Letter pubs.acs.org/NanoLett
Remarkable Order of a High-Performance Polymer Christopher J. Takacs,† Neil D. Treat,∥,⊥ Stephan Kram ̈ er,§ Zhihua Chen,‡ Antonio Facchetti,‡ ∥ ,† Michael L. Chabinyc, and Alan J. Heeger* †
Department of Physics, Broida Hall, University of California, Santa Barbara, Santa Barbara, California 93106, United States Polyera Corporation, Skokie, Illinois 60077, United States § Materials Department, University of California Santa Barbara, Santa Barbara, California 93106, United States ∥ Materials Research Laboratory, Materials Department, Mitsubishi Chemical Center for Advanced Materials, University of California Santa Barbara, Santa Barbara, California 93106, United States ‡
S Supporting Information *
ABSTRACT: We directly image the rich nanoscale organization of the high performance, n-type polymer poly{[N,N′bis(2-octyldodecyl)-naphthalene-1,4,5,8-bis(dicarboximide)2,6-diyl]-alt-5,5′-(2,2′-bithiophene)} (P(NDI2OD-T2)) using a combination of high-resolution transmission electron microscopy and scanning transmission electron microscopy. We demonstrate that it is possible to spatially resolve “face-on” lamella through the 2.4 nm alkyl stacking distance corresponding to the (100) reflection. The lamella locally transition between ordered and disordered states over a length scale on the order of 10 nm; however, the polymer backbones retain long-range correlations over length-scales approaching a micrometer. Moreover, we frequently observe overlapping structure implying a number of layers may exist throughout the thickness of the film (∼20 nm). The results provide a simple picture, a highly ordered lamella nanostructure over nearly the entire film and ordered domains with overlapping layers providing additional interconnectivity, which unifies prior seemingly contradictory conclusions surrounding this remarkable, high-mobility material. KEYWORDS: Polymer morphology, TEM, long-range order, high-mobility polymer material
C
Studies of the pure thin-film morphology of P(NDI2OD-T2) have revealed unusual and striking behavior. While initially thought to be amorphous, using a combination of grazing incidence wide-angle X-ray scattering (GIWAXS) and atomic force microscopy (AFM), Rivnay et al. found a remarkable degree of molecular order.10 Moreover, crystalline domains adopted an unconventional “face-on” texture where the πconjugated backbone is parallel to the substrate when the films were spin-coated. This orientation corresponds to the alkylstacking peak being predominantly in-plane and the π−π stacking in the out-of-plane direction. In conjunction with AFM data, they postulated that the material was composed of platelet-like crystals with local alignment of the chain backbones and a high degree of interconnectivity. Film morphologies with “edge-on” polymer chains were observed after thermal annealing above the melting temperature (∼300 °C)11 or, more recently, for Langmuir−Shäfer deposited multilayers.12
onjugated polymers have generated intense academic and commercial interest. Their remarkable electronic properties and compatibility with large-scale processing methods have opened new possibilities for low-cost and flexible electronics. Because of the development of new materials and processing methodologies, the device performance of organic field-effect transistors (OTFTs) and organic photovoltaics (OPVs) continues to rise. Recently, the mobility of p-type polymeric OTFTs has exceeded 10 cm2 V−1 s−1, a major milestone1and the efficiencies of OPVs are now nearing 10% power conversion efficiency (PCE).2,3 With these rapid improvements in the performance, fundamental studies of the electronic properties and morphology are essential steps needed to direct the next generation of materials. Yan et al. recently introduced poly{[N,N′-bis(2-octyldodecyl)-naphthalene-1,4,5,8-bis(dicarboximide)-2,6-diyl]-alt-5,5′(2,2′-bithiophene)} (P(NDI2OD-T2)), an n-type polymer with a mobility approaching 1 cm2 V−1 s−1 (Figure 1).4 It has shown promise in all-polymer OPVs as a fullerene replacement reaching PCEs of 1.4% with P3HT5 and more recently, a PCE of 4.2%.6 From a fundamental perspective, charge transport measurements of its electron mobility show an unusually low degree of electronic disorder for a polymer semiconductor.7−9 © XXXX American Chemical Society
Received: February 13, 2013 Revised: April 29, 2013
A
dx.doi.org/10.1021/nl4005805 | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
Letter
damage and the conjugated backbone is expected to be able to quickly delocalize charge reducing the chance for damage. This adds an additional challenge in the study of continuous organic thin films; cross-linking of the side-chains can cause local warping of the sample structure during the exposure. This can severely degrade the recorded image and lead to an underestimation of the local degree of order.15 Sample degradation can be particularly problematic for HRTEM, where the detector, a CCD camera, integrates over several seconds, limiting imaging even before the chemical and structural changes are severe. With this in mind, STEM is used as a complementary imaging technique. The image in STEM is formed by focusing the probe beam and collecting scattered electrons as the beam is scanned across the sample. The dwell time of the beam ∼105 times shorter than the exposure time for HRTEM and much less than the time scale for mechanical drifting, wrinkling, and twisting of the film over the dimensions of probe beam. We outline the relevant issues here and point the interested reader to the excellent review by Martin et al. and the references within on the methods, challenges, and rewards of high-resolution polymer imaging.16 Figure 2 shows a STEM image of the P(NDI2OD-T2) thinfilm. Lattice planes corresponding to the alkyl stacking
Figure 1. Perspective view of a linear conformation of P(NDI2ODT2) with each monomer repeated in the same orientation and shown in a different color. The thiophene units are coplanar but are twisted relative to the NDI units due to steric interactions. The molecular structure appears in the inset.
