Photolithographically patterned silver nanowire electrodeposition

Article abstract of DOI:10.1016/j.electacta.2010.02.075

We report the fabrication of silver nanowires using lithographically patterned nanowire electrodeposition (LPNE). The LPNE synthesis of silver nanowires proceeds by lithographically patterning, and then etching an evaporated nickel film to produce a nickel nanoband 20-80 nm in height. This nanoband, which traces the perimeter of the exposed region, is recessed by ≈500 nm into the photoresist producing a horizontal trench. A silver nanowire, of controlled height and width, is formed within this trench by electrodepositing silver from either of two aqueous solutions at the nickel nanoband. Silver nanowires with controlled widths ranging from 100 to 400 nm were obtained and the height of silver nanowires was independently controllable over the range from 20 to 80 nm. The LPNE process is wafer-scale and continuous silver nanowires millimeters in length are readily obtained. Data for the characterization of these nanowires using AFM, TEM, and XRD is presented.

Full text of DOI:10.1016/j.electacta.2010.02.075

Electrochimica Acta 55 (2010) 8074–8080
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Photolithographically patterned silver nanowire electrodeposition
S.C. Kung, W. Xing, K.C. Donavan, F. Yang, R.M. Penner∗
Department of Chemistry, University of California, Irvine, CA 92697-2025, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
We report the fabrication of silver nanowires using lithographically patterned nanowire electrodeposi-
tion (LPNE). The LPNE synthesis of silver nanowires proceeds by lithographically patterning, and then
etching an evaporated nickel film to produce a nickel nanoband 20–80 nm in height. This nanoband,
which traces the perimeter of the exposed region, is recessed by ≈500 nm into the photoresist producing
a horizontal trench. A silver nanowire, of controlled height and width, is formed within this trench by
electrodepositing silver from either of two aqueous solutions at the nickel nanoband. Silver nanowires
with controlled widths ranging from 100 to 400 nm were obtained and the height of silver nanowires
was independently controllable over the range from 20 to 80 nm. The LPNE process is wafer-scale and
continuous silver nanowires millimeters in length are readily obtained. Data for the characterization of
these nanowires using AFM, TEM, and XRD is presented.
Received 17 December 2009
Received in revised form 18 February 2010
Accepted 21 February 2010
Available online 6 March 2010
Keywords:
Silver
Nanowire
Metal
Electrodeposition
Polycrystalline
Photolithography
© 2010 Elsevier Ltd. All rights reserved.
1. Introduction
coming this issue for two types of plating solutions – a sulfate-based
plating solution containing saccharine and a commercial, cyanide-
Among the noble metals, silver occupies a position of promi-
nence because of its plasmonic properties [1–9]. The exploitation
of these properties has motivated efforts to create silver nanos-
tructures using wet chemical methods [10], hydrothermal methods
[11], vapor–liquid–solid (VLS) growth [7], metal amplification [12],
and electrodeposition within porous templates [13–16]. Earlier, we
developed methods for preparing ensembles of silver nanowires
by electrochemical step edge decoration (ESED) [17–20], but the
minimum diameter of these nanowires was in the 100 nm range.
Moreover, ESED does not enable the control of nanowire position,
orientation or number density – a serious liability.
Recently, we have developed a new method for the wafer-scale
patterning of metal and semiconductor nanowires on dielectrics
[21–25]. The purpose of this paper is to describe the adaptation of
this method, called lithographically patterned nanowire electrode-
position (LPNE), to the growth of silver nanowires. Previously, LPNE
has been used to prepare metal nanowires composed of platinum
[21,23,25], palladium [21,23–25], gold [21,23,25–27], and bismuth
[21]. The application of the LPNE method to the growth of sil-
ver nanowires presents a new challenge as compared with these
other metals: the facile electron transfer kinetics of silver promotes
a rapid, diffusion-controlled growth and a propensity towards
dendritic deposition. In this paper, we describe methods for over-
free plating solution for silver (Clean Earth SolutionsTM, 45.220).
