Seawater electrolysis by adjusting the local reaction environment of a catalyst | Hacker News Reader

Seawater electrolysis by adjusting the local reaction environment of a catalyst

nature.com

181 pointsCharlesW87 comments

I’m not very deep into the topic, but this new process might positively affect the economics and scalability of pilot plants like these as well:

https://techcrunch.com/2022/12/20/porsche-pumps-first-synthe...

With advances in nuclear fusion or other technologies that lower the price of CO2 neutral electricity, we just might be able to build plants that produce conventional fuel, using seawater and removing CO2 from the atmosphere at the same time. This would allow scalable production of conventional fuel, without affecting our drinking water or having to replace several billion cars all over the world.

I’m probably simplifying here way too much, but exciting developments in that space.

Synthesizing conventional fuel makes a lot of sense for airplanes. But the overall efficiency of internal combustion fuel in cars is awful.

Missing out on the potential air quality improvements would be disappointing too.

Fully agree with you on the significant efficiency advantage of electric cars vs. ICE-cars.

There might be several use cases, airplanes is one for sure, given the energy density per kg is too low in batteries vs. kerosene today. Pure hydrogen planes have big risks associated to them.

Potentially not having to drill into the ground anymore to extract oil for fuel production is another one. Producing conventional fuels and plugging them into the existing distribution system is beneficial in terms of how rapidly we could replace CO2-adding fuel with CO2-neutral fuel. The market would take care of this as soon as synthetic fuels are cheaper than “old fuels”. This is especially relevant if you think about the billions of people in the developing world that today cannot afford electric cars or the country doesn’t have the infrastructure to support electric cars. Batteries also still have cons in their production process (extracting lithium for example), and recycling is not solved neither. Again, this might be solved at some point, but scaling existing battery tech today to billions of cars would have its own side effects / feasibility issues.

Another benefit I could think of is we would reduce our reliance on certain countries that own most of the oil, geopolitically a very important factor as well.

Just thinking out loud here. Increasing our odds to potentially produce billions of liters of conventional fuel that might be cheaper than “old fuel” at some point, while taking CO2 out of the atmosphere, sounds promising to me.

> There might be several use cases, airplanes is one for sure, given the energy density per kg is too low in batteries vs. kerosene today. Pure hydrogen planes have big risks associated to them.

Is ethanol biofuel not an option? I imagine that actual production would be the limiting factor but it seems to me that should be a workable solution but would probably require significant retooling and redesigning of engines.

I'd guess we'll be looking at a combination of various fossil fuel replacements in the future. EV's are great, but it'll take up towards a century before combustion engines are a thing of the past.And I suspect that's being optimistic.

I seem to remember (don't have the source anymore, if someone can find it again I'd be grateful) that half of the total fuel produced worldwide is used to litterally take the other 50% to where it's needed. Delocalization of fuel production would be a BIG deal.

What do you mean by "efficiency" here?

Energy efficiency. Electric to gasoline to wheels is probably something like 75% * 50% * 30% == 11% efficient (roughly, very roughly). Electric to battery to wheels is approximately 77% efficient.

We'd need 5 times the power generation capacity, or more, just to support that use case.

The fact that we burn over a million tons of gasoline a day (in the US alone) is a secondary issue, but even by itself that's a hell of an engineering problem to solve. Aviation use is less, but still a lot.

Aha, thanks for clarifying; now I see what you mean. I'm not used to thinking about gasoline as something we spend electricity to produce!

It should be said that the overall efficiency of airplanes is also terrible. It's just that there's no drop-in replacement, and therefore people are looking at e-fuels for planes. But this doesn't change a simple fact: This is very inefficient and won't be cheap.

This seemed ... odd. Surely being crammed in like a cow in an abattoir truck gets airplanes something. So I looked it up https://en.wikipedia.org/wiki/Energy_efficiency_in_transport

For the curious: Boeing 747-400: 3.1 litres / passenger 100km, or 91 passenger miles per US gallon.

Better than ICE car with 1 person in it, and modern planes do better. It turns out the lower temperatures at 10km above sea level helps both the Carnot efficiency and drag. But the 747-400 travels just under the speed of sound, which hurts. Slower airplanes (but still faster than any land based transport), get you a 2 or 3 fold improvement.

