Making steel without emitting CO2 | Hacker News Reader

Making steel without emitting CO2

science.org

236 pointsttfkam117 comments

The chlor-alkali process is responsible for something on the order of 80%+ of what we think of when we say "toxic waste" - dioxins and polychlorinated compounds. I don't think it's going to be a panacea for iron refining to switch to electrolysis. That said, it's very interesting research that probably works for a lot of other non-bulk metals that need to be reduced to elemental metals.

> The Oregon group’s setup generates essentially as much chlorine gas as it does iron

I'd much rather be dealing with industrial quantities of CO2 than Cl2, if that stuff gets into the atmosphere in bulk I doubt it'd end well. The article seems to suggest that the chlorine can be solved commercially so the argument seems to be that it is economical to sequester the chlorine rather than the carbon dioxide.

There are some obvious risks of being left with large amounts of Chlorine to try and find a home for, but the idea is at least plausible on face value.

I mean people are intentionally making industrial quantities of chlorine, via electrolysis, and have been for an century.

I did some quick searches thinking I'd find we produce vastly more iron than chlorine, and we surprised to learn that we only use sixteen times more iron.

Being so reactive I imagine. If chlorine were much much cheaper we would find many more uses.

https://en.wikipedia.org/wiki/Polyvinyl_chloride

Big use-case, problem is that in case of fire lots of dioxins are produced.

Yes, and chlorine is the stuff that no one wants, so it's turned into polyvinylchloride, which isn't really useful unless with copious quantities of plasticizer.

For something nobody wants it costs several hundred dollars per ton to buy

The price of chlorine though is much lower still and it has been very stable until recently. How elastic is demand for PVC?

I imagine Cl2 being way easier to capture by virtue of its weight. It is also a very useful byproduct for downstream applications.

It's also incredibly reactive, so it's going to be a lot easier to recover in general.

Isn't Cl gas commercially valuable, people are making it. edit: the article says that this process would make too much...

The process results in a lot of sodium hydroxide as well. Mix it with chlorine and water, and you're back to what you started with: seawater.

And energy! Maybe you can even recover some of the energy from your new warm seawater.

but at scale? how much cl2 are we talking about?

> I'd much rather be dealing with industrial quantities of CO2 than Cl2

The costs of CO2 - the real costs, not the subsidized ones - are extraordinarily high, possibly leading to the worst global catastrophe in human history. What is the cost of Cl2?

Parts of your lung flying out of your mouth. It was the first chemical weapon for a reason.

Is gassing the frog or slowly boiling it crueler?

With the passing of time the answer might not be so clear.

Well, you gas your frogs and I put mine in 2 degrees warmer ponds and we see which one are happier ?

Time for some dark humour! After careful analysis of the situation...

The frogs in the 2 degree warmer pond will be complaining a lot more than the other frogs, so that must be worse.

Last I checked (many years ago) the estimated long term cost from climate change due to all CO2 so far released was only in the low trillions (making WW2 a much bigger catastrophe), do you have an updated figure?

You talking about the area under an initial section of a hockey stick curve.

Swapping all the CO2 in the atmosphere for Cl would kill us all within 30 minutes (at 400ppm). It's way more dangerous.

I realise this is a silly thought experiment but the point of comparison should be, at the very most, the human industry generated CO2, not all natural CO2. The CO2 from this particular process is probably more relevant and since this process generated 1 ton of Cl per ton of Steel and the normal process generated 3 tons of CO2, we should give it a 66% discount.

> I realise this is a silly thought experiment but the point of comparison should be, at the very most, the human industry generated CO2, not all natural CO2

The pre-industrial levels of CO2 were around 280ppm, while the current CO2 levels are around 440ppm.

For Cl2, in occupational-health situations the permissible exposure is around 1ppm per hour, or 3ppm for 15 minutes. So the 160ppm we'd throw into the atmosphere in this thought experiment (or even the 52ppm at the suggested discount) is going to cause some pretty big problems pretty rapidly.

