McPhy Energy premieres first industrial system coupling electrolysis and solid state hydrogen storage
California Governor signs 6 bills to support burgeoning EV market, including $2B AB 8

Kawasaki Heavy to build first ocean-going liquid hydrogen tanker with demo in 2017; H2 for transport, industry, power in Japan

KHI’s view of a “CO2-free hydrogen chain”. Source: KHI. Click to enlarge.

The Nikkei reports that Kawasaki Heavy Industries Ltd. (KHI) will build the first ocean-going ships to carry liquefied hydrogen (LH2), with plans for a demonstration test by 2017 in which liquefied hydrogen will be shipped from the state of Victoria in Australia to Japan. The project will cost ¥60 billion (US$610 million), according to the report.

As part of Japan’s WE-NET (World Energy Network) research program of the New Sunshine Project begun in 1993, Kawasaki and its other industrial colleagues in Japan have been considering the large-scale marine transportation of liquid hydrogen for some time (e.g., Abe et al., 1998). KHI has previously discussed the concept of such a hydrogen-carrying vessel as part of its Business Vision 2020.

KHI’s view of the timeline of the CO2-free hydrogen chain (from a 2012 presentation). Click to enlarge.

The notion there is to create a “CO2-free hydrogen chain” in which hydrogen is produced in a resource producing country such as Australia; shipped to Japan via tanker and then distributed via trucks and large-scale stationary tanks; and then used for industrial, transportation and power generation applications.

KHI is envisioning gasifying Australian brown coal (with carbon capture and sequestration) to produce the hydrogen, followed by liquefaction and shipment via the LH2 tankers.

One of the key technical challenges is the development of liquefied hydrogen containers for marine transport requiring advanced heat sealing technologies to enable high efficiency in the transportation of the cryogenic liquid.

KHI has already been working on liquid hydrogen containers for multi-modal transportation in Japan.

For the initial pilot phase of operation to begin in 2017, KHI envisions building two small ships with 2,500 m3 of LH2 capacity (2 storage units, along with stationary storage tanks for the liquefaction and regasification bases of 3,000 m3 and a liquefaction plant capacity of 10 tons/day. The pilot ships could thus carry the equivalent of 2,700 tons per year, or about enough to run 35,000 fuel-cell cars for a year.

By 2025, KHI hopes to move from pilot to demonstration phase of the hydrogen chain. This would entail:

  • Liquefaction plant capacity of 770 tons/day
  • Storage tank capacity of 250,000 m3
  • Ocean-going carrier capacity of 160,000 m3 (4 x 40,000 m3 spherical tanks witha boil-off rate of 0.2%/day.

Regulation will prove a critical factor; large bulk transportation of LH2 is not fully covered by any existing codes, although aspects of liquid hydrogen carriers have been considered. The Nikkei report notes that the Japanese government plans to support hydrogen procurement by drawing up safety standards in the near future.

Superconducting motor for the carrier. Earlier this year, KHI announced it had achieved the world’s highest power density with a prototype 3MW superconducting motor. Kawasaki has been developing the superconducting motor to help save energy and reduce the size of propulsion systems powering offshore vessels, special-purpose vessels and merchant ships.

Kawasaki’s prototype 3MW superconducting motor is half the size of conventional motors. The significant reduction in size was achieved by placing a superconducting coil in the rotating part of the motor, and injecting cryogenic gas to cool the coil.

The newly developed superconducting motor will enable such improvements as reduced power consumption and a smaller footprint, allowing greater flexibility of layout and adoption of a hull form that reduces underwater drag. If incorporated into the propulsion system of a typical diesel-powered ship, the overall improvement can amount to some 20% reduction in fuel consumption, according to the company.

Further, if the superconducting motor is applied in the anticipated liquid hydrogen carriers, the boil-off gas generated during transport can be effectively utilized to cool the motor, allowing for a propulsion system with even greater efficiency, KHI noted.


  • A. Abe, M. Nakamura, I. Sato, H. Uetani, T. Fujitani (1998) “Studies of the large-scale sea transportation of liquid hydrogen”, International Journal of Hydrogen Energy, Volume 23, Issue 2, Pages 115-121 doi: 10.1016/S0360-3199(97)00032-3



Oh yeah, great idea: turn your backs on nuclear and build floating bombs instead. *facepalm


Perhaps you would put some numbers on your notion that this is more dangerous than existing NG carriers.


