Obviously hydrogen has it's uses, but a large part of me feels the energy industry is stalling for time. Oil and gas is basically a subscription model, whereas electric cars and heat pumps are buy once, upgrade in x years or when broken model. With enough renewable energy and a few nuclear power plants, electricity should be ridiculously cheap and if they've got the space, the consumer can even generate their own. That's more money and self-enpowerment for the consumer, and far less for the energy companies. They obviously don't like that, which is why they keep pushing for hydrogen to keep people dependent.
> few nuclear power plants, electricity should be ridiculously cheap
Nuclear has never managed to be cheap. The really big nuclear buildouts always had a national security subsidy: either for energy security (France), or for nuclear weapons (US, UK, USSR, China, and also France). This is partly why non-weaponizable reactor designs never became popular either.
> They obviously don't like that, which is why they keep pushing for hydrogen to keep people dependent.
This isn't quite it. Energy production is always going to be dominated by capital owners because it's a capital-intensive business. Doubly so for renewables. No, the real reason fossil fuel companies keep pushing for hydrogen is so they can sell hydrogen produced from natural gas as "green", after they've dumped the inconvenient carbon atoms into the atmosphere.
The medium term real, important uses for renewable hydrogen are (1) Haber process (3H2 + N2 -> 2HN3) and (2) steel production by reducing (removing oxygen from) raw iron ores.
As much as I’ve been rooting for nuclear, I think the ship has sailed. If newer advances can’t make it cost competitive with renewables plus some storage (I’ll get to that in a minute) then why would you ever deploy it.
Renewables are really cheap and getting cheaper, so some combination of overbuild with storage is probably the winning formula.
Storage is a little too expensive as lithium ion batteries, but there’s literally a hundred alternative there. I’m fond of pumped hydro, which we can do anywhere with a height difference, we don’t need to restrict that to existing dams. But there’s lots of options there and many are very viable.
So nuclear (including hypothetical nuclear fusion) will probably not acquire significant market share. Unless you can somehow make it cost competitive.
If you look at planned energy projects, the market has very clearly spoken. That doesn’t mean it can’t change, but I don’t think it’s likely.
> So nuclear (including hypothetical nuclear fusion) will probably not acquire significant market share. Unless you can somehow make it cost competitive.
Ontario is a great counter example.
We have something like 60-70% of our power from nuclear, representing most of the base load. We're building more plants soon too.
Electricity is cheap, and used to be even cheaper before it was privatized.
And we have no nuclear weapons program supporting these plants.
Ontario has cheap nuclear because it's nuclear fleet was build in the best possible time, the past.
Even china [0] renowned for pushing projects through and "getting stuff done" hasn't been able to push the price and construction time of nuclear down enough to make them cheap and easy to build.
Ontario is a great example where no no nuclear projects have been built or proposed at a competitive price. I stand by my assertion that nuclear is dead in the West until shown contemporary numbers to the contrary.
>If newer advances can’t make it cost competitive with renewables plus some storage (I’ll get to that in a minute) then why would you ever deploy it.
Because it can work rain or shine. Wind or no wind.
What happens if we have no wind and have a lot of clouds for a few weeks in an area? The storage is limited and nuclear can still produce. Yes, I understand other areas can likely pick up the slack, but nuclear doesn't have that issue. At a minimum nuclear is a good thing to have as a back up even if we have sufficient renewables.
Also, don't forget climate change is supposed to cause more extreme weather events. We don't know what impact that will have on wind and solar.
The answer is a combination of storage, overbuild, long distance interconnections, and diversification ( which mostly means we keep some natural gas plants around, not new nuclear.)
Solar can produce on cloudy days, but can be greatly reduced. I've seen estimates at 10-25% of the normal rate on a very cloudy day. If there is no wind being generated at the same time it could cause issues.
Sure overbuilding might work, but only getting 10% of the solar and 0% of the wind would require such massive overbuilding it would probably not be practical.
