 They're asking, I'm going to talk here about microelectronics. I'm an engineer and I have been working in microelectronics for now more than 20 years. And I would like to show basically what we do at the lab and also maybe the conscious that we're starting to have in the field with my colleagues and also the students of course that we have in our classroom. And it's the reason that the title here in my presentation will be No Digital Society with all sustainable information technology. And this is the outline of my presentation. I will say a few words about what we call a digital society, but I'm pretty sure that everyone here in the audience knows about this digital society. But I would like to insist on the impact of those technology on the economy, of course, with the environment and also on people. I'll shortly talk about planned obsolescence because it's reality and I will show you different ways to basically have these planned obsolescence in our, let's say, daily life. And I will, in the second part of my presentation, I will insist on the need to educate, think, design, innovate, and act differently as a scientist. And I will finish up my talk talking about what we do at UC Louvain at the European level in that field. You all know that starting at the, let's say with microelectronics at the 60s, 70s with the first microprocessors. And today we have all of us, I suppose you have a laptop or a computer at least, you have a smartphone. And started to see around as more and more connected object, what we call the Internet of Things. And started this idea in 2009. And today ten years after, we can count more than 50 billions of objects that they are connected and all of those that communicate wirelessly. And it means a lot of, as you will see, a lot of energy is used to produce those electronic devices. A lot of materials are needed to build up those devices. And on top of that, I'm not going to talk about that today. But we are really surrounded by microwaves. My research is what I'm going to describe shortly now. It's really about microelectronics for high-frequency applications. And believe me, here, we are really this full of electronic waves at different frequencies. The full spectrum is covered up to approximately four or five gigahertz. And we talk now about 5G. It's really the reality in the lab. And it's 28 gigahertz, 39 gigahertz, 77 gigahertz, 94 gigahertz, 120 gigahertz. And we see with the improvement of electronics, we start to have applications at those frequencies. Of course, if you talk with people working in the field, including myself, we will tell you that all of the benefits to build up those different kind of smart objects. We can, of course, you have heard about what we call smart home, where we say that we will increase security, monitoring, wellness. Autonomous vehicle is going to be a reality certainly in 10 years from now. But all the technology is ready to have autonomous vehicles. And of course, the advantage which is shown on the market is to say, OK, you will have a security removal human errors. You will have less traffic jam. And all the time when we talk about those new technologies, we show the good face, basically. Prevention for the health, monitoring, health care at home for the building, then saving energy for sure, and smart cities with also energy distribution, waste management, emergency services, and so on and so on. And you can hear a lot of advantages to have more and more connected objects around us. And of course, partially, those statements are true, of course. And you can see big players like Microsoft or Cisco that they know that basically the interest of the internet of things may not be in the object itself, but will lead to the data. And similar to all the applications that you have today is the data that people will sell and make money in the close future. And those big actors, they know it, and they have a lot of investment to the microelectronic industry to produce those objects. And afterwards, they will get the money back just by selling data. In my lab, I'm really a scientist working between material science and electronics. Then in my case, a transistor. I will shortly describe what is a transistor, but it's really basically, to my point of view, the system on which I'm working. I'm trying really to find good materials to fabricate novel type of transistors. Another type of sensors. And just here, a few examples of sensors that we have fabricated in our lab. At UC-Lovain, we have clean rooms where we can fabricate those micro and nano devices. And we have, of course, all of the tools to make characterization of those. Different types of sensors. You can see gas sensors, flow sensors, humidity sensors, pressure, DNA. And all of those sensors, they just take a few microns square. And the idea, of course, is to integrate them in different kind of object, or even at the wall, pretty transparent to the users. And we can distribute those sensors all over. And they consume just a few microwatts for most of them. And the idea is to harvest the energy that you have in the environment to avoid batteries. And this is the idea of all of those sensors. This type of sensor as well that we have fabricated that basically in the field of what we call micro-electromechanical system, when you deposit thin materials in microelectronics, you build up some internal strain in materials. And you can imagine, you start with a silicon substrate, you deposit a layer, let's say, silicon dioxide. You have to put this like cooking when you work in microelectronics. And you have to put the wafer, and the wafer is that size, 300 millimeters of silicon with approximately one millimeter thick. You put it in the oven to have silicon dioxide, it's pretty easy. You put the silicon at high temperature around 900 degrees C, and you have a gas flow, oxygen, for example, and oxygen will react with silicon at the surface of your silicon wafer, and you would grow that silicon dioxide layer. But you do it at 900 degrees C, and of course, you have to take the wafer out of the oven, and then it means that you put the wafer at room temperature, starting from 900 down to room temperature. You have now a composite materials. You have silicon and silicon dioxide. They have two different thermal expansion coefficient. It means that when you take back the substrate because the substrate is one millimeter thick of silicon and silicon dioxide might be a few nanometers. Of course, the contraction of the silicon substrate, I mean of the substrate, would be governed by silicon, not by the silicon dioxide. And you start to build up some internal stress because they have different thermal expansion coefficients. Okay, you understand the principle? Then you have a lot of different layers in which you have traction or compression stress in those thin films. And in microelectronics, this internal stress is seen really as a nightmare. A lot of researchers, maybe a thousand of researchers, they try to review, and we were also part of this type of research, try to reuse as much as we can the internal stress in thin films. But in that