Introduction To EDS Analysis

Introduction to
EDS analysis
Reference Manual
Bruker Nano GmbH
Schwarzschildstr. 12
12489 Berlin
Tel. +49 (30) 670 990-0

Fax +49 (30) 670 990-30
E-mail service-berlin@bruker-nano.de
Internet www.bruker.com
The reproduction, transmission or use of this document or its contents is not permitted without express
written authority. Offenders will be liable for damages. All rights reserved.
We have checked the contents of this manual for agreement with the hardware and software described.
Since deviations cannot be precluded entirely, we cannot guarantee full agreement. However, the data in
this manual are reviewed regularly and any necessary corrections are included in subsequent editions.
Suggestions for improvement are welcome.
Order no. DOC-M82-EXX053.

© 2011 Bruker Nano GmbH, Berlin, Germany.
All trademarks and registered trademarks are the sole property of their respective owners.
Printed in the Federal Republic of Germany.
Contents QUANTAX
Contents
1 Introduction .................................................................................................................................. 5
2 Microanalysis Basics .................................................................................................................... 7

2.1 Overview ...................................................................................................................... 9
2.2 Quantitative Analysis .................................................................................................. 10
2.2.1 True Standardless Analysis with P/B-ZAF ..................................................... 10
2.2.2 Standard-based Analysis ............................................................................... 11
2.2.3 Reference-based Analysis ............................................................................. 12
2.2.4 Standard-based P/B-ZAF-Analysis ................................................................. 12
2.2.5 Combined Analysis Methods ........................................................................ 13
2.3 Imaging Functions ...................................................................................................... 13
2.3.1 Standard Image Capture................................................................................ 13
2.3.2 ColorScan ...................................................................................................... 14
2.3.3 Mapping and Element Imaging .................................................................... 14
2.3.4 Hyperspectral Mapping ................................................................................. 15
2.4 Object Analysis, Multipoint, and Line scan ................................................................ 15
2.4.1 Object Analysis .............................................................................................. 15
2.4.2 Multipoint Analysis ........................................................................................ 15
2.4.3 Line Scan ....................................................................................................... 16
2.5 Analysis Limits ........................................................................................................... 16
2.5.1 The Analysis Volume ..................................................................................... 16
2.5.2 Spatial Resolution .......................................................................................... 17
2.5.3 Analyzing Inhomogeneous Samples ............................................................. 17
2.5.4 Sensitivity of X-ray Microanalysis .................................................................. 18
2.5.5 Limit of Detection .......................................................................................... 18
2.5.6 Accuracy and Precision ................................................................................. 18
2.5.7 Special Consideration for TEM Analysis........................................................ 19
3 Performing X-ray Microanalysis ............................................................................................... 23

3.1 Preparing Microanalysis Samples .............................................................................. 25
3.1.1 Embedding and Contacting ........................................................................... 25
3.1.2 Grinding and Polishing ................................................................................... 25
3.1.3 Coating Samples ........................................................................................... 25
3.1.4 Sample Mounting .......................................................................................... 26
3.2 Setting up the Microscope ......................................................................................... 26
3.3 Microanalyzer Settings ............................................................................................... 28
3.3.1 Image System Settings ................................................................................. 28
3.3.2 Detector Position ........................................................................................... 28
3.3.3 X-ray Spectrometer Functions ....................................................................... 29
3.3.4 Parameter Data Transfer ............................................................................... 29
3.4 System Calibration ..................................................................................................... 30
3.5 Avoiding Analysis Errors............................................................................................. 31
3.5.1 Checking Sample State ................................................................................. 31
3.5.2 Maintaining System Calibration ..................................................................... 31
3.5.3 Understanding Spurious Peaks ..................................................................... 32
3.5.4 Avoiding Carbon Contamination .................................................................... 32
3.5.5 Detecting Inadequate Settings ...................................................................... 33
3.5.6 Using the Right Analysis Method .................................................................. 34
Appendix ................................................................................................................................... 37
3
User Manual Contents
History of Energy Dispersive X-ray Microanalysis ...................................................... 39

The Sources of X-rays................................................................................................. 40
Absorption and Fluorescence ..................................................................................... 42
Semiconductor Detector Effects ................................................................................ 44
Conventional Quantitative Analysis Procedures ......................................................... 46
P/B-ZAF Spectrum Analysis Background.................................................................... 46
Further Reading .......................................................................................................... 50
Index .................................................................................................................................... 51
4
1 Introduction
This manual offers an introduction to the basic principles of energy dispersive
spectrometry (EDS). The main goal is to provide some general information on the analysis
possibilities when using an EDS system and to summarize some background knowledge
that will help to optimize analysis results.
The first chapter deals with different strategies for quantitative analysis, from true
standardless to standard-based quantification, as well as combined analysis methods.
This is followed by a description of the typical imaging functions of an X-ray microanalysis
system, together with information on the acquisition of element distribution images,
hyperspectral maps and line scans. A discussion on the EDS analysis limits rounds off this
section.
The second chapter is dedicated to the explanation of different factors that have to be
taken into consideration when performing practical X-ray microanalysis. These factors
inlcude sample preparation, setup of the electron microscope, spectrometer settings and
system calibration. A list of frequent analysis errors and the way to avoid them closes this
part of the manual.
Finally, the appendix contains some general EDS information including topics such as the
history of energy dispersive X-ray microanalysis, absoprtion and fluorescence, detector
effects, etc. in form of an extended glossary.
2 Microanalysis Basics
2.1 Overview ...................................................................................................................... 9

2.2 Quantitative Analysis .................................................................................................. 10
2.3 Imaging Functions ...................................................................................................... 13
2.4 Object Analysis, Multipoint, and Line scan ................................................................ 15
2.5 Analysis Limits ........................................................................................................... 16
User Manual 2 Microanalysis Basics
8
2.1 Overview QUANTAX
2.1 Overview
X-ray generation. Scanning electron microscopes
(SEM) or similar devices scan the specimen with a
beam of high-energy electrons. By interaction of
these electrons with the atoms of the specimen,
secondary electrons, backscattered electrons,
characteristic X-ray photons, and bremsstrahlung are
generated. While the different electron detectors of
the microscope (SE, BSE) deliver images of the
specimen - revealing surface structure and topology -
the X-ray spectrometer adds detailed information
about the chemical composition for a given measure-
ment spot or area.
Qualitative analysis. The characteristic X-rays

consist of narrow emission lines, which are
characteristic for the chemical elements contained in
the sample. The energy of these lines is nearly
Fig. 2.1-1 Typical analysis geometry independent of the chemical bonding state of the
affected atoms; the electron probe microanalysis
(EPMA) is element sensitive. In the resulting energy
dispersive spectrum the characteristic X-rays
correspond to visible peaks. Identifying these peaks
yields information about the elements that are
present in the sample.
General spectrum analysis steps Quantitative analysis. Quantitative data about the
sample composition is derived from the different
 Spectrum acquisition peak intensities by an extensive mathematical
 Correction of the detector effects process, often referred to as ZAF or PhiRhoZ matrix
(Escape, Shelf, Tail, Shift) correction. Bremsstrahlung X-rays, which form a
continuous spectrum background, are used as
 Identification of the elements and
selection of a line series for each additional source of information during the spectrum
element analysis with self-calibrating methods (P/B-ZAF, see
2.2.1).
 Calculation of the bremsstrahlung
background radiation (e.g.
bremsstrahlung) Different approaches with and without reference to
standards are commonly used and supported by the
 Deconvolution of overlapping peaks
QUANTAX software. They are described in more
and estimation of the net intensities
detail in the next sections.
 Calculation of concentrations with
or without reference to standards
Automatic and interactive operation. Fully
 Result formatting and presentation automatic spectrum analysis is used in routine
analysis in industry as well as in many laboratories.
The input necessary for controlling the analysis is
contained in filed method settings, provided by the
vendor or user.
In complicated cases the expertise of the operator

may be required.
9
Thereby complete interactive control over critical

analysis steps helps expanding the methodical limits.
Position referred methods. Combining electron

beam scanning and X-ray spectrometry allows -
besides classical qualitative and quantitative point
analysis - analysis of randomly shaped objects,
automatic multipoint analysis, recoding of
concentration profiles, as well as advanced imaging
methods like mapping, and element imaging. The
necessary access to the electron beam deflection is
normally accomplished by a secondary scan system
(external scan generator).
2.2 Quantitative Analysis
2.2.1 True Standardless Standardless analysis. Standardless electron probe

Analysis with P/B-ZAF microanalysis provides quantitative composition data
from evaluating a recorded X-ray spectrum by means
of fundamental physical formulas and extensive
atomic databases. Semi-standardless analysis, which
is still being offered sometimes, substitutes part of
these fundamental data and relations with spectral
references provided by the manufacturer.
QUANTAX analysis strategies
Modern standardless analysis is sufficiently accurate
 P/B-ZAF (true standardless) and reliable for a wide range of applications in
 P/B-ZAF (standard-based) research, education, environmental protection,
production, and many other fields. The ease of use
 P/B-ZAF (direct reference)
and reliability of standardless electron probe
 PhiRhoZ (standard-based) microanalysis are unique features, which, among
 PhiRhoZ (direct reference) others, account for the wide acceptance of this
analysis method.
 Film analysis
PB-ZAF. P/B-ZAF standardless analysis is a
fundamental parameter (FP) based true standardless
and self-calibrating spectrum analysis procedure,
based on modified ZAF matrix correction formulas.
Absolute concentration values are obtained without
the use of explicit or implicit standards and without
system factor calibration.
Using the P/B-ZAF algorithm the characteristic X-ray

intensities (i.e. the net peak areas) are calculated in
relation to the mean level of the simultaneously
recorded bremsstrahlung background. Therefore the
general system factors (electron beam current, solid
angle) and a number of adverse influences cancel
out each other and therefore do not influence the
result of quantification. The theory of the P/B-ZAF
10
2.2 Quantitative Analysis QUANTAX
algorithm is discussed in the Appendix.
Internal reference. The internal reference to the

bremsstrahlung is the key factor of P/B-ZAF analysis.
It compensates effects resulting from absorbing
surface layers and changes of effective take off
angles. This successfully allows the analysis of
unprocessed, rough surfaces.
Analysis conditions. The bremsstrahlung contains

information on variations of detection conditions
which are difficult to handle, for example excitation
energies and local surface tilt. Deviations of the
effective excitation energy, e.g. caused by sample
charging, can be taken into account by interactively
or automatically evaluating the higher energy
bremsstrahlung distribution (Duane Hunt Limit).
Local tilt angles can be evaluated and corrected by
fitting the low energy side of the spectral
background.
Automatic and interactive operation. P/B-ZAF

standardless analysis can be done automatically or in
interactive mode. In either case all information
needed for quantification is obtained solely from the
spectrum recorded by the X-ray spectrometer. No
reference measurement or determination of system
factors is necessary using standardless P/B-ZAF
analysis.
2.2.2 Standard-based Standard-based analysis is traditionally preferred in

