Fabrication and structural characterization of TiO nanoparticle soft-landed on substrate by the magnetron sputtering-gas aggregation method

Article abstract of DOI:10.1140/epjd/e2005-00123-8

We have succeeded to fabricate the thin film mainly or entirely consisting of nano-crystallites TiO on a substrate at room temperature by using the magnetron sputtering combined with the gas aggregation method which has the capability to deposit grown small particles and/or clusters of the desired compound material on the substrate in soft-landed mode. The important experimental parameters to control the growth of TiO particles are identified as partial pressure of additional oxygen gas, growth-reaction distance L _td, and LN₂ cooling of growth chamber. The smallest averaged diameter of TiO nano-particles is 5.62 nm and size distribution being from 2 to 8 nm under 50 mm of L_td. The compositional uniformity is confirmed by EDS and EELS. The chemical state of the film observed by Ti 2P XPS peak confirms the primarily growth of TiO and indicates the formation of minor content of amorphous Ti_1-xO_x and/or Ti_1-xC _x particularly at the surface of the film. EDP Sciences, Societa? Italiana di Fisica, Springer-Verlag 2005.

Full text of DOI:10.1140/epjd/e2005-00123-8

Eur. Phys. J. D 34, 79–82 (2005)
DOI: 10.1140/epjd/e2005-00123-8
THE EUROPEAN
PHYSICAL JOURNAL D
Fabrication and structural characterization of TiO nanoparticle
soft-landed on substrate by the magnetron sputtering-gas
aggregation method
S. Tanemura1,a, K. Kajino1, L. Miao1, S. Koide1, M. Tanemura1, S. Toh2, K. Kaneko2, and M. Mori3
1
Dept. of Environmental Technology and Urban Planning, Nagoya Institute of Technology, Gokiso-cho, Showa-ku,
Nagoya 466-8555, Japan
Dept. of Material Science and Engineering and Lab. HVEM, Kyushu Univ. Higashi-ku, Fukuoka 812-8581, Japan
Materials Research Laboratory, NGK Insulators Ltd., Mizuho-ku, Nagoya 467-8530, Japan
2
3
Received 6 September 2004
c
Published online 13 July 2005 – ꢀ EDP Sciences, Societ`a Italiana di Fisica, Springer-Verlag 2005
Abstract. We have succeeded to fabricate the thin film mainly or entirely consisting of nano-crystallites TiO
on a substrate at room temperature by using the magnetron sputtering combined with the gas aggregation
method which has the capability to deposit grown small particles and/or clusters of the desired compound
material on the substrate in soft-landed mode. The important experimental parameters to control the
growth of TiO particles are identified as partial pressure of additional oxygen gas, growth-reaction distance
Ltd, and LN2 cooling of growth chamber. The smallest averaged diameter of TiO nano-particles is 5.62 nm
and size distribution being from 2 to 8 nm under 50 mm of Ltd. The compositional uniformity is confirmed
by EDS and EELS. The chemical state of the film observed by Ti 2P XPS peak confirms the primarily
growth of TiO and indicates the formation of minor content of amorphous Ti1−xOx and/or Ti1−xCx
particularly at the surface of the film.
PACS. 81.15.Cd Deposition by sputtering – 81.10.-h Methods of crystal growth; physics of crystal growth
– 61.46.+w Nanoscale materials: clusters, nanoparticles, nanotubes, and nanocrystals – 61.66.-f Structure
of specific crystalline solids
1 Introduction
are difficulties to control precisely the ratio of metal vapor
flux and oxygen vapor flux as well as the densities of those
fluxes to assure homogeneous growth of TiOx (x = 1)
compound with very narrow range of x according to the
equilibrium-phase diagram of titanium/oxygen system.
The aim of the present study is to examine the
possibilities of the magnetron sputtering combined with
the gas aggregation method [7,8], which has the ca-
pability to deposit the small particles and/or clusters
of the desired compound materials as developed orig-
inally by Harberland et al. [9], for the fabrication of
the thin film consisting mainly or entirely of nano-
particles of TiO phase on a substrate at room temper-
ature. Their crystal structure and morphology, composi-
tional uniformity, and chemical state of the film, have been
identified by transmission electron microscopy (TEM), se-
lected area electron diffraction (SAED), energy disper-
sive spectroscopy (EDS), electron energy loss spectroscopy
(EELS) and X-ray photo-electron spectroscopy (XPS) re-
spectively.
Among transition metal oxides, conductive titanium
monoxide (TiO) with NaCl type crystal structure, which
has low electrical resistivity combined with a high anti-
diffusion efficiency to prevent the interface degradation [1]
and a lattice parameter close to MgO used as a substrate
for epitaxial growth of lead zirconia titanate (PZT) [2], has
recently considered as favorable thin films materials for
the application in low resistance contact metallization in
microelectronics layered structures [3]. Particularly, TiO
as an intermediate layer has promising role to enhance
the growth of the appropriate PLZ phase and to reduce
the mutual diffusion of the different layer materials for
the performance improvement of the layered ferroelectric
device [4].
The following three works have been only known as
the TiO thin film fabrication: (1) double step by reactive
evaporation at 300 ◦C followed by annealing at 550 ◦C [5],
(2) one step by reactive ion beam sputtering at 600 ◦C [6],
and (3) one step b◦y oxygen-ion-assisted physical vapor
deposition at <100 C [3]. The applied high temperature
in the first two methods is disadvantage for the fabrication
of the layered ferroelectric device. In third method, there
2 Experimental
The nano particles of TiO were grown by the magnetron
sputtering combined with the gas aggregation method by
a
e-mail: tanemura.sakae@nitech.ac.jp

