WO2025175348A1 - Blast furnace with capture and recycling of carbon monoxide from top gases and methods for improved blast furnace operational efficiency - Google Patents

Abstract

A blast furnace smelter operation comprising: a blast furnace comprising an upper raw material feed, an intermediate reaction zone and a hearth having an iron product outlet and a slag outlet; a top gas outlet for removal of a CO/CO₂ containing top gas from said blast furnace; a CO/CO₂ separator for separating CO gas from said CO/CO₂ containing top gas; and at least one tuyere for injection of separated CO gas back into said blast furnace.

Description

BLAST FURNACE WITH CAPTURE AND RECYCLING OF CARBON MONOXIDE FROM TOP GASES AND METHODS FOR IMPROVED BLAST FURNACE OPERATIONAL EFFICIENCY

FIELD OF INVENTION

The present invention relates to a Blast Furnace (BF) with capture and recycling of Carbon Monoxide (CO) from Top Gases (TGs) and methods for improved BF operational efficiency.

More specific embodiments of the invention relate to a BF with elimination of Nitrogen (N2) in the “Blast”, capture of TGs, separation of those TGs into gaseous CO and Carbon Dioxide (CO2), recycling of the CO to replace the Pulverised Coal Injection (PCI) and air blast, introducing gaseous CO2 into the coke oven bed or using liquid CO2 instead of water to quench the coke to obtain more CO for injection higher up in the BF, and other methods for improved BF operational efficiency.

BACKGROUND ART

Referring to Figure 1 , a conventional BF operation is illustrated. The way BFs are currently operated, including process flow quantities for a nominal 4.4 MTPA of Pig Iron, Project Iron Boomerang (PIB) standard smelter, are shown in Figure 1. In that regard, reference herein to Project Iron Boomerang (PIB) refers to a plan to connect northern Queensland’s coalfields to northern Western Australia’s iron ore reserves with a railway that transports the minerals both ways, supplying steel mills at both ends.

CO is the main reductant in BF processes. It is the CO that reacts with the various iron oxides to remove the Oxygen (O2) in the iron oxides from the iron. The BF produces CO from the partial combustion of Carbon, which then progressively reacts with the Oxygen in the iron ore to produce CO2 and pig iron.

The Carbon (C) in an existing BF comes from the:

• Pulverised Coal Injection (PCI) coal injected with the “Blast”, the hot air pumped into the base of the BF; and

• Coking Coal fed in with the “Charge” at the top of the BF.

Complete combustion can occur resulting in CO2, or the CO from partial combustion can pick up an Oxygen atom from iron oxides in their various states in the BF Column to form CO2. The CO and CO2 rise through the Column. Some of the CO picks up Oxygen from the iron oxides, some of the CO2, reacts with the C in the BF, converting the CO2 into CO and any H2O present to H2. The CO and CO2 remaining exit the top of the BF as TGs.

The source of the C for the breakdown of the CO2 comes from the Coke as it progresses down through the porous and permeable solid burden in the BF shaft and from the PCI injected at the base of the BF.

The source of the H2O is mainly from the moisture content in the PCI, with minor contributions from the moisture in the iron ore, measured as Loss on Ignition (LOI). The reactions are:

C + CO2 = 2CO.

C + H2O CO +H₂

Some of the CO produced in the BF does not react and exits the Blast Furnace in the TGs. The TGs are typically produced at a rate of 2.5-3.5 tonnes/tonne of liquid steel (tls) and typically contain 20-30% CO, 20-25% CO2 and 2-6% Hydrogen (H2). The remaining 48.5% is Nitrogen (N2) (Huth and Heilos, 2013).

The loss of the CO in the TGs is not just a significant loss of reductant material but also results in the need for more coal and coke within the BF to replace the Carbon within the CO that is lost. Many BF operations burn this CO to recover energy. It is often burnt to preheat the air in the Blast. The presence of N2 in the TGs, due to the use of air containing N2 in existing BFs, makes potential separation and recovery of the CO from the TGs very difficult.

The subject matter claimed herein is not limited to embodiments that solve any disadvantages or that operate only in environments such as those described above. Rather, this background is only provided to illustrate exemplary technology areas where some embodiments described herein may be practiced.

Various aspects and embodiments of the invention will now be described.

SUMMARY OF INVENTION

As mentioned above, the present invention relates generally to a BF with capture and recycling of CO from TGs and methods for improving BF operational efficiency.

According to one aspect of the invention there is provided a blast furnace smelter operation comprising: a blast furnace comprising an upper raw material feed, an intermediate reaction zone and a hearth having an iron product outlet and a slag outlet; a top gas outlet for removal of a CO/CO2 containing top gas from the blast furnace; a CO/CO2 separator for separating CO gas from the CO/CO2 containing top gas; and at least one tuyere for injection of separated CO gas back into the blast furnace.

In a preferred embodiment, a feed for the blast furnace inlet is N2 free and comprises the separated CO gas, optionally further comprising PCI and/or BioChar, preferably BioChar. If PCI is part of the feed, a much smaller amount of PCI is used compared with that currently used in conventional BF operations. Preferably, the top gas is N2 free and CO gas separation comprises a vapourliquid separation. This is possible due to the different vapour-liquid properties of the two gases. In that regard, the CO/CO2 separator may comprise a pressure vessel that condenses CO2 gas to CO2 liquid facilitating separation of CO gas from the CO/CO2 containing top gas.

In certain embodiments, the CO/CO2 containing top gas is pre-cooled before vapour-liquid separation, preferably in a heat exchanger, where it can help to preheat the recovered CO being fed back into the BF as discussed below.

In an alternative embodiment, the top gas is N2 free and CO gas separation comprises an amine absorption separation. The amine absorption separation may comprise feeding said CO/CO2 containing top gas to an absorber containing an amine solvent to absorb CO2, recovering CO gas from the absorber, and feeding CO2 laden amine solvent to a desorber to recover CO2 gas. According to this embodiment, the method preferably comprises solvent cooling to 40°C to facilitate CO2 absorbance and reboiling the CO2 laden amine solvent in a reboiler to a temperature of about 120°C prior to desorbing.

The amine solvent may be selected from primary, secondary and tertiary amines, for example the amine solvent may comprise monoethanolamine (MEA), methyldiethanolamine (MDEA)Zpiperazine, and 2-amino-2-methyl-1- propanol (AMP).

CO2 recovered from the CO/CO2 separation may be collected for external applications. In certain embodiments, CO2 liquid recovered from the vapourliquid separation is used, at least in part, for coke quenching in a coke oven of the blast furnace smelter operation, and/or CO2 gas recovered from the CO/CO2 separation mat be used prior to quenching to obtain more CO, prior to feeding coke to the raw material feed of the blast furnace. For example, the CO2 liquid may be fed to the coke oven by expansion into the base of the coke oven or dropped as a liquid onto the top of the coke as presently occurs with water quenching, or wherein the CO2 gas is pumped into the base of the coke oven and converted to CO and collected from a top of the coke oven, prior to quenching.

CO gas formed from the coke quenching may be fed back into the blast furnace, preferably through at least one tuyere at one or more location from one third to two thirds up the blast furnace. The CO from the coke ovens may be injected at any level, including above that where the CO injected at the base has been consumed. The higher the level of injection, the less pressure and preheating required. In some embodiments multiple injection levels may be used.

In certain embodiments, before injection back into the blast furnace, the separated CO gas is re-heated in a heat exchanger with the CO/CO2 containing top gas removed from the blast furnace.

In order to minimise addition of water to the blast furnace, iron ore fed to the raw material feed is preferably preheated to reduce or remove water content.

Preferably, the at least one tuyere for injection of separated CO gas back into the blast furnace is located at an upper region above the hearth. The at least one tuyere but generally the use of multiple tuyeres for injection of separated CO gas back into the blast furnace generally comprises the use of existing tuyeres used to inject the PCI and the blast in existing conventional BFs.

