An efficient method based on the photothermal effect for the release of molecules from metal nanoparticle surfaces

Full text of DOI:10.1002/anie.200805303

Communications
DOI: 10.1002/anie.200805303
Photothermal Release
An Efficient Method Based on the Photothermal Effect for the Release
of Molecules from Metal Nanoparticle Surfaces**
Amir Bahman Samsam Bakhtiari, Dennis Hsiao, Guoxia Jin, Byron D. Gates,* and
Neil R. Branda*
New methods for the spatially and temporally controlled
delivery of chemical and/or biochemical species using low-
energy visible light offers greater opportunities in photo-
dynamic therapy,[1] chemical synthesis,[2] and photolithogra-
phy.[3] The fact that designer metal nanoparticles absorb light
of specific frequencies (at their plasmon resonance) and
efficiently convert it into heat (through the photothermal
effect) is a particularly appealing phenomenon that has been
effectively used to trigger cell death[4] and melt ice[5] within
the vicinity of the nanoparticles. However, to date, the heat
generated when visible light is absorbed by such structures
has not been used to initiate delivery events. This is surprising,
as this technique would provide a convenient and general
method to break chemical bonds and release biologically
relevant payloads from the nanoparticle surfaces. Both
detrimental (poisonous) and beneficial (therapeutic) agents
could be released on command by using such a universal
method.
is activated can be easily tuned by modifying the structure of
the two chemical components (the diene and dienophile).[6]
This cycloaddition reaction is particularly well-suited for
surface-localized chemistry and has been used to selectively
decorate[7] and control nanoparticle aggregation[8] by directly
applying heat to suspensions of nanoparticles. The photo-
thermal effect has yet to be taken advantage of in this context.
We identified a specific retro-Diels–Alder reaction (shown in
the substructure in Figure 1a). Our choice was based on the
fact that the bond-breaking reaction of 7-oxa-bicyclo-
[2.2.1]hept-5-ene-2,3-dicarboxylic imide to release a furan
and a maleimide component barely occurs at room temper-
ature but proceeds above 608C. Model studies on both the
endo and exo isomers of the depicted bicyclic system suggest
that, as expected, the use of the former isomer is advanta-
geous[9–11] since is cleanly converts into its two components at
We recently set out to develop a new method to harness
the local heat that dissipates from the nanoparticles when
they are stimulated with visible light, in order to trigger
thermally activated bond-breaking reactions close to the
nanoparticle surfaces without significantly increasing the
temperature of the surrounding environment. This approach
has a benefit over others in that specific agents can be
delivered to a cell without damaging it. Herein, we describe
how we can trigger the release of a fluorescent dye from core–
shell nanoparticles as the first illustration of a generalized
approach to photothermal release.
We chose the retro-Diels–Alder reaction to demonstrate
our new release method because the temperature at which it
[*] A. B. S. Bakhtiari, D. Hsiao, G. Jin, Prof. B. D. Gates,
Prof. N. R. Branda
4D LABS at Simon Fraser University
8888 University Drive, Burnaby, BC V5A 1S6 (Canada)
Fax: (+1)778-782-3765
E-mail: bgates@sfu.ca
nbranda@sfu.ca
[**] This research was supported by the Natural Sciences and Engi-
neering Research Council (NSERC) of Canada, the Canada Research
Chairs Program, and Simon Fraser University (SFU) through the
Community Trust Endowment Fund. This work made use of 4D
LABS shared facilities supported by the Canada Foundation for
Innovation (CFI), British Columbia Knowledge Development Fund
(BCKDF), and Simon Fraser University. We thank Prof. Gary Leach
(SFU, Chemistry Department) for helpful discussions and for
access to his laser spectroscopy laboratory.
Figure 1. a) Release of the fluorescein dye from the surface of a silica–
gold core–shell nanoparticle by using the photothermal effect to
induce the retro-Diels–Alder reaction of an anchored 7-oxa-bicyclo-
[2.2.1]hept-5-ene-2,3-dicarboxylic imide derivative. UV/Vis absorption
spectra of aqueous dispersions with TEM images of each type of
particle shown as insets: b) 200 nm diameter core–shell gold nano-
particles (inset scale bar 200 nm)and c) 16 nm diameter solid gold
nanoparticles (inset scale bar 50 nm).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200805303.
