Valorization of Coal Gangue and Vanadium-titanium Slag into Glass-ceramic Coating for Oxidation Resistance of 60Si2Mn Spring Steel at High Temperature

Abstract

Glass coatings and ceramic coatings are usually used to protect the slabs from oxidation during the reheating process of 60Si2Mn spring steel before hot rolling. In view of the characteristic that the primary glass is able to transform from amorphous phase to ceramic phase during the reheating process, glass-ceramics are potential to become a new type of coating materials with both the advantages of the glass coatings and ceramic coatings. In this paper, glass-ceramic was firstly prepared with solid wastes including vanadium-titanium slag (VTS) and coal gangue (CG). With the increasing content of VTS, the main crystalline phase of the glass-ceramic transformed from cordierite to spinel and hercynite, and the crystallization activation energy (E_k) increased. New glass-ceramic coatings were prepared with the as-prepared primary glass powders with different proportion of VTS. When the proportion of VTS within the primary glass powders was 6% by weight, the glass-ceramic coatings possessed the best anti-oxidation effect of 74.4% at 1100°C for 2 h. The protective mechanism of glass-ceramic coatings was discussed with the results of SEM and EDS. Before crystallization, the amorphous phase within primary glass contributed to good sintering performance, and thus the coatings formed a dense film, slowing down the oxidation caused by the direct contact between oxygen and the slabs. With the increasing temperature, amorphous phase gradually transformed to ceramic phase, preventing the oxidation resulted from the diffusion of Fe ions.

1. Introduction

Carbon steel manufacture always involves a slab reheating process before hot rolling. During the reheating process, slabs suffer serious oxidation at high temperature based on different kind of steel,^1,2) such as 60Si2Mn, which should go through a treatment at 1100°C for 2 h. The oxidation gives rise to weight loss of slabs and energy waste.^3,4) Otherwise, the deciduous scale generated by the oxidation will result in the pollution and corrosion to furnace.⁵⁾ Temporary protective coating, an effective and economical method, is considered as an alternative solution to prevent the slabs from oxidation during reheating process.^6,7,8)

Generally, the protective coating usually involves glass coating and ceramic coating. The glass coating has a low melting point (<800°C), so it can easily form a hermitic film on the substrate surface to hinder the oxidation caused by the direct contact between oxygen and substrate.^9,10,11,12) However, some components within the glass coating, such as borate and phosphate, are prone to react with the substrate at high temperature, and the corrosion brings about surface quality problems. Therefore, the glass coating is only suitable for some special steel, of which the reheating treatment is conducted at low temperature. The ceramic coating (based on MgO, Al₂O₃, SiO₂, CoO, ZnO, ZrO₂, etc.), similar to the refractory, has a relative high melting point and chemical stability.^2,8,13) The main crystalline phase of ceramic coatings are usually the cordierite, spinel^14,15) or hercynite,¹⁶⁾ etc. Compared with the oxide layer formed on the surface of the substrate, the diffusion of Fe ions through the ceramic coatings is much more difficult. As a result, ceramic coatings have a significant inhibitory effect on the oxidation caused by the diffusion of Fe ions.²⁾ However, the ceramic powders are hard to sinter by a solid-phase reaction without any compaction process. Therefore, it is difficult for ceramic coatings to form a dense structure at a relative low temperature to protect the slabs from the oxidation, which is caused by the direct contact between the oxygen and slabs. In view of the characteristic that the primary glass is able to transform from amorphous phase to ceramic phase during the reheating process, the protective coating with glass-ceramic as principle materials is potential to integrate both the advantages of glass coating and ceramic coating.^17,18)

In order to reduce the cost of the glass-ceramic coating, some solid waste rich in silicon and aluminum oxide can be used as raw materials. Coal gangue (CG), which mainly contains silicate and aluminium oxide, is produced by coal mining and dressing. Statistically, the accumulation of coal gangue has already exceeded to one billion tons at present in China, and the amount is still increasing at a rate of one hundred million tons per year.¹⁹⁾ Most of them are discharged into landfills and occupy great quantities of land.²⁰⁾ Additionally, it poses a potential fire hazard due to the flammability. There are varieties of utilizations based on CG,^21,22) while the utilization rate is still below 15%.²³⁾ Vanadium-titanium slag (VTS) is the solid residual generated from the sodium-translated roasting pelletizing process after smelting of vanadium-titanium ore, which mainly contains iron oxide, silicate, titanium oxide and alkali salt.^24,25) It has accumulated more than three million tons of storage every year.²⁶⁾ The unexploited VTS occupy a large amount of industrial and agricultural land, so safety problems and environmental pressure are presented. Rich-iron slag, such as VTS, could be utilized as additives for pellets roasting. However, the alkali salt in VTS would reduce the melt point of pellets and result in scaffolding. Besides, the valorization of VTS into ceramics for the preparation of decorative materials seems as a feasible solution. However, in order to enhance the decorative performance, solid waste should be de-ironed by the acidification method, which is a high-cost treatment and causes secondary pollution. Therefore, the utilization of VTS is limited and an economical approach is urgently required.

