Experimental Methods in Chemical Engineering:

X‐Ray Photoelectron Spectroscopy‐XPS

Josianne Lefebvre,1 Federico Galli,2 * Claudia L. Bianchi,2 Gregory S. Patience3 and Daria C. Boffito3
1. Department of Engineering Physics, Polytechnique Montréal, C.P. 6079, Succ. CV Montréal, QC, H3C 3A7, Canada
2. Dipartimento di Chimica, Universitá degli Studi di Milano, via Golgi 19, 20133 Milano, Italy
3. Department of Chemical Engineering, Polytechnique Montréal, C.P. 6079, Succ. CV Montréal, QC, H3C 3A7, Canada

X‐ray photoelectron spectroscopy (XPS) is a quantitative surface analysis technique used to identify the elemental composition, empirical
formula, chemical state, and electronic state of an element. The kinetic energy of the electrons escaping from the material surface irradiated by an
x‐ray beam produces a spectrum. XPS identifies chemical species and quantifies their content and the interactions between surface species. It is
minimally destructive and is sensitive to a depth between 1–10 nm. The elemental sensitivity is in the order of 0.1 atomic %. It requires ultra high
vacuum (1 × 10−7 Pa) in the analysis chamber and measurement time varies from minutes to hours per sample depending on the analyte. XPS
dates back 50 years ago. New spectrometers, detectors, and variable size photon beams, reduce analysis time and increase spatial resolution. An
XPS bibliometric map of the 10 000 articles indexed by Web of Science[1] identifies five research clusters: (i) nanoparticles, thin films, and surfaces;
(ii) catalysis, oxidation, reduction, stability, and oxides; (iii) nanocomposites, graphene, graphite, and electro‐chemistry; (iv) photocatalysis,
water, visible light, and TiO2; and (v) adsorption, aqueous solutions, and waste water.

Keywords: depth profiling, nanocomposites, nanoparticles, photocatalysis, photoelectron peaks

INTRODUCTION This article reviews essential features of XPS and is part of a

‐ray photoelectron spectroscopy (XPS) is a surface‐ series that describes the most frequent analytical techniques,

X sensitive quantitative analysis technique.[2] It probes the

surface chemistry of materials and reports the elemental
composition, empirical formula (without hydrogen), and chemi-
experimental methodologies, and statistical approaches in chemical
engineering.[26,27] Together with some theory, it summarizes the
science disciplines that apply this technique the most, highlights
recent work published in The Canadian Journal of Chemical
cal and electronic state of the elements, with an average analysis
depth of 1–10 nm.[3] X‐ray beams irradiating a material’s surface Engineering, and reports sources of error and uncertainties.
generates a spectra of electrons with a characteristic kinetic Description
energy. Each element produces a set of peaks at characteristic
A low‐energy achromatic or monochromatic x‐ray source
binding energies (the energy required to eject an electron from
(most often AlKα or MgKα ) irradiates a sample in a ultrahigh
an atom and depends on the element, orbital, and chemical
vacuum, which ejects core‐level electrons from the atoms
environment). The peaks refer to the electron configuration
(Figure 1). The kinetic energy of a photo‐emitted core electron
within the atoms, and the number of detected electrons
is a function of its binding energy in the atom and depends on
(intensity) is related to how much is present. XPS software
the elements that compose the material. As a dominant
includes libraries of empirical or calculated relative sensitivity
relaxation mechanism at x‐ray energies typical of XPS, an
factors (RSF) for each element and each orbital and compute the
outer electron fills the formed core‐hole and the emission of an
element content from XPS peak areas. We calculate the atomic
Auger electron balances the transition energy. An electron
percentage on the surface layer based on the normalized
energy analyzer and an electron detector count the emitted
corrected signals. The analysis chamber operates at a ultra‐high
photoelectrons and Auger electrons as a function of their
vacuum (1 × 10−7 Pa) to minimize sample surface contamination
energy. The spectrum represents the material surface composi-
and maximize the number of photoelectrons reaching the
tion. The peak position on the energy scale refers to the
analyzer. It analyzes low vapour pressure solids such as
element and the peak area indicates its relative amount.
inorganic compounds,[4] metal alloys,[5] semiconductors,[6]
polymers,[7] elements,[8] catalysts,[9] glasses,[10] ceramics,[11]
dry paints,[12] papers,[13] dry inks,[14] woods,[15] plant parts,[16]
bones,[17] medical implants,[18] bio‐materials,[19], glues,[20] soil *Author to whom correspondence may be addressed.
particles,[21] and ion‐modified materials.[22] It also analyzes E‐mail address: federico.galli@unimi.it
some low vapour pressure liquids such as viscous oils[23] and Can. J. Chem. Eng. 1–6, 2019
ionic liquids.[24] In near‐ambient pressure, XPS equipment can © 2019 Canadian Society for Chemical Engineering
DOI 10.1002/cjce.23530
also analyze a variety of liquids and higher vapour pressure Published online in Wiley Online Library
materials.[25] (wileyonlinelibrary.com).