It's more dangerous in that when/if the hydrogen is allowed to enter into a gaseous state (note that "boil-off" gas is being used to cool the electrical motor) it's much harder to prevent leaks as hydrogen can tunnel through any material.


The key to handling hydrogen safely is to remember the phrase "Up, up, and away."

Hydrogen has been portrayed in the popular press as a relatively more dangerous fuel, and hydrogen in fact has the widest explosive/ignition mix range with air of all gases except acetylene. However this is mitigated by the fact that hydrogen rapidly rises and disperses before ignition, and unless the escape is in an enclosed, unventilated area, it is unlikely to be serious.

Demonstrations have shown that a fuel fire in a hydrogen-powered vehicle can burn out completely with little damage to the vehicle, in stark contrast to the expected result in a gasoline-fueled vehicle.


I do however question the idea that this hydrogen will be "CO2 free." Just how effective is the CCS?

Roger Pham

Good point, GreenPlease.

Actually, the whole exercise of transporting LH2 transoceanically is questionable. H2 is better off produced locally from excess solar and wind electricity, and definitely NOT from brown coal all the way in Australia. Alternatively, the coal can be transported to Japan produce H2 locally.

Perhaps there is lack of solar and wind potential in Japan?

Roger Pham

Furthermore, liquefying H2 will rob about 30-40% of its energy, not to mention significant cost investment in cryogenic machinery and storage vessels. Simply making H2 locally and store it in underground caverns would be best. However, in earth quake prone Japan, that, too, might be dangerous in certain area.


As Al notes, some aspects of dealing with hydrogen may be more difficult than NG, but others are easier.
We have decades of experience in dealing with hydrogen in quantity a it is extensively used and piped large distances for refining and other industries.

There would seem no good reason to use the language of hysteria and FUD to describe transporting it by ship.



Like many of these 'cunning plans' for very high percentages of renewables, it all looks great until you start considering the cost of equipment and capacity utilisation factors.

How are you going to amortise the electrolysers if you are only running them in the limited amount of time you have surpluses in wind and solar?

For solar the times are pretty clear, it would be in the middle of the summer, in the middle of the day.

It sounds as though your electrolysers will be idle much of the time.

For wind, supply is very lumpy.
For instance Spain, which on average gets around 10% of its total electricity from wind, can have a lull of a week or more.
Conversely in very windy conditions the percentage of grid supply can reach over 100%.
Clearly those gales do not generally last a high percentage of the time, although they comprise a high percentage of total energy.

Again, you would have a lot of equipment idle a lot of the time.

Roger Pham

H2 is most useful when made from RE or nuclear energy. Some solar PV panels and wind turbines can be dedicated solely for H2 production, and the output of either can be used to make H2. As such, the electrolyzer does not have to sit idle for long. Also, electrolyzer cost will come down along with the cost of FC, so, eventually, we will be able to afford to have plenty of electrolyzers sitting idle a lot of time.

Looking at the development time table of this LH2 ocean carrier, by the time the carrier is ready for operation beyond 2020-2030, that FC and electrolyzers will be very cheap and common place, and so will solar PV's and wind turbines, with tremendous amount of surplus energy, so there will be no need to go all the way to Australia to pick up the LH2 at much lower efficiency and higher cost than domestically-made H2 from local solar and wind energy.


Solar panels dedicated solely to H2 production?
That still would not run the electrolysers all the time, or would need extensive battery back up, with annular variation still making the electrolysers under used in winter.

I don't know where you are getting your costings from, to say that electrolysers will get so cheap that you don't have to worry about not using them much of the time, as you don't detail your costings, and AFAIK that is not remotely in prospect, so perhaps you would specify your sources.

Your assumptions on the costs of wind and solar seem equally optimistic.
Panel costs have dropped a lot, but the same doesn't hold for the rest of the system, and since many of the most favourable sites for wind are already taken and offshore wind is much more expensive, I can see little support for your notion that all the costs will be so low that normal methods of working out how efficient a total system is, and ensuring high capacity utilisation of assets, do not apply would not seem to me to have foundation.


