Strange Tech from the Quantum Realm!
have you ever wanted to visit the quantum realm the good news is you're already there there's no such thing as you know like the quantum realm and the classical realm everything is quantum bottom to top it's just that the quantum part is sometimes hard to notice there are laws of physics everyone intuitively knows laws that Newton first described with precise mathematical language among these rules everything has a well-defined position no thing spontaneously jumps from one place to another there is no fundamental limit to how fast something can move or how gently it can be nudged these seem so obvious it sounds silly to list them out but the remarkable thing is that they are only approximations of true reality we call these rules of thumb classical physics but beneath them lies a deeper fundamental theory of nature that we call quantum mechanics and it's far stranger than you could ever imagine take spin for example most people think that when you spin something there is no limit to how slow it can revolve but most people are wrong spin comes in irreducible chunks the slowest a basketball can rotate is one turn every 250 seextilian years the next fastest speed is two turns every 250 seextilian years these spin chunks are too small to notice in a basketball but the effect is more obvious with lighter weight objects a virus can only spin at a rate of once per second two turns per second three etc but no in between spins can exist looking at yet smaller sizes it's more obvious that an electron's qualities like spin charge or magnetic flux are irreducibly chunked or quantized electrons don't actually spin at all yet still have an intrinsic angular momentum called spin up or spin down and the weirdness doesn't stop there you can learn quantum mechanics which is kind of the the the base code of the universe and you can learn how that's done in a book and you can even do some some basic experiments but you don't really appreciate it until you try to make something new out of it today scientists aren't limited to watching nature do amazing and weird things they can engineer systems using these properties they can make something be deloized neither exactly in one place nor the other nor exactly in two places at once what physicists call a superp position they can entangle two things such that neither is independently real and only their whole combination has an independent reality they can even tunnel or quantum leap something through a barrier and have it appear on the other side as if by magic superp position entanglement quantum leaps a computer designed using the strange quantum nature of reality could have unprecedented power and everyone wants one quantum computers are rapidly coming into their own we can see a path towards them replacing very specific computing hardware for problems like simulating large quantum systems right now they're science experiments with a big dream the word computer is a little misleading it makes us think that the computers are just like the classical systems that we have but they're nowhere near that developed yet so we are still trying to build the very basic building blocks in classical computers information is at its most basic level a bit either a zero or one state that's like your light switch being on or off or more likely in a computer it's a magnetic domain that's oriented up or down when you're talking about a quantum bit you now not only have something that's analogous to zero and one up and down but you also have the ability to be in superp positions of up and down quantum bits or Qbits can not only be in two states at the same time they can share or entangle their states across many cubits increasing their computing power you want a a quantum superp position that is shared across many cubits probably hundreds or thousands until we have a a quantum computer that beats what we can do with classical computers in classical computers operations are performed on bits using tiny electrical switches called transistors more and more of which have been squeezed onto a chip as transistors shrink quantum effects become more noticeable and can even cause classical designs to fail but if you're ready for those effects you can use them to your advantage and even make a single atom the basis of a quantum bit this however is more easily said than done it's hard to get an atom to talk with another atom um people are doing it that is one form of quantum computing but uh a lot of the quantum computing that goes on in this lab is trying to use electronic circuits to emulate the same behavior as atoms the electronic properties of atoms can't be customized circuits on the other hand can be made any way scientists like we've become able to make pieces of metal that also have these discrete sets of energy states at least when we make them cold enough the key ingredient to accessing a circuit's quantum properties is temperature if you freeze some metals colder than the vacuum of space they become superconducting the name literally refers to how you can conduct electricity through a metal with no loss and the first time I heard about it I thought that was the most important property of superconductors but I would call it a macroscopic quantum state the idea is that when you get cold enough all these electrons start moving in lock step all doing exactly the same thing right down to the quantum mechanical level so even though it's it's a huge microscopic