Beyond Zero interviews Miss Rebecca Dunn on solar thermal storage technologies

Beyond Zero's Matthew Wright speaks to Rebecca Dunn from the Australian National University about solar thermal storage technologies.

Rebecca Dunn's interview

download

Transcript

Matthew: Today we are going to be talking to Rebecca, a solar thermal storage expert at the Australian National University. She’s a PhD candidate in solar ammonia thermochemical storage. More broadly she’s participated in papers on all things solar thermal and storage in particular. She’s been touring solar plants around the world. In fact she’s been to the United States, to Spain, to Germany and to France. Anywhere else you’ve been to solar thermal plants, Rebecca?

Rebecca: No, I think that covers it, apart from Australia.

Matthew: Which ones have you been to in Australia?

Rebecca: Well we don’t really have …

Matthew: Research ones?

Rebecca: I’ve been to the Liddell demonstration loop, the old AUSRA linear Fresnel loop up at Liddell in the Hunter Valley.

Matthew: That’s basically a very small booster to a coal-fired power plant?

Rebecca: Yes, that’s right. I think it was about three megawatts thermal when we were there. Then there’s the CSIRO facility. But it’s been a couple of years since I’ve been there. I think they’re just …

Matthew: … commissioning a new tower.

Rebecca: Yes, so that’s exciting, and they’re going to do some storage work there as well. Then, of course, the big dish at ANU.

Matthew: That’s where you work. There are two dishes there.

Rebecca: There are three.

Matthew: Yes, that’s true. The third one’s a little dish. You claim not to own it but I call it Rebecca’s dish.

Rebecca: Yes, I guess it’s whoever’s dish is working on it at the time.

Matthew: So it’s Rebecca’s dish at the moment. You’re the custodian of the little dish.

Rebecca: Something like that.

Matthew: Basically Australia has a research program. There’s another company called Lloyd Energy Storage.

Rebecca: Unfortunately I haven’t made it to any of their plants yet.

Matthew: Interesting story there. I was concerned that perhaps that project wouldn’t come off but I ran into somebody in Mullumbimby, of all places. He was a consultant to them, a consulting engineer from an engineering practice, and he reckoned that that plant’s almost ready to go.

Rebecca: That’s good news.

Matthew: It’s near Parkes – Lake Cargelligo. I’ve actually been offered a tour there so I’ll have to get you along. It’s out past where you’re from, in Wagga, an hour or two, a country mile!

Rebecca: Actually it’s north of there, but yes.

Matthew: Can you give us a recap on where solar thermal is around the world. I think, for listeners, I’ll just mention that, at the moment today, there’s four main commercial renewables. There’s five actually. We’ll go with five. First, there’s conventional geothermal, which isn’t hugely deployed. There’s about 10,000 megawatts around the world or about half of Australia’s average – I hate to use the term – average baseload.

Then there’s hydro. There’s a lot of hydro around the world. It’s significant. China just has massive hydro projects. There are some massive hydro projects in Brazil. Obviously it does have some environmental concerns associated with it. The dams can silt up and can release a lot of methane which is not a good solution for climate change.

Basically then you’re left with wind power – that’s the next most commercial. Wind power has been used as a fuel saver around the world quite a bit initially. They think of it in terms of energy security. If you are using wind in Germany or Denmark, then you’re not importing gas from Russia.

Rebecca: Which is a good thing.

Matthew: A very good thing. Then, most recently – it’s been cooking away for a long time – there’s photovoltaics. They have really got off.

Rebecca: Have come down a lot in price.

Matthew: Yes and the deployment level’s become significant – really big deployment of solar photovoltaics. So the final commercial renewable is solar thermal with storage. That’s pretty much the newest one that’s gone commercial on any sort of scale. It has a benefit over the others.

Rebecca: It’s complementary.

Matthew: Yes. It probably does the same thing as hydro in a way. But hydro’s not suitable for deserts. And the other thing with hydro, being commercial way back to year one of electricity generation, is that many of the sites that were going to be deployed or exploited, probably already have been. What sort of scope is there to use solar thermal with storage? Molten salt power towers for instance, to actually repower the world?

