85 comments

[ 3.0 ms ] story [ 170 ms ] thread
I don’t get this. We already use direct reduction using about half hydrogen, and that can be increased to over 90%. Producing ammonia via the Haber process means losing 40% or so of the energy (and potentially more as you convert it back), so why not just use hydrogen directly? Simply because moving hydrogen is harder than ammonia? I think it makes way more sense to just make and use the hydrogen on-site.
You said it: transporting, storing, and otherwise handling ammonia is easier than H2. Amd handling a lump of coke is even easier, which is why we started there.
If hydrogen is electrolytic you can as well make it on site. Steel mills are continuous operations and there is not much need to pool/smooth demand.
Shifting electric prices due to wind power is a strong motivation to pool it though.
So far it's easier and more efficient to reserve/contract some hydro (dam) storage than store/pool hydrogen for continuous operations.
That sounds very location dependant and not necessarily feasible at all given higher and higher share of wind power being used. That hydro will be needed during periods of low wind to stabilise everything else.
The whole point of the grid is to de-localize power.

Direct hydrogen reduction plants pretty much don't exist yet, might as well build them where electricity can be stored efficiently. Some of the biggest hydro plants were built specifically to power aluminium smelting nearby.

i think the question is how far you are shipping the hydrogen. If, say, people are making hydrogen in (say) the Middle East and shipping it to (say) Europe then the overhead of liquifying or compressing H2 is on the same order as converting to ammonia. In that paper they demonstrate that you can just use the ammonia directly to reduce iron and not have a separate system to convert it back.

If you have a big wind power or solar complex like the ones being built in the North Sea or Australia you might be better off using hydrogen directly.

I would guess that even better than shipping ammonia or hydrogen to European steel plants would be to build new steel plants near the hydrogen producers, wherever they may be, and shipping iron ore there while shipping steel back out. Since iron ore and steel are much denser than either ammonia or hydrogen and do not need pressure vessels or chilling they can be shipped at lower speeds (save transport energy consumption) and save money too.
I'm asking this 100% from a place of curiosity because I don't know the answer. From iron ore to steel, how much waste is there? If the waste fraction is large, people might balk at the idea of either leaving that waste behind in the hydrogen-host country or burning fossil fuels to ship it around to have a carbon-free extraction process.
Anywhere you have water you have hydrogen.
Interestingly, the paper mentions that the reactions do produce (some) hydrogen, which could perhaps be recaptured.

However, a bigger concern upon a glance is that this process does produce NOx emissions...

NOx from big industrial processes tends not to be an issue. Catalytically reducing it to N2 and O2 is easy, and at the same time, you get out 'free' high grade heat, which there is usually some use for elsewhere in the plant.
> Interestingly, the paper mentions that the reactions do produce (some) hydrogen, which could perhaps be recaptured.

The hydrogen is used for reducing the iron oxides, that's the whole point of the process! That is,

FeXOY + H2 => Fe + H2O

> From iron ore to steel, how much waste is there?

Depends on how we define 'waste' and at what part of the process.

When they start with the rock from the ground/pit, the rocks/etc are often crushed, running through some sort of slurry while basically separating the 'ore' out from silicates/etc that will be around them. I'm guessing this is already done close to the site, since transport cost could be fairly high even by past standards.

What you wind up Iron ores that are considered 'worth' mining, the actual Iron content is anywhere from ~48% to ~72%. They'll typically have Oxygen, Possibly also Carbon or hydrogen as the 'impurities'.

So, there's still a lot of potential waste in transporting all of that.

[0] - Also, that would theoretically be useful in filling the pit back up, one would hope. But not sure on that one.

I work in the steel industry so can provide some info here.

In the mining of iron ore there is a step called "beneficiation" which involves the wet processing of minerals - this is not unique to iron ore and is common to many mining operations for a wide variety of materials, wet processing produces a waste slurry known as "tailings" which are typically stored in a dam or similar (https://en.wikipedia.org/wiki/Tailings_dam) - tailings dams have been the cause of major environmental issues in the past. Such as the Brazilian Dam failure in 2019 which killed 250+ people.

After beneficiation Iron ore is transported to a steelworks where it is smelted. The smelting process produces a further waste product "slag" (https://en.wikipedia.org/wiki/Slag) how much slag is generated depends on a few factors such as the efficiency of the plant and the quality of the raw materials. Steelmaking slags can have some further use, for example it can be used in the concrete and cement industry and as a road base but how much if any of it is recycled very much depends on the country and the steelworks in some countries, particularly those with looser environmental restrictions slag is sent to landfill.

