Patrick Molloy at Rocky Mountain Institute runs through the pros and cons of hydrogen fuel cell vehicles (FCEVs). The big pluses are that hydrogen has an energy density of around 120 MJ/kg, almost three times more than diesel or gasoline. Half the energy generated by an internal combustion engine is wasted as heat, whereas electric drivetrains used by FCEVs only lose 10%. Nikola Motors, a U.S. maker of hydrogen trucks, claims its vehicles can get 12 to 15 mpg, well above the average 6.4 mpg for a diesel truck. But hydrogen production needs to scale up rapidly and reduce costs. The fuel needs transportation and storage infrastructure (though producing hydrogen “anywhere” or using existing gas pipelines can solve this). And diesel trucks beats hydrogen on range (though 500+ miles seems plenty). What are the drivers? The EU has committed to removing gasoline and diesel vehicles by 2030. California and Canada have similar ambitions. Japan currently has 3,400 hydrogen vehicles on its roads and wants this to increase to 800,000 by 2030. China’s target is 1 million by 2030.
Although hydrogen fuel cell vehicles (FCEVs) have been around since the 1960s, they have recently emerged as a potential solution to decarbonise heavy transport. Nikola Motors just announced it has raised $1 billion in funding for its hydrogen vehicle technology, adding some substantial new partners including CNHI and Bosch. Earlier this year, the company also launched a daring roadmap for 700 fuelling stations across the USA and secured an 800-vehicle partnership with Anheuser-Busch to help decarbonise its freight fleet.
What makes FCEVs a good choice for decarbonising heavy transport? Let’s examine the similarities, advantages, and challenges of FCEVs compared with conventional internal combustion trucks.
Hydrogen fuel pumps: same as diesel, gasoline
One of the benefits of FCEVs is that hydrogen uses a fuelling infrastructure that’s similar to conventional trucks. This means that FCEVs could be refuelled at existing truck stops across the country and the fuelling experience would be similar. A truck can be filled with hydrogen in less than 15 minutes and the process of fuelling a FCEV is similar to fuelling a diesel truck; hydrogen gas is pumped into the vehicle tank using a gas pump and nozzle that is similar to a traditional diesel pump.
SOURCE: Nikola Motors
Higher energy density
Another advantage is hydrogen’s energy density. Diesel has an energy density of 45.5 megajoules per kilogram (MJ/kg), slightly lower than gasoline, which has an energy density of 45.8 MJ/kg. By contrast, hydrogen has an energy density of approximately 120 MJ/kg, almost three times more than diesel or gasoline. In electrical terms, the energy density of hydrogen is equal to 33.6 kWh of usable energy per kg, versus diesel which only holds about 12–14 kWh per kg.
What this really means is that 1 kg of hydrogen, used in a fuel cell to power an electric motor, contains approximately the same energy as a gallon of diesel. Taking this into consideration, Nikola claims its vehicles can get between 12 and 15 mpg equivalent, well above the national average for a diesel truck, which is around 6.4 mpg.
Much less heat loss
Electric drivetrains are also more efficient than internal combustion engines. With an internal combustion engine, approximately 50% of the energy generated is transferred to heat; but electric drivetrains only lose 10% of their energy to heat. This efficiency difference shows just how much consumers are losing with less efficient internal combustion engines.
Getting cheaper
Price is another attractive attribute of hydrogen. Diesel prices are currently hovering close to $3.00 per gallon, and with the recent curtailment of Saudi Arabian oil production, it’s reasonable to expect further price increases for diesel. But on the hydrogen front, recent analysis from Bloomberg New Energy Finance suggests the per-kilo production price for hydrogen could be as low as $1.40 per kilogram in about a decade.
SOURCE: Nikola Motors
Lighter fuel cell than batteries, longer range
When it comes to heavy transport, weight matters. FCEVs offer the same high torque that comes with battery electric vehicles, but at lower weight. An example is the estimated weight difference between the battery electric Lion 8 and the hydrogen fuel cell Nikola One; the Lion 8 has a 480-kWh battery pack with a 250-mile range, which equates to about 2–5 tons. A Nikola One, with a range of about 500-750 miles, is estimated to have a 250-kWh battery pack, which would likely weigh around 2.5-3 tons.
