Compatibilization of All-Conjugated Polymer Blends for Organic

Aug 2, 2016 - Compatibilization of an immiscible binary blend comprising a conjugated electron donor and a conjugated electron acceptor polymer with s...
0 downloads 7 Views 2MB Size
Compatibilization of All-Conjugated Polymer Blends for Organic Photovoltaics Florian Lombeck,†,‡ Alessandro Sepe,§ Ralf Thomann,∥ Richard H. Friend,† and Michael Sommer*,‡,∥,⊥ †

Cavendish Laboratory, Department of Physics, University of Cambridge, 19 J J Thomson Avenue, Cambridge CB3 0HE, United Kingdom ‡ Makromolekulare Chemie, Universität Freiburg, Stefan-Meier-Straße 31, 79104 Freiburg, Germany § Adolphe Merkle Institute, Chemin des Verdiers 4, CH-1700 Fribourg, Switzerland ∥ Freiburger Materialforschungszentrum, Stefan-Meier-Straße 21, 79104 Freiburg, Germany ⊥ FIT, Freiburger Zentrum für interaktive Werkstoffe und bioinspirierte Technologien, Georges-Köhler-Allee 105, 79110 Freiburg, Germany S Supporting Information *

ABSTRACT: Compatibilization of an immiscible binary blend comprising a conjugated electron donor and a conjugated electron acceptor polymer with suitable electronic properties upon addition of a block copolymer (BCP) composed of the same building blocks is demonstrated. Efficient compatibilization during meltannealing is feasible when the two polymers are immiscible in the melt, i.e. above the melting point of ∼250 °C of the semicrystalline donor polymer P3HT. To generate immiscibility at these high temperatures, the acceptor polymer PCDTBT is equipped with fluorinated side chains leading to an increased Flory−Huggins interaction parameter. Compatibilization in bulk and thin films is demonstrated, showing that the photovoltaic performance of pristine microphase separated and nanostructured BCPs can also be obtained for compatibilized blend films containing low contents of 10−20 wt % BCP. Thermodynamically stable domain sizes range between several tens of microns for pure blends and ∼10 nm for pure block copolymers. In addition to controlling domain size, the amount of block copolymer added dictates the ratio of edge-on and face-on P3HT crystals, with compatibilized films showing an increasing amount of face-on P3HT crystals with increasing amount of compatibilizer. This study demonstrates the prerequisites and benefits of compatibilizing all-conjugated semicrystalline polymer blends for organic photovoltaics. KEYWORDS: compatibilization, ternary conjugated polymer blends, organic photovoltaics, donor−acceptor conjugated block copolymers, semifluorinated alkyl side chains, phase separation, PCDTBT, P3HT

T

thermodynamic factors resulting in kinetically trapped nonequilibrium morphologies. Meanwhile, all-conjugated polymer polymer blends, which commonly exhibit macrophase separation leading to morphologies that are yet more difficult to control, have very recently matured from moderate to excellent performance.15−22 Employment of acceptor polymers in allpolymer solar cells has several advantages compared to widely employed classical polymer−fullerene analogs, for example, widely tunable absorption and energy levels. However, control over domain size, domain shape, and domain orientation remain an issue especially in all-polymer blends. This can, in

he efficiency of solution processable organic photovoltaic (OPV) devices fundamentally relies, among other parameters, on the morphology of the ∼100 nm thin active layer, which must exhibit a phase separated, cocontinuous network of interpenetrating donor and acceptor domains commensurate to the exciton diffusion length. Additionally, a high interfacial area for efficient charge separation, continuous percolation pathways for unhindered charge carrier transport through the film and successful collection at the electrodes is key.1−5 Several strategies have been successfully applied to gain control of the active layer morphology of polymer fullerene blends at an empirical level, such as careful choice of appropriate casting solvents,6,7 additives,8,9 active material ratio and concentration,10,11 thermal annealing,12,13 and solvent vapor annealing.14,15 These approaches rely on the complex interplay of kinetic and © 2016 American Chemical Society

Received: June 27, 2016 Accepted: August 2, 2016 Published: August 2, 2016 8087

DOI: 10.1021/acsnano.6b04244 ACS Nano 2016, 10, 8087−8096

Article

www.acsnano.org

Article

ACS Nano

Figure 1. Chemical structures of P3HT, various PCDTBT materials and the all-conjugated block copolymer SF-PCDTBT-b-P3HT (SF-BCP).

polymer/polymer blends with suitable optoelectronic properties for photovoltaics is yet to be demonstrated. Poly[N-9′-heptadecanyl-2,7-carbazole-alt-5,5-(4′,7′-di-2thienyl-2′,1′,3′-benzothiadiazole)] (PCDTBT) is an ambipolar semiconducting polymer, and, in combination with fullerene derivatives, a widely used p-type semiconducting polymer for OPV devices that exhibits excellent stability and an outstanding long-term device operation lifetime.45 Because of its ambipolarity, PCDTBT can also act as efficient electron acceptor in combination with strong donors such as poly(3-hexylthiophene) (P3HT).9 PCDTBT behaves similar to PFTBT, the latter of which giving power conversion efficiencies of ∼3% in BCP photovoltaics32 and ∼2.7% in P3HT blends.46 In this work, we demonstrate compatibilization of an immiscible allconjugated polymer blend composed of P3HT and a PCDTBT-based acceptor with the corresponding all-conjugated BCP PCDTBT-b-P3HT.36 We further show that in order to compatibilize all-conjugated polymer blends usually exhibiting small degrees of polymerization, macrophase separation in the melt is required, which is induced by usage of semifluoroalkyl (SF) side chains to increase dissimilarity of P3HT and PCDTBT. An increasing amount of all-conjugated BCP leads to continuously decreasing domain sizes. For 10−20 wt % all-conjugated BCP, the same device characteristics can be obtained as for the pristine all-conjugated BCP. BCP addition also enables beneficial reorientation of P3HT crystallites from edge-on in pristine blends to face-on orientation in ternary compatibilized blends, showing that all-conjugated BCP compatibilizers allow optimizing several parameters simultaneously.

principle, be solved by using donor−acceptor block copolymers (BCP) with suitable electronic properties enabling tunable and thermodynamically stable morphologies with domain sizes on the order of 10−20 nm.23−32 The rational design and the synthesis of well-defined all-conjugated BCPs exhibiting stateof-the art high performance building blocks as well as phase separation is synthetically yet more challenging due to mostly used step growth reactions and the concomitant reduced degree of control over molecular structure, rendering these sophisticated materials cost-intensive.32−37 To this end, usage of a small percentage of all-conjugated BCPs as compatibilizers appears a promising strategy to control, predict and reproduce the size of thermodynamically stable domains. However, though compatibilization is a classical strategy to control domain size in conventional polymer blends, this concept has not yet been demonstrated with all-conjugated polymer blends exhibiting suitable optoelectronic properties and promising device performance. The addition of a few percent of conventional BCPs to the corresponding polymer blend reduces its interfacial energy thus enabling control over domain size and leads to thermodynamically stable morphologies.38−40 A further advantage is the potential formation of a microemulsion, which is an isotropic structure and hence does not require vertical alignment in thin films for efficient charge transport.41 Although conventional polymer blends generally favor demixing,42 this situation is less clear and more challenging to achieve in all-conjugated polymer blends as a consequence of their typically small degrees of polymerization and intrinsic similarities, namely a rigid, aromatic backbone flanked by solubilizing alkyl side chains.36 In case of often used semicrystalline conjugated polymers, a prerequisite for meltcompatibilization is macrophase separation at temperatures above the melting point to provide a driving force for the BCP to locate at the polymer polymer interface, which in turn requires systems with sufficiently large Flory−Huggins interaction parameters. In case crystallization of one or two components drives phase separation from a homogeneous melt, the resulting morphologies are a strong function of the kinetics of crystallization, and hence empirically well-optimized annealing protocols need to be established in order to reproducibly control morphology.43 Studies in which two electronically active polymers have been compatibilized to control domain size for photovoltaics have been presented,44 but a clear demonstration of compatibilizing all-conjugated

