Content uploaded by Adam Georg Balogh
Author content
All content in this area was uploaded by Adam Georg Balogh on Dec 17, 2014
Content may be subject to copyright.
This content has been downloaded from IOPscience. Please scroll down to see the full text.
Download details:
IP Address: 195.113.27.90
This content was downloaded on 17/12/2014 at 14:43
Please note that terms and conditions apply.
Hydrogen storage in Ti, V and their oxides-based thin films
View the table of contents for this issue, or go to the journal homepage for more
2015 Adv. Nat. Sci: Nanosci. Nanotechnol. 6 013002
(http://iopscience.iop.org/2043-6262/6/1/013002)
Home Search Collections Journals About Contact us My IOPscience
Review
Hydrogen storage in Ti, V and their oxides-
based thin films*
Z Tarnawski
1
, K Zakrzewska
2
, N-T H Kim-Ngan
3
, M Krupska
3
, S Sowa
3
,
K Drogowska
1
, L Havela
4
and A G Balogh
5
1
Faculty of Physics and Applied Computer Science, AGH University of Science and Technology, 30-059
Kraków, Poland
2
Faculty of Computer Science, Electronics and Telecommunication, AGH University of Science and
Technology, 30-059 Kraków, Poland
3
Institute of Physics, Pedagogical University, 30-084 Kraków, Poland
4
Faculty of Mathematics and Physics, Charles University, Ke Karlovu 5, 12116 Prague, Czech Republic
5
Institute of Materials Science, Technische Universität Darmstadt, 64287 Darmstadt, Germany
E-mail: tarnawsk@agh.edu.pl
Received 9 November 2014
Accepted for publication 20 November 2014
Published 16 December 2014
Abstract
We have investigated the hydrogen storage ability and the effect of hydrogenation on structure
and physical properties of Ti/V and their oxides-based thin films. A series of Ti–TiO
2
and VO
x
–
TiO
2
thin films with different layer structures, geometries and thicknesses have been prepared by
the sputtering technique on different (Si(111), SiO
2
, C) substrates. For the Ti–TiO
2
–Ti films up
to 50 at.% of hydrogen can be stored in the Ti layers, while the hydrogen can penetrate without
accumulation through the TiO
2
layer. A large hydrogen storage was also found in some V
2
O
5
–
TiO
2
films. Hydrogen could also remove the preferential orientation in the Ti films and induce a
transition of V
2
O
5
to VO
2
in the films.
Keywords: titan oxides, vanadium oxides, multilayers, crystal structure, hydrogen storage
Mathematics Subject Classification: 2.00, 5.00, 5.13
1. Introduction
Hydrogen has been generally accepted as a fuel in the future.
There is currently a lot of hope in hydrogen-based energy.
The construction of hydrogen-based energy systems brings up
new issues, such as the interaction of hydrogen and matter in
the solid state, the search for hydrogen storage materials etc
This work is a review of our recent work of investigations of
the hydrogen storage as well as the effect of hydrogenation on
structure and physical properties of Ti and TiO
2
-based thin
films. Some new data obtained for V and VO
x
-based thin
films will be also presented.
Titanium dioxide, TiO
2
, has been widely used as pig-
ments and paints [1], in the optical coatings for laser mirrors,
interference filters, etc [2,3]. Its photocatalytic properties (i.e.
breaking water into hydrogen and oxygen on TiO
2
electrodes)
have been discovered by Fujishima and Honda since 1972
[4]. TiO
2
nowadays finds various novel applications in pho-
toelectrochemistry, photocatalysis, solar cells and gas sensors
[5–7]. For an extensive overview of the development of TiO
2
-
based photocatalysis and its future prospects, see [8–10].
An increased interest in TiO
2
has been recently focused
on the development of its nanostructured forms (nanotubes,
nanorods, etc) for renewable energy sources and hydrogen
economy. For example, TiO
2
nanotubes grown on Ti sub-
strates by anodization could be used for hydrogen storage
[11]. Metal–insulator–metal (MIM) structures such as Ti/
TiO
2
/Pt were proposed for resistance random access
|Vietnam Academy of Science and Technology Advances in Natural Sciences: Nanoscience and Nanotechnology
Adv. Nat. Sci.: Nanosci. Nanotechnol. 6(2015) 013002 (8pp) doi:10.1088/2043-6262/6/1/013002
*Invited talk at the 7th International Workshop on Advanced Materials
Science and Nanotechnology IWAMSN2014, 2-6 November, 2014, Ha
Long, Vietnam.
