Thermodynamic Modeling of the Solid State Carbothermic Reduction of Chromite Ore

Abstract

Chromium has a wide range of applications, including as an alloy addition in various steels and also as a corrosion resistance coating. Carbothermal reduction of chromite ore (FeCr₂O₄) in a submerged arc furnace is an important industrial process for extracting chromium, but the energy consumption is excessive. It is suggested that one area for future research is the low temperature carbothermic solid state reduction of chromite to produce an intermediate product, which can subsequently be upgraded to ferrochromium . In this regard, a thermodynamic model has been developed to investigate this process and the effects of temperature , carbon additions and ore composition on the recovery of chromium and the grade of the ferrochromium , have been studied. Further development of the model may allow it to be applied to the simulation of other processes for the recovery of chromium from chromite ores.

Introduction

During the last century, chromium has developed into a strategic metal and the global demand for chromium and its compounds has noticeably increased. Chromium is used in the manufacture of stainless steels, numerous alloys, catalysts and other various products. Generally, the source of the chromium units is ferrochromium or ferrochrome, which is primarily obtained from chromite ore . This ore represents a significant proportion of the known chromium reserves and also accounts for a large percentage of the current output [1, 2]. Metallurgical grade ore is utilized for ferrochromium production and in 2014 this accounted for 96% of the chromite ore consumption [3, 4]. The major mineral in the ore is iron chromite , which is a spinel and has the formula FeCr₂O₄ or FeO˖Cr₂O₃. However, various impurities can substitute for iron and chromium in the tetrahedral and octahedral sites and thus a more accurate formula is (Fe, Mg, Ca) [Cr, Al, Fe]₂O₄ [5, 6]. The amounts of MgO, Fe₂O₃, and/or Al₂O₃, can have a significant impact on the reduction process.

Traditionally, chromite ores have been processed using a number of various pyrometallurgical methods such as the submerged arc furnace (SAF) [7, 8]. Chromite ore is added along with a carbon-containing reducing agent and a flux , typically silica and then smelted in the SAF. This process is very energy intensive, requiring 4 MWh per tonne of ferrochrome alloy produced. There are some disadvantages of this process, such as the limited use of ore fines and friable ores as well as its dependence on both expensive metallurgical coke and the very high electrical energy consumption [9, 10]. Solid state carbothermal pre-reduction can decrease both the energy requirements and the cost [11–14]. The carbothermal reduction of chromite ore is complex and there are many kinetic and thermodynamic factors such as: ore composition, the Cr/Fe ratio, the amounts of ferrous and ferric iron , particle size, reaction temperature and time, and type and amount of reducing agent [6, 10]. Metallization of the chromite ore to ferrochromium in the solid state followed by concentration provides an attractive alternative to submerged arc furnace smelting .

In the current work, firstly previous thermodynamic studies of the solid state carbothermic reduction of chromite ore were reviewed. Then, an equilibrium model was developed to simulate the reduction process. The effects of temperature , carbon additions and ore composition on the chromium recovery and the grade of the ferrochromium were investigated. Subsequently, the simulated results were compared to those of previous thermodynamic and experimental studies.

Reduction Mechanisms

The reduction of chromite ore to ferrochromium can involve a large number of steps. Firstly, with carbon as the reducing agent, iron oxide in the iron chromite is directly reduced to metallic iron as follows [6, 15]:

${\text{FeCr}}_{ 2} {\text{O}}_{{ 4({\text{s}})}} + {\text{C}}_{{({\text{s}})}} \to {\text{Cr}}_{ 2} {\text{O}}_{{ 3 ( {\text{s)}}}} + {\text{Fe}}_{{({\text{s}})}} + {\text{CO}}_{{({\text{g}})}}$

(1)

Subsequently, the chromium oxide will be reduced by carbon to chromium carbide as follows [16]:

$7 {\text{Cr}}_{ 2} {\text{O}}_{{ 3 ( {\text{s)}}}} + 2 7 {\text{C}}_{{ ( {\text{s)}}}} \to 2 {\text{Cr}}_{ 7} {\text{C}}_{{ 3 ( {\text{s)}}}} + 2 1 {\text{CO}}_{{ ( {\text{g)}}}}$

(2)

Additionally, at higher temperatures, the chromium oxide can be reduced by chromium carbide as follows:

${\text{Cr}}_{ 7} {\text{C}}_{{ 3 ( {\text{s)}}}} + {\text{Cr}}_{ 2} {\text{O}}_{ 3} \left( {\text{s}} \right) \to 9 {\text{Cr}}\left( {\text{s}} \right) + 3 {\text{CO}}_{{ ( {\text{g)}}}}$

(3)

It should also be noted that some of the carbon dioxide can react with solid carbon to form carbon monoxide by the Boudouard reaction:

