TECHNICAL FIELD

The present disclosure relates to direct reduced iron pellets and uses of such pellets. More specifically, the disclosure relates to direct reduced iron pellets and uses of such pellets as defined in the introductory parts of the independent claims.

BACKGROUND ART

Steel is the world's most important engineering and construction material. It is difficult to find any object in the modern world that does not contain steel, or depend on steel for its manufacture and/or transport. In this manner, steel is intricately involved in almost every aspect of our modern lives. In 2018, the total global production of crude steel was 1810 million tonnes, by far exceeding any other metal, and is expected to reach 2800 million tonnes in 2050 of which 50% is expected to originate from virgin iron sources.

Direct reduction is an increasingly prevalent means of processing iron ore to produce the crude iron required for steelmaking. In direct reduction, the ore is reduced in a solid-state reduction process at temperatures below the melting point of iron. Shaft-based direct reduction processes utilize pelletized iron ore as the feedstock and produce a porous crude iron product known as sponge iron or direct reduced iron (DRI).

Most present shaft-based direct reduction plants are part of integrated steel mills and the DRI produced is utilized directly on-site in steelmaking. However, some DRI is also transported to remote steel mills or sold to third parties, and such DRI must be capable of being easily handled, transported and stored. As production of DRI becomes more prevalent, such commodity use of DRI is expected to increase. For example, the iron ore producer LKAB recently announced a strategy whereby they will switch from ore pellet production to hydrogen-based sponge iron production by the 2030s. The cold DRI pellets (cDRI) produced by shaft-based direct reduction plants is not typically well suited for such purposes. Due to its high porosity, low density, large surface area and low thermal conductivity, it has a propensity to undergo rapid corrosion and reoxidation reactions. Many of these reactions are exothermic, leading to self-heating and eventually self-ignition and fires if not controlled. Corrosion and oxidation reactions of DRI can also produce hydrogen, an explosive gas which is lighter than air, and carbon monoxide, a highly toxic gas. These problems are compounded by the fact that DRI is typically relatively weak and tends to break down during handling to produce dust and fines. DRI dust tends to be even more reactive than the bulk DRI and has a high propensity to self-heat and cause fires. For example, DRI dust that is dispersed in air can ignite in a flash fire or explosion.

Hot briquetted iron (HBI) was developed in response to the difficulties in shipping and handling cDRI. HBI is produced by compressing DRI to briquettes at high temperature (>650 °C). The compaction of DRI into a dense briquette increases its strength and decreases its reactive surface area, meaning that HBI has a much lower reactivity, and therefore is safer and more practical to handle and ship than cDRI.

Further relevant considerations for the handling, transport and storage of cDRI are detailed in the report "Direct Reduced Iron (DRI): Guide to Shipping, Handling and Storage (April 2022)" from the International Iron Metallics Association. This report states i.a. that DRI is relatively weak in comparison with iron oxide pellets and many other common bulk materials, and tends to break down during handling to produce dust and fines. Moreover, the report states that carbon content (in the form of iron carbide/cementite/Fe3C) decreases reactivity of DRI.

Kim and Pistorius (Kim G, Pistorius P.C., "Strength of Direct Reduced Iron Following Gas-Based Reduction and Carburization", Metallurgical and Materials Transactions B, 2020, volume 51, pages 2628-2641) describe a study of the effects of reducing gas composition, extent of reduction and carburization degree on the compressive strength of DRI pellets. Various industrial and laboratory-reduced DRI pellets were tested. Carbon monoxide in the reducing gas was found to contribute to pellet strength development, possibly by formation of "internal whiskers" in the DRI. DRI that was reduced using hydrogen only (or a mix of hydrogen and steam) was found to have poor strength, and the authors conclude that the low strength of pellets produced without CO might imply that weaker DRI might be an inherent feature of fossil-free ironmaking processes that would use H2-based direct reduction.

There remains a need for direct reduced iron products suitable for handling, transport and storage.

SUMMARY OF THE INVENTION

The inventors of the present invention have identified a number of shortcomings with prior art means of producing DRI that is suitable to handle, transport and store. As described above, traditional cold DRI is not particularly amenable to such purposes, and extensive precautions must be taken when shipping such a product. Moreover, the cold DRI typically requires passivation in a controlled atmosphere for a number of days post-production in order to decrease reactivity to a manageable extent, further adding to the expense of the process. Briquetting of DRI to produce HBI effectively addresses the reactivity problems, but at the cost of adding additional steps to the manufacturing process, resulting in additional expense.

