“a s 2008”

Nicolaus Copernicus University in Toruń Wrocław University of Technology

University of Technology and Life Sciences, Bydgoszcz

PROCEEDINGS OF THE XXIII INTERNATIONAL SYMPOSIUM

ON PHYSICO-CHEMICAL METHODS OF SEPARATION

“A RS SEPARATORIA 2008”

JULY 6-9, 2008, TORUŃ, POLAND

2008

Editors

Stanisław Koter Izabela Koter

Reviewers

Józef Ceynowa, Stanisław Koter, Wojciech Kujawski, Jadwiga Ostrowska-Czubenko, Artur P. Terzyk,

Władysław Walkowiak, Romuald Wódzki

Preparation for printing

Barbara Łukasik-Gilewicz, Stanisław Koter

© No part of the publication may be reproduced in any form by print, photoprint, microfilm or any other means without permission from

the author or publisher.

ISBN 978-83-231-2208-1

NICOLAUS COPERNICUS UNIVERSITY IN TORUŃ

WROCŁAW UNIVERSITY OF TECHNOLOGY UNIVERSITY OF TECHNOLOGY AND LIFE SCIENCES, BYDGOSZCZ

TORUŃ - WROCŁAW - BYDGOSZCZ, 2008

International Advisory Board B. Buszewski, Poland W. Charewicz, Poland G. Cote, France P.R. Danesi, Austria K. Inoue, Japan A. Larbot, France F. Mačasek, Slovakia A. Maruška, Lithuania A. Narębska, Poland A. Sastre, Spain J. Schauer, Czech Rep. S. Schlosser, Slovakia S. Siekierski, Poland O.A. Sinegribova, Russia G. Sulaberidze, Russia M. Streat, England W. Walkowiak, Poland R. Wódzki, Poland W. Wójcik, Poland

Organizing Committee

Beniamin Lenarcik – honorary chairman Stanisław Koter – chairman Józef Ceynowa – co-chairman

Andrzej W. Trochimczuk – co-chairman Wojciech Korpal

Izabela Koter – secretary Monika Kultys – finance

Barbara Łukasik-Gilewicz – materials editor Piotr Adamczak – IT coordinator

Dariusz Czerwiński Magdalena Gierszewska-

DruŜyńska

Wojciech Kujawski ElŜbieta Radzymińska-Lenarcik

Address: Nicolaus Copernicus University,

Faculty of Chemistry, 7 Gagarina Str.

87-100 TORUŃ, POLAND Tel. (048) (056) 611-43-18, Fax. (048) (056) 654-24-77

http://www.ars_separatoria.chem.uni.torun.pl

XXIII ARS SEPARATORIA – Toruń, Poland 2008

15

TROUBLESOME PHENOMENA ENCOUNTERED IN

HYDROMETALLURGICAL SOLVENT EXTRACTION PLANTS: FROM THE

LABORATORY TO THE REALITY OF THE INDUSTRIAL SCALE

G. COTE, A. CHAGNES, B. COURTAUD, J. THIRY

a Laboratoire d'Electrochimie et de Chimie Analytique, Ecole National Supérieure de Chimie de Paris-ENSCP, UMR 7575-CNRS-ENSCP-Université Pierre et Marie

Curie-Paris6 11, Rue Pierre et Marie Curie, 75231 Paris Cedex, France b AREVA-NC, Service d’Etudes de Procédés et Analyses (SEPA), B.P. 71, 87250

Bessines sur Gartemps, France.

Solvent extraction is a well established technology in various fields including the hydrometallurgy of base and strategic metals (Cu, Ni, Co, Ga, PGM’s, U, etc.), the reprocessing of used nuclear fuels (U and Pu), the petrochemistry, the pharmaceutical and food industries.

In its principle, solvent extraction is a simple technology consisting of mixing two non miscible liquid phases to achieve the selective transfer of a solute of interest from one phase to the other. The extracted solute is then stripped to allow its recovery and the recycling of the extracting phase. In hydrometallurgy, the feed solution is aqueous in nature and contains the metal values to be recovered, whereas the extracting phase is usually an organic phase containing an extractant and a phase modifier dissolved into an organic diluent.

The chemistry of metal extraction is based on the formation of metal entities including simple or mixed ligands metal complexes, solvates, ions pairs, etc. At the scale of a solvent extraction plant, the chemistry is limited by technological, economical and environmental constraints. For instance, the feed solution containing the metal values of interest is usually issued from previous leaching stages which impose important parameters such as the nature of the aqueous phase (sulfate, chloride, nitrate, phosphate, fluoride, etc.), the level of acidity and the ionic strength. Thus, most of the chemistry should be transferred on the design of the organic phase via the selection of appropriate extractants/co-extractants, diluents and phase modifiers.

In the life of a solvent extraction plant, the process may deviate from its nominal characteristics for various reasons including the ageing of the


16

solvent extraction phase and a change in the composition of the feed solution resulting from changes in the composition of the ores to be exploited. This may lead to enhanced solvent degradation, shift in selectivity, enhanced formation of muds, difficulties in metal stripping and solvent recycling. These phenomena can be coupled and may induce a significant decrease of the efficiency of the plant. The purpose of the present paper is to illustrate such troublesome phenomena with two examples: - extraction of nickel and cobalt from acidic leach solutions of laterites, - extraction of uranium from acidic sulfate solutions with Alamine® 336.

REFERENCES [1] S. Facon, F. Adekola, G. Cote, Stripping of copper from CYANEX® 301 extract with

thiourea–hydrazine–sodium hydroxide solution, Hydrometallurgy, 2007, 89, 297. [2] A. Chagnes, G. Cote, B. Courtaud, J. Thiry, Role of Vanadium(V) on the Ageing of the

Organic Phase in the Extraction of Uranium(VI) by Alamine®336 from Acidic Sulfate Leach Liquors, Proceedings of The International Solvent Extraction Conference ISEC 2008, Tucson, Arizona (USA).

[3] A. Chagnes, G. Cote, J. Thiry, B. Courtaud, Chemical Degradation of Trioctylamine in n-Dodecane modified by Tridecanol in Presence of Chromium(VI), Proceeding of the 11th International Mineral Processing Symposium, Belek-Antalaya (Turkey).

[4] A. Chagnes, G. Cote, J. Thiry, B. Courtaud, Implementation of New Flowsheets in Uranium Solvent Extraction Plants: Enhancement of the Production and the Sturdiness against Extraction Solvent Ageing, Proceeding of the 11th International Mineral Processing Symposium, Belek-Antalaya (Turkey).

[5] S. Collet, A. Chagnes, B. Courtaud, J. Thiry, G. Cote, Computer Simulation of Flow Sheets for the Solvent Extraction of Uranium: A New Route to Delay the Effect of the Chemical Degradation of the Organic Phase on Uranium Recovery from Acidic Sulfate Media, to be published.

[6] A. Chagnes, B. Courtaud, J. Thiry, G. Cote, Chemical Degradation of trioctylamine-Tridecanol-n-dodecane by Cr(VI), to be published.


17

APPLICATION OF PERVAPORATION AND OSMOTIC

MEMBRANE DISTILLATION IN FRUITS AND FRUIT JUICES

PROCESSING Wojciech KUJAWSKI1*), Magdalena GIERSZEWSKA-DRUśYŃSKA1),

Anna SOBOLEWSKA1), Agnieszka DOBRAK2), Carme GÜELL3) Montse FERRANDO3), Francisco LOPEZ3), Justyna WARCZOK4),

1) Nicolaus Copernicus University, Faculty of Chemistry, Toruń, Poland, *) [email protected]

2) Katholieke Universiteit Leuven, Faculty of Bio-eng. Sciences, Leuven, Belgium 3) Universitat Rovira i Virgili, Dept Chem. Eng., Tarragona, Spain

4) Dow Water Solutions, Tarragona Technical Centre, Tarragona, Spain

Abstract

In the food industry the traditional separation processes are progressively substituted by membrane filtration to improve the quality of the product. Osmotic membrane distillation (OMD) experiments with polytetrafluoroethylene (PTFE) hydrophobic porous membrane were performed to determine the efficiency of concentration of apple, white grapes and red grapes juices. Juices concentrations between 5 and 30ºBrix and 50 wt. % calcium chloride solution were used as a feed and a stripping solution, respectively. Results show that water permeate flux depended on the initial juice concentration in feed solution and on the type of the juice. Moreover, osmotic membrane distillation did not affect the analyzed quality parameters of the concentrated juices. It was shown that pervaporation with hydrophobic membranes can be effectively applied for the reconcentration of spent solutions from osmotic dehydration and OMD processes.

1. INTRODUCTION

Fruits and vegetables are essential ingredients of human diet but they are seasonal. Drying of fruits and vegetables and/or production of concentrated juices are well known methods of food preservation.

Osmotic dehydration (OD) process is based on the product (food) immersion in a hypertonic solution (e.g. concentrated sucrose solution in a case of candying). During OD candying, water flux from food to the solution occurs with a counter-current flux of sugar from the solution to the food stuff. This results in the dilution of the hypertonic solution with simultaneous increase of its mass and decrease of its dewatering potential [1]. The diluted hypertonic solutions must undergo a reconcentration process before they are re-used in a new OD cycle.


18

Production of concentrated fruit juices is mainly carried out nowadays by a multistage vacuum evaporation. However, this thermal technology implies a partial loss of important juice constituents (e.g. flavors, vitamins and/or antioxidant activity), and it can also result in cooked taste and color degradation, thus reducing the quality of the final product [2,3].

The food industry is interested in novel processing technologies that may result in economical and improved quality products. Osmotic membrane distillation (OMD) is a membrane contactor technique, which applies porous hydrophobic membranes, and allows the dehydration of solutions at constant temperature under atmospheric pressure [1-5]. During the OMD process, both sides of the porous hydrophobic non-wettable membrane are in contact with two aqueous solutions of different water activity, e.g. juice solution or sucrose solution as feed solution and concentrated CaCl2 or NaCl solutions as stripping ones. The water activity difference between the two liquid phases generates a flux of water vapor from the feed solution to the stripping solution.

Pervaporation (PV) is a membrane technique used for separation of liquid mixtures or to remove solvents from the solutions of non-volatile solutes [6,7]. The liquid feed is in contact with one side of a non-porous membrane, while the volatile components of the feed are evaporated on the other side of the membrane into a vacuum chamber or a sweeping gas [7].

The efficiency of OMD process for reconcentration of sucrose spent solutions from the OD process was presented in a previous paper [2]. This paper presents results on the application of membrane processes in the food processing as well as in the management of spent solutions. OMD process was used to fruit juices concentration, whereas PV process was applied to reconcentration of sucrose spent solution from OD process and to reconcentration of CaCl2 diluted solution from OMD process.

2. EXPERIMENTAL

Osmotic membrane distillation experiments have been carried out using a setup described in the detail elsewhere [2]. The membrane module consisted of two symmetrical compartments, separated by a hydrophobic membrane. Polytetrafluoroethylene membrane (Sartorius, PTFE 11806) was used in OMD experiments. PTFE 11806 is an asymmetric hydrophobic membrane with the mean thickness of 75 µm and an average pore size diameter of 0.45 µm on the skin side.

The following feed solutions were studied in OMD experiments: apple, red grape and white grape juices with initial concentrations of 5, 10 and 30oBrix. The 50 wt.% CaCl2 solution (awater=0.2) solution was applied as the stripping solution. During OMD experiments water flux from feed to


19

stripping solution was measured. The final juice concentration resulting from the OMD process, as well as some selected quality parameters of juices (e.g. color, total polyphenolic content and antioxidant activity) were also determined.

PV experiments were carried out in the laboratory-scale PV system described in [7]. The feed solution was circulating over the membrane, and the permeate was collected in cold traps cooled by liquid nitrogen. During experiments the upstream pressure was maintained at atmospheric pressure, while the downstream pressure was kept below 1 mbar by using a vacuum pump. As the feed mixture contained water and the non-volatile solute, the permeate constituted of pure water only. Two types of hydrophobic membranes were used in PV experiments: PDMS-PAN and PDMS-PAN-NF (Pervatech, the Netherlands). Both membranes possessed a selective layer made of polydimethylsiloxane (PDMS), but NF membrane had a slightly porous structure of the skin layer. The effect of the non-volatile solute content on water flux was determined for each membrane in contact with water-sucrose (10-60oBrix sucrose solutions) and water-calcium chloride solutions (30-45 wt.% CaCl2 solutions). All OMD and PV experiments were performed at temperature of 35oC.

3. RESULTS

Concentration of fruit juices by the OMD process

0

1

2

3

4

5

6

7

Wat

er flu

x J

[kg m

-2 h

-1]

Sucrose Apple juice Red grapejuice

White grapejuice

Fig. 1. Water permeate fluxes during OMD process of different 30oBrix solutions.

Fig.1 presents water permeate fluxes obtained during the concentration

of 30oBrix fruit juices. Results are compared with the water permeate flux obtained for 30oBrix sucrose solution, used to model a fruit juice. It can be seen that the water permeate flux was the highest for the sucrose solution. For fruit juices the flux varied, depending on the non-soluble species content (e.g. pectins).


20

Final concentration of OMD treated juices was influences by the dehydration process time and initial juice concentration. Results (Fig.2) clearly indicate that OMD can be efficiently used for the concentration of fruit juices. Moreover, since the driving force in OMD is not a hydraulic pressure difference, a very high final concentration of juice can be easily achieved, depending only on the configuration of the system and the duration of the process.

0

5

10

15

20

0 100 200 300 400 500

Time [min]

Juic

e co

nce

ntr

atio

n [

oB

rix]

5Brix, 0.45um

10Brix, 0.45um0

200

400

600

800

1000

1200

1400

0 2 4 6 8 10 12

Juice concentration [oBrix]

TP

I [m

g G

AE

/L]

Initial juice concentration 5oBrix

Fig. 2. Resulting red grape juice concentration during OMD process as a function

of the OMD process time.

Fig. 3. TPI as a function of the final red grape juice concentration after OMD process.

The OMD process should not affect the quality parameters of the

processed juices, like color, turbidity, total polyphenols index (TPI) or the antioxidant activity (given by the TEAC equivalent) [2,3,5,8]. Fig.3 presents the TPI of red grape juice as a function of the final juice concentration obtained after OMD. It can be seen that the TPI depends linearly on the final juice concentration, meaning that water removal by OMD process does not affect the amount of polyphenolic compounds. The preservation of other quality parameters (e.g. color or antioxidant capacity) after concentration by OMD was also proven in the investigated systems (results not shown). Reconcentration of spent solutions from OD and OMD processes by PV

The spent solutions from osmotic dehydration processes, i.e. diluted sucrose solution for candying, can be reconcentrated using OMD process [2]. However, the application of OMD process results in the generation of a new spent solution, i.e. diluted calcium chloride solution. To avoid this problem, pervaporation of spent sucrose solutions can be used. Results of the dehydration of sucrose solution by pervaporation using hydrophobic


21

membranes are presented in Fig.4. It was found that only water is transported through the pervaporation membrane and that the permeate flux

0

200

400

600

800

1000

1200

0 20 40 60

Sucrose concentration in feed [oBrix]

Per

mea

te w

ater

flux

J [g

m-2

h-1

]

PDMS-PAN-NF

PDMS-PAN

Fig. 4. Water permeate flux in pervaporation process of sucrose solutions.

was dependent on the sucrose concentration in the feed solution. In contact with PDMS-PAN-NF membrane the water permeate flux was higher comparing with PDMS-PAN membranes. The difference resulted from the more open structure of the selective layer of former membrane. For low sucrose concentrations (10-40ºBrix) higher water permeate flux was obtained for PDMS-PAN-NF membrane. However, once the sucrose concentration increased (40-60ºBrix) the differences in flux diminished reaching comparable values for both pervaporation membranes at 60ºBrix. The flux decrease was caused mainly by the concentration polarization effect and, to a much smaller extent, by the decreasing pervaporation driving force. In the industrial application of PV, the spent sucrose solution should undergone the microfiltration prior to pervaporation, in order to remove any suspended particles or microorganism (cold pasteurization).

Pervaporation process was also effectively used for the reconcentration of the spent hypertonic solution from the OMD process. Fig. 5 presents results of water pervaporation flux, when PDMS-PAN membrane was contacting calcium chloride solutions of different concentrations. Although permeate fluxes are lower than in case of sucrose solutions, they are still acceptable for the practical use. It should be however noticed that in the countries with a wide access to the solar energy, the reconcentration of calcium chloride solution could be also performed by a thermal evaporation.


22

0

100

200

300

400

500

600

700

800

20 30 40 50

Feed concentration [wt.% CaCl2]

Wat

er p

erva

po

ratio

n fl

ux

[g m

-2 h

-1]

Fig. 5. Water permeate flux in pervaporation process of calcium chloride solutions.

Membrane: PDMS-PAN.

4. CONCLUSIONS

Membrane separation processes like osmotic membrane distillation and pervaporation can be efficiently implemented in fruits and fruit juices processing. The OMD process did not affect the quality of processed juices. Pervaporation can be applied for reconcentration of the spend solutions from osmotic dehydration and osmotic membrane distillation processes.

Acknowledgements

A. Dobrak and A. Sobolewska. wish to express their thanks for the Erasmus mobility grant, which enabled them the research training at Universitat Rovira i Virgili (URV, Tarragona - Catalunya, Spain). W. Kujawski kindly acknowledges the research scholarship at URV granted by the Catalan Government (Beques de Recerca per a Professors i Investigadors Visitants a Catalunya - 2006PIV0016).

REFERENCES

[1] J.Warczok, M.Ferrando, F.López, A.Pihlajamäki, C.Güell, J. Food Eng. 80 (2007) 317. [2] J.Warczok, M.Gierszewska, W.Kujawski, C.Güell, Sep. Pur. Technol. 57 (2007) 425. [3] M.Dalla Rosa, F.Giroux, J. Food Eng. 49 (2001) 223. [4] M.Gryta, J. Membrane Sci., 246 (2005) 145. [5] V.D.Alves, I.M.Coelhoso, J. Membr. Sci. 208 (2002) 171. [6] W.Kujawski, Polish J. Chem. Technol. 5 (2003) 1. [7] W.Kujawski at al., Desalination 205 (2007) 75. [8] A.Zulueta, M.J.Esteve, I.Frasquet, A.Frigola, Food Chemistry 103 (2007) 1365.


31

FLUX AND CLEANING ENHANCEMENT IN A SPIRAL

MEMBRANE ELEMENT USING CONTINUOUS INFRASONIC FREQUENCY BACK PULSING

R.D. SANDERSON, D. KOEN, F. BESTER and D.S. MCLACHLAN

UNESCO Associated Center for Macromolecules & Materials, Dept. of Chemistry and Polymer Science, University of Stellenbosch, P. Bag X1, Matieland 7602,

South Africa

Abstract Experiments are described which show that continuous 1 Hz back pulsing,

with a peak pulse pressure that gives rise to a negative TMP and a duty cycle of 17%, can even with Dextrin (which has strong fouling tendencies), give rise to strongly enhanced continuous flux values.

1. INTRODUCTION

The supply of cheap potable water in sufficient quantities is a world wide problem. Polymer membranes provides a good solution to obtaining potable water, but membranes always foul during the water purification, and the flux drops with time. Physical cleaning methods, which do not require the plant to be shut down for lengthy periods (if at all), are very attractive for this purpose and also they do not generate any waste fluids.

There exists a limited amount work, where the pulsing has been applied to the permeate space, but to date experiments have only been performed on flat sheet membranes [1, 2]. In this paper rapid, continuous back pulsing into the permeate space is found to be very effective in cleaning a spiral wrap element. It is found that, in agreement with the flat cell experiments [3] currently being performed at this centre, the more adhesive foulants and/or those in the pores, can only be effectively removed if the peak back pulse has sufficient amplitude to cause a negative Trans Membrane Pressure (TMP) for part of the cycle.

No theory is given in this paper because there is no fundamental theory for describing the effects of back pulsing into the permeate space of spiral wrap elements. This is due to the extreme complexity of the system, which consists of a number of membrane envelopes spirally wound onto the central permeate tube. The backpulses have to propagate down the envelopes in order to clean the surface by distorting the pores by backwards flow through the membrane.


32

2. EXPERIMENTAL APPARATUS AND PROCEDURE

A fairly standard spiral wrap filtration plant was used in these experiments together with an Alpha Laval GR40PP 100000 MWCO element containing a 0.6 m2 polysulphone membrane. A mono spiral pump provided the feed and there was a direct digital readout Wika pressure gauge (0 to 600 kPa) near the ports of the feed and permeate lines. The pressure and flow into the feed space was controlled by a throttling and a bypass valve and the permeate pressure by a further throttling valve.

The flow was recorded using Burkert flow meters, a (100-1000lt/hr) one in the feed input line and a (2-100 lt/h and a (100-1000lt/hr) in the permeate line. As a control some the permeate flow measurements were also made using a measuring cylinder and a stop watch. Note that during the pulsing experiments the pulsing had to be switched off for 30 seconds so that the “true” permeate flow rate could be recorded. A clean water (Reverse Osmosis (RO)) value was recorded before the pulsing was switched on, whereupon the feed was switched to the Dextrin solution. The data between the first and second points, includes the period during which the RO water was exchanged for the foulant feed solution (500 gms. per 1000lt) and where the very rapid initial fouling layers were formed. The flow rate after the first 5 minutes includes all these factors. Thereafter data was recorded every 5 minutes.

0 100 200 300 400 5000

20

40

60

80

100

120

140

Mill

i-Vol

ts

Milli-seconds

Fig. 1. Traces of the voltages recorded on the oscilloscope as a function of time. The upper trace is from the pressure transmitter at the permeate port and the peak

pressure seen on this trace is 110 kPa. The pulse starts form the equilibrium pressure in the feed space which is about 10kPa. The voltages in the bottom trace, recorded at the fed inlet port, are 10 times lower than indicated and the trace starts

the 100 kPa feed pressure level and the peak value is about 1 kPa above this.


33

The pressure pulses, using water from the permeate, were generated using a fast action (25 milliseconds) solenoid valve, between a constant pressure space (200 and 350 kPa), containing permeate water and the permeate port. Two further Wika pressure (0-600 kPa) transmitters were placed at the permeate and feed inputs of the element. The voltage–time (V-t) traces on the oscilloscope, were printed, scanned and digitized, an example of which is shown in Fig.1.

3. RESULTS

0 200 400 600 800 100040

60

80

100

120

140

onon

offoff

on

off

Flu

x (L

t/Hr/

m2 )

Time (Minutes)

Fig. 2. Permeate flow rate as a function of time, using a cleaning pulse, with a 170

ms pulse width, a 1second repartition rate (1 Hz) and the peak primary pulse pressure was 110 Kpa (A negative 10 kPa TMP). The pressure in the permeate

pressure reservoir for these experiments was 350 kPa.

During the experiments, the flux was measured for 1 minute at regular 5 minute intervals (with the pulsing off) and recorded as the flux against time plots. In all experiments the feed pressure was 100kPa and the flow rate 1000 lt/hr. Three fouling curves, which were obtained without back pulsing, showed saturated flux values that lay between 30 and 33 lt/hr/m2.

Fig. 2 shows the best results and one can see that the flow can be maintained at close to 90% of the clean water RO water value. When the pulsing is switched off, rapid fouling occurs. However when the pulsing is


34

switched back on, the 90% value can be recovered in about 150 minutes. Other experiments were done with the pressure in the reservoir at both 200 and 350 kPa, 2 Hz with a pulse width of about 70 milliseconds and 1, 0.5 and 0.01 Hz all with a pulse width of about 170 milliseconds. Further discussion of these and non optimal results will be given in the presentation

4. DISCUSSION

Continuous back pulsing, with a peak pulse pressure that gives rise to a negative TMP and a duty cycle of 17%, even with Dextrin, which has strong fouling tendencies, gives strongly enhanced continuous flux values. Back pulsing, such as this has also been used in the regeneration of fouled membranes using soap solutions. This resulted in reproducible clean water fluxs of about 135 lt/hr/m2, which is close to the initial clean water value.

REFERENCES

[1] P. Czekaj, F. López, C. Güell, J. Food Eng., 201, 49 25. [2] P. Czekaj, F. López C. Güell, J. Membr. Sci., 2000, 166, 199. [3] D. S. McLachlan, E Shugman, R. D. Sanderson, these proceedings, p. 73.


41

SEPARATION OF ENANTIOMERS BY CHIRALLY

MODIFIED MEMBRANES

Izabela KOTER

Nicolaus Copernicus University, Faculty of Chemistry, 7 Gagarin St., 87-100 Torun, Poland

Abstract

The use of various membrane techniques for enantiomer separation was discussed. Enantioselective transport of some amino acids and other model compounds in chirally modified membranes was investigated. The chiral selectors (cinchonidine, methyl-β-cyclodextrin) were immobilized in mesoporous ceramic membranes and polymer flat membranes. The influence of some factors as membrane pore size, immobilization type etc. on enantioselective permeation of enantiomers through the chiral membranes was studied.

1. INTRODUCTION

The preparation of chiral compounds is an important area of contemporary synthetic organic chemistry. In spite of the advances in asymmetric synthesis of pure enantiomers, the resolution of racemates is still the most important industrial approach to the synthesis of enantiomerically pure compounds. Among the methods of enantiomer separation are diastereomer crystallization, high performance liquid chromatography (HPLC) using chiral stationary phase and kinetic resolution. An alternative is the use of synthetic enantioselective membranes to transport selectively a desired enantiomer from the racemic mixture. The advantages of membrane techniques over traditional methods are their ease of use, low-energy consumption, large processing capacity, continuous operation mode, etc.

The membranes used for chiral separation can be liquid, including supported liquid membranes, or solid. The liquid membranes [1-3] containing enantiomer recognizing carriers such as chiral crown ether or cyclodextrins show highly enantioselective permeability but usually low durability because of losses of the liquid and carriers. This greatly limits their applications to a large extent. Solid membrane is more stable and therefore a durable enantiomer separation process is possible.

Solid membranes for enantioselective processes can be categorized into several types:


42

a) chiral polymer membranes, obtained by polymerization of chiral monomers (e.g. chirally derived polysulfone [4,5], chiral polysaccharides, e.g. chitosan, polyglutamates, etc. [6-8]),

b) membranes with immobilized chiral selectors, such as cyclodextrins, chiral ligands or proteins, e.g. BSA, incorporated into non-chiral membranes [9-11]

c) molecularly imprinted polymer membranes [12]. A molecularly imprinted polymer (MIP) is a synthetic polymer possessing selective molecular recognition properties because of recognition sites within the polymer matrix that are complementary to the analyte molecule in the shape and positioning of functional groups.

d) Polymer membranes with immobilized chiral catalysts, mostly enzymes, have been used widely for kinetic resolution of enantiomers [e.g. 13-15]. A kinetic resolution is defined as a process where the two enantiomers of a racemate are transformed into products at different rates.

The ideal enantioselective membrane should combine good transport rate with high selectivity. However, it is well known, that usually higher transport rates imply low selectivities and vice versa. It is important therefore to find a good compromise to optimize the separation process.

A variety of chiral selectors have been employed in enantioselective membrane separation including tartaric acid derivatives, chiral crown ethers, cyclodextrins, quinine-type compounds, hydroxyproline derivatives etc. The chiral selector can be immobilized on membrane covalently or just by entrapping it during membrane formation. The first method needs additional immobilization procedure, but it gives more stable membrane avoid the loss of the selector during the process.

In this work, a few examples of enantioselective membrane separations by chiral selectors immobilized in solid membranes has been discussed.

2. PREPARATION AND CHARACTERISTICS OF CHIRALLY

MODIFIED MEMBRANES

2.1. PREPARATION OF A CINCHONIDINE IMMOBILIZED MEMBRANE

Application of optically pure alkaloids as resolving agents is a very useful method of racemic resolution. These alkaloids react with racemic acids to form diastereomeric salts of differing solubility. (-)-Cinchonidine is also an alkaloid used frequently in catalytic asymmetric hydrogenations as a chiral metal surface modifier. It has been applied as a chiral carrier in separation of amino acids in liquid membrane system [1]. The anchorage of


43

cinchonidine to an inorganic suport such as silica can be used to allow its application in continuous processes.

Scheme 1. Scheme of (-)-cinchonidine molecule.

The commercial ceramic UF disc membrane was modified by

depositing a thin layer of silica of define pore size. Mesoporous silica of pore diameter 24-40 Å was obtained by sol-gel method with nonionic surfactants of varying length of alkyl chain. The chiral selector was then immobilized covalently on the inner surface of the pores using triethoxychlorosilane. Membranes of varying pore sizes and ammount of the chiral selector immobilized were thus obtained. The membranes were used in enantioseparation of phenylalanine.

Fig.1. Immobilization of a functionalized cinchonidine in the pores of a ceramic membrane.


44

2.1.1. RESULTS

The transport rate for both amino acid enantiomers through chiral activated membrane, and the membrane enantioselectivity were estimated. The typical results were depicted in Fig.2.

Fig.2. Enantioselectivity of the permeated phenylalanine vs. time of the process at

varying amino acid initial concentration. The effect of the active layer pore size at similar chiral selector

loading per cm2 of the active surface is shown in Table 1. Enantioselectivity of the membrane increases slightly with decreasing pore size and membrane permeability. It is understandable, as chiral membrane discriminate enantiomers through their interactions with the chiral selector attached to the pore wall. Small pores enable better contact between the solute and the chiral residue.

Table 1. Enantioselectivity of membranes with immobilized cinchonidine for

various pore sizes. Pore diameter

[nm] Amount of immobilized

CD [mg/m2] αα

2.4 0.17 1.24 3.1 0.16 1.21 3.9 0.19 1.15

a Calculated as a ratio of L and D-isomer fluxes, respectively.


45

2.2. METHYL-ββββ-CYCLODEXTRIN IMMOBILIZED MEMBRANES

Cyclodextrins (CD) are cyclic oligosaccharides consisting of d-(+)-glucopyranose units, and have the ability to include various organic molecules in the central hydrophobic cavities. Currently, chiral separation is one of the most significant applications of cyclodextrins and their derivatives. Compared with other chiral selectors, cyclodextrins have a relatively lower cost, wider applicability and higher tolerance under various environments. Usually, the immobilization of cyclodextrin onto the solid membrane could be achieved by adding cyclodextrin into membrane casting solution, as well as by covalent binding between cyclodextrin and support.

In the present work, both methods were applied. A polymer membrane with Met-β-CD was prepared by casting a polyamide solution containing ca. 2% (w/w) of methyl-β-cyclodextrin. As a result, an UF membrane with asymmetric structure was obtained. Methyl-β-cyclodextrin was also immobilized covalently in a silica layer of a ceramic flat membrane according to the procedure analogous as for cinchonidine. Selective properties of the Met-β-CD-modified membranes were studied in separation of 1-phenylethanol and its ester.

2.2.1. RESULTS

The polymer membrane with Met-β-CD immobilized by simple entrapping exhibited enantioselectivity ca. 1.2. What is more important, selective properties of the membrane seemed to decrease with time (Fig.3). The reason of such a behaviour could be leaching of the chiral selector from the membrane during the process.

The membrane with CD derivative immobilized covalently was more stable. After 5 runs enantioselectivity was practically the same and was higher than in the case of a Met-b-CD immobilized polymer membrane.


46

Fig.3. Enantioselectivity of the separation of (R,S)-1-phenylethylpropionate in

the Met-β-CD immobilized membranes.

3. CONCLUSIONS

Immobilization of a chiral selector in a porous membrane is one of the methods to provide enantioselective transport of enantiomers through the membrane. Enantioselectivity of such a system depends on character of a selector, its loading density, method of immobilization, membrane pore size and its transport properties.

REFERENCES

[1] D. Stella, J.A. Calzado, A.M. Girelli, S. Canepari, R. Bucci, C. Palet, M. Valiente, J. Separation Sci. 2002, 25(4), 229.

[2] P.J. Pickering, J.B. Chaudhuri, J. Chem. Eng. Sci, 1996, 52 (3), 377. [3] M. Bryjak, J. Kozlowski, P. Wieczorek, P. Kafarski, J. Membr. Sci. 1993, 85, 221. [4] T. Gumi, C. Minguillon, C. Palet, Polymer 2005, 46, 12306. [5] M. Yoshikawa,, K. Murakoshi, T. Kogita, K. Hanaoka, M.D. Guiver, G.P. Robertson,

European Polym. J. 2006, 42, 2532. [6] C. Thoelen, M. De bruyn, E. Theunissen, Y. Kondo, I.F.J. Vankelecom,P. Grobet, M.

Yoshikawa, P.A. Jacobs, J. Membr. Sci. 2001, 186, 153. [7] J.H. Kim, J.H. Kim, J. Jegal, K-H. Lee, J.Membr. Sci., 2003, 213 (1-2), [8] H. Swapnali, J. Membr. Sci. 310 (2008) 174. [9] S.E. Snyder, J.R. Carey, W.H. Pirkle, Tetrahedron 2005, 61, 7562. [10] Y. Xiao, T.-S. Chung, J. Membr. Sci. 2007, 290, 78. [11] Y. Matsuoka, N. Kanda, Y.M. Lee, A. Higuchi, J. Membr. Sci. 2006, 280, 116. [12] Z. Jiang, Y. Yu, H. Wu, J. Membr. Sci. 2006, 280, 876. [13] S.-D. Choi, G.-J. Kim, Catalysis Letters, 2004, 92, 35.. [14] D.R. Wu, S.M. Cramer, G. Belfort, Biotechnol. Bioeng. 1993, 41, 979. [15] J. Ceynowa, I. Koter, Sep. Sci. Technol. 2001, 36 (13), 2885.


48

SELECTIVE EXTRACTION OF PALLADIUM(II) FROM

CHLORIDE MEDIA WITH PHOSPHONIUM IONIC L IQUIDS

Anna CIESZYŃSKA, Magdalena REGEL-ROSOCKA and

Maciej WIŚNIEWSKI

Institute of Chemical Technology and Engineering, Poznan University of Technology, Poznan, Poland, e-mail: [email protected]

Abstract The selective extraction of Pd(II) in the presence of Ni(II), Cu(II), Pb(II),

Fe(III), Rh(III), Ru(III) and Pt(IV) from chloride media with trihexyl(tetradecyl) phosphonium chloride (Cyphos®IL101) and trihexyl(tetradecyl)phosphonium bis(2,4,4-trimethylpentyl)phosphinate (Cyphos®IL104) in toluene was investigated. Over 99% of Pd(II) can be effectively extracted with Cyphos®IL101 and Cyphos®

IL104 from 0.1 M HCl in the presence of Ni(II), Cu(II), Pb(II), Fe(III), Rh(III) and Ru(III). While less than 10% of the other metals is transported to the organic phase. Separation between Pd(II) and Pt(IV) is not so effective and about 70-75% Pd(II) and 60-68% Pt(IV) is extracted. The selectivity of Pd(II) extraction over Pb(II), Fe(III) and Pt(IV) depends upon acidity of aqueous solution and along with increase of HCl concentration selective extraction decrease.

1. INTRODUCTION

At present, a demand for noble metals, particularly platinum group metals (PGM), such as Pd(II), Pt(IV), Ru(III), Rh(III), ect. is observed. These metals are used in electronic, chemical, pharmaceutical and petroleum industries on a large scale. They also found applications as vehicle catalytic converters, catalysts in organic processes and value added components in metal alloys. A gap between demand and natural sources, which are limited, must be replenished by recycling of spent materials, especially spent vehicle catalysts, containing these metals [1, 2]. In recent years solvent extraction has become a suitable method, among a variety of methods, for PGM recovery from low concentrated sources [3]. Various extractants have been studied and proposed for extraction of Pd(II): hydro-xyoximes [4], dialkyl sulphides [5], hydrophobic amines [6,7], pyridinecar-boxamides [8,9], and quaternary ammonium salt, such as Aliquat 336 [10].

Extraction of Pd(II), especially with dialkyl sulphides is a very slow process, which results not only from symmetrical Pd(II) chlorocomplex


49

(PdCl42-) but also from high hydrophobicity of the used extractants. For this

reason new organic reagents are in wide demand to enable fast Pd(II) extraction. Recently, some ionic liquids have been applied as solvent for the extraction of metals and organic compounds [11, 12].

The aim of the work is to study the selectivity of Pd(II) extraction over Ni(II), Cu(II), Pb(II), Fe(III), Rh(III), Ru(III) and Pt(IV) from chloride media with two phosphonium ionic liquids (Cyphos®IL101 and Cyphos®IL104). Cyphos®IL101 has been examined as extractant of Pd(II) from chloride solutions, previously [13].

2. EXPERIMENTAL

Two pure phosphonium ionic liquids: Cyphos®IL101 and Cyphos®IL104, produced by Cytec Industry Inc. (Canada), in the presence of toluene were used as extractants. The structures of the extractants used are presented in Table 1.

Table 1. Structures of extractants used

P+C6H13

C6H13C6H13

C14H29

Cl-

P+

C6H13

C6H13C6H13

C14H29

O

O

P-

CH3

CH3

CH3

CH3

CH3

CH3

CH3

CH3

Cyphos®IL 101 Cyphos®IL104

Extraction was carried out in a typical way with five millimolar solutions of IL in the presence of toluene as an organic phase. Aqueous feeds containing 2.5×10-3 mol dm-3 of each of metals: Pd(II) and Ni(II), or Cu(II) or Pb(II) or Fe(III) or Rh(III) or Ru(III) or Pt(IV), added as chloride in 0.1 - 3 mol dm-3 HCl. Equal volumes (5 cm3) of phases were mechanically shaken for a period of time between 30 s and 10 min at 20ºC and left to stand for phase separation. Metal concentrations were determined in the aqueous solutions by ICP (JY24, Jobin-Yvon, France), at the wave length 340.5, 352.5, 223.0, 220.4, 259.9, 365.8, 240.3, 214.40 nm for Pd(II), Ni(II), Cu(II), Pb(II), Fe(III), Rh(III), Ru(III) and Pt(IV), respectively.

3. RESULTS AND DISCUSSION

The selective extraction of Pd(II) in the presence of other metals from 0.1 and 3 mol dm-3 HCl was studied. Extraction of Pd(II) from acidic media is fast, efficient and selective. Selectivity (SPd/M) is defined as the ratio of distribution ratios of Pd(II) and the other metal according to the formula:

MPdMPd DDS // = . Both Cyphos®IL101 and Cyphos®IL104 are very


50

effective and selective extractants of Pd(II) in the presence of other metals from 0.1 mol dm-3 HCl. Almost 100% of Pd(II) can be effectively extracted in the presence of Ni(II), Cu(II), Pb(II), Fe(III), Rh(III) and Ru(III), while less than 12% of the other metal is transported to the organic phase. Separation of Pd(II) over Pt(IV) is worse and about 70-75% of Pd(II) and 65-68% of Pt(IV) is extracted (Fig.1).

0 2 4 6 8 10

0

1

2

3

4

log

S P

d/M

Time, min.

0 2 4 6 8 10

0

1

2

3

4

log

S P

d/M

Time, min.

Fig. 1. Selectivity of Pd(II) extraction vs. time of extraction with a) Cyphos®IL101

and b) Cyphos®IL104 from 0.1 M HCl solution in the presence of: (■) Ni(II), (●) Cu(II), (▲) Pb(II), (▼) Fe(III), (♦) Rh(III), (◄) Ru(III), (►) Pt(IV).

0 2 4 6 8 10-2

-1

0

1

2

3

4

log

S P

d/M

Time, min.

0 2 4 6 8 10-2

-1

0

1

2

3

log

S P

d/M

Time, min.

Fig. 2. Selectivity of Pd(II) extraction vs. time of extraction with a) Cyphos®IL101 and b) Cyphos®IL104 from 3 M HCl solution in the presence of:

(■) Ni(II), (●) Cu(II), (▲) Pb(II), (▼) Fe(III), (♦) Rh(III), (◄) Ru(III), (►) Pt(IV)

The selectivity of Pd(II) extraction in the presence of Pb(II), Fe(III) and Pt(IV) depends on the concentration of hydrochloric acid. Pd(II)

a) b)

a) b)


51

extraction amounts nearly 90-96% in the presence of Ni(II), Cu(II), Rh(III) and Ru(III) from 3 M HCl, while less than 10% of added metals is transported to organic phase by these phosphonium extractants. The separation between Pd(II) and Pb(II), Fe(III) or Pt(IV) is worse in these conditions. Value of logSPd/M is below 1.3 for Pd(II) over Pb(II) and below 0 for Pd(II) over Fe(III) and Pt(IV) (Fig.2).

4. CONCLUSIONS

The results presented here provide evidence for high selectivity of Pd(II) extraction over Ni(II) and Cu(II), Pb(II), Fe(III), Rh(III), Ru(III), from 0.1 mol dm-3 HCl with Cyphos®IL101 and Cyphos®IL104. Separation for Pd(II) over Pt(IV) is not so effective. With the increase in HCl concentration, to 3 mol dm-3 the selective extraction of Pd(II) in the presence of Pb(II), Fe(III) and Pt(IV) with these extractants decreases. Extraction of Pd(II) from acidic media is fast, efficient and selective, and for this reason examined phosphonium ionic liquids can be applied as extractants.

Acknowledgements

We thank Cytec Industries Inc. for providing us with free samples of Cyphos®IL101 and Cyphos®IL104. This work was supported by the grant No. 32-139/2008-DS.

REFERENCES

[1] M. Wisniewski, J. Szymanowski, Pol. J. Appl. Chem., 1996, 40, 17. [2] P. Young, N.A. Rowson, J.P.G. Farr, I.R. Harris, L.E. Macaskie, Environ. Technol.,

2003, 24, 289. [3] F. L. Bernardis, R. A. Grant, and D. C. Sherrington, React. Funct. Polym., 2005, 65, 205. [4] M. J. Cleare, R. A. Grant, P. Charlesworth, Separation of Platinum Group Metals by Use

of Selective Solvent Extraction Techniques, Extractive Metallurgy, IMM, London, 1981. [5] J.S. Preston, A.C. du Preez, Solvent Extr. Ion Exch., 2002, 20, 359. [6] M. Rovira, L. Hurtado, J.L. Cortina, J. Arnaldos, A.M. Sastre, React. Funct. Polym.,

1998, 38, 279. [7] A. Zhang, G. Wanyan, M. Kumagai, J. Solution Chem., 2004, 33, 1017. [8] I. Szczepanska, A. Borowiak-Resterna, M. Wisniewski, Hydrometallurgy, 2003, 68, 159. [9] M. Regel-Rosocka, M. Wisniewski, A. Borowiak-Resterna, A. Cieszynska, A.M. Sastre,

Sep. Purif. Technol., 2007, 53, 337. [10] P. Giridhar, K.A. Venkatesan, T.G. Srinivasan, P.R.. Vasudeva Rao, Hydrometallurgy, 2006, 81, 30. [11] M. Regel-Rosocka, K. Cieszynska, M. Wisniewski, Przem. Chem., 2006, 85, 651. [12] D.C. Stepinski, M.P. Jensen, J.A. Dzielawa, M.L. Dietz, Green Chem., 2005, 7, 151. [13] A. Cieszynska, M. Regel-Rosocka, M. Wisniewski, Pol. J. Chem.Technol., 2007, 2, 99.