${\text{C}}_{{ ( {\text{s)}}}} {\text{ + CO}}_{{ 2 ( {\text{g)}}}} \to 2 {\text{CO}}_{{ ( {\text{g)}}}}$

(4)

Wang et al. performed a kinetic analysis of the reduction of a synthetic chromite . Additional reduction reactions between carbon and oxide species for the formation of carbide and chromium metal were provided as follows [17]:

$7 {\text{Cr}}_{ 2} {\text{O}}_{{ 3 ( {\text{s)}}}} + 2 7 {\text{Fe}}_{ 3} {\text{C}}_{{ ( {\text{s)}}}} \to 2 {\text{Cr}}_{ 7} {\text{C}}_{{ 3 ( {\text{s)}}}} + 8 1 {\text{Fe}}\left( {\text{s}} \right) + 2 1 {\text{CO}}_{{ ( {\text{g)}}}}$

(5)

In carbothermal reduction , metal carbide formation is unavoidable. Metallic chromium is produced by the reaction of chromium carbide with chromium oxide [18]. Thermodynamically, the formation of mixed carbides of iron and chromium favors the reduction of chromium. Both the liquid alloy and the carbides can act as reducing agents and carriers of carbon, thereby enhancing the rate of reduction . In chromite reduction , the metal carbides form first, and then a ferrochrome solution is produced, at a sufficiently high reduction degree, by the reaction between metal carbides and metal oxides.

Katayama and Tokuda studied the reduction behaviour of a synthetic chromite containing both magnesium and iron spinels. They found that the pure iron chromite (FeCr₂O₄) decomposed into FeO and Cr₂O₃ at temperatures around 1150 °C. Subsequently, the FeO would quickly be reduced to metallic iron by carbon. Then, the iron would be converted into an iron carbide with a structure similar to that of the chromium carbides. Complex chromites, containing both the spinels of iron and magnesium, were found to have two distinct reduction stages, the first being the decomposition of iron chromite at 1150 °C, and the second being the decomposition of magnesium chromite at 1250 °C. For complex spinels containing some alumina (Al₂O₃), the decomposition temperature was increased to 1330 °C. Finally, the two researchers determined that (Fe, Cr)₇C₃ was the final product of the reduction of the complex spinels. However, for the chromites without iron , such as the magnesium and magnesium aluminum chromites, then the final product was Cr₃C₂ [15].

Thermodynamics

With regards to the reduction of these ores, there have been a number of studies, which have examined the effects of reducing conditions, reaction temperatures, reductant type and phase transformations on the composition and the metallization degree of the chromium-iron alloy. Reduction of chromite by carbon has been studied by several researchers [19–24]. It appears that the composition of the carbide phase is a function of the availability of carbon during reduction . For carbon-rich conditions, Cr₇C₃ is formed, while for carbon-poor conditions, Cr₂₃C₆ will be formed. Several studies also reported Cr₄C, although it is unclear which conditions favor its formation over Cr₇C₃. This confusion is likely due to the overlap between the x-ray diffraction (XRD) patterns of the carbides, complicating their positive identification. It is likely that in many cases a mixed carbide is formed.

Zhang et al. developed a thermodynamic model of the carbothermic reduction of FeCr₂O₄ using FactSage [25]. The amount of carbon was varied in the temperature range of 900 °C to 1600 °C in an argon atmosphere. It was reported that metal carbides were more stable at lower temperatures and higher C/O ratios, and the Fe-Cr-C liquid phase formed at higher temperatures and lower C/O ratios. The model predicted that at C/O = 1.5, the oxygen could be completely removed from the Cr₂O₃ at 1200 °C. They noted that the amount of products was temperature dependent. At C/O = 1, Cr₄C is the most stable carbide at 1300 °C and above. Fe₃C₇ was produced at about 1050 °C and decomposed completely at 1400 °C. The FeCr₂O₄ decomposition started at about 1030 °C and it disappeared quickly and produced Cr₂O₃ and iron carbides. They noted that the reaction products were: Cr₇C₃, Fe₃C₇, Cr₄C and Fe-Cr-C solution [25].

Hino et al. (1998) considered the chromite ore to be a pseudo ternary spinel structure which consisted of FeCr₂O₄, MgCr₂O₄ and MgAl₂O₄ [26]. By calculating the free energy of formation of FeCr₂O₄, they determined the activity coefficients of FeCr₂O₄ in FeCr₂O₄-MgCr₂O₄ and FeCr₂O₄-MgCr₂O₄-Al₂O₃ at 1300 °C [27, 28]. Their results showed that the solid solution could be considered as a regular solid solution and the activities of each component could be expressed in terms of temperature and composition. They used these activity coefficients to study the effects of both magnesium oxide and aluminum oxide on the activity of chromium oxide. They found that the chromite spinel system had negative deviation from ideality, which became more pronounced as MgCr₂O₄ was replaced by MgAl₂O₄ [26].