MONTH 2019 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING 1

Figure 1. Solid surface electron emission. X‐rays penetrate several atomic
layers down to some μ m; however, the top 2 to 3 layers contribute most
to the spectra.

Only electrons from a maximum depth of 10 nm contribute to

the signal, while a deeper electron emission contributes to the
spectrum background. Even though the background is usually
neglected, it contains information on sample structure.[28] Survey
scans identify the elemental composition of 1 nm to 10 nm of the
analyzed surface. High resolution spectra evaluate the chemical Figure 2. Photoelectric effect. In x‐ray nomenclature, the 1st shell is
state of each element through a core electron binding energy designated as K (n = 1) that holds up to 2 electrons. The second shell is
shift of 8.0 × 10−20 J up to 8.0 × 10−19 J. Curve‐fitting routines L (n = 2) with 2 + 6 electrons and the third shell is M (n = 3) with
determine binding energies, which shift due to the atoms 2 + 6 + 10 electrons. The valence band is the outer shell of electrons and
oxidation state, chemical bonds, or crystal structure. The tunable the conduction band.
x‐ray energy from synchrotrons radiation improves the photo‐
ionization cross‐section for various elements and core levels.[29]
Also, spectra with constant kinetic energy from synchrotron of the C1s from adventitious carbon contamination to a binding
radiation ensures a constant and tunable information depth for all energy of 4.6 × 10−17 J and background subtraction using a
elements.[30] Shirley baseline. Quantification results were obtained from the
Al2p, C1s, O1s, F1s, Fe2p3/2, and Co2p1/2 photoelectron peaks.
Theory We identified cobalt from the Co2p1/2 peak instead of the more
When an electromagnetic wave hits a solid surface, the latter typical Co2p or Co2p3/2 peaks because the Co2p3/2 peak overlaps
emits electrons, which is known as the photoelectric effect.[31] with the O KLL peaks and using it will introduce errors in the
Metals absorb incident energy, which Einstein described as a quantification results.
corpuscle of energy hν ; the material absorbs part of the energy ϕ
(work function), while the residual is the kinetic energy (Ek )
carried by the electron (hν = ϕ + Ek )[32] (Figure 1). APPLICATIONS
Einstein’s photoelectric effect (Figure 2) is the basis of
photoelectron spectroscopy theory. Photons with an energy Industry professionals and academics apply XPS to study surface
greater than the energy holding the electron in the atom ionize layers and thin films, nano‐materials, photovoltaics, catalysis,
it. The excess energy transforms into electron kinetic energy. corrosion, electronic devices and packaging, magnetic media, and
Jenkin et al. recorded the photoelectron spectra of several coatings. In 2016 and 2017, over 18 000 articles indexed by WoS
metals.[33] Later, Siegbahn[34] included a high‐resolution spectro- mentioned XPS,[26] and most of the articles appeared in physical
meter to identify characteristic peaks of electrons in the shell and chemistry with 3687 occurrences, followed by multidisciplinary
chemical bonding. Considering the energies before and after materials science (2535), multidisciplinary chemistry (1578),
photoemission, the electron binding energy (Eb ) is the difference chemical engineering (1515), and applied physics (1503). Chemi-
between the ionized atom final energy state e (E f ) and the target cal engineers publish articles most in the journals that specialize
atom initial energy state, (E i ) (Equations (1) and (2).

hν + E i = Ek (e−) + E f (1)

hν − Ek (e−) = E f − E i = Eb (2)

where Ek is the photoelectron kinetic energy.