RESULTS AND DISCUSSION In this study, we use two semifluoroalkylated PCDTBT-based block copolymers SF-PCDTBT-b-P3HT (SF-BCP) of different molecular weight (MW) and composition, the synthesis of which is described elsewhere.36 The chemical structures of the polymer blend components and the block copolymer are shown in Figure 1. We note that both the P3HT homopolymer as well as the P3HT block in SF-PCDTBT-b-P3HT contain a distributed tail-to-tail regiodefect, although this is not shown in Figure 1 for clarity.47 The energy levels of these materials match the energetic requirements for an OPV device, namely a low band gap to absorb light in the visible region of the solar spectrum, sufficient environmental stability and LUMO− LUMO and HOMO−HOMO offsets greater than 0.3 eV to 8088

DOI: 10.1021/acsnano.6b04244 ACS Nano 2016, 10, 8087−8096

Article

ACS Nano overcome the Coulombic attraction of an electron−hole pair to generate free charge carriers. It should be noted that the HOMO level of SF-PCDTBT is lowered by 0.3 and 0.4 eV compared to the parent hexyl-PCDTBT and H-PCDTBT, respectively, as a result of an additional surface dipole moment caused by the SF side chains pointing away from the surface.48 However, the band gap is unaffected by the presence of the SF side chains due to electronic insulation from the conjugated backbone by a butylene spacer, as evidenced by identical absorption spectra of both copolymers.36 This fact in principle should result in a lower open-circuit voltage (VOC) of the device. Although knowing the absolute energy levels of pristine materials via UPS and UV−vis spectroscopy, the energy levels of donor and acceptor at DA interfaces are difficult to determine since they rely among other factors on the nature of the interface (Table SI-1).5 Composition and molecular weights of all materials were determined by 1H NMR and size exclusion (SEC) analysis, respectively.36 The number-average molecular weights of P3HT, SF-PCDTBT, hexyl-PCDTBT, and H-PCDTBT are Mn,SEC = 19,1 kg/mol, 37.1 kg/mol, 46.9 kg/mol, and 22.0 kg/ mol, respectively (see also Table SI-1). Two SF-BCPs were used in this study, referred to as SF-BCP-13 and SF-BCP-20. SF-BCP-13 exhibits an Mn,SEC = 35.8, was prepared from a P3HT-Br macroreagent with Mn,SEC = 13 kg/mol and consists of 66% wt % SF-PCDTBT and 34 wt % P3HT. SF-BCP-20 has Mn,SEC = 43.2 kg/mol, was made from P3HT-Br with Mn,SEC = 20 kg/mol and is composed of 60 wt % SF-PCDTBT and 40 wt % P3HT. Influence of the Side Chains of PCDTBT on Phase Separation of PCDTBT:P3HT Blends. To develop a conjugated polymer blend system which phase separates in the melt, the influence of the side chain of PCDTBT on the phase separation of PCDTBT:P3HT blends was investigated first. A set of three PCDTBT acceptor polymers in which the substituents on the thiophene rings was varied. Namely, HPCDTBT (R = H, Figure 2), hexyl-PCDTBT (R = n-hexyl, Figure 2) and SF-PCDTBT (R = −C4H8C4F9, Figure 2) were blended with P3HT in a 1:1 ratio and annealed above the melting point (Tm) of P3HT, that is, at 260 °C, for 30 min (Tm (P3HT) = 225 °C). Using optical microscopy to image thick, drop-casted films (∼1 μm), the hexyl-PCDTBT:P3HT blend reveals a featureless morphology (Figure 2a). Transmission electron microscopy (TEM) and atomic force microscopy (AFM) were used to image thermally annealed thin films (260 °C, 65−70 nm), confirming a featureless, homogeneous structure of the hexyl-PCDTBT:P3HT blend films, indicating miscibility of the two polymers (Figure 2d,g). Obviously, compatibilization under such circumstances is not possible as no defined interface is present at which BCPs are able to locate. In addition, strong photoluminescence (PL) quenching (Supporting Information, Table SI-2) corroborates the finding of a substantially intermixed morphology, caused by the 3hexyl-thiophene motif present in both P3HT and hexylPCDTBT. In line with this scenario, usage of HPCDTBT:P3HT blends led to significant coarsening of morphology (Figure 2b,e,h). Although the TEM image (Figure 2h) does not reveal an indication for a phase separated structure, the optical micrograph (Figure 2b) along with the AFM image (Figure 2e) suggest domain sizes in the submicrometer range accompanied by moderate PL quenching (Supporting Information, Table SI-2). In strong contrast, a well-defined macrophase separated structure with domains in

Figure 2. Micrographs of the PCDTBT:P3HT 1:1 blends. (a)−(c) Optical microscopy of 800−1000 nm thick films in the melt at 260 °C. (d)−(f) Atomic force microscopy (phase images, film thickness 65−70 nm). (g)−(i) Transmission electron microscopy images (film thickness 65−70 nm, films stained with iodine, dark: P3HT). (a), (d), (g) hexyl-PCDTBT:P3HT; (b), (e), (h) HPCDTBT:P3HT; (c), (f), (i) SF-PCDTBT:P3HT.

the micrometer length scale is observed when SFPCDTBT:P3HT is investigated (Figure 2c,f,i). Furthermore, the PL of SF-PCDTBT in SF-PCDTBT:P3HT is only weakly quenched compared to the analogous hexyl-PCDTBT:P3HT blend as a consequence of a substantially reduced interfacial area between SF-PCDTBT and P3HT. Hence, semifluoroalkylation of one of the blend components significantly increases dissimilarity between PCDTBT and P3HT, that is, the Flory− Huggins interaction parameter, and therefore enables macrophase separation in the melt, providing sharp interfaces at which the corresponding BCP can locate. Furthermore, the excellent solubility of SF-PCDTBT allows solution blending with a variety of conjugated polymers in common organic solvents due to the presence of both, SF- and alkyl side chains at the TBT and carbazole repeat unit, respectively. All these properties make SF-PCDTBT:P3HT an ideal test system to explore the possibility of compatibilization, in contrast to the synthetically much more simple H-PCDTBT:P3HT blend, the latter of which showing a lower tendency of phase separation as well as limited solubility. Compatibilization of SF-PCDTBT:P3HT Blends in Bulk. To investigate the capability and efficiency of SF-BCP, which consists of the two immiscible blend components covalently linked together, to locate at well-defined interfaces in phaseseparated SF-PCDTBT:P3HT melts and thus act as compatibilizer, bulk experiments on thick (∼1 μm) films were performed. Pristine and compatibilized blends SF-PCDTBT:P3HT (1:1 by weight) were heated to 260 °C, which is above Tm of P3HT (Tm= 225 °C). Thick films obtained via drop casting showed featureless structures (Figure 3a). After surpassing T = 225 °C and isothermal annealing at 260 °C for 30 min, the bulk morphology changed instantaneously and drastically. Although 8089