52

KINETICS OF ZINC AND IRON ION EXTRACTION FROM

ACIDIC SOLUTIONS WITH PHOSPHONIUM IONIC LIQUID

Barbara MARSZAŁKOWSKA, Anna SOBCZYŃSKA, Maciej WIŚNIEWSKI


Abstract Studies on kinetics of zinc(II) and iron(III) extraction from model

hydrochloric acid solutions with phosphonium ionic liquid CYPHOS®IL 101 was carried out using modified liquid/liquid interfarcial system: Lewis cell with individual stirring of both phases without interrupting the constant interfacial area. Results of kinetics investigations in Lewis cell proved, that stirring velocity affect zinc(II) and iron(III) transfer.

1. INTRODUCTION

High amount of zinc(II) and iron(III) in spent pickling solutions from hot-dip galvanizing plants is considerable industrial problem. Continuous development of hot-dip galvanizing economy enforces to search advantageous and innovative recycling processes. Solvent extraction with ionic liquids as solvents for metal ion separation is one of currently resolved techniques which can be used to recover Zn(II) and Fe(III) from such solutions [1,2]. Preliminary studies indicated trihexyl(tetradecyl)pho-sphonium chloride (CYPHOS®IL101) as a prospective extractant [3]. However, it is necessary to know a mass-transfer mechanism of extraction with the novel extractant.

From the point of view of process and equipment design it is essential to get to know kinetics of extraction. The most important parameters that affect mass transfer include: interfacial area of mass transfer, liquid phase mixing intensity, presence of interfacial contaminations and disturbances and migration of ions [4-6].

The aim of the work is to determine kinetics of zinc(II) and iron(III) extraction from model HCl solutions with CYPHOS®IL 101. A Lewis cell was used as a contactor with mixing of both phases and keeping a stable interfacial surface [7, 8]. Performed experiments permitted to estimate the mass transfer coefficients.


53

2. EXPERIMENTAL

The composition of model acidic solutions was as follows: 5 g/dm3 (0.077 M) Zn(II) or 5 g/dm3 (0.089 M) Fe(III), 1.8% (0.58 M) HCl, 5 M Cl¯ (adjusted with NaCl). Trihexyl(tetradecyl)phosphonium chloride (CYPHOS®IL101) was used as an extractant (0.08 M). It was dissolved in toluene to overcome drawbacks caused by high viscosity of the IL.

Extraction was carried out in a Lewis cell, which was designed on the base of Plucinski and Nitsch’s data. The volumes of each phase were equal to 0.095 dm3. The interfacial area amounted 0.189 dm3. The mixing rates of both phases were constant (60-120 rpm). Each experiment lasted six hours. Samples (1 cm3) collected from the organic phase were stripped with double excess of ammonia buffer with pH = 10 (Zn(II)) or 1 M sulfuric acid(IV) (Fe(III)) during 30 minutes. Next the contents of metal ions in the aqueous phases were analysed by colorimetric titration.

3. RESULTS

Previous investigations proved that the metal ion transport from the aqueous to the organic phase in Lewis cell can be described by the following differential equations (the initial conditions: 0)0( ww CC = ,

0)0( oo CC = ):

- concentration of metal ion in the aqueous phase:

−⋅−=D

tCtCSk

dt

tdCV w

ww

)()(

)( 0 (1)

- concentration of metal ion in the organic phase:

−⋅=D

tCtCSk

dt

tdCV w

oo

)()(

)( 0 (2)

where: C - concentration, D - distribution coefficient, S - interfacial surface, k - mass transfer coefficient, V - phase volume, t - time, “0” - initial value, o and w state for the organic and aqueous phase, respectively [9].

Assuming that there are small changes of Zn(II) and Fe(III) concentrations and constant distribution coefficients in the experiment, we can apply the following equation:

+⋅−

+−

++

+=

wo

wo

wo

owo

wo

wooow VDV

VDVSk

VDV

CDCV

VDV

VCVCtC

)(exp

)()(

0000

(3)


54

+⋅−

+−

−++

=wo

wo

wo

owo

wo

wwooo VDV

VDVSk

VDV

CDCV

VDV

VCVCDtC

)(exp

)()()(

0000

(4)

Eqs.3 and 4 are applied for estimation of mass transfer coefficients by fitting theroretical results for experimental data (Fig.1). The mass transfer coefficient calculated by the least square method is shown in Tab.1.

a) b)

00

.02

0.0

4

0 12500 25000Time, s

CZ

n,o,

mo

l/dm3

00.

040.

080 12500 25000

Time, s

CF

e(I

II),

o, m

ol/d

m3

Fig.1. Zn(II) (a) and Fe(III) (b) concentrations in the organic phase as a function of time and stirring velocity equal for both phases: ▲ - 60rpm, ♦ - 80 rpm, ○ - 100

rpm, � - 120 rpm , ■ - 140 rpm; lines - calc., symbols – exper.

Table 1. The mass transfer coefficients for extraction of Zn(II) and Fe(III) for

different stirring rates

Stirring velocity rpm kZn, m/s kFe, m/s

60 80 100 120 140

8.44×10-7

1.46×10-6

1.95×10-6

1.97×10-6

2.04×10-6

1.73×10-6 1.66×10-6 3.17×10-6 3.77×10-6 4.7×10-6

The same volumes of both phases ( ow VV = ) and zero concentration of

metal ions in the organic phase ( 0)0( =oC ) enable to estimate the mass transport flux according to the following equation:

+−= t

DV

DSkCkj i

wi

)1(exp0 (5)

where: subscript i = Zn(II) or Fe(III).


55

0.0E+00

5.0E-08

1.0E-07

1.5E-07

2.0E-07

2.5E-07

3.0E-07

40 60 80 100 120 140 160

Stirring velocity, rpm

j, m

ol/m

2 s

Fig. 2. Zinc(II) (♦) and iron(III) (■) molar fluxes versus stirring velocity

(range 60-120 rpm).

4. CONCLUSIONS

Kinetics of extraction in the Lewis cell for the systems studied is controlled by diffusion. Accepted model fitts the experimental data very well. The mass transfer coefficients and fluxes indicate that the iron(III) transport to the organic phase is twice efficient than of zinc(II). Both zinc(II) and iron(III) transfer to the organic phase increases with increasing stirring velocity of both phases. It can result from interfacial diffusion impact, which will be soon investigated.

Acknowledgements

The work was supported by the Polish Ministry of Science and Higher Education with its grant No. 1T 09B 081 30.

REFERENCES

[1] H. Zhao, S. Xia, P. Ma, J. Chem. Technol. Biotechnol., 2005, 80, 1089. [2] G.-T. Wei, Z. Yang, C.-J. Chen, Anal. Chim. Acta, 2003, 488, 183. [3] M. Regel-Rosocka, K. Cieszyńska, M. Wiśniewski, Przem. Chem., 2006, 85, 651. [4] K.M. Samant, K.M. Ng, AIChE Journal, 1998, 44, 2212. [5] H. F. Aly, J.A. Daoud, J. Rad. Nucl. Chem., 1996, 208, 47. [6] S. Tsukahara, Anal. Chim. Acta, 2006, 556, 16.

[7] J. Niemczewska, R. Cierpiszewski, J. Szymanowski, Physicochem. Problems Min. Proc., 2003, 37, 87.

[8] J. Niemczewska, R. Cierpiszewski, J. Szymanowski, Desalination, 2003, 162, 169. [9] A. Sobczyńska, J. Niemczewska, S. Stelmaszyk, K. Alejski, Przem. Chem., 2006, 85,

1142.


56

COMPUTER SIMULATION OF FLOW SHEETS FOR THE

SOLVENT EXTRACTION OF URANIUM: A NEW ROUTE TO DELAY THE EFFECT OF THE CHEMICAL DEGRADATION

OF THE ORGANIC PHASE ON URANIUM RECOVERY FROM ACIDIC SULFATE MEDIA

Alexandre CHAGNES1),*, Solène COLLET1), Bruno COURTAUD2),

Jacques THIRY2), Gérard COTE1) 1) Ecole Nationale Supérieure de Chimie de Paris - ENSCP Université Pierre et Marie Curie – Paris6 - Laboratoire d'Electrochimie et de Chimie Analytique -

UMR7575 CNRS-ENSCP-Paris6 ENSCP, 11 Rue Pierre et Marie Curie, 75231 Paris Cedex 05, France.

2) AREVA-NC, Service d’Etudes de Procédés et Analyses (SEPA), B.P. 71 87250 Bessines sur Gartemps.

Abstract Solvent extraction is the most common process implemented in

hydrometallurgy because it is a sturdy and economic process but many problems occur during the operation of solvent extraction plants: low solvent loading, poor quality product, formation of cruds, precipitates, and emulsions, radiolysis or chemical stresses, etc. All of these problems may be responsible of an increase of the operating cost of the plant and a drop of extraction efficiency.

For example, during solvent extraction of uranium by trioctylamine in tridecanol/kerosene solvent, a chemical degradation of Alamine® 336 occurs in presence of vanadium (V) in the feed solution as follows:

(i) Vanadium (+V) is extracted by Alamine® 336 from aqueous phase to organic phase.

(ii) The strong oxidation power of V(+V) is responsible of the oxidation of the modifier (tridecanol) to carboxylic acid according to radical mechanism. The reduced form of V(+V), i.e. V(IV), is spontaneously transferred from organic phase to aqueous phase due to low interaction between alamine and V(+IV).

(iii) The preceding radicals react quickly with the extractant (Alamine® 336) to form dioctylamine as a degradation product.

The degradation of Alamine® 336 to dioctylamine is responsible of a drop of extraction efficiency as uranium (VI) extraction by trioctylamine is less efficient than uranium (VI) extraction by dioctylamine.

The modification of the flow sheets implemented for the extraction of uranium(VI) is a new route for delaying the drop of extraction efficiency induced by the chemical degradation of the extraction solvent. In this work, the sturdiness of various flow sheets in the face of the chemical degradation of trioctylamine is investigated by computer simulation.


61

REMOVAL OF HEAVY METAL IONS FROM AQUEOUS

SOLUTIONS USING CHELATING RESINS WITH IMINODIACETATE GROUPS

Maria Valentina DINU1), Ecaterina Stela DRAGAN1) and

Andrzej TROCHIMCZUK2) 1) ”Petru Poni” Institute of Macromolecular Chemistry, Aleea Grigore Ghica Voda

41A, 700487 Iasi, Romania, Tel: +40-232217454, Fax: +40-232211299 2) Faculty of Chemistry, Wroclaw University of Technology, Wyspianskiego 27,

50-370 Wroclaw, Poland

Abstract

A chelating ion exchange resin bearing iminodiacetate groups derived from an acrylonitrile – divinylbenzene (AN-DVB) copolymer was synthesized in our group and tested as sorbent for some heavy metal ions like: Cu(II), Co(II), Ni(II), Pb(II), Cd(II) and Zn(II) from aqueous solutions by a batch equilibrium technique. The influence of different parameters like: pH, metal ion concentration on the sorption capacity of the resin for the Cu(II), Co(II) and Ni(II) was examined. The overall tendency of the CIE resin toward Pb(II), Cd(II) and Zn(II), under non-competitive conditions, followed the order: Cd(II) > Pb(II) > Zn(II).

1. INTRODUCTION

The contamination of water resources by heavy metals is a serious environmental worldwide problem. The conventional technologies for the removal of heavy metals from wastewater include chemical precipitation, ion exchange, adsorption, membrane processes and evaporation. Adsorption has attracted attention because of new material types available for the removal process according to application. Some of these material types are zeolites, active carbon, fly ash, biosorbents, chelating resins [1].

Organic chelating resins, by their high selectivity in binding metal ions and metal complex ions, have a major implication in concentration control and in inorganic analytical chemistry. Numerous studies and reviews concerning the synthesis and characterization of the selective chelating sorbents, and the wide applicability of these resins in the removal of toxic metals and complex ions from industrial effluents, as well as in selective metal ion recovery processes, have been recently published [2-4]. Therefore, the adsorption properties of a novel chelating ion exchange (CIE) resin bearing iminodiacetate groups derived from an acrylonitrile -


62

divinylbenzene (AN-DVB) copolymer with 10% of DVB for the removal of heavy metal ions like: Cu(II), Co(II), Ni(II), Pb(II), Cd(II) and Zn(II) from aqueous solutions by a batch equilibrium technique were investigated in this work.

2. EXPERIMENTAL

2.1. MATERIALS

The metal salts used were CuSO4⋅5H2O, CoCl2⋅6H2O, (CH3CO)2Ni⋅4H2O, (CH3CO)2Cd⋅2H2O, Zn(NO3)2⋅6H2O and Pb(NO3)2 (all from Aldrich). Hydrochloric acid, potassium chloride, citric acid and dibasic sodium phosphate (Na2HPO4⋅12 H2O), analytical grade were used for the preparation of HCl-KCl and citrate-phosphate buffer solutions.

2.2. METHODS

The synthesis and the morphological characterization of the AN-DVB copolymer with 10% of DVB and CIE resin were performed as previously shown [5]. The cation exchange capacity was determined according to the method previously presented [6]. The CIE resin was regenerated with HCl 1 M/NaOH 1 M before the metal ion adsorption test. Study of the metal ion retention properties of the CIE resin was carried out using a batch equilibrium procedure. Thus, 0.5 g of dry CIE resin was placed in a flask and contacted with 50 mL of the aqueous solution of each metal ion: Cu(II), Co(II), Ni(II) at the concentration of 0.07 mol/L, each. The influence of pH on Cu(II), Co(II) and Ni(II) metal ions retention was studied for CIE resin and the pH of the metal-ion solution was adjusted before equilibration over a range of 1.2 – 2 with a HCl-KCl buffer solution, and over a range of 2.5 – 6 with a citrate-phosphate buffer solution. The resin was filtered off, and the residual concentration of the metal cation remained in the filtrate was measured by the UV-VIS spectroscopy at 510 nm for Co(II), 720 nm for Ni(II) and 775 nm for Cu(II).

The performance of CIE resin for removal of Cd(II), Pb(II) and Zn(II) metal ions was measured by contacting the known amount of swollen polymer with solution containing metal ions; an amount of resin containing 10-fold molar excess of ligand relative to the metal cations in solution was shaken with 25 mL of the solution for 24 h and then metal concentration was determined using atomic absorption spectroscopy (AAS) with wavelengths set at 228.8 nm for Cd(II), 213.9 nm for Zn(II) and 283.3 nm for Pb(II).


63


CIE resin was prepared according to the method previously presented [5]. The main physical and chemical features of this resin in the Na+-form are as follows: surface area: 5.67 m2/g; mean pore radius: 7.058 nm; cation exchange capacity per dry resin: 4.885 meq/g; water regain: 1.385 g/g.

3.1. ADSORPTION PROPERTIES OF CIE RESIN FOR Cu(II), Co(II), Ni(II)

The preliminary studies [5] prompted us to investigate in detail the sorption properties of heavy metal cations by CIE resin. Fig.1a shows sorption isotherms determined for Cu(II), Co(II) and Ni(II) at 25oC.

a) b)

0,00 0,02 0,04 0,060,0

0,8

1,6

2,4

3,2

4,0

q e, m

mol

M(I

I)/g

resi

n

Ce, mol/L

Cu(II)Ni(II)

Co(II)

0 2 4 6 80,0

0,8

1,6

2,4

3,2

4,0

q,

mm

ol M

(II)

/g re

sin

pH

Cu(II)

Co(II)

Ni(II)

Fig.1. a) M(II) retention as a function of the equilibrium metal concentration for

the CIE resin; b) Metal ion retention as a function of pH, at an initial concentration of 0.07 mol/L, T = 25oC

The maximum uptakes of Cu(II), Ni(II) and Co(II) was calculated using Langmuir equation and were equal to: 3.257, 2.809 and 2.512 mmol per gram of dry resin, respectively.

From Fig.1b, it can be observed that M(II) could hardly be adsorbed by CIE resin when pH < 3. With the increasing pH value, the M(II) retention increased gradually, the optimum adsorption pH of M(II) being located at 5. At pH > 5 the M(II) retention decreased probably because small amount of M(II) started to deposit as M(OH)2. The CIE resin showed a good retention ability for the metal ions in the following order: Cu(II) > Ni(II) > Co(II), at pH = 5 and at the concentration of 0.07 mol/L of metal ions, each.

3.2. ADSORPTION PROPERTIES OF CIE RESIN FOR Pb(II), Cd(II), Zn(II)

Pb(II), Cd(II) and Zn(II) were tested for their affinity to the iminodiacetate ligand anchored to an AN-DVB matrix. Batch-mode


64

sorption studies were performed separately for each cation at 10-fold excess of the ligand relative to the amount of metal cations present in solution. The distribution coefficients of the CIE resin for the cations studied under non-competitive conditions are presented in Tab. 1.

Table 1. Distribution coefficients of the CIE resin for Pb(II), Cd(II), Zn(II)

Resin Cation Distribution coefficient (k)

CIE

Cd(II) Pb(II) Zn(II)

2240 727 292

The distribution coefficients (k) were calculated as the ratio of the amount of metal adsorbed by 1 g of resin and the amount of metal remaining in 1 mL of solution after sorption. It can be seen that the overall tendency of CIE resin toward Pb(II), Cd(II) and Zn(II) under non-competitive conditions follows the order: Cd(II) > Pb(II) > Zn(II). The retention capacity of CIE resin for Cd(II) and Pb(II) ions increased with the increase of the equilibrium metal concentration according to a type I isotherm and the maximum adsorption capacity calculated with Langmuir equation were equal to: 0.812 mmol/g, 0.514 mmol/g and 0.224 mmol/g for Cd(II), Pb(II), and Zn(II), respectively.

3. CONCLUSIONS

The adsorption of Cu(II), Co(II), Ni(II), Cd(II), Pb(II) and Zn(II) on a novel CIE resin bearing iminodiacetate groups was reported in this work. The retention capacity for the metal ions on CIE resin, under non-competitive conditions, showed the following order: Cu(II) > Ni(II) > Co(II) > Cd(II) > Pb(II) > Zn(II).

Acknowledgements

The authors gratefully acknowledge the financial support of this research from the Project CEEX-MATNANTECH C-50/2006.

REFERENCES

[1] R.L. Alberti, M. Casciola, U. Constantino, Encyclopedia of Analytical Chemistry, Academic Press, London, 1995.

[2] A. Kałedkowski, A.W. Trochimczuk, React. Funct. Polym., 2006, 66, 957. [3] C. Noureddine, A. Lekhmici, M. S. Mubarak, J. Appl. Polym. Sci, 2008, 107, 1316. [4] A. M. Donia, A. A. Atia, A. M. Heniesh, Sep. Purif. Technol., 2008, 60, 46. [5] E.S. Dragan, E. Avram, M.V. Dinu, Polym. Adv. Technol. 2006, 17, 571. [6] S. Dragan, G. Grigoriu, Angew. Makromol. Chem. 1992, 200, 27.


65

LIQUID MEMBRANE EXTRACTION OF LIPOPEPTIDES

Krasimir DIMITROV1,2), Frederique GANCEL1), Ludovic MONTASTRUC1), Pascal DHULSTER1), Iordan NIKOV1)

1) Laboratoire ProBioGEM EA 1026, Polytech’Lille, USTL, Lille, France 2) Institute of Chemical Engineering, Bulgarian Academy of Sciences, Sofia,

Bulgaria

Surfactants of biological origin are of increasing interest for many industries due to their chemical diversity, multifunctional characteristics and low toxicity in comparison to synthetic, petrochemical-derived surfactants. A lipopeptide surfactin is one of the most powerful biosurfactants. Microbiological productivities, properties and applications of lipopeptides, including surfactin, have been extensively studied [1-3]. However, there is a lack of information about their separation, purification and concentration. In fact biosurfactants are not yet widely available because of their high production costs, which results primarily from low strain productivities and high recovery expenses.

The selective recovery and concentration of such lipopeptides from fermentation broth largely determines the production cost. The low concentrations and the amphiphilic character of these compounds pose serious limitations to their efficient recovery. Thus, development of efficient separation technologies is of growing interest. The commonly used methods for biosurfactants recovery are foam separation, acid precipitation, and solvent extraction [4]. The latter technique provides higher biosurfactant purity comparing to the other two methods [5]. The main inconvenient of solvent extraction is the problem with regeneration of the loaded solvent, and therefore the use of important quantities of solvent. In addition, the most efficient and generally used for lipopeptides recovery solvents, such as chloroform, methanol, and acetone, are known to be toxic and harmful to the environment and human health.

A prospective trend in biotechnological production of lipopeptides is the process known as in situ product removal (ISPR) in which the product is removed from the bioreactor during its production by an appropriate separation technique. Several integrated bioprocesses have been proposed to optimise productivity and cost-effectiveness of low and high molecular weight molecules [6]. Recently, the interest of coupling of production of surfactin by fermentation with its simultaneous recovery by adsorption was demonstrated [7, 8]. The purity of surfactin isolated after desorption with methanol and solvent evaporation is high, but the process is relatively long.


66

A potential advance for lipopeptides recovery from fermentation broth is the application of the low-cost liquid membrane process. This separation technique, based on solvent extraction, is called pertraction and operates in three-liquid-phase systems. Pertraction process is a combination of extraction and stripping operations performed simultaneously in one stage [9]. The main advantages of pertraction towards classical liquid-liquid extraction are the use of smaller quantities of organic solvent due to continuous regeneration of the solvent, as well as the possibility to recover the target species even in cases of low distribution coefficients [9]. Pertraction allows producing of valuable products of high quality at reduced costs, because of possibility to use as liquid membranes less powerful but more selective, less toxic and less expensive solvents than in the case of conventional solvent extraction. The interest of liquid membrane process for recovery of fermentation products have grown rapidly. Liquid membrane technique was successfully applied for recovery of some bioactive substances from fermentation broths [10, 11], but there are no data on lipopeptides recovery by using pertraction processes.

The recovery of the microbial lipopeptide surfactin from model aqueous solutions was studied. To confirm the applicability of the liquid membrane process to surfactin isolation from aqueous media, including fermentation broth, some properties, in particular, solubility and pH stability of the surfactin were studied. Usually, the B.subtilis straines used for surfactin production have been cultivated in medium with pH = 6.0-8.5 [7]. Consequently, lipopeptide extraction from such media was studied. To improve surfactin recovery, a possible small modification of pH was envisaged, too. However, acidification of the aqueous media containing surfactin was quite limited, because of its precipitation at pH < 5.5.

The effect of pH on the equilibrium distribution of surfactin between various organic solvents and aqueous solutions was studied. The most polar from studied solvents 1-octanol provided practically complete lipopeptide extraction from aqueous media in all studied pH-interval (pH = 5.5-9.0). However, the back extraction of lipopeptide into an aqueous solution and therefore, the regeneration of the loaded organic phase after the extraction was very difficult. In contrast, the non-polar n-heptane and n-octane were clearly less efficient solvents, but the degree of surfactin removal into these solvents was found to be strongly affected by the aqueous solution acidity. For both studied alkanes, the degree of surfactin extraction was relatively high from slightly acid (over than 80 % at pH = 5.5) or slightly basic (over than 60 % at pH = 9.0) aqueous solutions, while from neutral aqueous solutions the extraction was limited (less than 10 % at pH = 7.0-7.5). Consequently, the studied alkanes are suitable for liquid membrane permeation of surfactin, providing conditions suitable for lipopeptide


67

extraction into organic solvent (at pH = 5.5-6.0), but also conditions favourable for its back extraction into an aqueous solution (at pH = 7.0-7.5). The observed unusual pH effect of relatively high extraction degrees from both acid and basic media and noticeably reduced degree of extraction from neutral media could be explained to the different conformations of lipopeptide in these media. The observed minimum of degree of extraction from neutral media could be attributed to the higher micropolarity of the β-sheet micelles formed by surfactin molecules at these conditions. This configuration is characterised by an exposure of a large number of carboxylic groups on the micelle surface which could explain the relatively polar character of surfactin. In both acid and basic media, surfactin conformation alters from β-sheet to α-helices [3]. At this configuration, the non-polar ends of lipopeptide molecules are more exposed to contact the organic solvents and, as result, higher extraction degrees were obtained.

Surfactin permeation through a liquid membrane of n-heptane was studied in a laboratory rotating discs contactor. Batch pertraction process was carried out at different acidities of the feed solution. The obtained results of lipopeptide transport in the three-liquid-phases system show that surfactin can be successfully recovered from slightly acid media (pH = 5.5-6.0), including fermentation broth, by means of pertraction. The process efficiency grows with decrease of pH of the feed solution (83 % recovery at pHfeed = 6.05 and 97 % at pHfeed = 5.65 after 4 h pertraction). The pertraction process was very rapid: about 90 % of surfactin was removed from feed solution in 30 min only.

The efficient permeation of surfactin through a liquid membrane offers a new opportunity to isolate lipopeptide from fermentation broth. A further coupling of fermentation with pertraction in a new ISPR process could provide a relatively low-cost production of lipopeptides with high purity. This integrated bioprocess could also contribute to resolve the problem with foam formation during fermentation of biosurfactants.

REFERENCES

[1] E. Akpa et al., Appl. Biochem. Biotechnol. - Part A, 2001, 91, 551. [2] M.S. Yeh, Y.H. Wei, J.S. Chang, Biotechnol. Prog., 2005, 21, 1329. [3] M. Osman, H. Hoiland, H. Holmsen, Y. Ishigami, J. Peptide Sci., 1998, 4, 449. [4] J.D. Desai, I.M. Banat, Microbiol. Mol. Biol. Rev., 1997, 61, 47. [5] H.L. Chen, R.S. Juang, Biochem. Eng. J., 2008, 38, 39. [6] K. Schügerl, J. Hubbuch, Current Opinion in Microbiology, 2005, 8, 294. [7] T. Liu, L. Montastruc, F. Gancel, L. Zhao, I. Nikov, Biochem. Eng. J., 2007, 35, 333. [8] L. Montastruc, T. Liu, F. Gancel, L. Zhao, I. Nikov, Biochem. Eng. J., 2008, 38, 349. [9] L. Boyadzhiev, Z. Lazarova, Liquid membranes (liquid pertraction), in: Membrane

Separation Technology. Principles and Applications, Elsevier, Amsterdam, 1995. [10] J. Zigova, E. Sturdık, D. Vandak, S. Schlosser, Process. Biochem., 1999, 34, 835. [11] M. Di Luccio et al., Desalination, 2002, 148, 213.


68

NEW APPROACH TO HEAVY METALS EXTRACTION

FROM SEWAGE SLUDGE

Myroslav SPRYNSKYY and Bogusław BUSZEWSKI

Chair of Environmental Chemistry and Bioanalytics, Faculty of Chemistry, Nicolas Copernicus University, 7 Gagarina Str., 87-100 Toruń, Poland

tel. (48)(56)6114330, fax. (48)(56)6114837, e-mail: [email protected]

Abstract

This contribution presents the results of the study of the selected heavy metals removal from the sewage sludge using a new technique of solid- liquid-solid extraction with the adsorption-diffusion column filled by the zeolite. The metal extraction onto the zeolite from aqueous system of the clinoptilolite and the sludge composition is characterized by three stages: intensive extraction, inversion and stabilization with the moderate extraction increase.

1. INTRODUCTION

Use of natural sorbents or related materials for removal of heavy metals or immobilization of them seems to hold much promise for sewage sludge treatment. Application of different additives to sewage sludges for immobilization of the excessive amounts of heavy metals is one of the ways for improvement physical and chemical properties of the substratum [1-4].

The aim of the present work is to investigate removal of the heavy metals (Cr, Cu, Zn, Ni and Pb) from the sewage sludge using the raw natural zeolite (clinoptilolite rock) and pre-treated zeolites in kinetic and equlibrium studies. New technique of solid-liquid-solid extraction has been used for heavy metal uptake by zeolite from medium of sewage sludge – zeolite solution.

2. MATERIALS AND MRTHODS

The sludge samples have been collected from the postfermentation tanks of the communal wastewater treatment plant, Torun (Poland). The dewatered product contains 3.1 % of total nitrogen, 2.3 % of total phosphorus (P2O5), 20.5 % of total organic carbon, 0.82 % of potassium, 8.6 % of calcium, 3.7 % of magnesium and total concentrations of the heavy metals, mg/kg: Pb - 38.1 ±1.1; Cd - 13.0 ±0.9 Cr - 303.4 ±6.8; Cu - 240.40 ±1.2; Ni -402.8 ±5.4.


69

Clinoptilolite rock from Sokyrnytsya deposit (the Transcarpathian region, Ukraine) containing about 75% of clinoptilolite proper was used. This sorbent should be resolutely referred to natural micro - mesoporous materials with the polymodal pore size distribution. Heterogeneity and hierarchy of porosity of the clinoptilolite rock is shown in Fig.1. The modified clinoptilolite samples were obtained by three different methods including chemical treatment by HCl and NaCl as well as ultrasonic treatment.

Fig.1. Heterogeneity and hierarchy of porosity of the clinoptilolite rock

In kinetic study 10 g samples of dewatered sludge with 120 mL of distilled water and the perforated polyethylene columns filled by 1 g of the natural zeolite (grain size 0.71-1.00 mm) were placed into seven 200 mL polyethylene bottles (Fig.2).

Fig.2. Installation for solid-liquid-solid extraction of heavy metals.


70

The perforated polyethylene columns filled by zeolite played a role of adsorption-diffusion medium for heavy metals removal from sewage sludge. This process may be defined as “solid-liquid-solid" extraction, because heavy metals pass from organic and mineral particles of sewage sludge to the zeolite particles and aqueous solution is a carrier in the studied medium. The obtained three-component compositions were stirred by a mechanical shaker for 0.25, 0.5, 1, 2, 4, 8 and 24 hours. After that the zeolite-filled columns were withdrawn and washed with 10 mL of distilled water with following addition of the washing water to sludge. Then sludge was dried at 105oC and mineralized. The obtained mineralization solutions were used for analysis of the heavy metal contents by FAAS (AAS Analyst 800 Perkin Elmer, Shelton, USA)..


Extraction of the heavy metals by the clinoptilolite from aqueous solution of the clinoptilolite and the sludge composition is characterized by three clearly defined stages: the first stage of intensive extraction, the second stage of inversion with prevalence of the metals desorption, and the third stage of stabilization with the moderate extraction increase and attainment of steady state (Fig.3). The rate coefficients estimated by pseudo-second order kinetic and intraparticle diffusion models during the first stage are 5 - 10 times more than the same during the third stage.

Fig.3. Kinetics curves of heavy metals extraction (a) and the scheme of the mass transfer in ‘clinoptilolite – sewage sludge aqueous solution’ (b).

Efficiency of the metal uptake by the clinoptilolite evaluated in the

equilibrium study grows with increase of the zeolite dose added to the sewage sludge. Addition of 25% of the zeolite provides extraction efficiency of cadmium and lead of about 84 %, chromium, copper and nickel of 66, 61 and 50% relatively. Addition of 10 % of the zeolite previously treated by 2M solutions of NaCl, HCl and by ultrasound shows


71

increase of efficiency of the metals uptake for 2 - 14% . The estimated values of -∆G show that the heavy metal extraction by the clinoptilolite should be considered as a physical adsorption. According to the added zeolite amounts Gibbs energy of extraction varies in the range of 19 - 44, 14 – 49, 8 – 15, 16 - 20 and 7 - 10 kJ/mol for Pb, Cd, Cu, Cr, Ni. The values of -∆G suggested the following selectivity series of the zeolite: Cd > Pb > Cr > Cu >Ni.

It is apparent that mass transfer of the heavy metals onto the clinoptilolite are caused by difference in energetic potentials of particles of the sewage sludge and the zeolite and their seeking an energetic equilibrium under the specified thermodynamic conditions (see Table 1). In such a manner energetic difference in system ‘clinoptilolite – water - sewage sludge’ causes processes of redistribution of energy towards to its decrease for all components of the system. This redistribution of energy is accompanied by redistribution of substances as an energy carrier in the system and by exchange between a complex of the sorbed cations of the zeolite, the heavy metals of the sewage sludge and the dissociated water molecules.

Tab.1. The standard Gibbs energy (kJ/mol) for mineral-water interactions in dilute

systems [5, 6]

Clino-ptilolite [5]

Fe-Mg smectite [5]

Ca- saponite [5]

Lime [5]

Iron (III) hydroxide [5]

Soil humic substances [6]

19021 1262 1345 604.0 696.5 6752

The heavy metals are accumulated onto the clinoptilolite thanks to

their ability to more significant decrease of free energy of the clinoptilolite in comparison with its exchangeable cations of alkaline and earth-alkaline metals.

4. CONCLUSIONS

The obtained data suggested that adsorption-diffusion column filled by the zeolite may be used to effective removal of the heavy metals from sewage sludges. The difference between energetic potentials of the zeolite and sludge particles is a driving force of the metal redistribution in the aqueous system “sewage sludge – zeolite”. This process may be defined as solid-liquid-solid extraction, because heavy metals pass from particles of sewage sludge to the zeolite ones and aqueous solution is a carrier in the studied medium.

The technique of solid-liquid-solid extraction used in this study may be also applicable for heavy metals removal from other type of sewage


72

sludge as well as adsorption-diffusion columns may be filled by other sorbents.

REFERENCES

[1] H.W. Ryu, H.S. Moon, E.Y. Lee, K.S. Cho, H. Choi, J. Environ. Qual., 2003, 32, 751. [2] Wang Yi, Yang Mu, Zhao Quan-Bao, Yu Han-Qing, Sep. Purif. Technol., 2006, 1. [3] X.D. Li, C.S. Poon, H. Sun, I.M.C. Lo, D.W. Kirk, J. Hazard. Mater., 2001, 82, 215. [4] J.W.C. Wong, A. Selvam, Chemosphere, 2006, 63, 980. [5] Qualification of Thermodynamic Data for Geochemical Modeling of Mineral-Water

Interactions in Dilute Systems. ANL-WIS-GS-000003 REV 00 4-12 US, Nevada 89144, 2004, Available via http://www.osti.gov/bridge/purl.cover.jsp

[6] H.-R. Schulten, Environ. Toxicol. Chem, 1999, 18, 1643.


77

BORON REMOVAL FROM SEAWATER REVERSE OSMOSIS

PERMEATE BY SORPTION-MEMBRANE FILTRATION HYBRID METHOD

Marek BRYJAK1), Nalan KABAY2), Enver GULER2),

Jakub PIEKACZ3), Idil YILMAZ-IPEK 2), Mithat YUKSEL2)

1)Wroclaw University Technology, Depart Polym Carbon Mater, Wroclaw, Poland 2)Ege University, Depart Chemical Eng, Izmir, 35100 Turkey

3)Nicolaus Copernicus University, Chemistry Faculty, Torun, Poland

Abstract The boron problem associated to the seawater reverse osmosis (SWRO)

desalination has not been solved yet. The rejection of boron by RO is not sufficiently high; so about one third of feed boron content (~1.5 mg B/L) is normally found in the obtained permeate. When every country adopts the WHO recommendation for permissible boron concentration, it is not difficult to predict the problems that will be faced by managers and engineers of desalination plants. The main objective of this study is to evaluate the possibility to use sorption-membrane filtration hybrid system for boron removal from first stage seawater reverse osmosis (SWRO) permeate.

1. INTRODUCTION

The reverse osmosis (RO) membrane desalination process is an efficient and reliable technology for the production of drinking water from seawater. However, some serious limitations had recently been discovered during field practice. Among them, the boron problem seems to be a critical issue. The WHO recommends maximum concentration limit of boron as 0.5 mg/L for drinking water. It is reported that, this limit is rarely reached for conventional reverse osmosis desalination plants equipped with commercially available membranes. The associated boron problem with the seawater reverse osmosis desalination process had been detected clearly during the last few years. Seawater contains approximately 5 mg/L of boron. The rejection of boron by RO is not sufficiently high, so about one third of boron content (~1.5 mg/L) is normally found in the produced permeate.

There are several methods applied for seawater desalination and removal of boron from wastewaters: i) to use of improved RO membranes with higher B-rejection, ii) to increase pH of the water to be treated by


78

caustic soda (or other base) prior to RO membrane, and reacidifying the treated water after the membrane to bring it to the desired acidity, iii) to pass the desalinated water through two extra passages of RO treatment, iv) to apply electrodialysis after RO treatment.

In this study, an advanced separation process for boron removal has been put forward by combining boron sorption on a fine powdered boron selective resin with a complex separation on microfiltration membranes (Figure 1). The proposed membrane-based hybrid process integrates sorption efficiency with membrane separation of the selective resin in one step. It is shown that use of sorbent in the form of very fine particles improves the separation efficiency such that sorbent shows higher uptakes and enhances the rate of the separation process.

Fig. 1. Flow-sheet of the sorption-membrane filtration hybrid process.

2. EXPERIMENTAL

RO tests were performed by reverse osmosis system which was established in Izmir-Urla region. Dowex (XUS 43594.00) ion exchange resin with a particle size of 0-20 µm was employed for boron removal from RO permeate. A submerged-style hollow-fiber module containing two polypropylene membranes (diameter: 1.3 mm, thickness: 200 µm, pore diameter: 0.4 µm) was used throughout the sorption-membrane filtration hybid system. The tests were performed using boron selective ion exchange resins at different concentration in suspension and also at different flow

Seawater RO Permeate (SWP)

W – pure water B – boron brine

S – regenerated Binding material

Sorption B+S→→→→BS

Membrane Separation

Regeneration BS→→→→B+S

Membrane Separation

Complex BS

Mixing SWP+S+B


79

rates of fresh and saturated resins. At a certain time intervals, permeate samples were taken from the system and the boron concentration was determined. The analysis of boron was performed spectrophotometrically using Azomethine-H method (λmax: 415 nm).

3. RESULTS

As seen in Figure 2, boron selective resins (Dowex XUS 43594.00) showed a great performance for elimination of boron from natural seawater RO permeate containing 1.7 mg B/L when sorption-membrane filtration hybrid process was employed. The concentration of the boron in the product water was decreased to 0.3 mg B/L in 5 min. The boron level was almost kept constant during 3 hrs of operation. The scattered values were considered to be due to the small differences between the flow rates of fresh and saturated resins. In order to check the dilution effect in the submerged hollow fiber module, some tests were carried out with only RO permeate without using boron selective resin. As shown in Figure 2, boron concentration decreased to 1.5 mg B/L from 1.7 mg B/L in 30 min and almost kept constant at 1.4 mg/L throughout the whole test period (blank test). Totally, 17% of decrease was obtained due to the dilution effect. However, the decrease in boron level reached to 82% in 5 min when boron selective resins were used in the hybrid system.

0.000.200.400.600.801.001.201.401.601.80

0 20 40 60 80 100 120 140 160 180Operation Time [min]

B [mg/L]

Blank

Dowex (XUS-43594)

Fig.2. Boron removal from SWRO permeate by means of sorption-membrane

filtration hybrid process (Blank: data obtained without using boron selective resin)


80

4. CONCLUSIONS

It was shown that sorption-membrane filtration hybrid process could be an alternative method to the commonly used fixed-bed column-mode operation for boron removal from natural seawater RO permeate. Boron selective resins remove boron very quickly from RO permeate due to their large available surface areas and significantly improve the efficiency of the water deboronation process.

Acknowledgement

This study was supported by MEDRC (04-AS-004) and Ege University Scientific Research Project (EU-2007-MUH-015). We thank Ege University, Faculty of Fisheries for opportunity to conduct our SWRO tests in Urla. JP appreciates Socrates-Erasmus Student Exchange Programme for supporting his study at Ege University. We are also grateful to Dow Chem for Dowex ion exchange resins.

REFERENCES

[1] N.Kabay, I.Yilmaz, M.Bryjak M.Yüksel, Desalination, 2006, 198, 74. [2] I.Yilmaz, N.Kabay, M.Bryjak, M.Yuksel, J.Wolska, A.Koltuniewicz, Desalination,

2006, 198, 310. [3] I.Yilmaz, N.Kabay, M.Yuksel, R.Holdich, M.Bryjak, Sep. Sci. Technol., 2007, 42(5),

1013. [4] N.Kabay, M.Bryjak, S.Schlosser, M.Kitis, S.Avlonitis, Z.Matejka, I. Al-Mutaz,

M.Yuksel, Desalination, 2008, 223, 38. [5] M.Bryjak, J.Wolska, N.Kabay, Desalination, 2008, 223, 57. [6] N.Kabay, S.Sarp, M.Kitis, H.Koseoğlu, O.Arar, M.Bryjak, R.Semiat, M.Yuksel,

Desalination, 2008, 223, 49. [7] N.Kabay, O.Arar, F.Acar, A.Ghazal, M.Yuksel, Desalination, 2008, 223, 63. [8] H.Koseoglu, N.Kabay, M.Yuksel, M.Kitis, Desalination, 2008, 223, 126. [9] H.Koseoglu, N.Kabay, M.Yuksel, S.Sarp, O.Arar, M.Kitis, Desalination, 2007, 227,

258. [10] N.Kabay, S.Sarp, M.Yuksel, O.Arar, M.Bryjak, React Func Polym, 2007, 67, 1643. [11] I.Yilmaz Ipek, R.Holdich, N.Kabay, M.Bryjak, M.Yuksel, React Func Polym, 2007,

67, 1628. [12] M.Bryjak, G.Pozniak, N.Kabay, React Func Polym, 2007, 67, 1635. [13] H.Parschova, E.Mistova, Z.Matejka, L.Jelinek, N.Kabay, P.Kauppinen, React. Func.

Polym, 2007, 67, 1622.


