861: Engineering Novel Solutions for Data Storage and Energy Management in Electronics - Dr. Eric Pop

0:00Hello everyone, I'm glad to have you here for episode 861 of People Behind the Science. I'm your host, Dr. Marie McNeely, and this week we're rebroadcasting our interview with our guest, Dr. Eric Popp. Listeners, Eric's research spans electronics, electrical engineering, physics, nanomaterials, and energy. He's interested in applying materials with nanoscale properties to engineer better electronics, such as transistors, circuits, and data storage mechanisms. In addition, Eric is investigating ways to better manage the heat that electronics

0:32generate. And in our conversation, he shared some phenomenal stories and experiences from his life and career. So listeners, get ready to meet another one of our fantastic people behind the science. Every day, discoveries are made that will change our understanding of the world around us. Dr. Marie McNeely is here to bring you the brilliant minds who are making these discoveries so they can share their incredible stories and take you on an amazing journey. Welcome to People Behind the Science. Hello, everyone, and welcome to the People Behind the Science podcast. Today, I am delighted to

1:15speak with our guest scientist, Dr. Eric Popp. So Eric, welcome to our show today. How are you? Thank you. I'm doing great. Wonderful. We are excited to chat with you today about you and your work, but I'd love to take a moment to tell our listeners a little bit about your background first. So listeners, Eric is an associate professor of electrical engineering as well as material science and engineering at Stanford University. He received his bachelor's degree in electrical engineering as well as a bachelor's degree in physics and a master's in electrical engineering from MIT. He was awarded his PhD in

1:47electrical engineering from Stanford University. And afterwards, Eric conducted postdoctoral research at Stanford before accepting a position as a senior engineer at Intel. Prior to joining the faculty there at Stanford University, Eric served on the faculty at the University of Illinois at Urbana Champaign. And he has received numerous awards and honors throughout his career, including the 2010 Presidential Early Career Award for Scientists and Engineers, Young Investigator Awards from the Navy, Air Force and DARPA, as well as an NSF Career Award. And in our interview today, Eric, we want to learn

2:18more about you as a scientist and engineer, but also more about you as a person. So can you start by telling us what you like to do when you're not very busy with research? Sure. Well, since it's the winter season here in California, I'm typically up in the mountains snowboarding maybe every other weekend or something like that. So that's primarily my wintertime occupation outside of science. Year round, I would say my other primary occupation outside of science is soccer. I don't play as much as I used to, but I'm certainly a pretty fanatic soccer fan. I follow

2:51most of the Champions League and Premier League, as well as MLS games whenever I can. At some point, had season tickets here to the San Jose Earthquakes, but somehow I got too busy to go to all the games. So I gave up on that. Yeah. Understandable. But I would say other than travel and science, my two other hobbies are snowboarding and soccer. I love it. Great way to get out of the lab, see the beautiful scenery out there. Yep. That's it. Awesome. Well, Eric, great to hear about life outside the lab, but I would love to talk about your work next. I know the science and the engineering work, I'm sure, keeps you quite busy,

3:24but how do you describe it to someone who's not familiar with your specific area? We're primarily electrical engineers in my group. And typically, I think when people think about electrical engineers, they think about people who work with circuits, computers, perhaps as computer engineers. And we do that as well, but we're a little bit focused more on the physics side of things, physics and maybe the materials that make up these things. My own background is physics as well as electrical engineering. I've spent some time in chemistry labs as well. So we are particularly

3:56interested in how we apply materials that have nanoscale properties to engineer better electronics. One example of this is how you would use some fancy new material that maybe chemists have recently developed to develop better circuits, better transistors, better data storage for anything that we store data, such as our computers and phones and obviously the internet. We also focus on heat, not just electricity. So we are interested in what one might do to manage the heat that a lot

4:27of electronics put out. So for example, you know, when your cell phone is downloading something or playing a video or something like that, it is getting quite warm. Your computer, when you're working with your laptop on your lap, it is getting quite warm. So what we're working on is essentially trying to manage that heat and make sure that heat is actually not as damaging to the electronic components or even to us as humans as it might be otherwise. So we're working with new materials for the heat dissipation, but we're also thinking about how to make electronics in general more energy efficient

4:58so they dissipate less heat to begin with. Excellent. Well, I know my electronics are key in my life. So these are topics near and dear to my own heart and our listeners as well. I look forward to chatting more about these projects as we go through our interview today. But I'd love to take a moment to talk about what motivates you, Eric. I think life can get a little bit crazy in science. So I love hearing about what motivates and inspires and kind of keeps scientists and engineers going. So do you have a favorite quote or a saying or a force that motivates you? I don't know if I have a favorite quote. I'm certainly a big fan of many

5:31scientists before me and a few musicians here and there and a few sports people. But I think what motivates me is curiosity on one hand. So I think I started out as a curious kid, wanted to know how the world worked. So curiosity on one hand, and that's really the physics part of me, and then usefulness on the other hand, basically making sure that what we do is actually useful to people. And that's really the engineering part of me, right? We really want to make sure that what we're doing is not necessarily just the lab curiosity. It's also potentially can impact the way we use electronics,

6:04the way we view the world and so on. Absolutely. Well, I think these are two important forces in science. I think that curiosity is key to kind of keep you diving into new areas, but also grounding yourself in that usefulness, I think, ensures that you're doing relevant work. So you mentioned that you are fascinated by or maybe inspired by a variety of different scientists, musicians and sports people who came before you. So who are some of these inspirational figures in your life? Well, I think the most important scientist that I've been inspired by is an Italian scientist named

6:35Enrico Fermi. And Enrico is actually coincidentally the Italian version of my name, but that's just a coincidence. I wasn't named after him. My parents like to joke that I was named after Eric Clapton, speaking of musicians. But Enrico Fermi was an Italian scientist that lived between about 1900 and I think 1955. And he was one of those people who was, of course, he had a Nobel Prize and all that. And I think he got it when he was 30 something. So he was one of the early quantum theory scientists. But I think

7:06what really impressed me about him when I was reading his biography was that he was one of the few scientists who was equally strong in experimental work, such as in the lab, as he was in theoretical work, meaning with a blackboard, with equations, with pencil and paper, essentially. And the more I read about other scientists or the more you sort of think about scientists in general, you realize that very often they distinguish themselves either as theorists or as experimentalists. And I think Fermi was certainly the prime example that I can think of in the last hundred years that were considered to be

7:37sort of top Nobel Prize quality, both experimentally as well as theoretically, as opposed to picking one versus the other. That kind of versatility is very, very impressive. Absolutely. So does your own work bridge these two different worlds of theoretical and experimental work? It does. And maybe that's why I look up to him. So I started out as a kid playing with lots of things, trying to understand how things work, actually really just learning about the earth and the stars. And later on, I really like building things. I like building radios. I like building telescopes.