It would be advantageous to achieve a means of overcoming, or at least alleviating, at least some of the above mentioned shortcomings. In particular, it would be desirable to provide a DRI product that is readily amenable to handling, transport and storage, without needing to resort to the additional expense of briquetting the DRI. Moreover, it would be desirable if such a DRI product could be obtained by a process that is readily amenable to continuous large- scale production and is environmentally more benign. In order to better address one or more of these concerns, a direct reduced iron product having the features defined in the independent claims is provided.

According to a first aspect there is provided direct reduced iron pellets, wherein the DRI pellets have an average metallization of greater than or equal to 97%. The DRI pellets are either essentially free of carbon, or the DRI pellets comprise less than or equal to 2 wt% carbon. The DRI pellets are further characterised in that they have a median pore diameter of greater than or equal to 1.5 pm; and/or (ii). have an average BET surface area of less than or equal to 0.5 m ² /g; and/or

(iii). have an average porosity of less than or equal to 60%.

That is to say that the DRI pellets can be further characterised by any of (i); (ii); (iii); (i) and (ii); (i) and (iii); (ii) and (iii); and (i), (ii) and (iii). Preferably, the DRI pellets are characterised by at least (i), i.e. (i); (i) and (ii), (i) and (iii); or (i), (ii) and (iii).

Such DRI pellets may be obtained by using hydrogen as the reducing gas in the industrial direct reduction process. It has surprisingly been found that DRI pellets meeting the above specification demonstrate superior mechanical and ageing (reactivity) properties as compared to traditional DRI pellets produced using fossil-based direct reduction. More specifically, such DRI pellets demonstrate better cold compression strength, better tumbling index, slower ambient ageing and slower accelerated aging in water as compared to traditional DRI produced using fossil-based reducing gases. Such DRI pellets may possess low reactivity already upon discharge from the DR shaft, and may not necessarily require any further specific passivation procedure.

According to some embodiments, the DRI pellets may have an average metallization of greater than or equal to 98%, such as an average metallization of greater than or equal to 99%, such as an average metallization of greater than or equal to 99.5%.

According to some embodiments, the DRI pellets may have a median pore diameter of greater than or equal to 2.0 pm, such as greater than or equal to 2.5 pm. Greater median pore diameter has been found to provide DRI pellets with decreased reactivity (i.e. superior ageing properties) and improved strength (i.e. superior mechanical properties).

According to some embodiments, the DRI pellets may have average BET surface area of less than or equal to 0.4 m ² /g. A low surface area is expected to correlate to reduced reactivity of the DRI pellet.

According to some embodiments, the DRI pellets may have an average porosity of less than or equal to 58%. Low porosity is expected to correlate to reduced reactivity of the DRI pellet. This also serves to further distinguish the DRI pellets from pellets produced in laboratory scale that are not amenable to large-scale production and do not necessarily possess the same beneficial attributes. The DRI pellets may have an average porosity of less than or equal to 56%.

According to some embodiments, the DRI pellets may have an average total iron content of greater than or equal to 94 wt%, such as greater than or equal to 96 wt%, such as greater than or equal to 98 wt%. Use of such low-residual DRI allows for the production of low-residual steels such as exposed auto sheets, whilst still providing great leeway for use of higher- residual scrap in the melt.

According to some embodiments, the DRI pellets may comprise on average less than or equal to 3 wt% FeO. Low wustite content has been found to correlate with excellent mechanical properties, specifically high DRI compressive strength. The DRI pellets may comprise on average less than or equal to 2 wt% FeO, such as less than or equal to 1 wt% FeO, such as less than or equal to 0.5 wt% FeO.

According to some embodiments, the DRI pellets may comprise on average less than or equal to 0.5 wt% FesO4. Low magnetite content has been found to correlate with excellent mechanical properties, specifically high DRI compressive strength.

According to some embodiments, the DRI pellets may be obtainable by direct reduction in a countercurrent flow direct reduction shaft, in a reducing gas comprising hydrogen greater than 90 vol% hydrogen, and optionally steam and inert gas. The reducing gas may consist essentially of hydrogen, and optionally steam and inert gas. It has been found that direct reduction in hydrogen under appropriate conditions provides DRI with superior mechanical and ageing (reactivity) properties as compared to DRI produced using a fossil-based reducing gas such as natural gas or syngas.