Taking these factors into consideration, there is a clear pathway for hydrogen to be a low-carbon, low-cost, low-weight alternative fuel for heavy-duty trucks.
However, FCEV trucks are not without their challenges.
Not enough “green hydrogen” production. Yet
Even though hydrogen gas has no colour or odour, to support the decarbonisation of heavy transport we will need green hydrogen and a lot of it. Green hydrogen, also called renewable hydrogen, is hydrogen that’s made using only renewable energy, typically through the process of electrolysis. Electrolysis of water uses electricity to separate water into gaseous hydrogen (H2) and oxygen (O2), converting electrical energy to chemical energy. There are still questions around how quickly the production of green hydrogen can scale; the manufacturing capacity for electrolysers is only starting to significantly ramp up.
SOURCE: Nikola Motors
Transportation and storage. How?
The main challenges with hydrogen come down to transportation and storage. Hydrogen is produced in gaseous form, and it needs to be stored under pressure or liquified directly. Both of these processes require additional energy, which may or not be from renewable sources.
There are emerging methods that use chemical bonds (typically referred to as liquid organic hydrogen carriers [LOHCs]) or ammonia to transport hydrogen in a stable state. These methods don’t require pressure or cryogenic liquification, and therefore require less energy to transport and store the hydrogen. However, the technology is still in a relatively early stage of development and is not ready for large-scale adoption.
Produce hydrogen locally? Use existing gas pipelines?
Another solution to transportation and storage challenges has been to focus on localised production. Nikola has partnered with Nel and Bosch to deliver a network of local hydrogen production stations that utilise renewable energy sources and electrolysers, thus cutting out the logistics chain of conventional diesel and gasoline supply.
In the future, we could also potentially use natural gas infrastructure to transport hydrogen, reducing the need for large infrastructure development. This could also offer a means to provide hydrogen from central production hubs rather than localised builds.
Diesel beats hydrogen on range
One other disadvantage of hydrogen is the range. According to Nikola, the range of a fuel cell truck is 500–750 miles, depending on load and terrain; Toyota Kenworth FCEV trucks have a range of about 300 miles. This pales in comparison with diesel trucks, which can go well over 1,000 miles without refuelling. However, with drivers limited to 500 miles a day, this factor may not cause a significant disruption to standard practice.
Clean fuel policies, investments
Even though there are challenges, the time for hydrogen is now, and here’s why:
We are seeing increased regulatory pressure and industry demand. The European Union has committed to removing gasoline and diesel vehicles by 2030. At the same time, clean fuel standards and associated investment in California and Canada are creating the policy basis for change. Hyundai is planning for the production of up to 700,000 FCEVs per year by 2030, and Japan is targeting 800,000 FCEVs by 2030. And, with technology costs that are anticipated to reach break-even with diesel trucks in several markets, there is significant momentum and investment in hydrogen.
The more projects that increasingly use fuel cell technologies, the more potential for cost reduction and investment in the technology. China’s commitment to get 1 million fuel cell vehicles onto the roads by 2030 (with $7.6 billion being invested in heavy-duty trucking) offers huge potential for significant advancements in the efficiency and cost points for fuel cell vehicles.
Hydrogen has seen false dawns before, but this low-carbon alternative is being pushed by some of the largest companies on the planet across multiple sectors. Toyota Kenworth has a long track record of developing trucks using fuel cell technology and in 2019 it added 10 T680s to be used at the Port of Los Angeles and throughout Southern California. Shell has recently invested heavily in large-scale hydrogen electrolysers, which offer a zero-carbon option for hydrogen production. Earlier this month, Cummins acquired a market-leading electrolyser and fuel cell manufacturing firm, Hydrogenics, for $290 million. These are all signals of serious commitment by industry leaders to move into the hydrogen and fuel cell space.
Rocky Mountain Institute (RMI) is working to identify the opportunities for green hydrogen to accelerate decarbonisation in sectors that have struggled to make progress, and we are only now starting to see the role and position this technology can have in decarbonising the freight sector. You can watch a panel discussion on hydrogen in trucking, hosted by the North American Council for Freight Efficiency (NACFE).
