Hydrogen: Not the Vehicle Fuel of the Future

Merrick, 23rd January 2008ce

Hydrogen: Not the Vehicle Fuel of the Future
Hydrogen; the smallest, lightest and first element on the periodic table, and most abundant element in the universe. As we bask in the radiance of the vast hydrogen reactor at the centre of our solar system, hydrogen has a poetic rightness as an alternative vehicle fuel. It burns cleanly and the only exhaust gas is water vapour. All our climate and energy security problems solved at a stroke.

Sadly, as is often the case with the nice-feeling alternative technologies, it’s not that easy. It would be colossally expensive to introduce, doing so would commit us to long-term fossil fuel consumption and, most importantly, we can’t make it without significant climate impact worse than if we just carried on using petrol.


There aren’t any naturally occurring deposits of hydrogen for us to tap into like coal or oil, nor does it flow freely and abundantly around us waiting to be used like wind and sunlight. Hydrogen only comes bonded to other molecules; it takes energy to separate it. Like a battery, it needs a primary energy source to make it from – gas, coal, or something made into electricity – so it is only an energy carrier, rather than an energy source in the true sense.

Hydrogen is not a new product at all. We have over a century’s experience of industrial production (it’s used in the production of nitrate fertilisers and oil refining among other things). We’ve done much of what could be done to economise. It is still very expensive.

Manufacturing it from natural gas is the cheapest and most experienced method. But gas is rapidly being depleted. We are on course to hit ‘peak gas’ before mid-century, after which demand will outstrip supply and the price will go through the roof. We will hit it sooner if we use more of it, such as a switch away from coal to gas for electricity, or a switch to using it for making large amounts of hydrogen.

But there’s another consideration, the whole point of us even thinking about it in the first place; the climate angle. Manufacturing hydrogen means separating it from the CO2 in natural gas. This ‘clean’ fuel only gives off water vapour from a car exhaust, but that’s because the carbon’s already been emitted at the factory. This climate friendly renewable fuel is actually a carbon-emitting fossil fuel. The climate doesn’t care where you emit CO2, only that you do it at all.


The science is clear. In order to prevent runaway climate change, carbon emissions need to be stabilising within ten years, and we need at least a 60% global cut within 30 years (which means the over-emitting nations – ie the major car-driving ones – cutting by at least 90%). So if a technology can’t be developed and deployed in the next decade or so, it’s of no use to us as a response to climate change.

This means that the people currently touting nuclear power as our primary solution are wrong, as it cannot be on-stream quickly enough. It also means that the roll-out of hydrogen as a vehicle fuel – even if it were magically carbon-neutral to manufacture – cannot be any of use to us either; the safety and engineering issues would take too long to surmount.

Dr Joseph Romm served in the US Department of Energy during the Clinton administration when the ‘hydrogen economy’ became big news. Running the Office of Energy Efficiency and Renewable Energy from 1993-98, he oversaw significant increases in funding for hydrogen fuel R&D. Yet, although he’s a believer in the possibility of clean hydrogen, he’s firm in his belief that in 2030 we’ll have less than 5% of vehicles powered by it1. There’s not a credible voice that disagrees.

Since it cannot be part of a 90% emissions cut in thirty years we should, at least for those three decades, turn our attention elsewhere. But there are also other reasons – practical, engineering and economic – why hydrogen can never work.


There is a way of making it without using fossil fuels as the raw material; electrolysis of water. Put simply, an electrical charge breaks the bond in H2O, separating it into hydrogen and oxygen.

Because the raw material is water rather than gas or coal, this is often touted as ‘carbon-free’. Except that the electricity is coming from the national grid, which is not carbon-free, it’s mostly fossil-generated. This isn’t preventing emissions therefore, it’s merely displacing them, just like the ‘clean’ car exhaust that emitted all its CO2 at the factory.

As electrolysis uses so much electricity, once again the emissions are greater than if we were using a petrol vehicle. Powering BMW’s new hydrogen car with electrolysis hydrogen made from the UK grid would create around four times the emissions of its petrol equivalent2.

The only genuinely carbon-free hydrogen would be using renewable electricity to power electrolysis of water. But if we were to do this, we increase the overall demand for electricity. What we give to hydrogen from renewables makes a shortfall in the grid that will be taken up by extra fossil generation. Again, it just displaces emissions.

The only time it becomes genuinely carbon-free is when the whole grid is powered by renewables and we have spare capacity to start powering our vehicles. Even then, hydrogen isn’t the best option. Rather than losing half the energy of electricity making hydrogen and liquefying it, why not just use that electricity directly?

In his book Heat, George Monbiot advocates a proposal by Dave Andrews for using electric cars3. The problems with electric cars are their comparatively short range, and the long time they take to recharge. If we have a lot of renewable generators, we’ll have a lot of unused electricity - the wind and waves keep going through the night when our electricity demand is low. So, we use that off-peak power to charge batteries. When your battery runs low, you pull into a filling station and the battery is removed and swapped for a charged one. It would take the same time as refilling with petrol. More, it would do away with tankers entirely – the vehicles themselves are the delivery fleet.

