Shear-Force Based Stainless Steel Slag Modification for Chromium Immobilization

Home , Gravity separation, Ostwald ripening

ISIJ International, Vol. 59 (2019),ISIJ International,No. 3 Vol. 59 (2019), No. 3, pp. 583–589

Qing ZHAO,1)* Chengjun LIU,1) Longhu CAO,1) Maofa JIANG,1) Baokuan LI,2) Henrik SAXÉN3) and Ron ZEVENHOVEN3)

1) Key Laboratory for Ecological Metallurgy of Multimetallic Mineral (Ministry of Education), School of Metallurgy, Northeastern University, Shenyang, 110819 China. 2) School of Metallurgy, Northeastern University, Shenyang, 110819 China. 3) Thermal and Flow Engineering Laboratory, Åbo Akademi University, Åbo/Turku, 20500 Finland. (Received on November 23, 2017; accepted on November 5, 2018; J-STAGE Advance published date: December 20, 2018)

Immobilization of chromium in a stable spinel by modification is a powerful way to prevent chromium pollution of stainless steel (SS) slags. The precipitated spinel grain size is usually smaller than 30 μm, however, which limits the effectiveness of the modification. In the literature, very few efforts have been reported on promotion of spinel growth rate by optimizing the dynamic conditions. In this study, the effects of shear force on the spinel grain size evolution under isothermal conditions and during cooling were investigated. The experimental results indicate that the employment of shear force significantly changes the growth behavior of spinel at 1 500°C. The growth mechanism of spinel was studied by using crystal size distribution (CSD) theory, showing different regimes of supply-controlled Ostwald ripening, surface-controlled growth with decaying nucleation rate, and constant rate nucleation and growth at shear rates of 0 s −1, 10.83 s −1, and 21.67 s −1, respectively. On the other hand, at a cooling rate of 5°C·min −1, the shear force was found to have little effect on the crystallization behavior of spinel. The results of static leaching tests indicated that hardly any chromium had leached, which makes the modified SS slag more environmentally friendly when used as a raw material.

KEY WORDS: stainless steel slag; spinel; crystal growth; modification; shear force; chromium immobilization.

ration impossible.5) Acid leaching or biological methods that 1. Introduction have often been used to extract chromium from Cr-bearing China has for more than a decade been the main producer residue are not suitable for the SS slag, since most of the of stainless steel and today stands for more than a half of acid is consumed by the Cr-free phases.6,7) Researchers the global production. Approximately one ton of Cr-bearing have reported that chromium can exist in a number of min- stainless steel (SS) slag is generated when three tons of eralogical phases, and the existence of unstable phases like stainless steel are produced.1) Chromium in initial (i.e. dicalcium silicate, merwinite, periclase, and lime enhance before cooling) SS slag is generally bivalent or trivalent. the possible leaching of chromium elution into seawater or Cr2+ is unstable and may be easily oxidized to Cr3+ or Cr6+ acidified water bodies.8,9) However, a spinel is considered during the cooling process or during the storage period. Cr3+ to be a suitable stable phase for controlling the leaching and Cr6+ have different toxicities: Cr3+ is an essential nutri- of chromium from SS slag.20) The results of earlier work ment at small amounts, while Cr6+ is very toxic and may generally agree on that the activity of chromium is strongly be easily released which can lead to severe environmental decreased when bound in a spinel because of the strong pollution or health risk issues.2) Therefore, a treatment of bonding of chromium in that.11–13) Therefore, the leaching the SS slag that converts or stabilizes harmful components of chromium from SS slag is expected to be limited or even is essential prior to the disposal of the slag in a landfill or negligible if all of the chromium would exist in a stable possible application in other processes.3,4) spinel phase. Chromium may be partially recovered from the SS slag Adding spinel-forming agents, such as MgO and Al2O3, by magnetic separation or gravity separation, but the extrac- may promote Cr-containing spinel crystallization.14,15) After tion efficiency is unsatisfactory since some chromium exists such a slag composition adjustment, the grain size of the in the silicate phases making recovery by mechanical sepa- spinel mainly depends on the temperature to which it is exposed and the duration of the exposure.16) Controlling the * Corresponding author: E-mail: [email protected] cooling conditions can affect the mineral transformation of DOI: https://doi.org/10.2355/isijinternational.ISIJINT-2017-678 slag and the growth behavior of spinel, although it is dif-

