Incorporation of zinc in ferrous sulfate monohydrate

Incorporation of zinc in ferrous sulfate monohydrate

Hydrometallurgy, 33 ( 1 9 9 3 ) 301-311 301 Elsevier Science Publishers B.V., A m s t e r d a m Incorporation of zinc in ferrous sulfate monohydrat...

577KB Sizes 0 Downloads 26 Views

Hydrometallurgy, 33 ( 1 9 9 3 ) 301-311

301

Elsevier Science Publishers B.V., A m s t e r d a m

Incorporation of zinc in ferrous sulfate monohydrate F. Elgersma, G.J. Witkamp and G.M. van Rosmalen Delft University of Technology, Laboratory for Process Equipment, Leeghwaterstraat 44. 2628 CA Delft, The Netherlands (Received July 11, 1992; revised version accepted September 10, 1992)

ABSTRACT Elgersma, F., Witkamp, G.J. and Van Rosmalen, G.M., 1993. Incorporation of zinc in ferrous sulfate monohydrate. Hydrometallurgy, 33:301-311. Continuous crystallization experiments on FeSO4" H20 in aqueous solutions containing Zn 2+ were carried out, at temperatures of 140-165°C, in a stainless steel/glass autoclave. The incorporation of Zn :~+ in FeSO4"H20 as a function of process parameters such as the temperature and the residence time was determined. Additionally, the solubility of FeSO4"H20 in this temperature region in the presence of ZnSO4 was determined. The crystallization of FeSO4-HzO yielded a product which conrained between 2 and 3 wt% Zn 2+, depending on the process conditions. The incorporation of Zn 2+ was also quantified using a partition coefficient which, together with the calculated results for the supersaturation in the various experiments, revealed that the incorporation of Zn 2+ in FeSO4-H20 increased with the supersaturation, but was mainly influenced by the thermodynamic value of the partition coefficient; that is, a value equal to 0.3 at zero supersaturation. The separation of Fe 2+ and Zn 2+ from aqueous solutions by crystallization of FeSO4"HzO has too low a selectivity to justify its use in an integrated hydrometallurgical jarosite treatment process.

INTRODUCTION

In hydrometallurgical zinc winning processes large quantities of iron-containing residues are formed because 2-12 wt% iron is present in the zinc concentrate. The iron residues jarosite (MFe3 (SO4) 2 ( O H ) 6 with M = NH + or Na + ) and goethite (FeOOH) have not found any applications because of their contamination with heavy metals such as lead, zinc and cadmium ,and with the metalloids arsenic and antimony. Therefore, they have to be disposed of in HDPE-lined ponds. Despite the fact that protective measures are taken to prevent leakage o f contaminated water from the disposal ponds to the soil, more and more pressure is nowadays exerted on zinc-winning comCorrespondence to: F. Elgersma, Delft University of Technology, Laboratory for Process Equipment, Leeghwaterstraat 44, 2628 CA Delft, The Netherlands.

0 3 0 4 - 3 8 6 X / 9 3 / $ 0 6 . 0 0 © 1993 Elsevier Science Publishers B.V. All rights reserved.

302

F. E L G E R S M A E T AL.

g salt/k{} H20 8C0-

6C0

~

°S0

4O0-

200 -

0 0

L~O

80

120

160

200

240

Temperature ('C)

Fig. 1. The solubilityof pure FeSO4"H20 and pure ZnSO4"H20 as a function of temperature. panies to search for processes which yield environmentally more acceptable residues. In the Netherlands, even the jarosite residues disposed of in the past will have to be reprocessed in the near future. This calls for residue treatment processes aimed at converting iron residues formed in hydrometallurgical zinc winning to products which may find application in civil engineering or in (non-ferrous) metal industries. Recently, five residue treatment processes aimed at processing a jarosite residue [ 1 ] have been described in the literature. One of these processes, the integrated hydrometallurgical jarosite treatment process, includes a process step in which a ferrous and zinc sulfate solution is crystallized in a cascade of autoclaves at a temperature of about 160 ° C. Under these conditions ferrous sulfate monohydrate (FeSO4-H20 ) is formed which contains a small amount of zinc. This means that the remaining mother liquor has become enriched in zinc compared to the feed. The process concept is based on the considerably lower solubility of pure FeSOa-H20 compared with pure ZnSOa-H20 at 160°C, which should lead to a comparatively high selectivity for the crystallization of FeSOa-H20. Figure 1 shows the solubility of pure FeSOa.H20 and pure Z n S O 4 ° H20 as a function of temperature [ 2 ]. This paper discusses the equipment required to perform the crystallization experiments and the results obtained. THEORY The driving force for the crystallization of FeS04- H20 is quantified by defining a dimensionless supersaturation, fl:

