See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/299575213

Chemical bonding states and solar selective characteristics of unbalanced

magnetron sputtered TixM1-x-yNy films

Article  in  RSC Advances · April 2016

DOI: 10.1039/C6RA02550A

CITATIONS READS

32 903

14 authors, including:

M Mahbubur Rahman Zhong-Tao Jiang

Jahangirnagar University Murdoch University
104 PUBLICATIONS   1,243 CITATIONS    186 PUBLICATIONS   2,999 CITATIONS   

SEE PROFILE SEE PROFILE

L.s. Chuah Z. F. Zhou

Universiti Sains Malaysia City University of Hong Kong
97 PUBLICATIONS   499 CITATIONS    170 PUBLICATIONS   3,898 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Population genetic analysis of Hilsa shad (Tenualosa ilisha) in Bangladesh View project

Solar selective surfaces View project

All content following this page was uploaded by Amun Amri on 06 May 2016.

The user has requested enhancement of the downloaded file.

 RSC Advances
View Article Online
PAPER View Journal | View Issue
Published on 01 April 2016. Downloaded by MURDOCH UNIVERSITY LIBRARY on 18/04/2016 03:29:28.

Chemical bonding states and solar selective

characteristics of unbalanced magnetron sputtered
TixM1
x
yNy films†
Cite this: RSC Adv., 2016, 6, 36373

M. Mahbubur Rahman,a Zhong-Tao Jiang,*a Paul Munroe,b Lee Siang Chuah,c

Zhi-feng Zhou,d Zonghan Xie,ef Chun Yang Yin,g Khalil Ibrahim,a Amun Amri,h
Humayun Kabir,i Md Mahbubul Haque,j Nick Mondinos,a
Mohammednoor Altarawnehk and Bogdan Z. Dlugogorskik

Received 28th January 2016
Accepted 14th March 2016
DOI: 10.1039/c6ra02550a
www.rsc.org/advances
Transition metal nitride TixM1
x
yNy (M ¼ Al or AlSi) based thin films are evaluated as solar selective surfaces by
correlating their spectral selective features with their crystal structure and chemical bonding state including
mechanical strength. Ti0.5N0.5, Ti0.25Al0.25N0.5, and Ti0.25Al0.2Si0.05N0.5 films were synthesized on AISI M2
steel substrates via closed field unbalanced magnetron sputtering technology. These were investigated
using XRD, SEM, XPS, UV-Vis, FTIR and nanoindentation techniques. Analysis of the optical properties
showed the solar absorptance, in the visible range, of the TixM1
x
yNy films improved significantly from 51%
to 81% with AlSi-doping and an increase of solar absorptance of up to 66% was recorded from films doped
with Al. Moreover, the Al doping can reduce the thermal emittance in the infrared range from 6.06% to
5.11%, whereas doping with AlSi reduces the emittance to ca. 3.58%. The highest solar selectivity of 22.63
was achieved with TiAlSiN films. Mechanical studies showed enhanced hardness by 32%; enhanced yield
strength by 16% and enhanced plastic deformation by 110% of Al and AlSi doped TiN matrix.

possesses two characteristics: high absorptance, a in the visible

Introduction region of solar spectrum (0.3 to 2.5 mm) and low emittance, 3,
A spectrally selective surface is, generally, used to improve the (i.e., high reection) in the infra-red (IR) region ($2.5 mm) at
photothermal conversion performance and, as such, it operating temperatures. An excellent selective surface maxi-
mises the absorption of incoming photons in the visible region
and minimises photon emission through thermal radiation in
a
Surface Analysis and Materials Engineering Research Group, School of Engineering the IR energy region. Such a surface can be designed by an
and Information Technology, Murdoch University, Murdoch, Western Australia absorber–reector assembly. In such an approach, the reector
6150, Australia. E-mail: Z.Jiang@Murdoch.edu.au is coated with a highly absorbing layer over the visible solar
b
School of Materials Science and Engineering, University of New South Wales, NSW
spectrum, while the infrared region is made transparent.
2052, Australia
c
Various types of metal nitrides based selective solar surfaces
Department of Physics, School of Distance Education, Universiti Sains Malaysia,
11800 Minden, Penang, Malaysia such as TiN, ZrN, HfN, TiAlN, TiAlON, NbAlN, NbAlON, MoAlN,
d
Department of Mechanical and Biomedical Engineering, City University of Hong Kong, and WAlN have been investigated by numerous groups.1–11 Over
Kowloon, Hong Kong, China the past few years, transition metal nitride based thin lms have
e
School of Mechanical Engineering, University of Adelaide, SA 5005, Australia attracted signicant research interest as selective solar surfaces
f
School of Engineering, Edith Cowan University, Joondalup, WA 6027, Australia in solar thermal conversion devices. Generally, the energy
g
School of Science and Engineering, Teesside University, Borough Road, conversion performance of a selective solar surface depends on
Middlesbrough, TS1 3BA, UK
h
the lm materials, lm design and fabrication technique used.
Department of Chemical Engineering, Universitas Riau, Pekanbaru, Indonesia
i
A multi-layer lm stack with mixture of metal nitride, metal
School of Metallurgy and Materials, University of Birmingham, Edgbaston,
Birmingham, B15 2TT, UK oxide and metal oxynitride lms e.g., TiAlN/AlON, and TiAl/
j
Materials Science Division, Atomic Energy Centre Dhaka, Dhaka 1000, Bangladesh TiAlN/TiAlON/TiAlO has been explored for the potential
k
School of Engineering & Information Technology, Murdoch University, Murdoch, commercial development of selective solar surfaces.3,12 TiAlN/
Western Australia 6150, Australia TiAlON/Si3N4 selective absorbers have been produced on
† Electronic supplementary information (ESI) available: The details of the various substrates such as copper, nickel, stainless steel, glass
deposition parameters; and bonding states, elements/compounds, binding
and nimonic alloys.13 However, these materials are yet to be
energies, FWHMs, and percentages of the different components from XPS
tting curves of these lms are presented in Tables S1 and S2. See DOI:
commercialized for solar energy conversion applications.3,9,14 In
10.1039/c6ra02550a

This journal is © The Royal Society of Chemistry 2016 RSC Adv., 2016, 6, 36373–36383 | 36373