Ironically we don't produce enough CO2 (in concentrated form) to do that on a large scale. This means that we might have to use direct air capture to get the CO2, which is quite expensive. It also begs the question: Since the CO2 is going to end up in the atmosphere anyway, then why go atmosphere -> CO2 -> synfuel -> atmosphere instead of fossil fuel -> atmosphere AND atmosphere -> CO2 -> underground storage. The second might be cheaper and the end result in terms of CO2 in the atmosphere will be the same.

You may be interested that Porsche basically dumped that project recently, though it didn't get a lot of media attention.

The current Haru Oni plant is a really tiny pilot plant, but it always came with the promise that it'd be scaled up by many orders of magnitude. Der Spiegel recently reported that the plan to build a large wind farm to power the next stage has been cancelled: https://www.spiegel.de/auto/volkswagen-vw-boss-oliver-blume-... (sorry, paywalled, but I don't have another source)

> advances in nuclear fusion or other technologies that lower the price of CO2 neutral electricity

We are still far from the point where anybody serious could reasonably affirm that advances in nuclear fusion would be expected to lower the price of CO2 anywhere in the foreseeable future.

That does not mean we should slow down R&D in nuclear fusion in any way but rather bet on the fast deployment of renewables if the objective is to lower the price of C02 neutral electricity in the coming decade(s).

Synthetic fuels will be nothing but greenwashing for the next 20 years: there is massive demand for industrial hydrogen (most fundamentally for synthetic fertilizers), and as long as that demand is not met with hydrogen from renewable sources, it will be supplied (as today) from from methane-steam reforming -- using the exact fossil hydrocarbons that synthetic fuels would replace.

But yes, synthetic fuels would allow people with money to go to Davos and run their yachts while claiming total greenery, so I have a feeling we will end up with green synthetic fuel and black synthetic fertilizers even if causes larger net emissions.

Ironically, as production of hydrogen increases and therefore the availability increases and price drops, more and more industries will want to make use of it, reducing availability as well.

Same with renewable energy; over here, even before an offshore wind park was completed, Microsoft swooped in and bought up its capacity for one of their datacenters.

> even before an offshore wind park was completed, Microsoft swooped in and bought up its capacity for one of their datacenters.

Why is that bad? Isn’t it great that such an investment is so quickly finding a customer?

It's very questionable if this helps at all. The energy needed to desalinate water is tiny compared to the energy needed for hydrogen production via electrolysis. You're saving small amounts of energy in exchange for a more complex process.

Here's a more detailed explanation: https://www.linkedin.com/posts/paul-martin-195763b_new-elect...

Here's the crucial sentence: "Desalination of seawater by reverse osmosis takes only about 0.035 kWh/kg of H2 produced, i.e. a trivial fraction of the energy (50-65 kWh/kg) required to produce a kg of hydrogen by electrolysis"

Compared to conventional alkaline electrolysis at around 70% apparently.

Doesn't brine electrolysis also produce chlorine, or has Factorio fed me some horseshit?

Pretty sure it usually produces chlorine gas and caustic soda, but I could be wrong.

Ah yeah right of course, it's something like Na+ Cl- H2O --> O2 NaOH Cl2 (sorry chemists, I know your eyes are probably bleeding)

But if they manage to get H2 from it then the sodium, oxygen, and chlorine have to bind into something else than usual I guess.

There is a proposed method to sequester carbon and reduce ocean acidification by doing this process, extracting the Hydrogen and Chlorine (or hydrochloric acid) for industrial purposes, and releasing the sodium hydroxide to absorb dissolved CO2 (carbonic acid) into sodium carbonate.

You're probably thinking of this paper from 2007:

"Electrochemical Acceleration of Chemical Weathering as an Energetically Feasible Approach to Mitigating Anthropogenic Climate Change"

https://pubs.acs.org/doi/10.1021/es0701816

It was mind blowing for me at the time! Subsequent approaches to accelerated silicate weathering like Project Vesta dropped the chemical component and just used mechanical crushing of rocks to accelerate weathering. The all-mechanical approach is less complicated and energy intensive.

If you're extracting chlorine to react it with rock it probably makes more sense to just react the rock directly. But if you can use the chlorine industrially it makes more sense not to involve the mining and transport of rock. There's currently a significant industry of chlorine and HCl production that isn't linked to a carbon sequestration process that we can supplant.