If required, O2 gas mixed with CO2 gas drawn from the TG separator and combined in a mixture that replicates the combustion characteristics of air and PCI coal may be fed to the blast furnace during startup of the blast furnace smelter operation or as a replacement of the air and PCI during a transition period.

According to another aspect of the invention there is provided a method of operating a blast furnace comprising: capturing a CO/CO2 containing top gas from the blast furnace; separating CO gas from the CO/CO2 containing top gas; and injecting separated CO gas back into the blast furnace. There is further provided a method of retrofitting a blast furnace comprising: providing a CO/CO2 separator for separating CO gas from CO/CO2 containing top gas collected from the blast furnace; and providing a recirculation loop for recirculating separated CO gas to the blast furnace.

Again, a feed for the blast furnace containing the separated CO gas is preferably N2 free and may further comprises small quantities of PCI and/or BioChar, preferably BioChar.

Likewise, the top gas is N2 free and CO gas separation comprises either a vapour-liquid or amine capture separation. The above methods may comprise condensing CO2 gas to CO2 liquid for quenching of the coke. The methods may comprise cooling the CO/CO2 containing top gas before vapour-liquid or amine capture separation, preferably in a heat exchanger. As with the previous aspect of the invention, the top gas may be N2 free and CO gas separation may comprise amine absorption separation. Again, the amine absorption separation may comprise feeding said CO/CO2 containing top gas to an absorber containing an amine solvent to absorb CO2, recovering CO gas from the absorber, and feeding CO2 laden amine solvent to a desorber to recover CO2 gas. The method preferably comprises solvent cooling to 40°C to facilitate CO2 absorbance and reboiling the CO2 laden amine solvent in a reboiler to a temperature of about 120°C prior to desorbing. The amine solvent may be selected from primary, secondary and tertiary amines, for example the amine solvent may comprise monoethanolamine (MEA), methyldiethanolamine (MDEA)Zpiperazine, and 2-amino-2-methyl-1 -propanol (AMP).

Once again, the methods may comprise recovering CO2 from the CO/CO2 separation for external applications.

The methods may also comprise using CO2 liquid recovered from the vapourliquid separation, or liquified CO2 from the amine capture, at least in part, for coke quenching in a coke oven associated with the blast furnace, and/or using CO₂ gas recovered from the CO/CO2 prior to quenching to obtain more CO, prior to feeding coke to the blast furnace. For example, the methods may comprise feeding the CO2 liquid to the coke oven by expansion into a base of the coke oven or by dropping onto the coke as currently occurs with water quenching operations, and/or pumping the CO2 gas into the base of the coke oven and converting to CO and collecting the CO from a top of the coke oven, prior to quenching. The methods may further comprise feeding CO gas formed from the coke quenching back into the blast furnace, preferably through at least one tuyere disposed at a location from one third to two thirds up the blast furnace, generally above the level where the CO injected at the base is expected to have been consumed.

In certain embodiments the methods comprise re-heating separated CO gas in a heat exchanger with the CO/CO2 containing top gas removed from the blast furnace before injection of the separated CO gas back into the blast furnace.

The methods may comprise preheating an iron ore feed to reduce of remove water content before being fed into the blast furnace.

In preferred embodiments, the separated CO is injected back into the blast furnace at an upper region of a hearth of the blast furnace, generally through existing tuyeres used to inject the PCI and the blast in existing conventional BFs.

The methods may further comprise feeding O2 gas mixed with CO2 recovered from the TG separation process in a mixture designed to replicate the combustion characteristics of air and PCI coal to the BF during startup of the BF or transitioning an existing BF to the new system of operation.

According to yet another aspect of the invention there is provided a method of reducing PCI use and GHG generation in a blast furnace comprising: capturing a CO/CO2 containing top gas from the blast furnace; separating CO gas from the CO/CO2 containing top gas; and injecting separated CO gas back into the blast furnace. According to still another aspect of the invention there is provided use of CO gas separated from a CO/CO2 containing top gas of a blast furnace as a reductant feed in the blast furnace, thereby reducing PCI use and GHG generation in the blast furnace.

BRIEF DESCRIPTION OF ACCOMPANYING DRAWINGS

To further clarify various aspects of some embodiments of the present invention, a more particular description of the invention will be rendered by references to specific embodiments thereof, which are illustrated in the appended drawings. It should be appreciated that these drawings depict only typical embodiments of the invention and are therefore not to be considered limiting on its scope. The invention will be described and explained with additional specificity and detail through the accompanying drawings in which:

FIG. 1 illustrates a sectional view of the operation of a conventional blast furnace smelter of a size that would produce 4.4 MTPA of liquid pig iron.

FIG. 2 illustrates a sectional view of the operation of a blast furnace smelter according to an embodiment of Stage 1 of the present invention - Capture & Recycling of CO from Top Gases.

FIG. 3 illustrates a graph of CO2 Phases as determined by Temperature & Pressure.

FIG. 4 illustrates an embodiment of a CO/CO2 Separation System.

FIG. 5 illustrates an embodiment of a Top Gas/ Recovered CO Heat Exchanger.

FIG. 6 illustrates a sectional view of a blast furnace smelter according to an embodiment of the present invention - Increased Throughput from Recycling CO from Top Gas. FIG. 7 illustrates an embodiment of a CO2 Quenching Coke Oven.

FIG. 8 illustrates a sectional view of the operation of a blast furnace smelter according to an embodiment of the present invention - Increased Throughput from Recycling CO from Top Gas and CO from Coke Ovens.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

Hereinafter, this specification will describe the present invention according to the preferred embodiments. It is to be understood that limiting the description to the preferred embodiments of the invention is merely to facilitate discussion of the present invention and it is envisioned without departing from the scope of the appended claims.

Capturing and Recycling of BF TGs

The TGs in an existing BF are generated from a combination of hot Air injected into the base of the BF (the “Blast”), with Pulverised Coal Injection (PCI), coke and the combustion/reduction processes in the BF.

The N2 in the TGs is injected as part of the hot air in the Blast and passes through the BF as a chemically inert material that plays no chemical role in the BF. It absorbs heat from the processes in the BF or from preheating. In many operations, some of this heat is later recovered and used to preheat the blast air. Due to some efficiency losses in the heat transfers, the N2 has an overall negative impact on the BF processes.

As stated above, the range of TG production across different BFs is 2.5 to 3.5 t/tls. A Mass Balance, reproduced in Table 1 , has been undertaken for both the existing processes and that proposed for the purposes of illustrating the potential benefits arising:

Referring to Table 1 , adopting an average TG quantity of 3 t/tls results in the Project Iron Boomerang (PIB) production expectation of 44 MTPA of pig iron from 10 smelters with 4.4 MTPA capacity each as a base, producing 3 * 44 = 132 MTPA of TG.

The Mass Balance calculations are based on averages of different BFs and feed stock specifications. The advantages expected are believed to be significantly greater than the likely error arising from using averaged data.

Table 2 below uses the average of the range of TG concentrations to list the proportions by volume, and by using the molecular weight of each gas, determines the relative proportions by weight and actual weight of each of the TGs that would be produced per year by PIB if conventional BFs were used.

Note that this table does not include Argon (Ar), which makes up almost 1 % of air. Ar plays a similar role to N2 in the BF, being chemically inert, but impacting on temperatures due to absorbing heat. Ar is taken into account in the Mass Balance calculations in Table 1 for existing BFs but will not occur in the proposed operation because the proposed blast will not contain Ar as it will not use air.

Nitrogen is 79% of air by volume and 76.5% of air by weight. Table 2 shows 44.4% of the TG is N2 by weight, and the total quantity of TG is 132 MTPA. The air injected can thus be calculated to be 44.4% * 132 MTPA / 76.5% Nitrogen proportion in air, = 76.69 MTPA of air in the Blast.