Vitrification is recognized as an effective way to utilize silicon-based solid waste.^{27,28,29,30,31,32,33)} According to the compositions of VTS and CG, vitrification, especially the valorization of solid waste into glass-ceramics to obtain functional materials, is considered as an applicable way for the utilization of both the solid waste. Silicate, aluminium oxide and alkali salt in CG and VTS are the ideal components for the preparation of primary glass,^34,35) and titanium oxides in VTS are effective catalysts for the crystallization of glass-ceramics.³⁶⁾ The Fe³⁺ and Fe²⁺ in VTS could replace the Si⁴⁺ within glass matrix and weaken the stability of the glass structure. Therefore, crystallization temperature of the primary glass would reduce with the VTS as additives.^37,38)

In this paper, glass-ceramic of MgO–Al₂O₃–SiO₂ system was prepared with VTS and CG as principle raw materials. The influence of raw material proportion on the crystallization behavior was investigated. Then, a new protective coating was prepared based on the as-prepared primary glass powders to prevent 60Si2Mn spring steel from oxidation at 1100°C for 2 h. The anti-oxidation performance of the glass-ceramic coating was examined and the protective mechanism was discussed as well. Some achievements in this paper were not only applied to anti-oxidation of 60Si2Mn spring steel, but also provided new ideas for waste management, such as the comprehensive utilization of vanadium-titanium slag and coal gangue.

2. Materials and Methods

2.1. Preparation of Steel Sample

Spring steel, 60Si2Mn, with a specimen size of 10 × 10 × 10 mm³ was used for evaluating the oxidation resistance performance of glass-ceramic coating. The composition of 60Si2Mn was listed in Table 1. The samples were polished by abrading with SiC papers from 200# grit to 1200# grit, and then cleaned through an ultrasonic treatment.

Table 1. Chemical composition of spring steel 60Si2Mn (mass fraction, mass%).

Element	C	Si	Mn	S	P	Cr	Ni	Cu	Fe
mass%	0.6	2	0.78	0.035	0.035	0.35	0.35	0.25	balance

2.2. Preparation of Primary Glass Sample

Coal gangue (CG) in this work was collected from a coal dressing plant in Hebei province in China, and Vanadium-titanium slag (VTS) was collected from a smelting plant in Sichuan province in China. Both the solid waste was treated by a shock crusher until the particle size reached above 200# grit. In order to remove the remaining carbon and other decomposable components, coal gangue powders were dehydrated at 120°C for 2 h and heated at 950°C until the residual weight was constant. The compositions of CG and VTS were detected by X-ray fluorescence (XRF; AXIOS, Panalytical, Amsterdam, The Netherlands), and the results were shown in Table 2.

Table 2. Chemical compositions of VTS and CG (mass fraction, mass%).

Raw material	Chemical composition of raw materials
	SiO2	Al2O3	MgO	Fe2O3 and Fe3O4	TiO2	R2O	V2O5	CaO	Cr2O3	MnO2
VTS	17.1	1.5	1.1	54.2	10.4	4.2	1.2	1.6	2.1	6.6
CG	64.7	27.4	0.3	2.2	1.5	2.7	–	1.2	–	–

In order to investigate the influences of raw material proportion on the main crystalline phase and crystallization kinetics, four groups of primary glass samples with different raw material proportions were prepared. It was noted that although the mass proportion was different, the molar ratio of 2:2:5 (MgO:Al₂O₃:SiO₂) maintained constant which was balanced by the addition of Al₂O₃ powders (a-Al₂O₃ in Table 3) and MgO powders (a-MgO in Table 3), in correspondence with the stoichiometric ration of cordierite (2MgO·2Al₂O₃·5SiO₂). The components of the four primary glass samples were shown in Tables 3 and 4. All the components of raw materials was mixed and ground by ball-milling for 6 h until the particle size reached above 400# grit, and thereafter melted in muffle furnace at 1600°C for 2 h to ensure homogeneity. After water quenching, the frit of primary glass was obtained. The process route of primary glass and glass-ceramic sample preparation was shown in Fig. 1.