Some hydrogen enthusiasts have suggested similar ideas for using off-peak renewable electricity to make hydrogen from electrolysis of water, but this ignores the huge inefficiency.

If you make hydrogen from renewable electricity then, because you save the emissions that would have come from a petrol car, you save about 225kg of CO2 emissions per megawatt-hour of electricity used. But if you use that electricity directly as electricity, you save 367kg CO2 if you're replacing gas-fired power, and 890kg if you're replacing coal. So for maximum carbon savings, we need to use our renewable electricity as electricity, not make it hydrogen.

To replace our vehicle fuels with electrolysis hydrogen would take more than our present electricity consumption4.

Do we think we can double electricity generation whilst doing away with fossil burning? Or is electrolysis hydrogen as a vehicle fuel a non-starter?

When the Bush administration used the 2003 State of The Union address to announce the kickstart of the hydrogen economy for vehicles, they neglected to mention any of the emissions that come with making it. Under their National Hydrogen Energy Roadmap plan, fossils would be the source of the vast majority of hydrogen, but 10% would come from electrolysis of water, powered by dedicated nuclear power plants5. As if there weren’t enough safety issues with manufacturing hydrogen already.


The most prevalent idea for using hydrogen in a vehicle is the fuel cell. Fuel cells are essentially a battery that can be continually charged up. As a technology they’re well established, actually pre-dating the internal combustion engine. Hydrogen is fed through, producing an electric charge and also heat. Very large cells used to power buildings can also use the heat (making them quite efficient and more plausible as a future technology), but in a vehicle this heat – much of the energy we’re getting from the hydrogen - is simply wasted.

Then there’s the issue of having a hydrogen tank in the vehicle. At room temperature and pressure, hydrogen has one three-thousandth of the energy of petrol. Assuming you’re not going to have a fuel tank a couple of hundred times the size of your car, your hydrogen needs to be either compressed or liquefied.

To be liquefied, hydrogen needs to be cooled to -253 degrees centigrade. The energy used to do this is equivalent to 30- 40% of the energy the hydrogen contains6.

Let’s look at that from a climate perspective. It takes 12.5-15 kilowatt-hours of electricity to liquefy 1kg of hydrogen7. With the UK’s emissions from generating electricity, that’s 6kg-7.2kg of CO2 emissions8. Burning a gallon of petrol - which contains around the same amount of energy as 1kg of hydrogen - releases around 8.8kg of CO29.

In other words, hydrogen causes 68-82% of the emissions of burning petrol just for the liquefaction process! This is before we count the emissions of the raw material involved in production (if it came from natural gas, that's another 9kg of CO210). This fuel is worse than petrol.

To avoid the practical problems and monstrous energy consumption of cooling to -253 degrees and keeping it there, hydrogen can instead be left at room temperature as a gas but compressed. Compression takes less energy than liquefaction, but then compressed hydrogen contains less energy than liquefied hydrogen. The energy to compress it to 5,000lbs per square inch is only 4-8 percent of the energy it contains11.

However, this isn’t much use for a car; by volume, it contains one-tenth of the energy of petrol. A fuel tank ten times the size of current ones is out of the question, but then again having such a short driving range that you need to refuel ten times as often is utterly impractical. You’d also need many more tankers, pipelines and the rest of the distribution kit.

There is also a solid-state form of storage, ‘metal hydrides’, materials impregnated with hydrogen that release it when reacted with water. It’s a no-hoper for vehicles. The reactions involve very high temperatures, the hydrides are heavy, are highly prone to leaks and we have yet to develop a practical way to remove and recycle the spent hydrides from a vehicle.

As the American National Academy of Engineering concluded, ‘no hydrogen storage system has yet been developed that is simultaneously lightweight, compact, inexpensive, and safe’12. As those are the four key factors for a vehicle fuel, it’s a practical non-starter even before we consider the emissions issue.

As a liquid, it needs to be maintained at -253 degrees centigrade constantly until the point of use. This can be equivalent to another 10-15% of the embodied energy. By now, not only are the emissions the same or worse than burning petrol, but we’ve lost half the energy we put into making the stuff. It’s like a battery making machine that uses two batteries for every one battery it manufactures.

There are certainly good cases for having a few very inefficient devices – it’s far handier to power your camera off a battery than a small diesel motor. But using grossly inefficient technology for such a huge energy consumer as our vehicle fleet is an extravagant waste of resources that we can’t afford.

As Alec Brooks put it, ‘fuel cell vehicles are energy pigs. Fuel cell vehicles that operate on hydrogen made with electrolysis consume four times as much electricity per mile as similarly-sized battery electric vehicles’13.


So much for hydrogen fuel cells. But in the no-idea-too-profligate world of hydrogen vehicle fuels, there is an even worse concept. BMW have launched their hydrogen vehicle, the H7. As mentioned earlier, there can be no doubt that it is several times worse for CO2 emissions than the worst 4x414. It uses hydrogen not to power a fuel cell, but burns it directly in an internal combustion engine. Being even more inefficient than a fuel cell, it means it has a smaller driving range, a mere 125 miles on 8 kilos of hydrogen. The fuel is kept in liquid form in an insulated tank.

