Praxair starts up new Gulf hydrogen plant; largest SMR Praxair has built to date
SAE releases technical white paper on two new low-GWP refrigerants; alternates to R-1234yf

ITM Power reports its estimated cost of producing hydrogen via electrolysis down significantly from last year

ITM Power has provided an update on the cost structure of hydrogen generated by its HFuel electrolysis platform. The new estimated cost —US$4.13/kg after capital amortization—incorporates efficiency improvements, cost reduction of its HGas platform and data provided by Hyundai for the ix35 fuel cell electric vehicle (earlier post).

ITM Power projects hydrogen cost at £4.19/kg (US$6.44/kg), a 32.7% reduction from last year’s £6.23/kg (US$9.57/kg), within a 10-year capital amortization period and £2.69/kg (US$4.13/kg), a 22.9% reduction from last year’s £3.49/kg (US$5.36/kg), after capital amortization.

ITM Power recently announced an overall performance improvement to its rapid response stack platform, enabling hydrogen output to be increased by 11%. ITM Power previously described its HGas (25 kg/day PEM) stack module as being capable of self-pressuring up to 80 bar and responding to changes in power in less than 1 second.

Each stack is now able to generate up to 27.9 kg/day at full load at an efficiency of greater than 77%. This increased stack output means the hydrogen produced from a 1MW system consisting of 16 such stacks has increased from 400 to 446 kg/day.

Further, stack response time has improved. The response of an electrolyser stack is a function of both the electrochemistry of the stack and the power electronics. As part of the continuing assessment of the HGas stack platform, ITM Power has applied a suite of aggressive on/off cycles. These involve stepping the stack from 100% to 0% and back to 100% capacity every ten minutes to measure reaction time and any changes in performance. Changes in current are observed to take place in 0.2 seconds and are not changing during the testing to date.

For the cost analysis, ITM Power used a 1MW (446 kg/day) HGas Generation Module comprising 16 HGas electrolyzer stacks and associated balance of plant. In addition to capital cost and electricity consumption, the analysis includes an indicative annual maintenance cost and utilization factor. The assumptions used are:

Generation capacity 446 kg/24h
Amortization period 10 years
Electricity price 3.5p/kWh (US 5¢/kWh)
Water price 0.13p/liter (US 75¢/gal US)
System efficiency 55 kWh/kg
Annual Service 5% of sale price
Utilization factor 70%

The European cost targets for hydrogen generation are €9.90/kg (£7.92/kg, US$13.07/kg) in 2015 and €5.50/kg (£4.40/kg, US$7.26) in 2025, ITM Power notes. The refueling equipment is quoted separately and is typically tailored to the user’s specific requirements. The costs associated with shipping the unit to site are not included as they are location specific.

Increasing the utilization factor to 100% reduces the hydrogen cost to £3.52/kg (US$5.41/kg), a 29.6% reduction on last year’s £5.00/kg (US$7.68), during a 10-year capital amortization period and £2.47/kg (US$3.79/kg), a 19.8% reduction on last year’s £3.08/kg ($4.73/kg), thereafter.

The key to a low hydrogen price is a high utilization of assets combined with a low electricity price achieved by grid balancing payments, the company noted. One way of achieving high utilization is by combining Power-to-Gas energy storage with refueling. (Earlier post.)

Fueling cost comparison. The Hyundai ix35 FCEV has a hydrogen storage tank of 5.64kg with a driving range of 594 km, representing hydrogen consumption of 0.95 kg per 100 km. At a hydrogen price of £4.19 per kg that wrks out to a price of 4p per km (22.7¢ per mile). The diesel version of the Hyundai ix35 (2.0 liter CRDi 4WD) costs 7.9 pence per km (45.4¢/mile) to operate based on a fuel consumption of 5.7 liters of diesel per 100 km at a diesel price of 139.2 pence per liter (US$8.09/gallon US).



In case electricity would be used for BEV instead of hydrogen at stated 3.5p/kWh (US 5¢/kWh) price to cover the 100 km distance would cost $1.05 exactly. Instead targetet hydrogen cost for Europe for 100 km coverage $7 in 2025!!!! Don't people see hydrogen madness?