Needs massive energy input which would come from where? Also, massive amounts of toxic chemicals. Use biology to solve climate change by growing biomass (bio CCS) in the form of kelp, diatoms, and other high growth life (maybe GMO) to sink to the bottom of the ocean.

Nuclear plants can provide plenty of excess electricity as well as heat for thermal desalination.

Sounds like wishful thinking/hand-waving. Plus, the risks of and limitations of nuclear energy.

Biological processes are far more efficient at fixing carbon into biomass and using unlimited free solar energy. Bio CCS.

One could also release the chlorine into the atmosphere to destroy atmospheric methane. Elemental chlorine in sunlight is rapidly (within minutes) broken down into chlorine atoms. These atoms, being free radicals, efficiently extract hydrogen from methane molecules, starting a chain of reactions that converts the remaining fragment to CO2 and water.

You'd need a hell of a lot of chlorine to compensate for current methane injection, though.

No , cholrine radicals will also deplete ozone layer , Besides I don't think releasing highly reactive gases at any concentration into the atmosphere is a good idea, there can be other effects we haven't studied well enough.

They wouldn't make it to the ozone layer, if released in sunlight.

If you could release it close to methane concentrations. Otherwise it's going to react with anything close to it. Soil, lungs, vegetation, etc.

Methane is present throughout the troposphere -- it has an atmospheric lifetime of something over a decade and becomes well mixed. You'd want to release the chlorine in a sufficiently dilute and dispersed form that it didn't overwhelm the methane in the air it which it was released.

> But if they manage to get H2 from it then the sodium, oxygen, and chlorine have to bind into something else than usual I guess.

It's the same electrolysis process used today in swimming pools to generate chlorine. The hydrogen evaporates, the chlorine ends up in the water, and the chemical reaction ends up producing the same salt it started with. You don't have to add salt[0] and the chlorine pretty quickly evaporates and breaks down in sunlight if you don't put cyanuric acid in the water to bind it.

Mostly it could be ignored, though if the chlorine level got high enough it would kill the organic things.

[0] except for losses due to other reasons than chlorine generation

Not with this process, that’s the breakthrough.

Yea it produces chlorine. Nuclear subs need to use reverse osmosis first to get around that problem when they want to produce Oxygen from seawater. https://www.youtube.com/watch?v=g3Ud6mHdhlQ

This process avoids producing chlorine, and doesn’t require purification of the input either.

It seems quite promising.

If this new way is 100% efficient wouldn't it break conservation of energy to produce chlorine (and instead it should remain salt)?

(Edit n/m, 100% efficiency was now removed from the headline)

What mods are you playing that have chlorine?

Sounds familar to me. I think bobs, maybe angels.

It's one of those reactions that depends on concentration and acidity, basically not going to be a concern for seawater electrolysis. Making chlorine gas is a pretty involved process, the NaCl salt has to be highly pure, etc.

> "In all cell processes, the filtered brine is heated and passed through a bed of salt in a saturator in order to increase its salt concentration before feeding it to the electrolyzers. In some plants, the brine feed is acidified to improve the cell current efficiency. The acidification reduces the alkalinity, which would otherwise react with the chlorine in the anolyte compartment, forming chlorate."

https://knowledge.electrochem.org/encycl/art-b01-brine.htm

At neutral-alkaline pH with seawater you can get hypochlorite (ClO-), basically relatively weak chlorine bleach formation at 0.5-1%, this is used when seawater is used for industrial cooling systems (see nuclear power plants etc.):

> "A more realistic picture of the problems DSS faces can be gained from on-site hypochlorite generators, a technology established since the early 1970s for industrial water cooling systems. There, low concentrated hypochlorite is used to avoid the growth of marine organisms as they tend to foul equipment and worsen heat transfer. The hypochlorite is formed by direct oxidation of Cl- from the filtered seawater feed. The electrolysers are one compartment cells made of titanium, and the electrodes are comprised of a titanium core with a precious metal oxide coating (Ru, Ir, Pt)."

source: Hausmann et al. (2021). Is direct seawater splitting economically meaningful? (sci-hub)

Hydrogen from seawater directly seems pretty speculative, be interesting if it works, you could have hydrogen production at sea maybe. Scale might be an issue (giant barges with solar panels making hydrogen for fueling shipping?)