Here we report a survey spectrum of a bimetallic catalyst of
10 % mass Fe and 10 % mass Co supported on alumina (FeCo,
Figure 3).[35] A VG ESCALAB 3 MKII recorded the spectra. The
x‐ray source was Mg Kα. It operated at a power of 300 W and a
spectrometer pass energy of 1.6 × 1017 J. The chamber pressure
was 4 × 10−7 Pa and the electron takeoff angle between surface
normal and entrance to energy analyzer was 0 rad. The analysis
area was 2 × 3 mm. The Thermo Avantage v4.78 software
quantified the survey results applying a homogeneous specimen Figure 3. X‐ray photoelectron survey spectrum obtained from a catalyst
model and Wagner sensitivity factors, after the charge correction with a mass fraction of 10 % of both Fe and Co supported on alumina.

2 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING MONTH 2019

Figure 4. XPS bibliometric map of 107 keywords in the top 10 000 cited articles indexed by WoS in 2016 and 2017;[1,37] the VOSViewer groups
keywords into 5 clusters (indicated by the colours), where the font size and circle size are proportional to the number of occurrences: nanoparticles
appears in 1474 articles followed by performance (1268); adsorption (891); surfaces (775); and composites (774). The least frequent of the top 100
keywords appear less than 120 times: in situ (119); selective oxidation (118); mixed oxides (117); polymers (116); and sensitized solar cells (115). The
lines represent citation links between articles.

in these WoS these categories.[36] Over 120 of the 252 WoS Chemical Engineering Journal (266), Electrochimica Acta (246), ACS
categories cite the technique at least once. Applied Materials & Interfaces (237), Journal of Physical Chemistry
The VOSViewer software groups bibliometric data according to C (214), International Journal of Hydrogen Energy (191), Ceramics
citation links and identified five major research clusters for XPS International (170), Journal of Colloid and Interface Science (9139),
(Figure 4).[37] Nanoparticles, thin films, and surfaces are the most and Catalysis Science & Technology (134).
frequently cited keywords of the red cluster; it has 32 of the top As of June 2018, the article entitled “n,p‐Codoped Carbon
107 keywords. The green cluster has the second most number of Networks as Efficient Metal‐Free Bifunctional Catalysts for Oxygen
keywords (29) and concentrates on catalysis: performance, Reduction and Hydrogen Evolution Reactions” was cited 145
oxidation, reduction, stability, and oxides. Nanocomposites, times.[39] The authors developed a 3D porous carbon network and
graphene, graphite, and electro‐chemistry make up the keywords included keywords from the blue and green clusters: electrocatalysis
of the third ranked cluster (blue) with 22 keywords. The fourth (blue); graphitic carbon (blue); hydrogen evolution (green); oxygen
cluster (yellow) includes photocatalysis, water, visible light, and reduction (green); Zn air battery (blue); N2 doped graphene (blue);
TiO2 and contains 12 keywords. The magenta cluster also has 12 electrocatalysis (blue); energy conversion; oxides (green);
keywords including adsorption, aqueous solutions, waste water, nanosheets (blue); nanotubes (blue); melamine; sheets; and water
and removal. XPS analysis solves problems with existing surface (yellow). The keywords in the second most cited article (“Advanced
interactions or investigates new materials. With respect to Electrochemical Energy Storage Supercapacitors Based on the
catalysis, it identifies features and changes related to both the Flexible Carbon Fiber Fabric‐Coated with Uniform Coral‐Like
active phase (transition metal particles, for example) and an inert MnO2 Structured Electrodes”)[40] include MnO2 nanostructures (red
support (alumina, metal oxides, TiO2 , zeolites, and carbon). XPS cluster), carbon fabric (blue), hydrothermal reaction (green), Li‐ion
evaluates how the active component is dipsersed across the batteries (blue), catalysis (green), high‐performance (blue), fuel cells
surface. For example, Mazzieri et al.[38] studied the properties of (blue), graphene (blue), nanotubes (blue), composites (blue), electro-
Ru∕Al2 O3 catalysts and XPS revealed Ru in RuO3 or RuO4 form catalysis (blue), and nanofibres (blue). In the 3rd most cited paper
and a decrease in the Ru/Al atomic ratios after calcination, which (“Macroscopic and Microscopic Investigation of U(VI) and Eu(III)
resulted from a migration of ruthenium species to the interior of Adsorption on Carbonaceous Nanofibers”),[41] the XPS and XANES
the pellets. analyses indicated that OH and COOH groups of the carbon
Of the top 10 000 cited articles in WoS, RSC Advances published nanofibres adsorbed the U(VI) and Eu(III). This research relates
695, followed by Applied Surface Science (657), Journal of Alloys mostly to adsorption (magenta) and included sorption (magenta) and
and Compounds (307), Applied Catalysis B‐Environmental (293), graphene oxide nanosheets (blue) among the keywords.

