Crystallization Behavior and Growth Mechanisms of Spinel Crystals in Vanadium-Containing Slags

Abstract

To improve vanadium leaching and yield rates during wet vanadium extraction processes, the crystallization behavior and growth mechanism of spinel crystals in vanadium-containing slags were investigated by the mathematical models and high temperature experiments. The results show that the suitable temperature range for spinel crystallization was 1473–1573 K. And the spinels’ crystallization abilities were ranked FeCr₂O₄ → FeV₂O₄ → Fe₂TiO₄. Although the concentration of chromium in the vanadium-containing slags was relatively low, the crystallization rate of all spinel crystals was greatly affected by the chromium spinels. The distribution of spinel diameters differed under the actual smelting condition, with the most common diameter range being 10–15 µm. In addition, the mean diameter and volume fraction of spinels increased with prolonging holding time and decreasing temperature in the actual smelting temperature range. Spinel crystals grew rapidly with time at 1623 K, the mean crystal diameter changed from 15.0 µm at 0 min to 24.9 µm at 60 min. And the mean diameter and volume fraction of spinels treated at 1523 K for 30 min were 28.1 µm and 46.3%, respectively. The mean diameter and volume fraction of spinel crystals increased with prolonging holding time and by decreasing the temperature in the actual smelting temperature range (1573–1673 K). Thus, the remaining slag process is beneficial to spinels growth in industrial vanadium-extraction process.

1. Introduction

Vanadium is one of the most important alloying elements and is widely used in metallurgical, chemical engineering, and aerospace applications for its ability to enhance mechanical properties such as tensile strength, hardness, and fatigue resistance.^1,2) With the gradual increase in special steel production that has occurred over the 21st century, vanadium demand is rapidly increasing. Up to 85% vanadium is added as an alloying element in the special steels used in automobiles, ships, aircraft, and spacecraft.^3,4) Natural vanadium mainly exists as vanadium-titanium magnetite (VTM), which is found in China, South Africa, Russia, and the USA.^3,4,5) In general, VTM is reduced in a blast furnace to produce vanadium-bearing hot metal, which is then pre-oxidized in a vanadium-extraction converter (VEC) by blowing oxygen through it to obtain semi-steel and vanadium-containing slag.^6,7) Then, vanadium-containing slags are treated by wet extraction in a three-step process: salt roasting and leaching, solution purification and, finally, precipitation and V₂O₅ fusion.^7,8) Consequently, the phase structure and chemical components of vanadium-containing slags have vital effects on the subsequent vanadium extraction process. According to the actual production and related literatures,^{4,7,8,9,10,11)} the influence of each oxidation in vanadium-containing slag on the subsequent smelting was summarized in Table 1. As listed in Table 1, it is known that non-vanadium-containing oxides in the slag have a negative effect on the wet extraction process. Therefore, it is necessary to improve the quality of vanadium-containing slags and reduce the contents of other oxides beforehand to improve vanadium leaching and yield rates and reduce the amounts of sodium salt and waste discharged.^12,13,14,15) Thus, the improvement of vanadium-containing slags during vanadium extraction processes in converters has been well studied; however, only limited improvements have been obtained.^2,3,4,5,6) In addition, the quality of the vanadium-containing slags is largely determined by the proportion and grain size of the vanadium-containing spinels (V-spinels). These factors are also closely related to the vanadium yield rate, the amount of energy consumption, and the amount of waste discharge in the salt-roasting step.^8,9,10) As V-spinel particles are wrapped in silicate phases in the vanadium-containing slags, larger grain size can increase the exposed area to improve the vanadium oxidation and yield rates during wet extraction process, which can effectively reduce the required sodium salt dosage and roasting temperature.^16,17,18,19) Therefore, increasing the size of V-spinels at a certain amount of vanadium in the slag is very important before conducting the wet vanadium extraction process. In addition, many studies on the solidification of a multicomponent stainless steel slag have been studied to obtain a high yield ratio of valuable elements.^20,21) The effect of the slag composition on the formation of spinel-type inclusions has been investigated to improve the purity of liquid steel.^22,23,24,25) Formation of V-rich spinel phase in CaO–SiO₂–MgO–FeO_t–Al₂O₃–V₂O₃–P₂O₅ slags at 1573 K has been discussed to separate the vanadium-concentrating phase.²⁶⁾ Meanwhile, many researchers have reported that the suitable of crystallization temperatures of spinels in vanadium-containing slags was 1373–1573 K, but the vanadium-containing slags’s viscosity could significantly worsen at temperatures below 1473 K.^{18,19,27,28,29)} In addition, the evolution of the bath temperature during the extraction process is at the range of 1473–1673 K, and the smelting time is merely 15 min. Under high vanadium-extraction temperature (1473–1673 K) and short smelting time (15 min), spinels in vanadium-containing slags are not conducive to growth to cause smaller grain size, which seriously affects the subsequent vanadium-extraction process. Therefore, it is necessary to study and illuminate the crystalline behavior of spinels in vanadium-containing slags in combination with the actual smelting condition.

