Plenty of Sun To Go Around

Special Issue Volume A Number 1

Print this Article

E-mail this Article

Hong Hou, Ph.D., is President and CEO of Emcore Corporation, Albuquerque. He co-started the company's photovoltaics division and subsequently managed the company's digital fiber optic products division. He was previously a principal member of the technical staff at Sandia National Laboratories.

Craig Tyner, Ph.D., is Senior Vice President, engineering at eSolar, Inc., Pasadena, Calif., a producer of modular, scalable solar thermal power plants. He was previously at Sandia National Laboratories where he he conducted research on oil shale and solar thermal technologies and managed Sandia's solar, geothermal and licensing programs.

TERRY PETERSON: I'd like to begin by asking each of you to tell us a few words about yourself and what your company is doing.

HONG HOU: I've lived in Albuquerque for a total of eleven years on and off. Emcore, is a publicly traded company. We are nationwide with about 800 employees. We have two areas of businesses. The first area is in fiber optics, telecom, datacom, transceivers and transmitter receivers. Our fiber optics business today represents about 75 percent of our revenue.

The remainder of our revenue comes from the solar side of the business, which we started that about ten years ago with licensed technology from Sandia National Labs and the National Renewable Energy Laboratory.

Our initial business focus in the solar area was for space power applications. And today about 49 satellites and space stations have been using Emcore-made solar cells and solar panels, and we're happy to say there's zero in orbit failure. About four years ago, we started using this advanced technology concept for land-based solar applications. It's called CPV, concentrated photovoltaic. We are the world leader today; we supply CPV components to about 28 customers throughout the world. We are the only one vertically intake rated. This is an up-and-coming technology; it's different from silicon or thin film and we believe we will be able to grow our business with a very substantial share for the future.

CRAIG TYNER: I manage the engineering and development activities at a Pasadena-based, startup solar company called eSolar. We've been in business about a year and a half now, and are just completing our first commercial demo project, north of the Los Angeles area. We're in the business of building modular scalable solar thermal plants for utility scale applications. We have one large-scale power purchase agreement in Southern California Edison for one of our early commercial plants, as well as a number of others under negotiation.
I have a long history in solar thermal, having worked in the alternative energy area at Sandia Labs for 29 of the 31 years I was there, most of that time in solar thermal programs.

Our first question is "PV is about four or five dollars per watt. What will change that?" What will change that is more deployment of systems, more practice in running large-scale fabs. Basically learning by doing. It's an aspect of Moore's Law that applies very well to all solid-state technologies, including photovoltaics. Part of the perception that solar costs too much is an inappropriate comparison of capital costs and solar technologies, and which technologies such as gas turbines that burn fuel. When you build the solar plant, you're investing essentially in 30 years' production of electricity in that capital cost. Whereas when you build a gas turbine plant, the capital cost is lower but it's only the tip of the iceberg, in terms of the operating and generation costs, over the lifetime of that gas turbine plant. So there's a tendency to look at that capital cost of four to five thousand dollars a kilowatt and compare that to the approximately one thousand dollars a kilowatt for gas turbines. But when you look at the fuel costs the initial capital investment of the solar system doesn't look as bad as it did at first.

HOU: Today, the cost is a little high, but we need to do a fair comparison. When I lived in California I enjoyed that five-tier pricing. At tier one it was 12 cents per kilowatt hour, but it'd barely light up a couple light bulbs. If you do anything, like turn on your computer, you've kicked it up to tier two. Tier four and five are at 28 cents to 32 cents per kilowatt hour. And I think from that point of view, the solar generated power is competitive. If you compare it to the peak demand, it's competitive to date already. Fortunately, the PV-generated power works out the best when it's at the peak demand time. So it goes well with the demand cycle.

Also, photovoltaic is basically semiconductor-based technology. The cost will go down with innovation and with economy of scale. Think about your computer and how the processing speed and capability has advanced in the last 10 years. I firmly believe with a reasonable policy and with deployment we will be able to get to grid parity in ten years. So it's a little high right now, but indeed, with a favorable policy giving us a jump start, that will help us get through and accelerate a cost reduction curve.

