J. Phys. Chem. C 2010, 114, 17829–17835
17829
Distance-Dependent Fluorescence Quenching of Conjugated Polymers on Au/Ag Striped Nanorods Weiming Zheng and Lin He* Department of Chemistry, North Carolina State UniVersity, Raleigh, North Carolina 27695 ReceiVed: July 28, 2010; ReVised Manuscript ReceiVed: August 30, 2010
To understand the effect of metal surface-polymer separation on fluorescent signal intensity, we describe here our initial investigation on the fluorescent behavior of conjugated polymers near metal nanorods by attenuating the thickness of SiO2 coatings as the spacer. Our preliminary results show that the amount of fluorescence of conjugated polymers quenched by Au/Ag striped nanorods exponentially decays away from the surface in the range of 0-20 nm. This distance-dependent fluorescence quenching phenomenon is consistent with that of small organic dye molecules with a similar fluorescent profile near a metal surface. Furthermore, the results confirm that it is a valid assumption that the previously proposed superquenching model for conjugated polymers in solution is applicable to the polymers attached to a solid surface. The absolute size of polymers is irrelevant in quenching studies. Similar distance-dependent quenching trends are observed for conjugated polymers placed near metal surfaces of different length scales. However, theoretical fitting of the experimental data illustrates subtle differences in the quenching efficiency offered by the metal substrates of different sizes to the polymer. The finding from this study allows us to obtain fluorescence measurements in a reproducible fashion, which is critical in using conjugated polymers as the signaling probes in quantitatively reporting the amount of target analytes bound to metal particles. Introduction The delocalized electronic structures of conjugated polymers allow effective electronic coupling and efficient intrachain and interchain energy transfer, leading to either superquenching or superamplification of fluorescent signals.1-3 The use of fluorescent conjugated polymers as optical probes in biosensing, therefore, has attracted much attention and has undergone substantial growth.4-8 Of various fluorescent conjugated polymers reported to date, cationic polyfluorene9-11 and polythiophene derivates12-18 are the most widely used materials in the detection of DNA, proteins, viruses, and small molecules.19-21 We recently reported the use of a polythiophene derivative in conjunction with metallically striped nanorods to detect multiple target DNA sequences simultaneously.22 The method combines the modification-free merits of positively charged conjugated polymers with the multiplexing capability of encoded particles where different striping patterns of the particles provide a means to distinguish the identities of the capture probes immobilized on each particle.23,24 The change in optical signatures of conjugated polymers when they bind to ssDNA or dsDNA allows highly sensitive and specific detection of DNA hybridization. Detection sensitivity at an attomole level has been successfully demonstrated, and single-base mutation in the target DNA sequence has been distinguished. However, during the assay development, it was found that fluorescence of conjugated polymers was strongly quenched by the particles when the polymer-DNA complexes were immobilized directly on the metal surface. It was not unexpected, giving the presence of ample reports in the literature on fluorescence quenching by a metal surface. A common means to circumvent quenching is to introduce a thin layer of an inorganic coating (e.g., SiO2) between the metal surface and * To whom correspondence should be addressed. E-mail: Lin_He@ NCSU.EDU. Tel: 919-515-2993. Fax: 919-515-8920.
dye molecules, which was adopted in our previous study to restore fluorescent signals.22 Although it was successful in the concept-proof experiment, the distance-dependent fluorescence behavior of conjugated polymers on or near the metallic striped nanorods is left unaddressed and the optimal thickness of the spacer is yet to be determined. Florescence quenching of conventional organic dye molecules by metal surfaces, such as gold or silver, has been extensively studied in the literature.25-45 Although different theories are still under intense debate, nonradiative energy transfer seems to be the dominating model to explain the phenomenon in which an electron-rich metallic surface (acceptor) is suggested to serve as an energy sink to excited dye molecules (donor). A fourth power equation has been used to model the distance-dependent fluorescence quenching between a dipole (small dye molecule) and a metal surface.44,46 The model suggests that, for a given dye molecule, the quenching efficiency of the metal surface