9700
J. Phys. Chem. B 2008, 112, 9700–9708
Fluorescence Excitation and Emission Spectroscopy on Single MEH-PPV Chains at Low Temperature Florian A. Feist and Thomas Basche´* Institut fu¨r Physikalische Chemie, Johannes Gutenberg-UniVersita¨t, 55099 Mainz, Germany ReceiVed: March 25, 2008; ReVised Manuscript ReceiVed: June 9, 2008
Fluorescence emission and excitation spectra of single MEH-PPV polymer molecules dispersed in thin PMMA films have been recorded at 1.2 and 20 K. We observe single as well as multichromophore emission in single chain emission spectra, whereby the relative fractions depend on the two different molecular weights (50 and 350 kDa) studied. The molecular weight also affects the distribution of peak emission maxima, which is monomodal (bimodal) for the low (high) molecular weight. The appearance of an additional “red” subpopulation for the high molecular weight sample is attributed to interactions of multiple chromophores from a sufficiently flexible single chain. The comparison of emission spectra appearing in the “blue” as well as “red” subpopulations suggests that these intrachain interactions rather lead to ground-state aggregates than excimers. Independent of the molecular weight, large variations in spectral shape and apparent line width in the emission spectra have been observed. Occasionally, we find very narrow purely electronic zero-phonon lines both in emission and in excitation spectra, with line widths down to the instrumental resolution. In accordance with earlier literature data it is argued that linear electron-phonon coupling should be quite strong for MEH-PPV in PMMA, leading to only a small fraction of chromophores exhibiting zero-phonon lines. In addition, spectral diffusion, which manifests itself by several time-dependent line shifting and broadening phenomena, contributes to the substantial variations of spectral shapes. Excitation experiments with particularly stable chromophores provide an upper limit for the optical line width (∼0.1 cm-1), which at 1.2 K can actually approach the lifetimelimited homogeneous width. Raising the temperature to 20 K leads to line broadening and typically, to disappearance of zero-phonon lines. The failure to observe zero-phonon lines of chromophores supposedly serving as donors in intramolecular energy transfer is tentatively attributed to fast transfer rates, resulting in strongly broadened lines which escape detection even at 1.2 K. 1. Introduction The optical properties of organic semiconductors, especially conjugated polymers, emerge as a field of growing scientific interest, since they are crucial for a wide range of opto-electronic applications, including light-emitting diodes,1,2 solar cells,3,4 organic lasers,5–8 organic thin-film transistors 9,10 and chemical sensors.11–13 According to our present understanding, conjugated polymers are considered as multichromophoric systems, each chromophore formed by a certain number of repeat units.14 Some of their characteristic photophysical properties are a direct reflection of the distribution of conjugation lengths which translates into a distribution of transition energies of the chromophores. Consequently, light can be absorbed by a multitude of chromophores and the excitation energy thereafter is transferred to few or even only one lowest-energy site.15 Optical spectroscopy has been crucial to elucidate the electronic level structure and the intra- and intermolecular excitation dynamics in conjugated polymers. For the widely used poly(2-methoxy-5-(2′-ethylhexyloxy)-p-phenylene-vinylene) (MEH-PPV) profound insights into the photophysics have been gained from studies at the ensemble level in dilute solutions and thin solid films. Results include the determination of excitedstate lifetimes,16–18 the study of energy transfer processes 19,20 and the interconnection of photophysics and morphology.21 Regarding the latter issue, the existence of low energy excited * To whom correspondence should be addressed. E-mail: thomas.basche@ uni-mainz.de. Telephone: + 49 (0) 6131 39 22707. Fax: + 49 (0) 6131 39 23953.
