SCIENCE AT THE SHINE DOME canberra 3 - 5 may 2006

Symposium: Science on the way to the hydrogen economy

Friday, 5 May 2006

Dr Wes Stein
CSIRO Energy Technology

Wes Stein is the manager of CSIRO’s Renewable Energy and the National Solar Energy Centre. He has recently completed a study for the World Bank suggesting strategies for progressing solar thermal electricity. He is task leader for the International Energy Agency’s Solar Power and Chemical Energy Systems Program and was a member of the Bioenergy Australia Management Committee. He previously worked with Pacific Power for nearly 20 years and is the recipient of a number of awards and scholarships in the field of sustainable energy.

Making hydrogen from the sun

(Click on image for a larger version)

Firstly I would like to present Wes Stein’s apologies. He would have loved to be here but couldn’t make it. Therefore I will endeavour to step into the breach and do some justice to his slides.

In the slide at the top of the screen you can see the newly constructed heliostat field at the National Solar Energy Centre in Newcastle. The Energy Centre is itself only a couple of years old, and this picture would have been taken about two months ago. So it is still very early days for us, and very exciting times.


(Click on image for a larger version)

This afternoon I would like to provide a quick overview of the National Solar Energy Centre and then move on to the real questions that involve solar energy, natural gas and hydrogen. I will talk about the solar resource that we have in Australia, and the natural gas resources (which, as David Trimm has alluded to, can be nicely matched). I will also discuss some possible infrastructure that could well be useful. I will then talk about our solar gas concept – a little about the reactor, the chemistry, some of the features of the field – and our vision for the future.


(Click on image for a larger version)

The National Solar Energy Centre was opened on 31 March, and the idea of the Centre was to have quite a wide range of solar collectors and solar technologies so that it can be used as a combined demonstration and research facility. We are very keen to promote collaboration, both within Australia and with international researchers, and are already talking to some people overseas, particularly DLR - who are big players in the solar energy field.

To begin with we have two initial research thrusts, with two technologies that are being applied on the ground at the moment. Firstly, there is a trough field, which has long curved mirrors with a single line of focus. These operate well for 250°, or so. Then we are hooking this system up to a gas turbine which is uses an organic Rankine cycle to generate electricity.

The interesting aspect of this particular technology is that it is possible to take heat from anywhere because we are just using a thermal energy source. So this approach opens the door to integrating these systems with diesel generators where you can use the heat from them and may also source this from microturbines.

One of the possibilities we want to explore down the track arises from the fact that we have got microturbines onsite, as part of our power generation. (We also have a lot of photovoltaic (PV) cells.) So we are going to take the waste heat from the microturbines and integrate it with our solar field, to give it a bit of a boost in this way. The output of the microturbine is energy and electricity and this system is a 10kW electrical system. Interestingly, this sort of system can also be used in remote areas for power generation, heat and also cooling, for desalination, and for water pumping. So it’s pretty good stuff.


(Click on image for a larger version)

The other thrust of our activities revolves around the big player – this is the heliostat field, which is essentially comprised of a whole lot of mirrors. There are a couple of hundred mirrors, each one with dimensions of 1.8 by 1.4 metres. So for this field, the combined thermal output is about 500kW on the reactor, which is not actually installed at the site yet. Thus the little white square object seen in this slide is actually a temporary target, where we are going to mount the reactor. The objective of this system is to produce solar gas, which I am going to be talking about, as well as solar hydrogen, and this will be the subject for the remainder of this talk.


(Click on image for a larger version)

I guess that we all know, and especially because we have talked about it today, that solar energy is the largest sustainable energy resource. In fact, Australia is particularly well placed with respect to this type of energy source. It has got the highest average solar exposure any continent or significant group of countries. Yet, as David has just pointed out, most of the solar energy is where Australian people are not located. Now you might think that this is a bad thing, but it actually can be quite a good thing because if you are looking to produce a lot of energy or electricity from this source, then it is important that the land is cheap. In terms of the amount of land area that might be required, it turns out that a 50 kilometre by 50 kilometre area could supply all of Australia’s electricity needs by 2020. This also assumes fairly low conversion efficiencies.


(Click on image for a larger version)

The map on this slide is presented to stress that as the area becomes more remote, then the more prevalent sunshine appears to be.


