It is possible to power internal combustion engines that would normally consume petrol or diesel from hydrogen gas with minimal modification. Burning hydrogen to generate a mechanical force is not necessarily the best use of expensive-to-produce green hydrogen, but it may be justified as a transition measure which helps move consumer behaviour towards the routine use of hydrogen. Eventually, of course, all vehicles will use hydrogen with on-board fuel cells.
A typical 4-stroke engine cycle is shown in the animation above left. With the air intake and exhaust valves closed (left and right respectively), the trapped air is compressed by the upward movement of the piston. Just before maximum compression (top dead centre) the fuel injector adds diesel aerosol to the hot gas. It spontaneously combusts driving the piston in reverse. At the lowest point, the exhaust valve is opened and the upward movement of the piston (powered by other cylinders connect to the drive shaft, or crankshaft, and operating out of phase) forces out the exhaust gas. At the top of the cycle, the exhaust valve closes and, shortly afterwards, the air inlet valve opens. In the down stroke, air is drawn in (at external pressure if there is a turbocharger fitted) until the bottom of the stroke when all valves are closed. The cycle then begins once again. In a petrol engine, the only differences are that the injector is replaced by a spark plug and the fuel comes in pre-mixed with the intake air at the carburettor. A petrol engine can be powered entirely off hydrogen, whilst the diesel engine needs some diesel to start ignition (the ignition temperature for hydrogen is higher than the compression temperature). The focus in hydrogen ICE conversions is usually on heavy duty diesel vehicles because these are the engines in lorries, buses, ships and large electricity generators.
The conversion is very easy: hydrogen can be fed into the air intake at the correct time (a trivial modification) synchronised with the movement of the camshaft or under the control of the engine management unit. A more complex (but possibly more effective) modification is to fit a modified injector that will deliver both hydrogen and diesel.
There have been a number of trials and burning hydrogen results in reduced carbon dioxide emissions, but the economics are not favourable. One interesting but controversial idea is the use of HHO devices on board which are claimed to reduce fuel use. With this device, essentially a low cost electrolyser, energy from the engine is used to make a hydrogen-oxygen mix which is then fed back into the engine. In energy terms this should not work because more fuel is needed to produce hydrogen that would be the case if the electrolyser were not there. And the energy that is recovered when hydrogen is fed back into the engine is less because the electrolyser used will only be 65% efficient. In addition, a deionised or distilled water supply is required. Nevertheless, there are some ways this hybrid approach could be beneficial and claims need be evaluated fairly.
The first key claim is that hydrogen disperses in the cylinder rapidly and this along with the high flame speed ensures all the hydrocarbons are consumed. Whilst this address a problem with older engines, it is unlikely to be applicable to modern engines where the precise time control ensures almost all the fuel is burnt.
It is also suggested that burning hydrogen results in a higher combustion temperature hence the Carnot efficiency rises. We can check this, but bear in mind that higher temperature causes nitrogen to burn more readily with rising NOx emissions, and possible long-term damage to the engine unless the cooling system is boosted. This is in addition to the possible embrittlement effects.
We can do a quick and rough calculation. Assuming a stoichiometric mix and full combustion, 2 g of hydrogen releases about the same heat as 6 g of diesel (C12H23). Diesel will require 112 g of air to provide the necessary oxygen resulting in 7.6 g water vapour and 19.4 g carbon dioxide after combustion with the nitrogen content largely unaffected. Using the published specific heats for the gases, the final gas temperature is calculated be 2150 K. 2 g hydrogen combustion requires 86 g of air, resulting in the production of 18 g of water vapour with 70 g of nitrogen before and after combustion. The energy released will heat the gases to 2500 K. This is a significant increase, but it can only be achieved with proper control of the mixture – excess air will pull the temperature down; the increase is much reduced as the excess gas will take up much of the heat. Note that water vapour will tend to dissociate at this high temperature, and an excess of oxygen is really needed to ensure recombination as the gas cools in the expanding cylinder. One additional point is that injecting hydrogen and oxygen (HHO) to contribute 20% of the energy in a diesel-hydrogen mix will increase the temperature to 2450 K. But even this does not ensure reduced fuel use overall. To make the hydrogen that will produce 20% of the energy (0.4 g hydrogen in the calculation above) requires 5.5 g of extra fuel to be consumed – the hydrogen saves only 1.25 g and it is not possible to make up the rest by increased efficiency. 5.5 g diesel at 33% efficiency makes 72 kJ of electricity and using a 65% efficient electrolyser generates 0.4 g of hydrogen gas.
A third approach is to generate hydrogen when the engine is idling and the efficiency is very low, then use this electricity when there is high demand and the efficiency is naturally high. The efficiency of a typical engine can be represented by the following equation (efficiency is the fuel energy to brake power ratio - the power available at the driveshaft):
η = 0.33( 1 - exp(-4f))
where f is the power as a fraction of the maximum rated power of the engine. We can model the basic idea by looking at the New European Driving Cycle (NEDC) particularly the Extra-Urban portion (EUDC). When the vehicle is stopped, it may be assumed the engine is immediately switched off rather than ticking over. For a particular vehicle, the power is estimated from the acceleration and we can decide whether or generate or consume hydrogen. The cycle is 400 seconds – a longer cycle is not feasible as it is unsafe to store large quantities of HHO for long periods. This is modelled above.
The speed is displayed along with the probable efficiency. Note that when decelerating it is assumed the power output is zero, but power is required to maintain the vehicle at a constant velocity because of friction effects.
You can do the detailed calculation yourself with the figures provided but the increase in efficiency when the electrolyser is added as a load is not sufficient to make the hydrogen production and consumption strategy outlined above viable. The conclusion is that combining hydrogen with diesel can be beneficial, but there are significant challenges and any gains are likely to be marginal.
Investigations