Single-Particle Spectroscopy on Conducting Polymer-Fullerene Composite Materials for Application in Organic Photovoltaic Devices

The study of the photophysical and optoelectronic properties of a functioning conducting polymer device is extremely complicated and is hampered by the complex nanostructure and overall morphology of the conducting polymer materials applied in these devices. Here we discuss a novel approach to investigate this issue spectroscopically in terms of bulk-heterojunction organic photovoltaic devices. Novel composite nanoparticles of the conjugated polymers MEH-PPV and P3HT blended with the fullerene PCBM were fabricated and are observed to be excellent simplified model systems for the study of molecular-scale morphology effects at play in these complex nanostructured materials. Single-particle spectroscopy reveals the extent to which variations in polymer-chain folding and interactions between polymer chains and fullerenes affect material morphology, spectral properties, and optoelectronic properties, providing a detailed molecular scale insight into the morphological effects at play in the active layers of bulk-heterojunction organic photovoltaic devices that otherwise would be masked by the presence of the bulk.

Conducting polymers have been at the forefront of materials research on novel semiconductors given the exciting prospect of plastic electronics that can be built from solution-processed materials with low-cost manufacturing processes. The photophysical and optoelectronic properties of these materials have been investigated extensively for potential application in organic light-emitting diodes (OLED) (1,2), which for conducting polymer devices also are referred to as polymer light-emitting diodes (PLED), organic field effect transistors (OFET) (3–5), and organic photovoltaic (OPV) devices (Figure 1) (6–9). Despite intense and sustained research on conducting polymer materials and devices, there are no products slated for commercialization within the next few years that are based upon conducting polymers, in contrast to small-molecule organics that have been commercialized successfully in organic flat-panel displays. Unfortunately, the study of the photophysical and optoelectronic properties of a functioning conjugated polymer device is extremely complicated. Several factors contribute to difficulties with developing a detailed understanding of conducting polymer materials and devices. First, in a functioning device, there are a number of different excitations that exist at the same time, such as singlet excitons, triplet excitons, and polarons. These all interact with each other and are distributed heterogeneously throughout the device. Second, conducting polymer materials are enormously heterogeneous. This is complicated further by the fact that conducting polymers can fold into complex nanostructured particles for which the different resulting morphologies can lead to varying photophysical and optoelectronic properties. This was demonstrated clearly by characterization of MEH-PPV/C60 OPVs in relation to the morphology of the active layer (10). Third, there are particular issues related to charge transport through conducting polymers. Deep charge trapping in itself leads to unexpected observations when characterizing conducting polymer devices and has been cumbersome, particularly in terms of developing suitable device models (11,12). The result of these issues is that device behavior and performance incorporating these materials is unpredictable at best. In particular, the low photostability of conducting polymers and the ease with which they participate in electrochemical processes lies at the basis of the latter issue.

