Quantum Computing on Cusp

EE Times - Januray 7, 2017

LAKE WALES, Fla. — Today only a single company — D-Wave Systems — produces a commercial quantum computer, and even D-Wave admits its latest “2X” is no substitute for a supercomputer (except for a small set of optimization tasks). Within five years, however, all that may be changed.

Quantum computing uses qubits (quantum bits) on the atomic scale; it is predicted to be faster and able to store more data than transistor-based computing. Some researchers are predicting that the market for “universal” quantum computers that do everything a supercomputer can do plus everything a supercomputer can not do — in a chip that fits in the palm of your hand — will be burgeoning. The rise of quantum computing may be as important a shift as John von Neumann’s stored program-and-data concept.

Here are some of the scientists and breakthroughs that will enable this shift.

Yale, Quantum Circuit Inc.
Spawned from two decades of research in the Yale University labs of Professors Michel Devoret and Robert Schoelkopf is a new company, Quantum Circuits Inc. (New Haven, Conn.). Along with his collaborators, Schoelkopf claims a number of “world’s firsts,” the latest of which is the longest “coherence time” for a quantum superposition.

 Yale)<br />
Multilayer microwave integrated quantum circuit (left) uses silicon wafers with features etched using MEMS techniques to create enclosures that serve as high-Q resonators as well as providing shielding. Superconducting metalization (blue) covers the walls of these enclosures to provide low-loss wafer-to-wafer bonding. A cross-section of the rectangular cavity resonator (upper right) shows interlayer aperture coupling between the cavity and transmission lines above. 3D superconducting transmission lines (lower right) could be constructed using membranes (green) in the MEMS structure where qubits and act as a compact low-loss quantum bus. 
(Source: Yale)

Professor Robert Schoelkopf, chief architect and co-founder of Quantum Circuit Inc., explains the importance of quantum computing breakthroughs to the World Economic Forum. (Source: Yale)

Superposition is the heart of the quantum computing paradigm shift. It enables the simultaneous representation of both a “0” and a “1” in the same quantum bit (qubit). The reason that is important, is that qubits in such superposition states can calculate the whole realm of possibilities, say in the next chess move, then pick from among them the optimal one in a single cycle. This “optimization” solution, in fact, is the one quantum operation that D-Wave’s quantum annealing device can perform.

The ultimate goal of D-Wave, Quantum Circuits, Rigetti Quantum Computing, IBM, Google Labs, Microsoft Labs, the U.S. National Institute of Standards and Technology (NIST), the European Union, Imec (Belgium), Rikken Research (Japan), many other worldwide computer makers and hundreds of independent research groups is this: a “universal quantum computer” that can perform tasks that supercomputers perform — but in the blink-of-an-eye instead of hours or days — along with all the so-called NP-hard (non-deterministic polynomial-time hard) problems that no supercomputer can solve no matter how much time you give it today. (Plus as a bonus universal quantum computers can create impossible to crack encryption algorithms).

Quantum Circuits is perhaps furthest along in this quest measured by the number of “world’s first’s” it has accumulated plus one more thing — a real-life product. Despite the fact that thousands of research labs worldwide are researching the same problems in quantum computing, there are surprisingly few products — only D-Wave’s “optimizer” and Quantum Circuits’ “amplifier.”

What does a quantum amplifier do? Well one of the peculiarities of quantum qubits is their fragility. Almost anything — from heat to a cell phone’s microwaves — will cause their state of superposition to collapse (called decoherence) spoiling their ability to calculate. Even reading their value destroys them (which is why quantum encryption codes are uncrackable). To maintain “coherence” in the superposition state of a qubit, quantum devices are supercooled to almost absolute zero — to keep their atoms from vibrating — and shielded from every kind of electromagnetic waves in existence. Nevertheless, even so after decades of work the state of coherence has only been able to be boosted from nanoseconds to milliseconds, when ultimately its coherence needs to be measured in tenths of seconds (at least).

Even measuring the state of a qubit is one of those things that can disturb them and that is what a quantum amplifier minimizes. Quantum Circuits’ first product — the second in the world after D-Wave’s optimizer — is a quantum amplifier (unless you count single-photon emitters but that is a story for another time).

“Our team has been developing quantum amplifiers for several years because they didn’t exist, but were necessary for our research, and will be an essential component in future quantum computers. Today they are so useful and reliable, that they are becoming something we and others around the world would like mass-produced,” Schoelkopf told EE Times in an exclusive interview in October 2016.

Scanning electron microscope (SEM) image of using a MEMS etch to suspend a transmon (top) with a detailed view of the supporting pedestals (lower right) and a detailed view of the suspended Al-AlOx-Al Josephson junction (lower right). 
(Source: Yale)

The ability of Quantum Circuits amplifier led to its latest “world’s first” — an error-corrected quantum memory that can sense when a qubit has gone awry into decoherence and reinstate it to its former state of superposition. The amplifier and the error correction unit are both essential elements being sought by labs worldwide to eventually build the holy-grail — the aforementioned “universal” quantum computer. And Quantum Circuits plans to make available those functioning units needed by other labs as products.

From the accompanying images of Quantum Circuits “suspended” Josephson Junction (JJ) superconducting transistors — configured in pairs called a transmon — you can see that one of the keys to their success is the ability to harness microelectromechanical systems (MEMS) techniques to make their fragile qubits more bullet-proof. The techniques they use to etch their suspended JJs is licensed from Bosch — its well-known dry reactive ion etching (DRIE) process that here evacuates the silicon underneath the JJ. This is one ingredient in the development of a new technology for integrated quantum circuits that the team is pursuing.