***
Patrick Molloy is a Senior Associate, Industry and Heavy Transport, at Rocky Mountain Institute
This article is published with permission. Copyright 2019, Rocky Mountain Institute
I keep wondering where this recent move ‘chearleading for hydrogen’ from the RMI came from. The article lacks even one small mention of increasing specs of batteries. State of the art batteries are at 250 Wh/kg which results in the stated weight in this article. With solid state electrolyte this could probably double in the coming decade. On TCO this is one serious risk to the deployment of hydrogen in (heavy) transport. Hydrogen needs to focus on the R&D of the SOEC electrolyser, not on the massive upscale of it’s use. There is still the lack of sufficient green electricity, something that will also take the coming decade to reach 70+% of the market.
In order to fast-charge a battery (which would be necessary), the electricity requirement is 1MW per truck. Thus for 100 trucks you need 100MW of electricity, available instantly; which means you need a gas power station to supply that demand. So its not cheap.
With hydrogen, the electricity provided by low-cost, off-peak renewables is stored. During times of peak electricity demand, the electrolysers are powered off. The utilisation rate that Nikola and others are working with is about 70% as far as I know. So electrolysis makes a lot of sense in combination with renewables.
The reason REs are not being developed even faster is because of infrastructure and storage constraints; and so electrolysers and hydrogen pipelines will allow a significantly larger expansion of the industry. Pipelines carry 10 times more energy than HVDC for the same cost, and don’t have planning restrictions, which can take a decade or more.
The logic of hydrogen is twisted. If it is less efficient than electricity, how can it be that if I want to charge electric 100MW is a problem, but if I need (by definition) more than 100MW to make hydrogen it is not a problem?
The argument that hydrogen can be made from off peak renewable electricity is the same nonsense. Why can’t I use an electric truck at that moment to charge, but I can use an electrolyser at that moment? Apart from the fact that electrolysers would want to run as many hours as possible in a free market. After all businesses are run on marginal costs.
If you want to move as fast as possible to a decarbonized world, it makes sense to use the most efficient applications.
Even though HVDC carries less energy for the same amount of money, over their lifetime the avoided energy losses should be taken into account. So an TCO perspective is appropriate here. And I can’t imagine pipelines not subject to planning…
The majority of large truck operation is during daytime hours. Not long haul sleeper cab type work.
That means that if the truck has enough range for a day’s work (Tesla’s 300 mile daycab version) then it can charge during late night hours when wind turbines are often producing a lot and regular demand is low.
Obviously this article was written by a non truck driver. In the article it says that it seems like 500+ miles should be plenty to paraphrase. Well as a truck driver that drives 700 miles a day in all types of weather and wind which cuts the fuel economy we need a built in “cushion ” for distance. In our world of time is money we can’t stop and fuel. Unfortunately that is the facts. We use large tanks of fuel to be able to fuel at home terminals which cuts fuel cost or in the case of over the road drivers not have to buy fuel in high cost of fuel states. Just 500 miles won’t cut it if you want wide spread adoption of this new technology.
Are you not required by regulation to take rest breaks? IIRC drivers are required to take a 30 minute break in the first eight hours of driving and are restricted to eleven hours per day driving.
With a Tesla sleeper you start with 500 to 600 miles of range. To hit your 700 mile goal you’d have to pick up another couple hundred miles of charge which you could do in your required stop. Then fully charge during your required down time.
It is much less simple and obvious than this.
The best existing fuel cell electric vehicles are perhaps 64% efficient when you consider the fuel cell as well as the electric motor. https://www.truckinginfo.com/330127/how-nikola-plans-to-make-hydrogen-the-truck-fuel-of-the-future There are more efficient fuel cells known, eg molten carbonate, but they operate under conditions that makes them unsuited to vehicles. https://www.californiahydrogen.org/wp-content/uploads/files/doe_fuelcell_factsheet.pdf Doubtless there will be improvements in future.
Then there are substantial further energy losses in making the hydrogen in a zero carbon manner. So the end-to-end efficiency from basic energy to moving vehicle is lower than an internal combustion engine. When you consider how much electricity you need to generate to have lots of vehicles running on electrolysed hydrogen at this efficiency, it comes to staggering amounts, so if this becomes a widespread method substantial grid upgrades are needed.