Figure 1

Previously, we have investigated these issues by means of novel single-molecule spectroscopic techniques fluorescence-voltage single-molecule spectroscopy (FV/SMS) and fluorescence-voltage time-resolved single-molecule spectroscopy (FV-TR-SMS) (13,14). The application of SMS to current problems in integrating conducting polymer as semiconductors in plastic electronics is motivated by the fact that complex chemical and physical events in heterogeneous systems often are hidden in bulk ensemble experiments or only yield an averaged value. By simplifying the complex bulk system to single molecules and nanoparticles, the complications presented by the presence of bulk material can be avoided, and prohibitively difficult kinetic and mechanistic problems at the bulk scale can be simplified to manageable proportions at the single molecule and nanoparticle level (15–20). As such, SMS presents a unique opportunity to investigate the morphological and photophysical properties of nanostructured conducting polymer materials at the molecular level, at which excited state interactions and energy- and charge-transfer processes can be investigated and understood in great detail. With FV/SMS, in which fluorescence intensity of single molecules is correlated with applied bias on an OLED-type device, conducting polymer molecules embedded in functioning devices were investigated (13,21,22). These studies revealed that conjugated polymer photooxidation can proceed through a reversible process caused by a single electron transfer event (that is, reversible oxidation of the conjugated polymer chain). Oxygen was shown to receive an electron from the conducting polymer MEH-PPV (poly[2-methoxy-5-(2-ethylhexyl-oxy)-p-phenylenevinylene]), forming a MEH-PPV⁺ –O₂^– complex. While in this state, the MEH-PPV molecule could be reversibly reduced and reoxidized with an applied bias to the device. In addition, the presence of O₂^– was found to create a deep charge trap state in MEH-PPV, which reoxidized spontaneously to the cation at zero applied bias. FV-TR-SMS experiments on these devices with embedded single-conducting polymer molecules further helped elucidate the photophysical and optoelectronic processes at play in conducting polymer molecules. In FV-TR-SMS, a laser light source coupled into a single-molecule detection confocal fluorescence microscope is pulsed while a varying bias voltage is synchronously stepped with each laser excitation window that is applied to the single conducting polymer molecule. The result is that the presence of singlet excitons, triplet excitons, and hole polarons on the conducting polymer chain can be detected simultaneously. Singlet–triplet energy transfer, singlet–hole energy transfer, and triplet–hole electron exchange interactions could be quantified by this method and were found to be highly efficient. These experiments illustrate the complexity associated with the presence of multiple excited-state and polaronic species in a conducting polymer device. More recently, single-molecule spectroelectrochemistry (SMS-EC) was developed by Palacios and colleagues (23). SMS-EC enables the study of charging and discharging of individual conjugated polymer nanoparticles. It was found that the hole-injection into the nanoparticle is reversible (that is, the charges are not trapped in the nanoparticle). After several cycles of charging and discharging, some charges could no longer be removed easily from the nanoparticles, and this result was attributed to deep charge trapping.

From an application point of view, PLEDs have been surpassed by small-molecule OLEDs for display applications because of the issues described previously. Now, however, there is a strong interest in conducting polymer OPVs for solar energy conversion applications (6–9). One of the reasons these polymers might succeed in this field is their electrical conductivity. While for an OLED (electricity to light conversion) the active layer can be practically an insulator and still work as an OLED, the function of OPVs (light to electricity conversion) relies heavily upon efficient charge transport to the electrodes, which is determined mainly by the conductivity of the active layer in the device. In fact, significant research effort has been focused on developing bulk-heterojunction photovoltaic (PV) cells utilizing conducting polymer–fullerene blends (Figure 1) due to advantages such as flexibility of the devices and ease of fabrication at low cost (7,24). The function of organic photovoltaic devices is based upon charge separation from an excited state across an interface, followed by charge transport to the electrodes. Photoinduced electron transfer from a light-absorbing donor (conducting polymer) to an acceptor (fullerene) is studied widely as a pathway for inducing charge separation from an optically excited state.

Despite the efficiency of charge separation being close to unity in this type of PV devices, the device energy conversion efficiency is too low for practical applications. While an interpenetrating network of electron donor and acceptor materials provides a large interfacial area for efficient photoinduced charge separation, one of the most severely limiting factors in these PV devices is the resulting complex nanostructured morphology of the active layer that is created by mixing of the light-absorbing electron donor (conducting polymer) and acceptor (fullerene) molecules. As a result, low carrier mobility and inefficient carrier transport typically are observed.

Single-molecule and nanoparticle spectroscopy provide access to quantitative morphological, photophysical, and kinetic parameters that are not readily accessible for complex bulk systems. These methods are applied to investigate the properties of such nanostructured materials by the study of single nanoparticles. The composite conducting polymer–fullerene nanoparticles developed in this contribution are a novel material system that represents structures in between single molecule and bulk that enable us to study the effects of varying material morphologies and heterogeneity with respect to photophysical and optoelectronic properties at the nanoscale. The nanoparticles allow for a detailed investigation of molecular conformations and interactions as they are expected to appear in the bulk, where energy and charge transport, exciton dissociation, and charge injection are subject to molecular-scale interactions. Characterization of the nanoparticles with solution- and single-particle spectroscopy reveals the extent to which variations in polymer chain folding and interactions between polymer chains and fullerenes affect material morphology and photophysical properties, providing a detailed molecular-scale insight into the morphological effects that are at play in nanostructured bulk materials for plastic solar energy conversion devices.

