Fuel Cells are at the leading edge of alternative fuel technology.
For the automotive sector, technology to date incorporates platinum catalysts and there is no substitute so far. Loadings are, however, dropping fast.
They have been a long time coming, but now the technology is well advanced and vehicles are already on the road. The latest car is competitive in the small-medium car range, in terms of both cost and performance.
By 2010, automotive platinum demand in fuel cells should be between 500,000 and 650,000 ounces per annum.
And as a foot note, a fuel cell battery for mobile phones has just been unveiled, which could raise the game considerably.
The Environmental Saviour?
An Old Invention Now Epitomising New Age Technology
The first fuel cell was developed by Sir William Grove in 1839. It took over 120 years before it started being used in what might be thought of as a “commercial” application, when fuel cells, as with so many other examples of modern technology, were put into active use in the NASA space programme. They were employed as a compact and emission-free means of developing power. They are still used in space today, providing electricity and water for the space shuttle.
What Is A Fuel Cell?
A fuel cell works on the principle of hydrolysis, whereby hydrogen and oxygen are combined, with the aid of a catalyst, to produce water with energy as a by-product. Chemists among our readers will no doubt think back nostalgically to their schooldays on the sight of the chemical equation:
2H2 + O2 ⇒2H2O
This is the key to the fuel cell’s modern-day applications. Fossil fuel technology brings with it noxious emissions and the ever-increasing need to control atmospheric pollutants is hastening the search for alternative, environmentally-friendly, means of power generation. A variety of fuel cells have been developed, or are in the process of development, the most important of which are listed briefly below. Their applications vary according to the nature of the cell and while we list all of those currently in the public domain, the cells in which there has to date been most public interest, which would seem to be the closest to “mass usage” and which also involve the use of platinum as the catalyst, are those cells that are viable for transport purposes. These use the Proton Exchange Membrane (PEM) technology, using a solid electrolyte and may also be known as the solid polymer fuel cell. Later in the note we outline how this end-use has developed and look at the current state of play with respect to the transport industry as a whole, with a view to attempting to quantify the future impact on platinum demand.
Table 1. Major varieties of fuel cell
||Approx Operating temperature
|Proton Exchange Membrane (PEM)
||Transport, small-scale domestic goods
|40-50% (cf. <20% for internal combustion)
||Coming into commercial use
|Phosphoric acid (PAFC)
||Buildings, back-up generators.
Also heavy vehicles
||In use; 85% efficiency with cogeneration
|Molten Carbonate (MCFC)
||Molten Carbonate salt
||Can use unreformed fuel.
Slow delivery and response time
|Solid oxide (SOFC)
||Large, high-power utilities
||Two 25 kW cells in use; 100 kW underway
||Small scale military, aerospace
||Up to 70%
||First in use; NASA programme.
High purity feedstock.
||Very early stages.
Water feedstock; recycles water by product
Before we move on to look at the PEM cells which, at present, look like being the predominant alternative fuel source for transport vehicles, it is perhaps worth a brief look at the others in order to gain some perspective of how and where they may be used.
One drawback with fuel cell technology is that is still expensive, although considerable effort is being put into reducing costs, both for stationary and non-stationary applications. Prices are variable, although it has been reported that a power plant was offered in 1998 at roughly US$3,000 per kW, by comparison with US$1,000 per kW for existing power supply systems. While information on the development of the technology is to a large extent proprietary, it is expected that this can, at least, be halved. We must assume that costs in early 2000 are en route for the predicted US$1,500/kW level, but have been unable to ascertain prevailing rates. We shall see below how the costs for automotive cells are falling.
The PAFC Cell
The phosphoric acid fuel cell was already in use in prototype by the mid-1980s, when cells delivering as much as 11MW were installed in the US and Japan, with particular emphasis on use as back-up generators for hospitals and so forth. Commercial production was underway by the start of the 1990s and several hundred are now in use. The Japanese are particularly interested in the development of this technology due both to considerations of space and heavy Japanese dependence on oil imports. The same considerations underscore the aggressive work by Japanese car companies on developing the use of automotive fuel cells. The PAFC cell is the most commercially attractive for small-scale operators.
The molten carbonate and solid oxide cells take a long time to build up to critical operating rates, and the molten carbonate cell can take time also to respond to changes in output requirements. Because of their high operating temperature, each is capable of operating on unreformed fuel and the solid oxide cell can utilise comparatively impure fuels. The primary purposes for these cells would be in utility applications for power provision.
