Selective Morphology Control of Bulk Heterojunction in PolymerSolar Cells Using Binary Processing AdditivesYen-Sook Jung,† Jun-Seok Yeo,§ Nam-Koo Kim,‡ Sehyun Lee,† and Dong-Yu Kim*,†

†School of Materials Science and Engineering, Research Institute for Solar and Sustainable Energies, Gwangju Institute of Science andTechnology (GIST), 261 Cheomdan-gwagiro, Buk-gu, Gwangju 500-712, Republic of Korea‡Department of Nanobio Materials and Electronics, Gwangju Institute of Science and Technology (GIST), 261 Cheomdan-gwagiro,Buk-gu, Gwangju 500-712, Republic of Korea§Carbon Convergence Materials Research Center, Institute of Advanced Composite Materials, Korea Institute of Science andTechnology (KIST), Wanju-gun, Jeollabuk-do 55324, Korea

*S Supporting Information

ABSTRACT: We report the effect of binary additives on the fabrication of polymer solarcells (PSCs) based on a bulk heterojunction (BHJ) system. The combination of 1,8-diiodooctane (DIO), a high-boiling and selective solvent, for fullerene derivatives andpoly(dimethylsiloxane) (PDMS) precursor, a nonvolatile insulating additive, affordscomplementary functions on the effective modulation of BHJ morphology. It was foundthat DIO and PDMS precursor each play different roles in the control of BHJmorphology, and thus, the power conversion efficiency (PCE) can be further enhancedto 7.6% by improving the fill factor (FF) from 6.8% compared to that achieved using aconventional device employing only a DIO additive. In the BHJ of the active layer, DIOsuppressed the large phase separation of PBDTTT-CF and PC71BM while allowing theformation of continuous polymer networks in the donor polymer through phaseseparation of the PDMS precursor and BHJ components. The appropriate amount ofPDMS precursor does not disturb charge transport in the BHJ despite having insulatingproperties. In addition, the dependence of photovoltaic parameters on different lightintensities reveals that the charge recombination in the device with DIO and PDMS precursor decreases compared to thatachieved using the device with only DIO.

KEYWORDS: polymer solar cells, bulk heterojunction, PDMS resin, DIO, binary additives

■ INTRODUCTION

In bulk heterojunction (BHJ) polymer solar cells (PSCs), themorphology of the photoactive layer influences the separationof excitons and transport of separated charge carriers to theirrespective electrodes. Therefore, morphology control is acrucial factor that determines device performance.1,2 The BHJconsisting of electron donor and acceptor materials shouldform nanoscale interpenetrating networks due to excitondiffusion length within 10 nm.3 To control the nanoscalemorphology of the active layer, various methods have beenused such as thermal annealing, solvent annealing, andintroduction of interfacial layer and solvent additives.4−6

Among the methods to control the morphology of activelayer, the introduction of solvent additives is not only veryeffective but also one of the simplest. Various additives such as1,8-octanedithiol (ODT), chloronaphthalene (CN), and 1,8-diiodooctane (DIO) have been employed to improve phaseseparation and to control the morphology by enhancing thecrystallinity and by suppressing oversized phase separations.7−9

In many cases, solvent additives are essential, for example, forthe BHJ-containing a poly(thieno[3,4-b]thiophene/benzodi-thiophene) (PTB7) system because the large phase separation

between the donor and acceptor materials results in poordevice performance.10 DIO in particular has been used as asolvent additive to prevent large phase separation between thedonor and acceptor while significantly changing the morphol-ogy of the active layer by enhancing the intercalation of[6,6]phenyl C71 butyric acid methyl ester (PC71BM) intoPTB7.11−13 Also, the insulating additives such as poly-(dimethylsi loxane) (PDMS),14 ,15 (3-chloropropyl)-trimethoxysilane (CP3MS),16 and poly(methyl methacrylate)(PMMA)17 have been used to decrease the film roughness andto induce a more homogeneous morphology with fewervacancies and less charge recombination.Even though introducing additives is an attractive method to

improve device performance, a single additive dominantlyinfluences one phase, making detailed control of themorphology difficult. Recently, binary additives have beenintroduced to further optimize and to more finely tune themorphology.18 When binary additives are introduced, the

Received: August 25, 2016Accepted: October 19, 2016Published: October 19, 2016

Research Article

www.acsami.org

© 2016 American Chemical Society 30372 DOI: 10.1021/acsami.6b10715ACS Appl. Mater. Interfaces 2016, 8, 30372−30378

appropriate combination of additives should be carefullyconsidered to improve device performance.19,20 Liu et al.have investigated the effect of binary additives in PSCs basedon a low band gap polymer by changing the ratio of CN andODT. The crystallinity of the donor polymer when theoptimum ratio of CN and ODT was applied facilitated chargetransport and led to high short-circuit current (Jsc) and fillfactor (FF).21 Therefore, the introduction of well-chosen binaryadditives has the potential to further optimize deviceperformance.Herein, we fabricated PSCs with a poly[4,8-bis-substituted-

benzo[1,2-b:4,5-b′]dithiophene-2,6-diyl-alt-4-substituted-thieno[3,4-b]thiophene-2,6-diyl]fluorine (PBDTTT-CF):PC71BM BHJ and various mixtures of the binary additives,DIO and PDMS precursor. The device performance isenhanced from 6.8% to 7.68% when fabricated with binaryadditives because the addition of DIO and PDMS precursorfurther optimized the morphology of the BHJ active layer. DIOsuppressed the large phase separation of PBDTTT-CF andPC71BM, and PDMS precursor induced the formation ofcontinuous polymer networks in PBDTTT-CF through phaseseparation. In addition, we found that charge recombination inthe device fabricated with DIO and PDMS precursor is reducedcompared to that in the device fabricated only with DIObecause of the dependence of photovoltaic parameters ondifferent light intensities.

