J. Phys. Chem. A 2010, 114, 13291–13295
13291
Fluorescence Enhancement by Symmetry Breaking in a Twisted Triphenylene Derivative Jack W. Levell,† Arvydas Ruseckas,† John B. Henry,‡ Yi Wang,‡ Andrew D. Stretton,‡ Andrew R. Mount,‡ Trent H. Galow,‡ and Ifor D. W. Samuel*,† Organic Semiconductor Centre, SUPA, School of Physics and Astronomy, UniVersity of St. Andrews, North Haugh, St. Andrews, Fife KY16 9SS, U.K., and School of Chemistry, EaStCHEM, The Joseph Black Building, King’s Buildings, The UniVersity of Edinburgh, Edinburgh EH9 3JJ, U.K. ReceiVed: July 16, 2010; ReVised Manuscript ReceiVed: October 1, 2010
1,4,5,8,9,12-Hexamethyltriphenylene (HMTP) shows a high photoluminescence quantum yield (PLQY) of 31% in the solid state, making it of interest for luminescence applications. The detailed photophysical properties of HMTP have been investigated by using time-resolved and steady-state luminescence, PLQY, and molar absorption coefficient measurements. An enhancement of the transition dipole moment for fluorescence and absorption was demonstrated compared to the case of unsubstituted triphenylene, which resulted in a 20-fold increase in the radiative decay rate. This is attributed to a breaking of triphenylene symmetry as a result of the necessarily twisted structure induced by steric crowding. In addition, it was shown that HMTP shows similar photoluminescence energies in solution, powder, and film, indicating a reduced propensity for intermolecular π-stacking compared to the case of triphenylene, as a result of this twisted structure. This work also develops a method for calculating the photoluminescence quantum yield of powders by using a calibrated photodiode in combination with an uncalibrated CCD spectrometer. 1. Introduction Emissive molecules often suffer from reductions in luminescence as well as red shifts in their emission spectrum due to concentration quenching or excimer formation in concentrated solutions,1,2 and these problems become worse in the solid state.2-4 This effect can be particularly pronounced in large planar conjugated systems. These intermolecular interactions are undesirable for solid-state light-emitting devices and organic electronic applications and so considerable effort has been expended to control them. Strategies to minimize interactions include blending materials into higher energy hosts,5 using bulky side groups in conjugated polymers4 or making light-emitting dendrimers6 to increase the spacing between choromophores, or using the conformation of the molecule to prevent π-stacking by twisting the molecule out of a single plane, for example, in spiro-fluorenes.7 Triphenylene (Figure 1A) is a weakly fluorescent material that has a low dipole moment for its S1 to S0 transition because it is forbidden by symmetry.8-10 The material also has a planar conformation that readily leads to π stacking, and so derivatives of this material have widely studied liquid crystalline phases that may be useful for singlet excitation transfer.8,11-14 In these cases triphenylene is normally functionalized at the 2, 3, 6, 7, 10, and 11 positions to introduce spacers between the triphenylene columns while leaving the planar structure mostly unaffected. X-ray crystallography studies have shown that the triphenylene molecule is not exactly planar but instead exhibits a slight tendency to twist out of plane with displacements of ∼0.1 Å.15 This can be increased by methyl substitutions at the 1, 4, 5, 8, 9, and 12 positions to give 1,4,5,8,9,12-hexamethyltriph* Corresponding author. E-mail:
[email protected]. † University of St. Andrews. ‡ The University of Edinburgh.
