,,
!?
BNL-66825
CHAPTER 8
SYNCHROTRON-RADIATION
INDUCED X-RAY EMISSION (SRIXE)
Keith W. Jones
Brookhaven National Laboratory
Upton, New York 11973
To be Published in
Handbook of X-Rav SRectrometrv, Second Edition
Edited by
Ren6 E. Van Grieken and Andrzej A. Markowicz
Marcel Dekker, Inc., New York, New York
1999
By acceptanceofthisarticle,thepublisherand/orrecipientacknowledgestheUSGovernment’srightto retaina nonexclusive,royalty-free
Iicensein andto any copyrightcoveringthii paper.
DISCLAIMER
This report was prepared as an account of work sponsored
by an agency of the United States Government.
Neither
the United States Government nor any agency thereof, nor
any of their employees, make any warranty, express or
implied, or assumes any legal liability or responsibility for
the accuracy,
completeness,
or usefulness
of any
information, apparatus, product, or process disclosed, or
represents that its use would not infringe privateiy owned
Reference herein to any specific commercial
rights.
product, process, or service by trade name, trademark,
manufacturer, or otherwise does not necessarily constitute
or imply its endorsement, recommendation, or favoring by
the United States Government or any agency thereof. The
views and opinions of authors expressed herein do not
necessarily state or reflect those of the United States
Government or any agency thereof.
DISCLAIMER
Portions
of this document
in electronic
produced
document.
image
may be illegible
products.
Images
from the best available
are
original
!’
?)
CHAPTER 8
SYNCHROTRON-RADIATION INDUCED X-R4Y EMISSION (SRIXE)
K. W. Jones
Brookhaven National Laboratory
Upton, New York 11973 USA
OUTLINE
I.
Introduction
II.
Properties of SynchrotronsRadiation
III.
Description of SynchrotronsFacilities
IV.
Apparatus for X-Ray Microscopy (X04)
A. Collimated X-Ray Microscopes
B. Focussed X-Ray Microscopes
C. Experiments at a Distance
v.
Continuum and Monochromatic Excitation
VI.
Quantitation
WI.
Sensitivities and Minimum Detection Limits
VIII.
Beam-Induced Damage
IX.
Applications of SRIXE
A. Archaeology
B. Biology/Medical: Calcified Tissue Studies
c. Geology and Environmental Sciences
D. Materials and Chemical Sciences
x.
Tomography
XI.
Extended X-Ray Absorption Fine Structure (EXAFS) and
X-Ray Absorption Near-Edge Structure (XANES)
XII.
Future Directions
XIII.
Selected References
I.
INTRODUCTION
Elemental analysis using emission of characteristic x rays is a well-established scientific
method. The success of this analytical method is highly dependent on the properties of the source
used to proc[uce the x rays. X-ray tubes have long existed as a principal excitation source, but
electron andl proton beams have also been employed extensively.
The development of the
synchrotronsradiation x-ray source that has taken place during the past 40 years has had a major
impact on the general field of x-ray analysis. Even tier 40 years, science of x-ray analysis with
synchrotronsx-ray beams is by no means mature. Improvements being made to existing synchrotrons
facilities and the design and construction of new facilities promise to accelerate the development of
the general scientific use of synchrotronsx-ray sources for at least the next ten years.
The effective use of the synchrotronssource technology depends heavily on the use of highperformance computers for analysis and theoretical interpretation of the experimental data.
Fortunately, computer technology has advanced at least as rapidly as the x-ray technology during
the past 40 years and should continue to do so during the next”decade. The combination of these
technologies should bring about dramatic advances in many fields where synchrotronsx-ray science
is applied.
It is interesting also to compare the growth and rate of acceptance of this particular research
endeavor to the rates for other technological endeavors. Griibler [1997] cataloged
the time
required for introduction, diffi.wion,and acceptance of technological, economic, and social ch~ge
and found mean values of 40 to 50 years. The introduction of the synchrotronssource depends on
both technical and non-technical factors, and the time scale at which this seems to be occurring is
quite compatible with what is seen for other major innovations such as the railroad or the telegraph.
It will be interesting to see how long the present rate of technological change and increase in
scientific use can be maintained for the synchrotronsx-ray source.
A short summary of the present state of the synchrotronsradiation-induced x-ray emission
(SRIXE) method is presented here. Basically, SRIXE experiments can include any that depend on
the detection. of characteristic x-rays produced by the incident x-ray beam born the synchrotrons
source as they interact with a sample. Thus, experiments done to measure elemental compositio~
chemical state, crystal, structure, and other sample parameters can be considered in a discussion of
SRIXE. It is also clear that the experimentalist may well wish to use a variety of complementary
2
L
techniques for study of a given sample.’ For this reason, discussion of computed microtomography
(CMT) and x-ray diffraction is included here.
It is hoped that this present discussion will serve as a succinct introduction to the basic ideas
of SRIXE for those not working in the field and possibly help to stimulate new types of work by
those starting in the field as well as by experienced practitioners of the art.
The topics covered include short descriptions of (1) the properties of synchrotronsradiation,
(2) a description of facilities used for its production, (3) collimated microprobe, (4) focussed
rnicroprobes, (5) continuum and monoenergetic excitation, (6) detection limits, (7) quantitation, (8)
applications of SRIXE, (9) computed microtomography (CMT), and (1O)chemical speciation using
x-ray absorption near-edge structure (XANES) and extended x-ray absorption fme structure
(EXAFS). An effort has been made to cite a wide variety of work Ilom different laboratories to
show the vital nature of the field.
There are many review articles and books that cover all aspects of the production and use of
synchrotronsradiation [Kim, 1986; Kunz, 1979; Margaritondo, 1988; Winick and Doniach, 1980].
The early article on SRIXE by Sparks [1980] is a usefil source of information on various aspects
of the use of SFUXEas part of a high-resolution x-ray microscope (XRM) system. Several more
recent review articles have covered new developments [Chen et al., 1990; Jones and Gordon, 1989;
Knochel, 1990; Vis, 1990; Vis and van Langevelde, 1991].
Microscopy using x rays can be carried out in several ways. A large effort is aimed at
producing extremely high resolution maps of the linear attenuation coefilcient for low energy x rays
in biological materials [Sayre et al., 1988]. Another approach is based on detection of electrons
emitted from the specimen [Ade et al., 1990]. Elemental detection can be accomplished by mapping
above and below the absorption edge for the element. Rarback et al. [1987] have shown that this
may be a preferable method for elements lighter than about calcium because of the small values of
the fluorescence yield. This approach has been used by Kenney et al. [1985] to study the calcium
distribution in a bone specimen using the Ca L absorption edge for the image formation.
Image formation using fluorescent x rays is advantageous for the high-Z elements in terms
of attaining the best possible values for minimum detection limits (MDLs) combined with the best
possible spatial resolution. It is not possible to use x-ray detection in some situations where the size
of the specimen is much bigger than the absorption depth of the fluorescent x rays. In this case also,
3
1’
the use of the above- and below-edge imaging using the techniques of computed microtomography
is possible. The use of these methods in common on a single specimen can be helpfhl. In the
discussion that follows, the use of all these methods is discussed, although the major emphasis is
placed on SFWH3.
Images may also be formed using other types of information obtained through probing the
sample. X-ray difiiaction can be used to determine crystal structure. Information on the chemical
state of the elements and position in the lattice can be obtained by use of XANES or EXAFS. These
are, of course, methods widely applied with both conventional and synchrotronsx-ray sources. A
merging of these methods with technologies for producing micrometer-si~d
beams is a natural
direction for development of the instrumentation to follow. This is beginning to happen now, and
the rate of development is bound to increase in the I%ture.
No matter what mode is used for image formation or to characterize a single volume element
in a specimen, the work is dominated by the fact that only limited numbers of atoms are present to
produce a signal. As an example, consider the number of zinc atoms present in an organic material
as a fimction of the specimen volume probed for a constant weight fraction of one part per million
@pm). Values are shown in Figure 1, and it can be seen that there are only 104atoms present in a
1- m3 volume. Detection of such a small number is a technical challenge no matter what mode of
detection is used.
II.
PROPERTIES
OF SYNCHROTRONS RADIATION
It has been known for almost one hundred years that the acceleration of a charged particle
will result in the radiation of electromagnetic energy. The development of the betatron and
synchrotronselectron accelerators fifty years later led to the experimental observation of radiation
from electrons circulating in a closed orbit @31deret al., 1947] and to naming it specifically
synchrotrons radiation after the accelerator used to produce it. The first recognition of the unique
properties of the synchrotronsradiation by Tombulian and Harman [1956] then brought about an
explosive development of activity in constructing improved sources for production of the radiation
and to using the radiation in experimental science.
The cuiginal synchrotrons were designed for use in nuclear physics research. Later f=ilities
were designed to optimize the conditions for production of x rays. The components of the new
4’
facilities include asourceofelectrons orpositrons andanaccelerator toproducehigh-energy beams.
~sti@tbe
done byuseofa
linemaccelerator toproduce energies ofmowdlOOMeV.
These
beams are then injected into a synchrotronsand boosted to energies in the GeV range. Finally, the
beam is stored in the same accelerator used to attain the final energy or in a separate storage ring.
An acceleration field in the ring is used to supply energy to the beam to compensate for the radiation
energy loss. The lifetime of the stored beam is many hours, so that in practice the synchrotrons
source is almost starting to become similar to a standard x-ray tube in use.
The synchrotron-produced x-ray beams have unique properties that make them desirable for
use. They have a continuous energy distribution so that monoenergetic beams can be produced over
a wide range of energies. The photons are highly polarized in the plane of the electron beam orbit
which is extremely important for background reduction in SRIXE-type experiments in particular.
The x rays are emitted in a continuous band in the horizontal direction, but are highly collimated in
the vertical direction. It is therefore possible to produce intense beams with little angular divergence.
The source size is small, and as a result, the production of intense beams of small area is feasible.
The synchrotronssource is a pulsed source because of the nature of synchrotron-type accelerators.
The x rays are produced in narrow bursts, less than 1 m in width, and have a time between pulses
of around 20 ns or more.
The main parameters of interest in defining the synchrotronssource are then
1. Magnitude of the stored electron/positron current. Typically, currents are in the range
from 100 to 1000 mA. Lifetimes of the stored bem are many hours. The lifetimes for
stored electrons at the National Synchrotron Light Source (NSLS) at Brookhaven
National Laboratory, Upton, New York, USA, have typical values of around 24 hours.
At Laboratoire pour L’Utilisation de Radiation Electromagnetique (LURE), Orsay,
France, where positron beams are used, the lifetimes are even longer since the positive
beam does not trap positively charged heavy ions produced from the residual gases in the
vacuum chamber.
2. Assuming that the size and angular divergence of the electron beam are not important,
the source brightness is defined as (in units of photons-’ mr-’ (O.lVObandwidth)):
(1)
5
where E is the energy of the electron beam in GeV, i the current in amperes, H2(0/co2)
is a fimction tabulated in ref. 4,0 is the photon angular frequency, and COC
is the critical
frequency which splits the emitted power into halves and is given by the expression
34y3c/2pwhere y is the electron energy in units of the electron rest mass, c is the velocity
o~light, and p is the radius of curvature of the electron path. The angles 6 and ~ are the
angles of emission in the plane of the electron orbit and perpendicular to that plane,
respectively.
3. The total photon emission is found by integrating over 1$and is given by (in units of
photons S-lrnr-l (O.1% bandwidth)-l:
——
‘iI = 2.457
Lie
x
10el E(GeV) i(~) G~~-
(2)
where G1(u/@ is tabulated by Kim [1986].
For electron beams with non-zero emittance (ftite area and angular divergence), it is
necessary to define another quantity, the brilliance, which is the number of photons emitted into
angular intewals de and d~ at angles 6 and ~ from an infinitesimal source area (in units of
photonss-l rm-2mm-2(O.1% bandwidth)-l). The values of brilliance and brightness are important in
evaluating the performance expected for focussed and collimated x-ray microscopes. Kim [1986]
provides a detailed discussion of the emittance effects, polarization, and performance of wiggler and
undulator insertion devices.
Values attained for these quantities at the 2.5-GeV x-ray ring of the NSLS are shown in
Figures 2-4. [ts x-ray beams are typical of those produced by second-generation synchrotronsstorage
rings. The ring energy is high enough to produce x rays over an energy range sufficient to produce
K-x rays from elements to about Z = 40 with good efficiency and L-x rays throughout the periodic
table. Thus it is highly suitable for use as the basis of a system for x-ray microscopy-based SRIXE.
The brilliance obtained horn two third-generation synchrotrons is shown in Figures 5-6 as a fimction
of the photon energy for several types of devices. The brilliance obtained from an undulator
insertion device at the 7-GeV Argonne National Laboratory Advanced Photon Source is shown in
Figure 5. The brilliance for different bending magnet, undulator, and wiggler devices at the 8-GeV
6
,,
t,
Spring-8 facility in Japan is given in Figure 6. It can be seen that the third-generation storage rings
have gained approximately 3-4 orders of magnitude in brilliance when compared to that found from
bending magnets at the NSLS. Similar gains are found for third-generation rings operating at lower
energies although the spectrum of x rays is naturally affected by the energy of the stored
electron/positron beam. This increase translates to a gain in x-ray intensity that enables new types
of x-ray fluorescence experiments.
The high degree of linear polarization of the x rays flom the synchrotronssource is a major
factor in making the synchrotrons XRM a sensitive instrument.
The physics describing the
interaction of polarized x rays with matter is therefore an important topic. Hanson [1990] has carried
out an extensive examination of the scattering problem and given methods for assessing particular
geometries used in the XRM. A second point is the continuous x-ray energy spectrum extending
from energies in the i&red
to hard x rays with energies of several hundred KeV. The tunable
energy is important for many types of experiments.
III.
DESCRIPTION OF SYNCHROTRONS FACILITIES
There are now many synchrotronsfacilities located around the world that are suitable for use
in various types of x-ray spectrometry measurements. They fall roughly into three classes.
First-generation synchrotrons were often built primarily as high-energy physics machines and
were used secondarily for synchrotronsradiation production. The Cornell High Energy Synchrotrons
Source (CHESS) at Ithac% New York, USA, is an example.
Second-generation synchrotrons were optimized as radiation sources, and as a result, produce
x-ray beams with superior brilliance and brightness characteristics. The two rings of the Brookhaven
NSLS fdl in this category.
Finally, the third generation of synchrotrons is now being designed. They will be the first
sources intended to incorporate insertion devices, wigglers, and undulatory in the design phase. In
some cases the ring energy is increased to give better performance with the insertion devices.
Synchrotrons designed with these features will begin operation in the latter part of the 1990s. The
European SynchrotronRadiationFacility (ESRF) at Grenoble, France, the Super PhotonRing-8 GeV
(SPring-8) in I&m@ Japan, and the Advanced Photon Source (APS) at Argonne National
Laboratory, Argonne, Illinois, USA, illustrate this case.
7
Alistingofanumber ofsynchrotronlaboratories producing high-energy x-ray beams suitable
for use in SRIXE is given in Tables 1 through 3 [Fuggle, 1990; Jackson,1990; Winick,1989 and
1990]. The number of XRM beam lines is growing rapidly and their employment for research as a
result is becoming widespread.
Iv.
APPARATUS FOR X-RAY MICROSCOPY (XRM)
The apparatus that is used for x-ray microscopy measurements varies substantially from
Laboratoryto laboratory. A schematic diagram of the components of a very comprehensive system
is, shown in Figure 7. All the components are not necessarily used in a specific instrument. The
mos~important differences between systems lies in the treatment of the incident beam. The simplest
approach is to use the white beam (fill-energy spectrum) and a collimator. The more complex
systems use fbcussing mirrors to collect more photons and demagnifj the beam and monochromators
to produce rnonoenergetic beams. At present, the performance of the various systems is quite
comparable in terms of spatial resolution and MDLs. Thus, a versatile and flexible approach to the
choice and design of the components is advisable. A number of the different instruments are
described briefly in the following sections.
A.
