ISO 14701:2011
(Main)Surface chemical analysis — X-ray photoelectron spectroscopy — Measurement of silicon oxide thickness
Surface chemical analysis — X-ray photoelectron spectroscopy — Measurement of silicon oxide thickness
ISO 14701:2011 specifies several methods for measuring the oxide thickness at the surfaces of (100) and (111) silicon wafers as an equivalent thickness of silicon dioxide when measured using X-ray photoelectron spectroscopy. It is only applicable to flat, polished specimens and for instruments that incorporate an Al or Mg X-ray source, a specimen stage that permits defined photoelectron emission angles and a spectrometer with an input lens that can be restricted to less than a 6° cone semi-angle. For thermal oxides in the range 1 nm to 8 nm thickness, using the best method described in the standard, uncertainties, at a 95 % confidence level, could typically be around 2 % and around 1 % at optimum. A simpler method is also given with slightly poorer, but often adequate, uncertainties.
Analyse chimique des surfaces — Spectroscopie de photoélectrons par rayons X — Mesurage de l'épaisseur d'oxyde de silicium
General Information
Relations
Standards Content (Sample)
INTERNATIONAL ISO
STANDARD 14701
First edition
2011-08-01
Surface chemical analysis — X-ray
photoelectron spectroscopy —
Measurement of silicon oxide thickness
Analyse chimique des surfaces — Spectroscopie de photoélectrons par
rayons X — Mesurage de l’épaisseur d’oxyde de silicium
Reference number
ISO 14701:2011(E)
©
ISO 2011
---------------------- Page: 1 ----------------------
ISO 14701:2011(E)
COPYRIGHT PROTECTED DOCUMENT
© ISO 2011
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means,
electronic or mechanical, including photocopying and microfilm, without permission in writing 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 2011 – All rights reserved
---------------------- Page: 2 ----------------------
ISO 14701:2011(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.
International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2.
The main task of technical committees is to prepare International Standards. Draft International Standards
adopted by the technical committees are circulated to the member bodies for voting. Publication as an
International Standard requires approval by at least 75 % of the member bodies casting a vote.
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.
ISO 14701 was prepared by Technical Committee ISO/TC 201, Surface chemical analysis, Subcommittee
SC 7, X‑ray photoelectron spectroscopy.
© ISO 2011 – All rights reserved iii
---------------------- Page: 3 ----------------------
ISO 14701:2011(E)
Introduction
The measurement of the thickness of silicon oxide at the surface of silicon wafers has been conducted in
the past by many methods. These generally apply to oxide layers thicker than 20 nm. It is often important to
measure thicknesses in the range below 10 nm, and this International Standard addresses the range below
8 nm using X-ray photoelectron spectroscopy. Problems arise in measuring film thicknesses in this thickness
range since, for a layer to bond well to the substrate, it must form strong inter-atomic bonds at the interface
so that a monolayer or more of layer and substrate interfacial material exists there. This material would not
necessarily be a thermodynamically stable bulk material. Additionally, if the layer is reactive, its outer surface
might have reacted with the environment and so be changed between fabrication and measurement. For the
particular case of silicon dioxide on silicon, at the interface there is approximately a monolayer of sub-oxides
and, at the surface, adsorbed materials containing carbon, oxygen and probably hydrogen atoms. These
effects lead to offsets for the thicknesses deduced from many methods that, whilst reliably measuring changes
in thickness between one specimen and another, have difficulty in defining an absolute thickness.
