Development of Chromite Sinter from Ultra-Fine Chromite Ore by Direct Sintering

Abstract

A significant quantity of chromite ore is available in form of fines and is friable in nature. Agglomeration is necessary for utilizing these fines. Briquetting gives green agglomerates with inferior high temperature properties. Pelletization requires further grinding of the naturally available ore fines and the subsequent firing of the green pellets for strength development which make it energy-intensive and complex process. In contrast, sintering can be done directly on the as-received friable chromite ore in the presence of coke breeze (as in case of iron ore sintering), which is likely to be free from the above limitations. In the current work, an optimum combination of temperature, flux (added to increase the relative quantity of molten phase required for sintering) and coke (added as fuel to attain the sintering temperature in a sinter pot) was computed using the thermochemical software, FactSage 6.1 and enthalpy balance calculation. Sintering of a mixture of chromite ore fines and flux with the optimum composition at 1600°C was carried out in; (i) resistance furnace (100-g scale), without using carbon and (ii) a pot sintering set-up (10-kg scale), using the computed quantity of carbon. A good correlation between experimental result and predicted equilibrium phases has been observed. The characterization of the sinters prepared in pot sintering set-up was done by conducting shatter, tumbler, and abrasion tests, and their phase identifications by XRD and EPMA/EDS. The developed sinter was found to possess adequate handling strength that would be well acceptable to produce ferrochrome in the SAF.

1. Introduction

World’s total generation of chromite ore is more than 30 million tons per annum¹⁾ at present, out of which more than 80% is available in form of friable fines. The fines can not be directly used for smelting in electric arc furnace (EAF) or submerged arc furnace (SAF) due to difficulty in charging and poor permeability of the bed during reduction. A significant portion of these fines flue off from the furnace with gas also. To alleviate the above problem, a suitable agglomeration techniques such as briquetting, palletizing or sintering is necessary. The present practice of ferrochrome production²⁾ is to grind the chromite ore fines to finer particles and pelletize them. This is followed by the firing of the pellets and their smelting in an electric arc furnace. Pelletizing of chromite ore fines followed by heat hardening can produce good quality agglomerate, but a considerable size reduction by grinding is necessary prior to the pelletization, which consumes much energy. Chatterjee³⁾ reported on the difficulty of making strong pellets from the Indian high-grade chromite ores. The difficulty could be overcome, according to him, by using a considerable quantity of bentonite as binder. However, this would unduly increase the slag volume and result in greater specific power consumption in the submerged arc furnace.

Indian Charge Chrome Limited⁴⁾ made cold-bonded briquettes from chromite ore fines of varying fineness using binders. They utilized these briquettes in the submerged electric arc furnace to produce charge chrome. However, high curing time and use of binder are important concern from process point of view. Nandy et al.⁵⁾ made cold-bonded chromite briquettes and compared their performance in the EAF with the existing practices of charging lump ore alone or combined charging of fired pellet and lump ore. They found poor performance of these briquettes than the fired pellets. Though, briquetting is a simple cold-bonded process with the advantages of having low operation cost and being environment-friendly, it has the limitation for its poor high temperature performance and is inferior to the fired pellets.

Several investigators have palletized and pre-reduced the pellets prior to charging in SAF. The friable chromite ore fines were first ground to suitable size fractions to make pellets. The pellets were pre-reduced at 1300–1400°C in a rotary kiln, followed by their smelting in the EAF to make the ferrochrome. The effects of composition, grain size of chromite fines and recycled fines on the compressive strength of the fired and pre-reduced pellets in varying diameters were studied.^6,7) The possibility of pre-reduction of a variety of chromite ores with a low-grade carbonaceous reductant was also examined.⁸⁾ These studies have shown that pre-reduction of chromite ore increases the productivity and decreases the electrical energy consumption, compared to the use of the unreduced material in the SAF. However, the heating and prereduction of chromite ore fines, as found by the preceding investigators, is energy intensive, and needs some added operational measures.

Nandy et al.⁵⁾ also observed that it was possible to agglomerate chromite ore fines through sintering. Performance study of this sinter in a laboratory-scale EAF showed that the specific power consumption in the melting was comparable with that of the heat-hardened chrome ore pellets and lower than the briquettes and lump ores. They also reported that chromite sinter can reduce the unit electricity consumption by 200–300 kWh/t for smelting due to its more open structure.⁵⁾ Zhu Deqing et al.⁹⁾ studied the sintering of chromite ore fines and the effect of various sintering parameters, and found that the normal sintering process is suitable for the agglomeration of chromium ore fines. They reported a sinter yield of up to 70.6%, a tumbler index of 57.27% of the produced sinter and the incipient formation of 20% molten phase (olivine).