2043-6262/15/013002+08$33.00 © 2015 Vietnam Academy of Science & Technology1
memories (ReRAM) [12,13]. Light induced metal-semi-
conductor reversible transition at room temperature has
been demonstrated for TiO
x
nanoparticles making it a pro-
mising material for high density storage media [14].
The vanadium oxide family (V
2
O
5
,V
2
O
3
,VO
2
)is
among the most attractive class of smart materials exhibit-
ing semiconductor–metal transition. Vanadium pent-oxide
(V
2
O
5
), the most stable vanadium oxide having a wide band
gap and being an n-type semiconductor material, has been
used especially as optical and electrical devices, e.g. V
2
O
5
thin films can be applied in electrochromic and electro-
chemical devices [15] and microscale batteries [16]. A large
interest is focusing on investigations of TiO
2
–V
2
O
5
thin
films to gain the optimal electrochromic properties [17]
regarding their potential applications for electrochromic
smart windows. The interaction of V
2
O
5
with TiO
2
would
strongly affect the structural properties and reactivity of
V
2
O
5
; in particular, it leads to V
2
O
5
dispersion [18,19]. We
notice here that due to oxygen reduction by using e.g.
hydrogen, V
2
O
5
can be converted to vanadium dioxide VO
2
and a complex mixture such as V
4
O
7
and V
5
O
9
before V
2
O
3
is reached. The promoting effect of a titanium oxide support
on the catalytic properties of vanadium oxide is widely
recognized [20]. Investigations of thermochromic properties
have revealed that VO
2
–TiO
2
multilayers have a higher
luminous transmittance than that of single VO
2
film and
could yield a large change of solar transmittance at both
temperatures below and above semiconductor–metal tran-
sition temperature (∼333K)ofVO
2
[21].
Introduction of hydrogen into the crystal lattice in
general leads to a modificationofbothcrystalandelectronic
structure. Hydrogen absorption brings a relatively small
perturbation to the system (e.g. the lattice expansion and the
hydrogen bonding with other atoms in the lattice). It,
however, often implies some new and interesting features in
the systems. Thin films and multilayers often play an
important role in the improvement of hydrogen absorption
rate as well as the chemical and crystal structure stability.
On the other hand, the atom mixing, diffusion across the
interfaces and precipitation of nanoparticles may also affect
the hydrogen uptake-release cycling as well as the ther-
modynamic properties of the films.
We aim at characterization of the film structure and
properties of the TiO
2
-based thin films, in particular the
interlayer properties and the influence of hydrogen intake on
the microstructure and electronic structure of the films.
Numerous TiO
2
-based thin films, consisting of TiO
2
, Ti and
VO
x
layers with a different layer structure (e.g. with single-,
bi- and tri-layer structure of different layer sequences), have
been prepared by means of sputtering technique and investi-
gated. Selected films subjected for hydrogenation were then
investigated focusing on film stability, hydrogen uptake and
hydrogen storage under different conditions. Details of the
experiments and the main outcome of our study are presented
in the next sections.
2. Experimentals
Thin films consisted of Ti, TiO
2
and VO
x
layers with different
layer sequences and different layer-thicknesses have been
deposited by means of magnetron dc pulse sputtering system
[22] on Si(111), silica SiO
2
and C-foil substrates. Our ana-
lysis of all investigated layers indicates the presence of TiO
2
rutile, while several vanadium oxide forms (V
2
O
5
,VO
2
…)
can exist. For the selected film, an additional Pd layer was
deposited on the film surface by MBE technique.
Selected films with chosen film-structure have been
subjected to hydrogenation at atmospheric atmosphere (1 bar)
and at 300 °C (so-called hydrogen charging) and/or at high
hydrogen pressure up to 102 bar and at room temperature
(RT) with different times. For the VO
x
–TiO
2
series, some
samples were charged with hydrogen at 1 bar only once,
while some were charged twice. Each charging was for 3 h.
For investigating the possibility of enhancement of
hydrogen absorption in the thin film system, an additional
palladium layer was deposited on a chosen film by molecular
beam epitaxy (MBE) technique.
The film chemical composition, depth profile, layer
thickness and structure were determined by combined ana-
lysis of x-ray diffraction (XRD), x-ray reflectometry (XRR),
Rutherford back-scattering (RBS) and optical spectro-
photometry. In order to compare the layer-thickness deter-
mination from three independent complementary methods, we
also evaluated the layer thickness in nm from RBS data. More
details of RBS analysis are described elsewhere [23,24].