I did mention interconnection. While it can help, there can be extreme weather over large parts of the country at one time. We have fires in Canada blocking out some parts of the North East. Imagine if there was a large fire happening in California and a hurricane in the south. As climate change continues that is only going to get worse blocking out solar in large chunks of the country.
The problem is determining the correct amount of storage. If we get 10% of the normal solar along with no wind for a week or two would there be enough storage? I'm guessing not.
It is good you agree that we need some diversity. So many people are radicals and say we don't need an alternative. I agree with your sentiment, but think we should also have nuclear not just natural gas. Nuclear is reliable and clean. Why use natural gas if we don't need to?
> I agree with your sentiment, but think we should also have nuclear not just natural gas. Nuclear is reliable and clean. Why use natural gas if we don't need to?
Because natural gas is cheap and nuclear is not. If you’re using tax payer dollars, you absolutely can be wasteful and choose the more expensive option. That may even make sense when you factor in the cost of the carbon dioxide pollution. We’re not there yet, but it could come back around, especially if you implement a high enough carbon tax, which I’ve always advocated for.
I think you can get potentially very long term energy storage via pumped hydro, so I expect that would help as well. But natural gas can also be fired up occasionally on those very cloudy days when the wind is also not blowing. The pollution might not matter if you offset it via other means or you use green hydrogen or ammonia or something to that effect.
Since we already have a lot of natural gas power plants, we might not need to build any more, just maintain the more efficient ones in working order.
Rooftop solar in Australia is now cheaper in some jurisdictions than transmission per kWh. Meaning, if I tried to build a nuke or any other centralized generator and I gave the electricity away for free, rooftop solar would still be cheaper per kWh because consumers don’t need to pay to maintain transmission infrastructure..
Of course, only when the sun is shining. But it’s an incredible Lego block we’ve got to play with in building this new energy system, zero marginal cost generation.
There’s sodium ion, molten sodium, iron air, probably various options with iron or aluminum. There’s literally dozens of promising battery chemistries for grid storage where weight and density don’t matter.
> If newer advances can’t make it cost competitive
Did they ever get rid of the rule that prohibits nuclear from being cost competitive? Its opponents got it so if they ever found a way to make it cost less the money explicitly had to be spent on new safety measures. Which obviously not only makes it impossible to reduce the cost but also removes any incentive to try.
Google doesn't want to find a more specific link right now, but you can see the implication from the definition: If you find a way to make nuclear cost less than something else, now it's economical to make it cost more in order to reduce radiation exposure, with no lower limit where you can stop.
When I say nuclear, I don’t mean as the primary fuel source, but as a few strategically placed plants to serve as back up until large scale battery storage gets sufficiently commonplace until it is no longer needed. Eventually every house will have its own large battery and solar panels on its roof. In combination with grid battery storage, increased panel efficiencies and huge amounts of solar and wind farms eventually the need for backup nuclear energy plants will be next to zero and would probably only exist as small, mobile units that could be moved to disaster zones. Obviously I’m talking on a massive timescale here, we probably won’t get to this point this century unless someone really does make a room temp superconductor and then we can just pipe energy from the equator to where it’s needed.
> and then we can just pipe energy from the equator to where it’s needed.
Superconductors aren't the limiting factor for that, on paper we can already make a 40,000 km long 0.5Ω power line for costs comparable to current annual fossil fuel mining (Chinese annual coal alone is expensive enough for the aluminium).
The problem is geopolitics and that it's a megaproject.
Interesting, I knew there was talks about piping energy from Morocco/Sahara to the UK but didn't know anything about projected costs or feasibility. Politics is the worst.
Sahara to UK (or equivalent) is a necessary, but not sufficient, early step to learning how to do a full-scale planetary grid.
It will highlight many of the engineering issues and some of the political ones, and ideally will seem like an overpriced mess when we reach the level where we even want to optimise design and process for a full-size grid.
For a true planet scale grid, think half a trillion dollars of aluminium, 3.75 years of current annual worldwide production: its doable, just not what you should jump into without smaller scale experiments.