underneath is dictated by the overlap of energy bands of the acceptor and the donor, the size and shape of the metal surface, the orientation of the donor dipole with respect to the dye-metal axis, and, last but not the least, the distance between the dye molecule and the surface. A similar nonradiative energy dissipation pathway is suspected for metal surface-based quenching of conjugated polymers. However, many questions remain: Will the quenching efficiency of polymers follow the same decay trend as the small dye molecules, regardless of it being 100+ times larger in size? Will the stretched linear conformation of polymers have an impact on the quenching efficiency? Will the so-called “superquenching” effect due to their delocalized electron structure affect the decay profile?6,7,47-49 In this report, we systematically investigated distance-dependent fluorescence quenching of conjugated polymers on Au/Ag striped nanorods by introducing an optically transparent silica spacing layer to fine-tune the separation between polymers and the metal surface. A silica
10.1021/jp1070693 2010 American Chemical Society Published on Web 09/16/2010
17830
J. Phys. Chem. C, Vol. 114, No. 41, 2010
coating was deposited via a sol-gel process, and its thickness was controlled from 0 to ∼80 nm by varying sol-gel deposition conditions. For comparison purposes, distance-dependent quenching of small dye molecules on metal nanorods and polymers on metal substrates of different sizes was monitored to examine subtle differences in their distance-dependent decay profiles. Materials and Methods Materials. The fluorescent conjugated polymer used in this study, poly(1H-imidazolium,1-methyl-3-[2-[(4-methyl-3-thienyl)oxy]ethyl]-bromide), was prepared according to the published procedure.12,13 Au/Ag striped nanorods patterned 000100 (0 represents a 1 µm segment of Au and 1 represents a 1 µm of Ag) were synthesized following the literature protocol.50,51 Gold nanoparticles (GNPs) (13 nm) were prepared in house.52 Gold substrates (50 Å of chromium, followed by 1000 of Å gold on a float glass) were purchased from Evaporated Metal Films (Ithaca, NY). Tetraethoxysilane (TEOS) and 3-aminopropyltrimethoxy silane (APTMS) were purchased from Gelest (Morrisville, PA). Ammonium hydroxide (29.5%) was purchased from Fisher (Fairlawn, NJ). DyLight 488 NHS ester and sulfosuccinimidyl 4-[N-maleimidomethyl]cyclohexane-1-carboxylate (sulfo-SMCC) were purchased from Pierce. Poly(vinylpyrrololidone) with average molar masses of 10 kg/mol (PVP-10) and (3-mercaptopropyl)trimethoxysilane (3-MPTMS) were purchased from Sigma-Aldrich (St. Louis, MO). DNA capture probe (5′-TAA CAA TAA TCC CTC A20-C3-S-S-C3OH), target DNA (5′-GAG GGA TTA TTG TTA), and dual functional ssDNA A (5′-NH2-C6-TAA CAA TAA TCC CTC A20-C3-S-S-C3-OH) were purchased from Integrated DNA Technologies (Coralville, IA). Upon receipt of thiolated DNA, the disulfide was cleaved with 100 mM dithiothreitol (DTT) and purified with a micro biospin 30 column from Bio-Rad (Hercules, CA). Silica Coating on Various Metal Surfaces. Silica was deposited onto Au/Ag striped nanorods via a modified sol-gel process.53 Briefly, 300 µL of nanorods (∼3 × 108 particles), 40 µL of TEOS, 160 µL of H2O, 490 µL of ethanol, and 10 µL of 29.5% ammonium hydroxide were mixed and subjected to sonication for 1 h. The nanorods were then washed three times with ethanol by centrifugation before being suspended in 1 mL of ethanol for later usage. This standard procedure was known to deposit a uniform silica layer of ∼20 nm on the particles’ surface.51 A thinner silica coating can be achieved by reducing the reaction time, whereas thicker silica layers were achieved by multiple coating reactions to avoid formation of silica particles without nanorods inside. GNPs (13 nm) were coated with silica of different thicknesses following the published procedure.54 Briefly, 45 mg of PVP-10 was dissolved in 2 mL of water by ultrasonication of the solution for 15 min. The PVP solution was then added into 10 mL of 10 nM GNPs and stirred at 600 rpm for 24 h at room temperature. After the mixture was centrifuged at 13 000 rpm for 30 min, the clear supernatant was discarded and the precipitate was redispersed in 0.6 mL of water. The aqueous dispersion of PVP-coated GNPs was then added to 3 mL of ethanol dropwise under vigorous stirring, followed by adding 150 µL of 29.5% ammonium hydroxide. TEOS (10 vol % in ethanol) was immediately added where the total amount of TEOS added was decided by the desired thickness of the silica shell. The reaction mixture was stirred for another 12 h. The reaction was stopped by centrifugation to separate particles from the reaction mixture. The particles were then washed with ethanol and dispersed in 3 mL of ethanol for later usage.
