View Article Online

View Journal

PCCP
Physical Chemistry Chemical Physics
Accepted Manuscript

This article can be cited before page numbers have been issued, to do this please use: C. Maheu, L.
Cardenas, E. Puzenat , P. Afanasiev and C. GEANTET, Phys. Chem. Chem. Phys., 2018, DOI:
10.1039/C8CP04614J.

Volume 18 Number 1 7 January 2016 Pages 1–636 This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
PCCP
Physical Chemistry Chemical Physics Accepted Manuscripts are published online shortly after
[Link]/pccp

acceptance, before technical editing, formatting and proof reading.

Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.

You can find more information about Accepted Manuscripts in the

author guidelines.

Please note that technical editing may introduce minor changes

to the text and/or graphics, which may alter content. The journal’s
ISSN 1463-9076 standard Terms & Conditions and the ethical guidelines, outlined
PERSPECTIVE
in our author and reviewer resource centre, still apply. In no
Darya Radziuk and Helmuth Möhwald
Ultrasonically treated liquid interfaces for progress in cleaning and
separation processes
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.

[Link]/pccp
 Page 1 of 10 Physical
PleaseChemistry Chemical
do not adjust Physics
margins
View Article Online
DOI: 10.1039/C8CP04614J

Journal Name

Physical Chemistry Chemical Physics Accepted Manuscript

ARTICLE

UPS and UV spectroscopies combined to position energy levels of

Published on 17 September 2018. Downloaded by Gazi Universitesi on 9/24/2018 [Link] AM.

TiO2 anatase and rutile nanopowders.

Received 00th January 20xx,
Accepted 00th January 20xx Clément Maheu,* Luis Cardenas,* Eric Puzenat, Pavel Afanasiev, Christophe Geantet
DOI: 10.1039/x0xx00000x An accurate experimental determination of electronic structure in semi-conductors nanopowders is a challenging task. We
[Link]/ propose here to combine UPS and UV-Vis spectroscopies in order to get the full description of electronic bands alignment
of powder samples, TiO2 rutile and anatase. For UPS measurements, two preparation methods, namely dropping method and
electrophoretic deposition, were used to prepare layers of titania powders on a conducting substrate, ITO or Ag. Both
methods lead to comparable results, with a quantitative description of the energy levels from the valence band. Combining
these results with the UV-Vis spectra of the same powders, enables to determine the absolute position of the valence band
maximum and of the conduction band minimum. Combined UPS–UV-vis spectroscopy provides a better insight into the
properties of a powdered material which can differ from single crystal model systems. It can be also used to predict the
electronic transfer in mixed phase systems during, in photocatalytic processes.

Introduction
During the last decades, extensive efforts were made to
determine precisely the energy levels and band structure of
titania (TiO2).1 Titanium oxide is used in many areas as
memresistors, optoelectronics, photovoltaics or
photocatalysis.2–4 Determination of electronic structures and Figure 1: Possible cases of relative positions of rutile and anatase energy levels.
carriers transport properties is crucial for these applications. Type I, included; Type II, staggered with energy levels of anatase above or reverse
Regarding photocatalysis, titania powders is the most widely case.
studied class of materials to the present day.5 In the 70s,
The relative positions of the energy levels determines the
Teichner discovered and conceptualized heterogeneous
direction of the electronic transfer of a photocatalytic material
photocatalysis with TiO2.6,7 Since, a tremendous amount of
based on a heterojunction between TiO2 and a metal11 or a
studies have been devoted to this oxide and its photocatalytic
semiconductor.12 A particular interest is also devoted to the
properties.7,8
materials composed of different titania phases. The most known
Among the many polymorphs of TiO2, rutile and anatase are
one is Degussa P25 titania which is a mixture of rutile and
the most studied.2 For the photocatalytic applications of these
anatase.13 The exact nature of its high photocatalytic activity
crystalline structures, determination of the position of the
still remains not completely understood, but the energy
energy levels is essential. Position of the valence band (VB), of
diagrams are obviously crucial to explain its properties.14 The
the conduction band (CB) and the size of the optical band gap
bands alignment between the two phases in Degussa P25 was
(Eg) are dependent of the crystalline phase. These parameters
claimed to enhance the charge careers separation.15 However,
influence the amount of photons absorbed upon illumination
there is a controversy in the literature concerning this band
and the respective oxidation and reduction potentials of holes
alignment. The band gaps of rutile and anatase TiO2
and electrons, generated by light. Presence of defects or
polymorphs are of 3.0 and 3.2 eV, respectively. However, their
dopants as well as the particle size may also play a key role and
relative position is still under debate. Three possibilities are
allow tuning the electronic properties of titania materials.10
generally discussed (Figure 1). The energy levels of both
phases can be included as the Eg,rutile < Eg,anatase (type I).15 It can
Université de Lyon, Institut de Recherches sur la Catalyse et l'Environnement de also be staggered alignment with the CB and the VB of anatase
Lyon - IRCELYON - UMR 5256, CNRS-UCB Lyon 1, 2 avenue Albert Einstein 69626 lying above those of rutile (type II-anatase) or with the CB and
Villeurbanne cedex, France.
† Electronic Supplementary Information (ESI) available: XRD pattern of TiO2
the VB of rutile lying above those of anatase (type II-rutile).16
powders. Methodology of UPS calibration, background subtraction and absolute As the bands are close, Mi et al. proposed two additional
UPS measurement. UPS spectra of conducting substrates and of the different TiO2 particular cases with the VB or the CB of the two phases being
layers (rutile and anatase with various preparation methods). UV-vis diffuse
reflectance spectra. See DOI: 10.1039/x0xx00000x aligned at the same value.17 However, there are no physical

This journal is © The Royal Society of Chemistry 20xx J. Name., 2013, 00, 1-3 | 1

Please do not adjust margins

 Physical
PleaseChemistry Chemical
do not adjust Physics
margins Page 2 of 10
View Article Online
DOI: 10.1039/C8CP04614J
ARTICLE Journal Name

methods accurate enough to prove the exact match of the materials, prepared under controlled atmosphere and the true
energy levels. electronic structure of powdered samples.
The bands positioning of type II-rutile or of type I should In this paper we introduce a new methodology to study
induce an electronic transfer from anatase to rutile.18–21 For a photocatalytic powders, in our case titania, with UPS and
type II-anatase, the charge transfer is in the opposite provide a combined UPS-UV-vis spectroscopic study of rutile