They are not selling you power for 5 cents, they are MAKING it for 5 cents and want to sell it for 20 cents. Instead they sell it to their own hydrogen stations at 5 cents to sell you $5 hydrogen.


Tanking the costs apart I can't see how for the UK they can argue for 3.5p/kwh and 70% Utilisation at the same time.
Checking on their website for the origin of their assumptions we find:
Which shows the cost of their equipment in 2012 but not where the electricity cost comes from.
The earliest reference to hydrogen costs on the site is:

'HFuel can be switched on/off in a second and can be demand side managed as a smart load which could potentially result in electricity costs of less than 4p/kWh and potentially as low as 0p/kWh or negative electricity prices in some parts of Europe utilising a high percentage of intermittent renewable power.'

Now that is well and good, but I can't see that you are likely to get 70% utilisation that way.

I have contacted them to seek clarification.
For comparison the average price of industrial electricity in the UK is around 7-8p/kwh.

This would put the cost of running a car on hydrogen in the UK at around the same level as for diesel, but the diesel pays tax at around 60% of the retail price in the UK.

Surprisingly the economics in the US are rather more favourable.
This is primarily because of the lower price of electricity.
For light vehicles the fairer comparison is petrol, not diesel, models which get less per gallon.
Even if the US wanted to switch to diesel in a big way, the way the barrel of oil is refined can't simply be swapped to petrol instead of diesel.
I don't know enough on the subject to comment definitively on whether it would be possible to switch world use over to diesel, but it would certainly not be trivial.
For a petrol i35, the comparable figures to the ones used in the analysis seem to be around 8.5 litres/100km:

So even with electrolysis using average US electricity rates but loaded downward somewhat to allow for some TOU credit we are perhaps looking right now at costs per mile in the same ball park as petrol.

For those who are curious why I bring these issues up, since I am usually regarded as a hydrogen advocate, I actually don't have a dog in the fight, and simply call things the way I see them.

In my view it is in any case pretty easy if it is cheaper that way to combine a plug in car with fuel cells, and in any case we are very far indeed away from any level of bottoming of cost for hydrogen production by electrolysis, and in the case of California for instance only around 1/3rd of hydrogen is mandated to come from non-fossil fuel sources.

Roger Pham

Good point, SJC.
Additionally, Darius, you have not factored in the battery amortization cost and depreciation cost of a battery pack of a BEV, due to battery wear and battery aging. If you do, the total energy cost would come out quite comparable. H2 fuel tank will last many times over the lifespan of a FCV, and the same for the FC stack.

Furthermore, the cost of distilled water at $0.75/gallon is too much. A kg of H2 requires 8 kg of water at a cost of $1.5 is unacceptable. Distilled water can be made at cost of under $1 per cubic meter(1000 kg), or $0.001 per kg, or $0.004/gallon. If one would knock off $1.5 in cost per kg of H2, you can see that H2 is actually quite a lot cheaper than ITM Power's conservative estimate. The eventual true cost of H2 will be well under $3 USD, which is in line with my previous projections.

Why pay more for fuel?


Your arguments work fine if, like me, you advocate a massive build out of nuclear.
They don't work if you actually fancy renewables, as they can't effectively store them.
That is why all those who want renewables as a high percentage of the grid, such as the Germans, are moving towards hydrogen.

Believe it or not, you are not the first one to have thought of the points you raise.

Do do some research on the actual, messy, lossy, business of running a grid using a lot of renewables.


At $6.44/kg and 0.95 kg/100 km, the Hyundai costs $6.12/100 km or 9.84¢/mile for hydrogen.  Also, if it's going to be used as load-balancing for large amounts of RE, utilization will drop quite a bit and costs jump upwards again.

Electricity for a PHEV costs about 2¢/mile where I am.  Gasoline for a hybrid at 50 MPG and $4/gallon is 8¢/mile.  Then there are the additional road and sales taxes which have to be applied to the hydrogen to finance the transport infrastructure.

Electrolytic hydrogen still comes out much more expensive than electricity, and is barely competitive with petroleum at today's prices even without taxes.