Table 1. Influence of chemical composition on vanadium-containing slags quality.

Chemical composition	Effect on roasting rate of V2O5	Sources and causes
V2O3 (13.5–19.0%)	The effective composition directly affects the roasting rate	Vanadium content in molten iron
SiO2 (11.8–18.6%)	Silicate easily forms a cladding layer to the spinel phase in slag, causing oxidative difficulty for V3+. Meanwhile, SiO2 easily generates insoluble sodium silicate with vanadium and Na2O in slag to reduce the leaching rate of vanadium and block equipment during filtration	Molten iron with slag, [Si] oxidation in molten iron and coolant containing silicon
CaO (0.6–3.2%) MgO (0.3–5.5%)	Silicate compounds generated by CaO, MgO, and SiO2 in the slag are wrapped on the surfaces of V-spinels. This phase is very stable and destroying it requires much time and high temperatures to reduce the conversion rate of vanadium. In addition, CaO reacts with vanadium in slag to generate water-insoluble calcium vanadate to reduce the leaching rate of vanadium	BF slag brought in by hot melt. The converter lining is corroded during the vanadium extraction process
Cr2O3 (0.9–4.6%) TiO2 (6.9–14.4%)	Plenty of Cr2O3 and TiO2 can significantly improve chemical stability at the boundary of the spinel phase and reduce the conversion rate of vanadium	Oxidation of [Ti] and [Cr] in molten iron
Fe (7.0–10.0%)	Iron oxidizes exothermically during the roasting process to cause agglomeration of materials	Separation of slag- iron is poor during the vanadium extraction process
Fe2O3 (32.0–26.0%)	Vanadium dissolves in Fe2O3 and affects the yield of vanadium	Oxidation of FeO in vanadium-containing slags
MnO (7.4–10.7%)	Compounds form a reddish-brown film that causes filtration difficulty during the leaching process	Oxidation of [Mn] in molten iron
P2O5 (≤0.2%)	P2O5 in slag is easy to leach out with V2O5, which affects the purity of V2O5	Oxidation of [P] in molten iron

According to above discussion, this study comprehensively explores the crystallization ability, precipitation sequence, and optimum crystallization temperature range of spinel nucleation and growth in vanadium-containing slags. On this basis, the crystallization behavior and growth mechanism was studied and illuminated according to the actual smelting condition, with the aim of optimizing the vanadium extraction process and improving the quality of vanadium products.

2. Experimental Section

2.1. Mathematical Model

According to the classical nucleation and growth theory of Turnbull and Fisher,^18,19,27,28) the crystallization of spinels can be divided into two stages: nucleation and growth. The nucleation rate per unit volume I can be calculated by Eq. (1):   

$$I=\frac{N_{V}kT}{3\pi a^3\eta }exp\left[\frac{-16\pi {\alpha }^3\beta }{3T_r{(\mathrm{\Delta }{T}_r)}^2}\right]$$(1)

where N_V is the number of molecules (or atoms) per unit volume, m⁻³; k is Boltzmann’s constant, J/K; T is the absolute temperature, K; a is the lattice parameter of the crystal, Å; η is viscosity, Pa∙s; α is the reduced crystal/liquid interfacial tension, N∙m/J; β is the reduced molar heat of fusion, −; T_r=T/T_m is the reduced temperature, −; T_m is the melting point of the crystal, K; and ΔT_r=1 − T_r, −.

