Nanophotonics: Ready for life and work

Xinghuo Yu:My name is Xinghuo Yu. I'm the chair of RMIT Professorial Academy, which was just recently established by the Vice-Chancellor basically to give university an opportunity to have this kind of think tank to advise on university policies, to look at the features, and also to be ambassador to promote university.

Xinghuo Yu:Today is the one occasion, one the thing we do is really to celebrate some of the achievement of outstanding researchers, and also this lecture is another way is promote particular strong areas, and in particular is the impact for directions.

Xinghuo Yu:It give me a great pleasure today is to introduce distinguished Professor Min Gu, who is going to deliver the first inaugural RMIT Distinction Lecture. This is a great honor, but I don't want to read Min Gu's very long CV. Certainly he's one of the top researcher in the country or in the world as well in photonics.

Xinghuo Yu:He has many honors. As you know, that he is the fellow member of the Australian Academy of Science and Australian Academy with Technological Engineering and Science and also for member of the Chinese Academy of Engineering.

Xinghuo Yu:Today he's going to talk about Bio/Nanophotonics: Ready for Life and Work. I think this is going to be very easy to understand. I know there's concern about this, so sophisticated. Even all distinguished professor found it a bit hard, but I think Min is going to explain it very plain term, and particularly this is Friday, just to make sure that we do something entertaining. Without any delay, please join me to welcome Prof. Gu.

Prof. Min Gu:Thank you. Can you hear me? Okay. Thank you very much for coming to this inaugural lecture for RMIT Professorial Academy. I was very honored to be invited to give a first talk when Xing came to my office and say, "Well, you should give the first talk." I said, "Someone else should do it." I thank you very much for I'm pleased to come here to talk about very difficult topic, nanophotonics, and I'm trying to make it as simple as possible.

Prof. Min Gu:Now, whilst I was prepare this talk, I trying to minimize the mathematics. I know people hate mathematics, but eventually I found I have to include one mathematical formula. I'll hope you all enjoy that mathematical formula and why this is important for Ready for Life and Work. You will see mathematics. If you coming to say, "I don't want to see mathematics," close your eyes for this.

Prof. Min Gu:Let me first start that I would like to acknowledge the people of the Kulin Nations on whose unceded lands we are meeting today. I will respect if you acknowledge their elders' past and present for this particular occasion.

Prof. Min Gu:On this occasion, I also want to acknowledge that the support I have from my research group. It is this group, and they make me possible today I'm still speak whilst I'm still doing the ... and commit a lot of time in the administration roles in my RMIT. Thank you very much to those people in the audience and I really appreciate over the last three years the support.

Prof. Min Gu:I came to RMIT beginning at 2016. Thanks, RMIT, for the support. I had the lab opening at the end of 2016. You can see the photos. [Karin 00:04:05], you're there, and the Vice-Chancellor there. We had a lab open at the end of 2016, but we created a world-class facility in order to make this complicated nanophotonic device.

Prof. Min Gu:Having been in RMIT, it is important to aware that the Ready of Life is the strategy. Last year, we create a second laboratory, which is called the Photonics Technology Translation Facility. What I'll talk about today, most of the work has come from the second lab, but it's very important to have the first lab in order to produce the excellence. Then we can translate into the impact of what we talk about.

Prof. Min Gu:Now, let me first start to say, what is the photonics? In fact, the photonics is the science of light. We have light in the room, and that's the photonics. That's very easy. Now, how important the photonics. Two or three weeks ago, there was a Prime Minister Award. If you look at the award, the most important award, the Prime Minister Award Prize for Innovation.

Prof. Min Gu:This year, this prize awarded to these four gentlemen. If you look that what they have done for this country, switching light for fast and more reliable internet. It's very important to realize, when we talk about the light, immediately is related to the life we are living, and it's also the concept immediately coming to the mind the internet.

Prof. Min Gu:These four gentlemen, what they have done is, if we think about what we learned in high school, Newton's Prism. If we want to resolve the sunlight into multiple color, seven colors, so we learned in high school, but what they say is they can resolve into 100 colors. Color is very important.

Prof. Min Gu:What that means, they can actually send these multiple colors through this optic fiber. That's where link everywhere from the city, Melbourne, from the internet, from the data center, to the homes. It is important, notice this fiber is now enabling our everyday life, but it's also we should actually acknowledge that the person who invented the fiber is actually receive the Nobel Prize in 2010 as the physics. They can see how mathematics and physics important. He passed away, unfortunately, last year.

Prof. Min Gu:That is light, internet, fast, reliable. That's the way how this important for these four gentlemen. In fact, one of the people, Steven, actually helped RMIT to evaluate some of RMIT's technology in these areas as entrepreneur.

Prof. Min Gu:Now, I just touch base that color is important. More color, more information, more movie, more data you can do, and faster. Now, there is a limit. That's why we have a bandwidth, broadband network. Bandwidth means how many colors you can do.

Prof. Min Gu:Apparently you can say, "Why don't we increase the color? Thousands of color?" Unfortunately, we do not have this capacity. There is a limit. How many colors we can send through the fiber, limited by the material, limited by the capability how many we can produce the color.