MONTH 2019 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING 3

 Can. J. Chem. Eng. published 33 articles in 2016 and 2017 that peaks because the latter are from higher kinetic energy electrons
contained XPS among the keywords, which ranks it in the top that have a longer inelastic mean free path and thus are less
third of all analytical techniques.[26] Feng et al.[42] investigated attenuated traveling through the over layer. Recently, we analyzed
the features of ZnO/red clay sorbents. XPS characterized their homogeneous MoO3 films and found that adventitious carbon
surface properties before and after regeneration and established contamination made up 30 % of the elemental composition of the
the chemical state of Zn, O, and S. Moreover, it showed that the analyzed volume. The calculated O/Mo ratio with the standard
surface was hydrated with water molecules and hydroxyl groups. homogeneous analyte was 3:1 but the actual ratio was closer to
Zuo et al.[43] prepared spherical supported catalysts, ZSM‐5, 4:3 (based on XPS MultiQuant software that considers hydro-
γ ‐Al2 O3, MCM‐41, SBA‐15, and β ‐zeolite, to produce methyl carbon contamination). Good practice requires researchers to
acrylate (MA) with phosphorous as the active component. The determine the specimen morphology and to apply appropriate
P2p peak intensity increased proportionally with the P loading equations or models for quantification analysis.
until it reached the maximum feasible loading. Khalid et al.[44] Published XPS data and results also contain errors. Seldom do
investigated the gold dissolution from sulphidic minerals. XPS researchers evaluate uncertainties, and results are often reported
characterized the surface‐obstructing species formed during the with more significant figures than warranted considering experi-
cyanidation process, and the single spectra of each metal mental parameters. When researchers report the experimental
disclosed the chemical state and composition responsible for the conditions accurately, readers better assess the validity and
retarding effect on gold leaching. In the case of bimetallic quality of the results. High resolution XPS data are interpreted
catalysts, XPS reveals the electron transfer between metals and the majority of the time by producing a synthetic spectrum
the consequent formation of an alloy,[45] as the binding energy of composed of a series of functions representing individual peaks.
alloying is lower or higher than that of a standard pure species, The sum of these peaks resembles the experimental spectrum.
proving that partial electron transfer occurred. Binding energy The analyst assigns to each of these individual peaks the chemical
shifting also proved the decomposition from an oxidized species group or the oxidation state of the corresponding elemental
to new compounds. Ansaloni et al.[46] concluded that the Mo spectra based on the peak’s binding energy, with the help of
oxidation state was unaffected in the hydrodeoxygenation of previously published work or databases. This analysis is called
guaiacol because the binding energies of the main peaks were peak fitting or curve synthesis (not deconvolution). XPS decon-
unchanged after the reaction. Yang et al.[47] determined the volution methods calculate the energy loss spectrum from the
chemical nature of the nitrogen‐containing functional groups of original data or enhance spectral resolution by deconvoluting
activated carbon by XPS. Peak fitting revealed five nitrogen‐ spectral broadening due to instrumental factors.
containing species that are related to catalytic activity. Wang
et al.[48] studied how additives changed the metal valence of Au
Case Study
and Cu ions on catalyst for the acetylene hydrochlorination at
165 °C by XPS. Tables 1 and 2 compare systematic uncertainties to statistical
uncertainties survey quantification results. The statistical uncertainty
Δmass calculated from N1 ∕ 2 is negligible. For most peaks, the sample
UNCERTAINTY standard deviation smass determined from the five independent
measurements is significantly larger. This is due to systematic
Sources of Errors and Limitations uncertainty contributions. Even in the case where the same spot of
Sources of errors in an XPS experiment include mishandling the same sample is measured five times consecutively under the same
specimens, contaminated vacuum condition in the XPS analysis conditions, there remains systematic uncertainty due to the choice of
chamber, specimen damage during the measurement, and experimental parameters like step size while scanning the binding
questionable analytical strategies.[49] Strategies to limit surface energy and data treatment by the analyst. In the case where five
contamination include handling specimens from edges with clean specimens of the same sample are analyzed (Table 2), the sample
tweezers and avoiding direct contact of the surface region with standard deviation is larger due to additional systematic uncertainty
anything, including gloves. Plastic containers transfer contami- introduced from specimen positioning and, possibly, from differences
nants to the specimen surface such as plasticizers. Similarly, between the specimens themselves, which occur if the catalyst sample
residues such as oils from the hand also contaminate the surface is slightly heterogeneous. We also note that the relative standard
of a sample. Another unexpected source of contamination is deviation is larger for lower intensity peaks like C1s, F1s, Fe2p3/2, and
demonstrated in the FeCo catalyst survey data (Figure 3), in which Co2p3/2 even though they have a relatively low statistical uncertainty.
fluorine appears as a contaminant. The origin of this contamina- We interpret this as being the result of additional uncertainty in
tion is the PTFE coating of a magnetic stirrer when the operator
prepared the sample. Table 1. XPS survey quantification results from five consecutive
XPS spectra are repeatable (precise) yet can be inaccurate. measurements of a single specimen with a mass fraction of 10 % of Fe
Statistical uncertainties in elemental quantification are ± 1 % or and Co supported on alumina (x = mass fraction; Δmass = uncertainty;
lower for large spectral peaks and greater for small peaks or noisy and smass = sample standard deviation)
spectra. Systematic uncertainties reach 50 % in the worst case and
Name Atomic (%) x (%) Δmass (%) smass (%) smass (%)
contribute the most to inaccuracy, particularly due to applying an (relative)
incorrect model. The peak area ratio (normalized by sensitivity
factors) is the most common quantification method that assumes Al2p 25.8 33.7 0.19 0.3 0.01
the sample is homogeneous, which is often erroneous for samples C1s 1.6 0.9 0.02 0.3 0.28
exposed to air because adventitious carbon contaminates the O1s 65.7 51.0 0.11 0.4 0.01
surface and/or a surface oxide layer forms. Furthermore, most F1s 2.4 2.2 0.02 0.5 0.22
samples are heterogeneous. Overlaying layers reduce the intensity Fe2p3/2 2.1 5.7 0.04 0.3 0.05
Co2p1/2 2.5 6.5 0.06 0.6 0.09
of high binding energy peaks more than low binding energy