With O2 being 22% of air by weight, the air in the conventional blast for a PIB furnace would contain 16.97 MTPA of O2.

It is important to understand the BF process to appreciate the potential gain from recovering and recycling the CO in the TG.

Four main reactions are occurring in a BF to achieve the reduction of iron ore to pig iron (Yang et al., 2014). Haematite Iron Ore, as is planned to be used in PIB, is predominantly made of Fe2Os. The University of Birmingham's 2023 paper on Perovskite splitting of CO2 explains the BF process as:

Two reactions are occurring below 570°C (843°K), i.e., in the upper sections of the BF:

(1) 3Fe₂0s + CO -+ 2Fe3O4 + CO2

AH843K = - 30.050 kJ

(2) FezOt + 4CO 3Fe + 4Cft

AH843K = - 44.478 kJ

Both of these reactions are exothermic, helping to heat the charge as it progresses downwards through the Column. The extraction of molten pig iron and slag from the base of the BF draws this heated material deeper into the BF column where other reactions occur.

As the charge into the BF drops lower in the column and the temperature increases above 570°C the following two reactions become dominant: (3) FesC>4 + CO — > 3FeO + CO2

AH1073K = + 9.707 kJ

(4) FeO + CO ^ Fe + CO2

AH1273K = - 15.653 kJ, AH1473K = - 16.484 kJ

It is clear that in all four of these reactions the CO is the reductant.

Not all of the CO reacts with the iron ore. The BF process is designed to have a surplus of CO to maximise the chance of a CO molecule contacting an iron oxide molecule for reduction of the iron ore. The surplus CO generated to achieve this is lost to the TGs.

The Mass Balance attached in Table 1 shows that 29.99 MTPA of the CO generated is lost in the BF TG emissions.

It is proposed here to capture and recycle the CO as illustrated in Figure 2, and this is expected to significantly improve the BF efficiency and reduce Greenhouse Gas (GHG) emissions.

As shown in Table 1 , the O2 content in the blast is 16.97 MTPA. If all the CO in the TGs is captured and recycled, the 29.99 MTPA of CO would be comprised of 12/28 C, or 12.85 MTPA C, and 16/28 O2, or 17.14 MTPA O2.

This means that the CO recovered from the TGs can replace a similar quantity of both the C in the PCI and the O2 in the air injected at the base of the BF in the “blast”.

The 12.85 MTPA of C in the CO replaces the 13.31 MTPA of C in the PCI and the 17.14 MTPA of O2 in the CO replaces the 16.97 MTPA of O2 in the Blast. There may still be a need to inject some Carbon with the CO to provide some of the C in the Pig Iron. Some believe that the 4% C in the Pig Iron needs to be physically forced into the molten iron because, being of lower density, it floats on top of the molten iron. If this is the case, and a small quantity is required, a BioChar could be used instead of PCI so that there is no moisture, Sulphur or Phosphorous introduced into the BF.

As there is molten slag sitting above the molten iron, it is more likely that the C in the Pig Iron is absorbed from the coke and the presence of C within the Pig Iron lowers the melting point and reduces the viscosity that assists it in sinking though the Column. The fluxing of the molten slag means that this and the molten iron are immiscible liquids, and due to the higher density of the molten iron, it sinks through the molten slag.

It is important to note that the invention does not require all of the PCI to be eliminated. It only requires the capture of CO from the hot TGs and reintroduction of the captured CO by injection into the BF.

It is envisaged that some, most or even all of the PCI can be replaced, but simulation modelling followed by physical testing is required to determine the proportion of PCI that can be replaced with the recirculated CO.

The Mass Balance calculations in Table 1 show that there is 21.12 MTPA of coke containing 18.48 MTPA of C, and 14.78 MTPA of PCI containing 13.31 MTPA of C. The total C contained within the coke and PCI in an existing operation is therefore 31 .79 MTPA.

If all of the CO from the TGs can be captured and recycled, this would reduce GHGs by 12.85 I 31 .79 MTPA of PCI and coke = 40.4% and therefore that up to 40.4% of the GHGs produced by the BFs could be avoided.

As the C and O2 in the CO are a good match for the C and O2 in the PCI and the air in the Blast, it is proposed as a first step to replace all of the PCI and all of the air, with recirculated CO, captured and recycled from the TGs as illustrated in Figure 2. If a small quantity of C is needed, then a small quantity of PCI or preferably BioChar, can be injected with the recirculated CO.

In addition to the mass balance, the energy balance also needs to be considered. This is best done with simulation modelling as there are so many factors involved. The considerations below therefore are best considered as approximate.

The heat of combustion for each of the C to CO, C to CO2 and CO to CO2 are listed below.

2C + O2 = 2CO is 220 Kj/mol

2C + 2O2 = 2CO₂ is 393.5 Kj/mol

2CO + O₂ — 2CO₂ is 566.0 Kj/mol

In the best case, if 40% of the Carbon is removed by eliminating all of the PCI, then 40% of the 220 Kj/mol generated by converting C to CO is lost. The ultimate reaction to turn C to CO2 produces 393.5 KJ/mol. The two reactions total 613.5 Kj/mol. The reaction to turn CO into CO2 generates 566 KJ/mol, so only 40% of (613.5 - 566) = 40% of 47.5 KJ/mol = 19 KJ/mol is lost. This is 3.1% of the total 613.5 kJ/mol generated.

In the worst case, it is envisaged that it could be that 40% of the 220 kJ/mol generated are lost. This would increase the percentage heat lost to 40% of 220 kJ/mol = 88 kJ/mol. This could be 88 1566 = 15.5% of the heat generated.

If all the PCI is eliminated, the Mass Balance calculations in Table 1 shows that there is 1.48 MTPA less PCI ash, an approximately equivalent amount of limestone less due to less fluxing required, 58.66 MTPA less N2 and 1 .06 MTPA les Ar. This totals 62.68 MTPA of inert material that needs to be heated in the BF. With 192.34 MTPA of total material passing through the BF, the inert material comprises approximately 33% of the total feed. This means that the 3.1% to 15.5% less energy needs to heat up 33% less material, so it appears likely that higher temperatures can be generated.

Most of the heat is absorbed by the solids, which reduce by 22%. This also exceeds the potential 15.5% less heat generated.

As the column will be dense and chemical reactions are affected by pressure, generally proceeding more quickly, throughput time is also likely to reduce.

The actual result will be dependent on the different specific heat values for the different materials being heated as well as the different heat of combustion for the reactions with the iron oxide minerals as opposed to just the heat generated by burning C versus CO.

Again, simulation modelling will be required to forecast the ultimate result more accurately, but it appears likely that the lesser heat generated in total will be offset either fully or in part by the lesser quantities of materials to be heated combined with the opportunity to pre-heat the feed to produce the temperatures necessary for the most efficient operation of the BF.

If the modelling shows an excess in heat required, lesser volumes of CO can be used in the blast to regulate the temperature. If the model shows a shortfall in heat required, it is envisaged that just a small volume of PCI, or preferably BioChar and perhaps some more CO or even O2 will be required to make up the shortfall. It is considered that the BioChar is a better solution, as this has been devolatilised and is therefore, unlike the PCI, free of moisture, Sulphur and Phosphorous. This has significant benefits with respect to the steel quality that can be achieved for a given cost.

With regard to the startup of the BF, it is important to note that there will be a need to initially use O2 in the Blast at either startup, or as a replacement for an Air Blast, in combination with CO2 recovered from the TG separator in a mixture that replicates the combustion characteristics of air. In either case, once the BF is operating with no N2 in the TGs, the O2 and CO2 mixture in the Blast can be replaced by just the CO captured and recycled from the TGs.

