Rolls-Royce drives up the pressure on SOFC costs1 October 2003
Rolls-Royce's fuel cell development activities took a significant step forward in September with the announcement of European Commission (EC) funding for a further Rolls-Royce led programme on SOFC (solid oxide fuel cell) research. The new programme is concerned with the development of pressurised SOFC technology. It is funded under EC Framework 5 and will build on two previous EC Framework 5 programmes due to conclude in 2003, MF-SOFC (stack development) and IM-SOFC-GT (integrated modelling study of fuel cell/gas turbine (FC/GT) hybrids).
The total value of the new programme called PIP-SOFC (Pressurisation of Integrated Planar - SOFC) is 310 million, including contributions from the EC, Rolls-Royce and its European partners, the University of Genoa, Gaz de France, Morgan Crucible and Risø National Laboratory.
The aim of PIP-SOFC is to develop and demonstrate a 10 kW SOFC stack basic building block with integral internal reforming. This block is envisaged as the basic module for FC/GT hybrids of various sizes.
SOFC cost reduction
For the last ten years or so Rolls-Royce has concentrated its efforts on SOFC technology, with UK stack and systems development efforts now combined into Rolls-Royce Fuel Cell Systems Ltd (RRFCS). The first product is expected to be a 1 MW pressurised hybrid system fuelled by natural gas for stationary power generation applications. This initial RRFCS product will comprise four complete 250 kW systems. Other applications are envisaged at a later stage.
A key advantage of the SOFC is fuel flexibility in that it requires no hydrogen infrastructure as ultra clean hydrogen is not needed. The SOFC operates on CO and the high temperature is ideal for internal reforming of hydrocarbon fuels. A wide range of fuels are possible, including hydrocarbons from Fischer-Tropsch processes, biofuels and waste gases.
The challenge for SOFC designers is to get the costs down so the technology can compete with alternatives such as diesel reciprocating engines and gas turbines. Cost reduction has been central to the Rolls-Royce SOFC development effort from the outset. In particular the use of exotic materials ("raiding the periodic table", as Gerry Agnew of RRFCS calls it) has been avoided, power density has been increased through good packing and the manufacturing processes used are low cost and highly scaleable, eg screen printing and extrusion.
A key element of the Rolls-Royce approach to maximising cell life is to make cells cheap so can have a lot of them and therefore low current density.
Low cost materials are used for most of the fuel cell mass. For example, and perhaps most striking, the support tubes, which dominate the stack weight, are made from commercial grade material similar to that used in the manufacture of bathroom tiles.
Rolls-Royce has pioneered the integrated-planar (IP) stack design, which aims to combine the benefits of planar and of tubular SOFC stack designs. The basic building block (or module) consists of series connected cells on a flat ceramic tube. The tubular overall geometry minimises sealing requirements, while the planar surfaces simplify manufacture and allow dense packing of the fuel cell active area.
The Rolls-Royce approach is to have a standard sized basic fuel cell from which power plant systems of various sizes can be built up.
Individual cells deposited on the surface of the ceramic flat tube are small in area and connected in series. This results in low current and high voltages, leading to the same savings in materials as achieved in high voltage transmission lines.
The ceramic tubes therefore have a very low inventory of high value materials. Indeed, the high value materials are confined to very thin layers deposited on to the low cost ceramic support tubes. All of the manufacturing steps need only conventional low cost manufacturing methods and novel expensive high-vacuum based methods are completely avoided.
The incoming natural gas is mixed with steam to reform it to a hydrogen and carbon monoxide rich mixture. The fuel cell stack converts most of the reformed fuel to electricity at high efficiency, with only about one third of the energy appearing as heat. This heat must be dissipated to protect the stack. Reforming is an endothermic reaction, taking place near stack temperature and absorbing a significant portion of this heat.
The Rolls-Royce IP-SOFC stack uses a mixture of indirect internal reforming (IIR) and direct internal reforming (DIR). Other fuel cell makers have shown that DIR of natural gas is possible in SOFCs, but in practice there are serious problems such as large thermal gradients which can damage cells and anode deactivation resulting from carbon formation. Carrying out IIR in a dedicated reformer can get round these problems, but the size of the reformer needed for 100% conversion, and therefore the costs, are prohibitively high.
The Rolls-Royce anode's tolerance of hydrocarbons allows a combination of IIR and DIR to be used, with a large fraction of fuel reformed in the reformer unit and the remaining few per cent converted on the anode. The result is once again significant cost reduction, says Rolls-Royce.
