Bjorn’s Corner: Sustainable Air Transport. Part 15. Hydrogen propulsion system choices.

By Bjorn Fehrm

April 15, 2022, ©. Leeham News: Last week, we examined different airliner types’ power requirements and the importance of their size classes in the market.

Now we look at what propulsion system alternatives are available when using hydrogen as the energy source and their principal advantages and disadvantages.


Figure 1. CO2 emission by airliner segments. Source: EU Hydrogen-powered aviation report.

What power levels do we need for Green airliners?

We listed the typical power levels for different airliner classes last week, Figure 2.

Figure 2. Shaft power to the propeller/fan for different airliners. Source: Leeham Co.

We also looked at the expected new deliveries per market segment based on a market study from the JADC.

Figure 1 shows the contribution of CO2 emissions by market segment. We see the main emissions (43%) come from the 166 to 250 seat segment, with long-range over 250 seat second (30%) and short-range (81-165 seats) third at 24%. The segments below 80 seats represent less than 4% of CO2 emissions.

Consequently, our primary need for a green propulsion system is in the 10 to 20 MW range on an aircraft level. Propulsion systems below 5 MW on the aircraft level are not helping us reach 2050 emission goals.

Propulsion alternatives for hydrogen-fueled airliners

We have seen in earlier articles that battery-based or hybrid-electric aircraft are not viable solutions to help with Greenhouse gas emissions from air transport.

Sustainable Aviation Fuel, SAF, is the easiest and most straightforward solution. Still, as its volume production path (the biomass path is too limited) is additional processing of green hydrogen, its cost will be higher than hydrogen fuel.

A hydrogen propulsion system can burn H2 in a modified gas turbine or generate electric energy in a fuel cell that drives an electric motor to drive the fan/propeller.

We will deep dive into both alternatives in subsequent Corners. Let’s examine on a high level how these alternatives can meet our powerplant objectives for our airliner segments.

Hydrogen burning gas turbine

As I’ve outlined before, the world’s first jet engine, von Ohaine’s S-1, started on hydrogen in 1937 as he had difficulty getting it to run on kerosene. Hydrogen is in many respects a better fuel for gas turbines than our kerosene-based jet fuel:

  • Hydrogen allows a leaner and cleaner burn, resulting in no CO2 emissions, no soot, and five times lower NOx emissions.
  • Liquid hydrogen (LH2) offers a cool sink that can be used to cool the gas turbine and increase its efficiency.

Hydrogen as a fuel for gas turbines has been used in ground-based gas turbines for decades and has been trialed several times in airborne test aircraft.

The advantages of a gas turbine as a hydrogen propulsion system can be summarized as:

  • It eliminates CO2 emissions and reduces NOx emissions five times. It also eliminates CO, SOx, UHC, and Soot emissions.
  • It’s proven technology. It’s been used in ground installations for many years and has flown in trail aircraft since the 1950s.
  • It combines the gas turbine’s high volumetric/mass efficiency with low energy consumption. As we shall see, new technology can drive the core efficiency when using LH2 over 60% (today’s turbofan cores max out at 55%).
  • We reach the same power levels as today’s turboprop/turbofan engines with the same installation envelopes.

But the LH2 gas turbine alternative also has disadvantages:

  • It emits water vapor that forms contrails. Because no soot is emitted, the ice crystals formed will be larger and fewer. It’s not clear how big a problem this will be. Experts say flight trials measuring the impact is necessary. It explains the importance of Airbus’ 2026 flight trails with an A380. With an H2-burning engine placed on the roof of the A380, the engine’s contrails can be measured by a trailing aircraft.
  • There is still a reduced NOx emission.
  • The engines are as complex as today’s turboprops/turbofans to design and produce. Thus only engine OEMs can develop these propulsion systems.
Hydrogen fuel cell system

A fuel cell system combines H2 with air in a fuel cell and forms heat, electricity, water, and oxygen-depleted air as outputs. The electric output then drives an electric motor to turn the propeller/fan.

Fuel cells for ground use are established and mature technology. For airborne use at the power levels we need, they pose thermal and electric management problems combined with high mass and volumes. Recent trials with small fuel cell propelled aircraft (H2FLY, MAHEPA) confirm their advantages and problem areas.

The fuel cell alternative’s advantages can be summarized as:

  • The output is free from Greenhouse gases.
  • The fuel cell’s water output is water, not vapor. Thus it doesn’t create contrails.
  • It’s easy to adapt fuel cells and electric motors to different power levels. It’s, therefore, the preferred option for upstarts targeting smaller, first-step aircraft.

