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 no bigger than a single atom. This is interesting because there are hundreds of quantum computers being built all over the world and each uses different methods for building qubits.
Qubit is a quantum analog of a classical bit that leverages the principles of quantum mechanics the most popular and mature Quantum technology right now is superconducting Cubit, and this has been pursued by companies 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 performance, currently the largest quantum computers run on a few thousand cubits which is still very far from making it practical.
Scalability
However, scalability is limited by many factors and one of them is the size of cubits. So what’s wrong with it? you know our traditional silicon chips like GPUs CPUs AI chips 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 a 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 have impurities in them.
In this work the researchers from MIT took a diamond and imbued it with teen atoms they bombarded the diamond with tin ions as a result we have a diamond with some implanted tin ions and many vacancies afterward the diamond was heated and so-called teen vacancy centers were formed and these act as a single entity that has Quantum properties. We can control these properties with electromagnetic waves.
For example, a microwave at just the right frequency can flip a vacancy in the center from zero to up or down. So in this way, we get qubits that we can entangle and we can compute with them. These structures are just the 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 such teen vacancy cubits. Over the last couple of years, 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 the way classical computers do because they use quantum bits for both storing and computing the information. This quantum system in the chip features qubits and interconnects, so we connect many such chips to scale it up. In the paper, they mentioned that it’s possible to connect a thousand such chips to come to a 1 million Cubit Milestone.
What is so important here is 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. These diamond-colored centers are solid-state systems. They are 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 these 1024 cubits fit into a 500 Micron by 500 Micron area and what is fascinating is that this qubid density is 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 notes 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 is 4 cm to 4 cm in size only for 49 measly qubits.
It’s worth mentioning that the above was one of the older Intel Quantum chips and now Intel is working on Silicon Quantum dots technology which is an extremely promising technology.
There have been some very exciting advances. We’ve discussed two critical challenges remain: cooling and error correction 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.
N = kTB
Here’s a table based on the equation N=kTBN = kTBN=kTB, where:
- N is the dependent variable
- k is a constant
- T is temperature
- B is another variable, possibly bandwidth (if this is related to thermal noise)
Let me know if you have specific values you’d like to include. Here’s a generic table with some example values:
| k (Constant) | T(Temperature) | B(Bandwidth) | N (Result) |
|---|---|---|---|
| 1.38×10^−23 | 300 K | 1 MHz | 4.14×10^−15 |
| 1.38×10^−23 | 400 K | 2 MHz | 1.10×10^−14 |
| 1.38×10^−23 | 500 K | 5 MHz | 3.45×10^−14 |
This assumes k is Boltzmann’s constant (1.38×10^−23 J/K)
You can very well see it from this equation 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 temperature 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%. A 10% error rate is 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. A 10% error rate is disastrous. In general modern quantum computers have error rates from 1% to .1% and we will talk about it later 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 logical cubits 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 the rapid spread of information that is hard to verify.
Quantum Teleportation
Silicon Quantum Dots and Their Importance
In general, what we want to do in the Quantum ecosystem is 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, there has 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, and diamond vacancies that we’ve just discussed, 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 layman’s terms the idea is to take a transistor and inside isolate a single electron in the channel and then use its pin as a state of a cubit so this technology is 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 area of silicon.
The Role of T-Centers in Quantum Communication
So this technology is scalable for mass production and exactly this technology is behind this experiment so what’s neat about the photonic platform is they have what’s called T-Center. At the core of their device, they have this T-Center and it has a mixture if you will of types of cubits right it has not only what we call a spin qubit but it also can emit a photon. Because it can emit a photon, a photon can travel, and its light can travel in fiber optics and so what photon has been able to do is demonstrate this in their platform.
The Quantum Teleportation Experiment
For the first time they have two separated cryostats there are about 40m of fiber optics between them. Imagine a 40m of fiber connecting these two cryostats and inside these cryostats are these T-centers. A Quantum device in each one and so those Quantum devices you can shine light you know shine a light to excite and release a photon so each release a photon and once they are detected at the same time essentially then you perform in each cryostat a series of operations on each T- Center and this enables the entanglement across these two cryostats. We call this distributed entanglement or the ability to do remote entanglement. And you do it right without having to interact with what’s inside those two cryostats.
They interact via the photons the light that travels across the fiber. Just like classical computers quantum computers perform operations on logic gates and these logic gates convert input into a certain output. One type of Quantum logic gate is a 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.
Significance and Future of the Quantum Internet
The other bit now in our case these are qubits but still the same idea 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 a controlled knot operation between the cryostats. 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. Recently, Photonic and Microsoft successfully implemented it. This is a significant milestone in quantum entanglement as we have these networked machines and longer distance connection ultimately will look like a Quantum Internet that will you know essentially run alongside your classical internet. it doesn’t replace your classical internet but gives additional capabilities on top of your classical internet.
New Quantum Devices
New Quantum Devices which is extremely promising for building practical quantum computers. It’s a relatively new flavor 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 cubit is particularly interesting because unlike other types 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 face of matter with our topological qubits. These are based on a very new type of idea is actually to create a new phase. It’s called a topological phase.
Understanding the Topological Phase of Matter
When you think of phases of matter liquid, gas, and solid. We engineer a new phase of matter in the device it’s called a topological phase of matter. This is a very new property it’s essentially a nanowire it’s 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 simply, we have a nanowire and on both sides of it we have Quantum dots which practically work like a gate in a classical transistor.
How Topological Qubits Work?
It controls the flow of electrons through this wire and when we close this gate some of the electrons are trapped in the wire and the number of these trapped electrons defines the state of cubit. 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 of this wire.
Enhanced Noise Resistance with Majorana Particles
At the same time, topological cubits promise to be 100 to a 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.
Error Rates and Performance Improvements
I’ve simplified it a lot of course but this is the basic idea behind By having this natural protection at the hardware level, we can start at a better air rate right so that physical Cubit promises to have 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. We call that a 10 – 5 or 10 – 6 air rate.
Scalability and Future Potential
So that’s several you know orders of magnitude better than many of the other cubits in existence. Today this new type of Cubit essentially promises great scalability. Because it has the right speed, the right size the right controllability and its fidelity is much better than other types of cubits out there today. And so 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.
Leave a Reply