i s Djjb as a funct ion of concentrat ion of n i t r i c acid for o­aminophenol ( o ) , p­aminophenol (A) , and o ­ n i t r o a n i l i n e ( o ) . Aqueous phase was: O.t ml 0.9x10"*K " z r in 6 M И90, + 0.1 ml 1.7x10~

2

M Zr(K0 3 ) 4 ­ 5 H 20 i n 6 M HS0 3 + z ml 1.5 M HHO, (z was varied from 0.B6 ml t o 4.2 m l ) . Total volume of aqueous phase was 5 .0 ml. Organic phase was 5.0x10

_ 3

M phenol + 5.0хЮ"г

Н HDCC (d icarbo ly l c o b a l t i c a c i d ) .

166

Dj b as a function

of concentration of nit rle acid (o) o-amlnophenol; (u) f-aminophenol; (a) o-nitroanlllne

0 6.5 1.0 [HKtjtMl.te'3)

References 1 . SeluckJ P . , Vanura P . , Bale J . , Kyre M.//Badiochem.Badioanal.

Latere 1979- V o l . 3 8 . P.297. 2 . Warner B . P . , Naylor A . , Duncan A. , Wilson P . P . / / P r o c . I n t e r n .

Solvent Extract ion Conf.ISBC 74, byon.London.Chem.Soc. 1974. Vol . 2 . P .1461 .

3 . Mowak M., Kowak 2 . , Bochon A. / /Radiochem.Hadioanal .Letters .1974. V o l . 2 0 . P .47 .

4 . Macaaek P . , 8eoh E./ /Radlat.Pnys.Chem. 1984 .Vol .23 . P .473 . 5. «aceeek P. / / I b i d . 1904. 7 0 1 . 2 3 . P . 4 8 1 .

75

50

167

KOBIPICATION OP AQUEOUS REPROCESSING TECHHOLOGY TO REDUCE iiXTSRHAL NUCLEAR FUEL CYCLE

10­14

A.S.Ifikiforov, B.S.Zakharkin, E.V.Henard, A.M.Rozen, V.S.Vlasov, E.A.Pilippov, A.A.Pushkov, G.I.Kuznetsov, E.Ya.Smetanin,State Committee for utilization of Atomic Energy, Moscow, USSR The paper discusses possible modifications of the aqueous technolo­

gy for reprocessing high burn­up (—100 QWday/t) and short cooled (3­b months) nuclear fuel;presents the results of H&D work carried out at pilot plants. The major difficulty ­ the possible destruction of an extractant during processing solutions of the activity of the order 3

10 Ci/l­is obviated through the UBage of short­residence time centii­lugal extractors ICE). The studies into the extraction kinetics showed tnat the chemical reactions of TBP extraction of many elements is ac­complished in the time of order tenth parts of a second [i]. This fact permitted us to expect rather a high extraction of valuable elements in a cascade of CE,

High efficiency CE's for the radiochemical processes (Fig.l) are Deing developed from the base designs[2^ana °an be serviced with mani­pulators. Feed solutions are fed through pipes 1 to a labyrinth mixer­blower 2 where they are agitated and forced into a mixing chamber 3. Emulsion is fed to rotor 4 where it is divided under the action of | centrifugal forces. The divided liquids are directed into circular collectors of a fixed body 5»then they gravitate froo the extractor. Rotor unit 4 with drive 8 is remotely replaced by a reserve one. The extractor can be also used for slow mass transfer processes [3} The needed time of phase contact is ensured through an installation of mi­xing chambers 3 having different volumesjfl.The mass transfer effective­ness in CE using a 30% TBP in synthin­uranium­HHO, system is more than

J> 9555. The effectiveness can be raised through recirculation [3} Tjje disperse aqueous phase re­

circulation,e.g. «during U0„ + extraction from a sulphuric acid solution by DaEHPB increases the mass transfer effectiveness from 75$ to 94$ at

_J3 the time of a pha^se contact 7s. A severe prob­

lem is a possible formation of a second orga­nic phase at a high Pu concentration that is ? caused by a limited solubility of TBP­Pu solva­tes in a diluent (normal para:ffine hjdrocarbons

^ ­ИРН). A series of experiments were carried out (

to select extractants having «in extraction abi­lity at the level of TBP,but forming solvates with actinides(IV) compatible with a diluent. On the basis of work [5] a conclusion is drawn ^ [6­9] that trialkyl phosphates (TAP) are advisab­^ le,if their hydrocarbon chain is optimized,i.e.

168 1. Centrifugal

extractor

elongated and the iBostructure is used. The first member of the series ­ trlisoamyl phoaphate(TIAP,nc=15) was found to give solvates with Pu(IV) that are compatible with a diluent,however,solvates with Th(IV) form a second organic phase. Diisobutyllsooctyl phosphate(DIBIOi,n«t6) turned out to he the first versatile extractant for actinides(IV);its solvates with both Pu(IV) and Th(IV) are compatible v Л ЫРН;solvates of actinides(IV) with triisohexyl phosphate(TIHP,n =18) do not give the second phase [6­9] . The above three extractants were tested in a cascade of OE along with TBF (a further increse of n seemed to be inadvisable due to difficulties involved in washing from long­chain destruction products). It should be noted that In the USA TIAP ПНР and triisooctyl phosphate(n0"?4) were recommended. Succepsful checking of the latteT in hot cells was reported fo] .methods of the effective washing from destruction products were developed.

Mow consider the results of the experiments. Puel used. The fuel(20?S Pu,80# U) of the BR­10 and BOR­60 of the

burn up to 10?S(~100 GWday/t) cooled for 3 months to 2 years was repro­cessed.

Shearing of fuel before dissolution. The method of chopping can be used and in some experiments it was used,however,when handling fuel cooled for less than 1 year modification of some aggregates that would provide for high­speed chopping,or thermal methods with the help of which a fused structural material would be separated from fuel fll] are required. Being an alternative to mechanical processes these methods are applied to fuel held practically for any period of time. As the experiments performed in pilot facilities show,the fusion of stainless structural materials at 1450­1550°C in the inert atmosphere makes it possible to reach adequate process parameters;the capture of steel by fuel does not exceed 3%>Plutonium losses with the separated steel are 0.01$,those with ceramic lining are e.02%;the total yield of fuel com­ponents is 99.9%.

The combination of shearing with the oxidation permits removal from the fuelCbefore its dissolution) and localization, of up to 99% tritium as well as a larger portion of Kr.Xe, " l , °I,Ru and other nuclides.

The dissolution „as performed in 8­9 mole/1 HMO, at 100*5°С for 6 hours. The S:L ratio was chosen based on condition of receiving 250 g/1 (U+Pu) in a final solution (prior to mixing with scrubbing water). Additions of F were required for the final dissolution of hardly so­luble residues aa dependent on their Pu content.

The clarification of solutions was performer in two stages: a coa­rse one (up to 30­100 mg/1 content of dissolver residues) on a centri­fuge,a fine one(up to 5 mg/l) using filtration through a metal powder.

The clarification centrifuge development [12] showed that the separa­tion of fine dissolver residuces from the feed solution (about 90$ particlates<3 mem dia) the separation factor of the order of 10­* is

169

MOO r—**T^I ' *л »­г : м Н Е Й '•> "У, ,"?,'."?, , . , , ."!*. |т|»|в|г»|;||ф|я|ю|ф1а|и|)(|[)| \к]гЯ

j ~~"Ц""г%

Pig. 2. The scheme of the process (the dotted line ­an extractant).Г ­ a tho­rough gamma­absorbtiometer; A ­ a thorough alpha­radio­meter on organic (An) and aqueous (A_) flows

adequate to clarify solutions up to 30­100 mg/1. After centrifugation the feed solu­tion contained up to 100 mg/1 residuces. The filtration characteristics of the so­lution were determined using stainless steel powders of twc fractions (grain dia 0.3­0.4mm and 0.4­0.6mm). At the layer height of 400mm and specific throughput 6 m

3

/m the filtration rate wee 1.3 and 2.2 m/h (P­300mm water column),respectively! based on these data the solution was fil­tered through a three layer powder filter 32mm dia,with the total layer height being 600mm. The essentially full clarification was reached under the load 1­5 m 3

/m2 (the

<v Table 1.

capacity of 1 Si

was checked in the Composition of Feed Flows

V h ) . (the first cycle)

43 CB's cascade. The

Feed to(Index( stage M

Plow Chemical composition Flow rate 1/h

F Feed aqueous 4 . 3 m/1 НН0Э,165 g / i ц,23 g / 1 Pu 0.97 ~~ZZ Fresh ( P e c y o ­ 30­40* TAP mixed with normal

r

"' l i n g ) . x t r a c ­ paraf f ins ( C 1 r . ­ C . , ) ( P u < 0.1 mg/1, 2.0 tans IK1.5 mg/1)

1 U 1 b

13 P_2 Strong acid scrubbing

T5 Р­Э Weak ac id scrubbing

10 m/1 HN0 3(10 g/1 NH 4V0 3) 0.12

0 .5 m/1 ЕЖ) 0.27

TT P­4 Fresh extractant Composition see P­1 0.80 21 P^5 A c i d i f i c a t i o n 10 m/1 ШЮ­,0.05 и/1 Н НЛГО,

~ЯГ of U scrubb: ling

ant P­fc Main reductant 0 . 6 m/1 НИ0.­75 g / 1 U,inc luding 0.2b 70 g / 1 U ( I Y

J

j , 0 . 2 m/1 H 2 H 5 H0 3

A u x i l l i a r y 0.15 m/1 HN0,,11 g/1 U,inc luding 7Г77" raductant 10 g / 1 D(IV)fo.1S m/1 И H HO,

u

'4 0

2­5—2 32 P­7

40 P­8 Uranium stripping agent 0.015+0.005 m/1 HNO, 3.2

P­9 Washing 5* Ha,C0 43 2 3 0.40

Outcoming Flows Index Flow U \ Pu |HN04 m/l) Others

100 Aqueous­tail solution 2.4 mg/1 0.5+0.4 mg/1 4.3+0=2 loo U,Pu ­ extract 68 g/1 12 g/1 300 Pu ­ product (strip) 0.4 g/1 35 g/1 7.1+0.3 0.07 m/1" 400 Soda solution from 5 mg/1 1 mg/1

54 g/l 0.06 mg/1 30 mg/1 IPO.

washing 600 U ­ product (strip)

800 Recycling extractant 1.5 mg/1 0.09 mg/1

170

52 mg/1 IPO 41

achene (Fig.2) Incorporated the co­extraction of (Г and Pu (atage6 1 ­10),two­zonal scruDbing­10 mole/1 HNOjdi­13) and 0.5 mole/1 HS0.J (14) separation of U/Pu using U(IV)'as a reducing agent (17­32),stripping of D (33­40) and soda washing of an extraotant (41­43)• The composi­tion and the flow rate are given in table 1;up to 10 g/1 of ammonium vanadate were introduced into the strong­acid sorubbing flow to oxidi­ze Np to Hp(IV).

Contact time (residence time in a mixing section) was 2­3s at sta­ges 1­16 and Z1­32;5­6s at stages 17­21 and 33­40;10s at stages 41­43­The separation time was 20±5s.

The results of the experimenta performed using TBP.TIAP and TIHP are listed in table 2. The data of the DIBIOP extraction differred insignificantly from the presented ones.

Table 2. Characteristics of I cycle performed in CE and mixer­settlers using spent FBR fuel,TBP,TIAP and TIHP

Characteristics Experiments in mixer­settlers

Extraetant ЭОЭб IBP 0,052 0,0056

<1,9

Experiments in СБ 30* TBP 1 36* «APJ4C* М Ю

Pu Ир

0,007 0,004

<4,6 0,008 0,004

<5 0,006 0,004

<4,6 Losses with aqueous and soda solutions ,f>

Decontamination factor of U from Pu (DP U/Pu) 6­10

S up to »,5­?05

DP Pu/U uo to 15 40 ­ 660 DP U/r 2.5­10

5 1.2­1.5 10 5 1.5­10

5 1.6­10

5

DP РиЛГ 1.3­10+ 1.. ) ­ 4.0­10+ ОТ 17/Но >100 >11 >12 >12 Consumption of U(IV) per strlooine.kK/ke 1,4 1,0 1,0 1,0 Saturation of extractant in I unit.e/1 of (U+Pu) 83 82±2 83 82 Accomulation of acid alkyl­phosphates in extractant ( prior to washing),mg Р0д/1

' £22 <10 <10 <11

Content of these products in recycling extractant after washing,mg Р0л/1

4­8 5.1 5,0 6,0

U content of recycling extractant. rnx/1 1,6 1,5 1,5 1,4 Frequency of phase exchange in I cycle »12 70­26­87 for I­II­III units

The direct extraction from the feed aqueoun solution of the 300 ­900 Ci/1 activity was for U and Pu ­ 99.998,Np ­ up to 98* (goes to the plutonium product solution). In the CE experiments the decontami­nation factors from y,S­emitting fission products ­ were: for If­REE > 7.7*10

u

iRu, im 3.1x103

;Cs>5.1x106

;So>6.1x105

;Zr ,Nb ­ 3.0x10+. Рог

i7 i

Pu­REE >2x106

;Ru,Rh ­ 1,8x103

;Cs > 7.8x105

;Sb > 9.3x104

;Zr,lib ­ 4.5x102

. In the experiment described the number of phase exchange was 60­100 (in the I unit),20­40 (in the II unit) and 80­110 (in the III unit). See fig.3 for the concentration profile in unit I under one of the conditions.

Stage Number Fig.3. Distribution of lT,Pu,HH0, over the stages of the first unit

(the regime with 15Й Pu (У1):х|},0 = 4 mole/l.XjjjjQ (P2)­(6 mole/1),3%= 125 g/l.Xp^ 13 g/1). The dots ­ 3

the experiment,3

tbo lines ­ the calculation by the model [17] with the refining in equation 21 the ionic strength >'=Х

Щ(0 + ^pn without taking account of uranium

On the Reductive stripping. A decrease of the consumption of tf(IV) from 1.4 to 1.0 leg/kg and a decrease of the phase contact time by a factor of several tens as compared to the experiments in the mixer­settler did not interfere with the high­quality purification of U from Pu and actinides from fission products.

On the distribution of Up. In the absence of an oxidation agent the short contact time in the extraction zone did not allow an adequate co­extraction of Up with U and Pu,due to this fact more than 5056 Np remained in the aqueous­tail flow. The introduction of a weak oxida­tion agent ­ V(V) ­ resulted in essentially a quantitative extraction of Ир with U and Pu in unit I. The oxidation process is so fast that the . ,'dation agent was worked for a short period of contact on 13 stat,—* (only 30 s in the mixing chambers). Further on Wp left together with Pu product. Comparisons between the reprocessing of fast breeder reactor mixed fuel in pilot mixer­settler and centrifugaltextrmctore (table 2) show that the latter units solve not worse than mixer­sett­ler the problems of quantitative extraction,decontamination from fis­sion products and mutual separation of actinides using the same num­ber of stages ­ and all this is done at 2­3 s phase contact in the CE mining chamber. As compared to the results received in mixer­settlers the less radiation­chemical destruction of an extraotant (when going from pilot to full­scale extractors this difference increases several

172

times),the practicability of quantitative extraction of Mp,decrease of U(IV) have been confirmed. The version of the electro­chemical se­paration of U/Pu was also successfully checked in the СБ cascade. The 2­nd extraction cycle was also checked (the additional uranium DP 10 was reached) . The use of TIAP and TIHP practically did not change the performance of the flow sheet,the use of DIBOP slightly improved it. The limits for Pu broke away.

Finally,experiments were performed on high extraction of trans­plutonium elements and lanthanides from the first cycle raffinates (directly and after concentration by evaporation) using bidentate organophosphorus compounds (for their properties see the stand paper by Rosen A.M. et al at this conference as well as {l3,14j­ Using tetratolymethylene diphosphine (0.5 mole/1 in tricolorоbenzene) prac­tically full extraction with 5­fold concentration was reached in a single stage,the extraction with ditolylbutyl­carbamoylphosphine oxi­de (tolg/bUo) required 5­6 stages.

Preparation of mixed fuel. Mono­ and polydisperse fractions of oxide microspheres prepared by a sel­gel procesa„are suitable for the re­fabrication of fuel in the form of granules and pellets. The ab­sence of a duet permits the activity to be localized within the wor­king equipment (see [15,16] ). Among the versions of sol­gel proces­ses the method of internal gel­formation seems promising. True or colloidal mixed high concentration U and Pu solutions are subjected to spheroidization in the organic phase. The solid phase precipita­tes as microspheres as a result of the pre­introduction of donors of ammonia ­ hexamethylenetetramine and urea­ into the aqueous phase, containing metals. With a temperature increase both inorganic and or­ganic constituents polymerize quickly in a drop of the working solu­tion. An organo­mineral phase forms 'hich as a result of washing with an aqueous solution of ammonia and treatment transforms to a stoi­chiometric compound (oxide).

The main difficulty was in the engineering for the prooess. The fact that drops of an aqueous working solution formed above the sur­face of a heated disperse phase resulted in the formation of very fine drops­satellites that interfered with the hydrodynamic and gra­vitational conditions of the operation of a gelation column. The un­loading of the microspheres,their washing with ammonium,transporta­tion to the subsequent technological operations were performed batchwise,which did not provide for the remote control of the process as a whole. The quality of the mixed U­Pu oxide microspheres was not stable which did not permit the guarantee of the quality of the pro­duct fuel. The investigations under way are primarily aimed at the preparation of monodisperse fractions of microspheres through a new

173

organization of a di • formation process,the continuous unloading of microspheres from the gelation column.washing and transports' ion to the subsequent thermal operations.

Conclusion. The experiments carried out in the pilot facilites show that the pyrochemical removal of clads.the use of short­resi­dence time centrifugal extractors and extractants with the optimized hydrocarbon chciu permit with confidence the employment of the aqueous reprocessing technology in the shortened version of the exter­nal fuel cycle.

References I.Tarasov V,V.,Yagodin G.A. et al. Kinetika extrakcii neorg. ves­cestv.'/ Itogi Nauki. Keorg.chimia.Part II.Moscow:VIHITI,1984.P.171

2.Kuznetaov G.I.,Pushkov A.A. et al./At.energija.1086.61.И 1.Р.23. 3.Xuznetsov G.l.,Puehkov A.A. et al./4sEC­80.P.2(r 4.Kuznetsov G.I.,Pushkov A.A. et ai. // Issledovat ..ja v oblasti pere­rabotki obluchennogo topiiva (IV Symposium SEV . Part З.КАБ С 3SH, Praga,1985.P.55.

5.Rozen A.M. et al.//J. Inorg.Mucl.Chem. Supplement.1976.P.220. 6.Rozen A.M.,Hikiforov A.S. et al. Avt.svid. Ur

SR N 841140;Bull. isobret.­1982.K 14.P.319.

7.Hikiforov A.S.,Shmidt V.S.,Rozen А.И. // XIII Mendel.s'jjezd. II. Nauka,1981.N 1.p.182; RadiochimiJa,1982.24,N 5. P.633.

S.Rozen A.M..Kifciforov A.S. et al. //DAM USSP. 1985.274,M 5.P.1139. 9.Rosen A.M.,Nikiforov A.S. et al./At. ene iija.1985.59_.P.413. Ю.Сгоиае D.J..Arnold W.D ,Hurst P.//1SEC­8 p.451. 11.AgeeEkov A.T.,Bibikov S.E. et al./At. < iergi3a.1977.42,H 5.P.378. 12.Rozen A.M. .Levishev A,K. et al.//At. eziergi;Ja.1976.40,K 6.P.467. 13.Rozen A.M.­. et alJ/Radiochimi3a.19S6.28,M 3.P.407. H.Rozen A.M. et al. Avt.svid. USSR N 601971;Bull.isobret. 1979.» 35.

P. 257. 15.Filippov E.A.,Papkov A.S. et al/'te .iochlmi;[a.1980.22,N З.р.305. l6.Pilippov E.A.,Karel±n A.I. et al./ iadiochimi;Ja.1984.N 2.P.225i

III Vsesojuzn. conf. po chimii uruna. H.: Nauka, 1985. P.100. 17.Rozen A.M.,Andruzky L.C.Vlasov A.S.//At. energija.1987.62.N 4.P.227.

174

с

HOW TO AVOID Pu­ACCUHULATIONS IN PUREX EXTRACTORS 10­16

G. Pet rich and H.Schmieden lns t l tu t fur H«l9e Chemie,

КегпГог*с1ит$вгетпж Karlsruhe, Postfach 3«40 ; D­7600 Karlsruhe, PRG

The effect of Pu accumulations in the Purex first cycle high active extraction and

scrub extractors HA and HS « a s first described In detail by Rozen ill A varie ty of process conditions may cause such accumulations: reduced solvent flow, reduced TBP

concentration. Increased hiph active feed concentrations or flow, low acidity etc. The

Pu accumulations in ft counter ­cur ren t extractor occur by loading the solvent close to

i ts theoretical limit. An Insufficient solvent supply creates a "bottleneck" for

organic Pu t ranspor t under usual flowsheet conditions and Pu is forced to the aqueous

phase. Close to the raffinate loading Is low and Pu Is extracted again. T h b recycling

allows Pu to accumulate. The smeller the solvent deficit, the slower the bui ld­up cf

the Pu peak, but the larger i t s maximum concentration. A large solvent deficit causes

a rapid propagation of the peak towards the aqueous raffinate outlet . The peak

deflates ra ther rapidly when i ts tall reaches the raffinate. As a consequence the

maximum peak concentrations increase with increasing extractor length.

Recently Sch6n et al. [2l observed peak concentrations of up to 40 g Pu,l (L.WR feed

composition, conventional flowsheet) in the miniature pulsed column facility MINKA

(column length 360cm). Theoretical calculations using the computer code VISCO [3)

predict possible t ransient peak concentrat ions for technical size columns in excess of

80 g Pu/1 under the most unfavorable conditions. Even though the bui ld­up time gf

several days will hardly pass unnoticed a potent ial crlt lcali ty problem exists and

requires additional design restr ict ions or et least highly reliable safety

instrumentation.

Standard Purex flowsheets for the HA/HS extractors favor the possibility of Pu

accumulations: 1ДУ1) has the highest extractabl i l ty and expels ell other species from

the solvent a i high loadings. Extensive studies using VISCO in order to minimize pu

accumulations predicted a possibility to completely avoid this risk. The overall HA

extractor temperature or at least the temperature at the location of the "bottleneck"

has to follow:

T Ь Ко.ОбвТби­О.ЗЗбТР­зат^Ю­ехр^.ООООв^&Н­Ш+РМ.Н­ 0­* 3"! + 401 (equ.l)

T is in *C, U and P are the aqueous U and Pu feed concentrations resp. in g/(, H is

the НЖЪ feed concentration (M.'l) This minimum extractor temperature does not

depend on TBP concentration. It reflects the temperature and concentration conditions

for which the distribution coefficient of Pu(lV) becomes larger than the distribution

coefficient of U(V1). If the condition In equ.i is met, Pu(lV) forms the strongest

TBP­complex of all species invoiced. It therefore expels U(VI) from the organic phase

and Pu cannot be recycled any more within the extractor: Pu accumulations in the V\ extractor are reliably avoided. There may be uranium accumulations Instead but even

for FBR feed composition they will amount to only a few g/1 and will be o/ no

practical significance.

Equ­l reflects the fact t ha t the distribution coefficient ratio DM/Do

Increases with Increasing temperature. Increasing acidity and Increasing metal

175

concentration. The equality in equ.l holds if DpB/Du=l defining the critical temperature T c for Pu accumulation. No Pu accumulations can occur a t DP»/DE^I (or ТйТс). Те will be the lower the larger the feed concentrations are. For composite extractors such as HA and HS the composite feed concentrations (e.g. HAF blended with HS­reflux) have to be insert ' in equ. l .

Fig.l shows the U and Pu concentration profiles in g/1 for a small lab

i scale B­stage mixer set t ler operated at 50*C. НЛХ OOVTBP, 2S8mLh) enters at stage l . HAF (150ml,h. 211 gLU, 2.35 gPu/i, o . ^ H HNu3) enterb ... si­as^ t'. Theory

I апи • •L.Cn'H'it ' j r e :

aou с б : .iu lines and triangles,

v i s c o organic = dashed lines and circles.

Fig.2 shows measured and calculated concentration profiles for an operating temperature of lS-ZO'C with a/1 other conditions kept identical to fig.J. While the feed composition Ь.-Ри is 90, raffinate If, Pu ir 2300{i) in the 50"C experiment shown in f.^.J, and 60 in the low temperature test of fig.2. A further tes t run at 20*C and 3M HNOa <Tc=45 :C) increased Pu со и". Pu=i5 in the raffinate and showed the expected Pu accumulation of 7g/l in stage 4. The 18­20'C experiment shown in fig.2 was operated

almost at Tc=20.4"C. U and Pu show a very similar behavior anf Caere is ever, a slight indication of a Pu peak. A further experiment in the 350cm A column of MfNKA was run at a

•ninal 20 JC using the feed composition of

i figs.l­2. But In this column the heat of extraction resulted in a

v i s c o temperature of 37'C at the location of the

"bottleneck" which is in favor of the new process: no Pu accumulation was observed and the raffinate ratio wa' L'/Pu^iao,

These results demonstrate tha t not only Pu accumulations are rellahly avoided by the new process but tha i in the case of a maioperatlon Pu is even depleted in the

176

— aqu — org Fig. 2

raffinale stream (it is by no means necessary to have U losses as large as in the tes t s of f igs . ] -2 to avoid Pu peaks). This allows to exploit the advantages of an ultra high loading process reducing the risk of Pu losses, AS demonstrated in the ORNL work [4] the position of the U extraction flank can be kept constant by automatic control using suitable in- l ine sensors (temperature, light absorption u c > . Instead of keeping the flank position close to the feed point [4) it can be moved close to the raffinate. dependent on how large U losses can be tolerated. The reward is a highly enhanced fission product decontamination.

Fig.3 shows measured and calculated U and Pu profiles for an 8-s tage HA extractor (stages 1-8 and an 8-stagfi HS extractor (stages 9-16) . HA is operated at 50*C. HAX (29VTBP. 327 ml/h) enters at stage l, HAF (145 ml/h. 230 gU/1. 2.87 gP':.l. !20 mgNp/1, 1 gZr/1, 1 gNb/1, 320 mgRu/1. 5.05M HNOa) at stage 8, HAS (65 ml/h. 3M HNOj) at stage 16. No Pu accumulation is observed in HA where the new process conditions apply (Tc=33'C). HS is operated at 2o'C. while Tc for the HS extractor is 5 9 T and Pu accumulates indeed.

F i g Q

10"

10

Fi£.4 shows the measured concentration profiles of Zr and Nb. Ordinates are % of the feed concentration. Both t,pe ies form weaker TBP complexes than U(VJ) and therefore accumulate in the lower stages of Нл where solvent loading decreases. The mechanism for this bui ld-up is the same as for PutIV) in a Ю conventional flowsheet as described above.

Excellent decontamination is achieved since the tuils of thu accumulations do not extend to the organic product. Ru does not even accumulate and is found 1004 in thi* raffinate as was to be expected.

In the example of figs.3-й Np(Vi) and Np(V) are reduced in HS predominantly ю 12.3ак.382 177

Z r Nb

9 13 T - a q u

1 org

i3 : v i s c o

Ия-4.

N p u l ) by operating the HS scrubber with ejectroreduction at a current density of about 6 IDA/cm*. Under these conditions Pu(lV> and U(V'l) are also reduced in sraa)] quanti t ies but they are rapidly reoxidized in the HA extractor as can be seen fro» the small Pu losses to the aqueous raffinate. Fig.5 shows the measured concentration profiles along with Np calculated as Np(IV) (left plot) and Np(Vl) (right plot); Np redox reactions are not yet Implemented in VISCO. From the Np slope in the HS part (stages 9-16) of fig.5 It Is obvious t ha t l i t t le Np(YI) Is present In the scrubber.

whereas in HA N'p is 1 predominantly hexavalem.

— a q u — o r g

I Np(lV) entering HA reoxldizes, the tailing in stages I and 2 is at t r ibuted to the almost inextractable Np(V) leaving with the raffinate- The Kp accumulation is caused mostly by Np(Vl) but due to electroreductlan a Np(Vl) tail cannot extend into the organic product: Np is quant i ta t ively removed from the product

Tc behavior was also studied in the new process. High temperature and high acidity are opposed to a good separation in the presence of Zr [5] and Tc was almost completely found In the product. Electroreduction did not reduce msjor amounts of Tc(VH) in the scrubber. Without Zr about 80% Tc was removed. Further experiments to reduce the Tc extraction factor at a lower organic to aqueous flow ratio are in preparation.

Increased solvent degradation was expected for the new process. The measured concentrations of HDBP were 20 to 30 mg/1 and seem to be only s i g h t l y increased as compared to the conventional process.

The new process will be run with high burn-up/shor t cooling time '-'BR fuel in 1988. By the presented resul ts we are encouraged to continue our efforts to design a one cycle Purex process 16[.

References 1. A.M.Rosen//Atomnaya energiya £, 27? (1969). 2 . J.Sehon, H.J.Bleyl, M.Kluth//Extraction'B7, ICE Symp.Ser.103, Dounreay 1987-3. G.Petrich, U.Galla, H.Gotdacker, H.Schmleder//Chem.Eng.Scl 41., 981 (1986). 4r D.E.Benker et al.^ENC'86 Transactions Geneva, June 1-6, 1986, vol.4, 109. 5 . J.Garraway/VExtractlon'84, ICE Symp.Ser.88, Dounreay 1984-6, H.SchraJeder, H.J.Bleyl, U.Galla, H.Goldacker, G.Petrich, L.Flnsterwalder//

lnt.Conf.-RECOD 87". Paris Aug.23-27, 1987.

178

ACOTMULATICB OP TEmmLEHT ACTTJTmS НЕСГСШ) IH THE 30 VOL.* ТИВ0ТПЯаЭЗШАТВ­К.1ХЮК;АНЕ AND HITRAIE­ SYSTEM: EXPERIMHJTS AND ITS ' ' ARAH5IS BY MATHEMATICAL MODEL S­Tachimori, H.Ami, S­Usuda, S.Sakurai, Japan Atomic Ihergy Research Institute, Tokai­mura, Naka­gun, Ibaraki­ken, 319­11» Japan

IHTROOTCTICH In the POREX process of the spent fuel reprocessing, the actinide elements and the

fission products distribute between the organic and the aqueous phases depending on the concentrations of various components involved in the system. Therefore i t is not surpr is ing to have been pointed out tha t some elercents e.g., plutonium(l­5), zirconium{6), are entangled to the ertract ion­reextract ion cycle in the extraction scrub stages by the action of the other components which change th<u.­ dis t r ibu t ion r a t ios , and resu l t in accumulation in the process or deter iora t ion o­f D?. ^hus a process design of PUHEX inevitably needs an assessment of sucr. ­>>. nomena for c r i t i c a l i t y eafety and high DP. One of *he novel methods to analy;. the recycle phenomena is a simulation study of the process by computer code. A mathematical model for HJREX process EKTRA.M has been developed to analyse transient benavior of uranium, plutonium and ni t r ic acid in the mixer­settlers.

In order to clarify parameters which control accumulation of Pu(IV) or U(I7) in the JO vdl-ft TBP­n^odecane­BHCy^I(VI) system, a parameter survey was carried out using mathematical models DIST and EXTRA.M. Then the computed results of u"(IV) buildup were compared with the results of laboratory experiments by miniature mixer set t lers being put special emphaslB on the modification of the EXTRA.M to improve simulation.

• RESULTS AHD DISCUSSION Effects of Parameters on Recycling of U(IV^ or Pu(IV)

Specifically, plutonium builds up in a co­decontamination process of PUREX due to high loading of U(VI) of the organic solvent and subsequent s p l i t of Pu(IV) to the aqueous phase. Thus the negative salting­out effect of U{VI) on distribution of Pu(IV) plays an essential role for recycle of Pu(IV).

Our distribution model DIST(7) expresses distribution ratios of metal M as

b„ =P1G/ 2/i1 + P3C/4 + F5c/ 6 Cy 6 + 5 ^ } (1)

where CJJ and C g denote the aqueous concentrations of to ta l nitrate ions and uranyl ion, respectively. The PI , P2, '" are parameters of which values are determined by the best f i t procedures in the DIST based on the experimental data, and X f signifies

; the degree of (self­)saltlng out effect of the other components than U(VI). Buildup of Pu(IV) or U(IV) w i l l occur provided tha t a highest D value and D value

! at the highest organic concentration of U(VI)(Cn;g) of the extract ion process are respectively higher than and lower than 1. Magnitude of D depends mainly on parameter PI, and a dependency of D value on C^g i s re la ted to parameters P5 and P6 of ~uhe

I eq.(1} expressing the salting out effect of U{VI). As an example of DIST calculation, i Pig.1 a),b) show dependencies of Dj4 and Dp^ on Cgg as a function of P5 value for 30

179

: a) LJ,­805/1 "

­ | . • •

: \*\ ! '

.

VoJ 1 г 3 4 5 6

t a Pi 0 0.1 o.s 1.6b Irele 3.3 10

„ «, и ) \ 6 \ 1

• , . . . , >

« Cu ( ­BOg«

, 1

\\Г 3 = 0.5 \ \ 4 • VDZIrefem* Volue) \ \

П f Cu6 Ig / t l Cub lg/Л

PJR.1 . Depender.cies of Ьцд and Ьр^л on C g BS a function of P5 value in the extraction system of 30 vol.* TBP­n.paraffin­3M HND3­ a: U(IV), b: Pu(IV)

vol.#TBP­n.paraffin~3M ШО3 system. The reference values adopted in the EJETRA.K are listed in Table. Prom curves 4 of Fig.1 a),b), i t is understood that the highest Cyg necessary for D(IV) or Pu(IV) Ъиг­ldup in the process is larger than 40 g/1 or 102 g/1. respectively. And in case of an assumed process of which highest O g value is 80 g/1, accumulation wi l l occur with P5 value larger than 0­5(curves 3 to 6) for U(IV) and 5(curves 6 and 7) for Pu(IV), where D i s lower than 1. Figure 2 shows G g prof i le of the assumed process and variat ion of Сщ prof i le with P5 value­ I t is clear ly shown; that only curves 3 to 6 have shape of accumulation mode.

Parameter P1 determines not only magnitude of D^ but also slope of a curve of С profi le appeared in front of the high Cgg zone­ Thus C p ^ prof i le (P1=22­9) is more sharpened in shape than tha t of Cy> (PI =4.06). For the accumulation of Pu(IV), al highest value of Cgg must reach near saturat ion of 30* TBP solvent. The Cyg profile] under that condition of the process is usually unstable, because the amount of U(VI) input by the feed flow does not balance to that flown out with the organic solvent and a high Cyg aone with а С­вд peak in the front moves between the scrub section and the organic i n l e t stage during operation. Since D g defines Cyg, par t icu la r ly at the or^nic outlet stage, i t determines the mass balance of U(VI) of the process and consequently width of the high Crjg zone.

Reference parameter values obtained by DIST and used for EXTRA.M computation

Reference parameter value PI P2 P3 P4 P5 P6

%4

22.9 2.90 5.38 1.85 1.02 2.28 4.06 2.68 2.14 2.12 1.65 2.35

Fifi.2. Cn/| and C g profiles of the assumed extraction process showing variation of ° U P r o f l l e w i ' t n ^ v ^ x i e

180

: 0) т 1

Ftcd C M . 150 g / 1 Сщ­А.О fl/1 с„­з.о м

V : ^ : I I ­

/ Cw

i 6/J / bjJj^v

! ­/ \ \ \"

! ­ /P \ \ / V

Comparison of Experimental Results and ЕХТНД.М Calculation for U(IV)­Buildup Counter­current extraction of U(H) and U(VT) waa carried outwith mini­mixer

sett lers. Reference flowsheet is shown in Fig­3* Buildup of TJ(IV) would not occur i.f Су6 of the feed is lower than 75 g/l» because Сц6 will not exceed 40 g/l where D^ is approx. 1 as seen from curve 4 in Fig.1 a). During the course of the extraction based on the reference flowsheet, the accumulation of U(IV) was observed in front of the ^Sh Сц£ zone, peaks of Сщ and C™ appeared at 8th and 9th stages, respectively. Time dependency of both Сцд and Сш profiles in the extractor agreed very well with those calculated during whole period of operation.

To examine r e l i a b i l i t y of parameters of Dyg used for EXTRA.M, the flowsheet conditions changed after concentrations of the components in the mixer set t lers having reached steady state so as to observe growth of the high Cgg zone with time, which is important in case of Pu(IV) buildup. Conditions changed are

i ) 3Q£ increase of feed flow, i i ) 3C# decrease of solvent flow,

i i i ) decrease of scrub acid from 3M to 0­5H. Figure 4 shows an example of the results

of experiments(points) and that of EXTRA.M (lines). After being decreased solvent flow ra te , the high C^g zone expanded toward solvent inlet stage with time as seen Fig.4 a) and U(IV) accumulated at the front top as seen in Fig­4 b). Measurement of temperature along stages, Fig.5 a), demonstrated that a maximum temperature rise occured at the stage where an exhaust U(Vl)­extraction took place and consequently the highest C^/, appeared (Fig.5 b)). Because calculated ehapes of both Сц and Сщ profiles agreed very well with the experiments, parameters of Dm were considered to be almosx sa t i s f ac to r i l y r e l i ab le . However, behavior of the expanding high Cyg zone did not always agree well with prescribed by EXTRA.M.

n­p»cifltr>e

ISO g/l

1.0 H

Fig.3' Reference flow sheet for experiments of U(IV) accumulation

•

* \ f tV

ъ) I 5 h /

У » V \ \ ^

•13 b

V °/ f /

/9h

/Th Ab

5h : о V

9 h : • Jth: * 13h • I5h: * ,

1 2 3 4 5 6 7

Sw^« iwnber

Hg.4* Change of Cjjg and C ^ profiles with time by a 30# decrease of solvent flow rate from the reference flowsheet, a: U(VI), b: tt(IV)

181

Aa being discueaed above, behavior of the high C jg zone ia determined by the саза balance of TJ(VI) of the process. Assuming that the feed composition and i t s flow rate are constant, the mass balance depends mainly on the 5gg value and over­flow rate of the organic product. Measurement of Cyg value at the 16th stage during the run showed some differences from those calcu­

lated and f luctuations to some extent;. Since the organic overflow rate varies with the operation factorp, i.e., flow rate 2f the feed solvent, interfacial level of th. s e t t l e r s etc. , a key parameter for the U(rv) buildup during very stable operation might be I^g, especial ly in the system of high Cjjg. Therefore, parameter values of Dg6 equation were reassessed by DIST solely with the experimental data оГ high &jg systems. The EXTRA.M computation with the revised parameter values improved the simu­

lation of the behaviors oi the high C g zone in the extraction. I t must be pointed out tha t growth of the th i rd phase of TJ(IV) during the ext rac­

tion run is important for the analysis of abnormal situation. Upon forming, the third phase, i t appeared i n i t i a l l y at the s e t t l e r where the highest Сц^ was observed, but splix to several stages with time and prediction of the fate of ТЦПГ)Д.е., concentra­

tion profile, a maximum concentration to be reached, and rate of growth of the peak, by the present model became impracticable, because the ЕГГОА.М includes no information about chemical charac te r i s t i c s of the th i rd phase of U(IV) and Pu(IV). One of the mostly needed subjects for the assessment of Pu buildup in the PUREX process from a view point of c r i t i c a l i t y safety ie to es tabl i sh a precise model to simulate the behavior of the third phase of tetravalent actinide elements.

Fig.5.Change of temperature profilf, during extraction showing the highest temperature appearing at t i e Btage of a highest Сщ. Numerals on the figure ; stage number

REPEHHJCES 1. Rozen A.M.,Zelvenskii МЛа.//Radiokhimiya. 18.572(1976). 2 . idem.//At.Energ.,46,333(1979). 3 . Poczynajlo A.,Zalew9ki J.//Nukleonika,24,1193(1979). 4. Ochsenfeld W. et al.//Proc. Symp. on Past Reactor Fuel Reprocessing. Dounreay(UX)

Hay 15­18, 1979. 77. 5. Schon J.,Bleyl E.J.,Ertel D.//Proc. Intern. Solvent Extraction Conf.(lSEC),

Mmchen, Sept. 11­16, 1936. 1­399­

6. Garraway J.//Proc. Symp. on Liquid­Liquid Extraction Science (Extract ion '84), Dounreay, Nov. 27­29, 1984, I.Chem­E.?ymp.beries Но.Вв, p.47 (1984).

1. Tachimori S. Intern. Conf. on Nuclear and Radiochemistry. Beijin(R.C), Sept.1­5, 1986. Abstract p­72.

182

10­18 PREDICTING ТЯБ BEHAVIOUR OF NEPTUNIUM DORINH HDCLEMI FUEL REPROCESS D*G v. A . Drake* Harwell Laboratory, United Kingdom Atomic Energy Author i ty , Didcot, UK

In t roduc t ion , Яр­237 i s formed from r eac t i ons of neutrons with uranium during I r r a d i a t i o n of nuclear fuel and as the daughter product , by a­decay, of Am­241. During POREX r ep rocess ing , neptunium w i l l be d i s t r i b u t e d throughout the Haste and product streams of t he proces s . For t he development of waste management s t r a t e g i e s , i t has become e s s e n t i a l t o be ab le t o p r e d i c t the p a t t e r n of neptunium d i s t r i b u t i o n in a given reprocess ing flowsheet so t h a t poss ib l e r e ­ r o u t i n g can be considered.

HeptunluE. Chemistry and Behaviour. In n i t r i c acid s o l u t i o n s , neptunium can occur s imul taneously in 3 s t a b l e oxida t ion s t a t e s : Np(IV), Hp(V) and Np(VI). These i n t e r ­

conve r t i b l e oxida t ion s t a t e s , e x h i b i t d i f f e r e n t e x t r a c t i o n behaviour о the re >cessing solvent (TBP­diluent) r e s u l t i n g in t he d i spe r s ion of neptunium through­

out the proces s . The rou t e taken by neptunium in reprocess ing i s t he re fo re not simple and w i l l depend upon t he p r e c i s e condi t ions of t he process . Four key aspec t s of t he solvent ex t r ac t i on process need t r be examined:

( i ) t he equi l ibr ium d i s t r i b u t i o n с neptunium oxida t ion s t a t e s r e s u l t i n g from d i s s o l u t i o n of t he fuel and chemical cond i t ion ing .

The d i s s o l u t i o n of spent nuclear fuel in n i t r i c acid and under re f lux cond i t i ons , provides an oxid i s ing medium t h a t w i l l r e s u l t in the higher oxida t ion s t a t e s , Np(V) and Np(VI) as the predominant neptunium s p e c i e s . Chemical condi t ioning with n i t r i t e , used s u b s t a n t i a l l y to conver t a l l plutonium t o Pu(IV), w i l l provide condi t ions for the reduct ion of Np(VX) t o Np(V). The r e l a t i v e propor t ions of Np(V) and Np(Vl) in the feed a re gene ra l ly dominated by the n i t r i c ­ n i t r o u s acid redox e q u i l i b r i a :

2Np02 ­* NOT + 3H+ ? = i 2Npoi + + ЙГГЭг + H 2 0 . (1) The equi l ibr ium cons tan t , v , i s given by:

[Np(VI>] / [HNOz] V* 1 1 (2)

[Np(V)] \[й+] 3[нОз] where К inc reases from 3.2 x 10 M ' " a in 1 M acid to 9.3 x to M in 4 н acid a t

( i i ) the d i s t r i b u t i o n of neptunium oxida t ion s t a t e s between the organic phase and

reprocess ing a c i d s . Under reprocess ing cond i t ions , Np{v) remains in the aqueous phase while Np(IV) and

Np(Vl) are ext rac ted i n to TBP­n a lkane . D i s t r i b u t i o n measurements between n i t r i c (2)

acid and d i f f e r i n g compositions of TBP­diluent have been repor ted : Figure 1 i l l u s t r a t e s t y p i c a l da ta p o i n t s .

In the UK, t he use of su lphur ic acid to s t r i p Pu(IV) from the solvent has been s tudied for f a s t r eac to r fuel (PFR) reprocess ing . The presence of sulphate anions wil l a l so с Figure 2j

f i i i ) the oxidat ion of Np(V) by n i t rous acid in the ex t r ac t i on stages of the process .

ш

Rg^U Distribution of Np(IV), Np (V) and Np (VI) " ~ ~ " " batwaan nitric acid and TBP/n-alkana

solutions of varying composition (Т = 20-Э0*С)

Flu, a . Distribution of Np (IV) and Np (VI) batman 20% and 30% TBP/rv-afkarw and mixed nttric-siriphuric add solutions (T = 21 °C)

During the ex t r ac t i on s tages of reprocess ing , Np(V) wil l tend t o be c o n t i n u a l l y

oxidised t o Np(VI) and ext rac ted as i t moves towards t he aqueous r a f f i n a t e end of •

contac tor in an attempt t o r e - e s t a b l i s h equi l ibr ium (equation 1 ) .

The r a t e ->f Np(V) oxida t ion has been measured by Moulin (5 ) . fie found:

.8 x 1 O " 3 [ H + ] 1 , 3 [ N O T ] 2 I N P { V ) ] I 'M

M.min (3) dt 1 [тюг]

a t 25°C, с = 62.79 KJmol" 1 .

Since the exten t of oxidat ion i s governed by the aqueous phase res idence time in

the e x t r a c t c o n t r a c t o r , the propor t ion of Np(VI) derived by oxidat ion of Np(V) can be

deduced }

(iv) the e f fec t on neptunium of tf-Pu p a r t i t i o n i n g r e a g e n t s .

Reagents t o reduce Pu(IV) t o i n e x t r a c t a b l e Pu(I I I ) a re t e t r a v a l e n t uranium ions

(usual ly s t a b i l i s e d by hydrazine) hydroxylamine and fe r rous i ons . Np(vi) i s r ap id ly

reduced t o Np(V) by a l l t h ree r eagen t s . Further reduct ion t o Np(IV) by U(iv) and

fer rous ions i s slower and by hydroxylamine i s n e g l i g i b l e . In reduct ive p a r t i t i o n i n g ,

the reduced plutonium may a l so p a r t i c i p a t e in the reduct ion of neptunium.

Comparison of Predic ted Neptunium Behaviour with Plan t Measurement. Measurements of

neptunium in f a s t r eac to r fuel reprocess ing have been repor ted in the UK for fuel

from the Dounreay PFR r eac to r , and in Prance for fuel from the pmatix r e a c t o r

At Dounrc-ay the fuel i s reprocessed using a i x e i - s e t t l e r c o n t a c t o r s . The flowsheet .("

184

uses sulphate and small concentrations of ferrous ions in most of the stripping steps. The equilibrium ratio of Hp(VI) and Hp(V) in the feed can be predicted from equation (2J , shown as a plot in Figure 3. Figure 4 shows the oxidation of Np(V) to NplVI) for the first extraction cycle, plotted using equation (3)» Essentially complete oxidation and therefore lew loss of neptunium to the cycle I aqueous raffinate would be expected in the long residence times of mixer­settlers. The use of ferrous ions in the stripping stages ensures complete reduction to Np(IV), while the presence of sulphate renders it virtually inextractable (see Figure 2 ) . In the second and third cycles, Np(IV) is the prodominar.t

HHO,mot»nty

L 3 . Pwcttrtag» Np (VI) at equilibrium In conOmomd 1—6 QH*] ­ 3­D­» 5M, [HO" J ­ 4 ^ ­ 5 ­ 4 * 1

Fig. 4 . Oxktetion of Np (V) to Np (VI) In U рпнмм ot « c m пНтаик acid

neptunium species?, however, the particular nitric­sulphuric acid conditions used result in partial extraction in both cycles. A summary of predicted neptunium behaviour and actual plant measurements is presented in Table 1 and shows reasonable anreaeent.