80
The European Physical Journal D
Fig. 1. Schematics of the magnetron sputtering-gas aggrega-
tion source and the experimental set-up for the film deposi-
tion. The distance between the Ti target and the aperture is
denoted as Ltd. The distance between the aperture and the
substrate is fixed to 30 mm. The shutter is provided in the
front of the substrate holder. The deposition chamber (CH-A)
is separated from the outer chamber (CH-B) by the diaphragm
with 2 mm φ hole.
Fig. 2. Low-magnified
TEM images of the de-
posited thin film un-
der Ltd = 50 mm (a),
HRTEM image of the
same sample (b), and
SAED ring pattern (c).
(c)
(a)
(b)
following our previous reports [8,9]. The source consisted and EELS (JEOL 2010 FEF at 200KV, and JEOL JEM
of a DC magnetron sputtering gun with Ti target, a cylin- 4000EX at 400KV for high resolution TEM (HRTEM)
drical reaction-chamber for TiO formation under the oxi- image) and XPS (Physical Electronics, ESCA-5700ci.) re-
dation reaction of the sputtered Ti atoms/clusters, and a spectively. In EELS, Ω-type energy filter was installed to
liquid nitrogen reservoir to cool down the generated com- the JEOL 2010 FET. The samples used for those char-
pound particles at high temperature for the control of the acterizations by TEM and the related analysis were soft-
particle size distribution has shown schematically in Fig- landed TiO particles and/or clusters directly on a micro
ure 1. The base pressure in the source was 3 × 10−4 Pa. grid and the sample for XPS was those on a glass.
The surface of Ti target (50 mm φ) was cleaned by pre-
To confirm the optimal experimental parameters for
the deposition of the film mainly or entirely consisting
of nano-particles of TiO crystal, the three experiments
under the following different conditions were additionally
carried out: (1) zero oxygen partial pressure, (2) differ-
ent Ltd at 150, 100, 50 and 20 mm, and (3) no coolant
of LN2. The results of those experiments are summarized
in Section 3.4.
sputtering with Ar gas in prior to the deposition.
Firstly Ti atoms and clusters was sputtered off the
Ti target by DC magnetron sputtering under 2 × 102 Pa
in Ar gas pressure with constant electric power of 300 W.
The hot sputtered atoms and/or clusters was cooled down,
aggregated and reacted by the collision of the surround-
ing mixture gas of Ar with 2.6 × 102 Pa and oxygen with
2.6 Pa in partial pressure. The pressure value of oxygen
was set at the maximum to keep the continuous and sta-
ble sputtering ignition. Consequently, the nano-sized TiOx
compound particles were grown in the reaction-chamber.
3 Results and discussion
3.1 Crystal structure
Then, the flux of the particles was spouted to the adja-
cent vacuum vessel kept at 3 × 10−4 Pa through the aper-
ture with 2 mm in diameter at the end of the source and Figures 2a, 2b and 2c show the low magnification TEM im-
directly being deposited on a substrate at room tempera- age, HRTEM image, and SAED ring pattern of the sam-
ture apart from the aperture by 30 mm in soft-land mode. ple deposited under Ltd = 50 mm respectively. The
The substrates used were either copper mesh with micro film are identified as polycrystalline TiO from Figure 2c,
grid carbon film (hereafter called “micro grid”) or glass. where (111), (200) and (220) lattice plane of TiO with
To control the particle size of TiO, the growth reaction dis- NaCl crystal structure are indexed. From Figure 2a, the
tance between the sputtering target and the aperture (Ltd) deposited film is composed by many nano-crystallites
was set at 50 mm as an optimal condition and the wall TiO with the particle diameter of 5.62 nm averaged over
of the reaction-chamber and aperture were cooled down 225 particles. There are no particles with a distinct faceted
to LN2. The crystal structure, morphology, compositional shape. The particle diameter is distributed from 2 to 8 nm
uniformity, and chemical state, of the deposited particu- as shown in the histogram of Figure 3. A HRTEM lat-
lates thin film were characterized by TEM, SAED, EDS, tice image of Figure 2b is indexed as TiO (220). The high