A]2 --=#=l. RT

_{". aFez+asoz4-aHzo "~ . . . k(aFe2+aso?~-aH2o)eq)

(1)

INCORPORATIONOFZINCIN FERROUSSULFATEMONOHYDRATE

303

where: ai = the activity of species i in the solution; A/~= the driving force for the phase transition (J mol-~); R = t h e gas constant 8.314 J mol-1 K - l ; T = the absolute temperature (K); the subscript eq = the equilibrium situation, which is strongly dependent upon temperature. The activity of a species, i, is given by: ai = YiCi

(2 )

where: 7i = the dimensionless activity coefficient; Ci = the concentration (mol 1-1). Calculation of the activity coefficients for FeSO4" H 2 0 at elevated temperatures, using Pitzer's equations [3,4 ], requires knowledge of the constants in these equations for FeSO4 and ZnSO4. The constants for FeSO4 are not known at temperatures beyond 70°C and extrapolating these temperature-dependent constants to 140-165 ° C seemed unacceptable. Consequently, it was not possible to calculate the activity coefficients for this FeSO4/ZnSO4 mixture at temperatures of 140-165 ° C. For this reason, and because the water activity of the supersaturated solution is not expected to be much different from that of the saturated solution, eq. ( 1 ) was simplified to: fl=ln(

CFe2+Cs°~k ( CFe2+ Cs042- )eq ]

(3)

Here the ionic product at equilibrium no longer equals the thermodynamic solubility product. In order to calculate fl it is essential to know the ionic product of FeSO4. H20 in the FeSO4/ZnSO4 solutions used in the experiments at temperatures from 140°C to 165°C. The solubility of FeSO4.H20 is, however, affected by the presence of zinc sulfate in the solution, owing to the presence of the additional sulfate ions, and can, therefore, not be calculated from solubility data obtained for pure ferrous sulfate solutions. For this reason, the solubility data were obtained from a solution with a Z n 2+ / F e 2+ molar ratio of 0.1, which was the same as the solution fed into the continuous crystallization experiments. The incorporation of zinc in FeSO4-H20 was quantified by a partition coefficient, D, as:

D-- [Zn]ff [Fe]s [Zn]ff[Fe]t

(4)

These D values were determined from continuous crystallization experiments.

304

F. E L G E R S M A ET AL.

EXPERIMENTAL

Determination of the solubility of FeSO4.H20 in aqueous FeSO4/ZnS04 solutions The solubility of FeSO4-H20 in the presence of Zn 2+ was determined from two batch experiments. A feed solution was used which contained 1.62 mol 1-1 FEZ+, 0.162 tool 1-1 Zn2+ and 1.809 mol 1- ~ SO 2- . In order to minimize oxidation of F e z+ during the preparation of the solution, H2SO4 w a s added. This caused the SO 2- concentration to be slightly higher than the sum of the Fe 2+ and the Zn a+ concentrations. During the preparation of the solution, as well as during the experiments itself, N2 gas was also used to minimize the oxidation. The solution was heated in a 1.5 1 autoclave with a heating jacket filled with oil to control the temperature within _+0.3°C. The solution was stirred at 400 m i n - ~with a two-blade impeller. In the first experiment the solution was heated to 140°C ( d T / d t = 1 °C m i n - 1) and then kept at 140 ° C for 3 h. The time o f 3 h included a safety margin, because similar experiments [ 5 ] have reported that equilibrium is reached after 2 h. After 2, 2.5 and 3 h solution samples were taken. The total concentrations o f iron and zinc in the solution samples were determined by atomic absorption spectroscopy (AAS). After 3 h at 140°C the solution was heated to 150 ° C, then to 160 ° C and finally to 170 ° C. In the second experiment the solution was heated to 165 °C (d T/dt = 1 °C min -~ ) and kept at 165°C for 3 h while solution samples were taken. The sampling procedure was repeated after the solution was cooled to 155 °C and thereafter to 145 ° C.