I only played vanilla, but my pool at home with salt system has taught me that. It uses a 48v cell with 3000 ppm salt to create chlorine to sanitize the pool water. Many pools now days use this kind of system rather than chlorine tablets or whatever.

As far as I can tell, this process doesn't make any electrolysis process more energy-efficient. It merely allows the use of untreated seawater, thus sparing possibly scarce freshwater resources.

It's not nothing, but it's also not a giant leap forward AFAICT.

I'd assume doing this process on salt water is a lot cheaper and doesn't stress fresh water supplies.

Agreed: another milestone not worth mentioning. I'm not sure why these trivial feats keep appearing on the front page.

My understanding is that the cost of producing hydrogen is not one of the major bottlenecks about why those vehicles are slow to be adopted?

I mean, good to have such breakthroughs, but this is not (right now) a huge limiting factor on why we don't drive more H vehicles, is that right? It's more the storage of it? And specialty fuel cells needed to power the cars?

There are many, many uses for green hydrogen unrelated to road vehicles. Long-term power storage (use cheap summertime solar to bank hydrogen for utility power in the winter), ammonia synthesis for fertilizer, replacing fossil fuels for high-temperature heat to drive industrial processes such as steel and cement production, possibly as fuel for long-haul shipping and aviation (either directly as hydrogen, or after conversion to other forms such as ammonia or hydrocarbons), etc.

Reduction of iron ore to metallic iron (DRI, "direct reduced iron").

> replacing fossil fuels for high-temperature heat to drive industrial processes such as steel and cement production

IIUC, 1 kg of hydrogen has close to the same energy as 1 gallon of gasoline - which has 120k btu.

It takes ~39 kWh with perfect efficiency to produce 1 kg of hydrogen (120k btu).

1 kWh = 3412 btu

39 kWh = 133k btu

Wouldn't you always be better off converting kWh to heat than kWh to hydrogen and then burning hydrogen?

I think the key word here is high heat, I'm not sure electrical heating can be used in the same places?

I believe the hydrogen is actually part of the chemical process for steel production, not just a source of heat.

For steel making, hydrogen is used as a reducing agent to reduce iron oxide to iron. I think arc furnaces are the usual heat sources if you're not using coal or natural gas.

I work in the steel industry so I can provide some background.

An Electric Arc Furnace (EAF) uses electricity and solid feed such as scrap metal and/or DRI (Directly Reduced Iron Briquettes) to produce Steel. The electrical arcs basically melt the solid feed producing liquid steel.

Note that an Arc Furnace needs highly metallic feed, Iron Ore which you dig up from the ground is an oxide (i.e. it is not metallic) it needs to be chemically reduced. You cannot feed ore directly into an Arc Furnace to produce steel. Arc Furnaces can recycle steel scrap into new steel but they are not suited to making virgin steel.

The most common way of producing virgin steel from Iron Ore is through what is known as an integrated steel plant which combines two processes

1. A Blast Furnace - which uses chemical reduction to produce molten liquid Iron.

2. A Basic Oxygen Furnace (also known as an LD converter) which injects oxygen at supersonic speed into the liquid Blast Furnace iron to remove impurities such as Carbon and Phosphorus. This produces liquid steel at the end of the process

Blast Furnaces use Carbon (Typically the carbon comes in form of coke, which is basically highly refined coal) for the chemical reduction of iron ore. CO2 is a byproduct of this reaction.

There are some alternates to a Blast Furnace Such as DRI (which uses a gas such as Natural Gas rather than Coke).

It is possible to chemically reduce Iron Ore using Hydrogen rather than Carbon (Thus avoiding CO2 as a byproduct) but for a variety of complicated technical reasons it is not as simple as just swapping the coke in a blast furnace one for one with hydrogen.

Hydrogen is widely seen within the industry as being the future of steelmaking, there is massive ongoing effort currently underway to develop capability etc the industry is very much going through a transitional period at the moment. I know of one plant in Sweden which is doing some pretty cutting edge work in this area currently.

Thanks for the detailed reply! Is hydrogen being used as the heat source and the reducing agent, or is it exclusively for reduction?

Inefficiency of producing hydrogen relative to storing energy in batteries is one of the major points that detractors of hydrogen cite.

yes but it is not and has never been an "EITHER / OR" scenario.

be wary of others that claim it is.