TYNER: Let me just add a little bit to your description of the resource and its potential impact worldwide. By 2050, we're projected to use worldwide primary energy on the order of a few tens of terawatts. And the potential for several terawatts exists just in the United States. Let me first state that I'm a big fan of diversity of energy; it's going to take all kinds of solutions to work our way through this energy problem. Solar is one of those. But on the scale of the resource, among renewable energies in particular, to address this several tens of terawatt need, most renewables are sort of projected to be in the range of a few terawatts.

Solar, however, is available on the scale of tens of thousands terawatts, which is an absolutely huge resource. And that resource is only beginning to be tapped today. There are a few gigawatts on line, if you total up solar generation around the world. There are a few tens of gigawatts now in the pipeline of technologies, either in construction or waiting to be built, again, globally. In the U.S. we have not only the solar resources, but also the industrial capability to build those solar technologies on the scale of hundreds of gigawatts, to have an impact of perhaps fifty percent of our energy needs sometime in the not-too-distant future. Over the next several decades the potential impact of solar on our economy here in the U.S. can be huge. And internationally, it's just that much bigger.

Related to that question of cost is the question of do these systems take a lot of energy to make? And do they ever actually recover the energy that's put in?
HOU: There are three different technology flavors. Silicon poly, silicon based, and thin film based. In a CPV, that's what we do, up and coming emerging technology. But for silicon based, and the energy consumed to make the poly silicon raw material to make the solar cells, and to make the solar panels, it probably takes about one and a half to two years to pay back. And everybody's designing the service life for a silicon-based panel at least for 20 years. So we have a net eighteen years of power contribution.

For a thin film, it's probably even lower. We know for the CPV we use very little semiconductor and that pay-back period is probably between one to one-and-a-half years. So you're not spending ten years to get the energy consumed to make this hard work back, it's only a couple years. And what about CSP? I don't know.

TYNER: Well, CSP is actually very similar. We did a lot of work in this area when I was at Sandia Labs, and typically the number you got for a pay back was a year-and-a-half or so. The parabolic trough large scale CSP plants that were built in the Barstow area in California, in the late 1980s and early —€˜90s, are still all operating today. With more than a 20-year life on some of those plants—€”you can see the energy pay back is good.

This, however, is a question that's very relevant to people when they come to eSolar, talking about our technology, they want to know "how much does it cost and when is it going to be available?" The question they ask after that, in some cases, is "what is the energy payback?" In fact, I've got some of my staff right now working on that problem for a customer who's really driven by that motivation. We want to make sure that we can deliver a lot more energy than it takes to build these plants.

I've heard people mention numerous times, often in connection with wind, but also with solar, that utilities need fossil back up for any capacity that they put on of an intermittent nature. How would you respond to that?

TYNER: With solar energy you have the issue that the sun doesn't shine at night so you're not going to collect any solar energy or make any. You're not going to collect any solar energy, but I wouldn't say you won't make any power from it. The first thing to note is that on the electric grid, something approaching half of the power needs are not base load needs, but are peaking at intermediate load needs, which start to crank up some time in the mid-morning, reach their peak in the afternoon and evening, and then tail off in the late evening. And to some extent, any solar technology, even without storage, matches a good part of that peaking period.

Some technologies, such as power towers and parabolic troughs, in particular, have the capability to integrate, very cost effectively, some number of hours of storage; typically three to six, to maybe even ten hours of storage. And what that does, is allow you to collect energy in the morning without generating electricity. And then turning on your turbine, generating power throughout the afternoon and evening hours, when it's most valuable to the utility unless they're willing to pay the highest rates for it, but also helps them offset capacity requirements with other technologies.

