Open Access
Add to BibSonomyAbstract
Cyanine dye J-aggregate films are a class of absorbing and luminescent materials which have been extensively applied in the polariton-based research. Here we systematically study the DEDOC cyanine dyes J-aggregate films made by layer-by-layer assembly and spin-coating processes to establish a clear correlation between the film structure and the absorption and luminescence properties. From detailed analyses of morphology, optical spectra, and light-emitting diode characteristics, we demonstrate that layer-by-layer assembled film has higher degrees of homogeneity and molecular packing quality than spin-coated film, leading to a higher absorption coefficient, more uniform luminescence, and a greater electroluminescence quantum efficiency with maximized thickness.
© 2014 Optical Society of America
1. Introduction
Cyanine dye J-aggregates are an attractive class of materials for the research of organic exciton-polariton due to their unique characteristics of large oscillator strengths and narrow exciton linewidths [1,2]. With simple solution processes the dye molecules can be made into different forms of J-aggregate films for use in various strongly-coupled devices. For example, in numerous studies regarding the optical characterization of polariton states, a process commonly implemented is to fill the whole cavity with a thick J-aggregate film by spin-coating a blend solution of dyes and insulating polymers [3–8]. This process allows the J-aggregates-forming dyes to be well dispersed in the thick polymer matrix and leads to an intense absorption and stable luminescence, but is unsuitable for electrically-driven applications due to the very low conductivities of polymer matrix. Realization of optoelectronic polariton devices is more technically challenging since the J-aggregate film needs to have sufficient conductivity and acts as an active medium to generate emission of polariton states via electrical excitation. This requires the appropriate processes to develop a high-density packing of J-aggregates in a relatively thin film while avoiding degradation of other device components on substrates. In the studies of electroluminescence of organic polaritons so far, layer-by-layer (LBL) assembly and spin-coating of dyes in aqueous solutions are the two mainly used processes for fabrication of J-aggregate thin films. LBL assembly is a gentle deposition method involving a sequential adsorption of cationic polyelectrolyte and anionic dyes. In 2005, Bulovic et al. demonstrated a high absorption coefficient of LBL J-aggregate films in nanometer-scale thickness [9]. The achievement of ultra-thin, highly absorptive LBL J-aggregate film not only leads to their success on the first demonstration of the electroluminescence of organic polaritons [10], but also provides flexibility of material combinations and cavity architectures in the design of novel polariton devices [11,12]. On the other hand, spin-coating has also been used as a convenient, time saving method to deposit J-aggregate thin films. Recently, Lidzey et al. investigated the electroluminescence of a cavity OLED with a spun J-aggregate film [13]. They pointed out that the radiative decay of exciton and polariton states is limited by charge transport and recombination, indicating the urgent demand to improve the charge mobility and luminescence quantum efficiency of J-aggregate films for practical device applications.
