As the world grapples to eliminate fossil fuels from our energy diet, electric cars have seen an incredible boom over the past few years. Last year, over one million electric cars were sold around the world. The number of Nissan Leafs, Teslas, and other electric vehicles in circulation worldwide is now more than three million.
And while there are many brands of electric car to choose from, there are only two choices when it comes to powering electric vehicles: fuel cells or batteries. Both produce electricity to drive electric motors, eliminating the pollution and inefficiencies of the fossil fuel powered internal combustion engine. Both hydrogen and electricity for batteries can be produced from low or zero carbon sources, including renewable energy like solar and wind, and therefore both are being pursued by car manufacturers and researchers as the possible future of electric vehicles.
However, a great debate is being waged by supporters of each technology. Elon Musk has called hydrogen fuel cell technology “incredibly dumb,” claiming they’re more of a marketing ploy for automakers than a long-term solution. In contrast, Japan has announced its intention to become the world's first hydrogen society, with the Japanese government and the auto industry working together to introduce 160 hydrogen stations and 40,000 fuel-cell vehicles by March 2021. So which is actually better?
At first glance, hydrogen seems like an extremely clever way to power a car. Compressed hydrogen has a specific energy (aka energy per unit mass) of nearly 40,000 watt hours per kilogram. Lithium ion batteries at best have a specific energy of just 278 wh/kg, but most fall around 167 wh / kg. That's 236 times as much energy per kg for hydrogen. And because of its energy density and lightweight nature, compressed hydrogen and fuel cells can power cars for extended ranges without adding much weight.
The designers of electric vehicles are caught in a catch 22 with energy density and range. Each extra kilogram of battery weight to increase range requires extra structural weight, heavier brakes, a higher torque motor, and in turn more batteries to carry around this extra mass, This weight compounding limits how far a battery powered vehicle can travel, until new technology can help reduce the weight of the batteries.
For hydrogen fuel cell vehicles, this weight compounding is not an issue. Additionally, a hydrogen fuel cell vehicle can be refueled in under 5 minutes, where a battery powered electric vehicle, like the Tesla model S, takes over 3 hours to fully recharge. When looking at the range and refuel times hydrogen can offer, you can see why some car manufacturers are investing in this technology. On the face of it. Hydrogen is a clear winner, but it falls behind when we start considering the end-to-end production process. While both batteries and hydrogen fuel cells are both forms of electricity storage, the cost differ drastically.
Fully charging a Tesla Model 3 with a 75 kiloWatt hour battery, costs between 10-12 dollars depending where you live. With a rated range of 500 kilometers, that’s between 2 and 2.4 cent per kilometer. A great price. The hydrogen from this station cost $85 dollars to fill the 5 kg tank of the Toyota Mirais on site, which had a range of 480 kms. That’s 17.7 cent per kilometer, 8 times the price. And here lies the problem, Hydrogen simply requires more energy to produce.
To understand the economic viability of hydrogen let’s dig deeper into the production process. Before any hydrogen vehicle can hit the road, you first need to produce the hydrogen, but hydrogen is not a readily available energy source. Even though hydrogen is the most abundant element in the universe, it is usually stored in water, hydrocarbons, such as methane, and other organic matter. One of the challenges of using hydrogen as an energy storage mechanism comes from being able to efficiently extract it from these compounds.
In the US, the majority of hydrogen is produced through a process called steam reforming. Steam reforming is the process of combining high-temperature steam with natural gas to extract hydrogen. While steam reforming is the most common method of industrial hydrogen production, it requires a good deal of heat and is wildly inefficient. Hydrogen produced by steam reforming actually has less energy than the natural gas that the steam reforming began with. And while hydrogen fuel cells themselves don’t produce pollution, this process does. So if we want to assume a future scenario with as little carbon emission as possible, this method won’t cut it.
Another method to produce hydrogen is electrolysis - separating the hydrogen out of water using an electric current. While the electricity needed for this process can be provided from renewable sources, it requires even more energy input than steam reforming. You end up losing 30% of the energy from the original energy put in from the renewables when you carry out electrolysis.
So we are sitting at 70% energy efficiency from hydrogen fuel cells if traditional electrolysis is used, before the car even starts its engine. A slightly more efficient method of producing hydrogen is polymer exchange membrane electrolysis. Using this method, energy efficiencies can reach up to 80%, with the added benefit of being produced on site, which we will get to in a moment. But this is still a 20% loss of energy from the original electricity from the renewables. Some experts say the efficiency of PEM electrolysis is expected to reach 82-86% before 2030, which is a great improvement, but still well short of batteries charging efficiency at 99%. A 19% difference in production costs doesn’t explain the difference in costs yet, so where else are we losing energy.
The next hurdle in getting hydrogen fuel cell vehicles on the road is the transport and storage of the pure hydrogen. If we assume the hydrogen is produced on site, like it was for this petrol station, then we eliminate one energy sink, but the cost of storage is just as problematic. Hydrogen is extremely low density as a gas and liquid, and so in order to achieve adequate energy density, we have to increase its actual density. We can do this in two ways. We can compress the hydrogen to 790 times atmospheric pressure, but that takes energy, about 13% of the total energy content of the hydrogen itself. Alternatively we can turn hydrogen into liquid, cryogenically. The advantage of hydrogen liquefaction is that a cryogenic hydrogen tank is much lighter than a tank that can hold pressurized hydrogen. But again, hydrogen's physical properties means hydrogen is harder to liquefy than any other gas except helium. Hydrogen is liquified by reducing its temperature to -253°C, with an efficiency loss of 40%, once you factor in the added weight of the refrigerators and the refrigeration itself. So pressurisation is a better option at a 13% energy loss.