Figure 1. Molar extinction coefficient and chemical structure of (a) triphenylene and (b) HMTP in THF solution (0.01 g dm-3). The S1 transition in triphenylene has been magnified 100 times. Dotted lines are Gaussian functions fitted to the S0 f S1 features in the spectra. The structure in (b) shows the C2 conformation of HMTP.16
enylene (HMTP, Figure 1B), which forces the molecule to adopt a twisted conformation.16
10.1021/jp106622n 2010 American Chemical Society Published on Web 12/06/2010
13292
J. Phys. Chem. A, Vol. 114, No. 51, 2010
In this paper we report enhanced luminescence of HMTP due to the altered symmetry of the molecule. We also show that this twisted HMTP structure has significantly reduced interchromophore interactions compared to those of triphenylene. 2. Experimental Methods Solid triphenylene was purchased from Sigma-Aldrich. HMTP was synthesized following a method described elsewhere.16 Films of triphenylene and HMTP were prepared on cleaned quartz disks from a 20 mg/mL concentration THF solution by spin coating at 2000 rpm for 1 min. The thicknesses of typical films were measured using a Veeco DekTak 3 surface profilometer to be 150 ( 10 nm for both HMTP and triphenylene. Absorption spectra of films and solutions were recorded using a Varian Cary 300 UV-vis spectrophotometer. Powder absorption could not be measured because of the difficulty in accounting for the large amount of scattered light. Emission spectra were recorded on a Jobin Yvon Fluoromax 2 fluorometer with both excitation and emission monochromators set to 1 nm bandpass. The effect of concentration on the emission spectrum of triphenylene was investigated in the range 0.04-0.2 g dm-3 (∼(1.3-8.5) × 10-4 M). The measurements were made with the fluorometer in a front face geometry, using a mirror to collect the light emitted from the excited face of the cuvette to minimize self-absorption and the inner filter effect. In fact, at the highest concentration, absorption through the 1 cm path length of the entire cuvette is calculated to result in less than 10% absorption at the energy (28 570 cm-1) corresponding to the blue shoulder of triphenylene emission. Time-resolved luminescence was measured using the timecorrelated single-photon counting (TCSPC) technique. For these measurements the sample was excited by a Nd:YAG microchip laser operating at 15 kHz, using either 266 nm (twice frequency doubled) or 355 nm (frequency tripled) light. The emitted light was collected through a monochromator and detected with a cooled Hamamatsu RU-3809 U-50 microchannel plate photomultiplier tube. With the exception of triphenylene solution measurements scattered excitation light was removed using a UV long pass filter. The count rate of detected photons was less than 5% of the repetition rate of the excitation laser. The instrument response function full width at half-maximum was measured to be ∼0.5 ns. Film samples were measured in a vacuum while powdered samples were measured in air. Solutions were prepared in THF and then freeze-thaw degassed inside quartz degassing cuvettes. They were optically dilute, with an absorbance of 0.1 at 285 nm (37 590 cm-1) and of less than 0.01 at 355 nm (28 170 cm-1). This corresponds to a concentration of 6.0 × 10-6 M for triphenylene and 2.1 × 10-6 M for HMTP. Fluorescence decay curves were fitted using iterative convolution to minimize the χ2 parameter, assuming Poissonian errors. Solution photoluminescence quantum yield (PLQY) measurements were made by comparison of the integrated emission spectra of samples with fluorescence standards and compensating for the effects of refraction and differences in absorption17 For 285 nm excitation 2-aminopryidene in 0.05 M sulfuric acid was used as a standard and taken to have a PLQY of 60%.18 For 360 nm excitation, quinine sulfate in 0.5 M sulfuric acid was used as a standard and assumed to have a PLQY of 55%.1 The samples and standards were made up to an absorbance of 0.1 at the excitation wavelength, freeze-thaw degassed, and their emission spectra measured by a Jobin Yvon Fluoromax 2 fluorometer.
Levell et al. Film PLQY was measured by exciting a sample in an integrating sphere using a He:Cd laser operating at 325 nm. The resulting fluorescence was detected using a calibrated photodiode behind a UV filter, which cuts out the excitation light. The fraction of light absorbed by the film and corrections for the reflected and transmitted excitation light were made by measuring the reflected and transmitted beams and applying the method of Greenham et al.,19 which consists of using an integrating sphere and a calibrated photodiode. Absolute PLQY mesurements of powders present an additional challenge of determining the amount of light absorbed by the sample as reflection and scattering are diffuse rather than specular. We therefore adapted the method of Greenham et al.,19 drawing on the work of the work of de Mello et al.20 A quartz cuvette containing the powder was suspended inside the integrating sphere and excited by a He:Cd laser at a wavelength of 325 nm. The intensity of the laser light in the sphere was measured using an Andor Model DV420-BV CCD spectrometer to determine the fraction of laser light absorbed. As the CCD was only used for comparing laser light intensities at a single wavelength, a calibration of its spectral response was not required. The full de Mello method was not used as calibration lamps traceable to a standards laboratory were not available to calibrate the spectral response of the CCD spectrometer in the UV. We have therefore developed a method for measuring PLQY by using a calibrated photodiode in combination with an uncalibrated CCD spectrometer. As noted by de Mello, the fraction of scattered laser light cannot simply be determined from the ratio of the intensity of laser light detected when the sample is in the beam of the laser (Csample) to the excitation light intensity in the sphere with no sample present (Claser).20 This is because this neglects the fact the spectral response of the sphere at the excitation wavelength is significantly altered by the presence of the absorbing sample. To avoid this problem, our method involves measuring the amount of laser light detected with the sample in the sphere but out of the beam (Csphere). The ratio Csphere/Claser then gives how much of the initially scattered light would not be reabsorbed by the sample. Increasing the ratio of Csample/Claser to account for the altered responsivity of the sphere with the sample present thus allows us to determine the true fraction of excitation light that was scattered initially. This involves dividing Csample/Claser by Csphere/Claser, thus the fraction of excitation light initially scattered by the powder sample (S) can be calculated using eq 1.