Experimental

Materials. MEH-PPV (poly[2-methoxy-5-(2-ethylhexyl-oxy)-p-phenylenevinylene]), P3HT (poly-3-hexylthiophene), PCBM (1-[3-methoxycarbonylpropyl]-1-phenyl-[6.6]C₆₁), and tetrahydrofuran (anhydrous, 99.9%) were purchased from Sigma Aldrich (St. Louis, Missouri). MEH-PPV was purified by hot acetone wash before use and vacuum dried for 24 h. P3HT and PCBM were used as received.

Nanoparticle preparation. Nanoparticles were prepared using the reprecipitation method. The materials were dissolved in a good solvent (tetrahydrofuran) that is miscible with water. The tetrahydrofuran solutions (1 mL) were rapidly injected into 4 mL of water, leading to a rapid aggregation of the materials in the aqueous environment. Composite nanoparticles were prepared by starting from mixed molecular solutions of conducting polymer and PCBM in tetrahydrofuran. The amount of PCBM needed to achieve 5 and 50 wt % of doping level in the doped nanoparticles was calculated by using the concentration of MEH-PPV in tetrahydrofuran, which was obtained from the absorption spectra. As soon as the amount (in grams) of the MEH-PPV was determined in the solution, the amount (in grams) of PCBM for each doping level was calculated.

Nanoparticle Characterization

Dynamic Light Scattering. The undoped and doped nanoparticle suspensions were diluted by placing a drop of sample into a 1-cm quartz cuvette using a glass pipette, and the remainder of the cuvette was filled with deionized water. A total of 30 trials were run (each trial contains 30 runs) using a dynamic light scattering system, model Precision Detector 2000 DLS (Bellingham, Massachusetts). These values were averaged and standard deviation was calculated.

Transmission Electron Microscopy. The size of the undoped and doped nanoparticles was determined using transmission electron microscopy (TEM) with a JEOL (Peabody, Massachusetts) model 1011 instrument. A 3-μL volume of nanoparticle suspension was drop cast onto the TEM grid and vacuum dried for 15 min.

Atomic Force Microscopy. Atomic force microscopy was run on a Veeco Multimode system with NanoscopeIIIa controller (Santa Barbara, California). A 3-μL volume of undoped and doped nanoparticle suspensions in water was dropped onto freshly cleaved mica substrates. These samples were vacuum dried for 15 min before imaging.

Bulk Solution Spectroscopy

UV–vis. The UV–vis absorption spectra were taken using a 1-cm path length quartz cuvette with a Varian (Palo Alto, California) Cary 300 Bio UV-vis scanning spectrometer.

Fluorescence. Fluorescence emission spectra were taken using a 1-cm pathlength quartz cuvette with a Nanolog Horiba Jobin Yvon (Edison, New Jersey) fluorometer. The excitation wavelength was set at 488 nm. The excitation and emission slits were 2 nm.

Single-particle imaging and spectroscopy. Samples were prepared by spincoating a solution of nanoparticles in 4% w/w polyvinylalcohol (Sigma Aldrich) in water at 2000 rpm for 60 s on glass cover slides. The samples were sealed from oxygen by thermal evaporation of 200 nm of aluminum to overcoat the samples. Single-particle fluorescence images were acquired using a home-built sample-scanning confocal microscope (Zeiss Axiovert 200, Peabody, Massachusetts) (Figure 2). The sample was raster scanned using a Mad City Labs (Madison, Wisconsin) piezoelectric stage (Nano-LP100). The 488-nm line of a Melles Griot (Carlsbad, California) 43 series argon-ion laser was used as the excitation source and was sent into the back port of the microscope. The laser was focused to a spot size of ~300 nm using a Zeiss 100x Fluar objective lens (NA 1.3, WD 0.17 mm). The single-particle fluorescence was detected using a PerkinElmer (Shelton, Connecticut) SPCM-AQR-14 avalanche photodiode. Images were scanned over a 10–20 μm range with dwell times of 10 ms over 100 pixels. The single-particle fluorescence spectra were obtained by diverting the signal through a spectrograph with the grating (150 g/mm blaze: 500 nm) centered at 600 nm (PI Acton SP-2156, Acton, Massachusetts), which was coupled to an electronically cooled electron multiplying charge coupled device (EM-CCD, Andor iXon EM+ DU-897 BI, South Windsor, Connecticut). Each fluorescence spectrum was collected with 10-s exposure times with three consecutive exposures and then averaged. A total of at least 100 spectra were acquired for each nanoparticle sample. Laser powers of 0.7 W/cm² and 7 W/cm² were used for imaging and collecting spectra for undoped and doped nanoparticles, respectively.