(Click on image for a larger version)

An interesting observation is that there is a reasonable match between the location of Australia’s natural gas reserves and the available infrastructure. Consequently, the infrastructure is already present that is required to bring the natural gas to market. So it be a great thing if we could grab the sunshine, react it with the natural gas, and then use that distribution infrastructure to get this product to market, and especially get it to the seaboard, where all the people are located.


(Click on image for a larger version)

Now I will talk about our SolarGas and solar hydrogen technology. The idea is that this technology is transitional and modular. We are not attempting to go the whole way towards producing fully solar hydrogen at this stage. It is recognised that fossil fuels are going to be around for a while yet. So instead, we are trying to upgrade these systems by embedding a component in them that is solar.


(Click on image for a larger version)

The process that I will discuss in relation to this slide is pretty similar to conventional steam reforming. Specifically, natural gas, water, and heat are used in the process. However, the difference from conventional forms in this case, is that the heat is sourced from the sun rather than from the combustion of fossil fuels, thereby producing a syngas. So in this way solar energy is captured in a form that is now transportable and dispatchable to market.

I have now just talked about SolarGas^TM or SolarHydrogen, and the distinction here is that in one case we are just making a solar syngas, which is carbon monoxide and hydrogen (even though our numbers are a little bit different from David’s and I’m not quite sure why that might be the case). It is a strongly endothermic reaction but instead of burning fossil fuels to provide the necessary energy for the reaction, we use the energy of the sun.

So we produce a syngas that can be used for electricity generation for domestic or industrial applications and interestingly perhaps, this is a precursor to liquid fuel manufacture through Fischer-Tropsch synthesis. This area is where we really start to get into some interesting ‘bottled sunshine’ applications.


(Click on image for a larger version)

The second way of operating these types of systems is to take the syngas, which has got carbon monoxide, and carry out a water gas shift in the reaction that David has already talked about. This reaction is slightly exothermic, and so there is some loss of energy. However, it is possible to separate out the carbon dioxide and sequester it if this is an intention, and you can also generate a pure hydrogen fuel which is then useful for fuel cells, transport applications if we ever get hydrogen-fuelled buses on a large scale, and also for refining heavier crude oils.


(Click on image for a larger version)

Now, we will take a quick diversion into the area of thermodynamics. As we all appreciate, the higher the temperature, the greater the theoretical conversion to useful work –as explained by Carnot. However, if we were to directly make electricity from the sun by making steam or high temperature air, then to produce the high efficiencies that we need at very high temperatures, we will need high optical performance in the system which comes at a cost. So our concept is to use methane reforming, which essentially uses the carbon available in the process to make the thermodynamics a bit more favourable and allow us to capture the energy at a much lower temperature.


(Click on image for a larger version)

We have achieved this type of process at Lucas Heights at around 850°C. However, in our research at Newcastle we want to push this process down to temperatures in the low 600°C s. We also want to carry out our chemical reaction and the capture of energy from the sun at these sorts of temperatures where we can achieve reasonable efficiencies.

So these issues essentially form the crux of our challenge. Other issues are less problematic since our utilisation is done within a gas turbine or in a fuel cell where actual recovery of this energy is very good. So in theory this complete system has the possibility of providing much higher efficiencies than those where we are just trying to do, say, a steam cycle based on energy from a solar field.

The other consideration is that because we are operating at lower temperatures, our thermal losses are lower and the solar field doesn’t have to be as complex or as nice optically, so in this way it is possible to cut down on capital costs.


(Click on image for a larger version)

Now I will just talk quickly about optimising the solar field because that was one of our critical design parameters. It turns out that there are some conflicting demands to be managed in these systems. The ideal solar field needs a large aperture at which to aim, because then the field doesn’t have to be too precise, and it also requires a cavity that is facing at a reasonable sort of acceptance angle. Then, of course, the reactor has demands that conflict with these characteristics. It requires a small aperture, and needs a cavity that is facing straight down so that there is no convective heat loss. The reactor also needs to operate at as low a temperature as possible. So all of these factors are then driven by thermal losses.

