Carbothermal Reduction of Stainless Steel Dust and Laterite Nickel Ore: Slag Phase Behavior Regulation and Self-pulverization Control Mechanism

Abstract

Stainless steel dust is a type of high basicity iron containing solid waste from the iron and steel smelting industry. In this study, the carbon-containing composite briquettes were prepared by adding low basicity laterite nickel ore and stainless steel dust, and the valuable metals Fe, Cr, and Ni were recovered by carbothermal reduction. The results show that the addition of laterite nickel ore into CBSL (carbon containing composite briquette of stainless steel dust and laterite nickel ore) effectively improves the recovery of metal Fe, Cr and Ni and the self-pulverization rate of reduction slag. The reduction temperature is 1450°C, the reduction time is 30 min, the FC/O (molar ratio of fixed carbon to oxygen) is 1.0, and the laterite nickel ore addition is 6%, the recoveries of metal Fe, Cr and Ni are 92.1%, 90.1% and 91.6%, respectively, and the self-pulverization rate of the reduction slag reaches 91.5%. The Ca₂SiO₄ phase content of the reduced slag at high temperature is 71.8%, and the reduced slag system is located in the orthosilicate Ca₂SiO₄ region of the phase diagram, which effectively improves metal recovery, enables efficient separation of the reduced product and reduces energy consumption for subsequent processing.

1. Introduction

The gradual consumption of iron ore and the large amount of CO₂ emission are the major problems to be solved under the concept of sustainable development.^1,2,3) In view of the current metal smelting industry, actively developing efficient smelting processes and recycling wastes are effective ways to effectively reduce resource consumption and alleviate ore depletion.^4,5,6) As a harmful secondary solid waste resource in the iron and steel industry, stainless steel dust contains a large amount of valuable metal Fe, Cr and Ni oxides, and has high recovery value.^7,8,9,10,11) However, due to its high basicity, there are still some problems to be solved in a large number of treatments at present.

At present, pyrometallurgical process is the main method to treat iron containing solid waste.^12,13,14,15) In the reduction process for the treatment of iron containing solid waste, in addition to the decisive effect of reduction conditions on the metal recovery, the composition of reduction slag also has a significant effect on the metal reduction and subsequent treatment process.^{16,17,18,19,20,21)}

Tang et al. carried out the thermal reduction experiment of mixing and compacting iron containing dust and coal into pellets at high temperature. The effects of C/O ratio (carbon oxygen ratio), slag basicity and temperature on metal recovery and reduction product separation efficiency were studied. It has been shown that the reduced product can be naturally separated with the slag basicity greater than 1.80, and that the optimal C/O ratios for the slag are 1.00 and 2.00, respectively. The Fe recovery ratio can reach more than 95% after the reduction temperature is higher than 1350°C.²²⁾ Zhang et al. studied the reaction mechanism of smelting slag through thermodynamic analysis and coal based direct reduction experiment. The recoveries of metal Fe are 91.55%, 44.53% and 88.48%, respectively, corresponding to the basicity of slag of 1.00, 2.00 and 3.85, respectively. The basicity of the slag is 3.85, the reducing slag is mainly composed of C₂S and C₂A₇, with a mass fraction of about 85%.²³⁾ Wu et al. prepared stainless steel slag systems with different basicity using chemical reagent powder and carried out melting crystallization experiments at 1400°C. The slag with high basicity (1.25 or 1.50) is easier to form spinel phase, and the slag with low basicity (1.00) is easier to form amorphous structure, which is conducive to the precipitation of Cr₂O₃ into Mg(Al, Fe, Cr)₂O₄ spinel solid solution and metal reduction.²⁴⁾ Kapure et al. studied the process of recovering Fe, Ni and Cr from Sujinda chromite aggregate by carbon composites at high temperature. The recoveries of Fe, Ni and Cr were more than 90%, 90% and 40%, respectively.²⁵⁾ Wu et al. prepared a synthetic stainless steel slag system with different basicity by using reagent powder, and carried out a crystallization experiment from melt to 1400°C in N₂ atmosphere. The results show that the low basicity is beneficial to the precipitation of Cr₂O₃ into Mg(Al, Fe, Cr)₂O₄ spinel solid solution, and to the reduction of metals.²⁶⁾ Liu et al. carried out research on the preparation of Fe–Cr–Ni–C alloy by stainless steel dust and laterite nickel ore. The ratio of stainless steel dust to laterite nickel ore is 94%: 6%, the reduction temperature is 1400°C, the reduction time is 20 min, and the FC/O is 0.8, the metal and slag of the reduced product can be naturally separated after cooling; The recoveries of Fe, Cr and Ni were 90.6%, 90.1% and 91.2% respectively. The Fe, Cr and Ni grades in the Fe–Cr–Ni–C alloy are 62.7%, 18.9% and 4.1% respectively. The alloy has low levels of the harmful elements S and P.²⁷⁾ Chen et al. prepared carbonaceous pellets with zinc-containing dust and chromium-containing sludge as the main ingredients. Thermodynamic analysis and XRD were used to characterize the mechanism of carbothermal reduction. Based on the non-isothermal thermogravimetry, the kinetics of the reduction process at heating rates of 15°C/min, 20°C/min and 25°C/min was stuied. The results show that ZnFe₂O₄ and FeCr₂O₄ phases can be effectively decomposed into iron oxide and chromium oxide at 1000°C. The kinetic results show that the reaction process can be divided into five stages, the main three of which follow the three-dimensional diffusion model.²⁸⁾ Liu et al. investigated a low-energy and high-efficiency direct reduction process for stainless steel dust and chromium-containing slag. The optimum process conditions are as follows: the ratio of stainless steel dust to chromium slag is 94%: 6%, the reduction temperature is 1400°C, the reduction time is 25 min, the FC/O is 0.9, and the recovery ratios of Fe, Cr and Ni in the reduction products are 90.5%, 90.6% and 91.2%, respectively. The self-pulverization rate of the reduced slag reached 92.8%, and the residuals of the metals Fe, Cr and Ni in the reduced slag were 134 ppm, 96 ppm and 129 ppm, respectively.²⁹⁾ Zheng et al. explored the possibility of producing fine near-spherical metal particles by carbothermal reduction of stainless steel dust. The high temperature and long reduction time favor stainless steel reduction. After 10 min at 1300°C, the reduction is almost complete. The primary M₇C₃ carbide and residual oxide are responsible for the irregular metal particles. The chemical composition of the residual oxides affects the size of the metal particles. The metal particles surrounded by SiO₂-rich oxides are much larger than those surrounded by MgO-rich oxides.³⁰⁾ Wang et al. prepared strong oxidation pellets by SRC-EF (Recycling of SSSW through carbothermal reduction of chromite-electric furnace) process using SSSW (stainless steel solid waste) and ferrochrome concentrate. The hardening properties and migration behavior of hazardous elements during pellet oxidation have been investigated and revealed. It was shown that 100% SSSW oxidized pellets have low mechanical strength. With preheating at 1000°C for 12 min and roasting at 1275°C for 15 min, increasing the proportion of chromite concentrate from 0% to 80% can significantly increase the pellet strength, from 460 N/pellet to 1868 N/pellet. Oxidized pellets prepared from SSSW and chromite concentrate are predominantly solid-phase consolidated. For the pellets composed of 32% SSSW and 68% chromite concentrate, 39.96% F, 91.06% S and 66.73% Cl can be removed by preheating at 1000°C for 12 min and roasting at 1275°C for 15 min.³¹⁾ In general, the current treatment processes have the problems of high energy consumption and low metal recovery ratio.^32,33,34,35) Laterite nickel ore, as a kind of low basicity, low grade and refractory ore, also has the problems of low metal yield and high energy consumption in the smelting process.^36,37,38,39)

