All these anodic TiO2nanotubes have already been used for a wide range of different applications, such as those based on the semiconductive nature of TiO2.21 Namely, dye-sensitization,22
Trang 1Multilayer TiO2–Nanotube Formation by Two-Step Anodization
J M Macak,*S Albu, D H Kim, I Paramasivam,
S Aldabergerova, and P Schmuki**,z
Department of Materials Science, University of Erlangen–Nuremberg, D-91058 Erlangen, Germany
In this work we report on the growth of a closely stacked double layer of a self-organized TiO2nanotubes For that we first anodize
Ti in acidic electrolyte containing hydrofluoric acid to form thin nanotube layers Afterwards we start a second anodization in a
different electrolyte, a glycerol/NH4F mixture This procedure allows us to grow the second layer directly underneath the first one.
From scanning electron microscopy and transmission electron microscopy investigations we revealed that the second growth
occurs via the tube bottoms of the first layer These stacked multilayers generate new possibilities to vertically tailor the properties
of the self-organized TiO2nanotube layers.
© 2007 The Electrochemical Society 关DOI: 10.1149/1.2737544兴 All rights reserved.
Manuscript submitted February 5, 2007; revised manuscript received April 8, 2007 Available electronically May 15, 2007.
Formation of porous alumina based on anodic oxidation of
alu-minium has been investigated and well understood already for many
years.1,2 But only about one decade ago, extremely ordered and
self-organized porous alumina structures could be formed using a set
of specific electrochemical conditions using optimized potential,
temperature, electrolyte composition, etc.3-7In 1999, Zwilling et al.8
showed that Ti can also be converted to highly ordered nanotubes
共in contrast to alumina nanopores兲 using self-assembly during
an-odic oxidation Since then there have been many efforts to tailor the
morphology of the TiO2 nanotube toward enabling potential
applications.9-13Later on, other valve metals have shown the ability
to form nanotubes.14-16In all these works, fluoride-anion-containing
electrolytes were used to selectively dissolve the anodized metals
under anodic bias applied for several hours, leaving nanotubular
layers on their surfaces Although there is still very little work done
on Ti and other metals compared to Al, reports showing significant
improvements in length9-13 and tube diameter17,18 have been
re-cently published by our group as well as by others.19,20Typically,
the diameter of tubes is controlled by the applied anodization
voltage17and various lengths can be obtained using different
elec-trolytes All these anodic TiO2nanotubes have already been used for
a wide range of different applications, such as those based on the
semiconductive nature of TiO2.21 Namely, dye-sensitization,22
doping,23-25 photocatalysis,26 electrochromism,27 and photolysis28
have been demonstrated, as well as others based on catalysis29 or
sensing.30Due to high biocompatibility of TiO2, other reports
tar-geted growth of a hydroxyapatite layer on the nanotubes31as well as
their formation on Ti alloys.32,33 It has also been shown that the
structure of the anodized tubes is always amorphous and can be
converted by annealing to anatase21,34,35 or, e.g., BaTiO336 or
Ba共Sr兲TiO337upon hydrothermal alkali treatment Additionally, by
electrochemical deposition into the tubes,38properties such as the
magnetic behavior of nanotube layers can be modified.39
In the present work we show that even multistacks of TiO2
nano-tubes can be grown directly by a two-step anodization process
Experimental
