The electric arc furnace (EAF) is currently the technology with the lowest carbon footprint for steelmaking, with an average of 500-800 kg of CO2 produced per tonne of steel produced.
In addition to 100% scrap-based EAF steelmaking, the EAF production route has shown that it can produce the same high-value products as the long-line (blast furnace/converter with carbon reduction) route if up to 40% of the metal feedstock is OBMs produced directly from ore (pig iron, direct reduced iron (DRI) or hot briquetted iron (HBI)). The circular economy of scrap steel depends on the use of raw material OBMs produced from ore to meet product performance requirements.
Static electric arc furnace energy balance：
Figure 1 Schematic diagram of static energy balance of electric arc furnace (EAF) smelting (left side: power input: reaction, natural gas burner, metal oxidation, scrap volatile combustion, electrode consumption oxidation. Right side: energy output: flue gas takes away heat, furnace shell and furnace cover water cooling, electric furnace slag, molten steel.
2. Source of Carbon in Electric Arc Furnace
The current CO₂ emissions are classified as follows:
•Scope 1 – covers direct emissions from owned or controlled sources.
•Scope 2 – covers indirect emissions from purchased electricity, steam, heating and cooling consumed by the reporting company.
•Scope 3 – includes all other indirect emissions occurring within a company’s value chain.
Currently, most long-process steelmaking companies focus on scope 1 emissions, with less scope 2 emissions. For steelmakers, smelting fluxes, ferrous metal feedstocks and even logistics can significantly increase scope 3 emissions, which may affect the process technology used in the future.
The first step to reducing the carbon footprint of EAF is to identify the various sources of carbon entering the process. The following are the main sources of CO2 emissions from electric arc furnaces:
• Power generation.
• Natural gas, used as fuel.
• Oxygen-carbon reaction in the molten pool.
• Combustion of oil, grease and other combustible materials mixed with waste.
• Carbon contained in various scrap steels.
• Carbon contained in ore-based metal materials OBM.
• Consumption of graphite electrodes.
• Consumption of refractory materials – such as MgO-C bricks.
• Reaction of carbon injection in electric furnaces.
• Reaction of carbon injection in electric furnaces.
• Evolution of CO₂ during lime calcination.
Figure 2 shows the amount of CO₂ that each source contributes to the overall carbon footprint of the EAF process. When we seek to eliminate these CO₂ emission sources from the EAF, we must remember why they were implemented in the process in the first place and what function they provide.
3. Forms of Fossil Fuel Energy in Electric Furnaces
The energy input of the electric furnace process comes in two forms: electrical energy input and chemical energy input. Chemical energy input is provided by the following equipment operations:
• Natural gas oxygen burners.
• Oxidation reactions necessary for steelmaking – C, P, Si, Al – must be taken into account.
• Oxidation reactions in the molten pool during the smelting process result in oxygen being injected into the molten pool to react with Fe and Mn, generating heat energy that enters the molten pool.
• FeO in the slag is reduced by carbon, which is an endothermic reaction. This reaction helps the slag “foam”, which helps shield the arc and greatly improves energy efficiency.
• Graphite electrode oxidation and electrode tip sublimation.
The concept of foamy slag evolved from the use of DRI in electric furnaces, where it was recognized that under the right conditions, the evolution of CO in the steel bath and at the steel-slag interface would result in slag foaming up to three times that of unfoamed slag.
This phenomenon was observed to result in better arc shielding, better arc stability and much improved efficiency of energy transfer to the bath.
Foaming of the slag is a natural consequence of using DRI in an electric furnace, as this material contains carbon and unreduced iron (FeO). As the material melts in the slag, CO bubbles are generated and the slag foams, and research into this mechanism has led to the injection of more oxygen in the electric arc furnace and carbon spraying to artificially create foamy slag and achieve the associated benefits.
As the operation of electric furnaces has evolved, the tap-to-tap time has become shorter and shorter, and it has been recognized that there are “cold zones” in the furnace where the scrap does not melt as quickly as elsewhere in the furnace.
Therefore, even though most of the metal in the furnace has melted, the melting time is extended to melt the scrap in these cold spots, which results in increased power-on time, reduced furnace capacity, overheated zones in the furnace, and reduced overall energy efficiency.
Oxygen burners are used in the furnace to accelerate the melting of scrap in the cold zones. Although the additional energy used in the cold zones is relatively small, the impact on efficient melting operations is huge.
