This article is part of the
Surface Enhanced Raman
Spectroscopy web themed issue
Guest editors: Professors Duncan Graham,
Zhongqun Tian and Richard Van Duyne
All articles in this issue will be gathered together online at
www.rsc.org/sers.
Published on 15 February 2011. Downloaded by TU Berlin - Universitaetsbibl on 31/03/2016 14:45:23.
View Article Online
/ Journal Homepage
/ Table of Contents for this issue
This journal is cThe Royal Society of Chemistry 2011 Chem. Commun., 2011, 47, 3553–3555 3553
Cite this:
Chem. Commun
., 2011, 47, 3553–3555
Functionalized Ag nanoparticles with tunable optical properties for
selective protein analysiswz
Arumugam Sivanesan, H. Khoa Ly, Jacek Kozuch, Murat Sezer, Uwe Kuhlmann,
Anna Fischer and Inez M. Weidinger*
Received 19th November 2010, Accepted 31st January 2011
DOI: 10.1039/c0cc05058j
We present a preparation procedure for small sized biocompatibly
coated Ag nanoparticles with tunable surface plasmon
resonances. The conditions were optimised with respect to the
resonance Raman signal enhancement of heme proteins and to
the preservation of the native protein structure.
Surface plasmon resonances (SPR) of noble metal nano-
particles are able to enhance the oscillating electric field of
incident light. This near-field enhancement can be used to
selectively detect and analyse molecules in the vicinity of the
metal surface. SER (surface enhanced Raman) spectroscopy is
the most prominent analytical technique exploiting this effect
since it can provide structural information about the adsorbates.
The sensitivity of this approach can be further increased for
molecules with electronic transitions in the visible spectral
range. In this case, optimum enhancement of the Raman
signals can be achieved if the excitation line is in resonance
with both the SPR and the electronic transition of the
immobilised molecules (SERR—surface enhanced resonance
Raman). Thus, the central challenge is to tune the optical
properties of metal nanoparticles such that the SPR coincides
with the electronic transition of the adsorbate. The SPR position
of noble metal nanoparticles such as Ag and Au can be
controlled by variation of particle size, shape and dielectric
encapsulation.
1,2
However, the spectral tuning of Au is restricted
to wavelengths larger than 500 nm
1
such that this metal is not
applicable for SERR spectroscopy of chromophores in the blue
and violet regions. Ag nanoparticles (AgNPs) exhibit plasmon
resonances up to the near UV and show even a higher intrinsic
field enhancement than Au. However, the synthesis of stable
monodisperse AgNPs is much more difficult to achieve.
In view of the steadily growing importance of SERR
spectroscopy in biological application, the preservation of
the native structure constitutes an additional requirement
on the design of nanoparticles. Biocompatible coatings on
nanoparticles that may protect proteins against direct and
usually harmful interactions with the metal can be obtained
by self-assembled monolayers (SAMs) of o-functionalised
mercaptoalkanes.
3
Again, this method has been established for Au but, up to
now, only one SERR spectroscopic study of proteins on
SAM-coated AgNPs has been reported.
4
However, in that work
rather unstable AgNP aggregates were used. In addition, the
yield of SAM coating was very low and no information about the
integrity of the immobilised protein could be provided.
In the present work, we have thus established a procedure to
prepare small sized SAM-coated AgNPs, starting with citrate-
capped seeds that were subsequently grown for controlled size
adjustment
5
and post-functionalised for biocompatible protein
attachment. Using cytochrome c(Cyt-c), a well-characterized
model protein frequently used in SERR spectroscopy,
6
we
have systematically investigated how selective sensitivity of the
particles can be improved and structural integrity of immobilised
proteins is ensured. The optical properties of the AgNPs were
tuned such that their plasmonic resonance matches the Soret
transition of the Cyt cheme cofactor at 410 nm.
7
First, citrate-capped AgNPs were prepared by borohydride
reduction of Ag
+
ions in the presence of citrate (Fig. S1, ESIz).