Arguably the most deadly part of CO2 is that it kills sooooooo slowly. Vent CO2 and nobody blinks. Vent Cl and everyone is gonna hate you immediately for immediately present negatives.

That's not the alternative option; nobody is suggesting swapping. Swapping CO2 for fission waste products would also kills us. Swapping it for H2O would drown us!

I think there's like 3 orders of magnitude more water in our atmosphere by weight than CO2, so a swap wouldn't drown us. But the point you were trying to make is clear.

Yes, but as long as we're having some fun with the concept: The problem with CO2 isn't the absolute CO2 in the atmosphere; most(?) of it was already there. It's the changes caused by the added CO2.

The problem with H2O wouldn't be the absolute amount, but the changes brought on by the added H2O. For one thing, it would rain pretty much 24/7 in Florida. :) (Yes, I know it's not at all that predictable.)

It won't necessarily accumulate though - there's a reason it was quickly abandoned as a chemical weapon. I would rather worry about its compounds, as it would quickly react with anything it touches.

You can avoid any CO2 production in the iron production process via replacing the natural gas inputs to the Direct Reduced Iron (DRI) method with hydrogen. To compare DRI with carbon vs only hydrogen:

with carbon and hydrogen from natural gas/coal gasification: CH4 + CO2 -> 2CO + 2H2 Fe2O3 + 3CO -> 2Fe + 3CO2 (exothermic in >50% H2 environment)

with water-sourced hydrogen: 2H2O -> O2 + 2H2 Fe2O3 + 3H2 -> 2Fe + 3H2O (endothermic, requires energy input)

There are two issues with using hydrogen, which can be overcome:

>"The energy balance of the shaft furnace is affected by the absence of the exothermic carbon monoxide reduction. . . Thus, it is necessary to add energy to the shaft furnace to carry heat in the burden."

>"The second issue is the resulting DRI carbon content; the DRI will have 0% carbon with pure hydrogen. The majority of DRI is used in EAFs, and EAF steelmaking practice generally employs carbon addition... Under current melting practices, it will be necessary to add hydrocarbons at some place in the process to achieve the desired carbon level... However, this added carbon will then be converted to CO2 in the EAF... Alternatively, carbon from a renewable source (like biomass) could be used."

https://www.midrex.com/tech-article/ultra-low-co2-ironmaking...

It really seems like any level of CO2 production from fossil fuels is now dangerously toxic too though? Swapping one problem off another yes but we have zero time to waste.

As all factorio players know, upgrading your steel production from coal-based to electric smelters is great to keep pollution down, and you also save on coal transport belts this way. Driving the smelters off solar panels is a good choice too.

As a factorio player, by the time I'm upgrading steel production I'm wondering whether I even care about pollution. The #1 benefit of electric smelters is that I can beaconize and fill it with prod mods, to get 140% the output at only the minor cost of a fuckton of pollution and electricity use.

In fact, the real score for reducing pollution here is efficiency mods - electric smelters plus steam engines apparently produce the exact same amount of pollution as just using lv2 smelters directly so you'll need those efficiency mods anyway (unless you've already switched to solar+nuclear), but you're better off putting your initial efficiency mods into your miners and oil.

Solar and nuclear are both a pain in the ass to switch to, though - nuclear takes stacks and stacks of copper to build and is a PITA to set up centrifuges and pipe down hydrochloric acid to uranium, and solar needs bots to place with any speed. It's just easy to slap down steam engines fed by either coal or solid fuel. So if you prioritize player-time over strict efficiency then the quickest way of reducing pollution is almost certainly efficiency mods into pre-existing buildings (i.e. miners and oil), and building electric furnaces will in-practice force you to build more power plants anyway. At least, in my experience. But maybe that's due to the prod-mods and beacons.

In the UK we like to pretend to not emit CO2 by closing down our own steel production, and importing it instead.

https://www.theguardian.com/business/2024/jan/18/tata-steel-...

