Toyota believes hydrogen fuel cells are the future | Hacker News Reader

Toyota believes hydrogen fuel cells are the future

autoweek.com

88 pointslawrenceyan186 comments

Someone high up at Toyota bet on fuel cells 20 years ago and can't back down now without losing too much face. That's the only explanation. Hydrogen just doesn't make sense from any engineering viewpoint other than "make sure there are still gas stations and a fuel distribution network no matter the cost."

I think they see the writing on the wall for their services business with solid state tech and wanting to fight the trend. Or, building engines are just such a core part of their company that they’re essentially petrol heads looking for a replacement as near as possible. But I’m low information on that and just speculating.

Fuel cells are neat tech though. I also understand the argument they made with respect to using nickel hydride batteries over lithium. See their CFOs recent remarks.

Hydrogen Fuel Cells are electric.

Yes, and they wear out and must be replaced. The drive train is powering an electric motor. Toyota might be expecting to get more service work out of fuel cells than batteries. Just throwing things at the wall for their motives though.

> Yes, and they wear out and must be replaced.

Doesn't everything? Some production fuel cells have a service life of >50000 hrs. That's way more than you can expect out of batteries.

They should have been well placed to adopt EV's with their experience making the Prius. It's mind boggling to me that they didn't take advantage of that.

I can see a hybrid vehicle with both batteries and fuel cells as a future option however. I don't think we're doing to slay this problem with just one technology.

Why on earth would you need that for a normal car?

same reason you'd have a plug in hybrid.

Because the power density for fuel cells is pretty meager, so you use batteries as an energy buffer to smooth out the demand.

Why on earth was this downvoted?

Because I think that concept already exists as prototypes.

A small fuel cell for example can serve as a range extender.

So for short drives you charge and use the batterie (cheap energywise). For long ranges the fuel cell adds power to charge the batterie. Methan or hydrogen is fast refueld and has a much bigger energy storage capacity compared to batteries.

If you look at the range of Teslas, it is close to the one of hydrogen fuel cell vehicles. The problem is, that hydrogen pressure tanks are large. Compare a Mirai to a Tesla Model 3. Even the recharging of the Tesla is close to the speed of hydrogen stations. (The refueling with hydrogen is faster, but then there are about 15 minutes in which the fuel pump has to repressure itself)

I am not saying the concept is superior right now. I am saying it is worth thinking about. And when you have the fuel cell only as addition, maybe you don't need the big pressurised tanks.

Maybe a small methan tank.

Maybe even then it is still not worth it regarding the extra weight and cost. I don't know. But I do know that the involved technologies are advancing and combining things makes sometimes sense.

edit: and the market is big. there are also people living in remote areas who sometimes want and need range extension in which case they could pack in some extra tanks.

> ”I can see a hybrid vehicle with both batteries and fuel cells as a future option”

All fuel cell vehicles already have batteries, of course.

But if you mean plug-in fuel cell hybrids, I don’t think that will happen.

Manufacturers of fuel cell vehicles need to create demand for hydrogen in order for hydrogen filling networks to expand, in turn stimulating demand for more hydrogen cars.

But if you put a plug on the car, owners will prefer to use the batteries for their daily driving, only using expensive hydrogen occasionally as a backup or for long-distance driving. This would greatly reduce the demand for hydrogen, so filling networks would struggle to be viable.

Also, from an engineering perspective, if you’ve gone to the trouble to add a plug and charging equipment, you already have an EV. Why then add expensive, complex fuel cells, bulky hydrogen tanks, huge air filters and all the rest of it? It’s just going to be easier and cheaper to use a bigger battery!

Nope. Hybrids are already starting to phase out.

Unified fully electric on batteries is so much cheaper that once it achieves critical distance (somewhere north of 150 miles per charge), hybrid adds far too much manufacturing and engineering cost.

If we are to do that on a big scale it required massive infrastructure investment. Something we are nowhere near yet. Plugin hybrids offer a useful stepping stone.

VW have been making a lot of noise about their ID range lately but if you look at what they have in the pipeline it's mostly hybrid.

Not as much as people think. The existing grid, in the UK at least, can cope with only minor changes, so long as cars are smart enough to react to pricing signals as they charge. This is quite simple to do as the grid already implements a market system with prices available in real time.

The big challenge will be providing charging for people (in the UK about 40% of the population) who do not have the opportunity to park their car off road. Quite a lot of this demand could be met by simply providing normal domestic sockets with RFID controlled breakers on posts at 6 m intervals along urban streets and in every public parking space. This is relatively simple and scalable and does not need large capital expenditure to get started.

Any maybe you really believe in hydrogen, but you're the ~largest automaker, you can afford to also bet on batteries.

The major hurdle for EVs was getting energy density, capacity, and power from batteries, all problems that Toyota's hybrid technology solves with the ICE instead (the Prius battery's primary function is to reclaim braking energy and assist the ICE).

Therefore despite their lead in hybrids, that tech wasn't directly transferable to EVs.

That's not the Toyota way. From a Westrum "Three Cultures Model" perspective Toyota would fall into the Generative classification, not Pathological or Bureaucratic.

Based on the discussion in the article I think Toyota is saying they've learned EVs are not scalable and are not a solution to all customer needs Toyota currently services. They're betting hydrogen is the solution.

Part of the reason that Toyota continues to pursue hydrogen fuel has to do with a Japan’s energy needs. It’s a pragmatic approach that acknowledges that energy needs to be stored and shipped long distances in order to avoid generating fuel in-country. Also, the idea that Toyota isn’t all over battery tech is a complete misunderstanding. All one needs to do is read their public strategy, which is available. They’ve been working on personal and ultra compact EVs, people movers, trucks, etc. They tend to release domestically first and move thoughtfully (cf. Tesla), as US consumers have proven to be very fickle, prone to nationalism, and so on.

I think there is a future in hydrogen fuel cells, but not for personal transportation.

