Enhanced Separation of Sulfur and Metals from Polymetallic Sulfur Slag through Recrystallizing Regulation of Sulfur Crystals

Abstract

Elemental sulfur is an important non-metallic material that is widely used in various fields including chemical, metallurgical and sulfuric acid industries. Polymetallic sulfur slag (PSS) is an important secondary resource for the recovery of sulfur. However, separation of sulfur from PSS is difficult due to the tight binding of some thiophilic metals to sulfur. Herein, we proposed a recrystallization method for the effective separation of sulfur and metals by using organic solvents to control the particle size of sulfur crystals. It was suggested that the particle size of sulfur recovered in the close to saturation state of toluene is about 2000 μm, while the purity is as high as 99.6%. Moreover, the recovery rate is over 95%, which is more than the current commonly used flotation method. The growth mechanism of sulfur crystals under different saturation conditions lays a foundation for the deepening of the growth theory of large-size sulfur crystals. This method realized the effective separation of sulfur from metal sulfides such as FeS₂ and ZnS, which provided an important guidance for the recovery of sulfur resources through a novel method of recrystallizing regulation.

1. Introduction

Sulfur is an important industrial raw material that is widely used in chemical, metallurgy and other industries [1]. Although China has abundant sulfur reserves, its production of sulfur is relatively low. A large amount of sulfur needs to be imported every year. However, the chemical smelting industry produces a large amount of sulfur-containing waste (the annual production is about 7.395 million tons). Among them, the zinc smelting leaching project in China can produce about 600,000 tons of polymetallic sulfur slag (PSS) per year, with a sulfur content exceeding 70% [2,3]. These sulfur-containing wastes causes serious environmental pollution and are a waste of large amounts of sulfur resources. Therefore, sulfur recovery from waste, especially from PSS, is a necessary way to solve the shortage of sulfur resources.

Wet leaching technology has been considered an effective method to enrich sulfur into waste residue for further extraction and recovery in recent years [4], which allows the comprehensive use of sulfur resources without producing SO₂ pollution. However, due to the influence of various metal impurities, the recovery rate and purity of sulfur are quite low (only 80% and 95%, respectively), which is the main problem to be solved in the sulfur recovery process [5]. The key to this problem is that sulfur exists in the form of fine particles, which is easy to adsorb metal impurities. Therefore, sulfur recovered from PSS needs to be recrystallized into large particles with smooth surfaces to reduce binding with metals. It is worth noting that sulfur is easily dissolved in specific solvents at high temperatures, while the recrystallization process occurs during cooling to room temperature [6]. Therefore, recrystallization is expected to be an effective technique for sulfur separation and purification. Shevchenko et al. [7] demonstrated for the first time the synthesis of sulfur nanoparticles with a diameter of 10–20 nm in a H₂O-DMSO (DMSO means dimethyl sulfoxide) dispersion system by using a microemulsion method, revealing that the size of sulfur nanoparticles depends on the concentration of the cosolvent. In addition, Shamsipur et al. [8] electrochemically synthesized sulfur nanoparticles by electrolysis of sodium thiosulfate solution, which suggested that the particle size of sulfur nanoparticles can be adjusted by changing the initial concentration of thiosulfate. However, most studies focus on the crystallization regulation of small-sized sulfur, which is not conducive to the recovery of high-purity sulfur from PSS.

Actually, the crystallization process mainly includes the formation of oversaturated solutions, the appearance of crystal nuclei, the growth of crystals and recrystallization (or aggregation) [9,10]. There are many factors affecting these processes, including solvent [10,11,12], oversaturation [13,14,15], temperature [16], additives [17,18,19], impurities [20,21,22] and external force fields (microwave, ultrasound, magnetic field, etc.) [23]. Among them, controlling the crystallization process by regulating oversaturation is the most widely used. Tsukamoto [24] analyzed the crystal growth mechanism by measuring the growth rate and supersaturation. It was found that the crystal grows spirally at low oversaturation, while the crystal grows in 2D nucleation mode at critical oversaturation. When the oversaturation increases, the crystal grows adhesion on the rough surface. The crystallization of sulfur in specific solvents undergoes a similar process. Sulfur is a non-polar S₈ molecule with π-electrons. According to the principle of similar miscibility, the non-polar characteristics of solvent molecules are expected to promote the dissolution of sulfur [25,26]. Thus, non-polar organic solvents with high-density aromatic π electrons, such as toluene and p-dimethylbenzene, are considered to be good solvents for sulfur. Then, the particle sizes and surface properties can be regulated during the recrystallization of sulfur. This method is expected to achieve the separation of sulfur and metal impurities in PSS to improve the purity of sulfur.

