Production of Ferro-Chrome


Production of Ferro-Chrome

Ferro-chrome (Fe-Cr) is an alloy comprised of iron (Fe) and chromium (Cr) used primarily in the production of stainless steel. The ratio in which the two metals (Fe and Cr) are combined can vary, with the proportion of Cr ranging between 50 % and 70 %.

Fe-Cr is frequently classified by the ratio of Cr to carbon (C) it contains. The vast majority of Fe-Cr produced is the ‘charge chrome’. It has a lower Cr to C ratio and is most commonly produced for use in stainless steel production. The second largest produced Fe-Cr ferro-alloy is the ‘high carbon Fe-Cr (HC Fe-Cr) which has a higher content of Cr and is being produced from higher grade chromite ore. Other grades of Fe-Cr are ‘medium carbon Fe-Cr’ (MC Fe-Cr) and ‘low carbon Fe-C (LC Fe-Cr). MC Fe-Cr is also known as intermediate carbon Fe-Cr and can contain upto 4 % of carbon. LC Fe-Cr typically has the Cr content of minimum 60 % with C content ranging from 0.03 % to 0.15 %.  However C content in LC Fe-Cr can be upto 1 %.

Ferro-chrome (Fe-Cr) alloy is essential for the production of stainless steel and special steels which are widely used and are of high quality, typically characterized by a high corrosion resistance and a low tendency to magnetization. The processing cycle of Fe-Cr involves the chemical reduction of the chromite ore.

Smelting of HC Fe-Cr ferro-alloy

HC Fe-Cr and charge chrome are normally produced by the conventional smelting process utilizing carbo-thermic reduction of chromite ore (consisting oxides of Cr and Fe) using an electric submerged arc furnace (SAF) or a DC (direct current) open arc electric furnace.

In SAF, the energy to the furnace is predominantly supplied in a resistive heating mode. The main features of this mode are the electrical resistivity of the slag and the slag liquidus temperature that are strictly selected to operate the process comfortably. These two parameters also impose some restrictions to the smelting process in terms of operating temperature.

SAF used for the smelting of chromite ores are of two types namely (i) closed type, and (ii) open type. Closed type furnace offers the opportunity to collect the carbon mono oxide (CO) rich off gas for preheating and partial pre-reduction of the burden.  Preheating and partial pre-reduction of the burden results in a significant reduction of the electric energy consumption. However, the operation of a closed furnace demands more care in the burden preparation for a smooth production.

The DC arc furnace uses a single, central hollow graphite electrode as the cathode, with an electrically conducting refractory furnace hearth as the anode. The furnace operates with an open bath, so there is no problem with overburden, and the chromite ore fines, together with coal and fluxes are fed directly into the bath through the hollow electrode. The furnace has a closed top. Some of the characteristics of DC arc furnace operation are (i) use of fine ores without agglomeration, (ii) use of cheaper reductants and hence there is greater choice of reductants, (iii) higher recoveries of Cr, (iv) deliberate changes in the charge composition are reflected rapidly in the slag or ferro-alloy, and (v) closed top operation allows furnace off-gas energy to be used.

The energy to a DC open arc furnace is mainly supplied in an arcing mode. This energy is largely independent to the slag chemistry. Thus it provides more freedom in the selection of the slag composition and process temperature. This freedom has conferred to the DC open arc furnace a greater ability to control silica (SiO2) reduction more closely by slag chemistry. A reasonably large range of process temperatures can be achieved in DC open-arc as compared to SAF. Hence in DC open arc furnace sub-liquidus and superheated slags can be produced in the smelting process. Superheated slag improves to a certain extent the kinetics of chemical reactions and the slag-metal separation whilst sub-liquidus slag has favourable effect on the furnace lining. Though change to the slag chemistry is possible, this is generally restricted by the economics of the process.

The conversion of chromite ore to Fe-Cr is dominated by SAF smelting where the electrodes are buried in the burden of lumpy material comprised of chromite ore, carbonaceous reductants, predominantly coke, and fluxes to form the correct slag composition. The electric current is 3-phase alternating current (AC) and the furnace has three equally spaced consumable self-baking graphite electrodes in a cylindrical, refractory lined container with a bottom tap-hole. Characteristics of the SAF for smelting chromite ore include (i) relatively easy to control provided the charge is well sorted to maintain a permeable overburden which allows easy escape of the gases produced, (ii) self-regulating with power input determining the rate of consumption of charge (overburden), and (iii) some pre-heating and pre- reduction of the overburden by the hot ascending gases. SAF can be open, semi-closed or closed top with correspondingly better thermal efficiency and the ability to make use of the energy in the off-gases from the closed top furnace.