2. Experimental methods
2.1. Nanowire fabrication
The seven-step LPNE procedure was used exactly as previously
described [21,23,25] to produce the silver nanowires described
here (Fig. 1) – except for the fifth step of the process involving
silver electrodeposition. The evaporated nickel film, which func-
tions as the working electrode for nanowire growth, has a (1 1 1)
domain orientation and a grain diameter of ≈15 nm, based on XRD
analysis. Silver nanowires were electrodeposited in the lithograph-
ically produced trench using a one-compartment three-electrode
electrochemical cell controlled by a Gamry G300 potentiostat in
conjunction with a computer. During this deposition, the nickel
nanoband electrode functions as the working electrode. The entire
lithographically patterned region was immersed in an aqueous
plating solution with the exception of one edge where an electrical
contact to the nickel layer was established with an alligator clip. A
saturated mercurous sulfate electrode (MSE) and a 1 cm2 platinum
foil were used as the reference and the counter electrode, respec-
tively. All potentials were given with respect to this reference elec-
trode. Either of two unstirred plating solutions were used for silver
deposition: solution 1 – Aqueous (Millipore Milli-Q, ꢀ > 18 Mꢁ cm),
1 mM silver sulfate (Aldrich, 99.999%), 1 mM saccharine (Aldrich,
98+%), and 100 mM sodium sulfate (FisherBiotech, 99.0%) at pH 1.7
obtained by the addition of sulfuric acid (J. T. Baker, ULTREX ultra-
pure); solution 2 – a commercial silver plating solution (Clean Earth
∗
Corresponding author at: Department of Chemistry, University of California,
1102 Natural Sciences II, Irvine, CA 92697-2025, USA. Tel.: +1 949 824 8572;
fax: +1 949 824 8125.
E-mail address: rmpenner@uci.edu (R.M. Penner).
0013-4686/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2010.02.075

S.C. Kung et al. / Electrochimica Acta 55 (2010) 8074–8080
8075
Fig. 1. Schematic diagram of the lithographically patterned nanowire electrodeposition (LPNE) process, which involves seven-step fabrication process.
SolutionsTM, 45.220). The procedures for silver nanowire electrode-
position were optimized for these two plating solutions as follows:
for solution 1, a nucleation pulse of −0.3 V vs. MSE for 10 ms was
applied followed by potentiostatic growth at −0.070 V vs. MSE for
50–400 s, depending on the desired width. For solution 2, a 750 ms
nucleation pulse of −1.4 V vs. MSE was applied followed by poten-
tiostatic growth at −0.475 V vs. MSE for ranging from 50 to 500 s.
electrode for all of these metals except bismuth where silver was
substituted. Nickel is used in this application because it is capable
of functioning as an efficient working electrode and it can be selec-
tively removed at the end of the deposition process using nitric acid
(Fig. 1, step 7) which does not dissolve noble metals. For bismuth
nanowires, the silver sacrificial electrode is removed using a perox-
ide/ammonia etch. Nickel, however, is fully compatible with silver
nanowire growth.
The silver nanowire electrodeposition process is examined next.
