Bjorn’s Corner: Sustainable Air Transport. Part 19. Fuel Cell propulsion systems

By Bjorn Fehrm

May 13, 2022, ©. Leeham News: Last week, we looked at advanced developments for hydrogen-burning gas turbines.

Now we look at the alternative hydrogen-based propulsion system, which uses a Fuel Cell to convert the energy in hydrogen to electric power that drives motors to spin propellers or fans, Figure 1.

Figure 1. The principal parts of a fuel cell propulsion system compared with other electric motor-based systems. Source: Leeham Co.

Fuel Cell propulsion system

Commercial fuel cells are almost 100 years old. I described the working principle here and the different fuel cell types and their characteristics here.

As a recap (Figure 2), the gaseous hydrogen enters on the Anode side, where its positive protons diffuse through the Proton Exchange Membrane (PEM), and the electrons take the external path to the Cathode side (thus forming the fuel cell current), where the reduction of H2 and O to H2O takes place.

Figure 2. The working principle of a PEM fuel cell. Source: NASA Fuel Cell report.

The PEM fuel cell type can be developed to have acceptable volumetric and mass characteristics so that it can be used in aeronautical applications.

There are two types of PEMs fuel cells for our purposes, the classical or Low-Temperature fuel cells (called LT-PEM fuel cells or just fuel cells from now on) and the High-Temperature PEM fuel cells (called HT-PEM fuel cells).

The cells have in common that they produce more heat than electric energy from the process H2+O = H2O. For air vehicle applications, the heat management of the fuel cell propulsion system is its major problem. The magnitude is dependent on what type of PEM fuel cell is used.

Low-Temperature PEM Fuel Cells

In the classical low-temperature PEM fuel cell, the Proton Exchange Membrane is dependent on stringent water management inside the cell for a proper function; not too little (conductivity suffers), not too much (access of O2 to the process suffers). The water management limits the cell’s operating temperature to around 80°C and makes it sensitive to freezing.

The large heat flow with a low temperature from the cell makes heat management difficult for LT-PEM cells in aircraft applications. It creates a mass, volume, and drag problem (large air-cooled heat exchanger surfaces are necessary). The heat is manageable for stationary and rolling vehicle applications as these have fewer constraints on mass and volume for the heat management system.

The LT-PEM cell is a mature product with several manufacturers and application areas (ground transport, portable power generation). The key problem area for ground transportation (trucks, buses, cars..) is the cost, partly because of the platinum used as a catalyst in the reaction process. The research focus for LT-PEMs outside aerospace is, therefore, cost down, not mass or volume down (as aerospace would like it).

High-Temperature PEM Fuel Cells

The HT-PEM cell has a typical heat exhaust temperature of 180°C, and its PEM is not dependent on the correct humidity. The PEM uses absorbed Phosphoric acid as the electrolyte. The type has been developed over the last 30 years for stationary applications, primarily as it’s less dependent on the purity of the hydrogen. It can use derivative fuels like methanol that is reformed into a hydrogen-rich gas.

The higher operating temperature eases the cell’s cooling as the heat exchanger surface area depends on the temperature difference, all other parameters being equal. Therefore, the HT-PEM is interesting for aeronautical applications. It’s a less mature variant, therefore it needs further development before the first aircraft applications.

The balance of plant

A PEM fuel cell needs several external systems to operate correctly. The management of these systems and the fuel cell is called Balance Of Plant, BOP.

Figure 3 shows the systems needed for an LT-PEM system. The graph is from the same NASA report as Figure 2. I included the figure text as it explains the different circuits of air and heat management necessary for the function of the stack.

Figure 3. The Balance Of Plant systems needed for an LT-PEM system. Source: NASA fuel cell report.

We need:

  • A compressor, cooler (compressed air gets hot), and humidifier for the external air that enters the cathode side of the stack to supply the O2 for the process.
  • We also need a turbine that sucks reacted air and H2O vapor through the humidifier and dumps it overboard.
  • A water extraction collector, tank, and pump for the PEM water management.
  • A liquid high capacity high temperature (80°C) cooling system to transport the excess heat from the stack.
  • A supply of gaseous H2 at the correct temperature to the stack from our liquid (-253°C) hydrogen tank.

All these loops must be controlled to give the fuel cell stack optimal working conditions at all phases of flight.

An HT-PEM system can skip the humidifier in the air supply loop, and the water out from the stack is an open-loop waste process. Its heat management system can also be made lighter and smaller.

In the following Corners, we look deeper into PEM fuel-based propulsion systems and their integration into an aircraft.

20 Comments on “Bjorn’s Corner: Sustainable Air Transport. Part 19. Fuel Cell propulsion systems

  1. All those coolers look very ‘lossy’. Why not use the liquid hydrogen fuel to do the cooling of the air going into the stack – then none of the heat would have to go overboard. It probably wouldn’t need so much humidification either.
    Also, why have an electric motor running a compressor – why not bleed compressed air from a gas turbine core and use that (Fig 1 does not show this arrangement)?
    There’s lots that could be done with hot HP air/vapour exhausted from the stack – for instance, putting it through a turbine/generator to make more electricity or find a way of getting it into the turbine exhaust to increase the mass flow & thus its output.