According to some embodiments, the reducing gas may have a temperature of greater than or equal to 750 °C at a reducing gas inlet of the direct reduction shaft. It has been found that higher reducing gas temperatures assist in providing appropriate conditions for the production of the superior highly metallized DRI. The reducing gas may have a temperature of greater than or equal to 800 °C at a reducing gas inlet of the direct reduction shaft, such as greater than or equal to 850 °C, such as greater than or equal to 900 °C, such as greater than or equal to 950 °C. The DRI pellets may comprise less than or equal to 1.5 wt% carbon, such as less than or equal to 1.0 wt% carbon. Since carburization is not an integral part of the direct reduction process, the carbon content may be controlled independently of other properties such as metallization. This is an advantage since the carbon in DRI is typically lost during a subsequent melting process, and it may therefore be desirable to provide a DRI containing only the carbon strictly required for subsequent processing steps.

According to some embodiments, in the cases where the DRI pellets comprise carbon, such pellets may be obtainable by carburization in a carburizing gas subsequent to direct reduction. It has been found that performing carburization subsequent to reduction in hydrogen is not detrimental to the mechanical and ageing properties of the DRI, in contrast to performing simultaneous reduction and carburization in a carburizing gas (i.e. traditional fossil-based direct reduction). In some cases a carbon-containing DRI may be desirable, for example as a drop-in replacement for DRI produced by traditional fossil-based direct reduction.

The carburizing gas may comprise or consist essentially of a gas selected from methane, ethane, propane, butane, carbon monoxide, hydrogen, nitrogen and combinations thereof, with the proviso that it comprises at least 5 vol% of a carbonaceous component, such as at least 10 vol%, such as at least 20 vol%, such as at least 30 vol%.

According to some embodiments, the DRI pellets may have an average cold compression strength of greater than 160 daN as measured by the method of ISO 4700:2015.

According to some embodiments, the DRI pellets may have a tumbling index of greater than or equal to 96% as measured by the method ISO 3271:2015. The tumbling index may be greater than or equal to 97%, such as greater than or equal to 98%.

According to some embodiments, the DRI pellets may have a loss of metallization of less than 1% upon storage for 28 days sheltered from precipitation at ambient temperature. The DRI pellets may have a loss of metallization of less than 1% upon such storage.

According to another aspect there is provided use of DRI pellets according to the first aspect as a feedstock in a melting furnace for the production of steel. The DRI pellets may not be briquetted prior to such use. The melting furnace may be located at a distance of at least 100 kilometres from the location of production of the DRI pellets. Since the DRI according to the first aspect has superior mechanical and ageing properties as compared to traditional DRI pellets (type (B) DRI), it is eminently transportable without the need for prior briquetting to HBI (type (A) DRI). The melting furnace may be located at a distance of at least 500 kilometres from the location of production of the DRI pellets, such as at least 1000 kilometers. The DRI pellets may be stored for a duration of at least 30 days prior to feeding to the melting furnace. Since the DRI according to the first aspect has superior mechanical and ageing properties as compared to traditional DRI pellets (type (B) DRI), it is eminently storable and is convenient to handle without the need for prior briquetting to HBI (type (A) DRI). The DRI pellets may be stored for a duration of at least 60 days prior to feeding to the melting furnace, such as a duration of at least 90 days, such as a duration of at least 120 days.

Further objects, advantages and novel features of the present invention will become apparent to one skilled in the art from the following detailed description.

BRIEF DESCRIPTION OF THE DRAWINGS

For a fuller understanding of the present invention and further objects and advantages of it, the detailed description set out below should be read together with the accompanying drawings, in which the same reference notations denote similar items in the various diagrams, and in which:

Fig 1 is a chart illustrating the metallization values, CCS values and carbon content (x 10) of various exemplifying samples;

Fig. 2a illustrates the microstructure of H2-reduced pellets;

Fig. 2b illustrates the microstructure of natural gas-reduced pellets;

Fig. 3 is a chart illustrating the results of tumbling tests determined using the methods of ISO 3271:2015 for a number of exemplifying samples; and

Fig. 4 is a chart illustrating the change in weight of various sample pellets upon aging under various conditions. DETAILED DESCRIPTION

The present invention is based upon the surprising discovery that highly metallized DRI pellets produced using hydrogen as reducing gas by continuous shaft-based direct reduction on an industrial scale have superior attributes that make such pellets highly suitable for storage, handling and transport. These superior attributes are improved mechanical strength and improved resistance to aging as compared to DRI pellets produced using traditional fossilbased reducing gases. This is in contrast to received wisdom whereby carbon incorporated in the DRI during reduction is considered to improve the strength and ageing of the DRI.

The present disclosure will now be described with reference to performed experiments in which preferred example embodiments of the disclosure are described. The disclosure may, however, be embodied in other forms and should not be construed as limited to the herein disclosed embodiments. The disclosed embodiments are provided to fully convey the scope of the disclosure to the skilled person.