As such, the use of O2 and CO2 mixture in the Blast is only necessary for a short period of time. It is not required during normal operations — only during startup, or transition from an existing operation using an Air Blast until such time as there is no N2 in the TGs.

O2 is expensive to produce, but the quantities required during startup or transition are comparatively very small.

In addition to the anticipated savings in cost from eliminating the use of PCI, there are several consequential benefits arising from doing this. These are:

• The inert ash within the PCI, which absorbs heat to become part of the slag, is eliminated from the process. This improves thermodynamic efficiency.

• The moisture in the PCI, often the largest source of H2O in the feed, absorbs heat and consumes some of the C in the process to produce H2 + CO, is eliminated from the process. This improves thermodynamic efficiency.

• Any Sulphur (S) or Phosphorous (P) in the PCI, both of which have deleterious quality effects on the steel produced, are also eliminated. This improves the quality of the steel produced and eliminates the need for pretreatment to remove S & P between the BF and the Basic Oxygen Furnace (BOF). This helps reduce costs and offset additional energy that may otherwise be required.

• Table 1 lists the densities of the various solid inputs and divides the weight for each to determine the volume of the solids. Eliminating the PCI, due to its much lower density than the iron ore and reducing the limestone by the same tonnage as the ash in the PCI, results in 22% less volume of solid in the BF. This means that 28% more material will fit within the BF. • The volume of gas in the BF reduces by replacing the N2 and Ar in the Air in the blast with the recirculated CO.

• The reduction of both solid and gaseous material in the column has two consequences, both of which will potentially increase the throughput capacity of the BF: o The higher density for the same chemical reactions, combined with the elimination of some of the inert material is likely to increase the pressure and temperature, and if this is the case, it is likely to speed up the chemical processes. o Reducing the volume of the solid material in the column by 22% means that the charge will take up less room, allowing 28% more material to be added. cture and Recycle the TG CO

N2 from the BF

The first step in this process is to eliminate the N2 in the Blast. This can be achieved initially by using a mixture of O2 and CO2 designed to replicate the combustion characteristics of air in the Blast instead of air at startup or during transition. As the O2 and CO2 mixture is only required during startup or transition until there is no N2 in the TGs, the quantity of O2 required is comparatively small.

PIB will have a very large O2 plant, so the source of the O2 for the startup or transition of the BF is not a cost or availability problem. As there are 5 steel smelters proposed for each Steel Park, the Oxygen Plant capacity required for 5 Basic Oxygen Furnaces can be used as each BF is progressively started one at a time. At other mills, it may be necessary to build storage facilities for the O2 to build up the initial amount required to eliminate the N2 from the blast.

The O2 and CO2 mixture used in the Blast at startup or transition with the normal amounts of PCI coal would only be required for a very short period. Once the BF is generating N2 free TGs, the Blast will be progressively swapped to the recycled CO and the PCI injection progressively reduced. This would occur one tuyere at a time.

If there is no N2 in the Blast, then there will be no N2 in the TGs. As N2 is 48.5% of current TGs, this almost halves the quantity of gases in the TGs to be processed, as well as making it easier to separate the two predominant gases, i.e., CO2 and CO that remain.

CO and N2 both have molecular weights of 28, so gravity separation will not be viable.

The boiling point (82 °K) and melting point (68 °K) of CO are very similar to those of N2 (77 °K and 63 °K, respectively), making it difficult to separate these two gases using different temperatures and pressures to control the different states of each gas.

Removing the N2 from the TGs, by not putting it in there in the first place, makes the separation of the CO from the CO2 much easier.

Liquid-Vapour Separation of CO and CO2

Figure 3 illustrates the Pressure/Temperature relationship for the various states of CO2. The “Triple Point” is at -56.6°C and 5.1 bar. The “Supercritical Point” is at 31 .1°C and 73.8 bar.

CO2 is a liquid at any combination of temperature and pressure within the zone in the image labelled “Liquid”. In contrast, CO’s condensation point is - 191 °C. This means that if the CO, CO2 mix is compressed and cooled to form liquid CO2, that at any combination of temperature and pressure in the liquid zone as shown in the image above, that the CO will be present as a gas, while the CO2 will condense as a liquid at the base of the mix.

A possible system to separate the CO from the CO2 is shown in Figure 4. As the mixture is compressed there will be a temperature increase due to the adiabatic heat of compression. Additional heat exchangers could then utilise the adiabatic heat generated from the compression of the mixture, to further heat the CO recovered prior to injection as the Blast.

If the pre-cooled TGs are compressed and injected into the upper levels of the liquid in the base of a pressure vessel, the CO2 will rapidly liquefy, while the CO bubbles up and out of the upper level of the liquid CO2. Alternatively, if injected into the gaseous CO, the CO2 will condense and rain down into the liquid CO2 below. The liquid CO2 can be continuously drained out of the base of the separating pressure vessel and stored for later use.

The CO and other minor constituents in the TGs will remain in their gaseous form, above the liquid CO2. The gaseous CO can then be continuously recovered from the top of the pressure vessel.

Having removed the CO2, the CO can be heated and recycled by injecting into the base of the BF instead of the previously used Air and PCI in the “blast”. The hot (-500°C) TG gas feed to the CO recovery unit will be precooled using a heat exchanger interacting with the cooler CO recovered from the separation plant. An indicative arrangement is shown in Figure 5.

There are minor quantities of H2 (0.26%) in the TGs as shown in Table 2 above. The source of the H? is predominantly from the minor quantities of moisture in the solid feedstocks. This is described as “LOI” for the iron ore and “moisture” in the PCI. There is very little water in the coke. There is also some moisture in the blast air. Most of the moisture is from the PCI. If there is less PCI used, there will be less moisture introduced into the BF.

This small quantity of water introduced with the charge, PCI and air, is rapidly heated to a steam, that then reacts with the hot C to form CO and H2. As most of the water comes from the moisture in the PCI coal and air, the quantities of H2 will reduce significantly when the use of PCI and air in the Blast reduces.

Any small quantities of H2 remaining will be captured with the CO and recycled through the BF. The formation of H2O is minimised provided there is a surplus of hot C in the charge to split any H2O that does form.

If there is any tendency for the H2 to accumulate in the process, this can be offset by preheating the iron ore to remove moisture, which will also help with improving BF efficiency. This can be done in a rotating trommel using waste heat from the BF process. Any accumulations of H2O could be removed as water, which due to liquid H2O having a lower density than liquid CO2, would float on the CO2 within the TG separation pressure vessel. The water can then be removed through a third “Tap” in the pressure vessel, just above the liquid CO2 level.

The CO recovered from the TGs will pass through a heat exchanger with the TGs that feed the CO recovery plant. This will heat up the CO prior to being injected in the Blast, making the BF process slightly more efficient. It will also cool down the TGs on their way to the CO recovery unit.

Further heating of the CO can be achieved with additional heat exchangers established around the base of the BF or from the cooling systems used to reduce the adiabatic heat of compression n the CO/CO2 separator.

It is considered that this can all be done relatively cheaply, compared to the savings in coal purchases envisaged, in the bulk quantities expected.

Amine Separation of CO and CO2

By way of example, the scale of BlueScope’s lllawarra Steelworks is employed below to outline the potential economic outcomes arising from using Amines to separate CO and CO?. The below discussion also introduces involvement of potential large scale consumers of CO2, as using the CO2 is better than sequestration or paying Carbon Credits.

The following discussion assumes that:

• lllawarra Steelworks produce 2.6 MTPA of steel;

• Half the energy can come from co-generation and the other half produced by gas turbines burning coke oven gas at a cost of $A0.10 per kWh;

• lllawarra Steelworks consume 1 tonne of coal per tonne of steel and that 25% of the coal used is PCI, or 2.6 * 0.25 = 650,000 tpa; and

• The cost of PCI coal is $A250 per tonne.

