Greg Glatzmaier - solar thermal storage and heat transfer fluids

You are missing some Flash content that should appear here! Perhaps your browser cannot display it, or maybe it did not initialize correctly.

-->

0MB | download

Transcript

Scott Bilby: This morning we're talking to Greg Glatzmaier, he's a senior engineer at the USA's National Renewable Energy Laboratory. He's part of their Concentrating Solar Power team and he works on systems analysis, heat-transfer fluids, and thermal storage concepts for concentrating solar power technologies. We've been reading all about this this morning and so we've got lots of questions fired up for Greg. Greg joins us live from Golden, Colorado.

And hello Greg, are you there?

Greg Glatzmaier: I am, good morning Scott.

Scott: It's great to talk to you Greg, and obviously it's afternoon for you there but it's a very cold and foggy morning here…

Matthew Wright: Not a great solar resource in Melbourne this morning.

Greg: Well, it's also kind of cold and rainy here, but we've had a nice spring.

Scott: Nice to hear. Now Greg, we're pretty impressed reading your bio. You joined NREL, the National Renewable Energy Laboratory, in 2007 doing the work I just mentioned, but you were also there back from 1987 onwards for a while. Can you tell us about the sort of work you did then, just quickly?

Greg: Well, I've always been involved in solar thermal energy research. Back in the mid '80s and early '90s when I was here the first time there was research going on in solar electric power generation which is what I'm working on now. But at the same time back then there was also kind of a diversity of research that covered other things, like solar material processing and things like that. So at that time I worked on a variety of different technologies.

There is so much emphasis right now and there's such a strong need for generation of electricity using solar energy that that's pretty much the only technology that we're working on right now, and that's what I'm doing.

Scott: Okay, and obviously at Beyond Zero we're a big fan of solar power and we really do think it's going to just skyrocket as far as its potential and its capacity over the next few years, and we're very interested in hearing about the thermal storage concepts you're working on and also the heat-transfer fluids. Could you just explain for the audience what thermal storage is actually?

Greg: Well, I think I'll start by just explaining a little bit about what the general solar plant consists of. The main plant right now that's being commercialised and is the one that's going to be used in the near term (and when I say 'near term' I'm talking about the next five to ten years), it's called the parabolic trough power plant. It consists of a field of solar collectors that collect the sunlight and then transfer that heat energy to a heat-transfer fluid. From there the heat-transfer fluid goes to a heat exchanger where it generates steam, steam goes to a turbine and from there it generates electricity. So that's the basic configuration or design of a solar thermal power plant.

In addition to that you can add thermal storage to the plant, and in that case the heat-transfer fluid can either go to the heat exchanger to generate steam and electricity or that thermal energy can be transferred to the thermal storage unit and then be recovered later for electricity production at a later time.

Scott: So the thermal storage concepts you're working on, is there a particular type of storage you're working on, like a particular number of tanks or anything like that? Currently what do they have, and where are you hoping to take us to make them more efficient or cheaper or whatever?

Greg: The standard right now for the solar industry is what's called two-tank molten salt storage, and in that system, as the name implies, there are two tanks; one tank holds the storage fluid which is hot, and then the second tank holds the same fluid but it's at a colder temperature. And the way the system works is that when you're charging the system or when you're adding thermal energy to the system, you pull fluid from the cold tank and then you heat it up using solar energy and then you deposit it in the hot tank. And then when you're retrieving the energy you basically run that process in reverse; you pull the fluid out of the hot tank, recover the thermal energy and then put it back in the cold tank. So that's the current technology and it works very well.

The storage fluid that's used in that system is actually a molten salt and it's not table salt but it's a nitrate salt which is mined. It's a commodity product so its price tends to fluctuate up and down, so that system works very well and it's being used in several new plants that are being built in the United States and also in Spain, but its cost can fluctuate, and particularly if the cost of the storage fluid goes up then the cost of the whole system goes up. So we're looking at some alternatives to that to try to keep the price down.

Matthew: And what sort of alternatives are you looking at?

Greg: The lead candidate for alternatives is the single-tank storage system, it's also called the thermocline. In that system both the hot and cold fluid are located in the same tank, the hot fluid is on top and the cold fluid is on the bottom, but they're in contact with each other and because the hot fluid has a lower density it tends to float on top of the colder fluid. And when that system is charged and discharged, the line between the hot and cold fluid, that temperature partition kind of moves up and down.