Fe-Cr smelting (Fig 1) is an energy intensive process. The energy is transferred to the process via self-baking carbon electrodes. The electrodes are laterally covered by the lumpy burden which has a higher electric resistivity than the molten bath below the electrode tips. Hence it is assured that the electric current flows to the hot reaction zone. The energy consumption is relatively high, and for HC Fe-Cr the electrical energy consumption varies between 2000 kWh/ton of Fe-Cr with pre-reduction to 4000 kWh/ton of Fe-Cr without pre-reduction and feed preheating. Because of the complexity of feed structure and electrical-thermal-chemical interactions, large temperature gradients exist in the furnace from a few hundreds at the surface of the burden to well over 2000 deg C around the electrode tips. This leads to various zones in the furnace and different reduction mechanisms.

Fig 1 Schematics of Fe-Cr smelting in SAF

In the furnace, a single chromite pellet or lumpy ore experiences an increasing temperature environment while the charge is descending, and is reduced by the ascending CO gas and promoted by contacted coke particles. It is apparent that temperature profile in the SAF has a big influence on the reduction rate and the production efficiency. Due to the sensitivity of the electrode control system to the distribution of the furnace temperature, the temperature distribution within the feed and various reaction zones are usually not symmetrically distributed. This uneven temperature distribution causes the difficulties in furnace control, product quality, and furnace efficiency.

Zones in a SAF

From investigations of excavated quenched SAFs, and overviews of the technology and the process steps, six numbers of the idealized reaction zones can be identified. However the exact positions of these zones can vary with furnace design and operating practice. The zones need not necessarily follow a simple layered structure. Schematic diagram of the reaction zones in submerged arc furnace for Fe-Cr production is shown in Fig 2.

Fig 2 Schematic diagram of the reaction zones in submerged arc furnace for Fe-Cr production

The first zone (zone 1) is the upper furnace zone. This zone has loose charge which extends from the top of the charge layer down to near the tip of the electrode. The activities which are taking place in this zone are (i) preheating of the charge (ii) decomposition (calcination) of fluxes e.g. lime stone, dolomite etc., (iii) gasification of carbonaceous material due to the reaction with air and carbon di-oxide (CO2), and (iv) gaseous reduction of chromite ore and partial metallization of Fe and Cr oxides. In fact, most of the volume in the SAF is having the loosely sintered burden. The average retention time in this zone has been estimated to be 24 hours, but only around 20 % of reduction of the charge takes place in this loose charge zone and no liquid slag is formed. The burden material in this zone usually descends in a V shaped distribution, and the rate of descent reaches a maximum at positions between the furnace walls and the electrodes and between the electrodes themselves. Data related to the temperature profiles and excess gas pressures in this zone show that the 1600 deg C isotherm is achieved only close to the electrode tips, and that above 1400 deg C the gas pressure rises rapidly, the later temperature corresponding to the onset of slag formation.

Zone 2 to zone 6 exists in the lower part of the furnace. The activities which are taking place in lower part of the furnace are (i) slag formation, (ii) dissolution of chromite ore in the slag, (iii) reduction of metal from the slag phase and metal alloy formation, and (iv) separation of alloy and the slag.

Zone 2 consists of sidewall slag, metal, ore, and coke. It has banks of rigid, partially fused and partly reduced materials which are formed adjacent to the furnace walls. These banks are thickest at distances furthest from the electrodes.

Zone 3 consists of sidewall slag and metal. It is the material below zone 2 and contains mixtures of slag and metal.

Zone 4 is the beneath the electrodes. The material present immediately under the electrode tips has some uncertainty. The zone under one of the electrode generally does not connect with the similar zones under the other two electrodes. The presence of a void can be due to the contraction of the bed during cooling of the furnace. There is also possibility of slag and coke under the electrodes. The other possibility can be the presence of a coke bed, containing a mixture of melted gangue minerals, fluxes and magnesia (MgO) and alumina (Al2O3) liberated from the chromite ore during its reduction. Due to the formation of partially solidified charge materials around the electrodes (zones 2 and 3) the active slag reduction zone is restricted in size. The residence time in the high temperature smelting zone, defined here as the coke bed (zone 4), is relatively short, possibly of the order of 30 minutes to 40 minutes.