583 © 2019 ISIJ ISIJ International, Vol. 59 (2019), No. 3 ficult to realize the growth of spinel to a grain size above was injected into the furnace at a rate of 500 mL·min −1 to 30 μm.17,18) Until today, very few efforts have been made prevent Cr6+ from forming during the modification process. to optimize the dynamic conditions of spinel growth via a The temperature was first elevated to 1 600°C for sample modification process. pre-melting, and then decreased to 1 500°C for shear-force Chromite with a common formula of (Mg,Fe2 + ) based modification. When the temperature had decreased to (Cr,Al,Fe3+) is the most important Cr-rich mineral world- the set point, the molybdenum agitator was moved down wide, and has a crystal structure and chemical composition until it was fully immersed in the molten slag, followed by similar to the spinel in SS slag.19) Furthermore, chromite stirring and timing. Different stirring rates of 0 rpm (refer- normally crystallizes as an early phase from mafic magmas, ence experiment), 25 rpm, 50 rpm, 75 rpm and 100 rpm which is similar to the crystallization process of spinel from were investigated in this study. The mean shear rate (γ, s −1) molten SS slag.20) Single ore bodies of chromite may hold is used to characterize the shear force exerted on the sample, a few million tons, and the geological shearing force has which is defined as been reported to play a significant role in the chromite min- KN ...... (1) eralization and growth processes by improving the kinetic conditions of crystallization.21,22) Inspired by these research where K is a constant which should be K =13 r −1 for the findings in chromite mineralogy, it is innovatively proposed impeller,23) N is the stirring rate as revolutions per second. in this paper that exerting a shear force during melting mod- Thus, the corresponding shear rates studied were 0 s −1, ification of SS slag may improve the chromium stabilization 5.42 s −1, 10.83 s −1, 16.25 s −1 and 21.67 s −1, respectively. and promote the spinel growth process. Cooling treatment was carried out at a rate of 5°C·min −1 In the present study, a shear force was applied to tradi- after 60 min of isothermal treatment at 1 500°C. The agita- tional modification process of SS slag, and the effects of tion was stopped and the molybdenum agitator was taken the shearing rate on the grain size evolution of spinel under out of the sample when the temperature had decreased to isothermal as well as cooling process conditions were inves- 1 300°C. Sampling was performed after 0 min, 15 min, tigated. Furthermore, the growth mechanism of spinel under 30 min, 45 min and 60 min during the isothermal process various shear force conditions was studied using crystal size and at 1 400°C, 1 350°C, 1 300°C, 1 250°C, 1 200°C and distribution (CSD) theory. A static leaching test at labora- 1 150°C during the cooling process using a quartz tube, and tory scale was carried out to evaluate the leaching property the obtained samples were immediately quenched in water. of chromium from such treated SS slag. Below 1 150°C, SS slag rapidly solidified and was cooled to room temperature in a furnace. The experimental apparatus used in this study is shown in Fig. 1. 2. Material and Methods CSD theory for particulate processes in chemical engi- 2.1. Materials neering24) is widely used in investigations on the growth Reagents of calcium oxide (CaO ≥ 98.0 wt.%), magne- sium oxide (MgO ≥ 99.9 wt.%), silicon oxide (SiO2 ≥ 99.0 wt.%), ferrous oxalate (FeC2O4·2H2O ≥ 99.9 wt.%), aluminum oxide (Al2O3 ≥ 99.0 wt.%), chromium oxide (Cr2O3 ≥ 99.0 wt.%), and calcium fluoride (CaF2 ≥ 98.5 wt.%) were used to prepare a synthetic SS slag, in which some aluminum oxide was employed as a spinel-forming agent based on earlier findings.13) The content of aluminum oxide is normally below 10 wt.%, but it was increased to 12 wt.% in this study. The raw materials were dried at 110°C for 10 h before weighting and mixing. The chemical composition of the target SS slag studied in this work is shown in Table 1. All reagents were purchased from Sinopharm Chemical Reagent Co., Ltd., China.

