Home Search Collections Journals About Contact us My IOPscience

Capillary effects, wetting behavior and photo-induced tube filling of TiO2 nanotube layers

This article has been downloaded from IOPscience. Please scroll down to see the full text article.

2008 Nanotechnology 19 305710

(http://iopscience.iop.org/0957-4484/19/30/305710)

View the table of contents for this issue, or go to the journal homepage for more

Download details:
IP Address: 137.189.43.142
The article was downloaded on 14/09/2010 at 09:17

Please note that terms and conditions apply.

IOP PUBLISHING NANOTECHNOLOGY
Nanotechnology 19 (2008) 305710 (5pp) doi:10.1088/0957-4484/19/30/305710

Capillary effects, wetting behavior

and photo-induced tube filling of
TiO2 nanotube layers
Doohun Kim, Jan M Macak, Felix Schimidt-Stein and
Patrik Schmuki
Department of Materials Science, WW4-LKO, University of Erlangen-Nuremberg,
Martensstraße 7, D-91058 Erlangen, Germany

E-mail: schmuki@ww.uni-erlangen.de

Received 19 March 2008, in final form 21 May 2008

Published 16 June 2008
Online at stacks.iop.org/Nano/19/305710

Abstract
The present work addresses effects associated with the electrolyte penetration kinetics into TiO2
nanotube layers. In particular, it is shown that the electrolyte uptake kinetics affects the
magnitude of the measured photoresponse. We demonstrate that for aqueous electrolytes the
penetration of the electrolyte into a TiO2 nanotubular layer is comparably slow and may take up
to several hours. The electrolyte uptake kinetics can significantly be accelerated by UV
illumination. We ascribe this to a light-induced change in the wetting properties on the inside of
the TiO2 nanotube surface. This effect can be exploited to achieve photo-induced filling of the
nanotubes by a secondary material.
(Some figures in this article are in colour only in the electronic version)

1. Introduction UV-treated TiO2 surface is hydrophilic [16]. Wang et al

TiO2 is a widely used functional material due to its
semiconductive properties, chemical and wear stability, non-
toxicity and low-cost preparation [1]. Recently, a novel
approach to form self-organized and highly ordered TiO2
nanotube layers based on the anodization of Ti in F-ion-
containing electrolytes has been reported [2–9]. Since then,
many efforts have been carried out to tailor the dimension of
the nanotubes, to improve ordering and to exploit them for
functional applications—an overview is given in [10].
A key parameter for efficient use of the nanotubes is
their wetting behavior. A good access of electrolyte into
the tubes is crucial for the effectiveness of many applications
such as in photocatalysis [11, 12] and tube filling [13] or
for in situ characterization such as photoresponse [14] or
impedance measurements [15]. In particular, to incorporate
specific properties to the inside of a tube by a secondary
material, for example, to add p–n semiconductor junctions,
magnetic properties or use the tube as a drug carrier, ‘good’
wetting properties are essential.
The wetting behavior of TiO2 can be influenced by light
irradiation. Among other approaches, it was found that a
reported a surface wettability switching on TiO2 surfaces by
alternating ultraviolet (UV) irradiation and dark storage. This
phenomenon was elucidated as the formation and extinction
of surface hydroxyl groups [16, 17]. Controlling and/or
modifying the surface wettability of solid substrates are
important in many cases from the scientific and technological
viewpoints. Several commercial products have been designed
and produced for practical uses such as exploiting UV-
illuminated TiO2 for antifogging and self-cleaning effects
on the basis of photo-induced surface wettability conversion
phenomena [18, 19]. Up to now, there have been a number
of reports on wetting properties of nanoparticle layers and
flat TiO2 . The wetting behavior of TiO2 nanotubes has only
scarcely been explored.
In recent years, we have reported that the surface wetting
properties of a nanotube layer significantly depend on nanotube
diameter and can be strongly altered using organic monolayer
coatings combined with UV light [20, 21]. However, the
wetting properties in view of electrolyte uptake were not
studied. In the present work, we study the effect of
electrolyte uptake into tubes characterized by photoresponse
measurements as well as measured by the tube filling

0957-4484/08/305710+05$30.00 1 © 2008 IOP Publishing Ltd Printed in the UK

Nanotechnology 19 (2008) 305710 D Kim et al

Figure 1. SEM top and cross-sectional view for a TiO2 nanotubular layer formed in 1 M H2 SO4 + 0.16 M HF at 20 V during 2 h after the
potential ramp from OCP to 20 V (sweep rate 500 mV s−1 ).

characteristics. Furthermore we explore light-induced effects

on the tube filling kinetics.

