Influence of Surface Energy on Organic Alloy Formation in Ternary

Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

Influence of Surface Energy on Organic Alloy Formation in Ternary Blend Solar Cells Based on Two Donor Polymers Nemal S. Gobalasingham, Sangtaik Noh, Jenna B Howard, and Barry C. Thompson ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.6b10144 • Publication Date (Web): 23 Sep 2016 Downloaded from http://pubs.acs.org on September 23, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Influence of Surface Energy on Organic Alloy Formation in Ternary Blend Solar Cells Based on Two Donor Polymers Nemal S. Gobalasingham,‡ Sangtaik Noh,‡ Jenna B. Howard, and Barry C. Thompson* Department of Chemistry and Loker Hydrocarbon Research Institute, University of Southern California, Los Angeles, California 90089-1661 Keywords: solar cell, ternary blend, semi-random copolymer, organic alloy, open-circuit voltage ABSTRACT: The compositional dependence of the open-circuit voltage (Voc) in ternary blend bulk heterojunction (BHJ) solar cells is correlated with the miscibility of polymers, which may be influenced by a number of attributes, including crystallinity, the random copolymer effect, or surface energy. Four ternary blend systems featuring poly(3hexylthiophene-co-3-(2-ethylhexyl)thiophene) (P3HT75-co-EHT25), poly(3-hexylthiophene-co-(hexyl-3-carboxylate)), herein referred to as poly(3-hexylthiophene-co-3-hexylesterthiophene) (P3HT50-co-3HET50), poly(3-hexylthiophenethiophene-diketopyrrolopyrrole) (P3HTT-DPP-10%), and an analog of P3HTT-DPP-10% with 40% of 3-hexylthiophene exchanged for 2-(2-methoxyethoxy)ethyl)thiophen-2-yl (3MEO-T) (featuring an electronically decoupled oligoether sidechain), referred to as P3HTTDPP-MEO40% are explored in this work. All four polymers are semicrystalline, rich in rrP3HT content, and perform well in binary devices with PC61BM. Except for P3HTTDPP-MEO40%, all polymers exhibit similar surface energies (~21-22 mN/m). P3HTTDPP-MEO40% exhibits an elevated surface energy of around 26 mN/m. As a result, despite the similar optoelectronic properties and binary solar cell performance of P3HTTDPP-MEO40% compared to P3HTT-DPP-10%, the former exhibits a pinned Voc in two different sets of ternary blend devices. This is a stark contrast to previous rr-P3HT-based systems and demonstrates that surface energy, and its influence on miscibility, plays a critical role in the formation of organic alloys and can supersede the influence of crystallinity, the random copolymer effect, similar backbone structures, and HOMO/LUMO considerations. Therefore, we confirm surface energy compatibility as a figure-of-merit for predicting the compositional dependence of the Voc in ternary blend solar cells and highlight the importance of polymer miscibility in organic alloy formation.

Introduction Organic solar cells continue to be a versatile and promising platform for energy conversion due in part to their compatibility with simple processing methods and flexible substrates,1,2 as well as their ability to be readily implemented into existing infrastructure as a complement to inorganic solar cells.3,4 The practical efficiency limit for single-junction bulk heterojunction (BHJ) solar cells is around 12%,5–8 with efficiencies as high as 11.7% recently reported.9 Tandem solar cells, which can achieve practical efficiencies of 14-15%,6,8,10,11 sacrifice processing simplicity for improved performance. Ternary blend BHJ solar cells featuring two donors and an acceptor (D1x:D2(1-x):Ay, where 0 < x < 1) or one donor and two acceptors (D1x:A1y:A2(1-y) , where 0 < y < 1) are an attractive and effective strategy for improved efficiencies.12–15 Comprised of a single absorbing layer, this architecture combines the simple fabrication and processing ease of single-junction solar cells with the potential efficiency boost of the tandem architecture, with single-layer ternary blend systems predicted to enable an almost 40% increase in device efficiency compared to binary blend organic photovoltaics (OPVs).16