Titanium foils共0.1 mm, 99.6% purity, Advent Materials兲 were
degreased by sonication in acetone, isopropanol, and methanol prior
to electrochemical experiments, afterward rinsed with deionized
共DI兲 water, and finally dried in nitrogen stream The samples were
pressed together with a Cu plate against an O-ring in an
electro-chemical cell 共1 cm2 exposed to the electrolyte兲 and anodized at
20 V in 1 M H2SO4electrolytes containing hydrofluoric acid共HF兲
共0.16 M兲 for 2 h to grow a 500 nm thick TiO2nanotube layer After
this the nanotube layers were rinsed and dried and a second anod-ization step was performed in glycerol electrolytes containing NH4F 共0.27 M兲 at 20 V for several hours In some cases 共to achieve longer tubes兲, substrates for the first layer were grown in 共NH4兲2SO4/NH4F mixtures For the electrochemical experiments, a high-voltage po-tentiostat Jaissle IMP 88 and a conventional three-electrode configu-ration with a platinum gauze as a counter electrode and a Haber– Luggin capillary with Ag/AgCl共1 M KCl兲 as a reference electrode were used All electrolytes were prepared from reagent-grade chemi-cals Selected samples were sonicated in the ultrasonic bath共power output 100 W兲 A scanning electron microscope 共Hitachi FE-SEM S4800兲, a transmission electron microscope 共CM 30 T/STEM兲, and
a high-resolution transmission electron microscopy共HRTEM兲 Phil-ips CM 300 UT were used for the morphological and structural characterization of the TiO2nanotubular layers
Results and Discussion
Figures 1A-1C show examples of two types of arrays of self-organized TiO2 nanotubes used in this study and a double-layer structure consisting of these two types The first type of layer共layer
1, Fig 1A兲 consists of 500 nm long tubes formed in a mixture of
1 M H2SO4and 0.16 M HF at 20 V during 2 h with a diameter of approx 100 nm.40The second type of layer共layer 2, Fig 1B兲 con-sists of longer tubes formed in glycerol electrolyte containing 0.27 M NH4F at 20 V To form the multilayer structure of Fig 1C, the growth of the two different types of tubes is combined in a two-step anodization process For this, a titanium sheet is anodized
to make the short tubes as described above Then, after rinsing and drying, the sample is anodized in the glycerol electrolyte at 20 V From Fig 1C it is clear that the dimensions of the second layer are the same as for nanotubes formed directly in this electrolyte.12 Fig-ure 1D shows polarization curves共recorded with a sweep rate of 0.1 V/s兲 and current transients recorded for the first and second layer growth For the first anodization in HF electrolytes leading to short tubes 共Fig 1A兲, the current reaches a maximum during the potential sweeping; this situation is typical for a self-organization sequence involving the formation of a compact oxide and irregular pores prior to ordered pore formation.10,40For the second anodiza-tion in glycerol soluanodiza-tions leading to long and smooth tubes, the same sequence occurs, but the tube self-organization kinetics is much slower and therefore the drop in the current appears in the anodiza-tion process later, in this case in the potentiostatic phase of the anodization process.41At the end of the anodization experiments, stirring of the electrolytes was applied for 15 min The resulting current behavior is shown as an inset in Fig 1D Just after the beginning of the stirring共angular speed ⬃150 rpm兲, the currents in both cases started to increase, and as soon as the stirring was termi-nated, they returned to their original values This can be ascribed to the fact that the steady-state current density is diffusion limited The
* Electrochemical Society Student Member.