Table 3. Raw material proportions of four groups of primary glass sample (mass fraction, mass%).

Sample No.	Raw materials proportion
	a-Al2O3	a-MgO	VTS	CG
1	12.6	12.4	1	74
2	12.1	11.9	6	70
3	11.6	11.4	12	65
4	11.6	10.4	18	60

Table 4. Chemical compositions of four groups of primary glass sample. (mass fraction, mass%).

Sample No.	Chemical composition of four groups of primary glass sample
	SiO2	Al2O3	MgO	Fe2O3 and Fe3O4	TiO2	R2O	V2O5	CaO	Cr2O3	MnO2
1	48.049	32.891	12.633	2.17	1.214	2.04	0.012	0.904	0.021	0.066
2	46.316	31.37	12.176	4.792	1.674	2.142	0.072	0.936	0.126	0.396
3	44.107	29.59	11.727	7.934	2.223	2.259	0.144	0.972	0.252	0.792
4	42.374	28.569	10.77	10.556	2.683	2.361	0.204	1.004	0.357	1.122

Fig. 1.

Process route of the experiment.

2.3. Preparation of Coating Slurry

The frit of primary glass was dried to constant weight at 120°C for 4 h and then ground by ball-milling until the particle size reached above 325# grit. Then, the primary glass powders were obtained for the preparation of coating slurry. The composition of coating slurry was shown in Table 5. It was mentioned that the coating slurry samples were same with the proportion of raw materials, while different with the composition of primary glass powders, as listed in Tables 3 and 4. Colloidal silica (40 mass%) was used as binder agent to enhance the adhesive strength, and sodium polyacrylate acted as the surface active agent to assist the dispersity of coating slurry. All the raw materials were mixed in a beaker under magnetic stirring. The prepared slurry was coated on carbon steel by spraying gun, which was connected to an air compressor with pressure of 8 bar. The thickness of the coating was 0.4 mm.

Table 5. Compositions of coating slurry (mass fraction, mass%).

Components	Primary Glass Powders	Colloidal Silica (40 mass%)	Sodium Polyacrylate	Water
Mass Fraction, mass%	60	20	5	15

2.4. Materials Characterization

The crystallization kinetics of glass-ceramic was investigated through differential thermal analysis (DTA), which was operated with thermal analyzer (DTA; STA449, Netzsch, Nuremberg, Germany). The primary glass samples were heated in flowing air from room temperature to 1250°C at a rate of 5°C min⁻¹, 10°C min⁻¹, 15°C min⁻¹ and 20°C min⁻¹, respectively.

The crystalline phases of primary glass samples after heat-treatment were identified by X-ray diffraction (XRD; X’Pert Pro, Philips, Amsterdam, The Netherlands).

The scanning electron microscopy (SEM; JSM-6700F, JEOL, Tokyo, Japan) equipped with an Energy Dispersive X-ray Spectroscopy (EDS, NORAN, Thermo Fisher, USA) was employed to investigate the micrograph of the crystalline phase within glass-ceramic samples and the element distribution within the coatings. The frit was heat-treated to 1100°C for 2 h at a rate of 10°C min⁻¹ to obtain the glass-ceramic samples, here, the parameters of heat-treatment was same to that of spring steel 60Si2Mn during reheating process. Before the investigation, the surface of glass-ceramic sample was smoothed by SiC papers and planed by diamond abrasion paste, followed by chemical etching in the hydrogen fluoride solution (5 vol. %) for 1 min.

2.5. Evaluation of Anti-oxidation Performance

Continuous thermos-balance (RZ, Luoyang Precondar, Luoyang, China) was employed to evaluate the anti-oxidation performance of glass-ceramic coatings. Bare sample and coated samples were heated from room temperature to 1100°C at 10°C min⁻¹ and maintained for 2 h. During the heating process, the weight gains were continuously recorded.

3. Results and Discussion

3.1. Microstructure and Crystalline Phase of the Primary Glass Sample

In this paper, the microstructures of glass-ceramic samples were investigated with SEM. Images in Fig. 2 indicated that there were homogeneously distributed plate-like phase and granular phase. With an increasing proportion of VTS, the plate-like crystal gradually disappeared, while the granular ones increased. The X-ray diffraction patterns of samples with different proportion of raw materials were shown in Fig. 3. Results in Fig. 3 indicated that with an increasing proportion of VTS in raw materials of primary glass, the main crystalline phase gradually transformed from cordierite to spinel and hercynite.

