![[Go to Home page]](../images/smallseg.gif){kind=small} |
Australian Academy of Science Science at the Shine Dome Canberra, 3-5 May 2006 |
|
Full listing of papers
|
SCIENCE AT THE SHINE DOME ANNUAL SYMPOSIUM Science on the way to the hydrogen economy 5 May 2006
Fuel cells
 (Click on image for a larger version)
Thank you. It is a great pleasure to be here amongst the talents, ‘the cream of the country’ – so to speak, and teachers and students. I am going to provide a very, very quick and fast overview of fuel cell technology today, so my apologies if I move a bit too fast.
As we heard in earlier talks, William Grove and Schönbein invented the principle of fuel cells in 1839, and there is some reference to work by Ritter in 1801. Fuel cells have since provided on-board power for the Gemini and Apollo space craft, and also for the Space Shuttle. Fuel cells have also been in use for defence applications for many years. Now they are undergoing early trials for stationary and transport applications as well as for portable power applications. I will now look at each application individually.
There are a number of fuel cell cars made by different manufacturers.
General Motors’ HY-WIRE fuel cell vehicle was brought to Australia last year. These slides show Toyota cars, Daimler and Honda, General Motors and then Daimler and Toyota fuel cell buses.
This auxiliary power unit, which fits in the boot of a car to provide the electric needs of the vehicle, is a fuel cell unit that generates a few kilowatts.
So, almost every major automotive manufacturer has a program for developing fuel cell technology. Indeed over 40 buses are undergoing trial in different parts of the world, including three in Perth. There are even fuel cells in scooters, wheelchairs and bicycles.
Someone even thought of putting a fuel cell in a plane. So there is enormous excitement.
So how does a fuel cell work in a fuel cell engine? Firstly, there is an on-board hydrogen storage system; or alternatively a fuel tank and then there is a reformer/gas cleaning system needed to produce pure hydrogen. That hydrogen then goes into the fuel cell stack, along with air, to be converted into electricity. Water is the by-product. Then some of the electricity is stored in a battery/super-capacitor system; while the rest goes through a power control unit to the electric motor. This energy system powers the car and looks fairly simple.
In this slide, the real system can be seen as Toyota’s PEM fuel cell drive train. Here a methanol fuel tank, fuel reformer/gas cleaning system provides hydrogen to the fuel cell stack that is then connected to a control unit and storage system.
So why is there so much excitement over fuel cells?
First, as you heard this morning, fuel cells can be designed to give really high efficiency compared with the internal combustion engine. In fact, we might expect the improvement in efficiency rates to be a factor of 2 to 2.5 as compared with internal combustion engines.
If we look at overall well-to-wheel efficiency with this data sourced from Toyota, then we can see that the current fuel cell vehicle with hybrid control provides something like a net efficiency of 29 per cent. This estimate is based on natural gas, with quite high compression losses. In future, once we obtain good storage technologies, the losses will decrease and fuel cell efficiency will increase. Now comparing this with renewable energy options, it can be seen that well-to-wheel efficiency of fuel cells can increase even further.
Fuel cells will extend the life of dwindling fuel reserves and provide an alternative. The real problem with liquid fuel reserves is that there are many forecasts which indicate that they are going to peak in the next 10 to 20 years. I have come across only a few data which indicate that they are going to peak in 2035 or 2040. Most predictions indicate that this is much nearer than many people think, and so the situation is in fact, quite serious.
If we look at well-to-wheel carbon dioxide emission levels, and compare petrol vehicles with current generation Toyota fuel cell vehicles that run on natural gas (those that use hydrogen from natural gas) then we see a significant reduction in emissions. In the future, once we move to renewable energy, obviously there is going to be a substantial reduction in CO2 emission levels. Fuel cells will contribute to a reduction in greenhouse gases, and help stabilise global warming. Currently, the surface temperature of the Earth is rising; and here are two different graphs that show a clear increase in the surface temperature of the Earth.
If we look at total emissions generated from passenger cars per annum, then apart from carbon dioxide, the other emissions such as NOx, SOx, carbon monoxide, and especially particulate levels (and note that we compare the petrol car with the natural gas fuel cell car) indicate an enormous reduction. Specifically, the particulate levels are reduced substantially. This is important because fine particulate emissions are a real public health concern. These particles move into people’s lungs and so on, and cause serious damage.