In this work, we regulated the recrystallization growth of sulfur in organic solvents by controlling its saturation. Firstly, the effect of sulfur recovery from PSS by recrystallization with organic solvents was investigated. Then, the corresponding morphology and phase changes of sulfur crystals under different saturation conditions were studied by optical microscopy, scanning electron microscopy, X-ray diffractometer and Raman spectrometer, in which the possible growth mechanism was revealed. Finally, the recovery mechanism of sulfur in PSS was explained by phase analysis. The findings are of great significance for the recovery and application of sulfur resources from solid waste through hydrometallurgical leaching.

2. Experimental

2.1. Materials

Sublimated sulfur was analytically pure (AR) and purchased from Aladdin Chemistry Co., Ltd. in China. Toluene (AR) was purchased from Sinopharm Chemical Reagent Co., Ltd. in Beijing. Tetralin, tetrachloroethylene and styrene were all analytically pure and obtained from Macklin. Polymetallic sulfur slag (PSS) came from China Zhongjin Lingnan Danxia Smelter.

2.2. Experimental Methods

2.2.1. Determination of Solubility

In this paper, the solubility of sulfur in organic solvents was determined by the crystallization precipitation method. In a certain volume of solvent, a certain amount of solute was dissolved to make it an unsaturated solution. When the solution slowly cooled down and began to precipitate crystals (the solution became saturated), the temperature of the solution was measured at the same time and the solubility at this temperature can be calculated.

2.2.2. Regulation of the Growth Process of Sulfur Crystals

A certain mass of sublimated sulfur was mixed with 10 mL of organic solvent (toluene), heated and dissolved at a certain temperature (80–120 °C) for 1 h and naturally cooled to room temperature to observe the recrystallization of sulfur crystals.

2.2.3. Recovery of Sulfur from Polymetallic Sulfur slag (PSS) by Organic Solvent Method

A certain mass of PSS was mixed with 10 mL of organic solvent (toluene), heated and dissolved at 120 °C for 1 h. The thermal filtration and non-filtration treatment were carried out, respectively. The thermal filtration treatment was to achieve solid–liquid separation through the thermal filter funnel at high temperature. The non-filtration treatment was to mix sulfur and slag together without separation treatment when the solution was naturally cooled and crystallized. Finally, the sulfur samples were recovered by ethanol washing and drying at 60 °C.

2.2.4. CS₂ Dissolution Test

The dissolution test of CS₂ was mainly carried out according to the national standard method. Firstly, 0.5 g~1 g samples were weighed, accurate to 0.0001 g, and placed in a glass sand core crucible dried to a constant amount. Secondly, the glass sand core crucible containing samples was installed on the filter bottle in the ventilation cupboard, and an appropriate amount of carbon disulfide was added to the crucible with a dropper. The sulfur was dissolved by stirring with a glass rod, and the vacuum pump was opened at a suitable rate for suction filtration. After most of the sulfur was dissolved, the crucible wall and its bottom were washed with carbon disulfide for suction filtration. Then, the glass sand core crucible containing the residue was moved into a constant temperature drying oven at 60 °C for 24 h. Finally, the glass sand core crucible was taken out, cooled to room temperature and weighed accurately to 0.0001 g. According to the above operation, the carbon disulfide treatment was repeated until the difference between the two consecutive weighings was not more than 0.0003 g.