Physical Chemistry Chemical Physics Accepted Manuscript

direction.16,22–24 and anatase polymorphs in order to establish the absolute
The first paper to advocate a type II-rutile alignment came energy band positions.
from Mott-Schotky analysis by Kavan et al.18 This hypothesis
was supported by experimental data19,21 and calculations.20
However, recently it was shown that the Mott-Schotky analysis Experimental Section
Published on 17 September 2018. Downloaded by Gazi Universitesi on 9/24/2018 [Link] AM.

is strongly influenced by the surface charges.16,25 Adsorption of Materials

ions at the surface of TiO2 leads to appearance of a dipole layer
The two TiO2 polymorphs studied are commercial powders.
which interferes with band alignment analysis provided by this
Anatase was purchased from Cristal (CristalACTiVTM PC500)
technique.
and rutile was purchased from US Research
The hypothesis of the type II-anatase bands alignment was
Nanomaterials (99.9% - 30nm). The XRD patterns were
mainly supported by XPS measurements.23,24 The offset
measured on a Bruker D8 Advance A25 diffractometer
between the EVB of two phases was measured and then, the ECB
equipped with a Ni filter (Cu Kα radiation at: 0.154184 nm).
offset is deduced from the known Eg difference. Scanlon et al
Crystallite size was estimated from X-ray diffraction (XRD)
reported a difference of 0.4 eV23 and Pfeifer et al. of 0.6 –
lines broadening using Scherrer equation and found to be
0.7 eV.24 Gao and co-workers claimed similar band alignment
around 7 nm for anatase and 25 nm for rutile as detailed in the
using Kelvin probe force microscopy. This work has the
Supporting Information (SI), Figure S1.
particularity to evidence band alignment for a real
heterojunction, previously it has been shown only on separate
Ultraviolet Photoelectron Spectroscopy
TiO2 phases.26
Recent theoretical experiments support the type II-anatase UPS measures the kinetic energy of photoelectrons emitted
band alignment but with a higher band offset.27 Ko et al. using by solid surfaces upon irradiation with ultraviolet light, in order
the relativistic DFT calculation suggested that the type of to determine electronic structure of the valence band.37 UPS
alignment depends on the size of rutile and anatase particles.28 analyses were conducted in an integrated ultrahigh vacuum
It means that sizes should be taken into account as well as the system, connected to an A Axis Ultra DLD spectrometer
presence of surface states, impurities or structural defects. (Kratos Analytical). UPS spectra were obtained with a He lamp
Indeed, between two powders, even within the same phase, providing two resonance lines, He I (hν = 21.2 eV) and He II
these parameters can induce variation of the electronic (hν = 40.8 eV). To separate the secondary edges of sample and
structure. Therefore, clear and accurate experimental protocols analyzer, a negative potential about 9.1 V was applied to the
are needed to determine the absolute electronic structure of any sample. Prior to measurements, an Ag foil was cleaned by
TiO2-based powder materials (extendable to any powder argon ion sputtering. The bias was optimized to shift the Ag
materials). To this day, no such protocol is reported in the spectrum in the linear region of the analyzer (0~10 eV kinetic
literature. Two spectroscopic techniques were used, the UV-vis energy). A work function of 4.2 eV was obtained, in good
Diffuse Reflectance Spectroscopy (DRS) to study transitions agreement with the literature value around 4.26 eV (See SI).38
from the VB to the CB and Ultraviolet Photoelectron The spectrometer has pass energy of 10 eV and spatial aperture
Spectroscopy (UPS). of 110 µm. Under these conditions, the intrinsic resolution of
The latter is a surface sensitive tool, widely used to probe the spectra was measured at the Fermi edge of the Ag foil at
the electronic valence band.29,30 UPS provides information on room temperature (See Figure S2), as reported by Schalf et al.39
the distribution of electronic density in VB as well as on the It is estimated to be 0.15 eV. For comparison, the intrinsic
work function. UPS has been extensively applied to undoped resolution of a lab-base XPS set-up using a monochromatized
and doped model titania surfaces.2,31,32 The UPS results source is more than 0.5 eV.
obtained on the TiO2 (001) monocrystalline surface highlight At high binding energy (BE), the cut-off energy (Ecut-off) is
the formation of new structures with narrower band gap after determined using the first derivative of the UP spectrum, as
reducing or oxidative treatments33,34 or after nitrogen detailed in the SI. At lower BE a strong contribution of the
implantation.35 Formation of new electronic states located secondary electrons required a careful background subtraction.
above the O 2p band after N-doping has been evidenced with In the literature, different types of background are used for UPS
this method. As another example, UPS on VOx/TiO2 (001) spectra interpretation. Fleming et al. advised a “a + cxd”
showed the presence of intermediate states near the Fermi form40, Laidani and co-workers recommended a Shirley type
level.36 background41–43 and Payne et al used a Tougaard type
However, the most efficient TiO2 photocatalysts are not background.44 Those types of background are more suitable for
single crystals but nanopowders. Morphological and chemical higher photon energies. Li et al. proposed a background
differences may induce differences between the model dedicated to UPS analysis.45 It takes into account the energy
dependence of the cross-section for the creation of secondary

2 | J. Name., 2012, 00, 1-3 This journal is © The Royal Society of Chemistry 20xx

Please do not adjust margins

 Page 3 of 10 Physical
PleaseChemistry Chemical
do not adjust Physics
margins
View Article Online
DOI: 10.1039/C8CP04614J
Journal Name ARTICLE

electrons by primary electrons. It also considers the secondary- Then, a constant voltage of 14.9 V was applied during 1 min
electron cascade process, i.e. that secondary electrons can in with a stabilized power supply (type elc AL 936N).
turn generate other secondary electrons. The latter induces an
exponential shape which is not represented by standard core- UV-Vis absorption Spectroscopy
level background. Comparison of the Li background with a

Physical Chemistry Chemical Physics Accepted Manuscript

UV-Vis diffuse reflectance was measured at room
core-level type background, Figure S4, advocates the use of Li temperature with a Perkin Elmer Lambda 45 spectrometer
background. After background subtraction, a decomposition of equipped with an integrating sphere (RSA-PE-20). Pure BaSO4
the signal was done using Lorentzian Assymetric Finit (LF) was used as a reference and to dilute TiO2 powder. It has a total
function. reflection of the light in the wavelength range of interest 200-
To determine the valence band maximum (EVB), we 700 nm. If the sample is thick enough (here 4-5 mm), no light is
Published on 17 September 2018. Downloaded by Gazi Universitesi on 9/24/2018 [Link] AM.

