 All right. Well, thank you, Nick, and it's a pleasure to be here. I changed the title of my talk a little bit to hope to give a slightly broader perspective of some of the ongoing activities that we have in my group in trying to control thermal energy or thermal radiation, rather, for energy application. So I think it's probably not surprising, but it's worth re-emphasizing that thermal radiation is really an ubiquitous aspect of nature. And if you look around ourselves, you actually almost see anything emitting thermal radiation, and the characteristics of these radiations are characterized actually by its temperature. So, for example, the sun, which we talk extensively about in this symposium, is a wonderful energy source because it has a fantastically high temperature of 6,000 Kelvin. And on the other scale, you can talk about ourselves, which turned out to be a very good thermal emitter as well at the room temperature of 300 Kelvin. And lastly, one can talk about the sink of the sky, and that's probably a 3 Kelvin. So what you generally see is that there are wide ranges of thermal body that they all emit. And therefore, the ability to manipulate the thermal radiation is going to be extremely important since all of these has important energy implication. Now, if you open any textbook on thermal radiation, they tend to teach you about black body. And the conventional view about a black body, well, of course, is that it's black. And so by black, what you really mean is that it's absorbing over a very wide range of wave fences. If you take a carbon black, for example, it absorbs both the infrared and the thermal photon. Consequently, when you heat it up, you have the standard textbook of black body radiation. And I'm showing here a tungsten light bulb to remind you how that look like. Generally speaking, these radiation are very broadband. They span both the thermal wavelength range up to maybe 10 micron or so, all the way into visible wavelength range if you heat it up to a temperature that's high enough. Also, these are broad angle emission. In the case of a light bulb, of course, that's great. You can see it from any angle you see. Now in the past decade or so, what has been emerging is that in fact, even though this is what's being taught in the textbook, this is not what the thermal radiator needs to be. And in fact, one can engineer very strongly both the spectral and the angular properties of thermal emitter. One of the early work in this field, take a silicon carbide, it rules a micron periodicity grating on it and show that this kind of structure actually has very strong angular response in its emissivity. One could also create very sharp spectral emissivity peak as well. And this is a work where one take basically now what we call a matter surface by essentially a gold micron scale antenna and was able to get very narrow emissivity peak or absorptivity peak by detail balance in a few microns. In this case, you can see around 6 microns to about 8 to 9 microns wavelength range. And in this case, the emissivity peak is entirely controlled by the geometry of these gold antenna. So the point here is that one really now has the possibility to shape thermal radiation by nanophotonic design. And in doing so, therefore, allow us to rethink about many of the aspect of thermal radiation in energy context. For example, the first thing I'm going to show you is a simple example that you can tailor a thermal emissivity to look like this that you can design a matter that has essentially very low or near zero emissivity all the way everywhere except in the wavelength range of 8 to 13 microns. Now, you might wonder why on earth would you care about this kind of profile? And it turned out what we will argue is that you can build something like that and put on top of a building. You can quote down the building yourself. And so the argument is a simple one. The atmosphere actually is transparent in the 8 to 13 micron window. And so any emitter that is strongly emitting in that wavelength range would be able to radiate the heat out into the outer space and, therefore, try to establish a thermal equilibrium with the outer space. And the outer space being extremely cold then provide a wonderful heat sink for the thermal energy that's inside the body on earth. And so we construct one of these things. So this is a picture of an 8-inch wafer, a silicon wafer, with these multilayer structures that we build on top of it. And under sunlight, this looks very much like a mirror. And I have two of my postdocs in my group, Ashwas Rahman and Linxiao Zhu, actually reflected in that mirror. And if you measure its solar reflectivity spectrum, it's strongly reflecting on average its 96% reflection over the entire solar wavelengths. On the other hand, it's a very strong thermal emitter in the 8 to 13 micron window. So if you take this thing and put on the roof of a building, and this is the Packer Electrical Engineering Building, and even with a very simple setup as we see here, mostly consists of woods and polystyrene films, you could actually get a temperature reduction below ambient. So what we see here, and that I showed before, the black curve is the ambient air temperature, the blue curve is the radiative cooler temperature, and the green curve here is the solar irradiance on the sample, what the sample will see. And at a peak solar irradiance of 900 watt per meter square, you will actually have a temperature that's five degrees Celsius below ambient. So that's some of the initial result that we show that you really can do cooling actually passively by simply shape the thermal radiation spectrum. One of