Figure 2

Results and Discussion

To relate material composition and morphology with photophysical and optoelectronic properties and develop structure-function relationships for conducting polymer materials applied in OPVs as the active layer, we devised the fabrication of novel composite nanoparticles consisting of MEH-PPV and PCBM, as well as of P3HT and PCBM. These materials are used most commonly at this time in the OPV field. The reprecipitation method, first introduced by Nakanishi and colleagues (25), was modified to enable the fabrication of these novel composite nanoparticles. Figure 3 illustrates the fabrication process schematically. A molecularly dissolved solution of the conducting polymer in tetrahydrofuran is injected into water under continuous stirring. The miscibility of tetrahydrofuran and water causes the conducting polymer to be transferred into water, which is a nonsolvent for these materials. As a result, the conducting polymer molecules form nanoparticle aggregates consisting of ~100 molecules. When starting from a mixed conducting polymer-fullerene solution in tetrahydrofuran, novel composite nanoparticles are fabricated.

Figure 3

With this method, MEH-PPV and P3HT nanoparticles doped with 0 wt%, 5 wt%, 10 wt%, and 50 wt% PCBM were prepared and characterized for size, morphology, and photophysical properties. Dynamic light scattering experiments show that the MEH-PPV nanoparticles have a size of approximately 30 nm that is only minimally affected by doping with PCBM. However, for P3HT, the nanoparticle size is found to increase significantly with PCBM doping. Undoped P3HT nanoparticles have a similar size to those of MEH-PPV, but the 50 wt% PCBM-doped P3HT nanoparticles have a size of approximately 80 nm, indicative of a stronger tendency of PCBM to aggregate with itself in the presence of P3HT compared to MEH-PPV. TEM data depicted in Figure 4 show well-separated individual nanoparticles on the TEM grid. The electron-dense nature of the organics in the nanoparticles ensures that a reasonable contrast can be obtained during imaging. The nanoparticle size measured from TEM data corroborates the findings from dynamic light scattering experiments. The particles have a nearly spherical shape, and although not monodisperse, appear to have a reasonable size distribution.

Figure 4

The optical properties of the nanoparticles were studied by bulk solution UV-vis absorption and fluorescence spectroscopy, and by single-particle imaging and spectroscopy. Absorption spectra of MEH-PPV nanoparticles in water are shifted slightly compared to the absorption spectra of the molecular solutions in tetrahydrofuran (Figure 5a). A unique observation is that for different samples prepared at different times, the nanoparticle absorption spectra vary between blue shifting and red shifting absorbance maxima compared to the molecular solutions in tetrahydrofuran, indicative of variations in single polymer chain morphology upon aggregation into nanoparticles between the different samples. The blue-shifted UV-vis spectra have maxima ranging from 492 to 498 nm and are attributed to a polymer chain conformation with kinking and bending of the chain, leading to a blue shift due to the reduced conjugation length (26,27). The red-shifted UV-vis spectra have maxima ranging from 502 to 515 nm that previously have been assigned to intra and interchain interactions (26,27). This new finding is illustrative of how the single polymer chain conformation is frozen upon assembly into the material: a quasi-instantaneous collapse of the polymer chain onto itself results in either a polymer chain conformation with kinking and bending of the chain, leading to a blue shift due to the reduced conjugation length, or otherwise a polymer chain conformation with extended straight conjugated segments that allow for pi-stacking interchain contacts and thus show red-shifted spectroscopy (28,29). In comparison, the P3HT absorption spectra (Figure 5b) consistently red-shift by 50 nm compared to P3HT solutions in tetrahydrofuran, which is attributed to increased interchain interactions in the densely compacted nanoparticles. In addition, for the nanoparticle suspensions, a shoulder in the absorption spectrum at 600 nm is apparent, which is assigned to interplane interactions (30,31). Neither the MEH-PPV nor the P3HT samples showed obvious variations in absorption spectra for different PCBM doping levels outside of the appearance of the PCBM absorption peak at 330 nm. As such, there is no apparent ground state interaction between conducting polymer and fullerene in the composite nanoparticles.