case, we say, okay, we know that we will have internal stress in the thin films, that why not use it to do something new? And it's the reason that here we build up what we call the three-dimensional MEM sensors. And it's not because microelectronics, it's a 2D technology. You deposit the thin films, you pattern and you etch, you deposit another one, and you build up the device in 2D by 2D layers. But in that case, the idea was to say, okay, let's play with the internal stress in the thin films and we deposit the different layers, we'll have internal stress. And when you release that, when you etch away the substrate, this multi-layer with a certain gradient of stress over the thickness can go up or down. And then we create those 3D MEM sensors, again to sense, for example, flow, to sense temperature, electromagnetic waves, and so on and so on. You see, when you are in front of a difficulty something that you try to avoid, you can also think about what it could be the interest of this drawback of the technology. And that's what we have done here. And of course, all of the sensors, they will be used for this IoT, for this internet of things. And we work on material and device. And another thing that we do is having access to this internal stress in thin films. It means that we have a force on chip and what we do is I suppose that you know about what we call the traction test when you want to qualify the mechanical properties of the materials. At the micro scale, we just use two anchors and you apply a force or a displacement and you measure the deformation of an object. And thanks to the internal stress in thin films, this is what we have built. Then we have here on chip, you see this is a top view where we have let's say in blue, this is the material in which you have a lot of internal stress. And in gray, this is the material that we like to test and have those two beams. And when you add your way by chemical reaction, you can add your way, what we call the sacrificial layer, in that case is the substrate. You add your way the substrate. It means that you can relax the internal stress that you have in these blue beams. And you have a unique actual traction test on the material which can be deposited at nanoscale. And of course, when you have just one structure, you can see here one structure is going to be one point in the stress train curve of your materials. But you design, this is the beauty of micro electronics. We can have very simple devices, but we can have 1000 of those very simple devices. And one device will give you just one point on the curve that if you build up 1000 of devices, you can really draw the complete curve, the stress train curve of your materials. And that's what we have done here. We have an array of simple test structures and we can build up, we'll need to extract the mechanical properties of thin films at nanoscale. Which is also needed when you start building micro electronics or nano electronics and those different types of sensor because they use materials at nanoscale. And properties, mechanical, optical, electronic properties of those materials might change when you move from a few microns down to a few nanometers. And this is the things that we do at the lab. The last thing that I want to describe here that we do at the lab, you see this is a cross section here of typical integrated circuits. What do we have? We have here what we call transistor. You see one unique transistor when you have the gate, the source, the drain, and the current is going to flow from the source to the drain and controlled by the gate potential that you apply here. The transistor is really a switch. If I apply here positive bias for n-type, I will have the current flow. If I don't apply here any bias on the gate, I don't have any current. It's really a switch, okay? This is typical transistor. It's what we use by billions on the chip to have microprocessor. You have the transistor. And here in that field, I will explain you all of the, let's say, all of the materials that we use today to build up this transistor. There are a lot of researchers working on the transistor itself and the materials that we integrate to build up transistors. You have what we call the backend offline. It's all of the internet connections. You have transistors, but you have billions of transistors. You have to connect them together to build up a function, a microprocessor, for example. And we have level of metallic layers here, which is copper. And you have maybe 15 layers of metallic lines on top of your transistor to make the connections. And underneath the transistor, you have the substrate. And when Alain said that you have in your smartphone new type of substrate, the work of my group was really on that part of the integrated circuit. Of course, we work on that, but the invention that now it's all in the smartphone is really about what is happening here in the substrate. You have to know that basically people in microelectronics, they have considered all the time what we call the substrate, but basically the materials that you have underneath the transistor has just a mechanical support because just to give you dimensions, the transistor here today, the thickness is approximately 10 nanometers of materials, just 10 nanometers. The gate oxide is one nanometers. Then the full stack that you have here for the active part is maybe just a few 10s of nanometers. Of course, you cannot process, you cannot manipulate a few 10s of nanometers. But you need to have the substrate, or if you want the handle substrate, this is the support to hold all of the electronics, okay? And in microelectronics, people have been considering that substrate just a mechanical support, just because it's necessary to have it. But when you go to higher frequency, you know that the current flow and the electromagnetic waves, they will not be contained only in the active area. The electromagnetic wave will propagate and we'll see also the substrate effect, you see? And when you are in static or low frequency, we don't care about the substrate. The substrate is just a mechanical support. But when we go to higher frequency and I'm talking about wireless communication, I'm talking about this connective object, they need to work at high frequencies. And all of the electromagnetic waves will see the substrate. And if you have a semiconductor, semiconductor that is conductive, it means that you will have some electromagnetic losses inside the substrate, okay? And silicon is a semiconductor, that if you don't do anything, you cannot produce microelectronic devices working at high frequency if you don't do anything about the substrate. They did not wait, the industry did not wait my group to have working devices at very high frequency. But at that time, they were using three, five materials. Then gallium arsenide, in-jump phosphide, this kind of material that are pretty expensive, they are critical, they are toxic. And our idea was to say, okay, let's see what we can do with silicon. And basically what we have done, we have used high resistivity substrate and we have introduced a lot