2.2. Methods A tube furnace was modified by equipping it with a shear force input device (i.e., a system containing a molybdenum agitator and control components), and employed for modification of SS slag. A sample of 140 g was poured into a molybdenum crucible, which was then placed into a graphite crucible in the modified tube furnace. Argon

Table 1. Chemical composition of SS slag studied in this work, Fig. 1. Illustration of the experimental apparatus used for modifi- wt.%. cation process. 1-cap; 2-furnace cover; 3-corundum tube; 4-insulation layer; 5-thermocouple A; 6-molybdenum agi- CaO SiO2 MgO FeO Al2O3 Cr2O3 CaF2 tator; 7-stainless steel slag; 8-refractory; 9-gas inlet; 10-thermocouple B; 11-graphite crucible; 12-molybdenum 40.8 27.2 9.0 3.0 12.0 5.0 3.0 crucible; 13-MoSi2 heating elements; 14-gas outlet.

© 2019 ISIJ 584 ISIJ International, Vol. 59 (2019), No. 3 history of mineral phases, including illite,25) shergottite,25) 0.01 mg·L −1. pyroxene,25) plagioclase,28) and volcanic lava,29) provid- ing valuable information about the complex interaction of 3. Results and Discussion nucleation and growth. Different shapes of the CSD curve of minerals imply different crystal growth mechanisms. In 3.1. Grain Size Evolution this work, CSD analysis was adopted to study the effects To investigate the effects of shear force on the growth of shear force on the spinel growth behavior during the behavior of spinel during the modification process, the modification process. Metallographic microscope analysis spinel grain size was measured and the mean diameter was was performed, and the mean diameter (D) of the spinel was calculated. Figure 2 shows the variation of mean diameter determined by analyzing spinel in sixty-four different fields with modification duration at various shear rates of 0 s −1, of view using Image-Pro Plus 6.0 software. The number- 5.42 s −1, 10.83 s −1, 16.25 s −1 and 21.67 s −1 at 1 500°C. It based mean diameter (D) was calculated as was found that the spinel grew slowly when no shear force was applied on the molten slag: the mean diameter had 1 n increased by only about 1 μm after 60 min of modification. D B Di ...... (2) n i 1 The employment of shear force significantly changed the where n is the number of spinel crystals in all fields of view th (-), and Di is the diameter of the i spinel (μm). Further- more, several samples were polished using various types of abrasive paper to get a smooth flat surface. Finally, gold spraying was applied on the flat surfaces to give sufficient conductivity for scanning electron microscopy-energy- dispersive X-ray spectroscopy (SEM-EDS) analysis. In order to evaluate the leaching property of chromium from the modified SS slag, a powdered sample was prepared by grinding and screening, and used for static a leaching test at laboratory scale. A sample (5 g) of slag was poured into 50 mL acid solution (with a mass ratio of sulfuric acid and nitric acid of 2:1) at a controlled pH value of 3.2, and the leaching test was conducted according to the Standard of Environmental Protection Industry of People’s Republic of China HJ/T 299-2007. After 18 h, the slurry was filtrated and a leachate was obtained. The leached quantity of chromium was determined using induced coupled plasma-optical Fig. 2. Variation of mean diameter of spinel with modification duration in the isothermal process (1 500°C) at shear rates emission spectroscopy (ICP-OES) with a detection limit of of 0 s −1, 5.42 s −1, 10.83 s −1, 16.25 s −1 and 21.67 s −1.

Fig. 3. SEM images of SS slag modified at shear rates of 0 s −1, 10.83 s −1 and 21.67 s −1.

585 © 2019 ISIJ ISIJ International, Vol. 59 (2019), No. 3 spinel growth behavior, with different outcomes for different shear rates. The mean diameter increased from 7.4 μm to 9.3 μm when the shear rate was 5.42 s −1, and rose to 11.7 μm when the shear rate was increased to 10.83 s −1. However, shear rates beyond 10.83 s −1 did not further improve the growth rate. By contrast, the growth was restricted when the shear rate reached 21.67 s −1. SEM images of SS slag modified at shear rates of 0 s −1, 10.83 s −1 and 21.67 s −1 are given in Fig. 3. Furthermore, the spinel grain size distribution was studied by SEM analysis, with results shown in Fig. 4. It can be seen that the grain size increased with process duration when no shear force was exerted on the molten slag. About 80% of the spinel grains are in the size range of 4–16 μm after 60 min of modification. When a shear force was employed at the rate of 10.83 s −1, the spinel grain size increased more quickly and some large crystals (with size around 20 μm) formed in the slag. In contrast, no notable change in grain size was found when the shear rate applied was 21.67 s −1: After 60 min, more than 90% of the spinel grains were still smaller than 10 μm.