Heavy machinery operated in remote areas like farming equipment have pretty intense energy needs and downtime for charging could ruin a harvest. Batteries are also very heavy and soil compaction is a real issue for modern farming.

I think we're approaching a time where the marginal cost of energy will be 0, but there will still be costs associated with using energy at a specific time or place. I would like to see a lot more dollars being spent on technology and facilities that can ramp their energy consumption up and down extremly quickly to absorb extra power on the grid. Hydrolysis seems like a fantastic candidate since the process is highly interruptible and the result is portable.

Trains too, there are some experimental fuel cell lines replacing diesel locomotives in germany and they ordered more, so it seems to be working. Presumably it's cheaper than electrifying branch lines.

Battery electric trains also greatly reduce the cost of rail electrification. They can charge when running under the wires, then use battery power on unnelectrified sections and branch lines.

I’d imagine you would only need a relatively short section of powered rail around each station, plus maybe some boosters for long inter-station runs or hills. I’m surprised this approach isn’t more common.

I've always wondered why you couldn't just string a wire from a tower down to a combine, tractor, or other piece of farm equipment and run it right off the grid. Using very high voltage could allow the wire to not be that heavy.

Because farms are huge, and any given unit of area needs to be touched relatively infrequently.

Cables can be thin and long. Look at high tension wires which carry vastly more power. It stands to reason that a much thinner lighter cord could run a combine.

I won’t even use a corded lawn mower. I highly doubt I’d put up with dragging a wire behind a combine through a 400 acre field.

Not dragging but way overhead from something about the height of a radio tower. There could be one about every square mile in a grid. It could have a smaller battery too to permit driving between them. Drive up, hook on, unhook, go to next.

It occurs to me that because I’ve operated farm equipment, maybe one should ignore my poo-pooing. Seriously, sure, I can tell you eight ways to Sunday this won’t work, and some smarty-pants can show me how I’m wrong. But right now I’m just stuck in a mode of “that’s more hassle than filling it with diesel and being done with it”. I might be right, but that’s not where innovation comes from, so I’ll be quiet.

It probably is more of a hassle but it would also be cheaper and lower emission. Of course I dont know if fuel is a significant cost to farmers.

There may be ways to make it less of a hassle by automating it somehow. I'm thinking of some standard hook on/off system. It's a tough problem though.

Maybe batteries will just keep getting better and lighter and cheaper and we can just go that way.

You're suggesting a large capital investment, with dubious utility, in a leveraged, low margin business with little access to technologically aggressive capital funding.

Furthermore, it's predicated on the supplier developing and marketing a class of equipment for which little demand currently exists.

Against that, the price of diesel seems relatively small.

Some alcohol probably makes more sense in a near zero cost energy world than hydrogen. It's easier to store and blah blah blah.

It's much easier to add some carbon into your hydrogen than it is to deal with all the problems that hydrogen bring.

This might be a dumb question: what if you could build a reactor that electrolyzes water but then also converts it into methane? If the electricity used for the process was derived from renewable sources, the process would still be carbon neutral, yeah?

I'm assuming the carbon itself could be harvest through non-fossil fuel means (biological perhaps? burn cellulose?).

Pilot “methanation” plants are already producing synthetic methane from hydrogen and CO2:

https://www.uniper.energy/news/methanation-plant-in-falkenha...

This is a good article on zero carbon farming utilizing hydrogen and ammonia powered tractors, using solar to produce the hydrogen.

https://spectrum.ieee.org/energy/environment/a-retired-jpl-e...

As an EV owner who also owns a 7klb Kubota CUT diesel is still ideal here for a few reasons. First it's easy to handle and transport, hydrogen has no shortage of issues there. It has a heck of a shelf life, my understanding is hydrogen will leak out and is a massive pain to store.

Unless you're doing serious ground work an EV based platform could be fine with solar. JCB already has a few electric excavators and they work pretty well.

I also with there was more focus on biodiesel, that has the aspect of being carbon neutral while working with a lot of existing infrastructure.

Biodiesel doesn’t solve air pollution, which is a huge health issue in many cities.

Admittedly it’s not such an acute problem for farms and rural areas, though, where biodiesel does make sense.

Yeah, that also depends on how modern your diesel is. The Kubota is much, much better on the particulate scale vs out old Ford thanks to a substantial DPF filter.

Newer diesels, when not modified are much better than they used to be.

Theres also this thing: https://www.slashgear.com/hydrogen-fuel-cell-ev-charger-afc-...

I have no idea if the economics of that actually will work.

> I think we're approaching a time where the marginal cost of energy will be 0

I strongly believe that we will continue to find arbitrary uses for available energy, e.g. Bitcoin mining. I can buy that current normal household energy use will approach 0 cost, but I think there will be players that scale use in such massive amounts that marginal costs will still matter to them.

> Hydrogen just doesn't make sense from any engineering viewpoint other than "make sure there are still gas stations and a fuel distribution network no matter the cost."

There are gas stations and a fuel distribution network, so it makes sense to use the existing infrastructure.

Our current fuel containers are completely unable to hold hydrogen. Neither can the fuel transportation infrastructure.

Fuel cells might be able to compete in that small slice of the market that needs deep-cold compatible vehicles. There are millions of people who live in places like Edmonton or Novosibirsk where temps can drop below -40. I don’t know if current HFCs can handle that but since gas solubility is higher at low temperatures it seems like a possibility. Lowering battery operating temperatures seems to be a hard problem.

Battery management systems deal with this by using some energy to heat the pack. So dealing with cold temperatures is just a measure of insulation and some vampire drain. While driving, or parked in a garage, there are no issues of course. Note that Norway is the European country where Tesla cars are most popular, so the problem seems to be nonconsequential in practice.

Norway is rather warmer than Novosibirsk. Annual average temperature in Drammen is 6 C, Novosibirsk 1 C. January average in Drammen is -3.6 C, Novosibirsk -17.3 C. See https://en.climate-data.org/asia/russian-federation/novosibi... and https://en.climate-data.org/europe/norway/buskerud/drammen-9...

