Slashdot Mirror
Tesla Co-Founder Says Hydrogen Fuel Cells Are a 'Scam' (electrek.co)
Marc Tarpenning, co-founder of Tesla, believes hydrogen fuel cells are a "scam". Tarpenning, who is not with Tesla anymore appeared on Internet History Podcast last week to outline a number of issues with hydrogen fuel cells. He said (via Electrek blog): If your goal is to reduce energy consumption, petrol or whatever resource, you want to use it as efficiently as possible. You don't want to pick something that consumes a lot for whatever reason, and hydrogen is uniquely bad. There's a saying in the auto industry that hydrogen is the future of transportation and always will be. It's a scam as far as I can tell because the energy equation is terrible. People will say that hydrogen is the most abundant element in the universe, but it's abundant out there in the universe not here. We live on a planet where hydrogen is super reactive -- it's bound up into everything. It's bound up into water, wood and everything else. They only way that you get hydrogen requires you to pour energy into it to break it from the chemical bonds. Electrolysis is the most common method. You put electricity in water and it separates it, but you are pouring energy in order to make hydrogen, and then you have to compress it and that takes energy, and then you have to transport it to wherever you actually need it, which is really difficult because hydrogen is much harder to work with than gasoline or even natural gas -- and natural gas is not that easy. And then you ultimately have to place it into a car where you'll have a very high-pressure vessel which offers its own safety issues -- and that's only to convert it back again to electricity to make the car go because hydrogen fuel cell cars are really electric cars. They just have an extraordinary bad battery.Here's the podcast.
And a poor way at that. Cannot agree more with his assessment. I have been saying things like this for years, could never understand why people thought it was a great idea.
Hydrogen is an energy *vector*, not an energy source. The energy must come from somewhere - natural gas usually - and, as TFA's author points out, the efficiency of the entire chain from energy source to the wheels is quite insanely bad.
"A door is what a dog is perpetually on the wrong side of" - Ogden Nash
There have been recent developments in using a catalytic reaction to generate H2O2 (hydrogen peroxide) from sea water and sunlight. Hydrogen peroxide is much more reactive (i.e. easier to separate the hydrogen) than water, and much more energy dense at room temperature and atmospheric pressure than hydrogen alone. This is the only promising solution to the hydrogen problem that I've seen.
Hydrogen has advantages and disadvantages. This post seems to list all the disadvantages, and none of the advantages.
As noted, hydrogen isn't an energy source-- it's an energy storage medium. But then, Tesla's batteries aren't an energy source either-- they are an energy storage system that takes energy from somewhere else. Hydrogen can be produced remotely, and shipped to where it can fill up cars in gas pipelines. Electricity can be produced remotely, and shipped to where it charges car batteries by wires. Same principle, different medium. (In principle, electricity could be shipped to the charging station, and produce hydrogen on-site by electrolysis-- but it's probably more efficiency to make the hydrogen remotely.)
Hydrogen's advantage is that it is extremely light: you can react it with air, you don't have to carry the air around, and hydrogen is the lightest thing there is to react with air. Weight-wise, hydrogen is the best possible fuel.
Hydrogen's disadvantage is that it is extremely light: it is hard to store a lot of it because the density is very low. You can do a little better if you go all the way to liquid hydrogen-- but nobody is going to do that for a car (not, at least, until cryo storage gets a lot better)-- and even liquid hydrogen has about the density of the lightest grade of styrofoam. So, the tanks are either big, or high pressure-- or both.
Advantage and disadvantages. This is what makes an engineering trade off.
With current technology, I'd go with batteries. Two or three generations down the line? Your estimate of technology progress is probably as good as mine.
http://www.geoffreylandis.com
Completely agreed that hydrogen fuel cells don't make sense from an energy standpoint (unless you're liberating the hydrogen from a high Gibbs free energy source like methane, or if you're getting the energy from a non-polluting source like nuclear or wind (in which case the hydrogen is basically acting like a battery). The transport argument is more specious. Yes transport and storage is worse than for gasoline (pretty much everything is worse, which is why we use gasoline). But electricity isn't much better - easier to transport, more expensive to store, and much harder to transfer from one storage medium (the charging station) to another (the car battery).
From an energy efficiency standpoint. the cost advantage of operating an electric car is only slightly due to improved energy efficiency. The vast majority of the price differential is due to the extremely low price of coal and natural gas relative to gasoline.
An ICE engine can hit about 30% efficiency. An automatic transmission is about 90%-95% efficient (pretty impressive considering it's just squirting fluid at a turbine).
Newer coal plants are about 40% efficient. Natural gas plants are about 60% efficient. Split the difference and go with 50%. Power lines are about 98% efficient. Real-world charging efficiency of the Tesla is about 80% (1/1.26 = 0.79). That is, 80% of the electricity from your wall socket goes into the battery, the other 20% becomes heat. I can't find any numbers for discharge efficiency, so let's call it 100% for now. And electric motor efficiency is about 90%-95%.
Electrolysis of hydrogen from water is about 65% efficient in the lab, closer to 30% in practice. Efficiency of hydrogen fuel cells is close to 90% in the lab, but is closer to 50% for industrial applications like a car motor. Tally it up and you get:
ICE: 30% * 92.5% = 27.8% efficient
EV: 50% * 98% * 80% * (100%) * 92.5% = 36.3% efficient.
