Just How Energy Efficient is the Toyota Mirai? Time to do Some Maths

The all-new 2016 Toyota Mirai Fuel Cell Sedan is making its North American debut this week at the 2014 LA Auto Show, where journalists and consumers will be given their first real chance to inspect Toyota’s plans for transitioning from gasoline to hydrogen fuel and perhaps even put the Toyota Mirai through its paces on the test track.

The 2016 Toyota Mirai has a claimed range of 300 miles per 5 kg fill of hydrogen. But how energy efficient is it?

The 2016 Toyota Mirai has a claimed range of 300 miles per 5 kg fill of hydrogen. But how energy efficient is it?

But while Toyota is finally able to talk about the Mirai’s specifications — including its 5 kilogram compressed hydrogen capacity, 113 kilowatt electric motor and 9-second 0-60 mph time — Toyota hasn’t yet released any kind of fuel efficiency figures for the hydrogen fuel cell electric car.

Luckily, our friends over at Autobloggreen have done some back-of-the-napkin maths, and think they have a realistic ballpark figure of 60 MPGe — or miles per gallon equivalent — fuel economy reading for the all-new fuel sell sedan.

As Autobloggreen points out, the U.S. Department of Energy states that one kilogram of hydrogen is equivalent to around one gallon of gasoline, which means that it’s just about possible to divide the total number of miles travelled on a tank of hydrogen in a fuel cell vehicle by the capacity of the fuel cell vehicle’s on-board hydrogen tanks.

To test its theory on cars with known fuel economy figures, Autobloggreen tried its theory out on hydrogen fuel cell cars with known EPA-approved fuel economies. The 2015 Hyundai Tuscon Fuel Cell vehicle for example, has a 5.64 kilogram tank of hydrogen, and a range of 265 miles per fill. That equates to a fuel economy of 47 miles per kilogram, a little less than the 49 miles per kilogram combined figure from the EPA and Hyundai’s ‘official’ 50 Miles Per Gallon equivalent quote.

The maths suggests an economy of just 60 MPGe

The maths suggests an economy of just 60 MPGe

Similarly, the 2014 Honda FCX Clarity — a vehicle we drove a few weeks back — works out to an efficiency of 59 miles per kilo using Autobloggreen’s maths, which as we’ve already explained, is roughly equivalent to 59 MPGe.

Put Toyota’s own figures for range and hydrogen tank capacity (300 miles and 5 kilograms respectively) and we’re left with a ballpark figure of 60 MPGe.

Naturally, these aren’t official figures in any way, shape or form. But given automakers tend to be optimistic ahead of official EPA figures on range and efficiency, we’re fairly confident that our colleagues over at Autobloggreen are right.

If they are, that means the Toyota Mirai is about as energy-efficient as the promised next-generation Toyota Prius hybrid will be, and nearly a third less energy efficient on paper than the Tesla Model S electric car.

Factor in the electricity needed to make the hydrogen, and things get much worse.

Factor in the electricity needed to make the hydrogen, and things get much worse.

Unfortunately however, it’s not that easy. To make one kilogram of hydrogen from electrolysis of distilled water — a process that’s only about 70 percent efficient — you need nine litres of water and 56 kilowatt-hours of electricity. To make the entire 5 kilogram tank of hydrogen on which the Toyota Mirai can theoretically travel 300 miles, you need to use 280 kilowatt-hours of electricity.

If we take that 280 kilowatt-hours and put it in the Tesla Model S — which itself is one of the least energy-efficient plug-in cars on the market today due to its large size — you’d be able to travel 800 miles.

Fuel efficiency figures — as always — aren’t as straight forward as they might first appear.


Want to keep up with the latest news in evolving transport? Don’t forget to follow Transport Evolved on Twitter, like us on Facebook and G+, and subscribe to our YouTube channel.


Want to keep up with the latest news in evolving transport? Don’t forget to follow Transport Evolved on Twitter, like us on Facebook and G+, and subscribe to our YouTube channel.

You can also support us directly as a monthly supporting member by visiting Patreon.com.

Share on FacebookTweet about this on TwitterShare on Google+Share on LinkedInDigg thisShare on RedditEmail this to someonePin on Pinterest

Related News

  • Michael Thwaite

    I’m always open to a good persuasive argument but, what positive features does this car have? Is range the only benefit over a traditional BEV? That would be ironic as, range is the only problem with EVs but this sacrifices so many of the positives.

