 Well John, I cannot take credit for what the Chinese did to bring the cost of silicon solar cells down, but thank you anyway. So yeah, today I'll talk about something new for my research group. We've been having a lot of fun making windows with dynamic tinting, and there's really no better way to start than to show you the windows. And on the left it's in its transparent state, and then you see that we can put the window in a partially tinting state, and it's got a nice neutral gray color to it, and then you see that after about a minute the the window absorbs almost all of the light, and it's even dark enough for privacy applications. We're doing this by electroplating metal, which is very effective at blocking light, and then you see that we can strip the metal and make the window go transparent again, and we've cycled these windows up to 5,500 times without observing degradation. So that in a nutshell, that's what we've done, but let me step back and motivate the application and tell you the other approaches that people are taking to it. So that one's one of my least favorites, the metal slat blind. I've never seen a science fiction movie in which the spaceships have metal blinds in them, and it's hard to imagine that in 10 to 15 years we'll be going around in self-driving vehicles and pulling on strings to rotate metal slats when we have too much glare in a room. It's not about as ridiculous as putting that on your glasses and using it instead of sunglasses. It's obviously much better to just adjust the tinting, and you see an example of this over here, and the great thing about that is that you can still see through the window. Blinds and curtains block your view, and we have a beautiful campus here, and so many people have their blinds closed because of the glare, and their view is lost. So it's clearly preferable to just adjust the tinting without blocking the view. From an energy savings point of view, studies have shown that you can save about 20 percent on the heating and cooling and ventilation costs, and also lighting. You can optimize the flow of lighting into a building with this kind of a technology, but I think you can't even fully capture the value here. People just enjoy natural lighting, and the future is certainly towards having more and more window area in buildings, and a solution to glare is certainly needed. This is from the View website. They're one of the leading companies in developing dynamic glass, and you can see a window switching here. One of the things that I think is really crucial for saving energy is that it can be automated and practice on a cold winter night. We could all put our window blinds down to hold heat in at night, but I think it's safe to say that approximately zero people on our campus do that on a nightly basis. So it needs to be automated, and if you look on this video, I didn't think it was appropriate to play a five or six minute commercial here in the seminar, but you can look at it yourself, and they talk about how you could have a light sensor on top of a building, and a computer would have the data from that. It would also have weather data, it would predict when clouds are coming, and it would know the layout of the building, and it would very intelligently optimize the tinting. First, prioritizing comfort for the people inside, minimizing glare, particularly on computer screens, and then prioritizing energy efficiency. But of course, you would always be able to override it and get whatever tinting you wanted. So since our work has come out a few weeks ago, a lot of people have been asking various questions, and the number one question is, how are our windows different from photochromics? So I'll start there. I'm wearing a pair of glasses with photochromic lenses. They go dark when they're exposed to UV light, and so it's great. I go outside. My glass is darkened. There's no switch, no power or anything like that. Those are advantages. However, they won't darken in a car because car windows have UV filters, and so there's no UV light to trigger the glasses to go dark. You just don't have control, and they're not dark enough. If you were to try to go skiing with a pair of these glasses, it's just too bright. The glasses don't have a wide enough dynamic range. For buildings, you really want to have that control, and so photochromics are not a great solution. Technology that's already being used in high-end cars, I think there's four companies, including Mercedes, that puts dynamic glass in their skyroves, and some of them use suspended particle devices, and here they're just particles in there that can scatter light, and with an applied field, the particles will line up, and light is able to go in between some of these columns of particles, and so it works reasonably well, but it's inherently hazy. In both of these states, light is scattered somewhat, and it's not ideal for skyroofs, and it's just not acceptable for the windows in buildings. Another technology that it's been available for at least 30 years, but I personally have never seen it, which means that it certainly hasn't caught on very well, are liquid crystals, and these are molecules where the refractive index is different in one direction compared to the other, and if you line the molecules up, you can get a state where light goes through unscattered, and it's clear, and in the other state it scatters, but it really doesn't work for energy efficiency. It's just for privacy. All it really does is scatter up the light, so you might use it if you want privacy in a conference room, but it doesn't really solve the problem that a lot of people want to solve. The most promising solution that the biggest companies are pursuing are so-called electro-chromics, and these are materials like tungsten oxide, that change color