DOCTORA L T H E S I S

Luleå University of TechnologyDepartment of Chemical Engineering and Geosciences

Division of Mineral Processing

2008:17|: 102-15|: - -- 08⁄17 --

2008:17

Physicochemical Studies ofOxide Zinc Mineral Flotation

Universitetstryckeriet, Luleå

Seyed Hamid Hosseini

Seyed Ham

id Hosseini

Physicochemical Studies of O

xide Zinc M

ineral Flotation 20

08:17

DOCTORAL THESIS

Physicochemical Studies of Oxide Zinc Mineral Flotation

Seyed Hamid Hosseini

Division of Mineral Processing Department of Chemical Engineering and Geosciences

Luleå University of Technology

March 2008

To My Wife & My Son

AbstractAt the present, zinc is produced mostly from zinc sulphide ores because the sulphides are easy to separate from the gangue by conventional flotation techniques. In the case of oxide zinc ores, there is often no selectivity in terms of zinc recovery. The objective of the present study is to investigate the influence of different cationic, anionic and mixed collectors on the flotation of smithsonite mineral and an oxide zinc ore at various concentrations and pH values.The present thesis consists of three parts: i) characterization of smithsonite mineral and oxide zinc ore from the Angooran ore deposit, Iran, ii) physicochemical studies on smithsonite sample including zeta-potential, contact angle, microflotation tests and adsorption studies using diffuse reflectance FTIR (DRIFT) and X-ray photoelectron spectroscopy (XPS) techniques in the presence of cationic, anionic and mixed collectors (cationic/anionic) and iii) flotation behavior of oxide zinc ore from the Angooran ore in the presence of cationic, anionic and mixed collectors. The results of XPS and EDX on ore samples in different size fractions showed no significant variations in zinc percentage on bulk and surface of samples. The smithsonite flotation in the presence of dodecylamine after sulphidization using sodium sulphide shows a maximum recovery (94 %) at around pH 11.5. Among the anionic collectors used i.e. oleic acid, hexylmercaptan, KAX alone and KAX in the presence of sodium sulphide and copper sulphate, oleic acid indicates a maximum flotation recovery (93 %) at around pH 10. Regarding the mixed collector flotation, the recovery increases with increasing KAX concentration i.e. the flotation recovery enhanced to 96 %. Also the recoveries and contact angles are much higher compared to KAX and DDA alone are used.The spectroscopic data are interpreted in the light of flotation, zeta potential and contact angle results. The qualitative and quantitative evaluation of the adsorbed layer which determines the extent of surface hydrophobicity or hydrophilicity, are performed using spectroscopic methods (FTIR and XPS). The DRIFT and XPS spectra confirm the adsorption of dodecylamine, oleic acid, hexylmercaptan, KAX and mixed collector (KAX+DDA) on the smithsonite surface. These results are consistent with the zeta-potential, contact angle and also microflotation results. The studies on oxide zinc ore indicate a maximum bench scale flotation recovery of 84.5 % with zinc grade 24.5% in the presence of dodecylamine at pH 11.5. There are no significant variations in recovery of cationic and anionic and mixed collectors, however, amine and mixed collector flotation show to be more selective than anionic collectors. The recoveries with mixed collectors show that increasing KAX leads to improve the flotation recovery.

Keywords: Smithsonite, Oxide zinc ore, Cationic collector, Anionic collectors, Mixed cationic/anionic collector, Zeta-potential, Contact angle, Microflotation, Batch Flotation, FTIR, XPS

I

Acknowledgements

I would like to express my true gratitude to my supervisor Prof. Eric Forssberg for his continuous guidance, encouragement and invaluable discussions and also for giving me the opportunity to pursue my doctoral studies. I acknowledge Prof. K. Hanumantha Rao, Luleå University of Technology, Division of Mineral Processing, for his valuable comments and also the review of my doctoral thesis. I would like to thank Prof. Pär weihed, Head of Department of Chemical Engineering and Geosciences and Dr. Bertil Pålsson, Head of Mineral Processing Division, for their helps to finalize my doctoral thesis. I would like to thank Calcimine Co., Zanjan, Iran for providing the samples, which were used in this work, and also for giving some suggestions concerning my research work. I am very grateful to Prof. M. Sadrzadeh, Tehran University, Tehran, Iran for providing me an opportunity to do chemical analyses. I acknowledge Mr. Urban Jelvestam and all colleagues at the Department of Material Science, Chalmers University of Technology, Gothenburg, Sweden for their helps in the laboratory for XPS measurements and special thanks to Dr. Karl Arnby who was PhD student at the Department of Chemical Engineering, Chalmers University of Technology, Gothenburg, Sweden, being very friendly and for his help to measure zeta-potential. I would like to thank Dr. Andrei Shchukarev, Department of Inorganic chemistry, Umeå University, Umeå, Sweden for the XPS measurements. I would like to give my thanks to Mrs. Maine Ranheimer for her help in my FTIR experiments and to Mr. Ulf Nordström for his help in the laboratory. I wish to convey my thanks to all of my friends and colleagues at Departments of Chemical Engineering and Geosciences, Luleå, Sweden, especially the fruitful discussions with Dr. Nourreddine Menad concerning my research works have been very helpful and my thanks also extend to all Iranian friends at Luleå University of Technology, especially to Dr. Hamid Reza Manouchehri, Peyman Roonasi for their attentions and to Pejman Oghazi for his chatting and his kind helps in my everyday life. I would like to thank Dipl. Ing. Majid Ghameshlu, Siemens A.G., Vienna, Austria for encouraging me to start my PhD studies. I am very thankful Dipl. Ing. Nasser Hoshyar, Abniroo Consulting Engineering Co., Tehran, Iran for his supports during my life and also this research work.I am very grateful to my parents, my brother, my sister and also my wife’s parents for their blessings for the success in my life. Finally but most of all, I am deeply indebted to my wife, Pantea and my sweet son, Parsa for their constant love and patience during my research works.

Seyed Hamid Hosseini December 2007 Luleå, Sweden

II

List of Publications

This thesis is based on the work contained in the following papers, referred to by Roman numerical in the text:

Paper IAdsorption Studies of Smithsonite Flotation Using Dodecylamine and Oleic acid S. Hamid Hosseini and Eric Forssberg Minerals and Metallurgical Processing, SME, Vol.23, No.2, 2006, 87-96

Paper II Smithsonite Flotation Using Potassium Amyl Xanthate and HexylmercaptanS. Hamid Hosseini and Eric Forssberg Mineral Processing and Extractive Metallurgy, Trans. Inst. Min Metals. C, Vol.115, No. 3, 2006, 107-112

Paper III XPS & FTIR Study of Adsorption Characteristics Using Cationic and Anionic Collectors on SmithsoniteS. Hamid Hosseini and Eric Forssberg Journal of Minerals and Materials Characterization and Engineering, Vol.5, No.1, 2006, 21-45

Paper IV Physicochemical Studies of Smithsonite Flotation using Mixed Cationic/Anionic Collector S. Hamid Hosseini and Eric Forssberg Minerals Engineering, Vol.20, Issue 6, 2007, 621-624

Paper V Flotation Behavior of Oxide Zinc Ore from Angooran Deposit, Iran in the Presence of Cationic/Anionic and Mixed (Cationic/Anionic) Collectors S. Hamid Hosseini and Eric Forssberg European Journal of Mineral Processing & Environmental Protection, Vol.6, No.3, 2006 (In press)

Paper VI Comparison Between the Bulk & Surface Composition of the Samples from Angooran lead & zinc Mine, Zanjan Province, Iran S. Hamid Hosseini, Ulf Södervall and Eric Forssberg Proceedings of the 6th Conference on Environment and Mineral Processing, Ostrava, Czech Republic, June 2002, pp.289-294

III

Paper VII Mineral Processing Possibility of Oxide Lead & Zinc Minerals from Angooran Deposit in Zanjan province, Iran S. Hamid Hosseini and Eric Forssberg Proceedings of the IX Balkan Mineral Processing Congress, Istanbul, Turkey, September 2001, pp.221-226

List of abbreviation

Abbreviation Meaning

DDA DodecylamineDRIFT Diffuse Reflectance Infrared Fourier Transform EDX Energy Dispersive X-ray AnalyzerFTIR Fourier Transform Infrared Spectroscopy HLS Heavy Liquid Separation HMS Heavy Media Separation HM Hexyl mercaptan or Hexanethiol KAX Potassium Amyl Xanthate OA Oleic acid XPS X-ray Photoelectron Spectroscopy XRD X-ray Diffraction XRF X-ray Fluorescence

IV

Table of Contents

1. INTRODUCTION ........................................................................................................................................1

2. BACKGROUND...........................................................................................................................................2

2.1 SOLUBILITY AND STABILITY OF OXIDE ZINC MINERALS IN AQUEOUS MEDIA.........................................22.2 SOLUTION CHEMISTRY OF REAGENTS ....................................................................................................2

2.2.1 Cationic Collector (dodecylamine).................................................................................................................... 3 2.2.2 Anionic Collectors ............................................................................................................................................ 5 2.2.2.1 Oleic acid.................................................................................................................................................... 5 2.2.2.2 Hexylmercaptan.......................................................................................................................................... 5 2.2.2.3 Potassium amyl xanthate ............................................................................................................................ 6 2.2.2.4 Mixed Collectors ......................................................................................................................................... 7

2.3 FLOTATION OF OXIDE ZINC MINERALS ..................................................................................................7

3. MATERIALS AND METHODS ..................................................................................................................9

3.1 MATERIALS .............................................................................................................................................9 3.1.1 Smithsonite and ore samples .............................................................................................................................. 9 3.1.2 Heavy liquids ................................................................................................................................................... 10 3.1.3 Reagents ........................................................................................................................................................... 10

3.2 METHODS ..............................................................................................................................................11 3.2.1 Characterization of Ore Samples....................................................................................................................... 11 3.2.1.1 Mineralogical Studies ............................................................................................................................... 11 3.2.1.2 Liberation Characteristics Studies ............................................................................................................ 11 3.2.1.3 Grindability Studies.................................................................................................................................. 11 3.2.1.4 Heavy Liquid Tests................................................................................................................................... 12 3.2.1.5 Surface and Bulk Composition Studies .................................................................................................... 12 3.2.2 Experimental Methods...................................................................................................................................... 12 3.2.2.1 X-ray Powder Diffraction Measurement................................................................................................... 12 3.2.2.2 Zeta-potential Measurements ................................................................................................................... 13 3.2.2.3 Contact Angle Measurements................................................................................................................... 13 3.2.2.4 Diffuse Reflectance FTIR Measurements................................................................................................. 13 3.2.2.5 X-ray Photoelectron Spectroscopy (XPS) Measurements......................................................................... 14 3.2.2.6 Microflotation Tests ................................................................................................................................. 14 3.2.2.7 Bench Scale Flotation Tests of Ore Samples ............................................................................................ 14

4. RESULTS AND DISCUSSION .................................................................................................................15

4.1 MATERIAL CHARACTERIZATION............................................................................................................15 4.1.1 Mineralogical Studies........................................................................................................................................ 15 4.1.2 Liberation Characteristics ................................................................................................................................. 15 4.1.3 Grindability Studies ........................................................................................................................................... 17 4.1.4 Heavy Liquid Separation Tests .......................................................................................................................... 17 4.1.5 Surface and Bulk Composition........................................................................................................................... 18 4.1.6 X-ray Powder Diffraction Measurements........................................................................................................... 21

4.2 PHYSICO-CHEMICAL STUDIES WITH CATIONIC COLLECTOR ....................................................................22 4.2.1 Zeta-Potential Measurements............................................................................................................................. 22 4.2.1.1 Effect of sodium sulphide .......................................................................................................................... 22 4.2.1.2 Effect of dodecylamine .............................................................................................................................. 23 4.2.2 Contact Angle Measurements ............................................................................................................................ 24 4.2.3 Adsorption Studies............................................................................................................................................. 24 4.2.3.1 Diffuse Reflectance FTIR Measurements .................................................................................................. 24 4.2.3.2 X-Ray Photoelectron Spectroscopy (XPS) Measurements......................................................................... 28 4.2.3.2.1 XPS Spectra of Smithsonite ............................................................................................................ 28 4.2.3.2.2 Adsorption of dodecylamine ........................................................................................................... 29

4.3 PHYSICO-CHEMICAL STUDIES WITH ANIONIC COLLECTORS.....................................................................31 4.3.1 Zeta-Potential Measurements .............................................................................................................................. 31 4.3.1.1 Effect of oleic acid ..................................................................................................................................... 31

V

4.3.1.2 Effect of KAX............................................................................................................................................ 31

4.3.2 Contact Angle Measurements............................................................................................................................... 32 4.3.3 Adsorption Studies ............................................................................................................................................... 32 4.3.3.1 Diffuse Reflectance FT-IR Measurements ...................................................................................................... 32 4.3.3.1.1 Adsorption of oleic acid .................................................................................................................. 32 4.3.3.1.2 Adsorption of hexylmercaptan ........................................................................................................ 36 4.3.3.1.3 Adsorption of KAX......................................................................................................................... 38 4.3.3.2 X-Ray Photoelectron Spectroscopy (XPS) Measurements.............................................................................. 40 4.3.3.2.1 Adsorption of oleic acid .................................................................................................................. 40 4.3.3.2.2 Adsorption of KAX......................................................................................................................... 42 4.4 PHYSICO-CHEMICAL STUDIES WITH MIXED COLLECTOR ............................................................................43 4.4.1 Zeta-Potential Measurements................................................................................................................................ 43 4.4.2 Contact Angle Measurements ............................................................................................................................... 44 4.4.3 Adsorption Studies................................................................................................................................................ 44

4.5 MICROFLOTATION TESTS ..........................................................................................................................48 4.5.1 Effect of dodecylamine .......................................................................................................................................... 48 4.5.2 Effect of oleic acid ................................................................................................................................................. 48 4.5.3 Effect of hexylmercaptan ....................................................................................................................................... 48 4.5.4 Effect of KAX........................................................................................................................................................ 49 4.5.5 Effect of mixed collector........................................................................................................................................ 51

4.6 BENCH FLOTATION TESTS OF ORE SAMPLES .............................................................................................52 4.6.1 Effect of dodecylamine .......................................................................................................................................... 52 4.6.2 Effect of oleic acid ................................................................................................................................................. 53 4.6.3 Effect of KAX........................................................................................................................................................ 55 4.6.4 Effect of mixed collector........................................................................................................................................ 56 4.6.5 Design of Flowsheet............................................................................................................................................... 58

5. CONCLUSIONS.............................................................................................................................................61

6. PLAN FOR FUTURE WORK.....................................................................................................................61

References ...............................................................................................................................................................................62

Appendix

Paper I Paper II Paper III Paper IV Paper V Paper VI Paper VII

VI

1 Introduction Angooran oxide lead and zinc mine located in Zanjan province, Iran, is one of the greatest lead and zinc mine in the Middle East. The mine is affiliated to the mines & industry ministry, Islamic Republic of Iran. The lead and zinc beneficiation plant at Dandy with a capacity of about 140 tons per hour is located 100 km southwest of Zanjan town. Oxidised lead and zinc ore from the Angooran mine (situated at a distance of 20 km from the Dandy plant) is used as the feed for this plant. Oxide zinc minerals such as smithsonite, hemimorphite, willemite, hydrozincite and zincite have also long been an important source of zinc ores. Table 1 shows the different types of oxide zinc minerals and their zinc contents.

Table 1. The different kinds of oxide zinc minerals

Oxide Zinc Minerals Chemical formula Zn %

Smithsonite ZnCO3 52.2Hemimorphite Zn4Si2O7 (OH)2 (H2O) 54.3

Willemite Zn2SiO4 58.7

Hydrozincite Zn5 (CO3)2 (OH) 6 59.6

Zincite ZnO 80.3

There are several investigations on the relations between flotation behavior and surface properties of carbonate minerals (Rey, 1953; Rey et al., 1954; Gaudin, 1957; Abramov, 1961; Rey et al., 1961; Billi and Quai, 1963; Rausch and Mariacher, 1970; Glembotskii, 1972; Rey, 1979; McGarry and pacic, 1981; Bustamante and Shergold, 1983; Weiss, 1985; Ozbayoglu et al., 1994; Herrera et al., 1998 and 1999; Keqing et al., 2007). According to Bustamante and Shergold (1983), the flotation of oxide zinc minerals with conventional collectors, such as long chain alkyl amines and fatty acids is not very selective because there are some similarities in the surface chemistry of the zinc and gangue minerals. However, the difficulties of oxide zinc ore flotation with amines are numerous. Slimes, especially those produced from clay minerals, are detrimental to flotation (Rey, 1953). Xanthates, thiocarbonates, mercaptans and dithiocarbamates can collect oxidised lead, copper and zinc minerals. According to Gaudin (1957), the process is not selective enough for practice, when hexyl or amyl xanthate is used for collecting of smithsonite. Mixture of amines and xanthates can be also used as mixed collector for smithsonite flotation (Tarjan, 1986).The overall aims of this study are:

To characterize the zinc minerals from the Angooran Mine To investigate the amenability of the desired ore gravity concentration To define the surface and bulk composition of liberated smithsonite particles by means of EDX and XPS To define the physicochemical properties of smithsonite using zeta potential and contact angle measurements To determine the adsorption of some reagents on the smithsonite surface using spectroscopic methods e.g. FT-IR (DRIFT) and XPS To investigate the floatability of smithsonite and ore samples using microflotation and bench scale flotation tests

1

2 Background

2.1 Solubility and Stability of Oxide Zinc Minerals in Aqueous Media Investigation of oxide zinc minerals are based on the assumption of an analogy between complexing power of metal ions in solution and on the surface of the crystal lattice. There is, however, no complete analogy between the properties of metal ions in solution and on mineral surfaces, due to the various degrees of coordination saturation caused by surrounding anions. In the case of carbonate minerals e.g. smithsonite, the surface activity of mineral in water increases, and the adsorption of water molecules resembles chemisorption. This high activity to water dipoles is one of the main reasons for low natural floatability of smithsonite compared with sphalerite. Hence, there is a lower possibility of effective adsorption of flotation reagent molecules (Glembotskii, 1972). The hydrolysis of Zn2+ occurs by the following reactions and equilibrium constants (Hu et al., 1995):

ZnOHZn 2 OH K1'=105 (1)

) Zn(OH 2Zn 2(aq)2 OH K2'=1011.1 (2)

H) Zn(O 3Zn -3

2 OH K3'=1013.6 (3) H) Zn(O 4Zn -2

42 OH K4'=1014.8 (4)

Smithsonite is a semi-soluble type salt mineral with solubility product constant of 1.46 ×10 -10 M(Lide, 2005). Fig. 1 shows the solubility of smithsonite at different pH versus concentration, which is a replot of data cited by Stumm and Morgan (1970). According to the results depicted in Fig.1, smithsonite is also thermodynamically stable solid at pH values less than 7.4 but between this value and near to pH 12.4, smithsonite partially hydroxylates to hydrozincite and above pH 12.4, Zn (OH) 2 amorphous is predominant. Bustamante and Shergold (1983) have studied the fundamental solution chemistry of the zinc oxide minerals. According to their results, the dissolution of smithsonite in aqueous solution can be described by the following equilibria:

-2 3

2(S)3 CO Zn ZnCO (5)

According to Bustamante and Shergold (1983), the zinc oxide minerals e.g. smithsonite, hemimorphite and willemite are negatively charged at pH values greater than 5. When alkali is added to an aqueous smithsonite suspension (pH more than 9), zinc hydroxide is the stable solid. The obtained results are summarized in Fig. 2.

2.2 Solution Chemistry of Reagents Solution chemical equilibria are one of the most important factors in determining the behavior of minerals in flotation system. Mineral-solution equilibria and mineral-surfactant interactions can influence on the behavior of various species at the solid-liquid interface. The interactions of dissolved species with mineral surface should be considered in flotation system.

2

0

2

4

6

84 6 8 10 12 14

pH

-log

Zn

SmithsoniteZnCO 3

HydrozinciteZn 5(O H) 6(CO 3) 2

Zn(O H)

7.4

12.4

Fig. 1 Solubility products of ZnCO3(s) and (Zn5(OH) 6(CO3)2 (s))(Replotted from Stumm and Morgan, 1970)

Fig. 2 Logarithmic concentration versus pH diagram for smithsonite- hydrozincite- zinc hydroxide system in closed container (Bustamante and Shergold, 1983)

2.2.1 Cationic Collector (dodecylamine) Amines, amine salts and organic ammonium compounds are the common cationic collectors. Cationic collectors are very sensitive to the pH of the medium. Besides, they are most active in slightly acid solutions and inactive in strongly alkaline and acid media (Wills, 1997).

3

According to Aplan and Fuerstenau (1962), the solubility of molecular species of dodecylamine and CMC value are 5×10-5 M and 1.3×10-2 M respectively.Various equilibria of dodecylamine were suggested by Somasundaran and Wang (2006), as follows:

10 S RNH RNH -4.692(aq)2(s) (6)

104.3K OHRNH OH RNH -4b

-322(aq) (7)

37.3pK OH RNH OH RNH b22-

3 (8) 63.1037.314pK a bW pKpK

Fig. 3 represents the species distribution diagram of a 1×10-4 M dodecylamine solution. It was constructed using the thermodynamic data given as follows (Pugh, 1986):

69.4pK RNH RNH sp2(aq)2(s) (9)

63.10pK H RNH RNH a23 (10)

08.2pK RNH 2RNH D2233 (11)

12.3pK )RNH (RNH RNH RNH BD3223 (12)

The pH substantially changes the form of dodecylamine solution, predominating in equal amounts at pH values less than neutrality, and being present in equal

amounts at pH 10.6 and the quantity of dropping rapidly with increasing pH. According to Laskowski (1989), the pK

RNH 3

RNH 3 RNH 2(aq)

RNH 3

b value representing the dissociation of DDA ion (DAH+) to the non ionized species (DDA) is 10.6 such that with an increase in pH, the concentration of DAH is reduced. At high pH >10.6 at a concentration about 10-4 M, separation of amine can occur leading to droplets of liquid DDA and a water rich phase. The predominant dissolved species of dodecylamine below pH 10.6 is 3RNH and above the mentioned pH, .2(aq)RNH

Fig. 3 Species distribution diagram of dodecylamine at concentration of 1×10-4 M (Smith and Akhtar, 1976)

4

2.2.2 Anionic Collectors 2.2.2.1 Oleic acid The solubility of the fatty acids is low and varies according to the length of hydrocarbon chain. The important factors like dimerization equilibria, micellisation, and precipitation are involved and controlled by the solution pH, which controls the surface charge of the adsorption layer itself (Gaudin, 1957; Jung et al., 1987). According to Somasundaran and Wang (2006), the solution equilibria of oleic acid are as follows:

6.7(aq)(l) 10S HOl HOl (13)

4.95 pK Ol H HOl AD-

(aq) (14)

0.4pK Ol 2Ol D-2

2- (15)

4.7 pK HOl Ol HOl im-

2-

(aq) (16)

Fig. 4 depicts the species distribution diagram for concentrations of 10-4, 10-5 and 10-6 M oleic acid solution. Oleic acid forms in various species e.g. undissociated acid (RCOOH), oleate ion (RCOO-), oleate dimer and the acid soap .The predominant dissolved species of oleic acid at below pH 5 is HOl

-22(RCOO) H(RCOO) -

2

(aq) and above this pH is Ol-. In the acidic region, RCOOH (l) and RCOOH (aq) are the predominant species, whereas species such as (RCOO-), and to a lesser extent exist in the basic region. According to Shibata and Fuerstenau (2003), the pH values where oleic acid may start to precipitate are 8.3, 7.5 and 6.5 for the initial oleate concentrations of 10

-22(RCOO) H(RCOO) -

2

-4, 10-5 and 10-6 M, respectively. It is reported that the activities of the ionic monomer and ionic dimer increase up to the above mentioned pH and remains practically constant above it (Somasundaran and Ananthapadmanabhan, 1979; Vijaya et al., 2002).

pH

LOG

CO

NC

ENTR

ATI

ON

Fig. 4 Species distribution diagram of oleic acid at concentrations of 10-4, 10-5 and 10-6 M (Shibata and Fuerstenau, 2003)

2.2.2.2 HexylmercaptanHexylmercaptan with formula of CH3 [CH2]5 SH, is a weak acid and the ionization of a mercaptan in water is according to the following equation (Dalman and Gorin, 1960):

OH RS OH RSH 3-

2 (17)

5

The various equilibria of hexylmercaptan are proposed as follows: - H RS RSH (18)

10.6pK RSH / H RS K a-

a (19)

The solubility of hexylmercaptan in water is 3.2×10-4 mol/lit and the pKa is reported 10.6 (Dalman and Gorin, 1960).

2.2.2.3 Potassium amyl xanthate There are a few investigations regarding KAX solution chemistry. KAX is hydrolysed depending on the following equilibria (Guo and Yen, 2002):

CSKOH ROH OH ROCSSK 22 (20)

There are a large number of xanthate species which can be identified in the flotation pulp including xanthate ion (ROCS2), monothiocarbonate (ROCOS-), xanthic acid (ROCS2H), carbon disulphide (CS2), and dixanthogen (ROCS2S2COR)( Bulatovic, 2007).In sphalerite and KAX system, Zn(OH)AX(s) is predominant at pH 8-10 and below pH 8 Zn(AX)2(s) is predominant. Fig. 5 illustrates AX- predominance diagram for sulphide system with amyl xanthate (Pålsson and Forssberg, 1989). The possible mechanism of reactions for Zn ions and AX- can be expressed as follows:

2-2 Zn(AX)2AX Zn (21)

22 (AX)2ZnAX 2Zn(AX) (22)

Fig. 5 Amyl xanthate predominance diagram for sulphide system (Pålsson and Forssberg, 1989)

6

2.2.2.4 Mixed CollectorThe catanionic collectors are made from the mixtures of cationic and anionic surfactants. They are commonly used in many practical surfactant applications because the solution behaviors of mixed surfactants can be complementary and often behave synergistically and may provide more favorable properties than constituent single surfactants. The effect of a collector mixture depends on many factors such as the type of bonding of a collector, the structure of collector and their mixing ratio. When surfactants are added together in water, several physicochemical properties of the mixed system compared to those of the single surfactants system are changed due to the fact that there is a net interaction between the amphiphiles (Khan and Marques, 2000) and the interaction between oppositely charged surfactants will be strong due to electrostatic interactions and ion -pair association and complex coacervate formation will occur when they are mixed in aqueous solutions (Tomlinson et al., 1979). According to Helbig et al. (1998), a model for adsorption and solubility behavior of mixtures of anionic and cationic collectors is suggested. This model shows that the process and phenomena must be taken into consideration at the interface and also in the solution phase e.g. formation of a mixed film at the liquid-gas interface, formation of mixed micelles, interaction between anions and cations and formation of insoluble complexes, and precipitation of molecules by multivalent cations. Fig. 6 illustrates their proposed model.

Fig. 6 Model for the adsorption behavior of anionic and cationic mixtures collectors (Helbig et al., 1998)

2.3 Flotation of Oxide Zinc Minerals There are some methods that have been reported for the flotation of oxide minerals of the base metals. The most important methods which have long been used are as follows:

Sulphidization using sodium sulphide and flotation with a cationic collector such as amines , etc. (Castro et al., 1974; Yamada et al., 1976; Caproni et al., 1979; Önal et al., 2005; Pereira and Peres, 2005; Keqing et al., 2007) Using fatty acids (Rey et al., 1954; Nagano et al., 1974; Kiersznicki et al., 1981)

7

Sulphidization and activation using a metal ion in flotation with a sulphydryl collector e.g. xanthates (Gaudin, 1957; Billi and Quai, 1963; Nagano et al., 1975; Yamada et al., 1976; Yamazaki et al., 1978; Herrera et al., 1999) Other sulphydryl collectors such as mercaptans (Eyring and Wadsworth, 1956; Gaudin, 1957)Chelating agents and dyes (Rinelli and Marabini, 1973; Fuerstenau and Palmer, 1976; Barbery et al., 1977; Marabini et al., 2007)

Flotation of oxide zinc is much more difficult than the flotation of corresponding sulphide minerals. Furthermore, the solubility of oxide zinc minerals is high, thus greatly increasing the amount of reagent required for flotation (Marabini et al., 2007). The most common flotation technique used commercially is sulphidization with Na2S, followed by treatment with conventional collector, a long chain primary amine for oxide zinc minerals (Rey et al., 1954; Weiss, 1985). Several amines were tested but a dodecylamine derived from vegetable oil proved to be the most suitable (Billi and Quai, 1963). Sodium sulphide proved to be the most satisfactory sulphidizing agent, both from the point of view of being cheaper than sodium hydrogen sulphide and also by virtue of generating a high pH (Rausch and Mariacher, 1970). Other applicable sulphides include, for example, calcium sulphide, barium sulphide and ammonium polysulphide. Slime is one of the important factors affecting directly on the grade and recovery, as no selectivity is achieved without desliming. In order to decrease the extent of loss of valuable mineral contents, the slime is removed for oxide zinc minerals (McGarry and Pacic, 1981). The amount of sulphidizing reagent and pH of the pulp must be carefully controlled. The feed to the oxide zinc flotation circuit requires careful desliming prior to flotation to remove iron oxide and clay minerals (Glembotskii, 1972). Reagent consumptions are usually of the order of 1000 g/t to 7500 g/t sodium sulphide or sodium hydrosulphide, and 50 g/t to 300 g/t cationic collectors (Rey, 1953; Herrera et al., 1998). In smithsonite flotation, the equilibrium of the reaction is different, due to the fact that ZnS is much more soluble than PbS, so that Na2S is not consumed as in the case of cerussite sulphidization (Rey, 1979). The activating effect of sodium sulphide is strongly time dependent. The increase in sulphidization leads to an increase in the hydrophobicity of the surface of minerals. Excess of sodium sulphide acts as depressant for oxide lead and zinc minerals because adsorption of divalent sulphide ion on the surface of lead oxide minerals increase the negative charge which prevents the adsorption of collector (Ozbayoglu et al., 1994). Flotation by using carboxylic acid is applicable for oxide zinc ore, if the gangue is silica or clay minerals (Gaudin, 1957). If there is limestone or dolomite as gangue minerals, these are also floatable during smithsonite flotation (Rey, 1979). Laurylamine acetate as a collector can be used after sulphidizing the mineral with sodium sulphide but the flotation properties of minerals depend on the pH (Abramov, 1961). Mercaptans with roughly six carbon atoms have reasonable solubility in water and can be used as a collector for oxide base minerals e.g. chalcocite, azurite, malachite, calamine, zincite, smithsonite and willemite. However, the mercaptan consumption is so large to float oxide zinc minerals (Gaudin, 1957; Aplan and De Bruyn, 1963). The application of the mixture of amines and xanthates were investigated as mixed collector in combination of solvents, emulsifiers or organic colloid e.g. starch, tannin, etc. These are added to the pulp in order to regulate froth quality or to minimize the harmful effects of fine slimes (Tarjan, 1986).

8

3 Materials and Methods3.1 Materials 3.1.1 Smithsonite and ore samples The smithsonite and oxide zinc ore samples were collected from the Angooran zinc ore, Iran. Table 2 shows the chemical analysis of smithsonite samples using X-ray Fluorescence (XRF). The mineral samples were wet ground in porcelain mill. The products were wet-sieved to obtain -125 +75 μm size fractions. A portion of -75 μm was further ground and size classification by gravity sedimentation to obtain -5 μm size fractions. The coarser size fractions of -125 +75 μm were employed for microflotation tests. The final product (-5 μm) was used for XRD, FTIR, XPS, zeta-potential and contact angle measurements.

Table 2. Result of XRF for smithsonite samples

Oxides Weight %

ZnO 58.50Na2O 3.41 PbO 1.06CaO 0.84

Fe 2O3 0.26SiO2 0.29L.O.I 35.64

Tables 3-4 represent the chemical analysis of ore samples from Angooran mine, Iran using wet chemical analysis and XRF. The ore sample, after being crushed, was ground in a laboratory stainless steel ball mill at a critical speed of 80%. After grinding, the ore samples were deslimed at -30 μm size using a sedimentation technique and were used for bench scale flotation tests.

Table 3. Result of wet chemical tests for oxide zinc ore samples

Oxides Weight %

SiO2 62.5Al2O3 16.5K2O 2.25TiO2 0.45PbO 1.36

Fe 2O3 1.85ZnO 9.60

Others 5.49

9

Table 4. Result of XRF tests for oxide zinc ore samples

Oxides Weight %

Elements( Oxides)

Weight %

Elements(Oxides)

Weight %

SiO2 61.90 F 0.21 MnO 0.02Al2O3 15.20 As2O3 0.18 V2O5 0.02ZnO 8.90 CdO 0.14 ZrO2 0.01K2O 2.15 S 0.10 Cl 0.01

Fe2O3 1.95 NiO 0.05 Sb2O3 0.01PbO 0.85 P2O5 0.05 La2O3 0.01TiO2 0.65 Cr2O3 0.04 Rb2O 0.01MgO 0.60 Co3O4 0.03 SrO 0.01CaO 0.30 WO3 0.03 L.O.I 6.57

For bulk and surface composition studies, a representative sample has been used to determine the chemical surface and bulk composition of the particles including some minerals in three size fractions of - 250 +200 μm; -200 +150 μm and -150μm. For this purpose, the wet sieve analysis of the ore samples is employed using standard sieves i.e. 150, 200, 250 μm. There are a number of methods that can be used to mount the powders for analysis with XPS. Perhaps the most widely used method is to use of conductive tape. Since the ore sample used is non-conductive, the samples were prepared and coated with a thin layer of carbon on EDX analyzer. Metallographic embedding, polishing, and sectioning are available for samples requiring special preparation. Samples are usually mounted and coated and introduced into the vacuum chamber (Goodhew, 1983). The specific surface area of a sample was measured by Flow Sorb II 2300, Micromeritics Co., Ltd., USA. This instrument is designed to take the measurement on bone-dried powders by N2gas adsorption and desorption in liquid nitrogen temperature and room temperature (BET), respectively. A representative amount of sample was dried in an oven at 110 oC for 24 hours in order to remove the residual moisture in the sample prior to measurements. The mean value of the adsorption and desorption specific surface areas of a sample was regarded as its real value. The measured surface area was 0.43 m2/g.

3.1.2 Heavy liquids Carbon tetrachloride S.G 1.59, tribromoethane S.G 2.80, tetrabromoethane S.G 2.96 and methylene Iodide S.G 3.31 were used in heavy liquid tests. Liquid having densities 2.7, 2.6 were obtained by mixing different proportions of tribromoethane and carbon tetrachloride.

3.1.3 ReagentsDodecylamine and hexanethiol were obtained from Fluka Chemie, Switzerland. Potassium amyl xanthate was purchased from Shandong Qixia Flotation Reagent Co. Ltd, China and was purified by recrystallization from acetone. Oleic acid and sodium sulphide (Na2S) and copper sulphate were procured from Merck KGaA, Germany. Pine oil used as a frother from Hung Kuk Co., China. Analar grade HCl and NaOH were used for pH adjustment in all experiments. Table 5 represents the purity of used reagents in this work. Deionised water (specific conductance, 0.4-0.7 μS.cm-1) was used in all experiments and tap water was used for bench flotation tests. Table 6 indicates the chemical analysis of tap water used for bench scale flotation tests (Total hardness of Tehran drinking water as CaCO3 is 300 g/m3).

10

Table 5. Chemical reagents used in this study

Reagents Purity %

Dodecylamine 99.0Hexanethiol 95.0

KAX 90.0Oleic acid 99.9

Sodium sulphide 99.0Copper sulphate 99.0

Pine oil 99

Table 6. Chemical analysis of used tap water for bench flotation tests

Elements Element (g/m3)

Fe 0.5Cu 0.75Pb 0.03Mn 0.3Zn 2.25Ca 30Mg 50

3.2 Methods3.2.1 Characterization of Ore Samples 3.2.1.1 Mineralogical Studies The mineralogical properties of ore sample were determined by means of different methods e.g. XRF, XRD, wet chemical assay and also ore microscopy.