Model Development

In the present work, a thermodynamic model was developed using the Equilibrium module of HSC Chemistry 7.1 and solved using the Gibbs Solver [29]. The Gibbs Solver uses the Gibbs free energy minimization technique to determine the equilibrium concentrations of the stable species at various temperatures and pressures. The elements considered were as follows: Cr, Fe, Si, Mg, Al, O and C. Inputting these elements generated a list of 190 possible species, from which 25 were selected based on their stability over the selected temperature range and pressure. Table 1 shows the species grouped into four phases: gases, oxides, alloy and carbon. The various spinels were included with the oxides and the alloy consisted of both iron and chromium carbides. It is understood that the model has a number of limitations. Firstly, being a thermodynamic model , it does not take into account kinetic considerations. Secondly, it is based on the thermodynamic data that is currently available. Thirdly, it assumes a closed system, while an actual system would likely be open. Fourthly, it assumes ideality unless activity coefficient information has been inputted.

Table 1

Phases (in bold) and species considered in the thermodynamic model

Gases	Oxides	Oxides (Spinels)	Alloy
O₂	Al₂O₃	FeCr₂O₄	Cr₄C
CO	Cr₂O₃	MgCr₂O₄	Cr₂₃C₆
CO₂	Fe₂O₃	Fe₃O₄	Cr₃C₂
Carbon	FeO	Mg₂SiO₄	Cr₇C₃
C	CrO	MgFe₂O₄	Fe₃C
	MgO	MgAl₂O₄
	MgSiO₃	FeAl₂O₄
	SiO₂	Cr₃O₄

Activity coefficient data was obtained and inputted based on the information available in the literature. For those cases where the activity coefficients were reported at only one temperature , the regular solution model was used to obtain values for the activity coefficients at other temperatures as follows:

$ln\gamma_{1} = \left( {\frac{{T_{2} }}{{T_{1} }}} \right)ln\gamma_{2}$

(6)

The activity coefficient of Fe₃O₄ was adapted from the results of Petric and Jacob (1982) at 1400 °C as follows [30]:

$ln\gamma_{{Fe_{3} O_{4} }} = \left( {\frac{0.8}{{0.075 - X_{{Fe_{3} O_{4} }} }} + \frac{3.2}{{4.7 - X_{{Fe_{3} O_{4} }} }}} \right)X_{{Fe_{3} O_{4} }} \left( {\frac{1673}{T}} \right)$

(7)

The activity coefficient of MgFe₂O₄ was adapted from the work of Katayama and Iseda (2002) at 1000 °C. The activity coefficient equation is as follows [31]:

$ln\gamma_{{MgFe_{2} O_{4} }} = \frac{{1.0 + \left( { - 0.92 \cdot X_{{MgFe_{2} O_{4} }} } \right)}}{{1 + \left( {4.6 \cdot X_{{MgFe_{2} O_{4} }} } \right)}} \cdot \frac{1273}{T}$

(8)

The activity coefficients for FeCr₂O₄, MgCr₂O₄, and MgAl₂O₄ were obtained from the equations developed by Hino et al. (1998) as follows [26]:

$ln\gamma_{{FeCr_{2} O_{4} }} = \frac{1}{RT}\left( { - 12800X_{{MgCr_{2} O_{4} }}^{2} - 92000X_{{MgAl_{2} O_{4} }}^{2} - 74800X_{{MgCr_{2} O_{4} }} X_{{MgAl_{2} O_{4} }} } \right)$

(9)

$ln\gamma_{{MgCr_{2} O_{4} }} = \frac{1}{RT}\left( { - 12800X_{{FeCr_{2} O_{4} }}^{2} - 30000X_{{MgAl_{2} O_{4} }}^{2} + 49200X_{{FeCr_{2} O_{4} }} X_{{MgAl_{2} O_{4} }} } \right)$

(10)

$ln\gamma_{{MgAl_{2} O_{4} }} = \frac{1}{RT}\left( { - 92000X_{{FeCr_{2} O_{4} }}^{2} - 30000X_{{MgCr_{2} O_{4} }}^{2} - 109200X_{{FeCr_{2} O_{4} }} X_{{MgCr_{2} O_{4} }} } \right)$

(11)

The input to the system was 100 kg of ore with the following composition: 43.7% Cr₂O₃, 20.1% Fe₂O₃, 15.6% MgO, 13.5% Al₂O₃ and 7.2% SiO₂ . This ore composition was chosen so as to represent Black Thor chromite ore . The X-ray diffraction analysis of the ore showed that the major minerals were iron chromite (FeCr₂O₄) and clinichlore ((Mg, Fe)₅Al(Si₃Al)O₁₀(OH)₈). From the ore composition, the moles of chromium and iron were 0.575 and 0.252, respectively, and thus the ratio of the chromium to iron of 2.28 was greater than the stoichiometric value of 2. This could be due to the reported non-stoichiometry of these types of ore [32]. From the model , the recoveries and grades of both chromium and iron in the ferrochromium product, as well as the carbon content of the ferroalloy were determined.

