REDUCTION PROCESSING OF METAL-CONTAINING ORES IN THE PRESENCE OF MICROWAVE AND RF ENERGY

The present invention relates to the treatment of metals in ores or concentrates (for example metal-containing oxides) and, in particular, a process for the chemical reduction of iron, for example, in ores or concentrates containing iron- oxides .

Extractive metallurgy is concerned with converting ores such as metal oxides, sulphides, chlorides and carbonates into the metal component of the ore. This is usually achieved by a process involving chemical reduction. Of critical importance in all extractive metallurgical processes is the need to achieve reduction of the metal ore on a large scale economically.

The reduction of iron in various ore forms is important in many industrial processes, for example in the production of iron and non-ferrous metals, such as nickel, vanadium and titanium. Examples of ores and concentrates from which iron may be extracted by reduction means include ilmenite, titaniferous magnetite, vanadiferous magnetite, iron ore, limonitic laterites and saprolitic laterites . In some industrial processes, it is desirable to eliminate a metal from an ore or concentrate, for example, the elimination of iron impurities from ilmenite. In other industrial processes it is the metal extracted from the ore or concentrate which is the desired product, for example extracted iron from iron ore.

There are currently several process routes that can be applied to separating iron from titanium oxide in ilmenite. These routes include the following: a) Fusion of the ilmenite ore with alkali metals and hydroxide to produce acid soluble salts; b) Direct leaching of the ilmenite using mineral acids,- c) Reduction of the ilmenite followed by leaching of the reduced iron and impurities; d) Carbothermic reduction and smelting to produce pig iron and slag containing titanium dioxide; and e) Selective chlorination of the iron by segregation.

Traditionally, conventional heating processes and chemical methods have been used to reduce the iron. Conventional heating processes encompass heating by convection, conduction and/or radiation. Typically, temperatures of at least 1000°C are required to process iron-containing ores. For example in the direct reduction in iron ore: (i) the SL/RN process requires a temperature of 1100°C; (ii) the DRC process requires a temperature of 1100 ⁰ C; (iii) the HyL process requires a temperature of 1000°C; and (iv) the MIDREX process requires a temperature of 760 to 930°C.

More recently, it has been suggested to use microwave (MW) energy to heat the ore. In particular, US 4,906,290 and US 6,277,168 describe processes for the reduction of metal ores

which involve the application of MW energy. US 4,906,290 describes a leaching or smelting precursor method of drying and heating particulate ores or concentrates which have been previously intimately admixed with either an already active form of carbon or with some other carbon-containing material which can be readily dried and heated to charring temperatures by MW energy comprising irradiating the composite with MW energy. US 6,277,168 describes a method for the direct preparation of a metal from a metal- containing ore by applying MW energy to extract metal from masses made by forming a powder of the ore and an optional reducing agent .

However, the use of dielectric heating (eg MW heating) has not proved economically viable on a large scale in view of the complication and expense of heating substantial volumes of ores to high temperatures (> 1000 ⁰ C) . There is therefore an economic prejudice in commercial extractive metallurgy to rely on heating methods other than convection, conduction and/or radiation.

It is an object of the present invention to address at least some of the problems and disadvantages of the prior art and to provide an energy efficient process for the chemical reduction of metal-containing ores and concentrates.

Accordingly, the present invention provides a process for reducing a metal-containing material, the process comprising: providing a metal-containing material, heating said metal-containing material by convective and/or conductive and/or radiative means,

exposing said metal-containing material to microwave (MW) energy, exposing said metal-containing material to radio frequency (RF) energy, and . exposing said metal-containing material to a reducing agent .

In another aspect the present invention provides an extractive metallurgical process for the chemical reduction of an iron-containing ore or concentrate, the process comprising: providing an iron-containing ore or concentrate, heating said ore or concentrate by convective and/or conductive and/or radiative means, exposing said ore or concentrate to a reducing agent, whereby some or all of said ore or concentrate is reduced to iron, characterised in that said ore or concentrate is also exposed to microwave (MW) energy and radio frequency (RF) energy concurrently with said heating of said ore or concentrate, the MW and RF energy levels being selected so that there is little or no additional heating of said ore or concentrate, and in that the temperature of said ore or concentrate does not exceed 85O ⁰ C, preferably 650 ⁰ C, during the reduction process.

The present inventors have surprisingly found that the application of microwave (MW) and radio frequency (RF) energy, together with conventional heating means, achieves reduction and/or metallisation of a metal in a metal- containing material at a lower temperature and at a shorter residence time than is possible using the conventional

_— C _

processes. As a consequence, the present invention enables the reduction of metal ores/concentrates without the need to heat the ore/concentrate to a high temperature. The process is therefore energy efficient and environmentally friendly because energy consumption is reduced and the large capital expenditure on equipment, which is required for high temperature operation, is obviated.

In the present invention, the heating of the metal- containing material by convective and/or conductive and/or radiative means preferably excludes the use of MW or RF energy as a significant source of heating energy.

Advantageously, exposing the metal-containing material to MW and/or RF energy (dielectric energy) raises the temperature by less than 50 ⁰ C, more preferably by less than 20 ⁰ C, still more preferably less than 10 ⁰ C, most preferably less than

5°C.

The application of microwave (MW) and radio frequency (RF) energy, together with conventional heating means has been surprisingly found to achieve direct reduction and/or metallisation of a metal in a metal-containing material at a lower temperature and at a shorter residence time than is possible using the conventional processes. This surprising effect is analogous to the MW and RF acting as a "catalyst" for the reduction of the material in question.