4 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING MONTH 2019

 Springer Science & Business Media, New York 2012,
Table 2. XPS survey quantification results from the measurement of
105–195.
five different specimens of a sample with a mass fraction of 10 % of Fe
and Co supported on alumina [3] B. D. Ratner, D. G. Castner, “Electron Spectroscopy
for Chemical Analysis,” Surface Analysis: The Principal
Name Atomic (%) x (%) Δmass (%) smass (%) smass (%)
Techniques, J. C. Vickerman, I. S. Gilmore, Eds., John Wiley
(relative)
& Sons, Hoboken 2011, 47–113.
Al2p 25.3 33.3 0.18 0.6 0.02 [4] L. Cizaire, B. Vacher, T. L. Mogne, L. Cizaire, B. Vacher, T. L.
C1s 2.0 1.1 0.02 0.5 0.46 Mogne, J. Martin, L. Rapoport, A. Margolin, R. Tenne, Surf.
O1s 66.1 51.7 0.11 1.1 0.02 Coat. Tech. 2002, 160, 282.
F1s 2.4 2.2 0.02 0.3 0.11
[5] A. Aric, A. Shukla, H. Kim, S. Park, M. Min, V. Antonucci,
Fe2p3/2 2.0 5.4 0.04 0.4 0.07
Co2p1/2 2.2 6.2 0.06 0.8 0.13 Appl. Surf. Sci. 2001, 172, 33.
[6] E. A. Kraut, R. W. Grant, J. R. Waldrop, S. P. Kowalczyk, Phys.
Rev. Lett. 1980, 44, 1620.
[7] G. Beamson, D. Briggs, High Resolution XPS of Organic
determining the limits of the peaks in data treatment for low intensity Polymers: The Scienta ESCA300 Database, Wiley,
peaks compared to higher intensity peaks. Hoboken 1992.
The experimental uncertainty calculated from the standard
[8] J. Moulder, J. Chastain, Handbook of X‐ray Photoelectron
deviation from five repeated measurements does not provide
Spectroscopy: A Reference Book of Standard Spectra for
much information about the accuracy of the quantification
Identification and Interpretation of XPS Data, Physical Electro-
results. Indeed, we applied a homogeneous quantification model,
nics Division, Perkin‐Elmer Corporation, Eden Prairie
while the specimens studied are actually far from homogeneous
1992.
in the analysis volume. We might be able to model them as
alumina spheres with islands of Fe and Co on the surface, [9] K. R. Priolkar, P. Bera, P. R. Sarode, M. S. Hegde, S. Emura,
however, this would still be innacurate as the surface of alumina R. Ku‐Mashiro, N. P. Lalla, Chem. Mater. 2002, 14, 2120.
is rough and porous. This example shows that, in XPS analysis, [10] A. Mekki, D. Holland, C. McConville, M. Salim, J. Non‐Cryst.
accuracy is easy to obtain and quantify, though good repeatability Solid. 1996, 208, 267.
and precision in addition to chemical state identification make it [11] J. Das, S. Pradhan, D. Sahu, D Mishra, S. Sarangi, B. Nayak,
an interesting technique to compare samples. S. Verma, B. Roul, Physica B 2010, 405, 2492.
[12] C. Altavilla, E. Ciliberto, Appl. Phys. A-Mater. 2004, 79, 309.
CONCLUSIONS [13] S. Chen, H. Tanaka, J. Wood Sci. 1998, 44, 303.
XPS analysis identifies elements or chemical groups based on [14] J. Dalton, J. Preston, P. Heard, G. Allen, N. Elton, J. Husband,
binding energy at which photoelectron peaks are detected. We Colloid. Surface. A 2002, 205, 199.