The thing with internal combustion engines is that the fuel is, well, combustible. You only want it to burn in a controlled way, rather than have it explode at an inopportune moment. With petrol cars, there’s a problem if the fuel and its gases get too hot; leave it parked somewhere on a blazing summer’s day and it could explode. Fortunately, there’s a little safety valve built in to release it safely. Hydrogen combustion cars regard anywhere above -253 degrees as a blazing hot sunny day.

The H7’s tank is not cooled – how would you find an on-board energy source for such refrigeration? – but is wrapped in layers of fibreglass and aluminium. The insulation cannot prevent the fuel warming, only slow it down. Which means that it reverts to gas and pressure in the tank increases. This isn’t a safety issue as, like a petrol car, there’s a valve that allows it to escape. It is, however, an economic issue. In case it wasn’t already expensive and wasteful enough, after about a day your H7 is preprogrammed to start jettisoning your fuel. A full tank empties itself completely in 10-12 days.

Joseph Romm, despite being something of a hydrogen advocate, is incredulous at the BMW combustion idea, commenting, ‘BMW has managed to develop the least efficient conceivable vehicle that you could invent’15.

This horrendous inefficiency means that it is more expensive for the owner, worse for the climate and cannot be taken seriously. As David Talbot put it in Technology Review, ‘a car like the Hydrogen 7 would probably produce far more carbon dioxide emissions than gasoline-powered cars available today. And changing this calculation would take multiple breakthroughs - which study after study has predicted will take decades, if they arrive at all. In fact, the Hydrogen 7 and its hydrogen-fuel-cell cousins are, in many ways, simply flashy distractions produced by automakers who should be taking stronger immediate action to reduce the greenhouse-gas emissions of their cars’16.


Even with the American gusto for the hydrogen economy, how likely is it to happen, how quickly and how much would it cost?

With current technology, the infrastructure to supply just 40% of the light-duty vehicles in the USA alone has been estimated at over 500 billion dollars17.

Don Huberts, CEO of Shell Hydrogen, confirms this estimate. ‘The initial investment has been estimated by Shell at around USD 20bn for the U.S. alone, to supply 2% of the cars with hydrogen by 2020 and to make hydrogen available at 25% of the existing gasoline retail stations. In the subsequent decades, further build-up of the hydrogen infrastructure will require hundreds of billions of US dollars’18.

Even these mindboggling figures are calculated using the cheap methods like gas which result in CO2 emissions comparable to or worse than burning petrol and do not address the secondary issue of ‘energy independence’, reducing reliance on imported sources of energy.

When we’re throwing half a trillion dollars (and that’s for the USA alone) into climate change mitigation and emissions reduction, we need to ensure the best bang for our bucks. We could do an awful lot more with an awful lot less money, and see results an awful lot sooner. Hydrogen is not something we can have before we’ve done serious emissions cuts, and certainly not as part of them.

Even if it were a clearly good idea, it’s hard to see how it would happen. It’s a chicken and egg problem. Who would shell out hundreds of billions of dollars to supply a fuel nobody uses yet? Yet who would buy a car that you can’t readily refuel? Nobody will be a customer until there’s the infrastructure. But nobody builds pipelines without customers.

It’s not like the growth of car use, comparatively slow and piecemeal. To have any part in climate impact mitigation (let’s just pretend that it even could), this needs to be huge and rapid. Yet with numerous other alternatives looking just as viable and only one or two ever going to be the winner, who’s going to throw hundreds of billions of dollars at what will may well turn out to be the automobile industry’s answer to Betamax?19

Then there’s the resistance from the old guard to deal with, inhibiting any quick deployment and uptake on this or any other climate response industry. Both the construction of the infrastructure and the take-up by the public will be stymied by those old industries who stand to lose. As if the fossil barons are going to take this lying down.

Just as they bought the politicians to scupper the Kyoto treaty and then did it again last December in Bali, so they’ll move their muscle to stop any transition away from their present dominance, and even resist any incursion to clean up their own industries as such action reduces profits which, if their behaviour is anything to go by, are much more important to them than human survival.

BP, feted as part of the vanguard of the hydrogen industry, are clear that using renewables for electrolysis of water is not part of their plan. ‘We view hydrogen as a way to really grow our natural-gas business,’ said Lauren Segal, BP’s general manager of hydrogen development20.


At present, hybrid petrol/electric cars are doing well commercially because they are more economical. It’s easy to sell someone the environmentally friendly option when it’s going to be cheaper for them. No driver, irrespective of their environmental feelings, will buy a car if they can’t afford it.

In December 2002 Yozo Kami, Honda’s engineer in charge of hydrogen fuel cells, said it would take at least ten years to get the price of a hydrogen car down to $100,000 (around £50,000). This from the people making one of the cheapest prototypes21.

Fuel cells of the type used in cars (proton exchange membrane cells) have a short lifespan too. The industry is aiming at around 4,000 hours of use, which might equate to ten years of driving. As it stands, a good prototype can only manage about 1,000 hours22. Buying a car that costs £50,000 and will be useless in three years isn’t going to appeal to anyone.

