ISO 14595:2014

INTERNATIONAL ISO
STANDARD 14595
Second edition
2014-10-15
Microbeam analysis — Electron probe
microanalysis — Guidelines for the
specification of certified reference
materials (CRMs)
Analyse par microfaisceaux — Microanalyse par sonde à électrons —
Lignes directrices pour les spécifications des matériaux de référence
certifiés (CRM)
Reference number
ISO 14595:2014(E)
©
ISO 2014

---------------------- Page: 1 ----------------------
ISO 14595:2014(E)

COPYRIGHT PROTECTED DOCUMENT
© ISO 2014
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized otherwise in any form
or by any means, electronic or mechanical, including photocopying, or posting on the internet or an intranet, without prior
written permission. Permission can be requested from either ISO at the address below or ISO’s member body in the country of
the requester.
ISO copyright office
Case postale 56 • CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
E-mail copyright@iso.org
Web www.iso.org
Published in Switzerland
ii © ISO 2014 – All rights reserved

---------------------- Page: 2 ----------------------
ISO 14595:2014(E)

Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Preparation of the research material . 2
4.1 Selection of material. 2
4.2 Preliminary inspection of the material . 2
5 Heterogeneity of material . 2
5.1 Sample preparation . 2
5.2 Sample size . 2
5.3 Test conditions . 3
5.4 Test procedure . 4
5.5 Statistical evaluation of data . 4
5.6 Criteria for certification. 8
6 Stability of the research material . 8
7 Determination of the chemical composition of CRMs . 9
7.1 Classification of CRMs . 9
7.2 Determination of classification of CRMs . 9
7.3 Selection of analytical method . 9
7.4 CRM material tested by EPMA only . 9
8 CRM specimen preparation, packaging, transportation, and storage .9
8.1 Preparation of CRM specimen . 9
8.2 Packaging . 9
8.3 Storage .10
8.4 Repolishing and recoating of CRMs.10
9 CRM certificate .10
9.1 Classification of CRM .10
9.2 Contents of the certificate .10
Annex A (informative) Spreadsheet instructions for the statistical evaluation of
heterogeneity data .11
Annex B (normative) Suggested classification of CRMs for EPMA .14
Annex C (informative) Example of a certificate for EPMA CRMs .15
Bibliography .16
© ISO 2014 – All rights reserved iii

---------------------- Page: 3 ----------------------
ISO 14595:2014(E)

Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards
bodies (ISO member bodies). The work of preparing International Standards is normally carried out
through ISO technical committees. Each member body interested in a subject for which a technical
committee has been established has the right to be represented on that committee. International
organizations, governmental and non-governmental, in liaison with ISO, also take part in the work.
ISO collaborates closely with the International Electrotechnical Commission (IEC) on all matters of
electrotechnical standardization.
The procedures used to develop this document and those intended for its further maintenance are
described in the ISO/IEC Directives, Part 1. In particular the different approval criteria needed for the
different types of ISO documents should be noted. This document was drafted in accordance with the
editorial rules of the ISO/IEC Directives, Part 2 (see www.iso.org/directives).
Attention is drawn to the possibility that some of the elements of this document may be the subject of
patent rights. ISO shall not be held responsible for identifying any or all such patent rights. Details of any
patent rights identified during the development of the document will be in the Introduction and/or on
the ISO list of patent declarations received (see www.iso.org/patents).
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation on the meaning of ISO specific terms and expressions related to conformity
assessment, as well as information about ISO’s adherence to the WTO principles in the Technical Barriers
to Trade (TBT) see the following URL: Foreword - Supplementary information
The committee responsible for this document is ISO/TC 202, Microbeam analysis, Subcommittee SC 2,
Electron probe microanalysis.
This second edition cancels and replaces the first edition (ISO 14595:2003), which has been technically
revised. It also incorporates Technical Correction ISO 14595:2003/Cor 1:2005.
iv © ISO 2014 – All rights reserved

---------------------- Page: 4 ----------------------
ISO 14595:2014(E)

Introduction
For electron probe microanalysis (EPMA), a comparative quantitative analytical method used throughout
the world, certified reference materials (CRMs) play a crucial role in the analytical accuracy.
This International Standard has been developed to facilitate international exchange and compatibility
of analysis data in EPMA.
It gives guidance on evaluating and selecting reference materials (RMs), on evaluating the extent of
heterogeneity and stability of RMs, and it gives recommendations for the determination of the chemical
composition of RMs for production as EPMA-certified reference materials.
© ISO 2014 – All rights reserved v