This is how we will get to net zero and it's so sad.

This represents a long term critical security risk for Europe as a whole.

Steel is not magic. People don't get this excited over UK copper and lead production. The UK was a steel leader due to easily accessible iron resources, which are greatly depleted now.

It is worth considering the environmental footprint of our trading partners, but note that they are rapidly improving as well as renewables come down the cost curve.

Countries like China are just building more coal power stations, while the UK trashes its small remaining industry, pretending to save CO2.

https://energyandcleanair.org/publication/chinas-new-coal-po...

China also installed 200GW of solar. https://www.euronews.com/green/2023/12/29/europe-us-china-wh... ; the cost curve is turning and there are central directives from the country to use a minimum quota of renewable energy in production.

The UK industry is just uneconomic, it's not a green issue.

Yeah, its the same with most manufacturing in western countries. Then politicians can pat themselves on the back for "doing something to fight climate change" when they probably just lead to a net increase in emissions.

https://upload.wikimedia.org/wikipedia/commons/1/15/World_fo...

Same with plastic recycling, if it's packaged up and exported you get the tickbox. It's a great example of Goodheart's Law being applied, focusing on the local metric (reduced emissions in our neck of the woods! Look at us meeting the Paris agreement!) without solving or even exacerbating the underlying problem (steel production in countries with less carbon emission rules, added emissions due to shipping, and big ships use really shitty bunker fuel anyway)

It's a little misleading to suggest that "co-production" of iron and chlorine can meaningfully save energy. Annual production of steel is about 2000 Mt, of Cl2 is about 100 Mt (a similar figure for NaOH). So you'll get many times more chlorine than the economy can presently use. You could neutralize it by a variety of processes[1], but this isn't thrifty.

IIRC there was already a thread some time ago about electrolysis of iron ore in sodium hydroxide (as reusable solvent) where I was wondering about the cost of NaOH and how much would have to be replenished per cycle. But that process easily whoops this one.

1: Cl2 + MgO >> MgCl2 + O2, for one. (Stoichiometry exercise for the reader.)

If we release Cl2 into the daytime troposphere, it will be rapidly photolyzed to chlorine radicals, which rapidly react with methane. It would be a way of scrubbing methane out of the atmosphere.

Will we all turn blonde though?

Our lungs will probably be bleeding first.

It would be best to release the gas at low concentration (there's only a little methane in any parcel of air), and it would probably be released over oceans to avoid too much human exposure. A bigger concern could be making sure very little was transported up to the stratosphere before it reacted.

What are the products of the reaction ?

Methyl radical and hydrogen chloride. The former is oxidized further to CO2; the latter gets rained out eventually. If this rained out acid goes into the ocean, the sodium hydroxide can be released there to neutralize it, although first using the NaOH to capture atmospheric CO2 could be beneficial.

At what ratio does this process produce steel and Cl2? I'm guessing it's not a ratio of 20:1, but I don't know. The final paragraph of the article suggests it's not an awful ratio:

> Even so, scaling production to match industrial chlorine gas needs would still produce tens of millions of tons of CO2-free iron and chlorine annually

The article mentions it's 1:1.

> The Oregon group's setup generates essentially as much chlorine gas as it does iron, notes Iryna Zenyuk, a chemical engineer at the University of California, Irvine.

1:1 by mole, but not 1:1 by mass (which is what the industrial usage ratios are in).

Chlorine is 1.6 times lighter than iron. Depending on how many atoms of iron are reduced per one atom of chlorine exhausted, the balance may be not as starkly bad.

But likely replacing 100% of steel industry with this process would still produce more chlorine than can be profitably sold currently. Maybe it would do with replacing 30%, for starters.

Interestingly with steel being made with electricity instead of coal, the best places for steel mills are deserts on the equator with lots of solar, like North Africa, the Middle East.

The best place is next to nuclear power plant. Nuclear power is good because it has few restrictions on locations.