RSC Advances
solar selective surface applications, transition metal oxides
based thin lms also received signicant research interests.15–19
Various synthesis methods such as evaporation, electrode-
position, chemical conversion, chemical vapour deposition and
magnetron sputtering have been employed in manufacturing
selective solar surfaces. Owing to its advantages in large area
deposition, dry, clean and environment friendly, magnetron
Published on 01 April 2016. Downloaded by MURDOCH UNIVERSITY LIBRARY on 18/04/2016 03:29:28.
sputtering technique is widely used for synthesizing selective
solar absorbers. Moreover, magnetron sputtered lms have
already shown promising optical behaviour that potentially
allow them to be utilized in solar selective surfaces.13,20 The
optical properties of transition metal based nitride lms can be
regulated by selecting and tuning the stoichiometry of the
nitrides. Furthermore, the density of free electrons in the d-
bands of transition metals is impacted by stoichiometry.11
Introduction of an appropriate dopant (e.g. Al or AlSi) in the
transition metal nitride alters the bonding structure of the
matrix which, in turn, modies the optical properties and
electrical behaviour of such lms. For this reason, a high solar
absorptance and low thermal emittance can be achieved by
TiAlN and TiAlSiN lms. Electronic properties of magnetron
sputtered Al, Si and Ni doped CrN thin lm coatings have been
reported in previous studies.21–23 The design, fabrication and
characterization of a diverse range of oxides, nitrides and oxy-
nitride based functional thin lm coatings have been exten-
sively studied for their structural, thermal, optical, electrical,
nanomechanical, photocatalytic and magnetic properties by
numerous authors.24–29
The optical properties can be also regulated by integrating
another metal component into the transition metal nitride
systems which transforms the bonding nature from metallic to
covalent.11 Since the solar selective surfaces are the easiest and
most direct way of harvesting solar energy, in recent years,
transition metal nitrides have received signicant attention for
mid and high temperature applications.3–6,8,9,20,30–34 Spectrally
selective behaviour of transition metal nitride lms have been
previously studied.10 Studies of magnetron sputtered TiAlN/
TiAlON lms synthesized onto a metal substrate exhibited
high absorptance and low emittance were reported by Barshilia
et al.5 The absorptance of the lms was substantially enhanced
to 0.95 by adding a Si3N4 anti-reective coating on the top of
these lms.5 A higher thermal stability and high solar selectivity
of 13.57 was achieved by this tandem selective surface. The
electronic properties of titanium nitride-based materials have
been extensively studied under different experimental and
theoretical approaches, but research on optical properties of
these materials have been relatively scarce35–37 and this is
especially true for (the lack of) studies on optical selectivity.5,38–42
In a recent study, we report on the post annealing effects on the
solar selective characteristics and chemical bonding states of
sputtered TiAlSiN thin lm coatings.43 Considering the above
circumstances, the objective of the present study is to investi-
gate the potential development of magnetron sputtered Tix-
M1
x
yNy (where M ¼ Al, AlSi, x ¼ 0.50, 0.25 and y ¼ 0.50) lms
as the selective solar surfaces. UV-Vis and Fourier transform
infrared reectometry (FTIR), X-ray diffraction (XRD), X-ray
photoelectron spectroscopy (XPS), band-gap analysis,
36374 | RSC Adv., 2016, 6, 36373–36383
View Article Online
Paper
scanning electron microscopy (SEM) and nanoindentation
measurements will be undertaken to investigate and charac-
terise the crystal structure, surface bonding structure, surface
morphology, optical and mechanical parameters of the as-
deposited lms.
Experimental section
Thin lm deposition technique and synthesis conditions
Three thin lm lms (with nominal compositions Ti0.5N0.5,
Ti0.25Al0.25N0.5, and Ti0.25Al0.2Si0.05N0.5) were deposited onto
AISI M2 tool steel substrates via closed eld unbalanced
magnetron sputtering system (UDP650, Teer Coating Limited,
Droitwich, Worcestershire, UK). The magnetron sputtered
system is equipped with a four-target conguration. The
substrates were ultrasonically cleaned in acetone and methanol
solution and then dried with nitrogen gas, before putting them
into the vacuum chamber. A background pressure of 2  10
6
Torr was attained in the chamber before the deposition process.
The substrates were sputter cleaned with Ar plasma at a bias of

450 V for 30 minutes. In order to achieve a homogeneous lm
thickness, the substrates were rotated at a speed of 10 rpm
throughout the synthesis process. The Ti0.5N0.5 lm was
produced at a Ti target current of 8 A, while the Al target current
was varied between 0 and 7 A for the synthesis of the TiAlN, and
TiAlSiN lms. However, for the TiAlSiN lms, a Si target current
of 1.5 A was also applied. The nitride lms were prepared in
a mixed Ar + N2 mixed gas atmosphere. Prior to any measure-
ments, all the samples were mechanically cleaned in ultra-high
vacuum via a diamond needle le and tungsten-wire brush. The
details of the deposition parameters for the sputtered lms are
displayed in Table S1 (see the ESI†).
XRD studies
X-ray diffraction (XRD) measurements were performed to
determine the identity of the various phases in the lms. A
Bruker Advance D8 X-ray Diffractometer equipped with a Lyn-
xEye detector was used for this purpose. The XRD machine was
operated at 40 kV and 40 mA at room temperature. The XRD
data of the lms were acquired with Cu-Ka radiation (l ¼ 1.5406
Å) over a 2q range of 30 to 50 with a step size of 0.01 .
Mechanical properties of the sputtered TixM1
x
yNy lms
Mechanical properties, such as Young's modulus and hardness,
of the thin lms were measured by a depth-sensing indentation
system (Ultra-Micro Indentation System, UMIS-2000, CSIRO,
Australia) equipped with a Berkovich indenter. The indenter tip
was calibrated by conducting single-cycling indentation tests on
fused silica (a standard material having a known modulus of 72
GPa) at different loads. Load–unload tests were run in a closed-
loop under load control to determine the mechanical properties
of the samples. A maximum load of 20 mN was applied in 10
increments. The loading rate set to 2.5 mN s
1 represents the
static response of the materials. Following each increment there
were 10 decrements, from which the Young's modulus and
This journal is © The Royal Society of Chemistry 2016