But generally speaking you are correct. My Tesla S 70D has active battery temperature management and can be set to precondition the battery and heat the cabin at set times every day so that it is ready to go when you commute to work.

A problem with that idea is that the total numbers would be so small that there would never be any real economies of scale, and so the cars would be too expensive for most people to afford.

>Someone high up at Toyota bet on fuel cells 20 years ago and can't back down now without losing too much face

That's it exactly. At the time li-ion batteries were pretty poor, so fuel cells seemed like a good bet. Since then batteries have progressed much faster than fuel cells, and now are way ahead, and no doubt that will continue in the future.

Many observers have commented over the decades that Japanese corporations have great difficulty admitting they have made a mistake, and this seems to be another case of that.

Germany, Korea, Australia and China are also all over hydrogen.

Does Australia even have a car industry?

It did. It died.

I don't see much of hydrogen here in Germany. Volkswagen, the biggest car manufacturer has clearly set the bets on battery electric. Mercedes has a hydrogen car, but only leases it out, doesn't sell it. Of course, everyone who missed the battery electric car market so far is still talking about hydrogen as "the future".

How many hydrogen charging stations are there in Japan and all those countries combined, as compared with how many electric charging stations?

Li-ion ev's outsell hydrogen ones about a hundred to one, and the numbers are going to greatly increase in the coming years. Hydrogen for cars has never gone anywhere and it never will.

I thought the Hydrogen thing was a ruse to kill battery electric vehicles, years ago, invented by fossil fuel lobbyists.

It works like this: Hydrogen seems plausibly usable, but nobody is sure. It allows continued use of internal combustion engines so you can carry on as normal for now. Then in 10 years when nobody has figured out how to make Hydrogen work it's too bad we didn't focus on battery electric. And of course you can repeat and rinse because perhaps in another 10 years Hydrogen may be made to work.

We're two cycles in at this point.

It also serves to preserve the current refueling experience from fuel delivery to point of sale.

Electric cars are mostly charged at home and at work/parking areas which flips everything on its head.

There are a lot of industries threatened by this.

One consideration is the current world political order is built around supplies of oil. Battery powered transport dramatically changes that.

That was definitely part of it. There was an industry-wide push to work on a "hydrogen economy" as a stall to stave off the CARB regulation which was going to force a certain percentage of vehicle sales to be electric.

It did make more sense in the late 90s when the predominant battery tech was NiMH. Fuel cells were actually competitive on price and performance, but can't really hold a candle to modern LiPoly or LiIon batteries.

U.S. car manufacturers dropped fuel cells like a hot potato as soon as the CARB mandate was repealed, and crushed their BEVs to destroy any evidence.

But the strategy failed because enough companies went ahead with battery electrics.

I always think of gasoline, hydrogen fuel cells, and batteries as chemical energy storage.

Gasoline had the advantage of also being a primary resource. And can be pumped. And stored in a thin walled metal tank. The disadvantage is it requires a jankie otto cycle engine to convert that energy to mechanical power. Which was both unreliable and polluting.

Hydrogen shared with batteries the disadvantage of not being a primary resource. And requires a heavy storage tank and a jankie temperamental fuel cell, batteries and an electric motor. Transport and fueling is sketchy with a capital E.

Batteries while having lower power density and high costs had the advantage of being really simple. Which I think explains their original success and eventual return. Consider a lithium ion battery needs bulky cathodes and anodes. While hydrogen needs a heavy high pressure tank. Not clear which of those is a winner. Yet the batteries functional simplicity is a win hands down.

Storing H2 is really hard, I wouldn't bet on it.

It's easy to criticize Musk, but I'm a big admirer of his pragmatism. If I were a billionaire I'd pour money on super capacitors, for instance. Much more exciting and the payoff is potentially huge.

Instead, he goes for 20th century tech and infrastructure that's proven to work and tries to squeeze every efficiency drop out of it. We have a problem that needs to be solved now. From a risk assessment point of view, it's the best call.

> Storing H2 is really hard, I wouldn't bet on it.

That's not even the biggest issue for "the future": the "hydrogen economy" is based on cracking fossil fuels, producing CO2, so it's not exactly green.

Electrolysis is energy-intensive (50kWh/kg) and quite expensive, it's used for a small fraction of the production (2%) and mostly for uses which need very pure hydrogen (cracking fossil fuels yields pretty "dirty" hydrogen which is unsuitable for some applications).

I guess at capacity that would make it suitable as a sink to dump excess electricity from renewables in, but then you hit the storage issues you mentioned.

It’s perfectly feasible to capture carbon from a natural gas refinery producing liquid H2 and sequester it safely. Doing that safely and consistently is going to be an engineering challenge, but so are the advances in battery technology and the need to scale lithium production in clean, safe ways that batteries would require. If anyone’s done the work on which approach is better, I’d be interested to see it.

Perhaps, but the refining companies would much rather lobby the US Senate to ignore Global Warming treaties, it's their modus operandi.

If you want to get off of carbon, you have to get away from the people making trillions off of carbon.

You make an interesting point but I think it actually works the opposite way from what you think. If the only alternative to oil and gas is battery-electric power, oil and gas companies will use their political power to obstruct climate change mitigation altogether. If there’s an alternative that not only includes them as partners but opens up new lines of business for them (eg carbon capture and sequestering), then they will apply their political power in favor of that alternative out of self-preservation.

Of course, the kind of argument you’re making doesn’t really seem to stem from trying to solve the problem in the first place—it stems from viewing the world in terms of conspiracy theories and making black-and-white judgments of who are the good guys who deserve to be supported and who are the bad guys who deserve to be destroyed. But even then, if you have enough power to achieve your fantasies of destroying the oil and gas companies, you also have the power to force them to move into the hydrogen and carbon-capture-and-sequestration industry.

I think what's going to happen, is happening is governments that have no stake in the oil economy are going to head for the exits. Oil producing countries are powerful but not powerful enough to stop that from happening.

