Trichlorosilane

Transcription

[MUSIC] So this is the agenda and structure of my talk I wanted to cover today. first, I will describe briefly Solexel and our value proposition. then, I think it will be very helpful to to give you an overview of Crystalline Silicon PV Technology landscape and, and, a little bit of overall PV market data. our focus is thin silicon and, and really. Looking at crystalline silicon, it's, you can call it ultra thin because we are, we are operating below 50 microns of crystalline silicon absorber. So I'm going to explain the motivation behind using thin crystalline silicon as it relates to high cell and module efficiency. then we will do a little bit more of deep dive into Solexel's disruptive technology and, and our progress at the company. And then I will. Leave you with some, takeaways, and, and, a conclusion. So, a little bit of, overview on, on Solexel. I started the company, over six years ago, and, and then incorporated, and, close the Series A financing in the first part of 2007. the Series A financing was led by Kleiner Perkins. Professor [UNKNOWN] has been involved from from very early days as a special adviser. we do have a fully operational pilot fab. it's a pilot engineering transitioning to production fab in, Milpitas. along with complete tests and reliability engineering tests lab. So, our technology platform is very IP-rich. and we do have a, a portfolio worth 170 inventions. they are primarily mostly in-house patents. we have also license to. portfolio from some key institutions. besides the venture financing we have received $17 million of uh,r and d grants, mostly from DOE, Department Of Energy. Actually, the last of these we received less than a year ago from DOE as part of the, sunshots program. it's a $12.9 million program based on which we are developing, we are applying our technology high efficiency technology to building integrated photovoltaic cells roof top shingles working with our partner on [UNKNOWN]. we have also received a significant incentive package from Government of Malaysia to to set up mass production lines there. so far we have raised $150 million in three rounds Series A $50 million, followed by Series B $98 million, and now Series C is in progress. We are about 2 3rds of the way through and, and the proceeds from series C We are using towards demonstration of the high volume manufacturability of our technology on the coffee exact high, high volume manufacturing tool set. So, Solexel as Solexel are, are. Technology actually has a unique attribute. We essentially eliminate the performance versus cost trade-off that has been governing and constraining the mainstream traditional PV technologies. on, on one end of the spectrum You have thin film leader First Solar, with the cadmium telluride technology. relatively modest modular efficiency today, about around 12.7%, and at least up until recently the lowest cost in terms of module cost, manufacturing cost per watt. A level 72 cents per watt this year. Efficiency is relatively modest, so really the ASC with that level of efficiency don't command any premium. On the other end of the spectrum we have a company like Sun Power, the, provider of high-performance modules with industry leading efficiencies today at the maybe over 20% and by next year actually moving to their height efficiency 21.2% panel efficiency but at manufacturing costs, which is you could say significantly higher on a per watt basis than the best of thin-film. on the other hand a company like SunPower is able to command the. Premium on their pricing. So the ASV, compared with thin film, is much higher and that is because each percentage absolute efficiency point has a value of five to seven cents per watt. At the installed system level. so, so, basically let's say a seven point or eight point efficiency advantage there. At the installed system level at the minimum has a value of $.40 per watt. So, you can actually take some of that as a, as a pricing premium. So. What and then you have the, the main crystalline silicon population, which is somewhere here, and, and and I'll get a little bit more into that. what we do at Solexel, we really have a platform technology that enables us to to combine the best of both worlds. So, we our architecture enables us to operate in the essentially elite high performance group, in terms of efficiency like Sun Power, but at a cost as we scale that is. Actually significantly below the best of thin film, in this case for solar. and, and really the, all the very important attributes are always efficiency, manufacturing cost, energy yield. And I'll get into some of the other very interesting attributes for our technology, later on. So, almost 90% of the crystalline silicon, 90% of the PV market is served with crystalline silicon PV. And and essentially maturity of the manufacturers go through the same macro. Flow and, and supply chain as I've shown here, so you start with purified silicon source gas typically Trichlorosilane or, or silane. And then from that you form, manufacture poly, poly silicon. from poly silicon either monocrystalline ingots which [INAUDIBLE] or multicrystalline silicon with casting technology and, and then those are sliced into wafers using water saw. you have to remove the water saw damage and then make the cell, the cells are assembled into modules and the modules are ready to be installed in the field. So, looking at the industry costs, essentially today, the leaders tier one manufacturers, the, the modules now costs a little bit over 70 cents, 70, 72 cents per watt to manufacture and, and at the installed system level that today enables an installed system cost which is a little below $1.60 per watt. Now. if you take the existing way of making the cells