But there are also disadvantages:

  • A fuel cell produces more heat than electric power. The management of heat is a major system complication. Studies and tests show that heat management is the heaviest part of a fuel cell system for aircraft.
  • The fuel cell alternative has all the complications of high electric power handling in the thin air where an airliner operates.
  • The mass and volumes of a fuel cell alternative are problematic. As we go up in power to 10 or 20 MW systems, we must use advanced technology such as superconduction and new packaging techniques to create acceptable installations.

We will go through these alternatives in more depth in subsequent Corners.

22 Comments on “Bjorn’s Corner: Sustainable Air Transport. Part 15. Hydrogen propulsion system choices.

  1. LH2 burning engines are manageble technical challanges designing burnes, fuel controls, engine driven fuel pump, fuel oil cooler, fuel driven actuators like VSV system. The rest of the LH2 system that is aircraft mounted is not as easy with tanks, pumps, pipes, valves, contianment requirements and emercency equipment like safety valves, dump valves and the certifications requirements for all these. Using LOX/LH2 rocket components gets too expensive so lots of work for suppliers like Liebherr, Hamilton Sundstrand, Argo Tech..

    • Hi Claes,

      I agree. The Airbus CRYOPLANE study thoroughly reviewed what it entailed to convert a gas turbine to an H2 run. It was a straightforward task. The challenge is in the rest of the system as you write. I see the burn alternative as the one we know works, its only issue is the contrails, and we don’t know how big a problem it is until we fly and measure.

      • Can you expand on what the problem with contrails is? Do they increase global warming, or is it an aesthetic problem, or something else?

  2. I would not rule out the use of hydrogen fuelled internal combustion engines on aircraft; both (SI) Spark Ignition and (CI) Compression Ignition(Diesel). Toyota, Liebherr and Bosh all have prototypes. Toyota and Subaru have cars. Slowey but surely it seems the problems have been solved despite YouTube experts.
    -One technique for SI engines is to use a prechamber with a rich mixture of hydrogen that is ignited by spark which produces a jet igniting an ultra lean mixture in the main chamber that will ignited reliably and be very efficient at partial loads while not producing NOX. Valves are phased to keep the exhaust valve cool and prevent preignition and of course high pressure allows direct injection which intrinsically prevents preignition.
    -For CI Diesel engines it seems simpler. The engines can be 100% Hydrogen or duel fuel with a almost infinite variation between liquid fuel and hydrogen.
    -I do not see great long term success for conversion of existing types from gas turbine to fuel cell. I would expect that the waste heat of the fuel cells would simply be used to heat the leading edges of wing surfaces to deice them. It may turn out to be a positive given that we have new ways of fabricating wings structures and fine channels.

    • With the flame speed of hydrogen LH2 piston engines will behave like HCCI engines (Homogeious charge compression ignition) but with igniters to control at different power levels the benefit of igniting the whole charge at close to top dead center is combusiton at max pressure, then just after t.d.c. the burn is over and the pressure just pushes the piston down, like on F1 engines. Moden diesels don’t want this pressure spike “single knock” but injects diesel in 5-7 burst to control the pressure and temperature profile on the ignition stroke for optimum emissions/temperatures/pressure loads/sound/life and power/fuel burn. For jet engines it is benefical with a fast burn, limiting the axial size of burner (engine length) and reducing time for air in extreme heat to form NOX, hence the LH2 engine can be made shorter and lighter and more easy pass altitude reignintion tests. For some reason both PWA and I suspect RR want to have the LH2 get into H2 phase before injection into the engine.

      • Unfortunately SI ICE seem to be limited to about 1MW-2MW with 12 cylinders for gasoline, more for diesel and it doesn’t get us into the 20mw-50mw we need for the 150 passenger seat market. Electric turbo chargers and turbo compounding should improve things. I see them as potentially superior than fuel cells as they may exceed 50% thermal efficiency.

  3. As hinted this is as much an economic issue as a technical challenge. Many things can be made to work, the kicker is making it viable financially. We see that the electrification of land transport (BEV, electrified rail) requires investment, but ultimately it delivers a lower TCO due to reduced energy and maintenance costs. Therefore the key challenge of decarbonising aviation will be lowering the cost of fuel. Hence Roll’s Royce’s recent investments in Nuclear.