states has been postulated, arising from electronic interactions between different chromophoric units, either located on the same or on separate polymer chains. For MEH-PPV in solid films, it has been suggested that such interchromophoric interactions give rise to the formation of excimers (interaction only of an excited with a ground-state chromophore without a stable ground-state dimer).22 Recently, optical studies have been extended to the level of single polymer chains, whereby the obscuring effect of ensemble averaging is avoided. These studies have given additional insights into basic spectroscopic properties and energy transfer processes of conjugated polymers. Collective behavior like single-step photobleaching and discrete fluorescence intensity jumps has been found, giving evidence for strong communication between the individual chromophores along the polymer chain.23–28 The time scale of these fluorescence intensity fluctuations allowed conclusions about typical energy migration lengths and electronic energy diffusion constants.29 Furthermore, polarization-dependent measurements suggested possible conformations adopted by single polymer chains, which accommodate straight chain segments and therefore localized chromophoric units with effective π-conjugation.30 Many spectroscopic signatures (e.g., intermittent fluorescence,31–33 emission spectra,31,33 and photon correlation32) crucially depend on the chain conformation due to its effect on the efficiency of intramolecular energy transfer. Taking advantage from this fact, chain conformations have been controlled by varying polarity of solvents31,32 or molecular weight and polarity of the host matrix.33 A considerable number of experiments have been performed at
10.1021/jp802585m CCC: $40.75 2008 American Chemical Society Published on Web 07/23/2008
Single MEH-PPV Chains low temperatures (down to 5 K), because these measurements profit from significantly narrower emission spectra compared to room temperature and the freezing of conformational dynamics. Therefore, they give access to the distribution of localized emitting states. It could be verified that the fluorescence of a single polymer chain originates from one or (in the case of several independent energy transfer channels) few chromophoric units,27,34 their superposition leading to the inhomogeneously broadened ensemble spectra. Furthermore, it was shown that the average number of emitting sites is related to the chain length.35 Moreover, spectral diffusion, which appears to be a common feature of conjugated polymers, has been studied in detail both in inert polymer hosts and in host-matrix free environments.25,36–38 There have been a number of discrepancies between results of MEH-PPV single chain emission spectroscopy obtained by different groups.27,36,37,39 It has been argued that sharp emission spectra observed by Schindler et al.36,37 were blue-shifted compared to a distribution obtained by Barbara and co-workers, possibly caused by partial photodegradation of the polymer.40 Mirzov et al. have compared data from different groups, taking into account different experimental conditions and techniques of sample preparation, and concluded that the reported discrepancies were probably due to the sample structure, experimental procedures and different temperature.41 Concerning emission line widths, it was not specified what kind of line width was meant. In the present study it will be shown that at very low temperatures (1.2 K here or 5 K in refs 36 and 37) sharp zerophonon lines (ZPLs) can be observed which typically disappear in the broad phonon wing at 20 K. Therefore, any discussion of line widths has to take into account the actual temperature of the investigation. Recently, we have reported the observation of very narrow line widths in optical spectra of single MEH-PPV chains imbedded in PMMA at 1.2 K.42 Both in high-resolution emission spectra and in fluorescence excitation spectra, we were able to detect purely electronic ZPLs, their widths being orders of magnitude smaller than the narrowest lines reported so far,36,37 yet still limited by the experimental resolution. The rare observation of sharp ZPLs in the excitation and emission spectra had been mainly attributed to spectral diffusion.42 Although several novel experiments presented here indeed give proof that spectral diffusion has a considerable influence on the spectral shape, reconsideration of earlier work suggests that linear electron-phonon coupling is expected to be quite strong in MEH-PPV.43,44 Accordingly, the low fraction of single polymer chains in which chromophores with ZPLs are observed is attributed to the influence of electron-phonon coupling as well as spectral diffusion. Another major objective of this study is to compare the results of low temperature (1.2 K) single chain spectroscopy for two different molecular weights of MEH-PPV, both dispersed in PMMA. We first address the question to which extent the fractions of single and multichromophore emissions and the distributions of peak emission maxima depend on the molecular weight. We then study the spectral shapes of single chromophore emission spectra, which reveal the percentages of chromophores with ZPLs. Furthermore, temporal variations of emission spectra are tracked for longer observation times. To get insight into line broadening processes, single polymer chains are studied at different temperatures (1.2 and 20 K) for the first time. In a next set of experiments we employ fluorescence excitation spectroscopy to conduct more accurate line width measurements than can be achieved by emission spectroscopy. Finally, we
J. Phys. Chem. B, Vol. 112, No. 32, 2008 9701 discuss our results including a comparison to previous work on low temperature single chain spectroscopy. We emphasize unifying and opposing tendencies in the spectroscopy of the two different molecular weight samples. Interestingly, the appearance of “blue” and “red” subpopulations for the high molecular weight sample and the basic similarity of the corresponding single chromophore emission spectra permit conclusions about the nature of aggregate emission in MEHPPV. 