Hacker News Reader

Comment by pfdietz | Hacker News Reader

One can make natural gas from CO2 captured from air and with hydrogen electrolysed form water. But then one has to ask: why not just use hydrogen directly and skip the inefficiency and cost of direct air capture of CO2 and of making methane? If one is dead set on making a carbon-containing synfuel, why methane and not something more storable, like methanol or higher hydrocarbons? Or, why not use the hydrogen to deoxygenate biomass (including waste biomass like paper) instead of laboriously collecting completely oxidized CO2 from air?

View on HN

> why not just use hydrogen directly and skip the inefficiency and cost of direct air capture of CO2 and of making methane?

Broadly speaking, one key reason is that we've already got the infrastructure in place for using methane (and other hydrocarbons) whereas we do not have this for hydrogen.

Another point is that this really isn't an either-or proposition: if people want hydrogen, then the Terraform electrolyzer can in principle provide it.

In the UK, our gas infrastructure actually used to be used for hydrogen[1], so a lot of it should still work if we were to switch back.

[1] Actually "town gas" which was a mixture of hydrogen and carbon monoxide. But I imagine not including the CO would not be a problem.

Neglecting the CO will absolutely be a problem. The material properties required for handling hydrogen are very different than those for other gasses, especially methane. Increasing the concentration of hydrogen significantly increases the difficulty.

But if the pressure of hydrogen was the same as the partial pressure of hydrogen before, wouldn't it be the same? I mean, if the materials could stand that partial pressure, why wouldn't they be able to stand that pressure in the absence of CO?

Unless of course that actually took it below atmospheric pressure so all the forces went in the opposite direction.

Because the economics of storing and transporting hydrogen are just not great. You need a lot of energy and infrastructure to do it. And while the density per kg is great. The density by volume is absolutely terrible because hydrogen is the smallest and lightest molecule. In liquid and gas form it's nearly 3x the volume of methane (for the same amount of energy). If you are transporting it in compressed gas form (like is common to supply hydrogen vehicles), it's about 18x the volume of diesel. Needless to say, this doesn't scale for things like supplying fuel stations. In liquid form, it's better but we're still talking 3x that of methane and 4-5x that of diesel. And to keep hydrogen liquid, you typically boil some of it off continuously: it doesn't keep for very long. Methane has the same problem except the temperatures and pressures are such that it's not an enormous amount. Liquid hydrogen has to be chilled to a few degrees Kelvin. If you ship it around, you'll lose some non trivial amounts over the whole journey.

Most hydrogen produced today is used onsite to produce something else for this reason. Moving it around just adds cost and complexity.

Terraform seems to have made some progress with hydrolizers. But before people start popping champagne bottles, their synthetic gas still is an order of magnitude more expensive than natural gas. So, the value proposition of replacing one of the cheapest (but dirty) fuels with a clean alternative that is much more expensive is limited. And of course the process of using the captured carbon, releases all of it back to the atmosphere. For most use cases, switching to synthetic gas would push those use cases into the deeply unprofitable region. I.e. you'd be considering many other alternatives before committing to that.

Green hydrogen at the price levels they are citing is ballpark getting to a stage where it could replace grey hydrogen and be worth the extra cost just to clean up existing uses of hydrogen. So, things like fertilizer production and other chemical processes. It's still more expensive but the difference could be bridged with subsidies and incentives.

The economics of storing hydrogen for long periods is just fine, compared to the economics of storing energy in batteries for long periods.

> compared to the economics of storing energy in batteries for long periods

Which doesn't say much, since practically nobody is doing that.

Also the need for this long term storage is never quantified by those we need it. Put a twh number on it and the problem melts away. We're actually producing batteries by the twh per year now.

Relevant graph

https://www.reddit.com/r/dataisbeautiful/comments/17r9q6s/oc...

People (particularly nuke bros, including some at MIT) are doing it in their arguments purporting to show 100% renewable grids are not feasible.

> their synthetic gas still is an order of magnitude more expensive than natural gas

No it's not. their projections are for $10/kcf.

Another point others haven't made, which Terraform Industries points out: that using solar to make Natural Gas is going to be the cheapest form of natural gas production.

You can effectively short circuit the existing fossil fuel industry and pull the hydrocarbons from the air instead of the ground to stay carbon neutral. No need to re-invent industry.

But if the main problem with industry is that it leaks then maybe you do still have to reinvent it.

Seriously, how do you know whether the supply chain is truly using one process to synthesize, vs. Adding/cutting with extracted fuel.

On the demand side, I imagine you'd be drawing fuel gas from a rig that's like, in your neighborhood, being powered by the solar panels on your roof. So you'd probably notice if there were tanker trucks showing up to fill it.