Collimated X-Ray Microscopes
A highly effective XRM system can be made by simply collimating the white beam
(continuous energy) of x rays produced by the synchrotrons.This approach has been followed mainly
by groups at the Hamburg storage ring &ochel
et al., 1983], HASYLAB, and at the Brookhaven
NSLS [Jones et al., 1988]. The use of white radiation is feasible because the ~gh brightness gives
a high flux ofphotons in a small area, and the high polarization of the synchrotronsbeams mininizes
scattering horn the sample into the detector. White radiation also makes possible efilcient multielement detection over a very broad range of atomic number.
The collimators for the instrument can be made from a set of four polished tantalum strips.
The strips can be spaced apart with thin plastic or metal foils to produce apertures useable to a beam
size of 1 micrometer. Alternatively, the slits can be attached to individual stepper motors for
producing variable-sizedbearns of size greater than about 10micrometers. The collimation approach
is ultimately limited by beam-spreading related to finite source and pinhole dimensions and to
dil%action.
The collimated XRM (CXRM) was first operated at the now defunct Cambridge Electron
Accelerator by Horowitz and Howell [1972]. They used a pinhole made by evaporation of a thick
gold layer around a 2- ~m quartz fiber followed by subsequent etching away of the quartz. The x-ray
energy was about 2.2 keV, and image contrast was achieved using determination of the linear
attenuation coefficient. The specimen was moved past the incident beam, and fluorescent x rays
were detected with a proportional counter. A spatial resolution of 2 j.tmwas measured with this
pioneering early apparatus.
Later versions of the CXRM have been put into operation at Hamburg and Brookhaven.
Some of the details of the operations are given to illustrate some of the more important operational
details.
It is oflen best to maximize the photon flux on the sample by employing a white beam of x
rays. The flux available at a point 9 m from a NSLS bending magnet source is shown in Figure 8.
The flux ilom this source, integrated over the entire energy spectrum, is about 3 x 108photons/s/ mz
with a 100-mA stored electron current.
For best operation of the CXRM, care must be taken in shaping the incident x-ray spectrum
using filters. The influence of the beam filters has been studied by the Hamburg group ~ochel
et
al., 1983]. Their results are shown in Figure 9. By varying the effective energy, it is possible to
tailor the beam to give the best possible MDL for a given atomic number. Filters on the detector can
be used in some cases to reduce the effect of a major element. In experiments that examine the
distribution of trace elements in bone, a filter of polyirnide can reduce the high rates caused by the
calcium in the bone, but will not have a large effect on the x-rays horn iron, copper, and zinc.
An equally critical task is-the alignment of the energy-dispersive x-ray detector. Kwiatek
et al. [1990] have reported on this phase of the optimization procedure. The importance of aligning
the detector in the horizontal plane can be seen by reference to Figure 10. In addition, the energy
spectruni and degree of polarization change as a fi.mctionof the vertical distance from the plane of
the electron orbit in the storage ring. Figure 11 shows the relative photon flux for the two
polarization states. Examination of the curves shows that the alignment needs to be made to an
accuracy of better than a few hundred micrometers to get best reduction of scattered background.
9
Resuhs of the experimental background/peak ratio determination as a fimction of vertiud
displacement are shown in Figure 12 for elements horn calcium to zinc in a gelatin matrix and for
palladium in a pyrrhotite matrix. The dependence on scattering angIe in the horizontal plane is
shown in Figure 13. The details of the methods used to make the alignments are given by Kwiatek
et al. [1990].
The Elrookhaven work has shown that it is possible to achieve spatial resolutions below the
10-micrometer range using the CXRM.
The spatial ‘~esolution of the instrument has been
demonstrated by scanning a thin-evaporated gold straightedge through the beam and recording the
intensity of the L-x rays as a fimction of distance. The resolution in this case was 3.5 pm. Some
control over the resolution for a given collimator is obtained by inclining the collimator with respect
to the incident beam.
The 13rookhaven device is routinely used for trace element measurements with a spatial
resolution of less than 10 micrometers. The sensitivity and MDL obtained under these conditions
have been reported by Jones et al. [1990b] and their results are displayed in Figures 14 and 15.
It is interesting to realize that an absolute determination of the elemental concentrations can
be made bawd either on the theoretical estimates of photon flux or from a direct determination using
an ion chamber. Kwiatek et al. [1990] point out that the expected counting rate for a given element
is given by:
(3)
where Y(EZ)I is the count rate for an element with a characteristic x-ray energy EZ,NAis the number
of target atoms within a beam spot, e~is the detector efficiency2dW4x the solid angle, N(E) the
photon flux \[nurnberof photond(s mm20.1 keV bandwidth)], w(E)the linear attenuation coefficient
for Al and p(EZ) that for polyimide, ~ the AI thickness, x~ the polyimide thickness, ofl(E) the
fluorescent cross section, E,~the energy of the absorption edge, and E the photon beam energy.
Results of a comparison of measured and calculated rates for a section of gelatin with known
amounts of iron and zinc are shown in Table 4 for three different incident beam spectra. The results
of the comparison are excellent for iron, less so for zinc. It is clear, however, that it is feasible to
make determinations of the abundances of the elements without reference to standards.
10
B.
Focussed X-Ray Microscopes
Focussed x-ray rnicroprobes (FXRM) have been the subject of great interest over the years.
The first version appears to have been developed by Sparks et al. [1980] and was used at SPEAR.
In an interesting applicationjit was used in a search for superheavy elements [Sparks et al., 1978].
A variety of different schemes have been used in the intervening years. A summary of recent
approaches at different laboratories is given hereto illustrate the directions now being taken by this
approach to production of intense x-ray beams. Early estimates of MDLs for specific configurations
have been given by Gordon [1982], Gordon and Jones [1985], and Grodzins [1983].
1. LURE (1987)
The equipment used for SRIXE at LURE has been described by Chevalier et al. [1987] and
Brissaud et al. [1989]. The white beam from the storage ring passes through a beryllium window
and is incident on a curved crystal of mosaic pyrolytic graphite. The size of the incident beam is 2
cm 1 cm before the monochromator which produces a final focussed beam of 14-keV photons with
a size of 1 mm x 1 mm. The fluorescent x-rays are detected with a Si(Li) detector collimated using
a 2.8-mm aperture. The MDLs achieved are around 1ppm for thick geological specimens and much
less for organic matrices. The probe has only been used for measurements which do not require high
spatial resolution.
2. Photon Factorv (1990)
The group working at the Photon Factory in Japan have made seminal contributions to the
development of SRIXE and of SXRM. An early instrument used Welter type focussing optics and
achieved spatial resolutions of around 10 ~m x 30 pm ~ayakawa et al., 1989]. A later version was
developed to provide higher photon flux and give improved values for the MDLs ~ayakawa et
,
al.,1990].
The design chosen was similar to one described by Jones et al. [1984] which has been
partially implemented at the NSLS. A schematic diagram of the Photon Factory apparatus is shown
in Figure 16. The platinum-coated ellipsoidal mirror was designed to produce a demagnification of
the beam of 1:9.5. The mirror accepted 0.6 mrad of beam in the horizontal direction and 0.05 rnrad
in the vertical direction and was run at a grazing angle of 7 mrad. A gain in intensity of a factor of
170 was found for a monoenergetic beam of 8 keV. The system produced a photon flux of about 1.4
11
1,
103photonsl(s. m2.rnA). The ability to locate the system in close proximity to the electron beam is
a key factor in maximizing the photon flux.
3. ~
(1990)
The instrumentation at the Daresbury SRS has been developed by a group of collaborators
from the Free University in Amsterdam and Warwick University at Coventry. The latest device uses
a silicon crystal as both a monochromator and focussing device wan Langevelde et zd, 1990~ Van
Langevelde et al., 1990b; Van Langevelde et al., 1990c]. A diagram of the instrument is shown in
Figure 17. The crystal is a 100-micrometer thick silicon crystal bent with a radius of 100 mm in the
sagittal plane and 5740 mm in the meridional plane. The beam is focussed .to a spot size of about
15 ~m x 20 pm at an energy of 15 keV. The demagnification is 1000 in the horizontal plane and of
15 in the horizontal plane with a photon flux increase of greater than 104. A flux of 15 keV photons
at the specimen of 104photons/(s- m2-mA) is obtained. This is an impressive result when it is
remembered that the device is located about 80 m from the electron beam.
4. Novosibir& (1989)
Baryshev et al. [1989] working at the Novosibirsk VEPP-3 synchrotronshave used both
monoenergetic and white beams of x rays. A single-crys~ pyrolytic graphhe monochromator was
used to produce monoenergetic beams with energies between 8 and 35 keV. The spatial resolution
was 60 pm for the monoenergetic beam and 30 pm for the white beam. The MDL for the
monoenergetic beam was 10 ppm for elements horn iron to strontium for a 1-3 s run.
5. ~
(1988)
The group at the Lawrence Berkeley Laboratory has developed a focussing system based on
the Kirkpatrick-Baez geometry [Giauque et al., 1988; Kirkpatrick and Baez, 1948; Thompson et
al.,1 987; Thompson et al., 1988; Underwood et al., 1988, Wu et al., 1990]. A schematic diagram
of their system is shown in Figure 18. The system uses a parallel beam of photons to produce an
image whichIis demagnified by about a factor of 100 to produce final images of a few micrometers.
The use of multi-layer coatings of tungsten carbide on the mirrors gives aquasi-monoenergetic beam
with a bandwidth of about 10VOat 10 keV and a high throughput. Much of the experimental work
has been carried out at the NSLS in collaboration with the BNL group. It is thus possible to make
comparisons in performance between the CXRM and FXRM on the same storage ring.
12
The LBL FXRM has produced photon fluxes of about 3 105photons/(s-mA- m2) at the 10
●
keV energy. It is important to note that the use of a final collimator is not required in this apparatus.
Improvements in the device will make it possible to reach higher x-ray energies and provide easy
tunability of the x-ray beam energy.
6. LBL at the Advanced Photon Source (ALS) (1998)
A much more elaborate imaging capability has now been developed by LBL at the ALS
[Warwick et al., 1998]. A number of different beam lines can be used to do experiments with
energies ranging from 100 eV to 15 KeV. The spatial resolutions vary from 0.1 ~m to about 2 pm
depending on the beam energy and imaging mechanism. Some of the techniques used are Fourier
transform infia-red spectroscopy (FTIR), scanning transmission x-ray microscopy, photoemission
electron microscopy, x-ray microdiffraction, and micro x-ray fluorescence. A schematic diagram
of the apparatus developed to cover the,x-ray energy range from 2- to 15-KeV is shown in Figure 19. The chosen hardware configuration uses Kirkpatrick-Baez focusing mirrors as did the earlier
system described above. Beam sizes as small as 0.8 pm have been achieved with an experimental
setup.
7. Cornell High Energy Svnchrotron Source (CHESS)
Focussing of both monoenergetic and white x-ray beams can be accomplished by total
external reflection of the x rays in tapered glass capillaries ~ilderback et al., 1994a; 1994b; 1994c;
1995a, 1995b; Hoffinan et al., 1994; Thiel et al., 1992].. The Cornell group has been one of the
leading proponents of this approach over the past ten years. It has also been actively developed at
other laboratories and installed on beam lines at several different synchrotronsfacilities. One such
installation at the ESRF is described below.
The experiments at CHESS have shown that it is
possible to produce beams with a diameter as small as 0.05 pm @offman et al., 1994]. A scan
across an edge demonstrating this spatial resolution is shown in Figure 20. The x-ray flux is
increased by factors of up to 1000. It seems fair to say that this experiment marks a major step
forward in x-ray microscopy techniques. The initial experiments showed that the capillary system
could be used for Laue di~action measurements and for x-ray radiography at this size scale.
Preliminary work done at the BNL X26A beam line showed that SRIXE experiments can be
.successfidly performed with the capillaries [Sutton, Hoffman, Bajt, Jones, Bilderback, private
communication, 1994]. SRIXE experiments are made difficult by the large divergence of the x-ray
13
beam after it emerges horn the capillary tip. In order to achieve the best spatial resolution, it is
necessary to place the sample within a few mm of the capillary and at the same time arrange optimal
detection geometry for the x rays from the sample at a 90 degree angle to the incident beam.
8. Eurouean Svnchrotron Radiation Facilitv (1994)
Themicrofocusbeam line on ID 13-1 developed by C. Riekel @ivatecomrnunication, 1995]
and collaborators is intended mainly for microdiffraction experiments. It has also been used for
SRIXE and fluorescent computed microtomography experiments. The beam line uses an undulator
x-ray source and is designed for operation between 6 and 16 keV. The optics use either silicon or
multi-layer rnonochrornators and an ellipsoidal mirror to produce a focal spot in the experimental
hutch. Collimation can be used to produce a 7-pm full-width-at-half-maximum focal spot with a
beam flux of 2 x 1011photons/sat 13keV. The silicon channel-cut monochromator can be combined
with capillary focusing optics to reduce the spot size to about 2 x 2 pm2. The photon flux is also
about 2 x 1(O11
photons/s in this conilguration. A schematic diagram of the beam line optics is shown in Figure 21. The photon flux that can be used for SRIXE experiments is around 4 orders of
magnitude higher than the values found for the NSLS X26A beam line. The expected detection
limits at the ESRF are then about 2 orders of magnitude lower. The high flux at the ESRF has made
it possible to achieve detection limits around 100 ag for transition elements [Janssens, private
communication, 1998].
9. Advanceci Photon Source (1998)
The IJniversity of Chicago Center for Advanced Radiation Sources is operating several beam
lines at the Al% with capabilities for SRIXE, computed microtomography, EXAFS,XANES, and
high pressure experiments. The SRIXE apparatus is located on an undulator beam line and uses
cryogenically-cooled Si monochromators to cover a photon energy range from 4.5-21 keV and 15-80
keV. The rnonochromator design and arrangement is such that both can be used simultaneously.
Focussing of both white and monoenergetic beams is done with two l-m long silicon KirkpatrickBaez type mirrors that reduce the beam size by factors ranging from 3 to 10. Experiments on this
beam line will be microprobe experiments to study various parameters of geological and
environmental samples including compositions of fluid inclusions and trace element partitioning.
The experimental techniques include SRIXE, EXAFWX.ANES,and computed rnicrotomography.
14
,,
1,
Summary Beam Line Desire Comments
Beam lines and experimental apparatus used for SRIXE and related experiments during the
past 25 years have been discussed here. ‘It is clear that very major developments have taken place
in the experimental capabilities during that time. The obvious foundation for the developments has
been the rapid improvements in the synchrotronsx-ray source through the introduction of the wiggler
and undulator insertion devices. However, the parallel improvement in x-ray optics for focussing
these beams and in x-ray detectors has been necessary to make possible the most effective use of
these beams for experiments. Indeed, fhture progress in the latter fields maybe the most important
factor for the actual application of these techniques in experiments.
c.
Experiments at a Distance
The effects of the remarkable increase in computer communications through the intemet and
World Wide Web are visible everywhere. In the past few years there has been a development of interest in making use of these technologies to help make facilities usefi.dto researchers at remote
locations. One example has been the DOE 2000 initiative of the US Department of Energy ~S
DOE, 1997]. There are two components of this initiative that are related to use of synchrotronsbeam
lines. First, there is the concept of national collaboratories. These are “Laboratories without walls
that unite expertise, instruments, and computers, enabling scientists to carry out cooperative research
independent of geography.” Second, is to set up pilot projects that are “virtual laboratories that give
scientists the technology to collectively observe and attack problems using combinations of ideas,
methodologies, and instrumentation that do not exist at any single location.”
Examples of this idea exist for several types of instrumentation.
An example of the
application to a synchrotronsbeam line is that of the “Spectro-Microscopy Collaborator
Advanced Light Source: A Distributed Collaborator
at the
Testbed [Agarwal et al., 1997]. This
prototype of a collaborator is supposed to enable operation of equipment on anundualtor beam line
at the ALS from the University of Wisconsin-Milwaukee. Some of the topics of concern for the
collaborator are security mechanisms, +ta dissemination, remote monitoring and control, safety
mechanisms for beam line operation, resource arbitration to decide location of control, video
conferencing and remote viewing, network needs, and shared electronic notebooks for data handling.