To date, a deep understanding of the structural factors that affect the optical and electrical properties of J-aggregate films is still lacking. In this paper, we present a systematic study of the correlation of the structure, absorption, photo- and electroluminescence properties of cyanine dye J-aggregate films. A cyanine dye molecule [5-chloro-2-(2-[(5-chloro-3-(3-sulfopropyl)-2(3H)-benzoxazolylidene) methyl]-1-butenyl)-3-(3-sulfopropyl)-benzoxazo-lium inner salt, sodium salt] (DEDOC, purchased from H. W. SANDS CORP) was chosen as the model material [14]. LBL assembly and spin-coating processes were used to prepare the J-aggregate films with different structures. In LBL assembly process, the polyelectrolyte poly(diallyldimethylammonium chloride) (PDAC, Mw = 400-500 kD, purchased from Sigma-Aldrich) solution was prepared in deionized water with different concentrations (1.7, 2.5, 5 mg/mL) to modify the thickness of adsorbed PDAC layer. The DEDOC cyanine dye solution was prepared by dissolving the dyes in alkaline water (pH = 10, adjusted by adding NaOH) at a fixed concentration of 4 × 10−3 mg/mL. The substrate was first treated with UV Ozone for 15 min, then sequentially immersed in PDAC and DEDOC aqueous solutions for cationic and anion adsorptions, and finally finished with PDAC adsorption. This will produce an assembly of adsorbed PDAC and DEDOC J-aggregate layers in a sequence of PDAC-(DEDOC-PDAC)N (denoted as “P(DP)N”, N is the number of adsorption cycles). The thickness and absorption intensity of LBL J-aggregate films can be precisely controlled by varying N, but the upper limit for the high-quality layered growth is around N = 4~5 [14]. Therefore, in this work we fabricated the LBL J-aggregate films with N = 4 as a maximum. In spin-coating process, the DEDOC solutions were prepared at various concentrations in the range of 1-10 mg/mL in alkaline water (pH = 12). Since spin-coating of aqueous solutions is highly sensitive to the wettability of substrates, we made the normal (deposited on the bare substrates) and the modified spin-coated films (deposited on the substrates pretreated with a PDAC adsorbed layer and then capped with another PDAC adsorbed layer, structurally analogous to the LBL film with N = 1) for comparison. With the different process conditions to control the film structures, our paper will provide a clear insight of how the film morphology and molecular packing quality determine the absorption and luminescence properties of J-aggregate films.
2. Process dependent morphology of J-aggregate films
Figure 1 shows the tapping-mode atomic force microscopy (AFM) images of the LBL-assembled and spin-coated DEDOC J-aggregate films on glass substrates. To obtain a high-quality J-aggregate film, it is crucial to prepare the substrates with high wettability. The contact angle of water measured on bare glass after ozone treatment is ~7° but decreases down to <5° after deposition of a PDAC layer. Hence, pretreatment of a cationic PDAC layer on glass not only enhances the adsorption of anionic dyes, but also improves the surface wettability for better formation of J-aggregate films. In LBL assembly process, the slow adsorption of DEDOC monomers on a PDAC layer favors the growth of J-aggregates into a rather uniform film with a dense distribution of nodulelike structures [Fig. 1(a)]. The RMS roughness of the first DEDOC adsorbed layer is 3-4 nm. For fast spin-coating process, the morphology of J-aggregate films is highly sensitive to the substrate wettability and the DEDOC solution concentration. The bare glass is not sufficiently wettable to form J-aggregate films when using low concentration solutions (<5 mg/mL). The spin-coated film with high concentration (10 mg/mL) on bare glass consists of randomly stacked flakelike structures and a high population of discontinuous defects, with the RMS roughness of ~5 nm [Fig. 1(b)]. On the other hand, spin-coating a low concentration solution (1 mg/mL) on highly wettable PDAC-treated glass could render a uniform J-aggregate film, which shows the smaller and more densely packed nodulelike structures as compared to the LBL film [Fig. 1(c)]. When increasing concentration to 2.5 mg/mL the nodulelike structures start to aggregate and the number of structural defects slightly increases. For even higher concentrations (5-10 mg/mL) the fibrillarlike structures appear and become entangled into large-sized clusters. The variation of morphology from 1 mg/mL to 10 mg/mL corresponds to the increased RMS roughness from 2 nm to 3 nm.
Fig. 1 AFM images (5 μm × 5 μm) of DEDOC J-aggregate films on glass substrates, made by (a) LBL assembly (PDAC/DEDOC), and spin-coating on (b) bare glass with 10 mg/mL DEDOC solution concentration and (c) PDAC-treated glass with various concentrations.
