 Rhaid i'w rhai, rhaid i'w Rhaid i'w Rhaid i'w Brifysg Senf's Association. Rhaid i'w rhaid i'w Radu Sprir, dweud o'n pethoedd y gweithio eich Llyfrgellau. Rhaid i'w Rhaid i'w Brifysg Senf's Association, i'w Radu i'w rhaid i'w rhaid i'w Llyfrgellau i'w rhaid i'w Brifysg Senf's Festival. Rhaid i'w Llyfrgellau i'w rhaid i'w Llyfrgellau i ddych chi'n gweithio fan hynny o'r lech. Felly, i'r bydwch ar gyfer y gwirthod yng Nghymru, gweithio'r pethau ar gweithio yma. Mi'n ddechrau i'r swansie, yn September. Rydych yn gweithio. Felly mae'r gweithio siart, a'r bydwch ar y gyfer y gweithiau, felly mae hi'n ddweud yw'r ffordd ar y gweithio ar y gweithio, ac efallai efallai i'r gweithio, mae'r gwrth yn beithio. Felly, rydyn ni'n gweithio Radu. Rydyn ni'n gweithio Radu, esbryr o'r rhesaith yma at the University of Surrey. He researches flexible electronics and there are many applications. So now, can we introduce him? Let's give him a welcome. Let's give him a round of applause. Thanks very much. Thank you. Hello everyone. Hi. I can't see your faces. I'm just assuming that you're here really. I'm going to start straight on because the time is quite short for this one. This is a talk about plastic electronics. This is a talk about electronics. I thought we'd start in the time and place which was absolutely essential for the beginning of electronics, which is, of course, the capital of Romania, Bucharest, where I come from. So in 1998, in about September, one Saturday, I was coming from a film that I saw with my girlfriend and I stopped at the bus stop while waiting for the bus and I bought my first copy of this computer magazine. And then I spent most of the evening reading it and in there, the most fascinating thing possible was this. The Pentium II 433 MHz processor. This was an absolute behemoth and I thought this is it. This is the future. This is the pinnacle of semiconductor engineering. So I was completely enthralled by this, but then I got a bit sad. Mainly because the film we went to see was Titanic. But it was more than that. You see, I wanted to be part of this thing. I wanted to be part of the story of electronics. I wanted to create things. But unfortunately, I was just a child. I was just going to school. So I wasn't able to do anything. And this was the future. And it was happening now and I'd missed it. But was it? Let me remind you how the 1990s looked. This is how computer graphics looked in the 90s. This is how mobile communications looked in the 90s. And this is how web design looked in the 90s. So was that the future? Well, it was in some sense, but not quite. It was, in fact, a small cog, a small part of a chain that started with the development of the first integrated circuit in the 50s by a bunch of guys that looked like this. One guy in the picture, probably to your left, is Gordon Moore. Now, Gordon Moore became famous for Moore's Law, which is really an observation. It's not a law at all. This guy at some point looked at the amount of stuff that was going into an integrated circuit and he said, okay, there were five or six or seven components up to about 20, up to about 30 by the mid-60s, so he just drew a dotted line. And that dotted line went on for 50 years to the point that these days, computer chips have parts that are a hundredth of a millionth of a metre across and there are many billions of them in a single chip, the size of a fingernail. And we're, of course, concerned with digital circuits where we only care about the zeros and the ones. We only care about the extremes of what the data can represent. Analog is getting a bit left behind, I suppose, in the public's imagination in the digital world. But I'm going to tell you in a second that analogue is, or rather plastic electronics and the future of electronics is all about analogue. So in analogue, we're really concerned with all of the values that a signal can take from its minimum to its maximum. And there are plenty of heroes for digital, but there are a lot of unknown heroes of analogue as well. Who is this? Do you know who this is? Bob Widdler. Bob Widdler, fantastic. So this guy was instrumental, again, in creating fantastic electronic circuits, amplifiers, current sources, you name it, analogue electronic circuits in the 60s and 70s and 80s. And he was not quite the advocate of digital as this picture shows. Right, analogue is something that I hold very dear. And in fact, eventually, to go back to the story on my first slides, eventually I was able to work on some fairly clever analogue circuits still in Bucharest and still something that I'm quite proud of. The thing though is that all of these circuits, analogue and digital, usually are made of minute specs of silicon. So the chips are quite small as you can see. But there are plenty of electronic circuits which are large by definition. And we're talking about what's technically called large area electronics, but mostly display screens, right? They're meant to be large because that's their thing. They need to be large. So the circuit has to go on a fairly large substrate, usually glass. And this is the genuine size of the glass that people make TV screens and things on. Now, of course, glass has a bit of a problem. And I'll show you what the problem is with this simple video. And I'll explain. The phone there is just used for contrast. It's not part of the circuit. But I'm going to make a very simple electronic circuit that I'm going to, rather electric circuit, that I'm going to tie together through a piece of metalized glass. So there's nothing very clever about it. It's just a piece of glass with metal on closes the circuit. When that happens, an LED lights