But then we have to make hydrogen at scale. We have insufficient understanding of how to do electrolysis at scale to count on it being practical in the near future. Also, making hydrogen only when there is surplus wind/sun means having expensive plant that has a poor load factor. Steam methane reformation is implementable at scale, and can be made carbon neutral with carbon capture and storage or reuse (CCSR), which is implementable but expensive.
It will be worth keeping an eye on whether carrying on using diesel and making it carbon neutral through CCSR by air capture, which might be paid for by a charge on diesel purchase, might prove competitive.
Combustion engines are between 17-24% efficient. Fuel cells are 60% efficient as you have mentioned. If you want to add the efficiency of extracting fossil fuel, refining it, shipping it from Saudi Arabia, and then trucking it to the fueling station, then the overall well-to-wheel efficiency is about 5%.
With renewables and an electrolyser; all that is required is a cable connection to the wind or solar farm, and the efficiency of the electrolyser which is about 80% for modern designs and some are now 96% using an overpotential which would work very well with midday solar for example.
There are many large-scale electrolysis manufacturing facilities now under construction, with a 360MW per year production facility being built for Nikola alone by Nel, and ITM are building a 1GW per year plant. Costs are expected to reduce similar to wind and solar over the decade.
So a 20% energy loss going from wind turbine or solar panel to hydrogen. Then another 10% loss for compression. 100 MWh of starting energy is now down to 72 MWh. And a further 40% loss converting from hydrogen back into electricity. And a 10% drivetrain loss. 72 MWh drops to 39 MWh of kinetic energy.
There’s an additional energy loss for hauling the hydrogen to the fueling station unless each station is also a generation plant.
Take that 100 MWh and use it to charge batteries. Down10%. Lose another 10% to drivetrain friction and there’s 81 MWh of kinetic energy pushing the truck down the highway. 2.3x times more efficient.
And that’s assuming a 60% efficient fuel cell.
You guys are really funny. You think the electricity magically makes it from the wind turbine/solar panel to the truck at the exact moment the truck needs it to do its duty. Oh how convenient that is that just when I need to charge up the truck, the electricity is available from solar and/or wind. This despite what is obvious to any thinking person that these technologies do not always provide energy on our “schedule” and vary significantly by season.
I know, you are going to say build transmission to shift the energy as needed. You ever cost out how much it takes to build high voltage high tension lines nowadays, assuming the NIMBYs will even allow it? You realize that, though convenient, electric transmission lines are some of the most inefficient ways to transport energy. The electricity transmission/distribution system needs as much as 7 PERCENT, that’s right 7 PERCENT of the energy just to get to end user. A pipeline, unit tanker train, ship/barge etc. uses less than 1% of the energy in the fuel for transmission/transportation and delivery. Even over the road tanker trucks use around 1-2% of the energy for a mid distance haul.
What is the plan for inevitable failures in transmission/distribution lines owing to any number of factors like disasters, war or simple wear and tear. Fluid fuels, including H2, though not as great as gasoline/ diesel, nat. gas etc. can be carted around by myriad means in the event of emergencies. You can pipeline them in, put them on barges, tanker ships, tanker trains , tanker/tube trucks, etc. as necessary and they will be available before the grid is available because transport infrastructure is usually the first thing to be attended to in an emergency.
Electricity is the highest quality energy there is (look up on ‘exergy’). So actually it’s the most efficient form of energy. Yes, there are transport losses, but in Europe these tend to be lower and reliability of the grids up to 99,99% (yearly 1 hour of loss on average). So with increased reliance on electricity only a good reason to improve on the electricity infrastructure in the US. And that doesn’t take into account that some of the renewable energy in the form of distributed energy systems is closer to the point of consumption.
And yes you generally first want to increase the share of renewable electricity before increasing your use of electricity. But this article states 2028 and by that time this % might be hovering around 70%. Even on current grids electric vehicles (and so trucks) are more CO2 efficient than fossil fuel counterparts.