Paper
hardness were calculated. Five indentation tests were per-
formed on each specimen.
SEM studies of the lms
The surface morphology of the lms were investigated with
scanning electron microscopy (SEM) using a PHILIPS XL 20
Published on 01 April 2016. Downloaded by MURDOCH UNIVERSITY LIBRARY on 18/04/2016 03:29:28.
(Eindhoven, The Netherlands) instrument. The SEM was oper-
ated at 25 kV. The elemental composition of the lms was ob-
tained by energy dispersive spectroscopy (EDS) from an
OXFORD Link ISIS unit attached to the SEM.
XPS analysis of sputtered TixM1
x
yNy thin lms
XPS analysis provides comprehensive information on the
chemical state, bonding structure and elemental composition
of the outermost layers of surfaces. XPS data of all thin lms was
acquired using a Kratos Axis Ultra XPS spectrometer (Kra-
tos,Manchester, UK) with Al-Ka monochromatic radiation (hn ¼
1486.6 eV) source and at an operating power of 10 mA and 15
kV. The XPS machine is also equipped with a cold stage, and an
Ar ion gun for etching and depth proling. The samples were
mounted on steel sample holder. The chamber pressure was
reduced to 2.9  10
9 Torr and maintained at that level during
the analysis. XPS survey spectra were acquired with both etched
(6 minutes of ion sputtering) and unetched samples. High
resolution XPS spectra were recorded immediately aer 6
minutes of Ar+ sputtering. Etching was done to remove any
surface oxide layers and to reduce the Ti4+ ions in order to lower
oxidation states. A pass energy of 20 eV was used for high
resolution Ti2p, Al2p, Si2p, O1s, and N1s photoelectron lines
measurement. CASA XPS (Version 2.3.15) soware was used for
processing and tting the peaks of the XPS spectra.
Solar selectivity studies of the sputtered thin lms
Optical properties of the lms were analysed by measuring the
optical reectance as function of wavelength from ultraviolet
through visible to the infrared range via UV-Vis and FTIR
spectrometers. A double-beam UV-Vis spectrophotometer
(Model: UV-670 UV-Vis spectrophotometer, JASCO, USA)
equipped with a unique, single monochromator design
covering a wavelength range from 250 to 2500 nm measured
solar absorptance of the lms. The monochromator is designed
with 1200 grooves per mm grating and a photo multiplier tube
(PMT) detector. A PERKIN Elmer Spectrum 100 FTIR Spec-
trometer (USA) was used for measuring the reectance of the
lms over the wavelength range from 2.5–15 mm.
Results and discussion
X-ray diffraction (XRD) data for Ti0.5N0.5, Ti0.25Al0.2N0.5, and
Ti0.25Al0.2Si0.05N0.5 magnetron sputtered thin lm lms are
displayed in Fig. 1. XRD peaks at approximately 2q ¼ 44 are due
to the steel substrate. Ti0.5N0.5 lms have a single TiN phase,
with a face centred cubic (FCC) structure, and a maximum
intensity peak at 2q ¼ 36.63 consistent with a (111) Bragg
reection with a minor peak at approximately 42.4 (JCPDS 87-
0633). In contrast, extra phases were identied for the
This journal is © The Royal Society of Chemistry 2016
View Article Online
RSC Advances
Fig. 1 XRD spectra of the as-deposited TixM1
x
yNy sputtered films
(TiN, TiAlN and TiAlSiN respectively stand for Ti0.5N0.5, Ti0.25Al0.25N0.5
and Ti0.25Al0.2Si0.05N0.5).
Ti0.25Al0.2N0.5, and Ti0.25Al0.2Si0.05N0.5 lms. The main phase for
all samples is TiN with a (111) preferred orientation. However,
with the addition of Al and the simultaneous addition of Al + Si
to TiN, the principal (111) peak shis towards higher Bragg
angles by 0.68 and 0.52 respectively, due to Ti sites being
occupied by smaller Al atoms to form a TiAl(Si)N solid solution.
These peak shis reveal lattice contraction, of the TiN phase, by
replacement of Ti atoms with Al and Si atoms, and are listed in
Table 1. Doping the cubic TiN sample with Al and simultaneous
Si + Al show extra XRD peaks at approximately 2q ¼ 36.06–36.71
and 38.28 that are identied respectively as hexagonal AlN
(JCPDS 89-3446) and hexagonal Si3N4 (JCPDS 83-0701) phases.
The formation of h-AlN in Ti0.25Al0.2N0.5 and simultaneous
formation of h-AlN and h-Si3N4 in Ti0.25Al0.2Si0.05N0.5 suggests
characteristic evidence of spinodal decomposition in these
lms through the addition of these elements to TiN. This type of
spinodal decomposition of TiAlN reported in previous
studies44,45 is also associated with the grain renement and
modications of microstresses within AlN and Si3N4 domains.
XRD analysis of the TixM1
x
yNy lms
XRD data was used to estimate the average crystallite size for the
TiN phase by the Debye–Scherrer equation,
0:9l
D¼ (1)
b cos q
where l is the X-ray wavelength, b is the full width at half
maximum (FWHM) and q is the diffraction angle. Estimates of
the crystallite size are listed in Table 1.
Table 1 Structural properties, for TiN phase, of TixM1
x
yNy films
Grain size (10%), Lattice parameter,
Samples D (nm) a (nm)
Ti0.5N0.5 24 0.4257
Ti0.25Al0.25N0.5 22 0.4215
Ti0.25Al0.2Si0.05N0.5 19 0.4195
RSC Adv., 2016, 6, 36373–36383 | 36375

RSC Advances
Analysis of Table 1 shows a decrease in lattice parameter and
grain size of the TiN phase as Al and Si dopants are added to
TiN. The contraction of lattice parameter is due to the extra h-
AlN and h-Si3N4 phases, formed from Al and Si interaction
with N, around the grain boundaries of the TiN phase and the
atomic radii of Al (0.143 nm) and Si (0.134 nm) being smaller
than Ti (0.147 nm). The difference in atomic radius between
Published on 01 April 2016. Downloaded by MURDOCH UNIVERSITY LIBRARY on 18/04/2016 03:29:28.
titanium and the dopants (Al and/or Si) results in the solution
strength which causes shrinkage in the lattice parameters and
thereby a change in the internal stress of these lms.46 A subtle
peak broadening and reduction of the residual stress in the
crystal lattices also results in a slight reduction in grain sizes of
the lms.
Analysis of the mechanical properties
The mechanical properties e.g., hardness and Young's modulus,
of the as-deposited thin lms, were determined from nano-
indentation testing using a depth indentation system and are
presented in Table 2.
Table 2 shows that aluminium addition, increase the hard-
ness (H) of TiN by 32%, (i.e., from 17.38 to 23.08 GPa), whilst
the Young's modulus (E) is enhanced by 23%. Using the
measured H and E values, the yield strength (H/E) and plastic
deformation factor (H3/E2) of the lms were calculated. Whilst
the additions of Al and Si to TiN led to a 8 and 16% increase
in yield strength; and a 52 and 110% increase in plastic
deformation factor. These data suggest that these lms also
possess good wear resistance. With additions of Al and AlSi,
formation of AlN and Si3N4 phases (XPS results) help inhibit the
grain boundary sliding and thereby increasing toughness of
these lms. Furthermore the addition of subsequent dopants
(Al/AlSi), the compressive residual stress of the lms is
increased which also enhances hardness.47 Thus, the higher
hardness values developed in the TiAlN and TiAlSiN lms are
due the combined inuence of modication of the grain
boundary structure and enhanced residual stresses.
Investigations of the surface morphology
The surface morphology of the lms were analysed by SEM in
order to explore the effect of surface structure on optical prop-
erties. The TiN lms exhibited a number of randomly dispersed
pores around the surface of the lms, together with a distribu-
tion of non-uniform grains. The average grain size of this lm,
within 15% measurement error, was observed to be around 40
nm (15%) (see Fig. 2(a)). With both the addition of Al and the
simultaneous addition of Al and Si to TiN, most of the pores
were no longer present and grains were rened (see Fig. 2(b)
Table 2 Mechanical properties of the TixM1
x
yNy films achieved via nanoindentation
Hardness H Young's
Samples (GPa) modulus E (GPa)
Ti0.5N0.5 17 281
Ti0.25Al0.25N0.5 23 347
Ti0.25Al0.2Si0.05N0.5 27 378
36376 | RSC Adv., 2016, 6, 36373–36383
View Article Online
Paper
and (c)). Addition of Al reduced the grain sizes down to 30 nm
(10%), while simultaneous addition of Al and Si resulted in
further decrease in grain size to 27 nm (10%). Concurrently
some ne grains are also developed around the surfaces of all
the lms. The SEM studies are consistent with XRD results,
where the average grain sizes of these lms were found to be
reduced systematically with the addition of Al and AlSi dopants.
Moreover with the addition of dopants, the surface morphology
and shape of the grains are both altered with irregular mould-
like structures seen around the surface of the lms. Further
inspection conrms that the higher surface roughness is
attained by the TiAlN lms. It suggests that Al and AlSi addi-
tions hinder the diffusion of Ti and N atoms during deposition
processes and thereby stimulate grain growth eventually acting
as nucleation sites for lattice distortions. Similarly, in TiAlSiN
lms, Al and Si reduce the mobility of both Ti and N atoms
thereby slowing down their diffusion.
The surface roughness produces a gradient in the optical
constants, rather than a sharp discontinuity at the interface.
Kang and Kim (1999) indicated that the gradient in the optical
constants strengthen optical behaviour.48 It was further noticed
that the surface of TiAlN and TiAlSiN became more compact
and denser with the addition of Al to the Ti and AlSi to Ti,
respectively. The dense and closely packed structures show
more stability as a selective surface at high temperatures. These
kind of sputtered lms are mechanically more reliable than
similar lms with large columns and open pores.49
EDS provided approximate elemental compositions of these
lms (see Table 3 below). The data are broadly consistent with
the expected compositions of these lms.
Energy band-gap analysis of the lms
The wavelength of light that a material absorbs is the charac-
teristic of its chemical assembly. Specic regions of the electro-
magnetic spectrum are absorbed by exciting specic types of
molecular and atomic motion to higher energy levels. Absorption
of ultraviolet and visible light is related with excitation of elec-
trons, in both atoms and molecules, to higher energy states. Most
molecules require very high energy radiation. The dependence of
optical absorption coefficient on the photon energy helps to
study the band structure and the type of electron transition
involved in absorption process.50 In crystalline and amorphous
materials the photon absorption obeys the Tauc relation,
ahn ¼ A(hn
Eg)n (2)
where, a is the absorption coefficient, hn is the incident photon
energy, h is Planck's constant, n is the frequency of incident
Yield strength Plastic deformation
H/E factor H3/E2 (GPa)
0.062 0.067
0.067 0.102
0.072 0.141
This journal is © The Royal Society of Chemistry 2016