S)
Csample Csphere
(1)
This initially scattered light produces secondary fluorescence from the sample when it is absorbed after subsequent reflections within the sphere. The Greenham method19 uses a measurement of the secondary fluorescence produced when the sample is out of the beam to subtract contributions from transmitted and reflected excitation light. By simply using S in place of the terms for the fraction of reflected and transmitted light (R+T), the Greenham formula can be used to calculate the powder PLQY. Absorption and emission transition dipole moments were calculated using the method of Strickler and Berg.21 Absorption dipole moments |da|, in units of Debyes (D), were determined by fitting the molar extinction coefficient ε(ν˜ ), in units of M-1 cm-1, in wavenumber (ν˜ ) space. This was done for the S0 f S1 transitions, assuming Gaussian vibronic peak shapes. The
Fluorescence Enhancement in a Twisted Triphenylene Derivative
J. Phys. Chem. A, Vol. 114, No. 51, 2010 13293
absorption dipole moment was then calculated using eq 2 below, where n is the refractive index of the solvent.
|da | 2 ) 9.186 × 10-3n
∫ ε(νν˜˜ ) dν˜
(2)
The emission dipole moment |de|, again in D, was calculated from eq 3 with the radiative lifetime τR, in seconds, the emission spectral intensity I, in photons per unit energy, the energy E, in joules, the permittivity of free space ε0, in F m-1, the reduced Planck constant p, in J s, and the speed of light c, in ms-1. Both HMTP film and powder were taken to have a refractive index of 1.8, which is a typical value for organic semiconductors.22-24
|de | ) 2
3πε0p4c3(
∫ E-3I(E) dE/ ∫ I(E) dE)
3.33564 × 10-30nτR
(3)
The radiative lifetime τR was calculated using eq 4 where τ is the observed lifetime and Φ is the measured PLQY.19,25
τR ) τ/Φ
(4)
Although the derivation of eq 3 depends on the assumption that the transition is strongly allowed, Strickler and Berg21 have noted that the formula may still be reasonably precise in the case of weaker transitions. In the case of triphenylene we have calculated the dipole moments to demonstrate the magnitude of the marked enhancement found for HMTP and so a high degree of precision is not required.
Figure 2. (a) PL spectra of triphenylene in THF at 0.04 g dm-3 (dashed line) and 0.2 g dm-3 (solid line) and in spin-cast film (thick red solid line). (b) PL spectra of HMTP in THF at 0.04 g dm-3 (dashed line) and 5 g dm-3 (solid line) and in spin-cast film (thick red solid line). Spectra are normalized at the peak.
3. Results The dilute absorption spectra and chemical structures of triphenylene and HMTP are shown in Figure 1. The absorption of the triphenylene is highly structured, showing numerous vibronic peaks. The relatively weak peaks around 30 500 cm-1 are attributed to the symmetry forbidden S1 transition. This attribution is based on previous quantum chemical calculatations on triphenylene9,10 and calculations on hexaalkoxy-substituted triphenylenes.8 HMTP, in contrast, shows smoother broadened absorption features with little evidence of vibrational structure. The spectrum is well fit by three broad peaks at 31 500, 35 300, and 38 000 cm-1. The lowest energy peak is attributed to the S0 f S1 transition. From the spectral intensity it can be seen that the S1 absorption is considerably stronger in HMTP than triphenylene. The photoluminescence spectra of triphenylene and HMTP are shown in Figure 2. The dilute (0.04 g dm-3) triphenylene solution has an emission peak at 27 100 cm-1, whereas the HMTP peak is at lower energy at 23 040 cm-1. As with the absorption spectrum, the triphenylene spectrum is structured, with many identifiable peaks, whereas the HMTP emission is relatively featureless. Triphenylene also shows a noticeable red shift of 1500 cm-1 in its emission peak on increasing the concentration from 0.04 to 0.2 g dm-3, and the peak is further shifted to 24 750 cm-1 in films. In contrast, this concentration dependence is not present in HMTP solutions even up to 5.0 g dm-3 and no shift is seen in HMTP films. The time-resolved luminescence of HMTP and triphenylene in solution and film, along with HMTP powder, are shown in
Figure 3. Fitted time-resolved luminescence decay data excited at 266 nm for dilute triphenylene solution in THF (at 390 nm, open red circles, black fit line) and triphenylene film (at 390 nm, filled red squares, black fit line), dilute HMTP solution in THF (at 425 nm, open black squares, red fit line), and HMTP film (at 425 nm, black horizontal crosses, red fit line). Also shown is HMTP powder data, excited at 355 nm (at 425 nm, gray diagonal crosses, black fit line).