3. Absorption and photoluminescence of J-aggregate films
To exploit the capability of high absorption of J-aggregate films for strongly-coupled applications, it is important to maximize both the film thickness and peak absorption coefficient (α). In LBL assembly process, the film thickness is mainly controlled by the number of adsorption cycles N, but is also affected by other parameters such as the concentrations of PDAC and DEDOC solutions. Here we investigate the effect of tuning the PDAC concentration on the absorption of LBL J-aggregate films. As can be seen in Fig. 2(a), increasing the PDAC concentration from 1.7 mg/mL to 5 mg/mL in the LBL films with N = 4 leads to a notably enhanced peak absorption at 546 nm from 40% to 50%. The absorption spectra were extracted from the reflectance (R) and transmittance (T) spectra by 1-R-T. Figure 2(c) further summarizes the peak absorption of LBL films versus N for different PDAC concentrations. To quantitatively evaluate the thickness of each PDAC and DEDOC adsorbed layer varied with the PDAC concentration, we perform an iterative algorithm for optimal fitting of the R and T spectra based on the Kramers-Kronig transformation and transfer-matrix method [15], in which the thicknesses per PDAC and DEDOC layer are the fitting parameters and the (n,k) values of DEDOC can be modified with the K-K transformation. We also make two assumptions: (1) the PDAC concentration only affects the thicknesses but not the (n,k) values of PDAC and DEDOC layer, and (2) the thicknesses and the (n,k) values of each DEDOC and PDAC layer are invariant when progressing more adsorption cycles. Eventually, a reasonably good fit for all the LBL films can be obtained with the universal (n,k) spectra of DEDOC as shown in Fig. 2(b). The peak k value at λ = 546 nm corresponds to a high peak absorption coefficient α = 4πk/λ~1.03 × 106 cm−1. The fitting thicknesses of each PDAC and DEDOC layer reveal a subtle but consistent variation with PDAC concentration (see the inset table of Fig. 2(c)). Indeed, immersion in a higher PDAC concentration solution can produce a slightly thicker PDAC layer with more polyelectrolytes, and thereby adsorb a thicker DEDOC layer for enhanced absorption. Due to the high peak absorption coefficient of LBL film, even a sub-nanometer variation in each DEDOC layer would make a remarkable difference in thepeak absorption intensity. It is thus important to optimize the PDAC concentration to maximize the DEDOC thickness and absorption. Overall, the P(DP)4 LBL film produced with 5mg/mL PDAC concentration exhibits the highest peak absorption up to 50% and corresponds to the maximum total DEDOC thickness of 11.2 nm as determined by the product of the adsorption cycle (N = 4) and the thickness of each DEDOC layer (2.8 nm).
Fig. 2 (a) Absorption spectra of P(DP)4 LBL films on glass with different PDAC solution concentrations. The absorption spectra were extracted from the reflectance (R) and transmittance (T) spectra by 1-R-T. (b) (n,k) spectra of DEDOC J-aggregates in LBL films. (c) Peak absorption of LBL films versus N for different PDAC concentrations. The solid lines indicate the optimal fitting of peak absorption in comparison with the experimental data (solid circles). The inset tables summaries the fitting thicknesses of each PDAC and DEDOC layer. (d)-(f) show the similar characterizations of spin-coated DEDOC J-aggregate films on PDAC-treated and bare glass. From the peak k value in (e) we extract α~8.1 × 105 cm−1 for spin-coated films on PDAC-treated glass. The fitting thicknesses of spin-coated films (solid lines) are well matched with the AFM measurements (solid circles).
Similarly, we characterize the absorption spectra, optical constants, and physical thicknesses of spin-coated J-aggregate films on bare and PDAC-treated glass for different DEDOC concentrations, as the results shown in Figs. 2(d)-2(f). The absorption of spin-coated films is found to depend on not only the solution concentration but also the substrate wettability. For the same solution concentration, the spin-coated films on more wettable PDAC-treated glass normally exhibit the peak absorption around 10% higher than those on bare glass [Fig. 2(f)]. By fitting the optical spectra we extract a larger film thickness and a higher peak absorption coefficient α~8.0 × 105 cm−1 at 546 nm for the spin-coated films on PDAC-treated glass (α~6.9 × 105 cm−1 for the films on bare glass). The fitting thicknesses are well matched with the AFM measurement [Fig. 2(f)]. This analysis suggests that PDAC pretreatment on substrates can improve the adsorption of J-aggregates during spin-coating process, and facilitate the formation of thicker and denser films with significantly enhanced peak absorption. Nevertheless, the peak absorption coefficient of spin-coated films on PDAC-treated glass is still lower than that of LBL films. The highest peak absorption of the spin-coated films can reach 52% with 10 mg/mL DEDOC solution on PDAC-treated glass, which is similar to the best of the P(DP)4 LBL films but corresponds to an approximately two-fold thickness (~23 nm).
