Green steel without green hydrogen – can it work? | Hacker News Reader

Green steel without green hydrogen – can it work?

canarymedia.com

62 pointsorangebanana174 comments

So, the process used in the article is molten oxide electrolysis - using a raw iron oxide ore in molten form in a bath with its other (oxygen-bound) impurities - aluminum oxide, burnt lime (CaO), etc.

The neat thing here is that the impurities stay behind in the electrolyte bath after the iron is removed (unsure if it's a gravity mechanism or a cathode attraction, but the molten iron ends up on the bottom of the electrolytic cell). As the article explains, this means that even cheap, low-grade ore with lots of impurities can be used with this technique.

The other advantage is that the direct chemical reduction of iron eliminates the multi step (ore in a blast furnace to get pig iron, pig iron + coke, etc) procedure in traditional production. Among other benefits, you're now only heating the material once instead of 2-3 times, and as a result, you actually consume less total energy in this process even though you have to reach a higher temperature in the single heating.

I imagine the carbon content of the final steal output might still be hard to control. In which case further treatment would be required, though perhaps that could be done directly with the molten output from the cell (e.g. by blowing CO or O2 through the molten iron depending on whether carbon should be added or removed.

You don't target cheap steel market with this, you go after VAR, and ESR steel.

> Among other benefits, you're now only heating the material once instead of 2-3 times, and as a result, you actually consume less total energy in this process even though you have to reach a higher temperature in the single heating.

Combined steel+iron plants were around 60+ years

The key development that allowed this was the 2013 development of chromium anodes that tolerate the insane conditions of the process:

https://pubmed.ncbi.nlm.nih.gov/23657254/

So, that's why MOE wasn't around before 2013.

MOE wasn't, but colocating blast furnaces with steel converters is a very old idea.

Aluminum and calcium have substantially stronger affinities for oxygen than iron does, so they'll tend to stay in oxide form even when you're successfully reducing the oxygen.

This is mentioned in the article:

> All of these oxides are more stable than iron oxide, so the iron oxide is the first to separate when exposed to electric charge, breaking down into pure oxygen and iron. The iron, still liquified, sinks to the bottom where it can be tapped out and turned to steel.

There are other elements in ore that will tend to reduce before the iron, like nickel, cobalt, lead, and copper, but they're a lot less abundant than iron and may not be harmful to the iron produced.

I meant when you're successfully reducing the iron. (And the answer for why it goes to the bottom is that it's two or three times as dense as the oxides.)

The higher temperature doesn't consume more energy theoretically as long as there is some way to use the heat of the produced iron and oxygen to help heat up the incoming iron oxide.

Not many heat exchangers can survive 1100 degrees C, but even a rudimentary 'fan blows air over the produced iron and then directs that through a ceramic pipe and through the incoming iron ore' heat recovery system should extract a large percentage of the thermal energy.

The problem with this is that it doesn't work as well with intermittent source of electric power. The pot has to be kept hot. Contrast this with production of hydrogen. A hydrogen electrolyser can ramp up and down very quickly and is not damaged by long periods in an off state. The hydrogen can then be cheaply stored for a very long time, even seasonally, allowing the direct iron production facility to be operated continuously.

I was thinking the same thing. Though I am surprised to hear that seasonal storage of hydrogen is doable. I imagined that the size of pressure vessels required, as well as the leakage losses of hydrogen in general, would make that infeasible.

Hydrogen can be stored underground just like natural gas. The cost is very low, particularly if there are salt domes available where large cavities can be solution mined. For example, the salt formation at Delta, Utah has room for 100 such cavities that could store enough hydrogen that (if burned in combined cycle power plants) could power the entire US grid for 30 hours. The estimated storage cost is as little as $1/kWh of storage capacity (power related costs are extra).

Here is a commercial example of underground hydrogen storage that has been in service since 2007. It is a salt cavern in Texas. It is part of extensive hydrogen infrastructure connecting many industrial users (600 miles of pipeline)

https://investors.linde.com/archive/praxair/news/2007/praxai...