Fig. 2.

Morphology of samples with different proportion of raw materials: (a) sample with 1% VTS, (b) sample with 6% VTS, (c) sample with 12% VTS and (d) sample with 18% VTS.

Fig. 3.

X-ray diffraction patterns of samples with different proportion of raw materials. (Online version in color.)

It was widely accepted that titanium oxide was an effective catalyst for crystallization under heterogeneous nucleation mode. Before the nucleation of primary glass, small amorphous droplets (Ti-rich phase) formed and uniformly distributed among the glass melt (Si, Al-rich phase). Interface emerging between Ti-rich phase and glass melt (Si, Al-rich phase) contributed to heterogeneous nucleation in crystallization process.^39,40) According to a pertinent literature,³⁷⁾ when the content of Fe ions was within a certain range, Fe ions firstly destroyed the glass structure by replacing Si ions. When the concentration of Fe ions was obviously excessive, the Fe ions would exchange with the Al ions and Mg ions within the cordierite, causing the lattice distortion and giving rise to the transition of crystalline phase from cordierite to spinel and hercynite.

3.2. Performance of the Coating

The anti-oxidation performance of protective coatings was evaluated by continuous thermos-balance. The bare and coated samples were heated from room temperature to 1100°C at 10°C min⁻¹ maintaining for 2 h, and the weight gains were continuously recorded. The anti-oxidation performance was evaluated by the weight gain per unit area of steel samples during heating process. The result of non-isothermal investigation from room temperature to 1100°C was presented in Fig. 4(a), and the result of isothermal investigation at 1100°C for 2 h was shown in Fig. 4(b). According to the results in Fig. 4(a), the non-isothermal anti-oxidation performance increased with the increasing proportion of VTS, while the isothermal anti-oxidation performance of coating during the temperature-holding stage at 1100°C decreased with the increasing proportion of VTS based on the result in Fig. 4(b).

Fig. 4.

(a) Non-isothermal anti-oxidization performances of the glass-ceramic coating samples (b) Isothermal anti-oxidization performances of the glass-ceramic coatings at 1100°C. (Online version in color.)

The opposite result above mainly resulted from the crystallization of the primary glass powders, which could be explained by the crystallization kinetics. Four groups of primary glass samples were heated in flowing air from room temperature to 1250°C at a rate of 5°C min⁻¹, 10°C min⁻¹, 15°C min⁻¹ and 20°C min⁻¹, respectively. The curves of DTA versus temperature (K) were shown in Fig. 5 and the corresponding crystallization peak temperatures were listed in Table 6.

Fig. 5.

Curves of DTA versus temperature (K) of four primary glass samples: (a) sample with 1%VTS, (b) sample with 6% VTS, (c) sample with 12% VTS and (d) sample with 18% VTS. (Online version in color.)

Table 6. Crystallization peak temperatures (T_p) of four primary glass samples.

Sample No.	Tp (K)
	α= 5°C min−1	α= 10°C min−1	α= 15°C min−1	α= 20°C min−1
1	1275	1298	1322	1334
2	1221	1246	1262	1274
3	1098	1110	1120	1128
4	1028	1038	1045	1052

JMA equation was widely applied to analyze non-isothermal kinetics of primary glass sample,^41,42) which could be expressed as Eq. (1):   

$$ln\alpha =-\frac{E_k}{R}\cdot \frac{1}{T_p}+\frac{AR}{E_k}$$(1)

Here, α was the heating rate of DTA, E_k was the crystallization activation energy of primary glass sample, R was the gas constant (8.314 J·mol⁻¹·K⁻¹), T_p was the crystallization peak temperature corresponding to different heating rate, A was the frequency factor.

Based on the JMA equation, values of α and T_p could be obtained from Table 6 and a linear fitting curve of ln α versus 1/T_p could be plotted. As a result, E_k could be calculated from the slope of the curves. The linear fitting curves of JMA equation were shown in Fig. 6 and the results were listed in Table 7, which suggested that the crystallization activation energy (E_k) gradually increased with the increasing proportion of VTS in raw materials of primary glass.

Fig. 6.

Linear fitting curves of ln α versus 1/T_p. (Online version in color.)

Table 7. Results of crystallization activation energy of primary glass samples.

