The Winding Road toward Zero-Carbon Iron



The steel industry’s biggest product today isn’t steel—it’s carbon. 

Roughly 1.8 tons of CO2 are released into the atmosphere for every ton of liquid steel produced on the still predominant, integrated route. And this number is based on calculations that assume an average, modern blast furnace operated in OECD Europe—many plants around the world emit three or more tons of CO2 per ton of steel. These facts are hardly news to decision makers inside the industry. But they are only beginning to dawn on governments around the world who are now weighing their options for bringing about the massive reductions in greenhouse-gas emissions mandated by the Paris Climate Agreement. As some of these governments strive toward the even more ambitious goal of net-zero carbon emissions by 2050, scrutiny is inevitably going to increase and pressure is going to build on steel producers to bring down emissions in drastic ways. Emissions-trading schemes are going to be implemented and widened in scope, carbon taxes are looming, and consumers will ultimately show concern for the carbon footprint of steel products. 

Producers in some parts of the world are already feeling the painful pinch of carbon pricing. The development of prices for European CO2 Emissions Allowances, as illustrated in Fig. 1, is a prime example. At their peak in mid-2019, prices had seen a sixfold increase over a two-year period. And they remain near record highs as the EU emissions trading scheme is poised to enter its next phase in 2021—when several measures kick in that will progressively tighten the supply of allowances.

In anticipation of such regulatory measures and market pressures, steel producers around the world are racing to deploy new technologies aimed at reducing carbon intensity in iron and steelmaking. Merely shifting production from coal and coke-based blast furnaces to direct reduction based on natural gas will not be sufficient. The industry will need to develop other energy sources without a direct carbon footprint, such as hydrogen, to commercial scale and in a manner that is economically feasible.


As Fig. 2 shows, the DR-EAF route, applying natural gas in the process, cuts carbon intensity for liquid steel by almost 50 percent in comparison with the conventional blast furnace to basic oxygen furnace route. By using green hydrogen instead, emissions can be reduced by 75 percent. Most emissions at this point are actually attributable to electricity production for the EAF process. The calculations are based on a grid emissions factor of 0.452 kg of CO2 per kWh—the average value for OECD Europe. Applying the emissions factor of Sweden (currently at 0.023 kg of CO2 per kWh) would bring emissions down to only 181 kg CO2 per ton—a whopping 90 percent reduction compared to the BF–BOF route. From there, all that stands in the way of true zero-carbon iron would be the provision of fossil-free energy for electricity, heat, and transport.

Debug: 5


Needless to say, establishing a ready supply of hydrogen that can make iron and steelmaking truly carbon-­free will be a major challenge. One of the key barriers is the sheer volume needed to support a massive upscaling in use by the industry. The amount that will be consumed is tremendous. Converting a typical integrated plant with a production of 5 million tons per year from coal and coke to hydrogen will require a supply of at least 480,000 Nm³ (equaling 44 tons) of hydrogen per hour. To put this into perspective, the largest proton exchange membrane (PEM) electrolyzer currently in operation generates only about 1,200 Nm³ (i.e. 0.1 ton) of hydrogen per hour. The largest one currently in construction is a plant in Canada that is to produce 3,000 Nm³ (0.25 tons) per hour, starting in late 2020. Not only will vast quantities of additional hydrogen be needed to support the steel industry, but it will also have to be produced using alternative processes. Currently, around 95 percent of hydrogen is “gray”—meaning it is produced by extracting gas from fossil fuels. It is possible to use carbon capture, utilization, and storage (CCUS) technology to prevent emissions from being released into the atmosphere, resulting in "blue" hydrogen. But this only makes economic sense where a high volume of CO2 can be captured at a single site. And it is only feasible in areas where there are geologically safe places to store captured carbon, such as beneath the sea or deep underground.