This is how Iceland exports the bulk of its hydroelectric energy, but they use aluminum
I was going to correct you with geothermal, but then I looked it up and it turns out that the majority of Iceland's electrical production is indeed hydro rather than geothermal (TIL).

In finding the source for this, https://en.wikipedia.org/wiki/Energy_in_Iceland aluminum was the thing of note there. I was aware that this was the main power consumption (tangent to reading Artemis by Andy Weir).

> ... This trend continued and increases in the production of hydroelectric power are directly related to industrial development. In 2005, Landsvirkjun produced 7,143 GWh of electricity total of which 6,676 GWh or 93% was produced via hydroelectric power plants. 5,193 GWh or 72% was used for power-intensive industries like aluminum smelting.

Yes! It's the similar with Norway. One of the largest exporters of oil, but entirely hydroelectric for electricity generation. It's the fjords!
Anhydrous ammonia’s volatility is on the order of that of propane/LPG (although a lot more hazardous to inhale), so the containment is easier than what the words “pressure vessel” might evoke.
Exactly, I struggle to see any situation, barring mismanagement or disaster, where you should be shipping tankers full of hydrogen byproducts like ammonia and losing most of the energy in the process.

We are currently importing all the natural gas in Europe, and both the price and carbon footrpint are more than double of what pipeline delivered from Russia.

You could build a pipeline from the middle east to Europe for hydrogen. We are already building powerplants in Sahara to export energy to EU. But I do not see why you should ever need to.

Iron ore and Aluminium ore is literally everywhere. We could move all of primary metal refining close to equator for solar power. China is close enough to equator, and produces huge quantities of Iron. Australia could be producing iron, they have plenty of sun.

Or Europe could produce hydrogen in the summer and store for the winter to keep refineries running.

Hydrogen, being so small, penetrates directly through the molecular structure of steel and in the process causes embrittlement of the steel. Is a long distance H2 pipeline a solved problem using alternate materials?
It is not. Thankfully the electricity required for hydrolysis can either be moved around or created sufficiently close to a steel plant.
thats one reason why ammonia is a good choice. the storage requirements are much kinder to the transporter
But the mass of the extracted oxygen exceeds the mass of the ammonia used, and we're only talking about 250 PSI to liquify anhydrous ammonia at room temperature. A large pressure vessel's mass is going to be negligible compared to the mass of the ammonia it holds. So, the same displacement ship traveling at the same speed can supply the production of more steel if you ship the ammonia to the iron ore instead of the other way around.
From this (https://www.sciencedirect.com/science/article/pii/S095965261...), they estimate a scaled up process requiring ~50kg of hydrogen to generate 1000kg of steel (which would be about 1500kg of iron ore).

50kg of hydrogen should be equivalent to like ~600kg of ammonia (if I did my math right). So there's a substantial mass difference in one way shipping of 600kg of ammonia versus two way shipping of ~1000kg of solids to consider. In addition, since the tanker is empty on the return trip, if we were willing to play the travel time game, we could reduce tanker return speeds to try to compensate for running faster while carrying.

So I don't think it's completely cut and dry that shipping the iron ore/steel makes more sense.

Thanks for the numbers! You and KMag have changed my mind. This is a lot less mass to ship around.
I think it's more rational to ship the hydrogen as methane, rather than ammonia. The energetics are comparable IIRC (Sabatier for methane, Haber-Bosch for ammonia, but it's more or less the same kind of high-pressure moderate-temp chemistry pipeline). Methane from CO2 + H2 vs. ammonia from N2 + H2, it's just that the latter technology has seen more research and investment.
The ammonia has a purpose on its own, as fertilizer. The nitrogen used to make it, is also freely pulled from the atmosphere
It’s also great for making explosives.

What?

A knife is great for spreading butter on toast.. also really good for stabbing as well.

your point is kinda silly, if thats the point you were trying to make.

>90% of the world's iron ore comes from very sunny places. Easier to just ship steel rather than iron ore and ammonia. For the last 10%, move the iron ore the other way (or still use local solar energy, because it is still cheaper than shipping ammonia).
> Easier to just ship steel rather than iron ore and ammonia.

Australia has a lot of high grade coal, which can be used for steel making (unlike low grade coal). And Australia has enormous quantities of high grade iron ore (as in you can weld to the rocks found in some parts of the country). Yet to the frustration of many Australian's we don't make steel. Instead we export the coal and ore to countries that make the steel, and they export the finished product back to us.

Easier to shift the blame for producing 10% of the world's emissions with 0.3% of the world's population that way.