when they're reduced, and so you can have a device where there's two conducting electrodes, something like indium-tin oxide, and on one of them you have the electrochromic material. On the other, you also have a material that's capable of being oxidized and reduced, and then you have an electrolyte in there with the ions, and you can apply a voltage and put the lithium into the tungsten oxide and darken it, and then when you want to, you can apply the opposite voltage and make it go clear again. The biggest companies are using tungsten oxide, and it's a very durable material. Everything in their stack is inorganic. Tests they like to run are 50,000 cycles under one sun at 85 degrees C, and they are able to pass that test, and they're expecting to have 20, 30, 40-year lifetime based on that accelerated testing. But I think if you ask the question, wait a minute, they have it. They have a window that switches well. Why don't all the buildings have this? A big answer is that it's around $50 per foot squared, and a lot of that is all of these layers are sputtered in high-vacuum tools, and there's microns of material there. Also, if there's a pinhole, the window is unacceptable. You just can't have a spot that doesn't change color. It's too noticeable when the windows are in their dark state, and it's challenging to make large windows in a sputter tool. Not impossible, but it does bring the yield down, and ultimately I think that adds to the costs. And so that's probably the main reason that the technology hasn't really exploded. There is a lot of interest. Three of the biggest players are Conestral View and Sage, and Conestral has raised at least 100 million view, up around 650 million. Sage Glass is also very large, and they were acquired by Saint Gobain. There are other smaller startups as well. So you can see there's starting to be some excitement, but we were also very surprised that hardly any professors work in this space, and this is a 40-year-old technology, and it does have some inherent flaws. It's not color neutral. The windows let through a little bit of blue light in the dark state, and a lot of people we've talked to find that unacceptable, and the expense is relatively high. Arguably, the most successful company is Gentex, and they use an electrochromic molecule, methylviologen, and I think they made a very clever choice of where to start their business. They do rear-view mirrors that have a sensor, and if someone is blinding you from behind, it'll sense that, and it'll darken the mirror, and they do about 1.5 billion in sales. Someone told me this, and this is John Reynolds, and I said, John, how come I've never seen this? And we walked outside, and it only took him about 20 seconds to find these cars. If you don't have a lever at the bottom to rotate your mirror, then you probably have this technology in your car. And then they're also doing the windows in the Dreamwiner, so if you've flown one of those recently, you probably noticed that there's a switch, and you can adjust the tinting here. People really like this, again, because you can still see through the window and get the view, even when it's in the darkened state. And the pilot has the ability to make all the windows go clear, which they're required to do by law when they land the airplane. And if they want the cabin to go dark, because most people are trying to sleep, they have the ability to make the whole cabin go dark, but then people can override it with their own switch if they want. I made my trip home from Europe a little longer than it needed to be a couple of months ago, so that I could get on a Dreamwiner. And the other passengers remarked on my enthusiasm for the window, and I took quite a number of pictures and videos. And I would say there's a lot of room for improvement here. The window takes minutes to switch, and so everyone just pushes the button over and over and over again, because they don't think that the window is working. And I don't know how they got that photograph, because it's not that dark. It didn't seem anywhere near that dark when I took the pictures. And it's clearly blue. In fact, this is what my hand looks like when the window is in the clear state, and that's what my hand looks like with the filtered blue light. They only get away with this because the sky is blue, and people, you know, it's not so ridiculously noticeable. So they're clearly, and we've talked to window companies, and yeah, they need a better color to put this into homes and to put it into office buildings. So to overview where the needs are, you want uniform switching, and let me elaborate on why this is a challenge. And I'm going to, you know, full disclosure, we're going to show you that it works awesome at a small size. It's challenging when it's large, because when it's large, you're drawing current, a lot of current, and that means you have a voltage drop across your transparent electrode. And when you don't have the same voltage everywhere, then it doesn't switch at the same speed everywhere. And so right now, you know, the really large windows, they take about 20 minutes to switch. That would be a good problem to solve, although we're told that for a lot of buildings, people would accept that. However, for cars, they need it to switch very quickly. And so, yeah, speed is nice, but the high contrast and color are really important. And particularly, sometimes you need privacy. Sometimes you don't. If you need privacy, you have to go below 0.1%. And a lot of the technologies are not able to go that low. Of course, you need it to be very durable. You'd like it to not draw power unless you're switching it. And then it's very important that they're below haze. You want a nice, clear view. So we decided to take a completely different approach to solving this