DOI: 10.1021/acsnano.6b04244 ACS Nano 2016, 10, 8087−8096

Article

ACS Nano

higher MW BCPs exhibiting more efficient compatibilization, i.e. less BCP is required in order to achieve the same reduction or stabilization of domain size.49 SF-BCP-13 and SF-BCP-20 differ in both P3HT block length and overall MW, but have similar composition. Interestingly, SF-BCP-20 has the same P3HT block length than the used P3HT blend component of about 20 kg/mol, whereas SF-BCP-13 has a smaller P3HT segment length compared to the P3HT homopolymer. Hence, an interesting question arises as to whether or not compatibilization with SF-BCP-20 is different. When discussing the different segment lengths of P3HT in homo and block copolymers, chain folding that sets in for Mn,SEC ∼ 13 kg/mol needs to be kept in mind, which coincides with the MW of P3HT in SF-BCP-13. 50,51 Compatibilization of SFPCDTBT:P3HT with SF-BCP-20 exhibited qualitatively a similar trend compared to SF-BCP-13 (Figure 4). Domain size decreased continuously upon increasing addition of 1−5% SFBCP-20 (Figure 4c−e). However, for larger weight fractions of SF-BCP-20 of 10−50 wt % (Figure 4f−h), domain size is smaller compared to the analogous experiment with SF-BCP-13 (compare to Figure 3f−h). From these experiments, we conclude that a higher content of SF-BCP-13 is needed to reach the same domain size reduction as with SF-BCP-20, suggesting more efficient compatibilization of the latter. Thin Film Compatibilization of SF-PCDTBT:P3HT Blends. To investigate phase separation behavior of the binary and ternary blends in thin films relevant to organic photovoltaics, films were prepared via spin coating to yield thicknesses in the range of 65 nm−70 nm. All films exhibited smooth surface topographies with rms roughnesses of ∼5.6 nm before thermal annealing (Figure SI-1). As can be seen from AFM phase images (Figures 5 and 6), the topographical modulation of thin films follows a similar trend that was also observed in bulk. In contrast to bulk samples, thin film morphological features are about 1 order of magnitude smaller as a consequence of geometric confinement.52 The binary blend exhibits a featureless and smooth surface morphology after spin coating (Figure 5a). Contact angle measurements and etching with selective solvent followed by AFM and UV/vis measurements are consistent with the formation of a bilayer structure, in which P3HT wets the SiO2 substrate and SF-PCDTBT forms the top layer (Figures SI-2 and SI-3). This is expected, given that fluorinated materials exhibit low surface tension and hence are drawn to the polymer−air interface.53 Upon thermal annealing the pristine blend at 260 °C, phase separation into ∼1 μm domains occurs (Figure 5b). Consistent with optical microscopy, a similar trend can at first glance be extracted from AFM analysis of thin films when using SF-BCP-13 as compatibilizer. Domain size decreases continuously from ∼1 μm (0 wt % SF-BCP-13, Figure 5b) to ∼300 nm (10 wt % SF-BCP-13, Figure 5e). For all higher SF-BCP-13 contents, domain size could not be extracted from AFM analysis (Figure 5f−h). The domain size of pristine SF-BCP-13 could also not be extracted; however, previous GISAXS experiments have indicated a phase separation on the ∼10 nm scale.36 The same thin film experiments were performed with SFBCP-20 as compatibilizer. In this case, domain size decreased as well, but a continuous trend was not observed (Figure 6b−d). We assume that additional processes, possibly dewetting phenomena, are at play, which overlay a continuous reduction of domain size with increasing block copolymer content. The origin if this behavior is unclear at present. Interestingly, at the

Figure 3. Optical micrographs of bulk samples of ternary SFPCDTBT:P3HT 1:1 blends compatibilized with different amounts of SF-BCP-13. All samples were taken at 260 °C after annealing for 30 min except a), which is the micrograph of the binary blend as prepared. Image dimensions are 75 μm × 100 μm. (a) 0% SF-BCP13; (b) 0% SF-BCP-13; (c) 1% SF-BCP-13; (d) 2% SF-BCP-13; (e) 5% SF-BCP-13; (f) 10% SF-BCP-13; (g) 20% SF-BCP-13; (h) 50% SF-BCP-13, and (i) 100% SF-BCP-13.

macrophase separation occurred for the binary blend leading to >50 μm large domains (Figure 3b), the addition of only 1% SFBCP-13 reduced P3HT domain size to ∼30 μm (Figure 3c). Further increasing the SF-BCP-13 content decreased domain size to ∼10 μm (2 wt % SF-BCP-13, Figure 3d), to ∼8 μm (5 wt % SF-BCP-13, Figure 3e), to ∼4 μm (10 wt % SF-BCP-13, Figure 3f) and to ∼3 μm (20 wt % SF-BCP-13, Figure 3g). For 50 and 100 wt % SF-BCP-13 no defined structures were discernible from optical microscopy (Figure 3h,i). To investigate the effect of BCP MW on compatibilization, the same set of experiments was performed with SF-BCP-20 (Figure 4). Effects of MW are generally important, with usually

Figure 4. Compatibilization with SF-BCP-20. Optical micrographs of ternary blends with varying percentage of BCP recorded in the melted state after 30 min at 260 °C, except (a), which is a micrograph of the binary blend as prepared. The image dimensions are 75 μm × 100 μm. BCP weight-percentage for the ternary blends are (a) 0% (as-prepared), (b) 0%, (c) 1%, (d) 2%, (e) 5%, (f) 10%, (g) 20%, (h) 50%, and (i) 100%. 8090

DOI: 10.1021/acsnano.6b04244 ACS Nano 2016, 10, 8087−8096

Article

ACS Nano

Figure 5. AFM phase images of thin films of ternary SF-PCDTBT:P3HT 1:1 blends compatibilized with SF-BCP-13. All films were annealed for 30 min at 260 °C except (a), which is an AFM phase image of the binary SF-PCDTBT:P3HT 1:1 blend as spun. (a) 0% SF-BCP-13; (b) 0% SF-BCP-13; (c) 2% SF-BCP-13; (d) 5% SF-BCP-13; (e) 10% SF-BCP-13; (f) 20% SF-BCP-13; (g) 50% SF-BCP-13; and (h) 100% SF-BCP-13.

