 So I'm actually going to start this talk with a small confession and that confession is although while I was here doing my PhD in energy technology for the longest time I didn't actually know how I was using electricity and that was pretty embarrassing and so eventually I picked up one of these it's called a kilowatt and what allows you to do is plug in whatever electronics that you're interested in and measure the power out and so I got really into this like embarrassingly into this annoying my housemates sort of into this because I was just running around I'm plugging things you know leaving the refrigerator running through this for multiple days to get like a better data and what I learned was actually kind of surprising to me what I learned was that the things that I associate with you know modern life the trappings the things that allow me to access the internet like my laptop or a monitor these all use roughly for me about 50 watts even the bathroom fan that we had was around there and when you add up all the light bulbs that we had in our house was up there as well and I expected our refrigerator to just be a huge power draw but then after getting that data over several days it averaged over the course of you know a day something like double that amount so your refrigerator might be newer it certainly actually probably is and your laptop might be a little bit more efficient but the exact number is an exact of the point the point is that we don't generate on the sort of 50 watt scale and we don't even generate a thousand times and often not even a million times that size and that's because this is still the state of the art mechanical engines just described to you you're all experts now they transport us they keep us cool and they generate over 80 of the electricity in the united states but they sort of create a problem and the problem is that they're so big in order to connect all of this generation to the sort of 50 watt devices that we all have in our houses well it requires an enormous amount of infrastructure and so this creates also an opportunity an opportunity not just for the third of the planet the living countries that use 10 times less electricity per capita than the united states and not just for the billion plus people who still don't have access to a reliable electric grid but it even creates an opportunity for us here because in places like new york city over half of the cost of electricity is due to distribution and not to generation itself so what if we could produce power at the human scale and how what would a technology that could take advantage of this opportunity look like well we think it needs to be three things first of all it needs to be distributed generate electricity where you want it at the levels that you need the second is that it should be pretty efficient because for the sake of our climate and our future probably shouldn't be more carbon intensive than what we're already going for and the third thing is we think it should be simple not just technologically but also from the user perspective should be able to plug in when you want and get the electricity that you want and so obviously to uh uh re-envision the sensual electric grid we're looking to half-century old technology but in seriousness half a century ago is the height of the space race and we needed a solution for power and space for the simple reason the steam turbines don't work as well there uh and so what we did is uh nasa poured in a huge amount of uh uh resources in order to develop technologies that would actually even today seem pretty high-tech solar panels as we heard earlier today even fuel cells and thermal electrics all of these fluent space for the first time in the 1960s but of all of these sort of space race technologies one of these at least for us kind of stood above the rest from the raw power the raw performance perspective and it's one you probably haven't heard of it's called thermionic energy conversion so a thermionic energy converter is probably the simplest heat engine that you can imagine it's got a hot side it's got a cool side and a vacuum gap in between that thermally insulates them and because of that vacuum gap you can get really high temperature differences that as we just learned can drive a lot of power and so back in the 1960s uh researchers demonstrated efficiencies there were three times higher than the competing technologies like solar panels or thermo electrics and power densities they were just colossal like a thousand times higher than a battery so then what happened well at the end of the space race thermionics kind of got forgotten it got a reputation of being an old-timey vacuum tube technology that had been tried before and so we believe that this creates an opportunity to apply decades of innovations and materials and microfabrication in order to re-envision this technology and overall more than double the performance and so that's where I sort of came in at stanford along with professor zx and nick moulach and roger howe i studied new applications and new materials new innovations towards thermionics and in those days we were focusing on basically power towers out in the desert combined cycles to increase the efficiency and therefore drive down the cost of these sorts of power plants but through the course of my phd and especially towards the end we realized that the innovations that we had made were far broader than just increasing the efficiency of steam turbines out in the desert and if you scale it down just the device itself on a standalone basis could be really pretty incredible what we realized was that we were really building towards a power plant on a chip something that could have the potential to become the smallest lightest engine on the planet something that could be competitive with virtually any mechanical engine but with no moving parts so let's imagine that we're successful that your next generator is a spark reactor why would that matter well for one the lack of a small light engine is why your quadcopter flies for maybe 20 minutes but your helicopter can fly you up and down a mountain for hours and setting our sites even higher we've learned through conversations with nasa that the promise of thermionics for space reactors is even larger today than it was in the 1960s in fact the energy conversion branch chief also chief technologist of the power and propulsion division at nasa thinks that our technology could be enabling to a man mission of mars because within a single calendar year you could not only send someone there but also bring them home and finally once they're home back on earth that's actually some of the most exciting opportunity i think because you know some of our first prototypes which we're just making now look a bit like this and while we're not there yet from the historical power density of thermionics something this size could put up up to 50 watts we're on our way there and if we get there it's not important just because that would power my old laptop but because the international energy agency defines a family of five in a rural area is being on the grid at 250 kilowatt hours per year when you translate that down to what a generator would have to operate at you know during the waking hours it's about 50 watts so according to the i a one or two of these devices could be the difference between a family being off the grid or on the grid but all without a grid so we're just getting started at spark and i thank you for your attention next