Plus doing anything else might see Australia benefiting from their natural resources. Last time a government tried that, the CIA and Britain delivered a new one.

This isn't using ammonia to transport hydrogen; the nitrogen is what's reacting with the iron.

And yes, ease of transportation and reactivity is a big motivator.

From the abstract:

"Ammonia is an annually 180 million ton traded chemical energy carrier, with established transcontinental logistics and low liquefaction costs. It can be synthesized with green hydrogen and release hydrogen again through the reduction reaction."

Also: "The authors show that ammonia-based reduction of iron oxide proceeds through an autocatalytic reaction, is kinetically as effective as hydrogen-based direct reduction, yields the same metallization, and can be industrially realized with existing technologies."

>the nitrogen is what's reacting with the iron.

Only in a supplementary manner to form a nitride coating as rust proofing - the primary reducing agent is hydrogen, forming water.

relevant snippet:

>>The nitride formation is another key advantage of ADR, as nitriding improves the aqueous corrosion resistance of iron.[29] The nitride passivated the otherwise highly active reduced iron, offering a safety-critical benefit for handling and logistics. Otherwise, for the downstream processing of the reduced material, the porous sponge iron is prone to re-oxidation and strong exothermic reactions with oxygen or moisture due to its high surface-to-volume ratio (typically above 40 vol% porosity[4]). Thus, the sponge iron produced by HyDR must be compacted into hot briquetted iron to reduce the porosity for shipping and handling, which is not necessary with ADR.

> Producing ammonia via the Haber process means losing 40% or so of the energy

This seems wrong. I believe that green HB process would be 80%-90% efficient since the heat is re-used? I have read conflicting papers.

Of course, HB is still a capex-heavy process.

Maybe the 40% number is for fossil methane to ammonia?

I also think so. Rather than move hydrogen, move the electricity or the iron. Produce hydrogen on site and even on demand to get rid of most of the need to store/transport it.

As to cost, who can quantify me the risks of having and transporting so much ammonia?

What about electrolysis?
The question is "what is the most efficient way of getting hydrogen at the source?"

Shipping it directly has problems with hydrogen being such a small gas and it is damaging to the vessel that it is transported in ( https://en.wikipedia.org/wiki/Hydrogen_embrittlement ).

Pushing electricity to the source needs a lot of power to work on the scale of steel production. Generating electricity from further away means power loss. Pumping water from the ground for hydrogen gives you ground water depletion (a serious issue in many parts of the world).

Ammonia is already a well understood compound for shipping. It doesn't pose the losses of shipping hydrogen and is already something needed in various places (fertilizer production).

Note that part of this is indeed doing electrolysis - at the location the ammonia is made rather than at the steel foundry. ( https://en.wikipedia.org/wiki/Ammonia_production#Haber-Bosch... )

In https://onlinelibrary.wiley.com/cms/asset/2299bce2-54a9-43ce... you will note that the cost of hydrogen production is the same - but the other costs of conversion and import / export terminals is much more.

> The question is "what is the most efficient way of getting hydrogen at the source?"

Who said anything about hydrogen? You can directly reduce iron electrolytically: https://www.siderwin-spire.eu/

From article:

" Important examples are hydrogen-based direction reduction (HyDR),[4] hydrogen plasma smelting reduction,[5] and various electrolysis processes (e.g., molten oxides’ electrolysis,[6] molten salt electrolysis,[7] waster-assisted molten salt electrochemical reduction,[8] and electrowinning of solid iron from aqueous solutions[9]). Among these alternatives, the HyDR approach has today reached the highest technology readiness level (TRL 6–8) and is currently being deployed at industrial scale.[2, 10] In this process, green hydrogen should be ideally used, i.e., hydrogen that has been produced using renewable energy sources, generating water instead of carbon dioxide as redox product.[4]

"

This article is trying to make HyDR more green since it's already deployed more at an industrial level compared to alternatives at the moment.

Reduction of iron ore with carbon monoxid in closed loop:

• Decarbonisation of BF-BOF through thermochemical closed carbon looping.

• Demonstration of mass and energy flows of thermochemical BF-BOF system.

• 88% emissions reduction of UK steel industry through £720 million investment.

• Decarbonisation without retiring of existing BF-BOF, reducing stranded assets.

• After 5 years, £1.28 billion savings and total UK-wide emissions reduction of 2.9%.

“if the thermochemical closed reactors were exclusively powered by electricity, it would require 607 kWh/t liquid steel.”