Existing aircraft will be wholly unable to compete on any route where LH2 aircraft operate, but it will take a long time to bring up the infrastructure for it. Expect it to appear first on select freight routes.
Oddly, H2 aircraft seem to be promoted with inboard tanks. The natural place for the tanks is in nacelles slung under the wings, for safety. (Hydrogen would not fit in the wings.) Hydrogen tanks in an enclosed cabin is a formula for disaster.
I am fairly sure this is due to the necessity of having pressurized cylindrical storage for the hydrogen. Fitting that kind of tank into the wings is going to be hard.
Tanks would be wholly impossible to fit into the wings. Thus, the under-wing nacelles. But pressure cylinders are a non-starter, because they are heavy. Expect to see, instead, insulated, unpressurized LH2 tanks.
The common criticism of LH2 economics is the energy required for liquification. Are you assuming that this is taken care of by overbuilding solar/wind and creating LH2 opportunistically?
Given the power requirements of flight — they're fuel efficient per passenger-kilometre, but they have a lot of passengers and go a long distance — planes are the only place[0] where I think "overproduction by nearby solar farms" just isn't going to be a thing.
OPEC paving their deserts with PV and synthesising fuel (whatever that is: hydrogen, Sabatier methane, aluminium for burning) or just exporting that electricity along a 2m^2 cross section solid aluminium rod to the other side of the planet? Sure, plausible.
[0] I was going to say "and rockets", but then I realised we don't launch anything like as many rockets as we fly planes, so even then rockets might still be running on green hydrogen or methane derived from it.
Notably, SpaceX is making no visible effort to synthesize methane for their cans. The only gesture in that direction is an announced plan to buy a pre-fab methane refinery that could possibly be adapted to run Sabatier; but no hint at a solar farm to power it, or a place to put one. I guess they could build one across the border in Mexico? CH4 for Cape Canaveral is a separate problem. More likely the refinery will just purify mined LNG.
Solar and wind farms supplying international airports would probably need to send power via HVDC transmission lines. But, yes, the airports will need much more than just overage from the farms, and probably booster shipments of LH2 from farms in the tropics, besides. Imagine how big must be the project of refining, transporting, storing, and distributing kerosene to gates, today. Yet it is made almost invisible.
> Existing aircraft will be wholly unable to compete on any route where LH2 aircraft operate
? Are you suggesting this will be a result of regulatory action? Since it would most likely be more expensive for the first decade, even if I gave you a tap on the airfield labelled "free H2"
It is because the very large difference in fuel weight for a flight leaves radically increased capacity for carrying paying freight. As LH2 produced on-site at airports gets cheaper than Jet-A fuel, the gap widens.
Volume is cheap, on aircraft. If a normal freighter today takes off with 25t of kerosene, the similar-sized LH2 craft needs only 10t of LH2. That means it can take an additional 15t of paying cargo.
That tendency will delay the transition. But the big economic benefit arises from being able to carry 40% more cargo because the fuel is lighter. Passenger aircraft capacity is maxed out when the seats are full, now that carrying freight in the same aircraft is not allowed. Also, it will take a long time for safety worries to be satisfied.
I think hydrogen for aviation is probably a dead end. I suspect the long term solution for aviation, at least long haul, is biofuels. Battery tech is edging into the realm where it could be used for short haul flights, but we have quite a ways to go before you'll be able to fly across the Pacific, or even Atlantic, on a single charge with reasonable passenger comfort.
Biofuels today have issues, but there's no fundamental problem preventing them from being green. It's just a side effect of shoehorning them into our existing agriculture systems.
In the future when we have more solar power than we know what to do with during the day it may become economical to run the Sabatier reaction with hydrolyzed seawater and atmospheric CO2 to make methane, which can be burned in a lightly modified aircraft turbine.
What do you mean? The fuel itself should be carbon neutral, even if it requires extra energy to create. I don't see any reason you couldn't make carbon neutral bio fuel, although I'd net on syngas instead.