Zheng and He Silica films were deposited on the gold substrates using a combination of self-assembly and sol-gel techniques according to the published procedure.55,56 In particular, gold substrates (∼1 cm2) were first cleaned in a piranha solution (70% H2SO4/30% H2O2) prior to use (Caution: piranha solution is hazardous and corrosiVe. Handle with care!). The substrates were then immediately immersed in 10 mL of 20 mM 3-MPTMS for 3 h, followed by rinses in ethanol and water. Subsequent hydrolysis and condensation of 3-MPTMS on the surface was carried out by immersing the substrates into 10 mL of 0.1 M HCl for 12 h, followed by rinses with water. A TEOS sol-gel solution was prepared by mixing 163 µL of H2O, 80 µL of ethanol, 81 µL of 0.1 M HCl, and 3-60 µL of TEOS, followed by rigorous shaking of the mixture for 30 min to facilitate TEOS hydrolysis prior to deposition. The prehydrolyzed TEOS solution (∼60 µL/ cm2) was added to 3-MPTMS-modified gold substrates and allowed to spin at ∼3400 rpm for 1 min on a spin-coater (Laurell Corp, North Wales, PA). The silica-coated substrates were then stored in a desiccator at room temperature for 2 days to complete condensation and to remove any residual solvent. The thickness of the silica shells on the outside of the particles was measured manually using particle TEM images, and the thickness of the silica layer on the flat surface was measured using an ellipsometer (more details in the Instrumentation section). DNA Immobilization and Hybridization. Cationic conjugated polymer (50 µL, ∼700 µM) was first mixed stiochiometrically on a repeat unit basis with 50 µL of capture DNA probe (20 µM) in order to form the polymer-ssDNA duplexes before surface attachment. Silica-coated nanorods were first modified with primary amino groups by adding 15 µL of APTMS into 150 µL of silica-coated nanorods and 235 µL of ethanol. The mixture was vortexed for 30 min and then rinsed three times with ethanol, followed by two rinses with 10 mM CHES buffer (pH 9.0). A solution of 1 mg of sulfo-SMCC in 400 µL of CHES buffer was added into the mixture and vortexed for another hour. After rinsing the nanorods twice with the CHES buffer and twice with 10 mM PB buffer (pH 7.4), the preformed polymer-DNA duplexes were added into the mixture and vortexed for another hour, followed by three rinses in the PB buffer and resuspension in 400 µL of PB buffer. For uncoated nanorods, conjugated polymer-DNA duplexes were directly attached to nanorods through thiol-Au interaction: 100 µL of preformed polymer-capture DNA duplexes was mixed with 50 µL of nanorods in 0.3 M NaCl/PB buffer for 3 h under vortexing. After the reaction was completed, nanorods were rinsed three times and resuspended in 400 µL of PB buffer. Attachment of polymer-DNA complexes on GNPs was similar to the procedure described above. For silica-coated flat gold substrates, the surfaces were first modified with amino groups by immersing the substrates in 10% (vol) APTMS/ ethanol for 30 min. After the substrates were rinsed with ethanol and CHES and dried under N2, 0.25 mg of sulfo-SMCC in 100 µL of CHES buffer was deposited and left for 2 h, followed by rinsing with CHES buffer then PB buffer. Immediately following SMCCactivation,100µLofpreformedconjugatedpolymer-capture DNA duplexes (concn of DNA ) ∼2 µM) was spotted onto the substrates. After 3 h of incubation, the substrates were rinsed with PB buffer. DNA hybridization on particles was carried out by adding 10 µL of 1 µM target DNA to 20 µL of bare or silica-coated particles preattached with conjugated polymer-capture DNA duplexes. The mixture was vortexed for 30 min at 40 °C to allow hybridization of target DNA to reach completion. For flat
Distance-Dependent Fluorescence Quenching
J. Phys. Chem. C, Vol. 114, No. 41, 2010 17831
Figure 1. Conceptual illustration of the surface assembly of conjugated polymer-dsDNA complexes and their immobilization on particles. The corresponding reflectance (A, C) and fluorescence (B, D) images of nanorods with polymer-dsDNA complexes without (A, B) and with (C, D) a 20 nm silica coating. Image scale bar ) 5 µm.
Au surfaces, DNA hybridization was carried out by spotting 100 µL of 1 µM target DNA onto the polymer-capture DNA bound substrates and allowing them to incubate at 40 °C for 1 h, followed by rinsing with the PB buffer. Small Dye Molecule Attachment. DyLight 488 was first tagged to dual functional DNA A by mixing 2 µL of 10 mg/ mL DyLight 488 NHS ester, 10 µL of 200 µM DNA A, 33 µL of H2O, and 5 µL of 10× sodium carbonate buffer (1.0 M, pH 9.0) together for 1 h. After purification using a micro biospin column (Bio-Rad), DyLight 488-tagged DNA A was attached to silica-coated nanorods via thiolated ends through SMCC coupling, as described in the previous section. For direct dye attachment on particles without DNA linkers, 50 µL of 0.1 mg/ mL DyLight 488 NHS ester was directly mixed with 50 µL of APTMS-functionalized silica-coated nanorods in the sodium carbonate buffer for 1 h. All particle solutions were eventually exchanged with PB buffer three times to remove unreacted reagents, and the particles were resuspended in 50 µL of PB buffer. Instrumentation. A JEOL JEM-100 CXII transmission electron microscope (TEM) was used to characterize the silicacoated nanorods and silica-coated gold nanoparticles. Samples were prepared by dropping 10 µL of particle solutions onto carbon-coated copper grids, followed by drying overnight. The thickness of the silica shells was measured by manually measuring the thickness of the gray portion of each particle. In each image, 2-3 particles were randomly measured to yield the averaged silica coating thickness. A Zeiss Axivert 35 inverted fluorescence microscope was used for optical imaging. The microscope was equipped with a bright-field reflectance filter set (Chroma, D495/40x, Q660DCLP dichroic, and 0.3 ND) for reflectance imaging of nanorods, a fluorescence filter set (Chroma, D405/40x excitation, Q460DCLP dichroic, and HQ510/50 m emission) for fluorescence imaging of conjugated polymer, and another fluorescence filter set (Chroma, HQ470/40x excitation, Q495LP dichroic, and HQ525/