General

Unless otherwise stated, the properties of the various DRIs tested are determined using standard methods known in the art. Where several methods are in conventional use for determining a single property, variations in the determined property are typically within the limits of experimental error.

Metallization is defined in a manner conventional within the art as (Fe _m etaiiic / Fe _to tai) x 100. Metallization was determined using X-ray diffractometry (XRD), but may also be determined using other methods. Such other methods include:

ISO 2597-1:2006 (Iron ores — Determination of total iron content — Part 1: Titrimetric method after tin(ll) chloride reduction) in combination with ISO 5416:2006 (Direct reduced iron — Determination of metallic iron — Bromine-methanol titrimetric method); and

ISO 10276-1:2000 (Chemical analysis of ferrous materials — Determination of oxygen in steel and iron Part 1: Sampling and preparation of steel samples for oxygen determination) in combination with ISO 10276-2:2003 (Chemical analysis of ferrous materials — Determination of oxygen content in steel and iron — Part 2: Infrared method after fusion under inert gas). Composition of the tested DRIs, such as carbon-content, may be determined using elemental analysis (LECO analysis). Relevant standards for such determination include:

ISO 15350:2010 (Steel and iron — Determination of total carbon and sulfur content — Infrared absorption method after combustion in an induction furnace);

ISO 10036:1989 (Chemical analysis of ferrous materials — Determination of total carbon in steels and irons — Gravimetric method after combustion in a stream of oxygen); and

ISO 9556:2001 (Steel and iron — Determination of total carbon content — Infrared absorption method after combustion in an induction furnace).

Setup

Unless otherwise stated, all DRI samples tested were produced at the Hybrit pilot direct reduction facility in Lulea. In brief, the pilot facility comprises a direct reduction shaft having a total height of approximately 9.3 meters, a widest diameter of approximately 1.22 meters and a total volume of approximately 7.6 cubic meters. Considering only the section of the shaft constituting the reducing zone, this zone has a height of approximately 3.0 meters and a diameter of approximately 0.94 m. The shaft is of a conventional design. That is to say that it is a solid-gas countercurrent moving bed reactor, whereby a burden of iron ore is charged at an inlet at the top of the reactor and descends by gravity towards an outlet arranged at the bottom of the reactor. Commercially available KPRS direct reduction pellets from LKAB were used as the iron ore burden in all studies described herein. However, the same or similar results as described herein may be obtained using any suitable iron ore pellets as the starting material. The DR shaft comprises a reducing zone, an isobaric (transition) zone, and a conical cooling zone tapering towards an outlet of the DR shaft. The shaft has a nominal production capacity of approximately 1 ton DRI/h. The operational pressure in the reactor may be varied up to about 4 barg.

A heated reducing gas may be introduced into the reducing zone in order to reduce the iron ore burden. The reducing gas may for example comprise or consist essentially of hydrogen, carbon monoxide, natural gas, and mixtures thereof. Reducing gas flow may be varied from about 1500 Nm ³ /h to about 3000 Nm ³ /h, and inlet temperature may be varied from about 550 °C to about 1000 °C. A cooling gas may be circulated in the cooling zone in order to cool the DRI after reduction and prior to discharge. Suitable cooling gases include, for example, nitrogen, hydrogen or a combination thereof if a carbon-free DRI is to be produced, or natural gas (diluted as appropriate) if a carbon-containing DRI is to be produced. Cooling gas flow may be varied from about 400 Nm ³ /h to about 1000 Nm ³ /h.

In some cases, no cooling gas is circulated in the cooling zone and the hot DRI is instead discharged to a separate shaft where it is cooled and optionally carburized using a circulating gas. Such a separate shaft arrangement is disclosed in W02021/225500 Al, which is hereby incorporated by reference. Study 1 - Cold Compression Strength of DRI

Cold compression strength (CCS) is a measure of the compressive load required to cause breakage of pellets. Such compressive loads may for example arise during handling, transport or storage. The cold compression strength was determined for a number of DRI samples produced in the DR pilot plant under various conditions using either natural gas or hydrogen as reducing gas. Mean CCS was determined from measurement of 60 pellets for each sample, in accord with the method of ISO 4700:2015 "Iron ore pellets for blast furnace and direct reduction feedstocks — Determination of the crushing strength". The results are shown in Table 1 and Figure 1.