Grossly higher capital cost of batteries per kWh of storage capacity is one of the major points that defenders of hydrogen respond with.

You're having a dig at someone who's on your side! His point was that this improvement helps negate that criticism.

Appreciate you sticking up for me, but I didn't interpret the parent's comment as a dig. Just an attempt to add some additional context to the "debate", which is always welcome.

As for what side I am on - you're right that I am excited about hydrogen as an energy storage medium due to its high energy density, and am in favor of developing the technology further. The challenges are real, and I don't know if it will ever be practical for powering personal transportation (though I wouldn't rule it out either), but there are many other applications beyond just cars where that energy density could prove useful.

I do find it disheartening that battery vs. hydrogen has become some sort of holy war instead of viewing them as two complementary technologies.

Indeed. They are complementary. Batteries for rapidly cycling storage where efficiency is important, hydrogen for long term or rarely used storage where efficiency is less important than minimizing capital cost.

Yes but in a lot of instances the cost of the batteries will be greater than the cost of the electricity you store in them.

Isn't the issue with hydrogen more storage/transportation than production?

The production of hydrogen is already 70-80% efficient. The conversion back to electricity seems like a bigger bottleneck for vehicles at ~40-60%.

Wikipedia claims that fuel cells can hit 85% efficiency by using cogeneration to repurpose the waste heat, so to me it would seem that hydrogen is more interesting at grid scale.

With hydrogen, you're just trading the problem of storing energy for the problem of storing hydrogen, which always leaks and can penetrate and weaken metal storage containers, valves and so on.

Hydrogen and electric are complementary strategies, not competitors.

I think Toyota is working on a hydrogen combustion engine. Maybe this plus that could be interesting

There's potentially some merit there, since the expense of the catalyst (and its decontamination process) is a significant barrier to the use of hydrogen fuel cells. That said, I'd expect it to be dramatically less efficient than a fuel cell system.

One of the things I've seen Tata Motors talking about re: Hydrogen ICE as an intermediate technology is that the fuel purity requirements are much lower for an ICE than for a fuel cell.

I have my doubts that hydrogen combustion will ever be viable for consumer vehicles. The energy output is far too low, and storage is vastly too problematic.

They have a fuel cell vehicle - https://www.toyota.com/mirai/

Toyota (in collaboration with Yamaha) is developing a V8 hydrogen combustion engine [1]. BMW also developed a V12 hydrogen combustion engine back in 2005 [2].

"Engineering Explained" has put out videos on both, and neither of them sound very appealing to me. [3][4]

[1] https://www.topgear.com/car-news/future-tech/toyota-and-yama...

[2] https://en.wikipedia.org/wiki/BMW_Hydrogen_7

[3] https://www.youtube.com/watch?v=vJjKwSF9gT8

[4] https://www.youtube.com/watch?v=AouW9_jyZck

(a) There aren't enough hydrogen charging stations, even in California.

(b) I'd rather drive around with a Lithium battery than a pressurized tank of hydrogen.

(b) you don't seem to understand how hydrogen functions as a fuel. It can be burned but it can also be used to make electricity. "it depends"

(d) sintered metal storage of hydrogen for ..

(e) electric motors where the regen of electricity from H2 produces water as a byproduct of electricity from H2

(a) who cares about ordering?

(f) yes there is hydrogen combustion outcomes for H2 powered cars too and yes early demonstrators do H2 in tanks but the scale industry here isn't targetting Gaseous H2 tanked vehicles AFAIK.

If there's a pressurized tank of hydrogen and something pierces the car, it's kaboom.

If something pierces a lithium battery you at least have a few seconds to get out.

If there's a pressurised tank and something pierces the car. its not necessarily kaboom. H2 is lighter than air. It disperses. Its burn patterns are different. I don't doubt its scary but the assumption it's worse than other fuels is an open question. It might be better than LNG in some circumstances.

As I said, the goal is not H2 tanks. Did you read that? do you understand how sintered metal H2 storage works? or NH storage? (NH is btw, also scary bad. people used to die when ammonium refrigerators leak)

Here's a video released by Toyota where they shot a hydrogen tank used in the Mirai with a 50-caliber bullet [1]. The bullet pierced the tank, but it didn't go kaboom. The hydrogen just leaked out.