of traps, a lot of defects, if you want, in the substrate. Don't have time to explain all of that, but if you are interested, we can discuss about this work. But just keep in mind, we use silicon and we know that we have three carriers in silicon, the room temperature, okay? And the idea was to say, okay, we cannot fight against physics, we will have at room temperature those three carriers. But if we put a lot of defects, a lot of trapping mechanism, we would drastically reduce the mobility of those carriers. But basically, thanks to the defects, they are not going to be free carriers anymore, that they are not going to be losses in high frequency. You see the idea? And this is what we have done. And of course, at the beginning in the community, just to tell you, in the community, when you say to people working in microelectronics for more than 50 years, they know that they work hard to have perfect crystal. They work hard to have perfect interfaces. We thought, any defects. And you come up with that idea, you have to put a lot of defects. And we had to really explain and make some demonstration in our lab to prove that indeed, of course, we need to have very good crystal and very good interfaces here where we have the transistor. But underneath the transistor, it has to be rich of traps. And it works. Okay, then you see how excited we are. We are so happy to work with science and technology. It's just amazing. You try, and we are very lucky in Louvain with all of those facilities and so on and so on. So excited by this technology. But when you see all of the object that we produce, all of those new sensors, new transistors, this is really the top of the iceberg. We work on those objects and we could see that some, let's say, company is what they do with these, let's say, fancy nano-electronic devices. And I'm just mentioning here as one of the connected toothbrush, this kind of, you know, sorry, stupid object. And you will see many, many, many more coming in the future. And of course, my group and many other groups in the world will participate to this kind of object that they will not bring you any, you know, pleasant or pleasure, whatever. This is just the object, but this is not the end of the story. You have to see all of the energy that is consumed and I will give you a few examples. When you use all of those smart devices. This is the power of course consumed by the smart devices, but as you will see, most of the power is not really consumed by your device. Of course, there is the power and the material that you use to fabricate the device. And also under operation, this is the energy that you have to use for the infrastructure for the network, basically, and to store all of the data in the data centers. And in terms of traffic, the traffic data you see data, sorry, it's increasing year after years. And the main, what I wanted to highlight here, it's video, video is really quite amazing. The quantity of data which only, I mean, I'm not saying that it's not interesting. I'm looking also video, but you have to realize that video is really a huge part of the data traffic on the internet. This is more than 50% of the Wi-Fi exchange. Mobile communication to us is 20% today, but we can really see a very high growth rate. This year it's 48% increase compared to the last year. Then the number of video exchanges by mobile phone, for example. Now if we see the consumption, electrical consumption for these type of, let's say applications. I'm not going to detail all the numbers, but you can see here this is the electrical power for in megajoule for gigabyte. And you can see that this is basically what you see. Let's say with your smartphone, when you download one gigabyte, it's going to be 0.02 megajoule of energy. But it's nothing compared to, let's say the 1.9 megajoule, which is needed for the infrastructure. And the 0.66 megajoule, which is needed to store the information in the data center, you see. And we have the impression that we don't consume anything with our terminal, but you have to think that about all of the energy which is used for this simple operation. One other things that people do usually also is to watch TV, watch a movie with their laptop. And in that case, depending on the way that you watch this movie, if you take, for example, a high definition, you see 4K, have to realize that just two hours, I mean, this is the power here, but just the instant energy that you use and the power, it corresponds to 4,500 watts, which is quite close to the washing machine when it's operating. And it's, again, this is not the power that you consume on your laptop, it's going to be a few tens of watts. But it's all the power that you consume with the network, the data center, this is really the power that you use. Now, if you reduce the definition, and this is what we can advise to people when they watch a movie on the laptop, the division of the screen doesn't need to have 4K, and there are different possibilities, of course, to reduce this power consumption. But you see simple things that we, some of you do maybe daily or sometimes, but it consumes quite a lot of energy. Yeah, and of course, the cheapest one is to use a DVD, because in that case, the only power that you would consume is the power of your own laptop. Listening music in my research group, most of them, they listen music when they are in the open space lab, and all of them, they are quite conscious and they do what they can. They come by bike or whatever. And when you see that at the end of the day, if they have eight hours music, approximately nine gigabytes from YouTube, because most of them, they don't use Spotify, they can go to YouTube, but please go to Spotify or other provider music, because with YouTube, you have anyway this video that you even don't watch. But it consumes, you see, energy, and if you made a math to have an equivalent of 10 kilometers driving by car every day. And they come by bike, but basically, by listening to music, they also participate unfortunately to the increase of CO2 in the atmosphere. Okay, this is the calculation here for Spotify. Okay, then I think that we are convinced that the impact of this technology on the energy, we need more energy. Just to give you a number, it's difficult at the, let's say at the world level, but it's approximately five to 10% of electrical power, which is used today only for IT, okay? This is the part of electrical power used only for IT. Just in the case of France, just to give, this corresponds to 14% of the electrical power which is used only for the IT than for the information technologies, which corresponds to approximately seven nuclear reactors. And just one thing here to notice, you know that the crisis between 2007 and 2017, at these last, let's say, 10 years, even if we were in the crisis period, as we said, in here in the occidental countries, we still have increased by 6.6% the electrical power consumption. And even also then the crisis, and we can add on top of that, most of the people now they buy these appliance, A++, A++, A++, A++, try to reduce, and this is good, try to reduce the