Huge Breakthrough in Quantum Computing. New Phase of Matter
hi friends in this video I will discuss new scalable Quantum technology huge milestone in Quantum Teleportation and then some brand new Quantum devices that use new phase of matter very exciting first of all researchers from MIT made a breakthrough in Quantum Computing they've developed a Quantum system on a chip that features a special type of qubits that are no bigger than a single atom and this is really interesting because you know there are hundreds of quantum computers being built all over the world and each use different methods for building qubits Qubit is a quantum analog of a classical bit that leverage the principles of quantum mechanics the most popular and mature Quantum technology right now is super conducting Cubit and this has been pursued by company like Google D-Wave and many other startups one of the biggest problems with this technology is scalability so the ability to scale to a larger number of cubits in a system without compromising the performance currently the largest quantum computers run on a few thousand cubits which is still very far from making it practical however scalability is limited by many many factors and one of them is the size of cubits so what's wrong with it you know our traditional silicon chips that I cover in depth on my channel like gpus CPUs AI chips they are made up of billions of transistors fitting on a very tiny piece of silicon the fact is that most of these chips are much smaller than individual superconducting cubits so how can we build the quantum system with millions of interconnected qubits that we can control this is exactly the problem the researchers from MIT are solving with tin vacancy cubits let me explain how it works let's start with the fact that we all love diamonds and what is so beautiful about them is of course their quantum properties a diamond so a pure diamond is made of a repeating crystal of carbon atoms but those diamonds that shine the most they have impurities in them and in this work the researchers from MIT took a diamond and imbued it with teen atoms they were basically bombarding the diamond with tin ions as a result we have a diamond with some implanted tin ions and many vacancies afterwards the diamond was heated and so-called teen vacancy centers were formed and these act as a single entity that has Quantum properties and we can actually control these properties with electromagnetic waves for example so let's say a microwave at just the right frequency can flip a in vacancy in center from zero to up or down so in this way we get qubits that we can entangle and we can compute with them you see these structures are just a size of a single atom so they have much better scaling perspectives than any other type of cubits now let me know what you think in the comments so in this work the team from MIT created the quantum system on a chip that features 1024 of such teen vacancy cubits you know over the last couple of years I talked a lot about this system on a chip concept like Apple's M chips with the latest M4 in 3 nm system on a chip refers to integrating all the components like memory processor and iOS into a single piece of silicon and now we have a Quantum system on a chip of course it doesn't have a memory in a way classical computers do because they use quantum bits for both storing and computing the information so this quantum system in the chip features qubits and interconnects so we connect many of such chips together together to scale it up and in the paper they mentioned that it's possible to connect a thousand of such chips to come to a 1 million Cubit Milestone what is so important here that we want more than just 1 million cubits we want 1 million qubits with good fidelity so that they are accurate and reliable and that's what's so hard to achieve we will talk more about that later in the video another aspect I find super interesting about this work that because these diamond colored centers are solid state systems they are actually compatible with our seamless fabrication process and this is the process that we've mastered so very well over the past couple of decades now it's getting very interesting because this 1024 cubits fit into a 500 Micron by 500 Micron area and what is fascinating is that this qubid density is actually close to the transistor density of the most one of the most advanced CMOS process nodes they write in the paper that it's comparable to TSMC's and three note wow it doesn't get any better than that to give you a feeling of what this is like here is an example of an older Quantum chip with 49 cubits that is 4 cm to 4 cm of size only for 49 measly qubits how does this sound let me know your thoughts in the comments it's worth mentioning that what you saw it was one of the older Intel Quantum chips and Now intel is working on Silicon Quantum dots technology which is extremely promising technology and there have been some very exciting advances so we're going to talk about it in the second part of the video apart of the scaling challenges we've discussed two critical challenges remain it's cooling and error correction the cooling is required because qubits are quite fragile so any noise or vibrations from the environment can destroy the information contained in the cubits that's why we always try to protect them and that's why when you're in the room with a working quantum computer you have to stay quiet joking actually by noise here we don't mean any loud sound we mean random bits of energy which can be in the form of microwaves or heat most of the time the main problem is thermal noise and you can actually very well see it from this equation you can see that it's proportional to the temperature and from this you can understand why we need to try to keep it cool so actually at temperatures closest to absolute zero at which Quantum systems typically operate the thermal noise is very low and this is also the case for vacancy qubits but these can operate at a slightly higher temperatures at 4 Kelvin which is about 1,000 times better but still far from a dream of making a quantum computer working at room temperature now researchers from MIT are focusing on further scaling the system of course but especially on the error correction algorithms because at the moment they're experiencing error rates of about 10%. 