New research shows hydrogen-reduced iron has superior properties | Hacker News Reader

New research shows hydrogen-reduced iron has superior properties

group.vattenfall.com

112 pointstda57 comments

Here's a good steel production flowchart that shows where this process would be used (replace the natural gas-powered direct reduction feed-in to the electric arc furnace with a hydrogen-powered direct-reduction step). Overall, it's part of the electric arc process which avoids the need for coke from coal in blast furnaces.

https://www.steel.org/steel-technology/steel-production/

Of course, making this fossil-fuel free requires significant hydrogen production from sources like hydropower (or nuclear) powered electrolyis, or my favorite, photoelectrochemical reduction of water, such as:

https://techxplore.com/news/2017-08-decades-technology-nrel-...

As far as the claims about producing a superior product with hydrogen compared to natural gas, that's hard to evaluate without more data. I imagine quality of the incoming ore is a major factor. If you want a technical paper on it, here's something recent from what looks like a China-Germany research collaboration:

https://sci-hub.se/10.1016/j.actamat.2021.116933

"(2021) Influence of microstructure and atomic-scale chemistry on the direct reduction of iron ore with hydrogen at 700°C"

P.S. The issue of hydrogen embrittlement of steel doesn't apply here, although it is a major issue that doesn't bode that well for replacing natural gas with hydrogen anywhere other than in industrial processes where the hydrogen is being consumed almost as fast as it is produced. Making synthetic natural gas from atmospheric CO2 and water-sourced hydrogen is a (currently expensive) option, however.

Methane digesters would probably be more efficient, but I’m not sure about the volume.

It is substantially more efficient, by unit of collecting area, to produce electricity using a regular solar panel and then electrolyse water for hydrogen. But of course collection area efficiency is not what matters, cost is. We don't know what the electrophotochemical equipment costs, or how that compares. It does seem more elegant, at a remove.

In practice, of course, you drive your electrolyser from solar, wind, hydro, geo, tidal, numerous forms of storage, or an unholy mix of all of them at different times -- whatever comes off the grid. At first, it will include NG sometimes, and nukes, until those are priced off of the market.

The photoelectrochemical reduction of water looks interesting -- what do you think the achievable efficiency of it might be and is there an existing path to commercialize the technology?

Also, would that work in a concentrated solar setup?

I always found the process promoted by Boston Metal (https://www.bostonmetal.com/) to be more interesting.

No need for a hydrogen in-between step, just electricity.

There is huge boom at wind power around Gulf of Bothnia, and we need to dump all that excess power somewhere so I think hydrogen would come there.

The press release from Vattenfall relates to a research program called Hybrit. Vattenfall (an energy company), together with the iron ore company LKAB, the steel manufacturer SSAB, and Luleå University of Luleå (LTU) are partners in the project.

For more details about Hybrir see: https://www.hybritdevelopment.se/en/a-fossil-free-future/ https://www.hybritdevelopment.se/en/research-project-1/

The research group list of papers (not all of them related to Hybrit): https://www.ltu.se/research/subjects/Processmetallurgi/Publi...<

A working paper related to Hybrit: https://cdn.sei.org/wp-content/uploads/2020/07/bigger-is-som...

Can this reasonably be more efficient than the direct electrolysis process? There was a major breakthrough in 2013 with a ferrochromium anode:

https://europepmc.org/article/med/23657254

which led to an NSF grant:

https://www.nsf.gov/awardsearch/showAward?AWD_ID=1534664

which spawned a startup that is now seeking private investors:

https://www.bostonmetal.com/transforming-metal-production/

Maybe not but the hydrogen process is probably a game changer regardless, because there is less research and productisation needed. It is similar to existing processes.

It’s also perfectly positioned to take advantage of wind power. The gas storage tanks can make up for variations in electricity production. It’s all set to be deployed at a large scale already.

Watt for watt it might not (I have no idea) be as efficient as electrolysis, but as a total solution I see how it could be deployed at scale. Combined with the advances in electric mining equipment like what Volvo is offering, the industry is all set to go fully carbon neutral and electric. (No tech hurdles left.)

It's not just a question of efficiency. How expensive is the equipment? And in particular, if it's using intermittently available power, how expensive is the part that must be ready to absorb that power? For hydrogen it's electrolysers (and perhaps compressors); for that scheme it would be the oxide electrolysis cells.

