Quantum Computing Hardware and Platforms | Superconducting Qubits, Trapped Ions, Majorana

A Quantum Engineer's Guide to Superconducting Qubits: https://arxiv.org/abs/1904.06560

Oxford website on trapped ions:
https://www2.physics.ox.ac.uk/research/ion-trap-quantum-computing-group/intro-to-ion-trap-qc

Steane paper on trapped ions: https://arxiv.org/pdf/quant-ph/9608011.pdf

Join this channel to get access to perks like behind the scenes and support my channel!
https://www.youtube.com/channel/UCzaYH6WeohiHKj3Ih_GdZdQ/join
or the same perks over on Patreon:
https://www.patreon.com/amarchenkova

0:00 Quantum hardware
1:23 Superconducing Qubits
3:45 Trapped ions
5:38 Photonic Qubits
6:22 Majorana Fermions
7:07 Quantum Annealers vs Universal Gate Quantum Computers
8:44 Quantum Volume and Algorithmic Qubits
11:29 Quantum Platforms
13:38 What's next?

To understand what makes some quantum computers more powerful than others, the best place to start is the qubits. Even though a qubit just means a two level system, the physical implementation of the actual physical qubit can be very different depending on the quantum system you’re looking at -- and there are pros and cons of each!

One hardware implementation for qubits is a superconducting qubit, which is a design based on a Cooper pair (which is a pair of electrons, or another fermion, fermion means particles with half-integer spins) combined with a Josephson junction.

Trapped ions are another hardware implementation of quantum systems. These systems usually consist of Ytterbium atoms.

First, as part of the trapping process, an electron has to be removed from the neutral ytterbium atom to create the ion. So these ions are now positively charged.

That charge on the ions allows us to trap them in an electrical field to form a line of qubits which can be controlled and measured by shining lasers at the individual qubits. So, these ions are now levitating in this vacuum environment, above a special chip, so they are “laser-cooled” since they are nearly stationary.

Once the ions are trapped, an array of individual laser beams, each targeted onto an ion, plus one “global” beam, is used to control the quantum system. The interference between the two beams controls the state the qubit is in. The quantum state is stored in two electronic states of each ion.

Another approach is to use photons, light particles, as photonic qubits.

These computers use an optical circuit, based on squeezed light or “single-photon” light pulses, light beam splitters, and photon counters, and work at room temperature.

These quantum states of light are controlled by laser pulses.

The best part of the quantum computing revolution is that quantum systems are available in the cloud, for anyone to use, without having to spend millions of dollars. So don’t take my word for it, or anyone’s word for it - log into a quantum cloud platform, and write some code for a real quantum computer!

Microsoft has spent the past years pursuing another approach to quantum computing based on Majorana fermions, a theoretical set of particles that are their own antiparticles. In 2018, Microsoft researchers claimed strong experimental evidence that they’d created the particle and would be able to build a quantum system out of these Majorana particles.

The idea was that two Majorana particles, that behave like a half an electron, are formed on the ends of a semiconductor wire wrapped in a superconductor. To make a qubit using these particles, you would swap the positions of the two half-electrons on the wire, like braiding hair.

Last month, however, Microsoft researchers published an official retraction to their discovery, citing “insufficient scientific rigor”.

Now, how do we compare these hardware implementations? What makes a collection of qubits useful?

We have the DiVincenzo criteria for the physical implementation of a quantum computer.
There are 5 factors:
- A scalable physical system with well characterized qubits
- Be able to put the qubits in the initial states accurately
- Long relevant decoherence times, so long times before the quantum information is lost, compared to gate operation times
- A “universal” set of quantum gates
- We have to be able to measure individual qubits

Pretty much every scientist will tell you that there are multiple good candidates for quantum computing technology. Since right now we’re in the early days of quantum computing, error mitigation is key to building useful algorithms. Once the native errors on existing systems are beaten down, the goal is then to build increasingly error corrected quantum computers that can solve some of the most important problems in the world that classical computers just can’t solve, from materials science, machine learning, and optimization.

#quantumchips #quantumcomputers #quantumhardware

Видео Quantum Computing Hardware and Platforms | Superconducting Qubits, Trapped Ions, Majorana канала Anastasia Marchenkova