LETTER pubs.acs.org/NanoLett
Transparent Plasmonic Nanocontainers Protect Organic Fluorophores against Photobleaching Soraya Zaiba,†,‡ Frederic Lerouge,§ Ana-Maria Gabudean,†,# Monica Focsan,†,# Jean Lerme,|| Thibault Gallavardin,§ Olivier Maury,§ Chantal Andraud,§ Stephane Parola,§ and Patrice L. Baldeck*,† †
Laboratoire Interdisciplinaire de Physique CNRS UMR 5588, Universite Grenoble 1, Grenoble, F-38402, France Quantum Electronics Laboratory, Physics Faculty USTHB, El-Alia Bab-Ezzouar, 16111 Algiers, and Physics Department, Science Faculty UMBB, 35000 Boumerdes, Algeria § Universite de Lyon, ENS Lyon, Universite Lyon 1, CNRS, Laboratoire de Chimie (UMR 5182), 46, Allee d’Italie, 69364 Lyon Cedex 07, France Universite de Lyon, Universite Lyon I, CNRS, Laboratoire de Spectrometrie Ionique et Moleculaire (UMR 5579), B^at. A. Kastler, 43 Bld du 11 novembre 1918, 69622 Villeurbanne Cedex, France # Babes-Bolyai University, Faculty of Physics and Interdisciplinary Research Institute in Bio-Nano-Sciences, T.Laurian 42, 400271 Cluj-Napoca, Romania
)
‡
bS Supporting Information ABSTRACT: Numerous research efforts are investigating the possibility of using light interactions with metallic nanoparticles to improve the fluorescence properties of nearby molecules. Few investigations have considered the encapsulation of molecules in metallic nanocavities. In this paper, we present the optical properties of new hybrid nanoparticles consisting of gold nanoshells and fluorescent organic dyes in their liquid cores. Microspectroscopy on single nanoparticle demonstrates that the extinction spectra are in good agreement with Mie’s theory. Finite difference time domain (FDTD) calculations reveal that excitation and emission radiations are efficiently transmitted through the thin gold nanoshells. Thus, they can be considered as transparent plasmonic nanocontainers for photoactive cores. In agreement with FDTD calculations, measurements show that fluorophores encapsulated in gold nanoshells keep their brightness, but they show fluorescence lifetimes 1 order of magnitude shorter. As a salient consequence, the photoresistance of encapsulated organic dyes is also improved by an order of magnitude. This unusual ultraviolet photoresistance results from the reduced probability of triplet singlet conversion that eventually exposes dyes to singlet oxygen photodegradation. KEYWORDS: Gold nanoshell, fluorescent molecules, photoresistance, excited-state lifetime, FDTD calculations, surface plasmon resonance
I
n the past decade, noble metal nanoparticles have sparked wide interest for developing new optical applications in the context of bioimaging,1 3 photothermal therapy,2,4,5 optical sensors,6 and solar cells.7,8 This interest is essentially due to their unique optical properties related to their localized surface plasmon resonances (LSPR). The interaction of visible or near-infrared light with the free electrons of these nanoparticles leads to strong scattering and near field enhancements. Especially, in the LSPR spectral range, near field intensities several orders of magnitude larger than those of the incident driving fields can be generated in the vicinity of metal nanoparticle surfaces. This optical antenna property at the nanoscale has been used to boost the optical properties of molecular systems located close to noble metal nanostructures such as for instance, in surface-enhanced Raman scattering (SERS) experiments.9 Numerous research works are investigating the possibility of using the light interaction with metallic nanoparticles to improve r 2011 American Chemical Society
the fluorescence properties of nearby molecules.10 The coupling of fluorophores with nanoparticles increases their absorption and spontaneous emission rates. Up to now, strategies to design nanomaterials have focused on hybrid geometries with molecules localized outside the metallic nanostructures. Only a few works have focused on the encapsulation of molecules inside a metallic nanoparticle.11 13 Theoretically it was predicted that, depending on the overall radius and shell thickness, both an increase in the fluorescence brightness and an improvement of the photostability occur for a single molecule located inside a metallic nanocavity.14 Recently, we have reported an “emulsion” process for elaborating gold nanoshells with liquid cores.15 After entrapping fluorescent molecules within these nanoshells the final object Received: February 10, 2011 Revised: March 22, 2011 Published: April 08, 2011 2043
