Microfocusing
Mirrors Advance Materials Science
Editor's note:
The following article describes an ORNL technology development that
won an R&D 100 Award in 2000. The awards are presented annually
by R&D Magazine
in recognition of the year's most significant technological innovations.
ORNL's 107 R&D 100 awards place it first among DOE laboratories
and second only to General Electric.
Research
Challenges | Origin
of the X-ray Machine | X-ray Microbeams at APS
A new era of X-ray science
is being ushered in by an advance in X-ray mirror technology in which
an ORNL scientist played a key role. Gene Ice, leader of the X-ray Research
and Application Group in ORNL's Metals and Ceramics Division, worked
with Beamline Technology Corporation of Tucson, Arizona, to develop
advanced X-ray microfocusing mirrors, which received an R&D 100 Award
from R&D magazine in 2000. The new technology, combined with
computer software to analyze X-ray patterns, opens a new frontier for
materials scientists, permitting direct observation of the building
blocks of most materials. The technology will allow scientists to characterize
material defects down to less than a millionth of a meter (a micron).
It will also help them determine which manufacturing processes make
more reliable electronic circuits, more efficient materials for energy
production, and stronger, lighter transportation materialssuch
as aluminumfor car bodies.
|
Gene
Ice inspects two advanced elliptical X-ray microfocusing mirrors
used for focusing X rays to submicron spots. The mirrors received
an R&D 100 award in 2000. (Photo by Curtis Boles.)
|
The building blocks of
most materials are small, single-crystal grains packed together in a
complicated network that includes intersections called grain boundaries.
"Our 3D X-ray crystal microscope will allow researchers for the first
time to see the three-dimensional crystal structure of each grain making
up a material of interest," Ice says. "We will now be able to look at
a grain's orientation, size, and shape. We can determine if a grain
points in a different direction, becomes deformed, or breaks up into
smaller subgrains as a result of internal forces induced by a manufacturing
process. We can observe the changes in grains that cause the material
to fail."
Research
Challenges
In the mid-1990s, Ice
was traveling around the nation to try to win funding for materials
research experiments that would use brilliant X rays from the Advanced
Photon Source (APS), which began operating in 1997 at Argonne National
Laboratory. During his visit to Bell Laboratories, he learned of an
urgent technical challenge. A major factor in the reliability of computers
and computer chips is strain in integrated-circuit wires. The tiny polycrystalline
aluminum or copper wires in the integrated-circuit chips expand and
shrink differently from the silicon bulk material as temperature changes
during processing or computer operation. Together with the mechanical
forces associated with current flow through the wires, this effect results
in strain, which varies at the micron level because of distortion in
individual crystalline grains. This strain can affect the mechanical
evolution of the aluminum or copper interconnect wires during operation
or even when a chip is made. Mechanical evolution is a change in the
shape, size, rotation, and strain (internal stretching) of a grain.
It is difficult to study strain, however, because the interconnect wires
are buried under an amorphous silicon dioxide film. Today Ice and Ben
Larson of the Solid State Division (SSD) and their ORNL colleagues are
using the 3D X-ray crystal microscope at APS to study strain in integrated-circuit
wires, as well as other materials science problems.
|
X-ray
image of grains in aluminum interconnect wires in a silicon semiconductor
chip used in a computer.
|
For example, John
Budai of SSD has been leading an effort to advance the technology of
making effective high-temperature superconducting materials in which
a thin superconducting single-crystal film is deposited on a buffer
layer that coats a metal substrate. In the best superconducting wires,
the film grains are deposited on aligned metal substrates and actually
increase their alignment as they grow so that most of the crystals point
in the same direction. To determine which processes repeatedly produce
superconducting materials with the best alignments, ORNL researchers
are using the 3D X-ray crystal microscope to examine many different
samples produced by various processes. Their results may help lead to
the development of commercial superconducting wires and devices.
Origin
of the X-ray Microscope
In the late 1940s and
early 1950s, P. Kirkpatrick and Albert V. Baez (father of singer Joan
Baez) demonstrated that two mirrors could be arranged to focus an X-ray
beam at a small spot on a target material. Such "crossed mirrors" came
to be known as Kirkpatrick-Baez mirrors. In the past 50 years, X-ray
mirror technology has rapidly progressed, especially in the production
of mirrors with flat and spherical surfaces. (A mirror with a spherical
surface can be visualized as a small strip from a hollow glass sphere).
To focus the beam
on a spot small enough to "observe" crystalline grains, it was necessary
to fabricate and polish mirrors with elliptical surfaces. The ideal
elliptical mirrors can be visualized as strips cut out of the surface
of glass shaped like a symmetrical egg. However, the technology does
not exist to polish an elliptical mirror to get the required X-ray quality.
In recent years, flat mirrors have been elastically bent to produce
elliptical surfaces. Although bent mirrors can produce beams that focus
to a spot less than a micron in diameter, they are sensitive to mounting
stresses and thermal loads.