Using expensive electricity for an energy intensive process seems like a good way to lose money.

Nuclear puts out a lot of heat. Retooling iron/steel foundries to use that heat directly (instead of converting it into electricity and back into heat) seems like the way to go.

Color me skeptical. Waste heat from a nuke plant is very low quality compared to the extremely high heat needing to make molten steel. We aren't taking about warming potato greenhouses here.

...low quality heat? Nuclear reactors produce nothing but heat. And it's all just heat, how you transfer it is very independent of the production process

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

This is very wrong. Different processes require different temperatures. Iron and steel making requires very high temperatures. Heat from nuclear reactors cannot be used for this directly, as nuclear reactors do not get that hot, because they would melt. They are kept much cooler, usually at under 500 C.

Current reactor designs cannot get that hot without melting. I bet we could design a molten salt reactor that could safely get there, especially if electricity production is not a design goal.

I seriously doubt that. What would you make the reactor of, so that it withstands temperatures that melt steel? How are you going to move that coolant around? There are hard physical limits here. In reality, nobody would bother, because you can just make electricity and use it to melt steel instead.

Nuclear lightbulb engine - but for process heat!

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

Like it was not crazy dangerous enough. ;-)

I wonder how you’d transfer the heat without exposing the steep to radiation. I assume that would be bad anyways. I wish I paid attention more in physics, would love to know how such a system could be built.

A trivial heat exchanger does that, and every nuclear power plant already has one.

Today we avoid radiating electric generation equipment by running two steam loops through a heat exchanger; one that interacts with the hot reactor and a second that spins the generators. Perhaps you could do similar with molten salt loops? Might be tough finding a material that would work as an exchanger at the required temperatures though.

Just using electricity is probably cheaper and easier; large power plants are crazy efficient already.

Thorium-based reactors famously are constrained by the lack of such material which would withstand the contact with hot molten salt, especially somewhat radioactive.

As anyone with a boat knows, salt water is insanely corrosive and gets into everything. Even large container ships have to be regularly serviced.

I can only what hot molten salt does to industrial processes.

Optimally, next to a source of water that can be split into hydrogen, ready to be used for the chemical process producing the pure iron. (Not the process in TFA.)

An array of SMRs (small modular reactors) located at the steel factory could be used – and would be sufficient – both for heating and producing the electricity without interruptions caused by fluctuating prices or blackouts.

Wouldn't the need for labor, a supply of ore, and a means of delivering the product all argue for a port city? There aren't many railroads in those deserts. Or steelworkers.

There are plenty of ports in those areas.

Not inherently, nothing is cheaper than hydropower.

How reliable is hydropower now though? I’m living in a place which normally sees huge snow falls each year, this melting snow runs hydro, guess what ? We have about 15% of our usual base and next week it's going to be +10c average.

Not necessarily true at this point for new projects. IRENA indicates wind projects commissioned globally in 2022 had a weighted average LCOE of $33 USD/MWh. Hydropower projects over the last decade or so (2010–2021) were around $39 to $48. (2022 average was $61, but obviously recent market conditions were far from normal)

AFAIK, LCOE are known to be misleading when comparing electricity sources with different power characteristics and applications.

https://www.wri.org/technical-perspectives/insider-not-all-e...

Wind, with it's availability of ~60% is not massively different from other energy plants with their availability of 80 - 92%.

Solar is worse, but since most modern high quality tables include LCOE's for "solar" and "solar+battery" with very different numbers, it shouldn't be misleading.

Shipping of raw resources is cheap. Much of the bauxite used in aluminum cans in N America is shipped to Kitimat BC [0] to be smelted. Aluminum is notoriously electrically expensive.

0 https://en.m.wikipedia.org/wiki/Kitimat

Iceland is a player too IIRC.

> Shipping of raw resources is cheap

Cheap in terms of money, maybe.

Less so in terms of emissions: https://www.statista.com/statistics/1291468/international-sh...

