PAPER www.rsc.org/materials | Journal of Materials Chemistry

Dow

nloa

ded

by U

nive

rsity

of

Que

ensl

and

on 1

7/04

/201

3 10

:07:

38.

Publ

ishe

d on

16

Sept

embe

r 20

09 o

n ht

tp://

pubs

.rsc

.org

| do

i:10.

1039

/B91

3478

FView Article Online / Journal Homepage / Table of Contents for this issue

Exploiting optothermal conversion for nanofabrication: site-selectivegeneration of Au/TiO2 inverse opals†

Ivano Alessandri*a and Matteo Ferronib

Received 7th July 2009, Accepted 22nd August 2009

First published as an Advance Article on the web 16th September 2009

DOI: 10.1039/b913478f

Optothermal conversion mediated by gold nanoparticles was exploited to generate localized Au/TiO2

inverse opal structures in direct opal substrates. The local enhancement of the electromagnetic field

made these nanostructures active sites for SERS and contributed to boosting the efficiency of some

reactions, such as the photodegradation of methylene blue, under both UV and Vis irradiation.

1. Introduction

Metal nanoparticles (NPs) are attracting ever-increasing interest

in nano- and bio-technology due to the synergic combination of

their versatile surface chemistry and optical properties, which

may be controlled directly by synthesis.1,2 In addition, metal NPs

can be exploited as very efficient photon-thermal converters to

generate localized heat at the micro- and nanoscale (optothermal

conversion).3 In particular, the heating effect is strongly

enhanced when the frequency of the incident light matches the

plasmon resonance of the metal NPs, so that it is commonly

referred to as ‘‘plasmonic heating’’.4 This key property is

currently being investigated for a number of important applica-

tions in various research fields, including drug delivery,5 cancer

diagnostics and therapy.6,7 Further important sectors that can

benefit from plasmonic heating are micro- and nanofabrication4

and plasmon-assisted chemical vapour deposition.8,9 Recently,

we used Au NPs as light harvesting centers to bring extremely

localized heating into colloidal particles and colloidal assemblies,

obtaining a selective modification of their morphology.10 More-

over, we demonstrated how plasmonic heating can be harnessed

and conveniently employed to yield ‘‘hot’’ sites for surface

enhanced Raman spectroscopy (SERS) which were based on

in-situ generated metal oxides.11,12 Optothermal conversion

becomes actually useful for nanofabrication when one can fully

take advantage of some of its key features, such as precise spatial

localization of heating effects and selective activation of specific

areas which have been properly functionalized.

In the present work optothermal conversion is exploited to

generate localized inverse opal structures into direct opal

substrates. Opals and inverse opal structures have been inten-

sively investigated in view of achieving photonic crystals with 2-

and 3D photonic bandgaps that can be tuned depending on the

aINSTM and Chemistry for Technologies Laboratory, University ofBrescia, via Branze 38, 25123 Brescia, Italy. E-mail: ivano.alessandri@ing.unibs.itbCNR-INFM and Department of Chemistry and Physics for Engineeringand Materials, University of Brescia, via Valotti 9, 25123 Brescia, Italy

† Electronic supplementary information (ESI) available: SEM image(enlarged view) of the resulting structures, microRaman monitoring ofUV-induced photodegradation of MB onto a 50 nm thick anatase filmdeposited onto a planar support, and microRaman monitoring of MBphotodegradation upon laser irradiation at the power of 0.05 mW. SeeDOI: 10.1039/b913478f

7990 | J. Mater. Chem., 2009, 19, 7990–7994

colloid size.13 Mallouk and co-workers14 and, more recently,

Corma and co-workers15 studied the photonic crystal topology

for applications in photoelectrochemical solar cells. Ozin’s group

proposed to exploit stop-band reflections to achieve optically

amplified photochemistry.16 This goal can be reached since, at

the frequency edges of these stop bands, photons propagate with

strongly reduced group velocity, so that the effective optical path

length of the system can be significantly enhanced. Chen et al.