Analysis certain cases. In well-controlled applications
standard-based electron probe microanalysis yields
high precision results that cannot be achieved with
any other analytical method likewise. Inevitable
prerequisites are well-prepared, i.e. embedded and
polished samples, high-class standard materials, very
stable analytical SEMs or microprobes, and well-
trained personnel.
Standards libraries. Data of standards for all

elements to be analyzed by standard-based analysis
is normally maintained in a so-called standards
library. State-of-the-art standard-based analysis is
highly automated, that means managing of numerical
entries like net peak area, k-Ratios or other
parameters is not required from the user any more.
Normalizing the system. With samples prepared

accordingly and an appropriate standards library at
hand, the only additional requirement for PhiRhoZ
based quantitative analysis - in contrast to
11
standardless analysis - is to measure a system

factor, which is mainly determined by the chosen
electron beam current, the detector size and
detector position. Occasionally performing a
calibration measurement with a proposed calibration
sample or the input of a so-called beam current
factor - derived by measuring the electron beam
current with a Faraday cup – is sufficient.
Accuracy. In most cases systematical errors of

about 1% can be achieved even when using pure
element standards or simple compounds. This
presupposes tightly controlled excitation conditions,
smooth polished samples, standards with matching
surface conditions and - if necessary – coatings.
2.2.3 Reference-based A special case of standard-based analysis is the

Analysis direct reference measurement, - for instance -oftenly
used in production control within metallurgical
plants. In this case a stable composition of a certain
product is verified by comparing it with certified
samples of the same type.
Current References. Reference-based analysis uses

the same principles as methods based on standards
libraries. The only difference is that the pre-defined
standards library is replaced by current reference
measurements using an available standard or a
reference sample. If the unknown specimen is
measured right after the corresponding reference
sample and using the same settings, a number of
adverse influence factors will be avoided.
If the current reference is of same type as the

unknown sample and all interesting elements can be
related to known elements within the reference
sample, no standard library has to be used. The
system normalization can be omitted and
quantification can yield good results even with
critical specimens, like powder samples or ceramics.
2.2.4 Standard-based P/B- This way, the remarkable insensitivity of P/B-ZAF

ZAF-Analysis analysis regarding measurement conditions is
maintained thus providing all advantages of standard-
based or direct reference analysis strategies.
QUANTAX integrates this method together with a
flexible library manager and reference administration.
Using the self-calibrating P/B-ZAF standard-based

analysis no system calibration measurement has to
be performed.
12
2.3 Imaging Functions QUANTAX
As with any standard-based method, analysis results

will be most accurate when measurement
conditions, compositions and sample states of
reference samples match those of the current
measurement. However, with P/B-ZAF standard-
based analysis these requirements are substantially
reduced, achieving nearly the simplicity of
standardless analysis.
2.2.5 Combined Analysis Combined analysis provides the possibility to choose

Methods the appropriate analysis method on an element-by-
element basis. Therefore it may be sufficient to
provide standards only for certain critical elements,
while for all other - bulk or minor - components
standardless analysis is adequate. This substantially
simplifies the standard-based analysis. Combined
analysis is also valuable in cases when standards are
hard or impossible to get.
2.3 Imaging Functions
2.3.1 Standard Image Image capture. State-of-the-art microanalyzers are

Capture equipped with means for capturing the electron
image additionally to the image display on the
console of the electron microscope. This is useful to
ease documentation and archiving. It provides
means for directly setting analysis points and regions
from the microanalyzer’s workspace.
 The QUANTAX scan system External scan system. When using advanced
is housed in the QUANTAX server analysis methods separate scan systems are
PC. typically used, allowing synchronization between
electron beam scanning and X-ray acquisition. To
ensure exact alignment of displayed image and
chosen analysis position the same external scan
system is also used for image capture.
The external scan system provides image resolution
and scan speed setting for microanalysis, which
work independently from normal image capture on
the microscope. Different noise reduction and slow
scan modes can be selected. Synchronizing of X-ray
data and beam positions is provided in all modes.
General control features, like magnification settings

and focusing, can normally still be controlled from
the microscope console.
13
2.3.2 ColorScan ColorScan is an enhanced acquisition method which

is provided additionally to standard image acquisition.
While normal imaging only utilizes the electron
detectors for capturing data, ColorScan also
 QUANTAX provides evaluates the output of the X-ray detector, allowing
ColorScan already for the primary indicative colors to be added to the image. This way
captured image, i.e. prior to an the different chemical compositions of phases,
element analysis or mapping.
features or particles throughout the imaged area are
indicated with different colors. ColorScan allows
detecting inhomogeneous regions in samples and
helps selecting adequate positions for analysis in
order to prevent severe analysis errors.
ColorScan requires no user input whatsoever.

Neither elements, X-ray window settings or
assignment of colors have to be selected. In
connection with the extraordinary pulse processing
power of the XFlash® detector, ColorScan can
produce high-definition color images in very short
capture time.
2.3.3 Mapping and Mapping. In X-ray microanalysis mapping means

Element Imaging that point-by-point an X-ray spectrum is accumulated
and evaluated to determine the content of certain
predefined chemical elements. Each resulting
element map represents the two-dimensional
concentration distribution of the regarding element
over the sample surface. According to the
preferences of the user, element maps are displayed
individually or color-coded and superimposed to form
one multi-color image.
Element mapping is an adequate procedure e.g. in
case of inhomogeneous specimens like metal alloys
or ceramics. EDS analyzers, in contrast to WDS
spectrometers, are generally able to generate maps
simultaneously for a great number of different
chemical elements.
Element images. An element image is a

topographic survey of the specimen surface
(captured by use of the electron detectors), which is
colored according to the local sample composition.
Like with the tool ColorScan, element images are
assembled by mixing the high-resolution electron
image with individual element maps. In this case
however, the user controls which element maps
shall contribute to the final image and adjusts the
image mixing parameters. This way high quality
images, ready for publishing, can be achieved
without manual post processing, marking or coloring.
14
2.4 Object Analysis, Multipoint, and Line scan QUANTAX
2.3.4 Hyperspectral Hyperspectral mapping (also known as Position

Mapping Tagged Spectrometry - PTS) is a data acquisition
method at which the acquired X-ray spectrum
information is stored tagged to the pixels of the
digitized sample image. In the QUANTAX system
this functionality is called HyperMap. The spectrum
database continually builds up during scanning of the
complete sample surface or a defined mapping area.
By post processing this kind of element maps, object
spectra, line scans, and other can be derived from
the database.
Like direct scanning position related analysis outlined

above, hyperspectral mapping is useful in connection
with inhomogeneous samples that require phase
reconstruction, feature recognition, multipoint
analysis or similar.
2.4 Object Analysis, Multipoint, and Line scan
2.4.1 Object Analysis With object analysis differently shaped areas of the
sample surface can be defined as area for analysis.
In contrast to point analysis, where the electron
beam remains steadily on one spot during spectrum
acquisition, object analysis lets the electron beam
constantly move over the whole object area while
the X-ray spectrum is acquired. This helps reducing
the strain on the sample surface and smoothes out
local impurities like dust particles or scratches.
Scanning of enlarged objects, however, is not
suitable for inhomogeneous or rough samples;
quantitative analysis of spectra averaged over
differently orientated or composed sample locations
normally will lead to erroneous results because of
non-linear spectrum mathematics.
2.4.2 Multipoint Analysis With multipoint analysis a number of separate spots

are defined on the sample surface for analysis. The
spectrum of each spot is evaluated separately by
means of batch processing. Different statistical
options for evaluating the results are provided.
Analysis results can be averaged to yield the mean
composition, or sorted into different categories
(phase analysis), naming only a few options.
Multipoint analysis can be combined with object
analysis to multi object analysis. Full quantitative line
scans and maps can be defined as special forms of
multipoint analysis too.
15
2.4.3 Line Scan During line scans, the electron beam scans the
sample continuously along a selected straight line.
Doing so, X-ray spectra are acquired and evaluated.
The resulting concentration profiles for selected
chemical elements are plotted in a diagram.
In contrast to element mapping or recording a full

spectral data base over the sample area
(hyperspectral mapping), line scanning is fast and
provides an intuitive display for quantitative data. It is
the standard method for examining cross sections of
functional layers, welding bonds, and much more.
Prerequisite for performing line scans is a separate

scan system of the microanalyzer or other adequate
means for synchronizing scan and spectrum
acquisition.
2.5 Analysis Limits
2.5.1 The Analysis Volume X-ray microanalysis is surface sensitive. The

information depth, i.e. the thickness of the surface
layer from which most of the analytic signal
originates, is defined on one hand by the ability of
the high-energy electron beam to penetrate into the
sample, on the other hand by the length of path the
generated X-rays can travel back through the
specimen. Since this path is substantially longer than
depths secondary and backscattered electrons can
penetrate, the information depth of microanalysis is
accordingly larger than that of SEM imaging.
Origin of the signals. The figure on the left side

