A New Era for Solar - Sarah Kurtz, PhD, NREL

and she was involved in development of various photovoltaic technologies and then went into supporting the concentrator for rotate industry and then in the PV performance and reliability she's going to be moving from NREL to California at a university position so we're lucky to get a chance to have her share her extensive wisdom and I think I've previewed the slides there's a lot of good meat in there she did jointly received the dam David prize in 2007 and the cherry award in 2012 she currently does manage the PV reliability group in row so I want you all to please welcome dr. Sarah Kurtz [Applause] this one well it's a real pleasure to be here with all of you as I reviewed some of the presentations that you've seen in the last few months I rapidly became convinced that you're the best educated audience in the nation possibly or at least in the state of Colorado and I'm not sure that I'll have much new to tell you relative to what everyone else has shared with you recently but hopefully I'll bring a good message it's a special opportunity to to be with the color at a renewable energy Society I've been a contributor I'm not sure if I'm technically a member or not but I've been a contributor for quite a number of years and I very much appreciate all the volunteers who come and knock on my door and I tell them that I'll do my contribution in some other ways so finally now I'm doing my real contribution tonight I think so I'd like to talk about a new era for solar first I should mention that I am still working at NREL but this has not been reviewed or approved by all the animal management these are my own thoughts that I'll be sharing with you tonight I have four primary areas to talk about tonight one is solar is coming of age how it's a new era then I'll talk some about the PV technology development where have we been and what will it take if we want to introduce a new technology and is the next step to look for new applications to transform the world so that's a little bit of a quick Odyssey we'll do we'll start by talking about how PV is growing so rapidly but how can it be both a small amount and a large amount at the same time we're entering a new era for solar that broadens the research agenda and well I'll show you how Hawaii and California are already already showing how that new era is being manifested so the the PV industry has grown to be huge if you haven't been following if you've probably seen this graph before but if you go back to 2006 is when the amount of silicon used by the solar industry surpassed that used for the micro electronics for the silicon that goes into your computers and things and you can see that relative to 2006 in the last 10 years it's grown a lot now the micro electronics industry has also grown since that time but nevertheless Solar has grown much faster and what I'll come back to this a number of times during the talk that time when the need for silicon for solar surpassed that of the micro electronics industry really turned things upside down because the people who had been investing in the silicon purification hadn't been anticipating that solar would be that important and so they didn't plan ahead to give enough supply enough silicon for the community now this is taking a step back and looking at the the bigger picture for electricity in the world so this is the global these are global numbers and these are capacity numbers so I think probably in this room you all know the difference between power and energy so if you have a light bulb that burns a hundred watts the hundred watts is the power and if you burn it for ten hours that will give you one kilowatt hour so this is looking at the capacity and you can see the solar line at the bottom has been increasing very quickly and actually if you've been looking at the news one of the things that's come up very recently is that the total value here and I wish I had a pointer that I could easily point up there but the pointer doesn't quite work but the the solar value is now catching up with nuclear and this year it looks like the number of gigawatts of a solar or PV capacity that are installed in the world will by the end of 2017 equal the amount of nuclear power so historically you've you've heard people talk about how solar is too expensive yeah it can do little things you know we can put it on a roof here and there but it's never going to give us a significant amount of electricity but this year it will be as big as nuclear I'll come back to this graph in a bit but I wanted to talk about how can solar be both 1% and 29% at the same time and if you were paying attention on the first slide I had 1% 20% because the numbers going into 2015 were a ratio of 1 to 20 approximately this year I just updated this slide last night to put the 2016 numbers in and now we have the comparison between 1 and 1/2 percent of the electricity and that's the number of kilowatt hours that of electricity from solar relative to the net capacity expansion from 2015 into 2016 was actually about 29% and so you hear people talk about how solar it's so small it's inconsequential and when you think about the 1% of the 1.5 percent it is very small and is not really making a big difference but if you look at the changes that are being made you see that 29 percent is a big portion of the Hopis and so this is what puts us into a new era if you go back to the graph I had on the the previous and now I'm just looking at the numbers for the total and for the solar and if you look at the the solid lines the red line if you extrapolate it look how much it increased from say 2005 to 2010 if we relatively on this logarithmic graph if that slope of that line were constant after 2010 you extrapolate that line out as you see over to the right and extrapolate where the total amount of electricity capacity in the world has been heading historically and what you find is that those two lines look like they could meet in just a few years it if you work these numbers out it could be just like 2020 sort of timeframe these two would mean now what does that mean if solar is growing so fast that it equals the total how can you have one component be as big as the total so there are only two possibilities here one is that solar in our new era now because we're getting so big is either going to grow more slowly we will need to shift our other energy usage over to electrification so that the top the black line at the top of this starts increasing faster to enable the solar to continue to grow at the rate it has been when I looked at this graph I went and my husband and I went out and bought an electric car so hopefully all of you will consider doing that and actually since we bought the electric car last November and it's really wonderful now when I rent a car someplace else it's like this car is really clunky why would anybody want to drive anything other than an electric car so I do think that there this is when I say there's a new era it's something really for us to celebrate because Solar has gotten so big that we now to continued as the success at the rate we've had it in the past we have to begin to change other things in the way we're living so one way to look at this this particular set of graphs which comes from a study by Paul Denholm and they were modeling in California what it would look like on a spring day in the different scenarios with the different penetration levels and just take note of the right-hand graph there that if you had on an annual basis 10% of the electricity in California coming from solar you see that in the middle of the day on a nice spring day when you've got lots of sunshine but not a whole lot of air conditioning load you can see that the amount of solar is so big that you're now having to shut off a lot of your other electricity so this is the the scenarios leading up to 10% there was a study done by Berkeley recently to look at the value of PV and other technologies as a function of the penetration level and what you see on the left is that wind is pretty much useful to us even as the penetration goes higher you see that the the solar the PV actually has more value at the low penetration than wind because there's a really good alignment between when the load is during the day and when the sun is shining but as you get to higher penetration and you meet the need of the load during the day then the value falls off and so one of the questions is can we do something that would retain the value of PV at high penetration or another question here is and I'll come back to this in in a few more slides is a concentrating solar power that can build in storage or other technologies that can build storage now become more valuable as we go to the higher penetration and so you'll see that the technology mix we need in this new era will probably shift one of the ramifications of the high energy of the high solar generation is that in locations where there is a high penetration the value of electricity during the middle of the day becomes less and so you can see here in Hawaii they have set up rates where in the middle of the day it's only 10 cents per kilowatt hour but when about the time the Sun is setting