4166
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 4166 –4169

Angewandte
Chemie
lower temperatures (63% yield after 5 h for the endo isomer
compared to 45% after 8 h for its exo counterpart) with no
observable side-products.[12] In our example, the bicyclic
system has a fluorescein dye attached to the furan compo-
nent; the dye acts as a useful read-out signal that can be easily
detected owing to its high fluorescence quantum efficiency
and emission of light in the visible region of the electro-
magnetic spectrum.
Conveniently, the fluorescence emission from the dye
molecule is efficiently quenched by the gold layer surround-
ing the core–shell nanoparticle as long as it is kept close to the
surface of the metal.[13,14] As soon as the dye is released from
the nanoparticle surface, fluorescence quenching is reduced
and the emission is turned “on”, which provides a means to
monitor the success of the photothermally induced retro-
Diels–Alder reaction by comparing the fluorescence before
and after stimulation. The role of the maleimide component is
to anchor the bicyclic system to the nanoparticle surface
aqueous dispersions of the respective nanoparticles and
transmission electron microscopy (TEM) images reveal that
both types of particles have monodisperse shapes and sizes
(see insets in Figure 1b and Figure 1c).
The 200 nm diameter gold nanoparticles were functional-
ized with the fluorescein dye by treating aqueous nanoparticle
suspensions with an excess of the bicyclic retro-Diels–Alder
system in its disulfide form.[28,29] The reaction mixtures were
best left to equilibrate overnight at 48C to ensure maximum
coverage of the nanoparticle surface with the organic system
À
through formation of Au S bonds. Any excess bicyclic
disulfide could be removed by repeated centrifugation–
resuspension (using 18 MW water) cycles[30] until all free
fluorescein (not bound to the nanoparticles) was removed (as
monitored by fluorescence spectroscopy).[31] The final sus-
pension[32] of decorated nanoparticles retained only a very low
level of emission, which arose from the quenching of the dyeꢁs
fluorescence through photoinduced energy transfer to the
nanoparticle,[33] thus providing a useful baseline with which
the release event can be monitored (Figure 2a).
À
through the Au S bond that is provided by a long-chain
alkane thiol (the synthesis of the fluorescent bicyclic system is
described in the Supporting Information). A core–shell metal
nanoparticle, in which a thin layer of gold is wrapped around a
silica sphere, is shown in Figure 1a.[15] The use of a solid gold
nanoparticle[16,17] for photothermal release is also described.
As mentioned above, the photothermal process leads to
the radiation of heat from the surface of metal nanoparticles
as a result of phonon relaxation from an excited state that is
accessed by irradiation of the surface plasmon resonance
(SPR) band of the nanoparticle.[18–20] The position of the SPR
band in the electromagnetic spectrum is dictated by the size,
shape, and composition of the nanoparticles, all of which can
be programmed into the nanoparticle synthesis.[21] SPR bands
typically appear within the visible and near-infrared (NIR)
regions of the spectrum (from 400 nm to greater than
1100 nm),[22] with the longer wavelengths being favorable in
release applications (especially biomedical-related), since
lower energy light will penetrate deeper into tissue while
minimizing damage to the surroundings. The longer wave-
lengths are also less efficiently absorbed by the dielectric
coating of the nanoparticle surface or by the medium in which
the particles are dispersed.[23,24] Our studies focus on the use
of decorated core–shell nanoparticles ((200 Æ 10) nm diame-
ter) that are composed of silica spheres wrapped in a (10 Æ
2) nm thick shell of gold and have two SPR bands (700 and
1000 nm) in the NIR region of the spectrum (Figure 1b). The
synthesis of these particles started with the preparation of
silica colloids by a modified Stꢀber technique,[25] followed by
attachment of an amino-terminated silane to the surface of
the SiO2 particles.[26] This surface modification is necessary in
order to attach the 3 nm diameter gold seeds[27] onto the
spherical silica templates needed to initiate the growth of the
10 nm thick gold shell by reduction of tetrachloroauric acid.[15]
The optical properties of solid gold nanoparticles with a
diameter of (16 Æ 3) nm are shown in Figure 1c, which
illustrates how the SPR band is significantly shifted to
higher energy (520 nm) because of the nanoparticle size and
composition. These pure gold nanoparticles are synthesized in
one step by a thermally induced reduction of tetrachloroauric
acid with trisodium citrate.[16,17] Both syntheses produce stable
Figure 2. a) Representative spectra showing changes in the fluores-
cence intensity when an aqueous dispersion of 200 nm diameter gold-
coated silica nanoparticles functionalized with the fluorescein dye is
irradiated with NIR light (800 nm, 1 kHz, 700 mW, 100 fs). Increase in
fluorescence intensity when aqueous dispersions of b) 200 nm core–
shell and c) 16 nm solid gold nanoparticles decorated with the
&
fluorescein dye are heated in a water bath at 608C ( ) or irradiated
with 800 nm light (1 kHz, 700 mW, 100 fs) for the core–shell particles
(ꢀ) and 532 nm light (10 Hz, 100 mW, 4 ns) for the gold nanoparticles
*
( ).