3.2. Crystallization Mechanism A main reason for the enhanced crystal growth is the improved kinetic conditions of the modification process caused by an appropriate shear force. The diffusion of growth units from the bulk solution to the crystal surface is apparently enhanced when the molten slag is subjected to a force, which is beneficial for the spinel growth. Fur- thermore, crystal nucleation may also be affected by the shear force, which is usually characterized by the secondary nucleation rate, B (m −3·s −1), given by

jl b BK b MCT ...... (3) −1 where Kb is the nucleation rate constant (kg ), MT is the suspension density (kg·m −3), ∆C is the degree of super- saturation (-), while constants j, l, and b are impact factors. Therefore, the spinel crystallization rate may be affected by the nucleation rate if a strong shear force is exerted on the molten slag and, as a consequence, the spinel growth is eventually restricted. In order to analyze the nucleation behavior of spinel at different shear rates, the number density was studied. Fig- ure 5 shows the results for share rates of 0 s −1, 10.83 s −1 and 21.67 s −1. The number density of spinel was found to decrease by about 45% during the 60-minute modification process for the case without shear force applied, and to decrease by about 67% at a shear rate of 10.83 s −1. How- ever, the trend was changed when the shear rate was at a high level (21.67 s −1): the number density increased during the first 30 min, implying that nucleation was promoted, which is unfavorable to spinel growth. Fig. 4. Variation of grain size distribution of spinel with modifi- 30) According to the literature, three basic types of CSD cation duration in the isothermal process (1 500°C) at are encountered: asymptotic, log-normal and steady-state. shear rates of 0 s −1, 10.83 s −1 and 21.67 s −1. The CSDs in Fig. 4 are log-normal-like distributions with sharp cutoffs toward smaller crystal sizes and a long tail toward larger crystal sizes. For the log-normal case, the 1 F(ln)D 2 V f ()D exp G 2 W ...... (4) occurrence frequency of spinel grains of different diameters D2 H 2 X is given by where α is the mean of the logarithms of the spinel crystal dimensions and β2 describes the variance of the logarithms.

Fig. 6. Schematic diagram of various growth mechanisms: (1) surface-controlled growth; (2) supply-controlled growth; (3) supply-controlled Ostwald ripening; (4) constant rate nucleation and growth.

Fig. 7. Relationship between α and β2 at different shear rates at 1 500°C.

nucleation rate, in which β2 increases linearly with α, (3) supply-controlled growth, in which β2 remains constant, (4) supply-controlled Ostwald ripening, in which β2 decreases with α. A schematic diagram of β2 versus α for the four growth mechanisms is given in Fig. 6. The relationship between α and β2 for the experiments at different shear rates at 1 500°C is plotted in Fig. 7. Since the experimental system studied here is neither an open nor Fig. 5. Number density of spinel at shear rates of 0 s −1 (top panel), closed system, the CSDs in Fig. 4 exhibit pseudo-standard 10.83 s −1 (middle panel) and 21.67 s −1 (bottom panel). distributions. Diao et al.32) reported that some spinel crystals can aggregate and grow together to become spinel clusters These two parameters are given by (cf. Fig. 3) resulting in pseudo-standard CSDs, therefore the relationship between α and β2 becomes more complex. In