50 m emission) for fluorescence imaging of DyLight 488. A 100 W mercury lamp was used as the light source throughout the experiments. Samples were prepared by first dropping 10 µL of the particle solution onto a glass slide. The nanorods were allowed to settle to the bottom of the slide for at least 2 min, followed by placing a coverslip atop. The assembly was sealed and flipped over for epi-imaging. All images were taken using a 63× oil immersion lens. Image J analysis software (NIH) was employed to quantify the fluorescence intensity of conjugated polymers on each particle. Briefly, 5-10 nanorods were randomly selected from each image and analyzed with Image J, which yielded a table of fluorescence intensity values of all pixels associated with those particles. The intensity values were then averaged to generate the mean fluorescence intensity and the statistic error at a 95% confidence interval. After subtracting the background, the mean intensity values were used in data plotting. The same method was used to estimate fluorescent signal intensities of the flat gold substrates where the mean value was averaged from several randomly selected regions of imaging (ROI). A PerkinElmer LS 50B luminescence spectrometer was used to acquire the fluorescence spectra from solutions of conjugated polymer-DNA-bound gold nanoparticles before and after hybridization with target DNA. The excitation wavelength was set at 420 nm with an excitation slit at 10 nm and an emission slit at 10 nm. An AutoEL-III Automatic ellipsometer (Rudoph Research) was used to measure the thickness of the silica film deposited on the gold substrates. The instrument irradiated the substrates at an incident angle of 70°; a reflective index of 1.457 was used for the silica films. Results and Discussion A typical assembly of polythiophene-dsDNA complexes on the Au/Ag striped nanorods during DNA detection is illustrated
17832
J. Phys. Chem. C, Vol. 114, No. 41, 2010
Zheng and He
Figure 2. Representative TEM images of silica-coated nanorods with the SiO2 thickness varied from 5 (A), 10 (B), 20 (C), and 40 nm (D) to 80 nm (E). The light gray layer is SiO2 due to its low electron density, whereas the black core is the metallic particle.
in Figure 1: a capture oligonucleotide sequence was premixed with cationic polythiothene derivatives to form a weakly fluorescent species in solution. The formed polymer-ssDNA duplexes were then covalently attached to the Au/Ag striped nanorods of a specific pattern. Subsequent hybridization of the oligonucleotide of the complementary sequence (i.e., target DNA) to the one on the particles resulted in formation of polymer-dsDNA triplexes, in which polymers underwent a conformational change that made the triplexes fluorescent strongly upon excitation. As expected, direct immobilization of polymer-dsDNA triplexes experienced strong fluorescence quenching by the Au and Ag surfaces underneath (Figure 1B). Moving the triplexes away from the metal surface, however, successfully restored the fluorescence (Figure 1D).22 To provide guidance for optimization of conjugated polymer-based bioassays on Au/Ag striped nanorods and gain insights on nonradiative energy transfer between an excited organic macromolecule and the nearby metal surface of similar length scales, we believe that quantitative characterization of the fluorescence behavior of conjugated polymers on or near metal surfaces is imperative, hence the following studies. Giving that the fluorescent quenching efficiency by a metal surface critically depends on the spacing between the excited species and the surface, silica layers of controlled thicknesses, ranging from 5 to 80 nm (sporadic coating for shells thinner than 5 nm), were prepared using the Sto¨ber method to tune the distance between the conjugated polymers and the metal surface.57 The thickness of the silica layer was first controlled by changing the deposition time, which increased from 10 min to yield a 5 nm coating to 60 min to achieve a 20 nm coating. Because further elongation of reaction time was known to result in formation of free silica particles without nanorods encapsulated, the particles precoated with 20 nm silica shells were used as the seeding materials to yield ∼40 nm coating after 60 min of incubation. Particles with an ∼80 nm coating were prepared with a third coating cycle. Figure 2 shows typical TEM images of silica-coated nanorods of various thicknesses. The silica coating in all cases was uniform across the particle surface. Little increase in surface roughness was observed, which eased concerns on arbitrary fluctuation in fluorescence signals caused by variations in surface roughness. It is also worth pointing out
that few free SiO2 particles, the primary byproduct of the sol-gel reaction, were observed in the TEM images. Polymer-dsDNA triplexes were assembled on the silica-coated nanorods using SMCC as the bridging reagent. Figure 3 shows the resulting fluorescence images of the nanorods attached with the same amount of polymer-dsDNA complexes but separated with SiOx of different thicknesses. It is clear that the measured fluorescence intensities increased rapidly as the silica coating grew thicker. Figure 3B shows a quantitative plot of averaged fluorescence intensity of the nanorods as a function of the silica shell thickness. As expected, the fluorescence intensity was increased steeply on the nanorod surface initially as the silica thickness increased from 0 to 20 nm, confirming the effective quenching of fluorescence by the Au/Ag striped nanorod surface within this range.34 When the silica thickness increased beyond 20 nm, the fluorescent signal continued to increase, but at a much slower rate, until the curve gradually leveled off. For further quantitative characterization of distance-dependent fluorescence quenching phenomena, a more commonly used term, quenching efficiency (QE), as QE ) 1 - Id/I∞, is used, where Id is the fluorescence intensity at a given separation distance and I∞ is the fluorescence intensity at a distance far from