Recovery of Antimony by Hydrolysis in Side Streams from Copper Production

Recovery of Antimony by Hydrolysis in Side Streams from Copper Production 1 Recovery of Antimony by Hydrolysis in Side Streams from Copper Production Eduardo Díaz*, José Antonio Maldonado**, Antonio Paúl*, José María Gallardo*, Irene Ruiz**, Guillermo Ríos ** *Department of Materials Engineering and Science and Transportation, University of Seville Avda. Camino de los Descubrimientos s/n, 41092, Seville Spain **Atlantic Copper S.LU. Company Avda. Francisco Montenegro s/n 21001 Huelva, Spain Keywords: Antimony metallurgy, Antimony oxychloride, Hydrolysis, Copper electro-refining. Abstract Atlantic Copper S.L.U. is a company dedicated to the treatment of copper ore concentrates and their transformation into anodes and cathodes in order to obtain high quality copper. To achieve this objective, it is crucial to control minor elements such as antimony, bismuth, and arsenic that are incorporated into the electrolyte from the anode. An ion exchange resin plant is used to control antimony and bismuth in the electrolyte. During the resin regeneration stage, called elution, a secondary stream with a significant antimony content is obtained. This metal is of interest, as it is classified as a critical raw material. This paper presents the results of the antimony extraction carried out by hydrolysis using water. The pH and temperature conditions that maximise yield extraction have been found. Also, the role of Bi and As present as impurities has been identified. 1. Introduction Atlantic Copper Metallurgical Complex in Huelva (Spain) was commissioned in 1970 with a capacity of 40,000 tpy of copper from concentrate. In 1975 the old blast furnaces were replaced by the Outokumpu FSF technology and the copper refinery capacity was increased up to 108,000 tpy. From 1994 to 1996 an expansion project was carried out so that the copper from concentrate and cathodes increased up to 250,000 tpy and 215,000 tpy respectively. This expansion consisted basically of a new sulphuric acid plant, replacement of the anode furnaces by two 400 t units, increase of the electrolytic cells up to 1120 and implementation of the ISA permanent cathode technology. The successive expansions of the metallurgical complex and some of the changes made in the process led to a progressive increase in the contents of As, Bi, and Sb. The presence of this kind of metals or semimetals, ç adversely affects the current efficiency and quality of the cathodes. Therefore, Atlantic Copper has conducted several studies with the aim of controlling the content of these impurities in 2 the electrolyte. In [1] , the relationship between the electrolyte content of As/(Sb+Bi) was determined to prevent the phenomena that affect the quality of the copper cathode. In a subsequent work, different procedures were studied for the control of impurities, mainly Sb and Bi. In later years and given the need to treat minerals with high antimony and bismuth content, Atlantic Copper addressed the possibility of installing an electrolyte treatment plant to control these elements, assessing in this case adsorption techniques, specifically electrolyte resins. Ion exchange and activated carbon since molecular recognition is ruled out by the already mentioned fact of effectiveness only in the removal of bismuth. In [2] tests were carried out in a pilot plant for ion exchange resins, in which it was concluded that the use of resin with aminoethyl phosphoric acid functional group is effective for the retention of Sb and Bi. In subsequent studies, also carried out in pilot facilities [3], the operating parameters are optimized given the variability of the starting concentrations of the electrolytes, and taking into account the possibility of resin poisoning due to the presence of Fe (III). The study concludes that an initial stage of pre-treatment for the removal of iron through a reduction with metallic copper. In addition, alternative procedures for the regeneration of the resin are proposed, with sulfuric acid and HCl mixtures with the addition of thiourea. Finally, the plant is put into operation in 2018, and its final process scheme is described below (Figure 1). Recovery of Antimony by Hydrolysis in Side Streams from Copper Production 3 Figure 1: Simplified process block diagram SbBi plant. In the resin regenerated stage the retained cations were removed by elution with HCl in accordance with Eq. (1): 2SbR3+ 6HCl→2SbCl3 +6(HR) (1) The product or effluent at this stage, commonly referred to as an eluate, consists of an acid chloride stream with chloride concentrations of approximately 170 g/l and a variable content of Sb, Bi, and other impurities. The typical composition of eluate is shown in Table 1. The eluate temperature is 35°C. 4 Table 1: Typical composition of ion-exchange facility eluate (g/l). As Bi Sb Cl- Others 2-5 8-10 9-10 170 <1 The eluate is currently neutralized using Ca(OH)2, and the resulting solids are currently treated as waste. Therefore, the search for an alternative processing is of great interest. Furthermore, antimony is classified as a critical metal in the European Union [4]. There are several reasons for this decision: (i) China’s majority in production, its mining of more than 52% of the global primary antimony, and its ownership of most of the declared resources; (ii) the 1lack of political and social stability of other producing countries; (iii) antimony economic importance; (iv) abundance; (v) the existence of possible substitutes; and (vi) the low recycling rate and high environmental risks, among other factors. Hence, this papers reflects the results of antimony recovery from the eluate by means of water hydrolysis. The effect of temperature, pH, as well as the effect of the presence of arsenic and bismuth and their effect on the process yield has been studied. 