and modules you're going to be essentially limited by the, by the costs of materials at the cell and module level, without introduction of any disruptive features then you can have the floor on the, on the cost, that might be somewhere in the range of 60 cents per watt. And, and that would enable maybe a reduction on the order of another 35 cents per watt going down to $1.25 with the mainstream technology. So, if you look at this flow. The, these three steps, polysilicon, ingot, and wafering, actually are the most costly and energy intensive steps. they account for almost 40% of the module manufacturing cost. And, what we do at Solexel, we actually address, beside the cell architectural innovation that I'll briefly describe, we also address these three steps. We eliminate these three steps, and, we we replace that with a direct-gas phase thin silicon growth. And this actually, in effect makes us Independent of the traditional silicon supply train. With input to our factories is purified We actually substantial reduced net amount of silicon that is consumed in our cells by at least a factor of ten. Significantly lower Two to three x lowers capex. If you compare that with the same value chain for the mainstream crystalline silicon, less energy usage that enables us to take the energy payback time in most installations to well below a year and, and make it comparable to the best of Kim Prong. And, and essentially it enables us, as we establish scale, to achieve, actually an installed system costs are 80 cents per watt. This is because we have a clear path to a module manufacturing cost with this technology, that are around 40 cents per watt at scale. So the balance assistance costs includes everything typically what, what we're what I'm showing here is also sometimes is separating inverter from balance assistance installation. In this case this is for large commercial installation it includes everything. So I'll, I'll get into that because one of the key factors there is is high efficiency and the other part is, is that we actually not only can make ourselves standard glass-covered modules but very lightweight Flex modules that don't have glass. And actually this significant reduces the, the balance system cost as well. So, let me move on to uh,give you an overview of crystalline silicon PV technology. A little bit of update on the overall PV market as well So if you look at the global PV demand, this is you know, looking both at the actuals and also a circle actuals and also forecast the PV market has been growing and is projected to grow at, at essentially compound annual growth rate of about 40 percent. Now. You cannot find very many technology in the technology sector, very many sectors growing at this at this healthy rate. This is the growth in terms of gigawatts of demand. this is actually despite the challenging conditions in the, in the industry since since 2009 and one of the, if you look at the size of the market this year, it's projected to be around 38 gigawatts and over the next four years again projection is that it will be essentially almost doubled. One of the interesting observations also here is that you see that market demand is transitioning from a Dominate the market to a market that is more globally, where demand is more globally distributed. So another way to look at this is look at the cumulative global installed base of PV. everything: crystalline, silicone and, and a key milestone that is forecast to be achieved this year is that the global install base is going to each about 100 gigawatts and and if you look at the really portion of market share of thin film it, it's it's a fairly small fraction of the overall installed base, less than 15%, and at least it's not projected that any time soon thin film non-crystalline thin film technologies are going to challenge crystalline silicon type technologies. But when I say KIM form typically KIM form means reference to anything, although we are you know, strictly KIM form but we are categorized as KIM form crystalline. I'll get into that a little bit more, but KIM form is really Catel six. more for silicon organic, PV you know, essentially non-crystalline semiconductor category. And, and really you know, the installed base reaching about 100 gigawatts globally this year. If you look at the PV opportunity, it is truly a multi-terrawatt opportunity. And I, you know I believe at least when I look at the, two, three, four decades ahead, at least a ten terrawatt opportunity. So this really this really represents in my view you know, basically 1% of what could be the the market opportunity for PV capacity globally. So really another interesting trend to look at is the progression of cost in in crystalline silicon PV. you know, any companies that started back in 2006, 2005, 2007. With the assumption that you know, polysilicon will be 50, 100, 200 dollars a kilogram. You know, a lot of them are not around today. So just looking at the just the last four years. If you look at the cost of, compared to the spot prices today. If you look at the cost of polysilicon, it has dropped by a factor of 20. And the cost of wafers It has dropped by, by a factor of ten. So you know, today, you can buy full-sized RAM 226-format wafers and on a dollar cost and these used to be just three, four years ago were over ten, ten dollars per wafer so. The spot prices for modules are below $0.70 per watt. in many cases actually these are below the actual cost. maybe it represents cash, cash cost. So so this has been really great for for acceleration of PV cost reduction and, and grid parity. Of course it has also caused a lot of pain particularly in the last three years in the, in the industry. But overall it's, it's a great acceleration for the industry. So