Sample NO.	Linear fitting equation	Adj. R-Square	Ek (kJ/mol)
1	y = −36152.71x + 15.72	0.97692	300.57
2	y = −38267.85x + 18.74	0.99928	318.16
3	y = −54881.99x + 37.64	0.98207	456.29
4	y = −61023.37x + 47.15	0.98259	507.35

According to the result of crystallization kinetics of the primary glass samples, as the proportion of VTS within the primary glass powders increased, the crystallization activation energy (E_k) increased, and thus the proportion of residual amorphous phase within the coatings increased. As a result, the sintering performance of the coating and the anti-oxidation ability of the coating at the initial stage of oxidation were improved, especially during the temperature-rising process. However, the excessive amorphous phase accelerated the diffusion of Fe ions during the temperature-holding stage, and therefore the anti-oxidation effect of the coating decreased with increasing VTS proportion of the primary glass powders.

Considering the contrary effect of VTS proportion in raw materials on the non-isothermal and isothermal anti-oxidation performance, the proportion of VTS had an optimum value. As was demonstrated, when the proportion of VTS was 6% by weight, the optimal anti-oxidation effect of glass-ceramic coating for 60Si2Mn spring steel was 74.4% at 1100°C for 2 h.

3.3. Protective Mechanism of the Protective Coating

Generally, the oxidation of metals was simultaneously affected by the direct contact between oxygen and the substrate as well as the ions diffusion within the oxide layers. At the initial stage of oxidation, the metal surface first formed a layer of porous oxide layer, and the metal oxidation at this time followed a linear law, indicating that the metal oxidation at the initial stage was mainly caused by the direct contact between the oxygen and the substrate. With the further progress of the oxidation, the oxide layer gradually became dense and reached a certain thickness, and the oxide layer was a structure mainly made of FeO and Fe₃O₄, both of which were p-type semiconductor. As a result, the oxidation process was mainly controlled by the external diffusion of Fe ions within the oxide layer. At this time, the metal oxidation followed the parabolic law.^43,44) The design of anti-oxidation coating was based on the oxidation mechanism above. Taking the coating with 6% VTS as an example, the protective mechanism of glass-ceramic coating was discussed based on the results of SEM and EDS.

In order to verify that the coating had good sintering performance and could form a compact structure to prevent the oxidation caused by direct contact between oxygen and the substrate, the surface morphology of the coating at different temperatures was analyzed by SEM. As shown in Fig. 7, when the temperature ranged from 800°C to 1100°C, the coating gradually sintered and turned dense. The surface of the coating was porous when the sample was heated at 800°C and 900°C. When temperature reached 1000°C, visible pores disappeared and the surface turned smooth, indicating a well-sintered structure formed and gave the possibility of a good anti-oxidation performance. When the temperature rose to 1100°C, particle morphology reappeared on the surface of the coatings, which was caused by the crystallization of the primary glass. The change of surface morphology of the coating was consisted with characterization of the crystallization kinetics of the primary glass powders. Combined with the protective effect of the coating during the temperature-rising stage, it was concluded that the coating with primary glass powders as raw materials exhibited excellent sintering performance and the formed compact structure contributed to prevent the oxidation caused by the direct contact between oxygen and the substrate when the temperature was above 1000°C.

Fig. 7.

The surface micrograph pictures of coating with 6% VTS at different temperatures: (a) 800°C, (b) 900°C, (c) 1000°C, (d) 1100°C.

In order to clarify the inhibition effect of the coating on the external diffusion of Fe ions, SEM with EDS method was used to analyze the distribution of Fe elements within the coatings and the oxide layer of bare sample. The x-axis of the spectrogram was the value of characteristic X-ray energy, which was used to identify the element composition of the area. The y-axis value of the spectrogram represented the counts of the characteristic X-rays collected by the instrument. The value of the intensity (counts) was positively relative to the constant of the element and the semi-quantitative analysis of element content was based on it. According to the result in Fig. 8, the coating was a structure consisted of the cordierite particles and Si-rich phase distributing among the particles. With the diffusion of Fe ions within the coating, a transition layer formed between the coating and the substrate. By comparing the EDS data at the corresponding positions within the coating and the transition layer, it could be deduced that only a small amount of Fe ions doped into cordierite particles, while the majority of Fe ions diffused through the Si-rich phase among the particles. As shown in Fig. 9, the oxide layer of bare sample was a structure of multi-layer and iron elements distributed uniformly within every layer, which was different with the diffusion of Fe ions within the coatings. The EDS lining micrograph picture of cross section of the coated samples and bare samples at 1100°C for different duration time were shown in Fig. 10. The result indicated that the coating could effectively prevent the external diffusion of Fe ions, and with the increase of the duration time, the inhibition effect of the coating on external diffusion of Fe ions became more obvious.