The different peak absorption coefficients of LBL and spin-coated J-aggregate films can be closely correlated with their different morphology and molecular packing quality. To further resolve the structures of J-aggregate films made by different processes, we perform the measurements of photoluminescence (PL) images and the corresponding spectra. Figures 3(a)-3(c) show the micrometer-scale PL images performed by confocal microscope with piezo-driven stage scanning configuration and spectrometer. The excitation Argon ion laser sources (Melles Griot) with a wavelength of 488 nm was reflected by a dichroic beamsplitter and then focused on the specimens by a 100X objective (Nikon) with a numerical aperture of 0.9. The PL signal was collected with the same objective and passed through a dichroic beamsplitter and a 488 nm long pass edge filter. It was coupled into an optical fiber connected with a spectrometer to map the PL images and spectra. The effective pixel size was 50 nm and the integration time of each pixel was 36.15 ms. Figures 3(d)-3(f) show the PL images recorded by the same system in a single measurement with a modification of 20X objective (Nikon). Overall, the LBL film exhibits the most uniform PL image [Figs. 3(a) and 3(d)]. The spectra corresponding to the regions with maximum and minimum PL intensity (marked by red and blue circles in Fig. 3(a)) reveal the similar lineshape (nearly identical after normalization) and the smallest contrast in intensity [Fig. 3(g)]. This clearly demonstrates the homogeneous microstructures and a uniform distribution of J-aggregates in the LBL film. On the other hand, both the spin-coated films with high DEDOC concentration (10 mg/mL) on bare and PDAC-treated glass (see the corresponding AFM images in Figs. 1(b) and 1(c)) apparently have boundary defects or discontinuities with little luminescence between clusters or stacked flakelike structures, such that the PL uniformity is poorer than the LBL film [Figs. 3(b) and 3(c)]. In particular, the spin-coated film on bare glass exhibits the worst PL uniformity [Fig. 3(e)]. The spectral difference between the regions with maximum and minimum PL intensity is extremely large, and it appears that very few J-aggregates are located in the minimum PL region [Fig. 3(h)]. By comparison, the spin-coated film on PDAC-treated glass shows more intense PL [Fig. 3(f)] and also an improved PL uniformity as evidenced by a reduced spectral contrast between the regions with maximum and minimum PL intensity [Fig. 3(i)]. This result again verifies that the PDAC pretreatment facilitates the formation of thicker J-aggregate films with more homogeneous microstructures and a better coverage on substrates.
Fig. 3 (a)-(c) PL images of DEDOC J-aggregate films on glass substrates performed by confocal microscope with scanning configuration. (a) P(DP)4 LBL film, (b) spin-coated film on bare glass, and (c) spin-coated film on PDAC-treated glass. Both the spin-coated films were prepared with 10 mg/mL DEDOC concentration. (d)-(f) show the large-scale PL images corresponding to (a)-(c) recorded from the same system in a single measurement. (g)-(i) PL spectra corresponding to the maximum (red) and minimum (blue) PL intensity in the images of (a)-(c).