Paper
Published on 01 April 2016. Downloaded by MURDOCH UNIVERSITY LIBRARY on 18/04/2016 03:29:28.
Fig. 2 SEM of sputtered: (a) Ti0.5N0.5 (b) Ti0.25Al0.25N0.5, and (c) Ti0.25Al0.2Si0.05N0.5 films.
light, A is an energy independent constant, Eg is the band-gap
energy and n is an index depending on the type of optical
transition involved in photon absorption.
The index n ¼ 1/2 and 2 for allowed direct and indirect
transitions respectively. Thus, the direct transition energy gap
(direct band-gap energy) can be obtained by plotting hn vs.
(ahn)2 and extrapolating the linear portion of the curve to (ahn)2
¼ 0 in the energy axis. Similarly the indirect energy band-gap
can be found by plotting hn vs. (ahn)1/2 and extrapolating the
linear portion of the curve to (ahn)1/2 ¼ 0 in the x-axis.
In the present study UV-visible absorptance spectra, pre-
sented in Fig. 3, were used to investigate the solar absorptance,
optically induced transition and to acquire information about
the band structure of the lms. Plots from Fig. 3 display that in
the visible range of the spectra, the solar absorbance of these
lms increases with the increase in wavelengths and a broad-
ening of the absorption peak. At a wavelength of 500–530 nm,
the absorbance values attain their maximum and then decrease
linearly until the near infra-red region. The spectral dependence
of a on the photon energy (hn) helps to understand the band
structure and the type of electron transition involved in the
absorption process. Photon absorptions in crystalline and
amorphous materials are known to follow the Tauc relation as
given in eqn (2). From this equation, the band-gap involved in
direct transition is estimated by plotting the photon energy hn
vs. (ahn)2 and then extrapolating the linear portion of the curve
to (ahn)2 ¼ 0. Fig. 4 indicates two different slopes in the
experimental photon energy ranges of these lms which can be
interpreted as the presence of direct and indirect optical tran-
sitions involved in these lms. Plots of hn vs. (ahn)2 for TiN,
TiAlN and TiAlSiN lms are shown in Fig. 4(A) and energy gaps
involved in direct transitions were estimated from the
Table 3 EDS analysis for the elemental compositions of the as-deposited TixM1
x
yNy magnetron sputtered films
Element %
Sample compositions of Ti
Ti0.5N0.5 48.4
Ti0.25Al0.25N0.5 23.5
Ti0.25Al0.2Si0.05N0.5 22.9
This journal is © The Royal Society of Chemistry 2016
View Article Online
RSC Advances
intercepts of the linear part of the curves extrapolated to zero in
the photon energy axis.
From Fig. 4(B), hn vs. (ahn)1/2, the intercept of the linear part
of the curves extrapolated to zero in the x-axis gives the energy
gap involved in indirect transition. The energy values computed
for direct and indirect transitions for TiN, TiAlN and TiAlSiN
lms are presented in the Table 4. The results indicate a slight
decrease in the energy band-gaps as the Al and Al + Si are
incorporated in the TiN system.
XPS analysis of the sputtered TixM1
x
yNy thin lms
Table 5 summaries the details of the atomic compositions of the
lms as obtained from the XPS studies both before and aer
etching. It is apparent that 6 minutes of Ar+ etching removed the
surface contamination (see the 3rd and 4th columns in Table 5).
That is, the oxygen content of these lms decreased from 50
at% before etching to 10 at% aerwards.
High resolution XPS spectra of the characteristic Ti2p, Al2p,
Si2p, N1s, and O1s peaks at different binding energies positions
are shown in Fig. 5–7. The Shirley method51 was used to remove
the background of spectra. The curve-tting of the high reso-
lution peaks was performed using a nonlinear least squares
tting of a mixed Gaussian–Lorentzian product function via
CASA version 2.3.5.1 XPS soware. The photon energy scales of
all the spectra were calibrated by using the C1s peak of C–H (or
C–C) at 285 eV. The appearance of oxygen peaks reveals the
existence of residual oxygen in the vacuum chamber, or
subsequently following heat treatment in air.
Curve tting of the core-level photoelectron lines of Ti, Al, Si,
N and O was used to estimate the relative contributions of Ti–N,
Al–N, Si–N, Ti-related oxides, Al-related oxides, Si-related oxides
Element % Element % Element %
of N of Al of Si
51.6 — —
47.50 29.0 —
45.2 26.8 5.1
RSC Adv., 2016, 6, 36373–36383 | 36377
 View Article Online

RSC Advances Paper

Table 4 Band-gaps of different sputtered TiN, TiAlN and TiAlSiN films

for direct and indirect allowed transitions (TiN, TiAlN and TiAlSiN
respectively stand for Ti0.5N0.5, Ti0.25Al0.25N0.5 and Ti0.25Al0.2Si0.05N0.5)

Band-gap Band-gap
Films name (direct transition), eV (indirect transition), eV

Ti0.5N0.5 2.65 0.90

Published on 01 April 2016. Downloaded by MURDOCH UNIVERSITY LIBRARY on 18/04/2016 03:29:28.