In this study, stainless steel dust was used as the raw material and laterite nickel ore was used as an additive to adjust the basicity of the slag component in the reduction process, and a composite carbon containing compact was prepared. Laterite nickel ore was added at 0%–10%. Efficient recovery of Fe, Cr and Ni from metal oxides is achieved by carbothermal reduction. The carbothermal reduction temperature is 1450°C, the reduction time is 30 min, and the FC/O is 1.0. At the same time, the effects of the addition of laterite nickel ore on the metal recovery and slag phase self-pulverization rate were investigated, and the phase composition of the reduction products was analyzed by XRD (X-ray diffraction). At the same time, the phase composition of the slag phase was calculated and analyzed using the thermodynamic computing software Factsage7.2 for different laterite nickel ore additions during the reduction process.

2. Experiment

2.1. Raw Materials

The stainless steel dust, laterite nickel ore and reduction coal used in the experiment came from Taiyuan Iron and Steel Company, Indonesia and Chengde Iron and Steel Company, respectively. The contents of the main chemical components of stainless steel dust and laterite nickel ore were obtained by chemical quantitative analysis and are listed in Tables 1 and 2, respectively, and the results of industrial analysis of reduction coal are listed in Table 3.

Table 1. Chemical composition of stainless steel dust (wt/%).

	Fe2O3	FeO	Cr2O3	NiO	CaO	SiO2	MgO	Al2O3	ZnO	P2O5	SO3
Average	38.09	18.67	13.20	2.73	15.01	4.15	2.87	1.13	0.35	0.03	0.28
Error	±0.3%	±0.3%	±0.3%	±0.3%	±0.3%	±0.3%	±0.3%	±0.05%	±0.02%	±0.0001%	±0.001%
Relative standard deviation	3%	3%	3%	3%	3%	3%	3%	3%	3%	3%	3%

Table 2. Chemical composition of laterite nickel ore (wt/%).

	Fe2O3	FeO	Cr2O3	NiO	CaO	SiO2	MgO	Al2O3	ZnO	P2O5	SO3	CoO
Average	40.74	0.21	1.40	2.31	0.09	34.78	12.98	4.00	0.51	0.01	0.07	0.26
Error	±0.3%	±0.3%	±0.3%	±0.3%	±0.3%	±0.3%	±0.3%	±0.05%	±0.02%	±0.0001%	±0.001%	±0.02%
Relative standard deviation	3%	3%	3%	3%	3%	3%	3%	3%	3%	3%	3%	3%

Table 3. Industrial analysis and ash analysis of reduction coal (wt/%).

	FC	Volatile	Ash	Moisture	Ash analysis
					CaO	SiO2	MgO	Al2O3	Fe2O3	Others
Average	60.58	28.47	10.20	0.75	10.33	40.82	2.92	30.87	5.89	9.17
Error	±0.3%	±0.3%	±0.3%	±0.3%	±0.3%	±0.3%	±0.3%	±0.3%	±0.3%	–
Relative standard deviation	3%	3%	3%	3%	3%	3%	3%	3%	3%	–

As can be seen from Table 1, the stainless steel dust is primarily composed of 38.09 wt.% Fe₂O₃, 18.67 wt.% FeO, 13.20 wt.% Cr₂O₃, 2.73 wt.% NiO, 15.01 wt.% CaO and 4.15 wt.% SiO₂. The basicity is high, that is, the CaO/SiO₂ ratio is large. As can be seen from Table 2, the laterite nickel ore consists mainly of 40.74 wt.% Fe₂O₃, 0.21 wt.% FeO, 1.40 wt.% Cr₂O₃, 2.31 wt.% NiO, 0.09 wt.% CaO, and 34.78 wt.% SiO₂, with a low basicity of small CaO/SiO₂ ratio. It can be seen from Table 3 that the fixed carbon, volatile matter, ash and moisture in the reduced coal are 60.58 wt.%, 28.47 wt.%, 10.20 wt.% and 0.75 wt.%, respectively, and the ash contains 10.33 wt.% CaO, 40.82 wt.% SiO₂, 2.92 wt.% MgO, 30.87 wt.% Al₂O₃, 5.89 wt.% Fe₂O₃ and 9.17 wt.% other components.

