View Article Online View Journal

Nanoscale Accepted Manuscript

This article can be cited before page numbers have been issued, to do this please use: J. K. Park, G. Y. Lee, J. Kinam, D. Ko, I. K. Han and H. Ko, Nanoscale, 2015, DOI: 10.1039/C5NR02813B.

This is an Accepted Manuscript, which has been through the Royal Society of Chemistry peer review process and has been accepted for publication. Accepted Manuscripts are published online shortly after acceptance, before technical editing, formatting and proof reading. Using this free service, authors can make their results available to the community, in citable form, before we publish the edited article. We will replace this Accepted Manuscript with the edited and formatted Advance Article as soon as it is available. You can find more information about Accepted Manuscripts in the Information for Authors. Please note that technical editing may introduce minor changes to the text and/or graphics, which may alter content. The journal’s standard Terms & Conditions and the Ethical guidelines still apply. In no event shall the Royal Society of Chemistry be held responsible for any errors or omissions in this Accepted Manuscript or any consequences arising from the use of any information it contains.

www.rsc.org/nanoscale

Please doNanoscale not adjust margins

Page 1 of 5

View Article Online

Journal Name COMMUNICATION

Published on 03 July 2015. Downloaded by Freie Universitaet Berlin on 07/07/2015 04:51:51.

Enhanced triplet–triplet annihilation in bicomponent organic system by using gap plasmon resonator Received 00th January 20xx, Accepted 00th January 20xx

Jun Kue Park,

a,b

a

a

c

a

a

Gi Yong Lee, Kinam Jung, Doo-Hyun Ko, Il Ki Han, and Hyungduk Ko *

DOI: 10.1039/x0xx00000x www.rsc.org/

The triplet–triplet annihilation (TTA) efficiency in bicomponent organic systems is investigated by employing a gap plasmon resonator. In our structure, strong absorption peaks arising from coupling between localized surface plasmons and surface plasmon polaritons closely overlap the Q band of porphyrin, leading to higher triplet concentrations within the film. We find that at ultralow excitation intensities on the order of watts per square −2 centimeter (Wcm ), TTA becomes predominant for the organic system on a gap plasmon resonator. A strong surface-enhanced Raman scattering intensity is observed in this substrate, verifying the near-field enhancement.

Photon upconversion phenomena have recently been suggested as alternatives to overcome the Shockley–Queisser solar power 1–7 conversion limit. Studies on upconversion in bicomponent organic systems leading to the upconversion have thus been widely pursued owing to its potential application in photovoltaic devices 1–4 and photocatalytic electrochemical cells. The upconversion mechanism in bicomponent systems is associated with the triplet– triplet annihilation (TTA) process, in which a combination of two molecules in excited triplet states gives rise to both ground state 1,3,5 and higher-energy singlet products. In solid-state systems, however, the upconversion efficiency is markedly decreased (by a 3 2,6,7 factor ~10 ) compared to that in solutions. The low efficiency in solid-state systems is mainly the result of the much lower molecular mobility in solid-state systems than in solutions. The lower molecular mobility results in a short exciton diffusion length, limiting the energy transfer as well as TTA, in turn causing a low 2,4,6,7 quantum yield. Thus, realization of efficient solid-state systems showing upconversion remains an important goal. Plasmonic nanostructures have been widely studied to improve the optical properties of the emission intensity of fluorophores a.

Nanophotonics Research Center, Korea Institute of Science and Technology, Seoul 136-791, Republic of Korea Korea Multi-purpose Accelerator Complex, Korea Atomic Energy Research Institute, Gyeongju 780-904, Republic of Korea c. Department of Applied Chemistry, Kyung Hee University, Yongin 446-701, Republic of Korea b.