proposed to use the second derivative of the spectrum acquired transmitted and the Kubelka-Munk (KM) function F KM can be
at low binding energy. EVB is the energy for which the first applied. The KM equation is valid for weakly absorbing
electron is photogenerated. Therefore, the EVB and the inflexion material, whereas for the values of reflectance R ∞ < 0.6,
point should match. It provides similar results as the method deviations from the theory occur. Therefore, highly diluted
proposed by Sayan et al.46 samples are required for eliminating the specular reflection and
minimizing the grain size effects.57 To prepare highly diluted
UPS sample Preparation samples, the standard procedure was to mix 30 mg of TiO2
Two different types of substrates and two sample powder with 1200 mg of BaSO4 (mass ratio of 1:40). Then 30
preparation methods were used. The first substrate is a mg of the latter mixture was again mixed with 1200 mg of
commercial conducting ITO layer on glass with RS=30-60Ω/sq, BaSO4 to get a 1:1600 ratio (625 ppm). This iterative procedure
purchased from Sigma. The second is an Ag foil purchased was repeated to get samples with a mass weight of 0.39 ppm.
from Alfa Aesar, previously cleaned with Ar ion sputtering. Such a procedure allows obtaining high quality UV-vis spectra.
The UP spectra for both substrates were acquired as references Additionally, dilution of the sample permits to approximate
(Figure S2 and S3). The Ecut-off measured at 17.0 eV for the ITO the scattering coefficient by that of BaSO4. The absorption
agrees with the one reported by Beerbom et al.47 It confirms the coefficient is then easily obtained from reflectance
accurate calibration of the analytic system. ITO is a suitable measurement (See SI). The latter is used to obtain the band gap
system for UPS measurements since there is no major of both anatase and rutile using Tauc’s relationship58 as
electronic contribution in the range of 10-0 eV which is our recommended by López et al.59. A further decomposition of a
area of interest. On the contrary, the d bands of Ag are located UV spectrum (FKM(hν)) can be obtained after normalization
in this energy range, but Ag foil permits an absolute according to the equation:
measurement of EVB. ௛ఔାௗ௛ఔ
The samples for UPS measurements must be thin in order to ‫׬‬௛ఔିௗ௛ఔ ‫ܨ‬௄ெ ሺℎߥሻ݀ℎߥ
‫ݐ‬ሺℎߥሻ = ஶ
minimize the charging effects. For the first preparation method, ‫׬‬଴ ‫ܨ‬௄ெ ሺℎߥሻ݀ℎߥ
the powders were ultrasonically dispersed in heptane or in
pentane (4 g/L). Then, several drops were deposited on the This new t(hν) function can be decomposed with several
conducting ITO substrate. This preparation method was Gaussian contributions. Details of the fitting parameters are
proposed for zeolites48 and further applied to other oxides such reported in the SI.
as V2O5 or bismuth molybdate.49 For the second preparation
method, a homogeneous thin film of TiO2 powder was prepared
by electrophoretic deposition (EPD). This technique was first Results and Discussion
used to prepare samples of dye-sensitized solar cells.50 The Valence Band Maximum (EVB): UPS
particles of a stable suspension were deposited applying a DC
While UPS allows getting a higher resolution and better
electric field.51 EPD duration, electrode distances, and
sensitivity than XPS, the sample preparation in case of non-
composition of the solution can influence the thickness and the
conductive powders is more demanding as concerns the layer
morphology of the film.52 Because of the key role of the
homogeneity and its electric contact with the conducting
charging solution, several solvents were tested.53–55 One of the
support. Two different preparation methods of samples for UPS
most efficient systems is a mixture of iodine, acetone and
measurements are compared in the present study. In the first
water.56 Therefore, EPD technique enables the preparation of a
approach, a suspension of TiO2-anatase in alkanes was dropped
homogeneous, pure and well-controlled TiO2 layers on a
on ITO substrate. Pentane and heptane suspensions were
conducting substrate.
studied. As UPS is sensitive to impurities, the first has the
To prepare the samples by EPD, 50 mg of iodine was
advantage to be easily removed. In the second approach TiO2-
dissolved in 100 ml acetone upon stirring. Then, a suspension
anatase was deposited by electrophoresis. All the UPS spectra
of titania powder (2-3 g/L concentration) was added, the
are reported in the SI. They have similar shape at low BE, as
suspension was sonicated for 10 min and leaved for
the occupied bonding and non-bonding states due to TiO2 are
approximately 1 hr. It was poured in a glass vessel containing
observed.2,57 At higher BE, the secondary edge is observed and
two parallel ITO/glass substrate bars, separated by ca. 1 cm.
corresponds to the Ecut-off.

This journal is © The Royal Society of Chemistry 20xx J. Name., 2013, 00, 1-3 | 3

Please do not adjust margins

 Physical
PleaseChemistry Chemical
do not adjust Physics
margins Page 4 of 10
View Article Online
DOI: 10.1039/C8CP04614J
ARTICLE Journal Name

For the spectra shown in the SI (Figures S5 and S6), anatase

EVB values are similar. EVB varies from 2.9 to 3.1 eV; the Ecut-off
differs from one preparation type to another. The presence of an
interfacial dipole between ITO substrate and the TiO2 powder
may induce a global shift of the spectrum. This dipole is

Physical Chemistry Chemical Physics Accepted Manuscript

influenced by the preparation method. A method to confirm the
absolute position of Ecut-off and EVB will be discussed below.
UPS spectra obtained for anatase and rutile are shown in
Figure 2. The Ecut-off is measured for both polymorphs at 17.2
eV but a shift is observed at low energy. The measured EVB
Published on 17 September 2018. Downloaded by Gazi Universitesi on 9/24/2018 [Link] AM.