Growing food and converting it to fuel is not carbon neutral. Farming is very resource intensive at the scales needed. In fact, it is literally the same idea as carbon capture, but via biology instead of synthetic. As a result, it is entirely a bad idea.
A catch is that hydrogen does become much safer in high-pressure tanks, because, when such tanks are breached, hydrogen escapes so fast that it can't react with oxygen fast enough to burst into flame. To prove this, Toyota literally shot their hydrogen tank with .50 cal[1].
Practically no one died in the Hindenberg disaster. There were more people injured than died. Most people involved just walked away from it. It was televised spectacle much more than a factual tragedy.
(It's a bit like saying we shouldn't use AC power because look what Edison did with those poor elephants. It's interesting anti-technological propaganda that made sense socio-politically at the time but isn't that useful to today's discussions.)
Hydrogen can be generated on-site (and in quite a few cases, that's preferable over filling up elsewhere).
So if the efficiency of generating hydrogen is increased, that's a win for endusers no matter if they do it themselves or pump it somewhere from 3rd party.
You are correct about the subscription model though.
Once GPU production is fully ramped up, I would expect AI to become energy bound. Can we install enough renewable energy and nuclear power plants to fulfill the demand to the point that energy will be ridiculously cheap?
AI as currently modeled can pretty much never get enough data, so mostly data processing, and model generation.
when a query is asked of an AI it has to generate a response from all of that data and the query and response themselves become data
running an LLM on local consumer hardware can take upwards of 20 minutes for a single query, so an AI service that may be responding to up millions of requests a day would need a massive hyper-parallelized server infrastructure
It’s energy bound in the sense that Datacenters only have so much energy supply, and heat dissipation capabilities. And the grid these DCs are hooked to only have so much excess generation that can be tapped.
But there’s ways to solve that through energy generation and DC investments.
It will, but that's also an x/y problem. We're seeing ridiculous energy costs and supply chain issues for crypto and AI because they're using the wrong architecture: GPU/SIMD.
I got my computer engineering degree back in the 90s because superscalar VLSI was popular and I wanted to design highly-concurrent multicore CPUs with 256 cores or more. Had GPUs not totally dominated the market, Apple's multicore M1 line approach with local memories would have happened in the early 2000s, instead of the smartphone revolution which prioritized low cost and low energy use above all. We would have had 1000 core machines in 2010 and 100,000-1 million core machines for 2020, for under $1000 at current transistor count costs. Programmed with languages like Erlang/Go, MATLAB/Octave, and Julia/Clojure in an auto-parallelized scatter-gather immutable functional programming approach where a single thread of execution distributes all loops and conditional logic across the cores and joins it under a synchronous blocking programming model. Basically the opposite of where the tech industry has gone with async (today's goto).
That put us all on the wrong path and left us where we are today with relatively ok LLMs and training data drawn from surveillance capitalism. Whereas we could have had a democratized AI model with multiple fabs producing big dumb multicore CPUs and people training them at home on distributed learning systems similar to SETI@home.
Now it's too late, and thankfully nobody cares what people like me think anyway. So the GPU status quo is cemented for the foreseeable future, and competitors won't be able to compete with established players like Nvidia. The only downside is having to live in the wrong reality.
Multiply this change of perception by all tech everywhere. I like to think of living in a bizarro reality like this one as the misanthropic principle.
nuclear isn't cheap if you are thinking about the cost to build out a new plant under current design.
but the per kw cost is relatively low as i understand it since the fuel is so efficient and its getting cheaper as new reactor designs get cheaper and safer
BEVs are the biggest example of greenwashing today. You will need a vast increase in raw material demand and it is arguably completely unsustainable. Especially once you realize that we are making battery powered SUVs.
In reality, people are spreading a conspiracy theory to mentally distract from this fact. They do not want to admit that they have been fooled by battery makers, so they create a narrative that the alternative is somehow an ever bigger scam.
