fuel cell focus
Domestic generation as baseload: the CFCL vision1 April 2007
Australia-based Ceramic Fuel Cells Ltd is developing a highly efficient SOFC for domestic microCHP, initially focused on northern European residential markets. It has recently launched what it calls a “commercial ready” metal–ceramic stack, with a very low ratio of heat to power production (0.5 to 1), designed to operate 24/7 and be of interest to utilities as a source of high efficiency baseload generation.
“We are aiming at the home but not necessarily the home-owner,” says Brendan Bilton, CEO of Ceramic Fuel Cells Ltd (CFCL) Europe. He explains that CFCL’s current vision is that its 1 kWe fuel cell module – suitably integrated with a boiler to create a domestic microCHP system – will be in year-round 24/7 continuous operation, with any surplus electricity not needed in the home provided to the grid and the distributed modules comprising a “virtual power plant”, centrally controlled, probably via the internet.
In CFCL’s view its fuel cell potentially represents a new form of high efficiency baseload power generation, which is why the company sees utility companies interested in distributed generation as its primary target audience rather than the individual home-owner.
The efficiencies being talked about are very high. As you would expect with a cogeneration scheme potential overall fuel utilisation is excellent. But it is the high electrical generation efficiency of the new, “commercial ready”, version of CFCL’s SOFC-based system – certainly over 50% and perhaps as high as 60% – that is one of the striking and distinguishing features of the CFCL business model.
If the electricity is used in the house where it is generated, with the consequent elimination of line losses – a major attraction of distributed systems – CFCL contends that its technology will deliver an efficiency “at the plug” of about 55%, much better than the best combined cycle technology (which delivers less than 50% efficiency at the plug, when transmission losses are taken into account).
The high efficiency of CFCL’s new metal-ceramic fuel cell stack means that the ratio of heat to electricity is very low, less than 0.5 kW thermal to 1 kW electrical. This is small even by fuel cell standards – with Ceres Power’s similar, metal supported, 1 kWe SOFC system, for example, producing about 1 kWt and CFCL’s own previous all-ceramic cell also achieving a ratio of about 1:1. It is particularly low by Stirling engine standards, which typically achieve a heat to power ratio of about 6-8 kWt per 1 kWe generated.
It is envisaged that the CFCL fuel cell will eventually be incorporated in a compact (wall mountable) micro CHP unit, the other main component of which will be a gas-fired condensing boiler serving both hot water and central heating needs. The relatively small amount of heat from the CFCL fuel cell would be “dumped” into the boiler to meet basic hot water needs (less than 0.5 kWt for a typical German home, see graph, below). This is why the fuel cell can be operated continuously all year round, in the baseload mode that CFCL believes utility customers will find attractive and which generates maximum value from the high electrical efficiency.
CFCL thinks that the competing distributed generation systems essentially produce too much heat, requiring them to be turned off during the summer months in Europe, for example, and undermining the economics.
Although unlikely to be sufficient for the likes of billionaire eco-warrior Al Gore (reputedly consuming about 221 000 kWh per year in his Tennessee mansion) – and probably most other American householders for that matter – the 1 kWe capacity of the CFCL system is considered to be adequate for the average European home, which typically has an annual consumption of around 4500 kWh.
The model for how the microCHP units are actually deployed commercially and installed in people’s homes may vary between markets. However the unit is likely to be owned and operated by the energy company, with the costs of the system offset by revenues from electricity fed into the grid (which could be tax exempt under microgeneration incentive schemes of the type recently proposed in the UK). In this model, the utility will probably install the microCHP unit in the home for no charge, when the home-owner signs up to a contract for bundled services (power, heat, perhaps telecommunications and other services) for say 3-5 years. The utility would bear the capital cost of the unit, and make money out of selling the unit’s power back into the grid. The utility would give the home-owner a share of this value as the “price” of the utility putting the unit in their home – say, cheaper power or ‘free’ heat, or fixed discounted prices or other ‘green’ credits. At a conceptual level, it could be quite similar to how mobile phones have been deployed – customers get the phone for ‘free’ by signing up to a contract.
Since CFCL was founded in Australia back in 1992, a joint initiative of energy and industrial companies, government bodies and the Australian national science agency CSIRO (Commonwealth Scientific and Industrial Research Organisation), the focus of the operation, not surprisingly, has changed.
The initial aim was to develop fuel cell units in the 200-400 kWe size range but over the years the target unit capacity has reduced markedly and the commercialisation strategy has evolved. It is now firmly focussed on the European microCHP sector, which Brendan Bilton describes as “the low hanging fruit” for initial deployment, with a number of European governments very much in favour of microgeneration.