BlnDf()D ...... (5) Fig. 7, β2 is seen to decrease almost linearly with α when no shear force was employed. It can be inferred that the growth 2 2 B(ln)Df ()D ...... (6) mechanism of spinel in an isothermal process at 1 500°C is supply-controlled Ostwald ripening: some small spinel crys- As for crystal growth in an open or closed system, the tals dissolved in the molten slag later redeposited on the sur- common CSD-theory based crystal growth mechanisms can faces of larger spinel crystals since larger particles are ener- be summarized as the following four types:31) (1) constant getically favored over smaller particles. When a shear force rate nucleation and growth, in which β2 increases exponen- with a shear rate of 10.83 s −1 was applied, β2 and α show tially with α, (2) surface-controlled growth with decaying an almost direct proportionality. The growth mechanism of

587 © 2019 ISIJ ISIJ International, Vol. 59 (2019), No. 3 spinel in this case is therefore likely to be surface-controlled of 0 s −1, 10.83 s −1 and 21.67 s −1. with a decaying nucleation rate. The crystallization rate of The apparent crystallization behavior of crystals can be spinel is limited by growth rate rather than by the diffusion characterized by the crystallization rate (r, s −4), which is or transport of growth units to the spinel crystal surface. defined as32,33) When a high shear rate (21.67 s −1) was applied, β2 increased almost exponentially with α. It is suggested that nucleation rI U 3 ...... (7) and growth rates remain constant during the crystallization 3 process period studied in the current work. A regime switch where I is the nucleation rate (m −3·s −1) and U is the growth from surface-control to supply-control may occur at a later rate (m·s −1), which can be expressed stage if the material requirement for growth exceeds the 31321 supply capacity. However, this switch is postponed or even IN 0kT()3abexp[ ()TTrr ] ...... (8) prevented by applying the shear force. 11 UD fa 1exp[HTmr()RT ] ...... (9) 3.3. Cooling Process 20) Zhou et al. reported that a chromium-containing spinel where N0 is the number of molecules (or atoms) per unit grew fast if the rate of the cooling process was reasonable. volume defined by 1/a3 (m −3), k is Boltzmann’s constant Inspired by these earlier findings, the effects of shear force (J·K −1), T is the absolute temperature (K), a is the lattice on the growth behavior of spinel was here studied within parameter of the crystal (m), η is the viscosity of slag (Pa·s), the temperature range 1 500–1 150°C at a cooling rate of b = 16π/3 (-), α is the reduced crystal/liquid interfacial −1 5°C·min . The evolution of the spinel mean diameter tension (-), β is the reduced molar heat of fusion (-), ΔTr = under various shear force conditions is shown in Fig. 8. It 1−Tr (-), Tr is the reduced temperature defined by T/Tm (-), can be seen that the sets of observations exhibit a similar Tm is the melting point of the spinel (K), D is the diffu- rising trend, including a rapid growth period from 1 350 to sion coefficient (m2·s −1), f is the fraction of acceptor sites 1 200°C. For cooling treatment without a shear force appli- in the crystal surface (-), ΔHm is the molar heat of fusion cation, the mean diameter showed a slight increase from (J·mol −1), and R is the gas constant (J·mol −1·K −1).34,35) 8.4 μm to 11.1 μm in the cooling process between 1 500°C Several physical property parameters of MgCr2O4 can be and 1 350°C, rising to about 28 μm when the temperature found in Table 2. decreased to 1 150°C. Employment of a shear force did not The results of the calculated nucleation rate, growth rate essentially change the grain size evolution, demonstrating and crystallization rate of MgCr2O4 with respect to tempera- that the driving force of spinel crystallization induced by ture are given in Fig. 9. It is seen that the optimum tempera- the temperature plays a more significant role than the shear ture for MgCr2O4 crystallization is around 1 280°C (1 553 force. The increase in viscosity of the molten slag with K), thus the rapid growth period of spinel (1 350–1 200°C) decreasing temperature clearly limited the effects of the observed in this study suggests a suitable range for spinel shear force. After the same cooling treatment to 1 150°C, crystallization. The enrichment of chromium in the spinel the mean grain diameters obtained were about 32 μm and was studied by using SEM-EDS analysis, indicating that 23 μm for the shear rates 10.83 and 21.67 s −1, respectively. almost all of the chromium existed in the spinel when Further cooling below 1 150°C had only a slight influence the temperature decreased to around 1 300°C, no matter on the grain size. When the samples had cooled to room whether the shear force was applied or not. On the basis of temperature, the mean diameters were found to be about 29

μm, 34 μm and 24 μm for the experiments with shear rates Table 2. Physical property parameters of MgCr2O4.

a, m α β Tm, K 8.32 ×10 −10 1/3 1 2 623

Fig. 8. Variation of spinel mean diameter during the cooling process with shear force applied at shear rates of 0 s −1, 10.83 Fig. 9. Nucleation rate, growth rate and crystallization rate of −1 −1 s and 21.67 s . MgCr2O4 with respect to the temperature.