The hydrogen profile was determined by means of a
secondary ion mass spectroscopy (SIMS) and N-15 Nuclear
Reaction Analysis (
15
N-NRA method). SIMS was carried out
by using Cs
+
primary ions recording positive secondary ions
by a CAMECA ims 5f equipment. The reaction
15
N+
1
H→
12
C+α+γ(4.965 MeV) at a resonance energy of 6.417 MeV
was used for the
15
N-NRA method to obtain the results. For
data evaluation the computer code SRIM was used.
3. Results and discussion
3.1. Hydrogen storage in Ti–TiO
2
films
Hydrogen storage in these systems have been thoroughly
investigated and reported in our previous publications [25–
27]. Our investigations revealed that:
I. On the as-deposited films of Ti-TiO
2
system:
1. Single Ti nucleates and grows as a compact layer on
the well-defined (111) plane of Si wafer–Ti/Si(111).
Interdiffusion was not found at the Ti/Si interface,
i.e. a sharp interface was always obtained. This Ti
film exhibits a strong preferred orientation with
(00.1) plane parallel to the substrate,
2. A small amount of interdiffussion was found at the
Ti–TiO
2
interface of the bi-layer film and at both Ti–
TiO
2
and TiO
2
–Ti interfaces of the tri-layer film. The
intermediate TiO
2
layer exhibits columnar structure
2
Adv. Nat. Sci.: Nanosci. Nanotechnol. 6(2015) 013002 Review
and there is some intergrowth between the Ti and
TiO
2
layers. The interdiffusion is attributed to Ti
diffusion along the channels formed between the
TiO
2
columns,
3. More precise determination of layer thickness and
element concentration, in particular the oxygen
concentration, can be obtained for films deposited
on C-foils. However, in this case a strong carbon
diffusion (up to 10 at.%) into the film was observed.
If the film is thin (the layer thickness <30 nm), the
carbon can be found even in the surface layer due to
carbon segregation from the substrates,
II. On hydrogen-charged Ti–TiO
2
films:
4. High hydrogen concentration (storage), up to a value
of over 40–50 at.%, was obtained for the top Ti-
layer,
5. Palladium could act as a good catalyst for hydrogen
diffusion in the Ti–TiO2–Ti films. Without covering
the film surface by palladium, the hydrogen
concentration in the bottom Ti layer has reached
only 15 at.%, whereas it increases up to 40 at.%
when the film was covered by palladium,
6. Hydrogen could be moved through TiO2 layer
without any accumulation there,
7. The preferential orientation in the Ti films was
destroyed/disappeared by hydrogen charging under
high pressures (pH2 = 100 bar),
8. Large swelling effect was observed for the thick Ti
layer (>240 nm) after hydrogen charging at 100 bar.
The enhanced hydrogen concentration (enhanced
storage) leads to an increase of the film thickness up
to 150% of its original value.
The hydrogen profile determined by
15
N-NRA for the Ti/
TiO
2
/Ti/Si(111) films revealing the large hydrogen storage up
to 40–50 at.% is shown in figure 1. More detailed results can
be found in our previous publication [26].
3.2. Hydrogen storage in VO
x
–TiO
2
films
Prior to the hydrogenation experiments, we have studied
thoroughly the properties of layered structures of numerous
VO
x
–TiO
2
films with different film-geometry and thickness.
The measured and simulated RBS spectra for the 5 films
deposited on silica SiO
2
substrates are shown in figure 2.
Estimated layer composition and thickness for each layer in
the films are given in table 1.
The RBS spectrum of V
2
O
5
/SiO
2
film (VS1) is char-
acterized by a large V-peak at energy around 1250 keV from
the film, a steep Si- and O edge from the substrate, respec-
tively, at 960 keV and 616 keV. Oxygen is present both in
the film and in the substrate, thus no separated oxygen signal
from the filmwasobserved;itspresenceimpliessome
increase of the signal on the left-hand side of the O-edge
(below 616 keV) in the RBS spectra. The film structure can
be described as follows: 1) the upmost surface layer of the
film indeed consists of the most stable and common vana-
dium oxide V
2
O
5
with a thickness of 13 nm (denoted as
layer 1 of sample VS1 in table 1), 2) beneath the V
2
O
5
layer
is a thick layer with a composition of V
2
O
5−x
(layer 2) with x
increasing from 0 to 3.0 and thickness of 51 nm, 3) beneath
the V
2
O
5−x
layer is the stoichiometric VO
2
layer with a
thickness of 15 nm (layer 3) and 4) an interface layer con-
sisted of a mixture of VO
2
+SiO
2
wasformedasacon-
sequence of V diffusion deeply into the SiO
2
substrate
(revealed by the non-zero background between the V-peak
and Si-edge). A more detail analysis of the mixture layer
V
2
O
5−x
(layer 2, thickness 51 nm) indicates that it consists of
two sub-layers, the first one with a thickness of 24 nm
beneath the surface V
2
O
5
layer1)isasamixtureofV
2
O
5
,
V
2
O
7
(i.e. x=1.5) and V
5
O
9
(x= 1.4), while the second one
(above the VO
2
layer3)withathicknessof27nmisa
mixture of V
2
O
3
layer (x=2.0) and VO (x=3.0).