The procedures described in this International Standard provide methods to measure the thickness with high
accuracy (optimally 1 %) and also, more rapidly and simply, at lower accuracy (optimally 2 %). It could also form
a basis for the measurement of many film thicknesses on substrates, but, without considerable further work,
the uncertainties will be undefined.
iv © ISO 2011 – All rights reserved
---------------------- Page: 4 ----------------------
INTERNATIONAL STANDARD ISO 14701:2011(E)
Surface chemical analysis — X-ray photoelectron
spectroscopy — Measurement of silicon oxide thickness
1 Scope
This International Standard specifies several methods for measuring the oxide thickness at the surfaces of (100)
and (111) silicon wafers as an equivalent thickness of silicon dioxide when measured using X-ray photoelectron
spectroscopy. It is only applicable to flat, polished specimens and for instruments that incorporate an Al or
Mg X-ray source, a specimen stage that permits defined photoelectron emission angles and a spectrometer
with an input lens that can be restricted to less than a 6° cone semi-angle. For thermal oxides in the range 1 nm
to 8 nm thickness, using the best method described in this International Standard, uncertainties, at a 95 %
confidence level, could typically be around 2 % and around 1 % at optimum. A simpler method is also given
with slightly poorer, but often adequate, uncertainties.
2 Symbols and abbreviations
2.1 Abbreviations
HPLC high-purity liquid chromatography
IPA isopropyl alcohol
2.2 Symbols
The term intensity is used below and elsewhere. This refers to a measurement of peak area in the spectrum
after relevant background subtraction.
� total oxide thickness
oxide
thickness contribution to the Si O peak
�
2
Si O
2
� thickness contribution to the SiO peak
SiO
thickness contribution to the Si O peak
� 2 3
Si O
23
thickness contribution to the SiO peak
� 2
SiO
2
� intensity of the Si contribution to the Si 2p peak
Si
intensity of the Si O contribution to the Si 2p peak
� 2
Si O
2
� intensity of the SiO contribution to the Si 2p peak
SiO
intensity of the Si O contribution to the Si 2p peak
�
2 3
Si O
23
intensity of the SiO contribution to the Si 2p peak
�
2
SiO
2
L attenuation length for Si 2p electrons in Si
Si
© ISO 2011 – All rights reserved 1
---------------------- Page: 5 ----------------------
ISO 14701:2011(E)
attenuation length for Si 2p electrons in Si O
L 2
Si O
2
L attenuation length for Si 2p electrons in SiO
SiO
attenuation length for Si 2p electrons in Si O
L
2 3
Si O
23
attenuation length for Si 2p electrons in SiO
L 2
SiO
2
intensity normalization parameter for the Si O contribution to the Si 2p peak
� 2
Si O
2
� intensity normalization parameter for the SiO contribution to the Si 2p peak
SiO
intensity normalization parameter for the Si O contribution to the Si 2p peak
� 2 3
Si O
23
intensity normalization parameter for the SiO contribution to the Si 2p peak
�
2
SiO
2
� uncertainty contribution, at a 95 % confidence level, for the spectrum measurement statistics
n
�
uncertainty contribution, at a 95 % confidence level, for θ
θ
� uncertainty contribution, at a 95 % confidence level, for the analyser electron optics defining the
A
solid angle of acceptance
� uncertainty contribution, at a 95 % confidence level, for the validity of the equations for thicknesses
E
� uncertainty contribution, at a 95 % confidence level, for peak synthesis without the intermediate
F
oxides
� uncertainty contribution, at a 95 % confidence level, for the attenuation length
L
θ
angle of emission of electrons measured from the surface normal
3 Outline of method
Here, the method is outlined so that the detailed procedure, given in Clause 4, can be understood in context.
Typical spectra are available in the literature and given later in Figures 3 and 4.
The initial step of cleaning the specimens, if necessary, is given in 4.1. In 4.2 and 4.3, the specimens are
mounted and suitable spectrometer settings chosen. In 4.4 and 4.5, the procedures for recording the data
and measuring the intensities are given. Finally, in 4.6 and 4.7, the oxide thickness and its uncertainty at a
confidence level of 95 % are calculated. In 4.5 and 4.6, two methods are provided for calculating the oxide
thicknesses from the data: a more complex method with better uncertainties and a simpler method with poorer
uncertainties. The more complex method might achieve uncertainties as low as 1 %, but the simpler method is
restricted to uncertainties that are greater than 2 %. This greater figure is often adequate for many purposes,
however. The sequence of steps is illustrated in the flowchart in Figure 1. It might be useful to refer to this during
use of this International Standard.