Particularly bullish is the UK’s DTI, which has recently suggested that microgeneration could provide 30-40% of the UK’s electricity needs by 2050. But it is a challenging field, which has been afflicted with a number of false dawns and embarrassing failures, most recently the demise of BG’s Stirling engine based Microgen programme and before that Sulzer’s abandonment of its SOFC based Hexis technology (eventually salvaged by a foundation, which has resumed development work on the system).
CFCL has been in the fuel cell business for a relatively long time and not surprisingly its technology has evolved steadily, with concerted efforts to increase power density, raise efficiency, reduce unit size, improve reliability and lengthen operating lifetime – although the basic concept has always been SOFC. In parallel CFCL has been aiming to build demand, by demonstrating real world operating experience in field trials with potential utility customers and developing products in co-operation with partners in the domestic appliance business, as well as developing plans for large scale fuel cell manufacturing facilities and the necessary supply chain, including the making of consistently high quality ceramic powders, a critical element in SOFC production.
During 2006 and the first months of 2007 major progress on all these fronts was seen.
In March 2006 CFCL was listed on the London AIM (following an oversubscribed placement raising £37 million), in addition to its existing listing on the Australian stock exchange.
In June 2006 German utility EWE (which over the past ten years or so has trialled a range of fuel cell based microCHP systems, including two from CFCL, 33 from Sulzer Hexis and seven from Plug Power/Vaillant) placed an order with CFCL for NetGen fuel cell demonstration units.
In March 2007 CFCL further developed its relationship with EWE (one of Germany’s top five energy suppliers), with the announcement that it had signed a new three-way product development agreement with the utility and with German domestic boiler manufacturer Bruns Heiztechnik GmbH. Under the agreement CFCL will work with Bruns to develop a 1 kWe micro CHP unit to meet EWE’s specifications for the German market. The aim of the first stage, to be completed by around mid 2007, will be to build and operate an ‘alpha’ unit consisting of a NetGen unit, a Bruns condensing boiler and a thermal store integrated and operated via a single control unit. The second stage will aim to create a near-commercial beta system, comprising a single unit in which a CFCL fuel cell module is fully integrated with a Bruns boiler, without the thermal store. EWE’s participation is intended to help steer the project towards a “customer orientated and economically viable” product.
The new agreement with EWE and Bruns is similar to another product development agreement signed in December 2006 with Gaz de France and domestic boiler supplier De Dietrich Thermique to develop a microCHP unit for the French residential market, also aimed at integrating a fuel cell module with a condensing boiler. GDF has a particular interest in distributed generation, as evidenced by its leadership of the EU’s DEEP programme on decentralised energy, while De Dietrich Thermique is reportedly the largest supplier of gas heating systems to the French market, with an extensive network of installers.
The French and German product agreements are both examples of CFCL’s commercialisation model of working with utilities and appliance partners to develop and trial its systems.
At the end of 2006 CFCL also took an important step into the area of volume manufacturing facilities with the announcement that it is to develop a large scale facility at the Oberbruch industrial park in Heinsberg, Germany. This park, interestingly, is owned and managed by Netherlands-based Nuon – suggesting the possibility that Nuon, which is very active in Germany, could be a future utility partner.
Technologically by far the most important recent milestone was CFCL’s launch (in September 2006) of what it calls the “commercial ready” version of its fuel cell technology, which achieves a power density of 400 mW/cm2, about twice that of the previous design. This means that, for a given power, a much smaller fuel cell module is required, which is easier to integrate into a compact microCHP system for domestic use.
The fuel cell fuel utilisation is increased from 65% to 85%, thus achieving over 50% electrical efficiency. The new stack employs 88 cells with an output of 14 W per cell to deliver the 1 kWe net output (compared with 280 cells of 5.35 W per cell to provide the 1.5 kWe of the previous, all-ceramic, offering). Among the key features of the new design are the following:
• A switch from electrolyte-supported to an anode-supported cell (see diagram opposite). This allows the ceramic electrolyte (which is made of yttrium stabilised zirconia (YSZ)) to be very thin – about 10 microns – which greatly reduces the resistance encountered by the oxygen ions flowing through it and results in higher power density. The anode, which consists of a porous YSZ–Ni mixture, 15 microns thick, is supported on a 200-300 micron thick “substrate”, also made of porous YSZ–Ni mixture. The cathode is 50 microns thick and is made of Sr-doped lanthanum cobalt perovskite. The individual cells have performed well in extended tests at 750ºC.
• A move away from the previous all-ceramic stack to a combination of stainless steel and ceramics, increasing compactness and reducing costs. The new system comprises four 70mm x 70mm square ceramic fuel cells arranged in a window frame type of array, separated by metal interconnect plates. A stack is made up of a number of these layers. The placing of cells in a metal frame structure has made it possible to move from a single-cell to multiple-cell layers, with four parallel-connected cells per layer, contributing to reduced stack size. The previous design had one circular cell per layer. The adoption of a square shape also contributes to better packing density, improved fuel distribution, easier manifolding and better uniformity of temperature throughout the stack.