the interesting questions about these systems is, well, what exactly is the limit in terms of the performance in this kind of system? In this case, of course, the limit really is the sky. And we know that the universe itself is as cold as three Kelvin, so in fact, theoretically, if you have a window of transmission that's perfect, then you should be able to get three Kelvin. Now, more realistically, of course, we don't have a perfectly transmitting atmosphere, but if you calculate based on the transmission spectrum of Stanford here, theoretically, you should get a temperature depending on the condition that somewhere on the order of 60 to 80 degrees Celsius below the ambient air temperature. And that's something with the support of GSAP that we are trying to actually probe. So in order to do so, we, of course, need to very significantly improve our thermal design and Dr. Zhen Chen who is sitting on the back is leading in this project. So what we do is we actually build a vacuum chamber surrounding the thermal emitter and we also ensure that we also build the entire setup, including the sunshade to block out the direct sunlight that may impinge on the emitter. And in doing so, we have the experimental results as shown here and where on average, we have close to where we have somewhere on the order of 40 degrees Celsius below the ambient air temperature entire passively over a 24-hour period. So in other words, one could really do deep freezing. This is way below the freezing point of water entirely passively without any electricity input. And one of the interesting point about this system is that the maximum temperature reduction actually occur when the ambient temperature is the hottest. In other words, if you look at this curve, the maximum temperature reduction actually occur around noon time. And this makes sense because if you cut out all the other heating pathway into the sample, then the sample is trying to pin its temperature to the universe and doesn't care about the ambient temperature. And by this argument then, if the ambient temperature is hotter, the temperature reduction is actually larger as well. So I think this point to some of the very unusual opportunity that you can think of about using these passive structure for thermal management purposes. The radiative cooling that I show here requires sky access. My group has been on the roof of Packer Engineering Building for about two or three years by now. So on the other hand, there are also a large class of application for cooling where you do not have a sky access. And one of the interesting possibility is something called personalized thermal comfort. And this has to do with indoor cooling. So in a typical environment, we are hotter than the ambient air and in this case, we dissipate our heat into the ambient environment. In a typical office environment, 40 to 60% of the heat dissipation of the human body is through thermal radiation. So being able to control the thermal radiation therefore would drastically influence our personal capability of dissipating heat into the surrounding region. Now as it turned out, a typical textile that we wear, for example, cotton, by design is used to absorb and therefore block thermal radiation. And so consequently, if you care about cooling inside a room, in fact, wearing something like cotton is suboptimal. Now we all have the experience that if we don't wear anything inside a house, we will feel colder. And so that's a natural cooling mechanism. Of course, this is not something that you can routinely do in the office environment. So by this argument, what you would like therefore is a textile structure that retain the ordinary textile capability and properties, both their mechanical and other properties as well as their visual appearance. But on the other hand, allow thermal radiation to pass through. And this is a work in collaboration with Professor Yi Cui who is sitting there. So what we did recently is to explore this class of structure called nanoporous polyethylene. And these are made of polyethylene but with some micron scale nanopores. And because these pores are at a wavelength scale, at a visible wavelength scale, they strongly scatter visible light. But on the other hand, these pores are at a subway and scale compared to thermal wavelengths. And therefore it doesn't scatter much the 10 micron thermal radiation and therefore the 10 micron thermal radiation can pass through. So on the left hand side of this figure, what you see is that it look white. It look almost like cotton. On the other hand, the paramaterial polyethylene itself of course is transparent to a visible wavelength range. And this is indeed a structure that is very transparent in the thermal wavelength range but in the visible range is actually opaque. And Professor Cui's group has done actually a lot of work further engineering these kind of polyethylene structure for example for weakening purposes to get water vapor through for mechanical stability. And we are also, this is a nonwoven structure but we're also very interested in pursuing this further into woven structure. I hope to up to this point to give sort of a broad range of overview of some of the activities that's going on in my group in thinking about thermal radiation. In the last maybe eight or nine minutes or so, I like to switch gear a bit and talk about specifically a project that we are starting to use the thermal radiation approach to think about improving solar energy conversion. And so the idea here comes back to the standard Shockley-Quietha limit. And in a Shockley-Quietha limit, we know that if you consider a single junction solar