Figure 5

As shown in Figure 5c, both the MEH-PPV and composite MEH-PPV/PCBM nanoparticles with red-shifted and blue-shifted absorbance show red-shifted fluorescence spectra. The emission maximum is at 593 nm, representing a red-shift by an additional 13 nm compared to single-molecule spectra of the defect cylinder conformation found in polymer host matrices, and is red-shifted by about 30 nm compared to the fluorescence emission maximum of MEH-PPV in tetrahydrofuran solutions (29). The nanoparticle emission maximum is comparable to values reported for spectra collected on bulk films (32). The red-shift of the fluorescence spectra for both types of particles is explained by what was previously designated as red sites in the folded MEH-PPV chains, which are accessed by fast energy transfer from higher energy (blue) chromophores to a limited number of lower energy red sites (29). Fluorescence spectra of solutions of MEH-PPV nanoparticles doped with 50 wt% PCBM have the same shape and emission maxima as undoped nanoparticles in the bulk measurements. The difference lies in the intensity, which is reduced by 90% compared to undoped nanoparticles due to excited state charge transfer and energy transfer from MEH-PPV to PCBM. Figure 5d shows bulk solution fluorescence spectra of P3HT solutions in tetrahydrofuran, and undoped as well as 50 wt% PCBM doped nanoparticle suspensions in water. The P3HT emission maximum is located at 570 nm in tetrahydrofuran, compared to which the nanoparticles are red-shifted by 80 nm. The emission maxima of P3HT films typically are located at 660 nm, indicating that the P3HT chains are not stacking ideally with each other in the confined nanoparticle structure. In addition, the fluorescence intensity of the undoped nanoparticle suspension is lower than expected from the large number of chromophores present in these nanoparticles, while the fluorescence intensity of the 50 wt% PCBM doped composite nanoparticles is reduced by a factor of two compared to the undoped nanoparticles. It is well known that P3HT will self-quench due to the formation of interchain excitations (33,34). This explains why the nanoparticle intensity is relatively low. In addition, excited state charge transfer and energy transfer from P3HT contributes further to fluorescence quenching of the composite nanoparticles. It deserves mention that excited state interactions between conducting polymers and fullerenes in our composite nanoparticle systems can occur via charge transfer or Dexter energy transfer. It has been shown in literature through various electron spin resonance experiments that signal from electrons in charge separated states can be detected (35). Thus, although energy transfer between conducting polymer and fullerene is a feasible competing mechanism, charge transfer has been proven experimentally to be the dominant process.

The fluorescence properties of the undoped and composite nanoparticles were studied in further detail with single-particle imaging and spectroscopy. Single-particle fluorescence images of composite MEH-PPV/PCBM and P3HT/PCBM nanoparticles are shown in Figure 6. The nanoparticles can be imaged at very low excitation energies because of the large number of chromophores per nanoparticle. The detected fluorescence intensity of the composite nanoparticles is substantially lower than that of the undoped polymer nanoparticles, and in fact requires a 10-fold increase in laser excitation intensity to achieve comparable emission count rates. As discussed previously, this is related to a combination of charge transfer and energy transfer processes that nonradiatively deactivate the conducting polymer excited state.