> Electrolysis is energy-intensive (50kWh/kg) and quite expensive

Solid oxide electrolysis cells have been producing hydrogen at >93% efficiency (LHV) for over a decade now. Some research cells have reached 99%. It is still considered expensive but only because hydrogen derived from natural gas is cheaper. I personally have produced hydrogen via electrolysis at a cost of $2.72/kg. After a 60% efficient conversion back to electricity, that's $0.116/kWh.

You might know: why do large electrolytic hydrogen projects being built or planned today still use alkaline or PEM electrolyzers? I haven't seen any announcements where projects are using solid oxide cells. I assumed that the technology wasn't commercially available yet.

I'm not aware of any large electrolysis projects using PEMs, unless the purpose is for research and development. I know of a few using alkaline cells, but those are quite efficient. At least as efficient as solid oxide cells, just a bit harder to work with. Most of the research and development for solid oxide cells is for fuel cell usage, with a focus on usage of liquid or gaseous hydrocarbon fuels, due to the easier technological transition (I'm working on a project like this).

Without subsidies or carbon taxes, electrolysis is likely to be done only on a small scale. At large scales, hydrogen from methane is still cheaper, and there is no penalty for carbon emissions, and consumers typically buy in large quantities at limited endpoints.

However, if fuel cell vehicles (more likely to be commercial vehicles than personal vehicles at this point, as the economics are more in their favor) become more common, there will be growing benefits to electrolysis because the logistics are so much simpler for a large amount of end points.

As in this recent story:

"World's largest green-hydrogen plant begins operation in Austria"

https://www.rechargenews.com/transition/worlds-largest-green...

It's using PEM technology from Siemens. It reports that an even larger PEM project should finish construction next year. It mentions one other project, even larger but further out, using alkaline electrolyzers. They are all part of decarbonization research projects, but operating at megawatts to tens of megawatts scales.

Well he started off working on super capacitors I think in college. Or he said he wanted to, I forget. More recently, Tesla has purchased a super capacitor company (Maxwell). I’m using some supercapacitors in a project at work. For $500 we get a super capacitor with the capacity of a $12 battery. For the time being, batteries are way, way ahead on cost.

I don't know much about costs, but papers like this make me pretty giddy at the prospect of supercapacitor-only cars.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6409971/#sec1-n...

That energy density is at least as high as current generation lithium ion, which is obviously already feasible for production cars.

Sure, but what’s the short circuit current rating from a super capacitor vs a battery. We have used them in thermal battery powered supplies for pulsed electronics.

They’ed be useful in cars for acceleration and regenerative breaking, allowing the use of lower current batteries.

> Instead, he goes for 20th century tech and infrastructure that's proven to work and tries to squeeze every efficiency drop out of it.

Ironically, this is Musk being more SoftBank than SoftBank.

Given our current and near term technology level, what industries have attainable new local maxima/minima, for want of only capital?

Nobody thought lithium ion batteries were a bad idea, they just weren't willing to pour crazy money into them.

In classic “Innovator’s Dilemma” style, Tesla spotted an opportunity to sell a technology slightly before it was ready for the mass market, by focusing on a small and undemanding (in price-performance terms) group of consumers that the big players couldn’t be bothered with.

I don't think electric cars are the product: they're the consumer-funded demand.

If storing H2 really cannot be solved (but I think it was claimed they did), there is allways methan as an alternative.

or ammonia.

https://en.wikipedia.org/wiki/Ammonia#As_a_fuel

What do you mean? Lithium is not 20th century tech, it's a major driving force behind much of technological innovation since 2000.

Tanaka explains their strategy pretty clearly here, and it’s about functioning in an entirely renewable energy context.

Hydrogen is better than batteries for longer term energy storage. To run a battery powered system twice as long, you need twice as much battery. To run a fuel cell twice as long, you need a bigger tank.

Japan is an energy insecure nation. Post Fukushima, Japanese companies are dealing with figuring out how to store energy from intermittent or interruptible supplies: not for hours but for weeks or months.

Hydrogen does that. It is a synthetic fuel that can be made from electricity and water. Fuel tanks scale well, and a large tank with thick walls can store a lot of hydrogen for quite a while.

So the strategy is anchored in a hydrogen energy architecture on a grid level. Mirai is a technology demonstrator for that. Critiques based on the current lack of hydrogen infrastructure are missing the point.

As a desert dweller, I prefer a drive train that outputs water just in principle. We need water out here, it’s only getting more dry, and there will be no big project coming to pipe water inland. Buuuut, batteries and electricity are here today.

The principle is cute, but on a regional environmental level the water emitted won’t make a difference.

And batteries are here already. Buy a Tesla.

People have been saying the same about carbon emissions for a long time.

Yes, and if we were talking about releasing fossil water into the atmosphere there might be something to the comparison. But water emissions from a hydrogen car are just moving existing water around and don't compound over time. It's minuscule compared to the hydrologic cycle even over desert regions.

How does the efficiency work out, considering you're using electricity to make hydrogen. Does the hydrogen you create put out more energy than the electricity you used to make it?

Toyota is ready to make BEVs at scale: https://www.thedrive.com/tech/28424/toyota-subaru-suzuki-and...

And is investing in making solid-state batteries affordable at scale: https://www.sae.org/news/2019/09/battery-show-solid-state-ba...

Many companies are “ready” to make BEVs. Where are all the BEVs, then?

They’re coming. Mini announced all of their models would be electric in 3-5 years, I think. Most or all have some plan for BEVs in coming car generations.

Your exact comment could have been posted three years ago, and three years before that, with Chevrolet, or Audi, or Volkswagen named.

And yet, here we are, and only one company in the world is actually building the factories you’d need to sell a half million EVs in a year.

There’s too much money in not making EVs for any of these companies to start making them a day before the market forces them too.

After 10 seconds of Googling, here are the new (BEV: 11 models / PHEV: 9 models) models shipping in the US next year: https://evadoption.com/future-evs/new-electric-vehicles-in-2...