Ok, so I shouldn't have said "floating bombs" but, still, this is a terrible idea for so many reasons. They should just build some MSRs and a couple of breeders and call it a day. It's their best option.

Steve Reynolds

Even liquid hydrogen is not very heavy. It seems a hydrofoil would work well (high speed, low energy use) for a ship like this.

Roger Pham

When both solar PV and wind turbines outputs are dedicated toward the same set of electrolyzers, you can see that they will complement each other nicely! Hydroelectricity with built-in pumped storage can further modulate RE output in order to minimize investment in electrolyzers. Thus, not too much electrolyzer capacity will be needed because when too much wind and solar energy are available, pumped hydro can store away a good chunk, then release this energy when both solar and wind outputs are low, in order to keep the electrolyzers occupied most of the time.

Looking at the steady declining prices of fuel cells in recent past, even when they were not even in mass production level yet, gives me reason to believe that electrolyzer's cost will follow the same trend.

The beauty behind H2 via electrolysis is that it can be produced anywhere and anytime with minimum hardwares in comparison to any other fuels. Petroleum fuels requires complicated refining process, and even NG requires significant purification to remove unwanted components.
Liquefying H2 and thereby robbing 30-40% of its energy value and dragging it across the ocean is really missing the advantages of H2 as mass storage medium of RE.

Nick Lyons

So does Kawasaki Heavy Industries own brown coal mines in Australia or what? This is a loopy idea in so many ways.


They should just build some MSRs and a couple of breeders and call it a day. It's their best option.

Of course it is, but after Fukushima the public will never allow it.


'When both solar PV and wind turbines outputs are dedicated toward the same set of electrolyzers, you can see that they will complement each other nicely!'

With no personal disrespect, it is loose armwaving like this that makes me tear what little remains of my hair out.

So you are assuming that a fairly close contiguous area has both good solar AND good wind resources?
Or alternatively you are conjuring up a grid capable of transferring sudden massive loads to the central site of the electrolysers, and incidentally negating one of their great advantages, that they can generate the power pretty well where needed.

With a wave of the other arm, still more geographically limited pumped storage is created, when the suitable locations are pretty fully occupied providing peaking power.

I've been arguing against those seeking to dismiss hydrogen as an energy source here and on other sites for years, but it seems to me to be way, way to optimistic to simply assume that costs will fall so far in the supply chain that normal criteria for capacity utilisation do not apply, or that alternatively a whole bunch of other capital intensive and geographically constrained equipment can be deployed and amortised.

The easy way of doing things is to use HT reactors such as China's PBR and produce hydrogen utilising the remaining waste heat in combined heat and power, so that the electrolysers etc can run 24/7, but what we are actually going to be stuck with is a very messy set of compromises.

My guess is that renewables will be used to some extent, but what will actually provide the grunt for the hydrogen economy is reformed gas obtained by fracking, and within 10 years or so a large input from methane hydrates and maybe coal bed gassification, and global warming will be largely disregarded.

That would still be a lot more efficient and release a lot less CO2 than present systems.

For instance fuel cell transport after reforming losses is around twice as efficient as petrol.

The Californian alternative fuels mandate specifies 30% should be from renewable resources, which if you are using NG to keep the electrolysers running when wind and sun are not available is do-able.

So emissions would be around a third of present levels, which is pretty good going.

Roger Pham

Yet, with another big wave of the arms, let's consider that FC now costs about $49 US/kW. Let's assume that electrolyzer will eventually cost the same per input kW. Assuming 4,500 hrs/year usage, which is slightly above 50% of the time, for 20 yrs, which will be 90,000 hrs, or 90,000kW input. Assuming 75% efficiency, which will take 53 kW/kg of H2. So, with 90,000kW input, 1,687 kg of H2 will be output. Divide $49/1,687kg = $0.029/kg of H2. So, the cost of electrolyzer will add only under 3 cents USD per kg. Let's say total cost of electrolysis facility cost will double the electrolyzer cost, that will be only 6 cents/kg. Or, if electrolyzer only last for 45,000 hrs instead of 90,000 hrs, in that case, the amortization cost of electrolysis will be only 9 cents/kg of H2, at about 50% utilization including all facility costs.

