Applications open for the fifth Volkswagen and BASF “Science Award Electrochemistry”; new special award for applied research
2017 Lincoln MKC adds available stop-start for 2.0L AWD

Toyota details design of fuel cell system in Mirai; work on electrode catalysts

While other major automakers have either introduced (Hyundai, Honda) or are in serious development of new hydrogen fuel cell vehicles for the market, Toyota continues to take the point in not just promoting, but also supporting the broader technical (and infrastructure) development required for a large-scale realization of hydrogen-based electromobility.

At the 2015 CES, Toyota announced royalty-free use of approximately 5,680 fuel-cell-related patents held globally, including critical technologies developed for the Mirai fuel cell vehicle. (Earlier post.) At the SAE 2015 World Congress, Toyota presented a set of four technical papers detailing some of the technology innovations used in Mirai fuel cell stack. (Earlier post.) And again at this year’s 2016 SAE World Congress, Toyota presented three more papers: one detailing the development of Mirai’s Toyota Fuel Cell System (TCFS) and two dealing with the critical issues of the fuel cell catalysts.

The Toyota papers were part of the larger World Congress technical session on practical hydrogen fuel cell technology: PFL 720, Advances in Fuel Cell Vehicle Applications, chaired by Jesse Schneider of BMW.

Toyota Fuel Cell System (TFCS)

The 2015 papers provided technical details on the high performance fuel-cell (FC) stack; specific insights into FC separator, and stack manifold; Mirai’s newly developed boost converter; and the new high-pressure hydrogen storage system with innovative carbon fiber windings.

The 2016 paper by Hasegawa et al. describes the details of the larger Toyota Fuel Cell System (TFCS) from the standpoints of improved efficiency and reliability, as well as the simplification of the fuel cell (FC) system.

FC Boost Converter. Hybrid systems are a core technology that Toyota is adopting in every form of its next-generation vehicles; the Mirai uses the same motor, power control unit (PCU), and hybrid battery as existing Toyota hybrids. To enable this technology sharing in the Mirai, Toyota had to make a design advance from Toyota’s earlier Toyota FCHV-adv fuel cell vehicle, launched on a limited lease basis in 2008 (earlier post) with the introduction of an FC Boost Converter.

In the 2008 vehicle, the fuel cell stack and inverter were directly connected, using the same voltage; this required dedicated designs for both the traction motor and inverter to match the unique characteristics of the FC (i.e., low voltage and high current).

By contrast, Mirai’s FC boost converter adjusts the voltage difference between the motor and inverte, enabling the adoption of a traction motor and inverter already in mass production. It also eliminated the restrictions caused by the gap between the FC and traction motor voltages, allowing the number of layered cells to be optimized and reducing the size of the FC stack.

FC System Configuration. Compared to the FC system in the FCHV-adv, the FC system in the Mirai is simplified and more reliable. The most important modifications included achieving a humidifier-less system by eliminating the external humidifier; changing the type of the air compressor; consolidating the functions of the valves (the stack inlet shut valve and the flow diverter shut valve, as well as the stack outlet shut valve and the pressure adjustment valve); eliminating the hydrogen diluter, and reducing the number of hydrogen tanks from four to two.

Configuration of TFCS in the Mirai. Hasegawa et al. Click to enlarge.
In PEM fuel cells, the humidity of the electrolyte membrane must be controlled to ensure sufficient proton conductivity; this function is usually performed by an external humidifier.

To eliminate the external humidifier, Toyota migrated water generated at the cathode to the anode where it uniformly distributed the water onto the surface of the anode membrane electrode assembly (MEA). This was done by developing innovative technology for the stack structure as well as modifying the anode operating conditions.

For the stack, Toyota:

  1. Reduced the thickness of the electrolyte membrane, promoting the diffusion of the water in the air system.

  2. Humidified the system using moisture at the anode—this humidifies the cathode inlet by flowing H2 and air in counter directions.

  3. Reduced evaporation by increasing coolant flow at the cathode inlet.

At the anode, Toyota increased the amount of H2 circulation based on driving conditions. Reducing the anode inlet pressure after ensuring the required amount of circulation promotes the evaporation of moisture and enhances the movement of generated water onto the anode surfaces.