Focusing instead on the organization of the polymer backbones, Sciascia et al. used polarized scanning transmission X-ray microscopy (STXM) to spatially resolve the order parameter.13 Using the orientation and chemical specific nature of the fine-structure near the carbon K-edge, the orientation of the polymer backbones can be studied at the nanoscale without the requirement of crystallinity. The polarized soft X-ray beam is focused and the sample is scanned to form an image pointby-point. The transmitted intensity is recorded for several inplane polarization vectors and used to construct the molecular director and spatially dependent order parameter with a resolution less than 90 nm. It is important to note that this represents a projection of the sample structure through the thickness of the film. While demonstrating impressive longrange correlations, they concluded the only a small fraction of the chains were actually aligned locally. In this report, scanning transmission electron microscopy (STEM) and high-resolution transmission electron microscopy (HRTEM) are used to examine the morphology of P(NDI2OD-T2) processed in the “face-on” orientation. The “face-on” crystallites give rise to a strong in-plane alkyl-stacking peak directly resolvable with TEM. These lattice planes are observed to cover almost the entire sample. While the nanostructure often fluctuates in the degree of local order, the polymer backbones show long-range orientational order on length-scales approaching one micrometer. These oriented regions are frequently observed to overlap suggesting a layered structure. The results represent an important step in documenting the rich and peculiar nature of nanoscale organization for this polymer. While STEM and HRTEM have comparable resolution, each has its own practical and conceptual advantages for image acquisition. STEM imaging of the lattice planes tends to have fewer drift/damage/charging issues but the signal-to-noise ratio is lower than with HRTEM. Electron microscopy in polymers is typically limited by electron beam damage.14 Electrons that scatter inelastically as they pass through the sample can directly eject light atoms or create high-energy excitations that cascade into many low-energy excitations with enough energy to do chemistry. The side-chains tend to be most sensitive to reactive
Figure 2. STEM image of P(NDI2OD-T2) thin-film. Lattice planes with d = 2.4 nm are discernible over almost the entire region of the film. The power spectrum of the image is shown in the insert. The scale bar is 50 nm.