Would not have expected to see Delta, UT on HN. I spent a significant amount of time there as an undergrad.

So maybe the intermittent sources of clean electric power aren't the best way to go for absolutely everything.

That depends on how cheap they are. Intermittency is a negative, but it's not an infinitely costly negative. Here, the question is how expensive the electrolysers would be -- and they are quickly falling in price now, due to China.

https://www.rechargenews.com/energy-transition/will-us-and-e...

So, all you need is a bunch of batteries or other storage plus some links to back up power supply from the grid (wind, hydro, geothermal, etc.).

Hydrogen is basically just a really inefficient battery. Nice if you really need it for e.g. energy density reasons but basically sub-optimal for other things.

The challenge with hydrogen hydrogen is energy losses in producing it are about 3-4x (so use 4 kwh of electricity to produce 1kwh of hydrogen). And then you lose more actually burning it. And storing and transporting it add to those losses. Using the electricity directly for heating the iron/steel is much more efficient and potentially a lot more cost effective since you effectively use at least 3-4x less MWH. That's a lot of cost savings. And those can finance a lot of batteries and other solutions.

Hydrogen has many other uses than for energy storage. And, when top-line energy production gets cheap enough, round-trip losses come to matter less than other things, such as raw usefulness.

Hydrogen is directly useful in electrical synthethis of methane and ammonia, besides myriad current industrial uses, and, as LH2, is disruptive as an aviation fuel: LH2 aircraft will be impossible to compete with, wherever they are available.

LH2 has very low density, so it cannot be used in an aircraft that looks like what we're currently using. I suspect synthetic hydrocarbons will be used instead. Hydrogen is likely essential in making those.

Low density is also a big reason why using LH2 in the first stage of a rocket launch vehicle is a dumb idea.

It will need new airframes and new cryo infrastructure, which is why it has been slow to happen. Once it does, building out will take a long time. But the competitive benefits are big enough to move a lot of activity.

In the meantime, we will need synthetic hydrocarbons, using captured carbon, for extant airframes. It will need to get (or be made, through carbon taxes) cheaper than the mined, refined, and transported stuff.

Having a big hydrogen synthesis infrastructure in place feeding hydrocarbon synthesis will then ease the transition off of hydrocarbons, as equipment that can use hydrogen directly comes online. Stationary uses will come first, displacing natural gas electric generation.

I've seen a couple articles which push hydrogen but with economics that don't make sense for "green" hydrogen. (Produced by electrolytic cracking of water with solar power)

Instead these are advertisements for "blue" hydrogen-- cracking natural gas and then injecting the CO2 back into the ground. It's a campaign by fossil fuel companies to preserve some of the value of their capital base, rather than just being shut down entirely. Blue hydrogen will be a lot cheaper, at the cost of fossil emissions from methane and CO2 that escape from the equipment or leak from the wellhead.

There's no economics in which blue hydrogen is cheaper than green hydrogen.

So it's basically just good old fashioned lies.

It takes ~1.33 kWh to make 1 kWh of hydrogen.

I think the bigger problem right now is that the pilot scale reactor didn't work. Cf:

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

>Bench scale experiments made use of an externally heated reactor while pilot scale experiments conducted under this grant used a reactor that was self-heating.

>At the bench scale, oxygen was produced with minimal corrosion of the anode material. At the pilot scale experiments in the self-heated reactor were not able to demonstrate oxygen production at the anode, or the production of iron as measured by tracer dilution. Further work is necessary to elucidate the difference between bench and pilot scale results.

It is not trivial to store hydrogen. It takes a very large and expensive tank to store it at any practical level.

And it has a tendency to leak through most materials.

https://planetforlife.com/h2/h2swiss.html#:~:text=Hydrogen%2....

"B&E estimates that a liquid hydrogen tank designed for automobile use will loose about 5% of its capacity every day, which is to say that all of it will be gone in 20 days. Losses of this magnitude are acceptable for, say, a taxicab fleet, but unacceptable to most people."