We explored the properties of two plating solutions for silver
nanowire growth: solution 1 – a sulfate-based solution contain-
ing saccharine, and solution 2 – a commercial, cyanide-free plating
solution (Clean Earth SolutionsTM, 45.220). Two different electrodes
were employed in these cyclic voltammetric investigations: evap-
orated nickel films on glass (area ≈3 cm2, Fig. 2a and d) and LPNE
patterned nickel nanoband electrodes (Fig. 2b and e). One impor-
tant difference between the two plating solutions can be seen
by comparing the stripping potential for silver in Fig. 2a and b
2.2. Microscopy
Optical micrographs were acquired using
a Carl Zeiss
Axioskop2 equipped with bright field objectives. Scanning elec-
tron microscopy (SEM) images were obtained using a Philips
XL-30 FEGSEM (field emission gun scanning electron microscope)
equipped with the elemental analysis capability of energy-
dispersive X-ray spectroscopy (EDX), operated at 10 keV. All
samples were sputter coated with Au/Pd alloy prior to imaging
to eliminate charging. Transmission electron microscopy (TEM)
images and selective area electron diffraction (SAED) were acquired
on a Philips CM 20 TEM at a 200 keV accelerating voltage. In order
to prepare TEM specimens, silver nanowires were produced on a
photoresist (PR) pre-coated surface that facilitate the release of
these nanowires by dissolution of the photoresist. PR-supported
silver nanowires were transferred onto a carbon film coated Cu grid
(Ted Pella, Inc.) using an acetone stream (which dissolved the PR
layer). Transferred nanowires were dried overnight before imag-
ing. Noncontact mode atomic force microscopy (AFM) analysis was
conducted using a MFP-3DTM stand alone atomic force microscope
(Asylum Research) with Olympus, AC160TS tips in a laboratory air
ambient.
(solution 1) and that seen in Fig. 2d and e (solution 2). This poten-
ꢀ
tial – which approximates EAog
in these solutions – is shifted
+
0
/Ag
in the negative direction by approximately 0.40 V for solution 2.
This shift reveals the presence of complexing agents for Ag+ in
this commercial plating solution; in solution 1, silver is uncom-
plexed. Other differences in the kinetics of silver deposition at these
two electrodes from these two plating solutions are also appar-
ent. In solution 1, silver electrodeposition proceeds on a clean
evaporated nickel surface at −(0.10–0.15) V (Fig. 2a, scan 1). A
somewhat more negative onset for silver deposition is seen for the
CVs recorded at the patterned nickel nanoband, with deposition
onsets in the range from −(0.20–0.25) V (Fig. 2b, scan 1). For both
electrodes, a more positive onset for silver deposition is seen on
scans 2 and 3. This positive shift is attributed to the presence on the
nickel surface of residual silver that facilitates the renucleation and
growth of silver on these subsequent scans. Thus, the observation
of this shifted silver deposition onset implies that silver is incom-
pletely removed from the nickel surface by anodic stripping in
solution 1.
2.3. X-ray diffraction
Grazing incidence X-ray diffraction (GIXRD) was performed
on arrays of thousands of silver nanowires patterned on glass
at 2–5 ␮m pitch, using a Rigaku Ultima III X-ray diffractometer
employing parallel beam optics with an incident angle of 0.3◦. The
Cu K␣ irradiation was operated at 40 kV and 44 mA. The specimens
were scanned from 2ꢂ = 15◦ to 85◦ with step-scan increments of
0.1◦ and a dwell time at each increment of 20 s.
Of course in solution 2 the onset for silver deposition is shifted
negatively as compared with solution 1 because of the −0.40 V
ꢀ
shift in Eo . However beyond this thermodynamic shift, the over-
ꢀ
potential for silver deposition (referenced to Eo ) is larger than the
3. Results and discussion
overpotentials observed in solution 1 at both types of electrodes
indicating that silver nucleation is impeded in solution 2. For exam-
ple, at the flat nickel electrode no significant plating current is
seen on the first scan until −0.80 V (Fig. 2d) – an overpotential of
∼−0.40 V – which compares with an overpotential in solution 1
of just ∼−0.10 V (Fig. 2a). The onset for silver deposition on subse-
3.1. Silver nanowire deposition
LPNE (Fig. 1) has been used previously to deposit nanowires
composed of platinum [21,23,25], palladium [21,23–25], gold
[21,23,25–27], and bismuth [23]. Nickel was used as the “sacrificial”

8076
S.C. Kung et al. / Electrochimica Acta 55 (2010) 8074–8080
Fig. 2. Cyclic voltammograms (50 mV/s) (a) for the nickel film and (b) for the nickel nanoband in solution 1, and (d) for the nickel film and (e) for the nickel nanoband in
solution 2. Pulse waveforms employed to produce silver nanowires from (c) solution 1 and (f) solution 2. Solution 1 – the sulfate-based aqueous bath containing 1 mM silver
sulfate, 1 mM saccharine, and 100 mM sodium sulfate at pH 1.7; Solution 2 – Clean Earth SolutionsTM, 45.220.