However, as discussed previously, the briquetting of chromite ore fines requires costly binders, exhibits poor performance in the ferrochrome production. The making of green pellet with suitable cold handling strength from a chromite ore is difficult. Also, the crushing of raw materials for pelletization and the subsequent firing of the green pellets for strength development is an energy-intensive and complex process. In contrary, sintering of friable chromite ore can be done directly without any grinding. Sintering of chromite fines possesses some advantages, such as uniform size distribution in SAF bed and better metallurgical performance in the submerged arc furnace (SAF). Electrical resistivity of chromite sinter at high temperature is much higher⁵⁾ and the unit electricity consumption is also lower than lump ore or cold bonded briquettes due to more open structure. In this respect, direct sintering of chromite ore fines in the presence of coke breeze (as in iron ore sintering) may be an alternative, cost-effective process of agglomeration which is likely to be free from the preceding limitations. However, sintering of chromite fines is still challenging, because the melting point of chromium spinel is too high to form a molten phase required for the bonding during the sintering. A systematic study for development of good quality sinter is imperative.

The objective of the current work is to develop good quality chrome sinter using the Indian chromite ore fines, which will be suitable for cold handling and easy smelting in the electric arc furnace, producing ferrochrome as the final product. For this purpose, an optimum combination of temperature, fluxes (added to increase the relative quantity of molten phase required for sintering of mass) and coke (added to attain the sintering temperature in a sinter pot) should be computed using the thermo-chemical software FactSage 6.1 and enthalpy balance calculation. Accordingly, sintering experiment of a mixture of chromite ore fines and fluxes with the predicted optimum composition at the optimum temperature should be carried out in (i) an electrical resistance furnace (100-g scale), without using carbon and (ii) a pot sintering set-up (10-kg scale), using the computed quantity of carbon to develop a good quality sinter with adequate amount of melt phase formation. Also, the characterization of the produced sinters in terms of their mechanical properties and phase identifications should be done.

2. Experimental

The chromite ore fines of Sukinda valley, India was used in this study. The size fractions of the as-received ore, lime, sand, bentonite and coke breeze are shown in Table 1. The chemical analyses of as-received ore, lime, sand, bentonite and coke breeze used in this study are shown in Table 2.

Table 1. Size distributions of raw materials used in sintering (wt%).

Size fraction, mm	Chromite ore fines (Wt%)	Lime (Wt%)	Sand (Wt%)	Bentonite (Wt%)	Coke breeze (Wt%)
10.0–8.0	–	–	–	–	–
8.0–5.0	7.75	–	–	–	–
5.0–3.0	2.92	–	–	–	25.16
3.0–1.0	14.8	12.20	16.6	–	38.01
1.0–0.125	41.8	50.76	79.98	3.85	26.61
<0.125	32.73	37.04	3.42	96.15	10.22

Table 2. Chemical analysis of raw materials used in sintering.

Mass%	Chromite ore	Lime	Sand	Bentonite	Coke breeze
Cr2O3	46.27	−	0.1	0.1	−
Fetotal	16.83	−	2.5	2.9	1.7
CaO	0.39	94	5.3	1.2	0.7
SiO2	4.76	1.5	87.7	60.84	11.5
MgO	10.4	−	0.7	2.57	0.2
Al2O3	13.02	1.2	1.9	17.81	4.5
LOI	−	3	−	9.59	−
F. Carbon	−	−	−	−	80
K2O+Na2O	−	−	−	1.61	−
Rest	Oxygen with iron and other minor gangue	other minor gangue	other minor gangue	other minor gangue	other minor gangue

Theoretical calculations were done using the thermochemical software FactSage 6.1. Equilib module for open system was used to predict the proportion of different molten phases formed on heating chromite ore fines at different temperatures and at one atmosphere pressure. The calculations were done with the Fact and FToxide database. Flux requirement for the formation of molten phases in varying proportions during sintering were also estimated from this calculation.