At Marcoule fuel has been reprocessed in a pulsed column pilot plant which uses a mixture of hydroxylamine, hydrazine and UtIVl in the u­Pu partitioning step* ftgain. Figure 3 can be used to estimate the Np(VI) at equilibrium in the feeds: using a 'best guess' estimate of 10~ 3 M BNO2, then ^ 20% of the r.otal neptunium in the feed will be as extractable. Np(VI) and "*< 80% as inextractanle Np(VJ . However, because of the short residence times in pulsed columns oxidation of Np(V) to Mp{Vl) in the extraction stages will be incomplete. For a cycle I residence time of 5.5 mins, Figure 4 shows that ^ 67% of the Np(V) will be oxidised to Np(VI) and for a cycle II residence time of 2 mins/ 1 эп1у * 25% of NplV) to this cycle will be oxidised. The predicted and actual neptunium distribution is summarised in Table. These are in good agreement.

t estimated from reference (7)

185

ficperiaaotal and predicted Hp distribution for the procesa

2 , ;

np=% ficperiaaotal and predicted Hp distribution for the procesa plant itriii Dounreay Plant Karcoule Plant

Measured Predicted Measured Predicted

Peed 100 100 100 100 Cycle I aqueous raffinate 10.3 •v­0 10 10 Cycla I aolvant ra£flnat« <5.1 2.0 ­ ­Cycle II aqueous rafflnata 51.2 57.0 18­58 54 Cycla II aolvant raffinate <4.2 l.O ­ ­Cycle III (Pu) purification: ­ ­ 32­72 36

aqueous raffinate 12.5 22.0 ­ ­

Pu product 20.5 18.0 ­ ­

Cycla III (0) purification ­ ­ O.l ­

Conclusions. On the basis of the analysis described, it is possible to predict the route followed by neptunium in a specific process and also to explore the likely effects of changes in process variables.

References Gourisse D Oxidation of Np(V) by aqueous solutions of nitric acid in the presence of nitroua acldf/J. Inorg. Nucl. Chen. 33, Б31. 1971. Petrich G., Kolarik Z. The 1981 Purex distribution data index. KfK­3o8o, Karlsruhe, 1981. Gouxisse D., Bathellier A., Blanchard J. C. Use of sulphate in solvent extraction separation processes // Proc. Int. Solvent Extraction Conf. 1971. Vol. 1, Society of Chemical Industry, London, 1971* 14. Moore V. A., Scarglll D. The extraction of Np{IV) and Np(VI) into 20% TBP/OK from nitric acid and nitric­sulphuric acid solutions containing uranium, AERE­R 11463. Harwell, England, 1985. Moulin J. P­ Oxidation­reduction kinetics of Np in nitric acid solution. CEA­T­4912. I lay­aux­Roses, France, 1978. Heptunrum in fast reactor fuel reprocessing. ND­R­1262(D) . Dounreay, UK, 1986. Charvillat H. H. A fiv 4 years experio.ice of ^ l s e columns extraction cycles for the reprocessing of fast breeder reactor fuels at the Marcoule pilot plant^ Extraction '84, Proc. Symp. Liquid­Liquid Extraction Science, Dounreay, 1984, Inst. Chem. Eng. London, 1984, Series 88, 1­17.

186

10­19 TECHHETIUM BEHAVIOUR IH POREX PROCESS G. A. Akopov, A. P. Krinitsyn, A. P. Tsarenko, Khlopin Radium Institute, Leningrad, USSR The degree of Tc recovery into uranium and plutonium со extract in

PUBEZ process is by several times greater than that determined by its static distribution coefficients in the system tributyl phosphate (IBP) ­ uranium ­ nitric acid [l, 2] . The results of recent investiga­tions have shoim that tUe Tc distribution coefficient depends not only on uranium and plutonium concentrations but also on the concentration of zirconium [3, 4] • Tc and Zr contraction mechanism is rather comp­lex and inperfectly understood.

In this connection we have studied the influence of Zr, nitric acid and U on the extraction of Tc with 30 56 TBP in dodecane at (20+1) °C.

The dependence of Tc distribution coefficient on Zr concentration in the organic phase at constant concentration of nitric acid in the aqueous phase (3.1 U) in logarithmic coordinates represents a straight line with the slope equal to 1, which points to the possibility of a mixed solvate formation with components ratio equal to 1.

Nitric acid has a strong effect on extraction both of Zr and Tc. Logarithmic dependence of the Zr distribution coefficient on the acti­vity of nitric acid represents a straight line. For a given TBP con­centration its value can be calculated from the formula:

°^Zr " 1 \Ш0

^1 1

'7

У В

1 , 7

Л1

) where K^ ­ effective extraction cons­tant of Zr| у ­ stoichiometric activity

coefficient of nitric acid [5]. In tnis case ';he variation of Zr activity coefficient can be neg­

lected since according to the results of [6] its value rather weakly depends on the ionic strength (yu>1).

Tc extraction in the presence of Zr is described by a more complex dependence (Pig. 1, curve 1). At nitric acid concentration within the range 0.5 to 4.0 li the Tc distribution coefficient decreases in accor­dance with hydxate­solvate extraction mechanism. Host of investigators believe that in the absence of Zr, Tc is extracted in the form of tri­soivate or tetrasolvate [2, 7, 8].

Prom the results shown in Table 1 it follows that the effective ex­traction constant of "c in the form of tetrasolvate does not vary in a wide range of nitric acid concentrations. This confirms the supposi­tion that Tc is extracted as tetraeolvate. The Tc distribution coeffi­cient ir this case is determined from the formula:

(1 + К3[НН03] 2; "f nitric acid [$]; S Q ­ 1.08 U ­ initial

where K, ­ concentration extraction constant

ЛВР concentration. 187

[WNO,J

Fig. 1. Dependence of Tc distribution coefficient on nitric acid concentra­tion. Initial Zr concentration 1.6 g/1. Ourvt ­ calculation; points ­ experi­ment

Рог nitric acid concent­ration above 2.0 H the Tc distribution coefficient sharply encreases which seems to be explained by co­extraction with Zr.

Logarithmic dependence of Tc (i stribution coefficient on TBP concentration in orga­nic phase is nonlinear, which is likely to be connected with the variation of the solvent activity coefficient and with the formation of solvates having different Tc to IBP ratios. However, at TBP concentration about 30 % the slope is close to 2.

Table 1. Tc extraction in the absence of Zr

Hitric acid concent­ration,

M

Tc distri­bution co­efficient,

Concent­ration extrac­tion constant

(Kg)

Effective extraction constant

0.50 1.29 0.97 1.82 0.83 1.07 1.05 1.97 1.61 0.54 1.12 1.75 2.37 0.35 1.91 2.23 4.89 0.025 5.13 2.07

Average 1.96+0.17

Based on the re­sults obtained the following mechanism of Tc and Zr coex­traction can be pro­posed ­

2+ ZrO + HO. + 2TfflJo­

3 T

TEFT к;

zrotJio,;Tco. 2(ЯИ63­тар) (2)

Taking into account that Zr extraction is described by the

formula (1), we derive the expression for Tc distribution coefficient:

3L . M" y=Cs„ ­ sY 1 + К^ШОзГ^.Уз' 7

(3) where K. ­ effective con­stant of Tc and Zr coex­traotion; [z^^Jl ­ zir­

conium initial concentration, II; S ­ concentration of free TBP, II. The formula (3) describes well Tc extraction at high concentrations

of nitric acid. It is evident that in a wide interval of nitric acid concentrations the Tc distribution coefficient can be calculated tak­ing into account combined effect of the hydrate­solvate mechanism and

188

the coextraction mechanism, i. e. o £ T c = ci% + oi2 (4). It ia вееп from Pig. 1 that the calculated curve agrees well with experimental points.

Uranium, as would be expected, displaces Zr from the organic phase, and Tc distribution coefficient decreases to the value deter­mined by the mechanism of its coextraetion with tl (Table Z).

It should be no­ Table 2. Influence of U on the extraction of Tc and Zr, (Zr^ НПО, » 4.6 Ю

0.8 g/1;

U con­centra­tion in organic phase, g/1

Zr con­centra­tion in organic phase, g/1

Tc di­stribu­tion co­efficient

Zr di­stribu­tion co­efficient

0 10 20 3S 76 108

0.33 0.32 0,25 0.20 0.10 0.06

0.75 0.73 0.69 0.67 0.52 0.35

0.70 0.70 0.45 0.33 0.13 0.07

108 0 0.36 ­

ted that the loga­rithmic dependence of Zr distribution coefficient on free TBF concentration represents a straight line with the slope equal to 2. This fact confirms the supposition of Tc and Zr coextraction in the form of di­aolvate.

The studies per­formed allowed to calculate the degree of Tc recovery in PURE" process at U and Pu coextraction and its distribution on the ь\.с •< T jrrtracSion battery for steady­stat­г conditions. The calcu­

was iuade using formulae derived in [10], assuming that the '" Tc distribution coefficient in the extraction section is de­

i -. by simultaneous action of hydrate­solvate mechanism and that oi •; ad Zr coextraction formula 4), whereas in the scribbing sec­'on t is determined by x _har: j; of coextraction with U, studied in

detail by P. Macasek [г]. Ini­.al data for calculation ­ the concent­rations and flow ratios ­ were taken from [l]. The Zr concentration was evaluated based on the average burn­up of the fuel reprocessed (~30 kg/t U).

Pig. 2 shows the distribution of Tc through the stages of battery. The agreement between calculated curves and experimental points con­firms the correctness of the proposed model. The degree of Tc reco­very in coextract is about SO % which also agrees with experimental data [l].

Cur results show t'oat it is possible to control the degree of Tc recovery. In order to increase To recovery it is necessary to keep

189

the acid concentration in the extraction aection within the range 4­6 X, and to renove most part of Tc in raffinate the Zr content

should be tilnlmlzed using,

«л

xn ГГТ зр^ gig. 2. To distribution in aqueous (1) abd in organic (2) phases of eoextrac­tion battery. Curves ­ calculation; points ­ experiment

for example, a conplexing agent for Zr.

The experiments have shown that for equivalent content of Zr and fluoride­ions (­ 10~ 2 M) the Tc di­stribution coefficient de­creases almost by a factor of 10. Such a suppression of co­extraction effect reduces, according to calculations, | the degree of Tc recovery in J FUSES, process to 30 %. j

References 1. Aiopov G. A., Erinitsyn A. P., Isarenko A. P.//Iesledovaniya v

oblaoty pererabotki obluchennogo topliva i obezvrezhivaniya ra­dioaktivnykh otkhodov. Haterialy Pyatogo Simpoziuma SEV. Praga, ChKAE, 19B1. T. 3. 0. 264.

2. Hacasek ?., Eadrobova J.//J. fiadioanal. Chem. 1979. Vol. $1, S 1. P. 97.

3. Garraway J., Wilson P. D.//J. Less­Common Metals. 1965. Vol. 106, S 1. P. 183.

4. Jassim T. H., Person G., Libjenzin J. 0.//Solv. Extr. Ion Exch. 1984. Vol. 2, If 7/8. P. 1079.

5. Dawis W., Be Bruin H. J.//J. Inorg. Nucl. Chem. 1964. Vol. 26, Я 6. P. 1069.

6. Korovin S. S., Lebedeva E. H., Rosen A. M.//ZhFXh. 1966. T. 40, H 7. S. 1508.

7. Pruett D.//Radiochim. Acta. 19B1. Vol. 29, Я 1. P. 107. 8. Lieser K. H., Singh R. H.//Radiochim. Acta. 1983. Vol. 32, H 2. ,

P. 203. 9. Hikolotova Z. I., Kartashova H. A. ­ Ekstraktiya neitral'nymi

organicheskiai soedineniyani. Spravochnik po eketraktsii. T. 1. X.: Atomizdat, 1976.

10. Alders L. Zhldkostnaya ekstraktaiya. H.: IL, 1962.

190 Г:

THE SEPARATION OP URANIUM PROM A SOLUTION OP FISSION PRODUCTS

10­20

O.Hladik, V.Boeasert, G.Bernhard and T.Grahnert, Central Institute of Nuclear Research, Rossendorf, GDR

A procedure is described for the solvent extraction of enri­ched uranium (35 % U) from the waste solution of fisaion mo­lybdenum production. The large scale production of molybdenum­99 is characterized by the use of ordinary fuel elements of a rese­arch reactor as starting material /1/. After the separation of fiseion molybdenum the fuel element solution is deposited in a storage for U years. Owing to the special composition of the feed solution, presented in the Table, and the technological conditions the well­known PUREX­proceas had to be modified. The activity of the solution is about 0.3 Ci/L.

Composition of the feed aolution

The schematic flowsheet of this separation process is shown in Pig. The main steps are: storage of fuel element solution in spe­cial tanks, feed adjustment, solvent extraction of ura­nium with tri­n­butylphos­phate (TBP) diluted in tetrachloroethen and scrubbing of the extract, backextraction of uranium,

aoiution and waste treat­ment. The extraction ргосевв takes place in counter­current extraction. The organic solvent ib the heavy phase. The extraction and the backextraction are carried out in a new mixer­settlcr­unit. Plow conditions are regulated in such a way that uranium is extracted almost quantitatively, leaving most of the fission products and other impurities in the aqueous phase. The recovered uranium is well suited for fabrioation of new target elements using in fission molybdenum production. The facility was put into operation in 1985.

Substance Concentration ( g / i :

»1(N0 3) 3 163.3 H 0 2 ( N 0 3 ) 2 4.3 Hg(N0 3) 2 0.15 Cd(.10 3) 2 8 k HN0 3 7.5 Si/Si0 2 0.5 PlBBion products 0.015 Pu 0.0005

191

Schematic flowsheet: inlet solvent; 1-2 - inlet scrub solution; 1.3 - inlet backextraction solution; l.i - inlet reducing agent; 2.1 - 2.6 - storage tanks; 3-1 - slab vessel for 'J0 2(N0 3) a org. ; 3-2 - slab vessel for U0p(N0_)„ aq.; h - centri-fuge; 5 - extractor; 6 - backextractor; 7 - storage for U0_(NQ)p aq.; 8 - evaporator

References 1. Utilize R., Hladik 0., Bernhard G., Boessert V.

Int.J.Appl.Radlat.Isot. _35_(1984).749. Schwarzbach R.//

192

SOLVENT EXTRACTION OF DIBUTYLPHOSPHATE BEARING ALKALINE WASTES FROH PUREX PROCESS 10­21

. .<*> *) (*) D. PAREAU , A. CHESNE ' , S. CHEVALIER 4

" , J . T . PASQUIOU4

L a b o r a t o i r e de Chimie N u c l e a i r e et I n d u s t r ­ i e l l e

Ecote CentraLe des A r t s et Manufac tu res

92295 CHATENAY­MALABRY, FRANCE

Commissar ia t a I " E n e r g i e Atomique BP n ° 6

92265 FONTENAY­AUX­ROSES, FRANCE

I n t h e Purex p rocess t r i b u t y L p h o s p h a t e undergoes chemica l and r a d i o l y t i c

a t t a c k Leading t o t h e f o r m a t i o n of a c i d i c d e g r a d a t i o n p r o d u c t s , main ly d i b u t y l ­

phosphate (DBP) and t o a l e s s e r extent monobuty lphosphate (MBP). These a l f c y l ­

p i o s p h o r i c compounds are e x t r a c t a n t s and may a l s o g i v e i n s o l u b l e complexes w i t h

s e v e r a l c a t i o n s of f i s s i o n p r o d u c t s . Thus t h e i r e l i m i n a t i o n f rom t h e o r g a n i c

phase by a l k a l i n e s c r u b b i n g o f the s o l v e n t i s n e c e s s a r y . The a l k a l i n e s o l u t i o n

i s g e n e r a l l y made of c a r b o n a t e , i n o r d e r t o keep uran ium and P l u t o n i u m , which

can be p r e s e n t i n sma l l q u a n t i t i e s , under a s o l u b l e f o r m . The d e s t i n a t i o n of

t h i s aqueous s o l u t i o n i s u s u a l l y t h e h i g h or t h e medium a c t i v i t y was tes . Recy­

c l i n g a c t i n i d e v a l u e s f rom t hese e f f l u e n t s o l u t i o n s i n t h e process i s t o be

cons ide red t o lower ^­activity i n t he wastes ( f i g u r e 1 ) . H o r w i t z and coworkers

[ l ] d e s c r i b e d a s o l v e n t e x t r a c t i o n process <Aralex> t o recove r a c t i n i d e s f rom

t nese s o l u t i o n s a f t e r t h e i r a c i d i f i c a t i o n and a subsequent phosphorous compounds

e x t r a c t i o n by 2 e t h y l 1 h e x a n o l .

Carb.sol.from Purex to process

­*и­to ш OK нш

Pig, if. DBP clean up cycle. 1 - pH adjustment; 2 ­ DBP extraction; 3 ­ amine solvent treatment

Another way t o hand le t h i s prob lem c o n s i s t s of s e l e c t i v e l y e x t r a c t i n g t h e

d i s t u r b i n g organophosphorous compounds f rom n e u t r a l or a l k a l i n e s o l u t i o n s where

t h e y a r e under an a n i o n i c f o r m . The a c t i m ' d e s w i l l s t a y i n t h e s e aqueous s o l u t i o n s

which can be c o n c e n t r a t e d and r e c y c l e d w i t h o u t any r i s k o f p r e c i p i t a t i o n . Amine

t ype e x t r a c t a n t s a r e known t o be p o w e r f u l an ion exchangers and t h e y have c o n s e ­

q u e n t l y been s t u d i e d .

13. 3 I K . 382 193

Tertiary «nines usually display a better chemical and radiolytic stabi l i ty compared to other types of amines. So we f i rst choose the tri­iso­octyLa»ine ; it is diluted in a mixture of dodecane and octanol to prevent third phase formation during the neutralization by an acid.

the extraetant is used under the form of tri­isc­oct>4a*Mioniu» n i t ra te , which is obtained by the reaction.

T\OA •*• H* + Nu"5*rT.0AH*N0j • И )

The extraction of dibutylphosphate by the alkylammoniun nitrate is supposed to obey the following anion exchange reaction :

А0 )

г р

(DBP~) The formation of the tri­iso­oetyLamraoniurfl dibutylphosphate obviously depends on : ­ the pH of the aqueous solution; ­ the aqueous concentration of the nitrate ions* ­ the dibutylphosphate anion concentration* All the experiments were carried out. at room temperature.

Influence of the aqueous phase pH

Br,

Pig*2* DBF partition coefficient versus pH

, In these experiments trioctylamine nitrate (0,43 mol I ) in a mixture of dodecane <82 X) and octanol (18 X by volume) is con­

tacted with various aqueous solutions. pH is measured at equilibrium. The results are shown on figure 2. A very sharp decrease of the OBP partition coefficient CD) around the neutrality is noted. This is consistent with equations (1 and 2) the concentration of the "anion exchanger" sharply decreasing. Influence of the nitrate concentration , We used the same solvent than above. No measu­

rable influence on DBP extraction has been observed when increasing the NO, concentra­

tion from 0 to 0.4 aol*I at pH 3,7. Com­

bining equations (1) and (2) Leads to the following relationships О = ((..К­П^ОЛнЧ where К, and К. are the equilibrium cons­

tants of the two reactions. The experiment tat results are consistent with this tela* tionship.

194

Influence of DBP concentration . The aaueous phase is a muture of hydrazine carbonate and hydrazinium nitrate in which the DBP concentration varies fro»

The aqueous solutions are contacted with the tri­iso­oetyla­nnonium nitrate in dodecane­oetanol ; the pH at equilibrium is always 7. No variation of ^ D

has been noted ; i ts value is 2 ­ 0.2. Optimization of the organic solution: . The preliminary results show that amines may be efficient extractants for OBP in neutral or slightly alkaline solutions. An optimization of the organic system was however undertaken, namely by study­

ing the influence of the nature of the amine, i ts concentration and the nature of the diluent, the purpose beeing to find a possibility of cLeaning up aLka­

line solutions from OBP at higher pH. For sake of comparison several alkylamnonium nitrate have been tested in

the following conditions : extractant concentration 0.1 mol­L in dodecane­

octanol, pH of the carbonate aqueous phase 8 . 1 , DBP concentration 5.10 mol­

I . Extraction is performed with a volume ratio of 1 . The DBP partition coef­

ficient is equal to : 0.41 for trilaurylammonium ni t ra te , 4 for Aliquat 336 (commercial trilaurylmethylammonium) ni t rate , 17.5 for Primene JMT <1B to 22 carbon atoms primary amine) nitrate. The primary amine nitrate looks very favourable compared to others. So we focused our studies on this extractant.

The natufe of diluent plays an important role in the extraction by ion pair formation. For simitar conditions (pH 8.2 , DBP 5.10 mol l , extrac­

tant 0.1 mol*I ) diluents of various polarity were compared :

Table 1 . Influence of diluent on №P partition coefficient

Diluent Xylene Chloroforme n­Dodecane HTP HTP­octanol Mixtures X­octanol

D

i * p 0,43 0,37 0,W 0 , « Ч5

0.8513.9

10 7.2

го 1 5 . 1

A diluent of relatively high polarity is needed to give a good extraction and to prevent third phase formation.

The variations of Б DBP versus the extractunt concentration were determined

at pH 8. ^ 0 В р is roughly l inear, the slope depending on the diluent. This is comistent with the supposed 1 : 1 stoechiometry. Counter current operation . A counter current test was worked out on aqueous solutions corresponding to the alkaline scrubbing of TBP Leaving a plutonium purification cycle. The general conditions were the following :

•^u»ous_phase : hydrazinium carbonate 0.7 mo'.4

1 nitrate 0.96 mol* L

О.'­б mjl ' t •1

­ 1

hydrazine DBP"

the pH was adjusted to 8.1 by addition of nitric acid.

•I 195

organic_ghase : Prinene JMT (NO, form) 0.12 И dodecane ­ 2 octanol (1S X by volume)

The extraction was performed in four stages of a Laboratory i The H o w rate ratio was OVA = 1/5.

«er s e t t l e r bank .

2 i> 4

ofiqaru.c

OQU.Z.0U4 2 i> 4

~^ botuXLo

&otuXi.on

2 i> 4

The DBP d i s t r i b u t i o n between t h e tuo phases i s g i v e n on t h e f o l l o w i n g t a b l e ;

Table I I . DBP d i s t r i b u t i o n between phases i n each s t a g e

Stage 1 2 3 4

ОвР 1 0 ~3 n o l l ~

org 32 16 M 1,3

DBP 1 0 ~3 mol. Г '

aq г 1 0,54 0,13

D

D B P is 16 13 ,7 14,6

In t h e s e c o n d i t i o n s t h e measured d i s t r i b u t i o n c o e f f i c i e n t i s o f t h e o r d e r of 10

f o r Pu( IV ) and 0.4 f o r R u t h e n i u m . Pu can be comp le te l y removed f rom t h e s o l v e n t

by a one s tage s c r u b . Amine s o l v e n t can be reused a f t e r an a l k a l i n e s c r u b b i n g

t o e l i m i n a t e D8P (and Ru i f p r e s e n t ) . The r e s u l t i n g aqueous e f f l u e n t w i l l be

a c t i n i d e f r e e and be sent t o medium a c t i v i t y w a s t e .

Reference

1 . HORWirz E.P. et a l . / /

1 7 7t h ACS N a t i o n a l H e e t i n g ­ H o n o l u l u , 1 9 7 9 ,

A c t i n i d e separa t ions­ACS Synpos ium.Ser ies 117, p , 4 7 5 ­ 4 9 6 .

196

ABOUT THE FORMATION OP INTERPHASE SPECIES BASED ON SILICIC I I ACID IN THE EXTRACTION PROCESSES OP NUCLEAR FUEL REPROCESSING L!_I . V. I. Andreev, R. I. Lyubtsev, G. S. Markov, IS. U. Hoshkov, B. V. HikLpelov, Khlopin Radium Institute, Leningrad, USSE

Appearance of interphase films, stabilized emulsions and precipita­tes (often termed as "crud") in the extraction processes of nuclear fuel reprocessing саивев a number of difficulties: incomplete phase separation, worsening of hydrodynamic conditions in extractors etc. Occurrence of crud is commonly related to the formation of stabilized emulsions which accumulate during demulsification of emulsified two­phase system in the region of phase boundary and subsequently change to interphase gelatinous precipitates. It is known that silicic acid can be one main components of interphase precipitates £1­3]. Dissolved si­licic acid has the property of forming, especially in the presence of tri­n­butyl phosphate (TBP), stable interphase films significantly re­ducing the velocity of demulsification [4­6]. However, the mechanism of formation and the nature of interphase precipitates containing silicium have not yet been studied enough.

We have investigated the conditions of appearance and the nature of interphase precipitates formed in the extraction systems with partici­pation of dissolved silicic acid. Host detailed investigation has been made of conditions leading to the production of stable practically un­destroyable in time emulsions formed at emulsification of the phases as a result of interaction of TBP molecules with intermediate species of polysilicic acid insoluble in organic phase, which appear in aqueous solution of nitric acid in the process of polymerization of monomeric silicic acid or its low­polymerized forms. With the nitric acid con­centration of 1­3 H and silicic acid concentration of 1.0 g/1 such in­termediate species can exist in solution for a long time ­ over 40­50 days. As it was found, polysilicic acid particles with different

(degree of polymerization participate in the formation of stabilized /'emulsions. During initial period of aqueous silicic acid aging the sta­bilized emulsion contains mainly /3­form of silicic acid. As the time of aging increases the particles of ­form take ever growing part in pro­duction of the stabilized emulsion. At the same time the volume of the .­stabilized eaulaion decreases. Later, the transition of polysilicic ;acids to highly polymerized forms leads to a complete cessation of sta­bilized emulsion formation (Fig. 1).

197

Based on oux data the formation of particles entering into the com­position of stabilized emulsions may be written as:

HHO,, polymerization H.SiO. *

+ TBP, diluent (S) r(Si02­nH2O)•ySi03H2«yTBP­ZS,

where x/y = 30­1000, and Z ­ 2­6. According to IH­spectrometric

study in the above species a re­latively strong hydrogen bond ex­ists between the phosphoryl group of the extr&ctant and hydroxyl groups of polysilicio acid. It has been also found that the pre­sence in the aqueous phase toge­ther with silicic acid of hydro­lyzed forme of zirconium, ferrum and aluminium intensifies the process of stabilized emulsion production. The ability of aque­ous solutions of silicic acid to form stabilized emulsions especi­ally grows in the presence of zir­conium hydrolyzed forms and radia­tion­chemical degradation products of TBP ­ alkyl phoshoric acids and H,P0.. Formed in this case finely dispersed precipitates of zirco­nium allcyl phospthates together with polysilicic acid particles concentrate after emulsification at the phase boundary in the form of voluminous stabilized emulsion. It should be noted that the life

time of the polysilicic acid active forms responsible for the genera­tion of stabilized emulsions essentially grows in the presence of hyd­ .. rolyzed zirconium species, so that the formation of stabilized emulsion^­at emulsification of two­phase system is observed even in some months of aqueous solution aging. It is belived that this phenomenon is conn­ected with the formation in the aqueous solution of stable polymeric heteronuclear compounds of silicium with zirconium. The determination of silicium to zirconium molar ratio in samples of stabilized emulsion obtained from aqueous solutions with different time of aging shows that

в Time.br

Fig. 1. Kinetic curves of silicic acid aL-, /3­ and g'-torae variation in solutions before and after the formation of stabilized emulsions in the systems 30 % IBP + 70 %

3.0 M HS0 3 + 0.7 g/1 Curves: I, I' ­o^­form; 2,

2> -£-torm; 3, 3' ­J'­form; 4 ­time variation of stabilised emul­sion volume; 1, 2, 3 ­ before the formation of stabilized emulsion; 1', 2', 3' ­ after the formation of stabilized emulsion

°&6

the Initial Si/Zr value of 4-5 gradually increases to 10-15. This is likely to be explained by the increase of fraction cf more polymerized forma of silicic acid in the emulsion composition.

It is interesting to note that the additions of small amounts of hydrogen fluoride (0.01-0.1 g/1) to aqueous solution cause an intensive formation of the stabilized emulsion with a volume much greater than that of stabilized emulsion formed in the absence of hydrogen fluoride. But the period during which the initial solution of silicic acid pre­serves its ability for stabilized emulsion production sharply decreas­es. Additions of hydrogen fluoride in great quantities ( >1 g/1) can prevent t^om formation of stabilized emulsions (Pig. 2). Thus, the re­sults obtained indicate that the formation intensity of interphase stabilized emulsions based on silicic acid is primarily determined by the factors affecting the generation of active species of polysilicic acid responsible for appearence of stabilized emulsions. As it has been found in our work these factors include the total content of si­licic acid in aqueous solution, the conditions of preparation and the time of solution aging before extraction, the concentrations of nitric acid, uranyl nitrate and nitrates of other metals, temperature as well as the type of enulsion formed at mixing and the time of phase contact. In addition to these factors the intensity of stabilized emulsion for­mation depends on the nature of diluent and extraetan.. The character of stabilized emulsion volume variation depending on IBP content in extraction mixture essentially changes when bensene, chloroform or car­bon tetrachloride are used instead of n-pareffines as diluents. In the

Pig. 2. Effect of HgPp on the formation of stabilized emulsions in the system: 30 f> 4SX + 70 % 0 1 2 H 2 6 + 3.0 M HHO-j + 1.0 g/1 H^SiO^. Curves: 1 - without addition! 2 - 0.001 g/1; 3 - 0.01 g/1; 4-0.1 g/1; 5 -1.0 g/1; 6 - 2.0 g/1

199

сага of carbon tetrachloride the onset of stabilized emulsion 1'o.r.aation corresponc.3 to the TBP volume fraction of 60­70 % £"7j. The substitution of triisoaiayl phosphine oxide for TBP at the same experimental condi­tions increases markedly the volume of stabilized emulsion which seems to be connected with the increase hedrogen bond strength between the extractant molecules and the particles of polysilicic acid.

The effect of ^"­radiation with absorbed dot^s up to ™ 2 MGy on ex­traction mixtures TBP + diluent has small influence on the ability of extraction system to form stabilized emulsions. On exposure of an in­terphase stabilized emulsion to "­radiation the greatest emulsion drops are destroyed, and the stabilized emulsion gradually transforms into a more compact interphase precipitate consisting of polysilicic acid gel and organic phase microemulsion. As a whole, the examination of the results obtained allows us to conclude that the problems of reducing or avoiding crud formation are connected not only with the degree of purity of solutions fed to extraction. Along with the selec­tion of an extractant and a diluent much importance is attached to the development of optimum extraction onditions taking into account all factors affecting the formation of stabilized emulsions.

Reference 1. Hardewell F., Hatchen G. // Khimiya yadernogo goryuchego. M.:

GHTIChL, 1956. S. 54­9. 2. Solovkin A. S., Yagodin G. A.//Ttogi nauki i tekhniki. Seriya "Me­

organicheskaya khimiya". 1959. '£. 2. M.: VIHITI, 1970. S. 46. I 3. Zimmer E., Borehardt J.//Mucl. Technology. 1986. Vol. 75, It 3. .

P. 332­337. 4. Skrylev L. D., Sazonova V. F.//ZhPKh. 1976. T. 49, Vyp. 6.

S. 1419­1421. 5. Chizhevskaya 3. V., Sinegribovn 0. A., Danilova S. S., Koro­

vin Yu­ т. Yagodin G. A.//Izvestiya vuzov SSSH. Khimiya i khimi­che'' .;.­, ­. ?khnologiya. 1977. T. 20, Vyp. 5. S. 694­697.

6. Kud­ . S. G., Sinegribova 0. A.//Trudy MXhTI im D. I. Mendele­ev». •:•"••­.•. Vyp. 414. S. 49­54.

7. Hikol ^kii B. P., Hikipeliv B. V., Moshkov M. M. i dr.//Atomnaya energia. 19B3. T. 55, Vyp. 5. S. 315­316.

200

EXPERIMENTAL PUREX FLOWSHEET STUDIES OF SMALL SCALE PULSED | " COLUMNS IJ°~ H. Schon, H.-J. Bleyl, H. Schmiedi-.-r, Institut ftir Keisse Chr mie, Kernforschungszentrum Karlsruhe, D-7 50U Karlsruhe, FRG

Introduction MINKA, a small scale pulsed column facility has been installed at the "Institut fur Heisse Chemie" at the Nuclear Research Center Karlsruhe, in order to study the behavior of Pu, U and HNO3 at the Purex process under standard and off-standard conditions. Decreased HNO3 concentration in the scrub solution and increased feed flu., or decreased extractant f.JW were chosen as parameters to establish off-standard conditions.

The influence of decreasing HNO3 concentration in the scrub solution has been studied in f\,2j't this plus simultaneously changed hold up of the dispersed j^ueous phase on extraction behavior were reported in fjj• First results showing the influence of lowering the extractant flow on the extraction behavior of Pu and U were given in OJ.

This paper presents the transient extraction behavior of Pu and U in a strongly disturbed Purex-codecontamination cycle. Experimental. Three pulsed sieve plate columns were connected in series for the demonstration of a codecontamination cycle. The data of the columns and the conditions of the flowsheet are summarized in Table 1. A detailed description of the MINKA facility is given in OJ'. Results. The extraction behavior of U and Pu at the conditions of an off-standard flowsheet was studied in the extraction and scrub-column only. The backextraction column was used as service unit.

The maximum of the concentration of U and Pu appears in the aqueous phase. Therefore in the following figures the metal concentration of the aqueous phase is plotted only.

The concentration profiles of U and Pu for the two flowsheet are shown in Flg.l. The solid line represents the extraction behavior for the flowsheet given in Table, AX:AF=2.G6 '. The dashed line shows the data of the codecontamination cycle after additional reduction of the flow ratio AX:AF-2.52. In tne first experiment the flowsheet has reached the equilibrium after a period of 62 hours. The maximum of the accumulation of U was located at the middle of the S-column, whereas the maximum of the Pu accumulation was placed at the bottom of the S-coluirn. This value was measured to 21.3 g Pu/1 which is equal to В times of the Pu concentration in the feed solution. The Pu mass recycled with the raffinate stream of the S-column into the A-column amounts 350% of the Pu mass supplied with the feed.

The recycled amount of the metal does not influence the

201

D"

*Г я

_ <

\ \

Ftow­ulio а • ы ss

. i г м ­ о . » • 1 . г.5г ­о.«

'170*0/1 I иге и/1

го 40 О so «ю so ffl/l gU/l

U, Pu concentrat ion p r o f i l e s

an газ

extraction behavior of the A­column significantly. In the AW stream the loss of Pu was determined to 5 mg Pu/1 comparing with about 3 mg Pu/1 AW in earlier experiments with a flow ratio of AX:AF=2.60.

The decrease of the flowrate AX:AF­2.66 to the ratio 2.52 initiated a major change in the concentration profile of U and Pu (dashed line ir Fig.l). In this experiment the Pu loss in the AW оtream increased to 170 mg Pu/1 within a period of 9 hours. Therefore this experiment was discontinued. At the end of the experiment the U and Pu accumulation took p.1 асе in both the A and S­column. For both elements the maximum concentration in the S­

Floweheet of the Codecontamination Cycle

column A S С function extraction scrub strip conti phase organic organic aqueous diameter /cm/ 2.6 2.6 3.6 pulsation amplitude /cm/ 1.5 1.5 2.5 frequency /Hz/ 1.2 1.2 1.0 sieve plats (23% free area) space /cm/ 3 3 5 holes /cm/ 0.15 0.15 0.32 flowsheet stream relative flow composition

К HNO, 2.96

J g 0/1 g Pu/1

AF 1 К HNO, 2.96

J 256 2.31

SS 0.4 0.2 AX 2.66 0.4 30% TBP/Alkane

202

£ во г

ит AW

ы ПОШЛ

— о 170 9 39 6 6 0 4

Э mtMqfii A­eoMm AMR.

/ i Г/ Г: .'! ­

£i!Li.A­column, Pu­ break through vl

ftm AW (4 <»Л О 0.4S 9 <0.06 6 <0.06 0 <0.06

1 • HtKjftt « 4 m M­SBl

f '9­ 3i A­columfii u­break (htough column was measured at the bottom of the column. Still bigger accumulations of the metals appeared in the A­column. The Pu accumulation was established at the 1­1.5 m level of the A­column (37.1 g Pu/1 together with 135 g U/l). The concentration of TJ increased with the length of the column. Its maximum value of 226 g U/l was reached at a height of 3.5 m.

The transient behavior of Pu in the A­colunn is given in Fig. 2. Flg.3 shows the corresponding data for O. The maximum of the U accumulation was located at the top of the A­column at every time in the experiment. With increasing concentrations the U extraction eone was broadened downward.

In front of the U extraction zone the Pu accumulation took place. Its extraction behavior initiated the formation of the Pu peak.

The pu loss did not rise in the first 6 hours of the experiment, but later the losses increased rapidly.

The displacement of the U and Pu extraction front in the A­column can be clearly followed by the change of the temperature profile. The maximum of the temperature was found in the section of the column containing 80 ­ 100 g U/l in the agueous phase.

The evaluation of the experiment is still under way so that no statement can be made concerning the staxicuia Pu accumulation.

References 1. J. schon et.al.//ISEC.lS86, Hilnchen. 2. J. schon, H.­J. Bleyl, M. Kluth//Extraction.1987, Dounreay. 3. J. sch&n et.al.//Summer School of Extraction. Toulouse, 1987.

203

SMDLATIOB ОГ ОЮигаМ/ПДГГОНПМ S F L i r r i W . DT A POLSED COLOHH ] " i ш THE FOBEX PBOCESS i 1 ( > ~ 2 4 j F. Baroo and J. Duhaeet ­ П — i n i il a i'Energie Atoedque , IRDI­Угапсе

The reductive plutonium stripping operations carried out in the Purex process, such as uranium/plutoniura splitting, are sometirces difficult to ^et under control. When these operations are performed in pulsed columns, added to the complexity of the chemical reactions involved, are the specific features of this type of contactor, whose performance is governed by a set of equally complex factors.

The perfect mastery of this process, both in the design phase and in the operating phase, requires the availability of a sijnulation tool incor­

porating the various aspects of the operation. The French .Commissariat a l'Energie Atonique accordingly set up a research programme aiming to develop a uatheeatical model simulating the behaviour of the different species concerned in unsteady state conditions. This modelling was carried out in several steps :

• elaboration of a preliminary mod.­.i, simulating uranium/plutonium split­ting in mixers/settlers (already reported elsewhere <1>) ;

• elaboration of a model for simulating simple extraction operations in a pulsed column, accounting for the specific features of this contactor ;

• build­цр of the final complete model.

The Modelling of the uraninm/plutonium splitting operat jn in mixers/set­tlers required the following in succession :

• the inventory of the chemical species to be taken into account (nine were selected : U(IV), U{VI), Pu(III), Pu(IV), HNO,, HMO,,N 3H s*. Tc(ox), Tc(red));

• the compilation of data concerning the partition of each of these spe­cies between the aqueous nitric phase and the solvent (30% TBP, diluent) (interdependent partition, described by a semi­empirical formula for the extraction mechanisms) ;

• the inventory of the chemical redox reactions ('functional', 'inter­fering' or 'useful* reactions) involving the various components,

J

n both phases ;

• the formulation of the kinetic aspects relative to the interfacial transfer of the species and the main chemical redox reactions involved.

These details were obtained either from the compilation of published results, or, above all, from experimental determinations made in the CEA laboratories. They hence provide the basis for a model simulating a parti­tion operation with uranous nitrate in a compartmented extractor. This model was validated successfully by comparing the results of its operation :

• with results of specific experiments performed in a suitable alpha installation (laboratory mixers/settlers) ;

• with operating results of industrial units.

The BodalliBg of sueple extraction operations (not involving chemical redox reactions) in • polled colon required consideration of <

• Data relating to partition and to the transfer kinetics for the species concerned (U(VI), Pu(IV), H N 0 3 ) , previously obtained.

204

FJK­2 ­ M e a

" HETSC*) versus total specific w flow rate (uranima extraction)

f*)HETS : Height Equivalent to Theoretical Sfage

0 г.о 4.0 Specific flowrate (l­h~1

. • Hechanisma specific to liquid/liijuid extraction columns which are

reflected ort the whole by axial mixing effects in both phases. We thus devuloppeo. a multi­component dispersion model which, after the

determination of the axial dispersion coefficients, allows the calculation of simple extraction operations encountered in nuclear fuel reproces­sing.

The validation of the model required the use of extensive experimental facilities, ranging from laboratory pilot plants to prototype industrial installations. The conducted tests consisted in the simultaneous determi­nation of :

• concentration profiles in both phases, for different extraction opera­tions ;

• axial dispersion coefficients by injection of tracers into each phase.

The lover performance observed at low flow rate is well simulated by the model. Figure 1 shows the variation in the mean HETS as a function of total specific flow rate, calculated for the same set of parameter;!. The compa­rison of the experimental and calculated concentration profiles {Figure 2) is also satisfactory.

Fi%, 2 . Calculated and on) experimental UIV1) profiles (extracti

(*) TSF : Total Specific Flow rate (l.h­^.cm­O

O.OlL

0.0001 4 z (ш)

205

ТЬ+ ca^>letioa of the foregoing t*o developments served to undertake the construction of a model able to simulate the metabolism of the different chemical species involved in a reductive stripping operation on pluconium in a pulsed column. The use of this apparatus as compared with nixers/settlers leads to signifiant differences in the operation of uranium/plutonium split­ting. The competition between the various chemical reactions is intensified, both phases remaining in permanent contact. A number of relatively negligi­

* Ы е mechanisms in mixers/­

settlers Assume greater impor­

Calculated and experimental hi profiles (U/Pu splitting)

bO tance here­ This meant that the corresponding kinetic drta used in the first model had to be

40 a justed. The model thus obtained served

to reproduce the behaviour des­, 20 cribed above, and yields very

satisfactory results, as shown by the successful comparisons of the computer code with the refe­rence experiments performed in a suitable test facility (Figure 3).

Ifence we nov have an qualified tool for calculating the most complex ex­traction operations conducted in the Purex process. Its field of application extends to :

• The optimized definition of process flow sheets as well as equipment, • The simulation of incidental operations in order to predict possible

developments, the means for detecting them (optimization of control sys­tems), and the measures required to return to the normal state : for exam­ple, the model helps to carry out the safety analysis which may be required to guarantee the safe operation of the facility concerned.

Finally, it provides & solid basis of knowledge that could be exploited advantageously, both for training operators and for the elaboration of ef­ficient tools for aid in facility control (expert system).

Boullis B. and Загоп Р. Modelling of uranium/plutonium splitting in Purex process Л I. Chen».B.Conf.Extraction '87, Dounreay, June 1987.

206

10­25 AQUEOUS PHASE ENTRAIHKEHT ID PULSED PLATE COLUMNS

J.A. Jenkins, D.H. Logsdall, E. Lyall, i.E. Myers, B.A. Partridge and P. Ml•try

UK Atomic Energy Authority, Harwell Laboratory

Introduction • Operation of liquid­liquid extraction equipment can give rite to the entr aliment of very aaall drop» of one phase dispersed in the other phase, which adversely affects product purity. For example, In the reprocessing of spent nuclear fuel uranium (0) and plutonlum (Pu) are extracted from highly active aqueous feed* Entralnment of aqueous phase In the organic product could result in carry­over of 'non­extractable' fission products with the 17 and Pu, which would put an undesirable burden on downstream purification and waste treatment processes.

This paper describes experiments with a pulsed plate column to study;

• the mechanisms which produce aqueous phase entralnment;

• the size characteristics of the entrained drops;

• their effect on the carry­over of a 'non­extractable' impurity.

The system studied was uraayl nitrate (UH)/nitric acid ­ ZOX tributyl phosphate in odourless kerosene (TBP/OK) which is used to simulate Pure* type flowsheets. Approximately 2 g/Jt of non­radioactive Cs were added to the UN feed as a tracer for f

non­extractable' fission products*

Experlaent * The column waa operated on a compound extract/scrub process and had a 10 и high and 50 mm diameter section fitted with stainless steel nozzle plates, with approximately 6 n height for extraction, and 4 m height for scrubbing. A settler for phase disengagement at the top of the column наа approximately i m high and 100 ma diameter.

The size distribution of aqueous drops entrained in the organic phase *a» measured by a Malvern laser spectrometer. A sample of the organic phase from the column was placed In the path of a parallel beam of monochromatic laser light sod the diffraction pattern produced by the sample was measured by a diode plate and modelled automatically by the spectrometer using scattering theory fl].

Mechanisms which produce entrained aqueous drops. Experiments show thct two mechanisms arise in the Purex system;

1. Column pulsation ­ which produces a small volume fraction of fine drops at the bottom end of the drop rflze distribution;

7. Precipitation of water from the organic phase ­ which occurs whan И, or U and Pu, la extracted into the organic phase.

207

Figure 1 shows how the concentrations of dissolved and entrained water in the organic phase varied with distance along the colutm. The dissolved water concentration la constant along «out of the extract section but decreases sharply at the U feed point, where most of the U Is extracted. Consequently, the entrained water concentration, which was constant and <0.I vol X along nost of the extract section where the first mechanism predominates, increased to ­ 0-7 vol X at the U feed point because of the second mechanise. However, Host of the precipitated water la removed in the scrub section of the column and the organic product from the column contained <0.1 vol X entrained water.

U feed Scrub feed

Scrub

г и б в to Distance from organic feed I m )

Fig.1. Vorialion of water concentrations along extract / scrub column

Predicted from literature (2)

Figure 2 shows how the diasolved water concentration in the organic phaee varied with U concentration. The data show a good correlation, dissolved water concentration varying from < 0.1 vol 2 when the organic phaee was almost saturated with U to - 0.4 vol Z when no U was present. The experimental data agree well with values calculated from the literature^ assuming that the concentration of "free" TBP is equal to the total concentration of TBp less that bound to U as the dlsolvate (U0 2(H0 3) 2.2TBP).

U concentration [M ) Fig. 2. Variation of dissolved water

concentration with U concentration.

Site cUarectarlatlce of entrained drop»» Measurement! with the laser spectrometer indicate chat the precipitated drops are much smaller than those produced by column puliation* A* Figure 3 «hows, the entrained drop* near the D £«td point vary fro» approximately 2 to 10 um diameter with a mean of 3 ця* Т Ы » ia in contraat to the entrained drop* 1 a below the U feed point, where little U m e atill present in the aqueous phase and the mechanism of column pulsatien therefore predominated, whose diameter varied from 2 to 190 tt* with an average of SO u,a.

208

_QCQ==LXI1J 10 го юо 200

Drop diameter I (iml

Fig 3 Size distribution of drops tnlrainta in onjonic phcse near U feet!

Saaplee of the organic produce leaving the settler at the top of the column contained entrained drops with a broad size distribution but, as Figure h shows, there were two peaks in diameter of approximately 7 да and 130 цш.

Comparison of Figures 3 and 4 suggest that the drops entrained in the organic product have the characteristics of both mechanisms of drop formation and therefore comprise;

(a) a snail fraction of precipitated drops which have passed through the scrub section of the column without either coalescing with large drops of scrub acid or coalescing with each other sufficiently to attain a size (~ 200 ци) able to sediment down the с п Ь ш against the flow of organic phase;

(b) drops formed by column pulsation.

In both cases some coalescence of the entrained drops during passage through the column and top settler is evident because the peaks in che drop diameters in the organic product are a factor of 2 ­ 3 greater than they are near the U feed point i.e. 7 urn compared with 3 ил for precipitated drops and 130 из compared with 50 ив for mechanically formed drops.

^ггпПгТттт­Drop Лот*fir (дт)

^_Sil« distribution of tfrop* •nrramvtf ii *~ product from lop wltlor.

20»

Dependence of С» carry­over on aqueous Phase entralnment. Cs can be carried over In the organic prodact because It la dlaaolved in the organic phase and/or dlaaolved In the aqueous phase entrained in the organic phase. Samples of the organic phase from the scrub taction of the column were analysed therefore for their total and dissolved Ca concentration*, the difference between the two values being that due to entralnment of aqueous phase.

As figure 5 shows, the total concentration of Cs in the organic phase decreased frost 4 x 10~

3 g/A near' the U feed point to» on average, 2.3 x 10"

5 g/Jl in the organic product from the top of the column. The concentration of dissolved Cs in the samples was below the limit of

2 ( 6 Distance from U feed { m I

detection of 1.1 x Ю "5 g/l­ The

FjqJ.Cs concentration profile in scrub section.«rry­over of Cs in the organic • product due to aqueous phase

entralnment therefore appears to be, on average, between 1.2 and

2.3 x tO­ 5 g/Jt (the latter value if It la assumed that there is no Cs

dissolved in the organic phase). As Table shows, a sample of organic product which did not contain a detectable amount of total Cs had an entrained water concentration which was an order of magnitude less than the other samples.

Concentrations of entrained water and Cs in samples of organic producT

Sample №

Entrained water (vol Z)

Total Ca

(g/1)

1 2 3 4

0.002 0.02 0.07 0.04

<1.1 x 10" 5

3.1 x 10" 5

2.2 x 10" э

2.6 x 10" 5

Conclusions, There are two mechanisms which produce entrained aqueous drops in the extraction stages of the Purex process; drop shear during mixing and chemical precipitation of dissolved water during uranium extraction. Entrained drops formed by water precipitation are much smaller than those formed mechanically (typical mean diameters being 3 ua and 50 urn respectively).