direction are visible. This peak has previously been assigned as the (100) direction and has a d-spacing of ∼2.4 nm. It noteworthy that the lattice planes are significantly different than a simple crystal: the lattice planes are not long, parallel lines like those expected in a perfect crystal but are instead curved. This curvature is likely related to defects, twisting, dislocations, slip of the backbone.17 The computed power spectrum appears in the inset. Over this length-scale, the diffraction pattern is a pair of arcs, characteristic of locally oriented “crystallites.” The larger bright and dark regions appear to be a combination of diffraction and mass−thickness contrast. An STEM image of a B
dx.doi.org/10.1021/nl4005805 | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
Letter
scattering from an “edge-on” population will also appear inplane. This scattering should be observable if the imaging conditions can be improved. At the present, the data cannot yet give useful insight into the existence of a small population of “edge-on” oriented crystallites in these thin-films. The structure is often seen to be more complex than as represented in Figures 2 and 3. An example is shown in Figure 4. On the large scale, we observe overlapping lamella, implying a layered structure may exist within parts of the ∼20 nm thick film. The layered structure may have a beneficial role by improving the local interconnectivity and electrical transport properties. The existence of well-ordered lamellar structures that overlap is likely also important for the interpretation of projection based techniques like STXM. In the worst case, the projected order parameter of orthogonal overlapping lamella would be zero and the material would be interpreted as “amorphous”. Examples of such structure are apparent in Figure 4 and suggest the degree of order is higher than previously suggested by STXM measurements.13 On the nanoscale, the lattice planes are often observed to slowly meander or to disappear and then later reappear in nearly the same direction. The similarity of the observed structure with the STEM images suggests electron beam damage is not the cause of the disorder. The observed long-range correlations of the backbone coupled with the propensity to meander suggest the polymer chains are locally linear/rodlike but are prone to disorder. The transitions between ordered and disordered states are likely driven through defects, boundaries between polymorphs, and other structural defects.17 This structural disorder is expected to have important consequences on the electrical properties.19 Ignoring the flexible, insulating side-chains, it has been previously noted that P(NDI2OD-T2) has two energetically low-lying structural conformations giving a linear polymer backbone. These structures differed by the orientation of the thiophenes, which are twisted about the polymer long-axis and not coplanar with the NDI units due to steric interactions.18 This twist also slightly inclines the NDI units with respect to the polymer long-axis in the solid-state.20 Both of these conformations are formed by repetition of a single monomer as illustrated in Figure 1. The zigzag, which involves two monomers in the repeat unit, is another simple conformation that will result in a linear structure (Supporting Information). Of course, the self-assembly in the solid-state is complex. Even if the polymers chains are locally aligned, packing is likely to be complicated by the 3D nature of the polymer backbone and its reduced symmetry; a simple rotation or inversion about the molecular long-axis produces a nonequivalent structure. We postulate that the linear backbone favors local alignment and that the availability of energetically low-lying polymorphs, conformations that may be accessible without large changes in the backbone direction, allows additional local structural freedom to accommodate some packing frustration and improve the electronic properties. Visualizing the complex molecular order encoded in the (100) alkyl stacking peak, a d-spacing of ∼2.4 nm, on lengthscales greater than a few hundred nanometers requires additional analysis. To analyze this ordering, we start by spatially decomposing the amplitude of the (100) peak and then computing the molecular director averaged over a length scale of ∼50 nm. The calculated molecular director field for the region shown in Figure 4A is presented in Figure 4B. It represents a spatial mapping of the average direction of the polymer backbones and is based on a modified form of the
larger area is presented in the Supporting Information as a reference. We note that over larger length scales, these bright and dark regions bear a striking resemblance to the fibular structure observed with AFM.10,18 A possible complex relationship between the semicrystalline structure and the fibrils has been discussed by Schuettfort et al.18 Here, the resolution is sufficient to state that the “face-on” crystal population is in general correlated with the observed fibrils; however, the relationship is far from perfect and, as demonstrated below, the nanostructure is generally more complex. At higher magnification, additional reflections beyond the (100) peak may also be visible; however, the scattering is weaker and generating sufficient contrast becomes more difficult. Figure 3 shows a small region of a HRTEM image
Figure 3. HRTEM image of P(NDI2OD-T2) resolving both the (100) and (001) reflections corresponding to the alky stacking and backbone (i.e., along-the-chain) stacking, respectively. While the scattering of the (100) is locally a well-defined peak, we find the scattering along (001) is broadened suggesting local slippage of the chain backbones. We note the degree of order in the sample is likely higher than pictured as some evidence of electron damage is present. The scale bar is 20 nm.