I will leave as an exercise for the two of you to Google:

1. How many new EVs arrived in 2019, in the US and worldwide? 2. How many are committed for 2021, in the US and worldwide?

You might start with VW committing to 27 models in 2022: https://www.greencarreports.com/news/1118857_vw-plans-27-ele...

None of those models are planned to sell 100k let alone 500k. They are compliance cars—a tax break, basically, so they can keep selling internal combustion engines without paying penalties.

I think the answer is to time travel to the year 2020 (it should take you about two weeks) and to also be in Europe. Here's just Volkswagen's EV models for the next year or so:

- VW e-Up!: https://www.carscoops.com/2019/09/vws-updated-e-up-offers-16...

- VW e-Golf (sold over 100,000 but will be replaced by the ID.3): https://www.youtube.com/watch?v=ah4lrqWx8E0

- VW ID.3: https://www.youtube.com/watch?v=op4HO6GHC8Q

- VW ID.4: https://www.autocar.co.uk/car-news/motor-shows-frankfurt-mot...

- SEAT Mii: https://www.electrive.com/2019/09/11/seat-mii-available-to-o...

- SEAT el-Born: https://www.youtube.com/watch?v=DZI7WFtwc8g

- Cupra Tavascan (maybe, not confirmed): https://www.youtube.com/watch?v=YNYHRKp4n1w

- Skoda Citigo iV: https://www.autocar.co.uk/car-news/new-cars/new-skoda-citigo...

- Skoda Vision iV: https://www.youtube.com/watch?v=-f1g9xl6W_E

- Audi e-tron (and also the new "sportsback" variant): https://www.audiusa.com/models/audi-e-tron

- Audi e-tron Q4: https://www.youtube.com/watch?v=DiwevzHsCbU

- Audi e-tron GT (maybe 2021): https://www.youtube.com/watch?v=tMEdiq2xTbQ

- Porsche Taycan (and the Sports Turismo wagon variant in future): https://www.porsche.com/usa/models/taycan/

And other manufacturers are delivering more EVs in 2020.

Lots of models, but where will the batteries come from? Only Tesla seems to have the proper vertical integration although I've heard VW group is investing in battery factories.

Sweden, Korea, and China: https://www.reuters.com/article/us-volkswagen-electric-batte...

Hydrogen definitely is the future, it just may not be the future for cars. Much more likely to be used for transportation modes that have long running and consistent speeds. Long distance trucks, cargo ships, planes, etc..

The problem with cars is that they're so intermittently used and the loads are so variable. The only fuel cells that are really up to that task in particular are PEM fuel cells, which are expensive and inefficient compared to solid oxide and alkaline cells. Solid oxide cells might have a chance if used in a more plugin-hybrid form, where the fuel cell is used as a range extender.

There's a real prospect of a worse is better outcome. Hydrogen is way more energy dense than oil-based fuel, never mind battery tech, and can be refilled just as fast. But it's a bigger technical leap, and people won't be able to refuel at home. If most vehicles are used for commutes, I can see battery tech winning the war, except perhaps for trucks, which today may take LPG and can more easily live with bigger distances between refuels.

Hydrogen fuel cells may be the solution to aviation though.

Water and electricity flow to many homes. It isn't certain that people can't refuel at home.

Hydrogen can be refueled at home. Electrolysis is pretty damn easy.

Currently the best electrolysis can do is 50kWh/kg of hydrogen, which is about equivalent to a gallon of gas. That's 7c/kWh to match a $3.6 gallon (as found in CA or HI). And that's before the storage issue is even considered.

That's the best that PEM electrolysis can do. Solid oxide cells have been >93% efficient for over a decade now, and some research cells are 99% efficient. If you are willing to supply heat to run in endothermic mode, they are >100% efficient on an electricity basis (meaning they take less than 39kwh of electricity to make a kg of hydrogen, with the rest made up via heat).

I personally have made hydrogen with SOECs for $2.72/kg using industrial rates for electricity. After a 60% efficient conversion back to electricity, that comes out to $0.116/kWh, which is better than the residential rates for nearly any state.

Electrolysis may be easy, but pressurizing the hydrogen to 700 bar is not. That would probably require a relatively bulky and noisy multistage compressor system.

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

They're pretty efficient, compact, and also no moving parts and no noise.

Depends on how you view density. Energy per unit mass - better than hydrocarbons. Energy per unit volume - worse.

They're technically wrong as energy density is per volume, "specific energy" or "massic energy" is per mass.

Also while the specific energy of hydrogen is good, you have to account for the weight of the container. You can store fuel in a bucket, not so hydrogen.

Compressed hydrogen containers are pretty heavy and that's a bummer, but they're still better than lithium ion by an order of magnitude.

Compressed hydrogen at 700 bar is 2x better volumetric energy density than the best lithium ion batteries on the market today. It's a 4-5x worse than hydrocarbons, but still workable.

2x better than a technology that benefits from the enormous, monstrous economies of scale that lithium ion batteries do is a folly. Every portable electronic device benefits from lithium ion improvements. That's how Tesla was able to launch: use the commodity 18650s from laptops as a relatively low cost cell.

There's a fundamentally better argument to shifting everyone to hybrids (extreme bang for buck for CO2 emission reduction) than FCEVs.

I think for most electric vehicles, batteries are good enough, and probably the better choice. But if you're going hybrid for the extra range, fuel cell technology is basically already there and makes a great range extender.

Battery tech has already won for general automotive applications, even though the number of cars sold is still small. The problem is that while Toyota is a big company, they’re not big enough to deploy the massive hydrogen distribution and fueling infrastructure that would be required in order to make their cars commercially viable around the world. To make this work, they’d need basically every other major auto manufacturer to commit to hydrogen fuel cells as their future non-ICE alternative. Unfortunately for Toyota, virtually every other auto manufacture has already bet heavily on BEV, and against hydrogen. Even if the bet ultimately proves to be a bad one on technical grounds (and I don’t think it is), it seems very unlikely that any set of circumstances can undo it at this point.

