The greatest hurdle for hydrogen fuel is the infrastructure. Every major car manufacturer has built or is designing a fuel cell vehicle. Yet there are only a handful of refueling stations in the United States. None of them are in the state of Washington or its neighbor states: Idaho and Oregon.
The problem is that creating fuel-worthy hydrogen is more complicated than just making hydrogen gas. At ambient pressure, hydrogen takes up a lot of space. 5kg of hydrogen (which contains about the same energy as a tank of gas) would take up 60 cubic meters (2000 cubic feet), more than the volume of 20 cars! To solve this, hydrogen is compressed so that it will take up less space. The DOE recommends 700 bar as the recommended pressure for refueling, which takes up about 1/8th of a cubic meter (4.5 cubic feet). For comparison, a household air compressor may only reach 6 bar. Current refueling stations use large hydrogen compressors to achieve this pressure. They are expensive and break often.
But there is another option. Here at Washington State University, a new method, called cryogenic thermal compression, is being tested by students. This avoids the expensive compressor and may drop the cost of hydrogen fuel significantly. It can even dispense liquid hydrogen with minor modifications. Soon, we will have the first refueling station in Washington State.
With the intent of improving fueling stations, the Hydrogen Student Design Contest was created. Washington State University entered the 2014 contest. Our design for a low-cost, modular, drop-in refueling station took the grand prize. Our approach was to deliver liquid hydrogen to the site on tankers, where our method of cryogenic thermal compression would deliver hydrogen gas at the required pressure to the consumer. Our system was designed to be between 1/4 and 1/8 the cost of current hydrogen refueling stations, lowering the barrier to construction. All of the equipment would fit inside of a standard 40-foot shipping container and could be installed in any location in 24 hours. The fuel cost is expected to be $9.62/kg, or about 16 cents per mile when used in a hydrogen-vehicle. A refuel should take less than 5 minutes, comparable to refueling a gasoline or diesel car. The contest details can be seen here, and the project poster can be viewed here.
This victory paved the way for further research into hydrogen refueling. We are currently working on a second station design which involves the same shipping-container base design and cryogenic thermal compression. But this design involves making hydrogen and liquefying it on the spot. This eliminates the need for refueling tankers and allows fuel-grade hydrogen to be produced and dispensed anywhere in the world.
When a liquid boils, it tends to expand. But if the liquid is sealed in a container where the volume can’t change, it will build up pressure instead. That is the basic idea behind cryogenic thermal compression. We take hydrogen gas, cool it until it liquefies, then seal it in a container and allow it to boil off. Once it has finished boiling, it will reach a pressure far greater than the 700 bar we need to refuel. By using this method, our only need for compression is to run the cooling cycle. This requires less than 100 bar.
This method does not come without its own difficulties. Cooling anything to liquid hydrogen temperatures is difficult, especially on an industrial scale. At atmospheric pressure, hydrogen becomes a liquid at 20 Kelvin (-250°C, -420°F). We can cool it most of the way with liquid nitrogen (which is cheap because it is a byproduct of oxygen purification). But to get the final cooling, we need to be more creative. Current cycles use throttle valves or expansion turbines. But throttles only work in certain temperature ranges and expansion turbines are difficult to engineer for such low temperatures. So we are researching the use of a Heisenberg Vortex Tube.
The Heisenberg Vortex Tube (also known as the Ranque-Hilsch vortex tube or simply, vortex tube) is a device which takes an inlet gas and separates it into two streams: one which is hotter than the incoming gas and one that is colder. By controlling how much of the gas goes out either end, the cooling power can be adjusted. Vortex tubes work at any temperature and require no moving parts. You can read more about vortex tubes here and here.
Currently, the H2-Flo project is constructing the cold-end of the liquefier. We are nearly done building a proof-of-concept for nitrogen. We will use one of our gas boosters as well as our throttle valve, heat exchangers, and vortex tubes. Without any coolant, we expect to liquefy nitrogen. Once this is achieved, we can begin testing with hydrogen, using the same cycle except with liquid nitrogen as a pre-coolant. Recent advancements include:
Heat Exchangers: A counter-flow, parallel-tube-type heat exchanger was designed using EES (Engineering Equation Solver). This was built and tested for performance. It achieved an effectiveness of 91-97% at liquid nitrogen temperatures. It should be higher at liquid hydrogen temperatures.
Gas Boosters: Although we don’t need as high of a pressure as current refueling stations, we still need a compressor. From older projects at WSU, we have inherited two Haskel gas boosters. These use compressed air at 90psi to compress another gas to a much higher pressure. One of them has been pressure tested, as seen here.
Dewar: A 250L liquid hydrogen container has been purchased to contain our product. A special flange has been designed and built for mounting on all of our attachments.
Purifier: An improvised hydrogen gas purifier has been constructed and tested. It makes use of liquid nitrogen. This is to decrease the temperature so that any gaseous contaminants (nitrogen, CO, CO2, etc.) will liquefy and fall out of the gas, improving purity.
Vortex Tube: Vortex tubes are continuing to be modeled and improved, as explained here.
3D Printed Micro Liquid Hydrogen Tank: A miniature liquid hydrogen tank has been developed which can be 3D printed. This allows easy transport of fuel for small vehicles such as drones. Read about the design here.