8:07I like building small gizmos around the house. And as many probably young scientists like taking things apart as well. So taking apart household electronics and small computers and so on. But for some reason, when I went to college, I got really involved in computational and theoretical research. So my first part of my formal education, at least in college, was actually fairly theoretical and fairly computational. So I used to joke with my friends that I sort of became an accidental theorist, not necessarily a pencil and paper theorist, but I sort of became somebody who was focused on the more theoretical, more computational aspects of the work, meaning that

8:41I never really stepped into a lab until I was actually probably in my maybe mid to late 20s. Okay. Really to do research. I mean, of course, I stepped in labs in classwork and so on. But then what happened is probably towards my late 20s, as I was finishing up my PhD, I actually realized that I was missing something. So I stepped into a lab. I was fortunate enough to actually work with a chemistry professor at that point as a postdoc at Stanford. And one of the deals that we made is that I was going to go into the chemistry lab to really learn some materials chemistry. And I was at the same time going to work on some theoretical

9:16investigations that they found very, very useful in terms of the problems they were working on. So it was sort of a very nice collaboration. Of course, I was a postdoc, so I was fairly mature, but at the same time, I was still trying to learn things. So then by the time I was in my 30s, and I had my own research group, I had already worked in industry a little bit when I was at Intel, I had sort of done a combination of experimental and theoretical research, particularly on data storage at that point. And by the time that I was running my own lab, first at the University of Illinois and later at Stanford, I had realized that the most interesting

9:49work, at least to me, combines both experimental subjects. When you're in the lab, you're building things, you're measuring things. Sometimes things don't work. And on the days that things don't work in the lab, you go back to the computer and you maybe try to compute, try to simulate, try to write equations about why maybe something didn't work. Why didn't you measure what you were thinking you were going to measure? So our work today does include both pretty heavy experimental as well as fairly heavy computational and theoretical aspects.

10:22Excellent. And now you mentioned that you had a position at Intel. Can you tell us a little bit about what it was like to work there? So Intel is a big company. It's probably over 100,000 people. I don't know how many PhDs, maybe 5,000 PhDs, certainly a lot of top-level scientists. And I joined Intel straight out of my postdoc. So I was probably 30 years old. And coming out of my postdoc, I felt like a fairly experienced researcher. But actually, it turns out I was by far the youngest person in my Intel research group and the most inexperienced. So these were all people with 20 or more years of

10:56experience over me. So it was an incredible learning experience. These were incredibly experienced and seasoned researchers, if you will. And the subject was data storage. The subject was really the future of data storage. How do we store data, meaning in our computers, laptops, cell phones, in a more energy-efficient way? This was 2006, 2007. And Intel was working on a new technology called Face Change Memory. In the meantime, they've actually come up with a product, which you can go buy it on Amazon. Intel makes this stuff commercially. It's still in development. And I'm happy to say that

11:31a small part of the effort that I contributed about a little more than 10 years ago has actually made it into a commercial product that Intel sells right now. And it's a data storage technology that is very different. It's quite different from all the other data storage that we have in electronics. Most data storage in electronics is either magnetic, if you think about hard drives or tapes, or it's electrical, meaning we store charge, such as most of the stuff in our computers is still charged today. We store a small amount of charge in a capacitor. That's kind of how my music collection is stored on my phone. That's how the data is stored on my computer and so on.

12:06So this technology that Intel was working on in 2006, and it's starting to appear in products right now, actually relies on a material called Face Change Material. And a Face Change Material actually stores the data in the phase of the material. And what we mean by phase of the material is whether the material is crystalline or amorphous. So you're literally scrambling the atoms in the material to store the data that you're trying to store. In an amorphous material, atoms are scrambled. They're not arranged in any sort of logical way. They're just random.

12:38In a crystalline state of the material, of course, it's a crystal. So the atoms have a very nice crystalline arrangement. And it turns out these particular materials are switchable between the amorphous and the crystalline state. And this is a reversible, very fast transition that can be done at remarkably low power if you do it the right way. So my group has played a role even after I left Intel in sort of establishing how this can be done, what are the fundamental limits, and how this can be improved beyond even what we can do today in industry. Excellent. Well, thank you so much for

13:10telling us a little bit about some of the cool things that you were able to work on at Intel. And I know there's kind of a notion out there in academia, particularly that once you go into industry, it can be very hard to come back then to academia. So what kind of prompted this career switch for you in transitioning back into a faculty position? Well, I would say your comment is true if the work that you do in industry is a little bit more secretive. So if you are an industry and you are working on things that you really cannot talk about, then your academic profile sort of goes a little

13:43bit silent, right? And academics love to present things at conferences. They love to be very open about their research. So I think in my way, I was a little bit lucky in the sense that the work that I was doing at Intel was actually fairly open-ended research. And perhaps I took on the project because it was fairly open-ended because it sort of matched my personality. I love to talk about my research. I love to present at conferences. I probably would not be very well suited working on a top secret project deep inside the bowels of a company. So the nature of my work at Intel was not

14:15very secretive. We were allowed to talk about it with other academics. We were not necessarily the most secretive group at Intel because it was a fairly far-reaching type of research. Now, the way I ended up back in academia, frankly, I had been at Intel for just over a year when I was invited to apply for various academic positions by people who knew me. So again, it was sort of a partly serendipitous kind of transition. And ultimately, I received the invitation to join the faculty at University of Illinois in Urbana-Champaign, which is an excellent engineering school. And I decided,