Prof. Min Gu:There is one things over the last few years. These things. This is actually invented about four, five years ago. You can see it's twisting. Twisting, it means a very important, another equivalent color. You can think about color, but it's not called color. It is actually call the angular momentum. It is angular momentum that increase the additional capacity.

Prof. Min Gu:This would be another revolution in telecommunication, but the issue is, how can we put this twisting, which is called the optic angular momentum, into the fiber? Now what RMIT's contribution 2016 is we know the group from Boston, they can put this twisting, which is angular momentum, into the fibers. That means that we now extend the bandwidth, but the question here is, can you use that twisting to carry the information?

Prof. Min Gu:What we did is say, "Well, yes, we can actually now coding the information into this kind of twisting light through the fiber." That's important to prove that concept. It's very fundamental science concept, but the next question is, can we detect it? Can we see it?

Prof. Min Gu:Now if you can transmit we cannot see, then, this useless. This is why recently over the last few days there's another round of media RMIT released that we have detector, we play the detector. How small the detector? Like a human hair. We can actually reduce traditionally to detect this kind of light from the tabletop, dining table, to now the human hair size. That is very important, another step as we are now we achieve this year.

Prof. Min Gu:Now where this go? What this eventually give to the Life and the Work, what this go? This is the future. Li-Fi. If you haven't heard that, this emerging technology is coming. What that means is that everything you see here, the light in this room, eventually becomes the Wi-Fi.

Prof. Min Gu:To do that, we need significantly extended bandwidth. Color is not enough. We need a different means. Anything related to physically you can actually design to this information, they are important. Angular momentum is emerging as very important. In fact, this project is being fund by the ARC Discovery Grant this year, start from this year.

Prof. Min Gu:I'm not talking this. I'm going to stop here, change to more general: the impact of the light. Now I talk about this impact and introduce that the light is very important related to internet, but in fact, if you look at the light, they're all related to more broader topic. They're related to telecommunications, solar energy, sensing, autonomous car, AR/VR, brain science, big data.

Prof. Min Gu:I will talk about in the next rest of the talk to cover these three topics we are working in RMIT. At the end of these three topic, I will point to the why this is important to brain science and how this artificial intelligence coming into the picture for all this journey we are doing.

Prof. Min Gu:Big data. I don't need to emphasize how important. It's not because engineering science. It is now very important to any aspect in social science, in the humanity, and in any other research area, but if I start asking where the data is, how much the data, and I do want to go to that, as I said. Let's look at the chart.

Prof. Min Gu:This is basically we are saying, every second year, we double it. This looks not difficult, but if you start calculating this every second year doubled, it's a huge number this what we are facing. This is the way the problems. What is more problem is, every country contribute to this, and I was trying contribute roughly 4.5% of the data.

Prof. Min Gu:For any developed economy, roughly similar to order of magnitude of 4% or 5% to the global data, but the problem is that we produce so much data, only 70%, the 70% of data cannot be stored. We don't have a space. We talk about cloud, the way as a cloud. Cloud is physically is a house you have to store somewhere in the device. Only 70% of the data can be currently stored.

Prof. Min Gu:Also, among the stored data, there is another issues. 70% of data stored, you not use it. It's called the cold data. You don't use it. For some reason, maybe two years, three year, maybe 10 years, they are useful. They have to keep it here. That's where the current problem I was talk about what's the problem.

Prof. Min Gu:We have cloud. Cloud is a house inside a house lots and lots and lots of device. Then they take the energy away. How much energy? 3%, 4% of the national and electricity use. It's huge. When you think about the data center, it cost 3% or 4%. That's a huge energy consumption. I'll come back to this.

Prof. Min Gu:There's also environment. These are huge data centers like football stadium. Lots of material, lots of infrastructure you have to committed to that. This is not environment-friendly and there's a lot of cables material you have to also to consume.

Prof. Min Gu:One of the problem is the longevity, because the current device that you go to data center, they will tell you every two to three years or every three to five years, depending on what media they use, you have to change that. You can see, if you change the recording media how much energy you needed to consume, you do this complete again. Think about that. Every three to five years, this is our problem.

Prof. Min Gu:The energy consumption is a problem. How do we compare this energy consumption? This is the data as talk about say, if you believe now every second year data is double that. Now this is the oil, petroleum. If you turn all the annual yield of oil into the electricity, then you will see that this is the electricity needed to store all the data.

Prof. Min Gu:There will be very soon, in 10 years' time, there will be a problem that you don't have enough petroleum if you turn all the petroleum into electricity. This is really the problem. That's all the big data center they are facing, that whilst we are promote digital economy, big data, there is the pressing issues, how do we make this sustainable?

Prof. Min Gu:This, as I said, if you take analogy 2011 or if you store all string data, that means that consuming the whole national electricity consumption. That's the type of the issues we are facing. How do we solve the problem?

Prof. Min Gu:I go back to the optics now. Let me go back to the equation I want to show you. First of all, if you go to Jena, you see the memoriam in the downtown city park. See how important to remember this gentleman's called Ernst Abbe. If you look that Abbe, what they put under monument.

Prof. Min Gu:This is interesting to see that equation. That is the discover. Very simple. What that means, it means that if you go send a light through the lens, the lens, we all use the lens. It's piece of lens. What that lens for, to see small detail. How small the detail you can see? That is the formula tells you what is the smallest feature you can see through the lens.