The growth rate U can be expressed as:^18,19,27,28)   

$$U=\frac{fkT}{3\pi a^2\eta }\left[1-exp\left(\frac{-\beta \cdot \mathrm{\Delta }{T}_r}{T_r}\right)\right]$$(2)

  

$$f=\{\begin{array}{ll}1& \mathrm{\Delta }{H}_{\text{m}}<2R{T}_{\text{m}}\\ 0.2\mathrm{\Delta }{T}_r& \mathrm{\Delta }{H}_{\text{m}}>4R{T}_{\text{m}}\end{array}$$(3)

where D is the diffusion coefficient, m²/s; and f is the fraction of acceptor sites on the crystal surface, −.

According to previous reports,^9,29,30) the vanadium-containing slags are mainly composed of spinel phases and silicate phases. V, Ti, and Cr are mainly concentrated in the spinel phase and exist in the form of Fe-spinels because the concentrations of Mg and Mn are much lower than that of Fe. Therefore, only Fe-spinels were considered in the present study, namely Fe_xV_3−xO₄ (1 ≤ x ≤ 3), Fe₂TiO₄, and FeCr₂O₄.

Considering that the viscosity of vanadium-containing slags was calculated by the National Physical Laboratory model (NPL model),^30,31) the optical basicity values of the slag components were obtained from the literature.^31,32,33,34) The calculation parameters of the spinel crystals are given in Table 2.^{18,19,27,28,35)}

Table 2. Parameters of the structure and melting point of three kinds of spinel crystals.

Spinel type	a (Å)	Tm (K)	NV (m−3)	α (N∙m/J)	β (–)	f (–)	k (J/K)
FeV2O4	8.543	2023	1.60387×1027	1/3	1	0.2ΔTr	1.38×10−23
FeCr2O4	8.378	2273	1.62317×1027
Fe2TiO4	8.509	1668	1.70051×1027

2.2. General Methods and Materials

The slag sample used in the present work was taken from a steel plant in China. Its chemical composition was investigated by X-ray fluorescence spectroscopy (XRF; Shimadzu Lab Center XRF-1800) and is presented in Table 3 with standard deviation of ±0.1%. The mineralogical phases of the vanadium-containing slags were examined by X-ray powder diffraction (XRD; X’pert PRO, PANalytical, Netherlands) using Cu Ka1 radiation (λ = 1.5406 Å) with a step of 0.02° (2θ) and scanning rate of 2° min⁻¹ in a range of 10° to 90°. The micro-structures, as well as a compositional analysis of the phases in the samples, were determined by a scanning electron microscope (SEM; SSX-550, Shimadzu, Japan) with an attached energy dispersive X-ray analyzer (EDS) (with operating condition EHT=15.00 kV, WD=7.9–10 mm, Time=20–30 s).

Table 3. Composition of Cr₂O₃-bearing vanadium-containing slag.

Component	FeO	SiO2	V2O3	MnO	Cr2O3	TiO2	MgO	Al2O3	CaO
wi/%	35.72	16.51	14.55	9.20	5.26	12.34	2.37	1.61	2.44
xi	0.381	0.211	0.074	0.100	0.027	0.118	0.046	0.010	0.033

10 g of vanadium-containing slag (particle size <0.1 mm) was charged into a MgO crucible (inner diameter = 15 mm, height = 30 mm, with purity of 99.9 mass% and operating temperature >2073 K) and heated at 1823 K for 30 min under Ar gas (purity of 99.999 vol%, a flow of 400 ml·min⁻¹) in a resistance furnace to ensure fully melting. Thereafter, the melted slag was sequentially cooled to target temperatures at a cooling rate of 2 K/minute and maintained for the required temperature for the required time. After heating, the sample was rapidly taken from the furnace and cooled according to a designed procedure.

2.3. Spinel Morphology and Mean Diameter

The morphology of the slag was studied by SEM and EDS. According to previous studies for vanadium-containing slags,^3,9,35,36) the white phase was made up of spinel crystals, while the dark phase was the silicate phase as shown in Fig. 1(a). As the areas of spinel crystals could be easily measured, the mean diameter d of the spinel crystals was calculated by sequentially measuring spinel crystal areas in 20 different SEM fields of view with image analysis software (Image-Pro Plus 6.0) according to Eq. (4)¹⁹⁾   

$$d=\frac{1}{n}\sum _{i=1}^n{d}_i=\frac{1}{n}\sum _{i=1}^n\sqrt{\frac{4S_i}{\pi }}$$(4)

where n is the number of spinel crystals in the field of view, d_i is the mean diameter of spinel crystal i, and S_i is the area of spinel crystal i, where i = 1, 2, … n. In addition, joint crystals were separated manually, as shown in Fig. 1.