2. Experimental Details MEH-PPV with an averaged molecular weight of Mw ∼ 200 kDa and large polydispersity (PD ) 5) was purchased from Aldrich. Two fractions with Mw ∼50-60 kDa and PD ) 1.1-1.3 (henceforward referred to as low molecular weight, LMw) and Mw ∼350 kDa and PD ) 1.2 (henceforward referred to as high molecular weight, HMw) were selected by gel permeation chromatography. The MEH-PPV was first completely dissolved in tetrahydrofuran (THF) and afterward appropriate concentrations were prepared by dilution with toluene containing poly(methylmethacrylate) (PMMA) at a concentration of 20 g/L. From these solutions, thin films (thickness ≈ 100 nm) were spin-coated onto thoroughly cleaned glass cover slides. During and after the spin coating process, the samples were constantly kept under argon. The sample to be investigated was transferred into an optical cryostat under constant argon stream and kept under vacuum conditions. For the measurements at 1.2 K, the sample was immersed in superfluid liquid helium. For the measurements with variable temperatures (2-20 K), the sample was cooled by a stream of He gas and the temperature regulated by a heat-exchanger. Temperature was measured by a calibrated Si diode being mounted in close proximity to the sample. The fluorescence imaging and spectroscopy of single polymer molecules was conducted using a home-built confocal microscope.45 The fluorescence emission spectra were taken with an excitation wavelength λexc ) 488 nm and a laser intensity in the range of 25 W/cm2 to 13.5 kW/cm2. All emission light transmitted by a 500 nm long pass filter was imaged onto the entrance slit of a spectrograph equipped with a liquid nitrogen cooled CCD camera. The spectrograph was operated in low (150 grooves/ mm grating; resolution 20 cm-1) and high (1800 grooves/mm; resolution 2 cm-1) resolution mode. The fluorescence excitation spectra were recorded by using a tunable, linearly polarized ring dye-laser (line width ∼0.1 cm-1) operated with pyrromethene 546. While scanning the dye laser across its tuning range (525-545 nm), all emission light transmitted by a 555 nm long pass filter was collected with an avalanche photo diode (APD). The excitation spectra were taken with an intensity of ∼ 5-150 W/ cm2. 3. Results A selection of emission spectra of MEH-PPV in PMMA recorded at 1.2 K is presented in Figure 1. The spectra shown in Figure 1a,b and Figure 1c-f are attributed to multichromophoric emission and emission from single chromophoric units within the conjugated polymer molecules, respectively. While these examples have been chosen from the HMw sample, the appearance of the spectra is also representative for the LMw sample which again is in accordance with earlier results42 for a slightly different low molecular weight sample. In case of single chromophore emission it is assumed that the multitude of chromophores excited by the laser (488 nm) funnel the excitation energy to one low-energy chromophore, which then relaxes to
9702 J. Phys. Chem. B, Vol. 112, No. 32, 2008
Feist and Basche´
Figure 2. Distribution of peak emission maxima observed for MEHPPV at 1.2 K (gray bars): (a) LMw; (b) HMw. The drawn lines are single (a) or double (b) Gaussians fitted to the data. The black bars give the subpopulation of molecules for which ZPLs could be detected.
Figure 1. Emission spectra of single MEH-PPV chains in PMMA taken with an excitation wavelength of 488 nm at 1.2 K. Substantial variations in spectral shapes are evident. In part a, two emitting chromophores are visible; in part b, the exact number of emitting chromophores is not determinable. The spectra in parts c-f are attributed to emission from single chromophores. While in parts e and f, a narrow ZPL is visible, the spectra in parts c and d are substantially broadened. All examples are taken from the HMw sample, but are also representative for the LMw sample. (Iexc ) 2.7 kW/cm2, resolution 20 cm-1).
the ground-state through fluorescence. On the other hand, multichromophoric emission is expected when either certain chromophores could not transfer their excitation energy despite the presence of lower-energy sites on the chain (e.g., for orientational reasons) or even different “energy funnels” were present on a single polymer chain.28 In our experiments, multichromophoric emission was found in the majority of cases (58% LMw, 74% HMw). While the spectrum shown in Figure 1a clearly appears as a superposition of the emission of two chromophores, the number of simultaneously emitting chromophores is basically undeterminable for the example given in Figure 1b. Multichromophoric emission often was subject to rich dynamics, including relative intensity variations and (dis-) appearance of particular chromophores. Nevertheless, in particularly favorable cases up to four chromophores could be assigned to a single chain emission spectrum. By analyzing the emission spectra of 266 single polymer molecules of the LMw sample and 146 single polymer molecules of the HMw sample, distributions of peak emission maxima were obtained which are shown in Figure 2. Please note that these distributions were built from single as well as multichromophore emissions. The emission spectra of the LMw sample are distributed in a monomodal way and are well described by a Gaussian with maximum at 18250 cm-1 and full width at half-maximum (fwhm) of 850 cm-1. This width is considerably larger than for organic dye molecules in PMMA and reflects the distribution of chromophore sizes and the different environments of these chromophores (static disorder). The HMw sample clearly gives rise to an additional contribution at lower energies, leading to a bimodal distribution, reported for MEH-PPV single chain emission spectra before.39,46 Accordingly, the histogram was fitted with a combination of two Gaussians with maxima of 18100 and 16750 cm-1 and fwhms of 1050 and 850 cm-1 for the high- and low-energy populations,
Figure 3. (a) Emission from a single MEH-PPV chain dispersed with a low-resolution grating (resolution 20 cm-1). (b-f) Series of emission spectra of the same molecule recorded with a high-resolution grating (resolution 2 cm-1), exhibiting sharp lines of varying number and position (λexc ) 488 nm, Iexc ) 2.7 kW/cm2).
respectively. Surprisingly, the ratio of the “blue” and “red” subpopulations was subject to pronounced variations when comparing samples prepared by nominally the same procedure. Therefore, the contribution of “red” spectra seems to be linked not only to the molecular weight, but also to other experimental details during sample preparation, possibly influencing the chain morphology and thus altering the degree of interaction between chromophores. So far, we have not been able to identify the origin of this stunning observation. A remarkable feature of the emission spectra, which can be easily perceived in Figure 1, is the large variation of spectral shapes. The spectra in Figure 1e,f exhibit sharp, purely electronic ZPLs of the (S0 r S1) transition. In high-resolution emission spectra (see Figures 3 and 4) the widths of the ZPLs can become quite narrow reaching the instrumental resolution limit (2-3 cm-1).42 Sharp spectral features, however, could only be observed for a minority of investigated molecules (20% in the LMw and 19% in the HMw sample, respectively). The latter percentages relate to results obtained by low resolution emission spectroscopy. Occasionally, it happened that for a given
Single MEH-PPV Chains
J. Phys. Chem. B, Vol. 112, No. 32, 2008 9703
Figure 5. Emission spectra belonging to the low-energy subpopulation of the HMw sample with (a) and without (b) ZPL. In part a, the vibrational features summarized in Table 1 are highlighted. (λexc ) 488 nm, Iexc ) 2.7 kW/cm2, resolution 20 cm-1).