I recently came across Boston Metal’s proposed use of this inert anode material but I couldn’t find much detail on whether the “breakthrough” claim was genuine. It still seems early days, though if true it would be a transformative enabler for the production of low-carbon steel.

The claim is that "Hydrogen-reduced carbon-free DRI is highly metallized and has superior mechanical and aging properties compared to direct reduced iron using fossil-based reducing gas such as natural gas." I wish there were more details.

Agreed. This claim is so vague, and it sounds like this is just one part of the smelting process.

Yes, there is little value in a press release about research that does not include a link or refence to said research.

Hydrogen is well known and studied to cause many issues with steel, generally brittleness and crystal structure defects. In situations where it’s important that cracking doesn’t occur, it’s common for low/no hydrogen electrodes in welding be required, for instance.

That using a method of production that naturally exposes it hydrogen would produce higher quality steel is surprising! We should still definitely have a link to the study!

This process produces direct reduced iron, not steel.

What do you think Steel is, but reduced Iron (to a specific level)?

Cast Iron is also impacted by hydrogen embrittlement, albeit iron is usually already pretty brittle so folks don’t use it where that kind of failure matters as much.

It produces an intermediate product. You can't use it without further processing. In any case, the hydrogen won't be a problem.

You mean, cast iron is pretty brittle. Pure iron is not brittle.

If you're talking pure iron, then yes - it's very ductile and malleable. However, the low end of carbon allowed while still calling something steel is .3%, so with a few exceptions (wrought iron, which is arguably very soft steel), it's rare to run across it.

There are of course ways to treat cast iron so it's decently durable (ductile iron), but it's still pretty brittle compared to 99% of steel in the real world.

> low end of carbon allowed while still calling something steel is .3%

Incorrect. That is the maximum for low carbon steel (also called mild steel - the most common form of steel).

Wikipedia says “The carbon content of steel is between 0.002% and 2.14% by weight for plain carbon steel”. “Plain carbon-iron alloys with a higher than 2.1% carbon content are known as cast iron.”

https://en.m.wikipedia.org/wiki/Steel

https://en.m.wikipedia.org/wiki/Pig_iron

https://en.m.wikipedia.org/wiki/Cast_iron

https://en.m.wikipedia.org/wiki/Carbon_steel

https://en.m.wikipedia.org/wiki/Alloy_steel

https://en.m.wikipedia.org/wiki/Iron

This is a process for refining iron ore, as opposed to using a blast furnace, or using the same process with decomposed natural gas instead of just hydrogen. In steel production it would be subsequently smelted.

Do you think that removes the concern? I don’t see how it would.

This process produces sponge iron, which is then fed into an electric arc furnace. There is no free hydrogen around at that point. Whatcomes out of an electric arc furnace is carbon steel.

The issue with hydrogen, which was first noted IIRC in the high-pressure Haber-Bosch ammonia production process, in which H2 and N2 at high pressure over a catalyst in a steel chamber forms NH3, is that free hydrogen reacts with the carbon in the steel under these conditions, which caused the pressure chambers to regularly explode (solution was a sacrificial lining of the chambers which was regularly replaced). For low-pressure H2 it may not be much of a problem, but that's not very efficient for transport.

Thanks for the background! So likely due to the complete remelt and re-alloying expected, it should work out?

Speaking of ammonia, any thoughts to share on an ammonia economy vs hydrogen economy? The former seems a lot more practical.

Ammonia as fertilizer is the only ammonia economy that makes sense; hydrogen economy doesn't make much sense either. Methane economy is the optional long-term solution for chemical energy sources, just CO2(atm) + H2 (water) -> CH4.

Ammonia combustion generates NOx smog, and ammonia is pretty toxic to store and transport as a gas. For fertilizer it's stored as (NH4)(NO3) or similar.

Hydrogen production has a few major use cases, but replacing natural gas is not one of them. Iron reduction, nitrogen reduction, carbon reduction, in industrial settings, is about it.

Thanks!

For what it's worth to reduce surprise/skepticism, a hydrogen atmosphere is already commonly used for annealing electrical steels (eg. motor laminations or magnetic parts).

https://www.tlclam.net/annealing-process/

https://www.vac-met.com/hydrogen-annealing/

Using natural gas introduces carbon. After the DRI stage, further treatments would be required to remove the carbon as it oxidizes. With less carbon there should be more consistent oxidizing of the iron, especially since it looks like carbon monoxide(from the natural gas) is not required which would introduce unwanted oxygen. I wish there were more details as well.