Modeling of Carbothermic Reduction of Chromite Ore

Effect of Temperature

As discussed previously, for the Black Thor chromite ore under consideration, the iron chromite (FeCr₂O₄) was not stoichiometric as the chromium to iron ratio was 2.28. Therefore, the first series of modelling calculations were performed at a carbon addition of 1.145 kmol/100 kg of ore, which is the stoichiometric carbon requirement for the reduction of the iron and chromium oxides according to the following reactions:

${\text{FeO}} + {\text{C}} = {\text{Fe}} + {\text{CO}}_{{ ( {\text{g)}}}}$

(12)

${\text{Cr}}_{ 2} {\text{O}}_{ 3} + 3 {\text{C}} = 2 {\text{Cr}} + 3 {\text{CO}}_{{ ( {\text{g)}}}}$

(13)

Figure 1 shows the behaviour of the major oxide species, particularly those containing magnesium, as a function of temperature . Of particular interest are the behaviours of the iron chromite (FeCr₂O₄) and the magnesium chromite (MgCr₂O₄).

../images/468727_1_En_74_Chapter/468727_1_En_74_Fig1_HTML.gif — Fig. 1
Effect of temperature on the equilibrium amounts of the major oxide species for the stoichiometric carbon addition

From 200 °C, the amounts of these species decrease slowly with increasing temperature . At about 900 °C, the amount of the iron chromite decreases rapidly according to the following reaction:

${\text{FeCr}}_{ 2} {\text{O}}_{ 4} = {\text{FeO}} + {\text{Cr}}_{ 2} {\text{O}}_{ 3}$

(14)

This liberates considerable chromium sesquioxide and the wustite can be reduced as follows:

$3 {\text{FeO}} + 4 {\text{C}} = {\text{Fe}}_{ 3} {\text{C}} + 3 {\text{CO}}_{{ ( {\text{g)}}}}$

(15)

Some of the chromium oxide can react with the free magnesium oxide according to the following reaction:

${\text{MgO}} + {\text{Cr}}_{ 2} {\text{O}}_{ 3} = {\text{MgCr}}_{ 2} {\text{O}}_{ 4}$

(16)

Although this solid-solid reaction is thermodynamically favourable, it would not be expected to occur from a kinetic perspective as the temperature is relatively low and the reacting solids involved have very high melting points. Most of the FeCr₂O₄ has disappeared by about 1150 °C. MgCr₂O₄ is significantly more stable than FeCr₂O₄ and only begins to decrease rapidly at about 1100 °C, but continues to exist even up to about 1350 °C.

The behaviours of the chromium-containing species are shown in Fig. 2. Due to the significant decomposition of iron chromite at about 900 °C, Cr₂O₃ is liberated in substantial amounts and reaches a maximum at about 1120 °C, before starting to decrease. Over this temperature range some Cr₂O₃ reacts with MgO to form MgCr₂O₄ as given by reaction (16). However, the rapid decrease in the amount of Cr₂O₃ is primarily associated with the formation of chromium carbides, mainly Cr₃C₂ with some Cr₄C as follows:

../images/468727_1_En_74_Chapter/468727_1_En_74_Fig2_HTML.gif — Fig. 2
The equilibrium amounts of the chromium-containing species as a function of temperature for the stoichiometric carbon addition

${\text{Cr}}_{ 2} {\text{O}}_{ 3} + 6 {\text{C}} = 3 {\text{CO}}_{{ ( {\text{g)}}}} + {\text{Cr}}_{2} {\text{C}}_{3}$

(17)

CrO begins to form at about 1000 °C followed by Cr₃O₄ at about 1100 °C. Above about 1200 °C, the amount of Cr₂O₃ decreases more slowly and the amounts of CrO and Cr₃O₄ continue to increase. The carbon in Cr₃C₂ is consumed as the temperature increases as a result of the following reaction:

$2. 8 {\text{Cr}}_{3}{\text{C}}_{ 2} + {\text{Cr}}_{ 2} {\text{O}}_{ 3} = 2. 6 {\text{Cr}}_{ 4} {\text{C}} + 3 {\text{CO}}_{{ ( {\text{g)}}}}$

(18)

Therefore, Cr₃C₂ begins to decrease and is replaced by an increasing amount of Cr₄C. Consequently, at 1600 °C the major species present are CrO, Cr₄C and Cr₂O₃. There is also some Cr₃C₂ and some Cr₃O₄. Thus, for a carbon addition of 1.45 kmol/100 kg of ore, conditions are not sufficiently reducing to permit the recovery of all the chromium from the chromium oxides.