In pulsed column tests most of the drops formed by chemical precipitation were removed by the scrub process. The concentration of Cs in the organic product after scrubbing was very low (*» 10" 5 of the concentration in the U fead) but appeared to depend on the concentration of entrained aqueous phase.

Experiments are In hand to study how the two mechanisms of entrained drop formation determine carry­over of non­extractable impurities and to develop methods by which aucralnmeat can be reduced.

References 1. Manual for Hodal 6000 Laser Spectrometer, Malvern Instruments,

Spring Lane, South Malvern, Worcs, DT. 2. Schulz, W.tf. and Navratll, J.D. (eds.) Science and Technology oi Tributyl

Phosphate, Vol. I, OtC Press, 1905

210

BRADSIM­PREDICTION OF SOLUTE CONCENTRATION. TEMPERATURE AND 1Q­26 PHYSICAL PROPERTY PROFILES ALOHC PULSED PLATE COLUMNS I ' D.n. Logsdall (UKAEA), S.F. Evana (BHFL, UK), J.A. Jenkins (UKAEA) and I.J.Smith <BHFL, UK)

Introduction , A dynamic model of pulsed plate coluam operation, BRADSIH, ha» been developed jointly by BRFL and the UKAEA, to enable designers and operators of pulse.' colurns for reprocessing Irradiated nuclear fuel to predict column performance during transient operation. The model cakes into account the change* in the hydraulic behaviour and mass transfer performance of a coluam fro» the beginning of an operational transient until steady­state Is reached.

The pu.sed plate column is a differential contactor in which mechanical energy la applied via liquid pulsing to enhance its extraction performance. It can be used for a wide range of processes, but probably Its best known use ha* beea la the nuclear tadttetry.

The pulsed column comprises a vertical, cylindrical column containing a series of equally spaced horizontal perforated plates arranged over its length. The liquids are fed countercurrently to the column and a pulsing action applied to the base of the column. If the light phase Is to be dispersed It Is forced through the plate perforations on the upstroke and the continuous phase flows downwards on the dowustroke.

This paper, which is the first of several to be published, describes the BRADSIH model of column operation and gives examples of simulations for a PUREX­type system for reprocessing reactor fuels. Predicted profiles of •olute concentration and liquid physical properties are presented and compared with experimental data from a 50 mm diameter column In the Harwell Solvent Extraction Pilot Plant ( 1 J and a 300 mm diameter column in the Sellafleld Pulsed Column Test R i g " ) . The extraction systems studied were uranyi nitrate (VH) ­ nitric acid/201 and 30* tributyl phosphate/odourless kerosene (ТВР/ОК).

Description of the model. The model simulates the dynamic response of a pulsed plat* column to perturbations in operating parameters and variableo. It takes account of the column hydraulic behaviour, multi­solute mass­transfer, non­linear equilibrium relationships for the solutes and volume changes resulting from solute transfer.

The model divides a column into a cascade of non­equilibrlun well­mixed elements, with additional end ^laments for the Introduction and withdrawal of the Appropriate feed, product and reffinate streams. (An element does not necessarily coincide with those formed by two adjacent plates). A system of dynamic mass balance equations for all the elements of the cascade Is developed In which backmixing la accounted for by the Inclusion of a backflow fraction of each phase between elements.

The equations take Into account the Interdependence of hydraulic behaviour and mass transfer performance. Рог example, they take Into account: the effect of changes In the phyaical properties of the solutions, due to solute transfer, on the volumetric flowrates of the phases and the dispersed Phase hold­up (DPNT), and the effect of changes in the concentration end flowrates of the feeds, and hence the DPHU, on solute concentration profiles.

The rates of solute transfer In each element чге described In terms of the film mass­transfer coefficient for both phases» the Interfaclsl area per unit volume In each element and the solute overall concentration driving force •cross the films.

211

The ability to predict temperature profiles Is Important with respect to monitoring column response to transients. Temperature has been Included In the model as en Integration variable and Is treated In a similar way to DPHU and solute concentrations so that temperature profiles can be calculated during transient behaviour.

An essential requirement for the design of pulsed columns is a knowledge of the physical properties of the solutions. These will vary signiftcantly along the column as the solutes are transferred from one phase to the other. Thus, equations have been included in the model for calculating profiles of phase densities, viscosities and the lnterfaclal tension.

Comparision of predictions with experimental data

(i) Harwell Solvent Extraction Pilot Plant ~ Experimental and predicted uranium concentrations la the solvent product stream under start­up conditions for uranium extraction using 20* TBP/OK art shown in Figure 1. These show good agreement between predicted and experimental values, and in both cases the time to steady­state from start­up is approximately 0.5 h.

Experimental and predicted uranium concentration profiles at steady­state in both solvent and aqueous streams are shown in Figures 2 and 3

Key + experimental — predicted

0­5 10 1­5 Time I h I

Fig. 1. U concentration in solvent product vs time.

103

i 7 0

I i с «I

I'0

"' ЕЮ'* Э

Ю"

10" limit of detection

T­>rV 0 2 i 6 8 10 Distance from top of column fmj

Fig.2. U concentration profiles ii " " "

—

~ " solvent phase

respect ively . The model correctly predicts a concentration of < К ГЦ g 0/Л. in

the aqueous raf f in i te 4 ш fro» the U feed point end a concentration of ­ 60 g Ч/Л in the solvent in the scrub section of the column. However, the rate of iiass transfer in the extract section appears to be s l i g h t l y under­predicted (because the predicted U concentrations in both phases occur on average 0.2 m further from the V feed point than they did experimentally) and the V concentration In the aqueous phase In the scrub sect ion i s rather over­predicted.

The profile» of densi ty , v i s c o s i t y and lnterfac la l tension in the solvent and aqueous phaaes at steady­atate are shown in Figure 4. Good agreement with the experimental data la evident , which confirms the accuracy of the correlat ions used in the model.

212

Key predicted experimental

Viscosity

Ю3

U leed

h Density Int tension

Key + experimental Ь _ _ ^ r — ^ ^

о

1 1

in­' С " — predicted

+\ \ Solvent \ feed

V l

h I

2

I A

С S

n ­2

e

IV a

С " — predicted

+\ \ Solvent \ feed

V l

h I

2

I fea 860 900 DensitytKg/m

3

! S­5 1 6 V7 1­е 1­9 2 0

Viscosity ImPa.s I

^ limit ot detec

— predicted

+\ \ Solvent \ feed

V l

ц­5

, ,

— predicted

+\ \ Solvent \ feed

V l

(a ( 2 i 6 8 10 Distance from top of column{m) (a 1 Solvent phase.

"mm 1200 i3oo Density IKg/m

3

) fe OS TO V2

Viscosity ImPo.sl

10 U IS

F ig .3 . U concentration profiles in aqueous phase.

In ter fac ia l tensiontmN/m)

f b j Aqueous phase

Fig, д. Predicted and experimental physical property profiles.

(11) Sel la f ie ld Pulsed Column Test Rig (PCTR) ­ Good agreement Is evident between the experimental and predicted concentration prof i les (or uranium and n i t r i c acid at steady­state ualng 301 TBP/OK shown In Figures S and 6. It should be noted that at the time of acquiring these experimental data the PCTR used a side­entry feed­pipe which gave Ineff ic ient feed dis tr ibut ion . However, by adjustment of the model i t was possible to match the observed prof i l es and thus highlight the use of BRADSM ав a means of Identifying non­Ideal hydraulic behaviour. Currently such large scale columns are f i t t ed with feed distr ibutors to a l l e v i a t e th i s e f f e c t .

_ ЙЙГГ}«— 5

1* Г"^ ! •

1

­ J 1 к.,

л Experimental 1 . Pr.dM.d / * 4 » ~ »

• Experimental I Predlrt.d I a»»»

1

Г""'­,,

""*\ ­­ ­ г ­ " . . . . .

Distance (ram lop «I column (m)

FI6. 5. U CONCENTRATION PROFILES

Distance (rem tap ol column (m>

FIG. 6. NITRIC ACID CONCENTRATION PROFILES

213

Conclusions. A model, BRADSIH, has been developed which predicts the dynamic response of a pulsed place column, fro» the onset of an operational transient to steady­state.

Overall, the agreement of the model predictions with experimental data for V extraction by 202 and 301 TBP/OK is good la both pilot plane and large industrial contactors.

additionally it should be noted that, since the correlations for the mass transfer coefficients used la the model are fundamental In nature and not specific to che Purex flowsheet conditions, the oodel can be applied to a vide range of flowsheet conditions and other systems.

Experimental and industrial columns will exhibit non­ideal flow behaviour which will depend on the column geonetry. BRADSIM can be used to diagnose this behaviour and predict Its effect on column performance.

Acknowledgements . BRADSIM Is based on Initial work at the University of Bradford by Professor W.L. Wilkinson and Dr. J. Ingham. The experimental work was undertaken by teams led by Mr. С Phillips at Sellafield and Mr, J.A. Jenkins at Harwell. A valuable contribution has been made by Mr. D.A. Houlton (Bradford university Research Ltd) . Long­term suppot,. has been given by Mr. A.L. Mills (Harwell) and, Dr. W. Batey and Dr. P.J.Thompson (Dounreay).

References

1. Jenkins J.A. et аУ/Proceedlngs of "Recod '87", Vol,II­ P«1219, Paris, France, August 1987,

2. Philips C.//Chenlstry and Industry, 2 September 1987. P*577­582.

214

?[:

EXTRACTION OP CESIUM, STRONTIUM, RAPE EARTHS AND I 10—27

TRANSPLUTONIUM ELEMENTS PROM LIQUID HIGHLY RADIOACTIVE I WASTES BY EXTRACTANT BASED ON COBALT DICARBOLYDE B.Ya.Galkin, V.M.Bsimantovskii, L.N.Lazarev, R.I.Lyubtsev, 7.N.R0­manovskii, D.N.Shishkin , Khlopin Radium Institute, Leningrad, USSR: L.Kadletsova, M.Kyrsh, I.Rais, P.Selutskii, Institute of Nuclear Research, Biei, CSSR

The need for processing of liquid highly radioacti ,e wastes (HAW) arising from radiochemical technology is determined jy the require­ments for long­term safe storage of these wastes and by the demand tor some radionuclides.

90 1Y7 HAW contain the following main radioactive elements: 'Si, J Cs,

TPE C 2

*1

' 2« A B , 24+ C B ) i E E ( Н 4 0 в ( 144j X t 147 Р т > 1 5 1 S B > 154, 155 E u ).

The present report concerns the summary results of the systematic­investigation on oAiracting properij.es 01 i.~^tu ax. lj...

.J I

"'3 )

"1

'2

~C

2B

9H

1lj2C o

J solution In nitrobenzene. The diearbolyde polyhedral' completes of some transition metals are known to possess strong hydrophobic features. Complex anion salts are readily soluble in polar organic solvents of nitrobenzene type and largely dissociat­ed 1л the organic pht.se. This is due to inert shell of the anion which interacts rather weakly with positive ions, Carborane complex anion form is synthetized from ortocarborane by ­ BH­group elimina­tion on boiling in methanol in the presence of NaOH and as a result of reaction between the formed acid and cobalt dichloride in strongly al­kaline medium.

Metal cations in the presence of cobalt dicarbolyde in anion form (D~) are extracted into nitrobenzene as dissociated ion pairs accor­ding to the following reaction:

H+* + nD~ =* M + n + nD" . (1) eq * ""aq ••­ org * ^org* v

" The extraction nrcceede by substitution of И + for H + between the

aqueous and organic phases: "taXrg­Cg + aV C2) In the case C^+n <*C H+ D­,[H

+

] o r g ­ [»"] o r g = C H

+

D~« the distribu­tion coefficients are

°H+D" »n • K

SH 3 — • u) she:a Itjg is an equilibrium constant of extraction exchange.

21S

Рог log к£ в ~ 10 2

, i.e. foi effective extraction of cesiuo froc; acidic solutions (up to 4­6 !l HHO,) , the extractant concentration should be within 0,01­0,1 11. The alkali­earth elements are extrac­ted rather weaker. In order to extract strontium from 0,5 M ЕШ0, at the distribution coefficient "­2, the extractant concentration should be relatively high (­0,5 M).

Constants of extraction of cations from the aqueous phase into ni­trobenzene are defined by hydrate number of a specific cation in ni­trobenzene! the greater the hydrate number of cation is, the lower is the constant of its extraction. This dependence explains the great difference in recovery of alkaline, alkali­^arth and rare­earth ele­ments in the presence of cobalt dicarbolyde in nitrobenzene, as the equilibrium constants of extraction exchange differ markedly for these element groups. So, in contrast to cesium, for the exchange reaction

B

< q + 2 H

org * B a

org + 2 H

aq (4) the value of log Kgj is 0,69+0,94 (hydrate number hfia+2=11,5+1) even a t C

HH0 3^1 »• Por the exchange reaction Ce+% + 3H+ 5*Ce+ + 3H+ (5) aq ^ org org aq

the value of log K°| = 0,5141,31 (hydrate number hCe+3=16,2+2) at °H

+

D " ""bin 0,04­0,3 X and С ы о within 0,3­0,5 ». Futher Investigation disclosed a synexgetic effect resulting in

complex formation of strontium, barium and trivalent elements with compounds of linear polyether type as for example polyethylene gly­cols (P)• These complexes pass into the organic phase more effectively than free ions according to the following reaction:

S*ll + pn.» + 2D~ ^SrP+fL + 2D" „ . (6)

aq org aq org org The presence of polyethylene glycols Increases significantly the

distribution coefficients of strontium, barium and trivalent elements and affects slightly the extraction of cesium.

The extraction of trivalent elements is characterized as follows! ­ the relations between log D_ and log Сд + may be expressed as

a line with a slope ­3l ­ the relations between log D № and concentration of extractant

are expressed also as straight lines with elope +2. These results should be accounted for by the following proposed

extraction mechanism: K\ * P

a, • org— **%T& * K4' ™ where №*^ la a complex in which one hydrogen at on Is clearaged from polyethylene glycol or hydrate sphere of «atml cation.

216

The composition of complexes 1л the organic phase waa not estima­ted and could be fairly complicated. The complexes BuP+

­^ and FuP*? prevail at loner concentrations of polyethylene glycol in the organic phase. At higher concentrations of polyethylene glycol these complexes react with next 1­2 molecules and give г!ве to some other compounds with probable composition BuP^3

, EuPjH^, EuPj 3

. Extraction equilibria in the case of trivalent elements are more

complicated as compared with extraction of mono­ and divalent catiore. In s+udles of radiation stability of dicarbolyde complexes the high

Jf­radiatlon resistance of extractant was revealed and an interesting phenomenon of radiation synthesis was detected which resulted in con­version of the main anion in irradiated organic phase to halogen­de­rivative compound with higher chemical stability.

In the last few years the extensive investigations have been under­taken with the aim of examination of the properties and possible use of chlorinated cobalt dicarbolyde for technological recovery of cesi­um, rtrontium, lanthanoids and actinoids from the liquid HAW.

The data obtained suggest that this extractant offers many advan­tages over the known extractants for cesium and strontium. It can be used in strongly acidic media, it is chemically and radiation resis­tant and provides good efficiency for recovery and separation of Cs, Sr, TPE and RE.

On this basis a technological flowsheet for cesium and strontium recovery was elaborated and tested with actual HAW solutions in a la­boratory installation mounted in hot cells. The flowsheet envisaged the recovery of cesium by extractant containing 0,C6 U chlorinated cobalt dicarbolyde in mixture with nitrobenzene and carbon tetrachlo­ride, extract washing and stripping of cesium by nitric acid. Stron­tium was recovered from raffinate by similar extractant with addition of 1 vol. 56 of "Slivafol 909". The teet of strontium and ceBium reco­very flowsheet was performed by using the raffinates arising from the first oycle of WEB spent fuel reprocessing without any preliminary adjustment. The recovery of Cs and Sr in extract was 99,8 %, the to­tal losses did not exceed 0,3 %. The decontamination factors for ac­companying ­nuclides were over 500. An essential drawback of the flowsheet consists in the use of two extractants for Cs and Sr recove­ry.

The further improvement of Sr and Cs recovery with chlorinated co­balt dicarbolyde leads to development of a technological flowsheet with one extractant.

The test data on simulated solutions indicated that cesium recovery was about 98 % and the decontamination factors for main radionuclides ( * 4

Ce, °Eh, 1 5*Eu) ware ~" 10 3

. The strontium recovery degree was

2>7

more than 95 X, the decontamination factor for cesium was ~10 , for fission products N 5 0 0 . The 4­fold concentration of strontium aa сов­pared with the feed simulated solutions was achlevedi experiments with cesium concentration failed.

The positive test results for the simulated solutions enabled to proceed to laboratory experiments on real rafflnate from spent fuel reprocessing. The recover; of cesium and strontium was about 98 % with rather high decontamination from accompanying radionuclides.

The concentrations of main macrolmpuritles in cesium and strontium reeztracts did not exceed their content in feed solutions. The reco­ver; and decontamination characteristics were not subjected to signi­ficant changes under the process conditions and this is sufficient demonstration of steady operation of the extraction flowsheet.

Along with the development of the processes for cesium and stron­tium recovery, the recovery of trivalent elements by the same extrac­tant was investigated. As a result, a complex flowsheet was proposed which made it possible to recover Sr, Cs, RE and THE from HAW. The la­boratory experiments provide data on the main extraction and technolo­gical parameters of the flowsheet.

Thus, the use of chlorinated cobalt dicarbolyde as extractant per­mitted to develop a technological flowsheet with one extractant which afforded the fractional recovery of Sr, Cs and RE­TPE sum.

References 1. Gerow I.H., Smith I.E., Davis M.w. // Separ.Sci.Technol. 1981.

Vol.16. P. 519. 2. Blasius J?.. Hillea K.H. // Badiochim Acta. 1984. Vol, 35. P.173J \

1985. Vol.36. P. 207. 3. Hishida E., Talcada H., Yashimusa. // Bull. Chem.Soo. Jap. 1984.

Vol.57. P. 2600. 4. Sohulz W.W., He Ysaac L.D. // Feport ABH­SA­263. 1977. 5. Havratil J.D., Martella Ь.Ь. // Separ. Sci.Teohnol. 1981. Vol.16.

P. 1147. 6. Yakahin V.V. et al. // Dokl. AIT SSSB. 1984. Vol.277. P. 1417. 7. Galkln B.Ya., Lazarev L.U. et al. // Issledovaniya v oblastl pere­

rabotki topllva . Prague, 6sSB. 1981. Vol.3. P. 272. B. Galkin B.Ya., Beimantovskiy V.H., Kyis H. at al. // Int. Conf. on

Hucl. and Badiochem. Lindau, PEG. 1984. P. 105.

216

SEPARATION AND RECOVERY OF TRANSURANIUM ELEMENTS FROM LIQUID WASTES PRODUCED BY THE CASACCIA PLUTONIUM PLANT

M. C a s a r c i , R . C h i a r i z . i a , G . M . G a s p a r i n i , G . P u z z u o l i , G . V a l e r i a n i ENEA F u e l C y c l e D e p t . , C . R . E . C a s a c c i a r R o m a , I t a l y

At t h e E n e r g y R e s e a r c h C e n t e r C a s a c c i a ( R o m a ) t e n c u b i c m e t e r s of a l p h a c o n t a m i n a t e d l i q u i d w a s t e s , o r i g i n a t e d by t h e U-Pu mixed f u e l f a b r i c a t i o n a t t h e P l u t o n i u m L a b o r a t o r y / a r e s t o r e d . As t h e i r Pu-Am c o n t e n t i s t o o h i g h t o be c o n s i d e r e d a LLLW, a s i m p l e p r o c e s s p e r m i t t i n g t h e t o t a l e l i m i n a t i o n o f a l p h a e m i t t e r s i s u n d e r i n v e s t i g a t i o n .

The p r o c e s s , n a m e d TESEO, b a s e d on t h e l i q u i d - l i q u i d e x t r a c t i o n t e c h n i q u e , w i l l u t i l i z e a new n e u t r a l b i f u n c t i o n a l e x t r a c t a n t , t h e o c t y l f p h e n y l ) - N . N ' - d i i s o b u t y l c a r b a m o y l m e t h y l p h o s p h i n o x i d e (CMPO):

C 8 H 1 7 ( C 6 H 5 ) P ( 0 ) - CH 2 - C(0) - N = ( C H 2 C H ( C H 3 ) 2 ) 2 '

T h i s c o m p o u n d , d e v e l o p e d a t t h e C h e m i s t r y D i v i s i o n of A r g o n n e

N a t i o n a l L a b o r a t o r i e s { A N D , C h i c a g o ( U S A ) , e x t r a c t s a l l a c t i n i d e s i n

d i f f e r e n t v a l e n c e s t a t e s f rom a c i d i c m e d i a . T h e TRUEX p r o c e s s ,

d e v e l o p e d a t ANL f o r t h e t r e a t m e n t of TRU c o n t a i n i n g w a s t e s [ 1 ] , i s

b a s e d on t h e u s e of CMPO.

Aim o f t h i s a c t i v i t y h a s b e e n t o d e v e l o p a TRUEX-lifce p r o c e s s f o r t h e : • selective separation of alpha emitters (mainly U, Pu, Am), in order to decrease the alpha activity of the liquid wastes at less than 370 Bq/g (10 nanocuries/c of disposed form}

* recovery of Pu (and in some casa of U •") i n a f o r m suitable to allow its use in the fabrication plantj

• separation of the alpha emitters in a small volume.

The items investigated have been the: - classification of waste solutions with regard to their chemical

composition in four types: nitric, oxalic, alkaline, analytical waste solutions(2J< Table )J

- verification, with simulated solutions of the possibility of mixing these different waste solutions without formation of solids or third phases. The mixing of • he wastes has the aim to avoid t-he study of four different flowsheets,one for each single waste[31j

- purification of commercial :MP0 produced by MST Chemicals Inc.,Rohway,(N.J.)070 95, with .ifferent methods(cristallization and ion-exchange) and study of cs impurities[2]\ choice of the oraanic solvent composition(0.25M CMPO,1M TBF in tetrachloroethylene(TCE)13]j

2 9

ALPHA LIQUID WASTES TO BE TREATED

a c i d i c waste oxa l i c waste a l k a l . waste a n a l y t i c a l wastes

Type l HNO3 Pu U

Type 2 Type 3 3-7 M Oxalic Ac. 0.6 M NH4OH 11M

24 HNO3 1 M N H 4 N 0 3 0.5M

50. i g CCI4 iraces 9.14д T H F A O 0.5M

Pu 14g

Type 4a HSCN 0 0 2 M

43.3 g Pu U

F e ( S 0 4 ) 3 0.03M H 2 S 0 4 0.7M Н 3 Р 0 4 2М HNO3 '0.06M U 815g Pu 20g others(") iraces

Volume 934 L Volume 153 L Volume15.1 L Volume 50 L v o l u m e 36 L

Type 4b АдЫОз 20g/l F e ( S 0 4 ) 3 0.01М U 1017g Pu 12.56 g othersO traces

П tetrahydrofurturylic alcohol; ("J ( N H 4 ) 6 M 0 7 O 2 4 - 2 4 H 2 0 , ( C e H 5 N H C 6 H 4 S 0 3 ) j B a . К2СЮ7 О АЦЫОзЬ. HSCN. O e 2 ( S 0 4 ) 3

s e l e c t i o n of seven samples of r ea l waste so lu t i ons with d i f f e ren t chemical composi t ion; de te rmina t ion of t h e i r radiochemical and chemical composition [3] ', e x t r a c t i o n (F ig . l )and s t r i p p i n g ( F i g . 2 ) t e s t s in batch with s u i t a b l e mixtures of these r e a l waste s o l u t i o n s [ 3 ] ; modelling of a reference flowsheet obtained by means of the SOLVEX computer c o d e [ 3 ] ( F i g . 3 ) j experiments in mini mixer s e t t l e r s with mixtures of r e a l waste s o l u t i o n s ' definition of the parameters of the TESEO Process.

о 1 2 Э 4 s 6

Contact number

Ptg.1. EXTRACTION TEST FOR A TYPICAL MIXTURE OF REAL WASTE SOLUTIONS

220

7 8 Г 'Р и S , n p

О

Contacts number

Fig. 2. SELECTIVE STRIPPING TEST Stripping solulion:0.04M HN0 3 for Am(III) and 0.04М HN03+ 0.05М HF for Pu(IV)

Mixing W*tt*s

Chemical­physical Characterize Hon

' • . ChtfniCKl Atfj slm>

RUralfon and/or CmirihigaUMi HNO3 0J2M

B.A.N. ILW 1 1 ; ILW

1 1 ; Extraction Sedton

• Бмдп 0/A ­ 0.15

1 S e n * Stcltofl 4Siagaa (VA ­ 3

Extraction Sedton • Бмдп

0/A ­ 0.15

S e n * Stcltofl 4Siagaa (VA ­ 3

Extraction Sedton • Бмдп

0/A ­ 0.15

S e n * Stcltofl 4Siagaa (VA ­ 3

HN03 0.04 M HFO.OSM

Am Strip Section в Stagn ОГЛ ­ 4

I . V CMPO 0.25 M TBPl M T»trachtoro«(hy1en*

Pu Slrip Section 4Stagei 0 / A ­ 4

Am Product

Fig. 3. TESEO PROCESS SCHEME to Solvent Regeneration and Recycle in Extraction Section

22t

The basic TESEO process flowsheet involves the following operations: waste solutions of different type in 2 liter plastic containers are introduced in a first glove box,where the feed mixture is prepared.To this purpose calculated amounts of solutions with appropriate chemical compositions are mixed at predetermined ratios.After chemical adjustment and filtration, the feed solution is trasterred into a second glove box,where the extraction is carried out. An 8+4 stages mixer-settler battery(efficiency 90%> is utilized for the extraction and for the scrubbing sections.The scrub solutions are 0. 2H HNO3 and 0.5M basic aluminum nitrate. In laboratory experiments a volume reduction factor of 6.25 was reached in the extraction-scrub section and 4 in the stripping section. It is possible to obtain a total Pu+Am stripping in a 4 stage mixer-settler or a selective stripping of Am (with 0.04м HN03;4 stages) and Pu (with 0.04M HNO3+ 0.05M HF,-4 stages) as shown in Fig.3.Uranium is recovered during the solvent regeneration with a Na2CC>3 solution.The decontamined solutions can directly be transferred to a cementation unity.

The planning,the installation and the operation of a small scale plant are foreseen during the years 1988-1989.

This activity is partially supported by the European Community in the frame of the RfcD Programme on Management of Radioactive Wastes.

лжгжхшсжэ 1 . R.A Leonard,G.F.Vandegrift et a l . ANL-87-3 (1987). 2 . M . c a s a r c i , R . C h i a r i z i a e t a l - CEE Contract n° FI1W-0012-T.-

I°Semestral Report July-Dec 1986. 3 . M.Casa rc i ,R .Ch ia r i z i a e t a l . CEE Contract n° FIlw-0012-T;

H°Semestra l Report Jan-Jun 1987.

222

UJUKIUM ЕХТЯаСПОН КНОИ PHOSPHORIC ACID USb.""} DEHSELY PACKED [7^^] MSPIRSIC»S ' ' M. Erotkl and a. acvorj­ta Institute еГ CheKioal Engineering and Beating Equipment, Technical University of Wroclaw, «roolaw, Poland Introduction. In procsssee Tor uranium recovery from phospborlo

acid liquor* the improvement of axtraotion equipment la «till a atda development araa. ТЪа majority of uranium solvent axtraotloa plant* oea bank* of Mixer ­ aattlar* which taka op auoh apao* especially for tbe •operation of phaaaa after mixing.

To reduoe the то1ши of equipment, centrifugal extraotora and aoae type* of ooluan* bare been need froai time to time but tbere oeenred •oee difficulties with emuleifloatlon or phase*, a tendonoy to create • table dl*peralon* to obaraoteriatio for uranlaa extraction *7ate*i* and reeulte from low ooalaaoenoe rate of drop*. On the other band tbi* disadvantageous feature oan be utilised purpoeefnl «ben opera­tine tbe extraction prooesa In a «pray oolumn in which a densely pac­ked diap*r*ion 1* maintained. Suoh aode of prooea* operation i* also oallad extraction with dense packing of drop* and It 1* suitable to liquid *y*teaa Just with low ooalaeoenoe rate.

Tb* experimental investigation* of epray ooluas working witb dense packing of drop* abowed that high holdup values v up to about 70£ could ba obtained [1,2,3] with resultant high interfaoial araa. Fur­thermore, low baokmlxlng ClJ or even abaanno of baofcmixing [5] in both phases had also been found. Therefore, the use of deneely paoked dispersion* seems to be an interesting propose 1 for the extraction of uranium. The aim of this work was to test this experimentally.

apparatus. The densely paoked dispersions were obtained in two ex­perimental oolumna of the same diameter but different height*, a* abown in Fig. 1. The first column was a standard «pray column of Elgin design. It wa* constructed from glass and the only material* in con­tact with tbe liquid* were glass, stainless steel and teflon. The height of the column proper waa 1,0 a and the diameter wee 0.0*1 m. Tbe distributor waa plaoed in the upper expanded part, being a eet of nozzle* with diameter of opening* equal to 1,0 am.

For the conventional operation with looaely paoked diaperelons and lowar hold­up the ooalesoenoe interface was kept constant in the ex­panded bottom vessel. To form a densely packed dispersion the lnter­faoelevel wae raised into the extended lower part of the oolimn pro­per. The interface area available for coalescence waa thus decreased and the drops began to form a dense dlsperelon.

The column of type II waa ef the «erne diameter but its height was *,0 a. In that ooluan toe conventional distributor was replaced

223

ft»

«40 Ф40

й with я mechanical dispersing de­trice to avoid crusting «ad clog­ging of the smallest orof* sec­tions, which is * well known phe­nomenon in the phosphoric acid induetry. The droplet dispersion was produced by means of я set of 5 reciprocated sieve plates with diameter of hol*s equal to 8 mm. The lower part of the column was modified using a conical section, within the coalescence interface was maintained. This design is an improved concept of Kehat's and Letan'a [6] recomendation.

Results. In this work U(IV) was extracted from phosphoric acid with nonylphenylphosphoric acid (NPPA) dissolved in kerose­ne* The concentration, of the ex­tractant in the organic phase was 0,25 шо1/1.

For both column­* uraniusi ex­traction efficiency tf as well as the basic hydrodynamio parame­

ters, i.e. holdup € and the height of densely packed dispersion bed H, had been determined* Exemplary results for a constant organic to water phase ratio 0/V =1*5 are presented la Figs. 2 and 3 for оо1ш ' 1 and II respectively. On the diagrams the total column loadings

interface

Fig.l. Extraction columns used in investigations

I ­ a standard spray column of Elgin type;

II ­ s modified column

С + " d ) were plotted on the abscissa axis againts the parameters H and £ successively. 9

The uranium extraction degree wing equationi

°R

had bean calculated fron the follo­

(1)

where o p I °ц •*• uranium! oonoentrationa In the water feed pbaaa and in raffiliate phaea, whereae oL la the equilibrium concentration in raffinate phaae related to the oreenio phase enterinic the column.

Pie. i a preeenta a oompariaon of uranium extraction effiolenoiea «кап firat the arcmnlo and next the water pnaae were dlaperaed in the column of type I. When the organic phaaa was diaperaad in the water phaae a oonatant extraction decree of oa. 13# waa obtained, indenpen­dentlr of the oelumn Loadlnc (line 1). In that oaae the oolunn worked

224

with 1Mealy paoka>d dlepere ione . The e f f i c i e n c y of uraolua e x t r a c t i o n could be inoreeaed by «boat k to 6 t l « e e while working with deneely pecked diaperslona (ettrrt 2 ) . ТЫ» wee poeaible whan tha we tar phaaa «•a tba diaperaed one» As l a аеетъ In Fig . 2a, the extrftotion e f f i c i e n ­

cy In coluem I Inoroaeod with lnoreeaad t o t a l flow ratea (u + u ) f

i 1 E 1

о)

<r У 0

] о л

1 1

т 0/W = 1:5

1

о ­A>—\ \

wbereea In ooli

<4 0.8

0.6

0.4

02

°. Him)

0,8

0.6

0.4

0,2

°< Z

08

0.6

0A

0.2

0

О a)

0/W = 1:5

И th la depaodanoa 1* ra»araed, as ahom In r i g . Ja.

Ф 0,8

0.6

0A

02

0

Him]

4.0

ao 20

1.0

0

1.0 2,0 3.0 4.0 ­10* 2,0 3,0 4,0 5,0 6.0­10"3

2 b)

f I E

I A 0/W: 1:5 / o I

1 I 1

Ы n

4

0/V /=1:5

1.0 2,0 30 4.0Ю9 20 30 4,0 5.0 6.0 Xf

о I c)

* v*1 "E

о

/ ^ H.1

0/W :1:5

/ V a I „ 1

1,0 2D 3.0 4,0'Kf u^+Uc | i t i ! n i i ­ s ' ]

e Q8

0.6

04

0.2

0

c)

0/V ' = 1:5

r i g , a. Coeparlaon of a * urenltm e x t r a c t i o n e f f i o i e n ­

oy ф , b ­ denaely peoked bed height H

and о ­ holdup 6

2.0 30 4.0 5,0 60­Ю* ц,+и е tm'­m'*­S

1

] У1д. 3, Effaot of tatal oolnnn

­ uraniua extraction erfi­olenoy f ,

> ­ height of tha denaely pao­ked bad Я and

for loaaly and danaalr paoWad dla­ ° ­ boldup t , paraion la ooluan of tjrpa I. f o r denial? paokad dlaparatona ourva 1 ­ danaa diaper» ion» l n oolunm of type II ourva 2 ­ looaa dlapereiona

lS.3aic.382 225

ТЫ» may be м Ш г explained by the fact that in tb» nalltr column the height of tb» dlaparaion bad inoraised (PI*. 2b) «blob causad alao an lnoraaaa In neldup value» (Fie. 2o). On tb» other hand, Uaa eeeand ooluan m operated with oonatant height of dispersion bad С Pig. Tb) end holdup vmluea deereeeed under tbaaa condition» (PI»;. 3o). Suoh a holdup dependenoe la oheraoterietio for tha operation «1th dan»» dla. paraloa» [1 a3t^] .

In tba ooluan of type I the bad of densely paokad dlaparalon beca­aa eatabiliohed at a certain height for aaoh pair of flow rataa u and u.» however tha height oould not ba aa hnalnail oonatant for a long

d time. Tb» gradual aceanlatioa of iapurltl*» at tha lntarfaoa oauaad tba rata of ooalaaoanoa to ohange and thua, tba height of dlaparaion bad to daoraaaa «1th time. Thla inoonvinienoe was eliminated in the bigger ooluan, in «hloh a oonloal bottom section wa» uaad. Such a con­•truotion made poaalbla tba eaay oontrol of ooalaaoanoa rata at tha Interfeoe and therefore, tba aaay aalntananoa of tha diaparalon bad at a oonatant height. Thii daai pn «a» tested and worked extremely veil.

Conoluaion». к aaall ooalaaoanoa ability, oharXbtariatio for ura­nium extraction syateme may ba utilised purpoeeful to create denaely paokad diaperalona, Which are oharaoterixed by low baokmlxlng and nigh hold­up» of tha diapei­aad phase. The raaulta of mass transfer inveati­gatiena showed that the uranium extraction efficiency could be lnorea­•ed by 4 to б tlaea by applying tha danaa dispersions, in relation to the conventional axtraotion with loose dispersions. In oolumns of f,0 and 4,0 • height wbloh had baan uaad in experiments the values of ига nlua axtraotion affloienoy varriad In tha rang» 55­80)6 and 75­99* for tha eaaller and the bigger ooluan raapeotlvaly. High uranium extrao­ > tion effioieaoy obtained in experiaenta may be interesting praatioally although columns of greater diameter should be tested before suoh an operation oan be uued industrially. It la planned to be done aa aoon as the corresponding equipment i» obtained.

Rererenoe» 1. Letan R., (afaat R^AJChE Journal. 1967. V o l . 1 3 , » 3 ­ P.VfJ. 2 . Loutaty R.,Vignea «.„//Oenle Cbiaiqu». 1969.Vol.101 ,N 2 . P.231. 3 . Howsryta A.. ,Kr6tki H.// Chea.Bloohea.&ig.Q. 1987, Vol . 1,N. 2 / 3 . I*. Parrut H.,Loutaty R. ,Le Qoff P^Chem.aog.Sol . 1973. Vol.2S. P.1541. 5 . Ste lnar L. ,Ugarolo N. ,Hartland S . / / Proo. 6­th CHISA Congr*»»,f978 6. Kenat £ . , l * t a n R.//Proo.De».and Do v. «968 .vo l . 7,N 3, P. 385.

226

10­ЭО DECREASE OP EHTRAIUHEHT LOSSES IN THE URAHIOM RECOVERY FROM' MET PHOSPHORIC ACID PROCESS A.Spasic, H.Djokovic, N.Canic, H.Babic Institute for Technology of Nuclear and Other Mineral Raw Materials, Beograd, Yugoslavia A long termed research programs is in progres to ascertain the redu­

ction of entrainsent losses and capital costs. The separation of entra­ined organic pbas"t «as examined in the pilot plant with organic phase loop presented in Fig.l.

The extraction of uranium was performed in four 3tages countercurr­ent "pump­mix" mixer­settler. Continuous organic phase was synergistic mixture DEHPA (0,5 M) ­ Т0Р0 (0,125 M) in kerosene and aqueous disper­

1. Organic phase tank 2. Four mixer­settler

extraction units 3. Three mixer­settler

reductioti stripping units 4. Lamellar соа1езсег with

perforated corrugated fillings

5. Four air induced flota­tion units

6. Settler unit 7. Adsorption column 8. Rafinat tank 9. Organic phase tank

Fig. 1. Organic phase loop

The aqueous phase entraine, at 90% confidence, 533+ 103 PP" of or­ganic phase in mixer­settler units. The aqueous phase was lead in the lamellar coalescer which was chosen and designed (1,2,3,4,5) for the reasons of simplicity and possibility of direct зса1е up. Thereafter, decrease of the organic phase content in the rafinat was effected in four air induced flotation units. The flotation overflow was split in the

n

3ettlern

and aqueous phase part was lead in the adsorption column. The separated organic phase contain some uranium and was lead back

in the mixter­3ettler units for reductive stripping.

227

Determination of the entrained organic phase content was realised by following methods: ваз cbronatography (i) and centrifugal jeparation in the bottles with capilary tubee (y),(6>. By gas chromatography beside the entrained organic phase the dissolved DEHPA and TOPO in the aqueous phase ыеге determined. As their sollubility is neglectible the correla­tion of these methods is possible. Regression analysis of 17 parallel measures give the following relation:

у = 0,75 X + «9,70 with correlation coefficient r = 0,959 what does mean that the confide­nce of the find relation is higher than 99,9*. The method of centrifu­gal separation (y) was used for the contents of organic phase in the range of 50­800 ppm (vol/vol) and the method of gas chromatography (i) for the contents of organic phase in the ra>.ge of 20­70 ppm (vol/vol).

On th« basis of the obtained results material balance for the rafi­nat and entrained organic phase is evaluated and presented in Table 1.

The efficiency of the lamellar coalescer unit was 62%. Thereafter, efficiency of the flotation units was 77*. Feed content in adsorption column was small; about 100 ppm (vol/vol) and it's efficiency was 63%. The overall efficiency of the examined conception for the separation of entrained organic phase was 91%­

The Fig.2. presents егозз sections of designed lamelar соа1езсег as well as corrugated perforated filling plate. The air induced flotation units where of "Denver sub A № 5 " type and the adsorption colucn was filled with activated carbon.

r 1 a t>. r

iffejl I ^ ^

r

*

! • ! I D D D " ^ 7 ­ |Jo"o

z •Mitt

d

Fig. Z. Cross­sections оГ lamellar coalescer a,b and corrugated perforated filling plate c,d.

228

Table 1. The material balance for the rafinat and the organic рпазе

Bquipnent and

place

Foss i t ion on toe F i ­gure 1

Voluae of toe raf inate

Distr ibu­t ion of the raf inate

Content of the organic phase in the rafinate

Voluae or the orga­n ic phase

Distribu­t ion o f the orga­n i c phase

(1) ?(<яА M ( v o l ) OPJ(vob'vol) VtdB3

) M(vol>

Соа1езсег 1 _ ­ ­ _ ­Input 2 . 1 . 111,0000 100,0000 0,0533 0,059163 100,00 axierflow 11.1 110,9636 99,9672 0,0205 0,0227*5 38,46 overflow 4 . 2 . 0 ,036

й 0,0328 99,9999 0,036408 61,54

;­jur f l o t a ­

t'. *i се11з 5 ­ ­ ­ ­ ­input 4 . 1 . 110,9636 99,9672 0,0205 0,022755 38,46 underflow 5 ­ 1 . 99,8672 89,9704 0,0040 0,003994 6,75 overflow 5 . 2 . 11,0963 9,9967 0,1690 0,018760 31,71

Secondary coalescer 6 ­ ­ ­ ­ ­input 5 . 2 . 11,0963 9,9967 0,1690 0,018760 31,71 underflow 6 . 1 . 11,0887 9,8808 0,0100 0,001109 1,88 overflow 6 . 2 . 0,0176 0,0159 99,9999 0,017650 29,83

Colum with activated carbon 7 ­ ­ ­ ­ ­input 6 . 1 . 11,0787 9,9808 0,0100 0,001109 1,88 output 1 7 . 1 . 11,0780 9,9801 0,0037 0,000410 0,70 output 2 7 . 2 . 0,0007 0,0007 99,9999 0,000699 1,18

Accroding to the results obtained from such pilot plant the residual content of entrained organic phage, at *)0% confidence, was 53+16 ppni (vol/vol), therefore the losses of organic phase wero 8,63* (vol).

The challenge from the technical and economical point of view could be investigation of the new conception of organic phase loop. The sup­stltution of the flotation units by densely packed coalescer could be examined. This question have to be realised with respect of specific characteristics of the (DEHPA­TOPo­keroaeneJ ­ phoshoric acid system.

The brife comparative analysis of this two methods will be discu­ssed. The considerations affecting the choice of the method and equi­pment from the process, operation, design and economic points of view are presented in Table 2.

229

Table 2. Comparative analysis of selected methods

Hethods Tor secondary separation Flotation Coalescence

(densly packed)

Ргосез considerations ­ efficiency ­ residual content of entrained org.ph.

­75* ­ 25­50ppn (vol/vol)

­90* ­20ppm (vol/vol)

Operation considerations ­ different characteristics ­ requirement of stable

feed content (one safe­ty settler)

­ sensitive to the temperature chan­ges

­ separation of tbe flo­tation overflow (one "settler")

­ sensitive on the occurance of sta­ble third phase

­ sensitive to the temp­erature changes

­ requirement of fre­quent regenerations

­ throughput ­ big ­ small ­ maintenance ­ easy ­ uneasy Design considerations ­ information status ­ lot of patents ­ not enough ­ improvements ­ limited ­ possible Bconoaic considerations ­ capital £osts ­ very high ­ high

Tbe answer wbicb method is more convenient have to be searched on pi lot­plant scale under consideration of all specific problems which could appear within tbe scale up.

References 1. Micbel P^ugoslav­Frencb.sytap. on modern technologies. Dubrovnik,

1987. 2. Daviater A., Hartin G.//ISEC'86. Hunchen, 1986. 3. Josa J.*., Moral А./ДБЕС"80. Liege. 1980. 1. Spasic A., Djokovie H., Canlc N.. Babic H., Tolic A./*HISA'87.

Prague. 1987­5. Spasic A., Djokovic M., Babic M. MCCE. Barcelona. 1987. 6. Lewis I.E.//CM, spec. 21, ISEC'77. 1977.

230

REEXTRACTION OF URANIUM (VI) FROM 0,3K D2EHPA • 0.075H TOPO IN KEROSENE WITH AMMONIUM CARBONATE SOLUTIONS

I. Fatovic, S. Meles, M. Radosevic, J. Romano INA­Researeh 4 Development, Zagreb, Yugoslavia

10­31

Introduction. We have recently studied the mutual influence of extraction of uranium (VI) and iron ( I I I ) and the effect of removal of iron from the organic pha­

se before reextraction of uranium (VI) with ammonium carbonate solution in the pro­

cess of uranium recovery from wet phosphoric acid O ) . The purpose of this investigation i s to determine the conditions of uranium (VI)

reextraction with ammonium carbonate from the organic phase obtained by the extract­

ion of wet process H­jPO , without previous removing of iron ( I I I ) .

Experimental. Crystalline uranyl nitrate was used for a l l uranium solutions. Di­2­ethyl­hexyl phosphoric acid (D2EHPA) and trioctyl­phosphineoxide (TOPO) were provided as a high purity product by Mobil Oil Co. Organic solutions were prepared with D2EHPA and TOPO in nonaromatic hydrocarbon diluent Mobil 190/210 and by dis­

solving uranyl nitrate and Fe ( I I I ) chloride, and the solution was purified by mul­

tiple water washes. Aqueous solutions of selected concentration were prepared from the (NH1,)_C0T analytical grade "Kemika" and carbamate solution from industrial plant Kutina (intermediate of urea). The measurements of distribution rat ios were perfor­

med in the thermostated bath. Equal volumes of aqueous and organic solutions were mixed mechanically 10 minutes at 50 C. I t has been previously established that the temperature does not effect the efficiency of reextraction, but influences only on the speed of phase separation. The concentration of U(VI) and Fe(III) in the aque­

ous and organic phases were determined by spectrophotometry (2).

Results and discussion. The dependence of distribution ratios on the concen­

tration of ammonium carbonate is shown in the figure 1. Increasing ammonium carbo­

[(NH4)2C0£molclm

Fig.1. Distribution ratios of U(VI) vs. initial concentration of (NH^J­CO­: 1 "Kutina" and 2 anal, grade,

=3,4.10" mol.dm"

1 2 3 4 [U(VlJ Ю

2

, mot dm3

Fig.2. Distribution ratios of U(VI) vs. ini t ial concentration of U(VI): 0,4М(Щд)2С03; с и ( у 1 ) = З А 10~*mol.dm

U(VI)" 231

nate concentration results in the increase of distribution ratios in all cases. The effect of carbonate composition on the efficiency of reextraction is also presented. The valves of distribution ratios of uranium are higher if the reextraction is per­formed with carbonate 1. The typical curves of dependence on the uraniim concentra­tion are shown in figure 2. The increase of value of distribution ratios of U(VI) at lower in i t ia l uranium concentrations is more expressed with carbonate 1.

Considering the results of previous investigations of U(VI) and Fe(III)ccextrac-tion (U the dependence of uranium reextraction on the concentration of extractant has been tested. The values of U(VI) distribution rat ios decrease with the increase of extractant concentration in both cases (Figure 3).

The dependence of the distribution ratios of U(VI) on the phase volume ratio a t various concentrations of ammonium carbonate are shown in figure *4, Further re­sults have shown that with the phase volume ratio aq = 2/1 the efficiency of reext­raction i s > 99%, i f in organic phases re(III) i s present or not. However, the FeClIIJ which is present in organic phase causes the precipitation of Fe(OHK in solutions of ammonium uranyl tricarbonate in the second or in the third stage. The precipitation depends on the concentration of Fe(III) in organic phase, concentra­tion of aqueous phase and in i t i a l an(i equilibrated pH values.