Figure 6

The strength of single-particle spectroscopy is that spectral differences and features that are masked by the bulk can be revealed. Spectral differences between the composite and undoped nanoparticles that were undetected in the ensemble solution measurements were revealed by single-particle spectroscopy for which data are shown in Figure 7 and Figure 8 for MEH-PPV and P3HT, respectively. From the data for MEH-PPV (Figure 7), there are two key observations to be made. First, the emission maxima of the single-particle ensembles gradually blue-shift with increasing doping level of PCBM. The spectra of the undoped nanoparticles show maxima at 594 nm, consistent with the ensemble solution studies. The PCBM-doped nanoparticles show maxima at 594 and 585 nm for 5 wt% and 50 wt% doped composite MEH-PPV/PCBM nanoparticles, respectively. In addition, a blue shoulder at 540 nm is present in the heavily doped (50 wt%) composite nanoparticle spectra. These data suggest that PCBM molecules interacting with and dispersing in between the conjugated polymer chains interrupt intra- and interchain interactions, leading to blue-shifted emission. Furthermore, a reduced conjugation length might exist as a result of steric effects caused by the presence of PCBM. Exciton transport also might be limited due to compartmentalization of the polymer chain segments, preventing the exciton from reaching and emitting from low-energy sites. Second, the 0-1 vibronic peak in the single-particle ensembles is gradually suppressed with increasing doping level of PCBM. This observation further supports the notion that increasing doping levels of PCBM cause the MEH-PPV chains to become isolated while also significantly reducing exciton transport (36).

Figure 7

Results of single-particle spectroscopy on P3HT composite and undoped nanoparticles are summarized in Figure 8. The single-particle ensemble spectra appear to be relatively unaffected by increasing doping levels of PCBM. However, through single-particle spectroscopy, it is not only possible to obtain the ensemble of single-particle spectra, but also their distribution. The inset in Figure 8 shows how the relative distribution of single-particle emission maxima changes with varying doping levels of PCBM. As the PCBM doping of P3HT nanoparticles is increased from 0 wt% to 10 wt% and 50 wt%, the occurrence of spectra with emission maxima at 660 nm decreases, while the opposite is observed at 720 nm. The single nanoparticle spectra were sorted in two subensembles according to the observed emission maxima at 660 nm and 720 nm. The top panel of the inset in Figure 8 shows the resulting subensemble spectra and reveals that there are two types of nanoparticles present in the samples with two discrete emission maxima. The difference between the spectra lies in the relative intensity of the vibronic peaks and again indicates how the morphology of the conducting polymer chains and the composite material vary with PCBM doping level, resulting in variations in conjugation length, intra- and intermolecular pi-stacking interactions, and exciton transport.

Figure 8

Figure 9 illustrates the changes in molecular conformations and intermolecular interactions within the conducting polymer nanoparticles with varying PCBM doping levels. For MEH-PPV (left column) extended chains with long conjugated segments (blue), defect cylinder coils with long conjugated segments that fold back onto themselves to form pi-stacked red sites (red), and globules (blue) in which the polymer chains have a reduced conjugation length are expected to be present in the nanoparticles. A significant presence of globules will lead to blue-shifted absorption spectra, an observation that applies to the various PCBM doping levels in MEH-PPV nanoparticles (Figure 5a). At 0 wt% PCBM doping, intra- and interchain interactions are abundant and will lead to red-shifted emission compared to MEH-PPV in THF solution (Figure 7). At 5 wt% PCBM doping, the PCBM molecules can intersperse in between pi-stacked chains, but the doping level is too low to affect the nanoparticle fluorescence spectra significantly. However, with 50 wt% PCBM doping, intra- and interchain interactions are interrupted severely, leading to more disordered chain morphologies with blue-shifted fluorescence spectra due to reduced exciton transport to low energy sites. The vibronic structure of the spectra for these nanoparticles confirms that polymer chains have limited interactions between them. For P3HT (Figure 9, right column), the presence of two types of nanoparticles as evident from the emission spectra (Figure 8) stems from variations in interaction between polymer chains. A combination of emission from more extended lamellar-type structures (blue) and aggregated chains (red) is observed. In addition, both structures lead to red-shifted absorbance spectra (Figure 5b). With increasing PCBM doping, the P3HT chains experience a poorer solvent environment, leading to the formation of folded chains. In addition, the possibilities for interchain interactions that force the polymer chains into planar extended conformations are reduced substantially. Therefore, the abundance of nanoparticles that contain a large fraction of collapsed P3HT chains that are folded onto themselves increases with increasing PCBM doping levels, which is observed as an increase in the number of nanoparticles that have their fluorescence emission maxima at 720 nm.