2.3. Characterization Methods

The phase composition of the samples was analyzed by a Bruker D8 Advance powder X-ray diffractometer (XRD) (Cu Kα, λ = 0.15406 nm; operating voltage 40 kV, current 40 mA, scanning range 10~80°); the structure of the sample was analyzed by a Renishaw laser confocal Raman spectrometer (Raman) (model: INVIA, spectral range: 200 nm–1000 nm, spectral resolution: 1 cm⁻¹, laser: 532 nm); the microstructure of the samples was analyzed by a Japan Electronics JSM-IT300 scanning electron microscope (SEM), Japan. An OLYMPUS Inverted Microscope IX73 was used to observe the sulfur crystal growth process.

3. Results and Discussions

3.1. Separation of Sulfur from Polymetallic Sulfur Slag by Recrystallization

Sulfur is readily soluble in non-polar solvents containing functional groups such as benzene ring, C=C bond and chlorine-containing groups. Therefore, organic solvents containing these groups in their molecule structure, including toluene, tetrachloroethylene, styrene and tetrahydronaphthalene, were chosen as sulfur dissolving solvents. The solubility of sulfur in these organic solvents was determined by the crystalline precipitation method. As shown in Figure 1a, the solubility of sulfur in all of the four solvents was greatly affected by temperature, which increased sharply after 100 °C. Among them, solubility changed in toluene under different temperatures exhibited the most obvious difference. Although tetrachloroethylene dissolves more sulfur at low temperatures, its volatility is stronger than that of toluene. Moreover, toluene itself is a by-product of the coking plant and raw materials are easily available. Therefore, toluene was selected as the solvent of elemental sulfur. The subsequent investigation of a sulfur-related growth mechanism was mainly carried out with toluene as the solvent.

In order to reduce the loss of sulfur resources in the zinc smelting industry, the typical PSS formed in the oxygen pressure leaching process of zinc hydrometallurgy was selected as the research object. The XRD results (Figure 1b) show that the main phases of PSS included ZnS, FeS₂, PbSO₄, CaSO₄, aluminosilicate, etc.. Refinement of XRD indicated that the sulfur content in PSS was about 57%, which is similar to the results obtained by the CS₂ dissolution test (59%). Then, the PSS was dissolved in toluene with different saturation. The sulfur recovery effect via thermal filtration and non-filtration was studied, respectively. The corresponding purity and recovery rates of recovered sulfur are shown in Figure 1c,d. It can be seen from Figure 1c that the purity of sulfur recovered by thermal filtration is higher than that of sulfur recovered without filtration, and the purity can reach industrial grade (99%). At the same time, it was found that the purity of sulfur recovered from PSS by thermal filtration was relatively higher at close to saturation state (0.3 g S/10 mL Toluene) than in other states. It can be seen from Figure 1d that the recovery rate of sulfur recovered by thermal filtration was relatively low (about 95–97%), indicating that most of the sulfur had been completely dissolved and only a very small amount of sulfur was not effectively recovered. The recovery rate of sulfur was only about 90–94% without filtration, which may be due to the incomplete separation of partially crystallized sulfur. At the same time, it was found that the recovery rate of sulfur was relatively higher in the slightly oversaturated state (0.4 g S/10 mL toluene) than in other states, which could reach more than 94%. The next highest recovery rate of sulfur was found in the close to saturation state (more than 92%). In summary, the analysis shows that the purities of sulfur recovered in the close to saturation state after thermal filtration were higher than those recovered in other states. The recovery rate of sulfur recovered in the close to saturation state after thermal filtration was close to the sulfur recovery rate under other states. This indicates that thermal filtration after regulating sulfur recrystallization in the close to saturation state is a suitable method for sulfur recovery from PSS.