up. But you'll see in a second that glass has its problems. So that's just a simple circuit. If we use this glass slide with aluminium on top, we just close the circuit. And of course, the LED lights up. Right, now the problem is glass very frequently comes into contact with fairly rigid stuff. And we all know if we had a mobile phone be dropped on the pavement, that that happens quite frequently and quite tragically. So you've seen that circuit dies. The circuit is broken. So glass has its problems. But what we do with our screens is we protect them by encasing, say, our phones into some sort of plastic cover. And this is exactly what I was doing when I was making this video. I was using some sort of plastic protection from my eyes. And we do this all the time for our electronics. But why not make them out of plastic to start with? So here's a bit of plastic that I've metalized only on one surface. And we still close the circuit the way we did. And long story short, this can take a lot of abuse as you might think. There you go. Right, so why don't we make all our electronics on plastic rather than on glass, our large area electronics? Well, the problem is, of course, that plastic doesn't really like high temperatures and a lot of our processing of the materials that gets into our electronic circuits has to happen at 400 or 600 degrees Celsius. A lot of the time glass barely can stand that, plastic wouldn't stand a chance. Right, so we need to change completely the system of our fabrication. We need to use other materials, other machines, other processes to create electronics on plastic. Ideally, that can withstand low temperatures. So we talk about plastic electronics, but we can also call them printed electronics or flexible or solution process in the sense that the materials that we're putting down are now put down at very low temperatures in the form of things like inks, sorry, pastes and gels, by printing, stamping, or even spraying. So this is extremely low cost fabrication at low temperatures for future electronics. And there are of course advantages, things like cost and area. You can do this on a vast area without having a large production machine. You can do it a bit like you print newspapers and magazines. And of course you can print a lot of them. The point is that soon enough we should be able to print electronics on flexible plastic at the same speed as we print, I know, Hello Magazine. However, there are downsides. The biggest downside really is the speed. And by this I mean the speed of our circuits. So the problem here is that the materials that we use, the semiconductors that we now use in the place of silicon are now a thousand to a million times less able to pass current. So the transistors that we're making and the circuits that we're making are going to be slower by that same amount. And yield is a problem. If you're using these very fast production techniques, a lot of the time your patterning is going to be mediocre to say the least. And you are going to end up with either completely failed or circuits with very, very different specifications to what you expected. Now, I work at the University of Surrey in the Electronic Engineering Department and the approach that we've taken to solving these problems of yield and speed and uniformity for plastic electronics is to completely rethink the design of the building block of the transistor. And we've come up with this thing called a source gated transistor. And it's very, very similar to the conventional transistor that we use in display screens, except this time we're using the contact between a metal and a semiconductor and the energy barrier that forms naturally there to restrict the current. And this gives us a lot of very beneficial advantages, I suppose. So for a start what we do to look into these devices is we create computer models, we simulate, we fabricate and then we measure or we test these devices. So the way we do this is we use computer software that will solve some fairly complex equations in semiconductor physics. We discretise these so we tell the computer where precisely to solve the equations. Of course you want better accuracy in the places where you want the data to be quite precise. And then you get something like this. And this is a bunch of electrical characteristics of a transistor working in an analogue fashion. So as you sweep the voltage on the X scale, you turn the current up from virtually zero to something that you could call sizeable, I suppose. And then we make them. We make them with fairly conventional technologies like this one with photolithography, the way we've always made larger electronics, but we make them with printing as well. And then we test them and then we get fairly similar electrical characteristics. So I mentioned printing and the idea here with this device, the cleverness of it, is that it can withstand significant tolerance or process variations. So here what I'm showing, which you can't really see I suppose, is the two curves are for two devices that are ten times different. One is about 20 microns, one is about 200 microns long. And we get exactly the same characteristics, which is not something that you normally get from a conventional transistor. And this is fairly important because what this means is that when you're printing at very high speed and the picture on the bottom really shows a deliberately bad print of two parallel lines, when you're printing at extremely high speed, this transistor will have exactly the same performance regardless of the waviness of the lines or the distance between them. So this means very robust electronics on plastic, on paper, you