---------------------- Page: 5 ----------------------
INTERNATIONAL STANDARD ISO 14595:2014(E)
Microbeam analysis — Electron probe microanalysis —
Guidelines for the specification of certified reference
materials (CRMs)
1 Scope
This International Standard gives recommendations for single-phase certified reference materials
(CRMs) used in electron probe microanalysis (EPMA). It also provides guidance on the use of CRMs for
the microanalysis of flat, polished specimens. It does not cover organic or biological materials.
2 Normative references
The following documents, in whole or in part, are normatively referenced in this document and are
indispensable for its application. For dated references, only the edition cited applies. For undated
references, the latest edition of the referenced document (including any amendments) applies.
ISO Guide 31:2000, Reference materials — Contents of certificates and labels
3 Terms and definitions
For the purposes of document, the following terms and definitions apply.
3.1
heterogeneity
measured variation in compositions of elements measured from a group of specimens
Note 1 to entry: The contributions to heterogeneity include the uncertainties in the measurements from specimen
to specimen, from micrometre to micrometre within each specimen, and from the test procedure itself.
3.2
research material
material that appears to have the physical and chemical characteristics required of a CRM, but which
is to be examined in detail, including the determination of chemical composition, stability, and micro-
heterogeneity and macro-heterogeneity, before certification as a CRM
3.3
stability
resistance of a specimen to chemical and physical change during long-term storage at normal
temperature and pressure
3.4
stability
resistance of the material to changes in chemical composition during electron bombardment,
i.e. the resistance to change of the intensity of the relevant characteristic X-rays observed during the
time the specimen is exposed to the electron beam
3.5
uncertainty
quantitative statement that provides a value for the expected deviation of a measurement from an
estimate of the value of the specific measured quantity
© ISO 2014 – All rights reserved 1

---------------------- Page: 6 ----------------------
ISO 14595:2014(E)

4 Preparation of the research material
4.1 Selection of material
The research material used for the preparation of a CRM should exhibit little or no heterogeneity on
a micrometre scale, should be free from unwanted inclusions, and should be sufficiently dense (such
that voids, if present, can be readily avoided during testing and analysis) and stable under prolonged
electron bombardment.
The mounted research material should be of sufficient size to provide several areas suitable for point
beam analysis; each area should be approximately 20 µm or more in diameter. At a minimum, the size
should be at least twice the area of X-ray emission.
The quantity of research material should be adequate for the preparation of certified specimens.
In the case of a synthetic RM, a detailed description of the preparation technique should be provided. In
the case of minerals, the geographic origin, the source, and the separation process should be specified.
4.2 Preliminary inspection of the material
Initial inspection of a possible research material for a CRM should be made using a binocular optical
microscope to evaluate the material for the presence of unwanted inclusions, voids, or other phases, and
if these are found to be sufficiently abundant to interfere with EPMA of the major phase of interest, i.e. to
prevent a clean sampling of the major phase at multiple points with a 1 µm electron beam, the material
should be rejected.
Further inspection for the possible presence of very small inclusions or other phases should be carried
out on polished sections in reflected and/or transmitted light. An electron microprobe or a scanning
electron microscope with secondary electron and backscatter electron detectors might be needed.
Material of known composition with inclusions or other phases should only be considered suitable if the
inclusions or other phases can be easily identified and clearly marked on accompanying documentation
so that they can be avoided during use.
Material found suitable after preliminary inspection should subsequently be processed for further
determination of heterogeneity and stability.
5 Heterogeneity of material
5.1 Sample preparation
The CRM should be stable under the electron beam. It should not charge under required test conditions,
though in some cases, a conductive coating might be required. It should be in such a physical state that
it can be mounted and polished if necessary without rapid surface deterioration on exposure to the
atmosphere or vacuum.
The research material should be in the same or similar physical orientation as that proposed for the
CRM, e.g. if the CRM is to be cut or cleaved so that flat surfaces are to be used by the analyst for EPMA,
then the research material should be mounted in the same manner as that used to obtain heterogeneity
data.
5.2 Sample size
The number of specimens selected for testing will depend upon the number, size, and composition of the
individual specimens in the sample group.
For a large number of specimens, such as 200 or more seemingly identical specimens already cut or
cleaved and ready for distribution, testing of all specimens would be prohibitively time consuming. A
statistically representative number of randomly selected specimens should be selected for testing. If the
2 © ISO 2014 – All rights reserved