Figure 4. (a) Low-resolution emission spectrum of single MEH-PPV chains at 20 K, including a sharp feature. (b) Sequence of highresolution emission spectra of the same molecule taken at different temperatures. At 2 K, the spectrum shows a narrow ZPL, which does not appear at 20 K. (λexc ) 488nm, Iexc ) 1.3 kW/cm2, LMw sample).
molecule a ZPL was observed in high resolution but not in low resolution mode (see Figure 3). These ZPLs were typically prone to spectral diffusion which most probably prevented their observation in low resolution spectra. Therefore, the percentages given above have to be considered as lower limits, although the numbers should represent reasonable estimates. Most emission spectra exhibited rather broad spectral lines with substantial variations of the apparent widths as shown in Figure 1c,d. ZPLs were also observed in multichromophoric emission spectra (see Figure 1a) and occasionally several chromophores in such spectra exhibited a ZPL built on their electronic origins. In Figure 2, we have highlighted within the full distributions of emission maxima the spectral positions at which ZPLs were found. For none of the two samples, the observation of narrow spectral features is related to spectral position. Therefore, we rule out the possibility of photodegradation as cause for the narrow emission lines, which should lead to a substantial blueshift of the spectra as has recently been argued.40 The single chromophore spectra from the HMw sample presented in Figure 1 all belong to the “blue” subpopulation. The same variety of spectral shapes was also observed in the “red” subpopulation. Two representative “red” spectra of the HMw sample are shown in Figure 5, both with peak wavelengths >600 nm. One exhibits a ZPL and has a shape similar to the spectrum shown in Figure 1e. The other one is substantially broadened and resembles the spectrum from Figure 1(c). The comparison of the spectra from the “red” and “blue” subpopulations points to a basic similarity: the casual occurrence of zerophonon transitions and vibrational substructure. In Table 1, the ground-state vibrational frequencies as taken from the emission spectra of two chromophores from the “blue” and “red” subpopulations, respectively, are given. Only sufficiently intense vibrational lines were considered, for the “red” spectrum, they
Figure 6. Fluorescence emission (a, c; λexc ) 488 nm, Iexc ) 2.7 kW/ cm2, resolution 20 cm-1) and excitation spectra (b, d; Iexc ) 150W/ cm2, resolution 0.1 cm-1) of single MEH-PPV molecules from the LMw (a, b) and the HMw (c, d) samples. Please note the differences in wavelength scale. The inset in part d shows the ZPL with increased resolution (Iexc ) 7W/cm2) and a Gaussian fit to the data.
TABLE 1: Vibrational Modes of Single MEH-PPV Molecules from the “Blue” and the “Red” Subpopulation (in cm-1) blue red
1
2
3
4
5
6
7
127 135
221 244
1111 1116
1315 1301
1434 1415
1577 1570
1710 1681
are highlighted in Figure 5(a). It is seen in Table 1 that there is good agreement between the two data sets. In addition to emission spectra, we also have measured excitation spectra of single MEH-PPV chains. In these experiments a principal limitation was given by the scan range (525-545 nm) of the dye laser, which restricted excitation spectroscopy to only part of the spectral distributions shown in Figure 2. Yet, this particular laser dye allowed to record excitation spectra of the LMw as well as the HMw sample. We found similar behavior in both samples, two examples of which are shown in Figure 6. In particular, only in a very small number of cases ( 6 no ZPLs can be detected and furthermore mentioned that at the time of publication no zero-phonon features had been reported for dopant molecules with ∆µ values exceeding 2-3 D. Putting all this information together (and momentarily not considering spectral diffusion), one immediate conclusion is that the observation of ZPLs for MEH-PPV embedded in an amorphous host polymer (here PMMA) should actually be unlikely. Indeed, in the bulk site-selective fluorescence experiments by Pauck et al. mentioned above no ZPLs have been observed for OPV-5 and PPV.44 Taking into account that in case of our MEH-PPV samples the probability of observing a ZPL in a single chain experiment was roughly 20%, it can be extrapolated that in a bulk site-selective experiment the observation of ZPLs might be rather difficult. In general terms, this comparison nicely demonstrates the ability of single molecule spectroscopy to observe rare events which would be drowned by ensemble averaging. In the present case, it is reasonable to assume that the (linear) electron-phonon coupling strengths of the chromophores in the chains are widely distributed as has been shown recently for a ladder-type conjugated polymer.49 Hence, in the single molecule experiment we can search for chromophores with larger values of the Debye-Waller factor, while in the ensemble experiment the most probable value, which may be too small for a significant ZPL intensity, will inevitably dominate. Unfortunately, for the MEH-PPV in PMMA system it is not possible to extract reliable values for the Debye-Waller factor of the [0,0]-electronic transition from our measurements. To give an example, in the spectrum shown in Figure 5a the true phonon sideband of the sharp ZPL at 605 nm is buried under the broad wing extending