89

SEPARATION OF Cu(II) AND Pb(II) BY PLASTICIZED

MEMBRANES WITH β-CYCLODEXTRIN DIMER AS THE ION CARRIER

Cezary KOZŁOWSKI1) and Władysław WALKOWIAK2)

1) Institute of Chemistry and Environment Protection, Jan Dlugosz University of Czestochowa, 42-201 Czestochowa, Armii Krajowej 13, Poland

2) Chemical Metallurgy Division, Faculty of Chemistry, Wroclaw University of Technology, 50-370 Wroclaw, Wybrzeze Wyspianskiego 27, Poland

Abstract In the present work the hydrophobic β-cyclodextrin dimer have been used as

macrocyclic ion carriers for separation of metal ions from dilute aqueous solutions by transport across polymer inclusion membranes. The dimer used as ionic carrier for competitive transport of Cu(II) and Pb(II) ions shows preferential selectivity order: Cu(II) > Pb(II). The selectivity of Cu(II) over Pb(II) in the transport through polymer inclusion membrane from aqueous nitrate solutions was found to be high, especially for pH of source phases equal 5-6.

1. INTRODUCTION

Crown ethers, calixarenes, and cyclodextrins (CDs) are typical examples of metal ion receptors in host-guest systems [1]. Due to high selectivity of crown ethers, they have been extensively used as extractants in solvent extraction and as the ion carriers in transport across the liquid membranes [2-4].

Recently, a novel type of liquid membrane system, called polymer inclusion membranes (PIMs) has been developed. The PIMs are appropriate for facilitated transport of ions and consist of a polymer (for example cellulose triacetate - CTA), a plasticizer (for instance o-nitrophenyl alkyl ether), and an ion carrier. The resulting membrane is used to separate the feed and receiving aqueous phases. For example, CTA membranes were used for carrier mediated transport of metal ions from aqueous solutions [5,6].

Kozlowski et al. [7] reported a new type of macrocyclic carriers based on polymers of β-cyclodextrin with alkenyl (nonenyl and dodecenyl) succinic anhydride derivatives. Several cyclic oligomers were synthesized and studied in CTA/ o-nitrophenyl pentyl ether (ONPPE) membranes for the competitive transport of Cu(II), Co(II), Ni(II) and Zn(II) and gave the selectivity order Cu(II) > Co(II) > Ni(II) > Zn(II). The authors suggested


90

that transport involves the formation of ion pairs between hydroxyl groups on the polymer and the metal cation. The addition of dinonylnaphthalenesulfonic acid to the membrane produced a synergistic effect that helped effectively in the removal of Cr(VI), Cu(II) and Cd(II) from industrial wastewaters and municipal sludge.

Recently, it was reported that β-CD polymers prepared by cross-linking of β-CD with 2-(1- docosenyl)-succinic anhydride derivatives show selectivity for lead(II) over other metal ions in the transport across plasticized membranes. For competitive transport of Pb(II), Cu(II), and Zn(II) ions through PIM the selectivity order was found to be as Pb(II) >> Cu(II) > Zn(II) [8].

In the present work we prepared β-CD dimer and used it as the ion carrier. β-CD dimer formed by cross-linking of β-CD with p-toluene-sulphonic anhydride was synthesized. The compound was used as the ion carrier in plasticizer membranes for competitive transport of Pb(II) and Cu(II) from nitrate aqueous solutions.

2. EXPERIMENTAL

Reagent analytical grade inorganic chemicals, i.e. Cu(NO3)2, Pb(NO3)2, CH3COONa, and HCl were obtained from POCh (Gliwice, Poland). Organic reagents, i.e. cellulose triacetate (CTA), o-nitrophenyl penthyl ether (ONPPE), 4,4’-methylene-bis(benzenesulpholyl chloride), β-cyclodextrin and dichloromethane were also purchased from Fluka and used without further purification.

Synthesis of β-cyclodextrin dimer: To a suspension of β-cyclodextrin in pyridine, 4,4’-methylene-bis(benzenesulpholyl chloride) was added. Mixture was vigorously stirred in 250 mL round bottomed flask for next 24 hours at 60 0C. Transfer the filtrate to conical flask and add access acetone. After the product, β-cyclodextrin dimer, was washed by acetone.

CH2O S CH2 S CH2O

O

O O

O

tosyl β-cyclodextrin dimer

The solution of CTA was prepared by dissolving an appropriate

amount of CTA in 15 cm3 of chloroform. A separate methyl chloride solution containing known amounts of ONPPE and β-CD dimer was


91

prepared. These solutions were mixed (total volume 25 cm3) and ultrasonicated for 15 min. to form the casting solution of PIM. Finally, this casting solution was spread on a 9 cm diameter flat-bottom glass Petri dish which was kept on a leveled surface. The Petri dish was covered with a glass plate and in such a way that the aeration was possible, and any cross-contamination was prevented.

The transport experiments were carried out in a permeation module cell described in our earlier paper [8].

To describe the efficiency of metal ion removal from the feed phase, the recovery factor (RF) was calculated:

% 100⋅−

=i

i

c

ccRF (1)

where c is the metal ion concentration at given time in the feed phase (M), ci is the initial concentration of metal ion in the feed phase (M).

From the slope of the straight line obtained when representing the metal concentration in the stripping phase in function of time, the flux (J) can be calculated according to the following equation:

⋅

=dt

dc

S

V J (2)

where V is the volume of the aqueous stripping phase, S is the exposed surface area of the membrane and C is the concentration of metal ions at elapsed time.

3. RESULTS

The competitive transport of Cu(II), and Pb(II) cations across PIM with β-CD dimer was investigated. In the case of PIM with β-CD dimer, the exponential increase of total flux with the solute concentrations in the feed phase was observed. The maximal values at 0.3 M β-CD dimer concentration in the membrane was achieved, and equal to 9.82 µmol/m2s.

The selectivity of Cu(II) over Pb(II) in the transport through polymer inclusion membrane from aqueous nitrate solutions was found to be high, especially for pH of source phases equal 5-6. In the case of competitive transport of Pb(II), and Cu(II) through plasticized immobilized membranes the selectivity of process is controlled via formation of ion pairs of ionizable groups with metal cations. In Fig. 1 the recovery factors for Cu(II) and Pb(II) vs. pH of source phase are shown; the selectivity order was found to be Cu(II) >> Pb(II).


92

1 2 3 4 5 6 7

20

40

60

80

100

RF

, %

pH of source phase

Cu(II)

Pb(II)

Fig. 1. The recovery factors (RF) for Cu(II), and Pb(II) in the transport process

across PIM with 0.30 M β-CD dimer. Feed phase: 0.0010 M metal ions, pH was stabilized by pH-sat. Receiving phase: 1.0 M HCl. PIM: 4.0 cm3 ONPPE / 1.0 g

CTA, 0.30 M carrier concentration; VF (cm3) : VR (cm3) = 5000 : 50

Acknowledgements

Financial support of this work was provided by the Polish Science Foundation (grant 4T09C 03230).

REFERENCES

[1] W. Sliwa, T. Girek, Heterocycles, 2003, 60, 2147. [2] W. Walkowiak, G. Ndip, R.A. Bartsch, Anal. Chem., 1999, 71, 1021. [3] R.A. Bartsch, E.-G. Jeon, W. Walkowiak, W. Apostoluk, J. Membr. Sci., 1999, 159,

123. [4] A. Gherrou, H. Kerdjoudj, R. Molinari, P. Seta, E. Drioli, J. Membr. Sci., 2004, 228,

194. [5] C.A. Kozlowski, J. Kozlowska, W. Pellowski, W. Walkowiak, Desalination, 2006, 198,

149. [6] M. Ulewicz, U. Lesinksa, M. Bochenska, W. Walkowiak, Sep. Purif. Technol., 2007, 54,

299. [7] C.A. Kozlowski, T. Girek, W. Walkowiak, J.J. Koziol, Sep. Purif. Technol., 2005, 46,

136. [8] C.A. Kozlowski, W. Walkowiak, T. Girek, J. Membr. Sci., 2008, 310, 312.


97

THE EFECT OF V, W AND MO SPECIATION ON THEIR SORPTION ONTO SORBENTS HAVING POLYHYDROXY

FUNCTIONAL GROUPS

Luděk JELÍNEK, Richard BURDA, Jana ĎURIŠOVÁ, Helena PARSCHOVÁ and Eva MIŠTOVÁ

Department of Power Engineering, ICT Prague, Technická 5, 166 28 Czech Republic

Abstract In this work, speciation of metal oxoanions is discussed with respect to

formation of isopolyanions and changes in metal valence. Oxoanions of all the studied metals are able to polymerize in acidic solution. Sorption properties of bulk polyanions are different from that of monomeric oxoanions. Though the gel diffusion through the sorbent particle is slower, the amount of adsorbed metal is higher. In the case of vanadium, it is possible to electrochemically reduce V(V) to V(IV). Adsorption of V(IV) onto studied sorbent was negligible.

1. INTRODUCTION

Adsorption of metal oxoanions onto sorbents having polyol functional groups was thoroughly studied in our group on both natural and synthetic sorbents [1,2]. Tungsten oxoanions were studied in connection with their possible clinical applications [3,4]. During these experiments interesting sorption properties of bulk tungsten heteropolyanions were observed. The work was then continued to elucidate the sorption kinetics and other properties [5].

Similar properties can be found in the sorption of molybdenum. However, unlike the case of tungsten, Mo sorption is more pH sensitive. In strongly acidic solutions molybdate(VI) oxoanions can turn to molybdenyl cations MoO2

2+ [6]. At higher pH molybdenum polyanions are turned into simple molybdates MoO4

2-. This can have strong impact on molybdenum adsorptivity onto polyol sorbents.

Chemistry of vanadium is even more complicated than that of molybdenum. Like molybdenum, vanadate(V) can be turned into a form of vanadyl(V) cation VO2

+. Yellow vanadate(V) oxoanions can also be reduced into blue vanadyl(IV) cations VO2+ [7].

Changes of adsorptivity of above mentioned species can play an important role in their mutual separations.


98

2. EXPERIMENTAL

For sorption experiments sorbents having 1-deoxy-1-(methylamino)-D-glucitol functional group Purolite S-108 and Purolite D-4123 (Purolite International) were used along with experimental sorbents having pyrocatechol and pyrogallol functional groups (Fig. 1). Sorbents were protonized by 1M HCl solution, washed thoroughly by deionized water and used in batch-wise and column experiments.

a) b) c)

Fig. 1. Structures of used polyol sorbents a) glucitol b) PK c) PG

Metal solutions were prepared from their sodium and ammonium salts of analytical grade purity. pH of sorption solutions was set up with HCl and NaOH solutions. If necessary, solution was buffered by HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid) buffer.

Vanadium was reduced at 3D cathode made of carbon felt in laboratory electrolyzer designed at Department of Inorganic Technology, ICT Prague. Reduction was carried out in potentiostatic mode. Potential of electrode was measured by Ag/AgCl reference electrode and controlled by potentiostat Voltalab PGZ 301 (Radiometer Analytical, France).

Metal concentrations were determined by ICP-OES Optima 2000DV (Perkin-Elmer Instruments, USA). pH was measured by pH meter InoLab Level 1 (WTW, Germany).


Polyoxoanions

Formation of oxoanions can be illustrated by pH dependence of Mo sorption that is shown in Figure 2. As can be seen, optimal pH for Mo sorption is about 3.5 for both pyrocatechol (PK) and pyrogallol (PG). It is in accordance with theory. At lower pH, sorption decreases as Mo polyoxoanions are turn to molybdenyl cation, whereas at higher pH sorption steeply decreases.

In the case of tungsten there is no decrease in acidic region and the decrease of sorption with increasing pH is only slight.

Sorption of vanadium is similar to the case of Mo. However, photo-catalyzed reduction of vanadium on the loaded ion exchanger accompanied by the color change was observed. Contrary to the expectations, it resulted in an increase of sorption capacity. As can be seen in Figure 3, an increase of V(V) concentration is very slow. Formation of mixed-valence oxoanions

N CH2

CH3 H

OH

OH

H

H

OH

H

OH

CH2OHPolymerOH

PolymerN

OH

OHCH2

CH3

OH

OH

PolymerN

OH

OHCH2

CH3


99

had probably positive effect on the complex formation with glucitol sorbent. Vanadium can also be adsorbed from neutral and alkaline solution onto sorbent in free base form.

Fig. 2. Effect of pH on Mo(VI) sorption Fig. 3. Sorption of V(V) and V(IV) onto phenolic sorbents onto glucitol sorbent

Valence change

Unlike in the case of spontaneous reduction on loaded adsorbent, pure V(IV) prepared by electrochemical reduction of V(V) showed almost no adsorptivity on the glucitol sorbent. As can be seen in Figure 3, V(IV) breaks through the column readily and reached the inlet concentration within 60 BV.

4. CONCLUSIONS

All the studied oxoanions can be adsorbed onto polyol sorbents from mildly acidic solutions. Anionic species of tungsten(VI) are adsorbed in widest range of pHs 2-6. In the case of molybdenum(VI), optimal pH range is about 3-4. Sorption of vanadium(V) is complicated by its reduction to V(IV) on the loaded sorbent. This reduction is photo-catalyzed. Pure V(IV) has very low adsorptivity onto glucitol sorbents.

Both pH dependence of complex formation and valence changes can be used in mutual separations of above mentioned metals.

Acknowledgement

The financial support from Research Plan MSM CZ 6046137304 is gratefully appreciated.

pHe

2 3 4 5 6 7 8

% A

0

20

40

60

80

100

PKPG

BV

0 200 800 1000

c/c 0

0.0

0.2

0.4

0.6

0.8

1.0

V(IV)V(V)


100

REFERENCES

[1] Matejka, Z., Ruszova, P., Parschova, H., Jelinek, L., Kawamura, Y., Advances in Chitin Science, 2002, 6, 213.

[2] Matejka, Z et al.: Fundamentals and Applications of Anion Separations, Chapter 15, 249-261 ACS, Kluwer Publ. House, 2004.

[3] Jelinek, L. et al.: Proc. Ars Separatoria 2003, Zloty Potok n. Czestochowa, Poland, June 2-5, 2003, p. 74-76.

[4] Stephan, H. et al.: Proc. ICSS&T Congress, Supramolecules, 5-11. 9. 2004, Prague, Chem. listy, 2004, 98(S), s35-s36.

[5] Jelinek, L. et al.: Proc. Ars Separatoria 2004, Zloty Potok n. Czestochowa, Poland, June 10-13, 2004, p. 86-88.

[6] Bartecki, A. and Dembicka, D.: J. Inorg. Nucl. Chem., 1967, 29, 2907. [7] Cotton, F.A. and Wilkinson, G.: Advanced Inorganic Chemistry, 5th edition, John Wiley

& Sons, 1988.


111

URANIUM SORPTION ON ORGANOBENTONITE

Marek MAJDAN, Agnieszka GAJOWIAK,

Agnieszka GŁADYSZ-PŁASKA, Stanisław PIKUS

Faculty of Chemistry UMCS, PL MC Skłodowskiej 2, 20-031 Lublin, POLAND

Abstract

The data concerning the uranium (VI) adsorption on organobentonite are presented. The concentration of uranium (VI) in the bentonite phase was evaluated based on the equation referring to the molar fractions of the complexes: UO2

2+, UO2 (OH)+, UO2 (OH)2 , (UO2 )2 (OH)2

2+, UO2 (OH)3- , UO2 (OH)4

2- , (UO2 )3 (OH)5

+, (UO2 )3 (OH)7- .

1. INTRODUCTION

The elimination of the uranyl ions UO22+ from the aquoeus

environment appears in connection with the reprocessing of the radiochemical wastes [1], as well as from these originating from the uranium mining industry . The natural adsorbents, among them also bentonite [2], are the good alternative for the conventional ion-echange resins. There are several papers concerning the sorption of uranyl ions by montmorillonite [3], but the explanation of the sorption mechanism is unambiguous. The aim of this report is the investigation of the pH influence on the UO2

2+ ions retention by the bentonite modified by hexadecyltrimethylammonium bromide (HDTMA).


The organobentonite was prepared by its sodium form with hexadecyltrimethylammonium bromide solutions according to procedure given in [4]. The comparison of the pH influence on the uranium sorption on organobentonite is given on Fig. 1. The percentages of the adsorbent modification related to its cation exchange capacity CEC were: 117, 96, 68, 12% The following conditions for the sorption were preserved: time of the phases equilibration 6 h, solid /liquid ratio-0.1g/100 ml, temperature 230C. The pH of the aqueous phase was controlled through the addition of HCl or NaOH. The concentration of the total U(VI) concentration in the aqueous phase was determined spectrophotometrically using Arsenazo III method. The concentration of the initial solutions solutions prepared from UO2(CH3COO)2 was always 0.0001 mol/dm3. The increase of the uranium


112

sorption in the region 3 to 6 pH is noticed, then depending on the adsorbent modification, the decrease of the uranium concentration in the bentonite phase or its constancy is noticed. The increase of the U(VI) sorption in the acid pH range results from the partition of the cations: UO2

2+ , UO2OH+, (UO2)2(OH)22+, (UO2)3(OH)5

+ , UO2(OH)2 to the bentonite phase. In turn the decrease of the cb in the basic pH range or its constancy appears as the consequence of the existence of UO2(OH)3

- , UO2(OH)42- , (UO2)3(OH)7

- anionic complexes, which are present in the aqueous phase and can be attracted by the positively charged bentonite surface, modified by quaternary alkylammonium cations. The Fig.2 and Table 1 are the result of the quantitative evaluation of U(VI) sorption for the HDTMA-bentonite ( modification 68%). The solid line refers to cb values calculated based on the semiempirical equation:

log cb = log[ K αUO22+ + K11 αUO2(OH)

+ + K12 αUO2(OH)2 + K22 α (UO2)2(OH)22+

+ K13 αUO2(OH)3- + K14 αUO2(OH)4

2- + + K35 α(UO2)3(OH)5

+ + K37 α(UO2)3(OH)7-] (1)

where α refers to the molar fraction of the particular uranium hydroxocomplex in the aqueous phase, calculated based on the stability constants of the complexes using Medusa Program [5-6]. The symbols K denote the proportional factors referring to the particular complexes. It results from Table 1 that complexes UO2(OH)+ , (UO2)3(OH)7

- do not play essential role in the uranium sorption, because their proportional factors are negative. The neutral complex UO2(OH)2 plays the dominant role in the uranium transfer to the solid phase, since its proportional factor is the highest One can conclude that the charged surface of HDTMA-bentonite is heterogeneous in the sense of the sign of the charge. In some places of the surface the charge is positive, because the proportional factors of the negatively charged complexes are positive.

Table 1. The proportional factors of the uranium complexes appearing in the aqueous phase.

Complex Proportional Factor Value [mol/g] UO2

2+ K 0.000013 UO2 (OH)+ K11 -0.000088 UO2 (OH)2 K12 0.00024

(UO2 )2 (OH)2 2+ K22 0.00012

UO2 (OH)3- K13 0.000047

UO2 (OH)42- K14 0.00011

(UO2 )3 (OH)5+ K35 0.00013

(UO2 )3 (OH)7- K37 -0.000036


113

Fig.1. The change of the uranium concentration in the bentonite phase with pH (the

numbers denote the percentage of bentonite modification referring to cation exchange capacity CEC).

.Fig. 2. The result of uranium sorption modeling on HDTMA-bentonite (68% modification; solid line refers to Eq.(1)).


114

3. CONCLUSIONS

The effective removal of the uranium from the aqueous solutions using HDTMA-bentonite is possible in the basic pH range. The semiquantitative evaluation of the adsorption data shows that the bilayer of the surfactant cations on the adsorbent surface starts to form, when the percentage of the adsorbent modification is below the CEC.

REFERENCES

[1] H. Rameback, Y. Albinsson, M. Skalberg, U.B. Eklund, L. Kjellberg, L. Werne, J. Nuclear Mat., 277 (2000) 208-214.

[2] M.T. Olguin, M. Solache-Rios, D. Acosta, P.Bosch, S. Bulbulian, J. Radioanal. Nucl. Chem., 218, 1 (1997) 65-69.

[3] E.R. Sylwester, E.A. Hudson, P.G. Allen, Geochimica et Cosmochimica Acta, 64 ,14 (2000) 2431-2438.

[4] M. Majdan, O. Maryuk, S. Pikus, E. Olszewska, R. Kwiatkowski, H. Skrzypek, J. Mol. Struct., 740 (2005) 203-211.

[5] I. Puigdomenech, "INPUT, SED, and PREDOM: Computer programs drawing equilibrium diagrams", Technical Report TRITA-OOK-3010 (ISSN 0348-825X), 12 pp., Royal Institute of Technology, Dept. Inorg. Chem., S-100 44 Stockholm (1983).

[6] I. Puigdomenech , "Windows software for the graphical presentation of chemical speciation", in: 219th ACS National Meeting. Abstracts of Papers, Vol.1. American Chemical Society, San Francisco, CA, March 26-30, 2000. Abstract I&EC-248.


115

TETRAD EFFECT IN THE ADSORPTION OF THE

LANTHANIDES ON ZEOLITE A FROM CHLORIDE MEDIUM

Agnieszka GŁADYSZ-PŁASKA, Marek MAJDAN, Darek STERNIK

Faculty of Chemistry UMCS, 20-031 Lublin, PL MC Skłodowska 2, Poland

Abstract In this study the changes of the distribution constants Kd of the lanthanide

chlorides in the system: zeolite A (solid phase) - sodium chloride (aqueous phase) were investigated. The evident convex tetrad effect in the change of logKd values within the lanthanide series was noticed and the attempt of its explanation through the comparison of covalence in Ln-O bonds existing in ≡Al-O(1/3Ln)-Si≡ species found in the zeolite phase and Ln(H2O)x

3+ complexes formed in the aqueous phase was presented.

1. INTRODUCTION

The aim of this paper is to show, that the adsorption of the lanthanides on the zeolite would be an effective tool in the investigation of the tetrad effect and that the existence of convex and concave character of the effect is very complicated in nature and needs further explanation.

We decided to use the zeolite A in the investigation, because it is known in nuclear industry as a good material for the immobilization of the different radionuclides coming from the reprocessing of nuclear fuel, among them also lanthanides. Therefore thorough understanding of rules governing the lanthanides adsorption on zeolite A is necessary.

The tetrad effect in the lanthanide chemistry discovered independently by Siekierski [1] on the one side and by Peppard [2] on the other, relies on the characteristic changes of the physical and chemical properties of the different lanthanide compounds with the division of the lanthanide series into four subgroups (tetrads): La-Nd, Pm-Gd, Gd-Ho, Er-Lu. The effect was explained from the viewpoint of the ligand field theory, which takes into account the interaction of the ligands with f-orbitals of the lanthanide tripositive ions in their complexes and from the viewpoint of parameters of interelectronic repulsions (known as Racah parameters) in the lanthanide ion, or rather their changes during formation of the covalent complex species. The problem is quite carefully verified for the liquid– liquid systems, particularly between organic phase and water phase, whereas the tetrad effect in the adsorption of lanthanides on different types of sorbents


116

or generally at solid – water interface was reported slightly [3, 4, 5]. The occurrence of this effect in the adsorption systems from aqueous solutions depends mainly on many external factors, because the adsorption mechanisms of Ln3+ from aqueous solution are obviously more complex than those in the liquid - liquid and precipitate - liquid systems. The tetrad effect in the adsorption of lanthanides from aqueous solution needs further theoretical and experimental study.

2. EXPERIMENTAL

The aqueous phases (100 cm3 volume) containing lanthanide chlorides (0.0005 mol/dm3; 99.9% purity, Sigma Aldrich) dissolved in sodium chloride with concentrations ranging from 0.001mol/dm3 to 2 mol/dm3 (pure, Sigma Aldrich) were equilibrated through 4h with 100 mg samples of the sodium form of the zeolite A (PQ Corporation) in temperature 23±10C. The aqueous phase was separated from the solid residue by filtration (paper filter Filtrak 390, Polskie Odczynniki Chemiczne) and the concentration of the lanthanides was determined spectrophotometrically using Arsenazo III [6]. The concentration of the lanthanide in the solid phase cZe was found as the difference between the initial concentration cin and the concentration in the equilibrium aqueous phase caq. The initial and equilibrium pH values were controlled using combined glass electrode (Sigma Chemical Co.) connected to the pH meter (CX-731 type, Elmetron Co.)


The change of the distribution constants of the lanthanides in the system LnCl3-NaCl – zeolite A with the chlorides concentrations is given in Fig. 1. The distribution constants Kd are defined as:

Kd = (cZe/ caq)V/m (1),

where: V and m denote the volume of aqueous phase and the mass of the adsorbent respectively.

The nonmonotous change of log Kd values with the atomic number of the tripositive lanthanides is evident. One can note a positive deviation from the straight-line relationship: log Kd vs. Z for the concentration of chlorides in the aqueous phase ranging from 0.01 to 2 mol/dm3 for I (La-Nd), III (Gd-Ho) and IV (Er-Lu) tetrad. With the rise of Cl- ions concentration the tetrad effect is more convex. Second tetrad (Pm-Gd) was not considered since we do not have the data for Pm. The appearance of the convex tetrad effect for the lanthanides: La-Nd (I tetrad), Gd-Ho (III tetrad) and Er-Lu (IV tetrad), visible on Fig. 1., is the consequence of the stronger


117

covalent Ln-O bonds in products of the adsorption reaction than in the substrates. The aquoions Ln(H2O)x

3+ present in the aqueous phase can be treated as substrates. In other words the covalence of the Ln-O bond in the products of the reaction phase ≡Al-O(1/3Ln3+)-Si≡ in the zeolite is higher. With the rise of chloride concentrations the Ln-O bond in the substrates of reaction is more ionic, because aquaions Ln(H2O)x

3+ and LnCl2+ are formed.

La Ce Pr NdPmSmEu Gd Tb Dy Ho Er TmYb Lu

log

Kd

-1,0

-0,5

0,0

0,5

1,0

1,5

2,0

0,001mol/dm3

0,01mol/dm3

0,1mol/dm3


log

Kd

-0,6

-0,4

-0,2

0,0

0,2

0,4

0,6

0,8

0,2mol/dm3

0,4mol/dm3 0,6mol/dm3 0,8mol/dm3 1,0mol/dm3


log

Kd

-0,4

-0,2

0,0

0,2

0,4

0,6

1,2mol/dm3 1,4mol/dm3

1,6mol/dm3 1,8mol/dm3 2,0mol/dm3

Fig.1. The changes of distribution constants (Kd) of the lanthanides in the system NaCl – zeolite A with chloride concentrations.

According to Kawabe suggestions [7-9], the tetrad effect appears when

there is noticeable difference in Racah parameters of f electrons in products and substrates of complexation reaction. In connection with this one can conclude that Racah parameters in aquoions of the lanthanides Ln(H2O)x

3+,


118

treated as substrates of the adsorption reaction, are higher in comparison with products of the reaction in the zeolite phase ≡Si-O(1/3Ln3+)-Al≡; in the other words the covalency in the products of the adsorption is higher than in substrates.

REFERENCES

[1] S. Siekierski, Polish J. Chem., 1992, 66, 215 [2] D.F. Peppard, G.W. Mason, S. Lewey, J. Inorg. Nucl. Chem., 1969, 31, 2271. [3] F. Coppin, G. Berger, A. Bauer, S. Castet, M. Loubet, Chem. Geology 2002, 182, 57. [4] M. Majdan, A. Gładysz-Płaska, S. Pikus, D. Sternik, O. Maryuk, E. Zięba, P. Sadowski,

J. Mol. Struct. 2004, 702, 95. [5] A. Gładysz-Płaska, M. Majdan, S. Pikus, W. Lewandowski, J. Colloid. Interf. Sci. 2007,

313, 97. [6] Z. Marczenko, M. Balcerzak, "Spektrofotometryczne metody w analizie nieorganicznej",

Wydawnictwo Naukowe PWN SA, Warszawa 1998, p.353. [7] A. Ohta, I. Kawabe, Geochemical Journal, 2000, 34, 455. [8] A. Ohta, I. Kawabe, Geochemical Journal, 2000, 34, 439. [9] I. Kawabe, A. Ohta, N. Miura, Geochemical Journal 1999, 33, 181.


119

MODIFICATED POLYMERS TOWARDS RHENIUM SORPTION AND DESORPTION

Joanna DĄBROWSKA, Dorota JERMAKOWICZ-BARTKOWIAK

Wrocław University of Technology, Faculty of Chemistry, Department of Polymer and Carbonaceous Materials, WybrzeŜe Wyspiańskiego 27, 50-370 Wrocław,

Poland; e-mail [email protected]

Abstract The rhenium sorption from hydrochloric and sulphuric acid solutions was

studied on anion exchangers. Resins were obtained by modification of vinylbenzyl chloride/divinylbenzene 2% (VBC/DVB) copolymer with cyclic amines: cyclohexylamine, hexamethyleneimine and cycloheptylamine. Resins have the following properties: capacity 2.54, 3.35, 1.21mmol/g, and 0,001 M HCl regain:0.56, 3.67, 0.45g/g, respectively. Sorption ability of rhenium from 0.1 M HCl, 1 M and 3M H2SO4 acid solutions was investigated. It was evaluated rhenium extraction from multicomponent solution of metals: Au, Pt, Pd, As, Re, Mo in 0.1 M HCl synthetic solution. The initial concentration of Re ions was 50 mg/l. The sorption decreased in the order: (Resin 1) Au>Re>Pd>Pt>As>M, (Resin 2) Au>Pd>Pt>Mo>As>Re, (Resin 3) Au>Pd>Pt>As~Mo>Re, respectively. The desorption rhenium was carried out with using 5% ammonium hydroxide solution.

1. INTRODUCTION

Rhenium is the last naturally occuring element to be discovered that is the one of the rarest elements in nature. It occurs in scattered form, and because of complicated and costly methods for its separation, it belongs to group of the most expensive metals on Earth [1,2].

Commercial source of rhenium are mainly copper and molybdenum ores [4,5], but perrhenate ions are also in seawater and many minerals [3]. World-wide supply of this metal is assessed to 17000 tons [1,6]. Rhenium is produced as a by-product of molybdenum production in the form of pure salts [4], metallic rhenium, ammonium perrhenate and perrhenic acid [1,2].

There are many methods for obtaining rhenium. Now preferred technology of rhenium recovery is selective sorption of perrhenate anions on functionalized resins follow by adsorption by ammonium hydroxide. As a product, ammonium perrhenate is obtained [1,3]. The aim of this work is to present the properties of prepared resins for rhenium sorption from acidic solutions.


120

2. EXPERIMENTAL AND RESULTS

Resins were prepared by exchange of chlorine atoms by cyclohexylamine, hexamethyleneimine and cycloheptylamine. The synthetic route to amino functional resins can be described schematically in Figures 1-3.

Fig. 1. Reaction modification of VBC/DVB using cyclohexylamine, (Resin 1).

Fig. 2. Reaction modification of VBC/DVB using hexamethyleneimine, (Resin 2).

Fig. 3. Reaction modification of VBC/DVB using cycloheptylamine, (Resin 3)

The properties of obtained anion-exchangers were characterized by water regain, ion-exchange capacity, sorption and desorption of rhenium. Batch method was applied to sorption - desorption studies under static conditions and room temperature during 48h. Schematic reactions of sorption and desorption of rhenium are presented in Reaction 1.

R-NH2·HCl → R-NH3+·Cl-

R-NH3+·Cl- + ReO-

4 → R-NH·3+ReO-

4 + Cl- [7]

R-NH3+ReO-

4 + NH4OH → R-NH2 + NH4ReO4 + H2O

Reaction 1. Schematic reactions of rhenium sorption and desorption.

Characteristic and properties of obtained resins are shown in Table 1. The effect of acid concentration on sorption of Re(VII) was also investigated. As presented in Figs. 4-6 Resin 2 shows particularly strong preferences for


121

Re(VII) sorption in 0.1 M HCl solution. It is sorbed 500 mg Re/g resin. It is seen that sorption from 1M and 3M H2SO4 4 solutions was lower. Resin 1 reveals sorption of 292 mg Re/g from 1M H2SO4 solution and Resin 2 can sorb 169 mg Re/g from 3M H2SO4 solution. The best sorption was obtained in 0.1M HCl solution.

Table 1. Characteristic and properties of resins.

Regain Re

Capacity H2O

0,001 HCl

Maximal* sorption

Sorption** Desorption** Resin Ligand

mmol/g g/g g/g mg/g %

1 2 3

Cyclohexylamine Hexamethyleneimine

Cycloheptylamine

2.54 3.35 1.21

0.40 0.96 0.43

0.56 3.67 0.45

387 500 375

42 29 19

18 43 39

* data from the isotherm sorption in 0.1 M HCl of [Re] 50-1500 mg/L ** multicomponent solution in 0.1M HCl of [Re] 50 mg/L, [Au] 47 mg/l, [Pt] 48 mg/L, Pd [26] mg/L, [As] 22 mg/L , [Mo] 158 mg/L

Co, Re mg/L

0 200 400 600 800 1000 1200

Rhe

nium

sor

ptio

n m

g/g

dry

poly

mer

0

100

200

300

400

0.1M HCl1M H2SO4

3M H2SO4

Co, Re mg/L

0 200 400 600 800 1000 1200 1400

Rhe

niu

m s

orp

tion

mg/

g dr

y po

lym

er

0

100

200

300

400

500

0.1M HCl1M H2SO4

3M H2SO4

Fig.4. The isotherms sorption of

rhenium for Resin 1. Fig.5. The isotherms sorption of

rhenium for Resin 2.

Co, Re mg/L

0 200 400 600 800 1000 1200 1400

Rhe

nium

sor

ptio

n m

g/g

dry

poly

mer

0

100

200

300

400

0.1M HCl1M H2SO4

3M H2SO4

Fig.6. The isotherms sorption of rhenium for Resin 3.


122

3. CONCLUSIONS

Incorporation of cyclohexylamine, hexamethyleneimine and cycloheptylamine ligands to copolymer VBC/DVB matrix leads to series of anion exchange resins with capacity of 1.2 – 3.5 mmol/g.. The Resins are useful for selective sorption of ReO4

- from multicomponent solution of Au,

Pt, Pd, As, Re and Mo. The best selectivity was determined for Resin 1 modificated with cyclohexylamine. The results of this investigation indicate that potential use of these resins for separations of rhenium over other metals from acidic solutions.

Acknowledgements

We thank the Minister of Science and Higher Education for supporting this work through Grant N205 046 31/2046.

REFERENCES

[1] D. Jermakowicz- Bartkowiak, ARS Separatoria 2007, 106 . [2] K. Leszczyńska- Sejda, G. Benke, A. Chmielarz, S. Krompiec, S. Michalik, M. Krompiec, Hydrometallurgy, 2007, 89, 289. [3] G. Benke, K. Anyszkiewicz, K. Hac, K. Litwinionek, Lejda. Leszczyńska- Sejda, Przem. Chem., 2006, 8-9, 793-797. [4] A.G. Kholmogorov, O.N. Konopa, S.V. Kachin, S.N. Ilyichev, V.V. Kryuchkov, O.P. Kalyakina, G.L. Pashkov, Hydrometallurgy, 1999, 51, 19-35. [5] T. Sato, K. Sato, Hydrometallurgy, 1990, 25, 281-291. [6] R. Champer, Z. Śmieszek, A. Chmielarz, K. Anyszkiewicz, G. Benke, K. Litwinionek, R. Kalinowski, Rudy i Metale NieŜelazne, 2004, 9, 441-444 . [7] G. Huifa, S. Jinglan, M.A. Hughes, Hydrometallurgy, 1990, 25, 293-304.


123

SELECTIVE SORPTION OF GERMANIUM OXOANION BY

COMPOSITE SORBENTS WITH HYDROUS OXIDES OF CERIUM AND ZIRCONIUM

Eva MIŠTOVÁ1), Helena PARSCHOVÁ1), Luděk JELÍNEK1), Ferdinand

ŠEBESTA2) 1) Department of Power Engineering, ICT Prague, Czech Republic

2) Center of radiochemistry and radiating chemistry and Department of Nuclear Chemistry, CTU Prague, Czech Republic

Abstract In this work, composite sorbents with hydrous oxides of cerium and zirconium

were used for selective removal of Ge oxoanion. Experiments were carried out by dynamic column sorption. The sorption capacity of Ge(IV) anion was the best at the pH of 9 for both sorbents, when sorption capacity for cerium oxide sorbent was about 1100 mg/L and for zirconium oxide sorbent 4300 mg/L. For regeneration of cerium and zirconium oxide sorbents 10 and 15 BV of 1 mol/L HCl was used, respectively.

1. INTRODUCTION

Germanium is naturally occurring in argyrodite (Ag8GeS6), zinc ores and coal (coal ash as much as 20 – 280 mg/kg) [1,2]. Majority of germanium is produced as a by-product of zinc refining.

Germanium has many applications, principally in fibre optics communication networks, infrared vision system and polymerisation catalysts. Certain Ge compounds are often used in medicines and nutriments because of their activity against certain bacteria. These applications can be the source of Ge in surface and waste waters.

Ge can form complexes whit –OH groups of organic compounds [3], as catechol (1,2-dihydroxybenzene) [4]. Recovery of germanium from water solution can be achieved by polymeric sorbents containing e.g. N-methylglucamine [5-7] or iminodiacetate (IDE) [7]. Inorganic sorbents, such as hydrous oxides of Al, Fe, Zr and Ti [8] or activated carbon [9] can also be used.

In this work, composite sorbents with hydrous oxides of cerium and zirconium were used for selective removal of Ge oxoanion. Sorbent CeO2.nH2O/XAD-7 is composite of nonionogenic resin Amberlite XAD-7 and cerium oxide. Sorbent ZrO–PAN is composite of zirconium oxide and


124

polyacrylonitrile matrix. Both sorbents have amphoteric functional groups. For the sorption of Ge oxoanion, sorbents were conditioned by hydroxide solution.

2. EXPERIMENTAL

Experiments were carried out by dynamic column sorption. For the experiments concentration of Ge(IV) in the feed solution was 1 – 3 mg/L and concentration of chlorides and sulfates was 100 – 1000 mg/L. The experimental pH values were 6, 9 and 11. The effects of pH, flow rate, concentration of accompanying anions in the feed solution and desorption of Ge(IV) were studied.

3. RESULTS AND DISCUTION

The influence of pH on the Ge(IV) oxoanion sorption, when c(Ge) = 3 mg/L, c(Cl-, SO4

2-) = 100 mg/L and specific flow rates s = 6 BV/h, was studied at pH = 6, 9 and 11. The sorption capacity of Ge(IV) anion was highest at the pH of 9 for both sorbents, where the sorption capacity was 1060 mg/L for CeO2.nH2O/XAD-7 and 4325 mg/L for ZrO–PAN.

900

1500

2100

2700

3300

3900

4500

0 200 400 600 800 1000

c(Cl-, SO42-)[mg/L]

q [m

g/L

]

Ce

Zr

Fig.1 Effect of concentration of accompanying anions in the feet solution to

sorption capacity

In the case of sorbent ZrO–PAN increased concentration of accompanying anions had a negative effect on the sorption capacity (Fig. 1). When the concentration of sulfates and chlorides in the feed solution was increased from 100 mg/L to 500 mg/L at pH = 9 the sorption capacity


125

was similar (about 4300 mg/L) but further increase of concentration of accompanying anions to 750 mg/L and 1000 mg/L resulted in decrease of sorption capacity by 15% and by 45%, respectively. In the case of sorbent CeO2.nH2O/XAD-7 the negative effect of concentration of chlorides and sulfides was not observed. The selectivity of ZrO–PAN was lower than that of CeO2.nH2O/XAD-7.

The effect of flow rate was studied at the specific flow rates 3 BV/h to 24 BV/h. In the case of the sorbent CeO2.nH2O/XAD-7 the differences among sorption capacities, was maximal about 35%. For sorbent ZrO–PAN was maximal about 70%.

Recovery of Ge(IV) from CeO2.nH2O/XAD-7 provided good yields (Fig.2). Ge was completely striped within 10 BV of 1M HCl regeneration solution. In the case of recovery of Ge oxoanion from ZrO–PAN 15 BV of 1M HCl had to be used. However, only about 70% of adsorbed Ge was recovered. During the next sorption cycle after regeneration, increased concentration of Ge in the eluent was found in first 60 BV. The recovery about 90% was obtained using of 15 BV of 2M HCl. However, the sorbent was not very stable in more concentrated solution of acid and hydroxide. Frequent use of these solutions limited the life time of this sorbent.

0

200

400

600

800

1000

1200

0 5 10 15BV

c(G

e)[m

g/L

]

Ce

Zr

Fig.2 Desorption curves of Ge – 1M HCl


126

Acknowledgements

This work was carried out with financial support of Research proposal (MSM 6046137304) from Ministry of Education, Youth and Sports of the Czech Republic.

REFERENCES

[1] H. Remy, Anorganická chemie, SNTN Praha, 1972. [2] P. Pitter, Hydrochemie, VŠCHT Praha, 2nd edition, 1990. [3] M. Mikešová, M. Bartušek, Coll.Czech.Chem.Comm., 1979, 44, 3256. [4] Jpn. Kokai Tokkyo Koho , JP 58213838, 1983. [5] Z. Matějka, H. Parschová, P. Ruszová, L. Jelínek, P. Houserová, E. Mištová, M. Beneš,

M. Hrubý, Fundamental and Applications of Anion Separation, Abstracts of Papers, 222nd ACS National Meeting, Chicago, IL, United States, August 26-30, 2001

[6] U. Schilde, E. Uhlemann, Reactive Polymers, 1994, 22, 101. [7] I. Ozawa, K. Saito, K. Sugita, K. Sato. M. Akiba, T. Sugo, Journal of Chromatografy A,

2000, 888, 43. [8] A.D. Abbasov, F.S. Mammadova, Kimia Problemlari Jurnali, 2006, 2, 257. [9] J.P. Marco-Lozar, D. Cazurla-Amoróz, A. Linares-Solano, Carbon, 2007, 45, 2519.


127

SORPTION OF HEAVY METAL IONS BY

POLYELECTROLYTE COMPLEX HYDROGEL MEMBRANES

Magdalena GIERSZEWSKA-DRUśYŃSKA and

Jadwiga OSTROWSKA-CZUBENKO

Chair of Physical Chemistry and Physicochemistry of Polymers, Faculty of Chemistry, Nicolaus Copernicus University

Gagarina St. 7, 87-100 Toruń, Poland

Abstract

Ionically crosslinked chitosan/tripolyphosphate membrane (Ch/TPP) was sy-nthesized and used to adsorb heavy metal ions: Cu(II), Zn(II), Cd(II) from aqueous solutions. Batch adsorption experiments were carried out and the effect of pH of solution and an initial metal ion concentration Co were analysed. It was shown that the amount of metal ion adsorption increased with increasing pH and Co.