Fuel economy. Mirai has a lower hydrogen storage capacity than the Toyota FCHV-adv. To achieve a similar cruising range as a gasoline vehicle, the fuel economy of the Mirai was improved by 20% compared to the Toyota FCHV-adv.

Two of the main measures enabling this are the reduction in H2 permeability due to crossover, and the use of a Roots-type blower for the air compressor instead of the conventional scroll type.

H2 crossover refers to the permeation of hydrogen from the anode through the electrolyte membrane to the cathode, creating an H2 partial pressure gradient between the anode and cathode. The amount of crossover is proportional to the permeability coefficient of the electrolyte membrane.

Because Toyota reduced the thickness of the electrolyte membrane by

two-thirds to realize the external humidifier-less system, crossover increased substantially. To counter this, and to improve fue economy, Toyota changed the anode control by optimizing H2 partial pressure control and H2 circulation control.

The H2 partial pressure required to prevent degradation of the anode catalyst due to insufficient H2 was reduced by using the H2 circulation pump to ensure an even H2 concentration at the anode. Along with enhanced membrane physical properties, this measure improved fuel economy by roughly 6.5% while achieving both the humidifier-less system and a reduction in H2 crossover.

To encourage the widespread adoption of FCVs, it is necessary to reduce the vehicle price, build an extensive hydrogen infrastructure, and enhance vehicle appeal. The technologies described in this paper enhanced the freedom of vehicle packaging by reducing the size of the system, and ensured that the Mirai has an equivalent cruising range to gasoline vehicles.

—Hasegawa et al.

Work on the Electrode Catalyst

The other two Toyota papers presented during the World Congress session took a less vehicle-specific view, and addressed the large general problem of enhancing the performance of the electrode catalyst, as well as providing insight on the degradation process.

Mizutani et al. reported on two examples of efforts to improve the performance of a platinum-cobalt (PtCo) catalyst in fuel cell vehicles. PtCo alloy nanoparticles have demonstrated much better activity over traditional Pt nanoparticles—but mainly towards the reduction of activation overpotential. An issue with their use is operation at high current when mass transfer process becomes the limitation.

  • By lowering acid treatment voltage, the team enhanced the effectiveness of the removal of unalloyed cobalt, leading to less Co dissolution during cell operation and about 40% higher catalyst mass activity.

  • The use of nonporous carbon support material promoted mass transfer and resulted in substantial drop of overpotential at high current and low humidity. This may suggest an effective strategy towards the development of fuel cell systems that operate without additional humidification, the authors said.

In his paper, Kato reported on an in situ transmission electron microscopy (TEM) method that enables real time, high-resolution observation of carbon-supported platinum nanoparticles in liquid electrolyte under working conditions.

On the cathode of a fuel cell, protons and oxygen molecules receive electrons, resulting in a chemical reaction that forms water molecules. To promote this chemical reaction, platinum particles of several nanometers are used as a catalyst. Because platinum is a rare element and expensive, it is necessary to reduce its use.

One of the factors behind the large usage of platinum is its degradation. It is known that platinum nanoparticles degrade while the fuel cell is generating electricity, thus reducing its output. Therefore, it is necessary to elucidate this degradation mechanism and implement a countermeasure. It would not be an overstatement to say that the very quality of automobiles depends on nanoscale analysis technologies. However, the mechanism behind the degradation phenomenon of platinum nanoparticles has not been fully understood. This is largely due to the technical difficulty in directly observing the nanometer-sized platinum particles in liquid electrolyte under applied electrochemical potentials.


By improving the design of the Micro Electro Mechanical Systems (MEMS) sample holder, the migration and aggregation of neighboring platinum nanoparticles could be visualized consistently and correlated to applied electrode potentials during aging process (i.e., cyclic voltammetry cycles).

Hypotheses of platinum nanoparticle degradation mechanism. Based on in-situ TEM observations, Kato observed platinum nanoparticle degradation via dissolution into the electrolyte and aggregation. Using the improved MEMS chip, he could stably perform cyclic voltammetry measurements. This enables the analysis of the relationship between the potential control and platinum nanoparticle degradation in detail based on real-time observation of degradation. Kato (2016) Click to enlarge.