14:49hey, this is a pretty unique opportunity for me to run my own research group when I'm still fairly young. So I just took that opportunity. And that's when the switch happened. Excellent. Well, Eric, it's great to hear more about your career experiences. And I think this helps our listeners out there who might be trying to figure out what their next steps are. Just to interject very quickly, one of the things that I do as part of my Stanford job is I actually speak to graduate students, postdocs, or other young scientists about career trajectories. And I very often use my own example, but also the example of other friends I

15:20have to essentially give career advice on how to tackle various jobs in either industry or academia. So this is something that I've actually thought about as well as spoken publicly about. Certainly. Well, Eric, wonderful to hear about your career experiences and to share these stories with our listeners out there. But I do want to make sure we give you an opportunity to talk about your work because I'm sure there are so many things that you're excited about that you're working on in your lab. So do you have a particular project that is your absolute favorite right now that you want to tell me and our listeners about?

15:51If my students are listening to this, if I pick a favorite, I'm going to get in trouble.

15:56Perhaps just an interesting project example. Well, truthfully, it's difficult to pick a favorite project because we have a lot of things I'm really, really excited about. And I think 2019 is actually going to be probably one of the best years in terms of my research group. We have a lot of things that we've been working on for three to four years, in fact, which are reaching fruition and maturity coincidentally right now in 2019. And really, I am excited about actually multiple projects. Maybe I can tell you about two of them

16:28just to pick sort of off the top of my head. And there's a bunch of other projects, which if you give me a call in two years, maybe I'll say, hey, those things that are really not working back in 2019 are looking great right now. And check it out. Look what's happening. But what I'm really excited about right now is actually things that we really started three, four years ago. And this is quite a journey. And a lot of these projects were actually started from a very, very small observation that either I or one of my graduate students or one of my postdocs made in the lab. It wasn't anything very big to begin with. And that's why I often emphasize to

17:00my students, you see something funny in the lab, say something, just tell the rest of us, look, I saw this funny signal. What does it mean? In fact, you asked me for a quote earlier, and I would say probably my favorite quote in the lab is what does this mean when we see a new phenomenon or a new device or a new behavior, and we literally don't know what it means. And then sort of like a nice detective process begins unfolding. And maybe more than half the time, it doesn't mean anything. Maybe it was just noise. It's not repeatable. But then my next question is, is it repeatable? Have you been able to see this multiple times, multiple days?

17:34A lot of our work is actually fairly scientific in that sense. There is a good amount of scientific discovery that goes on in our lab. And some of it is pretty serendipitous because we're combining different materials and trying to make electronics out of them. But sometimes we observe something that simply we were not expecting. Anyway, just to go back to maybe one or two of my favorite recent projects, one of them actually has to do with heat. And it's sort of a funny project in the sense that it was inspired by basically the way people block sound in real life. So if you think about airplanes

18:08or music studios, I take a lot of planes to fly places. And I think at some point, I took one of these newer airplanes that are much more silent than the older planes. And I sort of began thinking about what goes into designing these quiet planes. And then I remembered one of my hobbies or probably my major hobby as a PhD student was working at a radio station. Which is really cool, I must say. So I didn't do as many interviews as you probably do, but I have done a few interviews, mostly with musicians. Mostly I played music, I DJed my own sets and so on. But it sort of took me back

18:42to how the music studio was designed, right? It had multiple layers of glass. And not only did it have multiple layers of glass, but it had really thick glass, but it also had, if I remember correctly, other type of padding, like the soft, squishy padding that you see on music studio walls. So I realized that to block sound, really what people use is the combination of hard, heavy materials and soft, squishy materials. So I realized that we could replicate this at the nanoscale with some of the nanomaterials that we work with in our lab. And some of the materials

19:13that we work with are literally one or two or three atoms thick, which is incredible. We have these materials laying around, we've been working with them for a few years. But I realized that these materials actually have very, very different densities. Graphene is the lightest one of them. Graphene is just one sheet of carbon atoms. So carbon is element number six in the periodic table. It's very, very light. But then we work with other materials like tungsten selenide. It's actually even difficult to say almost. Tungsten selenide is WSE2. Both tungsten and selenium are actually very, very heavy materials. And if you layer

19:48one layer of graphene with one layer of tungsten selenide, and then we threw in a bunch of other of these atomically thin materials. Again, I want to emphasize, these are atomically thin layers, one to three atoms thick. Wow. So we realized that if we layer these things, we can block not sound necessarily, but heat. Heat essentially is just a manifestation of sound essentially on the nanoscale. Sound, as you know, is the vibration of air molecules. And heat turns out to be the equivalent vibration of molecules or atoms inside a solid material. So the atoms in a piece of glass are vibrating a

20:22certain way, and that's what's carrying the heat. Same with plastic, same with wood, same with the materials that we're working with. So what was the realization that I had about sound installation, music studios, airplanes, and so on, became essentially a demonstration of how to block heat, but over dimensions that are only a few atoms in thickness. So we demonstrated at some point late last year, using a material that's about 10 atoms thick, we were able to block heat as well as other

20:53materials which are hundreds, if not thousands of atoms thick. So we could block heat maybe almost a hundred times more efficiently than even some of the best thermal insulators that we have around in the lab. That's amazing. I think it's a very cool thing. And I like this project because it sort of started from a very benign real world observation. Hey, how do people block sound in real life? And then we applied that in the lab to the atomic scale. And it's also one of these projects which has both an experimental component because we demonstrated in the lab, but it also has a nice theoretical component because

21:25you're sort of wondering, gee, why the hell do these atoms block heat so well if you just layer them with varying atomic masses? Why are the different densities of atoms really blocking heat so well? So we actually had to go through quite a bit of computational and simulation work and some pencil and paperwork to really try to prove to ourselves that what we are measuring was not just a fluke. And it's also one of these projects where it's kind of a neat fundamental thing where, look, gee, we were able to block heat 100 times better than some other way is possible.

