12)X-ray Spectrometry– HVC Capacitor, HV Ceramic Capacitor to build All kinds of X-ray machine.
For the purpose of light element analysis of microscopic
particles, the samples are collected on conductive substrates and
measured without conductive coating. The usually short measur-
ing times for computer-controlled single-particle analysis help to
avoid a large charge buildup on coarse ( d > 2 ím) particles.
However, automated particle analysis by EPMA raises two
practical challenges: the accuracy of the particle recognition and
the reliability of the quantitative analysis, especially for micrometer-
sized particles with low atomic number contents. Choe¨l et al. ( G9)
proposed the combination of using B as the substrate material
and applying a reverse MC quantification procedure for fully
automated elemental analysis of several thousands of micro-
particles. The automatic analysis was tested on standard particles
ranging from 0.25 to 10 ím. Quantitative determination of low-Z
elements in microparticles was achievable using B as a substrate
material, and highly accurate results could be obtained using the
proposed automatic data processing compared to conventional
ZAF methods. The environmental interpretation of elemental data
for a large number of individual aerosol particles is more relevant
but often becomes more complicated when concentration values
are available for low- Z elements as well. Yuan et al. ( G10 ) pointed
out that internally mixed particles often cannot be classified to
groups separated by clear borders. In such cases, much useful
information can be lost by the unsupervised use of traditional
statistical techniques, such as factor analysis and cluster analysis.
By revealing the multiple functional relationships between the
elements, a graphical technique that uses binary and ternary plots
offers more insight into the groups of particles and the transitions
between them. However, the use of a graphical method for data
interpretation is very time-consuming for the operator.
For this reason, the development of a method that fully extracts
chemical information contained in the elemental concentration
data of large numbers of individual particles is important. An
expert system that can rapidly and reliably perform chemical
speciation from the elemental concentration data obtained by
single-particle, low-Z EPMA was presented by Ro et al. (G11 ).
Using a training data set of expected chemical species in
atmospheric aerosol, the expert system became capable of
calculating molecular fractions of the chemical species present
in each individual particle. The expert system is also capable of
classifying particles based on the constituting major chemical
species. Its feasibility was confirmed on elemental concentration
data collected for various types of standard particles and a real
atmospheric aerosol sample.
EPMA of porous, highly divided materials such as heteroge-
neous catalysts, by applying the traditional ZAF or æ(Fz) quanti-
fication procedures (calculating oxygen concentration from stoi-
chiometry), generally yields a systematic deficit in metal concen-
trations. The physical effects possibly responsible for this signal
loss were investigated by Sorbier et al. ( G12 ) using MC simula-
tions. An MC program was written that uses the PENELOPE
package and takes into account porosity, roughness, and energy
losses at interfaces, as well as charging effects. The authors
compared the simulated results with measurements obtained on
a mesoporous alumina. Their simulation results showed that none
of the four mentioned physical effects is responsible for the
observed signal loss. Using X-ray intensities of C and O measured
by WDX instead of stoichiometric considerations, the signal loss
could be explained by a composition effect due t o a C contamina-
tion brought by the sample preparation. Because of the high
specific surface of the porous sample, even a very thin surface
contamination layer can cause large quantification errors if the
contaminant is not analyzed. Light element analysis is therefore
very important for this kind of study.
X-ray microanalysis using transmission electron microscopy
(TEM) offers the possibility to perform quantitative analysis with
a spatial resolution in the nanometer range. Boon and Bastin
(G13 ) reported on an adaptation of the æ(Fz) quantification method
widely used for bulk EPMA measurements to thin-film analysis
in TEM. The advantage of the æ(Fz) method over the thin-film
approximation is that it does not require thin-film standards.
However, the measurement of bulk standards in the TEM requires
orders of magnitude lower beam currents; therefore, an accurate
measurement of the beam current is necessary. The validity of
the æ(Fz) model under TEM conditions was checked by perform-
ing bulk analyses on AlNi and AlTi samples and thin-film analyses
on an AlNi TEM specimen. The authors showed the applicability
of the æ(Fz) method to line scan analysis on a GaAs specimen
with GaAs/Ti interface. Simultaneous composition and sample
thickness information could be derived using only bulk standards.
Benhayoune et al. (G14 ) proposed a method for thickness or
composition determination by EPMA of thin films sitting on a bulk
substrate based on the fundamental expressions of Auger electron
spectrometry. When the detected element is only present in the
very thin coating, it is sufficient to use the surface value of the
depth distribution of the X-ray productionæ(0). The method can
take into account variations in substrate composition without
knowledge of the local substrate composition, based on the
measured backscattering coefficient. The analytical expressions
for æ(0) include the secondary fluorescence effect caused by
characteristic and Bremsstrahlung radiation originating from the
substrate. This method was successfully applied to thickness maps
and concentration maps of heterogeneous coatings based on an
iterative algorithm.
The use of environmental scanning electron microscope
(ESEM) with EDX detector is promising for the examination of
nonconductive samples, however, with difficulties of quantitative
chemical analysis. Carlton et al. ( G15 ) described a sample surface
charge neutralization using a path to ground near the specimen
surface. The procedure involves setting the current flow in the
ground path device to zero by adjusting the voltage to the gaseous
secondary electron detector. To minimize the skirt effect, the gas
pressure of the sample chamber was 1 -2 Torr. The authors
showed the possibility of quantitative X-ray analysis in the ESEM
using the proposed surface charge neutralization scheme, with
relative errors within 5%.
For use as a portable measurement technique closely related
to EPMA, Mainardi et al. ( G16 ) tested the capabilities of X-ray
analysis using excitation with high-energy â particles (10
2
-10
3
keV), allowing the ionization of K shells of high-Z elements. They
named the excitation procedure as BIXE (â-induced X-ray emis-
sion). Using a transparent source arrangement with a thin
â-emitting foil and an HPGe detector, a minimum detection limit
in the weight percent range could be reached. Since the energy
spectrum of â-particles used for excitation is polychromatic, the
authors thoroughly tested the calibration curves using binary and
multielemental metallic foils in the atomic number range of 22-
75. Applying the ZAF corrections usually used for EPMA,
semiquantitative results could be obtained. Further developments
of BIXE are needed in order to obtain a low-cost technique suitable
for in situ studies.