quent scans is shift positively by more than 0.30 V, again suggesting
that silver is not quantitatively removed from the nickel surface on
the positive-going voltammetric scan. In solution 2, a diffusional
wave for silver deposition – a prominent feature of the deposition
in solution 1 – is hardly observed on these subsequent scans sug-
gesting that the rate of silver deposition is reduced by additives
present in solution 2. The same −0.40 V overpotential is seen for
scan 1 at the patterned nickel nanoband (Fig. 2e), but a large over-
potential of −(0.30–0.35) V persists for scans 2 and 3 – in contrast
to the behavior seen for the nanoband in solution 1. Thus, the pro-
prietary additives present in solution 2 are effective in suppressing
the rate of silver deposition at LPNE patterned electrodes even after
an initial voltammetric scan.
Three conclusions based on these observations are the follow-
ing: (1) The onsets for silver deposition at the LPNE patterned
electrode are more negative by ∼−0.10 V relative to the flat nickel
electrode for both solutions. One explanation for this shift is the
rapid depletion of Ag+ from the narrow trench present in the
LPNE template, coupled with slow Ag+ transport into this trench
from the bulk solution. (2) Overpotentials for silver nucleation are
significantly larger in solution 2 as compared with solution 1 at
both electrodes. (3) The rate of silver deposition from solution 2 is
impeded as compared with solution 1 at both electrodes.
Based on the disparate voltammetric behavior of these two sil-
ver plating solutions, two pulse waveforms (Fig. 2c and f) optimized
for these two solutions were developed to efficiently nucleate and
grow silver nanowires in the LPNE template. For nanowire growth
from solution 1 (Fig. 2c), this program involves the application of a
nucleation pulse of −0.30 V × 10 ms followed by growth at −0.070 V
for a duration appropriate for the desired wire width (see below).
In solution 2 (Fig. 2f), a much more aggressive nucleation pulse
of −1.4 V × 750 ms followed by growth at −0.475 V was employed
to compensate for the greater nucleation barrier revealed by the
voltammetry results. The voltage range from −(0.470–0.475) V rep-
resented the most positive values for which silver nanowires were
obtained. Within this potential range, the dendritic growth of sil-
ver was suppressed and the width uniformity of the deposited
silver nanowires was optimized. These two deposition programs
produced silver nanowires that were remarkably similar, in terms
of shape and dimensional uniformity but the microstructure
of these nanowires were distinctly different, as documented
below.
3.2. Dimensional control
Atomic force microscopy (AFM) images of an array of linear sil-
ver nanowires prepared from solution 2 are shown in Fig. 3a and
b. AFM images of nanowires prepared from solution 1 were indis-
tinguishable from those shown here. These images reveal that the
nickel template and photoresist layers can be cleanly and selec-
tively removed from the surface following silver nanowire growth
(steps 6 and 7 of Fig. 1) without damaging the silver nanowires
and without releasing these nanowires or altering their position
on the glass surface. The thickness uniformity of these nanowires
can be assessed by measuring cross-sections using AFM (Fig. 3a–c).
The wire height seen in these cross-sections is constant to within
10–15% of the mean height. Moreover, the nanowire total height
equals the thickness of the nickel film (Fig. 3d) deposited in the
first step of the LPNE process (Fig. 1, step 1).