If a kg of H2 is to be sold at $3 USD/kg, then the 9 cents /kg cost of electrolysis is quite small. With solar PV cost of $0.8/W installed, it will cost only over $1 USD for raw electricity per kg of H2. Adding electrolysis cost and other costs, and H2 can be produced at the cost of $1.50 USD /kg, yet H2 will have a competitive market value of $3-8 USD, depending on local cost per gallon of gasoline. So, you can see the potential for large profit margin for H2 production from renewable energgy via electrolysis.

With wind electricity at 4 cents/kWh, the cost of H2 production will be $2.5/kg via electrolysis, but in comparison to the competitive market value of H2 at $3-8/kg, the potential profit margin is still very high. When considering that FCV having 2-3x the efficiency of ICEV, then the market can bear even higher cost for H2 at $6-16/kg, so the economic potential for retail market of H2 is very good!


Well Roger, for a start you have used the DOE's projected costs for PEM fuel cells at a production rate of around 500,000 car fuel stacks a year.
PEMs don't hit anything like 73% efficiency, and is even pushing it for SOFC's, whose costs have precious little to do with that for PEMs.

You then hypothesised that the total system costs would be double that of the stack, but forgot about that when you decreased the stack life to only (!) 45,000 hours.

I can't be bothered to dig out the actual figures, but stack life is currently more like 12,000 hours than 45,000.

AFAIK there is not a wind facility in the world which hits anything like 50%, and 30-35% is going some.

The wind for the electrolyser will not be available anything like as much as that though, as the 'cunning plan' is to just use the surplus, and most of the time when there is power from the wind it will just be fed to the grid, not used for electrolysis to hydrogen.

So it seems to me that you have just put a set of extraordinarily bad figures to the notion that if all the costs get really cheap, it will end up really cheap.

One thing even on these premises it won't do is get as cheap as running the electrolysers 24/7 from nuclear, or as reforming NG all the time.

The notion that it is in some weird way economic to use solar and wind to produce hydrogen is dependent on their already having a heavy level of subsidy and mandates, so there is a massive sunk cost, and so surplus's are at the margins free as they would otherwise be thrown away.

True enough, in a rather Alice in Wonderland sense, but that does not make it other than ruinously expensive to build out wind and solar further to generate bigger excesses to make hydrogen.

So IOW generating enough hydrogen by this means to really make a substantial difference to energy consumption would cost enormous amounts of money, as a quick check on the energy losses would show.

None of this means that no hydrogen at all could be generated in this way, but I don't see your notion that it would be in any way cheap, and it would take a lot better figures than you have shown to persuade me otherwise.

Of course, if they can up artificial photosynthesis to an efficiency of around 10% from the present 5%, the ball game changes, but that is a different subject to electrolyis from renewables, or at least a totally different technology.

Bob Wallace

"AFAIK there is not a wind facility in the world which hits anything like 50%, and 30-35% is going some."

The Danish offshore wind farm, Ronland 1, averaged 50% nameplate capacity in 2007 and 2012. It has consistently performed at over 40% capacity per year. There are three other Danish offshore farms that return >40% capacity each year.

The EIA reports at least one US wind farm has averaged 50.4% for a year. Median output capacity (capacity factor) for US wind farms is 38% with the Q3 point at 43%.

50% of all US wind farms report 38% or higher capacity.

25% report 43% or higher capacity.


A tank full of liquid ammonia at 70°F contains roughly 30% more hydrogen than LH2 at its atmospheric boiling point.  Shipping LH2 trans-oceanically is ridiculous from a great many points of view.

Of course it is, but after Fukushima the public will never allow it.

I have had some communications with a Japanese who showed all the signs of hysteria over the issue, including illness due to induced stress.  The actual casualty toll of the Fukushima radiation releases is 3 injuries from beta burns, all workers at the plant, all recovered.  Proper public education would fix all of that.  If the Japanese people won't accept the truth, I'm afraid they deserve the results.