** Electrochemical Society Active Member.
z
E-mail: schmuki@ww.uni-erlangen.de
Electrochemical and Solid-State Letters, 10共7兲 K28-K31 共2007兲
1099-0062/2007/10共7兲/K28/4/$20.00 © The Electrochemical Society
K28
Trang 2reason for the smaller increase in the glycerol electrolyte can be
ascribed to a significantly longer diffusion path共longer tubes兲 and to
the high viscosity of the electrolyte
In order to evaluate the growth of the second layer after the first
layer has been formed, we performed detailed scanning electron
microscopy共SEM兲 investigations of the interfaces between the
lay-ers Figure 2 shows a scheme of the formation of second-layer
growth, which is based on the SEM observation shown in Fig 3 We
first grow an initial nanotube layer in the acidic electrolyte共Fig 2a兲
Afterwards we clean and dry it, immerse it in the glycerol
electro-lyte, and start the second anodization process After several minutes,
there are very small channels or holes etched in the bottoms of the
first tube layer共Fig 2b兲 The etching takes place preferentially at the
bottom of the tubes From the images in Fig 3A and 3B that were
taken from sample anodized for about 20 min in the glycerol
elec-trolyte共to form the second layer兲, it is evident that the etching front
penetrates the bottoms and new and somewhat irregularly
distrib-uted tubes are formed 共Fig 2c兲 The new tubes are at this very
moment competing for available space and current Growth of some
tubes is terminated after a while, because there is not sufficient
space available After about 40 min, the newly formed tubes are
already self-organized共Fig 2d兲, with only small variations in
diam-eter共40 ± 10 nm兲 as shown also in Fig 3C From the images in Fig
3A and 3B it can be seen that the width of the channels drawn in
Fig 2c is in the range between 15 and 25 nm The number of
chan-nels is typically between 3 and 5 and the number of the newly
formed tubes from the first layer 共one bigger tube兲 is 3 or 4, as
shown also in Fig 2d From Fig 1A it can be seen that the original
outer diameter of the tubes 共layer 1兲 was about 130 nm and the
newly formed tubes共layer 2兲 in Fig 1B have outer diameters of
Figure 1 SEM top-view images of the tubes used for the growth of共A兲 the first and 共B兲 the second layer and 共C兲 their interface after two-step anodization The first tubes are 500 nm long with a diameter of 100 ± 10 nm, and the second tubes have a diameter of 40 ± 10 nm and length dependent on the anodization time; 共D兲 polarization curves 共left part, sweep rate 0.1 V/s兲 and current transients 共right part兲 recorded for anodization of Ti sample at 20 V during the first layer growth 共in acidic electrolyte兲 and during the second layer growth 共in glycerol electrolyte兲 共Insets兲 Magnification of the current transients during electrolyte strirring introduced for 15 min 共angular speed ⬃150 rpm兲.
Figure 2 Schematic drawing demonstrating four steps in the formation of
the second tubular layer during the second anodization step.
K29
Trang 350 ± 10 nm Using a longer tube geometry共more than 500 nm兲 as
the upper nanotube layer is possible, but wetting of such tubes after
drying becomes a more severe issue Remarkably, there is no
sig-nificant chemical dissolution apparent of either the first layer or the
second layer One can expect that the electrolyte also filled the space
between the nanotubes; however, it seems that the gaps of the first tube layer not the most reactive sites under present conditions, as we did not observe any nanotube growth there
To obtain more insight into the formation of the second layer, we performed some additional characterization by TEM Figure 4 shows TEM images of the second-layer nanotubes共A兲 and a single tube共B兲 of the first layer, with several channels at the bottom de-scribed and shown in Fig 2 and 3 From Fig 4A one can clearly see that the second-layer tubes are smooth and without ripples on the walls Furthermore, the presence of open channels for mass and current flow from the first to the second layer is confirmed by Fig 4B Figure 4C shows a high-resolution TEM image of the tube bot-tom of the second layer Clearly, an amorphous structure is present
as confirmed by selected area diffraction pattern共SAED兲, shown in the inset
Figure 5 shows the influence of anodization time on the thickness
of the second layer After 10 min of anodization there are only short
Figure 3 SEM images of the interface between both TiO2nanotubular
lay-ers 共A兲 in the top- and 共B兲 in the cross-sectional view showing channels at
the bottoms of the first layer; 共C兲 comparison between the tube outer
diam-eters during the transition from one 共outer diameter approximately 130 nm兲
to the other type 共approx 70 nm兲 Space occupied originally by one tube is
typically used for three tubes of the second layer.
Figure 4.共A兲 TEM images of bundles of smooth TiO 2 nanotubes from the
second layer; 共B兲 single tube depicted from the top showing etched bottom
with channels; 共C兲 HRTEM image of the tube bottom of the second type
showing amorphous structure Inset in c shows SAED with diffuse rings
matching amorphous structure.