---------------------- Page: 7 ----------------------
ISO 14595:2014(E)

measured heterogeneity between and/or within specimens is observed to be greater than 1 % relative
after taking account of counting statistics for the elements being certified, testing of more specimens
might be needed.
Where there are fewer specimens, typically 5 to 20, which can be tested before being cut into smaller
specimens for distribution, each specimen may be analysed before being cut, provided that the
preparation process does not change the composition in any way.
Consultation with an experienced statistician is strongly recommended before data acquisition is begun.
Detailed rules regarding the sample size are avoided here to allow the analyst flexibility in designing the
testing procedures since decisions will depend upon the characteristics of the material and the number
of specimens available.
5.3 Test conditions
If the extent of heterogeneity is being determined on the micrometre scale, a 1 µm (point) beam should
be used for the analysis. In some cases, where there might be damage to the specimen by the electron
beam, a defocused beam, typically 5 µm diameter, may be used. Such samples should, therefore, be
certified for use only with a defocused beam.
Wavelength-dispersive X-ray spectroscopy (WDX) is the preferred method for heterogeneity
determinations because the high X-ray peak rates obtainable with the technique expedite the acquisition
of statistically useful data. Energy-dispersive X-ray spectrometry (EDX) can be applied by using
integrated X-ray peak intensities, but the data acquisition process is significantly longer. For specimens
sensitive to the high current needed for WDX, EDX can be the only choice.
Ideally, the excitation voltage used for the analysis should be about two and a half times the critical
excitation energy of the X-ray line of the element being analysed, although this can be difficult if several
elements are analysed simultaneously. As a compromise, the selected excitation voltage should be
sufficient to excite the X-ray lines of the elements used in the testing with an adequate overvoltage of at
least 1,5 times the critical excitation potential.
The selected X-ray lines used to acquire the heterogeneity data should not overlap any X-ray lines of
other elements in the specimen. This can be ascertained from wavelength dispersive spectrometer
(WDS) scans of the pure elements (or appropriate well-characterized compound specimen in which
overlap does not present a problem) and of the RM.
The current used will depend upon element concentrations, the stability of the specimen to the electron
beam, and the count rate desired.
The count rate should provide acceptable counting statistics. The count rate should not be so high
that the dead time of the WDS proportional counter will increase beyond the normal working range. A
normal proportional counter dead time is 1,2 µs or less. For energy dispersive spectrometer (EDS), the
dead time should be approximately 30 %.
NOTE Acceptable count rates will also depend upon tolerable counting uncertainties. From Poisson counting
statistics, the standard uncertainty in the counts obtained from an X-ray measurement is equal to the square
N
root of the total number of X-ray counts, . A 1 % error can be obtained when the total number of counts is 10
000, but this relative error can be reduced by increasing the number of counts. At 100 000 counts, the relative
error is reduced to 0,3 %. For an EDS, the number of counts refers to the counts in the window of interest or
integrated peak counts, not the total spectrum counts. This test uncertainty will be present regardless of the
extent of heterogeneity and can be minimized by increasing the integral number of counts through increased
current and/or counting time at a given excitation voltage. Both ultimately depend on the specimen stability,
while the counting time will also be limited by test practicality.
Knowing the estimated count rate, R, and the desired relative error, σ, the counting time, T, required to
2
achieve that relative error can be calculated from the equation T = 1/σ R. This equation is derived from
the Poisson estimate of the relative error due to counting statistics,
11/(NR)/= ()T

© ISO 2014 – All rights reserved 3

---------------------- Page: 8 ----------------------
ISO 14595:2014(E)