    • I’m not sure how you mean ‘key challenge of decarbonising aviation […] hence investments in nuclear’. Nuclear can provide continuous energy over several years, but it is not particularly light weight or neatly packaged. Pentagon is building a microreactor that produces up to 5 MW, but weights 40 tonnes and is packaged in three 20-foot shipping containers (quite voluminous).

      • OK maybe I didn’t explain that too well!

        Ground based Nulcear can be very efficient for the production of Hydrogen and SAFs, because the surplus heat can be used to increase the efficiency of hydrolysis. RR are in a fortunate position because they already have a long established (military) nuclear reactor business, but they have explicitly stated that they are most interested in developing civil SMR solutions due to their potential for lowering production cost of Hydrogen, SAF/Synfuels, the goal being reaching parity of production costs with fossil fuels. Which (indirectly) supports the future of RR aero/turbine business interests. Hope that’s clearer!

        • The thing about Rolls Royce’s SMR “Small Modular Reactor” is that at 440MW it isn’t actually small or what is regarded as an SMR. It’s a normal 1980 sized reactor that is however half the size of the 1.3GW units that came in in the 1990s such as AP1000. NuScale’s SMR is a real SMR at 60MW. It is completely convection cooled and needs no pumps.
          However the Rolls Royce SMR borrows from SMR ideas.
          The key idea of the SMR is to build and operate it like aircraft are: A certified design, a certified test campaign, a type certificate, certified workers and production facilities. To do this the SMR should be small enough to be shipped by rail complete.
          The way to NOT build reactors is to modify the design every time for the bespoke requests of the customer and to use different contractors to fabricate and erect the reactor on site. It should all be done in the factory.
          The other reason to make SMR small is apart from transport of the new reactor is the possibility of transport of the shut down reactor when decommissioned. Small reactors can escape the need for forced cooling and this is a key safety feature that controls costs.

        • Incidentally Nuscale’s reactors won’t usually be operated individually. They’ll be in batches of 4 to 8.
          Hydrogen for airports can be easily transported to site by pipeline. It can also be made by electrolysis on site. In some cases from nearby renewables. Aircraft such as the B737 have done vertical dives into the ground and left 9n craters. I can guarantee you that the nuclear regulator is going to look at the case of a fully loaded and fuelled B777-9 or XF doing that straight into the reactor so I think the idea of reactors at airports is a non starter. Now we also have to worry about a hypersonic “Russian” Kinzahl missile (a 1980s Islander IRBM air launched). The beauty of electro SAF is they can be made remotely anywhere and shipped.

      • I think Sonic means that RR and other small modular nuclear reactors that produce electricity and heat are useful to produce LH2 and they can be placed at airports and close to cities that need the power. The certification requirements for different types of these I guess is in the works. After the first few are certified and have run thru the first fuel change and heavy check we will know its true costs per MWhr and if they can beat the 14-15 MW sea based windmills now getting into serial production by GE, Vestas, Siemens-Gemesa and others. If the high tech windmills are the winners over nuclear one ould expect RR and Pratt&Whitney to get involved and maybe buy the others.

        • Yes, exactly, thanks for clarification. The point about placing reactors close to airports is also a good one because as Bjorn has discussed before, Hydrogen is not so easy to distribute as liquid fuels. Ammonia is proposed as a more easily transportable fuel to replace heavy oil in surface shipping, so it would be interesting to know if there are any possibilities to use this fuel in aviation? But intermittent renewables will struggle to compete I think, because the much lower capacity factor (vs Nuclear) increases greatly the amortization period of the expensive generation and electroliser assets.

          • Yes, still many airports are located close to water and can get LH2 delivered by ship or piped a short distance like Newark or O’Hare. But having a small modular reactor at the airport making LH2 or SAF increases the reliability of the supply system. Having the reactor produce at full capacity no matter the airliners needs at the airport allows the reactor to backfeed the natural gas grid increasing it H2 content a bit.

  4. Ohain wasnt the first to run a bench test jet engine.

    Whittle bench tested his design , which he had a large contribution in making as he had a great deal of practical training and experience, in April 12 1937.