Another example involves the University of Florida and the APS [CAT Communicator, 1998]. This
15
‘
!,
project is supposed to allow remote operation of a materials research beam line. The computer
components needed are similar to those listed for the LBL-Wisconsin demonstration.
v.
CONTINUUM AND MONOCHROMATIC
EXCITATION
Successfid XRMs have been put into operation using both continuum and monoenergetic
synchrotron-produced x rays. The continuum radiation is extremely convenient to use since it is
easy to construct a CXRM with a minimum of equipment and achieve excellent performance.
Further, the broad-band excitation means that measurements can be made for essentially all elements
in the periodic table in a single exposure. Monoenergetic radiation can be used at an energy
optimized fcmproduction of a given element thus reducing radiation darnage in organic materials.
Counting-rate limitations in energy-dispersive detectors are reduced because of the elimination of
scattered x-my events.’ The energy can be tuned to eliminate interferences, e.g., lead-arsenic, and
to eliminate excitation of elements with Z higher than that of the element of interest. Maps can also
be constructed by subtraction of images obtained above and below absorption edges.
Successful XRMs have also been produced in collimated and focussed modes employing
either continuum or monoenergetic radiation. The brute-force type collimated continuum radiation
microprobe employed at Brookhaven and Hamburg has been comparable to the other types of probes
in terms of spatial resolution and MDL. The performance of the Brookhaven instrument has been
compared with the performance of the LBL Kirkpatrick-Baez XRM operated on the same NSLS
beam line [Giauque et al., 1988]. A comparison of results obtained with the CXRM positioned at
10 m from the source with the FXRM at 20 m from the source is given by Rivers et al. [1992].
Figure 22hows the results of the comparison. The MDL obtained with the FXRM is somewhat
better than for the CXRM, but the wider energy range of the CXRM is a substantial advantage in
some cases. The quality of the Kirkpatrick-Baez optics makes it feasible to dispense with the use
of a collimator placed close to the specimen which is an advantage in design of the experiments in
some cases.
For ultimate pefiormance in terms of the spatial resolution FXRM might be the best choice
since the construction of good collimators on the 1-pm2 size or smaller scale is difficult. However,
it is feasible to fabricate capillaries that act essentially as a collimator. In that case the beam
16
divergence is reduced and the capillary can produce a small beam without the need to place the
sample close to its end.
The challenge is thus to:
1. obtain better optics to improve the spatial resolution and
2. improve the efllciency of the optics to get higher photon fluxes.
These goals can be addressed most effectively by use of focusing optics. An alternative to
the capillary approach has been demonstrated by Padmore et al. [1997] who used an elliptical
focussing mirror and Kirkpatrick-Baez mirrors to obtain a white beam with a size of 0.8 ~m. SRIXE
systems with sub-micrometer spatial resolution are not now in use on a routine basis. However, it
appears that the combination of improving x-ray optics and the high flux x-ray beams produced by
third-generation synchrotronsundulatory will bring this advance about in the next few years.
VI.
QUANTITATION
Methods for making quantitative elemental determinations using x-ray fluorescence have
been developed over many years. These approaches are discussed for conventional x-ray sources
in Chapters 5 and 6. Some of the approaches used at synchrotronssources are given here to show
how the methods developed for use with conventional tubes have been used with the new radiation
source.
Giauque et al. [1986] measured several US National Bureau of Standards (NBS), now
National Institute of Standards and Technology (NIST), and the Japan National Institute of
Environmental Science (NIES) standard reference materials (SRM) at the 54-pole wiggler beam line
at the Stanford SynchrotronsRadiation Laboratory. They referred their measurements to a copper
standard prepared.by evaporation where the weight was determined from gravirnetric measurements.
Multi-element standards were prepared by dissolving known weights of an element in an acidic
solution. Monoenergetic radiation was used for the work at energies of 10 and 18 keV. The beam
size was defined by an aperture 3 mm in size. The differences in sample mass were accounted for
by normalizing to the intensity of the scattered radiation [Giauque et al. 1986]. The results obtained
for three NBS materials are shown in Table 5. The table also includes the ratio between the NBS
value and the value obtained with SRIXE to demonstrate that there is excellent agreement.
17
I
The work ofGiauqueetal.[1986]
addressed quantitation in thin biological specimens where
I
matrix effects were negligible and where the Compton scattering could be used for normalization
I
are many geological experiments where this situation holds. Methods for quantitation have been
I
I
I
I
I
I
I
I
of masses. It is necessary to extend this approach if thick specimens are to be investigated. There
discussed by Brissaud et al. [1!989]and by Lu et al. [1989]. Corrections are made for attenuation of
incoming and fluorescent x rays by the sample matrix and by any filters employed as well as for
secondary fluorescence. The composition of the major elements is generally known for geological
materials; hence, concentrations can be referred to that of one of the major elements.
Brissaud et al. [1989] compared the SRIXE results with several standards. The results are
shown in Table 6. The table gives the recommended concentration, the SRIXE value and the ratio
of the two.
It can be seen that agreement is quite good.
comparatively large 1-*
In this case, as noted before, a
beam was used. Lu et al. [1989] used a microprobe with a beam size of
30pm x 60 Ipmto analyze different specimens of feldspars. They compared their SRIXE results
with values c~btainedusing an electron probe and atomic absorption spectroscopy. The agreement
with the electron probe was good, but the atomic absorption values tended to be systematically lower
than the SRIXE values. Results for the comparison with the electron probe are shown for iron and
strontium in Figure 23.
For thin biological specimens, concentrations can be established by use of a sensitivity curve
such as the cme displayed in Figure 17. Corrections for differences in thickness can be made by
normalizing to the intensity of the scattered incoherent peak if a monoenergetic beam is used or to
the continuum comprising both incoherent and coherent scattering if white beam is used [Giauque
et al., 1979].
Quantit.ation of SRIXE draws on many years of experience gained using tube-excited x-ray
fluorescence. The main difference between the x-ray sources occurs when SRIXE is used with
microbeams with dimensions on the micrometer scale. In this case strict attention must be given to
the uniformity of the standards used and to the experience gained in calibration of the electron
microprobe [men et al., 1979] and proton microprobe [Johansson and Campbell, 1988].
18
,,
●’
VII.
SENSITIVITIES
AND MINIMUM DETECTION LIMITS
The related questions of sensitivities and MDLs have been addressed by calculations based
on the known physical parameters of the XRM systems and by empirical determinations. The
detailed understanding of the x-ray production process using synchrotronsradiation is helpfid in
assessing the sensitivities and MDLs that can be achieved in SXRM. The results cited for the
sensitivities and MDLs should be taken as representative of the current situation. The actual values
depend on the particular experimental conditions, synchrotronsring currents, spatial resolutions, etc.
so that exact comparisons are not terribly meaningful.
Sparks [1980], Gordon [1982], Gordon and Jones [1985], and Grodzins [1983] have
presented calculations of the minimum detection limits to be expected using the second-generation
synchrotronsradiation sources such as the NSLS to produce x-ray microbeams. Many experimental
determinations have been made for the two quantities. The results obtained by different groups
using different types of XRMs are given here.
Figures 14 and 15 show the values obtained by Jones et al. [1990a] using the collimated
microprobe at the NSLS. Their values are relevant to thin specimens with an organic matrix. Values
for the calculated MDLs extrapolated from the work of Gordon and Jones [1985] are included in
Figure 15. It can be seen that the experimental values are in the same range as the calculations
although the experimental system was not the same as the one considered theoretically.
The
sensitivities that are shown in Figure 15 show the agreement between predicted (solid curve) and
measured (data points) counting rates.
The curves were calculated fi-om knowledge of the
experimental arrangement, synchrotrons energy spectrum, and specimen parameters.
The
experimental points show the measured values assuming the same specimen parameters.
Figure 24 shows the results obtained for the MDLs by Giauque et al. [1988] using the
Kirkpatrick-Baez XRM at the NSLS. The sensitivity curve has been previously discussed in
comparing sensitivities for the collimated and focussed instruments (see Figure 22). Jones et al.
[1988] compared the MDL values for the collimated white beam and monoenergetic focussedbeam
approaches and showed that the MDL values were very similar. Ketelsen et al. [1986] have also
compared MDL values for white beams and monoenergetic beams. The results are shown in Figure
25 and also demonstrate that the two approaches give comparable results. The absolute values of the
19
,
two experiments cannot be directly compared because of differences in the beam size and ring
operating conditions.
Few experiments appear to have been devoted to measuring detection limits at the thirdgeneration sources. Results from the ESRF have been discussed in comparison to the values found
at the NSLS in the survey of beam lines given in Section IV-B. Improvements in both spatial
resolution and./or detection limits can be generally assumed. These improvements should be by
about a factor of 100 when comparing use of undulator beam lines at the new facilities with their
bending magnet performance.
VIII. BEAM-INDUCED DAMAGE
Passage of a photon beam through a specimen results in energy deposition through the
photoelectric effect and Compton scattering processes. This energy deposition results in a breaking of molecular bonds as the secondary electrons produced lose their energy by further ionization or
scattering prc~cesseswith other atoms. These effects have been examined in great detail for electron
beams used in electron microscopy. Much less has been done for the XRM. This is not surprising
since the field is much younger and less extensively developed than is the field of electron
microscopy. Another important reason is that the photon-beam fluences employed thus far have
been substantially smaller than those employed in the electron microscope and as a result the
magnitude of beam-induced damage has not been so important.
Biological and other orgtic
materials are generally more susceptible to beam-induced
darnage than are other materials. For this reason the discussion is limited to these materials. In the
future, when more intense synchrotronsx-ray sources are available, it will be necessary to expand the
list of materials considered.
A qualitative calculation can be done to illustrate the energy deposition for various energy
photon beams. Assume that the photon flux used for a typical fluorescence experiment is 10G
photons/(s-pm2). Typical run times are 10 minutes or less. The maximum photon fluence for
current XRM instruments is thus about 6 108photons/( mz). It is interesting to note that these
●
fluences are now starting to approach the range of the fluences found in use of the electron
microprobe. In order to estimate the energy deposition, it is assumed that attenuation coefficients
can be represented by the photoelectric process attenuation coefficient only and that all of the energy
20
(,
of the photoelectrons is absorbed in the volume considered. The Compton scattering process is
relatively small and can be neglected in a qualitative estimate of the dose. The results of the
calculation are shown for photon beams with energies less than 20 keV.
The x-ray dose needed to kill living biological systems has been examined in detail over the
years ~
and Sayre, 1980; Sayreetal.,1978; Sayre et al., 1977; Slatkin et al., 1984; Themner et
al., 1990]. A dose which exceeds”1 Gy is likely to cause serious damage to a biological cell or
system.
Thus, there are limitations to the use of x-ray beams for the examination of living systems,
as there are, of course, for W other beams. The limitations depend on the absorbed dose. One way
that this can be done is by measuring the linear attenuation coei%cients in CMT or in projection
radiography type experiments. Spanne [1989] has calculated the dose given to an object in obtaining
an image with a signai-to-noise ratio of 5 for a water phantom with a contrasting detail with a
diameter 0.005 of the phantom diameter. The results of Spanne for objects with different diameters and compositions are shown in Figures 26a and 26b. It can be seen that the dose is strongly energy
dependent and that for relatively small objects (1 mm), the observed dose at the optimum energy is
about 102 Gy. Examination of living systems using CMT with resolutions on a level of 5-10
micrometers thus may not be practicable. However, relaxation of the resolution criterion would
reduce the dose to a point where in-vivo examination of living systems is f&sible. It can also be
seen from Figure 26 that the optimum energy for examining very thin specimens (cells) becomes
very low. This represents a separate field of research and is not considered here. The use of
projection radiography methods should also be usefid. The dose is much less since fewer photons
are needed to form an image. In this case the examination of living systems will be easier.
If x-ray beams are used to make fluorescence measurements onnon-living systems, aloss-ofmass and possibly of trace elements can occur. Slatkin et al. [1984] investigated changes in the
morphology of human leukocytes and showed that severe damage resulted for fluences of 15-keV
photons of about 1017to 10*9photons/cm 2. Figure 27 shows a photomicrograph of the leukocytes
after bombardment by fluences from 0.4 to 2.4 x 109photons/cm2. Damage to thin sections of
kidney (10 ~m) was much less. Mass loss in other types of organic materials was measured by
Themner et al. [1990]. They determined mass loss by measuring the change in scattered radiation
counting rate as a function of total dose. The results that they obtained for irradiation of a skin
21
.
sample is shown in Figure 28. It can be seen that changes can be observed at dose wdues
comparable to those used in many fluorescence type experiments. Mass loss, therefore, should be
carefi.dlymeasured if the scattered radiation is used as a measure of the specimen areal density for
quantitation purposes.
The effects discussed can lead to loss of trace elements and to errors in the assignment of
concentrations, as is weIl known from the case of electron or proton microprobe.
Just as in those
cases, measurements need to be made of the yield of characteristic x rays as a fimction of photon
fluence for fluorescence measurements. One such examination shows the yield of chromium x rays
observed in the bombardment of a section of rat kidney which was investigated in a study of the
toxic effects of chromium. It can be seen that there are no indications of loss of the chromium under
the conditions used.
Diffraction experiments are possibly one type of experiment that will be very sensitive to
radiation damage effkcts. Protein crystallography experiments at the NSLS [Sweet et al., 1995]
showed that focussed bending magnet radiation from the NSLS was sul%cient to render samples
useless in a short period of time. Difllaction of 13 keV x rays by thin sections of polyethylene were
studied at the ESRF microfocus beam line. In this case the diflkaction patterns were essentially
destroyed in 30 to 60s [Jones, Engstrom, Riekel, Connor, private communication, 1995].
In summary, while beam-induced damage can be observed, it does not seem to have been a
problem in experiments conducted to date for experiments that are not sensitive to crystal structure.
However, since much higher fluences are to be expected as focussing methods are improved and as
the synchrotiron source itself becomes more powerfid, the beam darnage effects will become of more
central importance. Further studies of these effects should be carried out.
IX
APPLICATIONS
OF SRIXE
There are now examples of the uses of the XRM in many different fields. A brief description
of the diverse applications is given hereto illustrate the rapid development of the field and to give
an idea of the ways the XRM may be used in the fiture. The examples presented cover mainly
activities during the past 10 years. The total number of experiments performed during this time is
22
,x
.’
,,
very large. The intent is to give an example of work covering a variety of topics. There is no intent
to give an encyclopedic review of all applications.
A.
Archaeology
Brissaud et al. [1989] made low-resolution measurements on a number of different materials.
One case involved examination of three Gallic coins using synchrotronsradiation and a comparison
with the results obtained using proton-induced x-ray emission and neutron activation analysis. The
x-ray and proton beams probe material close to the surfhce of the coin, while the activation approach
gives the bulk concentration. The results given in Figure 29 show differences in the concentrations
found with these methods and also show that the activation approach is not applicable to all
elements. This straightforward example shows that the synchrotronscan be used to good effect in
studying archaeological and other materials with spatial resolutions of the order of 1 mm. It is,
however, a type of experiment that should be viewed as a bridge between the use of conventional x-ray tube sources and the synchrotronssources with high brilliance.
B.
Biology/Medical: Calcified Tissue Studies
The distribution of trace elements in bone and other calcified tissues is generally of great
interest since the concentrations of the essential trace elements are relevant to bone growth and
disease. Therapeutic agents used to treat disease states may modifi the trace element concentrations
and will deposit in particular patterns in the.tissue themselves; and, ftily,
toxic elements such as
lead are stored in the bone in localized patterns. Several experiments can be cited to illustrate these
points.
Tros et al. [1990] have examined neonatal hamster tooth germs using both microPIXE and
the XRM approach.
In this work neonatal hamsters were treated with a fluoride compound
commonly used to inhibit tooth decay. A tooth was obtained from the hamster after a one-day
interval and thin sections were measured using the two types of beams. The fluorine concentrations
were found using the proton microprobe at Amsterdam and the zinc distribution by -
at
Daresbury. The XR.Mwas chosen for the determination of the essential trace elements because of
its high sensitivity and low beam-induced damage. The proton beam was suited to the determination
of the fluorine since the cross sections for inelastic proton scattering are high. The combined use
of the two methods shows that ion beams can be used effectively in combination with the photon
23
.