electrical consumption of the appliances. But even on, you know, this is what you have to keep in mind, just one, another thing. I'm working at a very high frequency communication. You have heard about 2G, 3G, 4G, now we talk about 5G, okay? And people talking about 5G, they will tell you, oh, 5G is much better. You will reduce by a factor of 10, the power consumed by the electronics. You will reduce the electromagnetic waves in buildings in the street. It's partially true, yes, yes, I agree with them. But the problem is the new technology will not replace the other ones. The other one will stay there because the demand, the increase of data rate traffic, we need anyway to have all of those technologies. You understand? And basically you end up all of those different technologies. It's the same thing with the energy. You can see now people talking about, it was the fossil energy. We move to nuclear energy, but we keep fossil energy even in increase. The same with renewable energy. We had the impression that we can stop with nuclear. No, you can see. These years 13 new nuclear plants are on the construction over the world. Then this is what we have to keep in mind. There is no decrease. New technologies are coming, but they don't replace the former one most of the time. Impact of the technology, the need for new materials. Again, here just for the information of technology, this is just for four different metals here. You can see just the electronics use 30% of silver, 12% of gold, 30% of copper, and 80% of indium. Then microelectronics is a big consumer of those materials. Now in microelectronics, I don't know who knows the Moore's law. Who knows? Nobody? And you have to know in microelectronics, the Moore's law has been written in 72, 1972, by Moore, Moore, it's a co-inventor of Intel. And he wrote down this law in 72 saying that, saying that basically if you want to have this technology, economically good, you need to multiply the number of transistors per chip by a factor of two every 18 months. Then the objective of this, let's say this law is nothing based on science, nothing based on physics, nothing based on, nothing is just economic law. Just if you want to have sufficient financial benefits producing microelectronics devices, you have to increase the density of your secrets by a factor of two each 18 months. And it's what the industry, it's what researchers like me, we contribute to that because when you say that you have your roadmap, no difficulties to find some good scientific questions and problematic because it's very hard, very hard to keep that growth rate. As you can imagine here, today we have microprocessor with approximately 20 billion transistors over one centimeter square and for the memories, we reach now with Sony and Toshiba, they now propose memories with 400 billion transistors, also over approximately one centimeter square and 400 billions, it's the number of stars in our galaxy just to give you a little bit of number here. That's quite amazing. But to do that, a lot of pressure on scientists, a lot of pressure on resources as I'm going to show you. Just to show you here, the Mendeleev table that Alan described in details. Thank you, Alan. You can see here at the 80s, it was the elements used in microelectronics. You see that the channel is in silicon, the dopants that we use as phosphorus, boron, the metallic lines used for the interconnections is aluminum. You see just a few elements are used for building microelectronics chip. 10 years after, just a few more, you can see here cobalt titanium, they are used mostly to reduce the parasitic resistance of your transistor by seducidation. And you want to basically reduce the parasitics. You want to have access to your transistor with low resistive path and you metalize basically part of your transistor with cobalt and titanium. 2000, very breakthrough in microelectronics. Very, very big change. Today we use, as you can see here, mostly all of the stable elements of the Mendeleev table. And the reason this is the following, because this microelectronics industry is based on silicon transistor, the CMOS transistor, okay? And you reach some physics at one point. You shrink and shrink and shrink and shrink your device. And at one point it's so tiny that you cannot control properly the current in your transistor. And what you have to do is to change the material. You keep silicon for the substrate, but for the few tens of nanometers, you have a lot of possibility that you change the materials. Again, I can give you a lecture on that, it's not the purpose here. We can, we choose other materials to, for example, you have just one example. You have the, I can use the board maybe. Do you see on the board? Both. And we have here the gate, the source, the drain of the transistor, and the current is going to flow like this, okay? To be sure that the gate is going to control the flow of electrons from the source to the drain, we have to be sure that the capacitance of the gate is still able to control this flow. And it's the reason that here, the thickness of the oxide, when you reduce and reduce, this is the length of your transistor. You reduce, you reduce, you reduce the length. At the same time, you have to reduce the thickness here of the dielectric, okay? To have, still have an electric field to turn on and off your transistor, okay? You have, then basically just keep in mind when we start to reduce the size of the transistor, we have to reduce all of the sizes of your transistor, all of the dimensions have to be reduced. But if you reduce an oxide, here it's a field effect transistor. We don't want to have current going to the gate. We just want to have current flowing in the channel. But if you reduce so much, the thickness of the dielectric here, you will have some tunneling current. You know the effect, tunneling current, okay? Then you will have current going to the gate. If you have current going to the gate, it means that you are going to lose power not to control your switch, but just lose power. That it means you will lose also the logic of your transistor, that you cannot accept that. And the idea, the trick was to say, okay, let's replace this silicon dioxide because it was silicon dioxide by another material, which will give you the same control of the gate over the channel. It means the same capacitance, but much thicker, because as you know, the tunneling current is directly proportional to exponentially proportional to the thickness of the dielectric, okay, of the barrier. And if I can thicken here the barrier, I'm going to reduce this leakage current, this tunneling current, okay? And but you have to use materials with higher permittivity because silicon dioxide is 3.9. I have to use, by example, afneum oxide. Afneum oxide, you have a permittivity, which is approximately 20. That it means that you can thicken by a factor of five and reduce drastically this leakage current through the gate without losing the control of your channel of your transistor. Just to give you a simple example, okay? You see the motivation. We need to introduce new materials to be able to build those very tiny devices and still keep the performance of the device. Just now with the example, just to give numbers of the phone, you can see here the 56, we had just those elements used for your phone, then 12 elements, no critical element. And this is the number of phone which were sold per year and 20 millions. And the lifetime was 15 years. 