The Blast Furnace (BF) TG is typically produced at a rate of 2.5-3.5 tonnes/tonne of liquid steel (tls) and typically contains 20-30% CO, 20-25% CO2 and 2-6% H2 by volume, with the remaining 48.5% being Nitrogen (N2) (Huth and Heilos, 2013).

Table 3 below uses the midpoints of quantity and concentration from the above to approximate TG composition and quantity. With 3 tonnes of TG per tonne of liquid steel and 2.6 MTPA of liquid steel, the TG totals 7.8 MTPA.

Table 3 lists the proportions by volume, and by using the molecular weight of each gas, determines the relative proportions by weight and actual weight of each of the Top Gases produced per year.

Table 3: Top Gases Produced per Year

Table 3 shows 2.53 tonnes of CO2 contained in the TGs, which constitute 0.97 tonnes of CO2 per tonne of steel, whereas most sources quote CO2 emitted per tonne of steel as being, for BFs, typically in the range of 2.0 to 2.3.

It is understood here that the total emission figures normally quoted include the CO in the TGs after they are burnt to produce energy within the steelworks, or alternatively, the CO2 generated in producing electricity for the steelworks. Adding the CO after it has been burnt adds 2.8 MTPA of CO2, which when added to the 2.53 MTPA of CO2 in Table 1 results in 5.33 MTPA, which is a ratio of 2.05 tonnes of CO2 per tonne of steel produced.

With 48.5% of the TG being N2, and N2 being 79% of air, and the total quantity of TG being 7.8 MTPA, the air injected can be calculated to be 0.485 * 7.8 / 79% Nitrogen proportion in air = 4.8 MTPA of air in the blast.

With 20% of the air in the blast being Oxygen, the Oxygen content of this much air would be 0.96 MTPA. Each tonne of O2 will make (32 (O2 AW) + 2 * 12 (equivalent C AW)) = 56 / 32 = 1.75 tonnes of CO. 0.96 MTPA of O2 will therefore make 1 .68 MTPA of CO. This CO will include 12 / 28 = 42.85% Carbon.

42.85% of 1 .68 MTPA CO = 0.72 MTPA C. This is very similar to the C in the 650,000 tonnes of PCI that it is assumed BlueScope is already using.

The Oxygen content in the 1.68 MTPA of CO would be 16 / 28 * 1.68 = 0.96 MTPA. This neatly equates to the Oxygen content in the air used in the blast. The Carbon is just a little more than that injected as PCI coal. This means that recycling the CO from the TGs could eliminate the need to inject air and PCI into the base of the BF. The CO neatly replaces the O2, and the slight increase in C compared to that currently consumed means after combustion to create heat, that more CO is available in the hearth area of the BF to reduce the Iron Ore.

The slightly higher quantity of Carbon in the CO means that the reactions in the base of the blast furnace may be a little hotter and hence faster.

Replacing the PCI with the recycled CO from the Top Gases would reduce Carbon consumption and hence CO2 emissions by 25%.

As the quantities of Oxygen and the Carbon in the PCI in the Blast are not very different to that in the recycled CO, there is not expected to be much of a change in the composition of the TGs.

As it is intended to replace the PCI rather than the Coke and remove the N2 from the blast, BF efficiencies are likely to increase.

There will obviously be efficiency changes within the BF, as the equilibrium temperature points within the BF change. When it is considered that a lot of the CO in the TGs is generated in the upper sections of the BF, where the lower temperatures do not utilise it to the same extent as would occur if injected at the base, it appears that BF efficiencies could improve significantly. This arises because the upper levels of the BF could effectively be used to generate CO from the Coke, to be captured and recycled. This concept is similar to the “breeder reactors” used in Nuclear Power Stations which can manufacture their own fuel.

Another consequence is that 2.6 MTPA of coal will probably be comprised of 0.65 MTPA of PCI and that the remaining 1 .95 MTPA of coking coal will make say 1.56 MTPA of coke, assuming that the coal used has an average of 20% volatiles. As the coke density is about 1 , this will take up 1 .56 Mm³. As the PCI density is typically 1 .4, this would take up about 0.5 Mm³. If the iron ore is 60% Fe, there will be 4.3 MTPA of Iron Ore. As the Iron Ore has a density of 4.5, the volume of the Iron Ore will be 0.96 Mm³.

With 0.7 MTPA of limestone and a typical density of 2.1 , there will be 0.3 Mm³ of limestone used per annum. This means that the total processed volume per annum will be 3.72 Mm³, of which the PCI will be 0.5 Mm³, or about 14% by volume.

The BF throughput is a function of both temperature and volume. If the PCI comprises 14% of the volume and is removed, the volume of other solid materials flowing through the BF would be 1 / 86% = 1.16.

When this is combined with the higher temperatures expected from eliminating N2 from the BF and slightly increasing the carbon ratio in the blast, a production throughput improvement for the same sized BF of 16% appears to be possible.

At lllawarra steelworks, this would lift the annual capacity from 2.6 MTPA to ~ 3 MTPA. Eliminating the PCI also removes the moisture. Sulphur and Phosphates that they carry. This further improves BF efficiencies and also eliminates the pre-treatment processes to remove Sulphur and Phosphorous between the BF and the Basic Oxygen Furnace (BOF).

Removing these processes reduces the re-heating requirements applied between the BF and the BOF, further improving overall efficiencies.

Regarding the CO2 produced, it is noted that in the manufacture of Urea using cold plasma technology, one of the feed stocks required is CO2.

Australia consumes about 3 MTPA of Urea and it is understood that certain operators require 733 kgs of CO2 to produce 1 tonne of Urea. 3 MTPA of Urea will require 2.2 MTPA of CO2 or most, (87%), of the 2.53 MTPA of CO2 in the TGs.

Recycling the CO recovered from the Top Gases, could save 650,000 tonnes of coal * $A250 per tonne = $A162.5M per annum.

With 1 kWh = 3.6 MJ, projections of Finn Andrew Tobiesen et al, Modelling of Blast Furnace CO2 Capture Using Amine Absorbents, Ind. Eng. Chem. Res. 2007, 46, 7811-7819, would require 2.53 MTPA * 1.1 MJ/kg * $0.10 I 3.6 MJ/kWh = $77.3M.

It is envisaged that this embodiment of the invention may involve:

• Building the separator.

• Using the separator to obtain CO2.

• Making a mixture of CO2 and Oxygen to replicate the combustibility characteristics of air and use this mixture to replace the air in the blast.

• Running this mixture with the PCI coal until the top gases are Nitrogen free. This is expected to take - 2 hours and will mean that the top gases become predominantly just CO2 and CO.

• Recovering the CO from the Amine absorption separator and replacing the PCI and the CO2/O2 mix with just the CO.

If the CO2 was sold to a Urea producer for $A75/t, this would generate 2.53 MTPA * $75/t = $190M per annum. If BlueScope can replace all of its PCI purchases, it will save an estimated 650,000 tonnes * $A250/t = $162.5M per annum. The combined savings of $A352.5M per annum far exceeds the projected cost of $77.3M, with a healthy margin of $A275.2M per annum to cover the capital cost and a substantial profit.

Generating CO from Coke Oven Gases

A second source of CO within the Steel Park can come from the Coke Ovens.

Coke Ovens work by heating Coking Coal in the absence of air, then rapidly quenching the hot coke with a liquid. The rapid quenching causes a “glass” like structure that is strong enough to hold its form in the BF, necessary to create porosity and permeability for the gases in the BF to rise through the column. The quenching liquid currently used is water (H2O).

When water is used, there are several main chemical reactions. The predominant reaction is:

C + H2O = CO + H₂

This is also known as “Town Gas” or “water gas” and was used to supply gas to Newcastle for instance, prior to the conversion to Natural Gas in the 1970s.