The term "direct reduction" is used to describe the reduction of a metal in a metal-containing material wherein the metal-containing material remains in a solid-state throughout the reduction process.

As used herein the term "metallisation" is used to describe the reduction of a metal in a positive oxidation state (higher than 0) to a metal having a zero oxidation state.

Each aspect as defined herein may be combined with any other aspect or aspects unless clearly indicated to the contrary. In particular any feature indicated as being preferred or advantageous may be combined with any other feature or features indicated as being preferred or advantageous .

The inventors have also observed enhanced reaction kinetics and an improved extent of reduction and/or metallisation when only one of MW or RF energy is applied to the metal containing material, compared to conventional reduction and/or metallisation processes. However, surprisingly, the improvements over conventional methods are much larger when both MW and RF energy is applied.

The process for reducing metal-containing materials as herein described has surprisingly been found to improve the reducing process of metal-containing materials over conventional methods. While not wishing to be bound by theory, it is believed that exposing the metal-containing material to MW and/or RF energy changes the structure of the material to make it more amenable to being reduced by the reducing agent. Accordingly, it will be appreciated that the main process steps can be applied to any process in which a change in the structure of the material is desired. Accordingly, the present invention also provides a process for treating a metal-containing material, the process comprising: providing a metal-containing material,

heating said metal-containing material by convective and/or conductive and/or radiative means, exposing said metal-containing material to microwave (MW) energy, and exposing said metal-containing material to radio frequency (RF) energy.

This process may, for example, be used to oxidise, roast, calcine and/or dehyroxylate a metal-containing material.

Metal-containing materials for use in the present invention may include, for example, ores and concentrates.

The metal-containing material comprises at least one metal in a positive oxidation state, for example Fe ³⁺ in Fe ₂ O ₃ , which may be reduced to a lower oxidation state, for example Fe ²⁺ or Fe ⁰ , using the reduction process of the present invention.

Preferably, the metal-containing material comprises a transition metal, for example iron, nickel, cobalt, vanadium, copper, titanium, chromium, zinc. The transition metal may be in the form of a transition metal oxide and/or sulphide. More preferably, the metal-containing material comprises an iron-containing ore, most preferably an iron oxide-containing ore. Examples of metal-containing ores for use in the present invention include ilmenite, titaniferous magnetite, vanadiferous magnetite, iron ore, hematite, limonitic laterites and saprolitic laterites, including mixture of two or more thereof. A preferred ore is or comprises ilmenite or iron ore. The composition of ilmenite is briefly discussed below.

The term ilmenite as used herein includes the mineral FeTiO ₃ . However, it will be understood that the composition of ilmenite from mineral sand deposits rarely conforms to the stoichiometric composition, FeTiO ₃ (47.4% FeO and 52.6% TiO ₂ ) . Accordingly, it will be understood that the term ilmenite as used herein also encompasses non-stoichiometric compositions .

Furthermore, natural ilmenite is a mineral rich in FeTiO ₃ and contains small amounts of MnO, MgO, Fe ₂ O ₃ and/or Ti ₂ O ₃ in its structure, where some of the ferrous iron is replaced by Mn ²⁺ and Mg ²⁺ . The resulting general formula is (Fe ₇ Mn ₇ Mg)TiO ₃ .

It is well known that ilmenite in beach sand deposits undergoes a progressive alteration in composition and structure. The Fe ²⁺ in ilmenite is gradually converted to Fe ³⁺ as a result of prolonged atmospheric exposure. The Fe ³⁺ can be leached out of the mineral grains by natural waters.

This process is called weathering. Weathering enriches ilmenite in titanium. Weathering of ilmenite leads to the formation of pseudorutile according to Reaction 1.

Reaction 1 : 3FeTiO ₃ + 2H ⁺ + 0 . 5O ₂ = Fe ₂ Ti ₃ O ₉

+ Fe ²⁺ + H ₂ O

Alteration continues with the leaching of more iron from the pseudorutile, leading to the complete breakdown of the pseudorutile and the formation of anatase, as shown in

Reaction 2.

Reaction 2 : Fe ₂ Ti ₃ O ₉ + 4H ⁺ = 3TiO ₂ + 2Fe ²⁺ +

2H ₂ O + 0 . 5O ₂

In extreme cases, weathering results in the formation of an amorphous material with a composition close to TiO ₂ . This amorphous material is generally referred to as leucoxene.

Accordingly, it will be understood that the term ilmenite as used herein also encompasses compositions comprising ilmenite together with other ores, for example ilmenite- hematite solid solutions, rutile and variable amounts of other oxide and silicate mineral impurities .

The metal-containing material will typically be provided in the form of small pieces of crushed ore having diameters of less than approximately 100 mm, and preferably less than 50 mm. The metal-containing material may be provided in the form of pellets (pelletised fine particles) with diameters from 5 to 30mm, and preferably from 10 to 20 mm. In addition, or alternatively, the metal-containing material may be provided in the form of individual particles having diameters less than 5 mm, preferably less than 1 mm, more preferably less than 0.5 mm.

The metal-containing material may be pre-oxidised prior to the reduction process of the present invention. It will be understood that a variety of methods may be used to pre- oxidise the metal-containing material. For example, the metal-containing material may be oxidised by exposure to oxygen (either as air or an oxygen enriched stream) . This oxidation may be carried out at, for example, a temperature

greater than 800 ⁰ C in a suitable reactor (a rotary kiln, multiple hearth roaster or a circulating fluidised bed) .