Fig. 8.

The EDS result and SEM micrograph picture of cross section of the descaled coating at 1100°C: (a) cordierite particle within the original coating, (b) cordierite particle within the transition layer, (c) Si-rich phase within the original coating, (d) Si-rich phase within the transition layer. (Online version in color.)

Fig. 9.

The EDS mapping micrograph picture of cross section of coated samples and bare samples at 1100°C for different duration time: (a-1) coated sample for 0 min, (a-2) bare sample for 0 min, (b-1) coated sample for 30 min, (b-2) bare sample for 30 min, (c-1) coated sample for 60 min, (c-2) bare sample for 60 min, (d-1) coated sample for 120 min, (d-2) bare sample for 120 min. (Online version in color.)

Fig. 10.

The EDS lining micrograph picture of cross section of coated samples and bare samples at 1100°C for different duration time. (Online version in color.)

The effective process of the coating could be summarized as follow. Since colloidal silica was used as binder agent, amorphous primary glass powders could be attached to the substrate by mechanical occlusion. When the temperature increased, the primary glass powders began to melt on the surface of the substrate. The substrate was covered by a compact molten film, which was generated by the viscous flow densification of the amorphous phase. Although some pores occurred, the substrate was still insulated from the oxygen. As temperature continued to rise, the crystalline phase generated and grew within the glass melt. The proportion of amorphous phase started to decrease and the molten film turned into a ceramic layer. As a result, the external diffusion of iron ions was impeded by the crystalline phase.

4. Conclusions

In this paper, glass-ceramics were successfully prepared with coal gangue and vanadium-titanium slag as principal raw materials, and the properties and the crystallization kinetics of the glass-ceramic were characterized. A new type of coating was prepared with the as-prepared primary glass powders as raw materials, and the anti-oxidation ability of the coating was tested. In addition, a probably protective mechanism of glass-ceramic coating was discussed with these results of SEM and EDS. The following conclusions could be drawn from this study.

(1) With the increasing proportion of VTS in raw material, the main crystalline phase transformed from cordierite to spinel and hercynite, and the crystallization activation of the primary glass sample energy increased.

(2) Glass-ceramic coatings prepared by primary glass with higher proportion of VTS in raw materials exhibited better anti-oxidation performance during temperature-rising stage, while the coatings with lower proportion of VTS in raw materials showed better protective effect during the temperature-holding process. Combined with the non-isothermal and isothermal anti-oxidation performance, the optimum anti-oxidation effect of glass-ceramic coating was 74.4% when the proportion of VTS was 6% by weight.

(3) A compact molten film which was generated by the viscous flow densification of the amorphous phase contributed to prevent the oxidation caused by the direct contact between oxygen and the substrate. According to the fact that Fe ions could only partially dope into cordierite by replacing Mg ions and Al ions or diffuse through the Si-rich phase between cordierite particles within the coating, the external diffusion of Fe ions was slowed down by the crystalline phase.

Acknowledgements

This work was supported by the “The Key Research Program of the Chinese Academy of Sciences” (Grant No. ZDRWZS201812).