a metal surface. The value of QE defines the relative amount of fluorescence intensity lost, in this case, primarily due to quenching by the metal surface. In an imaging mode, I∞ cannot be directly measured. Rather, the fluorescence output from particles coated with an 80 nm silica coating was used instead, primarily because the fluorescence intensity at an 80 nm silica coating was close to leveling off with little fluorescence increase when the spacing between the polymer complexes and the metal surface increased. Figure 3C shows a plot of calculated QE against silica shell thickness. More than 75% of the fluorescence was lost when the polymer complexes were placed within 20 nm of the metal surface. Although the fluorescence decay has been mostly fitted by the fourth power of the spacing distance between a dipole and an infinite metal surface in the literature,46 we found the decay curve here is best fitted by an exponential decay model where QE ) QE0 × exp(-d/d0),58 where QE0 is the quenching efficiency when polymers were directly attached to the metal surface, d0 the distance constant within which fluorescence quenching occurs, and d the separating distance between the polymer and the metal surface, that is, the silica coating
Distance-Dependent Fluorescence Quenching
J. Phys. Chem. C, Vol. 114, No. 41, 2010 17833
Figure 3. Distance-dependent fluorescence quenching of conjugated polymer-dsDNA complexes on Au/Ag striped nanorods. (A) Fluorescence images of nanorods with polymer-dsDNA triplexes attached. The thickness of the silica coating was varied from 0 to 80 nm, as specified on each panel (scale bar ) 5 µm). (B) A plot of fluorescence intensities per pixel as a function of silica coating thickness. The solid line is a simple connection of experimental data points. (C) A plot of the quenching efficiency of conjugated polymer-dsDNA complexes on the metal surface against the thickness of the silica coating. The solid line is a mathematical fitting of QE ) 1.03 × exp(-d/17.85).
thickness. The fitting yielded an equation of QE ) 1.03 × exp(-d/ 17.85). The estimated QE0 value was close to a unit, indicating complete quenching of conjugated polymers on the metal surface at direct contact. The fitted d0 value of 17.85 nm was also in agreement with the theoretical and experimental values obtained from quenching of small organic dyes on the metal surface.39-41 It is important to point out that the plot used the silica shell thickness to represent the separation distance between conjugated polymer complexes and the metal surface, regardless that the overall polymer complexes could stretch ∼10 nm away from the surface due to electrostatic repulsion. This assumption was made based on the delocalized electronic structure of conjugated polymers, which allowed electrons to freely migrate along the backbone with little energy barrier. The well-fitted exponential decay curve confirms that the spacing between the attaching point of the polymer on silica and the metal surface is the most relevant in energy transfer between two species; hence, it is used in further study. To further examine the similarity of the quenching behavior between small organic dye molecules and conjugated polymers, we selected DyLight 488, which offers a similar emission profile (Abs, 493 nm; Em, 518 nm) to that of polythiophene. Quenching of DyLight 488 was studied by attaching dye molecules to the particles coated with silica either directly via its amino-reactive groups or using dual functional ssDNA A as the bridge. The latter attachment ensured that the surface density of small dye molecules is the same as the conjugated polymers, which was controlled by the surface density of DNA where electrostatic repulsion dictated the final packing density. The use of DNA molecules as the bridging materials further eliminates any possible ambiguity raised from the contribution of DNA
molecules in quenching. Figure 4A shows fluorescence intensities of dye-attached nanorods versus the thickness of the silica coating. With or without DNA, both systems exhibited the expected monotonic increasing profile and, interestingly, the same slowing-down point around 20 nm of silica coating. For dye molecules attached through ssDNA A, fluorescence was not completely quenched on bare particles due to the dead-length of DNA molecules, which placed the dye molecule away from the metal surface even in the absence of the silica layer.32-34 Note that, at the higher silica coating thickness (>20 nm) where the fluorescence quenching is minimized, the fluorescence intensity of the dye via direct attachment was still much lower than that with the DNA spacer. It is attributed to the relatively high density of dye molecules immobilized through direct attachment, which may experience intermolecular self-quenching when the dye probes were crowded. Regardless that the dye molecules were attached directly on a surface or through DNA molecules, the initial fast fluorescence recovery, followed by a gradual leveling-off, qualitatively resembled that of polymer complexes-attached particles (Figure 4A). Figure 4B shows the quenching efficiency profile of DyLight 488 on nanorods. For comparison, the quenching profile of conjugated polymer complexes was also plotted in black squares. It is interesting to note that, for dye molecules directly attached to the particle surface (green triangles), the quenching efficiency showed a similar exponential decay trend to the increase of the silica thickness (QE ) 1.04/exp(-d/19.7) and overlapped well with that of conjugated polymers. As aforementioned, the seemingly less effective quenching observed from the dye molecules attached through the DNA spacer was due to the additional
17834
J. Phys. Chem. C, Vol. 114, No. 41, 2010
Zheng and He
Figure 5. Comparison of distance-dependent fluorescence quenching profiles of conjugated polymer-dsDNA complexes on Au/Ag striped nanorods (black square), 13 nm GNPs (red dot), and 2-D flat gold substrates (green triangle). The solid lines are mathematical fittings of various systems with the equations labeled in the legend.