Nanoscale Accepted Manuscript

DOI: 10.1039/C5NR02813B

through excitation of localized surface plasmons (LSPs) or propagation of surface plasmon polaritons (SPPs). In particular, in recent years, the metal-nanostructure-induced upconversion luminescence has been extensively studied to enhance the emission 5,8,9 intensity. However, there have been few reports on the plasmon-enhanced TTA process, whereas most plasmonic studies 8,9 have examined the lanthanide phosphors. Poorkazem et al. recently reported plasmon-enhanced TTA using Ag nanoplates in 5 which higher overall triplet concentrations were observed. They explained that the strong overlap between the LSP bands and the Q band of porphyrin improves absorption in the film owing to local 1,5 field enhancement near metal nanoparticles. However, research on the LSP-induced upconversion luminescence, including studies of lanthanide phosphors and the TTA process, has not yielded a notable luminescence enhancement. Therefore, stronger local field enhancement in the excitation wavelength range is required to increase the absorption in the fluorophore. In this study, we investigate the TTA efficiency in thin films containing platinum octaethylporphyrin (PtOEP) as a triplet sensitizer and 9,10-diphenylanthracene (DPA) as a fluorescent emitter on the plasmon substrates deriving the gap plasmon resonance. In this structure, the constructive interference that gives rise to standing-wave resonances is expected to exhibit a more enhanced absorption cross section than the resonance of only LSPs 10–12 or SPPs. If the absorption bands with strong resonance can be well matched with the Q band of porphyrin, higher triplet concentrations can be created in the film, enhancing the 4,5 upconversion luminescence. As a result, we demonstrate that the upconversion luminescence is efficiently increased by realizing an optimized gap plasmon mode overlapping the Q band of porphyrin by fabricating a metal–insulator–metal (MIM) structure based on an array of triangular silver nanoplates (silver nanoplate array, AgNA). Figure 1a schematically shows the structure of the bicomponent organic film formed on the gap plasmon resonator. It is composed of a DPA:PtOEP film, AgNA, SiO2 film, and Ag thin film. As shown in the scanning electron microscopy (SEM) image in Figure 1b, the AgNA appears to be well-aligned periodically, and the nearestneighbor distance between two particles in the AgNA is around 240 nm, which is consistent with the designed schematic, as indicated

This journal is © The Royal Society of Chemistry 20xx

J. Name., 2013, 00, 1-3 | 1

Please do not adjust margins

Please doNanoscale not adjust margins COMMUNICATION

Page 2 of 5 Journal Name View Article Online

Figure 1. (a) Schematic of gap plasmon resonator (AgNA/SiO2/Ag). (b) SEM image of AgNA on SiO2 spacer (40 nm) and Ag film. (c) Surface morphology of gap plasmon resonator obtained by AFM. Lower graph shows cross section of the Ag array along red line. The average height of the silver nanoparticles is 50 ± 10 nm, and the nearest-neighbor distance between the particles is ~240 nm, which is consistent with that in Figure 1a. (d) Absorption spectra of DPA:PtOEP film, SiO2/Ag, and AgNA/SiO2/Ag substrates. The AgNA/SiO2/Ag substrate exhibits double bands centered at ~455 and ~530 nm. Vertical arrow denotes the excitation wavelength of λexc = 532 nm for photoluminescence measurement.

in Figure 1a. The AgNAs are 50 nm thick with diameters of ~120 nm. The particles are plate-like in structure and pseudotriangular in shape. Unlike the case of spherical nanoparticles, the anisotropic shape of the AgNAs may shift the position of the LSP resonance band toward the visible region, producing larger near-field 5,13,14 enhancements. To study the bicomponent organic film on the nanostructured surfaces, our proposed structure was compared with those on glass and SiO2/Ag substrates. The absorption spectra for the DPA:PtOEP film, SiO2/Ag, and AgNA/SiO2/Ag substrates are shown in Figure 1d. The AgNA/SiO2/Ag substrate clearly exhibits a strong plasmon 4,5,15 resonance band centered at 530 nm. Among the AgNA/SiO2/Ag substrates, that with a 40-nm-thick spacer exhibits the strongest absorption (see Supporting Information, Figure S1). The absorption peak for the AgNA/SiO2/Ag substrate at a wavelength of 530 nm closely overlaps the Q band of the DPA:PtOEP film centered at 540 nm. Thus, the AgNA/SiO2/Ag substrate is expected to efficiently yield near-field enhancement at the excitation wavelength of 532 nm, critically increasing the density of the excited triplet states in 5 the DPA:PtOEP film. In addition, the AgNA in the MIM structure is expected to produce efficient scattering due to plasmon resonant coupling with the excitation wavelength, enhancing the absorption in the DPA:PtOEP film as a result of the increased optical path 10 length. To ensure the plasmon resonance feature of the AgNA/SiO 2/Ag substrate, we analyzed the electromagnetic field distribution of the

Nanoscale Accepted Manuscript

Published on 03 July 2015. Downloaded by Freie Universitaet Berlin on 07/07/2015 04:51:51.

