ARTICLE pubs.acs.org/est
Removal of Arsenic and Antimony from Anode Slime by Vacuum Dynamic Flash Reduction Deqiang Lin and Keqiang Qiu* College of Chemistry and Chemical Engineering, Central South University, Changsha 410083, Hunan, China ABSTRACT: Anode slime is an important material of recycling precious metals. Up to now, treating the arsenic- and antimony-rich anode slime by conventional processes has the following problems: its economic and environmental effect is less than satisfactory, and the removal effect of arsenic and antimony from anode slime in present processes is not all that could be desired. Therefore, vacuum dynamic flash reduction, a new process for treating arsenic- and antimony-rich anode slime, was investigated in this work. During vacuum dynamic flash reduction, silver from the arsenic- and antimony-rich anode slime was left behind in the distilland as the silver alloy, and trivalent oxides of arsenic and antimony were evaporated in the distillate. The experimental results showed that the evaporation percent of the arsenic- and antimony-rich anode slime was 65.6%. Namely, 98.92% by weight of arsenic and 93.67% by weight of antimony can be removed under the following experimental conditions: temperature of 1083 K, vacuum evaporation time of 60 min, and air flow rate of 400 mL/min corresponding to the residual gas pressure of 250 Pa. Moreover, vacuum treatment eliminates much of the air pollution and material losses associated with other conventional treatment methods.
1. INTRODUCTION In China and elsewhere many lead ores contain arsenic and antimony, which enter crude lead as impurities in the leadmaking process. If crude lead is refined by electrolytic process, impurity arsenic and antimony enter again the lead anode slimes, which is a byproduct of electrolytic refining of crude lead, and an important material of recycling precious metals. Anode slimes are characterized by high concentrations of gold, silver, lead, and copper. Slimes without the lead or copper are then enriched in gold and silver. To obtain gold and silver, the other elements must be separated from anode slimes. In this case, arsenic and antimony have a significant impact on the process for treating anode slime. According to the content of arsenic, the anode slimes are usually divided into arsenic-rich anode slime and arsenic-low anode slime. There are many studies about the treatment of arsenic-low anode slime based on conventional pyrometallurgical process or hydrometallurgical process.15 However, treating the arsenic- and antimony-rich anode slime by conventional processes has the following problems: its economic and environmental effect is less than satisfactory, and the removal effect of arsenic and antimony from anode slime in present processes is not all that could be desired, because arsenic and antimony is widely distributed in various parts of the process, which result in not only some trouble in recovery of arsenic and antimony but also serious environmental problem.6 On the other hand, as is well-known, arsenic and arsenic compounds are especially potent poisons. However, they have high application value in the manufacture of pesticides (including r 2011 American Chemical Society
wood preservatives), electronic components (semiconductors), glass, alloys, pigments, and pharmaceuticals. 7 To turn the hazardous material containing arsenic and antimony into a good one, it is of great interest to develop and implement both an environmentally friendly and technically viable process for recycling arsenic and antimony from anode slime. That is the reason we undertook an investigation of this topic. The previous selective removal of arsenic and antimony may contribute to reduce environmental pollution and simplify the process for treating the arsenic- and antimony-rich anode slime. Vacuum method is considered to be feasible to solve this problem, which shows many advantages such as simple technological flow sheet, no or low environmental pollution,and low consumption of raw material and energy.8,9 Moreover, according to the characteristics of arsenic- and antimony-rich anode slime and the difference in saturation vapor pressure of each substance in anode slime, vacuum dynamic flash reduction was investigated in this paper. During vacuum dynamic flash reduction, As2O3 and Sb2O3, as soon as produced by reducing the high valence oxides of arsenic and antimony, are evaporated at once and stop being reduced to metals further. Arsenic and antimony can be removed from arsenic- and antimony-rich anode slime by this new method. Received: October 10, 2010 Accepted: March 21, 2011 Revised: March 16, 2011 Published: March 29, 2011 3361