86, we had 29 elements, five critical, and the lifetime, it's five years. It was the first, let's say, solar off phone. And today the smartphone, we have 55 elements. And you see in microelectronics now, as I told you, we will use the full mandatory of table. 55 elements, 20, what we say critical, it means that you have possible problematic related to the supply chain, or they might be also toxic. It might be difficult also to have them when you try to recycle part of your electronic appliance. You see the number of smartphone, one billion sold per year. And the mean lifetime, it's only 18 months. And when we see those numbers, we also wonder about what is going to happen with what we call the Internet of Things with those, so these 50 billion devices, built and sold per year. And there is a very big question mark here about the lifetime of those objects that we're going to produce. I don't know how much time do I still have? 10 minutes? 10, 15? 10 minutes? Okay. Okay, this graph here to show that all of those metallic compounds that we use, just to show that, of course, this is the abundance on Earth crust. Of course, I think that Johann will talk much more about that, but just to show here that from one ton of ore, we have 330 kilograms of iron. And I'm referring here to the difficulty to extract all of the elements that we need in microelectronics. I'm sorry. And you can see here just for platinum, platinum which is used for different types of electrodes in sensors, in biochemistry or biomedical devices. We use a lot of platinum as an electrode, an electrode of reference. And you see here in that case, we have from one ton of ore, we can extract only three grams of platinum compared to the 330 kilograms of iron. It means that you have, you need to have more and more and more energy to extract those elements that are needed for microelectronics. This was really the elements extracted for different products in the 1950. And today, for all of the product that you see on the market, basically we need to extract all of those materials. That means to use quite a lot of energy again to produce those products. It has been already mentioned by Alain. If you check a little bit all of the main elements that we use in a smartphone, you can see that basically except boron, which is extracted in Turkey, you see that most of the rare earth materials are extracted in China. And we don't have nearly anything here in Europe. And I don't know if Johan is going to talk about that, but the reason is not that we don't have anything in our soil. The main reason is that population, people don't want to have any more mines because you have a lot of pollution related to those mines. And this is the real problem because I took that slide from Eric Pirra from the University of Liege, making that exercise to extract basically one meter deep of 1,000 meter square of land in Belgium. And this is the content in terms of elements in that soil. And okay, I don't have time to detail all of them, but basically when you examine that table, you have everything that you need to build microelectronics devices. It means basically that we have everything that we need in Europe, for example, it's a question of decision, question of political decision to see if we reopen some mines, or we say, okay, it's going to be done in Asia most of the time or in Africa for the extraction or South America. And you see South America, Africa, some countries in China, where people still accept today to have that pollution. But soon, I'm pretty sure, very soon they will not accept anymore as we did not, you know, we decided not to accept anymore here in Europe. And there's something here to, what do you think about the, everything that we produce, everything that, a lot of, let's say, polluting industry has been, you know, displaced to those countries. And here we have the impression everything is clean. We talk about this digital society. We talk about the industry 4.0. Everything has to be digitalized and blah, blah, blah. Okay, but it means that other countries, they have to provide the materials, they have to produce, they have also, you will see, to recycle some of our e-waste and so on and so on. Plan obsolescence, very briefly. There are different possibilities, of course, for the producers to force you, basically, to buy new objects. And there are a series here. I mean, engineers, they are part of that, but they are not the only player, I would say, that they contribute to this plan obsolescence. Can be, of course, a functional defect. Could be the fact that you cannot find any more some pairs of parts if you want to repair your device. By notification, the very famous one is with the HP printer, where you have, basically, a memory mentioning that after 100,000 pages have to say that you cannot print any more and so on and so on. And of course, one maybe of the most important today is these obsolescence, psychological or cultural obsolescence, where you have advertising that telling you that basically your device is obsolete and you have to change it for many different reasons. But for sure, we are working with sociologists to see how we can really fight this kind of, let's say, advertising campaigns. And I was saying here that we have to keep in mind that technology is never neutral, never, never. And this is the type of electronic waste. You have to realize that with smartphone, just two person are recycling. And it's not that it's impossible. I mean, it's not that people don't try that is extremely difficult. Umicore, it's a big company in Belgium and they try to recycle smartphones and they can extract five elements out of the 55 elements. The other one I just lost is what we call in microelectronics a dispersive technology because it's so tiny quantity of materials and there's just like a big soup. It's a mixture of so many different elements. It's extremely hard. Anyway, it takes a lot of energy to separate elements and the elements that they can separate. I don't remember exactly, but gold for sure, copper, zinc, tin, they are the materials. And it was discussing with Umicore and as you know, in microelectronics we try to reduce as much as we can gold because it's very expensive materials. And the person, the researcher at the company told me if you stop putting gold, we will never recycle any smartphone because gold pay for the others. You see? It's amazing. You try to do something good by removing gold but if you remove gold, they will not even try to recycle the materials in your smartphone. And this kind of things that you are not aware. And of course, yes, electronic waste. I don't know here in Italy, but in Belgium, we pay Récupel, we have to pay when you buy an electronic device a few euros because a company, a