Prior to these advances in EAF technology, chemical energy accounted for approximately 15-20% of the net energy input to the EAF. With the injection of large amounts of oxygen into the EAF, the net contribution of chemical energy to the EAF increases to 40%, and in some extreme cases to 50-60%.
As the heat generated by the carbon-oxygen reaction within the EAF increases, the productivity of the EAF also increases, but at the expense of energy efficiency, as the amount of energy lost to the flue gases also increases significantly. In modern high-efficiency EAF operations, it is common for flue gas energy losses to be between 30% and 50% of the total equivalent energy input to the EAF.
EAF Operations – Future Challenges
The most logical approach to reducing the carbon footprint of an EAF is to evaluate the various sources of CO₂ in EAF operation and then determine how the fossil fuels associated with that function can be replaced or reduced without negatively impacting process efficiency.
4. Foam slag – shielding arc
The foaming of arc furnace slag to form foamed slag has proven to be one of the important technological advances in electric furnace steelmaking.
Poor arc shielding coverage can result in 50-70% heat transfer efficiency for input electrical energy, but good foamed slag shielding to control the arc can achieve an electric arc furnace electrical efficiency of up to 93%.
In some smelting operations, the efficiency of fully submerged arc approaches 100%. Good foamed slag shielding of the arc is most important when the arc furnace is operated under flat bath conditions. In the initial stage of scrap melting in a conventional electric furnace, the electrode penetrates the well and the arc is contained by the scrap. For flat bath operation throughout the tap-to-tap cycle, this foamed slag shielding of the arc is essential for efficient operation.
For most EAF operations, foaming slag is the most effective way to shield the arc. However, it is important to note that this is not the only option. If the foamy slag height is sufficient to shield the arc, it may not be necessary to continue foaming the slag to increase the height. This requires the EAF to be designed to handle a larger slag volume and brings some additional challenges, such as:
• Dephosphorization of steel may be more difficult.
• Larger slag volumes will require additional energy to heat it to the required smelting temperature. Each metric ton of EAF slag contains about 740 kWh/metric ton of energy. If the amount of slag in the EAF is doubled, the energy requirement may increase by 60-80 KWh/metric ton of liquid steel.
• Larger slag volumes may cause slag to be tapped.
• Lumpy slag formed by excess steel in the EAF may become a problem, especially when the EAF has a cold zone.
If the gangue content in DRI/HBI increases due to a shortage of low-gangue DR pellet feedstock (i.e. resulting in the use of lower grade iron ore pellets), the resulting increase in slag volume may actually improve EAF operation.
However, if the phosphorus content of the pellets is high, a significant problem arises in that there is not enough FeO in the slag, which, combined with high slag basicity, may not reduce the phosphorus content in the steel to acceptable levels. If oxygen usage is limited, it may be necessary to add scale or iron ore fines to the slag to provide sufficient FeO to facilitate phosphorus removal.
One of the biggest challenges associated with large slag volumes is the potential for reduced iron recovery. This would not be a problem if the EAF was designed to accommodate a larger slag volume. In some past operations, scale from the steel rolling process was added to the slag in order to reduce iron losses and improve metal recovery.
Maintaining tight endpoint control is critical to controlling iron losses. If oxygen is overblown and the slag volume is large, iron losses are greatly increased. Of course, iron losses due to slag are less significant if oxygen usage in the EAF is significantly reduced.
Slag foaming also facilitates efficient use of gas and solid matter injected into the EAF, and its elimination may result in greater process inefficiencies and lower recovery rates.
Excessive foamy slag in EAF operations may also cause problems with the lid feeding of raw materials such as DRI, HBI, granulated pig iron (GPI) and fluxes.
To promote rapid melting and iron entry into the bath, it may be necessary to operate at lower slag basicity and/or modify the slag to reduce viscosity and facilitate iron discharge into the bath. Similarly, if the material contains high phosphorus content, high iron losses may occur and dephosphorization operations may not be possible at low slag basicity.
Many EAF operators overlook the impact of gangue and trash fouling on the carbon footprint of EAF operations.
The greater the amount of trash/gangue associated with scrap/OBM, the greater the amount of metal that must be charged to the EAF to produce a given weight of ladle steel.
Of more concern is the fact that the higher the trash/gangue content, the more basic fluxes (lime, dolomite) need to be added to the EAF. Although lime/dolomite does not contribute to a steel plant’s CO2 emissions, its production from limestone/dolomite feedstock is a major source of CO₂ emissions and must be addressed to some extent, which is beyond the scope of this article and will not be discussed here. The combined impact of higher trash/gangue content on iron production, flux requirements, energy consumption, etc. can have a considerable impact on the carbon footprint of steelmaking and must be addressed as soon as possible.