5.4 Test procedure
Before heterogeneity testing is begun, the edges of bulk specimens should be analysed and compared
to the specimen interior to determine whether there might be a consistent difference in element
concentrations in the two locations. Occasionally, differences can result from the manufacturing process
of materials such as metal alloys or synthetic crystals. If the edges are different from the specimen
interior, they should be removed before samples are taken for bulk quantitative analysis and before
specimens are mounted and polished for heterogeneity studies. In some specimens, differences might
also be due to mounting and polishing procedures; if this occurs and cannot be remedied, the certificate
should include instructions to the analyst to avoid using the material within a specified minimum
distance from the edge.
Specimens that are being compared should be mounted together in the same sample mount or block, if
possible. Carbon coating, if necessary, should be applied to all specimens simultaneously.
Tests should be designed to efficiently acquire the data needed to determine the extent of the within-
specimen and between-specimen heterogeneity, to determine the experimental uncertainty, and to look
for gradual increasing or decreasing concentration changes on the micrometre scale using 50 µm to
100 µm line scans. Examples of tests are given below, but they may be modified depending upon the
individual material or group of specimens being analysed. The beam current should be monitored to
provide a value corresponding to each data reading enabling subsequent current drift corrections to be
carried out, if necessary.
NOTE It is advisable to collect data in an ASCII format that can be easily put into a spreadsheet for subsequent
processing.
For each specimen being tested, X-ray counts for several, randomly selected points (typically 7 to 10 or
more depending upon the size of the specimen) should be acquired. These data should be acquired at
least in duplicate i.e. integral X-ray counts should be acquired and recorded at least twice on each point
without moving the specimen or electron beam between acquisitions. Specimens should be analysed in
a random order and preferably, each specimen should be analysed twice, each time in a different order. It
may be worthwhile for different operators to take data for duplicate analyses, using a different random
[6]
sampling plan for each. Refer to ISO Guide 35 for sampling procedures and methods of evaluating
results. The data from this type of test is used to calculate the within-specimen and between-specimen
uncertainties, as well as the test uncertainty after beam current drift corrections are made. When
background data are obtained for each element, the uncertainties can be expressed as a mass fraction.
The formulae used for these calculations are given in 5.5.
To test for the presence of concentration trends within each specimen, which might not be detected by
random sampling, line profiles of the points less than 5 µm apart and 50 µm to 100 µm in length should
be prepared. Two-line profiles normal to one another are recommended. For specimens of 1 cm to 2 cm,
a set of two-line profiles should be prepared from at least two different locations on the specimen. After
current corrections, data should be plotted (distance against X-ray counts) for each element to expose
variations in concentrations that might be present. Such trends might not preclude the certification
process if they are within the 99 % confidence limits or ±3 times the Poisson counting error (square root
of the integral number of X-ray counts).
5.5 Statistical evaluation of data
The uncertainties in the element concentrations resulting from heterogeneity within specimens and
between specimens and in the test acquisition can be obtained from the procedures described above
using the following calculations.
[1][2][3][4]
NOTE There are several examples of the use of test procedures and calculations similar to those
described here; the statistical notation has been simplified for this document to facilitate its usage. The statistical
[5][6][8]
approach used here is called a nested design that is described in detail in other references. The procedures
described have been developed in collaboration between the National Institute of Standards and Technology
(NIST), Gaithersburg, MD, USA and the National Physical Laboratory (NPL), Teddington, Middlesex, UK and have
been used successfully. Other validated test and statistical procedures may be used, provided that they are
described in full in the CRM certificate.
4 © ISO 2014 – All rights reserved

---------------------- Page: 9 ----------------------
ISO 14595:2014(E)

Let w be the true mass fraction of a particular element in the RM. Any single micrometre scale
0
measurement, w, expressed in weight percent taken from a randomly selected point of a randomly selected
specimen will deviate from w because of the variation between specimens (macroheterogeneity),
0
variation within specimens (microheterogeneity), and the measurement error. The deviation, w – w ,
0
may be viewed as a sum of random effects, as shown in Formula (1):
w= w +S+P+E
0
(1)
where
is the true mass fraction in the selected specimen;
w +S
0
w +S+P is the true micrometre scale mass fraction concentration at the selected point of the
0
selected specimen;
E is the measurement error.
2 2 2
The components of variance σ , σ , and σ are the variances of the random effects S, P, and E,
S P E
w w w
2
respectively. The variance, σ , of the measurement w is given by Formula (2):
w
2 2 2 2
σσ=+σσ+ (2)
E
w S P
w
w w
If n independent measurements are made at each of n randomly selected points of each of n randomly
E P S
selected specimens and if w denotes the kth replicated measurement at point j of specimen i, then the
ijk
grand mean given by Formula (3):
n n n
S PE
1
w = w (3)
∑∑ ∑ ijk
nn n
()
PS E
ij==11 k=1
has a variant, given by Formula (4):
2 2 2
σσ σ
S P E
2
ww w
σ =+  + (4)
w
nn nn nn
S SP SP E
assuming the design is balanced. Thus, the uncertainty in the mean measurement W can be determined
2 2 2
from estimates of σ , σ , and σ . An approximate 95 % or 99 % confidence interval for the mean
S P E
w w w
micrometre scale concentration is respectively
1/2
2 2 2
 
