Layer-by-Layer Nanotube Template Synthesis

Article abstract of DOI:10.1021/ja049537t

Electroless deposition of gold on the pore walls of polycarbonate templates is currently the best known method for controlling inside diameters of template-synthesized nanotubes. It would be very useful to have alternative template-based synthetic chemistries that yield nanotubes composed of other materials, but which still allow for precise control over the nanotube wall thickness and i.d. A film-formation process that is based on layer-by-layer deposition of the film-forming material along the pore walls of the template membrane provides this desired alternative synthetic chemistry. We describe here the use of Mallouk's α,ω-diorganophosphonate/Zr layer-by-layer film-forming method for preparing nanotubes within the pores of alumina template membranes. We have found that this method allows accurate, quantitative, and predictable control over the wall thickness, and thus i.d., of the layered nanotubes obtained. Copyright

Full text of DOI:10.1021/ja049537t

Published on Web 04/20/2004
Layer-by-Layer Nanotube Template Synthesis
Shifeng Hou, C. Chad Harrell, Lacramioara Trofin, Punit Kohli, and Charles R. Martin*
Departments of Chemistry and Anesthesiology and Center for Research at the Bio/Nano Interface,
UniVersity of Florida, GainesVille, Florida 32611-7200
Received January 26, 2004; E-mail: crmartin@chem.ufl.edu
There is increasing interest in template-synthesized nanotubes.
Recent examples include templated carbon nanotubes for electro-
1
chemical energy production, phospholipid nanotubes for biochip
2
and biosensor applications, silica nanotubes for biocatalysis and
3
biorecognition, and nanotube-containing membranes for biosepa-
4
5
rations. An important feature of the template synthesis method
for all of these applications is the ability to control the dimensions
of the nanotubes obtained. The outside diameter of the nanotubes
is determined by the diameter of the pores in the template, and the
length of the nanotubes is determined by the thickness of the
template. It is, however, more difficult to control the inside diameter
(i.d.), or correspondingly the wall thickness, of template-synthesized
nanotubes, and, for some applications, precisely controlling the i.d.
is absolutely essential. Nanotube membranes for bioseparations are
a good example, because it has been shown that the transport
selectivity observed is profoundly influenced by nanotube i.d.4,6
Electroless deposition of gold on the pore walls of polycarbonate
templates is currently the best known method for controlling
nanotube i.d.6 However, this method is limited to nanotubes
composed of Au or other metals. It would be useful to have
alternative template-synthetic chemistries that yield nanotubes
composed of other materials, but which still allow for precise control
over the nanotube wall thickness and i.d.
Figure 1. (A) SEM of nanotubes with N ) 25 wall layers. (B) TEM of
nanotube with N ) 25 wall layers. (C) SEM showing nanotube length. (D)
SEM of collapsed nanotubes with thin (N ) 5 layer) walls.
and had 126 ( 10 nm-diameter pores. Prior to template synthesis,
both faces of the alumina membrane were sputtered with ultrathin
(
∼5 nm) films of Au. These Au films are too thin to block the
pores at the membrane surfaces; however, they prevent the
adsorption of the first layer of the R,ω-diorganophosphate to the
faces of the membrane. As a result, layered R,ω-diorganophos-