values are 2.9 eV for anatase and 2.5 eV for rutile. This
difference of 0.4 ± 0.2 eV agrees with the previous result Figure 3: He I UP background subtracted spectra of the anatase/ITO (drop
reported by Scanlon et al. and Pfeifer et al.23,24 However, it is method) (A) and of the rutile/ITO (EPD method) (B) and their decomposition into
the first time that this difference is measured on powdered four asymmetric contributions.
titania with UPS method. Note that a same band offset between band is composed of energy levels, due only to the non-bonding
the two polymorphs was obtained with the method determining (π) O 2p orbitals (t1g and t2u levels). Then, the hybridization
EVB as the intersection point between the zero-intensity between the Ti orbitals with the O 2p orbitals forms, depending
constant backgrounds and a linear extrapolation of the valence of their respective orientation, non-bonding orbitals (3t1u and
band decay (Table S1).61,62 (1t2g) and deeper in energy, bonding orbitals (σ) (2t1 u, 2eg and
2a1g).64
Experimental DOS feature More recently, Van de Krol for the rutile65 and Asahi et al.
To have a better understanding of the valence band from an for the anatase66 used the MO theory to build an energy band
experimental point of view, the low BE range was studied in diagram. The latter decomposed the VB in three main regions:
more detail. In both spectra baseline was corrected (see SI for the first one attributed to the O 2p π orbitals whereas π-
background treatment) using the background function hybridized and π-hybridized orbitals contribute to the second
developed by Li et al.45 As proposed by Fleming et al.40, from and the third regions. It was also correlated with DOS
empirical reasons, the resulting spectra can be fitted with four calculations. Indeed, UPS probes the DOS below the Fermi
components (Figure 3). Fleming’s interpretation was deduced level and a better understanding can be obtained from the
from UPS analysis on TiO2/Si (100) grown by molecular beam partial density of states (PDOS) description reported for
deposition.40 anatase66,67 rutile55, both23,24 or specific planes of titania.70
The state observed at BE = 11 eV (IV) is assigned to Recently, DFT calculations on TiO2 nanospheres corroborated
surface OH groups.60 Above this first contribution, the three states decomposition of UPS spectra.71
decomposition to the remaining three states can be qualitatively In the range between 0 and -6 eV, below the valence band
explained on the basis of localized molecular orbital (MO) maximum, DOS calculations predict three bands with a major
models63, The local electronic structure was associated with the contribution of O 2p orbitals. PDOS calculations clearly
MO-like state contributions composed from the Ti 3d and O 2p evidence that the contribution located at -1 eV below the EVB
orbitals. (sate I) is due to O 2p π orbitals. Below this feature, the PDOS
For titania, molecular energy-level diagram of the valence of Ti 3d increases with the decrease of energy (bonding states).
The state I is consequently due to a contribution from the
hybridized non-bonding orbitals and the state II is due to the
hybridized bonding orbitals.

Band Gap (Eg): UV-vis Diffuse Refelectance Spectroscopy

Overall, UPS accurately describes the occupied levels
composing the valence band. However, the positions of empty
states are not probed by UPS. The band diagram can be
completed to reach the level of the conduction band by using
the data of UV-Vis DRS.
UV-Vis spectra are reported for the two phases in
Figure S8. They show that, for a similar dilution in BaSO4, the
maximum of absorption is 2.5 higher for anatase than for rutile,
due to the difference between the absorption coefficients of the
two phases. It also demonstrates that using highly diluted
Figure 2: He I UP spectra of anatase/ITO (black) and rutile/ITO (red); both
samples is crucial to resolve fine transitions. Individual peaks
prepared by EPD method. Inset – zoom on the low BE area for anatase (black) and could be resolved in the UV-vis spectra for the 0.39 ppm
rutile (red). Black and red dotted lines correspond to the second derivative of the
corresponding spectra.

4 | J. Name., 2012, 00, 1-3 This journal is © The Royal Society of Chemistry 20xx

Please do not adjust margins

 Page 5 of 10 Physical
PleaseChemistry Chemical
do not adjust Physics
margins
View Article Online
DOI: 10.1039/C8CP04614J
Journal Name ARTICLE

Physical Chemistry Chemical Physics Accepted Manuscript

Published on 17 September 2018. Downloaded by Gazi Universitesi on 9/24/2018 [Link] AM.

Figure 4: t(hν) versus energy spectra of TiO2-anatase (A) and TiO2-rutile (B) diluted with BaSO4 (dilution of 0.39 ppm) (black dots). Violet curve is the fitting result and
bell shape curves are the different contributions (α, β, γ, δ, ε).

diluted sample whereas for a more concentrated sample and using the two Eg values determined by UV-vis absorption,
(16 ppm) this feature is hidden by peak broadening (Figure S8). the ECB levels were deduced. The absolute electronic structure
The decomposition of t(hν) spectrum is illustrated by for both rutile and anatase is shown on Figure 5. To the best of
Figure Figure 4. Four contributions are observed for the anatase our knowledge, for the first time, the band positions and shape
sample and five for the rutile. For the anatase sample, an of the VB of two powder photocatalysts were determined with
additional small sharp peak, probably related to excitons, at such accuracy. A band offset of 0.4 eV for the valence band and
3.25 eV is observed (insert of Figure 4A, obtained with a of 0.25 eV for the conduction band was determined between the
dilution of 9.8 ppb). two phases.
The four contributions around 4.2 eV (α), 4.5 eV (β), 5.0 eV The reasons of this offset between the two polymorphs are
(γ) and 5.5 eV (δ) are common to the two phases. A peak at due to both a shift of the different contributions and also a
lower energy is observed at 3.7 eV for the rutile (ε). It may be broadening of the O 2p state. Indeed there is a 0.2 eV shift
due to a state near the EVB of the rutile and nonexistent in the between the VB states of the rutile and of the anatase (See
anatase. For the two polymorphs the optical band gap was Table S4) In addition, the full width at half maximum for the
measured, Eg,anatase = 3.15 eV and Eg,rutile = 3.00 eV (See Figure contribution I is of 1.3 eV for anatase and 1.6 eV for rutile. To
S9). The error is estimated at 0.05 eV. achieve the MO-like description of UPS spectra, it was argued
that close to the VB and CB edges, the local structure
Absolute Electronic structure of anatase and rutile dominates over the effects of periodicity in determining the
The relative positions of VB determined from UPS and the optical properties of TiO2.40 Such analysis allows a qualitative
Eg value from UV-visible spectra give the overall picture of attribution of the corresponding components in the UP spectra
bands positions corresponding to the staggered alignment in of the valence band. However, it cannot be applied to explain
which anatase bands are lower than rutile ones (type II- the VB offset. Indeed, the local environment of Ti atoms in
anatase), in agreement with the previous findings.16,22,23 It must anatase and rutile is very similar and cannot account for the
be noted that their EVB was determined separately, so a band 0.4 eV differences of the bands positions. Obviously, structure
alignment between the two contacting phases may shift those and symmetry related differences in the VB and CB bands
values. shape explain this energy offset. Due to such symmetry
To properly describe the electronic structure of both rutile differences, rutile is a direct semiconductor with a transition at
and anatase, an absolute measurement of their energy levels the G-point, whereas anatase is an indirect one.72 The
was done. Dropping suspension of TiO2 in alkane on a silver distribution of projected DOS in anatase and rutile VB is
substrate allowed measuring a UPS spectrum with both similar. However, a simple qualitative interpretation of their
contributions of Ag and of TiO2 (Figure S7). Thus, on the same band structure difference can be given. The valence bands
spectrum we observed the Fermi edge of Ag foil and some primarily consist of O 2p states and a small Ti 3d contribution,
contributions of its d-states as well as characteristic peaks indicating the presence of some p–d hybridization (which
belonging to TiO2. corresponds to the component I in Figure 3). Hence, the
The VB maximum for the anatase phase is measured 2.9 eV difference between anatase and rutile is related to the O 2p –
below the Fermi level of the silver. Additionally, work function like states at the VB edge. Among two phases, rutile is a stable
of this Ag foil was measured at 4.2 eV. As a result, EVB for polymorph with a higher density and therefore a higher overall
TiO2 is -7.1 eV below the vacuum level. Using the offset interaction between the Ti and O valent electronic manifolds.
between the EVB of the two polymorphs, determined by UPS Due to that, the topmost VB band related to the O 2p states is