Our results indicate that during film deposition, first the
VO
2
layer was formed on SiO
2
substrate (with an oxygen
content of 33.3%). With increasing deposition time, the
Figure 1. Comparison of the hydrogen profile determined by
15
N-NRA for the Ti/TiO
2
/Ti/Si(111) film (a) without covering Pd and with Pd
cover (b). The hydrogen charging was at 1 bar and at 300 °C.
3
Adv. Nat. Sci.: Nanosci. Nanotechnol. 6(2015) 013002 Review
oxygen content increases to a higher value to form the mixed
V
2
O
3
+ VO layer (with the oxygen content of 40–50%).
Increasing further the deposition time, the oxygen reduction
leads to a formation of layer with a complex mixture of
V
2
O
5
,V
2
O
7
and V
5
O
9
until the stable V
2
O
5
(with the
oxygen content of 28.6%) is reached and formed on the film
surface. In all cases, from the estimated value of metal (M)
andoxygen(O)contentbySIMNRA,thetypeofoxides
consistinginthelayercanbeeasilydefined. For instance,
the M and O content in M
2
O
5
is respectively 28.6% and
71.4%, while it amounts to, respectively, 33.3% and 66.7%
for MO
2
.TheSIMNRAfit for RBS spectrum is simulated
for (Nt) product, i.e. the areal density (the number of target
atoms per unit area). The values of layer-thickness in (nm)
are converted from the simulated (Nt)values(in10
15
atoms cm
−2
) using the conversion coefficient and estimated
percentage (%) of each oxide presented in the layers. For
instance, for the stoichiometric V
2
O
5
layer (1) in the sample
VS2
⎡
⎣⎤
⎦
=−
d
nNt[nm] 10 at cm ,
15 2
while for layer (2) in the sample, VS2 consisted of 85%
TiO
2
,9%VO
2
and 6% SiO
2
⎡
⎣⎤
⎦
=++
−
d
mnpNt[nm] (0.85 0.09 0.06 )10 at cm ,
15 2
where m,n,pare the corresponding conversion coefficients
estimated using the bulk density of different oxides (e.g. ρ
(TiO
2
) = 4.23 g cm
−3
,ρ(VO
2
)=4.57gcm
−3
,ρ
(V
2
O
5
)=3.36gcm
−3
). In most cases, the thickness could be
estimated with a good accuracy, since the layer either con-
sists of only stoichiometric oxide (such as TiO
2
,V
2
O
5
,VO
2
)
or the mixture of oxides of the same type (such as
(TiO
2
+VO
2
)or(VO
2
+SiO
2
) mixture). For such a mixture,
since the oxygen content is the same (66.7%), the percentage
of different oxides can be easily estimated based on the ratio
Figure 2. Random RBS (markers) and SIMNRA (lines) simulated spectra for (a) 79 nm thick V
2
O
5
/SiO
2
(sample VS1) and 122 nm thick
V
2
O
5
/TiO
2
/SiO
2
(VS2) and (b) 117 nm thick TiO
2
/V
2
O
5
/SiO
2
(VS3), 184 nm thick TiO
2
/V
2
O
5
/SiO
2
(VS4) and 193 nm thick TiO
2
+ V/SiO
2
(VS5) after deposition. RBS experiments were performed with the incident He
+
ion energy of 1.7 MeV and the backscattering angle of 171°.
The films are denoted by nominal chemical composition and the estimated total thickness (see table 1). The curves were normalized and
shifted for clarity.
Table 1. Chemical composition of TiO
2
–VO
x
films deposited on silica SiO
2
substrates after deposition (as-deposited films): V
2
O
5
/SiO
2
(denoted as sample VS1), V
2
O
5
/TiO
2
/SiO
2
(VS2), TiO
2
/V
2
O
5
/SiO
2
(VS3 and VS4, with the same layer geometry but different thickness) and
TiO
2
+ V/SiO
2
(VS5). The layer-thickness of each layer (d(nm)), the percentage (%) of each type of oxides (TiO
2
,V
2
O
5
,VO
2
) presented in
the film are estimated using SIMNRA and the mass density of the bulk (see text). Layer (1) denotes the surface layer. The layer (2) of sample
S1 consists of V
2
O
5−x
with x=0−3.0 (see text).