1. INTRODUCTION

Fast industrial development is connected with increasing amount of waste waters that are highly dangerous for our environment. Heavy metal ions are main components among others water contaminations. They are not biodegradable and tend to accumulate in living organisms, causing various diseases and disorders. Adsorption is one of the well known method used for waste water treatment and heavy metal ion sorption [1].

Till now, activated carbon was widely used as a sorbent for industrial effluents treatment. Because of it’s high cost and requirement of complexing agents for improving its removal performance the new adsorbent materials are searched. Such materials like fly ash, silica gel, zeolites, lignin, clay materials etc. have been extensively investigated for the removal of heavy metal ions, like Cd(II), Cr(III), Hg(II), Pb(II) [2]. A special group able to metal ion sorption are natural polymers, including chitosan [1]. In comparison to the other natural polymers obtained from seafood wastes chitosan has highest chelating ability.

To overcome small resistance of pure chitosan membranes in acidic waste water solutions, the crosslinking process has been proposed [1]. In this work ionically crosslinked chitosan membranes were synthesized using sodium tripolyphosphate (TPP) as the crosslinking agent. The sorption


128

capacity of Cu(II), Zn(II), Cd(II) ions by Ch/TPP hydrogel membranes was studied.

2. EXPERIMENTAL

2.1. MATERIALS

Two different commercially available chitosan samples: high molecular weight chitosan (Ch-HMW) and medium molecular weight chitosan (Ch-MMW), sodium tripolyphosphate (TPP) were purchased from Sigma-Aldrich (Germany). Acetic acid (HAc), zinc nitrate (Zn(NO3)2), copper nitrate (Cu(NO3)2), cadmium nitrate (Cd(NO3)2), sodium acetate (NaAc), sodium hydroxide (NaOH), hydrochloride acid (HCl) and nitric acid (HNO3) were analytical grade and were purchased from POCh (Poland). Chitosan selected for this study varied in their molecular weight (MW), but had similar degree of deacetylation (%DD). Degree of deacetylation of chitosan, determined by potentiometric titration method [3], was equal to 75.7 ± 3.8 (Ch-MMW) and 79.5 ± 1,5 (Ch-HMW). The viscosity average MW, determined by viscosity analysis of chitosan solutions according to Il`ina and Varlamov [4], was equal to 730 kDa (Ch-MMW) and 980 kDa (Ch-HMW).

2.2. POLYELECTROLYTE COMPLEX HYDROGEL MEMBRANES PREPARATION

Pure chitosan membranes were prepared by casting and solvent evaporation technique. Filtered, degassed 1 wt.% chitosan solution in 2 wt.% aqueous HAc solution was cast as film on clean glass plate and evaporated to dryness in an oven at 300 K, then further dried under vacuum at the same temperature.

Chitosan/sodium tripolyphosphate (Ch/TPP) membranes were prepared by dipping pure chitosan membranes in 1.3 wt.% aqueous TPP solution for a proper time period. Conditions of polyelectrolyte complex preparation was as follows: crosllinking time 1h, pH of TPP solution 5.5 (adjusted by adding a small amount of concentrated HCl solution), temperature 277 K. The obtained Ch/TPP membranes were additionally thoroughly washed in deionised water and then dried similarly as chitosan films. Dried polymer films were stored in a dessicator over P2O5 at ambient temperature.

2.3. SORPTION EXPERIMENTS

Aqueous solution of Cu(II), Zn(II), Cd(II) ions was prepared by dissolving metal nitrate salt in deionized water. The total metal concentration varied from 3.2 mM to 16 mM. The initial solution pH was


129

adjusted in the range 2.2-8.2 by adding a small amount of HNO3 or NaOH. The solution pH was selected such that no metal hydroxide precipitations were experimentally found in bulk solution.

In adsorption experiments the mixture of dry membrane (0.05g) and metal ion solution (75 cm3) was placed in a 0.1 dm3 glass flask. The flask was shaken and placed for 48 h in thermostatic bath (T = 298.0 ± 0.1 K). After equilibrium the concentrations of metal ions were analyzed using an atomic absorption spectrometry. The amount of metal adsorbed qM (mol/kg) was obtained using Eq.(1):

W)VC(Cq MM /0 −= (1)

where C0 and CM are the initial and equilibrium concentrations in aqueous phase (mol/m3), respectively, V is the volume of the solution (m3) and W is the weight of dry membrane (kg). Each experiment was triplicate at least under identical conditions.


Fig. 1 shows the effect of pH on adsorption of Cu (II) ions (qM) onto Ch-MMW/TPP polyelectrolyte membrane.

Fig. 1. Equilibrium adsorption of Cu (II) ions at different initial metal ion concentration on Ch-MMW/TPP membrane

The obtained results indicate that the adsorption capacity both the

Cu(II) ions, as well as Zn(II) and Cd(II) ions, depends on the initial metal ion concentration and pH of solution. Value of qM increases with increasing pH, up to specified pH value (pH~4.7 for Cu(II) ion adsorption). It is due to


130

competitive adsorption of protons and metal ions. Amine and hydroxyl groups of chitosan and phosphate groups of TPP may interact with metal ions through different mechanisms (chelation and ion exchange/ electrostatic attraction) depending on pH, as was shown in Fig. 2 [5,6].

Fig. 2. Interaction mechanisms of Cu(II) ions with Ch/TPP hydrogel membrane: chelation (a), ion-exchange/electrostatic attraction (b, c)

4. CONCLUSIONS

The results obtained in this study show that chitosan/tripolyphosphate membrane can be used as an effective material to remove heavy metal ions from low concentrated salt solutions. The adsorption capacity of Cu(II), Zn(II) and Cd (II) ions is highly dependent on pH and initial metal ion concentration.

REFERENCES

[1] G. Crini, Prog. Polym. Sci., 2005, 30, 38. [2] S. Babel, T.A. Kurniawan, J. Hazard. Mater., 2003, B97, 219. [3] A. Alagui, M. Vincedou, P. Vottero, Polymer, 2000, 41, 2463. [4] A. V. Il’ina, V. P. Varlamov, Appl. Biochem. Microbiol., 2004, 40, 300. [5] E. Guibal, Sep. Purif. Technol. 2004, 38, 43. [6] S.-T. Lee, F.-L. Mi, Y.-J. Shen, S.-S. Shyu, Polymer 2001, 42, 1879.


147

REMOVAL OF HEAVY METALS FROM STRONG ANIONIC

COMPLEXES

Helena PARSCHOVÁ, Eva MIŠTOVÁ and Luděk JELÍNEK

Department of Power Engineering, ICT Prague

Abstract This work is concerned with the selective removal of copper from waste or

process streams containing strong anionic complexes (NTA, EDTA). The conditions for efficient sorption of copper from NTA (EDTA) complex were determined on oligo(ethyleneamine) resins in the free base form.

The sorption of metals on standard carboxylic or iminodiacetate cation exchangers was not effective for removal of metals of NTA or EDTA complexes. The sorption was only effective when a suitable oligo(ethyleneamine) ligand was added into the treated solution containing anionic complex. The conditions and mechanism of this recomplexation process were studied in this work.

1. INTRODUCTION

The uptake of heavy metals on various cation exchange resins proceeds predominantly through electrostatic attractive forces. But on carboxylic cation exchanger and on chelating iminodiacetate cation exchange resin the coordination of heavy metal cations to N- and/or O- atoms of resin’s functional groups also contributes to the uptake of metal by ion exchanger. The uptake of heavy metals by oligo(ethyleneamine) resins (OEA) is based on the coordination bond of heavy metal cation to N-atom of resin’s functional group in the free base form [1].

The application of OEA resins in the free base form is efficient for sorption of copper (breakthrough capacity 0.72 eq.L-1).

The sorption of copper or nickel from EA ligands on sulfonic, carboxylic and iminodiacetate cation exchangers is effective. The break-through capacities were greater than 1.24 eq.L-1 [1]. Very low breakthrough capacities (0.02-0.11 eq.L-1) were determined for the sorption of copper or nickel from oligo(ethyleneamine) ligand on oligo(ethyleneamine) resins in the free base form.

The sorbent having OEA group in the free base form is able to remove effectively heavy metal-cations from neutral and alkaline solutions containing EDTA chelating agent. [2].


148

Iminodiacetate chelating cation exchanger [3] is used quite commonly for the removal of heavy metals from citrate-containing waste water but weakly acidic carboxylic cation exchanger is not efficient for uptake of metals from citrate solutions [4] or NTA complexes [1].

2. EXPERIMENTAL

The following resins were used in the study: a) special acrylamide resins having oligo(ethyleneamine) moieties in the free base form with functional groups:

TETA = triethylenatetramine (Lewatit E-2) TEPA = tetraethylenepentamine (Lewatit E 1, Purolite S 960, Ionac A-305) PEHA = pentaethlylenehexamine (Lewatit 6718 HLH)

b) iminodiacetate cation exchanger in the Na+ form (Lewatit TP-207) c) carboxylic cation exchanger in the Na+ form (Duolite C433)

All presented results were obtained by dynamic column experiments that were carried out using synthetic metal solution containing 1 mmol.L-1 of metal ions and 2 mmol.L-1 EA and 2 mmol.L-1 NTA (EDTA). pH value of loading solutions was 8. The specific flow rate (s) of loading solution was 6 BV.h-1. Column inner diameter was 12 mm, bed volume 30 mL and bed height 26.5 mm. The sorption run was terminated at metal breakthrough concentration 1mg.L-1. The metals were analyzed by AAS.

The regeneration of carboxylic cation exchanger and on chelating iminodiacetate cation exchange resin consisted of two steps. In the first step 3 BV of 2M HCl and in the second step 5 BV of 1M NaOH were applied at a flow rate of 3BV.h-1. The regeneration of resins having oligo(ethylene-amine) moieties consisted of three steps. In the first step 5 BV of 1M Na2SO4 in the second step 5 BV of 1M H2SO4 and in the third step 5 BV of 1M NaOH were applied at a flow rate of 3 BV.h-1.


The application of carboxylic or iminodiacetate cation exchangers was not efficient in the case of NTA and EDTA complexes [5] because of high stability (Tab.1.) of these anionic complexes in the treated solution.

Table 1. The logarithmic values of stability constants of Cu complex [6]

Ligand

Log K

Ligand

Log K

EDA DETA TETA TEPA

10.7 16.1 20.4 23.1

PEHA IDA NTA

EDTA

26.2 10.3 12.9 18.2


149

Experimental results showed that oligo(ethyleneamine) resins in the free base form were efficient for sorption of copper from NTA or EDTA complex (Tab.2.).

Table 2. The sorption of copper from NTA or EDTA complex

An addition of suitable oligo(ethyleneamine) ligand into the treated solution containing anionic complex made sorption onto these cation exchangers possible. Anionic metal complexes were gradually and continuously transformed into cationic complexes by addition of oligo(ethyleneamine) ligand until the heavy metals were completely removed from the solution by cation exchangers.

Heavy metals present as cationic ethyleneamine complexes were then removed from solution quantitatively and without problems by cation exchangers. Carboxylic cation exchanger (Tab.3.) removed copper from NTA solution when TEPA ligand was added into loading solution. Iminodiacetate cation exchanger took up copper quantitatively when EDA ligand was added into loading NTA solution (0.64 eq.L-1).

Table 3. The sorption of copper from NTA and EA complex on carboxylic exchanger

Loading solution Breakthrough capacity

[eq.L-1]

Cu + NTA + DETA Cu + NTA + TETA Cu + NTA + TEPA

0.03 0.07 0.14

The sorption of copper from EDTA complexes on iminodiacetate and carboxylic cation exchanger was quantitative after addition of TETA and TEPA ligand, respectively (Tab.4.).

Breakthrough capacity [eq.L-1] Ion exchanger

NTA EDTA

Lewatit E-1 (50 % TEPA) Lewatit E-2 (70 % TETA) Purolite S-960 (TEPA) Ionac A-305 (TEPA) Lewatit 6718 HLH (PEHA) Lewatit TP-207 (IDA) Duolite C433

1.02 0.72 0.20 0.78 1.14 0.08 0.00

0.49 0.50 0.11 0.20 - 0.00 0.00


150

Table 4. The sorption of copper from EDTA and EA complex on IDA exchanger

Loading solution Breakthrough capacity

[eq.L-1]

Cu + EDTA + TETA Cu + EDTA + TEPA

0.10 0.22

The effect of recomplexation process was studied also on the sorption of copper from solution NTA and EDA ligands on oligo(etylenamine) resins. This sorption was not effective because copper with NTA ligand forms stronger complex than copper with EDA ligand (Tab.5.).

Table 5. Sorption of copper from NTA and EDA on oligo(etylenamine) resins

Oligo(etylenamine) resin Breakthrough capacity

[eq.L-1]

Lewatit E-1 (50% TEPA) Lewatit E-2 (70% TETA)

0.03 0.02

4. CONCLUSIONS

The standard carboxylic or iminodiacetate cation exchangers can be efficiently used for removal of copper from NTA or EDTA complexes when stronger oligo(etylenamine) ligand is added into the loading solution.

Acknowledgement

The financial support from Research Plan MSM CZ 6046137304 is gratefully appreciated.

REFERENCES

[1] Parschová H.: Thesis, Praha, 1998 [2] Matějka Z., Zítková Z.: Reactive & Functional Polymers, 1997, 35, 81-88 [3] Kahovec J.: Chemické listy, , 1981, 75, 398-404 [4] Matějka Z., Weber R.: Reactive Polymers, 1990, 13, 299-308 [5] Přibyl R.: Komplexometrie, SNTL, Praha 1977 [6] Martel E.A., Calvin M.: Chemistry of Metal Chelates, ČSAV, Praha, 1959


151

PYRIDYL KETONE OXIMES AS EXTRACTANTS OF

ZINC(II) FROM CHLORIDE SOLUTIONS

Karolina KLONOWSKA – WIESZCZYCKA, Andrzej OLSZANOWSKI, Anna PARUS

Institute of Chemical Technology and Engineering, Poznan University of Technology, Pl. Skłodowskiej – Curie 2, 60 – 965 Poznan

1. INTRODUCTION

Solvent extraction and ion exchange can be used to remove the zinc(II) from aqueous solutions. These techniques are used in many processes such as treatment of lean grade ores, spent catalysts, scrap, complex sulphides and in recycling wastewater. The use of various extractants has been proposed: D2AHPA, Cyanex 272, Cyanex 301, Acorga ZNX 50, Kelex 100.1 Pyridyl ketone oximes could be the new extractants to recover metals from aqueous solutions.

Pyridyl ketone oximes have the ability to coordinate metals such as: copper2, zinc3, cobalt4, nickel5 and others and can form the complexes with metal ions by solvating and/or chelating mechanisms. The formation of the complexes of pyridineketooxime with metal ions depends on the nature of solvents, the location of the oximes group in the pyridine ring and the concentration of metal ions6. Well-known hydrophilic oximes with 2-pyridyl groups depending on pH, can bind ion of metal in different modes: as neutral (HL) and/or deprotonated (L) ligands. The formation of complexes depends on the coordination tendency of metal, structure of ligand and composition of aqueous solutions7,8

The aim of this work was the synthesis of hydrophobic model extractants of pyridyl ketone oximes and the extraction of zinc(II) from chloride solutions by synthesized extractants. The effect of the position oximes groups in the pyridine ring and the length the alkyl group on zinc(II) extraction from chloride media were studied.

2. EXPERIMENTAL

Reagents: Pyridyl ketone oximes were obtained by a reaction of hydroxyloamine with corresponding ketones, which were synthesized from pyridinecarbonitrile with proper Grignard reagents. The structures of pyridyl ketone oximes are shown below.


152

N(CH2)11CH3

NOH N

(CH2)9CH3N

OH

N

(CH2)11CH3N

OH

N

(CH2)13CH3N

OH

2PC12 4PC10 4PC12 4PC14

The extraction was carried out at constant water activity aw = 0.8352, the constant total concentration of ions and molecules dissolved in the aqueous solution σ = 8.0M. NaCl, NaNO3, LiNO3 and NaClO4 were used to adjust to constant water activity to aw = 0.8352 and the ionic strength I = 4 M. In all experiments the pH of the aqueous phases was 3.5-3.8. The studies were carried out at constant zinc concentration equal to 0.01 M. The concentration of extractant in the organic phase was 0.1 M. Chloroform and toluene with 10% addition of decanol were used as diluents. The extractions were carried out in a test tube using equal volumes (5 mL) of both phases. The test tube was shaken for 30 min. Metal concentration in the aqueous feed was determined by the titration with EDTA. The extraction percentage was calculated from: E = (1-caq,be/caq,af)*100%, caq,be, caq,af – concentration of metal in aqueous phase before, after the extraction, respectively. All other chemicals were of analytical purity grade.


The extraction of zinc(II) was observed for all synthesized pyridyl ketone oximes. The extraction of zinc from chloride solutions with pyridyl ketone oximes depend on chloride concentration and the type of the used diluent. The effect of chloride concentration on extraction of zinc from chloride solutions at the constant water activity aw = 0.835 and the total concentration of ion and molecules dissolved in aqueous solution σ = 8.0M is shown on Fig. 1. The results obtained indicate that the 2-pyridyl ketone oxime (2PC12) in chloroform extracted zinc(II) in all range of chloride solutions and did not depend on its concentration. Different results were obtained when toluene with 10% addition of decanol was used as the diluent. The degree of zinc(II) extraction increased with chloride ion concentration up to 1M Cl- and above 1M Cl- the degree of extraction decreased. The extraction of zinc was observed for 4-pyridyl ketone oximes in chloroform. The degree of extraction increased at low chloride concentrations (up to 1M Cl-) and decreased at high chloride concentrations (above 1M Cl-). The decrease of zinc extraction at high chloride concentrations is probably connected with the increase of the ZnCl3

- and ZnCl4

-2 ions concentrations which depressed transport of metal.


153

0

20

40

60

80

100

0 0.5 1 1.5 2 2.5 3 3.5 4

[Cl- ] mol dm-3

extr

actio

n [%

]

2PC12 Ch

2PC12T+D

4PC10 Ch

4PC12 Ch

4PC14Ch

Fig.1. The extraction of zinc(II) from chloride solutions with 2- and

4- pyridylketoximes (aw = 0.835 and extractant concentration – 0.1M) (Ch – chloroform; T + D – toluene + 10% v/v decanol)

The effect of chloride concentration on the zinc extraction from chloride solutions at the constant ionic strength I = 4 is presented at Fig. 2. Only the extraction of zinc using 1-(2-pyridine)tridecane-1-one oxime (2PC12) in chloroform did not depend on chloride concentrations. But the use of toluene + 10% v/v decanol, as solvent, for 1-(2-pyridine)tridecane-1-one oxime (2PC12) pointed out that the degree of zinc extraction depends on the concentration of chloride ions. However, the extraction of zinc(II) was observed for 4-pyridyl ketone oximes (4PC10, 4PC12 and 4PC14) only for solutions of oxime in chloroform. Similarly to the extraction of zinc(II) from chloride solution at the constant water activity, the extraction of zinc increased at low chloride concentrations and than decreased with the increase of chloride concentrations.

0

20

40

60

80

100

0 0.5 1 1.5 2 2.5 3 3.5 4[Cl-] mol dm-3

extr

actio

n % 4PC12

4PC10

4PC14

2PC12 Ch

2PC12 T+D

Fig.2. The extraction of zinc(II) from chloride solutions with 2- and

4- pyridylketoximes (I =4 and extractant concentration – 0.1M) (Ch – chloroform; T + D – toluene + 10% v/v decanol)

The results obtained indicate that 2-pyridyl ketone oxime (2PC12) extracts more zinc(II) from chloride solution than 4-pyridyl ketone oximes (4PC10, 4PC12, 4PC14). The highest degree of zinc(II) extraction was


154

observed for 1-(2-pyridine)tridecane-1-one oxime (2PC12) in chloroform (~90%) but the lowest value of extraction (~50%) was observed for 1-(4-pyridine)undecane-1-one oxime (4PC10). Among 4-pyridyl ketone oximes the highest extraction of zinc was observed for 1-(4-pyridine)tridecane-1-one oxime (4PC12).

Zinc(II) could be effectively stripped from organic solution by water in the two-stage process. The first stage allows to strip up to 60% of Zn(II) from organic phase for 1-(2-pirydine)tridecane-1-one (2PC12) and 40 – 50% of Zn(II) from solution of 4-pyridyl ketone oximes (4PC10, 4PC12, 4PC14). In the second stage Zn(II) was stripped in 40% for 1-(2-pirydine)tridecane-1-one (2PC12) and 50 – 60% for 4-pyridyl ketone oximes (4PC10, 4PC12, 4PC14) from organic phase. The oximes, after striping by water, have similar extraction activity as before the extraction.

4. CONCLUSION

The 1-(2-pyridine)tridecane-1-one (2PC12), 1-(4-pyridine)undecane-1-one (4PC10), 1-(4-pyridine)tridecane-1-one (4PC12) and 1-(4-pyridine)-pentadecan-1-one (4PC14) oximes extracted zinc(II) from chloride solutions in all range chloride concentration but the process depended on ionic strength, chloride ion concentration and type of used diluents.

The extraction of zinc(II) was observed for 2-pyridyl ketone oxime in both diluents, but the extraction of zinc for for 4-pyridyl ketone oximes were observed for chloroform solutions. The extraction of Zn(II) for 2-pyridyl ketone oxime in chloroform did not depend on the chloride concentration. However, when used toluene with 10% addition of n-decanol as diluent the extraction of zinc(II) increased up to 1 M Cl-

, and thereafter decreased. Zinc(II) could be effectively effectively stripped from organic phase by water.

Acknowledgements

The work was supported by the Polish State Committee for Scientific Research grant BW 32/003/08.

REFERENCES

[1] G. Cote, A. Jakubiak, Hydrometallurgy, 1996, 43, 277. [2] P. Chaudhuri, Coord. Chem. Rev., 2003, 243, 143. [3] M. Salonen, H. Saarinen, M. Orama, J Coord Chem, 2003, 56(12), 1041. [4] T.C. Stamatatos, A. Bell, P. Cooper, A. Terzis, C.P. Raptopoulou, S.L. Heath, R.E.P.

Winpenny, S.P. Perlepes, Inorg. Chem. Commun., 2005, 8, 533. [5] K. Riggle, T. Lynde-Kernell, E.O. Schlemper, J. Coord. Chem., 1992, 25, 117. [6] J. C. Milios, T. C. Stamatatos, S. P. Perlepes, Polyhedron 2006, 134-194. [7] M. Salonen, H. Saarinen, M. Orama, J Coord Chem, 2003, 56(12), 1041. [8] C.B. Aakeroy, A.M. Beatty, D.S. Leinen, Angew. Chem., Int. Ed., 1999, 38, 1815.


155

THE COMPARISON OF HYDROPHOBIC PYRIDINE-2-

CARBOXAMIDES AND ALKYL 2-PYRIDYL KETOXIMES AS EXTRACTANTS OF COPPER(II), ZINC(II) AND CADMIUM(II) FROM CHLORIDE SOLUTIONS

Aleksandra BOROWIAK-RESTERNA, Karolina KLONOWSKA-WIESZCZYCKA, Andrzej OLSZANOWSKI, Anna PARUS and

Marta TOMASZEWSKA

Institute of Chemical Technology and Engineering, Poznan University of Technology, Pl. Sklodowskiej-Curie 2, 60-965 Poznan, Poland

1. INTRODUCTION

Solvent extraction is a very well known technique for the recovery of metal ions from aqueous solutions. Industrial application of the process continues to grow with improvements and developments of new organic extractants. The extraction of such metals ions as Cu(II), Zn(II), Cd(II), Pd(II) etc. from chloride solutions is particularly interesting. During the last decades amines, quaternary ammonium salts, organo-phosphorous compounds, benzimidazole derivatives and crown ethers have been used as extractants for the separation of cadmium, copper and zinc ions.

The pyridine derivatives, which form complexes with Cd, Cu, Zn, are potential extractants of these metals from acid chloride solutions. Acorga CLX-50 has been proposed by ZENECA (early ICI) for the extraction of copper(II) from concentrated chloride solutions, obtained by leaching of sulphide ores with aqueous solutions of iron or cupric chloride [1,2]. The active component of the extractant is diisodecyl pyridine-3,5-dicarboxylate. Other derivatives of pyridinecarboxylic acids (esters, amides, and ketoximes) for the extraction of Cu(II) from chloride solutions are proposed in [3-9]. Hydrophobic pyridine-3 (and -4)-carboxamides can effective extract Cd(II) [10] and Pd(II) [11] from acidic chloride solutions.

The aim of this work is a comparison of the pyridine-2-carboxamides and alkyl 2-pyridyl ketoximes as extractants of copper(II), zinc(II) and cadmium(II) from chloride solutions.

2. EXPERIMENTAL

Pure individual hydrophobic pyridine derivatives, three pyridine-2-carboxamides and two 2-pyridyl alkyl ketoximes, were used as extractants of copper(II), cadmium(II), and zinc(II) from chloride solutions. Synthesis,


156

purity and analytical data for the studied extractants were described in previous papers [7, 9, 10]. The structures of these compounds are as follows:

N C

N

R

OH

N C

O

NR1R2

DA - R1 = H, R2 = C12H25

EHA - R1 = H, R2 = C4H9CH(C2H5)CH2

DHA - R1 = R2 = C6H13

10KO - R = C10H21

12KO - R = C12H25 The extraction was carried out at constant water activity aw = 0.835

constant total concentration of ions and molecules σ = 8.0M, and constant copper concentration equal to 0.01 M. In all experiments the pH of the aqueous phase was close to 3.5-3.8. The concentration of extractants in the organic phase was changed from 0.01M to 0.2 M. Toluene and toluene with 10% addition of decan-1-ol were used as diluents. The extraction was carried out in a test tube using equal volumes (5 mL) of both phases. The test tube was shaken for 30 min. Metal concentration in the aqueous feed was determined by the titration with EDTA.

3. RESULTS

The synthesized pyridine derivatives are the weak organic bases. Acid transfer to the organic phase by these compounds depends on the concentration of hydrogen ions and the structure of substituent in the pyridine ring (Table 1).

Tab.1. Acid transfer to the toluene phase by extractants ([amide] = 0.2 M, [oxime] = 0.1 M, [HCl] + [NaCl] = 4 M)

HCl concentration (M) Organic phase

Aqueous phase DA [7] EHA [10] DHA [10] 10KO [9] 12KO [9]

1.0 0.0009 0.0001 0.006a 1.5 0.0007 0.0014 - 2.0 0.0014 0.007a 3.0 0.0016a 0.009a 4.0

EMULSION 0.0012

0.0019a

Dissolving of protonated ketoxime in

aqueous phase 0.01a

a – the protonated extractant precipitates

The obtained results pointed out that compounds with oxime or N-dodecylamide group at 2 position in the pyridine ring were easily protonated by HCl. The protonated 1-(2-pyridyl)-undecan-1-one oxime


157

(10KO) was dissolved in the aqueous phase. Hydrochlorides of 1-(2-pyridyl)tridecan-1-one oxime (12KO) and N-dodecylpyridine-2-carboxamide (DA), due to a low solubility in toluene, precipitated forming a third phase. Acid transfer to the organic phase by N-(2-ethylhexyl)- and N,N-dihexylpyridine-2-carboxamides (EHA and DHA, respectively) is low.

The extraction studies were carried out at constant water activity (aw = 0.835) and constant total concentration of ions and molecules dissolved in the aqueous solution (σ = 8.0 M) which made possible to maintain constant activity coefficients in the aqueous phase [5]. Our studies showed that the extraction of copper(II) from chloride solutions with alkyl 2-pyridyl ketoxime did not depend on the chloride concentration (Table 2); at the whole range of chloride concentration copper(II) was extracted to 100%. The measurement of the pH value pointed out, for both used ligands, the increase of the acidity of aqueous phases after extraction. The changes of pH were probably combined with deprotonation of oxime moiety allowing on chelating complexes formation. Similar high copper(II) extraction was observed when as extractant N-dodecylpyridine-2-carboxamide was used (Table 2). The amide probably forms with copper(II) oligomeric complexes such as (CuCl2)xL2 [7]. The amide with one branched carbon chain at the amide nitrogen (EHA) is weaker extractant but the compound with two long carbon chain in the amide group (DHA) is a very weak extractant for the recovery of copper. For DHA the extraction percent maximum (at 1 M Cl- and at [amide]/[Cu(II)] = 20) is equal 11%. Probably it is a result of the strong steric effect of the amide group.

Tab.2. Influence of Cl- concentration on Cu(II) extraction from weak-acid chloride solutions ([DHA] = [EHA] = 0.2 M, [10KO] = [12KO] = 0.1 M, [DA] = 0.02 M,

diluent: toluene or toluene with 10% addition of decan-1-ol (12KO))

Percentage extraction [Cl-] (M) DA [7] EHA DHA [12] 10KO [9] 12KO [9] 0.1 95.3 emulsion 5.9 0.5 96.4 98.0 6.8 1.0 97.5 97.1 10.9 2.0 97.2 94.5 9.6 3.0 97.0 93.2 9.3 4.0 98.5 79.9 3.7

100

100

Alkyl 2-pyridyl ketoximes are effective extractants for recovery of

cadmium(II) from chloride solutions at pH > 3. The extraction maximum is observed at 1 M Cl-. At chloride concentration above 1 M the complex-forming power of ketoximes decreases due to an increase in CdCl3

- and CdCl4

2- concentration in the aqueous phase. Pyridine-2-carboxamides are not effective extractants of Cd(II). EHA and DHA do not extract Cd(II) at


158

the whole studied range of chloride concentration. DA forms complexes with Cd(II) but these complexes precipitate in the extraction system.

Zinc(II) extraction ability rises for studied compounds in the order: DHA < EHA < 10KO < 12KO. The extraction maximum is observed at 0.5-1 M Cl- or at 1.5-2 M for DHA and others extractants, respectively. The slopes of the straight lines, showing the dependence log D = f (log[amide]), suggest that all studied extractants form complexes with Zn(II) in which the molar ratio of metal to ligand is equal to 1 : 2 (D – the metal distribution ratio).

4. CONCLUSIONS

Alkyl 2-pyridyl ketoximes extract more zinc(II) and cadmium(II) than hydrophobic pyridine-2-carboxamides at the same extraction conditions. Copper(II) is effectively eliminated from chloride weak-acid solutions with both alkyl 2-pyridyl ketoxime and N-alkylpyridine-2-carboxamide with strength carbon chain.

Acknowledgements

The work described in this paper was supported by 32-003/08-BW.

REFERENCES

[1] R.F. Dalton, R. Price, E. Hermana, B. Hoffman, Min. Eng., 1988, 40, 24. [2] R.F. Dalton, G. Diaz, R. Price, A.D. Zunkel, J. O. M., 1991, 43, 51. [3] R.F. Dalton, R. Price, P.M. Quan, D. Stewart, Eur. Patent 57, 797, 1982. [4] J. Szymanowski, A. Jakubiak, G. Cote, D. Bauer, J. Beger, in: Logsdail, D.H., Slater,

M.J. (Eds.), Solvents in the Process Industries, Proc. ISEC’93. SCI-Elsevier Applied Science, London, 1993, p. 1311.

[5] G. Cote, A. Jakubiak, D. Bauer, J. Szymanowski, B. Mokili, C. Poitrenaud, Solvent Extr. Ion Exch., 1994, 12, 99.

[6] A. Borowiak-Resterna, Solvent Extr. Ion Exch., 1994, 12, 557. [7] A. Borowiak-Resterna, Solvent Extr. Ion Exch., 1999, 17, 133. [8] A. Borowiak-Resterna, G. Cote, J. Szymanowski, in: Cox, M., Hidalgo, M., Valiente,

M. (Eds.), Solvent Extraction for the 21st Century, Proc. ISEC’99. Soc. Chem. Ind., London, 2001, p. 1197.

[9] K. Klonowska-Wieszczycka, A. Olszanowski, A. Parus, B. Zydorczak, Solvent Extr. Ion Exch., in press.

[10] M.Tomaszewska, A.Borowiak-Resterna, A.Olszanowski, Hydrometallurgy, 2007,85,116. [11] I. Szczepańska, A. Borowiak-Resterna, M. Wiśniewski, Hydrometallurgy, 2003,68,159. [12] A. Borowiak-Resterna, Przem. Chem., 2006, 85, 569.


159

SOLVENT EXTRACTION OF Cu(II) COMPLEXES WITH

1-ALKYL-4(5)-METHYLIMIDAZOLE

ElŜbieta RADZYMIŃSKA-LENARCIK1, Beniamin LENARCIK2 1Department of Inorganic Chemistry, University of Technology and Life Sciences

Bydgoszcz, Seminaryjna 3, 85-326 Bydgoszcz, Poland

2Higher School of Environment, Fordońska 120, 85 – 739 Bydgoszcz, Poland

1. INTRODUCTION

2-Alkylimidazoles have been found to exhibit steric effects in complexation reactions with metal ions [1,2]. The steric hindrance differentiates properties of the complexes such as their stability, structure of the co-ordination sphere, nature of the metal-ligand bond and even their interaction with solvent molecules [1-7]. This offers the possibility of separation of Co(II) and Zn(II) from other cations by using the solvent extraction method [4, 6, 7]. Recently, it has been demonstrated that 1-alkyl-2-methylimidazoles could be used for selective extraction of Cu(II) [8]. A similar influence on the complexation reaction of the transition metal ions has been reported for the methyl group in isomeric bases at position 4 of the imidazole ring in 4(5)-methylimidazoles [9]. The steric effect in the complexation reactions with metal ions should be exhibited by 4-methyl-substituted isomers only, as the 4-methyl substituent stands at α-position to the pyridinic (electron-donating) nitrogen atom.

The primary purpose of this study was to check the influence of the steric hindrance and the alkyl chain length on the stability and extraction of the Cu(II) complexes using the partition method. Seven 1-alkyl-4(5)-methylimidazoles have been chosen for the experiments with n-butyl through n-decyl substituents at the position 1. Toluene, p-xylene, dichloromethane, chloroform, and 2-ethylhexanol were used as extractants.

2. EXPERIMENTAL

All studies were carried out at 298K at constant ionic strength of the aqueous phase equal to 0.5 kept by KNO3 and HNO3. Before extraction the concentrations of Cu(II) and the nitric(V) acid in the aqueous phase were constant (0,01 M and 0,1 M respectively). The extraction measurements were carried out after adding an equal volume of organic solution of variable concentrations of 1-alkyl-4(5)-methylimidazoles to the water phase. The test tube were than shaken for 30 min. After equilibrium was


160

achieved both phases were separated and pH of the aqueous solution was measured, as well as the Cu(II) concentration (by the atomic absorption spectrophotometry). The VIS absorption spectra of Cu(II) complexes in both phases were registered.


The partition of copper(II) in the systems studied was calculated on the basis of the Cu(II) concentrations in the aqueous phase before and after attaining partition equilibrium from the following equation:

M

M0M

)aq)(II(Cu

)org)(II(CuM C

CC

C

CD

−== (1)

where: 0MC and MC denote analytical Cu(II) concentrations in the aqueous

phase before and after attaining partition equilibrium, respectively. The magnitude of variable DM depends on the 1-alkyl-4(5)-

methylimidazole concentration in both phases, and consequently on the pH. The progress of Cu(II) extraction in the systems studied is controlled

by complexes being formed in both phases as well as by their stability and extraction properties. Partition ratio, DM, is related to stability constants of the complexes and their partition constants by the following equation:

∑=

=

+++ +++= Nn

n

nn

NNN

ccc

ccc

M

L

LPLPLPD

0

111

][

][.....][][

β

βββ (2)

where: βn and βc are cumulative stability constants of the complexes in the aqueous phase, Pc are organic solvent/water partition constants of the complexes, (Pc=[ML c](org)/[ML c](aq)), [L] is the free ligand concentration (mol/L) in the aqueous phase and c is the number of ligands molecules in the first Cu(II) complex that is so hydrophobic that it freely passes into the organic phase [10,11].

4. CONCLUSIONS

1. The methyl substituent in 1-alkyl-4(5)-methylimidazoles has been found to be responsible for the steric effect in their reactions with the Cu(II) ion. Furthermore, the effect results in depressing the stability of the complexes formed. Stability constants, βn, of the complexes in the aqueous phase match those of the 1-n-alkyl-2-methylimidazole complexes of Cu(II).


161

2. Owing to the steric effect, elongation of the n-alkyl chain together with pronounced electron-donor properties of solvents result in species of presumable co-ordination number 4 having the shape of a distorted tetrahedron.

3. The phenomenon has been noticed at the second and third complexation steps (ML2 and ML3 species). The 4-co-ordinate Cu(II) species being formed are more readily extractable as demonstrated by large magnitudes of partition constants P2 and P3 increasing rapidly with elongation of the 1-n-alkyl substituent.

4. All those phenomena are of practical importance owing to depressing the magnitude of pH1/2 of the extraction thus causing it to proceed at lower extractant (ligand) concentrations in both the organic and aqueous phase, this, in turn, being favourable for the separation of the Cu(II) complexes from those of other metals.

REFERENCES

[1] B. Lenarcik, J. Kulig, P. Laidler, Roczniki Chemii, 1974, 48:1151. [2] B. Lenarcik, B. Barszcz, J. Chem. Soc. Dalton Trans., 1980, 24. [3] B. Lenarcik, K. Kurdziel, Polish J. Chem., 1981, 55:737. [4] B. Lenarcik, J. Glowacki, M. Rzepka, Solv. Ext. Ion Exch., 1979, 14: 37. [5] J.G.H. du Preez, Ch. Mattheüs, N. Sumter, S. Ravindran, C. Potgieter, B.J. van Brecht,

Solv. Ext. Ion Exch., 1998, 16 (2): 565. [6] B. Lenarcik, A. Adach, E. Radzyminska-Lenarcik, Polish J. Chem., 1999, 73: 1273. [7] B. Lenarcik, A. Kierzkowska, Sol. Extr. Ion Exch., 2006, 24, 433-445. [8] E. Radzyminska-Lenarcik Sep. Sci. Technol., 2007, 42, 2661. [9] B. Lenarcik, K. Oblak, Roczniki Chemii, 1977, 51:2079. [10] J. Rydberg, Acta Chem. Scand., 1950, 4: 1503. [11] F.J.C. Rossotti, H. Rossotti, The Determination of Stability Constants. McGraw-Hill:

New York. 1961.


162

SOLVENT EXTRACTION OF CU(II) COMPLEXES WITH

1-ALKYL-2-ETHYLIMIDAZOLE

ElŜbieta RADZYMIŃSKA-LENARCIK1 and Beniamin LENARCIK2 1Department of Inorganic Chemistry, University of Technology and Life Sciences

Bydgoszcz, Seminaryjna 3, 85-326 Bydgoszcz, Poland 2Higher School of Environmental, Fordońska 120, 85 – 739 Bydgoszcz, Poland

Formation of Cu(II) complexes of 1-alkyl-2-ethylmidazoles (where alkyl = propyl, butyl, pentyl, hexyl, and octyl) has been studied by using the liquid-liquid partition method, at 25oC and a fixed ionic strength of the aqueous phase (I = 0.5; (HL)NO3, KNO3). The complexes were extracted with 2-pentanone, 2-butanol, isoamyl alcohol, 2-ethyl-1-hexanol, dichloromethane, trichloromethane, and toluene. The length of the 1-alkyl group and the nature of solvent have been shown to influence the extraction process. Extraction curves (log DM vs. pH) are displaced towards lower pH’s with increasing chain length of the 1-alkyl substituent and donor number of the solvents. Stability constants of the complexes in aqueous solution were determined as well as their partition ratios between the aqueous and organic phase. The stability of the Cu(II) complexes increased with increasing 1-alkyl chain length. The stability constants are comparable with βn ones for the Cu(II) complexes of 1-alkyl-2-methylimidazoles, but smaller than those of the Cu(II) – 1-alkylimidazole counterparts. The partition ratios, Pc, increased with increasing 1-alkyl chain length and the donor number of the solvents. Four- and six- coordinate Cu(II) complexes with 1-alkyl-2-ethylimidazoles passed to the organic phase. They were identified thus revealing a structural change of the co-ordination polyhedron into a pseudo-tetrahedral one. This process turned out to be favorable for the extraction as indicated by the highest values of the P3 ratio. This finding offers the possibility of extraction of the Cu(II) ions from a mixture cations.


163

THE INFLUENCE OF ORGANIC SOLVENTS ON

EXTRACTION OF Cd(II) COMPLEXES WITH 1,2-DIMETHYLIMIDAZOLE

ElŜbieta RADZYMIŃSKA-LENARCIK1, Beniamin LENARCIK2

1Department of Inorganic Chemistry, University of Technology and Life Sciences Bydgoszcz,

Seminaryjna 3, 85-326 Bydgoszcz, Poland 2Higher School of Environment, Fordońska 120, 85 – 739 Bydgoszcz, Poland

1. INTRODUCTION

Previously it was demonstrated that the steric effect, due to the methyl or ethyl group in position “2” of the imidazole ring, favoured the formation of tetrahedral Co(II), Cu(II), Zn(II) and Cd(II) complexes in aqueous solutions [1,2,3]. This effect facilitated the transfer of the cations to the organic phase as compared to Ni(II) and other metals [4,5].

In this work we have investigated Cd(II) complexes with 1,2-dimethylimidazole and their extraction process. Alkyl group at the nitrogen atom does not affect the donor properties of 1,3-diazoles, but only suppress their solubility in water, thus increasing the hydrophobicity of the complexes [6,7].

The purpose of this work was to investigate the influence the organic solvents on the extraction of complexes of Cd(II) with 1,2-dimethylimidazole. 2-Ethyl-1-hexanol, 2-butanol, isoamyl alcohol, 2-pentanone, 2-hexanone and dipropyl carbonate were used as organic solvent.

2. EXPERIMENTAL

The measurements were run at 298K at constant ionic strength of the aqueous phase (0.5) kept by KNO3 and HNO3. Before extraction the concentrations of Cd(II) and the nitric(V) acid in the aqueous phase were constant (0,02 M and 0,1 M respectively). The extraction measurements were carried out after adding an equal volume of organic solution of variable concentrations of 1,2-dimethylimidazole to the water phase. The test tubes were then shaken for 30 min. After equilibrium was achieved both phases were separated and pH of the aqueous solution was measured, as well as the Cd(II) concentration (by the atomic absorption spectrophotometry or titration with standardized EDTA solution).


164


The extraction process was characterized by inspecting the dependence between distribution coefficient of metal ion DM and pH. The free ligand molar concentration in water phase [L] was calculated from the pH values using:

a

3

K [HL ][L]

[H O ]

+

+= (1)

where aK is the acidity constant of the protonated ligand, and [HL ]+ is the

concentration of the conjugate acid equal to analytical concentration of nitric(V) acid in aqueous solution. Necessary pKa values were taken from literature [8].