problem. Metals are very attractive for a blocking light. Only need 20 to 30 nanometers of metal to almost completely block light. And then I worked on organic semiconductors for 20 years and tried to build solar cells and make them last for 25 years. This time I decided to use metal. Metal can handle UV light. It can handle heat. It's not going to degrade. And you're going to see here that we just take a transparent electrode and inject a gel electrolyte on it. So it's very, very inexpensive. And I'm reluctant to put out cost numbers at this stage in the research, but I think it should be significantly cheaper than an electrochromic device with several microns of sputtered material. So Chris Barile was a postdoc in the group. He's now a professor at the University of Nevada. And he brought the electrochemistry expertise that our team needed to be successful. And he put together this table of candidates that we could use. And at the bottom you have some metals that would be easy to reduce. But Chris pointed out they can oxidize, and they're not so reversible. The opposite of that, you'll never oxidize gold or platinum but you can probably guess why we don't want to use those. They're very expensive. And in between are some metals that are attractive in every regard. We decided we'd start with copper because people probably know more about electroplating copper than any other metal. But copper, as you know, is reddish and we wanted a neutral color. So we needed to put another metal in. We started with lead because lead is very well understood for car batteries. And we got the message very, very clearly that people did not want lead in their windows. So we moved on to silver. And most of what I'll show you today is an alloy of copper and silver. I think this SEM here is actually copper lead. And these are ions in water. And one of the first things we learned is that if you try to plate these metals on ITO you don't get perfectly uniform coverage. If you look in an SEM you see wide spacing in between the crystals. And ultimately the light just goes in between the crystals. And so it doesn't work great. And we also found that as we would cycle it more and more we'd get a very different morphology. And so if we just kept cycling it, plating for 60 seconds, stripping for 60 seconds, the minimum transmission would change considerably over time. And so basically that first generation didn't work very well. And then Chris added some platinum nanoparticles using this molecule here with carboxylic acid to anchor to the ITO and a thiol to anchor to the platinum. And we got dramatically better results. And if you ask the question maybe why didn't people do this a long time ago, people did do this a long time ago but they didn't use the platinum. And the platinum is really helpful for a lot of the properties. And now you see we have a much higher nucleation density. And as we're plating further you're getting more and more metal and moving to a more opaque window. And now when you cycle it a thousand times you see that you're getting a very consistent morphology. We believe after each stripping cycle we've completely removed the metal and then we're starting all over again. And so here is the when we do an experiment where we plate for 60 seconds and then strip for 60 and plot the maximum transmission and the minimum. It's very stable. There is some conditioning of the electrolyte that occurs at first and then it's very stable. Later I'll show even more cycles. Right now it's not yet a window, this is actually in a beaker and I'll show you the window in a couple of slides. Here is the transmittance versus time and you see it just as we plate more and more metal the transmission goes down so you can just stop when you get the transmission that you want. And an important thing is we can go really quite low by plating a lot of metal. With the electrochromics there's only so much lithium that can go in and then you're done. And if you try to make an extremely thick electrochromic to get high darkness it's getting expensive and it's also getting hard for the lithium to go into such a thick film. Here you see the spectrum. So up top that way that's essentially just the spectrum of the of the indium tin oxide electrode and then as it's getting darker we're keeping a nice flat spectrum. And really the cover slide you saw it's a very nice black color. And then as we strip it you just go right back to the original transparency. Over here you see what happens if we plate metal and then remove the power source nothing happens. The metal stays put and then after a day we stripped it and and of course the metal ions will not just spontaneously plate on there. And so it draws no power unless you're actively switching the device. So it's a very very low overall power consumption. Here's how we the process flow to make a window. You take the ito coated glass and put a rubber edge seal on there. It's probably not the perfect material but it's what we use to package solar cells and it worked pretty well for a prototype. We put some copper tape around the edge here and put more rubber on. Put a second piece of glass over that. Inject the gel electrolyte. Here we're using as a sort of as a thickening agent high hydroxyl ethyl cellulose. And I've been informed that this is hair gel so I can no longer say that I've never used hair gel. Just don't use it for what most people use it for. And so that's how we make the the window and here's the video. I think this is 2x speed so you see the window go dark and then we'll strip it here in a little while and there it strips. You can see it did it it stripped at the edges a little bit faster than it stripped at the center. And here's some of the optics measured by Dan Slotkovich and the here's the transmission. And so in the in