Early reports of enhanced efficiencies in ternary-blend solar cells were attributed exclusively to the increase in the spectral response upon infusion of another photon absorber, whose absorption profile was complementary to the parent donor-acceptor pair and enabled improved short-circuit photocurrents (Jsc).17–23 Until fairly recently, the open-circuit voltage (Voc) of these devices was predicted to be pinned by the higher of the highest occupied molecular orbitals (HOMO) of the two donor components and the lowest unoccupied molecular orbital (LUMO) of the acceptor component (in D1x:D2(1-x):Ay systems, or vice versa).23 Unfortunately, because the third component often featured a narrower bandgap for an extended spectral response, it typically had a higher lying HOMO compared to the parent donor component. As a result, the Voc of these ternary-blend devices was generally smaller than that of the corresponding parent binary device.19,23–26 Recently, our group demonstrated that ternary-blend solar cells need not all exhibit this unfavorably pinned Voc value.13,27–33 Indeed, with the appropriate materials, the Voc can be tuned across a range of Voc values (confined by the individual binary devices) purely by controlling the composition of the ternary blend. This compositional depend-

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ence of the Voc has enabled a paradigm-shifting strategy for optimizing and developing ternary blend solar cells across a broad range of compositions while maintaining excellent fill factors (FF). Broadened absorption profiles, and the corresponding increase in Jsc, can now be pursued without sacrificing the Voc (Figure 1). With this emerging strategy, the only limitation compared to tandem solar cells is the confined thicknesses that can be achieved for optimal device performance, as two absorbing materials share a single active layer (whereas tandem cells have individual active layers connected in series or parallel); however, optimizing thicker films in polymer solar cells is an ongoing endeavor.34,35

Figure 1. Cartoon illustration of J-V curves of a low bandgap polymer (red) and a wide bandgap polymer (blue). The purple dashed line represents an ideal ternary blend, with high fill factors, tunable Voc, and increased Jsc from multiple absorbers, which would occupy a thicker single active layer within a simple device architecture.

Since this discovery, numerous other combinations of materials with tunable Voc have been reported;28,29,36–42 however, an equally large array of systems have also demonstrated pinned Voc values.25,43–47 Understanding the underlying morphological considerations for these observed phenomena has been a major emphasis in ternary blend research. An organic alloy model has been proposed to describe systems that exhibit tunable Voc.29,31 The usage of the term alloy in this model is analogous to the variation of valence and conduction band energies observed in inorganic semiconductor alloys with changing composition. In the case of organic systems, however, it is based on the compositionally-dependent averaging of HOMOs and LUMOs in the blends as expounded by photocurrent spectral response (PSR) measurements29 and ionization potential (IP) measurements,33 observations that provide strong support for the alloy model.31 Another model, referred to as the ‘parallel junction’ model, proposes ternary blends behave as a collection of individual binary cells; however, as noted by Kemerink et al.,48 it is unlikely that different regions can act as independent binary cells that are electronically decoupled (with isolated BHJ interpenetrating networks). Furthermore, the