2.1. Experiments in Electrical Resistance Furnace

Based on such a calculation for the formation of 25% molten phase, experiments were performed in a resistance furnace by taking a mixture of 100 g chromite ore fines, 0–5 g pure quartz, 0–10 g pure lime and 1.5 g bentonite. All the raw material was mixed and balled up using requisite quantity of water with the help of a pelletizer into pellets of 10–12 mm size. The dried pellets were put in a recrystalized alumina crucible to heat at the rate of 10°C/min up to 400°C, then 7°C/min up to 1000°C and 4°C/min beyond it and soaked in a chamber furnace at three different temperatures, 1500, 1550, and 1600°C for 20 minutes of retention. The heat-treated pellets were tested for cold crushing strength (CCS) by Hounsfield materials testing machine (Model: H 10K-S). Characterization of these sintered pellets was done by XRD (Brucker D8 Discover model, Cu K_α source to identify the phases), EPMA/EDS (JEOL JXA-8230 EPMA, 20 KeV) and image analysis (by LEICA MW software interfaced with the microscope) to determine the extent of sintering and pre-reduction at each temperature. The optimum sintering temperature (1600°C) was estimated from the preceding results. Enthalpy calculations were made to estimate the coke requirement to achieve this sintering temperature (flame front temperature) in a pot sintering set-up. For this purpose, a 25% heat loss by radiation and conduction from the system, enthalpy changes of different reactions and the enthalpy of formation of 25% molten phase etc. were considered. The details of which are presented in the subsequent section.

2.2. Experiments in Sintering Pot

Sintering experiments were conducted in a 10-kg pot sinter machine, based on the estimated coke and flux requirement to achieve 1600°C flame front temperature. The schematic of sintering in the pot sintering set up is given in Fig. 1. The 10 kg of sinter mix containing chromite ore fine, coke fine, and fluxes, were balled in a rotating drum (pelletizer) with the required quantity of water spray. The mixed mass gradually bonded to make balls (pellets) of average diameter 2–8 mm. This balled mix was next charged in the pot sintering set-up of 380 mm bed height and 140 mm diameter to make the sinter. Ignition was done at the top and the sintering was conducted at the suction pressure of 600 mm water column (WC). Temperatures at different heights of pot were recorded to observe the progress of the flame front. The temperature of the middle of the bed and the highest exit gas temperature were found to be about 1615°C and 260°C, respectively. The sintering was considered to be complete when the temperature of the bottommost thermocouple started to decrease after attaining the maximum temperature and the time elapsed up to this point was termed as ‘sintering time’. The sintered mass was cooled down to get outlet gas temperature below 100°C and then taken out of the pot through side door. The cold sinter was subsequently put to shatter, tumble, and abrasion tests. The shatter test was performed, as per the Bureau of Indian Standard Number IS: 9963, by dropping a 2 kg of chromite sinter in the size range +10 to –50 mm from a height of 2 meters on a steel plate, and screened to collect the +5 mm size fractions. The percentage of +5 mm size fraction was reported as shatter index. The tumbler and abrasion index of the chromite sinter were also done according to the Bureau of Indian Standard (IS: 6495). The sinter was analyzed chemically and phase identification was done by the EPMA-EDX and XRD analyses. The relative amounts of the molten and solid phases formed during the sintering were determined from these analyses.

Fig. 1.

Flow sheet for chromite sintering in sinter pot.

3. Result & Discussion

Chemical composition close to that of the chromite ore (Table 2) was considered for the calculation of relative amounts of different phases at equilibrium at different temperatures, using the software FactSage 6.1. The results are given in Table 3.

Table 3. FactSage 6.1 calculations of the relative amounts of solid and molten complex oxides at equilibrium at different temperatures.

Phases	Melting Point °C	Weight% of Phases at different temperatures, °C								Remarks
		1100	1200	1300	1400	1500	1600	1700	1800
FeCr2O4_solid	1900	60.8	60.8	60.8	60.8	60.8	38.45	15.7	9.97	solid phases
MgO.Cr2O3_chromite	2300	0	0	0	0	0	19.2	43.7	43.65
MgAl2O4_spinel	1995	14.19	14.19	14.19	14.19	14.19	0	0	0
Mg4Al10Si2O23_sapphire	2000	3.36	3.36	3.36	3.36	3.36	3.36	0	0
Mg2SiO4_forsterite	1890	9.99	9.99	9.99	9.99	9.99	9.99	0	0
FeAl2O4_hercynite	1750	0.19	0.19	0.19	0.19	0.19	17.5	15.7	15.7
Ca2MgSi2O7_akermanite	1400	0	0	0	0	0	0	0	0	molten phases
CaAl2Si2O8_anorthite	1450	3.22	3.22	3.22	3.22	3.22	3.22	3.22	3.22
FeO_wustite(s)	1377	0	0	0	0	0	0	9.9	9.9
Mg2Al4Si5O18_cordierite	1460	0	0	0	0	0	0	9.3	9.3