now it goes up to say 47 cents per kilowatt hour the article that I copied this graphic out of was saying with that big of a difference it actually makes economic sense along with the cost of the batteries that are out there now for you to go out and buy batteries buy up the electricity during the middle of the day and put it back onto the grid in the evening because the the electricity is more valuable then I also recently saw this graph an article that claims that in some places in California now that ratio get could get as high as 18 a factor of 18 that early in the day the electricity is quite cheap and then later in the day when there becomes more of a shortage the price or the value of the electricity becomes much higher now California has been adopting solar very rapidly this was a key reason why I'm moving to California I'd like to be part of this and see if we can make it happen even faster you see that in 2016 the state of California got 13% of their tricity from solar now if you remember back a few slides ago I showed you at the 10 of 10 percent penetration level that had been modeled for the state of California you were getting to the point where in the middle of the day there was more solar than you knew what to do with so indeed here is a data for april 5th of this year and you can see the big nice yellow hump in the middle at the top there is all of the electricity generated from solar at the bottom you can see there the wind and other things and on the the lower right you can see all of the different kinds of electricity there so just to kind of calibrate here if you look at noon there is about 9 gigawatts of solar electricity going on to the grid there's a total of about on this particular day every day is different 11 gigawatts of renewable electricity and the total demand at noon was about 25 gigawatts so we're getting close to the point where the renewable energy is supplying and that actually in California they don't consider the the big hydro to be renewable and I think it's mainly a matter of as they decided to adopt renewables they wanted to do new renewable that hadn't been there and the big hydro had been there for many years so it didn't feel like they should be counting that so actually if you add the big hydro and they're getting in the middle of the day more than 50 percent of their electricity from renewable energy well so what happens now as we increase the amount of renewable energy that we're putting out there if you now double or triple the amount that is going on to the grid what are you going to do with that electricity and so that's where we're in a new era we have to be thinking about new things I just wanted to show you a couple other days just for the front of it here's a day what it looks like when it's windy and you can see then the total amount of renewables is even greater and then this week because many of you probably liked it there was the Eclipse and you can see what happened to that nice hump now it has a notch out of it and in fact the graph on the right Tim de RAF shared this graph with me you can see a day that's a more normal day and then a day with the Eclipse and in a very clean site where you can see there's a nice notch out of the output of that system interesting to look at though that I think they got through that day just fine they there were headlines about how there's going to be a problem when this happens but you can see that the the rate of decrease that they found during the Eclipse it was comparable to what happens every day when the Sun sets so it wasn't like it was a dramatically different thing than what they control every day now they've been worried about this for years I think probably most of you have seen this picture of the duck curve and the the concern they have is that in the middle of the day when there's so much electricity coming from solar the net demand is going to dip and then as the sun's going down you'll have this big ramp in demand and so I looked up on this April 5th day I already showed you the data for you can see that indeed the data that they had predicted years ago is actually really very close to what is being observed in terms of the in fact the belly of the duck on this particular day may be a little bit deeper than what they had anticipated the ramp looks like it's not quite as steep as they had predicted but nevertheless obviously if it's your job to make sure that there's just the right amount of electricity flowing onto the grid at all times you can imagine that every day as the sun's going down you're making sure that you're bringing some new plants online to make up as the solar goes off but because they have to install many more sensors on the grid to have the a good idea of what's happening at all times I think it is actually going in general very smoothly we haven't heard of significant problems coming from that however we do see that there is curtailment out there so the graph you see here shows that in general the curtailment so curtailment is when you just tell the PV systems or the wind systems to turn off because you don't know what to do with that electricity and you can see that typically that happens during the spring so during the winter there isn't that much sunshine then during the summer the demand during the middle of the day for air conditioning goes up but during the spring we have quite a bit of sunshine and not a lot of idea of what we're going to do with that in some cases and you can see by the two blue bars over on the right that in particular in March and April of this year you can see that the amount of curtailed energy really took a big jump up and down at the bottom you can also see the the the numbers for all the different months of this year so far and you can see that March and April or some of the bigger ones here we have that same data but now plotted as a function of the time of day and you can see that really the time of day when we're turning it off varium closely matches the shape of the solar profile and so it's clear although we actually see both wind and solar getting curtailed at some points in general it's the the solar that is causing more of the the curtailment in general if I just did a quick estimate on this that if you look at so far this year in in California they've had 250,000 megawatt hours curtailed so this is essentially electricity they've thrown away because they didn't know what to do with it if you just to try to get a ballpark order of magnitude number assume that that electricity was worth four cents a kilowatt hour that totals up to about ten million dollars so and on the grander scheme of things ten million dollars isn't that big of a number but you can see it's actually getting significant and the projections have been that as we go into future years as California continues to install more and more solar unless they can also install enough storage to take care of this they could expect that the amount of curtailment could get quite large so that motivates again it's a new era we need to look at the what we are going to do with all this electricity so I know that recently you had a talk about hydrogen hydrogen at scale one of the questions is during the day when we have just more electricity than we know what to do with and we're beginning - shut the plants off could we use that to generate hydrogen and those who do the economics say that the cost of the electricity to generate the hydrogen is actually a significant fraction of the cost of hydrogen and if you had hydrogen that could be if you could have electricity essentially at zero cents a kilowatt hour that could be very useful on the other hand you won't necessarily be able to use that the hydrogen generation capability to 24 hours a day and so therefore the cost of the capital investment is maybe not as useful but nevertheless were in a new era where we need to be thinking about how are what are we going to do with this electricity rather than throwing it away can we find a good way to use it so just some of the statistics for the curtailment in California currently the maximum curtailment at any one point in time this particular year was in March about two gigawatts there had been a prediction from the article I had seen earlier from the state of California that the curtailment this spring could have gone as high as six or eight kilowatts a gigawatts excuse me and the prediction there is that it could get up to 13 gigawatts by 2024 and to put that in in context because most of us don't think in terms of gigawatts we might think about how many watts are in a light bulb but we don't normally think about how many gigawatts it takes to run California a typical nighttime total in California is about 20 gigawatts and the typical maximum total usually sometime during the day is more likely to be about 28 so you can imagine that if at some point in 2024 we could be curtailing 13 gigawatts that's getting close to 50% of the total amount of electricity that it takes to Jenna to run California that's actually a big number I think at that point it will add up to more than the 10 million I estimated this time so moving on to the next part of my presentation I'd like to talk now about the PV