    • lee colleton

      Yes, it’s my understanding that CARB is now basing their ZEV credits on range alone. Energy efficiency and emissions are hidden from the equation by the “Long Tailpipe”.

    • Surya

      It would appear that is indeed the only benefit. The performance won’t be top notch and the price certainly won’t be competitive.

    • mvetsel

      1. greater energy storage: 100+ kwh of electricity => longer range: 325 miles per EPA
      2. faster fueling: 3 – 5 minutes for fill up vs. 40 minutes to 8 hours
      3. lower cost: $60k versus $80k – $120k for Tesla (none of which hold 100 kwh of electricity, BTW).
      4. durability: greater tolerance for heat extremes
      5. longevity: range does not degrade every year
      6. lower weight: fuel system weighs a fraction of an equivalent li-on battery and takes up less space as well

      • Michael Thwaite

        There’s the thing, that extra storage isn’t a landslide – it’s not comparable to a gasoline fuel tank – it’s only 20% more than a Tesla Model S and cost-comparable to an i3 with range extender as the author suggests.

        The re-fuelling is not right – 3-5 seems to be 10-15 in reality compared to 20-40 for an EV, again, it’s not a landslide and, in most cases, it’s moot as EVs come fully re-fueled every morning.

        Cost isn’t a fair comparison though – Toyota is loosing money on every car, what’s the real cost comparison? Is the Mirai more expensive than a P90D? What would a Mirai cost if it had a fuel cell and battery combo that could match the breath-taking acceleration of the Tesla? Or even, a BMW i3 which the subsidized retail price of the Mirai matches?

        Heat extremes isn’t fair either, both suffer in the cold but the fuel cell completely shuts down.

        I can’t speak to longevity except for rumors of gradual degradation in fuel cell performance as pollutants in the air build up on the fuel cell over time. Perhaps they can be cleaned out?

        I’m not sure that I can concede the weight argument as we don’t seem to have any direct comparable vehicles but again, I’m not sure how long, if at all, any advantage is going to last. Fuel cells are already using carbon fiber tanks, can they get lighter as battery storage density increases? How much, how heavy or, is it even possible to make a fuel cell that delivers 600kW to launch a Mirai at 1.1G acceleration to match that of a BEV?

      • frittsinn

        You are dead wrong on most of your points.

        1. yes, the hydrogen car has slightly better energy storage capasity as of now.
        2. With Teslas battery swap stations you fill up 2-3 cars in the 3-5 minutes you need to fill your hydrogen car, you loose.

        3. BEV cars are far cheaper to own due to 5-10 times cheaper fuel.
        4. Durability: neither cold or hot climates are a problem for good ev cars like Tesla. FuelCells are extreamly sensitive to contamination in the fuel, with dire consequenses.
        5. Longevity: FuelCells degrade faster than batteries when you look at how mouch power they can deliver.
        6. maby lower witght, but fuel cell drivetrain takes up FAR more space than BEV cars. Have you even seen a Toyota Mirai? Its a 4 seater, and have almost no room in the trunk due to the hydrogen tanks. The Tesla model s has 7 seats, and great luggage capacity both in the trunk and the front trunk (where the miray has the fuelcell).

        7. You forgot to mention that the hydrogen car is way more complex and has a total cost of ownership far beond any EV car on the marked, including Tesla Model X P90D signature model.
        8, you also forgot to mention that you cant fuel your hydrogen car at home for 1/10th the cost of filling at a station.
        9. you forgot to mention that hydrogen cars gos into “limp mode” when you drive them hard, due to the hydrogen fuel cell not being able to charge the micoscopic battery in the hydrogen car fast enough to keep up.

  • JohnCBriggs

    Well done. This thread of the argument has not been picked up by GCR…yet.

  • David Peilow

    The “equivalent to one (US) gallon” statement is the literal energy content of the hydrogen. It ignores the input energy upstream.nnThe best case for efficiency and low emissions comes from the steam reformation of methane – a fossil fuel – but this has to be refined centrally and trucked in. In liquid form a tanker might transport 3 tonnes of hydrogen. For compressed hydrogen it’s about a tenth of that. That’s a lot of trucks. To liquify the hydrogen, 40% of the energy it contains is needed.

    • lee colleton

      Then we need to account for the build-out and operation of fueling infrastructure. Steel for storage tanks require energy to forge and install. Electricity is needed for pumps and filling control systems. Energy spent driving to and from the filling stations should also be considered: No one will have hydrogen fueling stations installed at their home.