number of electrons it's behaving like one electron going around an atom something you can almost get your hands on and touch and so an artificial atom when we're talking about quantum computing is usually a solid state system in this lab usually we meet a transmon cubit which is a capacitor with an inductor these type of circuits are known as LC oscillators because the energy oscillates back and forth between two components like a bell where the capacitor stores energy in the electric field and the inductor stores energy in the magnetic field and the energy sloshes back and forth it's like the energy goes to the inductor slashes to the capacitor back and forth but this is no ordinary LC oscillator its cold superconducting metal suppresses the background noise to such a degree that the circuit's quantum nature becomes accessible and at -459° F the discrete energy levels become clear like the electron energies in an atom but with properties we can design name and manipulate if you take all the energy out you can so there's no sloshing uh we just call that zero cuz we're unimaginative and if you add the smallest amount of energy you can to that we call that one operations are performed on these cubits not by arranging transistors like in a classical computer but by nudging the cubits with microwave pulses it's kind of like trying to make a wine glass ring it's not going to ring at any frequency that you can make similarly our cubits are oscillators that have a defined resonant frequency and it's only when we're hitting that frequency that they respond say we wanted to push uh a system from the zero state to the one state we would put in a little tone of just that frequency to get that up to the first one so far so good now we're in the one state if we want to go from the one to the two then we put in just the frequency that would go from one to two but that's the same as the frequency that goes from one back to zero you you could go either way it might even be a superp position and then you would just kind of lose track of where you are with evenly spaced or linear energy levels a cubit would be impossible to control so scientists use an unconventional nonlinear inductor with energy level spacings that are not alike so what we need is we need each of those energy splittings to have unique frequency that's what the nonlinearity does for us um so the 0 to1 there's a 0 to1 frequency there's a 1:2 frequency a 2 to three frequency and so on you need to be able to individually address those energy levels the way you individually address them is by having unique frequencies for every transition and the way you have different frequencies every transition is to have a little nonlinear element this nonlinear part of a cubit is incredibly small in fact it's invisible to the naked eye so those two parallel dark lines make a planer capacitor a capacitor is a device that will allow you to store electric energy and then there's a gap in them where it looks like they're not connected but in reality they are they're connected by a very very thin trace of metal this bridge is an inductor but it's a very unique one to understand why you need higher magnification you need the power of a scanning electron microscope [Music] zooming into the cubit reveals a strange stack of overlapping metal this is called a Josephson junction the Josephson junctions are the magic that make our cubits work it is a layer of superconducting metal it's just like a little layer cake uh a layer of oxide that does not conduct electricity at all and then another if I had a third hand a superconducting layer on top of that the way we make Josephson junctions is a bit complicated uh we have to take an aluminum layer very well orderered aluminum layer then we oxidize it uh oxidizing is kind of a messy process uh it's essentially allowing it to rust so we take an aluminum layer we allow it to rust and then we put another aluminum layer on top and that seems pretty uninteresting um because you you can't have any current go through a uh an insulating gap by definition um but if you make it thin enough you get another one of these quantum mechanical effects where there's no such thing as uh the electron concentration just stopping hard if they overlap enough then you can get what's called tunneling where they will just quantum mechanically you know quantum leap whatever you will that's a term uh jump from one side to the other this quantum tunneling is what gives the cubit its nonlinear energy levels uh that in essence means that when you change their amplitude their frequency changes the equation for them is actually identical to the equation of pendulum i have a pendulum right here and we can perform an experiment if I go and I start oscillating this pendulum very slightly you'll notice that it's going back and forth pretty quickly uh if I go and increase the rate of oscillation the amplitude of oscillation it takes longer to do a full loop so the fact that I can change the amplitude and that changes the frequency is what makes our quantum computer controllable meaning that if I want to switch one quantum bit from zero to one I have a microwave pulse that corresponds to that if I want to put two quantum bits into a super position I have another set of multiple wave pulses that do that action it is that set of pulses that do something that I know that we call a logical gate so why are quantum computers still experimental well um on paper quantum computing