Figure 5 Dependence of the second-layer thickness on the anodization time
during the second anodization step shown 共a兲 as a sequence of SEM cross-sectional images and 共b兲 as a plot Linear growth with constant of 300 nm/h
is achieved within first 12 h.
K30
Trang 4tubes; no visible alteration of the tube bottoms has occurred After
approximately 30 min of anodization there is already a second tube
layer formed with a thickness of about 100 nm The tubes keep
growing and the tube layer thickness increases with time as shown
in Fig 5b In the early stages共first 12 h兲, the thickness increases
linearly, and for longer times a deviation to an apparently slower
growth rate is observed After 24 h of anodization the second tube
layer thickness is reaching 5m From the cross-sectional SEM
images we can estimate the growth rate in the early stages of
ap-proximately 50 nm per 10 min共300 nm per h兲 This means that the
growth of nanotubes with the same length is somewhat slower
com-pared to tubes that were grown directly,12,41i.e., without the
pres-ence of the upper tube layer For example, in our previous work we
were able to grow 7m tubes within 13 h under the same
electro-chemical conditions as used here.12This can be described to
ham-pered diffusion of the chemical species through the narrow channels
between the first and second layer Multilayer structures, as formed
here, may have significant applications, for example, in
size-selective reactive filtration or Bragg-stack structures.42,43
Conclusions
The results of the present work show that multilayer stacks of
highly ordered and self-organized TiO2 nanotubular layers can be
grown by two-step anodization under different electrochemical
con-ditions From SEM evaluation we revealed that the growth of tubes
of the second layer starts at the bottom of the first tubes by narrow
channels being formed in the early stages Further, we show that the
thickness of these second layers can be in the range of several
mi-crometers, depending on anodization time Clearly, the diameter of
the tubes corresponds to the formation conditions of the individual
nanotube layers The growth of the second layer in length is
some-what slower than for its isolated formation The feasibility to form
two distinct layers may be exploited in view of optical properties or
for size-selective reactions
Acknowledgments
The authors acknowledge DFG for financial support Hans Rollig
and Martin Kolacyak are acknowledged for valuable technical help
University of Erlangen assisted in meeting the publication costs of this
article.
References
1 F Keller, M S Huntler, and D L Robinson, J Electrochem Soc., 100, 411
共1953兲.
2 J W Diggle, T C Downie, and C W Goulding, Chem Rev (Washington, D.C.),
69, 365共1969兲.
3 H Masuda and K Fukuda, Science, 268, 1466共1995兲.
4 G E Thompson, Thin Solid Films, 297, 192共1997兲.
5 O Jessensky, F Müller, and U Gösele, J Electrochem Soc., 145, 3735共1998兲.
6 S Ono, M Saito, and H Asoh, Electrochim Acta, 51, 827共2005兲.
7 W Lee, R Ji, U Gösele, and K Nielsch, Nat Mater., 5, 741共2006兲.
8 V Zwilling, E Darque-Ceretti, and A Boutry-Forveille, Electrochim Acta, 45,
921 共1999兲.
9 J M Macak, K Sirotna, and P Schmuki, Electrochim Acta, 50, 3679共2005兲.
10 L Taveira, J M Macak, H Tsuchiya, L F P Dick, and P Schmuki, J
Electro-chem Soc., 152, B405共2005兲.
11 J M Macak, H Tsuchiya, and P Schmuki, Angew Chem., Int Ed., 44, 2100
共2005兲.
12 J M Macak, H Tsuchiya, L Taveira, S Aldabergerova, and P Schmuki, Angew.
Chem., Int Ed., 44, 7463共2005兲.
13 S Albu, A Ghicov, J M Macak, and P Schmuki, Phys Status Solidi (RRL), 1,
R65 共2007兲.