Sometimes the sun doesn't shine because of weather, clouds and other conditions, and so you're not always going to be online; but the work that we've done in the past suggested, typically in the Southwest U.S. and good solar sites, solar plants are able to meet 80 percent of the capacity they commit to.

If you look in the broader context of beyond a single solar plant that can do that perhaps 80 percent of peaking needs, and look at a range of plants spread across the Southwest, it's not always sunny in any given area, it's not always cloudy in any given area, so if you average that integration, you can actually get something that approaches a much higher percentage of those peaking needs.

Turning to the challenges to commercialization and adoption. We've already talked about the high initial investment and how that's often improperly perceived. I've talked a little about policy measures that can help. Hong, you are concerned about the inconsistency of energy policy, which we've seen in the past.

HOU: Solar, because of its high cost, is still a policy-driven industry. There are two types of incentives. Germany, Spain, Asia and South Korea have a feed-in tariff policy. They will buy the offtake, the production of the power, at higher than the market price. That's one way to give the incentives and to drive the investment in a solar farm and installation of the solar modules.

The other way is the way we do it here. Through the tax credit, investment tax credit, a new market tax credit, state, tax exempt, all the different things. We encourage the policy makers to look at different things. We really can operate much more efficiently when we have consistent policy. We were pleased the investment tax credit was extended for another eight years.

I mentioned that the plants in Southern California were completed some time ago and there's been scrutiny of their operation in the last twenty years. Can you comment on what that has meant in terms of lessons learned on reliability and O&M costs?

TYNER: Reliability is certainly a big issue with solar plants. By definition you put them outdoors in very harsh environments, typically in the desert. They don't necessarily see a lot of rain, but they see enough, and they see very harsh conditions of dust and many other things. So making those plants last for their design lifetime, which typically we think of as twenty years, is really a big deal.

A good example of success in that area, is a plant built by LUZ Industries back in late 80s, early 90s, near Barstow. They are called SEGS plants or solar energy generating stations. There were nine of those plants built altogether.
And they were built over a course of five to seven years. The first plants were really the first large-scale solar thermal plants built anywhere, and they had some problems, and required some retrofits and a lot of work over the years. It's been a struggle to keep those plants on line, but they do keep generating electricity and are still making money for their investors today.

The third, fourth and fifth plants, actually, all those beyond the first two out of those nine, took advantage of improvements and lessons learned in those first two plants, and turned out to be much more robust. They're very well maintained.
Those plants have operated extremely well and have set the stage for where we are now in solar thermal. There are plants of that general technology nature that are called parabolic trough plants that are being built around the world.
There was a new plant dedicated in 2007 outside Las Vegas. There are several plants that have either been commissioned or under construction right now in Spain, and many planned for other parts of the world, taking advantage of the proven nature of that technology.

Most of us tend to think of solar power as being clean, almost the paragon of clean. But as one of the speakers earlier today mentioned, there's really no such thing as an energy generation technology that has no environmental impact.
Can you share some of your thoughts on what to do about mitigating impacts or what impacts you envision?

HOU: I can talk about PV, photovoltaic technology. It's very low maintenance, very low impact to the environment, if any. You don't really need water or anything to operate. You do need, from time to time, some operating and maintenance. But compared to other technologies, operating and maintenance cost is very, minimal. It's a very environmentally friendly technology.

TYNER: But it is interesting that there are still a lot of environmental issues that we face in building these large-scale plants. Terry talked about the amount of desert land that was needed, and while it's relatively small, a lot of that land is pristine desert, and there are serious environmental concerns about developing that land and what impact it might have on the desert.

So some plants are required to buy two or three times the area of the land for their plant to offset the land that they're disturbing.

At my company of eSolar we've taken the approach of looking primarily, at least in the U.S., not at these virgin desert lands, but at previously disturbed land, land that is used previously for farming, or other applications no longer in use, and thus environmental impacts are much less. It means the permitting process is much simpler, and the concerns that you might get from the environmental community much less.