Layer V
2
O
5
(%) TiO
2
(%) VO
2
(%) SiO
2
(%) d(nm) D(nm)
V
2
O
5
/SiO
2
1 100 ———13 79
(VS1) 2 100 (V
2
O
5−x
)———51 —
—3——100 —15 —
V
2
O
5
/TiO
2
/SiO
2
1 100 ———68 122
(VS2) 2 —85 9 6 54 —
TiO
2
/V
2
O
5
/SiO
2
1—100 —— 53 117
(VS3) 2 100 ———64
TiO
2
/V
2
O
5
/SiO
2
1—100 —— 113 184
(VS4) 2 100 ———71 —
TiO
2
+ V/SiO
2
(VS5) 1 —97 3 —193 193
4
Adv. Nat. Sci.: Nanosci. Nanotechnol. 6(2015) 013002 Review
between different M components, e.g. t(%) TiO
2
=u(%)Ti/
33.3(%) where u(%)Ti is the estimated percentage of Ti in
the layer and 33.3% amounts to the summation of percen-
tage of all metal contents in the layer (u(%)Ti + v(%)V + w
(%)Si = 33.3(%)). Some difficultyarisesinthethickness
conversion for the layer consisting of a mixture of different
oxides, such as layer (2) of sample VS1 (V
2
O
5−x
). In this
case, it is more difficult to estimate the exact ratio of dif-
ferent oxides based on only the ratio of M contents, since the
O content is different for each oxide. We notice here that
besides uncertainty in thickness conversion, the ambiguity in
determination of the layer thickness is also related to the
layer quality itself. Namely, the sputter deposited films may
have some porosities or defects and thus the mass density of
the film is certainly different from that of the bulk.
The RBS spectrum of V
2
O
5
/TiO
2
/SiO
2
film (VS2) is
characterized by a large V-peak at an energy around
1250 keV and a large Ti-peak at an energy around 1200 keV
from the film, a steep Si- and O edge from the substrate (at a
lower energy than that for the VS1 film, since the film
thickness is larger). The thickness of the V
2
O
5
and TiO
2
layer
are quite similar and so are the V- and Ti-content in the film,
thus a clear minimum between the V- and T-peak was
observed. The TiO
2
was first deposited on the SiO
2
substrate
and then the V
2
O
5
layer followed in this case. No stoichio-
metric TiO
2
was found. Instead, a mixed layer
(TiO
2
+VO
2
+ SiO
2
) with a thickness of 54 was formed (layer
2, sample VS2) as a consequence from some Si diffusion (6%
SiO
2
) and V-diffusion (9% VO
2
) into the TiO
2
film. How-
ever, the surface layer with a thickness of 68 nm consists of
only stoichiometric V
2
O
5
.
In the case of two TiO
2
/V
2
O
5
/SiO
2
films (VS3 and VS4),
i.e. deposition sequence is first the vanadium and then the
titanium oxide, our analysis reveals that each film consists of
only stoichiometric V
2
O
5
and TiO
2
layer. For the VS3 film,
the thickness of TiO
2
(64 nm) and V
2
O
5
layer (53 nm) is in
the same order of magnitude. Besides, the film is quite thin.
Thus, the V- and Ti- signal was combined into one large peak
at around 1200 keV in the RBS spectrum. For the VS4 film,
the thickness of TiO
2
layer is estimated to be 113 nm, while it
equals 71 nm for V
2
O
5
layer (see table 2). Both layers are
thicker than those of VS3. Besides, the thickness of TiO
2
film
is about 1.5 times larger than that of the V
2
O
5
one, i.e. the Ti
content in the film is much larger. This leads to the wide
shoulder (V-signal) and the large peak (Ti signal) in the RBS
spectra. In the case of TiO
2
+V/SiO
2
film (VS5), the V content
is estimated to be 1% for the entire film, i.e. the film com-
position is 97% TiO
2
+3% VO
2
. Since the film is thick
(193 nm) and the V-content is small, only a broad peak was
observed in the RBS spectrum.
Estimated layer composition and thickness of VO
x
–TiO
2
films after each charging with hydrogen are given in table 2.
We focus on analyzing the hydrogen charging results on
sample VS2 (with V
2
O
5
as the surface layer) and VS4 (with
TiO
2
as the surface layer). These films were charged by
hydrogen twice denoted as H(1) and H(2). The increased
thickness of the film (%) due to hydrogenation was estimated
for the total film thickness (D(nm)) with respect to that of as-
deposited film (=(D(after)-D(before))/D(before charging)).