Fig. 1.

Method for determining the mean spinel diameter d from SEM images. (Online version in color.)

3. Results and Discussion

3.1. Mineralogical Phases in Vanadium-Containing Slags

It is widely known that vanadium is a multivalent element that has multiple oxidation states, including VO, V₂O₃, V₂O₄, and V₂O₅.^3,9,37) However, during the vanadium extraction process in the converter, V is dominantly 3+ and reacts with iron oxides to form a stable spinel phase (FeV₂O₄). The XRD result for typical vanadium-containing slag is shown in Fig. 2, it can be observed that the main mineral compositions of vanadium-containing slag were spinel solid solution (FeV₂O₄ with Cr₂O₃, TiO₂ and a small amount of MnO) and fayalite solid solution (Fe₂SiO₄ and FeSiO₃ with MnO and small amounts of MgO, CaO and TiO₂).

Fig. 2.

XRD pattern of typical vanadium-containing slag.

Figure 3 is an SEM photomicrograph that illustrates the morphology of vanadium-containing slag. Figure 3(a) shows that the phases of vanadium-containing slag were mainly composed of three mineral phases, namely, phase 1 (bright white), phase 2 (light gray), and phase 3 (dark gray). According to energy dispersive spectroscopy (EDS), the bright white phase was a solid solution of FeV₂O₄, in which Cr₂O₃, TiO₂ and a small amount of MnO were contained. In other words, titanium and chromium mainly existed in the V-spinel phase. The light gray phase was a solid solution of 2FeO·SiO₂, in which MnO and a small amount of MgO were dissolved. The dark gray phase was FeO·SiO₂ with a high silica content. These results are consistent with the XRD analysis. In addition, the spinel phases that were wrapped in silicate phases were mainly represented as irregular triangles and quadrangles in the SEM image, as shown in Fig. 3(a). Meanwhile, the grain sizes of the V-spinels in the vanadium-containing slag were mainly in the range of 15–30 μm, and the mean diameter and volume fraction of spinels were about 21 μm and 34%, respectively.

Fig. 3.

Microstructure of vanadium-containing slag (1 = spinels, 2 = 2FeO·SiO₂, 3 = FeO·SiO₂). (a) SEM photo of vanadium-containing slag; (b) Spinel; (c) 2FeO·SiO₂; (d) FeO·SiO₂. (Online version in color.)

Therefore, it can be inferred from the above analysis that the phase composition of vanadium-containing slag in the high-temperature melting state (above 1573 K) was mainly a coexistence of spinel phases (namely FeV₂O₄, FeCr₂O_4, and Fe₂TiO₄) and a liquid phase. In order to better understand the formation mechanism of spinel phases in the vanadium-containing slags, the crystallization ability, precipitation sequence and grain size of spinels at different temperatures were investigated.

3.2. Optimum Temperature Range for Spinel Crystal Growth

The nucleation and growth rates of these three kinds of spinels with various temperatures were calculated by Eqs. (1), (2) as shown in Fig. 4. The suitable growth temperature ranges for FeCr₂O₃, FeV₂O₄, and Fe₂TiO₄ were 1700–1900 K, 1550–1750 K, and 1300–1500 K, respectively, and their optimum growth temperatures (namely the peak temperatures T_U) were about 1805 K, 1630 K, and 1380 K. The suitable nucleation temperature ranges for FeCr₂O₃, FeV₂O₄, and Fe₂TiO₄ were 1200–1400 K, 1100–1300 K, and 900–1100 K, respectively, and their optimum nucleation temperatures (namely the peak valves T_I) were 1320 K, 1190 K, and 1000 K. The favorable temperature ranges for the nucleation and growth of spinels in the vanadium-containing slag were ranked T_FeCr2O3 > T_FeV2O3 > T_Fe2TiO4.

Fig. 4.

Effect of temperature on nucleation and growth rates of spinels in vanadium-containing slag. (a) Fe₂TiO₄ spinels; (b) FeV₂O₄ spinels; (c) FeCr₂O₄ spinels. (Online version in color.)