»Cost effective decarbonisation of blast furnace – basic oxygen furnace steel production through thermochemical sector coupling« -- https://doi.org/10.1016/j.jclepro.2023.135963

The UK steel industry is 0.1% of the UK economy, at around 7 million tonnes per annum.

China producuces a little over 1 billion metric tonnes per annum

By what measure is this not futile?

So the UK, with about 4.7% the population of China, produces 0.7% as much steel, i.e. the ratios are within an order of magnitude. And presumably what works in the UK will work in China too.
Almost every individual action by a single emitter makes a tiny contribution to AGW.
If the technology is developed and proven in a smaller economy, then it can be expanded to the bigger one by said smaller economy.
Doesn't China have a free pass with their emissions through to 2030, due to their "developing country" status?

Pretty sure that's the case, though they have to start reducing their emissions after that like the rest of us.

>In conclusion, this paper aims to demonstrate the first principal calculations of coupling a thermochemical carbon dioxide splitting cycle with a steel production facility for cost-effective steel decarbonisation.

It's an interesting idea. But let's be clear: it's an idea. The scientific literature contains a lot of good ideas.

By contrast the OP paper is actually an experimental demonstration of reduction with ammonia. A bird in the hand is worth two in the bush.

Lots of refs here to the Haber-Bosch process in creating the ammonia... for those interested in a break down of what it is - https://www.carbonbrief.org/qa-what-does-the-worlds-reliance...

...sort of feels like kicking the can down the road if this is how the ammonia is generated for "Green Steel".

From article:

" Currently, ammonia is synthesized through the Haber–Bosch process by converting hydrogen and nitrogen into ammonia. In this process, hydrogen is mainly produced via steam methane reforming. This fact makes the process of fossil-fuel-based ammonia synthesis very carbon dioxide intensive, accounting for ≈1% of the global carbon dioxide emissions.[11, 16] Yet more sustainable ammonia synthesis pathways are under development to mitigate carbon dioxide emissions in the ammonia industry.[11, 16] For instance, the electrically driven hybrid Haber–Bosch process (via replacing the steam methane reforming by water electrolysis to obtain green hydrogen and coupling with an ammonia synthesis reactor in the Haber–Bosch process) or direct electrosynthesis using renewable energy (via nitrogen reduction reaction) enables the production of green ammonia.[11, 16] "

Not sure how that is kicking the can.

I work at a company that is creating better electrolyzers for this process. Ammonia producers are clamoring for this technology. It works, and it is happening!
same :-)

This is not just for green steel, it's largely used for fertilizer and some other industrial purposes. Making it green is going to be big business!

Could ammonia be used for stable seasonal storage of energy?

The abstract suggests that ammonia is much nicer to transport and store than hydrogen. If we can use it to store summer solar to run winter heat-pumps with a low roundtrip efficiency but even lower costs per kWH of capacity it could really help.

Ammonia can indeed be used as a longish term energy storage. It's not as great as methane (that can be pumped into underground caverns), but it liquefies at room temperature at just around 9 atmospheres or at -33C at atmospheric pressure.

Though it's pretty poisonous if it leaks.

It's corrosive to humans, toxic to aquatic life, but once you dilute it, humans survive easily.

The dangers are in confined spaces.

it's caustic if it leaks, but in the environment it quickly reacts to produce nontoxic products
It could work theoretically. The only question is if it is economical to do so. And the answer to that is probably not. It ranks pretty low on the list of possible solutions in terms of cost and efficiencies.

Hydrogen and ammonia (you typically generate one to generate the other) are interesting as a fuel in some use cases (anywhere the weight of lithium ion batteries is a problem basically). Main use cases seem to be shipping, and maybe long haul aviation. Probably not for road transport (battery electric seems adequate there for most vehicle categories).

But as a battery/energy storage solution it makes less sense. The round trip from solar/wind energy to hydrogen to ammonia and back to electricity loses most of the energy in the process. It's doable but there are probably more efficient and cheaper ways to store the energy. You lose about half (at least, that's a super optimistic percentage) of the energy creating the hydrogen. Then some more creating the ammonia. And then some more converting that back to electricity. It's pretty easy to waste less energy than that.

For heat, simple thermal mass is very efficient, low tech, and has already been demonstrated to work for seasonal storage. Throwing away half the energy to create ammonia simply makes no sense. All you need for thermal mass is some basalt, sand, etc. with a lot of mass, a container to put it in, and some cheap way to insulate it (wool would do the job). Heat it up in the summer, extract heat in the winter. It scales. The raw materials are dirt cheap (because they are literally dirt), the complexity is low (pipes, plumbing, insulators, sand/rock).