In the 1990s,
with the construction of the APS it was apparent that improved X-ray
optics technologies would be needed to exploit the brilliant, laser-like
X rays expected from the APS. Ice pushed for improvements in X-ray optics
as early as 1990. His sense of urgency grew when he became part of a
group that would have a dedicated beam line for experiments at APS,
thanks to internal ORNL funding and a collaborative agreement in 1996
with Howard University, a historically black university. Ice called
the X-ray optics experts he knew around the country. Eventually, he
found a company willing to try making an elliptical mirror to his specifications.
The company, Beamline
Technology Corporation of Tucson, Arizona, was interested in depositing
a coating of varying thicknesses to build a mirror with the required
shape. Andrew Lunt, general manager of Beamline Technology, told Ice
that the company had already developed a differential deposition technology
for use in correcting small shape defects in X-ray mirrors.
To make the desired
elliptical mirror surface, a Beamline Technology team cut out a small
strip from the surface of an X-ray-polished spherical mirror made of
ultra-low-expansion glass or single-crystal silicon. Scientists at Argonne
measured the surface shape, and Ice calculated the required deposition
profile. Then a layer of chrome was deposited to bind the differentially
deposited coating to the substrate. To build an elliptical surface in
the shape specified by calculations, a thin layer of gold was differentially
deposited on the chrome coating. Gold atoms were sputtered from a source
using a computer-controlled machine; the thickness of the deposited
gold coating was adjusted by changing the machine's power-the higher
the power at any one point, the thicker the coating there. A final layer
of palladium was deposited on the gold coating to make the mirror highly
reflective of X rays up to 24,000 electron volts (24 keV).
Throughout the
process, many measurements were made to determine the mirror shape.
Beamline Technology sent samples to ORNL for Ice to test at APS. The
differentially coated mirrors were also tested successfully by researchers
at the National Synchrotron Light Source at Brookhaven National Laboratory,
where Ice worked for five years. Based on feedback from the researchers,
Beamline Technology perfected the differentially deposited mirror technology
for the X-ray crystal microsope.
X-ray
Microbeams at APS
Ice, Larson, Budai, and
their ORNL colleagues Jon Tischler, Eliot Specht, Jin-Seok Chung, Nobumichi
Tamura, Wenge Yang, Ki-Sup Chung, and Mirang Yoon have performed experiments
at the APS using the X-ray crystal microscope. Although the microscope
is based on a well-known X-ray dif- fraction technique called polychromatic
Laue diffraction, no one had ever pushed the technology to obtain such
detailed quantitative information. In fact, several years ago X-ray
experts advised ORNL that the technology was not possible. The microscope
is possible, Ice says, because of the availability of ultra-brilliant
synchrotron radiation at the APS and three other major technical advances.
One advance is novel achromatic focusing mirrors that allow the microscope
to achieve high spatial resolution with broad spectrum X-ray beams.
Another is an innovative X-ray scanning monochromator built at Howard
University that allows for the rapid measurement of absolute stresses
(internal forces) in crystals. A third is specialized pattern analysis
software that determines the type, orientation, and stress of individual
grains from overlapping multi-grain X-ray scattering patterns.
|
The
advantage of bombarding a material with the Advanced Photon Source's
(above) high concentration of X rays of many different wavelengths
is that it increases the chances of satisfying Bragg's Law, a
statement of the conditions under which a crystal will reflect
a beam of X rays with maximum intensity. Because the grains in
a material may be pointing in different directions and because
it is slow and impractical to rotate the sample, focusing a finely
collimated beam of polychromatic X rays on each single-crystal
grain should produce detailed information regardless of the grain
orientation.
|
The X-ray diffraction
patterns are captured by a charge-coupled detector (CCD), which is carefully
calibrated to determine the precise direction of the scattered X-rays.
From the pattern on the CCD, the number, orientation, and distortions
of the crystal grains can be determined. This information is essential
to understanding the forces driving the evolution of the grains at the
subgrain level. A complete description of the local forces that are
deforming each grain can be measured using the Howard University micromonochromator.
Ice credits Jim Roberto,
then director of SSD, for encouraging him and Ben Larson to apply successfully
for internal funding from the Laboratory Directed Research and Development
(LDRD) Program at ORNL to build an X-ray microprobe for mesocale studies.
With LDRD funding and
through collaborations with Howard University scientists, Larson and
Ice put together all the pieces for a working 3D X-ray crystal microscope.
Early results provided essential support for a successful two-percent
initiative proposal from ORNL to DOE's Office of Basic Energy Sciences.
In addition to
winning an R&D 100 Award, the new elliptical mirrors have another claim
to fame. They have beaten the bent mirror focus record (0.8 x 0.8 mm2)
by achieving a focused spot size of ~0.4 x 0.5 mm2.
Clearly, these achievements reflect well on the sponsors of the development
of the 3D X-ray crystal microscope.
Beginning
of Article
Related Web
sites
ORNL's
X-ray Research and Application Group
ORNL's
Metals and Ceramics Division
ORNL's Solid State
Division
Advanced Photon
Source (APS)
National
Synchrotron Light Source
DOE's
Office of Basic Energy Sciences