demonstrated that the photoactivity of anatase TiO2 inverse

opals is remarkably enhanced by using slow photons with ener-

gies close to the electronic bandgap of the semiconductor.16 In

particular, they showed that TiO2 inverse opals, generated as

inverse replicas of colloidal crystals made of polystyrene nano-

spheres 150 nm in diameter, exhibited a significant enhancement

in photodegrading methylene blue (MB) under white light irra-

diation in comparison to conventional nanocrystalline TiO2. In

a following paper,17 the same authors demonstrated that

a certain degree of microstructural disorder, which is inherently

associated with these materials, can be tolerated without pre-

venting their use for some important applications, such as puri-

fication of water from environmental pollutants. Another way to

increase the efficiency of oxide-based photocatalysts is by

exploiting the enhancement of electromagnetic field that takes

place in the presence of metal NPs, as proposed by Awazu et al.18

The synergic combination of amplified photocatalytic properties

of metal oxide inverse opals with localized surface plasmons of

metal NPs might strongly enhance the efficiency of many reac-

tions occurring under UV-Vis-NIR irradiation. In addition, the

integration of photoactive metal oxide inverse opals into micro-

reactors or smart devices represents an important goal for

nanotechnology. However, the selective generation of metal-

doped inverse opals and extended defects into functional oxide

matrices (e.g. TiO2, ZnO, etc.) still remains a challenging task.

Here we demonstrate that Au NPs can be used as nanoheaters in

order to open up macroporous windows at selected areas of PS/TiO2

core/shell colloidal crystals. The resulting Au/TiO2 inverse opals

were active sites for SERS and exhibited enhanced photocatalytic

properties when illuminated under both UV and Vis conditions.

2. Experimental

The colloidal crystals formed by PS-Au/TiO2 nanospheres were

prepared as follows. Monodisperse PS nanospheres (diameter:

This journal is ª The Royal Society of Chemistry 2009

Dow

nloa

ded

by U

nive

rsity

of

Que

ensl

and

on 1

7/04

/201

3 10

:07:

38.

Publ

ishe

d on

16

Sept

embe

r 20

09 o

n ht

tp://

pubs

.rsc

.org

| do

i:10.

1039

/B91

3478

F

View Article Online

150 � 6 nm) suspended in milliQ water (10% wt) were assembled

as multilayered colloidal crystals on both Si (100) and micro-

scope glass slides by either sedimentation or crystallization under

capillary forces. The substrates were previously UV-cleaned for 5

minutes to remove organic contaminants from the surface and

improve hydrophilicity. The colloidal crystals were coated by

sputtering with a thin layer of gold (about 12 nm thick), con-

sisting of nanoislands agglomerated in the form of gold nano-

particles. Details on Au deposition can be found in reference 10.

These Au nanoshells act as active substrates (nanoheaters) for

triggering the formation of the inverse opals. The 50 nm thick

amorphous titania layer was deposited in a Savannah 100 ALD

flow reactor (Cambridge Nanotech Inc., MA), using TDMAT

(tetrakis-dimethylamido titanium, Aldrich, Germany) as the

titanium source and water as the oxygen source. The deposition

temperature and pressure were respectively 90 �C and 0.5 Torr.

TDMAT (99.999%) and H2O were evaporated from stainless

steel reservoirs held respectively at 80 �C and at room tempera-

ture, and led into the reactor through solenoid valves. Nitrogen

was used as a precursor carrier and purge gas. The processing

cycle consisted of a TDMAT pulse of 0.1 s, a 5 s purge pulse of

N2, a 0.1 s pulse of water vapour and a 5 s pulse of N2. The

deposition rate was 0.0667 nm per cycle. To generate Au/anatase

inverse opals, the colloidal substrates were set upon the motor-

ized x-y stage (resolution: 0.1 mm, scanning range: 100 �100 mm) of a high resolution Raman microscope (Horiba/Jobin-