Without wishing to be bound to any particular theory, it is thought that pre-oxidation may enhance the reduction of metal oxides in, for example, ilmenite by causing a change in the structure and reactivity of the oxide surface.

It will be understood that the percentage of metal in the metal-containing material will vary depending on the particular material. Typically, the metal oxide and/or sulphide will comprise up to 95 wt.% of the metal-containing material, for example from 20 to 60 wt.%.

Typically, electrical energy (e.g. electrical resistive heating elements) or fuels in the form of a solid, liquid or gas are be used to heat (convective and/or conductive and/or radiative) the metal-containing material. These methods are well known in the art.

Examples of suitable solid fuels include one or more of coal and cellulosic materials such as timber or saw dust. Coal includes lignite, bituminous coal, coke, anthracite and carbonised coal.

Examples of suitable liquid fuels include one or more of fuel oil, light fuel oil, heavy fuel oil, diesel and kerosene .

Examples of suitable gaseous fuels include natural gas, which may be treated to form synthetic gas, producer gas or reformer gas .

The metal-containing material is heated to a temperature lower than the melting point of the material, so that the metal-containing material is in the solid-state throughout the reduction process. Preferably, the metal-containing material is heated by said convective and/or conductive and/or radiative means to a temperature that does not exceed approximately 1,200 ⁰ C, more preferably a temperature that does not exceed approximately 85O ⁰ C, most preferably a temperature that does not exceed approximately 65O ⁰ C. For example, the metal-containing material may preferably be heated by said convective and/or conductive and/or radiative means to a temperature in the range of from 400 to 850°C.

The inventors have observed that the most significant improvement in the kinetics and extent of metallisation and/or reduction are observed at low temperatures, preferably less than 85O ⁰ C, more preferably less than 65O ⁰ C.

At these low temperatures using conventional reduction methods, iron oxide which has been weathered to form predominantly Fe(III) oxide will be metallized by both conventional and dielectric methods. However, iron- containing ilmenite, which is essentially FeO. TiO ₂ , in which the iron has not been weathered (oxidised to Fe(III)) is not observed to be reduced by conventional methods . It is only observed to be reduced under the dielectric conditions of the present invention.

In the present invention, the heating of the metal- containing material by convective and/or conductive and/or radiative means preferably excludes the use of MW or RF

energy (dielectric heating) to significantly heat the metal. Advantageously, exposing the metal-containing material to MW and/or RF energy (dielectric energy) raises the temperature by less than 50 ⁰ C, more preferably by less than 20 ⁰ C, still more preferably less than 10 ⁰ C, most preferably less than 5°C.

The MW and RF energy used in the present invention are chosen such that upon exposing the metal-containing material to MW and RF the material and/or system is not significantly heated.

Depending on the material to be processed, the inventors have found that there is a window of MW and RF energies within which there is observed a significant increase in the kinetics and/or extent of metallisation without substantial heating of the material and/or the system. Outside this window, the application of MW and RF energy heats the material to a significant degree. It will be appreciated, that the power densities for the MW and RF are lower than would be required to heat the material.

It is well known that RF and MW radiation occupy adjacent sections of the electromagnetic spectrum, with MWs having higher frequencies than radio waves. Several frequencies have been set aside by International Treaties, such as ISM (Industrial, Scientific and Medical) . These are internationally agreed and recognised frequency bands, known as ISM bands, or Industrial, Scientific and Medical Bands. ISM defines RF bands as:

(i) 13.56 MHz ± 0.05%

(ii) 27.12 MHz ± 0.6% (iii) 40.68 MHz ± 0.05%

and MW bands as:

(i) 896 MHz

(ii) 915 MHz

(iii) 2450 MHz ± 50 MHz

It has been found that exposing a metal-containing material to RF and/or MW ISM bands described above results in an increase in the kinetics and/or percentage of the reduction and/or metallisation reaction without significantly heating said material and/or the system. Preferably, the RF band is 13.56 MHz ± 0.05%.

In the case of large, industrial scale machines the MW band is preferably 896 MHz or 915 MHz. However, in smaller laboratory scale machines, it may be preferable for the MW band to be 2450 MHz ± 50 MHz.

MW and RF can be differentiated with reference to the methods of producing the high frequency fields. Microwave systems utilise magnetrons and klystrons to generate the electromagnetic field, and resonant and non-resonant cavities to apply the field. Radio frequency systems use high power valves to generate the electromagnetic field, and capacitors and electrodes to apply the field.

The MW and RF energy may be applied independently of one another, or in combination, over a range of electric field strengths and field strength ratios.

In the prior art relating to processing minerals in a microwaved energised fluidised bed, it is taught that high electric field strengths, in the region of 10 ⁶ Vm ^"1 are desirable. Furthermore, it is taught that Q values should be as high as possible. The "Q" factor is the ratio of electromagnetic energy stored to electromagnetic energy- dissipated. Typically the prior art teaches towards minimum Q values of 20,000.

In contrast to the prior art, the inventors have found that low Q values, preferably in the region of 1 to 1,600 are advantageous. The specific range of Q values for a given experiment will depend on the materials present in the cavity. Furthermore, it is preferable for the RF and MW to have low electric field strengths .

Preferably the peak electric field strength of the MW energy applied to cavity does not exceed 5,000 Vm ^"1 , more preferably it does not exceed 1,500 Vm ^"1 , most preferably it does not exceed 850 Vm ^"1 .

Preferably the peak electric field strength of the MW energy applied to the cavity is in the range of from 300 to 5,000 Vm ^"1 , more preferably from 500 to 3,000 Vm ^"1 .