As an example, a comparison of the measured and simulated
RBS spectra for the V
2
O
5
/TiO
2
/SiO
2
film (VS2) before and
after hydrogen charging are shown in figure 3. The Ti-peak
Table 2. Effect of hydrogenation on selected films: V
2
O
5
/TiO
2
/SiO
2
(VS2), TiO
2
/V
2
O
5
/SiO
2
(VS4). They were charged by hydrogen twice
(denoted, respectively, by H(1) and (H(2)), each charging was at pressure of 1 bar, at temperature of 300 °C and for 3 h. The layer-thickness
of each layer (d(nm)), the percentage (%) of each type of oxides (TiO
2
,V
2
O
5
,VO
2
) presented in the film were estimated using SIMNRA and
the mass density of the bulk (see text). Layer (1) denotes the surface layer. Due to enhanced Si diffusion from the SiO
2
substrate into the film,
the mixed TiO
2
–VO
2
–SiO
2
layer was formed at the film–substrate interface in all cases. The increased thickness of the film (%) due to
hydrogenation was estimated for the total film thickness (D(nm)) with respect to that of as-deposited film.
V
2
O
5
/TiO
2
/SiO
2
(VS2) Layer V
2
O
5
(%) TiO
2
(%) VO
2
(%) SiO
2
(%) d(nm) D(nm) Increased thickness
as-deposited 1 100 ———68 122 —
—2—85 9 6 54 ——
hydrogenation-1 1 100 ———62 131 7%
H (1) 2 62.5 26.5 11 56 ——
—319196213——
hydrogenation-2 1 100 ———62 140 15%
H (2) 2 —48 36 16 37 ——
—3—7187541——
TiO
2
/V
2
O
5
/SiO
2
(VS4) Layer V
2
O
5
(%) TiO
2
(%) VO
2
(%) SiO
2
(%) d(nm) D(nm) increased thickness
as-deposited 1 —100 —— 113 184 —
—2 100 ———71 ——
hydrogenation-1 1 —100 —— 112 184 0%
H (1) 2 100 ———58 ——
—3——15 85 14 ——
hydrogenation-2 1 —100 —— 111 187 2%
H (2) 2 —94 6 47 ——
—3——15 85 29 ——
5
Adv. Nat. Sci.: Nanosci. Nanotechnol. 6(2015) 013002 Review
and V-peak with almost equal intensity before hydrogen
charging was observed. The effect of hydrogen charging on
this film is revealed by a lowering of the Ti-peak and a
widening of this peak at the left hand side. It is caused by a
decrease of the percentage of TiO
2
in layer 2 as well as the
appearance of an extra mixed layer in the interface (see
table 2). The total film thickness is increased, respectively, by
7% and 15% after the first and second hydrogen charging.
The thickness change is large enough in this case, which can
be seen in the RBS spectrum. No visible hydrogen effect can
be seen in RBS spectra from hydrogen charging for the other
samples (VS1, VS3, VS4 and VS5). A comparison of the
measured and simulated RBS spectra for sample VS4 before
and after hydrogen charging is shown in figure 4. The RBS
spectra before and after hydrogen charging are similar. The
effect from hydrogen charging is mostly revealed by the
change of the layer content and layer thickness, but the
change in the total thickness is in the range of e.g. 2–3% for
VS3 and VS4 film.
We concentrate on analyzing the most visible effect from
hydrogen charging, i.e. on the V
2
O
5
/TiO
2
/SiO
2
film (VS2).
The hydrogen charging leads to a decrease of the layer
thickness of the stoichiometric V
2
O
5
layer (layer 1) from
68 nm to 62 nm, while that of the mixed TiO
2
+VO
2
+SiO
2
layer (layer 2) increases from 54 nm to 56 nm after the first
charging. The TiO
2
percentage in layer 2 is only 62.5%, much
lower than that before charging (85%). The VO
2
percentage
increases from 9% to 26.5%. Besides, an extra layer with the
same mixture as that of the layer 2 but with a lower percen-
tage of TiO
2
(19%) and VO
2
(19%) and a higher percentage
of SiO
2
(62%) does appear (with a thickness of 13 nm). It can
be explained as an enhancement of both Ti diffusion into the
SiO
2
substrate and Si diffusion out from substrate into the
film. We notice here that the film is heated up to 300 °C for
3 h upon hydrogen charging. It certainly promotes such dif-
fusion and as a consequence increases the thickness of the
interface layers. Indeed the second hydrogen charging indu-
ces a larger enhancement; the percentage of SiO
2
in the
interface layers is largely enhanced and reached even 75%.