And even once the industry were scaled up, as a vehicle fuel hydrogen is still likely to be at least twice as expensive as petrol23.

This is compounded by the poor efficiency of the hydrogen vehicles. Alec Brooks again; ‘the commonly held belief is that fuel cell vehicles will have two to three times the fuel economy of gasoline powered vehicles. But so far, fuel cell vehicles are losing. The mid-sized petrol powered Toyota Prius has 13 percent better fuel economy than the subcompact Honda FCX fuel cell vehicle’24.

Alec Brooks is a big electric vehicle advocate, but his view is shared not only by the research scientists quoted above, but by those with a vested interest in hydrogen. Shell Hydrogen’s CEO Don Huberts bluntly conceded, ‘at the end of the day, hydrogen and other alternative fuels will be three to four times as expensive as oil based products, and if no one wants to pay for that, we can't make those fuels’25.


Being a small molecule and very light, hydrogen is particularly leak prone. It is also odourless. Natural gas can have varying or no odour, so an odorising agent is added to it. That smell we think of as gas isn’t gas at all, it’s a chemical blend made in a factory in West Bromwich and added to the supply. It must be weird living round there, how would you tell a gas leak from the smell of the factory?

But anyway, an odoriser cannot be added to hydrogen as not only might it damage the fuel system technology (especially in sensitive fuel cells) but it wouldn’t actually work - the hydrogen would be substantially lighter and separate from its smell.

This gets worse. Not only is it leaky, invisible and odourless, but it burns invisibly too. The first you’d know about a raging fire would be when you stepped into it and went up in flames.

NASA use a lot of hydrogen (it’s in those fat tanks on the side of the space shuttle). Being NASA, all high-tech and with as much experience of handling hydrogen as anyone, they developed a special device for detecting burning leaks issued by their Office of Safety and Mission Assurance. Walk round pushing a broom in front of you and see if the bristles catch fire26.

It is a very dangerous substance to be handling in large quantity, in populated areas, at thousands of forecourts with untrained members of the public.

Some safety factors work in hydrogen’s favour – it is essentially non-toxic and dissipates rapidly in the air, making its way swiftly upwards and bonding with oxygen. If only pools of petrol did the same instead of lingering round waiting for a fag end.

However, you’re in real trouble if the hydrogen leak is already burning. And that’s easily done. It is flammable over a wide range of concentrations and has ignition energy twenty times smaller than natural gas or petrol. ‘Operation of electronic devices (cell phones) can cause ignition’, and ‘common static (sliding over a car seat) is about ten times what is needed to ignite hydrogen’27. Electrical storms several miles away can generate enough static to ignite hydrogen28.

But surely, with it being so volatile and also with it being a new technology needing public confidence, the motor manufacturers have taken extra care and got all this covered, right?

Wrong. In May 2003 Toyota recalled all its hydrogen vehicles after a leak was discovered in the tank of one. Not by engineers, but by the driver noticing a strange noise when refuelling29.

Safety isn’t just an issue for the filling stations and vehicles. Just think about the tankers on the roads.

A study examining the possibility of trucks carrying compressed hydrogen in tubes showed that for every 200 miles the truck drives, it uses energy equivalent to 20% of the fuel it delivers30. Because it carries so little fuel, more tankers would be needed. The study said ‘it would take 15 tube-trailer hydrogen trucks to serve the same number of vehicles that are nowadays energized by a single 26 ton gasoline truck’.

Fifteen times the number of tankers on the road isn’t just an emissions nightmare, it’s a serious safety issue too. They went on, ‘today about one in 100 trucks is a gasoline or diesel tanker. For surface transportation of hydrogen one may see 115 trucks on the road, 15 or 13% of them transporting hydrogen. One out of seven accidents involving trucks would involve a hydrogen truck. Every seventh truck-truck collision would occur between two hydrogen carriers’.

All the new infrastructure would need serious safety testing before rolling out nationally and globally. Even with a massive and unflinching political, industrial and financial push, we’re talking a couple of decades turnaround. From a climate perspective, we don’t have that sort of time to tackle vehicle emissions.


There are other safety and cost factors too. Hydrogen is highly reactive; it bonds easily with other substances so it doesn’t exist in isolation anywhere on earth and has to be ‘manufactured’ by splitting it from whatever it’s bonded to. It’s reactiveness causes metals, including steel, to become brittle. Pipelines, tanker trucks and other things for storage and supply would need to be made of higher grade materials and/or replaced more frequently. The infrastructure costs would be astronomical.

‘Higher strength materials are more susceptible to hydrogen embrittlement,’ says Jim Campbell of hydrogen manufacturer Air Liquide31.

Lower strength materials are, of course, more susceptible to rupture. Any potential solution would have to see the high strength material lined with a less reactive low-strength one, adding more R&D time and costs on to this already slow and prohibitively expensive plan.

There are the previously mentioned problems about the high number of hydrogen delivery trucks and their likelihood to be in accidents – one in seven accidents involving a truck would involve a hydrogen transporter.