2 © ISO 2011 – All rights reserved
---------------------- Page: 6 ----------------------
ISO 14701:2011(E)
Figure 1 — Flowchart of the measurement process
(Y and N at decision points are the usual “yes” and “no”, respectively)
Subclause 4.3 requires the angle of emission to be set accurately, and it is often the accuracy of this setting
that limits the final accuracy. Users of this procedure will have to ensure that the accuracies of these settings
are known in order to evaluate the final uncertainty. The settings can be checked to an adequate level using
[1][2]
reflectors mounted on the specimen stage, a laser beam and standard geometrical relationships .
© ISO 2011 – All rights reserved 3
---------------------- Page: 7 ----------------------
ISO 14701:2011(E)
4 Method for measuring the oxide thickness
4.1 Cleaning and preparing the specimen
4.1.1 For cleaning and preparing the specimens, gloves and uncoated stainless-steel tweezers are required.
In selecting gloves, care shall be taken to avoid those with talc, silicone compounds or similar contaminants.
“Powder-free” gloves have no talc, and fresh polyethylene gloves, or gloves of a higher quality, shall be used in
specimen handling. Do not use moulded gloves, for example vinyl, which will probably be covered with highly
contaminating release agents. Tweezers that are of uncoated stainless steel shall be used.
4.1.2 To manipulate specimens, the gloves are used to hold the tweezers and not the specimen. Avoid any
wiping materials, sometimes used to handle specimens, as they might result in unwanted contamination of the
specimen surface. Unnecessary contact of the specimen with the gloves shall be avoided. Specimen mounts
and other materials used to hold specimens shall be cleaned regularly whenever there is a possibility of cross-
[3]
contamination of specimens. The use of tapes containing silicones and other mobile species shall be avoided.
4.1.3 Specimens shall be prepared and mounted with clean tweezers to ensure that the surface is not altered
prior to analysis and that the best possible vacuum conditions are maintained in the analytical chamber. Use the
gloves to handle the tweezers to avoid contaminating them or any cleaning equipment with finger grease. Clean
the tweezers by one of the following two methods:
a) Immerse the tweezers before use for 16 h in high-purity liquid chromatography (HPLC), or equivalent,
grade (>99,5 %) isopropyl alcohol (IPA) that leaves no significant residue. Next, remove the liquid, renew
the IPA, agitate ultrasonically for 1 min, rinse in fresh IPA and remove the excess liquid using a jet of pure,
dry argon or another inert gas.
b) Boil the tweezers in ultra-high-purity water for 10 min.
Grip the specimen at the edge only, in a region that will not be analysed. Avoid breathing or speaking over the
specimen. Keep these tweezers in a clean, glass container for future use. Tools shall not unnecessarily touch
the specimen surface to be analysed.
4.1.4 Inspect the specimens for any scratches, blemishes or marks on the polished surfaces. Finger marks
should not be present but, if they are, may be removed as described in 4.1.6. Note the condition of the surface.
It should be featureless. Identify the side of the specimen for analysis. This is usually the polished side. If
the unpolished side is to be analysed, this International Standard is not applicable. If the specimen is too
large for insertion into the instrument, a smaller portion will need to be cut from it. To do this, material with a
(100)-orientated surface may be cut to form a suitably sized rectangular portion by cleavage along (111) planes.
In this way, a square of side 10 mm, bounded by <110> directions, may be conveniently produced. For those
specimens with a (111)-orientated surface, a similar cleaving along (111) planes forms equilateral triangles, also
bounded by <110> directions. Triangle sides of length 15 mm are convenient. The scribing for cleaving often
leaves very small fragments of Si on the specimens. These fragments shall, as far as possible, be removed. The
cleaning procedure described in 4.1.6 is often found sufficient for this purpose.
NOTE The <110> directions are usually indicated by flats cut into the sides of (100)- and (111)-orientated wafers.