Table 3 shows that the phases FeCr₂O₄, MgCr₂O_4, MgAl₂O₄, Mg₄Al₁₀Si₂O₂₃, Mg₂SiO_4, FeAl₂O₄ with melting points above 1700°C are termed as solid phases; the phases Ca₂MgSi₂O₇, CaAl₂Si₂O₈, FeO and Mg₂Al₄Si₅O₁₈ with melting points below 1500°C are termed as molten phases. According to the heat calculations up to 1600°C, the proportion of the molten phases is very low (~3.5%). However, at 1600°C, a good quantity (17.5%) of the solid phase FeAl₂O₄ (m.p., 1750°C) is formed. The percentage of the molten phases increases at 1700°C and above, but sintering at this high temperature will be almost impracticable. Therefore, the possibility of sintering at the temperature of 1600°C was assessed through the calculation. At 1600°C, the complex oxide CaAl₂Si₂O₈ is the lone molten phase, which accounts for 3.22% of all phases formed. To increase the extent of formation of the molten phases, the calculation were redone at 1600°C with varying quantities of CaO and SiO₂ added to the ore, as shown in Table 4.

Table 4. Effect of flux addition to the proportion of molten phase formed during sintering at 1600°C, as calculated by FactSage 6.1.

Composition	%SiO2 with respect to Cr-ore fines	%CaO with respect to Cr-ore fines	Carbon	% of liquid phases at 1600°C
1	0	0	0	3.22
2	0	2 to 12	10.8	12–24
3	1 to 6	10	10.8	24–32

Table 4 shows that the addition of flux increases the relative amounts of molten phases. Figure 2 shows that with increase in concentration of lime up to 10%, the proportion of molten oxide phases increases to about 24%. Further increase of lime addition doesn’t increase the percentage of molten phase. Notably, a large quantity CaO is neither desirable because that will increase the slag volume, increasing the cost of production and causing problem in the downstream processing. With the use of about 10% CaO, 0% SiO₂ and 10.8% carbon (which on combustion provides heat to a sinter pot allowing it to attain the sintering temperature), the molten phases of total at 24% is formed that includes CaAl₂Si₂O₈_anorthite (melting temp 1553°C), Ca₂Al₂SiO₇_gehlenite (melting temp range 1410–1590°C), and MgCa₂Si₂O₇_akermanite (melting temp range 1350–1450°C) as shown in Fig. 2.

Fig. 2.

Effect of CaO addition to chromite ore, on the proportion molten phase formation at 1600°C as calculated by FactSage 6.1.

Possibility of further increase in the proportion of molten phase was explored by the addition of SiO₂, along with the 10% addition of CaO. The calculated result is given in Table 4 and Fig. 3. By increasing the SiO₂ addition from 1 to 6%, as shown in Fig. 3, the percentage of molten phase formed increases from 24% to 32%. Silica lowers the melting point by breaking the chromite spinel structure into the low-melting iron-silicate, calcium-alumino-silicate, and magnesium-calcio-silicate. However, further increase in SiO₂ addition does not increase the extent of the formation of the molten phase. Thus, according to the FactSage 6.1 calculation, and as shown in Fig. 3, addition of 10% CaO, 5% SiO₂, and 10.8% C to the chromite ore is likely to produce 30% molten phase during sintering. The remaining 70% (solid phase) will be constituted by the high-melting MgCr₂O₄, MgAl₂O₄ and FeCr₂O_4. This composition of the charge material (chromite ore + 10% CaO + 5% SiO₂ + 10.8% C) was considered as the optimum charge composition and served as the reference for the subsequent sintering study at 1600°C in (a) an electrical resistance furnace (100-g scale), without using the carbon and (b) a pot sintering set-up (10-kg scale), using the carbon.

Fig. 3.

Effect of SiO₂ addition to chromite ore, on the proportion molten phase formation at 1600°C as calculated by FactSage 6.1.