Here’s some news for you. Almost all the time in the US if you flick the light switch in your house the light comes on. Or grid is not as reliable as some in Europe but it’s pretty good. When it’s time to charge a truck electricity will be available, almost always. And if the grid is down don’t expect a hydrogen filling station to function.
If you want to change the topic to how a 100% RE grid would operate you can. But it has nothing to do with H2 FCEV vs. battery powered trucks.
Despite an attempt to appear balanced, this is pretty clearly a puff piece. The first three points in favor of hydrogen are all misleading at best.
1) Hydrogen fuel pumps are definitely NOT the same as diesel and gasoline. Hydrogen is delivered to the truck as a highly pressurized gas, not an ambient temperature liquid. The new infrastructure required to support hydrogen storage for fueling would be extensive, costly, and dangerous. The only practical way to handle enough hydrogen to supply a hundred trucks a day would be to build a pipeline from a remote storage facility. High powered booster pumps at the fueling point would be needed to compress it from the 20 bars or so in the pipeline to the 700 bars tank pressure in the truck. Active cooling systems to dissipate the heat of compression of the gas would be needed.
Granted, that’s all assuming that nothing better than the 700 bar graphite-wound pressure tanks now used in hydrogen FCEVs comes along. DOE research been diligently trying to crack the hydrogen storage problem since Jimmy Carter was in office. In all that time, the best they’ve managed is to reduce the cost of super high pressure tanks.
2) Higher energy density? One really has to be digging for reasons to support hydrogen to raise that old canard. Sure, it’s true if you’re only looking at the basic chemistry and molecular weights. But that’s totally ignoring the weight of the hydrogen storage system. The best systems today — aside from rocket tanks briefly holding liquid hydrogen — mass about 16 times the hydrogen they hold. The effective energy density for hydrogen when the tanks are included is less than a quarter of gasoline or diesel.
3) Much less heat loss? Hah! The 90% efficiency figure claimed is for the electric drive system proper, and doesn’t account for losses in the fuel cells and required auxiliary equipment, much less the losses in production and delivery of the hydrogen.
I could go on, but I think the drift is clear. All that said, I’m not totally down on hydrogen. What I’m down on is irresponsible hype. Hydrogen IS clean, and there are situations where it makes sense as a transport fuel. To much to get into here, but I’ve written about it elsewhere. Ask google.
Hydrogen can work for decarbonizing trucking, but it won’t be nearly as easy, as cheap, or as efficient as this article suggests.
It’s absurd to cite, as an advantage of hydrogen, that it has a super-high mass energy density. State-of-the art for hydrogen storage tanks is about 16 kilograms of storage tank for every kilogram of hydrogen stored. Add tankage into the balance, and hydrogen’s 4x the energy density of gasoline or diesel and 40x that of lithium-ion batteries drops to 1/4 that of gasoline or diesel, and only 2x that of current generation lithium-ion batteries. With developments that are on the horizon, lithium-ion batteries could soon match the energy density of hydrogen + tankage, and with a hell of a lot more efficiency.
There’s a possible way to double the mass energy density of hydrogen + tankage. That’s very cold compressed hydrogen. If the gas is cooled to around 140 degrees K — roughly halfway between dry ice and liquid nitrogen — the same volume and pressure of tankage will hold twice the mass of hydrogen. That brings the effective energy density of hydrogen + tankage to half that of diesel fuel. At that performance level, long haul HFC trucks with 1000-mile range are no sweat, and 1500-mile HFC airliners are probably feasible.
I don’t know if anybody is actively pursuing that approach. It isn’t particularly suitable for cars, because they may sit for long periods soon after being fueled. Even if insulated, their cold tanks of hydrogen would warm up. As much as half their load of hydrogen would need to be vented. But the approach is fine for long haul trucks and aircraft. In both cases, the vehicles will begin using fuel shortly after filling up. The drop in tank pressure as hydrogen is consumed will be greater than the rise in pressure from the tank’s warming.
Then there’s Tesla battery powered semi. Close to 600 miles range fully loaded. Tesla has a new battery chemistry that should give batteries a million mile lifespan.