Eqs. (7)–(9) and these experimental findings, the crystal- Acknowledgements lization rate of spinel in the cooling process appears to be The authors gratefully acknowledge supports by the independent to the shear force. National Natural Science Foundation of China (No. 51704068), the National Key R&D Program of China (No. 3.4. Leaching Test 2017YFC0805100), the Fundamental Research Funds for In order to evaluate the leaching behavior of chromium the Central Universities (No. N172504020). from the modified SS slags, static leaching tests were conducted. The experimental results showed that after 18 h of REFERENCES leaching treatment by sulfuric acid-nitric acid solution at 1) H. Shen, E. Forssberg and U. Nordström: Resour. Conserv. Recycl., pH = 3.2, no chromium could be detected by ICP-OES in 40 (2004), 245. any of the leachates, i.e., the concentration of chromium 2) P. Chaurand, J. Rose, J. Domas and J. Y. Bottero: J. Geochem. Explor., 88 (2006), 10. ions was below the detection limit of the equipment (0.01 3) R. Baciocchi, G. Costa, E. D. Bartolomeo, A. Polettini and R. Pomi: mg·L −1). Under the Standard of Environmental Protection Waste Biomass Valoriz., 1 (2010), 467. 4) E. Kim, J. Spooren, K. Broos, P. Nielsen, L. Horckmans, K. C. Industry HJ/T 301-2007 of the People’s Republic of China, Vrancken and M. Quaghebeur: Chem. Eng. J., 295 (2016), 542. the leaching limits of chromium for the utilization of Cr- 5) H. Shen and E. Forssberg: Waste Manag., 23 (2003), 933. bearing slag in cement and brick industries are 0.30 mg·L −1 6) E. Ahmed, H. M. Abdulla, A. H. Mohamed and A. D. El-Bassuony: −1 Process Saf. Environ. Prot., 104 (2016), 1. and 0.15 mg·L , respectively. Therefore, the modified SS 7) Q. Zhao, C. Liu, B. Li, R. Zevenhoven, H. Saxén and M. Jiang: slag can be utilized in several industrial sectors by enrich- Process Saf. Environ. Prot., 113 (2018), 78. 8) M. Guo, D. Durinck, P. T. Jones, G. Heylen, R. Hendrickx, R. Baeten ing the chromium in the stable spinel phase, and there is no and P. Wollants: Steel Res. Int., 78 (2007), 117. doubt that promoting spinel crystallization improves its sta- 9) M. Gelfi, G. Cornacchia and R. Roberti: 6th European Slag Conf. bility, which broadens the allowable application area of SS (EUROSLAG), The European Steel Association, Brussels, (2010), 1. 10) D. Mombelli, C. Mapelli, S. Barella, C. D. Cecca, G. L. Saout and slags. Specific investigations of the effects of crystallinity E. Garcia-Diaz: Process Saf. Environ. Prot., 102 (2016), 810. and grain size of spinel in SS slag on the ability to release 11) Q. Zhao, C. Liu, P. Shi, B. Zhang, M. Jiang, Q. Zhang, H. Saxén and R. Zevenhoven: Int. J. Miner. Process., 130 (2014), 95. chromium under various conditions, such as in strong acid 12) Q. Zhao, C. Liu, B. Zhang, M. Jiang, J. Qi, H. Saxén and R. solution and seawater, have been conducted and will be Zevenhoven: Steel Res. Int., 86 (2016), 1541. 13) L. Cao, C. Liu, Q. Zhao and M. Jiang: J. Iron Steel Res. Int., 24 reported in forthcoming papers. (2017), 258. 14) H. Cabrera-Real, A. Romero-Serrano, B. Zeifert, A. Hernandez- Ramirez, M. Hallen-Lopez and A. Cruz-Ramirez: J. Mater. Cycles 4. Conclusions Waste Manag., 14 (2012), 317. 