The width of these silver nanowires is most accurately measured
using scanning electron microscopy (SEM). Again, no significant dif-
ferences between the nanowires obtained from solutions 1 and 2
were observed. Nanowires deposited at 5 ␮m pitch from solution
2 are shown in Fig. 4. The width of these nanowires was adjusted
by varying the duration of the growth time. Plots of wire width
vs. growth time (Fig. 4e) are approximately linear and virtually

S.C. Kung et al. / Electrochimica Acta 55 (2010) 8074–8080
8077
Fig. 3. Atomic force micrographs at low magnification (a, 40 ␮m × 40 ␮m) and higher magnification (b, 5.0 ␮m × 5.0 ␮m) of silver nanowires. The black arrow indicates the
growth direction. (c) Amplitude traces of three kinds of silver nanowires prepared using nickel films with the thickness of 20 nm, 40 nm, and 80 nm, respectively. (d) Plot of
the mean height for produced silver nanowires reveals the linear dependency of the nickel layer thickness. (For interpretation of the references to color in this figure legend,
the reader is referred to the web version of this article.)
identical for these two plating solutions. This result is somewhat
surprising since the nucleation pulse of 750 ms used for solution
2 is a factor of 75 longer than that used in solution 1. We specu-
late that the reason silver nanowires prepared from solution 2 are
not wider than those obtained from solution 1 is that both of these
nucleation pulses substantially depletes silver from the solution
between neighboring nanowires, which are spaced by 5 ␮m. This
explanation is supported by the following analysis: the excursion
was somewhat better for nanowires prepared using solution 2. The
most important conclusion of these data is that the duration of
the growth pulse affords control over the nanowire width over the
range from 100 to 400 nm.
3.3. Structural characterization
Grazing incidence X-ray diffraction (GIXRD) provides a powerful
method for the characterization of the nanowires produced using
LPNE that does not require the removal of nanowires from the glass
surface. Samples suitable for GIXRD measurements were prepared
by patterning linear silver nanowires at high number density – typ-
ically with an interwire spacing of 2–5 ␮m. An optical micrograph
of a silver nanowire array suitable for this purpose (Fig. 5a) shows
hundreds of linear silver nanowires. GIXRD patterns were acquired
by fixing the incident angle for X-rays, collimated using parallel
beam optics, at a value below the critical angle and translating
of the diffusion layer from the nickel edge, x, can be approximated
√
as: x = 2Dt, where the diffusion coefficient, D, can be approxi-
mated at 1 × 10−5 cm2 s−1 and t is the nucleation pulse duration.
For t values of 750 and 10 ms, x is 39 and 4.5 ␮m, respectively.
Thus, in both cases the diffusion layer can be expected to encom-
pass the entire interwire distance and silver ion is likely to be
depleted from this volume by both of these nucleation pulses. Error
bars shown in Fig. 4e document the standard deviation of the wire
width and provide an indication of the wire width uniformity which

8078
S.C. Kung et al. / Electrochimica Acta 55 (2010) 8074–8080
Fig. 4. Scanning electron microscopy (SEM) images of (a) a silver nanowire array at a 5 ␮m lateral pitch. (b) Different widths of silver nanowires by different deposition
time. (c) The quasi-linear correlation can be observed between the mean widths of the generated nanowires and the corresponding deposition time from a sulfate-based
aqueous bath (red) and the commercial silver plating solution (blue). (d) The energy-dispersive X-ray (EDX) analysis confirms the composition of the silver nanowire. (For
interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
the analyzer through an arc of 2ꢂ to record the diffraction pattern
(Fig. 5b) [23]. GIXRD patterns (Fig. 5c) for the two plating solutions
both show five reflections assignable to fcc silver as indicated. The
intensities seen for both experimental patterns track the intensi-
ties expected from the model pattern (Fig. 5c, bottom) indicating
that silver grains are isotropically oriented with respect to the scat-
tering direction [15]. A difference in the diffraction line widths,
however, is observed with nanowire deposited from solution 1
producing narrower diffraction lines than those obtained from
solution 2. If it can be assumed that X-ray line broadening is dom-
inated by the crystallite diameter as opposed to lattice strain, the
average crystallite size can be estimated from the Debye–Scherrer
equation [28]:
0.9ꢃ
d =
ˇ · cos ꢂ
where d is the mean crystallite diameter, ꢃ is the X-ray wave-
length, ˇ is the full width of the diffraction line at half its maximum
intensity, and ꢂ is the diffraction angle. Silver nanowires deposited
from solution 1 have a mean grain diameter of ∼24 nm whereas
those deposited from solution 2 have a mean grain diameter of
∼15 nm. It is important to note that the scattering vector for the
Fig. 5. (a) Optical micrograph of an array of silver nanowires prepared using LPNE and used for the acquisition of grazing incidence X-ray diffraction (GIXRD). The distance
between nanowires in this array was ≈5 ␮m pitch. (b) Schematic diagram showing the geometry of the GIXRD experiment involving a fixed 0.3◦ angle of incidence, ꢄi. (c)
GIXRD patterns for nanowire deposited using solution 1 (labeled SO42−) and solution 2 (labeled Clean EarthTM). The assignment of reflections to fcc silver (JCPDS 71-4613)
is shown at bottom. A broad reflection at 25◦ is produced by diffuse scattering from the glass surface.

