METHOD FOR PROCESSING MATERIALS CONTAINING IRON AND ZINC

Abstract

A method for processing a source material containing zinc ferrite that includes the following step: B. Partially reduce source material using a reducing gas containing hydrogen to form a reduced material; where step B is carried out at below 1000 C. using a reducing gas containing at least 0.25% (by vol.) and up to 70% (by vol.) hydrogen in a carrier gas.

Claims

1. A method for processing a source material containing zinc ferrite that includes the following step: B. partially reduce source material using a reducing gas containing hydrogen to form a reduced material; wherein the reducing gas includes water.

2. The method as claimed in claim 1 wherein, step B is carried out at or below 1000 C.

3. The method as claimed in claim 1 wherein, the source material is a particulate material that is not pelletised before processing.

4-5. (canceled)

6. The method as claimed in claims 1 wherein, the source material is Electric Arc Furnace Dust (EAFD).

7-8. (canceled)

9. The method as claimed in claim 1 wherein, step B is carried out at between 500 C. and 750 C.

10-11. (canceled)

12. The method as claimed in claim 1 wherein, the method includes a step C, where step C cools the reduced material to a temperature no higher than 350 C. in an inert or reducing gas environment.

13. The method as claimed in claim 1 wherein, the hydrogen content of the reducing gas is between 0.25% and 10% by volume.

14. The method as claimed in claim 1 wherein, there is a step A preceding step B, where in step A the source material is analysed to determine the total zinc and/or zinc ferrite concentrations, wherein, Step A determines proposed processing conditions to be used in Step B, including reduction time (Rt), reduction temperature (RT) and reducing gas flowrate (GF).

15-20. (canceled)

21. The method as claimed in claim 1 wherein, the water is added to the reducing gas at a predetermined time after step B commences.

22. The method as claimed in claim 21 wherein, the predetermined time a. is reached before the temperature has reached a value selected from the list consisting of 500 C., 550 C., 600 C., and 650 C. or b. is reached when the temperature has a value within a range selected from the list consisting of 500 C. to 550 C., 550 C. to 600 C., 600 C. to 650 C., 500 C. to 600 C., 550 C. to 650 C. and 500 C. to 650 C.

23-28. (canceled)

29. The method as claimed in claims 1 wherein, after step C there is a step D, where step D is an acid leaching step.

30. The method as claimed in claim 29 wherein, the acid leaching step uses an acid leach which includes one or more acids selected from the list consisting of sulphuric acid, phosphoric acid, hydrochloric acid, hydrofluoric acid, and an organic acid with less than 15 carbon.

31. The method as claimed in claim 29 wherein, step D is carried out at a pH of less than 3.

32. The method as claimed in claim 31 wherein, step D is carried out at pH 2.

33. The method as claimed claim 30 wherein, the acid leach used contains sulphuric acid.

34. The method as claimed in claim 1 wherein, step B is monitored and/or controlled by one or more methods selected from the list consisting of directly or indirectly measuring the extent of reduction by determining the forms of iron present, determining the composition of the reducing gas within the furnace and determining the composition of an exiting gas stream.

35. (canceled)

36. The method of claim 1, wherein the ratio of water to hydrogen (H.sub.2O:H.sub.2), on a molar basis, is between 1:20 and 7:1.

37. The method of claim 1, wherein the water to hydrogen ratio is between 1:2 and 2:5.

38. The method of claim 1, wherein the method reduces the zinc ferrite to magnetite and zinc oxide whilst substantially avoiding the formation of wustite, iron or zinc.