dx.doi.org/10.1021/nl2004847 | Nano Lett. 2011, 11, 2043–2047
Nano Letters exhibits luminescent features. In this paper, we present the experimental and theoretical luminescence properties of these fluorescent fluorophore@Au nanoshells. The interaction between the encapsulated fluorophores and the gold metallic shell decreases the excitedstate lifetime of the fluorophore and improves its photoresistance. These results are well rationalized by FDTD simulations. Hollow spheres are prepared by stabilized emulsions without surfactants obtained with fast mechanical stirring.15 Gold salt HAuCl4 is dissolved in water and mixed with a solution of tetraoctylammonium bromide (TOAB) in toluene. Dodecanethiol is then added to the organic phase. Then a benchtop homogenizer is used to provide fast stirring and an aqueous solution of NaBH4 is quickly added. During this step an emulsion of nanodroplets of water containing NaBH4 is formed in the organic phase. The small nanoparticles and the nanocapsules can be easily separated by washing with ethanol and centrifugation. Colorless colloidal suspensions of nanocapsules can be isolated and characterized. Entrapment of chromophores inside the nanocapsules is also possible since the reaction occurs in a biphasic liquid liquid medium. The required condition is that the chromophore remains soluble in the droplets phase during the emulsion formation. Rhodamine 610, also named Rhodamine B, was chosen for several reasons, among which is its well-known spectroscopic properties. During synthesis, the chromophores are added to the emulsion. Samples are treated several times under ultrasound in ethanol to make sure that any physisorbed organic moieties are washed out from the outer layers of the nanoshells and the solvent. Transmission electron microscopy (TEM) analysis of the nanoshells shows typical spherical nanoshells similar to those observed in ref 15. Diffraction patterns of the wall from the gold nanoshells are consistent with Au(0) structure. Dynamic light scattering (DLS) measurements are
Figure 1. TEM image of a nanoshell and high-resolution TEM (HRTEM) of the wall of the shell.
LETTER
consistent with TEM micrographs showing particles with a diameter around 200 nm with a good size distribution. Extinction spectra of Rhodamine 610@Au nanoshells are measured using a multimodal microspectroscopy setup. The nanoshells are dispersed in a spin-coated film of bovine serum albumin (BSA). The microspectroscopy setup is based on an Axiovert 200 Zeiss microscope and an Ocean Optics S2000 spectrometer. A multimode optical fiber (diameter 100 μm) collects the magnified (Zeiss objective Apochromat 100) optical signals (transmission, fluorescence, and scattering) of single nanoshell in the camera port, with a spatial resolution of 1 μm. Typical extinction spectra of different nanoshells are shown in Figure 2A. A two-band structure in the 650 800 nm range is observed, which is in agreement with the dipole and quadrupole contributions to the plasmon resonance. Theoretical computations based on Mie theory and developed by Bohren and Huffman16 have been performed (Figure 2B) on metallic shells (90 nm external radius and 20 nm shell thickness (bulk gold), with a water core (dielectric index n = 1.33) and embedded in a thin film of BSA (n = 1.45) as surrounding medium). The theoretical result is in rather good agreement with experimental observations confirming the spectroscopic shell properties of the synthesized nanoparticles. It is then possible to assign the highest wavelength band to the dipole band while the short wavelength band can be associated to the quadrupole resonance. Noticeable discrepancies in the low and high sides of the large bell-shaped resonance, as well as differences in the quadrupolar mode bandwidth, are observed when we compare the two spectra. This could probably be due to the shell structure differences between the perfect geometry assumed in the modeling and the
Figure 3. FDTD calculations of field intensity distributions around and inside the metallic shell at (A) the excitation (B) emission wavelengths of Rhodamine 610.