shows an electron beam that hits a specimen
surface. The multi colored region is the volume that
is penetrated by the multiply scattered primary
electrons and excited to produce characteristic X-ray
radiation.
The yellow and red ranges are the volumes from

which electrons can escape from the surface.
Backscattered electrons are detected by the BSE-
detector of the electron microscope. Secondary
Fig. 2.5-1 Excitation and escaping volumes electrons, which have less energy than
backscattered electrons, originate mainly from the
yellow part (secondary electrons are the most
common signal source for SEM imaging).
16
2.5 Analysis Limits QUANTAX
The red and blue areas are the origin of most of the
characteristic X-rays and bremsstrahlung.
Depth distribution function. The information depth

spans from part of a micron to some microns
depending on the high voltage of the electron gun,
the mean density of the specimen matter, and the
element content. Calculated depth distribution
functions for a certain sample are available after
quantitative spectrum analysis (refer to the
QUANTAX user manual for more details).
Lateral scattering. In deeper regions the electrons

are scattered in lateral direction forming a pear
shaped volume in which atoms are excited to
produce X-rays. This lateral size in combination with
the depth distribution defines the analysis volume,
which is normally in the order of magnitude of 1µm³.
2.5.2 Spatial Resolution The three dimensional size of the analysis volume
limits the spatial resolution of X-ray microanalysis to
values much lower than the electron image
resolution of modern electron microscopes. At a
certain point the spatial resolution becomes
independent of decreasing spot size. In contrary,
intentionally defocusing the electron beam can be
used to decrease sensitivity to small artifacts.
Moderate magnification values are adequate for X-
ray mapping and similar applications.
As outlined above reducing the high voltage of the

electron gun enhances the spatial resolution.
Excitation using low energy characteristic X-ray lines
(L- and M-line series), which will travel only a short
distance in the sample, can further enhance lateral
resolution. But in this case detection of X-rays
becomes much more critical, and analysis will be
increasingly sensitive to surface contamination and
specimen condition.
2.5.3 Analyzing Inhomogeneous samples require special

Inhomogeneous consideration when using X-ray microanalysis. Point
analysis inadvertently performed on textured
Samples material will lead to random results.
Averaging over a larger sample surface (object

analysis) with inhomogeneous samples will generally
not result in mean concentration values. First of all
this is due to the fact that numerical two-dimensional
distribution of the constituents neither matches
volume contents nor mass concentrations. Secondly,
non-linear relationships of spectrum analysis prevent
17
simple averaging of spectra from being adequate.

The second reason also applies when averaging on
rough sample areas, even if individual point analysis
results are correct.
ColorScan, mapping and element imaging are means

to detect inhomogeneous samples or to find phases
and suitable spots for analysis. Automatic multipoint
analysis can be an adequate method to measure
mean concentrations or statistical distribution
parameters.
Microtextures. Special cases are samples with

microtextures in the range of the spatial resolution of
X-ray microanalysis. Here the measurement
conditions can only be improved by reducing the
analysis volume, by using low-kV excitation or
special methods like thin sectioning.
2.5.4 Sensitivity of X-ray Because of the small analysis volume X-ray

Microanalysis microanalysis is extraordinarily sensitive. Absolute
sensitivity in the order of 0.1 pg and well below can
be achieved in many cases. In connection with thin
sections - as used for TEM analysis - even total
amounts of 10-19 g can be detected. This
corresponds to only a few thousand atoms.
2.5.5 Limit of Detection Despite of the extraordinary low absolute detection

limits of X-ray microanalysis, the limit of detection
(LOD) of EDS in terms of concentration is restricted
by the bremsstrahlung background, which is always
present. In average and when using moderate
measurement times, the detection limit for trace
elements is roughly 0.1% mass concentration.
Of course this value is modulated by the matrix,
being lower within biological or geological matrices,
higher within heavy absorbing matrices. Optimum
analysis conditions and very long spectrum
acquisition times can shift the limits.
2.5.6 Accuracy and Precision and accuracy. The precision of

Precision quantitative EDS is mainly governed by counting
statistics of the discretely counted X-ray quanta.
Using fast X-ray detectors statistical errors well
below 1% can be easily achieved even at moderate
collecting times.
Accuracy is influenced by a great number of factors
ranging from the property of the sample itself, the
selection of elements to be analyzed, the possibility
of severe X-ray line overlap, to the selection of the
general quantification method. In average cases
advanced standardless X-ray microanalysis has
18
proven to be reliable and accurate, allowing error

levels as low as 3 to 5% relatively.
Basically, the accuracy of standard-based analysis is
only limited by the quality of standards and the
stability of the beam at the SEM. However, error
levels down to statistical limits require suitable
samples, skilful selection of standards, and optimum
operating conditions.
Fig. 2.5-2 Limit of detection for different X-ray
line series
2.5.7 Special Consideration Microanalysis with transmission electron

for TEM Analysis microscopes (TEM, STEM) is special in many
different ways. The most important are summarized
below.
High voltage. TEMs often use extremely high-

energy electron beams (100 keV and more) in order
to increase image resolution and penetration of the
sample. High excitation voltages give access to K-
line series radiation for almost all elements of
interest. On the other hand, high-energy electrons
and X-rays can produce a number of side effects that
are commonly not taken into consideration when
using standard microanalysis procedures.
Thin sections. TEM analysis is restricted to the

examination of ultra thin sections of a sample that
can be transmitted by electrons. Because of this
transmission of electrons, the formulas normally
used for calculating X-ray generation are not valid
anymore. Because of only marginal electron stray
processes within the sample the distribution of
bremsstrahlung becomes anisotropic and therefore
hard to predict.
Whether a sample can be classified to be thin or not
depends on the high voltage used and the
specimen’s material.
Detector position. TEMs are often not optimally

adapted for X-ray microanalysis and support only
horizontal detector ports. Tilting of the sample by
using a goniometer stage is necessary for X-ray
analysis in this case. Because of generally low beam
currents, optimum detector positioning is particular
important when using a TEM.
Supporting grid. TEM thin sections that are not

self-supporting are placed on supporting grids. The
supporting grid also serves abducting the heat
generated by the electron bombardment and is
therefore normally made of carbon-coated copper. X-
19
rays originating from the supporting grid do overlay

the measured spectra. Special consideration has to
be taken, when the supporting grid construction
material contains elements to be analyzed in the
sample.
Quantification. For standardless and standard-based

quantification of specimens at TEM a number of
special methods do exist. The analytical questions
depend on the type of sample.
The internal features of a microanalyzer can cover

only general tasks. Program options and offline
quantification by specialized software may apply for
e.g. biological samples. Standard-based offline
quantification normally uses so-called k-ratios as
interface. K-ratios are derived by comparing net peak
areas of the unknown sample to standards.
20
21
3 Performing X-ray Microanalysis
3.1 Preparing Microanalysis Samples .............................................................................. 25

3.2 Setting up the Microscope ......................................................................................... 26
3.3 Microanalyzer Settings ............................................................................................... 28
3.4 System Calibration ..................................................................................................... 30
3.5 Avoiding Analysis Errors............................................................................................. 31
User Manual 3 Performing X-ray Microanalysis
24
3.1 Preparing Microanalysis Samples QUANTAX
3.1 Preparing Microanalysis Samples
3.1.1 Embedding and In average cases, samples to be analyzed with

Contacting QUANTAX P/B-ZAF standardless analysis do not
require other preparation than normally required for
quality imaging. However, because microanalysis is
surface sensitive, clean and unspoiled surfaces are
prerequisites.
Samples for electron microscopy are usually

 QUANTAX interactive embedded and contacted either using conductive
spectrum analysis provides means for resin or silver ink as conductor. For microanalysis,
detecting and compensating sample
proper contacting is particularly important, on one
charging.
hand because of the relatively high electron beam
currents used, on the other hand because sample
charging would depreciate quantitative results.
Isolating or poorly conducting samples need

conductive coating (see section 3.1.3).
3.1.2 Grinding and In any case grinding and moderate polishing of

Polishing samples will improve precision of the quantitative
analysis because of the implied cleaning, removing
of oxide layers, and the well-defined local take off
angel (TOA). Flawless metallographic or petrographic
surfaces are not necessary if measurement spots
are carefully selected to avoid scratches or other
artifacts.
With standard comparison methods, normally used

for high precision applications, polishing is
mandatory unless the sample shows a
microscopically flat surface by nature. In either case
the procedure for embedding or mounting the
sample must prevent inadvertent surface tilt in
excess of about one degree.
Embedded samples are to be contacted using e.g.

conductive resin or silver ink conduction paths. For
poorly conducting or isolating samples see below.
3.1.3 Coating Samples Isolating samples. Isolating or poorly conductive

samples (glasses, minerals, ceramics, plastics) must
be conductively coated whatever analysis method is
used. The only alternative to coating such samples is
using the variable pressure (VP) mode that is
provided by certain SEMs. Variable pressure mode,
however, is complicated to handle from the
quantitative microanalysis point of view.
Carbon Coating. Carbon coating is a common
25
method and, because of the low X-ray absorption

factor of carbon, the first choice for microanalysis. A
variety of carbon coaters, working on the principle of
vaporizing or sputtering, are commercially available.
 For optimum standard-based The coating must be as thin as possible while being
analysis, the sample and the standard
homogeneous and contiguous. In praxis, coatings of
are coated at the same time.
10 - 30 nm are proved suitable.
An alternative coating material - e.g. aluminum - can

be considered in case carbon itself is to be analyzed
precisely. However, coating calibration may also
solve the problem.
Other coatings, like gold or platinum, may be better

for high-definition imaging, but the impact on
quantitative results in general will become
unpredictable.
For standard-based analysis coating of specimen and

standards must match tightly.
3.1.4 Sample Mounting For average purposes and if the SEM provides an
elevated detector port (detector elevation angle
≥30°), horizontal mounting of the sample is
adequate. In special cases where self-absorption of
X-rays within the specimen is the limiting factor,
tilting the sample towards the detector, 90° take off
angle (TOA) approximately, is recommended.
Tilting of sample, of course, is always necessary
with horizontal detector ports that are on the same
level as the specimen.
Submitting the correct geometry data to the
software is mandatory for successful quantitative
analysis.
The sample holder must be electrically connected to
the sample stage. When a sample current meter is
provided, it must be connected and in operation in
order to put the stage on ground potential.
Otherwise the stage is to be grounded via bypass.
3.2 Setting up the Microscope