S.C. Kung et al. / Electrochimica Acta 55 (2010) 8074–8080
8079
GIXRD measurement is not fixed along the surface normal, as it is
in the conventional Bragg–Brentano geometry, but it nevertheless
has its principle component perpendicular to the surface so to a
first approximation, it is the grain dimension perpendicular to the
surface that is obtained using this measurement.
As compared with GIXRD, transmission electron microscopy
(TEM) provides complimentary information relating to the grain
structure of these nanowires both because it is a real-space imag-
ing technique and because it is the in-plane, not out-of-plane, grain
dimensions that are seen. TEM images of nanowires obtained from
solution 1 (left) and solution 2 (right) are compared in Fig. 6a.
Dark lines in these images are associated with grain boundaries.
For either of these nanowires, if an individual grain is examined at
high magnification lattice-fringes are observed (e.g., Fig. 6b). Here
the distance between adjacent lattice planes is ∼2.40 Å, close to that
expected for the d-spacing along the (1 1 1) plane of silver (2.36 Å,
JCPDS71-4613). Selected-areaelectron diffractionanalysis (Fig. 6c),
acquired from a 10 ␮m diameter area of the surface encompass-
ing several nanowires deposited from solution 2, shows concentric
rings of diffracted electron intensity as expected for polycrystalline
materials.
A histogram of crystallite diameters can be constructed by
acquiring TEM images of the same nanowire section at a series of tilt
angles relative to the electron beam axis. In each of these images,
a unique subset of the grains meets the condition for diffraction
of the electron beam and these grains appear black in the trans-
mission image [29], enabling the accurate measurement of grain
dimensions. These histograms (Fig. 6d) reveal a significant differ-
Fig. 6. (a) Transmission electron micrographs (TEM) of two silver nanowires, both 20 nm in thickness. The wire at left was obtained from solution 1 and the wire at right from
solution 2. (b) high magnification TEM image showing closest packed silver atomic layers along (1 1 1) within a single crystalline grain. This lattice resolved image reveals
the interplane distance of 2.40 Å consistent with fcc silver. (c) Selected-area electron diffraction (SAED) pattern for the nanowire prepared from solution 2. The pattern seen
for the wire deposited from solution 1 was virtually identical. Concentric rings, characteristic of polycrystallinity, are assigned to the reflections of fcc silver (JCPDS 71-4613).
(d) Histograms of the crystallite size for silver nanowires prepared from the two plating solutions, determined by tilt-resolved TEM image sequences.

8080
S.C. Kung et al. / Electrochimica Acta 55 (2010) 8074–8080
ence between nanowires deposited from solution 1 and solution 2.
Specifically, nanowires deposited from solution 1 show a bimodal
grain diameter distribution with mean values near 30 and 80 nm.