Rebecca: Well, pretty much you can do it in any country that gets good direct sunlight. So, direct sunlight is, if you’re casting a shadow from the sun, then that’s direct sunlight. So you can do it in any sunny countries.

Matthew: And diffuse sunlight – plants can use that? They can grow without direct sunlight?

Rebecca: Yes, plants can use diffuse sunlight. Often if it’s a cloudy place and it’s somewhere that’s good for growing crops and stuff, that’s good too, because you can get fed. And your photovoltaics still work when you’ve got diffuse light, if it’s coming through clouds and it’s not really casting a shadow. It will still work. But of course, if there’s less light, they are not putting out so much power as they would otherwise.

Matthew: What sort of figures, like putting out half their power?

Rebecca: It depends on how much light they are receiving. But you were asking about solar with storage. So the advantage that we haven’t really talked about for concentrating solar power with storage is that it can provide dispatchable power. Basically that means, when you want to produce the power, the concentrating solar plant can do that on demand.

Matthew: So, the thing about power sources is you’ve got inflexible power sources and flexible power sources. The word in the power industry for a flexible power source is a ‘dispatchable’ one or something like that?

Rebecca: Yes, that’s right.

Matthew: Dispatchable ones are often, but not exclusively, the same as peaking power sources.

Rebecca: Similar, yes.

Matthew: So they’re available on demand. Which one is of higher value to a person who runs an electricity grid, a baseload power source or a dispatchable power source?

Rebecca: Well ‘baseload power source’ generally means that it doesn’t stop, so that’s more inflexible. For example, for coal-fired power you’ve got all your conveyor belts bringing the coal from the mine, crushing equipment to crush the coal before it goes into the furnace and all that kind of thing. So you’ve got all this auxiliary – well you wouldn’t just call it auxiliary equipment because it’s quite large – machinery going on in the background that you don’t want to have to slow down and ramp up. That makes it inflexible for ramping it up and down. That’s why you end up with pretty much constant output from it.

Matthew: What sort of time does it take to get those plants from cold start to up and running?

Rebecca: OK, you’re drawing on my memory from a couple of years ago when I visited the Liddell Power Station. I think the operator said about 24 hours from cold start to full operation. Maybe you can do it a little bit quicker but you are probably sacrificing turbine life or something, if you do that.

Matthew: We’re speaking to Rebecca. She’s a solar thermal storage expert at the Australian National University. She’s a PhD candidate in solar thermal storage. Now, Rebecca we just talked about the ramp time. So how would a representative coal plant, taking say 24 hours, or some fairly long period, compare to a hydro dam with a couple of hundred metres of head [the vertical distance between the intake and the turbine]?

Rebecca: A hydro dam’s ramp time is in minutes if not seconds. You just have to open the gate and let the water through.

Matthew: I guess there’s a little bit of a delay just as the water runs through the pipes. And gas peakers?

Rebecca: Well I actually worked at a gas peaking station one summer up in Queensland. They are pretty quick to ramp up. It would be not quite as quick as hydro, maybe 15 minutes or so.

Matthew: To go from a cold start?

Rebecca: If you are going from a cold start, maybe it’s 30 minutes.

Matthew: Fairly quick. A gas peaker is just like a jet engine on an aeroplane. You get the jet engines warming up on the tarmac.

Rebecca: Except there’s a generator attached, instead of a wing.

Matthew: Yes and I guess it’s attached to a big lump of concrete in the ground so the jet engine doesn’t fly away.

Rebecca: But talking about baseload, baseload’s inflexible. Dispatchable power is when you can ramp it up and down quickly. I spent a year in Belgium and in Belgium they have a lot of nuclear power at the moment, which I’m pretty sure they are going to be switching off now that we have had the Fukushima disaster.

Matthew: The Belgians haven’t made an announcement yet but the Swiss, the Italians …

Rebecca: … the Germans …

Matthew: Yes, and I think we are going to see some action at a few plants in Spain coming off. There’s one built precariously in an earthquake-prone zone. That one might be gone.

Rebecca: Yes, so the Belgians have a lot of nuclear power. Actually, probably some of your listeners know this, you can see Belgium quite clearly from space at night because they have lit roads everywhere. The reason they have so much lighting everywhere is because they have excess power from the nuclear power plants. They don’t know what to do, except to light every single little bit of road everywhere.

