David Trimm was born in England and worked at Imperial College in London until 1976. He then became Professor of Petrochemistry at the University of Trondheim in Norway before leaving for Australia in 1979, when he discovered that low temperatures and high taxes were a bad combination. He was appointed Head of the School of Chemical Engineering and Industrial Chemistry at the University of New South Wales and was recruited by CSIRO’s Division of Petroleum Resources in 2001. He was appointed as a Federation Fellow in 2002. His scientific interests have focused on heterogeneous catalysis, mainly in the context of energy-related research. His current responsibilities include research into the conversion of remote natural gas to synthetic liquid fuels, that involves the conversion of natural gas to synthesis gas – a mixture of carbon monoxide and hydrogen – which can be used as a source of synfuel or of hydrogen. Trimm is a Fellow of the Australian Academy of Technological Science and Engineering.

Catalysis and syngas for the production of hydrogen
by Professor David Trimm

(Click on image for a larger version)

Thank you very much indeed. I also have to thank the Academy very much for the invitation to speak. I say ‘very much’ because I have links with the field of chemical engineering, and everyone knows that a good scientist wouldn’t let their daughter marry a chemical engineer!

Today I am going to talk to you about a future problem but not quite so far set in the long-term future.

(Click on image for a larger version)

It is fairly obvious that we will need hydrogen in the long term, but there are a few problems en route. The basic problems that concern this issue mainly involve energy density.

(Click on image for a larger version)

1. Production costs

Now if you look at production costs, the problems are easy to see. For example, in this slide you can see figures from the Economist. So from coal, gas or oil you are looking at roughly US$1–5 a gigajoule. Then if you take away the CO₂, you then arrive at up to US$8–10. Then comparing these figures with solar, the costs reach up to US$25–50. Comparing these figures then with those for biomass, we see figures of between US$12–18. So there is an immediate advantage based solely on cost, for hydrogen that is derived from coal, gas or oil.

(Click on image for a larger version)

2. Efficiency and Delivery – well-to-wheel

Then if you look at efficiency and delivery, an interesting comparison can be identified that is presented on this slide in the right-hand column. These figures are presented as grams of CO₂ per kilometre and you can see that hybrids and fuel cells are not too far different, with the biggest difference being that, if you really want a fuel cell car, then you are going to pay a fortune for it when you can go out and buy a hybrid for a more reasonable cost. So one is really saying there is a cost-benefit that is present, even on CO₂.

(Click on image for a larger version)

3. Production and supply

Another point is that must be considered is what the costs are likely to be and the direct production costs towards getting hydrogen to be competitive. It is certainly possible to make hydrogen for about the same price as you can make petrol. For example, petrol is roughly US$5 and hydrogen is about US$5.4. This estimate is based on US$25 a barrel However, in practical terms, if you were to fuel your vehicle in a garage then you would get the petrol there for US$7 and the hydrogen into the vehicle for US$20. This is mainly due to the problems associated with distribution and loading.

(Click on image for a larger version)

The other problem is that we have essentially got, at the moment, a hydrogen production ratio of 17 trillion standard cubic feet. If we switch completely  to a hydrogen economy, then we will be looking at 233 trillion cubic feet. In America alone, this produces some interesting figures.

(Click on image for a larger version)

To achieve adequate supply of hydrogen we will need an extra 6,000 chemical plants. Alternatively 9,000 nuclear plants would be needed – and in the USA that means about one at every 100 kilometres around the coast – or about 220,000 square kilometres covered in solar cells. I suspect that this will eventually happen, but there are problems, as Kuwait, where I did a lot of work for a time. Essentially, they put in a solar cell and it really worked tremendously well, until the first sandstorm, when all the mirrors were very nicely abraded and the whole thing collapsed to 0.1 per cent efficiency.

(Click on image for a larger version)

So it can be said that we have got a long way to go. Indeed we really have to consider what is going to happen in the future, and in my estimate, about the next 50 years. Essentially, this comes down to present practise , and how to get the  most efficiency out of our present processes, while at the same time pushing our research and development efforts onwards to new areas.

Of course, this system will involve CO₂ sequestration, which I won’t talk about, though CSIRO has contributed to such a major effort, and one CRC has also been involved. To some extent this is political; however at the same time we know it works – the Norwegians are sequestering the gas – and so it’s really a question of collecting it and disposing of it. That is a problem, particularly for vehicles.

(Click on image for a larger version)

Shown here are the reactions for some familiar reforming processes. The first one is steam reforming, the second one is called dry reforming or CO₂ reforming. In relation to this process, there is a plant that is now being built in Iran, because they have got a high-CO₂ gas reservoir under construction. However, in practical terms, the rate of steam reforming is much higher than the rate of dry reforming. So as soon as some water is generated by the process, it will switch from dry reforming to steam reforming.