D2EHPA,mot 0mJ

Fig.3. Distribution ratios of U(VI) va. concentration of extractant:

mol.dre

The results show that for the optimal uranium reextraction i t is necessary to control the pH values and to main­tain them within the limits 8,0*8,4(3). The efficiency of reextraction is not satisfactory, below the value pH=8 and above pH=8,4 Fe(OH-J is being pre­cipitated in the solution of ammonium uranyl tricarbonate (Figure 5,Table ). Inthe figure 5 the slight decreasing of U(VT) distribution ratio caused by the presence of Fe(III) in organic phase is also visible. The concentration profile of s ta t io­nary state after twelve hours of con­tinuous counter current extraction from 0.3M D2EHPA + 0.075M TOPO with 0.38M (NHy)2C03 in mixer set t ler units is shown in figure 6. The phase volu­me rat io i s o/aq=2,5/1 and the reac­tion temperature (working temperature)

232

Mult is tage r e e x t r a c t i o n of CKVI) Г on 0,3M D2EHPA * 0.O75M TOPO

with O.IMCNH^Oy o/aq=2/1 , ^п.*3,*.10~'вА.11а~';

4e(III )= 9

'5

­1 0

"m o l

­d j , 1

":

'

Stage DU D Fe pH l n l t . pHeq. % r e e l . ( I ­ i r i )

I X 0 ,010

0,030 0,003

0,000* 0,0005

9,76 9,76 8,30

7,77 7,76 6,29

I I *5,21

1,16 0,08

8,15 8,12 7,53

I I I x 3 1 1 , 0

152,0 5,2

Fe(OH) 8,56 8,53 8,01

99,95 99,87 83,39

* without F e ( I I I ) In organic phase

ofag IK «5 1,tt Ifi 1.S

pHintt/pHeg

F l g . 1 . D i s t r i b u t i o n r a t i o s of UtVI) v s . phase volume r a t i o o/aq: атц)гС0у о 0,1М; • 0,5M; D0.75M

"­"(VI) =3,1.Ю" mol.dm

F i g . 5 . D i s t r i b u t i o n r a t i o s of U(VI) v s . pH i n i t . / p H eq. r a t i o s :

a n a l , grade;

^ F e ( I I I ) :9 ,5 .10 mol.dm

4HVI) =3,1.10 mol.dm

;зз

O I C . V I S H 1 1С E I S T H C T I D I

*«2«/L О <W6g/L Fe

3

QflSg/L Fe

Q 1 9 « / L P 2 0 5

*0Og/L U QP5K/L Fe QOtg/L Р г 0 5

V 6 e / L Р г 0 5

В г О ft 7 L/H

462g/L U 007g/L Fe QD9g/L F«

ft«7g/L P ? 0 5

196g/L U Q9«g/L U C£7e/t­ D <V>5g/L U

V ) 5 g / L U

Q59g/L P ? 0 5

025g/L С Q,1*«/L Fe

H . P O . U / H

S T R I P P I N G REGENERATION

<tfOg/L U A04g/L Fe • e ' l ­ P 2 O 5

2 ^ l 4 g / L U

PTOg/L U qplS/L. Fe

'A • g / L P _ 0 ,

4 0 2 C / L U

0P5S/L U AOlg/L­ F« •g /L РгО«.

Q 2 9 g / L U

org.phase 5 L/H 0 , 0 8 g / L U

Q02g/L Fe

B g / L P ^ 0 5

0.38И (нн4)гса3

г L/н

Fig.6. The concentration profile of secondary extraction/atrip system

has been ^9 to 50 C, and the pH values in the course of process have been maintained within the limits from 8 to 8,45. This has been achieved hy adding 20% NH^OH and HC1 (1:1). The efficiency of reextraction has beer. 99% and the precipitation of Fe(0HK in the solutions of ammonium uranyl tricart­'nate has not been observed.

Uranium peroxidhydrate which has been precipited from such obtained solutions has been of high purity.

References

1. S. Meles, I. Fatovic, M.V. Prostenik//J. Romano. ISEC'86. J_*2?

5( 198b). 2. I. Brcic, £. Polla and M. Radosevic//­Analyst. 170. 7349,14630985?. 3. W.W. Berry, A.V. Henrickson. US Pat.Ц,302,427;1981).

234 I

HYDROMETALLURGY

GOLD RECOVERY FROM ACIDIC LEACH SOLUTIONS USING AS 1 1 _ g

EXTRACTANTS TRIALKYLAHINES OR N,N' DI ALKYL ALIPHATIC | _ AMIDES . . F.Baroncelli,D.Carlini,G.M.Gasparini,E.Simonetti ENEA­C.R.E.Casaccia, PB 2400, Roma,ITALY

I M T K O D D C T I O M . in this contribution the results of the experimental work on some liquid­liquid extraction processes for the recovery of gold from HN03/HC1 leaching solutions of Au minerals are summarized.

The solvent extraction of gold from HN03/HC1 solutions represents an alternative process when he cyanoalfcaline leaching and the carbon­in­pulp adsorption cannot be employed. Some Cu, As and Sb containing pyritic minerals and some carbonious rocks need expensive roasting pretreatments and sometime do not allow an economic gold recovery "1 '*', On the contrary, the aqua regia attack of these minerals permits good gold recoveries. Unfortunately selectivity problems, due to the presence of many dissolved chloroanionic complexes, can arise.

As gold extractants we have chosen a trialkylaliphatic amine, the TriOctylAmine(TOA) and a di­substituted aliphatic amide, the N,N­Di­n­ButylOctanamide(DBOA). Both extractants have been examined in batch and in mini mixer­settlers experiments using leachates of Peruvian and bolivian concentrates.With these minerals, very rich in sulfur (pyrites,stybine), a 90­95* gold recovery in 12­24 hours has been reached, by leaching with 4M aqua regia (HC1 3M /HN03 1М) at room temperature and with a 1/3 solid/liquid ratio. On these leachate solutions ( 2­3M total acidity, 10­60 ppra of Au) the "two processes with TOA fGAMRX PRQTESSI (3

J and with DBOA fAOMiPEX PROCESS)I

4

] have been tested and compared.

*** "*¥Bt PftflCTnS• This process, utilizes аз solvent for gold extraction a solution of TOA in a commercial mixture of aromatic diluents. A 0.14 M(0.5tw) extractant concentration assures high separation factors from Fe(III),Sb(V),Zn,Cu,Pb,Sn and alkaline earths, and high Au loadings (« lOg/l). Tertiary amines are typical liquid anion exchangers 5 ' .""heir selectivity can be attributed to the presence into the organic phase of TOA­HNOj, the predominant species in equilibrium with the 1:3 HNO3/HCI leaching solutions. This species exhibits a high affinity for Auchlorocomplexes and a low extracting ability for chlorocomplexes of other metal ions.

Co­extracted metals are easily scrubbed from the organic phase with 0.­5M HNO3. A 0.25M thiourea in 0.5M HC1 stripping solution

236

a J lows a very good Au recovery in aqueous phase, fol lowed by e i e c t r o l y t i c a l p r e c i p i t a t i o n t o obtain a metal of ­hiqh pur i t y . With the flowsheet shown in Fig­ 1, t e s t e d in l abora tory sca l e m i x e r ­ s e t t l e r s , a 99% Au recovery from very d i l u t e d feed so lu t ions and e leva ted concent ra t ion fac to r s have been reached.

Org I18ml/h TOA 5%w(0.14M in Sofvesso 100

4 5

Feed 598mVh Au = tOppm H

3 ­ 3.3M Sb = 5.3 g/l Fe = 12 g/l Cu,Zn,Sn,As and others еЫ Ig/l

Raff. 64Bml/h.

Aus 0.1 ppm

t—I I I I I I h 1 2 3 4 5 6

Sonc 59ml/h . \u =lC0ppm Fe = 178ppm St = 5ppm

Scrub 59mim

НГЩ = 0.5M

Org I18m(/h Au<Q.2ppm Fe =* 37ppm

Strip 59ml/h

HCI « 0.25 M T

hioureae0.25M

Fig. 1 . THE GAMEX PROCESS WITH TRI­n­OCTYL AMINE

THE АПМ1РЕХ PROCESS. T h i s p r o c e s s i s b a s e d on t h e u s e / a s s o l v e n t , O f a 0 . 2M <5ftw> solution of Ы . H ' ­ d i ­ n ­ B u t y l o c t a n a m i d e fDBOA) i n m e s i t y l e n e . T h i s compound, a l s o e m p l o y e d i n U­Fu s e p a r a t i o n s ' 6 ^ , i s a p o w e r f u l g o l d e x t r a c t a a t f r o m a q u a r e g i a s o l u t i o n s . T h e p h y s i c a l ­ c h e m i c a l c h a r a c t e r i s t i c s of DBOA a r e r e p o r t e d i n T a b l e . T h e

P h y s i c a l ­ c h e m i c a l c h a r a c t e r i s t i c s of DBOA

N , N ' ­ d , j ­ n ­ b u t y l octana,mj.d,e h a s t h e f c l l o w i n g s t r u c t u r e : C H 3 ( C H 2 ) 6 ­ C ( 0 ) ­ N ( C H 2 ­ CH2 - C H 2 ­ CH3}2

I t i s a c o l o r l e s s m o b i l e l i q u i d w i t h t h e s e c h a r a c t e r i s t i c s : M o l e c u l a r Weight 2 5 5 . 4 5 Boiling Point 100­120°C Density 0.861g/ l Refractive index n 20 1.452

­ 5 .

ImmHgl

i n n ­ d o d e c a n e :<*>;

d i b u t y l [6]

Solubility ( m o l e s / 1 a t 21°C) i n w a t e r : 2 . 3 10 i n m e s i t y l e n e : «> . I t i s p r e p a r e d by r e a c t i o n o f o c t a n o y l c h l o r i d e and amine J y i e l d > 8 0 % ) , b u t o t h e r p r e p a r a t i o n m e t h o d s a r e a v a i l a b l e Chemical stability: N , N ' d i ­ n ­ b u t y l o c t a n a m i d e i s a v e r y s t a b l e compound w i t h r e g a r d t o l i g h t , t e m p e r a t u r e and a c t i o n of a c i d i c or b a s i c , o x i d i s i n g o r r e d u c i n g a g e n t s . The p r i n c i p a l h y d r o l y s i s p r o d u c t s a r e octanoic a c i d and d i b u t y l a m i n e .

237

extraction ability of these substances is attributed to a solvation process similarly to other "oxygen donors" ligands like TBP. DBOA exhibits a good selectivity as shown by its extraction behavior with leachate solutions of gold containing minera.s. Concentrated solutions of FeUII),Hg(II) ,Sn(II) ,Zn,Cu are easily separated from gold in HNO3/HCI solutions. As with tertiary amines the only troublesome ions are Bi(lll) and sbUID.In Fig.2 the flowsheet of the process is reported.

Org 116m№ DBOA S%(0.2M; In mesitylene

Feed S98ml/h Au=21ppm Fe -15 g/1 S b - 6дЛ H

+ - 3. 2 M others elements - ' о л

Rtff. 648m№ Au -t ppm

TT

Scrub SKril/h

HCI - 1.Э5М

Org liaml/h

Au s 0.1 Fe-nd

Strip SSffllfo

на .02s м Thiourea-0.25M

rig.2. THE AUMIDEX PROCESS WITH N,N' DI n-BUTYL OCTANAMIDE

еоисьпягпия. The results of the experimental work with TOA or DBOA as extracting agents of gold from acidic leachates,strongly support the possibility of using these extractants on an industrial scale. At laboratory scale both processes, tested with various leachates solutions obtained from different minerals gave at least a 98% gold recovery and concentration factors > 10.The tests were run with mini mixer-settlers and without .reflux of the organic phase. With an appropriate "in stage" or " in battery" reflux the concentration factor may increase of one or two orders of magnitude.

Both extractants exhibit good stability toward chemical degradation. The introduction of 1-2 g/1 of urea into the feed solution assures ar even higher stability of the extractants and prevents the chemical attack of the diluent.

The main differences between the two flowsheets lies in the scrubbing sections. In the AUMIDEX process, the necessity to scrub with HCI solutions in order to keep the Au distribution ratio

238

sufficiently high, favorably affects the removal of the coextracted metals from the organic phase.

Fe(III) at high concentrations interfere with the gold separation process. However the different stability of the chlorocomplexes in aqueous phase allows high Au/Fe separation factors.

The presence of high concentrations of Bi fIII) can also originate problems. A small amounts of EDTA in the scrub solution is effective in removing Bi (III) from the organic phase,thus reducing its interference.

The efficiency of gold reextraction from the organic phase with thiourea was excellent in both processes. The stripping kinetics was also adequate to obtain more than 98% of gold recovery in few stages.

The good hydrodynamic behavior of the phases in the mixer-settler experiments (fast transfer kinetics,no cruds at the interphase, no losses of organic phase for solubilization or entrainment) permits the organic phase recycling after a suitable solvent regeneration section. The use of mesitylene as diluent of the amide reduces the coalescence time of the phases.

RIKMMCgS

1. C.W.A.MUIR,R.D.STETT,C.J.PRIDDLE// Proc .of 3 t J l I n t e r . Symp.on Hydrometallurgy-ll2 t n AIME„ Annual Meet ing.Atlanta ,Georgia; March 6-10 (1963) .

2. J.A.EISELE,A.F.COLOMBO,G.E.MCCLELLAND / / Sep .Sc i . and Techn. 18 (12-13) 1081-1094(1983).

3. I t a l . p a t . n°48615 A /86 (Nov,3,1986). 4. I t a l . Pat.n°4P667 A /85 (Oct ,15,1985). 5* I.D.MILLER, M.B.MOOIMAN// Sep .Sc i .and Techn. 19(11-12) B95-909

(1984-5). 6. G.M.GASPARINI,G.GROSSI / / S o l v . E x t r . a n d Ion Exch.4<6). 1233-1271

(1986).

239

THE SEPARATION OF NIOBIUM AND ZIRCONIUM nv SOLVENT EXTRACTION | ]

"3 |

D.S. Flett and D.H. West, Warren Spring Laboratory, Stevenage, Herts, U.K.

As part of a project to recover valuable materials from the residues arising frotn the chlorination of titaniferous ores, Warren Spring Laboratory was commissioned to study the feasibility of using solvent extraction for zirconium and niobium separation end recovery. The feed material for this study was a crude pulp produced by HC1 dissolution of the chlorination residue followed by selective precipitation of the zirconium and niobium as sulphates. These pulps contained on a wet basis up to 101 Zr and 32 Nb; varying quantities of Ti, V, Fe, Cr as impurities; 3­52 sulphate and 0.15­0.251 chloride. The pulps were soluble in both HC1 and H2SO4. Thus solvent extraction from both chloride and sulphate solutions had to be considered.

The Chloride System From the literature [1] i*" was known that from HCl­HF mixtures niobium was readily

separated from zirconium and titanium with a secondary amine, LAl. However for the current project the introduction of HF into the flowsheet caused problems because of the need for complete HF recovery and recycle. Consequently this approach was not examined in any detail. Preliminary testwork on the extraction of zirconium and niobium from chloride solution with amines, Table, showed that high extraction could only be achieved at high chloride concentrations «ith tertiary and quaternary amines

Extraction of Zirconium and Niobium from Chloride Solution with Amines Feed Solution: Approx 13 g l ­ 1 Zr ai.d 4.9 g l _ 1 № a e C 1

" <HC1 + HgCl 2) Extractant : 30 v

/o amine in Solveseo 150; Modifier ­ leodecanol, 10 v

/o

Amine CI in Feed soln. Extract ion X Separation Factor, Zr/Nb

M Zr Nb Prinene JHT 8.6 0.0 0.0 ­

10.5 1.9 16.9 0.098 12.0 14.5 38.9 0.266

LA­2 8.6 0.0 0.0 ­10.5 8.9 16.0 0.51 12.0 33.7 34.9 0.95

Alamine 336 8.6 19.7 6.7 3.1 10.5 68.3 22.6 7.3 12.0 87.4 46.7 7.9

Aliqu«t 336 8.6 13.9 5.9 2.4 10.5 78.4 20.0 13.1 12.0 93.1 39.5 20.7

which showed moderate selectivities for zirconium which increased with increasing chloride concentration. No vanadium or chromium were coextracted but iron w o strongly coextracted. Although the results of this work looked promising HC1/C1"

240

would need to be recycled to minimise process costs. To do this Nb must first be removed and it is not clear how this can be done effectively and completely from the zirconium extraction raffinate. Furthermore, the feed liquor contains appreciable amounts of SO^™ which will build up on recycle and would have to be removed by precipitation to maintain the chloride system. Thus this approach was also abandoned.

While the literature [2­M reports favourably on the use of alkyl phosphorous acids for selective zirconium extraction, preliminary testwork with both chloride and sulphate pulp leaches shoved considerable crud formation and very poor phase breaks. Consequently this approach was not pursued.

Tr butyl phosphate on the other hand gave excellent and selective extraction of zirconium from niobium from chloride solutions but because of the reported rapid degradation of TBP to n­butyl chloride in the presence of HC1 solutions containing zirconium [S] this approach, too, was abandoned. The Sulphate System

Unlike the extraction behaviour found in the chloride вувгзщ, amines were found to extract both zirconium and niobium from low strength sulphuric acid solutions. Comparative tests on liquors produced by H2SO4 leaching of the pulps showed that Primene JHT gave the best performance, closely followed by Alanine 336. Both the secondary amine LA­2 and the quaternary amine Aliquat 336 gave inferior performances, Fig 1. Tt.e separation factor for Zr from Nb at leach acid concentrations of 2M and 3M H2SO4 was "^31 with a Zr extraction o f ™ 70X for Primene JKT. Thus this reagent was selected for further study. Distribution isotherms were developed for zirconium using both the acid treated and free base forma of the extractant. The feed liquor for this work contained 21.9

Plg»1.Extraction of zirconium from sulphuric acid solutiona of a mixed Zr/Nb feedstock with amines. Ejctractant: 30v

/o amine in solvasso 150; modi­fier, isodeoanol, 10v

/o

H.g>g. Zirconium and niobium extraction isotherms from sul­phate solution with Prlmene JUT. Bxtraotant: 30V

/o Primene JMT In eolveeao 150: modifier1 10v

/o i . .ecan^l; feed Zr 21.9; Nb 6 Л g.l"1

Г1ЙССИИИ M WUCOUS N A S I . I •"

241

MfPiNATE FEED LIQUOR

RECVCLE TO PULP LEACH

Plg.3« Suggested flowheet for the separation and recovery of zirconium and niobium

Ti, 0,3 gl" 1 Fe, 0,5 gl~ Cr. Both modes of operation gave organic phase loadings of > 8 g l

­ 1 Zr for 30 v

/o Prinene JWT in Solveseo 150 with 10v

/o isodecanol e i modifier, Pig 2. No significant difference in separation factors between the .wo systems was noted. In presence of a l*rge excess of organic phase, titanium, ir ­>n and chromium were all extracted with the zirconium. On increasing the ratio of aqu*ous to organic phase ( A

/0) t however, these metal ions could be effectively rovded out of the organic phase. A McCabe Thiele construction on Fig 2. shows effec vely 100Z Zr removed in 2 stages. Coextracted niobium could be

cfe. .­.. л.. i.v i f r o :­ m „ nmnnir phase vi»*> а ястиЬ feed containing 16,3 ((1 Zr in 2M H2SO4 at an A

/ 0 ratio of 0.1 in two crosscurrent stages.

242

Stripping of the zirconium fro* the loaded organic phase could be carried out in several way». Sod it­* carbonate (2/0 gave 95X recovery in a single stage vhili 3M KCl gav« 98Z recovery in a single stage also. In practice an ammoniacal/ammonium ct­ >nate strip would be favoured as the ammonia and carbon dioxide could be thermally released, collected and recombined for recycle thus precipitating a ZrO^ product, a small portion of which could be «dissolved in H2SO4 for scrubbing purposes.

Although little work was carried out on this aspect, niobium wao readily extracted with Primene JWT from the raffinate from zirconium extraction. From other work [6] niobium can be readily stripped from the loaded organic phaee using a NaOH/H2°2 strip. Thermal decomposition of the peroxy complex will yield an impure № 2 0 5 product Tfhich can be further purified by conventional means.

Thus a tentative flowsheet for processing the Zr/Nb pulps produced from the liquor obtained from leaching the residues from chlotination of titaniferoua ores can be envisaged as in Pig Э. From the available data about two extract stages, two or three scrub stages and two strip stages would be required for the zirconium circuit and the proposed flowsheet is considered to be technically feasible. The authors acknowledge the financial support of Tioxide (UK) Ltd for this work. References 1. Merill C.C. and Couch D.E. US. Bureau of Mines Rep, Invest. 7671. 2. Sato Г. and Nakamura X.U J. inorg. nucl. Chem. 1971. Vol 33. P.10B1. 3. Witte M.K, and Frey C.C. Eur. Pat. Appl. 0.154 445 A2 20 Feb. 1985. 4. Sella С and Bauer D. Separation Processes in Hydrometallurgy, Ellis Horwood,

Chichester, 1987. p.207. 5. Hoffat, A.J. and Thompson, R.G.//J. inorg. nucl, Chem. 1961.Vol,l6. P.365. 6. Warren Spring Laboratory, unpublished data.

243

A STUD'!' OK TKREE­EXIT EXTRACTION SEPARATION TECHNIQUE OP RARE EARTHS WITH HEH(EHP)* I 11­4

Han Li, Weixing Zhang, Zhichuan Chen, Shulan Meng, Changchun Institute of Applied Chemistry, Academia Sinlca, Changchun, China The separation process of counter current extraction is an effective

method for separating and purifying substances, and it is widely used for the separation of rare earths. However, for a given process, only two components can be separated, when it is used for multi­component systems of rare earths, more reparation ргосевэев will be needed, and the separating circle must be very long. Literature [13 reported a three­exit method in which three components can be separated simulta­neously, and the separation results of La­Pr­Nd were published. During the same time, we put forward another kind of three­exit extraction se­paration technique 121, and with this technique, the separation experi­ment in the system of (M­Tb­D7)Cl3­HCl­HEH(EHP)­Kerosene was carried out. 1. Experiments

Extractant: ammoniatsd HEH(EHP)­Kerosene Aqueous feed: mixed rare earths chloride solution, T04O7 55Й Scrubbing solution: IC1, 1.0­3.0 mol/1 The separation ехрег­ment of Gd­Tb­Dy was carried out in separatory

funnels. The rare earth concentrations were analysed by EDTA capacity method. The compositions of rare earths both in feed and in stages were analysed with VP­320 У­iay fluorescence spectrometer. The purities of the outlet products were determined by GPS­2 emission spectrometer. 2. The process of three­exit extraction separation technique

In order to improve the separation efficiency for rare earths, a three­exit extraction separation technique, based on experiments and computer simulation, has c­een developed (Pig.1).

VS1­

VLXVFi— NF MW­I

sectioN i

3= VSItVSl vLUvta

SectioN 2

vSjf I VU

Flg.1. The scheme of three­exit extraction separation NF,N ­ the feeding stage num ar and the total stage number 1,N0+1,NO ­ the first, весопо, and third exit stage numbers NW.VFa ­ the back flowing sta^e number and the volume of aqueous feed VS1.VS2 ­ the volumes of extractant VL1,VL2 ­ the volumee of scrutbing solution

244

5. Separation of Gd­Tb­Dy with the three­exit technique Stage шшЬегв: H­38, 1ГР­11, 1W­19, HO­50 М.СЧ ratios: Y?:VS1:VS2:Vb1:VL2­6.6:18.4:5.4:4.0:4.5 The separation results are shown in Table 1 and Fig.2­3

Table 1. The parities of the three outlet products

Exit (­a­Gd)X TW (Бу­lu)* 1 2 3

99.99 3.34 /

0.01 68.90 0.20

1 27.76 99.80

а щ ш ,SH

«X I» 1 \ if

fe

*> V­v л Js \

» ^ * 4 > ­ r '

Plg.3. The compositions of rare earths in organic phase

1 ­ (La­Gd)*; 2­TW, 3 ­ (4y­Lu)*

Fig.2. The coapositione of rare earths in aqueous phase

SN ­ stage number; 1 ­ (La­Gd)jS; 2 ­ To*', 3 ­ (Dy­Lu)#

4. The advantages of the three­exit technique

Prom Pig.1­3­ it is seen that stage 1­18 (ssctiu.­. 1) functions as a process of Gd­Tb separation, stage 19­38 (section 2) acts as a

process of Tb­Dy separation. Sat the three­exit process is not simple combination of the two processes. There are two main differences bet­ween the». The first is that, the scrubbing solution for section 2 includes the solution for section 1, thus the rescrubbing ratio of section 2 is larger than that of commonly used Tb­Dy separation pro­cess, and the number of stages can be reduced. The second difference is that in stage NW­1, the scrubbing solution for section 1 contains rare earthe. This results that the concentrations of rare earths in scrubb­ing section for section 1 are larger than that of general Gd­ТЪ separa­tion process so the number of stages can be further reduced.

Acknowledgement We are indebted to Chunfu He, Hongxing Ren, Yulan Tong and Xuedong

Liu for providing X­ray fluorescence analysis. References

1. Tunyu Shen, Jinghong Jin //Journal of the Chinese Rare Earth Society (Chinese). 1985­ Vol.3, H 2. P.1.

2. Han Ы , Zhichuan Chen, Weixlng Zhang//Proceedings of the First Symposium on Extraction Chemistry in China (Chinese), 1985. P. 12B, The Project Supported by National Natural Science Foundation of China

245

CXTf iACTIOf . FXTHOOS OF E V O L U T I O N , SEPARATING ATJD CONCENTRATING OF j 11­5

RARE AND DISPERSED ELEMENTS­COMPANIONS OF LEAD ANO ZINC '

G.L.Pashkuv, S.T.Takezhanou , G.P.Giganav, G.A.Yagodin, A.S.Kulenov, A.I.KhDlkin, V. U.Sergiyei.'Sky, К.Z.Kuanysheva, 8.N.Alibayev, T.Sh.Bayanzhanov, Institute of Chemistry and Chemical Technology of Siberian Branch of Academy of Science of the USSR, Krasnoyarsk and Ust­Kamenogorsk Lead­Zinc Interprice, Ust­Kamenogorsk, USSR

The use of extraction methods has resulted in effective euolu­tion of rare and dispersed elements from the complex salt solutions of lead­zinc production. Indium, thallium and cadmium obtained after hydrometallurgical treatment of lead powders mere suggested to be extracted with tributyl phosphate containing molecular iodine /1,2/. To form extracted complexes reactant­reductant of iodine, e.g. sodium sulfite, is introduced into the extraction system. Then the extrac­tion process is described by the following reaction equations:

CdSO^ + J 2 + 2L + (Na2S03 + H 2 0 ) ^ CdJ^ 2L + NajSO^ + H 2SG u; In­(504)3 + TjJ + бГ + 3(Na2S03 + h ^ O ) ^ 2InJ3>3L + 3Na2SQ4+ 3H 2S0^

T12S0^ + 3 ^ + 2L + tNa2S03 + Н 2 0 ) ^ 2ТЦЛ­J2J­L + Na 2SO u + H 2S0 4 >

where: L ­ (R0)3P(0)R3_n; R = C^Hg * C 8H 1 ?; n = 0 + 3. In organdie phase mononuclear nonhydrated complexes of nonelectro­

lytes (CdJ, L­), (lnJ,L3) and (T1J,L) are formed. Tautomeric equi­librium exists in organic phase /l/;

T1{J­J2)­L zZ T 1 J

3 ' ­ •

Extraction by 1,6 PI snlution of tributyl phosphate,containing 0,35 14 J_, in kerosene allows during 6­S cauntercurrent stages to extract in organic phase more than 99,9% of thallium, 39,5­99,9% of cadmium end 92­96% of indium. To avoid the formation of thallium precipitation the equilibrium organic phase must contain not less than 0,1 (4 of free molecular iodine. Separation coefficients of Cd­Zn, Cd­Cu are at the level of 50, and Cd­Fe(II},Cd­As(III) and Cd­Sb(lll) at the level of 10 . Separation coefficient for Tl­Zn is more than 10 .

The stripping of elements may be carried out using sulphate hyd­rogen peroxide solution. For example, stripping of cadmium nay be dpscribed by the equation:

^~ ~ CdSQ, ~~

246

.;.,­ •t._r ff vctive extraction constants {LgK) of (rieL J^)

complexes at 25° С (diluent ­ n­octane, R =* C^Hg)

L lgK

L T I L I 3 C d L 2 J 2 i n L 3 a 3

(R0) 3 PD B.7 ­1P7 3 , 9 . 1 0

4 1 , 9 . 1 03

(H0) 2 P(D)R 1 . 3 ­ 1 08 9 , 5 . 1 0 ' B , 3 . 1 0

3

(RO)P(0)R 2 1 ,7 .1DS

2 , 5 ­ 1 0b 4 , 9 . I D

4

П3РО 2 , 3 . 1 0е б . В ­ Ю

5 3 , 4 . 1 0s

Varying * 2

S 0

4 a n d H

2°2 concentration the selectivity at stripping may be achieved. Values S5­!!7" of cadmium is stripped uiith S^ solu­tion of H 20, in 15% solution of H 2S0 4, and then values 98­99,1 of in­dium and thallium are hollowed out from organic phase uiith 7:' solu­tion of H,0, in 30% solution of lUSO^.

" 2U

2 S o l u t i o n

E x t r a c t i o n o f C d . T l . I n

S ^ H 2 O 2 + T S ? H Z S O 4

S t r i p p i n g E x t r a c t „ of Cd

7^н 2о 2+зо?;н 2зп 4

^ T r i p p i n g of I n , T l

R a f F i n a t e I I n . T l ­ s t r i p ' T o d u c t Cd­stripproduct for obtaining of Cd forobtaining of In,Tl

Extraction of J"

^ Н00Н L(J,­ L) RaffinateJI

(ZnSO. solution) for obtaining of In

Technological diagram of the extraction of cadmium, thallium and indium.

Table 2.Compositions of products in Cd, Tlf and In­extraction from zinc sulphate solutions

Produc t C o n c e n t r a t i o n , o/dm •

Produc t Zn 1 Cd I Tl ' F e ( I I ) l R s C l I l ) l s b ( I l H b ~ l ­USD,.

S o l u t i o n 50­80 1­15 0,3 0 ,1 2­10 1­5 0,25 0,01 15­25

R a f f l n e t B I 60­80 0,2 0 ,01 0,005 2­10 1­5 0,25 2­5 20­35

25­40 H a f f l n a t e I I 60­80 0 . 1 0,005 0,005 2­10 1­5 0,25 0,01

20­35

25­40

C d ­ a t r i p p r o d u c t 2­3 12D­15C D,02 0,005 0 , 1 0 ,1 0 ,01 0,01 15­20

T l , I n ­ s t r i p p r o d u c t 1­2 1,0 B­20 2­6 0 , 1 0,05 0 ,01 0,01 2 8 0 ­

320

References I.Takezhanov S.T,, Giganov G.P., Paahkov G.L. et al.//Proc. of the VHUTavetmet. M.:Metallurglya. 1968. H 12. P.5­10.

2. Takezhanov S.T., Getekin L.S., Paehkov G.b.//IMa. P.15­19.

247

EXTRACTION IK HYDROHETALLURGY OP BISHUTH n_6 I Yu.M.Yukhin, A.I.Hikhailichenko, T.A.tfdalova, Yu.A.Taylov, L.I.Drozdova, Institute of Solid State Chemistry, Siberian Branch of the Academy of Sciences of the JSSR, Novosibirsk; State Research and Design Institute of Rare Metal Industry, Koscow, USSR

Hydropietallurgtc methods of bismuth extraction are baeed on the nitrate and chloride solution processing /^/. Kitric acid is the best solvent for metallic bismuth, its alloys and com­pounds. For leaching of bismuth bearing ores and concentrates one generally uses hydrochloric acid, mixtures of HC1 and CaClp or HpSCL and NaCl. Tbus obtained nitrate solutions contain (g/1): 1­10 Bi, 4­40 Fb, 0.2­1 Ag, 0.15­0.4 Си, 0.01Ч).05 Zn, 0.0005­0.002 Fe, 5­40 Ш10,, and chloride solutions contain: 1­3 Bi, 1­2 Cu, 1­2 Pb, 0.5­1 Zn, 3­6 Fe, 0.02­0.05 is, 60­80 CI.

These solutions should be treated by the liquid­liquid extraction technique to obtain better concentration and purifi­cation degree.

The comparison of cation exchange, neutral and anion ex­change extractants capacity has shown that the first ones are of practical interest while treating nitrate bismuth bearing solutions. These extractants provide bismuth concentration and purification and allow quantitative reextraction with diluted solutions of mineral acids. Bismuth distribution coefficients are reduced in series of dialkyldithiophosphoric > dialkylthio­phoephcric > thionaphtenic > alkylphosphoric > dialkylphoaphoric acids > alkylmerkaptanes > naphtenic > carbonic acids. To increase sbirply the extragent capacity it is necessary to substitute sulphur for ox/gen in it. Thus, on di­2­ethylhexylphosphoric acid (D2EUPA) transition to di­2­ethylhexyldithiophoaphoric acid (D2EHDTPA) the bismuth extraction constante increases 16 orders (lg

KD2EHPA­'

,i

''' l B K

Ti3KHTyppi*1

9­5) «ben using octane «8 » diluent. The D2EHPA solution in kerosene was taken as an extractant

to treat bismuth containing nitrate solutions. Bismuth extraction from UNO, solutions with D2EHPA may be expressed as:

Bi­* + 2.5(H 2B 2) 0 . BiB3.2EB0+3i£

D **/' F witn constant К ­ я­»— , where F accounts for Bi and

^aVo'5

nitrate ions complex formation. In technological solutions nitrate­ion concentration is 3­* И and for extraction of Bi

248

with Dg. л. 10 one should take an extragent with lg К equal to 3* For dialkylphosphoric acids if heptane acts as a diluent, Lg К is equal to 5.3 (D2EHPA), 4.5 (DBPA), 4.? (IMPA), 4.8 (DAPA), 4.9 (DHPA), 5.1 (DHPA).

The advantage of D2S1PA over extractants of ofier classes is Bi reextraction by relatively diluted НПО, solutions, as well as a possibility of its effective purification from impure metals (Pb, Cu, Ag, AZn, Ca, Mg, Ma, etc.) both during Bi extraction and organic phase washing with diluted (0.2­1.0 M) HMO, solutions CZ]. The 5­5ii IBP addition to the D2EHPA. solutions in kerosene has accelerated the process of phase separation and slightly in­creased Dg.. That is why the given mixture may be used for bismuth extraction from HNO, solutions with a higher (to 4 K) concentration.

Bi шау be reextracted from organic phase with 2.5­5.0 H HBO, or 1­5 H HC1 solutions. Since fli forms stable water phase comp­lexes with nitrate and cnloride ions, it is advisable to carry out 3i reextraction from the cation exchange reagents with a mixture of an acid and its ammonium salt 5/.

This decreases acid and alkaline reagent consumption during the reextract processing.

Industrial tests has showed the possibility of complex processing of bismuth­, lead­ and silver­containing nitrate so­lutions. In this case bismuth was extracted by the mixture of di­2­ethylhexylphosphoric acid and 5­butylphospnate, while silver was extracted by di­2­ethylb.exeldithiophosphoric acid.

Comparison of extractants of different classes has shows that neutral ertractante may be effectively used to treat chloride solutions that were obtained during the copper­bismuth concentrate leaching. In this case bismuth distribution coeffi­cients are reduced in series of; threeoctylphosphinoxide >• dioctylaulfoxide > oilaulfoxide > threebutylphosphate > hexa­cyclic > methylisobutylketone > dioctyleulphide > n­octenoic n­hexanoic > isoamilic ulcohole > dibutyl ether. TBP was used to extract bismuth from chloride solutions /4/. Dgj is high ( 10) in the 10­80 g/1 acid concentration range; Dgj reduces when HC1 content increases (Fig.2).

Bi purification can be carried out on its extraction from diluted (5­10 g/1) HC1 solutions as at these conditions Ou, Fe, 2n, Pb end Aa distribution coefficients are 10 . At Bi reext­raction by hydrochloric solutions of ammonium chloride it is possible to concentrate it and purify from Fe and Zn impurities since D„ and B„ are higher then Bg, in this case. Bi can be

249

Pig. г

Flg.1. Bi extraction from HBO, solutions with В2ШРА, TBP and their mixtures, and its reextraction by ammonium salt (4,7), E i П.; 2,0 D2BHPA, 53* TBP (1); 1,0 D2EHPA 67% TBP (2), Ю 0 * TBP (5); 1,0 D2EHPA, % TBP (4,7); 1,0 D2EUPA, Tf> TBP (5); 1

»0 D2EHPA (6)i (4­7)­dilution ­ kerosene. 3 g/1 Bi w (1­5,5,6), 24 g/1 Bi Q (4,7), 4 ­ HH4H0j (1,1 M HHOj), 7 ­ NH^Cl (0,8 И HGl)

Flg.2. Extrat ion of Fe (III) (1); Zn (2); Bi (5); As (III) (4); Cu (5); Pb (6) with TBP from solutions of HC1 (a) and Iffl Cl (70 g/1 HCl) (b);[He]l g/1

obtained ав л hydroxide on processing the organic phase with the «quiTalent volume of sodium hydrate solution (40 g/1).

The method of bismuth oxide production from Cu­Bi concent­rates (0.4­0.6Й Bi) based on Bi leaching with the hydrochloric acid at L:S «2:1 and temperature 90*3°C followed with its concent­ration and purification by TBP extraction has been industrially tested /5/.

Extraction can be used to determine Bi in different materials and technological solutions, while Bi extraction with diluted cation exchange reagent Dg, comes to 10* and this allows to avoid impurity influence on Bi determination. Bismuth is extracted quantitatively (Bgi 10 5

) by the 0,05 N D2SIDTPA solution from 0.1­8,7 И НС10#, 0,1­10 И Я^Оц, 0,1­6 M НТО,, as well as from 0,1­6 И HCl and 0,1­4 И SBr solutions, in which Bi forms stable halogenide complexes. During extraction by 02ШВТРД or its salt (indium, plumbate) Bi can be se­parated from alkaline, alkaline­earth, rare­earth elements, Al, Ga, Ti (IV), Zr, Cr (III), Mn (II), Pe (II), Fe (III), As (V), As (III),

250

Zn, Cu, Cd, XI (I), So (II), Co, Hi, Ag, Au (I), Hg (II). In this case Bl may be determined directly in the organic phase by adsorp­tion of its dialkyldithiophoephate (as J50 im 3i Q > 1 nkg/ml and ЧО0 nm Bi Q ^ 5 mkg/ml). At the D2EHPA extraction from diluted (0,1­1 M) acid solutions Bi can be quantitatively separated fro» alkaline, alkaline­earth elements, Zn, Cd, Bi, Co, Си, Fe (II), Cr (III), Hn (II), As (III), As (V), ]*b and during reextraction it may be separated from In, Ga, Tl (III), Fe (III), Sn (IV), Sb (III), which are extracted together with Bi. At D2EHPA extraction Bi may be defined in reextraets (4­6 И НН0, or ?­8 N HjSO^) by the iodate method directly in the organic phase adding dithizone, catechold violet, 1­2­pyridylazo­2­naphtol, methyltymol blue and thinorea to ?i­containing extract. Besides the extraction ­ photometric method di determination at D2EHFA and D2£BDM?A extraction is possible by the atomic absorption method, and by spraying an extract in the air­acetylene flame.

References 1. Polyvyahny I.R., Ablanov A.D., Batyrbekova S.A., Sysojev L.H.

Metallurgy of bismuth. Alma­Ata: Science, 1975­2. A.s. 452&Ч­2 (USSR). The method of bismuth extraction from nitrate

acid solutions /Yukhin Yu.M., Levin I.S., Korgov A;p. et al. Publ. in B.I., 1976, N 9. P.208.

5. A.s. 10?0192 (USSR). The method of bismuth extraction from nitrate acid solutions /Yukhin Yu.M., Kikhailichenko A.I., Tsylov Yu.A., et al. Publ. in B.I., 1984, N 4, P.101.

4. A.s. 1117529 (USbSR). The method of bismuth extraction from contair ing­chloride solutions /Yukhin Yu.M., flavtanovich M.L., Marinkina G.A. et al. Publ. in B.I., 1984­, N 57, P..74.

5. Yukhin Yu.M., Korzhenevsky L.A., Drozdova L.I. et al. //Complexs use of mineral raw materiale. 1987. N 2. P.69.

25»

COPPER ELECTROLYTE PURtFICATIOfr PROM ARSENIC Ш I ~~ CENTRIFUGAL EXTRACTOR I ——' V.V.Kalujta, G.I.Kuanetsov, A.N.Kravchenko, V.P.Lanin, G.P.MIraevsky, A.A.Puahkov, V.F.Travkin, L.I.^Shkl'yar, Mendeleev Institute of Chemical Technology, Moscow, USSR Tor the last years, low­grade ores and ores that are more difficult to

be upgraded have been involved in copper processing which affects the quality of cathode copper. One of the raost harmful impurities accumulated in the electrolyte is arsenic, and one of the most efEective methods of its removal is solvent extraction [l, 2]. The method is already used in industry [3, V . When copper is electrolytically processed, arsenic occurs in the sulphuric acid solutions in the form of arsenic acid and is only quite effectively extracted by undiluted TBP.

When using such extractant, one has to deal with a very siow emulsion separation process during the extraction and stripping stages. The utilization of centrifugal extractors in this case is superior to any other type of ex­tractors. The high capacity, low extractant volume and its minimum losses, characterize the centrifugal extractor.

The present paper considers the results of semi­industrial testing of copper electrolyte purification in centrifugal extractors. Design features and technical characteristics of a single­stage extractor, which has a rotor diameter of 125 mm (CE­125), used in the study are described previously [5]. The extractor (Fig.l), consists of a fixed body 1, in which a rotor unit 6 with drive 5 and bearing­support 4 are placed. The feed solutions are fed to the mixing chamber 9 where they are mixed with a stirrer 8. The resulting emulsion is fed to the rotor by a screw pumping device 7, where it is separat­ed under the action of centrifugal forces. The separated liquids are directed into the circular collectors 2. Then they gravitate from the collectors to the adjacent extractor. The loL.­ion of the interface in the separation cham­ber Is provided by the selnction of diameter of the overflow ring 3, installed at the heavy phase outlet. It was preliminarily determined that arsenic ex­traction with 1007. TBP and stripping with water are proceeding very rapidly: one second phase contact is sufficient to reach equilibrium. The extraction is increased with the growth of aqueous phase acidity (Fig.2) and with the decreasing of temperature (Fig.3), The conditions of purification processing were established on the basis of enlarged laboratory study with a CE­33 [6].

The semi­Industrial testing was carried out on an installation comprising six CE­125: A ­ on extraction and 2 ­ on stripping. The electrolyte after filtration is corrected with a view to reach a sulphuric acid contents up to 170 ­ 180 g/1 and is fed to the fourth stage of the extraction cascade. The ext.­actant is fed to the first stage. Stripping is carried out with water of a temperature of чО ­ 50 С which is fed to stage 6. The batchwise washing of extractant from acid products of its hydrolysis is done by a 1­2% soda solution in two CE­125. On the basis of phase entrainment, the capacity of the whole installation is approximately equal to 0.5ra /hour and is determined

252

PlK.1. Centrifugal Extractor

a.01

0,6

2.0 3.0 4,0 X SO 70«, cH?so(,K/i

Fig. 2. Effeot ">f copper electrolyte acidity on arsenlo extraction by IBP Plg.3. Effect of temperatute of arsenic extraction by TBP

253

by the operation of extraction. The phase entrainment on each stage was not highi­r than 0.037, when the ratio 0:A = 1 and the temperature of the solution equals to 30 ­ 45 С Increasing 0:A up to 3 and simultaneously decreasing the :enperature to 23 ­ 25 С leads to a rise of extractant entrainment into the electrolyte up to 0.5?.. This is the rtsull of temperature decrease, as we П as the inversion of emulsion—type in the mixing chamber [7].

The characteristics of the feed solutions and the distribution of arsenic on process flow are presented in the table. One can see that increasing the eLectrolyte acidity from 95 ­ 110 g/1 to 170 g/I, leads to an increase of ars­enic extraction from 45 ­ 56% to 70 ­ 75% (when the ratio 0:« equals 1.5). Increasing the flow ratio up to 2 leads to a higher extraction of 80%. The contents of arsenic in purified electrolyte ­ when the sulphuric acid concen­tration is 170 g/1 ­ is in avb.­?5e 2.6 g|l which corresponds to a separation factor of 3.7. Its contents in the recirculated extractant reaches 0.15 g/1 or 2.3*/. from a total contents in the feed solution. Copper and sulphuric acid contents ii the arsenic strip solution are 1.1 g/1 and 31.3 g/1 corres­ponding 3.0 and 18.47. accordingly and can be returned into the process by the additional operation of extraс tant scrubbing.

A long­time operation of this installation shows reLiable work of the centrifugal extractors and good purification of copper electrolyte from arsenic.

Characteristics of Feed Electrolyte and Distribution of Arsenic Among Flows

feed e ;ectrolyce Raffinite Scrip Recirculated E~tractant

Volumetric flou ratio 0 A

Concentration R/1 H

2S 0

4 Cu Ni As As As As extraction stripping 110,3 _ _ 11.4 6.2 4.8 0.01 1.5 1.5 95,6 ­ ­ 13.0 5.3 7.7 0.5 2.0 2.0 97,0 _ _ 11.3 7.7 8.3 0.8 1.0 2.0 136,2 33.7 15.5 8.1 4.5 3.2 0.04 1.5 1.0 169,0 38.1 16.2 9.3 г. а 5.6 0.2 1.5 1.5 170,0 33.7 14.6 8.2 1.7 7.0 0.4 2.0 2.0 172,0 41.1 20.3 9.8 3.1 7.0 0.05 1.5 1.5 170,0 ­ ­ 9.3 2.3 6.5 0.1 1.5 1.5 173,0 34.3 ­ 9.0 2.2 6.6 0.2 1.5 1.5

Note: Th e temperature oi all Clows Is 25 ­33 C, except arsenic strip vh ich Is 37 ­ 46° С

References 1. Giganov G.P., Travkin V.F., Anikina H.P.// Tsvetnye metally, 1978, N.8, p.27­29. 2. Giganov G.P., Travkin V.F. et al.//Bull Tsvetnye metally, 1978, N.3, p.21­24. 3. Shepper k.ft Papers presented at the Extraction Metallurgy '85 Symposium

organized by the Inst, of Mining and Metallurgy London, 9­12 September 1985, p.167­188.

4. 0'Kane J., Middlin В., Royson D.// ibid. p.951­965 5. Kuznetsov G.I., Pushkov A.A. , Belyakov S.M.//Atomna ja Energija 1986, 6J_,1, p.23­27. 6. Pushkov A.A., Travkin V.F. et a1.// Trudy Moscow Mendeleev Institute of Chemical

Technology, 1986, V.143, p.38. 7. Kuznetsov G.I., Pushkov A.A. et al.//Trudy VI Symposium SEV "Issledovanija in

oblasti pererabotlci obluchennogo topliva i obess&razhivanija radioactivnyh othodov", Praga, CKAE, 1985. V.3, p.55­70.

254

11­е SULPHIDE STRIPPING IN SOLVENT EXTRACTION K. Quon and P.A. Ois t i n , Department of Mining and Metal lurgical Engineering, McGill Universi ty, Montreal, Canada Summary. I t is possible in some solvent extract ion systems to s t r i p loaded organic

with hydrogen sulphide giving a sulphide prec ip i ta te and regenerated extraccant. Situations in which th is technique could be considered are b r i e f l y discussed. Da.a are presented showing the response of several chelates of KELEX ICO to sulphide str ipp ing using hydrogen sulphide at elevated temperatures.

Int roduct ion. Many solvent extract ion c i r cu i t s incorporate acid st r ipp ing of loaded organic followed by metal electrowinning. Al te rna t ive ly , i t i s sometimes possible to precip i tate metal sulphide powder from loaded organic using hydrogen sulphide [ l ] , or metal powder d i rec t l y using hydrogen [ 2 ,3 ,4 ] . Examples of thp re­

levant reactions f o r metal M loaded in to extractant HL are: ML2 + H2S ­v MS + 2HL, ( 1 )

M L

2 + H

2 * M + 2 H L

­ ^ The circumstances i n which reactions 1 and 2 might be technical ly and economically feasible involve many fac tors . The st r ipp ing reaction must simultaneously produce powder and regenerated extractant at acceptable rates and y ie l ds . Also* the organic structure must be stable under the st r ipp ing condit ions, while the powder product should be readi ly f i l t e r a b l e .

Assuming the above c r i t e r i a are s a t i s f i e d , sulphide str ipp ing i n par t icu lar could be a viable option when the loaded organic contains small concentrations of a high value metal with a redox potent ial at which hydrogen reduction i s not possible. Dissolution of the sulphide product in a small volume of acid would give higher metal concentrations fo r subsequent pur i f i ca t ion and electrowinning than obtainable by acid s t r ipp ing . In add i t ion , some metal separations may be possible through select ive prec ip i ta t ion as sulphide.