For the purpose of X-ray microanalysis electron
microscope settings differ from what is normally
used for imaging purposes. Special consideration is
required with the settings summarized below.
Magnification. Although the microanalyzer can

capture high-resolution images, the finite X-ray
26
3.2 Setting up the Microscope QUANTAX
interaction volume (analysis volume) in the order of

magnitude of 1 µm³ (see also section 2.5.1) limits
the spatial resolution of microanalysis, depending on
high voltage (kV) and material density. Thus, useful
magnification with respect to object analysis, line
scan, or mapping, is limited normally to a few 1000x.
In order to perform point analysis without external
scanning the magnification can be set as high as
possible to reduce the area of the sample rastered
by the electron beam.
Working distance. In contrast to normal imaging,

the working distance (WD) cannot be freely selected
when performing X-ray analysis. The correct working
distance for microanalysis is a constant and depends
on the design of the X-ray detector and sample
chamber. The exact value is determined during
system installation. Only at the specified working
distance optimum detection efficiency and correct
quantitative results can be achieved.
Inadequate settings of the working distance can lead
to spurious peaks in the spectrum (e.g. silicon
fluorescence from the supporting grid of the
detector window), excess spectrum background by
malfunction of the magnetic electron trap, spectrum
distortion, and pulse loss. The calculated take off
angle is also only correct with the approved working
distance
High voltage. The high voltage should, as a rule of

thumb, be twice the energy (in kV or keV) of the
highest peak to be evaluated. In other words, the so-
called overvoltage ratio should not be less than two
for all elements of interest. Reduced voltages can be
used in connection with L- and M-series analysis to
increase spatial resolution. However, this is at the
expense of quantitative accuracy and higher
susceptibility to surface contamination.
Electron beam current. The current of the electron

beam should be set according to the desired pulse
density of the X-ray spectrometer (see also section
3.3.3). For high speed analysis, line scan, and
mapping and with respect to sample endurance and
pulse load capacity of the detector, the beam current
should be as high as possible.
27
3.3 Microanalyzer Settings
3.3.1 Image System During installation, image area, brightness, and

Settings contrast on the screen of the microanalyzer are set
to match the display on the SEM console as
precisely as possible. Parameters like focusing,
magnification setting, and image adjustment will be
controlled via the SEM console. The result can then
be observed on the SEM or microanalyzer display.
Image resolution. For the purpose of microanalysis

image resolution, scan speed, and acquisition mode
(line and frame averaging) can be adjusted
independently of the SEM settings. Settings can be
made within wide ranges according to the
preferences. High image resolution settings, of
course, will raise acquisition and processing times as
well as memory consumption drastically when
performing mapping and element imaging.
Scan mode. Noise suppression, which is often

necessary for high-definition imaging, reduces the
scanning speed. Mappings require even longer scan
times in most cases. The measurement time can be
increased, for example, through higher dwell time
per pixel (for slow scan modes it can even be several
seconds). In contrast to extreme slow scan, frame
averaging with several fast scans succeeding one
another will reduce the temporary sample strain
resulting from the electron bombardment.
However long frame-averaging times require

extreme stability of the scanning system in order to
prevent blurring of the image. Therefore line
averaging is normally the method of choice. Using
this mode a defined horizontal line is scanned
several times before proceeding to the next line.
3.3.2 Detector Position Most EDS-detectors can be retracted, since the

optimum position for microanalysis may interfere
when performing high-resolution imaging. For X-ray
spectrometry, line scan, and element mapping the
detector should always be positioned to the closest
position, just not touching the pole piece or the
sample. In this position the solid angle is at its
maximum and the lowest beam current is required
for the desired X-ray signal intensity.
In some cases the X-ray detector is provided with an

oblong orifice at the flange allowing shifting the
detector perpendicular to its axis. This is useful to
28
3.3 Microanalyzer Settings QUANTAX
optimize the correct working distance for EDS (see

section 3.2), which is normally done only during
installation.
3.3.3 X-ray Spectrometer Settings of the X-ray detector and spectroscopy

Functions signal processor influence the analysis performance.
With QUANTAX most settings are either automatic
or otherwise made during system installation,
freeing the user from further consideration. Detailed
discussion is added here only for the sake of
completeness.
Maximum Throughput. The type of X-ray detector

and setting of the spectroscopy signal processor
determine the maximum pulse throughput of the
spectrometer. The actual count rate (pulse rate) in a
given situation determines how long a spectrum is to
be accumulated in order to become statistically
relevant. Higher throughput values correspond to
reduced energy resolution and may lead to a rise of
the lower energy threshold.
The appropriate compromise between energy

 Bruker XFlash® detectors resolution and pulse throughput depends on the
have very high throughput capabilities measurement task: light element analysis will
and can be operated at optimum
require best energy resolution at low count rates,
energy resolution in most cases.
peaks in the upper energy range can be well
separated at moderate settings, while element
mapping will benefit from the highest throughput
setting available for the given spectrometry signal
processor.
Energy range. The maximum energy the

spectrometer can record should be chosen to
enclose all relevant parts of the spectrum, especially
the full bremsstrahlung background (the upper end
of the bremsstrahlung background yields information
about the effective excitation voltage). If, on the
opposite, the energy ranges are much larger than
necessary this may affect accuracy and resolution of
the energy scale.
Cooling mode. The thermoelectrically cooled

XFlash® detector provides different cooling modes:
maximum cooling will provide optimum energy
resolution, whereas thermostat mode may be
chosen if the stability of the system parameters is
predominant.
3.3.4 Parameter Data Detector parameters. Detector parameters (crystal

Transfer type, window type) are fixed by the system design
and installation. The correct definition within the
29
microanalysis software at the time of installation is

crucial for quantitative analysis.
Geometry parameters. The detector elevation and

azimuth angles are fixed depending on the design as
well. These parameters are also defined during
installation. The current tilt angle can only be
automatically transferred if tilting is performed by a
motorized goniometer stage. In other cases manual
input to the microanalyzer before starting the
spectrum acquisition is mandatory.
Microscope data. The values for high voltage,

magnification, working distance and - if provided -
sample tilt are often automatically transferred to the
microanalyzer. In any case, the user should check
correct data entry, because incorrect data will lead to
erroneous quantitative results or other malfunction.
3.4 System Calibration

Different levels of system calibration are provided.
The need for recalibration is determined by technical
conditions, the precision requirement of the
application, and - eventually - laboratory or legal
regulations.
Image calibration. The image calibration

recalculates the current magnification value reported
from the electron microscope to obtain the scaling
factors needed for the QUANTAX image capture
procedures. Graphical means are provided to update
the calibration from the QUANTAX system
environment. Image calibration influences scale bars
as well as measuring functions.
X-ray spectrometer calibration. The energy scale

of the EDS detector can be calibrated from the
QUANTAX system environment. To perform the
calibration a test sample of known composition is to
be provided for spectrum acquisition. Spectrum
acquisition, analysis and maintenance of calibration
data are fully automatic.
System factor calibration. For standard-based

analysis using the PhiRhoZ matrix correction a
system factor (or beam current factor) has to be
determined. The measurement is performed using a
specified calibration sample directly from the
workspaces. Beam current meters (Faraday cup) can
30
3.5 Avoiding Analysis Errors QUANTAX
be used instead.
3.5 Avoiding Analysis Errors
3.5.1 Checking Sample Cleaning. Samples for X-ray microanalysis shall

State always be clean and properly degreased. Organic
reminders can increase the carbon signal and
produce incorrect quantitative results.
Severe oxidation is to be removed by grinding,

polishing or edging. However the native oxide layer
on most metals is thin with respect to the
information depth of X-ray microanalysis at moderate
or high excitation voltages and can be tolerated with
normal requirements with respect to accuracy.
For sample preparation see also section 3.1.
Spot selection. Scratches, local impurities, or dust

are not problematic if the analysis spots are carefully
selected to avoid these artifacts. With PB-ZAF,
spectrum analysis requirements regarding surface
flatness are generally rather low.
If the inhomogeneities of samples are larger than the

X-ray spatial resolution (analysis volume), the
selection of proper analysis points is very important.
Simply averaging over different areas by object
analysis, defocusing of the electron beam, or free
scanning modes of the SEM will generally lead to
incorrect quantitative results. Manual or automatic
multipoint analysis can solve the problem (see also
section 2.5.3). Analysis of sample regions with
inhomogeneities in the region of (or just below) the
size of the analysis volume will generally lead to
reduced accuracy.
Contacting. Microanalysis samples must be

properly contacted to avoid charging up. For sample
coating see section 3.1.3. Also dust particles on the
sample may charge up and bend the electron beam,
even if the sample itself is sufficiently contacted.
Any instability of the SEM image, flicker or gutter
may be an indication of possible local charging up.
3.5.2 Maintaining System Incorrect calibration of the energy axis of the X-ray
Calibration spectrometer will affect auto identification of
chemical elements. Especially in cases where
deconvolution of overlapping X-ray lines is necessary
for precise quantitative results the exact energy
calibration is crucial.
With standard and reference-based quantitative
31
analysis (other than standard-based P/B-ZAF) the

calibration of the system factor directly influences
non-normalized results.
Correct image calibration of course is necessary for
all image related measurements.
3.5.3 Understanding Spurious or “false” peaks can be observed in