Galliumnitratei sanother therapeutic elementthatisused totreataccelerated boneresorption
found in cancer patients. The mechanisms by which gallium interacts with the bone are as yet poorly
understood. Beckman et al. [1990] used the NSLS X26 XRM to study sections of rat tibia obtained
from animals treated with gallium nitrate. A scan across the tibia from periosteurn to endosteum
giving the distribution of C% G% and Zn in a 12- m thick section of bone is shown in Figure 31 (a).
A two-dimensional map of the gallium and calcium distributions in a fetal rat ulna bone that was
exposed to gallium (25 uM) in culture medium for 48 hours is shown in Figure 31 (b). The bone
structure is Shown in an electron micrograph in the center section of the figure. The non-calcified
portion accumulates little gallium compared to the calcified portion where the gallium accumulates
in the metabolically active regions where new bone matrix is being formed. The method has been
used to study the kinetics of gallium absorption in the bone for different doses and for different states
of the bone metabolism. Changes in the concentrations of iron and zinc as a fi.mctionof the gallium
treatment indicate that it may be possible to infer particular enzymes, which are targets for the gallium.
Lead can be used to show the uses of the XRM in the study of the effects of toxic metals.
Lead is a m?jor public health problem in many countries. Most of the lead in the body is stored in
the skeleton and can be released to cause serious health effects under certain conditions. It is known
that it causes neurological and other problems in children and it is associated with @dney and
cardiovascular disease in adults. Understanding the deposition patterns and kinetics of lead in bone
are therefore of great importance from a practical stand point. Jones et al. [1990b, 1992] have
reported mea~urements on lead distributions in the human tibia and in sections of deciduous teeth.
Figure 32 shows the distribution of lead in such a tooth. The end-objective in this case will be to
attempt to ccmelate the distributions with the blood-lead concentrations at birth from examination
of the enamel (formed by the time of birth) and the dentine (later exposures). Knowledge of the
time-integrated lead exposure can then be related to neurological deficits and other effects.
An initial experiment to investigate the kinetics for the accumulation and resorption of lead
in bone was mrried out in a controlled experiment that investigated chick tibia [Jones et al., 1997].
The measurements were performed the proximal end of thin (6Opm) sections of tibia. Chicks raised
on diets that contained normal amounts of Ca and Ca-deficient diets and lead were analyzed. Maps
over the bone were made with a step-size between points of 60 pm. Analysis of these maps showed
24
,!
.
,,
on diets that contained normal amounts of Ca and Ca-deficient diets and lead were analyzed. Maps
over the bone were made with a step-size between points of 60 pm. ArmJysis”ofthese maps showed
that Pb and Ca depositions in bone are similar and that the effects on bone growth and mineralization
caused by Pb influences the distribution of both cations. This experiment shows that SRIXE can be
used as a valuable complement to more standard techniques used in biomedical investigations.
The effect of lead on the developing brain has been investigated by Cline et al. [1996] and
Jones et al. [1997]. They studied the retinotectal system of frog tadpoles as a Iinction of exposure
to various levels of lead during growth; The lead content in the optic tectum of the exposed tadpoles
was measured using SRIXE at the NSLS X26 beam line with white light and a collimated beam.
Lead levels of about 200 ppb were successfidly detected. Use of SRIXE was advantageous because
it can be used with the very small amounts of sample material available. Atomic absorption
spectroscopy could not be used for that reason. The results of the overall experiment demonstrated
the impact of lead on the developing brain through the structure of the retinal axons. Further, work
with the chelating agent, 2,3-dimercaptosuccinic acid, showed that the effect of lead on the neuronal
structure could be reversed at a level superior to removal of the lead source alone. These conclusions
are relevant to lead-induced cognitive deficits in humans. The results demonstrate that SRIXE can
be a usefid measurement technique in many different types of biomedical applications where it is
necessary to measure trace concentrations of an element in thin tissue sections where sample size
or handling preclude other conventional approaches. Lower detection limits at undulator beam lines
will increase the scope of this type of analytical experiment.
Osteo-arthritic degeneration of articular cartilage is a degenerative disease of major
importance which can be brought about by senescence and trauma. It causes changes in the cartilage
and in the bone. The metacarpal joints of horses affected by osteoarthritis can be used as an animal
model for study of the disease. Rizzo et al. [1995] have studied 5-pm thick cartilage slices obtained
from normal and arthritic horses using x-ray microbeams (10pm size) at the X26A beam line at the
NSLS to map the spatial distribution of S, Ca, and Zn along a line through the section. It was found
that there were differences in the results for the normal and arthritic tissue for all three elements. It
is argued that the zinc maybe contained in the metalloenzyrne alkaline phosphatase. This result may
be of use in understanding the metabolic processes of importance in osteoarthritis.
25
-
c.
Geology and Environmental
Sciences
The XRM has been used for many different types of geological and environmental
experiments. A survey of geological and environmental applications of synchrotronsradiation has
been given by several authors [Smith, 1995; Jones, 1999]. A few examples of different applications
are given here.
Scientific questions related to the environment are generally very complex and require
understanding on a number of spatial scales. Synchrotronsradiation analytical techniques can be used
to effectively investigate many of these problems. This has been recognized explicitly at the ALS
[Robinson, 1997] where they have considered how to integrate synchrot.ron techniques with
environmental projects. ‘A particular virtue of photon use is ease of analyzing wet samples as
compared with electron spectroscopy. It was suggested that an appropriate name for the field would
be Molecular Environmental Science (~S)
in the Soil X-ray Region although it could be argued
that the use of “molecular” is too restrictive. In any case, some of the experiments described here
fit well into such a conceptual framework.
One topic that has been the subject of many investigations has been the uptake of metals by
plants and trees from contaminated soils and water. Experiments have recently been can-ied out at
the ESRF and at the NSLS which are relevant to this topic. The ESRF experiments were carried out
using high-resolution beams formed by capillary focussing to determine wood density and elemental
composition. Figure 33 shows results obtained for the distribution of Mn and the density in the
region analyzed. The area examined is at the boundary between winter and onset of growth in the
spring. Measurements at the cellular level make it possible to examine pathways for accumulation
of toxic elements from the environment. The experiments at the NSLS have been airned at looking
at accumulations of metals over a long-term period at trees growing in close proximity to a metal
smelter ~artin
et al., 1998]. The focus of this work was to attempt to determine the relationship
between the time varying environmental conditions and the composition of the wood. The results
obtained using SRIXE and SIMS to examine thick wood specimens showed that the distributions
were very heterogeneous and that growth cycles were not easily distinguished from the trace element
concentrations.
In geology, many of the experiments used thin seetions of rocks for studies of specific types
of minerals in a heterogeneous matrix. Examples of this type of application are the study of zoned
26
carbonate gangue cements found in Tennessee [Kopp et al., 1990]. The XRM measurements were
used in an attempt to interpret the effects of trace amounts of Mn and Fe on cathodoluminescence
in carbonates. Further, the variations in the trace element concentrations should aid in deducing the
source and direction of flow of fluids responsible for the formation of dolomites.
Small particles can also be effectively examined using XRM. Sutton and Flynn [1988] and
Flynn and Sutton [1990] have carried out a series of experiments dealing with analyses of
extraterrestrial particles. These analyses were carried out on piyticles of sizes less than 100 pm.
Minimum detection limits less than 10ppm were obtained for particles less than 20 ~m in size using
the BNL X26 XRM. Tuniz et al. [1991] also used this equipment for examination of fly ash taken
from different types of power plants and incinerators. They measured elemental composition in
individual fly-ash particles with sizes down to a few pm and also made two-dimensional maps of
the distributions of the elements in 10-~m thick sections of particles produced by a lapping
technique. The results may be useful in verifying models for production of toxic compounds in
incinerators based on the presence of specific metals in the ash ~arasek and Dickson, 1987;
Hagenmaler et al., 1987; Altwicker et al., 1990].
D.
Materials and Chemical Sciences
r
Fluorescence can also be used to advantage in the material sciences. For example, Isaacs et
al. [1991] have studied the concentration gradients produced in a solution during the localized
corrosion of stainless steels. The combination of high spatial resolution and excellent detection
sensitivity enabled them to study the variation in the nickel concentration above a stainless steel
surface immersed in a bulk chloride electrode. Figure 34 shows the electrochemical cell used in the
work and the observed variation of nickel concentration above the stainless steel surface. From
study of the concentration gradients, Isaacs et al. [1991] were able to identifi effects arising from
silicon in the steel. In-situ studies of kinetic effects should be of increasing interest, not only in
corrosion measurements, but also for other types of chemical reactions.
SRIXE can also be applied to the measurement of relevant to supported catalysts.
In
particular, it can be used to measure the distribution of Cr catalyst distributions in polyethylene
polymerization particles [Jones et al., 1997]. Th< use of SRIXE makes it possible to study the
catalyst distribution at high yields which are not possible by other methods. The experiment was
done on microtomed sections of polyethylene particles using the NSLS X26A XRM. A map of the
27
<
Cr distributicm is shown in Figure 35. Sharp peaks observed at the periphery of the particle are
consistent tikh observations made using computed microtomography.
The presence of rather
uniform concentration of catalyst through the particle has not been observed previously.
The
observation shows that it is possible to investigate the Iiagmentation of the catalyst during the
polymerization process by making sequential measurements as the polymerization process evolves.
Use of higher spatial resolution and improved detection limits feasible with new XRM equipment
will make it possible to substantially improve this type of investigation.
x.
Tomography
Computed microtomography (CMT) is an important approach to nondestructive analysis and
has been extensively developed using conventional x-ray sources. The synchrotronssource gives
substantial advantages because of its high brilliance and continuous x-ray spectrum. The superior
properties of the synchrotronssource have led to CMT instrumentation capable of superior spatial
resolution and shorter data acquisition times.
Almctst all the work that has been done has concentrated on the use of CMT in the
attenuation mode where determinations are made of the linear attenuation coefficients.
The
*
absorption mode is a simple and effective approach to CMT imaging. It can be used on samples
with a varie~y of sizes and attenuation coefficients by choosing the appropriate x-ray energy. The
use of SRIXE is more restricted since the specimen must be small enough to allow the escape of the
characteristic x ray of interest. However, as is the case with EXAFS and XANES, the absorption
and emission (SR.IXE) approaches are complementary and both need to be available as part of the
basic XRM.
Refmlements to the method are necessary in order to get better information on the elemental
composition of the materials. The absorption approach can be refined by producing tomograms
above and below an elemental x-ray absorption edge. Subtraction of the tomograrns gives the
concentration of that element. The difference technique is valuable for study of major and minor
elements to the O.10/0level.
The imaging of trace elements often necessitates the use of SRIXE so that specific elements
are selected through the detection of their characteristic x rays. The detection system must be chosen
for high efficiency and high counting-rate capability.
28
The type of specimen which can be
J,
investigated is limited by the attenuation of the fluorescent x rays in the sample being investigated.
The sample dimension is strongly constrained because of this.
Several groups have done CMT work with synchrotrons. Flannery et al. [1987] developed
a third-generation type system at the NSLS. They used a x-ray magnification system with a
scintillation detector coupled to an image intensifier to produce images with a claimed resolution
down to 1 pm. A similar approach was used by Bonse et al. [1986] and Kinney et al.[1988] at
SSRL. Spanne and Rivers [1987] demonstrated a first-generation system at the NSLS X26 beam
line. Later work with the apparatus has produced images with a spatial resolution of 1 pm x 1 pm
and a slice thickness of 5 pm quite comparable to the results from the third-generation devices.
More recently, a third generation apparatus has also been put into operation at BNL ~owd et al.,
1998]. It can be used for analyses with voxel sizes as small as 2.7 pm3.
The f~st-generation approach takes longer to produce images, but has the advantage that
beam scattering effects in the sample are eliminated and SRIXE measurements can be performed to
produce elemental maps. In the third-generation systems, elemental maps are made by subtracting
images taken above and below the absorption edge of interest. The MDL for such an approach is
about O.10/O.
Spanne [1990] has carried out a pilot study with the aim of evaluating the potential for
mapping of light elements at the cellular level in the rat sciatic nerve using fluorescence CMT. A
comparison of the mean free path for characteristic x rays from potassium and typical sciatic nerve
sizes shows that it is feasible to make corrections for the attenuation of the potassium K-x rays in
the nerve. The computed emission tomogram of the distribution of potassium in the epineurium of
a rat sciatic nerve given in Figure 36 illustrates this point. Note that the short escape depth for the
potassium x rays that are observed necessitates special measures during the reconstruction of the
image. Saubermann @rivate communciation, 1989] points out that fluorescence CMT makes
possible studies of elemental distributions in unsectioned samples. Examination of unsectioned
samples also makes feasible in-vitro analysis of sections of nerves several millimeters long.
Longitudinal distributions of elements can then be conveniently studied by scanning at different
heights, and it is even possible to go back to a previously mapped region for a more detailed
examination if steep concentration gradients are discovered. Longitudinal concentration gradients
29
!,
have been demonstrated in the rat sciatic nerve, although with a very poor longitudinal resolution,
and may be clf significance in nerve injury ~oPachin et al., 1988; LoPachin et al., 1990].
EXXOID[Auzerais
et al., 1996] and BNL [Spanne et al., 1994, Coles et al., 1996; Coles et al.,
1998a, b] groups have investigated topics related to the microgeometry of sandstones in parallel
experiments
id
the NSLS. In addition to measuring properties of the sandstones such as porosity,
permeability, tortuosity, and connectivity attention was given to the dkplacement of oil by water and
modeling of flow through the experimentally-determined structures. A structure of Fontainebleau
sandstone measured in the experiment of Spanne et al. [1994] is shown in Figure 37. The structure
was analyzedl to give a theoretical representation of the structure that could be used for predicting
fluid-flow at larger size scales.
Feng, et al. [1999] have exarninedthe structure of micrometeorites using the BNL equipment.
Previous examinations of the particles have shown that there are void spaces and small nuggets of”
Pt in the interior of some of these particles. This information was gained by laborious sequential
sectioning of particles with a diameter of about 400 pm. The use of CMT to gain the same
.
itiormation represents a major step forward in analytical technique for the field. A representation
of a Pt nugget observed in a micrometeorite is displayed in Figure 38. The results show that it will
be possible tc~systematically examine large numbers of specimens to obtain significant information
about the nuggets and about the history of the particle as it passes through the earth’s atmosphere.
Several groups at the ESRF and NSLS have been concerned with the development of
improved methods for obtaining contrast between materials with similar x-ray absorption
coefficients [see, for example, Raven et al., 1997]. Bufflere et al. [1998] have applied the phase
contrast technique, based on diffraction patterns produced at discontinuities in the sample, to
measure damage in metal matrix composites. This was an ingenious experiment that investigated
the development of cracks in the composite as a fi.mctionof tensile forces applied to the specimen
in situ. The experiment revealed the cracking of SiC particles, decohesion between sample phases,
and propagation of pores.
The development of CMT at synchrotrons has been very rapid over the past ten years from
the standpoint of instrumentation. Applications have also been developed, but perhaps at a slower
pace. This will change in the fiture with an increase in the availability of the technique and
improvements that can be made with the increasing use of third-generation synchrotronsundulatory.
30
,,
It is apparent that both absorption and emission approaches are needed and that both approaches are
required at any synchrotronsXRM facility. The emission technique will be greatly assisted by the
use of undulator beam lines to obtain higher counting rates.
XI.
EXAFS AND XANES
EXAFS and the related XANES have been widely applied to give information on the
chemical state of elements in many different materials [Winick and Doniach, 1980; Koningsberger
and Prinz, 1988]. Many of the experiments that have been carried out have used relatively large xray beams and thick specimens to make absorption measurements on a time-scale of minutes
possible. This approach is not usefid for elemental concentrations less than about 0.1 to 1%. The
use of EXAFS and XANES at lower concentrations can be achieved by use of fluorescent x-ray
detection. Cramer et al. [1988] have developed a 13-element Si(Li) x-ray detector which gives a
high effective detection efllciency for this type of application. Other workers have deveioped means
of acquiring EXAFS spectra on a ins-time scale [Tolentino et al., 1990] and have made
measurements of chemical state on a finer spatial scale [Iida et al., 1989]. The fbture development
of x-ray microscopy can thus be seen to include the use of SRIXE with EXAFS and XANES and
development of new techniques to make it possible to work with improved MDLs and beam sizes
at the micrometer level.