DOI: 10.1039/C5NR02813B

Figure 2. (a) Schematic of a 3D electromagnetic simulation model for a AgNA MIM plasmonic structure with pitches in the x and y axes when unpolarized light is irradiated along the –z axis. (b) Simulated absorption spectrum of AgNA MIM structure showing two distinct surface plasmon resonances at λ = 435 and 570 nm. (c) Simulated E-field distribution of AgNA MIM: at λ = 435 nm in xy (top left) and xz planes (bottom left), and at λ = 570 nm in xy (top right) and xz planes (bottom right).

AgNA MIM structure using the three-dimensional (3D) finitedifference time-domain (FDTD) method (FDTD Solutions, Canada). AgNA patterns that were modeled to have flat and rounded corners based on the SEM images in Figure 1b were imported into the FDTD model. Because of the symmetries along the x and y axes of the plasmonic AgNA structure and the imposed plane-wave source in the structure, antisymmetric or symmetric boundary conditions (BCs) were applied to the x and y axes by considering the periodicities of the structure. For the transverse magnetic (TM) or xpolarized case, we set the antisymmetric BC on the x axis and the symmetric BC on the y axis; for the transverse electric (TE) or ypolarized case, we set the antisymmetric BC on the y axis and the symmetric BC on the x axis. A perfectly matched layer BC was implemented in the z direction for the TE and TM simulations to absorb all of the back-reflected electromagnetic waves from the 2 boundary. By averaging the E-field intensities (|E| ) for normal incident TE and TM polarized plane-wave sources in the visible range of 350–700 nm propagating in the –z direction above the plasmonic structure, we could realize unpolarized light conditions for the FDTD simulation. Two-dimensional power monitors that record the amount of normalized power transmitted through them were inserted into the model to evaluate the absorption spectra of the structure. We also employed a 3D field profile monitor to detect the electromagnetic field distribution of the plasmonic structure. The simulated absorption profile of the AgNA MIM in Figure 2c has two prominent surface plasmon peaks at wavelengths of 435

2 | J. Name., 2012, 00, 1-3

This journal is © The Royal Society of Chemistry 20xx

Please do not adjust margins

Page 3 of 5

Please doNanoscale not adjust margins

Journal Name

COMMUNICATION View Article Online

Nanoscale Accepted Manuscript

Published on 03 July 2015. Downloaded by Freie Universitaet Berlin on 07/07/2015 04:51:51.

DOI: 10.1039/C5NR02813B

Figure 4. Obtained power dependence of samples on (a) bare, (b) SiO2/Ag, and (c) AgNA/SiO2/Ag substrates. Upconversion and phosphorescence peak intensities were measured at 430 and 650 nm, respectively.

Figure 3. Emission spectra of samples on SiO2/Ag, AgNA/SiO2/Ag, and bare substrates showing both (a) DPA fluorescence originating from TTA and (b) PtOEP fluorescence (I) and phosphorescence (II). (c) Schematic energy level diagram of TTA-supported upconversion process. The similar triplet levels of PtOEP (1.91 eV) and DPA (1.78 20 eV) permit highly probable triplet energy transfer. S, sensitizer; E, emitter; S0, ground state; S1, first excited singlet state; T1, first excited triplet state; ISC, intersystem crossing; TET, triplet energy transfer; TTA, triplet–triplet annihilation.