The distribution ratio of metal ions between organic and aqueous phases (DM) was found from the measured concentration:

0Cd(II)(org) M M

MCd(II)(aq) M

C C CD

C C

−= = (2)

where 0MC and MC denote analytical concentration of the metal ion in the

aqueous phase before and after reaching partition equilibrium, respectively. Generally, the complexes were formed during reaction of the organic

solutions of the ligand with metal nitrates. The extraction results for all systems are presented as dependence of distribution coefficient of metal ion log DM versus pH. The extraction curves of Cd(II) complexes with 1,2-dimethylimidazole in all organic solvents are presented in Fig.1.

The extraction efficiency (DM) of those complexes decreased in the following rank order of solvents: 2-ethylhexanol > 2-butanol > 2-hexanone > 2-pentanone > isoamyl alcohol > dipropyl carbonate. The curves for the ketones are situated at low pH range. The difference in pH1/2 between 2-hexanone and dipropyl carbonate is quite large, amounting to 1.8 pH units.

To demonstrate the influence of the solvents on the extraction process of Cd(II) a dependence between pH1/2 and the Gutmann’s donor number [9,10], DN, of the solvents used was investigated. As seen, the pH1/2 value for each extraction system decreases with increasing donor number of solvent. To find the reason for this behavior, it is necessary to get parameters determining the extraction process in the systems studied.


165

-0,8

-0,6

-0,4

-0,2

0

0,2

0,4

0,6

0,8

1

1,2

4 4,5 5 5,5 6 6,5 7 7,5

pH

log DM

dipropyl carbonate 2-ethylhexanol 2-pentanone 2-hexanone 2-butanol isoamyl alcohol

Fig.1. Influence of organic solvents on extraction process of Cd(II) complexes with 1,2-dimethylimidazole

REFERENCES [1] B. Lenarcik, K. Kurdziel, Pol. J. Chem., 1981, 55, 737. [2] B. Lenarcik, K. Kurdziel, Pol. J .Chem., 1982, 56, 3. [3] B. Lenarcik, A. Adach, E. Radzymińska-Lenarcik, Pol. J. Chem., 1999, 73, 1273. [4] B. Lenarcik, J. Głowacki, M. Rzepka, Sep. Sci. Technol., 1979, 14, 37. [5] B. Lenarcik, K. Kurdziel, R. Czopek, Pol. J. Chem., 1991, 65, 837. [6] B. Lenarcik, B. Barszcz, Pol. J. Chem., 1979, 53, 963. [7] J. Kulig, B. Barszcz, B. Lenarcik, Pol. J. Chem., 1992, 66, 79. [8] B. Lenarcik, P. Ojczenasz , J. Heterocycl. Chem., 2002, 39, 287. [9] Y. Marcus, Chem. Soc. Rev., 1993, 409. [10] J. Rydberg, C. Musakis, G.R. Chopin, Principles and Practices of Solvent Extraction.

M. Dekker, Inc.: New York, 1992 vol.1: 22-24.


166

RATE OF Cd(II) EXTRACTION BY HYDROPHOBIC

PYRIDINECARBOXAMIDES FROM CHLORIDE AQUEOUS SOLUTIONS

Aleksandra BOROWIAK-RESTERNA1) , Michalina SIKORA1), Ryszard

CIERPISZEWSKI2), and Krystyna PROCHASKA1) 1) Institute of Chemical Technology and Engineering, Poznan University of

Technology, Pl. Sklodowskiej-Curie 2, 60-965 Poznan, Poland 2) Poznań University of Economics, Faculty of Commodity Science,

al. Niepodległości 10, 60-967 Poznań, Poland

Abstract The equilibrium and the initial rate of extraction of Cd(II) from chloride

solutions by N,N-dialkylpyridine-3-carboxamide and N,N-dialkylpyridine-2-carboxamide were examined. For kinetics studies a Lewis cell method was used. It was found that N,N-dihexylpyridine-3-carboxamide is stronger extractant than N,N-dihexylpyridi-ne-2-carboxamide in studied systems. The obtained results indicated that the rate of extraction of Cd(II) is controlled by diffusion and depends on the concentration of amide in organic phase as well as choride ions in aqueous phase.

1. INTRODUCTION

Cadmium is a toxic metal which naturally exists as a contaminant of other metals ores and as a component of various materials and goods, e.g. alloys, dental materials, nickel–cadmium batteries. There are many papers reported the method of removing this metal from chloride, sulfate or phosphate solutions. Amines, quaternary ammonium salts, organo-phosphorous compounds and crown ethers. N-alkyl- and N,N-dialkylpyridinecarboxamides with the amide group at the 2nd, 3rd or 4th position were also used as cadmium(II) ions extractants . It was found that such compounds (diluted in toluene) are effective extractants of cadmium(II) from acidic chloride solutions [1].

The aim of this work was to study the rate of cadmium(II) extraction from aqueous chloride solutions by two hydrophobic derivatives of pyridinecarboxyamide, i.e. N,N-dialkylpyridine-3-carbox-amide and N,N-dialkylpyridine-2-carboxamide. Moreover, the influence of composition of an aqueous phase as well as the type of organic diluent on the rate of Cd(II) extraction was studied.


167

2. EXPERIMENTAL

N,N-dihexylpyridine-3-carboxamide (DH3) and N,N-dihexylpyridine-2-carboxamide (DH2) as model extractants were used. The methods of synthesis of these compounds and their characteristics were described in our previous paper [1, 2].

DH3

6 13 2

N

N(C H )

O

DH2 N

N(C H )

O

2136

The equilibrium of extraction was carried out in separatory funnels using the same volume (10 ml) of the organic (toluene or toluene and di-2-ethylhexan-1-ol in ratio 4:1 v/v) and aqueous phases. The phases were mechanically shaken for 1 hour and then allowed to phase separate (usually more than 1 hour). Demineralized water from reverse osmosis (10 cm3) was used for stripping of Cd(II) from the organic phase.

The kinetics experiments of Cd(II) extraction were investigated by a Lewis-type cell (Fig.1.). The apparatus provided a constant interfacial area of 16.6 cm2 and the same volume of the aqueous and organic phases (95 ml). The rotation number of the stirrers were kept in the range of 30÷120 rpm. to maintain a flat and stable interface. Cd(II) concentration in the aqueous phase was determined by atomic absorption spectroscopy using a Varian SPECTR AA800 [3].

Fig.1. Schematic view of the standard

stirred cell; 1 - organic phase, 2 - aqueous phase, 3 - organic/aqueous interface, 4 - two teflon tubes, 5 - water jacket, 6 - hole for taking samples.

3. RESULTS

The investigated reagents extract cadmium(II) in different way. The extraction ability of DH3 is much higher than in the case of DH2. The same effect was observed for both types of organic diluents considered.


168

The extraction behaviors of amides significantly depends on their concentration in the organic phase as well as on chloride and acid concentration in the aqueous phase. At 4 M chloride concentration and at [amide]/[CdCl2] = 20, DH3 extracts above 90% of Cd(II) even at low hydrochloric acid concentration (0.15÷0.5 M). At the same conditions DH2 can extract only small amounts of this metal. For DH2 Cd(II) extraction of about 30% was obtained in system with more than 1.7 M HCl in the aqueous phase.

The extraction of Cd (II) with DH3 can be described by Eq.(1) [1]:

oo CdClLHCdClHL )()(22 242

24

−+−+ =++ (1)

whereas in the case of DH2 the extraction process probably can be described by Eq.(2):

oo LHClCdClLHClCdClHL ))(()(33 242

24

−+−−+ =+++ (2)

When the chloride concentration in the aqueous phase is reduced to 3 M and at [amide]/[CdCl2] = 5, the extraction of Cd(II) decreases. At HCl concentration lower than 2 M DH3 extractant transfers into the toluene phase less than 50% of Cd(II), whereas DH2 extracts very small amounts of cadmium(II) independently on the HCl concentration in the aqueous phase.

The addition of 20% (v/v) of 2-ethylhexan-1-ol (as a modifier of extraction process) to toluene slightly decreases the extraction ability of the studied amides. The differences between results obtained for systems with and without modifier are higher for DH3.

In the kinetics studies the rate of extraction of Cd(II) by N,N-dihexyl-pyridine-3-carboxamide was investigated. It was found that the increase of flux of metal ion extracted to the organic phase is affected by the concentration of DH3 in the extraction system as well as by the concentration of chloride ions. The amount of hydrochloric acid present in the extraction system is very important parameter, too. Moreover, the effect of the rotation speed of the stirrer on the rate of Cd(II) transfer from aqueous phase to the organic phase was observed. It was found that the concentration of Cd(II) ions in the organic phase increases almost linearly with the time in the period up to about 30 minutes (Fig.2). From the slope of linear relations presented in Fig.2 the fluxes of Cd(II) transfer from the aqueous phase to the organic can be determined. Moreover, the initial rates of extraction were estimated. Obtained results showed that the initial rate of extraction was approximately proportional to the stirring speed up to 120 rpm.


169

0

100

200

300

400

500

600

700

800

900

0 10 20 30 40 50 60 70

Time, min.

Cd

(II)

co

nce

ntra

tion

in o

rga

nic

ph

ase

, g/

dm3

Fig.2. Effect of stirring speed on the initial rate of extraction of Cd(II) by DH3;

(●)30 rpm, (■)60 rpm, (▲)90 rpm, (♦)120 rpm.

4. CONCLUSIONS

N,N-dihexylpyridine-3-carboxamide is more effective extractant of Cd(II) from acid chloride solution than N,N-dihexylpyridine-2-carboxamide. The obtained results indicate that the extraction ability of N,N-dihexylpyridine-carboxamides decreases when the distance between the amide group and the pyridine nitrogen decreases. The obtained results are in agreement with the amides basicity and with their ability to proton transfer to the organic phase [1]. The study of kinetics of Cd(II) extraction indicated that the reaction is controlled by diffusion. It was found that the rate of cadmium(II) extraction is affected mainly by the concentration of Cl¯ in the aqueous phase.

Acknowledgements

The work was supported by 32-270/08-BW and 32-003/08-BW.

REFERENCES [1] M. Tomaszewska, A. Borowiak-Resterna, A. Olszanowski, Hydrometallurgy, 2007, 85,

116. [2] A. Borowiak-Resterna, Solvent Extr. Ion Exch. 1994, 12, 557. [3] Varian Australia Pty Ltd Analytical Methods – Flame AAS, 1989. [4] M. Tomaszewska, A. Jeschke, A. Borowiak-Resterna, R. Cierpiszewski, K. Prochaska,

Przem. Chem. 2006, 85(6/7), 668.


172

PREPARATION OF MICROCAPSULES CONTAINING TRI-n-OCTYLAMINE FOR EXTRACTION OF Pd(II)

BY SPG MEMBRANE EMULSIFICATION AND in situ POLYMERIZATION METHODS

K. SHIOMORI1), K. MINAMIHATA 2), S. KIYOYAMA2),

M. YOSHIDA3), Y. HATATE3) 1) Dept. Applied Chem., Univ. Miyazaki, Miyazaki, 889-2192, Japan

2) Dept. Chem. Sci. Eng., Miyakonojo NCT, Miyazaki, 885-8567, Japan 3) Dept. Applied Chem. Chem. Eng., Kagoshima Univ., Kagoshima 890-0065,

Japan

Abstract Microcapsules containing tri-n-octylamine as an extractant with the diameter

less than 25µm were prepared by using SPG membrane emulsification and in situ polymerization methods. The average diameter of the microcapsules was about half of the pore diameter of SPG membrane. The encapsulation efficiency of tri-n-octylamine was approximately 100%. All microcapsules prepared reached equilibrium in the forward extraction of Pd(II) within 5 minutes and the forward extraction ratio reached nearly 1. The back extraction of Pd(II) from the microcapsules was carried out using 0.1M HCl solution containing thiourea. The back extraction occurred promptly and the back extraction ratio was high enough to elute all Pd(II) from the microcapsules. Furthermore, no leakage of tri-n-octylamine was observed during repeating experiment of extraction and back extraction.

1. INTRODUCTION

Microencapsulation of extractants, which are used for liquid-liquid extractions of metal ions, organic acids, amino acids and various valuable compounds, is one of the most effective method to overcome some disadvantages in the extraction process, such as difficulty in phase separation by formation of stable emulsion or third phase between aqueous and organic phases, and use of large amount of organic solvent [1-3]. We have investigated on microcapsules containing tri-n-octylamine, hereafter TOA, for the extraction of Pd(II) from hydrochloric acid solution [4,5] and have shown that the microcapsules are effective extraction media for Pd(II). However, the microcapsules of larger diameter than 300µm and/or high content of TOA show very slow extraction rate. To solve this slow rate problem, we attempted to reduce the diameter of the microcapsules. In this


173

study, preparation of the microcapsules containing TOA with a diameter less than 25µm by using SPG membrane emulsification and in situ polymerization methods and their extraction properties of Pd(II) was investigated.

2. EXPERIMENTAL

2.1. PREPALATION OF MICROCAPSULES

An organic phase was composed of TOA as an extractant, DVB as a wall material of the microcapsules, toluene as a diluents, and ADVN as an initiator. Gum arabic solution at 2wt% was used as an outer aqueous phase. The organic phase was poured into the outer aqueous phase and emulsification was carried out using SPG membrane emulsification method. After emulsification, in situ polymerization was carried out at 343 K for 5 hours under N2 atmosphere with stirring at 400rpm. Then the microcapsules were obtained by filtration, washing with distilled water and finally drying under vacuum.

2.2. EXTRACTION PROPERTIES OF Pd(II) BY MICROCAPSULES

The microcapsules were pretreated to form TOA-HCl complex with 2M HCl for 24 hours at 298K. The pretreated microcapsules were dispersed into 0.1M HCl solution containing PdCl2 and then shaken at 150 rpm at 298K. The solution was collected by filtration at prescribed interval and the concentration of residual Pd(II) in the solution was measured with ICP spectrometer. The forward extraction ratio, Ef and the initial forward extraction rate, Rf were defined as follows;

( )2/'2/'

/ ,,,,

E

C

E

MVCCE

MCPdMCsolsolPdinisolPdf

ext=−

= [-] (1)

Rf =

CPdext,MC,∆t

∆t [mmol/g·s] (2)

where E’ is molar quantity of encapsulated TOA per 1g of microcapsules. The microcapsules extracting Pd(II) complexes were put into 10mM of thiourea in 0.1M HCl solution, and back extraction was carried out with shaking at 150 rpm at 298K. The solution was collected at subscribe interval and the concentration of Pd(II) in the solution was measured using ICP. The back extraction ratio, Eb and the initial back extraction rate, Rb were defined as follows:

MCiniMCPd

solsolPd

b MC

VCE elu

⋅=

,,

, [-], Rb =

CPdelu,sol,∆t

∆t [mmol/l⋅s] (3,4)


174

3. RESULTS AND DSICUSSION

3.1. PREPALATION OF MICROCAPSULES

The SEM photograph of the typical microcapsules prepared in this study is shown in Fig. 1. The microcapsules have a smooth surface and a spherical form. The effect of the pore diameter of SPG membrane on properties of microcapsules are shown in Fig.2. The average diameter of microcapsules was about half of the pore diameter of SPG membrane. It is possible to control the size of microcapsules by changing the pore diameter of SPG membrane. The encapsulation efficiencies of TOA were nearly 100% for all the microcapsules prepared. This shows no loss of TOA during the preparation of microcapsules. Furthermore, DM did not affect DP, S and also the pore distribution of microcapsules. This shows that the inner structure of microcapsules was not influenced by the size of microcapsules and influenced by the polymerization condition such as the concentrations of monomer and toluene as diluents.

3.2. EXTRACTION PROPERTIES OF Pd(II) BY MICROCAPSULES

The results of forward extraction of Pd(II) using the microcapsules having various diameters are shown in Fig.3. The forward extraction of Pd(II) was occurred qulickly, and all microcapsules reached the forward extraction equilibrium within 5 minutes. Furthermore, the forward extraction ratios were almost 1 for all the microcapsules. Therefore, almost of TOA encapsulated in the microcapsules react with Pd(II). It is possible to prepare microcapsules having high extraction ratio and fast extraction rate by using SPG emulsification method.

The effect of DM on the initial forward extraction rate of Pd(II) is shown in Fig.4. Rf decreased with an increase in DM, and reached constant value. The increase of DM is considered to cause the decrease in the surface area per unit mass of microcapsules. The less surface area means the less TOA�HCl salt exists near the surface of the microcapsules. So, the

0

10

20

30

DE o

r D

M [µ

m]

0 10 20 30 40 50

DP,SPG [µm]

0

50

100

E [%

]

0

Ctolu = 25wt%CTOA/CDVB = 0.5

Key

� › D E

� ˘ D M

0

5

10

15

20D

p [n

m]

0100200300400500

S [m

2 /g]

00 10 20 30 40 50

DP,SPG [µm]

Fig.1. SEM observation of the microcapsules prepared.

Fig.2. Effect of SPG pore diameter on the characteristics of microcapsules.


175

extraction rate would be decreased. According to the results in Figs.2 and 4, the extraction rate is dependent on only the surface area of microcapsules excluding the pores. Therefore in the case of microcapsule prepared in this study, the diffusion of Pd(II) ions into the pores on microcapsules has little influence on the extraction of Pd(II). The internal diffusion of Pd(II) ions into microcapsules seems to be taking place mainly through the surface of microcapsules. And in a range of DM is less than 20µm, no correlation was observed between Rf and DM. This result is considered that Rf in this range was too fast to determine the difference between each size of microcapsules by using the batch-wise experiment.

All palladium extracted in the microcapsules was back-extracted

effectively using a low concentration of thiourea HCl solution.

4. CONCLUSIONS

Microcapsules containing tri-n-octylamine with the diameter less than 25 µm were prepared by using SPG membrane emulsification and in situ polymerization methods. The microcapsules show high extraction capacity and fast rate for the extraction and back-extraction of Pd(II).

REFERENCES

[1] A. Laguecir, B. Ernst, Y. Frère, L. Danicher, M. Burgard, J. Microencapsulation, 2002, 19, 17.

[2] E. Kamio, K.Kondo, J. Chem. Eng. Jpn., 2007, 35, 574. [3] Y. Frère, L. Danicher, A. Laguécir, J. M. Loureiro, M. Burgard, Int. J. Pharm., 2002,

242, 393. [4] K. Shiomori, H. Yoshizawa, K. Fujikubo, Y. Kawano, Y. Hatate, Y. Kitamura, Sep.

Sci. Technol., 2003, 38, 4059. [5] S. Kiyoyama, S. Yonemura, M. Yoshida, K. Shiomori, H. Yoshizawa, Y. Kawano, Y.

Hatate, React. Funct. Polym., 2007, 67, 522.

Fig.3. Effect of SPG pore diameter on the extraction properties of Pd(II).

Fig.4. Effect of microcapsule diameter on the forward extraction rate of Pd(II).

0

0.5

1

Ef [

-]

0 5 10

t [min]

� � ž

DM [µm]DM [µm] DP,S PG [µm]

6.797.7811.818.0

15203040

�¤ 23.5 50

0

0.005

0.01

0.015

Rf [

mm

ol/g�

Es]

0 25 50 75 100 125

DM [µm]

open : MCs prepared by SPG menbraneclose : MCs prepared by homogenizer


176

APPLICATION OF HEDP IN SORPTION OF HEAVY

METAL IONS ON POLYSTYRENE ANION EXCHANGERS

Dorota KOŁODYŃSKA1) and Zbigniew HUBICKI

Department of Inorganic Chemistry, Faculty of Chemistry, Maria Curie-Sklodowska University, Maria Curie-Sklodowska Sq.2,

20-031 Lublin, Poland, Tel.: +48 (81) 5375736; Fax: +48 (81) 5333348, 1) e-mail: [email protected]

Abstract

The polystyrene anion exchangers were used as materials for removing some toxic metals from wastewater streams containing HEDP (1-hydroxyethane-1,1-diphosphonic acid). The sorption properties of Amberlite IRA 402 and Lewatit MonoPlus MP 500 were studied as a function of phase contact time, pH value, metal concentration and temperature. The studies using the column experiments were also carried out.

1. INTRODUCTION

The most commonly used phosphonates are structural analogues to the well known aminopolycarboxylates such as ethylenediaminetetraacetate (EDTA) and nitrilotriacetate (NTA). The environmental fate of these aminopolycarboxylate chelating agents has received considerable attention. Much less is known about the fate and behaviour of the corresponding phosphonates in the environment. Phosphonates are complexing agents containing one or more C–PO(OH)2 groups. Phosphonates have three main properties: they are effective chelating agents for two- and trivalent metal ions, they inhibit crystal growth and scale formation and they are quite stable under harsh chemical conditions. Important industrial uses of phosphonates are in cooling waters, desalination systems and in oil fields to inhibit scale formation. In pulp and paper manufacturing and in textile industry they are used as peroxide bleach stabilizers, acting as chelating agents for metals that could inactivate the peroxide. In detergents they are used as a combination of chelating agent, scale inhibitor and bleach stabilizer. Around 16,000 tons of phosphonates are used annually in Europe and 30,000 tons in the USA [1, 2].

One of the complexing agents from the group of phosphonic acids used in detergent formulation is 1-hydroxyethane-1,1-diphosphonic acid (HEDP). Although HEDP has a very low biodegradability it is almost


177

totally decomposed in the surface water when exposed to the sunlight. It also participates (in the abiotic conditions) in the mobilisation of heavy metal ions Cu(II), Zn(II), Ni(II), Cd(II), Pb(II), Cr(III) from industrial waste streams. As follows from the literature data, complexation properties of this ligand towards various ions determined in the pH range of natural waters equal from 5 to 9 and with the ion and ligand concentration corresponding to 2 x 10-3 M and their ratio M:L=1:1 was found to be as follows: Fe(II) > Cu(II) >> Zn(II) > Cd(II) > Ni(II) >> Ca(II). It was also found that at basic pH, at high concentrations, 98% of Cu(II), Fe(II) and Zn(II) cations are linked to HEDP. At low concentrations it decreases to 23% for Zn(II) and 93% for Fe(II) at pH 7.0. Insoluble products of HEDP are also formed with heavy metals such as Pb(II) and Cd(II). The structure of HEDP is presented below:

The study allows to determine the influence of chemical conditions on

the ion exchange capacity and on the kinetics of copper(II) and zinc(II) uptake by the strongly basic gel and macroporous anion exchangers Amberlite IRA 402 and Lewatit MonoPlus MP 500.

2. EXPERIMENTAL

2.1. REAGENTS

The studied anion exchangers are produced by Rohm & Hass and Bayer Chemicals. The HEDP was supplied by POCh. All the chemicals were analytical grade reagents. The complexed solutions of Cu(II) and Zn(II) were prepared by dissolving the appropriate chloride salts with 2% excess of stoichiometric quantities of HEDP solution at pH 6.0.

2.2. COLUMN AND BATCH METHODS

The column method was used to study the sorption process of Cu(II) and Zn(II) in the presence of HEDP on Amberlite IRA 402 and Lewatit MonoPlus MP 500. From the obtained breakthrough curves, the weight (Dg) and bed (Dv) distribution coefficients were calculated according to the equations:

OH

P

O

HO

OH

C

CH3

OH

P

O

OH

m

VUUD 0

g

−−=j

v V

VUUD

−−= 0


178

0 20 40 60 80 100 1200

2

4

6

8

10

Cu(II) IRA 402 Zn(II) IRA 402

qt [mg/g]

t [min]0 20 40 60 80 100 120

0

2

4

6

8

10

Cu(II) MP 500 Zn(II) MP 500

qt [mg/g]

t [min]

where: Ū – the effluent volume at c=co/2 (mL), U0 – the dead volume in the column (liquid volume in the column between the bottom edge of anion exchanger bed and the outlet) (mL), V – the void (inter particle) anion exchanger bed volume which amounts to ca. 0.4 of the bed volume (mL), m – the dry anion exchanger weight (g), Vj – the bed volume (mL). In the studies carried out by the batch method the influence of different parameters affecting the sorption capacities was established by shaking 0.5 g of anion exchanger with 50 mL of appropriate solution until the metal complex reached the equilibrium between the solution and the resin phase. The amount of heavy metal complexes sorbed was calculated according to:

mVccq tt /)( 0 ⋅−=

where: c0 – the initial concentration of M(II) in the aqueous phase (mg/L), ct – the concentration of M(II) in the aqueous phase at time t (mg/L), V – the solution volume (L), m – the dry anion exchanger weight (g). The Radiometer pH meter (Model PHM 82) was used for pH measurements. Metal determinations were performed by the AAS method.

3. RESULTS AND DISCUSION

The sorption of studied metal ions in the presence of HEDP can be described as the function of the phase contact time as shown in Fig. 1.

Fig. 1. The influence of the phase contact time on the effectiveness of sorption of Cu(II) and Zn(II) complexes in the presence of HEDP on Amberlite IRA 402 (left)

and Lewatit MonoPlus MP 500 (right).

The sorption equilibrium for Cu(II) and Zn(II) in the presence of HEDP was achieved between 60 and 120 min. Quick large decrease in metal concentration in the aqueous phase followed by a nearly constant value of percentage of metal ion removal equals 99 %. The sorption experiments were also repeated at various temperatures in the range from


179

0 20 30 40 50 600

1

2

3

4

5

6

7

Cu(II) IRA 402 Zn(II) IRA 402 Cu(II) MP 500 Zn(II) MP 500

qt [mg/g]

T [oC]

20 to 60oC and the results are presented in Fig. 2. The percentage of metal removal slightly increases with the increasing temperature up to 60 oC so it can be stated that in this range of temperature is not a decisive factor for sorption effectiveness. As follows from comparison of sorption in different pH values by increasing pH from 5 to 12.5 the effectiveness of sorption rises form 80% to 100% (not presented). The exemplary breakthrough curves of Cu(II) and Zn(II) complexes in the presence of HEDP on the strongly basic polystyrene anion exchangers are presented in Fig. 3.

Comparisons of the calculated values of the weight (Dg) and bed (Dv) distribution coefficients for the Cu(II)-HEDP and Zn-HEDP complexes in the above mentioned system show that the studied anion exchangers exhibit higher affinity for [Cu(hedp)]

2- complexes than for these of [Zn(hedp)]2-.

3. CONCLUSIONS

The studied strongly basic gel polystyrene anion exchangers with respect to their applicability in sorption of Cu(II) and Zn(II) complexes with HEDP can be arranged as follows: Lewatit MonoPlus MP 500> Amberlite IRA 402. Particularly effective sorption results of the above mentioned complexes of Cu(II) and Zn(II) were obtained at pH above 11.0

REFERENCES

[1] V. Deluchat, J.C. Bollinger, B. Serpaud , C. Caullet, Talanta, 1997, 44, 897. [2] B. Nowack, A. T. Stone, Water Res., 40, 2006, 2201.

Fig. 2. The influence of the temperature

on the effectiveness of sorption of Cu(II) and Zn(II) complexes in the

presence of HEDP on Amberlite IRA 402 and Lewatit MonoPlus MP 500.

Fig. 3. The breakthrough curves of Cu(II) and Zn(II) complexes in the

presence of HEDP on Amberlite IRA 402 and Lewatit MonoPlus MP 500.

0 2000 4000 6000 8000 10000 120000,0

0,2

0,4

0,6

0,8

1,0

Cu(II) IRA 402 Zn(II) IRA 402 Cu(II) MP 500 Zn(II) MP 500

c/co

V [mL]


180

INFLUENCE OF THE FUNCTIONAL GROUPS OF ION EXCHANGERS ON SORPTION OF Cu(II) AND Zn(II)

COMPLEXES WITH METATARTARIC ACID

Dorota KOŁODYŃSKA1), Zbigniew HUBICKI and Marzena GĘCA

Department of Inorganic Chemistry, Faculty of Chemistry, Maria Curie-Sklodowska University, Maria Curie-Sklodowska Sq.2,

20-031 Lublin, Poland, Tel.: +48 (81) 5375736; Fax: +48 (81) 5333348, 1) e-mail: [email protected]

Abstract

The studies show that metatartaric acid is an effective complexing agent for removal of copper(II) and zinc(II) ions using polystyrene and chelating ion exchangers. The effect of important parameters such as the value of pH, the metal(II) ion and ligand concentration as well as their molar ratio and type of functional group of ion exchanger used was studied. The obtained results indicate that the effectiveness of sorption depends on the kind of complex form in the resin phase and the structure of the ion exchanger.

1. INTRODUCTION

Dicarboxylic acids such as oxalic, malonic, succinic, glutaric, fumaric etc. as well as hydroxycarboxylic ones such as tartaric, citric are known as complexing agents. It is well known that the most important factors controlling the process of complex formation are the type of complexing agents and pH. Carboxylic acids form complexes of moderate strength with most heavy metal cations. They are widely used in many industries. Apart from oxalic acids, which form insoluble compounds with a large number of divalent metals such as Cu(II), Zn(II), Ni(II), Co(II), Sn(II), Pb(II) in the wide range of pH, their chemical interaction for example with Cu(II) are different. Formation of Cu(II)-malonic acid precipitates is very slow and begins only after 2-3 days and pH about 5. Forming of Cu(II) precipitates with glutaric, tartaric and citric acids is also slow [1,2]. Therefore in our investigations metatartaric acid (obtained by heating tartaric acid) was chosen as a complexing agent for coper(II) and zinc(II) ions. Metatartaric acid (H2Tar) and its salts occur in various ionization forms. In the solution the following two-step equilibrium exists:

H2tar = H+ + Htar- and Htar- = H+ + tar2-.


181

In this paper the continuous column method and the batch technique were used to understand the influence of pH, molar ratio and concentration on sorption effectiveness of two selected metal ions in the presence of metatartaric acid.

2. EXPERIMENTAL

2.1. MATERIALS, REAGENTS AND INSTRUMENTS

The anion exchangers Lewatit MonoPlus M 500, Lewatit MonoPlus MP 64, Lewatit MP 62 as well as the chelating ion exchanger Lewatit TP 207 were supplied by Bayer Chemicals. The metatartaric acid (2,3-dihydroxy-disuccinic acid) C4H6O6 was supplied by Begerow. All the reagents used were of analytical grade. Metal solutions were prepared from concentrated stock solutions of Cu(II) and Zn(II) chlorides and the appropriate amount of H2tar. The Radiometer pH meter (Model PHM 82) was used for pH measurements. Metal determinations were performed with Varian Spectr AA-880.

2.2. PROCEDURES

The column method was used to study the sorption process of Cu(II) and Zn(II) on the selected ion exchangers. From the obtained breakthrough curves, the sorption parameters the weight (Dg) and bed (Dv) distribution coefficients as well as the working ion exchange capacities (Cw) were calculated. In the studies carried out by the batch method the effect of pH, the M(II):tar2- ratio and concentration of metal ions was studied. The sorption capacities at the specific time (qt) and at the equilibrium was established according to:

where: c0 – the initial concentration of M(II) in the aqueous phase (mg/L), ct, ce – the concentration of M(II) in the aqueous phase at time t or equilibrium (mg/L), V – the solution volume (L), m – the dry anion exchanger weight (g). The copper and zinc concentrations in the influent, effluent and rafinate were determined by the AAS method.

3. RESULTS AND DISCUSION

3.1. EFFECT OF pH

The influence of pH on the sorption of metal ions on polystyrene anion and chelating exchangers was studied. The obtained results are presented in Fig. 1.

mVccq tt /)( 0 ⋅−= mVccq ee /)( 0 ⋅−=


182

Fig. 1. The influence of the pH value on the effectiveness of sorption of Cu(II) complexes in the presence of H2tar on Lewatit MonoPlus M 500, Lewatit

MonoPlus MP 64 and Lewatit MP 62 as well as Lewatit TP 207. As expected in the case of the strongly basic anion exchanger Lewatit MonoPlus M 500 the amount of sorbed ions increases with the increasing pH whereas for the medium and weakly basic anion exchangers decreases with the increasing pH. For the chelating ion exchanger with the iminodiacetate functional group in the range from 3 to 5 it decreases and then slightly increases.

3.2. EFFECT OF THE TYPE OF FUNCTIONAL GROUP

The ion exchangers used in the paper possess the following functional groups: Lewatit MonoPlus M 500 – trimethylammonium (type 1), Lewatit MonoPlus MP 64 – trimethylammonium and tertiary amine, Lewatit MP 62 – tertiary amine as well as Lewatit TP 207 – iminodiacetate.

The exemplary breakthrough curves of Cu(II) complexes in the presence of metatartaric acid on the strongly, moderately and weakly basic ion exchangers as well as on the chelating ion exchanger are presented in Fig. 2.

Based on the obtained sorption parameters these functional groups of ion exchangers with respect to their efficiency in sorption of Cu(II) from Zn(II) complexes with metatartaric acid can be arranged as follows:

trimethylammonium (M 500) > iminodiacetate (TP 207) > trimethylammonium and tertiary amine (MP 64) > tertiary amine (MP 62).

These results were confirmed by those obtained by the static method. The sorption capacities (qt) increased with the increasing time of equilibration and reached the plateau value at 240 min for Cu(II) and Zn(II) complexes (data not presented).

1 2 3 4 5 6 7 8 9 100

1

2

3

4

5

6

7

8

M 500 MP 64 MP 62 TP 207

q e [m

g/g]

pH


183

Fig. 2. The breakthrough curves of Cu(II) complexes in the presence of H2tar on Lewatit MonoPlus M 500, Lewatit MonoPlus MP 64 and Lewatit MP 62 as well

as Lewatit TP 207.

The adsorption data obtained for these two cations were explained well for the Langmuir adsorption data model rather than the Freundlich model and the adsorption capacities obtained were found to be 11.12 mg/g and 10.24 mg/g for Lewatit MonoPlus M 500.

3. CONCLUSIONS

The studied gel and macroporous polystyrene ion exchangers with respect to their applicability in sorption of Cu(II) and Zn(II) complexes with metatartaric acid can be arranged as follows:

Lewatit MonoPlus M 500 > Lewatit TP 207 > Lewatit MonoPlus MP 64 > Lewatit MP 62.

Taking into account the overall results, it may be stated that the strongly basic polystyrene anion exchanger as well as the chelating ion exchanger seem to be promising in the treatment of wastewaters containing both heavy metal ions and metatartaric acid.

REFERENCES

[1] R.S. Juang, F.C. Wu, R.L.Tseng, Water Res., 1999, 33, 2403. [2] K.K. Wong, C.K. Lee, K.S. Low, M.J. Haron, Chemosphere, 2003, 50, 23.

0 5000 10000 15000 20000 250000,0

0,2

0,4

0,6

0,8

1,0

M500 MP 64 MP 62 TP 207

c/co

V [mL]


184

BAYPURE CX 100 AS A NEW GENERATION COMPLEXING

AGENT USED IN REMOVAL OF HEAVY METAL IONS FROM WASTEWATERS

Dorota KOŁODYŃSKA1), Halina HUBICKA and Zbigniew HUBICKI

Department of Inorganic Chemistry, Faculty of Chemistry, Maria Curie-Sklodowska University, Maria Curie-Sklodowska Sq. 2,

20-031 Lublin, Poland, Tel.: +48 (81) 5375736; Fax: +48 (81) 5333348, e-mail address: 1) [email protected]

Abstract

In the presented paper the sorption of heavy metal ions in the presence of iminodisuccinic acid (IDS) on polyacrylate anion exchangers from aqueous solutions was studied. It is a complexing agent of new generation as it undergoes biodegradation thus being an alternative for the reagents of the EDTA or NTA type. Based on the research, the applicability of gel and macroporous polyacrylate anion exchangers with different functional active groups was determined by the dynamic technique. Batch experiments were also carried out to determine the factors affecting sorption and kinetics of the sorption process.

1. INTRODUCTION

Tetrasodium salt of N-(1,2-dicarboxyethylene)-D,L-aspartic acid (sodium iminodisuccinate) produced since 1998 by the Lanxess (Bayer AG Leverkusen, Germany) as a Baypure CX 100 (IDS), is a synthetic compound readily biodegradable, which is made from maleic anhydride, ammonia and sodium hydroxide solution. In water equal amounts of the four possible stereoisomers: two identical meso forms RS'-IDS and SR'-IDS and two forms RR'-IDS, SS'-IDS are found [1,2]. The structure of IDS can be described as follows:

It forms chelates of octahedral structure with many metal ions. The reaction between the metal ion and the anion of IDS acid is reversible and occurs with the ratio 1:1 [3]:

Mm+ + idsn- ⇄ [M(ids)](n-m)- (1)

HOOC

HOOC

COOH

COOHN

H


185

IDS is characterized by low remobilization of heavy metal ions and is superior to the conventional complexing agents because of its excellent ability as complex iron(III), Cu(II) and Ca(II) ions and its very good biodegradability. For these reasons sodium iminodisuccinate seems to be an important alternative soon for the traditional agents of the EDTA type.

The aim of the presented paper was to examine the possibility of using polyacrylate anion exchangers for removal of heavy metal ions such as Cu(II), Co(II) and Ni(II) from aqueous solutions in the presence of IDS.

2. EXPERIMENTAL

The following polyacrylate anion exchangers were used in the investigations: Amberlite IRA 458, Amberlite IRA 958 and Amberlite IRA 67. Prior to the use they were washed with 1 M NaOH, 1 M HCl and deionized water. The resin was finally converted to the appropriate forms. The following heavy metal ions were selected - copper(II), cobalt(II) and nickel(II) in the nitrates form. The initial concentration of metals ions was 0.1 M. The aqueous solutions of each heavy metal ions in the presence of IDS were prepared by dissolving the stock solutions.

In order to measure affinity of Cu(II)-IDS, Co(II)-IDS and Ni(II)-IDS complexes the breakthrough curves were determined. The frontal analysis process was carried out in glass columns. The prepared solutions were passed continuously downward through the resin beds keeping the flow rate at 0.8 mL/cm2⋅min and the effluent was collected in fractions. The weight (Dg) and bed (Dv) distribution coefficients as well as the working ion exchange capacities (Cw) of M(II) were calculated. The recovery factors (%R) were determined by means of the static method (0.5 g of appropriate dry anion exchanger was placed in a 100 mL stoppered conical flask containing 50 mL of solution and shaken at the constant temperature in the three parallel series). pH was measured with a Radiometer pH meter (Model PHM 82). The concentrations of Cl-/NO3

- ions were not determined. The contents of each metal in the raffinate and eluate were determined by the AAS method (Varian SpectrAA- 880).

3. RESULT AND DISCUSION

The sorption process of cobalt(II) and nickel(II) from the single-metal ion solutions in the presence of IDS was conducted by the dynamic method on Amberlite IRA 458, Amberlite IRA 67 and Amberlite IRA 958. As follows form the obtained breakthrough curves (Fig.1) and the calculated weight (Dg) and bed (Dv) distribution coefficients the nickel(II) complexes with IDS exhibit higher affinity for these anion exchangers than that for the cobalt(II) complexes.


186

0 4000 8000 12000 160000,0

0,2

0,4

0,6

0,8

1,0

Amberlite IRA 458 Co(II)-IDS=1:1 Ni(II)-IDS=1:1 Co(II)-IDS=1:2

c/co

V [mL] 0 4000 8000 12000 160000,0

0,2

0,4

0,6

0,8

1,0


c/co

V [mL]

0 4000 8000 12000 160000,0

0,2

0,4

0,6

0,8

1,0


c/co

V [mL]

0 50 100 150 200 2500

1

2

3

4

5

6

7

Cu(II)-IDS=1:1 Amberlite IRA 458 Amberlite IRA 67 Amberlite IRA 958

qt [mg/g]

t [min]

0 50 100 150 200 2500

1

2

3

4

5

6

7

Ni(II)-IDS=1:1 Amberlite IRA 458 Amberlite IRA 67 Amberlite IRA 958

qt [mg/g]

t [min]

0 50 100 150 200 2500

1

2

3

4

5

6

7

Co(II)-IDS=1:1 Amberlite IRA 458 Amberlite IRA 67 Amberlite IRA 958

qt [mg/g]

t [min]

It can be also stated that the sorption effectiveness depends on the M(II):L molar ratio and the form in which these anion exchangers were used (data not presented). As follows from the studies in the pH range from 2.0 to 6.0 using the static method, at the constant phase contact time equal 6 h, the values of recovery factor (%R) of metal ions on Amberlite IRA 458 and Amberlite Amberlite IRA 958 increase and reach the plateau at pH above 5. In the case of weakly basic Amberlite IRA 67 these values slightly decrease. Therefore in the next stage, the investigations of sorption by the static method depending on the phase contact time were carried out in the M(II):IDS=1:1 and 1:2 systems at pH without adjustments. The values of anion exchangers capacities (qt) determined for the Cu(II), Co(II) and Ni(II) complexes with IDS are presented in Fig. 2.

Fig.1. The breakthrough curves of Co(II) and Ni(II) complexes with IDS on Amberlite IRA 458, Amberlite IRA 67 and Amberlite IRA 958.

Fig.2. The effect of the pH value on the sorption of Cu(II), Co(II) and Ni(II) complexes with IDS on Amberlite IRA 458, Amberlite IRA 67 and Amberlite IRA 958.


187

As follows from the comparison of the obtained results, the recovery factors assume the values in the range from 60 % to 100% and they are much more differentiated for the Co(II) complexes than for Cu(II) and Ni(II) ones. The adsorption data for a wide range of adsorbate concentrations are most accurately described by the adsorption isotherms, such as those of Langmuir or Freundlich, which relate adsorption amount to the equilibrium adsorbate concentration in the resin phase. As in Table 1, evaluation of the equilibrium adsorption constant and the maximum sorption capacity was readily obtained from the experimental data by plotting the reciprocal amount of adsorbed metal ions. The highest amounts of metal ions adsorbed per unit weight (mg/g) of the anion exchanger are for the Cu(II) and Co(II) complexes on Amberlite IRA 458 and Amberlite IRA 958.

Table 1. Comparison of Cu(II), Co(II) and Ni(II) complexes with the IDS amount adsorbed per unit of weight of anion exchanger (mg/g) in the M(II)-IDS=1:1

system.