The widespread use of hydrogen fuel cell vehicles will reduce pollution, because fuel cells themselves generate no pollution or greenhouse gases at end use sites. Here you see some of quite heavily polluted cities. We need to clean them up. There is quite a bit of excitement about hydrogen fuelling stations. Some people think that we need to build a brick or concrete wall to contain a hydrogen station. In fact, that is not the case. There are already stations like these located in urban areas in Japan. Indeed at the moment you can just go in there, as you would go into a normal petrol station, and fill up your car.
Currently, there is talk about California planning a hydrogen highway; and Canada is looking at a similar hydrogen highway for the Winter Olympics in 2010 to link Vancouver airport with Whistler. For these applications, it is possible to fill your car in three or four minutes.
There are also mobile hydrogen filling stations, and then various hydrogen dispensers which measure hydrogen in kilograms rather than in litres.
You can see here in this slide a liquid hydrogen fuelling station in Japan.
There are lots of stationary applications for fuel cells: emergency lighting, standby power, premium power, residential power, and remote area power supplies. These systems can be put anywhere people wish, indeed wherever people need energy. Other applications include large buildings, hospitals, and office blocks. Let’s now examine how a stationary fuel cell power plant looks. It is different from the fuel cell plant for a car, but it is much simpler than the conventional coal-fired or natural gas-fired power plant.
Firstly, there is a fuel processing/cleaning unit that generates hydrogen as fuel. This is fed into the fuel cell stack along with air, and then this generates DC power which is connected to a power conditioning unit. Furthermore, for stationary applications, it is possible to recover the heat generated inside the fuel cell stacks. Thus these systems are very good for co-generation of both heat and electricity. It has been estimated that system efficiencies (with heat recovery) of up to 85 to 90 per cent can be obtained using fuel cell units.
This slide shows a number of smaller units (generating 1–10kW) that have been developed for residential-type applications by various companies.
This slide shows some more solid oxide fuel cells that operate at around the 1-5kW range.
This slide illustrates some medium-sized plants. One is the 250kW Westinghouse combined cycle system, with a net efficiency of around 80 per cent (53 per cent electric or 27 per cent thermal). The Ballard unit generates 250kW, and you can also see the ONSI now United Technology Fuel Cells unit.
In this slide you can see the Siemens Westinghouse 100kW solid oxide fuel cell system, which has run for over 20,000 hours in Europe – the Netherlands and Germany – with a total system efficiency of 76 per cent.
This slide illustrates Fuel Cell Energy’s 300 kWe molten carbonate fuel cell system. A number of these units are currently being installed for demonstration purposes. Shown in this slide are some units which are a bit larger. Specifically, these include a 2 MWe molten carbonate fuel cell unit, a 1 MWe phosphoric acid, and a 1 MWe molten carbonate fuel cell in Japan.
Eventually, people will have fuel cells for residential use. This can be achieved by supplying natural gas to a house and then meeting all heating, cooling and electrical demands of the house with a fuel cell unit. A 1kW unit will generate 150 to 200 litres of hot water per day at 60 to 70°C if these employ a polymer electrolyte membrane fuel cell.
Fuel cells for use as portable power now promise six to seven times longer operational and standby time. They also have much shorter recharging times which occur in seconds as opposed to hours for batteries. The balance of a plant is bare minimum. Lifetime of two to three times that of rechargeable batteries, and significantly higher power density are expected. So consequently, there is enormous excitement about using fuel cells for portable power applications: laptops, PDAs, mobile phones, cameras and so on.
This slide summarises the applications of fuel cells. I won’t spend too much time on this, but you might look at some of the small and portable fuel cell applications for defence and space, and also for traffic signals, electronic tolls, traffic cameras, traffic volume sensors and counters, as well as variable message or emergency signs. You can really put them anywhere. Such fuel cells range in size from one watt to maybe up to a few hundred watts. There are lots of advantages of fuel cell technology:
 (Click on image for a larger version) There are number of different types of fuel cells, to further confuse the situation. There is not just one type of fuel cell that is available. For instance, alkaline fuel cells have been used in space missions. Then there is the direct methanol fuel cell, the polymer electrolyte membrane fuel cell, the phosphoric acid fuel cell, the molten carbonate fuel cell, the solid oxide fuel cell, and then direct or indirect methanol/ethanol PEM fuel cells.