Prof. Min Gu:Now I just rewrite this formula slightly different. This is the modern way to write. This is engraved on the memoriam. This is important. Now if he didn't pass away before the first Nobel Prize award, and he would be the first person win the Nobel Prize. You see how important, why this is important, because this is lay the foundation the microscope. The microscope are now we are still using after hundred years, still this is enabling because of that simple formula.

Prof. Min Gu:Another German scientist, Maria Goeppert. In her PhD, she discovered that so-called two-photon citation. Now forget about it. Just remember the name. What that means? If you combine with the microscope with this kind of discovery, and you will see that these are difference.

Prof. Min Gu:If you take Maria's discovery seeing through the microscope, you will see a tiny dot, but if you don't, then you see a blurred thing. Now this is important. How important? Very simple. I'll you example. Opera House. We build Opera House seven years, using seven years.

Prof. Min Gu:This, we build up a mini opera house, 30 minutes. That is the foundation of nanotechnology. Nanofabrication now. 3D fabrication can go to the nanoscale. It is because of that discovery. She was awarded the Nobel Prize in 1963. How small this, the Australian kangaroo? We actually fabricated this. The background of this is the human hair.

Prof. Min Gu:Another thing so I want to introduce with optics. Let's now go to the optic disk to touch the topic. The optic disk is invented after almost 100 years of Ernst Abbe by this American entrepreneur. What he says, "Okay, that's good, so let's think about this formula."

Prof. Min Gu:What that means, it gives you smallest the feature you can see. What happen if I use this smallest feature? Burn a mark on the disk. That is the CD burner. That's the concept from very simple turn the imaging, see through to the burning. We burn the mark. That's where using this formula.

Prof. Min Gu:That means we can see, we can burn a very, very small dot on the disk. That's where the first concept, the first-generation CD, based on this concept. The industry, you can see, if you remember 20 years, 30 years ago, CD was really the place lots of things in every aspect.

Prof. Min Gu:Then we went to this Blu-ray. Now what is the Blu-ray? Basically make the, see whether we can work around that formula make it a little bit smaller. You can actually think about that make this more clever not using one layer, you can do this multiple layer.

Prof. Min Gu:The capacity, smallest feature. Remember, this is smallest feature you can put it in, and that means roughly 300 gigabyte. That will be our maximum. That was the reason DVD now not being popular, because the good things become better things under the current situation.

Prof. Min Gu:2005, at that time, we proposed a concept. Thanks to ARC after accept that concept, we got the ARC Discovery Grant, and we call this five-dimension data storage. Whether you can see that we actually break that ceiling, so the bottleneck issues, 300 gigabytes, we now can do terabytes, 10 times more than information you can put it in.

Prof. Min Gu:Now this marked beginning of the nanophotonics. Very important, the nanophotonics. What we do in this case is using this tiny nanoparticles, now it's very popular, this kind of nanoparticle is the metal particle. We use gold because gold is just available at that time. You can use other material.

Prof. Min Gu:They are not sphere. They elongate. Why they elongate? Because if I have a sphere turn around in the space, you will not see you turned. You don't identify the difference, but if I have lost, if I turn around, you will see you turn around. That means the equivalent of the color.

Prof. Min Gu:It's exactly the same thing as want to see multiple difference, then I can put the information into that. We use this different orientation, you can detect and we can actually achieve the optic data storage. This is over the years quickly snatched up, along that journey, first concept published in 2009. In fact, the work started 2005.

Prof. Min Gu:Over the years, we proved that this kind of a storage is secure and could be ultra-secure because you have amount of information. Then, we also, 2017, we can reduce. Remember, energy consumption is the issues. We reduce the consumption per bit, each bit, and to a very small value.

Prof. Min Gu:Now if you couldn't imagine how small this value is, this is equivalent to the brain consume energy. We want to eventually go to that region, because our body is actually the best efficient can use the energy. We do not consume too much energy in the brain.

Prof. Min Gu:Another things we achieve the last year is we can now store the information up to 650 years. 650 years. Think about that. Someone will argue, why I need? I will show you why we need these things, and for the long data storage, but I also list here a couple of things. This is for this Ready for Life and Work.

Prof. Min Gu:First things, we submit the patent in 2013. Two years later, we submit another patent. This year, we also submit the patent on this one. I'll talk about this while we needed these three patents. Go back to this formula. Now remember that this is the Ernst Abbe engraved on the monument.

Prof. Min Gu:This is this gentleman and he was awarded Nobel Prize 2015 for Chemistry. Not for Physics, for Chemistry. Now if you look at that, what is it he discover? He discover this formula. How close these two formula? The only additional things here is little. What that means? Why this got a Nobel Prize?

Prof. Min Gu:Now remember, these are still the smallest of feature that becomes better things. Better is in the biology, you cannot see inside the cell because you have to go into the cell, have a look at what cancer or cancer. You also want to see nanoparticles, small, very small. You can't do it using optical microscope.

Prof. Min Gu:What this gentleman says is, well, how about I do these things? When you see small things, using eraser to erase this fluorescence along the age of that small spot, very simple thinking. He invented this idea when he was a post-doc.