The results of XRD analysis of stainless steel dust and laterite nickel ore are shown in Figs. 1(a) and 1(b). The main constituents of stainless steel dust are the oxides of the valuable metallic components Fe, Cr and Ni produced during smelting, mainly in the form of Fe₃O₄, Fe₂O₃, NiFe₂O₄, FeCr₂O₄ and Fe₂MgO₄. The state of occurrence of the phases in laterite nickel ore is relatively complex and occurs mainly in the form of complex substances coexisting with metals and gangue. The metal Fe, Cr and Ni mainly exist in the form of Fe₂O₃, Fe₃O₄, FeOOH, Ni₃Si₂O₅(OH)₄, Mg₃Fe₂O₅(OH)₄, Fe₃Si₃O₇(OH)₄, NiCr₂O₄ and Fe₂SiO₄.

Fig. 1.

XRD energy spectrum analysis of stainless steel dust (a) and laterite nickel ore (b). (Online version in color.)

2.2. Experimental Procedure

The flow of the process for the carbothermal reduction experiment is shown in Fig. 2. The experimental raw materials included stainless steel dust, laterite nickel ore and reduction coal. Put the experimental raw materials in the drying oven, set the temperature at 110°C to dry for 3 h, and pass the 200 mesh sieve after crushing and screening (ensure that the particle size of the raw materials is less than 74 μm). Fully mix 70 g of dried raw materials, weigh 7 g and place it in a stainless steel mold, heat it in a 200°C high temperature resistance furnace for 20 min, and press it into an oval cylindrical block with a diameter of 35 mm and a height of 20 mm under a pressure of 35 MPa (not used in the binder forming process), which is called the carbon containing composite briquette of stainless steel dust and laterite nickel ore (CBSL), and its average compressive strength exceeds 800 N/pellet. The basicity in the research is the binary basicity of the mass ratio between CaO and SiO₂ in CBSL, and the FC/O (molar ratio of fixed carbon to oxygen) refers to the molar ratio of fixed carbon in CBSL coal to oxygen in Fe, Cr and Ni metal oxides.

Fig. 2.

Process flow chart of carbothermal reduction experiment. (Online version in color.)

The experiments were performed in a vertical high temperature furnace. The maximum heat-resistant temperature of the high-temperature furnace is 1600°C, and the heating element is MoSi₂. At the same time, two thermocouples were equipped to measure the furnace body temperature and the experimental sample temperature, respectively. The experimental temperature measurement error is less than 1°C. After the set temperature is reached, the heat preservation treatment is performed according to the carbothermal reduction time. Wearing high-temperature fire-resistant gloves and high-temperature test protective clothing, the corundum crucible containing the CBSL sample was placed in the high-temperature measurement position of the standpipe high-temperature furnace and the carbothermal reduction test was performed with an iron long-handled clamp. Argon with a flow rate of 1.3 × 10⁻⁵ m³/s is blown into the corundum tube of the high temperature furnace to maintain an inert atmosphere (the purity of argon is more than 99.9%). After reaching the reduction time, wearing the same protective equipment, the crucible was removed with a long-handled clamp and quickly placed in an argon cooling tank filled with 10 L/min for rapid cooling. The reduced high-temperature furnace is located close to the argon cooling tank to reduce and ensure the same exposure time. After cooling to room temperature, the crucible is removed and the experiment is completed.

2.3. The Calculation Procedure for Metal Recovery and Self-pulverization

In this study, the recovery ratio of metal Fe, Cr and Ni and the self-pulverization rate of reduced slag were investigated during the carbothermal reduction of stainless steel dust and laterite nickel ore. The calculation formula of metal Fe, Cr, Ni recovery ratio and reduced slag self-pulverization rate is as follows:   

$$R_r=\frac{m_r}{M_r}\times 100\%$$(1)

Where R_r represents the Fe, Cr and Ni metal recovery ratio, m_r represents the mass of metal in the reduced product, and M_r represents the mass of metal in the raw material.   

$$S_s=\frac{m_s}{M_s}\times 100\%$$(2)

Where S_s represents the self-pulverization rate of the reduced slag, m_s represents the mass of slag with granularity less than 74 μm, and M_s represents the total mass of slag.

2.4. Detection Method

20 g stainless steel dust and laterite nickel ore were randomly selected, weighed, and used potassium dichromate titration, wavelength dispersive X-ray fluorescence spectrometry, flame atomic absorption spectrometry, high-frequency combustion infrared absorption spectroscopy, and chemical instrument X-ray fluorescence spectrometer to calculate the mass loss in a professional research institute of chemical analysis (Analysis and testing center of Northeastern University, China), The chemical composition of stainless steel dust and laterite nickel ore was quantitatively analyzed and evaluated. The testing equipment mainly includes the X-ray fluorescence spectrometer (ZSXPrimus II) of Rigaku Corporation, the flame atomic absorption spectrometer (Z-2300) of Hitachi in Japan, and the infrared carbon sulfur analyzer (CS230) of LECO in the United States.