nonprofit company will take care of your electronic appliance and they promise you they will take care of it that they will recycle the object. But they confess they cannot recycle those objects and some of the objects are shipped to Africa most of the time, Ghana, Rwanda and other countries in Africa or also in Pakistan or India. They have big ship going to those countries and this is what you see here. I went there, different countries in Africa and it effectively, this is the disaster that we see and then people try to get copper, gold and this small quantity of metal but in those conditions with very toxic smoke and so on, you can imagine that this is really a human disaster for those people. Okay, now I don't want to be pessimistic, not at all. As I told you, I love science and technology and I think that we can do a lot of things to improve the situation. But I'm pretty convinced that, okay, when you see all of these exponential use of natural resources and energy, people are still thinking that there is no limit, that the planet is not finite and we will have to always have the possibility to use more materials and use energy. I don't believe that. But I'm pretty sure that human being will go until the end. They will try always to push and to push the limits. We will see what is going to happen. But I think that working as a scientist and a researcher and also a professor at university, what I would like to try is to at least increase the conscious of our scientists and students. Even if we go to the world, I don't care. I don't want just to hear crying people say, I didn't know. No, you knew it. But we will go, why not? We will go and crash on the world. But we will know why. This is the negative situation. I'm hope that we will be able to do something more than that. And it's the reason I think that we have to think, design, innovate and act differently to have a sustainable development. And there is a very famous professor in Cambridge University, Mike Ashby. And he developed a methodology in five steps because we as scientists will like to have something very clear and it's true that I've started six years ago. I would explain to you how to try to discuss with economists, with anthropologists, with sociologists and all of those people, quite exciting, quite interesting. But okay, you drink, you drink at a bar, discuss with those people, fantastic. But afterwards, you are still an engineer and you don't know exactly what you can do. You don't have any tools basically to act differently, to design differently. And the beauty of the work of Mike Ashby was indeed in his field to develop a methodology. He's working in methodology and he proposed a software that will help you to make decision. And okay, this is just the definition of sustainable development, okay, you know that. And basically in the software, they try really to start from the beginning, from the extraction of the material. To see the different, let's say, way to refine the materials, to produce the materials to use and really try to calculate the energy and the resources used during the full life cycle of the materials until the end of the first life, we call the first life. You can have landfill when it's really not possible to recycle. You can use also as a combustion for combustion. And of course, try to maximize the possibility of recycling, but also maybe re-engineer or reconditioning of the materials and also the reuse of parts of your object. And this is what he proposed in this book and he showed, for example, one of the book, he showed 10 case studies and explained basically what the product, the way that product has been designed and he showed that it could have been differently. It could show that if we, let's say, if you can really think about all of these different constraints, because of course, he applies some supply chain constraints. He applied also constraint about the criticity of the material if it's toxic. Restriction, all of the distillations in Europe, in Asia, in the US, also the registration related to the disposal. Then I don't know if someone that already used this software, Granta, never heard about that. I would invite you to check on the website, Granta. And you have to imagine the idea is to say, okay, I want to design, I don't know, for example, a bridge, okay? And if you want to design a bridge, you need to have, let's say, metallic structure, and that software will help you to decide the design, the material to use, we'll calculate, we propose you the process to use to form the different metallic pieces that you need for your bridge. They will show the water that you use, the carbon dioxide that you will use for the, depending on the process that you use, and depending where you are going to implement the bridge, they will give you some constraint about the distillations. It's amazing, it's a huge database, which is not only technical, okay? And it's what I think we would need for all of us in our own field. I'm working in microelectronics, and now with the group that I founded three years ago, it is what we try to do, to develop this type of tools for microelectronics. Just an example here that we can already do. If you want, for example, to define here an accelerometer, this is the cross section of the accelerometer, an accelerometer that you have, for example, for the airbag in the car, and the way that it works is pretty simple. You see, this is the silicon substrate here. You etch, then you remove part of the material of your silicon substrate. You see that cavity, which has been etched. You have then that beam, the beam is going to be the spring, if you want. And here you have the mass. And when you have a very strong distillation, you will have the mass here on the vibration. You see, the mass is going to vibrate, and you detect that vibration thanks to the piezoelectric layer that you have here on top of your bridge. You see, this is the principle of the accelerometer. What you need to have here, material, which is going to be very sensitive to those vibrations. It's a piezoelectric material. And, for example, with that software, we can plot now, you have a library with all of the piezoelectric materials that you can imagine. And what you do here, you plot the coefficient of off-interest, which is the piezoelectric charge coefficient. And it's the change of charge on the material for a certain force applied to the material. And what you observe here, this is all of the materials that you can get. But now, if you apply some constraint, you see, usually as engineers, the only constraint that we have, it's coming from, you try to think and to design something. And after you go and you show to the, let's say the marketing department or the supply chain or whatever, and they would tell you, no, no, no, this is too expensive. You have to gain a few dollars on that, blah, blah, blah. The economy is going to put your account trained. But this is nearly the only one. Here, the idea is to take into account other constraints. For example, here, you try to say, okay, I don't want to have any toxic materials. And then you