At present, high-efficiency electric arc furnaces operate with a carbon injection rate of 10-20 kg/metric ton, producing 37-74 kg of carbon dioxide per metric ton of molten steel.
There are several alternatives to carbon injection, and recycled plastics have been proven as a foaming agent for slag formation in EAF operations around the world.
The main issues with this technology include proper separation of waste plastics (based on content) to ensure that hazardous byproducts (such as dioxins/furans) are not produced in the EAF, and the fact that the plastics decompose very quickly after being injected into the EAF.
This creates a problem with the continuous release of CO bubbles to foam the slag. The rapid burning of plastics often results in a surge of CO gas, which is not conducive to a good and stable slag foaming operation. Therefore, 20-35% recycled plastics are usually used to replace the injected carbon.
Another possibility is to use biomass instead of carbon injection. To be effective as a slag foaming agent, the biomass must be converted to biochar. In the past, some EAF operations have used charcoal made from roasted coconut shells as an effective slag foaming agent, and other waste biomass sources may be suitable. The key to good slag foaming operation is to achieve the correct size and density of these materials, and to allow continued pyrolysis combustion to produce CO gas to foam the slag.
In the case of recycled plastics and biochar, considerable work needs to be done to determine the optimal size and properties to provide high recovery rates and sustained slag foaming.
A major challenge to biochar utilization in electric furnaces is that biomaterials tend to have very high moisture contents, so moisture must be removed from the material before it can be used in the electric furnace process.
The energy costs required to remove the moisture may affect the economic viability of using this material.
In addition, some potential sources of biochar may contain phosphorus, nitrogen or sulfur, which are detrimental to the steelmaking process. It will be important to fully understand the origins of various biomaterials to find the sources that are most suitable for producing biochar. In order for biochar to be truly carbon neutral, the rate of production (plant growth) must match the rate of consumption, and it remains to be seen whether this balance can be maintained.
Slag viscosity poses a potential problem for feeding DRI/HBI through the lid. A well-foamed slag will allow the material fed above the lid to penetrate into the slag where melting occurs.
Conversely, if a more foamy slag is used as a means of shielding the arc, higher iron losses may occur. In addition, if oxygen utilization is reduced, the FeO content in the slag may not be high enough to form a fluid slag.
The feed rate above the lid may be limited and DRI/HBI may accumulate on the slag, resulting in delayed melting of the material and possible metal particles entering the flue gas system, resulting in reduced yield.
5. Auxiliary Energy Requirements
To ensure uniform melting of the scrap in an electric arc furnace, it is necessary to provide some form of energy input in the cold zone of the furnace.
The simplest solution is to replace the natural gas currently used in electric furnaces. Possible “green” alternative fuels could include hydrogen, ammonia, biogas, synthesis gas, coal gas or gas from municipal solid waste.
Ammonia is essentially an alternative to using hydrogen as a fuel, but with the benefit that it is safer, more stable, and easier to transport and store than hydrogen. However, the burner design needs to minimize NOx formation, as high NOx levels negate the benefit of reducing CO₂ emissions.
Another option is to input energy into the cold zone using a non-transferred arc plasma torch.
This technology has been used in the past to heat the steel in the tundish of a continuous caster. Plasma torches usually cost much more than burners of the same heat rating, but the advantage of plasma technology is that it is powered by electricity.
6. The role of oxygen in low-carbon steelmaking
Oxygen currently has several functions in the EAF. Over the past 25 years it has become the primary energy source for EAF production. However, secondary roles include the reaction products promoting slag foaming and oxidative metallurgy to remove harmful impurities from the steel.
• The reaction of oxygen with carbon in the EAF bath also provides a means of removing dissolved nitrogen and hydrogen from the steel. If CO generation in the bath is limited, the result may be high levels of dissolved nitrogen in the steel at tapping time. This can be addressed by additional treatment of the steel using a vacuum degasser in downstream processes, but for some grades requiring very low nitrogen levels (<25ppm), the required conditions may be difficult to meet.
7. Low-carbon scrap steel smelting
Many predictions for a carbon-neutral steelmaking future call for utilizing DRI/HBI produced by hydrogen reduction. No doubt, as more and more end-of-life scrap is recycled, some form of OBM combination will be needed to facilitate scrap recycling. Full hydrogen-based DRI/HBI has zero carbon content, and if some natural gas is used in the reduction process, the material may contain up to 1% carbon.