Continuous crystallization of FeSO4" H20 at 140-165 ° C A diagram of the equipment used is given in Fig. 2. A detailed description of the equipment is given elsewhere [6]. The m e m b r a n e pump (F) could yield a flow rate which corresponded with suspension residence times, varying between 0.5 and 2.0 h. The inaccuracy of the flow was below 1 wt%. The reactor (G) was a 1.5 1 autoclave with a jacket filled with heating oil. For safety reasons, the m a x i m u m temperature difference between the oil in the jacket and the suspension temperature was 50 ° C. This limited the temperature in the reactor to 150 ° C for a suspension residence time of 0.5 h, to 155 oC fi~r 1 h, to 160°C for 1.5 h a n d t o 165°C for 2 h. Nine experiments were carried out in which a ferrous and zinc sulfate solution, initially containing 1.62 mol 1-1 Fe:+ and 0.162 mol 1-1 Zn2+, was crystallized under the conditions given in Table 1. All experiments lasted nine residence times or even longer. Liquid samples were taken at least every resi-

INCORPORATION OF ZINC 1N FERROUSSULFATEMONOHYDRATE

305 V17

VlO

-

-

;,',~

4

¥16

V18

t3

@@

Kj

t

Fig. 2. The e q u i p m e n t used.

35

TABLE 1 Results of continuous crystallization experiments of FeSO4-H20 in the presence of Z n S O 4 at 140165°C Test

Temperature (°C)

Residence time (h)

Conversion 1 (%)

Supersaturation

Zn load (g/kg solids)

D

cl c2 c3 c4 c5 c6 c7 c8 c9

140 140 150 145 150 145 155 160 165

0.5 1.5 1.5 1.0 0.5 1.0 1.0 1.5 2.0

24 60 83 67 65 58 81 85 89

1.80 0.74 _2 0.76 1.22 1.12 0.22 0.014 0.002

19 23 28 25 25 25 28 30 30

0.41 0.35 0.31 0.36 0.34 0.35 0.30 0.31 0.31

J The conversion is defined as (Cve f e e d - C~e)ICFefeed. 2 For unknown reasons a negative supersaturation was calculated for this experiment.

dence time and a solid sample was taken from the fifth residence time, onwards. The molar ratio and concentration o f Fe 2 ÷ and Zn 2 ÷ were chosen such that

306

F. ELGERSMA ET AL.

the solution resembled a solution leaving a reductive leach autoclave in a hydrometallurgical jarosite treatment process [ 1 ]. RESULTS A N D D I S C U S S I O N

Solubility of FeSO4"n20 in the presence of Zn 2+ Two experiments were carried out to determine the solubility of FeSO4" H20 in the presence of Zn 2÷ in aqueous solutions at temperatures ranging between 140°C and 170°C. Figure 3 shows the results obtained for the Fe 2÷ concentration in experiment 1, (where the solubility was measured using a rising temperature program of 140-150-160-170 ° C ) and those obtained in experiment 2 (under a decreasing temperature program of 165-155-145 ° C ) . Identical solubility curves were obtained on heating and cooling. Figure 4 shows the results obtained in both experiments for the Zn 2+ solution concentration. The results in experiment 1 show a higher Zn 2÷ concentration than those obtained in experiment 2. The dashed line shows the 'average' value used in the calculations of the supersaturation, where the Zn 2÷ concentration is used as a measure for its contribution to the overall SO ]- concentration. The difference in the results of experiments 1 and 2 is explained by the fact that, in experiment 1, the experimental procedure included a fast rise in the suspension temperature to 140°C; whereas in experiment 2 the suspension temperature rapidly increased to 165 °C. In the latter experiment the major part of the crystals were formed at temperatures between 140°C and 165 ° C. The rapid increase in the temperature from 140 ° C to 165 ° C resulted in a high supersaturation value, which caused a higher uptake of zinc in FeSO4" H20. Consequently, the FeSO4. H20 crystals grown in experiment 2 contained more 20 Fe

(g/L)

• healing ocooling

~ 15 •

10I 6

c~

130

150

q70 Temperature I'O

Fig. 3. The equilibrium solubility Fe 2+ concentration as a function of temperature.