with the computed power spectrum displayed in the inset. Over this small area, the alkyl peak is diffracted into two relatively narrow spots. A weaker reflection with a d-spacing of ∼1.4 nm appears distributed over a range of angles nearly orthogonal to the (100) peak. This distance has been previously assigned to (001), the reflection arising from periodic structure of the polymer backbone. The distribution of scattering angles relative to the sharp alkyl peak could suggest local slip of the polymer backbones (Supporting Information). The slip of the backbones is expected to change the relative angle between the (100) and (001) peaks in a manner similar to what is observed. This image was taken at a higher electron dosage where signs of electron beam damage were observed; however, this explanation is consistent with the already partially disordered nature of the sample and the presence of a relatively broad peak with d ∼ 1.4 nm observed in X-ray scattering experiments of similarly prepared thin-films.10 It is worth noting that the (001) C
dx.doi.org/10.1021/nl4005805 | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
Letter
Figure 4. (A) A region of complex, overlapping structure as imaged by HRTEM. Visible lattice planes (d = 2.4 nm) appear, disappear, and meander making straightforward identification of crystals and grain boundaries difficult. Such regions are commonly observed. The power spectrum is shown in the inset. (B) Computed director field of the image shown in part A. Short lines are drawn parallel to the direction of the observed lattice planes and the width is set in proportion to the local intensity. (C) A reconstruction of the average polymer direction using continuous lines with the local slope taken from the director field. This represents the average direction of the polymer chains and is a useful method to visualize long-range organization in the film. Scale bar is 50 nm.
Figure 5. An attempted reconstruction of the average polymer backbone orientation based on the molecular director field. Frequent bend and splay are observed along with a highly overlapping structure. Scale bar is 250 nm.
algorithm used by Sun et al.21 A similar algorithm has also been used in the analysis of liquid crystalline-like dispersions of carbon nanotubes.22 It is important to properly accommodate the overlapping structure; thus, we allow the molecular director to take on multiple directions at each location to reflect the layered nature. Using the local value of the molecular director as the tangent vector, we draw continuous lines representing the average orientation of the polymer backbones as shown in
Figure 4C. As these lines are propagated through the director field, they bend as to follow the slow contours of the director. As shown in Figure 5, long-range ordering of the lamella along with gradual bending of the director is readily apparent on length-scales approaching a micrometer. Within the large oriented regions of P(NDI2OD-T2), small fluctuations in the direction of the lines are often observed and highlight the propensity for local disorder. The complex nanostructure of the D
dx.doi.org/10.1021/nl4005805 | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
Letter
HRTEM images were acquired using a low-dose methodology similar to Sun et al.21 The incident electron beam is set to be approximately parallel. The sample is mechanically translated to an undamaged area of the sample. The minimal defocus required to capture the spatial frequencies of interest is applied and the sample is allowed to mechanically stabilize. The electron beam is then shifted off the optical axis and several images are taken in rapid succession. As the diffracted angles corresponding to the spatial frequencies of interest are small, the effect of off-axis aberrations is negligible. Multiple exposures of the same area are useful in postprocessing to evaluate stability and damage-related effects. The electronic shifting and imaging acquisition is repeated 8 times about the central spot. About 20 unique positions across 3 different samples were imaged in this study with no substantial variation in the behavior observed. For STEM imaging, the microscope was operated in uProbe mode: the probe convergence angle was set to 2.3 mrad and a 768 mm camera length to give an approximately 0.6 nm focused probe. All images were acquired on an FEI Titan microscope operating at 300 kV.