technology development I think most of you know that at NREL we have for the last 30 years or so been working on different PV technologies I'll give you some historical trends in efficiency and and deployment volumes and some other things so one of the questions that's been out there for years is will there be one winner or will we have many technologies so for many years we've had many different battery technologies and if you looked in your car or in your watch or wherever you'd look chances are you had a different kind of battery for different kinds of applications now the lithium-ion batteries are beginning to take over in a broader range but the same the same question comes up with solar will we have one technology that is the winner or will we have a whole suite of technologies so probably most of you have seen this chart many times Keith emery has been maintaining this chart for many years looking at the efficiency of all the different kinds of technologies and so the first thing you note is this is a really busy graph that has a lot of different data on it because there are many different kinds of solar cells out there so I wanted to show you here how the evolution of the industry has been by technology so on this graph you can see that the dominant two colors the black which is mono crystalline silicon and the blue which is the multi crystalline silicon or polycrystalline silicon are really the two dominant factors and now if you remember a little bit ago I was talking about how in 2006 was the time when the demand for silicon to meet the needs of the solar industry surpassed that of the micro electronics industry well what happened is that after 2006 wiebe we developed a shortage of silicon you know there's plenty of silicon in the world you can go get silicon from the sands of the beaches but the the equipment that you need to take the sand from the beaches and turn it into the purified silicon that's used in the semiconductor is actually quite large and the amount of time that it takes to get the investment lined up to get the permits to build the plants to get the plants built and actually get some new purified silicon on the market at that time anyway was something more like maybe four or five years and because people had been investing at a rate to meet the needs of the micro electronics industry but not the needs of the solar industry we came up short and so we had a shortage market back around the time of 2008 or so and what we found at that time was First Solar it's a company here in the United States probably many of you have heard of them that they make the cadmium telluride technology they didn't need this they purified silicon and they had a technology they had developed which they found they could put into factories and ramped up their production faster than what people could do purified silicon and so they went around people had been during that time people are very excited about solar and people had written contracts to say I'm gonna install for you a five megawatt plant or a ten megawatt plant I'll install it for you two years from now and people just assumed they would be able to buy the modules that they needed but it turned out that because of the shortage of silicon the companies who had not planned ahead and signed the contracts for this the silicon were not able to meet they weren't able to ramp their production in the way they had planned to and so first Solar went around and bought up all those projects and you can see they grew very quickly if you look at the green bars on the green stripe on this they grew very quickly and in about three years time they went from a small startup company to the biggest company in the whole world and they didn't hold that lead for very long because then the market went shifted more from a shortage market more to a surplus market and we've now been in a surplus market for quite some number of years that's made it much more difficult for the the thin film and the other technologies out there I'll come back to this general theme that you see understanding the history of how things have evolved will be important specifically back in 2012 as in in the introduction you may have noted I had worked on the the CPB industry the concentrator PV and the idea here is to take the very high efficiency cells you can make with the multi-junction cells take a cheap lens or a cheap mirror and focus the light on a very teeny tiny cell you can reach them a higher efficiency without having to have the cost of the semiconductor material and so I routinely in 2012 was giving talks about how exciting was about how the CPB industry was growing and you can see 2009 2010 2011 2012 on this graph it was growing very nicely but at that time they should the surplus market I'm sorry the shortage market switched over to the surplus market and during that then after 2012 it became much more difficult just as in the here you can see that in around the 2012 sort of timeframe the first solar with the the green bars started to also lose market share it became much more difficult here for the the new technology in the concentrator PV industry and the numbers now have dropped down quite dramatically now a similar kind of story came comes through in the concentrating solar power industry they had been growing back up till about the 2012 sort of timeframe that was during the time when there was the shortage market for the silicon other technologies were very able to compete with silicon because it was difficult sometimes to line up the supply of the silicon that you needed and so the CSP plants were growing quite well and then around 2012 2013 kind of as good a time frame as the silicon modules became more available the rate of growth dropped off quite dramatically now an interesting question here is I don't have the projections for you for the CPB but the CSP the concentrating solar power which uses the concentrated light to make something very hot and then use the hot fluid to run a conventional kind of power plant where it uses the heat to run the turbines in this case now if you can heat that liquid in and keep it hot and then after the Sun Goes Down use it at that time to generate electricity you can build some storage in without needing to add the batteries and so what we have right now is china has gotten very interested in CSP China's has done remarkably well at taking the photovoltaic technology squeezing the cost down figuring out how to ramp it up and produce the modules at a much lower cost than what people had thought they would be able to do and now they are looking at doing that for CSP and they want to do that because if if you're looking at the curtailment that I quoted for you in California that pales compared to the curtailment that's happening in China right now there are some regions in China that may be curtailing as much as more than 20% of the electricity being generated by the electric power plants and in fact they have shifted their incentive programs to say we're not going to install any more solar in this area because we can't get the electricity out to the the load centers where they need that electricity so they're looking at if we could put CSP in and get shift the solar energy and deliver electricity later in the day that would be very useful to us so this is again an a ram off' occation of a new era for solar we view that PV is effectively one if you go back and if you look at these numbers for what's happened with CSP the rate of deployment of CSP relative to the rate of deployment of PV they're both solar technologies but pb's been growing very very fast and CSP is pretty much dried up and so there's some people are tempted to say it's dead forget about it but you need to look at that we're in a new era and now the ability to have to deliver the electricity in the evening could make a big difference will China be successful in getting the cost of CSP down so that it becomes the dominant technology we don't know yet but it'll be very interesting thing to follow in the in the next few years certainly this projection here that came from IEA would suggest that that the trend of drying up is going to reverse so I wanted to talk next about some module efficiencies and again it's fun to look at this data in a historical perspective so this is a graph that we've been working on Keith has actually helped me a little bit with this project and it's something I need to finish up in the next month or so before I head off to California to do the graph now for the modules which is different than the cells so the cells maybe say one square centimeter or maybe a hundred square centimeters the modules might go up to more like a square meter or something so the again we have the different technologies by the different colors here I want to take you through the different technologies to give you a little bit of the history so an interesting thing with this the silicon efficiencies and recall from the graphs I showed you earlier the black and the blue so the the black bar that I showed you earlier in the deployment graph is now the top line of this and you can see that the efficiencies have increased over time the size of the symbols here is proportional to the