21:58But it has applications as well. And actually, the applications are maybe not immediately obvious because I've noticed when I talk to my colleagues about it, my colleagues' first answer is, gee, Eric, why would you want to block heat? Don't you just want to disperse the heat more quickly instead of blocking it, right? And yes, of course, in many, many applications, you would say, well, I just want a better heat sink. I want to slap on some copper or some diamond that will just disperse the heat a little bit better. But in some cases, you actually want to block the heat. And the applications for that are twofold. One is actually going back to what I used to work on at

22:32Intel and phase change memory. To make that memory more energy efficient, you actually need to be able to thermally insulate it better. Because the phase transition, the actual crystalline to amorphous and reversible change is actually done by heating the material through a quick voltage pulse. And the better you can thermally insulate the material, the lower the voltage pulse that you have to put in to basically change the phase transition. So it's very, very important for a certain type of energy efficient memory. And then the other application is, again, in electronics,

23:04notice most of the applications that we can think of certainly are electronically driven because we are electrical engineers, is you have, again, cell phones, laptops, and so on. And sometimes you really, really need to insulate a certain component from the heat that is produced by another component. And just like you would need an electrical insulator, like for example, if you've ever done any kind of wiring work in your home, or you've seen that the copper wires are insulated in a plastic jacket. Same thing for heat, right? If we have a really, really good thermal insulator,

23:34we are A, able to protect ourselves from the heat that is generated somewhere. But also, we are able to essentially divert that heat to go in other ways. And being able to not only protect, but also divert the heat is actually very, very important, especially in very, very tight geometries like inside your cell phone. If you take apart your cell phone, you realize it's completely crammed with components. And you don't want the heat from one component to necessarily fry the other component or fry your display or fry your hand. So being able to actually route heat almost at will inside a

24:08very tightly packaged electronic component is actually quite important. Because heat, unlike electricity, is actually a lot more difficult to control. Certainly. Well, Eric, this is a really cool project. And I know this is only one of two. So what is the second one you wanted to mention today? Right. Yeah. I said, if I pick favorites and my students are listening to this, then I'm going to get in trouble. Well, this project really comes to mind because it's something that we've actually recently submitted for publication. And it's actually still in review. I mentioned that we work with these materials, which are one to three atoms thick, graphene being

24:43one of them. Another one of these is molybdenum sulfide. That's a molybdenum metal in the middle and a sulfur atom above and below it. And this is a fairly new field in the world. Up until maybe about five, six years ago, these materials were not even being studied, at least not for electronics. They were being studied for other applications, more for catalytic applications. But really for electronics, they really just sort of exploded onto the scene a few years ago when people were able to isolate them in single layers, meaning a single layer of this material is really

25:14three atoms, one of the molybdenum atom in the middle and then sulfur above and below. It's moly sulfide top and bottom, MOS2. And there's already been proposals that these materials actually have pretty good electronic properties. People have even said, hey, maybe you can use this to sort of make a nanoscale form of circuits out of it. People make nanoscale circuits out of silicon, but of course, we're not using three atom thick silicon to make the circuits. The silicon is still pretty thick. But people have said, maybe you could use three atom thick MOS2 to make nanoscale circuits. And there's been lots of

25:49demonstrations of this actually over the last five, six years, but none of them have actually been nearly as good as silicon. And yet people persist. In fact, the theorists, people who do computational materials, they predicted, hey, this material actually might be better than silicon in some applications, but nobody had been able to demonstrate this experimentally, at least not fully. And I don't know if I necessarily claim that we've been able to do it, but in a certain way we have. So very recently, we've been able to get this material really in the shape of nanoscale

26:19transistors that are maybe just 10, 20 nanometers across, which are much, much thinner than the silicon counterparts. So remember, these transistors are basically three atoms across, which is about as thin as you can make any electronic component. Yet they behave almost as good and in some ways better than the silicon transistors that we have in today's electronics. So again, I think there's still a lot of work to do because there's still some challenges in really dialing in the properties. But I think we've recently, just in the past one year or so, gotten these three atom thick materials where

26:53they're actually about as good as silicon and really in some ways actually even better. And the ways they're actually better is that transistors are electricity switches, right? So when a transistor shuts off, it really needs to shut off electricity. And it turns out nanoscale silicon transistors, they're sort of like leaky faucets these days. They don't fully shut down the flow of electricity. So when you put your computer to sleep or your cell phone to sleep, it actually still very slowly eats away at your battery, which is very, very important because you might think your laptop's

27:25asleep, but actually in a couple of days, maybe up to a week, it'll drain the battery because all the components inside are actually leaking very slowly, but leaking. By the way, this is just quantum mechanics at work. Sure, sure. So what we've demonstrated is that the way in which these materials actually truly are significantly better than silicon is when you shut them off, when you shut out the transistors, they really shut off. And by that, I mean, they shut off like about not a factor of two, not a factor of a hundred, but a million times better than silicon. Oh, wow.

27:56So when they turn on, they're not necessarily better than silicon. Maybe they're comparable to silicon, maybe within a factor of 20% of silicon or something like that, right? So in terms of raw performance, we're not necessarily improving upon silicon by a whole lot. We're fortunate to have gotten roughly where silicon is. Now, of course, I should mention that silicon's been around 50 years and there's a trillion dollar industry behind it. And of course, a lot of optimization, academic groups and people in general have been working on these three atom thick materials for only five or six years. So there's obviously a lot to be learned and done, but we're almost there in terms of performance.

28:30And there's still some problems, but we demonstrate that when you shut them off, these things turn off really well. And fundamentally, it has to do with the way these atomically thin materials carry electricity and how they actually really block electricity. So maybe it's a coincidence. Maybe it's just the fact that you're talking to me in March of 2019. But in a way, the two projects that I'm actually, in a way, I'm most excited about, they both involved, one involved actually blocking heat and the other one involves blocking electricity in some very, very creative way that is in some way a world record or something like that.