• Lowering of the operating temperature, from 800-850ºC to about 700-750ºC, but without losing the capability to internally reform gas (which would entail a penalty in terms of efficiency, complexity and cost). Among the advantages flowing from the lower temperature are an increase in cell lifetime and reduced cooling requirements. (For comparison the Ceres metal-supported SOFC, for which low temperature has been a key selling point, operates at about 550ºC.)
The metal–ceramic stack is in fact nothing new to CFCL, which over the past 15 years has gained experience with both all-ceramic and metal–ceramic configurations. The new metal–ceramic stack builds on this experience but makes us of new lower-cost high temperature ferritic stainless steels with a low coefficient of thermal expansion, matching that of the cell, and low interfacial resistance. These steels have only become commercially available in the past two to three years.
To tackle the problem of poisoning of the cathode by volatile chromium species from chromium oxide layers (which form on ferritic steels at the operating temperatures envisaged for the CFCL fuel cell) a CFCL-developed chromium-manganese spinel coating is applied to the metal. The spinel layer, which was discovered and patented by CFCL in the mid 1990s, eliminates this poisoning and keeps contact resistance low.
The stack also employs a CFCL-developed glass sealing concept (adapted from that used in its previous all-ceramic stack), which avoids the need for welding and, as with the steels, is optimised to match coefficients of thermal expansion.
Initial production of the new, metal–ceramic, stacks was due to start in April 2007, to be followed by their incorporation in a new generation of fuel cell modules (stack plus core balance-of-plant).
These new-generation fuel cell modules will benefit from further advances in fuel cell core balance-of-plant systems, including a 50% smaller steam generator, 40% smaller heat exchanger, 60% smaller burner, improved insulation leading to 50% lower heat loss and a 75% reduction in air flow through the system (which is important as the air blower is the main parasitic load). Contributing to the reduction in air flow requirements are the lower operating temperature, the improved thermal conductivity due to the use of metal in the stack, and improved air manifolding.
The new fuel cell modules will be integrated into NetGen microCHP units to be delivered to EWE (and perhaps other customers) for field trials to start during 2007. Of the ten demo NetGen units ordered by EWE in June 2006, two (already delivered) will have the all-ceramic stack while the rest will have the new, metal-ceramic, stack.
The NetGen systems, described as “pre-commercial demonstration units”, are the latest of several iterations of fuel cell microCHP system that CFCL has developed, starting with TD1 (Technical Demonstrator 1), for proof of concept, followed by CHP1 (prototype stand alone CHP unit), and then CHP2 (CHP unit for field trials). CFCL has installed five CHP2 trial units, two in Germany (EWE), two in Australia (Szencorp and ETTA) and one in New Zealand (Powerco). They are grid connected and centrally controlled via the internet from CFCL’s Melbourne headquarters (further demonstration that the basic principle of the “virtual power station” is viable, by the way).
But by the end of this year, says Brendan Bilton, CFCL would prefer not to be in the business of providing microCHP systems. Instead, its ambition is to be a supplier of just the fuel cell modules (ie stack, core balance of plant and control system). The job of integrating the fuel cell module with condensing boiler technology to create the microCHP appliance itself will be that of the domestic boiler makers with whom CFCL is currently in the process of nurturing partnerships, while testing will be done in conjunction with utility partners, EWE, GDF and others yet to be named – but likely to include one or more in the UK.
A major challenge for the integration process will be coming up with a system that is compact enough. As the pictures show the current NetGen units are floor mounted, which may be acceptable in some applications, eg certain types of new build, but ideally microCHP systems need to be available in wall-mountable form. So there is probably a fair way to go yet, in terms of design and development as well as field testing, before we get to the final product, which must be attractive to the homeowner as well being economic and reliable. This is where the domestic boilermakers, with whom CFCL is partnering have a crucial role to play.
Typical residential energy consumption for heating and hot water (German home, source Deutsche Energie Agentur) SOFC cell structures. The new CFCL module uses the anode supported type There are four square fuel cells in one of the layers that make up the new metal-ceramic stack (compared with one circular cell per layer in the previous design) Powder manufacture New metal–ceramic fuel cell module. Note square fuel cell stack, with four cells per layer, as opposed to the one cell per layer of the previous, cylindrical, stack Trial microCHP unit with metal–ceramic fuel cell module Possible scheme for integration of fuel cell module with condensing boiler and hot water storage New core balance of plant Karl Foger, chief tecchnology officer of CFCL, with a NetGen microCHP system. In the background is a CHP2 system, showing the size reduction achieved