The UV-vis spectrum displays a sharp SPR peak at 390 nm,
indicating that no aggregates are formed. The NPs were almost
spherical with an aspect ratio, defined as the ratio between the
longer and the shorter particle axis, of R=1.040.005. For the
longer axis a mean diameter of 12 0.4 nm was determined
(Fig. S2, ESIz). These particles were subsequently used as seeds
forpreparingAgNPsoflargersizevia the seeding growth
method:
8
AgNO
3
was added to the seed solution containing
citrate-capped AgNPs and ascorbic acid (AA) as the reducing
agent. The growth of the particle could be controlled by the
concentrations of AgNO
3
and AA (Fig. S2 and S3, ESIz). The
AgNPs were functionalised by stirring a solution containing
citrate-capped AgNPs and mercaptoundecanoic acid (MUA) at
different concentrations over night. In the last step, Cyt cwas
added to a final concentration between 0.05–1.3 mM. SERR
spectra of Cyt cwere measured with the 413-nm excitation line
of a Kr
+
laser using a confocal Raman spectrometer or a
spectrograph operating in a 901scattering geometry (for further
experimental details, see ESIz).
Technische Universita
¨t Berlin, Institut fu
¨r Chemie, Sekr. PC 14,
Straße des 17. Juni 135, D-10623 Berlin, Germany.
Tel: +49 3031422780
wThis article is part of a ChemComm web-based themed issue on
Surface Enhanced Raman Spectroscopy.
zElectronic supplementary information (ESI) available: Experimental
details and additional TEM and spectroscopic data for further AgNP
characterisation. See DOI: 10.1039/c0cc05058j
ChemComm Dynamic Article Links
www.rsc.org/chemcomm COMMUNICATION
Published on 15 February 2011. Downloaded by TU Berlin - Universitaetsbibl on 31/03/2016 14:45:23.
View Article Online
3554 Chem. Commun., 2011, 47, 3553–3555 This journal is cThe Royal Society of Chemistry 2011
The SPR position of a given nanoparticle ensemble can be
monitored by UV-visible absorption spectroscopy. Fig. 1 (top)
shows the UV-vis spectra of 4 different nanoparticle batches
obtained with different growth solutions. As a consequence of
the particle seeding and coating, the SPR is different for each
batch, i.e. at 403, 413, 418 and 427 nm. The red shift from
390 nm to 403 nm can solely be attributed to the MUA coating
that increases the local dielectric constant of the surrounding
medium. Correspondingly for all nanoparticle batches a shift
of 13 nm was observed after MUA addition. Further SPR
tuning to 413, 418 and 427 nm was achieved by adding AgNO
3
of different concentrations to the nanoparticle seed solution
which led to an increase in average size and aspect ratio
(Fig. S3, ESIz). Both effects are generally assumed to cause
a red shift of the SPR position.
1,2,9
Addition of Cyt cto the AgNP solution leads to a further
red shift of the SPR by 8–10 nm (Fig. S3, ESIz), indicating that
the protein was effectively adsorbed on the NP surface. In all
cases, a time-dependent drop in plasmon absorption intensity
was observed which is attributed to a slow aggregation of the
nanoparticles upon Cyt cbinding. Surprisingly the Ag
413
/Cyt c
complex was stable even after 90 min whereas the Ag
427
nanoparticle batch completely lost plasmonic activity already
15 min after Cyt caddition.
Fig. 1 (top squares) also shows the normalised SERR
intensity of Cyt cadsorbed on the various MUA-coated
AgNPs. The highest intensity is obtained for the Ag
413
NPs
where the SPR maximum coincidences with the excitation
energy. The SERR intensity strongly decreases for NPs with
SPR maxima at lower and higher wavelengths. In the latter case,
namely for Ag
418
and Ag
427
, the weaker signal enhancement may
be mainly due to the strong tendency of the particles to aggregate
upon Cyt caddition. Nevertheless, the results clearly show that a
good SERR intensity can only be achieved in a small spectral
window where the SPR matches the molecular electronic
transition of the heme cofactor of the protein.