deduce elemental content from the area of the peaks. The choice [15] G. N. Inari, M. Petrissans, J. Lambert, J. J. Ehrhardt,
of suitable experimental parameters and acquisition of spectra P. Grardin, Surface Interface Anal. 2006, 38, 1336.
with adequate signal to noise ratio minimizes uncertainties. [16] T. Hietala, N. Mozes, M. J. Genet, H. Rosenqvist, S. Laakso,
Accuracy improves by choosing an appropriate quantification Colloid. Surface. B 1997, 8, 205.
model. Recent commercial XPS systems perform depth profiling [17] A. M. Arroyo, M. L. Ruiz, G. V. Bernabeu, R. S. Román, M. G.
on a variety of materials including organics with new ionic Morales, L. Straus, J. Archaeol. Sci. 2008, 35, 801.
species for etching and chemical state imaging to a spatial [18] Y. Okazaki, Y. Ito, K. Kyo, T. Tateishi, Mat. Sci. Eng.‐A Struct.
resolution down to 3 μ m. New higher energy monochromatic 1996, 213, 138.
x‐ray sources such as AgLα are also available, which increase the
[19] B. Nisol, S. Watson, A. Meunier, D. Juncker, S. Lerouge,
information depth and create opportunities for non‐destructive
M. R. Wertheimer, Plasma Process. Polym. 2018, 15, 1700132.
depth profiling surface layers. Synchrotron radiation is the best
x‐ray source for XPS because the monochromatic beam produces a [20] G. Z. Xiao, M. Delamar, M. Shanahan, J. Appl. Polym. Sci.
narrower peak and, therefore, reduces peak overlap. For example, 1997, 65, 449.
the Ru 3d5 ∕ 2 and Ru 3d3 ∕ 2 binding energy of Ru metal is [21] W. Amelung, K. Kaiser, G. Kammerer, G. Sauer, Soil Sci. Soc.
4.49 × 10−17 J and 4.55 × 10−17 J, respectively, and 4.50 × 10−17 J Am. J. 2002, 66, 1526.
and 4.57 × 10−17 J, respectively for Ru4+, while it is 3.98 × 10−17 J [22] K. Sahre, K.‐J. Eichhorn, F. Simon, D. Pleul, A. Janke,
for carbon. With standard commercial Al anodes, the C1s peak G. Gerlach, Surf. Coat. Tech. 2001, 139, 257.
with a maximum at 4.56 × 10−17 J and a width of 1.6 × 10−19 J at [23] T. Mundry, P. Surmann, T. Schurreit, Fresenius’ Journal of
half the peak height, overlaps with the Ru3d peaks. This leads to Analytical Chemistry 2000, 368, 820.
increased uncertainty in the quantification of these elements.
[24] O. Hfft, S. Bahr, M. Himmerlich, S. Krischok, J. A. Schaefer,
V. Kempter, Langmuir 2006, 22, 7120.
[25] S. Kaya, H. Ogasawara, L. Kenslund, J.-O. Forsell, H. S.
REFERENCES Casalongue, D. J. Miller, A. Nilsson, Catal. Today 2013,
[1] Clarivate Analytics, Web of Science Core Collection, accessed 205, 101.
on 25 June 2018, http://apps.webofknowledge.com. [26] G. S. Patience, Can. J. Chem. Eng. 2018, 96, 2312.
[2] J. D. Andrade, “X-ray Photoelectron Spectroscopy (XPS),” [27] G. S. Patience, Experimental Methods and Instrumentation
Surface and Interfacial Aspects of Biomedical Polymers: for Chemical Engineers, 2nd edition, Elsevier, Amsterdam
Surface Chemistry and Physics, Joseph D. Andrade, Ed., 2017.