This journal is © The Royal Society of Chemistry 20xx J. Name., 2013, 00, 1-3 | 5

Please do not adjust margins

 Physical
PleaseChemistry Chemical
do not adjust Physics
margins Page 6 of 10
View Article Online
DOI: 10.1039/C8CP04614J
ARTICLE Journal Name

Physical Chemistry Chemical Physics Accepted Manuscript

Published on 17 September 2018. Downloaded by Gazi Universitesi on 9/24/2018 [Link] AM.

Figure 5: Electronic structure of the two TiO2 anatase and rutile. Ethr corresponds to the deepest state that can be excited in a photocatalytic test (Ephotons ≤ 4.3 eV).

broader in rutile which explains both the increase of VB edge

energy and the decrease of Eg as compared with anatase Conclusions
counterpart. The bands in the CB are narrower since they are Powders of the two main polymorphs of TiO2, anatase and
related to the antibonding states. The difference of CB rutile, were studied by UPS and UV-vis spectroscopies. Up to
properties is therefore lesser and the values for two polymorphs now, UPS on powders was only qualitatively interpreted. It can
are close. provide a quantitative description of the energy levels
We can now reconsider the UV Vis electronic transitions composing the valence band. Combination of UV-vis and UPS
observed on Figure 4 and assign them in terms of a simplified allowed positioning the energetic levels and discuss the
MO approach. Obviously α and β transitions (as well as e for electronic structures of the two solids.
rutile) are issued from non bonding O 2p orbitals. Above, γ and In the case of anatase and rutile powders the absolute EVB
δ transitions are energetic enough to photoexcite either non positions were measured at -7.1 eV and -6.7 eV, respectively.
bonding O 2p or hybridized orbitals. The resulting electronic structure obtained one at a time for
anatase and rutile supports the hypothesis of a type II-anatase
In a typical standard photocatalytic test, pyrex glassware energy bands alignment. In a mixture of both polymorphs the
filters the incoming photons below 290 nm (Ephotons = 4.3 eV).73 transfer of the electrons from rutile to anatase particles should
This allows to define an energy threshold (Ethr.) corresponding be favoured. Considering photocatatytic experiments, efficient
to the deepest energy level above which electrons can be photoelectrons for this process are excited from the top of VB;
photoexcited. . This value is located at -8.25 eV for anatase and attributed to a narrow contribution originating from the non-
-8.0 eV for rutile. Our work demonstrates that under UV bonding O 2p orbitals.
illumination, electrons promoted from VB to CB come mainly A full description of the electronic structure of the powder
from the non-bonding O 2p orbitals (contribution I) and samples would be obtained using complementary techniques
contributed to photocatalysis. In order to optimize the such as inverse photoemission spectroscopy or
photocatalytic efficiency, only those states must be targeted. spectroelectrochemical analysis to describe the CB states.75,76
The type II-anatase alignment supported by our results
suggests that photogenerated holes in anatase should have a
higher oxidative potential than those in rutile and Conflicts of interest
photogenerated electrons in rutile should have a higher There are no conflicts to declare.
reduction potential than those in anatase (Figure 1). The higher
difference of oxidative potential counterbalance the difference
of reduction potential and reinforces the conclusion that anatase Acknowledgements
is considered as the most active TiO2 polymorph.74
C. M. acknowledges financial support from the French
Ministry of Education.

6 | J. Name., 2012, 00, 1-3 This journal is © The Royal Society of Chemistry 20xx

Please do not adjust margins

 Page 7 of 10 Physical
PleaseChemistry Chemical
do not adjust Physics
margins
View Article Online
DOI: 10.1039/C8CP04614J
Journal Name ARTICLE

References 16 Y. Nosaka and A. Y. Nosaka, Reconsideration of Intrinsic

Band Alignments within Anatase and Rutile TiO2, The
1 M. Kapilashrami, Y. Zhang, Y.-S. Liu, A. Hagfeldt and J. Journal of Physical Chemistry Letters, 2016, 7, 431–434.
Guo, Probing the Optical Property and Electronic Structure of 17 Y. Mi and Y. Weng, Band Alignment and Controllable
TiO2 Nanomaterials for Renewable Energy Applications, Electron Migration between Rutile and Anatase TiO2,
Chemical Reviews, 2014, 114, 9662–9707. Scientific Reports, 2015, 5, 11482.

Physical Chemistry Chemical Physics Accepted Manuscript

2 U. Diebold, The surface science of titanium dioxide, Surface 18 L. Kavan, M. Grätzel, S. E. Gilbert, C. Klemenz and H. J.
Science Reports, 2003, 48, 53–229. Scheel, Electrochemical and Photoelectrochemical
3 K. Nakata and A. Fujishima, TiO2 photocatalysis: Design and Investigation of Single-Crystal Anatase, Journal of the
applications, Journal of Photochemistry and Photobiology C: American Chemical Society, 1996, 118, 6716–6723.
Photochemistry Reviews, 2012, 13, 169–189. 19 T. Miyagi, M. Kamei, T. Mitsuhashi, T. Ishigaki and A.
4 S. Almosni, A. Delamarre, Z. Jehl, D. Suchet, L. Cojocaru, Yamazaki, Charge separation at the rutile/anatase interface: a
Published on 17 September 2018. Downloaded by Gazi Universitesi on 9/24/2018 [Link] AM.