the clear state it's it's fairly clear in the visible out in the infrared the the TCO blocks. In fact the just the TCO is is sort of a bare bones low e coding. And then in of course we can plate for different amounts but for this one particular plating we've got a fairly flat 10 percent transmission. And then we wanted to know are we reflecting the light or are we absorbing the light? And the answer is we're reflecting a little bit you know maybe 10 percent but mostly we are absorbing the the light. And it's not completely obvious why that would be the case. I mean after all mirrors are made with metal. But because of the roughness and the nanoparticle nature it the light is likely being scattered a lot and there are many plasmonic resonance resonances in there. And so we're mostly getting an absorption. And it it it there are some people who have said well we you know don't you want to reflect the the light and keep the heat away. But it turns out that for most applications there are some limits on how much reflectance like in Singapore you're not allowed to reflect more than 30 percent. And so in a lot of ways absorption is preferred. Originally like when we conceived of the project we used Beers law to calculate how thick the metal was going to have to be. But Beers law really turns out not to apply here. Beers law would have been the red curve. And the reason is most of the light that's getting through isn't going through the metal it's going in between the metal particles. And taking into account how much area we had between the particles and assuming 100 percent transmission there we were able to come up with a simple model that was able to explain the the data. And I would say there's good and bad news associated with that. The the good news is I I think our windows look more uniform because the Beers law curve is really steep and you can see that if you had just one nanometer more metal in one place than another you'd have a noticeable transmission. So I think the shallower curve makes the window look more uniform. On the other hand if we if we could plate uniformly we would block light more effectively and that would mean we could plate less metal which would mean less current and that would mean less voltage drop and that would mean less irising. So we would be able to switch the window you know bigger windows faster if we can learn how to plate the metal uniformly. So that's something we'll think about. And yeah here's the durability for our copper silver gel electrolyte window. After 5500 cycles we needed the potentiostat for other experiments. We weren't seeing any any signs of degradation with with cycling. We still need to do things like higher temperature and UV light but we don't seem to have the problems that you might expect with batteries and the reason in a battery you would be repeatedly intercalating lithium in and out and and you keep expanding and contracting the material and you can get failure. We don't have anything like that here and we also don't have a metal that we're completely tearing down and then replating. So it seems to cycle quite well. So to wrap up compared to other electrochromics we think some of the big advantages are that it's gray in the partially tenting state and then black when it's fully opaque and I'll just say for now cheaper. I'm not exactly sure what the cost is going to be but I think it'll be well under $50 per square foot and I think we can get a higher range because we can make our windows darker than most of the electrochromics. So I think sort of the challenges that we're working on now I'm not going to say how but we're trying to figure out how to make these windows switch really quickly when they're large without irising and we need to do all the long-term stability tests because these windows will need to last 30 to 40 years. So I really want to thank an excellent team for doing all of this work. I've never actually seen them wear these sport coats and ties but maybe they put them on or maybe they photoshopped it but I think most of them are in the audience and Chris is now at Nevada and there'll be some new chemistry we'll announce next year that Tyler and Michael and Teresa have been developing this summer and I really want to thank the Stanford Precourt Institute for Energy for funding this. National Science Foundation did not get excited about this and two of them have fellowships and so we were able to really go a long way with a little bit of money and now that it's working people are a lot more excited about it so we really appreciate the seed grant. With that I thank you for your attention and look forward to answering your questions. So thanks for leaving so much time for questions I think that's great. By tradition we actually start with student questions so any student questions, comments? Sir. In leasing micro battery and there will be for when you are playing a micro on the actual there will be a dangerous problem and the danger you are causing the two reactors is this problem existing in the smart computer? No we don't see dendrites and because we didn't see them I haven't had to learn that much about them but my understanding is there's some layer that grows on the electrolyte that his name I can't remember and sometimes you can punch through it and you get the dendrite we don't have any such layer and you saw our SEMs you saw the you know the way it grows. Way in the back. Mike how good is the control between on and off? Can you stop at some partial transmission value reliably? Yeah we can stop anywhere we want along the way in an application I don't know if well you could you can integrate probably what you would do is integrate the current and that would be telling you how much you had deposited possibly there'd be a sensor but probably you would