Spurious Peaks different situations. It is crucial not to misinterpret
them considering them characteristic X-rays of
elements contained in the sample.
Pile-up peaks. At high count rates residual pile-up

peaks at about the double energy value of strong
peaks may be observed. To check for pile-up reduce
the beam current by a factor of e.g. two and
compare the acquired spectra. All peaks whose
relative height with respect to the rest does drop by
about the same factor are probably pile-up peaks.
Escape peaks. Escape peaks are completely

removed by QUANTAX automatic spectrum
correction.
Silicon fluorescence. Silicon fluorescence peaks

(not originating from the sample) are mainly
observed with incorrect detector positioning or
wrong working distance. Misalignment of analysis
spot and detector axis can also result from beam
deflection at very low magnifications and the use of
electrical beam shift features. Analyzing at moderate
magnifications and using the sample stage to select
the analysis spots is recommended.
With proper adjustment false silicon peaks should
contribute well below 1% mass concentration to
quantitative results.
Stray radiation. Backscattered electrons can reach

the borders of the sample chamber and the pole
piece of the electron microscope. Unresolved iron
and chromium fluorescence peaks can be expected
to originate from stainless steel instrument
components. In certain cases stray radiation from
aluminum or brass sample holder may also lead to
spurious spectrum peaks.
3.5.4 Avoiding Carbon Carbon contamination. Carbon contamination,

Contamination which leads to false carbon peaks and excessive
absorption of X-rays, is inadvertently produced during
scanning in the SEM. Residual organic gases in the
vacuum chamber, possibly originating from
outgassing of the sample itself, are cracked by the
electron beam. Atomic carbon and other residues are
32
deposited on the sample. If no special technical

means of avoiding carbon contamination are at hand
(cryotrap, air jet), reduce the measurement time to
the indispensable limit. With homogeneous or
partially homogeneous samples enlarge the
measurement area e.g. by object analysis.
To check if carbon contamination is a problem at all,

perform a series of subsequent measurements on
the same sample location and compare the height of
the carbon peaks or the quantitative analysis results.
3.5.5 Detecting Inadequate A proper setup of the analysis parameters has been
Settings discussed in the sections above. Some common
errors will be summarized here:
Too short measurement time or low count rate.

Too short measurement times with respect to the
actual count rate leads to insufficient spectrum
statistics. Especially the self-calibrating P/B-ZAF
analysis requires relatively high overall counts (about
100 000) in order to become statistically stable.
Check the estimated errors, which are output during
quantitative spectrum analysis.
Too high count rates. Too high count rates can lead
to reduced energy resolution of the detector, which
may affect peak separation especially in the range of
light elements. Residual pile-up peaks will also rise
with high count rates.
Pay also attention to the lower energy threshold,
which may automatically rise with very high count
rates. Ultra high throughput settings of the
spectroscopy pulse processor generally lead to
threshold values that make carbon or even nitrogen
peaks inaccessible.
Too low overvoltage. Low settings of the high

voltage of the electron gun limit the useful X-ray
range in general. Analyzing characteristic X-rays at
overvoltage ratios below 2 leads to rising
dependence on the actual excitation energy. Check
the bremsstrahlung background in the recorded
spectra (Duane Hunt limit) for the effective energy;
especially with low overvoltage ratios the reading of
the SEM may prove not to be sufficiently accurate.
Too high accelerating voltage. High accelerating

voltages reduce the spatial resolution and increase
the information depth. Especially when analyzing
coatings and galvanic layers, too high accelerating
voltages can cause the excitation of the bulk
33
material, which may lead to wrong quantitative

results.
Wrong working distance. Wrong working distance

will lead to reduced detector efficiency by partially
shadowing the detector and furthermore to
distortions of the spectrum, e.g. by altered take off
angle. If necessary, check thoroughly which working
distance performs best (normally the one that yields
the highest count rate at otherwise equal
conditions).
Wrong parameter data. Always check system

settings and automatically transferred parameters.
Wrong spectrum parameters will lead to erroneous
quantitative results. Check spectrum parameters
especially with spectra imported from unfamiliar
sources.
3.5.6 Using the Right Unfortunately the right X-ray spectrum analysis
Analysis Method method that covers all different analysis situations
does not exist. Besides selecting point, multipoint, or
object mode and proper analysis parameters, also
the spectrum evaluation strategy influences the
accuracy and precision of the results.
Generally use true standardless analysis (P/B-ZAF) in

cases of unknown, unprocessed, or generally rough
samples. Use standard-based analysis only if reliable
standard databases, measured on the same
machine, are available.
Some complicated analysis tasks cannot be properly

handled by automatic spectrum analysis. In
ambiguous cases perform or correct element
identification interactively. When performing trace
element analysis select the elements that are not
found by auto identification manually. With
complicated samples (many contained elements)
adapt the background fitting ranges according to the
known peaks. Use interactive tilt angle correction
and correction of excitation voltage if applicable.
Select the result normalization appropriate for the

desired purpose. Especially with limited statistics
(e.g. during quantitative mapping), result
normalization can improve the precision drastically.
34
35
Appendix
History of Energy Dispersive X-ray Microanalysis...................................................... 39

The Sources of X-rays ................................................................................................ 40
Absorption and Fluorescence..................................................................................... 42
Semiconductor Detector Effects................................................................................ 44
Conventional Quantitative Analysis Procedures ........................................................ 46
P/B-ZAF Spectrum Analysis Background ................................................................... 46
User Manual Appendix
38
Appendix QUANTAX
EDS Background Knowledge
History of Energy Dispersive The invention of the energy dispersive X-ray

X-ray Microanalysis semiconductor detector (EDS) in the early sixties
opened a way to perform electron probe
microanalysis on virtually any existing electron
microscope. The main advantages of energy
dispersive detectors are their ability to
simultaneously span a wide energy range of x-rays,
the unique detection efficiency, and affordable costs.
In a short period of time, the existing physical

models for analysis by comparison to standards, the
conventional ZAF-matrix correction methods, were
adapted to energy dispersive spectrometry.
However, standard comparison methods did not
match the ease of use of energy dispersive
Fig. 3.5-1 First generation energy dispersive spectrometers that well since they are time
detector consuming and bound to extraordinary stable and
precise instrumentation.
Numerous approaches to standardless analysis by

fundamental parameter calculations - often backed
up by stored spectra libraries – failed to provide the
desired accuracy and reliablilty. This led to the
common reputation of standardless X-ray analysis as
“semi-quantitative”. Only the recognition of the so-
called bremsstrahlung as valuable source of
information rather than being only a nuisance
enabled the development of true standardless
quantitative EDS.
Using this information the P/B-ZAF approach

substantially improved repeatability and accuracy of
analysis results. Only using this method quantitative
analysis of rough surfaces und irregular shaped
particles became feasible.
Modern EDS microanalysis features detecting limits

of less than 10-12g for bulk samples to amazing 10-19g
in thin sections. EDS now is common in nearly every
field of science, medicine and technology.
New developments in the field of EDS were mainly

influenced by more and more powerful computers
and growing software complexity. However, the
improvement of the most critical part of the system,
the semiconductor detector-crystal, soon met its
physical limits and stagnated.
39
This situation changed when the silicon drift detector

(SDD) was invented. These chips were used for the
development of the new XFlash® detector
generation. These new detectors are maintenance
free, do not require extensive cooling means and
allow - because of their extremely fast operation- a
significant reduction of measurement time. The
detection speed is the key feature for element
imaging, a method that produces colorful images
Fig. 3.5-2 Modern 12-fold SDD
within measurement times not much longer than
ordinary image acquisition.
In 1979, during the Workshop on Energy Dispersive

X-ray analysis, K.F.J. Heinrich concluded:
“If solid-state detectors of adequate energy

resolution that do not require liquid-nitrogen cooling
were developed, this would considerably extend the
useful range of solid-state X-ray detection”.
About 25 years later, the proof was at hand.
The Sources of X-rays X-rays of two different sources are present in an

electron microscope: characteristic X-rays and
bremsstrahlung.
Primary ionization. Specimen atoms are ionized

when hit by high-energy electrons. When inner shell
electrons are to be released (e.g. from the K-shell, as
shown in Fig. 3.5-3) the kinetic energy of the
electron must be substantially higher than the critical
excitation energy Ec of the corresponding shell,
which is specific for every type of atom. The ratio
between the incident electron energy E0 and the
critical energy Ec is called overvoltage ratio U0= E0/Ec.
The probability of ionization referring to a certain
Fig. 3.5-3 Ionization and X-ray emission shell (also described as ionization cross section) has
its optimum at overvoltage ratios of 2 to 3.
The total count of ionizations caused by a defined

number of incident electrons is additionally
influenced by the backscatter coefficient and the
electron stopping power of the specimen matter.
These factors are summarized by the Z-factor of
classical ZAF matrix correction.
Characteristic X-rays. Subsequently following

Fig. 3.5-4 Ionization cross section with varying primary ionization an electron from an outer shell fills
overvoltage the vacancy releasing a characteristic X-ray quantum.
The quantum energy hereby equals exactly the
difference of excitation energies of the two shells
40
Appendix QUANTAX
involved, which are – at least for inner shells –

virtually independent of the chemical state of the
regarding atom.
Depending on which electron shell was holding the

primary vacancy, we talk about K-, L-, or M- X-ray line
series. Looking at the shell (or sub-shell) from which
the electron originates that fills the vacancy, different
lines can be distinguished within the groups
mentioned before. The common terminology,
invented by Siegbahn, reads Ka1, Ka2, Kß1, Ll, Mz ...
accordingly. Meanwhile more lines and sub-lines are
known than can be expressed by Siegbahn’s
empirical notification. All characteristic X-ray lines
can be classified using the official IUPAC notification,
simply stating the type of atom, the destination, and
source shell name (e.g. Pb M1N3).
Which X-ray lines can be used for analysis mainly

depends on the excitation energy available and the
Fig. 3.5-5 Selected transitions properties of the X-ray detector.
Moseley’s law. Moseley’s law (Moseley, 1913)

states a simple monotonic relationship between the
atomic number and the energy of a characteristic X-
ray line. Nowadays exact energy values are tabulated
in atomic databases and managed by software.
However, Moseley’s plot allows an intuitive
prediction of the probability of line overlap for the
different X-ray line series.
Typical intensity relations of characteristic X-ray lines,