is solved the theory which is a mathematical description and the hardware which is a piece of metal in this case don't line up to every single level and so anything where the theory doesn't match the physical system is something we call noise we have noise from everything around us uh if you look at room temperature we have uh infrared cameras that allow you to see in the dark that for us is noise infrared would cause a whole bunch of noise on our cubits that's why we cool them down to very low temperatures to try and isolate them from the room temperature noise noise not only introduces errors into quantum computers it can unintentionally change a cubit's delicate quantum state a process called [Music] decoherence so to separate them from their environment scientists run their experiments in giant refrigerators that are more complicated than you'd imagine so if we're going to send a signal into our fridge here it all starts on the room temperature racks on my right uh they come up overhead and come in through the top of the fridge through this very top plate which is always at room temperature on the outside is the regular lab space on the inside when all the shields are on will be under vacuum um vacuum is important because if there's any air in the fridge that would transmit heat from room temperature to the colder stage inside we pump out all the air and then each stage gets a little bit colder uh all the way down to the bottom so there's uh 37 lines coming in the input here our signal will follow one of them little coaxial lines and they all have a wiggle to them um that's for two reasons one is the the fridge will expand and contract a little bit as it changes temperature and we don't want to put any strain on these so that just can take a little flex and it doesn't damage the cables if they change in length slightly the other is we're paranoid about any stray light getting in some infrared light or something like that light won't bend around corners and send any photons we don't want into the fridge this stage is at about 30 Kelvin that is not a temperature that comes up in everyday life uh the nitrogen in the air that you're breathing will turn to liquid at 77 Kelvin oxygen a little below above that air is ice the next stage is 3 Kelvin at this stage even helium the last gas of anything uh becomes liquid helium never becomes solid fun fact at this point gases don't exist at this temperature at 3 Kelvin now um you're a solid or liquid there's no other choices we go to even skinnier cables as we get down uh skinnier just less metal we'll conduct less heat we we want to make sure that we don't conduct any heat from stage to stage um these are a metal that's a very low thermal conductivity a copper nickel alloy I think uh and then we get colder to about um 1 Kelvin this is the still uh this stage we actually add heat to for all the all the hard work we do to try to get cold this this is the stage that we actually heat uh to get it to just the right temperature next stage 100 ml stage uh each one gets a curly queue down to the mixing chamber which is at about 10 michel um which is uh you know 30,000 times colder than room temperature on an absolute scale and everything at this plate and down is at 10 mic keelvin so we have a lot of room to mount our little devices and we typically we have I don't know maybe 20 independent experiments on this at a time you can see all manner of different filters and things here just to throw out any frequencies you don't want let's just follow this one for example it comes in you see it goes through this little collar and inside this collar would be inside a shield that is a magnetic shield it's different than a Faraday cage that's something where you can just block electric fields it's actually not possible to block magnetic fields you can only redirect them and so it's actually a kind of metal that kind of sucks all the magnetic field lines into it so that there's a lower magnetic field inside and outside the can in the walls of the can the magnetic field will actually be enhanced we just kind of redirect all those lines into the walls of the can and then the space inside will be a very uh low magnetic field so we might take some of this and mount it just so put some screws on there onto the paddle and then I can open up inside i'll just take the lid off and this is a typical device um it has six coaxial connectors um you have a little chip with some quantum devices in the center these uh six ports might for example be in this case the top and the bottom ones are the input and output for the RF so our output lines uh once we have uh some some signal from our from our cubits they'll typically go through a quantum limited amplifier and then go through these isolators these are double junction single junction isolators basically one-way valves cryogenic one-way valves so that your signal uh to the hotter stages can get out but any uh thermal noise from the upper stages can't get down to the lower stage and then if we follow that signal chain out up to the uh 3 Kelvin stage and this is where we have the first conventional amplifiers we're usually trying to measure kind of single photon events at room temperature even in the microwave range there's a lot of thermal radiation around so we got to boost that signal up to a level to where something uh conventional room temperature electronics back at the racks like this can detect it and distinguish it from just a thermal