39. The method of claim 1, wherein the mass loss during step B is controlled to no more than about 2% (2%+/0.5%).

Description

BRIEF DESCRIPTION OF DRAWINGS

[0036] By way of example only, a preferred embodiment of the present invention is described in detail below with reference to the accompanying drawings, in which:

[0037] FIG. 1 is a flowchart of the method:

[0038] FIG. 2 is a H.sub.2/H.sub.2O phase equilibria diagram for various iron forms at various temperatures:

[0039] FIG. 3 is a CO/CO.sub.2 phase equilibria diagram for various iron forms at various temperatures:

[0040] FIG. 4 is a TCD (Thermal Conductivity Detector) graph comparing various water to hydrogen ratios with a synthetic Electric Arc Furnace Dust (EAFD);

[0041] FIG. 5 is a TCD (Thermal Conductivity Detector) graph comparing real Electric Arc Furnace Dust (EAFD) to a synthetic Electric Arc Furnace Dust (EAFD); and

[0042] FIG. 6 is a TCD (Thermal Conductivity Detector) graph comparing synthetic Electric Arc Furnace Dust (EAFD) to a real Electric Arc Furnace Dust (EAFD).

DEFINITIONS

[0043] EAFD: Electric Arc Furnace Dust. [0044] FRANKLINITE: Some references use the term franklinite to refer to ZnFe.sub.2O.sub.4 (zinc ferrite) present in EAFD due to the form being similar. In many EAFD's (and the mineral form) franklinite contains ZnFe.sub.2O.sub.4 and ZnMnFeO.sub.4, as such herein where franklinite is used it contains zinc ferrite and may contain ZnMnFeO.sub.4. [0045] MAGNETITE: Fe.sub.3O.sub.4 [0046] SULPHURIC ACID: this is intended to include oleum. [0047] Water: this term is intended to include all fluid forms of water such as water vapour (steam) and liquid water as determined by the local environmental conditions. [0048] WUSTITE: sometimes called non-stoichiometric iron oxide (Fe.sub.0.947O), both Fe(II) and Fe(III) are present in the oxide. Sometimes written as FeO. [0049] ZINCITE: essentially zinc oxide (ZnO).

[0050] As the materials being processed by this method are complex where a chemical formula is used this may simply be to refer to the predominant species present.

[0051] For clarity the term about is normally used when the parameter is within +/15%, and approximately is used when the parameter is within +/10%, unless specified otherwise.

[0052] First Mode for Carrying out the Invention

[0053] Though described with reference to EAFDs the method is believed to be applicable to other zinc rich dusts from the steel or iron making process, for example dust from basic oxygen furnaces, where ZnFe204 is present. It is also felt that any process or industry that produces material containing zinc ferrite could also use the method to recover zinc, for example material from the mining and refining of zinc ore.

[0054] Referring to FIG. 1 the core steps of the method for processing a source material (1), most likely EAFD, are shown in the form of a flowchart.

[0055] For this variant the steps, in order, are: [0056] A. Analyse the source material (1); [0057] B. Partially reduce source material (1) using a reducing gas (2) containing hydrogen to form a reduced material (3); [0058] C. Cool reduced material (3); and [0059] D. Acid leach reduced material (3).

[0060] Step A is optional and will not be present in all variations, it is at present used to determine the optimum processing variables. Steps C and D are preferred but also optional.

[0061] Step A is the analysis of the source material (1) to determine the amount of zinc and zinc ferrite present so that the processing parameters can be determined. These parameters include the reduction temperature (RT), the reduction time (Rt), the hydrogen concentration (HC) of the reducing gas (2) and the reducing gas (3) flowrate (GF). Total zinc is at present determined by X-Ray Fluorescence (XRF) or a standard chemical analysis and the zinc ferrite fraction by X-Ray Diffraction.

[0062] Once step A has been completed the reduction conditions in step B can be determined. The aim is to reduce the zinc ferrite to magnetite and zinc oxide whilst avoiding the formation of wustite, iron or zinc. Wustite and iron are soluble in acid and as zinc is molten at around 420 C. this can cause problems in the reduction furnace.

[0063] Initially a bench scale reduction on a sample will be carried out based on the analysis from step A. The bench scale results and/or the equilibrium calculations will be used to determine the reduction time (Rt), reduction temperature (RT) and reducing gas flowrate (GF) for the bulk material reduction carried out in step B. Most likely this will be determined by setting the conditions to ensure the mass loss during the sample reduction is less than about 2% (2%+/0.5%).