The reduction product is crushed, ground and sieved through a 200 mesh sieve (hole diameter is 74 μm). The enrichment of the metals Fe, Cr, and Ni in the reduced products is also obtained by the above detection, and the recovery ratios of the metals Fe, Cr, and Ni are then obtained by weighing and computing. Use a 200 mesh sieve (hole diameter is 74 μm), the reduction products are sieved, and then the self-pulverization rate of the reduction slag is calculated by balance weighing and formula. The reduction product is broken and sieved through a 200 mesh sieve (hole diameter is 74 μm) , placed on the sample carrier plate and flattened, and analyzed the phase composition with X-ray diffractometer (MPDDY2094, PANalytical B.V., Almelo, Netherlands). Copper Ka radiation (40 kV, 40 mA, wavelength 0.154 nm) is used as the X-ray source, the scanning angle is 5° to 90°, and the scanning speed is 0.2°/s.

2.5. Thermodynamic Calculation of Metal Oxide Reduction

The oxides of metal Fe, Cr and Ni are the valuable metal oxides mainly existing in stainless steel dust and laterite nickel ore, and the reduction reaction formula and basic thermodynamic data of CO reducing Fe, Cr and Ni compounds mainly involved in the reduction process are listed in Table 4. The Gibbs free energy change and gas phase equilibrium diagram of the reduction reaction of Fe, Cr and Ni compounds were drawn using the reaction module in the thermodynamic software Factsage 7.2, as shown in Figs. 3(a) and 3(b) respectively. It can be seen from Fig. 3(b) that the reduction reaction between Cr oxide and CO requires very high CO/(CO+CO₂) concentration, so the reduction reaction of Cr compounds can only be carried out when the CO partial pressure is very high. Combined with the data obtained from thermodynamic calculations, it can be seen that oxides of the valuable metals Fe, Cr, and Ni can be efficiently reduced at sufficient reduction temperatures and in a reduced atmosphere.

Table 4. Reduction reaction equation and thermodynamic data of Fe, Cr and Ni compounds and CO.

Reactions	ΔG=f (T), J/mol	Number
3Fe2O3+CO=2Fe3O4+CO2	−36750−51.37·T	(3)
Fe3O4+CO=3FeO+CO2	187.8−13.95·T+9397000/T	(4)
1/4Fe3O4+CO=3/4Fe+CO2	−12270+12.64·T+1414000/T	(5)
FeO+CO=Fe+CO2	−19640+23.3·T	(6)
FeCr2O4+CO=Fe+Cr2O3+CO2	59800+2.947·T−3331000/T	(7)
7Cr2O3+33CO=2Cr7C3+27CO2	18594+30.94·T	(8)
Cr2O3+3CO=2Cr+3CO2	90933+3.23·T−72867/T	(9)
NiO+CO=Ni+CO2	50460+2.52·T+0.0001238·T2+1249000/T	(10)
0.75NiFe2O4+CO=0.75Ni+0.5Fe3O4+CO2	−28430−16.56·T	(11)

Fig. 3.

Thermodynamic calculation of metal oxide in the stainless steel dust and laterite nickel ore. (Online version in color.)

3. Results and Discussion

3.1. Effect of Laterite Nickel Ore on Metal Recovery of CBSL

The carbothermal reduction temperature is set at 1450°C, the carbothermal reduction time is 30 min, and the carbon oxygen molar ratio FC/O is 1.0. The influence trend of different laterite nickel ore addition on the recovery of metal Fe, Cr and Ni in the carbon bearing briquette CBSL formed by stainless steel dust and laterite nickel ore is shown in Fig. 4. The addition of laterite nickel ore in CBSL can effectively reduce the overall basicity. At this point, the basicity is the binary basicity of the ratio of all CaO and SiO₂ in stainless steel dust, laterite nickel ore and reduction coal. The total binary basicity in the CBSL are 3.0, 2.6, 2.3, 2.1, 1.9 and 1.7, respectively, corresponding to the conditions that the laterite nickel ore addition are 0%, 2%, 4%, 6%, 8% and 10%, respectively.

Fig. 4.

Effect of laterite nickel ore additions on metal recovery of CBSL. (Online version in color.)

Figure 4 shows that each reduction condition provides the maximum reduction efficiency under the same reduction procedure conditions. The recovery ratios for the metals Fe, Cr, and Ni show a gradual increase at first followed by a sharp decrease as the addition of laterite nickel ore increases. The recoveries of metals Fe, Cr and Ni increased from 83.6%, 81.3% and 86.9% to 92.1%, 90.1% and 91.6%, respectively, before declining to 81.8%, 82.4% and 82.6%, respectively, with the addition of laterite nickel ore increasing from 0% to 6% and then to 10%. The recovery ratios for the metals Fe, Cr, and Ni reached their maximum values, respectively, with the addition of the laterite nickel ore at 6%. The addition of laterite nickel reduces the overall basicity of the reduction system and modulates the slag phase properties to a certain extent. The change in basicity affects the difficulty of the metal oxide reduction process. The recovery ratio of the metal in the metal oxide was effectively enhanced by the addition of 6% laterite nickel ore at the appropriate base conditions.

3.2. Effect of Basicity on Phase Transition

The XRD phase analysis method was used to analyze the composition of the reduced product and clarify the effect of the basicity change on the phase transition of the reduced product under different conditions for the addition of laterite nickel ore. The XRD spectrum results after reduction at FC/O 1.0 and 1450°C for 30 min with different laterite nickel ore additions are shown in Fig. 5.

Fig. 5.

XRD analysis of reduction phase change of CBSL with different laterite nickel ore additions at FC/O 1.0, 1450°C for 30 min. (Online version in color.)