integrate the current to know when to stop and then a little bit of a problem if you go let's say you go down to 10 and then you want to come back up to 40 we have to go back up to 90 and then down we need to strip it and and and start you saw how it iris now of course even that we stripped too fast maybe you know maybe if we worked on the right procedures we could work that out but right now we would take it all the way back clear and then go to the new setting. I have two questions the first one is what is the highest level of transmission I mean the maximum transmission versus the minimum transmission that you achieve actually with this device you have shown some data where when you did the 500 cycles the data going from 80 percent down to 20 and towards the end we saw actually a slight rise of the of the transmission you know on this curve so what is the lowest level of transmission because you said you you could achieve a really dark yeah so what is the number there I mean Dan what's the lowest we've ever gone we go because if we play long enough we just basically have a solid layer of metal right and as long as you have a solid layer of metal you're below 0.01 percent okay but but in this case what would be the higher the maximum transmission so it's going to be the transmission of ito and you know so around 90 and that yeah when we I mean we've only done a handful of these cycling tests and just to kind of make it run a little faster we only darkened it down to about 25 percent and we just haven't had time to do a cycling where we're going down to say one you know what we need to do is get a whole bunch of channels you know so that we can test a large number of of devices my second question is I'm sorry maybe my second question was related to so I'm sure you don't have the answer for that but just trying if you take a larger sample which you don't have now what would be the how how would it darken I mean would you see actually going from your burst bar on the side towards actually the center or would it be actually too fast to show actually any non-uniformity in the way it darkens well if you just take our window the way it is right now and you make it a meter by a meter and and you don't change anything else and you try to run it at the same conditions it's probably going to plate you know roughly a few inches on the outside and it's not gonna it's not gonna plate very well in the center of a window that large so you're gonna have to do one of two things you're gonna have to run the window slower and and accept a lower current so that there will be a smaller voltage drop and that that's what most of the companies are doing with the conventional electrochromics or we need a significantly better electrode and we're working on that up here on the left and then back yes okay there is quite a bit of absorption across the entire light spectrum so my question is the the absorbed energy has to heat up the window and and the window heats up when the transmission is very very low so it would we radiate like at least like a black body equally in both directions of the window so the space inside the window in the building does get heated up so have you guys measured the temperature rise of the window as a function of transmission no we haven't I'm I'm sure that the the companies that are further along have done that a lot and I have heard that they will often in it they'll have a double pane window and the dynamic part part will be with the pane of glass that is on the outside and then they'll put a low e coating on the other pane to to catch that radiation we had one back up on the left and then down here tell us about the changes going on as a counter electrode right so it's I'm I mean right now it's it's it's copper tape out on the edge um we have made some bigger windows and we we use a like a copper mesh and that works really well and you actually it's fine enough that you don't you don't really see it very well um it yeah you you you you can't just have the counter electrode be on the edge for a one by one meter and so um yeah we're working on a counter electrode grid um that um will be very hard to see and and we've we've we have metal lines that um I mean you you you look at them really hard you have to get the lighting at just the right angle um you know to be able to see them um and and it you know so it can work as the counter electrode it it's it's um it's hard to hard to pick one moment but it's it's been fantastic uh I mean this project is is only a little more than a year old and and we've gotten this far I've certainly never operated at this speed uh before and and sort of knock on wood we haven't encountered problems that we couldn't solve yet so you know the whole thing um it's been a lot of fun can you talk a little bit more about the platinum treatment are you actually passing charge through these platinum species when you're doing your electroplating or as you know what effect that's having on the the deposition um well yeah I think it's fair to say that current almost has to be going through the ito and and then through the platinum and and that the metal is you know being plated off of the platinum and um um I um I'm not a very good chemist and I've asked quite a number of chemists why platinum is the best catalyst for pretty much everything and no one's been able to explain it in terms that I could understand um but even I know that platinum is a good catalyst and uh it works um I don't know what you know more uh triple just say about it than that how does it again is the power requirement to switch the windows and if you were able to change the design such that it required more power you could you could cycle more quickly or go for larger area would that be an issue I mean power consumption is really not an issue at