Different method, but related goal: I enjoyed this episode on making carbon-negative portland cement for construction https://wondery.com/shows/how-i-built-this/episode/10386-lit...

Sublime Systems is working in the same field: https://sublime-systems.com/

I was more impressed by the Science article from two weeks ago about cost effectively making iron from the toxic residue of aluminum refining: https://www.science.org/content/article/zapping-red-mud-plas...

A group in Norway ran a pilot plant producing iron from sulfide ores back in the 1950's. Their paper quotes achieving 85% efficiency and around 4.7kwh/kg. It feels like with eletrowinning you can get the cost down to a few hundred dollars a metric ton. Scrap is around $200/ton. So not grossly uneconomic.

I think the devil is really in pre-processing ore. Whatever process can do that most cheaply is likely to win out. If you can figure out how to use waste like red mud so much that better.

A phys.org article goes into more detail about its cost effectiveness: https://phys.org/news/2024-02-team-green-steel-toxic-red.htm...

Germany is pushing "green steel" very hard right now: https://www.reuters.com/markets/europe/german-econ-ministry-...

There is a company called Helios Space that found adding Sodium into the process solves a couple of different issues. And it can work to refine other elements as well.

https://project-helios.space/green-steel/

I'm flabbergasted that there is no mention in the article of the H2 direct iron reduction process which today is the contender when it comes to decarbonization of steel making (especially now that we realize that there are tons of white hydrogen everywhere)

Could they add H to CL then use the HCL and NAOH to drive a acid-base flow battery* during peak loads? I guess it'd be less efficient because the hydrogen would have to come from electrolysis.

https://www.cetjournal.it/cet/23/100/112.pdf

Hydrogen is an indirect greenhouse gas with a Global Warming Potential(100) of around 11.6¹. Meaning it's over 11x worse than CO2 in the atmosphere! The main problem with this is that hydrogen molecules are extremely tiny, light and volatile, thus having a very high diffusivity. So unsurprisingly, hydrogen transport and storage has extremely high leakage rates. It's hard to alleviate because of the fundamental structure of the atom/molecules. Hydrogen can even permeate solid metals. A lot of the "green hydrogen" pitches don't take this into account.

¹https://www.nature.com/articles/s43247-023-00857-8

Maybe this is a dumb question, but looking at GWP figures, it seems like there's nothing listed that's below CO2. Are there really no hypothetical gases to put in the air that is better than CO2 for global warming?

Some gases don't cause any warming at all. Helium for example is not a greenhouse gas. What you looked at probably ignored those, we generally only compare problematic ones that contribute to the greenhouse effect. CO2 has been chosen as the standard to compare everything else to, I'm not sure if there's any molecule that does cause some warming but has a lower 100-year potential than CO2. Interesting question, couldn't find an answer on the quick.

Plenty. But honestly, nobody cares about them. For example, helium is near zero.

CO2 by itself is not particularly bad. We are just emitting a LOT of it.

There are some zero or negative GWP gases actually, like water vapor which hovers around the zero point plus or minus epsilon, HFE 365mcf3 which is below 1, and obviously argon and nitrogen are at zero. Mostly people don't record GWP values that are below 1 since e.g. any common constituent of the atmosphere contributes negligibly.

It's remarkably hard to find organic compounds that are less-bad than CO2 though. It really is the very bottom of a local minimum.

And how much does it leak? If the amount leaking to the atmosphere is a thousand of the amount of CO2 that would have been released for the same process, it's still a huge net win.

Comparing the effect of one kg of H2 with one kg of CO2 is irrelevant if we don't know how much H2 is released compared to the amount of CO2 for corresponding processes to produce the same amount of goods.

But how long does bare H2 remain in the atmosphere, given that it's pretty reactive and light enough to leave entirely?

H2 does not cause any warming at all, it's the reactions that do. So you're correct but that's exactly the problem. That's why I worded it as 'indirect' greenhouse gas.