size of the module actually the little teeny markers here are just small cells and so you can see that the smaller devices tend to have the higher efficiencies than the larger devices both for the mono crystalline and the multi crystalline that you see at the bottom and also if the scales are the same on the top and the bottom the the mono crystalline always tend to have higher record efficiencies than the multi crystalline now silicon technology if you look on the the graph on the the cell graph it might look like there's been almost nothing going on in the silicon technology because you'll see that there was a record set many years ago and then nothing appeared to happen for many years but if you understand that there are many different flavors of the silicon technology there has actually been a lot going on and it's just not shown on the the record efficiency charts if we talk a little bit now about the CAD telluride efficiency history going back to you remember a back at the time of 2006 when the shortage market developed where there wasn't enough silicon and cadmium telluride was able to step in and fill the void at that time you'll see around 2006 on this graph that there weren't any efficiency record set and that was because first solar was primarily focused on ramping production building factories as fast as they could and they weren't looking that much at the efficiency then when we get closer to about the 2012 timeframe when you recall that now we were moving into a surplus market and cadmium telluride was having more trouble competing relative to the growth of the silicon they went back and said our efficiencies are lower than silicon maybe we need to improve that and so they began to invest more more of their profits into the R&D then what most of the the silicon companies have and you can see that that paid off that they made very fast progress in terms of improving the efficiencies you can see at the right side of this graph and both in this in this chart the open symbols are looking at this small area cells and the the solid symbols are looking at the large area modules and you can see that both the cells and the modules improved in efficiency very quickly in recent years and so now cadmium telluride is very close to being able to compete with the multi crystalline silicon in terms of efficiency at the module kind of level a similar story comes up with cigs that in the 2006 kind of timeframe there was a bit of a pause and the efficiency records came much more slowly and then when you get back into say the 2012 kind of time frame then the investment in making higher efficiency modules became much more important to be able to compete in the marketplace so just now looking at the big picture of the different kinds of technologies it's kind of interesting to see that back the the solid greens well here is six that's copper indium gallium diselenide and for many years it had a higher efficiency than the cadmium telluride it'd but the large investment that was made by First Solar and and other some other competing companies who eventually merged with them notably GE they managed to bring together the their expertise and increase the efficiency of cadmium telluride so now the cadmium telluride and cigs technologies are neck-and-neck or maybe the academy of telluride is a little bit ahead of 6:00 so coming back now to looking at the fractions of market share the Academy and Telluride has lost market share in recent years as you can see from the green stripe on this graph but because it's gained a lot in efficiency in recent years if there were a shortage in the market now cadmium telluride is really poised to grow rapidly and so question is could there be a shortage in the could we move back eventually to a shortage market and that would open up the possibility to reverse the trends and you see from history when there's a shortage market the competing technologies will gain relative market share but currently we're not in that situation but again we could be in a new era now where we could eventually give opportunities for a new technology just to look through a few other things the amorphous silicon was one of the early entrants if you actually look at the red curve on this one of the key takeaways you have here is that back in 2000 the early 2000s and actually if you go back in the 1990s amorphous silicon at one point had something closer to I think 10% of the market share and it's now effectively dried up you can see that a key reason for that is if you look at the efficiencies they're simply lower than all the other efficiencies they've never managed to get the efficiency up to the point that they can compete with the other technologies especially as silicon prices have come down so one question is with amorphous silicon now the small red bar you see at the top there with it having lost so much market share would it be possible to bring it back now now there's some new technologies out there the organic PV we often call Oh PV and paratus kites the profits are a new exciting material and you can see that the Prostate here shown in black has increased in efficiency very quickly just in their in the recent years to come up to efficiencies in fact over 20% which is pretty remarkable for these technologies where they some of these are actually fabricated at close to room temperature so if you envision that a future of PV that you'd really like to see is to be able to paint on the PV and it's an interesting thing that silicon to form the junction you heat it up to something like a thousand degrees centigrade when you cool it down it really locks in a lot of what had happened at a thousand degrees centigrade but if you try to make a semiconductor material at room temperature what's to prevent it to continue to do that initial growth process and if you think about it one of the few things in our world right now that we have that's a material that's made at room temperature that's very stable is paint and so what what you were trying to do with the peroxide is to effectively use a process that's kind of similar to paint at some level they you can you coat something and then you evaporate something out and then you can make maybe a stable layer now they're currently using the equivalent maybe of the oil-based peratt's kite if they could get it to the point where we had the equivalent of the latex paint where they were using it with water that might be better but they're looking at trying to get a process it's very cheap to make because if you just need to if you're gonna paint solar cells on your walls or something you don't have to invest in the the sophisticated equipment that are needed for semiconductors typically and so the pirata kites are very exciting from the possibility that it could eventually go to something where you went to the hardware store and got everything you needed to paint your walls with the paratus kite material make solar power from that we're a little ways away from that but it's the vision of what we'd hope to go to now three five efficiencies so this is a technology that I personally worked on for some time this is a technology that uses multiple junctions so here where you see one j2j this is multiple junctions and then the larger number of junctions tends to enable you to get to the higher efficiencies and also this these numbers are one Sun if you concentrate the light onto them you can get the efficiencies even higher and whereas we were talking before about efficiencies being really good if they got over twenty percent for these cells under concentration you can actually get over forty percent so that's close to double what you can do with the more conventional cells so these are very exciting the challenge though is to get the whole system to work because you have to concentrate the light so you can't use all of the sunlight you can only use the light that's in the direct beam that is the light that's not scattered and by the clouds or the the moisture or the dirt that's in the air but just the light that's coming straight through and so you lose some of the solar resource but if you can make it work in the sunny locations of the world you can still generate as much or more electricity in a given area than you can using a conventional approach here the technology is extremely strong but as I showed you before it became ready just at a time in the marketplace when the we ended up with a surplus market and so for business reasons if you're somebody out there buying a product would you prefer to buy the one that has been deployed in huge volume or would you prefer to buy the new exploratory one and there were some people who were willing to do it in 2012 but since then the number of people who've been willing to invest in the concentrator pv technology has decreased significantly if we had another shortage market I think it the demand for this technology would go way up very quickly and they would be able to read quickly I'm it's interesting to note here you can look at the efficiency limits theoretically and look at what we've achieved and look at the number of junctions on the left you have these this data for one Sun and on the right you have it for concentrated