Investigation of a Single Pd Nanowire for Use as a Hydrogen Sensor

communications
Hydrogen sensors
DOI: 10.1002/smll.200500365
Investigation of a Single Pd Nanowire for
Use as a Hydrogen Sensor**
Yeonho Im, Choonsup Lee, Richard P. Vasquez,
Mangesh A. Bangar, Nosang V. Myung, Eric J. Menke,
Reginald M. Penner, and Minhee Yun*
Interest in nanowire technology has grown significantly over
the past several years, due to their potentially broad applications in optics, electronics, and sensors (e.g., resonators,[1]
molecular detection,[2] nanoconnectors,[3] biosensors,[4] and
gas sensors[5]). Due to their small size, sensitivity, real-time
detection, and ultra-low power demands, nanowire sensors
are being investigated for detection of a wide range of
chemical and biochemical species. The techniques used to
fabricate these sensors include laser ablation cluster formation,[6] focused ion-beam etching of electrodes to create a
small gap ( 100 nm),[7] use of ion-track membranes,[8] use
of alumina templates,[9] and electrochemical step-edge decoration.[10] These techniques have drawbacks of limited controllability and manufacturability, therefore reliable and
controllable nanowire fabrication remains a significant challenge.
We have previously reported a method of electrodepositing Pd wires with micrometer diameters[11] and conducting
polymer nanowires of polyaniline and polypyrrole with diameters of 100 nm.[12] In this work, single Pd nanowires are
fabricated by electrodeposition in electrolyte channels patterned with electron-beam lithography, and hydrogen sensing is demonstrated. This fabrication technique can produce
nanowires with controlled dimensions, positions, alignments,
and chemical compositions, and potentially enables the use
of a wide range of materials for creating arrays of single-
[*] Prof. M. Yun
Department of Electrical and Computer Engineering
University of Pittsburgh, PA 15261 (USA)
Fax: (+ 1) 412-624-8003
E-mail: [email protected]
Dr. C. Lee, R. P. Vasquez
Jet Propulsion Laboratory, California Institute of Technology
Pasadena, CA 91109 (USA)
M. A. Bangar, Prof. N. V. Myung
Department of Chemical and Environmental Engineering and
Center for Nanoscale Science and Engineering
University of California at Riverside, Riverside, CA 90025 (USA)
E. J. Menke, Prof. R. M. Penner
Department of Chemistry, University of California at Irvine
Irvine, CA 92697 (USA)
Y. Im
Chonbuk National University
Jeonju, Jeonbuk (South Korea)
[**] This research was performed under a contract with the National
Aeronautics and Space Administration.
356
nanowire sensors. Additionally, the growth of nanowires
using this technique would allow for easy integration with
existing silicon technology.
Figure 1 shows SEM images of typical electrodeposited
nanowires between electrodes. The magnified image in Figure 1 b shows a single Pd nanowire with a diameter ranging
from 70 to 85 nm. An EDS (energy dispersive spectroscopy)
elemental analysis detected only Pd metal in the nanowire.
Figure 1. SEM images of a single Pd nanowire grown between gold
electrodes using an e-beam patterned electrolyte channel and electrodeposition (a) with a diameter of between 70 and 85 nm (b).
Figure 2 shows various nanowire diameters, ranging
from as small as 100 nm to as large as 1 mm, with lengths up
to 7 mm. Figure 2 a shows a nanowire with a diameter of
100–150 nm that is 3 mm long. This nanowire has a high surface area with a bristle structure due to the electric field
inside the nanochannel. I–V characterization shows ohmic
and metallic behavior with a resistance of 875 W at room
temperature. This is much larger than the value of about 1–
5 W expected for bulk Pd resistivity. The additional resistance may be attributed to nanocontacts and the grain-boundary scattering effect.[13] For small film thickness the mean
grain size is smaller, which leads to the presence of more
grain boundaries and, hence, higher resistivity.[14] The diameters of the other Pd single nanowires shown in Figure 2 are
approximately 100 nm to 1 mm (b), 100–300 nm (c), and
250–350 nm (d); the initial resistances of these nanowires
are 85, 248, and 235 W, respectively. In general, the resistance of a nanowire is dependent on the diameter of the
B 2006 Wiley-VCH Verlag GmbH & Co. KGaA, D-69451 Weinheim
small 2006, 2, No. 3, 356 – 358
tergrain contacts and closing
of the nanogaps when hydrogen absorption causes the Pd
grains to expand.[5]
Figure 3 c shows the measured output voltages when
hydrogen concentrations of
0.1, 1, 5, and 10 % are cycled
three times (Ex1-1, -2, and
-3), and with hydrogen concentrations of 0.1, 1, and 10 %
cycled three times (Ex2-1, -2,
and -3) to confirm reproducibility. There is little scatter in
the measured output voltages
between cycles and the hydrogen concentration can be
uniquely determined for hydrogen concentrations above
1 %. However, for concentrations of 1 % or below the
measurements are more variable and are not dependent on
Figure 2. SEM images of various Pd nanowires with diameters of: a) 100–150 nm, b) 100 nm to 1 microthe hydrogen concentration.