The work described below concerns the sulphide str ipp ing of gal l ium, aluminum and sodium loaded separately into an alkylated 8­hydroxyquinoline (KELEX 100 ­ Schering Aktiengesel lschaft) . These systems were chosen because, f i r s t l y , KELEX 100 has good thermal s t a b i l i t y at the elevated temperatures needed fo r reactions 1 and 2 [ 5 ] . Also, th is extractant loads gallium wi th high recovery from Bayer solutions contain­

ing small amounts of gallium but wi th a large excess of sodium and aluminum [ б ] . However, when contacted with Bayer so lu t ion , co­extraction of these l a t t e r two metals uses most of the extractant 's loading capacity. Thus the objectives of the present work were to determine i f gall ium, aluminum and sodium are sulphide stripped using hydrogen sulphide, and i f so, whether re la t ive reaction rates indicate that gallium could be select ive ly recovered w: n loaded from gal l ium, aluminum, sodium mixtures.

Experimental. Unloaded organic phase comprised (vo l . * ) 8.5 KELEX 100, 10.0 decanol dissolved in a low vapour pressure kerosene of aromatic content below ]%. Gallium or aluminum or sodium were loaded frs.n t i t r a t e solutions under conditions such that metal concentrations were simi lar to those obtained in the organic phase when loaded from a typ ica l Bayer solut ion af ter alumina prec ip i ta t ion (about 0.20 g . l " Ga, 1.5 g . Г A l , 2.0 g . l " Na)[6] . Solutions were analysed using atomic absorption spectrophotometry.

255

Sulphide s t r ipp ing tests were carried out using a standard 2 1. capacity stainless steel Parr autoclave. After f lushing a i r from the reactor using ni t rogen, loaded organic, contained in a glass l i n e r , was heated to the reaction temperature, then hydrogen sulphide was admitted be'ow the solvent level at a predetermined pressure. During sulphide s t r i pp ing , the organic was s t i r r e d , while both temperature and pressure were held constant. Analyt ical samples of the organic phase were withdrawn per iod ica l ly during react ion, at the end of which the reactor was cooled under hydro­gen sulphide. Af ter f i l t r a t i o n , the so l id product was retained for fur ther examina­t i o n , while the organic was recycled.

Results and Discussion. As shown in Figures 1 and 2, gallium can be sulphide stripped leaving residual solutions containing as low as Z pom gal l ium. The fastest reaction rate measured was at 170°C wi th 0.62 MPa hydrogen sulphide, essent ia l ly complete gallium recovery being obtained af ter 20 minutes. An important requirement of thp system is a b i l i t y to recycle sulphide stripped organic. Figure 1 includes data showing tha t , at least for reaction at 145°C with 0.62 MPa hydrogen sulphide, gallium loading levels and subsequent s t r ipp ing kinet ics are reproducible through three load-st r ip cycles.

g i g » 1 . Effect of temperature on sulphide s t r ipp ing of gallium with 0.62 MPa H S

F i g . 2 . _ Effect of hydrogen sulphide pressure on sulphide str ipping of gallium at 145°C

The conditions used to sulphide s t r i p gall ium also cause the equivalent reaction fo r aluminum, as seen in Figures 3 and 4. At dn i n i t i a l aluminum concentration of 2,1 g . l , 50% to 50% of th is amount precipi tates when about 2 ppm gallium is reached in the gaTium/KFLEX 100 system shown in Figures 1 and 2. However, in comparing gallium and aluminum st r ipp ing kinet ics using the present data, i t should be noted that free KELEX concentrations d i f f e r widely. When 8.5 vol.% KELEX is loaded wi th 0,22 g . l gallium or 2.10 g . l " aluminun, about 4% or 87% respectively of the load­ing capacity is used. Since the presence of unloaded extractant may suppress s t r i p ­ping ra tes, gallium st r ipp ing in par t icu lar may be faster when 0.22 g . l - 1 gallium and 2.1 g . l aluminum co-exist i n the same solvent,

256

•c

2 0

^

* 120 О 14% 0 170

1 5

Э ^ « s ^

1 0 \

0 5

\ о

\ 0

• 0 34 • a 0 62 ?\f о 097

fig.3. Effect of temperature on sulphide stripping of aluminum

Fig.4. Effect of hydrogen sulohide pressure on sulphide stripping of aluminum at 145 с

The behaviour of sodium at various temperatures and hydrogen sulphide pressures is shown in Figures 5 snd 6. About 80S to 85% o." the sodium precipitates under con­ditions which produce 2 ppm residual gallium, as seen in Figures 1 and 2. Based on the present results, it is therefore probable that gallium could be sulphide stripped at good yield from mixed gallium, aluminum and sodium containing extractant, and with partial selectivity over aluminum providing the reactions are immediately terminated at roughly the 2 ppm gallium level. However, most of the loaded sodium is expected to precipitate at high gallium recoveries. The sulphide stripping reactions are assumed to be:

2GaL 3 + 3H 2S ' G a 2 S 3 + 6HL

2A1L 3 + 3H ?S ­ A l j S 3 + 6HL

2NaL + H 2S * Na 2S + 2HL

(3) 'A-'.

(5)

Here, GaU, AIL, and NaL are the relevant metal chelates of KELEX (HL). The precipi tates formed as easi ly f i l t e r a b l e free powder with no attachment to

internal reactor surfaces. However, gelatinous products appeared when the reactions were carried out at above 170°C. X­ray d i f f r ac t i on analyses showed the s o l i d : to contain hydrated thiosulphates, these presumably being formed by sulphide oxidation in moist a i r fol lowing removal from the reactor. Powders produced in the gall ium/ KELEX 100/hydrogen sulphide system contained about 45 weight % gal l ium, which com­

pares with 59 weight * fo r stoichiometric Ga 2Sj. Al l the precipi tates were readi ly soluble in warm, d i lu te hydrochloric or sulphuric acids.

257

2 0

\ \ \

X • 120 a 145 0 170

l . i

\N

1 0 V о.ь 1

ч

— a — —

—* — o ^ — 0^—

F i g . 5 . Effect of temperature on sulphide str ipp ing of sodium

F i f e . 6 . Effect of hydrogen sulphide pressure On sulphide str ipp ing of sodium at 145°C

Conclusions. This work demonstrates a potent ial appl icat ion of sulphide st r ipp ing in which.a high value metal loaded at low concentration can be recovered as an acid soluble concentrate. Spec i f i ca l l y , gall ium can be sulphide stripped at high recovery from KELEX 100 usinq hydrogen sulphide at^elevated temperature and pressure. Tests with organic containing ei ther gall ium or aluminum or sodium suggest t ha t , i f loaded from Bayer so lu t ion , gall ium could be sulphide stripped with par t ia l se lec t i v i t y over aluminum, but nearly a l l the sodium would coprecipi tate. Further experiments using Bayer solut ion would determine i f the expected str ipp ing behaviour occurs. The resistance of the organic to degradation over a large number of recycles i s also un­

known. In addi t ion, the poss ib i l i t y of replacing tox ic hydrogen sulphide with an aqueous solut ion of sodium hydrogen sulphide or sodium sulphide should be studied.

References 1. i ler igold C,R. and Sudderth R.B.//Proc. In tern. Symp. on Hydrometallurgy. Chicago,

1973. P.559. 2. Burkin A.R. and Burgess J.E.A./ /Proc. F i rs t Annual Meeting of Canadian Hydro­

meta l lurg is ts , Montreal, 1971. P.51. 3. Van der Zeeuw A.J. and Gandon L.//Proc. Tenth In tern. Mineral Processing Congr.

London, 1973. P.1067. k. Demopoulos G.P. and Dist in P.A. / /J . Metals. 1985. Vol .11, N 7. P.46. 5. Demopoulos G.P. and Dist in P.A./ /J . Chem. Tech. Biotechnol. 1983. Vol.33A, P.249. 6. Leveque A. and Helgorsky J. / /Proc. In tern. Solvent Extraction Conf. (ISEC 77).

Toronto, 1977. Vol .2. P.439.

258

11­9 RECOVERY OF GALLIUH ЛЯ) VANADIUH FROM FLOE DUST IN THE ALUONIUfl INDUSTHY A FEASIBILITY STDDT O.B. Michelsen, Institute for Energy Technology. P.O. Воя 'Ю, N­2007 Kjeller. Norway

1. INTRODUCTION. Aluminium plants produce considerable quantities of flue dust during the process of electrolysis ­ for a comparatively large plant in Norway If may amount to sore than • thousand metric tons a year. The dust repre­sents both a storage and a waste problem. The sain constituents of the dust are common ill fnti such as aluminium, sodium, iron, sulphur, fluorine and carbon, the latter one accounting /or about one third ot the total «ass. In addition to these elements others and sore rare ones are also present, notably «allium and vanadium which seem to accumulate, possibly through a sublimation process, in the dust. The amounts вшу vary, and in the case of gallium figures in the range of 0.1 to 1% (Ga 0 ) have been quoted. Vanadiua as a rule seems to be somewhat шоге abundant. Table 1 gives the result of the analytical examination of the sample Investigated in this work.

Introductory work [1] on this sample had as its primary goal to recover Al

Table 1 Composition of flue dust sample. main elements

Elenent j Al Pe № F s /c V Ga Concn., % 12.7 ».o ДО.8 16. <l l».3 ­33 0.36 0.13

and F, preferably in the form of cryolite (Na A1F,} or chiolite {Na Al F , ) and pure enough to be recirculated. Hydrochloric acid was thought to be the most favourable leaching agent. It is known that GaCl is easily extracted by a number of organic extractants, so the resulting solution was expected to be a suitable medium also for separating off Ga.

This initial investigation was not successful, neither with respect to the recovery of Al and F in a usable form, nor with regard to Ga extraction. The HC1 solutions of Ga were shaken out with TBP (tributyl phosphate) of various con­centrations but with little effect, and the expected separation from Fe did not take place. Following these negative results the project vas shelved for a while. However, as interest in Ga for laser and other applications grew, it was decided to take up work again, but now with the principal objective of recover­ing Ga. V was Included later, more or less by accident.

Д. LEACHIWQ OF FLY DUST. 2,1. Leaching with water or HC1. Initial atteapts Ш to leach the dust with water at temperatures up 100 С and contact times up to 2k hours showed that about 25% of the dust went into solution in all experiments. At 100 С the dissolved material was almost pure Na SO , while at 20°C the Na SO was contaminated by Al and F. No Ga dissolved.

The hydrochloric acid leaching» were done in two ways. One way was to treat the duet with 6 N HC1 (300 ml acid to 100 g of dust) for 2 hours at 110°C. The yield of dissolved gallium was about 70%. The other way was to calcine the dust at 700°C and leach the residue with efther 3,7 N HC1 (500 ml acid to 100 g

259

residue) or with 6 N HC1 (300 ml acid to 100 g residue), both at boiling tem­perature for 2 hours. Leaching with 3­7 N HC1 gave a yield of only 52% Ga. while lea chin* with 6 N HC1 produced «bout 90% dissolved Ga.

The reason for the difficult extraction of Ga was thought to be associated with the presence of fluoride. Also chloride ions fore complexes with Ca. When the investigation was resumed it was first looked at Ga leaching by other mineral acids with less tendency to fore complexes with this element.

2.2. Leaching with HNO or H SO . 10 g samples were leached with 50 ml acid of desired concentration. The slurry was stirred for 4 hours at the temperature chosen. It was filtered when cold, washed with water, and the filtrate diluted to 100 ml before rnalyeis. Only Ga, Fe. and acidity were determined. The results are shown in Table 2.

Table 2. Result of leaching experiments with HNO and H SO^

Sample No.

Acid Initial concn., N

Te«p.. Final acid concn.,N

Undissolved matter, %

Concn. , mg/1 X dissolved Sample No.

Acid Initial concn., N

Te«p.. Final acid concn.,N

Undissolved matter, %

Ga Fe Ga Fe

1 2 3 4 5 6 7 8 9

HNO HNO HNO HNO HNO HNO KN0 HNO H

2

S 0

*

1.4 1.4 2.9 2.9 7­2 7.2 7.2 14.4 3.6

22 100 55 100 22 55 100 55 70

0.45 0.025 О.65 0.40 2.9 2.4 2.4 5 6 0.75

66.4 48.7 60.9 '37.7 753 61.9 40.4 79­9 48.7

24 45 86 110 42 90 111 74 104

600 1000 2200 2500 1000 2300 3000 ' 2000 36ОО

18.5 34.6 66.2 81.6 32.3 69.2 85.1 56.9 80.0

15.0 25.0 55.0 62.5 25.0 57.5 75.0 50.0 91.3

To find optimal conditions for leaching Ga more work remains to be done, but the experiments above indicate that Ga yields well above 80% can readily be obtained with the mineral acids tried in this study.

3. EXTRACTION EXPERIMENTS. 3.1 Materials and conditions. For these experi­ments 1000 g portions of dust were leached with HNO or H SO, of initial eoncen­

3 г и

trations 2­9 N end 3.6 N respectively. The leachates were not diluted with water, but the acidity was adjusted in each case by addition of NH .

HDEHP (hydrogen di­(2­ethylhexyl) phosphoric acid) extractions were carried out with 1 M solutions. The amine extractants Amberlite LA­1 (secondary amine from Rohm t Heas) and TI0A (tri­isooctyl amine) were 0.2 and 0.5 H respectively. All agents were diluted with Solvesso 150.

Element concentrations were determined by ICP spectroscopy and only in the aqueous phase.

3.2 Extraction of iron. Iron was thought to be the most difficult element to get rid of. HDEHP is known as a good extгасtant for Fe *, and extraction was tried from both hWO and H SO leachates. Equal volumes of aqueous and organic solutions were shaken for 5 rain, when samples were drawn for analysis. This was

260

repeated also after continued shaldnc for 15. 30. 60 «In. In addition to Ft Mid acicity the Initial and final solutions were analysed for Al. Ca. and V. The reaulta are shown In Tables 3 and 4.

Tadle 3. Extraction of Fe froa a HNO leachate

Shaking tlae, aln

Concn In aaueous Dhase Shaking tlae, aln

H'. N

Fe. 4/1

Al. as/I

Ga. se/1

V. a»/l

0 5 15 30 60

0.27 0.60 0.60 0.60 0.63

3230 2240 1540 1150 640

12400

12300

110

110

260

260

Ga and V are not extracted under these conditions and very little, if any. Al. Fe is known to extract slowly in a H SO, aediua, but it seems that even the HNO

2 f t 3

systea s t i l l i s i n a s t a t e of non­equilibriua af ter 60 Bin. Both aqueous phases were adjusted back to the i n i t i a l a c i d i t y and « e x t r a c t e d with fresh portions of HDEHP for another 6o a i n . The res t concentration of Fe was then found to be 10 ag/1 in the HNO system and 230 ag/1 in HgSO , which ind icates that Fe шву be d i f f i c u l t to remove completely in the l a t t e r sys tea .

Table 4 . Extraction ot Fe froa a H SO, leachate

2 4

Shaking tlae, min

Concn In aaueous phase Shaking tlae, min

H". N

Fe. Al, a*71

Ga, •8/1

V, ИК/1

0 5 15 30 60

0.24 0.36 0.44 0.51 0.57

5400 3600 2500 1500 620

17000

16900

175

175

390

390

3.3. Extraction of Ga and V. Normally HDEHP will, also extract Ga when con­ditions are suitable. Extraction experiments were carried out froa both HNO and H SO media of various acidities. Contact time was 5 «in. and the phase volume ratio 1:1. Tables 5 and 6 show the results of these experiments.

In experiments 3 and 4 of Table 5 and experiment 6 of Table б the acidity of the solutions was adjusted so as to bring the pH as close to the precipitation point as possible. It looked as If this point was higher in the SO^ systea than in the other. Not surprisingly, the results Indicate that lower acidity promotes extraction of both Ga and V. Both systems also extract Fe, but not to the extent that might be expected. Also a little Al is extracted at the lower acidities.

261

Table g. Er.trлеи on with HDEHP Tram UNO solutions

Table 6. Extraction with HDEHP froa H SO solutions

Expt. No.

State Conco. in aqueous Dhase Expt. No.

State H'. Fe, Al, Ga, v. N 4/1 •*71 ­C/l •*7l

1 I 0.67 2200 93 A 0.72 1600 93

2 I 0­13 1540 90 A 0.18 300 63

3 1 0.01 2500 32 A 0.08 200 48

4 I 0.05 <5 11900 105 200 A 0.17 <5 10500 38 95

I = II iltial A x After equilibration

Expt. Mo.

State Conco. in aaueous phase Expt. Mo.

State H'. N

Fe, •в/1

Al. •S/l

Ga. •g/1

v. •g/1

1 I 0.17 2250 88 A 0.37 1200 72

2 I 0.20 2800 96 A 0.24 1500 67

3 I pH= 3 0

2800 96

A 0.11 600 46 4 I

2.0 1170 70

A 0.13 300 12 5 I pH=

2.0 <10 9400 68

A 0.17 <10 8800 26 6 1 pH=

4.0 920 17500 185 370

A pH= 1.75

700 16500 65 55

I = InifcU il, A = After equilibration

3.4. Back extraction. Back extrac­tion was tried with H SO, and HC1. Low

2 4 acidity (0.05 N) strip solutions proved very inefficient for the elements under study. 3 N solutions, however, gave good recoveries of Ga and V, while Fe was only partly stripped. Al was stripped practically quantitatively by H SO , unexpectedly poorly by HC1, In Table 7 are the results of the back extraction experiment with 3 N HC1 since this is the important one with a view to the subsequent separation of Ga and V. The organic phase was stripped twice, each time with an equal volume of strip solution for 5 min. Initial concentrations (calc.) in the organic phase were: Fe: 320 mg/1. Al: 1000 fflff/1. Ga;­90 4ff/l. V; 310 mg/1.

Table 7­ Back extraction free HDEHP with HC1 3­5­ Separation of Ga and V. It is known [2] that Araberlite LA­1 extracts Ga from HC1 solutions. V can also be ex­tracted, but ex­traction starts at a higher acidity.

Strip solution No. 1 (Table 7) was shaken for 5 min. with an equal volume of the Aiaberlite LA­1 extractant. Afterwards the following concentrations were found in the aqueous phase: H*: 3­05 N. Fe: 40 a*/l. Al.loO mg/l. Ga <2.5 mg/1. V: 300 mg/1.

Strip No.

Concn. in aqueous Dhase Total amount strioped, % Strip No. H', Fe, Al, Ga, V,

N Bg/1 mg/1 Bg/1 nig/1 Fe Al Ga V

1 2

3.10 180 160 100 300 3.10 35 210 <10 10

56 67

16 37

100 97 100 100

262

Practically all the Ga ha* been extracted but no measurable amount of V (or Al). Also a considerable portion of the iron extracted, but that was expected since Fe forms negative complexes with the chloride ions.

3­6. Separation of V fro» Fe and Al. The best way of achieving this separa­tion was thought to be extraction of pentavalent V by TIOA free a H S0fc solu­

tion. For this purpose V had to be precipitated fro* its chloride medium­ This proved possible by oxidising the V from its tetravalent state, and adding NH, until precipitation occurred. V would then cose out with the hydroxides of Fe and Al 3

'. It was found that the degree of V coprecipltation was dependent on the concentrations of these ions. An experiment was carried out in which mixtures of V of constant initial concentration and Fe of varying concentration were pre­cipitated. The result is shown in Table 8. Initial V concentration was 2700 mg/1, and the tabulated Fe concentrations are those that would be if precipitation did not occur, V concentrations are those measured after precipitation.

To the aqueous phase remain­ Table B. Coprecipitation of V with ing after extraction of Ga {35} Fe

3

* by Ю Ц was added NaClO to oxidise V. Neutralisation of the solution with NH then produced a brown precipitate which was filtered off, washed with water and dis­solved in a minimum of 1 N H SO •

The solution was analysed with the following result: H*: 0.2 N. Fe: 190 mg/1. Al: 770 mg/1. V: 1375 •»/!• The solution was shaken for 5 min, • ith an equal volume of the TIOA extractant. Analysis of the resulting aqueous phase then showed: Fe: 185 mg/1. Al: 780 mg/1. V: 200 mg/1 Only small amounts of Fe and Al have been extracted while over 85% of the V has been transferred to the organic phase in this single equilibration. Amberlite IA­1 was also tried for this step, but proved far less effective.

4. PREPARATION OF CONCENTRATES OF Ga Ш ) V, 4.1 Ga. The H SO based leachate — — — 2 и

used for the experiment had the following composition with respect to the key elements: H*: 2 .1 N. Fe: 7­6 в / l ­ Al: 25 g/1­ Qa: 200 mg/1. V: 570 mg/1 with leaching

recoveries of 80 and 82% for 0a and V respect ive ly .

500 ml leachate was adjusted to an acid concentration of 0.3 N by adding

cone. NH . Xt was «haken out twice, each time with 500 ml HDEHP for 60 min.

Between the shake­outs ac id i ty was adjusted back to 0­3 N with NH .

The aqueous phase a f t e r t h i s operation was neutral ised with NH to the point

where permanent turbidity prevailed. pH was then 3 .7 . The so lut ion was e q u i l i ­

brated with 150 ml HDEHP by shaking for 5 min. pH was again adjusted, th i s time

i t ended up a t pH 4.0 . Extraction was repeated with another 150 ml portion of HDEHP

fhe combined organic phases (300 ml) were stripped twice, each time with 75

ml 3 N HC1 by shaking for 5 min. The etr ­p so lutut ions were also combined,

263

Element Concentrations, eg/l

Fe V

0 266 523 2700 2120 1870

1246 980

18 Ю 600

2314

325

ascorbic «cid was added 1л « l ight excess of the Fe present, «nd Ga extracted by shaking for 5 s i n . irfth 75 «1 Asoerl i te IA­1. The organic p b u t was washed with 50 . 1 3 N HC1 containing a l i t t l e (ca. 20 s c ) of ascorbic ac id . Final ly Ca was (tripped by shaking Гот 5 aitn. with 25 «1 uater. The atr ip so lut ion was analysed with the r e s u l t l i s t e d in Table 9­

Table 9. Analycia of s t r i p so lu ­

t ion containing; purified Ga

Element Fe Al Ga V

Concn., s«/l

25 <1 3480 <1

The purity of Ga with respect to the other element* in the table 1» 99­3%­ Recovery of GR through the extract ior. s t eps was 87%. and thus the overal l y i e l d including leaching was about 69­5%­

4 .2 V. To the aqueous phase remaining a f t e r extrat ion of Ga {4 .1) was added s u f f i c i e n t NaCIO to ox id i se the V and

also the Fe. Cone. NH was then added, u n t i l nothing ю г е prec ip i ta ted . The prec ip i ta te was f i l t e r e d of f and washed with water, and then dissolved in 1 N H S O . The so lut ion was di luted to 100 «1 (Measured ac id i ty 0.5 N) and shaken once for 5 s i n . with 50 sd 0.5 И TIOA. The organic phase was washed once with 25 ml 0.5 N H 2 S0 t ­

Suf f i c ient ascorbic acid to reduce a l l V5

* to V*4 was added to 50 ml 0.5 N

HC1 and the so lut ion used to s t r i p the vanadiub from the TIOA phase by contact­

ing the so lut ion for 5 min. Analysis of the resu l t ing aqueous phase gave the r e s u l t l i s t e d in Table 10.

Table 10. Analysis of s t r i p so lu ­

t ion containing purif ied V

Element Fe Al Ga V

Concn., •K/l

115 40 <1 4B40

Calculated in analogy with Ga the purity of V i s 96.93» and recovery through the separatirtg s t eps reckoned from the leach­

ate i s Ы>%, giving an overal l y i e l d of 75.5%.

5. COIMEWTS AND CONCLUSIONS­ The work presented above does not pretend to have found optimum conditions for recovering

Ga and V from aluminium f lue dust. I t shows, however, that both elements can be leached in r e l a t i v e l y good y i e l d s from the dust, and a l so that they can be separated out in good purity and acceptable y i e l d s through a s e r i e s of simple shake­out extract ion operations. The r e s u l t s can probably be considerably improved by applying multistage extract ion and scrub pr inc ip les such as i s nor­

mally done in industr ia l processes .

The economic implications of a process bawjd on the separation scheme f o l ­

lowed in t h i s work have not been evaluated. References

1. Gjelsvik N.. Dugan M.//IFE Report IFE/KFi/F­B2/0l4, September 1982 <conf.) . 2. Ishimori Т. , Nakakmura

I n s t i t u t e , 1963. E. JAERI 1047, .'арап Atomic Energy Research

264

MECHANISM OF SYNERGISTIC EXTRACTION OF PHOSPHORUS FROH SODIUM ' MOLYBDA^E SOLUTION BY PRIMARY AMINE AND TRIBUTYL PHOSPHATE I

1 <

~1 0

ZHAO YOUCA1, Yu SHVQ7V AND CHEN JIAYONG

Institute of Chemical Metallurgy, Academia Sinica, Beijing, China

The removal of impurities such as P, As and Si from alkaline leach­ing sciutions is an important step in hydro­metallurgy of W and Mo. The traditional method used in industry is by precipitation with the addi­tion of magnesium salt. This process has many problems such as low re­covery, disposal of precipitate, and pollution arising from the storage of precipitate.

Solvent extraction can be considered as a better method for impu­rity removal in comparison with precipitation method if an effective solvent system can be developed. Work on the extraction of P, As, Si and their heteropolyacid formed with W and Mod­2) hav*: been report­ed, which generally were carried out under acidic condition. It has been found that separation of W from Mo and Re(VII) from Mo can td carried­out satisfactorily in weakly alkaline solutions by synergistic extraction with primary amine and TBP{3­4). In the present paper, the mechanism of extraction of P from weakly alkaline molybdate solution, by primary amine and TBP was investigated.

EXPERIMENTAL PART

An electric shaker was used to mix aqueous and organic solutions. Primary amine(R­CHNH­ or simplified as RNH­ in the following expres­sions) with a total carbon number of about 21 was supplied by the Institute of Organic Chemistry, Academia Sinica, Shanghai, China. The other chemicals were of analytical grade purity. Mo and r' were analyz­ed by thiocyanate and molybdophosphate colorimetry respectively.

RESULTS AND DISCUSSIONS

The extraction of phosphorus . The extraction of P is a function of pH value of aqueous solutions, concentrations of Ко and P in the aqueous phase, temperature, phase ratio, and both type and concentration of amine and neutral donor extractants used. Experimental results show that synergism of amines with TBP decreases in the order of primary, secondary and tertiary amines. P can not be extracted without the presence of Mo in the aqueous solution. It can be proposed that P is extracted into the organic phase as molybdophosphoric acid, it was found that a mixture of 1­2% RNH, and 10­20% TBP(v/v) with n­hep­tane as diluent can be used to separate P from sodium molybdate

265

solution with equilibrium pH 6.4­6.6 very effectively when the initial concentration of P and Mo were in the range of about IOOmg/1 and higher than 40g/l respectively at 25°C or higher. The separation factor of P to Mo can be higher than IOOO and one stage extraction ;­f P can be better than 99% under above conditions.

Determination of the extracted species The extraction equilibrium can be written as

mH, t ( a ) т „ 1 Я ( а ) . nMo ( a )«cTBP ( o )+yRNH 2 *HmPMon.xTBP.yRNH2(o), (1) where some of reactants and products are represented in atomic form only because their exact molecular formulaes are unknown.

From equation (1 ) we can get LogK=LogD + mpH ­ nLog(Mo) ­ xLog(TBP) ­ yLoglRNH,), (2) where К is extraction equilibrium constant and D the distribution ratio

P of P. From the slopes of Fig. 1­4, n,m, у and x are estimated as 2. 3, 2 and 1, respectively and the equilibrium constant was calculated as 2.08x10 . It should be pointed out that the ratio of Mo/P in the organic phase is a function of that in the aqueous phase. The ratio of Mo/P in organic phase is always about ? when that in aqueous phase is less than about 6 which was used in the experiments. This is in agre­ement with the slope of Fig.1. The primary extraction reaction may be expressed as 3H

T K,PO: +2HMoo: 2 4 4

+ TBP

3 5 7 10Log(Ko)

Fig.1»Determination of n.

Initial organic phase: 1%RNH 2 + +15*TBP(v/v)

63 65 10pH

.2* Determination

67

Fii of m Initial organic phase:

1 1 7 3 ­10Log(TBP)

Fig.4. Determination of x.ENH2 1%(30.43mM) P ImM Mo 5.5mM pH 6.4­6.7

20 16 12 8 ­10Log(HHH2)

Fiq.3« Determination of У TBP 556 mM P ImM,Mo 6mM pH 6.4­6.8

All the experiments were carried out at room temperature and the mixing time was 10 minutest the equilibrium time of the extraction was about 2 minutes).

The infrared absorption band of stretching vibration of the ­NH group in a primary amine has been found to be much broader and the wave number of P=0 group in TBP has been shifted from 1275 to 1260 cm" after loaded with mo­lybdophosphoric acid as shown in Fig,5.

266

The liquid chromatography absorption peak of '.he extracted species has been found to be broadened after extraction which indicates that the extracted species consists of two or more substances whose polarities are very close but less than those of primary amine and TBP according to the decreasing order of polarity from right to left in the liquid chro­matography spectra as shown in Fig...

Dependency of molar conductivity on the concentration of the extracted species. The shape of line CA in Fig.7 suggested that the extracted species was weakly dissociated in a mixture of RNH?, TBP and n-heptane. When n-heptane was evaporated, more extracted species was dissociated and the conductivity increased rapidly because of higher polarity of RNH2 and TBP as solvent(see line AB in Fig.7).

5000 3000 1400- 1000 cm Fig.5« Infrared spectra of the ex­tracted species.Solid Line—1%RNH2 + 15%TBP in n-heptane; Dotted line— above organic solution loaded with P and Mo

Ely.. 6 * Liquid chromatography of the extracted species. Solid line--the spectra of organic solution with 1%RNH2+l5%TBP(v/v) in heptane; Dotted line--above organic solution loaded with P and Mo

Discussion of mechanism of .extractio'i and the structure of flxhrart-od species. The synergism of amines with TBP towards the extraction

of P from sodium molybdate solution decreases in the order of primary, secondary and ter*--..• amines. It can be considered that the number of active hydrogen atoms in amines plays an important role in the syner­gism, it can be suggested that the hydrogen bonds were formed between TBP, active hydrogens of primary amine and molybdophosphoric acid. This is in agreement with the previous work<3-4).

The extracted species dissociated in the organic phase may come from the direct dissociation of neutral molecules or the occurrence of ionic association. It ia resonable to propose that the occurrence of ionic asaioiation contributes little to the formation of extracted species.

267

Зо the mechanism of extraction le predominantly by solvation extraction.

О О О — Н­НЧН ' II II

Н­0­Ио­О­Р­О­Мо­О­Н I I RHH­H­0

n

9"4

I b

0­Р­0СлН, 4n

9

,Fiq^g.Proposed structure of the extracted species

100'Ca

Fig.7. Dependency of molar conducti­vity of the extracted species on its concentration C; The concentration of molybdo­phosphoric acid F: Molar conductivity of molybdo­phosphoric acid Temperature: 13°C

The structure of the extracted species can be expressed as shown in Fig.8 (the most probable structure).

CONCLUSTIONS

The extraction of P from weakly alkaline sodium molybdate solutions can be carried out very effectively with a mixture of primary amine and TBP as extractants. Phosphorus is most probably extracted into the or­ganic phase as 2­molybdaphosphoric acid. The mechanism of synergistic extraction has been studied. Hydrogen bonds are considered to be formed between TBP, primary amine and molybdophosphoric acid during extraction. The composition of the extracted species has been determine and its structure suggested.

References

1, Tatsuya Sekine and Yuko Hasegawa. Solvent Extraction Chemistry: Fundamentals and Apolications, New York, 1977.

2, V.L.Lakshmanan and B.C.Haider//J. Indian Chem. Soc. 1970.47J 3 ),231 3, Yu Shugiu and Chen Jiayong//Hydronuitallurgy, 1985t1_4 .115. 4, Yu Shuqiu and Chen Jiayong//ISEC'86, Preprints. Vol.2, p.573.

268

ON THE EXTRACTION OF COBALT(II), NICKELCII) AND IRONCIII)

FROM ACIDIC LEACH LIQUORS

Darko Maljkovic, Zdenka Lenhard and Milka Balen Faculty of Metallurgy, University of Zagreb, Siaak, Jugoslavia

For the purpose of recovering coba l t ( I I ) and n i c k e l ( I I ) from leach l iquors obtained from Yugoslav ore ( the Coles l o c a l i t y ) the p o s s i b i l i t y of applying the cconercial ext­

ractant Cyanex 272 (bis /2,4,14'triaethylpentyl /phospnlnic a d d ) наз examined. Cyanex 272 proved to have an advantage over : elated organophosphoric extractants D2EHPA ( d i ­ 2 / e t ­

hyl/phosphorie acid) and FC88A (2­ethylhexyl phosphonic acid mono 2­ethylhexyl es ter ) [1 ,2] . Extractions were carried out from aqueous sulphate, chloride and sulphate­

­chloride so lu t ions . The extraetant Cyanex 272 was better at separating cobalt­nickel from a sulphuric acid medium [1 ,2 ] . whereas chloride leach l iquore made simple remo­

val of i r o n U I I ) poss ib le [3,1] . A sample of Cyanex 272 was a g i f t or Cyanamid Ltd. Canada. The samples extracted

were leach l iquors or so lut ions prepared from pure chemicals. Ground ore or grain s i z e ­= 0.125 mm was leached in a device of own construction a t room or higher temperature (85°C). In Table an average content of metals determined in ore and leach liqu ­s i s given.

Average content of cobalt , nickel and iron in the ore from the Gole? r e a l i t y and in leach l iquors produced by di f ferent procedures [5­8]

Leaching agent

Temp., °C

Content of metal Leaching agent

Temp., °C Co Ni Fe

ore,% ­ ­ 0.06 1.3 21.3

leach

l iquors ,

g/dm3

8 M HC1 85 0.25 3.6 68.0 leach

l iquors ,

g/dm3

8 H HC1 room 0.23 2.6 40.0

leach

l iquors ,

g/dm3

1 M HgSO^ 85 0.10 2.0 17.0

leach

l iquors ,

g/dm3

1 м HjSO^­eMHci room 0.25 2.9 46.0

Extraction was carried out in graduated cuvettes (15 cm3

). By at1 orbance measurement ­

­ concetratiens of cobalt (using nitroso­R salt) and nickel (u' .ng dimethylglioxime and KjSjOg) were determined. Iron was determined by complexome ­ric titration and lar­

ger quantities of nickel gravimetrically.

In prepared samples of composition corresponding to the r stal content in ore leach liquors the effect of different diluents on cobalt extract' >n from aqueous sulphate solutions with Cyanex 272 was examined. The diluents were: n­hexane, benzene, 2,2',4­

­trimethyl pentane (i­octane), kerosene and a mixture of benzene and 2,2,4­triinethyl pentane. Experimental conditions (temperature, init ial pi, contact time, initial phase volume ratio) were chosen on the oasis of preliminary lamination. Figure 1 shows the results of extraction with the above diluents at dif f ~ent init ial phase volume ratios of organic to aqueous phase ( r

1

) . 269

E3

The histogram In Fig. 1. shone the IJ H am» with 2,2,l»­trimethyl pentane аз diluent to yield heat results. The recovery and separation of metals strongly depend on the pH of

Diluent:

1 ­ n­hexane ­, 2 ­ eyclohexane; 3 ­ benzene­, U ­ kerosene^ 5 ­ 2,2,it­triaethyl pentane; 6 ­ 2,2!t­tri»ethyl pentane

•• 5% benzine

Fi«. _K Recovery factor for cobalt extracted froa a sulphate medium with 10* Cyan» 272 in different diluents. Initial pH = 6.0, ei^O.08 »g/caJ

,

= 0.5 and 1.0 Fig. 2. The effect of initial pH value on ex­traction of cobalt with 10% Cyanex 272 In 2,2',4­1г1»еи1у1 pentane froa sulphate medium

= 0.08 mg/o

= 1.0

the extraction systems slth Cyanex 272 be­

cause of reaction mechanism. Under defined experimental conditions the most favourable init ial pH value for extraction of cobalt

was 8.0 (Fig. 2.) and corrpespondlna the nost favourable equilibrium pH value was 6.0 (Fig. 3 . ) .

At the same init ial value pH = 8 the change in init ial phase volume ratio had no great Influence on extraction, as was observed by a parallel investigation. At chosen experimental conditions (10* Cyanex 272 in 2,2;4­trimethyl pentane, Initial pH = 8.0 and r =1.0) extraction «Г cobalt ( c ^ = 0.08 mg/om>> and nickel (cj^ = 1.77 eg/cm») as well as the oobalt­nlokel separation were studied.

Contrary to toe results reported In literature, moat of which refer to a higher ten • perature, these results were obtained at 25°C.

During the extraction of cobalt from leach liquors some interferences caused by the presence of other components appeared. One was due to high concentration of Iron. In 1

270

Fig. 3­

•me effect of equilibrium pH value on extraction of cobalt with Cyanex 272 from sulphate medium. A compa­

rison with results reported in Ref. t.

Щ ­ 185 g/dmJ Cyanex 272 in Kermac

470B » 10» nonyl phenol 0 .015 B o l / d B

3

, t = 50°C, ""Co

i 1.0 (Ref. 2.)

10* Cyanex 272 in 2,2',U­tri­

Bethyl pentane = 0.08 g/dm

5

, t = 25°C, 4

0.9

0.7 /J 0.5

/ / 0.3 • J J 0.1 " •—­

• /

i i . .

=. 1.0 3 « 5 EQUILIBRIUM pH

order to solve the problea the posibility of extracting iron from leach liquors ­.."ore cobalt and nickel was studied. For the extraction of iron from leach liquors ­* ai­e­J by different procedure (as given in Table 1) the solvents used were: diisopr™.i­

etJier, ethyl ether and their mixtures with n­butyl alcohol. Best results wu г obtained from chloride and chloride­3ulphate media with the mixture of ethyl ether and n­butyl alcohol. As illustration, from a sample of chloride leach liquors with tl.e mixed sol­

vent ethyl ether • n­butyl alcohol (30%) in the first stage of extraction the recovery factor (H) was 0.988 and the separation factors iron­cobalt ( <C £) and iron­nickel (oC jjf) were 2008 and 1600 respectively.

Finally, i t a.juld be emphasized that fficaciou3 application of Cyanex 272 for the recovery of cobalt from the investigated ore from the Golea locality depends, in addi­

tion to the above, also on effective removal of other component3 present in leach l i ­

quors, in the first place on effective removal of aluminium. Investigations are in progress.

References 1. Ann.//Cobalt­Nickel Separation Using Cyanex 272 Extractant, Technical Bulletin,

American Cyanamid Company. New Jersey, USA, 1982. 2. Rickelton W.A., Flett D.S., West D.W.//Cobalt­Nickel Separation with Cyanex 272

Extractant, International Solvent Extraction Conference. ISE~'B3. Denver, Colorado, 1983. P.195.

3. Maljkovic D., Branica M.//Croat.Chera.Acta. 1966. Vol.38. P.65. 1. handbook of ЫопГеггоиз Metallurgy (Recov ­>f the Metals), Editor Donald M.Liddell.

New fork and London, McGraw­Hill Book, 19U5­

5. lenhard Z., Maljkovic Da., Balen M.//Tehnlka. Rudarstvo­geologija­roetalurgija. 1985. Vol.36. P.885.

6. Vueurovlc D.//Tehnika. Rudarstvo­geoiogiJa­metalurgi'­\. 1965. Vol.16. P.275. 7. Lenhard Z., Maljkovic Da., Balen M.//Metalurgija (Sisak). 1986. Vol.25. P.153. 8. Lenhard Z., Maljkovic Da., Balen M.//Metalurgija (Sisak). 19B7. Vol.26. P.3.

271

HOVEL TECHHOLOGY FOR SOLVEHT EXnuCTION­SIHULTAHEOUS 11­12 I SEPARATIOH OP Cu, Co, Hi WITH HEH(EHP)­KEROSEHE­H 2SV

Han LI, Toujun Fu, Welling Zhang, ZMchuan Chen, Changchun Institute of Applied Chemistry, Academia Sinica, Changchun, China

the three­exit extraction technology has been reported [1,2] for the separation of rare earths recently, but has not been reported in non­rare earth separation. In many cases, there are multi­component system, such as in the work of nickel electrolyte purification Cu, Co, Hi may coexist. Usually Cu and Co are treated as one uomponent, and separated tToa nickel. Яеге a new technology, by which the three elements can be separated simultaneously, is proposed. 1. The extraction and separation ability

Prom the effect of pH on the distribution ratio of each element in individual component system (Pig.1), it is determined tha*

рН^­З.З; рн£°­3.8; pHJP­4.7. The order of extractioc ability can then be formulated as Cu > Co > N1. And the elops have the same order. These determine the relation between the exits and elements in the technology.

x

Pig.1. The effect of pH on distri­bution ratio of Cu, Co, Hi

4"""

iS *» Ji P"

Fig.2. The effect of pH on separation factors

Tig.2 shows that both the separation factors (3 /_ and P C o/ H< hare the tendences to increase when the pH of the equllibrum aqueous solution increases.

The extraction percentage (E)0 at different ammoniated ratios of HEH(EHP) were measured when the initial metal ion concentration was constant. It is shown that there exists certain linear relation between

* The Project Supported by National Katural Science Foundation of China,

272

the extraction ratio and ammonlated ratio of HEH(ERP) (KHjX). Then »e can easily reason out the linear relation between the metal concentra­tion In organic phase and HH3*.

The effect of metal ion concentration on the separation factors la also studied. Experiments show that ? C u / C o has little variation with change of initial Co concentration when the Cu * Initial concentra­tion Is set constant. Whereas Р С ц / С о таг­ies greatly when the initial concentration of Cu 2 + changes and C 0 o2* keeps constant. Similar regu­larity exists in the system of Co and N1.

The effect of temperature on P C u / C o and Р С о л ц are shown In Pig.3. It can be seen that ? r.„/Co falls while ?co/Ki

r a i s e s w i t h t n e i n

~ crease of temperature.

Flg.3. The effect of temperature on separation factors

2. The technology of the simultaneous separation The simultaneous separation experiments of Cu, Co, N1 was carried

out at room temperature. The feed composition is Cu:Co:Ni»1:1:1 Technology 1 The scheme of technology 1 is as Fig.4. The conditions are H­14, NP­3, NG­6, NW«10 VOE:VOG:VAW:VAG:VAF­15.0:4.0:11.4:7.4:7.5

VAW

4. vff >jm

M M |NF| \U I

vow t i

N

k$* F1R.4. The sch erne of three­exit technology 1

18.3uc.382 273

The initial concentration of extractant 1в 0.74Н and aaaoniated ratio 1»' 68%. Sulphuric acid scrubbing concentration la 0.6m. Toe feed solution has the total concentration of 0.498И.

After equilibrium, the qualities of the three products are as follows

The first aqueous exit: Я1 > 90% The second aqueous exit: Co> 85X The organic exit: Co > 99.5*

Technology 2 The scheme of technology 2 is as Fig.5. The conditions are H­16, HF­3. KG»10 V0E:VAW:VAF:VAG»19.0:11.6:7.5:4.3 The initial concentration of extractant la 0.72M. Ammoniated ratio

is 12%. Sulphuric acid scrubbing concentration is 0.59N. Feed solution is 0.49M. The products of the three exits have the qualities as follows

The first aqueous exit: Ni > 99.55* The second aqueous exit: Co>97* The organic exit: Cu >99*

VAF V»E ­ « — • r­*­

i NF

VOW vAqr VME

Fig. 5. The scheme of three­exit technology 2

Nomenclature NF,N ­ feeding stage nuober and total stage number, VAF.VAW ­ volumes of feed and scrubbing solution) VAE,VAG,VOW ­• volumes of the first, second and third exit producte; VOE.VOG ­ volumes of extractant; NV ­ exit stage number of organic product for technology 1; NG ­ back flowing stage number for technology 1; exit stage number of

second aqueous product for technology 2.

Acknowledgement We are thankful to Chunfu He and Hongxing Ren for providing X­ray

fluorescence analysis.

References 1. Yunyu Shen, Jinghong Jin//Journal of the Chinese Rare Earth Society

(Chinese). 1985. Vol.3, N 2. P.1. 2. Han Li, Zhichuan Chen, Wcixing Bhang//Proceedings of the First

Symposium on Extraction Chemistry in China (Chinese). 19B5. P.128. 3. Chunman Gao//Nonferrous Hetals (Chinese). 1984, N 1. P.44.

274

THE DS« OF TEULKILBHZrLAMIOBIOII SALTS П HYDHOHETALLUBCT i ; — I.И. I тел от, L.H. Gindln, H.L. Havtanovlch, Institute of n

~ '3 .

Inorganic Chemistry of the Siberian Branch of the USSR Academy of Science» , Bbvosiblrsk; "Gipronikel", Leningrad, USSR Researoh work carried out by us enabled us to accumulate certain

experience In the use of trlalkylbenxylammonium salts (TABAS) for extraction, separation, and refining of metals. The use of nonpurifled tertiary amines and benzyl chloride in the synthesis of the extract­ant results in the final product containing some free amines. Their presence does not practically interfere with metal extraction from neutral and alkaline solutions whereas In acid solutions one has to take into account the possibility of an accompanying extraction of acids and metals by these amines. If required, the extractant can be easily freed from the amine impurities by extracting them with kero­sene from where they can be isolated into a separate phase by an acid treatment of the extract.

TABAS has a high viscosity and for extraction purposes it Is em­ployed In the form of solutions in nonaqueous solvents. The viscosi­ty of the solution depends on the properties of the solvent, the ex­tractant concentration, and the temperature of the solution. Owing to the tendency of TABAS to aggregation the solution viscosity stron­gly Increases with increasing extractant concentration.

The solution viscosity and the average degree of TABAS aggregat­ion ю in solutions can be estimated using equation 1 which relates the specific viscosity of a solution with the molecular mass of the polymer in this solution / 1 J

4x ­ K cx <•> \ (1)

where h » (h.S~')e / l > a n d

1 s a n

* Is a r e t n e v

iB O O B i t i e a o f

the solution and solvent, respectively, and "a" Is the permeability coeffioient of the polymer. For equilibrium systems the amount of TABAS aggregates will decrease with decreasing TABAS concentration and will tend to unity ( n^ — 1 ) in the region of diluted solutions. Equation (1) can be used to estimate the values of n. •

log r^x « log К + a log n x + log C x , (2) In the region of diluted solutions ( at n x — 1) the slope of the

log <^ т я • 1°* cx Pl°* will also tend to unity ( » log f\x I a log С

—­ 1). The increase in the slope Indicates the appearance of ag­gregates in the solution. Calculated values of n for TABAS solu­tions are given In Fig. "\. It is characteristic that with the same concentration of TABAS the viscosity of a solution is proportional

275

that of the pur* solvent. It can be assumed that the TABAS aggregates consist of chains or plane «peel»» sandwiched «1th the solvent • With increasing temperature there la a noticeable Increase In the solution viscosity Oig. 2 ).

if*

«0

Я

K g . ­1* ­W ­Д2 tyCw

1. Xhs degree of TASAS 20 re" TF

yig. 2. The viscosity of aggregation as a function of concentration

TABAS solutions as a fun­ction of temperature

In order to maintain high metal distribution coefficients TABAS ia used in the for* of solutions in aromatic hydrocarbons such as polyalkylbenEenes of various origin. But for many systems mixtures of kerosene with aliphatic alcohols are used.

The metal extraction isotherms, is particular of metal chlorides, «lth TABAS solutions are «ell described by the following empirical equation which takes into account non­ideality of the extractant so­lution

_ (a+D (m+1) m * X ( С u o t , ­ BC,^)» С,,,., К , . (3) **(o) ЩСЕ'

«here a is the metal cation charge, n is the charge of the anion being extracted and p » 1.6­1.8 in the extraction of Co(II), Cu(II), Hn(II) and other metals and p ­ 0.7­0.9 la the extraction of Pe(III) and Cu(I).

Using ХАЗЛЗ a technology has been developed for extractive puri­fication of nickel electrolytes producing pure nickel and copper together «lth oobait oonoentratee CQ . In addition, a method has been developed for extractive treatment of the cobalt concent­rate which, in oonbinatiori with the first scheme, form a close­circuit technology (Fig. 3).

276

Pig. 3. Flowsheet for se­lective extraction of Impu­rities from nickel electro­lytes

Arsenic is effectively extracted from acid electrolyte solut­ions with a molybdate form of TABAS.

(HE 4) 2H^lo 70 2 4 + HjAe04 — nm 4) 2Uo : c(As 20 5)0 y . (4) The cobalt i s s tr ipped from the extract with water t o give a c o ­

b a l t s o l u t i o n . The ix­on i s a l s o strippr­" with water but a f t e r i t s reduct ion with an i r o n po;?der/3/

2(HR 4 FeCl 4 ) + Fe — 2 BE^Cl + 3 FeClg. (5)

A method has been developed f o r waste water process ing which al lows crea t ion of a w a s t e ­ f r s e technology .