In addition, according to the nucleation and growth theory reported by Turnbull and Uhlmann,^27,28,38) the ability to form an amorphous solid substance can be calculated by (T_U‒T_I)/T_I. Although a larger value of (T_U–T_I)/T_I of a substance indicates a better ability to form an amorphous solid. That is to say, substances with smaller values of (T_U–T_I)/T_I can precipitate more easily. Therefore, the crystallization ability of the spinels in the vanadium-containing slag could also be deduced from Fig. 4. And the calculated results for crystallization abilities of three kinds of spinels in vanadium-containing slag were showed in Table 4.

Table 4. Crystallization abilities of three kinds of spinels in vanadium-containing slag.

Spinel type	Tu/K	TI/K	(Tu−TI)/TI
FeCr2O4	1805	1320	0.367
FeV2O4	1630	1190	0.370
Fe2TiO4	1380	1000	0.380

Overall, the (T_u–T_I)/T_I values of the three kinds of spinels were similar as listed in Table 4, which indicates that they also had similar crystallization abilities. Meanwhile, the (T_u–T_I)/T_I values of the three kinds of spinels in the vanadium-containing slag were low compared to those of other oxides.^27,28,38) In other words, the crystallization abilities of the three spinel types in the vanadium-containing slag were relatively good and the spinel phases were easy to crystallize. In addition, it could be seen from Table 4 that the crystallization abilities of spinels were ranked as follows: FeCr₂O₄ > FeV₂O₄ > Fe₂TiO₄. In order to further investigate the crystallinity of spinels, the crystallization rate r was confirmed by the spinel nucleation and growth rates, as per Eq. (5).³¹⁾   

$$r=\frac{\pi }{\text{3}}{I}^{3}U$$(5)

The total crystallization rate r_Total is defined by a function of the crystallization rates of the three kinds of spinel crystals, as per:   

$$r_{\text{total}}=\sum _i{w}_i{r}_i$$(6)

where r_i is the crystallization rate of the spinel crystal and w_i is its weight percentage. To simplify the calculation, it was assumed that only V, Cr, and Ti existed in the spinel crystals. The weight percentages of FeCr₂O₄, FeV₂O₃, and Fe₂TiO₄ were 12.14 mass%, 33.74 mass%, and 54.12 mass%, respectively.

The curves of crystallization rate r versus temperature are shown in Fig. 5. It was observed that the suitable crystallization rates of FeCr₂O₃, FeV₂O₄, and Fe₂TiO₄ were obtained at temperature ranges of 1150–1250 K, 1300–1400 K, and 1473–1573 K, respectively. The overall crystallization rate for vanadium-containing slag was dominated by the behavior of FeCr₂O₄ spinel crystals. This is because that the structure of a Cr-spinel is similar to that of a V-spinel; thus, some Cr-spinels in the slag provide nuclei that could facilitate V-spinel formation, which benefits the crystallization of vanadium from the slag. In addition, the mineralogical investigation indicates that almost of the vanadium, chromium and titanium in the vanadium-containing slag were formed spinel solid solution as shown in Fig. 2. The crystal structure of spinels is mainly tetrahedron and octahedron. The preference energy of a cation in a spinel is determined by a number of factors, including its ionic radius in both tetrahedral and octahedral coordination, and the difference in its CFSE (crystal field stabilization energy) in tetrahedral and octahedral coordination. As vanadium, chromium and titanium are adjacent in the periodic table, their atomic radius and physicochemical properties are similar, especially chromium and vanadium. Therefore, V³⁺ substitution by Cr³⁺ does not affect the Cr-octahedral or Cr-tetrahedral bond distance. During the smelting process, the octahedral/tetrahedral sites of Cr-spinel can be occupied by trivalent ions such as V³⁺ and Cr³⁺, forming Fe(V,Cr)₂O₃, which can greatly facilitate the precipitation of vanadium from molten slag. This meant that the suitable temperature range for the crystallization of spinel crystals in vanadium-containing slag was 1473–1573 K. At a temperature of about 1520 K, the total crystallization rate was optimum. In addition, the crystallization sequence of the three kinds of spinels was deduced as FeCr₂O₄→FeV₂O₄→Fe₂TiO₄ during the cooling process according to Fig. 5, which is in accordance with Fig. 4.