That's not quite correct.

Long term storage solutions are dominated by capital cost. You can use batteries to store electricity from day to night because you recoup part of your capital investment every day for ten years or so. But if you can only sell electricity once a year, then whatever profit you make it better be good, because you only get that 10 times in 10 years. The alternative is to have dirt cheap capital cost.

Pumped storage works where the lake and dam already exist, because the capital storage is essentially zero.

Everywhere else, nothing works. Batteries don't work, chemical storage (like ammonia or hydrogen) doesn't, compressed air, or molten salt, nothing works.

What will work is a way to "cheat". If you can mimic the day-night cycle of batteries and make a profit every day, you win. The way to do that is by long-distance transportation: you make ammonia in Australia, or Morocco, or Saudi Arabia and sell it in Japan, China or Europe. You may incur round trip losses of 80% or more, but if the price you pay for one kwh is one cent, and you sell for 10 cents, you can still make a profit.

Haber-Bosch seems to be only feasible at large scale. It would be great to make this scale down and be more efficient.

It's the secret to a whole lot of ways to combat rising carbon.

What's interesting is that H2 + N2 -> NH3 reaction is thermodynamically favorable, so in theory with a good enough catalyst it can be driven at mild conditions.

And this has actually been achieved back in early 2000-s! But the catalysts are very finicky and they get poisoned too quickly for industrial use. Additionally, it'd be nice to be able to use water instead of hydrogen directly.

There are some interesting developments in this area, like this one: https://www.nature.com/articles/s41467-022-34984-1 - they synthesized ammonia using visible light as an energy source.

And yet the creation of hydrogen through electrolysis of water is both well understood and potentially green if solar and wind electricity is used for electrolysis. This would also bypass either the capital intensive requirement to moderate solar and wind electricity with battery energy storage and/or the losses in transmitting solar and wind electricity over the grid.
I don't get it. Why not just use methane?

It can be synthesized easily, we already have infrastructure to ship it around the world (just ask Putin for his opinion on that), and it's much less hazardous than ammonia.

Moreover, it doesn't require plants to switch to green methane right away. It can be done gradually and in a distributed fashion.

Drilling for methane adds GHGs to climate change.

For DRI, use a solar power tower to directly heat ore with "syngas" made similarly with solar heating and a renewable carbon source.

Such a setup can approach closed-loop production and emissions capture.

Sure. But we also can synthesize methane from captured carbon dioxide and green hydrogen. It's still more expensive than fossil methane, but it's getting cheaper.

But we can start transitioning factories to use methane right now, starting from fossil methane and slowly moving to greener sources.

It's not particularly unlikely that we'll have to tear down or repurpose all that methane transportation infrastructure once we prioritize that global warming caused by leaks in that infrastructure.

Transporting ammonia doesn't have the leaks-make-global-warming problem

(comment deleted)
To eliminate rust on steel rather create it [edit: steel]: phosphoric acid. In gel form, it's called "naval jelly."
>Sustainable

Not in the financial sense of the word.

There is already direct hydrogen reduction, why bother with ammonia? At least hydrogen can be from solar electricity, not so much for ammonia.
Regarding your second sentence in particular, the Haber Bosch process already make ammonia out of hydrogen. That it usually take hydrogen produced out of methane is something that can be changed.
"every MWh of renewable electricity put onto the grid would displace ~3MWh of fossil fuels being burned to supply grid electricity and eliminate over a tonne of CO2 emissions." - https://johnmenadue.com/the-green-hydrogen-myth/

Electric arc furnaces are rife with issues.

All the ESG "let's minimize CO2 emissions" stuff had me going long levered on nat gas.

https://bit.ly/kumarngbull

Have a look at $UAN, $FCG, and $TELL(Gamble) -- I went long levered in 2019 onwards.

If I have to watch the train wreck, I deserve to be compensated accordingly.

Steel data: https://twitter.com/datarade/status/1642716122343526400?s=20

To those who doubt my line of thinking:

Is there as ingle commercial steel plant that minimizes iron ores via hydrogen? Furthermore, we know that most of the hydrogen is produced by nat gas reformation.

To make zero carbon iron you'd have to do mass scale electrolysis and it would need to be powered by renewables, right?

https://www.industrytransition.org/green-steel-tracker/ - here's some 50+ green steel projects tracked, I check it periodically.

Cheers.

I do believe electric arc furnaces will get ESG ayatollah money and will scale over time even if counterproductive to basic scientific production metrics against feedstock requirements, throughput, quality, and industrial requirements.

Another rust post making it to the front page...