Also new to me that it's a greenhouse gas: "Hydrogen cannot be vented to the atmosphere because it is an explosion hazard and because it is a greenhouse gas. The vented hydrogen must be burned. A continuously running gas stove with one burner set to "medium" would do it."

As I noted elsewhere, in large quantities it would be stored underground as compressed gas.

The world uses many millions of tons of hydrogen a year. There are already 1600 miles of hydrogen pipeline in the US. So, while materials must be chosen carefully, materials compatible with hydrogen obviously exist.

Transportation is not the only, or even a particularly favorable, use case for hydrogen. Pointing to it to paint all hydrogen uses as inadvisable is not honest, especially in a thread where we're talking about reduction of iron ore, not cars.

Hydrogen is a greenhouse gas in this sense: hydrogen consumes OH radicals in the atmosphere which would otherwise go to destroying methane. So, indirectly, hydrogen causes warming by reducing destruction of methane. The effect is not large, however. Carbon monoxide has a similar indirect effect.

Others have mentioned that molten oxide electrolysis isn't as flexible with electricity consumption as hydrogen based processes. The even bigger problem, IMO, is the incredibly demanding chemical and thermal environment. There are not many combinations of vessel/electrode materials that are stable toward these high temperatures, dissolution by the lava-like electrolyte, and oxidation at the anode.

According to the article their molten bath sounds like it is made of silicates. Electrolyzing molten silicates sounds a lot like NASA's "molten regolith electrolysis" [1] concept for in-situ resource utilization on the moon. The NASA concept valued the oxygen product while the produced metal is much more interesting here on Earth. Although conceptually simple, it has been a tremendous challenge finding materials that endure under these operating conditions. If Boston Metal has really cracked the problem, it bodes well for the future of terrestrial and off-planet resource extraction.

[1] e.g. https://ntrs.nasa.gov/api/citations/20120003037/downloads/20...

1600° is really about the same temperature as an ordinary blast furnace, and silicates have been commonly used as fluxes and refractories in steelmaking, even if basic oxygen steelmaking is more popular nowadays. If I understand correctly, Bessemer converters commonly used silicates. So I don't think the challenges you're pointing at, even if they have to be solved, are anything new for steelmaking.

However, the article suggests that actually the electrolyte might be neutral, containing both calcium and silicon oxides. That can't possibly be right if they're talking about only 1600° because larnite would precipitate out.

Electrodes would surely be carbon, just as they are in any old arc furnace. This does result in carbon dioxide emissions, as it does in the making of aluminum. Can they keep this down to an acceptable level? Is there an alternative electrode material, such as zirconia or carborundum? I'd like to know, but the article doesn't say.

I think carbon electrodes would dissolve in iron. If I recall correctly, it's some sort of oxide-coated metal electrodes.

Well, carbon electrodes don't dissolve fast enough in molten iron to keep them from being standard equipment for decades in EAFs, but you're right that that's not what they're using.

Scythe found this paper by Lan Yin, Antoine Allanore, and Sadoway from 02013 that says they're using 90% chromium, 10% iron, forming a refractory but conductive layer of eskolaite/alumina: https://pubmed.ncbi.nlm.nih.gov/23657254/ Boston Metal links this paper and calls out Allanore on their about page: https://www.bostonmetal.com/who-we-are/.

Sadoway is on Boston Metal's board, but Yin and Allanore evidently aren't involved.

Electric arc furnaces for steel don't typically immerse the electrodes in the molten metal (in the slag on top of the molten metal, yes.)

Oh, of course you're correct, thank you. How high is the solubility of graphite in molten iron?

Iron/carbon phase diagram: http://www.substech.com/dokuwiki/doku.php?id=iron-carbon_pha...

The eutectic of iron and carbon (mixture with the minimum melting point) is 4.3% carbon by mass.