and 570 nm, which coincide well with the experimental UV-vis absorption peaks in Figure 1d. Near-field enhancement around the plasmonic structures interacting with normal incident light at the two distinctive plasmon resonances is depicted in Figure 2b. At the first broad peak of λ = 435 nm, the near-field intensity of the plasmonic MIM structure in the xy and xz planes is highly confined in a nanoscopic area at the rounded corners of the triangular patterns owing to interaction between the LSP from the AgNA and 5 the incident light. Moreover, from the near-field profile in the xy plane at the second narrow peak at λ = 570 nm, concentrated local fields around the triangular pattern are observed. Further, from the

electric field in the xz plane around the nanoplates at a wavelength of 570 nm, we can observe a much stronger local electric field inside the SiO2, which arises from the gap plasmon mode. It is thus evident that the absorption peak around the wavelength of 530 nm is a result of the strong gap plasmon resonance mode, which closely 4,10,11,13–17 overlaps the excitation wavelength. Therefore, when an upconverting DPA:PtOEP film is formed on the plasmonic MIM substrate, the AgNA MIM substrate would be expected to effectively absorb and scatter the incident photons of the excitation wavelength within the medium. This would critically increase the density of the excited triplet states of the DPA:PtOEP molecules and extend the optical path length inside the molecules. Figure 3 shows the upconversion luminescence and phosphorescence of DPA:PtOEP on three different types of substrates under excitation at a wavelength of 532 nm. Figure 3a clearly shows that the intensity of the upconversion luminescence of the sample on a gap plasmon resonator is strongly enhanced. The emission was up to ~75 times higher than that of the sample on the bare substrate. The emission intensity of the sample on the SiO2/Ag substrate was also greater than that of the sample on the bare substrate. The fluorescence and phosphorescence of PtOEP appear at ~565 and 650 nm, as denoted by I and II, respectively, in Figure 3b. The highest ratio of fluorescence intensity to the phosphorescence intensity of PtOEP is obtained for the gap plasmon resonator, whereas the lowest appears for the bare substrate. The greatest intensity for the gap plasmon resonator is attributed to the enhanced absorption band due to coupling between LSPs and SPPs, which closely overlaps the Q band of 5,15,18 porphyrin. Direct transfer of the excited triplet level in PtOEP

This journal is © The Royal Society of Chemistry 20xx

J. Name., 2013, 00, 1-3 | 3

Please do not adjust margins

Please doNanoscale not adjust margins

Page 4 of 5 Journal Name

Figure 5. Raman spectra of DPA:PtOEP on bare, SiO2/Ag, and AgNA/SiO2/Ag substrates. Spectra of samples on bare and SiO2/Ag substrates were magnified by the indicated values to identify the peaks. To clarify the Raman signal, the fluorescence backgrounds were removed for the samples on AgNA/SiO2/Ag and SiO2/Ag substrates. The measurement was made with 532 nm excitation under the same conditions for all the samples.

molecules to the triplet state of the blue-emitting molecules is followed by effective TTA between the triplet states of the blueemitting molecules (see Figure 3c). When the triplet energy transfer is more efficient between the two molecules, the phosphorescence emission becomes weaker than the fluorescence emission from the 19,20 singlet state in the PtOEP molecule. Therefore, a higher ratio of the fluorescence intensity to the phosphorescence intensity of PtOEP on the gap plasmon resonator indicates efficient triplet energy transfer between the molecules, which in turn increases the upconversion luminescence. In addition, Figure S2 shows that the TTA efficiency depends on the thickness of the spacer; the substrate with a 40-nm-thick spacer shows more efficient upconversion via TTA than the other substrates. The higher upconversion intensity relative to the phosphorescence intensity indicates a higher triplet concentration in the film, which causes more efficient triplet energy 19,20 transfer. The low extinction coefficient of the Q band for porphyrins 2,5,6 results in low triplet concentrations and TTA efficiency. Thus, the main deactivation channel is unimolecular in nature with negligible TTA. At low excitation powers, unimolecular triplet decay yields a quadratic dependence on the excitation power, whereas at high excitation powers, TTA becomes favored, leading to a more linear 5,6 dependence on the excitation power. To probe the dependence of the TTA process on the PtOEP triplet concentration, the power

