Why You Don’t Want a Hydrogen-Powered Car
Or a Hydrogen-Powered Anything, Really
Let’s start with a bang. Okay, maybe it’s more of a fwoosh than a bang, but it’s still dramatic. In 2001, Michael Swain of the University of Miami produced a video comparing the effects of two fires: one in a hydrogen-powered car, and another in a gasoline-powered car. After about three and a half minutes, the gasoline-powered car is a smoldering husk, while the hydrogen car is unscathed. Since then, proponents have been pointing to the results of that experiment and claiming that it shows hydrogen cars to be safer, at least in a fire scenario.
It shows no such thing. The hydrogen leak was simulated by opening a valve that happened to be pointing straight up. Would that car look so safe had the 5-meter-long flame been aimed at the passenger cabin, or at a sidewalk full of pedestrians? I’m not at all suggesting the test was rigged, but it certainly wasn’t an accurate representation of the fire danger in hydrogen cars.
And fire is not the only threat posed by hydrogen. I work in a laboratory where we use compressed gases on a daily basis. For the most part, these gases are stored in heavy steel cylinders at 2000 psi or less, and we must undergo training before we’re allowed to handle them. Mythbusters did an episode that shows why: A punctured cylinder has enough energy to break through a cinder-block wall. A 2006 accident at Texas A&M involved a liquid nitrogen cylinder. When safety features were disabled and pressure climbed to around 1200 psi, the tank destroyed laboratories on two floors of the chemistry building.
The entrance door and wall of the lab were blown out into the hallway, [and] all of the remaining walls of the lab were blown 4-8" off of their foundations. All of the windows, save one that was open, were blown out into the courtyard.
Now consider: In order to carry a useful amount of hydrogen, a tank must be pressurized to 5,000-10,000 psi. Even an inert gas at that pressure has the potential to destroy a car — and any nearby pedestrians — if it’s catastrophically released.
Enough about danger. Aren’t hydrogen cars much more environmentally friendly? After all, the pollution created by hydrogen fuel cells or combustion is clean water. That certainly beats the carbon-dioxide-spewing cars that dominate our roads now. But there’s still the problem of getting that hydrogen into the tank to begin with, and the problem of how little energy we obtain from using it.
As an exercise, let’s use solar energy to power two different vehicles, and then decide which one we’d rather have. First, we’ll generate hydrogen for our car using photoelectrochemical (PEC) cells. Second, we’ll use photovoltaic (PV) cells to charge the batteries of an electric vehicle.
The best PEC cell in a laboratory today can convert about 3 percent of solar energy striking its surface into energy stored in hydrogen. Once the hydrogen is produced, it has to be stored, and I don’t have figures on how much energy it takes to run the necessary compressor. Let’s be generous and say we can use 10% of the gas we have to compress the rest to a useful pressure. In this exercise, we’re fueling a hydrogen car from our own (future) backyard PEC panels, so there’s no transportation cost. Now we have 0.03 × 0.9 = 0.027, or 2.7 percent of the energy we started with, and we’re ready to power the car.
BMW is getting just over 40 percent efficiency with hydrogen combustion, or we could use fuel cells at 40-70 percent efficiency to drive a 90-95 percent efficient electric motor. Using the most optimistic of those figures, 0.7 × 0.95 = 0.665, or 66.5 percent of the paltry 2.7 that we managed to store. That’s about 1.8 percent of incident solar energy actually moving the car. There are certainly more efficient ways to produce hydrogen than PEC cells, but you have to pay for the energy input to all of them, and the inefficiency of actually using hydrogen still weighs heavily.
Up next is our electric car. You can go out today and buy PV panels that convert 15-20 percent of incident solar energy into electricity. That electricity is easily and safely storable in a battery, and the charging process is 90-95 percent efficient. Again, assume we’re charging from our backyard PV panels, so there’s no transmission cost, and now the car is charged and ready to go, with (pessimistically) 0.15 × 0.9 = 0.135, or 13.5 percent of incident solar energy at our disposal.
Discharging the battery costs about 5-10 percent, and then turning an electric motor is another 5-10 percent hit. Being pessimistic again, 0.135 × 0.9 × 0.9 = 0.109, or 10.9 percent of the energy hitting our solar panel is being used to move the car.
Now it’s time to choose. Do you want an electric car that uses 10.9 percent of available solar energy, or an arguably dangerous hydrogen car that uses only 1.8 percent, taking 6 times as long to “charge”? Do you want 10,000 psi tanks in the hands of untrained people, or would you rather they used the batteries we all carry around in our cell phones?
The above exercises were played out under ideal conditions. In real-world use, hydrogen power is less than 40 percent efficient, from energy in to energy out. Do we really want to invest in new infrastructure for a technology that wastes 60 percent of its input? Do we really want a hydrogen-powered anything?
Email me when Tom Mangan publishes or recommends stories