Cost per mile fueling with hydrogen compared to running on electricity is likely to make hydrogen a non-starter. Between the high energy loss between producing hydrogen and converting it to kinetic and the cost of building and maintaining a hydrogen infrastructure companies that run H2 fleets just won’t be competitive.
Let’s see this mythical officially verified 600 mile range battery only truck that can haul any decent payload (20,000 lbs or more). I am very anxious to see it and will state loudly, my opinion will change upon observation of this reality. And you guys claim that hydrogen is pie in the sky!! You are really hilarious :).
To get a battery truck hauling 45,000 – 50,000 lbs over 600 miles, you need a battery with at least 2 MWH. Even assuming a doubling of Li battery density from the current 250 wh/kg (which by the way is not a durable version that will last 1,000,000 miles), you need at least 10,000 lbs just for the battery and not including other heavy components like drivetrain and auxiliary (about 1,000 – 2,000 more lbs.) for tractor weight well north of 30,000 lbs. If you make the supporting structure light to compensate for the heavy battery, your truck will not last even 100,000 miles and will fall to pieces from the first few uses.
We know the limitations of hydrogen, the efficiency penalty etc. but we at least know it can haul a truck the described distance with a given payload because we know the current state of the art for hydrogen tank and fuel cell system weight. For 6 – 7,000 lbs, we know such a truck can be built today with current technology. We even know, at the price of efficiency, that using LH2, the hydrogen truck can match and even beat diesel in terms of payload/range.
Tesla has built two prototypes and has used them for hauling between the Bay Area and their gigafactory outside of Reno. A number of major corporations have tested the semi and pre-purchased a bunch. At least 500 that have been announced. I think it’s safe to assume that there’s some there there.
Tesla has not released info about the size and weight of the battery pack so let’s use your numbers “at least 10,000 lbs just for the battery and not including other heavy components like drivetrain and auxiliary (about 1,000 – 2,000 more lbs.)”
The T-semi will use four Model 3 motors connected to drive wheels so outside of the motors there won’t be a lot of weight. How about we settle for no more than 1,000 lbs/0.5 tons? Plus 10,000 pounds for batteries takes us to 5.5 tons.
Here’s some data on diesel tractors from the DOE.
“Powertrain includes engine and cooling system, transmission and accessories.” 4,080 lbs.
“Miscellaneous accessories/systems includes batteries, fuel system, and exhaust hardware.” 3,100 lbs.
Not included in that 7,180 pounds are the drive axles and fuel. Fueled up a diesel tractor could have 9,000 pounds or 4.5 tons of stuff that wouldn’t be needed in a battery powered truck.
https://www.energy.gov/eere/vehicles/fact-620-april-26-2010-class-8-truck-tractor-weight-component
And this is from an analysis done by Randy Carlson who tends to get stuff right.
“each solid axle – hub motors – disk brake combination shown here weighs somewhat less than a corresponding diesel truck differential, axle, and brake equivalent”
So based on your numbers and the DOE numbers for diesel the T-semi would have about a one ton loss in cargo capacity out of about 24 tons or a 4% lower capacity than a diesel rig. Seeing how a number of shipments probably reach volume capacity before weight capacity it wouldn’t be the case of a constant 4% loss, but somewhere between 0% and 4%.
And I’m willing to guess that you might have gone light on the batteries. A battery powered semi might have to give up as much as 10% of weight capacity on some loads. But you’ve got to offset that against operating cost differences for fuel/maintenance and electricity.
You are right to be critical of current claims on 600 mile range and of course this should first be verified in the real world. That’s why you want to start with the short range, light weight vehicles / delivery vans first. Or even broader first 2 (or 3) wheelers, personal vehicles, small vans, light trucks, longer range trucks and lastly heavy duty trucks (or even beyond that ships or airplanes). However, this article states hydrogen in 2028 so it’s only fair to compare this to 2028 battery developments.
If you know much at all about Elon Musk then you’d know that he is a very truthful person. If he says that the semi Tesla will manufacture will have a range of over 500 miles, “closer to 600” then that’s a probable.
Tesla already has prototypes on the road and has used them to haul full loads up the Sierras from the Central Valley over the top to Reno. Given the itchy trigger finger of Trump’s SEC Musk has to be careful to not exaggerate.