dx.doi.org/10.1021/es103424u | Environ. Sci. Technol. 2011, 45, 3361–3366
Environmental Science & Technology
ARTICLE
2. EXPERIMENTAL SECTION 2.1. Experimental Materials. An arsenic- and antimony-rich anode slime from Hunan province in China was chosen for the experiments. Table 1 gives the main chemical composition of metallic elements in the arsenic- and antimony-rich anode slime. As shown in Table 1, there was 19.35 wt % of arsenic, 24.75 wt % of antimony, and 3.88 wt % of silver in the sample. The reductant used in this study was charcoal powder. The XRD analysis results of arsenic- and antimony-rich anode slime are showed in Table 2. The XRD shows that the main components in arsenic- and antimony-rich anode slime are Pb2As2O7 and AsSbO4. A rich (in metallic values) complex metal oxides including Pb3O4, Sb2O4, and Sb2O3, are also present in the slime. Silver is mainly presented as Ag3Sb in this sample. 2.2. Apparatus. The experiments were carried out on the selfmade apparatus shown in Figure 1. The middle section of a stainless steel pipe was heated to the desired temperature by an electric-tubular furnace to evaporate anode slime from a porcelain boat. The furnace temperature was controlled by a temperature controller connected with a thermocouple. The outlet of the stainless steel pipe was connected with the condensation cavity and vacuum pump. 2.3. Procedures. The porcelain boat loaded with 10 g of anode slime mixed with a certain dosage of reductant was placed at the appropriate position in the stainless steel pipe and was heated to certain temperature under vacuum, then the timing of the vacuum treatment process started. After the experiment, the cooled residue in the porcelain boat was taken from the vacuum chamber, weighed, and sampled. Silver alloy could be collected from the distilland. The content of As and Sb in the silver alloy was determined by titrimetric analysis. The sample of silver alloy was dissolved in sulfuric acid and potassium sulfate, then arsenic and antimony were analyzed, respectively, by potassium bromate titration and ceric sulfate titration. The evaporation percent (E) of anode slime and the removal percent (R) of arsenic or antimony were calculated by the following formulas:
m1 m2 100% m1
ð1Þ
m1 W 1 m3 W 2 100% m1 W 1
ð2Þ
E¼
R ¼
where m1 is the mass of the arsenic- and antimony-rich anode slime (g); m2 is the mass of the distilland (g); m3 is the mass of
the silver alloy; W1 is the content of arsenic or antimony in arsenic- and antimony-rich anode slime (wt %); and W2 is the content of arsenic or antimony in the silver alloy (wt %). 2.4. Exploratory Experiment. To research the removal effect of arsenic and antimony from anode slime under vacuum, four exploratory experiments were carried out which mainly consisted of two parts: one was carried out in static state, and the other was carried out in dynamic state. Two experiments were carried out in static state at 1033 K for 60 min with 10 wt % dosage of reductant or without reductant, respectively. “Static state” means that no flow of air was introduced to the system during the experiment. The other two experiments were conducted in dynamic state with 10 wt % dosage of reductant or without reductant, respectively. “Dynamic state” means that air with the flow rate of 400 mL/min was introduced through the pipe inlet to the system during the experiment. The residual gas pressure of the static state experiments was kept constant at 30 Pa while for dynamic state experiments it was kept at 250 Pa. The experimental results are summarized in Figure 2. As can be seen from Figure 2, if without reductant, the evaporation percent of anode slime was only 29.1% in static condition, and it reached 30.8% in dynamic condition; if with 10 wt % dosage of reductant, it increased from 51.9% in static state to 63.9% in dynamic state. It can be seen from this that the evaporation percent of anode slime can be increased in dynamic state, no matter whether with reductant or without reductant.