The fluorescence intensity of the fluorescein dye imme-
diately increased when aqueous dispersions of the core–shell
nanoparticles were irradiated at 800 nm (1 kHz, 700 mW,
100 fs) with a pulsed laser light that overlaps with their SPR
band. This increase indicates effective release of the dye into
the bulk solution. A representative example of the changes
typically observed is shown in Figure 2a. During this process,
there was no measurable change in temperature of the
aqueous suspension measured by using a thermocouple, even
though there was clearly enough heat generated at the surface
of the nanoparticles to break the bonds in the retro-Diels–
Angew. Chem. Int. Ed. 2009, 48, 4166 –4169
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4167

Communications
À
Alder system. No changes in fluorescence intensity were
nanoparticle through a combination of Au S bond cleavage
observed when a similar suspension is exposed to ambient
light, since this light source is insufficient to induce the retro-
Diels–Alder reaction through the photothermal effect. The
increase in fluorescence intensity was significantly more rapid
when the retro-Diels–Alder reaction is triggered with 800 nm
pulsed light than when it is induced by applying direct heat
(Figure 2b).
The photothermally induced release was almost complete
after only 30 min, while the equivalent release reaction
progressed to only a quarter of this extent when an identical
sample was incubated in a water bath at 608C, a temperature
at which significant damage to cellular systems would occur.
Similar trends were observed for the 16 nm gold nanoparticles
when 532 nm light (10 Hz, 100 mW, 4 ns) was used (Fig-
ure 2c). Once again, the photothermal release from the
nanoparticles is significantly more efficient than when the
sample was directly heated. We estimate that approximately
1600 dye molecules are released per gold nanoparticle.[34]
Initial inspection of Figure 2b,c suggests that the release
of the fluorescein dye from the 16 nm gold nanoparticles is
more efficient than the release observed for the core–shell
particles, therefore the former is preferable. This is in fact not
the case, and a closer look at the results suggests that there is
and the retro-Diels–Alder reaction. The lack of an observed
shift in the SPR band following photothermal release from
the core–shell particles suggests that the retro-Diels–Alder
reaction is the dominant release mechanism from these
particles.
The examples presented herein clearly demonstrate that
the localized heat generated by the photothermal effect of
metal nanoparticles can be harnessed to selectively break
chemical bonds and release anchored components. The
release event can be visually demonstrated by using fluores-
cence microscopy to compare aqueous samples of core–shell
nanoparticles before and after irradiation. Figure 4 clearly
À
substantial cleavage of the Au S bond during the photo-
thermal event. As observed for other systems, a blue shift
occurred in the SPR band when the 16 nm particles were
irradiated with 532 nm pulsed light (Figure 3a). This shift
correlates to the loss of the surface-capping group,[35] and is
Figure 4. Fluorescence microscopy images of capillary tubes contain-
ing 200 nm diameter core–shell particles decorated with the fluores-
cein dye before (top) and after (bottom) irradiation with light (800 nm,
1 kHz, 700 mW, 100 fs) for 10 min.
illustrates how aqueous dispersions of both the decorated and
nondecorated 200 nm core–shell nanoparticles do not signifi-
cantly fluoresce until they are exposed to pulsed 800 nm light.