Zone 5 is the region where there is the presence of a large region of not melted partly reacted lumpy ore between the slag and metal.

Zone 6 is the region of the formation of a distinct liquid Fe-Cr alloy layer at the base of the furnace.

Process reactions

After reaching the zone below the electrode tips (zone 4), all oxides are molten and the carbo-thermic reduction can take place with the solid coke particles, according to the equation Cr2O3 + 3C = 2Cr + 3 CO. The remaining oxides such as SiO2, Al2O3, CaO, and MgO go into slag. This slag forms a liquid layer on top of the liquid Fe-Cr. Droplets of liquid Fe-Cr descend through this slag layer and collect at the bottom of the furnace.

During the production of Fe-Cr through carbo-thermic reduction, metallic Cr which is formed tends to react further with the available C to form Cr carbides (Cr3C2, Cr7C3, and Cr23C6). Similarly metallic Fe reacts with the available C to form carbides of Fe (Fe3C and Fe2C). The presence of these carbides increases the total C content of the Fe-Cr ferro-alloy beyond the specified limits since the theoretical C content of these carbides ranges from 5.5 % to 13.3 %. The following are the simplified reactions which are taking place during the process.

Cr2O3 (l) + 3 C = 2 Cr (l) + 3 CO (g)

3 Cr2O3 (l) + 13 C = 2 Cr3C2 + 9 CO (g)

2 Cr2O3 (l) + 7 C = Cr4C + 6 CO (g)

7 Cr2O3 (l) + 27 C = 2 Cr7C3 + 21 CO (g)

23 Cr2O3 (l) + 81 C = 2 Cr23C6 + 69 CO (g)

Fe2O3 (l) + 3 C = 2 Fe (l) + 3 CO (g)

3 Fe2O3 (l) + 11 C = 2 Fe3C + 9 CO (g)

Fe2O3 (l) + 4 C = Fe2C + 3 CO (g)

The standard Gibbs free energies for the formation of these carbides (by reaction between Cr2O3 dissolved in the slag and solid, unreacted C) have been studied by several researchers. These studies show that the Gibbs free energy values for the formation of Fe3C is lower than that for the formation of metallic Fe for the entire temperature range studied (1500 deg C to 2000 deg C). Thermodynamics therefore favour the formation of Fe3C.

Similarly, the reaction Gibbs energy for the formation of Cr3C2 is lower than that of metallic Cr for the most of the temperature range (upto around 1920 deg C).  Cr7C3 and Cr4C can also form at temperatures of 1650 deg C and 1550 deg C respectively. Several carbides therefore form preferentially to the metallic Cr and Fe during the reduction process of chromite ores.

As a result, LC Fe-Cr cannot be produced directly through carbo-thermic reduction of the chromite ore. The products from the carbo-thermic reduction are therefore HC Fe-Cr or charge chrome depending on the Cr to Fe ratio in the chromite ore.

The coke particles (quantity and size) have a big influence on the electric resistance in the reaction zone and in the burden column. Therefore the coke has a dual function, for the smelting reaction and for the conversion of electric energy into thermal energy. It also helps in keeping the burden permeable for the ascending CO gas.

The formation of carbides in the carbo-thermic reduction of Fe-Cr starts already at low temperatures. Slag temperatures of upto 1650 deg C always result to C content of 7 % to 8 % in the Fe-Cr. Only if the MgO content of the ore is high and the slag temperatures exceed 1700 deg C then the C content is in the range of 4 % to 6 %. Lower C levels cannot be reached in the SAF and a second process step is needed.

Raw materials

The type and combination of charge materials (ores, reductants and the fluxes) used for the smelting of the Fe-Cr influence furnace operations and ultimately affect the electricity consumption. Slag volume produced during the process is highly dependent on the quality of the charge materials used for the production process. The consumption of electrical energy is influenced by the slag volume. The higher is the volume of the slag higher is the requirement of the electric energy.