[0064] The results from step A may allow the optimum selection of the hydrogen concentration (HC) in the reducing gas (3) as the equilibrium/maximum reduction is determined by this.

[0065] In most cases the reduction temperature (RT) and reducing gas flowrate (GF) for step B will be predetermined, and the reduction time (Rt) set by the bench scale test. This allows a batch furnace to be maintained at a preset optimum temperature and reducing gas flowrate (GF) to be fixed to minimise entrainment of the EAFD. For a continuous furnace it allows the temperature and reducing gas flowrate (GF) to be optimally set, and the transport speed of EADF through the furnace to be adjusted to give the optimum reduction time (RT).

[0066] Step B is the partial reduction of the zinc ferrite in the EAFD to magnetite and zinc oxide without the formation of significant amounts of wustite, metallic iron or metallic zinc. This process is carried out in a furnace of known type and configuration. To minimise the formation of wustite, metallic iron or metallic zinc it is likely the mass loss during this step will be controlled to no more than about 2% (2%+/0.5%). This is accomplished by using a reducing gas (2) including hydrogen in a carrier gas such as nitrogen or argon at temperatures below about 800 C. Most likely the processing will occur at between 500 C. and 750 C., and is believed that the optimum may lie between 650 C. and 750 C.

[0067] The fire or explosion risk will likely determine the maximum hydrogen concentration (HC) in the reducing gas (2). It is also likely that the maximum hydrogen concentration (HC) in the reducing gas (2) will vary with the processing equipment and safety requirements for the location.

[0068] Based on trials to date the reducing gas (2) is expected to be in a neutral carrier gas, such as nitrogen, containing from 0.25% to 10% hydrogen by volume, though a concentration of up to 70% hydrogen is felt to be the upper limit, and it is felt that below 66% is optimum.

[0069] Table 1, below, shows typical mass loss results for trial reductions of EAFD using a 5% (vol %) Hydrogen in nitrogen reducing gas (2).

TABLE-US-00001 TABLE 1 Reduction Reduction 5% H.sub.2 in N.sub.2 Mass Mass Mass Temperature Time Flowrate before After Loss C. Hours cc/min g g % Comparing reduction temperatures 600 1 100 2.11 2.076 1.6% 650 1 100 2.606 2.555 2.0% 700 1 100 2.7017 2.5438 5.8% 850 1 100 2.636 2.352 10.8% Comparing reduction times 650 1 100 2.606 2.555 2.0% 650 3 100 2.7424 2.5144 8.3%

[0070] As can be seen, using a reducing gas (2) with 5 vol % hydrogen in a neutral carrier gas (nitrogen) the mass loss for this sample of EAFD is below about 2% if the reduction temperature (RT) is below 650 C. and the reduction time (Rt) is around 1 hour. The mass loss is approximately linear with regards to time so the effect of increasing the time can be determined. Increasing the reduction temperature (RT) also changes the mass loss, though not in a linear manner as reaction kinetics are more sensitive to temperature changes.

[0071] A mass loss of about 2% has been found to reduce the franklinite to zinc oxide and magnetite, with a low risk of reducing the magnetite to wustite or iron which is undesirable. However, a mass loss of greater than 2% without the formation of metallic iron, metallic zinc or wustite, or the loss of zinc, could be tolerated.

[0072] Table 2, below, has XRD analysis results of the reduced material (3) at a variety of reduction temperatures (RT) and reduction times (Rt).

TABLE-US-00002 TABLE 2 Fe and Zn Reduction Reduction Gas flow Zinc after furnace Temperature Time (4% H2 in Ferrite treatment (magnetite (RT) (Rt). N2) (GF) reduction and ZnO preferred) [ C.] [mins] [cc/min] [] [] 600 60 100 Near No metallic Fe, complete some wustite 650 60 100 Complete Some metallic Fe, some wustite 650 180 100 Complete Significant metallic Fe, less wustite 700 60 100 Complete Some metallic Fe, some wustite 700 60 30 Complete No metallic Fe, no wustite 850 60 100 Complete Fe same as 700 C., but more Zn lost

[0073] As can be seen the amount of metallic iron and wustite formed is determined by the reducing gas flowrate (GF), reduction time (Rt) and reduction temperature (RT). More results are needed to determine the optimum parameters based on total zinc and zinc ferrite fraction analysis carried out in step A.