3.2. Crystallization Kinetics and Mechanism of Sulfur in Toluene

3.2.1. Crystallization Kinetics Analysis

Figure 2a shows the actual samples of sulfur with different saturations in toluene and their corresponding scanning electron microscope (SEM) images. It can be seen that the growth morphology of sulfur is different under different saturation conditions, with particle sizes all above the micron level, even the millimeter level. The growth size of sulfur crystal in this paper is given according to the longitudinal sizes. The samples synthesized in the unsaturated state (0.2 g S/10 mL Toluene) were mainly irregularly shaped, with the average size about 68 μm, which may be caused by obvious defects in the growth process. Samples synthesized in the close to saturation state (0.3 g S/10 mL Toluene) were mainly prismatic, with the average size around 2000 μm. In comparison, samples synthesized in the slightly oversaturated state (0.4 g S/10 mL Toluene) had a long prismatic shape that extended along the axis and diffused to both sides, while the average size of the obtained sample was around 1500 µm. There were also some longer sizes that exceeded the measurable range of SEM, with dimensions up to about 2500 µm according to the measurements of actual samples. Samples synthesized in the oversaturated state (0.5 g S/10 mL Toluene) were mainly sheet-like. The sheet-like sample was spliced together by the prism in a certain order, which was the size in the millimeter level. Due to the limited range of SEM measurements, only a part of the morphology is displayed in the diagram, with a particle size of about 600 μm. The contact area between sulfur crystals of different morphology and size and metal impurities is different. Small size crystals tend to adsorb impurities, while large size crystals have a weaker binding effect with impurities [27]. However, when the size reaches a certain level, the combination with impurities will be reduced during the recrystallization process, which also explains why the recovery rate and purity of sulfur crystals recrystallized from PSS in a close to saturation state are high.

In order to further analyze the recrystallization process of sulfur from toluene in PSS, the recrystallization process of pure sulfur system in toluene with different saturation was investigated by inverted optical microscope. As shown in Figure 2b, the samples close to saturation state begin to appear to have prism-like crystals and grow rapidly with hard texture after natural cooling for about 25 min. Samples in the slightly oversaturated state begin to appear to have crystals after natural cooling for about 16 min, while most of them grow in a needle-like structure along a certain direction. The crystals of the oversaturated sample start to appear after natural cooling for about 10 min, which are mainly grown in a sheet-like structure with slightly irregular shape influenced by different local concentrations. Although the sheets are spliced together by the prisms in a certain order, the texture is brittle, which is consistent with the phenomena observed in the actual samples and SEM. Therefore, according to the growth size of sulfur crystals in organic systems with different saturations, it can be found that the growth rate is much larger than the nucleation rate when the solution is close to saturation, while the nucleation rate may be much higher than the growth rate in the oversaturated state. According to the comparison of the size of sulfur crystals in the close to saturation state and oversaturated state, it can be found that the nucleation rate and growth rate are dependent on the oversaturated state of the solution. In summary, it can be seen that, as the solution saturation gradually increased, the distance and orientation between the grains attracted or repelled each other. This caused the grains connected together along one crystal plane and the relative growth rate to change. It affected the crystal growth morphology and formed sulfur crystals with different morphologies.

The growth kinetic process of sulfur crystals was further analyzed through inverted optical microscope and SEM. Growth rate curves were plotted according to the growth rates of sulfur in toluene with different saturations. Because the growth rate of the oversaturated state is fast, it is difficult to measure its growth rate quantitatively. Therefore, only the growth rate curves of samples in the close to saturation and slightly oversaturated state are given in Figure 2c. It can be seen from the figure that the growth process of the sample close to saturation state was relatively slow. The average growth rate was above 100 μm/min in the first 10 min, decreased to about 37 μm/min for 10–30 min and decreased to 24 μm/min after 30 min. The growth size tended to be stable until about 63 min with a value of 2200 µm, and it was larger than that of the slightly oversaturated samples (1900 µm). It can be also seen from the blue line of Figure 2c that the growth process of the slightly oversaturated sample was relatively fast. The average growth rate was above 75 μm/min in the first 7 min, decreased to about 43 μm/min for 7–35 min and decreased to 30 μm/min after 35 min. With the increase in growth time, the growth rate gradually decreased until the average size of the samples in the slightly oversaturated state stabilized around 43 min. It indicates that the average particle size of the sample close to saturation state was larger than that of the sample in the oversaturated state. It is speculated that the growth rate is larger than the nucleation rate at low saturation. The nucleation rate was relatively larger than the growth rate as the saturation increased; therefore, the average particle size of the sulfur crystals crystallized in the oversaturated state was smaller. The slow growth rate leads to the formation of small-sized crystals that tend to adsorb impurities on the surface during the growth process, inhibiting further crystal growth. This in turn reduces the purity and recovery of sulfur crystals.