I N C O R P O R A T I O N O F Z I N C IN F E R R O U S S U L F A T E M O N O H Y D R A T E

Zn

307

(g/L)

6

heating Ocooling

• N. \

5



0

o 0

i~o

I~o

60 Temperature I'[)

Fig. 4. The zinc concentration in the solution as a function of temperature, measured during the FeSO4-H20 solubility experiments. Dashed line = an estimated, average value. zinc than those grown in experiment 1 (with a slow increase of temperature in the range 140-165 °C and therefore a lower supersaturation). In subsequently decreasing the temperature via 155 ° C to 145 ° C in experiment 2, the Fe 2+ concentration reached a new, higher level of solubility, with a correspondingly higher Fe 2÷ concentration. The solid phase, however, still has its 'fast growth history' and, therefore, the Zn z+ concentration found in experiment 2, although increasing, remained lower than in experiment 1. Since the substitution of Fe e ÷ by Zn z + ions in the FeSO4. HeO lattice gives no change in the S O l - concentration in the solution, the growth history was not found to influence the FeSO4. HzO solubility.

Results of the continuous crystallization of FeSO4.n20 The results obtained in the experiments c 1 to c9 are summarized in Table 1.

Characterization of the product Figure 5 shows a representative SEM photomicrograph of a few FeSO4-H20 agglomerates. The rounding o f the crystal edges reveals that some dissolution occurred during the sampling. Using a Bfichner funnel it was qualitatively shown that the filtration behavior o f the solids was excellent. Some dissolution of fines could, however, have happened during sampling. Guinier de Wolff X-ray diffractograms indicate the presence o f FeSO4-H20 as the only crystalline phase.

308

F. ELOERSMA ET A L

Fig. 5. A secondaryelectronmicrographshowinga few representativeFeSO4-H20 crystals.The rough surfacesindicatethat during sampling some dissolution took place.

Conversion values The precipitation of FeSO4.H20 is higher at higher temperatures (compare, for example, experiment cl with c5) or longer residence times (compare experiment c I with c2 ). The reproducibility of the conversion achieved is, as is obvious from the experiments c4 and c6, about 10%. No explanation is available for the unexpectedly high conversion observed in experiment c3, which also resulted in the calculation of a negative supersaturation for that experiment, following from a too low Fe 2+ concentration in the solution.

Supersaturation values The supersaturation values were calculated from the Fe 2+ concentration determined in the solution using eq. (3). It was assumed that the SO 2- concentration was the sum of the Fe 2+ and Zn 2+ concentrations, plus the contribution of the H2804 addition. High fl values were obtained for experiments with a low conversion and low values for experiments with a high conversion.

INCORPORATION OF ZINC IN FERROUS SULFATE MONOHYDRATE

Incorporation

o f Z n 2+

309

in FeSOa'H20

The incorporation of Zn= + in FeSO4-H20 is expressed by two parameters: by the absolute quantity of zinc in FeSO4-H20 in grams per kilogram of solids, and by the partition coefficient, D, which is a more proper measure for the selectivity of the process. The reproducibility of the data on the incorporation, expressed in the partition coefficient, D, is +0.01, which is sufficiently good to distinguish between the various experiments. Experiment c 1 shows a comparatively high D value, which is related to a comparatively high value of the supersaturation (0.90). The experiments c2, c4, c5 and c6 together form a group of experiments with D values ranging between 0.34 and 0.36, and corresponding to supersaturation values of between 0.74 and 1.22. Finally, the experiments c3, c7, c8 and c9 form a group with D values of 0.30 and 0.31 and low supersaturation levels of 0.22 or less. Figure 6 shows the relation between D and fl for these experiments, assuming a fl value of 0.2 for experiment c3. Figure 6 obscures the effect of the temperature on the D value, because the D values depend on both the temperature and the supersaturation. It is not possible to distinguish between the influence of the temperature and the supersaturation. However, comparing the D values measured in experiments c 1 and c2, or c5 and c3, it becomes clear that these couples of isothermal experiments show considerable differences in D values. Thus, higher D values can only be related to the higher supersaturation levels in the various experiments. This provides evidence that the supersaturation probably has a larger influence on the D value than the temperature. This is disappointing, because upon increasing the temperature, a significantly lower D value was expected Partition coefficient 0 (-) 045

0,40

0.35

0.30

O.&

1.2

2

SUPERSATURATION ~ (-)

Fig. 6. The partition coefficient,D, as a function of the supersaturation.