The behaviours of wustite, iron carbide, carbon and carbon monoxide are shown in Fig. 3. As discussed previously, iron chromite begins to decompose at low temperatures as given by reaction (10) and this releases wustite. From about 925 °C to 1180 °C, the wustite is converted into iron carbide according to the following reaction:

../images/468727_1_En_74_Chapter/468727_1_En_74_Fig3_HTML.gif — Fig. 3
Equilibrium amounts of FeO, Fe₃C, C and CO_(g) as a function of temperature for the stoichiometric carbon addition

$3 {\text{FeO}} + 4 {\text{C}} = {\text{Fe}}_{ 3} {\text{C}} + 3 {\text{CO}}_{{ ( {\text{g)}}}}$

(19)

As already stated, chromium carbides, in particular Cr₄C begin to form at 1100 °C and therefore by 1180 °C all the carbon has been consumed and converted to Fe₃C and Cr₄C, plus some Cr₃C₂. These reactions result in the generation of a large amount of carbon monoxide over this temperature range. Above 1180 °C, further generation of carbon monoxide is due to the reaction of chromium carbides with chromium oxides as described by reaction (18). Additionally, above this temperature , since the carbon has been consumed and the conditions have become somewhat more oxidizing, then some of the Fe₃C is converted into FeO according to the following reaction:

$2. 6 7 {\text{Cr}}_{ 2} {\text{O}}_{ 3} + 2. 3 3 {\text{Fe}}_{ 3} {\text{C}} = 7 {\text{FeO}} + 1. 3 3 {\text{Cr}}_{ 4} {\text{C}} + {\text{CO}}_{{ ( {\text{g)}}}}$

(20)

Figure 4 shows the iron and chromium recoveries as a function of temperature . Reduction of iron oxide to iron occurs at 925 °C and the recovery increases rapidly to 96% at 1150 °C. At higher temperatures, the recovery decreases due to oxidation processes. Chromium oxide reduction begins at 1080 °C and the recovery quickly reaches a maximum of 56% at 1180 °C. Subsequently, the recovery decreases once again due to oxidation processes.

../images/468727_1_En_74_Chapter/468727_1_En_74_Fig4_HTML.gif — Fig. 4
Iron and chromium recoveries as a function of temperature for the stoichiometric carbon addition

Figure 5 shows the chromium, iron and carbon contents of the metal. At 925 °C only iron is present as Fe₃C and thus the iron content is about 93.3%. The carbon content is about 6.7%, until chromium begins to be recovered at about 1100 °C. Subsequently, chromium carbides began to form and consequently the chromium and carbon contents increased and the iron grade decreased. The carbon content reached a maximum of about 9.5% at 1180 °C due to chromium carbide formation and here the chromium grade was about 51%. Above this temperature , the carbon content decreased due to oxidation processes, the chromium content increased slightly, while the iron content remained relatively constant. At 1600 °C the carbon content was about 6%, the chromium content was about 55% and the iron content was about 39%.

../images/468727_1_En_74_Chapter/468727_1_En_74_Fig5_HTML.gif — Fig. 5
Iron , chromium and carbon contents of the ferrochromium as a function of temperature for the stoichiometric carbon addition

Since the stoichiometric carbon addition of 1.115 kmole/100 kg of ore gave chromium recoveries of only 56%, further calculations were performed at a much higher carbon addition of 1.450 kmol of C/100 kg of ore. Figure 6 shows the effect of temperature on the behaviour of the chromium oxides and carbides and can be compared with Fig. 2. The temperatures at which various species formed or where there were significant changes in behaviours, were not affected by the amount of carbon added. However, the amounts of the species were significantly changed. Starting from about 1200 °C, the amount of Cr₃C₂ for the higher carbon content was almost double that of the lower carbon content and these higher levels were maintained even up to 1600 °C. Also, the amount of Cr₄C was higher over the whole temperature range. At the higher carbon content, there was virtually no Cr₂O₃ at 1600 °C but there was some CrO. At the lower carbon contents, the amounts of both CrO and Cr₂O₃ were significantly higher. The behaviours of FeO, Fe₃C, C and CO are shown in Fig. 3 and can be compared with Fig. 2. Again the temperatures of the various reactions did not change significantly. However, for the higher carbon addition, the amounts of Fe₃C were higher and the corresponding amounts of FeO were lower. Thus, it can be seen that the increased carbon addition resulted in lower amounts of metal oxides, increased amounts of metal carbides and higher carbon contents of the ferrochromium .