  • Hasegawa, T., Imanishi, H., Nada, M., and Ikogi, Y. (2016) “Development of the Fuel Cell System in the Mirai FCV,” SAE Technical Paper 2016-01-1185 doi: 10.4271/2016-01-1185

  • Mizutani, N. and Ishibashi, K. (2016) “Enhancing PtCo Electrode Catalyst Performance for Fuel Cell Vehicle Application,” SAE Technical Paper 2016-01-1187 doi: 10.4271/2016-01-1187

  • Kato, H. (2016) “In-Situ Liquid TEM Study on the Degradation Mechanism of Fuel Cell Catalysts,” SAE Int. J. Alt. Power. 5(1) doi: 10.4271/2016-01-1192



It is amazing to see all the technology that Toyota used to improve the performance of the FCEV Mirai PEM.

It is very generous of Toyota to offer (Royalty Free) 5680 PEM/FCEV associated patents. It is an excellent way to promote the use of high efficiency PEM/FCs world wide. This will encourage others to do likewise to further improve PEMs and FCEVs in another 4 to 5 years.

Next generation FCEVs will certainly have lighter, more efficient, lower cost PEMs to support extended range with smaller H2 tanks and lighter more efficient vehicles.

High compatibility with HEV and PHEV drive trains will further reduce total FCEVs development and mass production cost.

A hand to Toyota (and others) for all the resources used to develop and mass produce affordable, practical, extended range, all weather FCEVs.

It is hard to see how such an incredibly complex system, with wear-prone moving parts like pumps and compressors, not to mention life-limited parts like 10,000 PSI storage tanks, will ever win a price competition with chemical batteries.


It is hard to imagine the effrontery of someone who runs another site devoted to the immoderate praise of all things Tesla trolling every hydrogen thread on another site.

You ill-manners are boundless, Jay.


Oops, too late. Hydrogen lost even to trucks:


Well said Davemart! I only read ECI's posts when I need a good laugh.

Account Deleted

This is Toyota’s pet project that will never sell but exists purely for green washing reasons in order to trick soft minded people into thinking that Toyota want a better world to happen.

It has been claimed that hydrogen cars had low weight so I looked it up. Turns out they are far heavier than gassers and just as heavy as long range BEVs.

Toyota mirai Weight 1,850 kg (4,078.6 lb) mid sized sedan. Power 113kW, 2WD.

Toyota prius Weight 1,379 kg (3,040 lb) mid sized liftback. Power 100kW, 2WD.

Tesla Model S Weight: 2,090 kg (4,608 lb) (70D) full size sedan. Power 386kw, 4WD.

Note that Model S 70D is bigger and over 3 times as powerful as the Mirai that has nearly the same weight and only has 2 wheel traction. Imagine the weight of the Mirai with 4WD in full size with 386kW power? I bet 2700 kg or 6000 pounds! So much for the myth about low weight FCV.


If you read the details in the article you will see that the progress from the Toyota FCHV-adv model is not all that impressive. 6.5% more efficient fuel cells and 20% overall. Most of the improvements are based on a more aerodynamic design of the car.

It seems to me that most of the efforts has been made to make the car less expensive by using as many Prius parts as possible.

I don't know if this information is very valuable for other car makers.

And I don't know why they insist to claim that the cruising range is the same as for gas cars. A typical gas car can drive > 1000 km on a full tank. Mirai can drive > 500 km.

Toyota has been doing testing for winter conditions in Norway this winter, and claims that the range is up to 400 km. Practical tests done by a news magazine concluded with a range of 340-350 km.


Henrik - your post is misleading. What matters is the marginal increase in weight per increase in range, not the absolute weight for today's cars. The fact is that adding batteries penalizes you a lot more than adding a little extra composite material for H2 tanks for a given size vehicle. Limiting climate change to 2 degrees means ALL vehicle categories need to be decarbonized, not just the smaller, urban ones in which EVs are currently used.