Figure 3. The fitting parameters are shown in Table 1 along with PLQY measurements and the resulting calculated emission dipole moment. 4. Discussion The concentration dependence of the triphenylene emission spectra in Figure 2a is attributed to the formation of lower energy aggregate states due to molecular π-stacking. X-ray diffraction has shown that HMTP does not π-stack in the solid state16 due to its twisted structure; HMTP photoluminescence (Figure 2b)
13294
J. Phys. Chem. A, Vol. 114, No. 51, 2010
Levell et al.
TABLE 1: Emission Dipole Moments for Triphenylene and HMTPa photoluminescence quantum yield
sample
lifetime/ns
triphenylene: THF solution
6% (285 nm)
triphenylene: film
13% (325 nm)
triphenylene: powder HMTP: THF solution HMTP: film HMTP: powder
12% (325 nm) 5% (285 nm), 5% (360 nm) 31% (325 nm) 31% (325 nm)
radiative rate/µs-1
emission transition dipole moment/D
10%, 2.3; 35%, 12.7; 55%, 38.6 average ) 25.8 (266 nm) 85%, 2.1; 11%, 11.1; 4%, 47.0 average ) 4.9 (266 nm)
1.7b
0.5b
(27b,c)
(1.9b,c)
5.4 (266 nm) 6.7 (266 nm) 6.2 (355 nm)
9.2 46 50
1.3 2.1 2.1
a
Photoluminescence quantum yield measurements and emission dipole calculations for triphenylene and HMTP in film, powder and solution. The excitation wavelengths used in each measurement are shown in brackets. b Calculated on the basis of average lifetime. c Triphenylene film emission is dominated by aggregates and so cannot be used to calculate the molecular dipole moment.
TABLE 2: Absorption Dipole Moments for Triphenylene and HMTPa transition
absorption peak average peak absorption transition position/cm-1 spacing/cm-1 dipole/D
triphenylene S0 f S1
29 090 29 820 30 490 31 160 31 770
HMTP S0 f S1
31 450
670
0.7
3.6
a S0 f S1 transition dipole moments calculated from the fitted absorption spectra of triphenylene and HMTP in THF solution (Figure 1).
does not change with concentration, showing that π-stacking is also prevented in solution. A simple model of conjugated molecules implies that a twisted structure would reduce the conjugation length and thus increase the energy of the excited states in the HMTP molecule. However, in fact, its emission spectrum is red-shifted with respect to triphenylene. This red-shifting is, however, consistent with the observation that oligo(phenylenevinylene) molecules undergo a red-shifting and broadening of their spectra when they are in a bent conformation.26 As there is only a small difference in the positions of the S1 absorption peak in both materials, it could be that the increased separation of absorption and emission in HMTP is the result of additional molecular reorganization in the excited state. The relative flexibility of the molecule is illustrated by conformational transitions observed in variable temperature NMR data and supported by quantum chemical calculations detailed in previous work.16 These transitions occur when the methyl substituents manage to exchange conformational positions by pushing past one another. Absorption dipole moments were calculated for the S0 f S1 transitions shown in Figure 1 and the results are given in Table 2. These calculations confirm that the triphenylene dipole moment of 0.7 D is significantly lower than the HMTP dipole moment of 3.6 D. The HMTP emission lifetime curves were all fitted well by single exponential decays; however, triphenylene curves required a multiexponential fit. This indicates that even the dilute triphenylene solution contained multiple emissive environments, most likely resulting from aggregates. HMTP’s single exponential decays are therefore further evidence for the suppression of intermolecular interactions. As the emission dipole moment can only be calculated for a single lifetime component, the dipole moment of triphenylene was estimated on the basis of an average lifetime determined
from the lifetime components weighted by their amplitudes. The best estimate of the isolated triphenylene molecule’s dipole moment comes from the dilute solution measurement as the film emission is dominated by aggregates. Triphenylene has been found by other authors to have a fluorescence lifetime of 37-39 ns in solution27 and 41 ns gas phase10 and so the long 38.6 ns lifetime component of the solution is most likely to correspond to the monomer. HMTP and triphenylene solutions have similar photoluminescence quantum yields (PLQYs) of 5 and 6%, respectively, in solution. However, Figure 3 shows HMTP to give much faster fluorescence decay, indicating that HMTP has a higher radiative rate. The average lifetime of triphenylene in solution is 25.8 ns, which is significantly longer than the HMTP single exponential lifetime of 5.4 ns. The