The NSLS X26 group has employed a simple channel-cut silicon monochromator with an
energy resolution of about 1.1 eV for several demonstration XANES experiments. The beam was
first defined with a four-jawed aperture whose size could be adjusted using computer-controlled
stepping motor drivers. This was followed by the monochromator placed about 10 cm upstream
fkomthe target. The beam moved vertically on the target by about 60 cm during the scan. This was
not important for the resolutions used, but could be easily compensated for by a correlated motion
of the target to keep the same spot under the beam. The frosttest used a thick specimen of NIST
SRM 1570 spinach leaves, which contained 550 ppm iron. The iron x rays were detected with the
XRM equipment described above. The work was done with a beam size of 2 mm2. The results of
a scan of a pure iron specimen (beam size 200 pm x 200 pm) and the spinach leaves are shown in
Figure 39. The spectra agree well with the results of a similar scan done on the NSLS X19 beam
line using the 13-element Si(Li) detector. Extrapolation from these initial values showed that work
31
,
with beam sizes down to 100pm and perhaps lower was feasible with the existing equipment for this
particular target.
A more stringent demonstration is the use of thinner specimens.
For this purpose
measurements were made on tie chromium contained in olivine and pyroxene components in a 30ym section of lunar mare basalt 15555 from Apollo 15 and in a 10-~m section of a rat kidney. The
lunar basalt study was undertaken since the oxidation state of the chromium could shed light on
conditions existing at the time of the formation of the mineral studied [Sutton et al., 1991]. The rat
kidney measurement was needed to cast light on nephrotoxic effects resulting horn environmental
exposures. It is hypothesized that the oxidation state of the chromium changed during its passage
horn lungs to kidneys with related implications for health effects.
The basalt specimen contained chromium at a level of about 1000 ppm and the kidney at a
level of about 50 ppm. Figure 40 shows the XANES spectrum for chromium in pyroxene and
200 pm. Figure 41
olivine contained in the lunar basalt taken with a beam resolution of 200 pm
x
shows the spectrum obtained for the rat kidney, but in this case with a 1 mm
1 mm beam size. The
x
beam was positioned over the medulla portion of the kidney. The XANES spectra show that
chromium exists primarily in the 2+and 3+in the lunar olivines but as the 3+state in the kidney.
The application of SRIXE to EXAFS and XANES experiments has expanded greatly on the
X26 beam line since these initial experiments. This has been especially true in the geological and
environmental fields. Delaney and co-workers have continued their examination of both geological
and extra-tenestrkd materials ~elaney et al., 1996, 1998a, 1998b; Flynn and Sutton, 1990]. A
number of experiments have been performed that relate to speciatiob of contaminants in soils and
sediments [Elertsch et al., 1994; Bajt et al., 1993; Sutton et al., 1994; Tokunaga et al., 1997, 1998].
Ilman et al. ~rivate comrnunicatioz 1996] have investigated the chemical state of Cr and As in
wood treatecl with chromated-copper-arsenate and the role of Mn in fungal diseases of wood. The
work, taken as a whole, serves as a an actual demonstration of the concept mentioned in Section IX
for definition of a field of Molecular Environmental Science ~obinson, 1997].
There are several examples of the way that these experiments can be extended at third
generation synchrotronsfacilities. Sarret et al. [1998] used a bending magnet beam line at the ESRF
to do experiments with spatial resolution of 300 pm
x
300 pm for examination of trace elements at
concentrations as low as 100 ppm. These parameters are similar to those at the NSLS X26 beam
32
,,
.,
.*
line. Measurements were made to study uptake mechanisms of Pb and Zn in lichens. Lichens were
of interest since they are commonly used as a biomonitor to assess environmental pollution. The
analysis of the EXAFS spectra obtained finm the lichens and comparison with reference compounds.
Analysis of these spectra made it possible to reach an understanding of how Pb and Zn are
transported and absorbed in the plant.
A demonstration experiment similar in intent to the work at the NSLS and ESRF has been
performed at the ALS ~arwick
et al., 1998]. In this example measurements were made with a
much higher spatial resolution. The oxidation state of Cr in a soil sample was determined by first
scanning a region of 60 pm x 80 pm. One region of high Cr concentration was found and chosen
for examination with XANES. The spectrum obtained from this region with an area of a few (pm)z
indicated that the Cr was in both Cr(III) and Cr(VI) oxidation states. The results of this experiment
are shown in Figure 43:
Another example is given by the work of Cai et al. [1998] at the APS. They describe a beam
line using an undulator source and zone plate focussing to obtain a beam size of less than 0.25 pm
and a photon flux of 5 x 1010photons/s at a bandwidth of 0.010/O. It is possible to do x-ray
difilaction, x-ray fluorescence, and computed microtomography simultaneously on this beam line.
Data is presented for spatial distributions of Mn, Fe, Cu, and Zn in a root, hydrated P. lanceolata,
infected by a fhngus, mycorrhizal fungus G. mosseae. A XANES spectrum for Mn obtained from
a location in the root at a spatial resolution of 1 ~m x 3 ym is shown in Figure 42.
The examples of work that has been done at the NSLS, ESRF, ALS, and APS on soils and
metal transport in plants during the past decade illustrates the usefulness of the synchrotronsXRM
for geological and environmental work. A similar collection of experiments can be assembled to
illustrate other applications of SIUXE/XANES/EXAFS using both second- and third-generation
facilities. It is clear that the scientific applications of these techniques will increase dramatically in
the future.
XII.
FUTURE DIRECTIONS
What does the fi.dmrehold for synchrotronsXRM? Several different developments can be
predicted for the fnst decade of the next millenium, 1999-2010, with some confidence.
33
,,
First, continuation of the present types of experiments will go on at an expanded rate as the
virtues of the method become better known. The applications will benefit fkomenhancements in the
facilities which will bring spatial resolutions to 1 ym2 for two-dimensional maps or tol ~m3 for
CMT. Relatively minor improvements to the existing systems are needed to achieve this level of
resolution. MDL values should also improve somewhat with the introduction of improvements in
the x-ray detector systems--again using known techniques. There will be a further merging of XRM
techniques with those developed for EXAFS and XANES.
Second, development of new focussing methods will begin and this will lead to the routine
.achievement of spatial resolutions of the order of 0.1 pm ky 2000-2005. An example is the
production of zone plates suitable for focussing of 8-keV ~ rays by Bionta et al. [1990] and the
application of capillary focussing by Bilderback et al. [1994].
Third., anew era in SRIXE research has been signaled by the operation of third-generation
synchrotronssources such as the Advanced Photon Source at Argonne National Laboratory and the
European Synchrotrons Radiation Facility at Grenoble.
These systems produce beams with
improvements of three to four orders of magnitude when using unduhutorinsertion devices and with
higher energies when using either bending magnets or wigglers. A general discussion of the impact
of these new rings on x-ray microscopy has been given by Sparks and Ice [1990], who found
detection limits of 108atoms in a solid and of 103free atoms for fluorescence detection with a l-~m2
beam and a data acquisition time of 1 s. An early discussion of the impact of the APS on the
geosciences Twasgiven by Sutton et al. [1988]. It remains to be seen how close these predictions will
approach reality.
It is presently clear that SRIXE is now a major anlaytical method for the
geological and environmental sciences and possibly to a lesser extent for other scientific fields.
Figure 44 summarizes the time dependence for the synchrotronsXRM spatial resolution. The
area resolution has decreased exponentially over the past fifteen years. Extrapolating into the future
assuming the same rate of improvement leads to an estimate of an area resolution corresponding to
a beam size of roughly 1 nm by the year 2000. Looking at what has been accomplished for
focussing low-energy x rays with zone plates and at the high energy end with capillaries, it is
probable that the XRM resolutions will approach a constant value of about 0.025 to 0.05 ~m.
We have already seen the ctiacteristics
of x-ray sources improve by many orders of
magnitude. There is a good probability that this rate of improvement is not about to come to an end.
34
There has been an intense effort to develop a free electron laser at a number of laboratories. For
instance, the Stanford Linear Accelerator Center (SLAC) has developed a design for a linac coherent
light source (LCLS) based on a 15-GeV linear accelerator [Comacchi~ 1998]. Figure 45 shows the
average and peak spectral brightness as a function of photon energy for the LCLS as compared to
the values for other operating and proposed facilities. Improvements of three or more orders of
magnitude should be attained. Many new types of experiments should be possible considering the
time structure of the photon beams and their intensity.
It seems very safe to conclude that SRIXE experiments should continue to bean area of
continuing high interest for the foreseeable fhture. This is based on the established value of the
analytical techniques and the prospects for major improvements in its capabilities from fiture
improvements in the x-ray source, x-ray optics and detectors, and in the interfacing of computational
methods with the expe~ments.
ACKNOWLEDGEMENTS
I amparticukwy indebted to my colleagues for stimulating discussions and interactions which
have influenced my views of x-ray microscopy over the years. Among them are: R. S. Beckman,
R. D. Giauque, Y. Gohshi, L. Grodzins, A. L. Hanso~ J. B. Hastings, K. Janssens, J. G. Pounds, C.
Riekel, M. L. Rivers, A. J. Saubermann, J. V. Smith, P. Spanne, S. R. Sutton, A. C. Thompson, S.
Torok, C. Tuniz J. H. Underwood, and R. D. Vis. I am particularly saddened to note that P. Spanne
died in the crash of Swiss Air Flight111, September 1998,
This work was supported in part by Processes and Techniques Branch, Division of Chemical
Sciences and the Division of Engineering and Geosciences, Office of Basic Energy Sciences, US
Department ofEnergy for development and application of analytical techniques under Contracts No.
DE-AC02-76CHOO016 and DE-ACO2-98CH1O886 and by the National Institutes of Health
Biotechnology Research Resource Grant No. P41RR01 838.
35
.’
REFERENCES
Ade, H., Kiriz, J., Hulbert, S. L., Johnson, E. D., Anderson, E., and Kern, D. Appl. Phys. Letts.
56:1841,1990.
Agarwal, Deb et al. An Overview of The Spectro-Microscopy Collaborator at the Advanced Light
Source: A Dktributed Collaborator Testbed. March 3,1997.
Altwicker, E, R., Schonberg, J. S., Konduri, N. V., and Millig~
M. S. Hazardous Waste &
Hazardous Materials 7(no. 1):73, 1990.
Auzerais, F. M., Dunsmuir, J., I?erreol,B. B., Martys,N., OISOWJ., Ramakrishnan, T. S., Rothman,
D. H., and Schwartz, L: M. Geophysical Research Letters 23:705-708, 1996.
Bajt, S., Clark, S. B., Sutton, S. R., Rivers, M. L., and Smith, J. V. Analytical Chemistry 65:1800,
1993.
Baryshev, V. B., Gavrilov, N. G., DarYin,A. V., Zolotarev, K. V., Kulipanov, G. N., Mezentsev, N.
A., and Terekhov, Ya. V. Rev. Sci. Instrum. 60:2456, 1989.
Bertsch, P. M., Hunter, D. B., Sutton, S. R., Bajt, S., and Rivers, M. L. Environmental Science and
Technology ~!8:980, 1994.
Bilderback, Donald. H.; Hoffinan, Stephen A.; and Thiel, Daniel J. Science 263:201-203, 1994a.
Bilderbac~ D. H., Thiel, D. J., PaM, R., and Brister, K. E. Journal of Synchrotron Radiation 1:3742, 1994b.
Bilderback, Donald H.; Hoffimm, Stephen A.; and Thiel, Daniel J. SynchrotronsRadiation News
7:27, 1994c.
36
.’
.*
Bilderback, Donald H.; Pahl, Reinhard; and Freelan~ Riley. Chess Newsletter-1995, pp. 41-43,
1995a.
Bilderbac~ Donald H. Rev. Sci. Instrum. 66:2058-2063, 1995b.
Bion@ R. M., Ables, E., Clamp, O., Edwards, O. D., Gabriele, P. C., Miller, K., Ott, L. L., Sku!b~
K. M., Tilley, R., and Via
T. Opt. Eng., 29:576, 1990.
Bockm~ R. S., Repo, M. A., Warren, R. P., Jr., Pounds, J. G., Schidlovsky, G., Gordon, B. M., and
Jones, K. W. Proc. Natl. Acad. Sci. USA 87:4149, 1990.
Bonse, U., Johnson, Q.jNichols, M., Nusshardt, R., Krasmaki, S., and Kinney, J. H. Nucl. Instr. and
Meth. in Phys. Res. A245:644, 1986.
Brissaud, I., Wang, J. X., and Chevalier, P. Journ. Radioanal. Nucl. Chem. 131:399, 1989.
Bufliere, J.-Y., Maire, E., Cloetens, P., Peix, G., Salom4, M., and Baruchel, J. ESRF Newsletter,
April 1998, pp. 20-21.
Cai, Zhonghou; Yun, Wenbing; Pratt, Stephen T.; Miller, Raymond M.; Gluskin, Efim; Hunter,
Douglas B.; Jarstfer, Amiel G.; Kemner, Kenneth M.; Lai, Barry; Lee, Heung-Rae; Legnini, Daniel
G.; Rodrigues, Willi=,
and Smi@ Christopher I.
Advanced Photon Source Research,
ANL/APS/TB-31 1(1):5, 1998.
CAT Communicator 8. Distance Links: Working Remotely at the APS. CAT Communicator 8,
May 1998,1-2, 1998.
Chen, J. R., Chao, E. C. T., Minkiw J. A., Back, J. M., Jones, K. W., Rivers, M. L., and Sutton, S.
R. Nuclear Instruments and Methods in Physics. Research B49:533, 1990.
37
‘
Chevalier, P,, Jehanno, C., Maurette, M., Sutton, S. R., and Wang, J. J. of Geophys. Res. 192(no.
B4):E649, 1987.
Cline, Hollis T.; Witte, Soni~ and Jones, Keith W. Proc. Natl. Acad. Sci. USA 93:9915-9920,1996.
Coles, M. E., Hazle@,R. D., Muegge, E. L., Jones, K. W., Andrews, B., Dowd, B., Siddons, P.,
Peskin, A., Spanne, P., and Soil, W. E. Proceedings of 1996 Annual Technical Conference and
Exhibition, Denver, Colorado, October 6-9, 1996, Paper SPE 36531, p. 413.
Coles, M. E., Hazlett, R. D., Spanne, P., Soil, W. E., Muegge, E. L., and Jones, K. W. J. Petroleum
Science & Engineering 19:55, 1998a.
Coles, M. E., Hazle% R. D., Muegge, E. L., Jones, K. W., Andrews, B., Dowd, B., Siddons, P.,
Peskin, A., Spanne, P., and Soil, W. E. SPE Reservoir Evaluation& Engineering, p. 288, August
1998b.
Comacchia, .Max. SynchrotronsRadiation News 11:28-33,1998.
Cramer, S. P., Tenth, O., Yocum, M., and George, G. N. Nucl. Instr. and Meth. in Phys. Res.
A266:586, 1’988.
Delaney, J. S., Bajt, S., Sutton, S. R., and Dyar, M. D. Mineral Spectroscopy, Special Publication
No. 5, 165,1996.
Delaney, Jeremy S.; Dyar, M. Darby; Sutton, Steven R.; and Bajt, Sasa. Geology 26:139, 1998a.
Delaney, Jeremy S.; Sutton, Steven R.; and Dyar, M. Darby. Lunar and Planetary Science=
#1241 (CD-IROM): March 1998b.
38
.’
.’
●,
Dowd, B. A., Andrews, A. B., Marr, R. B., Siddons, D. P., Jones, K. W., and Peskin, A. M.
Presented at 4P Annual Denver X-ray Conference, Colorado Spring, Colorado, August 3-7, 1998;
Advances in X-Ray Analysis, in press.