3.1. Study in Electrical Resistance Furnace

Experiments were performed in a resistance furnace by taking a mixture of chromite ore fines, pure quartz, pure lime and bentonite. Three compositions were considered to study the effect of composition, as shown in Table 5. The cold crushing strengths (CCS) of these sintered pellets were measured. The change of CCS with sintering temperature is shown in Fig. 4. CCS increases monotonically with increase in temperature for each pellet composition. As can be seen, when sintered at 1600°C, pellets made of chromite ore fines only (composition 1) have an average CCS of 150 kg, pellets of composition 2 of 280 kg, and pellets of composition 3 of 300 kg. The sintered pellet with higher flux content provides better strength than lower. This is due to higher amount of molten phase formation during heating as has been seen from FactSage calculation. Though, 280–300 kg CCS is available in sintered pellet, the sinter made with same composition of mixture in a sintering strand/pot may not provide so high strength, because of its less packing/open structure. Therefore, for the subsequent sinter making in sintering pot, composition 3 was considered to get a suitable strength for handling. Notably, the CCS values for any given pellet composition were very low for the two smaller temperatures (1500 and 1550°C). Further more there was an uncertainty of ±10°C in the measurement of temperature. Therefore, the actual (pot) sintering of chromite fines (the results of which will be discussed subsequently) was conducted at 1600°C.

Table 5. Composition of charge materials for chromite ore sintering in resistance furnace.

Composition No	%SiO2 with respect to Cr-ore fines	%CaO with respect to Cr-ore fines	%Bentonite with respect to Cr-ore fines
1	0	0	1.5
2	2.5	10	1.5
3	5	10	1.5

Fig. 4.

Cold crushing strength of sinter (pellet), in kg, vs temperature.

The XRD analysis of the sintered pellets, for each of the three temperatures, 1500, 1550, and 1600°C, are given in Fig. 5. It is evident that with the increase in temperature, the percentage of the molten phases (Fe₂SiO₄ (melting point, 1220°C), Ca₂Al₂SiO₇, and MgCa₂Si₂O₇) increases. It should be noted that the melting point of Fe₂SiO₄, not listed in Table 3, is 1220°C. To further confirm the formation of these phases in the sintered mass, microstructure analyses of the chromite ore fines and the chromite sinter were done. Figure 6 shows the back scattered electron image of the as received chromite ore fines which has Mg–Cr–Fe–Al rich and Fe–Si rich phase. The back scattered electron image with EDS analysis result of sintered pellet is shown in Fig. 7 that comprised of both solid and liquid phases. The area analysis of the micrographs shows 22–30% of area is molten phase (Fe₂SiO₄, Ca₂Al₂SiO₇_gehlenite, MgCa₂Si₂O₇_akermanite as evident from the XRD study) depending upon flux composition and remaining area is covered by solid phase (Mg, Fe)(Cr, Al)₂O₄, as evident from the XRD study). Based on the density data¹⁰⁾ of above three oxide phases (Fe₂SiO₄, Ca₂Al₂SiO₇, MgCa₂Si₂O₇), the density of molten phase comes about 3.0 g/cm³ and the density of solid phase, which is mainly FeCr₂O₄ is 5.0.¹⁰⁾ Because of this density difference mass percentage of corresponding 22% molten phase area would be, {(3×0.22)*100/(5×0.78 + 3×0.22)} = 14.5%. Hence, the mass percentage range of molten phase, for varying flux composition will be approximately, 14.5–20.5%. Therefore, around 20% molten phase was obtained for the pellet with flux containing 10% CaO and 5% SiO₂ on heating at 1600°C in resistance furnace.

Fig. 5.

XRD pattern of (fluxed chromite ore fines) sintered pellets at different temperature in resistance furnace.

Fig. 6.

Back scattered electron image of as received chromite ore fines.

Fig. 7.

Back scattered electron image with EDS analysis result of fluxed chromite ore fines pellet at 1600°C in resistance furnace.

3.2. Study in Sintering Pot

Experiments were also carried out in a sinter pot on 10-kg scale to study the sintering of chromite ore fines, starting with the equivalent flux composition as in the resistance furnace experiments, plus the carbon needed to attain the sintering temperature of 1600°C. Enthalpy calculation based on the sintering of one kg chromite ore fines was done to find the fuel (carbon) requirement by balancing the heat input and output as follows:

Heat Sink

i) Heat absorbed to heat the charge up to the sintering temperature

Heat absorbed by all the individual oxides viz. Cr₂O₃, FeO, Al₂O₃, MgO, CaO and SiO₂ present in the charge to heat up to 1600°C, was calculated using respective specific heat (C_p values) as per equation   

$$\mathrm{\Delta }{\text{H}}_{\text{T}2}^{°}=\mathrm{\Delta }{\text{H}}_{\text{T}1}^{°}+\underset{\text{T}1}{\overset{\text{T}2}{{\displaystyle \int }}}\mathrm{\Delta }\text{Cp}\mathrm{\hspace{0.17em} }\text{dT}$$(1)