The alkaline cell, although the first in “commercial” use, is one of the more exotic in that it is generally prohibitively expensive for large-scale applications as it needs high purity hydrogen and oxygen feedstock, although efforts are being made to enhance its competitiveness.
The PEM Cell
The Legislative Background
The United States, and especially California, is in the forefront of emission control legislation, with a view particularly to unburned hydrocarbons, nitrogen oxides and carbon monoxides (all of which are currently treated via PGM-containing emission control catalysts), plus carbon dioxide (which is not). The current requirements with respect to motor vehicles, as stipulated by the Environmental Protection Agency (EPA) involve two separate programmes; the Californian LEV II programme (Low Emission Vehicles) and the NLEV Programme (National Low Emission Vehicles), of which the former is the more stringent. Under the LEV programme, which applies to California and, by volition, New York, Massachusetts, Maine and possibly Vermont, 10% of all vehicles produced as of 2003 must be Equivalent Zero Emission Vehicles (EZEV). There are numerous complications to the emissions requirements, involving four other levels of emissions that are permissible in differing proportions of the car fleet, but for these purposes they may largely be ignored. The exception is the Super Low Emission Vehicles (SULEV’s), which would encompass those vehicles using fuel cells powered by organic fuel and using a reformer, since they would involve a small degree of pollution.
It is worth noting at this point that “vehicles” includes off-road vehicles such as tractors and fork-lift trucks for factory sites and therefore does not strictly imply 10% of passenger vehicles. (Equally, the attraction of fuel cells for stationary environments is the comparative lack of CO2 emissions). What is of clear importance is that emissions legislation can only become increasingly stringent and that the technology provided by fuel cells is, at present, at the forefront of alternative solutions to the prevailing fossil-fuel driven technology. As far as the automotive sector is concerned there is an alternative in the form of battery cars, typically using nickel-hydride batteries, but these require charging on a regular basis, which limits their scope. Fuel cells will run for as long as there is a fuel source – see below. While there are vehicles in the pipeline that could prove both viable and popular (including those that involve both batteries and fuel cells), the proven performance to date would point firmly in the direction of the fuel cell as the market leader.
Table 1 shows the comparative efficiency in energy delivery between a fuel cell engine and that from an internal combustion system. We show a more detailed comparison in the final section of the note, when looking at specifications and the implications for platinum usage. Suffice to say at this juncture that the performance specifications and the low fuel consumption and mileage range achieved between fuelling almost sounds too good to be true!
In a typical PEM cell, each electrode is a carbon-based substrate coated with a platinum catalyst which, at the anode, causes hydrogen to split into protons and electrons, its two major component parts. The protons carry a positive charge and move across the polymer electrolyte towards the cathode, while the electrons take a different route through the external electrical circuit set up between the anode and the cathode. Once in the cathode, the protons, electrons and oxygen react to form water and heat. These cells can be combined into a “stack” to deliver sufficient energy to the vehicle.
The feedstock oxygen for a PEM cell can obviously be taken from the air. Hydrogen can be delivered in its pure form, but an increasing number of cell designs incorporate a reformer, which allows the use of methanol, ethanol, petroleum or natural gas. (Alkaline fuel cells cannot use organically derived feedstock since these generate carbon dioxide, which reacts with the catalyst). There are additional alternatives for fuel cells operating in a stationary environment, such as gas taken from biomass or wastewater treatment plants. Vehicles operating from a pure hydrogen supply will qualify as “ZEV’s” (zero emission vehicles) under the US legislation, while those using hydrocarbons are likely to emit a small degree of CO2 and may thus have to qualify as one of the intermediary vehicles, the “SULEV” or Super Low Emission Vehicles. The emission level is almost negligible, however, and they may yet pass the requirements for ZEV’s.
It seems highly likely that the majority of fuel cells on board cars will operate from a liquid fuel and incorporate a reformer, if only because the energy density is higher from such a feedstock and the mechanics of delivering the fuel to the vehicle are easier. Johnson Matthey, is at the cutting edge of this technology and at the end of last year was awarded the prestigious Italgas Prize for its work on developing a natural gas fuel processor (the “Hotspot”) which can be used to generate hydrogen for fuel cells for cars as well as for small units for non-automotive domestic use.