this is also an interesting trend of the module and install system cost and, and the module cost. The install system cost, essentially the total install system cost represents the module cost, again, for commercial-scale installations and the balance of system costs, right? If you have higher efficiency, the balance of system costs goes down. If you have you know, lightweight, flexible modules that, that you know, require less mounting hardware et cetera, the balance of system costs goes down. So one of the really interesting milestones being essentially reached this year is that you see that silica thin film. Primarily, primarily for solar cathode modules, has been lower cost, in terms of module cost, and it's whole system cost crystalline silicone. This year, actually the trend has been reversed, so crystalline silicone module cost excuse me here. The crystalline silicon module costs on the average, is lower than ten form. The balance ten foam and the balance the total installed system cost. And, and, the trend in terms of installed system cost is expected to continue. So, they both go down where you going to have to. advantage in the case of crystalline silicon technologies. So, so let me switch gear from market now to technology. if you look at, and again the focus, my focus is on crystalline silicon technology. Look at the terestrial of solar radiation. Basically, with silicon, and again all the focus is on single junction, silicon solar cells. When you have photon energy less than silicon band gap, essentially those photons don't make any contribution to to your short circuit current or to internal quantum efficiency. The portion that really contributed is any photon energy above band gap, above 1.1 of E. Now any excess photon energy above band gap gets thermalized. So really that, the, the amount that can be effectively utilized and converted to electricity is shown in, in green here, which is essentially the, the portion that, that has energy above Van Graaff and also converted to electricity. So, it's also good to to know what soem of the key measurs and metrics are for crystalline silicon cells and modules. many of thse are acutally same metrics are shared almost all of them in the case of thin film modules. You hear a lot of short circuit current or current density, open circuit voltage and maximum power point which is typically where you operate the panels at different times during the day. the product of maximum power point, current over voltage is really your your maximum power that you extract from the panel. And fill factor, which is also a very important parameter. When you multiply fill factor by short-circuit current density and open-circuit voltage, that's your panel, or cell efficiency. Fill factor is a measure of ideality of your I-V, so the squareness of the I-V. And and it's the ratio of your maximum power divided by product of short circuit current and open circuit voltage. conversion efficiency, very important, the number one [INAUDIBLE] in terms of importance in PV cells and modules. Another important factor also for energy yield is temperature coefficent as you rise as you increase the temperature of the cells which is typical in the field operation how much power loss you have in your panels the less the better. Normal operating cell temperature, the cooler the better, again in terms of energy yield. internal and external quantum efficiency. This is really a measure of that, that the fraction of either absorbed photons or incident photons that are Converted to, or contribute to electron hold per generation are, are converted to electricity. And last, but not least, is energy yield. Very important, metric for installations in the field. So, Short circuit current or short circuit current densities is a very important parameter. It is again, looking at single junction, it really, it, it's directly dependent on the band gap of the semiconductor. Now, as you reduce the band gap of semi-conductor Then more and more of the photons in the solar spectrum are available to generate electrical current, so that's why, as you use the bandgap, the sthort circuit current density, max storage co-value increases now but on the other hand the trend is actually opposite for open circuit voltage and also fill factor so. Now if you look at the, for instance this is for germanium silicon gallium arsenide, theoretical value for silicon, the maximum Jsc, or short circuit current density about 46 million per square centimeter. as I mentioned, Voc, field factor have an opposite trend so actually when you go to a larger band gap. Semiconductor. Those values increase so you can expect that there is there is an optimum or there is an optimum range for Vanguard that gives you the maximum efficiency that's really predicted by the the classical shock equalizer efficiency limit. And and you see that If you look at the maximum efficiency, this is essentially thermodynamic efficiency as a function of bandgap. there is a, there is a maximum. About 33%. This really doesn't take into account some of the more practical limitations that I'll cover shortly. basically going to lower band gap, higher JSC, going, going to higher band gap, higher VOC and typically also field factor, and fortunately, actually when you look at this silicon and gallium arsenide in terms of max storage, total efficiency, pretty much fall on the peak or near the peak efficiency capability. So, so that's good news for both silicone and gallium arsenide. Of course the difference here is the silicone is an indirect band gap, gallium is direct band gap semiconductor. So you can use much less gallium arsenide. In the case of silicone you can, you can make it very thin but light management and light trapping becomes extremely important.