29:03And being able to conduct heat or conduct electricity is just as important as being able to shut it off and to shut it off really well when you don't want that electricity or that heat to flow in unwanted ways, particularly electricity, which flows in unwanted ways. It basically drains the battery of your devices. Certainly. Well, Eric, these are both really exciting projects. And I love that in explaining them, you touched on some of the challenges. And I think even just the time it takes for these projects to come to fruition is a challenge in the field. So I love talking about these challenges

29:34that scientists and engineers face. Do you maybe have a favorite story of a failure or a challenge or just a difficult time that you've been through in your career that you can tell me and our listeners about and give us a walkthrough on how you got through it? Well, sometimes I joke with my students that being able to survive failure is very important in our research. And I live in Silicon Valley, so I often hear startup people saying fail early and fail often or something like that, right? And fail cheaply too. So for every success that we have in

30:04our lab, we have many, many more failures. And I think that's an important thing to keep in mind because I think especially when we talk to new students, when we talk to even to each other, or when we talk to the media, or heck, when we talk to our parents and our friends, it sounds like we have all these great successes in the lab, but actually... They're much rarer. Yeah. Exactly. And actually, we have a lot more failures than we have successes. And what I tell my students is that really, you have to develop a certain thick skin as a researcher and a sort of ability to move on from these failures and not let them affect you,

30:37right? Because a lot of young researchers, and I'm hoping some of them are listening, may be extremely successful at doing homework. Our undergraduate education system in the US is great. We have some top institutions, top in the world. But if you notice, a lot of it is just homework, homework, homework, right? And then when these brilliant students go to graduate school, myself included, we go in the lab or we go in front of the computer and we're trying to solve something that, A, nobody knows if it has a solution. It's a research problem. So the advisor did

31:07not know if there is a solution, right? When you're working on homework, the TA knows there's a solution. The professor knows there's a solution. And not only do we not know if there's a solution, but whether there's a solution or not, maybe we're going to work on it for a few months or six months or even longer, and we're just going to fail. Then the solution we come up with will be basically not very good, not very optimal, or maybe not a solution at all. So sometimes I tell my students, look, if you fail, it's fine. I failed more times than I can remember. But it's very important to sort of always ask yourself, hey, what did I learn from that failure?

31:38And to apply that to your next project and to try to avoid that failure again, then sometimes even just recognize that you failed. And I think to be a successful scientist or engineer, it really takes a certain personality in a way. I know this maybe sounds cliche and it's maybe overplayed, but you need grit. You need a little bit of grit in the sense that you need to be able to say, I went to the lab today, nothing worked, but I actually learned a lot from that. And I'm going to do things better tomorrow. I'm not going to let this tell me that I'm a bad scientist or whatever, because all of us fail very often. And sometimes failure is due to,

32:13oh, we made a mistake because we were being naive. And we really ought to talk to one of our collaborators who has done something like this before. But very often is we're working on very, very difficult things that nobody has solved. And pretty much all great engineers and great scientists have failed a lot more than they've succeeded. But actually, the failures, I think, are equally important. And they're huge, huge learning opportunities. Certainly. Well, Eric, thanks so much for talking about how you view failure in your own laboratory and in some of your past experiences going through this process yourself. I think it's really helpful

32:47for our listeners out there to hear that maybe they're not alone in this going into the lab and feeling like you're failing so often. No, no, no. This is very important. Occasionally, I think scientists and more of us really ought to step out and literally just publish papers and publish books, maybe just about our failures, because I think you could fill up more books with failures than you could with successes. There's a funny example. A few years ago, there was an article about how J.K. Rowling, the writer of Harry Potter books, how she published all her rejection letters on

33:19her website. So before she was able to publish a single Harry Potter book, she had been rejected for something like 40 or 50 publishers. And she scanned in all the letters, the PDF, and basically posted them all online somewhere. Kind of an inspiration to young writers everywhere. To me, that actually was very, very inspiring in a way. And I forwarded that link to my students right away. And I said, look, this is one of the most successful writers of all time. And she got incredibly rejected at the beginning. And she probably went back and maybe improved the writings. Maybe she just persisted. Obviously, writing Harry Potter is not exactly the same as doing science, but there's

33:53certainly some similarities in the process that there are successes and failures, right? And persistence is very, very important. Absolutely. Well, Eric, again, thank you so much for talking about some of the challenges you face. But we don't just want to talk about what's hard. I'd love to talk about some of the successes, some of the wins that you've had. So do you have a favorite story of success you can share with me and our listeners today? I don't know about favorite, but there's certainly a couple of things that stick out, at least from our past research. One of the things that comes to mind

34:25is the time about eight, nine years ago, when we were working on a very, very interesting question about a certain data storage technology. And this data storage technology had been studied at Intel and Samsung and some other major companies. And in fact, it's actually already made into production. It's something called phase change memory. It's sort of an interesting way of storing data by looking at the phase, meaning the state of the material. So if the material is amorphous, then the resistance is high. If the material is crystalline, then the resistance is low.

34:57And therefore, you can store data that way, as you can imagine. I mean, you can read it out electrically, you can store it electrically just by converting this material with small heat pulses, voltage pulses between an amorphous and a crystalline state. And the interesting little success that we had several years ago is we were able to take this technology, which at the time, it was almost a commercial technology. Now it is. And we were able to shrink it by about a factor of a hundred. Oh, wow. Usually that's a pretty big number to take any kind of electronic or any kind of

35:30technology that's fairly current in an academic lab and then to shrink it by a factor of a hundred. So the idea that we had was to use carbon nanotubes, which are the smallest known metallic conductors and to use the carbon nanotubes as the electrodes for this memory. So industry typically uses metal wires as the electrodes and metal wires are actually difficult to shrink down to the nanoscale. But carbon nanotubes, they're basically just a few atoms across and they're just shaped in a little tube of carbon atoms. So the diameters are basically

36:00about one nanometer, which in perspective, you'd have to put 50,000 of these across to make the width of a human hair. So they're about 50,000 times narrower than a human hair. So we took these carbon nanotubes and we use them as electrodes. Like I said, sometimes the biggest breakthroughs come from the simplest ideas. So in this case, the simple idea was let's take the smallest metallic conductor that we have, which is a carbon nanotube, use that instead of the metallic electrodes, which industry was using for this memory. And therefore, we were essentially able to shrink the volume of the memory cell by about a factor of 100. And therefore, we also showed that

36:36we could shrink the programming energy of the memory cell by about a factor of 100, which very quickly was adopted both by industry as well as by academia as not just a world record in terms of what is possible with this memory technology, but also really pointing to the fundamental limits. We hadn't reached the fundamental limits of the memory. We could probably still make it a little bit smaller, but we got very darn close to them by using the carbon nanotubes as the electrodes for this memory technology. These results are still, in fact, the smallest ever produced with this memory technology.