Cu(II) Co(II) Ni(II) Amberlite IRA 458 65.88 60.65 23.96 Amberlite IRA 67 15.98 12.45 14.16

Amberlite IRA 958 56.78 51.00 33.33

As follows from the research the polyacrylate anion exchangers, according to their applicability for sorption of Cu(II), Co(II) and Ni(II) complexes in the presence of IDS in the 1:1 system can be put in the order:

Cu(II) Amberlite IRA 458 > Amberlite IRA 958 > Amberlite IRA 67, Co(II) Amberlite IRA 458 > Amberlite IRA 958 > Amberlite IRA 67,

Ni(II) Amberlite IRA 958 > Amberlite IRA 67 > Amberlite IRA 458.

3. CONCLUSIONS

The goal of this study was the assessment of the performance of commercial polyacrylate anion exchange resins, in order to recover the copper(II), cobalt(II) and nickel(II) in the presence of IDS from industrial effluents. The selected resins were Amberlite IRA 458 and Amberlite IRA 958 and the experimental results showed that both are effective in the removal of these heavy metal ions from aqueous solutions. The experimental data were well fitted by the Langmuir model.

REFERENCES

[1] Brochure of the company Bayer Chemicals AG (http://www.baypure.com). [2] T.P. Knepper, Trends in Analytical Chem., 2003, 23, 708. [3] A.M. Niećko, J. Pernak, Przem. Chem., 2006, 85, 635.


188

STUDIES OF WEAKLY BASIC ANION EXCHANGERS APPLICABILITY IN SORPTION OF PALLADIUM(II)

COMPLEXES FROM CHLORIDE SOLUTIONS.

Anna WOŁOWICZ and Zbigniew HUBICKI

Department of Inorganic Chemistry, Faculty of Chemistry, Maria Curie-Skłodowska University,

Maria Curie-Skłodowska Square 2, 20-031 Lublin, Poland

Abstract Weakly basic anion exchangers such as Amberlite IRA 95 and Amberlite IRA

96 were used in sorption of palladium(II) complexes from the 0.1 – 2.0 M HCl – 1.0 M NaCl – 100 ppm Pd(II); 0.1 – 2.0 M HCl – 2.0 M NaCl – 100 ppm Pd(II) solutions. Sorption of palladium(II) complexes was carried out by means of the static and dynamic methods. Dynamic processes were applied in order to determine the breakthrough curves of Pd(II) complexes. The working capacities, weight and bed distribution coefficients were determined from the Pd(II) breakthrough curves. Using the static metods, the amount of palladium(II) complexes sorbed onto anion exchangers under discussion was calculated by means of the mass balance-relationship. Moreover, the influence of concentration of hydrochloric acid and sodium chloride as well as the phase contact time was discussed.

1. INTRODUCTION

Anion exchangers are commonly used in sorption of palladium(II) complexes due to the fact that palladium(II) complexes exist in chloride solutions usually in the anionic form, therefore they are able to undergo the ion exchange reaction. In this group one can mention weakly, medium and strongly basic anion exchangers [1-5].

The aim of this work was to study the applicability of two anion exchnagers (Amberlite IRA 95 and Amberlite IRA 96) in sorption of palladium(II) complexes from chloride solutions in order to check which one is more promising.

2. EXPERIMENTAL

Anion exchangers. Amberlite IRA 95 and Amberlite IRA 96, weakly basic anion exchangers were used in sorption of palladium(II) complexes. Their characteristics are presented in Tab. 1.


189

Tab.1. Characteristics of the anion exchangers.

Description Amberlite IRA 96 Amberlite IRA 95 Producer Rohm & Haas Co., France

Type Weakly basic Functional groups Polyamine Tertiary amine

Ionic form as shipped - Free base Matrix Styrene - divinylbenzene

Structure macroporous Bead size [mesh] 16-50

Operating pH range 0-7 Max. operating temperature [K] 373

Total exchange capacity [meq/cm3] 1.3 1.25 Physical form and appearance Opaque spherical beads

- data not avaliable

Methods. The static method consists of the following stages: drying of anion exchangers, weighing of resins (0.5 ± 0.0005 g), shaking (laboratory shaker type 358 S, Elpin +, Poland) organic and aqueous phases (50 cm3) and then determination of concentration of Pd(II) complexes by means of the iodide method [6]. The amount of Pd(II) complexes sorbed onto anion exchange resins was calculated from the following mass balance-relationship:

WVCCq tt /)( 0 −= (1)

where C0 is the initial concentration of palladium(II) complexes in the aqueous phase, 100ppm, Ct is the concentration of palladium(II) complexes in solutions after time t, V is the volume of the solutions (dm3), W is the weight of the dry ion exchangers used (g) [7].

The weight distribution coefficient, Dw, was calculated from Eq.(2):

Dw = (U – Uo – V)/W (2)

where U is the effluent volume at C=0.5C/Co (cm3), Uo is the dead volume in the column (cm3), V is the void (inter – particle) ion-exchanger, W is the dry ion-exchanger weight (g), whereas the bed distribution coefficient, Dv, from Eq.(3):

Dv = Dw ⋅ dz (3)

where dz is the bed density (g/cm3).


190


Palladium can form stable chloro-complexes such as PdCl+, PdCl2, PdCl3

- and PdCl4

2- in the acidic chloride medium (Fig.1):

Pd2+ + Cl- � PdCl+ log K = 6.1

Pd2+ + 2 Cl- � PdCl2 log K = 10.7

Pd2+ + 3 Cl- � PdCl3 - log K = 13

Pd2+ + 4 Cl- � PdCl4 2- log K = 16

Pd2+ + 5 Cl- � PdCl5 3- log K = 14

Pd2+ + 6 Cl- � PdCl6 4- log K = 12

Fig. 1. Forms of Pd(II) chlorocomplexes, (Pd2+

TOT= 10.0 µM).

At 0.1 M Cl- and higher, the predo-minant species in solutions is PdCl4

2-.

Due to the fact that Pd(II) complexes occur in the anionic form, they are capable of undergoing anion exchange reactions.

Amberlite IRA 95 and Amberlite IRA 96 as the anion exchangers possess the polyamine functional groups. These groups contain one or more coordinating atoms, N donor atoms, therefore Pd(II) complexes are able to combine with resins by means of the solvent mechanism R.(1):

MClp(p-n)- + m (R)(o) ↔ MCln(R)m (o) + (p-n) Cl- (1)

Moreover, in acidic solutions polyamine groups are protonated, therefore the ionic interactions (R.(2)) between the protonated amines and the chloropalladium(II) complexes can be observed [8].

MClp(p-n)- + (p-n) (RH+Cl-)(o) ↔ (RH+)p-nMClp(o)

(p-n)- + (p-n) Cl- (2)

The dynamic processes were applied in order to determine the breakthrough curves of Pd(II) complexes which are shown in Fig.2.

a) 0 1000 2000 3000 4000 5000V [cm3]

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

C/Co

1.0 M NaCl0.1 M HCl - 1.0 M NaCl0.5 M HCl - 1.0 M NaCl1.0 M HCl - 1.0 M NaCl2.0 M HCl - 1.0 M NaCl

b) 0 1000 2000 3000 4000

V [cm3]

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

C/Co

2.0 M NaCl0.1 M HCl - 2.0 M NaCl0.5 M HCl - 2.0 M NaCl1.0 M HCl - 2.0 M NaCl2.0 M HCl - 2.0 M NaCl

Fig. 2. Breakthrough curves of Pd(II) complexes in the: a) 0.1 – 2.0 M HCl – 1.0 M

NaCl, b) 0.1 – 2.0 M HCl – 2.0 M NaCl systems for Amberlite IRA 95.


191

The breakthrough curves of Pd(II) complexes were used in order to calculate the weight and bed distribution coefficients and the working ion exchange capacities (Tab.2).

Table 2. Comparison of the working ion exchange capacity values and the weight and bed distribution coefficients of Pd (II) complexes in the 0.1 – 2.0 M HCl – 1.0

M NaCl and 0.1 – 2.0 M HCl – 2.0 M NaCl systems determined for the anion exchangers under discussion.

Cr Dw Dv Cr Dw Dv

2.0 M NaCl 0.0162 957.4 267.2 0.0132 706.7 206.7 0.1 M HCl – 2.0 M NaCl 0.0150 952.7 265.9 0.0103 698.5 204.3 0.5 M HCl – 2.0 M NaCl 0.0108 786.8 219.6 0.0103 610.9 178.7 1.0 M HCl – 2.0 M NaCl 0.0075 700.8 195.6 0.0048 524.4 153.4 2.0 M HCl – 2.0 M NaCl 0.0050 542.1 151.3 0.0036 402.7 117.8

The values of the working ion exchange capacities as well as the weight and bed distribution coefficients of Pd(II) complexes depend on the hydrochloric acid concentration in the systems under discussion and decrease with the increasing hydrochloric acid concentration. Decrease in the sorption yield of PdCl4

2- observed with the higher hydrochloric acid concentration in the systems under disccusion is caused by competitive sorption of Cl- and HCl2

– type ions. HCl2 – ion as the ion of the acid

stronger than HCl exhibits greater affinity for anion exchangers. Making comparison between the amount of palladium(II) complexes,

qt,calculated for Amberlite IRA 95 and Amberlite IRA 96 the higer values of qt are observed for Amberlite IRA 95.

REFERENCES

[1] El-Said N., Mekhael S., Khalifa S.M., Aly H.F., J. Nucl. Sci. Technol., 1996, 208, 243. [2] Els E.R., Lorenzen L., Aldrich C., Miner. Eng., 1997, 10, 1177. [3] Els E.R., Lorenzen L., Aldrich C., Miner. Eng., 2000,13, 401. [4] Z. Hubicki, M. Leszczyńska, Desalination, 2005,175, 227. [5] Z. Hubicki, M. Leszczyńska, Desalination, 2005,175, 289. [6] Z. Marczenko and A.P. Ramsza, Chem.Anal.,1978, 23, 23. [7] M. Özacar, I. A. Şengil, J. Hazard. Mater., 2003, B98, 211. [8] F.L. Bernardis, R.A. Grant, D.C. Sherrington, React. Funct. Polym., 2005, 65,205.

Systems Amberlite IRA 95 Amberlite IRA 96

1.0 M NaCl 0.0321 1540.7 430.0 0.0251 1143.2 334.4 0.1 M HCl – 1.0 M NaCl 0.0290 1379.1 384.9 0.0246 750.8 219.6 0.5 M HCl – 1.0 M NaCl 0.0208 1162.3 324.4 0.0182 905.6 264.9 1.0 M HCl – 1.0 M NaCl 0.0159 946.6 264.2 0.0132 1194.5 349.4 2.0 M HCl – 1.0 M NaCl 0.0097 716.2 199.9 0.0089 544.3 159.2


192

SORPTION OF PALLADIUM(II) ON LEWATIT MP 62 – EQUILIBRIUM AND K INETICS CHARACTERISATION.

Anna WOŁOWICZ, Zbigniew HUBICKI



Abstract Sorption of palladium(II) chlorocomplexes onto the macroporous anion

exchange resin Lewatit MP 62 from the chloride solution of the composition: 0.1 – 2.0 M HCl – 1.0 M NaCl – 100 ppm Pd(II); 0.1 – 2.0 M HCl – 2.0 M NaCl – 100 ppm Pd(II) was discussed. The sorption research of Pd(II) complexes on this resin was carried out by means of static and dynamic methods. The amount of palladium(II) complexes sorbed onto anion exchangers, the working ion exchange capacities and the weight and bed distribution coefficients were calculated. Moreover, the kinetics study of these systems was carried out by means of the pseudo first- , second- order and intraparticle diffusion kinetic models. The rate constants for the three models and the correlation coefficients have been calculated in order to assess which model provides the best fit of the predicted data with the experimental results.

1. INTRODUCTION

Sorption or ion exchange methods are widely used and are the most effective for preconcentration and separation of noble metal ions from aqueous solutions.

Both palladium(II) sorption and the study of sorption kinetics are very important in order to understand profoundly the reaction pathways and the mechanism of sorption and recovery reactions. Numerous sorption systems have been investigated particularly during the past 15 years. Chemical kinetics explains how fast the rate of chemical reaction proceeds and also the factors affecting the reaction rate. The nature of sorption process will depend on physical or chemical characteristics of the systems as well as on the system conditions. The most commonly used kinetic expressions to explain the solid/liquid sorption processes are the pseudo first-, and second-order kinetics [1-4]. The interparticle diffusion [5-7] model was also used to predict the sorption kinetics. The Lanergen pseudo first-order kinetics is not proved to be effective in representing the experimental kinetic data for the entire sorption period whereas the pseudo second-order kinetics was found to explain well the kinetics of most sorption systems for the entire


193

range of sorption period. Interparticle diffusion will represent the experimental kinetics where the effect of pore diffusion and the film diffusion are expected to be negligible.

The aim of this paper was to study sorption and sorption kinetics of palladium(II) chlorocomplexes onto Lewatit MP 62 from the 0.1 – 2.0 M HCl – 1.0 M NaCl – 100 ppm Pd(II); 0.1 – 2.0 M HCl – 2.0 M NaCl – 100 ppm Pd(II) solutions in order to predict which theoretical models provide the best fit of the predicted data with the experimental results.

2. MATERIALS AND METHODS

2.1. ION EXCHANGER

The macroporous, styrene–divinylbenzene, weakly basic anion exchanger Lewatit MP 62 was used in sorption of palladium(II) chlorocomplexes. This resin contains the tertiary amine functional groups. The ionic form as shipped is a free base. Lewatit MP 62 can be used in the whole pH range and in the temerature range from 274 to 347 K. Moreover, the total exchange capacity of this resin is equal to 1.7 meq/cm3.

2.2. EXPERIMENTAL PROCEDURE

A laboratory shaker (type 358 S, Elpin +, Poland) was used in the static method to shake the ion exchanger (0.5 g) and liquid phases (50cm3). After different periods of time, the ion exchanger was separated by filtration using filter paper and the supernatant liquid is analyzed by the spectrophotometric iodide method [8]. The ions concentration retained in the ion exchanger phase was calculated using the following mass balance-relationship:

WVCCq tt /)( 0 −= (1)

where C0 is the initial concentration of palladium complexes in the aqueous phase, 100 ppm, Ct is the concentration of palladium complexes in solutions after time t, V is the volume of the solutions (dm3), W is the weight of the dry ion exchanger used (g).

2.3. KINETIC MODELS OF SORPTION

In order to examine the controlling mechanism of sorption processes such as mass transfer and chemical reaction several kinetic models are used to test the experimental data.

A simple kinetic model is the pseudo first-order equation (the Legergren’s equation) in the following linear form:


194

303.2/log)log( 111 tkqqq t −=− (2)

where q1 and qt are the amounts of palladium complexes sorbed at equilibrium and at t (mg g -1), respectively, and k1 is the rate constant of the pseudo first - order sorption (min -1). The other model is the pseudo second-order kinetic equation which can be expressed as:

tqqkq

t

t 2222

11 += (3)

where q2 is the amount of palladium complexes sorbed at equilibrium (mg g -1), k2 is the equilibrium rate constant of the pseudo second-order sorption (g mg -1 min -1). The rate parameter (kint) for the interparticle diffusion can be defined as:

2/1int tkqt = (4)

where kint is the interparticle diffusion rate constant (mg g -1 min -1/2) [1-6].


In the chloride solution palladium(II) can form stable chloride, hydroxy-chloride and hydroxide complexes which occur in acidic chloride media mostly in the anionic form therefore they are capable of undergoing anion-exchange reactions. The major species in the acidic solution containing 0.1 M and higher chloride concentration is PdCl4

2-. The Pd-chlorocomplexes form ion pairs with the anion-exchange resins.

Lewatit MP 62 is a weakly basic anion exchanger which can be protonated significantly, and therefore behaves as ion exchangers, only at high acid concentrations. In acidic media, palladium(II) complexes are bonded through the protonated nitrogen of the amine group of the resins:

MClp(p-n)- + (p-n) (RH+Cl-)(o) ↔ (RH+)p-nMClp(o)

(p-n)- + (p-n) Cl- (1)

As follows from the research results presented in Tab. 1 the working ion–exchange capacities (Cr, g/cm3) as well as the weight (λ) and bed (λ’ ) distribution coefficients of Pd(II) complexes for Lewatit MP 62 depend on hydrochloric acid concentration in the systems under discussion and decreased with the increasing hydrochloric acid concentration. The distribution coefficients start at a maximum and drop with the increasing hydrochloric acid concentration due to the increasing competition between Cl- and metal-complex anions. These competitions lead to a reduced capacity, too.


195

Table 1. Comparison of working ion exchange capacity values and the weight and bed distribution coefficients of Pd(II) complexes in the 0.1 – 2.0 M HCl – 1.0 M

NaCl and 0.1 – 2.0 M HCl – 2.0 M NaCl systems for Lewatit MP 62.

Systems Cr λ λ' Systems Cr λ λ'

1.0 M NaCl 0.0269 1182.8 368.8 2.0 M NaCl 0.0175 758.5 236.5 0.1 M HCl –1.0 M NaCl 0.0205 973.1 303.4

0.1 M HCl – 2.0 M NaCl 0.0136 678.6 211.6

0.5 M HCl – 1.0 M NaCl 0.0185 871.1 271.6

0.5 M HCl – 2.0 M NaCl 0.0098 581.8 181.4

1.0 M HCl – 1.0 M NaCl 0.0143 795.7 248.1

1.0 M HCl – 2.0 M NaCl 0.0066 489.7 152.7

2.0 M HCl – 1.0 M NaCl 0.0127 708.8 221.0

2.0 M HCl – 2.0 M NaCl 0.0055 335.5 104.6

*dz = 0.3118 g/cm3,density of Lewatit MP 62; Cr, g/cm3 - working ion–exchange capacities; λ - weight distribution coefficient, λ' - bed distribution coefficient

The kinetic results (Tab. 2) show that of all examined kinetic models, the pseudo second-order kinetic model was found to agree well with the experimental data for all systems under discussion. Moreover, the correlation coefficients of the plot t/qt versus t, r2

2, achieve very high values.

Table 2. Kinetic parameters for palladium(II) complexes onto Lewatit MP 62.

First-order kinetic equation

Second-order kinetic equation

Interparticle diffusion Systems

qe,exp

qe k1 r12 qe k2 h r2

2 kint r int2

1.0 M NaCl 9.98 0.61 0.03 0.749 10.0 0.09 9.30 1.0 0.13 0.378 0.1 M HCl–1 M NaCl 9.17 0.96 0.01 0.544 9.19 0.06 4.97 1.0 0.15 0.455 0.5 M HCl–1 M NaCl 9.48 1.06 0.01 0.596 9.49 0.07 6.27 1.0 0.14 0.331 1 M HCl–1 M NaCl 9.69 0.96 0.01 0.683 9.70 0.07 6.86 1.0 0.13 0.380 2 M HCl–1 M NaCl 8.55 0.85 0.01 0.610 8.58 0.07 4.86 1.0 0.15 0.432

2.0 M NaCl 9.97 0.38 0.02 0.945 10.1 0.04 4.15 1.0 0.15 0.581 0.1 M HCl–2 M NaCl 9.12 0.42 0.01 0.766 9.20 0.02 1.83 1.0 0.22 0.700 0.5 M HCl–2 M NaCl 8.93 0.68 0.01 0.708 8.93 0.04 3.56 1.0 0.14 0.496 1 M HCl–2 M NaCl 8.58 0.62 0.01 0.647 8.58 0.04 2.90 1.0 0.16 0.528 2 M HCl–2 M NaCl 7.91 0.54 0.01 0.762 7.92 0.03 2.03 1.0 0.16 0.545

REFERENCES

[1] Y.S. Ho, J. Hazard. Mater., 2006, B136, 681. [2] Y.S. Ho, G. McKay, Process. Biochem., 1999, 34, 451. [3] Y.S. Ho, G. McKay, Can. J. Chem. Eng., 1998, 76, 822. [4] K.V. Kumar, S. Sivansesan, Process. Biochem., 2006, 41, 1198. [5] K.V. Kumar, V. Ramamurthi, S. Sivansesan, J.Colloid Interface Sci., 2005, 284, 14. [6] G.M. Walker, L. Hansen, J.A. Hanna, S.J.Allen, Water Res., 2003, 37, 2081. [7] G. McKay, S.J.Allen, Can. J. Chem. Eng., 1980, 58, 521. [8] Z. Marczenko, A.P. Ramsza, , Chem.Anal., 1978, 23, 23.


196

PRECONCENTRATION OF TRACES OF PALLADIUM(II)

ONTO CHELITE S CONTAINING CHELATING FUNCTIONAL GROUPS FROM CHLORIDE SOLUTIONS.

Anna WOŁOWICZ , Monika WAWRZKIEWICZ, Zbigniew HUBICKI



Abstract The extended use of palladium in automotive catalytic converters has led to

increasing concentration of this metal in the environment. In the determination of trace or ultra-trace amounts of noble metals in environmental samples the preconcentration procedure is required, even with the high sensitivity and selectivity of modern instrumental techniques.

Preconcentration of traces of palladium(II) onto Chelite S containing thiol, the chelating group using the ion exchange method from the 0.1 – 2.0 M HCl – 1.0 M NaCl – 100 ppm Pd(II); 0.1 – 2.0 M HCl – 2.0 M NaCl – 100 ppm Pd(II) solutions by means of the static method was discussed. Moreover, the amount of palladium(II) complexes sorbed onto Chelite S was calculated by means of the mass balance-relationship. The different types of the pseudo second-order kinetic model were applied in order to explain the mechanism of palladium(II) sorption onto Chelite S. The correlation coefficients, the rate constants as well as the initial sorption rate were calculated from these models to assess which one provides the best fit with the experimental data.

1. INTRODUCTION

Preconcentration is an inevitable step in the determination of traces of noble metals (PGMs) in geological, metallurgical and automotive catalyst converter samples [1-3]. Ion exchange is the most frequently used for matrix elimination and concentration of platinum and palladium from the environmental samples. Ion exchange methods are mostly based on the separation of the anionic chlorocomplexes of platinum metals from the matrix elements, which exist in the cationic forms in diliute acid solutions.

At present, preconcentration procedures employ various types of commercially availiable ion exchangers. Well known and selectively retained PGMs are chelating: –SH compounds, such as polyorganics, polyuretane resins and thiosemicarbazides [3-4]. Chelating ion exchangers are polymers with the functional groups able to form complexes with


197

selected ions. This type of resins combines ion exchange and complexing reactions, and hence can exhibit high selectivity for some ions or groups of ions.

The aim of this work was preconcentration of palladium(II) chlorocomplexes onto Chelite S containing thiol groups from the 0.1 – 2.0 M HCl – 1.0 M NaCl – 100 ppm Pd(II); 0.1 – 2.0 M HCl – 2.0 M NaCl – 100 ppm Pd(II) solutions using static methods and kinetic study of the preconcentration process.

1. EXPERIMENTAL

1.1. REAGENTS

The aqueous Pd(II) stock solution was obtained from solid PdCl2 (>99.99%, POCh, Poland) and standarized HCl solutions in the amounts required for maintaining H+ concentration at 0.1 M. Working solutions of palladium(II) were prepared by dilution with HCl in order to obtain a desired value of H+ concentration. The other reagents were of analytical grade from POCh (Poland).

The physical and chemical properties of Chelite S produced by Serva Feinbiochemica GmbH & Co., Germany are presented in Tab. 1.

Table 1. Characteristics of the ion exchanger.

Type Functional groups

Ionic form as shipped Matrix Bead size

[mm] Operating pH range

Thermal stability

[K]

Cationic - SH, thiol Na+ Polystyrene –

divinylbenzene 0.3 – 0.8 1 – 13 333

2.2. METHODS

Chelite S was dried at room temperature and then 0.5 g of the cation exchanger was put into the conical flask and 50 cm3 of the solution were added. The flask was shaken using a 358S laboratory shaker (Elpin type, Poland) from 60 to 43200 s at 298 K. Then the raffinate was separated from the ion exchanger by filtration and the content of Pd(II) was determined using the spectrophotometric iodide method.

2.3. KINETIC STUDY

The pseudo second-order model can be linearized to at least four different types. The details of these different forms of the linearized pseudo second-order equations are shown in Tab. 2.


198

Table 2. Pseudo second-order equations and their linear forms [5-7].

Type Linear form Plot Parameters

Type – I tqqkq

t

eet

112

2

+= t/qt vs. t

Type – II eet qtqkq

11)

1(

12

2

+= 1/qt vs. 1/t

Type – III e

e

t

e

q

qk

q

qk

t

22

221 −= 1/t vs. 1/qt

Type – IV e

tee

t

q

qqkqk

t

q 222

2 −= qt/t vs. qt

qe - the amount of palladium

complexes sorbed at equilibrium

k2 - the equilibrium rate constant of the

pseudo second-order sorption

t – phases contact

time


The static method was used to calculated the amount of palladium(II) complexes preconcentration on Chelite S from the following mass balance-relationship:

WVCCq tot /)( −= (1)

where Co is the initial concentration of palladium(II) complexes in the aqueous phase, Ct is the concentration of palladium(II) complexes in solutions after time t, V is the volume of the solutions (dm3), W is the weight of the dry Chelite S used (g).The plots of qt versus time (t) are shown in Figs. 1-2.

0 20000 40000 60000t [s]

123456789

10

qt1.0 M NaCl0.1 M HCl - 1.0 M NaCl0.5 M HCl - 1.0 M NaCl1.0 M HCl - 1.0 M NaCl2.0 M HCl - 1.0 M NaCl

0 20000 40000 60000

t [s]

123456789

10

qt 2,0 M NaCl0,1 M HCl - 2,0 M NaCl0,5 M HCl - 2,0 M NaCl1,0 M HCl - 2,0 M NaCl2,0 M HCl - 2,0 M NaCl

Fig. 1. The amount of Pd(II)

complexes versus time for the 0.1 – 2.0 M HCl – 1.0 M NaCl solutions.

Fig. 2. The amount of Pd(II) complexes versus time for the 0.1 – 2.0 M HCl –

2.0 M NaCl solutions.


199

Table 3. Kinetic parameters for palladium(II) preconcentration onto Chelite S.

Type I Type II Systems

qe,exp qe k2 h r2 qe k2 h r2 1.0 M NaCl 9.99 10.04 0.03 3.14 0.999 9.30 0.04 3.18 0.953

0.1 M HCl–1.0 M NaCl 10.0 10.07 0.03 2.67 0.999 9.15 0.04 3.08 0.942 0.5 M HCl–1.0 M NaCl 9.97 10.05 0.03 2.58 0.999 10.4 0.02 1.97 0.992 1.0 M HCl–1.0 M NaCl 9.99 10.06 0.03 2.84 0.999 9.39 0.03 2.84 0.965 2.0 M HCl–1.0 M NaCl 9.99 10.04 0.04 4.45 0.999 9.78 0.04 3.44 0.971

2.0 M NaCl 9.99 10.06 0.03 3.21 0.999 9.18 0.04 3.50 0.922 0.1 M HCl–2.0 M NaCl 9.96 10.09 0.02 1.70 0.999 9.37 0.02 1.61 0.991 0.5 M HCl–2.0 M NaCl 9.99 10.07 0.02 1.86 0.999 8.82 0.03 2.14 0.953 1.0 M HCl–2.0 M NaCl 9.98 10.11 0.02 1.63 0.999 9.09 0.02 1.73 0.981 2.0 M HCl–2.0 M NaCl 9.99 10.06 0.03 3.04 0.999 9.21 0.04 3.29 0.940

Type III Type IV Systems qe.exp qe k2 h r2 qe k2 h r2



qe,exp - the experimental amount of palladium(II) complexes sorbed at equilibrium, k2 –the rate constant of the pseudo second-order sorption, h – the initial sorption rate, r2 – the correlation coefficient

In the present study, the correlation coefficient, r2 was used to determine the best fit kinetic expression (Tab. 3). The r2 values for types II, III and IV of the pseudo second-order equations achieve lower values than for pseudo second-order equation type I. This fact confirms the fact that sorption of palladium(II) complexes onto Chelite S with thiol groups is pseudo second-order, type I. Moreover, the calculated values of qe for pseudo second-order, type I are very close to qe, exp.

REFERENCES

[1] P.Liu, Z. Su, X. Wu, Q. Pu, J. Anal. At. Spectrom., 2001, 17, 125. [2] A. Limbeck, J. Randl, H. Puxbaum, J. Anal. At. Spectrom., 2003, 18, 161. [3] J. Polacovičová, J. Medved’, V. Streško, J. Kubová, A. Čelková, Anal. Chim. Acta, 1996, 320, 145. [4]. M. Iglesias, E. Anticó, V. Salvadó, Anal. Chim. Acta, 1999, 381, 61. [5] K.V. Kumar, J. Hazard. Mater.,2006, B137, 1538. [6] K.V. Kumar, S. Sivansesan, , J. Hazard. Mater., 2006, B134, 277. [7] Y.S. Ho, Water Res., 2006, 40, 119.


224

CHEMICAL MODIFICATION OF POLYMERS IN ORDER TO

IMPROVE THE COMPLEXATION ABILITY OF HEAVY METAL IONS

Laurent DAMBIES2), Agnieszka JAWORSKA1), GraŜyna

ZAKRZEWSKA- TRZNADEL1) 1) Intytut Chemii i Techniki Jądrowej, Dorodna 16, 03-195 Warszawa

2) Visiting researcher in the scope of Marie Curie Transfer of Knowledge Project AMERC (MTKD-CT-2004-509226)

Abstract

Toxic heavy metals in water are global problems that create growing threat to the environment. For the removal of heavy metals from aqueous solutions it is possible to apply polymer enhanced ultrafiltration (PEUF). Strongly acidic water soluble polymers prepared from poly(vinyl alcohol) 10,000 can be used in PAUF (Polymer assisted ultrafiltration) to remove with high capacity Co2+ ions between pH 3 and pH 6 with the same efficiency, The retention of cobalt in polymer enhanced ultrafiltration was higher when modified polymer - sulfonated PVA as a complexing agent was applied.

1. INTRODUCTION

Membrane processes can be efficient methods for removal of metal ions, as well as radioactive species like cobalt-60, which is usually present in liquid radioactive waste in form of small ions. When ultrafiltration membranes are applied for radioactive wastes processing, the metal ions have to be formerly bound with macromolecular compound to form complexes that can be retained by the membrane.

Polymer assisted ultrafiltration (PAUF), also described as PEUF (polymer enhanced ultrafiltration), LPR (liquid phase polymer retention) or Polymer Filtration (PF) is a relatively new process of separation for purifying water containing metal ions. PAUF combines the ion exchange or chelating properties of a functionalized water soluble polymer with the sieving power of an ultrafiltration membrane.

Among soluble polymers, the most popular is polyacrylic acid used in its sodium or hydrogen form which can bind numerous metal ions such as Zn2+, Ni2+, Mg2+. Polyethyleneimine (PEI) is also quite effective in removing metal ions such as Hg2+,Cd2+,Cu2+ and Zn2+, Co2+. Another popular polymer is polyallylamine for the removal of Cu2+, Co2+, Ni2+ [1-3].


225

A survey of the papers on PAUF indicates that for a large number of them, commercially available polymers are used instead of tailored polymers.

As far as the synthesis of water soluble polymers concerned, two approaches can be used which are the functionalization of an existing polymer or the synthesis of the polymer starting with the individual monomer(s) by radical polymerization.

In the present work the first method using poly(vinyl alcohol) and poly(ethylene glycol) to attach series of different ligands (sulfonic, phosphoric, EDTA) was chosen.

2. POLYMERS SYNTHESIS

2.1. SYNTHESIS AND CHARACTERIZATION OF SULFONATED POLYMERS

For PVA polymers synthesis procedure is adapted from patent 2,531,468 [4] and is carried in three steps: pretreatment of the PVA polymer (10,000, 50,000 and 100,000 MW), preparation of the sulfonating agent and the reaction itself. 4.4 g of PVA (100 mmoles) are transferred in a 100 mL round bottom flask equipped with a condenser. 25 mL of pyridine were added and the mixture was stirred with a magnetic stirrer for 2h at 80°C then cooled down in the ice bath. 25 mL of pyridine were transferred to a 500 mL Duran bottle fitted with a small glass funnel. A stirring bar is added in the bottle and placed in an ice bath. 8 mL of chlorosulfonic acid were measured in a graduated cylinder and added drop by drop through the funnel in the pyridine solution during 30 min using a disposable glass pipet.

Sulfonating agent was redissolved in 25 mL of pyridine by warming up the bottle to 60 °C and added quite quickly to the PVA-pyridine mixture kept in the ice bath by using a disposable glass pipet. Round bottom flask was placed in an oil bath at 80 °C and reaction allowed to proceed at this temperature for 3 to 4 hours.

Sulfonates based on PVA 10,000 and 50,000 were soluble in water at room temperature whereas sulfonates based on PVA 100,000 were soluble at higher temperatures. Sulfonated PVA 10,000 had a solid yellow color while the 100,000 form had a very pale yellow color.

Table 1. Characteristics of the sulfonated polymers

PVA molecular weight Acid capacity of the final

polymer solution in (mmol/mL)

Acid capacity (mmol/g dry polymer)

10,000 0.429 4.93 50,000 0.177 3.8 100,000 0.00624 3.60


226

2.2. PREPARATION AND CHARACTERIZATION OF PHOSPHORYLATED PVA

4 g of PVA 100,000 were added in a 100 mL round bottom flask along with 50 mL of H3PO4 (85 % in water). Reaction was carried out for 5 hours at 65 oC then continued overnight at room temperature. At the end of the reaction a gold color supernatant was obtained and gummy light brown polymer (water soluble) attached in the bottom of the flask. Polymer was redissolved in water and filtered through a membrane of 10,000 MW cut off. The polymer solution obtained was titrated with 0.1 M NaOH. It had acid capacity of 0.00979 mmol/mL corresponding to 0.82 mmol/g. Acid capacity was quite low which might be related to the high molecular weight of PVA or the experimental conditions that were not optimal for this reaction.

3. Co2+ REMOVAL BY PAUF

The next step was ultrafitration process coupled with complexation by water soluble polymers. Tab. 2 and Fig. 1 show results of the experiment.

The 50 mL-ultrafiltration cell Amicon was used for all experiments. In the 100 mL-volumetric flask 1 mL of cobalt stock (1 mg or 0.017 mmole) with a suitable volume of polymer solution to give 0.17 mmole of acid groups were added. Solution was completed to 100 mL with distilled water, mixed and transferred in a beaker with a magnetic stirrer. The pH was adjusted to the desired value with NaOH 0.1 M or HCl 0.1 M. Solution was agitated for at least 2 hours and pH was checked and adjusted if necessary. 50 mL of the solution were then added to the ultrafiltration cell fitted with 10,000 MW cut-off membrane (PES Millipore). The remaining solution was kept to analyse the initial cobalt concentration. Ultrafiltration cell was closed and the assembly was put under pressure of 3 bars to carry out the filtration.

Retention coefficient of the Co2+ ions was investigated between pH 3 and 6 for different polymers based on poly(vinyl alcohol). Commercially available poly(acrylic acids) 100,000 and 15,000 MW were chosen as benchmark for this study.

Table 2. Characteristics of acidic polymers used for PAUF

Polymers Acid capacity

(mmol/g) Molar ratio H+/cobalt

Mass ratio Polymer/cobalt

PVA-SO3H 10,000 4.93 10 34.5 PVA-SO3H 100,000 3.60 10 27.4

PAA 15,000 H+ 12.9* 10 13 PAA 100,000 H+ 13.2* 10 13

PVA-PO3H 100000 0.82 10 208

*Theoretical acid capacity to PAA= 1000/72= 13.9 mmol/g


227

Fig.1. Retention factor of Co2+ as function of pH

4. CONCLUSIONS

The retention of cobalt in polymer enhanced ultrafiltration was higher when modified polymers as complexing agents were applied.

Strongly acidic water soluble polymer prepared from poly(vinyl alcohol) 10,000 can be used in PAUF to remove with high capacity Co2+ between pH 3 and pH 6 with the same efficiency. Sulfonated PVA 10,000 performed very well with a rejection rate above 95% between pH 3 and 6.

REFERENCES

[1] I. Korus, M. Bodzek, K. Loska, Sep. Purif. Tech. 17 (1999) 111. [2] Han S-C.,Choo, K-H., Choi,S-J. a, Benjamin,M.M., J. Membr. Sci. 290 (2007) 55. [3] R-S. Juang, C-H. Chiou, J Membr. Sci. 177 (2000) 207. [4] D.D. Reynolds, W. O. Kenyon, Polyvinyl sulfonates and process for their preparation.

US patent 2,531,468. (1949)


230

THE UPTAKE OF SAMARIUM(III) AND Eu(III) BY THE

THERMOSENSITIVE SULFONATED COPOLYMER OF N-ISOPROPYLACRYLAMIDE AND

STYRENE

Arleta BUTEWICZ, Marta BURBAN, Andrzej W. TROCHIMCZUK

Faculty of Chemistry, Wrocław University of Technology, WybrzeŜe Wyspiańskiego 27, 50-370 Wrocław, Poland ([email protected])

It is well known that some polymers, containing hydrophobic monomers, like for example N-isopropylacrylamide (NIPAM), alkyl acrylates and methacrylates etc. belong to the class of temperature-sensitive gels. The former monomer is used most often and its polymers are soluble at room temperature and start to precipitate at elevated temparatures due to the increase in the strength of hydrophobic interactions. When small admixture of crosslinking monomer (ethylene glycol dimethacrylate, poly(ethylene glycols) dimethacrylate) is used in the polymerization mixture the resulting polymer becomes insoluble and can only swell and shrink in water at temperature below and above the LCST (lower critical solution temperature). This is due to a phase transition between the hydrated and dehydrated states of the polymer chain. LCST of crosslinked poly(NIPAM) can be effectively controlled by incorporating charged mers as well as hydrophobic mers. Charged residues typically increase the LCST, while hydrophobic residues lower the LCST. The addition of charged monomer (i.e. possesing ion-exchange ability) allows for the preparation of ion-exchangers, which are sensitive to the temperature changes. In such material the ion-exchange groups can be effectively brought together at elevated (above LCST) temperatures due to the shrinking of polymeric network. Such fenomenon allows the simultaneous interactions of these groups and multivalent ions, the interactions, which is seriously impaired at room temperature due to the increased distance between groups.

In the current work we decided to introduce the sulfonic ion-exchange groups by sulfonating the N-isopropylacrylamide/styrene copolymer crosslinked with ethylene glycol dimethacrylate. The choice of ion-exchange groups was done in order to avoid the presence of groups like phosphoric, tested by us in the previous work, which were exhibiting both ion-exchange and coordinating properties depending on the pH of the solution. It turned out that the enthalphy of metal ions uptake indicated


231

mixed mechanism of interactions and thus made impossible applications of such materials in the separation of lanthanides [1].

Poly(NIPAM-co-St) sulfonated gel, as a thermosensitive gel, shows the conformational change of polymer network from swelling to shrinking with increasing temperature. If the ion-exchange molecules are placed in a stimuli-sensitive polymer gel, the salt formation between metal ion and these molecules may be controlled by the conformational change of polymer network.

In our research the extraction of europium, samarium and monovalent ions from an aqueous solution to a thermosensitive polymer gel, Poly(NIPAM-co-St) sulfonated with chlorosulfonic acid, was studied. The extraction of cations was carried out at two temperatures – RT and 50oC that is both below and above VPTT (volume phase transition temperature) of the gel. It was observed that polymer had affinity to monovalent ions but is was independent on the temperature, whereas extraction behaviour of europium and samarium was affected strongly by the phase transition of the gel.

REFERENCES

[1] Danko, B., Trochimczuk A.W., Samczyński, Z., Hamerska-Dudra, A., Dybczyński, R.J., React. Funct. Polym., 2007, 67, 1651.


230

THE UPTAKE OF SAMARIUM(III) AND Eu(III) BY THE

THERMOSENSITIVE SULFONATED COPOLYMER OF N-ISOPROPYLACRYLAMIDE AND

STYRENE

Arleta BUTEWICZ, Marta BURBAN, Andrzej W. TROCHIMCZUK

Faculty of Chemistry, Wrocław University of Technology, WybrzeŜe Wyspiańskiego 27, 50-370 Wrocław, Poland ([email protected])

It is well known that some polymers, containing hydrophobic monomers, like for example N-isopropylacrylamide (NIPAM), alkyl acrylates and methacrylates etc. belong to the class of temperature-sensitive gels. The former monomer is used most often and its polymers are soluble at room temperature and start to precipitate at elevated temparatures due to the increase in the strength of hydrophobic interactions. When small admixture of crosslinking monomer (ethylene glycol dimethacrylate, poly(ethylene glycols) dimethacrylate) is used in the polymerization mixture the resulting polymer becomes insoluble and can only swell and shrink in water at temperature below and above the LCST (lower critical solution temperature). This is due to a phase transition between the hydrated and dehydrated states of the polymer chain. LCST of crosslinked poly(NIPAM) can be effectively controlled by incorporating charged mers as well as hydrophobic mers. Charged residues typically increase the LCST, while hydrophobic residues lower the LCST. The addition of charged monomer (i.e. possesing ion-exchange ability) allows for the preparation of ion-exchangers, which are sensitive to the temperature changes. In such material the ion-exchange groups can be effectively brought together at elevated (above LCST) temperatures due to the shrinking of polymeric network. Such fenomenon allows the simultaneous interactions of these groups and multivalent ions, the interactions, which is seriously impaired at room temperature due to the increased distance between groups.

In the current work we decided to introduce the sulfonic ion-exchange groups by sulfonating the N-isopropylacrylamide/styrene copolymer crosslinked with ethylene glycol dimethacrylate. The choice of ion-exchange groups was done in order to avoid the presence of groups like phosphoric, tested by us in the previous work, which were exhibiting both ion-exchange and coordinating properties depending on the pH of the solution. It turned out that the enthalphy of metal ions uptake indicated


231

mixed mechanism of interactions and thus made impossible applications of such materials in the separation of lanthanides [1].

Poly(NIPAM-co-St) sulfonated gel, as a thermosensitive gel, shows the conformational change of polymer network from swelling to shrinking with increasing temperature. If the ion-exchange molecules are placed in a stimuli-sensitive polymer gel, the salt formation between metal ion and these molecules may be controlled by the conformational change of polymer network.

In our research the extraction of europium, samarium and monovalent ions from an aqueous solution to a thermosensitive polymer gel, Poly(NIPAM-co-St) sulfonated with chlorosulfonic acid, was studied. The extraction of cations was carried out at two temperatures – RT and 50oC that is both below and above VPTT (volume phase transition temperature) of the gel. It was observed that polymer had affinity to monovalent ions but is was independent on the temperature, whereas extraction behaviour of europium and samarium was affected strongly by the phase transition of the gel.

