East Sussex
WORKING as a microengineer can be hellishly frustrating. It isn’t hard to
build cogs and levers only a few micrometres across, but just try handling these
Lilliputian objects. “Picking them up, moving them around and putting them in
the right place can be a big problem,” says Derek Chetwynd of the University of
Warwick’s Centre for Nanotechnology. Microengineers are crying out for decent
tools, he says. Until now, they have been using the microscopic equivalent of
Neolithic flint tools in order to move their creations about.
But the Stone Age is over—the laser spanner has arrived. Its swirling
beam allows scientists to manipulate objects no larger than red blood cells with
startling precision. And it is more than just a tool. One day it may power
micrometre-sized dynamos or motors, spinning their cogs and wheels on a shaft of
light.
Advertisement
The laser spanner is the offspring of another technique, the laser tweezers.
Invented by Arthur Ashkin at Bell Laboratories in New Jersey about twenty years
ago, laser tweezers are now standard equipment in many biotechnology labs across
the world. They rely on the forces that arise from the variation of light
intensity within a laser beam. The intensity of a standard laser
beam—which physicists call Gaussian—is greatest along the beam’s
axis and trails off towards the edges. Any partially transparent object caught
in the beam experiences a small force pulling it inwards towards the brightest
point at the centre of the beam.
Trapped by light
The trapping force can be explained by refraction. When a laser beam hits a
particle slightly off-centre, refraction bends part of the laser light outwards,
away from the axis of the beam. Another way of looking at this is that the
photons gain momentum in that direction
(see Diagram). And momentum is always
conserved, so the particle gains an equal amount of momentum in the opposite
direction, and moves towards the centre of the laser beam. If it strays too far,
the laser light hits off-centre on the other side and is refracted in the
opposite direction, pushing the particle back towards the centre. So the
particle is trapped.FIG-21215301.jpg
The trouble is that, like a bead threaded onto a piece of string, it is only
held in two dimensions. Laser tweezers like these can move an object around on a
microscope slide, but they can’t pick it up. Cheating gravity and actually
lifting a particle up is a little more tricky. To do this, the laser light must
be tightly focused onto the object through a strong lens. Near the focus of the
beam, the photons are travelling diagonally towards the particle. On refraction,
they gain momentum downwards. To compensate, the particle gains momentum
upwards—it levitates, trapped just below the beam focus. “They are
three-dimensionally trapped,” says Halina Rubensztein-Dunlop at the University
of Queensland in Brisbane.
The most exciting scientific discoveries rarely come from carefully planned
research programmes, and it required just one more, unexpected step for the
researchers to come up with the laser spanner. In January 1994 Hao He, a
graduate student working with Rubensztein-Dunlop, was attempting to trap ceramic
particles in a special kind of laser beam. When projected onto a screen,
conventional laser beams create a circular dot of light. But He’s laser was
operating in what is called the Laguerre-Gaussian mode, which causes the laser
energy to spiral through the air rather than travelling along the beam’s axis. A
Laguerre-Gaussian laser beam has the profile of a ring doughnut: when projected
onto a screen, it makes a bright ring of laser light surrounding a dark centre.
Forces exerted by the laser light mean that a small object in the centre should
remain trapped. His trap worked, but not quite as he expected. “To his and our
surprise, he saw that the ceramic particles were spinning,” says
Rubensztein-Dunlop.
The technique is so new that few researchers have tried it. But physicists
Miles Padgett and Les Allen from the University of St Andrews in Scotland first
considered the idea of a laser spanner in 1993, while working on two other
projects. One involved optical tweezers and the other was to establish the exact
character of the Laguerre-Gaussian laser mode. “We were doing these projects
right next to each other for months before we wondered what would happen if we
put a Laguerre-Gaussian mode into the optical tweezers,” Padgett recalls.