[0074] The above said, it is felt that the optimum reduction temperature (RT) will be in the range of 500 C. to 800 C., most likely below about 700 C. to keep the metallic iron and wustite in the reduced material (3) low.

[0075] Example 1 has a theoretical thermodynamic analysis of the hydrogen reducing gas (2) results. If these calculations are confirmed with EAFD trials then to minimise wustite formation and effect a sufficient reduction of the zinc ferrite to zinc oxide then a molar excess of H.sub.2 is required. The theoretical analysis suggests that a reduction temperature of 400 C. would achieve the desired result with a reducing gas (2) containing a 3 molar excess of Hydrogen

[0076] Once step B has been completed and the desired degree of reduction has been achieved step C is undertaken. Step C is the cooling of the reduced material (3) under reducing or neutral conditions until it is below 350 C., possibly even lower. If the reduced material is cooled in air above 400 C. then there is rapid formation of ZnFe.sub.2O.sub.4.

[0077] After cooling the reduced material (3) undergoes an optional step D which is an acid leaching step. It is expected that this will be using sulphuric acid at a pH below about 3.

Best Mode for Carrying out the Invention

[0078] In a further variation of the method step A is again optional, though preferred, and the reducing gas (2) includes water at a molar ratio of H.sub.2O:H.sub.2 of 1:2 to 2:5 i.e. 2 molar parts hydrogen to 1 part water through to 5 molar parts hydrogen to 2 parts water. This changes the reductive effect of the reducing gas (2). It has been found that the water to hydrogen ratio range of 20:1 to 1:5 has the required properties.

[0079] The use of H.sub.2/H.sub.2O results in a very different phase equilibria diagram, see FIG. 2, to that when CO/CO.sub.2 is used, see FIG. 3. At equilibria the CO/CO.sub.2 system favours the reduction to iron. At equilibria a H.sub.2/H.sub.2O reducing gas (2) favours the formation of magnetite. As such depending on whether the reducing gas contains hydrogen or it contains carbon monoxide the iron equilibria tended towards will be different.

[0080] The use of hydrogen and water in a neutral carrier gas (most likely nitrogen) as the reducing gas (2) has been found to provide a means of adjusting the reductive power of the reducing gas (2). By controlling the water vapour present it is possible to limit the reductive potential of the reducing gas (2) lowering the formation of undesirable species.

[0081] The water vapour concentration can be measured for the reducing gas (2) into the furnace and the spent reducing gas (2) exiting the furnace and by controlling the water content of the reducing gas (2) the reductive power of the reducing gas (2) can be controlled. Example 2 has a theoretical thermodynamic analysis of the water/hydrogen reducing gas (2) results, if these are confirmed with EAFD trials then to minimise wustite formation the temperature will need to be maintained at below 650 C.

[0082] In a further variant it is possible to add a small amount of carbon monoxide to the reducing gas (2). It is thought that a H.sub.2:H.sub.2O:CO ratio of 2:(1 to 10):(0 to 10) at a concentration of between 0.25% to 15% (by volume) in a neutral carrier gas will allow additional control of the reduction reaction that occurs. Of course, using CO is undesirable due to the environmental concerns, and it is an odourless poisonous gas which can make handling difficult, but it may improve the reaction kinetics thus reduce the energy input required.

[0083] In some variants step A and/or step D are not present. If step A is not present then step B is controlled by directly or indirectly measuring the extent of reduction by measuring the forms of iron present, and/or measuring the composition of the reducing gas (2) within the furnace or using a Thermal Conductivity Detector (TCD) on the exiting gas stream. Step D may not be undertaken if the reduced material (3) is stored or sent off site for further processing.