    • You are right that having LH2 it is a powerful cooler and no need for cooling fans. Air motors is very hard to make as efficient as modern electrical motors. You can use hot and mosit air, one application could be the air cycle machine together wioth LH2 cooling as you need cabin air pressure and air exchange.

      • Still the volume -density problem isnt there and you want to add heat exchangers for the LH2 and the excess heat. Cooling fans would still be need to draw away the heat from the unit into the actual heat exchanger itself.

        From Bjorns previous look
        ‘. If we include system losses due to the fuel cell’s cooling and control, the system-level efficiency drops around 10%-15%.’

        Theres still a lot of work to be done having the entire system running in non pressurised part of the fuselage. Thats were the likely first start as an APU direct replacement will cover

        • -The first generation of fuel cell will almost certainly be gaseous compressed hydrogen LT-PEM ( 80 Celsius ) so using a cryogenic heat exchanger is moot. The cooling requirements would be challenging in a hot (middle eastern) takeoff with a temperature difference of 45C to 80C. That itself may may require cryogenic hydrogen or HT-PEM.
          -I suspect cryogenic systems for smaller aircraft will however develop faster than anticipated and be more cost effective than expected. Miniature liquefaction plants etc.
          -The development of ultra lean internal combustion engines (HCCI or RCCI) engines and ultra lean hydrogen internal combustion engines may be more attractive than LT-PEM given the efficiency will be as good and NOX emissions are controllable.

    • You can turbo compound anything but the absolute efficiency Carnot efficiency possible is
      n is efficiency
      Tc temperature of the cold side ie heat rejection temp in Kelvin.
      Th temperature of the hot side.
      With a working temperature of 80C or 353K and a heat rejection of say 30C or 303K there is not much to be recovered, nevertheless it has been considered for automotive use. With a working temperature of 180C or 453K the scope for turbo compounding is more significant. SOFC solid oxide fuel cells at 800C will be astoundingly desirable to turbo compound.

    • Interesting tidbit about the scaled flight test model from my old friends the NLR. A simple idea – make a windtunnel model remote-control flyable – instantly validating calibration between wind-tunnel testing and actual flight characteristics. So simple I’m surprised I’d not heard of (or thought about) it before.

      Thanks for that…

      • Its been worked in the reverse too:
        MBB project Lampirydiae scaled copy was flown in a windtunnel in the Netherlands.

        • @Uwe, could you clarify that for me? Was a full scale Lampirydiae or a subscale Lampirydiae flown in the Dutch wind tunnel?

          As you know Subscale models will not verify a full scale design because of Reynolds Number effects. The subscale model would need to be in a Variable Density Wind Tunnel (operated at increased pressure) because the transition from laminar to non laminar occurs at a different point of the wing which completely changes wing behaviour.

          No building Lampirydiae was a big missed opportunity for Europe to lead in stealth.

          • The US seems to have leaned massively on Germany after some US mil people had seen a mock up during a visit at ?Ottobrunn?.

            Stealth and Aero performance seem to have been well ahead of the “Have Blue” project back in the US.

            ( always amusing to have certain posters insinuate that inventions and research only happen under a NASA tag 🙂

          • I am rather certain the MBB people knew what they were doing.

            go down to “Lampirydiae”
            use google translate.

            seems to be the best write up available.

  2. A bit off-topic, I have a pet theory that a wood-burning boiler powering a steam engine (turbine or piston) would be more feasible than battery powered electric. I’d love to see some back of the envelope calculations for that!

    • I won’t bother trying to come up with any numbers since there’s no point… every stage of that pet theory – turning sunlight, CO2 and soil nutrients into wood over a couple of decades, harvesting and transporting the wood, extracting the wood’s energy into steam, converting the steam energy into electric – involve HUGE losses which are orders of magnitude less efficient than e.g. using wind/solar to get hydrogen from water. Plus wood burning is way more polluting (fun fact – most urban pollution in the UK is currently from new wood-burning stoves, not traffic!).

      Just think about the additional land use, time taken, transport, mass, volume (you want to store wood chips on a plane?!?). There are reasons we don’t ride steam trains any more and there are reasons steam trains (mostly) burnt coal instead of wood…

      • For most purposes you don’t care about the inefficiency of growing trees, since that is a carbon sink. It would just be SAF with fewer steps. (and, as you say, more pollution).

        The absurdity of a wood-burning steam powered aircraft is exactly the point: If you run the numbers and find that the steam powerplant and fuel weighs a lot less for a given endurance than battery-electric, it provides a reality check on battery-electric. A reality check that is sorely needed, because there seem to be many “lower carbon” projects that are being passed over in investment and attention for “zero carbon” battery electric projects that are totally impractical.

        • Take up of environmental concerns ( real ) happens to a large part as a religious movement ( gaga).