Matthew: Yes, and of course there wouldn’t be a huge market for it in France because they are full of nuclear plants as well. They are busy trying to dump their power on Italy and Spain and Germany and wherever they can, except the Germans have got too much power as well. Although that’s flipped around a little bit in the last month or so, when they turned off all those nuclear plants. So the solar thermal plant – it can be a drop-in replacement for the less flexible baseload plant and it’s a fast ramper as well?

Rebecca: That ‘s right. Just in the last month, on 24 May, we had an announcement from Torresol, the company that’s done Gemasolar, the first commercial molten salt power tower plant in Spain. It operates for 75% of the year. They have just announced they are selling power to the grid now. They’ve been officially operating since the beginning of May. So that’s pretty exciting. They started operation in March but have been selling power since May this year.

Matthew: If listeners want to check out the Beyond Zero Emissions website at beyondzeroemissions.org we have a previous interview with Santiago Arias. I think it’s almost about time to get him on again, to talk about the now operational Torresol Gemasolar.

Now you just did a paper on that for the very respected IEEE.

Rebecca: Proceedings of the Institute of Electrical and Electronics Engineers.

Matthew: That’s a very esteemed journal in the power and electronics industry. There would only be two or three top journals and that would definitely be one of them. You’ve had that paper accepted. Can you give us a bit of a preview of the Gemasolar power plant and what you’ve described in that paper.

Rebecca: First of all, for any listeners who aren’t familiar with the power tower, it’s a sea of mirrors focusing light up to the top of a tower. It gets hot at the top of the tower.

Matthew: Like a smoke stack, right, but no smoke?

Rebecca: Yes. In most cases it’s a concrete tower. So the sea of mirrors, called heliostats, focus the light up to the top of the tower. It gets hot there. It heats up the molten salt – it’s a nitrate salt, just like fertiliser. You start off with it in the cold tank at 290 degrees and pump it up to the top of the tower. It gets hot and then you store it in the hot tank at 565 degrees.

Matthew: Now 290 degrees is not very cold. What do you mean by that?

Rebecca: OK, ‘cold’ as in colder than the 565 degrees. The reason you start at 290 degrees, or 292, is so that it is already liquid, because obviously it’s harder to pump solid salt.

Matthew: Just like table salt, but obviously this is fertiliser – you wouldn’t sprinkle it on your dinner – but table salt is not molten when it’s sitting on your table. So this stuff’s only molten once it gets above …?

Rebecca: 220 degrees

Matthew: But they’re running the cold tank at 290?

Rebecca: So you’ve got a bit of a buffer.

Matthew: A safety margin?

Rebecca: Yes, that’s right.

Matthew: So you’re moving all that ‘cold’ (290-degree cold) salt through the top of that tower that’s receiving all that light and heat and it’s just being banked in the hot tank. Once the hot tank’s full, what happens?

Rebecca: Well, you don’t necessarily have to fill the tank up. The point is you are storing it in the hot tank until you want to dispatch your power. So, as soon as you want to release power, then you take the salt from the hot tank and run it through a heat exchanger to make steam.

Matthew: So you’re pumping it through the heat exchanger?

Rebecca: Yes. Then the steam drives your turbine which is coupled to the generator and makes electricity. You do that whenever you want to produce power. You take salt from the hot tank and so, maybe that’s running all night.

Matthew: Let’s say you’re in the north-west of Australia and there’s a 10-hour day. You’ve got to hour number 7 and there’s still 3 hours of sun to go and you’ve filled your hot tank. There’s no more salt, apart from safety amounts, sitting in the cold tank.

Rebecca: You’ve filled the tank and you’ve been running the turbine at the same time?

Matthew: No. Maybe the turbine’s been turned off because you don’t have so much power demand, because everyone decided to go on an air-conditioning amnesty. You’ve filled the hot tank, you’ve got three hours of sun to go. What are you going to do with your plant?

Rebecca: Good question, Matthew.

Matthew: I assume that, at that point you are going to idle your plant, not focus the mirrors on the tower.

Rebecca: Oh OK. Defocus your mirrors. Yes.