H2: 30% * 50% = 15% efficient (did I mention hydrogen doesn't make sense from an energy standpoint?)
So really not that big an efficiency difference between the EV and ICE. If battery discharge efficiency is also 80%, then the EV is basically identical to an ICE in overall energy efficiency. Yes if solar and wind come down in price to match or beat coal, then you can drop the 50% at the front. But wind is still about 1.5x-2x the price of coal, and solar about 5x-7x the price. Nuclear would be the obvious solution, but the people supporting EVs seem hell-bent on shutting down nuclear.
Now look at the fuel price side.
Coal costs about $50/ton, and contains about 24 GJ/ton. That's $2.08 per GJ. Gasoline costs about $2/gallon and contains about 120 MJ/gallon. That's $16.67 per GJ. Almost an order of magnitude more.
So there you have it. EVs are only 1.1x-1.3x more energy efficient than ICE cars. But their fuel source is 8x cheaper. That's why EVs are cheaper to operate than ICE vehicles. If more of our electricity production shifts away from fossil fuels and towards non-polluting sources, then that also makes the hydrogen economy more viable. EVs and hydrogen in inextricably linked in this way.
Bulk commercial electrolysis hydrogen could be very cheap is there was real demand for it. Steam electrolysis is "over 100% efficient" if you're using waste steam from the power plant (you steal back some of the huge wasted heat of vaporization, making the electricity input low), so making it in vast quantities as a byproduct of power generation would work well. But no one does it because there's no distribution network.
I'm not sure how much sense home electrolysis would make from an efficiency point of view, but from a "I want off the grid, even if it costs more" point of view it's great. You can power a heavy SUV or jacked-up 4x4 pick-em-up-truck with it, which we won't see with battery power any time soon.
Socialism: a lie told by totalitarians and believed by fools.
The Tesla batteries are actually outlasting their estimates. At 8 years and 100,000 miles (the standard warranty), they still hold over 85% of their charge; and the Tesla Model S usually only allows an 80% charge, unless you tell it to fill up for a long trip. For commuters or a system with a high availability of Tesla's 20-minute super chargers, you'd still be going 100 miles between a charge (about an hour and a half) at 30% battery life, 37 years into the car's life; commuters particularly are doing under 50 miles per day and coming home to charge in between.
In the short-term, long trips are unfeasible on a new electric car due to lacking infrastructure and long charge times; in the long-term, severely-degraded cars 40 years and half a million miles into their life could still make cross-country trips on the original battery.
A properly-maintained gasoline engine often can't make 250,000 miles without a rebuild; the car is considered old and dead after 100,000 miles, but that's kind of dumb. breaking 400,000 miles on a Tesla battery in the above scenario should be doable. My car, at 100,000 miles, is getting 78% of its original range; I'll probably have sunk around $7,000 total into maintenance (including transmission maintenance--Teslas don't have one) after I've had the engine's systems repaired to get it back to its full range of 320 miles on an 11 gallon tank. Mind you the car's 12 years old; it's been cheap to maintain. Just a 2004 Mazda 3, and that's still on the L-Series engine (a Ford make; Mazda switched to their own engine for the 3 series, which is superior in terms of early-life maintenance costs, and thus total lifespan).
Right now they're on rough par for lifespan; electric cars are doing better for fuel costs and maintenance in most use cases.
Support my political activism on Patreon.
Don't think about this like a financial engineer, not a civil engineer. It doesn't matter what makes the most sense from a technical perspective, what matters is not disrupting cash flow for entrenched industries.
Well I am an accountant and an engineer. You are right about progress following the path of least economic resistance but I think you have that path misidentified. Part of the flaw in your argument is in thinking there is just one big industry in the fight here. Basically you will be pitting the electric generation companies against the refining companies. Except not really or at least not immediately.
The least disruptive technology is actually plug in hybrids because it bridges both types of infrastructure. It can behave exactly as current gas powered cars do. As battery technology gets better you'll see the range of the electric vehicles go up and the charging times go down and the charging infrastructure get built out. Eventually you reach a tipping point.where it makes sense to go all electric and drop the second engine. In the mean time the gas station still sells fuel, the electric company gets time to beef up their already existing infrastructure and you don't have to introduce any truly different infrastructure like compressed gas or specialty chemicals.
To make fuel cell vehicles work you have to build out all new infrastructure everywhere all at once and to date they are behind the curve in performance. There is no consensus on what form hydrogen should be distributed in or how it should be implemented on the vehicle. With hybrids you can incrementally solve the problem today. I don't think it is very likely that fuel cells will make some miraculous technological leap that will make building out all that extra infrastructure economically worthwhile in the near future.
The primary constituents of lithium ion batteries are actually relatively straight forward to recycle (Lithium, Cobalt, Copper, Nickel, Graphite (carbon), Aluminum are the most common elements), and the batteries have useful lifetimes well beyond 10 years. However, the capacity degradation curve for Li-ion chemistries is mostly logarithmic, meaning even after it's done as an EV power store (say at 2/3s original capacity which might be reached after 10 years), it can last another 20+ years as grid storage, which doesn't require anywhere near the same weight/size to energy ratio.