Page 2 of 19

presence of metal contacts mandates that these subcells cannot be treated independently. Ultimately, the ability to predict and design custom systems that could exhibit organic alloy formation for increased Jsc, uncompromised Voc, and consistently high FF values would be incredibly beneficial in the pursuit of high performance solar cells featuring multiple donors and/or acceptors. To that end, more morphological studies are necessary to elucidate the interactions between components. Some ternary blend systems are investigated by combining two high performance systems to achieve a slightly higher percent conversion efficiency (PCE) followed by attempting to elucidate the cause of this improvement.49,50 Another strategy is to combine two structurally interesting polymers and observe the subsequent effect on device performance across a breadth of compositions. Our group recently applied this approach to further elucidate our two donor model system consisting of semirandom copolymer poly(3-hexylthiophene-thiophenediketopyrrolopyrrole) (P3HTT-DPP-10%) and random copolymer poly(3-hexylthiophene-co-3-(2ethylhexyl)thiophene) (P3HT75-co-EHT25), which exhibited tunable Voc with phenyl-C61-butyric acid methyl ester (PC61BM).28,29 The chemical structures of these components are provided in Figure 2. In a subsequent report, we studied ternary blends of perfectly alternating copolymer poly[N-9′-heptadecanyl-2,7-carbazole-alt-5,5-(4′,7′di-2-thienyl-2′,1′,3′-benzothiadiazole)] (PCDTBT) and P3HTT-DPP-10% which exhibited a pinned Voc.30 Surface energies of the polymers supported the observed miscibility of P3HTT-DPP-10% with P3HT75-co-EHT25 but not with PCDTBT. As a result, the surface energy of polymers was proposed to be a figure-of-merit for predicting alloy formation and compositional dependence of the Voc in ternary blends. However, a number of other factors could have also contributed to this non-alloying pair. In addition to different surface energies, the polymers also exhibit different backbones, one being a regioregular P3HT-based analog and the other being a carbazole-based perfectly alternating polymer entirely free of 3-hexylthiophene. Furthermore, P3HTT-DPP-10% is semi-crystalline in nature while PCDTBT is amorphous. Surface energy, the random copolymer effect (an observation that the random incorporation of common co-monomers engenders miscibil ity),51,52 and co-crystallization could all play a role in alloy formation. Recently, our group demonstrated that cocrystallization was observed in ternary blend systems of P3HTT-DPP-10% and P3HT75-co-EHT25, which is indicative of intimate mixing of the polymers; however, another semi-random P3HT-analog, poly(3-hexylthiophenethiophene-thienopyrroledione) P3HTT-TPD-10% did not exhibit co-crystallization despite demonstrating tunable Voc.33 Thus, the formation of an organic alloy is more nuanced that merely co-crystallization but both the random copolymer effect51 and surface energy remained potential figures-of-merit. A goal of the present work is the deconvolution of these other parameters in order to understand


Page 3 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2. Chemical structures of P3HTT-DPP-10%, P3HT75-co-EHT25, P3HTTDPP-MEO40% (with methoxyethoxy side-chain highlighted), P3HT50-co-3HET50, and PC61BM. Center left box details the surface energy of each polymer and illustrates the four sets of ternary blends evaluated in this work: A1 (P3HTT-DPP-10% and P3HT75-co-EHT25), A2 (P3HTT-DPP-10% and P3HT50-co3HET50), B1 (P3HTTDPP-MEO40% and P3HT75-co-EHT25), B2 (P3HTTDPP-MEO40%and P3HT50-co-3HET50). Top right box provides absorption profiles and coefficients (as determined by GIXRD reflectivity measurements) of neat films for PC61BM (brown, square) P3HT75-co-EHT25 (purple, circle), P3HT50-co-3HET50 (red, triangle up), P3HTT-DPP-10% (blue, triangle down), P3HTTDPP-MEO40% (green, diamond).

the influence of surface energy in organic alloy formation in ternary blends based on two donor constituents. Surface energy is increasingly recognized as an important parameter in solar cells.53–55 Recently, the impact of surface energy on the location of the third component has been investigated by Ohkita et al.56 with a dye molecule in the presence of P3HT and polystyrene (PS). By modifying the surface energy of dyes, they observed that a low surface energy dye aggregated near low surface energy regiorandom P3HT; whereas a high surface energy dye aggregated near high surface energy PS. Although small molecules are distinct from polymers, similar behaviors can be derived for the overall morphology of the ternary blend system. Our group recently demonstrated the ability to tune the surface energy of poly(3-alkylthiophenes) by modifying the alkyl side-chain with oligoether or semifluoroalkyl chains.53 By adding a carbon spacer between the thiophene and the relevant functionality of the side-chain, the polymers were electronically decoupled, and so exhibited optoelectronic properties such as absorption profiles, HOMO energy levels, and crystallinity that were comparable to P3HT. Depending on the co-monomer composition, however, the surface energy could be tuned from 14

