Quantum Computing: Part 2

Plenty of Room at the Bottom

It was late afternoon and I found myself sat alone in Edward’s office staring at a painting of an arthritic olive tree on his bookshelf. My mind summoning memories of Julys spent in Italy. Edward is a connoisseur of great literature: poetry, biographies, mythology, science fiction, and of course non-fiction. A particular book stood out on the floor-to-ceiling shelving that displayed his vast reserve. It was titled ‘The Feynman Lectures on Physics’. A textbook chronicling famous talks by the illustrious physicist. Feynman – one of the great scientific minds of the twentieth century – worked on the atomic bomb at Los Alamos during World War II. Shared the Nobel Prize for physics with Julian Schwinger and Shinichiro Tomonaga in 1965 for work on quantum electrodynamics. Was a leader in the field of subatomic particles – studying the building blocks of protons and neutrons. And even had a hand in solving the Challenger space shuttle disaster.

Seeing this book reminded me of an article I had read exploring one of Feynman’s talks. In this 1959 address titled ‘There’s Plenty of Room at the Bottom’, Feynman spoke of ‘man’ soon being able to manipulate matter down to the level of individual atoms. He continued to progress his ideas and spoke fervently during the twilight of his career about using Quantum Mechanics to perform calculations. Feynman’s talks and articles are the foundation of what we now call Quantum Computing. They were the first crow flying across the sky. ??

Science Advances by Accepting Absurdities

Quantum Computing has the qualities of a bestselling science fiction tale – reminiscent of work written by Isaac Asimov or Martha Wells – rather than credible science. But this is because it is a new technology that many of us (me included) have not yet come to terms with.

Science advances by accepting absurdities. History is replete with examples: the Earth revolves around the sun; time is curved; a black hole could fit in your pocket (a primordial black hole, that is); the varieties of defensive white blood cells that our bodies hold – some ten million – each designed to identify and destroy a particular classification of invader. These are all now widely accepted scientific phenomena, given the evidence in their favour. And you are in a small minority if you dispute them. Quantum Computing is just the next advancement that appears inane at first but that will yield fruit before long. ?

Part 1 – if you have not yet read the first of this two-part series, I advise that you do before continuing – acts as an introduction to this new technological frontier. Focusing on qubits, amplitudes, superposition, and entanglement. A focal point being that the central goal of Quantum Computing is to devise an algorithm that can be used to choreograph a pattern of constructive and destructive interference. So that for wrong answers (or for answers with a low probability of occurring) positive and negative amplitudes cancel each other out. Whereas for right answers (or answers with a high probability of occurring) positive amplitudes reinforce.

Part 1 is a good grounding in this rather opaque and nuanced topic. Part 2 builds on the subject’s underpinnings and discusses when a quantum computer will be built, quantum decoherence, the different ways to make a qubit, and Quantum Computing in pharmaceuticals. Consistent with Part 1, I continue to detail my conversations with Edward et al. ??

Early Supper

Ravenous (and slightly fatigued) after a day of discussing quantum physics, an early repast was in favour by all. ‘All’ being Edward (my redoubtable host), Charlie (the phlegmatic physicist), and Stephen (a man of letters and Edward’s close friend).

We arrived at a chic French bistro. A favourite of the coterie, who dined regularly together. Menus were not required. Each member of the group opting to have the chef’s special which was advertised in bold white lettering on the towering chalkboard that dominated this intimate venue.

After our orders had been taken and the waiter had exited the stage, I raised the next topic of discussion. Quantum Computing in Pharmaceuticals.

Quantum Computing in Pharmaceuticals

An area of interest – rather than expertise – within the group. But everyone agreed that pharmaceuticals would be one of the first realms to benefit from this emerging technology.

Pharma’s focus on molecular formations makes it well suited to the strengths of Quantum Computing. Molecules used for drug development are themselves quantum systems. What Quantum Computing can offer (hopefully in the not-too-distant future) is the ability to predict and accurately simulate the structure, properties, and behaviour/reactivity of molecules much more effectively than classical computing can. Target identification, drug design, and toxicity testing will all improve. Ultimately speeding up the research and development timeline.

Drug discovery is so costly, and time consuming*, that pharma companies have routinely been early adopters of technological advancements – molecular dynamics simulations, density functional theory, artificial intelligence – in an effort to reduce financial and time commitments. Quantum Computing is the next technology in line for adoption. ?