          Especially the christian framework connected religions are about untainted virginity.
          Thus the activity moves away from “sensible” improvements to absurd demands of “untainted fully carbonless” solutions ( that are either unattainable or counterproductive.

          Then most “anti” movements want their concern alive and not a really a closing solution.

        • @ Gareth: Ah right, I see. I was wondering how anyone could propose steam powered aircraft… 😀

          I think it’s pretty much agreed within the commercial aviation world that sustainably sourced hydrogen is the way we’re headed at the moment (maybe with some SAF low-hanging fruit along the way). The “problem” is all the startups hyping their quadcopter taxis etc. and the media totally buying into that (as it has more wow-factor). Obviously those supersized drones will never have the efficiency to carry large payloads over significant distances – can’t change physics – but let’s all just gloss over that, shall we…?!

          In the long term, with the rapid development of battery tech at the moment, a viable battery-electric system may emerge though. I wouldn’t rule it out.

      • -Biofuel has its place when the resource is a waste that wouldn’t otherwise be utilised. Biodigestors will turn manure, waste food crops, sewer, silage, damaged or sick crops into biomethane at nearly 60% efficiency. This can be converted to jet fuel at 60-65% efficiency.
        -However is silage (grass stalks from grain and maize crops) really a waste or is in an animal fodder? German farmers who started making biomethane from waste, manure etc are now earning money growing energy sugar beats, making it into biomethane and then running it in gas engines to sell subsidised ‘green electricity’. This doesn’t make sense at all.
        -So there are substantial agricultural, restaurant, domestic kitchen food wastes that should be used and we should be over producing food by 30% and letting it go to waste simply (used in bio digestors) so that there is excess for crop failure. (Putin’s war on Ukraine will remove 10% of globally traded grain crop).
        -If jet fuel is made it is looking more likely through the route “Lanza Tech” is using whereby different bacteria are used on wastes to produce an alcohol which is more efficiently (90% or more) converted to jet fuel.
        -All of the processes of fermentation and burning of wood produces CO2 which can be captured and utilised to make electro fuels.
        -Overall should make an important but minor contribution.

        Note read the nutrition information panel on your food in kj and you can calculate how much biomethane is produced. I’m probably wasting 25% of my food including peels, scraps, leftovers, spoiled foods so enough to make about 0.4L of jet fuel per week.

    • -At the moment Electric Aircraft would work viably if the flights were less than 100km-150km or so. Much over that distance the increasing weight requirements of the battery rapidly forces the aircraft design into a weight increase spiral to the point that the electric aircraft is 3 times heavier than a petrochemically fuelled aircraft and it doesn’t make sense anymore. Nevertheless batteries are improving all of the time. Commercial Lithium batteries with Silicon Lithium nanowire instead of Graphite electrodes are now available at 450W.Hr/kg for satellite use so that distance might double.
      -Of course there is a miniscule market for flights of 100km-150km from one airport to another but eVTOL looks like having a potentially huge market over these distances due to the relative ease of building vertiports.
      -Steam powered aircraft have been built. YouTube videos of a steam powered Travel Air 2000 are convincing. Steam power is quite viable for transport aircraft but in practice slightly inferior to gas turbines and piston engine aircraft.
      -As to coal or wood powered aircraft I think it would work: Pulverized coal works very well and was used in the final generation of locomotives. Notably Henschel Locomotives) When one see a flame thrower size blow torch in steam plant and the handling equipment one can see a wood steam aircraft could be built without having to throw logs or coke beads into a boiler. Alexander Lappish designed Ramjet powered (rocket accelerated) delta aircraft powered by pulverized coal and even one in which had a coal bed. Probably would have run but likely a poor weapon.
      -The German railways ran a constant volume gas turbine (two combustion chambers alternately pressurised by poppet valves and then ignited) in the 1930s that was tested on pulverised coal. (some erosion issues).
      -Today we have the Australian Ultraclean coal project which hopes to produce a refined purified coal that can be directly run in a conventional gas turbine to try and achieve the 60% efficiency combined cycle plants can running on oil.
      -Pulverised wood boilers and handling equipment exist so yes I think it can be done. I think it would be very limiting of course.
      -Wood, wilderness, forest are in short in the world. Even Scandinavia is reaching its limits. I would rather live in a world with a few wild Elke, Moose, Bears and Wolves than to destroy the last wilderness for such a tiny resource compared to global population.
      -If one had to the wood could be burned in a power station and the CO2 emissions captured and combined with renewable hydrogen to make SAF. Direct production of gasifier wood leads to contamination of the catalysts.

      • “As to coal or wood powered aircraft I think it would work”

        Huge difference between “it would work” and “more feasible than battery electric”…

        I also note that the single steam powered conversion is an anomaly – why would that be? – and was oil fired, certainly not wood.

      • Hi William,

        Thanks for that, very interesting. Wasn’t aware that a steam powered aircraft had already flown!

Leave a Reply

Your email address will not be published. Required fields are marked *