σσ σ
S P E
ww w
 
w ± 2+ + (5A)
 
nn nn nn
S SP SP E
 
or
1/2
2 2 2
 
σσ σ
S P E
ww w
 
w ± 3 ++ (5B)
 
nn nn nn
S SP SP E
 
2 2 2
Estimates of σ , σ , and σ can be obtained from the raw count data as follows.
S P Ew
w w
Let Y denote the kth count measured at point j of specimen i, and let B represent the background
ijk ijk
count associated with the measured count Y . Assuming a linear relationship between the number of
ijk
© ISO 2014 – All rights reserved 5

---------------------- Page: 10 ----------------------
ISO 14595:2014(E)

counts above the background count and the mass fraction on the micrometre scale, Y – B can be used
ijk ijk
to determine a mass fraction measurement, as given in Formula (6):
YB−
()
ijkijk
w = (6)
ijk
C
where C is a conversion factor which depends on test conditions such as the operating voltage, counting
time, etc.
If B is the mean background count:
n
E
1
is the mean count at point j in specimen i; (7)
= Y
Y
∑ ijk
ij
n
k=1
E
n
P
1
= Y
∑ ij is the mean count for specimen i; (8)
Y
i
n
j=1
P
n
S
1
Y = Y
is the grand mean count; (9)
∑ i
n
i=1
S
n
S
2
S = YY−
nn () is the between-specimen sum of squares; (10)
S i
∑
PE
i=1
n n
S P
2
S = YY−
n
() is the between-points within-specimen sum of squares; (11)
P ∑ ∑ ij i
E
i=1 j=1
n n n
S P E
2
SY=  −Y
()
E ∑ ∑ ∑ ijkij is the error and baseline sum of squares. (12)
i=1 j=1 k=1
The corresponding mean squares are
S
S
S =
MS between-specimen; (13)
n − 1
S
S
P
S =
MP
between-points within specimen; (14)
nn − 1
()
SP
S
E
S =
ME
residual. (15)
nn n − 1
()
SP E
6 © ISO 2014 – All rights reserved

---------------------- Page: 11 ----------------------
ISO 14595:2014(E)

Assuming Poisson variation for the background count and for replicated counts at each point of each
specimen, it follows that
(i)
n n n
S P E
2

  −
(Y
ijk Y )
∑ ∑ ∑ ij
222
i=1 j=1 k=1 ;
is an unbiased estimate of C σσ−
E B
=S
w
ME
nn n − 1
()
SP E
(ii) n n
S P
2
  −
n )
(Y Y
∑ ∑ ij i
E
222 2
i=1 j=1
is an unbiased estimate of Cnσσ+ −σ ;
()
E EP B
=S
ww
MP
nn − 1
()
SP
(iii)
n
S
2

− Y
()
nn Y
∑ i
PE
222 22
i=1
is an unbiased estimate of Cnσσ++nn σσ− .
=S ()
E EP PE S B
MS ww w
n − 1
()
S
2
where σ is the variance of the background noise.
B
The estimated components of variance are thus taken to be
 
S
 
E
 + B
 
nn n − 1
SB + ()
 
SP E
2  
ME
sˆ = = (16)
E
22
w
ˆˆ
CC
 
S S
 
P E
 −
 
nn n − 1
nn − 1
()
 ) 
(
SS− SP E
P
 s 
2 MP ME
sˆ == (17)
P
2 22
w
n ˆ
C
n ˆ E
C
E
 
S
S
 
S P
−
 
n −1
nn −1
 )
(
SS−
s P
 s 
2 MS MP
ˆ
s == (18)
S
22
w
nn ˆˆnn
CC
PE PE
where the conversion factor, C, is estimated using Formula (19):
ˆ
C= YB− /wˆ (19)
()
0
where w is equal to the certified mass fraction concentration determined by chemical analys
...