meter, c) 100–300 nm, and d) 250–350 nm.
Although the values of low
hydrogen
concentrations
cannot be determined, the Pd
nanowire sensors do detect the presence of hydrogen down
nanowire, with the narrowest region determining the final
to the lowest concentration obtainable (0.02 %) in this
resistance. Our nanowires have the same height, but differwork.
ent growth conditions give different lengths. The nanowires
In conclusion, this report describes the fabrication of
in Figures 2 a and b have lengths of 3 mm, whereas those in
single Pd nanowires with diameters ranging from 70 to
Figures 2 c and d have lengths of 5 mm between the elec300 nm and up to 7 mm in length using standard semicontrodes. Even though tapered structure electrodes (c and d)
ductor processing techniques and electrodeposition. This
have longer distances between electrodes, they show more
process can, in principle, produce sensor arrays that are castable nanowire growth.
pable of sensing multiple chemical species simultaneously.
The sensing of different hydrogen concentrations (beWe have demonstrated hydrogen sensing in the concentratween 0.02 and 10 % in N2 gas) was demonstrated with the
tion range of 0.02 % to 10 %. This Pd nanowire sensor exsingle electrodeposited Pd nanowire shown in Figure 2 a.
hibits a fast response time (< 300 ms) with ultralow power
Figure 3 a shows an increased output voltage (decreased
consumption (< 25 mW).
nanowire resistance) with hydrogen gas-flow and a return to
the original state when no hydrogen gas is present. The
sensor responds to hydrogen across the entire concentration
Experimental Section
range studied. We were not able to obtain hydrogen gas
concentrations lower than 0.02 % due to the limitations of
the hydrogen gas-flow controller, so the lower limit of sensiFabrication of the Pd nanowires was similar to the procedure
tivity could not be determined. However, the demonstrated
described previously.[10] A 0.5-mm-thick SiO2 insulator layer was
sensitivity is comparable to, or better than, previously regrown on a Si(100) wafer by wet oxidation. A Ti (25 ,) adhesion
ported H2 gas sensors such as those based on Pd,[15, 16] fieldlayer and thick Au (approx. 1500 ,) contact layer were then deeffect transistors,[17] and metal oxides,[18] with lower power
posited, with electrodes and contacts being patterned by liftoff.
consumption.
Thermally evaporated SiO ( 1500 ,) was then deposited on top
Figure 3 b shows the variation of output voltage in a
of the electrodes. A poly(methyl methacrylate) layer was selecsingle Pd nanowire while the hydrogen gas-flow is cycled to
tively opened by e-beam lithography, and electrolyte channels
yield hydrogen concentrations in the range 0.1 % to 10 %.
were formed in the SiO by reactive-ion etching. One drop of elecThe Pd sensor response is rapid, reversible, and reproducitroplating solution—[Pd(NH2)2(NO2)2] (10 g L 1) and NH4NH2SO3
ble, as shown in Figure 3 b. Hydrogen gas with concentra(100 g L 1)—was placed into each channel with a micropipette.
tions of 0.1, 1, 5, and 10 % was cycled three times in 370 s
The nanowire grew from the cathode to the anode through the
and each concentration showed about the same response bechannel when an electric potential was applied. The diameter
tween cycles. The observed decrease of the nanowire resistand length of each nanowire can be predetermined by the width
ance (output voltage increase) may be due to improved inof the electrolyte channel and the distance between the elecsmall 2006, 2, No. 3, 356 – 358
B 2006 Wiley-VCH Verlag GmbH & Co. KGaA, D-69451 Weinheim
www.small-journal.com
357
communications
measured the variation of output voltage caused by the resistance change in a single Pd nanowire to which a constant voltage
of 5 mV was applied. An ammeter with a current amplifier with a
gain of 106 was placed in series with the nanowire, and the
output voltage of the current amplifier was measured with a
fixed output resistance of 2 W.