The rate of photothermally induced release from the core–
shell nanoparticle is approximately four times that observed
for a thermally initiated release. Results obtained from
irradiation of a solid gold nanoparticle with visible light
suggest that the dye molecules are released through the
Figure 3. a) Changes in absorption maximum as the fluorescence
intensity increases when aqueous dispersions of 16 nm gold nano-
particles decorated with the fluorescein dye are heated in a water bath
À
breakage of Au S bonds. This type of release is, however,
&
*
at 608C ( ) or irradiated with 532 nm light (10 Hz, 100 mW, 4 ns) ( ).
viewed as detrimental to biological systems, as free thiols are
unstable and can lead to unwanted side-reactions. Future
studies will take advantage of the demonstrated selective
release of a molecular payload, and explore the possibilities
for both temporal and spatial control of release. This on-
demand release of a molecular payload will be highly
beneficial to drug delivery and other biomedical applications.
b) Absorption spectra of an aqueous dispersion of 200 nm core–shell
nanoparticles decorated with the fluorescein dye irradiated with NIR
light (800 nm, 1 kHz, 700 mW, 100 fs) over a period of 17 min.
Minimal changes are observed over this period.
not observed when the particles were directly heated. On the
other hand, the position, intensity, and shape of the SPR
bands recorded for the core–shell nanoparticles were con-
sistent, even after more than 17 min of irradiation with
800 nm light (Figure 3b). A significant increase in fluores-
cence intensity was observed during this period of irradiation.
These results indicate a different mechanism for the photo-
thermal release of the fluorescein dye from each of the two
types of gold nanoparticles. The photothermal effect most
likely leads to the release of the dye from the 16 nm gold
Received: October 29, 2008
Revised: March 29, 2009
Published online: April 30, 2009
Keywords: cycloaddition · fluorescence · nanostructures ·
.
photorelease · photothermal effects
4168
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 4166 –4169

Angewandte
Chemie
[24] A. M. Gobin, M. H. Lee, N. J. Halas, W. D. James, R. A. Drezek,
[1] T. Patrice, Photodynamic Therapy, Royal Society of Chemistry,
Cambridge, 2003.
[2] W. M. Horspool, F. Lenci, CRC Handbook of Organic Photo-
chemistry and Photobiology, Vols. 1 and 2, Routledge, USA,
2003.
J. L. West, Nano Lett. 2007, 7, 1929 – 1934.
[25] W. Stꢀber, A. Fink, E. Bohn, J. Colloid Interface Sci. 1968, 26,
62 – 69.
[26] A. van Blaaderen, A. Vrij, J. Colloid Interface Sci. 1993, 156, 1 –
18.
[3] S. P. A. Fodor, J. L. Read, M. C. Pirrung, L. Stryer, A. T. Lu, D.
Solas, Science 1991, 251, 767 – 773.
[27] D. G. Duff, A. Baiker, P. P. Edwards, Langmuir 1993, 9, 2301 –
2309.
[4] C. M. Pitsillides, E. K. Joe, X. Wei, R. R. Anderson, C. P. Lin,
Biophys. J. 2003, 84, 4023 – 4032.
[5] H. H. Richardson, Z. N. Hickman, A. O. Govorov, A. C.
Thomas, W. Zhang, M. E. Kordesch, Nano Lett. 2006, 6, 783 –
788.
[6] V. Lemieux, S. Gauthier, N. R. Branda, Angew. Chem. 2006, 118,
6974 – 6978; Angew. Chem. Int. Ed. 2006, 45, 6820 – 6824.
[7] M. Shi, J. H. Wosnick, K. Ho, A. Keating, M. S. Shoichet, Angew.
Chem. 2007, 119, 6238 – 6243; Angew. Chem. Int. Ed. 2007, 46,
6126 – 6131.
[28] The disulfide used to decorate the nanoparticles is not sym-
metrical and has one bicyclic system and one free maleimide
derivative attached to it.
[29] A. Ulman, Chem. Rev. 1996, 96, 1533 – 1554.
[30] Purification of the particles included at least three complete
cycles of centrifugation with decanting of the supernatant
followed by dispersion of the collected solids in 18 MW water.