Small quantities of CH4 can also be formed from the pyrolysis of the hydrogen in the coal as it is heated, prior to the quenching. Some CO2 formed subsequently reacts with the hot C in the coke to make CO, but the list below shows that 4% of CO2 is a typical outcome in the Coke Oven Gases.

When water is used to quench the coke, Coke Oven Gas (COG) typically contains (by volume), 60% hydrogen (H₂), 24% methane (CPU), 6% Carbon Monoxide (CO), 6% Nitrogen (N₂), and 4% Carbon Dioxide (CO2) (Yang et al., 2014).

The low proportion of CO in the Coke Oven Gas makes this less attractive as a source for CO compared to the quantities contained in the TGs. The H₂ and CH4 are very valuable for other purposes such as feed stock for fertilisers, explosives and Urea/ AdBlue manufacture and to burn to produce electricity.

The small proportions of N₂ and CO2 in the Coke Oven Gas do not prevent this gas from being used as a fuel to generate heat and/or electricity.

The Mass Balance calculations of Table 1 show that the capture and recovery of CO from the BF TGs, will produce an estimated 18.18 MTPA of liquid CO2.

Liquid CO2 is a versatile industrial material, used, for example, as an inert gas in welding and fire extinguishers, as a pressurizing gas in air guns and oil well recovery, as a supercritical fluid solvent in decaffeination of coffee, supercritical drying and meat packaging. It is also added to carbonated beverages for effervescence.

While it is not proposed here that the quality of the CO2 will meet food grade standards without further refinement, the scale involved may justify further refinement that could make this possible.

As the CO2 will be produced as a necessary byproduct of steel manufacture, it can be sold much cheaper than CO2 specifically made for industrial purposes.

Bulk applications such as for pressurising oil wells to improve recoveries could be serviced by transporting in existing LNG ships in a backhaul arrangement where the ship delivers CO2 to the oil/gas well and carries the LNG produced away.

Use of liquid CO2 to Quench Coke

It is possible that a lot of the CO2 could be compressed to a liquid and sold for commercial benefit. However, there is likely to be a significant surplus of CO2 remaining after sales and “giving away”.

Whilst the surplus could be sequestered in abandoned oil and gas wells as is currently proposed in Bass Strait, this would incur a cost. Whilst the cost to do so is likely to be less than the cost of purchasing Carbon Credits, it is proposed here that at least some of the liquid CO2 be used to quench some of the coke instead of water. Alternatively, gaseous CO2 could be pumped through the Coke Oven beds prior to quenching, which due to the Boudouard Reaction would convert most of this CO2 to CO.

If liquid CO2 is used to quench the coke, the pressure reduction will cool the CO2 due to the reverse of the adiabatic heat of compression, cooling whatever it touches and reacting with the hot coke. This adiabatic expansion should result in more rapid cooling than is possible using water, and hence a potentially stronger “glass” structure in the coke.

Designing the process to convert significant quantities of the CO2 to CO and collect a rich CO stream will require substantial development of the technology which will be required to show that the critical coke strength and reactivity properties are not impacted. The predominant chemical equation becomes:

C + CO₂ - 2CO

It is possible that the liquid CO2 can be dropped onto the hot coke as occurs with water quenching, but if pumped as either a liquid or gas into the base of the coke bed from below, and allowed to decompress, the very cold gas resulting could achieve a better outcome. Liquid CO2 has a higher density than water and has a higher specific heat, both of which suggest it will be more effective than water as a quenching agent.

Without the H2 contained in the quenching water, CH4 or H2 products are not possible other than as trace gases from moisture within the Coking Coal or the minor quantities of H2 within the coal itself.

This use of CO2 is economically viable because of the 1^st Stage or core of the present invention, where the liquid or gaseous CO2 is created as a byproduct from the recovery of the CO from the TGs.

There are many advantages and opportunities arising from using the CO2 to make CO at the same time as making the Coke:

® If for example half of the Coking Coal is replaced with CO, only half as many Coke Ovens are needed. The capital cost saving is probably similar to the capital cost of the TG CO/CO2 separation facility, meaning that there is unlikely to be much capital cost difference between this proposed process and that currently used. This of course only applies to new projects such as PIB. Older operations would need the coke ovens to be converted, or new coke ovens constructed and designed to use liquid or gaseous CO2 to produce more CO.

« If water is used to quench the coke, the Steel Park will need to consume more water. Using the liquid CO2 to quench half the Coke therefore reduces the water consumption required for the quenching of the coke by half.

® Replacing half the Coke also reduces the volume of solids within the charge, facilitating further throughput capacity increases for the same sized BF.

® It is much easier to preheat CO through heat exchanges around the base of the BF than it is to preheat solid material in the charge. This further increases the efficiency of the BF.

It is important to reiterate that it is not proposed to replace all of the coke in the Charge.

The coke is needed in the BF for other purposes than as a source of C to produce heat and make CO. The coke will, however, no longer be needed as the sole source of heat or to make CO, but we can’t obtain CO from this second source of CO without making coke, and the coke will still be required to achieve permeability within the BF for the free flow of the CO through the Column.

The coke volume in a BF typically accounts for 35%-50% of the total volume of the charge. Coke is relatively strong. As long as it remains blocky above the tuyere area, it will enable the column in the BF to have the good gas permeability necessary for the free movement of the CO so it can react with the iron oxides in the iron ore.

Less gas permeability is required if the CO in the TGs are captured and recycled in the Blast, because there will be no N2 in the Blast. Consequently, only about half the volume of gases need to be able to flow through the Column. A likely consequence of the higher density of the Column is that the CO will need to be pumped in at a higher pressure, but this is also expected to accelerate the chemical reactions as the additional adiabatic heat of compression will also improve efficiencies.

Table 1 shows that there is 18.18 MTPA of C within the CO2 in the TGs, and 18.48 MTPA of C within the Coke fed into the BF in the Charge. The quantities are very similar.

As there is only half the gas quantity in the TGs, only half the void volume would be required in the BF to retain similar levels of void space for the amount of gas present, that is, only half the porosity is required to deal with half the gases present to facilitate the same rate of gas flow through the Column.

As half the void can be tolerated because the TGs are halved, it is likely that little more than half the coke can be replaced with CO manufactured by using the liquid CO2 sourced from the CO TG recovery process to quench the coke. The best ratio applying can be determined first by simulation modelling and then using an iterative process to fine tune for the best outcomes.

If there is half the void volume however, the average Column density will increase, so a higher pressure will be required for the CO to be injected into the BF to offset the increased density in the Column.

It is envisaged that halving the Coke, using the CO generated in the coke ovens, will save a further 10.5 MTPA of solid low-density material in the Charge, further increasing the average density of the Column.

It is proposed that the CO recovered from the coke oven gases be injected one third to halfway up the BF, most likely in quantities that will result in the C/Fe ratio exceeding the previous amounts used. It is expected that a higher C/Fe ratio will result in higher efficiencies in the BF and any surplus CO in the TGs as a result of the higher proportion can in any case, be recaptured and recycled. Excess C in the BF was not previously considered because of the extra cost of the coal but becomes possible with the capture and recycling of the TGs. There will be an equilibrium balance achieved with each set of feed stock specifications and BF designs. The ability to vary the proportion of CO produced from the coke ovens will help adapt to each set of circumstances.

As the Coke typically produces 60% of the GHGs in the current process, halving the Coke will reduce the GHGs generated by a further 30%, taking the total saving of GHGs after the CO from the TGs recycling is included to 70%.

Whilst the numbers above work on averages, different BFs and feedstocks will require differences in the proportions of CO and Coke used. The flexibility to adjust the ratio can be achieved by having some of the Coke Ovens set up to take either water or CO2 as the quenching liquid.

It remains desirable to use water in at least some of the Coke Oven batteries to produce the H2 that is targeted for fertiliser, explosive and Urea/AdBlue manufacture.