Table 1

Example no. CCS Metallization C Pressure Red. gas Red. Temp

(daN) (%) (%) (barg) (°C)

1 155 89 1.9 3 NG 1080

2 134 85.5 2.8 4 NG 1040

4 169 99.7 1.55 3 H2 885

5 162 99.4 0 2 H2 900

7 148 93.4 0 3 H2 900

8 172 99.1 0 4 H2 1000

9 139 91.7 0 4 H2 800

10 126 95.2 0 2 H2 800

11 145 96.3 0 3 H2 800

13 169 98.5 1.1 3 H2 935

14 169 99.3 0 3 H2 935

16 165 97.6 0 4 H2 880

NG = Natural gas; H2 = Hydrogen

The DRI of examples 1, 2, 4, 5 and 14 was produced in single-shaft operation whereby a cooling gas was provided to the cooling zone of the DR shaft. The DRI of all other examples was produced in dual-shaft operation whereby cooling and optionally carburization was performed in a separate shaft. Examples 4 and 13 were cooled in natural gas and thus contain carbon. All other examples were cooled in a non-carburizing cooling gas such as nitrogen or hydrogen.

From the CCS values obtained, the following conclusions can be drawn. Higher CCS values can be obtained by reduction in hydrogen as compared to reduction in natural gas. A comparison of examples 1 and 2 shows that when reducing in natural gas, higher carbon levels in the DRI have a seemingly negative impact on CCS. A sample of industrially produced DRI purchased for reference had higher carbon and even lower CCS (85 daN) as compared to the pilot-produced examples. However, when pellets are reduced in hydrogen and subsequently carburized during cooling, the carbon level in the DRI seems to have little to no effect at the levels tested in the examples, as shown by comparison of examples 4, 5 and 13. A comparison of examples 13 and 14 shows that choice of single or dual-shaft operation does not seems to affect the CCS value. Although not shown in the present examples, it has been observed in other experiments that the iron ore source and properties also has an impact on CCS of the obtained DRI. Finally, the results indicate that higher metallization improves the CCS value for hydrogen-reduced DRI.

Looking more closely at the impact of metallization on CCS, Figure 1 shows a plot of the metallization values, CCS values and carbon content (x 10) of the various examples. Line 101 marks a metallization value of 97%. Line 103 marks a mean CCS value of 160 daN. It can be seen that for a metallization greater than approximately 97% a high mean CCS value is obtained, above about 160 daN in the present examples. Conversely, metallization below approximately 97% gives a lower mean CCS, below about 160 daN in the present examples. Metallization correlates to a certain extent with reduction temperature, with higher reduction temperatures typically giving higher metallization, although, as demonstrated by example 7, a high reduction temperature is no guarantee of high metallization.

Study 2 - Excavation studies

In order to investigate the properties of pel lets/DRI at various points when passing through the direct reduction shaft, an excavation was performed. This involved operating the DR shaft at a chosen stable state for a determined period of time, followed by quenching the shaft to halt reduction and subsequent excavation to retrieve samples at varying depths within the shaft.

The chosen operational point before the quench used hydrogen as reducing gas, had a reducing gas temperature of 935°C to the reactor, a system pressure of 3 barg and a reducing gas flow of 2450 Nm ³ /h. The DRI was cooled in the cooling zone with a flow of 775 Nm ³ /h nitrogen. The operational point before the quench had a stable period of approximately 28 hours. Key quality parameters of the DRI obtained at the reactor outlet are shown in Table 2 below (as determined by XRD). It can be seen that performing H2-bsed direct reduction under these specified conditions in a pilot direct reduction shaft as described under "setup" permits the production of DRI having very high total Fe and metallization. Such a DRI has the advantageous mechanical and ageing attributes as disclosed herein.

Table 2

Specification %

C 0

Fetot 98,4

Metallization 99,4

Fe20s 0.3

FesO4 0.4

FeO 0,006

Quench was performed using nitrogen gas. Once the reactor was quenched and cooled, excavation was performed. An excavation consists mainly of sampling along the shaft in both radial and vertical direction descending down into the shaft. The target layer-thickness in the reduction zone was set to 150mm, with thicker layers in the isobaric and cooling zones. 13 layer-samples were taken out for each layer. Each layer-sample weighed approximately 1200g. Composition and cold compression strength (CCS) was determined for pellets in each layer sample.

It was found that as the iron ore is progressively reduced from hematite (Fe2O3) via magnetite (FesO4) to wustite (FeO), the compressive strength of the pellets decreases, reaching a minimum of approximately 85 dN at a depth of 3 - 3.5 m into the DR shaft. It was found that once most oxides had been reduced, somewhere about the transition between the reducing zone and the isobaric zone, the strength recovers to approximately 150-170 dN. This indicates that the oxides magnetite and wustite play a central role in determining the compressive strength of DRI pellets.