all here the only application I can think of where it might be an issue is eyeglasses and um even there we calculated a pretty good number of switches off of a battery that could reasonably be fit in the in the size of the glasses um for the other applications it's it's just not going to be an issue at all now maybe the fact that it needs any power is an issue just because um well I mean you've you've either got to wire it up and there's some especially in a retrofit situation there'd be costs associated with that or you know people are thinking about um integrating solar cells and batteries into the windows there's a uh from Princeton there's a paper in nature materials a couple months back on a um transparent UV solar cell um combined with um an electrochromic um but it's um completely negligible amount of power really and and it's more um yeah the the issues are more the things I talked about like um um just you know get uniformity and and you know stuff like that we're talking about this like everyone should have these windows in like just like a few years or so like do you think that it would be possible to add this gel on existing windows in some way or do you think that I mean you don't usually don't change windows until like 50 years or so like you think there could be a way to just use the windows that are there and add something or do you have to have a new window I mean it's certainly more challenging when you retrofit but you could imagine um I don't see any reason why we can't do this on uh flexible um plastic or flexible glass and and then have an adhesive um you know so it's such that you can uh you know apply it um on there um I I suspect that it would have its challenges and and I've had to think about that in the context of solar a lot and it's dramatically easier to make something in a factory um you know than it is to do out in the in the field um but but you could imagine trying to to do that I think some are picturing it almost like a um a window blind that isn't even attached to the glass so now it's just a window blind that can go from transparent to um to opaque you know one of one of the fun things about the product it spans so much there's the electro chemistry and there's the the nano optics and then it goes all the way over into design and architecture and there's um there's there's so many different things you can imagine doing uh with dynamic glass actually since I've seen no more hands uh seeing our good friend Diane Gurnek former california energy commissioner sitting here on the second row next to sally I do remember reading in some of your materials um a statement that it might be possible with this technology to save 20 percent on building heating and cooling so what's up with those numbers and where are you in that kind of reckoning we have sally's got people who could probably help with those kinds of assessments I'm only reading other people's papers and and reporting the number that I I I can't I can't say that I've had the time you know to think about it and I see the numbers varying a bit and it obviously depends on um uh how many windows you have and and what direction they're pointed and and a number of things but um but the studies seem to show you know numbers um certainly within that zone of of 10 to 30 percent um uh energy savings Diane do you have any sense of the possible timing if everything goes well from when you might actually try a real world demonstration prototype and do you envision again hypothetically thinking about the cost that this would be something that you would just put when you're installing new windows or that you might be able to actually have it um I think for a moment ago maybe why something that with all the existing windows it can be added to well I um I I think that this could develop um relatively uh quickly certainly compared to what I'm used to with solar cells um I think the manufacturing process is not going to be as as difficult um there is some infrastructure out there that um the fact that other dynamic window companies are paving the way and building the industry and you know getting the architects thinking about it um is is great and and is likely going to speed things up for people that come in later and um I uh you know one of the things that I'm I'm pretty sure that Jim Sweeney would agree with me is that um people usually won't buy these products that improve efficiency um just because of the energy savings or the cost um usually they're going to buy it because there's something else about it that they really like and I think we absolutely have that here um I suspect that most people picking these windows won't even think um you know about the energy savings it it just would look so much better or you know imagine imagine a beachfront restaurant where people go to have dinner and watch the sunset but they're just completely blinded um you know the ability to go in and dim it um you know would would be wonderful and I think um I I think people would be willing to retrofit or you know maybe people who weren't willing to upgrade to double pane windows now they would um because um it's just it's just more compelling you know it's going to look so much nicer um you know with that um so I I I think some people will choose to retrofit but it's always easier to have it look just right and be perfect you know when you designed the building from the beginning you know to have the the dynamic windows okay great we're just about out of time so I think if you guys want to save your questions for after that would be good we had a couple hands go up uh at this point uh want to remind the students taking this class for credit to initial the attendance list in the back of the room where Ari is now and I'd like to thank Mike one last time for both the next thing