Padgett and Allen announced the idea of an optical spanner at a symposium in
the summer of 1994. After a year, they had their apparatus up and running,
except for one vital part—they couldn’t find the right object to
rotate.
Padgett and Allen wanted to lift their particle up into the air and set it
spinning at the same time. However, this required a particle with exactly the
right balance between absorption and transmission of the laser light they were
using. Too much absorption and no light would be refracted, so the particle
wouldn’t levitate. Too much transmission and it wouldn’t absorb the spiralling
photons which provided the spin. It wasn’t until early in 1996 that they found
what they were looking for.
Padgett had just bought a new bicycle, and with it came a little tube of
high-tech lubricant. “On the packet it said `micron-sized Teflon
spheres’—we thought we’d give them a go,” Padgett remembers. The results
were perfect. They illuminated a Teflon bead with their 25-milliwatt laser, and
watched as it rotated slowly about its laser axle.
The conversion from tweezers to spanners requires the addition of some
angular momentum, the property that an object gains through circular motion. A
standard laser beam, however, has an angular momentum of zero—the photons
that give the laser its power deliver their energy straight along the axis of
the beam.
Doing the twist
To give the beam angular momentum, its energy has to be given a twist. This
can be done using a modified diffraction grating, which is covered with
thousands of tiny parallel lines. As the beam’s photons pass through it, their
phase is shifted by diffraction. They no longer share the same phase, as
adjacent photons in an ordinary laser beam do, but each is given a slightly
different phase to its neighbour, so that the beam advances like a screw thread.
This is the Laguerre-Gaussian laser mode.
The beam’s energy now twists around its axis in a helical motion. As well as
having linear momentum along the beam axis, the spiralling photons have angular
momentum and it is this that set his ceramic particles spinning.
But angular momentum comes in two varieties—”orbital”, which arises
from the photon’s circular motion, and “spin”, which comes from a photon’s
rotation. To increase the angular momentum of the photons in a laser beam so
that more torque can be applied to a spinning particle, they can be set spinning
as they spiral. This is achieved by circularly polarising the laser beam using a
special type of polariser known as a quarter-wave plate. Linearly polarised
light has its electric field energy pulsing in only one dimension—up and
down or side to side. Circular polarisation sets the electric field rotating.
Photons in a circularly polarised beam have angular momentum from their spin
that, depending on the direction of rotation, either adds to or subtracts from
the beam’s orbital angular momentum.
The research groups at St Andrews and Queensland have both shown that by
adjusting the equipment slightly, the object trapped in the spanner can be
rotated clockwise or anticlockwise, quickly or slowly. Flip the modified
diffraction grating back-to-front and the direction of the photon’s orbital
angular momentum is reversed. Rotating the quarter-wave plate polariser changes
the sign of the photons’ spin. By delicately balancing the spin and orbital
angular momentum, the researchers can bring a rotating particle to a halt at
exactly the orientation they want. “You can spin them, change their direction of
rotation and stop them,” says Rubensztein-Dunlop.
This ability to give micrometre-sized particles a precise orientation could
be invaluable in many fields. It could certainly improve the future of magnetic
recording, says Ken Mackay of the Louis-Neel Laboratory in Grenoble, France.
Simply tweaking the properties of today’s magnetic materials will not be
sufficient to give us the next generation of high-density recording devices.
Nothing less than a revolution in recording technology is needed. “If you want
higher density recording, you’ll have to use nanostructures,” he says. These are
built from arrays of magnetic crystals or dots, with each dot representing one
bit of data. But according to Mackay, it is not simply a case of building these
arrays. “You have to know what the properties of each particle are,” he says.
And to do this you need something like a laser spanner.