background amplification is the name of the game not just amplification but very low noise amplification every kind of amplifier adds its own noise that's both just an engineering truth and we're at the level where we get the physics truth you can actually quantify how little noise nature allows an amplifier to add and we get as close to that limit as we can and then it goes all the way to room temperature and we have more amplifiers sitting on just conventional nice conventional amplifiers at the top until we finally make it back to our um equipment rack okay this is a VNA one of our workhorse instruments among many around here vna stands for vector network analyzer uh vector in the sense that uh if you've ever used a vector in math you know it's something that has both an amplitude and a direction um in our case we think of our microwave signals as vectors they have an amplitude and as it's going around in its cycle repeating again that's like 360° in a circle and we want to know exactly what the direction the the phase of the signal is that that's what makes it a vector analyzer what makes it a network analyzer is it has more than one port so it can this one has four we're just going to be using two in a little demonstration here um we're hooked up to ports labeled three and four right now um I can grab this guy this is where we're going this is one of our 3D cavities that we will mount on the fridge and we're just doing a little room temperature test on it um we're hoping that it has a frequency of around I think 6.7 gigahertz and we're just going to measure that before we cool down and make sure that's what we want we know where to measure when it gets cold uh this is something that does have a little um cubit inside but we won't be able to see anything quum about this at room temperature we're just looking at what the cavity does so putting that aside if my cables and connections are good that trace didn't change when I moved it and what this yellow trace is is just uh uh magnitude so this is really just kind of the length of the vector on the vertical scale and the horizontal scale is frequency you see this blue light here means we're sending signal out through port three we're measuring what comes back on port 4 so if I have this on a marker I just manually put that on there i'm reading out about yeah 6.65 GHz not bad it's about where I wanted it i can turn on another trace here and uh it's a very very sharp transition in angle so typically what we'll do if we're reading these out is we'll be taking a trace um very much like this but instead of scanning through many frequencies which takes a lot of time we want to read out very fast so we'll typically do is send in a single tone maybe just a little off center uh of what the center frequency of the cavity is to see the state of the cubit like a zero or one state of the cubit shifts the the frequency of the cavity just a little bit say this is when the cubit's in the zero state it's in the one state it might shift over to be where the one is so in that way we can send in a relatively large microwave signal into the cavity look at the phase of its return and depending on how the phase is shifted we know the state of the cubit uh without talking to it directly that's called dispersive readout you're kind of talking to the cavity around the cubit and the cubit has a very subtle effect on the cavity that we can read out with uh with high fidelity and that's how we usually talk to our cubits um in this case we're not going to see any dispersive readout because our cubit is uh at room temperature it's not even superconducting much less in a coherent quantum state one thing that the national laboratories do especially well is our ability to tackle big problems that have a long time scale developing the next generation of computing hardware is one of these problems and that sort of commitment and a devotion to solving the big problems that are needed to bring the technology to the next level is the reason that the national laboratories like Lawrence Livermore National Laboratory are an ideal space for us to be doing this work there are constantly new technologies that uh are coming around and that we're exploring having expertise in lasers allows us to quickly turn over and look at any applications that might come out with lasers having expertise in superconducting materials allows us to go and explore superconducting systems very quickly if we never iterated we wouldn't have this fridge open behind us right we would just run we got to open up and say "Oh uh okay we got a little data that gives me an idea uh I've got this other device from the from the fab room that I want to put in now let's warm up and swap it out with another one i've got five new things I want to try you don't know what you're going to learn till you try it that gives you some new ideas to try something else out so you got to have the amazing tools people who are good at using the tools very creative scientists who can imagine new things that we don't know to be true yet check that they're true and then when you discover something new well you've done some science quantum technology isn't science fiction an abstract concept or a microscopic world separated from our reality the truth is we've always lived in the quantum realm and we're just awakening to what that might mean for more information on the laboratory's quantum experiments visit science and technology review
2025-04-04 17:45