Elder, F. R., Gurenwitsc~ A. M., Langmuir, R. V., and H. C. Pollock. Phys. Rev. 71:829, 1947.
Feng, Huan, Jones, K. W., Stewart, B., Herzog, G. F., Schnabel, C., and Browrdee, D. E. Presented
at 30* Lunar and Planetary Science Conference, March 15-19, 1999, Johnson, Space Center,
Houston, Texas.
Flannery, B. P., Deckm@ H. W., Roberge, W. G., and D’Arnico, K. L., Science 237:1439, 1987.
Flynn, G. J., and Sutton, S. R. In Proceedings of the 20th Lunar and Planetary Science Conference,
Houston, Texas, 1990, p. 335.
Fuggle, J. SynchrotronsRadiation News 3(no. 1):24, 1990.
Giauque, R. D., Garrett, R. B., and God% L. Y. Anal.Chem.51:511,
1979.
Giauque, R. D., Jaklevic, J. M., and Thompsow A. C. Anal. Chem. 58:940, 1986.
Giauque, R. C., Thompson, A. C., Underwood, J. H., Wu, Y., Jones, K. W., and Rivers, M. L. Anal.
Chem. 60:855,1988.
Gordou B. M. Nucl. Instr. and Meth. 204:223,1982.
Gordon, B. M., and Jones, K. W. Nucl. Instr. and Meth.inPhys.Res.BIO/11
Grodzins, L. NeuroTox. 4:23, 1983.
39
:293, 1985.
.
Griibler, Amulf. Time for a Change: On the Patterns of Diflision of Imovation. Technological
Trajectories and the Human Environment. Washington, DC: National Academy Press, 1997, pp.
14-32.
Hagenmaler, H., Kraft, M., Brunner, H., and Haag, R. Environ. Sci. Technol. 21:1080,1987.
Hanson, A. I.. Nuclear Instruments and Methods in Physics Research A290: 167,1990.
Hayakawa, S., Iida, A., Aoki, S., and Gohshi, Y. Rev. Sci. Instrurn. 60:2452, 1989.
Hayakaw~ S., Gohshi, Y., Ii@ A., Aoki, S., and Ishikaw~ M. Nucl. Instr. and Meth. in Phys. Res.,
B49:555, 1990.
Hoffi-n~ Stephen A.; Thiel, Daniel J.; and Bilderback, DonaldH. Optical Engineering 33:303-306,
1994.
Horowitz, P,, and Howell, J. Science 178:608, 1972.
Hren, J. J., Goldstein, J. I., and Joy, D. C. eds. Introduction to Analytical Electron Microscopy.
New Yorlq New York: Plenum Press, 1979.
Iid% A., Takahashi, M., Sakurai, K., and Gohshi, Y. Rev. Sci. Instrwrn.60:2458, 1989.
Isaacs, H. S., Davenport, A. J., Cho, J. H., Hansom A. L., and Rivers, M. L. National Synchrotrons
Light Source Annual Report 1990, BNL 52272,1991, p. 348.
Jackson, A. SynchrotronsRadiation News 3(no. 3):13, 1990.
Johansson, S. A. E, and Campbell, J. L. eds. PIXE: A Novel Technique for Elemental Analysis,
New York, INewYork John Wiley& Sons, 1988.
40
,)
.’
,,
Jones, K. W., Gordon, B. M., Hanson, A. L., Hastings, J. B., Howells, M. R., and Kraner, H. W.
Nucl. Instr. and Meth. in Phys. Res. B3:225, 1984.
Jones, K. W., Kwiatek, W. M., Gordon, B. M., Hanson, A. L., Pounds, J. G., Rivers, M. L., Sutto~
S. R., Thompson, A. C., Underwood, A. H., Giauque, R. D., and Wu, Y. In: Barrett,C. S., Gilfiich,
J. V., Jenkins, R., Russ, J. C., Richardson, J. W., Jr., and Predicki, P. K., eds. Advances in X-ray
Analysis 31. New York, New York: Plenum Press, 1988, p. 59.
Jones, K. W., and Gordon, B. M. Anal. Chem. 61(no. 5), 341A, 1989.
Jones, K. W., Gordon, B. M., Schidlovsky, G., Spanne, P., Dejun, Xue, Bockm~
R. S., and
Saubermann, A. J. In: Williams, D. B., Ingram, P., and Michael, J. R., eds. Microbeam Analysis- 1990. San Francisco, California: San Francisco Press, Inc., 1990a, pp. 401-404.
Jones, K. W., Schidlovsky, G., Burger, D. E., Milder, f. L., and Hu, H. In: Yamumq
S., et al., eds.
New York, New York: Plenum Press, 1990b,p.281.
Jones, K. W., Sparme, P., Webb, S. W., Conner, W. C., Beyerlein, R. A., Reagan, W. J., and
Dautzenberg, F. M. Nuclear Instruments and Methods in Physics Research B 56/57:427-432, 1991.
Jones, K. W., Schidlovsky, G., Spanne, P., Dejun, Xue, Bockm*, R. S., Rabinowitz; M. B.,
Harmnon~ P. L., Bornschein, R. L., and Hoeltzel, D. A. In:
Michette, A., Morrison, G., and
Buckley, C., eds. X-Ray Microscopy 111.Berlin: Springer-Verlag, 1992, pp. 431-434.
Jones, K. W., Berry, W. J., Borsay, D, J., Cline, H. T., Conner, W. C., Jr., and Fullmer, C. S. X-Ray
Spectrometry 26:350-358 (1997).
Jones, K. W. In: Janssens, K., Rindby, A., and Adarns, F., eds. Microscopic X-ray Fluorescence
Analysis, Sussex, England: John Wiley & Sons Ltd, 1999, to be published.
41
,4
Karasek F. W., and Dickson, L. C. Science 237:754,1987.
Kermey, J. M., Jacobsen, C., Kirz, J., Rarback, H., Cinotti, F., Thomlinson, W., and Schidlovsky,
G. J. Microscopy 138(Pt. 3):321, 1985.
Ketelsen, P., KtMchel, A., and Petersen, W Fresenius Z. Anal. Chem. 323:807, 1986.
Kim, Kwang-Je Ki. Vaughan, Douglas, ed. Section 4, X-Ray Data Booklet. Berkeley, California:
Lawrence Berkeley Laboratory, Center for X-Ray Optics, Berkeley, 1986.
Kinney, J. H.,,Johnson, Q. C., Saroyan, R. A., Nichols, M. C., BonSe, U., Nusshardt, R., and Pahl,
R. Rev. Sci. [nstr. 59:196, 1988.
Kirkpatrick, P., and Bae~ A. V. J. Opt. Sot. Am. 39:766, 1948.
Kirz, J., and Sayre, D. In: Winick, H., and Doniac~ S., eds. SynchrotronsRadiation Research. New
York, New York: Plenum Press, 1980, p. 277.
Kndchel, A., Petersen, W., Tolkien, G. Nuclear Instruments and Methods 208:659, 1983.
Kndchel,A. Fresenius Z. Anal, Chem. 337:614, 1990.
Koningsberger, D. C., and Prim, R. X-ray Absorption: Principles, Applications, Techniques of
EXAFS, SEX=S,
and XANES. New York: Wiley, 1988.
Kopp, O. C., and Reeves, D. K. Chemical Geology 81:337-348, 1990.
KU
C. SynchrotronsRadiation: Techniques and Applications. Springer-Verlag, 1979.
42
,!
.*
Kwiatek, W. M., Hanson, A. L., and Jones, K. W. Nuclear Instruments and Methods in Physics
Research B50:347, 1990.
LoPac~
R. M., Jr., Lowery, J., Eichberg, J., Kirkpatrick, J. B., Cartwright, J., Jr., and Saubermann,
A. J. J. Neurochem. 51(no. 3):764, 1988.
LoPachin, R. M., Jr., LoPachin, V. R., and Saubermann, A. J. J. Neurochem. 54:320, 1990.
Lu, F. Q., Smith, J. V., Sutton, S. R., Rivers, M. L., and Davis, A. M. Chem. Geol. 75:123, 1989.
Margaritondo, Giorgio. Introduction to SvnchrotronRadiation. Oxford University Press, New York
(1988).
Martin, R. R., Sham, T. K., Won, G. W., van der Heide, P., Jones, K. W., Song, S.-R., and Protz,
R. Canadian J. of Forest Research, in press, 1998.
Padmore, H. A., Ackerman, G., Celestre, R., Chang, C.-H., Franck, K., Howells, M., Hussain, Z.,
Irick, S., Locklin, S., MacDowell, A. A., Patel, J. R., Rah, S. Y., Renner, T. R., and Sandier, R.
SynchrotronsRadiation News 10:18-26, 1997.
Rarback, H., Cinott, F., Jacobsen, C., Kemey, J. M., Kirz, J., and Rosser, R. Biol. Trace Element
Res. 13:103,1987.
Raven, C., Snigirev, A., Koch, A., Snigireva, I., and Kom
V.
Society of Photo-Optical
Instrumentation Engineers, San Diego, 1997.
Rivers, M. L., Sutton, S. R., and Jones, K. W. In: Michette, A., Morrison, G., and Buckley, C., eds.
X-Ray Microscopy III. Berlin: Springer-Verlag, 1992, p. 212.
43
.
I&m, R., Grimdolfo, M., Godeas,’C., Jones, K. W., and Vittur. F. Journal of Experimental Zoology
273:82, 1995.
Robinson, Art. SynchrotronsRadiation News 10:2-4, 1997.
Sarret; G. , Manceau, A., and Hazemann, J. L, Environmental Science. ESRF Newsletter No. 30,
April 1998, pp. 51-52.
Sayre, D. Kirz, J., Feder, R., Kim, D. M., and Spiller, E. Ultramicroscope 2:337,1977.
Sayre, D., K@ J., Feder, R., Kim, D. M., and Spiller, E. Ann. N. Y. Acad. Sci. 306:286, 1978.
Sayre, D., Howells, M., Kirz, J., and Rarback, H. eds. X-Ray Microscopy II. Berlin-HeidelbergNew York-London-Tokyo: Springer-Verlag, 1988.
Slatkin, D. N., Hanson, A. L., Jones, K. W., Kraner, H. W., Warren, B. P., and Finkel, G. C. Nucl.
Instr. and Mcth. 227:378, 1984.
Smith, J. V. The Analyst 120:1231, 1995.
Spanne, P., and Rivers, M. L. Nucl. Instr. and Meth. in Phys. Res. 1324/25:1063, 1987.
Spanne, P. F’hys.Med. Biol. 34(no. 6):679, 1989.
Spanne, P. Abstract of invited talk at Bioscience 90--Swedish National Convention of BioScientists, Malrno, Sweden-April 23-26,1990. See also Jones, K. W., Beckman, R. S., Gordon, B.
M., Rivers, M. L., Saubermann, A. J., Schidlovsky, G., and Spanne, P. In: Moro, R., and Cesareo,
R., eds. XIZF and PIXE Applications in Life Science. Singapore, 1990, p. 163.
44
,’
Spanne, P., Thovert, J. F., Jacquin, C. J., Lindquist, W. B., Jones, K. W., and Adler, P. M. Physical.
Review Letters 73:2001, 1994.
Sparks, C. J., Raman, S., Ricci, E., Gentry, R. V., and Krause, M. O. Phys. Rev. Lett. 40:507,
1978.
Sparks, Jr., C. J. In: Winick, H., and Doniac~ S. eds. SynchrotronsRadiation Research. New York,
New York: Plenum Press, 1980, p. 459.
Sparks, C. J., and Ice, G. E. In: Proceedings 15th Intern. Conf. on X-Ray and Inner-Shell Processes,
Knoxville, Tennessee, 1990.
Sutton, S. R., and Flynn, G. J. In: Proceedings of the 18th Lunar and Planetary Science
Conference, Houston, Texas, 1988, pp. 607-614.
t
Sutton, S. R., Jones, K. W., Gordon, B., Rivers, M. L., and Smith, J. V. Lunar and Planetary Science
XXII. Houston, Texas: Lunar and Planetary Institute, 1991, p. 1365.
Sweet, R., et al., 1995.
Thernner, K., Spanne, P., and Jones, K: W. Nucl. Instr. and Meth. in Phys. Res. B49:52, 1990.
l%iel, Daniel J.; Bilderback, Donald H.; and Lewis, Aaron.SPIE 1740:248-251,1992.
Thompson, A. C., Wu, Y., Underwood, J. H., and Barbee, T. W., Jr. Nucl. Instr. and Meth. in Phys.
Res. A225:603, 1987.
Thompson, A. C., Underwood, J. H., W% Y., Giauque, R. D., Jones, K. W., and Rivers, M. L. Nucl.
Instr. and Meth. in Phys. Res. A266i3 18, 1988.
45
,
Tokunag~ Tetsu K.; Brown, Jr., Gordon E.; Pickering, Ingrid J.; Sutton, Stephen R.; and Baj~ Sasa.
Environmental Science and Technology 31(5):1419, 1997.
Tokunag~ Tetsu K.; Sutton, Stephen R., Bajt, Sas~ Nuessle, Patterson, and Shea-McCarthy, Grace.
Environment
Science and Technology 32(8): 1093,1998.
Tolentino, H., Baudelet, F., Dartyge, E., Fontaine, A., Len% A., and Tourillon, G. Nucl. Instrum.
and Meth. in Phys. Res. A289:307, 1990.
Tombulian D. H., and Harrnan, P. L. Phys. Rev. 102:1423, 1956.
Tros, G. H. J., Van Langevelde, F., dd Vis, R. D. Nucl. Instr. and Meth. in Phys. Res. B50:343,
1990.
Tuniz, C., Zanini, F., and Jones, K. W. Nucl. Instr. and Meth. in Phys. Res. B56/57:877, 1991.
US Department of Energy.
DOE 2000.
National Collaboratoties, pp. 4-5; The Materials
Microcharacterization Laboratory pp. 6-7. DOE Publication-797, June 1997.
Underwood, J. H., Thompson, A. C., Wu, Y., and Giauque, R. D. Nucl. Instr. and Meth. in Phys.
Res. A226:296, 1988.
Van Langevelde, F., Bowen, D. K., Tros, G. H. J., Vis, R. D., Huizing, A., and de Boer, D. K. G.
Nucl. Instr.
i~d
Meth. in Phys. Res. A292:719, 1990a.
Van Langevelde, F., Tros, G. H, J., Bowen, D. K., and Vis, R. D. Nucl. Instr. and Meth. in Phys.
Res. B49:544, 1990b.
46
,,
r
,,
.
Van Langevelde, F., Tros, G. H. J., Bowen, D. K., and Vis, R. D. In: Jasiensk~ S., and
Maksymowicz, L. J., eds. Proceedings of XIIth IXCOM, Krakow, Poland, 1989. Institute of Physics
Publishing Ltd., 1990c, p. 453.
Vis, R. D. Fresenius Z. Anal. Chem. 337:622, 1990.
Vis, R. D., and Van Langevelde, F. Nucl. Instr. and Meth. in Phys. Res. B54:417, 1991.
Warwick, T., Anders, S., Hussain, Z, Lamble, G. M., Lorusso, G. F., MacDowell, A. A., Martin, M.
C., McHugo, S. A., McKinney, W. R., and Padmore, H. A. Imaging Spectroscopic Analysis at the
Advanced Light Source. SynchrotronsRadiation News 11:5, 1998.
Winick, H. SynchrotronsRadiation News 2(no. 2):25, 1989.
Winick, H Nuclear Instruments and Methods 291:401, 1990.
Winick, H., and Doniach, S., eds. SynchrotronsRadiation Research. New York, New York: Plenum
Press, 1980.
Wu, Y., Thompson, A. C., Underwood, J. H., Giauque, R. D., Chapman, K., Rivers, M. L., and
Jones, K. W. Nucl. Instr. and Meth. in Phys Res. A291:146, 1990.
.