310

F. ELGERSMAETAL.

due to the lower solubility of pure FeSO4-HzO relative to that of pure ZnSO4 °H 2 0 (Fig. 1 ). A relationship between D and the ratio of the solubilities of salts was given by Balarew [ 7 ] and Witkamp [ 8 ]. The zinc content of the FeSOa-H20 ranges between about 2 and 3 wt% and reaches the highest values for experiments with a high conversion. Despite the fact that the partition coefficient, D, decreases with increasing conversion, the zinc content of the FeSOa-H20 increases. This rather unexpected result is caused by the fact that the zinc content of the solid phase is a product of two terms, as shown by: [Zn2+ ]s D [ Z n 2 + ]1 ire2 + =

(5)

An increasing conversion is accompanied by a low equilibrium concentration, which yields a lower value of D. Simultaneously, an increasing conversion also leads to a large decrease in the Fe z+ concentration in the solution, which, despite a smaller decrease in the Z n 2+ concentration, increases the :ratio of Zn 2+ to Fe z+ in the solution. The resulting zinc content of the solid phase is higher because the decrease in the partition coefficient with a decreasing supersaturation is smaller than the increase of the Zn 2+ / F e z+ ratio :in the solution for high conversion experiments. The main conclusion which can be drawn from these experiments is that, despite the considerable differences in the solubilities of the pure salts FeSO4-H:O and ZnSO4-H20, the selectivity for the crystallization of FeSOa.H20 in the presence of Zn 2+ is poor. Although the incorporation of Zn 2+ is, to some extent, determined by the supersaturation, the relatively large value of the partition coefficient is mainly due to the high thermodynamic equilibrium concentration of Z n 2+ in FeSO4- H20. As follows from the material and energy balances calculated in [ 1], for D = 0.3, the selectivity of this separation process is too poor to justify its application as a key step in an integrated hydrometallurgical jarosite treatment process. CONCLUSIONS

The crystallization of FeSO4"H20 from an aqueous solution containing Zn 2+ in a continuously operated autoclave at 140-165 °C yielded a well-filterable, crystalline product, of which the filtration properties might have been obscured somewhat by the dissolution of fines. The incorporation of Zn 2+ in FeSO4-H20 ranged between 2 and 3 wt% and depended slightly on the process conditions, such as the temperature and the residence time. The partition coefficient, D, for the incorporation of Zn 2+ in FeSO4-H20 varied between 0.30 and 0.41 and was shown to increase with increasing su-

INCORPORATION OF ZINC IN FERROUS SULFATE MONOHYDRATE

31 1

persaturation. The partition coefficient was mainly determined by the D value at zero supersaturation and not by the crystallization kinetics. The separation of Fe 2+ and Zn 2+ from aqueous solutions by crystallization of FeSO4"H20 has too low a selectivity to justify its use in an integrated hydrometallurgical jarosite treatment process. ACKNOWLEDGEMENTS

The authors wish to thank the Dutch ministry of Housing, Physical Planning and the Environment for financially supporting this study and A.P.R. de Kort and A.H.W. Rooijmans for their contribution. The research of G.J. Witkamp was made possible by a fellowship from the Royal Netherlands Academy of Arts and Sciences.

REFERENCES 1

2 3 4 5 6 7

8

Elgersma, F. and Zegers, T.W., Integrating jarosite residue processing in hydrometallurgical zinc refining---comparison of five potential processes. In: R.G. Reddy, W.P. Imrie and P.B. Queneau (Editors), Residues and Effluents, Processing and Environmental Considerations. TMS-AIME, Warrendale, Pa. ( 1992 ), pp. 413-448. Bruhn, G., Gerlach, J. and Pawlek, F., Solubilities of salts and gases in water. Z. Anorg. Allg. Chem., 337 (1965): 68-79 (in German). Pitzer, K.S. (Editor), Activity Coefficients in Electrolyte Solutions. CRC Press, Boca Raton, Fla., 2rid Ed. ( 1991 ). Pitzer, K.S., Theoretical considerations of solubility with emphasis on mixed aqueous electrolytes. Pure Appl. Chem., 58 ( 1986 ): 1599-1610. Kammel, R., Pawlek, F. and Simon, M., Solubility experiments of FeSO4 in ZnSO4 containing solutions. Chem. Ing. Tech., 58 (1986): 323-325 (in German). Elgersma, F., Integrated hydrometallurgical jarosite treatment. Ph.D. Thesis, Delft Univ. Technology (1992). Balarew, C., Isomorphous and isodimorphous admixtures in minerals-crystal hydrate salts. In: Proc. 13th Gen. Meet. Int. Mineral. Assoc. Publishing House Bulg. Acad. Sci., Varna (1986), pp. 287-293. Witkamp, G.J., Crystallization of calcium sulfate and uptake of impurities. Ph.D. Thesis, Delft Univ. Technology (1989).