Yvon). The Au/TiO2 inverse opals were obtained by focussing

a He–Ne (l ¼ 632.8 nm) laser through 50� long working

distance (numerical aperture, NA: 0.5) and 100� (NA: 0.9)

objectives with different dwell times.10,12 The inverse opal

nanostructures reported in Fig. 3 (see later) were obtained by

irradiating for 10 s with a 100� objective. The power of the laser

at the surface of the sample was about 4.5 mW. MicroRaman

spectra of the Au/TiO2 were acquired using optical filters that

allowed the power of the laser to attenuate to 0.05 mW. The UV

photocatalytic degradation experiment was performed by drop-

ping 1 mL of a 10�6 M aqueous solution of methylene blue (MB,

C16H18N3SCl) onto the area of Au/TiO2 inverse opals upon

direct observation by means of an optical microscope. The

adsorption of the MB on the anatase surface was checked by

Raman spectroscopy (see the text for details). The photo-

degradation was carried out by irradiating the sample with

a Philips UV lamp, that emits from 340–410 nm, with a peak

maximum at 365 nm. The distance between the lamp and the

sample was 1 cm. The sample was directly UV irradiated without

moving the Raman stage, to ensure that all the Raman spectra

were acquired on the same zone. The scanning electron micros-

copy (SEM) images were recorded using a LEO 1521 high

resolution instrument equipped with a field emission gun and an

in-lens secondary electron detector. The SEM was operated in

the 1–2 keV beam energy range to prevent the specimen from

electrostatic charging and from excessive beam induced damage

and shrinking.

Fig. 1 Schematic of Au/TiO2 inverse opal generation by optothermal

conversion. Colloidal crystals made of PS-Au/TiO2 nanoshells are irra-

diated by a low power C.W. He–Ne laser focused onto a specific region

through the objective of a microscope. Upon irradiation, inverse opal

nanostructures are created by efficient photon-thermal conversion

mediated by Au nanoparticles. The lateral resolution that can be achieved

is less than 2 mm.

3. Results and discussion

The experimental procedure for generating Au/TiO2 inverse

opals (i.o.) localized at specific sites is displayed in Fig. 1.

Colloidal crystals (opals) made of polystyrene (PS) nanospheres

This journal is ª The Royal Society of Chemistry 2009

(diameter: 150 nm) were first coated with a 12 nm thick layer of

gold nanoparticles (Au NPs) and then with a 50 nm thick layer of

amorphous titania (a-TiO2) obtained by atomic layer deposition.

Selected regions of these opals were irradiated by a continuous

wave (C.W.) He–Ne laser (lmax: 632.8 nm) focused through the

objectives of a Raman microscope. The power of the laser was

4.5 mW. Details on preparation are reported in the experimental

section. As a general result, the irradiation yielded Au/TiO2

inverse opals whose size could be tuned by changing different

parameters such as time of exposure and laser penetration depth.

Moving the laser using a motorized x-y stage allows inverse opal

structures to be localized and patterned at selected areas with

a lateral resolution of about 2 micrometers.

Fig. 2a shows a lateral view of PS-Au/TiO2 opals before laser

irradiation. The colloidal crystals were formed by about 20 close

packed layers of PS nanospheres. This thickness was the same as

has been employed in other experiments aiming to exploit slow

photons for photocatalysis.16 Fig. 2b shows that the optical

extinction of these opals is characterized by two main bands

ranging from 300 to 400 and from 500 to 900 nm and centered at

about 340 and 640 nm, respectively. The first band is mainly due

to the absorption of a-TiO2, as shown by comparison with the

absorbance spectrum of a 50 nm thick a-TiO2 taken as a refer-

ence. According to Chen et al., the stop band of the resulting

inverse opal nanostructures that will be generated by laser irra-

diation should be located in this range of frequencies.16 The

second broad band, which is extended from the visible to the near

infrared range, is due to the plasmon resonance characteristic of

Au nanoshell-like structures formed by the PS–Au composite

nanospheres. In the present case the maximum of this peak

matches the wavelength of the exciting He–Ne laser, so that the

photon-thermal conversion process occurs under optimal

conditions. The most interesting aspect of using Au nanoshells as

substrates for optothermal processes is that the plasmonic band

can be shifted and broadened simply by changing the PS/Au

J. Mater. Chem., 2009, 19, 7990–7994 | 7991

Fig. 2 a) SEM section of a PS-Au/a-TiO2 colloidal crystal before laser

irradiation. Scale bar: 200 nm. b) Optical extinction of PS-Au/a-TiO2

colloidal crystals in the UV-Vis range. Dotted line represents the

extinction of a 50 nm thick a-TiO2 reference sample. The wavelength of

the incident laser is marked.