So it can be seen that the names of fuel cells stem from the electrolyte or fuel type that is used in the fuel cell. We will not go through all the fuel cell systems here. Now, you saw a diagram that was very similar to the one this morning, when Dr Crabtree explained how the fuel cell works. So I won’t go through this in detail, except to say that fuel cells are basically electrochemical devices that will generate electricity continuously without combustion, and by harnessing the energy created when hydrogen and oxygen are electrochemically combined. So it is an electrochemical reaction that combines the hydrogen and oxygen.
Typically, the open circuit voltage of the cell is given by the free energy of the fuel oxidation reaction, and you also saw the equation for this, earlier this morning.
Now we will compare fuel cells with batteries. Fuel cells combine the clean and quite efficient attributes of a battery with the refuelling capability of an internal combustion engine. So fuel cells are energy conversion devices, they are not energy storage devices. They are like power plants – if you keep on feeding in fuel, you keep on generating electricity. Typically, you have 0.6 or 0.7 volts per cell output when the cell is loaded. Open circuit voltage is typically around 1.0-1.1 volts. In reality, fuel cell structure is quite complex. There is an electrolyte which has to meet certain requirements. It has to be impervious to gases, allowing only ionic transport to go through the system. Then there is a catalyst layer that facilitates the fuel oxidation reaction. Fuel cells have different catalysts on either side of the unit, and they need to be distributed in the right places. Then there is a diffusion layer that is located at the back of the catalyst. This has to be conductive, porous, and allow all the reactants and products to diffuse in and out of the system. Then there is a bipolar plate which connects one cell to the next cell, to increase the voltage and distributes reactants to the reaction sites and remove products. So, on this diagram, the current flows downward and the reactants flow in and out of the cell. At this point it is also necessary to remove the heat. For instance, if we have 1kW generated as electric output, then we have a similar amount of heat in the fuel cell that needs to be removed. Keeping this fuel cell structure in mind, what can be seen over here at the left-hand side of the diagram? The five layers constitute what we call a membrane electrode assembly (this term is quite often used in reference to fuel cells); and then there is a bipolar plate to connect one cell to the next cell. As this animation runs, you can see the fuel cell stack with its bipolar plate, and then the five components of the membrane electrode assembly. The catalysts are deployed on the inside of the diffusion layer. So this is how the fuel cell works. The arrow on the right shows where hydrogen is supplied. (All the components are physically joined together),.
Then you stack a number of these cells, and so the power output increases as you increase the number of cells.
There are different cell reactions in different types of fuel cells, and for systems operating at different temperatures. Again I won’t go through all the reactions, but will just give you an overview. Basically, different fuel cells use different materials, they have different cell reactions and so on, and different operating temperatures. Three fuel cells have major commercial potential. The first type is the solid oxide fuel cell, which operates at around 800 to 1000°C. However this has a long start-up time and shut-down time. This type of fuel cell is good for base load applications. It can be used for residential applications, but only when there is a requirement for high thermal load and low electric load. The second type of fuel cell is the molten carbonate fuel cell, which operates at a somewhat lower temperature. Again it has a long start-up and shut-down time. This type is good for base load stationary applications operating in the hundreds of kilowatts to low megawatts.
The third type of fuel cell is the polymer electrolyte membrane fuel cell, with an operating temperature of around 80°C at this stage of development. It has a very fast start-up and shut-down time and can take fluctuating loads quite rapidly. It can be designed to generate a few watts or up to hundreds of kilowatts. The applications for this type of fuel cell include all types of transport applications, small stationary applications such as backup power and premium power; and it can be used as a UPS or for residential needs or can provide power to remote areas. Furthermore, micro fuel cells can be used, obviously as portable power packs. I just want to spend a few moments on this slide about fuel cell technology status. At this stage, today, there are about 600 to 700 fuel cell vehicles on the road that have been produced by all manufacturers. By comparison, there are 45 million to 50 million vehicles produced around the globe per annum. So there is a large gap. There are about 160 hydrogen refuelling stations worldwide, but in Australia alone there are about 8,000 petrol stations. For stationary uses, there is less than 100 MW of total installation capacity, and most of the units are phosphoric acid fuel cells. There are some molten carbonate fuel cells and PEM fuel cells. However in Australia alone there is a 40,000 MW power generation capacity. The R & D expenditure is about one billion-dollars per annum and shuld be much higher. Typical current sales figures are around half a billion dollars per annum.