Fig. 5.

Crystallization rates of three spinel crystal types in vanadium-containing slag, both individually (top three graphs) and overall (bottom). (Online version in color.)

On this basis, combined with actual smelting conditions and theoretical calculation values, the crystalline behavior and growth mechanism of spinels in vanadium-containing slags should be further studied and expounded to obtain the suitable crystallization conditions.

3.3. Growth Mechanism of Spinels According to Crystal Size

Figure 6 shows the distribution rate of spinel crystal size under heat treatment for 15 min at different temperatures (the actual time of vanadium extraction). Overall, the highest distribution rate occurred in the diameter range of 10–15 μm under different conditions. However, a few large-sized spinel crystals were produced, reaching diameters of 40–45 μm. This is caused by the following reasons: On the one hand, a small number of spinel grains meet and bridge to form a new large grain in the process of growth; on the other hand, most of spinel grains are small-angle grain boundaries with small mobility and relatively stable growth, but a few large-angle grain boundaries appear with high mobility to improve dynamic condition, and the corresponding grains can grow rapidly. Temperatures of 1473–1623 K benefited the growth of spinel crystals, especially the optimum temperature of 1573 K, where more large spinel crystals were generated.

Fig. 6.

Size distributions of spinel crystals formed at different temperatures for 15 min. (Online version in color.)

Figure 7 shows the size distributions of spinel crystals formed with different holding times at 1623 K. It can be seen that the distribution peak for spinel size differed according to holding time, reaching approximately 0.25 at about 10 mm after 0 min, 0.20 at about 15 mm after 15 min, 0.19 at about 27 mm after 30 min, 0.23 at about 29 mm after 45 min and 0.25 at about 29 mm after 60 min. Spinel crystals grew rapidly with holding time, with large ones reaching a diameter of 40–45 μm. The mean crystal diameter changed from 15.2 μm at 0 min to 24.9 μm at 60 min.

Fig. 7.

Size distributions of spinel crystals formed with different holding times. (Online version in color.)

Figure 8 shows the effect of temperature on the grain size and volume fraction of spinels with a holding time of 30 min. It can be seen that the temperatures range of 1473–1623 K benefited spinel crystal growth and precipitation, with the mean diameter of spinel crystals being > 22 μm and reaching a good size. Meanwhile, the volume fraction of spinel crystals first increased rapidly at temperature below 1523 K and then decreased gradually with temperature, finally decreasing rapidly at temperature above 1623 K. In addition, the mean diameter and volume fraction of spinel crystals at 1523 K reached good values of 28.1 μm and 46.3%, respectively. In general, the optimum precipitation and particle size for spinel crystals could be obtained at the temperature range of 1473–1623 K.

Fig. 8.

Effect of temperature on spinel grain size and volume fraction.

Figure 9 shows the grain size and volume fraction of spinels with holding time at various temperatures. Both were smaller at 1673 K than at 1573 K as the nucleation driving force was small with low undercooling. Specifically, the mean diameter and volume fraction increased from 10.8 μm and 27.0%, respectively, to 19.9 μm and 33.1% with a holding time increase of 0 to 90 min at 1673 K. This indicates that the grain size and low-volume fraction of the spinels hardly changed with increasing holding time at that temperature. However, the mean diameter and volume fraction of spinel crystals increased obviously from 15.8 μm and 34.9%, respectively, to 28.6 μm and 51.0% with a holding time increase of 0 to 90 min at 1573 K. In addition, the mean diameter and volume fraction of spinels at 1573 K increased rapidly within 30 min and more slowly afterward. In general, the suitable the mean diameter and volume fraction of spinels were more desirable at 1573 K than at 1673 K, which indicates that appropriate decreased temperature and extension time in smelting process are beneficial to the precipitation and growth of spinel crystals.

Fig. 9.

Effect of holding time on spinel grain size and volume fraction. (Online version in color.)