  

$${\text{C}}_{\text{p}}=\text{a}+\text{bT}-{\text{CT}}^{-2}$$(2)

T₁ was taken as 298 K (room temperature) and T₂ = 1873 K (sintering temperature).

ii) Heat of melting

Heat requirement for melting of the melt phases (viz Fe₂SiO₄, Ca₂Al₂SiO₇ and, MgCa₂Si₂O₇) was calculated from the respective latent heat of fusion from the equilib module of Factsage 6.1

iii) Heat of moisture evaporation

Heat necessary to evaporate the moisture was calculated as: heat required to raise the temperature of the water to 100°C plus latent heat of vaporization of water.   

$$\mathrm{\Delta }\text{H}=\text{m}\underset{298}{\overset{373}{{\displaystyle \int }}}\text{Cp}\mathrm{\hspace{0.17em} }\text{dT}+\text{m}\mathrm{\hspace{0.17em} }\text{*}\mathrm{\hspace{0.17em} }\text{Lvap}$$(3)

Where m = mass of water added to sinter mix and L_vap is Latent heat of vaporization of water

iv) Heat with off gas

Heat carried away by off gas was calculated from its C_p value and temperature of the exit gases.

v) Heat loss through wall

Heat loss due to conduction through wall and radiation was assumed to be 25% of the total input heat to the pot sinter.

Heat Source

i) Burning of coke

Heat available from the burning of carbon in coke breeze was calculated considering following reaction.   

$$\text{C}+{\text{O}}_2={\text{CO}}_2;\mathrm{\Delta }{\text{H}}^0{}_{{\text{CO}}_{\text{2}}}\text{at}\mathrm{\hspace{0.17em} }1\mathrm{\hspace{0.17em} }873\text{ K}=-397\mathrm{\hspace{0.17em} }048.6\mathrm{\hspace{0.17em} }\mathrm{J}/\text{mol}$$(4)

ii) Heat of formation of phases

Heat of formation of major complex phases (Fe₂SiO₄, Ca₂Al₂SiO₇ and, MgCa₂Si₂O₇) formed by the reactions with gangue materials (CaO, Al₂O₃, SiO₂ MgO etc.) at sintering temperature was calculated by using Reaction module of Factsage 6.1.

iii) Heat of exothermic oxidation of oxide

FeO in chromite may get oxidized to Fe₂O₃. Therefore the following reactions have been considered for the calculations.   

$$\begin{array}{c}2\text{FeO}+½{\text{O}}_2={\text{Fe}}_2{\text{O}}_3;\mathrm{\hspace{0.17em} }\hfill \\ \mathrm{\Delta }{\text{H}}^0{}_{{\text{Fe}}_{\text{2}}{\text{O}}_{\text{3}}}\mathrm{\hspace{0.17em} }\text{at }1\mathrm{\hspace{0.17em} }773\text{ K}=-275\mathrm{\hspace{0.17em} }549.8\text{ J}/\text{mol}\hfill \end{array}$$(5)

Equating the heat source and heat sink, amount of carbon to raise the temperature to 1600°C in the pot sinter has been calculated. This comes to 10.3 g per 100 g of chromite ore fines. Thus, the requirement of coke having 80% fix carbon will be 12.8%.

Based on the above calculation, the sintering of chromite ore fines with flux composition of 10.6% lime, 5.7% sand and 1.5% bentonite was conducted at 13% coke breeze with respect to chromite ore fines. It was also carried out in varying coke rate from 10–15% to identify the actual requirement of coke breeze. Reproducibility tests have also been conducted in varying coke breeze percentage with a total of around twenty four experiments in 10-kg scale. Figure 8 shows the temperature increase of a typical point in sinter bed with time during the sintering of chromite ore fines in the pot sinter. The curve shows that the flame temperature reaches beyond the desired temperature of 1600°C. The flame front movement up to the bottom was completed by 30 min (sintering time). A picture of the sintered mass inside the sintering pot is shown in Fig. 9. Figure 10 presents the photograph of chromite sinters on a tray. The molten phases bonded well with the solid phases to produce porous (and partly reduced) sinter mass. The strength of the chromite sinters was characterized by tumbler index, shatter index and abrasion index, and the results with different coke additions to the charge mix, are shown in Table 6. It is depicted from the table that both shatter index and tumbler index increases and abrasion index decreases with increase in percentage of coke in sinter mix. This is because; increasing amount of heat at high coke percentage increases the amount of slag phase (molten phase) in sinter which enhances the bonding between solid particles. All tumbler index, shatter index and abrasion index of the sinter with addition up to 13% coke is not so encouraging and may not be acceptable for a chromite sinter for handling in subsequent operation. With increasing amount of coke to 15%, both the tumbler index and shatter index reaches to 77% and 87.4%, respectively and abrasion index comes down to 5.8%. These values appear well acceptable for further use of chromite sinter to make ferrochrome in the submerged electric arc furnace. Chemical analysis of the produced chromite sinter was done and the result is given in Table 7. The produced sinter contains high percentage of gangue such as CaO and SiO₂. This is due to the addition of fluxes to increase molten phase formation in sintering. These fluxes will reduce the flux consumption in next step during smelting of sinter in SAF. X ray diffraction analysis of chromite sinter is shown in Fig. 11 to identify the phases formed in the actual sinter in the sintering pot.