../images/468727_1_En_74_Chapter/468727_1_En_74_Fig6_HTML.gif — Fig. 6
Equilibrium amounts of the chromium-containing species as a function of temperature for the non-stoichiometric carbon addition

Figure 7 shows the effect of temperature on the iron and chromium recoveries for the carbon addition of 1.45 kmole/100 kg of ore. Again the temperatures of the initiation of the reduction of iron oxide to iron and chromium oxide to chromium were not affected by the amount of carbon. At 1180 °C the iron recovery was 98% and this was slightly higher than the value of 96% at 1150 °C for the stoichiometric carbon addition. Subsequently, the chromium recovery remained relatively constant. On the other hand, the chromium recovery was 80% at 1180 °C and increased to 92% at 1600 °C. The incomplete recovery of chromium can be attributed to the presence of some chromium oxide (CrO) in the oxide phase. These values are significantly higher than the maximum recovery of only about 56% for the stoichiometric carbon addition.

../images/468727_1_En_74_Chapter/468727_1_En_74_Fig7_HTML.gif — Fig. 7
Iron and chromium recoveries as a function of temperature for the non-stoichiometric carbon addition

As shown in Fig. 8, at 1180 °C the chromium, iron and carbon contents of the metal for the carbon addition of 1.45 kmole/100 kg were 57, 33 and 10%, respectively. At 1600 °C the corresponding values were 63, 31 and 6%. From 1180 °C to 1600 °C the chromium grade increased slightly to 63% due to the increasing recovery of chromium. The higher carbon addition resulted in an increased recovery of chromium and thus a higher chromium grade. The behaviour of carbon was similar to that observed for the stoichiometric carbon addition.

../images/468727_1_En_74_Chapter/468727_1_En_74_Fig8_HTML.gif — Fig. 8
Iron , chromium and carbon contents of the ferrochromium as a function of temperature for the non-stoichiometric carbon addition

Effect of Temperature and Carbon

In order to more fully understand the effects of carbon additions and temperature on the chromium recovery and the grade of the ferrochromium , calculations were performed over a wide range of carbon additions and temperatures. Figures 9 shows the effects of carbon additions and temperature on the iron and chromium recoveries. It can be seen that both of these recoveries increased with temperature and carbon additions. For chromium, the recovery was more strongly dependent on the carbon addition than the temperature . For a given carbon addition the majority of the reduction occurred over the relatively narrow temperature range from about 1080 °C to 1180 °C. These chromium recoveries are in general agreement with the results reported in the literature, such as those of Chakraborty et al. (2005) [25]. On the other hand, the recovery increased almost linearly with carbon addition from about 1% at 0.5% carbon to about 90% at 1.5% carbon. At higher carbon additions, chromium recoveries approaching 100% could be achieved. Iron was recovered at lower carbon additions and temperatures than chromium. Again for a given carbon addition, the majority of reduction occurred over a relatively narrow temperature range from 900 °C to 1050 °C. Again, these iron recoveries correspond to those found in the literature [25]. For iron , over this temperature range, the recovery increased rapidly with carbon to about 80% for a carbon addition of 0.5%. Subsequently, the recovery increased slowly to about 95% for a carbon addition of 1.5%. Again as for chromium, the recoveries decreased due to reoxidation at low carbon additions.

../images/468727_1_En_74_Chapter/468727_1_En_74_Fig9_HTML.gif — Fig. 9
Effects of carbon additions and temperature on a the iron recovery and b the chromium recovery

Figure 10 shows the effects of temperature and carbon additions on the iron , chromium and carbon contents of the ferrochromium . At both low carbon additions and low temperatures, the iron content was high, as very little chromium oxide was reduced. However, with both increasing temperature and carbon additions an increasing amount of chromium oxide was reduced and thus the iron content decreased. Due to the relative high stability of chromium oxide as compared to iron oxide , the chromium grade of the metal exhibited a behaviour which was the inverse of that for iron . Thus the chromium grade was high at high carbon contents and high temperatures. Here the typical chromium content was in the range of 60–62%. The highest chromium content was 63%. At low temperatures and low carbon additions, only iron was reduced and Fe₃C formed and thus the carbon content was about 6.7%. At intermediate carbon additions and temperatures, chromium carbides formed and thus the carbon content was high at about 10% and decreased with increasing temperature . As a result of this decreasing carbon content, the highest chromium grade of 63% was achieved. At both high temperatures and carbon additions, the carbon content remained high at about 10%.