3.2.1.2 Liberation Characteristics Studies The liberation characteristics of ore samples were complex and extremely difficult to quantify because it was not easy to distinguish between the minerals microscopically. The characteristic sample has been collected from the sieve analysis of the above sample. These were then points counted under the microscope to find the degree of liberation where % liberation was defined as

the ratio of 100Particles All

Particles Liberated . In this method, all the numbers of particles on the cross

wire under the microscope were counted.

3.2.1.3 Grindability Studies The grindability of the ore samples is an important factor which can not be deduced from the mineralogical characteristics. Many theories have been proposed to determine the amount of required energy in a comminution operation as a function of the nature of material, size and shape of the minerals before and after of comminution. Rittinger (1867), Kick (1885), and Bond (1952) proposed the most important comminution laws. Since there are some problems with the standard Bond grindability test such as requiring a special mill, requiring a relatively large

11

amount of sample and also being relatively time consuming and tedious, in this work, a new size ball mill (NSBM) was used for determining the Bond work index. The ball mill used was scaled down with a coefficient of two-thirds from the standard Bond ball mill (200×200mm). The following equation was obtained to compute the Bond work index (Nematollahi, 1994):

F10

P10

10.75

iG

10.23

iP

11.76iW (23)

Where P¡ is the sieve opening at which the test is carried out (μm); G¡ is the NSBM grindability, net grams of ball mill product passing sieve size P¡ produced per mill revolution (g/rev); F is the feed size based on 80% passed (μm); and P is the product size based on 80% passed (μm).

3.2.1.4 Heavy Liquid Tests Generally heavy liquid tests indicate what theoretically can be expected from gravity separations. Chemical analysis of products for Pb and Zn do not, however give much information about the liberation of lead and zinc minerals. Chemical assay of sink products on the other hand, does not tell whether the heavy minerals are liberated from each other. For this reason degree of liberation was further studied using other methods. A representative ore sample from size fractions -212 +150; -150 +125 and -125 +75 μm were subjected to heavy liquid tests. The feed and product of used density fractions 3.31, 2.96, 2.8, 2.7, 2.6 were assayed for lead, zinc and SiO2.The curve by plotting cumulative float weight % Vs. S.G fractions shows that the ability of using the gravity separation for the examined sample; the slope of the curve indicates this ability. On the other hand, the steeper slope of the curve shows the ability of gravity separation (Gruender, 1963).

3.2.1.5 Surface and Bulk Composition Studies The X-ray photoelectron spectroscopy (XPS), also known as Electron Spectroscopy for Chemical Analysis (ESCA) is used to determine the surface composition of oxide lead & zinc minerals. This method determines the chemical composition of a surface using the photoelectric effect. XPS can determine the bonding state and/or oxidation states of materials and surface concentrations.In bulk composition studies, an energy-dispersive X-ray analyzer (EDX) is a common accessory, which gives the SEM a very valuable capability for bulk chemical analysis (Hren et al., 1979).

3.2.2 Experimental Methods3.2.2.1 X-ray Powder Diffraction Measurement XRD data have been recorded for smithsonite mineral and ore samples by means of a Siemens D5000 X-ray diffractometer. The instrument is equipped with a scintillation counter using Cu Kradiation and graphite monochromator. The scanning rate of the goniometer was 1.2 grad.min-1.This data is represented in a collection of single-phase X-ray powder diffraction patterns for the three most intense D values in the form of tables of interplanar spacings (D), relative intensities (I/Io) and hkl plane. This data is then analyzed for the reflection angle to calculate the inter-atomic spacing (D value in Angstrom units, 10-10 m). The intensity (I) is measured to discriminate (using I ratios) the various D spacings and the results are compared to this table to identify possible matches.

12

3.2.2.2 Zeta-potential Measurements The zeta potential of smithsonite fines in aqueous solutions was determined using a Coulter DELSA 440 instrument, which is based on the electrophoretic light scattering (ELS) technique, to directly measure the velocity of particles moving in an electric field and also zeta potential. In this study, the mineral suspensions of smithsonite finer than 5 μm with 0.1% solids were first conditioned for 2-4 minutes together with pH regulating reagents i.e. sodium hydroxide, HCl in the presence of different activators e.g. sodium sulphide, copper sulphate, if needed and then for 5-10 minutes with different reagents e.g. dodecylamine, oleic acid, hexylmercaptan, KAX and mixed cationic/anionic collector. It was dispersed in an ultrasonic cleaner for an additional minute. The suspensions were then transferred to the rectangular capillary cell of the instrument. The measurement was performed using an electric field of 10-20 V/cm, a frequency range of 500 Hz and duration of 100 (using 2 s on and 1 s off sequence altering the electric field polarity) at the upper stationary layers of the cell. The average value of zeta-potential from the four angles was reported as a final result. The temperature was kept at 25.0 ± 0.1 °C through the whole measurements.

3.2.2.3 Contact Angle Measurements Contact angle measurements are a powerful tool to characterize solids with respect to their wettability and their surface tension. The qualitative technique is using of the pressed powder as a pellet with the thickness about 1mm (Laskowski and Ralston, 1992). The Fibro DAT1100, Dynamic adsorption tester, Fibro System AB-Sweden, was used to measure the contact angle of the minerals. This instrument allows detailed, automatic analysis at full video speed of the interaction between a liquid droplet and a specimen surface. The preparation of mineral powders is as the same for zeta-potential measurements. Since the surface of the sample has to be flat, the treated minerals were compressed in a pellet and then transferred to the instrument. The interaction between a liquid droplet and a specimen surface can be accurately measured with millisecond precision within the first second of contact between the mineral surface and the liquid. Contact angles can be measured as a function of time.

3.2.2.4 Diffuse Reflectance FTIR Measurements In this study, the infrared spectra were recorded for all samples on the air-dried -5 μm powders before recording the DRIFT infrared spectrum. The FT-IR spectra were obtained with a Perkin Elmer Spectrum 2000 FTIR-Diffuse reflectance spectrometer. The Perkin Elmer Spectrum 2000 FTIR Spectrometer is capable of data collection over a wave number range of 370-7800 cm-1.This instrument can be configured to run in single-beam, ratio or interferogram modes. About 10 % by the weight of the solid samples were mixed with spectroscopic grade KBr with a refractive index of 1.559 and a particle size of 1-5 μm. These spectra were recorded with 200 scans measured at 4 cm-1 resolution. The adsorption studies of collectors were carried out using FT-IR spectra. The adsorption density was calculated according the measured specific surface area of smithsonite (0.43 m2/g). The statistical surface coverage ( ) of smithsonite can be calculated using Eq. 24 as follows (Kongolo et al., 2004):

= Qads. E. N/ Sp (24)

Where is the statistical surface coverage of the mineral using different collectors, Qads is the amount of collector adsorbed on mineral surface (mol/g), E is the cross-sectional area of the collector (m2), N is the Avogadro number, 6.022×1023 ,and Sp is the specific surface area of the mineral (m2/g).

13

3.2.2.5 X-ray Photoelectron Spectroscopy (XPS) Measurements X-ray photoelectron spectroscopy (XPS) is one of the major techniques in basic research of flotation-related surface studies of different minerals. The advantages of the method, in general, are good surface sensivity, rather straightforward elemental and chemical state analysis and reliable quantification of the data. The XPS spectra were obtained with an Axis Ultra electron spectrometer manufactured by Kratos Analytical. Ltd, UK.The vacuum in the sample analysis chamber was 10-9 Torr. A value of 285.0 eV was adopted as standard C(1s) binding energy. The measurements were performed on samples in DRIFT measurements.

3.2.2.6 Microflotation Tests Experiments were conducted with a 100 ml Hallimond tube. Generally one gram smithsonite sample with particle size of -125 +75 μm was conditioned for 2-4 min with sodium sulphide or copper sulphate, if needed. Then the suspension was conditioned in predetermined concentration of different reagents solution at the given pH for 5-10 minutes and transferred into the Hallimond tube and diluted to about 1% solids. The flotation was carried out at a constant airflow rate of 20 cm3/min. The float and sink products were filtered, dried and weighed to determine the flotation recovery of the mineral being studied. In all cases, pH was adjusted with NaOH or HCl in a beaker before conditioning with sulphidizing agents or collectors.

3.2.2.7 Bench Scale Flotation Tests of Ore Samples The bench flotation tests were run on oxide zinc ore samples. The experiments were conducted on 540 g samples in Denver flotation cell (1.5 liter) operating at 1500 rpm with a constant flow of air (the maximum permitted by the system). The ore sample, after being crushed, was wet ground in a laboratory stainless steel ball mill having a critical speed of 80%. Tap water was used at 20% solids, and the d80 of the products was around 100 μm. The various amounts of reagents used in oxide zinc flotation, were added at different rates of concentration. The conditioning time was 5-10 minutes. Sodium sulphide was added with conditioning time of 2-4 minutes at different concentrations. The copper sulphate was added 1500 g/t with 5 minutes of conditioning time. The pine oil as a frother and sodium silicate (100 g/t) were used with a conditioning time of 2 minutes. Flotation was conducted for 30 minutes, and the pH of the pulp was adjusted using HCl and NaOH for all flotation tests. The model with rectangular distribution of floatabilities has been applied in the evaluation of flotation results. This model is evaluated by fitting the flotation results from batch flotation tests. The mathematical form of this model can be written as follows (Huber-Panu et al., 1976 and Klimpel, 1980):

R= R (1-1/ kt [1-exp(-kt)]) (25)

Where R is the percentage recovery of the mineral at any time interval t; R , is the maximum percentage recovery of the mineral at infinite time t and k is the flotation rate constant (min-1).The R and k values for flotation recoveries have been estimated statistically and from the curves, the optimum flotation time has been determined graphically.

14

4 Results and Discussion 4.1 Material Characterization 4.1.1 Mineralogical Studies After preparation of thin and polished sections from the collected samples and observing it under optical microscope, the minerals of quartz, mica, sericite, smithsonite and clay minerals (Kaolinite) as major minerals and also the minerals of mimetite(Pb5AsO4)3Cl), pyrite(FeS2), iron oxides (goethite), rutile(TiO2), cerussite(PbCO3), galena(PbS) and sphalerite(ZnS) as minor minerals have been identified. Fig. 7 shows the thin section of ore samples. The smithsonite has been connected with quartz, sericite, mica and opaque minerals. Fig. 8 illustrates the polished section of ore sample. XRD studies have shown that quartz, sericite, smithsonite, kaolinite and mimetite have been identified in these ore samples. Combining the information obtained from chemical analysis, XRD and ore microscopy, the approximate mineralogical composition of the oxide zinc ore sample can be shown as follows:

Quartz 52.0 % Sericite 16.5 % Smithsonite 14.0 % Kaolinite 12.0 % Other zinc minerals 2.0 % Iron oxide (FeOOH) 1.5 % Lead minerals 0.6 % Rutile 0.4 % Other minerals 1.0 %

4.1.2 Liberation Characteristics The results show that the best degree of liberation, which the crushed product has not been overground, will be the size fractions of -212 +150; -150 +125 and -125 +75 μm. Fig. 9 depicts the smithsonite particles locked into quartz particle. The comparison between the size fractions for liberation studies is represented in Fig. 10.

15

51

3

2

4

Fig. 7 Thin section including 1-Smithsonite 2-Sericite 3- Mica 4- Quartz 5-Opaque minerals.

(LP -200 X)

μ 100 m

Fig. 8 Polished section of the ore sample including Iron oxide, rutile and pyrite. (PPL 32 × 8 Oil + Nicole)

μ 100 m

Fig. 9 Locked smithsonite particles in quartz

16

50 75 100 150 200 300 400 600 800 1200 2000 4000 0

10

100

Mean Particle Size (Micron)

Deg

ree

of L

iber

atio

n (%

)

Y1 = 100 exp(-0.0016181x)

Y2 = 100 exp(-0.0002178x)

Y3 =100 exp(-0.0012141x)

Y4 = 100 exp(-0.0005802x)

Degree of Liberation(Smithsonite) Degree of Liberation (Quartz) Degree of Liberation (Iron Oxide) Degree of Liberation (Schist) Mi l

Fig. 10 Comparison between degrees of liberation for ore samples

4.1.3 Heavy Liquid Separation Tests Fig. 11 illustrates the Henry Reinhardt diagram for size fraction -150 +125 μm. The results

of heavy liquid tests indicate that the most amount of curve slope occurs the more ability of gravity concentration for the size fraction -150+125 m as compared to other size fractions (paper VII, Figure 7).

Fig. 11 Henry Reinhardt diagram for –150 +125 μm

4.1.4 Grindability Studies The Bond Ball Mill Work Index provides a measure of how much energy is required to grind a sample of ore in a ball mill. The results of grindability studies for calculation of Bond index are

17

illustrated in Table 7. Based on the calculations of Bond Ball Mill Work Index (11.9 kWh/t), the oxidised ore can be categorized in medium category (9-14 kWh/t).

Table 7.Results of Bond work index

Feed Particle Size

(d80)(μm)

Product particle Size

(d80)(μm)

MeanGi(g)

P1(μm)

Work Index

F P GiP1 Wi

1800 122 0.36 150 11.9

4.1.5 Surface and Bulk Composition StudiesTables 8-10 indicate the results of XPS for three size fractions of ore samples and also Table 11 shows the result of EDX for finest sample. Figures 12 and 13 illustrate the peaks obtained on XPS and EDX spectra, respectively.

Table 8. Results of XPS on ore sample (-150 m)

Element Area(cts-eV/s)

Sensitivity Factor

Atomic Concentration (%)

Weight(%)

Fe2p3O1sAl2pSi2pClsK2pCa2pZn2p3F1sMg1s

6595352649 12476 35533 24347 13270 574185586 19141786

1.791 0.711 0.193 0.283 0.296 1.300 1.634 3.354 1.000 1.433

0.4560.89 7.9415.41 10.10 1.250.433.130.230.15

1481021621

100.50.5

Table 9. Results of XPS on ore sample (-200 +150 m)

Element Area(cts-eV/s)

Sensitivity Factor

Atomic Concentration (%)

Weight(%)

Fe2p3O1sAl2pSi2pClsK2pCa2pZn2p3F1sMg1s

4783263502 874427782 13485 9557427768699 15731311

1.791 0.711 0.193 0.283 0.296 1.300 1.634 3.354 1.000 1.433

0.4562.26 7.6116.49 7.651.240.443.440.260.15

3471022421

110.50.5

18

Table 10. Results of XPS on ore sample (- 250 +200 m)

Element Area(cts-eV/s)

Sensitivity Factor

Atomic Concentration (%)

Weight(%)

Fe2p3O1sAl2pSi2pClsK2pCa2pZn2p3F1sMg1s

5288272991 946829168 13531 9642421264855 1284964

1.791 0.711 0.193 0.283 0.296 1.300 1.634 3.354 1.00

0.4862.33 7.9616.73 7.421.200.423.140.210.11

1481023421

100.50.5

According to the obtained results, it can be found that these samples have no significant variations in elemental composition for three size fractions. The results of XPS show that there is no difference especially in Zn % and also for other elements in three size fractions. It can be concluded that the surface composition are approximately the same for all three size fractions. Table 12 represents a comparison between the data obtained from XPS (Surface composition) and EDX analysis (Bulk composition) so that there is some differences in the contents of Si, Fe and Al elements in the ore sample and these differences between XPS and other techniques should be based on the amount of oxygen for each technique (Paper VI).

Fig. 12 ESCA spectra of ore sample (-200 +150 m)

19

Fig. 13 Peak results of EDX on ore sample (-150 m)

Table 11. The results for bulk chemical analysis using EDX on ore sample (-150 m)

Element AREA AREA/BGND

%CONC. NORM.%

Al K 3788± 105 3.032± 0.412 23.72 29.41

Ti K 223± 55 0.129± 0.032 0.62 0.76

Si K 10586± 156 7.274± 0.692 41.47 51.42

Fe K 638± 71 0.531± 0.061 2.36 2.92

Zn K 1635± 102 2.256± 0.169 12.50 15.49

Table 12. The results for bulk & surface chemical analysis of ore samples

Ore sample (-150 m)

Ore sample (-200+150 m)

Ore sample (- 250+200 m)

Weight % Weight % Weight % Analysis Technique Si Al Zn Fe Ti Si Al Zn Fe Ti Si Al Zn Fe Ti

EDX 51.42 29.41 15.49 2.92 0.76 -- -- -- -- -- -- -- -- -- --

XPS(ESCA)

21 10 10 1 --- 22 9.73 11 3 --- 23 10 10 1 ---

20

4.1.6 X-ray Powder Diffraction MeasurementsThe strongest line in the X-ray powder pattern reference is 2.75(1), and the others are 3.558(0.5), 1.703(0.45), 2.328(0.25), 2.109(0.25) and 1.948(0.25). Fig. 14 and Table 13 illustrate the results of XRD measurements and library data for Siemens XRD instrument.

Fig. 14 XRD pattern of smithsonite sample

Table 13. XRD results for smithsonite sample

Interplanar spacings(d) experiment data

Interplanar spacings(d) Siemens library data

1.3609 1.35681.3769 1.37341.4131 1.40961.4954 1.49411.5175 1.51431.7089 1.7027(0.45) D (I/I )(hkl)3 o1.7802 1.77541.9493 1.94812.1142 2.10922.3304 2.32802.7603 2.7502(1) D (I/I )(hkl)1 o3.5621 3.558 (0.5)D (I/I )(hkl)2 o

21

4.2 Physico-chemical Studies with Cationic Collector 4.2.1 Zeta-Potential Measurements 4.2.1.1 Effect of Sodium Sulphide The sulphidization studies were carried out on 0.5 g of smithsonite samples in sodium sulphide solution at 2.6×10-2 M concentration. The electrokinetic behavior of smithsonite in distilled water, sodium sulphide solution and ZnS are shown in Fig. 15.

-60

-50

-40

-30

-20

-10

0

10

20

30

40

2 4 6 8 10 12 14

pure smithsonite

Smith+Na2S 2 g/l

ZnS(10-6 M NaCl

pH

4 6 8 10 12 14

Pure smithsoniteSmithsonite +Na2SZnS (10-6 M NaCl)

Zet

a po

tent

ial (

mV

)

2

Fig. 15 Zeta potential of smithsonite as a function of pH in distilled water, sodium sulphide solution, 2.6× 10-2 M and ZnS (10-6 M NaCl), Sphalerite data was replotted from Zhang et al., 1995.

Fig. 16 shows the different species of Na2S at various pH (Jones and Woodcock, 1978). The reaction can be caused by HS-, which predominates in pH range of 7-13. According to the obtained results from zeta potential measurements, the formation of ZnS partially is observed which is more negatively charged than unsulphidized smithsonite. Adding various amounts of sodium sulphide makes their zeta potential more negative and the value of the isoelectric point decreases. The pHIEP of ZnS is around 2. The pHIEP of smithsonite drops from 8 to 6.3 because chemical reactions of HS- with metal ions on the surface of minerals can occur and to form the metal sulphide film on the surface of smithsonite. Oversulphidization of pulp (3 g/l) is due to high HS- ion content, which are responsible for depression of the mineral and decreases the flotation recovery and controlled sulphidization appears to be the preferred method of adding sodium sulphide.

22

The most probable reaction on smithsonite is between HS- , which predominates in the pH range of 7-12, and surface layers of smithsonite, hydrozincite or zinc hydroxide, depending on the pH as follows:

HCO ZnS HS ZnCO -3(surf)

- (surf)3

(26)

OH OH ZnSHS Zn(OH) -2(surf)

- (surf)2

(27)

According to Malghan (1986), and Önal et al. (2005), the over sulphidization of pulp is due to high HS- ion content, which is responsible for depression of the mineral and decreases the flotation recovery.

Fig. 16 Species distribution diagram of sodium sulphide in aqueous solution as a function of pH (Jones and Woodcock, 1978)

4.2.1.2 Effect of Dodecylamine Generally the collectors are used as adsorbents at the solid-liquid and liquid-gas interface in order to change the surface charge and impart hydrophobicity to the mineral surface (Hunter, 1993). The adsorbed ions can be identified by their ability to reverse the sign of the zeta-potential. It is obvious that a physically adsorbed ion does not affect the pzc (i.e.p) but can reverse the sign of the potential, while a chemisorbed ion shifts the pzc and can remain adsorbed even when the underlying surface has the same sign as itself. The specifically adsorbed anion tends to make zeta-potential more negative, and a more positive surface potential is required to offset the effect (Hunter, 1981). Fig. 17 illustrates the zeta potential results of smithsonite as a function of pH in the presence of Na2S (2.6×10-2 M) at various concentrations of DDA.

23

-50

-40

-30

-20

-10

0

10

20

30

40

4 5 6 7 8 9 10 11 12 13

pH

Zeta

-pot

entia

l(mV)

Pure SmithsoniteZnCO3+ Na2S 2 g/lDDA 0.1 g/l +Na2S 2 g/lDDA 0.2 g/l +Na2S 2 g/lDDA 0.3 g/l +Na2S 2 g/l

Pure SmithsoniteZnCO3+Na2S ( 2.6×10-2 M )DDA( 5.4×10 -4 M )+ Na2SDDA (1.1×10 -3 M )+ Na2SDDA (1.6×10 -3 M )+ Na2S

Fig. 17 Zeta-potential of smithsonite as a function of pH in the presence of Na2S (2.6× 10-2 M) and DDA

According to Fig. 17, the surface charge of smithsonite is less negative at the highest DDA concentration (1.6×10-3 M), at pH 11.5. The zeta potential value decreases from - 47.8 to -26.75 mV which indicates the adsorption of cationic collector on the surface of the mineral. However, no significant variation of zeta potential between various concentrations is observed. These results reveal that the primary amine (DDA) is adsorbed predominantly on the negatively charged of smithsonite surface at alkaline pH after sulphidization with sodium sulphide.

4.2.2 Contact Angle Measurements Fig. 18 shows the result of contact angle measurements for smithsonite sample as a function of various dodecylamine concentrations in the presence of sodium sulphide. Hence, the highest contact angle using DDA attained was 115 º.

4.2.3 Adsorption Studies 4.2.3.1 Diffuse Reflectance FTIR Measurements The reference spectra of smithsonite, dodecylamine, potassium amyl xanthate, sodium oleate, hexylmercaptan, zinc oleate and zinc sulphide are presented in Fig. 19. The smithsonite spectrum displays several bands in the region 4000-400 cm-1. The infrared band at around 744 cm-1 (Ferraro, 1982; Frost et al., 2007; Hales and Frost, 2007) is assigned to the 4

in phase bending mode. Two bands are observed at 868.7 and 840 cm-1(841 cm-1: Gadsden, 1975 and 870 cm-1: Gadsden, 1975; Ferraro, 1982; Jones and Jackson, 1993; 864: Hales and Frost, 2007). These bands are assigned to the carbonate 2 bending mode. An intense broad infrared band at 1435.46 cm-1 is assigned to the 3CO3

2- antisymmetric stretching vibration (1427 cm-1:Ferraro, 1982; Jones and Jackson, 1993; 1440 cm-1: Farmer, 1974; Gadsden, 1975; Frost et al.,

24

2008). The second sets of bands occur at 2492, 2850 and 2920 cm-1. It is proposed that these bands are due to overtone and combination bands. For example the band at 2492 cm-1 is a combination of the 1 and 3 vibrational modes (Frost et al., 2007).

0

20

40

60

80

100

120

1.00E-05 3.10E-04 6.10E-04 9.10E-04 1.21E-03 1.51E-03 1.81E-03

DDA concentration(mol/l)

Con

tact

ang

le (d

egre

e)

DDA

DDA +Na2S (1 g/l)

DDA +Na2S (2 g/l)

DDA +Na2S (3 g/l)

DDA +Na2S (8 g/l)

DDA +Na2S (15 g/l)

10-5 3×10-4-4 6×10-4-4 1.5×10-3-41.2×10-3-49×10-4-4 1.8×10-3-4

without Na2SNa2S 1.3×10-2 MNa2S 2.6×10-2 MNa2S 3.9×10-2 MNa2S 10-1 MNa2S 1.9×10-1 M

Fig. 18 Contact angles of smithsonite as a function of various dodecylamine and sodium sulphide concentrations

The carbonate, hydroxyl carbonate, and hydroxide species are released, when smithsonite mineral is immersed in water. H2O groups physically adsorb on the surface of the mineral. The amount of physically adsorbed water decreased upon Na2S treatment which transforms into ZnS. According to Marabini and Rinelli (1986), there are no significant variations after sulphidization and this is expected because the ZnS and ZnCO3 bands are very similar. The ZnCO3 species after sulphidization will disappear and forms the ZnS coating in the case of very high sodium sulphide concentrations in the form of one monolayer or less.

25

Fig. 19 Reference DRIFT spectra of smithsonite, dodecylamine, KAX, sodium oleate, hexylmercaptan, zinc oleate and zinc sulphide

Fig. 20 displays the spectra of smithsonite sample treated with different concentrations of dodecylamine at pH 10.5-11. The spectra exhibit intense absorption band corresponding to (NH2)at a fixed concentration of sodium sulphide (2.6× 10-2 M). Table 14 shows the band assignments before and after adsorption of dodecylamine on to smithsonite surface. The reagent may be linked with Zn2+ ions through coordination bonds formed by N atoms and adsorbs to the smithsonite surface in the form of Zn-amine complexes or perhaps the hydroxyl ions presents as zinc hydroxyl species on the surface of smithsonite (Pascal, 1962; Healy and Moignard, 1976). Generally at the flotation pH of around 11, it can be assumed that the RNH2 becomes attached to the zinc on the surface in both the ZnCO3 and ZnS form through complexation bonds as follows (Marabini et al., 1984):

(28)

n

n

OHSorCO

ZnSorCO

RNHSorCO

)(|

|)(

3

3

23

26

Fig. 20 DRIFT spectra of smithsonite treated with sodium sulphide (2.6× 10-2 M) at pH 11.5 with increasing initial concentration of dodecylamine (a) 10-5 M b) 5.3×10-4 M c) 1.1×10-3 (d) 1.6×10-3 M

Table 14. Bands assignment before and after dodecylamine adsorption on smithsonite surface in the presence of sodium sulphide

Vibrating Mode Before adsorption Wavenumber

(cm-1)

After adsorption Wavenumber

(cm-1)4(CO3

-2) 744 ------2(CO3

-2) 869 ------(CO3

-2) Stretch 1092 ------3 (CO3

-2) 1435.5 1435Combination

bands2492, 2850, 2920 ------

(NH2) ------ 3330(CH3) ------ 2952

as(CH2) ------ 2916

s(CH2) ------ 2850(CN) stretch ------ 1043, 1660

(NH) bending ------ 1598

27

The adsorption model schematically is illustrated for dodecylamine on smithsonite surface after sulphidization with sodium sulphide in Fig. 21.The area under the alkyl chain bands (3000-2800 cm-1) and the adsorption density of DDA as a function of DDA concentrations are shown in Fig. 22. The results of DDA adsorption study by means of DRIFT spectra show the increasing DDA adsorption density with increasing initial concentration of dodecylamine from 0.63 mg/m2 to 5.95 mg/m2. The adsorbed amount of DDA is reached to 1.46×10-6 mol/g at low DDA concentration of 10-5 mol/l and the statistical surface coverage is calculated to be about =0.5assuming the cross-sectional of area of a vertically oriented amine species to be 25 ×10-20 m2

(Sabah et al., 2002) but it progressively increases up to 2.37×10-6 mol/g for an equilibrium DDA concentration of 1.35×10-4 mol/l ( =0.8), then the adsorbed amount of DDA increases rapidly to 5.95 ×10-6 mol/g at DDA concentration 2.7×10-4 mol/l ( >1). At higher DDA concentration (5.39×10-4 < C < 1.62×10-3 mol/l), the adsorbed amounts increase and reach to 1.38×10-5 mol/g ( >2).

Fig. 21 Schematic adsorption model for dodecylamine on smithsonite treated with sodium sulphide

4.2.3.2 X-Ray Photoelectron Spectroscopy (XPS) Measurements4.2.3.2.1 XPS Spectra of Smithsonite The XPS spectral results of smithsonite demonstrate the appearance of the Zn (2p), C (1s) and O (1s). The Zn (2p), C (1s) and O (1s) spectra of the smithsonite are presented in Fig. 23. The curve fitting unveiled that the Zn (2p) spectra consist of two components, at 1022.1 and 1045.1 eV which the both components are assigned to zinc in ZnO bond (Maroie et al., 1984). The C (1s) spectra show the signal at 285, 290 eV which is assigned to carbon in C-O and C=O in carbonate groups (Bichler et al., 1996). The O (1s) spectra at 532 eV which is relevant to C-O and C=O in carbonate groups (Bou et al., 1991). The atomic percentage of XPS spectral results shows the element percent of Zn 17.11%, C 6.65 %, and O 56.19 % (Paper III).

28

0

1

2

3

4

5

6

7

8

9

10

0.00E+00 5.00E-04 1.00E-03 1.50E-03 2.00E-03

DDA concentration (mol/l)

Alk

yl g

roup

s Are

a( c

m-1)

0

1

2

3

4

5

6

7

8

9

10

Adsorption density (m

g/ m 2)DDA AreaDDA+Na2S AreaAdsorption Density DDAAdsorption Density DDA+Na2S

0 5×10-4 10-3 1.5×10-3 2×10-3

Fig. 22 Adsorption density and area under alkyl chain bands (2990-2800 cm-1) of DRIFTspectra of smithsonite as a function of dodecylamine concentration at pH 11.5

in the presence or absence of sodium sulphide

4.2.3.2.2 Adsorption of Dodecylamine XPS analysis of samples treated with DDA shows nitrogen to be present on the surface of sulphidized sample and this shows that DDA is present with the formation of a DDA layer on the smithsonite surface. The C (1s), N (1s), S (2p) spectra of the smithsonite treated with DDA (1.62×10 -3 M) are presented in Fig. 24. The curve fitting unveiled that the C (1s) spectra consist of two components, at 285 (reference peak) and 288.1 eV. The first component is assigned to carbon in C-(C, H) bond and second one to the carbon in the C-N (Delpeux et al., 1998). The N (1s) signal of the amine groups at 399.6 eV can be assigned to nitrogen in R-NH2 bond and confirms the existent of DDA on the surface of mineral (Lim and Atrens, 1990). The S (2p) spectra of sample conditioned with 1.62×10 -3 MDDA show the broad peak at around 162.1 eV confirms that ZnS is present on the mineral surfaces (Brion, 1980).The XPS spectra results of smithsonite conditioned with Na2S (2.6×10 -2

M) and DDA (1.62×10 -3 M) demonstrate the appearance of the N (1s) signal of the amine groups and S (2p) signal of ZnS which increased in the intensity of the signal of C (1s) peak by adsorption of DDA on smithsonite with a simultaneous decrease in the peak intensities of oxygen.

29

Fig. 23 XPS Zn (2p), C (1s) and O (1s) spectra of smithsonite

Fig. 24 XPS spectra of smithsonite conditioned in solution of DDA at pH 11.5 after sulphidization with sodium sulphide

30

4.3 Physico-chemical Studies with Anionic Collectors 4.3.1 Zeta-Potential Measurements 4.3.1.1 Effect of Oleic acidFig. 25 represents the zeta potential of smithsonite at various pH values in the presence of oleic acid. As it can be seen, the negative charge values increased from -35 to -55.5 mV for oleic acid concentration (1.1×10-3 M); this is due to the adsorption of oleate anion at pH 10. It can be concluded that at higher concentrations the adsorption increases due to higher negative charge and subsequent adsorption of oleate anion at pH 10.

-70

-60

-50

-40

-30

-20

-10

0

10

20

30

40

0 2 4 6 8 10 12 14

Pure Smithsonite

OA 0.1 g/l

OA 0.2 g/l

OA 0.3 g/l

2 4 6 8 10 12 140

Pure smithsoniteOle ic acid, 3.5×10 -4MOle ic acid, 7.1×10 -4MOle ic acid, 1.1×10 -3M

Zet

a po

tent

ial (

mV

)

pH

Fig. 25 Zeta potential of smithsonite as a function of pH in the presence of oleic acid

4.3.1.2 Effect of KAX The electrokinetic potential studies were performed on smithsonite in distilled water and in the presence of Na2S, CuSO4 and various concentrations of KAX. The results are illustrated in Fig. 26. The isoelectric point of smithsonite is around 8.0, which agrees well with those reported in literatures (Quaresima, et al., 1991; Hu, et al., 1995). The reaction can be occurred between zinc and HS- , which predominates in pH range of 7-13. The zeta potential results that were obtained with sulphidized smithsonite show that ZnS is more negatively charged than unsulphidized smithsonite. The surface species formed during copper activation are covellite (CuS) and chalcocite (Cu2S). Some investigators studied that the activation product is covellite, while others suggested that it is chalcocite. The formation of CuS or Cu2S is the possible for this reaction. In basic solutions, most of the copper added as activator is precipitated as Cu(OH)2. According to Laskowski et al.,

31

(1997), the hydroxide slowly releases Cu2+ ions into solution, which in turn form a flotation active product such as (Zn, Cu)S. According to zeta potential measurements, it can be assumed that at higher concentrations more negative zeta potentials lead to higher adsorption. The negative charge of the surface is due to the adsorption of KAX anion are consistent with the contact angle and also microflotation results.The results of electrokinetic studies confirm that the sulphydryl collector (KAX) is adsorbed predominantly on the negatively charged surface of smithsonite at alkaline pH.

-60

-50

-40

-30

-20

-10

0

10

20

30

40

0 2 4 6 8 10 12 14

pure smithsonite

Smith+Na2S 2 g/l

KAX 0.1 g/l

KAX 0.3 g/l

KAX 0.6 g/l

0 2 4 6 8 10 12 14

Pure smithsoniteSmithsonite +Na2SKAX ( 4.94×10 -4 M )KAX ( 1.48×10 -3 M )KAX ( 2.96×10 -3 M )

Zet

a po

tent

ial (

mV

)

pH

Fig. 26 Zeta potential of smithsonite; smithsonite and Na2S; and smithsonite, Na2S,CuSO4 and KAX as a function of pH

4.3.2 Contact Angle Measurements The measured contact angle using oleic acid was 105º. The highest measured contact angle for KAX and hexylmercaptan were 99º and 92º respectively. Figures 27-29 illustrate the results of contact angle measurements.

4.3.3 Adsorption Studies 4.3.3.1 Diffuse Reflectance FT-IR Measurements 4.3.3.1.1 Adsorption of Oleic acid Fig. 30 illustrates the spectra of smithsonite treated with increasing initial concentration of oleic acid in the region 3400-1000 cm-1. Table 15 shows the band assignments before and after adsorption of oleic acid on to smithsonite surface. The peaks at 2924 and 2852 cm-1 (Socrates, 1980; Miller and Kellar, 1999) show that a long alkyl chain is present in oleic acid treated smithsonite samples. The appearance of peak 3411 cm-

1 shows that -OH on the surface of smithsonite has not reacted completely. It is known that RCOOH (l) and RCOOH (aq) are the predominant species, whereas species such as RCOO-, (RCOO)2

-2 and a lesser extent [(RCOO)2H] - exist in the basic region. The maximum adsorption of oleic acid on the smithsonite surface occurs around pH 10. It can be attributed to

32

interaction between RCOO- and zinc on the smithsonite surface and can be assumed that the adsorption of oleic acid takes place by an ion exchange mechanism as represented below:

OH RCOO Zn RCOO Zn(OH) -(surf)

--(surf)2 (29)

0

20

40

60

80

100

120

0.00E+00 2.00E-04 4.00E-04 6.00E-04 8.00E-04 1.00E-03 1.20E-03

Oleic acid concentration ( mol/l)

Con

tact

ang

le( d

egre

e)

0 2×10-4 4×10-4 6×10-4 8×10-4 10-3 1.2×10-3

Fig. 27 Contact angles of smithsonite as a function of various oleic acid concentrations at pH 10

30

40

50

60

70

80

90

100

0.00E+00 5.00E-04 1.00E-03 1.50E-03 2.00E-03 2.50E-03 3.00E-03

KAX concentration (mol/l)

Con

tact

ang

le (d

egre

e)

contact angle CuSO4 1 g/lcontact angle CuSO4 1.5 g/lcontact angle CuSO4 2 g/lcontact angle CuSO4 2.5 g/l

0 5×10-410-3 1.5×10-3 2×10-3 2.5×10-3 3 ×10-3

CuSO4, 6.3×10-3 M CuSO4, 9.4×10-3 MCuSO4, 1.3 ×10-2 MCuSO4, 1.6×10-2 M

Fig. 28 Contact angles of smithsonite as a function of various KAX concentrations in sodium sulphide solution (2.6×10-2 M) with copper sulphate at pH 10.5

33

Accordingly, the frequencies of the bands of carboxylate ion should show intermediate values between C=O and C-O. Bands between 1538-1650 cm-1 and 1360-1470 cm-1 (Smith, 1998) were assigned due to asymmetric and symmetric stretching vibrations of carboxylate ion. The asymmetric carboxylate vibration band can be attributed to chemisorbed oleate (Gong et al., 1992; Jang et al., 1995). If the oleate is present in the form of undissociated oleic acid (-COOH), the mean frequency of

C=O stretching vibration should appear around 1690 cm-1 for dimer and 1718 cm-1 for monomer. In the present investigation, the absence of band around 1690 cm-1 indicates that the dimer adsorption is not present.