To be a true game changer, the hydrogen used in the ironmaking process must be "green" hydrogen, generated by electrolysis from water, using fossil-free power only. To date, producing hydrogen in this way has proven too costly to be competitive. But this is changing, as the boom in renewable energy sources, like wind and solar power, brings down global electricity prices and the generation of hydrogen provides a way to store energy in times of excess electricity production—when the wind is blowing or the sun is shining during times of low demand. Establishing a policy and financial-incentive framework for adopting hydrogen is an essential part of this shift toward a more sustainable future, which gives industry leaders the confidence to invest in long-term hydrogen projects.

For example, a project called the Asian Renewable Energy Hub plans to generate green hydrogen with up to 15 gigawatts of renewable power from thousands of large wind turbines and photovoltaic solar panels in Western Australia. Local government is providing crucial support and MHI Group company Vestas, a leader in wind-turbine technology, is a part of the consortium. The first hydrogen could be tapped in 2026 and production capacity could reach 1.5 million tons per year. This would be enough to produce 25 million tons of DRI per year, which equals 25 percent of today's global production capacity.




  • H2 as a reducing agent

  • Direct use of iron-ore ­concentrate fines

  • No agglomeration

  • Modular Plant design


Hydrogen-based Fine-Ore Reduction (or HYFOR for short) is the world's first direct-reduction process for iron-ore concentrates from ore beneficiation that does not require any preprocessing of the material like sintering or pelletizing. Building on comprehensive experience from Primetals Technologies' Finmet and Finex processes, the new technology can be applied to all types of beneficiated ore. It works with particle sizes of less than 0.15 mm for 100 percent of the feedstock, while allowing a maximum grain size of 0.5 mm. Thanks to the large particle surface, the process achieves high reduction rates at low temperatures and pressures.


As a primary reduction agent, the new process uses H2 from renewable energy or alternatively H2-rich gases from conventional steam reformers. As yet another alternative, HYFOR can run on H2-rich waste gases. Depending on the source of the hydrogen, this leads to a low or even zero CO2 footprint for the resulting direct-reduced iron. The plant features a modular design with a rated capacity of 250,000 tons per module and year, making it suitable for all sizes of steel plants. A pilot plant for testing purposes is being constructed at voestalpine Stahl Donawitz, Austria and is due to be commissioned by the end of 2020.


HYFOR both drastically reduces CO2 emissions and helps producers to effectively deal with the challenge of reduced iron-ore quality, which has become more acute as of late—resulting in an increased need to beneficiate the ores. Rising demand for iron-ore pellets for blast furnaces and direct-reduction plants has led to higher prices for iron ore, especially pellet premium. With HYFOR, producers will be able to use pellet-feed fine ore directly and benefit from the rising global supply of ultrafines.


The HYFOR pilot plant at voestalpine Donawitz will consist of three parts: A preheating-oxidation unit, a gas-treatment plant, and the core—the novel and unique reduction unit. In the preheating-oxidation unit, fine-ore concentrate is heated to approximately 900°C and fed to the reduction unit. The reduction gas H2 is supplied over the fence from a gas supplier. A waste-heat recovery system that harnesses heat from the off-gas ensures optimal energy use and a dry-dedusting system takes care of dust emissions from the processes. The resulting hot direct-reduced iron (HDRI) leaves the reduction unit at a temperature of approximately 600°C and can subsequently be fed into an electric arc furnace (EAF) or used to produce hot-briquetted iron. 

The purpose of the pilot plant is to provide practical evidence for this breakthrough process and to serve as a testing facility, collecting enough data to set up an industrial-scale plant at a later stage.


As of today, most of the hydrogen for industrial use is produced by steam methane reformers (SMRs). Since the natural gas that feeds these reformers (methane—CH4) contains carbon, the resulting "gray" hydrogen causes a sizable amount of CO2 emissions. To fully decarbonize the process, hydrogen for ironmaking will have to be produced by electrolysis from water, using fossil-free energy. 