*The average cost of developing a new drug sits at c.$2.2B and takes north of seven years to create and launch (Deloitte).

A Main Course of Quantum Fuzziness

Following a toothsome starter – Galette de Bretagne (crepes with egg, bacon, and mushrooms) – the main course appears. A reverent silence falls around the table as four plates of Sole Meunière, accompanied by green beans almondine, buttery mashed potato, and cherry tomato confit are placed lightly in front of us. Pinot Bianco is adroitly poured to accompany the plat du jour. Our flowing conversation came to a grinding halt while eating. The four of us entranced by the culinary artistry and profusion of flavours. I broke the transfix by reciting,

?????????? ‘It is abundantly clear that when you travel down to the subatomic level, things get weird. It is unrecognisable. Our current understanding of reality is challenged. You can forgive a layperson for thinking that what Quantum Mechanics posits is preposterous.’ I asked laconically,

?????????? ‘Why do non-physicists – like me – have a difficult time understanding Quantum Mechanics, and by extension, Quantum Computing?’ Stephen jumped on the question saying,

‘Physicists also have a difficult time understanding and explaining Quantum Mechanics. To work in Quantum Computing, you need to be comfortable with uncertainty. The subatomic world is not deterministic, and Quantum Mechanics does not provide a deterministic description of the world. It is probabilistic and postulates through superposition that all possible outcomes have a probability of being realised upon measurement. Before the electron is measured, it is everywhere it can possibly be within the parameters of its situation. Quantum Mechanics tells a story that contains a beginning and end. Everything in the middle is blurry (known as quantum fuzziness). And what is really exciting is that physics does not yet have a definitive answer explaining the apparent loss of deterministic power. Many people find this uncertainty unpleasant.

To prove the probabilistic nature of quantum mechanics, the German physicist Werner Heisenberg constructed the Uncertainty Principle, asserting that we cannot know the values of certain pairs of physical variables (like position and velocity) with arbitrary accuracy.’ He continues by passionately acting out the following,

‘You might be thinking, ‘Wait! How can that be? My GPS on my phone frequently helps me to get to my desired destination – this wouldn’t be possible if it didn’t know my speed and location. A valid argument; however, remember, we are talking about the subatomic. The infinitesimal. The quantum uncertainty Heisenberg was referring to is inversely proportioned to the mass of the object. A car – because it is considerably heavier than a single electron (putting it mildly) has very small uncertainties in its speed and position. The electron has much larger uncertainties.

The takeaway: the bigger the mass is, the smaller the uncertainties in position and velocity; the smaller the mass is, the bigger the uncertainties are. This quantum uncertainty was a catastrophic blow to the old guard physicists (Einstein, Schrodinger, Planck) hanging on to notions of determinism.’ Stephen sips his premium fizzy water from a tall glass goblet and utters,

?????????? ‘Embrace uncertainty. Embrace the probabilistic nature of reality.’ ?

When Will Quantum Computers be Built?

My mind was now in too low a gear to continue a discussion on the inconsistencies of nature. My next question to the group was an attempt at having a lighter conversation. I said,

‘What developers and investors I speak to really want to know is when will a quantum computer be built? When could these machines be in their buildings?’ Edward said,

?????????? ‘This depends on what your definitions of ‘quantum computer’ and ‘built’ are. Quantum computers exist today; however, they are not the fully functioning, break the internet, solve climate change quantum computers that people are talking about when the topic of quantum is raised.

It is difficult to say when the fully functioning, error-free machine will be with us. Revolutions and technological advancements are often invisible in the moment. But when pressed, my stock answer is 10 years.’

Edward’s comment about scientific revolutions being inconspicuous (an idea that the philosopher of science, Thomas Kuhn, is famous for positing) is true. A century from now, scientists and historians will point to the period 1980 – 2040 and call it the quantum revolution. We are living through it. But many do not realise. Edward said,

?????????? ‘Asking when the quantum computer will be assembled is like asking when Quantum Mechanics was invented. Or when the classical computer was built. There are no specific dates. They happened in phases. And it will be the same for quantum computers.’

The next stage is to solve some of the Quantum Computing challenges that are hampering progress. One being ‘quantum decoherence’. ?

Quantum Decoherence

A principal challenge. Particles (like electrons) – which quantum computing is based on – exhibit some wave properties. When these waves are vibrating in unison (known as coherence), quantum mechanical calculations can be conducted. If coherence is not achieved, everything vibrates at a different frequency and calculations are erroneous.