That system also has in it…generally it has a filler material like a granular material, something as simple as sand, and the sand helps to stabilise the hot and cold fluid within the tank, and it also is a medium in which the energy is stored.

So the benefits of that system is that you only have one tank instead of two and that reduces the cost. Also the filler material, the sand, is much less expensive than salt, so that also reduces the cost. The issue with it is it's more complicated to operate and its performance can be not as good as the two-tank molten salt system.

Matthew: Obviously you're pursuing that, so are the savings expected to be significant? What sort of economies are you talking about if you successfully move to a single-tank system?

Greg: I think within the storage system itself, if you just look at those costs, you probably will see a reduction in the cost of the storage system itself that's probably in the order of about 25% to 40%, and how that translates into the cost of electricity that's generated by the whole plant, roughly that translates into reducing the cost of the electricity by about 1 cent per kilowatt hour.

Matthew: Yes, and every cent adds up. Now on that, what sort of price are we talking about to dispatch power under normal financing arrangements from solar plants at the moment, if you were to build one in the United States?

Greg:I’m sorry?

Matthew: What sort of cost is it to produce electricity if a company was to build a modern, proven solar thermal plant now in the United States?

Greg: The cost of the electricity that's produced…I guess there's several ways in which the cost is characterised, but the cost of the electricity that's produced from the new solar plant…you know, as I said, it depends upon the cost of the materials, which tend to fluctuate a little bit.

It also depends upon the size of the plant. If you build larger plants you have some economies of scale there which tend to help out.

But the range that we're looking at, our cost analysis says that the cost of electricity is somewhere in the range of 13c to 17c per kilowatt hour of electricity that's produced.

Matthew: With the addition of the molten salt storage to those kinds of systems that cost 13c to 17c, does that reduce the cost of the plant or increase the cost of the plant?

Greg: Well, it keeps it about the same. It's kind of interesting, if you start with a plant that has no storage, you'll kind of be in that price range, and then as you add storage, say you go from zero hours of storage out to about six hours of storage, the cost of the plant or the cost of the electricity that's produced by the plant actually drops. And the reason for that is that you're using your turbine to generate more electricity so that the cost of that turbine is distributed over more energy production. So when you go from zero hours of storage out to maybe five or six hours, the cost of the electricity actually drops.

When you try to add more storage than that, go out to maybe eight, 10, 15 hours of storage, the cost starts creeping up again. And the reason for that is that when you have…as you add storage you not only have to add your storage system to the plant, you have to also add additional collector area or additional solar collectors to the plant to collect extra energy for the storage system. And when you do that the extra collectors that are added to the plant are added at a further and further distance from the power plant, so your piping costs go up and then also your pumping goes up.

So eventually you get to a point where adding more and more storage actually becomes more expensive. But when you add the first several hours of storage, going out to about six hours, the cost of the electricity that's produced is actually less.

Scott: But it's obviously solar thermal power plants with storage, they can provide ultimately base load as well as dispatchable power and stuff like that, so obviously that increasing cost you were talking about by putting more mirrors out in the field, obviously you don't need to put so many out because you only need to cover eight, nine, ten hours or so, and potentially once it's on a grid system with lots of other energy sources you're going to have your constant power that you can plug in here and there whenever you want anyway, don't you?

Greg: Yes, I mean the markets that we're looking at right now are what are called the intermediate load market and those are energy demand for about 12 to 14 hours a day. Typically they start in the morning, maybe about eight o'clock, and then they go to eight or ten o'clock in the evening. It turns out that energy that's produced from these solar power plants can meet that demand very well. Basically when demand kicks in about eight or nine o'clock in the morning, that's when you have your sun, so you can generate power directly from the sunlight that's collected. And then if the plant has three, four or five hours of storage built into it, then in the afternoon when the sun starts to get low in the sky you can take thermal energy from the storage system and use it to generate electricity on into the evening. So that market is really the one that the solar industry, at least in the States, is looking to address.

There is some push here to try to get into the base load market but it's a much more difficult market to get into for solar technologies.