Fig. 3 a) SEM image showing Au/TiO2 inverse opals generated after

laser irradiation. Inset highlights the macroporous structure. Scale bar:

100 nm. An enlarged view of this nanostructure is reported in ESI.† b)

MicroRaman spectrum of the Au/TiO2 inverse opals.Dow

nloa

ded

by U

nive

rsity

of

Que

ensl

and

on 1

7/04

/201

3 10

:07:

38.

Publ

ishe

d on

16

Sept

embe

r 20

09 o

n ht

tp://

pubs

.rsc

.org

| do

i:10.

1039

/B91

3478

F

View Article Online

core/shell ratio, as reported by Halas and co-workers.19 This

allows the effects of plasmonic heating to be tailored depending

on the distance between the wavelength of the exciting laser and

the spectral position of the plasmonic resonance.

Fig. 3a shows an example of the local inverse opals obtained

by irradiating the PS-Au/a-TiO2 colloidal crystals for 10 s with

a 100� (N.A.: 0.9) objective lens. The effects of laser irradiation

encompassed a circular area of about 2 mm in diameter, which

corresponds to the size of the focused laser. The central part

shows a three dimensionally ordered macroporous structure

extended in depth to a few layers of the close packed colloidal

crystal. The mechanism of inverse opal formation is analogous to

that we reported in the case of formation of Au/TiO2 SERS-

active spots.11,12 The laser beam penetrates the amorphous TiO2

and is strongly absorbed by the gold interlayer. Au NPs act as

very efficient photon-thermal converters, so that a large amount

of heat is generated and propagated through the surrounding

medium. PS is a poor thermal conductor, thus heat cannot be

dissipated in an effective way, leading the local temperature to

increase up to the thermal decomposition of the polymeric

spheres. As this process is exothermic, additional heating is

supplied to the system, allowing the amorphous TiO2 to be

converted into the anatase phase (vide infra). Moreover, the

presence of Au may further contribute to the anatase crystalli-

zation at low temperatures.20

The heat generated by the tail of the laser beam, which is

assumed to be Gaussian, is enough only to lead to the sintering of

the nanospheres located at the peripheral region. On the other

hand, at the centre of the beam, the remarkable enhancement of

7992 | J. Mater. Chem., 2009, 19, 7990–7994

the local temperature gives rise to explosions of the polymer

cores, which cause the ejection of the composite material. Part of

the ejected Au re-crystallizes on the surface in the form of

spherical droplets which appeared as bright areas in the SEM

image. The explosive release of gases upon pressure buildup

provoked the partial destruction of the walls interconnecting the

topmost layers of the inverse opals. On the other hand, the

deepest layers exhibit integral, well-interconnected macropores

of about 120 nm in diameter.

In general, greater penetration depth can be obtained by using

lower numerical apertures, as we demonstrated in previous

works carried out on similar systems.10,11

The resonant excitation of surface plasmons allowed strong

heating to be achieved by means of low-power C.W. lasers.

Unlike conventional approaches, which utilize high energy

pulsed lasers to produce direct heating of the substrate, this low-

power, Au-mediated procedure offers higher lateral and depth

resolution which ultimately comes from the localized nature of

the optothermal conversion. In addition, this method allows

operation in a ‘‘direct-writing’’ mode, without needing to use

either lithographic masks or etching procedures.