up to 650 nm and being a superposition of contributions from inter- and intramolecular modes including low frequency (frustrated) torsional motions of the polymer chain. (Actually, to determine the Debye-Waller factor one preferentially would have to refer to high resolution emission spectra, which, however, would not change the principal problem.) As mentioned above, the actual value of ∆µ and therefore the strength of linear electron-phonon coupling do not only depend on the dopant but also on the chemical nature and phonon structure of the host.54 In this context it is interesting to compare our results in PMMA to a recent low temperature (T ) 5 K) single molecule study of MEH-PPV in polystyrene (PS).36,37 In contrast to our observations, in the investigations of Schindler et al.36,37 the finding of ZPLs represented the typical case. Possible explanations for these differences between the
9706 J. Phys. Chem. B, Vol. 112, No. 32, 2008 two host matrices might be a substantially weaker electronphonon coupling strength and less efficient spectral diffusion in the PS host favoring the observation of ZPLs. In addition to electron-phonon coupling, spectral diffusion strongly contributes to the variation of spectral shapes depicted in Figure 1. In accordance with other investigations,25,36 several time-dependent spectral shifting and broadening phenomena have been observed. It is, however, not easily possible to distinguish the relative weight of electron-phonon coupling and spectral diffusion contributions for a given spectrum of MEHPPV in PMMA. Furthermore, it is not clear to which extent both effects may be coupled since structural relaxation might not only change the transition frequencies of chromophores but also their electron-phonon coupling strengths. In the same direction, not only electron-phonon coupling but also spectral diffusion is expected to increase with ∆µ. Anyway, the combined action of both effects is ultimately responsible for the low fraction of ZPLs observed for MEH-PPV in PMMA. It is interesting that these fractions have been the same for both molecular weights. In the bulk investigation by Pauck et al. it was reported that ZPLs disappeared when going from OPV-3 to OPV-5.44 Despite a number of differences between both studies, this result qualitatively is in accordance with our observations. The two polymer samples used here are composed of many chromophores whose average size is assumed to be about 10 repeat units. Hence, there should be no difference in the probabilities for observing ZPLs which, however, should increase if oligomers with a small number of repeat units would be studied. Such experiments are planned in the future, because they might give additional insights into the size dependence of electron-phonon coupling. With respect to the single chromophore emission lineshapes, it is remarkable that no obvious differences have been found between the “blue” and “red” subpopulations of the HMw sample. In particular, zero-phonon transitions and comparable vibrational substructure could also be detected in the “red” subpopulation (cf. Figure 5a and Table 1). This close resemblance suggests that the low-energy excited states in the HMw sample may not be excimers, which were reported for MEHPPV in solid-state films.22 For an excimer, resulting from the interaction of an excited with a ground-state chromophore, which cannot form a dimer in the relaxed state, we would expect a spectrum substantially different from those belonging to localized excited states; in particular, we would expect a broad, unstructured emission spectrum. Thus, it appears to be more likely that the low-energy transitions (“red” subpopulation) are due to weakly coupled ground-state aggregates. The presence of vibrational substructure for aggregates and its loss for excimers has recently been shown experimentally for a similar conjugated polymer on the ensemble level.55 Considering the LMw sample, it has been concluded before that no obvious relation between the occurrences of multichromophoric emission and “red” emitting chromophores exists. Accordingly, a substantial number of chromophores (∼20) in a single chain must not necessarily lead to aggregation, even though our preparation conditions suggest a rather coiled chain conformation to prevail.31,32 It appears that the single chains must be long and flexible enough to drive the formation of aggregates. Interestingly, this process is sensitive to at present hidden parameters, since we could not really control the amount of “red”-emitting chromophores in the HMw sample even for nominally the same preparation conditions. With respect to aggregation, the behavior of single chains as studied here may be quite different to that of polymers in a thin solid film, where
Feist and Basche´ interchain interactions are abundant. Under such circumstances excimers may be more easily feasible than in a single chain. We would like to point out, however, that we cannot rule out the possibility of excimer formation in our samples completely. Provided that the fluorescence quantum yield of excimers would be significantly lower than the one for intrachain excitations and aggregates,22 polymer chains containing excimers would appear as spots of relatively weak intensity in the fluorescence images and thus might have been ignored during the experiments. An important parameter accessible in low temperature single molecule experiments is the homogeneous line width, which is intimately connected to the nature of the corresponding electronic transitions and their coupling to the environment. It is composed of two contributions, the lifetime of the excited state (T1) and the pure dephasing time (T/2):