S. Tanemura et al.: Fabrication and structural characterization of TiO nanoparticles
81
Table 1. Comparison between the lattice spacing appeared in
JCPDS card and that obtained from SAED ring pattern and
the lattice image in HRTEM.
˚
˚
A: JCPDS (A) B: Calculated (A) (A − B)/A (%)
TiO111
TiO200
TiO220
TiO311
TiO420
2.407
2.085
1.475
1.259
1.205
2.473
2.098
1.489
1.250
0.938
–2.7
–0.6
–0.9
–0.7
–0.4
Fig. 3. The distribu-
tion of the diameter of
the TiO particles de-
posited under Ltd
50 mm.
=
density and fine crystal quality of the fabricated nano par-
ticles by the present method are confirmed.
The lattice spacing values of TiO films obtained from
SAED and HRTEM lattice image agree satisfactorily
with those of bulk values [10] as shown in Table 1.
Consequently, the fabricated TiO particles and/or clusters
are grown without strain under the present experimental
conditions. This result strongly supports the feasibility of
our examined method for the fabrication of TiO thin film
with high crystal quality.
3.2 Compositional uniformity
The chemical composition of the obtained film was ana-
lyzed by EDS installed in the used JEOL 2010 FET TEM
Two kinds of EDS profiles were carried out: two spots and
line profiles shown as the bright field image of scanning
TEM (s-TEM) mode in Figure 4a. The representative spot
and line profile of EDS results are shown in Figures 4b,
and 4c respectively. The elements C, Ti, O, and Cu are
detected either in spot or line profiles. We can safely say
that Cu element come entirely from mesh of the micro
grid, while C comes mostly from micro grid substrate and
small portion of it is contamination from vacuum pump.
In the line profile, intensity profiles of Ti and O are quite
proportional and this confirms the compositional unifor-
mity of the deposited TiO thin film.
(a)
Figure 5 shows EELS of the deposited thin film. Ti-L2
and Ti-L3 peaks due to the core excitation from Ti 2p1/2
to 3d and from 2p2/3 to 3d as well as O-K peaks can
be clearly observed, although C-K peak is not appeared.
This strongly suggests the present film is composed of
Ti-O compound and coincides with SAED observation.
(b)
3.3 Chemical state
We measured Ti 2p core level of the deposited film by XPS
as shown in Figure 6 to investigate the chemical state. For
the comparison, XPS of the TiO2 thin film composed of
anatase type poly-crystals previously reported by Miao
et al. [11] is also cited. The observed Ti 2p peak values
of TiO are smaller than those of TiO2 by 1eV. The pre-
viously reported Ti 2p3/2 peak is 454.7 eV for the bulk
TiO sample prepared by high-temperature annealing at
(c)
Fig. 4. (a) Low-magnified bright field s-TEM image of the
deposited film giving the position of EDS profile, (b) a typical
spot EDS spectrum taken from position 1 indexed in (a), (c).
Line profile along the line from the edge of micro grid to the
inner part indicated in (a).
◦
1300–1500 C of materials which were initially prepared
by direct combination of high purity elements in Vycor
◦
glass or fused-silica tubes at temperature <700 C [12].