to 27 mN/m. In a subsequent report, we demonstrated that modifying the surface energy of P3HTT-DPP-10% did not adversely affect device performance with PC61BM.57 Herein, we evaluate the influence of surface energy on organic alloy formation to study Voc trends in ternary blend systems based on rr-P3HT analogs. We study two deep HOMO polymers, P3HT75-co-EHT25 and poly(3hexylthiophene-co-(hexyl-3-carboxylate)), herein referred to as poly(3-hexylthiophene-co-3-hexylesterthiophene) (P3HT50-co-3HET50)58 and two analogous polymers featuring distinct surface energies, P3HTT-DPP-10% and another analog with 40% of 3HT exchanged for 2-(2methoxyethoxy)ethyl)thiophen-2-yl (3MET) (featuring an electronically decoupled oligoether side-chain), referred herein as P3HTTDPP-MEO40%. Structures of these four polymers are provided in Figure 2. Except for P3HTTDPP-MEO40%, all polymers exhibit similar surface energies (~21-22 mN/m). P3HTTDPP-MEO40% exhibits an elevated surface energy of around 26 mN/m. However, all polymers exhibit rr-P3HT-rich backbones and semicrystallinity, with the DPP-containing polymers exhibiting a broader absorption (into the near-infrared region) that is complementary to the P3HT-based random copolymers (strongly in the visible region). Our findings ulti-



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

mately demonstrate that similar polymers can exhibit either tunable or pinned Voc values solely through modification of the surface energy, even when other attributes such as semi-crystallinity, optical band gaps, HOMO/LUMO energy levels, and binary device performances are comparable. Recently, it has been asserted that poor miscibility of the two donors would not lead to different behaviors as long as the donors mixed well with the fullerene (although only miscible polymers were examined).48 We investigate two analogous polymers that individually have been shown perform and mix well with fullerene57 but because of surface energy differences, exhibit distinct behaviors in ternary blend solar cells. We confirm that surface energy is a useful figure-of-merit for predicting organic alloy formation and our results ultimately demonstrate that some level of physical and electronic interaction is important for achieving compositionally-dependent Voc.

Experimental Section Materials and Methods. All reagents from commercial sources were used without further purification, unless otherwise noted. Solvents were purchased from VWR and used without purification except for tetrahydrofuran (THF), which was dried over sodium/benzophenone before distillation. All reactions were performed under dry N2 in glassware that was pre-dried in an oven, unless otherwise noted. Flash chromatography was performed on a Teledyne CombiFlash Rf instrument with RediSep Rf normal phase disposable columns. 1H NMR spectra were recorded in CDCl3 on a Varian Mercury 400 NMR Spectrometer or a Varian Mercury 600 NMR spectrometer. For polymer thin-film measurements, solutions were spin-coated onto pre-cleaned glass slides from odichlorobenzene (o-DCB) solutions at 7 mg/mL for P3HTT-DPP-10%, P3HTTDPP-MEO40%, P3HT75-coEHT25, and P3HT50-co-3HET50. For ternary blend thin-film measurements, solutions were spin-coated from the optimal D1x:D2(1-x):Ay ratio, where D1 was the DPP-containing polymer, D2 was a random copolymer, and A is PC61BM, resulting in four sets of ternary blends. UV-Vis absorption spectra were obtained on a PerkinElmer Lamda 950 spectrophotmeter. The thickness of the thin films and grazing-incidence X-ray diffraction (GIXRD) measurements were obtained using a Rigaku Diffractometer Ultima IV using Cu Kα radiation source (λ = 1.54 Å) in the reflectivity and grazing-incidence mode, respectively. Surface energy studies of neat polymer films and PC61BM were performed on a Ramé-Hart Instrument Co. contact angle goniometer model 290-F1 and analyzed using Surface Energy (two liquids) tool implemented in DROPimage 2.4.05 software. Films on pre-cleaned glass slides were prepared via spin-coating from 5 mg/mL oDCB solutions of P3HTT-DPP-10%, P3HTTDPP-MEO40%, P3HT75-co-EHT25, and P3HT50-co-3HET50. For PC61BM, a 5 mg/mL chloroform solution was utilized. Water and glycerol were used as the two solvents in the so-called twoliquid model to measure the contact angle, and harmonic

