ANYONE who has suffered a broken bone knows about X-rays. Because they pass
through soft flesh more readily than through bone, X-rays create shadow images
of structures inside the body. More than a century of work has gone into
developing techniques to exploit this phenomenon, yet X-ray shadow images are
primitive compared to the sophisticated imaging techniques that are possible
with visible light.
The problem is that X-rays cannot be focused in the way that a magnifying
glass focuses visible light. Without lenses, researchers cannot produce
magnified images of small objects, they cannot easily concentrate X-rays onto
small areas or collect light from faint sources.
But all that is set to change. At the European Synchrotron Radiation Facility
(ESRF) in Grenoble, France, a team of researchers led by Anatoly Snigirev have
discovered a simple way to build the X-ray equivalent of a magnifying glass. And
its potential is huge. X-ray lenses could produce magnified images of tiny parts
of the human body, perhaps even revealing the structure and workings of
individual cells. Arrays of focused beams could be used to destroy small tumours
without damaging surrounding tissue. And focused beams will help materials
scientists to analyse the structure of ever smaller samples of materials such as
crystals.
Advertisement
Beam bending
The key to all this is the development of a lens that refracts X-rays.
Refractive lenses work because electromagnetic waves slow down when they pass
from a vacuum into a material. And if the rays enter the material at an angle
other than 90掳, their change in speed alters the direction in which they are
travelling. The ratio of the speed of light in a vacuum to its speed in a
material is known as the material鈥檚 refractive index. The greater the refractive
index, the more the incoming beam is bent.
When rays pass from one refracting medium to another, it is the difference
between the refractive indexes that determines how much the rays are bent. For
visible light, the refractive index of air is close to 1, but the refractive
index of glass is about 1.5. This is why glass bends visible light rays so
strongly.
But X-rays behave very differently. At their much shorter wavelength, there
is little difference between the refractive indexes of glass and air. X-rays
pass through conventional glass lenses almost undeviated. Another obstacle to
designing an X-ray lens is that materials with high refractive indexes tend to
absorb X-rays. These problems led researchers to believe that refractive X-ray
lenses were impossible to build.
This has meant that for X-rays there are few of the sophisticated imaging
techniques that are possible with visible light. Researchers who want to focus
X-rays now use curved mirrors to reflect a beam towards a focal point. A tiny
circular diffraction grating, called a Fresnel lens, can also focus X-rays into
a bright spot. More recently, researchers have used capillary optics, a
technique in which X-rays pass through a tapered hollow tube, reflecting off the
sides in the same way that visible light does in an optical fibre. 鈥淚 have been
in this business fifteen years, so the field of microfocusing X-rays is not new
to me. But in that time, I have never come across any experiments done with
refractive lenses,鈥� says Snigirev.
The idea of refracting X-rays came to Snigirev鈥檚 team when they were working
on how to focus X-rays for holography. The key to making a lens is to find two
materials that do not absorb X-rays strongly but have different refractive
indexes. Snigirev and his team knew that two such materials are air and
aluminium and that, of the two, air has a slightly greater refractive power.
With a leap of imagination, they realised that a beam of X-rays travelling
through an aluminium block should be brought to a focus by lens-shaped pockets
of air.
But there was one problem. Although the difference in refractive indexes
between air and aluminium is greater than many other combinations of materials,
it is still tiny鈥攐nly 2.8 x 10-6. This means that a single lens would
have a focal length of more than 50 metres. The researchers got round the
problem by combining the lenses to make a compound lens. Five lenses, for
example, would give a focal length of one-fifth of the original. Snigirev鈥檚
team鈥檚 design uses 30 cylindrical lenses, which give a combined focal length of
only 1.8 metres.
Each prototype lens consisted of a hole drilled into an aluminium block.
While thin sheets of aluminium absorb only a small amount of X-rays, a large
block absorbs much more and drastically reduces the intensity of the beam. To
maintain the beam intensity, the researchers drilled the holes in the block so
that the slivers of aluminium between each lens measured just 10 micrometres
across.
Such cylindrical lenses focus only in one dimension, turning a beam of X-rays
into a wedge shape (see
Diagram). But by using two compound lenses at right
angles, says Snigirev, it will be possible to focus in two dimensions, like a
magnifying glass. A beam of X-rays could then be turned into a cone.
The team is pleased with its early results. 鈥淭he refractive lens has many
advantages over other optics,鈥� says Snigirev. The prototype produces a focal
鈥渓ine鈥� that is 8 micrometres thick, which is comparable with other focusing
techniques that have taken years to develop. 鈥淲ith further development, we think
it will be possible to create submicron focal points which are not possible with
other methods,鈥� he adds.
The X-ray lens is simple to manufacture. What鈥檚 more, the lens鈥檚 focusing
ability depends less critically on shape and smoothness than that of focusing
mirrors. As a result, the massive heat loading created by the world鈥檚 most
intense X-ray sources is less likely to damage the lens. At some laboratories,
heat loading can reach as much as 1 kilowatt per square millimetre, enough to
raise the temperature of an uncooled target by hundreds of degrees. This can
distort even cooled reflecting optics, ruining the focus.
Refractive X-ray optics opens up a number of exciting opportunities, says
Snigirev, especially in medicine. As the resolution of a focusing system depends
on the wavelength of the radiation used, X-rays will produce even finer details
of the cells of the body. 鈥淢ammography is an area that will benefit, in
particular,鈥� says Snigirev.
By using microfocused beams 鈥攁rrays of parallel beams a few micrometres
across鈥攖he researchers also hope to examine tiny volumes of sample. This
could be of particular benefit in developing new materials when samples are
available in minute quantities. 鈥淔or the microfocus work, we need to focus the
beam down to about a micrometre鈥攚e鈥檙e hoping to use the new lenses for
this sort of application,鈥� explains Snigirev. Microfocusing will allow
researchers to see in unprecedented detail the processes going on in a single
cell. It will also reveal the crystal structure of tiny grains of proteins or
other large molecules which are difficult to grow into large crystals.
In the future, Snigirev hopes to improve the lenses by using materials other
than aluminium. 鈥淏oron is more interesting,鈥� he confesses. 鈥淚t鈥檚 a less
absorbent material but the refraction is nearly the same as aluminium.鈥� Although
it is more difficult to machine, boron has a melting point of more than 2000
掳C, compared with 658 掳C for aluminium. So it is much better able to
withstand large heat loads.
Lenses in the future may even be able to focus in two dimensions, by using
spherical holes in the aluminium matrix instead of cylindrical ones. One way to
make spherical holes is to embed polymer beads into a metal block. Snigirev
hopes that because polymer beads can be fashioned into more conventional lens
shapes, his team can correct some of the aberrations his lenses suffer from.
鈥淭his will give us even smaller spot sizes. I鈥檇 say 500 angstroms [50
nanometres] might be possible,鈥� he says.
With a little technological development, refractive X-ray lenses may no
longer be the dream of a few, but reality for many researchers worldwide. But a
large part of the appeal of X-ray lenses lies in the simplicity of the concept.
Why hadn鈥檛 anyone thought of it before? 鈥淚 don鈥檛 know,鈥� says Snigirev, 鈥渕aybe
it鈥檚 just too simple.鈥