name it. So this device is particularly energy efficient, it gets to the problem of uniformity and it is high gain as in you can make very high amplification devices out of it and circuits. It has downsides as well of course. One is speed, so because we're restricting the current as it enters the device at the metal contact, we're probably losing about 90% of what this device could go at in terms of its switching speed. And because of the physics, the current passing through it is very temperature dependent. That is probably a major downside. And it's different. Probably its biggest problem is that it's different. Industry works on four or six or eight years timescales. And if you come up with a new idea that's good for, I don't know, next year's product, it's never going to be incorporated because people have thought of what they're going to be doing for a number of years. So it's quite hard to actually get this into industry. We've solved the temperature problem with some fairly clever simulations. I may show you if we have time. And the speed is not entirely problematic for the very simple reason that we're not trying to create processors or graphics card, we're trying to create very simple electronics which will probably go at a few hundred hertz or a few tens of kilohertz at the maximum. So this isn't really a problem in the real world. I'll show you really briefly how this behaves in comparison with a conventional transistor. So you can use the transistor as a current source. And ideally, a current source would look like this. Irrespective of the supply voltage, you should get exactly the same amount of current and you should start getting that current from the moment you apply a minute voltage across it. This is the ideal characteristic of a transistor as a current source. Current devices look a bit like this. So, of course, the flatter the characteristics are, the better. The source gated transistor looks like that. So what we're going to be doing, and I'm going to show you two videos of, again, lighting up a simple LED by passing current through this transistor and in turn driving the LED with current. So what we're going to do is we're going to sweep these characteristics. So we're going to go from large or high voltage all the way to zero and back again a number of times on the same curve, first on the TFT and then on the SGT. And what's going to happen is that the TFT, the conventional thin film transistor, is going to drive the LED with increasingly lower and lower current. So the brightness is going to drop until eventually it goes down to zero. And then you're going to see this happening a couple of times. The second film is about the source gated transistor where, as you might expect from the graph, the current is going to stay virtually the same for the majority of the period. So the LED is going to have the same brightness until very low supply voltages. So here it is. So you can see right at the bottom is the supply voltage, which is negative. And then you can see the tiny LED losing brightness and being off for the majority of this period. And if we go to an SGT driven, sorry, here we go. So it's the same period, the same time frame for this process, but the LED stays on for longer of the period and it doesn't lose its brightness. This is with the same technology in fabrication, but using this trick of controlling the current at this metal semiconductor contact. So, as I said, we can make these with conventional technologies, but also with printing. Desktop type inkjet printers, stamping, spray coating, that sort of thing. So the source gated transistor is a very useful device for flexible and printed electronics because it's robust and it's energy efficient and it has this property of uniformity. But it's only part of the story because apart from very good electronic devices, you need to come up with things or ideas that you couldn't realize before with conventional technology. I'm just going to show you a couple of things that people have been doing around the world in the area of printed electronics. So this was last year a fully functioning microprocessor made by inkjet printing in Belgium. Now the problem with this one is that it's about the same performance as something coming out of 1972 in silicon technology, but it's just the beginning and it's very likely that we're going to be able to do exactly the same thing akin to Moore's Law with these types of transistors and these types of technologies. In Japan, Takao Somea is working on very thin printed electronics. So this thing is an array of sensors that you'd put presumably on your body to monitor your health. The interesting thing about it is that it's just one micron thick. Again, when you're talking about nanotechnology, the stereotypical comparison is the width of a human hair, 50 nanometres or 80 nanometres. So this is absolutely feather-light electronics or feather-weight electronics. So this is how they imagine that this might work. You stick it on, it would send some data wirelessly, it would be powered wirelessly, and then when you don't need it, it would either peel off or even absorb. In the US at Urbana-Champaign, John Rogers is working on transient electronics. So these are things that can be dissolved either if you're ingesting them and they stain the body for a couple of weeks and then they just simply are dissolved or things that you'd put in the environment and they are either water soluble or with very, let's say, not nasty chemicals. So a great deal of effort is now looking into the ecology of these things because this could be a new beginning for the whole of the industry of