Our results clearly show the transition from V
2
O
5
to VO
2
,or
in other words, a reduction of V
2
O
5
and increase of VO
2
due
to hydrogen charging.
We did not construct the depth profile from the RBS data,
i.e. the concentration of each element as a function of the film
thickness, just because the relative change between different
layers is very small. For a clear demonstration of the hydro-
gen effect, we construct the film diagram, where the layer
thickness of different layers is drawn proportionally with
respect to the values given in table 2. Different compositions
of different oxides in the layers are presented by using dif-
ferent colours. The film diagram was shown for VS2, VS3
and VS4, respectively, in figures 5and 6, since the hydrogen
charging leads to a change of layer thickness (up to 15%) and
composition in these cases by hydrogen.
The obtained results on VO
x
–TiO
2
films revealed that:
(1) Stoichiometric V
2
O
5
and TiO
2
layers were obtained if
the deposition sequence was first the vanadium and then
the titanium oxide,
(2) The V
2
O
5
reduction upon hydrogen charging, i.e. the
V
2
O
5
–VO
2
transition, was always observed,
(3) In the case when the V
2
O
5
layer is on the surface, the
hydrogen charging effect is much enhanced, indicated
by a large increase of the total film thickness (up to 15%
Figure 3. Random RBS (markers) and SIMNRA (lines) simulated
spectra for V
2
O
5
/TiO
2
/SiO
2
(VS2) before (as-deposited) and after
two times of hydrogen charging (denoted, respectively, as H(1) and
H(2) ). RBS experiments were performed with the incident He
+
ion
energy of 1.7 MeV and the backscattering angle of 171°. The curves
were normalized and shifted for clarity. The hydrogen charging was
seen by e.g. the relative change in the peak-intensity of Ti- and V-
signal.
Figure 4. Random RBS (markers) and SIMNRA (lines) simulated
spectra for TiO
2
/V
2
O
5
/SiO
2
(VS4) before (as-deposited) and after
two times of hydrogen charging (denoted respectively as H (1) and H
(2)). RBS experiments were performed with the incident He
+
ion
energy of 1.7 MeV and the backscattering angle of 171°. The curves
were normalized and shifted for a clarity. No visible effect from
hydrogen charging was seen.
6
Adv. Nat. Sci.: Nanosci. Nanotechnol. 6(2015) 013002 Review
of its original value after hydrogen charging of 6 h). It
reveals the large hydrogen storage in the film,
(4) The V
2
O
5
can be well preserved upon hydrogen
charging if it locates on the film surface, as in the case
of V
2
O
5
/TiO
2
/SiO
2
film,
(5) In the case when the TiO
2
layer is on the surface, the
film thickness does not change much (only 2–3%) upon
hydrogen charging. However, a larger reduction of
V
2
O
5
is observed. Namely, after 6 h charging, a
complete transition of V
2
O
5
into VO
2
can be obtained.
It indicates that the TiO
2
layer acts as ‘hydrogen
catalyst’for such V
2
O
5
–VO
2
transition.
(6) The results obtained for TiO
2
+ V/SiO
2
film confirmed
that no hydrogen is accumulated in TiO
2
even if it is
doped with vanadium. However, in this case, since the
V-doping is very small (1%) and the film is thick, VO
2
(3%) exists in the film. Thus there is no possibility to
observe the V
2
O
5
–VO
2
transition.
4. Concluding remarks
The most important finding of our investigations is that
hydrogen can be stored largely in the Ti layer (with hydrogen
content to 50%) in thin films of Ti–TiO
2
system and that
palladium could act as a good catalyst for hydrogen diffusion
into the films. A large hydrogen absorption can be obtained in
the thin films of VO
x
–TiO
2
system if the surface layer is the
V
2
O
5
layer. Besides, the introduction of hydrogen could also
remove the Ti preferential orientation and/or induce a V
2
O
5
–
VO
2
transition in the films.
Acknowledgments
The authors highly acknowledged the great help and fruitful
cooperation of A Brudnik (AGH Krakow), S Flege and C
Schmitt (Damstadt University of Technology), D Rogalla and
H-W Becker (Dynamitron Tandem Lab, Ruhr-Universität
Bochum), R Kužel and V Sechovsky (Charles University).
We acknowledge the support of the Czech-Polish cooperation
by the Czech Ministry of Education (Czech-polish project
7AMB14PL036 (9004/R14/R15). N-THKN acknowledged
the financial support by the European Regional Development
Fund under the Infrastructure and Environment Programme.
References
[1] Yan Q, Lei Y, Yuan J and Coat J 2010 Technol. Res. 7229
[2] Jacobson P, Li S-C, Wang C and Diebold U 2008 J. Vac. Sci.