Matthew: That means you are going to turn them away from the centre?

Rebecca: Yes.

Matthew: My understanding from what you were telling me before is that, if you don’t have any salt in the cold tank, you can’t have the mirrors focusing on the receiver.

Rebecca: No, definitely not.

Matthew: What happens then?

Rebecca: If you don’t have salt going through the receiver, you want to defocus the mirrors.

Matthew: When you say ‘defocus the mirrors’, do you just mean taking them off the top of the tower?

Rebecca: Yes. Point them away from the top of the tower. It’s not very hard.

Matthew: It’s all computer-controlled?

Rebecca: Honestly your smart phone could probably run the whole field of mirrors. Yes, I reckon the computing power in a smart phone … actually you probably wouldn’t even need that. You’d probably only need about a thousandth of the computing power of a smart phone to run the mirrors.

Matthew: A paper I was reading from the seventies, when they were first investigating these power towers – actually they went back to the USSR in the fifties – but when they got serious about them in the seventies, they had a price stack. It measured the cost of all the different components. The receiver could melt if you didn’t have the salt running through it. That was fairly highly priced. Then the mirror field was fairly highly priced but the computing power was fairly highly priced. Now you’re telling me your $400 smart phone can do the job?

Rebecca: Yes, well you wouldn’t even need the $400 smart phone. You probably need an integrated circuit from it, or something.

Matthew: So what else can you tell us about these solar thermal plants, or about Gemasolar?

Rebecca: Well Gemasolar has 15 hours of storage. That’s important so that it can run all night. Because we’ve had trough plants like the Andasol plants 1 and 2. There are actually at least seven of these trough plants that are officially operating and probably some more that are in test operation in Spain: Andasol 1 and 2, Manchasol 1 and 2, Extresol 1 and 2, and La Dehesa. Oh sorry, there’s just Manchasol 1 at the moment.

Matthew: I think there’s almost 10 plants out of something like 20 that are operating and 60 that they are planning on having operating.

Rebecca: La Dehesa and La Florida as well. So these are parabolic trough plants.

Matthew: Those Spanish names Rebecca was mentioning are just the names of the solar plants.

Rebecca: So they are all 50-megawatt parabolic trough plants but they’ve all got seven and a half hours of storage. That’s sizable storage but in winter it’s not long enough to get you through the night. Whereas, with the 15 hours of storage at the Gemasolar molten power tower plant, you can go all through the night.

Matthew: We are speaking to Rebecca. She’s a solar thermal storage expert at the Australian National University and a PhD candidate in solar thermal storage. You’re expecting to finish that PhD at the end of the year or the beginning of next year?

Rebecca: Beginning of next year, yes.

Matthew: Now we were just talking about those different times. In the power industry, for listeners, they talk about availability factor and capacity factor. Capacity factor is what they talk about in Victoria and availability factor is what they talk about in NSW.

Rebecca: I think there’s a slight difference between them.

Matthew: There is a difference. I’ll digress and go into it. Victoria talk about capacity factor because they’ve got the most inflexible plants, that aren’t worth turning off. They are old plants so they try and run them flat chat forever. That means that their cheap power gets dumped via export on the NSW market, where they can turn the black coal plants off. But those black coal plants could have generated …

Rebecca: … and therefore their availability factor is higher and that’s why they want to quote that?

Matthew: Yes, that’s right. So they could have generated for 85% of the time but they actually only generate for 63%. Now people run around saying ‘baseload baseload baseload’. So let’s just define what a 63% capacity factor is. There’s 8,760 hours in a year.

Rebecca: 24 times 365

Matthew: If NSW coal plants have something like 63% capacity factor that’s … you’d be better at maths than me?

Rebecca: I’d have to use your smart phone!

Matthew: It’s something like 5,000 hours. I’m sure I’m slightly wrong because my maths is appalling. That’s why I do radio not maths. So that’s 5,000 hours [per year] for that plant but what about Gemasolar? Let’s go Andasol first, those trough plants that don’t perform so well in winter. What were they achieving once they added molten salt?

Rebecca: The Andasol plant with seven and a half hours’ storage, and all the plants that are similar to it, their capacity is about 41%. So I guess that’s around just under 4,000 hours out of the 8,760.