Since the year 2000, more than 230 hydrogen-­electrolysis projects (counting both those based on renewable as well as conventional sources of energy) have entered into operation around the world. Most of them, like the H2Future project in Fig. 4, are based in Europe. But several have been started or announced in Australia, China, and the Americas. Almost all have been at a scale of less than 10 MW, but a 20 MW plant is being constructed in Canada and lately there have been several proposals for plants exceeding 100 MW—chief among them is the Asian Renewable Energy Hub project in Western Australia (see page 41).

There are three main electrolysis processes that make all of this possible: alkaline, proton-exchange membrane (PEM), and high-temperature steam electrolysis. The most advanced technology at this point—and the one that most recent hydrogen-generation projects have favored, is PEM. It attaches electrodes on two sides of a solid polymer membrane, which acts as the electrolyte and as a separator to prevent the produced gases from mixing. Hydrogen ions form at the anode, pass through the membrane and combine with electrons from the cathode to form hydrogen gas. 

The PEM-type electrolyzer has several advantages: It is highly efficient, features high power density, and has an extended dynamic operating range, allowing it to be directly coupled to renewable sources—because it can quickly react to changes in the electricity supply. Modules are available in the range of 3 to 100 MW, producing up to 20,000 Nm³ (or roughly 2 tons) of hydrogen per hour.



Hydrogen is already part of the reduction-gas mix in mainstream direct-reduction processes—alongside carbon monoxide, of course. Most notably, perhaps, the Finmet (Finored) process, introduced by Primetals Technologies (VAI) in the 1990s, uses more than six parts of hydrogen for every part of carbon monoxide. 

This flexibility of direct-reduction processes provides an enticing way for plant owners to switch to hydrogen in a gradual fashion, increasing the addition of hydrogen over time as prices come down. It also underlines the value of new investment into direct-reduction plants: Whatever the future might hold in terms of emissions regulations or prices for raw materials, direct-reduction technology allows for maximum adaptability.


Midrex direct-reduction plants produce roughly 60 percent of the world's DRI and have been available from Primetals Technologies for more than 30 years. The reducing gas—mainly a mixture of H2 and CO—is produced from natural gas in a special CO2 reformer supplied by Midrex Technologies. Without any modifications to the equipment, the process allows up to 30 percent of the natural gas to be replaced by hydrogen. To take an example, 60,000 Nm³/h of hydrogen could be brought in to substitute 20,000 Nm³/h of 
natural gas in the process. With minor additions to the equipment (to protect the reformer), the rate can reach as high as 100 percent—the limit may be determined by the required carbon content in the final product. The process can easily accommodate fluctuations in the hydrogen-addition rate, allowing the plant to react to changing hydrogen supplies (which are to be expected when sourcing the gas from water electrolysis with renewables like wind or solar).


If hydrogen is to be used as the sole reductant permanently (Fig. 7), the natural-gas reformer can be replaced by a reduction-gas heater. H2 will be converted to H2O during reduction and condensed in the top-gas scrubber. Since there is no source of carbon monoxide in the process loop, there is no need for a CO2 removal system. The process uses approximately 550 Nm³ of hydrogen per ton of DRI for reduction. Additionally, it requires about 250 Nm³ of H2 per ton of DRI for heat—which can be accommodated by other energy sources.

Debug: 5


Using hydrogen to decarbonize ironmaking and other industrial processes is a powerful idea—but not actually a new one. Leaders in the industry who are now nearing retirement may remember it vividly from their student days. There were peaks of interest in the ’70s during the oil price shocks, the ’90s with incipient concerns about climate change, and in the early 2000s on the same note. None of these developments resulted in a breakthrough moment for hydrogen. However, this time around may well be different: As the International Energy Agency (IEA) notes in a recent report, last year marked a period of unprecedented momentum for hydrogen, combining a new depth of political enthusiasm with a breadth of new possibilities around the world.