phonate/Zr films (the nanotubes) are deposited only along the pore
walls and not on the faces of the membrane. Without these Au
films, the faces become preferentially coated, and nanotubes are
not obtained.
A film-formation process that is based on layer-by-layer deposi-
tion of the film-forming material along the pore walls of the
template membrane might provide this alternative synthetic chem-
istry. In this case, the wall thickness, and correspondingly the i.d.,
of the nanotubes would be determined by the number of layers of
the material deposited along the pore wall. This was attempted,
using the well-known layer-by-layer polyelectrolyte film-deposition
1
,10-Decanediylbis (phosphonic acid) (DBPA) was prepared as
10
described previously. A 1.25 mM solution of DBPA (adjusted to
7
process in the pores of a nanopore alumina template. However,
pH ) 6 with NaOH) was used to deposit the DBPA layers. A 5.0
the polyelectrolyte deposition process in the pores was found to be
different than that on flat surfaces. As a result, the ability to control
nanotube i.d. was not demonstrated, and only thick-walled nano-
mM solution of ZrOCl
2
‚8H
Zr layers. After the desired number of alternate DBPA/Zr emersion
cycles, the alumina template was dissolved in 27% H PO , and the
2
O was used to deposit the alternating
10
3
4
7
tubes were obtained. In related work, Guo et al. have used this
layered DBPA/Zr nanotubes were collected by filtration (Figure
1). We denote these nanotubes by the number (N) of DBPA/Zr
layers that make up the walls of the tubes.
8
method to coat the outer surfaces of Au nanowires. In addition,
Kovtyukhova et al. have used a method based on alternate SiCl
O deposition/reaction cycles to make silica nanotubes.9
We describe here an alternative layer-by-layer film-forming
4
/
H
2
Nanotubes with wall thicknesses of N ) 5, 15, 25, and 30 DBPA/
Zr layers were examined by both scanning (SEM) and transmission
(TEM) electron microscopy. Nanotubes with walls composed of
10 or more layers have outside diameters equivalent to the pore
diameter of the template (Figure 1A, B), and the hollow core and
wall thickness can be seen in the TEM images (e.g., Figure 1B).
Furthermore, low magnification SEM images show that the length
of these nanotubes is equivalent to the thickness of the template
membrane (38 µm, Figure 1C). Interestingly, nanotubes whose walls
are only N ) 5 layers thick collapse when collected by filtration
(Figure 1D).
method, Mallouk’s alternating R,ω-diorganophosphonate/Zr chem-
istry,10 for preparing nanotubes within the pores of alumina template
membranes. This simple method entails alternate emersion of the
nanopore template into a solution of the R,ω-diorganophosphonate
and then into a solution of ZrO2 to deposit layered R,ω-
diorganophosphonate/Zr nanotubes along the pore wall. We have
found that, in complete analogy to films formed on flat surfaces,10
this method allows for accurate, quantitative, and predictable control
over the wall thickness, and thus i.d., of the layered nanotubes
obtained.
+
When this DBPA/Zr deposition chemistry is used to deposit films
on flat surfaces, each layer is 1.70 ( 0.03 nm thick, and the total
The alumina template membranes were prepared in-house by
anodic oxidation of Al foil;11 these templates were 38 µm thick
10
film thickness (in nanometers) is 1.7 × N. Correspondingly, our
5674
9
J. AM. CHEM. SOC. 2004, 126, 5674-5675
10.1021/ja049537t CCC: $27.50 © 2004 American Chemical Society