Figure 2. (A) Measured extinction spectra of Rhodamine 610@Au shells of different sizes in BSA. (B) Theoretical extinction cross section (Mie’s theory) of a single water core gold nanoshell in BSA. 2044
dx.doi.org/10.1021/nl2004847 |Nano Lett. 2011, 11, 2043–2047
Nano Letters partial granular and/or porous structure of the synthesized gold shell resulting from the “emulsion” process.15 One major fact that has to be taken into account is whether the excitation wavelengths can or cannot go through the gold shell whatever the way (in or out). Figure 3 shows FDTD calculation results (Lumerical FDTD software17) of electric field intensities inside the nanoshells at the excitation and emission wavelengths of Rhodamine 610. Clearly, the visible irradiations (consisting on a wave plane propagating along the y axis and polarized in the z direction), can propagate through the thin gold shells. At these wavelengths, the metallic nanoshell acts as a transparent nanocontainer for the chromophores that are entrapped in the liquid core. Steady-state fluorescent properties of single Rhodamine 610@Au nanoshells are measured by epifluorescence with an excitation wavelength at 400 nm (Figure 4). In order to compare the emission properties of Rhodamine 610 either within the gold shell or in water, we have synthesized liposomes encapsulating Rhodamine. These objects present the same size as the shells and the same concentration of fluorophores. The emission spectrum of the Rhodamine 610 in liposomes is similar to the one of Rhodamine in water. The emission intensity of the hybrid nanoshells is at least in the same order of magnitude as the emission of the Rhodamine in liposomes under identical excitation conditions. It is important to note that empty gold
Figure 4. Typical emission spectra of different Rhodamine 610@Au nanoshells with water cores (dashed and dot curves) under the same excitation conditions. Black curve corresponds to the emission of Rhodamine 610 in liposomes.
LETTER
nanoshells do not show any fluorescence under the same conditions. The differences in the emission intensity observed for various Rhodamine 610@Au nanoshells result from the size dispersion of the synthesized nano-objects. The emission spectra of Rhodamine 610 in the water core of the gold nanoshell are blue-shifted in comparison to the emission of Rhodamine 610 in liposomes (Figure 4). In the gold nanoshell, the excited-state lifetime of the fluorophore is so short (see below) that the molecular geometry does not have time to relax before emission. Thus, the fluorescence spectrum appears as the mirror image of the absorption spectrum, i.e., it is not red-shifted by the conformation relaxation the same way as observed in the water for liposomes. Fluorescence lifetime measurements of Rhodamine 610@Au nanoshells embedded in a BSA thin film were performed using a fluorescence lifetime imaging microscope (FLIM) and a Ti sapphire femtosecond laser at 760 nm. FDTD calculations show that this near-infrared incident laser field also efficiently gets into the nanoshell core. The fluorescence of Rhodamine 610 is then induced by two-photon absorption. The fluorescence lifetime duration is less than the 150 ps minimum resolution of this FLIM system. Thus, it is at least 10 times shorter than the fluorescence lifetime of Rhodamine 610 in water (1.7 ns).18 The influence of the gold shell on the excited state lifetime and the quantum yield is studied with FDTD calculations by considering the emitting molecule as a classical dipole in the liquid core of the nanoshell.19 The radiative and nonradiative decay rates, γrad and γnrad, respectively, are calculated for different positions of the molecule inside the nanoshell. In Figure 5, where the theoretical results are summarized, the values normalized with respect to the free-space case (index “0”) are plotted. The fluorescence lifetime is 2.5 times shorter at the core center, and it decreases even more when the molecule reaches the inner metallic surface. The lifetime is decreased by about 10 times at 25 nm from the inner surface of the shell. The reduction of the fluorescence lifetime is due to the increase in the radiative and nonradiative decay rate when the molecule is nearby the metallic surface. In fact, when the fluorophore gets closer to the metallic surface the coupling with the plasmon becomes higher. The absorption and the scattering of the metallic nanoparticle contribute to the increase in the nonradiative and radiative decay rates which decrease the fluorescence lifetime. The fluorescence quantum efficiency Q, which corresponds to the ratio of radiative decay process to total decay decreases from
Figure 5. (A) Normalized radiative (left) and nonradiative (right) decay rates. (B) Normalized fluorescence lifetime (left) and quantum efficiency (right). 2045
dx.doi.org/10.1021/nl2004847 |Nano Lett. 2011, 11, 2043–2047
Nano Letters
Figure 6. Fluorescence photoresistance of Rhodamine 610 molecules encapsulated in Au nanoshell and liposomes.