To further investigate the effects of temperature and holding time on the grain size and volume fraction of spinels, the mean diameters and the amount of precipitation were studied using different treatment methods according to actual smelting condition. Based on the actual smelting temperatures (1673–1573 K), this study investigated the mean diameters and volume fraction of spinel crystals formed by different treatments, as shown in Fig. 10. It can be observed in Fig. 10 that the mean spinels diameter and amount of spinels in the vanadium-containing slag increased obviously from treatment 1 to treatment 4. Furthermore, it can be seen in Figs. 8 and 9 that the mean diameter and volume fraction of spinel crystals increased with prolonging holding time and by decreasing the temperature in the actual smelting temperature range (1573–1673 K), which is consistent with the present study’s (as shown in Fig. 10). Therefore, it can be concluded that a suitable holding time benefits spinel crystal growth. Therefore, the remaining slag operation in the vanadium-extraction process is beneficial to spinel growth in industrial production.

Fig. 10.

Mean diameter of spinel crystals under different treatment crafts. (a) SEM photo of vanadium-containing slag; (b) grain size and volume fraction of spinels. Note: 1 craft = 1673 K for 15 min; 2 craft = 1673 K for 15 min → 1623 K for 15 min; 3 craft = 1673 K for 15 min → 1623 K for 15 min → 1573 K for 15 min; 4 craft = 1673 K for 15 min → 1623 K for 15 min → 1573 K for 15 min → room temperature. (Online version in color.)

4. Conclusions

The crystallization behavior and growth mechanism of spinel crystals in vanadium-containing slags were investigated at various temperatures and holding times by the crystallization ability, precipitation sequence and optimum crystallization temperature range of spinel nucleation and growth, and the following conclusions can be drawn:

(1) The crystalline sequence and abilities of spinels in vanadium-containing slag was FeCr₂O₄ → FeV₂O₄ → Fe₂TiO₄. The suitable temperature range for crystallization of spinel crystals in vanadium-containing slag was 1473–1573 K. Although the concentration of chromium in the vanadium-containing slag was relatively low, the crystallization rate of all spinel crystals was greatly affected by the chromium spinel.

(2) The distribution of spinel crystal diameters differed according to the temperature of a 15 min treatment, with the highest frequency occurring for diameters of 10–15 μm. Spinel crystals grew rapidly with time at 1623 K, the mean crystal diameter changed from 15.2 μm at 0 min to 24.9 μm at 60 min.

(3) The optimal temperatures for spinel growth and precipitation were 1473–1623 K. And the mean diameter and volume fraction of spinels treated at 1523 K for 30 min were 28.1 μm and 46.3%, respectively.

(4) The mean diameter and volume fraction of spinel crystals increased with prolonging holding time and by decreasing the temperature in the actual smelting temperature range (1573–1673 K). Thus, the remaining slag operation in industrial vanadium extraction processes is beneficial to spinel growth.

Acknowledgment

This work was supported by the National Natural Science Foundation of China (No. 51804094).