M. Giteau, B. Behaghel, A. Julian, C. Ibrahim, L. Tatry, H. dominant factor of photocatalytic activity, Chemical Physics
Wang, T. Kubo, S. Uchida, H. Segawa, N. Miyashita, R. Letters, 2004, 390, 399–402.
Tamaki, Y. Shoji, K. Yoshida, N. Ahsan, K. Watanabe, T. 20 J. Kang, F. Wu, S.-S. Li, J.-B. Xia and J. Li, Calculating
Inoue, M. Sugiyama, Y. Nakano, T. Hamamura, T. Toupance, Band Alignment between Materials with Different Structures:
C. Olivier, S. Chambon, L. Vignau, C. Geffroy, E. Cloutet, G. The Case of Anatase and Rutile Titanium Dioxide, The
Hadziioannou, N. Cavassilas, P. Rale, A. Cattoni, S. Collin, Journal of Physical Chemistry C, 2012, 116, 20765–20768.
F. Gibelli, M. Paire, L. Lombez, D. Aureau, M. Bouttemy, A. 21 S. Shen, X. Wang, T. Chen, Z. Feng and C. Li, Transfer of
Etcheberry, Y. Okada and J.-F. Guillemoles, Material Photoinduced Electrons in Anatase–Rutile TiO2 Determined
challenges for solar cells in the twenty-first century: by Time-Resolved Mid-Infrared Spectroscopy, The Journal
directions in emerging technologies, Science and Technology of Physical Chemistry C, 2014, 118, 12661–12668.
of Advanced Materials, 2018, 19, 336–369. 22 P. Deák, B. Aradi and T. Frauenheim, Band Lineup and
5 J. Schneider, M. Matsuoka, M. Takeuchi, J. Zhang, Y. Charge Carrier Separation in Mixed Rutile-Anatase Systems,
Horiuchi, M. Anpo and D. W. Bahnemann, Understanding The Journal of Physical Chemistry C, 2011, 115, 3443–3446.
TiO2 Photocatalysis: Mechanisms and Materials, Chem. Rev., 23 D. O. Scanlon, C. W. Dunnill, J. Buckeridge, S. A. Shevlin,
2014, 114, 9919–9986. A. J. Logsdail, S. M. Woodley, C. R. A. Catlow, M. J.
6 M. Formenti, F. Juillet and S. J. Teichner, Photooxydation Powell, R. G. Palgrave, I. P. Parkin, G. W. Watson, T. W.
ménagée des parafines et oléfines sur l’anatase à température Keal, P. Sherwood, A. Walsh and A. A. Sokol, Band
ambiante, C. R. Acad. Sc. Paris, 1970, 270, 138–141. alignment of rutile and anatase TiO2, Nature Materials, 2013,
7 N. Serpone, A. V. Emeline, S. Horikoshi, V. N. Kuznetsov 12, 798–801.
and V. K. Ryabchuk, On the genesis of heterogeneous 24 V. Pfeifer, P. Erhart, S. Li, K. Rachut, J. Morasch, J. Brötz, P.
photocatalysis: a brief historical perspective in the period Reckers, T. Mayer, S. Rühle, A. Zaban, I. Mora Seró, J.
1910 to the mid-1980s, Photochemical & Photobiological Bisquert, W. Jaegermann and A. Klein, Energy Band
Sciences, 2012, 11, 1121. Alignment between Anatase and Rutile TiO 2, The Journal of
8 M. Pelaez, N. T. Nolan, S. C. Pillai, M. K. Seery, P. Falaras, Physical Chemistry Letters, 2013, 4, 4182–4187.
A. G. Kontos, P. S. M. Dunlop, J. W. J. Hamilton, J. A. 25 J. Kullgren, B. Aradi, T. Frauenheim, L. Kavan and P. Deák,
Byrne, K. O’Shea, M. H. Entezari and D. D. Dionysiou, A Resolving the Controversy about the Band Alignment
review on the visible light active titanium dioxide between Rutile and Anatase: The Role of OH–/H+ Adsorption,
photocatalysts for environmental applications, Applied The Journal of Physical Chemistry C, 2015, 119, 21952–
Catalysis B: Environmental, 2012, 125, 331–349. 21958.
9 Y. Ma, X. Wang, Y. Jia, X. Chen, H. Han and C. Li, Titanium 26 Y. Gao, J. Zhu, H. An, P. Yan, B. Huang, R. Chen, F. Fan
Dioxide-Based Nanomaterials for Photocatalytic Fuel and C. Li, Directly Probing Charge Separation at Interface of
Generations, Chemical Reviews, 2014, 114, 9987–10043. TiO 2 Phase Junction, The Journal of Physical Chemistry
10 Z. Zhang, C.-C. Wang, R. Zakaria and J. Y. Ying, Role of Letters, 2017, 8, 1419–1423.
particle size in nanocrystalline TiO2-based photocatalysts, 27 Y.-Y. Zhang, L. Lang, H.-J. Gu, S. Chen, Z.-P. Liu, H. Xiang
The Journal of Physical Chemistry B, 1998, 102, 10871– and X.-G. Gong, Origin of the type-II band offset between
10878. rutile and anatase titanium dioxide: Classical and quantum-
11 J. M. Herrmann, From catalysis by metals to bifunctional mechanical interactions between O ions, Physical Review B, ,
photocatalysis, Topics in Catalysis, 2006, 39, 3–10. DOI:10.1103/PhysRevB.95.155308.
12 M. T. Uddin, Y. Nicolas, C. Olivier, W. Jaegermann, N. 28 K. C. Ko, S. T. Bromley, J. Y. Lee and F. Illas, Size-
Rockstroh, H. Junge and T. Toupance, Band alignment Dependent Level Alignment between Rutile and Anatase
investigations of heterostructure NiO/TiO2 nanomaterials TiO2 Nanoparticles: Implications for Photocatalysis, The
used as efficient heterojunction earth-abundant metal oxide Journal of Physical Chemistry Letters, 2017, 8, 5593–5598.
photocatalysts for hydrogen production, Physical Chemistry 29 F. Reinert and S. Hüfner, Photoemission spectroscopy-from
Chemical Physics, 2017, 19, 19279–19288. early days to recent applications, New Journal of Physics,
13 A. Fujishima, X. Zhang and D. Tryk, TiO2 photocatalysis and 2005, 7, 1–34.
related surface phenomena, Surface Science Reports, 2008, 30 S. Hüfner, S. Schmidt and F. Reinert, Photoelectron
63, 515–582. spectroscopy - An overview, Nuclear Instruments and
14 Z. Zhang and J. T. Yates, Band Bending in Semiconductors: Methods in Physics Research Section A: Accelerators,
Chemical and Physical Consequences at Surfaces and Spectrometers, Detectors and Associated Equipment, 2005,
Interfaces, Chemical Reviews, 2012, 112, 5520–5551. 547, 8–23.
15 D. C. Hurum, A. G. Agrios, K. A. Gray, T. Rajh and M. C. 31 G. Liu, W. Jaegermann, J. He, V. Sundström and L. Sun, XPS
Thurnauer, Explaining the Enhanced Photocatalytic Activity and UPS Characterization of the TiO2/ZnPcGly
of Degussa P25 Mixed-Phase TiO 2 Using EPR, The Journal Heterointerface: Alignment of Energy Levels, The Journal of
of Physical Chemistry B, 2003, 107, 4545–4549. Physical Chemistry B, 2002, 106, 5814–5819.