Figure 6. AFM phase images of thin films of ternary SF-PCDTBT:P3HT 1:1 blends compatibilized with SF-BCP-20. All films were annealed for 30 min at 260 °C except (a), which is an AFM phase image of the binary SF-PCDTBT:P3HT 1:1 blend as spun. (a) 0% SF-BCP-20; (b) 0% SF-BCP-20; (c) 2% SF-BCP-20; (d) 5% SF-BCP-20; (e) 10% SF-BCP-20; (f) 20% SF-BCP-20; (g) 50% SF-BCP-20; and (h) 100% SF-BCP-20.

stage of 10 wt % added SF-BCP-20, smooth films are seen from which an extraction of domain size is no longer possible (Figure 6e). A similar AFM image was obtained from ternary films compatibilized with SF-BCP-13; however, here, 20 wt % had to be used to achieve the same smooth surface structure. This is in line with bulk compatibilization, where SF-BCP-20 was shown to be more effective and showed a stronger reduction in domain size compared to SF-BCP-13. We will show later that indeed SF-BCP-20 enables more efficient photovoltaic cells already for 10 wt %, whereas 20 wt % SF-BCP-13 need to be added to achieve the same effect. However, it needs to be noted that although the relative observations made with AFM and optical microscopy are consistent, AFM is only able to probe the structure of the top layer, and not the entire thin film morphology. Further experiments are underway to corroborate these results. Photoluminescence Quenching and Solar Cell Performance. The degree of photoluminescence quenching is a measure for the probability of an exciton to reach the donor− acceptor interface, thus being able to create a charge in an OPV device. The measured values for photoluminescence quantum efficiency (PLQE) go along with the observed reduction in domain size and increase in interfacial donor−acceptor area. For a given exciton diffusion length, smaller domains result in

an increasing probability for an exciton to reach a donor− acceptor interface and thus create charge. The initial binary blend exhibits a PLQE of 8.8% which is decreased by 70% to 2.6% in a ternary blend composition with 5% SF-BCP-20. Upon further addition of SF-BCP-20 the PLQE remained constant with small fluctuations between 2.6% and 3.6% for SFBCP-20 percentages between 10% and 100% (Figure 7). Hence, the PLQE is significantly reduced in all ternary systems, with larger amounts of SF-BCP-20 leading to stronger quenching. In the case of SF-BCP-13, the drop in PLQE upon initial BCP addition is not as strong as with SF-BCP-20, which is perfectly consistent with the difference in domain size reduction seen in both bulk and thin film experiments. Thus, the PLQE experiments confirm the more effective compatibilization behavior of SF-BCP-20 compared to SF-BCP-13. Nevertheless, both pristine BCPs show very similar photoluminescence quantum and therefore quenching efficiencies as a consequence of small domains and the enhanced probability of an exciton to reach a DA-interface prior to being quenched. (Figure SI-4, Table SI-2). The trend that can be extracted from PL quenching experiments can be observed in photovoltaic device performance in a similar fashion when using the standard device architecture (glass/ITO/PEDOT:PSS/active layer/Al). To 8091

DOI: 10.1021/acsnano.6b04244 ACS Nano 2016, 10, 8087−8096

Article

ACS Nano

crystals to the photocurrent in pristine SF-BCP-13 compared to the ternary systems (Figure 8b). The compatibilization profile of SF-BCP-20 is qualitatively similar compared to SF-BCP-13, but OPV parameters saturate already for lower contents of SF-BCP-20 (Figure 9). Here, already for 10−20 wt % SF-BCP-20 the same device characteristics as for pristine SF-BCP-20 can be achieved. Further BCP addition does neither increase the JSC nor the VOC. This result is consistent with the relative trends seen for the two block copolymers: Apparently, the development of the morphology and the optoelectronic properties of compatibilized thin films were a function of the size and the composition of the BCP. For an amount of, for example, 10 wt % BCP, the longer SF-BCP-20 enabled smaller domains as seen by bulk and thin film compatibilization, a more strongly reduced PLQE and a higher solar cell performance. Interestingly, a generally strong increase of the open circuit voltage VOC from 0.5 V only for the pristine blend to up to 1.2 V for compatibilized blends was found for both SF-BCP-13 and SF-BCP-20. Tentatively, this is ascribed to a smoother top layer composed of SF-PCDTBT in case of compatibilized thin films leading to reduced surface recombination. In binary blends, the different surface energies of SF-PCDTBT and P3HT (see Figure SI-2 for contact angles) lead to vertical phase separation and hence bilayer formation after spin coating (Figure 5a, Figure SI-3). However, after melt annealing, evolution of the observed lateral morphology can facilitate contact of the P3HT component with the cathode as well (Figure 5b), despite similar contact angles. Such a smoother top layer composed of SF-PCDTBT can additionally lead to reduced ohmic leakage currents. Figure SI-7 shows that the root-mean-square roughness (rms) obtained from AFM of the compatibilized thin films indeed decreases with increasing BCP content. The finer structures can simultaneously lead to better contact of the SFPCDTBT component with the aluminum cathode thereby reducing surface recombination. Similar effects have also been observed with ternary block copolymer/homopolymer blends.54 A detailed spectroscopic study is underway to confirm this assumption. Notably, photovoltaic devices with inverted device structure (glass/ITO/TiO2/active layer/MoO3/Au) gave only negligible photocurrents and open circuit voltages, presumably caused by the mentioned surface layer of the semifluoroalkylated side chains of SF-PCDTBT. Thus, a capping layer of the acceptor

Figure 7. Photoluminescence quantum efficiency (PLQE) of ternary SF-PCDTBT:P3HT 1:1 blends compatibilized with SFBCP. Black filled squares, SF-BCP-20; red filled circles, SF-BCP13.

illustrate compatibilization in OPV devices, the OPV parameters of solar cells made from compatibilized blend films, prepared in the same way as film for AFM and PLQE analysis, are plotted as a function of wt % SF-BCP-13 and SFBCP-20 (Figure 8a and 9a, respectively). The corresponding EQE curves with SF-BCP-13 and SF-BCP-20 are shown in Figure 8b and 9b, respectively (for J−V curves see Figure SI-5). All photovoltaic device characteristics are summarized in Table SI-3 and Table SI-4. The solar cells made from pristine SF-PCDTBT:P3HT 1:1 blends all resulted in poor device characteristics and a power conversion efficiency (PCE) of 0.09% only as a consequence of large domains and a minimal interfacial area, as expected. As the amount of SF-BCP-13 is successively increased, a strong improvement of all OPV parameters was found for low BCP contents up to 10 wt % SF-BCP-13, with a less strong increase for ∼20 wt % SF-BCP-13. It is noteworthy that performance continuously increased up to the pristine SF-BCP-13, which exhibited the best performance (Figure 8a). Similarly, the EQE curves exhibited an increase in the peak-EQE with increasing BCP content. This observation is in line with an increase in short-circuit current density. Although the EQE curves of the compatibilized blend devices show similar shapes, the EQE curve of the pristine SF-BCP-13 exhibited a distinct shoulder at lower energies, indicating a higher contribution of P3HT

Figure 8. Solar cell parameters of ternary SF-PCDTBT:P3HT 1:1 blends compatibilized with SF-BCP-13. (a) Short circuit current JSC, open circuit voltage VOC and power conversion efficiency PCE as a function of weight percentage SF-BCP-13 and (b) EQEs of SF-PCDTBT:P3HT (1:1) blends compatibilized with SF-BCP-13. 8092