Ti0.25Al0.25N0.5 2.45 0.45

Ti0.25Al0.2Si0.05N0.5 2.35 0.25

Table 5 Details of the elemental compositions of the TixM1
x
yNy

unbalanced magnetron sputtered films acquired via XPS survey scans

Atomic percentages of the

Fig. 3 Energy dependence of UV-Vis absorption spectra of sputtered
elements
TiN, TiAlN and TiAlSiN films (TiN, TiAlN and TiAlSiN respectively stand
for Ti0.5N0.5, Ti0.25Al0.25N0.5 and Ti0.25Al0.2Si0.05N0.5). Aer 6 minutes
Samples Elements Before etching etching

Ti0.5N0.5 Ti 21.7 45.5

N 27.5 44.6
O 50.8 9.9
Ti0.25Al0.25N0.5 Ti 8.3 22.3
N 24.9 42.7
Al 21.2 21.6
O 45.6 14.4
Ti0.25Al0.2Si0.05N0.5 Ti 13.2 22.9
N 25.3 43.1
Al 14.8 18.4
Si 3.2 4.2
O 43.5 11.4

Ti2p1/2 peaks separated by a binding energy of 5.6 eV. The

Fig. 4 Estimation of band-gaps of the sputtered TiN, TiAlN and TiAlSiN
films for: (A) direct and (B) indirect allowed transitions (TiN, TiAlN and
TiAlSiN respectively stand for Ti0.5N0.5, Ti0.25Al0.25N0.5 and
Ti0.25Al0.2Si0.05N0.5).
to the lms. The details on the bonding states, elements/
compounds, binding energies, FWHMs, and percentages of
the different components from XPS tting curves of these lms
are presented in Table S2 in the ESI.† The high-resolution Ti2p
spectra show the spin–orbit doublet which gives out Ti2p3/2 and
doublet separation value 5.6 eV between the two lines of Ti2p3/2
and Ti2p1/2 reveals that the bonding state of Ti is in the form of
TiO2.52 In the TiN lms (see Fig. 5); the deconvolution of the
Ti2p spectrum can be assigned as three different chemical
bonding states. The binding energy of the features ranging from
455.1 to 458.6 eV is assumed as the presence of the Ti2p3/2 state.
The rst component at 455.1 eV (Ti2p3/2 together with Ti2p1/2 at
460.7 eV) corresponds to Ti–N bonds in TiN.53 The second
component at 456.8 eV is related to Ti2O3 bonds and another
component at 458.4 eV may be attributed to suboxides based on
TiO2 compounds.44 This reveals the fact that the surface of these
lms has been oxidized. In the case of the Ti2p photoelectron
lines of TiAlN, and TiAlSiN lms, a shi to lower binding energy
and peak broadening indicates the appearance of reduced Ti
ions. This is because of a larger number of oxygen vacancies
around the surface of the lms. It is also expected that the
possible formation of bonding states of Ti–Al(N) and Ti–AiSi(N)
phases may cause the shiing and broadening of Ti2p peaks.
Furthermore, some Ti may also exist as Ti3+ ions in the form of
TiN. Aer a tting, the N1s core level photoelectron spectra
exhibited three components located at 396.8, 397.6 and 399 eV:
the rst component from the low binding energy side corre-
sponds to nitrogen in TiN, AlN, and/or SiN bonds, the second
fragment (in the middle) originates from titanium oxynitrides
and the relatively low intense component centred at 399 eV is
due to the presence of impurities/pores/defects in the lms

36378 | RSC Adv., 2016, 6, 36373–36383 This journal is © The Royal Society of Chemistry 2016

Paper
Published on 01 April 2016. Downloaded by MURDOCH UNIVERSITY LIBRARY on 18/04/2016 03:29:28.
Fig. 5 Core level high resolution XPS spectra of Ti0.5N0.5 films and the peak fittings of (a) Ti2p (b) N1s and (c) O1s.
such as free N2 or nitrogen associated with carbons. However,
incorporation of aluminium and simultaneous addition of
aluminium and silicon to the TiN system results in TiN/TiAlN,
AlN and Ti-oxynitrides or Ti-and/or metal oxynitride phases at
similar binding energy positions. In the case of TiN lms,
curves of O1s core level electrons were assigned into three
different fragments: the rst component at 530.7 eV is ascribed
to TiO2, the second component at 531.6 eV is assigned as the
elemental O (or physically adsorbed molecular H2O) and the
third component at 532.8 eV is identied to be originated from
TiON phase.
In the TiAlN and TiAlSiN lms, the deconvolution of Al2p
spectrum can resolved into two components: the rst one
located at 74.2 eV and the second component is seen at
Fig. 6 Core level high resolution XPS spectra of Ti0.25Al0.25N0.5 films and the peak fittings of (a) Ti2p, (b) Al2p, (c) N1s and (d) O1s.
This journal is © The Royal Society of Chemistry 2016
View Article Online
RSC Advances
a binding energy position of 75.0 eV. The rst component can
be assigned to AlN, while the component seen at 75.0 eV is
mainly due to the Al–O bonding in Al2O3.54 Similar features were
also observed in the Al2p spectrum of the TiAlSiN lms (as seen
in Fig. 6 and 7; and in Table S2 in the ESI†). The Si2p core level
electrons of the TiAlSiN lms are tted with two components:
the rst one at 101.7 eV is attributed to Si3N4 and the second
peak at 103 eV is identied to be the characteristic for Si–O
bonds in SiOx phase.
Solar selectivity results
Using the reectance, R(l) data acquired via UV-Vis and FTIR
measurements, the solar absorptance and the solar emittance
of a material can be conveniently estimated.55 For a range of
RSC Adv., 2016, 6, 36373–36383 | 36379
 View Article Online

RSC Advances Paper

Published on 01 April 2016. Downloaded by MURDOCH UNIVERSITY LIBRARY on 18/04/2016 03:29:28.