Still, there is enormous optimism. You might like to read the print at the top of this slide with the statements below. President Bush has said that ‘the first car driven by a child born today could be powered by hydrogen, and pollution-free’. Ford has said that ‘fuel cell vehicles could be commercially viable by 2015’. General Motors has forecast one million cars by 2020. There are many challenges for fuel cell technology. The infrastructure for hydrogen is non-existent – it will cost hundreds of billions of dollars to put such hydrogen infrastructure in place. Codes and standards for hydrogen are just beginning to be developed, and there is a long way to go. In terms of hydrogen storage technology for transport, as we heard this morning, we are too far away from meeting target specifications. Then fuel cell costs are too high. We can tolerate something like A$50 per kilowatt stack for transport, and about A$1,500 per kilowatt for the system for stationary applications. However these costs are already an order of magnitude lower than the cost of producing fuel cells today, so we need an order of magnitude reduction in fuel cell costs to achieve these targets, and a factor of 2 to 3 reduction in hydrogen production costs. Fuel cell life times are also too short. We need 4,000 to 5,000 hours of accumulated time for transport applications, but only about 2,000 hours have been demonstrated so far. It must also be recognised that moving from 2,000 to 4,000 hours is not going to be so simple - it is going to take a long time. For stationary applications, 50,000 hours are required. A plant life time of over 40,000 hours has been demonstrated for phosphoric acid fuel cell technology, however, this technology will not be with us for too long. Other technologies have much shorter life times. There are fuel cell challenges related to fuel cell stack cost and reliability, and this in turn relates to materials, fabrication technology and scale-up. So there are an enormous number of challenges.Each type of material has to meet many different specifications and requirements.
There is also the need to design an interface at the nanoscale. In these models, the catalyst distribution is quite critical and the electrode reactions occur in the space of a few Ångstrom (10–10 metres),very close to the interface between gas, the electrode and the electrolyte. Then, in order to reduce the cost of catalyst it will be necessary to use either cheap materials or minimise the amount of catalyst and ensure that it is distributed where it is most needed.
In summary, in the future there will be a need for an interface design at the nanoscale for maximum catalyst utilisation and durability. We will need good catalysts with high tolerance to impurities. We will also need stable membranes with a long lifetime, and the use of a holistic materials approach. It will also be important to have good predictive models to predict fuel cell life times, and so on. This slide summarises the CSIRO hydrogen fuel cell R&D program. In the past we set up a company called Ceramic Fuel Cells, which was a spin off from our research work in 1992 and now has about 100 employees. It is listed on the Stock Exchange.
Some of our current projects at CSIRO now relate to the polymer electrolyte membrane fuel cell that is for small-scale applications; micro fuel cells; distributed hydrogen generation that is linked to renewable energy for water electrolysis; and metal and ceramic membranes for hydrogen separation. So there is enormous emphasis on materials (membranes, electrode, interconnect and coatings), interface design, stacking technology, prototype development and demonstration. In conclusion then, we need a shift to a hydrogen economy. Sustainable energy is essential for the survival of our future generations. Fuel cells offer the best and optimal solution, as you have seen this morning. Fuel cells are considered the key technology for transport applications, to reduce our reliance on oil. However, we need enormous effort to reduce costs and increase fuel cell reliability, and commercialisation is quite a long way away. It is certainly not going to occur in the next 5 or 10 years. I will now just leave you with a very good quote which you may have time to read while I am answering questions. If you read the statement at the top of the slide, and relate it to what the company says at the bottom – I won’t name the company – you will find it quite interesting. It just shows you how much overselling of fuel cell technology has been occurring at this stage. On the one hand, the company says that they have ‘designed and built custom fuel cell power systems to the technical demanding requirements of world renowned customers’. Then, on the other hand, they talk about a 300-hour life time, with 5kW power over 300 hours. The target required is actually 50,000 hours. So this is the problem.