References

• 1)   X. S.  Li and  B.  Xie: Int. J. Miner. Metall. Mater., 19 (2012), 595.
• 2)   X.  Zhang,  B.  Xie,  J.  Diao and  X. J.  Li: Ironmaking Steelmaking, 39 (2012), 147.
• 3)   W. J.  Huang,  M.  Chen,  X.  Shen,  S.  Yu,  N.  Wang and  L.  Xu: Steel Res. Int., 88 (2017), 1600324.
• 4)   W. J.  Huang,  S.  Yu,  X.  Shen,  L.  Xu,  N.  Wang and  M.  Chen: Steel Res. Int., 87 (2016), 1228.
• 5)   X. R.  Wu,  L. S.  Li and  Y. C.  Dong: ISIJ Int., 47 (2007), 402.
• 6)   Y. C.  Dong,  X. R.  Wu and  L. S.  Li: ISIJ Int., 45 (2005), 1238.
• 7)   B.  Liu,  H.  Du,  S. N.  Wang,  Y.  Zhang,  S. L.  Zheng,  L. J.  Li and  D. H.  Chen: AIChE J., 59 (2013), 541.
• 8)   H. X.  Fang,  H. Y.  Li and  B.  Xie: ISIJ Int., 52 (2012), 1958.
• 9)   D. X.  Huang: Vanadium Extraction and Steelmaking, Metallurical Industry Press, Beijing, (2000), 56 (in Chinese).
• 10)   H. X.  Fang,  H. Y.  Li,  T.  Zhang,  B. S.  Liu and  B.  Xie: ISIJ Int., 55 (2015), 200.
• 11)   L. S.  Li,  X. R.  Wu,  L.  Yu and  Y. C.  Dong: Ironmaking Steelmaking, 35 (2008), 367.
• 12)   X.  Li,  H.  Yu and  X.  Xue: Procedia Environ. Sci., 31 (2016), 582.
• 13)   Z. H.  Liu,  Y.  Li,  M. L.  Chen,  A.  Nueraihemaiti,  J.  Du,  X.  Fan and  C. Y.  Tao: Hydrometallurgy, 159 (2016), 1.
• 14)   H. B.  Liu,  H.  Du,  D. W.  Wang,  S. N.  Wang,  S. L.  Zheng and  Y.  Zhang: Trans. Nonferr. Met. Soc. China, 23 (2013), 1489.
• 15)   X. S.  Li,  B.  Xie,  G. E.  Wang and  X. J.  Li: Trans. Nonferr. Met. Soc. China, 21 (2011), 1860.
• 16)   S. Q.  Zhang,  B.  Xie,  Y.  Wang,  T.  Guan,  H. L.  Cao and  X. L.  Zeng: J. Iron Steel Res. Int., 19 (2012), 33.
• 17)   L.  Yu,  Y. C.  Dong,  G. Z.  Ye and  S.  Du: Ironmaking Steelmaking, 34 (2007), 131.
• 18)   W.  Zhou,  B.  Xie,  W. F.  Tan,  J.  Diao,  X.  Zhang and  H.  Li: JOM, 68 (2016), 2520.
• 19)   L. Y.  Shi,  Y. L.  Zhen,  D. S.  Chen,  L. N.  Wang and  T.  Qi: ISIJ Int., 58 (2018), 627.
• 20)   S.  Jung,  G.  Kim and  I.  Sohn: J. Am. Ceram. Soc., 100 (2017), 3771.
• 21)   G.  Kim and  I.  Sohn: J. Hazard. Mater., 359 (2018), 174.
• 22)   J.  Shin and  J.  Park: ISIJ Int., 58 (2018), 88.
• 23)   L.  Jiang,  Y.  Bao,  Q.  Yang,  Y.  Chen,  G.  Liu,  F.  Han,  J.  Wei,  F.  Engström and  J.  Deng: Steel Res. Int., 88 (2017), 1700066.
• 24)   S.  Jung,  K.  Jung and  I.  Sohn: Metall. Mater. Trans. A, 48 (2017), 617.
• 25)   A.  Harada,  A.  Matsui,  S.  Nabeshima,  N.  Kikuchi and  Y.  Miki: ISIJ Int., 57 (2017), 1546.
• 26)   L.  Wu,  C.  Qi,  B.  Yan,  J.  Wang and  Y.  Dong: Metall. Mater. Trans. B, 50 (2019), 924.
• 27)   D.  Turnbull: J. Appl. Phys., 21 (1950), 1022.
• 28)   D.  Turnbull: Contemp. Phys., 10 (1969), 473.
• 29)   J.  Diao,  B.  Xie,  C. Q.  Ji,  X.  Guo,  Y.  Wang and  X.  Li: Cryst. Res. Technol., 44 (2009), 707.
• 30)   L.  Zeng,  Y.  Wang,  L. K.  Fan and  B.  Xie: J. Iron Steel Res. Int., 21 (2014), 910.
• 31)   K. C.  Mills and  S.  Sridhar: Ironmaking Steelmaking, 26 (1999), 262.
• 32)   E.  Bordes-Richard: Top. Catal., 50 (2008), 82.
• 33)   X. G.  Huang: Principles of Steel Metallurgy, Metallurgical Industry Press, Beijing, (2017), 134 (in Chinese).
• 34)   C. S.  Fu: Principles of Nonferrous Metallurgy, Metallurgical Industry Press, Beijing, (1993), 86 (in Chinese).
• 35)   D.  Lenaz and  V.  Lughi: Phys. Chem. Miner., 40 (2013), 491.
• 36)   J. C.  Li,  Z. C.  Guo and  J. T.  Gao: Ironmaking Steelmaking, 41 (2014), 710.
• 37)   L.  Whittaker,  J. M.  Velazquez and  S.  Banerjee: CrystEngComm, 13 (2011), 5328.
• 38)   W. A.  Johnson and  R. F.  Mehl: Trans. Am. Inst. Min. Metall. Pet. Eng., 135 (1939), 416.