Batteries or Hydrogen?


Riversimple is a sustainable car company, not a hydrogen car company.  This is a summary of the thought processes that we have been through in choosing to build a hydrogen fuel cell car – guided by the need to address the environmental impact of personal transport.

The question “Batteries or Hydrogen?” is actually the wrong question.  We need a mixture of different solutions.  We don’t argue over whether solar PV or wind turbines will ‘win’ the renewable energy race; they’re different and we need them both.  The same is true of cars; the answer is neither hydrogen nor batteries exclusively.  We always hope for simple, single, ‘silver bullet’ solutions but, as Albert Einstein observed, “Everything should be made as simple as possible, but no simpler“.

Efficiency is the driver

We need a mix of fuels and powertrains in the future, chosen for different needs on the basis of carbon emissions and energy efficiency.  In particular, the characteristics of battery electric vehicles (BEVs) and hydrogen fuel cell vehicles (HFCVs) are so different that we see the current debate as a false dilemma; we need both, for different roles.

Our primary concerns are:

– Carbon emissions

– Total energy consumption

These are loosely correlated but they are not the same; an inefficient car running on renewable sources of energy is ‘low carbon’ but we are not going to be able to have a renewable energy economy, living off energy revenue, if we use it profligately.

This does not mean that we are ignoring the embodied carbon in building cars but cutting energy consumption in use is the biggest win; reducing carbon emissions from use requires us to reduce weight (and thus embodied carbon) anyway, and increasing vehicle lifespan amortises embodied carbon over a longer period so focusing on use has a positive impact on lowering embodied carbon.

Key observations on the drivers of efficiency:

– Vehicle efficiency is highly dependent on vehicle weight

– Powertrain efficiency is not dependent on vehicle weight – and is therefore not very well correlated to vehicle efficiency

– Weight is highly dependent on a) the choice of powertrain and b) designed vehicle range

These factors explain why the choice of powertrain should be different for different applications.  As is frequently pointed out, the powertrain efficiency of a BEV is higher than other powertrains, but our concern is vehicle efficiency.  An oft-quoted calculation by Bossel (Fig. 1) is used to support BEV efficiency but ignores these and other points.

Why Bossel’s calculation is not the whole picture

Three key changes are required to give a representative picture of Well to Wheel (WtW)[1] energy efficiency – one at the beginning, one in the middle and one at the end.

Fig 1. Useful transport energy derived from renewable energy from Ulf Bossel, Does a Hydrogen economy make sense? IEEE 2006

Fig 1. Useful transport energy derived from renewable energy from Ulf Bossel, Does a Hydrogen economy make sense? IEEE 2006

a) Electricity is not the source of the energy. Both hydrogen and electricity are energy carriers – you can’t dig either of them out of the ground.  There are losses upstream of the tank, Well to Tank (WtT), and for some fuel paths they are worse for electricity, some for hydrogen.

Over 85% of the world’s hydrogen is generated from methane, not electricity.  Steam Methane Reformation (SMR) of natural gas to create hydrogen is 75% efficient whereas generation of electricity from natural gas, considered our cleanest form of fossil-generated electricity, is only 49% efficient in the UK grid.

b) If hydrogen is generated by electrolysis, it is done at the point of distribution so there are no transport transfer losses.

c) At the end of this diagram, there are no distance units!  It does not say how far the car can travel on those final kilowatt hours.  To understand vehicle efficiency, the Bossel diagram needs to include the kWh required per km.  This will vary for different applications, particularly influenced by the range of the vehicle.

– A heavy car can have a very high powertrain efficiency but a low vehicle efficiency, because powertrain efficiency says nothing about how much power is required at the tyre contact patch.

–  As the range for which the vehicle is designed increases, a BEV rapidly gets heavier and therefore less efficient[2]; this is due not only to the batteries but a stronger chassis, larger electric motors, brakes etc – mass compounding.

–  A BEV is therefore very efficient for short range but not for long range applications.

Bossel's diagram adjusted to show vehicle efficiency

Figure 2: Bossel adjusted to show relative mileages achieved from same energy input

The Bigger Picture

Our simulation of vehicle efficiency suggests that we can build a more efficient hydrogen vehicle for any range beyond 150 miles, although we also concede that you can build inefficient hydrogen vehicles.  For a vehicle of the range to which we have been accustomed, 300 miles plus, there is no solution on the horizon that can be remotely as efficient as a hydrogen car.


Fig 3: Relationship between energy efficiency and design range for different vehicle energy sources

Other factors in a transition to renewables

a) Source of hydrogen

Whilst the endgame of a decarbonised grid and green hydrogen is clear, there is less clarity over the transition strategy.  Our choice of energy source will change as the transition to a sustainable energy system progresses and the priority is to reduce carbon emissions:

–          as much as we can

–          as quickly as we can

All carbon free energy should be used to displace the most carbon intense sources of energy first.  While we are totally committed to carbon reduction, and renewable ‘green’ hydrogen in the long term, we believe that there is a role for hydrogen from natural gas to refuel cars in the short term.  This is because renewable electricity will displace more carbon if used in the grid rather than displacing petrol.

If all hydrogen for cars must be green, the development of FCEVs and green hydrogen are coupled, something that we have not done with BEVs and green electricity.  By coupling the fuel source and use, it will hobble the development of both.  Fuelled by hydrogen from natural gas, a car like the Rasa at 40gCO2/km still represents a 60% reduction on the WtW emissions of the lowest emitting cars on the market today.

As we decarbonise our energy system, we can progress seamlessly to ‘green’ hydrogen without any further investment in the vehicle technology or distribution infrastructure.  It is therefore an investment in the long term rather than an interim solution and buys us badly needed flexibility:

– Over time, to transition incrementally from 100% grey hydrogen to 100% green hydrogen as we develop and increase renewable sources.

– Geographically, to collaborate globally on technology standards whilst all regions can develop whatever their local mix of renewables is to generate hydrogen – and those renewable sources are distributed much more evenly around the planet than oil.

b) Decarbonising the grid

It is easy to forget that electricity for charging vehicles is all additional demand on the grid; the grid does not provide our petrol and so, whilst we are well short of decarbonising the grid, it is not helpful to increase demand.

c) The 80/20 rule for short journeys

We often hear that 80% of journeys are less than 20 miles.  This is probably true and BEVs are good in this role.  However, this is a sleight of hand in that the corollary is that 80% of miles are driven in the other 20% of journeys, so this is where 80% of the problem lies.  This is where we need FCEVs.


BEVs have two key advantages; they can be very efficient for short range applications and can contribute to stabilising the grid by charging off-peak at night.  They have a key role to play in an integrated approach to transport and energy.

However, if used for long distance applications, not only does the BEV become inefficient but it also relies on fast charging in peak demand hours; a Tesla supercharger at a motorway services draws 120kW and is usually used in the daytime.  The misapplication of a critical technology – an inefficient car that destabilises the grid – not only weakens but reverses the benefits.

It’s time to put the batteries or hydrogen question to rest.  The challenge now is to use both clean technologies appropriately so that we can cut carbon emissions as quickly as possible.

[1] Well to Wheel (WtW) efficiency is the critical measure of energy consumption; it is broken down into two key stages, Well to Tank (WtT), covering extraction of energy, refinement and distribution to cars, and Tank to Wheel (TtW), covering fuel consumption in the car.

[2] This is true to a small degree for any vehicle – the longer the range, the more energy must be carried – but it is very acute for BEVs, as batteries are much heavier than petrol or hydrogen per unit of energy stored.