Maintaining a state of coherence is akin to balancing a pencil on its tip. With no vibrations and external interferences, a perfectly balanced pencil would stay plumb. But when minuscule environmental disturbances are introduced, the delicate order is disturbed. The pencil falls and is in a state of decoherence. ??

Like the pencil, qubits must be shielded from noise (noise being described as heat, vibration, local radiation (WiFi, for example), cosmic rays, and even the Earth’s magnetic field). Failure to do so would flip zeros and ones and wipe out a crucial superposition. Or as Charlie put it, ‘uncontrolled noise will lead to your superposition collapsing like a soufflé.’ Increasing qubit coherence time and eliminating noise is a significant area of research. As is understanding the different ways to make a qubit.

More Than One Way to Make a Qubit?

There are a handful of competing visions for qubit construction being pursued by start-ups, academics, and tech giants. ‘Superconducting’ is the most popular, followed by ‘Quantum Dot/Silicon Chip’. Progress is being made with the ‘Linear Optical’ approach (using photons of light as the qubit), as well as with the ‘Trapped Ion’ approach. ‘Colour Centre/Nitrogen Vacancy’ and ‘Neutral Atoms in Optical Lattices’ are further approaches gaining momentum. But superconducting is the leading technique, and the superconducting ‘skeleton’ – the hanging muddle of gold wiring, analogous to a grand, modern-day chandelier – the most popular Quantum Computing image (see Fig 1).

This infrastructure is necessary to achieve the low temperatures – minus 460 degrees Fahrenheit – required to make durable qubits. This is colder than outer space. Two days is needed for Google’s cryostat to get the quantum chip down to this temperature (10 millikelvin). The chip – which is no bigger than a first-class postage stamp – sits at the bottom of the apparatus. This unimposing chip is what all the fuss is about and is what will permit scientists to accurately simulate the probability-based laws of Quantum Mechanics.

One of the most entertaining ways to create qubits that I have been informed about is the Electrons on Helium technique. In this approach, the qubits are single electrons that are trapped in a vacuum above the surface of superfluid helium and qubit states are created using microwaves. The method being a blend between the ion trap and superconducting approaches. The main benefit of which (I am reliably informed) is the long coherence times that are achievable. Charlie said,

?????????? ‘It is still unclear which approach will win the race. But the winning formula – either a single approach or a combination of a few – will become clearer. Hopefully within the next 12 months.’ ???

Final Thoughts

I have discussed Quantum Computing and Quantum Mechanics with a variety of different people while writing these articles. Quantum Computing is a fairly new topic (outside of the world of physics) and I have found that many non-specialists have struggled with the ‘common sense’ obstacle of the phenomenon. This is understandable – common sense is based on previous experiences. Introducing new ideas and technologies that are at odds with the way people see the world is one of the reasons why new ideas and technologies can take a while to gain traction (see the Diffusion of Innovations Theory). There are many people – like Edward, Charlie, Stephen, and Jaron – trained in the field of Quantum Mechanics/Computing that have developed a common sense for this vanguard of technological advancement, allowing them to make predictions with quantum theory. Even though these predictions are often at odds with Newtonian classical theory.

Richard Feynman famously said (in 1965) that nobody understands quantum theory. But quantum physics has come a long way since the mid-twentieth century. Many scientists today do understand quantum theory. Can use it, calculate with it, and make predictions with it that are provable in laboratory settings. As Jonathan Dowling says in his slick book ‘Schrodinger’s Killer App’, maybe what Feynman meant was that nobody understands ‘why’ quantum theory is weird the way it is – superposition, entanglement etc. Dowling goes on to observe that this is the case for many theories. Why is it that in electricity theory, there are opposite types of charges (positive and negative) that can either repel or attract each other. And in magnetism theory, there are opposite types of poles (north and south) that can either repel or attract each other. But in gravity theory, there is only one kind of mass that only attracts and never repels? Why is it that when we drop a pencil, it always falls down and never up? It is because this is the way things are. No experiment has ever demonstrated a pencil falling up. We accept it because there is no evidence to the contrary.

Learning about Quantum Computing over the past few months has been highly stimulating. I have thoroughly enjoyed speaking to computational mathematicians, quantum physicists, and erudite engineers – all of whom demonstrated a fantastic vehemence to realise the full potential of this technology. And if history has taught us anything, it is that if you give incredible tools to astute individuals, they will find something remarkable to do with it.

?

?

?