There's a chapter from Richard Preston's "American Steel" (a 1992 book that I highly recommend about the Nucor minimill in Crawfordsville, Indiana) where they find at one point during the startup of the then-cutting-edge continuous casting line that the steel in arc furnace had too little carbon. They decided to solve that (expensively, for the purpose of the testing) by dipping the electrodes into the steel and letting their ends dissolve.

It seems likely that almost any process that uses fossil inputs can be done reasonably well by simply subbing in electricity to fake the original process (e.g in this case making green hydrogen) but that there's a lot of synergies to be unlocked by revisiting from first principles.

It's like TV going from radio with pictures to its own thing, or horseless carriages becoming automobiles etc. A new paradigm for looking at things afresh.

You could even make a "fake internal combustion" engine that uses high-power electric arcs or lasers to super-heat the compressed air in the cylinders at the right moment, instead of combustion. Super inefficient and would likely spew out ozone and/or nitrogen oxides, but fun.

The article doesn't mention it: there is actually a research project in France doing iron oxide electrolysis, but in a different way than what Boston Metal does. It's called Siderwin and steel company ArcelorMittal is involved: https://www.siderwin-spire.eu/ I recently covered this in an article (though German, in case you understand that: https://www.golem.de/news/eisenoxid-elektrolyse-stahlherstel... ).

Main difference between this and Boston Metal: They use a relatively cold process (~100°C) while Boston Metal uses a hot process.

I guess it's good to try to make this work in different ways.

I happened across a great, long-form podcast that touches on this subject the other day. It offers a great survey of broad trends around decarbonizing heavy industry. https://www.volts.wtf/p/volts-podcast-rebecca-dell-on-decarb...

off-topic aside: whats the audio based equivalent to pocket or other "bookmark this to read it later" services.

I dont want to figure out specific audio hosting things for each site, I want to hit a button that means, I want to listen to this audio later and have some code figure out all the BS and put it in a list for me when I have time to listen to something.

That was an excellent listen, thank you for recommending it.

Glad you liked!

One upside I see with hydrogen is that it is easier to store on the scale of days.

Hence you can be more opportunistic with your hydrogen production. Whereas an electricity based process will require a constant supply of electricity, putting more strain on the grid, and on prices in moments where wind and solar are low in production.

In fact, I imagine the hydrogen production could help level out daily fluctuations in power availability.

Is hydrogen easy to store? I was under the impression that it had the tendency to leak out of whatever you store it in, due to its very small size.

Easier at a scale of "national grid for n hours" and beyond than batteries. Much harder than piling up unused coal for a decade or two, noticeably harder than the equivalent stash of hydrocarbon gases or liquids.

For these energy quantities and consistencies, wouldn't a flywheel or localized hydro storage solve for n hours? (Next to a local reactor obv)

Flywheels are quite limited in their storage capabilities. Because there is a limit to their RPM before they tear themselves apart from centrifugal forces. They are really interesting for energy storage on the scale of seconds, to smooth out spikes over that range. Much less interesting for storage on the scale of hours to days. Where batteries and hydrogen become more interesting. Pumped hydro is also quite nice on those scales, but cannot be implemented everywhere.

Hydrogen is much less efficient that batteries, but much more scalable. That makes it attractive for taking in large amounts of over-produced electricity, and for use-cases where you need hydrogen anyway rather than meaning to turn the hydrogen back into electricity.

There is one variety of flywheel that evades this problem. The magnetically confined kinetic energy storage ring levitates and confines a rotating ring via magnetic forces. The mass of the magnets and tunnel required scales in proportion to the radius, while the energy stored scales as the square of the radius. This does not violate the virial theorem, since the design is taking advantage of the underlying bedrock to carry the confining load.

It's a rather whacky idea, and as far as I know it hasn't been looked at for decades, but it is out there.

https://digital.library.unt.edu/ark:/67531/metadc173309/

Yes, in underground caverns, like natural gas is stored. This is a demonstrated technology.