A HB..SCH form of TABAS s e l e c t i v e l y e x t r a c t s some metals from n i t ­

r a t e and s u l f a t e s o l u t i o n s (Zn, Cd, Си, Со ) Л7» The NH.SCN extractant has been used i n a t e c h n o l o g i c a l scheme

i n the process ing of cobal t so3ut ions r e s u l t i n g from d i s s o l u t i o n of wastes with n i t r i c acid / 5 7 . Copper and i ron were removed from the s o l u t i o n with carbonic acid and cobal t was extracted by the r e a c t i o n

C o2 + + 2 HOj + 2 HR SCH ­ (HR 4 )gCo(SCK) 2 (H0 3 ) 2 . (6)

Str ipp ing of cobalt was carr ied out with a s o l u t i o n of ammonium o x a l a t e .

(HRjjSOT) s o l u t i o n s extract copper ( I I ) from s u l f u r i c acid s o l u t ­

i o n s by the reac t ion

Cu' ,2+ n 2 ­

+ H 2 S O 4 + 4 HR^SCH u. т n2f>v. т ч о»,!»! ­ 0ra 4) 2ou(scH). + 2 UH.HSO.. (7)

Due to the slow rate of the iron extraction copper is selective­ly recovered from acid solutions with pH > 1 (Fig. 4 ) &7, The distribution coefficients of copper are considerably increased when extraction is carried out from в mixed sulfate­ohloride eolation»

277

5 T,h

F i g . 4. The e x t r a c t i o n of Cu(II) and F e ( I I I ) with TABAS s o l u t i o n s as a funct ion of time

The HR.KO, form of the extractant has been used t o purify W and Mo from many i m p u r i t i e s . In H,^ s o l u t i o n s complex anions of the type MogO^T are formed which prevent formation of heteropoly compounds. The method al lows i high l e v e l of p u r i f i c a t i o n of metals from As, P, SI , Al , Ha and other .

Various s a l t forms of TABAS are the most promising ex trac tant s f o r ex trac t ion of anionic metal complexes from neutra l and a lka ­

l i n e media,.

References

1. Ivanov 1.Ы., Zai tsev et a l . / / I z v . S i b . Otd. Akad. Haufc SSSR. 1986. Ser. Chim. Hauic.J.. 16.

2 . Ivanor X.K. et al./?Hydromatallyrgy. 1979. £ . 377.

3. Kheifeta 7.X. et a l . / / Patent USSR. H 352955, B. 1972. H 29. 4. Ivanov 1.Ж. e t aLDep.7lHITI,1979, Ho 2441­78, 24 . 5 . Havtanovieh H.X. et a l . /Zhurn . Prikladnoj Khlm. 1977. 50. 3 .

660. 6. Ivanov I . I , et a l . / / Iav . S i b . Otd. Akad. Hauk SSSE. 1977.

Ser. Khim. Hauk.£, 70. 7 . Ivanov I .K. et a l . / / I zv . S i b . Otd. Akad. Hauk 8SSH. 1972, 7. 90.

278

^ EXIRICTIOI OF «TALS USI«G HAPHTBBIO ACIDS ГВОК SOUJTIOJB О? COMFOuTO 0СМКВИ1ОТ A.M.Berestovoy, V.D.Dvmidov, O.L.Belkova, H.N.Voronin, V.I.Tolstunov, I.E.Cherkasov, D.G.Petruuiak, L.A.Toronchihlna, Leningrad Mining Institute, Leningrad, USSR

She carboxylic acids зв the extractants are widely ueed in scien­tific investigation», technological testings and some of there are al­ready used on a limited scale for separation of metals in industry. Haphtenic acids represent one of the interesting extractant of this class. They are natural substancee having a number of properties which meet the most part of the industrial requirements for extractants: low salubility in water and high solubility in organic solvents, nontoxic!­ty, availability and low cost. £hey extract metals at ph value, close t« precipitation of their hydroxides fi,Z7­ these are the approximate coefficients of separation: Je(III)/Cu­ 300­700; Cu/Zn­ 100­350; Zn/Co­3­6; Hi/Ce­ 1,5­2,0; Co/Hn­ 1,0­1,5; Co/Ca­ 2­5. Ihe above indicates high efficiency of Fe(III), Cu and Zn separation. Hulti­stage extrac­tion needed for other metals (Zn/Co­12, Hi/Co­ 35­40 stages) is a drawback of this extractant. The possible ways of increasing the appli­cation efficiency of naphtenic acids are discussed below.

Studing the kinetic regularities at temperature 15­90°C we deter­mined Fi] considerable variations in extraction rates for non­ferrous metals (Cu, Ni, Zn) and Ca (fig.1). On this basis there had been work­ed out a process of refining these metals from Ca that permitted to increase the coefficient of sepa­ration Co/Ca to 15­45» The realiza­tion of this process in the one stage of mixer­settler centrifugal extractor (Mlot­in'" atrial scale) with low residence time, permitted to reach a deep refinement of Co and Ni from Ca and aig.

It is known that reducing Cu,Ni and Co by hydrogen in aqueous solu­tions is a highly efficient in­dustrial method of purification and producing them as powder metola

25 SO № W> "C. Pig.1. The conditional extraction constant of rate as a function of temperature.l­Ca,2­Mf3­ Co,4­Zn The realization of these processes

in organic extracts /47 permits to decrease pressure and temperature and to increase selectivity of Co and HI separation. Here is shown that these processes take place at t = 130­220°C and P„ = 1­й MPa, the

279

rate of processes being higher. Thus, in tne 33% solution of D2EGPA at temperature 140C

C and pressure 2l£Pa Hi is reduced in 30 minutes but Co remains in the extract. We studied Cu, Ni and Co reducing in naphte­nic extracts in an autoclave (3 1 volume). The parameters of the pro­cesses lie in the limits shown above. The naphtenic acids repeatedly u3ed in these processes did not change their extraction properties.

Impossibility of application of staniard equipment for the metal separation processes based on kinetic regularities and on processes complicated by the third solid phase formation brought about the neces­sity of the construction of the mixer­settler cenrifugal extractor (Fig.2) /3/. The rotor­separator has mud chambers on the peripheric part (inside) with fixed nozzles talcing out the solid phase together with the major part of water. Periodical washing out cf rotor is realis­

ed under the very low concentration of solid phase. The reguired residence time for reali? ing processes is achieved by mounting corres­ponding volumes of chamber mixers. The *• .lot models of extractors (Productivity IT m 0,5 to 1,5 m­Yh for all phases) have b ,n used to

­Г^Ь х^У гГ"3**~ test the technique and have show . satisfacto­** • ry results in the wild ratio range of flows

and densities of phases. During the testing of the geometrycally similar models it has been determined that with equal centrifugal factors of separating the ratio of their ca­pacities is equal to the ratio of the outside

Pig.2. The scheme of construction of mixer­settler centrifugal ext­ diameters of rotors to the third power: rator: 1 ­ driver; 2 ­rotor­separator; 3 ­mixer; 4 ­ frame with annular receiver; 5 ­chamber of mixing; 6 ­pumping over equipment

V«2 = (D­/Dg) J

. The equation is necessary for calculating the basic sizes and number of re­volutions in designing industrial extractors.

On the basis of distribution regularities and metal extraction kinetics the technologi­cal scheme for treatment of sulphuricsolution (contents g/1: 2­3 Cu; 3­4 Zn; 2­5 Co; 1­1,5

Ni; 0,3­0,5 Ca, 8­10 Mg) has been worked out and tested on a pilot­industrial scale (Fig.3). The worked out technology permits to obtain goods products in the sort of salts and oxides with 99.45S extraction of metals (92# Mg). The technology could be improved Ъу reduction of Cu, Hi and Co in organic extracts. It permits to cut down by more than 3 times the number of extraction stages (because of simplification of separation of Co,Ui,Zn)and by 2 timeerthe expenditure of reagents.

It is known that the use of the extraction processes is limited by concentration of metals it. the solution treated. The expenditures on the fill up (supplying) of losses of extractant at low concentration

280

Init ia l solution NH.R

4­

f n s

ih extract ion, tages

шАк

NH.R CuSO, __|Cu extraction, 13 stagrs

Jwashing, " I I at age

So, Ni extraction, 1 stage

washing, 6 stages

1 I HoSO,

I NiS04,CoR2,HgS04 H 2S0 4

Cu reex*ractioH' at age

F i NH.OH I CuS0 4 ' 5H 2 0

i/Co separation 35 stages

йп reextraction, 1 stage

» CoSO. solution NiSO.

4 4 T ""Г

i i

Loading of extrac­tant

KH.R to extraction 4

X CNH 4) 2C0 3

3o,Ni,Ca finish extraction __ Mg precipitation 6 stages

Г evaporation

MgO (NH 4) 2S0 4

Fig.3. The technological scheme of treatment sulphuric solution; water phase; organic phase

increased relatively to the cost of the good products which leeds to the reduction of economical efficiency of the processes. For naphte­nic acids this defficiency could be avoided by application of combin­ed flotation­extraction technology. The basis principles are the fol­lowing: flotation is applied for preliminary concentration of extract­ed metals; pneumatic product is treated by extraction methods; combi­nation of flotation and extraction is attained by use of the same re­agents as .the collector and extractant. The use of flotation in the heading operation provides a high productivity of the technique and permits to decrease a wastage of naphtenic acids from 50­100 mg/1 to 5­8 mg/1. This permits to trea. times more diluted solutions. The combined technique is developed for extraction of Cu from so Lution containing g/1 0,2 Ou; 0,5 Co; 2,0 Mn; 35,0 Mg /5/. Below is the suc­cession of operation (Fig.4): ionic flotation of Cu by naphtenic ammo­nium at ph • 6­7,5; dissolving of pneumatic product in the kerosine solution of naphtenic acids; treatment of the organic solution by the Vnown extraction processes (washing of CuS04 solution and reextraction jf Cu) and regeneration of collector. The regeneration of collector is resized ty its transfering into aqueous solution with the help of tre­atment of organic phase by BI^OH­aolution at ph.8,3­8,5 and t=50­.60°C.

281

Initial Solution flotation, 1С mln

E chumber product (Cu cleaned solution) {product

NH.OB

3 / [dissolving of pneumatic

pneumatic product

I

[regeneration of collectorU. I extraction treatment

нн 4н IHR CuSOjj/

Pig.4. The scheme of flotation­extraction iaolation of Cu from solution; water phase; organic phase

The expenditure on reagents, including the cost of wastage of the naphtenic acids, reaches 2,5% from cost of the goods CuSO..

The combined method permits efficiently use of naphtenic acids for utilization of non­ferrous metals in flotation technology for purifi­cation of the sawage /6/. The technique of purification of the waste mine waters for one of the mining plants (with utilization Cu and Zn to the good products in sort of sulphates) has been worked out too.

Thus, the combination of the considered processes and centrifugal extractor in the technological scheme permits: to increase the effi­ciency of application of the naphtenic acids; to realize the complex treatment of the different content solutions and concentrations of metals; to receive the wide range of the goods products in sort of salts, oxides, powder metals; to solve ecological problems.

References 1. Fletcher A.W., Plett D.S.//Proc. Int. Conf. Solv. Extr. Met. in

Harwell. London, 1965. P.359­375. 2. Illuvieva G.V.//Zapisky Leningrad, gorn. institute. 1966. T. XLVI,

H 3. P.95­109. 3. Berestovoy et al. // Goraorudnaja Prgioram v nauke i technike,

Symp. Prglbram,1986. P.745­756, 4. Burkin A.R.3ur.ges J.A. //Ingenieursblad. 1972. V.41, N 17. P.459­

463. 5. Auth.Cert,USSR. » 1235959. B.I. 198S. N 21. P.99. 6. Voronin K.N. et al.//Tezisy dekl. Vsesojuznogo sovesohanija po

technologii extracoionnogo razdelenija neorg. veschestv, Apatity, SSSR, 1986. P.38­39.

282

STUDIES AHD CBVELOJMKHTS Of HF'TAL SSTRACTIOH Ш HXDR0JI3TAI ~ ~ ~1~] HTDROISKTALLURGY OF HICKEL AHI COB'.LT 1 ­ 1 U.L.Havtanovich, L.S.I.utova, State Design and Research Institute for M.ckel­Cobalt Industry Jlpronickel, Leningrad, USSR

In the USSR studies on the application of extraction in hydrometal­lurgy of nickel and cobalt began with the use of carbozilic acids for removal of impurities from cobalt solutions [ 1 J and with the use of tertiary aliphatic amines for separation of cobalt from nickel [2J .

Studies of extraction processes from the point of view of electro­chemistry «ere carried out under the leadership of V.l.Kheifetz [ 3,4)­The potential jump at the extract­aqueous phase interface and its role in the extraction processes, as well as the influence of the electro­lytic dissociation ot extractanta and extractive compounds in the or­ganic phase on the processes of ion­exchange extraction were conai ­dered.

It is stateu that the selectivity of the cation­exchange extrac­tion increases with the decrease of the strength of the organic acid used as an extractant; electrophllic reagents reduce the values of the anion­exchange extraction of metals; the strength of the organic ba­ses (nltrlles, amides, amines) correlates with their extractive power towards oobalt and nickel chlorides being extracted from 4M solutions of BaCl. Despite their relatively ?.ow conductivities, the organic so­lutions of extractanta and extracts behave like typical electrolytes.

3+ # 2+ Oxldatlon­reduction potentials of Fe^VFe system in organic solu­tions of their salts with caprilio acid and D2BHPA I 5 ] and of Cu 2 +

/Cu+

system in organic solutions of QAB and tertiary amine salts [ б j *­­fe measured»

Hie electrochemical measurements are us'ful for the explanation of the mechanist of metal extraction and for the determination of the composition of extractive complexes. The above measurements are also effective for practical control of extraction processes because they make it possible to determine quickly the type of emulsion in an ex­tractor mixing chamber, phase interface in a settler and the oxida­tive state of metals in extracts.

Is selective cxtractants for nickel, long­chain dialkyl­ct­dioximes of the general formula R­C(HOH)­C(HOH)«R were synthesized and studied with the following R and R 1

: 1) CH­, C3H7, 2) CH3, C6H13, 3) CH3, C 7H 1 5, •) 0H 3, CeH­17, 5) C 2H 5, C3B7, 6) C3H7, C4kg, 7) 0 ^ 3 , C 7H 1 5, 8) C 7H 1 5, C 6H 1 7, 9) 0 8

H

17» С А э " 1 0 } °10Н

21« °11 Н

2з'­7

^ Dioximes extract nickel at pE­4.2 and copper at pH«5.4 from sul­

phate solutions. With the help of dioximes, nickel ia effectively se­parated from cobalt, copper, ferru», zinc, arsenic, cal"­ium, aagne­

283

slum and sodium In ammonia solutions. Among the synthe izet! dioxides, ethylpropyl­and metylgepthyl­oL­dioximes are the most effective ex­tractants of nickel. Stripping of nickel from organic phases is per­formed by aqueous solutions of mineral acids.

Extraction of heavy non­ferrous metals by trialkybenzylafflmonium rhodanide waa tested and the following order of extractiveness of the­ir salts was found» 2п(Н0з>2> ZnCl2 > Cu(N0 3) 2> Со(1ГОз)2 > CdCl 2> > Cd(H03)2 > CuClg > CoCl2>> ZnSO+ > CuS04? C0SO4 > CdS04 f 8 J . In the hydrometallurgy of nickel QAB rhodanides can be used for the se­lective extraction of impurities, in particular zinc from sulphate so­lutions.

Organic salts formed by a cation of quaternary ammonium or phospho­nium and by an anion of earboxylic acid, n(ZX)0+ (МУ П) 0 ,

2 > n(ZY>Q+ + (HIn)aq.. where Z =ЩЯ, R4P; Y­HCOO; И ­ metali X • CI", Ш 3 , SO42

", HSO4, were tested as extractants of non­organic salts [ 9] . Ehe extrac­tion prooeeds according to a reverse reaotion and depends on the tot&l concentration of salts in the aqueous solution, lion­organic cations and anions are extracted according to the cation­and anion­exchange extraction for organic acids and bases forming the extractant. Non­or­ganic ions in the extract are mobile and can be exchanged which makes it possible to carry out an extensive non­organic synthesis.

Hie crystallization of salts from extracts during the stripping process studied by us on the examples of nickel, cobalt and copper stripping from their salts with carboxyiic, naphthenic and di­£­ethyl­hexylphoaphoric acids by concentrated solutions of mineral acids is of practical significance [ 10 J . The following crystalline hydrates of salts were produced from extracts: CUSO4 • SHgO, 0u(HO,)? • ЗН^О, CuCl2­ ^ О , Co304­ 7H 20, HiS0 4 • 7^0.

ТЫ; • ­sti­ity of liquid­liquid extractiob of metalo is studied on '!'« »• ­ .со or extraction of cations, anion complexes and salts [llj.

• sre derived describing the dependence of the distribution o. j • Smi2/m1 on thr^e factors, which are the ratios of equilib­• • <­.. jits of extraction .'««actions (Ее), the activity coefficients ' nets, iquo complexes in \"i aqi» .lis phase ( JWo. ,) and the extrac­

tive complexes in. the organic ph j; (Уе.с.)> correspondingly: rt г Kg a YM«'I­ < Ye• С

The flowsheets tested on a pilot plant scale are shown in the table» Besides, a method of production of nickel electrolyte from the hydrate of nickel oxj.de dissolving it with the help of carboxyiic acids [ 12] and a method of copper­nickel raatte treatment which combines the pro­cess of electrolysis with extraction [13] are developed on a laborato­ j ry soale. In the last few years a number of new technologies using ex­ "

284

Extraction flowsheets [ 14 j SI. [no в

Feed Extractant Products Sumber of

.stages 1

1

Cobalt anodes, composition, %: 77 Co, 8.3 HI, 10 Pe, 0,14 Cu, 1.3 S

1. Trialkyl­amlne ­ (>%, 1. Electrolytic fatty alcohols ­ 10S6, cobalt kerosene ­ 8456 2. Hydrate of 9

2. Amine ­ 3036, alsohol* nickel oxide 1 i ­ 109S, kerosene ­ 60? i

2 Cobalt concentrate composition, %s 23 Co, 1.15 HI, 0.C3 Cu, 1.25 Pe

1. 6% solution of amine in aromatic hydrocar­bons

2. 30£ solution of amine in aromatic hydrocar­bons

1. Cobalt oxide 2. Hydrate of

nickel oxide 20

3 Concentrate, com­position,%i 19.8 HI, 13.7 Co, 12.3 Cu, 3.8 As, 2.0 Zn 0.5 Pe, 75 H20

1. Synthetic fatty acldal. Cobalt oxide Cy­Cg 2, Copper sul­

2. 50% solution of tri­ ' phate i alkylbenzylammonium

;

3. Hickel carbo­cblorlde in aromatic , nate hydrocarbons 1

31

4 Chloride solution, conposltlon, g/1: 102 Co, 4.6 11, 6.2 Cu, 2.7 Pe, 2.1 Oa, 0,4 Ни, 202 Cl~

1. 6% solution of amine in aromatic solvent

2. 50* solution of 0ЛВ chloride in aromatic solvent

1. Cobalt oxide 2. Hydrate of

nickel oxide 23

5 nitrate solution, composition, g/1: 67 Co. 6 Pe, 17Cu, 5,9 HI, 18.8 Ca, 37 Ha

5056 solution ot fatty acids Cj­Cq 1л kero­sene

1. Cobalt oxalate: 3. Pe­Cu cake 1 3. Ito­Ca cake '

2 4

4. Ammonium ni­trate

6 Sulphate solution, composition, g/1: 30­32 Cu, 15 Hi, 15 Co

509» solution of fat­ty acids Cy­Cg in ke­rosene

. Copper sul­ '• phate .

2. Solution, 30 1 2

g/1 Hi, 30g/l { Co

3. Sodium sulphate

7 Solution, 22­39g/3 Co |

Patty aoids C7­C9, Cobalt Sulphate 5

285

traction processes combined with autoclave leaching processes, electro­lysis or autoclave hydrogen redaction have been developed.

The advantages of the extraction technologies отег the settling ones lie in the elimination of labour­consuming operations of filtration, reducing of reagent consumption and technological equipment volumes , in the increase of metal recovery into finished products and in the im­provement of their quality at the expense of deeper separation of me­tals by extraction. In some cases the extraction technologies have no wastes, they provide the regeneration of the most part of reagents and eliminate the formation of salt effluents.

The economic calculations confirm the high effective та of ext­ractive technologies ol nickel and cobalt production.

References J. Giadia L.M., Bobikov P.I. et al.//Inventor's Certificate 44038.

Bulletin of Inventions. 1958. N 7. 2. loffe E.Sh., Dushkina L.V.// Inventor's Certificate 163356. Bulletin

of Invetions. 1964. N 12. 3. Kheifetz V.L., Gindin L.H.// Chemistry of Extraction Processes .In

Russian). M.: Nauka, 1972. P.11­19. 4. Kheifetz V.L., Navtanovich M.L.// Zh. Prikl. Khira. 1976. H 1,

138­142. 5. Harkovich G.M., Havtanovich M„L., Kheifetz V.L.// Izv. VUZOV. Khim.

i Khim.. Telchnol. 1977. N 12. p. 1833­1836. 6. Harkovioh G.M., Havtanovich M.L., Kheifetz V.L.// Izv. TOZOV. Khim.

i Khim. Telchnol. 1977. H 11. p. 1647­1650. 7. Havtanovich II.L., Lutova L.S. Chemistry of Extraction (_In Eussian).

Novosibirsk; Nauka , 1984. P. 171­175. 8. Bavtanovich H.L., Ilmofeeva T.I., Kheifetz V.L.// Zh. obshchey khim.

1977. N 3, P. 540­545. 9. Havatsnovich H.L., Kheifetz V.L.// Zh. obshchey khim. 1975. N 2.

p. 413­418. 10. Havtanovich H.L., Timofeeva T.I., Kheifetz V.L.// Uetallurgy of Ni­

ckel and Cobalt/ Trudy Inst. Gipronlckel, issue 60. L. 1974, P. 72­74.

11. Havtanovich M.L.// Zh. Prlkl. Khim. 1979. H 8. P. 1769­1773­12. Eonovslov V.L., NaTtanovich H.L. et al.// Inventor's Certificate

533651. Bulletin of Inventions, 1976, H 40. 13. Uarkovich G.H. et al.// Inventor's Certificate 657088. Bulletin of

Inventions. 1979. N. 4. 14. Konovalov V.L. et al.// Tzvetn. metally, 1984. N 8. P. 24­26.

286

RURIFICAT10N OF NICKEL ELECTROLYTE (HCUH ?SC, MIXED SOLUTION! | 11­l6 J BV SOLVENT EXTRACTION FOB GETTING HIGH­PURE (0 #

) ELECTROLYTIC NICKEL Li Zhou, Cao Chun­raan, Wu Hua­wu, Zhang De­leng, Bao Fu­Yi, Liang Zun­fu, Tsinghua University, Beijing, China Foi overcoming the disadvantages of purification process of nickel eletrolyte

mainly by precipitation method, a new process based on the solvent extraction purification method was developed.

The composition of nickel electrolyte is as follows: Ni 70­75^/1, Co O.15­O.20g/1, Cu 0.45­0.5g/l, Fe 0.15­0.20g/l, Zn 0.004­0.005g/l, Pb 0.002g/l, Na 40­45g/l, H 3 B 0 3 3­8g/l, Cl" 110­l20g/l, SO* 80­90g/l, PH 1,3­1.5

The new purification process are divided into the following four steps: 1. Oxidation of Cu +

, Fe to Cu , Fe by chlorine; 2. Extraction of

amidate sort); 3. Extraction of Co, Cu, Zn by EHPNA (its Chinese trademark is P507); 4. Removal of Pb by ion exchange with 701 resin. After the above­mentioned purification steps, the qualified nickel electro­

lyte is obtained for producing 0 electrolytic nickel. In this paper only the extraction of Co, Cu, Zn from Ni by EHPNA is mentioned. As known, EHPNA is a efficient extractant for the separation of Co from Ni

[1­5], but it is less investigated for the extraction process taken place in the mixed acid (HCl+H­SO.) medium.

At first, the extraction equilibrium studies were carried out, based on this studies extraction casecade and bench­scale extraction experiments were finished.

1. Extraction equilibrium studies. In this experiments the parameters affecting the extraction equilibrium, the PH value, concentration of EHPNA, operation temperature and acidity of stripping solution are studied.

The feed is prepared according to the composition of given nickel electrolyte, and all the reagents are analytical pure. EHPNA is che product of Shanghai Organic Reagent Factory No.4, the diluent kerosene 260 is produced by Shanghai Oil Refinery. For controlling the equilibrium PH value, 12.5N NaOH is added into the prepared organic phase for saponification.

Equilibrium studies were carried out in the special test tube with the volume of 25ml.

Based on the obtained experimental results, the extraction equilibrium math­ematic models of Ni. Co and Cu were developed as follows:

log D M i=­2,5995 ,+ 2.727B log[RH] + 0.951ЭРН with the average deviation 9.8% J logD C o=­1.9566 + 2.1127 log[RH] + 1,2373 PH with the average deviation 9.77. J logD C u=­l.B307 + 2.2109 log[RH ]+0.6928PH

with the average deviation 13.37.. and the following.main points can be concluded:

1. The equilibrium PH value is the main parameter affecting the extraction

2B7

equilibrium of various metals, and their extracted order, in the ^i*ed HC1 * H.SO, medium is similar to that in the sulphate medium [l]. It is easy to separate Zn.Cu.Co and Ni by controlling the equilibrium PH value, but for Lhe separation of Cu and Co there are sone difficulties owing to their similar extraction characteristics.

2. The concentration of EHPNA affects the extraction equilibrium of metals remarkably. At the same equilibrium PH value the extraction ratios of metals increase with the increasing of EHPNA concentration, but the extracted order of various metals has not change.

1. The operation tetnpexatute mainly affetss the extraction equilibrium of Co, and ­the distribution coefficient of Co increases wit"h the increas­ing of temperature, so it is benefit to the separation of Co from Mi со raise the temperature.

4. For complete stripping of Zn.Cu.Co.Ni in organic phase, the equilibrium PH value roust be controlled below 1, then the H­SO, concentration in

l ч stripping solution can be determined according to the concentration of metals in loaded organic phase and the volume ratio between organic and aqueous phases.

II,Extraction casecade experiments. The casecade experiments were carried out in the same test tube as used in the equibibrlum studies.

Because the separation factor of Co from Ni is high enough in the EHPNA extraction system, the simple counter­current method was adopted, and the optimized computation was carried out based on the extraction rate of metals > 997, according to the fallowing objective function N x [RH] X V => minimum The computation results are as follows,:

[RH].7.(v/v) Equilibrium PH V(L=1) N

10 4 . 4 0 . 2 5

10 4 . 4 0.15 7

15 4 . 4 0 . 2 4

15 4 . 6 0 . 2 3

According to the computation results the appropriate operation parameters for extraction casecade are determined as follows:

[RH] 157,(v/v) v/L 0.2 t 35 1 1°C N 6

A part of extraction casecade experimental results Is tabulated in Table 1.

The number of stripping stage is calculated by the following equation; ,1­fy.

lo

*\l­f J N = ­ 1

log I

288

Table 1. Experimental results of extraction casecade

[RH] ,

%(v /v )

F e e d ,

g / 1

E x t r a c t i o n

s t a g e

Aqueous p h a s e , g / 1 o r g a n i c p h a s e , g / 1 Equil ibria

PH

[RH] ,

%(v /v )

F e e d ,

g / 1

E x t r a c t i o n

s t a g e N i Co Cu N1 Co Cu

Equil ibria

PH

1 5 . 0 N 1 7 7 . 6 1 77.47 0.0665 0.0472 1.2750 0.8116 2.852 4.62

s a p o ­ CoO.152 2 n i f i c ­a t i o n = 7 6 . «

CuO.692

PH 3 . 6 4

3

4

77.46 0.CO91 0.0012 1.3700 0.1551 0.0335 4.80

5

6

77.43 O.0O12

0.00066

O.OCOld

0.00018

1.7250 0.0416 0.0023 4.82

4.97

and Che calculation results show that near four ideal stripping stages is needed for the complete stripping of Zn, Cu, Co.Ni. Under the experimental conditions of stripping phase ratio (to/A) = 10:1 and ambient temperature,after four stripping stages the concentration of Ni, Co, Cu in the organic phase is 0.00088g/l, 0.00013g/l and O.00007g/1, respectively.

Based on the casecade experiments, the following operation parameters are determined; for extraction step:

concentration of EHPNA 15%(v/v) 75% 3.5-3.7 ~35°C

5.0 1:5 6

saponification of EHPNA PH value in feed operation temperature PH value in rafflnate flow ratio (o/A) number of ideal stage

for stripping step:

operation temperature flow ratio (o/A) number of ideal stage

2N. ambient 10:1

III. Extraction bench-scale experiments. According to the determined operation conditions the bench-scale extraction and stripping expenments were carried out in a laboratorial mixer-settler with raixer volume 0.51 and settle ing volume 1.51. In this experiments the feed solution is the practical nickel electrolyte. The experimental results are tabulated in Table 2.

It can be =oe.n from this results that the purification of nickel (Co 0.0005g/l, Cu 0.00031g/l in purified nickel electrolyte) and recovery of Ni. Co, Cu (Ni>

99%, Co.Cu>99.9%) are satisfied, and the qualified nicked cathode plate was produced in the succeeding electrolytic process.

289

Table 2. Bench­scale experimental results (Extraction stages: 6, stripping stages: U)

•

s t a g e No.

in aqueous phase , ? /} in organic phase. g/I in a 4 .

t ,

О с

о/А •

s t a g e No. Hi Co Сь Hi Co Си

in a 4 .

t ,

О с

о/А

1

2

3

4

5

6

77.42 77.32 77.35 77.30 77.35 77.23

0.1108 0.0325 0.0160 0.0059 0.0017 0.0005

0.0402 0.0282 0.0094 0.00097 0.00079 0.00031

0.9000 1.4000 1.2300 1.5100 1.2300 1.8300

0.6920 0.5050 0.1575 0.0390 0.0107 0.0095

1.9860 0.9240 0.1686 0.0113 0.0025 0.0013

4.24 4.76 38­40 4.62 4.89 4.9C 5.03 40­46

1/5

V 2' 3 ' 4 '

2.9360 0.1025 0.0346 0.0042

10.164 1.3400 0.0415 0.0016

0.0475 0.0200 0.0001 0.0000!

0.0120 0.0008 0.0001 0.0001

0.0680 0.0010 0.0001 0.0001

Arab. 1 0 / 1

At the end of this paper it can be concluded that: 1. EHPNA is still an excellent extractant for purification of nickel ele­

ctrolyte in the mixed HC1+H­S0, system. 2. The purified nickel electrolyte may be used to produce 0 eletrolytic

nickel, and the loss of nickel in the extraction process is only 0.237.. 3. Owing to the high flow ratios adopted in the extraction and stripping

steps, the stripped solution of Co, Cu is concentrated by about 50 times, it is of benefit to be treated further.

References 1. Chengye Yuan et al./fNonferrous Smelting (Chinese), 1981.Vol.10,N 2.Р.1. 2. Komasawa I., Otake T. and Hattori I.//J. Chem. Eng. Japan, 1963.Vol.

36. p.210. 3. Komasawa I., Otake T. and Hattori I.//J. Chem. Eng. Japan. 1983.

Vol. 16.p.384. 4. Komasawa I., Otake T. and Ogawa V.//J. Chem. Eng. Japan., 1984.Vol.

17. p.410. 5. Komasawa I. and Otake T.*/J. Chem. Eng. Japan. 19&4. Vol. 17. p. 417.

290

A COriPARISQN OF 2INC EXTRACTION FROM AMMONIACAl. SOLUTION! j USING COMMERCIAL CHELATING REAGENTS 1 'I C.Abbruzzese, Institute of Mineral Processing,National Research liDjpcil .Rome, Italy An arnmoniaca1 1eeching process was developed to extract netal

values from oxidized zinc and lead­zinc ores not amenable to f lo ­tationfl). Leach solution arising out contain dissolved zinc as cationic amine complex and attempts were made to recover it by solvent extraction. The use of carboxylic acids, naphtenic acids and organophosphorous extractants results in reagent losses owing to their appreciable solubility in aqueous phase.

The recent application of hyHroxyoximes and 8­hydroxy­quino­

line derivatives has broadened the variety of commercial chela­

ting reagents available for the extraction of metal ions from both acid and alkaline solutions (2).

A detailed investigation was carried out and the extraction of zinc from an ammoniaca1 solution was studied as a function of contact time, aqueous solution composition and pH.

The chelating extractants employed far the experimental work are indicated in Table 1.

Table 4., Commercial chelating reagents for zinc extraction

R e a g e n t S u b s t i t u e n t s

R4 R*

OH NOH

R,

L I X ­ 6 5 N

S M E ­ 5 2 3

P 5 0

C

9H

1 9 C

6H

S

с 9 н 1 9 н

HO

К.в1вх 1 0 0 C

! 0 ­

The organic рЬазв for all the extraction tests was в 10\(by vo­lume solution of unpurified extractant dissolved in a diluent: ke­rosene for LIX­65N,Acorga P50 end SME­529 and toluene for K B I B X with 10* p­nonyl­phenol as modifier. All the extractions were par­formed at room temperature (about 2D*C) in a mixing vessel descri­bed previously (3), The ratio of the volumes of the organic phase to the aqueous phase was 1:1.

The concentration of the constituents in the feed liquor was chosen based upon available pilot plant data (21.3 g/1 Zn) and the physical constraints of the eystem. The extraction equi^ librium is attained after about 3' min for all the chelating rea­gent tested] however, equilibration was allowed to proceed for 30 min. After phase disengagement, duplicate aliquot of organic

2Я

Table 2. Zinc extraction from ammoniacal solutions with hydroxyoximes

Ex treetant (NH 4) 2C0 3,

g/1

Distrlo. Coeff.

D

Extraction

4-

LIX-64N 50

100 20D

0.21 1.34 0.ЭБ

17.0 57.2 26.2

ST1E-529 50 0.21

100 0.51 200 1.30

17.0 33.7 56.4

P-5D 50

100 200

0.44 I .41 0.23

30.6 58.5 18.7

and aqu by plas tics of

Tha

1±Zinc extraction Ке1вх 100

taKen for metal conten on spectrophotometry. D reproducible within les extraction tests was ru ng pH and ammonium cone that can be drawn from

s ( Table 2 ) is that z es CLIX-65N, P-50, SME te concentration ( 50 g th all the hydroxyoxime tration is involved.

(NH loC0,- concentrati • . 4 2 3 . . however, raises zinc all the hydroxyoximes Kedly with LIX-65N an SME-529 is utilised a zinc extraction is fu sed at 200 g/1 f N H ^ onversely in this cas tion is depressed wit hydroxyoximes ,

Good zinc extract! ved, however, with KB 1 demonstrates the in ammonium carbonate со pH on zinc extraction with increase in ammo tion And pH in solutii remaining high. The b ver 90% zinc recovery tained at pH 6.0 and carbonate.

t determi istributi

5%. an the

entration examinat

inc extra 529 ] is /11 at pH s when lo' An increa on to 100 extractlo

but mor d P-50. W s extract rther inc CD g in so

zinc ex h the oth

nation on га-Zn-NH -. The ion of ct ion poor 9.0 . am-

se in /1

n with & mar-hen ant, ree-

lution. trac -er

on can be achie-lex 100. Fig. fluence of ncentration and .This decreases nium concentre-on, while still est results С о-) have been ob-50 g/1 ammonium

The detailed examination of the results of the astractiDn teat of zinc from ammoniecal solution, allows to assert that hydroxy-qui noline (Kelex-100) extracts zinc in a bigger extent than hydroxyо ximes.

These results can be explained by reference to the coordination chemistry. In fact, according to the principle of soft and hard acids and bases (PSHAB1, zinc ion belongs to the bordedine class

292

and its behaviour is mora like that of the herd cations (4). The ligand ЛЖ also belongs to the hard class. Pearson's rule thus po­stulates that thB hard Lewis acids (or cations) react preferential ly with hard bases. This explains why zinc amine complexes form r£ adily in ammoniacal solution.

In the system examined ­hydroxy­oximes and hydroxy­quinoline­

the properties of the extractant depend on the properties of the nitrogen electron­donor atom, i.e. on the oxime group =N­dH and the nitro group «N­(5). The structure and composition of LIX­65N and Kalex­ЮО have received ample study (2,6). Kelex­iDO consists of p­alkenyl­8­hydroxy­quinoline; the alkyl chain contains about Ю carbon atoms and a least one double bond. LIX­65N, instead, is a p­hydroxybenzophenonoxime.

The formation of the mBtal chelate depends o n the ease of oro­ton relase by the extractant molecule and the facility with which the nitrogen atom of the quinoline derivative (in the case of K,e­lex 100) and the alkyloxime (where LIX­85N is concerned) can dona­te their electron pair to the empty orbitals of the zinc ion.

Regarding the first point, it must be considered that in Kelex 100 molecule electrons associated with the hydroxjl group occur in в aromatic ring and hence proton is given up more readly than in the case of the LIX­65N (alkyl­oxima) molecule. In fact tr= a­cldity of the extractants increases as fallows (6]:

LIX­S5N < SME­529 < P­50 < Kelex­100.

Turning now to the second point, it must be stressed that elec­tron­donor properties of the nitrogen atom are associated with the electron density on that atom. The latter ensues from the molecule (inductive effects) and steric factors, when bulky substituents агэ close to the atom which is the active centre of the reaction.

Taking due account of these phenomena, it is fairly evident why only Kelex­100 extracts zinc with sufficiently high distribution coefficients with respect to the hydroxy­oxlmes. More precisely, at a given pH, extraction with Kelex­ma is greater awing to the higher intrinsic acidity and the bigger electronic density in the hydroxy­quinoline molecule. So chelates are readily formed with zinc ion. LIX­6SN, instead, whose molecule is less aciL and who­se electronic flexibility is lower, does not form chelates with the zinc that is already in solution as stable amine complex Zn[NH­)**.

3 4 The effects of structural modifications of the hydroxyoximes

extractants on the zinc extraction are evident considering that

293

ii J.. II Ли

LIX­65N molecule contains two isomers: syn and anti L Fig.21 . The syn structure is inactive towards the coordination of the zinc atom because the nitrogen ion pair is not capable for co­ordination and a structural re­

arrangement in favour of the anti form occurs slowly '7) .

In conclusion zinc can be extracted from ammaniacal solutions using Kelex 1QD and the hydroxyoxime extractants are less effecci­ve. The extraction is sensitive to pH as well as NH concentration.

Fig.2.The isomers of LIX65N

References 1. Abbruzzese C//Paper at TN5­AINE Confarenca,Denver,1987,А87­ЭЗ, 2. Attwood R.L.,Miller J . D .//Trans.А1ИЕ,1973.Vol.254.P.319. 3. Coldenberg J .F.,Abbruzzese СУ/Int•J­Mineral Proc. 19B3.Vol.lQ,

P.711. 4. Pearson R.G.//J.Amer.Chem.Soc. 1963,Vol. 85.P. 3533, 5. Briegleb G.//Electronen­dohator­acceptor Komplexe,Berlin:Sprin­

ger Verlag,1961, 6. Ashbroofc A.W.//Hydrometallurgy.1975.Vol.1.P.5. 7. Ritcey G.M.,Aahbrook A.W.//Solvent extraction principles and

application to process metallurgy.Amsterdam:Elsevier,1984.Parti. Ch.3.

294 i

ZINC RECOVERY FROM BLEED OFF STREAMS BY SOLVENT EXTRACTION AND ELECTROWINNINGt*)

Dante BUTTINELLI, Carlo GIAVARINI, Carla LUPI, Department of Chemical Engineering, University of Roma " La Sapienza", Italy

Previous works showed that zinc can be extracted from industrial waste liquors by using DEHPA and naphthenic acid as solvents (1,2). When the liquors are strongly acidic, e.g. spent electrolytes of the zinc industry, sulphuric acid can be previously extracted with iso-butyl alcohol to reduce the acidity of the original liquor and to recover the aci^ itself (3). Zinc can also be recovered from acidic sulphate liquors by direct extraction with TBP (4), which is very selective and gives high recovery yields. A drawback of the TBP use is that a salting agent, such as NaCl, must be added , so that the extracted species are zinc chloride complexes not suitable for the sulphate electrowinning process actually used. However, electrowinning can be carried out from chloride electrolytes by using a diaphragm cell and DSA anodes for chlorine evolution (5). The purpose of this work was to prepare a suitable chloride electrolyte for zinc electrowinning by zinc extraction with TBP from an industrial bleed off stream and to experimentally study the whole process including both sections of solvent extraction and electrolysis. EXPERIMENTAL The investigations were carried out on a lab-scale pilot plant with flow rate of 1-2 1/h. The pilot plant essentially consisted of the following sections (Fig.l): - a battery of mixer-settlers ERIES AT-1 ( with 6-8 steps in the extraction stage, aad 4-8 steps in the stripping stage);

- an electrowinning cell with Al cathode and DSA anode (De Nora, Italy) separated by Nafion 120 diaphragm (Du Pont, Usa)? the anode compartment w^s closed and designed for chlorine collection. The anolyte ( "iaCl) was continuously recycled to the compartment after NaCl ma"

- an act • • >n treatment of the electrolyte from the stripping section •'••". .ce the TBP content to leas than 2 ppm.

Industrial -pent electrolytes were used containing (g/1): 14-15 Zn, 230-250 H2S04, 15-20 Mg, 1-3 Mn, and 4-10 rag/I Fe, together with CI and minor amounts of Cu, Co, Ni, Sb, Co etc.; tfaCl was added in various amounts to the bleed off .stream as a salting agent. The extractant was technical grade TBP containing Kerosene- (30-50% TBP by vol.); the stripping solutions were pure H2S04 solutions (25 g/1) or solutions containing various amounts of the exausted catholvte. All experiments were carried out for several months, frequently changing the operating conditions of the extraction section to prepare the most suitable zinc solution for the electrowinning cell, and the most favourable stripping system. Zinc deposits on Al cathode were stripped every day.

* Work supported by Ministero Pubblica Istruzione Italy.

295

HjSO,

Spent NBCI Electrolyte | 1 I

HT E o

Q,

. . , t r t 5 f Zn­itripping

§L

7T7^

Й [ So4v«m j 4 1TBP1 I

j l ^ Q

(to recovery J

Al I D.S.A.

Electrolyte (recycle 1

O^B ' Aqueous

raffinate

Fig. 1* Flowsheet of the pilot­plant

RESULTS AND DISCUSSION During a first set of experiments, carried out with 6 extractions steps and 4 stripping steps, as shown in Fig. 2, acceptable zinc extraction yields were obtained (87%). The extraction selectivity of TBP towards zinc was high: major impurities normally contained in industrial electrolytes were, in reality, not extracted; only few ppm of iron, which did not. effect zinc chloride electrolysis, were'extracted .

Exhausted Electrolyte (+35 g / l NaCI)

E X T R A C T I O N ( O / A = 1) Aqueous

Refflnate H o waste)

12.7 5.5 2 .9 2.5 2 .1 1.8 1.6

1 —- 2 3 1 5 6 1 —- 2 3 1 5 6 1 2 3 1 5 6

16.8 LOE

sol Kfcd

15.9 D.8 12.8

Solv recy

ent cle

5 . 7

4 . 1 3 *­4

22.2 Zn Soln. (to recovery) STRIPPING

ГО/А = 2 |

0 .0 H 2 SO„ O.SN

F i g . 2 . Zn­concent ra t ion prof i le of a typ ica l r u n w i t h 6­ i * steps

The extracted liquor, stripped by a sulphuric acid fresh solution, was treated with activated carbon and sent to the cell. TBP losses in aqueous phase were relatively low ( 0,06­0,06 g/l ), but a carbon treatment was all the sam^ necessary before the electrolysis. From the cell, the effluent electrolyte (containing 8­10 ?/l Zn) was sent to the first extraction step mixed with the spent industrial electrolyte.

296

The described procedure had the drawbacks of low zinc extraction yields,high zinc recycle (with the effluent electrolyte from the cell), and of relatively high TBP losses due to the greater amount of aqueous raffin»te. In an other set of experiments, the electrolyte from the cell was used as stripping solution; for this purpose, part of the electrolyte (40­«0«) was sent to the stripping section while an other portion was sent back to the cell together with the recovered zinc solution. Combining these two streams, the resulting 3n and acid concentration in the electrolyte to the cell was the most convenient. Electrolyte purge was provided to avoid build up of impurities, such as Pe and Ид. More steps (8+8 or 8+7t were necessary to reach good zinc extraction (Fig.3); zinc recovery yields of over 94% were obtained, leaving a residual zinc content of less than 1 g/1 in the aqueous raffinate, and reaching a Zn concentration of 14 g/1 in the solvent to be stripped. Advantages of such approach were the following: ­ more convenient utilization of the zinc solutions; ­ higher zinc recovery; ­ high purity of the electrolyte into the cell; ­ no sulphuric acid consumption; ­ lower TBP losses •> ­ lower cost of the carbon treatr. 2nt. The amount of the salting agent NaCl was varied during the tests in the range 35­50 g/1. The higher concentration (50 g/1) was the most favourable, giving zinc extraction yields between 90 and 94%.

Exhausted Electrolyte I +50 g/1 NaCl) 14.4 13.6 t i .7

Г Loaded solvent l ( 2

8.8

13.0 M.I 8.2

«.6 3.1

EXTRACTION (O/A = 1 I

6.1

5.5 3.3

1.6 0.7

Aqueous Raffinate (to waste I

1.1 0.8 3.» 2.3 1.1_t

1.7

0.4 Solvent

0.» recycle

35.0 22.7 ID.8 M.S 8.2 6.2 5.5 5.5 5.1 Zn Soln. Electrolyte Ito recovery) STRIPPING Recycle

(O/A = 2.21

Ftq. 3. Zn­conoentratlon profile of a typical run with 8­8 steps

rhe TBP concentration into the kerosene varied between 30 and 50 % by vol.: lower contents gave low zinc extraction yields, while higher concentrations were not interesting due to high viscosity, small Zn extraction increase, and high cost. Organic to aqueous ratios between 1 and 2.2 were used during stripping; 0/A=2 was the best ratio, giving a liquor which after mixing with the recycled cell effluent had the right Zn and acid contents for the electrolysis.

297

Considering the electrowinning section of the pilot plant, high current efficiency (92-93%) and low specific energy consumption (2.7-2-8 kWh/kg) were obtained during the best runs (5).

It should be pointed out that the current efficency is comparable with that obtained in the traditional sulphate electrolysis, and that energy consumption is appreciably lower.

A high quality zinc deposit on the cathode was obtained by using TBACL as additive at a concentration of 15-30 ppm, as suggested in Mac Kinnon and Brannen work (6). The overall zinc recovery from the bleed off stream was averagely between 85-90%.

In conclusion, it can be said that the whole process (i.e. solvent extraction, stripping by using the electrolyte from the cell, and electrolysis in chloride medium) looks promising. The evolved chlorine could be used to prepare ipochlorite solutions or for other purposes.

Preliminary cost extimates showed the advantages of zinc recovery by this way , due to the negligible cost of the raw materials (spent electrolytes). Moreover, the process partially solves the problem of waste disposal in zinc industry.

REFERENCES

1. D. Buttinelli and С Giavarini // CHIMICA E INDUSTRIA, 68, 563 (1981).

2. D. Buttinelli, C. Giavarini,and A. Mercanti //REAGENTS IN MINERAL INDUSTRY /Ed. M. J. Jones and R. Oblatt. IMM, London, 1984, p.233.

3> D. Buttinelli, C. Giavarini, and A. Mercanti //PROCEEDINGS OF ISEC 83 , Denver, August 19B3, p. 422

4. D. Buttinelli and C. Giavarini // CHEMISTRY & INDUSTRY. 6. 395 (1981).

5. D. Buttinelli, c. Lupi, and A. Riesenkamp:, Zinc electrowinning from aqueous chloride electrolytes //Journees d'Electrochimie '87 , Juin 1987, Dijon.