Scott: So is that only more difficult at this current time and space when there's so much subsidy for coal and stuff like that? But if the world gets realistic about this and puts a price on coal, you'll end up with…you could have base load solar and that can be backed up by other solar plants producing intermediary load power and peaking load power.

Greg: Well, that can certainly happen, and if the need for solar energy gets to that point where people want base load power provided by solar plants, it can happen. Technically it's certainly feasible, but the price for that power tends to go up, and the reason is…it's kind of an interesting thing, if you size a solar plant to collect enough energy to operate 24 hours a day or to generate electricity 24 hours a day, if you design the plant to do that in the winter when the days are short and you don't have a lot of sunlight, then you need a very large collector field to do that.

And what happens is that in the summer when you have longer days and a lot of sunshine, then your collector field is oversized for that time of year. So a significant portion of your collector field has to go idle during the summer, and that tends to drive up the cost of the electricity because you have to pay for all of that collector field up front, and the way to keep the cost down in terms of electricity production is to utilise those collector fields as much as possible. So if you have part of them sitting idle in the summer, then that tends to drive up the price of the electricity that's produced. But there's no inherent or technical reason why it couldn't be done.

And in terms of supplying base load power from solar energy, you're right…the scenarios that are being looked at right now to meet that need are if you combine solar energy or electricity from solar energy with electricity from other renewable resources, and the main one that we have in mind right now is wind power…and there is some evidence that says that there's kind of a complimentary interaction between solar energy which is obviously available during the day and there's some areas, at least in the United States, where the wind tends to be more active at night. So there's kind of a natural compliment between those two technologies, and using those two together I think we can get closer to meeting this base load demand.

Scott: Now, we’re speaking to Greg Glatzmaier. He’s a senior engineer at the USA’s National Renewable Energy Laboratory.

Matthew: Greg, on that wind power, would there be an ideal sizing where perhaps you'd be able to get that winter ride-through with wind power but use the solar thermal with storage for firming power, for the energy security, to make sure that you're always delivering the power when it's needed?

Greg: I think that kind of gets into a lot of technical detail. Certainly the wind farm and the solar power plants can be designed so that they compliment each other and so that they generate power that meets a certain demand profile.

Scott: Greg, I've got a question for you. Obviously the general topic is concentrated solar power, big solar thermal plants with…

Matthew: ...large mirror fields that produce power in the conventional way with steam and…

Scott: Yes, and they've got multiple heat exchange fluids running through them, and they could have two tanks, they could have one tank if you're storing stuff, and you're obviously working on single-tank technology. In a single-tank technology is it…so basically ideally is that leading towards an end result where you're only using one heat-transfer fluid through the entire system?

Greg: That's another configuration that we're looking at and, as you said in your introduction, we work on thermal storage technologies here and we also work on advanced heat-transfer fluids. Right now the current configuration for these parabolic trough power plants is one in which there is a heat-transfer fluid that flows through the collector field and collects the solar energy. That fluid is an organic liquid, it's kind of similar to motor oil although it is thinner, it has a lower viscosity, and it also has a higher temperature limit, but that fluid moves through the collector field. Then if you generate electricity with that directly then you transfer the energy from that fluid to the steam, and that goes to the turbine. So you have the heat-transfer fluid which is organic, you have the steam which is in the turbine, and then if you want to store that energy in a storage system, the two-tank system that I talked about, that storage fluid is molten salt.

So it's a complicated system. It works, but it's complicated because you have energy being transferred between all these different fluids and there's actually efficiency losses associated with those transfers. So ideally you would like to have, say, a single fluid that acts as both a storage fluid and the heat-transfer fluid that circulates through the field. There are some people and some companies that are looking at molten salt, that's the storage fluid, also using that in the field to collect the solar energy.

The problem with the molten salt is that it has a high freezing point. The current salts that are being used freeze at about 120 degrees Celsius or about 240 degrees Fahrenheit. So if you have that fluid circulating through the collector field, in the evening if it's there it's going to cool down and it's going to freeze in the pipes and that's just a major problem, that's enough to basically shut down the plant.

So some of the things that we're doing here at NREL and also at Sandia Labs down in Albuquerque, that's another lab that does renewable energy research, we're looking at fluids that have basically a wider temperature range in which they're a liquid. We want fluids that stay liquid at low temperature, basically at room temperature or temperatures that one of these plants would see in the evening or at night, and then we also want them to stay liquid and be thermally stable at these higher temperatures when they're collecting the thermal energy.