Figure 5 shows that with the increase of the proportion of laterite nickel ore, the significant changes of diffraction peaks of each phase are mainly concentrated in the slag phase of the reduction product, and the main slag phases of the reduction product are Ca₃SiO₅, Ca₂SiO₄ and Ca₃Mg(SiO₄)₂. The metal phases in the reduction products are mainly Fe, (Fe,Cr)₇C₃ and Fe–Cr–Ni–C. As the amount of laterite nickel ore added increases, the overall basicity of the raw material decreases and the corresponding Ca₂SiO₄ diffraction peak intensity increases in the slag phase of the reduced product. The amount of laterite nickel ore added amounted to 6%, and the grade was reduced to 2.1. A large number of Ca₂SiO₄ diffraction peaks are present in the analytical results. With the continuous increase of laterite nickel ore addition, the basicity decreases to 1.9 and 1.7, and the low melting point compound Ca₃Mg(SiO₄)₂ appears in the slag phase of the reduction product. The presence and content of the low melting point phase during high temperature reduction will improve the fluidity of the reduced product. A good fluidity will prevent the aggregation and growth of the metallic phase during the reduction process, as well as the separation of the metallic and slag phases in the reduction product. The Fe–Cr–Ni and residual C phases appear after the metallic phases in the reduced products are reduced to 1.9 and 1.7, which is also the main reason for the sharp drop in the corresponding metallic recovery.

3.3. Slag Phase Reaction Mechanism

The slag phase in stainless steel dust is dominated by CaO, SiO₂, MgO, and Al₂O₃, with high CaO and SiO₂ content. The stable compounds Ca₂SiO₄ and Ca₃SiO₅, formed from CaO and SiO₂, are the main constituents of the slag phase at high temperatures, and they are also key factors in achieving the final efficient separation of the reduced product. Table 5 lists the main slag phase reactions and thermodynamic data, and the corresponding variation of the Gibbs free energy is shown in Fig. 6.

Table 5. The main reduction chemical reactions and thermodynamic parameters of the slag phase.

Reactions	ΔG=f (T), J/mol	Number
2CaO+SiO2=Ca2SiO4	−128300+1.186·T, T<1120 K −92640−30.33·T, T>1120 K	(12)
Ca2SiO4+CaO= Ca3SiO5	76470−64.01·T, T<1160 K 8163−5.16·T, T>1160 K	(13)
3CaO+SiO2=Ca3SiO5	−68540−44.53·T	(14)
2CaCO3+SiO2=Ca2SiO4+2CO2	217200−295.4·T	(15)
3CaCO3+SiO2=Ca3SiO5+3CO2	444100−473.7·T	(16)
CaCO3=CaO+CO2	80960−68.77·T	(17)

Fig. 6.

Gibbs free energy of main chemical reactions of slag. (Online version in color.)

The Gibbs free energy ΔG change plot corresponding to the main reaction of the slag phase during the carbothermal reduction of stainless steel dust and laterite nickel ore is shown in Fig. 6. The Gibbs free energy of all reactions decreases with the increase of temperature in the selected temperature range (0–2000°C). The Gibbs free energy of reactions (12) and (14) is less than 0 in the whole temperature range, indicating that the reactions (12) and (14) have been spontaneous and forward. The Gibbs free energy ΔG of reaction 3 is less than 0 as the temperature is higher than 897°C, and the reaction proceeds forward. At the same time, it reacts with SiO₂ for (15) and (16) to form Ca₂SiO₄ and Ca₃SiO₅. The Gibbs free energy ΔG of reaction (17) is less than 0 as the temperature is higher than 1303°C, the reaction proceeds in a forward direction, and CaCO₃ decomposes. Throughout the high temperature reactions, reactions (15) and (16) are easy to occur, and Ca₂SiO₄ and Ca₃SiO₅ are easy to form. The Gibbs free energy of reaction (13) is small. Although it is less than 0 at high temperatures, it is spontaneous in the forward direction, but Ca₃SiO₅ is not easy to form. In general, Ca₂SiO₄ and Ca₃SiO₅ are easily formed in the reaction process of the slag phase, which is also the basis for the efficient separation of slag and metal by carbothermal reduction of stainless steel dust and laterite nickel ore.

3.4. Self-pulverization Effects of Reduced Product Slag under Different Basicity Conditions

Define that the percentage of the mass with particle size less than 74 μm in the reduced slag product to the mass of the reduced slag is the self-pulverization rate of the reduced slag. Under the conditions of reduction temperature of 1450°C, reduction time of 30 min and FC/O of 1.0, the change of the self-pulverization rate of the carbothermal reduction product slag under the conditions of different laterite nickel ore addition is shown in Fig. 7.

Fig. 7.

Effect of the addition of laterite nickel ore on the self-pulverization rate of CBSL reduced slag. (Online version in color.)

As can be seen from Fig. 7, the basicity of the slag phase system gradually decreases with the addition of laterite nickel ore, the self-pulverization rate of the reduced slag under the same reduction system first increases and then decreases with the addition of laterite nickel ore. The self-pulverization rate of the reduced slag increased from 56.4% to 91.5% and then decreased to 69.9% as the laterite nickel ore increased from 0% to 6% and then increased to 10%.

The addition of the laterite nickel ore modifies the basic composition of the entire slag system. The self-pulverization of the reduced slag reached a maximum of 91.5% when the amount of laterite nickel ore added reached 6%. The slag phase composition in the raw material is dominated by CaO and SiO₂, and the basicity of the reduced slag is defined as CaO/SiO₂. The addition of laterite nickel ore up to 6% gives a corresponding slag phase basicity of 2.1 and a corresponding ratio of CaO to SiO₂ close to 2:1, which favors the formation of Ca₂SiO₄ and Ca₃SiO₅ in the reduced product. Based on the production of Ca₂SiO₄, the self-pulverization rate of the reduced slag was achieved. Ca₂SiO₄ is formed at high temperature, and the crystal form will change from α-Ca₂SiO₄ to β-Ca₂SiO₄ and then to γ-Ca₂SiO₄ during cooling. The cooling temperature is lower than 500°C, β-Ca₂SiO₄ will undergo a sharp transformation to γ-Ca₂SiO₄, and the density of β-Ca₂SiO₄ will change from 3320 kg/m³ to 2960 kg/m³ corresponding to γ-Ca₂SiO₄, resulting in volume expansion.⁴⁰⁾ In general, the composition of the slag phase is destroyed and a large amount of self-pulverization slag is generated, which can effectively separate the self-pulverization slag from the metal particles in the reduced product and effectively reduce the energy consumption during subsequent processing.