References

• 1)   K. Y.  Frolenkov,  L. Y.  Frolenkova and  I. F.  Shadrin: Prot. Met. Phys. Chem. Surf., 46 (2010), 103. https://doi.org/10.1134/S2070205110010156
• 2)   M. D.  Nguyen,  J. W.  Bang,  Y. H.  Kim,  A. S.  Bin,  K. H.  Hwang,  V. H.  Pham and  W. T.  Kwon: Ceram. Int., 44 (2018), 8306. http://doi.org/10.1016/j.ceramint.2018.02.017
• 3)   G. Y.  Fu,  L. Q.  Wei,  X. M.  Zhang,  Y.  Cui,  Y. L.  Wang,  B.  Yu,  C. C.  Lv and  S. F.  Ye: ISIJ Int., 58 (2018), 929. https://doi.org/10.2355/isijinternational.ISIJINT-2017-602
• 4)   X. J.  Hu,  B. M.  Zhang,  S. H.  Chen,  F.  Fang and  J. Q.  Jiang: J. Iron Steel Res. Int., 20 (2013), 47. https://doi.org/10.1016/S1006-706X(13)60043-6
• 5)   P. A.  Munther and  J. G.  Lenard: J. Mater. Process. Technol., 88 (1999), 105. https://doi.org/10.1016/S0924-0136(98)00392-6
• 6)   G. Y.  Fu,  L. Q.  Wei,  X.  Shan,  X. M.  Zhang,  J.  Ding,  C. C.  Lv,  Y.  Liu and  S. F.  Ye: Surf. Coat. Technol., 294 (2016), 8. https://doi.org/10.1016/j.surfcoat.2016.03.053
• 7)   G. Y.  Fu,  L. Q.  Wei,  X. M.  Zhang,  Y. B.  Cui,  C. C.  Lv,  J.  Ding,  B.  Yu and  S. F.  Ye: Surf. Coat. Technol., 310 (2017), 166. https://doi.org/10.1016/j.surfcoat.2016.12.084
• 8)   M.  Torkar,  B.  Glogovac,  F.  Kaucic,  D.  Finzgar and  S.  Triglav: J. Mater. Process. Technol., 58 (1996), 217. https://doi.org/10.1016/0924-0136(95)02126-4
• 9)   M. H.  Chen,  W. B.  Li,  M. L.  Shen,  S. L.  Zhu and  F. H.  Wang: Corros. Sci., 82 (2014), 316. https://doi.org/10.1016/j.corsci.2014.01.033
• 10)   S. C.  Kuiry,  S. K.  Roy and  S. K.  Bose: Oxid. Met., 41 (1994), 65. https://doi.org/10.1007/BF01196643
• 11)   D. C.  Li,  N. N.  Han,  L.  Quagliato,  N.  Kim and  G.  Yu: Adv. Mater. Sci. Eng., 2019 (2019), 3864765. https://doi.org/10.1155/2019/3864765
• 12)   M. H.  Chen,  M. L.  Shen,  X.  Wang,  S. L.  Zhu and  F. H.  Wang: Surf. Coat. Technol., 216 (2013), 145. https://doi.org/10.1016/j.surfcoat.2012.11.045
• 13)   M.  Guenthner,  T.  Kraus,  A.  Dierdorf,  D.  Decker,  W.  Krenkel and  G.  Motz: J. Eur. Ceram. Soc., 29 (2009), 2061. https://doi.org/10.1016/j.jeurceramsoc.2008.11.013
• 14)   X.  Zhou,  S. F.  Ye,  H. W.  Xu,  P.  Liu,  X. J.  Wang and  L. Q.  Wei: Surf. Coat. Technol., 206 (2012), 3619. https://doi.org/10.1016/j.surfcoat.2012.02.055
• 15)   X.  Zhou,  L. Q.  Wei,  P.  Liu,  X. J.  Wang,  S. F.  Ye and  Y. F.  Chen: Chin. J. Process Eng., 10 (2010), 167 (in Chinese). http://www.corc.org.cn/handle/1471x/857644, (accessed 2020-01-10).
• 16)   H.  Odashima and  M.  Kitayama: ISIJ Int., 30 (1990), 255. https://doi.org/10.2355/isijinternational.30.255
• 17)   M. H.  Chen,  W. B.  Li,  M. L.  Shen,  S. L.  Zhu and  F. H.  Wang: Corros. Sci., 74 (2013), 178. https://doi.org/10.1016/j.corsci.2013.04.041
• 18)   A.  Majumdar and  S.  Jana: Bull. Mater. Sci., 24 (2001), 69. https://doi.org/10.1007/bf02704843
• 19)   C.  Li,  J. H.  Wan,  H. H.  Sun and  L. T.  Li: J. Hazard. Mater., 179 (2010), 515. https://doi.org/10.1016/j.jhazmat.2010.03.033
• 20)   N.  Zhang,  H. H.  Sun,  X. M.  Liu and  J. X.  Zhang: J. Hazard. Mater., 167 (2009), 927. https://doi.org/10.1016/j.jhazmat.2009.01.086
• 21)   Z.  Li,  Z. W.  Luo,  X. Y.  Li,  T. Y.  Liu,  L. M.  Guan,  T.  Wu and  A. X.  Lu: J. Porous Mater., 23 (2016), 231. https://doi.org/10.1007/s10934-015-0074-y