see, this is the reason that PZT, lead, lettonium, azuconium, titanium has been removed from the list here. Even if they were the best. But this is, of course, trade-off, okay? Then you remove those and the first one, it's barium, tetanide, which can be the candidate for this kind of application. Okay, just one example. I have more, more examples, but okay, I don't have time. Work on legislation. I think that public government can do also quite a lot to fight against obsolescence. Just to have transparency and information to the consumer, longer warranty show. Also, I'm talking with people at our university to have on the label of each equipment, just to mention if it's possible to repair or not. To, and for the guarantee to have more than three years, to have five years or even 10 years, to have all of the information if you want to repair yourself. This is things that doesn't exist today, simple things, but we have to fight to have that from the manufacturers. And there are people that they work on that. New economic models. Currently we are in the consumer market. The economy is based really on the ownership and the high replacement rate. What we would like to see, and there are already a few examples, is to move from the economy of ownership to the economics of functionality. And basically it's the economy of the use of the product. It's not really to own the product, but you use the product. And this is the case, for example, for your, in Belgium at least, for the Y5 box. You don't buy it. It's the provider of the wireless, let's say, communication that they provide the hardware, the box. And it's going to replace it when you want to change one component to improve the performance of your Y5 box. And I was in contact with a person working in the company and he told me you cannot imagine. Those box, they can stay for 15, 20 years. We just change a few components. We keep most of them. We just change the component that we need to change. And of course, we design the box, knowing that with new generation, that part is going to be improved and blah, blah, blah. That they know exactly before the design that to have something that is going to sustain, to design it differently. And you see the responsibility is on the producer, not on the owner. And it changed completely. And the engineer that designed it properly without obsolescence, plan obsolescence. They really designed for durability. And it's one of the economic model on which we have to work in microelectronics. And it's the reason that three years ago, we launched this European Nanoelectronics Consortium on sustainability, in which we have, of course, engineers, but we have also economics. We have philosophers. We have sociologists and so on. Then we have people coming from other field that they help us, basically, to define the technology of the future. And about the education. Of course, I think it's quite important to prepare our students to think differently, to get this holistic approach. And we have one lecture at UC Louvain, which is named Ingenieux Sud, where we have students in our university who collaborate with students in southern countries. And they really work on a very practical problem. They work together by changing IDs, prototypes, during the academic year. And there is summertime, the Belgian students, they go to different countries in Africa or South America to implement the technical solutions that they define together. And what we try to do there is to strengthen, really, the collaboration between young people working on very practical problems or needs and to respect themself, to see also the world differently, because in my case, I always been working with European people or American people, and we have the same way of consuming, the same way of thinking about life, about love, about nature, about work and whatever. But working with people from those different countries, they change their mindset. And also, for the southern students, they change their mindset as well. And it's really a very good project. And just to give you a few numbers, we started six years ago, approximately 100 projects. You see 700 students, they have already participated to this program. And it's a good thing here. I did not say that. I don't know if you know about the number of women and men in science and technology. What is the ratio you think? In science and technology. Engineers is 10% of women, engineers. But science and technology is much more than engineers. Science and technology is approximately one-third of women. One-third. But the disaster is when you see the number of managers, the number of people, directors of labs, of professors. Three percent of women, only. Three percent of women. And we are one-third of women in science and technology. We need more women. Yes, really. And the proof is there. This lecture, which is about collaboration, which is about taking care of others, we have 52% of women. We are in the sector of science and technology. 52% of women and 48% of men taking that class. Then, conclusions. I hope that you understood that to my point of view, science and technologies are really great. It helps you to explore, to better understand the world that's surrounding us. Try also to, of course, to transform the world in which we are living. And could be seen as really a great thing. Don't forget this never-neutral. When we change something around us, we will have a certain impact, not only positive. Technology cannot do anything alone. What I mean is the technology is not a question of engineers. Technology is a question of everyone. Everyone has to discuss, has to debate about the importance of the technology. And as I told you, we must think, research, teach, and act differently. Okay, there's a problem here with the slide, but just to say that up to now, as engineers, I have been working on problems. I tried to bring new technology to solve that problem. And as I told you, I always see economic as really a constraint and the only constraint. Not a partner, a constraint. And we are at the reflex, just, ah, you have a problem. I'm an engineer. I'm going to come with a solution. And you are a problem. What I hope is to transform that problem to a need. Really think about what are the needs of our society and discuss with, as I said, many different people from different disciplines to really define what are really the needs. And to, let's say, fool the gap, try to answer to those very important needs. Again, we will not need only engineers or scientists, but we will need all of those different disciplines to come up with a solution. And it means to adopt this holistic approach to embrace the complexity of the world in which we are living. And that means it's not reflex, but reflection that we need to start all together. Thank you. Thank you very much. Yes, under the thickness of the dielectric, we use different deposition techniques. There are some physical techniques by sputtering, for example, or the disoplasma that you have in the chamber. And then you accelerate some of the ions on the surface. If you, let's say, the materials that you want to deposit, then