0

10

20

30

40

50

60

70

80

90

100

0.00E+00 2.00E-03 4.00E-03 6.00E-03 8.00E-03 1.00E-02 1.20E-02

Hexylmercaptan concentration (mol/l)

Con

tact

ang

le (d

egre

e)

2×10-30 4×10-3 6×10-3 8×10-3 10-2 1.2×10-2

Fig. 29 Contact angles of smithsonite in the presence of hexylmercaptan at pH 9

Table 15. Bands assignment before and after oleic acid adsorption on smithsonite surface

Vibrating Mode Before adsorption Wavenumber

(cm-1)

After adsorption Wavenumber

(cm-1)4(CO3

-2) 744 -----2(CO3

-2) 869 ------(CO3

-2) Stretch 1092 10913(CO3

-2) 1435.5 1455Combination

bands2492, 2850, 2920 ------

as(COO-) ------ 1547(CH3) ------ 2946

as(CH2) ------ 2920

s(CH2) ------ 2850(C=O) ------ 1713

34

The adsorption model schematically is illustrated for oleic acid on smithsonite in Fig. 31. The area under the alkyl chain bands (3000- 2800 cm-1) and the adsorption density of oleic acid as a function of oleic acid concentrations are illustrated in Fig. 32.

Fig. 30 DRIFT spectra of smithsonite at pH 10 with increasing initial concentration of oleic acid (a) 4.42×10-5 M (b) 8.85×10-5 M (c) 1.77×10-4 M (d) 3.54×10-4 M (e) 5.31×10-5 M (f) 1.1×10-3 M

Fig. 31 Schematic adsorption model for oleic acid on smithsonite surface

35

The results of oleic acid adsorption studies by means of DRIFT spectra show the increasing oleic acid adsorption density with increasing initial concentration of oleic acid from 0.63 mg/m2 to 15.1 mg/m2. The adsorbed amount of oleate is reached to 9.6×10-7 mol/g at low oleate concentration of 4.42×10-5 mol/l and the statistical surface coverage is calculated to be about =0.25 assuming the cross-sectional of area of a vertically oriented oleate species to be 2 ×10-19

m2 (Shibata and Fuerstenau, 2003) but it progressively increases up to 3.09×10-6 mol/g for an equilibrium OA concentration of 8.85×10-5 mol/l ( =0.85), then the adsorbed amount of OA increases rapidly to 6.7 ×10-6 mol/g at OA concentration 1.77×10-4 mol/l ( >1). At higher OA concentration (3.54×10-4 < C < 1.1×10-3 mol/l), the adsorbed amounts increase and reach to 2.3×10-5 mol/g ( >2).

0

5

10

15

20

25

0 0.0002 0.0004 0.0006 0.0008 0.001 0.0012

Oleic acid concentration(mol/l)

Alk

yl g

roup

are

a (c

m-1)

0

4

8

12

16

20

Adsorption density( m

g/m 2)

Alkyl group areaAdsorption density

0 2×10-4 4×10-4 6×10-4 8×10-4 10-3 1.2×10-3

Fig. 32 Adsorption density and area under alkyl chain bands (2990-2800 cm-1) of DRIFT spectra of smithsonite as a function of oleic acid concentration at pH 10

4.3.3.1.2 Adsorption of hexylmercaptan Fig. 33 depicts the spectra of zinc mineral at pH 9 at various concentrations of hexylmercaptan. The bands related to the hydrocarbon chain assigned at 2850, 2952 cm-1. Since the S-H bond is weaker than C-H and has a lower stretching absorption 2600-2500 cm-1 compared to 2960-2840 cm-1 for the C-H bond(Leja, 1982). The poor double peak at 2940-2850 cm-1 is caused by C-H stretching frequencies associated with the methyl (CH3) and ethylene (CH2) groups (Tarjan, 1986). Clearly in all spectra the surface S-H stretching (2530 cm-1) bands are missing. Although, the band due to the S-H stretching vibration can be weak and even may be missed in dilute solution. The lack of S-H stretching bands and poor bands of C-H stretching is confirmed by some authors’ works (Eyring and Wadsworth, 1956; Roberts and Friend, 1988). They stated that the S-H bond in the mercaptan is destroyed upon adsorption. Gaudin and Harris (1954) studied the adsorption of hexylmercaptan on sphalerite, zincite, and willemite from aqueous solution, zinc minerals and between aqueous solution and the gaseous phase. They showed that the adsorption of mercaptan from the vapor phase prior to flotation

36

testing were effective in causing flotation and the adsorption of mercaptan corresponding to less than monolayer coverage. The presence of surface –OH groups is properly associated with satisfying of unsaturated valence forces when a surface is freshly fractured. The following reaction is represented for zinc carbonate at pH 9.

(30)33 HCO

OHZn

OHZn

HOHCO

Zn

Zn

The adsorbed mercaptan reacts with the surface –OH group forming the zinc mercaptan salt, splitting out a molecule of water in the process according to the following reaction (Cook and Nixon, 1950):

OHSRZnRSHOHZn 2 (31)

The mechanism for the adsorption of mercaptan on to malachite and chrysocolla could be similar to the mentioned reaction and the process could be extended to the attachment of mercaptan to any base-metal oxide or sulphide mineral (Aplan and Fuerstenau, D.W., 1984). The key point of the mercaptan adsorption process is the formation of a strong sulfur-metal (S-M) bond (Sardar et al., 2004).

Fig. 33 FT-IR spectra of smithsonite treated at pH 9 with hexylmercaptan concentrations (a) 8.4×10-4 M (b) 2.54×10-3 M (c) 4.2×10-3 M

(d) 6.77×10-3 M (e) 8.5×10-3 M (f) 10-2M (g) 1.1×10 -2 M

37

The adsorption model schematically is illustrated for hexylmercaptan on smithsonite surface in Fig. 34. The area under the alkyl chain bands (3000-2800 cm-1) and the adsorption density of hexylmercaptan as a function of various HM concentrations are shown in Fig.35. The adsorption density was calculated according to the measured specific surface area of smithsonite (0.43 m2/g). The results of hexylmercaptan adsorption studies by means of DRIFT spectra show the increasing HM adsorption density with increasing initial concentration of hexylmercaptan from 9.77 mg/m2 to 34.05 mg/m2. The adsorbed amount of hexanethiol is reached to 3.5×10-5 mol/g at low HM concentration of 8.46×10-4 mol/l and the statistical surface coverage is calculated to be about =0.5 assuming the cross-sectional of area of a vertically oriented hexanethiol species to be 2.5 ×10-21 m2 (Sui et al., 2003) but it progressively increases up to 6.19×10-5 mol/g for an equilibrium HM concentration of 2.54×10-3 mol/l ( =1), then the adsorbed amount of HM increases rapidly to 7.5×10-5 mol/g at HM concentration 4.23×10-3 mol/l ( >2). At higher HM concentration (6.77×10-3 < C < 1.1×10-2 mol/l), the adsorbed amounts increase and reach to 1.24×10-4 mol/g ( >3).

Fig. 34 Schematic adsorption model for hexylmercaptan on smithsonite surface

4.3.3.1.3 Adsorption of KAX Fig. 36 depicts the FTIR spectra of smithsonite treated with different concentrations of KAX in the presence of sodium sulphide and copper sulphate solutions. Clearly in all obtained spectra the surface, the dixanthogen (AX) 2 (1270-1240 cm-1) (Leja, 1982) bands are missing. For all surface coverage values, the spectra exhibit absorption bands at 1037 cm-1, assigned to the asymmetric stretching vibration of CS2 (Leja, 1982; Socrates, 1980) and 1137 cm-1, characteristics of C-O-C stretching in dixanthogen (Leja, 1982).The bands related to the hydrocarbon chain assigned at 2856, 2956 cm-1 (Socrates, 1980; Miller and Kellar, 1999). Table 16 shows the band assignments before and after adsorption of KAX in the presence of sodium sulphide and copper sulphate on to smithsonite surface. The interaction of amyl xanthate with the surface of cerussite represented by the following reaction (Fuerstenau et al., 1985; Fleming, 1953):

COPbAX 2AXPbCO (aq)-232(s)

-(aq)3(s) (32)

According to Equation 32, it can be assumed that the ion-exchange reaction produces solid zinc xanthate at the mineral surface and releases carbonate ions to the solution.

38

0

5

10

15

20

25

30

35

0.00E+00

2.00E-03

4.00E-03

6.00E-03

8.00E-03

1.00E-02

1.20E-02

Hexylmercaptan concentration (mol/l)

Ads

orpt

ion

dens

ity (m

g/m2 )

0

2

4

6

8

10

12

14

16

18

20

Alkyl groups area(cm

-1)

Adsorption densityAlkyl groups area

2×10-3 4×10-3 6×10-3 8×10-3 10-2 1.2×10-20

Fig. 35 Adsorption density and area under alkyl chain bands (2990-2800 cm-1) of DRIFT spectra of smithsonite as a function of hexylmercaptan concentrations at pH 9

Fig. 36 FT-IR spectra of smithsonite treated with sodium sulphide (2.6×10-2 M) and copper sulphate (9.4×10-3M) at pH 10.5 with increasing initial concentration of KAX (a) 4.94×10-4 M

(b) 9.88×10-4 M (c) 1.48×10-3 M(d) 1.98×10-3 M (e) 2.47×10-3M (d) 2.96×10-3M

39

The adsorption model schematically is illustrated for KAX on smithsonite surface after sulphidization with sodium sulphide and activation with copper sulphate in Fig. 37.

Fig. 37 Schematic adsorption model for amyl xanthate on smithsonite surface in the presence of sodium sulphide and copper sulphide

The area under the alkyl chain bands (3000-2800 cm-1) and the adsorption density of KAX as a function of KAX concentrations are shown in Fig.38. The adsorption density was calculated according the measured specific surface area of smithsonite (0.43 m2/g). The results of KAX adsorption study by means of DRIFT spectra show the increasing KAX adsorption density with increasing initial concentration of potassium amyl xanthate from 0.35 mg/m2 (4.94×10-4 M) to 9.35 mg/m2. The adsorbed amount of amyl xanthate is reached to 9.2×10-7 mol/g at low KAX concentration of 4.94×10-4 mol/l and the statistical surface coverage is calculated to be about =0.3 assuming the cross-sectional of area of a vertically oriented amyl xanthate species to be 29

×10-20 m2 (Kongolo et al., 2004) but it progressively increases up to 2.37×10-6 mol/g for an equilibrium KAX concentration of 9.88×10-4 mol/l ( =0.96), then the adsorbed amount of KAX increases rapidly to 8.7×10-6 mol/g at KAX concentration 1.48×10-3 mol/l ( >3). At higher KAX concentration (1.98×10-3 < C < 2.96×10-3 mol/l), the adsorbed amounts increase and reach to 2.46×10-5 mol/g ( >11).

4.3.3.2 X-Ray Photoelectron Spectroscopy (XPS) Measurements4.3.3.2.1 Adsorption of Oleic acid The XPS C (1s) and O (1s) spectra of smithsonite conditioned in a solution of oleic acid at pH 10 are shown in Fig. 39. The C (1s) spectra consist of peaks 285 eV (reference peak) and 290.2 eV, corresponding to alkyl and carboxylate carbon respectively. The O (1s) spectra measured consist of the peak 533.5 eV, corresponding to O (1s) in carboxylate group (Clark and Thomas, 1978). It is obvious that the most of the adsorption occurs at around pH 10 and this is because of the existence of RCOO- is predominant in solution and has interaction with mineral surface.

40

0

2

4

6

8

10

12

14

16

0.00E+00

1.00E-03

2.00E-03

3.00E-03

4.00E-03

KAX concentration (mol/l)

Ads

orpt

ion

dens

ity (m

g/m2 )

0

2

4

6

8

10

12

Alkyl groups area(cm

-1)

Adsorption densityAlkyl groups area

0 10-3 2×10-3 3×10-3 4×10-3

Fig. 38 Adsorption density and area under alkyl chain bands (2990-2800 cm-1) of DRIFT spectra of smithsonite as a function of KAX concentration at pH 10.5 in sodium sulphide (2.6×10-2 M) and copper

sulphate (9.4×10-3 M) solutions CS2

Table 16. Bands assignment before and after KAX adsorption on smithsonite surface in the presence of sodium sulphide and copper sulphate

VibratingMode

Beforeadsorption

Wavenumber (cm-1)

Afteradsorption

Wavenumber (cm-1)

4(CO3-2) 744 -----

2(CO3-2) 869 ------

(CO3-2)

Stretch1092 1091

3(CO3-2) 1435.5 1455

Combination bands

2492, 2850, 2920

------

(C-O-C) ------ 1137(CH3) ------ 2956

as(CH2) ------ 2917

s(CH2) ------ 2856

s(CS2) ------ 1037

as(CS2) ------ 1100

41

Fig. 39 XPS C (1s) and O (1s) spectra of smithsonite conditioned in solution of OA at pH 10

4.3.3.2.2 Adsorption of KAX The XPS S (2p) and Cu (2p) spectral results of smithsonite conditioned in solution of KAX (2.96 × 10-3 M) in the presence of sodium sulphide and copper sulphate at pH 10.5 are illustrated in Fig. 40. The C (1s) spectra consist of peaks 285 eV (reference peak), 286.4 and 288.4 eV, corresponding to alkyl, CS2 and CO carbon respectively (Liao et al., 1993; Wagner et al., 1979). The intensity of the characteristic peak in spectra for alkyl group (285.0) shows the increasing peak intensity in comparison to C (1s) spectra in smithsonite. It shows that there is adsorption to form a film at the surface of the mineral. These results are consistent with the contact angle and also microflotation results.The S (2p) spectra measured consist of the peak 161.6 eV, corresponding to S (2p) in ZnS layer onto the mineral surface (Briggs and Seah, 1993).The changes in the C1s and O1s emissions during the adsorption treatment show that adsorption of xanthate occurs. Slight changes in the positions and shapes of the components due to the xanthate group are observed during the adsorption process. The Cu (2p) spectra measured consist of the peak 932.1 eV, corresponding to Cu (2p) in CuS (Briggs and Seah, 1993) and 952.3 eV, corresponding to Cu2O (Maroie et al., 1984) layer on the mineral surface. The atomic concentration of the Cu peaks shows the higher concentration for CuS in comparison of Cu2O. It can be suggested that copper cations exchange with those of zinc during copper activation of smithsonite such as activation of sphalerite. It verifies the ion exchange of Zn2+ and Cu2+ and existence of CuS layer on the surface of mineral. The maximum adsorption occurs for the KAX concentration of 2.96×10-3 M at pH 10.5. The XPS analysis of samples treated with KAX shows that KAX is present with the formation of KAX layer on the smithsonite surface.

42

Fig. 40 XPS spectra of smithsonite conditioned in solution of KAX (2.96×10-3 M l), sodium sulphide (2.6× 10-2 M) and copper sulphate (9.4×10-3 M) at pH 10.5

4.4 Physico-chemical Studies With Mixed Collector 4.4.1 Zeta-Potential Measurements Fig. 41 shows the zeta potential of smithsonite as a function of pH in sodium sulphide solution and mixed collector (potassium amyl xanthate and dodecylamine). The isoelectric point of smithsonite is about 8.0, which agrees well with those reported by researchers (Quaresima, et al., 1991; Hu, et al., 1995). After addition of sodium sulphide, the solubility of smithsonite is reduced and is increased their negative surface charge. The pHIEP of smithsonite decreased from 8 to 6.3 because chemical reactions of HS- with metal ions on the surface of minerals can take place.

43

It is observed that after treatment with DDA at 1.1×10-3 M concentration, the zeta potential decreases from - 48 to -16.75 mV which shows the adsorption of cationic charge on the surface of the mineral. Hence, the zeta potential in mixed collector increased their negative charge with increasing the KAX concentration due to adsorption of C=S on the mineral surface. These results suggest that the mixed collector (dodecylamine and potassium amyl xanthate) is adsorbed predominantly on the negatively charged part of the smithsonite surface at alkaline pH after sulphidisation with sodium sulphide at alkaline pH.

-60

-50

-40

-30

-20

-10

0

10

20

30

40

4 5 6 7 8 9 10 11 12 13 14

pH

Zeta

pot

entia

l(mV)

in destilled water

in amine( 1.1*10-3M)and KAX(1.1*10-3M)

in amine(1.1*10-3M) and KAX (2.24*10-3M)

in amine(1.1* 10-3M) and KAx(3.3*10-3M)

in 2.96*10-3 M KAX

in 1.62*10-3 M Amine)

in distilled waterin DDA (1.1×10-3M) and KAX(1.1×10-3M)in DDA (1.1×10-3M) and KAX(2.2×10-3M)in DDA (1.1×10-3M) and KAX(3.6×10-3M)in 2.96×10-3M KAXin 1.1×10-3M DDA

Fig.41 Zeta-potential of smithsonite as a function of pH in the presence of distilled water, KAX only (2.96 ×10-3 M), DDA only (1.1 ×10-3 M) and mixed collector

(KAX+DDA) including 1.1 ×10-3 M of DDA and various KAX concentrations of 1.1 ×10-3 M, 2.2 ×10-3 M, and 3.6 ×10-3 M

4.4.2 Contact Angle MeasurementsFig. 42 show the results of contact angle measurements for smithsonite treated with the mixed collector. The contact angle measurements show the maximum value of 117º for mixed collector in DDA solution (1.1×10-3 M) in sodium sulphide solution at pH 12 and in absence of sodium sulphide at pH 9.5.

4.4.3 Adsorption Studies Fig. 43 represents the adsorption of KAX on the smithsonite surface at pH 9.5 in dodecylamine solution, 1.1×10-3 M in the alkyl group and functional group region, respectively. At the fixed dodecylamine concentration (1.1×10-3 M), the KAX concentration is progressively increased. Clearly in all obtained spectra the surface, the dixanthogen (AX)2 (1270-1240 cm-1) (Leja, 1982) bands are missing. For all cases, the spectra exhibit absorption bands at 1041cm-1, assigned to C=S stretching mode (Leja, 1982; Socrates, 1980) and 1090,1196 cm-1, characteristics of C-O-C stretching (Leja, 1982).The bands related to the hydrocarbon chain assigned at 2850, 2917 and 2950 cm-1 (Socrates, 1980; Miller and Kellar, 1999). The spectra also exhibit intense absorption band corresponding to (NH2) in the mixed collector. The FT-IR spectra revealed that no

44

adsorption of amyl xanthate when used alone but showed its coadsorption as amine-xanthate complex when used with dodecylamine. The presence of KAX increased the dodecylamine adsorption due to decrease in the electrostatic head-head repulsion between the two adjacent surface ammonium head group cations because the anionic KAX sitting between these two cations screens the electrostatic repulsion and increasing the lateral tail-tail hydrophobic bonds.

60

70

80

90

100

110

120

0.00E+00 1.00E-03 2.00E-03 3.00E-03 4.00E-03

KAX concentration (mol/l)

Con

tact

ang

le (d

egre

e)

in Amine (1.1*10-3M) andNa2s (2.96*10-2 M)in 1.1*10-3 M amine

10-3 2×10-3 3×10-3 4×10-30

in DDA (1.1×10-3M) and Na2S (2.6×10-2M)in 1.1×10-3M DDA

Fig. 42 Contact angles of smithsonite as a function of KAX concentration with DDA concentration 1.1×10-3 M in sodium sulphide solution (pH 12) and in absence of sodium sulphide at pH 9.5

Fig. 43 FTIR spectra of smithsonite in the presence of (a) 2.96 ×10-3 M of KAX only (b) 1.1 ×10-3 M of DDA only and mixed collector (KAX+DDA) including 1.1 ×10-3 M of DDA and variousKAX concentrations (c) 4.94 ×10-4 M (d) 9.88 ×10-4 M (e) 1.48 ×10-3 M (f) 1.98 ×10-3 M

(g) 2.47 ×10-3 M (h) 3.6 ×10-3 M, in the alkyl and functional group region at pH 9.5

45

Figs. 44 and 45 show the KAX adsorption on smithsonite at pH 12 in dodecylamine (1.1×10-3 M)and sodium sulphide solution (2.6 ×10-2 M), in the alkyl group and functional group region, respectively. The experiment was performed at the fixed dodecylamine concentration (1.1×10-3

M) and the KAX concentration is progressively increased. The dixanthogen (AX) 2 (1270-1240 cm-1) (Leja, 1982) bands are missing. The spectra exhibit absorption bands at 1041 cm-1,assigned to C=S stretching mode (Leja, 1982; Socrates, 1980) and 1090,1196 cm-1,characteristics of C-O-C stretching mode (Leja, 1982).The bands related to the hydrocarbon chain assigned at 2855, 2917 and 2950 cm-1 (Socrates, 1980; Miller and Kellar, 1999). The spectra also exhibit intense absorption band corresponding to (NH2) in the mixed collector. The FT-IR spectra show co-adsorption of the amine–xanthate complex using a collector mixture. The area under the alkyl chain bands (3000-2800 cm-1) of DDA only, KAX only and mixed collector at pH 9.5 as a function of KAX concentrations are shown in Fig.46. These results show that the areas under the alkyl chain bands are much lower for KAX and DDA when used alone. It also indicates the increase in area under the alkyl chain bands for the mixed collector with increasing initial concentration of KAX from 8.87 cm 1 (4.94 × 10 4 M) to 18.77 cm 1

(3.6 × 10 3 M) at pH 9.5. The area under alkyl chain bands at pH 12 in the presence of sodium sulphide shows the lesser adsorption bands. The initial concentrations of KAX are varied from 2.57 cm-1 (4.94× 10-4 M) to 7.3 cm -1 (2.96×10-3 M) (Fig. 47).

Fig. 44 FTIR spectra of smithsonite in the presence 1.1 ×10-3 M of DDA and sodium sulphide with various KAX concentrations in the alkyl group region at pH 12

46

Fig. 45 FTIR spectra of smithsonite in the presence 1.1 ×10-3 M of DDA and sodium sulphide with various KAX concentrations in the functional group region at pH 12

0

2

4

6

8

10

12

14

16

18

20

KAX/DDA concentration (mol/l)

Alk

yl g

roup

s are

a(cm

-1)

KAX onlyDDA onlyKAX+DDA

10-3 2×10-3 3×10-3 4×10-30

Fig. 46 The area under alkyl chain bands (2990-2800 cm-1) of DRIFT spectra of smithsonite in the presence of DDA only, KAX only and mixed collector (KAX+DDA) at pH 9.5

47

0

2

4

6

8

10

12

14

16

18

20

0.00E+00 1.00E-03 2.00E-03 3.00E-03 4.00E-03

KAX concentration (mol/l)

Alk

yl g

roup

s are

a (c

m-1

)

pH=9.5pH=12

10-30 2×10-3 3×10-3 4×10-3

Fig. 47 The area under alkyl chain bands (2990-2800 cm-1) of DRIFT spectra of smithsonite in the presence of DDA (1.1 ×10-3 M) at pH 9.5 and 12, in sodium sulphide solution as a function of KAX

concentrations

4.5 Microflotation Tests

4.5.1 Effect of Dodecylamine The results of microflotation tests for smithsonite with various concentrations of DDA in the absence and presence of various amounts of sodium sulphide are shown in Fig. 48. Flotation results revealed that zinc recovery sharply increased up to 94% at a sodium sulphide concentration of 2.6×10-2 M and a DDA concentration of 1.6×10-3 M and then gradually decreased such that the excess sodium sulphide could not improve the zinc recovery; instead, it results in a decrease of recovery. Therefore, this gradual decrease would be attributed to the depression of smithsonite due to reducing pulp potential values.

4.5.2 Effect of Oleic acid Fig. 49 shows the smithsonite flotation recoveries as a function of various concentration of oleic acid at pH 10. The flotation results revealed that increasing the oleic acid concentration to 1.1× 10-3 M causes higher recovery. Because pH of the medium is an important factor in flotation, its effect was investigated. The solution pH was varied from 8 to 11. The maximum recovery occurs at pH 10. The maximum flotation recovery using oleic acid was 93 %, which obtained at pH 10 when RCOO- is predominant in the solution. Flotation results show that the recovery increased with increasing collector concentration.

4.5.3 Effect of Hexylmercaptan Results of smithsonite microflotation tests with various amounts of hexylmercaptan are reported in Fig. 50. The maximum flotation recovery occurs at about 78 % at pH 9. As expected, the results show maximum flotation recovery occurs at highest concentration of hexylmercaptan (1.1×10-2 M) at pH 9 which are 78 %. According to Fig. 50, increasing HM concentration enhances the flotation recovery.

48

Fig. 48 Flotation results for smithsonite at various concentrations of DDA in the presence of Na2S at pH

11.5 and in the absence of Na2S at pH 9.5

4.5.4 Effect of KAX Microflotation tests were carried out in sodium sulphide solution (2.6×10-2 M) with various concentrations of copper sulphate and KAX. The flotation recoveries of smithsonite as a function of KAX concentration at pH 10.5 showed at lower KAX concentration (4.94×10-4 M), the rapid rise in smithsonite recovery begins and at higher level (2.6×10-3 M), the maximum flotation recovery occurs at 81% (Fig.51). In general, the recovery of smithsonite increases with increasing KAX concentration. The smithsonite recovery is enhanced from about 53 to 81% at higher KAX concentration. The results confirm that at lower KAX concentrations, smithsonite has low surface coverage and adding KAX concentration enhances smithsonite recovery.

49

0

10

20

30

40

50

60

70

80

90

100

0.00E+00 2.00E-04 4.00E-04 6.00E-04 8.00E-04 1.00E-03 1.20E-03

Oleic acid concentration ( mol/l)

Rec

over

y (%

)

0 2×10-4 4×10-4 6×10-4 8×10-4 10-3 1.2×10-3

Fig. 49 Flotation recoveries of smithsonite as a function of various concentrations of oleic acid at pH 10

0

10

20

30

40

50

60

70

80

90

100

0.00E+00 2.00E-03 4.00E-03 6.00E-03 8.00E-03 1.00E-02 1.20E-02

Hexylmercaptan concentration (M ol/l)

Rec

over

y (%

)

2×10-30 4×10-3 6×10-3 8×10-3 10-2 1.2×10-2

Fig. 50 Flotation recovery of smithsonite as a function of various amounts of Hexylmercaptan at pH 9

50

Fig. 51 Flotation recoveries of smithsonite as a function of various KAX concentrations in sodium sulphide solution (2.6×10-2 M) with copper sulphate at pH 10.5

4.5.5 Effect of Mixed Collector The flotation responses of smithsonite as a function of KAX in 1.1×10-3 M DDA concentration are given in Fig. 52. The results show that the smithsonite recovery increases with increasing the KAX concentration in both cases (pH 9.5 and 12). As, the maximum flotation recovery of smithsonite at pH 12 was 95% while at pH 9.5 amounts to 96% at high KAX concentration (2.96×10-3 M). However, there are no drastically significant variations for microflotation recoveries at pH 9.5 and pH 12 in high KAX concentration (2.96×10-3 M). The mole ratio of KAX: DDA and pH were found to be an important factor in the smithsonite flotation response.

0

10

20

30

40

50

60

70

80

90

100

0.00E+00 1.00E-03 2.00E-03 3.00E-03 4.00E-03KAX concentration (mol/l)

Rec

over

y(%

)

in amine(1.1*10-3M)and Na2S(2.96*10-2M)in 1.1*10-3 M amine

0 10-3 2×10-3 3×10-3 4×10-3

in DDA (1.1×10-3M) and Na2S (2.6×10-2M)in 1.1×10-3M DDA

Fig. 52 Flotation recovery of smithsonite as a function of KAX concentration with DDA concentration 1.1×10-3 M in Na2S solution (pH=12) and in absence of Na2S (pH=9.5), total concentration 3.87×10-3 M

51

4.6 Bench Flotation Tests of Ore Samples 4.6.1 Effect of Dodecylamine The zinc grade-recovery curves of bench flotation in terms of the quantity of collectors are given in Fig. 53. The results of oxide zinc ore flotation using various amounts of DDA in the presence of sodium sulphide show that the optimum concentration of Na2S is 2000 g/t. The recovery as well as grade fell sharply at 1000 g/t of Na2S. This indicates that this level of sulphidization is not sufficient for a feed containing high amount of smithsonite. When the amount of Na2S was increased, the recovery increased. At the highest concentration of Na2S (3000 g/t), the recovery fell off slightly, indicating depressing action of Na2S at high concentration (paper V). It is mentioned that controlling the exact amount of sodium sulphide is more important. It means that the sulphidization stage is critical in the sulphidization of oxide ores for flotation, because either too little or too much sodium sulphide gives poorer metallurgy than does an optimum addition (Castro et al., 1974). Barbaro, et al., 1997, also reported that the sulphidization step must be strictly controlled because an excessive dosage of sulphidizing agent results in drastic reduction in floatability. The optimum amount of the cationic collector was 300 g/t in the presence of Na2S; 2000 g/t at pH 11.5 with 84.5 % recovery and 24.5% zinc grade. The grade and the recovery of the smithsonite in the concentrate and the tailing in terms of the quantity of dodecylamine and sodium sulphide are given in Table 17. From Fig. 54, it can be concluded that the optimum flotation time is 15 min. The losses of zinc in tailing can not be ignored. According to the flotation results, it can be seen that the tailings were rather high in zinc content, ranging from 2.2 to 10 % Zn. It can be attributed this fact to the slimes because the slimes make the froth layers brittle and the recovery of coarse zinc minerals drops (Rey, 1953).

0

10

20

30

0 10 20 30 40 50 60 70 80 90 100

Zinc Recovery(%)

Zin

c G

rade

(%)

Na2S=1000;DDA=100 g/t Na2S=1000;DDA=200 g/t Na2S=1000;DDA=300 g/tNa2S=2000;DDA=100 g/t Na2S=2000;DDA=200 g/t Na2S=2000;DDA=300 g/tNa2S=3000;DDA=100 g/t Na2S=3000,DDA=200 g/t Na2S=3000;DDA=300 g/t

Fig. 53 Zinc grade-recovery curves for oxide zinc ore flotation investigating the effect of DDA in various amount of sodium sulphide in solution

Mineralogical studies of the tailings reveal that the oxide zinc minerals in the tailing are extremely finely disseminated smithsonite, which is not liberated from the gangues. Overall zinc flotation recovery with respect to the total zinc was estimated to be about 75 % according to the

52

contents of zinc in tailing and slimes. The particle size measurement shows that over 90% of smithsonite particles in slime are below 20 micron. This range of particle size is difficult to recover by froth flotation.

20

30

40

50

60

70

80

90

100

0 5 10 15 20 25 30 35

Rec

over

y(%

)

Flotation time(min)

Fig. 54 Zinc ore flotation at various flotation times (Na2S 2000g/t; DDA 300 g/t)

Table 17. Grade and recovery of bench flotation tests for various amounts of DDA and sodium sulphide

Na2S = 1000 g/t

Na2S = 2000 g/t Na2S = 3000 g/t

DDA (g/t) DDA (g/t) DDA (g/t)

Grade & Recovery

(%)100 200 300 100 200 300 100 200 300

34 45 44 42 68 84.5 34 62 64Recovery(%)

11.6 14.8 14 16.8 26.5 24.5 11.8 10 25.3 Zn % in Conc.

Zn % in Tails

8.8 7.4 7.7 7.3 4.0 2.2 8.7 9.0 4.6

4.6.2 Effect of Oleic acid The zinc grade-recovery curves of the bench flotation as a function of pH and reagent consumption are illustrated in Fig. 55. An examination of the results shows that the best flotation recovery in the concentrate was obtained at the collector concentration of 300 g/t at pH 10.It is seen that the recovery as well as grade enhances slightly at pH 9. The maximum flotation recovery occurs at pH 10. This indicates that the surface hydrophobicity was sufficient for a feed containing high amount of smithsonite. When increasing pH to 11, the recovery fell sharply, which indicating low adsorption of oleic acid on smithsonite surface. However, at the highest level of oleic acid addition at pH 10 the grade was low, which indicated, as expected, lower selectivity as the optimum concentration was exceeded.

53

However, it is expected according the obtained results from the Hallimond tube tests, zeta potential and contact angle measurements, FTIR and XPS studies that the optimum amount of the oleic acid as an anionic collector was 300 g/t at around pH 10 with 78.8 % recovery and 15.2 % zinc grade.Fig. 56 shows zinc ore flotation at various flotation times. The optimum flotation time is 20 min. The zinc grade and the flotation recovery in the concentrate and the tailing in terms of the quantity of oleic acid are given in Table 18.

Table 18.Grade and recovery of bench flotation tests with various oleic acid concentrations

pH 8 pH 9 pH 10 pH 11 Oleic acid (g/t) Oleic acid (g/t) Oleic acid (g/t) Oleic acid (g/t)

Grade & Recovery

(%) 100 200 300 100 200 300 100 200 300 100 200 300

Recovery(%)

68 71.6 74.8 73.8 75.2 72.8 72.8 76 78.8 48 49.8 54.6

Zn % in Conc.

15.4 15.3 17.1 15.3 16.3 16.2 15.6 15.5 15.2 14.3 14.7 14.1

Zn % in Tails

5.3 4.9 4.2 4.7 4.3 4.6 4.7 4.4 4.1 7.4 7.1 6.9

8

10

12

14

16

18

20

0 10 20 30 40 50 60 70 80 90 100

Zinc Recovery(%)

Zin

c G

rade

(%)

pH=8;OA=100 g/t pH=8;OA=200 g/t pH=8;OA=300 g/t pH=9;OA=100 g/tpH=9;OA=200 g/t pH=9;OA=300 g/t pH=10;OA=100 g/t pH=10;OA=200 g/tpH=10;OA=300 g/t pH=11;OA=100 g/t pH=11;OA=200 g/t pH=11;OA=300 g/t

Fig. 55 Zinc grade-recovery curves for oxide zinc ore flotation investigating the effect of oleic acid at different pH

54

20

30

40

50

60

70

80

90

100

0 5 10 15 20 25 30 35

Rec

over

y(%

)

Flotation time(min)

Fig. 56 Zinc ore flotation at various flotation times with oleic acid concentration 300 g/t at pH=10

4.6.3 Effect of KAX The zinc grade-recovery curves of the bench flotation tests as a function of pH and the reagent consumption are illustrated in Fig. 57. The examination of the tests shows that the best oxide zinc ore recovery in the concentrate was obtained at collector concentration of 600 g/t at pH 10.5. The trend shows increase of recovery with increasing the collector concentrations. It is seen that the recovery as well as grade enhances slightly and at KAX concentration 600 g/t, the optimum recovery (68 %) obtains with zinc grade of 16.2 %. However, at the highest value of KAX addition, the zinc grade was low, which indicated, as expected, lower selectivity as the optimum concentration was exceeded.Fig. 58 shows zinc ore flotation at various flotation times. From Fig. 58, it can be concluded that the optimum flotation time is 15 min. The zinc grade and the recovery of the smithsonite in the concentrate and the tailing in terms of the quantity of KAX are given in Table 19.