MicroRaman analysis revealed that laser irradiation caused

the formation of nanocrystalline anatase. Fig. 3b shows the

This journal is ª The Royal Society of Chemistry 2009

Fig. 4 Photodegradation of MB monitored by microRaman spectros-

copy (see the text for details) under a) UV and b) laser irradiation

(0.5 mW).Dow

nloa

ded

by U

nive

rsity

of

Que

ensl

and

on 1

7/04

/201

3 10

:07:

38.

Publ

ishe

d on

16

Sept

embe

r 20

09 o

n ht

tp://

pubs

.rsc

.org

| do

i:10.

1039

/B91

3478

F

View Article Online

presence of signals at about 149, 394, 512 and 636 cm�1, which

correspond, respectively, to the Eg(1), B1g(1), A1g + B1g and Eg(3)

modes characteristic of TiO2 in the anatase phase. The spectrum

was acquired using an optical filter in order to attenuate the laser

power below 0.05 mW and avoid any possible heating-induced

modification during the measurement. The broadening of the

Raman modes, as well as their slight shift compared to bulk

powder samples, suggested that the anatase was present in the

form of nanocrystals and might indicate diffusion of gold into

the anatase layer.

As we already observed for analogous systems,11 it is impor-

tant to note that, under typical conditions, the Raman detection

of such a thin (50 nm or less) anatase layer is prevented. In

contrast, in the present case, the gold aggregates gave rise to

suitable conditions for a strong enhancement of the Raman

signal, so allowing the detection of anatase. This confirmed that

the electromagnetic field was locally increased by laser excitation

of gold surface plasmons, and that a certain amount of Au NPs

remained tethered into the crystallizing anatase. Further in-

depth microstructural characterization is currently under way to

analyze the Au–TiO2 interdiffusion.

The SERS activity of the Au/TiO2 inverse opals can be

exploited for monitoring photo-induced reactions occurring at

these sites. We tested the photodegradation of methylene blue

(MB) under two different sets of operating conditions. On one

hand, a UV-induced photodegradation experiment was carried

out using a lamp emitting in the 340–410 nm range as an exci-

tation source. To avoid possible direct photobleaching of MB

during Raman acquisition the power of the He–Ne laser was

attenuated to 0.005 mW. In the present case, as the main

absorption band of MB falls within the frequency range of

excitation, the surface enhanced Raman experiment is carried

out under resonant conditions (SERRS). This enables one to

gain extra sensitivity compared to the surface-enhanced effect

taken by itself, so allowing the adsorbed dye to be detected at low

concentration (10�6 M, see experimental section).

Photodegradation was monitored by observing the Raman

modes located in the 450–504 cm�1 range which have been

attributed to C–N–C skeletal bending of dimer (450 and

504 cm�1) and monomer (490 cm�1) MB species.

The intensity of MB bands diminished progressively upon UV

irradiation, indicating that the photo-degradation process

occurred successfully at the Au/TiO2 i.o. sites (Fig. 4a). More-

over, photodegradation is faster than that observed for a refer-

ence sample made of a 50 nm thick film of anatase deposited by

ALD on a planar substrate (see ESI†).21 This improved kinetics

may depend on many factors, such as the presence of Au NPs

and the sample morphology, including both increased surface

area and possible enhancement of optical path length due to slow

photons. The latter factor has been demonstrated by Chen et al.

for analogous TiO2 i.o. systems,16 as we reported in the intro-

duction. On the other hand, the role of metal NPs towards

improving charge separation and promoting interfacial charge

transfer kinetics in semiconductor assisted photocatalysis has

been intensively investigated in the literature.22 For example,

Kamat and co-workers demonstrated that, in the case of TiO2/

Au NPs, the Fermi level is shifted to more negative potentials in

comparison to the values of pure TiO2. This accounts for

increased charge storage and improved charge separation

This journal is ª The Royal Society of Chemistry 2009

exhibited by these composite systems.23 Although the TiO2–Au

interface may offer many complex scenarios, photo-

electrochemical and photocatalytic measurements indicate that

the presence of Au NPs promotes charge stabilization in nano-

structured TiO2 and interfacial redox processes. In the present

case, the combination of Au/TiO2 i.o. structural and optical

properties may contribute to enhance photocatalytic efficiency.