4.1.5 Analyses show that wafers and many other materials such as metals accumulate organics, hydrocarbons,
silanes and phthalates from the environment. During storage of wafers, the thickness of these adsorbed layers
increases to around 0,35 nm on the polished surface in normal, uncirculated laboratory air after 100 days, but is
kept below 0,2 nm if a wafer container is used that has been kept closed and has not been exposed to excessive
[4]
heat (i.e. has been kept below ∼35 °C). In either case, the specimens should be analysed without cleaning.
If, however, there is evidence that they have been contaminated by organic contaminants (e.g. finger grease)
or the specimens have been cut to reduce their size, the contamination can be reduced to a thickness of about
0,14 nm by cleaning as described in 4.1.6.
4.1.6 If the specimens require cleaning, immerse them in a cleaned glass container in HPLC (or equivalent)
quality (>99,5 %) isopropyl alcohol (IPA) for 16 h (e.g. overnight). The top of the test tube can be conveniently
4 © ISO 2011 – All rights reserved
---------------------- Page: 8 ----------------------
ISO 14701:2011(E)
closed by a piece of clean aluminium foil. Next, remove the liquid, renew the IPA, agitate ultrasonically for 1 min,
rinse in fresh IPA and remove the excess liquid using a jet of pure (>99,9 % purity), dry (<0,01 % water) argon
or an equivalent rare gas. If IPA is not available, HPLC-quality chloroform or dichloromethane may be used. The
specimens are now ready for analysis.
[4]
NOTE The procedure using 16 h immersion in solvent leaves significantly less carbon than a simple ultrasonic rinse.
If HPLC-quality chloroform or dichloromethane is used instead of IPA, the level of carbon remaining is generally
about twice as high. However, the amount left depends on how the specimens have been contaminated in the
first place. Hence, chloroform and dichloromethane are not recommended unless IPA is unavailable. Note
that there are relevant safety requirements in using all solvents. Carbon deposited during any spectroscopic
analysis can be crosslinked by the radiation used, forming a tough adherent layer that cannot be removed
without compromising the oxide integrity. Do not use other cleaning methods, even if they are known to remove
[4]
contamination, since they might also change the oxide thickness .
If pure, dry argon or another rare gas is not available, do not use gas from pressurized cans that include a
propellant or from compressed-air lines, as these might deposit contaminants. Under no circumstances use
any proprietary cleaning agents or liquids containing surfactants.
4.2 Mounting the specimen
Mount the specimen on the specimen holder using fixing screws, or other metallic means, to ensure electrical
contact. Do not use double-sided adhesive tape. The (100)-surface specimens shall be mounted such that the
photoelectron angle of emission is set in the azimuth at 22,5° to one edge of the rectangular specimens and
the (111)-surface specimens shall be mounted such that the angle of emission is in an azimuth of one edge of
the triangular specimens. Set these azimuthal angles as accurately as possible and within 2° of their nominal
values. This is shown in Figure 2.
NOTE The reasons for selecting this geometry are described in detail in Reference [5]. This geometry sets the
emission direction as a single direction available to both the (100) and the (111) surfaces that is as far from any low index
directions as possible. In this direction, the Si substrate forward-focussed intensity is avoided.
4.3 Choosing spectrometer settings
−9
4.3.1 Achieve a good high vacuum with a pressure of less than 5 × 10 mbar. Operate the instrument in
accordance with the manufacturer’s documented instructions. The instrument shall have fully cooled following
any bakeout. Select the X-ray source. If a twin-anode source is available, it is usually best to use the Mg anode,
rather than the Al anode, since the former gives higher intensity and better energy resolution. If the instrument is
not equipped with a twin-anode source or if the monochromated source delivers more intense spectra than the
twin-anode source, then use the monochromated source. Ensure that the operation is within the manufacturer’s
recommended ranges for source power, counting rates, spectrometer scan rate and any other parameter
specified by the manufacturer. Ensure that the entrance solid angle for the spectrometer is set at a cone semi-
angle of less than 6°. Setting too small an entrance angle will limit the signal quality and the ultimate accuracy of
the measurement. Record a survey (widescan) spectrum to ensure that the only significant peaks are those of
Si, O and C. The intensity for peaks for all other elements shall not exceed 5 % of the intensity of the Si 2p peak
for the uncertainty analysis in 4.7 to be valid. Ensure that there are no significant peaks that are characteristic
of the specimen holder. Figure 3 shows an example of a widescan spectrum.