37:09They're not commercially feasible yet because it's very difficult to incorporate carbon nanotubes in a large-scale manufacturing operation. But what I like about this result is that, first of all, we took a fairly serious, almost mature technology and we showed, hey, look, it could be improved by another factor of 100, which is a big, big win. Secondly, that it was actually a fairly simple experiment in retrospect. So it's pretty simple and elegant and obviously very reproducible. And third is that it really pointed and actually still points the way towards the fundamentals

37:41of this memory. And that really got us to think about, you have this technology, really what is fundamentally the smallest memory bit that you could program with this material technology and really points the way forward towards the research that is still ongoing, both in our lab as well as other industrial labs and frankly, with a commercial technology. So those three things combined, I think, are very relevant and very interesting in terms of not only a nice scientific breakthrough, but also one with very significant technological

38:11relevance for industry. Certainly. Well, Eric, congratulations on this remarkable success. And we've talked about a lot of the cool projects you've been able to work on over the years. I know as a scientist, you're constantly immersed in the literature and reading every chance you get to stay up to date on what's happening. But I love to talk about what scientists are reading for fun as well. So do you have a favorite book that you can recommend to me and our listeners, whether it's related to science or not? I do, actually. And it's partially related to science, but it's not really the science that I practice. It's more related to statistics than anything else. And it's related to my favorite

38:46sport, I suppose you could say, outside of the lab. This is a book I was actually reading last summer. It's called The Numbers Game, Why Everything You Know About Soccer Is Wrong. And it's a very interesting book about soccer statistics and how explaining why certain soccer coaches make big differences, why certain soccer positions are valued more than others, why certain players may be underpaid compared to others, and really looking at decades and decades of data on soccer players. And it's fairly up to date. I think it goes up until the 2014 World Cup. So to me, that was a

39:20really fun read, not just because it's about soccer, but it's also because it's a very mathematical and statistical treatment of soccer and sort of explaining some of the trends that you see with hard numbers. I like it. Combining multiple passions of yours. Yeah. Awesome recommendation. I will put it out on our website for our listeners to find there. And we've chatted a little bit about some of the places you've been able to go for your works and the opportunities that you've had to train in different places. And travel is a big part of the life of a scientist. So Eric, do you have a particularly memorable travel experience that you'd like to tell me and our listeners about today?

39:53Exactly. As you pointed out, one of my favorite things about this job is receiving invitations to go to interesting places for conferences and sometimes just to visit collaborators. And I've been lucky to travel to Japan, South Korea, China, Brazil, many places in Europe, of course. But my favorite two trips, both happened in the past year, actually. One was there was a graphene and two-dimensional materials conference in San Sebastian in Spain. And San Sebastian,

40:23I had known much about it from the Anthony Bourdain show, as it turns out. Oh, yeah. Because he used to love to go there for the food scene. It's a beautiful, beautiful town on the northwest coast of Spain. Right near the border was France, actually. And also, not only was the subject of the conference obviously something very near and dear to my heart, but the food scene in San Sebastian is incredible. And also, I was lucky to be joined by two of my graduate students who were able to come with me on this trip. So the three of us went on hiking trips around town. We sat outside

40:54and enjoyed the pinchos, which are the local tapas. And we had really, really nice conversations with all kinds of new and old collaborators and just the environment in general. I remember one of my graduate students, it was actually his first conference trip. And he said something like, is this what all the conferences are like? I said, this is actually probably one of the best ones I've been to. So it's all downhill from here. I'm so sorry. You peaked early. You peaked very early. That's right. Yeah. But it was very nice. It was very nice to hang out with them and the collaborators in such a nice setting. The other trip, which actually I just took this

41:29past December, which I think the environment was not as nice, but it was very strange for us. And again, I took the trip with a postdoc and a graduate student from Stanford. So we hung out a lot and ate weird foods. There was a conference in Levy or Levi, which is a small ski resort in the north, north, north part of Finland. And because it was in December, it was dark except for roughly two hours during the day. So we had a bit of light between 11 a.m. and 1 p.m. and the rest of the day was dark and snowing. Oh, wow. Yeah. So first of all, not the most cheerful environment,

42:05but a very, very interesting, strange foreign environment for us. It was not very cold. In fact, the Finnish people were saying it was actually one of the warmer winters they have seen. It wasn't very snowy, but it was sort of the constant darkness and the fact that every meal contained reindeer. That was unusual to us. So breakfast, reindeer patties, reindeer salami, lunch, reindeer meatballs, dinner, reindeer hamburgers. We were surprised we were not served reindeer beer or something like that. I wouldn't say I would necessarily go back to the constant

42:38darkness of Lapland. But certainly as a memorable trip, it's definitely up there. And San Sebastiano, of course, is very beautiful, I think, year round. So happy to look for another conference there. Absolutely. And I think going to some of these places in the world that you might not otherwise think to travel is part of the cool thing about attending some of these conferences. That's right. I don't think I would naturally book a vacation in Lapland, Finland, not necessarily on my own. Right. Well, Eric, thanks for sharing some of these travel experiences. And I think it's fun to highlight just some of the opportunities that scientists get in this career, because listeners

43:08may not be familiar with what life is really like as a scientist, because I think sometimes the way that scientists and engineers are portrayed in the media is not necessarily true to life. We don't meet all of the stereotypes. So I love trying to break some of these stereotypes on our show by talking about this human side of science. So Eric, do you have some fun traditions or just quirky memories or funny things that have happened in the lab that go contrary to some of these main stereotypes? I think in general, I'm happy to say that my lab is pretty full of jokey people. There's a lot of

43:39laughter in group meetings and a lot of jokes being thrown around. People have nicknames for each other. But I think one of my favorite recent memories that come to mind, maybe I'll just mention two of them. One is the summer barbecues and winter ski trips that we take with the lab. They're not necessarily funny, but it always leads to laughter and a lot of jokes. And I jokingly say, of course, we ban talking about science for the barbecue. Like, that's it. Let's just play soccer and volleyball and talk about other things like what books people read and what sports they play.