Fig. 7 Core level high resolution XPS spectra of Ti0.25Al0.20Si0.05N0.5 and the peak fittings of (a) Ti2p, (b) Al2p, (c) Si2p (d) N1s and (e) O1s.

solar wavelengths, the total solar absorptance, a(l) is dened as the Duffie and Beckman method.55 The optical absorptance
a weighted fraction between absorbed radiation and incoming evaluated as an average value computed from 20 values corre-
solar radiation (Isol), while thermal emittance, 3(l) is dened as sponding to the solar spectrum range between 190 nm and 2500
a weighted fraction between emitted radiation and the Planck nm are shown in Table 6. As seen in Fig. 8(A), the lms exhibit
black-body distribution (Ip):56 moderate absorptance (a) of >70% to UV radiation, high
ð 2:5 absorptance of >80% in the visible spectrum and high reec-
Isol ðlÞð1
RðlÞÞdl tance (up to 94%) in the near-infrared region of the solar
a ¼ 0:19 ð 2:5 (3) spectrum. In the shorter wavelength region, all the reectance
Isol ðlÞdl spectra show a wavy curve with interference peaks along with
0:19
a sharp absorption edge. In the visible region, up to 650 nm,
ð 15:4 compared to the TiN and TiAlN lms, the reectivity of the
Ip ðlÞð1
RðlÞÞdl TiAlSiN lms was found to be decreased signicantly, whilst
3¼ 2:5
ð 15:4 (4) above the visible range, the reectance of the TiAlSiN lm was
Ip ðlÞdl quite steady over the near-infrared range up to 2500 nm. In all
2:5
three lms, the highest absorption of solar radiation was
measured in the visible spectrum and the lowest absorptance
The solar spectrum, Isol has an air mass (AM) of 1.5 in (>90% reectance) in the near-infrared region. The major
accordance with the ISO standard 9845-1 (1992). The gure of difference between the reectance spectra of the TiAlSiN lm
merit of a selective surface is expressed in terms of a parameter and the other two samples is that the reduction in infrared
known as the selectivity, s. The selectivity of a material is reectance is more pronounced. For this reason, the overall
dened by the ratio of the solar absorptance (a) to the thermal absorptance of the lm with AlSi-additions was enhanced
emittance (3):57 signicantly. Our observation is found to be consistent with
a previous investigations.13,58 High absorptance of the lm makes
s¼ (5) it an ideal candidate for use as a selective solar absorber. A
3
valuable approach of developing multilayer or gradient layer
structures of selective solar surfaces are thoroughly discussed in
The UV-Vis reectance spectra of the TiN, Ti0.25Al0.25N0.5 and
ref. 59. The increase in solar absorption is associated with the
Ti0.25Al0.20Si0.05N0.5 lms deposited on M2 steel substrates are
gradual increase in the refractive index of the lms from the
demonstrated in Fig. 8(A). The UV-Vis reectance spectra were
outer surface to the substrate. In the case of single layered lms
used to estimate the optical absorptance (a) of these lms using
on a steel substrate, we achieved the highest average solar

36380 | RSC Adv., 2016, 6, 36373–36383 This journal is © The Royal Society of Chemistry 2016

Paper
Published on 01 April 2016. Downloaded by MURDOCH UNIVERSITY LIBRARY on 18/04/2016 03:29:28.
Fig. 8 (A) UV-Vis and (B) FTIR reflectance spectra of the sputtered TixM1
x
yNy films (TiN, TiAlN and TiAlSiN respectively stand for Ti0.5N0.5,
Ti0.25Al0.25N0.5 and Ti0.25Al0.2Si0.05N0.5).
Table 6 Solar absorptance, emittance and solar selectivity of the
sputtered TixM1
x
yNy films
Solar absorptance Emittance Solar selectivity
Name of the samples (a) (3) (s ¼ a/3)
Ti0.5N0.5 0.51 0.0606 8.42
Ti0.25Al0.25N0.5 0.66 0.0511 12.92
Ti0.25Al0.2Si0.05N0.5 0.81 0.0358 22.63
absorptance of 81% which may be substantially enhanced by
designing multilayered transition metal nitride lms. Further-
more, an addition of a dielectric antireection layer on the top
of the outer lm surface would further increase the photo
absorption capacity of such lms.
The infrared reectance spectra of the TiN, Ti0.25Al0.25N0.5
and Ti0.25Al0.20Si0.05N0.5 lms measured via FTIR spectroscopy
are presented in Fig. 8(B). The infrared thermal emittance (3) of
these lms calculated via the Duffie and Beckman method55 are
displayed in Table 6. As can be seen from Table 6, the thermal
emittance of the lms decreased with the addition of Al and the
simultaneous addition of Al and Si to TiN. When Al and Si are
added to TiN, the emittance decreased because the formation of
the surface oxidation layer (Al2O3, SiO2) resulted in high
infrared reectivity.60 This is consistent with Al2O3, SiO2
observed the XPS analysis.
TiN possesses metallic character, while AlN and Si3N4 are
both dielectric in nature. Thus, the addition of Al and AlSi to the
TiN matrix also transforms the electronic properties of TiAlN
and TiAlSiN from metallic to a more dielectric character. The
solar absorptance and emittance of the lms is strongly related
to the band-gap of the lms. TiN matrix possesses a band-gap of
2.65 eV. The addition of Al and AlSi to the TiN phase reduces the
direct energy band-gap from 2.65 eV to 2.45 and 2.30 eV,
respectively. This designates that addition of Al and AlSi,
effectively modies the overall band structures of TiN lms, and
thereby the overall solar selectivity of the TiAlN and TiAlSiN
lms is increased. Moreover, with addition of Al to TiN lms
change their colour from gold to blue grey. This change in
colour also enhances the optical absorption of TiAlN lms and
thereby solar selectivity is increased. Further, the occurrence of
This journal is © The Royal Society of Chemistry 2016
View Article Online
RSC Advances
Al2O3 (as conrmed by XPS studies) on the lm surface, which
displays ionic character, has high electrical resistivity, good
chemical stability, and a refractive index between 1.5 and 1.9 in
the visible range of the solar spectrum also play signicant role
in improving the optical properties of these lms.61,62 In addi-
tion, Al and AlSi dopants result in a change in structure of
grains and grain boundaries, surface chemical bonding states
and surface oxide layers which contributes to the surface
morphology of the lms as observed from SEM studies. This
may also affect the overall optical constants, such as refractive
index, optical dielectric constant and optical conductivity of the
materials.
The solar selectivity (s) values of our lms are listed in Table
6. The solar selectivity of TiN lm is improved signicantly by
the addition of both Al and Al/Si. Barshilia et al.13 documented
sputtered TiAlN/TiAlON/Si3N4 lms on a stainless steel
substrate for solar selective absorbers and obtained a solar
selectivity of 7.4. These works utilized both a solar spectrum
reectometer and emissometer to compute the solar absorp-
tance in the visible spectrum and thermal emittance in the
infrared range. In addition to lms on stainless steel substrates,
other researchers also deposited the TiAlN/TiAlON/Si3N4 selec-
tive absorbers onto glass, copper and nickel substrates and the
solar selectivity values recorded were 13.3, 13.6 and 6.7
respectively.13 In the same study,13 TiAlN sputtered lms were
synthesized onto a copper substrate and a solar spectrum
reectometer and emissometer were used to measure the solar
absorptance in the visible spectrum and thermal emittance in
the infrared range. An overall solar selectivity of 12.9 was
attained by TiAlN solar selective lms. In comparison with
similar solar selective absorbers developed onto stainless steel
and other substrates, our lms were found to exhibit superior
solar selectivity. The solar selectivity of these lms depends on
the thicknesses of the lms, surface roughness, materials used,
processing techniques, processing conditions, and the reec-
tive nature of the substrate used. Thus, it has been clearly seen
that incorporation of the dopants (Al and/or AlSi) lead to
microstructural modication (from SEM observations), grain
renements (XRD data), lattice contraction (XRD data), reaction
between the elements to generate additional phases (XPS data),
alternation in electronic bonding states (XPS results) and band
RSC Adv., 2016, 6, 36373–36383 | 36381