overlapping regions is greatly simplified and we observe that the overlapping layers can bridge the large, oriented domains. Furthermore, we expect these layers are coupled electronically and improve the overall transport by reducing the effects of interdomain boundaries, which may explain the reported low energetic disorder. 7−9 Splay is also apparent but the interpretation is more difficult. It could be related to the segregation of chain ends or a spherulite-like structure but more information will be needed regarding the full 3D organization within the film. Nonetheless, while the volume fraction of this “face-on” population (i.e., the total crystallinity) is unknown, the long-range order, rodlike molecular structure and confined sample geometry suggest that material in orientations/ conformations not accessible with TEM will still be strongly influenced by the high degree of structural order of its neighbors. We add that long-range order is not unique to P(NDI2ODT2). Similar long-range order has been reported in both highmobility transistors and some small-molecule OPV devices.23−26 We are currently studying other semiconducting polymers using TEM and find similar, although less obvious, order.27 Here, because of the unconventional “face-on” texture of the lamella, we can directly resolve the remarkable nanoscale organization over length-scales ranging from nanometers to micrometers using a combination HRTEM and STEM. The structural organization of P(NDI2OD-T2) emphasizes that a wide-range of length-scales are important for both successful characterization and function of high-performance polymeric materials. One observes not only a highly ordered lamellar nanostructure over nearly the entire film, but highly oriented domains with overlapping layers providing additional interconnectivity. By accounting here for the regions of overlap of polymer chains, which have previously been assigned as amorphous regions,13 the overall degree of order in the sample is increased and in agreement with the previous X-ray scattering measurements.10 These results provide a simple picture that unifies the seemingly contradictory conclusions surrounding this remarkable material. The successful integration of these properties correlates with the remarkable transport properties of the polymer and serves as a useful guide for the design of the next generation of high-performance polymers. Significantly higher carrier mobilities should be achievable through simultaneously improvements of the local crystallinity, longrange order, and interconnectivity of domains. Experimental Details. Thin-films giving a “face-on” orientation were prepared similar to those in Rivnay et al.10 Samples were cast on cleaned silicon/SiO2 wafers. P(NDI2ODT2) (Mn = 33 kDa, PDI = 3.4) was dissolved in 70 °C orthodichlorobenzene at a concentration of 10 mg mL−1. Samples were spin-cast at 4000 rpm for 50 s and annealed at 165 °C for 10 min under N2 atmosphere on a calibrated hot-plate. The resulting films were ∼20 nm thick. The thin-film was then delaminated on to an air−DI water interface after dissolving the SiO2 layer in a 5% HF solution and transferred to a C-Flat TEM grid with 2 um diameter holes to allow for unobstructed view of the sample or an ultrathin carbon film with lacey support from Ted Pella and dried. Prior to the film transfer, the TEM grids were soaked in a solution of ∼15 nm citratestabilized gold nanoparticles, washed in isopropyl alcohol, and dried to aid microscope alignment and focusing of the electron beam.
■
ASSOCIATED CONTENT
S Supporting Information *
Large areas of HRTEM and STEM images along with DFT optimized structures of linear chains are available. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Present Address ⊥
Department of Materials, Imperial College London, London SW7 2AZ. Author Contributions
The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS C.J.T. would like to acknowledge Edward J. Kramer, David S. Cannell, Ben B. Y. Hsu, Loren G. Kaake, and Jacek J. Jasieniak for useful discussions. The TEM studies were supported by the National Science Foundation (DMR-0856060). We also acknowledge support from the UCSB Center for Scientific Computing NSF CNS-0960316 and the Materials Research Laboratory an NSF MRSEC (DMR-1121053). M.L.C. and N.D.T. were supported by NSF DMR-1207549.
■
REFERENCES
(1) Kumatani, A.; Liu, C.; Li, Y.; Darmawan, P.; Takimiya, K.; Minari, T.; Tsukagoshi, K. Sci. Rep. 2012, 2. (2) He, Z.; Zhong, C.; Su, S.; Xu, M.; Wu, H.; Cao, Y. Nat. Photonics 2012, 6 (9), 591−595. (3) Dou, L.; You, J.; Yang, J.; Chen, C.-C.; He, Y.; Murase, S.; Moriarty, T.; Emery, K.; Li, G.; Yang, Y. Nat. Photonics 2012, 6 (3), 180−185. (4) Yan, H.; Chen, Z.; Zheng, Y.; Newman, C.; Quinn, J. R.; Dotz, F.; Kastler, M.; Facchetti, A. Nature 2009, 457 (7230), 679−686. (5) Schubert, M.; Dolfen, D.; Frisch, J.; Roland, S.; Steyrleuthner, R.; Stiller, B.; Chen, Z.; Scherf, U.; Koch, N.; Facchetti, A.; Neher, D. Adv. Energy Mater. 2012, 2 (3), 369−380.
