2 November 2019 | New Scientist | 9IT COULD be decades before we
have quantum computers that
can do anything practical, but
once we get there, what will we
use them for?
The first application could
be in chemistry. As the physicist
Richard Feynman once said,
“Nature isn’t classical, dammit,
and if you want to make a
simulation of nature, you’d
better make it quantum
mechanical.”
The forces, molecules,
particles and more involved in
a chemical reaction all interact
in different ways. When trying
to simulate what will happen,
the combinations can quickly
get out of hand, at least on a
classical computer.
Quantum computers,
however, are particularly
good at dealing with this
combinatorial explosion.
According to researchers at
Microsoft, simulating a single
water molecule requires
16,000 bits on a classical
computer. But a quantum
computer would need just
24 quantum bits, or qubits.
The hope is that a quantum
computer could reveal someof chemistry’s most elusive
secrets, such as how to improve
the Haber-Bosch process. This is
used to help make fertiliser for
crops around the world, but it
is energy-intensive, generating
about 1 per cent of annual global
carbon dioxide emissions.
Because of this, chemists are
on the hunt for a more-efficient
replacement. The Microsoft
team estimated that just
100 qubits, double the number
in Google’s new quantum
computer, could be enough
to crack this nut.
There is also hope that
quantum computers could
boost our understanding
of superconductivity. This
strange but useful property
occurs when electrical charge
moves through a material
without resistance, and it could
vastly improve the efficiency
of power grids.
Currently, we can make
superconductive materials only
when they are extremely cold,
which requires lots of energy,
nullifying the advantage. By
looking at the huge number
of ways that atoms in a material
interact, quantum computersmay unlock a method of
making superconductors at
much higher temperatures.
The pharmaceutical
industry could also benefit
from quantum computing.
Drug design relies on predicting
how a protein folds based on its
building blocks and the external
forces acting on it. Quantum
computers could considerably
shorten the time required
to work out how different
proteins will act. Teams
around the world are working
on quantum algorithms
designed to do just this.
These examples are probably
just scratching the surface.
Classical computers were
first used for performing
calculations and solving
technical tasks, such as
attempting to break secret
codes. At the time, it would
have been hard to imagine
that, less than a century later,
billions of people would carry
hand-sized computers around
with them everywhere they
went. Imagining what they
would use them for would have
been even more difficult.
The same is true for quantum
computers. The first practical
quantum computers will be
extremely expensive and used
for niche applications. But with
so many possibilities beyond
those, predicting what will
happen is nearly impossible. ❚Cybersecurity Real-world applications
Donna Lu Chelsea Whyte
GOOGLE’S claims of quantum
supremacy have some people
worried that the internet is now
broken. Within hours of the news
breaking, US presidential candidate
Andrew Yang tweeted: “It means...
no code is uncrackable”.
But Yang isn’t quite right.
Although there is reason to believe
that quantum computers will
threaten encryption, there is still a
long way to go before that happens.
The internet uses encryption to
stop sensitive data, such as bank
details or private messages, being
read by prying eyes. The process
involves mathematically jumbling
up data before it is transmitted
and then deciphering it once it is
received – a process vulnerable
to quantum computers.
RSA cryptography, which is used
all over the internet, relies on it
being very difficult to find the prime
numbers that multiply together
to make up a large number. It could
take current classical computers
billions of years to factor numbers
that are 2048 bits – 617 decimal
digits – or longer. But using an
algorithm developed by computer
scientist Peter Shor in 1994, a
quantum machine could do it in just
8 hours, by one estimate. However,
it would need 20 million quantum
bits, or qubits, to achieve this.
Google’s quantum computer
only has about 50 qubits, so such
a device is a long way off. And
attempts are already well under
way to find encryption methods
that can resist quantum attack. ❚
Has Google
accidentally broken
the internet?
What can we do with
a quantum computer?
MATT LIMB OBE/ALAMY STOCK PHOTOQuantum computers could
find ways to improve the
efficiency of farming“Quantum computers
are starting to
approach the limit of
classical simulation”
IBM
In a blog post by Edwin Pednault,
John Gunnels and their colleagues2100
Bespoke quantum
computing chips are
integrated into our
devices. Who knows
what we will use
them for, but we
will surely think
of something