2. Experimental details

Titanium foils (0.1 mm thickness, 99.6% purity, Goodfellow,

UK) were degreased by sonicating in acetone, isopropanol
and methanol, rinsed with deionized water (DI) and dried in
a nitrogen stream. TiO2 nanotube layers were prepared by
electrochemical anodization using a high voltage potentiostat
Jaissle IMP 88 PC and a conventional three-electrode
configuration with a platinum gauge as a counter electrode and
a Haber–Luggin capillary with Ag/AgCl (1 M KCl) reference
electrode was used for the anodization.
The electrolytes were 1 M H2 SO4 + 0.16 M HF for
the growth of TiO2 nanotube layers which are 500 nm in
thickness [3]. Anodization was carried out at 20 V for 2 h, after
the potential was ramped from open circuit potential (OCP) to
20 V at a sweep rate of 500 mV s−1 . To grow nanotube layers
of a greater thickness, a glycerol + water (50:50 vol%) +
0.27 M NH4 F mixture was used as an electrolyte. Anodization
was carried out at 30 V for 3 h at a sweep rate of
250 mV s−1 [22]. To obtain a compact oxide TiO2 film,
anodization was carried out using the same set-up at 20 V for
3 min at a sweep rate of 100 mV s−1 in 1 M H2 SO4 electrolyte.
Then the samples were rinsed in DI water and dried in a
nitrogen stream. All the experiments were carried out at room
temperature.
To convert the nanotube layers into anatase, the samples
were annealed at 450 ◦ C for 1 h at the heating and cooling
rate of 30 ◦ C min−1 using a Rapid Thermal Annealer-Jipelec
JetFirst 100 [14]. After annealing, the samples were left at
room temperature in the dark for 48 h prior to any further
experiment.
For morphological characterization of the samples, a
field-emission scanning electron microscope (Hitachi FE-SEM
S4800) was used.
Photoelectrochemical characterization was carried out
with a set-up consisting of a 150 W Xe arc lamp (LOT-
Oriel Instruments) as an irradiation source and a Cornerstone
Motorized 1/8m Monochromator. Phototransient spectra were
acquired at 350 nm wavelength (UV intensity at 350 nm;
2.84 mW cm−2 ) with a 20 s light pulse in 0.1 M Na2 SO4 at
a potential of 500 mV (Ag/AgCl).
Figure 2. Change of the photocurrent magnitude for annealed TiO2
nanotubular layers with time (soaked in the dark and soaked under
illumination) and TiO2 compact oxide layer. Two insets show
photocurrent transients of the samples after 2 min soaking and
24 h soaking in the dark.
To obtain information on the penetration depth of aqueous
solutions, we used a lead marker and SEM observation. For
this, nanotube samples were soaked for different times in 1 M
Pb(NO3 )2 both with and without illumination. The samples
then were dried and the Pb penetration front was observed from
SEM images. The light source for illumination was a high
power Hg lamp (UV intensity; 15 W cm−2 ).
3. Results and discussion
Figure 1 shows scanning electron microscopy (SEM) images of
the top and cross-section of typical TiO2 nanotube layers used
for most of the present investigations. They consist of nanotube
arrays of approximately 100 nm tube diameter, 15 nm wall
thickness and 500 nm tube length (before and after annealing
to anatase).
With these samples, we performed electrolyte uptake
experiments. First we studied the photoresponse magnitude
with soaking time by measuring photocurrent transients at
different exposure times. To measure always under defined
conditions, all samples were left in the dark at room
temperature for 48 h prior to photoresponse measurements.
The phototransient measurements were carried out at a

2
Nanotechnology 19 (2008) 305710 D Kim et al

Figure 3. SEM top views and cross-sections used for Pb marker experiments. Tubes after soaking in the dark in 1 M Pb(NO3 )2 10 min
((a), (b)), for 24 h ((c), (d)) and after soaking under illumination with a high power Hg lamp for 10 min ((e), (f)) and for 1 h ((g), (h)) in 1 M
Pb(NO3 )2 . Evaluation of the Pb penetration depth obtained from SEM images for 500 nm tube length (i).