3.2.2. Phase Analysis

In order to explore the microstructure changes of sulfur crystals after recrystallization in toluene with different saturations, the phase and structure were studied by XRD and Raman. It can be seen from XRD results (Figure 3a) that the obtained samples belong to orthorhombic elemental sulfur (PDF No. 41-1488, a = 10.450 Å, b = 12.840 Å, c = 24.460 Å). With the increase in saturation, XRD peak intensity gradually decreased. The most obvious changes were at 2θ = 15.37°, 23.08°, 25.88° and 28.68°, corresponding to crystal planes (113), (222), (026) and (313), respectively. It indicates that the crystallinity of sulfur crystals decreased gradually and the particle size decreased gradually. It resulted in the growth of crystals into fine crystals, which is consistent with the phenomenon observed in SEM. At the same time, combined with the analysis of the results in Figure 1c and Figure 3a, it shows that the formation of crystals with high crystallinity and large grain size is more conducive to the recovery of sulfur. Raman results (Figure 3b) show that the peak intensity of sulfur after the recrystallization was weaker than that before the recrystallization. The peak intensity gradually decreased with the increase in saturation, which may be related to the decrease in the crystallinity of the samples. This result is similar to the XRD results. At the same time, the positions of the characteristic peaks of the sulfur crystals before and after the recrystallization changed as follows. When the displacement was 472.17 cm⁻¹, 438.64 cm⁻¹ and 246.10 cm⁻¹, the positions of the characteristic peaks belonged to the symmetric stretching vibration [28,29] and symmetric bending vibration of the S-S bond [30], respectively. The displacement of 217.41 cm⁻¹ and 152.04 cm⁻¹ corresponded to the symmetric bending vibration and asymmetric bending vibration of the S-S bond, respectively [31,32]. However, the characteristic peaks of samples after the recrystallization showed shifts to about 219.33 cm⁻¹ and 153.96 cm⁻¹, which may be affected by the crystal field effect of prismatic sulfur. The appearance of the wave number 184.79 cm⁻¹ may be due to the anti-resonance phenomenon caused by the ring structure of S₈, while all belonged to the external vibration of the S-S bond below the low frequency wave number 100 cm⁻¹ [33]. In conclusion, the microstructure of samples changes slightly before and after the recrystallization. It shows that recrystallization does not affect its phase change, and is an effective method for separating and recovering sulfur in PSS.

3.2.3. Growth Mechanism of Sulfur Crystals

In order to deeply understand the above growth process, the growth mechanism of sulfur in toluene with different saturations was further analyzed. According to Figure 4a, it can be seen that the elemental sulfur first dissolved in the solvent to produce sulfur clusters. Then, it was cooled to form small particles of sulfur with high specific surface areas, and solvent molecules were adsorbed on the surface of the sulfur particles. In the unsaturated state and the close to saturation state, the steric hindrance effect of solvent was stronger than the directional aggregation effect between sulfur particles. Therefore, a single prismatic sulfur particle formed. In contrast, the directional aggregation effect of sulfur crystals was stronger than the steric hindrance effect in the slightly oversaturated state and the oversaturated state. Consequently, the sheet shape formed by the directional aggregation of prisms.