There are some possible ways around this. There are some proposals to have filling stations producing their own hydrogen. This would remove the delivery fleet and pipeline issues. However, putting a hydrogen production plant in every filling station would be phenomenally expensive and, with production dotted at thousands of small sites scattered round the country, completely rule out any chance of ‘carbon capture and storage’ (technology for large fossil-using sites that could catch and bury their CO2 – more on that in a minute).

Also, the smaller scale plants at filling stations would be even more inefficient than making hydrogen centrally, so what we gain by not transporting it we lose in inefficiency.


Astonishingly, some manufacturers are looking at on-board hydrogen manufacturing from a variety of sources. As the manufacture of hydrogen takes place in the car itself, there is no possibility of either using renewable electricity or carbon capture and storage. Most flabbergasting is the suggested use of petrol as the source fuel. Rather than burning it directly, the car uses it to make hydrogen as fuel for its fuel cell. Renault have been working with a company called Nuvera Fuel Cells to develop this32.

Despite all that talk about hydrogen granting countries energy independence, they want to make it from the very oil we’re supposedly becoming independent of! And, in case the idea wasn’t mad enough already, a study found that energy consumption and greenhouse gas emissions of such a vehicle would be greater than a hybrid petrol vehicle33. (And there are plenty of existing cars that way outperform the hybrids, appalling guzzlers of oil products whose level of consumption is something that has to become a thing of the past if we’re to have any hope of tackling climate change34).

In-car hydrogen manufacture is the worst of both worlds, as the EU concluded; ‘Indirect hydrogen (via a liquid fuel and on-board reformer) combines high costs with low greenhouse gas savings except when the fuel is from biomass origin’35.


In theory, hydrogen from fossil sources needn’t involve carbon emissions. We could make hydrogen and have carbon capture and storage (CCS), whereby the CO2 is removed during production and piped to an underground geological gap such as an old gas field or saline aquifer.

In 2005 BP announced it would be piloting the world’s first industrial scale hydrogen production with CCS. The plan was to build a power plant taking natural gas, separating the hydrogen from the CO2, then immediately burning the hydrogen at the same site to generate electricity. This means no compression, liquefaction, storage or transport, making it far more efficient and safe. It also means that this is not a model for production of hydrogen as a vehicle fuel.

The captured CO2 would be stored in a North Sea oil field, pumping it down would flush out about 40 million barrels of oil that are not currently recoverable. In consecutive paragraphs of their press release, BP brag about the climate benefit of capturing CO2 and brag about the extra oil they’ll get to burn!36

The emissions from the oil would have been between a third and four-fifths of the CO2 being captured37. Just like ‘clean’ hydrogen for vehicles and biodiesel from palm oil, the result of this supposed solution would be considerable carbon emissions.

Even though they would only be capturing 90% of the CO2, and using that to generate more emissions so they'd be releasing 47-90% of the emissions of a normal power station, BP still called the plan ‘carbon free’38. No shame, no irony, no way they can be trusted on climate issues.

In May 2007 BP cancelled the plan citing governmental delays in approving the project and giving incentives. They were in a rush because the Miller Field is reaching the end of its life, so they want the CO2 in a hurry to extend it. The cancellation was met with dismay by all who spoke of it including those who should know better like Friends of The Earth39.

They’re looking to build these pilot CCS plants where there’s a dividend like bonus oil, yet we need fossil fuels to stay in the ground to avert runaway climate change.

Still, if Chinese power stations don’t get CCS sharpish then we face severe risk of climate catastrophe, and somebody’s got to develop the technology and deliver it to them. So even allowing for the extra oil, a pioneering project like that could perhaps have had some mitigating merit.

In February 2003 the US government’s Department of Energy announced the billion-dollar ten year FutureGen project to design, build, construct, and demonstrate a 275 megawatt prototype plant that would co-generate electricity and hydrogen and have Carbon Capture and Storage to sequester at least 90% of the CO2. It aims to validate the viability of the technology by 202040.

Assuming they manage it, getting it all mass produced and on stream would take around a decade, which means that, even with generous assumptions on the push behind it, we’re not looking at this being part of the solution until 2030. By which time we’re almost certainly past the climate’s tipping point. Which means this stuff is not going to help us.

We have to find another way – again I find myself thinking ‘leave this shit in the ground’ is the only one that seems safe and certain – and maybe, if this technology works and we have happy skippy clean hydrogen in future we can get our cars back out of the garage. But to keep going on the promise of something that can’t arrive in time is like saying to an alcoholic they should keep on drinking because in thirty years time we’ll be able to grow them a new liver in a laboratory.


Carbon Capture and Storage could be possible by 2030. In that time the world will have doubled its coal-fired generation; the lifetime emissions of those new coal plants are equal to half of the total carbon emissions from all fossil fuel use globally since the industrial revolution41.

Coal is by far the most CO2-intensive energy source. It emits 80 percent more carbon per unit of energy than gas and 29 percent more than oil. Half of Britain’s electricity already comes from coal, and power generation using coal is expanding. Even madder, the government is planning new coal-fired power stations. With oil and gas prices rising and energy security becoming increasingly problematic, the coal option looks more attractive to government. But if a new era of energy expansion is based on coal then we’re all toast.