RSC Advances
structure of the lms (band-gaps data). These structural factors
signicantly inuence the scattering and the reection of light
and therefore the overall optical properties of the materials are,
in turn, expected to be modied. As a result, the overall spectral
selectivity of these lms is enhanced.
Published on 01 April 2016. Downloaded by MURDOCH UNIVERSITY LIBRARY on 18/04/2016 03:29:28.
Conclusions
Magnetron sputtered Ti0.5N0.5, Ti0.25Al0.25N0.5 and Ti0.25Al0.2-
Si0.05N0.5 lms were developed for solar selective applications.
XPS measurements are in good agreement with the XRD anal-
ysis, in which, both Si3N4 and AlN phases were detected. No
signature peaks for either SiO2, Al2O3 were detected in XRD
suggesting that they are in low concentrations or in amorphous
form. In TiAlSiN lms, sp3 Si bonding with sp2 N was noted by
XPS suggesting the formation of Si3N4 is consistent with the
XRD studies. The appearance of a small amount of oxide indi-
cated oxidation has occurred during the synthesis (as well as
subsequent annealing process) of thin lms. With Al and Al + Si
dopants, the direct energy band-gap reduces from 2.65 eV to
2.35 eV; hardness, Young's modulus, yield strength and plastic
deformation factor increase. Additionally, with the dopants, the
lms show an increased reectance in the infrared region and
a decrease in the UV-Vis region signicantly. Therefore, these
lms can be used as selective solar surfaces that signicantly
can improve the solar selectivity efficiency with increased
mechanical durability and stability.
Acknowledgements
M. Mahbubur Rahman acknowledges Murdoch University for
providing with the nancial support under the Murdoch
International Postgraduate Research Scholarship (MIPRS)
program. The authors also acknowledge funding support
provided by IRU-MRUN Collaborative Research Program and
the School of Engineering & Information Technology, Murdoch
University. C.Y. Yin is supported by the Teesside University
Research Fund.
References
1 Q. C. Zhang and D. R. Mills, Appl. Phys. Lett., 1992, 60, 545–
547.
2 Y. Yin and R. E. Collins, J. Appl. Phys., 1995, 77, 6485–6491.
3 H. C. Barshilia, N. Selvakumar, K. S. Rajam and A. Biswas,
Sol. Energy Mater. Sol. Cells, 2008, 92, 1425–1433.
4 H. C. Barshilia, N. Selvakumar, K. S. Rajam and A. Biswas,
Sol. Energy Mater. Sol. Cells, 2008, 92, 495–504.
5 H. C. Barshilia, N. Selvakumar, K. S. Rajam, D. V. Sridhara
Rao, K. Muraleedharan and A. Biswas, Appl. Phys. Lett.,
2006, 89, 1919–191911.
6 M. Du, L. Hao, J. Mi, F. Lv, X. Liu, L. Jiang and S. Wang, Sol.
Energy Mater. Sol. Cells, 2011, 95, 1193–1196.
7 A. Satta, G. Beyer, K. Maex, K. Elers, S. Haukka and
A. Vantomme, Mater. Res. Soc. Symp. Proc., 2000, 612,
D651–D656.
36382 | RSC Adv., 2016, 6, 36373–36383
View Article Online
Paper
8 N. Selvakumar and H. C. Barshilia, Sol. Energy Mater. Sol.
Cells, 2012, 98, 1–23.
9 N. Selvakumar, N. T. Manikandanath, A. Biswas and
H. C. Barshilia, Sol. Energy Mater. Sol. Cells, 2012, 102, 86–92.
10 C. G. Ribbing and A. Roos, in Transition metal nitride lms for
optical applications, ed. R. L. Hall, San Diego, CA, USA, 1997.
11 A. Schüler, V. Thommen, P. Reimann, P. Oelhafen,
G. Francz, T. Zehnder, M. Düggelin, D. Mathys and
R. Guggenheim, J. Vac. Sci. Technol., A, 2001, 19, 922–929.
12 L. Hao, S. Wang, L. Jiang, X. Liu, H. Li and Z. Li, Chin. Sci.
Bull., 2009, 54, 1451–1454.
13 H. C. Barshilia, N. Selvakumar, K. S. Rajam, D. V. Sridhara
Rao and K. Muraleedharan, Thin Solid Films, 2008, 516,
6071–6078.
14 N. Selvakumar, H. C. Barshilia, K. S. Rajam and A. Biswas,
Sol. Energy Mater. Sol. Cells, 2010, 94, 1412–1420.
15 A. Amri, X. Duan, C. Y. Yin, Z. T. Jiang, M. M. Rahman and
T. Pryor, Appl. Surf. Sci., 2013, 275, 127–135.
16 A. Amri, Z. T. Jiang, P. A. Bahri, C. Y. Yin, X. Zhao, Z. Xie,
X. Duan, H. Widjaja, M. M. Rahman and T. Pryor, J. Phys.
Chem. C, 2013, 117, 16457–16467.
17 A. Amri, Z. T. Jiang, N. Wyatt, C. Y. Yin, N. Mondinos,
T. Pryor and M. M. Rahman, Ceram. Interfaces, 2014, 40,
16569–16575.
18 A. Amri, Z. T. Jiang, X. Zhao, Z. Xie, C. Y. Yin, N. Ali,
N. Mondinos, M. M. Rahman and D. Habibi, Surf. Coat.
Technol., 2014, 239, 212–221.
19 A. Millar, M. M. Rahman and Z.-T. Jiang, J. Adv. Phys., 2014,
3, 179–193.
20 C. Nunes, V. Teixeira, M. L. Prates, N. P. Barradas and
A. D. Sequeira, Thin Solid Films, 2003, 442, 173–178.
21 M. M. Rahman, Z. T. Jiang, Z. Xie, X. Duan, Z. F. Zhou,
P. C. Wo, C. Y. Yin, N. Mondinos, Q. Gu, H. Widjaja,
K. Jack, A. Yago and A. Amri, J. Phys. Chem. C, 2014, 118,
18573–18579.
22 M. Mahbubur Rahman, A. Duan, Z.-T. Jiang, Z. Xie, A. Wu,
A. Amri, B. Cowie and C.-Y. Yin, J. Alloys Compd., 2013,
578, 362–368.
23 M. M. Rahman, Z.-T. Jiang, X. Duan, Z. Xie, A. Tadich,
Z.-F. Zhou, N. Mondinos, C.-Y. Yin, M. Altarawneh and
B. Z. Dlugogorski, J. Alloys Compd., 2016, 661, 268–273.
24 C. Dhand, N. Dwivedi, X. J. Loh, A. N. Jie Ying, N. K. Verma,
R. W. Beuerman, R. Lakshminarayanan and S. Ramakrishna,
RSC Adv., 2015, 5, 105003–105037.
25 R. Ghosh, K. S. Asha, S. M. Pratik, A. Datta, R. Nath and
S. Mandal, RSC Adv., 2014, 4, 21195–21200.
26 I. Caretti, M. Yuste, R. Torres, O. Sanchez, I. Jimenez and
R. Escobar Galindo, RSC Adv., 2012, 2, 2696–2699.
27 D. Porwal, A. C. M. Esther, I. N. Reddy, N. Sridhara,
N. P. Yadav, D. Rangappa, P. Bera, C. Anandan,
A. K. Sharma and A. Dey, RSC Adv., 2015, 5, 35737–35745.
28 S. Rtimi, O. Baghriche, C. Pulgarin, R. Sanjines and J. Kiwi,
RSC Adv., 2012, 2, 8591–8595.
29 L. Zhao, Y. Di, C. Yan, F. Liu, Z. Cheng, L. Jiang, X. Hao,
Y. Lai and J. Li, RSC Adv., 2016, 6, 4108–4115.
30 C. G. Granqvist, Sol. Energy Mater. Sol. Cells, 2007, 91, 1529–
1598.
This journal is © The Royal Society of Chemistry 2016