Technol. B26 2236
[3] Ankerhold G, Mitsehke F, Frerking D, Ristau D, Mlynek J and
Lange W 1989 Appl. Phys. B48 101
[4] Fujishima A and Honda K 1972 Nature 238 37
[5] Linsebigler A L, Lu G and Yates J T Jr 1995 Chem. Rev.
95 735
[6] O’Regan B and Grätzel M 1991 Nature 353 737
[7] Varghese O K and Grimes C A J 2003 Nanosci. Nanotech.
3277
[8] Fujishima A, Zhang X and Tryk D A 2008 Surf. Sci. Rep.
63 515
[9] Henderson M A 2011 Surf. Sci. Rep. 66 185
[10] Baraton M-I 2011 Open Nanosci. J. 564
[11] Pillai P, Raja K S and Misra M 2006 J. Power Sources 161 524
[12] Choi B J et al 2005 J. Appl. Phys. 98 033715
[13] Kim W-G and Rhee S-W 2010 Microelectron. Eng. 87 98
Figure 5. Illustration of the influence of hydrogenation on the
structure and composition of V
2
O
5
/TiO
2
/SiO
2
film (VS2). The layer
thicknesses are drawn proportionally with respect to the values given
in table 2. The solid line indicates the original separation between the
film and the SiO
2
substrate. Different colours indicate different
compositions of different oxides (TiO
2
,VO
2
, SiO
2
) in the layers.
The dashed line reveals the film thickness after two times of
hydrogenation with respect to that of the as-deposited film (thin solid
line). The estimated total film thickness before and after hydro-
genation is also given.
Figure 6. Illustration of the influence of hydrogenation effect on the
structure and composition of two films with a similar layer geometry
TiO
2
/V
2
O
5
/SiO
2
but with a different thickness (VS3 (top), VS4
(bottom)). The layer thicknesses are drawn proportionally with
respect to the values given in table 2. The solid line indicates the
original separation between the film and the SiO
2
substrate. Different
colours indicate different compositions of different oxides (VO
2
,
SiO
2
) in the layers. The dashed line reveals the film thickness after
two times of hydrogenation with respect to that of the as-deposited
film (thin solid line). The estimated total film thickness before and
after hydrogenation is also given.
7
Adv. Nat. Sci.: Nanosci. Nanotechnol. 6(2015) 013002 Review
[14] Ohkoshi S, Tsunobuchi Y, Matsuda T, Hashimoto K,
Namai A, Hakoe F and Tokoro H 2010 Nat. Chem. 2539
[15] Rosaiah P and Hussain O M 2013 Advanced Nanomaterials
and Nanotechnology (Springer Proceedings in Physics) vol
143 (Berlin: Springer) p 485
[16] Zhang L T, Song J, Dong Q F and Wu S T 2009 Adv. Mater.
Res. 79–82 931
[17] Zhang Q et al 2001 J. Mater. Sci. Technol. 17 417
[18] Honicke D and Xu J 1988 J. Phys. Chem. 92 4699
[19] Xie Y C and Tang Y Q 1990 Adv. Catal. 37 1
[20] see e.g. Centi G 1996 Appl. Catal. A147 267
[21] Mlyuka N R, Niklasson G A and Granqvist C G 2009 Sol.
Energy Mater. Sol. Cells 93 1685
[22] Brudnik A, Czternastek H, Zakrzewska K and Jachimowski M
1991 Thin Solid Films 199 45
[23] Kim-Ngan N-T H, Balogh A G, Meyer J D, Brötz J,
Hummelt S, ZającM,Ślęzak T and Korecki J 2008 Surf. Sci.
602 2358
[24] Kim-Ngan N-T H, Balogh A G, Meyer J D, Brötz J, ZającM,
Ślęzak T and Korecki J 2009 Surf. Sci. 603 1175
[25] Drogowska K, Tarnawski Z, Brudnik A, Kusior E,
Sokołowski M, Zakrzewska K, Reszka A,
Kim-Ngan N-T H and Balogh A G 2012 Mater. Res. Bull.
47 296
[26] Tarnawski Z et al 2013 Adv. Nat. Sci.: Nanosci. Nanotechnol.
4025004
[27] Drogowska K, Flege S, Schmitt C, Rogalla D, Becker H-W,
Kim-Ngan Nhu–T H, Brudnik A, Tarnawski Z,
Zakrzewska K and Balogh A G 2012 Adv. Mater. Sci. Eng.
2012 art. 269603
8
Adv. Nat. Sci.: Nanosci. Nanotechnol. 6(2015) 013002 Review