wavelength of 350 nm. The results are shown in figure 2—
the two insets show typical photocurrent transients for the
annealed tubular layers. The shapes of these photocurrent
transients remained similar over the entire set of experiments
and show the typical recombination characteristics observed
for annealed TiO2 nanotubes [14]. The value of the
photocurrent was taken at the steady state after 20 s (for
example, this results after soaking for 2 min in a steady
state photocurrent of 7.6 µA). The results in figure 2
show the magnitude of the photocurrent as a function of

3
Nanotechnology 19 (2008) 305710 D Kim et al

the soaking time. This is shown for nanotube samples that

were exposed to light only during the photocurrent transient
measurements (a); for samples that were continuously exposed
to UV light (and in between transients were acquired) (b); and
as a reference for a compact TiO2 (anatase) layer (c). For the
nanotube layers in the dark, (a), the photoresponse increases
by approximately 42% compared with the initial value within
the first 24 h and then levels off at a steady magnitude. The flat
TiO2 sample shows over this time a constant photoresponse—
this indicates that the changes observed for the tubes cannot
be attributed to alterations in the TiO2 layer (for example,
structural changes or surface modification). Therefore the
findings suggest that this saturation effect in the photoresponse
is associated with a slow electrolyte uptake into the nanotube
layer due to the fact that the bottom of TiO2 nanotubes are
closed and thus the TiO2 nano-‘capillary’ acts as if one side
were closed. In order to explore if surface wettability can
be influenced by UV light [16–19] and can be exploited to
achieve faster tube filling, we performed experiments under
permanent UV light illumination. As shown in figure 2(b), the
UV-illuminated tubes show immediately a significantly higher
photoresponse and already only after 1 h it reaches steady
state. This strongly suggests that illumination alters the TiO2
nanotube inner surface to a more hydrophilic state, allowing
the capillary forces to fill the tubes.
To additionally demonstrate the wetting behavior and
the capillary filling effect of TiO2 nanotubes, we used a
Pb marker and we estimated the penetration depth directly
from cross-sectional SEM images. For this, TiO2 nanotubes
were soaked for different times in 1 M Pb(NO3 )2 with and
without illumination. The samples then were dried and the Pb Figure 4. (a) Penetration depth obtained from SEM images for
penetration front was observed from SEM images. In the case 1.8 µm long TiO2 nanotubes using Pb markers as in figure 3. The
inset shows a SEM cross-sectional view of the nanotubes that were
of soaking for 10 min, as shown in figures 3(a) and (b), most soaked in the dark for 24 h in 1 M Pb(NO3 )2 (as an example for the
of the soaked tubes are empty and only the top of the tubes evaluation of penetration depth). (b) EDX–intensity for different
were deposited with a little amount of lead. Obviously, the filling times corresponding to (a).
Pb(NO3 )2 solution could not sufficiently penetrate inside the
tubes within the given time. For samples soaked in the dark
for 24 h (figures 3(c) and (d)) the nanotubes are filled with lead with hydroxyl groups (–OH) on the surface, which causes
to about 70% of the nanotube length. In the case of the soaked the penetration velocity of aqueous electrolytes into the TiO2
samples under UV illumination for 10 min (figures 3(e) and (f)) nanotubes to accelerate.
and 1 h (figures 3(g) and 1(h)), they are clearly filled within a It is noteworthy that nanotubes with much smaller
considerably shorter time. As shown in figures 3(g) and (h), diameters can be produced [22, 23]—however, the lower limit
after 1 h illumination the TiO2 nanotubes are completely filled is currently around a diameter of 15 nm; thus still not entirely
with the lead marker. sufficient to achieve quantum size effects.
As an additional example of the tube filling with a different
thickness of nanotube layer, TiO2 nanotube layers of 1.8 µm 4. Conclusions
thickness were produced in a glycerol + water (50:50 vol%) +
0.27 M NH4 F mixture electrolyte. As shown in figure 4, after In the present work, the wetting behavior of TiO2 nanotube
24 h soaking in the dark, the tubes were not completely filled layers was investigated using the effect of tube electrolyte
with the lead marker. The inset SEM image clearly shows filling on the photoelectrochemical response. The results
the contrast for the Pb front (figure 4(a)). For the illuminated strongly suggest that changes of photoresponse with soaking
tube samples, after 1 h, the nanotubes are completely filled. time are due to a slow filling of a ‘dark’ tube. UV illumination
Figure 4(b) shows a corresponding EDX analysis performed of the nanotube layer drastically affects the wettability and
on the different samples. Clearly the intensity of the Pb is a highly effective treatment to achieve rapid filling of an
signal increases in line with the results of figure 4(a). It is aqueous electrolyte into the TiO2 nanotubes. This effect can
therefore evident from these results that illumination by UV be exploited to achieve photo-induced filling of the tubes with
light strongly accelerates the tube filling. a secondary material—thus facilitating applications such as
It may be suggested that by UV illumination the chemical p-type material deposition for more efficient solar cells or
termination of the wall altered to a more hydrophilic state incorporation of drugs for biomedical delivery purposes.