Figure 4. (A) Plots of the fluorescence intensity of DyLight 488 on nanrods versus the thickness of the silica coating with (red dot) or without (green triangle) DNA as the spacer and polymer-dsDNA complexes on the nanorods (black square). (B) Plots of the fluorescence quenching efficiency of conjugated polymer-dsDNA complexes (black square) and DyLight 488 on nanorods with (red dot) or without (green triangle) DNA as the spacer. The solid lines are mathematical fittings of various systems with the equations labeled in the legend.
separation, introduced by the DNA spacer, between dye molecules and the surface (red dots). The best-fitted equation (QE ) 1.04/exp((-d + 4.29)/16.05) suggests that the additional thickness introduced by DNA was ∼4.3 nm, which is consistent with the loose structure of ssDNA (35-mer) attached at its end point. Indeed, taking this estimated length of the ssDNA spacer (4.29 nm) into account, the adjusted plot overlapped well with that of the dye molecules directly attached to the silica-coated nanorods (Supporting Information, Figure S1). It is known, and has been numerously reported, that fluorescence quenching efficiency of small dye molecules absorbed on the metal surface varies when the size and shape of the metal substrate underneath changes.26 The molecular weight of conjugated polythiophene is hundreds of times higher than that of small organic dye molecules, and its linear structure often leads to a much bigger footprint when it comes to surface coating. Hence, an interesting question yet to be answered is how the relative length scale of the excited species to the metallic surface underneath affects local energy transfer between the donor and the acceptor. To study this question, two additional metal substrates were selected for comparison: (a) 13 nm gold nanoparticles and (b) flat 2-D gold substrates. Gold was selected as the metal medium because it was the primary metal in the nanorods we used (Au/Ag ) 5:1). In the designed experiments, the separation distance between the attached conjugated polymer complexes and the metal surfaces was again varied by changing
the silica coating of different thicknesses. In particular, by repeated addition of TEOS during shell growth, silica-coated GNPs with thicknesses of 5.0 ( 1.0, 10.0 ( 0.8, 20.0 ( 0.6, 30.0 ( 0.5, and 60.0 ( 0.3 nm were obtained (Supporting Information, Figure S2). Sol-gel deposition on flat Au substrates yielded film thicknesses of 4.6 ( 0.4, 9.5 ( 1.6, 21.6 ( 1.5, 32.5 ( 2.6, and 81.6 ( 6.9 nm (Supporting Information, Figure S3). The same attachment chemistry was used for all surfaces to ensure comparable attachment efficiency across. To quantify fluorescence intensities of conjugated polymers, the fluorescence spectra of conjugated polymer-dsDNA bound GNP solutions were collected and the peak intensities were used. It is important to point out that, in the case of the polymer complex-bound GNPs, the increase of the silica coating thickness resulted in an increase in the total surface area, that is, an increase in the absolute amount of polymer complexes bound. Because the fluorometer measures bulk fluorescent signals, this increase of an absolute amount of polymer complexes could not be easily separated from reduced quenching. To eliminate any ambiguity raised from the changes in surface loading, the amount of conjugated polymer complexes bound on each particle was controlled by introducing the same small amount of polymer-dsDNA complexes to the reaction mixtures. For flat surfaces, fluorescence images were captured directly, and the fluorescence intensities were averaged from a few randomly selected regions of interest (ROI). Figure 5 shows direct comparison of distance-dependent fluorescence quenching of conjugated polymers on three metal substrates of various sizes and shapes. All substrates showed similar exponential decay with fitting equations of QE ) 1.03 × exp(-d/ 17.85) for nanorods, QE ) 1.036 × exp(-d/19.75) for GNPs, and QE ) 1.03 × exp(-d/17.25) for flat gold substrates. The similar QE0 values for all three systems suggest that QE0 is primarily determined by the chemical nature of the donor (i.e., conjugated polymer) and the acceptor (i.e., Au surface). Two more points can be concluded from the fitting equations: (1) Free electron migration through the polymer backbone makes the fact that the absolute size of the conjugated polymers is much larger than the small dipole molecule irrelevant during surface quenching. (2) The subtle differences in the d0 value, 19.8 (GNP) versus 17.9 (nanorod) versus 17.3 (flat substrate), are in agreement with the literature where the larger donor/acceptor size ratio often leads to a larger d0 value.58 In other words, the smaller the energy acceptor, the less efficient it is to quench fluorescence.