Citrate-reduced AgNPs have been shown to afford high
SERR signal intensities of the adsorbed Cyt c.
10
However, the
high charge density of the citrate layer causes irreversible
denaturation of the protein which, in the case of Cyt c,is
reflected by the transition from the native six-coordinated low
spin (6cLS) configuration to a five-coordinated high spin
(5cHS) species in the SERR spectrum.
6
Replacing citrate by
MUA is expected to improve the stability of the native protein
structure. This is in fact confirmed by the present results
although the preservation of the protein structure depends
on the conditions of SAM formation, specifically on the MUA
concentration used for citrate/MUA exchange (Fig. 2). Using
a molar MUA/AgNP ratio of ca. 10
4
, the citrate/MUA
exchange is not complete as concluded from the residual
contribution of the 5cHS species in the SERR spectrum which
is reflected by the characteristic marker band at 1490 cm
1
(n
3
). Upon increasing the MUA/AgNP ratio to 10
6
, the SERR
spectrum of the bound Cyt cexhibits essentially the same
vibrational signature as the resonance Raman (RR) spectrum
of Cyt cin solution. The preservation of the proteins native
structure, however, is achieved at the expense of the SERR
Fig. 1 Top: UV-Vis spectra of MUA–AgNPs with l
max
= 403
(violet), 413 (red), 418 (blue) and 427 (green) nm. SERR intensity
(squares) of the maximum Raman band (n
4
) of adsorbed Cyt cas a
function of l
max
. Inset: TEM picture of MUA–AgNP with SPR at
413 nm. Bottom: UV-vis spectrum of Cyt cat the Soret band region
(solid line). Inset: SERR spectrum of Cyt c.
Fig. 2 (A) RR spectrum of Cyt cin solution (20 mM) compared with
the SERR spectra of Cyt cbound to SAM-coated AgNPs (1 mM Cyt c)
using MUA/AgNP ratios of ca. 10
4
(B) and 10
6
(C).
Published on 15 February 2011. Downloaded by TU Berlin - Universitaetsbibl on 31/03/2016 14:45:23.
View Article Online
This journal is cThe Royal Society of Chemistry 2011 Chem. Commun., 2011, 47, 3553–3555 3555
intensity which is ca. one order of magnitude lower than for
the citrate-capped AgNPs due to the larger separation of the
protein from the Ag surface.
The surface coverage of Cyt cwas determined by monitoring
SERR intensity of Cyt cfor different protein concentrations in
solution. The relative surface concentration is given by:
y=G/G
S
,whereGand G
S
refer to the surface coverage at a
particular Cyt cconcentration and to the saturation surface
coverage, respectively. ywas calculated by y=I
SERRS
/I
max
SERRS
and plotted as a function of initial Cyt cconcentration c
0
in
solution (Fig. 3) . At low c
0
(o1mM), yincreases linearly with c
0
butitlevelsoffforc
0
41mM. The data could be well described
by a fit of a Langmuir adsorption isotherm:
y¼Kc
1þKcð1Þ
where Kis the equilibrium constant and c=c
0
yG
s
corresponds to the Cyt concentration in solution after adsorption
equilibrium is established. The fit of eqn (1) to the data in Fig. 3
affords G
s
=0.86mMandK=1.210
8
M
1
(Fig. S4, ESIz).
Additional UV-vis spectroscopic determination of the solution
concentration of Cyt ccarried out with the supernatants of
centrifuged MUA capped AgNP suspensions (Fig. S5, ESIz)
yields essentially the same result for G
s
. Approximating an ideal
spherical shape for the AgNPs and a diameter of ca. 17 nm we
can estimate the NP concentration to be ca. 11 nM, leading to a
surface coverage of roughly 80 Cyt cmolecules per nanoparticle
in the saturation limit. If we consider that the particle diameter is
increased by ca. 4 nm due to the MUA coating the surface
concentration of Cyt cisestimatedtobe10pMcm
2
which is in
good agreement with the data obtained in a previous work on
MUA-coated Au electrodes.