The third reaction is the water-gas shift. Then the fourth reaction which I am going to talk about a little later on, involves the direct decomposition of methane.

(Click on image for a larger version)

If we now examine the first types of reforming reactions (which is how something like 98 per cent of all hydrogen is currently generated) then we are looking at autothermal systems. This is where part of the methane is burnt to produce heat, and this heat is used to drive the endothermic steam reforming or dry reforming processes.

So there are two zones. There is the production of heat in the oxidation part of the process, and the first problem that is faced here is oxygen separation. If air is used (one process uses air) then there is a need to take inert nitrogen through the rest of the processing stages. This then increases the required volume of the production plant, and the cost , because the cost of the plant roughly increases exponentially with pressure.

So it is really necessary to separate out the oxygen. In relation to achieving this requirement, there is work going on in America that makes use of an ionic membrane to separate oxygen, and this product appears to be almost on the point of commercialisation.

(Click on image for a larger version)

However, there are further problems since 35 per cent of the cost of production is tied to the oxygen involved in the process, while the remaining 65 per cent is related to the production of syngas (if the goal is to make synfuel). Then since the reactions are catalytic, there is the issue of carbon formation. This essentially involves managing the balance between carbon formation and carbon removal.

(Click on image for a larger version)

This diagram shows where the carbon limits can be found and illustrates the various reactions with steam or CO₂. The curved line represents the carbon formation limit. So there are very distinct limits within which it is possible to operate without carbon formation. One of the features of research in this area has been directed at reducing these limits. Our group has been fortunate in developing the advance of adding tin to the catalyst, which reduced the limit a bit further.

Nickel or rhodium are however the favoured catalysts for reasons which will be explained.

(Click on image for a larger version)

Now if we look at the other possibility, there is good potential for using partial oxidation to produce syngas. This involves a reaction between methane and oxygen, and it is carried out at very short residence times and quite high temperatures. The big advantage is that it is actually exothermic, so once you get the process going it will continue quite happily.

There has been a history of interesting developments in this area. The person who first developed the process was Lanny Schmidt. Then there are others who say that 'methane, oxygen, high temperatures, and short residence times' equals 'too hard to handle'. On the other hand, Conoco® has actually developed a catalytic process that is based on catalytic oxidation, which they are waiting to put into Qatar.

(Click on image for a larger version)

Our research led to the realisation that the throughput of these low residence times was roughly equivalent to that achieved by running a turbine, and this involved two approaches.

One involved putting the catalytic oxidation in front of the turbine and using the turbine to generate electricity. That worked reasonably well. It was found to increase the overall efficiency, and basically decrease greenhouse gas production per unit of energy.

The second possibility was more interesting, and this was where we actually deposited the catalyst on the blades of a turbine.

So we got very excited about this development and now we are talking to various people, mainly in America and Europe. It is an interesting concept, because it requires talking with either the turbine manufacturers or the people who will run the partial oxidation reaction.

(Click on image for a larger version)

If one generates syngas, then in order to successfully make hydrogen it is necessary to remove carbon monoxide. This is because carbon monoxide is a poison for fuel cells. To achieve this then the first step is to operate a water gas shift reactionThis is a reversible reaction, and CO removal is favoured at low temperatures. However, what is actually done  is to start at a high temperature and then proceed to a low-temperature water gas shift reaction.

These processes do not remove quite enough carbon monoxide. Indeed it is really necessary to remove the last traces, and this is mainly done by selective oxidation. This can be done by methanation, but selective oxidation enables the CO levels to be reduced to the point where a fuel cell will accept them.

(Click on image for a larger version)

In terms of CO_x removal, we began by taking a hard look at the high-temperature water gas shift. This is a very old reaction, it has been around a long time, and so we thought that we could perhaps make some improvements. Indeed we found that we could improve this process and have since developed an absolutely fantastic catalyst which is about 10 times more active and hence needs less reactor space. There is only one minor problem: the cost went from ca Euro 10  to  €340 a kilo. So, the bottom line is 'Always look at the dollars', or you can find yourself caught out quite badly.

(Click on image for a larger version)

The next step along in the process involves low-temperature CO removal, which is based on either some very nice, interesting new gold catalysts or some copper based systems.  The former involves  nanoparticles suspended in a metal oxide system, which are excellent catalysts. However, the reversible reaction has an equilibrium and so eventually, even at low temperatures, the processes will run with about 0.5 to 1 per cent carbon monoxide.