Recycled CO from TGs

The recycling of the CO from the TGs is expected to eliminate the PCI coal and about 1 .4 MTPA of the limestone as it will not be required to flux the ash in the PCI. Table 1 includes a line for the density of each of the solid materials fed into the BF. The weight of the material fed in is then divided by the density to determine the volume.

The ash in the PCI, totalling 1.48 MTPA is removed from the process. As the ash needs to be heated up to become molten slag, it also absorbs heat from the BF process. Removing the ash therefore makes the process more thermodynamically efficient. Removal of this ash facilitates the removal of a corresponding amount of limestone otherwise needed to flux this ash. There is 22% less volume in the solids in the revised feed quantities, meaning that for the same volume BF, that 28% (22/78) more material can be added. This means that the throughput of the BF can increase by 28% due to volume constraints alone.

These changes are expected to result in:

• Hotter average input temperatures.

® Increased average Column density.

® Increased average temperature within the BF.

« Faster chemical reactions

® Improved reaction efficiency.

These factors all work to reduce retention time, thus further contributing to potential throughput capacity increases. An overall 28% increase will lift the 4.4 MTPA capacity of the PIB BFs to 5.6 MTPA.

Figure 6 illustrates the BF operation with the 28% increased process flow quantities expected.

The inert Nitrogen, totalling 58.66 MTPA, plays no role in the BF reactions and is removed from the process. As the N2 needs to be heated up, it absorbs heat from the BF process. Removing the N2 therefore makes the process more thermodynamically efficient. This assists in increasing throughput quantities but has not been taken into account in the above 28% projected throughput increase.

CO Made by Quenching Coke with CO2

As with the removal of the PCI, halving the coke utilised is likely to increase throughput capacity due to a combination of even less solid material required, higher input temperatures and less inert material in the form of Coke ash and the corresponding amount of limestone flux required. Table 1 also includes a mass balance using CO generated by quenching the coke with liquid CO2, however it must be conceded that this is an even further extrapolation than that used to estimate the mass balance for recycling the CO recovered from the TGs.

The mass balance for the CO from the coke ovens was achieved by maintaining the same C in the output and then varying the proportion of CO and CO2 to get the O2 to match. Chemical equations don’t necessarily work this way. There is however an ability to vary the CO input and other controls to adapt the system to optimise the outcome and the approximate mass balance demonstrates the potential for further significant efficiency gains.

A possible arrangement to quench the Coke with liquid CO2 is illustrated in Figure 7.

The combined effect of capturing and recycling the TGs and sourcing additional CO from modified coke ovens reduces the volume of solids in the BF by 40%. If the volume of solids is the predominant constraining factor with regard to BF throughput, then the throughput, compared to the standard 4.4 MTPA per PIB BF capacity would increase by 67% to 7.35 MTPA. Increase Benefits

As detailed above, the combined potential of these proposals is to increase BF throughput capacity by 67%. Figure 8 illustrates a BF with the projected increased process flow quantities.

The increased throughput will reduce costs arising because:

• If all CO in the TGs is captured and recycled and can replace all of the PCI coal totalling 14.78 MTPA. Replacing the PCI with recycled CO avoids the purchase cost of the PCI, which if $250/t may save at PIB’s scale, $3.7B per annum. • If half the coke in the charge is replaced with CO manufactured by using liquid CO2 to quench the coke, there is 10.5 MTPA less coke required, which if $400/t may, at PIB's scale, save $4.2B per annum.

• If the quantity of limestone required for fluxing the ash is reduced in proportion to all of the ash in the PCI and half of the ash in the coke, there is a reduction of 3 MTPA of limestone, which if $40/t may, at PIB’s scale, save $120M per annum.

• If the increased BF throughput due to less feed materials required and increased efficiencies is 67% higher than the original design capacity and the labour cost is 25% of the operating cost, then the operating cost is likely to reduce by 16% due to labour savings alone. This is envisaged to save about $100 per tonne of pig iron produced, or at PIB’s increased capacity of 73.5 MTPA, a total of $7.35B per annum.

Total projected savings are of the order of $15B per annum at PIB’s scale. Any operating cost increases, such as those arising from capturing and separating the TGs, are likely to be offset in part, by the reduced coke oven operating costs. Any extra capital cost incurred is likely to be offset by the reduced capital cost required due to half as many coke ovens being needed.

Steel Quality Benefits

The most deleterious elements in a BF with respect to steel quality are Sulphur and Phosphorous. These are most often removed in a Basic Oxygen Furnace (BOF), using the Linz-Donawitz process developed in Austria. Where contamination is high, Sulphur and Phosphorous are often removed using pretreatment processes, which add cost and take time.

Sulphur and Phosphorous predominantly enter the BF as contaminants in the PCI. They are present in much smaller quantities in the Coke, as the process of making coke generally results in the SOX & POX being released as volatiles. If the SOX, POX and Silicon are high, they are often removed in a pretreatment process in the ladle, before the pig iron is converted to steel. In external desulfurizing pretreatment, a lance is lowered into the molten iron in the ladle and several hundred kilograms of powdered Magnesium are added. The Sulphur impurities are reduced to Magnesium Sulphide in a violent exothermic reaction and then raked off. Similar pretreatments are possible for external desiliconisation and external dephosphorisation using mill scale (iron oxide) and lime as fluxes. The decision to pretreat depends on the quality of the hot metal and the required final quality of the steel.

If the present invention is successful in eliminating the PCI coal, it removes the moisture in this coal as well as the Sulphur and Phosphorous.

There may be very small quantities of Sulphur and Phosphorous present in the Iron Ore and Limestone, but these would be very small quantities in comparison to the much higher levels of these undesirable constituents, typically brought into the BF as part of the PCI.

The Pig Iron needs to be Carbon saturated to about 4%C to keep the liquid iron temperature to a practical 1500°C so the slag is also fluid.

If not all of the PCI can be removed and/or if some Carbon is required at the base of the BF to achieve the 4% C in the Pig Iron, then the smaller amount of PCI otherwise required could be replaced with BioChar, as this has also had the Sulphur and Phosphorous volatilised out.

The BOF that follows the Blast Furnace process will therefore not require pretreatment to remove the Sulphur and Phosphorous. This saves costs as well as allowing the BOF to focus on just the Silicon and Carbon. This could mean less retention time in the BOF, less O2 consumed and higher throughput rates.

The total outcome from the above analysis suggests that:

• The Capex for the project is unlikely to significantly change as less coke ovens and less pretreatment work will be close to offsetting the cost of the TG separation and recovery. • Operating costs for a PIB scale operation due to the reduction in coal purchased and from lower labour costs per tonne of steel produced will be reduced by an expected ~$15B per annum, which provides plenty of benefits to pay for any increased costs incurred with the replacement processes.

• Steel Quality will be improved and or pretreatment costs significantly reduced.

• GHGs may be reduced by 70%.

Throughout this specification, unless the context requires otherwise, the word “comprise”, or variations such as “comprises” or “comprising”, will be understood to imply the inclusion of a stated step or element or integer or group of steps or elements or integers, but not the exclusion of any other step or element or integer or group of steps, elements or integers. Thus, in the context of this specification, the term “comprising” is used in an inclusive sense and thus should be understood as meaning “including principally, but not necessarily solely”.

Unless the context requires otherwise or specifically stated to the contrary, integers, steps or elements of the invention recited herein as singular integers, steps or elements clearly encompass both singular and plural forms of the recited integers, steps or elements.

It will be appreciated that the foregoing description has been given by way of illustrative example of the invention and that all such modifications and variations thereto as would be apparent to persons of skill in the art are deemed to fall within the broad scope and ambit of the invention as herein set forth.