On this slide, it may not be possible to read the graph at the top right of the slide. However, it basically says that as your aperture gets larger, your efficiency drops due to the thermal losses. The red line represents 800°C, while the blue line represents 950°C, and again there is a penalty for when the temperature rises further.

The graph below this one shows the other consideration in terms of looking at the angle of the aperture. Obviously, the more that the aperture is tilted, then the more that air can enter and pull heat out, thereby stopping you from converting this into something useful.

So we considered all of these factors when we were designing our reactor.


(Click on image for a larger version)

Now this isn’t completely blue-sky stuff. (Well on a light note, hopefully it is blue sky because that’s when the sun is shining.) So, more seriously, we have demonstrated that these systems work at a pilot scale at Lucas Heights, in Sydney. You see here the dish reactor.


(Click on image for a larger version)

Now we are working on transferring these principles to a heliostat field with central receiver, and this is quite a different kettle of fish. A dish is pretty easy: all you have to do is to point it at the sun and the whole unit moves together. It is easy to get the focus right because it can be seen when it is actually focussed. However, by comparison, when there are 200 mirrors that need to be trained onto a small target, to track the sun through the day, then this is quite a different sort of scenario. So the real challenge for us is to develop something that is cost-effective and can use a transitional phase.

Obviously, it is going to be a long time before we have fully solar hydrogen. However, we are trying to look at ways that we can get some sort of transitional approach that uses a fossil fuel within which solar energy can be embedded. For example, the fuel can have 26 per cent solar energy in syngas (as a transition). In addition, there appear to be some really good synergies between solar resources and natural gas resources, and the available infrastructure that is present to transport these around. The other interesting aspect is that these systems are quite small, that the field is quite and modular. So this means that if it is not big enough for you, then you just put another one next to it.

Another factor to be considered is that the capital cost is critical. To address this, we have gone for a design that is based on very cheap components – actually, just bathroom glass that is stuck onto a mild steel frame. This steel is actually rolled. So these structures have a focal length, but this is not precise. Then the glass is attached to the steel with a bit of glue, and then some sandbags are placed on top of this structure. So this is a completely different paradigm from conventional solar fields, where there are parabolic mirrors and a lot of money is spent on getting it perfect. We have said, "Let’s make it small, let’s make it cheap, and let’s make it work."

Another big difference is that we have tried to jam as many heliostats into a small area as possible, and our packing density is about twice what you would see in some of the examples overseas. Using our approach, we are trying to overcome the thermal loss issue by changing the design of our next-generation reactor so that we can remove the hydrogen product as we make it. By driving the equilibrium in this direction we can then operate the reactor at lower temperatures and thereby reduce thermal losses.


(Click on image for a larger version)

Now in this slide, we have some images of what concentrated sunlight can look like. The photograph shows a moving bar. This system maps the concentration of solar radiation around the reactor location. To obtain these types of images, we essentially use a camera which can, from the brightness that it observes, actually make a flux map of the way concentrated solar radiation appears.

These are actually individual heliostats, and you can see that some of them are excellent while others are pretty ordinary. That is one of the trade-offs that we have with this sort of system. However, overall we appear to be able to get quite a nice level of flux, and I guess that over the next few months (when the reactor arrives onsite) that we will have a better feel for managing this system.


(Click on image for a larger version)

This is the concept of the reactor. It contains concentric tubes where water enters in the inner tube and gas goes into the outer tube. The water goes through the system and exchanges heat with the produced gas. In this way, the water gets heated up, turns into steam and eventually meets the natural gas which then flows back through the outside ring which is packed with catalysts. Thus heat is absorbed from the sun and used to make hydrogen and carbon monoxide.


(Click on image for a larger version)

This slide shows the reactor awaiting installation on the tower. You can see the concentric rings and the aperture of the receiver, which has got mirrors on it that can be used to try and channel the light a little bit. It also has a water-cooled shield to hopefully prevent the reactor from melting. We were hoping to have the reactor placed on the tower towards the end of this month. However, this is now looking like it will happen more like next month.


(Click on image for a larger version)

Now we will just pull the picture together. The concept is basically to take methane and water, add sunlight, and get a syngas which then can be used in a variety of available ways. So we can make power directly with it using a gas turbine combined cycle. There are also other potential options, such as making a synfuel out of it, modifying gas to liquids, or if we want hydrogen, then this can be produced and we can also make carbon dioxide that can be sequestered. So it is a nice sort of system because there are quite a few options available for manipulation.