Aviation, yes. Aviation is particularly sensitive to mass density, not so much to volume, and the number of joules available per kilogram of hydrogen is nothing short of spectacular.

Unfortunately, while storing hydrogen in a much bigger tank makes economic sense, the lead time on designing, approving, and deploying new lifting-body airframes is too long to help much with the current crisis.

Boeing is not about to start studying lifting-body designs when they are unsure they can survive at all. Airbus, meanwhile, has reason to be complacent now, same reason.

Starting a new airliner company is extremely risky. A hydrogen bizjet would be safer, but wouldn't generate the investment to get hydrogen refueling infrastructure in enough places.

It's too bad, because the room for innovation in a hydrogen-powered air transport is inspiring. Imagine electric-only launch, and a jet engine that only needs to operate above 300 knots. Electric motor propellers mounted anywhere convenient, because of small size and disinclination to explode. Lifting-body designs with enormously more cargo space.

> the most cost-efficient method (of making hydrogen) has been using natural gas to reclaim the hydrogen.

So this wasn’t clean energy to begin with, but now they’re improving:

> solar energy and wind power and wind-generated energy, to make hydrogen using electrolysis

And this sounds definitely less efficient than using electricity directly.

Maybe hydrogen cells will have the density advantage, but at what cost to overall efficiency?

Efficiency only has meaning when applied to a metric. Hydrogen is more efficient for speed of refuel and for energy density, important for things like flying and other long distance work.

If it's a completely closed-loop process, efficiency doesn't matter nearly as much as scalability. Let's say the solar cells are 20% efficient, and electrolysis is 50% efficient...

So 10% of the solar energy makes it into the tank, and some subset of that makes it through the fuel cell...

But we started with water and photons and end up with evaporated water and heat. That's where we were going to be, anyway... If it scales, we're okay.

If that electricity is generated from some non-renewable resource, efficiency is of huge importance. If it's something like wind or solar, it can take a back seat.

If electricity costs $.15/kWh, efficiency might matter. If it costs $.03/kWh, there are way more factors that might matter a lot more.

> And this sounds definitely less efficient than using electricity directly.

Not necessarily. (1) There are losses transporting in storing electricity. If hydrogen was made on the spot, and if the losses storing and transporting hydrogen were smaller (unlikely), that would be a gain. (2) Hydrogen can be produced at solar peaks, when energy prices turn negative, and stored... huge (net) efficiency gain!

Why not charge 2-3x as many cars during the solar peaks? Hydrogen might be in the far future for long-term storage of electricity, but then it will get used to power the grid in times of lacking sun/wind, rather then propelling cars.

Seems like they could have some huge advantages for air travel.

Part of the calculations in flight is that your vessel weighs less as the journey progresses, and for maximum efficiency you really only want to carry as much fuel as you need to get from point A to point B. A plane powered by batteries would never stop being heavy.

Given the eye popping carbon costs of routine flights, clean hydrogen is worth pursuing for aircraft alone.

"> solar energy and wind power and wind-generated energy, to make hydrogen using electrolysis

And this sounds definitely less efficient than using electricity directly"

Of course not. But since you do not want to drive with a cable on your car you allways have to transform the energy.

And the overall efficiency with fuel cells is much lower, but it has other advantages. Like the potential unlimited storage element water (plus iron for the tank walls), compared to limited avaiable batterie ressources.

Just browsed google maps. By my count there are more public EV chargers in a mile radius of my house (east bay) than hydrogen fuel stations in San Francisco, San Mateo, Contra Costa, and Alameda counties combined.

It appears that even among early adopters of new energy sources, the market has spoken.

Hmmm. We could use this same reasoning to prove that the market has spoken in favor of gasoline instead of electricity... Not really a relevant measure when technologies are in transition.

Every house on my block can supply my car with electricity, and I don’t think our neighborhood is unusual. Knocking on someone’s door to borrow a cup of hydrogen would probably be less productive. Borrowing a gallon of gas might be possible, assuming my neighbor hasn’t purchased an electric mower.

Makes much more sense to keep striving for electric solutions so our whole infrastructure can be uniform.

As long as you don't get your hydrogen from natural gas it's just energy storage like batteries. You can make it from electricity or directly via solar-powered reactions.

How efficient is hydrogen as an energy storage medium compared to e.g. Li-Ion batteries?

Hydrogen is 3x more energy dense than diesel, and diesel is about 5 to 10x more energy dense than lithium ion batteries.

By mass or volume?

trying to split the difference between https://en.wikipedia.org/wiki/Toyota_Mirai https://en.wikipedia.org/wiki/Tesla_Model_S https://en.wikipedia.org/wiki/Tesla_Model_3

I mean different shaped/sized cars and all but probably a bit worse? range vs weight seems about on par with the S but worse HP?

but also https://en.wikipedia.org/wiki/Fuel_cell#Theoretical_maximum_... so ?

Efficiency as in energy/kg?

Tesla batteries: 207 watt-hours per kilogram (per google search)

Hydrogen: 120-142 MJ per kilogram (per google search)

Conversion from MJ to watt-hours: 277.778 watt-hours per MJ

120*277.778/207 ~ 161 more compact. Assuming energy conversion losses are equal.

Efficiency in terms of watts out from the medium divided by watts into the medium over a normal duty cycle.

As I said in another comment with some napkin calculations is about 30%. Another user gave a 22% number compared to 73% for batteries. The difference is hugely in favour of batteries.

>As long as you don't get your hydrogen from natural gas (...)

And that's the problem: hydrogen production is a huge waste of energy. Well to wheel efficiency of fuel cells is 22% vs 73% for battery EV (source: https://twitter.com/transenv/status/899976235794788352/photo...)

But BEV cannot really use intermittent energy sources. Not without severely inconveniencing drivers and that's impossible. Hydrogen could be made from intermittent energy sources.