• 22)   J. M.  Wang,  Q.  Qin,  S. J.  Hu and  K. N.  Wu: J. Clean. Prod., 112 (2016), 631. https://doi.org/10.1016/j.jclepro.2015.07.138
• 23)   D. X.  Li,  X. Y.  Song,  C. C.  Gong and  Z. H.  Pan: Cem. Concr. Res., 36 (2006), 1752. https://doi.org/10.1016/j.cemconres.2004.11.004
• 24)   Y. L.  Ji,  S. B.  Shen,  J. H.  Liu and  Y.  Xue: J. Clean. Prod., 149 (2017), 1068. https://doi.org/10.1016/j.jclepro.2017.02.177
• 25)   G. B.  Sadykhov and  I. A.  Karyazin: Russ. Metall., 6 (2007), 447. https://doi.org/10.1134/S0036029507060018
• 26)   D. S.  Chen,  L. S.  Zhao,  T.  Qi,  G. P.  Hu,  H. X.  Zhao,  J.  Li and  L. N.  Wang: Trans. Nonferrous Met. Soc. China, 23 (2013), 3076. https://doi.org/10.1016/S1003-6326(13)62836-8
• 27)   C. S.  Fan and  K. C.  Li: J. Clean. Prod., 57 (2013), 335. https://doi.org/10.1016/j.jclepro.2013.06.002
• 28)   J. F.  Lv,  H. Y.  Yang,  Z. N.  Jin and  M. L.  Zhao: Waste Manag., 76 (2018), 687. https://doi.org/10.1016/j.wasman.2018.03.019
• 29)   P. R.  Monich,  A. R.  Romero,  D.  Hoellen and  E.  Bernardo: J. Clean. Prod., 188 (2018), 871. https://doi.org/10.1016/j.jclepro.2018.03.167
• 30)   I.  Ponsot and  E.  Bernardo: J. Clean. Prod., 59 (2013), 245. https://doi.org/10.1016/j.jclepro.2013.06.029
• 31)   I.  Ponsot,  E.  Bernardo,  E.  Bontempi,  L.  Depero,  R.  Detsch,  R. K.  Chinnam and  A. R.  Boccaccini: J. Clean. Prod., 89 (2015), 224. https://doi.org/10.1016/j.jclepro.2014.10.091
• 32)   P.  Sarfo,  A.  Das,  G.  Wyss and  C.  Young: Waste Manag., 70 (2017), 272. https://doi.org/10.1016/j.wasman.2017.09.024
• 33)   S.  Varitis,  P.  Kavouras,  E.  Pavlidou,  E.  Pantazopoulou,  G.  Vourlias,  K.  Chrissafis,  A. I.  Zouboulis,  T.  Karakostas and  P.  Komninou: Waste Manag., 59 (2017), 237. https://doi.org/10.1016/j.wasman.2016.10.011
• 34)   A.  Rincon,  D.  Desideri and  E.  Bernardo: J. Clean. Prod., 187 (2018), 250. https://doi.org/10.1016/j.jclepro.2018.03.065
• 35)   R. K.  Chinnam,  A. A.  Francis,  J.  Will,  E.  Bernardo and  A. R.  Boccaccini: J. Non-Cryst. Solids, 365 (2013), 63. https://doi.org/10.1016/j.jnoncrysol.2012.12.006
• 36)   S. M.  Wang and  K. N.  Liang: Chemosphere, 69 (2007), 1798. https://doi.org/10.1016/j.chemosphere.2007.06.016
• 37)   M.  Romero,  J. M.  Rincon and  A.  Acosta: J. Eur. Ceram. Soc., 22 (2002), 883. https://doi.org/10.1016/S0955-2219(01)00395-8
• 38)   S. M.  Salman,  S. N.  Salama and  E. A.  Mahdy: Ceram. Int., 44 (2018), 14587. https://doi.org/10.1016/j.ceramint.2018.05.079
• 39)   P.  Amista,  M.  Cesari,  A.  Montenero,  G.  Gnappi and  L.  Lan: J. Non-Cryst. Solids, 192–193 (1995), 529. https://doi.org/10.1016/0022-3093(95)00451-3
• 40)   M. H.  Lewis,  J.  Metcalf-Johansen and  P.  Bell: J. Am. Ceram. Soc., 62 (1979), 278. https://doi.org/10.1111/j.1151-2916.1979.tb09481.x
• 41)   E. A.  Marseglia: J. Non-Cryst. Solids, 41 (1980), 31. https://doi.org/10.1016/0022-3093(80)90188-X
• 42)   Y. J.  Park and  J.  Heo: Ceram. Int., 28 (2002), 669. https://doi.org/10.1016/S0272-8842(02)00026-3
• 43)   A. S.  Nagelberg: J. Mater. Res., 7 (1992), 265. https://doi.org/10.1557/JMR.1992.0265
• 44)   Z.  Jiang: Hot Work. Technol., 44 (2015), 89 (in Chinese). http://doi.org/10.14158/j.cnki.1001-3814.2015.20.024