MONTH 2019 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING 5

[28] S. Hajati, S. Tougaard, Journal of Surface Analysis 2009,
15, 220.
[29] D. Briggs, M. Seah, Practical Surface Analysis: By Auger and
X‐ray Photo‐Electron Spectroscopy, Wiley, Chichester,
PA 1983.
[30] P.‐L. Girard‐Lauriault, T. Gross, A. Lippitz, W. E. Unger, Anal.
Chem. 2012, 84, 5984.
[31] H. Bonzel, C. Kleint, Prog. Surf. Sci. 1995, 49, 107.
[32] O. W. Richardson, K. T. Compton, The London, Edinburgh,
and Dublin Philosophical Magazine and Journal of Science
1912, 24, 575.
[33] J. Jenkin, R. Leckey, J. Liesegang, J. Electron Spectrosc. 1977,
12, 1.
[34] K. Siegbahn, Ark. Fys. 1954, 7, 86.
[35] P. Louyot, C. Neagoe, F. Galli, C. Pirola, G. S. Patience, D. C.
Boffito, Ultrason. Sonochem. 2018, 48, 523.
[36] G. S. Patience, C. A. Patience, F. Bertrand, Can. J. Chem. Eng.
2018, 96, 811.
[37] N. J. Van Eck, L. Waltman, Scientometrics 2010, 84, 523.
[38] V. Mazzieri, F. Coloma‐Pascual, A. Arcoya, P. LArgentire,
N. Fígoli, Appl. Surf. Sci. 2003, 210, 222.
[39] J. Zhang, L. Qu, G. Shi, J. Liu, J. Chen, L. Dai, Angew. Chem.
Int. Edit. 2016, 55, 2230.
[40] M. Cakici, K. R. Reddy, F. Alonso‐Marroquin, Chem. Eng. J.
2016, 309, 151.
[41] Y. Sun, Z.‐Y. Wu, X. Wang, Environ. Sci. Technol. 2016,
50, 4459.
[42] Y. Feng, Y. Li, L. Shi, M. Wu, J. Mi, Can. J. Chem. Eng. 2017,
95, 2087.
[43] C. Zuo, C. Li, T. Ge, X. Guo, S. Zhang, Can. J. Chem. Eng.
2017, 95, 2104.
[44] M. Khalid, F. Larachi, A. Adnot, Can. J. Chem. Eng. 2017,
95, 1875.
[45] F. Li, J. Liang, K. Wang, B. Cao, W. Zhu, H. Song, Can. J.
Chem. Eng. 2017, 95, 2012.
[46] S. Ansaloni, N. Russo, R. Pirone, Can. J. Chem. Eng. 2017,
95, 1730.
[47] G. Yang, H. Chen, H. Qin, X. Zhang, Y. Feng, Can. J. Chem.
Eng. 2017, 95, 1518.
[48] L. Wang, B. Shen, J. Zhao, X. Bi, Can. J. Chem. Eng. 2017,
95, 1069.
[49] C. J. Powell, M. P. Seah, J. Vacuum Sci. Technol. A 1990, 8, 735.

Manuscript received December 22, 2018; revised manuscript

received April 10, 2019; accepted for publication April 10, 2019.

6 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING MONTH 2019