Most of us when looking at the recent changes in the world's climate are reasonably convinced that global warming is a reality. Whether this is solely a man-made phenomenon or not, most governments now recognise the need to reduce environmental pollution and particularly the emissions of carbon dioxide.

There can be few people in the developed world who have not heard of the potential of hydrogen as a clean \\$quot;fuel\\$quot; which used in \\$quot;fuel cells\\$quot; or burnt in internal combustion engines, would produce power with only water as a by-product. More correctly, since hydrogen has to be made rather than extracted from the earth, it would be an energy vector like electricity, rather than a fuel. In which case, what is wrong with electricity? The simple answer is that whilst electricity can be made in a variety of ways that would reduce pollution, it is very difficult to store in any large quantities and unless battery technology undergoes a step-change in energy density, it is not very portable.

Hydrogen has been viewed as having the potential to provide both a means of transmitting energy and of being both storable and transportable.

The modern story really starts in California in the late 1980's when legislation was proposed to reduce pollution from urban transport against some key deadlines. Since then billions of dollars have been spent both by governments and private organisations to identify and test the best technological solutions. The driving force for much of the effort was not just the huge current energy demand, but the prospective demand as developing nations improved their standard of living. Few of us realise that over 1.6 billion people have no access to electricity and 2.6 billion have only biomass for heating and cooking. Projections, such as that by Shell above, look at how the energy need could be met, but do not comment on how the energy would be distributed. Many of the sources shown also have problems of seasonality or diurnal changes. The wind doesn't blow all the time and even in deserts night falls on solar arrays.

This emphasises the need to provide an energy storage system. The other trend that needs to be dealt with is that with development comes increased ownership of private transport and this becomes a significant source of pollution, particularly in the urban environment.

One proposed solution to this is to use \\$quot;fuel cells\\$quot; where hydrogen combines with oxygen from the air to produce electricity in an efficient manner. The concept was invented by Sir William Grove in 1839 and first used in the Apollo Space program in the 1960's. Significant progress has been made in recent years in increasing the compactness of fuel cells, increasing their lifetime and thus reducing their evaluated cost. Most major car manufacturers, energy companies and numerous technology companies have been involved in joint venture activities aimed at one of two ends; automotive systems or distributed stationary power systems.

There are a number of challenges to be met:

Fuel cells like the one shown in the diagram use hydrogen to generate electricity, and hydrogen is the least dense gas. A tonne of hydrogen has a volume of more than 11000 m3.

At atmospheric pressure compared with about 800 m3 in the case of air and just over 1 m3 in the case of a liquid fuel like gasoline. Even allowing for the fact that hydrogen delivers more than three times as much energy as gasoline by weight, it is obvious that one of the main challenges is storage. Significant research programs are underway to identify storage systems that are better than high-pressure cylinders or liquid hydrogen, which are the current options. The systems studied are either chemical or physical. In the former hydrogen is reacted with a carrier to produce a stable solid or liquid and this is then converted back to the carrier and hydrogen by some combination of catalysis, heat and pressure change. The carrier can either be recycled or replenished. Typical carriers are metals like nickel which form hydrides, chemicals like benzene which can be made into hydrogen-rich cyclohexane and simple molecules like nitrogen (to ammonia) and carbon monoxide (to methanol). Many of the carriers proposed have undesirable SHE side-issues, carry too little hydrogen or have poor economics. Nonetheless a number of them have found niche applications. The poor hydrogen density of nickel hydrides may be unimportant in stationary use or in ships where it could replace ballast. The boron hydride system which offers good potential for disposable plug-in cartridges has obvious benefits for military applications and may be developed to larger scale applications. The physical systems generally involve the adsorption of hydrogen onto or into a micro- or nano- porous body. In spite of hype about carbon nanotubes, the weight ratio of most systems fell significantly short of the 9% target set by the US DOE and did not compare with cylinders. However a recent development at the Technical University of Denmark have found a way of storing ammonia, which is an excellent but toxic carrier of hydrogen, on magnesium chloride tablets. This produces a safe storage system that holds 9.1% of hydrogen by weight and is now attracting significant interest.

