ICCT Thinks Green Hydrogen For Shipping Will Have More Energy Than Electricity Required To Make It

Sign up for daily news updates from CleanTechnica on email. Or follow us on Google News!

---

A month ago the International Council on Clean Transportation (ICCT) released a Europe-focused report on the total cost of ownership of various decarbonization pathways for freight trucking. It was deeply flawed, with unrealistically low costs for the hydrogen pathway as they put multiple thumbs on the scale. Perhaps their analyses of other transportation segments are better? Sadly, no.

To summarize the hydrogen oddities, the ICCT researchers gave hydrogen advantages of much cheaper electricity at truck stops than battery electric vehicles charging in the same locations, a €3 per kilogram subsidy hydrogen with no subsidies for electric trucks, expected no profits for hydrogen sellers but profits for electricity sellers to trucks, greater cost drops for electrolyzers than batteries, maintenance cost improvements for hydrogen trucks but none for battery electric trucks and more.

Did these massive and unlikely variances make hydrogen cheaper than battery electric vehicles? No, of course not. But it meant that the ICCT made three extraordinary claims.  The first claim is that manufacturing hydrogen at a refueling station from electricity and water would be cost competitive with diesel in 2030. The second claim is that the energy costs per kilometer will be only 50% more than using electricity in battery electric vehicles in 2030. The third claim is that hydrogen will be only 10% more expensive than using electricity in battery electric trucks in 2040 on a per kilometer energy basis.

These claims require massive manipulation of comparative costs in favor of hydrogen and still resulted in it being uneconomic. But the claims of being cost competitive with diesel soon and within the ballpark of electric for 2050 are deeply misleading and inappropriate conclusions. Were they transparent about this massive disparity in treatment? No. That something was deeply wrong was obvious from their lead graphical representation, but it took digging through two reports and their ‘rebuttal’ to find the full list.

As of December 12th, the original ICCT post quietly changed. They re-ran the numbers with the same artificially low cost of electricity for battery electric vehicles (something that they don’t require to be superior) but removed none of the other thumbs from the hydrogen side. The gap between battery electric vehicles and hydrogen vehicles increased substantially, although it’s still smaller than reality. And they changed their conclusions as well. They are still claiming that hydrogen, if given massive subsidies, would be cheaper than diesel, but are no longer claiming that energy costs for hydrogen and battery electric are remotely similar.

They still have a bunch of thumbs on the scale for hydrogen and the report remains weak and non-transparent, but its conclusions aren’t as egregiously and glaringly wrong and misleading any more. So, a bit of a win. For anyone interested, the updated report is here.

But this led me to wonder. The ICCT’s scope isn’t land transportation, but transportation. I’ve assessed all modes of transportation and published on them constantly over the past few years as I figured out what made sense and what didn’t. I’ve looked at hydrogen for aviation and shipping in detail as part of my analyses of what pathways make the most sense.

What does the ICCT have to say on the subject? Well, they are big fans of liquid hydrogen it seems.

Let’s start with shipping as I’ve been publishing a series on the subject on Forbes in the past couple of weeks. I’ve spent a lot of time on the subject, created an integrated projection of all shipping tonnage through 2100 on inland, short sea and deepwater vessels, communicate with experts and shipping professionals globally, spoke with the head of the Global Centre of Maritime Decarbonization in Singapore about their decarbonization pilots, debated maritime decarbonization in Glasgow for an audience of maritime engineers and designers and continue to stay up to date on efforts in the field. I’ve assessed a failed liquid hydrogen for shipping fuel initiative in northern Europe and the cost of shipping liquid hydrogen. And I’ve spent a bunch of time looking at ports and integrating shipping with ports so that I understand how solutions might work operationally.

That Norwegian failed liquid hydrogen for shipping fuel solution was interesting. It was going to be using Norwegian commercial electricity rates, which are very low, and liquifying the hydrogen at the dock for bunkering. My assessment of realistic costs for bunkered liquid hydrogen at that specific site was that it would be in the range of US$9,300 per ton. And to be clear, this is the cheapest possible cost without massive subsidies with the entire facility built beside the place ships would bunker, and it has thrown away easily half of the energy used to create it before it gets anywhere near a ship.

Is that a reasonable cost? By comparison a ton of diesel averages $500 in the USA. Liquid hydrogen has an energy density about 2.6 times higher by mass, but $9,300 is over five times as expensive per unit of energy. And it’s not as if fuel cells to electric motors are vastly more efficient than marine diesel motors. They are both in the 50% efficiency range, so there are no savings to be had there.

Then we get into boil off both in ports and at sea. While some LNG ships include reliquification for boil off of their massive, globular tanks and take advantage of it for methane engines, the much tinier hydrogen tanks required for fuel would boil off a lot more and the energy requirements for liquification are much higher for hydrogen than methane, making reliquification prohibitive. Transfer of liquified hydrogen between bunkering facilities and ships will involved boil off leakage as well.

