
Duncan Stewart sees himself as midway between a physicist and
an engineer. And that's exactly how he likes it.
His team in the Quantum Science Research group is sort of a hybrid
as well -- a mix of chemists, physicists, engineers and a computer
architect -- and he likes that, too.
"Most of the great advances in the future will come by combining
different fields of science," he says. "It's only by
mixing together different ways of looking at things that we really
are going to make significant progress."
Stewart, whose holds a PhD in engineering physics, is the Quantum
Science Research team's electronics tester. His job is to measure
and understand how electricity passes through the molecular circuits.
"My role is to determine what the thing is actually doing,
and then try to figure out why it's doing that," he says.
"The devices we're building behave in strange and wonderful
ways, and the truth is, we understand very little of that."
The goal: to understand the devices well enough so that they
can be controlled and mass produced. That's an incredibly complex
problem.
"At the very tiny sizes we're working with -- we're talking
20 or 30 atoms in one row -- it's very difficult to know whether
those atoms are in a straight line or crooked line, or if there's
another atom that found its way into there by mistake," he
says. "It's difficult to go in and characterize what we made
and to learn what is inside there."
Stewart uses several investigation techniques -- measuring how
devices behave when different voltages are applied, or when the
frequency of the electricity is changed, or when the temperature
changes, to name a few -- but none paints a full picture. At this
stage, the best he and his colleagues can do is put all of the
evidence together and guess.
Despite the team's recent success -- creating the highest-density
electronically addressable memory known to date -- the solution
is still a long way off.
"This particular demonstration is one successful milestone
on a long and torturous path that we hope leads to success,"
Stewart says.
"Each one of these devices is touched by at least three
people, usually four or five, and each of us contributes what
we're best at. And it works," he says. "The finished
product was better than any one of us could have built."

|