Plasma-Induced Formation of Ag Nanodots for
Ultra-High-Enhancement Surface-Enhanced Raman Scattering
Zhiyong Li,*,†William M. Tong,†,‡William F. Stickle,§David L. Neiman,§and
R. Stanley Williams†
Quantum Science Research, Hewlett-Packard Laboratories, 1501 Page Mill Road, Palo Alto, California
94304, AdVanced Materials Process Lab and AdVanced Diagnostic Lab, Hewlett-Packard Company,
CorVallis, Oregon 97330
Luke L. Hunter and A. Alec Talin
Sandia National Laboratories, LiVermore, California 94550
D. Li and S. R. J. Brueck
Center for High Technology Materials, UniVersity of New Mexico, Albuquerque, New Mexico 87106
ReceiVed December 20, 2006. In Final Form: February 13, 2007
We report here plasma-induced formation of Ag nanostructures for surface-enhanced Raman scattering (SERS)
with imprint lithography; then the substrate was further treated with an oxygen plasma to fracture the patterned
structures into clusters of smaller, interconnected, closely packed Ag nanoparticles (20-60 nm) and redeposited Ag
(1010) over the entire substrate for 4-mercaptophenol molecules. By comparison, Au patterned structures fabricated
with the same method did not undergo such a morphological change after the plasma treatment and showed no
enhancement of Raman scattering.
For 30 years, surface-enhanced Raman scattering (SERS) has
attracted considerable interest because of its great potential for
trace chemical analysis. One critical issue limiting the general
application of SERS is the economical fabrication of a uniform
for a uniform, repeatable, manufacturable SERS substrate with
enormous enhancement factors that were observed in random
“hot spots.”1The importance of geometry and the separation
between metallic nanostructures was recognized soon after the
initial emphasis of generating rough metal surfaces.2-4Both
theoretical and experimental studies indicated that a controlled
separation of less than 5 nm between aggregated nanoparticles
is one of the important characteristics of strongly enhancing
synthesized by a bottom-up approach, such as by Langmuir-
Blodgett films or by casting metal colloids onto surfaces, can
achieve regular arrays of metallic nanoparticles or nanowires
with small separation,6,7but this approach usually requires a
which can interfere with the Raman signal of the analyte, and
techniques, such as nanosphere lithography, based on the use of
also fail to provide the necessary nanometer-scale gaps between
choice for large-scale manufacturing of SERS substrates.
However, as we show in this paper, at the ∼150 nm feature size,
desirable for SERS detection. Rather, we report on a hybrid
method that utilizes first NIL to achieve coarse (∼400 nm)
reliable high enhancement factors for Raman scattering.
Patterned Au and Ag structures were fabricated on thermally
oxidized (100 nm SiO2) silicon substrates with NIL, which is a
that is either thermally or UV cured. NIL is ideally suited to
define nanoscale patterns over large fields and with high
* To whom correspondence should be addressed. E-mail: zhiyong.li@
‡Advanced Materials Process Lab, Hewlett-Packard Co.
§Advanced Diagnostic Lab, Hewlett-Packard Co.
(1) Nie, S.; Emory, S. R. Science 1997, 275, 1102.
(3) McCall, S. L.; Platzman, P. M.; Wolff, P. A. Phys. Lett. 1980, 77, 381-
(4) Moskovits, M. J. Chem. Phys. 1978, 69, 4159.
(5) Jiang, J.; Bosnick, K.; Maillard, M.; Brus, L. J. Phys. Chem. B 2003, 107,
(6) Wang, H.; Levin, C. S.; Halas, N. J. Am. Chem. Soc. 2005, 127, 14992.
(7) Tao, A.; Kim, F.; Hess, C.; Goldberger, J.; He, R.; Sun, Y.; Xia, Y.; Yang,
P. Nano Lett. 2003, 3, 1229-1233.
(8) Haes, A. J.; Haynes, C. L.; McFarland, A. D.; Schatz, G. C.; Van Duyne,
R. P.; Zou, S. MRS Bull. 2005, 30, 368.
(9) Schift, H.; Heyderman, L. J. In AlternatiVe Lithography: Unleashing the
Potentials of Nanotechnology; Sotomayor, C. M., Ed.; Kluwer Academic:
Dordrecht, The Netherlands, 2003; p 50.
(10) Chou, S. Y. In AlternatiVe Lithography: Unleashing the Potentials of
Nanotechnology; Sotomayor, C. M., Ed.; Kluwer Academic: Dordrecht, The
Netherlands, 2003; p 15.