    Ohain , who filed his patent after he and the Berlin patent office had seen Whittles patent, being a more theoretical person had the manufacture done by Ernst Heinkel and his aero engine machine shop. The He-S1 was running under its own power in Sept 1937. So was second

    Bjorn could you correct that claim that Ohains engine ran first on its own power

    • von Ohain and Whittle met. They compared notes and diaries. Ohains engine ran under its own power about 2 weeks before Whittles run. von Ohain gave Whittle credit for inventing the jet engine because he said Whittle had solved the problems of combustion of fluids first. The men became friends. I think Whittles response was “so you did…..” von Ohains engine ran on hydrogen first and the combustion problems are much less due to wider flammability limits and better mixing. Gas turbines ran and existed before either man. Sanford Moss of GE who invented the turbo charger was face palming himself for not having invented the Jet. His writing on the issue show he was somewhat upset with himself.

      • They never met before the war. Whittle wasn’t aware of Ohains work as it was classed secret in pre war period and of course the first flights – which Ohains engine was ahead- were also military secrets.

        Whittle should be given credit for his priority in patent and sustained test runs.

        • It’s fairly pointless assign credit to someone beating someone else. I give credit to Whittle for his fir tree roots, his excellent combustion chambers and mixing nozzles and organisational skills and foresight to get quality alloys developed. They were clearly working independently and I think von Ohain would have succeeded without Whittle in the same time frame. A short time thereafter the other German makers (Junker Airframe Division and BMW Airframe division). The Hungarians had a turboprop before 1939 and Vickers in UK and Lockheed L1000 in USA. It’s as pointless as the Leibniz–Newton calculus controversy. von Ohain must have been 14 years old when Whittle filed his first patent.

          • Ohains engine patent was in 1936.
            He actually filed a patent for a film sound process which came out of his PhD thesis he was awarded in 1933. The rights were sold for a considerable sum and it was largely still used up till the digital era.
            So he was very familiar with the patent process when he went on to work on gas turbine design for his ‘second PhD’ ( these days called post doc) and his second patent, for a jet engine design was 1936.

            Whittles actual jet engine patent came in 1932 ( when Ohain was a PhD student, the claim he was 14 is a diversion)

            if some one is going to say the successful bench test run of the Ohain engine was a GT first , its wrong and should be corrected

            Theres other work in aerodynamics by Lanchester on the theory of flight that was a first. The german theoretical experts tried to pretend they ‘didnt know about it’ inspite of Lanchester having a seminar at their own university at the time.

    • In regards to the patent examination claim. After von Ohain and Max Hahn started working with Heinkel the organisation did a patent search. They did find Whittles patent and it likely did effect von Ohains design. They also found other patents for gas turbines. von Ohain had actually chosen a centrifugal compressor with a single sided impellor. It’s not correct to say that the Germans started with axial and the British with centrifugal. Both parties pursued both types. Hans von Ohain chose a radial inflow turbine. In other words the turbine looked like a centrifugal compressor however with the flow inwards. He did this because a centrifugal compressor and a centrifugal/radial inflow turbine would match each other automatically and eliminate the need to test the turbine and compressors separately to adjust and match them. von Ohain used a annular combustion chamber . The effect of Heinkel seeing Whittle’ patent was that Heinkel changed the annular combustion chamber to a double reverse flow type annular chamber. The resulting engines was exceedingly short due the combustion chambers and the in flow turbine. Whittle had used a can-annular (not annular) combustion chamber with reverse flow whereas Heinkel was initially straight trough annular and then switch to double reverse flow annular which ran in late March 1937 on hydrogen. The reason for the use of double reverse flow by Whittle was to reduce the shaft length and allow machining of the turbine disk and shaft in one piece to prevent alignment issues. Engines intended for an actual production aircraft however abandoned the double reverse flow and used annular chambers. They looked quite different from the Welland.

      • The Welland , the Rolls Royce production engine for the Gloster Meteor still had many similarities to the Power Jets engines series bench tested by Whittle and then used in the first flights by Gloster experimental type

        Even Encyclopaedia Britannica credits the first running bench test
        jet to Whittle and thus ‘the invention of the jet engine’
        ‘He tested his first jet engine on the ground in 1937. This event is customarily regarded as the invention of the jet engine..’

  5. Hi Bjorn. My understanding of Fuel Cell challenges from driving an FCEV is that the primary issue is the temperature operating range, it needs to run at fairly low temperatures. Even we FCEV drivers in California have occasional issues with cooling. Not because it’s so much heat being developed, but because you need to get rid of so much of it to keep the Fuel Cell within operating range. Developing Fuel Cells that can run hotter and still keep their efficiency will be key, particularly for aviation.

Leave a Reply

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