Matthew: When you say 4,000 hours they’re not starting at hour number 1, then running and then stopping?

Rebecca: No, they are rated at 50 megawatts. So it means, over the course of the year, they are operating the equivalent of those 3,500 hours, or however many it is, at the full output. Actually, maybe they are doing more hours but it’s at half output or something like that.

Matthew: That’s where availability comes in I think. They could be actually generating at a lower output but they are available to generate at full output, if they were called upon by the grid.

Rebecca: I’m pretty sure availability just means, if we wanted to generate we could.

Matthew: You could be outputting at 50% but you are available to output at 100%.

Rebecca: No, I think if you are outputting … anyway, it doesn’t matter. It just means you are generating at some output level.

Matthew: But it is recorded as a percentage. So therefore it can be available for half the time.

Rebecca: Yes, but the definition of availability is, if it’s … for example, maybe you’re not generating because the grid operator tells you ‘no, we’ve got too much power’ but you could have been generating because everything’s running smoothly in your plant. That’s availability. Anyway, it doesn’t matter.

Matthew: Well, you’re the power engineer and you’ve worked in a power station. It’s a bit hard for me to argue. Anyway, going back to Gemasolar. This is a 20-megawatt plant?

Rebecca: Well, 19.9 but yes.

Matthew: Sometimes people will look at me and say that’s not very big but then they’ve got 5-kilowatt solar panels on their roof. How does it compare to my solar system? Say I’ve got one kilowatt of solar power on my roof. How many times bigger is this plant?

Rebecca: 20,000 times bigger

Matthew: So it is pretty big!

Rebecca: It’s not a huge area. People often have a misconception about how much area.

Matthew: It’s 600 metres from one edge of the mirror field to the central tower.

Rebecca: To the centre, yes.

Matthew: So, 20 megawatts. What else can you tell us about it? Build time?

Rebecca: About a year and a half from when they started the foundations. The other thing I was going to say, though, is that from 1996 to 1999 there was Solar Two which was a molten salt power tower operated by the US Department of Energy and associated labs and industry. It was a 10-megawatt plant with three hours’ storage. Now we have 20 megawatts with 15 hours’ storage, which is quite a big upsize from the 10 megawatts.

The next plants that are being developed by companies like Torresol and SolarReserve in the US are 50 megawatts, then 100 megawatts and onwards from there. In fact I think the first one that SolarReserve is going to build in Tonopah, Nevada is 110 megawatts. That’s the Crescent Dunes Project with eight hours of storage.

Matthew: They’ve got another Rice Project in California and a project in Arizona.

Rebecca: But I think the Crescent Dunes Project is due to start construction in July/August this year.

Matthew: Just thinking about the US Department of Energy project with three hours of storage, Solar Two. There’s five times the storage in Torresol’s Gemasolar and twice the size of the plant rating. So basically it’s a 10 times or more scale-up.

Rebecca: That’s right. The next one is going to be the SolarReserve plant that they’re breaking ground on in July or August. That’s 110 megawatts.

Matthew: That’s slightly less storage. If you wanted to do a one for one comparison, it’s equivalent to a 75. In other words …

Rebecca: Maybe it’s more like a 60 or something, but yes.

Matthew: So three times scale-up of Gemasolar’s. That’s pretty exciting. The maximum size would be about a 10 times scale-up of Gemasolar.

Rebecca: So 200 megawatts with 15 hours of storage. Yes. But the point is you just have a bunch of those plants.

Matthew: Then you can get any size you want. In fact, yesterday Patrick, who works for the Zero Carbon Australia Project – he’s a zero carbon research fellow at the University of Melbourne – was speaking to about 25 people from the Chinese State Government. He drew little squares, 130 kilometres on one side by 130 kilometres on the other side. He drew three of them. That would be the equivalent of China’s complete baseload power, not only their baseload, but their baseload and their peaking.

Rebecca: That’s pretty impressive.

Matthew: You could do that with many many many modules of 200-megawatt power plants.

Rebecca: But the point is it doesn’t take much area.

Matthew: We are going to have to leave it there, Rebecca. Thank you for joining us today and we hope to have you back on the show again.

transcript by Bronwyn L.