electronics, right? There are new materials, there are new substrates to put things on. Why not do it in an ecological fashion? And again in the US at Stanford, they're looking at stamping this array of electrodes onto some flakes of semiconductor which are quite randomly put around the substrate. But the idea is that if you need something extremely low cost, perhaps just sprinkling the semiconductor on a substrate at random and only working with the devices that work by accident is probably a good way of doing it. So this is Jan-An Bao who's doing this sort of research. And again, my friend Chiao Jun Guo at Shanghai Jiao Tong University, who is a Sari graduate, I should say. He's looking at a lot of anti-counterfeiting biosensors, chemical sensors on flexible and printed electronics. Again, extremely low cost and things that eventually you would be able to print in your own home. But speaking of printed transistors, this is the work of Henning Searinghouse when he was a postdoc about 15 years ago, and he developed the first fully printed organic transistor. And he just printed it with an inkjet printer like that, with a variety of materials, semiconductors, conductors and insulators. Fast for about 15 years, Henning Searinghouse is still one of the pioneers of printed electronics. He's in Cambridge and he started this company called Plastic Logic. And Plastic Logic made this, which is a fully bendable display screen, plastic display screen. So what you're looking at if you can even see it, it's a black and white image, a bit like a Kindle display, but this is the image latched into this flexible plastic display. Now this is from a couple of years ago, they've let me show it to people. More than welcome to have a look at it afterwards. These days they can do full colour, video rate display screens in this size. So this is absolutely fantastic because surely it means that this is the coming of age of plastic electronics, right? These things are happening, but the fact is we can't really buy them. So is this really the future? Well, it is the future in some sense, but if you think about it, if you were to compare the technology of communications to the technology of printed electronics, we're not even at this stage in terms of our development, we're more like at this stage. So even though there are plenty of companies and plenty of research groups doing amazing stuff with new materials and new techniques, we're hardly here. So to get here, we need a lot more understanding of the basic processes, a lot more understanding of the engineering, but a very good reason for doing things because in this era of commoditisation of electronics, it's very easy to say, okay, I'm going to build the same thing faster or cheaper, but eventually you're running out of reasons to make things. So the key is enabling completely new applications, things that you couldn't do with rigid or just conventional silicon technology. And part of this I think is doing or marrying old technologies with brand new ones. So I'm going to show you two demos which are very, very rough and ready of two technologies where printed electronics would work fairly well and would be seamlessly integrated in everyday products. So the first one is a photo book which has ambient sound and as you turn the pages, the sound changes for a soundscape. Of course you can do much more with it, but the demo is about that. So let's have a look. So the book effectively becomes the trigger for digital events, or it becomes the remote control for interacting with digital content. The other demo we've made for a little project last year was a packaging box which tells you in your own language instructions for putting together a piece of furniture, say, and the idea is that you can navigate by gestures through the steps of the building process. The point again is that we've made this with off-the-shelf components, but the idea is that you'd be able to integrate absolutely every bit of the electronic system into the cardboard in a printing process, so it would be dirt cheap eventually. Welcome to IKEA Trogan Steps. This recording, together with the manual supplied, should make the process of building your steps easier. If for some reason you are missing parts, you can easily order them on the website at IKEA. One, begin by tapping the dowels into the two step panels and attaching them to the side panel. Two, make sure the additional support bar is facing in the correct order. Paper and printed electronics is a big deal these days, mostly to do with recycling and with, again, the ecology side of things. In fact, the European Union, minutes after Brexit, announced that there will be a funding opportunity later this year for paper-based electronics, and we're trying to capitalise on that following on from these little demos. Right, so where do we go from here? My feeling is that the whole idea from now on, it will be from... Sorry, I should say that in a different way. What needs to happen for new applications, for completely new ways of using electronics, for giving us a reason to make these flexible electronics, is to marry the interaction with real-world objects, things like the internet of things, broadly speaking, with distributed data, as in the cloud and big data. And we can easily do this following the digital side of things with immense computing power that we have, and with the analogue, which is coming from the printed sensors and the printed electronics. So what's next? Well, at this point, genuinely, your guess is as good as mine, and for that reason, I'm going to ask you for your thoughts, perhaps, by the large tree outside. Thank you very much for listening.