Mackay is already collaborating with Padgett in a project to investigate the
properties of these individual magnetic crystals. To do this, he has developed
tiny magnetometers about one micrometre across, called micro-superconducting
quantum interference devices (microSQUIDs), which work at the temperature of
liquid helium. These U-shaped devices are etched onto silicon wafers which
resemble printed circuit boards. To measure the magnetic properties of a single
particle you have to get it onto the microSQUID, and until the invention of the
laser spanner, there was no reliable way to do this. “Until now we’ve been
pouring magnetic powder onto a silicon wafer covered with SQUIDS, washing it off
and hoping that one particle went in the right place,” he says. “With the
spanner we can take the particle that we want, orient it and put it exactly
where we want.”
Mackay envisages other applications for the spanner. A trapped magnetic
crystal could be at the centre of a highly sensitive magnetometer. The crystal’s
orientation would change whenever it was exposed to a magnetic field, and the
trapping potential would act as a restoring “spring”. According to Mackay’s
initial calculations, it would be almost as sensitive as the microSQUID, with
the big advantage that it would operate at room temperature.
Alternatively, Mackay says, if you set the magnetic crystal spinning inside a
micro-coil you would have the rudiments of a tiny dynamo. “It might be an
interesting way to power an independent micromachine,” he says.
Speed control
But there are still problems to overcome. Padgett’s Teflon beads absorb about
2 per cent of the laser beam’s power and spin at one or two revolutions per
second, exerting a tiny 10-17 newton metres of torque. “If you tried to drive a
gearbox with that, it would grind to a halt pretty quickly,” he admits. But you
can’t simply turn up the power of the laser. If the particle absorbs too much
energy, it just heats up and the energy is wasted.
Rubensztein-Dunlop believes that she may have found a solution. She has
replaced the ceramic particle with a calcite crystal which is a “birefringent”
material. As a linearly polarised laser beam passes through such a material, it
splits in two. These two beams will arrive at the other side of the crystal with
their light waves out of phase. If the thickness of the crystal is adjusted to
make this phase difference exactly 90 degrees, the beam that emerges is
circularly polarised—its photons have angular momentum, just as in the
Laguerre-Gaussian laser beam.
So the crystal spins as a reaction to its own action on the laser beam, like
a water sprinkler on a garden lawn. “It doesn’t absorb the light but, because
you must have conservation of momentum and energy, you’re giving a torque to the
crystal, which then turns,” says Rubensztein-Dunlop. “The more power you send
through it the faster it will spin.”
Rubensztein-Dunlop can increase the crystal’s rotation speed even more by
rotating the plane of the polarised light going into it. She does this by
mounting the polariser in a motor-driven mount. This allows you to precisely
control the speed at which it spins, she says.
Her experiments have spawned some wacky but feasible ideas. For starters,
what about micrometre-sized turbines? So far, her team has managed to spin a
single propeller-shaped crystal of calcite at 300 revolutions per second. The
researchers are excited about the future. “We are trying all sorts of things at
the moment—we believe that a micromachine is a possibility,” says
Rubensztein-Dunlop. “But I would really rather do a few experiments before
talking about great potential,” she cautions.
Like Rubensztein-Dunlop, Padgett believes that micromotors are a possibility,
but he is more optimistic about other uses. “I think it’s in the biological
applications that we’ll see big things happening—the ability to orientate
is very important, and you don’t necessarily need very much torque for that,” he
says. Orienting an enzyme to meet up with the right receptor, for example, could
prove an ideal problem for the laser spanner to tackle.
Justin Molloy, a biologist at the University of York, is interested in using
the spanner to measure the forces in tightly coiled molecules. “Looking at the
torsional rigidity of DNA would be interesting. You could stick a bead on the
end, twist it and unwind the helix, measuring the necessary force,” he
muses.
Grabbing and rotating a plastic bead is one thing, but rotating living
objects such as cells or DNA has proved a little harder—if they absorb too
much laser light, they are “opticuted”. Overcoming this problem may take a
little time and a lot of thought. But one thing is certain: the laser spanner’s
potential is a powerful stimulus for research in all fields. And microengineers
have finally got the set of tools they need to do what is surely the most fiddly
job in the world.