For comparative purposes, a similar quench and excavation was performed on a natural-gas (NG) based direct reduction. The reducing gas flow was 2500 Nm ³ /h and the reducing gas temperature was 1080 °C, but the process parameters were otherwise similar. Key quality parameters for the DRI obtained prior to quench are shown in Table 3 below. It can be seen that the total Fe and metallization are at the higher end of the range of what is typical for industrial DRI obtained by fossil-based processes, and that there are still relatively large amounts of residual oxides, particularly wustite.

Table 3

Specification %

C 2

Fe _to t 93,2

Metallization 87,3

Fe ₃ C 36,2

Fe ₃ O ₃ 0,08

Fe ₃ O ₄ 2,7

FeO 12,7

After quench and excavation, CCS and composition analyses were performed on excavated samples. It was found that the drop in compressive strength when using a natural gas-based reducing gas is more pronounced in the reducing zone, reaching a minimum of approximately 70 dN at a depth of 1 - 1.5 m. Although the strength appears to subsequently recover on increased reduction, reaching approximately 120-150 daN in the isobaric zone, it can be seen that the reduction never fully proceeds to completion, and even pellets in the isobaric zone exhibit significant residual quantities of the oxides magnetite and wustite. This tallies well with the quality of DRI obtained before quench.

Figures 2a and 2b illustrate the microstructure of the H2-reduced (2a) and NG-reduced (2b) pellets. It can be seen that the H2-reduced pellets contain very little residual oxides, and any oxides remaining are mainly located between grains. However, NG-reduced pellets still contain considerable amounts of wustite located inside each grain, indicating that NG-based reducing gas has difficulty in permeating to the grain centres.

To summarize, the excavation experiments indicate that the presence of oxides such as wustite and magnetite are detrimental to the compressive strength of DRI, and that reduction in a carbon-containing reducing gas may exasperate the detrimental effects of these oxides. Moreover, the experiments demonstrate that it is possible to obtain DRI having very high metallization and nearly no residual oxides by using hydrogen as reducing gas, whereas comparative experiments using natural gas as reducing gas resulted in DRI having more typical values for metallization and residual oxides.

Study 3 - Tumbling tests

In order to further investigate the effects of reducing gas composition and temperature on DRI mechanical properties, the tumbling and abrasion indexes of a number of DRI samples were obtained. The tumbling index provides an indication of the susceptibility of DRI pellets to break due to abrasion during handling and transportation. The tumble and abrasion indices of the tested DRIs and iron ore pellets were determined using the methods of ISO 3271:2015 "Iron ores for blast furnace and direct reduction feedstocks — Determination of the tumble and abrasion indices".

Figure 3 shows the results of these tests, as well as the metallization and carbon content of the various tested samples. The exact metallization and carbon content of Sample A (industrial reference), Sample C (NG excavation) and Sample D (H2 excavation) are unknown. To the left, natural gas based DRI are presented. The lowest value of 90.4 % > 6,3 mm after tumbling (TTH) is for Sample A, which is a purchased industrial DRI reference produced using fossilbased direct reduction and having rather high metallization degree and carbon content (exact composition unknown). This value can be compared to the pilot produced fossil-based reference, Sample B, with a TTH value of 95.3 % after tumbling. From the excavation performed after campaign K2 a sample of not fully reduced NG-DRI, Sample C, has also been tested as a comparison. It can be seen that the lesser degree of metallization appears to correlate with lower tumbling index.

For the hydrogen-reduced DRI, presented to the right in the graph, an excavation sample with not fully reduced H2-DRI is included (Sample D). This result is lower than the rest of the hydrogen-reduced samples that have metallization in excess of 98% and are produced using differing process conditions (Samples E-L). Such differing process conditions include i.a. varying reducing gas temperature between 800 to 900 °C and carbon-contents (carbon-free or post-carburized with natural gas to a carbon content of 1 %C ). The results from the tumbling of highly-metallized hydrogen reduced DRI is in all cases a tumbling index TTH of between 98 to 99 % TTH. By comparison, these values are much superior to those obtained from natural gas-reduced DRI (Samples A-C), and are even superior to the iron ore pellets used for direct reduction.

The abrasion index (ATH) is also shown, representing the percentage of a sample after tumbling that is less than 0.5 mm. It can be generally stated that the abrasion index inversely correlates to the tumbling index.