Further experiments aiming to elucidate the roles of these

different factors are currently under way.

Another interesting aspect of generating Au/TiO2 i.o. at

selected, micron-sized areas, is related to their possible integra-

tion into microreactors and microdevices. From this standpoint,

these nanostructures might be suitably employed as plasmonic

photocatalysts, enabling important reactions to occur with high

efficiency and minimal light exposure. Thus, in a second exper-

imental layout, the MB photodegradation was tested under

monochromatic irradiation at 632.8 nm. The power of the

exciting laser was set at 0.5 or 0.05 mW. Under these conditions,

reference MB solutions deposited onto glass substrates resulted

stable after prolonged irradiation. In contrast, Fig. 4b shows that

the spectra of MB adsorbed onto the Au/TiO2 sites changed

continuously as photodegradation proceeded. In particular,

J. Mater. Chem., 2009, 19, 7990–7994 | 7993

Dow

nloa

ded

by U

nive

rsity

of

Que

ensl

and

on 1

7/04

/201

3 10

:07:

38.

Publ

ishe

d on

16

Sept

embe

r 20

09 o

n ht

tp://

pubs

.rsc

.org

| do

i:10.

1039

/B91

3478

F

View Article Online

during the first stages of this process, MB species converted from

dimers to monomers (MB2 ¼ 2MB), as indicated by the

appearance of the peak at 490 cm�1, which becomes stronger, and

the corresponding weakening of dimer signals.24 Then, the

photoreaction proceeded until the MB was completely removed,

and no signals could be detected. Although the power of the laser

was set at very low values, the removal of MB through this

process is faster than that occurring under UV irradiation.

Moreover, MB removal was also observed even when the power

of the laser was further reduced to 0.05 mW, as shown in ESI.†

Again, Au plays a key role in enhancing the efficiency of this

process. During the first stages the heating generated by light

absorption under resonance conditions promotes the dimer-to-

monomer conversion, which is an endothermic reaction. There-

after, the monomeric species are completely removed in a few

minutes. As MB photobleaching usually occurs at higher laser

power, in the present case the efficient photon-thermal conver-

sion of Au is a decisive factor for driving the degradation of the

dye. This is an important outcome, as it may allow the produc-

tion of microreactors using low power-solid state lasers as exci-

tation sources.

4. Conclusion

In conclusion, we demonstrated that optothermal conversion

can be exploited to generate nanostructured Au/TiO2 inverse

opals in opal substrates with high spatial resolution. This process

was accomplished by means of a C.W. laser at low power, with

a very simple experimental setup. The localized nature of plas-

monic heating provides a mean for generating inverse opals

without heating the remainder of the substrate, thus enabling the

integration of inverse opals into microreactors and nanodevices.

Gold acted as a very efficient photon-thermal converter, allow-

ing the a-TiO2 to be transformed into anatase. The overall

thermal effect was enhanced by the exothermic decomposition of

the underlying polymer. The enhancement of the electromag-

netic field due to the presence of Au NPs allowed the in-situ

microRaman characterization of the inverse opal nano-

structures. The same effect was exploited both for monitoring

and at the same time assisting reactions occurring at these sites

under optimal conditions of efficiency and light exposure. This

approach may be extended to other composite systems, including

Au (Ag)/ZnO or Au (Ag)/CeO2, thus opening attracting

scenarios for fabricating nanostructures and in situ character-

ization of physical and chemical processes occurring at these

interfaces.25

7994 | J. Mater. Chem., 2009, 19, 7990–7994

Acknowledgements

This work was supported by Fondazione Cariplo. Marcello Zucca

is gratefully acknowledged for assistance in deposition of TiO2.

Prof. Laura E. Depero is acknowledged for valuable discussions.