Hydrogen Is Less Risky Than Has Been Feared
In addition, it has been argued against fuel cells that carrying a supply of hydrogen in an automotive vehicle involves an unacceptable degree of risk, and also that the cost of conversion of fuel supply sources would be prohibitive. In response to the former point, it is reported that Ford has conducted safety tests for the US Department of Energy, which suggests that the technologies under development for hydrogen storage in a fuel cell are safer than those in existence for storing gasoline. While the most likely method of powering the cells will be to carry a liquid fuel and use a reformer; some of the comments in the report make interesting reading:
The report says, inter alia, that “a fuel cell electric vehicle would carry about 0.8GJ of hydrogen energy in a four passenger car, or 0.2GJ per passenger… stored in fibre-composite tanks; could withstand 50-mph head-on collisions, engulfment by a diesel fire, and pressures at least 2.25 times design pressure without rupture”. In addition: “each class of tank is also subjected to gunfire and must not explode but leak only through the bullet hole”. Some test!
The company also submits that on collision, since hydrogen disperses far more rapidly than gasoline, the potential for fire (if the tank is ruptured) is considerably reduced and that a fuel cell vehicle would be carrying 60% less total energy than its petrol-driven equivalent, also reducing the fire risk. While liquid fuels seem to be the more likely, it would appear that pure hydrogen, if deemed viable, would pass a number of safety tests. Whether the risks of pure hydrogen leaking into the atmosphere on impact would be “politically acceptable” is perhaps a topic for debate.
Cost vs Environment
As far as delivery of the fuel, be it pure hydrogen or a liquid, is concerned, the consultancy Arthur D. Little has estimated that it would cost between US$50-100Bn to provide the infrastructure for converting 10% of the US car fleet to an alternative fuel. Equally, the US Department of Energy has estimated that conversion of 10% of the US fleet from petrol to fuel cells would reduce oil imports by 800,000bbl/day. At an oil price of, say, $20/bbl (against today’s level in excess of $25), that equates to almost US$6Bn per annum. Thus the saving on imports would take 10-20 years to cover the cost of conversion. This does not, on these parameters alone, look like a commercial proposition, but the Dept. of Energy also points out that a 10% conversion of the US fleet would cut air pollutants by one million tonnes per annum and that 60Mtpa of CO2 would be prevented.
The power of the oil lobby must not be underestimated and there is a feeling in some quarters that their market share may be under threat. The current ecological position is such, however, that the environment may well take the upper hand – as it did in the face of protest from the oil industry in the 1970s and 1980s over the elimination of lead in petrol in order to accommodate emission control catalysts. (It is not often recognised that the introduction of autocatalysts was the precursor to stripping lead from petrol; the “lead-free” issue grew up on the back thereof). It is of enormous potential significance that Exxon has decided upon entering partnerships with General Motors and Toyota with respect to developing fuel cell powered cars and the consortium is looking at developing cleaner petrol.
Moves Towards Fuel Cell Usage
Clearly the jury is out as far as a large-scale conversion is concerned, but the activity of the major car companies in company with organisations such as Ballard and Johnson Matthey is leading the way forward. It was originally envisaged that phosphoric acid fuel cells would be viable for automotive technology, but their power density is way too low and a cell large enough to power a car would have taken up most of the back seat! The PEM technology is a considerable improvement and hard work by the major car companies in alliance, largely, with the technology companies, has brought us to the stage that prototypes are now on show at car shows.
The Major Players
One of the most significant recent developments in the evolution of the automotive fuel cell was the formation of dbb, a joint venture between DaimlerChrysler, Ford and Ballard Power Systems. (In addition DaimlerChrysler and Ford now hold 20% and 15% of Ballard respectively). The alliance is working well, aided by a collaboration agreement signed in March 1999 with Johnson Matthey under which JM is dbb’s exclusive development partner for catalysts for hydrogen purification, along with several other relevant catalytic components. JM did already have a long term agreement with Ballard for the joint development and supply of catalysts and this link forges even closer ties between the two. Meanwhile there are any number of other companies looking at developing the technology; DaimlerChrysler is using fuel cells in the new A-class model, while Ballard has recently added Nissan to its ever-lengthening list of clients which already includes Honda, VW and Volvo.
A small number of buses are already on the road fitted with fuel cells and Ballard expects that by 2003 there will be 55 such vehicles on the streets of California. As far as cars are concerned it looks as if the first company into commercial-scale production will be DaimlerChrysler, which expects to have a version of its NECAR on sale in 2004, right on time to comply with the environmental legislation.