      • David Peilow

        If they make the renewable fuel argument, it just doesn’t scale. A single hydrogen pump at some future gas station requires a supply of 3.6MW to operate continuously. That’s the real world output of three large wind turbines – per pump.

        • lee colleton

          That can’t be right. You’re saying it’s 3.6 megawatts to run a hydrogen fuel pump at a filling station? Where are you getting that figure?nnnThere’s just no way a filling station could fit in the massive transformers required to handle that sort of energy load. Do you mean 3.6 kilowatts?

          • David Peilow

            No I mean 3.6 MWnnIf it takes 56 kWh to make 1 kg of hydrogen (source: ITM Power), 2.43 kWh to pump it (source: BOC Linde) and further electricity to compress it for storage, let’s say 60 kWh per kg in total.nnManufacturers claim a 3 minute fill up and between 4.75 and 5.6 kg per tank. Let’s say that’s 12 cars per hour at a busy location taking 5 kg on average.nnSo 60 kWh * 5 * 12 = 3600 kWh per hour = 3.6 MWnnnThe load might be less at night, but if it’s a busy motorway location then probably not much.

          • lee colleton

            The load to generate hydrogen would be run off-peak (save for the small amount to pump it into the car) so this would help with the problem of peak load. It’s still more efficient to charge electric cars but with comparatively lower transmission losses for hydrogen gas produced through steam reformation, hydrogen makes sense for those who are unconcerned about the risks of fracking.

          • David Peilow

            If the hydrogen was made off-peak, that exasperates the situation as the supply to the site would have to be even higher to make the same amount of hydrogen in a shorter time.nnnRegarding reforming fracked gas into hydrogen – yes the emissions are lower per unit of hydrogen made (as I said in the OP) – but you have to compress/liquify it for transport and with 40% losses to liquify I don’t think you can say transmission losses are lower.

          • lee colleton

            Shifting load away from peak usage doesn’t exasperate the problem; it lowers costs and emissions.

          • Kaiser

            Eight Tesla supercharger stalls charging simultaneously @ 120 Kwh would draw a megawatt. So either the problem is tractable with renewables, or Tesla’s approach is also not sustainable.

          • David Peilow

            Only if 8 cars arrived simultaneously and only for the first few minutes of the charge curve. Tesla already installed battery buffers at some sites to smooth this.nnAnd as we know, most EV drivers fill up overnight when the grid is less loaded. Only 10% of charging is done away from home or work, whereas 100% of hydrogen drivers will be filling up at service stations.nnYour own numbers, notwithstanding the flaws I pointed out above, show that one hydrogen pump needs 3x the power of a whole Tesla supercharger station and if we are trying to be sustainable in a world of limited renewable resources then that’s a huge difference. The hydrogen approach throws away two thirds of the energy generated.

          • Kaiser

            Honda has already prototyped a home hydrogen filling station. Even with inefficient electrolysis, a typical home solar install produces enough power to drive a fuel cell car 12K miles a year. nnOver the past ten years, solar power has dropped 7x in price. If that continues, by 2025 solar power will be 3x cheaper than any other form of electricity. If solar should become that cheap, hydrogen production from electrolysis, which is also dirt cheap, will drive solar adoption. Grids will have no ability to take excess solar power at noon, and storage becomes the issue. A $1K solar system with a $500 hydrogen generation and storage system beats a $1K solar system with a $6K battery. Cost trumps efficiency just with ICE and gasoline today.

          • David Peilow

            A typical 4 kWp system will produce enough energy for 12000 BEV miles a year now. One can buy $4000 of off-board batteries and a $1000 inverter now to store the amount generated daily.nnAlternatively if that energy was fed into an electrolyser and HFCV you would get 4000 miles. You would likely run out of roof space for your solar panels before reaching 12000 miles.nnBesides, tell me where I can buy this hypothetical $500 hydrogen manufacture and storage system?

          • Kaiser

            There are 250M cars in the US and 120K gas stations. Assuming a refill every two weeks, that’s only 150 cars per day for the entire station, which is 2 mW per hour continuous. To put that in perspective, the state of Masssachusetts generates 2,156 thousand MWh per year of electricity.

          • David Peilow

            You mean 2 MW.nnWhat is 2156 thousand MWh? 2156 GWh? Source please.nnAnyway, if we take your assumptions (a big if that it’s every two weeks for cars with < 300 miles range), that's 17,532 MWh per year, per station.nnThat's 2,103,840 GWh for all the gas stations in the US. Or roughly 1000x the annual electricity output of Massachusetts.nnThanks for proving my point 😉

    • Methane can be renewable.