REFERENCES

[1] Danko, B., Trochimczuk A.W., Samczyński, Z., Hamerska-Dudra, A., Dybczyński, R.J., React. Funct. Polym., 2007, 67, 1651.


235

SEPARATION OF Zn(II), Cd(II) AND Pb(II) WITH

PNP-LARIAT ETHERS

Cezary KOZŁOWSKI1), Jolanta KOZŁOWSKA1) and Władysław WALKOWIAK 2)

1) Institute of Chemistry and Environment Protection, Jan Dlugosz University of Czestochowa, 42-201 Czestochowa, Armii Krajowej 13, Poland

2) Chemical Metallurgy Division, Faculty of Chemistry, Wroclaw University of Technology, 50-370 Wroclaw, Wybrzeze Wyspianskiego 27, Poland

Abstract In the present work the PNP lariat ethers have been used as macrocyclic ion

carriers for separation of metal ions from dilute aqueous solutions by transport across polymer inclusion membranes. In the case of competitive transport of Zn(II), Cd(II), and Pb(II) ions through plasticized immobilized membranes the selectivity of process is controlled via interaction of lariat ether groups with metal ions. The lariat ethers with larger size of substituents shows preferential selectivity order: Pb(II) > Cd(II) > Zn(II).

1. INTRODUCTION

Macrocyclic compounds such as crown ethers, lariat ethers, azacrowns, cryptands, and calixarenes were applied for selective recognition of specific metal ions. Crown ethers as well as their derivatives such as lariat ethers containing nitrogen and oxygen as donating atoms form stable complexes with transition metal ions, and may by used as the ionic carriers. Recently, we reviewed the application of macrocycle compounds as the ion metal carriers [1]. A remarkable increase in the applications of the liquid membranes for separation processes is observed. Few papers have focused on the determination of the selectivity and efficiency of the carrier-mediated transport of metal ions through organic media into an aqueous receiving phase [2-4]. Recently polymer inclusion membranes, as the components of novel liquid membrane systems, have been reviewed [5]. We now report the results for competitive transport of Cd(II), Zn(II), and Pb(II) ions by the polymer inclusion membranes from dilute nitrate aqueous solutions with PNP-16-crown-6 derivatives.

2. EXPERIMENTAL

Reagent analytical grade inorganic chemicals, i.e. Zn(NO3)2, Cd(NO3)2, CH3COONa, and HCl were obtained from POCh (Gliwice,


236

Poland). Organic reagents, i.e. cellulose triacetate (CTA), o-nitrophenyl penthyl ether (ONPPE), and dichloromethane were also purchased from Fluka and used without further purification.

A solution of CTA, the ion carrier (1-4), and the plasticizer, o-nitrophenyl pentyl ether in dichloromethane was prepared. The structure of PNP-lariat ethers investigated in the present work is shown in Table 1. A portion of this solution was poured into a membrane mold comprised of a 9.0 cm glass ring attached to a glass plate with CTA-dichloromethane glue. The organic solvent was allowed to evaporate overnight and the resultant membrane was separated from the glass plate by immersion in cold water. The membrane was soaked in 1.0 M aqueous HCl for 12 hours and stored in water.

Table 1. Structures of investigated lariat PNP lariat ethers

PNP–laraiat ethers R No.

O

C(O)OC2H5

1

CH3-(CH2)7-NH- 2

O

CHN(CH2)7CH3

3 O

O

O

O O

N

P P

NP

RR

R R

O CH N(CH2)7CH3

4

The transport experiments were carried out in a permeation cell

described in our earlier paper [6]. From the slope of the straight line obtained when representing the

metal concentration in the stripping phase in function of time, the initial flux, J, can be calculated according to the following equation:

⋅

=dt

dc

S

V J (1)

where V is the volume of the aqueous stripping phase, S is the exposed surface area of the membrane and C is the concentration of metal ions at elapsed time.


237

Samples of 1 mL were periodically withdrawn from feed and stripping compartments and were analyzed for metal content by ICP-AES (Varian Liberty RL spectrometer).

2. RESULTS AND DISSCUSION

Ethers 1÷4 were used as the ion carriers for transport of Zn(II), Cd(II) and Pb(II) through PIM. The source phase was an equimolar mixture of metal ions in the 1.0⋅10-4M. In Table 2 the kinetic parameters, as well as selectivity order and selectivity coefficients of metal ions removed by PIM from containing the above lariat ethers are given.

Table 2. The kinetic parameters and selectivity coefficient of the Zn(II), Cd(II) and Pb(II))ions transport through PIM containing lariat ethers 31÷38. Source phase: 1.0⋅10-4M solution of metal ions (pH 5.0). Receiving phase: distilled water. PIM:

0.3M lariat ether and 4.0cm3 ONPPE for 1.0g CTA

Lariat ethers

Metal ions Flux

J, µmol/m2s Selectivity order and selectivity

coefficients

Zn(II) 0,100

Cd(II) 0,100 1

Pb(II) 0,018

Zn(II), Cd(II) > Pb(II) 5,5

Zn(II) 0,024

Cd(II) 0,011 2

Pb(II) 0,065

Pb(II) > Zn(II) > Cd(II) 2,7 5,9

Zn(II) 0,010

Cd(II) 0,012 3

Pb(II) 0,080

Pb(II) > Cd(II) > Zn(II) 6,6 8,0

Zn(II) 0,009

Cd(II) 0,015 4

Pb(II) 0,100

Pb(II) > Cd(II) > Zn(II) 6,6 11,1

As shown in Table 2, for lariat ethers the Zn(II) transport rate decreases in the order: 1> 2, 3, i.e. decreases with larger size of substituents


238

in the crown ring. The transport of Pb(II) ions with the use of 1 is slower than in the case of 3 and 4, bearing large substituents; the increase of the diameter of the crown cavity results in the higher efficiency of Pb(II) ions transport. The ethers 1-4 selectively removed Pb(II) ions, with only insignificant transport of other metal ions. The selectivity coefficients SPb/Cd obtained in the transport process through PIM containing ethers 3 and 4 have identical values equal to 6.6, since their substituents are isomeric.

Acknowledgements

Financial support of this work was provided by the Polish Science Foundation (grant 4T09C 03230).

REFERENCES

[1] W. Walkowiak, C.A. Kozlowski, Macrocycle carriers for separation of metal ions in liquid membrane processes, Membrane Science and Technology Conference of Visegrad Countries PERMEA 2007, September 2-6, 2007, Siofok, Hungary.

[2] L. Rainer, Fresenius J. Anal. Chem., 2000, 367, 103. [3] W. Walkowiak, R.A. Bartsch, C. Kozłowski, J. Gęga, W.A. Charewicz, B. Amiri-

Eliasi, J. Radioanal. Nuclear Chem., 2000, 246, 643. [4] Shamsipur M.G. Azimi, S.S. Madaeni, J. Membrane Sci., 165, 2000, 217. [5] L.D. Nghiem, P. Mornanea, I.D. Potter, J.M. Perera, R.W. Cattrall, S.D. Kolev, J.

Membrane Sci., 2006, 7, 281. [6] J. Kozlowska, C.A. Kozlowski, J.J. Koziol, Sep. Purif. Technol., 2007, 57, 26.


239

EFFECT OF DILUENTS AND TYPE OF POLYMERIC

MEMBRANE ON CARRIER MEDIATED TRANSPORT OF CHROMIUM(III) WITH D2EHPA

Katarzyna ROTUSKA and Wiesław APOSTOLUK

Chemical Metallurgy Division, Department of Chemistry, Wrocław University of Technology, WybrzeŜe Wyspiańskiego 23, 50-370 Wrocław

Abstract Chromium(III) mediated transport through a flat-sheet supported liquid

membrane has been studied using the di-(2-ethylhexyl) phosphoric acid as a carrier and Durapore and Celgard 2500 as a solid supports. The effect of diluents in the membrane phase on chromium(III) transport has been also presented.

1. EXPERIMENTAL

Chromium(III) feed solution was prepared from chromium(III) nitrate nonahydrate (Cr(NO3)3

.9H2O) and sodium perchlorate (NaClO4). The ionic strength of the initial feed solution containing 0.0048M of Cr(III) was equal to 1.0 M. pH of the feed solution was adjusted to 4.5 – 5 by means of 6M NaOH solution. As a receiving phase 1.5M H2SO4 solution was used.

Di-2-ethylhexylphosphoric acid was used as Cr(III) carrier in the membrane phase. D2EHPA was mixed in an appropriate ratio with organic diluents and 1-decanol as a modifier. The following diluents were tested: kerosene, benzene, dichloromethane, trichloromethane, tetrachloromethane and n-heptane, respectively. Two hydrophobic polymeric supports, namely polyvinylidene fluoride membrane (Durapore) of nominal porosity 75%, 125µm thick and of 0.22µm pore size as well as polypropylene membrane Celgard 2500 (Celgard) of nominal porosity 55%, 25µm thick and of 0.22µm pore size, were used. The SLM membranes were prepared from the indicated polymer supports impregnated for about half an hour in 1M carrier organic solution containing 5% (v/v) of 1-decanol.

All experiments were performed at room temperature applying a two compartment cell with SLM sheet assembled between the feed and receiving phases. Both, feed and receiving phases were mixed mechanically at 800 rpm. Chromium(III) concentration in feed and receiving phases was determined spectrophotometrically by means of diphenylcarbizide method [1].


240


At a constant elapsed time of 24 h the effectiveness of chromium(III) recovery from the feed solution in the systems studied is compared in Table 1. The obtained results prove that in each case the transport of Cr(III) applying Durapore support (PVDF) is worse than that for Celgard 2500 (PP), in spite of lower porosity of the latter.

Table 1. Cr(III) percent recoveries from feed phase, at different polymeric support impregnated in a 1M organic solution of D2EHPA containing 5% (v/v) of

1 - decanol

R, %

Diluent PVDF PP

trichloromethane 47,8 49,6

benzene 29,2 80,5

n-heptane 36,3 85,9

kerosene 80,4 95,1

The greatest impacts on membrane permeability have pore sizes and tortuosity. Both compared materials have the same pore sizes. Hence, the reason of disagreement can be found in their different tortuosity and/or in thickness. Note that the support PVDF is five times thicker than PP membrane. Fig. 1 shows cross-section of PVDF and PP membranes, respectively.

A

B

Fig.1. Cross-sections of PVDF (A) and PP (B) membranes


241

It is also evident that the observed difference between both materials in their behavior toward Cr(III) transport depend on the type of diluents. This conclusion is in good accordance with the results of Molinari and coworkers [2] who studied the effect of diluents on Cr(III) transport with dinonylnaphtalene sulfonic acid (DNNSA) in supported liquid membrane systems. Based on the results from Table 1 Celgard 2500 membrane was chosen as a support for further experiments.

As it comes from literature [2, 3] the properties of diluents can affect the parameters of the transport process and the membrane stability. From our results presented in Fig. 2 it is clear that the flux of Cr(III) depends on the nature of the diluents and decreases in the following order: kerosene > n-heptane > CCl4 ~ benzene > chloroform > dichloromethane.

Fig. 3 demonstrates that in the same order changes the yield of Cr(III) recovery from the feed solution.

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

J, g

/cm

2 h

dic

hlo

rom

eth

ane

tric

hlo

rom

eth

ane

tetr

ach

loro

met

han

e

ben

zen

e

kero

sen

e

n-h

epta

ne

Fig. 2. Cr(III) flux vs. different diluents in the organic phase The experimental fluxes of Cr(III) transport (Fig. 2) can be described

by the following dependence:

J = -0 .86 π* + 0.0023 (δH)2 (1)

R2 = 0.9802; SD = 0.08, F = 99.1, N = 6,

where Kamlet and Taft parameter π* denotes the dipolarity/polarizability of organic diluents while (δH)2 stands for their cohesive energy density [4]. This dependence fairly well confirms the sequence of diluents reported


242

above and is in good agreement with the results of Islam and Biswas obtained for chromium(III) extraction with D2EHPA solutions in different organic solvents [5].

0

10

20

30

40

50

60

70

80

90

100

R, %

dic

hlo

rom

eth

ane

tric

hlo

rom

eth

ane

tetr

ach

loro

met

han

e

ben

zen

e

kero

sen

e

n-h

epta

ne

Fig. 3 Percent recoveries of Cr(III) from the feed phase at a different composition of organic membrane

3. CONCLUSIONS

1. The material of polymeric matrix in supported liquid membrane has a great impact on chromium(III) recovery. Decrease of polymeric support’s thickness and tortuosity can favor permeability and faster transport of Cr(III) cationic species through the supported liquid membrane.

2. The nature and properties of diluents in organic phase affect the transport rate and recovery of chromium(III). The effect of aliphatic hydrocarbons on chromium(III) transport in the studied systems prevails over the effects of benzene and chlorinated solvents, respectively.

REFERENCES

[1] Z. Marczenko, Separation and Spectrophotometric Determination of Elements, Ellis Horwood, Chichester, 1986.

[2] R. Molinari, E. Drioli, G. Pantano, Sep. Sci. Technol., 1989, 24, 1015. [3] F.J. Alguacil, M. Alonso, Hydrometallurgy, 2004, 74, 195. [4] Y. Marcus, Chem. Soc. Rev., 1993, 22, 409. [5] F. Islam, R.K. Biswas, J. Inorg. Nucl. Chem., 1979, 41, 229.


247

TRANSPORT AND SEPARATION OF Zn(II) AND Cu(II)

IN AN AGITATED BULK LIQUID MEMBRANE

Piotr SZCZEPAŃSKI, Aneta RASZKOWSKA

Nicolaus Copernicus University, Faculty of Chemistry Gagarina Str. 7, 87-100 Toruń, Poland, e-mail: [email protected]

Abstract

Pertraction of Zn(II) and Cu(II) ions from binary solutions across a bulk liquid membrane (BLM) containing di(2-ethylhexyl)phosphoric acid (D2EHPA) as the carrier was investigated. The influence of the pH, D2EHPA concentration, feed and stripping phase concentration was evaluated. The enhancement of the separation effects in unsteady state pertraction of Zn(II) was observed.

1. INTRODUCTION

The application of liquid membranes is an important alternative to the standard processes of wastewater treatment or separation and recovery of metal cations. LM is composed of an organic liquid phase placed between two aqueous phases. Concentration of the feed and stripping solution, carrier concentration and the type of organic solvent are the most important factors influencing the pertraction process.

Organophosphorous compound, such as D2EHPA, have been widely used as the extraction agent (carrier) in transport and separation of a number of metals such as Zn(II), Cu(II), Co(II), Ni(II), Cd(II), etc. [1-3]. The separation of Zn(II)/Cu(II) cations has been investigated by many researchers because of great importance to hydrometallurgy and environmental protection [3-7].

The aim of this study was to investigate the influence of the carrier (D2EHPA) concentration, feed and stripping concentration and pH of the feed phase on transport and separation of Zn2+ and Cu2+ cations in BLM with D2EHPA as the carrier in non-stationary conditions.

2. EXPERIMENTAL

The experiments were performed in a beaker-in-beaker type pertractor at 25oC (Fig. 1). The experimental device consisted of two concentric beakers, containing the feed solution (external beaker) and stripping solution (internal beaker). Volumes of the feed (solution of Zn2+ and Cu2+ nitrates), organic (D2EHPA in kerosene) and stripping phases (sulfuric acid


248

solution) were 125, 30, 25 cm3, respectively. The membrane contacting area was 16.45 cm2 (f/LM interface) and 5.8 cm2 (LM/s interface). The solutions were agitated with a glass stirrer (LM) at 375 rpm and magnetic stirrer (the feed and stripping solution) at 150 rpm. The concentrations of metals in the feed and stripping solutions were determined by a Varian 20-ABQ absorption spectrophotometer.

Fig. 1. Scheme of experimental bulk liquid membrane (BLM), f - feed solution, LM–liquid membrane, s–stripping solution, MS–magnetic stirrer

3. CALCULATIONS

Under the experimental conditions studied, non-stationary pertraction of Zn(II) was observed. Therefore, nonlinear equation was applied for fitting the experimental results:

( )( ) 1)(

−+⋅+⋅= dd tbtcbatc (1)

The flux of Zn(II), J(t) was then calculated by differentiating the curve-fit functions, Eq. (1):

dt

)t(dc

A

V)t(J ⋅= (2)

where V denotes volume of the stripping solution, A – the interphase contact area. The plot of J(t) vs. t corresponding to Zn(II) pertraction exhibits a maximum characterized by Jmax. The stripping phase concentrations of Cu(II) ([M]s,Cu) was used for calculating the amount of species transported effectively from the feed to the stripping phase after the time t:

s

s

A

VQ

⋅⋅

=1000

[M] Cus, [mol/cm2] (3)

LM

MS

f

s

LM


249

where Vs denotes the volume of the strip solution, As denotes the output membrane area (cm2). In the case of a linear relationship Q vs. t, the transport was considered as quasi-stationary, and respective fluxes were calculated as:

Cu Q/J t= ∆ ∆ [mol/cm2s] (4)

4. EXPERIMENTAL RESULTS

The effect of the feed phase concentration (0.005 M-0.1 M) on the pertraction of Zn(II) and Cu(II) cations is presented in Fig. 2. Under these experimental conditions it was observed that the fluxes of Zn(II) are higher than the fluxes of Cu(II). An increase in the feed concentration causes an increase of Zn(II) fluxes and decrease in Cu(II) fluxes. This effect was interpreted elsewhere by Juang [4] in this way that decreasing of JCu at higher feed concentrations region is caused by much higher distribution ratio of Zn(II) than Cu(II) catons (at the F|LM interface). Also, for the extraction of Cu(II) in the system with D2EHPA a decrease of distribution coefficient of Cu(II) with increasing initial concentration was observed [8].

Feed concentration [mol/dm3]

0.00 0.02 0.04 0.06 0.08 0.10 0.12

Flu

x [m

ol/c

m2 s]

1e-9

2e-9

3e-9

4e-9

5e-9

6e-9

5.0e-12

1.0e-11

1.5e-11

2.0e-11

2.5e-11

Fig. 2. Dependence of Zn2+ and Cu2+ fluxes on the feed phase concentration in BLM with 0.1 M D2EHPA as a carrier

Results in Fig.3 indicate that the flux of Zn(II) and Cu(II) increases

with an increase of pH of the feed phase. In the case of low pH of the feed solution, the pertraction is governed by diffusion of complex in the membrane, and to some degree by an equilibrium distribution of transported species at the feed/membrane interface [9].


250

pHf

1 2 3 4 5 6

Flu

x [m

ol/c

m2 s]

0

1e-9

2e-9

3e-9

4e-9

5e-9

6e-9

0

5e-12

1e-11

2e-11

2e-11

2e-11

3e-11

Fig. 3. Effect of pH in the feed phase on the fluxes of Zn2+ and Cu2+

in BLM with 0.1 M D2EHPA as a carrier The influence of H2SO4 concentration in the stripping phase was

studied at the concentration range from 0.05 to 1 M. Under the conditions, an increase of JZn with increase of acid concentration in the stripping solution was observed. This effect results from an increase of the driving force for the pertraction.

Stripping concentration [mol/dm3]

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Flu

x [m

ol/c

m2 s

]

3.0e-9

3.5e-9

4.0e-9

4.5e-9

5.0e-9

5.5e-9

5.0e-12

1.0e-11

1.5e-11

2.0e-11

Fig. 4. Fluxes of Zn2+ and Cu2+ in BLM system with D2EHPA as carrier, at different H2SO4 concentration in the stripping phase

Fig. 5 shows the effect of the feed D2EHPA concentration on JZn and

JCu. The flux of Zn(II) and Cu(II) increases with increasing in the carrier concentration. Moreover, the fluxes reach a plateau in the region of which the pertraction process is controlled by diffusion of metal ions in the feed solution [4]. The transport properties of the liquid membrane itself become


251

more important in the case of lower D2EHPA concentrations. The largest difference between JZn and JCu was observed for 0.01 M D2EHPA.

Carrier concentration [mol/dm3]

0.0 0.1 0.2 0.3 0.4 0.5

Flu

x [m

ol/c

m2 s]

0

2e-9

4e-9

6e-9

8e-9

1e-8

0.0

5.0e-11

1.0e-10

1.5e-10

2.0e-10

Fig. 5. Dependence of Zn2+ and Cu2+ fluxes on the carrier concentration in BLM with 0.1 M D2EHPA as a carrier

The extraction of Zn2+ and Cu2+ from aqueous solution with D2EHPA

can be expressed as follows [4]: M2+ + [(2+n)/2]

2(HR) ↔ n(HR)MR 2 +2H+ (4)

where n=1 for Zn2+ and 2 for Cu2+. Overbars refers to the organic phase. The selectivity of Zn2+ over Cu2+ in extraction can be defined as [5]:

β1=DZn/DCu=2.14⋅103 2(HR)

-1/2 (5)

where D denotes a distribution ratio. A change of the monomeric D2EHPA concentration from 0.01 - 0.5 M lowers the value of β1 decreases from 957 to 135. These values are much higher than the ratio of JZn/JCu in the steady state pertraction through a supported liquid membrane [4]. Higher values of the ratio of JZn/JCu for concentration of D2EHPA below 0.25 M, were observed in the case of studies presented herein (non-stationary conditions for Zn(II) transport).

5. CONCLUSIONS

The influence of various experimental parameters on the transport and separation of Zn(II) and Cu(II) was evaluated. Zn(II) can be effectively transported and separated through a bulk liquid membrane containing D2EHPA as the carrier. The increase of the carrier concentration in the liquid membrane phase, counter-ion concentration in the stripping phase, concentration and pH of the feed solution led to an increase in flux of Zn(II). Under the conditions, it was found that a higher selectivity in the


252

separation of Zn(II) over Cu(II) can be obtained for high feed concentration (and pH ≈ 5), stripping phase concentration of 0.25 M and low D2EHPA concentration. Enhancement of the separation effects was observed for unsteady state transport of Zn(II) in BLM,

REFERENCES

[1] M. Resina, J. Macanas, J. de Gyves, M Munoz, J. Membr. Sci., 2006, 268, 57. [2] R.-S. Juang, J. Membr Sci., 1993, 85, 157. [3] R.-S. Juang, J.-D. Jiang, Sep. Sci. Technol., 1994, 29(2), 223. [4] R.-S. Juang, Ind. Eng. Chem. Res., 1993, 32, 911. [5] R.-S. Juang, H.-Ch. Huang, J. Membr Sci., 1999, 156, 179. [6] R.-S. Juang, H.-L. Huang, J. Membr. Sci., 2002, 208, 31. [7] R.-S. Juang, H.-L. Huang, J. Membr. Sci., 2003, 213, 125. [8] S. Raharimalala, E. Nakache, G. Cote, J. Chim. Phys., 1995, 92, 1286. [9] P. Plucinski, W. Nitsch, J. Membr. Sci., 1988, 39, 43.


258

TRANSPORTATION OF ZINC(II) IN BULK LIQUID

MEMBRANES CONTAINING PHOSPHONIUM IONIC LIQUID

Łukasz NOWAK, Magdalena REGEL-ROSOCKA, Maciej WIŚNIEWSKI


Abstract Investigations on trihexyl(tetradecyl)phosphonium chloride as selective

carrier to remove zinc(II) from HCl solution in bulk liquid membrane system were carried out. The use of trihexyl(tetradecyl)phosphonium chloride as a selective carrier of zinc(II) in the membrane process is disadvantageous. Zinc(II) and the carrier form strong bounded ionic pair that makes zinc stripping very difficult. And the obtained zinc(II) concentration in the stripping phase is small, what disqualifies the use of the system on a larger scale.

1. INTRODUCTION

The undertaken investigation of ionic liquids as selective zinc ion carriers in the extraction process proved to be very promising and even a small addition of these compounds causes very effective recovery of that metal [1].

The extraction process in liquid-liquid system is connected with necessity of in- between operations, e.g: scrubbing of the organic phase before the stripping. It causes energetic outlay expansion and the possibility of environment pollution.

The use of systems with liquid membranes (LMs) is more effective method than the classic extraction. LMs offer a possibility of applying new and more effective extractants. Moreover, using supported or emulsion membranes the ratio between active substance and feed is very low.

The process course investigation at a supported (SLM) or emulsion (ELM) membrane is difficult, e.g: problem with analysis of stripping phase in case of ELM [2]. It is the aim of the work to investigate the application of trihexyl(tetradecyl)phosphonium chloride (classified as ionic liquid) for removal of zinc(II) from HCl solution in bulk liquid membrane system.


259

2. EXPERIMENTAL

An initial zinc(II) solution with the following composition: 5 g/dm3 (0.077 M) Zn(II), 1.8% (0.58 M) HCl, 5 M Cl- (adjusted with NaCl), was submitted to a membrane process. 0.2 M trihexyl(tetradecyl)phosphonium chloride solution in toluene was used as a membrane solution [3]. Fig.1 represents the structure of this compound. Solutions of 1 M Na2SO4 and H2SO4 were used as a stripping solution.

P

C5H11

C5H11C13H27

C5H11

Cl-

Fig. 1. The structure of trihexyl(tetradecyl)phosphonium chloride.

Investigations were made in double Lewis cell presented in Fig. 2. The device is built of two cylinders connected with bridge. All three phases of the system were stirred with minimal speed to avoid disturbances.

Fig. 2. The scheme of double Lewis cell.

The process was carried out for 24 hours. During the process samples of feed and stripping phase (2 cm3) were collected every 30 minutes and after that both phases were replenished with equivalent amount of initial solutions. Zinc ions were indicated complexometrically with EDTA in the presence of eriochrom black T as an indicator.


260

3. RESULTS

The achieved results are used to create a profile graph of the change of zinc(II) concentration in individual phases of the system during the membrane process (Figs. 3 and 4).

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0 5 10 15 20 25

time [h]

conc

entr

atio

n Z

n(II

) [M

]

Fig. 1. Concentration of zinc(II) in bulk liquid membrane process when 1 M H2SO4

was used as a strip phase (● - feed, ▲ – membrane, ■ – strip).

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0 5 10 15 20 25

time [h]

conc

entr

atio

n Z

n(II

) [M

]

Fig. 2. Concentration of zinc(II) in bulk liquid membrane process when 1M

Na2SO4 was used as a strip phase (● - feed, ▲ – membrane, ■ – strip).

During first hours of the process there is a sharp decrease in zinc(II) ion concentration in the feed due to fast loading of the organic phase. The transport of zinc(II) ions to the stripping solution takes place after the loading of the organic phase. After 10 hours no significant change in zinc concentration is observed. The most visible changes take place during first 5-7 hours, when the biggest decrease in zinc(II) ion concentration in feed


261

and the highest increase in the stripping phase appear. During the next 24 hours the system is slowly approaching to equilibrium state. The stripping does not depend on the stripping solution used ( Na2SO4 or H2SO4).

The extraction process in bulk liquid membrane system with trihexyl(tetradecyl)phosphonium chloride as a carrier takes place quickly and soon after a few hours equilibrium state is achieved. The mechanism of zinc ion transport with the studied carrier is probably similar to base extractants and can be described with ion-exchange equation as fallows:

ZnCl42-

(w) + 2H+(w) + 2[R3R

’P+][Cl -](o) ↔

[(R3R’P+)2][ZnCl4

2-](o) + 2H+(w) + 2Cl-(w)

The stripping from the membrane to the stripping phase is slow and limits the rate of the whole process. The concentration of the zinc(II) ions in the stripping phase is constant and it does not depend on the concentration in the feed.

4. CONCLUSIONS

Resuming all achieved results we can state that the use of trihexyl(tetradecyl)phosphonium chloride as a selective carrier of zinc(II) in membrane process is disadvantageous. Zinc(II) and the carrier form strong bounded ionic pair that makes zinc stripping very difficult. And the obtained zinc(II) concentration in the stripping phase is small, what disqualifies the use of the system on a larger scale.

Acknowledgement

The work was supported by the grant No. 32-139/08-DS.

REFERENCES

[1] M. Kosmulski, B. Tendaj, Przem. Chem., 2002, 81, 106. [2] W. Cichy, S. Schlosser, J. Szymanowski, J. Chem. Technol. Biotechnol., 2005, 80, 189. [3] M. Regel-Rosocka, K. Cieszyńska, M. Wiśniewski, Przem. Chem., 2006, 85, 651.


262

RECOVERY OF Cu(II) AND Co(II) FROM CHLORIDE SOLUTIONS BY POLYMER INCLUSION MEMBRANES

WITH ALIQUAT 336

Beata POSPIECH1), Władysław WALKOWIAK2)

1) Department of Chemistry, Technical University of Czestochowa, 42-200 Częstochowa, Armii Krajowej 19, Poland, e-mail: [email protected]

2) Division of Chemical Metallurgy, Wroclaw University of Technology, 50-370 Wrocław, WybrzeŜe Wyspiańskiego 27, Poland

Abstract In this work the transport of copper(II) and cobalt(II) from aqueous chloride

solutions through polymer inclusion membranes (PIMs) is presented. Aliquat 336 has been applied as the ion carrier. The source phase was the aqueous HCl solution containing Cu(II), Co(II) and Ni(II). The selective transport through PIM containing cellulose triacatate (CTA) as the support, o-nitrophenyl pentyl ether (ONPPE) as the plasticizer and Aliquat 336 as the ion carrier has been studied. Cu(II) and Co(II) ions were effectively removed from the source phase by transport through PIM into 0.5 M CH3COONH4 as the receiving phase.

1. INTRODUCTION

Copper and cobalt are the important metals with several industrial applications. In the separation of these metal ions the transport through liquid membranes plays an important role. Copper(II) and cobalt(II) can be removed from chloride solutions using various reagents, such as: the tertiary amines, i.e. tri-n-octylamine (TOA) and triisooctylamine (TIOA) [1-2]. Aliquat 336 (tri-n-octylmethylammonium chloride) as a carrier species has been reported by several authors [3-4]

The separation of copper(II) from aqueous solutions by PIM containing Aliquat 336 as the specific metal ion carrier has been carried out by Wang and Shen [3]. They have investigated the stability of Aliquat336/PVC based polymer liquid membrane. The results show the membrane stability was likely to be dependent upon the metal species extracted. The membrane showed an excellent stability in the extraction of Cu(II), but a poor stability in the extraction of Cd(II).

Argiropoulos et al. [4] studied the transport of copper(II) and gold(III) ions from hydrochloric acid solutions by polymer membrane consisting of


263

Aliquat 336 immobilized in poly(vinyl chloride) (PVC). The data in their work show that the membrane provides a promising method for the selective separation of gold from chloride solutions. The membrane was able to transport Au(III) in the presence of high concentrations of Cu(II).

Transport of Cu(II) and Ag(I) through a plasticized cellulose triacetate membrane (CTA) containing macrocylic polyethers as the carrier and 2-nitro phenyl octyl ether (ONPOE) as the plasticizer has been investigated by Arous et al.. [5].

Kozlowski et al. investigated the transport of Cu(II), Co(II), Ni(II) and Zn(II) from aqueous solutions into distilled water through plasticized immobilized membranes. The hydrophobic β-cyclodextrin (β-CD) polymers have been used as the macrocyclic ion carriers. With the use of β-CD polymer as an ionic carrier in the competitive transport of metal ions the preferential selectivity order: Cu(II)>Co(II)>Ni(II)>Zn(II) was observed [6].

In this work the selectivity of Cu(II) and Co(II) extraction from acidic chloride solution with Aliquat 336 has been studied. The transport of copper(II) and cobalt(II) through PIMs containing CTA as a support, o-nitrophenyl pentyl ether (ONPPE) as a plasticizer, and Aliquat 336 as the ion carrier is performed.

The kinetics of PIM transport was described by a first order reaction rate:

ln ktc

c

i

−=

(1)

were c is the metal ion concentration (M) in the source phase at some given time, ci is the initial metal ion concentration in the source phase, k is the rate constant (s-1), and t is the time of transport (s).

To calculate the k value, a plot of ln(c/ci) versus time was prepared. The rate constant value for the duplicate transport experiment was averaged and standard deviation was calculated. The initial input flux (Ji) was determined as equal to:

ii ckA

VJ ⋅⋅= (2)

where V is volume of the aqueous source phase, and A is an area of effective membrane.


264


The competitive transport of Cu(II) and Co(II) from chloride solutions through PIM containing Aliquat 336 as the ionic carrier into 0.5 M CH3COONH4 as the receiving phase was investigated. The source phase was composed of 0.01 M Cu(II), 0.002 M Co(II), 0.02 M Ni(II), 1.0 M HCl and 4.0 M NaCl. The initial fluxes and recovery factors for competitive transport of Cu(II) and Co(II) are presented in Table 1. Ni(II) was not detected in the receiving phase. The obtained results are presented in Fig. 1. The fluxes of Cu(II) and Co(II) were obtained by using PIM containing Aliquat 336 as the ion carrier were equal to 5.59 and 2.48 µmolm-2s-1, respectively.

0 5 10 15 20 25 300.0

0.2

0.4

0.6

0.8

1.0

Cu(II) Co(II)

c/c o

time, h Fig. 1. Kinetics of Cu(II) and Co(II) transport through PIMs containing 2.5 M

Aliquat 336, 1.28 cm3 ONPPE/1 g CTA. Source phase: 0.01 M Cu(II), 0.002 M Co(II); receiving phase: 0.5 M CH3COONH4.

Wang et al. [7] has also studied the transport of Cu(II) from 2.0 M HCl

through liquid membrane containing Aliquat 336 as the ion carrier. In the result of investigation, the following mechanism of reactions with chloride complexes of copper(II) and Aliquat 336 was proposed:

AClm + [CuCl3]-(aq) ⇄ A[CuCl3]m + Cl-(aq) (1)

2 AClm + [CuCl4]2-

(aq) ⇄ A2[CuCl4]m + 2 Cl-(aq) (2)

where: A – Aliquat 336.


265

Table 1. Kinetic parameters for Cu(II) and Co(II) transport through PIM containing 2.5 M Aliquat 336, 1.28 cm3 ONPPE/1 g CTA.

Metal ions Permeability coefficient, P

(µms-1)

Initial flux, Jo (µmolm-2s-1)

Recovery factor, RF (%)

Cu(II) 0.56 5.59 63.47

Co(II) 1.24 2.48 86.45

As it can be seen in Tab. 1 the permeability coefficient and the initial flux of Cu(II) are higher in comparison with the initial flux of Co(II). The metal ions fluxes decrease in the order: Cu(II) > Co(II) > Ni(II).

3. CONCLUSION

The presented investigation shows that copper(II) and cobalt(II) ions can be effectively recovered from aqueous chloride solutions containing Ni(II) ions in the transport process using Aliquat 336 as the ion carrier. The results of this work suggest that Aliquat 336 is a very good ion carrier of Cu(II) and Co(II). Results of the membrane stability study will be described in the future work.

REFERENCES [1] B. Pospiech, W. Walkowiak, Sep. Purif. Technol., 2007, 57,461. [2] J. Marchese, M. Campderros, A. Acosta, J. Chem. Tech. Biotechnol., 1995, 64, 293. [3] L. Wang, W. Shen, Sep. Purif. Technol., 2005, 46, 51. [4] G. Argiropoulos, R.W. Cattrall, I.C. Hamilton, S.D. Kolev, R. Paimin, J. Membr. Sci.,

1998, 138, 279. [5] O. Arous, H. Kerdjoudj, P. Seta, J. Membr. Sci., 2004, 241, 177. [6] C.A. Kozlowski, T. Girek, W. Walkowiak, J. Koziol, Sep. Purif. Technol., 2005, 46, 136. [7] L. Wang, R. Paimin, R. Cattrall, W. Shen, S.D. Kolev, Proceedings of ISEC’ 1999, 1053.


266

STUDY ON APPLICATION OF ALKYL THIOBENZOIC

COMPOUNDS FOR PERTRACTION OF CATIONS IN LIQUID MEMBRANE SYSTEMS

Andrzej OBERTA1), Janusz WASILEWSKI2),

Marek ŚWIĄTKOWSKI1), Romuald WÓDZKI1) 1) Nicolaus Copernicus University, Faculty of Chemistry, Gagarin Str. 7,

87-100 Toruń, Poland 2) University of Warmia and Mazury, Department of Biochemistry, Faculty of

Biology, Oczapowski Str. 1A, 10-719 Olsztyn, Poland [email protected]

Abstract

Pertraction of metal ions, especially Pb(II), through a multimembrane hybrid system (MHS) with a liquid membrane consisting of 1,2-dichloroethane and 2-(octylthio)benzoic acid as an ion carrier, was investigated. The liquid membrane phase was dehydrated in time of the process and the amount of water permeated through pervaporation membrane was determined. The following order of selectivity was observed: Pb(II)>>Ca(II)>K(I)>Mg(II)>Na(I).

1. INTRODUCTION

Sulphur-containing compounds were reported as ion carriers of various heavy metals. For instance Lamb et al. described macrocyclic crown thioethers as the Pb(II) carriers [1]. Several other authors also reported on new selective ionophores for the removal of heavy metal ions of similar properties [2-4]. Sulphur-containing carboxylic acids are known to be the selective extractants of Ag(I) and alkali earth metals [5,6]. These properties can be related to the complexation ability of thio-derivatives of carboxylic acids towards different cations [7-9].

In this paper, the application of one of the new selective ionophores from S-alkyl acetic and benzoic acid derivatives is described for Pb(II) separation in the multimembrane hybrid system (cation exchange membrane|flowing liquid membrane|cation exchange membrane) [10]. The transport and selectivity properties of the first compound studied, i.e. 2-(octadecylsulphanyl)acetic acid, were reported recently [11] and the properties of 2-(octylthio)-benzoic acid are presented herein.


267

2. EXPERIMENTAL

2.1. CHEMICALS

1,2-dichloroethane (DCE, Riedel-de-Haën, Germany, extra pure, 65 cm3) was used as the organic phase. 2-(octylthio)benzoic acid (MSD-10) (Fig.1) was used as an ion carrier. The feed solution (0,01M Pb(II), Ca(II), Mg(II), K(I), Na(I)) was prepared by dissolving appropriate amounts of metal nitrates (POCh Gliwice, Poland) in twice distilled water. The stripping solution (1M HNO3, 60 cm3) was prepared by diluting 65% HNO3 (POCh, Gliwice, Poland). All the chemicals were used as received, without further purification.

Fig. 1. 2-(octylthio)benzoic acid (MSD-10)

2.2. SYNTHESIS OF 2-(OCTYLSULPHANYL)BENZOIC ACID

A mixture of 30.8 g (0.2 mol) of thiosalicylic acid, 23.6 g (0.42 mol) of potassium hydroxide and 38.6 g (0.2 mol) of n-octyl bromide was refluxed in 500 cm3 of absolute ethanol under nitrogen atmosphere for 5 hours. The mixture was then allowed to cool to room temperature, diluted with 500 cm3 of water and acidified with hydrochloric acid. The white precipitate formed was filtered, washed first with cold ethanol and several times with water, and air dried. After crystallization from aqueous ethanol, 39.6 g of the desired product (74.3% yield) was obtained. For physicochemical measurements, the product was additionally twice recrystallized from aqueous ethanol and dried in vacuum for 5 hours at room temperature. M.p.: 79.5-80 oC (lit. [12] m.p. 80 oC)

1H-NMR (300 MHz, CDCl3, δ): 0.89 (t, 3H, CH3-), 1.30 [m, 8H, -(CH2)4-], 1.49 (m, 2H, -CH2CH2CH3), 1.75 (quint, 2H, -SCH2CH2-), 2.93 (t, 2H, -SCH2-), 7.18 (t, 1H, aromatic), 7.33 (d, 1H, aromatic), 7.48 (t, 1H, aromatic), 8.13 (d, 1H, aromatic), 12.15 (s broad, 1H, -COOH)

13C-NMR (300 MHZ, CDCl3, δ): 14.06 (CH3-), 22.61 (CH3CH2-), 28.11, 29.11, 29.17, 29.20, 31.77 [-(CH2)5-], 32.03 (-SCH2-), 123.61, 125.40, 126.08, 132.50, 133.07, 143.44 (aromatic), 171.96 (-COOH)


268

2.3. MULTIMEMBRANE HYBRID SYSTEM (MHS)

The multimembrane hybrid system (MHS) was constructed as shown in Fig. 2. The organic phase circulated between MHS, pervaporation module, and reservoir, and its circulation was forced by a Teflon made piston pump (with pumping velocity of 150 cm3/min). The feed and stripping solutions circulated between MHS and the reservoirs with pumping velocity of 20 cm3/min.

Additionally, a pervaporation module was used for removing water from the organic phase. The receiver of the pervaporation unit was connected to a glass-made vacuum system, operating at pressure of ~1 mm Hg. Water pervaporated into lower-pressure side was frozen in one of the two receivers using solid carbon dioxide as the cooling medium. The amount of frozen water was determined by weighting. The ion-exchange membrane – Nafion-117 (Du Pont, USA) – supported by a porous metal plate was used as a pervaporation membrane. The same Nafion-117 membrane was used as the separator between the organic and aqueous phases.

At the zero-time of the transport run, membranes were always in their hydrogen form. After each experiment, Nafion-117 membranes were converted into their hydrogen form by soaking for a few hours in the 1M nitric acid solution, and then washed with water in order to remove free electrolytes. For each transport experiment, fresh liquid membrane solution was prepared.

Fig. 2. Scheme of the MHS multimembrane hybrid system: (1) pertractor, (2) feeding phase reservoir, (3) stripping phase reservoir, (4) organic phase

reservoir, (5) pervaporation module, (6) water receivers, (7) cooling medium, (8) vacuum system, (9) ion exchange membranes


269

2.4. PERTRACTION EXPERIMENTS

The Nafion-117 membranes in their hydrogen forms were clamped in a triple-cell compartment, with the total membrane working area of 12.57 cm2. Each membrane was in contact with the organic phase and the feed or stripping phase, respectively. The metal concentration was determined (AAS, SpectrAA-20, Varian Inc.) as a function of time in both the feed and receiving aqueous solutions by taking 0.5 cm3 and 1 cm3 samples at periodic intervals from the feed and stripping phase, respectively.