81012141618202224

0 10 20 30 40 50 60 70 80 90 100

Zinc Recovery(%)

Zin

c G

rade

(%)

KAX=100 g/t KAX=200 g/t KAX=300 g/t

KAX=400 g/t KAX=500 g/t KAX=600 g/t

Fig. 57 Zinc grade-recovery curves for oxide zinc ore flotation investigating the effect of KAX

55

10

20

30

40

50

60

70

80

90

100

0 5 10 15 20 25 30 35

Rec

over

y(%

)

Flotation time(min)

Fig. 58 Zinc ore flotation at various flotation times with KAX 600 g/t

Table 19. Grade and recovery of bench flotation tests for various amount of KAX in the presence of sodium sulphide (2000 g/t) and copper sulphate (1500 g/t)

KAX (g/t) Grade & Recovery

(%)100 200 300 400 500 600

Recovery (%) 45 48 52 51 60 68Zn % in Conc. 21.6 19.6 17.5 18.1 17.1 16.2Zn % in Tails 6.5 6.5 6.5 6.4 5.7 5.1

4.6.4 Effect of Mixed Collector The flotation experiments were performed with mixed collector concentrations and the effect of mixture ratio was investigated. The mixture ratio of KAX: DDA was changed as 3:1, 2:1, 1:1, 1:2 and 1:3. The zinc grade-recovery curves of the bench flotation using mixed collectors depends on mixture ratio are shown in Fig. 59. The present results show that the highest recovery for concentration ratio of 3:1 and 1:3 are 82 and 74 %, respectively. The zinc grade variations are from 17.2 to 25.4 %. The optimum mixture ratio of the mixed collector was KAX: DDA, 3:1 at pH 11.5 with 82 % recovery and 24.1% zinc grade. Fig. 60 illustrates the flotation time vs. recovery (KAX: DDA; 1:3). It can be concluded that the optimum flotation time is 20 min. The zinc grade and the zinc recovery in the concentrate and the tailing in terms of the quantity of mixed collector are given in Table 20.

56

81012141618202224262830

0 10 20 30 40 50 60 70 80 90 100

Zinc Recovery(%)

Zin

c G

rade

(%)

KAX:DDA(3:1) KAX:DDA(2:1) KAX:DDA(1:1) KAX:DDA(1:2) KAX:DDA(1:3)

Fig. 59 Zinc grade-recovery curves for oxide zinc ore flotation investigating the effect of mixed collector concentration ratio

20

30

40

50

60

70

80

90

100

0 5 10 15 20 25 30 35

Rec

over

y(%

)

Flotation time(min)

Fig. 60 Zinc ore flotation at various flotation times for mixed collector (KAX and dodecylamine, mole ratio 3:1) at pH 11

Table 20. Grade and recovery of bench flotation tests for various amounts of mixed collector (KAX and dodecylamine) in various mole ratio in the presence of sodium sulphide (2000 g/t)

KAX : DDA Concentration ratio Grade & Recovery

(%)3:1 2:1 1:1 1:2 1:3

82Recovery (%) Zn % in Conc. Zn % in Tails

24.12.4

7626.13.1

5525.45.4

6920.14.4

7417.24.3

57

4.6.5 Design of Flowsheet The lead and zinc beneficiation plant at Dandy village with a capacity of about 140 tons per hour is located 100 km southwest of Zanjan town. Oxide lead and zinc ore from the Angooran mine situated at a distance of 20 km from the plant is used as the feed for this plant. The primary minerals of this mine are cerussite and smithsonite. The plant has been designed to handle two types of feeds i.e. low and high grade zinc ores. The high grade ore contains 10% lead and 35% zinc while the low grade ore averages 7% lead and 22% zinc. The normal products of the plant include lead concentrate grading 60% lead, zinc concentrate with 38% zinc and a calcined zinc concentrate with 52% zinc. This plant includes the following sections: grinding, heavy media separation (HMS), milling, filtration and calcination. If the feed is of the low grade type, it is first upgraded by HMS cyclones and the concentrate obtained is introduced along with the high grade feed into the beneficiation circuit and subsequently concentrated in the flotation unit. Material < 2 mm in size which is not introduced into the HMS circuit, is milled and added to the low grade flotation route.Finally, after the rougher, cleaner and scavenger stages of the lead flotation route, a concentrate with 60% lead is obtained and the corresponding tailing is dewatered and sent to the zinc concentrate stockpile. A part of the zinc concentrate is sent to the calcination unit to obtain a calcined product containing 52% zinc. Fig. 61 shows the comparative results of using different collectors in oxide zinc ore flotation. The present results from laboratory flotation tests indicate that the zinc grade-recovery curves for flotation tests using DDA after sulphidization has highest grade and recovery.

0

10

20

30

40 50 60 70 80 90 100

Zinc Recovery(%)

Zin

c G

rade

(%)

KAX:DDA; 3:1 pH=10; OA=300 g/t Na2S=2000;DDA=300 g/t KAX=600 g/t

Fig. 61 Zinc grade-recovery curves for oxide zinc ore flotation investigating the effect of DDA, OA, KAX and mixed collector concentration ratio

Fine particles can be attached on coarse particles as slime coatings. These coatings can be detrimental to flotation. When the slime coating consists of gangue particles, they can prevent the attachment of air bubbles and cause low recoveries. Slime will seriously interfere with the flotation of oxide zinc ores and it should be removed prior to flotation of oxide minerals. The grinding feed conducted to desliming stage for the removal of slime. It was used one stage hydrocyclone to remove slimes. The deslimed underflow is fed to

58

smithsonite flotation circuit to float smithsonite and the overflow (wt. %:14.5) is discarded as slimes with average zinc grade of 4.5%. Overall zinc flotation recovery with respect to the total zinc was estimated to be about 75.5 % according to the zinc content in the slimes and tailings. The slime influence can be immediately seen from Table 21, which shows the comparison of oxide zinc ore flotation tests results with and without desliming. The dosage of reagents is the same in both cases (DDA= 300 g/t) except for sodium sulphide. Without desliming, the flotation of smithsonite consumes more sodium sulphide with a lower zinc grade and recovery than that with desliming operation. Fig. 62 shows the comparative results of oxide zinc ore flotation using DDA (300 g/t) as a collector after sulphidization with or without desliming of flotation feed. The present results from laboratory flotation tests indicate that the zinc grade-recovery curves for flotation tests using DDA after sulphidization with desliming has highest grade and recovery.

Table 21.The effect of desliming on grade and recovery of oxide zinc ore flotation using DDA concentration of 300 g/t and various Na2S concentrations

Without desliming With desliming

Na2S(g/t)Zn grade in Conc.

(%)

Zn grade in Tails.

(%)

Zinc Recovery

(%) Na2S(g/t)

Zn grade in Conc.

(%)

Zn grade in Tails.

(%)

Zinc Recovery

(%) 1000 12 9.1 26 1000 14 7.7 44.22000 15 6 62.5 2000 24.5 2.2 84.53000 16 7.5 41.2 3000 25.3 4.6 64.1

0

10

20

30

0 10 20 30 40 50 60 70 80 90 100

Zinc Recovery(%)

Zin

c G

rade

(%)

Na2S=1000 g/t ;With desliming Na2S=1000 g/t;Without deslimingNa2S=2000 g/t;Without desliming Na2S=2000 g/t;With deslimingNa2S=3000 g/t;Wihout desliming Na2S=3000 g/t;With desliming

Fig. 62 Zinc grade-recovery curves for oxide zinc ore flotation investigating the effect of desliming in DDA concentration of 300 g/t and various Na2S concentrations

The designed flowsheet for the treatment of the Angooran oxide zinc ore is shown in Fig. 63. The apparent advantage of this flowsheet is that the smithsonite as a valuable mineral in the ore is recovered. Besides, the desliming operation can remove most of the harmful slime and increases zinc flotation recovery from 62.5 to 84.5 %.

59

Ore feed

Grinding(d80 100 μm)

Tailings Wt.% : 65.4 (Zn%: 2.2)

(Zinc Rec. %: 15.5)

Desliming (d50 = 20 μm)

Slimes( -20 μm )

Wt.% : 14.5 Zn%: 4.5

Smithsonite Flotation DDA= 300 g/t Na2S= 2000 g/t

Pine oil = 100 g/t Sodium silicate =100 g/t

Zinc Conc. Wt.% : 34.6 (Zn%: 24.5)

(Flotation Zinc Rec. %: 84.5) (Overall Zinc Rec. % :75.5)

Fig. 63 The designed flowsheet and reagent system for the beneficiation of the Angooran oxide zinc ore

60

5 Conclusions

A. The zeta potential measurements of sulphidized (using a concentration of 2 g/l) smithsonite show that ZnS is partly formed on the mineral surface making the surface more negatively charged than the unsulphidized mineral. By using a higher Na2S concentration (3 g/l) the mineral becomes over sulphidized and the flotation recovery decreases. B. When smithsonite is treated with DDA (dodecylamine) in alkaline solutions, the negative zeta potential decreases resulting in absorption of dodecylamine on the mineral surface after a sulphidization treatment. This observation is confirmed by our FTIR analyses. The flotation of the mineral is enhanced and results in a recovery of 94%. C. When smithsonite is treated with oleic acid, the negative zeta potential increases due to adsorption of oleic anions on the mineral surface. As has been revealed by FTIR analyses the adsorption of oleic acid on the mineral surface is caused by an ion exchange mechanism. The maximum adsorption occurs at pH 10 and the resulting flotation recovery was then 93 %. D. When using hexylmercaptan, the maximum flotation recovery amounts to only 79 %. This can be explained by the formation of strong S-Zn bonds on the surface of the smithsonite mineral particles leading to a weakening of S-H bonds in the hexylmercaptan molecules. E. When smithsonite is treated with potassium amyl xanthate, KAX, the maximum flotation recovery occurs at pH 10.5 and amounts to 82%. The effect is devoted to that KAX is predominantly adsorbed on the negatively charged mineral surface present in an alkaline solution. XPS studies indicate that, when the mineral is activated by copper, an ion exchange occurs between Zn2+ and Cu2+ on the mineral surface resulting in the formation a surface layer of CuS.F. The presence of CS2 and NH2 on the surface of smithsonite revealed the coadsorption of mixed collector onto mineral surface as amine-xanthate complex. The presence of KAX increased the dodecylamine adsorption. G. Bench flotation tests show that the maximum flotation recovery for the smithsonite mineral (84.5 %) occurs by using dodecylamine after sulphidization treatment of the mineral surface.H. A flowsheet for an optimal and industrial flotation process for the smithsonite mineral to produce zinc ore, including desliming and sulphidization sub-processes, is presented in this thesis. The flowsheet is based on our research results and the bench scale flotation experiments.

5 Plan for future work Further work will be needed to study other collectors in sulphydryl groups e.g. dodecylmercaptan, hexylxanthate, dodecylxanthate and also hydroximic acid which may be have better performance in adsorption and flotation process. Since the ion exchange can play an important role in smithsonite-xanthate system, typical depletion adsorption tests can be done to give better information. The sensitivity of slimes to oxide zinc ore flotation with references to adsorption of collectors on fine and coarse particles can be investigated. In this case, some organic and inorganic depressants and flocculants can be used to depress specially slimes and the gangue minerals.

61

References

Abramov, A. A., 1961, ''Use of cationic agents for the flotation of oxide lead-zinc minerals'', Chemical abstracts, Vol. 55, 26910f.

Aplan, F. F., and De Bruyn, P. L., 1963, ''Adsorption of Hexyl mercaptan on gold'', Trans. AIME, 229, pp.235-242

Aplan, F. F., and Fuerstenau, D. W., 1962, ''Principles of non-metallic mineral flotation'', Fuerstenau, D.W. (Ed.), Froth Flotation 50th Anniversary Volume, AIME, New York, pp.170-215

Aplan, F. F., and Fuerstenau, D. W., 1984, ''The flotation of Chrysocolla by mercaptan'', International Journal of Mineral Processing, 13, pp.105-115

Barbaro, M., Herrera Urbina, R., Cozza, C., Fuerstenau D. W., and Marabini, A., 1997, ''Flotation of oxide minerals of copper using a new synthetic chelating reagent as collector'',International Journal of Mineral Processing, Volume 50, Issue 4, pp. 275-287

Barbery, G. Cecile, J. L., and Plichon, V., 1977, ''The use of chelates as flotation collectors'', Proceeding of XII International Mineral Processing Congress, Sao Polo, pp.19-34

Bichler, C. H., Bischoff, M., Langowski, H.-C., Moosheimer, U., 1996, '' The Substrate-Process Interface in Thin Barrier Film Coating'', Proceeding of 39th Annual Technical Conference of the Society of Vacuum Coaters, SVC publications, Philadelphia, PA.

Billi, M., and Quai, V., 1963, ''Development and results obtained in the treatment of zinc oxide ores at AMMI mines'', XI International Mineral Processing Congress, Cannes, pp. 631–649

Bond, F. C., 1952, ''The third theory of Comminution'', AIME, Vol.193, pp. 484-494 Bou, M., Martin, J. M., Le Mogne, T. H., Vovelle, L., 1991, ''Chemistry of the interface between

aluminum and polyethyleneterephtalate by X PS'', Applied Surface Science, Vol. 47, pp.149-161

Briggs, D., and Seah, M. P., 1994, ''Practical Surface Analysis'', Vol. 1, Second Edition, Chichester, John Wiley & Sons, Ltd.

Bulatovic, Srdjan M., 2007, ''Handbook of flotation reagents: chemistry, theory and practice'', Elsevier Science & Technology, Netherlands, 1st edition, ISBN: 9780444530295, pp.128-130

Bustamante, H., Shergold, M., 1983, ''Surface chemistry and flotation of zinc oxide minerals: I-Flotation with dodecylamine'', Transactions of the Institution of Mining and Metallurgy, 92, pp. C201-C203

Caproni, G., Ciccu, R., Ghiani, M., and Trudu, I., 1979, ''The processing of oxide lead and zinc ores in the campo Pisano and San Giovanni plants (Sardinia)'', Proc. XIII International Mineral Processing Congress, J. Laskowski (Ed.), Elsevier, Warzawa, Poland, pp.69-93

Castro, S., Goldfarb, J., and Laskowski, J., 1974, ''Sulphidizing reactions in the flotation of oxidized copper minerals, I. Chemical factors in the flotation of oxide copper oxide'', International Journal of Mineral Processing, Vol. 1, pp.141-149

Clark, D. T., and Thomas, H. R., 1978, ''Applications of ESCA to Polymer Chemistry'', XVII. Systematic Investigation of the Core Levels of Simple Homopolymers, Journal of Polymer Science, Polymer Chemistry Edition, Vol. 16, pp. 791-820

Cook, Melvin A., Nixon, John C., 1950, ''Theory of Water-Repellent Films on Solids Formed by Adsorption from Aqueous Solutions of Heteropolar Compounds'', Journal of physical and colloid chemistry, Vol. 54, pp.445

62

Dalman, G., and Gorin, G., 1961, ''Ionization constant of Hexanethiol from solubility measurements'', Journal of Organic Chemistry, ACS Publications, Vol. 26, Issue 11, pp. 4682- 4684

Delpeux, S., Beguin, F., Benoit, R., Erre, R., Manolova, N., Rashkov, I., 1998, ''Fullerene core star-like polymers-1. Preparation from fullerenes and monoazidopolyehers'', EuropeanPolymer Journal, Vol. 34, No.7, pp.905-915

Eyring, M., and Wadsworth, Milton E., 1956, ''Differential infrared spectra of adsorbed monolayers-n-Hexanethiol on Zn Minerals'', Mining Engineering, Transaction AIME, pp.531-535

Farmer, V. C., 1974, ''The infrared spectra of minerals'', Mineralogical Society, ISBN: 978-0903056052, pp.239

Ferraro, John R., 1982, ''The Sadtler infrared spectra hand book of minerals and clays'', Sadtler Research Laboratories, ISBN: 978-0845600801

Fleming, M. G., 1953, ''Effect of soluble sulphide in the flotation of secondary lead mineral'', Trans. IMM, London, 28, pp.C521-C554

Frost, Ray L., Martens, Wayde N., Wain, Daria L., and Hales, Matt C., 2007, ''Infrared and infrared emission spectroscopy of the zinc carbonate mineral smithsonite'', Spectrochimica Acta Part A, Molecular and Biomolecular Spectroscopy, In Press, Corrected Proof, http://www.sciencedirect.com/dx.doi.org/10.1016/j.saa.2007.10.027

Frost, Ray L., Hales, Matt C., and Wain, Daria L., 2008, ''Raman spectroscopy of smithsonite'', Journal of Raman Spectroscopy, John Wiley & Sons, Ltd., Vol. 39, Issue 1, pp.108-114

Fuerstenau, M. C., Palmer, B. R, 1976, ''Anionic Flotation of Oxides and Silicates Flotation'', A. M. Gaudin Memorial Volume., Vol. 1, American Institute of Mining, Metallurgical and Petroleum Engineers, New York, pp. 148-196

Fuerstenau, D. W., Sotillo, F., Valdivieso, A., 1985a, ''Sulfidization and flotation behaviours of anglesite, cerussite and galena'', Proceeding. XV International Mineral Processing Congress, Cannes, France, 2-9 June 1985. St. Etienne: GEDIM, pp.74-86

Gadsden, J. A., 1975,''Infrared Spectra of Minerals and Related Inorganic Compounds'', Butterworth & Co. Ltd., London, U.K., pp.66

Gaudin, A. M., and Harris, D. L., 1954, ''Adsorption of a mercaptan on zinc minerals'', Trans. AIME, 199, pp.925-930

Gaudin, A. M., 1957, ''Flotation'', McGraw Hill Inc, New York, pp.182-189 Glembotskii, V. A., 1972, ''Flotation'', Primary Sources, New York, pp.185-189 Gong, Wen Qi Parentich, A., Little, L., H. and Warren, L. J., 1992, ''Adsorption of oleate on

apatite studied by diffuse reflectance infrared Fourier transform spectroscopy'', Langmuir 8, pp 118 – 124

Gruender, Werner, 1963, ''Aufbereitungskunde'', Huebner Verlag, pp.262-282 Guo, H., Yen, W. T., 2002, ''Surface potential and wettability of enargite in potassium amyl

xanthate solution'', Minerals engineering, Vol. 15, pp.405-414 Goodhew, P. J., 1983, ''Specimen Preparations in Material Science'', Practical Methods in

Electron Microscopy, Vol 1, No. 1, Elsevier Science Ltd, ISBN: 978-0444104120 Hales, Matthew C., and Frost, Ray L., 2007, ''Synthesis and vibrational spectroscopic

characterization of synthetic hydrozincite and smithsonite'', Polyhedron, 26, pp.4955-4962

63

Healy, T. W., and Moignard, M. S., 1976, ''A review of electrokinetic studies of metal sulphides'', Flotation: A. M. Gaudin Memorial Volume, Fuerstenau M. C. (Ed.), AIME, New York, pp.334-363

Helbig, C., Baldauf, H., Mahnke, J., Stöckelhuber, K. W., and Schulze, H. J., 1998, ''Investigation of Langmuir monofilms and flotation experiments with anionic/cationic collector mixtures'', International Journal of Mineral Processing, Vol.53, pp.135-144

Herrera Urbina, R., Sotillo, F. J., Fuerstenau, D .W.,1999, ''Effect of sodium sulfide additions on the pulp potential and amyl xanthate flotation of cerussite and galena'' , International Journal of Mineral Processing , 55(3),pp. 157 - 170

Herrera Urbina, R., Sotillo, F. J., Fuerstenau, D. W., 1998, ''Amyl xanthate uptake by natural and sulfide-treated cerussite and galena'', International Journal of Mineral Processing, 55(2),pp. 113 – 128

Hren, J. J., Goldstein, J. I., and Joy, D. C., 1979, ''Introduction to Analytical Electron Microscopy'', Plenum Press, New York

Huber-Panu, I., Ene-Danalache, E., and Cojocariu, D.G., 1976, ''Mathematical Models of Batch and Continuous Flotation'', Flotation- A.M. Gaudin Memorial Volume, M.C. Fuerstenau, ed., AIME, New York, NY, Vol. 2, Chapter 5, pp. 675-724

Hunter, Robert J., 1993, ''Introduction to modern colloid science'', Oxford University Press, United Kingdom, ISBN 0-19-855386-2

Hunter, R. J., 1981, ''Zeta potential in colloid science'', Principles and Applications, Academic Press, London, pp.233-235

Hu, Yuehua, Luo, Lin, and Qiu, Guangzhou, and Wang, Dianzuo, 1995, ''Solution Chemistry of electrokinetic behavior of carbonate minerals'', Transactions of Nonferrous Metals Society of China, Vol.5, No.4, pp. 27-30

Jang, Woo-Hyuk, Drelich, J., and Miller, Jan D., 1995, ''Wetting Characteristics and Stability of Langmuir-Blodgett Carboxylate Monolayers at the Surfaces of Calcite and Fluorite'', Langmuir 11, pp.3491 – 3499

Jones, G. C., Jackson, B., 1993, ''Infrared Transmission spectra of carbonate minerals'', Chapman & Hall, London, ISBN: 978-0412546507

Jones, M. H., and Woodcock, J. T., 1978, ''Evaluation of Ion elective electrode for control of sodium sulphide additions during laboratory flotation of oxidized ores'', Trans. IMM 87, C 99-C105

Jung, R. F., James, R. O., and Healy, T. W., 1987, ''Adsorption, precipitation, and electrokinetic process in the iron oxide (goethite)-oleic acid-oleate system'', Journal of Colloid and interface Science, Vol. 118, No.2, pp.463-472

Keqing, F. A., Miller, J.D., and Guang-hui, Li, 2007, ''Sulphidization flotation for recovery of lead and zinc from oxide-sulfide ores'', Transactions of Nonferrous Metals Society of China, Vol. 15, No.5, pp. 1138-1147

Khan, Ali, and Marques, Eduardo F., 2000, ''Synergism and polymorphism in mixed surfactant systems'', Current Opinion in Colloid & Interface Science, Vol. 4, pp.402-410

Kiersznicki, T., Majewski, J. J., and Mzyk, J., 1981, ''5-alkylsalicyladdoximes as collectors in flotation of sphalerite, smithsonite and dolomite in a Hallimond tube'', International Journal of Mineral Processing, Vol.7, pp.311-318

Klimpel, R.R., 1980, ''Selection of Chemical Reagents for Flotation'', Mineral Processing PlantDesign, 2nd Edition, A.L. Mular and R.B. Bhappu (Eds.), Chapter 45, AIME, New York, NY, pp. 907-934

64

Kongolo, M., Benzaazoua, M., de Donato, P., Drouet, B., and Barres, O., 2004, ''The comparison between amine thioacetate and amyl xanthate collector performances for pyrite flotation and its application to tailings desulphurization'', Minerals Engineering, Vol. 17, pp.505-515

Laskowski, J. S, 1989, ''The colloid chemistry and flotation properties of primary aliphatic amines''. In: K. V. S. Sastry and M. C. Fuerstenau (Eds.), Challenges in Mineral Processing.Soc. Mining Eng., pp.15-34

Laskowski, J. S., Liu, Q. and Zhan, Y., 1997, ''Sphalerite Activation: Flotation and Electrokinetic Studies'', Minerals Engineering, Vol. 10, pp.787-802

Laskowski, J. S., and Ralston, J., 1992, ''Colloid chemistry in mineral processing'', Elsevier Science Publishing Co., pp. 428

Leja, J., 1982, ''Surface chemistry of froth flotation'', Plenum Press, New York Liao, H. M., Sodhi, R. N. S., Coyle, T. W., 1993, ''Surface composition of Al N powders studied

by x-ray photoelectron spectroscopy and bremsstrahlung-excited Auger electron spectroscopy'', J. Vac. Sci. Technol. A, Vol. 11, No. 5, pp.2681-2686

Lide, David R., 2005, ''CRC hand book of chemistry and physics'', a ready reference book of chemical and physical data, Editor in chief: David R. Lide, 85th edition, Boca Raton, Florida, CRC Press, ISBN: 0849304857, pp. 8-134

Lim, A. S., Atrens, A., 1990, ''ESCA studies of Nitrogen-Containing Stainless Steels'', Appliedphysics. A, Solids and surfaces, Springer, Berlin, Vol. 51, pp. 411-418, ISSN: 0721-7250

Malghan, S.G., 1986, ''Role of sodium sulfide in the flotation of oxidized copper, lead and zinc ores'', Minerals and Metallurgical Processing, Vol. 3, NO. 3, pp.158-163

Marabini, A. M., Ciriachi, M., Plescia, P., Barbaro, M., 2007, ''Chelating reagents for flotation'', Minerals Engineering, Vol. 20, pp.1014-1025

Marabini, A. M., Alesse, V., Garbassi, F., 1984, ''Role of sodium sulphide, xanthate and amine in flotation of lead-zinc oxide ores'', Jones, M. J., Oblatt, R. (Eds.), Reagents in the Mineral Industry, The Institute of Mining & Metallurgy, London, pp.125-136

Marabini, A. M. and Rinelli, G., 1986,''Flotation of Lead-Zinc Oxide Ores'', Advances in mineral processing, Proceedings of a symposium honoring Nathaniel Arbiter, P. Somasundaran (Ed.), Society of Mining Engineers, Littleton, CO, USA, pp.269-288

Maroie, S., Haemers, G., Verbist, J. J., 1984, ''Surface oxidation of polycrystalline "alpha"( 75% Cu et 25% Zn ) and "beta" ( 53% Cu et 47% Zn ) brass as studied by XPS : influence of oxygen pressure'', Applications of Surface Science,Vol.17,pp.463-476

McGarry, P. E., Pacic, Z., 1981, ''Flotation of non-sulfide zinc materials'', United States patent: 4253614

Miller, J. D., and Kellar, J. J., 1999,''Internal reflection spectroscopy for FTIR analysis of carboxylate adsorption by semi soluble salt minerals'', Advances in Flotation Technology. Society for Mining, Metallurgy and Exploration, Inc., Littleton, CO, ISBN: 9780873351843, pp.45-58

Nagano, J., Tanaka, M., and Saito, K., 1974, ''Ore Flotation'', Chemical abstracts, 82:114571 Nagano, J., Tanaka, M., and Saito, K., 1975, ''Flotation of zinc oxide ore'', Chemical abstracts,

83: 31384. Nematollahi, H., 1994, ''New size laboratory ball mill for Bond Work Index determination'',

Mining Engineering, SME, Vol. 46(4), pp.352-353 Önal, G., Bulut, G., Gül, A., Kangal, O., Perek, K.T., and Arslan, F., 2005, ''Flotation of Aladagˇ

oxide lead–zinc ores'', Minerals Engineering, Vol.18, pp.279-282

65

Ozbayoglu, G., Atalay, U., and Senturk, B., 1994, ''Flotation of lead and zinc carbonates ore'', Proceeding of International Conference on recent advances in materials and mineral resources, Penang, Malaysia, May 1994, School of Material and Mineral Resources Engineering, pp. 504–509, ISBN 938-9700-24-3

Pascal, P., 1962, ''Complexes du zinc'', in Compléments au Nouveau traité de chimie minérale, Masson et Cie, Paris, pp.318-321

Pereira, C.A., and Peres, A. E. C., 2005, ''Reagents in calamine zinc ores flotation'', Minerals Engineering, Vol. 18, pp.275-277

Pugh, R.J., 1986, ''The role of the solution chemistry of Dodecylamine and oleic acid collectors in the flotation of fluorite'', Colloids and Surfaces, Vol. 18, pp.19-41

Pålsson, B., Forssberg, E., 1989, ''Computer-Assisted Calculations of Thermodynamic Equilibria in Sphalerite Xanthate Systems'', International Journal of Mineral Processing, 26, pp.223-258

Quaresima, S., Sivadasan, K., Marabini A., Barbaro, M., and Somasundaran, P., 1991,'' Behaviour of colloidal suspensions of zinc carbonate in the presence of copolymers designed for selective flocculation'', Journal of Colloid and interface Science, Vol. 144, No.1, pp.159-164

Rausch, D.O., and Mariacher, B.C., 1970, ''Concentration of oxide ores at Tynagh Mining and concentrating of lead & zinc”, Rausch, D.O., Mariacher, B.C.(Eds.),AIME World Symposium on Mining & Metallurgy of Lead & Zinc, Vol. 1, Extractive metallurgy of lead and zinc, New York, pp. 721-731

Rey, M., 1953, ''The flotation of oxide ores of lead, copper and zinc'', Recent Developments in Mineral Dressing Symposium, IMM, London, pp. 541-548

Rey, M., 1979, ''Memoirs of milling and process metallurgy: 1-flotation of oxide ores'', Institution of Mining and Metallurgy, Section C, 88, pp.245-250

Rey, M., De Merre, P., Mancuso, R., 1961, ''Recent research and developments in flotation of oxide ores of copper, lead and zinc'', Colorado School of Mines Quarterly, 56(3), pp. 163 - 175

Rey, M., Sitia, G., Raffinot, P. and Formaneck, V., 1954, ''Flotation of oxide zinc ores'', Trans. AIME, Mining Engineering, 199, pp.416-420

Rinelli, G., Marabini, A. M., 1973, ''Flotation of Zinc and Lead Oxide-Sulphide Ores with Chelating Agents'', Inst. Min. Met., Proc. X International Mineral Processing Congress, London, pp.493-521

Roberts, Jeffrey T., and Friend, C.M., 1988, ''Spectroscopic identification of surface phenyl thiolate and benzyne on MO (110)'', American Institute of Physics, J. Chem. Phys., 88(11), pp. 7171-7180

Sabah, E., Turan, M., and Celik, M.S., 2002, ''Adsorption mechanism of cationic surfactants onto acid and heat activated sepiolites'', Water research, Vol. 36, pp.3957-3964

Sardar, S. A., Syed, J. A., Yagi, S. and Tanaka, K., 2004, ''Orientation of ethyl mercaptan on cu (111) surface'', Thin solid Films 450, pp.265-271

Shibata, J., and Fuerstenau, D.W., 2003, ''Flocculation and flotation characteristics of fine hematite with sodium oleate'', International Journal of Mineral Processing, 72, pp.30

Smith, R. W., and Akhtar, S., 1976, ''Cationic flotation of Oxide and silicates'', Flotation, M. C. Fuerstenau (Ed.), A. M. Gaudin Memorial Volume, Vol. 1, AIME Inc., New York, pp.87-116

Smith, B. C., 1998, ''Infrared Spectral Interpretation: A systematic Approach'', CRC Press, Washington, DC

66

Socrates, G., 1980, ''Infrared Characteristics Group Frequencies'', John Wiley & Sons, Ltd., New York

Somasundaran, P., and Ananthapadmanabhan, K. P., 1979, ''Solution chemistry of surfactants'', K .L. Mittal, (Ed.), Plenum press, New York, Vol.2, pp.777

Somasundaran, P., and Wang, Dianzuo, 2006, ''Solution chemistry: minerals and reagents'', Elsevier publisher, Development in mineral processing 17, pp.18-22

Stumm, Werner, and Morgan, J. J., 1970, ''Aquatic Chemistry'', Wiley InterScience, New York, 503p

Sui, G., Orbulescu, J., Ji, X., Gattas-Asfura, K.M., Leblanc, R.M., and Micic, M., 2003, ''Surface Chemistry Studies of Quantum Dots (QDs) Modified with Surfactants'', Journal of Cluster Science, Springer Netherlands, Vol. 14, No.2, pp.123-133

Tarjan, G., 1986, '' Mineral Processing'', Akademiai Kiado, Budapest, Hungary, p.782, ISBN: 9630522438

Tomlinson, E., Davis, S. S., Mukhayer, G. I., 1979, ''Ionic interaction and phase stability'', In: Mittal, K. L., (Ed.), Solution chemistry of surfactants, 1, Plenum Press, New York, pp.3-43

Tondre, Christian, and Caillet, Céline, 2001, ''Properties of the amphiphilic films in mixed cationic/anionic vesicles: a comprehensive view from a literature analysis'', Advances in Colloid and Interface Science, Vol. 93, pp.115-134

Vijaya Kumar, T. V., Prabhakar, S., and Bhaskar Raju, G., 2002, ''Adsorption of Oleic Acid at Sillimanite/Water Interface '', Journal of Colloid and Interface Science, Vol. 247, pp. 275-281

Wagner, C.D., Riggs, W. M., Davis, L. E., Moulder, G. F., 1979, ''Handbook of X-ray photoelectron spectroscopy'', Editor: G. E. Muilenberg, Eden Prairie, Minnesota, Physical Electronics Division, Perkin-Elmer Corporation

Weiss, N. L., 1985, ''SME Mineral Processing Hand book'', AIME, pp. 15-4 15-7 Wills, B. A., 1997, ''Mineral Processing Technology'', Butterworth-Heinemann, 6th Edition, pp.

262-267Yamada, M., Shoji, T., Onada, T., Shimoiizaka, J., 1976, ''Flotation of zinc carbonate'', Chemical

abstracts, 84:182894. Yamazaki, S., Matusi, N., and Ohtsubo, E., 1978, ''Addition of flotation agents'', Chemical

Abstracts, 89:63078. Zhang, Qingsong, Xu, Zhenghe, and Finch, J. A., 1995, ''Prediction of species distribution at

sphalerite / water interface'', Minerals Engineering, Vol. 8, pp.999-1007

67

Appendix

Paper I

Adsorption Studies of Smithsonite Flotation Using Dodecylamine and Oleic acid Seyed Hamid Hosseini and Eric Forssberg, Minerals and Metallurgical Processing, SME, Vol. 23, No.2, 2006, 87-96.

MINERALS & METALLURGICAL PROCESSING Vol. 23, No. 2 • May 200687

Adsorption studies of smithsonite flotation using dodecylamine and oleic acidS.H. Hosseini and E. ForssbergPh.D. student and professor, respectively, Division of Mineral Processing, Department of Chemical Engineering and Geosciences, Luleå University of Technology, Luleå, Sweden

AbstractThe interaction of various concentrations of sodium sulfide, dodecylamine (DDA) and oleic acid (OA) on smithsonite were investigated at different pH levels using zeta potential, contact angle, microflotation and diffuse-reflectance FT-IR studies. Flotation results show that the recovery and contact angle are enhanced to 94% and 115º, respectively, with a dodecylamine concentration of 1.6 × 10-3 M and a pH of 11.5. The optimum sodium sulfide consumption was found to be 2.6 × 10-2 M. Zeta potential measurements showed less negative charge after DDA treatment on the surface of pure crystalline smithsonite. The recovery and contact angle for oleic acid flotation rises to 93% and 105º, respectively, with an oleic acid concentration of 1.1 × 10-3 M and a pH of 10. The zeta potential in the case of using oleic acid showed a more negative charge after oleic acid treatment on the smithsonite surface. The FT-IR spectra studies of smithsonite conditioned with DDA confirmed the adsorption of DDA on the smithsonite surface. The spectra show that the mineral surface is changed partially to a ZnS layer after sodium sulfide treat-ment. The spectra confirmed the formation of zinc oleate on the smithsonite surface after oleic acid treatment. A comparison of the results using cationic and anionic collectors showed that the different adsorption densities of the reagents in two cases conferred different degrees of hydrophobicity on the smithsonite surface.

Key words: Smithsonite, FT-IR, Flotation, Zeta potential

Introduction

3 2 4

3 2

2 3 2

-

-

2

2

Paper number 05-316. Original manuscript submitted for review May 2005. Revised manuscript received and accepted for publication November 2005. Discussion of this peer-reviewed and approved paper is invited and must be submitted to SME Publications Dept. prior to Nov. 30, 2006. Copyright 2006, Society for Mining, Metallurgy, and Exploration, Inc.

May 2006 • Vol. 23 No. 2 MINERALS & METALLURGICAL PROCESSING88

-

-

Materials and methods

Fe2 3 2 2

-

-

-

2

-

2

Zeta-potential measurements.

-

-

3

-

2

Contact angle measurements.

-

cm

cm

Results and discussion

--2

2

2-+ -

1

HS

S H2

S -2

MINERALS & METALLURGICAL PROCESSING Vol. 23, No. 2 • May 200689

22

2-S2

2

-

2

2

2

2

2

2

-

-

of ZnS

-

-

-

2

Zeta-potential studies:

Figure 1 — Zeta potential of pure smithsonite as a function of pH in distilled water and sodium sulfide solution, 2.6× 10-2 M. ZnS (10-6 M NaCl) (replotted from Zhang et al., 1995).

Figure 2 — Species distribution diagram of sodium sulfide in aqueous solution as a function of pH.