Langmuir 2007, 23, 5135-5138
10.1021/la063688n CCC: $37.00© 2007 American Chemical Society
Published on Web 03/27/2007
resolution and throughput.11,12In this work we used a SiO2on
Si mold fabricated by a combination of laser interference
lithography and reactive ion etching.13The utilization of
interference lithography for mold fabrication has the advantage
PMMA at 130 °C and a pressure of 200 psi. The patterning of
PMMA only required ∼5 min. After imprinting, the substrates
were etched for 20 s in an oxygen plasma at 50 W to remove
residual PMMA; this was followed by evaporation of either 6
nm Cr/60 nm Au or 6 nm Ti/60 nm Ag. The film thickness was
monitored with a quartz crystal monitor and was independently
verified by cross-sectional scanning electron microscopy. The
bath. Parts a, b, and e of Figure 1 show SEM images of the
size of 150 nm in diameter and separation of 400 nm. Oxygen-
plasma treatment of the Ag and Au patterned structures was
of 0.5 Torr, 50 W power, for a duration of 1 min. Figure 1c is
the same patterned Ag structure after the oxygen-plasma
of the high-resolution SEM image (Figure 1d) revealed that the
interconnected nanoparticles 40 ( 20 nm in diameter, which is
typical of the grain size observed for 60 nm Ag films.14
Ag nanodots of average size of 10 nm and of average height of
etched by Ar ions, suggesting that Ar-plasma etching may also
patterned structures were not observed in the Au case, as shown
in parts e and f of Figure 1 for before and after the oxygen-
structures into distinguishable nanoparticles was likely due to
believe that the lack of large grain boundaries and the stronger
metallic bonding energy in Au are why its patterned structures
did not undergo the same fracturing.
On the other hand, the mechanism of formation for the Ag
nanodots was initially a mystery because of the difficulty in
resolving them even with a high-resolution scanning electron
microscope and also because they did not appear immediately
after the oxygen-plasma treatment. We compared Auger spectra
on the SiO2substrate area of the Ag samples immediately after
oxygen-plasma treatment and discovered the presence of Ag, as
shown in Figure 2. Despite the lack of any nanodots observable
by SEM, a silver film was redeposited from the patterned Ag
the film, which was likely to be composed of Ag2O during its
initial formation, decomposed into Ag on the substrate and left
a thin film of Ag behind. Like all metals, Ag has a very high
surface energy. However, unlike other metals, Ag has a high
surface mobility even at room temperature. It took less than a
by this plasma treatment was measured to be approximately 20
nm from the high-resolution SEM images.
(11) Chou, S. Y.; Krauss, P. R.; Renstrom, P. J. Science 1996, 272, 85.
J. O.; Nielsen, K. A.; Stoddart, J. F.; Williams, R. S. Nanotechnology 2003, 14,
(13) Brueck, S. R. J. Proc. IEEE 2005, 93, 1704.
(14) Maqbool, M.; Khan, T. Surf. ReV. Lett. 2005, 12, 759-766.
Figure 1. SEM images of (a) as-patterned Ag structures by
imprinting lithography, (b) a magnified view of the patterned Ag
structures as shown in (a), (c) an oxygen-plasma-treated sample,
and (d) a magnified view of the oxygen-plasma-treated sample,
showing the presence of small nanodots. SEM images of a gold
sample before (e) and after (f) the same oxygen-plasma treatment
are shown for comparison.
Figure 2. (a) Representative Auger spectra of the SiO2substrate
area from the red and blue square regions in (b) and (c), showing
were not observed. The oxygen peak came from the silicon oxide
energy of the samples before and immediately after oxygen-plasma
5136 Langmuir, Vol. 23, No. 9, 2007Li et al.
analyzed by X-ray photoelectron spectroscopy. As shown in
composition was not changed by the oxygen-plasma treatment.
Interestingly, the oxygen-plasma-treated sample showed much
structure (see the Ag 3d high-resolution spectra in the inset of
Figure 3), while much less intense oxygen 1s and silicon 2s and
2p peaks were observed. This can be attributed to the silicon
plasma treatment. A trace amount of fluorine 1s peak in the
spectrum of the oxygen-plasma-treated Ag sample was likely
diminished after the sample was simply rinsed with ethanol.