../images/468727_1_En_74_Chapter/468727_1_En_74_Fig10_HTML.gif — Fig. 10
Effects of carbon additions and temperature on a the iron grade of the ferrochromium and b the chromium grade of the ferrochromium and c the carbon content of the ferrochromium

Effect of Ore Composition

In addition to temperature and carbon addition, ore composition can also have a significant effect on the recovery of chromium and the chromium grade of the ferrochromium . Here a number of important compositional factors can be investigated such as the iron content of the ore and the magnesia content of the ore as shown in Fig. 11. The addition of magnesium oxide promoted the formation of magnesium chromite and hence the reduction of the chromium oxide became more difficult. The increasing iron content of the ore resultsed in increased reduction of iron oxide and hence an increased amount of Fe₃C. Hence, the activity of chromium was reduced in the Fe-Cr-C solution, chromium reduction was facilitated and therefore the chromium recovery increased.

../images/468727_1_En_74_Chapter/468727_1_En_74_Fig11_HTML.gif — Fig. 11
Effect of iron and magnesia contents of the ore on the chromium recovery at 1150 °C for the non-stoichiometric carbon addition

Conclusions

(1)
The literature pertaining to the thermodynamics of the carbothermic solid-state reduction of chromite ore has been reviewed. Although the chemistry can be considered complex, the reduction sequence can be described as follows. Firstly, iron chromite (FeCr₂O₄) decomposes to produce chromic oxide and wustite (FeO), which is reduced to iron and iron carbide (Fe₃C). Secondly, the chromic oxide is reduced and a Fe-Cr-C solution is formed. Other chromites, such as magnesium chromite (MgCr₂O₄) decompose at higher temperatures and chromium oxide reacts with the Fe-Cr-C solution and consequently the carbon content of the Fe-Cr-C solution decreases.
(2)
A thermodynamic model of the carbothermic solid-state reduction of Black Thor chromite ore has been developed using the Equilibrium Composition module of HSC Chemistry 7.1. The input data was the typical composition of the ore as determined by chemical analysis. In general, the model calculations were in agreement with the thermodynamic reaction sequencing and temperatures available in the literature. Iron chromite decomposed above 900 °C and iron carbide began to form at about 925 °C. Subsequently, chromium carbide began to form at the much higher temperature of 1100 °C and thus a Fe-Cr-C solid solution was produced. Magnesium chromite , being more stable than iron chromite decomposed above 1100 °C and a portion remained stable even up to 1350 °C. As the temperature increased, Cr₄C was produced in increasing amounts at the expense of Cr₃C₂ and thus the carbon content of the Fe-Cr-C solution decreased.
(3)
Chromium recoveries increased with increasing temperature and carbon additions. For chromium, the majority of reduction occurred from about 1080 °C to 1180 °C. Recoveries approaching 100% could be achieved at carbon additions of 1.5 kmole/100 kg of ore and temperatures over 1200 °C. Under these conditions, the chromium grade was 60 to 62% and the carbon content was about 10%. The model was utilized to predict the effects of both the iron and magnesium oxide contents of the ore on the chromium recovery . It is intended to apply the model to other processes for treating chromites, such as reduction by methane and the segregation process using calcium chloride .

Acknowledgements

The authors thank the Natural Resources and Engineering Research Council of Canada (NSERC) for their support of this research.