'Electrolytic hydrogen still comes out much more expensive than electricity, and is barely competitive with petroleum at today's prices even without taxes.'

I'd say that is pretty good at this stage of the game.
Saving the odd trillion dollars or so on wars in the Middle East would save folk a fair bit, and we have only just started on hydrogen cost reduction.

Even if 2/3rds of early hydrogen production comes from reforming NG we are still way ahead of the game, with greatly reduced energy requirements to power our car fleet in a much cleaner and more energy independent manner.

That is aside from fuel cells natural partnership with batteries.

Batteries for running around day to day and a fuel cell for longer journeys, with quite a bit of that produced by electrolysis sounds pretty good to me.


The problem is that even the $6.44/kg figure is a projection (for a date not given), and does not include the cost of the storage and dispensing apparatus.

By the time the date of the projection rolls around, batteries will have continued to improve by 10-15% per year.  Maybe we'll see some step-increases in performance due to e.g. silicon anodes.  I'm still driving on the same tank of fuel after 6 weeks and 660 miles, and my PHEV uses only the existing infrastructure.  If it came to that, my car could run its ICE on CNG or LNG instead of gasoline.  I don't see an opening for hydrogen.


A li-ion battery is over 400% the lead-acid energy density, yet EV battery cost calculations seem to 'discard' at 80% recharge-ability.

80% of 400% is still like the capacity of 3.2X new lead-acid batteries, which could be used in a thousand stationary power storage applications for many years.

The total cost of producing, storing, infrastructure, transporting H2 plus the initial fuel cell cost would have to undercut(significantly) existing electric power, battery, and EV costs.

Bob Wallace

What's the expected life of fuel cells (in miles)?



I've heard numbers between 60,000 to 200,000 miles.


Using the ACAL liquid electrolyte they hit over 300,000 miles:

The fuel cells being put into, for instance, the Hyundai i35 will not do nearly so well, but even excluding the AVAL technology durability is reaching viable levels.

It seems to me fair to look at best practise, just as for lithium batteries the Panasonic cells are a benchmark for energy density in discussions.


These are electrolysers actually in production, so costs are current, and they have based their costs on UK figures where it is more expensive than in the US.
I have noted above reservations on their methodology, on which I am hoping for a response, but US costs are certainly lower than those in the UK.


Ignoring the cost issues discussed above, I think it's worth considering the energy involved.

It takes 60 kWh of electricity to make 1 kg of hydrogen, so this fuel cell car travels about 1 mile per kWh of primary electricity. A similar size BEV travels 4 miles per kWh of primary electricity.

What country or economy in their right mind would throw away 80% of the hard won electricity? Or be forced to build 4 times as many power stations to run a fuel cell fleet?


Energywise hydrogen would be reasonable producing in CHP or CHHP (combined hydrogen and power or combined hydrogen heat and power) process when water spliting ocures at superhigh temperatures and process heat used for power generation and finaly low grade heat for process steem or district heating but I have no idea what investment cost would be.


I have no idea why you are quoting 60kwh/kg when the article above clearly shows it as 55kwh.
60kwh was the previous generation, so progress is rapid.

As for why they are doing it, when charging a battery is more efficient than storing electricity via electrolysis as hydrogen, that is fine if the electricity is happening when you need it.

Wind power and solar power largely is not happening when it is needed, so places which aim to have a lot of renewables are going for hydrogen.

It is somewhat inconvenient waiting for the summer to charge your car.


Hi Darius.
As you say, CHP is not taken into account in these calculations.
CHP is rarely used in the US with its low energy prices, but many of the countries keen on hydrogen have extensive networks, among them Germany and the Nordic countries.
Audi for instance pending the build out of hydrogen stations is producing e-gas in a CHP installation:

Not only can waste heat be used where hydrogen is produced by electrolysis, but also where natural gas is reformed, and the energy calculations bandied about purporting to show batteries as more efficient energetically than fuel cells already ignore NG reforming, as well to wheels fuel cell cars powered by hydrogen are already as efficient as battery cars using the US grid, without using the waste heat.