Figure 9

The morphological variations detected in our experiments are at play at the molecular scale, and performance of devices that use these materials as active layers will thus be affected strongly by molecular and nanoscale morphological, photophysical, and optoelectronic processes. It is therefore paramount to investigate such issues at the molecular- and nanoscale, for which the single-molecule and single-nanoparticle spectroscopic methods we have developed are valuable tools, especially in combination with our novel composite nanoparticle systems that have been shown through this work to be excellent model systems for the study of the relation between morphological and photophysical properties of materials that can be used as active layers in organic electronic devices.

Conclusion

A novel approach to investigate material issues related to the active layer of bulk heterojunction organic photovoltaic devices spectroscopically was discussed. The composite conducting polymer–fullerene nanoparticles developed in this work are a novel material system that represents structures in between single molecule and bulk that allow us to study the effects of varying material morphologies and heterogeneity with respect to photophysical and optoelectronic properties at the nanoscale. Single-particle spectroscopy reveals the extent to which variations in polymer chain folding and interactions between polymer chains and fullerenes affect material morphology, spectral, and optoelectronic properties, providing a detailed molecular-scale insight into the morphological effects at play in the active layers of bulk heterojunction organic photovoltaic devices that otherwise would be masked by the presence of the bulk.

Acknowledgment

The authors wish to thank the National Science Foundation (NSF) for financial support of this work through a Nanoscale Exploratory Research (NER, BES-0608870) grant and a CAREER award (CBET-0746210).

Andre J. Gesquiere, Daeri Tenery, and Zhongjian Hu are with the NanoScience Technology Center and Department of Chemistry, University of Central Florida, Orlando, Florida.

References

(1) M. Gross, D.C. Muller, H.G. Nothofer, U. Scherf, D. Neher, C. Brauchle, and K. Meerholz, Nature 405, 661–665 (2000).

(2) G. Gustafsson, Y. Cao, G.M. Treacy, F. Klavetter, N. Colaneri, and A.J. Heeger, Nature 357, 477–479 (1992).

(3) C.D. Dimitrakopoulos and P.R.L. Malenfant, Adv. Mater. 14, 99–117 (2002).

(4) G. Horowitz, Adv. Mater. 10, 365–377 (1998).

(5) H. Sirringhaus, N. Tessler, and R.H. Friend, Science 280, 1741–1744 (1998).

(6) S.E. Shaheen, C.J. Brabec, N.S. Sariciftci, F. Padinger, T. Fromherz, and J.C. Hummelen, Appl. Phys. Lett . 78, 841–843 (2001).

(7) G. Li, V. Shrotriya, J.S. Huang, Y. Yao, T. Moriarty, K. Emery, and Y. Yang, Nat. Mater. 4, 864–868 (2005).

(8) J.K.J. van Duren, X.N. Yang, J. Loos, C.W.T. Bulle-Lieuwma, A.B. Sieval, J.C. Hummelen, and R.A.J. Janssen, Adv. Funct. Mater. 14, 425–434 (2004).

(9) H. Hoppe and N.S. Sariciftci, J. Mater. Res. 19, 1924–1945 (2004).

(10) J. Davenas, P. Alcouffe, A. Ltaief, and A. Bouazizi, Macromol. Symp. 233, 203–209 (2006).

(11) P.H. Nguyen, S. Scheinert, S. Berleb, W. Brutting, and G. Paasch, Org. Electron. 2, 105–120 (2001).

(12) S. Scheinert, G. Paasch, and T. Doll, Synth. Met. 139, 233–237 (2003).