On the supply side, you could use things like methane sniffing satellites (https://www.nature.com/articles/d41586-024-00600-z) to find cases of extraction and tax (at first) or arrest (eventually) those participating in it.

It doesn’t matter. This isn’t driven by consumer choice, it’s driven by production cost.

At least that’s the idea.

Other commenters have pointed out that hydrogen may not be feasible, but the other direction -- methanol or higher hydrocarbons, seems way more interesting. Storage and density are orders of magnitude easier, and since we're synthesizing it from atmospheric component we don't have to be concerned with contaminants, especially sulfur compounds, that plague higher hydrocarbon use.

Even better in a lot of ways would be to move to amorphous carbon; generating coal from atmospheric CO2 would be a huge win in transportability especially around safety and reliability dimensions.

It's quite hard to create heavy hydrocarbons in a controlled fashion. That company is already full of technical risks. Adding an extra optional one would be crazy.

That said, if not them, somebody will probably crack that one problem in time. But it's best done as an independent development.

Methane is pretty directly usable. Generally it costs some additional processing to produce higher hydrocarbons, which may be worth doing. But a system that makes pure methane can be used as the first part of some other, optional, system to make other hydrocarbons from that. You don't need additional equipment and processing facilities, which is important for Terraform Industries because it wants the sytem to be as cheap as possible.

Making methane is the cheapest way to get a transportable, useful output (hydrogen is notoriously difficult to move around, methane is a lot easier, and of course propane etc is a lot easier still).

Hydrogen is much more difficult to store.

It escapes easily from any container you put it in and causes hydrogen embrittlement for any metal you try to use to contain it.

Apparently it can be stored in abandoned salt mines and we probably will use them for that eventually but you dont always have one of those handy.

All those points are overstated.

My guess is that the answer is taking advantage of existing NG infrastructure.

> why not just use hydrogen directly and skip the inefficiency and cost of direct air capture of CO2 and of making methane?

Because hydrogen sucks. Specifically, storing it sucks.

You'll either compress or liquefy. That costs energy. It leaks and embrittles containers. One of the best ways of storing it in a stable manner is to combine with carbon atoms. Voila, natural gas.

You repeat some of the anti-hydrogen memes. Hydrogen is actually quite storable, certainly much more easily stored than energy in batteries. The cost of underground storage caverns for hydrogen is two orders of magnitude cheaper than the cost of batteries storing the same energy.

That's great, but how do I take one of these underground caverns around with me? :P

More seriously, I'm sure hydrogen will make a valuable contribution to the stationary grid-level energy storage situation. I'm not convinced about other applications.

You don't, just as you don't take massive takes of compressed methane around with you (I know about CNG cars; they're not competitive with BEVs or liquid fueled cars.) Methane and hydrogen are competing in stationary applications.

Guess what the cost of an underground storage cavity is for natural gas? Very cheap, because they're everywhere.

Yes, but hydrogen has the advantage of not requiring CO2 to either be acquired or recovered.

Because we kinda know how to do natgas storage/transport/use at large scale, while we're not quite there on the hydrogen economy.

Higher complexity hydrocarbons likely have higher synthesis costs, and if you want to enter the market, you probably want your pricing as competitive as possible.

Same for the biomass approach.

It is ultimately an approach striving for as much simplicity as possible. That's firmly baked in their culture, too - see their home page :)

We use all that because natural gas is cheap. If hydrogen were cheaper than natural gas, we'd use that instead, at least for the industrial processes, and we'd switch away from combustion of a gas in the distributed residential/commercial applications.

And if wishes were horses... ;)

Kidding aside: That's the whole point. This is a project that can work without retooling infra all over the world. It's a drop-in replacement to get to carbon-neutral fuel.

This solves problems right now, with limited investment. This isn't about perfection. This is specifically about "works right now, cheap, no impact on the surrounding infrastructure".

(Also, strong doubts on "we'd use hydrogen instead", because storage is a beast. Until that's solved, there is no chance we'd use that)

I don't believe their cost figures (look above; their 250/t for CO2 doesn't jibe), and I think electrification of most residential/commercial uses will be the superior solution (it's more efficient even if the electricity is produced by burning natural gas). Industrially, hydrogen is just fine. After all, industry already manipulates vast quantities of hydrogen.

The fossil fuel uses that are the most difficult to displace involve liquid fuels for transportation; natural gas has limited use there.

I would guess because natural gas has hundreds of years of us handling it, and we already have a lot of things from power generators to domestic driers that consume it.

Reply on news.ycombinator.com