sunlight and we could spend quite some time on this but in the interest of time I'll move on just to the big picture here is that the debate for many years has been a revolutionary versus an evolutionary approach and people have said for many years that silicon couldn't make it we had actually recently in the 40th anniversary at NREL Dennis Hayes was one of the early directors of NREL and he told a story there where when he took over as director of NREL at that time it was Siri the Solar Energy Research Institute and the Department of Energy at that time had some skeptics in it you've heard a few skeptics currently in Washington today at the end doing studies maybe so they commissioned a study they said we're not sure we should be investing in the solar technology it's probably never going to go anyplace so they did a study and Dennis Hayes said the conclusion of that study was very clear that you could never get to $1 a watt with silicon he then he said that he did the numbers work the numbers and today where we are in the dollars of that time period would be 10 cents a watt so we've reached a price point that is 10 times lower than what the Department of Energy back in something like 1980 concluded was impossible fortunately at that time Dennis Hayes was able to convince the people ideally to overrule that study and go ahead and put some research into solar and because of that the United States has played a leadership role over the years in helping to develop the technology bring the cost down and enable what the the phenomenal success that I've been sharing with you today so this question of 10 silicon make it or not the word on the street has been it can't I remember when silicon got to $2 a watt and the silicon company stood up and said here's our roadmap to $1 a watt and the thin-film people said oh they can't do it no way and then when they got to $1 a watt the company stood up and said here's our roadmap to $0.50 a watt and again the thin-film people said now they can never do it today I hear that in India they sometimes sell models at 20 cents a lot so we as a community are terrible about predicting what price point you can get - it's not just for solar you've probably heard the stories with when Thomas Edison was working on the electricity and the light bulb came out and they said well light bulbs are wonderful but only the wealthy will ever be able to afford them yet the incandescent light bulb can be made now prepend so we as a society are actually remarkably good at being able to bring the cost of products down if we invest enough and get them to large enough scale manufacturing and put enough innovation into that so I think one of the the takeaway messages we should keep in mind both with solar and potentially for all the different renewable energy technologies we have the same thing going on right now with CSP to say CSP is too expensive it's way more expensive than PV so why don't we just use PV well it may be that PV and batteries batteries are getting a lot cheaper - it may be that PV and batteries will end up being cheaper than CSP but we don't know until we give CSP a chance and so in this new new era we have opportunities for different things to happen here is a graph this Accra this graph actually doesn't give you the most recent data showing how the prices dropped very dramatically when we shifted from the shortage market into the surplus market and here is what we like to present this as the experience curve that's looking at the price on a logarithmic scale as a function of the total number that had been deployed on the the x axis here again on a logarithmic scale and you can see that by extrapolating this out you can estimate where we might be able to be and the interesting thing here is that we move down this faster than we had expected you do see that the deviations here correlate with whether we had a shortage or a surplus market currently and I don't have the data point on this graph it goes kind of right in the middle of the box that says here 50 cents per watt at about one terawatt we currently have prices that are getting quoted at about 35 cents a watt one of the questions is why are we so far below and part of that is that we currently have a surplus market so that the prices are really low but there's also another theory I've heard and that is that we have all the industry and not all the industry but a good fraction of the industry is working in China and the companies there share information and a way that we in the United States probably would not support and so if if one group develops a new product chances are they go out so say they could make a better sell the company is selling the sells to the module makers will go out and sell that sell to every module maker around and they share the information so quickly that the innovation is actually accelerated and we've managed to bring the prices down faster than people would have predicted based on the the historical learning curve okay so moving next on to introducing new new PV technologies if you have a technology that's out there that's so well established the silicon a question is how could we bring in a new technology to compete with it so entry markets may be a key thing there may be other strategies as well I wanted to note here arpa-e is currently funding some various projects the mosaic and the focused project r2 that use ideas of concentrators no Rana is very much a supporter of the approach of using concentration they have a number of different projects you can go to the websites below if you're interested in seeing all these one here at the top is a planer light light guide another is the what they call the glint photonics which accepts light from a fairly wide angle into a stationary CPV system then here we have a couple of projects one from the Arizona State University where they're looking at having a mirror that reflects the the light and they have a silicon cell that captures some of the light at the back and then they have a light that reflects at a mirror that reflects the the light that has an energy that turns into heat but isn't a very good for a solar cell and they can have both in the hot liquid generated in addition to getting the electricity similarly there at the University of Tulsa they have a plasmatic mana particle where they're looking at enhanced liquid filters for optimal solar conversion now if we go back to the learning curve I showed you on the previous slide think about if you have a new product that's coming in at a cost that's less than what silicon is today the the blue data will give you on the right side what you're competing with on the left you have the the green curve that may be starting at the initial small volumes may be starting at a much higher cost or price even though it has the potential to beat the incumbent in the long term if it could scale up to get to the same size and so the question is how do you do this investment and one strategy is to use entry markets and I wanted to show you some examples you've probably seen the the little walk lights that's under the right side here the nice thing about these is that you can buy them at the hardware store and instead of needing to run an electric cord through your grass or whatever you just poke it into the grass and then it charges itself during the day and turns the light on at night on the left the lower-left here we have a drone when you're looking at a drone you would like to have a really lightweight high efficiency product and so if you can make a technology that has a high efficiency and the lower weight somebody might be willing to pay more for that than for the regular silicon cells because at the system level there's an advanced there was a an announcement just in the news this week that talked about how the one of the the manufacturers will be looking at putting gallium arsenide cells on the roof of cars I wanted to talk briefly about the in the introduction they noted that I will be moving to California and the next months and one of the reasons I'm moving out there is Roland Winston I don't know if any of you or how many of you know Roland but he was the inventor of non imaging optics and has been working on various projects with concentrating light for many years so this is a project here that Roland has gotten funding for recently and I wanted to share with you briefly how this works so he's looking at how you can use the different portions of the spectrum and these shows in the blue there is the the bright blue is the energy that he can get from the gallium indium phosphide cell the the bright green is the energy he can get from a gallium arsenide cell and then the bright red is the energy that his system will get in thermal energy so this is a hybrid system and for many years people have said it makes sense to make a system that can use all of the energy make some electricity and generate some thermal heat and that gives you a better usage of the total spectrum however the way the funding agencies go they find that if they duplicate efforts at all the people above them say hmm looks like you're duplicating an effort we could probably cut that so anytime they see something that might be considered