∆νhom )
1 1 + 2πT1 πT/
(1)
2
At very low temperature pure dephasing times can become very long and the line width then will be dominated by the excited-state lifetime contribution. From single molecule fluorescence lifetime measurements of MEH-PPV at 1.2 K, we have obtained values in the range of several 100 ps, which is in accordance with other single molecule56 and ensemble measurements.18 Assuming an average fluorescence lifetime of 500 ps, the purely lifetime limited line width should be around 0.01 cm-1. This value represents a lower limit because due to additional dephasing processes and/or spectral diffusion the homogeneous line width may be considerably larger.57 In addition, we cannot completely rule out power broadening, because excitation ZPLs typically disappeared, when attempting to measure a series of excitation power dependent spectra. In previous studies the narrowest line widths have been reported by Schindler et al., who found values of around 20 cm-1 for MEH-PPV in PS at 5 K.36,37 Although measured at 5 K, these values appear to be very large and seem to be far from the homogeneous line width. The narrowest line width obtained in our high resolution emission experiments were in the range of 2-3 cm-1. This value basically represents the instrumental resolution and therefore cannot be associated with the homogeneous line width of the optical transition. The largely improved resolution of laser-based fluorescence excitation spectroscopy is indispensable to reach the region of the true homogeneous line width. For the data presented in Figure 6, line widths of 0.14 and 0.3 cm-1 have been found. Again, the widths are ultimately limited by the instrumental resolution. Nevertheless, these results clearly indicate that by suitable experimental techniques, line widths approaching the range of the lifetime limited width can be found. Although the observation of such narrow lines is rare due to electron-phonon coupling and/or spectral diffusion, their finding strongly supports the molecular description of the low energy transitions in conjugated polymers. Due to the limited scan range of the dye laser (525-545 nm), for the MEH-PPV molecule portrayed in Figure 6c,d an excitation spectrum could only be taken for the chromophore whose emission maximum appeared around 541 nm. The other two chromophores contributing to the emission spectrum with electronic origins at 559 nm and 574 nm were simply not accessible by the dye laser. According to the foregoing estimate the average number of chromophores in a single chain of the HMw sample should be around 130. The fact that out of this large number only 3 chromophores emit independently is attributed to efficient electronic coupling between chromophores.
Single MEH-PPV Chains More difficult to understand is the observation that in all excitation spectra taken so far only a single ZPL was found. These excitation ZPLs always corresponded to an emission line (not necessarily a ZPL). Even when taking into account the limited scan range of the dye laser (and the probability of strong spectral diffusion and electron-phonon coupling), this is a surprising result in light of the estimated total number of chromophores (130). Obviously, so far only chromophores could be observed in the excitation spectra that served as acceptors in the electronic energy transfer process before relaxing through fluorescence, and therefore also appeared in the corresponding emission spectrum. ZPLs of higher-energetic chromophores, which transferred their excitation energy completely to lowerenergy sites, could not be detected. Such “donor” ZPLs might be appreciably broadened by rapid energy transfer making it very difficult to detect them with sufficient signal-to-noise ratio. This scenario basically agrees with the assumption that in our samples, which were spin-cast from toluene, rather coiled conformations prevail31,32 favoring fast energy hopping between the chromophores.
J. Phys. Chem. B, Vol. 112, No. 32, 2008 9707 are thought to be appreciably broadened, limiting on one hand their detection probability but on the other hand giving direct access to intramolecular energy transfer rates58 in MEH-PPV, numbers for which only rough estimates exist. Furthermore, the temperature dependence of the homogeneous line width would give unprecedented insight into the coupling to low frequency excitations. Acknowledgment. We thank Eva Wa¨chtersbach for GPC size fractionation of the MEH-PPV sample. Financial support from the Graduate School POLYMAT (F.A.F.), the German Science Foundation (SFB 625) and the Fonds der chemischen Industrie is gratefully acknowledged. Supporting Information Available: Text and figures showing experimental results on the time dependent width of emission spectra as well as a simulation verifying the emergence of broadened spectra by spectral diffusion. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes
5. Conclusion and Outlook We have conducted fluorescence emission and excitation spectroscopy on single chains of the conjugated polymer MEHPPV in PMMA at temperatures of 1.2 and 20 K. Two different samples were studied, allowing for comparison between different molecular weights (50 and 350 kDa). As was reported in other studies of MEH-PPV in PMMA, we observe single as well as multichromophore emission in both samples. In accordance with our previous work,42 we find a monomodal distribution of peak emission maxima for a low molecular weight sample. For a high molecular weight sample a bimodal distribution (“blue” and “red” subpopulations) emerges, indicating that the shape of such a distribution depends on the molecular weight.39,46 Our results further suggest that the “red” subpopulation is due to ground-state aggregates rather than excimers. This conclusion is supported by the observation that the emission spectra from the “blue” and “red” subpopulations share the same characteristics. The shape of emission spectra of single chromophores, which did not differ for the two molecular weights, were subject to pronounced variations, ranging from spectra with sharp, purely electronic ZPLs of spectrograph-limited width to very broad ones. We attributed these variations to a distribution of electron-phonon coupling strengths and spectral diffusion. The latter phenomenon was visualized by temporal fluctuations of spectral positions and widths. By increasing the temperature to 20 K, ZPLs typically disappeared due to the temperature dependence of the homogeneous line width and the Debye-Waller factor. We could also find ZPLs in excitation spectra of single MEH-PPV chains for both samples. These observations were very rare, however, and furthermore confined to emitting chromophores. For these transitions, line widths down to the instrumentally limited value of ∼0.1 cm-1 were measured at 1.2 K, giving an upper limit of the homogeneous line width of MEH-PPV chromophores. We have emphasized the importance of linear electron phonon coupling in MEH-PPV/PMMA, which generally limits the occurrence of zero-phonon features in amorphous host-guest systems. In future experiments we will study other amorphous host polymers as poly(styrene) or poly(norbornene) to check whether the fraction of chromophores with ZPLs can be significantly increased. More intense and stable ZPLs are a prerequisite to interrogate those chromophores which serve as energy donors in intrachain energy transfer. Their line widths
(1) Burroughes, J. H.; Bradley, D. D. C.; Brown, A. R.; Marks, R. N.; Mackay, K.; Friend, R. H.; Burns, P. L.; Holmes, A. B. Nature 1990, 347, 539. (2) Friend, R. H.; Gymer, R. W.; Holmes, A. B.; Burroughes, J. H.; Marks, R. N.; Taliani, C.; Bradley, D. D. C.; Dos Santos, D. A.; Bre´das, J. L.; Lo¨gdlund, M.; Salaneck, W. R. Nature 1999, 397, 121. (3) Lo, S.-C.; Burn, P. L. Chem. ReV. 2007, 107, 1097. (4) Gu¨nes, S.; Neugebauer, H.; Sariciftci, N. S. Chem. ReV. 2007, 107, 1324. (5) Moses, D. Appl. Phys. Lett. 1992, 60, 3215. (6) Tessler, N.; Denton, G. J.; Friend, R. H. Nature 1996, 382, 695. (7) Hide, F.; DiazGarcia; M, A.; Schwartz, B. J.; Andersson, M. R.; Pei, Q. B.; Heeger, A. J. Science 1996, 273, 1833. (8) Samuel, I. D. W.; Turnbull, G. A. Chem. ReV. 2007, 107, 1272. (9) Horowitz, G. AdV. Mater. 1998, 10, 365. (10) Murphy, A. R.; Fre´chet, J. M. J. Chem. ReV. 2007, 107, 1066. (11) Zhou, Q.; Swager, T. M. J. Am. Chem. Soc. 1995, 117, 7017. (12) McQuade, D. T.; Pullen, A. E.; Swager, T. M. Chem. ReV. 2000, 100, 2537. (13) Thomas, S. W., III; Joly, G. D.; Swager, T. M. Chem. ReV. 2007, 107, 1339. (14) Bre´das, J.-L.; Cornil, J.; Beljonne, D.; Dos Santos, D. A.; Shuai, Z. Acc. Chem. Res. 1999, 32, 267. (15) Bre´das, J.-L.; Beljonne, D.; Coropceanu, V.; Cornil, J. Chem. ReV. 2004, 104, 4971. (16) Smilowitz, L.; Hays, A.; Heeger, A. J.; Wang, G.; Bowers, J. E. J. Chem. Phys. 1993, 98, 6504. (17) Samuel, I. D. W.; Crystall, B.; Rumbles, G.; Burn, P. L.; Holmes, A. B.; Friend, R. H. Chem. Phys. Lett. 1993, 213, 472. (18) Gaab, K. M.; Bardeen, C. J. J. Phys. Chem. B 2004, 108, 4619. (19) Hayes, G. R.; Samuel, I. D. W.; Phillips, R. T. Phys. ReV. B 1995, 52, R11569. (20) Nguyen, T.-Q.; Wu, J.; Doan, V.; Schwartz, B. J.; Tolbert, S. H. Science 2000, 288, 652. (21) Schwartz, B. J. Annu. ReV. Phys. Chem. 2003, 54, 141. (22) Jakubiak, R.; Collison, C. J.; Wan, W. C.; Rothberg, L. J.; Hsieh, B. R. J. Phys. Chem. A 1999, 103, 2394. (23) Vanden Bout, D. A.; Yip, W.-T.; Hu, D.; Fu, D.-K.; Swager, T. M.; Barbara, P. F. Science 1997, 277, 1074. (24) Mu¨ller, J. G.; Lemmer, U.; Raschke, G.; Anni, M.; Scherf, U.; Lupton, J. M.; Feldmann, J. Phys. ReV. Lett. 2003, 91, 267403. (25) Pullerits, T.; Mirzov, O.; Scheblykin, I. G. J. Phys. Chem. B 2005, 109, 19099. (26) Mirzov, O.; Cichos, F.; von Borczyskowski, C.; Scheblykin, I. G. Chem. Phys. Lett. 2004, 386, 286. (27) Yu, Z.; Barbara, P. F. J. Phys. Chem. B 2004, 108, 11321. (28) Liang, J.-J.; White, J. D.; Chen, Y. C.; Wang, C. F.; Hsiang, J. C.; Lim, T. S.; Sun, W. Y.; Hsu, J. H.; Hsu, C. P.; Hayashi, M.; Fann, W. S.; Peng, K. Y.; Chen, S. A. Phys. ReV. B 2006, 74, 085209. (29) Hu, D.; Yu, J.; Barbara, P. F. J. Am. Chem. Soc. 1999, 121, 6936. (30) Hu, D.; Yu, J.; Wong, K.; Bagchi, B.; Rossky, P. J.; Barbara, P. F. Nature 2000, 405, 1030. (31) Huser, T.; Yan, M.; Rothberg, L. J. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 11187. (32) Hollars, C. W.; Lane, S. M.; Huser, T. Chem. Phys. Lett. 2003, 370, 393.