Put bluntly, we can’t wait for Carbon Capture and Storage to save the day because it can’t arrive in time. We need to avert new coal generation. This means efficient use of all non-coal sources of energy, which more or less puts hydrogen out of the window automatically. Any solution that would inevitably produce a huge leap in energy demand as well as a massive increase in heavy industry should be squarely placed in the ‘absolute last resort’ pile.

Hydrogen requires large amounts of energy to produce, large amounts of energy to store and large amounts of energy to distribute. To increase the demand for natural gas by making it the primary fuel for hydrogen means the price of gas would go up, making coal even more viable for electricity generation.

Yet the gas supply is only temporary too. If we invest in the trillions of pounds of infrastructure to make a global hydrogen economy then we are committing to many decades of hydrogen use. When ‘peak gas’ hits by mid-century, coal – which there is plenty of - will become the cheapest way of making hydrogen.

In the meantime, the emissions figures for gas are underestimates; they presume gas is piped in gas form. However, as the European fields reach the end of their life, gas is being shipped in liquid form from the Middle East. (Once again, the talk of the hydrogen economy creating energy independence and security is shown up as a lie). To be liquefied, gas is cooled to minus 162 degrees centigrade then kept at that temperature, with all the energy consumption that implies, so the emissions from using gas effectively increase.

Increasing hydrogen production – from whatever source – will have a further climate impact. Having so much hydrogen at so many sites will invariably involve leaks. Once it gets in the troposphere (the lower atmosphere) it reacts with hydroxyl to form water vapour. This makes hydrogen indirectly become a greenhouse gas.

Not only is water vapour a greenhouse gas, but hydrogen affects methane levels too. Methane is a greenhouse gas over twenty times as potent as CO2. The primary ‘sink’ for reducing methane in the atmosphere is reaction with hydroxyl. The more hydroxyl has reacted with leaked hydrogen, the less there is to clear away the methane42.


The vision of hydrogen as a vehicle fuel isn’t new. In 1874, Jules Verne wrote in The Mysterious Island, ‘yes, my friends, I believe that water will one day be employed as fuel, that hydrogen and oxygen which constitute it, used singly or together, will furnish an inexhaustible source of heat and light, of an intensity of which coal is not capable. Some day the coalrooms of steamers and the tenders of locomotives will, instead of coal, be stored with these two condensed gases, which will burn in the furnaces with enormous calorific power… water will be the coal of the future.’

It’s easy to dismiss a novel as an inaccurate prediction, and those who do – like those in the year 1984 who said George Orwell was ‘wrong’ because we weren’t living in the place his novel described – miss the point of fiction entirely. Verne’s work is staggeringly visionary to the point of feeling plausible.

But from a literalist standpoint the glaring problem is the absence of the primary source of energy that will separate and compress the hydrogen and oxygen.

Still, Verne’s imaginative work is extraordinary, and as a fiction author his pipedreams do us no harm. The same cannot be said of those who know better and do act in the real world on a grand scale.


Aware of the rising need to reduce carbon emissions, the oil companies have responded in several ways to appear as if they’re taking it seriously. BP have got a pretty sunflower logo, but in their move to produce hydrogen they’ve been outpaced by the oil industry’s PR master, Shell.

Shell are the most visible company pushing the idea that Iceland will be ‘the world’s first hydrogen economy,’ implying that it’s just Iceland leading the way and proving we can all do it. They opened a Shell hydrogen filling station in Reykjavik, and proudly said that they’d immediately signed up 4% of the city’s bus fleet.

That, though, is only three buses. Because this is the thing. Iceland is not only peculiar because it is sat on more renewable energy than it can use (a few huge hydroelectric plants and a hell of a lot of geothermal energy); it is also little more than a city state. It has a population the size of Bradford and two-thirds of them live in one city. So all you need is three or four filling stations and you’re covered. That simply cannot be scaled up to the UK, or anywhere else. The rest of us need a different solution.

Even in Iceland, it was only a gimmick. In 2007 one of the buses was retired to a transport museum, the other two were scrapped43.

In 2001 Shell produced a report called Energy Needs, Choices and Possibilities: Scenarios to 2050. They describe several possible futures. One talks of ‘developing a new “fuel in a box” for fuel cell vehicles’.

BMW’s admittedly inefficient hydrogen combustion car the H7 uses about a litre of fuel to go 2km44, yet in Shell’s scenario ‘a six-pack of fuel (12 litres) is sufficient for 400 km,’ or over 16 times the efficiency of the H7. Even more curiously, they talk of this fuel in a box as being ‘distributed like soft drinks through multiple distribution channels, even dispensing machines’45.

For Shell, all the safety problems of having hydrogen vending machines seemingly don’t exist. Presumably they disappeared around the same time as the mystery safe canister material was invented. They neglect to say whether these will be high-pressure gas canisters (with the risk of rupture and explosion), or ones holding liquid at-253 degrees (necessitating a hell of an electricity bill for the vending machine, or else a super-insulated canister that leaks the fuel and, if it’s been sat in some underused vending machine, may be empty by the time you buy it).