in the other guys territory they say no I don't want to touch that so if you put in a hybrid proposal the thermal people say that's got solar electricity that's to go in the solar electric and if you put it into the solar electric side they say oh that's got solar thermal it should go in the solar thermal side so nobody wanted to fund it there was a big void in between arpa-e is now chosen to fund things that are hybrid systems and so Roland has the funding to work on this and he has interesting design here where he's using a segmented system to reflect some of the lights he has the absorber perhaps it's easier to describe in this one he has the solar cells that are on the outside that generate the electricity and then some of the light is reflected to the tube in the center that collects the heat and then you can flow the fluid through the tube and take the heat out the other end and with this he thinks he will be able to get the higher efficiencies here you can see on the graph on the right how combined you'll get a higher efficiency than what you can get with either the solar electric or the solar thermal by itself and here you have actually some pictures where he's put together a prototype and has it actually functioning on the Sun has been measuring some of the output there there are other things that may happen though as the world is moving forward the aesthetics can be a real selling point for some things you can actually know they're people figuring out how to make the cells colored you can also look at reducing the weight this could be a good selling point here's some flexible modules that are out there another one is when you've put the modules on a roof you find that they tend to run hot when you put them straight as you integrate them into the roof and we find that some of the technologies are actually better at rejecting some of the heat and don't heat up as much finally in the last section of the talk I just want to talk briefly about how the world could be transformed and stimulate your vision of how things could be different in future years so you know that we have the the IOT the Internet of Things and sensors that are being deployed around the world they need a power source if we start small with those and then think big we can start with little devices like this that can collect the electricity the sunshine wherever we need it and generate electricity to power a device but then you can look at also thinking bigger so for example I think you agree that Tesla thinks big and here we have a pig one of their solar roots that they're looking at what if we could make a solar be so cheap that we could just build it into the building materials for every building that we put out there and that it didn't cost that much more to put it in so why not if we had a solar paving the whole world how would our world be different much as now that we have a cell phone in our pocket how is our world different than when we had a phone on our desk so think about what would the future look like if you could put PV everywhere so just in summary here I hope that I've convinced you that we're in a new era Solar has really grown up things are happening differently now than they were we need increased electrification and when I say electrification meaning to move our use of fossil fuels over to using renewable electricity the PV research is not done you see there's still lots of new opportunities and new directions we need batteries electrolyzers and other things introducing a new technology is going to require us to have into entry markets or some special way to sell that new technology because silicon is now so well established but PV can coat empower the whole world and providing entry markets for new technologies and I hope you'll join me in celebrating the success of PB as we move into this new era and thank you for your attention [Applause] well we get ready John's got a mic and I got a mic and we'll get to the questions but I'm gonna start off with one we find people in the audience I've really thought it was a fantastic talk you put a lot of great information that the normal even technically focused person doesn't get so thank you very much for that one question I know that the model crystal --it and the have a decline in performance over time and very good brands maybe do a quarter of a percent per year and more normal competitive six-tenths of a percent per year in degradation what are some of these other technologies looking at or is there enough experience to know about the degradation on them so it's interesting that once the technology has evolved to be a more mature you find that so for example the cigs and the cadmium telluride particularly the cigs in early days was often reported to degrade it's a 3% per year or something but after they put in the effort to make it better they found we found that actually some of the cigs products actually improve in the first few years and then they may eventually start maybe declining half percent per year or something now currently the perovskites the new material I was talking about this material is still struggling to get the stability that we really need but they've made actually a lot of progress in the last couple of years and the hope would be that eventually they will all manage to get to less than 1% per year one of the questions that I have is do you expect to see a simple multi Junction cell based off of traditional mono crystalline silicon with some simple coding to increase the spectrum that absorbs in the not-too-distant future and the reason I ask that is because from a financing standpoint we've seen the cost of solar come down exponentially because of the economies of scale and if we can leverage the entirety of the industry from a manufacturing standpoint and just add efficiency to the current design that will be the fastest way to reduce cost so if you look at the roadmaps that are shown like if you go to China and all the companies stand up and say this is what we're this is the technology we're working on this year and this is what we're going to work on next year this is what we're going to work on in the future way over on the right side all of them say tan themselves is where it's going to go we are going to eventually go to multi-junction cells but if you ask people how are we going to do that and make it work in a cost-effective way that will actually be able to compete with the single junction I think you'll be hard-pressed well the paratha sky people are ready to take over the world and they will tell you definitely that there's no question but what they're going to be able to make multi-junction perovskite cells or maybe peroxide on silicon or whatever and be able to do it at lower cost than what anybody else can do but they they haven't gotten yet most of the efficiency numbers that I showed you there they haven't gotten to the large area modules yet so there aren't really any commercial products out there yet so I wouldn't call them like next year or something but in general I do see multiple pathways to getting to multi-junction cells that could be cost-effective in the long run but I don't see any doing it in the next few years I think that's probably more than ten years away now I could be totally wrong on this but I think it's more than ten years away thank you for all your work on this and your many decades NREL and your presentation is just really impressive my question is what is the basic material and technology for the Tesla shingles that are out there and the long-term efficiency life line of those um so we work with many of the companies across the country Tesla and has been in a little different category maybe you've noticed Tesla just works in different ways and some of the other companies so I have not personally ever seen one in my hand I do read the same news reports that you do they might mean more to me than than to you but in the end they are partnering with Panasonic Panasonic does what's called a hit structure a hetero interface technology that looks at putting a taking a silicon crystal and coating it on the front and the back of the martha silicon layers so we don't know actually if the current products are using the hit structure or whether they are using a more conventional structure but we do know that the plan is to use a high-efficiency version of a silicon technology so they're not going to we can be quite confident that they're not going to gallium arsenide or any other semiconductor it will definitely be silicon but the details of exactly how it's all put together they have been very tight with so I have not seen and no one on my team has actually held one of these in their hands thanks sir the surprising sight for me was the one that showed the value of PV dropping off dramatically 40 I think was a number like 40 percent but also Jacobson that Stanford has been arguing I think for something closer to a hundred percent a PV maybe could you describe how competent people confident people are that that's the right answer there to go to 100% PV the reason that I'm just surprised that the curve looks like I haven't seen so these simulations are assuming California the people in Arizona hate it when we