Fig. 3.5-6 X-ray line energies according to within a certain series, are called relative emission
Moseley’s law
rates. Values are tabulated for all atoms.
Chemical shift. The energies of characteristic X-rays

and absorption edges that are in connection with
valence levels show some dependence on the
chemical state of the atom. These minor shifts are
barely visible to energy dispersive spectrometers
and therefore mostly negligible in EDX microanalysis.
The effects of chemical constitution on the fine
structure of absorption edges, however, give rise to
a number of advanced X-ray analysis methods (e.g.
EXAFS, XANES).
Fluorescence yield. With well-defined probability

when filling a vacancy in an inner shell the excess
energy is released by sending forth an outer electron
instead of an X-ray quantum. This is called
radiationless Auger transmission and reduces the
fluorescence yield, i.e. the measurable characteristic
41
X-rays. In the lower energy range the reduction is

substantial.
Bremsstrahlung. Bremsstrahlung (literally

deceleration radiation) is generated by deceleration
of high-energy electrons in the electric field of an
atomic nucleus. Thereby the electron looses a
substantial amount of energy which is released in
form of an X-ray quantum. Since any value of energy
loss - from zero to the full electron energy - is
possible, the bremsstrahlung quanta statistically
form a continuous spectrum background.
The primary bremsstrahlung distribution is a steady

function monotonically rising towards lower
energies. Because of absorption and self-absorption
the actually recorded bremsstrahlung background
has a more complicated shape and exhibits
discontinuities (edge jumps) that make fitting
complicated.
Bremsstrahlung generation is the main difference
between electron beam excitation and X-ray
fluorescence analysis (XRF).
Absorption and Absorption law. Absorption in the low and medium

Fluorescence X-ray energy range is basically determined by the
photoelectric effect, leading to fluorescence and
Auger electrons. The absorbing power of different
types of matter is described by energy dependent
mass absorption coefficients (MAC). These are
tabulated in atomic databases. The accuracy of the
MACs, which are generally submitted by a number
of different sources, is one of the main quality
factors of quantitative spectrum analysis software.
The absorption process follows the Lambert-Beer

law quite concordantly, which states an inverse
exponential relationship between transmitted X-rays
on the one hand and the product of MAC and length
Fig. 3.5-7 Soft X-ray window transmission of path on the other hand.
Detector window effects. X-rays must pass through

the radiation entrance window of the detector as
well as the functional and “dead” layers on the
detector surface before entering the active volume
of the detector. Thereby they undergo energy
dependent attenuation according to the mass
absorption coefficient and the effective thickness of
these different layers. The absorption curves exhibit
sharp edges (absorption edges) at energy values
corresponding to the critical energies of the window
material. Window absorption is the main reason for
42
Appendix QUANTAX
non-unity quantum efficiency of X-ray detectors in

the low energy range. This limits the detection of
low energy X-ray lines.
Two different kinds of detector windows are

commonly offered: beryllium windows for detection
beginning from sodium, and light element windows,
which can transmit boron (or even beryllium) K
radiation.
Self-absorption. Attenuation of the primary

generated X-rays (characteristic X-rays and
bremsstrahlung) also occurs within the specimen by
self-absorption. Self-absorption is one of the reasons
for the complicated non-linear relationship between
specimen composition and detected X-ray
intensities, which is a subject for matrix correction
routines. The effect of self-absorption is described
Fig. 3.5-8 X-ray absorption on rough surfaces
by the A-factor in the classical ZAF approach. Self-
absorption drastically increases at lower take-off
angles (TOA). Therefore take off angles below 30°
should be avoided for quantitative spectrum analysis.
Rough or contaminated surfaces are the main
problem for exact quantitative analysis using X-rays.
The reason for this is the unknown length of the
absorption path for a certain location in the sample.
That means self-absorption cannot be calculated
using geometrical considerations only. This problem
is substantially diminished by advanced P/B-ZAF
analysis (see below in this appendix).
Fluorescence enhancement. In photon induced X-

ray spectrometry (XRF) X-ray fluorescence is the
primary signal source. With EDS analysis secondary
fluorescence, caused by primary characteristic X-rays
and bremsstrahlung, is considered to be an effect of
second order normally enhancing characteristic X-
rays by only some percent. In some special material
compositions the characteristic X-rays of one
component tightly matches the absorption edge of
another. In these cases fluorescence enhancement
can rise to about 30%. Fluorescence enhancement
is accounted for by the F-factor in classical ZAF
matrix correction.
In case of inhomogeneous samples, especially when
small intrusions of light elements are in a heavy
matrix (e.g. the carbon content of spheroidal
graphite), secondary fluorescence may need special
consideration in order to avoid analysis errors.
43
Semiconductor Detector Si(Li) detectors. The classical energy dispersive X-

Effects ray detector build on of a bulk silicon semiconductor
crystal which exhibits an inert volume as X-ray
sensitive element. To maintain the inert volume the
crystal is drifted with lithium, that is why the name
Si(Li) is used. However, impurity compensation is far
from ideal. The detector, as well as the first amplifier
stage (field effect transistor), has to be cooled with
liquid nitrogen.
An X-ray quantum, absorbed in the volume of the

detector, causes a cloud of electrons and holes to be
set free by successive ionizations of the detector
material. Since this inert volume is virtually free from
other charge carriers, the electrons and holes
generated can be separated by means of an electric
field. Charge carriers traveling to the electrodes
located on both sides of the detector cause a
displacement current that is being measured by a
Fig. 3.5-9 Schematic drawing of Si(Li) detector charge sensitive amplifier. The amplifier outputs a
signal pulse, approximately proportional to the
energy of the corresponding X-ray quantum.
Because of inevitable electronic noise and the

statistical nature of charge generation the
characteristic X-ray lines, very narrow by nature, are
broadened in an energy dispersive collected X-ray
spectrum. The broadening described by the full
width of the peak measured at half maximum
(FWHM) depends on the energy of the quantum, in a
manner expressed by the Fano-factor.
XFlash® Detectors. The Bruker XFlash® detector is

an energy dispersive X-ray detector based on the
silicon drift detector (SDD) principle. A monolithically
integrated on-chip FET acts as signal amplifier and
allows unprecedented energy resolution.
The principle of sideward depletion of the active

detector volume in connection with the integrated
drift structure provides an extremely small detector
Fig. 3.5-10 Schematic drawing of SDD detector capacitance. This enables use of fast signal
processing techniques. The pulse throughput of
XFlash® spectrometers is by more than an order of
magnitude higher than what energy dispersive
systems can provide.
44
Appendix QUANTAX
The detector crystal is moderately cooled by

vibration free thermo-electric coolers; any heat
generated is dissipated by unforced convection with
no needs for external cooling means.
A super thin radiation entrance window separates

the sensitive detector area from the ambient
atmosphere and guarantees good transmission of
Fig. 3.5-11 SDD chip layout the X-rays of interest.
EDS-Detector artifacts. Due to the complex

interaction processes within the detector crystal the
pulse height distribution recorded in the multi
channel analyzer (MCA) shows a number of artifacts.
The most prominent effect - line broadening - can
clearly be observed on the screen and is tackled by
so-called deconvolution or peak fitting procedures
during spectrum analysis.
The main effects of second order are summarized

below. Nowadays they are physically well
understood and automatically compensated or taken
into account by state-of-the-art spectrometers and
microanalyzers.
Escape peaks. With a certain probability

fluorescent X-rays escape the silicon
detector material and thereby subtract
themselves from the signals measured.
Without Escape correction small peaks of an
energy level of 1.75 keV lower than the
signal peaks can be observed in the
spectrum.
Shelf. Shelf is a residual background continuum,

overlaid to bremsstrahlung, caused by
Fig. 3.5-12 Tail and shelf (uncorrected) incomplete charge collection (ICC) within the
detector.
Tail. Tailing is of the same nature as shelf, and

leads – if uncorrected - to slightly
asymmetric peak shapes.
Shift. Some Si(Li) detectors exhibit minor peak

shifts in the range of 200 ... 500 eV. These
are compensated during spectra collection.
Pile-up. Pile-up peaks can be observed at higher

count rates. They appear at about the double
energy value of prominent peaks. The height
of pile-up peaks rise with the count rate. A
special hardware (pile-up rejector)
45
suppresses pile-up peaks as far as possible.
Conventional Quantitative Conventional quantitative X-ray microanalysis bases

Analysis Procedures on measured intensities of characteristic X-rays
which are related to standard measurements, filed
reference spectra, or computed values. The intensity
quotients are commonly called k-Ratios. The
objectives of modern matrix correction are going far
beyond empirical approaches as e.g. proposed by
Bence and Albee (1968). Standard algorithms like
ZAF, PhiRhoZ and advanced approaches (PAP,
PROZA, XPP, etc.) try to model the fairly complicated
inter-elemental effects of X-ray self-absorption and
fluorescence enhancement by using empirical
formulas or relations derived from theory. Their aim
is to reduce the number of required references and
to improve accuracy.
Using modern calculation methods, internal

calibration curves can already be derived from only
one standard value per chemical element to be
analyzed. Composites as well as pure element
standards can be used as reference. Meanwhile the
application of the related calculation procedure is
completely automated. Still, prerequisites for precise
standard-based analysis are very stable and well
controlled working conditions and properly
processed samples.
In contrast to that, every day use of quantitative X-

ray microanalysis calls for reliability of analysis using
ordinary instruments with a great variety of
materials, which come as fragments, powders, dust,
or environmental samples of any kind. Frequently
these materials have to be analyzed in natural state.
With natural samples standard-based quantitative

analysis is likely to fail just because of significant
local variations across the rough or coarse sample
surface. Correct matching of standards and
specimens taking local structure, sample point,
feature size, etc., into account is virtually
impracticable.
P/B-ZAF Spectrum Analysis Self-calibration. The self-calibrating P/B-ZAF