14 H Tsuchiya, J M Macak, I Sieber, and P Schmuki, Small, 1, 722共2005兲.
15 H Tsuchiya and P Schmuki, Electrochem Commun., 7, 49共2005兲.
16 I Sieber, B Kannan, and P Schmuki, Electrochem Solid-State Lett., 8, J10
共2005兲.
17 S Bauer, S Kleber, and P Schmuki, Electrochem Commun., 8, 1321共2006兲.
18 H Tsuchiya, J M Macak, L Taveira, E Balaur, A Ghicov, K Sirotna, and P.
Schmuki, Electrochem Commun., 7, 576共2005兲.
19 Q Cai, M Paulose, O K Varghese, and C A Grimes, J Mater Res., 20, 230
共2005兲.
20 K S Raja, M Misra, and K Paramguru, Electrochim Acta, 51, 154共2005兲.
21 R Beranek, H Tsuchiya, T Sugishima, J M Macak, L Taveira, S Fujimoto, H.
Kisch, and P Schmuki, Appl Phys Lett., 87, 243114共2005兲.
22 J M Macak, H Tsuchiya, A Ghicov, and P Schmuki, Electrochem Commun., 7,
1138 共2005兲.
23 A Ghicov, J M Macak, H Tsuchiya, J Kunze, V Heaublein, L Frey, and P.
Schmuki, Nano Lett., 6, 1080共2006兲.
24 J M Macak, A Ghicov, R Hahn, H Tsuchiya, and P Schmuki, J Mater Res., 21,
2824 共2006兲.
25 J H Park, S Kim, and A J Bard, Nano Lett., 6, 24共2006兲.
26 J M Macak, M Zlamal, J Krysa, and P Schmuki, Small, 3, 303共2007兲.
27 A Ghicov, H Tsuchiya, R Hahn, J M Macak, and A G Munoz, Electrochem.
28 O K Varghese, M Paulose, K Shankar, G K Mor, and C A Grimes, J Nanosci.
Nanotechnol., 5, 1158共2005兲.
29 J M Macak et al., Electrochem Commun., 7, 1417共2005兲.
30 O K Varghese, D Gong, K G Ong, and C A Grimes, Sens Actuators B, 93, 338
共2003兲.
31 H Tsuchiya, J M Macak, L Muller, J Kunze, F Muller, S P Greil, S Virtanen,
and P Schmuki, J Biomed Mater Res., 77A, 534共2006兲.
32 H Tsuchiya, J M Macak, A Ghicov, and P Schmuki, Small, 2, 888共2006兲.
33 J M Macak, H Tsuchiya, L Taveira, A Ghicov, and P Schmuki, J Biomed.
Mater Res., 75A, 928共2005兲.
34 O K Varghese, D Gong, M Paulose, C A Grimes, and E Dickey, J Mater Res.,
18, 156共2003兲.
35 J M Macak, S Aldabergerova, A Ghicov, and P Schmuki, Phys Status Solidi A,
203, R67共2006兲.
36 N T Padture and X Wei, J Am Ceram Soc., 86, 2215共2003兲.
37 X Wei, A L Vasiliev, and N P Padture, J Mater Res., 20, 2140共2005兲.
38 J M Macak, B G Gong, M Hueppe, and P Schmuki, Adv Mater (Weinheim,
Ger.), Accepted.
39 V M Prida, M Hernández-Veléz, K R Pirota, A Menéndez, and M Vasquéz,
Nanotechnology, 16, 2696共2005兲.
40 R Beranek, H Hildebrand, and P Schmuki, Electrochem Solid-State Lett., 6, B12
共2003兲.
41 J M Macak and P Schmuki, Electrochim Acta, 52, 1258共2006兲.
42 H Tsuchiya, M Hueppe, T Djenizian, and P Schmuki, Surf Sci., 547, 268共2003兲.
43 G Vincent, Appl Phys Lett., 64, 2367共1994兲.
K31