Account Deleted

GM I can assure you that all vehicle categories can be decarbonized using battery electric energy storage. Tesla is selling a large SUV called the Model X and it is one of the fines vehicles on the planet and it is a BEV. It is not small but can seat 7 large adults. Tesla does not yet make any small BEVs but they will with the upcoming Model 3. Forget about all the other automakers BEVs they are halfhearted attempts and mostly compliance cars. The real comparison is with Tesla’s BEVs and the hydrogen cars that may be offered for sale one day but never actually sell in volume. The newest version of Model S 90D is rated for 296 miles range not far from the 312 miles range of the Mirai research vehicle. Tesla will launch a 100D Model S with over 330 miles range within a year. We will see if Toyota can ever make a FCV for sale with 500kw power that can beat that. I say they can’t.

Your definition of small must imply anything that weight less that 5000 kg or 11000 lb. Over that weigh FCV might be able to begin competing in terms of weight with equal range BEVs. However, self-driving vehicles that will be here before FCVs are commercially available will make that point non-important as well. Heavy duty trucks that are self-driving have plenty of hours during the week where they can charge themselves so we don’t need hydrogen here either in order to save weight.


I think the transportation sector gets a bad rap. Yes we can decarbonize the vehicles. But still there are other better approaches that will do much more. Cars are actually lower than homes or for food sources for energy consumed.

Vehicles are near the endlife with ICEs and into mature batteries, and still up and coming FC vehicles.

I think cars are just a punching bag for the worlds problems, as they are single handedly one of the most regulated industries on the planet. There are low hanging fruit still. If we really really cared, we'd focus on those. Also trains, maybe adding more lines to help goods travel more efficiently.

Depending on where you live it's possible that your car's is cleaner than the air around it.


Henrik - I'm with you that EVs are a great solution for most applications. But my question is: how many application/trips are NOT suitable for EVs? That's where people who only focus on EVs get it wrong, in my opinion.

2 degrees warming means 80% reduction in economy-wide GHGs relative to 1990 levels 2050. Agreed? Because some sectors are nearly impossible to decarbonize (e.g., agriculture) we may need even greater reductions in other sectors like transportation. After looking at lot of medium and heavy duty use data (like VIUS and Fleet DNA from NREL) I am totally unconvinced that we can get to 80% in the transportation sector with EVs alone, aside from having some technology that seems farther fetched than affordable fuel cells (like inductive charging). In particular, long-haul trucking is a major fraction of the 50B gallons per year of diesel. What are we going to do about those trucks? I'm open to suggestions.

GM> adding batteries penalizes you a lot more than adding a little extra composite material for H2 tanks for a given size vehicle.

The limitation for H2 tanks is volume, not weight. These tanks are not going to move past 10k PSI, so the choice is either larger tanks or more tanks. Larger tanks are very problematic in a passenger car without giving up trunk space. More tanks, in addition to taking up space, are prohibitively expensive.

As batteries energy density increases, volumetric and gravimetric gains are achieved.

Very little range increases will be acheived by FCVs past 300 miles. BEVs are already at 295 miles range and battery energy density is likely to double within the next 10 years.

These are limits and capabilities imposed by chemistry and physics, not an R&D budget.

Go ahead and pick your ponies, gentlemen. The score currently stands at 4000:1.


Could it be that e-c-i.c is under estimating near future FCs potential progress in order to favor 10-10-10 (2050) batteries?

Future FCs may use higher energy content liquid bio-fuels with onboard treatment. FC development will not stop in 2016.

Of course, batteries will also be developed and 5-5-5+ batteries may be around by 2035 or so. Only than, will we have affordable all weather extended range BEVs.

Meanwhile, a mix of FCEVs, HEVs, PHEVs and short range BEVs may have to be used to progressively replace ICEVs. to reduce oil consumption, GHG and pollution?


Henrik/ECI -

My question remains: what about medium and heavy duty vehicles? Long-haul trucks go a 1000 miles or more per day through rural regions and there's no way to electrify those miles (unless you install inductive roadway charging). They also account for about half of the diesel use in the U.S. You can't tell me that we can electrify these vehicles.

It would help to move this conversation forward if we all agreed that there's a place for EVs, but there are some vehicles/application in which EVs are unfeasible.

Account Deleted

E.c.i. you are spot on with volumetric energy density also being a show stopper for FCV. I knew that already but was surprised to discover today that gravimetric energy density is also a show stopper at least for “small” FCV below about 11000 pounds.