Now personally I think that a hydrogen leak could also be very dangerous. But so can a gasoline leak or a lithium ion battery if exposed to the air.

[1] https://www.youtube.com/watch?v=jVeagFmmwA0

In a car crash, there are frequently showers of sparks from metal bits grinding against pavement. I'd like to see that test performed with an ignition source near the tank.

Here's a video from 2003 where BMW stuck a hydrogen tank and a gasoline tank in a fire and compared the results [1]. Not exactly what you were asking for - but demonstrates hydrogen being released at high pressure and ignited.

The caption from BMW is:

> Fire behaviour test, comparing a petrol tank and a tank filled with liquefied hydrogen. The heat from outside causes a rise of pressure inside the hydrogen tank. The hydrogen gasifies and emerges through a safety valve into the air, where it burns off. From the fuel tank liquid petrol emerges and causes a surface fire. Statements by Dr. Joachim Wolf, Linde AG "What that means for the car manufacturer – or for cars in general – when we transfer this hypothetical situation into reality: an accident happens, petrol runs out, and a car drives into the flames. It’s not very nice when a car is on fire and people have to be rescued. That’s the case with petrol. With hydrogen, if the fuel escapes then it disperses upwards. That probably offers much better options for rescuing people who may be trapped in the vehicles.""Hydrogen is no more dangerous than petrol: I think this test shows that. We see a clear blue flame that doesn’t produce as much heat as burning petrol. Hydrogen is not more dangerous, but it’s also no less dangerous than petrol. It simply poses different potential hazards."

I more or less agree with the statement that Hydrogen is not necessarily any more dangerous than the cars we are already driving around, but poses a different a different set of potential hazards. That being said - I don't think I would like to be an early adopter of hydrogen cars.

[1] https://www.press.bmwgroup.com/embed/Ufc3E4x91b

The big industry aim is to cheaply produce hydrogen, transport and store as ammonia, convert back and use in large scale regional baseload power stations to buffer wind and solar.

ie. Still an electric car renewable future, but one with a more reliable supply of less fossil fuel baseload generation, less battery farms.

The future is mixed hybrid.

Storage, transport, and fueling are all big problems for personal vehicles.

It may still be useful for energy storage and industrial processes (see sibling comments). Better storage for excess solar/wind generation would be very helpful.

Perhaps it would also be useful for limited range applications where you wouldn’t need lots of spread out stations like for vehicles at a port that do a lot of work but never get very far from home.

Does 100% efficiency here mean that the energy of the hydrogen equals the energy used to produce it?

Unlikely. What with thermodynamics. I'd guess it's something like "nearly" hides some sins and ignoring things other than oxygen that result hide the rest.

I would think it means that (nearly) 100% of the energy input is used in useful chemical reactions doing work as opposed to generating waste heat or something else.

My understanding is that a perfectly efficient device would take ~39 kWh to produce 1 kg hydrogen [1].

So I'm imagining this is close to that?

IIUC, 1 kg hydrogen = ~33.6 kWh.

[1] https://wernerantweiler.ca/blog.php?item=2020-09-28#:~:text=....

Current battery storage costs are $143/kWh.

The LCOE of solar is <$0.04 now.

A hydrogen powered combined cycle peaker plant would have ~60% efficiency.

Meaning, you should be able to produce solar-hydrogen fuel for $0.04/0.6/(33.6/39) = <$0.08 per kWh.

Why isn't anyone doing this?

Natural gas costs >$3.45/MMBTU = $0.01 per kWh.

That sounds like it's 8x more expensive.

But the wholesale cost of Natural Gas electricity is $0.18 per kWh - meaning going to Hydrogen shouldn't even increase that by 40%. But in the EU, it's already higher than that!

There are short-term issues like capital costs and a need for contracts to be set up first. These things can't happen instantly.

But beyond that, it is entirely backwardness in thinking that is driving the opposition. Many people simply reject the evidence and are convinced that such low costs are impossible. It is a repeat of those who thought wind and solar could never be cheap. They will be embarrassed in the same way too.

Not much there to comment on.

The actual paper is at: https://www.nature.com/articles/s41560-023-01195-x

DOI for those who want to search other databases: https://doi.org/10.1038/s41560-023-01195-x

Reply on news.ycombinator.com