47
Table 1. First-GenerationSynchrotronsLight Sourcesa
Storage Ring (Lab)
Energy (GeV)
Location
ADONE (LNT)
1.5
Frascati,ITALY
DCI (LURE)
1.8
Orsay,FRANCE
VEPP-3(INP)
2.2
Novosibirsk,USSR
BEPC (IHEP)
2.2- 2.8
Beijing, CHINA
SPEAR(SSRL)
3.0- 3.5
Stanford,USA
www-ssrl.slat.stanford.edu/
welcome.html
ELSA(BonnUniv.)
3.5
Bonn,GERMANY
Www-elsa.physik..uni-bonn.de
DORM11(W4SYLAB)
3.5- 5.5
Hamburg,GERMANY Www-hasylab.desy.de
VEPP-4(INP’)
5.0- 7.0
Novosibirsk,USSR
CESR(CHESS)
5.5- 8.0
Ithac%USA
www.chess.comel] .edu
Ace. Ring (KEK)
6.0- 8.0
Tsukuba,JAPAN
www.kek.jp
Tristan (KEK)
25-30
Tsubuka.JAPAN
wvvw.lure.u-psud.fi
n Most of these early facilities were built to do research with the primary electron or positron beams.
Synchrotrons radiation research was parasitic. Today, all have become at least partly dedicated as
synchrotronsligM sources. See I-I.Winick, 1989, 1990; A. Jackson, 1990; J. FuggIe, 1990.
Lists of synchrotrons sites can be found from many of the home pages listed above.
48
Table 2. Second-GenerationSynchrotronsLight Sources”
Storage Ring (Lab)
Energy (GeV)
Location
CAMD (LSU)
1.4
Baton Rouge, USA
Www.camd.lsu.edu
DELTA (DU)
1.5
Dortmund, GERMANY
prian.physik.unidortmund.de
NSLS x-ray (BNL)
2.5
Upton, USA
Www.nsls.bnl.gov
Photon Fat. (KEK)
2.5
Tsukub% JAPAN
www.kek.iv
Synch. Rad. Source (SRS)
2.0
Daresbury, UK
www.dl.ac.uk
‘ This is the first generation of machines built to be dedicated synchrotronsradiation facilities. The use of
bending magnet ports is emphasized although some straight sections have insertion devices (e.g. undulatory
and wigglers). See H. Winick, 1989, 1990; A. Jackson, 1990; J. Fuggle, 1990.
Lists of synchrotronssites can be found from many of the home pages listed above.
49
‘
Table 3. Third-Generation SynchrotronsLight Sources’
Light Source
Energy (GeV)
Location
ALs
1.0- 1.9
Berkeley, USA
wwvv-als.lbl.gov
ASTRID
0.58
Aarhus, DENMARK
www.dfi.aau.dk
BSRF
1.4-1.55
Beijing, CHINA
solar.rtd.utk.edukchird
ins/iHEP/bsrf/bsrf.html
CLS
2.5- 2.9
Saskatoon, CANADA
sal.usask.caicls/cls. html
SLS
2.1
Villigen, SWITZERLAND
www.psi.chlsls
SRC
0.8- 1.0
Stoughton, USA
Www.src.wise.edu
SRRC
1.3
Hsinchu, ROC
210.65. 15.200/en/index.html
INDUS II
1.4
Indore, INDIA
MAXII
1.5
Lund, SWEDEN
www.rnaxlab.in.se
BESSY II
1.5- 2.0
Berlin, GERMANY
www.bessy.de
ELET’TRA
1.5- 2.0
Trieste, ITALY
www.elettra.trieste.it
LNLS
2.0
Campinas,BRAZIL
www.lnls.br
PLS
2.0
Pohang,KOREA
SIBERIAII
2.5
Moscow,USSR
ESRF
6.0
Grenoble,FIL4NCE
www.esrf.fi
APs
7.0
Argonne,USA
Www.epics.aps.anl.gov
SPring-8
8.0
Kansai,JAPAN
www.spring8.or.jp
DAPS
???
Daresburv.UK
www.dl.ac.uk
aThis is the newest generation of dedicated synchrotronsradiation facilities. The use of insertion devices (e.g.
undulatory and wigglers) is emphasized, but bending-magnet ports will also be available. See H. Winiclq
1989, 1990; A. Jackson, 1990; J. Fuggle, 1990.
Lists of synchrotrons sites can be found from many of the home pages listed above.
50
,’
.“.’
Table 4. Experimental and Predicted Count Rates for Different Beam Filtersa
Beam Filter
Thickness ~m Al]
Fe [cts/sl
Mb”
50
86.5
100
200
Zn [cts/sl
M
T
75.6
77.7
53.7
54.8
48.5
59.1
43.2
24.7
23.6
37.0
28.8
F
aExperimental errors at 10-15V0.
b Measured values.
‘ Theoretical values.
51
Table 5. ResultsDeterminedfor ThreeNBSStandardReferenceMaterials@g/g)
WheatFlour
SRM 1567
disld 60 mg/cm2
wtb43 mg
wt’400 mg
Non-FatMilk Powder
SRM 1549
disld51 mg/cm2
wtb37 mg
WP500 mg
Element
NBS
XRF
NBS
RiceFlour
SRM 1568
disk’60 mg/cm2
Wtb43 mg
wtc400 mg
XRF
NBS
XRF
K
16900 * 3000
17800 + 2000
1360 *40
1220 * 130
1120+20
1360+ 160
Ca
13000 * 500
12000 + 800
190+ 10
174+ 10
140 *20
158+ 14
0,0026 + 0.0007
Mn
0.26 * 0.06
0.33 + 0.12
Fe
(2.1)
2.30 % 0.16
Ni
<().3
<o,6
Cr
8.5 * 0.5
8.2 * 1.8
20.1 * 0.4
22.1 + 2.8
18.3 *1.O
17.1 *4.8
8.7 + 0.6
9.1 * 1.2
.
0.24 * 0.06
(0.18)
0.11 * 0.06
(0.16)
0.18 + 0.06
0.65 * 0.04
2.0 * 0.3
1.88 + 0.12
2.2 A 0.3
2.21 % 0.22
Cu
0.7 + 0.1
Zn
46.1 * 2.2
46.9 + 0.9
10.6* 1.0
As
(0.0019)
<(),()5
(0.006)
Se
0.11 +0.01
<0.4
10.3 * 0.4
<0.03
0.92 * 0.06
1.1+0.2
0.09 ~ 0.04
19.4* 1.0
21.9* 1.8
0.41 + 0.05
0.42 * 0.09
0.4 *o. 1
0.38 * 0.04
Br
(12)
12.1 * 0.2
(9)
8.5 * 1.4
(1)
1.19+0.17
Rb
(11)
13.1 +0.2
(1)
0.94 * 0.06
(7)
8.4 * 0.9
Sr
0.82 + 0.04
3,69*0.10
.
0.19 + 0.04
Hg
0.0008 * 0.0002
<(),1
0.001 + 0.0008
<o.06
0.0060 * 0.0007
Pb
0.019 * 0.003
<().1
0.020 * 0.010
<0.1
0.045 * 0.010
<o.08
0.10 + 0.09
“Massthicknessof disks.
bWeightof areascanned.
‘Recommendedsamplewt.
Source:R, D, Giauque,et al., 1986.
.
52
.
“.
Table6. 16.6and21.7keVSRXRFAnalysisof ThreeInternationalGeostandardsa
GSN
Element
BEN
SRIXE
SRIXE
16.6
21.7
3.84
3.54
3.34
Ca, %
1.78
1.68
Ti, %
0.41
0.41
K,
yO
SRIXE
GST
GST
..
MICA-Fer
GST
166
717
7.26
6.10
5.89
8.74
0.31
0.15
0.07
1,49
1.50
1.35
1.47
16.6
21.7
1.15
1.05
0.86
1.66
9.85
8.98
0.38
1.57
1.54
v
65
125
188
235
185
360
135
42
257
Cr
55
45
117
360
346
407
90
125
134
Mn
433
479
551
1540
1662
1679
2695
3006
3200
9.66
8.99
3.12
3.13
2.63
Fe, %
18.95
17.96
9.20
19.15
---
Ni
34
36
35
267
320
278
35
Cu
20
16
25
72
62
85
4
Zn
48
60
66
120
117
129
1300
1320
1501
Ga
22
21
27
17
10
21
95
79
118
Rb
185
115
189
47
27
61
2200
1013
2696
Sr
570
469
905
1370
1550
1759
5
1
5
Y
19
---
22
30
---
34
25
---
50
Zr
235
---
296
265
---
343
800
.-.
1058
Nb
23
...
32
100
---
136
270
...
375
Ba
1400
.. .
1103
---
---
---
...
...
---
w
470
500
610
30
30
...
---
...
Pb
53
51
82
4
2
5
13
23
17
Th
44
11
13
11
2
3
150
51
50
...
2
60
u
8
---
4
2.4
39
---
..-
14
33
‘ GST are the admittedvalues. Unitsare in ppm or ‘%0.The abbreviations,GSN,BEN,MICA-Ferarethe namesof standardsas givenin K. Gavindaraju,GeostandardsNewsletters,
J3 (1979; 4:49 (1980);& 173 (1984). Source: I. Brissaud, et al., 1989.
53
,
FIGURE CAPTIONS
Figure 1.
The number of zinc atoms contained in a given volume element is plotted as a
fi.mctionof volume. A constant zinc concentration of 1 ppm contained in an
organic matrix with a density of 1 g/cm3 is assumed. This shows that the
detection of trace amounts of an element at a given concentration level becomes
increasingly dii%cult as the volume probed decreases. (From K. W. Jones and J.
G. Pounds. Biol. Trace Elem. Res.12:3, 1987.)
Figure 2.
The brilliance of the NSLS x-ray ring is plotted as a fimction”of the x-ray energy
produced. The brilliance is important in deffig
limitations on production of x-
ray images using focussing optics. (From K. W. Jones et al. Ultrarnicroscopy
24:313,1988.)
Figure 3.
The brightness of the NSLS x-ray ring is defined in Eq. 1 and plotted as a function
of the x-ray energy produced. The brightness is defined in Eq. 1. The brightness
is of importance in defining the usefi.dness of x-ray beams produced by use of a
collimator. (From K. W. Jones et al. Ultramicroscope 24:313, 1988.)
Figure 4.
The x-ray flux produced by the NSLS x-ray ring is plotted as a function of the xray energy produced. The values are given after integration over the vertical
opening angle of the beam. The definition of the flux is given in Eq. 2. (From K.
W. Jones et al. Ultramicroscope 24:313, 1988.)
Figure 5.
The brilliance obtained from an undulator insertion device at the 7-GeV Argonne
National Laboratory Advanced Photon Source. (From Advanced Photon Source
Research, ANL/APS/TB-31 1(1):5, 1998.)
Figure 6.
The brilliance for different bending magnet, undulator, and wiggler devices at the
8-GeV Spring-8 facility in Japan. (From Y. Miyahara. SynchrotronsRadiation
News 9:24, 1996.)
54
,’
.
,*
Figure 7.
A comprehensive synchrotronsbeam line designed for use as a x-ray microscope.
Not all the components would be utilized at a given time in practice. The rather
varied uses of the system require a flexible approach to the design of the
equipment.
Figure 8.
Enery distribution of photon flux produced by a bending magnet on the NSLS xray ring at a stored electron current of 100 mA. (From K. W. Jones et al., 1990b.)
Figure 9.
Values for MDLs obtained for a white-light XRM are shown as a fimction of
atomic number. The change in the MDLs as a function of filtering of the incident
beam using aluminum filters is an important feature of this type of arrangement.
(From A. Knochel et al., 1983.)
Figure 10.
Dependence of incoherent scattering cross sections for x rays polarized parallel
and perpendicular to the plane of the stored electron orbit on the scattering angle.
Observation at a scattering angle of 90° gives optimum signal-to-background
conditions. (From K. W. Jones et al. Ultrarnicroscopy 24:313, 1988.)
Figure 11.
Polarization of NSLS x-ray beams is given as a fimction of distance from the
plane of the electron-orbit for x-ray energies of 10,20, and 30 keV. (From A. L.
Hanson, et al. Nucl. Instrum. and Methods in Phys. Res. B24/25:400, 1987.)
Figure 12.
The background-to-signal ratio is shown as a fimction of displacement of the
detector normal to the plane of the orbit of the stored electrons. The alignment
becomes increasingly critical at the higher x-ray energies. The curves shown in
Fig. 10a were obtained for C% Fe, Zn, Br, and Sr contained in a gelatin matrix.
The curve shown in Fig. 10b was obtained for Pd contained in a pyrrhotite matrix.
(From Kwiatek et al., 1990.)
55
,
Figure 13.
The background-to-signal ratio is plotted as a function of scattering angle in the
horizontal plane. The curve was obtained by displacement of the detector and the
equivalent angular range spanned was roughly
Figure 14.
●
4°. (From Kwiatek et al., 1990.)
The sensitivity of the BNL collimated XRM is given as a function of atomic
number for K- and L-x ray detection, Values are given in units of counts/@pm-s)
at a stored electron current of 100 mA with an acquisition time of 300s and a
beam size of 8 ~m x 8 ~m. A 3-mm diameter Si(Li) x-ray detector placed 40 mm
from the beam was used as a detector. The curves were calculated from basic
principles. (From Jones et al., 1990b.)
Figure 15.
MDLs measured for the BNL collimated XRM under the conditions described in
the text and for Fig. 14. (From Jones, et al., 1990b.)
Figure 16.
Schematic diagram of the focussing XRM used at the Photon Factory. (From
FIayakawa et al., 1990.)
Figure 17.
Schematic diagram of the focussing XRM in operation at the Daresbury XRM.
(From F. Van Langevelde, et al., 1990c.)
Figure 18.
Schematic diagram of the focussing XRM of the Berkeley group used at the BNL
X26 beam line. (From A. C. Thompson, et al., 1988.)
Figure 19.
Schematic diagram of the imaging apparatus developed by LBL at the ALS.
(From Warwick, et al., 1998.)
Figure 20.
The experiments at CHESS have shown that it is possible to produce beams with
a diameter as small as 0.05 ~m. A scan across an edge demonstrating this spatial
resolution is shown. (From Hoffinan, et al., 1994.)
56
,,
.’
,,
Figure 21.
ESRF schematic diagram of the beam line optics is shown. (From C. Riekel,
private communication.)
Figure 22.
Sensitivity of the BNL collimated white light XRM compared with the sensitivity
obtained ~th the Berkeley focussed XRM. (From M. L. Rivers, et al., 1992.)
Figure 23.
Comparison
of concentration
for
Sr and Fe in
feldspar obtained
using the
NSLS
XRM with those obtained using an electron microprobe. The solid lines show the
values expected for exact agreement between the two methods. The good
agreement validates the use of the XRM method for geological analyses. (From
F. Q. L% et al., 1989.)
Figure 24.
Relative sensitivity for determination of elements from K to Zn obtained using the
LBL Kirkpatrick-Baez XRM at the NSLS X26 beam line. NIST (NBS) thin glass
Standard Reference Materials 1832 and 1833 were used. (From Giauque et al.,
1988.)
Figure 25.
Detection limits observed for aerosol samples using the Hamburg synchrotrons
with a 70 mA stored beam of 3.7 GeV electrons for 300s. The open triangles
were obtained with a white beam filtered with 8-mm of Al, filled circles fora31 keV monoenergetic beam, and open circles for a 12.5-keV monoenergetic beam.
(From P. Ketelsen, et al., 1986.)
Figure 26.
(a) Absorbed dose at the center of a circular water phantom for detection of an
element of fat with a signal-to-noise ratio of 5 as a function of photon energy.
The cylinder diameters are shown on the figure. (b) Absorbed dose at the center
of 1-mm diameter water phantoms for detection of elements of fat, air, and
calcium with a signal-to-noise ration of 5. The element diameter is .005 of the
phantom diameter or 5 Vm. (From P. Spanne, 1989.)
57
,,
Figure 27.
Morphology of red blood cells is shown as a function of the fluence of incident
15-keV photons. The fluences for the different exposures areas follows: 0,0.4,
1.1, 1.6,2.0, and 2.4 x 1019photons/cm2 for frames (a)-(f), respectively. (From D.
N. Slatkin, et al., 1984.)
Figure 28.
