Microstructure Effects in Hydrogen-based green steel production

Steel production is a major contributor to global warming, accounting for 7%-8% of global CO2​ emissions and about 35% of all CO2​ produced in the manufacturing sector, primarily due to the use of fossil carbon-based reductants.

Hydrogen-based direct reduction (HyDR) is an attractive, alternative processing technology aimed at replacing carbon-based reductants with sustainably produced hydrogen.

Despite hydrogen diffusing faster than carbon-based reductants, the net reduction kinetics in HyDR remain extremely sluggish for high-quantity steel production, and the hydrogen consumption is substantially higher than the stoichiometrically required amount.

In our microstructure- and kinetics-oriented studies about hydrogen-based direct reduction (HyDR) of iron oxide we identified the importance of understanding heterogeneity at the iron oxide pellet and microstructure scales to improve efficiency and accelerate the sluggish net reduction kinetics of HyDR.

More specific we investigated commercial direct reduction (DR) hematite pellets at 700°C using techniques like synchrotron high-energy X-ray diffraction (HEXRD) and electron microscopy with EBSD and EDX.

Our ultimate goal is to develop tailored iron ore pellets for fast, sustainable, low price and efficient hydrogen-based direct reduction process, which is critical for achieving carbon-lean steel production.

Some main takeaways from our investigations are:

Role of Reduction Kinetics and Phases

    The overall reduction proceeds along the sequence Fe2​O3​ (hematite) →Fe3​O4​ (magnetite) →Fe(1−x)​O (wüstite) →α-Fe or γ-Fe at temperatures above 570°C.

The reduction from wüstite to α-iron (the final stage) started slow (∼0.6×10−3s−1) and continuously slowed down, indicating this stage is extremely sluggish, and complete metallization (98% reduction degree) required 52 minutes at 700°C.

The gradual deceleration of the transformation rate is a common feature within each phase regime, largely attributed to changes in the pellet's microstructure, such as the gradual loss of internal free surfaces (pores, cracks) that facilitate rapid diffusion.

Through-Pellet Spatial Gradient

A significant finding is the strong heterogeneity of the reduction rate along the pellet radius.

Near-Surface Region: Showed the highest metallization degree of 88.3 vol% α-iron with small fractions of wüstite and magnetite (5.6 vol% and 6.1 vol%, respectively). This is due to rapid hydrogen intrusion and fast oxygen removal/water formation on free surfaces.

Near-Center Region: Showed very low metallization, with only 3.6 vol% α-iron, and the majority consisting of wüstite (83.2 vol%) and magnetite (13.2 vol%).

This indicates a drastic difference in the reduction rate between the near-surface and interior regions.

Microstructure and Mechanism in Hydrogen-based direct reduction (HyDR)

Pores and Defects: In the early stages of individual phase transformations, high porosity, delamination at hetero-interfaces, and cracking facilitated rapid mass transport and removal of water, enabling fast nucleation and growth.

Encapsulation: As the reaction progresses, remaining wüstite islands become increasingly encapsulated by iron, impeding the outbound oxygen transport through the denser reaction products. This microscopic core-shell behavior is a key reason for the sluggish final reduction stages.

Iron Nucleation: Iron nucleation occurs adjacent to free surfaces (pores) because the nucleation barrier is likely smaller due to the heterogeneous nucleation advantage and the relaxation of elastic stresses from the large volume mismatch between iron and wüstite (over 40%).

Some further microstructure items to explore in Hydrogen-based direct reduction (HyDR)

The large through-pellet reduction gradient suggests that the current commercial pellet design and process optimization may not be fully suitable for HyDR processes, especially considering the need for efficient use of expensive green hydrogen. Future work might place focus on experimentally and theoretically assessing the effect of pellet size, porosity, and microstructure on gaseous percolation. This knowledge will guide the design of next-generation pellets for faster and more efficient HyDR.