6. M.J. Mac Kinnon and J.M. Brannen //J. APPLIED ELECTROCHEMISTRY 12. 21 (1982).

298

ТНГ LXTRACUON OF CADMIUM CHLORIDES Ш1ТН BIWArtY EXTRACTAUTS , [11­19

A.I.Khalkin, G.L.Pashkov, V/. N. Dokuchayev, U.I.KuZmii, U . V. Serge ye va , :i. V.Protasova, K.S.Luboshnikova, I.Yu.Fleitlich, L.K.Novikov, Institute of Chemistry and Chemical Technology, Siberian Branch of Academy of Sci­ences of the USSR, Krasnoyarsk, Institute "Gidrotsvetmet", Novosibirsk, USSR

The technique of cadmium extraction from zinc sulfate solutions uiith simultaneous removal of chloride­ions from the solutions should prove a good example of the advantages of binary extraction.

This method makes use of the separate stripping of chloride­iDns and cadmium uiith binary extractant £\J.

The stages of extraction and stripping of cadnuum and chloridb­ions are described by the equations (the solution and polymerization effects have been neglected):

" ( • q ) + 4 C 1

U q ) + ( (

V»2S D

4(o) = < V >2 M C 1

4 (о) + S 0

4(aq> '» (

R

aN )

2C d C 1

4 ( o )+ 4 H A

< o )+

"0 H

" ( a q ) 5 ? 2 V A

( o ) *C d A

2 ( o ) , 4 С 1 Г а Ч ) + 4 H 2 0 ;

" A 2 ( o ) + 2 H 4 N A ( o ) * « H | a q ) * S 0 j j a i i ) — ( H f t N ) 2 S D 4 ( o ) + * H A ( o ) • C d ^ q ) .

The process is described similarly in the case of utilization of tri­alkylammonium carboxylate H,NHA. Chloride­ions are stripped with NaOH solution fully uihen equilibrium pH­value of aqueous phase is equal to 5­7.The rise of HA concentration in the organic phase extends the range of pH­values retaining cadmium in the chloride strip product solvent, but this process rises the consumption of sodium hydroxidB (Table 1).

Cadmium is completely stripped with solution of H­SG, uhen equilib­rium pH­value in the aqueous phase 2 is reached. At the stage of strip­ping cadmium can be concentrated in the strip product solvent up to 150­180 g/dm , I.e., 15­30 times compared to the initial, solution. The chloride­ion concentration in the same strip product solvent reaches 75­90 g/dm . Sodium hypochlorite can be obtained from chloride strip product solvent by electrolysis L/hiiecadmium is extracted from its strip product solvent either by electrolysis or by electrolytic precipitation of zinc.

The isotherms of the extraction of cadmium and chloride­ions uiith binary extractant from 2inc sulfate solutions are shown in Figure i. The sequence of procedure of exttaction of cadmium from zinc sulfate solutions obtained in the processing of load powders is provided in the diagram in Fig.2.

The composition of the initial solutions (given, in g/dm ) it as fol­lows; 60­80 Zn, 5­15 Cd, 5­10 Fe, 1­2 As, 0,1­0,3 Sb, 3­5 Cl", 15­25

299

For the industrial process the parameters and composition of the products obtained by this technique are given in Table 2.

The enoineering method h<as been tested in pilot scale /2/. The extraction of cadmium from the initial solutions into strip product solvent makes up 98.9­99.9%, the purification of zinc sulfate solu­tion from chloride­ions reaches 96­9855. The engineering process can be applied in the processing of sulfate­chloride solutions resulting from salt leaching of lead­zinc sulphide concentrates as well as in the reduction of cadmium electrolytes in galvanic production and the extraction of cadmium from the slimes of blast furnace production.

Table 1. The effect of UIK­2 (HA) concentration in organic phase on the stripping of chloride­ions with Ma OH solution (0 : lib Т И ) The extract composition: 0,6 PI R,N with the 10?S addition iso­octenol; 5.0 g/dm3 C d 2 +

; 13.5 g/dm3 Cl"

RjN:HA In organic shase

C

NaOH,f) pH Concentra t ion aqueous phase f g/dm RjN:HA In organic shase

C

NaOH,f) pH Cd C l "

1:D.5 0.3 4 . 2 1.90 6,7 0.5 6,2 0,60 13.1 0 .7 B.8 0.02 13.5

t : 1 0.5 5 .0 1,80 1 1 , * 0.7 6 , * 0 ,60 13,1 0.9 8 . 0 0.03 13,5

1:2 0 .8 6 .9 0 , 5 0 13.3 1.1 7 .6 O.OS 13.4

1|2 8 . 1 0.03 13.S

Fig.1. The isotherns for the extraction of Cd (a) and those for chloride­ions (b) fro» zinc sulfate solutions (2f«nS0 4;0.znH 2S0 4) with binary axtractant: a) 0.6 П R­jNHCl and 0.6 ft HA (UIK­2)

b) 0,3 И (R NH) SO and 0.6 П HA (WIK­2)

300

Circulating extractants (R,NH) 2S0. + HA

Initial solution

Extraction of chlori­de­iona(R )

Extraction of Cd(R 2) Org.

°"ase. J_ Stripping

of chlori­de ions(R 3)

Org. phase,

Str ipp ing of Cd(R,J

Raff inate I I Extract for obta in­ing 2n

organic phase

C l ­ s t n p product solvent

Extragent regenera­t i o n (B s )

Cd­st r ip product solvent for obta in­ing Cd

H1O_ For obtaining NaOH For obtaining Na hypo­solution for stripping chlorite by electroche­

mical method

Fig.2. The technological diagram of the extraction of cadmium with binary extractant from zinc sulfate solutions with simultaneous re­moval of chloride­ions

Table 2. Technological parameters and composition of the obtained products

Stage Parame ters Products Concentr ation,g/dm Stage

Phase ra­tio 0:Ш

The num­ber of steDS

Products

Cd Cl~

R

i 1:3 5 Raffinate II 0,1 0,2 R

2 1:3 U Heffinate I G.I 1­2 R

3 5:1 2 Cl­strip pro­duct solvent

0.02 75­90

R

< 10:1 2 Cd­strip pro­duct SOlvBnt

1S0­1B0 0.5­1.0

R

5 3:1 3 Liquor 0.05 1­2

References 1. G.L.Pashkov, A»I.Kholkin, V.V.Sergeyeva et al. //Tsvetnyiye metally.

1966. N 2. P.29­30. 2. G.L.Pashkov. A.I.Kholkin, V.U.Sergeyeva et al.//Tsvetnyiye metally.

1966. N 3. P.34­36.

301

SOLVENT EXTRACTION TECHNOLOGY OP COPPER FROM LEACHIHG " 11­20

SOLUTIONS 1. 1 V.V.Tarasov, V.P.Travkin, A.M.Mirokchin, V.S.nijanov, L.I.Ruzin, A.A.Picbugin.V.M.Martinov, All­union Renearch Institute of Chemi­cal Technology, Moscow, USSR

Rapid development of extraction technology of copper is considerab­ly connected with a large range of selective extractants for it bet­ween which the reagents of oxyoximes class are most widely used in in­dustry fl,2 j . The extractant alkylbensophenoneoxime ­ ABP, making possible the selective extraction of copper from complex solutions is created in the USSR.

The Influence of nitrogen­containing additions such as: alyphatic amines and salts of quaternary ammonium bases and neutral phosphoror­ganic compounds ­ IBP, di­2­ethylhexylmethylphosphonate (D2EGMP) and trialkylphoephineoxide (TAPO) is described In present work.

It has been shown, that the acidity of initial water solutions containing copper can be rised. In this way pH of aqeous phase incre­ases from trialkylamine (TOA) to the salts of quaternary amnonium bases (TAMAS) and in the ranges TBP^­D2EGHF<TAP0 in accordance with basicity of reagents in it. The addittion of above said reagents doesn't lead to the decrease af selectivity, while kinetic parameters of the process are improved.

The fact of effective stripping of copper from organic phases con­taining mixtures of ABP with nitrogen ­ and phosphorus containing reagents by solutions of 100 g/1 H2 S 0

4 i s °* S e e a i interest.

Table. Stripping of copper by solutions of sulphuric acid С0:А»1:1;г"» 5 min; t » 22°C; copper concentration in organic phase

2,1­2,2 g/1) Concentration of a^SOe i n s t r i p p i n g s o ­l u t i o n s

Eff i c i ency of copper e x t r a c t i o n , % Concentration of a^SOe i n s t r i p p i n g s o ­l u t i o n s

0.21m ABP +0.18m

TOA

0 .21s jSTT» + 0.19m

ТАИАЗ

0. ?1ra ABP +0.21m

ГВР

0,?­>таАВР +0.19m DZEGtlS

0,?lm ABP + 0.18m

TAPO 203 180 151 130 112 91 73 51 32

8 1 . 9 7 5 . 3 7 1 . 3 6 8 , 2 51.6 45.5 30 .8 22 .8 16.1

83 .0 79.7 77.6 7 6 . 3 73.7 68 ,0 54.6 45 .4 30 .2

8 1 , 8 79.0 7 7 . 3 76 .2 74 .9 69.8 59.7 51­0 35 .6

82.5 78 .4 75.5 74 .5 73.0 67.1 51.8 42,0 2 8 . 7

80.9 77.6 74 .5 73­6 72­5 67­7 55 .2 43.8 32­9

8 2 . 8 78.5 7 7 . 2 75.8 75­0 69 .8 59.5 48 .7 36 .2

It ia known that hydroxyoxlaes are slowly­kinetic extractantо of oopper and this is a considerable detect of that class of reagents.

302

The addition of reagent being an extractant with higher kinetic para­meters is one of the most effective methods in order to improve copper extraction. For example extractant Ш 64 Я is a mixture of slowly­ky­netic alkylbenzophenoneoxime and aliphatic oxyoxime Ы Х 63 possesing high kinetic properties. The application of ortho­nitrosophenoles as kinetic addittions is described in present r?"*v. т+ у.»ч beer determi­ned that the extraction rate increases in 5­10 times in the presence of 0,1­1% of ortho­nitrosophenole. Probably the mechanism of cataly­zing influence of ortho­nitrosophenoles is like the mechanism realized when LIX 63 is used as a modificator of LIX 65N. So, rapid reaction uf copper extraction is taking plase at first with following exchange of ortho­nitrosophenole's extreme raical for tne nydrooxime one.

It should be noted that the saturation of org nic phase and acidity interval reguired for effective ex.raction and also the effciency of copper stripping by so­utiona of sulphuric acid 150­170 g/1 are not changed by using mixture of ABP wii.h ortho­nitroaophenole as an extrac­tion agent. Tne Selectivity of extraction of ferrou, zinc, nickel and cobalt doesn't change when this mixture is used.

The laboratory investigations of copper extraction from solutions with low acid content were carried out using micxture of 10% alkylben­zophenoneoxime ABP and 0,1% 2­(ot,o£ dimethylbenzyl)­4­methyl­6­nitroso­phenole.The extraction process and stripping of copper were carried out at o/a«1:1 and 10:1 in 2 and 3 stages using one minute agitation tijae. The extraction of copper into raffinate after stripping has reached 96,756 at initial content of 1,32 g/1 and pH 1,98. Using 10% ABP without ortho­nitrosophenole the extraction level of 73,5% was recieved.

Pilot­plant tests of copper extraction by ABP from the solutions after leaching of balaced out copper ores were carried out in accor­dance with technological line: leaching ­ extraction ­ electrowiiming. The productivity of this plant using 15?» solution of ABP in kerosene was 100­150 лг/day. Pulse columns with "KHIMZ" plates and mixer­set­tlers were used for extraction.

The analyses of the results shows that copper extraction largely depends on intensity of pulsation, flows ratio and temperature of solutions.

For example the increase of pulsation from 900 up to 1200 mm/min and detention of dispersed phase from 20 up to 325? leads to growth of copper extraction from 30 up to 6S% other things being equal(Fig.j,

The change of stream ratio from 0,5 up to 1,3 leads to increase of extraction from 40 to B8%. The growth of temperature from 22 to 28°C alao results in its inorease from 40 up to 70% that oan be explained, probably, by the improvement of systems kinetic parameters.

The comparison of main technological parameters of copper extrac­

303

«on in pulse oolmm with "KRIMZ" plates and In mixer­settler shows that the application of pulse ooltmns allows to decrease the load of organic phase by 20% and the losses of extractant.

r Dependence of the detenting capacity of pulee column on the extraction efficiency Diameter 0,9; effective height 5 m; aq * 5 m 3

/h; o:a > 0.9; t .= 27°C

Whereas the equipment of "mixer­settler" type is more safe and simple in work. Thus the choice of necessary equipment can be made only after some comparative calculations. One the base of this tests an optimal parameters of copper extraction in pulse columns were de­termined.

Extraction: frequency of pulsation ­ 1200­100 mm/min; summary spe­3 2

cific load ­ 16­20 m"7m 'hour; ratio o:a«1. Stripping: frequency of pulsation ­ 1800+ 200 mm/min; summary spe­

cific load ­ 15 лг/т .hour; ratio o:a=20:40. To reach extraction efficiency of 90­95% an overall height must be

above 12­15 m. An extractant losses with stripping solution 0,06­0,08 8/1.

An optimal parameters of copper extraction in "mixer­settler" were determined as following.

Extraction: agitation time ­ 3 min; line speed of agitation ­ 300­­350 m/min; summary settler load ­ 2,5 m­fym.h; ratio o:a»1.

Stripping: agitation time ­ 5 min; line speed of agitation ­ 300 m/min; summary settler load ­ 2,0 лг/m .h.

An extractant losses with raffinate 0,1­0,18 g/1; with solvent ­0,07 g/1.

Strip raffinate contains (g/I): Cu 39­41; HgSO. 180­200; Pe 0,31; Sn 0,3; Bi 0,003; Hi, Fb, So, As no more than 0,1.

Copper from the cathodes after electrowinning contained the follow­ing contaminants (%): Bi 0,00005; So 0,0004; As 0,0002; Fe 0,0006; Si 0,0004; ТЪ 0,0048; Sn 0,00006; Zn 0,0003. Lean electrolyte con­taining 20 g/I and 200 g/1 HgSO. was retimed to stripping. ABP expense by the extraction technology of oopper from solutions after leaching was 10­12 kg/It of cathode oopper.

Referenoes 1. Ple t t D.S.//Chen.lnd.1977. H 7. P.706. 2. Xatsutoahi J.//B|y4rometallurgy. 1984. Vol.13, M 1. P.73.

304

x'Tic:; FHt-czj'JiiiG OF •XI­JTAMIIJATED СОГРЕЙ ELECTROLYTE 11*21 I

. . ' *­.­­", A.V.JtryapHov, Khimiko­metalluv£­ichesky Institute , :. r.H::rJ:h::k-i' J.'JR, "ariganda State University, Karaganda,

rreccesir.f о Г lcvt-rre.de polyrcetallic ore rostO ted in considerable •:с2jr.\<'iri\ior. ox r.­ir­ful impurities ­ Pc, An, ib ­ in recycled soluti­on at rone copper­electrolytic plants, in fall of tcchnical­economi­;:•:] indices of electrolytic refining and in lowering of a product f. lity. As а rule purification works are not able to cope with the needed volumeи of solutions because of low capacity of traditional с chords o' neutralization by copper pellets and deep electrolytic decepperization. The processes proceed under hard sanitary conditions with hif,h material and energy expenses, copper and nickel salts being c: poor quality.

Hew hl.yi efficient technology of stepwise extraction of valuable components from strong acid copper electrolyte solutions has been de­veloped and tested. The first 3tep is the step of oulphuric scid se­paration with tertiary alkyl amine (alkyi C7­C„) solution in benzene 'in acid salts such es/(A.TiiUp30./_.H? 0^. To increase extraction pro­•

l :•' solubility it was proposed to rid:l to the extraction agent alky­lated phenols as more effective and cheaper amonf; known up to date solubilization additives­high molecular alcohols Ц 1 J,

Sulphuric acid reextraction proceeds with hot water as it is known that heating essentially improves the process as a result ol exother­mic reaction. By calculations and experiments in continuous regime on a pilot­scale plant made up of extractors such as mixer­settler with 4.5 l/h production over phase зит it was found that 5­6 steps of ex­traction with the mixture of 1 mole/1 of trialkylamine with the addi­

I tion of 5% alkylphenol (organicrwater volume ratio (3.5~4):l) were quite enough to decrease electrolyte acidity. After washing of the extract with water at the same phase volumes at 60°C regenerated sulphuric acid 125 g/1 and middle amine salt returned for extraction were obtained.

After hydrolytic refining electrolyte cornea in contact with the solution in кетозепе of cheap commercial extractant­naphthenic acids (molecular mass 230­250) tc which alkylphenols with synergistic addi­tive were added to improve ita extraction properties С 2 J» As a re­sult copper distribution coefficients are found to increase owing to alicylphenol molecule substitution for solvated naphthenic acid ones and more lyophilic mixed complex formation. As a result the time of segregation of phases reduces and useful capacity of extractant in­

« creases. Different methods of physico­chemical analysis allowed to ' determine the composition of the formed mixed complexes. Optimum con­

4 20, 3IK. 382 305

ditions of copper and nickel selective separation by mixed extr&ciant are determined on the basis of the calculations of entropy of mixinr, value which takes into account component redistribution between phascu

The developed technology was carried out in continuous regime on a pilot-scale plant which consists of tanks and extractors such as mixer -settler with 120 1/h output over phase sum. Mother solutions (pH 4) containing 50 g/1 of copper and 40 g/1 of nickel were used as midel ones. The schemes of crossing current and countercurrent flow were tested taking into consideration the high metal concentration in solu­tions. Practically quantitative extraction of copper was provided in four steps of extraction (pH 4.5-5.0), while nickel remained in raf­finate. Copper saturated organic phase was reextracted by equivalent amount of sulphuric acid (125 g/1) in two steps, tho extractant being recycled. Nickel (pH 6.5-7.0) was extracted by analogy, pK correction was carried out by addition of the soda solution (6%), All the proces­ses were characterized by stability without accumulation of detrimen­tal impurities and intermediate phases. Reagents rate turned out to be close to stoichiometrical one, losses of extractant being insigni­ficant,

Metals distribution in the products of technology {%) (entered with initial solution 100%)

1 Copper

extraction Cu Ni Nickel

extraction Cu Ni Technological extraction

1 Copper

extraction Cu Ni Nickel

extraction Cu Ni Cu Ni

Crossing current flow scheme Copper vitriol 56.25 - I 9 6 # 6 3

Mother solution 40.38 30.52 J Raffinate 3.37 69.48 Nickel ->

vitriol - 39.59 Mother ( 6 8 " 0 9

solution 3.37 28.50 J Raffinate - 1.39

Countercurrent flow scheme Copper -. vitriol 73.52 - I Hother J 9 8 ' 8 9

eolution 25.37 2.93 Nickel - 57.45 "1 Rafinate 1.11 97.07 vitriol >95.98 Mother I solution 1.11 38.53 J

Raffinate - 1.09

306

high­class commercial copper and nickel sulphates out of vaporized reextracts have been obtained, balance calculations give evidence of nigner counterflow scheme efficiency by which technolfgical copper extj­ection into end product amounts to 99#» nickel ­ 96%.

Combination of the extraction and electrolysis with insoluble anodes is more advantageous and economic method of metal extraction from the solution. In these schemes evolving at the electrolysis acid is consumed for saturated organic phase reextre.ction resulting in con­centration of recycled electrolyte with precipitated metal. On this scheme using naphthenic acids nickel has been obtained.

The tests were carried out in the cell with cloth countercurrent diaphragm, separating cathodic and anodic spaces to store up sulphu­ric acid in the latter. The composition of nickel reextracts met the requirements to obtain high­grade metal. Great role of organic impuri­ties , pH and electrolyte hydrodynamics in dense cathodic residue for­mation was determined.

pH fall of accessing catholyte up to 2­3 results in decrease of possibility of hydrateforraation and cathodic nickel cracking and in facilitating of natural solution convection on account of cathodic space expansion up to 4 cm. Electrolyte refining from organic impuri­ties was carried out by filtration through the column with activated coal BAU with the rate up to 20 volumes per hour amounting to 76 volu­mes. The coal has been easily regenerated with 1­2 volumes of sodium solution and water. Multiple long­term (3­5 days) experiments on the scheme "extraction­reextraetion­electrolysis" allowed to obtain ca­thodic nickel specimens of the mark H­I from spent copper­electrolyte solutions with 85­96% current efficiency and showed that detrimental impurities storage in the nickel electrolyte and on the coal did not take place.

The developed technology is based on the application of cheap and non­deficient reagents! the processes are stable and easily automated. Utilization of ammonia for electrolyte neutralization and recycle of obtained solutions for sulphate storage gives the opportunity to va­porize waste waters up to the products used in agriculture * Expected economical effect at Balkhash mining and metallurgical plant amounts to more than 300 thousand roubles.

References 1. Avt. evid. Я 611877 (USSR). Extractant for sulphuric acid extrac­

tion. /Buketov E.A., Stryapkov A.V. et al. B.I., 1978, N 23. 2. Avt.evid. N 542771 (USSR). A process for separation of copper and

nickel sulphates. /Buketov E.A., Stryapkov A.V. et al. Б.1., 1977, N 2.

307

..:: •'.::: • v.:n:(>'ir _ ';. Angelov, G. Kyutchouko­», L. Boyadzhiev, D. Elerkov, Institute ol Chemical Engineering, Bulgarian Acader.7 of Sciences, Sofia,Bulgaria Introduction «The waste flows of numerous technological processes

contain a significant amount of copper compounds. In printed circuit board production,for example,about 65& of copper cover is dissolved which makes its recovery worthy­Copper isolation from the spent etch­ants is complicated because of the multlcomponent composition of the solution.Some methods '­pply treating with chemicals. Generally, low quality product is obtained along with many chemical wastes |l~|. j

When electrolyse directly spent etchants many attendant processes take place, harmful gases are released,the process is unstable and the yield get down|2­4!. The above methods веек to recover the metal and not to regenerate wasteless the etchant in order to use it in a recyc­le. If apply liquid extraction methods|5],lt is possible to solve both problems. Copper can be partially extracted from the etching solution which is returned in the production cycle. By stripping the solvent, an acid solution of copper salt is obtained, from which copper can be easily recovered by electrolysis.

The aim of this work is to study several extractantB in order to test their ability for selective extraction of copper from spent etch­ants and to develop an appropriate continuously operating installation

Extractant studies.The main extractant properties needed for a suc­cessive resolution of the particular problem are:

­Selective extraction of copper from the rnultlcomponent mixture, 1. e only the reaction fcu(.NH,) 1 Cl2+2RHxR2Cu+2HH.Cl +(n­2)HH, takes place

­ High copper loading ­ It is favourable if the stripping could be carried out with 2­3;:

B„SO. according to the reaction R„Cu + H.SO.Ti CuSO. + RH in order to 2 4 2 2 4*^ 4

obtai a standartised solution for electrolypis. ­ Paet kinetic of both ргосезвев. ­ Low viscosity and good phase separation.The last two requirements

are connected with the maximum flowrate and the volume of the contact and phase separation units.

The laboratory studies for choosing appropriate solvents were carri­ed out with real spent etchants. The qualitative results for the gene­ral properties of 4 extractants are given in Table, Full quantitati­' ve results are listed 1пГб]. Evidently the most fitting are the prop­erties of L U 54, which was used as a solvent in the process carried out in the installation described below.

installation .Its scheme ie shown on f ig.1. According to the labora­ ~"' tory resultB, the extraction and stripping processes could be realized • in one step apparatus under intensive mixing for a good phese contact.

308

Extractant Copper loading Select ivi ty Extraction

kinetic Stripping

kinetic Phase

separat .

Fatty acids C13~°18 u p t 0 1 0 * Acorga P5100

up to 10% HS­LII­65H up to 15*

50 *

ЫХ 54 Ю0Й

+

XX

+

+

X

+

+ +

+

+

+

+

+

+ no stripping + at room temp.

* I * iExtraction of ammonia takes place. Soaps as well as etable d i s p e r s ­

ions are formed. X I Hign 0/W phase ra t io i s needed to reach the required ext­actuon.

In the Ins t a l l a t ion two similar contact units i and I I , 100 mm in diam. are used for extraction and stripping operations. Each of them Include a mixing zone(6) with a turbine stirrer(.7J and s e t t l i n g zones(.5,9) i n ­

corporated in the apparatus body. The exit flows pass through adhesive phase separators00,11) for fine phaee separation. The feediui J(4) is direc t ly in the mixing zone. In the apparatus I a reduction of copper contents in the spent etchant(3) takes place. In fact t h i s i s a rege­

nerat ion of the etching solution which i s returned in the etching bathe(1) . The organic solvent containing the extracted copper(12) i s stripped In apparatus I I by sulphuric acid solution(15) coming from the e lec t ro lys is baths af ter copper winning. The regenerated ext rac t ­

*Чк. 1 309

ant is fed into 1 for a new extraction cycle. The spent stripping flow (CuS04)+ KjSO^}, tank(16), is used for electrowinr.lng of copper.

This brief description gives an idea for the basic advantage of the applied principles­all flows are used in a closed cycle and no waste flows exist. Furthermore, the installation is simplified in constructi­on and easily controlled. As the stripping column is mounted in a '%h-

er level over the extractor, the solvent iB fed by gravity,avcidinb a feeding tank and a pump. Thus the solvent flcwrates in two apparatuses are selfsynchronized and no special mutual regulation is needed. The flowxate in two columns is specified by c; ntrolling a single pump(l3).

Determination of optimal processing conditions.For the considered particular case the efficiency of the units is provisionally defined as

ng/l copper extracted from the etchant(extraction case) or from the

organic solvent (stripping .лаве). The extraction efficiency E_ under various mixing intensities W is given in iig. ,.. u±**­..a± nyai­oivn ..­Ac reglm­eefgoad mixing and no phase separation problems) are created when the stirrer peripheral velocity W is 1­1,2 m/s. Significant phaee retention and unata'. le regimes , are seen when get over ;. 1,25 m/s ( f)

The extraction efficiency vs etchant flowrate F, resp. total throughput q is seen from fig.3.under optimal mix­ing regimes and 0/W ratio 2, chosen

, 4 «a 40 'jf

*< У, -30 y ^ ( /

?n ' \ -

by laboratory tests. It follows that throughput over 5­6 m'/ш2

^ means low efficiency regimes due to the shorter residence time.

Pig. 4 shows the extraction efficiency when vary 0/W ratio ф . The throughput and mixing regimes are fixed at optimal values according to л

previous tests. When ф is less thar 2, the extracted amount of copper J

310

in not enough. When ф is greater than 2,2(jU the apparatus is overloaded and flooding ef­fects are ;,een. Thus, the value of ~ 2 seems reasonable for apparatus operation and ensu­res the extraction of required quantity of copper.

The efficiency of stripping unit Eg at various mixing regimes is plotted on fig. 5. The eystem solvent­sulphuric acid allows more intensive agitation since phase separa­tion problems arise at peripheral stirrer velocity over 1,43 ra/s( f/j.

In table 2 results for stripping effici­ency at various phase ratio is given. It is worthy to operate at high­er 0/W flow ratio since lesser stripping solution is spent. But its

14 Ulmlsl Fig. 5

output concentration cannot over­come 55 g/1 H2SO. because crlsta­lleation takes place which might block the pipelines. Thus, suit­able values of phase ratio are 1,1­1,3. The flowrate of organic phase is the same for both units. The stripping solution flowrate is adjusted by choosing phase ra­tio as stated above.

Table 2 Phase r a t i o , 0/W

Stripped quantity,

g/1

Output cone .of s t r i p . so lu t i on ,

g/1 2,25­2,35

1,45 1,21

18,5­20,5 !8 17,5

54­61 55 44

1,11 17 35 1,07 16 30 0,83 14 25

Conclusion A solvent extraction procees for recovery of copper .from a lka l i so­

lutions with high contents of ammonia and ammonia s a l t s it' developed and continuously operating i n s t a l l a t ion i s constructed. Studies for Choosing of copper select ive extractants are carried out *s well as study on the optimal apparatus performance. An extraction str ipping technology with wasteless recycling of technological flows i s descr i ­

bed. I t 1B auccessfuly teBted for continuous regeneration of i ndus t r i ­

a l spent etching solut ions .

References 1 .Beeby K.E.,bee S,//Chem. Process. 12­16(1983;. 2 . tonstantonros E.,Hoelmueller H.//Ger.0ff. BE 3,305,319 (1984). 3 . Hll l i s M.R. UK pat .appl . GB 213 3806A (1984). 4 . Manning M.J.,Melsheloer S.J.^nd.Eng.Chem.(fund.;.22.311(1983). 5. Anderssen S., Reinhard H. / / Handbook Solv.Extrn. Wiley,N.T.751.1983. 6. Technical report N 1604, Inst.Chem. Eng.,BAS,1987.

311

> nlMiilMiP PKLC1P! TATU).\­l:XTRACTIu\ Mi.'liliMi Mil­: IkL.UINi. M.WCAXLSL­ M T R A T I . SOLUTIONS Mill I M P U R I T I L S <i| \dN­I I !<ki>U

I'.l'.ucr.ilov , M.Kazakova , Sc ieniific .\ Pro JUL" t ion i.ni erpr i Mining an J Proccs:" i n.c К­ г reus Met л I > Ore> , >u i i :i, i л 1.. и* i a D.llaiij lev , iHilyaiian Ли a J и my <JL ;>L K­IIL'L :•, ii... , ..(..

Krciiiikovtsi ore Jt'pos i t s слп be с ha r.'ictcr i :e.! a.­> In­. ­i;ni.)c po! >"'K ­t a 1 ] ic i ron ores with h igh content of manganese . iiecau>c о . the prfSi.T:. of copper, lead , z inc , mercury, a lka 1 inc­ca rtli atul n o b U met a 1 > , uhr •­. contents vary depend ing on the loca1ity, the hydrometal 1 urg ica 1 me t ':..<:•.

0 Г treatment prove to be the most с f Г ic Lent ones from t conomica] point с f v icw. Leach inti by sulphur i с acid III, nitric ac id I ­ 1 or comb inc J roasting and water leaching 151 were proposed as an initial stage in technologies of this type. Further utilization of manganese from the

1 each ing solution has been subject of a number oi" invest i gat ions 15.4, Ь.bl involving cither extraction processes U . M or purification by cementation and precipitation.

Л sui table method for product ion of high­grade iron concent rate con­sists of leaching with diluted nitric acid 1:1. The solution in additie:: to containing Mn conta ins also copper, zinc, lead, barium and other me ­tals of certain practical signii _­ancc. In the majority of cases tin­ i r concentrations arc quite low compared to those of manganese, and the Ji­rect application of extract ion processes for their separation requests scrubbing stages and large sized extraction equ ipment.

Another approach that could prove to be successful consists in conse­cutive alternation of precipitation and extraction. This allows to dimi­nish considerably the volumes of the treated solutions and respectively the equipment used and extractant losses during the technological pro­

cess. The aim of this work is to determine whether or not a combined precipitation­extraction method can be realized, thereby providing a manganese solution amenable to thermal treatment and allowing the reuse of the valuable metals.

The experiments were carried out by nitrate solutions, containing: 70­75 g/1 manganese, 0.75­0.S3 g/1 copper, 1.5­3.b g/1 lead, 0.15­0.16 g/1 zinc, 9­30 g/1 calcium, 4­5 g/1 magnesium with initial pH = 2.1­­.3. The test programme included experiments with model and models of real solutions. The organic phase consisted of S y% Lix 65N, 20 V% di­2ethyl­hexvlphosphoric acid (D2EHPA), with addition of 5 V% tributylphosphate as a;, emulsion inhibitor. All experiments in extraction were performed in AKUFUE 110 apparatus combined with a system for automatic control and pH regulation. The results obtained characterize the behaviour of lead, copper, zinc, barium, calcium, magnesium and manganese during precipita­

312

n­23

СоМэд MglNOfc

HjSCt

ti

CuS<^

^ ОЛЬ

СиЮн)г

PblDHl,

Co'!H)j

j B2EHRA.

Fig. 1. General Technological Flowsheet

ion and extraction. Fig. 1 shows a general technological flowsheet roposed o n the basis of the experimental results. The first stage of this flowsheet consists in precipitation of barium

s BaS04 with addition of concentrated sulphuric acid in amount 1/1000 f the initial volume. The precipitate obtained contains about 40-50S 'bSO. and the data for the concentrations of difft he solution after filtration are shown in Table. 'bSO. and the data for the concentrations of different metal cations in

Concentrations of metal cations in the filtrate solution (g/l)

Mn Cu Pb Zn ; Ca Mg

Operat ion

P r e c i p i t a t i o n by H,S0, up to pH = 0,62 70 0.75 1.5 0.15 30 4

P r e c i p i t a t i o n by lime milk up t o pH = 6,18 49 - 0.02 0.02 13.6 3

Solving the hydroxide precipitate by 2SI h^SO,, 39 3.8 0.014 0.6 0.008 0.43)

313

By addition of lime milk to the filtrate solution up to pH = 6.2­6.3 precipitation of all accompanying non­ferrous metals (copper, lead,zinc as well as a large part of the present calcium) is obtained.

The manganese solution, after removing the precipitate (see Table 1) is amenable to thermal decomposition.

The precipitate, containing hydroxides of copper, lead, zinc, calci­um and particularly of manganese, was treated by 25 V% sulphuric acid. In this manner PbS04 and CaSO^ remain in the precipitate and MnSO,, CuS04 and ZnSQfi go into the solution in concentrations as shown in Table.

The volume of the obtained solution is ten times smaller than this of the initial one. It was processed by extraction to separate and reuse the contained metals. A route that can prove to be successful is the separation of copper and zinc from manganese, the purified sulphate manganese solution being amenable to thermal dec imposition.

On the basis of the results obtained by Ritcey and Ashbrook 131 as well as results from performed experimentsi the extraction of copper prior to zinc­manganese separation is very suitable.For this purpose 5 V% solution of Lix 65N in normal paraffins C­I­­C­IC w a s used, and the equilibrium value of pH was maintained within the range of 1.85­1.95 where the separation factors are A c /Zn 5 and fi Cu/Mn 16 respecti­vely. Because of the low distribution coefficient of manganese m M =0.5 the two scrubbung stages of the extract allow its complete separation from copper, which can be utilized after stripping with 20 V$ sulphuric acid.

Figs. 2 and 3 show :he relationships lgm = f(pH) for the cases of zinc and manganese rec3very by 20 VI D2EHPA and 5 V% D2EHPA + 50 V% fatty acids.

.Щ. lg» СрЮ Fig. 3 . lgn f(pH)

314

It is clear that maximum separation between zinc and manganese in the first case is realized at pH = 3.3­3.5 where the zinc concentration is lower than ID mg/1 and the separation coefficient J£ 7 n / . M n = 135­198. However, regardless of this value, due to the high distribution factor of manganese, m ­ = 1,6 a significant amount of it goes into the ex­tract and can not be efficiently scrubbed.

An approach for separation improvement consits in application of the so called mixed extractants, in the present case D"EHPA and a mixture of fatty acids C..­C.g. The latter recover 2inc and manganese at quite higher pH, the pH values of semiextraction being 5,15 for zinc and 5,65 for manganese. Their introduction to the system should move apart the ranges of extraction of these metals­ In fact the replacement of D2EHPA by a mixed extractant on the basis of an ijiu^strial mixture of fatty acids C,4­C.g­oleic fraction + D2EHPA shows that the best separation of the two metals is obtained at considerably higher values of pH, i.e. pH = 4,20­4.40, the separation factor Q 7n/Mn D e

*n

£ much lower ­ 44. Hpwever the use of this extractant is justified because of the lower distribution coefficient of the manganese, Tl„ = 0,95, which allows its complete separation from zinc by four scrubbing stages of the ex t га с t.

The obtained MnSO. solution is suitable for producing high purity manganese salts.

It can be concluded that the general flowsheet, given in Fig.1 makes possible to obtain purified manganese solutions and at the same time assures the reuse of the valuable metals, accompanying the manganese in the initial solution.

References 1. Qazi М.АУ/Hydrometallurgy of Low­Grade Iron­Manganese Ore, Dept.

Mining and Metallurgy, M. of Alberta,Edmonton,Alberta, 1967. 2. Videnov N. //Cent.Inst.Chem.Technol.,Report 111,1984. 3. Ritcey G.M., Ashbrook A­V.Solvent Extraction Principles and Applica­

tions to Process Metallurgy. Metallurgy, Moscow,1983. p.155. 4. Ritcey G.M., Lucas B.H. Can.Met. Quarterly 10. 5. Guerzilov P.,Hadjiev D. Purification of manganese solutions contai­

ning copper, zinc, lead and other impurities by l iquid­liquid ex­

tract ion//Proc. 4th Med.Congr.Chem.Eng. Barcelona,1987.

315

RECOVERY OP HYDROCHLORIC ACID PROK ACIDIC IRON-CONTAINING EFFLUENT BY SOLVENT EXTRACTION OF LONG-CHAIN ALKYLAMINES

Dongxiu Sun, Hinbo Chen» Tianzhu Jin, Hai Guo, Fuan Liu, Nlng Xu, Jinguang Wu and C-uangxian Xu , Department of Chemistry, Peking University, Beijing,China

Hydrochloric acid is widely U3ed in rust-washing process in eteel factories and machine building plants. A large amount of acidic effluent from these plants containing various amount of iron in 2-3 M hydrochloric acid causes a serious pollution of environment. In order to recover the residual hydrochloric acid in the effluent and avoid the environmental contamination, high-temperature baking treatment of the acidic iron-containing effluent is currently used in large- scale plants. Hjwever, high -temperature process losses its feasibility in medium and small plants for its high investment and huge energy consumption. Therefore, we have developed a solvent extraction process by tertiary long-chain alkylaraine for the treatment of the acidic iron-containing effluent from rust-washing process, which is an easily feasible and economic method with high efficiency.

The acidic iron-containing effluent is a dark yellow-green liquid and its typical composition is 102.8 g/1 in Fe(ll), which makes up about 99Ji of the total amount of iron, and 2.60 M in hydro-chloricacid. Usually, divalent iron cation in aqueous solution can only be extracted by chelating extractants with low capacity as well as poor applicability. The Fe(II) in the effluent is then aubject'to oxidation to Fe(IIl) by chlorine gas or concentrated hydrogen peroxide. Experimental result indicates that 99# of the Pe(II) in the effluent can be oxidated in 2 hrs by bubbling chlorine gas at room temperature. Fe(III) can be extracted by many organic extractants such as quaternary and tertiary alkylaminos, neutral and acidic organo­phosphorous compounds, ethers, ketones and alcohols from HC1 solution as indicated in Fig.1. However, only tertiary and quaternary long-chain alkylamines show high distribution coefficients for Fe(IIi), Dp e(ni)i i n sufficiently wide range •of"concentration of HC1 and are used in the subsequent

* To whom correspondence should be sent. ** Project supported by National Education Commission.

11-24

316

i. Effect of concentration of HC1 in TuToue prtaae on dletrlButlon coefficient f trace amount of Ve(XII) by various ex­ractants (30 v/o In keroeene). . Tnbutylphosptaate; . n­Trloctylamine; . эес­Octanol; . N­263: . Methyl leohutyl ketone; . ?­507, ^oiaaerclal product equivalent

to 2­ethylhexyl(2­»thylheJcyl)phoipnonlc acid;

. iaopropylether;

. P­215. equivalent to dl(1­methylheptyl)­

phoapharlc acid; . P­204, equivalent to HDEHP

* — • / " * " "~­*. 2 * — • / " * " "~­*. 2

, Ъ 10

1 v / /

1/

-Г 4 ,6

Y

9 0 . 1 /

n ­2

laeij, и

experiments, rixperiraental results show that extraction equilibrium of Fe(III) can be established within 3­5 mln. for both Я­235 and N­263, which are the commercial products equivalent to Alamine­536 and Aliquat­336 respectively, and Fe(III) is prob.ifc'.y extra­

2 3­cted in the form of a anion Feci,­ or FeClg showing & very strong absorption peak at 381 cm" (Fig.2).

D

j e(III) * е с г е

&зез dramatically аз the concentration of iron(III) in aqueous phase increases (Fig.3). Ao the concentration of N­235 or N­263 in organic phase increases from 30 v/o to 45 v/o, D

p e(xil) increases from 0.827 or 1.10 to 2.02 or 6.00 respectively.

?tg. 3. Far lafrared absorption spectrin oj iron(III) In the phaacof N­235/»­octanol/k«rOBKM

317

л 5

г

^ °"5

в

Ч 1 0.5 1 5

. . .

rift. 3 t Kfr»ct е>Г concentration of 7»( I I I ) t t equi l ibr ia» ID « Q U M U pb*M on l t a own di s tr ibut ion co»ffi~t^Qt for A. ?0 ж/о M­Z35/50 T'O •­oet*noZA*ro*«n»; S. УО v/o Я­?63/30 W* *­осЪшяо1/катош»ы.

,­г o.i ЛОЛ) „ . И

Although result indicates that both 4­235 and H­263 enow suffi­cient extractrability for Fe(IIl) in HC1 eolation even in high concentration of Fe(III) in aqueoue phase, N­235 is superior over H­263 for its higher yield in multistage countercurrent stripping experiments (Table 1). Therefore, the tertiary long­chain alkyl(Cg­C10)amine N­235 is finally chosen as the extra­etant for Fe(III) and the HC1 left ia the raffinate Is recovered for recycling in rust­washing process. The final aqueoue solution after stripping (pH<1) contains 210 g/1 PeCl, with only trace amount of Cu.Mg and Si ( mg/1) and can be easily reprocessed to obtain solid PeCl, products for sale. The overall flowsheet proposed for recovering KC1 in acidic iron­containing effluent from rust­washing process of macnine building plants ia schema­tically indicated as follows.

—г •—

­«Htii it»»—bUgL ., —I Oxidation

toplMltn with ШС1

• U t l | U • H o t y t l o »

tatoa extraction! of Iron Г

1 Г Kaltl.tM" oo*kt«roumat

Л ок.

318

Strlpylag »f lr*a<IXI) ***• ог(а*1с рЬм» la multlMXmf eoantereurwat proc***

Йиж HO. 1 г 3 4

0 ~ . phaaa

30 r /o » ­ 2 f V К т /о a­aotanol

40 » /o «­263/ 35 r/o a­octaaol

30 »/o »­гз5 / 30 <r/o e­oetaaol

45 тЛ> «­235/ 35 r/o a­ocsase l

Vaaparatiara ( a) » 58 38 38

No. of atafea 6 8 6 a Str lPylaf n * | n t watar water water water

flow r a t i o (org . /** , . ) 2:1 2:1 2:1 2:1

C o » , of Г » ( 1 Ш i n Пя*1 org . аавае, H 0.184 0.320 0.100 0.021

C O M . o f T o ( I I I ) IB / i s a l 09 . рвааа, II 0.978 0.800 1.25 1.32

n . l d of atripptaf (X) 55.6 55.6 B2.4 91.1

References 1. I sh imor l e t e l . , JAERI­1(M7( 1965) J JABRI­1062( 1964)}

JAERI­1106(1966). 2. J .H.White , P.Ke l ly and N . C . L i / / J . Ino rg . Nucl. Ohem.

I f . 337(1961). 3 . A.D.NelBon, J .L .Fasch ing and R.L.KcDonald, i b i d . _27. 439(1965). 4. O.Levy, G.Markovitea and A.S.Ker tes , i b i d . 33. 551(1971). 5. Mary I , Good and Sara Е. H r y a n / / J . Aroer. Chem. So'­.

,82, 5636(1960).

319

PHASE INTERACTIONS IN THE SOLVENT-IN-PULP EXTRACTION OF BORON 11-25

A.W.L.Dudeney and K.Poslu, Department of Mineral Resources Engineering, Imperial College, London, S.W.7 2BP, UK

In common with many metal extractions, the solvent extraction of borates normally depends on careful prior solid-liquid separation. Solvent-in-pulp systems are rarely feasible because of solvent loss r\nd emulsification. However, incentive remains to avoid the costs of thickening, filtration and clarification, particularly in the Turkish borate industry, where borax deposits are interlayered with clay minerals and carbonates.

Previous reports ( 1,2) considered the solvent extraction of borax from kerosene solutions of 2-

chloro-4-(ltlt3,3-tetrameihylburyl)-6-methylolphenol (CTMP). The rate and equilibrium position of extraction were found to be largely unaffected by the presence of 5 weight percent of -75 micrometre quartz, calcite or kaolinite in suspension. The present paper deals with phase interaction and disengagement characteristics, which were affected by these solids.

Preliminary experimental work indicated that, when various mixtures of CTMP, kerosene, water, and aqueous borate were stirred together at 800 rpro in a baffled beaker, emulsification occurred above pH 6 and increased in intensity with increase in pH. On sealing, the phases slowly separated. After settling for 30 minutes at pH 8, the upper bulk phase still contained about 30% water and the lower phase a small concentration of oil microdroplets (giving it a hazy appearance). When solids were added at pH 8, and the mixing and settling procedures carried out as before: quartz remained predominantly in the aqueous phase with little alteration of the emulsified state without solids; calcite (or colemanite) was totally transferred to the oil phase giving an emulsion of enhanced stability above a clear aqueous phase having some 70% of its original volume; kaolinite (от bentonite) similarly 'oil-floated' with most of the aqueous phase incorporated in the emulsion; and talc formed a stable emulsified phase between relatively clear organic and aqueous phases. As expected, increased ionic strength (by additions of borax or halite) or decreased stirring rate reduced the intensity of emulsification. When the phases were mixed gently in fully filled cylindrical flasks, by means of a wheel rotating at 10 rpm for 30 minutes, and allowed to settle for a further 30 minutes, there was no significant entrainment of one phase in the other and no emulsification in the presence of solids. UV spectrophotornetric analysis of the aqueous phase for CTMP (.carried out by back extraction into kerosene and measurement in this phase at 285 nm) gave values similar to the solubility of the reagent, e.g., 60 ppm at pH 8. The same boron extraction was observed regardless of the mode of mixing.

From the preliminary results it appeared that solvent-in-pulp boron extraction might be feasible if the behaviour of the suspended solids was better understood (and controlled) and if extractions were carried out with sufficiently controlled mixing of the phases. More detailed studies were undertaken on the effects of CTMP adsorption onto solids, CTMP acid dissociation, interfacial tension, ionic strength, and mixing conditions on emulsification and phase disengagement

CTMP adsorption,Measurements of CTMP adsorption onto calcite, quartz and kaolini' represented (i) the relative tendencies of die minerals to become oleophilic and therefore to collect at t' oil-water interface or disperse in the organic phase and (ii) potential solvent losses by adsorption ir solvent extraction process.

A known mass (in the range 0.1-1.0 g) of mineral (nominal specific surface area 1,0 m2/g) w" mixed for 30 minutes with 50 ml aliquois of purified petroleum ether or water solutions of the reage (5-100 ppm CTMP)by initial ultrasonic dispersion for one minute followed by gentle shaking. Aft* centrifuging, the liquid phase was analysed spectrophotometrically and the CTMP adsorbed found by difference. The results obtained are shown in Fig. I. For increasing CTMP concentration greater than 10"4 M, adsorption increased rapidly for calcite dispersed in petroleum ether until saturation was

330

. 1 1 i

­

(В)

:

­

|гн=ыз<^££Е* 4­ i ­5.0 ­4.0 ­3.0

Log [CTMP(M)]

Fig.1. Adsorption of CTMP onto calcite О , kaolinite • and quartz • from (A) petroleum ether and (B) water at pH 9

eached. This saturation was taken to correspond to a monolayer coverage of СГМР molecules of 4.61 дто1 nr 2, assuming a functional head group area of 0.36 nm2 (estimated from a 3D model of the CTMP molecule). For kaolinite and quartz the increase was more gradual. The extent of adsorption decreased in the order calcite > kaolinite > quartz. In the presence of an aqueous phase, CTMP adsorption was Jess than 15% ofnwnolayercxaverageurKierall conditions, presumably because of competition by water molecules for surface active sites: at pH 9 the aqueous solubility of CTMP is about 100 ppm (3.4 x 1(И M) while the concentration of water molecules is 55 M. The same order of adsorption persisted.

At pH 8­9 the surfaces of calcite are approximately electroneutral, whereas those of kaolinite and quartz carry a net negative charge (3). Thus electrostatic repulsion should reduce adsorption of (negatively charged) CTMP species on kaolinite and quartz. Conversely, specific chemisorption at Ca2+

sites on calcite might be facilitated. From these results, the mass of CTMP adsorbed was less than 10% of that dissolved in aqueous

solution. Solvent losses due to aisorption would be relatively unimportant However, a small percent of monolayer coverage was clearly sufficient to cause significant changes in surface chemical properties, and thus a more stable interaction with an oil phase, as has been reported for other systems (4,5).