82
The European Physical Journal D
The averaged particle size increases up to 9.87 nm and
the size distribution expands from 4 to 14 nm without
LN2 coolant. This confirms the importance of the cooling
process during the growth of small particulates.
Consequently, the adopted experimental parameters;
oxygen partial pressure: 2.6 Pa, Ltd: 50 mm, and cooling
of growth chamber with LN2 coolant, are confirmed to be
optimal for the fabrication of the nano crystallites TiO.
Fig. 5. EELS of
the deposited film.
4 Conclusions
Using the magnetron sputtering-gas aggregation method,
we have succeeded to fabricate the thin film consisting
mainly or entirely of nano-crystallites TiO on the sub-
strate at room temperature in the soft land mode under
the optimal experimental parameters of partial pressure of
additional oxygen gas, growth reaction distance Ltd, and
LN2 cooling of growth chamber. The smallest averaged di-
ameter of TiO nano-particles is 5.62 nm and the diameter
distribution being from 2 to 8 nm under Ltd = 50 mm. The
compositional uniformity is confirmed by EDS and EELS.
The chemical state of the film from XPS of Ti 2p peak con-
firms the primarily growth of TiO in the film and indicates
the formation of minor content of amorphous Ti1−xOx
and/or Ti1−xCx particularly at the surface of the film.
This is the world first achievement to deposit TiO crys-
tallites on substrate at room temperature and the method
being prospect for the preparation of the layer material for
the ferroelectric device.
Fig. 6. Compari-
son between XPS
spectra of Ti 2p
from the depos-
ited TiO thin film
and that from the
polycrystalline an-
atase TiO2 film.
However, this reported value seems to be too low to refer,
when we take account the recent theoretically calculated
adiabatic electron affinities of TiO and TiO2 and TiO3
as 1.25, 1.60, and 3.34 eV respectively [13]. Those affinity
values support our presently observed shift in the binding
energy.
The shoulder observed at 455 eV suggests the possible
existing of amorphous T1−xOx and/or Ti1−xCx particu-
larly at the surface of the film. This topic is again discussed
in Section 3.4.
This work is partly supported by a grant from the NITECH
21st Century COE Program for “World Ceramics Centre for
Environmental Harmony”, and by “Nanotechnology Support
Project” of the Ministry of Education, Culture, Sports, Science
and Technology (MEXT), Japan.
References
1. C. Queniset, R. Naslain, P. Demoncy, Surf. Interf. Anal.
13, 130 (1988)
2. M.S. Ameen et al. MRS Sym. Proc. 200, 65 (1990)
3. I.N. Matrev, Vacuum 58, 327 (2000)
4. R. Ramesh et al. Appl. Phys. Lett. 61, 1537 (1992)
5. K.A. Grigorov, M. Stoyanova, I.N. Matrev, Bulg. J. Phys.
25, 177 (1998)
6. K.G. Grigorov, D. Bouchier, M. Stoyanova, G.I. Grigorov,
R. Caudano, R. Sporcen, C.R. Acad. Bulg. Sci. 50, 29
(1997)
7. H. Haberland, M. Karrais, M. Mall, Z. Phys. D 20, 413
(1991)
8. M. Gotoh, J. Murakami, Y. Tai, K. Yoshimura, K. Igarashi,
S. Tanemura, Z. Phys. D 40, 115 (1997)
9. M. Gotoh, J. Murakami, Y. Tai, K. Igarashi, S. Tanemura,
M. Kusunoki, Jpn J. Appl. Phys. 37, L1537 (1998)
10. JCPDS-International Centre for Diffraction Data,
PDF 08-0117
11. L. Miao et al. J. Cryst. Growth 260, 118 (2004)
12. H.F. Franzen, M.X. Umana, J.R. McCreary, R.J. Thorn,
J. Solid State Chem. 18, 363 (1976)
3.4 The influence of the experimental parameters
on TiO growth
In the case of zero oxygen partial pressure, the growth of
TiO crystallites is extremely reduced, whereas particles of
crystal TiC and amorphous Ti1−xOx and T1−xCx with
averaged diameter of nanometers size are characterized
by TEM, SAED, EDS, EELS and XPS in the deposited
films. This reveals that the addition of oxygen gas even at
low partial pressure is essential to enhance the oxidation
reaction of sputtered Ti element to TiO.
The average sizes of the particles increase up to
9.97 nm at Ltd = 100 mm and 10.28 nm at 150 mm
respectively, and the size distribution is becoming wider
than that observed for Ltd = 50 mm. Under Ltd
<
50 mm, amorphous Ti1−xOx is apparently fabricated, al-
though TiO crystallites with smaller averaged diameter
than 5.62 nm are observed. This implies that the critical
value of Ltd to deposit the film mainly or entirely consist-
ing of nano-crystallites TiO is 50 mm.
13. M.B. Walsh, R.A. King, H.F. Scaefer III, J. Chem. Phys.
110, 5224 (1999)