10% error rate is basically a probability of undesired change in the state of the qubit so 10% error rate means that every 10 out of 100 operations result in an error so 10% error rate is actually disastrous in general modern quantum computers have error rates from 1% to .1% and we will talk about it later in the video because there have been some very exciting progress but just to give you a feeling achieving Quantum Supremacy requires us to achieve an error rate of one failure per trillion quantum operations so we have far away moreover you typically need more physical cubits to build let's say 100 of logical cubits those that you then can access with algorithms and software you see that's why scaling is so essential now before we discuss a huge milestone in quantum teleportation and the special qubits that leverage a new phase of matter I think one of the big problems we face nowadays is a rapid spread of information that is hard to verify internet algorithms make it even worse most of the time we don't know where information is coming from and if it's reliable Ground News the sponsor of this video is solving this problem they gather articles from all around the world in one place so you can compare how different media outlets cover them they also provide context about the news sources political bias credibility and ownership so readers can see how these factors influence an article reporting and all of this backed by ratings from three Independent News monitoring organizations you can easily stay up to date on issues that matter the most to you that's why I use ground news to stay informed about what's happening in the world including technology and science I think they're a really powerful tool for making sure you see the bigger picture so check them out for yourself go to ground.news/anastasi or scan my QR code here for 40% off the same
Vantage plan I use for for unlimited access to all their features ground news is independent and supported by their subscribers so by signing up you directly supporting the development of more transparent and objective media landscape you know in general there are different things we want to do in Quantum ecosystem first of all we want to build cubits and then we want to build a quantum computer out of it and then we want to network these computers think about distributed computing and there have been huge news recently so Photonic in collaboration with Microsoft was able to transfer Quantum information between two distant cubits but before we talk about this success we have to talk about silicon Quantum dots as you know quantum computers use quits to store and process information and there are multiple types of cubits like super conducting qubits, diamond vacancies that we've just discussed in the first part of the video, and then there is silicon Quantum dots technology I'm keeping an eye on the last one because it's very promising and that's the technology that's based on Silicon now if I explain it in a layman terms the idea is to take a transistor and inside isolate a single electron in the channel and then using its pin as a state of a cubit so this technology is basically leveraging our standard CMOS technology to build cubits and this is something we know very well until now have managed to scale it down to 1.6 nm and pack hundreds of millions of transistors in a tiny tiny area of silicon so this technology is scalable for mass production and exact this technology is behind this experiment so what's really neat about the photonic platform is they have what's called T-Center so at the core of their device they have this T-Center and it has a mixture if you will of types of cubits right so it has not only a what we call a spin qubit but it also can emit a photon and because it can emit a photon right a photon can travel right it's light it can travel in a in in fiber optics and so what photon been able to do is demonstrate this in their in their platform right for the first time so they have two separated cryostats there's about 40m of fiber optics uh between them so imagine right a a fiber uh 40m of fiber connecting these two cryostats and inside these cryostats are these t- centers right a Quantum device in each one and so those Quantum devices um you can shine light you know shine a light to excite and and release a photon so each release a photon um and once they are detected at uh at the same time essentially then you perform in each cryostat a series of operations on each T Center and this actually enables the entanglement across uh these two cryostats so we call this um distributed entanglement or the ability to do remote entanglement um and you do it right without having to interact what's inside those two cryostats right they interact via the photons the light that travels across the fiber you know just like classical computers quantum computers perform operations on logic gates and these logic gates convert input into a certain output and one type of a Quantum logic gate is so-called controlled nod gate or C not gate this is a controlled nod operation so what it means is based on the value of one of the bits you want