Preferably the peak electric field strength of the RF energy applied to the cavity does not exceed 100,000 Vm ^"1 , more preferably it does not exceed 25,000 Vm ^"1 , most preferably it does not exceed 12,000 Vm ^"1 .

Preferably the peak electric field strength of the RF energy- applied to the cavity is in the range of from 3,000 to 100,000 Vm ^"1 , more preferably in the range of from 4,500 to 60, 000 Vm ^"1 .

Preferably, the electric field strength ratio of MW energy: RF energy is from 1:4 to 1:20.

Without wishing to be bound to any particular theory, it is believed that exposing the metal-containing material to MW and/or RF energy changes the structure of the material to make it more amenable to being reduced by the reducing agent. At relatively low power levels, it has surprisingly been observed that while there is little or no additional heating by the MW and RF sources, the metallisation of the ore or concentrate is significantly improved.

It is important to understand that there is a distinction between induction heating and dielectric heating. In induction heating, the workpiece or load to be heated forms a resistive secondary of a transformer which has the induction heating coil as its primary. In contrast, in dielectric heating, the load to be heated forms a lossy capacitor, which is coupled to an applicator capacitor. In induction heating it is the magnetic field from the coil which induces the electric currents in the resistive load whereas in dielectric heating, it is the electric field produced between the plates of the applicator which leads to the heating effect. As a consequence, induction heating is characterised by large electric currents (and magnetic fields) whereas dielectric heating is characterised by high voltages (and electric fields) . Preferably, inductive

heating, which is usually generated with a high frequency- source typically with oscillators of high power, operating at 10 ⁶ -10 ⁷ Hz ₇ is not used in the present invention to convectively and/or conductively and/or radiatively heat the metal-containing material in the present invention.

Inductive heating is excluded in the present invention from the term "exposing said metal-containing material to RF energy" . In the present invention, preferably dielectric heating is not used to significantly heat the metal- containing material.

The metal-containing material may be heated by said convective and/or conductive and/or radiative means before exposing said material to MW energy and/or RF energy. Alternatively, the metal-containing material may be exposed to MW energy and/or RF energy before the metal-containing material is heated by said convective and/or conductive and/or radiative means. Alternatively, or in combination, the metal-containing material may be exposed to MW and/or RF energy while heating said material by said convective and/or conductive and/or radiative means .

Preferably, the metal-containing material is exposed to MW energy, RF energy, and is heated by said convective and/or conductive and/or radiative means simultaneously.

The reducing agent may be a solid, liquid or gas. Preferably, the reducing agent is a solid or a gas, most preferably a gas.

Examples of suitable solid reducing agents include one or more of carbon, bitumen, coal, coke and anthracite.

Examples of suitable gaseous reducing agents include one or more of carbon monoxide, carbon dioxide, hydrogen, and hydrocarbons. Preferably, the reducing gas comprises carbon monoxide and/or hydrogen, more preferably the gas is or comprises hydrogen.

The reducing agent (solid and/or liquid and/or gas) may be comprised in the metal-containing material.

If the reducing agent is a gas, the gas may be continuously introduced into a reaction cavity, in which the metal- containing material is situated. The reducing gas reduces the oxidised metal in the metal-containing metal. The out- gas may be continuously removed from the reaction cavity.

Preferably, sufficient reducing gas is passed over the metal-containing material to maintain a stoichiometric excess of reducing gas to oxidised metal ions in the metal- containing material.

A preferred embodiment of the present invention concerns a process for reducing iron oxide in ores and/or concentrates. The reduction reactions are preferably conducted at a temperature of from 250 ⁰ C to 1200 ⁰ C, more preferably from

300 ⁰ C to 950 ⁰ C, and most preferably from 400 ⁰ C to 850 ⁰ C. The reduction reaction may be carried out in a reducing gas atmosphere, preferably in a hydrogen atmosphere. The metal- containing material is exposed to low intensity MW and RF frequency energy, preferably a MW energy of approximately 2450 MHz for small scale reactions and 896 MHz or 915 MHz

for large scale reactions, and a RF energy of approximately 13.56 MHz.

The present invention will now be described further, by way of example only, with reference to the following drawings, in which:

Figure 1 is a block flow diagram of one embodiment of the invention.

Figure 2 shows a comparison of the measured furnace and sample temperatures under an applied MW and RF field.

Figure 3 shows the kinetics for metallisation of ilmenite under hydrogen as a function of temperature (MW power is 900W and RF power is 300W) .

Figure 4 shows the effect of temperature on the extent of enhancement of the metallisation of ilmenite.

Figure 5 shows the effect of dielectric power on the reduction of ilmenite.

Figure 6 shows the metallisation kinetics of ilmenite at 850 ⁰ C under 70% CO and 15% H ₂ .

Figure 7 shows the metallisation kinetics of ilmenite at 900 ⁰ C under 70% CO and 15% H ₂ .

Figure 8 shows the metallisation kinetics of ilmenite at 950 ⁰ C under 70% CO and 15% H ₂ .

Figure 9 shows the metallisation kinetics of ilmenite at 650 ⁰ C under 100% hydrogen.

Figure 10 shows the metallisation kinetics of ilmenite at 750 ⁰ C under 100% hydrogen.

Figure 11 shows the metallisation kinetics of ilmenite at 850 ⁰ C under 100% hydrogen.

Figure 12 shows the effect of the preoxidation temperature on mass loss (metallisation) .

Figure 13 shows the effect of reduction temperature.