you can really, by erosion, have part of these materials that you want to deposit and is going to deposit on the substrate and can be controlled really, let's say, a subnanometer by subnanometer, depending on the power of your plasma and this kind of thing. This is one technique. We have also atomic layer deposition where you functionalize the surface and you put another gas, you have condensation of that new gas. And it's really atomic layer by atomic layer that you can deposit the materials. And those techniques are available today for nearly all of the materials that you need in microelectronics. And you can really find control of the thickness that you deposit. Another example is the growth of silicon dioxide from silicon. You fix the temperature and you have a gas flow of PPN of oxygen. And that's it. You know the cinematic and it would take maybe, let's say, 30 seconds to have one nanometer of oxide and you just control by timing and you end up with the right thickness plus, minus, you know, the, there you have a certain uncertainty, but yeah, this is for the thickness. After you have to pattern. But okay, for the control of the thickness it's pretty easy on the control, I would say. You mean that the replacement of the silicon dioxide, yes, we need to use, if we still want to shrink this technology, silicon dioxide will not work. And we need to change and use other materials and some of those, silicon dioxide is pretty, I would say, except the high temperature process that you have to use to grow the oxide. Afterwards, it's pretty safe materials. It's not toxic, it's not critical, it's easy to grow. But afnum oxide, it's already another problem. I mean, in terms of toxicity and so on. But this is the way if we want to, would you have this high integration which is not needed most of the time? Really, which is not needed? It's a nightmare when you work on nano-electronics because you start to work on what they call a node and we have been working, and the node is the size of the gate. Starting in my PhD thesis was one micron and after my postdoc it was 250 nanometers and now the channel length of transistor is approximately 10 nanometers. But with the technology which has 100 nanometers you can do maybe 99% of the applications. But okay, industry, they tell you that okay, you have to move to the new technology. But the other technology is still very good and it's still working properly, yeah. Yes, yes, what can we do? What I think is the way that we work, my dream is not to have individuals finding solutions. I like to see more group of people and when you are in front of this systemic crisis, you need to have this systemic approach and the center of gravity will not be one person. It will be really cooperative work for the disciplines and it means also it's not going to be a man only for two, I think they're going to change completely the way that we work. It's really more a teamwork than just the individuals. It's what I believe and then thanks to that I'm pretty sure that in the group we will find as you can see in the lecture that we have, we have more than 120 students a year. It's really a 60, 60 woman and man and it's really great and it's a good dynamic. And afterwards those students, just to tell you, master thesis, they have to do a master thesis because they attend that lecture the third year of engineers and master thesis, they have to do it at the fifth years and last week I received 33 students. They want to make, because they participate to Ageniusud and they now see, oh yeah, we need to adopt this holistic approach and they are looking for master thesis or PhD thesis where we allow them or we force them to think that way and you see 33. I cannot of course, supervise all of those students that I have to discuss with the other professor, but it's good. It means, and I'm also convinced, change will come from the students, from the young people. They will force us to teach differently. When we started, we started with my nanometers and course for nanometers, nanometers. Yes. But every time we have more performance here, it's a bit of a fault. Like for consumption in my area, I want to know if we can do that. So the technology is not against us. Exactly. I don't know if it helps us more. I said, I think the technology can really bring beautiful solution, but when you say that for example, it's going to reduce the power consumption, but yes, it's true. It's true, but you can look at the power consumption. Now it's starting to saturate because the power consumption, due to that just one point, increase, increase, increase the density, heating, joule effect. Now you don't have any integrated circuit working at room temperature. They work at 80 up to 100 degrees C. It means that you degrade the performance of the device and then you consume power also for heating power. And there is a trade-off. Of course, I agree that thanks to the reduction, you reduce the power consumption with the FinFAT and so on. I would say it's wrong due to the concentration. I'm really working at the state of the art technology, which is our industry with 28 nanometers fully depleted SOI, FinFAT, and we see this increase of the power consumption now due to the cell heating. Of course, they are working hard to reduce that. And I agree with you that we can reduce the power consumption, but the problem is you build more and more and more and more than the total global energy consume will still increase. This is what I've shown for different kinds of applications. Then I'm not saying that we don't have to pursue the downscaling of the transistor. Okay, we can discuss afterwards. Yes. The final message. Okay, just one example here. Smartphone, who knows Fairphone? Who knows Fairphone? Nobody. Okay. Thank you. Then, Fairphone is just to say that they are products that they have been thought differently. This is what we call modular electronics. I told you that with microelectronics, it's a dispersive technology that it's impossible to recycle, okay? And what we have to think is to repair, is to replace, and this is the idea of Fairphone. It's a Dutch company, a small Dutch company, and they produce a smartphone that works perfectly, but I can open it just with a simple screw. I can really change everything here. There is approximately 12 blocks, and you can replace the camera, the screen, whatever, et cetera. Then the idea is to have this modular electronics, then electronics that if I don't use anymore the camera for the smartphone, because I want to have one with higher quality, I can still use it for another applications, you see? This is the idea that it's not a monolithic device. This is something that can evaluate, can be upgraded, can be repaired, can be blah, blah, blah. And then you start to have different kind of electronic products with the same spirit. Then this is, I think, one of the way that we should think is can still reduce the size, but please think about the total lifetime of your product, and think to increase basically the lifetime by this kind of principle, for example. You can go and see Fairphone, they do a very good job.