“You can think of our qubit circuits as artificial atoms that display, at the macroscopic level, the same quantum effects that all matter exhibits at the atomic level,” Schoelkopf told EE Times. “We have now demonstrated the prototypes of all the subsystems we need to build a universal quantum computer. Now it is just a matter of time. Within the next few years we will scale up to the quantum equivalent of the first digital electronic computers. But even with only 50 or 100 qubits, these early systems will already unlock the true power of quantum information, and can have a computation power for certain tasks that exceeds even the largest conventional supercomputers.”

Scientists at the University of New South Wales (Australia) Centre for Quantum Computation & Communication Technology (CQC2T) explain how they fabricated quantum bit with a coherence time of 2.4 milliseconds. (Source: UNSW)

D-Wave was the first quantum computer company, because they decided to go with solving just the part of the problem that could be solved today, rather than wait for the “universal quantum computer” paradigm to be invented.

D-Wave’s quantum computer only solves optimization problems, epitomized by the traveling salesman problem (TSP) which asks the question, “what is the optimal (shortest) route for a salesman to take when visiting 10 or more cities?” The TSP is unsolvable even by the fastest supercomputers, unless you trivialize it to just a few cities, or use various “estimating” methods, because there are too many combinations to compare them all.

D-Wave’s quantum “optimizer” computer, however, uses a superconducting core to solve these problems. It works by generating the lengths of all the possible paths in a single step by using a superposition of them all, after which it picks the shortest route. NASA, Lockheed and Google have purchased D-Wave’s machine to solve optimization problems in their various domains.

Rigetti Quantum Computers
Officially named Rigetti & Co, Inc. (Berkeley, Calif.) Rigetti himself worked under professor Michel Devoret at Yale for his Ph.D., shortly thereafter founding Rigetti Quantum Computers. A generation or two behind Quantum Circuits, Rigetti’s group nevertheless is enlisting the help of the University of California at Berkeley’s graduates to build a universal quantum computer from the ground up, unlike D-Wave which built a special purpose quantum computer from available technologies.

University of New South Wales (UNSW, Australia) Centre for Quantum Computation & Communication Technology (CQC2T) scanning electron microscope (SEM) view of its new quantum bit with an unusually long coherence time of 2.4 milliseconds.<br />
University of New South Wales (UNSW, Australia) Centre for Quantum Computation & Communication Technology (CQC2T) scanning electron microscope (SEM) view of its new quantum bit with an unusually long coherence time of 2.4 milliseconds.

Chad Rigetti speaking at Startup School SV 2016 on Sep 29, 2016.
(Source: Regetti)

“The D-Wave system is a special-purpose tool that can solve a particular class of optimization problems. We’re building a universal quantum computer, which hosts a coherent quantum mechanical state with a gate set capable of performing a wide range of algorithms. Research groups around the world have demonstrated a variety of small quantum computers, typically comprised of just a handful of qubits. However, because the computing power of these algorithms scales exponentially with the number of qubits, most approaches to date have only provided a glimpse of what the technology could do,” Rigetti told EE Times.

The Rigetti approach is to build a quantum computer from the ground up to be scalable and coherent, according to its founder. The first prototypes will be important not just because of their sheer power, but because they will be built in such a way that rapidly increasing their scale will be straightforward, requiring Rigetti to solve many of the hardest engineering problems first in order to develop an architecture with long-term product evolution possibilities.

Founded by Rigetti in 2013, after a three-year stint at IBM researching quantum computers after graduating from Yale, the first few years were devoted to designing a product architecture and building what Rigetti calls “world-class” capabilities in building and measuring qubits. The company recently completed the characterization of a five-qubit chip, including the software stack that sits on top of this architecture that is currently in private beta with several customers.

“In the next 18-to-24 months, we are aiming to build systems large enough to perform computations out of the reach of the world’s most powerful classical supercomputers, at which point the economic and technical value of quantum computers will become extraordinary,” Rigetti told EE Times.

According to Rigetti, the advantage of quantum computers is that they store and process information on individual photons, whereas today’s digital computers uses one hundred billion photons for each individual bit processed.

 UNSW)<br />
Apparatus at University of New South Wales (UNSW, Australia) Centre for Quantum Computation & Communication Technology (CQC2T) to support its qubits-long-coherence time of 2.4 milliseconds. (Source: UNSW)

“We are building the most powerful computers in the world to solve humanities most important and pressing problems,” said Rigetti.

Researchers worldwide are aiming at the same goal, one step at a time. For instance, just last week Australian engineers at the University of New South Wales (UNSW) claimed to have produced a new quantum technology that uses a strong magnetic field to extend quantum superposition coherence to 2.4 milliseconds.

Quantum computing has become the “space race” of the 21st century, and every semiconductor maker, researcher and inventor knows it and wants their discoveries to be recognized as the “firsts.”

Quantum Circuits flux tunable suspended qubits in a side view scanning electron microscope (SEM) image realizing a suspended SQUID transmon.
Quantum Circuits flux tunable suspended qubits in a side view scanning electron microscope (SEM) image realizing a suspended SQUID transmon.

For more information read Suspending superconducting qubits by silicon micromachining just one of Schoelkopf’s excellent papers.

D-Wave is funded by Bezos Expeditions, Draper Fisher Jurvetson, BDC, GrowthWorks Venture Capital, IIU, Kensington Smart Alternatives, Harris+Harris Group, In-Q-Tel and Goldman Sachs.

Rigetti is funded by Y Combinator, SUSA Ventures, AME Cloud Ventures, Morado Venture Partners, Feicis Ventures, Fenox Venture Capital, Streamlined Ventures, Berggruen Holdings, The Alchemist Accelerator, Data Collective and AceQuia Capital.

Quantum Circuits was founded in late 2015 and is seeking funding under a Form D filing with the Securities and Exchange Commission (SEC).”

R. Colin Johnson -  EE Times