The two main vibronic features of the PL spectra, 0-0 emission at 546-549 nm and 0-1 emission at ~580 nm, can be further analyzed to gain insight of molecular packing quality in different films. It can be found that the intensity ratio of the 0-0 to 0-1 emission is largest in the LBL film, intermediate in the spin-coated film on PDAC-treated glass, and lowest in the spin-coated film on bare glass. According to the theoretical analysis by Spano and associates [16,17], for linear J-aggregates in the strong exciton coupling regime (as the case considered here when the main absorption is markedly red-shifted from monomers) the 0-0 emission from the bottom of the excited band to the vibrationless ground state is linearly enhanced by the coherence number Ncoh (the number of molecules over which the exciton wave function is delocalized), whereas the 0-1 emission arising from a constructive interference between one-particle (vibronic) and two-particle (vibronic/vibrational pair) excitations of the emitting exciton is weakly coherent with Ncoh. A simple expression can be derived for the intensity ratio of the 0-0 to 0-1 emission
where the λ2 is the Huang-Rhys (HR) factor. By substituting the measured PL ratio and assuming λ2 = 1 [18,19], we estimated Ncoh≈7 for the LBL film, Ncoh≈3-4 for the spin-coated film on PDAC-treated glass, and Ncoh≈2-3 for the spin-coated film on bare glass. The highest coherence number Ncoh in the LBL film can be interpreted as an evidence of the lowest degree of disorder that leads to the largest spatial extent of delocalized excitons. And possibly due to the lowest degree of disorder the excitons can migrate between different delocalization domains and relax towards the lowest lying excited states [20], as evidenced by the longest wavelength of 0-0 emission (548-549 nm) observed in the LBL film [Fig. 3(g)]. On the other hand, the smaller coherence numbers Ncoh in the spin-coated films suggest the higher degrees of microstructural disorder. The more localized excitation also results in the higher energy of 0-0 emission and smaller Stokes shift [Figs. 3(h) and 3(i)].From detailed analyses of PL measurements, we confirm that LBL assembly enables the formation of homogeneous film structure and high-quality molecular packing of J-aggregates. This is consistently correlated with the highest peak absorption coefficient extracted for LBL film. By contrast, spin-coating process generally yields the less homogeneous film structure and poorer molecular packing quality, which may account for the lower peak absorption coefficients of spin-coated films.
4. Electroluminescence of J-aggregate films
To further investigate the electroluminescence (EL) properties of DEDOC J-aggregate films, we fabricated the OLEDs with J-aggregate films as the luminescence layer sandwiched between a 60 nm thick poly-TPD [poly(N,N’-bis(4-butylphenyl-N,N’-bis(phenyl)benzidine)] as the hole transporting layer and a 60 nm BCP (bathocuproine) as the electron transportinglayer, a 6 nm thick Cs2CO3 as the electron injection layer, and a 80 nm thick Ag as the top cathode electrode [Fig. 4(a)]. The optoelectronic characteristics of the OLEDs was measured using an Agilent B1500A semiconductor parameter analyzer with a silicon photodiode (Hamamatsu S1133-01) which detected the photocurrent for light outcoupling through the glass substrates. The EL spectrum was recorded with an Ocean Optics HR4000 spectrometer using an optical multi-mode fiber. The external quantum efficiencies (EQE) were extracted from the measured photocurrent, photodiode’s sensitivity (0.3 A W−1 at 550 nm), and normalized EL spectrum [21].
Fig. 4 (a) Device structure of an OLED with DEDOC J-aggregate films as the luminescence layer. (b) EL spectrum and image of an OLED with P(DP)4 LBL film. The peak wavelength of EL spectrum is 551 nm. The device area is 2 mm × 5 mm.
