top of page

Fuel Cell Operation


There are a number of different types of fuel cell design, including:

  • Proton Exchange Membrane

  • Phosphoric acid

  • Alkaline (sodium or potassium hydroxide)

  • High temperature designs

  • Solid Oxide

  • Molten Carbonate

Fuels other than hydrogen can be used in fuel cells, in particular methanol, natural gas and methanol, either directly or by converting to hydrogen first. As an example, Ceres Power are developing a fuel cell for domestic CHP or transport applications using NG directly.

Here we are only looking at fuel cells operating on hydrogen, and will only consider PEM fuel cells as these are the type currently deployed commercially for road transport.

In thermodynamic terms a fuel cell is an electrolyzer in reverse and there has been some development of systems to achieve a dual duty/reversible operation using PEM technology and solid oxide cells. The interest in reversible fuel cells arises from potential applications in power-to-power systems since a single unit could reduce capital costs compared to having separate units for each duty. There has been some demonstrated operation of a solid oxide fuel cell operating reversibly, but the differing design requirements for fuel cells and electrolyzers suggest that the same unit is unlikely to be efficient in both operating modes.

A typical PEM fuel cell uses air as the oxygen source and requires a hydrogen supply of suitable quality and pressure. Losses of hydrogen occur due to diffusion of molecular hydrogen through the PEM, but otherwise utilization levels of hydrogen are very high due to recirculation of hydrogen around the cell; only a very small purge is required to limit the build up of nitrogen (which gets in the hydrogen by diffusion from the air side).

Hydrogen storage can be as either liquid or high pressure gas, but liquid storage is unlikely to be viable for road transport, and current transport applications use high pressure gas storage.

The standard cell voltage (Eo) is +1.23V. Operating cell voltages are usually in the range 0.65 to 0.8V, depending on the current density and will reduce slightly during long term operation. A catalyst (normally platinum) is used on the hydrogen side to promote hydrogen dissociation. A typical plot of cell volt v. current density is shown below (blue line). The graph also shows a typical curve of power as a function of current density (red line).

Typical Plot of Voltage v. Current Density.

( from ITM, High Power Density Fuel Cells, Hannover Messe, 2012 )

Vehicle manufacturers have to trade off performance factors. Operating at lower current density increases efficiency but reduces specific power output. To achieve a given overall power, as the current density reduced the fuel cell stack has to get bigger, increasing cost and weight of the fuel cell, but increasing range or allowing design for smaller hydrogen storage tanks. What can be expected is that as fuel cell costs come down this will allow a shift towards higher cell efficiencies. Efficiencies can also be expected to improve due to design and technology development.

For example:

  • Honda (2017) have increased the fuel cell power output per unit weight in the new Honda Clarity compared to their previous model, partly by improving cell voltage efficiency

  • Platinum use is a significant cost in fuel cells and much research is directed towards reducing the amount required. In September 2017 Ballard Power Systems announced a development said to replace most of the platinum used in earlier designs.

An example car fuel cell system is shown below – it’s a representation of the system used in the Toyota Mirai.

The Mirai cell voltage is estimated as 0.67V, a voltage efficiency of 54.5% (2017 DOE Hydrogen and Fuel Cells Program Review, James et al).

The overall system efficiency is lower than this due to parasitic losses arising from ancillary systems such as the air compressor, pumps and fans.

The net to gross power ratio for the Mirai is estimated as 80% at full power, rising to 91% for alternative designs that incorporate an exhaust air expander (reduces the compressor motor power) and ejectors (using the hydrogen feed from the high pressure tank as the motive fluid) to replace the hydrogen circulation blower (Ahluwalia et al, DOE Hydrogen and Fuel Cells Program Record,September 2016).

The parasitic losses are not a fixed proportion of the net power output. Some ancillary power usage will reduce at lower power outputs but overall at lower power outputs the losses will be a higher proportion of the system power.

Cell stack operating temperatures are typically around 95C but vary depending on the design. The hydrogen supply pressure to the cell is typically around 1.5 to 2.5 bara, with the air side at a slightly lower pressure. On the oxygen side the oxygen feed rate is typically 1.6 to 2 times the stoichiometric rate, so the exhaust air will contain roughly 10% or less of residual oxygen, plus water vapour, with the balance being nitrogen.

Most of the water produced by the reaction remains in the liquid phase but some will be lost as vapour in the exhaust air and hydrogen purge streams.

The thermodynamic data for a hydrogen fuel cell is shown below for the gas feeds as the pure gases (H2 and O2) at 2 bara and 1.8 bara respectively, at for H2 and O2, at both 25C and 95C (the nominal cell operating temperature). The numbers assume all water product leaves as liquid.


Hydrogen Fuel Cell : Thermodynamic Data

for the reaction : H2 (gas) + 0.5O2 (gas) = H2O (liquid)


Where:

The hydrogen feed rate to the cell is typically 1.6 to 2 times the reaction rate of hydrogen, so the conversion per pass is roughly 50 to 60%; unreacted hydrogen is recirculated.

A small purge of hydrogen is taken off this loop to limit the build up of nitrogen on the hydrogen side of the cell (nitrogen diffuses across the PEM from the air side). Loss of hydrogen in the purge is typically around 1% of the hydrogen system feed rate.

There is also loss of hydrogen by diffusion across the PEM to the air side. This hydrogen is oxidized and does not contribute to the cell power output. The rate of hydrogen permeation is very low, around 0.1 to 0.2% of the hydrogen feed rate.


For a reversible cell under typical conditions the power output is 229.0 MJ/kmol of hydrogen with the balance of 54.6 MJ/kmol as heat. The corresponding cell voltage is 1.19V, so for an operating cell voltage of 0.66V, this corresponds to a cell voltage efficiency of 55.6%, giving 127.3 MJ/kmol as power.

The energy not extracted as work (lost work) appears as additional heat (101.7 MJ/kmol), giving a total heat generation of 156.3 MJ/kmol.

This does not allow for the inefficiency due to hydrogen permeation to the air side but the rate is low enough that we can neglect this effect in these calculations.


We can define the system energy efficiency based on the exergy balance from the table above 5 and on the hydrogen reacted (i.e. excluding hydrogen lost in the purge) as:



where :

The purge hydrogen loss is better accounted for as a physical loss of mass rather than as a cell efficiency loss.


The waste heat from the fuel cell can be used to generate additional power. For waste heat at 95C a practical system would have an efficiency of around 7.3% (including allowance for parasitic losses).

For the efficiency figures given above the cell waste heat rate is 156 MJ/kmol and the potential power output 11.4 MJ/kmol, increasing the net power by 8.9% and the system efficiency to 55%.

This type of additional power recovery is unlikely to be practical in fuel cell systems used for transport applications due to the additional system weight and size,but it could well be cost effective for static industrial applications.



124 views0 comments

Bình luận


bottom of page