I can't speak to well to wheel, but at the very least current research tech electricity to hydrogen to electricity round trips are 75% efficient.

https://www.nature.com/articles/s41560-019-0333-2

That's pretty much irrelevant, though. The infrastructure for energy distribution is decoupled from its production. Even if hydrogen is split from water via gas today, it can be done by solar or wind tomorrow with no changes to the distribution network.

Current hydrogen production is largely from cracking methane, where natural gas is the feedstock (and also energy source) - it is much cheaper than production from electrolysis.

Mining and processing Lithium isn't that clean either.

There isn't enough infrastructure to support all electric transportation, yet. Also not enough cobalt for the current battery technology.

If you need to build additional infrastructure, why not use a technology that can "recharge" a car within minutes instead of hours?

If you need to build additional infrastructure why not use a technology that's used already.

The charging issue will be solved. There's lots of theoretical and lab proven work showing it can be done and will be achieved.

At least in Germany, the electric infrastructure is mostly up to the task, switching to all electric cars would take only minor improvements, mostly adding more local transformers as the electricity/household is going to raise most of all. Overall, it would mean a mere 20% of more electricity consumed. Also, as charging is a nicely controllable load, electric cars could stabilitze the grid.

Modern electric cars can be recharged in well under an hour, so "hours" is off - unless you charge over night, where it doesn't matter.

For hydrogen on the other side, you would need a complete new infrastructure. Hydrogen fuel stations have nothing in common with conventional gas stations and are really expensive ($1 million/pump). Also, there are restrictions in setting them up in residential areas and while you might be able to refuel a hydrogen car in not much more than 5 minutes, there is a dead time in which the pump cannot serve the next car until the internal pressure has been restored (like 15 minutes).

This is absolutely false. There are currently discussions about net instability due to the missing North-South connection for renewable energy sources. I think you underestimate the sheer amount of power needed for cars, especially in the densely populated southern areas.

This is NOT "absolutely false". There are no net instabilities. Yes, there are lacking north-south connections and they are being built. But they don't have to do with the net capacity as such. They have to do with the new requirement of distributing wind energy from the north to the south and solar energy northwards. This was previously not needed, as every country roughly generated enough energy for their needs and there was little far-distance transportation.

And I don't understimate the amount of power needed for cars, it can be easily calculated, as the numbers of the cars are known as well as the average distance travelled per car. Do the multiplication and you get about 120 TWh of electricity needed per year. That is for converting any single car to electric. Which will take more than 20 years from now on. Compare that to the 600 TWh of production.

This doesn't even calculate in all the electricity spent on fossile fuels, which would be saved.

I've heard people say that battery storage technology improves at 3% per year.

What's the rate of improvement in high-pressure gas or liquefied gas storage? Is the weight of the container decreasing at 3%+ per year?

Toyota is still pushing that? Despite the terrible sales figures?[1] The Toyota Mirai has been on sale in the US for over three years now, and the sales are dropping. Even though they throw in three years of hydrogen.

[1] http://carsalesbase.com/us-car-sales-data/toyota/toyota-mira...

If anything methanol would be a sensible replacement for cars. Much clearer transition path and same end result. But in personal transportation both of those will be a tiny part of the market. Alternative fuels transition should have started 30ish years ago.

I have to imagine that a large part of the interest in fuel cells comes from its defense applications (particularly for submarines). It's been clear for years that for cars it's a total dead end.

https://en.wikipedia.org/wiki/Compressed_air_car has more details

And what do you use to make Hydrogen?

Do fuel cell cars have instant torque like EVs? We recently got an EV, and whenever I drive my gas car, the delay from the gas pedal is extremely annoying now.

There was nothing in the interview about how to keep a hydrogen-fueled car from becoming a miniature Hindenburg disaster in a collision. Is that considered a solved problem?

It might surprise you to learn that gasoline-fueled cars do not usually explode in collisions despite gasoline being far more volatile than hydrogen.

Gasoline is not under pressure and does not burn. Only gasoline vapors burn.

It is certainly not a solved problem. At 300+ bars, just the rupture of a filled tank could cause a significant explosion, even without the hydrogen igniting. If the hydrogen stored in a car ever catches fire, there will be a very quick reaction. In Norway, they closed down all Hydrogen fuel stations after a minor explosion this year (fortunately the main tanks did not explode).

Producing hydrogen using electricity and then convert it back.

For a total efficiency of 22%...

Is there still concern with explosions with hydrogen cells?

Was there ever?

Is there with lithium batteries? Storing large amounts of energy is inherently dangerous.

Generally less than with gasoline. Both require oxygen but hydrogen being stored in a high pressure tank, in year 2000 design concepts with a burst disc going up

Excellent belief Toyota! now start experimenting with some Lithium Ion batteries and a motor. Your neighbor Mitsubishi might help you with those.

I'm not entirely sure if they're wrong. A lot of science for hydrogen fuel cell has not really been studied. I'm curious if we put equal funding into fuel cells instead of battery, today's footing of fuel cell will be much more different. Some cool technology development is the nickel based catalyst (compared to Platinum) but the development is slow since the industrial application is small. Also, there are ways to use fossil fuels directly (with huge caveats) instead of the standard hydrogen gas fuel cells.

The problem with fuel cell vehicles isn’t really the fuel cells. It’s physics. And economics.

Which part of the physics though? I am not entirely convinced that a standard hydrogen PEM fuel cell is the only way to generate energy for a car.

Current fuel cells are said to be in the 40-60% efficiency range. That’s actually not bad, especially if you can do something with the waste heat.

But when you look at the overall round-trip efficiency of the hydrogen lifecycle (production, compression/liquefaction, distribution, storage/leakage, compression again), it’s pretty woeful. Many estimates come in well below 20%.

If we came up with a more efficient fuel cell, it wouldn’t improve the overall efficiency much.

I don't see efficiency as nearly problematic as most seem to make it out to be (as far as the physics goes).