3. RESULTS AND DISCUSSION 3.1. Factors on Anode Slime Evaporation. The factors influencing the evaporation percent of arsenic- and antimonyrich anode slime were studied under different experimental conditions including reductant dosage, reduction temperature, gas flow rate, and vacuum treatment time. 3.1.1. Effect of Reductant Dosage. The effect of reductant dosage on the evaporation percent was investigated in the range from 2.5 to 12.5 wt %, at temperature 1033 K, vacuum treatment time 60 min, and air flow rate 400 mL/min, corresponding to the residual gas pressure of 250 Pa. The relationship between evaporation percent and reductant dosage is shown in Figure 3a. As shown in Figure 3a, the evaporation percent of anode slime increases from 38% to 63.9% when the reductant dosage changes from 2.5 to 10 wt %. Nevertheless, the evaporation percent drops
Table 2. Phase Compounds in Anode Slime element
Table 1. Chemical Composition of Arsenic- and AntimonyRich Anode Slime element
Pb
Sb
As
Ag
Cu
Bi
content (wt %)
17.01
24.75
19.35
3.88
1.92
3.28
form
Pb
Pb3O4, Pb2As2O7
As
Pb2As2O7, AsSbO4
Sb
AsSbO4, Sb2O3, Sb2O4, Ag3Sb
Ag
Ag3Sb
Figure 1. Schematic illustration of apparatus: 1, gas inlet; 2, stainless steel pipe; 3, furnace; 4, porcelain boat; 5, thermocouple; 6, water jacketed condenser; 7, gas outlet; 8, condenser pipe. 3362
dx.doi.org/10.1021/es103424u |Environ. Sci. Technol. 2011, 45, 3361–3366
Environmental Science & Technology a little with the growth of reductant dosage when the dosage is over 10 wt %. It can be explained that the metallic oxides are reduced to their metals due to the higher dosage of reductant. Because the volatility of these metals is poorer than that of their trivalent oxides, the evaporation percent of anode slime decreases certainly. 3.1.2. Effect of Reduction Temperature. The experiments were conducted in the temperature range from 883 to 1083 K, maintaining the reductant dosage at 10 wt %, vacuum treatment time at 60 min, and the air flow rate at 400 mL/min corresponding to the residual gas pressure of 250 Pa. The relationship between
Figure 2. Results of exploratory experiments.
ARTICLE
evaporation percent and reduction temperature is shown in Figure 3b. As can be seen from Figure 3b, the evaporation percent of anode slime increases obviously as the temperature is increased. When the temperature increases from 883 to 1083 K, the evaporation percent correspondingly increases from 49.6% to 65.6%. This can be explained by the fact that the vapor pressure of As 2 O3 and Sb2O3 increases as the temperature is increased. 3.1.3. Effect of Gas Flow Rate. The effect of gas flow rate was investigated by running experiments using the air flow rate in the range from 200 to 600 mL/min, when the temperature was 1033 K, reaction time was 60 min, and reductant dosage was 10 wt %. The relationship between evaporation percent and gas flow rate is shown in Figure 3c. When the gas flow rate increases from 200 to 400 mL/min, the evaporation percent of anode slime changes from 56.4% to 63.9%. This can be interpreted as follows: if the gas flow rate is small, the evaporation percent of anode slime increases, because the vapor molecule evaporated can be driven off by the gas flow and the partial pressures of evaporating substance immediately above the surface is reduced. When the gas flow rate is more than 400 mL/min, the evaporation percent gradually decreases with increasing the gas flow rate. This is because increasing the gas flow rate causes an increase in the residual gas pressure of system, which has weakened the function of the vacuum above the evaporation surface. 3.1.4. Effect of Vacuum Treatment Time. The object of the experiments in this series is to observe the influence of vacuum treatment time on the evaporation percent of arsenic- and antimony- rich anode slime. These experiments were carried out at 1033 K with reductant dosage of 10 wt % and air flow rate
Figure 3. Effects of (a) reductant dosage, (b) reduction temperature, (c) gas flow rate, and (d) vacuum treatment time on evaporation efficiency. 3363
dx.doi.org/10.1021/es103424u |Environ. Sci. Technol. 2011, 45, 3361–3366
Environmental Science & Technology
ARTICLE
Table 3. Chemical Composition of Silver Alloy element
Pb
Sb
As
Ag
Cu
Bi
content (wt %)
43.02
10.31
1.38
23.02
5.03
11.03
Figure 5. Relationship between the vapor pressure and temperature for various substances.
Figure 4. XRD patterns of (a) primary condensate from anode slime after vacuum dynamic flash reduction, (b) secondary residue from primary condensate after vacuum evaporation, and (c) secondary condensate from primary condensate after vacuum evaporation.