Prof. Min Gu:Now if you look at this picture carefully, had you recognized that's Australian? Phillip Island. He was in my lab when I was at Victoria University in 1997 and he was a post-doc. He created this idea. Now did everybody think he's crazy. How can you do that, achieve this kind of an action? He did.

Prof. Min Gu:Now of course if I simply say, because that formula, he got a Nobel Prize. No impact to Karin. There we have the impact. Formula. That's impact. For 15 years, everybody criticized. He actually worked with industry. Leica, the big microscope company developed this instrument. This will cost a half-million. If you buy a microscope for this one, it will be half a million.

Prof. Min Gu:Everywhere in biology that you go to, Melbourne University, any biologic lab, they will buy this microscope. It's called a STED microscope. It's how important the concept and turning into the product, and also how long the journey, 15 years, that he actually get this.

Prof. Min Gu:I'll come back to this formula now. He invented this. Also, because of that formula, smallest feature you can do, this becomes now the bottleneck issue of the Blu-ray. We think in same way. Along the 2010, we will say, "Well, why don't we do this thing?" We erase it. You record, we erased along the age, and we do it continual. Then we also turn that formula into slightly complicated formula, but we achieve that using this geometry, the method that we call "spin." He goes there and we call spin.

Prof. Min Gu:What that means? Of course in the laboratory, you have to do very complicated. These are the system we can achieve now, 33 nanometer. In fact, the smallest the feature is 9 nanometer. Now for those people who working in the nanotechnology, you know smallest the feature, the electro-microscope where you can see, you can give is along the 7 or 10 nanometer. This is optic method, gives you the smallest feature, 9 nanometer. This is actually achieved in a much cheaper format.

Prof. Min Gu:That is actually the times we realize this can change the data storage industry, we call the new era is coming, so which means using this 33 nanometer, we can actually better the capacity up to 340 terabytes, so thousand times high than the bottleneck issues in the Blu-ray, and this much better than the traditional in terms energy consumption, in terms environmental protection.

Prof. Min Gu:This is over the years we work on this journey. Now if you put this into this comparison, hard disk, fresh memory of the disk. The hard disk eventually reach the ceiling because the nanotechnology and because all this industry can do, so there's a ceiling going towards ...

Prof. Min Gu:You can actually do possible in few years' time the nanotechnology, nanofabrication becomes better, but still you can see there is a ceiling we can achieve for this hard disk industry. Then, optics, because of that formula, stops somewhere here. Now, with this technology, spin technology, we can assess to the nanoscale and we are actually going on this another disruptive change.

Prof. Min Gu:Long data storage. Remember that the 650 years, that is very important, the milestone we achieve at RMIT. Why we needed that? There are a number of area really need a long data storage. Brain research. Brain research, you record the conscious or the thinking, you need a long data storage. You need a high-capacity data storage and long time.

Prof. Min Gu:The other one is the telescope we are developing, Australians also part of the project, square kilometer array telescope. Large amount of data that is only meaningful if the 100 years long this kind of data. Then, there's a gravitation wave and an even longer time scale for this meaningful result.

Prof. Min Gu:In fact, recently, as I said, there's a more related to the other area, geology, biology, astronomy, history, they all need long data. You need it to data meaningful over the few generations. This is actually the reality how we want to do from their own words.

Prof. Min Gu:[Inaudible 00:28:17]. Large amount of data consume energy. We need to consume much less, so our ambitions through the ARC project, the Discovery one, the first one I got when I came to RMIT, we want to go into femtojoule. If you look at 2017, we achieve picojoule. We want another three order magnitude the reduction of the consumption. We were using nanotechnology. I'll come back to this one.

Prof. Min Gu:Next one, longer. Longer and higher capacity. That's the demand from industry. We are actually working on combination, the spin technology and the nanotechnology. Then faster. If you have a vast amount of information, we want fast. How do we do that? We want to have, again, the high capacity, and we need artificial intelligence. That's the way.

Prof. Min Gu:Let me give some flavor what we are now achieve along this direction. The first one, student. One of the student, Simon, I hope he's in the audience. He came from Italy. We just launched the patent. What we use is using very low-cost material. It's called the [inaudible 00:29:29] nanoparticle. These are the crystal.

Prof. Min Gu:We collaborate with Singapore National University and we achieve what he did and say, well, using this nano-crystal, we can now moving into the region where the consumption of energy can be reduced another three order magnitude, and also, nanotechnology. This is the patent we just submitted. He just show some success result there, and hopefully another six months he will conclude this piece of the work.

Prof. Min Gu:Then, artificial intelligence. We will do it, but I don't believe ... I won't do it. I do not have time to do these things, but we have to work on this complicated artificial neural network. This is the concept of how do you make things faster.

Prof. Min Gu:This is the calculation method. How do you calculate this multiple, the focus port, and through this artificial neural network things? That's too complicated. I can't do it. What I can do is we facilitate a workshop with the Computer Science people, computer scientists in the School of Science.

Prof. Min Gu:This is one of the workshop we facilitate. You see Sheldon Lee, [inaudible 00:30:41], they all are very enthusiastic. Very important we are now have a paper on my desk to submit through this workshop, and it will be submitted to a very important journal through this collaboration.