Estimates online roughly estimate that the amount of power from the sun that strikes the Earth in an hour is more than the entire world consumes in an year. Capturing just 0.0001% of that power would cover our needs. And I don't the worse efficiency of hydrogen being problematic in that regards.. but I can understand arguments regarding the economics.

How does this compare to the efficiencies of batteries and gasoline engines? I know otto cycle engines are limited to low 10s% efficiency for just the car as a system as a whole. I haven't seen much about the efficiency in terms of "round-trip" which would be interesting to compare to fuel cells.

I could believe it is better for the environment to make hydrogen powered cars over cars with batteries filled with toxic chemicals.

i don’t think hydrogen powered cars are the future but the next mirai looks much better than the current one https://www.motor1.com/news/375766/2021-toyota-mirai-fuel-ce...

This does look a lot better.

I often wonder if carmakers purposefully make non-IC cars ugly so too many people don't order them. And if that's true, it's probably main thing that Tesla changed: make those cars look awesome.

I still like my Model S quite a bit, but Tesla cars have always looked a little dorky to me, models S, 3, or X (I'm not sure how I feel about the Cybertruck just yet). I certainly didn't get mine for the look. They sure look much better than the i3 or Mirai though.

I always assumed they made their experimental cars ugly so people would notice them and talk about them.

... don't all fuel cells currently known become unusable after 100k miles or less?

Even if hydrogen is the future, Toyota must realize that the present is battery electric. They can continue research on future tech while they rise to the market demands in the present.

They’re wrong

I own 3 Toyota products and a Honda.

I have great confidence in Toyota's vision and engineering excellence. (This coming from a long-time motoring enthusiast.) If they see hydrogen as the future, I take it seriously.

Me too (Toyota owner all my life), although I can’t figure out/don’t understand the emissions fight they’re currently in regarding CA standards.

How do you explain that Toyota supports the Trump administration when they fight California for cleaner energy?

Because Toyota never makes mistakes?

As a fellow Toyota fan and owner I'll take them seriously when they start building industrial scale hydrogen refineries and transportation systems on par with Tesla's supercharger network or better.

Until then these are all compliance cars at best. Doesn't matter how well engineered the implementation is if the design is impractical or of negligible benefit.

The only argument that moves the needle for me on Hydrogen at all is that we have tons of it via seawater, whereas we're not sure if we have enough lithium to produce electric cars for everyone in perpetuity.

Why do they need to be industrial scale? Hydrogen electrolysis is fast, efficient, and can be done on any scale. Centralized production is likely never going to be as efficient as distributed production, simply because there are more losses in distribution than there are in electrolysis.

Hydrogen electrolysys can't be done efficiently at home. And the efficiency even in industrial plants is pretty bad, about 50% to produce hydrogen compressed for car use, and that doesn't include the fuel cell efficiency that brings the total efficiency to about 30%. Batteries have much better efficiency.

I have efficiently done it at an industrial lab with commonly available industrial electricity rates ($.065/kWh). 93% efficiency using a solid oxide cell that was made in 2008. I don't know where you're getting your information but it's wrong. $0.116/kWh after conversion back into electricity.

If I were doing it at home with residential rates, I would be paying $0.125/kWh, which is still feasible, but more expensive than gasoline at the moment. With the way solar prices are dropping, that won't last long.

https://en.wikipedia.org/wiki/Electrolysis_of_water#Efficien... "Efficiency of modern hydrogen generators is measured by energy consumed per standard volume of hydrogen (MJ/m3), assuming standard temperature and pressure of the H2. The lower the energy used by a generator, the higher its efficiency would be; a 100%-efficient electrolyser would consume 39.4 kilowatt-hours per kilogram (142 MJ/kg) of hydrogen,[23] 12,749 joules per litre (12.75 MJ/m3). Practical electrolysis (using a rotating electrolyser at 15 bar pressure) may consume 50 kW⋅h/kg (180 MJ/kg), and a further 15 kW⋅h (54 MJ) if the hydrogen is compressed for use in hydrogen cars."

What about your sources for 93% efficiency?

93% is what I got in my own experiments with an used cell I bought that was made in 2008. My cell is a slight materials variant of the NASA BSC cell which was also tested at 93% efficiency.

https://ntrs.nasa.gov/search.jsp?R=20090013708

Here is a commercially available cell stack that is getting 88% efficiency, despite being a reversible cell (ie not optimized for electrolysis).

http://www.helmeth.eu/index.php/technologies/high-temperatur...

This research system has already demonstrated 75% efficiency round-trip. That is electricity-to-hydrogen-to-electricity. 97% efficient in electrolysis alone.

https://www.nature.com/articles/s41560-019-0333-2

Even with a 93% efficiency adding the energy needed to compress the hydrogen and the fuel cell efficiency you arrive at an efficiency of 48%, pretty much half as the batteries.

I can dispute those numbers too but that is beside the point. Batteries are probably good enough in energy density for most car users, and they do have an efficiency advantage (not as great as you're trying to portray but still better). For 99% of car buyers, that's a better option.

But fuel cells still have a place where energy density is a major concern. That's likely to be the case for commercial vehicles, where long recharge times and low ranges are unacceptable.

I tend to agree - if Toyota wants to develop in hydrogen, why should it matter? And by the way li-on isn’t something you just walk down to Walgreens, buy a palette of a123 batteries and put them in your car. It’s a significant supply chain hurdle that Tesla still has a non trivial cost containment issue despite having a gigafactory or two, giving them an advantage over other manufacturers who don’t have access to the supply chain the way Tesla does. If you look at ev production at scale with numerous parties with at scale products, it’s unlikely that this picture is complete without a significant presence in China. So there you have the real reason why Toyota’s strategy makes sense... what if in ten years, you can’t buy the lithium to make your own batteries? What if you can’t even buy an assembled battery? What if you have to buy the entire vehicle from China to be in the ev space? Not convinced? I present to you as evidence, Tesla gigafactory 3... you guys might want to start thinking about Toyota’s point of view.

Honda is experimenting in hydrogen too, fwiw.

Reply on news.ycombinator.com