Figure 14 shows the effect of power on the reduction of ilmenite pre-oxidised at 95O ⁰ C and then reduced under hydrogen at 600 ⁰ C and 750 ⁰ C over a 60 minute period.

Figure 15 shows the reduction kinetics of pre-oxidised ilmenite calcine at 600 ⁰ C under hydrogen after pre-oxidation at 95O ⁰ C.

Figure 16 shows the effect of temperature on the reduction kinetics under hydrogen after pre-oxidation at 95O ⁰ C.

Figure 17 shows the effect of the reduction temperature on metallisation after pre-oxidation at 950 ⁰ C for 60 minutes.

Figure 18 shows the metallisation kinetics of pre-oxidised ilmenite calcine at 850 ⁰ C after pre-oxidation at 95O ⁰ C.

Figure 19 shows the particle size distribution for the titaniferous magnetite sample. The +106 μm -1700 μm fraction (shaded) was used in fluidised bed reactor experiments .

Figure 20 shows metallisation of magnetite under hydrogen at 600 ⁰ C

Figure 21 shows indicative enhancement ratio as a function of temperature.

Figure 22 shows the effect of MW power on the enhancement of magnetite metallisation.

Figure 23 shows the effect of RF power on the enhancement of magnetite metallisation. The mass loss is a direct measurement of the reduction and metallisation.

Figure 24 shows reduction of hematite iron ore fines under hydrogen. Forward P _MW = 900 W and P _RF = 300 W.

Figure 25 shows the effect of temperature on the extent and kinetics of the reduction of Hematite iron ore fines under carbon monoxide.

Figure 26 shows the comparison of the mass losses for the reduction of Hematite iron ore fines under hydrogen and under carbon monoxide .

Figure 27 shows the particle size distribution of the limonitic nickeliferous laterite sample showing the size fractions used in the test work.

Figure 28 shows the comparison of the reduction of ferric iron under conventional and dielectric assisted conditions.

Figure 29 is a comparison of the formation and reduction of ferrous iron under conventional and dielectric assisted conditions .

Figure 30 is a comparison of the formation of metallic iron under conventional and dielectric assisted conditions.

The present invention may be further understood with reference to the block diagram shown in Figure 1. This diagram shows an embodiment of the present invention comprising: a reaction cavity 1; a heating element 2; a MW source 3; a RF source 4; a mineral feed 5; a product discharge 6; a reductant off-gas 7; and a reductant feed gas 8.

Each component of this diagram will be discussed in more detail below.

The reaction cavity 1 is for containing the metal-containing material. It will also provide a confinement for both MW and RF energy.

The heating element 2 is positioned to enable the metal- containing material to be heated by convective and/or conductive and/or radiative, means . The heating element 2 may utilise resistive heating elements, gas, liquid or solid combustion.

The MW source 3 and RF source 4 are coupled to said cavity 1 to enable the metal-containing material, when positioned in said cavity, to be exposed to MW and RF energy.

The MW and RF energy are preferably controllable independently of one another, and independently of the conductive and/or convective and/or radiative heating means.

The mineral feed 5 provides an inlet into the cavity 1 through which the metal-containing material may be introduced to the cavity.

The product discharge 6 provides an outlet through which the reduced material may be discharged. Preferably this reduced material will comprise metal in the metallic state.

The inlet and outlet are designed to prevent, or at least minimise, MW radiation from leaving the cavity.

The reductant feed gas 8 may be continuously fed into the reaction cavity 1, where it reacts with the metal-containing material. The reductant off-gas may also be continuously removed from said cavity 1.

Although conventional furnaces may be adapted for use in the present invention, hybrid furnaces, such as those described in EP 0 914 752, are preferred.

EP 0 914 752 describes a RF and MW assisted furnace. Previous uses of this hybrid furnace have been limited to heating of ceramics, ceramic-metal composites, metal powder components, and engineering ceramics. The MW and RF sources

are independently controlled and are applied to the furnace chamber. The conventional furnace can be operated independently of the MW and RF sources .

In the prior art, MW energy is known to be used to preferentially heat selected mineral phases of ores. In contrast to this, the heating means used in the present invention substantially avoids preferentially heating specific phases of the metal-containing material. It can be seen in Figure 2 that the MW and RF energy applied to ilmenite does not appreciably heat the ilmenite above the furnace temperature .

The process of the present invention for reducing a metal- containing material has a number of advantages over the processes known in the prior art. One advantage is the enhanced reduction kinetics observed using the process as described herein. In particular, it has been observed that the reduction kinetics of reactions carried out at high temperatures and in the absence of dielectric fields (MW and RF) are slower than the reduction kinetics of reactions carried out in the presence of MW energy and RF energy at the lower temperatures required for the present invention. The corollary of this is significant energy savings in extractive metallurgical processes and reduced emissions.

In particular, reduced gaseous emissions are observed which is linked to the reduced specific energy consumption of the present invention. Moreover, the process achieves a greater degree of metallisation at a given temperature and also allows for metallisation at lower temperatures than currently possible.

This is exemplified in Figure 3, which shows a significant improvement in the kinetics and extent of metallisation of ilmenite at low temperatures compared with using conventional reduction processes (i.e. in the absence of MW or RF energy) . The improvement may be observed at 850 ⁰ C. However, the improvement is more significant at 750 ⁰ C, and even more so at 650 ⁰ C.

The present invention will be further illustrated with reference to the following non-limiting Examples.

A general procedure for the experiments as described herein is given below.

The MW generator (2.OkW at 2.450GHz) used in the experiments was a SAIREM Model GMP 20T.