(Click on image for a larger version)

This slide shows a potential vision of the future. This is actually a Chinese city since we had some Chinese guests visit us a little while ago and we were talking about the concept. You can see a few of our solar towers sitting on top of a building. One of the considerations that we are examining is whether these sorts of systems can actually be practically placed on the top of buildings and car parks to act as refuelling stations for hydrogen-powered vehicles at some stage in the future.


(Click on image for a larger version)

Just quickly, I want to mention that we fully realise this is a transitional phase. However, we are also mindful of the need to move towards fully renewable hydrogen production. Indeed, one of the things that we are looking at includes thermochemical cycles. For these, a lot of work has been done in the past but it has never really translated into any operating plants.

A thermochemical cycle is essentially a little black box containing our thermochemical cycle and we put water and heat into this so that we can then produce hydrogen and oxygen. It contains a series of chemical reactions where a simple reactant would be, say, zinc oxide. We then use heat to decompose this into zinc metal and oxygen, and then we throw away the oxygen. We then add the water to the zinc metal, and we can thereby produce hydrogen and zinc oxide again. (It’s not quite as simple as that, but this is the concept and the system does basically involve a series of redox reactions.)


(Click on image for a larger version)

The good thing about thermochemical cycles is that they only use thermal energy and reactants that are regenerated and recycled. We are thereby able to get away from having to use electrical energy and electrolysis, and we also overcome some of the thermodynamic limitations of directly splitting water. To try and do thermolysis of water is a pretty tall order.

Currently we are have only commenced screening potential cycles and doing some preliminary experiments.


(Click on image for a larger version)

In conclusion, I just want to say that the Solar Energy Centre is a parallel research and demonstration facility. At this facility, we have a range of collector technologies and we are planning to carry out demonstrations for a full sun-to–hydrogen-to-end use system. We are seeking collaboration, both here and overseas so these are very exciting times!


(Click on image for a larger version)

Thank you.

Discussion

Ben Hankamer – How many hours a day can you use your mirror system? (I presume it depends on the angle of the sun); and how you can target this system up to the site at the top.

Jim Hinkley – Yes. That is something we don’t really know yet - the transient effects. Obviously, there is going to be four hours in the middle of the day where the levels of sunshine are pretty good; what happens at either end of the day is yet to be seen, to some extent.

Ben Hankamer (cont.) – Do you have to be over, say, 45° in order to actually angle the sunlight to the receiver? As the sun comes over the horizon, will it actually start the system up, or do you have to wait quite a lot longer into the day?

Jim Hinkley – Oh no, we can pick it up from very early in the morning and put it on the target, right to the end of the day. The range of mirror actuation is 135°, both in terms of azimuth and zenith.

Ben Hankamer (cont.) – The other question I had is this: if you wanted to do direct splitting of water, I believe you would need about 1,500 Kelvin or something like that. Is that something that could be achieved with those types of systems?

Jim Hinkley – Well indeed, probably a lot more – 3,000 Kelvin. The other problem with direct water splitting is that you have got to get the hydrogen and the oxygen apart very quickly, or they will recombine. However, I guess our feeling is that this system is not a super-high concentration system, because it is designed to function at maybe 900° to 1,200°, not really beyond this to those very high temperatures and concentrations you mentioned.

Evan Gray – I notice that you are using a small rectangular concentrator. Have you considered using hyperboloid revolution, or some other kind of compound concentrator, which will give you the ability to gather the more diffuse input you are going to get when your mirror is located at a high angle early and late in the day? There is one in Australia, by the way – I used to run it!

Jim Hinkley – Well, that reactor is actually the Lucas Heights reactor. We have said there is enough uncertainty in moving to a heliostat array, as opposed to a dish concentrating system, so at the moment we will just take the same reactor and see how we go. We certainly intend to make sure that we understand the shape of the different images. One of the reasons I showed the images for the different heliostats is that we just don’t know yet what they look like yet.

You are quite right that we need to understand how it changes through the day, and what this impact is exactly on the reactor. Certainly, we are ruling nothing out at this stage.

Bob Watts – That was good, Jim. Thank you.