60% at the core center to 14% at 25 nm from the inner metallic surface and tends to zero close to the metallic surface causing fluorescence quenching. The fluorescence brightness is the ratio of quantum efficiency and lifetime. This is why molecules located at 25 nm from the inner metallic surface have a 10-fold shorter lifetime with a 2.4-fold brightness. In a different experiment, Sergei K€uhn and co-workers have already reported that the excited-state lifetimes of fluorophores can be reduced by 20-fold simultaneously to a 20-fold enhancement in the fluorescence intensity when approaching a gold nanoparticle at 12 nm.20 This can be correlated with our observations. Photobleaching consists of a photochemical modification of dyes resulting in the irreversible loss of their fluorescence ability. It is then a major drawback in microscopy and imaging.21 The design of a fluorescent molecule more stable and more resistant to photobleaching becomes of great interest. A few years ago it has been shown that decreasing the photobleaching or photolysis can be achieved by coupling a molecule to a metallic surface.22 We have studied the stability of the emission signal for long excitation times for both Rhodamine 610@Au and Rhodamine 610 encapsulated in liposomes under the same excitation conditions and same concentration (Figure 6). It appears that the fluorescence emission of Rhodamine 610@Au is relatively stable with time and the exponential decay fitting shows that the fluorescence decay time is around 37 min for Rhodamine 610@Au and 3.7 min for Rhodamine 610 in liposomes. Thus, the entrapment of organic fluorescent molecules inside gold nanoshells makes them highly photoresistant. In order to explain this phenomenon, excited states of the chromophores should be considered. When a fluorescent molecule is irradiated by light at its specific excitation wavelength, it absorbs the photon and goes to a singlet excited state. Via an intersystem crossing, it can undergo a transition to a triplet excited state if the duration of this conversion is less than the excited state lifetime. However, a molecule in the triplet excited state can lead to a permanent change in the molecule structure which causes the annihilation of the fluorescence called photobleaching or photolysis.23 The interaction between the chromophore triplet state and the molecular oxygen triplet state leads to the formation of molecular radicals and singlet oxygen. Localizing the fluorescent molecule near a metallic nanoparticle allows circumventing this drawback, since the excitedstate lifetime is strongly decreased. When the lifetime becomes
LETTER
comparable to, or shorter than, that of the intersystem crossing process, the transition to the triplet state becomes unlikely and, consequently, the probabilities of molecular oxygen attack and radical generation are reduced. Thus, metallic nanoshells act as an efficient protection against photobleaching. In summary, the fabrication of hollow gold nanospheres by the miniemulsion technique is a simple encapsulation method allowing one to study their plasmon interaction with inner fluorophores. Fluorescence imaging microscopy shows that these fluorophores can easily be excited and detected despite the gold shell. FDTD calculations confirm that the shell plasmon mediates the electromagnetic flux through the ultrathin metallic wall. Thus, gold nanoshells act as transparent plasmonic nanocontainers for photoactive materials. Under excitation wavelengths in the ultraviolet and blue range, there is no significant internal enhancement. The brightness of fluorescent nanoshells is at most a few times brighter than similar liposomes doped with the same molecular concentration. In contrast, the fluorescence lifetime is strongly reduced by the additional nonradiative pathway of Ohmic losses. Consequently, there is a reduced probability of triplet singlet intersystem crossing that exposes dyes to singlet oxygen photodegradation. The photoresistance of encapsulated organic dyes is improved by an order of magnitude. The gold nanoshell acts as a transparent plasmonic nanocontainer that also shields organic fluorophores against photobleaching processes. Further works are in progress to optimize this hybrid system in order to benefit from the high field enhancement that could be reached in such a nanoshell core.