This bimodal microstructure is already apparent from the TEM
image shown in Fig. 6a (left) where it can be clearly seen that
the grain size increases dramatically with distance away from the
nickel edge (from right to left in Fig. 6a). The largest grains seen
along the left edge of that nanowire are in the 70–100 nm range
whereas the smaller grains at the right edge are 10–40 nm in size.
The increase in grain diameter during nanowire growth from solu-
tion 1 could be the result of saccharine depletion since saccharine
is a nucleation promoter in this system. In contrast, nanowires
obtained from solution 2 (e.g., Fig. 6a, right) are composed entirely
of 10–40 nm crystallites. It is important to note here that we found it
impossible to produce continuous silver nanowires using solution 1
with the omission of saccharine so it plays an essential role. Further
investigations will be required to fully understand the mecha-
nism by which saccharine promotes continuous nanowire growth.
This result, together with the less profound grain size differences
exposed by GIXRD, suggests that additives present in the commer-
cial plating solution (solution 2) are more effective than saccharine
(present in solution 1) at promoting the formation of new silver
nuclei at the relatively positive potentials used to propagate silver
nanowire growth (Fig. 2c and f). The silver grain diameter seen at
the nickel edge can be used to estimate the minimum nanowire
width obtainable using this process and this minimum width is
in the 10–40 nm range, similar to the smallest gold and platinum
nanowires obtained using the LPNE process [21,23,25].
near the nickel edge to values near 80 nm at the other edge of
the nanowire, just 200 nm distant from the nickel edge. No evo-
lution in grain diameter with growth time was observed for the
nanowires obtained from solution 2 where a mean value near
30 nm applied across the diameter of the nanowire.
Acknowledgements
The authors gratefully acknowledge the financial support by the
National Science Foundation (CHE-0956524) and the School of Phys-
ical Sciences, Center for Solar Energy at UCI. Electron microscopy was
performed at the Materials Characterization Center, LEXI/CALIT2, of
the University of California-Irvine. The constructive comments of a
reviewer of this paper are also gratefully acknowledged.
References
[1] M.V. Sosnova, N.L. Dmitruk, A.V. Korovin, S.V. Mamykin, V.I. Mynko, O.S. Lytvyn,
Applied Physics B: Lasers and Optics (2009).
[2] H. van der Lem, A. Moroz, Journal of Optics A: Pure and Applied Optics 2 (2000)
395.
[3] H. Ditlbacher, A. Hohenau, D. Wagner, U. Kreibig, M. Rogers, F. Hofer, F.R.
Aussenegg, J.R. Krenn, Physical Review Letters 95 (2005) 4.
[4] A.L. Pyayt, B. Wiley, Y. Xia, A. Chen, L. Dalton, Nature Nanotechnology 3 (2008)
660.
[5] G. Schider, J.R. Krenn, A. Hohenau, H. Ditlbacher, A. Leitner, F.R. Aussenegg, W.L.
Schaich, I. Puscasu, B. Monacelli, G. Boreman, Physical Review B 68 (2003) 4.
[6] M. Moskovits, Reviews of Modern Physics 57 (1985) 783.
[7] P. Mohanty, I. Yoon, T. Kang, K. Seo, K.S.K. Varadwaj, W. Choi, Q.H. Park, J.P. Ahn,
Y.D. Suh, H. Ihee, B. Kim, Journal of the American Chemical Society 129 (2007)
9576.
[8] C. Zuo, P.W. Jagodzinski, The Journal of Physical Chemistry B 109 (2005) 1788.
[9] C.-W. Kuo, J.-Y. Shiu, K.H. Wei, P. Chen, Chemistry Letters 37 (2008) 610.
[10] N.R. Jana, L. Gearheart, C.J. Murphy, Chemical Communication (2001) 617.
[11] Z. Wang, J. Liu, X. Chen, J. Wan, Y. Qian, Chemistry – A European Journal 11
(2005) 160.