Background method starts with peak to background intensities as
primary input in contrast to absolute peak intensities
as with classical approaches. This reference to
relative data, i.e. ratios of peak intensities to the
background, only renders the method to be
completely independent from scaling factors like
beam current, acquisition time and solid angle of the
46
Appendix QUANTAX
detector. Moreover the considerable independency

of properly chosen intensity ratios from different
other influences allow accurate quantitative analysis
results even under conditions of unknown
absorption, which means the method is self-
calibrating to the current conditions.
True standardless. The necessary reference values

for determination of the fundamental peak to
background ratios are provided by characterization of
the bremsstrahlung, which is simultaneously
detected by the EDS-detector. The bremsstrahlung
normally seen as a disturbing background
phenomenon just to be removed turns out to provide
all necessary supplementary information that is
needed for a reliable analysis. Since no other input is
necessary P/B-ZAF fundamental parameter analysis
is true standardless. Both, characteristic X-rays and
bremsstrahlung are of the same physical nature,
apart from the generation process. Therefore
absorption effects, measurement errors, and
detector artifacts more or less cancel out each other
by calculation of the fundamental ratios, leaving only
a minor secondary influence of these effects on
quantitative results.
Theoretical derivation. The intensity of the

primarily generated characteristic X-rays can be
derived from the ionization cross-section of the
respective electron shell, the fluorescence yield and
relative emission rate with sufficient accuracy.
However, this radiation is attenuated by absorption,
possibly enhanced by fluorescence and detected at a
solid angle that is hard to control. Moreover, the
registered X-ray quantities are proportional to the
product of electron beam current and acquisition
time, which in total is expressed by the well known
fundamental equation of electron probe
microanalysis:
Ω
N ichar = ⋅ I 0 ⋅ t ⋅ ω i ⋅ ε i ⋅ qi ⋅ ci ⋅ (Z ⋅ A ⋅ F )i (1)
char
4π
The so-called factors Zchar, Achar and Fchar actually

represent complicated terms depending on the
composition of the sample. Within classic standard-
based analysis heuristic procedures are used to
extract the unknown concentrations from the
resulting set of nonlinear equations by means of
comparisons with values from the standard
47
database.
Nichar number of x-ray photons of

the characteristic energy of a The intensity of bremsstrahlung for a given energy
certain element i range can be expressed through an equation similar
Ω solid angle covered by the to equation (1).
detector
I0 electron beam current
Ω
N ibr = ⋅ I 0 ⋅ t ⋅ ε i ⋅ (Z ⋅ A)i
t acquisition time br
(2)
εi detector efficiency 4π
ωi fluorescence yield
qi relative emission rate
Within equation (2) the element concentrations ci are
ci concentration of element i only implicitly included in the form of the factors Zbr
within mass fraction and Abr, representing the fact that generation of
Zi atomic number factor bremsstrahlung is not specific for individual
Ai absorption factor
elements. However other factors can be found in a
Fi fluorescence factor
similar way as within equation (1), which makes the
combination of these equations a unique tool for
eliminating excess parameters.
Dividing equation (1) by equation (2) results in the

expressions for the so-called P/B-ratios, which are
the intensity ratios of particular characteristic X-rays
Optimized light element and low to bremsstrahlung for the same energy range.
energy line analysis (TQuant)
(Z ⋅ A ⋅ F )i
Due to statistical reasons, a high line char
density and possible deviations in the N ichar
( P / B) i = = ω ⋅ q ⋅ c ⋅ (3)
low energy background, P/B ratios
( Z ⋅ A) br
i i i
acquired at energies lower than 1 keV
N ibr i
are not suited for quantification.
Nevertheless, a net count based Within this set of equations factors representing
standardless analysis for low energy measurement conditions like solid angle, detector
lines below 1 keV combined with the
P/B ZAF algorithm for elements with efficiency and electron beam current are no longer
lines above 1 keV provides present. After determining the P/B-ratios for all
appropriate data for a standardless elements in question from a single measured
analysis. The net count based spectrum equation (3) can be solved for the required
analysis can even be used when a
concentrations. Wendt [1], Small [2], and Statham [3]
sample contains only light elements
or the spectrum contains only low propagated this approach already a long time ago.
energy lines For practical use Bruker continuously improves the
TQuant, an adaptive algorithm
fundamental procedures and adds new formalisms
developed by Bruker, chooses the for dealing with light elements.
right spectrum analysis strategy for
every type of sample. The solution of equation (3) yields absolute values for
concentrations without the need for normalization.
Therefore, checking the result quality is feasible by
critical assessment of the sum of element
concentrations, provided that all elements contained
in the analyzed specimen are accessible to X-ray
analysis. Otherwise a non-detectable element can be
determined by difference. In contrast to standard-
based analysis also in this case no reference
measurement or system normalization is required.
48
Appendix QUANTAX
Using the P/B-ZAF method, to enable standardless

analysis only the primary generation rates of
characteristic X-rays must be known from theory.
These can be obtained with sufficient accuracy from
available atomic databases. As in practical cases
absorption of characteristic X-rays can be regarded
as similar to absorption of bremsstrahlung (Achar ≈
Abr). The less accurately known and more complex
mass absorption coefficients of the different
chemical elements are excluded from the
calculation. Theoretical modeling of bremsstrahlung
and fitting to the measured spectrum, provides in
addition a convenient means for checking the
electron energy E0 (excitation energy) and - in case of
well known detector characteristics - for calculation
of a nominal local tilt angle.
49
Further Reading
1. Castaing, R., "Application des sondes 9. Small J.A. u.a., Nat. Bur. Standard
electronique a une methode d' Spec. Publ. 533 (1978) 29
analyse ponctuelle chimique et
10. Heckel, J., Jugelt P., X-Ray
cristallographique", Theses
Spectrometry 16 (1987) 67
Universität Paris (1951)
11. Strüder, L. et al., Microscopy &
2. Heinrich, K.F.J., Electron Beam X-ray
Microanalysis 4 (1999) 622
Microanalysis, Van Nostrand
Reinhold, New York, NY, 1981 12. Small, J. et al., J. Appl. Phys. 61
(1987) 459
3. Reimer, L., Scanning Electron
Microscopy, Heidelberg (1998) 385 13. Trincavelli, J. et al., X-Ray
Spectrometry 27 (1998) 81
4. Wendt, M., Kristall und Technik 13
(1978) 1259 14. Bowman, H. H. u.a., Science 151
(1966) 562
5. Wendt, M., Schmidt, A., phys. stat.
sol. (a) 46 (1978) 179 15. Reed, S.J.B., Journal of Physics E5
(1972) 997
6. Pouchou, J.-L., Standardless X-Ray
Analysis of Bulk Specimens, 16. Scholze, F., Nucl. Inst. and Meth. in
Mikrochimica Acta (1994) 114, 33 Phys. Res. A 339 (1994) 49
7. Heckel, J. Jugelt P., X-Ray 17. Statham, P.J., Journal of Microscopy
Spectrometry 13 (1984) 159 130 (1983) 165
8. Statham P.J., Microchimica Acta,
Suppl.8 (1979) 229
50
INDEX QUANTAX
Index
A D
Absorption Data
A-factor ..................................................... 43, 47 Transfer of SEM- ............................................ 29
Detector window............................................ 42 Database, Atomic- .............................................. 10
Mass absorption coefficient ........................... 42 Dead layer, Si- .................................................... 42
Sample coating ............................................... 25 Depth distribution function................................. 16
Self-absorption ............................................... 43 Detector ............................................................. 29
Surface layers ................................................. 11 Correction ...... → Correction of detector effects
ZAF matrix correction ..................................... 39 Effects............ → Correction of detector effects
Absorption edges ............................................... 42
Image- ................................... → Image detector
Accuracy of EDX........................................... 11, 18
Parameter ...................................................... 29
Acquisition ............................ → Spectrum-, Image-
Position .................................................... 19, 28
A-factor ............................................................... 47 Window.......................................................... 42
Analysis ................................ → Spectrum analysis Direct reference, analysis by- ............................. 12
Analysis volume ................................................. 16 Duane Hunt Limit ......................................... 11, 33
Atomic database................................................. 10
Auger electron .................................................... 41
E
B Electron beam
Control of- ...................................................... 15
Backscatter coefficient ....................................... 40 Current ..................................................... 25, 27
Beam .......................................... → Electron beam Electron detector ...................... → Image detector
Beam current factor ................... → System Factor Electron image .........................................→ Image
Beam current meter ..................................... 11, 30 Electron trap....................................................... 26
Beryllium, detector window ............................... 42 Electronic noise .................................................. 44
Bremsstrahlung .................................. 9, 12, 29, 42 Element identification .......................................... 9
Calculated ....................................................... 19 Element imaging .......................................... 14, 39
Duane Hunt Limit ..................................... 11, 33 Energy
P/B-ZAF .............................................. 10, 39, 47 Range ............................................................. 29
BSE detector ..................................................... 16 Resolution ...................................................... 29
Scale, calibration ...................................... 30, 31
C Threshold ................................................. 29, 33
Energy dispersive spectrometry ........................ 39
Calibration Error, statistical ............................................ 18, 33
Image- ...................................................... 30, 31 Escape correction .............................................. 45
Sample ..................................................... 11, 30 Escape peak ....................................................... 45
Spectrometer- .......................................... 30, 31 ESD ................................. → Electro static damage
System- .................................................... 11, 12 Excitation energy ................................... 11, 19, 26
Carbon coating ...................................... → Coating External scan system .................... → Scan System
Carbon contamination ........................................ 32
Characteristic X-rays ................................. 9, 40, 46
Charging of sample ............................................ 31 F
Chemical shift..................................................... 41 Fano-factor ......................................................... 44
Coating
Faraday cup ....................... → Beam current meter
Of samples ..................................................... 25
F-factor ......................................................... 43, 47
Color
Fluorescence................................................ 32, 43
Adjustment of-................................................ 14
Fluorescence yield ....................................... 41, 47
Image- ............................................................ 14
Frame average ................................................... 28
ColorScan ........................................................... 14
FWHM ............................................................... 44
ColorSEM ........................................................... 39
Concentration profiles ........................................ 16
Cool max ............................... → Maximum cooling G
Cooling mode ..................................................... 29
Geometry parameters ........................................ 29
Count rate .................................................... 29, 33
Critical energy..................................................... 40
Current reference ............................................... 12
51
User Manual INDEX
H N
High voltage ................................................. 19, 26 Net peak area ..................................................... 11
HV ................................................. → High voltage
O
I Object analysis ................................................... 15
ICC ....................... → Incomplete charge collection Overvoltage ........................................... 26, 33, 40
Identification of elements..................................... 9
Image P
Calibration ................................................ 30, 31
ColorScan ....................................................... 14 P/B-ZAF
Element- ......................................................... 14 Matrix correction .............................................. 9
Magnification .................................................. 26 Standard related analysis ............................... 12
Resolution ...................................................... 28 Standardless analysis ................... 10, 25, 34, 46
System settings ............................................. 28 Parameter
Image capture .................................................... 13 Detector ......................................................... 29
Image detector ............................................... 9, 14 Geometry ....................................................... 29
Incomplete charge collection ............................. 45 Phase analysis ................................................... 15
Information depth ............................................... 16 PhiRhoZ matrix correction .............................. 9, 46
Inhomogeneous samples ............................. 17, 31 Pile-up peak ................................................. 32, 45
Instrument factor ........................ → System factor Point analysis ............................................... 17, 26
Ionization cross section ................................ 40, 47 Polished sample ........................................... 11, 25
Position tagged spectrometry ............................ 15
Precision of results ...................................... 11, 18
K Preparation, sample- .......................................... 25
k-Ratios .................................................. 11, 19, 46 Primary energy ....................... → Excitation energy
kV→ High voltage Profiles ............................ → Concentration profiles
PTS ..................... → Position tagged spectrometry
Pulse
L Processor ...... → Spectrometry signal processor
Lambert-Beer law ............................................... 42 Throughput .................................................... 29
Library ....................................... → Standard library
Light element Q
Analysis .................................................... 29, 33
Window .................................................... 42, 43 Qualified sample ................................................ 12
Limit of detection ............................................... 18 Qualitative Analysis ......... → Element identification
Line average ....................................................... 28 Quantitative spectrum analysis ............................ 9
Line profiles........................................... → Profiles Quantum efficiency............................................ 42
Line scan ............................................................ 16
Line series .................................................... 17, 41
LN2 ............................................. → Liquid nitrogen
R
LOD ....................................... → Limit of detection Reference based analysis .................................. 12
Reference measurement ............................. 11, 12
Reference sample .............................................. 12
M Relative emission rate........................................ 41
Magnification setting .......................................... 26 Resolution
Mapping ............................................................. 14 Spatial ............................................................ 26
Mass absorption coefficient ............................... 42 ROI........................................ → Region of Interest
Maximum Rough surface ........................................ 11, 39, 43
Pulse throughput (setting) .............................. 29
Maximum cooling (detector setting) .................. 29 S
MCA ............................... → Multi channel analyzer
Microanalysis ....... → Electron probe microanalysis Sample
Moseley’s law .................................................... 41 Charging up .............................................. 11, 31
Multi channel analyzer ........................................ 45 Coating ........................................................... 25
Multiple Inhomogeneous ....................................... 17, 31
Analysis objects ............................................. 15 Polishing ........................................................ 25
Multipoint analysis ................................. 15, 17, 31 Preparation ............................................... 25, 31
Qualified- ........................................................ 12
52
INDEX QUANTAX
Strain ........................................................ 15, 28 Systematic errors ............................................... 11