The argument is invariably that 'eventually' we have to move on from fossil fuels.

Not only are the figures from PRESENT DAY electrolysis then used without amendment without taking into account high temperature electrolysis and CHP, but electrolysis is simply assumed to be the only route for non-fossil fuel hydrogen.

In reality, at that future date, we have no way of knowing whether direct solar to hydrogen or other technologies will not be the source.
For instance:

All in all I am pretty happy with the costs and efficiency of the electrolysis technology detailed in this article, at this stage of the game.

I have nothing at all against using batteries where they pan out, but fuel cells in one way or another are a very, very handy tool to have available.

If big battery pack BEVs work out, well and good.

A fuel cell car with or without plug in capability would also be a pretty good way of getting around though.

Bob Wallace

So a fuel cell could be toast and need full replacement at 100,000 miles? Any significant recycling value?

EV batteries would be down to ~80% capacity. That means that they should be worth at least 50% the cost of a new pack to a utility for grid backup.


No Bob. A fuel cell would NOT be toast and need full replacement at 100,000 miles. It would need "servicing."


Platinum on the membranes becomes contaminated, so think of 100,000 miles as a "rebuild" not a scraping. The value is still there, the catalyst needs to be replaced.


Places where CHP systems for local district heating can run a variant on this:

'Solid oxide fuel cell (SOFC) developer Ceramic Fuel Cells Limited (CFCL) held a media event on Friday 19th July where it introduced a new financing scheme for its BlueGen fuel cell system. Suitable applicants in the UK can now receive a free BlueGen fuel cell micro combined heat and power (micro-CHP) unit, instantly saving them money on their utility bills with no capital costs up front.

CFCL is targeting social housing schemes with its free BlueGen campaign where one or more of its 1.5 kW SOFC systems can provide baseload power and heating to schools, businesses or apartment blocks. Certain criteria must be met in order to qualify for this scheme and more information can be found on the BlueGen website.'

Of course, for a fuelling station much more than 1.5kw would be needed.
The point is that the station could reform NG releasing hot water for district heating, and then provide still more hot water through using the electricity in electrolysers.

This amortises the electrolysers, so that when surplus renewable power is available, from solar on summer's days, or wind, often at night, then this can easily be used instead of the NG.

What you then have is viable half way house between fossil fuels and renewables, equipment is used economically and efficiently and storage problems of renewables are solved.

Energy efficiency would be excellent.


What will you pay for this efficiency, I wonder?

All of this, to avoid building base-load nuclear plants.

Roger Pham

Base-load nuclear plants are great when winter heating is done by NG or oil or wood. However, when zero-carbon energy will be required, then base-load nuclear plants won't be sufficient to cover this heating energy requirement. So, storage of energy produced by these nuclear plants in the spring, summer and fall must be done for winter heating. Much less H2 storage will be needed when nuclear plants are used as compared to solar energy.

The decision is then depended upon overall costs associated with nuclear + [less]H2 storage + [less] FC-CHP vs. solar PV + [more] H2 storage + [more]local FC-CHP, and local preferences. In some geologic and/or cultural areas, nuclear may not be safe, while other geologic areas have low solar or wind potential and must depend on nuclear energy. The answer is not "either/or", but "all of the above".

Local expertise availability for nuclear energy, and worker reliability is another issue that can be solved by "all of the above" approach, that can utilize all segments of the working population, not employing just the most reliable workers for nuclear energy and not employing the rest for RE.


100% agreement here!

Unfortunately we have to make the best of the bad choices we are left with since nuclear has been nobbled by a coalition of fossil fuel interests and 'greens'.

Mind you, CHP using nuclear is interesting, as the Scandinavians transport hot water for tens of kilometres through insulated pipes.

That is a good way of getting something approaching 3GW of useful energy out of a 1GW rated nuclear plant!


I have posted description of hydrogen CHP - combined hydrogen and power generation. It is not traditional CHP. It means hydrogen is produced at 1000 Cº or higher and after that oxygen and hydrogen gases when cooling are heating steam via heat exchanger to 700 Cº. In that case 100% of heat energy would transformed into hydrogen.

The comments to this entry are closed.