4
Nanotechnology 19 (2008) 305710
Acknowledgments
The authors would like to acknowledge the DFG for financial
support, and Hans Rollig and Martin Kolacyak for valuable
technical help.
References
[1] Fujishima A and Honda K 1972 Nature 238 37
[2] Zwilling V, Darque-Ceretti E, Boutry-Forveille A, David D,
Perrin M Y and Aucouturier M 1999 Surf. Interface Anal.
27 629
[3] Beranek R, Hildebrand H and Schmuki P 2003 Electrochem.
Solid-State Lett. 6 B12
[4] Macak J M, Sirotna K and Schmuki P 2005 Electrochim. Acta
50 3679
[5] Macak J M, Tsuchiya H and Schmuki P 2005 Angew. Chem.
44 2100
[6] Ghicov A, Tsuchiya H, Macak J M and Schmuki P 2005
Electrochem. Commun. 7 505
[7] Macak J M, Tsuchiya H, Taveira L, Aldabergerova S and
Schmuki P 2005 Angew. Chem. Int. Edn 44 7463
[8] Albu S P, Ghicov A, Macak J M and Schmuki P 2007
Phys. Status Solidi 1 R65
[9] Raja K S, Misra M and Paramguru K 2005 Electrochim. Acta
51 154
[10] Macak J M et al 2007 Curr. Opin. Solid State Mater. Sci.
doi:10.1016/j.cossms.2007.08.004
D Kim et al
[11] Macak J M, Zlamal M, Krysa J and Schmuki P 2007 Small
3 300
[12] Albu S P, Ghicov A, Macak J M, Hahn R and Schmuki P 2007
Nano Lett. 7 1286
[13] Macak J M, Gong B G, Hueppe M and Schmuki P 2007
Adv. Mater. 19 3027
[14] Beranek R, Tsuchiya H, Sugishima T, Macak J M, Taveira L,
Fujimoto S, Kisch H and Schmuki P 2005 Appl. Phys. Lett.
87 243114
[15] Muñoz A G, Chen Q and Schmuki P 2007 J. Solid State
Electrochem. 11 1077
[16] Wang R, Hashimoto K, Fujishima A, Chikuni M, Kojima E,
Kitamura A, Shimohigoshi M and Watanabe T 1997
Nature 388 431
[17] Wang R, Sakai N, Fujishima A, Watanabe T and
Hashimoto K 1999 J. Phys. Chem. 103 2188
[18] Sakai N, Wang R, Fujishima A, Watanabe T and
Hashimoto K 1998 Langmuir 14 5918
[19] Fujishima A, Hashimoto K and Watanabe T 1999 TiO2
Photocatalysis: Fundamentals and Applications
(Tokyo: BKC Inc.)
[20] Balaur E, Macak J M, Taviera L, Tsuchiya H and
Schmuki P 2005 Electrochem. Commun. 7 1066
[21] Balaur E, Macak J M, Tsuchiya H and Schmuki P 2005
J. Mater. Chem. 15 4488
[22] Macak J M, Tsuchiya H, Ghicov A, Yasuda K, Hahn R,
Bauer S and Schmuki P 2007 Curr. Opin. Solid State Mater.
Sci. 11 3
[23] Bauer S, Kleber S and Schmuki P 2006 Electrochem. Commun.
8 1321