The challenge here is cheap electricity. We can do amazing things with cheap electricity (green steel, green hydrogen, etc), but we are moving backwards on electricity prices. For example, we are seeing large electricity prices increases in Europe.

Anything that relies on cheap electricity is not going to actually be very green.

Wind energy counts as green and will be used for this purpose.

Not necessarily. Northern most part of Scandinavia have among the lowest prices for electricity in Northern and Central Europe. Most of the electricity there comes from Hydropower. It's also the same area where Sweden makes it green steel. Cheap and green electricity is not an oxymoron or impossibility.

I'm pretty sure they're mostly counting on wind energy to provide the cheap electricity since 0 marginal cost while hydro serves as a reliable backup.

Traditionally Swedish electricity has been about equal parts nuclear and hydro. Nuclear usually running full out, with hydro used to regulate for daily and seasonal variation.

In recent years they have built a lot of wind, as of 2019 providing 12% of electricity, today almost certainly more. But yes, like you said, hydro is used as the backup for wind.

Iceland has solved the cheap electricity problem and has easy access to iron ore from the port of Narvik.

Currently we have a "problem" of solar energy being too cheap (even negative price) at peak.

Could a steel production still be viable if it could only run during daylight, or adjust to intermittent availability of wind+solar?

I know currently production is typically designed to run at maximum utilization 24/7, but maybe the solution is to rethink that assumption.

I recently heard that, even in Germany where there is lots of solar, there were only 9 days in the last year where spot prices were negative. Sadly my source is just "someone said".

As for intermittent steel production, from what I know about aluminum it is essentially impossible to turn the process off, because the molten stuff solidifies and is impossible to melt again.

Steel plants sometimes have deals to operate at certain times for load shedding anyway, but how about they use the times when we have an excess of solar/wind and its not well suited to steel production (maybe due to shift patterns) to generate onsite Hydrogen using that excess power? Then they have readily available onsite green hydrogen.

I think things like this will become an absolute necessity soon, using green power at its peaks to store resources for use later. (Obviously batteries will also be essential, although I would rather see gravity storage with big reservoirs like they have in Wales)

Largely the negative prices are a side-effect of feed-in tariffs, where (some) renewable projects get paid a fixed price/kWh regardless of what the actual spot market price is. So when there's an oversupply of these subsidized renewables it pushes the spot price to negative.

I think newer renewable support schemes have been designed to avoid this problem, either by cutting off the feed-in tariff when the spot price goes to zero, or then by restructuring the subsidies as some kind of investment support (state guaranteed very low interest loans or such) rather than paying for the generated electricity.

That of course doesn't change the fundamental issue, in that in a grid with increasing amounts of variable production renewables and the price determined by marginal cost of the most expensive producer, the price swings will be larger than what we previously were used to.

European electricity still depends on natural gas -- which is largely controlled by Russia in large swaths of Europe. (This is also why Russia might think it can get away with invading Ukraine.)

The cost of electricity once we've transitioned from fossil fuels is ultimately what counts -- and there is a lot of progress being made there, albeit slowly.

Russia's influence, frankly, should be another really big motivator for Europe to transition away from fossil fuels as quickly as possible.

Germany is especially linked to Russian gas for their electricity production, as they shut down all of their nuclear plants in some kind of attempt to be "more green" or something.

> For example, we are seeing large electricity prices increases in Europe.

Which is very good. New capacity will most likely be non-gas.

The electricity price is set by the most expensive marginal cost.

This made sense when renewables were more expensive on average than fossil fuels (which we want to discourage) but the rules probably need rewritten now that renewables are cheaper and continuing to get cheaper and fossils relegated to peaked roles.

At the moment it just means that renewable providers are getting a lot of extra cash for not doing anything different and then on occasions when we go 100% renewable the price will crash suddenly to near zero.

Dispatch based on marginal cost makes sense, as that ensures that the lowest marginal cost generators will be used to provide the demand.

But yes, increasing penetration of variable renewable energy like wind and solar will mean that the price will vary wildly. That means that players in the market will need to be hedged to insure them against too high/low prices.