Mass loss observed for skin sample irradiated with a 66-keV beam of photons is
shown as a fimction of the dose. The mass loss was determined by observation of
the scattered radiation intensity. (From Themner et al., 1990.)
Figure 29.
Chemical compositions of coins determined using SRIXE, PIXE (solid bar), and
neutron activation analysis (open bar). The SRIXE work was done at 17 keV
(diagonrd cross-hatch) and 35 keV (horizontal cross-hatch). (From Brissaud, et
-
al., 1989.)
Figure 30.
Scan over a 2-pm section of a neonatal hamster tooth germ showing the variation
of zinc (a) and calcium (b). The zinc is located on the outside edge of the calcium
distribution, but the reasons for difference remain to be determined. (From G. H.
J. Tros, et al., 1990.)
Figure 31.
(a) Scan over a thin section of the tibia diaphysis of a rat which had been treated
with gallium nitrate, showing the distribution of the C% Zn, and Ga. Some of the
variations come from irregularities in specimen thickness. The spatial distribution
is a fiction
of the gallium nitrate treatment of the rat. The step size was 9 Km.
(b) Map of the distribution of gallium (left) and calcium (right) in a fetal rat ulna
bone afier exposure to gallium (25 PM) in a culture medium. The light regioti
have the highest elemental concentrations. A scanning electron micrograph
(center) shows the bone structure in the same region. The dark portions at the top
of the figure are the non-calcified cartilaginous portions of the bone which
accumulates little gallium. Most of the gallium is distributed in metabolicallyactive regions of the metaphysicsand diaphysis. (From Beckman, et al., 1990.)
58
‘
,’
.
,’
Figure 32.
The relative concentrations of C% Zn, and Pb are for a scan across a section of a
child’s deciduous tooth which was thick compared to the absorption of the
characteristic x rays. The variation of the lead in the scan maybe useful in the
future in understanding the time-dependence of lead exposure and uptake by the
child. The spatial resolution and step size were both about 10 ~m. (From K, W,
Jones, et al., 1992.)
Figqre 33.
Shows results obtained for the distribution of Mn and the density in the region
analyzed in the ESRF experiments were carried out using high-resolution beams
formed by capillary focussing to determine wood density and elemental
composition. (From ESRF Highlights-1996/1997, p. 80, A. Rindby and P.
Engstrom, to be published.)
Figure 34.
The apparatus used for studying corrosion of stainless steel in a chloride
environment is shown on the left. The spatial distribution of nickel observed
above the stainless steel “isshown as a fiction
of distance above the metal-liquid
interface. (From H. S. Isaacs, et al., 1991.)
Figure 35.
Shows a map of the Cr distribution. The experiment was done on microtomed
sections of polyethylene particles using the NSLS X26A XRM. Sharp peaks
observed at the periphery of the particle are consistent with observations made
using computed microtomography. (From Jones et al., 1997.)
Figure 36.
Computed-emission tomogram showing the distribution in the epineurium of a
freeze-dried rat sciatic nerve. The pixel size was 3 ~m x 3 pm and a slice
thickness of 5 ~m. The matrix size was 175 x 175 pixels. (From K. W. Jones, et
al, 1990a.)
59
.
Figure 37.
View of the pore space of Fontainebleau sandstone obtained by CMT. The pore
structure is shown as white and the rock as black. That is, the pore structure is
opaque and the rock transparent. (From Spanne et aL, 1994.)
Figure 38.
CMT 3-D image of the interior of type-I deep-sea spherule with parts of surface
cut away. The compact bright spot in the upper right is a platinum group metal
nugget. The light mass in the foreground represents a hole in the interior of the
spherule. (From Feng, et al., 1999.)
Figure 39.
Fluorescent XANES spectrum obtained for Fe at the 550-pm level in a thick NIST
SRM 1570 (spinach) saxpple compared to the spectrum obtained under the same
conditions for an iron foil.
Figure 40.
Fluorescent XANES spectrum obtained for Cr at the 1000-pm level in a 30-~m
thick section of pyroxene and olivine contained in lunar basalt 15555. The spatial
resolution was 200 ~m x 200 pm. (From S. R. Sutton, et al., 1991.)
Figure 41.
FluorescentXANES spectrum obtained for Cr at the 50-ppm level contained in a
thin section of rat kidney. The spatial resolution was 1 mm X 1 mm. The
spectrum shows that there is little Cr VI present in this portion of the kidney.
Figure 42.
A XANES spectrum for Mn obtained from a location in the root at a spatial
resolution of 1 ~m x 3 pm is shown. (From Cai et al., 1998.)
Figure 43.
Elemental mapping and chemical speciation of Cr in a soil specimen on the
micrometer-size scale. (From Warwick et al., 1998.)
60
“
,’
..
,*
Figure 44.
The spatial resolution (expressed as the beam area) obtained with high-energy xray microscopes is plotted as a function of time. The best spatial resolutions in
1990 are around 1-4 ~m2. If improvements were to continue at the same rate in
the fiture, the resolution will approach atomic dimensions around the year 2000.
Figure 45.
The average and peak spectral brightness as a fimction of photon energy for the
LCLS as compared to the values for other operating and proposed fmilities.
(From Comacchi~ 1998.)
4n 1
-I;i::i::l;;i”l’
Iu
‘l”’l--lool-ml--l
“-1-”’-
10-’
jo-3
1 xl mm2beam
4
—----- 1OOX1OO p.m2 beam
~lOx10
pm2beam
,0-11
103
105
107
“lo”
109
Number of Zinc Atoms
10’3
10’5
10’7
.
Figure
1
.
.
1
1-
-+-
\
I 1
Ill
1111111
I 11111111
I 1 11111111
I 1111111117 .
co
m
*
cm
ol-
-0
“o
“o
“o
‘o
(tuw/zpuw/Ma
%l”o/OaS/qd
Om=
—
‘o
-o
-o
o
>
w.
Ill
---
--.,
...
hull
I
I
m
m
‘o
“o
I 1 1I I 1111111
1111111
I I 1111111
0
0
Ym
-0
“o
0
1
.
.
Io”g
I
I 1 I 11111
I
I I I 11111
1
I I I Ill
a
1L
10’2
G
10”
~
10’0
F
1
109
0.1
I I I
11111
1
I
I
I 111111
10
I
I
I
100
Energy ( keV )
Figure
4
r
L
.
I II111I
I
l“’’’’’
’’1’ ’’’’
’’’’
[’’’
’’’’”1
,’
‘“ “n
—
Original Design
------ Enhanced Performance
n=l
0“
o
10
20
30
40
50
Energy(keV)
Figure
5
*,
~
lo?’
3
n
:
10*O
:
10’9
um
g 10’8
E
g 10’7
a)
~ 10’6
0
g
10’5
1
‘“ 102
I
1
I
1,,
,1
,03
IIJIIA # , ,,,,1
,
,
,04
, , , 1111
105
,
,
I
,
,
106
Photon Energy (eV)
Figure
6
,,
\
,,
“,
\
.
.—
Electron Storage
l.—.
—
.—.
—
.—
..
-o
Ring
Vee Collimator
m
m
Mormchromator
7
8000 _
Focussing Mirror
Collimator
00
!
Pinhole Microscope /
E
E
Monochromator
ion Chamber
-
Ion Chamber
Target
Assembly
Si(Li)
Detector
Ion Chamber
PB Beam Stop
!
-9000
-
.
,,
a
.
-
r“’’”‘
Sllllll i sip!
Lf5
1 111111111 111111111
N
m
d
ml
o
x
Lq
b
--
---.-
o
Ln
..
JIJllll
m
o
I I
I I 1111111
11!11111
I 1111111
I
e
0
m
0
m
o’
111111111 10
~
“o
o
,,
.
.-.
——.
.
.
0.10
0.08
-
;
u
0.04
0.02
.
0.000
I
30
1
60
90
120
ScatteringAngle( deg)
150
Figure
10
*
I
pill
I
pllll
C9
o
E
o
L
0
.
,
,“
40.00
1
I
I
I
I
(a)
~“ 30.00
~
*
a
20.00
j
\
.5?
m 10.00
0.00
-3.0
-2.0
-1.0
0.0
1.0
2.0
3.0
Distance From The Beam Center (mm )
1.00
(b)
0.80
0
g
: 0.60
~
z 0.40
.5?
m
0.20
I
0.00
-3.0
-2.0
I
-1.0
1
I
0.0
1.0
I
2.0
3.0
Distance From The Beam Center (mm )
Figure
12
.2
.
0.11
0.10
.
0
.=
& 0,09
I
~ 0.08
m
0.07
0.06
-3.0
-2.0
-1.0
0.0
1.0
2.0
Distance Along The Incident Beam (mm )
3.0
Figure
13
If
.
.
●
3000
I
I
1
I
I
I
I
i
i
I
I
-0
I
/\’-\
[
I
I
I
\
/’
1
1
1’
\
I
\
1’
1’
●
\\
‘\
“i,
\\
\\\
‘,
La
✎
o
✎
\
\
\\
✎
0
I
)90’
10
20
30
40
50
1
60
I
70
I
I
80
1
I
90
1
100
Atomic Number
Figure
14
100 k
I
I
1
I
i
I
I
I
I
I
I
I
I
I
I
Beam,Size = 8 pm x 8 pm
Ring Current = 100 mA
10
I
b“
0.1
!
0.001 ~
La
. .. . .
XKM
\=-/
Calculated
J
F
~cooof
a
L—LdJJ—
10
20
30
I
1
I
40
50
60
Atomic Number
I
i
70
I
I
80
I
90
.,\
Figure
15
.
al
n
-J
-
G
Storage
Ring
r=15.38m
..
and
Be-window
I
.
‘“~/
/
\-
L/ser
“ror
Slits
Ellipsoid
Si(lll)
C,rystal
I
Si(Li)
n- L--*-..
n.
‘38m-7’7XYZ
Stage
\
\
Vacuum
Chamber
.
Figure
17
.
,,
kI
e-
Synchrotrons Source
(white radiation )
/
Solid State
Si ( Li ) Detector
mirrors
Focal Spot
Scanning
Stage
Figure
18
.’
Horizontal“focwsshg
KB mirror
Sample
Figure
19
b
2.4
1.6
/1
,
,
1
‘o’”+ l’0’d411
:
0.8
0
0
L
-0.8 r
-0.3
1
1
o
I
I
-0,15
1
I
0
1
I
I
I
0.15
I
1
0.3
Position (~m)
Figure
20
,,
s
Focus
P’
E
Shielding
low-p
Undulator
E
0
w
\
..
0
m
V&Ju
EllipsoidalMirror
Iv”l
o
26.0
Mulfiayers
27.6
28.9
31.0
z
w
34.1 [103mm]
Figure
21
,,
.
t
1200
I
I
I
I
I
I
I
I
I
I
I
I
I
Microprobe
800
600
400
200
1
0
2
I
4
1
I
6
1
I
8
I
I
10
1
I
12
I
I
14
I
16
Fluorescent Peak X-ray Energy ( keV )
Figure
22
,
6000
I
II
5000
.
4000
.
I
I
.’
I
.
9
3000 .
2000
1000
0
0
1000
2000
3000
4000
5000
6000
EMP Sr Concentration (ppm)
4000“
I
I
i
I
i
I
●
I
3000 -
●
2000 -
1000 “
o
0
1000
2000
4000
EMP Fe Concentration (ppm)
Figure
23
1000
I
I
I
I
I
I
{
Cu
800
600
400
.
200
.
0
3456789
10
Kab(keV)
Figure
24
—
.
“o
o
cy*
o
0
1
I
I
1
-“\.
m
U3
‘~
In
ml
o
1
ml
Q
-
>
4!?
w
(ha ) asoa paqJosq~
,
A
C5
-
E
Y
o
m
,
.,
:
Y
,
10pm
I
I
,,
.
.
1.0
0.9
0.8
0.7
:
I
--
20
I
.-
40
I
--
60
Dose ( 106 Gray)
Figure
28
*
1’
m
N
al
L1
m
===%
-1
n
u)
n
n
‘.
--
Countsper Pixel (x 1000)
0s
0
23
I
I
Concentration
(ppm)
w
0
I
=
WI.
0
Cn
o
A
4
0 0
o
0
&
m
o
!s
0
E
0
03
0
(i
Ld
o
I
1
!3
o
Iv
uc!!cl
o
0
*-,
H
.
.
0
.
0
Cs
o
0
0
00
,
8X104
E
o
L
o
80001
F
600CI
L
4000
h
1-
2000~
0
0
1
2
3
4
1
2
Distance (mm)
3
4
600
400
H
200
n
‘“o
\
Figure
32
. ,—,
I
Figure
33
(a)
Reference
Electrode
Counter
Solution
Electrode
z
8000
I
6000
I
I
I
I
(b)
7000 .
Ce
~
I
x
.
h
White
X-ray ~
Beam
Fluorescent
X-ray
~Thin
Plastic
Metal Foil
Working
Electrode
i
3000
200C
I
o
I
I
0.50
Distance
I
I
I
I
1.50
1.00
Above S.S. (mm)
I
2.00
Figure
34
I
.
0,0030
00025
0.0020
0.00”5
().00”0
0.0005
00000
Figure
35
0
0
C9
Figure
38
40
X103
35
Fe Foil
.
30
25
‘G 20
15
.
10 .
5 .
7000
7025
7150
7175
Energy (eV)
Figure
39
ChromiumK XANES
1.2
l“’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’’”
1.0
7
‘-
;*>.,
,’ #’
s
,
:
‘,
*
#
,
!’,’
,’
#
#
‘.
‘.
/
~
3
0.8 “-
o
v
.&
~
.,,
‘ . . .. -----
.. . .... .
““-::.\,*
-.
,.’f
0.6 “-
a
c
?
u
‘..
/,
~’
-0a)
.
●
,’ ,’
t
.,
/
,.’,’
,’
/
~-’~
0.4 “-
-1()
‘.
15555
–--–
H%TERITE
1
(syn. }
,;7
,J,
,
.’
.’ ,/
0.2 “-
0.0
‘.
PYROXENE
-----------
OLIVINE
..+?
‘.
I I i
o
*9
I I 11I 1I 11I 1I I I I I I I I 1I 11I t I I t I 1I I I I I I
40
20
30
10
ENERGY ( eV relative to 5989 eV )
Figure
40
I
1’
0,
.
0.
0.14
0.12
0.10
0.08
.
0.06
0.04
.
0.02
0.00
5920
I
1
I
I
1
5980
Energy (eV)
i
I
1
6010
I
1.0
F“’’’’’
0.0 1
’’’’’’’’’’’’hi
1
6520
6560
6600
Energy (eV)
Figure
42
Swdation
d Qm%amirwnts k Soil at the micron Jevel
of Cr
.
/M
aoo$
!
I
Colmrsn!$sec
k-
\
●
●
.
I-J”
0.002
O.ouQJ
---
eulo
elm
-W
i
-1
@m
Cr CW.Iy- l-2 mkmIPiIIxy
XIW shows Cr associationwith
Fe mineral
. p-xAs showspresenceof Cr(vl)
Figure
43
u
u
a
0
0
s
zm
&!
J81181 11111, ,,111111111141
0
!1111111
00
00
( yi
) WV
11111811111
0UJWS
.
.
.,-> *
..
I
124.0
I
Wavelength (rim)
12.4 1.24 0.1240.0124
I
I
{
124.0
,.24
I
Wavelength (rim)
12.4 1.24 0.1240.0124
1
1
,.22
s
.
g
‘b
E
s
.
1020
0
t
,020
DESY
TIF-FEL
..-” -0-●.-
m,.18
SJ
1
E
..- “-- ””....
E
spent.
I
~
10’6
LCLS’--
spent.
t
i
1010
CU-K
Mo-K
_
C-K
g108
W
g
%
‘T
w
:“:
;ss’7
c
E
UJ
%
75
=
I
*---
10610-3
PI-K
CU-L
1
1
,0-*
10-1
100
I
Bremsstrahlung-
101
a
Photon Energy (keV)
102 103
,.6
/
10-3
I
I
1O-*
10-1
I
100
I
lot
1
102
I
103
Photon Energy (keV)
Figure
45