CTMP acid dissociation. Measurements of acid dissociation — which occurs via the CTMP phenolic group — related to me effects (on phase interactions) of increased solubility and ionization of intcf acially adsorbed molecules at higher pH values.

The acid dissociation constant K a of R(OH)2, i.e., CTMP repres ;nted as a diol (1), was jeterminedby (i) equilibrating equal volumes of 0.50 M CTMP in petroleum ether with water at different M values in the range 3­10 and determining the aqueous concentration (solubility) ^pectiophomaietrically, as outlined above; (ii) calculating the distribution ratio as > = 0.50 / {[R(OH>2J+[R(02H)­]} at each pH; and (iii) interpreting a plot of log D vs pH, ~s shown in

Mg. 2(A). The theory behind this determination has been given by Poslu (6). Fig. 2(B) shows the »lculated relative concentrations of species basexf on K a = 1(Уа. It can be seen that the organic phase fcncentratioo {ЗДОгОДо remains essentially constant while [R(OH>2] falls above pH S and [ВДОДО­]

(together with the overall aqueous solubility) increases rapidly above pH 7. As will be seen below, the liquid­liquid interface is likely to be saturated with CTMP species. The

proportion of these in ionized form should also increase above pH 7. The ions ihould orientate with the

­5.0 ­4.0 ­3.0 Log [CTMP(M)]

321

3.5 -

Q g> 3.0

2.5 -

x>-i~t 1 ' 1 1 : c - - j

-0 \

-

-

'Л -

(A) ! *\ 1 I : i oi

5.0 7.0 9.0 11.0 Equilibrium pH

5.0 7.0 9.0 11.0 Equilibrium pH

Fig. 2. Logarithmic plots of distribution ratio (A) and CTMP concentration (B) vs pH indicating CTMP apparent pK = 8.0 and calculated concentrations of species

polar (ionized) part in the water phase (the minus charge being counter balanced by Na+ and H + ions in this phase) and the non-polar pan in the petroleum eiher, thus giving rise to an electrical double layer. In consequence, any globules in an emulsion would carry the same electrical charge and be mutually repellent, thus retarding aggregation and coalesence.

Increases in solubility and implied development of surface negative charges on oil globules accompanying acid dissociation are consistent with the observed onset of relatively stable emulsificaiion, and make practical solvent extraction difficult above pH 7-8.

Imerfacial tension and ionic strength - Measurements of interfacial tension (Kruss digital tensiometer) between water and СГМР in petroleum ether indicated (i) an upper limiting .value of 45 mN nv1 at zero CTMP concentration; (ii) a rapid decrease from 35 to 20 mN nr 1 in the range 10-3 - 1 0 2 M CTMP; and (Ш) the approach to a lower limiting value of 10- 12mNm-] at 0.1 - 1.0 M CTMP as the interface became saturated with CTMP species. Increases in ionic strength (by addition of sodium chloride at 0.2 -1.0 M) or decreases in pH (from 11 to 3) increased the interfacial tension marginally, by .ibout Ifo. Ionic, strength above 0.1 caused rapid separation of metasrable emulsions.

The small interfacial tension at industrially relevant CTMP concentrations (> 0.5 M) indicates a relatively large interaction at the interface and partly explains the observed tendency for emulsions to form easily widi this system. Interfacial tension cannot readily be manipulaied to advantage but.of course, ionic strength can be increased to aid phase separation.

Mixinp conditions , Photographic measurements of droplet size distributions after mixing at different stirring rates gave (i) relationships between energy input and intensity of emulsification / demulsification and (ii) information on emulsion structure without and with the presence of solids.

Equal volumes of 0.5 M CTMP in kerosene and 0.048 M boron as borax (at pH 7.5, 8.5 and 9.7) were stirred in a baffled beaker at 600, 800 and 1000 rpm for 10 rain. Samples of each resulting emulsion was immediately transferred to a glass cuvette and photographed (.SLR camera with macro lens and bellows) at suitable time intervals at the mid point of one of the optical flats of the cuvette. Postcard sized prints provided an overall magnification of 50 times. The droplet size distributions were determined by direct counting of 100 droplet images on each print

Pig. 3(A) shows the observed decrease in size (average droplet diameter) from about 125 to 65 yim on increasing the stirring rate from 600 to 1000 rpm. Increase in pH caused a marginal decrease in size. Also plotted (Fig. 3(B)) is the logarithm of the time taken for the static emulsions to break down completely against original stirring speed. The straight line relationship observed facilitates extrpolatior. to lower stirring speeds (f~r which the photographic technique is too slow). For instance, a stirring soced of 200 rpm gives en estimated demulsificanon time of 10 roin. Clearly, in a successful industrial

322

£&

3 -

E .£

о Ё

600 800 1000 600 800 1000 Stirring speed (rpm) Stirring speed (rpm)

3. Plots of average droplet size (A) and log (demulsification time) pH 7.5, 8.o and 9.7 respectively (B) against stirring speed: О , • , П

process the energy input from stirring would have to be relatively low, even in the absence of solids. In the presence of calcite or kaolinite (800 rpm, pH 8) the photographs showed little change in droplet size. The particles seemed to have been trapped between two or more organic droplets, but with the mineral staying mostiy in the aqueous phase, and tended to form a 'monolayer' of particles around each droplet. Such a collection of particles would cause a mechanical barrier to coalesence. The emulsions containing the mineral particles were stable for several days at rest, although they were readily disrupted by centrifuging for 2 min at 1000 rpm. With quartz, the mineral rapidly settled under all conditions and the demulsification times were similar to those without solids present. Evidently, chemisorption stabilised the droplet - mineral - water associations in the cases of calcite and kaolinite but -ot quartz.

To provide appropriate mixing conditions, a 15.2 cm diameter contacter (2,6), which was supplied by RTL Contactor Holding SA, was used in the semi-continuous, counter current, solvent-in-pulp extraction of boron. Under the conditions employed (rotor speed 5 rpm, flow rate 90 ml per minute, inlet boron concentration 1000 ppm adjusted to pH 10.8 with lime, solids content 1 % as a 50% mixture of calcite and bentonite, and CTMP concentration 0.3 M) a 93.6% boron extraction was achieved with an outlet pH of 8.4. Initial organic entrainment was about 1000 ppm. After settling for 30 minutes this was Teduced to 450 ppm and centrifuging readily reduced the losses to the CTMP solubility level (60 ppm). A preliminary flowsheet was designed to reduce the content of plant effluents to less than 10 ppm boron. This will be described elsewhere.

References; 1. Poslu K.andDudeney A.W.L. Hydrometall.. 1983JO. 47-60. 2. Poslu K., Dudeney A.W.L,and Gochin R.J. Trans. Inst. Min. MexalU 1985. j?f. C230-232. 3. Apian F.R and rucrstenau D.W. Principles of non-metallic flotation, 50th anniversary volume,

Ed., Fuerstenau D.W. A.I.M.E., New York, 1962. p,l70-274. 4. Stratton-Oawley R. and Shergold H.UColloids and surfaces. 1981.J. 253-265. 5. Shaw 0., Introduction to colloid and surface chemistry, 2nd Ed. Butterworths, London, 1975. 6 Poslu K. Solvent extraction of boron from clear solutions and slurries, PhD thesis. University of

I-ondon, 1983.

323

SOLVENT EXTRACTION ОТ ELEMENTAL SULPHUR IN A CONTINUOUS PROCESS FOR THE OXIDATIVE LEACHING 07 SULPHUR ORES

F. Pochetti, L. Того (*) and R. Lavecchia^

Dipartimento di Ingegneria Chiraica, University of Rome, Italy

(*) Dipartimento di Ingegneria Chimica e dei Materiali University of L'Aquila, Italy

INTRODUCTION

Oxidative leaching processes have recently gained an increasing interest in the hydrooe'tallurgy of sulphur ores. Nevertheless, at present, they are not applicable with profit to all minerals, nor are they free from drawtocks of technological nature. In particular, the formation of muds with unleashed or partially leached sulphides, inerts and precipitates makes quite expensive the separation of reaction products and the recovery of sulphur.

The Authors developed an original process for the continuous oxidative leaching of metallic sulphides in the presence of an organic solvent, of sulphur [l­3]. This process allows the total recovery, from the organic phase, of the elemental sulphur produced in the course of the reaction. The kinetic analysis of the process, with reference to syntetic and natural minerals, allowed to determine the dependence of the rate of sulphide dissolution from the operative parameters. The reaction rates observed were much higher than those obtainable in the absence of the solvent.

A pilot reactor­extractor plant, operating continuously in countercurrent, was then designed and built, on the basis of the preliminary results of batch runs carried out at constant temperature in shaken flasks.

MATERIALS

Two different minerals were used in the experimental runs: an Italian high sulphur content sphalerite (Raibl) and a nixed sphalerite­pyrite­chalcopyrite concentrate. This mineral was selected in order to investigate the possibility of selectively recovering copper and zinc in the course of the leaching process. Elemental chemical analyses axe reported in table.

Chvalcal analysis of the alnaxalB (vt x) A ­ aiUcd concentrate В ­ •phalarit»

' 1 A В

Cm S.H 0.O* Al l .ftt* If. o.*o 33.30 *t 7J>.*5 м > 0.COO5 1m 0.01B г.so Aa ».0t pp. 9 47.11 30.00 Co 0.001 1* 3*.7t 1,70 lit 0.О0Э Rl 0.0)1 N1 0.001s Ы 0.0]) 0ЛС AB 0.1S fa 0.1SI S» O.0I3 4 0.003 1.23 Q* 0.O* SI 0.C7J l . M Ca O.OOS ГМ 0.05 Sn

­0.00s

я* 0.ОИ CI 0.15 С 0.011 2.00 P 0.00s

11­26

324

The leaching solution consisted of ferric sulfate dissolved in distilled water. Carbon tetrachloride was used as solvent.

EXTRACTION APPARATUS

A reactor­extractor, of a glass column with diameter of i*k сто and length of 1700 mm was built up. the column axis a stainless

consisting internal

n overall Along steel

Fig. 1» Reactor­extractor

shaft provided with agitators and driven by a variable speed motor was placed. A schematic representation of the apparatus is shown in fig. 1. The leaching solution, which is the continuous phase, is fed from the bottom» while the suspension of sulphide minerals in the organic solvent, which represents the dispersed phase, flows in from the top.

The reactor­extractor can be divided into three zones. The upper part is water­cooled, in order to accomplish the separation of the phases: in fact density change with temperature for the solvent is much greater than for the aqueous phase. In the central zone, which has a volume of about 2.5 2, the reactions take place, while in the lower part of the column the organic phase accumulates.

A comprehensive investigation of the performance of the reactor involves both a fluodynamic and a kinetic characterization. In a preliminary approach to the analysis of the process we determined the dispersed phase hold­up and performed some kinetic runs.

HOLD­UP MEASUREMENT

The bold­up measurement was performed by closing the inlet valves of the aqueous and organic phases after they had reached steady­state conditions. Then thti motor was stopped, the two phases were drawn out of the column and th­*ir volume was measured. These runs were carried out at unitary conversion and at infinite recirculation ratio.

If V( is the total volume of the reaction zone, the dispersed phase hold­up can be calculated as:

9»­­V H/V„ (1)

where V^ is the dispersed phase volume. For the residence time of the phaues inside the reactor we have:

V d * c ­ <v

t V d)/F c <2)

where Fj and F c are the flowrates of the dispersed and continuous phase respectively.

325

At this point» i t is possible to evaluate the characteristic velocity, v k

from the relation:

v

k" V I V1 " *QH

+ V H1 " Ъд

г

)> (3)

where Vj and v c are the superficial velocities of the two phases, defined as the ratio of the flovrate to the coluan cross­sectional area.

Finally, the slip velocity can be determined by the following equation:

v

r " vd/9>d + v c/(l ­ y d ) (4)

v k, n/h In fig' 2 the characteristic velocity is plotted as a function of the slip velocity, at different flowrates.

KINETIC RUNS

Kinetic runs were carried out at the temperature of 70 °C and with a mineral to organic suspension ratio of 15 g/1. The conversion was determined by weighing the elemental sulphur after solvent evaporation. The experimental results were analyzed by the shrinking non­reacting care model. Its validity was verified in previous investigations [l~5] and justified by the non­porosity of both minerals and by the continuous dissolution of sulphur in the organic solvent.

The following equation was used to interpret the kinetic data:

Characteristic velocity patterns 1 (1 ­ X)' ,»/3.

VT J (5)

where X is the conversion» while conversion, which is defined as:

t is the time required for the complete

3/(r a n H ) . m In the above equation r represents the reaction rate, a 0 the surface area

and M the molecular weight of the mineral. In fig. 3 some values of т are reported as a function of the ferric sulfate concentration.

CONCLUSIONS

The use of a continuous process allows to carry out a correct kinetic analysis, under temperature and pressure conditions comparable with those used in batch industrial processes. The possibility of operating under pressure and in a continuous way is another innovative feature of the proposed process. Other non negligible advantages are the absence of air pollulants, the possibility of reacting low concentration ores and the economic recovery of elemental sulphur from the solvent by simple cooling and crystallization.

326

?e 2(S0 4) 3, mol/l Fife. 3« Experimental values of r

REFERENCES

1. Pochetti F., Того L. // Paper at 3rd Austrian­Italian­Yugoslav, Chem. Eng. Conf. Graz, 1982. ProceedingStP.693#

2. Pochetti F , Того L. // Paper at V Convegno sullo stato metallico Genova, 1983. Abstract 23.

3. Pochetti F., Того L. // Brev. Ital. n. 49195 A/83. U. Pochetti F., Того L., Huscari V. // La Chiraica e 1' Industrie 1984.

Vol.66. P.395 5. Lavecchia В., Pochetti F., Того L. // La Chimica e 1' Industrie. 1987.

Vol.69. P.B7

327

THE THIRD PHASE FORMATION AT PHOSPHORIC ACID WET PROCBSS 11­27 Z.S.Golynko, G.K.Tselistchev, All­Unlon Research Institute of Chemical Technology, State Committee on Utilization Of Atomic Energy,Moscow,USSR

The third (the second organic) phase formation at the phosphoric acid extraction by trial, ylphosphate is associated with limited solubi­lity of the solvates formed in an inert diluent and depends on the al­]£ylphosphate type, diluent origin, phosphoric acid concentration and extragent concentration.

It has been established that in the alkylphosphate­alkylphospninox­ide series the phosphona­ es of DAMP or DOUP acid type do not form the third phase in all studi: d concentration ranges of phosphoric acid and extragent, which appears to be associated with raised solubility of isomer­forming phosphona es, solvated by phosphoric acid, in an inert diluent.

The study has shown t at in the phosphoric acid concentration up to 15 mol/1 the pure (10056) TBP does not form the third phase. The third phase is formed in diluted TBP solutions, at that the starting point of the third phase is displaced into the region of smaller equilibrium concentrations of phosphoric acid in the aqueous phase and for more di­luted TBP solutions. The conditions for the starting point of the two

The third (lower) phase presents pure TBP, which is confirmed by the dis­

organic phases formation are given in Table. Table 1. Conditions fjr starting point

TBP, mol/1

H 3 P 0 4 , l o l / l TBP, mol/1

i n kerosene i n octane

0 .9 1.8 2.7

9.55 ' 9.61 9.85 10.20

10.20 | 13.20

of two organic phases formation tribution coefficient value, being constant within the range 0.4­0.42 for all concentra­tions TBP studied.

The origin of diluent is of certain significan­ce. TBP solution does not form the third phase in benzene.

Distribution of phosrhoric acid in aqueous and organic phases during the two organic phases ­jrmation is shown in Fig,

The data on phosphor, з acid distribution proves presence of solvates with different TBP and : ,P0. ratio. The character of solvatation can be assessed from the compo.ition of the third (second organic) phase. Ana­lysis of the third phas , obtained by treating 25St TBP solution in oc­tane with concentrated hosphoric acid, showed that molar ratio TBP : H

3*°4 " 1 : 3 and TBP : HgO = 1. Thus, one can suppose existence of hy­drosolvate TBP • 3K,P0„ H„0.

3 4 2 Analysis of the thirc phase obtained by treating 25% TAPO solution

in octane has shown thai the molar ratio TAPO : H,P0. = 1 : 4 , and one ,

328

H

Jr e

«,(°)

Distribution of phosphoric acid between aqueous and organic phases in the region of the third phase formation 1 ­ 100S8 TBP; 2 ­ 75SS TBP; 3­50!» TBPj 4 ­ 25* TBP in octane

can suppose presence of TAPO . 4H,P0. solvate. Thus, that can explain Intensified eitractability of phosphinoxides in regard of phosphoric acid.

Effect of eztragent concentration on the distribution coefficient of phosphoric acid leads to possibility of obtaining the second aqueous phase, e.g. variation of TBP concentration in octane from 100 to 25#V decreases the distribution coefficient from 0.16 to 0.005, i.e. 30 fold. The second aqueous phase contained 4.2 mol/1 H,P0..

An alternative to the process of the third phase formation is the process of mixing aqueous and organic phases with formation of homoge­neous aqueous­organic solution. The character of phase volumes variati­on at the extraction of phosphoric acid with butanol has shown that there exists a threshold concentration for phosphoric acid above which phosphoric acid of any concentration can mix with a diluent. It means that the process passes from a heterogeneous into a homogeneous stage. It has been pointed out that the point of displacement is affected by water contents in butyl alcohol. For the butyl alkohol containing 0.088 mol/1 HgO mixing starts at an initial phosphoric acid concentration of 7.1 mol/1, for the butyl alkohol containing 8.3 mol/1 HgO ­ 8.45 mol/1. The above relationships are characteristic for other mineral acids as well, at that the complete mixing depends on the nature of acid. The aeries of extractability of mineral acids regarding butanol is retained under conditions of complete mixing, and the better mineral acid is ex­tracted the lower concentration it takes to mix completely, e.g. for nitric acid concentration is 4.86 mol/1. 2Z Зак. 382 329

11­28 j WET PROCESS FOR PHOSPHORIC ACID AHD COMPARATIVE ASSESSMENT OF ORGANIC SOLVENTS B.N.Leskortn, D.I.Skorovarov, Z.S.GoIynko, G.K.Tselistchev, All­Unlon Research Institute of Chemical Technology, State Committee on Utilisation of Atomic Energy, Moscow, USSR

Extraction of phosphoric acid is gaining attention in regard of ma­nufacturing pure phosphoric acid and phosphates on its base from phos­phorites of various type.

The extraction oi phosphoric acid calls for utilization of solvents of various classes: aliphatic alcohols, organic phosphates, amines, eth­ers and their mixtures.

Phosphoric acid in aqueous solution is highly hydrated and, as a rule, the hydxosolvates of mS • nH^PO. . pHpO type are extracted. The mechan­izm of phosphoric acid extraction by organic amines is different in some way.

Comparative characteristics of extragents of various classes show that the extractability decreases in the series C.>Cc>Cg>...C10. This is relevant to the series of organic phosphates, too. Besides, extractabili­ty increases in the series R,P(0)> (R0),P(0). In other words, extracta­bility depends on the hydrocarbon chain in the organic solvent.

The class of aliphatic alcohols has a clearly expressed linear corre­lation between the distribution coefficient of phosphoric acid (D) and the number of carbon atoms in the hydrocarbon chain (n c) with a correla­tion coefficient г о 1. E.g. the extraction from 6 mol/1 H,PO. is des­cribed by equation D • 1/6.02 n„ ­ 18.58. It is worth noticing that the coefficients in the equation are interrelated and have to be determined on the basis of mathematical statistics lows. At extraction of phospho­ric acid by alcohols, e.g. by butanol, a certain linear correlation in many parameters is evident, such as one between phosphoric acid concen­tration in organic or aqueous phases (% weight) and density of the cor­responding phase (kg/m­3

), presented as Y Q = 6.94/H,P0./ + 838. The study on equilibrium phases composition in system BuOH ­ H,P0. ­ H_0 has shown that there exists a correlation between parameters /H ?0/ and /BuOH/^, the ratio approaches 1, the mean value ­ 1.006 with standard deviation В • 0.03. There is a correlation between lg H,0 /H PO, ­lg Bu0H/H,P0,. The obtained correlations lead to empirical equation for phosphoric acid extraction with butanol,

lg /Н эР0 4/ о . f.45 lg/K 3P0 4/ w ­ 0.45 lg/H 20/ w ­ 0.216 or lg/H 3P0 4/ o . lg/H 3P0 4/ w + ­0.015 /HjP0 4/ w ­ 0.833.

The joint solution of these equations brings about the possibility to determine the water contents in the aqueous phase and thue to calculate

330

the threefold diagram of the phase transformation In the system Бион­и к ­ н2о.

Phosphoric acid ie satisfactorily extracted with IBP. The differen­ce in the extractions!, ability of TBP according to its concentration ia evident, which can be seen from Table 1.

Table 1. Extraction of Н чРО л with TBP

Distribution coefficient for concentration of H,P0,

TBP concentration, vol.S Distribution coefficient for concentration of H,P0, 25 50 75 ;oo

1 mol/1 6 aol/1

0.005 0.04

0.02 0.11

0.06 0.19

0.15 0.28

For the phosphoric acid extraction with TBP the distribution coeffi­cient rises with the increasing equilibrium concentration of phospho­ric acid in the aqueous phase, such relation is characteristic for the phosphoric acid extraction with butanol, too, i.e. for the extragents having a concave isotherm. A sharp increase of the distribution coeffi­cients affected by TBP concentration is also evident.

It is known that the nature of diluent can considerably affect the distribution coefficients of phosphoric acid. The TBP solutions in hy­drocarbon diluents have higher D. The distribution coefficient in aro­matic hydrocarbons is lower. Still, it is noticed that with increasing concentration of phosphoric acid in the initial solution the phospho­ric acid concentration in the organic phase is practically independent of the diluent nature, which is supposed to relate to the character of solvation. The mechanism of the phosphoric acid extraction is complica­ted enough. The data on the phosphoric acid distribution point out to existence of solvates with various ratio between TBP and H,P0.. In the saturated state the ratio TBP : H,P0. • 1 : 3. Presence of the phospho­ryl group both in the TBP molecule and H­FO. molecule is the main rea­son of association in the system. The direct measuring testifies to it too. Viscidity of the saturated organic phase is aboux three tines gre­ater than viscidity of phosphoric acid 4 • 10 ш/а against t.4 • 10 m /и. The study on the third phase composition also points out to this ratio.

The character of phosphoric acid hydration in the organic phase can be determined on the basis of the direct water determination. It has been established that the constant ratio TBP : H„0 is retained through­out a wide range of phosphoric acid concentration in the organic phase, which makes one believe that phosphoric acid is hydrated with one mo­lecule of water and the composition of hydrosolvate is ТВР»ЗН,Р0Л'Н;,0.

331

Fig.1.H­P0, extraction with TBPO 1­0.4; 2­0.8; 3­1.6 mol/1 TBPO

The distribution character of phosphoric acid between its aqueous solutions by trialkylphosphinoxide shows that the curve of the D­H,P0, relation passes through the maximum and then decreases monotonously with increasing equilibrium concentration of phosphoric acid in the aqueous solution. The maximum position is Individual for each phosph­inoxide of corresponding concentration. The isotherm for extraction of phosphoric acid by trialkylphosphinoxides is convex and the section ran­nig parallel to the X­axis is absent, thus, the saturation of extragent with phosphoric acid is not reached. It is characteristic both for al­cohols and organic phosphates with the exception of amines. The increa­se of equilibrium concentration of phosphoric acid in the aqueous solu­tion is accompanied by its increasing contents in the extragent. This increasing concentration of phosphoric acid in the organic phase brings about oomplete miring of the two phases or formation of the third (se­cond organic) phase. If the data on the phosphoric acid extraction are expressed in the coordinates of D ­ Н,Р0,/ТЛР0 then two straight sec­tions will be eviaent, each one described by equation of the type С « mC^ (Pig,1). Some empirical equations are given in Table 2.

Table 2.H3P0. extraction with TBPO

!BPO concentr., mol/1

Diagram equa­tions

1.625 C 0 ­0 .738C W '1 0 5

C 0 * . 4 6 8 C ° /5 2 5

0.812 C0H3.317C^*1 2 0

C 0 * . 2 1 8 c £ '5 9 1

0.406 C o«0.0720^*0 9 0

С оЦ).0850° г*6 9 6

332

P i g . 2 . H,PO. e x t r a c t i o n l g 1­BuOHj 2­TBP; Э­ТВРО ­Oft

-an

-lo

1 2 3 4 H.PO., molA

The joint nolution of the equations makes it possible to find such pho­sphoric acid concentration at which the mechanism of phosphoric acid extraction with phosphinoxides is altered.

к comparative estimation for extractional ability with butyl alco­hol, TBP and TBPO is given In Pig.2.

Table 3. Capacity of extragents t o r phosphoric acid, kg/t

Comentration of H,P04, mol/1 TBP BuOH DBAA TBPO

1.0 16 18 66 1?0 2.0 51 52 150 187 3.0 78 101 233 220 4.0 92 140 275 248 5.0 109 193 316 297

A comparative assessment of extragent may be done proceeding from their capacity for phosphoric acid. Table 3 gives the comparison data.

333

EXTRACTIVE RECOVERY OP ORGANIC LIQUIDS PISELY DISPERSED IN WATER

G.Angelov, I n s t i t u t e ot Chemical Engineering, Bulgarian Academy of S c i e n c e s , S o f i a , Bulgaria

Organic-In-water d i spers ions are produced In many i n d u s t r i a l proc­e s s e s . Separation of such systems i s of ten needed in order to recover valuable organic l i q u i d s or to meet the r e s t r i c t i o n s for waste flow contamination». In the case of organlcs p a r t i a l l y so luble i n water the dispersed f r a c t i o n could he separated to some extent by us ing c o ­alescence-promoting d e v i c e s f l ] . But for a t o t a l separat ion , inc luding a l s o the d i s s o l v e d organic f r a c t i o n , l i q u i d ex trac t ion p r i n c i p l e s ehould be appl iedJz] . Most o f high e f f i c i e n c y conventional ex trac t ion equipment use I n t e n s i v e hydrodynamie regimes f o r c r e a t i n g o f great i n ­terphase s u r f a c e . This i s not convenient for the considered case beca­use of f i n e drople t s ye t e x i s t i n g i n the feed f low and the p o s s i b i l i t y for further unwanted drop breaking, which could cause a d d i t i o n a l pha­s e separation d i f f i c u l t i e s - The laboratory s c a l e s tudy[3] l ead to the development of equipment designed to f i t the requirements f o r contact and separat ion of polyphasic l i q u i d systems with f i n e granulometry. The evo lut ion o f apparatus cons truct ion from laboratory to I n d u s t r i a l s c a l e i s presented i n t h i s paper and r e s u l t s of l a r g e s c a l e t e s t s are g i v e n .

I f compare the separat ion e f f i c i e n c y of bed columns of var ious s i z e the c l a s s i c a l sca le -up problem i s c l e a r l y manifested - the greater the apparatus s i z e , the lower the separation e f f i c i e n c y . Trying to overcome t h i s problem, mult ip le construct ion of repeated modules was studied.The e l n g l e module was c y l i n d r i c a l car tr idge- type c o n s t r u c t i o n . The separated layers are superimposed on a perforated t u b e - s c e l e t o n . The d e t a i l e d descr ip t ion i s given i n [4]. An apparatus 100 mm i n diam. with one module incorporated was t e s t e d . I t s main features were s t a ­b l e ex trac t ion e f f i c i e n c y , not s t rong ly inf luenced by the throughput and a good a b i l i t y for separat ion of phaaeB In a heterogeneous system. but the o v e r a l l separat ion e f f i c i e n c y was not high enough mainly due to the incompleted mass t r a n s f e r . I t might be i n t e n s i f i e d when make interphase surface r e g u l a r l y renewed. Currently i t i s done by repeated mechanical mixing f o r d ispergat ion of phases fol lowed by s e t t l i n g and phase coa lescencefc ] . In our cas« t h e layer surface proper t i e s were used to c r e a t e a s i m i l a r e f f e o t . When the f low meets porous media with a l t e r n a t e l y changed surface proper t i e s (wetted-nonwetted) , c o a ­l e scence and rediapergat ion i s caused. Skis was the concept for con­s t r u c t i o n of c y l i n d r i c a l cartr idge with mul t i iayer wall .Considerable improvement of mass t r a n s f e r e f f i c i e n c y was obtained (over 90 % a t thr ighput 40-45 m'/m 2h.). The wal l th ickness was s i g n i f i c a n t l y t h i n -

11-29 I

334

an (л­25 шш) In comparison with the thickness of the horissintal la­yer of comparable efflci«ney(25­3t) cm).

The overall apparatue flowrate is Increased hy multlplyung the num­ber of single modules. What is necessary is to ensure relatively uni­form croessection 'low distributior in the apparatus body in order to provide similar near­to­optimal operating conditions for individual modules. This requirement might be siaply fulfilled by using a combina­tion of a horizontal filed bed section followed by a section of verti­cal cartridge modules. She higher pressure drop of I section equalizes approximately the crossflov distribution. Moreover, this bed contribu­tes additionaly to the mass transfer efficiency. The 71 section keeps its two main functions of efficient mass transfer and phase separation device.

400r­

f i g . 1. Separation e f f i c i e n c y v s throughput. Single module. Pine o i l ­ w a t e r d i s p e r s i o n . Solvent (petrolium e t h e r ) t o water r a t i o ф 1:50

О 1:100

ф 1:200

20O 30Q TOOWHATE, 1 / h

400

Ihe results for testing single multilayer modules are illustrated on Fig. 1. Ihe approximate range of operating conditions for a good performance, which could be maintained in a multimodale construction is: flowrate 300­400 1/h (throughput ~50 в'/ш2

Ь) under phase ratio 1:50­1;1oo.

Ih« total flow capacity of the industrial scale installation was chosen according to the particular requirements of a common essential oil production plant (~2500 1/h). More than 85 % recovery of the ess­ential ail IB required.

Ihe installation scheme is shown on M.g. 2. Ihe flow to be treated paaaesthrough the filter (1) for removal of solid particles. A part of the flow (2) is used to disperse the solvent in the injector (3). The solvent droplet size is controlled by regulation of this flow (2). Ihe extractive­separational unit includes the horizontal "equalizing" layer (4) and 7 identical modules (5) as described above. The deoiled water leaves the apparatue through the exit (6). In» light organic phase (7) containing extracted oil,enters the distillation unit (8),

335

where the eolvent i s dis t i l l ed , the «pours are condensed (9) and re­

cycled in the column. The unevaporated rest *" t h e extracted essential

o i l product (10).

rig. 2­ Principal scheme of the industrial installation 1­Miter; 2­Injecto* controlling valve; 3­InJector; 4­Horizontal equalising layer,400 аш in dlam.; 5­Extractive­separatio nal cartrid­

ges; 6­Bxit purified water", 7­ attract exit; 8­Diatillation unit; 9­ Condenser; 10­£xtracted product

Kg. 5. Separation efficiency versus flowrate Large scale lnetal le; ion: 1­ camomile oil­v?ter dispersion, solvent to water ratio 1s75s 2­ . i ­ l i oil­water uispersion, solTent to water ratio 1:50; 3 ­ levanda oil­water dispersion, eolvent to water ratio 1Я0О

336

Duxii g the Indus tr ia l t e s t s s tab le operat ion under continuous non stop regime was observed. General I l l u s t r a t i o n of the e f f i c i e n c y of e s s e n t i a l o i l recovery i s made on f i g . 3 .

The i ldUBtrlal performance data confirm "the expedient design ana e f f e c t i v e operation o f the I n d u s t r i a l s c a l e extrac t ion i n s t a l l a t i o n for t r e a t i n g of heterogeneous systems containing f i n e l y dispersed and d i s s o l v e d organic l i q u i d s .

In tbe considered p a r t i c u l a r case current ly decanters are used f o r the sept ration of t h e e s s e n t i a l o i l ­ w a t e r d i s p e r s i o n . When use e x t r a c ­

t i o n as a next s t e p of separat ion , near ly 70 % a d d i t i o n a l product (camooi: a o i l with a higher content of water so lub le components) was abtaln.ee [ б ] . The reported q u a n t i t i e s might be considered as recovered from » : t » f lows uning the proposed new technique .

QQHC DSIOE. The elaborated extrac t ion process and equipment enab­

l e s for a more complete separat ion of organic l i q u i d s f i n e l y dispersed and p a r t i a l l y d i s s o l v e d i n water . The d i s s o l u t i o n o f organic d r o p l e t s by the olvent i s much more e f f e c t i v e and f a s t e r than t h e i r c o a l e s c e n ­

ce and grav i ty s e t t l i n g . Moreover, the d i s s o l v e d i n water organics are a l so ex trac ted . The process i s low energy consuming [7J . Itoergy i s ueed only for regenerat ion of a r e l a t i v e l y email amount of l o w ­ b o i l i n g s o l ­

v e n t . ', eat ing of heterogeneous systems with dispersed phat" d e n s i t y above с .­ below water dens i ty does not require changes i n aj « r a t u s c o n s t n c t ion because dens i ty di f ference so lvent ­water i s e e c e n t i a l f o r a norm: . opera t ion . Ihe a p p l i c a t i o n of t h i s method to low dens i ty d i f ­

f e r «net systems ( t h e i r g r a v i t a t i o n a l s e t t l i n g i s slow and d i f f i c u l t ) i s very convenient and e f f e c t i v e . The sea l ing­up problems are o v e r c o ­

med by use of mult ip le module c o n s t r u c t i o n . Thi process and i n d u s t r i a l i n s t a l l a t i o n are appl ied i n the e s s e n ­

t i a l о 1 production technology . The main area of other use are: phar­

macy, food industry , petrochemistry, organic syntheses e t c . for p r o ­

' luc t lo j or recovery of valuable substances as well as f o r removal of harmful and t o x i c components from water f l o w s .

References 1 . Perrut M.//Revue I n s t . Franc. P e t r o l s , J J (5) .763 ( 1 9 7 2 ) . 2 . Tr<ybal a . Liquid ex trac t ion . Moscow, Khimiya, 1966. 3 . Angelov G., Dragostinov P. et al.^Hung. Joum.Chem.Ind. ^5 (2 )

16 (1987) . 4 . AD e l o v G., Boyadahiev L. et a l . / / B u l g . P a t . H 59826 (1983) . 5 . Sh b e l B.G., Kerr A.E. / /Ind.Sig .Chem. 42. 1048 ( 1 9 5 0 ) . e . Te;hnlcal Beport H 8, ObChB (1983) . 7 . Juigelov и./УСиеш.* Ind. 5C '"). ^93 (1986) .

337

REMOVAL OF ORGANIC ENTRAINMENTS FROM RAFFINATES [ 1 '~30

AND STRIPPING LIQUORS BY ADSORPTION ON MACROPOROUS POLYMEREE

Holger Stephen. Peter Muhl and Karsten Gloe Chemiefaserkombinat Schwarza, Rudolstadt­Schwarza ­ DOR­6822

(GDR) 2antraj.­netttut fur Festkfirperphysik und werkstofforechung, Akademie dar Wissenschaften der DDR, Dresden ­ 00R­S027 (GDR)

Liquid­liquid extraction processes for metal separation and winning are always connected with the problem of the entrainment of organic phase (extractants and diluents) into the aqueous process solutions and waste water, respectively. Entrainments in raffinatae or separate scrub raffinatea take a lot of efforts in waste water treatnent and organic impuri­ties in stripping liquors lead to difficulties in further pro­севе steps. Both, physically dissolved parts and mlcroemul­sions of extractants and diluents in the aqueous phase, must be removed to a required degree*

The object of the present investigations is the adsorption of di(2­ethylhexyl)phosphoric acid (D2EHPA) as extractant and aliphatic hydrocarbons as diluents from acidic zinc­bearing solutions, produced by zinc recovery from process solutions of rayon industry. On principle activated carbons and macroporous reeins ­ obtained by copolymerization of etyrene and divinyl­benzene ­ as adsorbents are suitable for the removal of orga­nic compounds from aqueous solutions. The mean advantages of macroporoue resins in comparison with the carbons are the simple regeneration process and the low energy consumption. Furthermore they are effective for many process cycles ^1­3/в

Investigations have been carried out with the nonionic macroporous resins Wofatit EP 61 (specific surface on ЗЕТ 380*480 m2

/g) and Y 77 (> 1000 m2

/g) ­ both from Chemiekombi­nat Blttarfeld (GDR) ­ in comparison with activated carbon (> 1000 • /g). The adsorption re*in Y 77 contains not only nacropores, which are typical, of well­known resins, but also many mesoporee and micropores* It has an excellent adsorption efficiency.

The adsorption of D2EHPA and n­dodecane ­ as example for technical hydrocarbons ­ by the various adsorbents havt b­jen •"­vestinatpd under variation of the H_S0.­ and ZnSO.­concen­trations as a function of time. Adsorption isotherms of 02EHPA and n­dodecane have been determined for typical ргосвзз solu­

338

t ions. The possibi l i t ies for the elution of these compounds from the adsorbents, which are ent i re ly eetureted with the ьхtreetent end the diluent, has been studied. Beet results were found using methanol or alkaline methanol solution, and in the case of steam stripping, respectively. The obtained eluete has a very high concentration of extгасtant or diluent, 1 . e« an economic recovery can be carried out» The best ad­

sorption eff iciency shows Wofatit Y 77. Acidic zinc s­ l fa te solutions with an i n i t i a l content of 10­500 ppn of organic compounds could be puri f ied to a level of lower than 1 ppm* The capacity of this resin is higher than that of the activa­

ted carbon ueed and is sufficient for application In an in ­

dustr ial process. Regeneration is possible without losses of capacity, A very f:i.gh s t a b i l i t y against osmotic end chemical stress was observed too.

The results, obtained on laboratory scale* have been con­

firmed on a pi lot equipment with two para l le l columns. This adsorpticn process can be successful applied for the

separation of surfactants from feed solutions.

References

1 . R» L. Gueteiecn, R, L. Albright, 3. Heieler, 3. A„ Li r io , 0, T. Reid j r , // Ind , Eng. Chem. Prod. Res. Develop»^7 (1968). 107.

2 . K. Gloe, P. Miihl, I . N. Kremenskaja, A. N. Turanov// Proc. XX. Symp. "Hcrnicka Pribram vs Vede a Technics**, Pribram (CSSR), 1981, P, 403,

3 , G* Dummer, L. Schmidhammar// Paper, Oehrestreffen dar Verfehrenslngenieure, Niirnberg (FRG) 1983.

339

AUTHOR ПГОЕХ

Abbruzzese С. - 291 Abishev D.H. 305 Abramov A. A. 161 Akopov O.A. 187 Alexandrov St.38,88 Alibayev В.Ы. 2*6 Alioarfo I. P. 6* Aly H.F. 1*0 Alymova A.T. 108 Ami N. 179 Andreev V.I. 197 Aneva Z. 88 Angelov G. 308, 33* Antonovich V.P. 71 Arpadjan S. 38, 88 Babic M. 227 Bsgreev V.V. 22 Balen M. 269 Balog I.S. 22 Bao Р.У. 287 Baron P. 20* Baroncelli P. 236 Bazel Ya.R. 22 Beklemishev M.K. 9* Belkova O.L. 279 Berestovoy A.M. 279 Berahard G. 191 Beyanzhanov T.Sh. 2*6 Breyl K.-J. 201 Be issert W. 191 Borkowaki M. 129 Bcrzova E.A. 8* Boyedzhiev L. 308 BukJia T.I. 150 Buscjunann H. 19 Buttinelli D. 295 Canic N. 227 Carlinl D. 236 Casarci M. 219 Cech R. 165 Cechova S, 165

ChanyBheva T.A. 90 Chaplygina H.I. 8* Charykov A.K. 53 Chen J.Y. 265 Chen H.B. 316 Chen Zh.Ch. 2**, 272 Cherkasov A.E. 279 Chesne A. 193 Chevalier S. 193 Chlarizia R. 10, 219 Chmutova Ш.К. 137 Czopek R. II* Danesi P.R. 10 Demidov V.D. 279 Distln P. A. 255 DjoPovic N. 227 tiokjchayev V.N. 299 Dosmagambetova S.S. 82 Drake V.A. 183 Drozdova L. I. 248 Dubrovina L.T. *J Dudeney A.W.L. 320 Duhamet J. 204 Dvorak Z. I** Elenkov D. 308 Ermolayeva T.N. 108 Esimantovskil V.M. 215 Evans S.F. 211 Fatovic I. 231 Filippov E.A. 168 Fleitlich I.Yu. 299 Flett D.S. 2*0 Frankovsky V.A. 3* Prenkel V.Ya. 1*7 Pu Y.J. 272 Fuchtner P. 157 Puks L. 129 Galkin B.Ya. 215 Gao Ch.M. 287 Garnak.S.A. 97 Gasparini G.M. 219 , 236

340

Germain M. 124, 193 Giavarlni С. 295 Giganov G.P. 2*6 Gighlrhanov M.CrA. J6I Gindin L.M. 275 Gloe K. 338 Golynko Z.S. 328, 330 Gorenatein L.A. 75 Grahnert T. 191 Grazhulene S.S. 84 Grote M. 45 Guerzilov P. 312 Guo H. 316 Hadjiev D. 312 Halonln A.S. 30 Hermann A. 157 Hladik 0. 191 Horwitz E.P. 10 Ivanov I.M. 275 Ivanov V.M. 64 Ivanova S.N. 90 Janson E.J. 97 Jedinakova V. 144 Jinkina J.A. 207, 211 Jin T.Zh. 316 Kabachnlk M.I. 137 Kadletsova L. 215 Kalinkina S.P. Ю 8 Kallsh N.K. 100 Kaluga V.V. 252 Karalova Z.K. 150 Karandaahev V.K. 84 Karavaeva L..V. 64 Kartasheva N.A. 133 Katykhin G.S. 41 Kazakova M. 312 Kettrup A. 45 Kholkin A.I. 246, 299 Kiah P.P. 22 Kolomleta L. L. 34 Kondratieva N.E. 41 Kopunec R. 14 Korda Т.Н. 100 Korenraan Ya.I. 108 Koayakov V.N. 153

Kovenia A.V. 53 Kravchenko A.K. 252 Krinitsyn A.P. 187 Krotki M. 223 Kuanyeheva K.Z. 246 Kulenov A.S. 246 Kunit3kaya I.S. 97 Kurdziel K. 114 Kuz'min N.H.. 94 Kuzmin V.I. 299 Kuznetsov G.I.. 168, 252 Kyrsh M. 215 Kyutohoufcov G. 308 Labuda J. 79 Lanin V.P. 252 Laskorln B.N. 330 Lavecchia E. 324 Lavrinovich E.A. 150 Lazarev L.H.2I5 Lensrcik B. 114 Lenhard Z. 269 Leonov V.A. 82 Leahchev S.M. 60 Li H. 244, 272 LI Zh. 287 Liang Z.F. 287 Liu P.A. 3I6 Lobanov P.I. 82 Logadail D.H. 207, 211 Luboahnikova K.S. 299 Luginin V.A. 41 Lupi С 295 Lutova L.S. 283 Lyall E. 207 Lyubtsev R.I. 197, 215 Macaaek P. 14, 165 Macklachcov A.G. J6I Maghrawy H.B. 140 Makaimova N..V. Щ 0 Haljkovie D. 269 Markov G.S. 197 Martynov V.M. 302 Meles S. 231 Meng Sh.L. 244 Mazharaup G.P. 97

Michelaen O.B. 25S Mikhaillchenko A.I.,'48 MikulaJ V. 14 Miraevsky G.P. 252 Mirokchin A.M. 302 Mistry P. 207 Mojski M. Ц 2 Molochnikova H.P. IV? Moshkov M.M. 197 Jfutol P. 338 Muslkas C. 124 Hyasoedov B.P. 118,137,147,150 Нуегз Р.К. 207 Havtanovich M.L. 275, 283 Nazarenko A.Yu. 49 Nazarenko V.A. 71 Nebel D. 157 Neaterova N.P. 137 Nifctforov A.S. 168 Niklpelov B.V. 197 Mkolotova Z . I . 133 Kovikov L.K. 299 Noworyta A. 223 Nurtaeva A.K. 82 Oslpov H.N. 53 Oveharenko V.P. 161 Panfllova S.E. 94 Pareau D. 193 Partridge B.A. 207 Paahkov G.L. 246, 299 Pasquiou J.Y. 193 Perkov I.e. 26 Petrioh G. 175 Petrukhin O.M. 4 Petrunlak D.fl. 279 Piehugin A.A. 302 Plekert G. 45 Poohettt P. 324 Pollshchuk S.V. 60 Fopandopulo Yu.I. 84 Poelu K. 320 Preenyak I.S. 71 Prlbylova G.A. 137 Protasova H.V. 299

Pushkov A.A. 168, 252 Puzzuoll G. 219 Pyatnltsky I.V. 34 Quon K. 255 Radosevic И. 231 Rals I. 215 Rajec P. 14 Rakhmanjko E.K. 60 Al-Rawi H. 129 El-Reefy S.A. 140 Renard E.V. 168 Rodionova T.V. 64 Romano J. 231 Romanovskii V.N. 215 Rozen A.M. 133, 168 Ruzin L.I. 302 Sakurai S. 179 Samollov Yu.M. 67 Sohnieder H. 175, 201 Schon H. 201 Schwartz J.M. 104 Selutakii P. 215 Sergeyeva V.V. 299 Sergiyevaky V.V. 246 Sevostyanov Yu.G. 161 Shatskaya S.S. 67 Shelikhina E.I. 71 Shevohuk I.A. 57 Shishkin Г.Н. 215 Shkinev V.H. 147 Shkl'yar L.I. 252 Shuvayeva O.V. 90 Simonetti E. 236 Skorovarov D.I. 330 Smetanln E.Ya. 168 Smith I.J. 211 Saulrek W. 129 Spasic A. 227 Splvakov B.Ya.4 Starobinete G.L. 60 Stephan E. 338 Stryapkov A.V. 305 Suohanov P.T. Ю 8 Sukhan 7.V. 49

342

Sun D.X. 316 Svec A. 14 Taehlmori S. 179 Takezhanov S.T. 246 Tananaiko Н.И. 75 Taraeov V.V. 302 Tolatunov V.I. 279 Torgov V.G. 100 Того L. 324 Travkin V.F. 252, 302 Taarenko A.R 187 Tse P.K. 10 Taellstchev G.K. 328, 330 Tevirko G.A. 60 Taylov Yu.A. 248 Udalova T.A. 248 Uljai v V.S. 302 Usuda S. 179 Valerian! G. 219 Vanfckova M. 79 Vaeileva E. 38 Vasilieva А. А. Ю 0 Visotskaya T.I. 75

Vlaaov H.H. 153 Vlasov V.S. 168 Voronehlhlna L.A. 279 Voronin H.H. 279 Weat D.W. 240 Wu H.W. 287 WJ J.G. 316 Xu G.X. 316 Xu Я. 316 Yagodin G.A. 246 Yakovlev K.G. 153 Yakshln V.V. 82 Yofa B.S. 161 Yu Sh.Q. 265 Yudelevich I.G. 90, 100 Yukhln Yu.M. 248 Zadnepruk L.V. 84 Zakharkin B.S. 168 Zelentsova L.V. 100 Zhang D.L. 287 Zhang W.X. 244, 287 Zhao Y.C. 265 Zolotov Yu.A. 94

343

С OBTESTS

Analy t i ca l Cheniatry 3

Nuclear ft-ooeaees 117

Sydroaetal lurgy. 2

3 5

Author Index 340

Научное издание

МЕЖДУНАРОДНАЯ КОНФЕРЕНЦИЯ ПО ЭКСТРАКЦИИ (1SEC88) Труды конференции, том IV (на англ. яэ )

Утверждено к печати Институтом геохимии и аналитической химии им. В.И. Вернадского АН СССР

Художник А.Г. Кобрин. Художественный редактор В.В. Алексеев

Н/К Подписано к печати 14.04.88. Формат 60 X 90 1/16. Бумага офсетная N" 1 Печать офсетная. Усллечл. 21,5. Усллср.-отт. 21,6. Уч.-издл. 22,7 Тираж 1000 экз. Тип. зак. 382. Бесплатно. Заказное Ордена Трудового Красного Знамени издательство "Наука'* 117864 ГСП-7, Москва В-485, Профсоюзная ул., д. 90 Ордена Трудового Красного Знмени 1-я типография издательства "Наука" 1990Э4, Ленинград В-34, 9-я линия, 12 Отпечатано с оригинала, подготовленного к печати Оргкомитетом Международной конференции по экстракции (1SEC88)