3.5. Thermodynamic Calculations for Different Reduced Slag System Compositions

Changes in the composition of the slag will directly affect the pulverization effect of the slag phase during the cooling of the reduced slag, and further affect the reduction recovery of metallic Fe, Cr, and Ni oxides. In order to show the phase composition of reduction slag system at high temperature under different ratio of stainless steel dust and laterite nickel ore, Factage 7 2 thermodynamic calculates and plots the theoretical composition of different CBSL slag system at 100–1500°C, as shown in Fig. 8.

Fig. 8.

Composition of CBSL theoretical slag system under different laterite nickel ore additions: (a): laterite nickel ore addition of 0%; (b): laterite nickel ore addition of 2%; (c): laterite nickel ore addition of 4%; (d): laterite nickel ore addition of 6%; (e): laterite nickel ore addition of 8%; (f): laterite nickel ore addition of 10%. (Online version in color.)

It is clear from Fig. 8 that the addition of laterite nickel ore has a significant effect on the change in the slag phase composition of the reduced product. In the temperature range of 100–420°C, the main phase is Ca₂SiO₄; in the temperature range of 420–1150°C, the main phase is Ca₇Mg(SiO₄)₄; in the higher temperature range of 1150–1420°C, the main phase is Ca₂SiO₄. The content of each phase changes significantly over the whole temperature range. It can be seen from Fig. 8(a) that the main products of CBSL slag phase under the high-temperature conditions without adding laterite nickel ore are Ca₂SiO₄, CaO, MgO and Al₂O₃ wherein the content of Ca₂SiO₄ is 56.7%. The presence of Ca₂SiO₄ is a key ingredient to ensure self-pulverization of the CBSL slag upon cooling. It can be seen from Fig. 8(b) that with the addition of laterite nickel ore, the content of Ca₂SiO₄, the main product phase of the reduction slag in the high-temperature section, increases significantly. In Fig. 8(e), the Ca₂SiO₄ content produced in the high temperature section reaches the maximum value of 76.9%, and the addition of laterite nickel ore reaches 8%. At the same time, it can be seen that the temperature range of high-temperature Ca₂SiO₄ generation is shorter, and the temperature range of a large number of low melting point compounds such as Ca₇Mg(SiO₄)₄ is expanding. It can be seen from Fig. 8(f) that with the addition of laterite nickel ore reaches 10%, the basicity decreases to 1.7, the content of Ca₂SiO₄ formed at high temperature decreases sharply, and a large number of low melting point compounds and liquid phases are generated, which hinders the pulverization of reduction slag during cooling. The thermodynamic calculation of the reduction slag shows that the basicity is 2.1 in Fig. 8(d), and the content of Ca₂SiO₄ generated in the corresponding slag phase at high temperature reaches 71.8%, the increase of Ca₂SiO₄ generation increases the amount of γ-Ca₂SiO₄ generated during the reduction process (<500°C), and promotes the self-pulverization rate of the reduction slag.

3.6. Phase Diagram Analysis of the Composition of the Slag System

The quaternary phase diagram of the CaO–SiO₂–MgO–Al₂O₃ system was plotted using the thermodynamic software Factsage7.2, and then the characteristics of the slag phase transition in stainless steel dust with different laterite nickel ore additions were clarified. The quaternary phase diagram of CaO–SiO₂–MgO–Al₂O₃ was drawn under the temperature range of 1000–2800°C, the addition of laterite nickel ore was 6%, and the corresponding Al₂O₃ was 4.47%, as shown in Fig. 9.

Fig. 9.

The phase diagram of CaO–SiO₂–MgO–Al₂O₃ slag system in 1000–2800°C. (Online version in color.)

Figure 9 shows the change trend of the different slag phase positions with decreasing basicity. As can be seen from the marked hexagons in Fig. 9, the corresponding positions of the cinders gradually shift from the bottom-left to the top-right corner of the phase diagram as the basicity decreases. The primary phases of the slag system corresponding to different laterite nickel ore additions are located in the monoxide region and Ca₂SiO₄ regions, with a general trend of a gradual transition from the monoxide to the Ca₂SiO₄ region. The basicity is high (the slag system composition positions corresponding to 3.0, 2.6, and 2.3), the corresponding position of the reducing slag composition is in the monoxide region; the basicity is low (the slag system composition positions corresponding to 2.1, 1.9, and 1.7), the corresponding position of the reducing slag composition is in the Ca₂SiO₄ orthosilicate region. According to the research results on the influence of basicity change on the self-pulverization rate of slag phase in Section 3.4, the basicity of slag system is 2.1 (the addition of laterite nickel ore is 6%), the self-pulverization rate of reduction slag reaches the maximum. It can also be seen from the thermodynamic calculation results of different slag system compositions in Section 3.5 that with the basicity of slag system is 2.1 (the addition of laterite nickel ore is 6%), the high-temperature reduction products of slag system are mainly Ca₂SiO₄. Therefore, in general, the basicity is 2.1 (the addition of laterite nickel ore is 6%), the theoretical composition of the corresponding slag system enters the normal silicate Ca₂SiO₄ area, which is conducive to the formation of Ca₂SiO₄ in the reduced product slag. The reduction of the basicity increases the generation of Ca₂SiO₄, improves the self-pulverization rate of the reduction slag, reduces the theoretical formation melting point of the slag phase, improves the fluidity of the reduction products, is conducive to the formation and growth of metals in the reduction process, and effectively improves the recovery ratios of the metals Fe, Cr, and Ni.