According to the above analysis, the relationship between the nucleation rate and the growth rate of sulfur under different saturation is concluded in Figure 4b. In Section 1, the morphology of sulfur in an unsaturated state was irregular, with generally small particle size, indicating that the nucleation rate (blue line) and growth rate (red line) of sulfur crystals were both low at this time. However, the crystal growth was low, so the growth rate may be greater than the nucleation rate. In Section 2, the morphology of sulfur crystals close to saturation state was regular prismatic with a large single particle, indicating that the growth rate of sulfur crystals was greater than the nucleation rate. The crystal growth was faster than that of sulfur crystals in the unsaturated state, indicating that the nucleation rate of crystals in the close to saturation state was higher than that in the unsaturated state. In Section 3, the morphology of sulfur crystals in the slightly oversaturated state was similar to the needle-like structure spliced together by a single prism along a certain direction. The average size of a single prism was smaller than that of the sample in the close to saturation state, indicating that the samples in the slightly oversaturated state had a relatively fast growth rate. In Section 4 of the oversaturated state, the morphology of sulfur crystals was similar to the sheet spliced together by multiple prisms in parallel along a certain direction. The average size of a single prism was smaller than that of samples in the slightly oversaturated state, indicating that the growth rate of the samples in the oversaturated state was much lower than the nucleation rate. Therefore, it can be concluded that to obtain sulfur with high crystallinity and large size, the solution should be controlled in the close to saturation state. The formed high crystallinity and large size sulfur crystals are more conducive to the high purity recovery of sulfur in PSS.

3.3. Recovery Mechanism of Sulfur in Polymetallic Sulfur Slag

Sulfur crystals with different morphology can be crystallized from toluene with different saturation, which may have different binding effects with metals. As mentioned above, this method can be used to guide the high-purity recovery of sulfur in PSS. The phase types affecting the purity of recovered sulfur were analyzed by XRD. The corresponding results are shown in Figure 5. Under the thermal filtration condition (Figure 5a), the recovered sulfur was pure yellow in appearance (illustration), and impurities were less. The main peaks of XRD were coincident with the main peaks of S₈ (PDF 074-1465), indicating that the main phase of the recovered sulfur was elemental sulfur. In addition, there were a few miscellaneous peaks, such as 2θ = 33.1, 47.7 and 51.9°, which corresponded to the (200), (116) and (008) crystal planes of FeS₂, ZnS and HgS, respectively. It was proved that the existence of these trace impurities seriously affected the high purity recovery of sulfur. However, under the condition of non-filtration (Figure 5b), the recovered sulfur was grayish-yellow in appearance (illustration), indicating more impurities than those of thermal filtration. XRD results also show similar results, with the recovered sulfur under the thermal filtration condition. In addition to the main peak of elemental sulfur, diffraction peaks of trace impurities such as ZnS, FeS₂, HgS and Ag₂S also existed. Compared with the original slag, it can be clearly seen that most of the sulfate and aluminosilicate impurities were almost removed from the recrystallized sulfur, so the purity of recovered sulfur can be improved. In summary, it can be seen that the key phases affecting the recovery purity are mainly sulfur-friendly metal sulfides such as ZnS, FeS₂, HgS and Ag₂S. Under the condition of thermal filtration, the sulfur recovered from the close to saturation state may be weakly bound to these sulfur-friendly metal sulfides, so the purity of recovered sulfur is relatively high.

4. Conclusions and Outlook

This work proposed a novel method to control the recovery rate and purity of sulfur from PSS through the recrystallization process. Sulfur and metals from PSS were separated via recrystallizing in toluene to achieve high-purity recovery of sulfur. The growth mechanism of sulfur with different saturation in toluene suggested a rapid growth of sulfur crystals in organic solvents. Such rapid growth had a self-exclusion effect, which contributed to the high-purity recovery of elemental sulfur in PSS under different saturation states. The experimental results show that the prismatic sulfur crystals of about 2000 μm were obtained by recrystallization of solvent toluene at the close to saturation state (0.3 g S/10 mL toluene). Under this condition, sulfur with a purity of about 99.6% can be recovered from polymetallic sulfur slag by thermal filtration, and the recovery rate can reach more than 95%. It realized the effective separation of sulfur from metal sulfides such as FeS₂ and ZnS. Moreover, the analysis of the physical phases before and after recrystallization revealed that the key factor affecting the purity was the strong binding effect of thiophilic metal sulfides. This result has important guiding significance for the high purity recovery and wide application of sulfur.