15) G. Albertsson, L. Teng, B. Björkman, S. Seetharaman and F. In this study, a shear force was applied in a modifica- Engström: Steel Res. Int., 84 (2013), 670. 16) F. Engström: Ph.D. thesis, Luleå University of Technology, (2007), tion process of stainless steelmaking slags, and the effects http://citeseerx.ist.psu.edu/viewdoc/download?doi= 10.1.1.573.8169 of the shear rate on the grain size evolution of spinel in &rep=rep1&type=pdf, (accessed 2018-06-06). isothermal as well as cooling processes were investigated. 17) M. Tossavainen, F. Engström, Q. Yang, N. Menad, M. L. Larsson and B. Bjorkman: Waste Manag., 27 (2007), 1335. The results indicated that the employment of a shear force 18) Y. Samada, T. Miki and M. Hino: ISIJ Int., 51 (2011), 728. significantly changed the growth behavior of spinel at 19) Q. Zhao, C. Liu, D. Yang, P. Shi, M. Jiang, B. Li, H. Saxén and R. Zevenhoven: Process Saf. Environ. Prot., 105 (2017), 91. 1 500°C, and showed different influences at different shear 20) M. Zhou, P. T. Robinson, B. Su, J. Gao, J. Li, J. Yang and J. Malpas: rates: the growth rate increased when the shear rates were Gondwana Res., 26 (2014), 262. moderate (5.42 s −1, 10.83 s −1 and 16.25 s −1), but decreased 21) J. Weertman: Rev. Geophys., 10 (1972), 287. −1 22) D. S. H. King, M. E. Zimmerman and D. L. Kohlstedt: J. Petrol., 51 when the shear rate was high (21.67 s ). The spinel growth (2009), 21. mechanism was studied using CSD theory, showing regimes 23) A. B. Metzner and R. E. Otto: AIChE J., 3 (1957), 3. 24) A. D. Randolph and M. A. Larson: Theory of Particulate Processes, with supply-controlled Ostwald ripening, surface-controlled 1st ed., Academic Press, New York, (1971), 1. growth with decaying nucleation rate, and constant rate 25) C. Brime and D. D. Eberl: Schweiz. Mineral. Petrogr. Mitt., 82 nucleation and growth at shear rates of 0 s −1, 10.83 s −1 and (2002), 203. −1 26) R. C. F. Lentze and H. Y. McSween, Jr.: Meteorit. Planet. Sci., 35 21.67 s , respectively. The mean diameter of the spinel (2010), 919. grains exceeded 30 μm when the shear rate was controlled 27) A. Torkian, N. Salehi and M. Kord: Petrology, 4 (2013), 33. 28) H. Pourkhorsandi, H. Mirnejad, D. Raiesi and J. Hassanzadeh: Geol. in a suitable range. During the cooling process (at a cooling Carpath., 66 (2015), 257. rate of 5°C·min −1), the shear force was found to have little 29) H. Yu, J. Xu, C. Lin, L. Shi and X. Chen: J. Asian Earth Sci., 58 (2012), 1. effect on the spinel crystallization behavior. Finally, a static 30) L. Li, X. Wu, L. Yu and Y. Dong: Ironmaking Steelmaking, 35 leaching test was conducted to evaluate is chromium could (2008), 367. be leached from the modified SS slag, but no chromium ions 31) D. D. Eberl, V. A. Drits and J. Środoń: Am. J. Sci., 298 (1998), 499. 32) J. Diao, Y. Qiao, X. Zhang, C. Ji and B. Xie: Crystengcomm, 17 were detected in the leachates. The experimental findings of (2015), 7300. this study indicate that chromium can be stabilized by con- 33) X. Zhang, B. Xie, J. Diao and X. J. Li: Ironmaking Steelmaking, 39 (2012), 147. trolling the modification conditions, and that the application 34) D. Turnbull: J. Appl. Phys., 21 (1950), 1022. of a shear force can promote spinel crystallization. 35) D. R. Uhlmann: J. Non-Cryst. Solids, 7 (1972), 337.