NOTE 1 The peaks visible should be the C 1s, O 1s and O Auger peaks and the Si 2p and 2s peaks. For good practice,
the height of the C 1s peak should be less than 30 % of the Si 2s or 2p peak heights although, even for significantly
[5]
larger amounts of carbonaceous contamination, the measurement should not be affected , except to the extent that the
important Si 2p signal intensity will be reduced, leading to an increase in the measurement uncertainty. Figure 3 shows a
typical result for a cleaned, but stored, specimen.
NOTE 2 The higher signal level for the unmonochromated Mg X-rays, compared with that for the monochromated Al
X-rays available in many instruments, is important in improving the accuracy of the final result. This choice leads to a
reduction in the term � discussed later.
n
4.3.2 Select slit width and pass energy settings to provide peak widths of around 0,6 eV to 0,9 eV for the
Si 2p and 2p elemental peaks.
3/2 1/2
© ISO 2011 – All rights reserved 5
---------------------- Page: 9 ----------------------
ISO 14701:2011(E)
4.3.3 For (100) specimens, set the photoelectron angle of emission to 34° and, for (111) specimens, 25,5°.
It has been found that the accuracy in setting this angle can limit the final accuracy of the thickness measurement.
It is recommended that users check that the calibration of the angle of emission setting is adequate for their
purpose. This can be done conveniently in spectrometers with suitable viewing windows by using a laser beam,
[1][2]
reflectors on the specimen stage and standard geometrical relationships .
NOTE It could be thought that normal emission is less sensitive to the accuracy of the emission angle. However, the
[5]
forward focussing along the crystal axis leads to non-linearity in the equations to be used, resulting in significant errors.
a) Plan view of (100) sample b) Side view of (100) sample from A
c) Plan view of (111) sample d) Side view of (111) sample from A
Key
1 to analyser
a
Analyser azimuth.
Figure 2 — Orientation of the specimens on the specimen stage
[Figures 2 a) and 2 b) show (100) surfaces with the photoelectron angle of emission at 34° from the surface
normal and with the analyser set in the azimuth P of the square Si specimen that is in its surface and at 22,5°
to the direction of one of the edges. Figures 2 c) and 2 d) show (111) surfaces with the angle of emission at
25,5° from the surface normal and with the analyser set in the azimuth Q of the triangular Si specimen that is
in its surface and parallel to the direction of one of the edges.]
6 © ISO 2011 – All rights reserved
---------------------- Page: 10 ----------------------
ISO 14701:2011(E)
Key
X binding energy
4
Y intensity/10 counts
1 oxygen Auger
2 oxygen 1s
3 carbon 1s
4 silicon 2s
5 silicon 2p
Figure 3 — Widescan (survey) spectrum of SiO on Si using Mg Kα X-rays
2
and before removal of the X-ray satellites
4.4 Recording data
Record the region of the Si 2p peaks, using the spectrometer settings selected in 4.3.1. If the instrument
incorporates a functioning twin-anode source with both Mg and Al X-rays, choose the Mg source, as it gives the
higher intensity and better energy resolution. For an unmonochromated source, record from a binding energy
8 eV lower than that of the elemental Si 2p peak at approximately 99,3 eV binding energy to a binding energy
7 eV higher than that peak. For a monochromated Al Ka source, the lower binding-energy limit may be reduced
from 8 eV to 3 eV. Note that the accuracy that can be achieved depends on the total signal level measured
for the Si 2p peaks. This signal level depends on the product of the count rates in the spectrum and the time
for recording that spectrum. For higher accuracy, use longer times. A spectrum acquisition time of 5 min to
10 min is often used. Longer times will improve the accuracy, but any improvement might not be significant
unless full consideration is given to all the uncertainties discussed in 4.7 to ensure that other uncert
...
Questions, Comments and Discussion
Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.