[0084] Monitoring and/or controlling the reaction in step B, whether other steps are present or not, can be by directly or indirectly measuring the extent of reduction by measuring the forms of iron present, and/or measuring the composition of the reducing gas (2) within the furnace and/or using a Thermal Conductivity Detector (TCD) on the exiting gas stream.

[0085] In some variants there is a hydrogen recycling loop which extracts the hydrogen from the exhaust gases. The exhaust gas is processed and this process returns either hydrogen for addition to a neutral carrier gas, or a reducing gas (2) containing hydrogen.

[0086] In some variants the composition of the reducing gas (2) in step B is intentionally changed over time, staying within the parameters given previously. This intentional change in the composition of the reducing gas (2) will most likely be accomplished by adding hydrogen or water at a predetermined time after step B commences, but before later steps occur. This predetermined time delay is most likely to be implemented when the method is implemented as a continuous process. For example the water could be added part way through step B to allow a rapid initial reaction to occur with the water added to mediate the reaction end point. This predetermined time delay is most likely to be determined before the temperature in step B reaches a certain value or falls within a certain range of values. It is expected that this predetermined time delay will be before the temperature reaches 650 C., 600 C., 550 C. or 500 C., and most likely when the temperature falls within a predetermined range for example 600 C. to 650 C., 550 C. to 600 C. or 500 C. to 550 C.

[0087] Thermodynamic simulations have predicted that the process will operate at up to about 1000 C. successfully, though the sample runs have been limited to 900 C. due to furnace limitations. We believe that the method will operate successfully, but not necessarily optimally, at temperatures up to about (+/10%) 1000 C. and this is supported by thermodynamic simulations, as such about 1000 C. appears to be an upper practical limit. Processing below 800 C. has significant advantages so it is believed that 800 C. or below is the optimum range, with 500 C. to 750 C. and 550 C. to 650 C. ranges providing additional benefits such as controllability.

EXAMPLES

Theoretical Examples

Example 1

[0088] Hydrogen reducing gas with no additional water, thermodynamic predictions for various reductants between 400 C.-700 C. are given below (FactSage calculation using the Equilib-Web):

[0089] The zinc ferrite reduction with hydrogen uses H.sub.2 per zinc ferrite:

3ZnFe.sub.2O.sub.4+H.sub.2.fwdarw.3ZnO+2Fe.sub.3O.sub.4+H.sub.2O ()

[0090] Reacting 1 mol ZnFe.sub.2O.sub.4 with 1/3 mol H.sub.2 (i.e. no excess) gives:

TABLE-US-00003 TABLE 3 Reduction of 1 mol ZnFe.sub.2O.sub.4 with mol H.sub.2 (1 atm) at different temperatures Temperature/ C. 400 500 550 600 650 700 ZnO/mol 0.53 0.75 0.83 0.89 0.93 0.95 ZnFe.sub.2O.sub.4/mol 0.47 0.25 0.17 0.11 0.07 0.05 Fe.sub.3O.sub.4/mol 0.35 0.50 0.55 0.59 0.62 0.63 FeO/mol 0 0 0 0 0 0 H.sub.2/mol 0.16 0.08 0.06 0.04 0.02 0.02 Zn/mol (vapour) 7 10.sup.11 5 10.sup.9 3 10.sup.9 1.2 10.sup.7 4 10.sup.7 1 10.sup.6

[0091] To reach 95% with stoichiometric H.sub.2 quantity, around 700 C. is needed.

[0092] Same calculations performed with 3excess H.sub.2:

TABLE-US-00004 TABLE 4 Reduction of 1 mol ZnFe.sub.2O.sub.4 with 1 mol H.sub.2 (1 atm.) at different temperatures Temperature/C. 400 500 550 600 650 700 ZnO/mol 1 1 1 1 1 1 ZnFe.sub.2O.sub.4/mol 0 0 0 0 0 0 Fe.sub.3O.sub.4/mol 0.67 0.67 0.67 0.67 0.59 0.51 FeO/mol 0 0 0 0 0.22 0.48 H.sub.2/mol 0.67 0.67 0.67 0.67 0.59 0.51 Zn/mol (vapour) 4 10.sup.10 9 10.sup.8 8 10.sup.7 6 10.sup.6 2 10.sup.5 7 10.sup.5

[0093] Conclusion. Thermodynamics says to operate at below 700 C. excess H.sub.2 is required (at 3excess the equilibrium conversion can reach 100% at 400 C.). This will be fine as long a H.sub.2 recycling can be implemented. This also shows that keeping temperatures below 650 C. could be useful to limit FeO (wustite) formation.