Claims

1 . A blast furnace smelter operation comprising: a blast furnace comprising an upper raw material feed, an intermediate reaction zone and a hearth having an iron product outlet and a slag outlet; a top gas outlet for removal of a CO/CO2 containing top gas from said blast furnace; a CO/CO2 separator for separating CO gas from said CO/CO2 containing top gas; and at least one tuyere for injection of separated CO gas back into said blast furnace.

2. A blast furnace smelter operation as claimed in claim 1 , wherein a feed for said blast furnace inlet is N2 free and comprises said separated CO gas, optionally further comprising PCI and/or BioChar, preferably BioChar.

3. A blast furnace smelter operation as claimed in claim 1 or 2, wherein said top gas is N2 free and CO gas separation comprises a vapour-liquid separation.

4. A blast furnace smelter operation as claimed in claim 3, wherein said CO/CO2 separator comprises a pressure vessel that condenses CO2 gas to CO2 liquid facilitating separation of CO gas from the CO/CO2 containing top gas.

5. A blast furnace smelter operation as claimed in claim 3 or 4, wherein said CO/CO2 containing top gas is pre-cooled before vapour-liquid separation, preferably in a heat exchanger.

6. A blast furnace smelter operation as claimed in claim 1 or 2, wherein said top gas is N2 free and CO gas separation comprises an amine absorption separation.

7. A blast furnace smelter operation as claimed in claim 6, wherein said amine absorption separation comprises feeding said CO/CO2 containing top gas to an absorber containing an amine solvent to absorb CO2, recovering CO gas from the absorber, and feeding CO2 laden amine solvent to a desorber to recover CO2 gas.

8. A blast furnace smelter operation as claimed in claim 7, comprising solvent cooling to 40 °C to facilitate CO2 absorbance and reboiling the CO2 laden amine solvent in a reboiler to a temperature of about 120 °C prior to desorbing.

9. A blast furnace smelter operation as claimed in claim 7 or 8, wherein said amine solvent is selected from monoethanolamine (MEA), methyldiethanolamine (MDEA)Zpiperazine, and 2-amino-2-methyl-1 - propanol (AMP).

10. A blast furnace smelter operation as claimed in any one of claims 3 to 8, wherein CO2 recovered from the CO/CO2 separation is collected for external applications.

11. A blast furnace smelter operation as claimed in any one of claims 3 to 10, wherein CO2 liquid recovered from the CO/CO2 separation is used, at least in part, for coke quenching in a coke oven of said blast furnace smelter operation, or wherein CO2 gas recovered from the CO/CO2 is used prior to quenching to obtain more CO, prior to feeding coke to said raw material feed of said blast furnace.

12. A blast furnace smelter operation as claimed in claim 11 , wherein the CO2 liquid is fed to said coke oven by expansion into a base of said coke oven or is dropped from above onto the coke, or whereinthe CO2 gas is pumped into the base of the coke oven and converted to CO and collected from a top of the coke oven, prior to quenching

13. A blast furnace smelter operation as claimed in claims 11 or 12, wherein CO gas formed from said coke quenching is fed back into said blast furnace, preferably through at least one tuyere disposed at a location from one third to two thirds up the blast furnace.

14. A blast furnace smelter operation as claimed in any one of the preceding claims, wherein before injection back into said blast furnace, said separated CO gas is re-heated in a heat exchanger with the CO/CO2 containing top gas removed from the blast furnace.

15. A blast furnace smelter operation as claimed in any one of the preceding claims, wherein iron ore fed to said raw material feed is preheated to reduce or remove water content.

16. A blast furnace smelter operation as claimed in any one of the preceding claims, wherein said at least one tuyere for injection of separated CO gas back into said blast furnace is located at an upper region above said hearth.

17. A blast furnace smelter operation as claimed in any one of the preceding claims, wherein a mixture of O2 and CO2 gases in a proportion designed to replicate the combustion characteristics of air is fed to said blast furnace during startup or transition of said blast furnace smelter operation.

18. A method of operating a blast furnace comprising: capturing a CO/CO2 containing top gas from said blast furnace; separating CO gas from said CO/CO2 containing top gas; and injecting separated CO gas back into said blast furnace.

19. A method of retrofitting a blast furnace comprising: providing a CO/CO2 separator for separating CO gas from CO/CO2 containing top gas collected from said blast furnace; and providing a recirculation loop for recirculating separated CO gas to said blast furnace.

20. A method as claimed in claim 18 or 19, wherein a feed for said blast furnace containing said separated CO gas is N2 free, and optionally further comprises PCI and/or BioChar, preferably BioChar.

21 . A method as claimed in any one of claims 18 to 20, wherein said top gas is N2 free and CO gas separation comprises a vapour-liquid separation.

22. A method as claimed in claim 21 , comprising condensing CO2 gas to CO2 liquid facilitating separation of CO gas from the CO/CO2 containing top gas.

23. A method as claimed in claim 21 or 22, comprising cooling said CO/CO2 containing top gas before vapour-liquid separation, preferably in a heat exchanger.

24. A method as claimed in claim 18 or 19, wherein said top gas is N2 free and CO gas separation comprises amine absorption separation.

25. A method as claimed in claim 24, wherein said amine absorption separation comprises feeding said CO/CO2 containing top gas to an absorber containing an amine solvent to absorb CO2, recovering CO gas from the absorber, and feeding CO2 laden amine solvent to a desorber to recover CO2 gas.

26. A method as claimed in claim 25, the method further comprising solvent cooling to 40°C to facilitate CO2 absorbance and reboiling the CO2 laden amine solvent in a reboiler to a temperature of about 120°C prior to desorbing.

27. A method as claimed in claim 24 or 25, wherein said amine solvent is selected from monoethanolamine (MEA), methyldiethanolamine (MDEA)/piperazine, and 2-amino-2-methyl-1 -propanol (AMP).

28. A method as claimed in any one of claims 21 to 27, comprising recovering CO2 from the CO/CO2 separation for external applications.

29. A method as claimed in any one of claims 21 to 28, comprising using CO2 liquid recovered from the CO/CO2 separation, at least in part, for coke quenching in a coke oven of said blast furnace smelter operation, and/or using CO2 gas recovered from the CO/CO2 prior to quenching to obtain more CO, prior to feeding coke to said blast furnace.

30. A method as claimed in claim 29, comprising feeding said CO2 liquid to said coke oven by expansion into a base of said coke oven or dropping said CO2 liquid from above onto the coke, and/or pumping the CO2 gas into the base of the coke oven and converting to CO and collecting the CO from a top of the coke oven, prior to quenching.

31. A method as claimed in claim 29 or 30, comprising feeding CO gas formed from said coke quenching back into said blast furnace, preferably through at least one tuyere disposed at a location from one third to two thirds up the blast furnace.

32. A method as claimed in any one of claims 18 to 31 , comprising reheating separated CO gas in a heat exchanger with the CO/CO2 containing top gas removed from the blast furnace before injection of the separated CO gas back into said blast furnace.

33. A method as claimed in any one of claims 18 to 32, comprising preheating an iron ore feed to reduce of remove water content before being fed into said blast furnace.

34. A method as claimed in any one of claims 18 to 33, wherein said separated CO is injected back into said blast furnace at an upper region above a hearth of said blast furnace.

35. A method as claimed in any one of claims 18 to 34, comprising feeding a mixture of O2 and CO2 gases in a proportion designed to replicate the combustion characteristics of air to said blast furnace during startup or transition of said blast furnace.

36. A method of reducing PCI use and GHG generation in a blast furnace comprising: capturing a CO/CO2 containing top gas from said blast furnace; separating CO gas from said CO/CO2 containing top gas; and injecting separated CO gas back into said blast furnace.

37. Use of CO gas separated from a CO/CO2 containing top gas of a blast furnace as a reductant feed in said blast furnace, thereby reducing PCI use and GHG generation in said blast furnace.