Paper
31 S. A. Kalogirou, Prog. Energy Combust. Sci., 2004, 30, 231–295.
32 C. E. Kennedy and H. Price, Proceedings of ISEC, 2005.
33 Y. Liu, C. Wang and Y. Xue, Sol. Energy Mater. Sol. Cells,
2012, 96, 131–136.
34 S. Yue, S. Yueyan and W. Fengchun, Sol. Energy Mater. Sol.
Cells, 2003, 77, 393–403.
35 J. P. A. M. Driessen, A. D. Kuypers and J. Schoonman, J. Vac.
Published on 01 April 2016. Downloaded by MURDOCH UNIVERSITY LIBRARY on 18/04/2016 03:29:28.
Sci. Technol., A, 2000, 18, 1971–1976.
36 M. Eizenberg, K. Littau, S. Ghanayem, M. Liao, R. Mosely
and A. K. Sinha, J. Vac. Sci. Technol., A, 1995, 13, 590–595.
37 G. G. Fuentes, E. Elizalde and J. M. Sanz, J. Appl. Phys., 2001,
90, 2737–2743.
38 D. Barbier, X. Orignac, X. M. Du, R. M. Almeida, X. Orignac,
D. Barbier, X. M. Du and R. M. Almeida, Proceedings of
Topical Symp. VII on Advanced Materials in Optics, Electro-
Optics and Communication Technologies, 1995.
39 H. Schroeder, Oxide layers deposited from organic solutions,
in Physics of thin lms: Advances in research and development,
ed. G. Hass and R. E. Thun, Academic Press, New York, 1969,
vol. 5.
40 X. Orignac, D. Barbier, X. M. Du and R. M. Almeida, Appl.
Phys. Lett., 1996, 69, 895–897.
41 R. M. Almeida, X. Orignac and D. Barbier, J. Sol-Gel Sci.
Technol., 1994, 2, 465–467.
42 Y. Sorek, R. Reisfeld, I. Finkelstein and S. Ruschin, Appl.
Phys. Lett., 1993, 63, 3256–3258.
43 M. M. Rahman, Z.-T. Jiang, Z.-F. Zhou, Z. Xie, C. Y. Yin,
H. Kabir, M. M. Haque, A. Amri, N. Mondinos and
M. Altarawneh, J. Alloys Compd., 2016, 671, 254–266.
44 J. T. Chen, J. Wang, F. Zhang, G. A. Zhang, X. Y. Fan, Z. G. Wu
and P. X. Yan, J. Alloys Compd., 2009, 472, 91–96.
45 D. G. Park, T. H. Cha, S. H. Lee, I. S. Yeo, J. W. Park and
S. D. Kim, J. Vac. Sci. Technol., B: Microelectron. Nanometer
Struct.–Process., Meas., Phenom., 2001, 19, 2289–2294.
This journal is © The Royal Society of Chemistry 2016
View publication stats
View Article Online
RSC Advances
46 Y. Wei and C. Gong, Appl. Surf. Sci., 2011, 257, 7881–7886.
47 J. G. Swadener, B. Taljat and G. M. Pharr, J. Mater. Res., 2001,
16, 2091–2102.
48 J. H. Kang and K. J. Kim, J. Appl. Phys., 1999, 86, 346–350.
49 M. Kotilainen, K. Mizohata, M. Honkanen and P. Vuoristo,
Sol. Energy Mater. Sol. Cells, 2014, 120, 462–472.
50 Z. T. Jiang, K. Ohshimu, M. Aoyama and T. Yamaguchi, Jpn.
J. Appl. Phys., 1998, 37, 4008–4013.
51 D. A. Shirley, Phys. Rev. B: Solid State, 1972, 5, 4709–4714.
52 I. Bertóti, M. Mohai, J. L. Sullivan and S. O. Saied, Appl. Surf.
Sci., 1995, 84, 357–371.
53 A. Vennemann, H. R. Stock, J. Kohlscheen, S. Rambadt and
G. Erkens, Surf. Coat. Technol., 2003, 174–175, 408–415.
54 L. Rosenberger, R. Baird, E. McCullen, G. Auner and
G. Shreve, Surf. Interface Anal., 2008, 40, 1254–1261.
55 J. A. Duffie and W. A. Beckman, Solar Engineering of Thermal
Processes, John Wiley & Sons Inc, New Jersey, USA, 3rd edn,
2006.
56 T. Boström, E. Wäckelgård and G. Westin, Sol. Energy, 2003,
74, 497–503.
57 T. Maruyama, Sol. Energy Mater. Sol. Cells, 1998, 56, 85–92.
58 V. Godinho, D. Philippon, T. C. Rojas, N. N. Novikova,
V. A. Yakovlev, E. A. Vinogradov and A. Fernandez, Sol.
Energy, 2010, 84, 1397–1401.
59 H. C. Barshilia, N. Selvakumar and K. S. Rajam, J. Vac. Sci.
Amp. Tech. A, 2007, 25, 383–390.
60 D. Gong, X. Cheng, W. Ye, P. Zhang and G. Luo, J. Wuhan
Univ. Technol., 2013, 28, 256–260.
61 H. Demiryont, L. R. Thompson and G. J. Collins, Appl. Opt.,
1986, 25, 1311–1318.
62 Z. W. Zhao, B. K. Tay, G. Q. Yu, D. H. C. Chua, S. P. Lau and
L. K. Cheah, Thin Solid Films, 2004, 447–448, 14–19.
RSC Adv., 2016, 6, 36373–36383 | 36383