Fuel cells directly convert chemical energy into electric energy. The conversion does not create any noise or vibrations. Whereas the efficiency of engines is limited by the Carnot process, the efficiency of a fuel cell is not.

Types of fuel cells

The following table gives an overview of different existing fuel cell types. Due to high power density, low process temperature and flexible use, the PEM type is seen as the most promising candidate for mobile applications. The following evaluation therefore refers to this type.

Source: Hörl et al. 2001

Principle of PEM fuel cell

A PEM fuel cell consists of two reaction chambers separated by an electrolyte. The two reaction chambers are fed with oxygen (or ambient air) and hydrogen (or hydrogen containing gas) respectively.

The electrolyte is realised by a polymer membrane plated with a thin platinum coating containing a catalyst. This membrane conducts only protons.

The power generating process is based on the following overall reaction:

2 H2 O2 ? 2 H20

This reaction can be split up into two sub-processes:

1. Hydrogen is ionised (split into electrons and protons) on the anode-side by a catalytic process according to the following scheme:

H2 ? 2 H 2 e-

2. The protons pass the membrane and react on the cathode-side with oxygen to form water:

½ O2 2 H 2 e- ? 2 H20

These reactions create an electron excess on one side and a corresponding lack on the other side thus causing a potential difference or voltage.

This process can be seen as a cold combustion of hydrogen with oxygen. This yields the following advantages over conventional combustion in an internal combustion engine:

• efficiency not limited by Carnot process
• no noise generated
• no harmful emissions
• process is easier to control

Voltage and power density

At its working point a single cell only produces a voltage of 0,7 V. In order to reach operating voltage desired for a given system, several cells have to be combined to form a fuel cell stack. The effective current of a cell is proportional to the membrane area. The power density is a function of current density and reaches a peak of 0,5 W/cm² at 900mA/cm².

Auxiliary devices

Apart from the fuel cell stack, a fuel cell system requires several auxiliary devices for fuel supply and heat removal.

The commercially available 230 kW fuel cell system P4/P5 (from XCellsis) includes the following components:

• Fan/radiator module
• Water/glycol cooling system system module
• Engine inverter/Controller Module
• Fuel Cell stack module
• fuel cell air system module
• fuel cell hydrogen regulating system and diffuser module
• P4VA auxiliary drive & electric motor
• P4T Auxiliary drive & electric motor including traction
• major pipe interconnection system
• major electric power interconnection system
• interfaces to the vehicles
• dynamic braking

Hydrogen supply

Fuel cell powered vehicles can be either supplied directly by hydrogen or by other fuels (especially methanol) to be reformed on-board before injection into fuel cell.

1) Hydrogen stored on-board: Hydrogen can be either stored in a pressurized or in a liquefied form to save volume.

2) Hydrogen can be generated from methanol or hadrocarbons by an on-board reformer. See below.

Fuel reforming to obtain hydrogen

The main advantage of an on-board reformer lies in the higher energy density of methanol and other "indirect" fuels. This leads to higher range compared to hydrogen storage.

A methanol reformer produces hydrogen and CO2 from methanol and water. In the process a number of by-products are produced some of which have a negative effect on the fuel cell catalyst. Therefore the reforming gas has to be cleaned before use inaside the fuel cell.

Types of fuel production and transport

Figure 1 gives an overview of the most important fuel supply chains for fuel cells.

Figure 1: Fuel supply chains for PEM fuel cells

Source: IZT

Layout of a fuel cell driven rail vehicle

A study by DB AG examined the layout of a fuel cell driven MU (DMU of type VT 610). The results for the fuel cell system NEBUS (by XCellsis) are given in the following table: 

Source: Hörl et al. 2001

Fields of application

Stationary: decentralised power generation

Mobile devices: energy supply for notebooks, camcorders etc.

Transportation: cars, busses, railways, vessels

Railway applications

The following areas of application for fuel cells in railways are conceivable:

• traction
• energy supply for auxiliaries and/or comfort functions
• stationary applications

The evaluation focuses on an application for traction.

Manufacturers

Ballard Power Systems (Canada)

Siemens

• Fuel costs: uncertain. Depends on primary fuel used. In the case of hydrogen, price will strongly depend on demand. Considerable scale effects seem likely in case of dynamic demand development.
• Niehues, Edwards expect maintenance costs to fall by a factor of four between 2000 and 2007.

Key factors for price development

Price effects will mainly come from three areas:

• Manufacturing processes (cf. Application outside railway sector - Potential for further development)
• A supply industry for individual components does not yet exist and has to be systematically developed in the future.
• Production numbers have to increase substantially.

Expected price development

In a publication from 2000 (Niehues, Edwards 2000), DaimlerChrysler and Adtranz expected the following relative price development for the complete 230 kW FC system including the components but excluding a traction system and hydrogen tanks:

1999: 16.

2002: 4.

2004: 3.

2007: 1

Thus expected price in 2007: 100.000 Euro.

Niehues, Edwards specify the conditions for such a devlopment:

"By the year 2004 the production of 100.000 fuel cell stacks (ca. 50 kW) for the automobile sector has to be achieved and a dedicated fleet of fuel cell busses of about 1000 vehicles has to be in place by 2007. In case serial production has not reached 100.000 items per year by 2004, the costs for the stacks will stay at their current high level."

In 2002, it is already clear that a production of 100000 items in 2004 will not be reached. The above scale effects are therefore too optimistic as well and have to be postponed by at least 3 years.

Fuel prices

Scale effects are to be expected for fuel prices as well.

In view of the present uncertainties concerning the price of both fuel cells and fuel, no general payback characteristics can be given.

The study by DB AG yielded the following economic feasibility (assuming two different investment costs 2000 and 50 EURO/kW):

Source: Hörl et al. 2001

This table shows that economic feasibility could be reached for investment costs of 50 EURO/kW and methanol operation.

A number of very advanced prototypes exist.

230 kW fuel cell system “P4/P5” (XCellsis) is commercially available.

Several vehicles by DaimlerChrysler and propulsion systems with fuel cells:

• NECAR I, NECAR II:
• NEBUS: modified city bus O 405 N equipped with PEMFC

Fuel cell-driven NECAR 5 (New Electric Car) from DaimlerChrysler and the Premacy FC-EV (Fuel Cell Electric Vehicle) from Mazda obtained a road permit from the Japanese Ministry for Infrastructure and Transport in Feb. 01. The fuel used in the system is methanol.

DMFCs are still in the stage of laboratory applications.

Main overall development targets for future fuel cells are:

• reduce size and weight (and thus increase power density)
• increase reliability and life-time

R&D is done in the following fields:

• New types of fuel cells
• Improvement of methanol reformers and the related cleaning processes
• Hydrogen generation by reforming commonly available hydrocarbons (gasoline, diesel or renewables hydrocarbons like biomass). An efficient and economic reforming of hydrocarbons could open new market segments for fuel cell technology especially in the ground transportation segment.
• DMFC (Direct Methanol Fuel Cell): New developments also permit the direct conversion of liquid methanol inside the fuel cell itself (as opposed to a separate reformer). Depending on future research progress, a compact system could become available to be refuelled directly with liquid fuel.
• Improvements for the complete fuel cell system. The dimensions and power consumption of the compressor could be reduced by lowering the pressure on the cathode side of the fuel cell. The decreased efficiency of the fuel cell itself at reduced operating pressure could be more than compensated by significant savings of compressor energy. Thus, such a measure would raise overall system efficiency.

Manufacturing processes:

At present there are still several components of a fuel cell which can only be produced manually. Despite intensive R&D efforts, convenient technology for mass production is not available yet. This is one of the key cost drivers.