From the tumbling tests it can be seen that hydrogen-reduced DRI has significantly improved mechanical properties as compared to natural gas-reduced DRI references. The excellent mechanical properties are obtained for H2-DRI over a range of reduction temperatures, and regardless of whether the DRI is subsequently carburized. However, the incompletely reduced H2-DRI sample from the excavation study was found to have inferior tumbling index compared to the fully reduced H2-DRI samples having metallization greater than or equal to 98%.

Study 4 - Ageing studies

In order to study reactivity and reoxidation of the produced DRI batches during storage, a number of ageing studies were performed, both under ambient conditions as well as under conditions expected to accelerate aging.

Bulk ageing studies

Aging studies were performed on a variety of DRI batches by filling a large bag (volume approximately 1 m ³ ) with each batch and then storing these bags sheltered at ambient temperatures. The batches studied were:

NG-reduced DRI (metallisation 87.6 %);

High-metallized H2-reduced DRI (metallization 99%, reduction temp. 900 °C); and

Carburized high-metallized H2-reduced DRI (metallization 99%, C 1.4 %, red. temp. 900 °C).

Changes in composition and metallization were determined by periodically sampling a number of pellets from each bag and analysing the pellets using XRD and LECO elemental analysis. A mean change in composition and metallization could then be determined for each period in time. It was found that NG-reduced DRI demonstrated a reasonably rapid loss in metallisation, losing approximately 1.6 % metallization in the first four weeks (28 days) of storage. However, no further significant drop in metallization was observed after storage fur a further 4 weeks (56 days total).

None of the high-metallized H2-reduced DRI's were found to have any detectable decrease in metallization, even after extended storage for a total of 180 days. This was the case regardless of whether the DRI was carbon-free (cooled in nitrogen) or carburized (cooled in natural gas).

Accelerated ageing in water

In order to further investigate the effect of metallization and carbon content on the ageing properties of hydrogen-reduced DRI, a number of H2-reduced DRI's were subjected to accelerated aging tests in water. The batches tested were the high-metallized H2-reduced DRI's as described in the aging experiments above (both carbon-free and carburized), as well as a further batch:

Mid-metallized H2-reduced DRI (metallization approx. 96%, red. temp. 800 °C).

Approximately 200 g of each DRI was put into separate buckets that were then filled with water until the DRI was completely covered. Samples were taken after 3 days, 2 weeks and 4 weeks, and analysed by XRD and LECO elemental analysis as previously described. Before the water-drenched samples could be prepared for analysis, they were dried at 105 °C for 24 h.

It was found that each of the highly metallized H2-DRI's had lost approximately 1-1.5 % in metallization after 28 days. This was the case regardless of whether the DRI was carburized or carbon-free. The mid-metallized H2-DRI showed an even larger decrease in metallization after 28 days, approximately 2-4 %.

Single pellet ageing studies

In order to further investigate the effect of metallization on aging, single pellet studies were conducted on the carbon-free (nitrogen-cooled) high-metallized and mid-metallized pellets. Individual pellets were stored either indoors, or outdoors protected from precipitation. At regular intervals, the pellets were weighed using a high precision scale. The total testing period was approximately 1 month. All increase in weight was assumed to be due to reoxidation of iron. The results are shown in Figure 4.

It was found that the high-metallized H2-DRI had very little propensity to gain weight over the test period, regardless of whether it was stored indoors or outdoors. The mid-metallized H2- DRI stored outdoors was shown to gain weight in a linear fashion throughout the test period, resulting in a total weight gain of approximately 0.4-0.5 % at the end of the test period. The mid-metallized H2-DRI stored indoors was found to relatively rapidly gain approximately 1.2 % in weight (after approx. 1 week), but did not gain any further weight after this initial increase.

In summary, hydrogen-reduced DRI was found to age more slowly than natural gas-reduced DRI. For hydrogen-reduced DRI, increase in metallisation was found to lead to less rapid ageing, both in ambient tests and in accelerated (water) tests. Carbon content of the hydrogen-reduced DRI was not found to have any significant effect on ageing, at least for the highly-metallized H2-DRI's that were tested.

Study 5 - Porosity and surface area measurements

Porosity and BET Surface area were determined for a number of the examples from Study 1, together with some further examples.

The porosity of the tested DRIs is determined by the methods of ISO 15901-1:2016 "Evaluation of pore size distribution and porosity of solid materials by mercury porosimetry and gas adsorption — Part 1: Mercury porosimetry". The mercury temperature was 20.0 °C and the pressure range was 0.10 to 61,000.00 psia.

The BET surface area of the tested DRIs is determined by the methods of ISO 9277:2010 "Determination of the specific surface area of solids by gas adsorption — BET method". Krypton at 77K analysis bath temperature was used in BET surface area determination.