References

1.
Independent Environmental Technical Evaluation Group (IETEG), Chromium (VI) Handbook (2005) CRC Press, Boca Raton, Florida
2.
Beukes JP, Du Preez SP, Van Zyl PG, Paktunc D, Fabritius T, Päätalo M, Cramer M (2017) Review of Cr (VI) environmental practices in the chromite mining and smelting industry–relevance to development of the Ring of Fire, Canada. J Clean Product 165:874–889Crossref
3.
U.S. Geological Survey. Mineral Commodity Summaries (2017) U.S. Geological Survey, United States of America
4.
International Stainless Steel Forum, Stainless Steel in Figures (2015) In: International stainless steel forum
5.
Tripathy SK, Banerjee PK, Suresh N (2015) Magnetic separation studies on ferruginous chromite fine to enhance Cr: Fe ratio. Int J Miner Metall Mater 22(3):217–224Crossref
6.
Chakraborty D, Ranganathan S, Sinha SN (2010) Carbothermic reduction of chromite ore under different flow rates of inert gas. Metall Mater Trans B 41B:437–444
7.
Vignes A (2013) Extractive metallurgy 3: processing operations and routes. ISTE Ltd and John Wiley & Sons Inc, New Jersey
8.
Daavittila JBJ (2013) “High carbon ferrochrome technology”, in handbook of ferroalloys—theory and technology. Butterworth-Heinemann, Kidlington, Oxford, pp 317–363
9.
Schulte RF, Taylor RD, Piatak NM, Seal II RR (2010) Stratiform chromite deposit model. U.S. Geological Survey, Reston, Virginia, Open-File Report 2010 1232
10.
Gujar AR, Ambre NV, Sridhar DI, Mislankar PG, Loveson VJ (2010) Placer chromite along South Maharastra, Central West Coast of India. Curr Sci 99(4):491–499
11.
Takano C, Zambrano AP, Nogueira AEA, Mourao MB, Iguchi Y (2007) Chromites reduction reaction mechanisms in carbon-chromites composite agglomerates at 1773 K. ISIJ Int 47(11):1585–1589Crossref
12.
Apaydin F, Atasoy A, Yildiz K (2011) Effect of mechanical activation on carbothermal reduction of chromite with graphite. Can Metall Q 50(2):113–118Crossref
13.
Lyakishev NP, Gasik MI (1998) Metallurgy of Chromium. Allerton Press, New York
14.
Weber P, Eric PH (2006) The reduction of chromite in the presence of silica flux. Miner Eng 19(3):318–324Crossref
15.
Katayama H, Tokuda M (1979) The reduction behavior of synthetic chromites by carbon. Tetsu- Hagane 65(3):331–340Crossref
16.
Hu X, Teng L, Wang H, Ökvist LS, Yang Q, Björkman B, Seetharaman S (2016) Carbothermic reduction of synthetic chromite with/without the addition of iron powder. ISIJ Int 56(12):2147–2155Crossref
17.
Wang Y, Wang L, Yu J, Chou KC (2014) Kinetics of carbothermic reduction of synthetic chromite. J Min Metall Sect B Metall 50(1B):15–21Crossref
18.
Vignes A (2012) Extractive metallurgy 2: metallurgical reaction processes. ISTE Ltd and John Wiley & Sons Inc, New Jersey
19.
Khedr MH (2000) Isothermal reduction kinetics of Fe₂O₃ mixed with 1-10% Cr₂O₃ at 1173-1473 K. ISIJ Int 40(4):309–314Crossref
20.
Qayyum MA, Reeve DA (1976) Reduction of chromites to sponge ferrochromium in methane-hydrogen mixtures. Can Metall Q 15(3):193–200Crossref
21.
Anacleto NM, Ostrovski O (2004) Solid-state reduction of chromium oxide by methane-containing gas. Metall Mater Trans B 35B:609–615Crossref
22.
Ebrahimi-Kahrizsangi R, Zadeh HM, Nemati V (2010) Synthesis of chromium carbide by reduction of chromium oxide with methane. Int J Refract Met Hard Mater 28(3):412–415Crossref
23.
Arvanitidis I, Artin C, Cleason P, Swartling D (1996) Study of the kinetics of reduction of iron chromate by hydrogen. Scand J Metall,. 25(4): 141–147
24.
Chakraborty D, Ranganathan S, Sinha SN (2005) Investigations on the carbothermic reduction of chromite ores. Metall Mater Trans B 36(4):437–444Crossref
25.
Zhang Y, Liu Y, Wei W (2013) Carbothermal reduction process of the Fe-Cr-O system. Int J Miner Metall Mater 20(10):931–940Crossref
26.
Hino M, Nagasaka T, Higuchi K, Ban-Ya S (1998) Thermodynamic estimation on the reduction behavior of iron-chromium ore with carbon. Metall Mater Trans B 29(2):351–360Crossref
27.
Hino M, Higuchi KI, Nagasaka T, Ban-Ya S (1994) Phase equilibria and activities of the constituents in FeO· Cr₂O₃–MgO· Cr₂O₃ spinel solid solution saturated with Cr₂O₃. ISIJ Int 34(9):739–745Crossref
28.
Hino M, Higuchi K, Nagasaka T, Ban-Ya S (1995) Phase equilibria and thermodynamics of FeO.Cr2O3-MgO.Cr2O3-MgO.Al2O3 spinel structure solid solution saturated with (Cr, Al)2O3. ISIJ Int 35(7):851–858Crossref
29.
Roine A (2006) HSC chemistry 6.0 - User’s guide. Outokumpu Research Oy, Poir, Finland
30.
Petric A, Jacob KT (1982) Thermodynamic properties of Fe₃O₄-FeV₂O₄ and Fe₃O₄-FeCr2O4 spinel solid solutions. J Am Ceram Soc 65(2):117–123Crossref
31.
Katayama I, Iseda A (2002) Thermodynamic study of spinel-type solid solutions of the Fe₃O₄-MgFe₂O₄ coexisting with Fe₂O₃ by emf method. Scand J Metall 31:374–378Crossref
32.
Chirasha J (2011) Ferrochrome smelting and smelting capacity investment in Zimbabwe. Southern African Pyrometallurgy, 77–81, Johannesburg, 6–9 March 2011