Crimson Lucre features a number of computers – in SaMMCon, aboard Prospector 1 and on Mars. But are they quantum computers? Probably not. For “simple” problems like navigation and orbital dynamics, classical binary computers are up to the challenge. But for cyber security, materials science, chemistry, pharmaceuticals, climate and weather modeling, the speed and computing power of a quantum computer is essential.
Quantum computers have been in the news lately. IBM, Google, Honeywell, Amazon &others offer quantum computer access via the cloud. Recently, great strides have been made in the number and stability of qubits available on quantum chips.
What’s the difference?
The basic unit of classical computing is the bit. A bit only possesses one of two states, 0 or 1. A computer with two bits can consider four different states(00,01,10,11). A classical computer must work through one computation at a time to solve complex problems, a quantum computer essentially solves its computations simultaneously.
The basic unit of quantum computing is the qubit. Qubits rely on two distinct quantum states of matter: Two of the many discrete spin states of an ion, the nuclear spin state of an atom, the spin states of an electron or the polarization of a photon. However, unlike the 0-1 logic of a classical computer, these quantum states are not strictly defined by up-down, left-right, or high-low, but by probabilities. As qubits are added to a quantum system, the number of possible probabilities becomes astonishing. Sepehr Ebadi, a Harvard graduate student and lead author of a paper announcing the creation of a 256-qubitquantum chip published in July’s edition of Nature noted in ScienceDaily that “"The number of quantum states that are possible with only 256 qubits exceeds the number of atoms in the solar system."
The answer to a problem posed to a quantum computer is not arrived at by a series of answers to binary questions, but rather on the arrangement and quantum states of a collection of qubits. This reliance on quantum processes is at the heart of the speed of quantum computing.
Today’s rudimentary quantum computers have already demonstrated the ability to solve in minutes, problems that take classical supercomputers days or years to solve.
What’s the Use?
A number of applications out of the reach of today’s most powerful supercomputers lend themselves to quantum computing.
Security. The symmetric-key algorithm employed by The Advanced Encryption Standard (AES)is the gold-standard of computer security. To crack an AES encrypted data set, it takes a supercomputer years to crunch through the permutations to arrive at the correct matching key. A quantum computer employing the requisite number of qubits could decrypt the data in minutes, if not seconds.
Most experts believe we are about ten years from entering this new (in)security paradigm. When we do, quantum keys will become the new normal in encryption. That, and a new class of security software that can track malicious malware back to its source and eradicate it—call it Smith.
Materials Science. Quantum computers have a clear advantage over classical computers because they use quantum processes (qubits) to describe and evaluate quantum processes (electromagnetic and orbital behavior at the atomic and subatomic level). Today, most advances in alloys, plastics, etc. is the result of painstaking trial and error. Quantum computing can model exactly what interactions are occurring within crystalline or amorphous solids to achieve a desired material property.
Chemistry. Many companies today utilize cloud-based quantum computing to understand interactions at active sites on molecules such as lithium hydride. For example, Mitsubishi is investigating lithium-oxygen batteries with greater energy density. Daimler has a program to develop the next generation of EV batteries.
Pharmaceuticals. Receptor sites on proteins and enzymes rely on delicate quantum interactions to achieve a given biological activity. Quantum computers have already been used to evaluate the active sites of numerous enzymes for drug research.
Climate and Weather modelling. Modern climate models can take supercomputers days to run. These models divide the atmosphere into a 3D grid starting at the Earth’s surface. Various algorithms calculate the transfer of heat, moisture, etc., between the grid dells. The smaller the cell size, the greater the precision of the model. But the smaller the grid, the greater the number of cells and the longer the run time. Today’s best models rely on a 1-mile square grid, but at the cost of accuracy (or resolution, in the modeler’s jargon). The goal is to achieve a grid size something on the order of 1 square meter – something that would take a binary supercomputer months, or even years to run. But a quantum computer with the requisite number of qubits could run such a model in minutes. Consider this: we may be able to predict which butterfly in which location will initiate the chain of events that cause the hurricane of the century. (Consider that when you admire that butterfly resting on your rose bush.)
Moore’s Law for Quantum Chips?
The original Moore’s law applied to the miniaturization (and thereby increased density and capacity) of transistor yes/no logic gates (i.e., the 1 or 0 state known as bits) on a silicon chip. Through the 80’s, 90’s and 2000’s, transistor capacity consistently doubled every two years. This became known as Moore's Law. Are quantum chips experiencing similar growth with their numbers of qubits?
There are two basic approaches taken by quantum chip developers - groups like IBM or Rigetti—who use superconducting qubits composed of a supercooled electron pair (known as a Cooper pair) trapped in an aluminum lattice and manipulated by magnetic fields—and groups like Honeywell or IonQ—who use ion-trap-based qubits that use light to excite ions, then detect the photon that is later released. The wavelength of the photon corresponds to the quantum state of the ion.
In 2016 IBM made its rudimentary quantum computer available on the cloud. That early quantum processor used 5 qubits. IBM also made available QISKit, a “toolkit” of open-source quantum algorithms. Research teams are constantly adding new quantum algorithms to the QISkit library.
In November 2020. IBM made available its 54-qubit quantum processor.
In September of last year, IBM introduced a 65-qubit processor. The industry buzz is that they will release a 127-qubit processor perhaps sometime this year. Two independent scientific teams announced the achievement of 256-qubit quantum chips just last month.
Recent advances in the number of qubits on quantum chips has indeed been increasing.
So What’s the Problem?
But…the greater the number of qubits on a chip, the greater the probability of decoherence - a quantum state other than what is called for by the problem posed – thereby introducing error into the solution. To date, the way to increase reliability is to add more qubits – essentially duplicate computations that can self-correct for errors. Qubit stability on the order of a second is adequate for quantum computers to function reliably. But coherence is often measured in microseconds
The primary hurdle for quantum chips is the basic instability of quantum elements needed to create and maintain the qubit.
Time for a Change?
This July, researchers at Google and several educational institutions announced the successful creation of a functional 20 qubit time crystal. A time crystal is a newly discovered state of matter that alternates between two quantum states –like a metronome. But unlike a metronome, a time crystal will alternate between those two states forever without the expenditure or input of energy. In essence, it’s a quantum perpetual motion machine. This property means that time crystal qubits modified by a quantum algorithm will remain stable indefinitely. The announcement, reported in this July’s edition of Quanta Magazine, notes that the process is scaleable - meaning its amenable to commercial development and fabrication.
So, will we see widespread use of quantum computers by 2035? The infrastructure is in place for major advancements: quantum chips powerful enough for meaningful computation, technology to produce stable qubits, and a rapidly growing open-source library of quantum algorithms that act as a bridge between binary computer code and quantum hardware.
So yes, quantum computing will be certain and commonplace in 15 years. Hold onto your hats. This will be quite the ride.
For further reading:
https://www.sciencemag.org/news/2021/07/physicists-move-closer-defeating-errors-quantum-computationhttps://www.eurekalert.org/news-releases/779236https://news.mit.edu/2020/scaling-quantum-chip-0708https://www.sciencedaily.com/releases/2021/07/210709104157.htmhttps://www.techrepublic.com/article/6-experts-share-quantum-computing-predictions-for-2021/https://www.quantamagazine.org/first-time-crystal-built-using-googles-quantum-computer-20210730/#:~:text=Like%20a%20perpetual%20motion%20machine,matter%20inside%20a%20quantum%20computer.