DOI: 10.1021/acsnano.6b04244 ACS Nano 2016, 10, 8087−8096

Article

ACS Nano

Figure 9. Solar cell parameters of ternary SF-PCDTBT:P3HT 1:1 blends compatibilized with SF-BCP-20. (a) Short circuit current JSC, open circuit voltage VOC and power conversion efficiency PCE as a function of weight percentage SF-BCP-20. (b) External quantum efficiency (EQE) as a function of weight percentage SF-BCP-20.

found, P3HT in the blend exhibited preferential edge-on orientation with a spacing of 1.59 nm. This suggests that covalent connectivity of SF-PCDTBT to P3HT leads to changes in both P3HT crystal orientation as well as in a slightly increased main chain side chain distance. For the ternary mixtures, a gradually reorientation of the P3HT crystals from preferentially edge-on to completely face-on was found. The ratio of the peaks corresponding to edge-on and face-on orientation is shown in Figure 11, from which the gradual reorientation of P3HT crystals becomes evident.

formed at the anode by virtue of self-assembly, which acts as a blocking layer inhibiting hole extraction. To probe the semicrystalline structure and orientation of P3HT contained in the compatibilized thin films, the binary and compatibilized ternary blends were additionally studied by grazing incidence X-ray scattering (GIXS), which is a powerful technique to study the structure of thin films.55 Figure 10

Figure 11. Ratio of face-on to edge-on oriented P3HT crystals in ternary, compatibilized films containing SF-BCP-20, SF-PCDTBT, and P3HT.

The observations of enhanced OPV device performance by improving open-circuit voltage and increasing short-circuit current densities can on the one hand be explained by the gradually reduction of domains size. A similar observation was made by He et al. for nanoimprinted devices with gradually smaller domains.56 On the other hand, the beneficial reorientation of P3HT crystallites from mainly edge-on toward fully face-on is observed. The origin of saturation of PLQE and device parameters in the compatibilization profile of SF-BCP20 remains open at this point and is subject of ongoing investigations.

Figure 10. GIXS: 1D intensity scattering profiles (region of the 100 reflection) along the edge-on (a) and face-on (b) direction with SFBCP-20 as compatibilizing agent. The upper curves are from pristine SF-BCP-20, the lowest ones from pure SF-PCDTBT:P3HT 1:1 blends.

shows GIXS data from thin films compatibilized with SF-BCP20. The scattering patterns show prominent Bragg reflections at positions which correspond to d-spacing and lattice structure of the corresponding thin films, thus employing a wide q-range where structural properties are investigated. Although for SFBCP-20, a face-on orientation of P3HT crystals with a main chain side chain separation (100 reflection) of 1.65 nm was

CONCLUSION We have shown that an immiscible binary blend of semiconducting polymers with suitable electronic properties for organic photovoltaics can be compatibilized by the addition of a block copolymer (BCP) composed of the same segments. 8093

DOI: 10.1021/acsnano.6b04244 ACS Nano 2016, 10, 8087−8096

Article

ACS Nano

SR303i, Andor Technologies). Three individual measurements were performed to calculated the quantum yield ϕ according to DE MELLO et al.61 In the ON measurement, the laser is focused on the film directly, whereas in the OFF measurement, the laser does not hit the sample but the reflective wall of the integrating sphere. The background is subtracted by measuring the integrating sphere without sample to account for impurities inside the sphere. The PLQE is calculated by the difference in the area of the laser peak of all three measurements, which gives the amount of photons absorbed by the film, and the difference in the signal area of the emission peak, giving the number of photons emitted. Device Fabrication. Photovoltaic devices were fabricated in the standard ITO|PEDOT:PSS| BCP|Al layer sequence. ITO on glass anodes were first cleaned via sonication in acetone and isopropanol, followed by oxygen plasma treatment. A ∼ 40 nm thick PEDOT:PSS layer was spin-coated onto the plasma-treated substrates, annealed at 230 °C for 10 min and transferred into a nitrogen-filled glovebox for the subsequent fabrication steps. Photoactive layers were spin-cast from chlorobenzene-solution (12 mg/mL) to give ∼65−70 nm thick films and thermally annealed at 260 °C for 30 min. Additional annealing was performed at 140 °C for 10 min after deposition of an Al electrode (100 nm). Device Testing. Device EQE was measured using a tungsten lamp with a monochromator at intensities of ∼1 mW cm−2. Short circuit currents were recorded using a Keithley 237 source meter. The current−voltage characteristics of the device were measured under simulated 100 mW cm−2 AM 1.5 illumination using an Abet Technology solar simulator. The spectral mismatch of the simulator was calibrated to a silicon reference cell. Atomic Force Microscopy. AFM capturing was carried out on a Veeco Dimension 3100 AFM, operated in tapping mode. Grazing Incidence Wide-Angle X-Ray Scattering. The GIWAXS measurements were carried-out at the beamline D1, Cornell High Energy Synchrotron Source (CHESS), Cornell University, Ithaca, NY, U.S.A. The wavelength was λ = 1.15 Å. The beam was focused to a size of 500 μm × 100 μm (horizontal × vertical) at the sample position. A beam stop for the primary beam and the smallangle X-ray scattering signal was employed. A CCD detector with a pixel size of 46.9 μm × 46.9 μm was used for GIWAXS with a sample− detector distance of 0.11 m. In order to choose the appropriate incident angle, an X-ray reflectometry scan was performed before recording the image. Images were then taken at an incident angle (αi) which is slightly higher than the critical angle of the film (αcf) and lower than the critical angle of the substrate (αcs). Thus, the entire film was penetrated, the internal film structures could be detected and the beam was fully reflected from the sample/substrate interface. Furthermore, at these incident angles the beam footprint on the sample is similar to the whole sample size. The samples were moved out of the beam after each measurement to avoid beam damage. The q-space calibration was performed fitting the characteristic scattering signal arising from silver behenate. The conversion of the 2D images from pixels to q-values as well as the analysis of the 2D scattering maps were carried out using Xi-CAM (http://www.camera.lbl.gov/#!xi-caminterface/z8vcm). The horizontal and vertical regions used for the angular integrations are described elsewhere.43

Phase separation in the melt is a prerequisite, which was achieved here by increasing dissimilarity of the donor P3HT and the acceptor polymer PCDTBT by means of semifluoroalkyl side chains (SF) in SF-PCDTBT. Binary blends SFPCDTBT:P3HT exhibited macrophase separation behavior in bulk as well as in thin film with micrometer sized domains. Increasing addition of the corresponding BCP SF-PCDTBT-bP3HT (SF-BCP) led to thermodynamically stable, continuously decreased domain sizes, increased photoluminescence quenching, and increased solar cell performance. Higher molecular weights of SF-BCP led to more effective compatibilization. Though for the shorter SF-BCP-13 (Mn,SEC = 35 kg/ mol) device performance continuously increased with increasing content of SF-BCP-13, only 10−20 wt % of SF-BCP-20 (Mn,SEC = 43.2 kg/mol) were needed to obtain the same OPV performance as for pristine SF-BCP-20. Additionally advantageous is that the presence of SF-BCP in ternary blends enabled beneficial reorientation of P3HT crystals from preferentially edge-on to fully face-on, demonstrating that various parameters can be optimized simultaneously. The herein presented concept contributes to the prevalent discussions related to morphology control and morphological long-term stability in OPV devices and can potentially be extended to any conjugated polymer/ polymer blend system suitable for photovoltaic application, in which control over domain size emerges as most critical parameter.