4. Conclusions

(1) The addition of laterite nickel ore to CBSL carbon bearing briquette improves the recovery of Fe, Cr and Ni after carbothermal reduction. The reduction temperature is 1450°C, the reduction time is 30 min, the carbon ratio FC/O is 1.0, the laterite nickel addition is 6%, and the recovery of metals Fe, Cr and Ni reaches 92.1%, 90.1% and 91.6% respectively.

(2) The addition of laterite nickel ore reduces the basicity of stainless steel dust and the slag system of carbon bearing briquette CBSL of laterite nickel ore, and improves the self-pulverization rate of reduction slag. The addition of laterite nickel ore is 6%, and the self-pulverization of the reduced slag reaches 91.5%. The metal alloy particles of the reduced product and the reduced slag can be naturally and efficiently separated, effectively reducing the energy consumption of the subsequent processing procedure.

(3) The basicity of the slag system of stainless steel dust and laterite nickel ore carbon bearing briquette CBSL is reduced to 2.1, and the theoretical amount of Ca₂SiO₄ in the reduction product under high temperature conditions reaches 71.8%. With the change of CBSL slag composition, after the basicity decreases to 2.1, the composition of the reduced slag system is located in the original silicate Ca₂SiO₄ region in the ternary phase diagram of CaO–SiO₂–MgO–Al₂O₃ system, which promotes the formation of Ca₂SiO₄ in the reaction process.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

The authors are especially grateful to National Natural Science Foundation of China (No. 51974077) and Xingliao Talent Plan (No. XLYC1902118), and special thanks are due to the instrumental analysis from Analytical and Testing Center, Northeastern University.