E
dx.doi.org/10.1021/nl4005805 | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
Letter
(6) (a) Facchetti, A. Mater. Today 2013, DOI: 10.1016/j.mattod.2013.04.005. (b) Facchetti, A. Presented at the 2012 Organic Photovoltaics XIII Symposium, SPIE Optics + Photonics Annual Meeting, San Diego, CA, August 14, 2012. (c) Drees, M.; Chen, Z.; Pan, H.; Facchetti, A. Unpublished work, 2013. (7) Caironi, M.; Bird, M.; Fazzi, D.; Chen, Z.; Di Pietro, R.; Newman, C.; Facchetti, A.; Sirringhaus, H. Adv. Funct. Mater. 2011, 21 (17), 3371−3381. (8) Blakesley, J. C.; Schubert, M.; Steyrleuthner, R.; Chen, Z.; Facchetti, A.; Neher, D. Appl. Phys. Lett. 2011, 99 (18), 183310−3. (9) Lange, I.; Blakesley, J. C.; Frisch, J.; Vollmer, A.; Koch, N.; Neher, D. Phys. Rev. Lett. 2011, 106 (21), 216402. (10) Rivnay, J.; Toney, M. F.; Zheng, Y.; Kauvar, I. V.; Chen, Z.; Wagner, V.; Facchetti, A.; Salleo, A. Adv. Mater. 2010, 22 (39), 4359− 4363. (11) Rivnay, J.; Steyrleuthner, R.; Jimison, L. H.; Casadei, A.; Chen, Z.; Toney, M. F.; Facchetti, A.; Neher, D.; Salleo, A. Macromolecules 2011, 44 (13), 5246−5255. (12) Fabiano, S.; Musumeci, C.; Chen, Z.; Scandurra, A.; Wang, H.; Loo, Y.-L.; Facchetti, A.; Pignataro, B. Adv. Mater. 2012, 24 (7), 951− 956. (13) Sciascia, C.; Martino, N.; Schuettfort, T.; Watts, B.; Grancini, G.; Antognazza, M. R.; Zavelani-Rossi, M.; McNeill, C. R.; Caironi, M. Adv. Mater. 2011, 23 (43), 5086−5090. (14) Egerton, R. F.; Li, P.; Malac, M. Micron 2004, 35 (6), 399−409. (15) Downing, K. H.; Glaeser, R. M. Ultramicroscopy 1986, 20 (3), 269−278. (16) Martin, D. C.; Chen, J.; Yang, J.; Drummy, L. F.; Kübel, C. J. Polym. Sci., Part B: Polym. Phys. 2005, 43 (14), 1749−1778. (17) Kübel, C.; González-Ronda, L.; Drummy, L. F.; Martin, D. C. J. Phys. Org. Chem. 2000, 13 (12), 816−829. (18) Schuettfort, T.; Huettner, S.; Lilliu, S.; Macdonald, J. E.; Thomsen, L.; McNeill, C. R. Macromolecules 2011, 44 (6), 1530− 1539. (19) Rivnay, J.; Noriega, R.; Northrup, J. E.; Kline, R. J.; Toney, M. F.; Salleo, A. Phys. Rev. B 2011, 83 (12), 121306. (20) Steyrleuthner, R.; Schubert, M.; Howard, I.; Klaumünzer, B.; Schilling, K.; Chen, Z.; Saalfrank, P.; Laquai, F.; Facchetti, A.; Neher, D. J. Am. Chem. Soc. 2012, 134 (44), 18303−18317. (21) Sun, Y.; Welch, G. C.; Leong, W. L.; Takacs, C. J.; Bazan, G. C.; Heeger, A. J. Nat. Mater. 2012, 11 (1), 44−48. (22) Shaffer, M. S. P.; Fan, X.; Windle, A. H. Carbon 1998, 36 (11), 1603−1612. (23) Zhang, X.; Hudson, S. D.; DeLongchamp, D. M.; Gundlach, D. J.; Heeney, M.; McCulloch, I. Adv. Funct. Mater. 2010, 20 (23), 4098− 4106. (24) Zhang, X.; Richter, L. J.; DeLongchamp, D. M.; Kline, R. J.; Hammond, M. R.; McCulloch, I.; Heeney, M.; Ashraf, R. S.; Smith, J. N.; Anthopoulos, T. D.; Schroeder, B.; Geerts, Y. H.; Fischer, D. A.; Toney, M. F. J. Am. Chem. Soc. 2011, 133 (38), 15073−15084. (25) Collins, B. A.; Cochran, J. E.; Yan, H.; Gann, E.; Hub, C.; Fink, R.; Wang, C.; Schuettfort, T.; McNeill, C. R.; Chabinyc, M. L.; Ade, H. Nat. Mater. 2012, 11 (6), 536−543. (26) Takacs, C. J.; Sun, Y.; Welch, G. C.; Perez, L. A.; Liu, X.; Wen, W.; Bazan, G. C.; Heeger, A. J. J. Am. Chem. Soc. 2012, 134 (40), 16597−16606. (27) Takacs, C. J.; Treat, N. D.; Brady, M. A.; Sun, Y.; Kim, H.; Lee, K.; Toney, M. F.; Kramer, E.J.; Chabinyc, M. L.; Heeger, A. J. In preparation.
F
dx.doi.org/10.1021/nl4005805 | Nano Lett. XXXX, XXX, XXX−XXX