Prof. Min Gu:We are actually trying to, through this conceptual development to the device, which is on the 13 floor, the translation lab, and turning into we call a box. Of course, industry needed to be onboard, and hopefully this can be go into this one.

Prof. Min Gu:Now, that's what we can see what's the future happening. With this Blu-ray technology, the people are producing now the optical media and data center for the reason that low energy consumption, but the size for this amount of information, 100 petabyte space capacity, we need roughly 300-meter-square space, but if we do our own, then we only need a one meter to achieve equivalent things.

Prof. Min Gu:This is the huge impact, of course, and the most important with this optic technology. Then you can see that the energy consumption can be much, much lower. I think we did the calculation few years ago, if we do the solar cell power of the building that is electricity we can generate, a solar panel can be used to power all this optic disk. This will be now beyond the force field. You can use energy outside earth to do the big data.

Prof. Min Gu:I won't be able to deliver this. The industry collaboration, as we can see, very important. The first things, as I said, around 2013, when we have the technology, this spin technology, this company is a spinoff company from Facebook staff member. They actually called me when I was in Singapore. They said, "Well, have you thought about this spin technology for application?" I said, "Well, it's fun. We published a paper." "Nice."

Prof. Min Gu:They actually came to [inaudible 00:32:52] to the patent for optic data storage. They just send the [inaudible 00:32:58] to the patent. Of course they wanted the patent licensed to OAI. A year later, the technology was bought by Sony. Sony acquired OAI.

Prof. Min Gu:About that time, we launched the second patent which is the long data storage idea, and this eventually, there's another company from China, they licensed before I left, actually, [inaudible 00:33:21] this license to another company in China.

Prof. Min Gu:Very interesting, when we publish the result, the 650 years, early this year, Microsoft actually approached us from Microsoft Cambridge and they want to know what we do with 650 years. Now, this is very important. I mentioned it. Sometimes we thought that this is the pathway. We do, but in fact, when I went there, everything I telling you, this possible or not possible.

Prof. Min Gu:Maybe this is not the industry want to go to the pathway. Why is that? When we go to the cloud, so the data center, they manage multiple data center around the world, and they not consider individual disk performance. If you improve individual disk performance, that not necessarily the best of pathway for the cloud storage.

Prof. Min Gu:In fact, they are working on this, a similar competing technology from University of Southampton in UK. They work on this concept, optic cloud. Because ND, I can't go to the detail, but this is actually the future and we've been invited to present a pitch to this concept, how do we eventually all the technology we develop can be used on cloud. This is the way optic data storage.

Prof. Min Gu:I want to quickly change the gear to another technology which is very important, again, invented by Dennis Gabor. Dennis Gabor invented this so-called holography. For that, he actually got a Nobel Prize in 1971. He was born in Hungary, eventually later living in UK, I think Cambridge or Oxford.

Prof. Min Gu:Now, this is his invention. Now I don't want to go into this detail, what that holography means, but I think that you have the experience that if you go to some of the museum, and they have nice this hologram, a 3D picture. It can be through 3D, but what is ultimately this technology can be impact is from this movie.

Prof. Min Gu:Have you see this movie? You saw this, 3D movie, the first 3D movie, Avatar. That's the movie. This actually show that the future, the 3D display, it should be like that. Important I want to emphasize, this person watch this 3D side view, not in the way we watch normally. This is important. The signal you will receive, your side. You're not face it. This, how to achieve, holography can do that.

Prof. Min Gu:Now, at this stage, this is the device. It's called a spatial light modulate, SLM. This is actually a piece of device currently used for generate this kind of a hologram. If you start to do a little bit analysis why this doesn't work, and if you look that SLM versus if you look at this viewing angle, I emphasize this viewing angle is important, if you want to do 3D, you have to be able to see, this is the screen and you can see somewhere here. The viewing angle becomes very important.

Prof. Min Gu:SLM can only give a few degrees. A few degrees. That's why we want to now design all these things we have to face it. You cannot achieve really 3D. In order to break this ceiling, then you needed to think about what's the problem? The problem is that we have to now moving from this region to the region more than 90 degrees.

Prof. Min Gu:You can see that the size predicted that it's very challenging, but this is actually we achieve a few years ago. We can use this laser printing in graphing. This is a graphing material. It's very popular, but we use the laser to produce this hologram, simply speaking.

Prof. Min Gu:What this advantage in this technology is we actually put multiple color. If you can put three color, it means that all these true color imaging you can achieve that. We achieved 52 degree. This is 52 degree, 52 degree in this technology. Now, going beyond that, I'll tell you why this is difficult. Let's first have a look at one of the movie which, at that time when we published result, it was list on the Time Magazine in the front page.

Prof. Min Gu:This is actually the movie. You can see, this test object three-dimension, you can see from minus-26 degree to positive-26 degree, around this, so 52 degree, you can see this. This is the first time. I guarantee that this is the real 3D, the imaging. You can see the wide angle.

Prof. Min Gu:The one step, the one step to reach that level. That is because the technology. Again, go back to why we cannot do that, because Ernst Abbe said, "The smallest the feature, there's a limit." You can't do more than that. You can't go more than this.