All this, remember, comes from the company whose current slogan is ‘real energy solutions for the real world’.

It’s all a load of fanciful nonsense made up out of thin air, deliberately ignoring the serious and presently insurmountable engineering and safety problems of hydrogen. It certainly isn’t a scenario that could possibly be in place by its projection of 2025. It’s a decoy to make the public think that there’s a safe, easy future and the oil companies have got it all figured out, so we leave them alone and don’t stop them producing oil, nor do we feel the need to cut back on our consumption of it.

Consumption itself is our problem. Even if there were some proven method of Carbon Capture and Storage that definitely locked it away for all eternity without major leakage, it would only address the carbon issue. Simultaneously, it would commit us not only to ongoing coal consumption and all the products that high energy use bring; all our other resource depletion and pollution issues would rampage onwards unchecked.


For hydrogen to be viable as a vehicle fuel it needs numerous technological breakthroughs in all major areas; production, distribution and storage.

The fuel and vehicles, even with some optimistic assumptions about technological breakthroughs, will be more expensive than conventional vehicles to buy, more expensive to refuel, and not last as long.

For it to be a climate technology, we need a glut of renewable electricity or universal effective Carbon Capture and Storage within a decade. The likelihood of the infrastructure that would make it work being put in place within four or five decades is slim. So, the technical and economic issues mean it cannot be a climate solution.

There won’t be enough renewable electricity to make carbon-free hydrogen possible for decades, if ever.

The hydrogen industry has no part to play in reducing emissions if there are fossil fuels anywhere in the equation. The emissions will be equivalent or worse than conventional vehicles, which are already enough to take us to hell in a handcart.

It cannot help us.


  1. Joseph Romm, The Hype About Hydrogen, Island Press, 2003
  2. A back of the envelope calculation. Electrolysis ‘requires 39 kWh of electricity to produce 1 kilogram of hydrogen at 25 degrees C, and 1 atmosphere’ [J. Levene, B. Kroposki, and G. Sverdrup, Wind Energy and Production of Hydrogen and Electricity — Opportunities for Renewable Hydrogen, US National Renewable Energy Laboratory, March 2006, p2. [link] ]

    The BMW H7 has an 8kg hydrogen tank.
    8kg=200km driving range [BMW press release, 14 May 2007 [link] ]
    312 divided by 200 = 1.56 kw/h per km
    UK grid emissions 480g kw/h (see footnote 8)
    480 x 1.56 = 749g/km to make the hydrogen gas

    Then it has to be liquefied:
    With UK energy mix, 6-7.2kg CO2 emissions per kg hydrogen (see footnote 8)
    8kg=200km, so 25km per kilo.
    6-7.2kg for 25km, 240-288g for 1km for liquefaction

    749 + 240-288 = 989-1033g/km

    For comparison, a Toyota Prius emits 104g/km, a Renault Megane emits 117g/km, a vicious gas guzzler like the Porsche Cayenne emits 310g/km. The car the H7 is based on, the BMW 750, emits 271g/km.
  3. George Monbiot, Heat: How To Stop The Planet Burning, Penguin/Allen Lane, 2006, p165
  4. Decarbonising the UK – Energy for a Climate Conscious Future, Tyndall Centre for Climate Change Research, 2005, p74 [link]
  5. United States Department of Energy, National Hydrogen Energy Roadmap, November 2002, p11 [link]
  6. US National Academy of Engineering Board on Energy and Environmental Systems, The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs, 2004, p38 [link]
  7. Raymond Drnevich of major American hydrogen supplier Praxair, Hydrogen Delivery: Liquefaction & Compression, May 2003, p8 [link]
  8. UK CO2 emissions – 480 g/kwh
    [table 3, Fuel Mix Disclosure Data Table, DBERR 2006-07, [link] ]
    480g x 12.5-15(see footnote 7) = 6-7.2kg
  9. CO2 emissions from gasoline 8.8 kg/gallon. US Government Environmental Protection Agency Office of Transportation and Air Quality, Emission Facts: Average Carbon Dioxide Emissions Resulting from Gasoline and Diesel Fuel p2, February 2005. [link]
  10. Manufacturing hydrogen from natural gas emits 9.1 kg CO2 per kg H2
    (IPCC, Special Report on Carbon Dioxide Capture and Storage, Cambridge University Press, 2005, p 131. [link]
  11. US National Academy of Engineering Board on Energy and Environmental Systems, The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs, 2004, p38 [link]
  12. US National Academy of Engineering Board on Energy and Environmental Systems, The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs, 2004, p27 [link]
  13. Alec Brooks, CARB's Fuel Cell Detour on the Road to Zero Emission Vehicles, Electric Vehicle World, 7 May 2004 [link]
  14. See footnote 2.
  15. Hell and Hydrogen, Technology Review, MIT, March 2007 [link]

  16. Hell and Hydrogen, Technology Review, MIT, March 2007 [link]
  17. Marianne Mintz et al, Cost of Some Hydrogen Fuel Infrastructure Options, Argonne National Laboratory Transportation Technology R&D Center, January 2002 [link]
  18. Don Huberts, testimony to House Science Committee, "The Path To A Hydrogen Economy", March 5, 2003 [link]
  19. For those born after 1975ish: In the early days of home video tape recorders there were three different competing formats: JVC’s VHS, Philips’ V2000, and Sony’s Betamax. You couldn’t play a tape of one format on the machine of another.