References

1 C. Burda, X. B. Chen, R. Narayan and M. A. El-Sayed, Chem. Rev.,2005, 105, 1025.

2 N. L. Rosi and C. A. Mirkin, Chem. Rev., 2005, 105, 1547.3 H. H. Richardson, Z. N. Hickman, A. O. Govorov, A. C. Thomas,

W. Zhang and M. E. Kordesch, Nano Lett., 2006, 6, 783.4 M. B. Cortie, N. Harris and M. Ford, Physica B, 2007, 394, 188.5 A. G. Skirtach, A. Mu~noz Javier, O. Kreft, K. K€ohler, A. P. Alberola,

H. M€ohwald, W. J. Parak and G. B. Sukhorukov, Angew. Chem., Int.Ed., 2006, 45, 4612.

6 C. Loo, A. Lowery, N. Halas, J. West and R. Drezek, Nano Lett.,2005, 5, 709.

7 L. R. Hirsch, R. J. Stafford, J. A. Bankson, S. R. Sershen, B. Rivera,R. E. Price, J. D. Hazle, N. J. Halas and J. L. West, Proc. Natl. Acad.Sci. U. S. A., 2003, 100, 13549.

8 D. A. Boyd, L. Greengard, M. Brongersma, M. Y. El-Naggar andD. G. Goodwin, Nano Lett., 2006, 6, 2592.

9 W. H. Hung, I.-K. Hsu, A. Bushmaker, R. Kumar, J. Theiss andS. B. Cronin, Nano Lett., 2008, 8, 3278.

10 I. Alessandri and L. E. Depero, Nanotechnology, 2008, 19, 305301.11 I. Alessandri, M. Ferroni and L. E. Depero, ChemPhysChem, 2009,

10, 1017.12 I. Alessandri and L. E. Depero, Chem. Commun., 2009, 2359.13 (a) G. Ozin, A. Arsenault and L. Cademartiri, Nanochemistry, RSC

Publishing: Cambridge, 2008; (b) A. Stein, F. Li and N. R. Denny,Chem. Mater., 2008, 20, 649.

14 S. Nishimura, N. Abrams, B. A. Lewis, L. I. Halaoui, T. E. Mallouk,K. D. Benkstein, K. van de Lagemaat and A. J. Frank, J. Am. Chem.Soc., 2003, 125, 6306.

15 I. Rodriguez, P. Atienzar, F. Ramiro-Manzano, F. Meseguer,A. Corma and H. Garcia, Photonics Nanostruct.: Fundam. Appl.,2005, 3, 148.

16 J. I. L. Chen, G. von Freymann, S. Y. Choi, V. Kitaev and G. A. Ozin,Adv. Mater., 2006, 18, 1915.

17 J. I. L. Chen, G. von Freymann, V. Kitaev and G. A. Ozin, J. Am.Chem. Soc., 2007, 129, 1196.

18 K. Awazu, M. Fujimaki, C. Rockstuhl, J. Tominaga, H. Murukami,Y. Ohki, N. Yoshida and T. Watanabe, J. Am. Chem. Soc., 2008, 130,1676.

19 S. J. Oldenburg, R. D. Averitt, S. L. Westcott and N. J. Halas, Chem.Phys. Lett., 1998, 288, 243.

20 C. Wang, J. Fu, G. M. Chow, P. C. Hsu and Y. K. Hwu, J. Am.Ceram. Soc., 2005, 88, 758.

21 I. Alessandri, M. Zucca, M. Ferroni, E. Bontempi and L. E. Depero,Small, 2009, 5, 336.

22 M. Jakob, H. Levanon and P. V. Kamat, Nano Lett., 2003, 3, 353.23 V. Subramanian, E. E. Wolff and P. V. Kamat, J. Am. Chem. Soc.,

2004, 126, 4943.24 S. H. A. Nicolai and J. C. Rubim, Langmuir, 19, p. 4291.25 A. V. Whitney, J. W. Elam, P. C. Stair and R. P. Van Duyne, J. Phys.

Chem. C, 2007, 111, 16827.

This journal is ª The Royal Society of Chemistry 2009