44:11In terms of more of a funny event, I guess in the recent ski trip, there were some funny events in the sense that one of my new students was from Bangladesh. It was the first time he had ever seen snow. And him and his wife, also from Bangladesh, who had also first seen snow. I have this really fun picture of the two of them in Lake Tahoe, surrounded by about 10 feet of snow. Oh, wow. Because it had been snowing that much. So it was also the first time they had seen not just snow on the ground, but snow falling from the sky. And it was sort of fun to hear the other students sort of explaining to them where this stuff comes from.

44:44All the different activities you can do. Yeah. That's right. And so we put them on skis as well. So the other students took care of the ski lessons and they tried to put them on skis and make sure they didn't get hurt. But in the lab itself, there's a student in the lab who is a pretty good doodler. She draws cartoons of people. So a few years ago, I walked into the student office and on the whiteboard that normally is covered with equations and device diagrams and whatnot, there was basically a fully formed cartoon of everyone in the group. So about 20 people that she had drawn on the board in stick

45:17figures. But these are stick figures that were so well drawn that you could immediately tell, oh, the person that always wears the headphones is Mike. The stick figure with a plaid shirt is Chris. You know, like they were very, very well drawn. And ever since then, that drawing sort of become like a group logo. I mean, it's not on the website or anything, but it shows up in people's presentations when they acknowledge the group. We've printed it out on mugs and given it to group alumni. The only thing I think we haven't done is put it on a T-shirt. I love that. It's like your ever-evolving lab family photo. That's fantastic.

45:48That's right. I think I told her this year she needs to update it with a few new lab members that have come in. So that needs about a yearly update, that cartoon. And I don't know what we're going to do once she graduates. We're going to have to continue the tradition somehow. Exactly. Find another artist to join the lab, right? That's right. Well, Eric, I love that your lab has this fun side to it. And you guys do a lot of things that are outside of these stereotypes. And it's not just all work. You guys are able to get out and hang out together at these barbecues and these ski trips and stuff. I think that's so important. And I know in the work that you do in the lab, you're very serious and you're addressing

46:18some amazing problems and answering some really interesting questions. And I imagine that there's a thing or two, Eric, standing between you and being able to answer the dream questions you would love to answer. So if we took away all of the constraints, if funding and staff and technology and feasibility were not issues today, what is the one question you'd want to answer or problem you'd want to solve? Oftentimes on a trip, when your mind just kind of wonders, this is exactly the kind of question that I ask myself, because it's very important to sort of take stock of what you're really doing with your research and say, look, what is the really big picture? What is the

46:52really important problem that I'm really trying to solve here? As opposed to a nitty-gritty that we might publish this week or the next month or whatnot. And the first thing that comes to mind is, again, related to these two-dimensional materials that we work with, like graphene and molybdenum disulfide. These are these materials which are one to three atoms thick. Graphene is one carbon atom thick. Molybdenum disulfide is three atoms thick. Molybdenum in the middle, sulfur top and bottom. And you can imagine these just being films, right? So these are material films that you can cover entire surfaces with. And these are not the only two materials that we work with,

47:27but these are sort of the two representative ones. And I specifically want to focus on molybdenum disulfide, which is a two-dimensional semiconductor. It's a three-atom thick semiconductor. And the question that I really, really would like to answer, and we've seen glimpses of this. This is why this is the first one to come to mind, because even though we have some rudimentary lab instruments to look at it, and we have sort of limited funding, obviously, to look at it, we've seen glimpses that this may be possible. I would really like to know what the fundamental properties of this material are from an electrical point of view. And we're starting to

48:01get there, but I think we're still pretty far from there. Meaning, how much better can this material get compared to the way it is in our lab right now, in terms of electrical properties? So that's one thing. The second thing is being able to grow this material on demand on essentially any substrate that will support the growth. Right now, this material is grown at fairly high temperatures, about 700 Celsius. Obviously, you can only grow it on substrates that support that kind of temperature. So glass, maybe silicon, maybe sapphire, quartz, these are substrates that will

48:35be able to go in the furnace and survive those temperatures. But I would love to bring the growth temperature down, preferably below 500 Celsius, in which case a lot more substrates, including essentially fully patterned silicon wafers will become an option, enabling us to basically integrate this material with silicon circuits. That kind of takes me to number three. Why I'd like this is because fundamentally, we believe this material and some of the other related materials that are in this family actually have much better electrical properties that people realize right now. And I

49:07personally, and a few others in the community, we don't necessarily see it as a replacement for silicon. I think silicon is here to stay. People have often said, well, are you going to replace silicon with graphene or with carbon nanotubes or with whatever other material? And I think the answer to that is probably not because silicon is very cheap and it's good enough for what it does. It's sort of like saying, are we going to replace the jet engine on airplanes? I mean, there's other ways to fly, but the jet engine has been here for 60 or more years with various

49:38improvements. It works. It works really well. It gets us from here to New York in six hours or so, and it is what it is. Sure, we have the Concorde and supersonic flight, but you see it's sometimes technological advances are just much more expensive and people are just not willing to pay for them. So there goes that. But going back to this material, I think it wouldn't necessarily be used to replace silicon, but it has certain properties, especially the thickness, the really good electrical properties in the three atoms thickness that silicon cannot match. And there's