Distance-Dependent Fluorescence Quenching Conclusion In this report, we have investigated the distance-dependent fluorescence quenching behavior of conjugated polymer-dsDNA complexes on the Au/Ag striped nanorods. Our results show that the quenching efficiency of conjugated polymers on the nanorods decays exponentially as the coating thickness of the silica layer increases. An inert coating of 20 nm is the minimal spacing needed for fluorescence restoration. The quenching efficiency of conjugated polymers is found to be comparable to that of small dye molecules atop metal surfaces where the natural length of the conjugated polymer is negligible. The finding does not contradict the fast electron transfer model proposed in other reports to explain the superquenching behavior of conjugated polymers. Similar distance-dependent quenching fitting equations are obtained for polymer-dsDNA complexes, regardless that they were attached on 13 nm gold nanoparticles, 6 µm nanorods, or 2-D flat gold substrates. The observation suggests that quenching efficiency is dictated primarily by the chemical properties of the donor and acceptor and, to a small extent, by the relative size of the donor-to-acceptor ratio. Acknowledgment. We thank Dr. Simon Lappi for the help in the use of the fluorescence microscope. We also thank Drs. Michael Natan and Gabriela Chakarova at Oxnonica Inc. for providing us with Au/Ag striped nanorods. This work was partially supported by the NSF (No. 0644865). Supporting Information Available: An adjusted plot of the fluorescence quenching of DyLight 488 on nanorods after taking the length of the 35-mer DNA spacer into consideration, TEM images of silica-coated gold nanoparticles of different thicknesses and the corresponding fluorescence spectra of bound conjugated polymer, and a plot of the measured silica film thickness against the amount of TEOS added during silica coating on the Au substrates and their corresponding fluorescence images of absorbed conjugated polymer. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Heeger, P. S.; Heeger, A. J. Proc. Natl. Acad. Sci. 1999, 96, 12219– 12221. (2) Chen, L.; McBranch, D. W.; Wang, H. L.; Hegelson, R.; Wudl, F.; Whitten, D. C. Proc. Natl. Acad. Sci. 1999, 96, 12287–12282. (3) McQuade, D. T.; Pullen, A. E.; Swager, T. M. Chem. ReV. 2000, 100, 2537–2574. (4) Thomas, S. W.; Joly, G. D.; Swager, T. M. Chem. ReV. 2007, 107, 1339–1386. (5) Liu, B.; Bazan, G. C. Chem. Mater. 2004, 16, 4467–4476. (6) Ho, H. A.; Najari, A.; Leclerc, M. Acc. Chem. Res. 2008, 41, 168– 178. (7) Feng, F.; He, F.; An, L.; Wang, S.; Li, Y.; Zhu, D. AdV. Mater. 2008, 20, 2959–2964. (8) Jiang, H.; Tarenekar, P.; Reyonlds, J. R.; Schanze, K. S. Angew. Chem., Int. Ed. 2009, 48, 4300–4316. (9) Gaylord, B. S.; Heeger, A. J.; Bazan, G. C. Proc. Natl. Acad. Sci. 2002, 99, 10954–10957. (10) Liu, B.; Bazan, G. C. Proc. Natl. Acad. Sci. 2005, 102, 589–593. (11) Baker, E. S.; Hong, J. W.; Gaylord, B. S.; Bazan, G. C.; Bowers, M. T. J. Am. Chem. Soc. 2006, 128, 8484–8492. (12) Ho, H. A.; Boissinot, M.; Bergeron, M. G.; Corbeil, G.; Dore, K.; Boudreau, D.; Leclerc, M. Angew. Chem., Int. Ed. 2002, 41, 1548–1551. (13) Dore´, K.; Dubus, S.; Ho, H. A.; Le´vesque, I.; Brunette, M.; Corbeil, G.; Boissinot, M.; Boivin, G.; Bergeron, M. G.; Boudreau, D.; Leclerc, M. J. Am. Chem. Soc. 2004, 126, 4240–4244. (14) Raymond, F. R.; Ho, H. A.; Peytavi, R.; Bissonnette, L.; Boissinot, M.; Picard, F. J.; Leclerc, M.; Bergeron, M. G. BMC Biotechnol. 2005, 5, 10. (15) Xu, H.; Wu, H.; Huang, F.; Song, S.; Li, W.; Cao, Y.; Fan, C. Nucleic Acid Res. 2005, 33, e83.