I can safely say the following things. No one in the shipping industry is considering liquid hydrogen as an alternative energy carrier for maritime freight vessels, with the very limited exception of organizations exploring shipping liquid hydrogen as an LNG replacement. Liquid hydrogen’s cost would be absurd. Hydrogen can be green, but it can’t be cheap without multiple massive subsidies like those granted to trucking by the ICCT. The industry is taking two synthetic fuels — green methanol and ammonia — seriously, especially the former, but are realizing this year that the vendors of those products have been telling them appealing lies about future costs.

To be clear, I’ve also assessed wind power assist technologies and have a perspective that given the plummeting of bulk cargo, the dominance of container shipping that results, the requirements of port integration and the complexity of folding rigid sails, the only sail power assist technology that makes much sense is autofurling, bow-mounted parafoils. Why? There’s no room for rigid sails on container ships because containers are stacked high on them and rigid sails must fold in order to not interfere with port cranes once again making them impossible to retrofit to container ships. This applies to Magnus effect rotors and complex rigid-wing folding sails like those recently launched with BAR designs.

So imagine my surprise when I found that the ICCT just published a US-centric report about liquid hydrogen powered shipping with rigid wing sails as the center piece of their analysis. Pacific North West, Great Lakes, Pacific and West Coast routes with liquid hydrogen demand approaching 60,000 tons a year.

A fuel that no one in the industry is using and technologies that don’t make much sense. However, the claim was retrofitting aging US vessels that were already in operation, so I hypothesized they were bulk carriers that would persist, so perhaps the sail aspect might be viable. I do keep trying to give the ICCT the benefit of the doubt.

The Midnight Sun is a vehicle carrier. That actually makes it a reasonable choice for assessing wings. It’s a roll off, roll on (roro) vessel and they typically have flat roofs with nothing much on them. The likelihood that they are suitable for retrofitting with high stress and strain sails is relatively low as they are tall, thin-skinned boxes, but at least it’s not a container ship and we will continue to have significant roro traffic.

Not so the Horizon Reliance. It’s a container ship, so there’s no sane integration of rigid sails with it that passes the slightest sniff test of useful integration into ports and operations. Ditto the Mahimahi.

The Badger is a car and passenger ferry with significant top deck infrastructure, passenger decks and lifeboats, a very typical configuration for the class. Once again, there’s no room for a new sail. A quick scan of the route at least didn’t find that the ferry goes underneath bridges, a common feature of ports for example the Port of Vancouver and the Lion’s Gate Bridge, another sail-related situation that makes folding them a requirement, not an option.

The port of Oakland, however, requires that ships pass under both the San Francisco – Oakland Bay Bridge, clearance 58 meters, but also the Golden Gate Bridge, clearance 67 meters, and the assessed routes include it. The Mahimahi is a container ship with five container above deck stacking, about 13 meters and about the same again to the water. The BAR rigid-wing sails by themselves are 37.5 meters. 26 meters plus 37.5 meters is 63.5 meters, so folding them is not an option but a requirement.

So exactly one of the four vessels chosen for the study are remotely suitable for adding rigid wing sails. It seems that adherence to realism isn’t a requirement for the ICCT.

Regardless, let’s play along. They are projecting significant cost savings from these non-viable sales for the cost of bunkering the liquid hydrogen. What costs are they using? Apparently they are using $4,300 per ton for liquid hydrogen, well under half of the best case scenario with cheap electricity in Norway. That’s odd.

They reference a paper by another ICCT pair of researchers for the costs, the same ones who found the absurdly low costs of electrolysing hydrogen in Europe at truck stops with an initial set of thumbs on the scales. The paper is Current and future cost of e-kerosene in the United States and Europe from March 2022. I’ll be returning to aviation in a subsequent piece, but suffice it to say that e-kerosene is synthetic kerosene made from hydrogen, and as such is always more expensive than hydrogen.

That paper shows the cost of production of green hydrogen to be $4.30 per kilogram in 2020, hence the $4,300 per ton, which is remarkable in and of itself, but might be somewhat more justifiable than the low-balled costs for hydrogen manufacturing at truck stops. There are some things worth noting.

They are, once again, assuming power purchase agreements (PPA) for green electricity with minimal transmission and distribution adders. They are using the levelized cost of electricity (LCOE) for renewables instead of actual PPAs, which is somewhat defensible but odd. They are picking the lowest of solar or wind for the LCOE. And they are excluding all compression and liquification costs explicitly. There are no profits for anyone in this mix. They are transparent about most of this as well, making their system boundaries fairly clear.

However, moving forward their costs per kilogram calculations fall apart. In 2020, the electricity price with transmission and distribution is $80 per MWh. For the cost of $4,300 per ton, that suggests 53.75 MWh per ton, which is within the realm of believability with costs of electrolyzers and 95% utilization. It aligns relatively well with Lazard’s levelized cost of of hydrogen for alkaline electrolysis, excluding pretty much everything except the electrolyzer.

However, as the decades unfold, the cost of electricity doesn’t drop much. In 2030, it’s $77 per MWh but only $3.1 per kilogram, asserting 40 MWh per kilogram. The higher heating value of a kilogram of hydrogen is 39.39 per kilogram. That means electrolysis is approaching 100% efficient and there are zero capital costs in 2030.