If the material contains no carbon, the melting temperature will be higher than that of carbon-containing metallic charges, which may result in higher power requirements for the electric furnace and longer power-on times.
In the past, Cleveland-Cliffs produced zero-carbon HBI at its Trinidad facility. Melting tests showed that this material melted quickly when layered with pig iron in a scrap basket. The high carbon content in pig iron allows for faster and more complete melting of the zero-carbon material. However, this can create a problem in low-carbon electric furnace operations, where the continuous feed rate on the furnace lid for zero-carbon HBI/DRI may be significantly lower than that for carbon-containing metallic charges.
8. Electric furnace design considerations
Based on some of the changes in the EAF process identified in the previous part of this article, it can be seen that some improvements to the EAF design are still needed. One of the changes is to deepen the furnace, which will be beneficial for the upper furnace slag to be used in the next batch of molten steel by retaining steel. If a large amount of DRI/HBI is fed through the furnace cover, this is also beneficial. However, a method must be provided to mix and stir the top feed in the molten steel pool, which can be achieved by gas stirring/electromagnetic stirring, or by injecting inert gas through the furnace wall gun. In any case, the energy consumption of the EAF will increase.
As oxygen and fossil fuel use in the furnace decreases, it may be possible to eliminate some water cooling components, for example if the energy input is primarily electrical and concentrated in the center of the furnace.
However, this approach must balance the energy concentrated in the center of the furnace with the formation of cold steel shells and cold zones around the furnace, and adjustments to the electrode circle may be required to achieve the correct heat transfer balance.
If the main form of energy input to an EAF is electricity, then several challenges arise. In the past, the use of chemical energy grew because it could be added to the EAF while it was powered. This allowed for shorter and shorter smelting cycle times. In the past, the limitation on the electrical energy input was related to the capacity of the EAF transformer.
In recent years, the capacity of the transformer has increased significantly and is no longer a bottleneck. However, if future EAF operations wish to maintain tap-to-tap times as low as 30-40 minutes, then graphite electrodes must be able to provide power densities twice as high as currently required.
In low/no chemical energy use electric furnaces, power demand could grow significantly, perhaps to 500-550KWh per ton of steel produced.
This would require a 60-90% increase in power density in the furnace to maintain the same furnace smelting cycle time. This would require stronger graphite electrodes and more powerful transformers. The physical properties of the graphite electrodes may need to be adjusted, and grid requirements may limit the capacity of the furnace operation (a function of the short circuit capacity (SCC) at the point of common coupling (PCC) and the amount of static VAR compensation provided). This is a particular concern for the current electric furnace industry, and many steel mills may have maximized the capacity of the transformer based on the current SCC of the PCC. If this is the case, static VAR compensation may be an option in the future, or a major upgrade of the grid may be required to allow more active power (MW) to be used to maintain current productivity levels. The most advanced static VAR compensation systems can provide quarter-cycle corrections, and in the future, it may be necessary to make corrections on a faster basis.
As the EAF process evolves, better process analysis tools combined with advanced instrumentation are needed to provide more complete and timely information to control the process and quickly identify the impact of process changes and equipment.
9. Impact of Flue Gas System
One of the biggest benefits of reducing carbon combustion in an EAF is that flue gas generation is significantly reduced. Depending on whether carbon is eliminated or replaced with biochar material, the requirements on the flue gas system may change. Not only will the volume of flue gas be reduced, but the energy content of the flue gas will also be much lower than with conventional EAF process operation.
In some cases, up to 50% of the energy goes into the flue gas, which includes both sensible and chemical heat. This has led to a great deal of interest in recovering energy from the flue gas, but it is a difficult task because the thermal energy contained in the flue gas can vary by as much as 300% throughout the entire tap-to-tap smelting cycle.
If flue gas generation and the associated heat losses were significantly reduced in an EAF, flue gas heat recovery would not be necessary and would result in higher overall EAF energy efficiency.
The lower flue gas flow rate in the furnace of the electric furnace can allow finer materials to enter the top of the furnace cover and prevent them from being carried away by the fast-flowing flue gas, thus reducing material loss.
It is worth noting that the current emphasis on decarbonizing steelmaking processes is focused on process and equipment changes.
In reality, the carbon footprint of any EAF operation can be reduced by about 20% through process optimization. Heat/energy recovery could further reduce the carbon footprint by 20-30% in the future. While both of these optimization options have been explored in the past, they have never been adopted by mainstream steel producers, despite offering a potential payback period of 2-5 years.