Prof. Min Gu:We will continue on. Last year, and this is another paper we published. The one step, we do step by step. We actually can produce a hologram and in very, very thin sheets. There is housing 1,000 hair size. It's very, very thin. That means you can extend this into the more broad application. One is this vision. This is really a number of industry now want us to demonstrate that with this kind of watching, and then you can actually achieve is a 3D display.

Prof. Min Gu:To do that, the first things, we needed to increase the viewing angle. The viewing angle, as I said, with the spin now, we should be able to achieve 90 degree. Now I do have the result, but I didn't have time to include. In fact, last week, one of the post-doctor achieved that this smallest feature would really go beyond other limit. This is doable. We want to also do flexible, and I'll show you the flexibility in a minute.

Prof. Min Gu:The next thing, so must be dynamic. If we don't do dynamic things, you can't see the movie, all these thing. For that, we launched the patent this year to make sure that the idea, before we can talk, the idea is being patented. Very soon, we will be able to now solve this problem. As I said, we want to go to the smallest feature and go to our mobile unit.

Prof. Min Gu:Just as one demonstration, you can see, this is the flexible material and we can actually produce a hologram into this material. Of course, at this stage it's still not high-quality, but it demonstrated that this 3D holographic imaging can be build up into this flexible structure, and this is the pathway is there.

Prof. Min Gu:Last topic. Super capacity. Karin. This is Karin's favorite topic. What I will only emphasize what the contribution, what we have contributed to this field. Now, what is a super capacity? There's a academic definition of "super" and there's also very easy to understand the "super." "Super," it means that quick charge, and you can have charge this capacity very quickly. You don't need to wait for hours to get the charging. That's super.

Prof. Min Gu:You can actually reuse, charge and recharge and recharge much more times than the current battery, and again, this is super. There is the possibility you can actually make this into flexible, and that means that all the stretchable device, flexible device we dreamed, this can be powered by this kind of a device. They are environmentally friendly, so it's much less damage to the community, to the environment.

Prof. Min Gu:There are the scientific definitions, "super," and this is probably from more layman point of view. I go back to the scientific comparison, do this easy one first. This is a very simple comparison. Notice, kind of also have similar slides on that.

Prof. Min Gu:Super capacity versus lithium-ion battery. This is the one we use in mobile phone similarly, so all this in mobile phone. We experience the problem, the safety issue, there's a chemical problem issues, all kind of issue, but this let's compares very important.

Prof. Min Gu:As I said, the charge cycle much better than the traditional one, and there's also the fast, fast charging. This is the very fast charging and lifetime longer, but there is one problem. The problem is energy density. Why is this so good? Why not we use it? Because we cannot put too much energy into this one. There is a problem. This is where you can see this currently the very low energy density if you want to do have a lift, you have to use many, many of these things in order to achieve the practical application.

Prof. Min Gu:Go to young people. We have [Litty 00:42:54]. Litty is in the audience? Not? This is another PhD student. She just finished the PhD last year. Now, this is her idea, independently proposed. It's not my idea. I just endorsed the idea. This is a good idea.

Prof. Min Gu:We were working these issues, how do we put the more electricity into this super capacity? Now what she come back the idea, say, "Why don't we learn from this tree?" Tree, the particular tree is called a fern tree. If you look at the tree, it's American fern tree, and under the microscope, then you see this very nice pattern.

Prof. Min Gu:This side under the ... close to the macro-nanoscale. They're the interesting things what she discovered. This pattern mathematically, let's go to the math, mathematically, it's equivalent to this fractal structure. Hubbard is very famous mathematician. He actually predict a lot of things. They're the pattern Hubbard fractal. They are equivalent, mathematically exact, the same.

Prof. Min Gu:Why we need this equivalence proof? Because now, this pattern, I can ask the laser to follow the trajectory. The laser can follow this pattern to produce electron which can store the energy. Now more than that, this pattern, actually the fractal is a cell for reproduce.

Prof. Min Gu:When you go to smaller, produce this same pattern, when you go to smaller codes in the pattern. We can actually really control at a nanoscale the size. This can be also done on the substrate. This actually fabricate a structure, and this on the substrate, flexible substrate.

Prof. Min Gu:She done very more since integrated this device on the solar cell panel. This on the top, there's a solar panel, and under that, there's super capacity. It's actually one girl you can actually make using this technology to finish this using layer-by-layer fabrication to achieve that.

Prof. Min Gu:That's Litty's PhD. She also actually done that increase the size moving to the slightly larger than the research size. This is another product that she produced in her PhD. At that time, the energy density is higher, it's higher than the published result, but it's not high than what we want to go. Litty's result received mainly inquiry from the industry even from Defense. For that, we'll launch the patent in March this year. Litty was awarded this RMIT Impact Award for HDR Student.

Prof. Min Gu:From there onwards, now on this journey, so first of all, using spin, we want to increase to this number. This will be equivalent to lithium-ion battery. Then we want to size the increase with this ceiling funding, and then we are actually also integrated with the solar cell for the panel for the design harbor. This is Litty. Just yesterday, she told me she can actually achieve this fractal pattern on the fabricates. This is the conductive fabrics and this can be done.