3. RESULTS

Transport results, shown as cumulative curves vs. time on the stripping side, are shown in Fig. 3. Selectivity of MSD-10 is presented in Fig. 4 as the separation factors of all the investigated cations. Separation factors

( iMMΣα ) were calculated in accordance with Eq. (1), where zMi is the charge

of cation Mi, and [M]i,f and [M]i,s are the concentrations of cation Mi in the feed and strip phase, respectively.

fjijfi

sjijsi

ii

iii

zz

zz

,M,M

,M,MMM ]M[]M[

]M[]M[

≠

≠Σ Σ

Σ=α (1)

Fig. 3. Cumulative stripping curves vs. time Fig. 4. Separation factor vs. time � Na(I), K(I), � Mg(II), � Ca(II), � Pb(II)

As it can be seen in Figs. 3 and 4, 2-(octylthio)benzoic acid exhibits both high transport performance and high selectivity for Pb(II), contrary to the


270

rest of the investigated cations. For a better description of transport rates, a maximum stripping flux for each cation was found. For that purpose, each cumulation curve was fitted with a sigmoid function in accordance with Eq. (2). The inflexion points of the given curves were found by double differentiation (using MathCAD), and the respective values of maximum fluxes were evaluated (Tab. 1).

++=

−−d

c

s bat

e1)t(Q (2)

Table 1 Maximum output fluxes, Jmax·1010 (mol·cm2·s-1)

Na(I) K(I) Mg(II) Ca(II) Pb(II) 0.002 0.03 0.003 0.17 6.20

As it can be interpreted from Tab. 1, the following order of the transport rates was observed Pb(II)>Ca(II)>K(I) ≈Mg(II)≈Na(I).

4. CONCLUSIONS

2-(octylthio)benzoic acid, used as an ion carrier, enhances transport of Pb(II) over Na(I), K(I), Mg(II) and Ca(II).

The multimembrane hybrid system with the liquid membrane containing an appropriate carrier can be applied for construction of an efficient apparatus for the separation of ions.

According to the high selectivity obtained for Pb(II) in the investigated system, its potential application for the removal of heavy metal ions from waste waters can be suggested.

REFERENCES

[1] J.D.Lamb, R.M.Izatt, P.A.Robertson, J.J.Christensen, J.Am.Chem.Soc.,1980,102, 2452. [2] A.A. Kalachev, L.M.Kardivarenko, N.A.Platé, V.V.Bagreev, J.Membr.Sci., 1992,75, 1. [3] S. Dadfarnia, M. Shamsipur, Bull. Chem. Soc. Jpn., 1992, 65, 2779. [4] A.J. Walsh, H.G. Monbouquette, J. Membr. Sci., 1993, 84, 107. [5] Y. Baba, K. Inoue, Solv. Extr. Ion Exch., 1984, 2(7&8), 1021. [6] K. Saito, I. Taninaka, S. Murakami, A. Muromatsu, Talanta, 1998, 46, 1187. [7] D.S Barnes, G.J. Ford, L.D. Pettit, C. Sherrington, J. Chem. Soc., 1971(A), 18, 2883. [8] G.J. Ford, P. Gans, L.D. Pettit, C. Sherrington, J. Chem. Soc., 1972, 16, 1763. [9] L.D. Pettit, C. Sherrington, J. Chem. Soc., 1968, 12, 3078. [10] R. Wódzki, P. Szczepański, J. Membr. Sci., 2002, 197, 297. [11] R. Wódzki, J. Wasilewski, Properties of acyclic ionophore with carboxymethylthio-

group in the liquid membrane pertraction of alkaline and alkali-earth cations, 8th Int. Sym. Sep. Sci., Toruń, 2002.

[12] S.E. Livingstone, J. Chem. Soc., 1956, 437.


271

ROTATING HYBRID PERTRACTOR

Aneta RASZKOWSKA and Romuald WÓDZKI

Nicolaus Copernicus University, Faculty of Chemistry, 87-100 Toruń, Poland [email protected]

Abstract The hybrid version of a rotating film pertractor with discs made of a cation

exchange polymer (CEM, Nafion) was constructed and tested in the process of Zn2+/Cu2+separation. The bulk liquid membrane composed of D2EHPA in kerosene was used. The separation coefficients and maximum output fluxes were calculated and compared to the blank system (without CEMs).

1. INTRODUCTION

Pertraction is a continuous extraction and re-extraction process, which occurs in a liquid membrane system composed of three phases, i.e.: an aqueous feed phase, organic phase (a liquid membrane phase with or without a carrier), and a stripping phase (an aqueous solution with a stripping agent) [1-3]. A liquid membrane used in this process can be arranged as a bulk liquid membrane (BLM), supported liquid membrane (SLM) or emulsion liquid membrane (ELM). Depending on a specific liquid membrane composition, the process offers the possibility to separate various metals, organic compounds, preconcentration of analytes or extraction of natural compounds from aqueous media of different origins [4-7]. To improve the stability of the liquid membrane operation, some multimembrane hybrid systems (MHS) were also proposed as a method for a practical application of bulk liquid membranes [8-10]. The idea of such systems is based on coupling the properties of physically and chemically different types of membranes, i.e. liquid membranes and polymer cation-exchange membranes [11]. It was demonstrated that such systems are very stable in time of operation and can be considered as a very useful method of BLM application in various configurations [8-10].

Another problem to solve is the construction of practical equipment to implement the pertraction idea. The related devices, called thereafter “pertractors” can be divided into the following types: creeping pertractor [12], rotating pertractor [13], flat pertractor [14], complex pertractor [15], hollow-fiber pertractor [16-17], spiral pertractor [18]. Similar solutions can be applied in the case of MHS [19]. While looking for improved laboratory devices (e.g. for analytical applications) we paid attention to the rotating


272

pertractor. Its first version was constructed and described by Schlosser et al. [13], and then widely applied by Boyadzhiev et al.[3-5]. The scheme of this device is presented in Fig.1.

Fig. 1. Scheme of rotating film pertractor: 1-rotating discs, 2-common shaft, LM-

liquid membrane, F-feed solution, S-stripping solution.

According to the scheme, in the classic rotating film pertractor the hydrophilic discs (e.g. made of cellulose) rotate between aqueous and organic phases. Adhering feed or striping solutions are transferred to the organic phase where respective extraction / re-extraction phenomena occur, the mechanism of which depends on the liquid membrane composition. Exploiting the idea of hybrid systems we constructed the hybrid rotating pertractor in which the discs are made of a cation-exchange polymer (Fig. 2).

Fig. 2. Scheme of hybrid rotating pertractor: 1-rotating discs made of cation-

exchange polymer (CEM), 2- mover, 3-common shaft, LM-liquid membrane, F/S-feed/stripping solution.

Cation-exchange discs fulfill different functions, i.e. (i) they can carry

adhering liquids from the feed or stripping phase to the membrane, (ii ) to concentrate cations in the their internal solution, and to separate cations from the feed according to their selectivity properties. In practice, due to high ion exchange capacity, the rotation of discs may be much slower than in the case of a classic device. The introductory testing of HRFP (hybrid


273

rotating film pertractor) was carried out, and some results are presented below.

2. EXPERIMENTAL

The pertraction cells were made of Plexi glass. The device was composed of two compartments, each of which comprising an upper and a lower part. An upper part of the pertractor was filled with the organic phase (a bulk liquid membrane with the carrier). A lower part was divided into two sections containing the feed or stripping solution (see Fig. 2). The rotating discs of 3.5 cm diameter and 7.2 cm2 area were made of the Nafion 117 cation-exchange membrane. The contact area of the organic and aqueous phases was 7,8 cm3. The compartments were filled with 50 cm3 of the feed or stripping solution which additionally were circulated from external reservoirs of 1000 and 100 cm3 volume by two peristaltic pumps. The initial compositions of the aqueous and organic phases was as follows:

• feed (donor) solution: Zn(NO3)2 and Cu(NO3)2 (P.O.Ch., analytical purity) each of 0.01 mol/dm3 concentration,

• stripping (acceptor) solution: H2SO4 of 1 mol/dm3 concentration, • membrane: 50 cm3 of 0,1 mol/dm3 solution of D2EHPA (Chemical

Co. SIGMA, purity 95%) in kerosene. The concentration of zinc and copper in the feed and stripping solution in time of experiment (75 h) was determined using the AAS method (SPECTRA-20ABQ, Varian Co.), The pH of the feed solution was controlled by the ELMETRON CPC-501 pH-meter. The disc rotation velocity was set at 6 rpm.

3. RESULTS

Two series of measurements were carried out: one with ionic discs as described above, and another, without ionic discs (blank experiment with a plastic handle only). The results are presented in Fig. 3 a-b as the plots of metals concentration in the stripping solution vs. time of experiment. As expected, according to the properties of D2EHPA, Zn2+ cations are preferentially transported into the stripping solution in the both systems. To compare the separation abilities of the both systems, the separation coefficients

[Zn] / [Cu]

[Zn] / [Cu]Zn s sCu

f f

α = (1)

were calculated and presented in Fig.4 a-b. Because the output fluxes in the system with the feed and stripping solutions of limited volumes are dependent on time with maximum at the inflexion point of curves presented


274

in Fig.3, the corresponding maximum fluxes Jmax,Zn and Jmax,Cu were calculated as characteristics of the systems studied. For this purpose, the experimental data Vs×[M] s/A (M=Zn or Cu) = f(t) were fitted (TableCurve 2Dv5.01) with a sigmoidal function, and then differentiated numerically to calculate instantaneous fluxes.

( )1

t c

d

bf t a

e− −

= ++

(2)

The values of maximum output flux are presented in Tab.1.

Fig. 3. Concentration of Zn2+ and Cu2+ in the stripping solution vs. time of transport: (a) HRFP with cation-exchange polymer discs, (b) blank experiment.

Fig. 4. Separation of Zn2+ and Cu2+ vs. time of transport: (a) HRFP with cation-exchange polymer discs, (b) blank experiment.


275

Table 1. Maximum output fluxes of Zn2+ and Cu2+.

Jmax (mol/cm2s) System

Zn2+ Cu2+

Hybrid rotating film pertractor Blank experiment

8,64 × 10-9 4,94 × 10-9

9,51× 10-11 7,40 × 10-11

4. CONCLUSIONS

The results of this introductory study show that the hybrid rotating film pertractor is an alternative device for pertraction and concentration of zinc from Zn(II)/Cu(II) mixtures in aqueous solutions with bulk liquid membranes. It was found that the process with HRFP allows a very selective zinc recovery over copper, i.e. the separation factors Zn

Cuα reach the

value 175, for HRFP and 130 for the blank system. Also, the maximum output flux of Zn(II) observed for the hybrid version of the rotating film pertractor was two times higher than the value for the blank system.

REFERENCES

[1] L. Boyadzhiev, Z Lazarova, Liquid Membranes (Liquid Pertraction), in Membrane Separation Technology (eds. R. Noble, S.A. Stern, Elsevier, Amsterdam, 1995, 283-352

[2] L. Boyadzhiev, N. Kirilova, Bioproc. Engng., 2000, 22, 373. [3] V.A.Deratani, S.Touil, J.Palmeri, S. Tingry, S. Bouchtalla, Desalination, 2006, 200, 103. [4] K.Dimitrov, S.Aleksandrova, L.Boyadzhiev, S.Ruellan, M.Burgard, Sep. Purif. Technol,,

1997, 12, 165. [5] S. Zhirkova, K. Dimitrov, G. Kyuchoukov, L. Boyadzhiev, Sep. Purif. Technol., 2004,

37, 9. [6] R. Wódzki, P. Szczepański, Sep. Purif. Technol., 2005, 21, 289. [7] R. Gawroński, Mat. Membrane School, Jachranka, Poland, May, 6-8, 1997. [8] R. Wódzki, G. Sionkowski, Sep. Sci. Tech., 1995, 30, 2763. [9] R. Wódzki, P. Szczepański, Sep. Purif. Technol., 2001, 22-23, 697. [10] R. Wódzki, J. Nowaczyk, M. Kujawski, Sep. Purif. Technol., 2000, 21, 39. [11] O. Kedem, L. Bromberg, J. Membr. Sci.,1993, 78, 255. [12] L. Boyadzhiev, E. Bezenshek, Z. Lazarova, J. Membr.. Sci., 1984, 21, 137. [13] B. Daniel, S. Schlosser, E. Kossaczky, Patent CSR, 235362, 1984. [14] W. Kamiński, W. Kwapiński, Polish J. Environm. Stud., 2000, 9, 37. [15] Membrany i Membranowe Techniki Rozdziału (ed. A. Narębska), NCU Press, Toruń

1997, 359-396 [16] Z. Lazarova, B. Syska, K. Schugel, J. Membr. Sci., 2002, 202, 151. [17] S. Schlosser, E. Sobolova, R. Kertesz, L. Kubisova, J. Sep. Sci., 2001, 24, 509. [18] J.A. Jonsson, L. Mathiasson, Trends Anal. Chem., 1992, 3, 106. [19] R. Wódzki, P. Szczepański, J. Membr. Sci., 2002, 197, 297.


280

MASS TRANSFER IN MEMBRANE ABSORPTION SYSTEMS

Alexander OKUNEV1) , Nikolay LAGUNTSOV2) , Ivan KURCHATOV1) ,

Sergey DEMCHENKO1) 1) JSC “Aquaservice”, Russia, Moscow, 115409, Kashirskoe sh., 31, e-mail:

[email protected] 2) Moscow Engineering Physics Institute (State University), Russia,

Moscow115409, Kashirskoe sh., 31

Abstract

The mass transfer in membrane absorber was investigated. The necessity of taking into account of nonequilibrium interfaces mass transfer has been revealed. Nonequilibrium of process on different stages of mass transfer was considered. Special mathematical models and experimental methodic were developed. Mathematical models were used to optimize the different parameters of membrane absorption systems such as geometric and process operation parameters and membrane choosing. Using the results of the research the special membrane absorbers and systems based on it was developed.

The main applications of the membrane gas absorption technology are

acid gases removal from mixtures, gas drying, liquid gasification and degassing. Membrane gas absorption is a successful hybrid of membrane technology with high flexibility and scalability and absorption technology with high separation factors.

Separation in membrane absorber (also known as membrane contactor) is realizing due to selective mass transfer through membrane between moving gas and liquid phases. The membrane is permeable for gases and impermeable for liquid phase.

Most researchers working with membrane gas absorption in mathematical modeling uses approach based on calculating macroscopic similarity to the operation mode in which mass transfer in liquid phase taking place in the narrow boundary layer near to membrane. This approach leads to good agreement of experimental and calculated results in case of high flow rates of liquid. More perspective is the approach based on calculating of microscopic material balances in gas and liquid phases [1]. In this case the results of separation can be obtained without intermediate experiments what greatly simplifies the optimization of the process.

To obtain right results of calculation mathematical model must include describing of all sufficient stages of the process. Particularly researchers


281

dealing with mass transfer in membrane absorbers assume local equilibrium on interfaces. But, it’s simple to show, that for oxygen transport process in membrane contactor with porous membrane and water as the liquid carrier this assumption doesn’t work. To understand this it’s enough to compare deoxygenating productivity of same membrane contactor with Palladium applied on the water side of porous membrane operating with excess gaseous flux of hydrogen and nitrogen [2]. Liquid flow rates in both cases are identical, that means identical diffusion process in the volume of the liquid, excess of gaseous fluxes means highest driving force of process if in this case interface equilibrium taking place the productivities must be the same. But this isn’t right, and applied Pd realizing additional interface process and increase productivity.

Solubility of oxygen in water is small that means high diffusion resistance of liquid phase and in this case sufficient influence of nonequilibrium interface mass transfer is very strange. In other cases of more soluble gases in liquids, for example CO2 absorption in water and water solutions of amines and carbonates sufficient influence of nonequilibrium absorption is more expectable and may provide high difference between calculated results and real data.

For correct describing of nonequilibrium mass transfer on interfaces and gas absorption/desorption through porous and nonporous membrane special approach have been developed [3]. The approach can be applied to most of membrane gas absorption processes including material transformations on interface. Theoretical investigations of obtained results shown that using of nonporous membranes can increase overall contactor productivity comparing with porous membranes and sorption on free surface. This conclusion contrary to common opinion in this question.

Taking into account the nonequilibrium character of interface mass transfer leads to dependence of nonporous membrane permeability not only on given gas but also on absorption liquid. This permeability should be measured in this exact system: certain gas, certain membrane, certain liquid carrier. For measuring this permeabilities and other kinetic parameters special methodic has been realized. The methodic based on minimizing of difference between theoretical and experimental results of gastransfer in membrane contactor with two independent gas phases.

Besides of the problem membrane choosing it is very important to realize geometric and process operation parameters for effective separation by membrane gas absorption. In case of binary gas mixture separation mathematical estimations have shown existence of regimes in which there are high liquid saturation degree by good soluble component and high purification degree of low soluble component in gas phase are simultaneously realizing.


282

To provide this regimes overall membrane permeability of good soluble gas must be high enough, liquid and gas must flow in countercurrent mode, liquid phase thickness must be small to facilitate transport in it and liquid flow rate must be small enough [1]. This requirements realizing on specially developed and realized flat sheet membrane contactor [4]. In this contactor nonporous PVTMS membrane is used. It has thin selective layer and good interface with water mass transfer properties that provide high gas permeability. Absence of through pores excludes liquid passing to gas phase in wide range of operation pressures.

To separate binary gas mixtures by gas absorption method it is also desorber needed that should be connected with absorber. They should be connected in cycle. Desorber provides liquid regeneration and obtaining of gas flux enriched by good soluble component. If in membrane desorber similar to described above absorber regime is realizing then in terms of diffusion in liquid phase it’s able to obtain two flows on the exits from the system, one of them highly enriched by low soluble component, other by high soluble component [5]. These regimes can not be calculated using approach of mathematical modeling based on macroscopic similarity to the operation mode in which mass transfer in liquid phase taking place in the narrow boundary layer near to membrane.

Similar regimes are realized using developed flat sheet membrane contactors in biogas separation process. Systems have important feature – they can operate without temperature difference between absorber and desorber. This sufficiently reduces energy consuming. CO2 in the systems can be obtained in several cases: first – technically pure gas, to realize it desorber vacuum pumping; second – dissolved in water and third – in the mixture with air. Last two forms do not need sufficient inputs on pumping and CO2 in this forms used to plant fertilization. This systems oriented on local using in agriculture to increase its autonomy.

Acknowledgements

The work was supported by the grant of Russian Foundation for Basic Researches #06-08-01626-а.

REFERENCES

[1] A.Yu. Okunev and N.I. Laguntsov, J. Eng. Ph. and Thermoph., 2006, 79, 781. [2] R. van der Vaart, B. Hafkamp, P.J. Koele, M. Querreveld , A.E. Jansen, V.V. Volkov,

V.I. Lebedeva, V.M. Gryaznov, Proc. Int. Conf. “Euromembrane 2000”, 359. [3] A.Yu. Okunev and N.I. Laguntsov, Theor. Found. of Ch. Eng., 2007, 41, 257. [4] Kojevnicov V.Yu., Levin E.V, Laguntsov N.I., Okunev A.Yu., Hafisov R.S., patent RF

№2304457 of 20.08.2007. [5] A.Yu. Okunev, Yu.P. Neshchimenko, A.I.Dianov, Proc.Sci.Session of MEPhI 2007,22.


365

MICELLAR AND POLYELECTROLYTE ENHANCED

ULTRAFILTRATION FOR REMOVAL OF MANGANESE AND IRON FROM AQUEOUS MEDIA

Gryzelda POŹNIAK

Wrocław University of Technology, Faculty of Chemistry WybrzeŜe Wyspiańskiego 27, 50-370 Wrocław

Abstract The removal of manganese (Mn2+, MnO4

-) and iron (Fe3+, Fe(CN)44-) from

aqueous media by micellar (MEUF) and polyelectrolyte (PEUF) enhanced ultrafiltration was investigated. For present study heksadecyltrimethylammonium bromide (CTAB) as the surfactant and polyethyleneimine (PEI) as the polyelectrolyte were chosen. The membrane was formed from aminated polyethersulfone. It was shown that MEUF and PEUF can be valuable alternative for the removal of manganese and iron from aqueous media.

1. INTRODUCTION

The intensive development of agriculture and wide application of chemicals led to the contamination of many natural water reservoirs and ground waters by heavy metal ions. The traditional treatments of metal-contaminated waste comprised such processes as: liquid-liquid extraction, adsorption, precipitation or ion-exchange. The mentioned above techniques posses some disadvantages - they are based on the heterogeneous reactions and usually eliminate but not recover the metals. Membrane separation processes, consuming less energy and being more ecological than conventional separation techniques, are receiving considerable interest in technologies important from environmental point of view.

Ultrafiltration (UF) is efficient in removal large molecular substances (polymers, colloids) of size ranging from 2 to 100 nm and molecular weight larger than 500 Da. The pore size of any ultrafilter is too large to reject small molecules like inorganic ions. In these cases simple ultrafiltration can be replaced by such non-conventional processes as Micellar Enhanced Ultrafiltration (MEUF) [1] and Polymer Enhanced Ultrafiltration (PEUF) [2]. MEUF and PEUF are the combination of two phenomena - ultrafiltration with solubilizing of metal ions by surfactant micelles (MEUF), or binding to a water soluble polymer (complexation) (PEUF).


366

Polyethersulfone, PES, is frequently used as a membrane material due to its thermal, biological, and chemical stability. Hydrophobicity, the most serious disadvantage of the PES membrane, can be overcome by introduction of ionic functional groups to the polymer. The porous ion exchange membranes form the new category of filtration media. The improvement of solute rejection and decrease of membrane fouling ability are the main profits coming from the use of such membranes [3-5].

The aim of this work was to test the usefulness of porous ion exchange membranes made of aminated polyethersulfone in the non-conventional separation processes (MEUF and PEUF) for removal of manganese and iron from aqueous media.

2. EXPERIMENTAL

2.1. POLYMER MODIFICATION

Polyeterosulfone (PES), Ultrason E2020P (Mw=58kDa, BASF) was chloromethylated using a mixture of methyl chloromethyl ether (50 mole per mer PES) with SnCl4 (5 mole per mer PES). The reaction was conducted for 24 hours at room temperature. The porous membranes were formed directly from the chloromethylated derivative. Aminolysis of chloromethylated polymer was carried out in 50 vol.-% solution of N,N-dimethylaminoethanol in 1:1 mixture of water and methanol. The reaction was completed within seven days at room temperature.

2.2. PREPARATION AND PROPERTIES OF MEMBRANES

Porous asymmetric membranes were formed by the phase inversion method from 16%-wt. solution of chloromethylated PES (after amination – porous anion exchange membrane), using water was as the coagulation medium. The ion exchange capacity was assayed by acid-base titration method [6]. Porosity was determined gravimetrically and average pore diameter was calculated according to Ferry-Faxen relationship. The Amicon 8200 dead-end cell with filtration surface area of 19.6 cm2 was used. The transmembrane pressure was equal to 0.1 MPa.

Mixtures of K4[Fe(CN)6] or KMnO4 (1 mM) with hexadecyl-trimethylammonium bromide, CTAB, (0.9 and 4.5 mM), and FeCl3 or MnCl2 (1mM) with polyethyleneimine, PEI, (4 mM) were filtered through APES membrane. The concentration of metals was determined using AAS method (AAnalyst 100, Perkin-Elmer). To evaluate the filtration efficiency in removal of manganese and iron, the rejection coefficient (R) was calculated:

R = (1 – cp/cf) x 100 (1)

where cp , cf are the ion concentrations in permeate and feed, respectively.


367

3. RESULTS

To reduce tendency of cationic surfactant to deposit on the membrane surface, aminated polyetherosulfone membrane was used (Table 1).

Table 1. Properties of ultrafiltration membrane from APES

Ion exchange capacity mmol/g

Porosity %

Average pore diameter

nm

Flux of water under 0.1 MPa

dm3 /m2 h

1.6 79 21 100

The porous structure of APES membrane is suitable for use in MEUF and PEUF processes - rejection of PEI and CTAB reaches values above 80%.

a) b)

0102030405060708090

100

Rej

ectio

n [%

]

0 0.9 4.5

CTAB concentration [mM]

KMnO4 K4[Fe(CN)6]

0102030405060708090

100R

ejec

tion

[%]

3 6 9

pH

MnCl2 FeCl3

Fig. 1. Rejection of manganese and iron as a function of: a) concentration of CTAB, b) pH of PEI

Manganese and iron are solubilized on the surface of CTAB micelles

due to strong ionic interaction between anionic form of investigated metals and cationic surfactant. At concentration of CTAB equal to 4.5 mmol/dm3 (5 cmc) almost 100% of both anions are rejected (Fig. 1a).

PEI is very effective coupling agent. Iron cations bond to PEI stronger than manganese ions. The best results for both cations have been obtained at pH = 9 (Fig. 1b).

During PEUF-MEUF hybrid process higher rejection of Mn(II) is obtained (Fig. 2). In this process sodium dodecyl sulfonate (SDS, anionic surfactant) and polyethyleneimine compete in Mn(II) cations bounding. After addition of PEI to SDS solution the critical micelle concentration is lowered [7]. The best result, almost 90% of Mn(II) rejection, was obtained for the mixture of 4 mM PEI and 4.05 mM SDS (0.5 cmc for neat SDS in distilled water).


368

0102030405060708090

100

Rej

ectio

n o

f Mn

Cl2

[%]

A B C

Fig. 2. Rejection of manganese as a function of kind of enhanced agent:

A. SDS, c=4.05 mM, B. SDS, c=8.1, C. SDS, c=4.05mM + PEI, c=4 mM

4. CONCLUSIONS

1. Micellar-enhanced and polymer-enhanced ultrafiltration can be valuable alternative for the removal of manganese and iron from aqueous medium.

2. Rejection of manganese and iron anions in MEUF depends on concentration of surfactant; almost full retention (100%) can be obtained with concentration of CTAB equal to 5 cmc.

3. Rejection of manganese and iron cations in PEUF depends on pH of PEI; the highest rejection can be obtained at pH=9.

4. The best results can be obtained for removal of manganese cations during PEUF-MEUF hybrid process.

Acknowledgements

This work was supported by the Polish Scientific Committee in the framework of a grant no 3 T09B 047 28.

REFERENCES

[1] B.L. Rivas, E.D. Pereira, I. Moreno-Villoslada, Prog. Polym. Sci., 2003, 28, 173. [2] R.S. Juang, Y.Y. Xu, C.L. Chen, J. Membr. Sci., 2003, 218, 257. [3] G. Poźniak, I. Gancarz, W. Tylus, Desalination, 2006, 198, 215. [4] G. Poźniak, Ars Sep. Acta, 2006, 4, 50. [5] G. Poźniak, R. Poźniak, M. Bryjak, Chemistry for Agriculture, 2007, 8, 188. [6] A. Hamza, G. Chowdhury, T. Matsuura, S. Sourirajan, J. Appl. Sci., 1995, 58, 613. [7] I. Benito, M.A. Garcia, C. Monge, J.M. Saz, M.L. Marina, Coll. Surf. A: Physicochem.

Eng. Aspects, 1997, 125, 221.


369

REMOVAL OF RADIONUCLIDES FROM WATER SOLUTION

IN ULTRAFILTRATION/COMPLEXATION PROSESS WITH POLYETHERSULFONE (PES), POLYSULFONE (PS) AND

SURFACE-MODIFIED (SMM) MEMBRANES

Mohamed KHAYET1), Marian HARASIMOWICZ2), Agnieszka JAWORSKA2), GraŜyna ZAKRZEWSKA-TRZNADEL2)

1) University Complutense of Madrid, Spain 2) Institute of Nuclear Chemistry and Technology, Dorodna 16, 03-195 Warszawa

Abstract Ultrafiltration coupled with complexation by water soluble polymers was

tested to assess the membrane surface modification and its influence on removal of radioactive metal ions from water solutions The experiments were conducted using porous membranes prepared from polysulfone (PS) and polyethersulfone (PES), and membranes made from the same polymers with modified surface by employing surface modifying marcomolecules (SMMs). After 2h operation, the adsorption of 60Co by SMMs modified membranes was found to be 60-75 % smaller than that of the unmodified membranes and the decontamination factor was 2-3 times higher for the SMMs modified membranes.

1. THE HYBRID ULTRAFILTRATION/COMLEXATION PROCESS

The hybrid ultrafiltration/comlexation process was proposed as an efficient method that can replace reverse osmosis (RO) applied for radioactive wastes processing [1]. Ultrafiltration coupled with complexation by water soluble polymers was tested to assess the membrane surface modification and its influence on removal of radioactive metal ions from water solutions.

The experiments were conducted by use of porous membranes prepared from polysulfone (PS) and polyethersulfone (PES), and membranes made from the same polymers with modified surface by employing surface modifying marcomolecules (SMMs). The membranes were manufactured by the phase inversion method [2] in Department of Applied Physics I (University Complutense of Madrid, Spain). Commercial PES membranes supplied by Millipore having molecular weight cut-off 10 kDa were used for comparison. A typical UF set-up equipped with stainless steel membrane module of effective membrane area of 14.52⋅10-4 m2, was employed in the experiments.


370

2. EXPERIMENTAL

Ultrafiltration experiments with different membranes were conducted in several steps:

1) determination of permeation flux for pure water for each membrane during 1h,

2) saturation of the system by using 1 g/l [Co(NO3)2.6H2O] aqueous solution, during 1h,

3) measurement of permeation flux (i.e. performance ratio PR) and determination of the separation factor for each membrane (Eq.(1)) using 0.1 g/l [Co(NO3)2.6H2O] aqueous solution, during 1h,

4) complexation by application of 1 g/l of polyethylineimine (PEIM) and determination of the permeation flux together with the retention factor,

5) Addition of 60Co radioisotope and determination of the decontamination factor according to Eq.(2) (2h experiment).

All experiments were carried out at 24 ºC. Retention factor was calculated according to the formula:

( ) 100/1 ⋅−= fp CCR (1)

where Cf and Cp represent the solute concentration of the feed and permeate, respectively.

The decontamination factor, which characterizes the efficiency of the membrane process was calculated as follows:

pf AAD /= (2)

where Af and Ap are the activity of the feed and permeate, respectively. The cross-flow filtration system was used to conduct ultrafiltration

(UF) experiments. The feed pressure was maintained at 0.22 MPa and the feed flow rate was 40 l/h.

Aqueous solutions containing non-radioactive [Co(NO3)2.6H2O] (0.1 g/l) (i.e. electrical conductivity of about 100 µS/cm) was used for saturation of the UF system. The permeation flux and the rejection factor (R) were determined. 1 g/l of polyethyleneimine (PEIM) was used for complexation of cobalt ions (Co2+). The pH value was adjusted to be 6 by using a solution of 10% HNO3. Again, the permeation flux and the rejection factor (R) were determined. Radioactive solution containing 60Co as 60Co(NO3)2.6H2O was employed as a model radioactive feed solution. The final activity of the feed was about 4000 counts/100s.


371

3. RESULTS

The experiments showed that the flux of pure water (PWP) was highest for PES commercial membrane – 244 ml/h and for PES unmodified, laboratory-made membrane-185 ml/h. For unmodified PS membrane the flux was 140 ml/h; for all SMMs modified membranes (SMM3/PES and SMM41/PES) permeate fluxes were lower and they reached 61 and 57 ml/h, respectively. It must be pointed out that for SMM3/PS and SMM41/PS membranes the PWP was very low and therefore these membranes were not used in ultrafiltration/complexation experiments.

In ultrafiltration/complexation experiments the fluxes for the membranes tested were as follows: PES commercial-121, PES unmodified -123, PS unmodified - 101, SMM3/PES - 38,5 and SMM41/PES - 34 ml/h. The membranes with modified surface exhibited the highest retention of cobalt nitrate: the retention of Co2+ was over 98% for SMM3/PES and ca. 97% for SMM41/PES. At the same conditions the retention of unmodified PES was 71%. Substantially high decontamination factors for radioactive 60Co were also obtained when modified membranes were employed in UF/complexation process: for SMM3/PES – 300, for SMM41/PES- 163. For unmodified membranes decontamination factors were as follows: commercial PES- 44, laboratory made, unmodified PES - 75 and unmodified PS - 36. The obtained results are showed in Tables 1 and 2.

Table 1. Pure water permeation flux and ultrafiltration results 0.1 g/l Co(NO3)2 aqueous solution with 60Co.

Membrane PWP(g/h) PR (g/h) Cf(mg/l) CR(mg/l) Cp(mg/l) R (%) 120.7 20.75 19.5 14.25 29.19 PES

Commercial 244.8

80.2 13 11 0.75 93.75 123.52 21.75 21.25 12 44.2

Unmodified PES 185.2 85.2 12.75 11.25 3.5 70.8 38.5 22 21.5 10.5 51.72

SMM3/PES 60.9 29.44 12.75 11.25 0.15 98.75 34.2 21.5 21.0 11 48.23

SMM41/PES 57.44 25.69 12.75 11.5 0.4 96.70 100.7 20.7 19.8 17.6 13.1

Unmodified PS 140 60.04 13.0 12.7 0.47 96.34

Cf: feed concentration, Cp: permeate concentration; CR: retentate concentration; R: retention factor


372

Table 2. Pure water permeation flux and ultrafiltration results 0.1 g/l Co(NO3)2 aqueous solution with 60Co.

Membrane Af

1 (counts/100s)

Ap (counts/100s)

D Am 7

(counts/100s)

PES Commercial

3691 R: 870

52 2 43.86 2941

Unmodified PES 3594

R: 1475 34 3 74.54 1532

SMM3/PES 3594

R:1463 11 4 229.86 401

SMM41/PES 3663

R: 1890 17 5 163.32 600

Unmodified PS 3574

R: 1902 77 6

35.56 1861

D: Decontamination factor; Am: membrane activity after 2 h UF experiment with 60Co radioisotope.

1- R: Retentate 2- Without radioactive 394. Distilled water activity: 377 (counts/100s). New membrane counts: 423 3- Distilled water activity: 406 (counts/100s) 4- Distilled water activity: 423 (counts/100s) 5- Distilled water activity: 406 (counts/100s) 6- Distilled water activity: 423(counts/100s) 7- Support paper activity: 445 (counts/100s)

4. CONCLUSIONS

Substantially high decontamination factors for radioactive 60Co were obtained when modified membranes were employed in UF/complexation process, however permeate flux was reduced. Modified membranes showed smaller adsorption of radioactive cobalt that was beneficial taking into account the future applications. After 2h operation the adsorption of 60Co by Surface modifying macromolecules membranes was 60-75 % smaller than the unmodified membranes (PES).

REFERENCES

[1] G.Zakrzewska-Trznadel, M.Harasimowicz, Desalination, 2002, 144, 207. [2] Khayet M. et al., J. Appl. Poym. Sci., 2003, 89, 2902.


373

REMOVAL OF COPPER(II) IONS USING MICELLAR-ENHANCED ULTRAFILTRATION

Katarzyna STASZAK and Krystyna PROCHASKA

Institute of Chemical Engineering and Technology, Poznań University of Technology, pl. M. Skłodowskiej-Curie 2, 60-965 Poznań, Poland

e-mail: [email protected]

Abstract

Removal of copper(II) ions from micellar solutions was studied in the cross-flow SEPA CF Osmonics module. In the case of high concentration of copper(II) ions (0.1 M) the efficiency of ultrafiltration is not satisfactory. That’s why the next step of this work is using the hydrophobic reagents to complexation the metal ions in order to improve the ultrafiltration process.

1. INTRODUCTION

The traditional techniques for the removal of metal ions from aqueous effluents like process of ion exchange, activated carbon adsorption and electrolytic removal are rather expensive but still used [1].

The use of techniques for the elimination of metal from liquid effluents based on separation by means of membranes is becoming increasingly more frequent. Reverse osmosis (or at least nanofiltration) can be used due to the size of the ions in aqueous solutions. But the usual permeate fluxes of RO membranes are limited and require high transmembrane pressure, which makes the process expensive [2]. In recent years Micellar-Enhanced Ultrafiltration (MEUF) has also been used. This technique combines the high resistance of reverse osmosis with the high relative flow of ultrafiltration.

The MEUF technique has been used in the elimination of microcontaminants in three ways: i) the use of an anionic surfactant allows micelles to formed where the organic part is facing the centre and the negatively charged hydrophilic part is facing out. The metals or cationic contaminants bind on this negatively charged surface; ii) the use of a cationic surfactant allows the formation of micelles, where the organic part is facing the centre and the positively charged hydrophilic part is facing out. The oxianions or anionic contaminants bind on the surface; iii) the water-repelling contaminants solubilize inside the micelles.


374

Many researchers have studied the application of MEUF for the separation of various ionic pollutants, but mainly in the much diluted systems (below 0.01M) [1, 3-5].

In the present work the retention of copper(II) ions with anionic surfactant – sodium dodecyl sulfate (SDS) were studied. The aim of this work is testing the usefulness of MEUF technique in the case of feed solution of high concentration of copper(II) ions (0.1M).

2. EXPERIMENTAL

2.1. REAGENTS

The following regents were used: Sodium Dodecyl Sulfate (>99% pure, Sigma Aldrich) and copper sulfate (CuSO4⋅5H2O, pure).

For the spectrometric measurements the ammonia and sulfate acid were used.

2.2. INSTRUMENTATION AND PROCEDURE

The ultrafiltration experiments were carried out in the SEPA CF Membrane Cell produced by OSMONICS, USA. The flat sheet polymeric membrane made of polyvinylidene fluoride. The effective surface area of the membrane was 0.0155 m2. The fluid was forced through the membrane at a pressure of 0.2 MPa. The inlet reservoir was initially filled with a 1000 ml feed solution and the process was stopped when 500 ml was taken as a permeate.

The CMC of surfactant and the concentration of surfactant in permeate was determined by the conductometric method [6].

The metal concentration in permeate was determined by UV spectroscopy using a Specol 1200, Analytic Jena, Germany by the complexation of copper ions by the ammonium solution [7].

The retention of metal was calculated from the equation:

NP ccR /1−=

where cP – concentration of metal ion in permeate, cN – initial concentration of metal in feed solution.


3.1. DETERMINATION OF SURFACTANT

During the ultrafiltration process the monomers of surfactant could be transported through the membrane. To check these phenomena the filtration of 5CMC SDS solution has been performed.


375

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 50 100 150 200

time, t [min]

CM

C o

f S

DS

12 13 14

87

0

10

20

30

40

50

60

70

80

90

R [%]

0.1M CuSO4 0.1M CuSO4

+ 5CMC SDS0.1M CuSO4

+ 10CMC SDS0.01M CuSO4

+ 5CMC SDS

Fig.1. Concentration of SDS in permeate

(CMC of SDS) as a function of time Fig.2. Average retention, R, of copper(II)

ions during the ultrafiltration process

Significant changes of surfactant concentration were observed during first 30 min of the separation process (Fig. 1). However, during the whole time of experiment (3 hours) the concentration of SDS was bellow its CMC.

3.2. METAL RETENTION

Four solutions of different Cu(II) concentration were filtrated. Three of them contained the total concentration of copper(II) ion equal to 0.1M and different concentration of anionic surfactant (0, 5 and 10 CMC) and in one solution with 0.01M CuSO4 and 5 CMS of SDS.

Fig.2 presents the variation of average retention in the case of UF of solutions of different composition. In dilute stream of copper ions the retention of metal is near 90%. However, in the case of high concentration of copper(II) ions the efficiency of ultrafiltration is not satisfactory. The retention is at the level of 12-14%. In order to enhance selectivity and efficiency of MEUF the ligand-modifed micellar enhanced ultrafiltration (LM-MEUF) should be used. This method involves additionally an amphiphilic ligand and a surfactant to the feed solution under conditions where most of the surfactant is present as micelles. The ligand reveals a high degree of solubilization in the micelles and a tendency to selectively complex the target metal ion [3]. Thus the next step of this work is investigation of MEUF with commercially available hydrophobic chelating extractant LIX 54 added to the separation systems.

4. CONCLUSIONS

The cross-flow micellar ultrafiltration experiments showed the usefulness of this method only for feeds of low concentration of copper(II) ions. In high concentration (0.1M) the retention is very small. However, the addition of anionic surfactant caused increase in retention of metal ions, but


376

the differences are over-slight. That’s way the next step of this work is using the hydrophobic reagents to complexation the metal ions in the order to improve the ultrafiltration process.

Acknowledgements

The work was supported by 32-270/BW/2008.

REFERENCES

[1] L. Gzara, M. Dhahbi, Desalination, 2001, 137, 241. [2] C. Guohua, J. Memb. Sci., 1997, 127, 93. [3] B.R Fillipi., J. F Scamehorn., S. D Christian., R.W Tayloret., J. Membr. Sci., 1998,

145, 27. [4] M. Aoudia, N. Allal., A. Djennet, L. Toumib, J. Memb. Sci., 2003, 217, 181. [5] K. Back, H. Leeb, J-W Yang, Desalination, 2003, 158, 157. [6] J. Garcia-Anton, J. L. Guinon, Colloid Surface, 1991, 61, 137. [7] A. Cygański, Metody spektroskopowe w chemii analitycznej, WNT, Warszawa, 1993.


377

REMOVAL OF CADMIUM IONS FROM AQUEOUS

SOLUTIONS USING CHITOSAN MEMBRANES

K. ZIELIŃSKA, A.G. CHOSTENKO, S. TRUSZKOWSKI

Chair of Nuclear and Radiation Chemistry Faculty of Chemistry, Nicolaus Copernicus University

ul. Gagarina 7, 87-100 Toruń, Poland e-mail: [email protected]

Cadmium is a metal commonly used in a series of industrial processes such as metallurgy, dye synthesis, surface treatment, pigments and battery production.

Pollution by metal ions is a major environmental problem. Toxic metal ions are difficult to remove from water, but their removal is fundamental to the preservation of the environment. Various methods exist for the removal of toxic metals from aqueous solution: precipitation, coagulation, sedimentation, ion-exchange, adsorption and electrochemical techniques. Adsorption is the most effective and widely used method of all, furthermore it is also considered as an economical method. Therefore biopolymers have been studied as adsorbents for the removal of heavy metals from water. Chitosan (Fig.1) is a natural polysaccharide that is gained in the process of deacetylation of the substance chitin. Due to its non-toxic, biocompatible and biodegradable characteristics, chitosan has been used in many industries, including water purification. The amine groups of chitosan are able to absorb metals through several mechanisms. The kind of interaction depends on the metal.

OH

H

H

NH2H

OH

CH2OH

H

OH

H

NHH

OH

CH2OH

HHO

O

nO

CH3

C

Fig. 1. The structure of chitosan


378

Although the adsorption properties of chitosan have been extensively studied by many researches, the mechanism of this process has not been well cheracterized.

The main purpose of this work was to determinate the capacity of the chitosan for adsorption of cadmium.

The measurements of the cadmium concentrations have been carried out with the use of an atomic absorption spectrophotometer. The results indicate that the chitosan membranes are effective adsorbents for the collection of cadmium ions. The amount of adsorbed Cd2+ was influenced by the initial Cd2+ concentration.

“a s 2008”

Documents