The RF generator (1 kW at 13.56MHz) used in the experiments was a RF Power Products Model RlOS (75 207 09 040).

The RF Matching Box (13.56MHz) was a RF Power Products Model SPR876 (76 226 59 010) .

The radiant heating system was a Kanthal Super 1800 rated at 11.8 V, 0.58kW. Six elements were required for a load of 3.5kW at 48A with the elements wired in series.

The radiant element dimensions were as follows: LU 180mm terminal length; LE 200mm heating zone length;

A 30mm distance between shank centres; and Approximately 69g in weight.

The control system used a Eurotherm Thyristor Model 461 240V, 25A. The Eurotherm dual loop temperatures controller model 90OEPC, and thermocouple type R was used.

The kiln had the following dimensions: 460 x 480 x 540 mm.

The cavity used in the experiments was designed to be an efficient applicator for the microwave field, minimising the losses of electromagnetic energy to the cavity shell and maximising the dissipation of electromagnetic energy into the cavity load. Additionally, the cavity contains the radio frequency applicator in the form of electrodes or otherwise. The cavity is designed in such a manner as to prevent the microwave field from accessing the radio frequency tuning components, such as inductors and matching capacitors.

Within the cavity there is insulating fibre board (Rath KVS 161/302), Superkanthal resistance heating elements, a quartz muffle, and when the sample is inserted, a quartz crucible and quartz gas supply line . The mineral sample of interest rests in the crucible.

The furnace temperature is set to the specified temperature. Time is allowed for the furnace conditions to stabilise. The sample is weighed and poured into the sample holder . The sample holder containing the sample is mounted onto the gas supply tube in the sample chamber and the sample chamber is closed. The sample chamber and muffle is purged with nitrogen before the appropriate gas mixture is introduced.

Where applicable MW and/or RF are introduced into the furnace cavity at this stage. In the examples when MW

and/or RF were applied, the MW and/or RF generators were switched on before insertion of the sample into the cavity. At the start of the experiment the sample holder is raised into the furnace. Relevant data is recorded during the experiment. After the specified reaction time the sample is lowered out of the furnace into the sample chamber. The sample chamber and muffle is purged with nitrogen and the sample is allowed to cool. The sample is weighed after the experiment and the weight loss is calculated. Selected samples are prepared for sample analysis.

With reference to field strengths, specific forward microwave and RF power levels were applied. The forward power levels from the generator and the reflected power levels were measured. In all experiments, unless stated otherwise, all power levels referred to are forward power levels. Additionally, MW power was adjusted to 900 W and RF power was adjusted to 300 W.

The actual electric field strength at the sample was not measured. Instead, the peak electric field strength was calculated according to the following equation:

E* V _n ωε _π

P = 'max ' C ¹ ^ ⁰ O

4β ₀

Where : E _max is the maximum electric field strength

P is the dissipated field power

Vc is the volume of the field cavity

Q ₀ is the "quality factor" of the field cavity. The Q factor for RF and MW assisted cavities when the cavity is empty is low, between 2 and 20, averaging 9, as most

energy is dissipated within the dielectric load and not stored in the cavity itself. ω 2πf, where f is the frequency f = 915 MHz or 2450 MHzfor microwave f = 13.56 MHz for radio frequency

In the following examples quantitative mass loss data and Mόssbauer analyses were used to determine the extent of metallisation. ⁵⁷ Fe Mόssbauer measurements were performed in conventional transmission geometry. A K3 Austin

Associates linear motor driven by a triangular reference wave-form was used to scan the resonance profile. A Kr-CO ₂ (2atm) proportional counter was used to detect the transmitted 14.4 keV resonance radiation from a 50 mCi ⁵⁷ Co(Rh) radioactive source.

Each sample was mounted in a specially designed powder-clamp holder cm). Sample quantities of 40-50 mg were ground under acetone in an agate pestle and mortar, thoroughly mixed with an inorganic buffer material and the mixture of sample and buffer distributed in the sample holder to form a disk of uniform thickness for transmission Mδssbauer measurements.

Typical count-rates in the discriminator window set to select the 14.4 keV resonance radiation were in the range 10000 counts/sec. Data acquisition of each spectrum and its mirror-image continued for a period of 6 - 12 hours to obtain several hundred thousand counts in each of 1024 channels of a PCA-based multi-channel analyser. X-scale velocity (energy) calibration was by means of a 25-μtn thick α-Fe foil measured before and after the series of

measurements which gave a line width of r=0.27mm/s. Prior to analysis each spectrum was folded with it's mirror image and adjacent channels subsequently added. This serves to remove geometrical base-line distortions and reduces the (VN) statistical scatter in the final data-set used for analysis. The fitting program, NORMOS, has been used for theoretical fits of the data with Lorentzian line shapes to deconvolute various sub-components (phases) in the spectrum.

In Examples 1 to 4 a sample of ilmenite concentrate was reduced using the process of the present invention. The results were compared with experiments conducted under the same conditions, but in the absence of MW or RF energy. The ilmenite sample used in experiments 1 to 4 is described below.

The particle size distribution of the ilmenite sample as determined by dry screening is shown in Table 1.

Table 1: Particle size distribution of the ilmenite sample.

+300 0.61

-300+212 1.27

-212+150 5.48

-150+106 42.22

-106+75 46.88

-75 3.54

The moisture content of the ilmenite sample, as determined by a Mettler Toledo HR73 moisture analyser at 120 ⁰ C, is 0.32%. Weight loss at 950 ⁰ C in a N ₂ atmosphere is 0.60%. The testing time- at 950 ⁰ C was 60 minutes.