    As a household would only buy one video machine, they would go for the one that had the best range of movies available, the longest record time on a blank tape, and cheapest retail price. Video rental shops found it expensive stocking every film on two or three formats. There was only going to be one winner. V2000 died a rapid death but Sony forged ahead for years throwing good money after bad plugging Betamax long after it was clearly going to lose.

  20. Barry C. Lynn, Hydrogen's Dirty Secret, Mother Jones, May/June 2003 [link]
  21. Business Week, Fuel Cells: Japan's Carmakers Are Flooring It, December 23 2002 [link]
  22. Joseph Romm, The Hype About Hydrogen, Island Press, 2003, p122
  23. Marianne Mintz et al, Cost of Some Hydrogen Fuel Infrastructure Options, Argonne National Laboratory Transportation Technology R&D Center, January 2002 [link]
  24. Alec Brooks, CARB's Fuel Cell Detour on the Road to Zero Emission Vehicles, Electric Vehicle World, 7 May 2004 [link]
  25. Looking Ahead: Fuel Producers Weigh in on Hydrogen’s Fit in Cleaner Energy Production, Fuel Cell Industry Report, January 2003
  26. NASA, Office of Safety and Mission Assurance, Safety Standard for Hydrogen and Hydrogen Systems, 1997, paragraph 601b(4) [link]
  27. Arthur D. Little, Inc. for US Department of Energy, Guidance for Transportation Technologies: Fuel Choice for Fuel Cell Vehicle, Final Report Phase Two, appendix p107, February 2002 [link]
  28. James Hansel of Air Products and Chemicals Inc, Safety Considerations for Handling Hydrogen, presentation to the Ford Motor Company, Allentown, Pennsylvania, 12 June 1998
  29. Toyota Recalls Fuel-Cell Cars Due to Hydrogen Leak, Agence France-Presse, 20 May 2003
  30. Bossel & Eliasson, Energy and the Hydrogen Economy, 2003, p19 [link]
  31. Jim Campbell, Hydrogen Delivery Technologies and Systems: Pipeline Transmission of Hydrogen, presentation to US Department of Energy, Strategic Initiatives for Hydrogen Delivery Workshop May 7-8, 2003, p6 [link]
  32. Fabien Boudjemaa, The Onboard Reformer: A Transitional Solution?, CLEFS CEA, no 50-51, winter 2004-2005, p35 [link]
  33. Weiss, M., Heywood, J., Schafer, A., Natarajan, V. Comparative Assessment of Fuel Cell Cars, Report LFEE 2003-001 RP, Massachusetts Institute of Technology, 2003 [link]
  34. George Monbiot and Merrick Godhaven, Greenwash Exposed – Toyota, Turn Up The Heat website, 10 May 2007 [link]
  35. Well-To-Wheels Analysis Of Future Automotive Fuels And Powertrains In The European Context, European Commission Joint Research Centre, January 2004 [link]
  36. Introducing hydrogen power: BP's plan to generate electricity from hydrogen and capture carbon dioxide could set a new standard for cleaner energy, BP press release, 30 June 2005 [link]
  37. Jim Bliss, Oil Companies and Climate Change Redux, The Quiet Road website, 20 March 2008 [link]
  38. Introducing hydrogen power: BP's plan to generate electricity from hydrogen and capture carbon dioxide could set a new standard for cleaner energy. BP press release, 30 June 2005 [link]
  39. BP pulls out of green power plant, BBC News website, 23 May 2007 [link]
  40. FutureGen - A Sequestration and Hydrogen Research Initiative, US Department of Energy Office of Fossil Energy, February 2003 [link]
  41. David Hawkins, director of Natural Resources Defense Council's climate center, testimony to U.S. House Committee on Energy and Commerce Hearing on Future Options for Generation of Electricity from Coal, June 24, 2003 [link]
  42. Anil Ananthaswamy, Reality bites for the dream of a hydrogen economy, New Scientist Magazine issue 2421, 15 November 2003 [link]

    See also ‘Atmospheric Impact of Hydrogen’, Platinum and hydrogen for fuel cell vehicles, AEA Technology, 2003 [link, or full report as pdf]
  43. Whatever happened to the hydrogen economy?, New Scientist, 28 November 2008 [link]
  44. A back of the envelope calculation. The 8kg tank gives around 200km driving range [BMW press release, 14 May 2007 [link] ]
    According to Ecoglobe, liquid hydrogen 1.8kg=25litres [link], making the tank about 111 litres. (Wikipedia suggests hydrogen is 0.07kg/litre, making the 8kg tank about 115 litres).
    111 divided by 8 = 13.875 litres per kilo.
    13.875 divided by 25 = 1.8km/litre.
  45. Energy Needs, Choices and Possibilities: Scenarios to 2050, Global Business Environment, Shell International 2001, page 48 [link]