Prof. Min Gu:This is the movie, I quickly show that, so that we can generate electricity from solar panel and then charge the super capacity. This is charging and this is super capacity. These, the fan, you will see that after minutes charging, so after charging ... Come on.

Prof. Min Gu:After the charging to the one volts and we turn off the solar cell, so disconnected this part and store the energy here, and then turn on the loading, the fan is there, so the electricity. We will scale up using this device to achieve this work with ...

Prof. Min Gu:Now, I want to wrap up my talk and just touch this topic. We all talk about artificial intelligence. Now, maybe the next few slides, my view will be very provocative. It's my personal view. This is where the direction to go. We all talk about artificial intelligence. I call this is the Phase 1. Based on this very simple older technology, still older, computer separated memory and calculation, but still like that, we still work on this architecture.

Prof. Min Gu:The only thing is this person, Turing, is very famous. He said we can do, using computer, do some testing. The human task make this so-called artificial intelligence. We can beat the game in the chess and this is the goal, our goal.

Prof. Min Gu:Everything we develop, this is all based on one's principle. It will not actually revolution. What is the revolution? It will be these things. The computer were based on the neural network. The chip will do the neural doing the job.

Prof. Min Gu:What that means, the neural will eventually collect a signal, do the calculation, pass on the signal, all and go within one cell. We have a millions, millions of the cell, and then it will make a decision. This is called the neuromorphic computing.

Prof. Min Gu:Now if the neuromorphic computing can be developed, this is really artificial intelligent, the computer. Now, by putting this biological model into the engineering model, and what that means is that we need as a device, you can calculate the waiting, the waiting from [inaudible 00:49:17]. We have to have a waiting like this one. Then we do this decision and then pass on, and on and on again. They produce this synapsis type of function and do this calculation. These are the two major functions, and it must be combined together.

Prof. Min Gu:This is over the few years, last few years, a very hot topic around the world. Again you can see this from Alan Turing and the same guys propose this artificial intelligence, the concept, the Turing test, he also proposed this concept so-called this neuromorphic computing. Then, this is the first person to use traditional computer to achieve this concept. It's huge, but you will not go into it because energy consumption issues, it's a huge energy consumption.

Prof. Min Gu:This is where the integrator circuit come into the picture, and that's where the magnet is smaller. Recently, now you can see photonics, RMIT, Princeton last year simultaneous published a paper going to neuromorphic photonic chip. Why that already emphasize photon carry much less energy, consume much less energy.

Prof. Min Gu:If you look at their result, what they are missing? They do this based on this calculation, and that's were achieve. The blue color means this based on the basically equivalent to [inaudible 00:50:43] communication type of a device, but they're missing this part. They do not have a synapsis.

Prof. Min Gu:While they do the synapsis calculation and the probability calculation, they still use traditional computer. It's not really neuromorphic things. What we recently proposed that the RMIT, the concept that we haven't going to the patent. If any in the audience, don't worry, I didn't disclose too much here.

Prof. Min Gu:We have a concept and I can see with joule here, by one of the early [inaudible 00:51:16] research, and through this device, we should be able to now produce this function. If we produce this function, connect it with this, and then, this will be really complete, the photonics or photonics neuromorphic computing with speed of light.

Prof. Min Gu:Now, I'm not trying to say, well, this is the only my view, it's happening in RMIT, in Princeton, in Stanford. In fact, the recent paper published University of California, they actually also moving onto this direction. It's publishing science, actually. It will happen, or optic computing with the speed of light with this kind of concept.

Prof. Min Gu:Just to finish up, this neuromorphic computing, if you use in traditional way, the first one, the energy consumption per signal pathway is somewhere here. Then using this electronic one, you'll reduce two to three order magnitude.

Prof. Min Gu:The real neuron is sitting here. There's six order magnitude that we have to find for, and then the photonics can come into this picture. Again, spin becomes very important in this case. The size, the energy consumption, that is the way we are targeting in this region.

Prof. Min Gu:Just want to summarize that I actually trying to say mathematics, science, they are very important, and this is the way we do lots of fundamental work, but in the meantime, we very well aware that the impacted is in each of the area. I didn't have time today talk about the house area. We do a little bit of work actually in this area, but I want to emphasize last, the one that all the work now point us into this, the real neuromorphical artificial intelligence, one I want to ... the group is eventually to move in this area.

Prof. Min Gu:Of course, industry is very important to take this all the result. These are all the industry, after RMIT, there's a number of industry coming to approach us, and specifically for the three technology, the optical data storage, the holographic, and the graphing super capacity.

Prof. Min Gu:I'll stop here. Thank you very much for the attention.

Xinghuo Yu:Thank you very much, Min, for the excellent talk. Now we have time for a few questions. Anybody have any questions, comments? Yes. I'm not sure if this one is working or not, but you can try. Otherwise, just shout. I think it's working.

Speaker 3:Thank you, Min. That was a fantastic talk. Fascinating. One of our question is, I'm particularly interested in the AI structure you mentioned. If you use the photonic technology, is the computer or the neuron will be binary or is it going to be like a decimal or even different type of representation maybe, analog?