One of the things that we're doing is we're trying find better salt formulations that have a lower melting point, and we've had some progress in that area. We have salt formulations right now that are freezing at about 240 degrees Fahrenheit. They (the new salt formulations) freeze at about 150 degrees Fahrenheit. Now, that's still too high, but we're getting that temperature down. So you're right in that if we could get a single heat-transfer fluid or a single fluid that would act both as the heat-transfer fluid and as the storage fluid, it simplifies the plant, it brings the cost of the equipment down and it also makes the plant run more efficiently, so that's one of the things that we're trying to do.

Matthew: We understand that a company in Spain, SENER, with Torresol and Solar Reserve are also looking at that in terms of power towers, which are a different kind of solar thermal mirror field set-up. Are you also working with these systems or are you only concentrating on parabolic troughs?

Greg: Those of us at NREL, just because of the way the research is set up, we work primarily on parabolic trough technology, and then Sandia Labs down in Albuquerque works on the power tower or the central receiver technology, but that's not completely exclusive, we do some things here in power tower research and they do some things down there in the parabolic trough research as well. But our focus is on parabolic troughs.

The power tower, it's another good technology. It's a little bit easier to use the molten salt in that technology because instead of having these pipes that go through a whole collector field and literally where you have miles of pipes, you have this central tower, and you have a field of mirrors that reflect and concentrate the sunlight up to a single location at the top of that tower. So your piping in that system is much less and more simple, you just have the piping that brings the fluid up to the top of the tower and back down again. So instead of having miles of tubing or miles of piping you have maybe several hundred feet. And because it's vertical piping it's very easy to drain when you're not producing electricity. So that system lends itself very well to using molten salt as both the heat-transfer fluid and the storage fluid.

There are a lot of advantages to the power tower concept. One of the benefits it has, and it's also one of its drawbacks, is that instead of…in the parabolic trough power plants the sunlight is focused to a line, which is the pipe that the fluid runs through, but in the power tower configuration the light is focused to a very central location. So the concentrations are higher, and so the temperatures at which the fluid is collecting energy are much higher, and that leads to more efficient generation of electricity in the turbine.

So that system has a lot of advantages, but probably the main drawback, as I said, that high temperature makes, it's a little bit of a tougher system to control. And if you have any deviation from perfect control, because you're dealing with these high temperatures it's easy to overheat the receiver and bust tubes and things like that. So because of that it's not quite as far along on the path to commercialisation, but as a technology it certainly has the potential to work at least as well as the parabolic trough and actually be quite a bit better.

Scott: At NREL you guys are working on a wide range of obviously renewable energy technologies, so what has the recent change to a Democratic president, President Obama, meant for you guys at NREL?

Greg: Generally it's been positive. President Obama has a vision for this country and for the world, and energy is a big part of that. He's certainly a big supporter of renewable energy. We've been going through a planning cycle in the last few weeks to try to get ready for the next year, and essentially all of our funding comes from the federal government, and it comes through the Department of Energy. And so we're getting some preliminary budget numbers for next year and there is a significant increase in funding for the CSP technology, the solar thermal technology, and then for most of the technologies, as far as I know. So it's a good thing for us.

Our funding levels in the past, they go up and down, and a lot of that depends upon the political influences in Washington, but it's probably driven more by energy prices. And back in the '90 when energy prices were relatively low, it just kind of follows that there was not a lot of public interest in developing new energy technologies, so as a result of that there wasn't a big push for funding some of these technologies. But then in the last few years where we've had this real run-up in energy prices, it just called a lot more attention to the issue.

And then also the climate change and global warming issue is also receiving a lot more attention these days. So all of those things are really having the effect of increasing the level of work activity at the labs here.

Scott: Greg, unfortunately we've run out of time, but thank you for talking to us this morning about the thermal storage concepts and heat-transfer fluids for concentrated solar thermal.

Greg: Well, thanks for having me, I enjoyed it.

Scott: The time just rocketed by, it went far too fast, and you obviously have a lot more to tell us. Thank you very much again for being on the show this morning.

We were just speaking to Greg Glatzmaier, senior engineer at the USA's National Renewable Energy Laboratory.

Transcript by Julie Burleigh