There is one hardly auto-relevant way to increase volumetric energy density of hydrogen and that is to use super cooled liquid hydrogen in cryogenic tanks. It is not relevant for vehicles because the fuel tank boils off after a few days regardless of whether the vehicle is used or not (not very practical or efficient for non-commercial vehicles not operated 24/7). It is used for space rockets today and it might become a renewable kind of fuel for large future airplanes and large commercial ships that are operated nearly 24/7. For large airplanes to work with hydrogen they would use low weight conventional jet engines instead of heavy fuel cells and electric motors. For shipping I imagine using liquid hydrogen to operate a combined cycle power plant and then use electric engines for propulsion.

GM the solution is to use a 500kwh battery or so in a fully autonomous heavy duty truck that can drive perhaps 100 miles for two hours in cold weather with up to 12 tons of cargo to the next supercharger station where it charges for an hour to drive another 2 hours and 100 mile and so forth non-stop all year round. Labor cost is not an issue as the truck is unmanned.

Harvey I suspect you must be an AI bot optimized for stupidity and hilarious speculation? A real human can’t be that good at it.

GM, I agree that current BEV technology is not going to solve long-haul over the road cargo applications. Truck BEVs are being used successfully for drayage.

Theoretical physics permit battery energy densities which, after accounting for total system efficiencies, exceed liquid fuels like gasoline and diesel. It is unknown how long it will take to achieve sufficient densities to make BEVs competitive for long haul trucking.

The case for H2 in trucking is stronger than for passenger cars, but in an industry where fuel cost is such a large part of COS, it's hard to see hydrogen being very competitive until it also has significant advances. People can speculate all day long about cheap H2, but until it actually happens, it's just so much gas. There is no trend line or power curve to support the bloviation.

Short term, it seems more likely for CNG to be the biggest challenge to diesel. It's only a half step, but at least the technology and infrastructure exists.

Roger Pham

@GM, Henrik and ECI,

Let's calculate to see what volume of H2 at 10,000 psi will it take to power a long-haul semitruck 500 mile range?

Let's recall the Daimler-Benz SuperTruck capable of 15 mpg. One gallon of diesel has about 13% more energy than 1 gal of gasoline, but FC-powered semitruck is also about 13% more efficient, so we can assume that FC-powered semi-truck to have 15 MPGe. So, 500-mi range / 15 MPGe = 33 kg of H2 required.
The Mirai has 122-liter H2 capacity to store 5 kg of fuel. So, storing 33 kg of H2 will require 805 liter capacity, or 215 gallons capacity. Many semitrucks these days sport 200-gallon of diesel fuel capacity to permit 1,200 mi range to take advantage of locations with cheap diesel fuel.

There should not be any problem for semi-truck's tractor to carry 215 gallons of H2 fuel.

With continual decline in Solar and Wind cost, there will be time when H2 from Solar and Wind will cost below diesel fuel. Making H2 from fresh air and sunshine available anywhere using just water and two electrodes has got to be far simpler than digging petroleum many miles deep, transporting it for thousands of miles, then refine it in huge and very expensive refineries...and then transport it again in thousands of miles of pipelines or tankers...
What do you guys think about that, eh?

I don't think it's impossible to package sufficient H2 in a tractor-trailer. But until you have a nationwide network of H2 stations actually dispensing cheap H2 fuel, no one is going to sell a lot of those trucks. Chicken, meet egg.

For long haul over the road trucking you need, all at once, a national network. Or it's just not going to happen. The "hub" approach used in LA for passenger cars just won't work for the trucking application. Huge capital expense.

The physics of H2 are bad. The economics are awful.


Good discussion. Henrik's suggestion of long-haul trucks using autonomous vehicles and recharging every couple hours is quite interesting (I need to read GCC more).

For ECI's last post -- first, long haul trucking with H2 is definitely possible from an energy storage perspective. DOE recently did the calculations and found there's plenty of space on a long-haul trucks using the space where diesel tank previously sat, and the space under the side rails behind the cab. They found it's even doable with 350 bar H2 to get similar range as a 200 gallon diesel truck (their paper is being presented at the EVS conference this summer if you're interested). There are also plenty of other medium and heavy duty vehicles where operators need flexible daily range and may have difficulty charging. The size of this non-electrifiable (is that a word?) segment is still unclear, but we know it exists.