triphenylene film showed a markedly decreased photoluminescence lifetime, compared to that for the triphenylene solution, with a much greater proportion of fast decaying components and has an increased photoluminescence quantum yield of 13% compared to 6%. This suggests the faster decaying components of the triphenylene emission relate to aggregated species that allow the triphenylene to overcome its unfavorable symmetry for light emission via aggregate-induced emission.2 This is supported by a red-shifting of the film emission spectrum compared to that for the solution, suggesting aggregated states, and is consistent with previous reports of red-shifted emission from triphenylene aggregates in organogels.12 In HMTP there is a much larger increase in PLQY on going from solution to film and powder. The PLQY increases over 6 times from 5% to 31% in the solid state. Increased PLQY in the solid state has been shown previously to result from increased molecular rigidity28 or aggregate-induced emission.2 For HMTP we rule out aggregate-induced emission as there is no change in the PL spectrum and the PL decay remains monoexponential on moving from solution to the solid state. The increase in PLQY could partly result from the fact that HMTP undergoes conformational transitions in solution (as indicated by NMR data).16 These conformational changes could give additional decay pathways in solution and thus increase the nonradiative decay rate. Such changes are prevented in the solid state; indeed X-ray diffraction shows that HMTP forms a regular crystal structure in the C2 conformation.16 However, this is unlikely to be the dominant effect, as time-resolved luminescence data only show a modest increase in lifetime from 5.4 to 6.7 ns going from solution to film, and this is not nearly enough to cause the observed change in PLQY by an decrease in the nonradiative rate alone. A possible explanation is that some or all of the alternate conformers of HMTP present in solution are able to absorb the
Fluorescence Enhancement in a Twisted Triphenylene Derivative incident light but are not emissive and so do not contribute to the PLQY. In the solid state the material is crystalline and only one conformer is present, as shown by X-ray crystallography.16 This is supported by the fact that the measured solution PLQY values result in an emission dipole of only 1.3 D, which is far from the measured solution absorption dipole moment of 3.6 D, whereas the measured film emission dipole moment is 2.1 D, more consistent with the solution absorption measurement. This suggests that absorbing dark conformers are present in the HMTP solution, and that the solid-state emission dipole moment gives the better measure of the emission dipole moment of the HMTP molecule. In contrast, the calculated triphenylene emission dipole moment of 0.5 D is consistent with the absorption dipole moment of 0.7 D; this is despite the fact that the Strickler-Berg equation for emission dipoles assumes a strongly allowed transition and that an average fluorescence lifetime has been used. Using the ∼40 ns lifetime, we attribute to the monomer on the basis of literature values10,27 instead of the average lifetime of 25.8 ns does not significantly change the estimated emission dipole moment, slightly reducing it from 0.5 to 0.4 D. Finally, we compare the radiative rate of the most dilute triphenylene solution (which should have the least aggregation) with that of the single HMTP conformer present in the solid state. On the basis of these measurements we can conclude that the twisted structure of HMTP increases the dipole moment compared to the case for triphenylene by a factor of ∼4 times and results in an increase in the radiative rate of ∼20 times. 5. Conclusion We have demonstrated that by using a twisted hexamethyltriphenylene derivative, we can alter the symmetry of the triphenylene S1 state and thus increase the radiative decay rate by a factor of 20. The twisting reduces intermolecular interactions, resulting in efficient photoluminescence in the solid state, with a quantum yield of 31%. The reduced tendency to π-stack is also demonstrated by the similarity of the photoluminescence spectra in dilute solution and neat film, in contrast to the situation in triphenylene. The PLQY of HMTP is 6 times higher in the solid state than in dilute solution, and we attribute this large difference to the presence of nonemissive chromophores in solution. Our results show that twisting is an effective molecular engineering strategy for control of intermolecular interactions that can enhance the performance of materials for solid-state luminescence applications.