The centrifugation step required a speed of 10000 rpm for
10 min for the gold nanoparticles and 4000 rpm for 10 min for
the core–shell nanoparticles.
[8] J. Zhu, A. J. Kell, M. S. Workentin, Org. Lett. 2006, 8, 4993 –
4996.
[9] K. Adachi, A. K. Achimuthu, Y. Chujo, Macromolecules 2004,
37, 9793 – 9797.
[10] J. R. McElhanon, D. R. Wheeler, Org. Lett. 2001, 3, 2681 – 2683.
[11] The exo isomer converts to its furan and maleimide components
at higher temperatures.
[31] Typically, the fluorescence emission intensity remained constant
after three purification cycles, which indicated that the
unbounded dye had been removed from the reaction.
[32] A time dependent study on the nanoparticles dispersed in water
(Figure S7 in Supporting Information) indicated a relatively
stable dispersion of gold nanoparticles over a period of 4 h.
These dispersions are stable for time periods longer than the
length of the photothermal release experiments.
[12] See the Supporting Information.
[13] N. Nerambourg, M. H. V. Werts, M. Charlot, M. Blanchard-
Desce, Langmuir 2007, 23, 5563 – 5570.
[33] K. G. Thomas, B. I. Ipe, P. K. Sudeep, Pure Appl. Chem. 2002, 74,
1731 – 1738.
[14] F. Cannone, G. Chirico, A. R. Bizzarri, S. Cannistraro, J. Phys.
Chem. B 2006, 110, 16491 – 16498.
[15] T. Pham, J. B. Jackson, N. J. Halas, T. R. Lee, Langmuir 2002, 18,
4915 – 4920.
[16] B. V. Enꢂstꢂn, J. Turkevich, J. Am. Chem. Soc. 1963, 85, 3317 –
3328.
[17] J. Kimling, M. Maier, B. Okenve, V. Kotaidis, H. Ballot, A. Plech,
J. Phys. Chem. B 2006, 110, 15700 – 15707.
[18] C. K. Sun, F. Vallꢃe, L. Acioli, E. P. Ippen, J. G. Fujimoto, Phys.
Rev. B 1993, 48, 12365 – 12368.
[19] T. S. Ahmadi, S. L. Logunov, M. A. El-Sayed, J. Phys. Chem.
1996, 100, 8053 – 8056.
[20] B. Khlebtsov, V. Zharov, A. Melnikov, V. Tuchin, N. Khlebtsov,
Nanotechnology 2006, 17, 5167 – 5179.
[21] P. K. Jain, K. S. Lee, I. H. El-Sayed, M. A. El-Sayed, J. Phys.
Chem. B 2006, 110, 7238 – 7248.
[34] The number of fluorescent molecules per particle was estimated
based on a quantitative analysis of the amount of dye released
and the number of particles in solution. For this experiment, a
dispersion of dye-coated nanoparticles was adjusted to pH 12
and heated to 808C in a water bath. The presence of a strong
base (e.g., KOH) efficiently hydrolyzes the ester linkage in the
dye bound to the nanoparticles. The release of dye molecules
was monitored by fluorescence spectroscopy. Upon reaching a
plateau in the overall fluorescence emission, the quantity of dye
released was determined using a series of fluorescein standards.
This quantitative analysis provided an estimate of the number of
dye molecules released into solution. The number of gold
nanoparticles in solution was quantified using the average
particle size, the density of gold, and the amount of gold in
solution (as determined by inductively coupled plasma MS
analysis).
[22] C. H. Chou, C. D. Chen, C. R. C. Wang, J. Phys. Chem. B 2005,
109, 11135 – 11138.
[35] P. K. Jain, W. Qian, M. A. El-Sayed, J. Am. Chem. Soc. 2006, 128,
2426 – 2433.
[23] L. R. Hirsch, R. J. Stafford, J. A. Bankson, S. R. Sershen, B.
Rivera, R. E. Price, J. D. Hazle, N. J. Halas, J. L. West, Proc. Natl.
Acad. Sci. USA 2003, 100, 13549 – 13554.
Angew. Chem. Int. Ed. 2009, 48, 4166 –4169
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4169