Example 2

[0094] Water and Hydrogen reducing gas, thermodynamic predictions for various reductants 600 C. are given below (FactSage calculation using the Equilib-Web):

[0095] The reaction is:

3ZnFe.sub.2O.sub.4+H.sub.2.fwdarw.3ZnO+2Fe.sub.3O.sub.4+H.sub.2O ()

[0096] Equilibrium calculations at 600 C. with 1 mol ZnFe.sub.2O.sub.4 with 1 mol H.sub.2 (i.e. 3excess) gives:

TABLE-US-00005 TABLE 5 Reduction of 1 mol ZnFe.sub.2O.sub.4 at 600 C. with different H.sub.2:ZnFe.sub.2O.sub.4 mol ratios (1 atm) H.sub.2:ZnFe.sub.2O.sub.4 mol ratios 1:3 1:2 1:1 1.17:1 3:2 2:1 (no excess) (1.5x excess) (3x excess) (3.5x excess) (4.5x excess) (6x excess) ZnFe.sub.2O.sub.4/mol 0.114 0 0 0 0 0 ZnO/mol 0.886 1.0 0.9999 0.9999 0.9999 0.9999 Fe.sub.3O.sub.4/mol 0.591 0.6667 0.6667 0.633 0.528 0.371 FeO/mol 0 0 0 0.100 0.415 0.887

[0097] Thermodynamics shows that above 1:1 mol ratios, at 600 C., reduction of iron oxides to FeO will occur. Steam (water vapour) can limit the further reduction of iron oxides. Calculations at 6excess (H.sub.2:ZnFe.sub.2O.sub.4 mol ratio=2:1) show this effect.

TABLE-US-00006 TABLE 6 Reduction of 1 mol ZnFe.sub.2O.sub.4 with 2 mol H.sub.2 at 600 C. with different H.sub.2O:H.sub.2 mol ratios (1 atm.) H.sub.2O:H.sub.2 mol ratios 1:20 1:10 1:5 1:2 5:1 7:1 ZnFe.sub.2O.sub.4/mol 0 0 0 0 0 0.47 ZnO/mol 0.9999 0.9999 0.9999 0.9999 0.9999 0.53 Fe.sub.3O.sub.4/mol 0.44 0.508 0.645 0.6667 0.6667 0.35 FeO/mol 0.681 0.476 0.065 0 0 0

[0098] Adding small amounts of steam (1:5=steam at 20% of H.sub.2) removes the reduction to FeO. Adding more than 6 mol steam (water vapour): 1 mol H.sub.2, starts to limit ZnFe.sub.2O.sub.4 reduction.

TABLE-US-00007 TABLE 7 Reduction of 1 mol ZnFe.sub.2O.sub.4 with 10 mol H.sub.2 at 600 C. with different H.sub.2O:H.sub.2 mol ratios (1 atm.) H.sub.2O:H.sub.2 mol ratios 0:1 1:10 1:5 1:2 5:1 7:1 ZnFe.sub.2O.sub.4/mol 0 0 0 0 0 0 ZnO/mol 0.9999 0.9999 0.9999 0.9999 0.9999 0.9999 Fe.sub.3O.sub.4/mol (Fe forms) (Fe forms) 0 0.6667 0.6667 0.6667 FeO/mol 0.567 1.32 2.0 0 0 0 ** at 7.5 mol H.sub.2O:1 mol H.sub.2, the zinc ferrite starts to not reduce. At 8:1, no zinc ferrite reduces

[0099] Tables 6 and 7 show that the addition of water vapour can prevent the over reduction of the zinc ferrites by favouring the formation of magnetite (Fe.sub.3O.sub.4).