The furnace feed typically consists of chromite ore (lumpy ore, pellets, and briquettes), reductants (anthracite, char, coke, and coal), and fluxes (quartzite, dolomite, and lime). Chromite ore particles do not necessarily have uniform composition since there are variations between ore bodies, and even between and within ore seams. The chromite ores are generally part of the spinel crystal family, having the general formula (Fe2+, Mg2+)O.(Al3+,Cr3+,Fe3+)2O3.

Through careful control of the size range and composition of the charge materials, ideal furnace conditions can be sustained. Close control of the raw material type and size range ensures good permeability within the packed bed, while the composition influences the slag properties. There are, however, a number of charge pre-treatments which can be utilized in order to improve the furnace stability and productivity.

The agglomeration processes which are generally used for chromite ores include (i) sintering, (ii) pelletizing, and (iii) briquetting. Also the efficiency of the process for the production of Fe-Cr depends on the type and pre-conditioning of the feed materials namely (i) pre-heating, and (ii) pre-reduction.

In case of friable chromite ores, it becomes necessary to pelletize them, after further grinding if necessary, with binder, reductant and fluxes and pass them through a rotary kiln where they are hardened (sintered), pre-heated and pre-reduced to a degree before charging to the SAF.

The chromite ore charge to the furnace is mainly in the form of lump ore and /or pellets. In case of the lump ore, the material consists of chromite particles surrounded by solidified host rock. Pellets are normally made from the concentrates of the chromite ore which has been agglomerated and pretreated to form spherical particles. The grain size and microstructures of the chromite ores vary from dense rounded grains in the lump to highly fractured acicular grains in case of the pellets. The particle size ranges of lump, pellet, and coke are controlled to maximize bed permeability.

Another method for the treatment of ore fines is by kiln pre-reduction where un-agglomerated chromite fines and low cost coal, with fluxes are used as the feed to the kiln. In this method, self-agglomeration of the fines is achieved close to the discharge from the kiln where the charge becomes pasty in a high temperature zone of around 1,500 deg C. Very high degrees of reduction are being achieved (80 % to 90%) thus reducing the loads on the downstream electric furnaces (SAF or DC arc) which then become basically a melting furnace.

A more recent approach, and one which is being installed by several plants, is again by pelletizing. Pellets are produced with coke included and these are sintered and partly pre-reduced on a steel belt sintering system. From there, the pellets are delivered to pre-heating shaft kilns which are usually placed above SAFs and which operate as direct feed bins, making use of the off-gas heat from the furnaces. Lump ore, coke and fluxes are also directed to the feed bins.

In addition to the above, there are some other approaches to the preparation of the chromite ores for smelting. These include rotary hearth sintering and pre-reduction of pellets, and fluidized bed pre-heaters for chromite fines.

Production process

Fe-Cr is produced essentially by a carbo-thermic reduction which is taking place at high temperatures. Chromite ore is reduced by coal and coke to form the Fe-Cr alloy. The heat for this reaction can come from several forms, but typically from the electric arc formed between the tips of the electrodes in the bottom of the furnace and the furnace hearth. This arc creates temperatures of about 2,800 deg C. In the process of smelting, a large amount of electricity is consumed.

Production process for the Fe-Cr is highly electric energy intensive since all the heat needed for the endothermic reduction reactions and to achieve thermodynamic equilibrium in the furnace is supplied through electrical energy only. Thus electrical energy is the most vital input in the process.

Tapping of the material from the furnace takes place intermittently. When enough amount of Fe-Cr has accumulated in the hearth of the furnace, the tap hole is drilled open and a stream of liquid alloy and slag rushes down a trough into a chill or ladle. The liquid Fe-Cr solidifies in large castings, which is crushed, sieved and packed or further processed.

The following are the important characteristics of production process of Fe-Cr in a SAF.