May 2006 • Vol. 23 No. 2 MINERALS & METALLURGICAL PROCESSING90

-2

-3

-3

-

-

-

-2 -3

-

-2

-

Figure 3 — FT-IR spectra of (a) pure smithsonite (b) smithsonite after treatment with Na2S (c) pure ZnS.

Figure 4 — Zeta-potential of smithsonite as a function of pH in the presence of Na2S (2.6 × 10-2 M) and DDA.

MINERALS & METALLURGICAL PROCESSING Vol. 23, No. 2 • May 200691

-3

-3

-

-

Figure 5 — Flotation results for smithsonite at various concentrations of DDA in the presence of Na2S at pH 11.5 and in the absence of Na2S at pH 9.5.

Figure 6 — Contact angles of smithsonite as a function of DDA concentrations in the presence of various amounts of sodium sulfide (pH 11.5) and in the absence of sodium sulfide (pH 9.5).

May 2006 • Vol. 23 No. 2 MINERALS & METALLURGICAL PROCESSING92

-3

2

Zeta-potential studies:

-3

-3

-3

-

-

-3

Reference spectra: -

Figure 7 — Zeta potential of smithsonite as a function of pH in the presence of oleic acid.

-

-

2

2

3 3

2

-

in

3 2)

s 2

2 s 2

2

MINERALS & METALLURGICAL PROCESSING Vol. 23, No. 2 • May 200693

Figure 10 — FT-IR spectra of pure smithsonite treated with sodium sulfide at fixed concentration of 2 g/L at pH 11.5 with increasing initial concentration.

Figure 9 — Reference DRIFT IR-spectra of smithsonite, dodecylamine, sodium oleate, zinc oleate and sodium sulfide.

Figure 8 — Contact angles and flotation recoveries of smithsonite as a function of various concentrations of oleic acid at pH 10.

2) to

-4

ions

-

2

3

May 2006 • Vol. 23 No. 2 MINERALS & METALLURGICAL PROCESSING94

Figure 11 — Adsorption density and area under alkyl chain bands (2,990-2,800 cm-1) of DRIFT IR-spectra of smithsonite as a function of dodecylamine concentration at pH 11.5 in the presence or absence of sodium sulfide.

Figure 12 — DRIFT IR-spectra of pure smithsonite at pH 10 with increasing initial concentration of oleic acid (a) 4.42×10-5 M (b) 8.85×10-5 M (c) 1.77×10-4 M (d) 3.54 ×10-4

M (e) 5.31×10-5 M (f) 1.1×10-3 M.

744.01 743 (Farmer, 1974) 742.9 (Gadsden, 1975) 744 (Ferraro, 1982) 745 (Jones and Jackson, 1993)

840.0 841 (Gadsden, 1975)

868.7 874.1 (Farmer, 1974) 870 (Gadsden, 1975; Ferraro, 1982; Jones and Jackson, 1993)

1089.88 1093 (Farmer, 1974) 1096 (Jones and Jackson, 1993)

1175 1170 (Jones and Jackson, 1993)

1435.46 1440 (Farmer, 1974; Gadsden, 1975) 1427 (Ferraro, 1982; Jones and Jackson, 1993)

1812.03 1830-1790 (Gadsden, 1975) 1810 (Ferraro, 1982) 1816 (Jones and Jackson, 1993)

2492 2493 (Ferraro, 1982; Jones and Jackson, 1993)

2850 2850 (Jones and Jackson, 1993)

FT-IR

experiment

data Reference data

Table 1 — Assignment of pure smithsonite bands according to literature and experimental data (wave number in cm-1).

MINERALS & METALLURGICAL PROCESSING Vol. 23, No. 2 • May 200695

-

2

2 2 -3

--

2-2

] -

-

-

n surf)- -

)

Figure 13 — Adsorption density and area under alkyl chain bands (2,990-2,800 cm-1) of DRIFT IR-spectra of smithsonite as a function of oleic acid concentration at pH 10.

2 2 -3

Conclusions

2-3

--

-

2-2

-3

-3 -

2-

-2 2

-3

2

2 -3

May 2006 • Vol. 23 No. 2 MINERALS & METALLURGICAL PROCESSING96

Acknowledgments

ReferencesAbramov, A., 1961, “Use of cationic agents for the flotation of oxide lead-zinc

minerals,” Chemical abstract, Vol. 55, 26910f.Billi, M., and Quai, V., 1963, “Development and results obtained in the treatment

of zinc oxide ores at AMMI mines,” IMPC, London, Paper 43.Bustamante, H., and Shergold, M., 1983, “Surface chemistry and flotation

of zinc oxide minerals: I-Flotation with dodecylamine,” Trans. Inst Min. Metal, p. 92.

Farmer, V.C., 1974, The Infrared Spectra of Minerals, Mineralogical Society, pp. 239.

Ferraro, J.R., 1982, The Sadtler Infrared Spectra Handbook of Minerals and Clays, Sadtler, 440 pp.

Gadsden, J.A., 1975, Infrared Spectra of Minerals and Related Inorganic Com-pounds, Butterworth, p. 66.

Gaudin, A.M, 1957, Flotation, McGraw Hill Inc., New York, pp.182-189.Gong, Wen Qi, Parentich, A., Little, L.H., and Warren, L.J., 1992, “Adsorption

of oleate on apatite studied by diffuse reflectance infrared Fourier transform spectroscopy,” Langmuir 8, pp. 118-124.

Healy, T.W., and Moignard, M.S., 1976, “A review of electrokinetic studies of metal sulfides,” Flotation: A.M. Gaudin Memorial Volume, M.C. Fuerstenau, ed., AIME, New York, pp. 334-363.

Hu, Y., Luo, L., and Qiu, G., 1995, “Solution Chemistry of electrokinetic behaviour of carbonate minerals,” Transactions of NF Soc.,” Vol. 5, No. 4, pp. 27-30.

Jang, W., Drelich, J., and Miller, J.D., 1995, “Wetting characteristics and stability of Langmuir-Blodgett carboxylate monolayers at the surfaces of Calcite and Fluorite,” Langmuir 11, pp. 3491-3499.

Jones, G.C., and Jackson, B., 1993, Infrared Transmission Spectra of Carbonate Minerals, Chapman & Hall, London, 200 pp.

Malghan, S.G., 1986, “Role of sodium sulfide in the flotation of oxidized copper, lead and zinc ores,” Minerals & Metallurgical Processing, SME, pp. 158-163.

Marabini, A.M., Alesse, V., and Garbassi, F., 1984, “Role of sodium sulphide, xanthate and amine in flotation of lead-zinc oxidized ores,” Inst. of Mining & Metallurgy, pp. 125-136.

Marabini, A.M., and Rinelli, G., 1986, “Flotation of lead-zinc oxide ores,” Advances in Mineral Processing, Proceedings of a Symposium Honoring Nathaniel Arbitor, P. Somasundaran, ed., pp. 269-288.

McGarry, P.E., and Pacic, Z., 1981, “Flotation of Nonsulfide Zinc Materials,” United States patent: 4253614.

Miller, J.D., and Kellar, J.J., 1999, “Internal reflection spectroscopy for FTIR analysis of carboxylate adsorption by semi soluble salt minerals,” Advancesin Flotation Technology, Society for Mining, Metallurgy, and Exploration, Inc., Littleton, Colorado, pp. 45-58.

Ozbayoglu, G., Atalay, U., and Senturk, B., 1994, “Flotation of lead and zinc carbonates ore,” Recent Advances in Materials and Mineral Resources,Penang, Malaysia, pp. 504-509.

Pascal, P., 1962, Complexes du zinc, Nouveau traite de chimie minerale, Mas-son, Paris, pp. 318-321.

Quaresima, S., Sivadasan, K., Marabini A., Barbaro, M., and Somasundaran, P., 1991, “Behaviour of colloidal suspensions of zinc carbonate in the presence of copolymers designed for selective flocculation,” Journal of Colloid and interface Science, Vol. 144, No. 1, pp. 159-164.

Önal, G., Bulut, G., Gül, A., Kangal, O., Perek, K.T., and Arslan, F., 2005, “Flotation of Aladag oxide lead–zinc ores,” Minerals Engineering, Vol. 18, pp. 279-282.

Ramachandandra, R.S., and Helper, L.G., 1977, “Equilibrium constants and thermodynamics of ionization of aqueous hydrogen sulphide,” Hydrometal-lurgy, Vol. 2, pp. 293-299.

Rausch, D.O., and Mariacher, B.C., 1970, “Concentration of oxide ores at Tynagh Mining and concentrating of lead & zinc,” AIME World Symposium on Mining & Metallurgy of Lead & Zinc, Vol. 1, Extractive metallurgy of lead and zinc, pp. 721-731.

Rey, M., 1979, “Memoirs of milling and process metallurgy: 1-flotation of oxidised ores,” Institution of Mining and Metallurgy, pp. 245-250.

Rey, M., 1953, “The flotation of oxidized ores of lead, copper and zinc,” Recent De-velopments in Mineral Dressing Symposium, IMM, London, pp. 541-548.

Rey, M., Sitia, G., Raffinot P., and Formanek, V., 1954, “Flotation of Oxidized zinc ores,” Mining Engineering, pp. 416-420.

Smith, B.C., 1998, Infrared Spectral Interpretation: A Systematic Approach, CRC Press, Washington, DC, 265 pp.

Weiss, N.L., 1985, SME Mineral Processing Handbook, SME, pp. 15-24.Zhang, Q., Xu, Z., and Finch, J.A., 1995, “Prediction of species distribution at

sphalerite/water interface,” Minerals Engineering, Vol. 8, pp. 999-1007.

Paper II

Smithsonite Flotation Using Potassium Amyl Xanthate and HexylmercaptanSeyed Hamid Hosseini and Eric Forssberg, Mineral Processing and Extractive Metallurgy, Trans. Inst. Min Metals. C, Vol. 115, No. 3, 2006, 107-112

Smithsonite flotation using potassium amylxanthate and hexylmercaptan

S. H. Hosseini* and E. Forssberg

The influence of potassium amyl xanthate (KAX) and hexylmercaptan (HM) adsorption on

smithsonite surface at various concentrations were investigated through using zeta potential,

contact angle, microflotation and diffuse reflectance FTIR studies at different pH. The zeta potential

measurements of KAX showed that the adsorption of ionic charge (more negative charge after KAX

treatment) takes place on the surface of pure crystalline smithsonite. The charges vary between238

and 245 mV at pH 10.5. Flotation results using potassium amyl xanthate reveal that the maximum

recovery of 81.3% and the maximum contact angle of 98.7u occurs at pH 10.5 at KAX concentration

of 2.9661023M in sodium sulphide (2.661022M) and copper sulphate (9.461023M) solutions. The

highest recovery and contact angle for flotation by means of HM occurs at pH 9 at values of 78.6%

and 92.3u respectively with HM concentration 1.161022M. The FTIR spectra studies of smithsonite

conditioned by KAX confirmed the adsorption of KAX and the presence of CS2 on smithsonite

surface. The FTIR spectra in HM studies showed the adsorption of RS2 on the oxidised zinc surface

and the S–H bond in the mercaptan is destroyed on adsorption. The comparison between the results

using anionic collectors showed that the presence of different amounts of reagents on smithsonite

surface in two cases confer different degree of hydrophobicity on the smithsonite surface.

Keywords: Smithsonite, Anionic collector, Thiols, KAX, Hexylmercaptan, Zeta potential, FTIR, Contact angle, Flotation

IntroductionSome methods have been reported for the flotation ofoxidised minerals of the base metals but the mostimportant methods which have long been used commer-cially are as follows:

(i) sulphidisation and flotation with a sulphydrylcollector, e.g. xanthates and activation bymeans of a metal ion1–3

(ii) sulphidisation using sodium sulphide and flota-tion with a cationic collector, e.g. amines4

(iii) selected collectors such as mercaptans5,6

(iv) chelating agents.7–10

Oxidised zinc minerals such as smithsonite (ZnCO3),willemite (Zn2SiO4), hydrozincite (2ZnCO3, 3Zn(OH)2),zincite (ZnO) and hemimorphite (Zn2SiO3.H2O) havebeen important source of zinc ores.

Barbery et al. in 1978 reported that xanthates can beused for the flotation of oxidised zinc ore. The activatingeffect of sodium sulphide is strongly time dependent.The increase in sulphidisation leads to an increase in thehydrophobicity of the surface of minerals. The excess ofsodium sulphide acts as depressant for oxidised lead andzinc minerals because adsorption of divalent sulphideion on the surface of lead oxide minerals increase the

negative charge which prevents the adsorption ofcollector onto oxidised zinc mineral surfaces.11

According to Joly et al., 2004, the possible mechanismof reactions for adsorption of metal ions to AX2 can beexpressed as follows12

M2zz2AX{'M(AX)2 (1)

2M(AX)2'2MAXz(AX)2 (2)

It was proposed that the mercaptan S–H bond isdestroyed in the adsorption process. The adsorbedmercaptan reacts with the surface –OH group formingthe zinc mercaptan salt, splitting out a molecule of waterin the process.13

It is reported the adsorption of hexylmercaptan (HM)on sphalerite, zincite and willemite from aqueoussolution, zinc minerals and between aqueous solutionand the gaseous phase that adsorption of the mercaptanfrom the vapour phase before flotation testing wereeffective in causing flotation and adsorption of mercap-tan corresponding to less than monolayer coverage.14

Adsorption of hexanethiol onto zinc oxide, zincsulphide and zinc silicate are studied. The results showedthat the S–H bond in the mercaptan is destroyed uponadsorption and a zinc mercaptan salt is formed at themineral surface.5

Hexylmercaptan can be used for flotation of syntheticzinc oxide, although, the requirement for mercaptan tofloat oxidised zinc mineral is strenuous. The preflotation

Division of Mineral Processing, Departments of Chemical Engineering andGeosciences, Lulea University of Technology, SE 97187 Lulea, Sweden

*Corresponding author, email hamid.hosseini@ltu.se

� 2006 Institute of Materials, Minerals and Mining and The AusIMMPublished by Maney on behalf of the Institute and The AusIMMReceived 12 May 2005; accepted 14 March 2006DOI 10.1179/174328506X109077

Mineral Processing and ExtractiveMetallurgy (Trans. Inst. Min Metall. C) 2006 VOL 115 NO 2 107

preparation of oxidised zinc ores with mercaptans canbe carried out under dry condition, which can befollowed by flotation in the customary manner.6

Hexylmercaptan is a powerful collector for gold.Mercaptans with roughly six carbon atoms have reason-able solubility in water (3.261024 mol L21).15

The adsorption of mercaptans on copper surface is ofinterest because the –SH functional group promotes theadhesion of organic layers onto metal surfaces. The keypoints of the mercaptans adsorption process are theformation of a strong sulphur–metal (S–M) bond. Thesulphur adsorption site and the corresponding S–Mbond distance determine the nature of the interaction.16

In the present investigation, the effects of variousamounts of potassium amyl xanthate (KAX) and HM atdifferent pH values on pure smithsonite using zetapotential, contact angle, microflotation and FTIRstudies are investigated. In the present paper, flotationproperties of oxidised zinc mineral are discussed andthe role of anionic collectors during adsorption isdelineated.

Experimental

MaterialsThe pure smithsonite samples (purity of 90%) wereobtained from Angooran region, Iran. The X-rayfluorescence (XRF) chemical analysis showed thatthe pure sample contains 58.50%ZnO, 0.26%Fe2O3,0.29%SiO2, 0

.84%CaO, 3.41%Na2O, 1.06%PbO, 0.34%others and 35.30% LOI (loss of ignition). The mineralsamples were wet ground in porcelain mill. The productswere wet sieved to obtain 2100z75 mm size fractions.A portion of 275 mm was further ground and sizeclassification by gravity sedimentation to obtain 25 mmsize fractions. This final product (25 mm) was employedfor zeta potential, contact angle and FTIR measure-ments. The coarser size fractions of 2100z75 mm wereused for microflotation tests.

The specific surface area of pure crystalline samplewas measured by Flow Sorb II 2300, BET,Micromeritics Co., Ltd, USA, which is an instrumentdesigned to take the measurement on powders bynitrogen gas adsorption and desorption respectively, inliquid nitrogen and room temperature. The measuredsurface area was 0.43 m2 g21.

ReagentsPotassium amyl xanthate with 90% purity was pur-chased from Shandong Qixia Flotation Reagent Co.,Ltd, China and was purified by recrystallisation fromacetone. Hexanethiol (95% purity) was obtained fromSigma Aldrich, Switzerland. Sodium sulphide andcopper sulphate anhydrous (99% purity) was procuredfrom Merck KGaA, Germany. Analar grade HCl andNaOH were used for pH adjustment in all experi-ments. Deionised water (specific conductance, 0.4–0.7 mS cm21) was used in all experiments.

MethodsIn all measurements using KAX as collector, the mineralsuspensions were first immersed at given pH with 1%solids using NaOH or HCl as pH regulators and thenconditioned with sodium sulphide as sulphidising agentand then treated with copper sulphate for 2–4 min

respectively. After conditioning for regulators, collectorwas added into solution and treated for 5–10 min. Therewas no need to use sodium sulphide and copper sulphatein HM flotation.

Zeta potential measurementsThe zeta potential of pure smithsonite in aqueoussolutions was determined using a Coulter Delsa 440instrument, which is based on the electrophoretic lightscattering (ELS) technique, to directly measure thevelocity of particles moving in an electric field andalso zeta potential. The Coulter Delsa 440 instrumentperforms a simultaneous four angle ELS measurementusing a rectangular capillary cell. The cell has two goldplated silver electrodes with the feature of electric fieldfocusing that reduces possible electrode surface electro-lysis and heat generation at high ionic strength andensures an undistributed and homogeneous electric fieldat the scattering cross-section.17

In the present study, the mineral suspensions ofsmithsonite 25 mm with 1% solids after treatment withNa2S, CuSO4 and KAX solutions are transferred to therectangular capillary cell of the instrument. Themeasurement was performed using an electric field of10–20 V cm21, a frequency range of 500 Hz andduration of 100 s (using 2 s on and 1 s off sequencealtering the electric field polarity) at the upper stationarylayers of the cell. Four angle ELS was simultaneouslyutilised. The average value of zeta potential from thefour angles was reported as a final result. Thetemperature was kept at 25.0¡0.1uC through the wholemeasurements.

Microflotation testsThe single mineral flotation tests were performed using a100 mL Hallimond tube. One gram pure sample withparticle size of 2100z75 mm after treatment is con-ducted into the Hallimond tube and diluted withdeionised water. The flotation tests were carried out ata constant air flow rate of 20 cm3 min21. The float andsink products were filtered, dried and weighed todetermine the flotation recovery of the mineral beingstudied. In all cases, for adjusting pH NaOH or HClwere used in a beaker before conditioning withsulphidising agent or collectors.

Contact angle measurementsThe Fibro DAT1100, Dynamic adsorption tester, FibroSystem AB, Sweden, was performed to measure thecontact angle of the minerals. This instrument allowsdetailed, automatic analysis at full video speed of theinteraction between a liquid droplet and a specimensurface. Since it needs to have a flat surface for sample,the treated minerals compressed in a pellet and thentransferred to the instrument. The interaction between aliquid droplet and a specimen surface can be accuratelymeasured with millisecond precision within the firstsecond of contact between the mineral surface and theliquid. As a result of wetting, contact angle can bemeasured as a function of time.

Diffuse reflectance FTIR measurementsIn the present study, the infrared spectra were recordedfor all samples on the air dried 25 mm powders beforerecording the DRIFT infrared spectrum. The FTIR

Hosseini and Forssberg Smithsonite flotation using potassium amyl xanthate and hexylmercaptan

108 Mineral Processing and Extractive Metallurgy (Trans. Inst. Min. Metall. C) 2006 VOL 115 NO 2

spectra were obtained with a Perkin Elmer Spectrum2000 FTIR-Diffuse reflectance spectrometer. The PerkinElmer Spectrum 2000 FTIR Spectrometer is capable ofdata collection over a wave number range of 370–7800 cm21. This instrument can be configured to run insingle beam, ratio or interferogram modes. About 10%by the weight of the solid samples were mixed withspectroscopic grade KBr with a refractive index of 1.559and a particle size of 1–5 mm. These spectra wererecorded with 200 scans measured at 4 cm21 resolution.

Results and discussion

Zeta potential studies in KAX flotationThe electrokinetic potential studies were performed onpure smithsonite in distilled water and in the presence ofNa2S, CuSO4 and various concentrations of KAX. Theresults are illustrated in Fig. 1. The isoelectric pointof smithsonite is y8.0, which agree well with thosereported in literatures.18,19

The reaction may be occurring between HS2, whichpredominates in pH range of 7–13. The zeta potentialresults that were obtained with sulphidised smithsoniteshow that ZnS is more negatively charged thanunsulphidised smithsonite.

Adding sodium sulphide (2.661022M) makes theirzeta potential more negative and the value of isoelectricpoint decreases. The pHIEP of smithsonite drops from 8to 6.3 because chemical reactions of HS2 with metalions on the surface of minerals can occur, and thenmetal sulphide film which shows the surface charactersof sulphides is formed. Oversulphidisation of pulp(3 g L21) caused by high HS2 ion content is responsiblefor depression of the mineral and decreases the flotationrecovery and controlled sulphidisation appears to be thepreferred method of adding sodium sulphide.

The following chemical reaction is generally acceptedfor smithsonite sulphidisation

ZnCO3(s)zNa2S(aq)~ZnS(s)zNa2CO3(aq) (3)

The following reaction suggested by some authors forcopper activation of ZnS surface as follows20,21

ZnS(s)zCu2z~CuS(s)zZn2z

(aq) (4)

The surface species formed during copper activation arecovellite (CuS) and chalcocite (Cu2S). Many investiga-tors have reported that the activation product is covel-lite, while others have suggested that it is chalcocite. The

formation of CuS and of Cu2S are the possible speciesfor the mentioned reaction. In basic solutions, most ofthe copper added as activator is precipitated asCu(OH)2. It is also suggested that the hydroxide slowlyreleases Cu2z ions into solution, which in turn from aflotation active product such as (Zn, Cu)S.20,22

According to zeta potential measurements, it can beassumed at higher concentrations is increased adsorp-tion and charge more negatively. The negative charge ofthe surface is resulted from the adsorption of CS 2 anion,which agrees well with flotation and contact angleresults. The results of electrokinetic studies suggest thatthe sulphydryl collector (KAX) is adsorbed predomi-nantly on the negatively charged surface of smithsoniteat alkaline pH.

Microflotation and contact angle studies in KAXflotationMicroflotation and contact angle studies were carriedout in sodium sulphide solution (2.661022M) withvarious concentrations of copper sulphate and KAX.The results can be seen in Fig. 2.

According to the contact angle measurements, thehighest contact angle measured which is 98.7u for KAXconcentration (2.661023M). The flotation recoveries ofsmithsonite as a function of KAX concentration at pH10.5 showed at lower KAX concentration (4.9461024M), the rapid rise in smithsonite recovery beginsand at higher level (2.661023M), the maximum flota-tion recovery occurs at 81.3%. In general, the recoveryof smithsonite increases with increasing KAX concen-tration so that the recovery smithsonite is enhancedfromy53.5 to 81.3% at higher KAX concentration. Theresults suggest that at lower KAX concentrations wheresmithsonite has partial flotation recovery and theincreased addition of KAX enhance smithsonite flota-tion recovery.

Microflotation and contact angle studies in HMflotationResults of smithsonite microflotation with variousamounts of HM are reported in Fig. 3.

According to Fig. 3, the maximum flotation recoveryoccurs aty78.6% at pH 9. As expected, the results showmaximum flotation recovery and contact angle occurs athighest concentration of HM (1.161022M) at pH 9which are 78.6% and 92.3u respectively. As it can beseen, at lower HM concentration where smithsonite haspartial flotation recovery and with increasing HMconcentration enhance the flotation recovery.

FTIR studiesWhen Smithsonite is immersed in water, there arecarbonate, hydroxycarbonate and hydroxide. WhenAlkali is added to an aqueous (pH.9) zinc hydroxideis the stable solid.23 The ZnCO3 species after sulphidisa-tion are disappeared and partially form the ZnS layer.

The reference DRIFT spectra of smithsonite, KAX,HM, sodium sulphide and copper sulphate are shown inFig. 4.

KAX adsorption studiesFigure 5 shows the FTIR spectra of smithsonite treatedwith different concentrations of KAX in the presence ofsodium sulphide and copper sulphate solutions.

1 Zeta potential of smithsonite as function of pH in pre-

sence of distilled water, Na2S (2.661023M), CuSO4

(9.461023M) and KAX

Hosseini and Forssberg Smithsonite flotation using potassium amyl xanthate and hexylmercaptan

Mineral Processing and Extractive Metallurgy (Trans. Inst. Min. Metall. C) 2006 VOL 115 NO 2 109

Clearly in all obtained spectra the surface, thedixanthogen (AX)2 (1270–1240 cm21)25 bands are miss-ing. For all surface coverage values, the spectra exhibitabsorption bands at 1037 cm21, assigned to C5Sstretching mode24,25 and 1137 cm21, characteristics ofC–O–C stretching in dixanthogen.22 The bands relatedto the hydrocarbon chain assigned at 2856 and2956 cm21.25,26

The interaction of amyl xanthate with the surface ofsmithsonite may be represented by the followingreaction

ZnCO3(s)z2AX{(aq)'ZnAX2(s)zCO{2

3(aq) (5)

It is pointed out that the similar equation was suggestedfor adsorption of xanthate on the cerussite surfaceby some researchers.27,28 This ion exchange reaction

3 Flotation recovery and contact angles of smithsonite

as function of various concentrations of HM at pH 9

4 Reference DRIFT IR-spectra of smithsonite, KAX, HM,

sodium sulphide and copper sulphate

2 Contact angles and microflotation recoveries of smithsonite as function of various KAX concentrations in sodium sul-

phide solution (2.661022M) with copper sulphate at pH 10.5

a 4.9461024M; b 9.8861024M; c 1.4861023M; d1.9861023M; e 2.4761023M; f 2.9661023M

5 FTIR spectra of pure smithsonite treated with sodium

sulphide (2.661022M) and copper sulphate (9.461023M) at pH 10.5 with increasing initial concentration

of KAX

Hosseini and Forssberg Smithsonite flotation using potassium amyl xanthate and hexylmercaptan

110 Mineral Processing and Extractive Metallurgy (Trans. Inst. Min. Metall. C) 2006 VOL 115 NO 2

produces solid zinc xanthate at the mineral surface andreleases carbonate ions to the solution.

The area under the alkyl chain bands (3000–2800 cm21) and the adsorption density of KAX as afunction of KAX concentrations are shown in Fig. 6.

The adsorption density was calculated accordingto the measured specific surface area of smithsonite(0.43 m2 g21). The results of KAX adsorption study bymeans of DRIFT IR-spectra show the increasing KAXadsorption density with increasing initial concentrationof KAX from 0.35 (4.9461024M) to 9.26 mg m22

(2.9661023M).

HM adsorption studiesFigure 7 shows the spectra of zinc mineral in HMconcentrations at pH 9.

The bands related to the hydrocarbon chain assignedat 2850 and 2952 cm21. Since the S–H bond is weakerthan C–H and has a lower stretching absorption 2600–2500 cm21 compared with 2960–2840 cm21 for the C–Hbond.24

The double peak at 2940–2850 cm21 is caused by C–H stretching frequencies associated with the methyl(CH3) and ethylene (CH2) groups.

29 Clearly in all abovespectra the surface S–H stretching (2530 cm21) bandsare missing. Although, the band owing to the S–Hstretching vibration can be weak and even may bemissed in dilute solution. The lack of S–H stretchingbands and poor bands of C–H stretching is confirmed bysome authors’ works. They stated that the S–H bond inthe mercaptan is destroyed upon adsorption.5

The adsorbed mercaptan reacts with the surface –OHgroup forming the zinc mercaptan salt, splitting out a

molecule of water in the process according to thefollowing reaction

..

.

. . . Zn � � � OH

..

.

zHzzRS�?

..

.

. . . Zn � � � SR

..

.

zH2O (6)

The mechanism for the adsorption of mercaptan on tomalachite and chrysocolla could be similar to thementioned reaction and the process could be extendedto the attachment of mercaptan to any base metal oxideor sulphide mineral.30

The key points of the mercaptans adsorption processare the formation of a strong S–M bond. The sulphuradsorption site and the corresponding S–M bonddistance determine the nature of the interaction.16

The area under the alkyl chain bands (3000–2800 cm21) and the adsorption density of HM as afunction of various HM concentrations are shown inFig. 8.

The adsorption density was calculated according tothe measured specific surface area of smithsonite(0.43 m2 g21). The results of HM adsorption study bymeans of DRIFT IR-spectra show the increasing HMadsorption density with increasing initial concentration

6 Adsorption density and area under alkyl chain bands

(2990–2800 cm21) of DRIFT IR-spectra of smithsonite

as function of KAX concentration at pH 10.5 in sodium

sulphide (2.661022M) and copper sulphate (9.461023M) solutions

a 8.461024M; b 2.5461023M; c 4.261023M; d6.7761023M; e 8.561023M; f 1022M; g 1.161022M

7 FTIR spectra of pure smithsonite treated at pH 9 with

HM concentrations

Hosseini and Forssberg Smithsonite flotation using potassium amyl xanthate and hexylmercaptan

Mineral Processing and Extractive Metallurgy (Trans. Inst. Min. Metall. C) 2006 VOL 115 NO 2 111

of HM from 9.77 (8.461024M) to 31.95 mg m22

(1.161022M).

ConclusionsAdding sodium sulphide, copper sulphate and KAXmake their zeta potential more negative. The negativecharge of KAX is caused by the adsorption of CS2anion, which agrees well with flotation and contact angleresults.

The flotation results using KAX reveals that themaximum recovery and contact angle occurs at aroundpH 10.5 which are 81.3% and 98.7u at 2.9661023MKAX concentration in sodium sulphide and coppersulphate solutions. The highest recovery and contactangle for flotation by means of HM occurs at pH 9which are 78.6% and 92.3u respectively at HM concen-tration of 1.161022M.

The FTIR spectra revealed that the presence of CS2on the surface of smithsonite. Therefore, this hasconfirmed the adsorption of KAX onto surfaces. TheFTIR spectra in HM studies show the adsorption ofRS2 on the oxidised zinc surface and the S–H bond inthe HM is destroyed upon adsorption.

Acknowledgements

We wish to thank Calcimine Co. for providing thesamples, which were used in the present investigation.

The assistance of colleagues at Chalmers University ofTechnology, Department of Chemical Engineering,Gothenburg, Sweden, specially Mr Karl Arnby is alsomuch appreciated.

References1. J. Nagano, M. Tanaka and K. Saito: Chem. Abstr., 1975, 83,

31384.

2. M. Yamada, T. Shoji, T. Onada and J. Shimoiizaka: Chem. Abstr.,

1976, 84, 182894.

3. S. Yamazaki, N. Matusi and E. Ohtsubo: Chem. Abstr., 1978, 89,

63078.

4. G. Caproni, R. Ciccu, M. Ghiani and I. Trudu: Proc. 13th Int.

Min. Process. Cong., Warsaw, Poland, June 1979, 69–93, Elsevier.

5. Eyring and Milton E. Wadsworth: Min. Eng., 1956, 8, 531–535.

6. A. M. Gaudin: ‘Flotation’, 182–189; 1957, New York, McGraw

Hill Inc.

7. G. Rinelli and A. M. Marabini: Proc. 10th Int. Min. Process.

Cong., London, England, April 1973, Institution of Mining and

Metallugy, 492–521.

8. M. C. Fuerstenau and B. R. Palmer: in ‘Flotation – A. M. Gaudin

memorial volume’, Vol. 1, ‘Anionic flotation of oxides and silicates

flotation’, 148–196; 1976, New York, American Institute of

Mining, Metallurgical and Petroleum Engineers.

9. G. Barbery and J.-L. Cecile: ‘Process for the concentration by

flotation of fine mesh size or oxidized ores of copper, lead, zinc’,

US patent 75 38393, 1978.

10. T. Kierszincki, J. Majewski and J. Mzyk: Rudy Met., 1979, 24, (10),

498–501.

11. G. Ozbayoglu, U. Atalay and B. Senturk: Proc. Int. Conf. on

‘Recent advances in materials and mineral resources’, Penang,

Malaysia, May 1994, School of Material and Mineral Resources

Engineering, 504–509.

12. H. A. Joly, R. Majerus and K. C. Westaway:Miner. Eng., 2004, 17,

1023–1036.

13. M. A. Cook and J. C. Nixon: J. Phys. Colloid Chem., 1950, 54,

445.

14. A. M. Gaudin and D. L. Harris: Trans. AIME, 1954, 199, 925–930.

15. F. F. Aplan and P. L. de Bruyn: Trans. AIME, 1963, 229, 235–242.

16. S. A. Sardar, J. A. Syed, S. Yagi and K. Tanaka: Thin Solid Films,

2004, 450, 265–271.

17. R.-L. Xu: Longmuir, 1993, 9, (11), 2955–2962.

18. S. Quaresima, K. Sivadasan, A. Marabini, M. Barbaro and

P. Somasundaran: J. Colloid Interf. Sci., 1991, 144, (1), 159–164.

19. Y.-H. Hu, L.-L. and G.-Z. Qiu: Trans. NFsoc., 1995, 5, (4), 27–30.

20. J. S. Laskowski, Q. Liu and Y. Zhan: Miner. Eng., 1997, 10, 787–

802.

21. A. M. Gaudin, D. W. Fuerstenau and G. W. Mao: Trans. AIME,

1959, 214, 430.

22. Z. Chen and R.-H. Yoon: Int. J. Miner. Process., 2000, 58, 57–

66.

23. H. Bustamante and M. Shergold: Trans. Inst. Min. Metall., 1983,

92, 201–208.

24. J. Leja: ‘Surface chemistry of froth flotation’; 1982, New York,

Plenum Press.

25. G. Socrates: ‘Infrared characteristics group frequencies’; 1980, New

York, John Willey & Sons.

26. J. D. Miller and J. J. Kellar: ‘Advances in flotation technology’,

45–58; 1999, Littleton, CO, Society for Mining, Metallurgy and

Exploration.

27. M. G. Fleming: Trans. IMM, 1953, 521–554.

28. D. W. Fuestenau, F. Stoillo and A. Valdivieso: Proc. 15th Int.

Miner. Process. Cong., Cannes, France, June 1985, Societe de

l’Industrie Minerale, Vol. 2, 74–86.

29. G. Tarjan: ‘Mineral processing’; 1986, Budapest, Akademiai

Kiado.

30. F. F. Aplan and D. W. Fuerstenau: Int. J. Miner. Process., 1984,

13, 105–115.

8 Adsorption density and area under alkyl chain bands

(2990–2800 cm21) of DRIFT IR-spectra of smithsonite

as function of HM concentrations at pH 9

Hosseini and Forssberg Smithsonite flotation using potassium amyl xanthate and hexylmercaptan

112 Mineral Processing and Extractive Metallurgy (Trans. Inst. Min. Metall. C) 2006 VOL 115 NO 2

Paper III

XPS & FTIR Study of Adsorption Characteristics Using Cationic and Anionic Collectors on SmithsoniteSeyed Hamid Hosseini and Eric Forssberg, Journal of Minerals and Materials Characterization and Engineering, Vol.5, No.1, 2006, 21-45.

Journal of Minerals & Materials Characterization & Engineering, Vol. 5, No.1, pp 21-45, 2006 jmmce.org Printed in the USA. All rights reserved

21

XPS & FTIR Study of Adsorption Characteristics Using Cationic and Anionic Collectors on Smithsonite

Hosseini S. Hamid *and Forssberg Eric

Division of Mineral Processing, Department of Chemical Engineering and Geosciences

Luleå University of Technology, SE –971 87 Luleå, Sweden

*Corresponding author: Tel.:+46-920- 491784; fax: +46-920-97364E-mail address: hamid.hosseini@ltu.se

ABSTRACT

The adsorption of cationic and anionic collectors on the surface of smithsonite was studied using diffuse reflectance FTIR (DRIFT) and X-ray photoelectron spectroscopy (XPS or ESCA) techniques.