’ ASSOCIATED CONTENT
bS
Supporting Information. Experimental and calculation details. This material is available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION Corresponding Author
*E-mail:
[email protected].
’ ACKNOWLEDGMENT This work has been partially supported by ANR P3N project nanoPDT # ANR-09-NANO-027-04 and MACODEV Rh^oneAlpes cluster. ’ REFERENCES (1) Patel, Y.; Saha, S.; Dimarzio, C.; O’Malley, D.; Nagesha, D.; Sridhar, S. Proceedings of the sixth IEEE international conference and symposium on biomedical imaging: from nano to macro 2009, 759–762. (2) Gindy, E. M.; Prud’Homme, K. R. Expert Opin. Drug Delivery 2009, 6, 865–878. (3) Bardhan, R.; Grady, N. K.; Cole, J. R.; Joshi, A.; Halas, N. J. Acc. Chem. Res. 2009, 3, 744–752. (4) Surbhi, L.; Clare, S. E.; Halas, N. J. Acc. Chem. Res. 2008, l41, 1842–1851. (5) Rozanova, N.; Zhang, J. Z. Sci. China, Ser. B: Chem. 2009, 52, 1559–1575. (6) Kvansnicka, P.; Homola, J. Biointerphases 2008, 3, FD4–FD11. (7) Akimov, Y. A.; Ostrikov, K.; Li, E. P. Plasmonics 2009, 4, 07–113. (8) Moulin, E.; Sukmanowski, J.; Luo, P.; Carius, R.; Royer, F. X.; Stiebig, H. J. J. Non-Cryst. Solids 2008, 354, 2488–2491. (9) Qian, X.-M.; Nie, S. M. Chem. Soc. Rev. 2008, 37, 912–920. 2046
dx.doi.org/10.1021/nl2004847 |Nano Lett. 2011, 11, 2043–2047
Nano Letters
LETTER
(10) Fort, E.; Gresillon, S. J. Phys. D: Appl. Phys. 2008, 41, 13001. (11) Zhang, J.; Fu, Y.; Lakowicz, J. R. J. Phys. Chem. C 2010, 114, 7653–7659. (12) Zhang, P.; Guo, Y. J. Am. Chem. Soc. 2009, 131, 3808–3809. (13) Zhang, J.; Gryczynski, I.; Gryczynski, Z.; Lakowicz, J. R. J. Phys. Chem. B 2006, 110, 8986–8991. (14) Enderlein, J. Appl. Phys. Lett. 2002, 80, 315–317. (15) Lux, F.; Lerouge, F.; Bosson, J.; Lemercier, G.; Andraud, C.; Vitrant, G.; Baldeck, L. P.; Chassagneux, F.; Parola, S. Nanotechnology 2009, 20, 355603. (16) Bohren, C. F.; Huffman, D. R. Absorption and scattering of light by small particles; Wiley: New York, 1983. (17) www.lumerical.com. (18) Magde, D.; Rojas, G. E.; Seybold, P. G. Photochem. Photobiol. 1999, 70, 737. (19) Bharadwaj, P.; Novoty, L. Opt. Express 2007, 15, 14266–14274. (20) K€uhn, S.; Hakanson, U.; Rogobete, L.; Sandoghdar, V. Phys. Rev. Lett. 2006, 97, 017402. (21) Deschenes, L. A.; Vanden Bout, D. A. Chem. Phys. Lett. 2002, 365, 387–395. (22) Muthu, P.; Gryczynski, I.; Gryczynski, Z.; Talen, J. M.; Akopova, I.; Borejdo, J. J. Biomed. Opt. 2008, 13, 014023. (23) Song, L.; Varma, C. A. G. O.; Verhoeven, J. W.; Tanke, J. H. Biophys. J. 1996, 70, 2959–2968.
2047
dx.doi.org/10.1021/nl2004847 |Nano Lett. 2011, 11, 2043–2047