4. Conclusion
The LPNE method outlined in this paper provides a robust
method for the fabrication of ultra-long, polycrystalline silver
nanowires. The following are the key conclusions pertaining to this
process:
[12] M. Barbic, J.J. Mock, D.R. Smith, S. Schultz, Journal of Applied Physics 91 (2002)
9341.
[13] Q. Zhang, Y. Li, D. Xu, Z. Gu, Journal of Materials Science Letters 20 (2001) 925.
[14] R.-L. Zong, J. Zhou, Q. Li, B. Du, B. Li, M. Fu, X.-W. Qi, L.-T. Li, S. Buddhudu, Journal
of Physical Chemistry B 108 (2004) 16713.
[15] G. Riveros, S. Green, A. Cortes, H. Gomez, R.E. Marotti, E.A. Dalchiele, Nanotech-
nology 17 (2006) 561.
[16] P.R. Evans, R. Kullock, W.R. Hendren, R. Atkinson, R.J. Pollard, L.M. Eng, Advanced
Functional Materials 18 (2008) 1075.
[17] E.C. Walter, B.J. Murray, F. Favier, G. Kaltenpoth, M. Grunze, R.M. Penner, Journal
of Physical Chemistry B 106 (2002) 11407.
[18] E.C. Walter, M.P. Zach, F. Favier, B.J. Murray, K. Inazu, J.C. Hemminger, R.M.
Penner, ChemPhysChem 4 (2003) 131.
[19] B.J. Murray, E.C. Walter, R.M. Penner, Nano Letters 4 (2004) 665.
[20] B.J. Murray, J.T. Newberg, E.C. Walter, Q. Li, J.C. Hemminger, R.M. Penner, Ana-
lytical Chemistry 77 (2005) 5205.
[21] E.J. Menke, M.A. Thompson, C. Xiang, L.C. Yang, R.M. Penner, Nature Materials
5 (2006) 914.
[22] Y. Yang, S.C. Kung, D.K. Taggart, C. Xiang, F. Yang, M.A. Brown, A.G. Guell, T.J.
Kruse, J.C. Hemminger, R.M. Penner, Nano Letters 8 (2008) 2447.
[23] C. Xiang, S.C. Kung, D.K. Taggart, F. Yang, M.A. Thompson, A.G. Guell, Y. Yang,
R.M. Penner, ACS Nano 2 (2008) 1939.
[24] F. Yang, D.K. Taggart, R.M. Penner, Fast, Nano Letters 9 (2009) 2177.
[25] C. Xiang, Y. Yang, R.M. Penner, Chemical Communication (2009) 859.
[26] C. Xiang, A.G. Guell, M.A. Brown, J.Y. Kim, J.C. Hemminger, R.M. Penner, Nano
Letters 8 (2008) 3017.
1. The LPNE process used for the fabrication of gold, platinum, and
palladium nanowires, involving a sacrificial nickel electrode, can
be adapted to the growth of silver nanowires using either of two
silver plating solutions.
2. Compared with the preparation by LPNE of gold, platinum and
palladium nanowires, we encountered greater difficulty obtain-
ing continuous nanowires of uniform width.
3. This issue was addressed by employing a nucleation pulse at the
beginning of the deposition process to promote silver nucleation.
The parameters for the nucleation pulses were optimized for the
two plating solutions employed here.
4. Control of the silver nanowire height is not affected by either
of these complications; as with other metals, the wire height is
controlled by the thickness of the evaporated nickel film over
the range from 20 to 80 nm.
5. The silver nanowires obtained from these two solutions are
indistinguishable externally, but they have distinctly different
grain structures. The nanowires obtained from solution 1 show
an increase in grain diameter from a mean value near 30 nm
[27] C. Xiang, J.Y. Kim, R.M. Penner, Nano Letters 9 (2009) 2133.
[28] A.L. Patterson, Physical Review 56 (1939) 978.
[29] D.T. Carpenter, J.M. Rickman, K. Barmak, Journal of Applied Physics 84 (1998)
5843.