Figure 5(a) shows the optoelectronic characteristics of the OLEDs with LBL J-aggregate films. For a series of LBL films made by 5 mg/mL PDAC solutions, the diode with only one adsorption cycle (N = 1) exhibits the highest current but the lowest EL intensity, and hence the lowest EQE. When more adsorption cycles are progressed, the current decreases while the EL intensity enhances, and the EQE increases accordingly. This can be understood since the electrons and holes passing through more assembled J-aggregate layers would have higher possibilities to recombine and to emit photons in one of layers. However, as shown in the inset of Fig. 5(a), the EQE increases sharply from N = 1 to N = 3 and then starts to level off for even higher N, similar to the variation of peak absorption with N [Fig. 2(c)]. The maximum EQE of LBL films is obtained in the order of 5 × 10−3% for N = 4. We speculate that when approaching the limit of layered growth the structural homogeneity of the top J-aggregate layer decreases, and the electron-hole recombination occurs mostly in the first few J-aggregate layers. In Fig. 4(b) we show a typical EL spectrum and image of an OLED with P(DP)4 LBL film. The peak of EL spectrum at 551 nm is slightly red-shifted with respect to that of PL spectrum (~549 nm, see Fig. 3(g)), reflecting the different PL and EL mechanisms. The PL is dominated by local excitation and exciton migration in neighboring domains. However, the EL involves long-distance transportation and recombination of non-geminate electrons and holes, which mainly occurs in the most ordered J-aggregates domains with high conductivities.
Fig. 5 (a)-(c) Optoelectronic characteristics of the OLEDs with DEDOC J-aggregates, made by (a) LBL assembly for various adsorption cycles N, (b) spin-coating on PDAC-treated Poly-TPD for DEDOC concentrations of 1-10 mg/mL, and (c) spin-coating on Poly-TPD without PDAC pretreatment for DEDOC concentrations of 5-10 mg/mL. The evaluated EQEs versus driving voltage are shown in the insets. (d) Summary of the maximum EQE values extracted above 12V versus the thickness of J-aggregate films for different processes.
Figure 5(b) shows the optoelectronic characteristics of the OLEDs with spin-coated J-aggregate films on a PDAC-treated Poly-TPD. For a series of DEDOC concentrations, the diode with the lowest concentration (1 mg/mL) exhibits the lowest current but the highest EL intensity, and the maximum EQE up to 4 × 10−3% is comparable to that of P(DP)4 LBL film. Interestingly, such a thin film with only 5 nm DEDOC can produce even higher EL intensity than the P(DP)4 film with ~11 nm DEDOC. As the film thickness increases with higher concentrations, the current density increases while the EL intensity decreases, and the EQE drops accordingly (the inset of Fig. 5(b)). This trend is opposite to that observed in LBL films, and can be attributed to the variation of morphology from low to high concentration films asshown in Fig. 1(c) (the morphology of spin-coated films on PDAC-treated Poly-TPD was found very similar to that on PDAC-treated glass). A very dense and uniform distribution of small nodulelike structures in the low concentration film allows the electrons and holes to be captured effectively within a few nanometers thickness. Once the large-sized clusters are formed in the higher concentration films, the increasing boundary defects between clusters would result in more pathways for current leakage through the films without efficient recombination and emission, and therefore the EQE decreases.
We also characterize the OLEDs with spin-coated J-aggregate films on bare Poly-TPD for comparison [Fig. 5(c)]. In general, for the same DEDOC solution concentration, the EL intensities and EQE values are apparently lower than the diodes with PDAC pretreatment, which is as expected due to the presence of more pronounced defects in the J-aggregate films. The maximum EQE of ~1 × 10−3% is obtained in the 5 mg/mL concentration film. When increasing concentration to 10 mg/mL both the current density and EL intensity decrease, and the maximum EQE drops to 6 × 10−4%. It is possible that the higher concentration film contains more microscopic defects, resulting in higher degrees of charge trapping and exciton quenching, and hence the overall decreased current and EQE.
In Fig. 5(d) we summarize the maximum EQE value versus the thickness of J-aggregate films for all the processes. The highest EQE of DEDOC in the present study is 5-6 times lower than the value reported for a well-studied dye molecule, TDBC [22]. One reason could be due to the lower lying highest occupied molecular orbital (HOMO) of DEDOC (~5.5 eV estimated from ultraviolet photoemission spectroscopy measurement) that leads to a larger hole injection barrier from Poly-TPD to DEDOC (~0.4 eV). For further studies Poly-TPD could be replaced by other hole transporting polymers with lower HOMO levels. Nevertheless, our result clearly shows that increasing the thickness of J-aggregate films while keeping the control of film morphology with minimized structural defects is a key factor for optimization of EL intensity and EQE.