Finally, selective oxidation is the last stage of the process. For industrial applications, this is usually carried out  using two catalyst  beds, and hence this area is really where some further research is needed. The reason is very simple – not that reaction is inefficient but that there is the requirement of controlling the temperature. If the temperature is allowed to increase, then the hydrogen will also oxidise. So the temperature has to be lowered sufficiently to enable adequate oxidation of CO but not hydrogen (ca  150°C). This is achievable and commercially this is done very efficiently with two platinum catalysts. Nevertheless, the difficulty is that these two catalysts are needed for temperature control and this involves twice the amount of platinum that should really be required.

(Click on image for a larger version)

Under these scenarios, it can be foreseen that over the next 50 years, hydrogen from oil, coal or gas is not going to be replaced. After this time, one can expect some changes – but it will take this long.

However, the other question we have to ask is: what about Australia? There are some very unusual problems in Australia.

(Click on image for a larger version)

The first of these problems is that Australia’s natural gas is, in fact, remote. There is about 130 trillion cubic feet of known reserves of methane and some industrialists will tell you that their expectation of further discoveries in Australia are about 10 times this amount. So we have very rich sources of natural gas, but these resources are located mainly offshore in Western Australia.

There are also coal seam methane reserves  on the east coast, and there is likely to be a pipeline to be built to Australia from East Timor, but ,as of today, most of the gas is over on the western side of Australia.

So it is necessary to think about distribution issues; and there is also the need to know about the potential for industrial processes which are going to produce greenhouse gases, whether we like it or not.

(Click on image for a larger version)

So, does Australia have natural resources? The answer is that there is a lot of solar energy, and this can be used in a process known as direct decomposition of methane, which CSIRO is currently working on.
Specifically, this process really involves looking at the decomposition of methane using solar heat. A solar collector works best at ca 750°C, which  provides the maximum irradiated area. Higher temperatures can be generated, but then concentrating mirrors have to be used  and so the irradiated area  decreases. Thus we should really be aiming at a temperature in the order of 750°C to decompose methane.

Another aspect that we understand is the thermodynamics of methane decomposition. Certainly if we work at 2,000°C we actually find that we get something like 98 per cent acetylene. However, normally to make hydrogen, we work in the range of about 1,000°C–1,300°C. So the question becomes: is there something we can add to this collecting system that will bring the temperature down to 750°C?

It has been found to be possible  to identify catalysts that promote the reaction. Iron and nickel are proving to be quite interesting, and carbon itself actually acts as an efficient collector of any further carbon that is produced. This is interesting, because there is an economic and greenhouse gas balance to achieve, in the sense that  one answer is to collect carbon, say on carbon, and then bury it. An alternative approach  is to collect the carbon on carbon and gasify, to produce energy, CO₂ and hydrogen. Economically, this is an important question, and gasification should be obvious. However if one looks at a greenhouse gas perspective that involves a carbon tax situation, then the benefit of the preceding option becomes much more questionable.

(Click on image for a larger version)

Is this process widely applicable? Well, it is going to be of importance in isolated circumstances, because if we look at the energy scenario for Australia we see that we are using 28 gigalitres of fuel every year. Bearing in mind that we are probably looking at a factor of four times as much energy content for fuel as that present in hydrogen, then we can easily calculate how much hydrogen we will need to produce, and where we need to produce it.

Thank you for your attention. I am happy to answer any questions if I can do so.

Discussion

Frank Larkins, University of Melbourne – David, in terms of the direct methane decomposition process (in an industrial sense), would there be a purity problem of doing that in a solar environment? Would this kill the economics because a need to obtain ultra-pure methane or something?

David Trimm – It is a good question. You can get methane that is reasonably pure, though there are still some unsaturated and saturated hydrocarbons that come through. What has been interesting is that Ken Hall, Texas A&M, has actually been working with some of these acetylenes to make gasoline.

I think that what is going to eventually become interesting in this particular process is whether we should go entirely to a synfuel process, or in fact go the hydrogen route, with all the problems of distribution and storage for a vehicle. It is going to be an economic balance, and it will be very, very interesting to see which products we will actually prefer. However acetylene does need higher temperatures and that means smaller irradiated areas.

Symposium program

Other speakers

Dr John Wright
Setting the scene: What is the hydrogen economy?

Dr George Crabtree
The two hydrogen economies

Professor Cameron Kepert
Hydrogen storage in nanoporous materials

Dr Sukhvinder Badwal
Fuel cells

Professor Andrew Dicks
Advanced nanomaterials for fuel cells

Dr Evan Gray
Hydrogen storage: status and prospects

Dr Ben Hankamer
Solar powered H₂ production from H₂O using engineered green algal cells

Dr Catherine Grégoire Padró
Production of hydrogen

Dr Wes Stein
Making hydrogen from the Sun

Professor Harry Watson
Hydrogen car prospects