The latest fuel cell system to be shown to the public by DaimlerChrysler is the Ballard Mark 900, which delivers 75kW on methanol and 80kW on hydrogen. It measures roughly one metre in length by 30cm in height and 45cm in width. This gives it a volume of roughly 77 litres – easily fitting into a standard engine compartment of today. This is roughly half the volume of the Mark 700 module, with a power density some 30% higher at 1.04kW/litre on hydrogen and 0.97kW/litre on reformate. The system on which the module is based is also reconfigurable for any car, according to size and power requirements. The module is ready for the production line, and all that may need to be changed is the nature of the seal of the cell, or the platinum loading.
Currently Viable For Small-Medium Cars
The typical fuel cell stack system designed for automotive use delivers 50-75kW; the latest system demonstrated by DaimlerChrysler delivers 75-80kW (see above), from a fuel cell stack with a power density of roughly 1kW per litre (a considerable improvement on the original designs). As 1kW is equivalent to roughly 1.34 horsepower, an 80kW stack delivers the equivalent of 107 horsepower. Most US vehicles are between 60 and 200hp, so this would place in the existing systems in the small-medium range. Top of the range equivalent would require a 150kW stack.
Comparison With Alternative Technology
The higher efficiency of a fuel cell by comparison with the internal combustion engine means that fuel consumption in general in a car powered by fuel cells is roughly one-third that of an equivalent petrol-powered car. Other specifications are also competitive. For example: The DaimlerChrysler NECAR 4 can attain 90mph, and has a range of almost 280 miles before it requires refuelling. The car can seat five with additional luggage space, as the fuel cell is mounted in the floor of the vehicle. Battery-powered vehicles, such as the GM EV1 tend to need recharging after only 90 miles and have severely restricted space (only two passengers). The latter are useful for small journeys, though, and are increasingly in use in fleets of postal delivery trucks, for example.
Platinum: Vital Component, But Loadings Are Falling Fast
The platinum loading is absolutely key. While the size of a typical fuel cell has been drastically reduced, so has its cost, but it could well need to be reduced again. In 1994, a typical fuel cell stack for a car would have cost in the region of US$30,000. By mid-1999 this was down to roughly $500, reflecting a drop by a factor of 60. Indeed Ballard, in its 1999 Annual Information Form, reports as follows: “In 1999, the Corporation demonstrated a 75% reduction in catalyst requirements and in 1995 a further reduction of 57% was demonstrated in 1997 the Corporation began to implement the low catalyst loading technology of 1995 / 96 in production fuel cells.
In 1998 the Corporation demonstrated a further decrease of 70% in catalyst loadings. The implication is that the 1998 technology has yet to be implemented in the cells in active use, but that it will not be long in coming. The cumulative effect of these reductions is a net drop, from the loading levels of the early 1990s, of 97%.
Much of the cost of a PEM fuel cell is the platinum, which in 1999 averaged $377.25/ounce, suggesting a loading in the latter part of the year of perhaps an ounce or so per stack. Latest published estimates from Johnson Matthey suggest that by the time fuel cell cars are in commercial-scale production, each will require between eight and ten grams of platinum. As we write this, platinum is in the mid $500 - $600 range, distorted by disruptions to Russian supplies. At our forecast level of $420/ounce as a long-term average, the simple spot price of a platinum loading would thus drop to $100. Certainly Ford and General Motors are of the view that it is possible to produce a fuel cell-powered car of the same cost as a model powered by a conventional internal combustion engine.
In 1999, 8.7M passenger cars were sold in the US. In order to derive an order of magnitude, we have assumed that by 2010 most US states will have followed California’s lead in terms of emissions and that 10% of the national fleet should be operating on fuel cells. Assuming modest growth in car production, but holding the rate of fuel cell usage at 10% only, fuel cell-powered vehicles in the US would be consuming between 280,000 and 350,000 ounces per annum. With Europe and Japan following suit, we can probably double this for worldwide application. This would be offset by the loss of perhaps 60,000 ounces annually from the lack of emission control catalyst fitments.
A Half Million Ounce Market For Platinum – Minimum
This must be viewed as a low case in terms of the number of cells being used; but it may be high in terms of platinum catalyst loadings. Great strides have already been made towards minimising platinum usage, but there is usually room in these designs for further stringency. We would expect that by 2010 the fuel cell industry should be adding at least 500,000 ounces of annual platinum consumption to the market, with the scope for further growth.