5. Conclusions
We present a correlative study of the structure, absorption, and luminescence properties of DEDOC J-aggregates films made by LBL assembly and spin-coating processes. These two processes lead to very different morphology and microstructure of J-aggregate films, and have a direct impact on the absorption, PL, and EL properties. We demonstrate that LBL assembly process enables to produce the J-aggregate films with delicately controlled thickness and high degrees of film homogeneity and molecular packing quality. LBL J-aggregate film shows a high peak absorption coefficient (~1×106 cm−1) and uniform luminescence. More importantly, the absorption and luminescence intensities, and also the EL quantum efficiency can be simultaneously maximized as progressing the adsorption cycles to the limit of layered growth. LBL J-aggregate film therefore could offer great advantages for use as both a strongly-coupled medium and a luminescent source in versatile optically and electrically pumped polariton devices. On the other hand, the structure of spin-coated film is a complex function of solution concentration and substrate wettability. Even with PDAC pretreatment to improve the substrate wettability and adsorption of dyes, spin-coated film generally has poorer film homogeneity and molecular packing quality than LBL film, and hence a lower absorption coefficient and less uniform luminescence. Increasing solution concentrations yields thicker films and enhanced absorption, but also induces more structural defects resulting in the decreased EL intensity and quantum efficiency. This is adverse to the optoelectronic applications and needs to be addressed further by taking into account more processing parameters, such as solvent polarity, pH value, and multistep spin-coating scheme, to better control both the film structure and thickness in the rapid deposition process.
Acknowledgments
The authors gratefully acknowledge the Ministry of Science and Technology of Taiwan for financial support under Contracts No. NSC100-2112-M-008-016-MY3, No. 102-2221-E-008-097 and MOST 103-2221-E-008 −003 -.
References and links
1. E. E. Jelly, “Molecular, nematic and crystal states of I: I-diethyl–cyanine chloride,” Nature 139(3519), 631–632 (1937). [CrossRef]
2. G. Scheibe, “Variability of the absorption spectra of some sensitizing dyes and its cause,” Angew. Chem. 49, 563 (1936).
3. D. G. Lidzey, D. D. C. Bradley, M. S. Skolnick, T. Virgili, S. Walker, and D. M. Whittaker, “Strong exciton-photon coupling in an organic semiconductor microcavity,” Nature 395(6697), 53–55 (1998). [CrossRef]
4. D. G. Lidzey, D. D. C. Bradley, T. Virgili, A. Armitage, M. S. Skolnick, and S. Walker, “Room temperature polariton emission from strongly coupled organic semiconductor microcavities,” Phys. Rev. Lett. 82(16), 3316–3319 (1999). [CrossRef]
5. D. M. Coles, N. Somaschi, P. Michetti, C. Clark, P. G. Lagoudakis, P. G. Savvidis, and D. G. Lidzey, “Polariton-mediated energy transfer between organic dyes in a strongly coupled optical microcavity,” Nat. Mater. 13(7), 712–719 (2014). [CrossRef] [PubMed]
6. N. Somaschi, L. Mouchliadis, D. Coles, I. E. Perakis, D. G. Lidzey, P. G. Lagoudakis, and P. G. Savvidis, “Ultrafast polariton population build-up mediated by molecular phonons in organic microcavities,” Appl. Phys. Lett. 99(14), 143303 (2011). [CrossRef]