The mineral analysis in Table 2 conducted on the ilmenite sample revealed that the main phases present are ilmenite (80.36%), altered ilmenite (4.73%), Ti-Hematite (11.48%) and rutile (0.41%). Mόssbauer analysis supports the mineral analysis indicating the presence of 73% Fe ²⁺ and 11% Fe ³⁺ as ilmenite and 16% Fe ³⁺ as hematite.

Table 2 Chemical and mineral analysis of ilmenite sample. fjliltlt

Chemical analysis

Ti 28.50

Fe 37.53

Si 0.99

Mg 0.28

Al 0.27

Mineral abundance

Ilmenite 80.36

Altered ilmenite 4.73

Ti - Hematite 11.48

Ruble 0.41

Iron metal 0.00

Fe phase abundance

Fe ⁰ as iron metal 0.00

Fe ²⁺ as ilmenite 73.00

Fe ³⁺ as ilmenite 11.00

Fe ³⁺ as hematite 16.00

Example 1

Reduction of ilmenite under CO/H ₂

Tests were conducted at temperatures between 85O ⁰ C and 1050 ⁰ C under a reducing atmosphere composed of 70% CO, 15% H ₂ and 2% CO ₂ with the balance comprised of nitrogen.

Samples were placed into a fluidised bed reactor, which was then inserted into the hybrid furnace. Reduction reactions were conducted under conventional conditions (i.e. in the absence of any MW and RF fields) and under dielectric

conditions (i.e. in the presence of MW and RF fields). The effects of temperature and dielectric power on the extent of metallisation and the metallisation kinetics were investigated and compared, as shown in the following figures (4 to 8) .

Example 2

Reduction of ilmenite under hydrogen Tests were conducted at temperatures between 65O ⁰ C and 850 ⁰ C under a reducing atmosphere composed of 100% H ₂ . Samples were placed into a fluidised bed reactor, which was then inserted into the hybrid furnace . Reduction reactions were conducted under conventional conditions (i.e. in the absence of any MW and RF fields) and under dielectric conditions

(i.e. in the .presence of MW and RF fields) . The effects of temperature and dielectric power on the extent of metallisation and the metallisation kinetics were investigated and compared, as shown in the following figures (9 to 11) .

Example 3

Reduction of pre-oxidised ilmenite under hydrogen Ilmenite was pre-oxidised at temperatures between 600 ⁰ C and 95O ⁰ C under air prior to being reduced at 600 ⁰ C under 100% hydrogen under conventional conditions (i.e. in the absence of any MW and RF fields) and under dielectric conditions (i.e. in the presence of MW and RF fields) .

The effect of the oxidation temperature and the application of dielectric assistance during the subsequent reduction on the extent of metallisation is shown in Figure 12.

Samples pre-oxidised at 950 ⁰ C under air over 60 minutes were placed into a fluidised bed reactor, which was then inserted into the hybrid furnace. The samples were then reduced under 100% hydrogen at temperatures between 550 ⁰ C and 750 ⁰ C. Reduction reactions were conducted under conventional conditions (i.e. in the absence of any MW and RF fields) and under dielectric conditions (i.e. in the presence of MW and RF fields) . The effects of temperature and dielectric power on the extent of metallisation and the metallisation kinetics were investigated and compared, as shown in the following figures 13 to 16.

Example 4

Reduction of the pre-oxidised ilmenite calcine under CO Samples pre-oxidised at 950 ⁰ C under air over 60 minutes were placed into a fluidised bed reactor, which was then inserted into the hybrid furnace. The samples were then reduced under 100% CO at temperatures between 750 ⁰ C and 950 ⁰ C. Reduction reactions were conducted under conventional conditions (i.e. in the absence of any MW and RF fields) and under dielectric conditions (i.e. in the presence of MW and RF fields) . The effects of temperature and dielectric power on the extent of metallisation and the metallisation kinetics were investigated and compared, as shown in the following figures 17 to 18.

In Example 5 a titaniferous magnetite ore was reduced under hydrogen and carbon monoxide using the process of the present invention. The results were compared with experiments conducted under the same conditions, but in the absence of MW or RF energy. The magnetite sample used in experiments 5 is described below.

Titanomagnetite and titanomaghemite are the main iron bearing phases in the ore, making up a combined 85% of the ore mass and containing more than 96% of the iron. The ore contains a substantial maghemite (γ-Fe ₂ O ₃ ) component. The maghemite is closely associated and intensely intergrown with the original magnetite. Maghemite is metastable and inverts to hematite (α-Fe ₂ O ₃ ) on heating (200° to 700 ⁰ C).

Titanium in the. ore is predominantly hosted by titanomagnetite and titanomaghemite (93 %) , with the remaining 7 % hosted by ilmenite. Combined, the titanomagnetite-titanomaghemite contains on average ±8.2% titanium by mass.

Vanadium in the ore is also predominantly hosted by titanomagnetite and titanomaghemite (>85 %) . No other well defined and distinct V-bearing or V-rich phases were identified.

The particle size distribution and the chemical composition of the ore are shown in Table 3 and Figure 19.

Table 3: Chemical composition of the +106 μm -1700 μm fraction of the titaniferous magnetite sample.

Example 5

Reduction under hydrogen

Tests were conducted at temperatures between 600 ⁰ C and 900 ⁰ C under a reducing atmosphere composed of 100% H ₂ . Samples were placed into a fluidised bed reactor, which was then inserted into the hybrid furnace. Reduction reactions were conducted under conventional conditions (i.e. in the absence of any MW and RF fields) and under dielectric conditions (i.e. in the presence of MW and RF fields) . The effects of temperature and dielectric power on the extent of metallisation and the metallisation kinetics were investigated and compared, as shown in the following figures 20 to 23.