50:08certain applications where we could integrate the material with silicon, such as for memory controllers, or really just for building out chips, not just in two dimensions. Today's computing chips are basically a circuit pattern, not a silicon chip, but it's a lateral circuit. There's no vertical circuit really. So I think this material, if it gets good enough, and if we can grow it at low enough temperatures with high enough quality, it will allow us to essentially embed it onto silicon chips and start building those chips vertically. So sometimes I like to remind my students and my colleagues that

50:42you've heard of Moore's law, obviously, and you've probably also heard that Moore's law is dead or dying or slowing down or whatever, because silicon is approaching whatever limits. And what I like to remind people is that Moore's law is not dead or dying. It's probably going to go 3D. If we have our way with these new materials, we're going to start building these chips vertically. And some of these materials we think, we hope, can be built at low enough temperatures where they're not going to disturb the silicon circuitry underneath. So we're still going to have the silicon circuitry underneath,

51:13but we're going to be able to build more circuitry and more complex functions, functionalities such as sensing, but even more memory and even computing units on top of the existing silicon wafers. Right now, your cell phone or your computer, whatever you have, many, many chips inside. The major ones being the computing chip and the memory chip. And actually, those two chips will merge, I think. And you'll actually really have tremendous amounts of memory directly on top of the computing chip. And that's going to not only give you a lot more data storage, meaning terabytes on your chip,

51:45but it will also reduce the energy consumption and therefore extend the battery life of everything that we have. So I'm pretty confident that technologically, this is all possible. What I'm not sure about, and if I had a crystal ball, I'd be great. But what I'm not sure about is, can we do it cost-effectively enough that it's going to survive the economics of manufacturing and all of that, right? And that's something that we don't really know yet. The other thing that I will mention very quickly is, we have this interest in thermoelectric energy

52:15harvesting. And if money and staff and bandwidth were not an issue, we would really love to push forward some of the thermoelectric energy harvesting materials that we work with, and really try to actually think about how we could harvest thermal energy cheaply and use it to power sensors, watches, and maybe even phones. Phones are a little bit more energy hungry, but I think if we combine some of the advances in energy-efficient electronics that I mentioned earlier with advances in energy harvesting, I wouldn't be too surprised if in a while we might be able to actually have

52:52ambient electronics, even cellular phones that are powered by just sort of ambient temperature gradients, even sort of the gradients of temperature from the human body. That actually can be done today on some level. People can buy thermoelectric, so basically heat-powered watches. So it can be done even today, but really doing it at a scale and at cost where we can power sensors, heart rate monitors, all these things that people talk about in terms of personalized medicine, all of that could actually be powered almost purely by body heat, which I think is a very, very interesting prospect

53:25that we would like to work on. Certainly. Well, something to look forward to. That would be absolutely amazing. And Eric, thanks for sharing some of these dream research projects with me and our listeners today. And you've also shared a ton of wonderful insights and stories with us. And I'd love to ask for one more thing, and that is advice. So is there one piece of advice that you received at some point that really helped you that you can share with our listeners today? One of the things, and you kind of hinted at this earlier in the conversation as well, is that scientists, maybe sometimes they're not portrayed accurately in the media. And I think part of that

53:59is our fault, obviously, because sometimes we maybe don't communicate what we do well enough. And this is why I think talking to you is so important. And by the way, I thank you for helping us communicate what we do. But I think some of the most important advice that I received, which maybe at the time, I didn't realize how important it was actually has to do with communication. It was sort of career defining advice in terms of how to more clearly communicate your research. And I don't even remember who gave me this advice or where I read it. But it was really about every

54:30time before you go out, and you give a presentation or a lecture or anything like that is really trying to understand your audience. Are you speaking to an audience of high school students? Are you speaking to an audience of undergraduates, your colleagues in the department, really trying to actually think as your audience? And I think it's ultimately about human empathy and trying to understand how do they process this information? What is the best way to transmit the really cool thing that I'm excited about to them? And again, I wish I could encapsulate it in a simple saying.

55:03But it's something that I picked up on, I think, pretty early in my career. And I've done my best to kind of read up on. And every time I come across a new link or story or video on the internet about how to communicate science better, you know, I forward it to my students. And I say, look, pay attention, you need to improve this. Because even if you let's say you don't work as an academic, and obviously, as an academic, part of my job is to be in science. Even if you are, let's say, deeply embedded in an R&D lab in industry. Communicating what you do is extremely important.

55:34It'll define your career. Being able to convince those around you that your ideas are worth pursuing or good, or maybe even if a lot of research is bad, is incredibly important. So I think really focusing on and improving our communication as scientists is probably one of the best advice that I've been given and I give to my students. And that obviously includes not just verbal communication, but it includes written as well as graphic communication, the way we present our figures, our research photographs, our manuscripts, the way we lay down our ideas, really trying to

56:05put them in the most clear, effective way possible. Excellent. Well, wonderful message to share with our listeners today. If they want to learn more about you and the wonderful work that you do, where should they go or how should they get in touch? I think the best way to start is our website, which is very simple. It's pop lab, pop like my last name, P-O-P lab, L-A-B in one word, dot stanford.edu. So again, that's pop lab dot stanford dot edu, or they could simply Google me and come across it. I'm a big believer in making

56:36our science accessible, even our sort of nitty gritty science. So a lot of our publications or almost all of them, wherever I can, are directly linked on our website. So you can download them directly. A lot of our presentations are there. Some of our research codes are downloadable. Research movies. And of course, you can read more about the people in the lab. There's links to our students, my own short profile, and obviously the contact emails for all of us. So if people really want to get in touch with us, email is probably the best way.

57:07Perfect. Well, listeners, definitely check out Eric's lab website. And Eric, we appreciate the outstanding work that you're doing, as well as these efforts that you've made to disseminate your work. And it's been such a pleasure to chat with you on the program today. So thank you for your time. Thanks a lot, Maureen. Absolutely. And listeners, great to have you here as well. We'll see you next time on another episode of People Behind the Science.