They now need to be revisited to see what can be achieved in the near term.
10. Summary and Conclusion
The purpose of this paper is to provide a catalyst for discussion on future EAF process operations, in areas where specific emphasis has not been given because the fundamentals driving the EAF process are not well defined.
It is recognised that the energy forum will play a key role in future ‘green steelmaking’ activities, but most steelmakers are more focused on the ‘finish line’ of CO2 reductions than on the steps that can be taken on the way to achieving those reductions.
This paper attempts to fill some of these knowledge gaps and identify some questions that we should be asking sooner rather than later. Other questions will certainly arise as we examine the technical challenges and requirements. This paper does not purport to provide all the answers, but rather to facilitate a meaningful discussion aimed at providing strong solutions for the sustainability of steel production. To summarise the main points:
•Carbon can never be completely eliminated from steelmaking because steel, by definition, is an alloy of iron and carbon.
• If the use of chemical energy in the electric arc furnace is minimized, the power demand will naturally increase. Based on data from historical operations, the demand is expected to be 500–550 KWh per ton of steel. This actually means an improvement in thermal efficiency of about 10-15% over current operations. However, based on some of the process options being considered, this figure could eventually be closer to 600 KWh/ton.
• Availability and specification of ferrous raw materials will be key to reducing the carbon footprint of EAFs. As discussed in the previous section, we must do a better job of separating end-of-life scrap and returning it to the appropriate steelmaking facility to match the residual element requirements of the steel product to be produced (e.g., remelting recycled rebar scrap to produce rebar).
It is hoped that technologies will be developed to remove harmful residual elements from scrap. But until this can be achieved, OBM will be required to dilute the residual element content within the scrap to the required product requirements. Better sorting and processing of scrap is also required to remove free copper components and retain the scrap value of high-quality scrap. Blending of various scrap types should only be done at the steel mill to meet the steel product specifications of the product.
• In the future, it is expected that most electricity will be generated from renewable sources, so the largest source of CO₂ emissions associated with EAF operations will be eliminated. Therefore, the increase in electricity demand will be more about the grid configuration and whether the short-circuit capacity at the point of common coupling can support the supply of the required energy. The CO₂ footprint of power plant generation will be less of an issue, the challenge will be technical and financial, and for many existing EAF plants this may require major upgrades to the grid to meet process targets.
• Carbon substitution with waste materials (e.g. plastics) and biogenic carbonaceous materials (biochar, other plant waste) is also feasible but does not actually reduce the carbon footprint of the EAF process unless the EAF is used at a rate that matches the efficiency of the electricity generation.
This is not the case with most biogenic carbonaceous materials and in the case of waste materials, although the resulting carbon footprint is real, the consumption of these materials is beneficial to society as a whole and perhaps these CO₂ can be grouped in a different grouping (e.g. CO₂ emissions providing social benefits – elimination or reduction of waste).
Removing moisture from biogenic carbonaceous materials prior to use as biogenic carbonaceous materials may result in significant energy absorption and process inefficiencies.
• Significant opportunities to reduce the carbon footprint of electric arc furnace steelmaking can be achieved in the near term through process optimization and the application of heat recovery technologies in steel mills. The development and application of appropriate process tools can greatly accelerate these improvements.
• Optimising raw material utilisation is critical to reducing the carbon footprint of EAF steelmaking. Iron yield is key to this discussion and the evolution of the EAF cannot be considered in isolation from raw material supply.
•Some aggressive steps must be taken in the short term to conserve higher quality scrap. Failure to do so will have a negative impact on any attempt to reduce the carbon footprint of electric arc furnace steelmaking.
• As arc furnace operations shift to electric power imports, significant changes to the power delivery to the furnaces may be required in the short term. This may require upgrades to the grid, switchyards, furnace transformers and even graphite electrodes.
There will be many possible innovations in the EAF process. It is important that we follow a structured approach to this evolution:
1. Understand the material inputs to the process and their associated CO₂ emissions.
2. Implement metal material replacement combinations in a way that does not affect the efficiency and capacity of the electric arc furnace.
3. Evaluate the need for physical changes to the EAF and related auxiliary equipment to accommodate the desired process improvements.
4. Continue to work towards improving efficiency through process assessment and optimization, identifying opportunities for optimization and energy recovery.
5. Improve instrumentation and data collection to improve process understanding. There is a great need to further embrace digitalization and develop new instrumentation accordingly.
6. Develop better process feedback tools to identify the impact of changes more quickly and comprehensively.