J. Phys. Chem. C, Vol. 114, No. 41, 2010 17835 (16) Duan, X.; Li, Z.; He, F.; Wang, S. J. Am. Chem. Soc. 2007, 129, 4154–4155. (17) Nilsson, K. P. R.; Inganas, O. Nat. Mater. 2003, 2, 419–426. (18) Zhang, Y.; Li, Z.; Cheng, Y. L. X. Chem. Commun. 2009, 3172– 3174. (19) Herland, A.; Ingana¨s, O. Macromol. Rapid Commun. 2007, 28, 1703–1713. (20) Abe´rem, M. B.; Najari, A.; Ho, H. A.; Gravel, J. F.; Nobert, P.; Boudreau, D.; Leclerc, M. AdV. Mater. 2006, 18, 2703–2707. (21) Wang, L.; Liu, B. Langmuir 2009, 25, 12787–12793. (22) Zheng, W.; He, L. J. Am. Chem. Soc. 2009, 131, 3432–3433. (23) Nicewarner-Pena, S. R.; Freeman, R. G.; Reiss, B. D.; He, L.; Pena, D. J.; Walton, I. D.; Cromer, R.; Keating, C. D.; Natan, M. J. Science 2001, 294, 137–141. (24) Keating, C. D.; Natan, M. J. AdV. Mater. 2003, 15, 451–454. (25) Moskovits, M. ReV. Mod. Phys. 1985, 57, 783–826. (26) Chance, R. R.; Prock, A.; Silbey, R. AdV. Chem. Phys. 1978, 37, 1–65. (27) Weiz, D. A.; Groff, S.; Gersten, J. I.; Nitzan, A. J. Chem. Phys. 1983, 78, 5324–5338. (28) Lakowicz, J. R. Anal. Biochem. 2001, 298, 1–24. (29) Lakowicz, J. R. Anal. Biochem. 2005, 337, 171–194. (30) Gryczynski, I.; Malicka, J.; Gryczynski, Z.; Lakowicz, J. R. J. Phys. Chem. B 2004, 108, 12568–12574. (31) Hong, B.; Kang, K. A. Biosens. Bioelectron. 2006, 21, 1333–1338. (32) Stoermer, R. L.; Keating, C. D. J. Am. Chem. Soc. 2006, 128, 13243–13254. (33) Dulkeith, E.; Ringler, M.; Klar, T. A.; Feldmann, J.; Javier, A. M.; Parak, W. J. Nano Lett. 2005, 5, 585–589. (34) Jennings, T. L.; Singh, M. P.; Strouse, G. F. J. Am. Chem. Soc. 2006, 128, 5462–5467. (35) Sokolov, K.; Chumanov, G.; Cotton, T. M. Anal. Chem. 1998, 70, 3898–3905. (36) Kulakovich, O.; Strekal, N.; Yaroshevich, A.; Maskevich, S.; Gaponenko, S.; Nabiev, I.; Woggon, U.; Artemyev, M. Nano Lett. 2002, 2, 1449–1452. (37) Malicka, J.; Gryczynski, I.; Gryczynski, Z.; Lakowicz, J. R. Anal. Biochem. 2003, 315, 57–66. (38) Schneider, G.; Decher, G.; Nerambourg, N.; Praho, R.; Werts, M. H. V.; Blanchard-Desce, M. Nano Lett. 2006, 6, 530–536. (39) Aslan, K.; Wu, M.; Lakowicz, J. R.; Geddes, C. D. J. Am. Chem. Soc. 2007, 129, 1524–1525. (40) Wokaun, A.; Lutz, H.-P.; King, A. P.; Wild, U. P.; Ernst, R. R. J. Chem. Phys. 1983, 79, 509–514. (41) Cheng, D.; Xu, Q. H. Chem. Commun. 2007, 248–250. (42) Tovmachenko, O. G.; Graf, C.; van den Heuvel, D. J.; van Blaaderen, A.; Gerritsen, H. C. AdV. Mater. 2006, 18, 91–95. (43) Chi, Y. S.; Byon, H. R.; Lee, B. S.; Kong, B.; Choi, H. C.; Choi, I. S. AdV. Funct. Mater. 2008, 18, 3395–3402. (44) Seelig, J.; Lesile, K.; Renn, A.; Kunh, S.; Jacobsen, V.; Van de Corput, M.; Wyman, C.; Sandoghdar, V. Nano Lett. 2007, 7, 685–689. (45) Kunh, S.; Mori, G.; Agio, M.; Sandoghdar, V. Mol. Phys. 2008, 106, 893–908. (46) Kuhn, H. J. Chem. Phys. 1970, 53, 101–108. (47) Swager, T. M. Acc. Chem. Res. 1998, 31, 201–207. (48) Swager, T. M.; Wosnick, J. H. MRS Bull. 2002, 27, 446–450. (49) Fan, C.; Wang, S.; Hong, J. W.; Bazan, G. C.; Plaxco, K. W.; Heeger, A. J. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 6297–6301. (50) Reiss, B. D.; Freeman, R. G.; Walton, I. D.; Norton, S. M.; Smith, P. C.; Stonas, W. G.; Keating, C. D.; Natan, M. J. J. Electroanal. Chem. 2002, 522, 95–103. (51) Nicewarner-Pen˜a, S. R.; Carado, A. J.; Shale, K. E.; Keating, C. D. J. Phys. Chem. B 2003, 107, 7360–7367. (52) Grabar, K. C.; Freeman, R. G.; Hommer, M. B.; Natan, M. J. Anal. Chem. 1995, 67, 735–743. (53) Sioss, J. A.; Stoermer, R. L.; Sha, M. Y.; Keating, C. D. Langmuir 2007, 23, 11334–11341. (54) Graf, C.; Vossen, D. L. J.; Imhof, A.; van Blaaderen, A. Langmuir 2003, 19, 6693–6670. (55) Kambhampati, D. K.; Jakob, T. A. M.; Robertson, J. W.; Cai, M.; Pemberton, J. E.; Knoll, W. Langmuir 2001, 17, 1169–1175. (56) Thompson, W. R.; Pemberton, J. E. Chem. Mater. 1995, 7, 130– 136. (57) Sto¨ber, W.; Fink, A.; Bohn, E. J. Colloid Interface Sci. 1968, 26, 62– 69. (58) Chhabra, R.; Sharma, J.; Wang, H.; Zou, S.; Lin, S.; Yan, H.; Lindsay, S.; Liu, L. Nanotechnology 2009, 20, 485201.
JP1070693