Roger Pham


If the lifespan of an electrolysis stack is only 12,000 hours, then whether the stack is used 24/7 or at 30-50% of the time, the cost per kg of H2 will still be the same, assuming negligible aging process when it is not used. The limitation here is in the limited number of hours that the stack can be used. If and when a stack will have infinite lifespan, then and only then cost saving can be realized by going 24/7 instead of part time.
At 45,000-hr lifespan of an electrolysis stack and at $49/kW of input power, the cost of electrolysis estimate per kg of H2 will be 9 cents when factor in the facility beside the stack will last for 20 years.

Whether or not a Solar PV farm or a wind farm is connected to the grid or to the electrolyzer will be up to the owner of the farms. If H2 will fetch higher revenue than feeding power to the grid, then that will be the way to go. Right now, NG prices in the US is too low, so H2 from RE is not competitive with H2 from NG. However, in Japan and Europe, the prices of NG is 2-4x the Henry Hub price of NG in the USA, so, at a certain point in the near future, H2 from RE will be more than competitive with H2 from NG, when the NG must be imported.

Efficiency of electrolysis at 75% is fairly typical of a larger scale setup. You are confusing this with the efficiency of a PEM fuel cell.

Why are you so pessimistic about H2 from RE? People in Germany are using surplus wind energy to make H2 and feed this into the NG system. This will also be the way that many H2 filling stations for FCV will operate, from solar PV or wind turbines, to produce H2 and stored locally for immediate dispensing to the FCV. Most major auto MFG's are announcing mass-production release of FCV's by 2015.

Roger Pham

The way to maximize revenue from a RE farm is to sell peak-pricing electricity to the grid at significantly higher rates than base rate, and then use off peak RE output for electrolysis to make H2. When the H2 is priced on competitive level as petroleum, then a lot of money can be made.

Please be reminded that the cost of transmission and distribution of electricity through the grid is ~3 cents/kWh. If you are using grid electricity to make H2, then you already are at a 3 cents disadvantaged per kWh of power consumed. For example, coal-fired electricity at 4-6 cents/kWh must be added to 3 more cents of transmission cost thru the grid. Instead, you use electricity from PV panels or wind turbines nearby and bypassing this 3 cents cost penalty, in order to get ahead on energy cost reduction for the production of H2. Using solar PV panels with raw output of electricity cost of 2 cents/kWh by bypassing transmission cost thru the grid and avoiding the use of grid-tied DC to AC inverter will reap you far more profit in H2 production. Ditto for the use of raw wind turbine output at 4 cents/kWh when bypassing the cost of grid transmission.

However, if your solar PV farm is producing a lot of output during peak electricity pricing, then you may use a limited amount of grid-tie inverter and to-the- grid transmission line to sell some of this extra output if the revenue of this peak pricing can outmatch what you will earn from H2 production. The advantage of having limited grid connection is to take advantage of low-cost night-time electricity in the grid for H2 production during times of simultaneously low solar AND wind availability. It's all in the economic calculations to determine what will bring back the most revenues.

FCV's demand for H2 will bring out an entire new economic model for RE that will be far more profitable than the previous model of just supplying energy to the grid. Remember that steam reformation of NG to H2 must be done in very large installation to realize cost-effectiveness. Then, you will have to include the considerable cost of transportation of the H2. With on the spot H2 production from solar PV and wind turbine farms nearby, disperse H2 filling stations can be economically built and will be 100% CO2-emission-free, and can spring up anywhere there are enough solar, wind, geothermal and hydroelectric potential.


The actual casualty toll of the Fukushima radiation releases is 3 injuries from beta burns, all workers at the plant, all recovered.

That's a strawman. People rarely get exposed to enough radiation to kill them outright but at lower levels it can still shorten your lifespan. Radiation kills like cigarettes do: In any population the average life expectancy is like a bell curve, exposure to things like radiation or toxins shift the curve in one direction while healthcare shifts it in the other. The Japanese know this from prior experience with radiation exposure.


In short: It will be years before they know what the actual casualty toll of the Fukushima radiation releases is. Years of waiting, years of stress & years of fighting any effort to build new nuclear plants. The Japanese need energy NOW, not after they've been re-educated by some lengthy, proper public education, campaign.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Your Information

(Name is required. Email address will not be displayed with the comment.)