J. Phys. Chem. A, Vol. 114, No. 51, 2010 13295 References and Notes (1) Melhuish, W. H. J. Phys. Chem. 1961, 65, 229. (2) Hong, Y. N.; Lam, J. W. Y.; Tang, B. Z. Chem. Commun. 2009, 4332. (3) Schouwink, P.; Schafer, A. H.; Seidel, C.; Fuchs, H. Thin Solid Films 2000, 372, 163. (4) Setayesh, S.; Grimsdale, A. C.; Weil, T.; Enkelmann, V.; Mullen, K.; Meghdadi, F.; List, E. J. W.; Leising, G. J. Am. Chem. Soc. 2001, 123, 946. (5) Tang, C. W.; Vanslyke, S. A.; Chen, C. H. J. Appl. Phys. 1989, 65, 3610. (6) Burn, P. L.; Lo, S. C.; Samuel, I. D. W. AdV. Mater. 2007, 19, 1675. (7) Salbeck, J.; Yu, N.; Bauer, J.; Weissortel, F.; Bestgen, H. Synth. Met. 1997, 91, 209. (8) Markovitsi, D.; Germain, A.; Millie, P.; Lecuyer, P.; Gallos, L. K.; Argyrakis, P.; Bengs, H.; Ringsdorf, H. J. Phys. Chem. 1995, 99, 1005. (9) Di Donato, E.; Vanzo, D.; Semeraro, M.; Credi, A.; Negri, F. J. Phys. Chem. A 2009, 113, 6504. (10) Kokkin, D. L.; Reilly, N. J.; Troy, T. P.; Nauta, K.; Schmidt, T. W. J. Chem. Phys. 2007, 126. (11) McKenna, M. D.; Barbera, J.; Marcos, M.; Serrano, J. L. J. Am. Chem. Soc. 2005, 127, 619. (12) Ikeda, M.; Takeuchi, M.; Shinkai, S. Chem. Commun. 2003, 1354. (13) Markovitsi, D.; Marguet, S.; Bondkowski, J.; Kumar, S. J. Phys. Chem. B 2001, 105, 1299. (14) Adam, D.; Schuhmacher, P.; Simmerer, J.; Haussling, L.; Siemensmeyer, K.; Etzbach, K. H.; Ringsdorf, H.; Haarer, D. Nature 1994, 371, 141. (15) Ahmed, F. R.; Trotter, J. Acta Crystallogr. 1963, 16, 503. (16) Wang, Y.; Stretton, A. D.; McConnell, M. C.; Wood, P. A.; Parsons, S.; Henry, J. B.; Mount, A. R.; Galow, T. H. J. Am. Chem. Soc. 2007, 129, 13193. (17) Demas, J. N.; Crosby, G. A. J. Phys. Chem. 1971, 75, 991. (18) Rusakowicz, R.; Testa, A. C. J. Phys. Chem. 1968, 72, 2680. (19) Greenham, N. C.; Samuel, I. D. W.; Hayes, G. R.; Phillips, R. T.; Kessener, Y.; Moratti, S. C.; Holmes, A. B.; Friend, R. H. Chem. Phys. Lett. 1995, 241, 89. (20) de Mello, J. C.; Wittmann, H. F.; Friend, R. H. AdV. Mater. 1997, 9, 230. (21) Stricker, S. J.; Berg, R. A. J. Chem. Phys. 1962, 37, 814. (22) Liu, Z. T.; Kwong, C. Y.; Cheung, C. H.; Djurisic, A. B.; Chan, Y.; Chui, P. C. Synth. Met. 2005, 150, 159. (23) Greenham, N. C.; Friend, R. H.; Bradley, D. D. C. AdV. Mater. 1994, 6, 491. (24) Khalfin, V. B.; Gu, G.; Burrows, P. E.; Garbuzov, D. Z.; Forrest, S. R. 1998, 58. (25) Samuel, I. D. W.; Crystall, B.; Rumbles, G.; Burn, P. L.; Holmes, A. B.; Friend, R. H. Chem. Phys. Lett. 1993, 213, 472. (26) Becker, K.; Da Como, E.; Feldmann, J.; Scheliga, F.; Csanyi, E. T.; Tretiak, S.; Lupton, J. M. J. Phys. Chem. B 2008, 112, 4859. (27) Wallace, W. L.; Vanduyne, R. P.; Lewis, F. D. J. Am. Chem. Soc. 1976, 98, 5319. (28) Harding, R. E.; Lo, S. C.; Burn, P. L.; Samuel, I. D. W. Org. Electron. 2008, 9, 377.
JP106622N