Also it wouldn't surprise me if various capacity market type mechanisms were to become more popular and constitute a bigger share of the money flows in the electricity market as a whole.

Recent article on this topic, summarizing a paper they co-wrote (linked at the end):

https://theconversation.com/renewables-are-cheaper-than-ever...

Grid prices and cost are two things. EU and US grid pricing is dominated by the most expensive things in the market (gas, coal, and nuclear) because once you use up all the cheap supply, people end up bidding for the expensive stuff. So whenever there's a shortage of that, prices go up.

Burning gas costs money. Shutting down or restarting a gas plant costs money too. And a gas plant that is not running still needs upkeep, maintenance, staffing, etc so it costs money. Some operators actually use negative rates to stimulate demand so they can keep their gas plants running and avoid shutting them down. Negative rates of course cost money, so that is added to the overall grid pricing. None of that has anything to do with the cost of renewables.

That same dynamic is also what makes renewable power very lucrative for operators. A low cost and a high market price just means a lot of profit. That's why world+dog is putting up windmills and solar parks as fast as they can. It's just that good of a deal. And of course the subsidies and positive press help.

If you are consuming a lot of power, that difference means investing in your own power generation makes a lot of sense. Which is why many plans for green steel plants involve plans for e.g. wind turbines and other solutions. So, they only buy from the grid when that supply is inadequate and actually supply to the grid when there is enough supply. Yes that's intermittent. But the connection to the grid isn't and the difference is just cost.

This is entirely going to depend on how governments treat 'blue hydrogen' (ie. hydrogen produced from natural gas where the carbon is injected back underground).

If blue hydrogen is treated as carbon-free by treaties and taxation schemes, then it will be a cheaper energy source than electricity, so it will be used to power steel, concrete, glass, and all the other heat and energy heavy industries.

If however blue hydrogen is put in the bin of dirty fossil power, then steelmaking via hydrogen will prove to be more expensive than direct electrolysis.

So what does Nucor think of this? Nucor, which started as a steel recycler, is the US's largest steelmaker. They use huge electric furnaces to melt down scrap. Scrap is 71% of their input. They now have some basic oxygen furnace operations, too, making new steel, but mostly it's the same steel going round and round.

If anybody is going to make basic steel with electric furnaces, it's likely to be Nucor. When they take this seriously, it's real.

> then refine it from iron oxide into pure iron and fortify it with small amounts of carbon. It’s a complex process that emits carbon at different stages. Some emissions come from the heating process, which usually involves burning a heat-refined form of coal called coke. A bit of the carbon from the coke gets dissolved into the iron, turning it into steel.

Very incorrect. It's not even a highschooler level mistake.

I worked in a steel mill a while back. While I agree it's incorrect, I would say that the average high school student knows next to nothing about making steel.

Here's how we made it: Take pig iron, put it into a Basic Oxygen Furnace. Add flux and inject supersonic oxygen. This removes impurities and carbon from the pig iron turning it into steel. Transfer to a huge crucible. Add any alloying elements, and cast it into slabs using a continuous slab caster.

> I worked in a steel mill a while back. While I agree it's incorrect, I would say that the average high school student knows next to nothing about making steel.

What country you are in?

Worth noting that today Green Steel investment at a facility that is already running (that will convert it to green steel) just happened. It is a $1.8B plan at the Steel facility in Hamilton, Canada. $400M kicked in by the Feds which should be operational by 2028.

Electric arc furnace for the moment is the plan and maybe in the future bringing in Hydrogen. So this is a project in the wild.

Ah yes this is Professor Sadoway's work.

Anyone know if this process could also help with copper contaminated iron scrap?

The melting temperature of copper is lower than that of iron. Most likely it will be possible, but instead of letting the iron flow out of the bath, you do it with the copper.

TLDR - Article Highlights https://www.kontxt.io/document/d/EhrawY-veGAyWlYU8-98OI5FKC3...

Reply on news.ycombinator.com