What could the state and local governments do to promote these gigawatt-scale deployments? And what would be the workforce needs for enhanced solar deployment? Those dovetail nicely together because I think the state can help with workforce requirements. Can you provide some thoughts on that, Hong?

HOU: New Mexico has a wonderful solar resource, especially for CPV. We have a very favorable policy; we have the state government, the city government, the community and the national labs. We have all the resources and support for renewable energy. And we have probably the best environment for that, but we haven't seen a lot of things happen yet.

I have been thinking about what can we do to accelerate the deployment of the renewable energy, specifically for the solar side, to make very significant economic, or social and environmental impact to the state. There are several pieces required in order to develop big projects like that. You need land, financing, technology, and work labor resources, but more importantly you need someone to buy the power. You need someone to want to buy the power, and of course the grid access as well. The power grid is mostly owned by the public utility companies, and some of them may have the capacity to absorb additional renewable energy. Some of them may not.

TYNER: There's really a huge impact that state and local governments can have on development of solar technology. At the federal level we get the investment tax credit, but we don't have anything in this country, as Hong mentioned, like the feed-in tariffs. So what causes utilities to pay more than the going rate for natural gas, for example, the bisolar plant? At this point it's state guidelines. There are portfolio standards, or renewable standards, that require so much percentage of energy to come from renewables by a certain date. That's a policy here in New Mexico. California has a very aggressive policy and utilities are really behind the curve in that regard and very anxious to be able to up their renewable contributions to meet these standards. And thus, they're willing to pay more than the going rate; they're willing to pay what it costs, basically, to get the technology. So the states have a lot of push.

There's also a lot of impact from local governments, in terms of things like property tax exemptions, sales tax exemptions, and also permitting issues. We had a problem with our commercial demo that we're building north of Los Angeles, with getting fire permits from the fire department to build certain portions of our plant. Because they looked at it as a power plant, thinking fuel, thinking various fire hazards, when in fact, its glass and steel and concrete out there, and not much that's going to burn. So it took us a while to convince them of that; but it's the kind of thing you have to deal very carefully and very thoroughly with at the local level.

To wrap up, I'd like to ask you to give us your thoughts about where you see the technologies and the industry in ten years, plus whatever thoughts you might have.

HOU: We have got all the great pieces and we're ready, willing and able to do it; we just need more favorable policy. We all talk about it and now it's about time to do it. We can use all the help we can get if we are committed to focusing our efforts in the state of New Mexico.

Solar, worldwide, is not only a solution, it's complimentary with our other renewable technology. Through bits and pieces, we'll get there. There's a lot more impact than just the dollar and cents on paper, and how many cents per kilowatt hours.

TYNER: Back when I was working at Sandia, we were one of the primary labs in DOE's solar thermal or concentrating solar power program, and in the late 1990s and early 2000s, our program was under a lot of pressure, budget pressure, and declining budgets every year. We were looking at ways to rejuvenate that program and rejuvenate congressional interest. Of course, at the time, not much was being built in the way of renewables, or there wasn't global warming pressure. Prices were low, the drivers weren't there, but we thought about how quickly those SEGS plants out in California got built and multiplying out how many manufacturers it would take at any given time, we concluded that it was possible to build 20gigawatts of concentrating solar power by 2020.

That sort of became the program slogan, "twenty gigawatts by 2020." And while you could imagine that it was possible to build that much hardware and get it in the field, it was hard to imagine an environment that would actually purchase those plants. It's interesting to look at the pipeline of plants under development, just in this country, and it is well beyond that twenty gigawatts by 2020. The current environment is going to get us there. The continuing driver of environmental concerns and global warming will continue to push technologies like solar and other renewable energies forward.

In the longer term, five to ten years out, I believe we'll compete much more broadly in the power markets without subsidies at all. I anticipate over the next dozen years or so, as the cost of collecting solar energy becomes cheaper, as the plants become more reliable, overall, the cost of collecting that energy goes down, we will start to apply that energy to making fuels. which will help, of course, our dependence on imported oil.