7. S. Hayashi, Y. Ishigaki, and M. Fujii, “Plasmonic effects on strong exciton-photon coupling in metal-insulator-metal microcavities,” Phys. Rev. B 86(4), 045408 (2012). [CrossRef]
8. Y. Obara, K. Saitoh, M. Oda, and T. Tani, “Anomalous reflection properties in high density limit fibril shaped PIC-J aggregates in microcavity structure,” Phys. Status Solidi 8(2c), 595–597 (2011). [CrossRef]
9. M. S. Bradley, J. R. Tischler, and V. Bulović, “Layer-by-layer J-aggregate thin films with a peak absorption constant of 106 cm−1,” Adv. Mater. 17(15), 1881–1886 (2005). [CrossRef]
10. J. R. Tischler, M. S. Bradley, V. Bulović, J. H. Song, and A. Nurmikko, “Strong coupling in a microcavity LED,” Phys. Rev. Lett. 95(3), 036401 (2005). [CrossRef] [PubMed]
11. M. S. Bradley and V. Bulović, “Intracavity optical pumping of J-aggregate microcavity exciton polaritons,” Phys. Rev. B 82(3), 033305 (2010). [CrossRef]
12. G. M. Akselrod, E. R. Young, M. S. Bradley, and V. Bulović, “Lasing through a strongly-coupled mode by intra-cavity pumping,” Opt. Express 21(10), 12122–12128 (2013). [CrossRef] [PubMed]
13. N. Christogiannis, N. Somaschi, P. Michetti, D. M. Coles, P. G. Savvidis, P. G. Lagoudakis, and D. G. Lidzey, “Characterizing the electroluminescence emission from a strongly coupled organic semiconductor microcavity LED,” Adv. Optical Mater. 1(7), 503–509 (2013). [CrossRef]
14. H.-S. Wei, C.-C. Jaing, Y.-T. Chen, C.-C. Lin, C.-W. Cheng, C.-H. Chan, C.-C. Lee, and J.-F. Chang, “Adjustable exciton-photon coupling with giant Rabi-splitting using layer-by-layer J-aggregate thin films in all-metal mirror microcavities,” Opt. Express 21(18), 21365–21373 (2013). [CrossRef] [PubMed]
15. R. Nitsche and T. Fritz, “Determination of model-free Kramers-Kronig consistent optical constants of thin absorbing films from just one spectral measurement: application to organic semiconductors,” Phys. Rev. B 70(19), 195432 (2004). [CrossRef]
16. F. C. Spano and C. Silva, “H- and J-aggregate behavior in polymeric semiconductors,” Annu. Rev. Phys. Chem. 65(1), 477–500 (2014). [CrossRef] [PubMed]
17. F. C. Spano and H. Yamagata, “Vibronic coupling in J-aggregates and beyond: a direct means of determining the exciton coherence length from the photoluminescence spectrum,” J. Phys. Chem. B 115(18), 5133–5143 (2011). [CrossRef] [PubMed]
18. D. Oelkrug, H.-J. Egelhaaf, J. Gierschner, and A. Tompert, “Electronic deactivation in single chains, nano-aggregates and ultrathin films of conjugated oligomers,” Synth. Met. 76(1-3), 249–253 (1996). [CrossRef]
19. J. Gierschner, H.-J. Egelhaaf, and D. Oelkrug, “Absorption, fluorescence and light sscattering of oligothiophene and oligophenylenevinylene nanoaggregates,” Synth. Met. 84(1-3), 529–530 (1997). [CrossRef]
20. I. G. Scheblykin, O. Yu. Sliusarenko, L. S. Lepnev, A. G. Vitukhnovsky, and M. Van der Auweraer, “Excitons in molecular aggregates of 3,3′-bis[3-sulfopropyl]-5,5′-dichloro-9-ethylthiacarbocyanine (THIATS): temperature dependent properties,” J. Phys. Chem. B 105(20), 4636–4646 (2001). [CrossRef]
21. J.-F. Chang, M. C. Gwinner, M. Caironi, T. Sakanoue, and H. Sirringhaus, “Conjugated-polymer-based lateral heterostructures defined by high-resolution photolithography,” Adv. Funct. Mater. 20(17), 2825–2832 (2010). [CrossRef]
22. J. R. Tischler, M. S. Bradley, Q. Zhang, T. Atay, A. Nurmikko, and V. Bulović, “Solid state cavity QED: strong coupling in organic thin films,” Org. Electron. 8(2-3), 94–113 (2007). [CrossRef]