Prof. Min Gu:Yes. It could be combination. It could be digital, it could be analog, and could be the combination, either digital or analog. In fact, you probably realize that another branch towards this neuromorphic computing is memories. If you look at the design in memory computing, and there are two ways still, but personally I believe that the digital probably is more advantage in terms of the signal and coding, and from our point of view.

Speaker 3:When you form the structure, can the structure be changed later on?

Prof. Min Gu:Oh, you have to design according to the ... Probably it's difficult. I think it will be related to here. Different approach probably need a different architecture.

Speaker 3:Thank you.

Xinghuo Yu:Thank you very much. Any other comments, questions?

Speaker 4:Thank you for the fantastic talk. Just a quick question. Can you comment on what you pointed out just now about neuromorphic computing? There's another quantum computing. Can you comment on any connections or any opportunities to different ideas or different concepts?

Prof. Min Gu:The quantum computing is actually still based on the current architecture thinking. You just increase the more states. If you do the job zero-one, the current computing, the quantum computing say, "Well, I can do zero-many, many states." It's not from zero-one. You can have a one. Quantum is a probability, so it actually could be unlimited, the states. Then, so that you can do the calculation much faster. That is from that angle.

Prof. Min Gu:Now, one of the problems I see, the energy consumption. Because you're still going on the traditional technology, silicon technology, all the technology, and you have to do the hardware to achieve this kind of things, the energy consumption possibly will be one of the problem.

Prof. Min Gu:Going to neuromorphic computing, because I have limited time, one of the important things, so this will combine into one unit. The calculation memory will be in one unit, and that will be the way to reduce the consumption. Now I also need to emphasize that, ultimately, we will not go through, I believe this eventually scientists who are not going to this way. This is still the concept [inaudible 00:57:17] this separately, but the way that one of the ... In fact, I was involved one of center of excellence UI this year.

Prof. Min Gu:Then finally, it must be related to the [inaudible 00:57:28] device. You have to follow the neural growth and then making sure that that will be the more sustainable approach, but this is a long way, probably another 50 years to go into that direction.

Xinghuo Yu:All right. Thank you very much. There's one or two more, then we ... I know I just remember you are standing between Min and the refreshment.

Speaker 5:I just have one question on this connection to computer science, machine learning, because on machine learning, you have this machine learning more like neural network, and you can do training. That means the connection waves can be adjust. If you can do this in, I don't know, in neuromorphic computing, is it possible to make them adjustable so that you can do them for this sort of learning or training?

Prof. Min Gu:Yes. Thank you. This is the way I found a few months ago when I was in Microsoft. Then I start see happening on the plane. There's 15 hours on the plane. Then you can start a speculation, think about all of these things.

Prof. Min Gu:This diagram, the dot is very important. This actually dynamically, we believe that very similar to the current memory disk, it's a photonic memory disk, so we can training this into the way to remember, and then do this machine learning thing. The learning function will be here.

Prof. Min Gu:I can talk to you which we collaborated, I want to show you in fact a very similar. This one, it means you can actually change status according to the output. We can achieve that loop, which is the learnings you need.

Xinghuo Yu:Thank you very much. Just the one last one. Anybody has ... ?

Speaker 5:Maybe I'll take the blame.

Xinghuo Yu:Oh, you'll take the blame.

Speaker 5:Sorry for holding you for the refreshment. Actually just a follow on Sheldon's question. In the recent years, in one of the bottleneck or one of the direction in AI or machine learning is the learning model is actually try to combine with memories.

Prof. Min Gu:Yes.

Speaker 5:I'm asking whether this structure can actually be a natural solution so that learn model also come with a memory.

Prof. Min Gu:Yes. The answer is yes, but there are two approach that currently in the future computer design. One is called in-memory computing. They follow the way the memory and the calculation combine together. This is the different approach. This come from the neuromorphic point of view, but they all achieve the same things.

Prof. Min Gu:In the end, these dots becomes the memory. On the chip, whatever happened, so this can remember, even you turn off the power, the state will be still remember, and then this becomes the memory which cannot do right now. Like we sleeping, we still have memory. This can be done. That's the way we want to achieve with the machine. This will be the on-chip memory simultaneously. These are the calculation.

Speaker 5:Yeah, but this is probably not exactly what I have in mind. When we talk about machine learning model has memory, they memorize variables before they're doing the decision?

Prof. Min Gu:Yeah. Our [inaudible 01:01:23] simply won't, but if you actually think about this array like the computer chip, and this can be proven.

Speaker 5:Ah.

Prof. Min Gu:I just simplify the ... I don't want to discuss all this. I have a much more complicated we have the calculation, and also each of the function can be train.

Xinghuo Yu:All right. You can continue discussion during the refreshment. Firstly, I would really like to thank Min for this very-

Prof. Min Gu:Endeavor.

Xinghuo Yu:... entertaining talk. I'm very enlightened, actually. Very nice on this Friday. I understand-

Prof. Min Gu:I saw the one mathematical formula.

Xinghuo Yu:I'm not sure, Min, I understood and all, but [inaudible 01:02:05]. Please join me to thank Min again for the excellent-

Prof. Min Gu:Well, thank you very much.

Xinghuo Yu:Thank you.