If you are interested in this, check out https://www.hybritdevelopment.se/en/

They are running something similar with goal of larger industrial scale.

There is already a method that seems promising:

https://bostonmetal.com/green-steel-solution/

One technology that I'd love to see is direct molten oxide electrolysis. It requires some unobtanium materials, but it seems to be possible: https://corporate.arcelormittal.com/media/press-releases/arc...

Direct hydrogen reduction has been there for a while.

The only issue is cost.

Energy intensity is an issue with hydrogen steelmaking. To produce hydrogen by electrolysis at the large scale needed for steelmaking requires a massive amount of electricity. Ideally it needs to be green electricity as well.

So if there is a lower energy way of producing green steel it is at least worth exploring.

What are you suggesting, electrolysis baths in every solar-powered home?

Meaningful solar, hydro, nuke and wind power plants are quite concentrated (100MW-few GW) anyway.

If anything, solar/wind are less suitable due to intermittent output and would require hydrogen buffer storage.

Experimental methods and data analysis are usually scrutinized in peer review, whereas wild, unsubstantiated economic claims such as the above are not. I think we need a higher bar for making claims like the above.

Sticking points for me were, massive chlorine gas production (extremely toxic...) and requiring extremely high purity iron ore, with costs to purify it hand waved away.

Swedish tax payers are already all-in on “green steel”. But it seems we are betting on a different process, using hydrogen gas to strip the oxygen from the iron. Paradoxically, if this new electrolytic process works it’s probably a (financial) disaster for us…

EDIT: For clarity, based on comment below.

Why would it be a disaster if the hydrogen based "green steel" were a success?

No I meant it would be a (financial) disaster if a cheaper way of making “green steel” was discovered. As I understand it this process (the OP) could be such a process.

Well, they also produce hydrogen in the process, so you could ask them kindly for the rest gases, and be on your way. And now, you are not producing one unit of steel thanks to the electrolysis process, but two. Win! ;-)

Can we extract iron ore without releasing CO2 and if so, which countries are mandating this?

The article is saying ‘kinda’. The process works in lab conditions with highly purified ore, but produces an equivalent amount of chlorine gas and iron.

The article is about iron smelting without CO2. The question is about iron mining without CO2.

Fortescue (www.fortescue.com) is actively working towards that goal.

To the extent they have stopped taking tax credits for diesel fuel as they start trialing electric/hydrogen/ammonia fueled equipment. (while also researching into green steal production and hydrogen electrolyzer production.)

Fortescue says regenerative “Infinity Train” may be on tracks in two years

https://reneweconomy.com.au/fortescue-says-regenerative-infi...

They're also hiring Software Architects and embedded engineers (along with Engineers)

https://careers.fortescue.com/job/Perth-Senior-Software-Arch...

https://careers.fortescue.com/job/Perth-Embedded-Test-Engine...

Perth, Western Australia based.

Their founder/CEO gives some very impressive climate talks.

I'm from the same state, it's 3x bigger than Texas with a population of only ~2 million and a bit, most of whom live near the one city.

His grand grand folks landed here as domestic servents to settlers, and their kids made out okay: https://en.wikipedia.org/wiki/John_Forrest

Twiggy's a rare breed of pastoral based hard rock mining billionaire and marine biologist who gives a toss about the environment .. which helps given the intersection of family lands and native title claims.

I saw an interesting comment about EV mining equipment recently.

The manufacturer initially sold them as reducing CO2 globally, and also reducing harmful emissions in the mine itself, and saving fuel and fuel transport costs to remote mines.

The miners reported that the key benefit was actually that the machine spit out a lot less heat in a confined space undeground. Previously they'd need to spend an hour using AC to reduce the temperature to a level that humans could cope with before sending workers in for the next stage. With the EVs that time was saved and work could begin immediately.

Can we extract iron ore while releasing more CO2? Just to piss of these techie armchair expert environmentalists.

Reply on news.ycombinator.com