The thing is you usually WANT Carbon in your Iron, so if you use Hydrogen DRI you have to further process the Iron to add the Carbon back in. There's a lot of potential for using mixtures of Hydrogen and Natural Gas to produce Iron with the exact desired Carbon content.

It seems to me that, when you want carbon in your iron (to make steel), you want a lot of control over how much carbon is in that steel. To start from a base iron that has no carbon in it whatsoever seems like it would be better for being more precise in your steel production.

While you certainly can add or remove Carbon during processing, you get the lowest costs if your input Iron is as close as possible to the desired Carbon content for your application.

Again, it seems to me that, if you care about extreme precision in your steel-making process, cost is the least of your worries.

Carbon is wanted for steel production, but at specific levels. DRI is further worked to oxidize carbon and remove it before processing. Eliminating the carbon/oxygen from natural gas reduction allows a more controlled introduction of carbon and or less processing to make ready for steel production.

Steel is an incredible massive category. For high quality steels (duplex, maraging, PH stainless), you actually don't want much carbon. Since they talk about "aging properties" in the article, I'd imagine they're targeting the nicer stuff. This doesn't seem like a process intended for cheap steel.

FWIW, iron reduced in hydrogen is known to have a very high magnetic permeability, surpassed only by one of the Metglas alloys.

This is just a PR release. I hope their claims prove out. But there is no evidence supplied and breathless PR releases seldom go anywhere.

From the article: "One element behind the high quality of the sponge iron produced is the high-quality iron ore from LKAB’s mine..." So would this process produce high quality sponge iron from iron ore of lesser quality from other mines?

What the link is "conveniently forgetting" to mention is that most industrial hydrogen production emits CO₂ (steam reforming: CH₄ + 2H₂O → 2H₂ + CO₂). And there's a good reason for that: electrolysis consumes a lot of electricity. I won't do the maths here but, if the electricity was generated through fossil fuels, I wouldn't be surprised if the process actually increased CO₂ emissions instead of reducing them.

But let's say that the hydrogen is from fossil-free electricity. You could be plugging that fossil-free electricity elsewhere instead. The press and media link does not mention that.

>Hydrogen-reduced carbon-free DRI is highly metallized

That can mean two things:

1. They're able to retrieve more iron from the oxide. Good, but it isn't enough to replace the current processes; at most to add hydrogen as the "chef's kiss" to the iron produced through another method.

2. Less cementite aka iron carbide aka the stuff that actually makes steel "steel" instead of plain iron. That's great or awful depending on application.

>has superior mechanical [...] properties

Again, it depends on application. I expect their iron to be rather soft and malleable, but lacking tensile strength.

>This new knowledge is a direct result of close value chain cooperation, determined innovative thinking and bold efforts in piloting new technology – a recipe to copy for other industrial sectors,”

>Hybrit Development AB has filed patent applications describing the included inventions to the European Patent Office.

"Guys, we made something to copy for other industrial sectors, except that we're smearing patents on it so you can't copy the process~".

There's another detail that the press and media link doesn't mention: hydrogen makes steel brittle.

______________________________________________________________

Take this link with a grain of salt.

Sweden, where this facility is supposed to be built, hardly has any fossil fuel electricity production.

Also you should compare the natural gas hydrogen process to the currently used coke process.

Correct.

https://en.wikipedia.org/wiki/File:Electricity_production_in...

>Sweden, where this facility is supposed to be built, hardly has any fossil fuel electricity production.

That would make their process suitable for countries that barely produce any iron to begin with. Unlike countries actually responsible for a big chunk of the world's iron production¹, and heavily reliant on fossil fuels; such as China, Japan, India, Russia.

>Also you should compare the natural gas hydrogen process to the currently used coke process.

No, I shouldn't. For three reasons: a) everybody knows that the "traditional" coke process is nasty, and is looking for alternatives; b) other alternatives already exist, as syngas-based³; and 3) the resulting iron is slightly different in properties.

Could the method from the link become more attractive in the future? Sure. Even then I strongly suggest everyone to hold their horses before going "WOOO THAT BIZNIZ SAID THAT THEY'RE CHANGING THE WORLD!!1". There's no "exciting" development yet.