In Examples 6 and 7 a hematite sample was reduced using the process of the present invention. The results were compared with experiments conducted under the same conditions, but in the absence of MW or RF energy. The hematite sample used in experiments 6 and 7 is described below.

The sample consists predominantly of hematite (+91%) , with the remainder comprising mainly silicates (7.1%) and

compound phosphates (1.0%). The silicates consist of mainly quartz, mica (muscovite) and kaolinite/pyrophyllite. The compound phosphate mineral appears to be a svanbergite- woodhouseite solid solution phase.

Hematite is thus the main iron bearing phase in the ore, with between 99 and 100 % of the iron hosted by hematite and very minor to trace amounts of associated goethite. The iron associated with the quartz and mica is mainly due to hematite, which is intensely intergrown with these phases. The chemical composition is shown in Table 4.

Table 4: Chemical composition of the iron ore (hematite) sample.

Example 6

Reduction of hematite (fines) under hydrogen

Tests were conducted at temperatures between 400 ⁰ C and 600 ⁰ C under a reducing atmosphere composed of 100% H ₂ . Samples were placed into a fluidised bed reactor, which was then inserted into the hybrid furnace. Reduction reactions were conducted under conventional conditions (i.e. in the absence of any MW and RF fields) and under dielectric conditions (i.e. in the presence of MW and RF fields) . The effects of temperature and dielectric power on the extent of

metallisation and the metallisation kinetics were investigated and compared, as shown in the following figure 24.

Example 7

Reduction of fines under carbon monoxide

Tests were conducted at temperatures between 600 ⁰ C and 800 ⁰ C under a reducing atmosphere composed of 100% CO. Samples were placed into a fluidised bed reactor, which was then inserted into the hybrid furnace . Reduction reactions were conducted under conventional conditions (i.e. in the absence of any MW and RF fields) and under dielectric conditions (i.e. in the presence of MW and RF fields) . The effects of temperature and dielectric power on the extent of metallisation and the metallisation kinetics were investigated and compared, as shown in the following figures 25 and 26.

In Example 8 a Limonitic Laterite sample was reduced using the process of the present invention. The results were compared with experiments conducted under the same conditions, but in the absence of MW or RF energy. The Limonitic Laterite sample used in experiment 8 is described below.

The sample was dried and the -1180+425 μm fraction removed for the required test work. This fraction was found to be most suitable in order to limit the loss of fines from the fluidised bed reactor. Unfortunately, the other size fractions were excluded from test work due to this equipment limitation. The particle size distribution of the limonitic

nickeliferous laterite sample is shown in Figure 27, illustrating the fractions used in the test work. The moisture content of the sample, as determined by a Mettler Toledo HR73 moisture analyser at 120 ⁰ C, is 3.23%. Weight loss (after 15 minutes) at 700 ⁰ C under a nitrogen atmosphere is 14.58%. The weight loss is not only attributed to loss of free moisture but also the loss of chemically bound water and dehydroxylation reactions.

Table 5: Chemical composition of the limonitic nickeliferous laterite.

Fe 46.20

Ni 0.37

Co 0.05

4.73

Mg 0.43

Cr 1.93

Al 5.37

Mn 0.40

Ti 0.28

0.88

0.04

The chemical composition of the head sample complies with the typical chemical composition of a limonitic nickel laterite ore (Table 5) . The Si content of the head sample is higher than the typical Si concentration of a limonitic nickel laterite ore.

The limonitic nickeliferous laterite sample consists mainly of goethite, hematite, magnetite, quartz and chromite. The sample is characterised by abundant limonite with lesser quantities of goethite, hematite, magnetite, quartz and Cr- spinel. Limonite is in general an ill-defined phase mainly

due to its amorphous nature. It is generally described as an intensely hydrated and amorphous iron oxide. Limonite forms a matrix within which grains and fragments of partly to completely hematitised magnetite, Cr-spinel and quartz are situated. In this sample, limonite is the dominant Fe and Ni bearing phase. It also contains relatively large quantities of Al and Si and often traces or minor amounts Ti and Cr. Goethite and goethite-hematite are present either within the limonite matrix or as discrete secondary oolitic nodules. The goethite nodules are usually surrounded by a film of limonite . The goethite hosts some Ni .

The magnetite fragments and grains were hematitised to variable degrees. Very fine Ni-metal inclusions occur within the preserved magnetite remnants and are possibly an inherent geological component of the rock.

Chromite (Cr-spinel) fragments and grains are quite common and mostly occur as inclusions within the limonite matrix. The chromite is often transected by partly hematitised magnetite veins. The chromite is occasionally surrounded by a Cr-rich goethite film. The Cr-spinel typically contains substantial amounts of Al, Fe and Mg.

Example 8

Reduction of the limonitic nickeliferous laterite sample

Tests were conducted 700 ⁰ C under a reducing atmosphere composed of 18% CO, 50% H ₂ , 2% CO ₂ , balance nitrogen. Samples were placed into a fluidised bed reactor, which was then inserted into the hybrid furnace . Reduction reactions were conducted under conventional conditions (i.e. in the

absence of any MW and RF fields) and under dielectric conditions (i.e. in the presence of MW and RF fields) . The effects of dielectric power on the extent of metallisation of the iron and the reduction kinetics for feric and ferrous iron were investigated and compared, as shown in the following Figures 28 to 30.