This journal is © The Royal Society of Chemistry 20xx J. Name., 2013, 00, 1-3 | 7

Please do not adjust margins

 Physical
PleaseChemistry Chemical
do not adjust Physics
margins Page 8 of 10
View Article Online
DOI: 10.1039/C8CP04614J
ARTICLE Journal Name

32 V. E. Henrich and R. L. Kurtz, Surface electronic structure of 49 M. Heber and W. Grünert, Application of ultraviolet
TiO2: Atomic geometry, ligand coordination, and the effect of photoelectron spectroscopy (UPS) in the surface
adsorbed hydrogen, Physical Review B, 23, 6280. characterisation of polycrystalline oxide catalysts. I.
33 A. Borodin and M. Reichling, Characterizing TiO2 (110) Methodics and analytical potential, Topics in Catalysis, 2001,
surface states by their work function, Physical Chemistry 15, 3–11.
Chemical Physics, 2011, 13, 15442. 50 D. Matthews, A. Kay and M. Gratzel, Electrophoretically

Physical Chemistry Chemical Physics Accepted Manuscript

34 J. Tao, T. Luttrell and M. Batzill, A two-dimensional phase of Deposited Titanium Dioxide Thin Films for Photovoltaic
TiO2 with a reduced bandgap, Nature Chemistry, 2011, 3, Cells, Australian Journal of Chemistry, 1994, 47, 1869.
296–300. 51 L. Grinis, S. Dor, A. Ofir and A. Zaban, Electrophoretic
35 M. Batzill, E. H. Morales and U. Diebold, Influence of deposition and compression of titania nanoparticle films for
Nitrogen Doping on the Defect Formation and Surface dye-sensitized solar cells, Journal of Photochemistry and
Properties of TiO2 Rutile and Anatase, Physical Review Photobiology A: Chemistry, 2008, 198, 52–59.
Published on 17 September 2018. Downloaded by Gazi Universitesi on 9/24/2018 [Link] AM.

Letters, 2006, 96, 026103. 52 V. Jouenne, J.-L. Duvail, L. Brohan, E. Gautron and M.
36 G. Silversmit, H. Poelman, D. Depla, D. Poelman, N. Barrett, Richard-Plouet, Low-temperature synthesis and
G. B. Marin and R. D. Gryse, UPS study of the thermal electrophoretic deposition of shape-controlled titanium
reduction of fully oxidized V2O5/TiO2 (001)-anatase model dioxide nanocrystals, RSC Adv., 2015, 5, 15118–15125.
catalysts, Journal of Electron Spectroscopy and Related 53 J.-H. Yum, S.-S. Kim, D.-Y. Kim and Y.-E. Sung,
Phenomena, 2005, 144–147, 377–380. Electrophoretically deposited TiO2 photo-electrodes for use in
37 R. H. Williams, G. P. Srivastava and I. T. McGovern, flexible dye-sensitized solar cells, Journal of Photochemistry
Photoelectron spectroscopy of solids and their surfaces, and Photobiology A: Chemistry, 2005, 173, 1–6.
Reports on Progress in Physics, 1980, 43, 1357–1414. 54 G.-S. Kim, H.-K. Seo, V. P. Godble, Y.-S. Kim, O.-B. Yang
38 A. W. Dweydari and C. H. B. Mee, Work function and H.-S. Shin, Electrophoretic deposition of titanate
measurements on (100) and (110) surfaces of silver, physica nanotubes from commercial titania nanoparticles: Application
status solidi (a), 1975, 27, 223–230. to dye-sensitized solar cells, Electrochemistry
39 R. Schlaf, P. G. Schroeder, M. W. Nelson, B. A. Parkinson, Communications, 2006, 8, 961–966.
C. D. Meritt, L. A. Crisafulli, H. Murata and Z. H. Kakafi, 55 S. Dor, S. Rühle, A. Ofir, M. Adler, L. Grinis and A. Zaban,
Determination of interface dipole and band bending at the The influence of suspension composition and deposition
Ag/tris (8-hydroxyquinolinato) gallium organic Schottky mode on the electrophoretic deposition of TiO2 nanoparticle
contact by ultraviolet photoemission spectroscopy, Surface agglomerates, Colloids and Surfaces A: Physicochemical and
Science, 2000, 450, 142–152. Engineering Aspects, 2009, 342, 70–75.
40 L. Fleming, C. C. Fulton, G. Lucovsky, J. E. Rowe, M. D. 56 N. Koura, T. Tsukamoto, HiromasaShoji and T. Hotta,
Ulrich and J. Lüning, Local bonding analysis of the valence Preparation of Various Oxide Films by an Electrophoretic
and conduction band features of TiO2, Journal of Applied Deposition Method: A Study of the Mechanism, Japanese
Physics, 2007, 102, 033707. Journal of Applied Physics, 1995, 34, 1643–1647.
41 N. Laidani, P. Cheyssac, J. Perrière, R. Bartali, G. Gottardi, I. 57 G. Kortüm, W. Braun and G. Herzog, Principles and
Luciu and V. Micheli, Intrinsic defects and their influence on Techniques of Diffuse-Reflectance Spectroscopy,
the chemical and optical properties of TiO2−x films, Journal Angewandte Chemie International Edition, 1963, 2, 333–341.
of Physics D: Applied Physics, 2010, 43, 485402. 58 J. Tauc, R. Grigorovici and A. Vancu, Optical properties and
42 I. Luciu, R. Bartali and N. Laidani, Influence of hydrogen electronic structure of amorphous germanium, physica status
addition to an Ar plasma on the structural properties of TiO2−x solidi (b), 1966, 15, 627–637.
thin films deposited by RF sputtering, Journal of Physics D: 59 R. López and R. Gómez, Band-gap energy estimation from
Applied Physics, 2012, 45, 345302. diffuse reflectance measurements on sol–gel and commercial
43 P. Pu, H. Cachet, N. Laidani and E. M. M. Sutter, Influence TiO2: a comparative study, Journal of Sol-Gel Science and
of pH on Surface States Behavior in TiO2 Nanotubes, The Technology, 2012, 61, 1–7.
Journal of Physical Chemistry C, 2012, 116, 22139–22148. 60 R. Sanjinés, H. Tang, H. Berger, F. Gozzo, G. Margaritondo
44 D. T. Payne, Y. Zhang, C. L. Pang, H. H. Fielding and G. and F. Lévy, Electronic structure of anatase TiO 2 oxide,
Thornton, Coverage-dependent two-photon photoexcitation at Journal of Applied Physics, 1994, 75, 2945–2951.
the H2O/TiO2 interface, Surface Science, 2016, 652, 189–194. 61 S. A. Chambers, T. Droubay, T. C. Kaspar and M. Gutowski,
45 X. Li, Z. Zhang and V. E. Henrich, Inelastic electron Experimental determination of valence band maxima for
background function for ultraviolet photoelectron spectra, SrTiO3, TiO2, and SrO and the associated valence band
Journal of electron spectroscopy and related phenomena, offsets with Si (001), Journal of Vacuum Science &
1993, 63, 253–265. Technology B: Microelectronics and Nanometer Structures,
46 S. Sayan, E. Garfunkel and S. Suzer, Soft x-ray 2004, 22, 2205.
photoemission studies of the HfO2/SiO2/Si system, Applied 62 A. Giampietri, G. Drera and L. Sangaletti, Band Alignment at
Physics Letters, 2002, 80, 2135–2137. Heteroepitaxial Perovskite Oxide Interfaces. Experiments,
47 M. M. Beerbom, B. Lägel, A. J. Cascio, B. V. Doran and R. Methods, and Perspectives, Advanced Materials Interfaces,
Schlaf, Direct comparison of photoemission spectroscopy and 2017, 4, 1700144.
in situ Kelvin probe work function measurements on indium 63 J. B. Goodenough, Metallic oxides, Progress in Solid State
tin oxide films, Journal of Electron Spectroscopy and Related Chemistry, 1971, 5, 145–399.
Phenomena, 2006, 152, 12–17. 64 D. W. Fischer, X-Ray Band Spectra and Molecular-Orbital
48 W. Grünert, R. Schloegl and H. G. Karge, Investigations of Structure of Rutile Ti O 2, Physical Review B, 1972, 5, 4219.
zeolites by photoelectron and ion scattering spectroscopy. 1. 65 R. Krol, in Photoelectrochemical Hydrogen Production, eds.
New applications of surface spectroscopic methods to zeolites R. van de Krol and M. Grätzel, Springer US, Boston, MA,
by a high-temperature measurement technique, The Journal 2012, vol. 102, pp. 13–67.
of Physical Chemistry, 1993, 97, 8638–8645. 66 R. Asahi, Y. Taga, W. Mannstadt and A. J. Freeman,
Electronic and optical properties of anatase TiO2, Physical
Review B, 2000, 61, 7459.