The FT-IR spectra studies of smithsonite conditioned using DDA (dodecylamine) show the presence of RNH2 on the surface of smithsonite and accordingly the adsorption of DDA. XPS results show the presence of a ZnS layer on the surface after sulphidising in amine adsorption. The appearance of the N (1s) signal of the amine groups and S (2p) signal of ZnS which increased in the intensity of the signal of C (1s) peak by adsorption of DDA on smithsonite. The presence of COO- on the surface of smithsonite after oleic acid treatment confirmed the adsorption of OA (oleic acid) onto the surface. The most adsorption occurs at around pH 10, when RCOO- is predominant in solution and has ample opportunities for interaction with the mineral surface.

The appearance of CS2 on the surface of smithsonite exposes the adsorption of KAX (potassium amyl xanthate) onto surface. XPS results confirm the presence of ZnS layer on the surface after sulphidising in amine adsorption and also the transferring the surface to CuS in KAX adsorption. It is suggested that copper cations exchange with those of zinc during copper activation of smithsonite such as activation of sphalerite.

Keywords: Smithsonite; Anionic collector; Cationic collector; FTIR; XPS

INTRODUCTION

A large investigation was devoted to finding the best collector for the recovery of smithsonite. Several amines were tested and a dodecylamine derived from vegetable oil proved to be the most suitable [1]. However, there is no complete analogy between the properties of metal ions in solution and on mineral surfaces, due to the various degrees of coordination saturation caused by surrounding anions. In the case of carbonate minerals

22 Hosseini S. Hamid and Forssberg Eric Vol. 5, No. 1

such as smithsonite, the surface activity of the mineral in water increases, and the adsorption of water molecules resemble chemisorption. This high activity to water dipoles is one of the main reasons for the low natural floatability of smithsonite compared with sphalerite. Hence, there is a lower possibility of effective adsorption of flotation reagent molecules on the smithsonite surface [2].

The design of reagent regimes for a selective hydrophobization of mineral surfaces by means of collectors is still an important subject in flotation research. The ionic surfactants are used with different structures for the selective flotation due to hydrophobization of the minerals. These collectors are adsorbed through electrostatic interaction of the polar group of the surfactant molecules with ions of the minerals’ surface [3].

The most important sources of zinc ore have been the oxidised zinc minerals e.g. smithsonite (ZnCO3), willemite (Zn2SiO4), hydrozincite (2ZnCO3 3Zn(OH)2), zincite (ZnO) and hemimorphite (Zn2SiO3 H2O). The flotation of oxidised lead and zinc, in particular oxidized zinc minerals, is much more difficult than the flotation of corresponding sulphide minerals [4].

Many researchers suggested that the oxidised zinc ores can be collected by flotation with long chain primary amines as collector after sulphidizing with sodium sulphide. They stressed that hexyl and amyl xanthate can be used for collecting smithsonite, but in practice the process is not selective enough. The amount of sulphidizing reagent and pH of the pulp must be carefully controlled. Sodium sulphide is a preferred soluble sulphide in comparison to other sulphides e.g. calcium sulphide, barium sulphide and ammonium polysulphide [5-8].

The activating effect of sodium sulphide is strongly time dependent. An increase in sulphidisation leads to an increase in the hydrophobicity of the mineral surface. Excess of sodium sulphide acts as a depressant for oxidised lead and zinc minerals because the adsorption of divalent sulphide ions on the surface of lead oxide minerals increases the negative charge which prevents the adsorption of collector [9]. XPS and FTIR have been found to be useful techniques for elucidating the surface properties of solids, which may be relevant in applied aspects of mineral processing [10].

In the present investigation, the adsorption behaviour effects of various amounts of cationic and anionic collectors on pure smithsonite surfaces was verified using diffuse reflectance FT-IR and XPS studies.

MATERIALS AND METHODS

Materials

The crystalline smithsonite samples (90% purity) were obtained from the Angooran zinc deposit, Iran. The XRF chemical analysis showed the pure sample contains 58.50 %

Vol.5, No.1 XPS & FTIR Study of Adsorption Characteristics 23

ZnO, 0.26 % Fe2O3, 0.29 % SiO2, 0.84 % CaO, 3.41 % Na2O, 1.06 % PbO, 0.34 % others, 35.30% L.O.I (Loss of ignition). The mineral samples were wet ground in a porcelain mill to obtain -5 μm size fractions. This final product (-5 μm) was employed for FT-IR and XPS measurements.

Reagents

Dodecylamine (99% purity) was obtained from Fluka Chemie, Switzerland. Potassium amyl xanthate with 90% purity was purchased from Shandong Qixia Flotation Reagent Co., Ltd, China and was purified by recrystallization from acetone. Oleic acid (99.9% purity), sodium sulphide and copper sulphate anhydrous (99% purity) was procured from Merck KGaA, Germany. Analytic grade HCl and NaOH were used for pH adjustment in all experiments. Deionised water (specific conductance, 0.4-0.7 μS cm-1)was used in all experiments.

Diffuse Reflectance FTIR Measurements

The mineral suspensions of smithsonite with -5 μm particle size and 1% solid ratio were first conditioned for 2-4 minutes together with NaOH or HCl as pH regulating reagent in the presence of various activators. Sodium sulphide and copper sulphate were used in KAX treatment whereas sodium sulphide as sulphidising agent was used in amine treatment. Then the samples were conditioned for 5-10 minutes with different reagents such as dodecylamine, oleic acid and KAX. The infrared spectra were obtained for all pure samples on the air dried -5 μm mineral powders before recording the DRIFT infrared spectrum.

The FTIR spectra were registered with a Perkin Elmer Spectrum 2000 FTIR-Diffuse reflectance spectrometer. This instrument is capable of data collection over a wave number range of 370-7800 cm-1 and can be configured to run in single-beam, ratio or interferogram modes. About 10 % by weight of the solid samples were mixed with spectroscopic grade KBr with a refractive index of 1.559 and a particle size of 1-5 μm. These spectra were recorded with 200 scans measured at 4 cm-1 resolution.

X-ray Photoelectron Spectroscopy (XPS) Measurements

X-ray photoelectron spectroscopy (XPS) is one of the major techniques in basic research of flotation-related surface studies of different minerals. The advantages of the method, in general, are good surface sensitivity, rather straightforward elemental and chemical state analysis and reliable quantification of the data.

The XPS spectra were obtained with an Axis Ultra electron spectrometer manufactured by Kratos Analytical. Ltd, UK. The vacuum in the sample analysis chamber was 10-9 Torr. A value of 285.0 eV was adopted as standard C (1s) binding energy. The measurements were performed on the samples used in FTIR measurement.

24 Hosseini S. Hamid and Forssberg Eric Vol. 5, No. 1

RESULTS AND DISCUSSION

FT-IR Studies of Pure Smithsonite

The smithsonite spectrum displays several bands in the region 4000-400 cm-1, Table 1 shows the assignment of pure smithsonite bands according to literature and experimental data. When smithsonite is immersed in water there are carbonate, hydroxyl carbonate, and hydroxide. H2O group physically adsorbs on the surface of the mineral. The amount of physically adsorbed water decreased by Na2S treatment which transforms into ZnS. Marabini and Rinelli stressed that there are no significant variations after sulphidization and this is expected because the ZnS and ZnCO3 bands are very similar. The ZnCO3species disappears after sulphidization and completely forms the ZnS coating in the case of very high Na2S concentrations as a form of one monolayer or a little more [11]. The reference DRIFT spectra of smithsonite, dodecylamine, KAX, sodium oleate, copper sulphate and sodium sulphide are shown in Fig. 1.

Table 1- Assignment of pure smithsonite bands according to literature and experimental data (Wave number in cm-1)

FT-IR experiment data Reference data

744.01 743 [12], 742.9 [13], 744 [14], 745 [15] 840.0 841 [13] 868.7 870 [13-15], 874.1 [12]

1089.88 1093 [12], 1096 [15] 1175 1170 [15]

1435.46 1427 [14-15],1440 [12-13] 1812.03 1830-1790 [13], 1816 [15], 1810 [14]

2492 2493 [14-15] 2850 2850 [15] 2920 2924 [15]

Adsorption of Dodecylamine

Since the functional groups of the collector absorption region closely match to the strong absorption region of smithsonite, it is difficult to identify any bands under mono layer adsorption. The absorption bands are often highly coupled and therefore it is not always possible to assign them to one specific vibration.

Figure 2 shows the spectra of pure smithsonite treated with sodium sulphide solution (2.6×10-2 M) at pH 11.5 with increasing initial concentration of dodecylamine.

Vol.5, No.1 XPS & FTIR Study of Adsorption Characteristics 25

0

1

2

3

4

5

6

7

8

9

10

60010001400180022002600300034003800

Wavenumber(cm-1)

Abs

orba

nce

C/S

Smithsonite

Dodecylamine

Potassium Amylxanthate

Sodium oleate

Copper Sulphate

Sodium sulphide

Fig. 1- Reference DRIFT IR-spectra of smithsonite, dodecylamine, KAX, sodium oleate, copper sulphate and sodium sulphide

The spectra exhibit intense absorption band corresponding to (NH2) in dodecylamine solution (1.6×10-3 M) after sulphidising with sodium sulphide (2.6×10-2

M). The reagent may be linked with Zn2+ ions through coordination bonds formed by N atoms and adsorbs to the smithsonite surface in the form of Zn-amine complexes or perhaps the hydroxyl ions presents as zinc hydroxyl species on the surface of smithsonite [16-17].

26 Hosseini S. Hamid and Forssberg Eric Vol. 5, No. 1

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

260027002800290030003100320033003400

Wavenumber(cm-1)

Abs

orba

nce

C/S

a

b

c

d

Fig. 2- FT-IR spectra of pure smithsonite treated with sodium sulphide solution (2.6×10-2 M) at pH 11.5 with increasing initial concentration of dodecylamine

(a) 10-5 M (b) 5.3×10-4 M (c) 1.1×10-3 (d) 1.6×10-3 M

Generally at the flotation pH of around 10 the RNH3+ is effectively absent and it can be assumes that the RNH2 becomes attached to the zinc on the surface in ZnS form through complexation bonds as follows [11]:

n

n

OHSorCO

ZnSorCO

RNHSorCO

)(|

|)(

3

3

23

(1)

Vol.5, No.1 XPS & FTIR Study of Adsorption Characteristics 27

The area under the alkyl chain bands (3000-2800 cm-1) and the adsorption density of DDA as a function of DDA concentrations are shown in Fig. 3. The adsorption density was calculated according the measured specific surface area of smithsonite (0.43 m2/g). The results of DDA adsorption study by means of DRIFT IR-spectra show the increasing DDA adsorption density with increasing initial concentration of dodecylamine from 0.63 mg/m2 (10-5 M) to 5.95 mg/m2 (1.62×10-3 M).

0

1

2

3

4

5

6

7

8

9

10

0.00E+00 5.00E-04 1.00E-03 1.50E-03 2.00E-03

DDA concentration (mol/l)

Alk

yl g

roup

s Are

a( c

m-1)

0

1

2

3

4

5

6

7

8

9

10

Adsorption density (m

g/ m 2)

DDA AreaDDA+Na2S AreaAdsorption Density DDAAdsorption Density DDA+Na2S

0 5×10-4 10-3 1.5×10-3 2×10-3

Fig. 3- Adsorption density and area under alkyl chain bands (2990-2800 cm-1) of DRIFT IR-spectra of smithsonite as a function of dodecylamine concentration at pH 11.5 in the

presence or absence of sodium sulphide

28 Hosseini S. Hamid and Forssberg Eric Vol. 5, No. 1

Adsorption of Oleic Acid

The peaks at 2924 and 2852 cm-1 [18-19] show that long alkyl chain is present in oleic acid treated smithsonite samples (Fig. 4). The appearance of peak 3411 cm-1 shows that -OH on the surface of smithsonite has not reacted completely.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

100014001800220026003000

a

b

c

d

e

Abs

orba

nce

C/S

Fig. 4- DRIFT IR-spectra of pure smithsonite at pH 10 with increasing initial concentration of oleic acid (a) 4.42×10-5 M (b) 1.77×10-4 M

(c) 3.54×10-4 M (d) 5.31×10-5 M (e) 1.1×10-3 M

Vol.5, No.1 XPS & FTIR Study of Adsorption Characteristics 29

It is known that RCOOH (l) and RCOOH (aq) are the predominant species, whereas species such as RCOO-, (RCOO)-2

2 and a lesser extent [(RCOO) 2H] - exist in the basic region.

The maximum adsorption of oleic acid on smithsonite around pH 10 may be attributed to interaction between RCOO- and zinc on the smithsonite surface and can be assumed that the adsorption of oleic acid takes place by an ion exchange mechanism as represented below:

OHRCOO Zn RCOOZn(OH) ---(surf)2 (2)

Accordingly, the frequencies of the bands of carboxylate ion should show intermediate values between C=O and C-O. Bands between 1538-1650 cm-1 and 1360-1470 cm-1 were assigned due to asymmetric and symmetric stretching vibrations of carboxylate ion [20]. The asymmetric carboxylate vibration band may be attributed to chemisorbed oleate [21-22].

If the oleate is present in the form of undissociated oleic acid (-COOH) ,the mean frequency of C=O stretching vibration should appear around 1690 cm-1 for dimmer and 1718 cm-1 for monomer. In the present investigation, the band around 1690 cm-1 and 1718 cm-1 was absent. It means that there is no dimmer adsorption. The broad band around 3411 cm-1 may be assigned to intermolecular hydrogen bonding of H2O molecules thus formed during adsorption.

The area under the alkyl chain bands (3000-2800 cm-1) and the adsorption density of oleic acid as a function of oleic acid concentrations are shown in Fig. 5. The results of oleic acid adsorption studies by means of DRIFT IR-spectra show the increasing oleic acid adsorption density with increasing initial concentration of oleic acid from 0.63 mg/m2 (4.42×10-5 M) to 13.1 mg/m2 (1.1×10-3 M).

Adsorption of KAX

Figure 6 shows the FTIR spectra of smithsonite treated with different concentrations of KAX in the presence of sodium sulphide and copper sulphate solutions. Clearly in all obtained spectra the surface, the dixanthogen (AX) 2 (1270-1240 cm-1) bands are missing. For all surface coverage values, the spectra exhibit absorption bands at 1041 cm-1,assigned to C=S stretching mode [3,18] and 1137 cm-1, characteristics of C-O-C stretching in dixanthogen [3]. The bands related to the hydrocarbon chain assigned at 2855, 2955 cm-1 [18-19].

30 Hosseini S. Hamid and Forssberg Eric Vol. 5, No. 1

0

5

10

15

20

25

0 0.0002 0.0004 0.0006 0.0008 0.001 0.0012

Oleic acid concentration(mol/l)

Alk

yl g

roup

are

a (c

m-1)

0

4

8

12

16

20

Adsorption density( m

g/m 2)

Alkyl group areaAdsorption density

0 2×10-4 4×10-4 6×10-4 8×10-4 10-3 1.2×10-3

Fig. 5- Adsorption density and area under alkyl chain bands (2990-2800 cm-1) of DRIFT IR-spectra of smithsonite as a function of oleic acid concentration at pH 10

The interaction of amylxanthate with the surface of smithsonite may be represented by the following reaction:

CO ZnAX2AXZnCO (aq)-232(s)

-(aq)3(s) (4)

It is pointed out that the similar equation was suggested for adsorption of xanthate on the cerussite surface by some researchers [23-24]. This ion-exchange reaction produces solid zinc xanthate at the mineral surface and releases carbonate ions to the solution. The area under the alkyl chain bands (3000-2800 cm-1) and the adsorption density of KAX as a function of KAX concentrations are shown in Fig.7.

The adsorption density was calculated according the measured specific surface area of smithsonite (0.43 m2/g). The results of KAX adsorption study by means of DRIFT IR-spectra show the increasing KAX adsorption density with increasing initial concentration of potassium amylxanthate from 0.35 mg/m2 (4.94×10-4 M) to 9.26 mg/m2(2.96×10-3 M).

Vol.5, No.1 XPS & FTIR Study of Adsorption Characteristics 31

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

9001200150018002100240027003000

a

b

c

d

e

Wavenumber (cm-1)

Abs

orba

nce

C/S

Fig.6- FT-IR spectra of pure smithsonite treated with sodium sulphide (2.6×10-2 M) and copper sulphate (9.4×10-3M) at pH 10.5 with increasing initial concentration

of KAX (a) 4.94×10-4 M (b) 1.48×10-3 M(c) 1.98×10-3 M (d) 2.47×10-3M (e) 2.96×10-3M

X-ray Photoelectron Spectroscopy (XPS) Measurements

XPS Spectra of Pure Smithsonite

The XPS spectral results of pure smithsonite demonstrate the appearance of the Zn (2p), C (1s) and O (1s). The Zn (2p) and C (1s) spectra of the smithsonite are presented in Fig. 8.

The curve fitting unveiled that the Zn (2p) spectra consist of two components, at 1022.1 and 1045.1 eV which the both components are assigned to zinc in ZnO bond [25].The C (1s) spectra shows the signal at 285, 290 eV which is assigned to carbon in C-O

32 Hosseini S. Hamid and Forssberg Eric Vol. 5, No. 1

and C=O in carbonate groups [26]. According to Fig. 8, the O (1s) spectra at 532 eV which is relevant to C-O and C=O in carbonate groups [27]. The atomic percentage of XPS spectral results shows the element percent of Zn 17.11%, C 6.65 %, and O 56.19 %.

0

2

4

6

8

10

12

14

16

0.00E+00 1.00E-03 2.00E-03 3.00E-03 4.00E-03

KAX concentration (mol/l)

Ads

orpt

ion

dens

ity (m

g/m

2 )

0

2

4

6

8

10

12

14

16

Alkyl groups area(cm

-1)

Adsorption densityAlkyl groups area

0 10-3 2×10-3 3×10-3 4×10-3

Fig. 7- Adsorption density and area under alkyl chain bands (2990-2800 cm-1) of IR-spectra of smithsonite as a function of KAX concentration at pH 10.5 in sodium sulphide

(2.6×10-2 M) and copper sulphate (9.4×10-3 M) solutions

Vol.5, No.1 XPS & FTIR Study of Adsorption Characteristics 33

Fig. 8- XPS Zn (2p), C (1s) and O (1s) spectra of pure smithsonite

34 Hosseini S. Hamid and Forssberg Eric Vol. 5, No. 1

Adsorption of Dodecylamine

XPS analysis of samples treated with DDA shows nitrogen to be present on the surface of sulphidised sample and this shows that DDA is present with the formation of DDA layer on the smithsonite surface. The C(1s), N(1s), S(2p) spectra of the smithsonite treated with DDA (1.62×10 -3 M ) are presented in Fig. 9.

The curve fitting unveiled that the C (1s) spectra consists of two components, at 285 (reference peak) and 288.1 eV. The first component is assigned to carbon in C-(C, H) bond and second one to the carbon in the C-N [28].

The N (1s) signal of the amine groups at 399.6 eV can be assigned to nitrogen in R-NH2 bond and confirms the existent of DDA on the surface of mineral [29]. The S (2p) spectra of pure sample conditioned with 1.62×10 -3 M DDA show the broad peak at around 162.1 eV confirms that ZnS is present on the mineral surfaces [30].

The XP spectra results of pure smithsonite conditioned with Na2S (2.6×10 -2 M) and DDA (1.62×10 -3 M) demonstrate the appearance of the N(1s) signal of the amine groups and S(2p) signal of ZnS which increased in the intensity of the signal of C(1s) peak by adsorption of DDA on smithsonite with a simultaneous decrease in the peak intensities of oxygen.

Figure 10 shows the atomic percent of elements at the mineral surface. As it can be seen, there is increasing the carbon atomic percent after conditioning the pure mineral which show increasing the adsorption of DDA on the surface of mineral. The maximum adsorption occurs at the DDA concentration (1.62×10 -3 M). The N (1s) spectra and also the nitrogen atomic percent show the same total atomic concentrations of nitrogen at higher concentration of DDA.

Figure 11 shows the effect of dodecylamine conditioned on sulphidised smithsonite with 2.6× 10-2 M of sodium sulphide.

The large increase in O/Zn and N/Zn values is noted with increasing the reagent concentration. There is always an excess of oxygen compared with that attributed to the smithsonite (ZnCO3), as a result of both reagents used together. Initially the S/Zn value decreases and then increases, which may be attributed to substitution of a pre-existing sulphide species. The small increase in S/Zn may be due to the adsorption of dodecylamine.

Adsorption of Oleic Acid

The XPS C (1s) and O (1s) spectra of pure smithsonite conditioned in solution of oleic acid at pH 10 are shown in Fig. 12. The C (1s) spectra consist of peaks 285 eV (reference peak) and 290.2 eV, corresponding to alkyl and carboxylate carbon respectively.

Vol.5, No.1 XPS & FTIR Study of Adsorption Characteristics 35

Fig. 9- XPS spectra of pure smithsonite conditioned in solution of DDA (pH 11)

36 Hosseini S. Hamid and Forssberg Eric Vol. 5, No. 1

0

20

40

60

80

100

Pure mineral 0.1g/l 0.2 g/l 0.3 g/l

%Zinc%Sulphur%Nitrogen

%Carbon%Oxygen

Pure mine ral 5.4×10-4 M 1.1×10-3 M 1.6×10-3 M

Fig. 10- Atomic percent elements on to smithsonite of surface after adsorption of DDA

The O (1s) spectra measured consist of the peak 533.5 eV, corresponding to O (1s) in carboxylate group [31]. It is obvious that the most adsorption occurs at around pH 10 and this is because of the existence of RCOO- is predominant in solution and has interaction with mineral surface.

Figure 13 shows the atomic percent of elements at the mineral surface. As it can be seen, there is increasing the carbon atomic percent after conditioning the pure mineral which show increasing the existence of alkyl chain at the surface and conclusively increasing the adsorption of oleic acid on the surface of mineral. The maximum adsorption occurs for the OA concentration of 1.1× 10-3 M at pH 10. The XPS analysis of samples treated with oleic acid shows that OA is present with the formation of OA layer on the smithsonite surface. Fig. 14 shows the effect of oleic acid conditioned on pure smithsonite with 1.1× 10-3 M of oleic acid.

Vol.5, No.1 XPS & FTIR Study of Adsorption Characteristics 37

0

1

2

3

4

5.00E-04 9.00E-04 1.30E-03 1.70E-03

DDA concentration(mol/l)

O/Z

n or

S/Z

n

0

0.5

1

1.5

N/Z

n

O/Zn

S/Zn

N/Zn

5×10-4 9×10-4 1.3×10-3 1.7×10-3

Fig. 11- The effect of DDA treatment on sulphidized smithsonite with 2.6× 10-2 M of Na2S

The increase in O/Zn and C/Zn values with increasing the reagent concentration at pH 10 shows the increasing of carbon content regarding carbon chain in reagent. There is always an excess of oxygen compared with that attributed to the smithsonite (ZnCO3), as a result of reagent used and the increasing of Zn content in media.

38 Hosseini S. Hamid and Forssberg Eric Vol. 5, No. 1

Fig. 12- XPS C (1s) and O (1s) spectra of pure smithsonite conditioned in solution of oleic acid (pH 10)

Vol.5, No.1 XPS & FTIR Study of Adsorption Characteristics 39

0

20

40

60

80

100

Puremine ral

pH 8 pH 9 pH 10 pH 11

% Zinc %Carbon%Oxygen

Fig. 13- Atomic percent elements on smithsonite surface after treatment with oleic acid (1.1×10-3 M)

40 Hosseini S. Hamid and Forssberg Eric Vol. 5, No. 1

0

10

20

30

40

50

60

7.5 8 8.5 9 9.5 10 10.5 11 11.5

pH

O/Zn

0

20

40

60

80

100

120

140

C/Zn

O/Zn

C/Zn

Fig. 14- The effect of oleic acid (1.1× 10-3 M) treatment on pure smithsonite at different pH

Adsorption of KAX

The XPS C (1s), S (2p), O (1s) and Cu (2p) spectral results of pure smithsonite conditioned in solution of KAX (2.96 × 10-3 M) at pH 10.5 are illustrated in Fig. 15.

The C (1s) spectra consists of peaks 285 eV (reference peak), 286.4 and 288.4 eV, corresponding to alkyl, CS2 and CO carbon respectively [32-33]. The intensity of characteristic peak in spectra for alkyl group (285.0) shows the increasing peak intensity in comparison to C (1s) spectra in pure smithsonite. It shows that there is adsorption to form a film at the surface of mineral.

Vol.5, No.1 XPS & FTIR Study of Adsorption Characteristics 41

Fig. 15- XPS spectra of pure smithsonite conditioned in solution of KAX (2.96 × 10-3 M) at pH 10.5

The S (2p) spectra measured consists of the peak 161.6 eV, corresponding to S (2p) in ZnS layer onto the mineral surface [34].

The changes in the C1s and O1s emissions during the adsorption treatment show that the adsorption of xanthate occurs. The detailed mechanism by which the exchange takes place is not clear and obviously cannot be concluded from the above observations. Slight changes in the positions and shapes of the components due to the xanthate group are observed during the adsorption process.

42 Hosseini S. Hamid and Forssberg Eric Vol. 5, No. 1

The Cu (2p) spectra measured consist of the peak 932.1 eV, corresponding to Cu (2p) in CuS [34] and 952.3 eV, corresponding to Cu2O layer on the mineral surface [25]. The atomic concentration of the Cu peaks shows the more concentration for CuS in comparison of Cu2O. It may be suggested that copper cations exchange with those of zinc during copper activation of smithsonite such as activation of sphalerite. Figure 16 shows the atomic percent of elements at the mineral surface before and after treating with KAX.

As it can be seen, there is increasing the carbon atomic percent and decreasing the oxygen content of mineral surface after conditioning which show increasing the existence of alkyl chain at the surface and conclusively increasing the adsorption of KAX on the surface of mineral. On the other hand, it can be observed the increasing the copper and sulphur percent of the surface with increasing the concentration of KAX in solution. This shows that the ion exchange of Zn2+ and Cu2+ and existence of CuS layer on the surface of mineral. The maximum adsorption occurs for the KAX concentration of 2.96×10-3 M at pH 10.5. The XPS analysis of samples treated with KAX shows that KAX is present with the formation of KAX layer on the smithsonite surface.

0

20

40

60

80

100

Pure mineral 0.1 g/l 0.3 g/l 0.6 g/l

%Zinc%Sulphur%Copper

i

%Carbon%Oxygen

Pure mineral 4.94×10-4 M 1.48×10-3 M 2.96×10-3 M

Fig. 16- Atomic percent elements on smithsonite surface after adsorption of KAX

Vol.5, No.1 XPS & FTIR Study of Adsorption Characteristics 43

CONCLUSIONS

1. According to the results, the presence of RNH2 on the surface of smithsonite was confirmed. Hence, this is showed the adsorption of DDA onto smithsonite surface. XPS results confirmed the presence of ZnS layer on the surface after sulphidising in amine adsorption. The appearance of the N (1s) signal of the amine groups and S (2p) signal of ZnS which increased in the intensity of the signal of C (1s) peak by adsorption of DDA on smithsonite with a simultaneous decrease in the peak intensities of oxygen.

2. The FTIR spectra revealed the presence of COO- on the surface of smithsonite and this is confirmed the adsorption of OA onto surface. It is obvious that the most adsorption occurs at around pH 10 and this is because of the existence of RCOO- is predominant in solution and has interaction with mineral surface.

3. Experimental findings showed the presence of CS2 on the surface of smithsonite and the adsorption of KAX onto surface. XPS results show the presence of ZnS layer on the surface after sulphidising and also the transferring the surface to CuS in KAX adsorption. It is suggested that copper cations exchange with those of zinc during copper activation of smithsonite such as activation of sphalerite.

Acknowledgements

We wish to thank Calcimine Co. for providing the samples, which were used in this investigation. The authors thank Dr. A. V. Shchukarev, Department of Inorganic Chemistry, Umeå University and also Mr. urban Jelvestam, Material Science Department, Chalmers university of Technology, Sweden, for the XPS measurements.

References

1. Billi, M., and Quai, V., 1963. Development and results obtained in the treatment of zinc oxide ores at AMMI mines'', IMPC, London, paper 43.

2. Glembotskii, V. A., 1972. Flotation, primary sources, New York, pp.185-189. 3. Leja, J., 1982. Surface chemistry of froth flotation. Plenum Press, New York. 4. Rey, M., 1954. Flotation of Oxidized Zinc Ores. Mining Engineering.5. Gaudin, A. M, 1957. Flotation. McGraw Hill Inc., New York, pp.182-189. 6. Weiss, N. L., 1985.SME Mineral Processing Handbook. AIME, pp. 15-4 15-7.7. Rey, M., and Raffinot P., 1953. The flotation of oxidized zinc ores, Recent

Developments in mineral dressing symposium. IMM, London. 8. McGarry, P. E., Pacic, Z., 1981. Flotation of Nonsulfide Zinc Materials. United States

patent: 4253614.

44 Hosseini S. Hamid and Forssberg Eric Vol. 5, No. 1

9. Ozbayoglu, G., Atalay, U., and Senturk, B., 1994. Flotation of lead and zinc carbonates ore. Recent advances in materials and mineral resources, Penang, Malaysia, pp.504-509.

10. Berry, Frank J., 1985. Mineral surfaces and Chemical bond, in Chemical bonding and spectroscopy in mineral chemistry. Frank J. Berry and David J. Vaughan (Ed.), London, Chapman and Hall, pp.293-315.

11. Marabini, A. M., Alesse, V., Garbassi, F., 1984. Role of sodium sulphide, xanthate and amine in flotation of lead-zinc oxidized ores. Inst of Mining & Metallurgy, pp.125-136.

12. Farmer, V. C., 1974. The infrared spectra of minerals. Mineralogical society, pp.239. 13. Gadsden, J. A., 1975. Infrared Spectra of Minerals and Related Inorganic

Compounds. Butterworth, pp.66. 14. Ferraro, John R., 1982. The Sadtler infrared spectra hand book of minerals and clays.

Saddler.15. Jones, G. C., Jackson, B., 1993. Infrared Transmission spectra of carbonate minerals.

Chapman & Hall, London. 16. Pascal, P., 1962. Complexes du zinc, '' Nouveau traite de chimie minerale, Masson,

Paris, pp.318-321.17. Healy, T. W., and Moignard, M. S., 1976. A review of electrokinetic studies of metal

sulfides, Flotation: A. M. Gaudin Memorial Volume, Fuerstenau M. C. (Ed.), AIME , New York, pp. 334-363.

18. Socrates, G., 1980. Infrared Characteristics Group Frequencies. John Wiley & Sons, Ltd., New York.

19. Miller, J. D., and Kellar, J. J., 1999. Internal reflection spectroscopy for FTIR analysis of carboxylate adsorption by semi soluble salt minerals. Advances in Flotation Technology. Society for Mining, Metallurgy, and Exploration, Inc., Littleton, CO, pp.45-58.

20. Smith, B. C., 1998. Infrared Spectral Interpretation: A systematic Approach. CRC Press, Washington DC.

21. Gong, Wen Qi, Parentich, A., Little, L., H. and Warren, L. J., 1992. Adsorption of oleate on apatite studied by diffuse reflectance infrared Fourier transform spectroscopy. Lagmuir 8, pp. 118 – 124.

22. Jang, Woo-Hyuk, Drelich, J., and Miller, Jan D., 1995. Wetting Characteristics and Stability of Langmuir-Blodgett Carboxylate Monolayers at the Surfaces of Calcite and Fluorite. Langmuir 11, pp.3491 – 3499.

23. Fleming, M. G., 1953.Effect of soluble sulphide in the flotation of secondary lead mineral. Trans. IMM, London, pp.521-554.

24. Fuerstenau, D. W., Stoillo, F., Valdivieso, A., 1985a.Sulfidization and flotation behaviour of anglesite, cerussite and galena. Proceeding. XV International Mineral Processing Congress, Cannes, France, pp.74-86.

25. Maroie, S. Haemers, G. Verbist, J. J., 1984.Surface oxidation of polycrystalline "alpha" ( 75% Cu et 25% Zn ) and "beta" ( 53% Cu et 47% Zn ) brass as studied by XPS : influence of oxygen pressure. Applications of Surface Science,Vol.17, pp.463-476.

Vol.5, No.1 XPS & FTIR Study of Adsorption Characteristics 45

26. Bichler, C. H., Bischoff, M., Langowski, H.-C., Moosheimer, U., 1996. The Substrate-Process Interface in Thin Barrier Film Coating. 39th Annual Technical Conference of the Society of Vacuum Coaters, Philadelphia.

27. Bou, M., Martin, J. M., Le Mogne, T. H., Vovelle, L., 1991. Chemistry of the interface between aluminium and polyethyleneterephtalate by XPS. Applied Surface Science, Vol. 47, pp.149-161.

28. Delpeux, S., Beguin, F., Benoit, R., Erre, R., Manolova, N., Rashkov, I., 1998.Fullerene core star-like polymers-1. Preparation from fullerenes and monoazidopolyehers.Eur. Polym. J., Vol. 34, No.7, pp.905-915.

29. Lim, A. S., Atrens, A., 1990. ESCA studies of Nitrogen-Containing Stainless Steels.Applied Physics A, Vol. 51, pp. 411-418.

30. Brion, D., 1980, ''Etude par spectroscopie de photoélectrons de la dégradation superficielle de FeS2, CuFeS2, ZnS et PbS à l'air et dans l'eau'', Applications of Surface Science, Vol 5, pp.133-152.

31. Clark, D. T., Thomas, H. R., 1978.Applications of ESCA to polymer Chemistry, XVII. Systematic Investigation of the Core Levels of Simple Homopolymers, Journal of Polymers Science, Polymer Chemistry Edition, Vol. 16, pp. 791-820.

32. Liao, H. M., Sodhi, R. N. S., Coyle, T. W., 1993. Surface composition of Al N powders studied by x-ray photoelectron spectroscopy and bremsstrahlung-excited Auger electron spectroscopy. J. Vac. Sci. Technol. A, Vol. 11, No. 5, pp.2681-2686.

33. Wagner, C.D., Riggs, W. M., Davis, L. E., Moulder, G. F., 1979. Handbook of X-ray photoelectron spectroscopy, Minnesota, Perkin –Elmer Corporation.

34. Briggs, D., Seah, M. P., 1993. Practical surface analysis, John Wiley & sons. Vol. 1, 2nd edition.

Paper IV

Physicochemical Studies of Smithsonite Flotation using Mixed Cationic/Anionic CollectorSeyed Hamid Hosseini and Eric Forssberg, Minerals Engineering, Volume 20, Issue 6, 2007, 621-624.

Technical note

Physicochemical studies of smithsonite flotation using mixedanionic/cationic collector

S. Hamid Hosseini *, Eric Forssberg

Division of Mineral Processing, Department of Chemical Engineering and Geosciences, Lulea University of Technology, SE-971 87 Lulea, Sweden

Received 17 May 2005; accepted 13 December 2006Available online 2 February 2007

Abstract

In this study, the flotation behaviour and surface adsorption of smithsonite were investigated using various concentration ratios ofpotassium amyl xanthate (KAX) and dodecylamine (DDA) in a surfactant mixture. The use of either KAX or DDA during flotationresulted in an increase in smithsonite recovery as the collector concentration increased. Further, the smithsonite recoveries were forthe most part less than 40% irrespective of collector concentration. However, when a mixture of KAX and DDA was used, smithsoniterecovery increased dramatically. The FT-IR spectra show co-adsorption of the amine–xanthate complex using a collector mixture. Thepresence of KAX in the mixture decreases the electrostatic head–head repulsion between the surface and ammonium ions and increasesthe lateral tail–tail hydrophobic bonds.� 2007 Elsevier Ltd. All rights reserved.

Keywords: Smithsonite; Mixed collector; Anionic/cationic collector

1. Introduction

The flotation of oxidized lead and zinc minerals, partic-ularly zinc minerals, is much more difficult than the flota-tion of corresponding sulphide minerals (Rey, 1953).Hexyl and amyl xanthates are capable of collecting smith-sonite; however, the process is not selective enough in prac-tice (Gaudin, 1957).