The second major challenge to the development of the hydrogen economy is the cost of fuel cells. Currently they cost about $3000+ per kW depending on type. Whilst this is acceptable for demonstration projects, the additional cost for a 40kW car drive system is clearly unacceptable. The costs need to fall to $60-$100 per kW for them to become acceptable for significant market penetration in the transport system. Significant progress continues to be made and fuel cells are finding applications in secure power systems for remote telecommunications centres. Fuel cell sales in 2004 were 3000 units world-wide so clearly there is a long way to go.

The final challenge is how do you generate and deliver hydrogen in a manner that is economic, compared to conventional fuels, and environmentally more benign over the whole energy cycle (well to wheel). There are clearly many ways of delivering hydrogen to fuel cells. Significant research has been spent on \\$quot;on-board\\$quot; processing of fossil fuels, like gasoline to produce hydrogen for the fuel cell drive system. Similarly, attention is focused on a new generation of high temperature fuel cells that could process hydrocarbon fuels directly. However none of these offer the same reduction in environmental impact that would be available from hydrogen produced from benign or renewable sources. Much of this research was to unlock the \\$quot;catch 22\\$quot; problem of no fuel, nobody buys fuel-cell cars; no fuel-cell cars, nobody builds fuel stations. Fortunately many of the demonstration projects have shown that it is possible to provide hydrogen filling stations which can operate safely and reliably. The first US public hydrogen filling station opened recently in Washington DC and there are many more world-wide.

The most common way that on-purpose hydrogen is produced today is by steam methane reforming or a variant of this. Smaller amounts of hydrogen are produced by electrolysis or by the dissociation of ammonia or methanol. If hydrogen is produced centrally, then the additional cost of transporting it to the filling site and storing it has to be less than the economies of scale compared to local production. In addition the environmental impact of any chosen scheme has to be minimised.

Numerous studies have been made of the cost and environmental impact of a number of schemes. The least impact comes from wind generation driving electrolysis at the filling station, but unfortunately this is also the least economic method. On the other hand, large steam reformer hydrogen production facilities with gas or liquid transport to filling stations offer the best economics and, with certain techniques for carbon dioxide capture, acceptable environmental impact, but they still use hydrocarbon feedstocks. The worst of all worlds is using fossil-fuel-generated electricity to power electrolysis at the filling station.

The graphs show the cost per kilometre of fuel cell system light vehicles compared with untaxed gasoline, using feedstock costs that are aligned. The various options considered include the average size of the filling station and, in the case of small local steam reformers, how many have been built, since this influence the capital cost. The second chart shows the carbon dioxide emissions per 1000 kilometres for various supply schemes all of which, except electrolysis from grid power, have lower emissions than gasoline.

///Keith Guy

///Keith Guy

So when will all of this come about? Recent announcements by the motor industry promise the first mass market hydrogen powered vehicles in 2008-2010. The early ones will be internal combustion hybrids, but the GM commitment for 2010 will be fuel cell powered. We can also expect to see further progress in the urban transport systems as the price of hydrogen powered buses comes down. Stationary power systems will also continue to find new applications. The challenge and opportunity is awesome. If the total volume of hydrogen sold today by the industrial gas industry were used for fuel cell energy generation it would only represent about 0.0004 exojoules compared with the 650 exojoules being used world-wide today. Another relevant fact is that modern cars have a very long life and any change to alternate fuel sources will be a slow process unless legislation forces the change. Such legislation would be political suicide. Government would be better advised to focus on regional standards for the supply of hydrogen and uniformity in regulations and restrictions. Currently a hydrogen powered vehicle would not be allowed to use the channel tunnel and even if it could there is no guaranty that the systems at the other end would be compatible. There are many positive developments in storage, fuels cells and small hydrogen delivery systems. There are also additional drivers in the political push for renewables, which will need energy storage and back-up systems. However, it will be at least 25 years before hydrogen starts to become a major player in the global energy scene and the impact on industrial gas sales will be very small until then.