cell, photons with energy below the band gap are not absorbed by the semiconductor. And photons that are above the band gap are absorbed but each photon contributes only part of its energy towards the open circuit voltage of the system. So consequently, if you run even in the ideal case of ideal semiconductor with no non-radiative combination, you will still have somewhere on the order of 30 to 40% efficiency depends on what semiconductor you use, depending on the concentrations and so on. So the concept of solar thermal photovoltaics, which by the way I think Professor Dick Swanson at the time at Stanford published one of the pioneering paper in this is a very unusual idea. And it's an idea to say that what you would like to do is to control the spectrum that's incident on the solar cell. In other words, to complement to many work that people have done to try to improve the solar cell itself, this is the approach in trying to engineer the light that hits the solar cell. And specifically what you would like to do is to be able to compress the solar spectrum. Into a narrowband radiation without losing much energy. And the way you can do it is through thermal radiation. So you imagine that you have a broadband absorber that absorb the sunlight and then heat it up. And then you conduct that heat into a narrowband thermal emitter with the emission of that emitter tailored to a PN junction's band gap. In doing so, you can take the same short-liquid cell as was done in the short-liquid analysis, but you just insert basically this intermediate absorber and emitter. And what you see is that you at least theoretically, you can get almost a doubling of the efficiency of the entire solar energy conversion system using only single junction cell. So of course, this is an idea that has been around, as I mentioned for quite a while. You may ask, well, okay, this sounds fantastic that you can double the efficiency of solar cell. Where is the experimental situation at the moment? And so here is, I think, one of the highest efficiency system that has been demonstrated up to now, and this is a work coming out of Evelyn Wong and Maureen Soliatris group at MIT, who have recently published a paper on solar thermal photovoltaics that reaching a solar to electricity, to electric energy conversion efficiency of 6.8%. There are a number of ways, of course, you can read the number of 6.8% efficiency in solar energy conversion. And my preferred way of reading this number is to say that compared this with the theoretical limit, there's therefore tremendous room for further innovation. And I'm very happy that GSEP actually takes the same interpretation of these results as we do. So now one should therefore go examine actually some of the inherent difficulties in these standard, in these current state-of-the-art solar thermal photovoltaic systems. And nearly as to say there are a lot of challenges if you want to think about these systems, I'll just highlight a few. And one of them is that the intermediate here really need to be heated to a very high temperature while being able to still shape thermal emission. The optimum temperature is talking about will be way beyond 2000 Kelvin. The second thing that will be interesting to highlight is the cell itself. Even at 2000 Kelvin, you will still be using a narrow band gap cell. And therefore, the optimization of these cells actually will be very important as that's something that has not been extensively carried out. So let me highlight the first part of the challenge here to try to design a meter that would be able to withstand high temperature. On the face of it, you would say, well, we have a meter that stands high temperature, tungsten libel can be heated up to almost 3000 C and it will still be stable for quite a while. However, typically what you are trying to do here is to shape the thermal radiation. The power of thermal photovoltaic is to be able to generate narrow band thermal radiation. In this case, you have to do something to it. Now, for example, you can do nanostructure tungsten, but as we've shown in this early experimental result, this is a result in collaboration with Paul Brown's group at UIUC. If you take a tungsten photonic crystal, one of the common thing that people do to shape thermal radiation, you heat it up to 1200 degree Celsius and it crumbles. And that's actually a very common difficulty associated with nanostructured thermal emitter. So one approach is a material approach. For example, Paul's group have shown that if you coat it, the tungsten was half an oxide, you could drastically enhance its thermal stability. The approach that we are pursuing right now is an alternative approach by realizing that in fact, the idea is that you can actually shape the thermal emissivity of tungsten without nanostructuring it. So the idea is to bring, for example, tungsten in close electromagnetic proximity to a dielectric structure that's used to shape its thermal radiation. In this case, the key point here is the dielectric structure that you can use can be largely transparent and therefore the heating on the dielectric structure will be minimized and moreover, these structures in principle need not be heated in the first place. So in this project, we also have a collaboration with Mark Brangasma's group who is trying to develop thermal metal surface structure for full control of thermal radiation. And he's envisioning a metal surface structure that's shown on the left here where you will have a titanium nitride-based resonator and these are high-temperature refractive materials where you can tune the emissivity peak of these titanium nitride structure