Keywords:
hydrogen · lithography · nanowires · palladium · sensors
Figure 3. Electrical signals as a function of time for various concentrations of hydrogen in nitrogen using a Pd single nanowire: a) signals for various concentrations between 0.02 and 0.2 % hydrogen,
b) a comparison of signals for concentrations of 0.1, 1, 5, and 10 %
over 360 s, and c) a comparison of signals for concentrations of 0.1,
1, 5 %, and 10 % hydrogen versus output voltage.
trodes; the channel also limits the dendrite branching during
growth.
Sensor measurements were carried out in a 160-mL aluminum cell with BNC electrical feed-troughs for gas outlets. These
measurements were made at room temperature and atmospheric
pressure, with ultrahigh purity nitrogen as the carrier gas. We
358
www.small-journal.com
[1] N. A. Melosh, A. Boukai, F. Diana, B. Gerardot, A. Badolato,
P. M. Petroff, J. R. Heath, Science 2003, 300, 114.
[2] B. Q. Xu, N. J. Tao, Science 2003, 301, 1221.
[3] B. H. Hong, S. C. Bae, C.-W. Lee, S. Jeong, K. S. Kim, Science
2001, 294, 348.
[4] Y. Cui, Q. Wei, H. Park, C. M. Lieber, Science 2001, 293, 1289.
[5] F. Favier, E. C. Walter, M. P. Zach, T. Benter, R. M. Penner, Science 2001, 293, 2227.
[6] M. Morales, C. M. Lieber, Science 1998, 279, 208.
[7] Z. Li, H. X. He, A. Bogozi, J. S. Bunch, N. J. Tao, Appl. Phys. Lett.
2000, 76, 1333.
[8] M. E. Toimil Molares, E. M. Hçhberger, C. Schaeflein, R. H. Blick,
R. Neumann, C. Trautmann, Appl. Phys. Lett. 2003, 82, 2139.
[9] J.-W. Kim, E.-A. Choi, S.-M. Park, J. Electrochem. Soc. 2003, 150,
E202.
[10] M. P. Zach, K. Inazu, K. H. Ng, J. C. Hemminger, R. M. Penner,
Chem. Mater. 2002, 14, 3206.
[11] M. Yun, N. V. Myung, R. P. Vasquez, C. Lee, E. Menke, R. M.
Penner, Nano Lett. 2004, 4, 419.
[12] K. Ramanathan, M. Bangar, M. Yun, W. Chen, A. Mulchandani,
N. V. Myung, Nano Lett. 2004, 4, 1237.
[13] C. Durkan, M. E. Welland, Phys. Rev. B 2000, 61, 14 215.
[14] A. F. Mayadas, M. Shatzkes, M. Janak, Appl. Phys. Lett. 1969,
14, 345.
[15] C. Wang, A. Mandelis, J. A. Garcia, Sens. Actuators B 1999, 60,
228.
[16] V. G. Ponomareva, G. V. Lavrova, E. F. Hairetdinov, Sens. Actuators B 1997, 40, 95.
[17] C.-C. Lu, F. Udreaa, J. W. Gardner, D. Setiadia, T. Dogarua, T. H.
Tsaia, J. A. Covington, Proc. SPIE 2001, 4408, 86.
[18] L. Bjçrnsson , E. G. Hçrnsten, B. Mattiasson, Biotechnol. Bioeng.
2001, 73, 35.
Received: September 24, 2005
Published online on January 13, 2006
B 2006 Wiley-VCH Verlag GmbH & Co. KGaA, D-69451 Weinheim
small 2006, 2, No. 3, 356 – 358