At least two pellets were tested for each example, and the values shown are the mean values for all pellets of each example. A compilation of the results is shown in Table 4.

Table 4 Example no. Red. Red. Pressure Metallization C Porosity BET gas Temp (barg) (%) (%) (%) Surface

(°C) (m ² /g)

1 NG 1080 3 89 1.9 56.08 0.39

2 NG 1040 4 85.5 2.8 54.12 0.39

3 H2 909 3 98.3 0 52.35 0.37

4 H2 885 3 99.7 1.55 55.14 0.39

5 H2 900 2 99.4 0

7 H2 900 3 93.4 0 54.61

8 H2 1000 4 99.1 0 59.78

9 H2 800 4 91.7 0

10 H2 800 2 95.2 0

11 H2 800 3 96.3 0 45.68

13 H2 935 3 98.5 1.1 55.30

14 H2 935 3 99.3 0 56.55

16 H2 880 4 97.6 0

LI H2/CO 800 - - 0.7 62.0 0.65

L2 H2 800 - - 0 60.3 0.62

L3 H2 800 - - 0 67.00

L4 H2 900 - - 0 65.85

L5 H2 1000 - - 0 69.98

Examples labelled Ln were obtained by lab-scale reduction of iron ore pellets in a flow of the relevant reducing gas heated to the relevant temperature. The iron ore pellets used are the same type as used in the pilot scale studies. Clear differences can be observed between the properties of laboratory-produced DRI produced by batch process and pilot DRI obtained by large-scale continuous process in a pressurized shaft. Pilot-produced DRI in general has a lower BET surface area and lower porosity than laboratory-produced DRI, typically porosity < 60% and BET surface area < 0.5 m ² /g. On the contrary, laboratory produced DRI typically has porosity > 60% and BET surface area > 0.6 m ² /g. No significant differences with regard to % porosity and BET surface area were found between

NG-reduced DRI and H2-reduced DRI.

However, in subsequent tests, the median pore diameters and total pore areas of NG-reduced and H2-reduced DRI samples were also determined by the methods of ISO 15901-1:2016. The samples tested were the same or similar to the industrial-scale samples listed in Table 4 above.

It was found that the tested NG-reduced DRI samples had median pore diameters ranging from 1.1 pm to 1.4 pm. The H2-reduced samples had median pore diameters ranging from 1.5 pm to 4.0 pm. Improved strength and ageing was found to correlate with greater median pore diameter. That is to say that H2-reduced samples having superior mechanical and ageing properties had median pore diameters greater than or equal to 2.0 pm, preferably greater than 2.5 pm. Such differences in median pore diameter were found irrespective of whether the H2-reduced DRI was carburized or not. That is to say that H2-reduced DRI that is subsequently carburized has a significantly larger median pore size than NG-reduced DRI, and may be distinguished from NG-reduced DRI in this manner. The H2-reduced samples having superior mechanical and ageing properties also were found to have a total pore area of less than or equal to 0.5 m ² /g.

Summary of Experimental Studies

Thus, to summarize, it has been found that highly metallized hydrogen-reduced DRI produced in pilot scale has superior mechanical and ageing properties as compared to traditional natural gas-reduced DRI as well as compared to hydrogen-reduced DRI with lesser metallization. The pilot scale DRI can be distinguished from laboratory scale DRI by porosity and BET surface area. Pilot scale high metallized hydrogen-reduced DRI is primarily distinguished from pilot scale natural gas-reduced DRI by its high metallization and corresponding lack of the oxides magnetite and wustite. The pilot scale high metallized hydrogen-reduced DRI is also distinguished from pilot scale natural gas-reduced DRI by its relatively large median pore diameter (> 1.5, preferably > 2.5). If the pilot scale high metallized hydrogen-reduced DRI is not carburized it may also be readily distinguished from pilot scale natural gas-reduced DRI by its lack of carbon.

The person skilled in the art realizes that the present disclosure is not limited to the preferred embodiments described above. The person skilled in the art further realizes that modifications and variations are possible within the scope of the appended claims. For example, the skilled person understands that DRI pellets having specific combinations of metallization, carbon content (or lack thereof), median pore diameter, porosity and BET surface area not specifically disclosed in the examples may be possible to produce under appropriate conditions.

Moreover, the skilled person understands that the favorable results obtained herein may be obtained using further suitable iron ore pellets and using further suitably dimensioned DR shafts than those specifically disclosed herein. Additionally, variations to the disclosed embodiments can be understood and effected by the skilled person in practicing the claimed disclosure, from a study of the drawings, the disclosure, and the appended claims.