References

• 1)   A.  Habib,  H. N.  Bhatti and  M.  Iqbal: Z. Phys. Chem., 234 (2020), 201. https://doi.org/10.1515/zpch-2019-0001
• 2)   G.  Calvo,  A.  Valero and  A.  Valero: Resour. Conserv. Recycl., 125 (2017), 208. https://doi.org/10.1016/j.resconrec.2017.06.009
• 3)   J.  Spooren,  K.  Binnemans,  J.  Björkmalm,  K.  Breemersch,  Y.  Dams,  K.  Folens,  M.  González-Moya,  L.  Horckmans,  K.  Komnitsas,  W.  Kurylak,  M.  Lopez,  J.  Mäkinen,  S.  Onisei,  K.  Oorts,  A.  Peys, et al.: Resour. Conserv. Recycl., 160 (2020), 104919. https://doi.org/10.1016/j.resconrec.2020.104919
• 4)   G.  Kim and  I.  Sohn: J. Hazard. Mater., 359 (2018), 174. https://doi.org/10.1016/j.jhazmat.2018.07.053
• 5)   H. T.  Shen,  E.  Forssberg and  U.  Nordström: Resour. Conserv. Recycl., 40 (2004), 245. https://doi.org/10.1016/S0921-3449(03)00072-7
• 6)   H. W.  Zhang and  X.  Hong: Resour. Conserv. Recycl., 55 (2011), 745. https://doi.org/10.1016/j.resconrec.2011.03.005
• 7)   J. A.  de Araújo and  V.  Schalch: J. Mater. Res. Technol., 3 (2014), 274.
• 8)   J.  Peng,  B.  Peng,  D.  Yu,  M. T.  Tang,  H. C.  Song,  J.  Lobel and  J. A.  Kozinski: Trans. Nonferrous Met. Soc. China, 14 (2004), 593.
• 9)   J.  Peng,  B.  Peng,  D.  Yu,  M. T.  Tang,  N.  Souza,  J. A.  Kozinski and  J.  Lobel: J. Cent. South Univ. Technol., 10 (2003), 301. https://doi.org/10.1007/s11771-003-0028-4
• 10)   S.  Ri and  M. S.  Chu: ISIJ Int., 55 (2015), 1565. https://doi.org/10.2355/isijinternational.ISIJINT-2014-845
• 11)   J.  Rieger and  J.  Schenk: J. Sustain. Metall., 5 (2019), 295. https://doi.org/10.1007/s40831-019-00220-2
• 12)   H. M.  Ahmed,  N.  Viswanathan and  B.  Bjorkman: Steel Res. Int., 85 (2014), 293. https://doi.org/10.1002/srin.201300072
• 13)   X.  Zhang,  G. J.  Ma,  Y. B.  Jin and  P. H.  Cheng: Ceram. Int., 40 (2014), 9693. https://doi.org/10.1016/j.ceramint.2014.02.050
• 14)   Y. S.  Li,  Z. L.  Xue,  E.  Tang,  Q.  Liu,  W. X.  Wang and  D. N.  Zhao: Adv. Mater. Res., 524–527 (2012), 2031. https://doi.org/10.4028/www.scientific.net/AMR.524-527.2031
• 15)   S. C.  Ri,  M. S.  Chu,  S. Y.  Chen,  Z. G.  Liu and  H.  Hong: J. Iron Steel Res. Int., 23 (2016), 314. https://doi.org/10.1016/S1006-706X(16)30051-6
• 16)   A.  Kasai,  H.  Toyota,  K.  Nozawa and  S.  Kitayama: ISIJ Int., 51 (2011), 1333. https://doi.org/10.2355/isijinternational.51.1333
• 17)   E.  Tang,  X. Z.  Liang,  J. Y.  Li and  Q.  Zhou: Adv. Mater. Res., 746 (2013), 505. https://doi.org/10.4028/www.scientific.net/AMR.746.505
• 18)   H. N.  Zhang,  L.  Hui,  J. H.  Dong and  H. H.  Xiong: High Temp. Mater. Process., 37 (2018), 785. https://doi.org/10.1515/htmp-2017-0024
• 19)   R. K.  Dishwar,  S.  Agrawal,  A. K.  Mandal and  O. P.  Sinha: Trans. Indian Inst. Met., 73 (2020), 537. https://doi.org/10.1007/s12666-020-01861-8
• 20)   H.  Cabrera-Real,  A.  Romero-Serrano,  B.  Zeifert,  A.  Hernandez-Ramirez,  M.  Hallen-Lopez and  A.  Cruz-Ramirez: J. Mater. Cycles Waste Manag., 14 (2012), 317. https://doi.org/10.1007/s10163-012-0072-y
• 21)   G. J.  Albertsson,  L.  Teng and  B.  Björkman: Miner. Process. Extr. Metall., 123 (2014), 116. https://doi.org/10.1007/s11663-013-9939-0
• 22)   E.  Tang,  X. Z.  Liang,  J. Y.  Li and  Q.  Zhou: Adv. Mater. Res., 746 (2013), 505. https://doi.org/10.4028/www.scientific.net/AMR.746.505
• 23)   Y. Y.  Zhang,  X. F.  Shi,  Y. H.  Qi,  Z. S.  Zou and  Y. H.  Guo: Iron Steel, 50 (2015), 17. https://doi.org/10.13228/j.boyuan.issn0449-749x.20140073
• 24)   X. R.  Wu,  X. M.  Dong,  R. T.  Wang,  H. H.  Lü,  F.  Cao and  X. M.  Shen: J. Residuals Sci. Technol., 13 (2016), S57 (in Chinese). https://doi.org/10.12783/issn.1544-8053/13/2/S10
• 25)   G. U.  Kapure,  C. B.  Rao,  V. D.  Tathavadkar and  R.  Sen: Ironmaking Steelmaking, 38 (2011), 590. https://doi.org/10.1179/1743281211y.0000000028
• 26)   X. R.  Wu,  X. M.  Dong,  R. T.  Wang,  H. H.  Lü,  F.  Cao and  X. M.  Shen: J. Residuals Sci. Technol., 13 (2016), S57.
• 27)   P. J.  Liu,  Z. G.  Liu,  M. S.  Chu,  R. J.  Yan,  F.  Li and  J.  Tang: Environ. Sci. Pollut. Res., 29 (2022), 65500. https://doi.org/10.1007/s11356-022-20420-w
• 28)   Z.  Chen,  R. Q.  Zheng,  D. C.  Ju,  R.  Mao,  H.  Ma,  H. B.  Peng and  W. T.  Du: J. Sustain. Metall., 8 (2022), 1001. https://doi.org/10.1007/s40831-022-00549-1
• 29)   P. J.  Liu,  Z. G.  Liu,  M. S.  Chu,  R. J.  Yan,  F.  Li,  J.  Tang and  J. G.  Feng: Process Saf. Environ. Prot., 164 (2022), 678. https://doi.org/10.1016/j.psep.2022.06.034
• 30)   L. C.  Zheng,  A.  Malfliet,  H. Y.  Ren,  Z. H.  Jiang,  B.  Blanpain and  M. X.  Guo: J. Sustain. Metall., 8 (2022), 297. https://doi.org/10.1007/s40831-021-00490-9
• 31)   X.  Wang,  J.  Pan,  D. Q.  Zhu,  C. C.  Yang,  Z. Q.  Guo and  G. H.  Xia: J. Mater. Res. Technol., 19 (2022), 2516. https://doi.org/10.1016/j.jmrt.2022.06.027
• 32)   H.  Zhang,  J. H.  Dong,  H. H.  Xiong,  Z.  Wang and  Y. M.  Lu: J. Alloy. Compd., 699 (2017), 408. https://doi.org/10.1016/j.jallcom.2016.12.362
• 33)   Y.  Lin,  B. J.  Yan,  T.  Fabritius and  Q. F.  Shu: Metall. Mater. Trans. B, 51 (2020), 763. https://doi.org/10.1007/s11663-020-01777-0
• 34)   S. S.  Jung,  G. B.  Kim and  I.  Sohn: J. Am. Ceram. Soc., 100 (2017), 3771. https://doi.org/10.1111/jace.14891
• 35)   P.  Gao,  W. T.  Zhou,  Y. X.  Han,  Y. J.  Li and  C. W.  Zhang: Miner. Process. Extr. Metall. Rev., 43 (2022), 1. https://doi.org/10.1080/08827508.2020.1793144
• 36)   J. S.  Shiau: Metall. Res. Technol., 117 (2020), 306. https://doi.org/10.1051/metal/2020024
• 37)   M.  Liu,  X. W.  Lv,  E. G.  Guo,  P.  Chen and  Q. G.  Yuan: ISIJ Int., 54 (2014), 1749. https://doi.org/10.2355/isijinternational.54.1749
• 38)   M.  Lei,  B.  Ma,  Y.  Chen,  W.  Liu,  B.  Liu,  D.  Lv,  W. J.  Zhang and  C.  Wang: ACS Sustain. Chem. Eng., 8 (2020), 3959. https://doi.org/10.1021/acssuschemeng.0c00219
• 39)   S.  Ilyas,  R. R.  Srivastava,  H.  Kim,  N.  Ilyas and  R.  Sattar: Sep. Purif. Technol., 232 (2020), 115971. https://doi.org/10.1016/j.seppur.2019.115971
• 40)   P. J.  Liu,  Z. G.  Liu,  M. S.  Chu,  J.  Tang,  L. H.  Gao and  R. J.  Yan: J. Hazard. Mater., 413 (2021), 125403. https://doi.org/10.1016/j.jhazmat.2021.125403