There are a lot of stupid things that happen in the name of protecting the environment (e.g. corn-based ethanol) that you just know that someone with power and influence is making money. So it is with Tokyo’s commitment to power the Olympics in 2020 in part with hydrogen. And this is a commitment backed up with $350 million in cash to subsidize hydrogen-powered cars and fueling stations. In addition they will be building a 6,000-unit Olympics village powered exclusively with hydrogen fuel cells. I’ll explain what all of this means, why it sounds like a good idea, and why it’s actually a bad idea.
What all of this means
Hydrogen is the smallest, lightest, and most common element in the universe. Two hydrogen atoms combine into a hydrogen molecule, which is typically a gas. Hydrogen combines with oxygen to make water, with carbon to make hydrocarbons (e.g. natural gas), with carbon and oxygen to make carbohydrates and the list goes on. Pure hydrogen molecules are very rare on Earth, because they would promptly react with oxygen molecules, resulting in water. So if you want hydrogen, you need to make it.
The most common way to make hydrogen is natural gas reforming, where natural gas is combined with steam in the presence of a catalyst to produce hydrogen molecules and carbon dioxide. The hydrogen molecules can then be separated and stored until needed. You can also run an electric current through water, “cracking” the H2O to make hydrogen and oxygen molecules. This is more expensive than using natural gas, but it is greenhouse gas neutral.
Hydrogen gas stores a lot of energy per weight, but very little energy per volume. If the hydrogen is going to be used to power a car, the energy per volume needs to increase, either by compression, which requires energy, or by cooling the hydrogen to near absolute zero where it becomes a liquid, which requires even more energy and expensive storage tanks.
The typical way that hydrogen is used is with a fuel cell, which combines hydrogen with oxygen from the air to generate electrical energy and water. The fuel cell acts like a battery in which you can keep adding fuel. This is not a new technology, the Gemini space missions had fuel cells on board for power, but it is not a mature technology. Currently fuel cells are expensive, but that could change if they were widely adopted.
Why it sounds like a good idea
The problem with most sources of zero-carbon power sources (e.g. wind, solar), is they are intermittent (nuclear isn’t intermittent, but it is too expensive): the wind might be blowing at 2 am when you don’t need power, but not blowing at 2 pm when you do. Hydrogen is not a source of energy, but a way to store electricity by running a current through water when you have excess. When you need more power, you just run the hydrogen through a fuel cell and generate zero-carbon electricity. In 2000 I was a big fan of this idea, and conceptually it can work.
Why it’s actually a bad idea
Like fusion energy, the promise of cheap and reliable fuel cells has been around for years, but it has always just beyond our reach. I remember the excitement around Ballard Power in 2001, when their stock price was $109. They’re still around today, trying to perfect fuel cells for cars, and their stock price is trading at $1.37. Fuel Cell Power, which works on stationary fuel cells like the ones that Tokyo will use to power their Olympic Village, traded at $577 in September 2001. Today they trade at $5.39.
What’s happened between 2001 and today to cause the bottom to fall out of the fuel cell market? Two words: lithium batteries. Looking at the chart below, you see that the cost per Watt-hour (a measure of how much energy a battery can hold) decreased quickly during the 1990s, driven by cell phones and laptops. At the same time the weight of the batteries were dropping. In 2003 Tesla Motors was founded and their Roadster was powered by a 6,831 of laptop battery cells. Prior to this, electric cars were powered by Nickel-Metal-Hydride batteries, which have about 1/4 the energy/weight and wear out much faster.
So why are batteries better than fuel cells? Assuming the hydrogen generated is zero-carbon, the input of both is the same: renewable electricity generated by wind and solar. The round-trip power loss in charging a battery is about 85%. The fuel cell cycle has many stages where energy is lost
- About 35% lost converting electricity to hydrogen
- About 35% lost compressing hydrogen
- About 60% lost converting hydrogen to electricity in the fuel cell
In other words, it takes over 4 times as much renewable electricity to get a fuel cell powered car to go a mile than a battery car like a Tesla. In addition, a fuel-cell car is much more expensive and requires much more infrastructure to be built. The costs of the fuel cell might come down and the efficiency get better, but since there are thermodynamic limits the fuel cell is unlikely to catch up with the battery.
The fuel-cell cars do have one advantage; you can fuel them more quickly (if you can find a hydrogen fueling station). But with 200-mile+ range EVs becoming common, charging overnight is almost always sufficient and the 20-minute level-3 chargers are sufficient in the rare cases when you can’t wait. It’s hard to imagine fuel cells ever catching up with EVs. And it’s hard to understand why the government in Tokyo thinks it’s worth spending over $350 million dollars on this dead-end.