_____________

1. Source: https://en.wikipedia.org/wiki/Iron#Industrial_production

2. Refer to https://ourworldindata.org/grapher/exports/share-electricity...

3. Note that syngas still uses coke, but as a DRI method it uses it more efficiently.

> I won't do the maths here but, if the electricity was generated through fossil fuels, I wouldn't be surprised if the process actually increased CO₂ emissions instead of reducing them. > But let's say that the hydrogen is from fossil-free electricity. You could be plugging that fossil-free electricity elsewhere instead.

This assumes a zero-sum situation where we have a fixed amount of fossil free electricity, but that's unlikely because fossil free energy (solar specifically) is the cheapest type of electricity generation we can build today.

Furthermore, the hydrogen can be electrolyzed at times when the supply of fossil free energy exceeds demand, thereby actually improving the economics of intermittent renewables by increasing their overall utilization, and hence incentivizing building more if it.

This is Sweden and almost at the most northern part of it. It is located a relative short distance from the polar circle. Solar is a very cheap form of electricity but it has some significant problems to overcome that far up north.

Hydro, nuclear and wind is what northern Sweden uses.

> This is Sweden and almost at the most northern part of it.

> Solar is a very cheap form of electricity but it has some significant problems to overcome that far up north.

Why would the hydrogen-reduced steel need to only be produced in Northern Sweden? It could just as well be produced somewhere with abundant solar.

That's where they have decided to build the plant.

Sure, but subsequent plants needn't be built in Northern Sweden. It would make a lot of sense to build them in other areas with lots of green power potential.

The plant is near the mines in northern Sweden. They could in theory be transporting the hydrogen, but hydrogen can be a bit tricky to transport and it obviously would cost more money. The green hydrogen process is also fairly expensive right now (despite what people might say), and the cost is not just attached to the electricity price. They had a person from the project discuss the value proposition just a few weeks ago in Swedish news. Green steel need both subsidies and green credits, but even so the end cost is landing at a few percent higher than regular steel. The argument was that even if it cost more its better for the environment, and this is a pilot program so maybe the cost will continue down in the future.

>This assumes a zero-sum situation where we have a fixed amount of fossil free electricity

I'm not assuming a zero-sum. The concern still holds without a zero-sum situation, as long as some commonly used sources of energy are not fossil-free.

>but that's unlikely because fossil free energy (solar specifically) is the cheapest type of electricity generation we can build today.

Higher demand can increase prices. Usually this wouldn't be a problem, but considering how big the iron/steel production sector is, the impact of the electricity used for the H₂ be measured, not assumed.

>Furthermore, the hydrogen can be electrolyzed at times when the supply of fossil free energy exceeds demand, thereby actually improving the economics of intermittent renewables by increasing their overall utilization, and hence incentivizing building more if it.

That is actually a fair argument. Unlike the above.

> Higher demand can increase prices.

Sure, but absent material supply constraints for production of PV and wind (of which none exist), supply will respond to that demand as it always has.

Using green hydrolyzed H2 for making steel makes more sense because our only alternative for steel is to use coal to produce it, which we know is terrible from a C02 emissions perspective.

Personally, I don’t see industrial processes like these as a priority for reducing CO2 emissions. If we get to a point where all the heavy emitters like cars and power plants are carbon neutral, things like steel production will be a tiny contributor to climate change, and can probably be better managed through capture and sequestration if the world decides every molecule of CO2 is harmful.

Concrete production is responsible for 7% of global CO2 emissions so there are a few key processes and material which really are worth a look.

I’d still put it below transit and electricity as far as priority.

Depending on which source you look at, steel production accounts for 8% - 11% of global CO2 emissions. According to Gate's 2021 book, 'getting around' (planes, trucks, cargo ships) accounts for 16% of CO2 emissions. So steel production is definitely a significant contributor.

I get that, but steel is kind of “worth it”. Civilization can’t function without steel right now. Meanwhile energy has a gazillion alternatives. Priorities.

Reading between the lines, it seems like this is intended for high quality steels. Think aerospace, not buildings. Actually in those applications, you don't really want carbon. The strongest steels have virtually no carbon. Hydrogen embrittlement and cracking have nothing to do with this.

Reply on news.ycombinator.com