Fig. 8.

Change of temperature of a typical point in bed with time during sintering in the pot sinter.

Fig. 9.

Picture of the produced chromite ore sinters inside the pot sinter machine.

Fig. 10.

Chromite sinters collected on a tray.

Table 6. Tumbler, shatter and abrasion index of chromite sinter as a function of % coke addition to the charge.

Sintering Experiment No	% coke breeze	Tumbler Index, %	Shatter Index, %	Abrasion Index, %
1	10	49.2	70.1	12.6
2	13	53.6	78.8	11.1
3	15	77	87.4	5.8

Table 7. Chemical analysis of the produced chromite sinter from sinter experiment 3.

Chromite Sinter	Cr2O3	Fe (T)	CaO	SiO2	MgO	Al2O3	Other gangue+ Oxygen with Fe
%	35.73	14.52	5.79	7.65	8.54	11.01	16.76

Fig. 11.

XRD pattern of as received chromite ore fines and its sinter.

The phases identified in the X-ray diffraction of the chromite ore sinter produced in the sintering pot are compared with those present in the chromite ore fines to confirm the formation of the molten phases, as shown in Fig. 11. The peaks marked as 1 and 2 in the diffractogram of chromite ore fines (at room temperature) confirm the presence of chromite spinel, PDF No 00-009-0353 and iron silicate (Fe₂SiO₄), PDF No 00-002-0184. These peaks were verified by the database and the primary phase (peak 1) was identified as magnesio-ferro-alumino-chromite. The sinter peaks 4 and 5 were identified as gehlenite (Ca₂Al₂SiO₇), PDF no 00-020-0199 and akermanite (Ca₂MgSi₂O₇), PDF No 01-076-0841, the two molten phases formed during sintering. Thus the phases found in XRD analyses were mainly (Mg, Fe)(Cr, Al)₂O₄ (solid) and Fe₂SiO₄, Ca₂Al₂SiO₇, and Ca₂MgSi₂O₇ (molten). The existence of different phases obtained under EPMA is shown in Fig. 12. The elemental analysis of the corresponding phases under EPMA-EDS is also shown in Fig. 12. A comparative picture on the existence of these phases in resistance furnace, actual sintering and theoretical calculation is given in Table 8. It shows that the phases present in the chromite sinter produced from the resistance furnace experiments are in agreement with the phases present in the chromite sinter produced from the pot sintering experiments. These experimentally found phases are also in compliance with the phases predicted from the theoretical calculations made by the software, FactSage 6.1.

Fig. 12.

Back scattered electron image with EDS analysis result of chromite sinter at 1600°C.

Table 8. Comparison of theoretical and experimental results on chromite ore sintering.

Melting point, °C	Phases predicted by FactSage calculation	Phases found in chromite sinter produced in resistance furnace experiments	Phases found in chromite sinter produced in pot sintering experiments
~ 2300	(MgO)(Cr2O3)_chromite	(Mg, Fe)(Cr, Al)2O4	(Mg, Fe)(Cr, Al)2O4
~ 1995	MgAl2O4_spinel
~ 1900	FeO.Cr2O3		FeO.Cr2O3
1410–1590	Ca2Al2SiO7_gehlenite	Fe2SiO4	Fe2SiO4
1553	CaAl2Si2O8_anorthite	Ca2Al2SiO7	Ca2Al2SiO7
1350–1450	MgOCa2O2Si2O4_akerman	Ca2MgSi2O7	Ca2MgSi2O7