METHODS Materials. All polymers were synthesized according to the literature.36,57−60 Solvents were purchased from Sigma-Aldrich and used as received. Gel Permeation Chromatography. GPC measurements of hexPCDTBT, SF-PCDTBT, and all BCP samples were carried out on two PL gel 5 μm mixed C columns with pore sizes ranging from 10 to 105 Å (Polymer Laboratories), connected in series with a SPD-M20A prominence diode array detector (Shimadzu) and a differential refractometer/viscometer model 200 (Viscotek) calibrated with polystyrene standards. THF was used as eluent at 22 °C with a flow rate of 1.0 mL/min. H-PCDTBT was measured on an Agilent PLGPC220 instrument equipped with a RI detector at 150 °C in 1,2,4trichlorobenzene (TCB) and calibrated relative to polystyrene. UV−Vis Spectroscopy. UV−vis spectra were recorded on a Lambda 650 S spectrometer from PerkinElmer in film or in chlorobenzene solutions (c = 0.02 mg/mL). Optical Microscopy. Optical micrographs were captured using an Olympus BX60 microscope equipped with various UPLFLN Series objectives and a transmission illuminator U-LH100. Images were captured with an Infinity 1 camera. Transmission Electron Microscopy. TEM was performed on a Zeiss LEO 912 Omega TEM at an acceleration voltage of 120 kV and zero loss filtering. Samples were spin coated onto PSS substrates, thermally annealed and floated onto water, collected by copper grids and stained using iodine. Ultraviolet Photoelectron Spectroscopy. For UPS measurements 7 nm of chromium and 70 nm of gold were evaporated on top of a clean silicon substrate. The ∼10−15 nm thin polymer film was deposited via spin coating. After thermal annealing at 160 °C for 10 min in a nitrogen-filled glovebox, the films were transferred into an ultrahigh vacuum chamber (ESCALAB 250Xi). UPS measurements were carried out using a pumped He gas discharge lamp emitting He I radiation (hν = 21.2 eV). Photoluminescence Quantum Efficiency. The measurement is carried out on films deposited on quartz glass (Spectrosil 2000) via excitation of the sample with a 523 nm laser. The sample is placed inside an integrating sphere. The integrating spherés inside is reflective and purged with nitrogen. The light collected from the integrating sphere is coupled through a fiber into a spectrometer (Shamrock

ASSOCIATED CONTENT S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsnano.6b04244. Additional characterization data (PDF)

AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest. 8094

DOI: 10.1021/acsnano.6b04244 ACS Nano 2016, 10, 8087−8096

Article

ACS Nano

(14) Verploegen, E.; Miller, C. E.; Schmidt, K.; Bao, Z.; Toney, M. F. Manipulating the Morphology of P3HT−PCBM Bulk Heterojunction Blends with Solvent Vapor Annealing. Chem. Mater. 2012, 24, 3923− 3931. (15) Sepe, A.; Rong, Z.; Sommer, M.; Vaynzof, Y.; Sheng, X.; MüllerBuschbaum, P.; Smilgies, D.-M.; Tan, Z.-K.; Yang, L.; Friend, R. H. Structure Formation in P3HT/F8TBT Blends. Energy Environ. Sci. 2014, 7, 1725. (16) McNeill, C. R.; Greenham, N. C. Conjugated-Polymer Blends for Optoelectronics. Adv. Mater. 2009, 21, 3840−3850. (17) Facchetti, A. Polymer Donor−polymer Acceptor (all-Polymer) Solar Cells. Mater. Today 2013, 16, 123−132. (18) Earmme, T.; Hwang, Y.; Murari, N. M.; Subramaniyan, S.; Jenekhe, S. A. All-Polymer Solar Cells with 3.3% Efficiency Based on Naphthalene Diimide-Selenophene Copolymer Acceptor. J. Am. Chem. Soc. 2013, 135, 14960−14963. (19) Hwang, Y.; Courtright, B. A. E.; Ferreira, A. S.; Tolbert, S. H.; Jenekhe, S. A. 7.7% Efficient All-Polymer Solar Cells. Adv. Mater. 2015, 27, 4578−4584. (20) Jung, J. W.; Jo, J. W.; Chueh, C.; Liu, F.; Jo, W. H.; Russell, T. P.; Jen, A. K. Fluoro-Substituted N-Type Conjugated Polymers for Additive-Free All-Polymer Bulk Heterojunction Solar Cells with High Power Conversion Efficiency of 6.71%. Adv. Mater. 2015, 27, 3310− 3317. (21) Kim, T.; Kim, J.; Kang, T. E.; Lee, C.; Kang, H.; Shin, M.; Wang, C.; Ma, B.; Jeong, U.; Kim, T.-S.; Kim, B. J. Flexible, Highly Efficient All-Polymer Solar Cells. Nat. Commun. 2015, 6, 8547. (22) Gao, L.; Zhang, Z.; Xue, L.; Min, J.; Zhang, J.; Wei, Z.; Li, Y. AllPolymer Solar Cells Based on Absorption-Complementary Polymer Donor and Acceptor with High Power Conversion Efficiency of 8.27%. Adv. Mater. 2016, 28, 1884−1890. (23) Scherf, U.; Gutacker, A.; Koenen, N. All-Conjugated Block Copolymers. Acc. Chem. Res. 2008, 41, 1086−1097. (24) Segalman, R. a.; McCulloch, B.; Kirmayer, S.; Urban, J. J. Block Copolymers for Organic Optoelectronics. Macromolecules 2009, 42, 9205−9216. (25) Hüttner, S.; Sommer, M.; Chiche, A.; Krausch, G. Controlled Solvent Vapour Annealing for Polymer Electronics. Soft Matter 2009, 5, 4206−4211. (26) Sommer, M.; Hüttner, S.; Thelakkat, M. Donor−Acceptor Block Copolymers for Photovoltaic Applications. J. Mater. Chem. 2010, 20, 10788−10797. (27) Lohwasser, R. H.; Gupta, G.; Kohn, P.; Sommer, M.; Lang, A. S.; Thurn-Albrecht, T.; Thelakkat, M. Phase Separation in the Melt and Confined Crystallization as the Key to Well-Ordered Microphase Separated Donor-Acceptor Block Copolymers. Macromolecules 2013, 46, 4403−4410. (28) Topham, P. D.; Parnell, A. J.; Hiorns, R. C. Block Copolymer Strategies for Solar Cell Technology. J. Polym. Sci., Part B: Polym. Phys. 2011, 49, 1131−1156. (29) Sommer, M.; Lindner, S. M.; Thelakkat, M. MicrophaseSeparated Donor−Acceptor Diblock Copolymers: Influence of HOMO Energy Levels and Morphology on Polymer Solar Cells. Adv. Funct. Mater. 2007, 17, 1493−1500. (30) King, S.; Sommer, M.; Huettner, S.; Thelakkat, M.; Haque, S. A. Charge Separation and Recombination in Self-Organizing Nanostructured Donor−Acceptor Block Copolymer Films. J. Mater. Chem. 2009, 19, 5436−5441. (31) Bu, L.; Guo, X.; Yu, B.; Qu, Y.; Xie, Z.; Yan, D.; Geng, Y.; Wang, F. Monodisperse Co-Oligomer Approach toward Nanostructured Films with Alternating Donor−Acceptor Lamellae. J. Am. Chem. Soc. 2009, 131, 13242−13243. (32) Guo, C.; Lin, Y. H.; Witman, M. D.; Smith, K. a; Wang, C.; Hexemer, A.; Strzalka, J.; Gomez, E. D.; Verduzco, R. Conjugated Block Copolymer Photovoltaics with Near 3% Efficiency through Microphase Separation. Nano Lett. 2013, 13, 2957−2963. (33) Verduzco, R.; Botiz, I.; Pickel, D. L.; Kilbey, S. M.; Hong, K.; Dimasi, E.; Darling, S. B. Polythiophene- Block -Polyfluorene and Polythiophene-b-Poly(fluorene-Co-Benzothiadiazole): Insights into