to flip the other bit now in our case these are qubits but still the same same idea right if this bit is zero then don't flip this C this qubit and if this qubit is one we're going to flip this qubit and so basically that's what's done here is is a controlled knot operation between the cryostats um but done by only locally operating on the cubits and so we call this a teleported controlled knot operation in order to scale quantum computers to larger systems we need to achieve entanglement not only between cubits in one chip but between cubits located in a two separate chips and recently Photonic and Microsoft successfully implemented it this is a significant milestone in quantum entanglement so as we have these networked uh machines right and longer distance connection ultimately will look like a Quantum Internet that will you know essentially run alongside your classical internet right it doesn't replace your classical internet but gives additional capabilities on top of your classical internet now there is another technology which is extremely promising for building practical quantum computers I don't know if you heard about it it's a relatively new flavour of qubits so-called topological cubits and it's very interesting because there have been some great breakthroughs recently first of all this type of cubits is particularly interesting because unlike other type of qubits we've just discussed that are based usually on particles such as ions electrons or photons these cubits are based on a topological state or um face of matter with our topological qubits um these are based on a very you know you you would say new type of physics right so the idea is actually to create um a new phase it's called a topological phase right when you think of phases of matter you have you know liquid gas solid we actually engineer a new phase of matter in the device it's called a topological phase of matter so this is a very new property it's it's essentially a nanowire um it's it's a um a superconducting nanowire essentially what you're doing is controlling these nanowires and driving them into this topological phase and then what emerges is the ability to use this as a cubit to put it simple we have a nano wire and on both size of it we have Quantum dots which practically works like a gate in a classical transistor it's controlling the flow of electrons through this wire and when we close this gate some of the electrons are trapped in the wire and actually the number of this trapped electrons defines the state of cubit and the quantum information in this case is stored on both ends let's say it's stored on both ends of this wire and these ends are about three microns apart and this is exactly what makes it resilient because it's very unlikely that this noise particle will hit at both ends uh of this wire at the same time so topological cubits promise to be 100 to thousand times better in terms of noise and this is huge for Quantum Computing the thing here is that the quantum information in this case is stored in the properties of the entire system rather than in the properties of individual particles or atoms so it's inherently more stable now these electrons are very sensitive to any noise from the environment or any radiation or waves or energy hits from the outside that's why they added special Majorana particles to the system these particles have some very unique properties that protect these electrons from the noise I've simplified it a lot of course but this is the basic idea behind so by by having this natural protection at the hardware level we can start at a a better air rate right so that physical Cubit promises to have say one in 10,000 only one fault in 10,000 operations at the physical level or even one fall in a million which is even better right so we call that 10 - 5 or 10 - 6 air rate so that's several you know orders of magnitude better than many of the other uh cubits uh in existence today this new type of Cubit essentially uh promises uh really great scalability right because it has the right speed the right size the right controlability and its fidelity is much better than other types of cubits out there today uh and so we believe this is you know a very promising approach to scaling up now of course they need to work on scaling it to a larger number of cubits and building logic gates out of it and eventually performing millions of quantum operations per second I'm pretty sure that Quantum Computing will bring us a lot of exciting surprises already in this century and this will be definitely fun to follow I hope you will stay with the channel so please consider subscribing not to miss the future episodes about the advances in Quantum Computing and leave me a comment below what you think about it as you may have noticed I took a very long break I was almost one and a half months away from YouTube and I have to say I missed you guys so badly very much uh but I was very busy working on a new project um I've made a huge change in my career and original plan was to share it in this video but I want to wait a bit until things are completely up and fully functional for now let's stay in touch let's connect on LinkedIn I personally love to talk to you on LinkedIn guys because I love to see who is behind the screen you know to whom I'm talking to so now I want to make a small giveaway of the book which I just read beautiful book one of the best reads this year uh it's called I may be wrong so I want to give it away among those who leave a comment under the video and you just mentioned that you want to win a book because I know not Everyone likes to read so uh and then I will send it to you thank you for watching and I will see you in the next video very soon ciao
2024-07-01 21:30