(13) A.J. Gesquiere, S.J. Park, and P.F. Barbara, J. Phys. Chem. B 108, 10301–10308 (2004).

(14) A.J. Gesquiere, S.J. Park, and P.F. Barbara, J. Am. Chem. Soc. 127, 9556–9560 (2005).

(15) W.E. Moerner and M. Orrit, Science283, 1670–1676 (1999).

(16) W.P. Ambrose, P.M. Goodwin, J.H. Jett, A. Van Orden, J.H. Werner, and R.A. Keller, Chem. Rev. 99, 2929–2956 (1999).

(17) X.S. Xie and J.K. Trautman, Annu. Rev. Phys. Chem. 49, 441–480 (1998).

(18) S. Nie and R. Zare, Annu. Rev. Biophys. Biomol. Struct. 26, 567–596 (1997).

(19) M. Orrit, J. Bernard, and R.I. Personov, J. Phys. Chem. 97, 10256–10268 (1993).

(20) P.F. Barbara, A.J. Gesquiere, S.J. Park, and Y.J. Lee, Acc. Chem. Res. 38, 602–610 (2005).

(21) A.J. Gesquiere, S.J. Park, and P.F. Barbara, Eur. Polym. J. 40, 1013–1018 (2004).

(22) Y.J. Lee, S.J. Park, A.J. Gesquiere, and P.F. Barbara, Appl. Phys. Lett. 87, 051906 (2005).

(23) R.E. Palacios, F.R.F. Fan, J.K. Grey, J. Suk, A.J. Bard, and P.F. Barbara, Nat. Mater. 6, 680–685 (2007).

(24) C.J. Brabec, N.S. Sariciftci, and J.C. Hummelen, Adv. Funct. Mater. 11, 15–26 (2001).

(25) H. Kasai, H.S. Nalwa, H. Oikawa, S. Okada, H. Matsuda, N. Minami, A. Kakuta, K. Ono, A. Mukoh, and H. Nakanishi, Jpn. J. Appl. Phys., Part 2 31, L1132–L1134 (1992).

(26) B.J. Schwartz, Annu. Rev. Phys. Chem. 54, 141–172 (2003).

(27) R. Traiphol, P. Sanguansat, T. Srikhirin, T. Kerdcharoen, and T. Osotchan, Macromolecules 39, 1165–1172 (2006).

(28) T. Huser, M. Yan, and L.J. Rothberg, Proc. Natl. Acad. Sci. U.S.A. 97, 11187–11191 (2000).

(29) D.H. Hu, J. Yu, K. Wong, B. Bagchi, P.J. Rossky, and P.F. Barbara, Nature 405, 1030–1033 (2000).

(30) H. Kim, W.W. So, and S.J. Moon, J. Korean Phys. Soc. 48, 441–445 (2006).

(31) Y. Kim, S. Cook, S.M. Tuladhar, S.A. Choulis, J. Nelson, J.R. Durrant, D.D.C. Bradley, M. Giles, I. McCulloch, C.S. Ha, and M. Ree, Nat. Mater. 5, 197–203 (2006).

(32) S.A. Amautov, E.M. Nechvolodova, A.A. Bakulin, S.G. Elizarov, A. Khodarev, D.S. Martyanov, and D.Y. Paraschuk, Synth. Met. 147, 287–291 (2004).

(33) M. Theander, O. Inganas, W. Mammo, T. Olinga, M. Svensson, and M.R. Andersson, J. Phys. Chem. B 103, 7771–7780 (1999).

(34) X. Bai and S. Holdcroft, Macromolecules 26, 4457–4460 (1993).

(35) V. Dyakonov, G. Zoriniants, M. Scharber, C.J. Brabec, R.A.J. Janssen, J.C. Hummelen, and N.S. Sariciftci, Phys. Rev. B: Solid State59, 8019–8025 (1999).

(36) S.Y. Quan, F. Teng, Z. Xu, L. Quan, T. Zhang, D. Liu, Y.B. Hou, Y.S. Wang, and X.R. Xu, J. Lumin. 124, 81–84 (2007).