ACKNOWLEDGMENTS S.M. Menke, F. Deschler, P. Kohn, U. Steiner and A. Hexemer are gratefully acknowledged for stimulating discussions. We thank D.-M. Smilgies, X. Sheng, and J. Dolan for their help during the D1 experiment at CHESS. Part of the work was conducted at beamline D1 at the Cornell High Energy Synchrotron Source (CHESS). CHESS is supported by the NSF & NIH/NIGMS via NSF award DMR-1332208. A. Hexemer and Xi-CAM are supported by the U.S. Department of Energy under Contract No. DE-AC02-05CH1123, the ECA award program and the LBNL LDRD “TReXS”. Financial support from the Fonds der Chemischen Industrie (FCI), the Research Innovation Fund of the University of Freiburg and the DFG (SPP1355) is acknowledged. F.L. acknowledges the EPSRC for funding. ABBREVIATIONS SEC size exclusion chromatography; P3HT poly-3-hexylthiophene; PCDTBT poly[n-9′-heptadecanyl-2,7-carbazole-alt5,5-(4′,7′-di-2-thienyl-2′,1′,3′-benzothiadiazole)]; SF semifluoroalkylated; BCP block copolymer; MW molecular weight REFERENCES (1) Hoppe, H.; Sariciftci, N. S. Morphology of Polymer/Fullerene Bulk Heterojunction Solar Cells. J. Mater. Chem. 2006, 16, 45−61. (2) Liu, F.; Gu, Y.; Jung, J. W.; Jo, W. H.; Russell, T. P. On the Morphology of Polymer-Based Photovoltaics. J. Polym. Sci., Part B: Polym. Phys. 2012, 50, 1018−1044. (3) McNeill, C. R. Morphology of All-Polymer Solar Cells. Energy Environ. Sci. 2012, 5, 5653−5667. (4) Li, G.; Zhu, R.; Yang, Y. Polymer Solar Cells. Nat. Photonics 2012, 6, 153−161. (5) Lu, L.; Zheng, T.; Wu, Q.; Schneider, A. M.; Zhao, D.; Yu, L. Recent Advances in Bulk Heterojunction Polymer Solar Cells. Chem. Rev. 2015, 115, 12666−12731. (6) Shaheen, S. E.; Brabec, C. J.; Sariciftci, N. S.; Padinger, F.; Fromherz, T.; Hummelen, J. C. 2.5% Efficient Organic Plastic Solar Cells. Appl. Phys. Lett. 2001, 78, 841. (7) Arias, A. C.; MacKenzie, J. D.; Stevenson, R.; Halls, J. J. M.; Inbasekaran, M.; Woo, E. P.; Richards, D.; Friend, R. H. Photovoltaic Performance and Morphology of Polyfluorene Blends: A Combined Microscopic and Photovoltaic Investigation. Macromolecules 2001, 34, 6005−6013. (8) Peet, J.; Kim, J. Y.; Coates, N. E.; Ma, W. L.; Moses, D.; Heeger, a J.; Bazan, G. C. Efficiency Enhancement in Low-Bandgap Polymer Solar Cells by Processing with Alkane Dithiols. Nat. Mater. 2007, 6, 497−500. (9) Liu, X.; Huettner, S.; Rong, Z.; Sommer, M.; Friend, R. H. Solvent Additive Control of Morphology and Crystallization in Semiconducting Polymer Blends. Adv. Mater. 2012, 24, 669−674. (10) Müller, C.; Ferenczi, T. A. M.; Campoy-Quiles, M.; Frost, J. M.; Bradley, D. D. C.; Smith, P.; Stingelin-Stutzmann, N.; Nelson, J. Binary Organic Photovoltaic Blends: A Simple Rationale for Optimum Conditions. Adv. Mater. 2008, 20, 3510−3515. (11) Cho, S.; Coates, N.; Moon, J. S.; Park, S. H.; Roy, A.; Beaupre, S.; Moses, D.; Leclerc, M.; Lee, K.; Heeger, A. J. Bulk Heterojunction Solar Cells with Internal Quantum Efficiency Approaching 100%. Nat. Photonics 2009, 3, 297−302. (12) Ma, B. W.; Yang, C.; Gong, X.; Lee, K.; Heeger, A. J. Thermally Stable, Efficient Polymer Solar Cells with Nanoscale Control of the Interpenetrating Network Morphology. Adv. Funct. Mater. 2005, 15, 1617−1622. (13) Verploegen, E.; Mondal, R.; Bettinger, C. J.; Sok, S.; Toney, M. F.; Bao, Z. Effects of Thermal Annealing Upon the Morphology of Polymer−Fullerene Blends. Adv. Funct. Mater. 2010, 20, 3519−3529. 8095