of 400 mL/min. Vacuum treatment time were changed from 30 to 90 min. The experimental results are shown in Figure 3d. As shown in Figure 3d, within 60 min, the evaporation percent of anode slime increased gradually as the vacuum treatment time was increased. For an increase in time from 30 to 60 min, the evaporation percent increased from 56.4% to 63.9%. When the vacuum treatment time was more than 60 min, the evaporation percents remained constant as the vacuum treatment time was increased. After the vacuum treatment time of 60 min, prolonging the vacuum treatment time will result in more energy consumption, which comes to naught. 3.2. Products. After the experiment, silver from arsenic- and antimony-rich anode slime was left behind in the distilland as silver alloy, and low valence oxides of arsenic and antimony was evaporated in the distillate, which were collected from the condensation cavity. 3.2.1. Silver Alloy. Silver can be absorbed by lead melt. After vacuum dynamic flash reduction was carried out at 1083 K for 60 min with reductant dosage of 10 wt % and air flow rate of 400 mL/min, 1.52 g of silver alloy was separated from slag and obtained at the bottom of the container. The chemical composition of silver alloy is given in Table 3. From Table 1 and Table 3, it means that 98.92% by weight of arsenic and 93.67% by weight of antimony can be removed through eq 2. The content of silver in the silver alloy from this experiment is almost six times as much
as that of the arsenic- and antimony-rich anode slime. Silver can be obtained from the silver alloy through a rotary oxidizing refining process and a electrolytic refining process. 3.2.2. Condensate. The XRD patterns of the distillate after vacuum dynamic flash reduction, namely, primary condensate, are shown in Figure 4a. As shown in Figure 4a, the primary condensate is a mixture of As2O3 and Sb2O3 according to the standard ICSD card. To separate the mixture of As2O3 and Sb2O3, primary condensate was treated again by vacuum evaporation at 623 K for 60 min under the residual gas pressure of 50 Pa, in which Sb2O3 from primary condensate was left behind in the secondary distilland and As2O3 was evaporated in the secondary distillate. Figure 4b and c show the XRD patterns of secondary residue and secondary condensate, respectively. According to Figure 4b and c, it can still be shown that the main composition of secondary residue and secondary condensate is Sb2O3 and As2O3, respectively. Meanwhile, little of impure peak is found in Figure 4b and c, which shows the purity of the secondary residue and the secondary condensate is quite good. This means that the primary condensate, namely, the mixture of As2O3 and Sb2O3, can be separated satisfactorily and recycled in an environmentally friendly manner by vacuum evaporation.
4. PRINCIPLE ANALYSIS The difference in saturation vapor pressure of each substance in anode slime at the same temperature is the basic principle for separating them from each other under vacuum. Based on the relationship between saturation vapor pressure and temperature,10,11 Figure 5 shows the saturation vapor pressure of As2O3, Sb2O3, PbO, As, Sb, Pb, Bi, Ag, and Cu in the temperature range from 650 to 1250 K. As shown in Figure 5, the saturation vapor pressures of substance is proportional to the temperature. It is an important fact that the saturation vapor pressures of As2O3 and Sb2O3 are much higher than those of others at the same temperature. Therefore, As2O3 and Sb2O3 with high vapor pressure can be evaporated into gas phase, while others with low vapor pressure remain in the distilland. It is easy to separate trivalent oxides of arsenic and antimony from the anode slime at the given reducing reaction temperature under vacuum. However, the XRD analysis results showed that this arsenic- and antimony-rich anode slime 3364
dx.doi.org/10.1021/es103424u |Environ. Sci. Technol. 2011, 45, 3361–3366
Environmental Science & Technology
ARTICLE
Figure 6. Schematic diagram of vacuum evaporation: (a) static; (b) dynamic.
Figure 7. XRD pattern of residue sample from exploratory experiment without reductant in dynamic state.
contains not only trivalent oxide of arsenic and antimony but also the high valence oxides of arsenic and antimony. Because the saturation vapor pressure of the high valence oxides of arsenic and antimony is very low, they can not be evaporated at the experimental temperature under vacuum. Therefore, it is hard to separate the high valence oxides of arsenic and antimony from the anode slime by vacuum evaporation. However, if the high valence oxides of arsenic and antimony from anode slime are reduced to their low valence oxides, the problem of separating the high valence oxides of arsenic and antimony from the anode slime can be solved. Besides, we still hope that the oxides of arsenic and antimony from the anode slime are reduced to metal arsenic and antimony with as little quantity as possible, because their saturation vapor pressure is much lower than that of their trivalent oxides. As2O3 and Sb2O3 have a very large intrinsic rate of evaporation at far below their normal boiling points. However, at atmospheric pressure, the large numbers of As2O3 and Sb2O3 evaporated are almost completely deflected back into the liquid surface by air molecules. As a result, an overall evaporation percent depends not only upon the intrinsic evaporation percent but also upon the ability of As2O3 and Sb2O3 to move away from the evaporation surface. The function of the vacuum above the evaporation surface is to remove air molecules, so that the chances of deflection of the evaporated As2O3 and Sb2O3 back into its source are reduced. Therefore, vacuum can create the following condition: As2O3 and Sb2O3, as soon as produced by reducing the high valence oxides of arsenic and antimony, are evaporated at once and stop being reduced to metals further. Because this reaction occurred in a flash, the process was defined as vacuum flash reduction.