For SERS evalulation, the samples were soaked in 1 µM
4-mercaptophenol in ethanol solution for 10 min, rinsed with
fresh ethanol, and blown dry with nitrogen. The Raman spectra
of the samples were collected on a Horiba-Yvon T64000 micro
Raman system, with three laser excitation sources at 532, 633,
and 785 nm. The intensity of the laser incident on the sample
surface was measured to be about 2.1, 1.2, and 2.2 mW for 532,
633, and 785 nm lasers, respectively, over a focused area of
Raman shift at 520 cm-1, assigned to the Si substrate phonon,
served as an internal standard for the calibration of both the
treated Ag patterned structures consistently yielded highly
for all three excitation wavelengths, as shown in Figure 4. In
comparison, the as-patterned Ag samples barely generated any
detectable signal associated with mercaptophenol molecules
whereas the silicon peaks were of comparable intensity for both
the as-patterned Ag samples and the plasma-treated samples.
as compared to those for the oxygen-plasma-treated sample.
To understand the relative contributions to the enhancement
by performing the oxygen treatment on the film before the lift-
off, thereby producing only the fractured nanoparticles without
the nanodots on the substrate. This substrate demonstrated very
weak Raman activity, thus suggesting that the enhancement
mainly originates from the nanodots on the substrates.
we also collected the Raman spectra of neat 4-mercaptophenol
in a quartz liquid cell using the same micro Raman system at
three excitation wavelengths. The enhancement factor was
power density for both the SERS substrate and neat sample,
respectively, and MSERS and Mneat are the total numbers of
molecules exposed under the laser spot.
on the Ag surface of 1013/cm2, we obtained an enhancement
factor of 6 × 109, 1 × 1010, and 7 × 1010for the peak at 1078
cm-1at 532, 633, and 785 nm excitation, respectively. An area
showed a standard deviation of 37% of the enhancement factor
in a 100 µm × 100 µm region, indicating a good uniformity of
the SERS effect in such a substrate. The stability of the plasma-
treated Ag substrate was also evaluated by storing the plasma-
treated substrate in an ambient lab environment over a month
on the aged sample.
The enhancement factor is partly contributed by an intense
electromagnetic field generated by the surface plasmons; large
signal enhancement is normally achieved at the excitation
Figure 3. XPS survey spectra of the as-patterned (red) and the
oxygen-plasma-treated (blue) Ag patterned structures. The inset
shows the high-resolution spectra of the Ag 3d region from the two
Figure 4. Raman spectra of 4-mercaptophenol adsorbed onto the
(middle), and 532 nm (bottom).
Formation of Ag Nanodots for SERS Substrates Langmuir, Vol. 23, No. 9, 2007 5137
coincidentwiththesurface-plasmonresonancefrequency,which Download full-text
is in turn determined by the size, geometry, and separation of
the Ag nanostructures.4,15-17For isolated Ag nanoparticles, the
factor for single Ag spheres was found for particles with a 20-
25 nm radius.18,19Only when there is a 1 nm separation between
aggregated Ag nanoparticles,5the enhancement factor can be as
high as 1010to 1012. For our nanodots induced by plasma
by large gaps (>20 nm), the local surface plasmon effect alone
with long-range photonic interactions, can generate a large
enhancement factor of 1013. The contribution other than the
nanodots themselves in the plasma-treated periodic Ag pattern
substrate is currently under investigation.
a SERS substrate with reliable and uniform ultrahigh enhance-
ment. This method was based on the simple oxygen-plasma-
induced formation of Ag nanoparticles and nanodots from Ag
patterned structures fabricated economically with nanoimprint
lithography. While we observed both the fracturing of the Ag
structures into nanoparticles and the redeposition of the Ag
of formation for the nanodots is believed to be the coalescence
treatment. The size and separation of the nanodots were dictated
by the diffusion-limited aggregation of Ag atoms.
Acknowledgment. Sandia is a multiprogram laboratory
operated by Sandia Corp., a Lockheed-Martin company, for the
U.S. Department of Energy National Nuclear Security Admin-
istration under Contract DE-AC04-94AL85000. The facilities
of the NSF-sponsored NNIN at the University of New Mexico
were used for the imprint master fabrication.
C. A.; Pond, S. J. K.; Marder, S. R.; Perry, J. W. J. Phys. Chem. B 2002, 106,
Sci. U.S.A. 2003, 100, 8638-8643.
(17) Yang, W. H.; Schatz, G. C.; Van Duyne, R. P. J. Chem. Phys. 1995, 103,
(18) Messinger, B. J.; von Raben, K. U.; Chang, R. K.; Barber, P. W. Phys.
ReV. B 1981, 24, 649.
(19) Wokaun, A.; Gordon, J. P.; Liao, P. F. Phys. ReV. Lett. 1982, 48, 957.
(20) Zou, S.; Schatz, G. C. Chem. Phys. Lett. 2005, 403, 62.
5138 Langmuir, Vol. 23, No. 9, 2007Li et al.