Thank you. Discussion Ken Shortman, Melbourne – I have just a basic question. Is the ratio between the electrical and heat output of a fuel cell something fundamental, or does it vary with the different types of fuel cells? Is it something you can alter? Sukhvinder Badwal – It varies with different types of fuel cells. It depends very much on the type of material you use. In the ideal situation, you need to get maximum electric efficiency from a fuel cell by optimising materials and reducing the thermal load. I was giving the example of a current status polymer electrolyte membrane fuel cell system. Peter Pockley – Could you perhaps give an idea of investment in these programs - in your presentation and spoke of CSIRO’s involvement. What is the extent of CSIRO’s funding of fuel cells at the moment? What trends have there been in reaching this point? Sukhvinder Badwal – At this stage, CSIRO does not invest an enormous amount of money in fuel cell technology. We will develop the technology to a certain stage and then it will be taken to industry, like Ceramic Fuel Cells. CSIRO would have invested may be a couple of million dollars up to the time the technology was given to Ceramic Fuel Cells, but the company has invested $150 million or more since then. A similar sort of thing would occur with other types of fuel cell technology. We are investing something like $2 million per annum in fuel cell and related technologies, but we are also talking with a number of investors. So there is strong potential for a spin-off to be formed. Then this company would invest an enormous amount of money, perhaps tens of millions of dollars per annum. Robert Hunter – Obviously, the big push on the transport side is the motorcar industry. I just wonder to what extent the competition between manufacturers is helpful. Is there any notion of consortia getting together, or does, for example, the patent literature give you enough information from one to the other to make this a more cooperative venture? If there are all these sorts of basic requirements before people get anywhere near what is needed, then one wonders whether the competitive procedure is the most useful way to go. Sukhvinder Badwal – In some cases there are partnerships being formed between various companies, especially if you look at the Ballard situation. Ballard Power Systems is forming an alliance with a number of transport companies; a number of transport companies are also joining up to develop the ‘Freedom Car’ project, and so on. Nevertheless, at the same time, you can read reports that General Motors – who I think were cooperating with Toyota, were working together to develop a fuel cell drive train – now I think that they have developed their own technology to a stage which they believe may be better than Toyota’s, so there is not much point in cooperating. So they are separating. I think that there should be some degree of cooperation. In terms of the patent literature, the reality of the situation is that people will file patents to cover themselves to some degree, and each company has a different patent strategy. Ballard would say, ‘We have a certain patent strategy,’ which would be totally different from CSIRO’s patent strategy, for example. Noel Hush, University of Sydney – There is quite a bit of discussion in America about methanol as the fuel for the fuel cell. It, of course, is liquid at ordinary temperature, which is a great help. Voller??OHLA® in America is, as you probably know, one of the proponents of this approach. There is a suggestion that this could be greenhouse gas-neutral, because you could produce the methanol from biomass and so on. What are your thoughts about this? Sukhvinder Badwal – There are two types of methanol fuel cell. One is a direct methanol fuel cell, where methanol actually participates in the electrochemical reaction. The other one is where you actually reform or convert methanol into hydrogen, and then feed hydrogen into the fuel cell. There is no problem with the second one. You can convert any fuel into hydrogen and then utilise this in the system. The difficulty is whether we can utilise this technology onboard a vehicle with a reformer/gas cleaning system and so on. At present, it is not possible to achieve this goal and yet for stationary applications this approach is fine. Direct methanol fuel cells are mainly considered for portable power applications, and the company that you mentioned is developing a portable power fuel cell. Yes, there is enormous potential for this approach. However, there are lots of problems associated with this technology as well. They produce carbon dioxide, so carbon dioxide has to be separated, and the electrolytes need to be continuously recycled to maintain the methanol concentration at fairly constant levels. Les Field – I think we need to curtail the discussion there. Thank you very much. Please join with me in thanking Sukhvinder for that talk.
Other speakers
Dr John Wright
Dr George Crabtree
Professor Cameron Kepert
Professor Andrew Dicks
Dr Evan Gray
Dr Ben Hankamer Dr Catherine Grégoire Padró
Professor David Trimm
Dr Wes Stein
Professor Harry Watson |
© Australian Academy of Science