dependence of all the samples was measured. FigureView 4 shows the Article Online DOI: 10.1039/C5NR02813B excitation power dependence of the upconversion intensity versus the phosphorescence intensity. The excitation power was controlled using six different combinations of neutral density filters. Because the phosphorescence intensity depends linearly on the excitation power, the plots in Figure 4 indicate the efficiency of the TTA process, correcting additionally for any photobleaching that 5, 6 may occur at the high power density used. In Figure 4a, the slope of the best-fit line for DPA:PtOEP on the bare substrate shows a roughly quadratic dependence of the TTA process. The value obtained is comparable to the value of 1.83 reported in previous 5 work. The slope for the sample on the AgNA substrate shows a decrease, suggesting that a greater concentration of the triplets 5,21 decays via TTA than via unimolecular triplet decay. In Figure 4c, the slope for the sample on the gap plasmon resonator is lower than that of the sample on SiO2/Ag, which is consistent with an increased level of light absorption for the gap plasmon resonator compared to the SiO2/Ag substrate, as shown in Figure 1d. The best-fit line for the sample on the gap plasmon resonance substrate has a slope of 1.03 ± 0.12, indicating a nearly linear dependence of the TTA process. The respective upconversion luminescence and phosphorescence as a function of the excitation power are displayed in Figure S3; DPA upconversion luminescence resulting from the TTA process is clearly observed for the system on the gap plasmon resonator in particular. To directly probe the electromagnetic field enhancement near the nanostructure surface, we employed Raman spectroscopy, 17,22 which specifies the vibration peaks of adsorbed molecules. Surface-enhanced Raman scattering (SERS) may show an increased scattering intensity by tens of orders of magnitude compared to 4,13,17 Raman scattering due to a weak process. In Figure 5, the Raman bands for the DPA:PtOEP sample on the gap plasmon resonator clearly appear. Most of the Raman bands indicated in Figure 5 appear to occur at approximately the same position (≤15 -1 23–25 cm ) as those observed for Ni- or Ir-octaethylporphyrin. The -1 Raman line at 677 cm is intensified for the octahedral structure of porphyrin but not for a square pyramidal structure owing to the 24,26 breathing-like motion of the 16-membered macrocycle. The SERS intensity from the gap plasmon resonator at the strongest -1 peak of 677 cm is 25 times greater than that of the SiO2/Ag substrate, whereas no Raman scattering signal was observed from the bare substrate. The gap plasmon resonator with a 40-nm-thick spacer shows the highest SERS intensity compared to the other ones, which is consistent with the efficiency of the TTA process (Figures S4 and S5 in the Supporting Information).

Experimental Section

Nanoscale Accepted Manuscript

Published on 03 July 2015. Downloaded by Freie Universitaet Berlin on 07/07/2015 04:51:51.

COMMUNICATION

To fabricate this structure, Ti (5 nm)/Ag (100 nm) layers were first deposited on a Si substrate using a thermal evaporator. Then, a bottom layer of SiO2 with a controlled thickness of 20, 40, 60, 80, or 100 nm was deposited on the Ti/Ag layers using plasma-enhanced chemical vapor deposition. To generate the AgNA, polystyrene (PS) spheres 420 nm in diameter were dispersed in a single layer onto the SiO2 spacer; then 50 nm of Ag was deposited on the PS sphere array by thermal evaporation. Finally, the PS spheres covered with

4 | J. Name., 2012, 00, 1-3

This journal is © The Royal Society of Chemistry 20xx

Please do not adjust margins

Please doNanoscale not adjust margins

Published on 03 July 2015. Downloaded by Freie Universitaet Berlin on 07/07/2015 04:51:51.

Journal Name

COMMUNICATION

Ag were removed, yielding gap plasmon resonators (AgNA/SiO2/Ag). For comparison, a separate bare substrate and a SiO2/Ag substrate 27,28 were also prepared. The fraction of PtOEP in the DPA:PtOEP dissolved in dichloromethane solvent was typically 2 wt.%, at which PtOEP 29 aggregations rarely influence the TTA efficiency. Thin films of DPA:PtOEP composite were prepared by spin-coating onto quartz, SiO2/Ag, and AgNA/SiO2/Ag substrates in a glove box to avoid oxygen contamination. The thickness of the DPA:PtOEP films for all the samples was carefully controlled to be 250±50 nm, and all the PL intensities measured were average of 5 separate samples. The emission spectra were measured at room temperature under -3 vacuum condition (

Enhanced triplet-triplet annihilation in bicomponent organic systems by using a gap plasmon resonator.

The triplet-triplet annihilation (TTA) efficiency in bicomponent organic systems is investigated by employing a gap plasmon resonator. In our structur...
1MB Sizes 0 Downloads 11 Views