  • David Peilow
  • Mark Benjamin David

    Somewhere I read where they claim hydrogen being more efficient than battery electric cars. B.S.nnnI don’t want to keep up, I know enough about hydrogen fuel cells to know it’s a waste of R&D time, money and resources, all of which should be going to advancing battery, super capacitor and solar PV tech.nnnLet’s keep hydrogen for rockets, get gas out of cars and make better batteries. I’ve been waiting far too long for mainstream battery electric cars. Reading all this stuff pushing hydrogen makes me angry. I rewrote this, it was twice as long.nnI can understand why oil companies want hydrogen in cars, but, the car companies? ANSWER: the only reason they won’t make our battery electric cars in high quantities is because they are too reliable.

    • There are many more smart and affluent people now, so capacitor tech getting buried in favor of fuel cells is unlikely. Even thorium came back from shelve.

  • Jones

    “To make one kilogram of hydrogen from electrolysis of distilled water u2014 a process thatu2019s only about 70 percent efficient u2014 you need nine litres of water and 56 kilowatt-hours of electricity.”nnNo, you most certainly do not need even a single kWh of electricity for the basic process of converting water into hydrogen, the only thing you need is HEAT. This heat does not have to be produced by an electric heater (even if perhaps this supplier produces the needed heat in this way)… This fact is something I think is sometimes a bit overlooked. From a purely thermodynamic standpoint “1 MPGe” would be more efficient if the “MPGe” is related to for example hydrogen (FCV) than if it is related to electricity (EV). nnAssume for example that we have an EV with 90 MPGe and a FCV with 60 MPGe, and that we want to use a nuclear power plant for producing energy to them. The efficiency of a nuclear powerplant is about 30-35%, producing electricity from uranium by heating water and driving a steam turbine etc. If we instead then want to use the powerplant to produce hydrogen, for example by high temperature electrolysis of water, the efficiency would be 40-50%, uranium heating water directly driving the electrolysis process. In the hydrogen case, we of course don’t need to transfer into electrical energy before the energy reaches the vehicle, where EPA MPGe “starts looking”, instead we transfer “only” into chemical energy, which is a “lower quality” energy than electrical energy, therefore it is reasonable that there will be less losses in this transfer. So the “well-to-wheel MPGe” in this case would according to the numbers in this example in fact become approximately the same for both vehicles, namely 30 “uranium-to-wheel MPGe”.nnI am not saying that MPGe is useless, but it of course does not describe the complete picture of well-to-wheel (nor is it meant to). All energy used for transportation will of course not come from nuclear power plants as in the example, nor wind power or gasoline. But what I AM saying is that if the second law of thermodynamics is anything at all applicable to the world we live in, which I am certain that it is, and that we can obtain heat in other ways than using an electric heater, which I think we can, it is wrong to assume that 1 MPGe for a hydrogen FCV should in any way be worse or less efficient well-to-wheel than 1 MPGe for an EV and if an assumption should be made, it should be the other way around.

    • BEVs still have less range and slower charging than FCEVs, and a LFTR is the best (near?) available power source, but would it not be safer to take the 10% loss to generate electricity instead of explosive H2 + O?

  • Israel Navas Duran

    Flawed logic: commercial hydrogen isn’t generated by electrolysis but by steam reforming. With the present natural gas prices (e.g. consumer price in SoCal at $2.78/GJ) you could drive 305 km (190 miles) per dollar of natural gas.

    • Commercial hydrogen fuel is produced from sunlight and water in Swindon.

  • Vulcan Logic

    The one thing that FCV tech has going for it that EVs do not is that FCV’s can be refueled in about 5min, comparable to the amount of time we spend at the pump now. The only way an EV can be a practical long-distance vehicle right now is a complex network of battery-sharing stations, which isn’t cost-effective in the long run. Swappable batteries are being explored by some EV manufacturers but the varying sizes and shapes of vehicles would make designing a “universal” battery pack difficult and impractical. While EVs are perfect for driving around town, getting from Florida to California in anything remotely resembling an acceptable amount of time will be the territory of FCVs.

  • Jim Seko

    Hydrogen fuel cell vehicles are a very complicated and expensive way of achieving LESS efficiency and MORE emissions than a gasoline powered car.

    • Fuel cells and electric motors are vastly superior to mechanical explosion engines.

Content Copyright (c) 2016 Transport Evolved LLC