The amount of sulphidizing reagent and pH of the pulpmust be carefully controlled in amine flotation (Rey et al.,1954). When the pH value decreases, there is a drop inrecovery (Rausch and Mariacher Burt, 1970).

A mixture of amines and xanthates can be used as a col-lector (Tarjan, 1986). A system that contains two surfac-tants of different charge is called a catanionic system. Theattraction between the differentially charged head groupswill lead to a decrease in the area per head group (Herring-ton et al., 1993).

The interaction between oppositely charged surfactants isstrong due to electrostatic interactions. Ion–pair associationand complex coacervate formation will occur when they aremixed in aqueous solutions (Tomlinson et al., 1979).

In the present study, the effect of various amounts ofmixed collectors (KAX + DDA) on smithsonite flotationhas been investigated by considering various parameterssuch as zeta potential, contact angle, microflotation anddiffuse reflectance FT-IR studies. The results are discussedand the role of mixed collectors in smithsonite flotation isoutlined.

2. Materials and methods

The smithsonite samples (purity of 90%) were collectedfrom the Angooran region of Iran. XRF chemical analysisindicated that the sample contained: 58.50% ZnO, 0.26%Fe2O3, 0.29% SiO2, 0.84% CaO, 3.41% Na2O, 1.06%PbO, 0.34% others, and 35.30% LOI.

KAX, purchased from Shandong Qixia FlotationReagent Co., Ltd., China, was purified using the recrystal-lization process from acetone. DDA (99% purity) was

0892-6875/$ - see front matter � 2007 Elsevier Ltd. All rights reserved.

doi:10.1016/j.mineng.2006.12.001

* Corresponding author. Tel.: +46 920491692; fax: +46 92097364.E-mail address: hamid.hosseini@ltu.se (S.H. Hosseini).

This article is also available online at:

www.elsevier.com/locate/mineng

Minerals Engineering 20 (2007) 621–624

obtained from Fluka Chemie, Switzerland. Analyticalgrade HCl and NaOH were used as pH modifiers. Deion-ised water was used in all experiments.

Fig. 1 shows the experimental procedure employed. A200 g sample of smithsonite was ground in a laboratoryball mill for 20 min. The ground smithsonite was wetscreened to extract the �106/+75 lm fraction for microflo-tation tests. The �75 lm fraction was ground further usinga disc mill. The �5 lm fraction was collected from theground product using a sedimentation technique.

A 1 g sample of the �106/+75 lm fraction suspended indeionised water was conditioned with pH modifier toachieve pH 9.5 before conditioning the slurry with collec-tor. Microflotation was conducted in a 100-ml Hallimondtube at a constant air flow rate of 20 cm3/min. The flota-tion concentrate and tailings were filtered, dried andweighed.

The zeta potential, contact angle and FT-IR spectra ofsmithsonite were measured on samples ground to �5 lm.The zeta potential of smithsonite at various pH valueswas recorded using a Coulter Delsa 440 instrument. Theaverage value of the zeta potential of four angles wasreported as the final result. The temperature was kept con-stant at 25.0 ± 0.1 �C.

The contact angle was determined on a Fibro DAT 1100dynamic adsorption tester. The interaction between aliquid droplet and the specimen surface was measured.

The FT-IR spectra were recorded with a Perkin–ElmerSpectrum 2000 FT-IR-Diffuse reflectance spectrometer.These spectra were recorded with 200 scans measured at4 cm�1 resolution.

3. Results and discussion

3.1. Microflotation and contact angle studies

The flotation recoveries and measured contact angles asa function of KAX or DDA concentration are given inFig. 2. The results indicate that the flotation recoveryand contact angle increase with increasing KAX concentra-tion in the mixed collector, and the flotation recovery andcontact angle are enhanced to 96.6% and 117.5�, respec-tively. Note, the recoveries and contact angles were muchlower for KAX and DDA when used alone.

Mineral Sample

GravitySedimentation

WetGrinding

-75 μm

Grinding

Wet Sieving

-106 +75 μm

Microflotation

FT-IR Zeta-

potentialContactangle

Fig. 1. Test procedure of this study.

0

10

20

30

40

50

60

70

80

90

100

KAX or DDA concentration (mol/l)

Flo

tati

on r

ecov

ery

(%)

0

20

40

60

80

100

120

140Re

c

0 10-3 2×10-3 3×10-3 4×10-3

Recovery in KAX+DDA

Recovery in KAX only

Recovery in DDA only

Contact angle for KAX+DDA

Contact angle for KAX only

Contact angle for DDA only

5

Contact angle (degree)

×10-3

Fig. 2. Flotation recovery and contact angle of smithsonite as a function of KAX or DDA concentration in the presence of KAX only, DDA only andmixed collector (KAX + DDA) including 1.1 · 10�3 M of DDA and various KAX concentrations at pH 9.5.

622 S.H. Hosseini, E. Forssberg / Minerals Engineering 20 (2007) 621–624

3.2. Zeta-potential studies

Fig. 3 shows the zeta potential of smithsonite as a func-tion of pH. The isoelectric point of smithsonite is about 8.0in distilled water, which agrees well with values reported byother researchers (Quaresima et al., 1991; Hu et al., 1995).The zeta potential in KAX or DDA when used alone

appears to be very similar to distilled water. This may beattributed to low adsorption of the desired collectors, whenused separately, on the smithsonite surface. Hence, the zetapotential in the mixed collector (KAX + DDA) increasedtheir negative charge with increasing KAX concentrationdue to adsorption of C@S on the mineral surface. Theseresults indicate that the mixed collector is adsorbed

-60

-50

-40

-30

-20

-10

0

10

20

30

40

4 5 6 7 8 9 10 11 12 13 14

pH

Zet

a po

tent

ial (

mV

)

in

in

in distilled water

in DDA (1.1×10-3M) and KAX(1.1×10-3M)

in DDA (1.1×10-3M) and KAX(2.2×10-3M)

in DDA (1.1×10-3M) and KAX(3.6×10-3M)

in 2.96×10-3M KAX

in 1.1×10-3M DDA

Fig. 3. Zeta-potential of smithsonite as a function of pH in the presence of distilled water, KAX only (2.96 · 10�3 M), DDA only (1.1 · 10�3 M) andmixed collector (KAX + DDA) including 1.1 · 10�3 M of DDA and various KAX concentrations of 1.1 · 10�3 M, 2.2 · 10�3 M and 3.6 · 10�3 M.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

1000120014001600180020002200240026002800300032003400

2950

2917

Wavenumber (cm-1)

Abs

orba

nce

C/S

2850

3325

1048

1098

1196

2847

2913

2943

3315

1189

1030

2950

Fig. 4. FT-IR spectra of smithsonite in the presence of: (a) 2.96 · 10�3 M of KAX only, (b) 1.1 · 10�3 M of DDA only and mixed collector(KAX + DDA) including 1.1 · 10�3 M of DDA and various KAX concentrations, (c) 4.94 · 10�4 M, (d) 9.88 · 10�4 M, (e) 1.48 · 10�3 M, (f)1.98 · 10�3 M, (g) 2.47 · 10�3 M, and (h) 3.6 · 10�3 M, in the alkyl and functional group region at pH 9.5.

S.H. Hosseini, E. Forssberg / Minerals Engineering 20 (2007) 621–624 623

predominantly on the negatively charged part of the smith-sonite surface at alkaline pH.

3.3. FT-IR studies

The adsorption ofKAX only (2.96 · 10�3 M), DDAonly(1.1 · 10�3 M) and also KAX adsorption in DDA solution(1.1 · 10�3 M) in the alkyl and functional group region isgiven in Fig. 4. At a fixed DDA concentration (1.1 ·10�3 M), the KAX concentration is increased in each exper-iment. Clearly in all obtained spectra, the dixanthogen(AX)2 (1270–1240 cm�1) (Leja, 1982) bands are missing.The spectra exhibit absorption bands at 1048 cm�1, assignedtoC@S stretchingmode and 1098 and 1196 cm�1, character-istic of C–O–C stretching. The bands related to the hydro-carbon chain occur at 2850, 2917 and 2950 cm�1. Theintensity of alkyl chain bands increases with increasingKAX concentration corroborating increased adsorption ofmixed collector. The spectra also exhibit an intense absorp-tion band corresponding to t (NH2) at 3325 cm�1. Theincrease inDDAadsorptionwith increasedKAXconcentra-tion indicates the decrease in the electrostatic head–headrepulsion between the surface and ammonium ions whilethe lateral tail–tail hydrophobic bonds increase. This con-firms the co-adsorption of amine–xanthate complex whenDDA and KAX were used as a mixed collector.

The areas under the alkyl chain bands (3000–2800 cm�1)of KAX only, DDA only and mixed collector (KAX +DDA) as a function of KAX or DDA concentration isdepicted in Fig. 5. These results show that the areas underthe alkyl chain bands are much lower for KAX and DDAwhen used alone. It also indicates the increase in area underthe alkyl chain bands for the mixed collector with increas-ing initial concentration of KAX from 8.87 cm�1 (4.94 ·10�4 M) to 18.77 cm�1 (3.6 · 10�3 M) at pH 9.5.

4. Conclusions

The flotation results confirmed that at a fixed amount ofDDA (1.1 · 10�3 M), increasing KAX concentration iseffective in recovery as the highest KAX concentration(3.6 · 10�3 M) enhanced the flotation recovery to 96.6%at 117.5� contact angle. This can be attributed to increasedadsorption of mixed collector on the smithsonite surface.

Using the mixed collector makes their zeta potentialmore negative in comparison with DDA or KAX alone.This proves the co-adsorption of cationic/anionic collectoron the smithsonite surface and increases the surface hydro-phobicity and accordingly enhances the flotation recovery.

The FT-IR spectra showed the presence of RNH2 andC@S on the surface of the mineral. The presence of KAXincreased the DDA adsorption due to the decrease in theelectrostatic head–head repulsion between the surface andammonium ions and increase in the lateral tail–tail hydro-phobic bonds. The adsorption study of KAX only, DDAonly and mixed collector (KAX + DDA) showed theincrease in mixed collector adsorption on the smithsonitesurface with increasing KAX concentration, which agreeswell with the flotation recovery, contact angle, zeta poten-tial and FT-IR results.

References

Gaudin, A.M., 1957. Flotation. McGraw Hill Inc., New York, pp. 182–189.

Herrington, K.L., Kaler, E.W., Miller, D.D., Zasadzinski, J.A., Chiru-volu, S., 1993. Phase behaviour of aqueous mixtures of dodecyltrim-ethylammonium bromide (DTAB) and sodium dodecyl sulfate (SDS).Journal of Physical Chemistry 97 (51), 13792–13802.

Hu, Yuehua, Luo, Lin, Qiu, Guangzhou, 1995. Solution chemistry ofelectrokinetic behaviour of carbonate minerals. Transactions of NFsoc5 (4), 27–30.

Leja, J., 1982. Surface Chemistry of Froth Flotation. Plenum Press, NewYork.

Quaresima, S., Sivadasan, K., Marabini, A., Barbaro, M., Somasundaran,P., 1991. Behaviour of colloidal suspensions of zinc carbonate in thepresence of copolymers designed for selective flocculation. Journal ofColloid and Interface Science 144 (1), 159–164.

Rausch, D.O., Mariacher, Burt C. 1970. Concentration of oxide ores atTynagh, mining and concentrating of lead and zinc, AIME WorldSymposium on Mining and Metallurgy of Lead and Zinc, vol. 1,Extractive metallurgy of lead and zinc, pp. 721–731.

Rey, M., 1953. The flotation of oxidized ores of lead, copper and zinc’,Recent Developments in Mineral Dressing Symposium. IMM, Lon-don, pp. 541–548.

Rey, M., Sitia, G., Raffinot, P., Formanek, V., 1954. Flotation of oxidizedzinc ores. Mining Engineering, 416–420.

Tarjan, G., 1986. Mineral Processing, Akademiai Kiado, Budapest,Hungary.

Tomlinson, E., Davis, S.S., Mukhayer, G.I., 1979. Ionic interaction andphase stability. In: Mittal, K.L. (Ed.), In: Solution Chemistry ofSurfactants, vol. 1. Plenum Press, New York, pp. 3–43.

0

2

4

6

8

10

12

14

16

18

20

KAX/DDA concentration (mol/l)

Alk

yl g

roup

s ar

ea (

cm-1

)

KAX only

DDA only

KAX+DDA

10-3 2×10-3 3×10-3 4×10-30

Fig. 5. The area under alkyl chain bands (2990–2800 cm�1) of DRIFT IR-spectra of smithsonite in the presence of DDA only, KAX only and mixedcollector (KAX + DDA) at pH 9.5.

624 S.H. Hosseini, E. Forssberg / Minerals Engineering 20 (2007) 621–624

Paper V

Flotation behaviour of oxide zinc ore from Angooran deposit, Iran in the presence of Cationic/Anionic and Mixed (Cationic/Anionic) collectors Seyed Hamid Hosseini and Eric Forssberg, European Journal of Mineral Processing & Environmental Protection, Vol. 6, No. 3, 2006 (In press)

1

Flotation behaviour of oxidised zinc ore from Angooran deposit, Iran in the presence of Cationic, Anionic

and Mixed (Cationic/Anionic) collectors

S. Hamid Hosseini * and Eric Forssberg Division of Mineral Processing, Department of Chemical Engineering and Geosciences

Luleå University of Technology, SE –971 87 Luleå, Sweden

ABSTRACT

In this study, the flotation of oxidised zinc ore from Angooran ore in the presence of cationic collector such as dodecylamine (DDA), anionic collectors such as oleic acid (OA) and potassium amyl xanthate (KAX) and mixed collector (cationic/anionic) was investigated. The parameters of the flotation process such as recovery and grade and the effect of using some collectors, sodium sulphide and copper sulphate at different pH were investigated. According to this investigation, maximum flotation recovery was found as 84.5% with 24.5% Zn content using cationic collector (dodecylamine) at pH 11.5. The flotation results using mixed collector (dodecylamine and KAX) showed when the KAX is increased, the recovery is increased but it has no significant variation in comparison with DDA flotation results. The lowest recovery is seen, when KAX used alone in batch flotation and it shows the poorer flotation results in comparing to the other reagents. The results of oleic acid flotation also have no significant variation for recovery in comparison with cationic flotation but no selectivity is observed.

Keywords: Angooran zinc ore; Mixed collector; Dodecylamine; Oleic acid; KAX; Bench flotation

1. INTRODUCTION

One of most important sources of zinc ore have also long been the oxidised zinc minerals such as smithsonite (ZnCO3), willemite (Zn2SiO4), hydrozincite (2ZnCO3.3Zn(OH)2), zincite (ZnO) and hemimorphite (Zn2SiO3.H2O). However, flotation of oxidised lead and zinc, in particular zinc minerals, is much more difficult than the flotation of corresponding sulphide minerals (Zhao and Standford ,2000).Oxidised zinc ores have long been an important source of zinc which can be floated with long chain primary amines as collector after sulphidizing with sodium sulphide (Weiss, 1985). Oxidised zinc ore may be concentrated by flotation in the presence of a soluble sulphide and soluble compound of an aliphatic amine containing from 8-18 carbon atoms (McKenna et al., 1949). Abramov also stated the best process for the flotation of smithsonite was by using of laurylamine acetate as a collector after sulphidizing the mineral with sodium sulphide. His results showed that the flotation properties of minerals depended on the pH (Abramov, 1961). Generally, flotation of oxidised lead and zinc, in particular zinc minerals, is much more difficult than the flotation of corresponding sulphide minerals. ____________* Corresponding author. E-mail: hamid.hosseini@ltu.se

2

The main flotation characteristics of these minerals are lower selectivity and higher reagent consumption. The most common flotation technique to be used commercially is sulphidisation with Na2S, followed by treatment with conventional collectors, namely xanthates for lead oxides and amine for zinc oxides (Rey, 1954). A large part of the test programme was devoted to finding the best collector for the Smithsonite. Several amines were tested but a dodecylamine derived from vegetable oil proved to be the most suitable (Billi and Quai, 1963). According to previous works, the size fraction of -125+75 μm to decrease the over grinding was selected as the best amount of the liberation degree and the result of HLS confirmed it. However, the applicability of the gravity concentration was in size fraction of –150+125 μm (Hosseini and Forssberg, 2001). The feed to the oxide zinc flotation circuit requires careful desliming prior to flotation and is then floated with a relatively large amount of sulphidizing agent and a cationic collector, with frother added as required. Investigators originally reported best results at pH levels between 10.5 and 11.5, although some ores respond well to the process at lower PH levels. Reagent consumptions are usually of the order of 1000 g/t to 7500-g/t sodium sulphide or sodium hydrosulphide, and 50 g/t to 300 g/t cationic collector. Gaudin stressed that hexyl and amyl xanthate can be used for collecting of smithsonite but the process is not selective enough for practice (Gaudin, 1957). Barbery et al., 1977, reported that xanthates, simple alkyl amines such as dodecylamine and also fatty acids can be used for flotation of oxidised zinc ore. Sodium sulphide proved to be the most satisfactory sulphidizing agent, both from the point of view of being cheaper than sodium hydrogen sulphide and also by virtue of generating a high pH. When the pH dropped there was a drop in recovery (Rausch and Mariacher, 1970). The amount of sulphidizing reagent and pH of the pulp must be carefully controlled (Rey, 1953). Sodium sulphide is a preferred soluble sulphide, other sulphides, which may be used, include, for example, calcium sulphide, barium sulphide and ammonium polysulphide (McGarry and Pacic, 1981) The activating effect of sodium sulphide is strongly dependent on time. The increase in sulphidisation leads to an increase in the hydrophobicity of the surface of minerals. Excess of sodium sulphide acts as depressant for oxidised lead and zinc minerals because adsorption of divalent sulphide ion on the surface of lead oxide minerals increase the negative charge which prevents the adsorption of collector (Ozbayoglu et al., 1994). Önal et al., 2005 showed that the increasing the sodium sulphide from 2.25 to 4.5 g/l decreased the zinc flotation recovery (Önal et al., 2005). In the present study, the flotation behaviour of oxidised zinc ore from Angooran deposit, Iran was studied using several cationic, anionic and mixed (cationic/anionic) collectors by means of bench scale flotation studies.

2. MATERIALS AND METHODS

The oxidised zinc ore samples were obtained from Angooran deposit, Iran. The chemical analysis showed the representative ore sample contains 9.6 % ZnO, 62.5 % SiO2, 16.5 % Al2O3, 2.25 % K2O, 1.85 % Fe 2O3, 1.36 % PbO and 0.45 % TiO2. Bycombining the information obtained from chemical analysis, XRD and ore microscopy, approximate mineralogical composition of the ore samples can be shown included smithsonite (16%), mimetite (0.6%), quartz (52%), sericite (16.5%), kaolinite (12%), iron oxides minerals (1.5%), rutile (0.4%) and other minerals (1%) (Hosseini et al., 2002).

The representative ore sample, after crushing, was ground in a laboratory stainless steel ball mill having a critical speed of 80%. After grinding, the material was deslimed at -20 μm size and used for bench scale flotation tests. Dodecylamine (99% purity) was obtained from Fluka Chemie, Switzerland. Potassium amyl xanthate with 90% purity was purchased from Shandong Qixia Flotation Reagent Co., 90%Ltd, China. Oleic acid (99.9% purity), sodium sulphide and copper sulphate (99% purity) were procured from Merck KGaA, Germany. Analytical grade HCl and NaOH were used for pH adjustment in all experiments. Tap water was used in all experiments. The chemical analysis of tap water showed that iron 0.5 mg/dm3, copper 0.75 mg/dm3, lead 0.03 mg/dm3, manganese 0.3 mg/dm3, zinc 2.25 mg/dm3, calcium 30 mg/dm3, magnesium 50 mg/dm3. (Total hardness of Tehran drinking water as CaCO3 is 300 mg/dm3). The bench flotation tests were performed with oxidised zinc ore samples. The experiments were conducted on 20% of solid in pulp ratio in Denver flotation cell operating at 1500 rpm with a constant flow of air (the maximum permitted by system). The ore sample, after being crushed, was ground in a laboratory stainless steel ball mill having a critical speed of 80%. The d80 of the products was -100 μm. The various amounts of reagents used in the oxidised zinc flotation, were added at different rates of concentrations .The conditioning time of collectors was 5 min .In amine flotation, the Na2S was added with conditioning time of 3 min at different concentrations. In KAX flotation, the copper sulphate was added 1500 g/t with 3 minute of conditioning time. The pine oil as a frother and sodium silicate (1000 g/t) were used with a conditioning time of 2 min. Flotation was conducted for 30 minutes and the pH of the pulp was adjusted using HCl and NaOH for all tests. The model with rectangular distribution of floatabilities has been applied in the evaluation of flotation results. This model is evaluated by fitting the flotation results from batch flotation tests. The mathematical form of this model may be written as follows (Huber-Panu et al., 1976 and Klimpel, 1980):

R = R (1-1/kt[1-exp(-kt)]) (1)

Where R is the percentage recovery of the mineral at any time interval t; R , the maximum percentage recovery of the mineral at infinite time t and k is the flotation rate constant (min-1). The R and k values for flotation recoveries have been estimated statistically and from the curves, the optimum flotation time has been determined graphically.

The experimental flow sheet illustrated in Figure 1, includes grinding, desliming and oxidised zinc flotation stage.

3. RESULTS AND DISCUSSION 3.1. Bench scale flotation with DDA

The bench flotation tests were performed with 100, 200 and 300 g/t of DDA concentrations and sulphidizing agent of sodium sulphide (1000, 2000, 3000 g/t) were used in the flotation experiments. The concentrate grade and recovery of zinc ore flotation for various amounts of DDA with sodium sulphide (2000 g/t) as sulphidizing agent are shown in Figure 2.

3

4

Figure 1. Experimental flow sheet

Tailing

Grinding

Hydrocycling

Zinc Flotation

Zinc Conc.

Feed

Slimes ( -20 μm )

0

10

20

30

40

50

60

70

80

90

100

0 1000 2000 3000 4000

Sodium sulphide concentration ( g/t)

Rec

over

y(%

)

0

5

10

15

20

25

30

Zinc grade (%

')

Recovery (DDA 100 g/t)Recovery (DDA 200 g/t)Recovery (DDA 300 g/t)Grade(DDA 100 g/t)Grade(DDA 200 g/t)Grade(DDA 300 g/t)

Figure 2. Concentrate grade and recovery of zinc ore flotation for various amounts of DDA in sodium sulphide solution (2000 g/t)

The maximum flotation recovery was found at 94% using DDA after sulphidisation at pH around 11 in single mineral flotation tests (Hosseini and Forssberg, May 2006). Hence, the results of zinc ore flotation using various amounts of DDA in presence of Na2S at various concentrations showed that the optimum concentration of Na2S was found at 2000 g/t. It is seen that at 1000 g/t of Na2S, recovery as well as grade fell sharply. This indicates that this level of sulphidisation was not sufficient for a feed containing high amount of smithsonite. When the amount of Na2S was increased, the recovery increased, but at the highest concentration of Na2S, the recovery fell off slightly, indicating depressing action of sodium sulphide at high concentration. It is mentioned that the controlling of the exact amount of sodium sulphide is more important. As, Castro et al., 1974, stated that the sulphidisation stage is critical in flotation of oxidised ores, because either too little or too much sodium sulphide gives poorer metallurgy than does an optimum addition. Barbaro et al., 1997, also reported that the sulphidisation step must be strictly controlled because an excessive dosage of sulphidizing agent results in drastic reduction in floatability. However, according the obtained results it is expected that the optimum amount of the cationic collector (DDA) was 300 g/t in the presence of Na2S 2000 g/t at pH 11.5 with 84.5 % recovery and 24.5% zinc grade. Amine collector at pH 11-11.5 is present essentially in undissociated form RNH2. The dissociation constant of dodecylamine is around 7.7. It has been assumed, in fact, that the amine becomes attached to the zinc surface through complication bonds (Marabini et al., 1994). Figure 3. depicts the collecting time vs. recovery (Na2S 2000 g/t; DDA 3000 g/t). This kind of exponential behaviour of the curve is well established and is in agreement with the first order with rectangular distribution of floatabilities. The R and k valueswere calculated by statistical software as 84.75 % and 0.728 (min-1), respectively. From Figure 3, it can be concluded that the optimum flotation time is 15 minutes.

20

30

40

50

60

70

80

90

100

0 5 10 15 20 25 30 35

Rec

over

y(%

)

Flotation time(min)

Figure 3. Zinc ore flotation at various collection times (Na2S 2000 g/t; DDA 300 g/t)

The grade and the recovery of the smithsonite in the concentrate and the tailing in terms of the quantity of dodecylamine and sodium sulphide are given in Table 1. The losses of zinc in tailing can not be ignored. The tailing should be reground to recover the zinc values. The loss of zinc was high. It can be explained with the size distribution of the minerals. According to flotation results can be seen that the tailings were rather high in zinc content, ranging from 2.2 to 10 % Zn. According to microscopic study on tailings, the presence of zinc minerals was found mainly as relatively coarse

5

6

(100 μm and coarser) and also liberated particles. It appears, therefore that zinc minerals of 100 μm and coarser in size were not in the suitable size range under the conditions as used in the tests. Rey, 1954, also attributed this fact to the slimes. He stated that the slimes make the froth layers brittle and therefore, the recovery of coarse zinc minerals drops. On the other hand, it can be concluded that the samples should be deslimed a little more (-30 μm or more). When the slimes are decreased, accordingly the flotation recovery of zinc minerals is improved.

Table 1 Grade and recovery of bench flotation tests for various amounts of DDA and sodium sulphide

Na2S = 1000 g/t Na2S = 2000 g/t Na2S = 3000 g/t

DDA (g/t) DDA (g/t) DDA (g/t)

Grade & Recovery

(%)100 200 300 100 200 300 100 200 300

Recovery (%) 34 45.5 44.2 42.4 68.5 84.5 34.6 46 64.1

Zn % in Conc.11.6 14.8 14 16.8 26.5 24.5 11.8 9.2 25.3

Zn % in Tails 8.8 7.4 7.7 7.3 4.0 2.2 8.7 10.0 4.6

3.2. Bench scale flotation with oleic acid

The bench flotation tests were performed with 100, 200 and 300 g/t of OA concentrations as a function of pH. The results are illustrated in Figure 4. The recovery of single mineral flotation (smithsonite) was found at 92.9% using oleic acid at pH around 10 (Hosseini and Forssberg, May 2006). The examination of the results shows that the best recovery of smithsonite in the concentrate was obtained at pH 10 and the collector concentration of 300 g/t. The trend shows the increasing of recovery with increasing of collector concentration. The grade and the recovery of the zinc mineral in the concentrate and tailing in terms of the quantity of oleic acid are given in Table 2. It is seen that at pH 9 recovery as well as grade enhance slightly and at pH 10, maximum recovery occurs. This indicates that this level of surface hydrophobicity was sufficient for a feed containing high amount of smithsonite. When increasing pH to 11, the recovery fell sharply, which indicating low adsorption of oleic acid on smithsonite surface.However, at the highest level of oleic acid addition at pH 10 the grade was low, which indicated, as expected, lower selectivity as the optimum concentration was exceeded. According to the obtained results the optimum amount of the oleic acid as an anionic collector was 300 g/t in presence at around pH 10 with 78.8 % recovery and 15.2 % zinc grade. Figure 5. shows the collecting time vs. recovery (oleic acid 300 g/t). The R and kvalues were calculated by statistical software as 86.73 % and 0.429 (min -1), respectively. From Figure 5, it can be founded that the optimum collection time is 20 minutes.

0

10

20

30

40

50

60

70

80

90

100

7 8 9 10 11 12pH

Rec

over

y(%

)

0

5

10

15

20

25

30

Zinc G

rade(%)

Recovery (OA 100 g/t)Recovery (OA 200 g/t)Recovery (OA 300 g/t)Grade(OA 100 g/t)Grade(OA 200 g/t)Grade(OA 300 g/t)

Figure 4. Grade and recovery of zinc flotation with various oleic acid concentrations

7

20

30

40

50

60

70

80

90

100

0 5 10 15 20 25 30 35

Rec

over

y(%

)

Flotation time(min)

Figure 5. Zinc ore flotation at various collection times with oleic acid concentration 300 g/t

Table 2Grade and recovery of bench flotation tests with various oleic acid concentrations

pH 8 pH 9 pH 10 pH 11 Oleic acid (g/t) Oleic acid (g/t) Oleic acid (g/t) Oleic acid (g/t)

Grade & Recovery

(%) 100 200 300 100 200 300 100 200 300 100 200 300Recovery

(%) 68 71.6 74.8 73.8 75.2 72.8 72.8 76 78.8 48 49.8 54.6

Zn % in Conc.

15.4 15.3 17.1 15.3 16.3 16.2 15.6 15.5 15.2 14.3 14.7 14.1

Zn % in Tails

5.3 4.9 4.2 4.7 4.3 4.6 4.7 4.4 4.1 7.4 7.1 6.9

3.4. Bench scale flotation using KAX

Smithsonite flotation recovery using Hallimond tube was determined 81.3% using KAX at pH around 10.5 (Hosseini and Forssberg, June 2006). The flotation recovery and grade curves of zinc ore as a function of KAX consumption are illustrated in Figure 6.

0

10

20

30

40

50

60

70

80

90

100

0 100 200 300 400 500 600 700

Rec

over

y(%

)

0

5

10

15

20

25

30

Zinc G

rade (%)

Recovery

Grade

KAX concentration(g/t)

Figure 6. Grade and recovery of zinc flotation for various KAX concentrations

8

The zinc ore flotation results show that the highest recovery of zinc mineral in the concentrate was obtained at pH 10.5-11. The optimum collector consumption was found 600 g/t. The trend shows the increasing of recovery with increasing of collector concentration. It is seen that the recovery as well as grade enhance slightly and at KAX concentration 600 g/t, the optimum recovery occurs which is 68.8 % with zinc grade of 16.2 %. However, at the highest level of KAX addition the zinc grade was low, which indicated, as expected, lower selectivity as the optimum concentration was exceeded.Figure 7. depicts the collecting time vs. recovery with KAX concentration 600 g/t. The R and k values were calculated by statistical software as 78.81 and 0.323, respectively. From Figure 7, it can be concluded that the optimum collection time is 20 minutes.

10

20

30

40

50

60

70

80

90

100

0 5 10 15 20 25 30 35

Rec

over

y(%

)

Flotation time(min)

Figure 7. Zinc ore flotation at various collection times with KAX 600 g/t

The grade and the recovery of the oxidised zinc mineral in the concentrate and tailing in terms of the quantity of KAX are given in Table 4.

Table 4 Grade and recovery of bench flotation tests for various amount of KAX in the presence of sodium sulphide (2000 g/t) and copper sulphate (1500 g/t)

9

3.5. Bench scale flotation with mixed collector

The flotation experiments were performed with mixed (Cationic/Anionic) collector concentrations. In these experiments, the effect of mixture ratios was investigated. The

KAX (g/t) Grade & Recovery

(%)100 200 300 400 500 600

Recovery (%) 45.7 48.6 52.2 51.7 60.5 68.8Zn % in Conc. 21.6 19.6 17.5 18.1 17.1 16.2Zn % in Tails 6.5 6.5 6.5 6.4 5.7 5.1

mixture ratio of KAX: DDA was changed as 3:1, 2:1, 1:1, 1:2 and 1:3. Smithsonite flotation recovery by means of Hallimond tube was determined 96.6% using mixed collector at pH 9.5 (Hosseini and Forssberg, 2007). The oxidised zinc ore flotation results are shown in Figure 8.

0

10

20

30

40

50

60

70

80

90

100

Rec

over

y(%

)

0

5

10

15

20

25

30

35

40

45

50

Zinc G

rade(%)

Recovery

Grade

3:1 1:1 1:3

M ixe d Colle ctors (KAX : DDA) mole ratio

1:22:1

Figure 8. Grade and recovery of zinc flotation for various mixed collector concentration ratio

As it is seen from Figure 8, the highest flotation recoveries were found 3:1 and 1:3 of concentration ratios (KAX: DDA) which are 82.65 and 74 %, respectively. The grade variation is from 17.2 to 25.4 % which for concentration ratio of 3:1 is 24.1%. Figure 9. depicts the collecting time vs. recovery (KAX: DDA; 1:3). The R and k values were calculated by statistical software as 87.59 and 0.775, respectively.

10

20

30

40

50

60

70

80

90

100

0 5 10 15 20 25 30 35

Rec

over

y(%

)

Flotation time(min)

Figure 9. Zinc ore flotation at various collection times for mixed collector (KAX and dodecylamine, concentration ratio 3:1) at pH 9.5

From Figure 9, it can be concluded that the optimum collection time is 15 minutes. The grade and the flotation recovery of the oxidised zinc mineral in the concentrate and tailing in terms of the quantity of KAX: DDA concentration ratios are given in Table 5.

Table 5 Grade and recovery of bench flotation tests for various amounts of mixed collector (KAX and dodecylamine) in various mole ratios in the presence of sodium sulphide (2000 g/t)

11

4. CONCLUSIONS

The batch flotation results revealed that the maximum flotation recovery and zinc grade were obtained using DDA which are 84.5% and 24.5% respectively. Hence, the optimum flotation recovery and grade using mixed collector shows the highest recovery for concentration ratio of 3:1 and 1:3 which are 82.65 and 74 % respectively. The grade variation is from 17.2 to 25.4 % which for concentration ratio of 3:1 is 24.1. The flotation results using mixed collector (dodecylamine and KAX) show that with increasing the KAX concentration, the recovery is increased. When it is used KAX alone in batch flotation, the poorer flotation results are obtained. The flotation results using oleic acid also have no significant variation for recovery in comparing with amine flotation but no selectivity is observed.

KAX : DDA Concentration ratios Grade & Recovery

(%)3:1 2:1 1:1 1:2 1:3

Recovery (%) 82.65 76.7 55.2 69.4 74.0Zn % in Conc. 24.1 26.1 25.4 20.1 17.2Zn % in Tails 2.4 3.1 5.4 4.4 4.3

12

ACKNOWLEDGEMENTS

The authors acknowledge Calcimine Co. for providing the ore samples, which were used in this investigation. The assistance of colleagues at Tehran Azad University, Department of Mining Engineering, Tehran, Iran is also much appreciated.