9708 J. Phys. Chem. B, Vol. 112, No. 32, 2008 (33) Sartori, S. S.; De Feyter, S.; Hofkens, J.; Van der Auweraer, M.; De Schryver, F.; Brunner, K.; Hofstraat, J. W. Macromolecules 2003, 36, 500. (34) Rønne, C.; Tra¨gårdh, J.; Hessman, D.; Sundstro¨m, V. Chem. Phys. Lett. 2004, 388, 40. (35) Schindler, F.; Jacob, J.; Grimsdale, A. C.; Scherf, U.; Mu¨llen, K.; Lupton, J. M.; Feldmann, J. Angew. Chem. 2005, 117, 1544. (36) Schindler, F.; Lupton, J. M.; Feldmann, J.; Scherf, U. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 14695. (37) Schindler, F.; Lupton, J. M. ChemPhysChem 2005, 6, 926. (38) Mirzov, O.; Pullerits, T.; Cichos, F.; von Borczyskowski, C.; Scheblykin, I. G. Chem. Phys. Lett. 2005, 408, 317. (39) Lee, Y. J.; Kim, D. Y.; Grey, J. K.; Barbara, P. F. ChemPhysChem 2005, 6, 2404. (40) Lee, Y. J.; Kim, D. Y.; Barbara, P. F. J. Phys. Chem. B 2006, 110, 9739. (41) Mirzov, O.; Scheblykin, I. G. Phys. Chem. Chem. Phys. 2006, 8, 5569. (42) Feist, F. A.; Tommaseo, G.; Basche´, Th. Phys. ReV. Lett. 2007, 98, 208301. (43) Wachsmann-Hogiu, S.; Peteanu, L. A.; Liu, L. A.; Yaron, D. J.; Wildeman, J. J. Phys. Chem. B 2003, 107, 5133. (44) Pauck, T.; Ba¨ssler, H.; Grimme, J.; Scherf, U.; Mu¨llen, K. Chem. Phys. 1996, 210, 219.
Feist and Basche´ (45) Christ, T.; Kulzer, F.; Weil, T.; Mu¨llen, K.; Basche´, Th. Chem. Phys. Lett. 2003, 372, 878. (46) Kim, D. Y.; Grey, J. K.; Barbara, P. F. Synth. Met. 2006, 156, 336. (47) Friedrich, J.; Haarer, D. Angew. Chem., Int. Ed. 1984, 23, 113. (48) Moerner, W. E. Ed. Persistent Spectral Hole-Burning: Science and Applications; Springer-Verlag: Berlin, 1988. (49) Hildner, R.; Lemmer, U.; Scherf, U.; van Heel, M.; Ko¨hler, J. AdV. Mater. 2007, 19, 1978. (50) Meier, H.; Stalmach, U.; Kolshorn, H. Acta Polym. 1997, 48, 379. (51) Grey, J. K.; Kim, D. Y.; Norris, B. C.; Miller, W. L.; Barbara, P. F. J. Phys. Chem. B 2006, 110, 25568. (52) Grey, J. K.; Kim, D. Y.; Donley, C. L.; Miller, W. L.; Kim, J. S.; Silva, C.; Friend, R. H.; Barbara, P. F. J. Phys. Chem. B 2006, 110, 18898. (53) Renge, I. J. Opt. Soc. Am. B 1992, 9, 719. (54) Renge, I. J. Chem. Phys. 1997, 106, 5835. (55) Huang, Y. F.; Shiu, Y. J.; Hsu, J. H.; Lin, S. H.; Su, A. C.; Peng, K. Y.; Chen, S. A.; Fann, W. S. J. Phys. Chem. C 2007, 111, 5533. (56) Yu, J.; Hu, D.; Barbara, P. F. Science 2000, 289, 1327. (57) Basche´, Th., Moerner, W. E., Orrit, M., Wild, U. P. Eds. Single Molecule Optical Detection, Imaging and Spectroscopy; VCH: Weinheim, Germany, 1997. (58) Me´tivier, R.; Nolde, F.; Müllen, K.; Basche´, Th. Phys. ReV. Lett. 2007, 98, 047802.
JP802585M