C O MMU N I C A T I O N S
Figure 2. Current-voltage curves for nanotubes with the indicated number
of layers comprising the nanotube walls.
Figure 3. Nanotube-membrane resistance data plotted as per eq 1.
Points ) experimental data. Solid curves were calculated using eq 1.
TEM images show that the wall thickness for the DBPA/Zr
nanotubes increases with the number of layers deposited on the
pore walls. However, it is difficult to quantify the nanotube wall
thickness from such images. We have used a simple electrochemical
method12 to prove that the wall thicknesses of our nanotubes (in
fn l aa nt os mu r ef ta ec r es s) .increase as 1.7 × N, exactly as per films deposited on
This method entails measurement of current-voltage curves
associated with ion-transport through electrolyte-filled DBPA/Zr
nanotube membranes.12 For these experiments, the alumina template
membrane was not dissolved away. Instead, the nanotubes were
left embedded within the pores of the template membrane, and the
nanotube-containing membrane was mounted between the two
halves of a U-tube permeation cell.12 Each half-cell was filled with
an expanded version of the data for low values of N. The solid
curves in Figure 3 were calculated from eq 1, and there are no
adjustable parameters in this calculation. These data prove that the
layer-by-layer film deposition process on the pore walls of the
alumina template is identical to deposition on a flat surface. These
data also show that this new nanotube synthesis method allows for
quantitative and predictable control over the wall thickness and
inside diameter of the nanotubes obtained. In addition, nanotubes
with very small inside diameters, for example, as small as 23 nm
in Figure 3, can be obtained.
This R,ω-diorganophosphonate/Zr chemistry should prove to be
a very versatile method for preparing template-synthesized nano-
tubes. This is because both the chemistry of the segment between
the terminal phosphonates and the length of this segment can be
varied at will. In addition, nanotubes with one chemistry on the
outer nanotube walls and a second chemistry on the inner walls
should be possible by simply forming the first layers of the nascent
nanotube using one R,ω-diorganophosphonate and changing to a
second R,ω-diorganophosphonate to deposit the subsequent layers.
0.5 M KCl, and a Ag/AgCl electrode was immersed into each half-
cell solution. A potentiostat was used to sweep the potential
difference applied between these two electrodes between values of
-1
-
1 and +1 V (sweep rate ) 100 mV s ) and measure the resulting
12
transmembrane ion current. This current is carried by ionic
migration through the nanotubes within the membrane.
In agreement with our studies on gold nanotube membranes, the
current-voltages curves are linear (Figure 2), and the slopes of
Acknowledgment. This work was supported by the National
Science Foundation and the Department of Energy.
12
these lines are the inverse of the membrane resistance (R
the wall thickness increases with the number of DBPA/Zr layers
deposited is clearly seen by the increase in R with increasing N
Figure 2). If we define the resistance of the as-synthesized
membrane (no DBPA/Zr within the pores) as Rbare, it is easy to
show that the ratio R /Rbare is related to the radius of the pores in
the as-synthesized membrane (r
m
). That
References
m
(
(1) Joo, S. H.; Choi, S. J.; Oh, I.; Kwak, J.; Liu, Z.; Terasaki, O.; Ryoo, R.
Nature 2001, 412, 169-172.
(
2) Smirnov, A. I.; Poluektov, O. G. J. Am. Chem. Soc. 2003, 125, 8434-
8435.
m
(
3) Mitchell, D. T.; Lee, S. B.; Trofin, L.; Li, N.; Nevanen, T. K.; Soderlund,
H.; Martin, C. R. J. Am. Chem. Soc. 2002, 124, 11864-11865.
o
) 63 nm) and the number of layers
making up the nanotube wall via
(4) Lee, S. B.; Mitchell, D. T.; Trofin, L.; Nevanen, T. K.; Soderlund, H.;
Martin, C. R. Science 2002, 296, 2198-2200.
(
5) Martin, C. R. Science 1994, 266, 1961-1966.
2
o
2
R /R ) r /(r - (1.7 × N))
(1)
(6) Jirage, K. B.; Hulteen, J. C.; Martin, C. R. Science 1997, 278, 655-658.
m
bare
o
(
(
7) Ai, S.; Lu, G.; He, Q.; Li, J. J. Am. Chem. Soc. 2003, 125, 11140-11141.
8) Guo, Y.-G.; Wan, L.-J.; Bai, C.-L. J. Phys. Chem. 2003, 107, 5441-
5444.
Equation 1 assumes that each DBPA/Zr layer making up the
nanotube wall is 1.7 nm thick, as per films deposited on flat
surfaces.10
Figure 3 shows the experimental data obtained from the slopes
of the current-voltage curves (points) plotted as per eq 1. The
standard deviations were obtained from replicate measurements
made on at least five different membranes. The inset in Figure 3 is
(
9) Kovtyukhova, N. I.; Mallouk, T. E.; Mayer, T. S. AdV. Mater. 2003, 15,
780-784.
(10) Lee, H.; Kepley, L. J.; Hong, H.-G.; Akhter, S.; Mallouk, T. E. J. Phys.
Chem. 1988, 92, 2597-2601.
(
11) Foss, C. A. J.; Hornyak, G. L.; Stockert, J. A.; Martin, C. R. J. Phys.
Chem. 1994, 98, 2963.
(
12) Harrell, C.; Lee, S.; Martin, C. R. Anal. Chem. 2003, 75, 6861-6867.
JA049537T
J. AM. CHEM. SOC.
9
VOL. 126, NO. 18, 2004 5675