REFERENCES

Abramov, A., Use of cationic agents for the flotation of oxide lead-zinc minerals. chemical abstract, 1961, 55, 26910f

Barbaro, M., Herrera Urbina, R., Cozza, C., Fuerstenau D., Marabini, A., Flotation of oxidized minerals of copper using a new synthetic chelating reagent as collector.International Journal of Mineral Processing,1997, Volume 50, Issue 4, pp. 275-287

Barbery, G. Cecile, J.L., Plichon, V., The use of chelates as flotation collectors. In:Proceeding of XII International Mineral Processing Congress, Sao Polo, 1977, pp.19-34

Billi M., and Quai V., Development and results obtained in the treatment of zinc oxide ores at AMMI Mines. IMPC, London, 1963, paper 43

Castro, S., Goldfarb, J., and Laskowski, J., Sulphidizing reactions in the flotation of oxidized copper minerals, I. chemical factors in the flotation of oxidized copper oxide, International journal of mineral processing, pp.141-149, 1974

Gaudin, A. M, Flotation. McGraw Hill Inc., New York, 1957, pp.182-189 Hosseini, S. Hamid, Forssberg, Eric , Mineral Processing Possibility of Oxidizing Lead &

Zinc Minerals from Angooran Deposit in Zanjan Province, Iran. In: Proceedings of the IX Balkan Mineral Processing Congress, Istanbul, Turkey, 2001, pp. 221-226

Hosseini, S. Hamid, Södervall, Ulf, Forssberg, Eric, Comparison between the Bulk & Surface Composition of the Samples From Angooran Lead & Zinc Mine, Zanjan Province, Iran , In: Proceedings of the 6th conference environment and mineral processing , Ostrava, Czech republic, 2002, pp. 289-295

Hosseini, S. Hamid, Forssberg, Eric, Adsorption Studies of Smithsonite Flotation Using Dodecylamine and Oleic acid. Minerals and Metallurgical Processing, SME, 2006 May, Vol. 23, No.2, pp.87-96

Hosseini, S. Hamid, Forssberg, Eric, Smithsonite Flotation Using Potassium AmylXanthate and Hexylmercaptan. Mineral Processing and Extractive Metallurgy (Trans. Inst. Min Metals. C), 2006 June, Vol. 115, No. 3, pp.107-112

Hosseini, S. Hamid, Forssberg, Eric, Physicochemical Studies of Smithsonite Flotation using Mixed Cationic/Anionic Collector'', Minerals Engineering, 2007, Volume 20, Issue 6, pp.621-624

Huber-Panu, I., E. Ene-Danalache, and D.G. Cojocariu, Mathematical Models of Batch and Continuous Flotation, Flotation- A.M. Gaudin Memorial Volume, M.C. Fuerstenau, ed., AIME, New York, NY, 1976, Vol. 2, Chapter 5, pp. 675-724

Klimpel, R.R., Selection of Chemical Reagents for Flotation, Mineral Processing PlantDesign, 2nd Edition, A.L. Mular and R.B. Bhappu, eds., 1980, Chapter 45, AIME, New York, NY, pp. 907-934

McGarry, P. E., Pacic, Z., Flotation of non-sulfide zinc materials. United States patent: 4253614, 1981

McKenna, W. J., Lessels, V., Petersson, E. C., Froth flotation of Oxidized zinc ores.United States patent: 2482859, 1949

Marabini, A. M., Alesse, V., Belardi, G., Spaziani, E., Effect of depressing agents on the

13

flotation of oxidized zinc minerals. Minerals and Metallurgical processing, SME, 1994, pp. 97-104

Ozbayoglu, G., Atalay, U., Senturk, B., Flotation of lead and zinc carbonates ore. In:Proceeding of International Conference on recent advances in materials and mineral resources’, Penang, Malaysia, May 1994, School of Material and Mineral Resources Engineering, pp. 504–509

Önal, G., Bulut, G., Gül, A., Kangal, O., Perek, K.T., Arslan, F., Flotation of Aladagˇoxide lead–zinc ores. Minerals Engineering, 2005, Volume 18, pp.279-282

Rey, M., The flotation of oxidized ores of lead, copper and zinc. In: RecentDevelopments in Mineral Dressing Symposium, IMM, London, 1953, pp. 541-548

Rey, M., Sitia, G., Raffinot, P., Formaneck, V.,, Flotation of oxidized zinc ores. Trans.AIME, Mining Engineering, 1954, 199, pp.416-420

Rausch, D.O., Mariacher, B.C., Concentration of oxide ores at Tynagh Mining andconcentrating of lead & zinc. AIME World Symposium on Mining & Metallurgy of Lead & Zinc, 1970, Vol. 1, Extractive metallurgy of lead and zinc, pp. 721-731

Weiss, N. L., SME Mineral Processing Hand book. AIME, 1985, pp. 15-4 15-7 Zhao, Youcai, Standford, Robert, Production of Zn powder by alkaline treatment of

smithsonite Zn-Pb ores. Hydrometallurgy 56, 2000, pp.237-238

Paper VI

Comparison between the bulk & surface composition of the samples from Angooran lead & zinc Mine, Zanjan Province, IranSeyed Hamid Hosseini, Ulf Södervall and Eric Forssberg, Proceedings of the 6th

Conference on Environment and Mineral Processing, Ostrava, Czech Republic, June 2002, pp.289-294.

Hosseini S.H., Södervall U., Forssberg E.: COMPARISON BETWEEN THE BULK & SURFACE COMPOSITION OF THE SAMPLES FROM ANGOORAN LEAD & ZINC MINE, ZANJAN PROVINCE, IRAN

COMPARISON BETWEEN THE BULK & SURFACE COMPOSITION OF THE SAMPLES FROM ANGOORAN LEAD & ZINC MINE, ZANJAN PROVINCE, IRAN

S. Hamid Hosseini*, Ulf Södervall ** and Eric Forssberg*

*Mineral Processing Division, Luleå University of Technology, S-951 87, Luleå, Sweden **Department of Microelectronics and Nanoscience, MC2-Microtechnology center at Chalmers,

Chalmers University of Technology, S-412 96 Gothenburg, SWEDEN____________________________________________________________________________

AbstractThe Angooran Lead & Zinc Mine located in Zanjan province, Iran is one of the largest ones of its kind in the Middle East. At present, a part of the ore body located within host rock mineralised schist is not included in the milling design. According to previous work, wet chemical assay, XRF, XRD and ore microscopy studies of the ore samples show that the ore contains Smithsonite (16%), Mimetite (0.6%), Quartz (52%), Sericite (16.5%), Kaolinite (12%), Iron Oxides Minerals (1.5%), Rutile (0.4%) and other minerals (1%). (Hosseini S.H. et al ,2001) In the present study, XPS (ESCA) and EDX have been utilized, in order to study the characterization of the surface and bulk composition respectively for three fraction of - 250, +200 μm; -200, +150 μm and -150μm. The results of XPS show that these have no difference especially in Zn percentage and also for other elements for three size fractions. It can be concluded that the surface composition are approximately the same for both three-size fractions. A comparison between the data obtained from XPS (Surface Composition) and from EDX Analysis (Bulk Composition) shows that there are some differences for the Si, Fe, and Al content in the ore sample and these differences between XPS and other techniques should be based on the amount of Oxygen for each technique.

Key words: Oxidized ore; EDX; XPS; Surface & Bulk Composition; Lead & Zinc; Zanjan; Iran

Introduction The Angooran Mine is located 100 Km. South west of Zanjan by road with an altitude of roughly 2950 m above sea level on a latitude of Approx. 47º 20´and a longitude of 36º 40´ and is one of the largest Lead & zinc deposit in Iran and also in the Middle East. In previous investigation, a representative sample, which collected from Angooran Lead & zinc mine, have been examined to determine the bulk chemical composition of particles including some minerals such as Smithsonite (ZnCO3), Quartz (SiO2), Goethite (FeOOH), Rutile (TiO2) and so on. Tables No 1 and 2 show that the result of chemical composition of Oxidized Lead & Zinc Ore. (Hosseini S.H. and Forssberg Eric, 2001)

Table No.1. Result of XRF Tests

Elements or Oxides Weight%

Elements Or Oxides Weight

%

Elements Or Oxides Weight

%SiO2 61.9 F 0.21 MnO 0.022 Al2O3 15.2 As2O3 0.18 V2O5 0.022 ZnO 8.9 CdO 0.14 ZrO2 0.016

K2O 2.15 S 0.102 Cl 0.013 Fe2O3 1.95 NiO 0.058 Sb2O3 0.007 PbO 0.85 P2O5 0.046 La2O3 0.006 TiO2 0.65 Cr2O3 0.044 Rb2O 0.0052 MgO 0.60 Co3O4 0.026 SrO 0.0041 CaO 0.30 WO3 0.025 L.O.I 6.57

289

290

6thConference on Environment and Mineral Processing _________________________________________________________________________________________

Table No. 2. Result of Wet Chemical tests

Oxides Weight % SiO2 62.5Al2O3 16.5K2O 2.25TiO2 0.45PbO 1.36

Fe2O3 1.85ZnO 9.60

XRD studies have showed that the minerals of Quartz, Sericite, Smithsonite, Kaolinite and Mimetite have been identified in these ore samples. By combining the information obtained from Chemical analysis (Table No.1, 2), XRD and optical microscopy, approximate mineralogical composition of the ore sample may be shown as follows:

Zn Minerals - 16 % Pb Minerals - 0.6 % Quartz - 52% Sericite - 16.5% Kaolinite - 12 % Rutile - 0.4% Iron oxide - 1.50%

Other minerals - 1%

Since in earlier experiments, XRF, XRD, Wet Chemical Assay and also Optical Microscopy were used for characterization of materials content. In the present work, XPS and EDX have been utilized in order to study the characterization of the surface and bulk composition respectively. The objective of this work is to determine the bulk and surface composition of Oxidized Lead & Zinc minerals and comparing the obtained data to each other .It is mentioned that this work will help us for studying of surface chemistry on flotation process in size fraction of 150-200 μm. (Vaughan D.J. and Pattrick R.A.D., 1995)

Experimental Materials A representative sample, which was collected in the Angooran Lead & Zinc mine, has been used to determine the chemical surface and bulk composition of the particles including some minerals in three size fraction of - 250, +200 μm; -200, +150 μm and -150μm. For this purpose it is used the sieve analysis of the samples with using Standard Sieves (150, 200,250 μm) so that there are three kinds of sample as follow: Sample No: H1250 (-250, +200 μm) Sample No: H2200 (-200, +150 μm) Sample No: H3150 (-150μm) There are a number of methods that can be used to mount the powders for analysis with XPS. Perhaps the most widely used method is too carefully and lightly dust the powder on polymer film based adhesive tape with camel hairbrush. The powder must be dusted on lightly, with no wiping stokes across the powder surface. It has been used successfully in the 10-9 Torr range. (Wagner C.D. et al., 1979)

291

Hosseini S.H., Södervall U., Forssberg E.: COMPARISON BETWEEN THE BULK & SURFACE COMPOSITION OF THE SAMPLES FROM ANGOORAN LEAD & ZINC MINE, ZANJAN PROVINCE, IRAN

The non-conducting samples in EDX analyzer are coated with a thin layer of carbon. Metallographic embedding, polishing, and sectioning are available for samples requiring special preparation. Samples are usually mounted and coated and introduced into the vacuum chamber. (Goodhew P.J, 1983)

Methods Surface Composition Study

The objective of this study is to determine the surface composition of Oxidized Lead & Zinc minerals. As we know, when a surface is bombarded with atomic and subatomic particles, ions, electrons and atoms are emitted. Bombardment with x-ray photons leads to the emission of electrons, these are called photoelectrons. These particles typically are emitted from an l0 nm surface layer. The chemical state of a surface can also be monitored by analyzing the photoelectrons emitted from a surface on bombardment with x-rays. This technique is called x-ray photoelectron spectroscopy (XPS). X-Ray Photoelectron Spectroscopy (XPS), also known as Electron Spectroscopy for Chemical Analysis (ESCA) determines the chemical composition of a surface using the photoelectric effect. The sample is irradiated with X-ray photons and electrons are emitted from the sample if the photon is of sufficient energy. Photoelectrons are counted at each kinetic energy value and a spectrum of intensity vs. binding energy is generated from the above equation. The binding energy of an electron is characteristic of the element, orbital and chemical environment therefore XPS can determine the bonding state and/or oxidation states of materials and surface concentrations. To produce the low energy X-ray, a 10 keV electron gun is aimed at an aluminum target. MgK X-rays (1253.6 eV) or AlK x-rays (1486.6 eV) are ordinarily used. Besides the normal peaks for elements in plotting spectrum, we can see auger peak shifts. The reason for that is the Auger electron is generated by the internal atom de-excitation (the atom recovers from a higher energy state caused by the loss of the photoelectron) The Auger electron kinetic energy is always independent from the source nature. At first glance it is not easy to distinguish on a spectrum Auger electron peaks from photoelectron peaks. Changing the source, i.e. changing excitation energy, will cause some peaks to be shifted some peaks not(Wagner C.D. et al, 1979).

Bulk Composition Study The objective of this Study is to determine the bulk composition of Oxidized Lead & Zinc minerals .An energy-Dispersive x-ray analyzer (EDX) is a common accessory, which gives the SEM a very valuable capability for bulk chemical analysis. (Hren J.J., Goldstein J.I and Joy D.C., 1979) The energy holding electrons in atoms (the binding energy) ranges from a few eV up to many kilovolts. Many of these atomic electrons are dislodged as the incident electrons pass through the specimen, thus ionizing atoms of the specimen. Ejection of an atomic electron by an electron in the beam ionizes the atom, which is then quickly neutralized by other electrons. In the neutralization process an x-ray with an energy characteristic of the parent atom is emitted. By collecting and analyzing the energy of these x-rays, the constituent elements of the specimen can be determined. (Norden H. and Thölen A., 1997)

Results and Discussion Tables No.3-5 show the results of XPS for three size fraction of samples and also Table No. 6 shows the result of EDX for finest sample. Figs 1, 2, 3 and 4 show the peak result of XPS and EDX respectively.

292

6thConference on Environment and Mineral Processing _________________________________________________________________________________________

Table No. 3. Results of XPS on Sample No: H3150

Element

Area(cts-eV/s)

SensitivityFactor

AtomicConcentration

(%)

Weight(%)

Fe2p3O1sAl2pSi2pClsK2pCa2pZn2p3 F1sMg1s

6595352649 12476 35533 24347 13270 5741

85586 19141786

1.791 0.711 0.193 0.283 0.296 1.300 1.634 3.354 1.000 1.433

0.4560.89 7.9415.41 10.10 1.250.433.130.230.15

1481021621

100.50.5

Table No. 4. Results of XPS on Sample No: H2200

ElementArea(cts-eV/s)

SensitivityFactor

AtomicConcentration

(%)

Weight(%)

Fe2p3O1sAl2pSi2pClsK2pCa2pZn2p3 F1sMg1s

4783263502 8744

27782 13485 95574277

68699 15731311

1.791 0.711 0.193 0.283 0.296 1.300 1.634 3.354 1.000 1.433

0.4562.26 7.6116.49 7.651.240.443.440.260.15

3471022421

110.50.5

Table No. 5. Results of XPS on Sample No: H1250

ElementArea(cts-eV/s)

SensitivityFactor

AtomicConcentration

(%)

Weight(%)

Fe2p3O1sAl2pSi2pClsK2pCa2pZn2p3 F1sMg1s

5288272991 9468

29168 13531 96424212

64855 1284964

1.791 0.711 0.193 0.283 0.296 1.300 1.634 3.354 1.000 1.433

0.4862.33 7.9616.73 7.421.200.423.140.210.11

1481023421

100.50.5

Hosseini S.H., Södervall U., Forssberg E.: COMPARISON BETWEEN THE BULK & SURFACE COMPOSITION OF THE SAMPLES FROM ANGOORAN LEAD & ZINC MINE, ZANJAN PROVINCE, IRAN

Fig No. 1 Peak results of ESCA on Sample H3150 Fig No.2 Peak results of ESCA on Sample H2200

Table No.6 The Results for Bulk Chemical Analysis by EDX

293

Spectrum File>LF011204 Live Time (Spec.)= 200 TILT =0.00 ELEV=30.00 AZIM =0.00 ENERGY RES AREA7.1 76.08 120609

Total AREA= 33838 Peak at 0.98 KeV Peak at 3.30 KeV Peak at 3.62 KeV FIT INDEX = 4.99ZAF-PB CALCULATIONS3 ITERATIONS G-FACTOR =7.993

SPECTRUM: All Elements analyzed

ELMT AREA AREA/BGND %CONC FST NORM. %

Al K 3788+- 105 3.032+- .412 23.72 0.671 29.41 Ti K 223+- 55 .129+- .032 0.62 1.293 0.76 Si K 10586+-156 7.274+- .692 41.47 .863 51.42 Fe K 638+-71 .531+- .061 2.36 1.542 2.92 Zn K 1635+-102 2.256+- .169 12.50 1.719 15.49

Fig No.3 Peak results of ESCA on Sample H1250 Fig No.4 Peak results of EDX on Sample H3150

294

6thConference on Environment and Mineral Processing _________________________________________________________________________________________

According to obtained results it is found out that these have no difference in elements percent for three size fractions and it can be said that approximately they are the same for both three size fractions .The desired size for flotation process which has the maximum degree of liberation is the size fraction (-200, +150 μm). In comparison of the data obtained from XPS (Surface Composition), EDX Analysis (Bulk Composition), it has been some difference between obtained data (table No.7). For example, the reason of differences of Zn % for XPS in comparing with EDX, it may be for Oxygen content, which has been determined by XPS method. It means that the Zn% in EDX is based on Oxygen content but in XPS the Oxygen % is determined individually. The results of XPS show that these have no difference especially in Zn percentage and also for other elements for three size fractions. It can be concluded that the surface composition are approximately the same for both three-size fractions.

Table No.7 The Results for Bulk & Surface Chemical Analysis

Sample No H3150 (Fine)

Sample No H2200 (Middle)

Sample No H1250 (Coarse)

Weight % Weight % Weight % AnalysisTechnique

Si Al Zn Fe Ti Si Al Zn Fe Ti Si Al Zn Fe Ti

EDX 51.42 29.41 15.49 2.92 0.76 -- -- -- -- -- -- -- -- -- --

XPS(ESCA)

21 10 10 1 --- 22 9.73 11 3 --- 23 10 10 1 ---

Conclusion A comparison between the data obtained from XPS (Surface Composition) and from EDX Analysis (Bulk Composition) shows that there are some differences for the Si, Fe, and Al content in the ore sample and these differences between XPS and other techniques should be based on the amount of Oxygen for each technique.

Acknowledgements The Authors wish to thank Calcimine Co. for providing the samples, which were used in this work. The assistance of Colleagues at the Chalmers University of Technology, Department of Materials Science, Gothenburg, Sweden is also much appreciated.

References 1. Goodhew P.J: Specimen Preparation in Material Science, North Holland/American Elsevier, 1983 2. Hosseini S. Hamid and Forssberg Eric. Mineral Processing Possibility of Oxidizing Lead & Zinc Minerals from Angooran Deposit in Zanjan Province, Iran, IX Balkan Mineral Processing Congress, Istanbul, Turkey, 2001 3. Hren J.J., Goldstein J.I and Joy D.C.: Introduction to analytical Microscopy, Plenum press, 1979 4. Norden H. and Thölen A.: Electron Microscopy and Microanalysis, Department of Physics, Chalmers University of Technology, Gothenburg, Sweden, pp.223-241, 1997 5. Vaughan D.J. and Pattrick R.A.D: Mineral Surfaces, Chapman & Hall, pp. 17-42, 1995 6. Wagner C.D. et al.: Handbook of X-ray photoelectron spectroscopy, Minnesota, Perkin –Elmer Corporation, 1979

Paper VII

Mineral Processing possibility of Oxide Lead & Zinc minerals from Angooran Deposit in Zanjan province, IranSeyed Hamid Hosseini and Eric Forssberg, Proceedings of the IX Balkan Mineral Processing Congress, Istanbul, Turkey, September 2001, pp.221-226.

MINERAL PROCESSING POSSIBILITY OF OXIDIZED LEAD & ZINC MINERALS FROM ANGOORAN DEPOSIT IN ZANJAN PROVINCE, IRAN

S.H.HosseiniMineral processing Division, Luleå University of Technology, Luleå, Sweden

K.S.E.ForssbergMineral processing Division, Luleå University of Technology, Luleå, Sweden

ABSTRACT: The Angooran Lead & Zinc Mine located in Zanjan province-Iran is the greatest one of its kind in Middle East. At present, a part of ore body located within host rock mineralized schist is not planned for milling project design. In the present study, the mineral processing possibility of mineralized schist deposit within Angooran Lead & Zinc Ore deposit was investigated. The ore microscopy studies were carried out to determine the mineral content of the ore deposit. The natural breakage and grindability were studied to determine the behavior of the desired ore during several stage breakages. The liberation degree of crushed ore in different size fractions was estimated by the Counting method. The Heavy Liquid Tests for three size fractions (-212+150; -150+125; -125+75 μm) were carried out in different density fractions and were assayed for lead, zinc and SiO2 content. The wet chemical assay, XRF, XRD and ore microscopy studies showed that the ore subjected to different stages included Smithsonite (16%), Mimetite (0.6%), Quartz (52%), Sericite (16.5%), Kaolinite (12%), Iron Oxides Minerals (1.5%), Rutile (0.4%) and other minerals (1%). The Locking Analysis for different size fractions showed that the size fraction of -125+75 μm to decrease the overgrinding was selected as the best amount of liberation degree and the result of HLS confirmed this, i.e. the most Zn content was in size fraction of -125+75 μm, while the applicability of gravity concentration was in size fraction of -150+125 μm. The mineral processing circuits for size fraction -150, +125 μm can be assessed for gravity concentration as a preconcentration process and for the size fraction -125, +75 μm in the flotation process in the next stages.

INTRODUCTION

The Angooran Mine is located 100 Km. South west of Zanjan by road with an altitude of roughly 2950 m above sea level on a latitude of Approx. 47º 20´and a longitude of 36º 40´ and is one of the largest Lead & zinc deposit in Iran. The Mine is affiliated to Mines & Industry ministry, Islamic Republic of Iran .The nearest major town is Zanjan so that a road, Approx. 100 Km. long maintained by the mine and Dandy mill plant links with Zanjan -Bijar road.

The daily feed to the Dandy plant which is located 20 Km from the mine, is Approx. 1000 ton. Table 1 shows the result of ore reserves in Angooran mine.

The major Zinc mineral of oxides zone is Smithsonite with Hemimorphite and Hydrozincite as minor minerals. Whereas, The major Lead mineral is Cerrusite with Mimetite as a minor one. Generally, the associated minerals of oxides zone

are Quartz, Mica, Hematite, Goethite, Kaolonite, and Montmorolinite. In sulfides zone the major Lead & Zinc mineral are Galena and Sphalerite respectively and the associated minerals generally are Quartz and Calcite.

Table 1.Ore reserves estimation of Angooran mine

Ore Type Zn(%)

Pb(%)

SiO2(%)

Estimated Reserves (1000 t)

Very Low grade

7.94 0.37 28.5 2600

MineralizedSchist

14.44 1.94 47.9 1900

High grade 31.68 5.54 13.2 1600

In general, the shape of this ore deposit is in the form of lentiform massive similar to egg which is located between schist and crystalline limestone with width of 100-300 m and 20 º dip towards east.

221

EXPERIMENTAL

Materials

According to the geological map of Angooran mine, the mineralized schist outcrop appears only at two benches of 2920 and 2930 levels and therefore, the samples collection is consequently confined only at these two levels. while the mineralized schist were partly covered by low grade oxides zone materials so for proper collection of samples, two trenches with average depth of 2m and a length of nearly 10 m were digged to achieve entirely mineralized schist deposit on the floor and wall of benches.

The sampling method adopted for this investigation was chip sampling on the floor and channel sampling for the wall on mineralized schist only. Since the samples are collected for the mineral processing test purpose, the selected chip sampling method were carried out to collect samples by scoop from benches floor proceeds in such a way what series of chips which are taken from all of parts outcrops within the defined areas of a square grid pattern of 1x1 m systematically. The average weight of increments in this method is about 0.5-1.5 Kg and consists of both coarse and fine materials so that the total weight of samples collected by chip sampling method is about 75 Kg. The channel sampling on the wall of trench (Bench No 2920) were carried out by digging a channel of about 3 cm depth and 10 cm width along the across of the trench and nearly 55 kg samples were collected .The total weight of samples collected by chip sampling and channel sampling methods of both trenches is 129.5 Kg. Carbon Tetrachloride S.G 1.59, Tribromoethane S.G 2.80, Tetrabromoethane S.G 2.96 and Methylene Iodide S.G 3.31 in Heavy Liquid tests are used. Liquid having densities 2.7, 2.6 are obtained by mixing different proportions of Tribromoethane and Carbon Tetrachloride.

Methods

Natural Breakage Studies

The objective of this test is the determination of the natural breakage characteristics of schist ore in coarse and fine size ranges based on the behavior of the desired ore in several stages of breakage so that the natural breakage characteristics of schist ore has been determined in range of lesser than 30 mm by using Jaw crusher STURTEVANT, Gape =165 x 95 mm with Open Setting of 40 mm and

Closed setting of 35 mm and also in range of lesser than 6 mm with Open Setting of 10 mm and Closed setting of 6 mm.

Mineralogical Studies

For determination of mineral composition of schist ore has been used from different tests such XRF, XRD, and Ore Microscopy.

Liberation Characteristics Studies

The Liberation Characteristics of this ore were complex and extremely difficult to quantify because it was not easy to distinguish between the minerals microscopically. The characteristic sample has been collected from some fraction size according the sieve analysis of the above sample. These were then points converted under the microscope to find the degree of liberation where % liberation was defined as the ratio of

100.Particles All

Particles Liberated

Fields of view were chosen at random and all the particles on the cross wires were counted. This method was time consuming and subject to errors such as diffusion, which occurred during the coloration of the minerals.

Table 2. Used polished sections for locking analysis

Polished Section

No

SizeFraction

(mm)

SizeFraction

(# ASTM) 1 -2.36 ,+1.18 -8,+162 -1.18 ,+0.85 -16,+20 3 - 0.85,+0.6 -20,+30 4 -0.6,+0.425 -30,+40 5 -0.425,+0.212 -40,+70 6 -0.212,+0.150 -70,+100 7 -0.150,+0.125 -100,+120 8 -0.125,+0.075 -120,+200 9 -0.075,+0.063 -200,+230

10 -0.063 -230

Grindability Studies

The grindability of the ores or minerals is an important factor that could not be deduced from their mineralogical characteristics. Many theories have been proposed to determine the amount of required work in a comminution operation as a function of the nature of material, size and shape of the minerals before and after of comminution. Rittinger (1867) proposes the most important laws, Kick (1885) and Bond (1952). According to Bond’s Theory, the work input is proportional to

222

the new crack tip length produced during the particle breakage. The work input is given by the following equation:

Where W is The energy input to the mill (kWh / st ); F is feed size based on 80% passing ,(μm); P is the product size based on 80% passed, (μm); and W¡ is the Work Index,(kWh/St).Since there are some problems with the Standard Bond Grindability Test such as requiring a special mill ,requiring a relatively large amount of sample and also being relatively time consuming and tedious. In this study, a new size ball mill (NSBM) for determining the Bond Work Index (H.Nematollahi, 1994) is used. The used ball mill is scaled down with a coefficient of two-thirds from the standard Bond ball mill (200×200mm). The following equation was obtained to compute the Bond work index:

Where P¡ is the sieve opening at which the test is carried out (μm); G¡ is the NBSM grindability,net grams of ball mill product passing sieve size P¡ produced per mill revolution (g/r); F is the Feed size based on 80% passed (μm); and P is the product size based on 80% passed (μm).

Heavy liquid Tests

Generally heavy liquid tests indicate what theoretically can be expected from gravity separations. Chemical analyses of products for Pb , Zn do not ,however give much information for the liberation of Pb , Zn minerals .Chemical assay of sink products on the other hand ,does not tell whether the heavy minerals are liberated from each other or not?! For this reason degree of liberation was further studied using other methods.

A representative sample from size fractions -212+150;-150+125; -125+75 μm were subjected to heavy liquid tests. The feed and product of used density fractions 3.31, 2.96, 2.8, 2.7, 2.6 were assayed for lead, zinc and SiO2

.

The Curve No.1 in figures related to Henry Reinhardt Diagram by plotting mean cumulative

float weight % Vs. Mean Zinc grade % with using the rectangles shows that the mean Zn content of used sample as if the under area of this curve moves down, naturally the Zn content in this sample is decreased. The curve No. 2 and 3 are plotted by the cumulative sink weight % and Cumulative floats weight % Vs. Zinc grade %. Therefore, it is determined the zinc grade % for such amount of sink weight % what Zinc grade % the specified amount of sink or floats weight % based on Curve No.2 and No.3 respectively.

)11(10FP

WW i

The curve No.4 for different size fractions by plotting cumulative float weight % Vs. S.G fractions shows that the ability of using the gravity separation for the examined sample so that the amount of the curve slope indicates this ability. On the other hand, the more slope of curve shows the ability of gravity separation (Gruender, Werner. 1963).

RESULT AND DISCUSSION

Natural breakage

FPGP

Wii

i 10101176.11

75.023.0

The results of natural breakage studies for the size fraction of greater than 30 mm and also greater than 6 mm has been showed in Figures 1 and 2. Figures 1 and 2 show that the relation between cumulative production of fines the lesser than 6 and 30 mm versus the cumulative feed greater than 6 and 30 mm. Based on Linear regression of obtained data for 30mm and 6 mm, the ratio of F and CL will be 0.3023, 231% for 30 mm and 0.3162, 216 % for 6 mm (F is Cumulative fines/Cumulative feed and CL is circulating load).

0

10

20

30

40

50

60

70

80

90

100

0 50 100 150 200 250 300 350

Cummulative Feed > 30 mm, Kg

Cum

mul

ativ

e Fi

nes

< 30

mm

, Kg

Figure.1-Stage crushing of the particle size fraction > 30 mm from the prescreening stage. Cumulative production of fines < 30 mm versus the cumulative feed > 30 mm

223

Figure.2-Stage crushing of the particle size fraction > 6 mm from the prescreening stage. Cumulative production of fines < 6 mm versus the cumulative feed > 6 mm

Mineralogical Studies

After preparation of thin and polished sections from collected samples and studying on it under Microscope ,the minerals of Quartz , Mica , Sericite, Smithsonite, Clay minerals (Kaolinite) as a large amounts and also the minerals of Mimetite, Pyrite , Iron Oxides (Goethite), Rutile , Cerussite, Galena , Sphalerite as a small amounts have been identified.Figure.3 shows that the thin section of desired ore sample which the Smithsonite has been connected with Quartz ,Sericite ,mica and opaque minerals. (Ln -200X) Figure.4 shows the polished section of ore sample. (PPL 32 x 8 Oil + Nicole) XRD studies have showed that the following minerals of Quartz, Sericite, Smithsonite, Kaolinite and Mimetite have been identified in these ore samples.

Figure.3 Thin section including 1-Smithsonite 2-Sericite 3-Mica 4- Quartz 5-Opaque minerals. (LP -200 X)

Table 3.Result of XRF Tests

0

5

10

15

20

25

30

35

40

45

50

0 20 40 60 80 100 120 140

Cummulative Feed > 6 mm, Kg

Cum

mul

ativ

e Fi

nes

< 6

mm

, Kg

Elements or Oxides

Weight %

Elements or Oxides

Weight %

Elements Or Oxides

Weight %

SiO2 61.9 F 0.21 MnO 0.022 Al2O3 15.2 As2O3 0.18 V2O5 0.022 ZnO 8.9 CdO 0.14 ZrO2 0.016 K2O 2.15 S 0.102 Cl 0.013

Fe2O3 1.95 NiO 0.058 Sb2O3 0.007 PbO 0.85 P2O5 0.046 La2O3 0.006 TiO2 0.65 Cr2O3 0.044 Rb2O 0.0052 MgO 0.60 Co3O4 0.026 SrO 0.0041 CaO 0.30 WO3 0.025 L.O.I 6.57

Table 4.Results of Wet Chemical tests Oxides Weight % SiO2 62.5Al2O3 16.5K2O 2.25TiO2 0.45PbO 1.36

Fe 2O3 1.85ZnO 9.60

According to XRD studies the minerals of Quartz, Sericite, Smithsonite, Kaolinite and Mimetite has been identified in these ore samples. By combining the information obtained from Chemical analysis (Table 3,4), XRD and Ore microscopy, approximate mineralogical composition of the ore sample may be shown as follows: Zn Minerals - 16 % Pb Minerals - 0.6 % Quartz - 52% Sericite - 16.5% Kaolinite - 12 % Rutile - 0.4% Iron oxide (FeOOH) - 1.50% Other minerals - 1%

Figure.4 Polished section of the ore sample including Iron Oxides, Rutile and Pyrite. (PPL 32 x 8 Oil + Nicole)

Degree of Liberation Studies

The results of counting the mineral particles and also degrees of liberation have been showed in

224

Table 5. The best amount of liberation degree, which our crushed product has not been overgrinding, is the size fractions of -212 +150; -150+125; -125+75 μm (Polished Sections No. 6-7-8). The comparison between the above three size fractions has been showed in Figure 5 (Abouzaid, A.Z., Mineral Processing Laboratory, 1990). Figure 6 shows the Smithsonite particles locked into quartz particle.

Table 5.Degree of Liberation for all content mineral Size

Fractions(mm)

Smithsonite (%)

Quartz(%)

FeOX(%)

Clay Minerals

(%)- 2.36,+1.18 7.55 66.42 30.77 44.44-1.18 ,+0.85 30.76 85.47 38.71 64.52- 0.85,+0.60 32.65 83.33 47.62 66.67

- 0.60,+0.425 36.60 90.43 50.70 69.57- 0.425,+0.212 57.80 93.14 54.05 71.25- 0.212,+0.150 69.74 97.75 62.34 75.25- 0.150,+0.125 79.25 98.75 71.80 81.55- 0.125,+0.075 88.70 99.57 87.80 92.56- 0.075,+0.063 96.20 100.00 96.39 96.77

Figure 5.Comparison between Degrees of Liberation

Figure 6.Locked Smithsonite Particles in Quartz

Grindibility Studies

The results of grindibility studies for calculation of Bond index show in table 6.

Table 6.Results of Bond Work Index Feed

Particle Size(d80) (μm)

Product particle

Size(d80) (μm)

MeanGi(g)

P1(μm)

Work Index

F P Gi P1 Wi1800 122 0.36 150 11.93

Heavy Liquid Separation Tests

The results of heavy liquid tests show that the most Zn content based on the most area of the curve No.1 occurs in Figure 7 for size fraction -125+75 μm as compared to the other size fractions and also the least area of the curve No.1 occurs in Figure 8 for size fraction of -212+150 μm. The concerning grade and recovery is considered with using the results of curves No. 2 and 3.

The results of heavy liquid tests indicate that the most amount of curve slope occurs the more ability of gravity concentration for the size fraction -150+125 μm as compared to others (Figure 9).

50 75 100 150 200 300 400 600 800 1200 2000 4000 0

10

100

Mean Particle Size (Micron)

Deg

ree

of L

iber

atio

n(%

)

y1 = 100 exp(-0.0016181x)

Figure 7. Henry Reinhardt Diagram for –125 +75 μm

Figure 8. Henry Reinhardt Diagram for –212 +150 μm

y2 = 100 exp(-0.0002178x)

y3 = 100 exp(-0.0012141x)

y4 = 100 exp(-0.0005802x)

Degree of Liberation SmithsoniteDegree of Liberation QuarzDegree of Liberation Iron OxideDegree of Liberation Clay Mineral

50 75 100 150 200 300 400 600 800 1200 2000 4000 0

10

100

y2 = 100 exp(-0.0002178x)

0 5 10 15

0

Mean Particle Size (Micron)

Deg

ree

of L

iber

atio

n(%

)

y1 = 100 exp(-0.0016181x)

y3 = 100 exp(-0.0012141x)

y4 = 100 exp(-0.0005802x)

10

20

30

40Degree of Liberation SmithsoniteDegree of Liberation QuarzDegree of Liberation Iron OxideDegree of Liberation Clay Mineral

50

60

70

80

0 5 10 15 20 25 30 35 40

0

10

20

30

40

50

60

70

80

90

100

Zinc Grade %

CummulativeMass%(SinksvsFl

Curve 1Curve 2Curve 3

20 25 30 35 40

90

100

Zinc Grade %

CummulativeMass%(SinksvsFl

Curve 1Curve 2Curve 3

225

0 0.5 1 1.5 2 2.5 3 3.5 4

0

10

20

30

40

50

60

70

80

90

100

S.G. Fraction (gr/cm3)

CummulativeMass%(Floa

Curve 4

Figure 9. Henry Reinhardt Diagram for –150 +125 μm

CONCLUSION

Results suggest that an excellent method for mineral processing of this ore deposit will be gravity separation for size fraction of -150+125 μm as preconcentration and also the best degree of liberation will be -125+75 μm. Thus according to the obtained data, the size fraction 125+75 μm can be processed by flotation for next stages.

ACKNOWLEDGEMENTS

The Authors wish to thank Calcimine Co. for providing the samples which were used in this work. They would also like to thank Prof. H.J.Steiner, Department of mineral processing, university of Leoben, Austria. The assistance of colleagues at the Islamic Azad University (AIU), Southern Campus of Tehran, Tehran, Iran is also much appreciated.

REFERENCES

Abouzeid, A.Z.M, Mineral Processing Laboratory Manual, Trans Tech verlag, (1990). Bond, F.C., The third theory of Comminution,AIME, Vol.193, pp. 484-494, (1952).Gruender, Werner, Aufbereitungskunde, Hübner Verlag, pp.262-282, (1963). Kick, F., Das gesetzt der proportionalem widerstand und seine anvendung, Leipzig, (1885).Nematollahi, H., New size laboratory ball mill for bond Work Index determination. Mining Engineering, (1994). Rittinger, P.R., Lehrbuch der Aufbereitungskunde,Berlin, (1867).

226