Once the hydrogen is produced and compressed to a liquid or gas, a viable hydrogen infrastructure requires that hydrogen be able to be delivered from where it's produced to the point of end-use, such as a vehicle refueling station. Where the hydrogen is produced can have a big impact on the cost and best method of delivery. For example, a large, centrally located hydrogen production facility can produce hydrogen at a lower cost because it is producing more, but it costs more to deliver the hydrogen because the point of use is farther away. In comparison, distributed production facilities produce hydrogen on site so delivery costs are relatively low, but the cost to produce the hydrogen is likely to be higher because production volumes are less.
While there are some small-scale, on-site hydrogen production facilities being installed at refuelling pumps. until this infrastructure is widespread, we have to assume that the majority of hydrogen is being transported by truck or pipeline, where we know that energy losses can range from 10% up to 40%.
In comparison, assuming that the electricity that we use for charging the batteries comes completely from renewable resources (like solar or wind), we just have to consider the transmission losses in the grid. Using the United States grid as a reference for typical grid losses, the average loss is only 5%. So in the best case scenario for hydrogen, using the most efficient means of production and transport, we lose 20% of energy during PEM electrolysis, and around 13% for compression and storage, amounting to a 33% loss. In other systems, this could be as much as 56%. For battery power, up to this point, we have lost just 6% to the grid and recharging. Bringing our best case efficiency difference to 27% and our worst case to 50%. The next stage of powering electric vehicles is what is called the tank to wheel conversion efficiency.
For hydrogen fuel cell vehicles, once the hydrogen is in the tank, it must be re-converted into electric power. This is done via a fuel cell, which essentially works like a PEM electrolyser, but in reverse. In a PEM fuel cell, hydrogen gas flows through channels to the anode, where a catalyst causes the hydrogen molecules to separate into protons and electrons. Once again the membrane only allows protons to pass through it, while electrons flow through an external circuit to the cathode.This flow of electrons is the electricity that is used to power the vehicles electric motors. If the fuel cell is powered with pure hydrogen, it has the potential to be up to around 60% efficient, with most of the wasted energy lost to heat. Like hydrogen fuel cells, batteries also come with inefficiencies and energy losses. The grid provides AC current while the batteries store the charge in DC. So to convert AC to DC, we need a charger. Using the Tesla Model S as an example, its peak charger efficiency is around 92%. The Tesla model S runs on AC motors; therefore, to convert the DC current supplied by the batteries into AC current, an inverter has to be used with an efficiency of roughly 90%. Additionally, lithium ion batteries can lose energy due to leakage. A good estimate for the charging efficiency of a lithium ion battery is 90%. All of these factors combined lead to a total efficiency of around 75%. However, hydrogen fuel cell vehicles also have some of these same inefficiencies. Any kind of electrolysis requires DC current, and therefore, a rectifier will be required to convert the AC current from the grid to DC. The conversion efficiency here is 92%. We also need to convert the DC current produced by the fuel cell to AC to power the motor through an inverter with an efficiency of 90%. Finally, the efficiency of the motor must be considered for both fuel cell and battery powered vehicles. Currently, this is around 90-95% for both of them, which is amazing when you consider that internal combustion engines running on petrol have an efficiency of only around 20-30%. If we add up all these inefficiencies and compare current generation batteries, to the best and worst case scenario of current gen hydrogen. We can see how they measure up. Even with the BEST case scenario. Not taking into account any transport due to onsite production, and assuming very high electrolysis efficiency of 80%, and assuming a HIGH fuel cell efficiency of 80%, hydrogen still comes out at less than half the efficiency. The worst case scenario is even worse off.
So while you may be able to go further on one fill-up of hydrogen in your fuel cell vehicle over a battery powered electric vehicle, the cost that is needed to deliver that one fill up would be astronomically higher compared to charging batteries due to these energy losses and efficiencies. Based on our worst case scenario, we would expect the cost per kilometre to be about 3.5 times greater for hydrogen, but as we saw earlier it’s actual 8 times the price. So additional costs of production unrelated to efficiencies are obviously at play. The cost of construction of the facility is one and the profit the station will take from sale is another. For now, these inefficiencies and costs are driving the market, where most investment and research is going into battery powered electric vehicles. So which wins?
Both are equally more green than internal combustion engines, assuming equal renewable resources are used to power them. Fuel cells allow for fast fill up times and long ranges; a big advantage. But battery powered vehicles might catch up in range by the time there are enough hydrogen stations to ever make fuel cell vehicles viable.
While fuel cells are efficient relative to combustion engines, they are not as efficient as batteries. They may make more sense for operation disconnected from the grid. For now, battery powered electric vehicles seem to be the sensible choice going forward in the quest for pollution free consumer transport. As battery-powered cars become more common, we’re also starting to see self-driving cars become the norm. Gradually, the job of driver is slowly automated away