Page 4 of 19

mean Wu model54,55 was used to calculate the average surface energy values for each film according to the following set of equations:

⋅ 1 + =

4 ⋅ ⋅ 4 ⋅ ⋅ + 1 + +

⋅ 1 + =

4 ⋅ ⋅ 4 ⋅ ⋅ + 2 + +

= + 3

= 72.8

; = 21.8 ; = 51.0

= 64.0

; = 34.0 ; = 30.0

where Zw is the contact angle with water, Zg is the contact angle with glycerol, γtot is the total surface energy, γp and γd are polar and dispersive surface energy components. Cyclic voltammetry (CV) was performed on a Princeton Applied Research VersaStat3 potentiostat under the control of VersaStudio Software. A standard three-electrode cell based on a Pt wire working electrode, a silver wire pseudo reference electrode (calibrated vs Fc/Fc+ which is taken as 5.1 eV vs vacuum),59,60 and a Pt wire counter electrode was purged with nitrogen and maintained under a nitrogen atmosphere during all measurements. Polymer films were made by drop-casting an o-DCB solution of polymer (10 mg/mL) and tetrabutylammonium hexafluorophosphate (TBAPF6) (30 mg/mL) directly onto the Pt wire and dried under nitrogen prior to measurement. Acetonitrile was distilled over CaH2 prior to use, and TBAPF6 (0.1 M) was used as the supporting electrolyte. Synthetic Procedures. Poly(3-hexylthiophenethiophene-diketopyrrolopyrrole) (P3HTTDPP-10%) (Mn = 16.5 kDa , Ð = 2.61), poly(3-hexylthiophene-co-3-(2ethylhexyl)thiophene) (P3HT75-co-EHT25) (Mn = 17.5 kDA, Ð = 2.23), poly(3-hexylthiophene-co-3hexylesterthiophene) (P3HT50-co-3HET50) (Mn = 73.3 kDa, Ð = 2.01), and poly(3-hexylthiophene-3methoxyethoxythiophene-thiophenediketopyrrolopyrrole) (P3HTTDPP-MEO40%) (Mn = 16.2 kDA, Ð = 3.1) were synthesized without modifications as reported in literature.57,58,61,62 Device Fabrication and Characterization. All steps of device fabrication and testing were performed at ambient temperatures and humidity in air. ITO-coated glass substrates (10 Ω/sq, Thin Film Devices Inc.) were sequentially cleaned by sonication in detergent, deionized water, tetrachloroethylene, acetone, and isopropyl alcohol, and dried under a nitrogen stream. PEDOT:PSS (Clevios™ PH 500, filtered with a 0.45 μm poly(vinylidene fluoride) (PVDF) syringe filter—Pall Life Sciences) was spin-coated on the pre-cleaned ITO-coated glass substrates and then


Page 5 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


annealed at 120 °C for 60 min under vacuum to generate a 40 nm thick film. Separate solutions of the polymers and PC61BM were prepared in o-DCB. The solutions were stirred for 24 h before they were mixed at the desired ratios and stirred for 24 h to form a homogeneous solution. Subsequently, the polymer:PC61BM active layer was spin-coated (filtered with a 0.45 μm polytetrafluoroethylene (PTFE) syringe filter—Pall Life Sciences) on top of the PEDOT:PSS layer. Concentrations of the binary and ternary blends were 11 mg/mL respective to the total polymer weight. For optimized conditions, devices of all the polymers were kept in a nitrogen box for 20 min after spin-coating and then placed in the vacuum chamber for aluminium deposition. The substrates were pumped down to high vacuum (

Influence of Surface Energy on Organic Alloy Formation in Ternary

Recommend Documents