put this up because they say in Arizona we have a very good match between when the sun is shining and when we need the electricity and so in Arizona the curves are much higher much flatter for PV what when I talk with the people about these graphs what they explained to me is that in the state of California at any given time the wind is blowing someplace but in the state of California at midnight the Sun isn't shining anywhere and so this is the problem that comes up with wind if you just put the turbines up higher and you have a big area like California you can find the wind someplace and you may have to transport that electricity around but with the the solar the question is what do you do at midnight Madonn have the unless you can store the electricity or there are lots of other strategies you can store chilled water you can store whatever if you're going to purify water again generate hydrogen if you can find ways to use that electricity during the day to give you value at night then you can move toward really high penetrations it's taking carbon out of the atmosphere so I think that we will want to use all the different renewable energy sources I don't I personally I don't see any reason why we would want to go to 100% solar I mean I think we could go to 100% solar if we wanted to but why do that when we have these other sources biomass I think will play a critical role the question is there if you an interesting question in the long term is to look at the energy payback and will the biomass in terms of the energy we need to put in to harvest the biomass bring it to the plant will we be able to do that and get as much energy back as what we can in other ways or will we need to use fossil fuels or whatever it is to actually make the the biomass work and I I'm honestly not an expert on the biomass I certainly think we should be looking at it but I think that probably looking at a solution that it uses all the different forms of renewable energy will be the best in the long term there are sectors that can be encouraged to spring up like mushrooms in your yard that would curtail the curtailment of very low-cost energy production when it's available so they're working on that right now with the batteries coming down so much in price if people have batteries that can get put onto the systems in places like Australia they are installing and in Hawaii they're installing the batteries much more so I think that that will happen very quickly but what we really need are the drivers and the this is actually something that gets debated a lot and there are those looking at net metering I had four years assumed that net metering was something that would have to go away at high high penetration but as I began to look into all this and understand if you couple net metering with time-of-use rates if we now go back to this graph if you give net metering now it's scary on the one hand for someone who has a PV system if you install the pv system at a time when the price of electricity in the middle of the day is twice the electricity at night and you have net metering that goes according to the the kilowatt hours where you may because the price is twice in the middle of the day then for one kilowatt hour you put into the grid during the day you can take two off at night because earlier it used to be that the the electricity in the middle of the day was highest now when you get to the high penetration levels if you drop the price of the electricity as you see here in the middle of the day and then at night you bring the price back up now you find that you motivate people to invest in all those technologies that we'll be able to use the electricity in the middle of the day and not use it at night so to do exactly what you just said what we want to do is to have both net metering and a time of use rate that shows you when it is we want you to use the electricity so the good news is that moving toward using that metering with time of use rates would do exactly what you just said it would enable all kinds of things to spring up whether it's investment in nest thermostats or or whatever it is but the problem with it for the people with PV on their roof is that if you switch to a time of use profile that looks like this and you keep the net metering you thought you put in a PV system is going to generate electricity just at the time of day when it's most valuable but at high penetration you now have taken a big hit in terms of the value of that electricity so it's a very tough pill for the solar industry to swallow as we move into this new era where our success has gotten so big that the value of PV is now dropping at the higher penetration levels as you see here the value at the low penetration you can see can easily be twice or even four times as much as what it might be at the higher penetration and that's a tough pill to take if you make an investment thanks again for all your years in this industry really appreciate it and good luck in California and that's my question are you going to UC Merced maybe Keith or others know this but it seems like a yes so it's University of California Merced that's perfect from your history and Roland is phenomenal so thank you excellent Sara a huge fraction of the energy used in the United States is is solar is heat is PV gonna be cheap enough to displace natural gas for say heating purposes hot water industrial process heat because I said about 30% of energy used in the United States is just to make heat and nobody is really looking at that market at all well it's it's interesting now that if you put in say a ground source heat pump to use like to heat a building like this you can put the the loops down into the ground when you build the building and then use a heat pump that can have a coefficient of performance of three or four and then use electricity to run that that's actually cost-effective for some sized buildings if you're doing it as a new building we looked into doing it at our house as a afterwards to get somebody to come in and drill down into the ground and do that after the building's in place actually adds quite a lot of cost but if you do it at the time that you're doing the foundation I think that's already cost-effective for to use the electricity relative so you can you can either do the comparison where you're looking at solar to natural gas directly or you can look at can natural gas compete with electricity out there today and then look at whether solar is a good way to provide that electricity and the conclusion is that at least in the locations such as in Colorado that it actually is cost-effective today to put in heating that way now that's a little different than the industrial heat I think the economics depend a little bit upon the the location the temperature the heat you need if you're going to need very high temperature heat you may want to go to a concentrating solar system where you could actually get the the very high temperatures quickly but the electricity because the heat pumps can be so efficient we're actually getting to that point where the electric approach can be superior in many cases Sarah one of the themes of your talk was the futuregrid as we build out more solar and what are your thoughts on the value of a solar system from a grid reliability standpoint in terms of a residential system versus a community solar system that serves several hundred houses what what what is a better benefit to the grid so community solar versus the residential I think it it may depend upon the details of the locality and there are their benefits in both ways in general the larger systems tend to be more cost effective and may be easier to maintain and control but at the same time people like to have them on their homes they should do that personally I have trees in my yard my husband wanted to put solar panels on the roof and I said where and he said well I think over in that corner and I said I think that will work very well and we went out and it was in the shade if we cut down our trees we could install PV panels and also install air conditioning and then we would need more electricity so I personally would say why don't we just put them on the roof of King Soopers over here they don't have any trees around them anyways so but but for the people who would like to have them on their home we're getting I mean in Australia right now it's simply cheaper for people to put in and in Hawaii as well in some cases it's cheaper to put the PV modules there with the batteries and then just have your home own home electric system so that wasn't a very satisfying answer I think well the technology is evolving so fast I'm I'm not sure and I'm not the best person to answer that what we know is that for example you can you can adjust the output of a PV plant faster than you can adjust the output of a conventional power plant and so if you're worried about grid stability having a PV plant out there that's running at say 90% of full power and being able to ramp it up makes it easiest you can do that whether you're using a residential system or a community system but now you're going to have to communicate with say 10 times or a hundred times as many residential systems as what