Second, agree that CNG is more "likely" for long-haul trucks but we need to separate what's more likely and what "needs" to happen. If we dismiss this segment then we, de facto, dismiss climate change goals. Other than renewable H2 or maybe AVs with charging every 200 miles, I don't see a solution.

Account Deleted

GM I agree that long-term we need to focus exclusively on zero emission solutions. Renewable hydrogen could be used for heavy duty vehicles on a large scale if we forgo fuel cells that we do not have enough platinum on the planet to deploy on a large scale and non-platinum fuel cells are inefficient and even heavier. It could be done using standard combustion engines instead that are optimized for hydrogen combustion.

However, it is still going to be electric autonomous trucks that will prevail because hydrogen is a very expensive fuel to make from renewable energy because of all the efficiency loses in the process and the expensive non-durable equipment that makes it possible. There are batteries that can be charged 10s of thousands of times before they wear out like Toshiba’s lithium titanate cells. They will be perfect for heavy duty trucks that drive 24/7 for many years before they are used up.

Another cost saving factor about using an autonomous heavy duty truck is that you can skip the cost of making a driver cabin with life support altogether. Just make a large skateboard shaped truck that can carry a standard container on its top and equip it with some robotic arms at the four corners to handle plugged autonomous charging and driving sensors. If you start developing this truck today it and its driverless technology could be ready in about 5 years time. Plan B for not being ready with a fully driverless truck in 5 years is that the truck is launched as a semi-driverless truck that is capable of following a human operated truck. Of cause you keep working on the driverless tech until it is fully autonomous and then relaunch the truck as such or if possible apply an OTA upgrade to the existing fleet of semi-driverless trucks in order to make them fully autonomous.

There are other potential solutions, GM. Carbon-neutral biofuels being just one example. There are other synfuels that use CO2 as feedstocks. Right now, they are expensive to produce. But they are much cheaper from a system perspective than hydrogen, especially hydrogen from RE.

Imagine you have a truck that can go 10 miles per kg. It's got a 1,000 mile route, say Seattle to San Diego. It would take the entire daily capacity of a $2,650,000 H2 refueling station to make that trip. Suppose you have two stations, one at each terminal. you can only support a total of four trucks.

And the fuel cost is greater than $8 per gallon equivalent.

What freight company is going to sign up for that deal?


H2 long haul trucks could carry enough H2 in SS compact H2 tanks for 1000+ Km.

A few as 120 large H2 stations, installed at major highway crossings, would be enough as an early thin H2 network for long haul trucks. This early network would cost about $240M to $360M or less than what USA is spending every month on Oil Wars.

Switching some $$$$ from Oil Wars to H2 stations (installation and operation) subsidies for 10+ years could lower H2 cost by 50% to 70%; progressively reduce oil imports, GHG and pollution.

Of course, the H2 network could/would be increased (2X to 3X) or so every 4 to 5 years or so to satisfy a fast growing FCEV fleet.

It is easily doable.

Harvey, I've already pointed out that the H2 production at these stations is only enough to support *four trucks each*.

Tell me again how 120 stations is going to work?

Tell me using real nameplate numbers, not numbers that just came out of your pipe.

Roger Pham

Sorry, battery just won't work for long-haul trucking. The high weight will cut into payload, and the high power required for charging will be very expensive, especially in rural areas. The battery pack is simply too big and bulky for rapid swapping.

ITM-Power data reveals a 446kg/d-H2 station costing a little above $2 Million. It has 1 MW power capability, so costing above $2,000 per kW, which is expensive when compared to Norks Hydro large-scale electrolyzer that has investment cost of $600-1000 per kW.
To cut investment cost, it would be best to have one big H2 electrolyzer plant and mobile trailerable tanker/dispenser to transport the H2 throughout the metroplex. Alternatively, local H2 piping can bring the H2 from the central H2 plant to local stations.

Eventually, H2 will cost around $4 per kg compressed and retail, to be equivalent to $2-3 petrol.

Roger, the station you specified could support only 4 trucks per day, 16 trucks in operation total. *One two million dollar station can support only 16 heavy vehicles!*

Truly, how would this work in the real world?

The comments to this entry are closed.