11
On the basis of these data, the
Raman enhancement factor (REF) is determined according to
REF ¼ISERRS cRR
IRR cSERRS kð2Þ
where c
RR
is the Cyt cconcentrationusedintheRRexperiment,
c
SERR
refers to the concentration of the adsorbed Cyt cin the
SERR experiments, determined according to eqn (1), and kis
a shielding factor that is assumed to be 0.25 as discussed
previously.
10
For determining the SERR and RR intensities of
the 1373 cm
1
band (I
SERR
,I
RR
), the measurements were carried
out with the AgNP/Cyt csuspensions and the Cyt csolution
adjusted to the same optical density at the excitation wavelength.
Thus, REF was determined to be (1.3 0.4)10
2
.Takinginto
account that the MUA coating causes an attenuation of
the surface enhancement by ca. one order of magnitude, the
enhancement factor for the protein directly attached to the metal
would be ca. 10
3
. This factor is the product of the enhancement
of the intensity of the incident and Raman scattered radiation at
l
exc
and l
Raman
, respectively. For the Ag
413
NPs considered here,
the surface plasmon absorption exhibits its maximum at l
exc
but
it has dropped to ca. 0.86 of this value at l
Raman
. Assuming that
the enhancement intensity scales with the surface plasmon
absorption, the enhancement factor for the incident electric field
at l
exc
is estimated to be 37. This value is in very good agreement
with the theoretically predicted enhancement of ca. 40 according
to Zeman and Schatz.
9
In conclusion, we have shown that by precise tuning
of the plasmonic properties of AgNPs, SERR detection of
specific cofactors in proteins is possible down to nanomolar
concentrations. Coating of AgNPs by SAMs with appropriate
head groups allows for binding of proteins under preservation
of the native structure.
The authors would like to thank Peter Hildebrandt for
helpful discussions and Soeren Selve from the ZELMI TU
Berlin for the TEM pictures. Financial support from the
Fonds der Chemie (IW), the Alfried Krupp Wissenschaftskolleg,
Greifswald (AS) and the DFG (Unicat) is gratefully
acknowledged.
Notes and references
1 S. Link and M. A. El-Sayed, J. Phys. Chem. B, 1999, 103,
8410–8426.
2 K. L. Kelly, E. Coronado, L. L. Zhao and G. C. Schatz, J. Phys.
Chem. B, 2003, 107, 668–677.
3 G. H. Woehrle, L. O. Brown and J. E. Hutchison, J. Am. Chem.
Soc., 2005, 127, 2172–2183.
4 A. Bonifacio, L. V. D. Sneppen, C. Gooijer and G. V. D. Zwan,
Langmuir, 2004, 20, 5858–5864.
5 K. R. Brown and M. J. Natan, Langmuir, 1998, 14, 726–728.
6 S. Oellerich, H. Wackerbarth and P. Hildebrandt, J. Phys. Chem.
B, 2002, 106, 6566–6580.
7 D. H. Murgida and P. Hildebrandt, Acc. Chem. Res., 2004, 37,
854–861.
8 N. R. Jana, L. Gearheart and C. J. Murphy, Langmuir, 2001, 17,
6782–6786.
9 E. J. Zeman and G. C. Schatz, J. Phys. Chem., 1987, 91, 634–643.
10 P. Hildebrandt and M. Stockburger, J. Phys. Chem., 1986, 90,
6017–6024.
11 S. Song, R. A. Clark, E. F. Bowden and M. J. Tarlov, J. Phys.
Chem., 1993, 97, 6564–6572.
Fig. 3 Relative Cyt csurface coverage as a function of protein
concentration in solution.
Published on 15 February 2011. Downloaded by TU Berlin - Universitaetsbibl on 31/03/2016 14:45:23.
View Article Online