DOI: 10.1021/acsnano.6b04244 ACS Nano 2016, 10, 8087−8096

Article

ACS Nano the Self-Assembly of All-Conjugated Block Copolymers. Macromolecules 2011, 44, 530−539. (34) Mulherin, R. C.; Jung, S.; Huettner, S.; Johnson, K.; Kohn, P.; Allard, S.; Scherf, U.; Greenham, N. C.; Sommer, M. Ternary Photovoltaic Blends Incorporating an All-Conjugated Donor-Acceptor Diblock Copolymer. Nano Lett. 2011, 11, 4846−4851. (35) Sommer, M.; Komber, H.; Huettner, S.; Mulherin, R.; Kohn, P.; Greenham, N. C.; Huck, W. T. S. Synthesis, Purification, and Characterization of Well-Defined All-Conjugated Diblock Copolymers PF8TBT-b-P3HT. Macromolecules 2012, 45, 4142−4151. (36) Lombeck, F.; Komber, H.; Sepe, A.; Friend, R. H.; Sommer, M. Enhancing Phase Separation and Photovoltaic Performance of AllConjugated Donor − Acceptor Block Copolymers with Semifluorinated Alkyl Side Chains. Macromolecules 2015, 48, 7851−7860. (37) Robb, M. J.; Ku, S.-Y.; Hawker, C. J. 25th Anniversary Article: No Assembly Required: Recent Advances in Fully Conjugated Block Copolymers. Adv. Mater. 2013, 25, 5686−5700. (38) Ryan, A. J. Designer Polymer Blends. Nat. Mater. 2002, 1, 8−10. (39) Ruzette, A.-V.; Leibler, L. Block Copolymers in Tomorroẃ s Plastics. Nat. Mater. 2005, 4, 19−31. (40) Utracki, L. A. Compatibilization of Polymer Blends. Can. J. Chem. Eng. 2002, 80, 1008−1016. (41) Bates, F. S.; Maurer, W. W.; Lipic, P. M.; Hillmyer, M. A.; Almdal, K.; Mortensen, K.; Fredrickson, G. H.; Lodge, T. P. Polymeric Bicontinuous Microemulsions. Phys. Rev. Lett. 1997, 79, 849−852. (42) Walheim, S.; Böltau, M.; Mlynek, J.; Krausch, G.; Steiner, U. Structure Formation via Polymer Demixing in Spin-Cast Films. Macromolecules 1997, 30, 4995−5003. (43) Kohn, P.; Rong, Z.; Scherer, K. H.; Sepe, A.; Sommer, M.; Müller-Buschbaum, P.; Friend, R. H.; Steiner, U.; Huettner, S. Crystallization-Induced 10-nm Structure Formation in P3HT/PCBM Blends. Macromolecules 2013, 46, 4002−4013. (44) Economopoulos, S. P.; Chochos, C. L.; Gregoriou, V. G.; Kallitsis, J. K.; Barrau, S.; Hadziioannou, G. Novel Brush-Type Copolymers Bearing Thiophene Backbone and Side Chain Quinoline Blocks. Synthesis and Their Use as a Compatibilizer in Thiophene Quinoline Polymer Blends. Macromolecules 2007, 40, 921−927. (45) Beaupré, S.; Leclerc, M. PCDTBT: En Route for Low Cost Plastic Solar Cells. J. Mater. Chem. A 2013, 1, 11097. (46) Mori, D.; Benten, H.; Ohkita, H.; Ito, S.; Miyake, K. Polymer/ Polymer Blend Solar Cells Improved by Using High- MolecularWeight Fluorene-Based Copolymer as Electron Acceptor. ACS Appl. Mater. Interfaces 2012, 4, 3325−3329. (47) Kohn, P.; Huettner, S.; Komber, H.; Senkovskyy, V.; Tkachov, R.; Kiriy, A.; Friend, R. H.; Steiner, U.; Huck, W. T. S.; Sommer, J.; Sommer, M. On the Role of Single Regiodefects and Polydispersity in Regioregular Poly (3-Hexylthiophene): Defect Distribution, Synthesis of Defect-Free Chains, and a Simple Model for the Determination of Crystallinity. J. Am. Chem. Soc. 2012, 134, 4790−4805. (48) Geng, Y.; Wei, Q.; Hashimoto, K.; Tajima, K. Dipole Layer Formation by Surface Segregation of Regioregular Poly(3-Alkylthiophene) with Alternating Alkyl/Semifluoroalkyl Side Chains. Chem. Mater. 2011, 23, 4257−4263. (49) Galloway, J. A.; Jeon, H. K.; Bell, J. R.; Macosko, C. W. Block Copolymer Compatibilization of Cocontinuous Polymer Blends. Polymer 2005, 46, 183−191. (50) Hamidi-Sakr, A.; Schiefer, D.; Covindarassou, S.; Biniek, L.; Sommer, M.; Brinkmann, M. Highly Oriented and Crystalline Films of a Phenyl-Substituted Polythiophene Prepared by Epitaxy: Structural Model and Influence of Molecular Weight. Macromolecules 2016, 49, 3452−3462. (51) Singh, C. R.; Gupta, G.; Lohwasser, R.; Engmann, S.; Balko, J.; Thelakkat, M.; Thurn-Albrecht, T.; Hoppe, H. Correlation of Charge Transport with Structural Order in Highly Ordered Melt-Crystallized poly(3-Hexylthiophene) Thin Films. J. Polym. Sci., Part B: Polym. Phys. 2013, 51, 943−951. (52) Zhu, S.; Liu, Y.; Rafailovich, M. H.; Sokolov, J.; Gersappe, D.; Winesett, D. A.; Ade, H. Confinement-Induced Miscibility in Polymer Blends. Nature 1999, 400, 49−51.

(53) Tada, A.; Geng, Y.; Wei, Q.; Hashimoto, K.; Tajima, K. Tailoring Organic Heterojunction Interfaces in Bilayer Polymer Photovoltaic Devices. Nat. Mater. 2011, 10, 450−455. (54) Singh, C. R.; Sommer, M.; Himmerlich, M.; Wicklein, A.; Krischok, S.; Thelakkat, M.; Hoppe, H. Morphology-Controlled Open Circuit Voltage in Polymer Solar Cells. Phys. Status Solidi RRL 2011, 5, 247−249. (55) Müller-Buschbaum, P. The Active Layer Morphology of Organic Solar Cells Probed with Grazing Incidence Scattering Techniques. Adv. Mater. 2014, 26, 7692−7709. (56) He, X.; Gao, F.; Tu, G.; Hasko, D. G.; Hüttner, S.; Greenham, N. C.; Steiner, U.; Friend, R. H.; Huck, W. T. S. Formation of WellOrdered Heterojunctions in Polymer:PCBM Photovoltaic Devices. Adv. Funct. Mater. 2011, 21, 139−146. (57) Loewe, R. S.; Khersonsky, S. M.; McCullough, R. D. A Simple Method to Prepare Head-to-Tail Coupled, Regioregular Poly(3alkylthiophenes) Using Grignard Metathesis. Adv. Mater. 1999, 11, 250−253. (58) Levesque, I.; Bertrand, P.; Blouin, N.; Leclerc, M.; Zecchin, S.; Zotti, G.; Ratcliffe, C. I.; Klug, D. D.; Gao, X.; Gao, F.; Tse, J. S. Synthesis and Thermoelectric Properties of Polycarbazole, Polyindolocarbazole and Polydiindolocarbazole Derivatives. Chem. Mater. 2007, 19, 2128−2138. (59) Lombeck, F.; Komber, H.; Gorelsky, S. I.; Sommer, M. Identifying Homocouplings as Critical Side Reactions in Direct Arylation Polycondensation. ACS Macro Lett. 2014, 3, 819−823. (60) Kim, J.; Kwon, Y. S.; Shin, W. S.; Moon, S.-J.; Park, T. Carbazole-Based Copolymers: Effects of Conjugation Breaks and Steric Hindrance. Macromolecules 2011, 44, 1909−1919. (61) DeMello, J. C.; Wittmann, H. F.; Friend, R. H. An Improved Experimental Determination of External Photoluminescence Quantum Efficiency. Adv. Mater. 1997, 9 (3), 230.

8096

DOI: 10.1021/acsnano.6b04244 ACS Nano 2016, 10, 8087−8096