you would do with the one community system and so it partly I with people actually this is a really interesting debate we have actually two choices to make this whole thing work as we go to the very high penetration one is you can set up a dynamic pricing system where the price of electricity is determined both in time and location according to the local availability of electricity at that moment and then anybody who wants to can invest in a technology that looks at that price signal and makes decisions and probably at a forecast of where the price signal would go so you may have on your washing machine the capability to delay the running of the washing machine until some time later and if you had a price signal going into the washing machine to say well the cost of electricity right now is the price is high but if you wait six hours it's going to drop to a third of its current price then you could just it could automatically figure out when it's going to be the lowest price and it can adjust my the advantage of that is that individuals I think like to kind of be in control of their lives and so to be able to have a system where you can go in and decide what's going to happen or how to program it may be advantageous the other is many you probably signed up for a program with Excel that allows them to turn off your air conditioner whenever they want and so the other is to manage the demand side so the only difference really in terms of stability of the grid with the community solar versus the residential is how many signals you're trying to coordinate with are you trying to turn off air conditioners in 20 different homes or adjust the output of 20 different systems are you trying to do it for one system I think that fundamentally the physics isn't really that different between the two there I'm sure there are some differences but it's more actually a challenge of figuring out how you decide who's going to turn off who's going to adjust to this current problem and that all needs to be automated and handled by the inverters anyway you know first a comment that you were speaking earlier about the ground-based heat run by electricity that this church building is part as part of its power from that source my question though has to do with kind of the other end of the scale which is PV on the really large scale like I read articles in Scientific American and whatnot about the possibility of huge PV farms you might call them in Arizona and you know in the southwest that would then potentially feed the power needs of Los Angeles and you know other places that would take a course a grid that could handle it and so I would like to hear your comments about Grid infrastructure and what in general we might be needing either for that or for other applications the analysis shows pretty clearly that transmission is cheaper than storage however installing the transmission can be a political nightmare so if you want to put it across and you're going to get have to have a hundred people who've maybe lived in in that house for generations and now they have to move that becomes very difficult but the the all the analysis I've seen is that at least currently now if battery prices come down as just the the solar panel prices have come down that could change but currently the transmission is cheaper that s her question leads right into mine we had a presentation here half a year ago the National Grid future plan and they were talking to transmit DC two million volts to five million volts across the country which would line up with a lot of future DC things are you involved with transmission future plans no I'm not personally involved in that okay if oh one more question right here thank you dr. Kurtz for being here today I'm sure everyone appreciated I'm not nearly as smart as any of the rest of these people here so bear with me for one of the questions I have about the concentrated solar the two production plants I believe that had been completed in California had great difficulty in ramping up their actual production level that they then ended up having in order to because of the contract they had with the utility in order to to deliver the appropriate amount of electricity they had to go in and install at extra cost a gas fired plant right next to the concentrated production because they had to be able to get that so it's like the opposite of the curtailment and I'm wondering if people are actually looking at these things upfront and preparing for them instead of installing and then going whoops we blew it okay and then the second question that I have for you purely because I don't know one of our biggest costs now in the solar and in the solar installation of itself is not necessarily the panels anymore and if you can increase the efficiency 10-15 percent that isn't going to decrease the panel cost all that much dramatically so our bigger cost now is the actual installation the racking costs that type of things so I'm wondering if the enroll was any kind of studies into reducing the cost of installation because ultimately that's going to make it more profitable for people to put it into them thank you yes so both of those I think it is quite normal for any product that comes out in the market that the first ones that are made always have some problems the big handicap that CSP has had over the the last few decades is that while PV works very well in small size systems so that we could get experience on a thousand small systems get all the kinks worked out all the bugs worked out so that when we went to the large systems that was quite robust with CSP you can't really make just a little tiny system you have to make a really big system and so if you look at the statistics there's still kind of in the infancy of making the first I don't know what the statistics are but you know the first two dozen systems or something they haven't got the kinks worked out because they haven't had that luxury of making a thousand systems and you probably need to make a thousand systems before you figure out all the difference the ways it can go wrong so the key thing that's different now for CSP and where if China manages to be successful which I don't know I was just at a plant where there they're putting together a test system and they view it as a test system they don't view it as a fully-developed they view it as a developmental sort of thing that they'll be looking at what works well and what doesn't but they're doing it at large scale now and it's not a big deal they've got they're putting in gigawatt sized fields of PV so to put in 100 megawatts or something of CSP doesn't feel like it's such a big thing whereas 20 years ago when they put in those CSP plants it was a big thing to in just those first ones and I don't think that it's realistic for us to expect that the first one we're going to put in won't have any problems and that we'll figure it all out before we start in my experience research doesn't usually work that way now going to the moon they had to make it work the first time but I remember actually being on risk mitigation panels they spend a lot of money on trying to think of what are all the ways it could go wrong so we've tested everything and and I don't think that that we're going to be able to do that in a cost-effective way with the plant like CSP then on to your second question yes we we've been moving over toward better understanding the the total system and there are lots of different ways that the costs of the at the system level can come down a lot of that is just for the installers to learn how to do it more quickly and more effectively and there's a lot of innovation going on if you go to the say the solar power International Exposition and look at the all they they they have all the different ways of mounting things and ways to put things on the roof ways to Brett both the racks together and it used to be that you would have like a little washer and then you would slide it into a rack and then try to get the bolt screwed in and you fiddle with it for like a minute for each of the different connections and now they figured out ways you just do it and go right down the line and so those things are coming the u.s. is still way behind Germany but I think we'll get there so we've had a number of different projects there I think that that's a fairly technical question that maybe I could take offline to talk with you later and I think we'll we're going to go out in the lobby I'm going to say just one thing to punctuate the soft costs having been in the hailstorm and having a 10k system I know the bid to get that off and back on and this doesn't involve the inverters it's just the panel's the racking off and back on it's 80 200 bucks for a 10k system or 80 cents per watt for the off and on just a one bullet point and so thank you so very much for coming and joining us and sharing your wisdom [Applause] you

Channel: Colorado Renewable Energy Society (CRES)

Views: 18,736

Keywords: solar power, photovoltaics, NREL, Sarah Kurtz, National Renewable Energy Laboratory, DOE, energy policy, solar thin film, multi junction cells, electrical grid

Id: MKzHSY6U2j4

Length: 84min 5sec (5045 seconds)

Published: Thu Sep 07 2017