2. Antimony hydrolysis under eluate conditions 2.1. Eluate composition Based on Atlantic Copper data, the composition of antimony, arsenic, and bismuth exhibits a high dispersion, inherent to industrial streams. However, a statistical study of the data enables delimitation of the compositions to represent all values, thereby limiting the scope of the tests. Results indicate that antimony has a nominal composition of 10 g/L, with maximum and minimum values of 15 g/L and 5 g/L, respectively. Bismuth occurs in similar concentrations, with a nominal composition of 7.5 g/L, a minimum value of 4 g/L, and a maximum value of 12 g/L. Finally, arsenic ranges from 1 to 6 g/L, with an average value of 3 g/L. Regarding the oxidation states, according to [5], antimony is found in the electrolyte in Sb(III) and Sb(V) oxidation states in an approximate distribution of 50%, while arsenic is mostly in As(V) and bismuth is fully Bi(III). According to [6], this kind of resin has the capacity to retain 100% of Sb(III) and Bi(III), but not of Sb(V), whose retention and elution capacity is lower and is generally associated with a decrease in resin durability. Consequently, it is estimated that, in the eluate, antimony is distributed 75%-25% Sb(III)-Sb(V) while Bi is 100% in Bi(III) form. The presence of other metals mainly depends on the ore concentrate feed at the head of the process. The most common elements are usually identified as being Cu, Pb, Ni, and Fe. Recovery of Antimony by Hydrolysis in Side Streams from Copper Production 5 2.2. Background of recovery process It has been well described by [7] that one of the typical routes of industrial production of antimony is a hydrometallurgy method, in which antimony in the oxide form is leached in chloride medium, transforming it into chloride according to the reaction (2): Sb2O3 +6HCl → 2SbCl3 +3H2O (2) This method can be followed by electrowinning in diaphragm cells, in which antimony is finally obtained in a metallic form. Alternatively, the antimony chloride present in solution can be extracted through hydrolysis by adding OH- ions, thereby generating solid oxychlorides according to reactions (3) and (4): SbCl3 +H2O → SbOCl + 2HCl (3) 4SbCl3 + 5H2O → Sb4O5Cl2 +10HCl (4) The processes and reactions involved in the hydrolysis of metal chlorides are mainly dependent on the concentration of metal, pH, and chloride concentration [8]. Moreover, the presence of solid intermediate products converts antimony trichloride hydrolysis in a particularly complex process. Theoretical studies of the species contained in the eluate have been carried out to limit the study ranges of laboratory tests, and mainly refer to arsenic, bismuth, and antimony compounds. By gathering data from various sources, it is possible to generate predominance diagrams that can yield information on the stability ranges of the different species. Figure 2 shows the results for antimony, bismuth, and arsenic on nominal conditions of the eluate. As can be observed, the appearance of the solid phase of antimony occurs at pH 0.3. Bismuth, whose concentration in the eluate is of the same order as that of antimony and undergoes hydrolysis in an acid medium [9]. As can be observed in Regarding arsenic, in the form of As(V), it will not precipitate under any of the eluate conditions. According to the diagram and also to the reviewed literature [10]As(V) forms H3AsO4 and H2AsO42- depending on the pH. Based on these diagrams, the studied pH range has been selected from 0.25 and 1. 6 Figure 2: Speciation diagrams in the form of a molar fraction as a function of pH for a solution containing 4.7 mol/l HCl and (a) Sb (III)-Cl-H2O system 0.08 mol/L Sb, (b) Bi (III)-Cl-H2O system 0.04 mol/L Bi, (c) As (V)-Cl-H2O system 0.08 mol/L As. 3. Materials 3.1. Materials The eluate sample was obtained directly from the industrial process at Atlantic Copper. Due to the varying ore mix fed into the copper-production plant, the eluate composition may exhibit some degree of dispersion. However, using historical compositional data provided by Atlantic Copper, a range of values has been established for the research test conditions. Moreover, to study the effect of arsenic and bismuth, samples of synthetic eluates have also been prepared. Table 2 shows all compositions of the eluates samples used in the tests reported in this paper. The synthetics eluate samples were prepared by using chemical analytical-grade reagents, in particular liquid matrix, hydrochloric acid (37% wt%) provided by GlobalChem (Seville, Spain) and distilled water (conductivity<5 μS/cm) have been used. The addition of Sb, Bi and As has been carried out by dissolving Sb2O3 (Acros Organic, Waltham, E.E.U.U) Bi2O3 (Scharlab, Barcelona, Spain) and As2O5 (Thermofisher, Waltham, E.E.U.U) respectively. To maintain consistency and prevent any chemical changes, each synthetic solution was prepared with a volume of 100 mL and the same concentration of chlorides where the metals were added. The solutions were stirred until the powders were completely dissolved. Table2 presents the compositions of all synthetic eluates that were tested, including the process eluate (a) (b) (c) Recovery of Antimony by Hydrolysis in Side Streams from Copper Production 7 Table 2: Composition of the main components of eluates used in this study (g/L). 