by simply changing its size. And moreover, one of the interesting ideas that they're pursuing is to do thermal metal surface with a phase gradient so that a metal surface like that will automatically focus. In other words, you heat it up and it will actually focus the thermal radiation by itself. And that's something that's potentially interesting in thinking about some of the thermal photovoltaic architecture. Now, as I mentioned, one of the important challenge here is in the solar cell itself for thermal photovoltaic application. Oh, the curve doesn't show, but that's okay. So now in our team, we have professor Jim Harris who have previously demonstrated one of the world record multi-junction cell. And they were able to show up to 43%, I think. It was on the slide before, of the efficiency out of these kind of systems. So the important point here, one of the important here is that the bottom cell of the system is precisely has the band gap that you would need for thermal photovoltaic application. So Jim's work, Jim's group has been starting to take their extensive event expertise in the design of semiconductors to develop semiconductor cells that are specifically optimized for thermal photovoltaic application. So with that, and I'm out of time so I'd like to briefly summarize, I hope to convey the message that there are really tremendous opportunity in engineering thermal radiation for a wide variety of energy applications. And I think this is really one of the very exciting opportunities that we are interested in pursuing. So let me start here and thank you for your attention. Presentation, questions. What months, please go ahead right here. So in terms of cooling by changing the body's radiative heat transfer is what you talked about, but what about the fact that a lot of the body cooling is actually done by evaporative heating, a cooling I should say, sweat? Right, so that's where the water vapor transport through them is so important. And what you would like to do and these kind of textiles will enable you to do is really in combining both of these approaches for cooling purposes. So in general, what is the breakdown between the various heat transfer mechanisms of the human body? Like I said, I think the numbers are about somewhere on the order of 40 to 60% who is through radiation. And the rest I think is dominated by convection when we're sitting here since we are not sweating. So the water vapor transport or wicking in those cases are largely for comfort issues rather than for cooling issue. But of course, if we run, if we work out, that's very different. It seems like this could have great applications in public health where there's no electricity and you need to keep vaccines or other medical supplies cool. Do you see that as being an economic? I think it's a very interesting opportunity. There are, that certainly is a very interesting opportunity. Obviously to do so one would need to make these systems a lot cheaper. And we are actually pursuing some of these work in part also because the vacuum system that we're doing is in fact, the same kind, well, similar kind of vacuum level that people use for example. In solar water heaters. So there I think is a pathway towards building this into a more cost-effective way for electricity-free freezing. Mike? Shanhui, for a reference point, could you tell us what temperatures you would have on the roof if instead of using these highly engineered multi-layers you just used individual materials and what would the temperature be if it were a good absorber or like a block material and what would be for cooling, what would be the best single-layer material? So I actually have a plot, I just didn't show it here. If you take something black on there, we did the experiment, it's about 80 degrees Celsius in the same situation. For single-layer material, we did an experiment with aluminum. Just put aluminum there. Basically because aluminum is a pretty good solar reflector. It has about something like 90% reflectivity. But when we put it up there, it still has a temperature somewhere on the order of 10 to 20 degrees Celsius above ambient. So the structure that we put is the first known structure that you put it under the sun and it actually has a temperature that drop below ambient air temperature. Question here? In the TPV application, if you look at a tungsten light bulb, it has glass and that holds a vacuum. So if I'm having sunlight coming to this emitter, doesn't it have to first transfer through glass and then be contained in a vacuum and how does that layer of glass affect the whole picture? So the glass of course can be and it is usually rather transparent to solar radiation. In these cases, what you really are after is to get the solar radiation in. So that shouldn't be a major issue. Right, so you certainly have a glass enclosed vacuum things and that's actually what we're thinking. Other questions, can I jump in and ask one here? So let's imagine I have this hot body and I'm doing some nanostructuring or something on front and then that restricts the emission out. How should I think about what is that material doing to the emission? Is it just reflecting it back or why is that changing the emission spectrum of that hot body? You mean if you change the absorption properties, why does it change? So the short answer is of course, that's what detailed atoms dictate. Maybe a slightly more microscopic answer is that the same process of electron oscillation is responsible for both absorption and emission. So if you influence one, you're gonna influence the other. Anything other questions? If not, let's thank Shanhuwee, fantastic talk.