According to the thermodynamics of evaporation, when a vapor is in equilibrium with a liquid, the rate of evaporation exactly equals the rate of condensation back upon the surface of the liquid; in other words, the net evaporation percent is practically zero (Figure 6a). If not in equilibrium, the evaporation percent depends upon the partial pressures of evaporating substance immediately above the surface under definite conditions of pressure and temperature. The smaller the partial pressures of the evaporating substance, the larger the rate of evaporation. To reduce the partial pressures of evaporating As2O3 and Sb2O3, a small air flow is introduced to blow their vapor molecules away from the evaporating surface during the period of vacuum reduction, as shown in Figure 6b. In this study, the process shown in Figure 6b was defined as vacuum dynamic flash reduction, while the process of vacuum flash reduction without a small air flow was defined as vacuum static flash reduction (Figure 6a). Therefore, the effect of separating As 2O3 and Sb2 O3 from the anode slime by vacuum dynamic flash reduction is better than that of vacuum static flash reduction. Figure 7 shows the XRD pattern of residue sample from the exploratory experiment without reductant in dynamic state under vacuum. As seen in Figure 7, the main components of the residue are the high valence compounds of arsenic and antimony. This indicates that low valence oxides of arsenic and antimony can be removed from anode slime even if without reductant, but not their high valence oxides. Therefore, the high valence oxides of arsenic and antimony should be converted into low valence oxides with reductant, or the removal effect of arsenic and antimony from arsenic- and antimony-rich anode slime is very bad by vacuum evaporation.
’ AUTHOR INFORMATION Corresponding Author
*Phone: þ86 731 88836994; fax: þ86 731 88879616; e-mail:
[email protected].
’ REFERENCES (1) Li, D.; Wang, Y. Extraction and Refining of Precious Metals; Central South University Press: Changsha, 2003 (in Chinese). (2) Ludvigsson, B.; Larsson, S. Anode slimes treatment: The Boliden experience. JOM-US 2003, 55 (4), 41–44. (3) D€onmez, B.; Sevim, F.; C-olak, S. A study on recovery of gold from decopperized anode slime. Chem. Eng. Technol. 2001, 24 (1), 91–95. (4) Fernandez, M. A.; Segarra, M.; Espiell, F. Selective leaching of arsenic and antimony contained in the anode slimes from copper refining. Hydrometallurgy 1996, 41 (23), 255–267. 3365
dx.doi.org/10.1021/es103424u |Environ. Sci. Technol. 2011, 45, 3361–3366
Environmental Science & Technology
ARTICLE
(5) D€onmez, B.; Ekinci, Z.; C-elik, C.; C-olak, S. Optimisation of the chlorination of gold in decopperized anode slime in aqueous medium. Hydrometallurgy 1999, 52 (1), 81–90. (6) Wu, J.; Lu, Y. Review on pre-treatment processes of arsenic-rich lead anode slime. Nonferrous Met. 2002, 54 (B07), 180–185in Chinese. (7) Choong, T. S. Y.; Chuah, T. G.; Robiah, Y.; Koay, F. L. G.; Azni, I. Arsenic toxicity, health hazards and removal techniques from water: an overview. Desalination 2007, 217 (13), 139–166. (8) Qiu, K.; Zhang, R. Research on preparation of nanometer antimony trioxide from slag containing antimony by vacuum evaporation method. Vacuum 2006, 80 (9), 1016–1020. (9) Hu, H.; Qiu, K.; Xu, G. Preparation of nanometer δ- and β-bismuth trioxide by vacuum vapor-phase oxidation. Trans. Nonferrous Met. Soc. China 2006, 16 (1), 173–177. (10) Kubaschewski, O.; Alcock, C. B. Metallurgical Thermochemistry; Pergamon Press: Oxford, 1979. (11) Speight, J. G. Lange’s Chemistry Handbook, 16th ed.; McGraw-Hill: New York, 2005.
3366
dx.doi.org/10.1021/es103424u |Environ. Sci. Technol. 2011, 45, 3361–3366