The mass percentages of different elements are in coherence with the stoichiometric percentage of the phases found in XRD study as above. The Fig. 12 also represents the molten and solid phase formed during sintering. The area percentages of solid and molten phases have also been measured by the area analysis software to get an approximate idea on quantity of molten phase. In Fig. 12, point 1 and point 3 represent the two molten phases Ca₂Al₂SiO₇ and Fe₂SiO₄, while point 2 and 4 represent the solid chromite spinel: (Mg,Fe)O.(Cr,Al)₂O_3. The area measurement of Fig. 12 yielded the area percentage of molten and solid phases as 28% and 72%, respectively. It may be noted that area percentage and mass percentage are not numerically same and the densities of different phases are also not same as mentioned earlier. However, a rough estimation of the mass percentage of molten phases was done using densities of different phases as mentioned in section 3.1 and it comes out to be approximately 19%. Quantitatively the mass percentage of molten phases calculated from experiments appears lower than the mass percentage of molten phases predicted from the theoretical calculations. This may be attributed to the fact that equilibrium was not obtained within the very short period of high temperature (~1600°C) in the sinter bed (Fig. 8). Kobayashi et al.¹¹⁾ also found a significant difference between the calculated and actual phase percentage for Ni ore reduction in rotary kiln due to the same reason.

4. Conclusion

Considering the importance of incipient formation of molten phases during sintering, thermodynamic possibility of the formation of different molten phases during the heating of chromite ore to high temperatures was calculated by the thermochemical software FactSage 6.1. Flux requirements for the formation of molten phases during sintering at the optimum temperature of 1600°C were also calculated by the software. According to these results, addition of 10% CaO, 5% SiO₂, and 10.8% C (as fuel) to the chromite ore is likely to produce 30% molten phase during sintering at 1600°C. This composition of the charge material was considered as optimum and served as the reference for the subsequent sintering at 1600°C in (a) an electrical resistance furnace (100-g scale), without using the carbon and (b) a pot sintering set-up (10-kg scale), using the carbon. The phases formed in these sintering experiments sinter were identified by XRD and EPMA. The mass percentage of the molten phase in produced sinter was found to about 19% and rest was solid phase. The molten phases were identified as Fe₂SiO₄, Ca₂Al₂SiO₇ and Ca₂MgSi₂O_7, which bound the solid phases, such as (Mg,Fe)(Cr,Al)₂O₄, in the sinter. Mechanical characterization of the produced sinter was done by shatter, tumbler and abrasion tests. Sinters were found to have adequate handling strength (for example, 85% shatter index, 75% tumbler index, and 5% abrasion index) to be acceptable for use in the downstream process of producing ferrochrome/charge chrome in the submerged arc furnace.

Acknowledgements

The authors would like to express their sincere gratitude to the Director, CSIR-National Metallurgical Laboratory, Jamshedpur, for his kind permission to publish this work.

References

• 1)  Indian Bureau of Mines: Indian Minerals Yearbook 2011, Part-II, 50th ed. India, (2011), 23-1.
• 2)   C. N.  Harman and  N. S. S.  Rama Rao: Proc. of INFACON XI, Macmillan Publishers India, Chennai, (2007), 67.
• 3)   A.  Chatterjee: Proc. of India-1st Ferro-Alloys, The Indian Ferro Alloy Producers Association, Mumbai, (2004), 1.
• 4)   C. R.  Ray,  P. K.  Sahoo and  S. S.  Rao: Proc. of INFACON XI, Macmillan Publishers India, Chennai, (2007), 63.
• 5)   B.  Nandy,  M. K.  Choudhury,  J.  Pal and  D.  Bhattacharjee: Metall. Mater. Trans. B, 40 (2009), 662.
• 6)   R. H.  Nafziger,  P. E.  Sanker,  J. E.  Tress and  R. A.  McCune: Ironmaking Steelmaking, 9 (1982), 267.
• 7)   H. G.  Vararlis and  A.  Lekatou: Ironmaking Steelmaking, 42 (1993), 42.
• 8)   O.  Soykan,  R. H.  Eric and  R. P.  King: Metall. Mater. Trans. B, 22B (1991), 53.
• 9)   D.  Zhu,  J.  Li,  J.  Pan and  A.  He: Int. J. Mineral Processing, 86 (2008), 58.
• 10)   J. W.  Anthony,  R. A.  Bideaux,  K. W.  Bladh and  M. C.  Nichlos: Handbook of Mineralogy, Vol. II, Mineral Data Publishing, Mineralogical Society of America, Chantilly, (2005), 122.
• 11)   Y.  Kobayashi,  H.  Todoroki and  H.  Tsuji: ISIJ Int., 51 (2011), 35.