  • Reducibility of chromite ore is important for process efficiency. It is determined by its mineralogy (MgO/Al2O3 ratio), the MgO structural form in the chromite ore, chromite grain size, the extent of grains distribution in ore and the ore melting temperature. The ore need to have porosity since solid-state reduction of chromite ore is of significance because of the reaction between gas and porous solid. Also, the ore has to have sufficient mechanical strength to withstand abrasion and crushing to reach to the furnace reduction zone. A high melting temperature of ore, if good reducibility in solid state exists, allows more time for the ore to be reduced before reaching the melting zone. This means less Cr2O3 losses to the slag.
  • Cr recovery mainly depends on reducibility of the chromite ore. From the input Cr to the furnace, other than recovered Cr to ferro-alloy, the balance mainly reports to slag and some to furnace dust. In case of hard and dense ore, the dust portion is low, but depending on ore reducibility and its size, the Cr2O3 content of slag can increase. Since reduction of chromite ore in solid state is significant, fine ores can be used in the furnace. The fine chromite ore is very readily reduced in solid state before it is melted, resulting in a Cr2O3 content in the slag in the range of 1.5 % to 4 %. Also it is important that there is complete separation of liquid ferro-alloy and slag during tapping so that the liquid Fe-Cr going to slag is minimal for the optimization of Cr recovery.
  • A good reducibility and especially solid state reduction help in the utilization of more CO gas for reduction of chromite ore. The ratio of MgO/Al2O3 in the slag equaling 2.1 corresponds to the lowermost melting point of the slag resulting into a minimum of energy consumption for the portion of slag to be melted. However, due to the presence of some Cr2O3 in the slag and other oxides, MgO/Al2O3 ratio has the typical value of 2.2. This also results into lower reduction of SiO2 to Si. Since reduction of SiO2 to Si being highly energy intensive there is good energy saving on this account as well.
  • Reduction at lower temperatures with a high MgO/Al2O3 ratio in the slag result into formation of higher amount of high C bearing carbides. The presence of Cr carbides such as Cr3C2 and Cr7C3 with less Cr23C6 indicates a better solid state reduction of the chromite ore. Cr/Fe ratio also plays a role in the determination of the C content of the Fe-Cr. As Cr has higher affinity to form carbides than Fe, a higher Cr/Fe ratio means a higher C content in the Fe-Cr.
  • A high ratio of MgO/Al2O3 in the slag and a basic slag inhibit reduction of SiO2 to Si. Higher temperatures are needed for the reduction of SiO2. Cr2O3 and FeO are reduced in solid state and in lower temperatures, thus there is very little opportunity for SiO2 to be reduced. Again a low Si content is a result of reduction of Cr2O3 and FeO in solid state and low temperatures.
  • Normally the sulphur (S) and phosphorus (P) content of the chromite ore is not high. Hence S and P in the Fe-Cr come predominantly from the reductants. A coke having the S content in the range of 0.6 % to 1 % contributes in the Fe-Cr a S content of 0.014 % to 0.025 %. For desulphurisation, basic slags, a reducing atmosphere and higher temperatures with respect to dephosphorization are necessary. These conditions do exist. For dephosphorization the requirements include basic slags, lower temperatures and an oxidizing atmosphere. From all of these, only a basic slag condition is available. An increase in the Si content of the ferro-alloy indicates a reducing atmosphere. Even though most of the P enters the ferro-alloy, the high basicity of the slag can result in a decrease in the content of P.
  • Slag composition under normal conditions has no big influence on the Cr2O3 content of the slag. A high content of Si in the ferro-alloy can reduce Cr2O3 dissolved in the slag by a silico-thermic reaction and in turn the Si content of Fe-Cr decreases, but this reaction is only significant at high Si contents. At this condition the colour of slag will become lighter.

Production of LC Fe-Cr

LC Fe-Cr is normally produced from HC Fe-Cr or charge chrome. The production of LC Fe-Cr is normally done by adding chromite ore or by blowing of oxygen (O2) in the HC Fe-Cr. However, these process are unattractive because of the high temperature (around 2100 deg C) needed for decreasing the C content to within the specification of LC Fe-Cr as well as due to the losses of Cr which takes place during the process. Hence, these methods for the decrease of C of the HC Fe-Cr are used mostly for the production of MC Fe-Cr and LC Fe-Cr is mostly produced by the metallo-thermic production processes.

In the metallo-thermic production processes, the reduction reaction is carried out with a specific metal, which has a negative Gibbs energy. Hence, the technically suitable metallic reductants which are produced in bulk are aluminum (Al), magnesium (Mg), manganese (Mn) and silicon (Si). The stoichiometric consumptions these metallic reductants per ton of Cr are 0.519 tons for Al, 0.701 tons for Mg, 1.585 tons for Mn and 0.405 tons for Si, though in actual practice the consumption can differ slightly to achieve targeted recovery of Cr. However, Si and Al reductants are more economical to use. Between the two, silico-thermic reduction process is more popular than the alumino-thermic process for the production of LC Fe-Cr.