It gets worse. In 2040, it’s $76 per MWh and $2.3 per kilogram for a miraculous 30 MWh per kilogram, 10% under the lower heating value of hydrogen of 33.33 kWh. More energy is being created than consumed. In 2050, a Nobel Prize worthy $72 per MWh results in $1.6 per kilogram hydrogen, for a truly miraculous 22.2 MWh per kilogram, a net return of 50% additional energy in the form of hydrogen. None of this is remotely realistic at the briefest of glances.

Is any of this explicable from the report? No, it has some probably unrealistic declines in the costs of electrolyzers and increases in the efficiencies, but electrolyzers don’t become free or break the laws of physics. They aren’t claiming IRA $3 per kilogram subsidies as the difference as they never mention the IRA or subsidies for green hydrogen. They are claiming some downstream e-kerosene subsidies, but those are tacked on at the end of the stream after they miraculously create more energy in the form of hydrogen than they consume in electricity.

By the way, this is all observable at a glance. This does not require arcane math skills or special knowledge. Their table of hydrogen costs in the appendix makes no sense in relationship to anything else in the appendix or report. This report had three authors and two reviewers, and apparently none of them noticed that they were projecting miracles. As with the trucking report, this should be deeply embarrassing for every one involved and for the ICCT.

They appear to assert that assumptions related to the costs are based on an earlier report by a third-party consultant but fully funded and under the supervision of the ICCT, Assessment of Hydrogen Production Costs from Electrolysis: United States and Europe. That study, however, comes out with a fairly reasonable median cost of hydrogen of $8.81 per kilogram in 2020, and even with significant cost reductions in pretty much every element, still close to $6 in 2050, about five times what the referencing study includes. That’s with minimal balance of plant, but includes compression for injection in a transmission pipeline, so it’s manufactured cost at the beginning of the product’s journey, not distributed cost with profits. Clearly that report got it fairly right, but the report that relies on it for some assumptions got it completely wrong.

What does this mean for the maritime shipping study? Well, to start with, the cost of $4,300 which they use for liquified hydrogen is missing massive costs. $4,300 is the cost at the electrolysis facility with no balance of plant, and there are 27 components in a plant without liquification. The liquification capital costs are very high, and liquification takes a third of the energy in the hydrogen to perform. There are no distribution costs which for hydrogen are a very big component of the cost. As a reminder of the ratios, a kilogram of hydrogen manufactured from natural gas without carbon capture in the USA or Europe costs $1 to $2 per kilogram to make, and costs $17 to $35 to buy at a hydrogen refueling station. That’s why over 85% of hydrogen consumed today is manufactured at point of use.

And then everyone involved in this requires a profit.

Let’s make another comparison. The Norwegian facility was likely going to be receiving electricity at the plant at local industrial rates of US$58 per MWh. This is 80% of the best electricity costs that the ICCT assumes in 2050, yet they arrive at hydrogen costs in 2020 that are less than half of the likely cost in Norway for liquid hydrogen, and hydrogen costs in 2050 that are a fifth of the Norwegian absolute best case costs.

Even at their low ball $4,300 per kilogram adding all of components for compression, liquification and profits would result in actual liquid hydrogen costs around the same as that completely economically non-viable liquid hydrogen in Norway, $9,300. So the ICCT is once again massively low-balling the costs for maritime shipping, something which is, again, completely unrealistic.

Is this the only report with unrealistic marine hydrogen costs? No, of course not.

Scaling U.S. zero-emission shipping: Potential hydrogen demand at Aleutian Islands ports uses a cost of $4.06 per kilogram in 2020 dollars, once again without any balance of plant, liquification or profits. Papers assessing China to USA shipping corridor liquid hydrogen infrastructure and viability don’t even bother to calculate any costs, a very odd type of viability assessment indeed. Another odd sail plus liquid hydrogen study which at least only used sails on bulk carriers didn’t bother to quantify the cost of liquid hydrogen at all, although it did price fossil bunker fuel.

All of this is to say that the ICCT’s over focus on hydrogen and inability to get its actual costs for transportation remotely correct cross over from trucking to maritime transportation, where they also combine it with completely non-pragmatic sail choices. They clearly over rely on internal hydrogen cost modeling which is flawed in and of itself, and then don’t bother to add costs for balance of plant, compression, liquification, distribution and profits for anybody involved.

So far, none of their hydrogen related reports for ground or marine transportation stand up to scrutiny. They vastly understate the actual costs of hydrogen as a transportation fuel, and it’s systemic across the organization. Next I’ll look at aviation, and frankly I expect to be disappointed there as well.

Once again, the ICCT’s explicit vision is policy and technology strategy and prioritization guidance around decarbonization of strategy. By messing up so badly and persistently on hydrogen costs, they are failing in their mission and causing active delays and confusion in realistic decarbonization policy. It’s unclear why this state of affairs has arisen, but they really need to address it to regain any credibility.