SCA ............................ → Single channel analyzers
Scan
Speed ............................................................. 28
T
Scan system ....................................................... 13 Tail correction ..................................................... 45
Scanning electron microscope ............................. 9 Take off angle .............................................. 11, 43
SDD detector...................................................... 44 TEM analysis ...................................................... 19
SE detector ....................................................... 16 Template .................................. → Report template
Self-absorption ................................................... 43 Thermostat mode (detector setting) .................. 29
Self-calibration ........................................ 10, 12, 46 Thin sections ...................................................... 19
SEM .................. → Scanning electron microscope Threshold ..................................................... 29, 33
Semiconductor detector, X-ray ........................... 39 Throughput............................. → Pulse throughput
Semi-standardless analysis ................................ 10 Tilt angle
S-Factor method ................................................. 12 sample- .......................................................... 25
Shelf correction .................................................. 45 TOA ............................................. → Take off angle
Si dead layer ....................................................... 42 True standardless analysis ................................. 10
SLEW ............................ → Light element window
Solid angle ........................................ 10, 28, 46, 47
Spatial resolution ................................................ 26 V
Spectrometer ....................................................... 9 Volume
Calibration ................................................ 30, 31 Analysis- ......................................................... 16
Functions ........................................................ 29 VP mode ...................... → Variable pressure mode
Spectrometry signal processor ........................... 29
Spectrum ............................................................ 14
Spectrum analysis W
Method selection ........................................... 34
WD ......................................... → Working distance
Multipoint ........................................... 15, 17, 31
WDX ...... → Wave length dispersive spectrometry
Quantitative ...................................................... 9
Scan area ........................................................ 15 Window.................................. → Detector window
Spot .......................................................... 17, 31 Working distance ......................................... 26, 33
Spectrum correction ....... → Correction of detector
effects X
Spectrum mapping ............................................. 15
Standard library................................................... 11 XFlash® detector .................................... 29, 39, 44
Standard related analysis .............................. 11, 12 X-ray analysis ........................→ Spectrum analysis
Standard sample X-ray detector.......................................→ Detector
Hard to get- .................................................... 13 X-ray lines
Standardless analysis ............................. 10, 11, 13 Characteristic- ................................................ 40
Stopping power .................................................. 40 X-ray spectrometer ...................... → Spectrometer
Strain, sample-........................................ 15, 27, 28 X-ray spectrum ................................... → Spectrum
Stray radiation, correction of- ............................. 32
Supporting grid ............................................. 19, 26
Surface
Z
Layers ............................................................. 11 ZAF matrix correction ......................... 9, 10, 39, 46
Sensitivity ....................................................... 25 Z-factor ......................................................... 40, 47
Tilt .................................................................. 25
System factor ......................................... 11, 30, 31
53

Introduction To EDS Analysis

Uploaded by

Copyright:

Available Formats

Introduction To EDS Analysis

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Introduction To EDS Analysis

Uploaded by

Copyright:

Available Formats

Introduction to

Tel. +49 (30) 670 990-0

Order no. DOC-M82-EXX053.

2 Microanalysis Basics .................................................................................................................... 7

3 Performing X-ray Microanalysis ............................................................................................... 23

History of Energy Dispersive X-ray Microanalysis ...................................................... 39

2.1 Overview ...................................................................................................................... 9

Qualitative analysis. The characteristic X-rays

In complicated cases the expertise of the operator

Thereby complete interactive control over critical

Position referred methods. Combining electron

2.2 Quantitative Analysis

2.2.1 True Standardless Standardless analysis. Standardless electron probe

Using the P/B-ZAF algorithm the characteristic X-ray

algorithm is discussed in the Appendix.

Internal reference. The internal reference to the

Analysis conditions. The bremsstrahlung contains

Automatic and interactive operation. P/B-ZAF

2.2.2 Standard-based Standard-based analysis is traditionally preferred in

Standards libraries. Data of standards for all

Normalizing the system. With samples prepared

standardless analysis - is to measure a system

Accuracy. In most cases systematical errors of

2.2.3 Reference-based A special case of standard-based analysis is the

Current References. Reference-based analysis uses

If the current reference is of same type as the

2.2.4 Standard-based P/B- This way, the remarkable insensitivity of P/B-ZAF

Using the self-calibrating P/B-ZAF standard-based

As with any standard-based method, analysis results

2.2.5 Combined Analysis Combined analysis provides the possibility to choose

2.3 Imaging Functions

2.3.1 Standard Image Image capture. State-of-the-art microanalyzers are

General control features, like magnification settings

2.3.2 ColorScan ColorScan is an enhanced acquisition method which

ColorScan requires no user input whatsoever.

2.3.3 Mapping and Mapping. In X-ray microanalysis mapping means

Element images. An element image is a

2.3.4 Hyperspectral Hyperspectral mapping (also known as Position

Like direct scanning position related analysis outlined

2.4 Object Analysis, Multipoint, and Line scan

2.4.2 Multipoint Analysis With multipoint analysis a number of separate spots

In contrast to element mapping or recording a full

Prerequisite for performing line scans is a separate

2.5 Analysis Limits

2.5.1 The Analysis Volume X-ray microanalysis is surface sensitive. The

Origin of the signals. The figure on the left side

The yellow and red ranges are the volumes from

Depth distribution function. The information depth

Lateral scattering. In deeper regions the electrons

As outlined above reducing the high voltage of the

2.5.3 Analyzing Inhomogeneous samples require special

Averaging over a larger sample surface (object

simple averaging of spectra from being adequate.

ColorScan, mapping and element imaging are means

Microtextures. Special cases are samples with

2.5.4 Sensitivity of X-ray Because of the small analysis volume X-ray

2.5.5 Limit of Detection Despite of the extraordinary low absolute detection

2.5.6 Accuracy and Precision and accuracy. The precision of

proven to be reliable and accurate, allowing error