8 | J. Name., 2012, 00, 1-3 This journal is © The Royal Society of Chemistry 20xx

Please do not adjust margins

 Page 9 of 10 Physical
PleaseChemistry Chemical
do not adjust Physics
margins
View Article Online
DOI: 10.1039/C8CP04614J
Journal Name ARTICLE

67 M. Harb, P. Sautet and P. Raybaud, Origin of the Enhanced

Visible-Light Absorption in N-Doped Bulk Anatase TiO2
from First-Principles Calculations, The Journal of Physical
Chemistry C, 2011, 115, 19394–19404.
68 L.-B. Mo, Y. Wang, Y. Bai, Q.-Y. Xiang, Q. Li, W.-Q. Yao,
J.-O. Wang, K. Ibrahim, H.-H. Wang, C.-H. Wan and J.-L.

Physical Chemistry Chemical Physics Accepted Manuscript

Cao, Hydrogen Impurity Defects in Rutile TiO2, Scientific
Reports, 2015, 5, 17634.
69 M. E. Arroyo-de Dompablo, A. Morales-García and M.
Taravillo, DFT+U calculations of crystal lattice, electronic
structure, and phase stability under pressure of TiO2
polymorphs, The Journal of Chemical Physics, 2011, 135,
Published on 17 September 2018. Downloaded by Gazi Universitesi on 9/24/2018 [Link] AM.

054503.
70 W. Chen, Q. Kuang, Q. Wang and Z. Xie, Engineering a high
energy surface of anatase TiO2 crystals towards enhanced
performance for energy conversion and environmental
applications, RSC Advances, 2015, 5, 20396–20409.
71 D. Selli, G. Fazio and C. Di Valentin, Modelling realistic
TiO2 nanospheres: A benchmark study of SCC-DFTB against
hybrid DFT, The Journal of Chemical Physics, 2017, 147,
164701.
72 J. Zhang, P. Zhou, J. Liu and J. Yu, New understanding of the
difference of photocatalytic activity among anatase, rutile and
brookite TiO2, Phys. Chem. Chem. Phys., 2014, 16, 20382–
20386.
73 J.-M. Herrmann, Heterogeneous photocatalysis: state of the
art and present applications., Topics in Catalysis, 2005, 34,
49–65.
74 T. Luttrell, S. Halpegamage, J. Tao, A. Kramer, E. Sutter and
M. Batzill, Why is anatase a better photocatalyst than rutile? -
Model studies on epitaxial TiO2 films, Scientific Reports,
2015, 4, 1038.
75 A. Calloni, A. Ferrari, A. Brambilla, F. Ciccacci and L. Duò,
Growth of stoichiometric TiO2 thin films on Au(100)
substrates by molecular beam epitaxy, Thin Solid Films,
2012, 520, 3922–3926.
76 M. Kobielusz, K. Pilarczyk, E. Świętek, K. Szaciłowski and
W. Macyk, Spectroelectrochemical analysis of TiO2
electronic states – Implications for the photocatalytic activity
of anatase and rutile, Catalysis Today, 2018, 309, 35–42.

This journal is © The Royal Society of Chemistry 20xx J. Name., 2013, 00, 1-3 | 9

Please do not adjust margins

 Physical Chemistry Chemical Physics Page 10 of 10
View Article Online
DOI: 10.1039/C8CP04614J

Physical Chemistry Chemical Physics Accepted Manuscript

Published on 17 September 2018. Downloaded by Gazi Universitesi on 9/24/2018 [Link] AM.

Absolute energy levels position and quantitative description of occupied levels obtained for TiO2
nanopowders, combining UPS and UV-Vis spectroscopies.