Example 3

[0100] A programmed reduction of synthetic EAFD using no water and water:hydrogen ratios of 1.6:1 and 2.1:1 is shown in FIG. 4.

[0101] In this example 50 mg of a synthetic EAFD was heated from 100 C. to 850 C. in a 5% H.sub.2 in Argon stream with a temperature increase of 20 C./minute. The outlet gas was monitored using a thermal conductivity detector. The detector effectively measures the hydrogen in the exit gas stream giving an indication of the reduction, the larger the number the greater the reduction that occurred.

[0102] In FIG. 4 line SNW is the synthetic EAFD where no water has been added, the line SD1 is where the water to hydrogen ratio is 1.6:1 (1.6 mole water to 1 mole hydrogen) and SD2 is where the water to hydrogen ratio is 2.1:1.

Effect of Water to Hydrogen Ratio:

[0103] As can be seen by comparing the SNW to SD1 or SD2 the water addition slows the reaction rate and the reaction starts at a lower temperature. However, counterintuitively increasing the water: hydrogen ratio from 1.6:1 to.2.1:1 increases the reaction rate and causes it to commence at a lower temperature.

Example 4

Real EAFD vs Synthetic EAFD

[0104] The experiment carried out in Example 3 was repeated with two real EAFD's and a synthetic EAFD with the same zinc content and the results graphed in FIG. 5. The water to hydrogen ratio used was 1.6:1.

[0105] Referring to FIG. 5 the lines shown are: [0106] SNW Synthetic EAFD with no water added; [0107] SD1 Synthetic EAFD with a water to hydrogen ratio of 1.6:1; [0108] RD1 Real EAFD sample 1 with a water to hydrogen ratio of 1.6:1; and [0109] RD2 Real EAFD sample 2 with a water to hydrogen ratio of 1.6:1.

[0110] As can be seen it appears that the synthetic dust reaction rate occurs in two stages with the highest reaction rate occurring at around 820 C. whereas the real EAFD's (based on two actual samples which are believed to be representative) do not exhibit this apparent two stage process and have peak reaction rates over the range of 700 C. to 720 C. This means that the reaction characteristics of synthetic EAFD are not a good match for real EADF. Based on FIG. 5 using the synthetic EAFD results to optimise a real EAFD processing plant would result in designing for around 820 C. resulting in significantly lower reaction rates and higher energy use than using real EAFD in the optimisation trials. Based on these results experiments using synthetic EAFD cannot be used to determine reduction conditions for real EAFD processes with any reliability.

Example 5

[0111] A programmed reduction of a synthetic EAFD and a real EAFD using no water is shown in FIG. 6.

[0112] In this example 50 mg of a synthetic EAFD and then a real EADF were heated from 100 C. to 850 C. in a 5% H.sub.2 in Argon stream with a temperature increase of 20 C./minute. The outlet gas was monitored using a thermal conductivity detector. The detector effectively measures the hydrogen in the exit gas stream giving an indication of the reduction, the larger the number the greater the reduction that occurred.

[0113] In FIG. 6 the lines shown are: [0114] SNW-synthetic dust, no water; and [0115] RNW-real dust, no water.

[0116] As can be seen the reaction peak for the synthetic EAFD is lower and occurs at a higher temperature (720 C.) than the real EAFD. The reaction peak of the real dust occurs at about 690 C. and is about 40% higher. Interestingly the real EAFD peak is much narrower than the synthetic dust peak and as such this may allow specific optimisation of the reaction conditions. Once again, based on these results, experiments using synthetic EAFD cannot be used to determine reduction conditions for real EAFD processes with any reliability.

KEY

[0117] 1 source material (EAFD, BOFD), [0118] 2 reducing gas; [0119] 3 reduced material (material after partial reduction of source material using hydrogen containing reducing gas); [0120] RT Reduction temperature [0121] Rt Reduction time [0122] HC Hydrogen concentration of reducing gas [0123] HF reducing gas flowrate