Eliminating heat exchangers
Minimising costs has also been a major driver in the Rolls-Royce approach to designing an integrated SOFC-based hybrid power plant. In particular the company is pursuing the concept of a pressurised SOFC to reduce the need for expensive heat exchangers.
The active surfaces of the fuel cells require an oxidising atmosphere everywhere on the air side, and a reducing atmosphere everywhere on the fuel side to avoid chemical damage. Consequently, some surplus oxidant and fuel pass into the stack exhaust where they are combusted, releasing additional heat. To achieve maximum efficiency this heat must be used to generate additional power via a bottoming cycle. There are many ways the SOFC stack and a Brayton cycle (gas turbine) can be integrated to realise this efficiency gain. The challenge, as Rolls-Royce sees it, is to devise a system that is "rugged, affordable and operable."
The stack is most efficient when operated at the highest possible temperature and performs best when the temperature is uniform. This means it must be fed with hot air. The conventional approach is to use a recuperator to heat the fresh incoming air by heat exchange with the stack exhaust air.
But according to Rolls-Royce there are serious problems with this approach because the top temperature in the recuperator is too high for affordable metals to be used and the heat exchange effectiveness needed is too much for compact ceramic units. There is also a low limit on acceptable cycle pressure ratio because the turbine exit temperature must be higher than the required stack entry temperature and the turbine entry temperature is limited to that acceptable by the stack exhaust. In other SOFC hybrid systems a pressure ratio of about 4 has been considered.
The Rolls-Royce approach recognises that high Brayton cycle efficiency can also be achieved by higher pressure ratio as well as recuperation. Higher pressure ratio causes the turbine to recover more exhaust heat and return it through the compressors rather than a heat exchanger. The added turbomachinery is cheaper and more rugged than heat exchangers, says Rolls-Royce.
To heat the incoming ambient air sufficiently to feed the stack purely by compression would require a pressure ratio of over 20. But this simple approach does not work when reforming takes place in the same zone as the stack because the working temperature is not high enough to prevent the high pressure from driving the reforming equilibrium backwards, allowing carbon-containing materials to block the active surfaces.
On the other hand to get enough reactant to the compact stack with low parasitic pressure loss, a relatively high working pressure is needed. So Rolls-Royce takes the approach that the working pressure should be the highest that is compatible with good reforming. This leads them to adopt a pressure of 7 bar.
Compression to this level raises the stack inlet air temperature to about half that required. The rest of the heating for the inlet air is done by making use of the fuel that has to be allowed to flow out of the stack to guarantee a reducing atmosphere at all points on the fuel side. It is combusted (well below NOx formation temperature) by a fraction of the stack oxidant stream exhaust and the products of combustion are then recirculated to mix with the compressor discharge air to achieve the required level of heating for the stack inlet stream. This heating technique requires a low-rise low stress impeller to drive the circulation.
The fuel-side recirculation loop provides the continuous supply of steam (reaction product) needed for reforming and ensures a much more even concentration of fuel throughout the fuel side of the stack. This maximises the fraction of fuel that can be used galvanically without incurring local depletion of the reducing atmosphere, further increasing efficiency.
Rolls-Royce estimates that one of its pressurised 250 kW SOFC units would have turbomachinery similar in size to a 50 kW microturbine.
For larger SOFC units, e fficiency improves and the compressor discharge temperature decreases, making it attractive to use a low effectiveness, moderate temperature, heat exchanger, recuperating from turbine exit to compressor discharge.
Speaking at the recent Grove Fuel Cell Conference in London, Gerry Agnew concluded that on the basis of the Rolls-Royce development work, the "low cost stack is within grasp, and improving fast" and that "system innovations at 1 MW level provide competitive installed plant costs." But he also intimated that "hybrids can get smaller as well as larger."
He suggested that the trade-off between efficiency and simplicity can be pushed further towards simplicity, particularly for pressurised small units, which could be implemented in thin wall pressure vessels.
One application of these small, say 10 kW, pressurised SOFC hybrids could be in portable power applications where there is lack of infrastructure. He showed a picture of a chainsaw to illustrate a possibility.
However, Rolls-Royce is specifically ruling out the residential sector as a market for its fuel cells because it is simply too difficult. "Domestic consumers don't go for cheaper electricity even if you offer it to them," Gerry Agnew said.