3.2. Methods The hydrolysis tests were carried out in glass flasks using 50 ml of solution. The flasks were placed in a thermostatic bath to maintain the reaction temperature within ±1 ºC, using continuous stirring. Distilled water with conductivity lower than 5 µS/cm was used to modify the reaction pH. Reaction time was 1 hour since the selected pH is reached. The pH and temperature were continually monitored using a pH-meter (Labprocess violab PH60. Barcelona Spain) with an accuracy pH of 0.01. The precipitated solid was vacuum filtered with 0.65-micron pore membrane filters and dried at 60 °C for 1 hour. X-ray diffraction (XRD) patterns of solid dry hydrolysed products were obtained in θ Chemical composition of hydrolysis dry products was carried out by means of X-ray fluorescence (XRF). Content of Sb, Bi and As in the liquid phase was measured by inductively coupled plasma optical emission spectroscopy (ICP). Ion chloride concentration of eluate samples was determined by titration with silver nitrate in an automatic tritator. Test have been conducted with pH values between 0.25-1, while discrete temperature values of 25, 30, and 35 have been chosen. In all cases, the chosen response has been the yield defined according to equation 4. 𝑆𝑏 𝑦𝑖𝑒𝑙𝑑 𝑒𝑥𝑡𝑟𝑎𝑐𝑡𝑖𝑜𝑛 % = 𝑆𝑏 𝑐𝑜𝑛𝑡𝑒𝑛𝑡 𝑖𝑛 ℎ𝑦𝑑𝑟𝑜𝑙𝑦𝑠𝑖𝑠 𝑝𝑟𝑜𝑑𝑢𝑐𝑡 (𝑔) 𝑆𝑏 𝑐𝑜𝑛𝑡𝑒𝑛𝑡 𝑖𝑛 𝑠𝑜𝑙𝑢𝑡𝑖𝑜𝑛 (𝑔) 𝑥100 (4) Element Sb Bi As Cl- Process eluate 8.9 9.5 2.3 170 Synthetic eluate Sb 10 Synthetic eluate SbBi 10 7.5 Synthetic eluate SbAs 10 3 Synthetic eluate SbBiAs 10 7.5 3 8 4. Results 4.1. Antimony extraction yield results The effect of pH and temperature has been studied with the aim of finding the conditions that maximize the antimony extraction yield. The isolated analysis of temperature shows no differences between the results obtained at 25 ºC and 35 °C; at 50 °C, the extraction yield decreased by 4%. Even though the maximum and minimum values overlap, the ANOVA analysis (p-value 0.008) and a subsequent TUKEY test confirm that the results show a significant difference in the case of the tests carried out at 50 °C with respect to those of 35 °C, with no significance in the results at the rest of the temperatures tested. Similar results are obtained in [11], where the decrease in extraction yield is attributed to the effect of temperature on the evaporation of the reagent and on the solubility of the hydrolysis products. Regarding the influence of pH in antimony extraction yield has two well-delim- ited trends. For pH values between 0.25 and 0.7, the extraction yield is high, ranging between 80% and 94%, with a slight increase as pH increases. On the other hand, for pH values between 0.7 and 1, the trend is inverted, and the antimony extraction yield strongly decreases with an increasing pH in a linear trend (These results are reflected in a paper that is currently undergoing peer review in Hydrometallurgy Journal). The effect of both variables has been studied by using the response surface method in the pH range of 0.25 to 0.7. The data was fitted to a quadratic model, and the statistical significance was evaluated using the F test and ANOVA, which produced an F value of 20.69 and an adequate precision of 14.650. As depicted in Figure 2, the model's predictions are similar to those obtained from analysing the individual variables. That is, the antimony extraction yield increases as pH rises and temperature enhances the yield until it reaches a maximum at 35 °C, after which it decreases. The ANOVA analysis reveals that temperature does not have a significant impact on antimony extraction yield in the model. When the model is employed to determine the optimal operating conditions that maximize performance, pH 0.645 and 34 ºC produce the best results, achieving 99.4% yield. The following highest yield point is at 25 ºC and pH 0.56, with an antimony extraction yield of 93.4%, which is very close to the optimal conditions. Recovery of Antimony by Hydrolysis in Side Streams from Copper Production 9 Figure 2: Response surface obtained from pH from (0.25 to 0.7) and temperature (25ºC to 50ºC). Synthetic eluates with the same antimony content as the process eluate but without arsenic and bismuth have been also tested. The tests were performed within the optimal pH range. As shown in Figure 3, the trend in extraction yield with respect to pH is similar in tests conducted on both eluates, with an increase in yield observed as pH increases. However, there are differences in the extraction yields obtained. The average differences range from 5% in the pH 0.5 to 0.6 group, to 9% in cases where the differences are close to 9%.