■ EXPERIMENTAL METHODSThe cleaned ITO/glass substrates were treated under UV/O3 for 15min. The PEDOT:PSS (Clevious P VP AI 4083, Heraeus) wasdeposited by spin coating at 5000 rpm for 40 s on the ITO substrates.Then, all substrates were annealed at 150 °C for 10 min in air. Toprepare the BHJ active layer, PBDTTT-CF (1-material, PDI = 2.2,molecular weight = 58 000), PC71BM (Nano-C), and PDMS precursor(Dow Corning Sylgard 184) were used. To optimize the performanceof the device, we fabricated devices with various combinations ofdonor and acceptor materials. For the most optimized device,PBDTTT-CF:PC71BM (9 mg:18 mg) was dissolved in 1 mL ofchlorobenzene with 5 vol % DIO and then was deposited on thesubstrates by spin-coating at 1500 rpm for 40 s. For devices usingPDMS precursor, 20 mg of PDMS precursor was dissolved in 1 mL ofchlorobenzene and stirred at 60 °C for 24 h. The PDMS precursorsolution was then blended from 1 vol % to 5 vol % in the samesolution used for the optimized device described above. The resultingsolution with PDMS precursor was then deposited by spin-coating at1500 rpm for 40 s on the PEDOT:PSS substrate. Finally, a 20 nm layerof Ca (1−2 Å/s deposition rate) and a 100 nm layer of Al (4−7 Å/sdeposition rate) were deposited through a shadow mask (with anactive area of 4.64 mm2) by thermal evaporation. The current density−voltage (J−V) of each device was measured using a Keithley 2400digital source measurement unit under 100 mW cm−2 irradiation witha simulated AM 1.5 G spectrum. This measurement system iscalibrated by standard Si solar cell certified by International System ofUnits (SI) (SRC 1000 TC KG5 N, VLSI Standards, Inc.). The externalquantum efficiency (EQE) spectrum was measured by a QEX-7 PVMeasurements Inc. spectral response system. To measure the mobilityof the hole and the electron, a hole only device with an ITO/PEDOT/active layer/MoO3/Ag architecture and an electron only device withan Al/active layer/Al architecture l were also fabricated. The surfacemorphology of the fabricated devices was imaged using transmissionelectron microscopy (TEM) (Technai G2 S-Twin 3600 keV electronmicroscope) and atomic force microscopy (AFM) (Veeco Dimension3100 atomic force microscope) using tapping mode. The atomconcentration was analyzed by X-ray photoemission spectroscopy(XPS) using a monochromatic Al Kα light source (AXIS NOVA,Kratos). The interface the films was measured by a contact angle

measurement system (Phoenix-300 Touch, SEO Co., Ltd., Republic ofKorea).

■ RESULTS AND DISCUSSIONFigure 1a presents the structural formulas of the photoactivematerials in the BHJ and of the additives used in this study,

with a schematic structure of the PSC provided in Figure 1b.We used a new type of insulating additive, PDMS precursor,with DIO to create a new binary additive for fabrication. In thisBHJ system, the DIO additive is essential to prevent largeaggregates of PC71BM from forming.10 The selective solubilityof DIO in PC71BM and its high boiling point (332.5 °C at 760mmHg) allow favorable intercalation of PC71BM acceptor intothe donor polymer domains, improving the surface area of theinterfaces between the donor polymer and PC71BM.11,12 ThisDIO-assisted BHJ morphology leads to efficient excitondissociation and charge transport, thereby resulting in highperforming PSCs. On the other hand, the PDMS precursorconsists of silicone elastomers for the polymerization ofpolydimethylsiloxane (PDMS). Previous studies have shownthat PDMS induces the fine and continuous interpenetratingnetwork in small molecule based organic solar cells.22−24 AlsoPDMS precursor facilitates the formation of networkedconjugated polymer pathways and exhibits improved mobilityin the organic field effect transistor and thus is an attractivecandidate to control the morphology of the BHJ active layer.25

On the basis of these results, we tried to further optimize theBHJ morphology using these complementary properties of thebinary additives.The PSC devices with active layers based on PDBTTT-CF

and PC71BM with binary additives were fabricated as follows.To finely control the amount added, 20 mg of PDMS precursorwas dissolved in 1 mL of chlorobenzene for 24 h at 60 °C, andvarious volumes of that solution from 1 vol % to 5 vol % wereblended into the active solution. Figure 2 provides

Figure 1. (a) Structural formulas of PBDTTT-CF, PC71BM, DIO, andPDMS precursor. (b) Schematic of the device architecture of the BHJsolar cell.

ACS Applied Materials & Interfaces Research Article

DOI: 10.1021/acsami.6b10715ACS Appl. Mater. Interfaces 2016, 8, 30372−30378

30373

representative current density−voltage (J−V) curves for a PSCsfabricated from PBDTTT-CF:PC71BM and our binary additivesby comparing the conventional DIO additive. Table 1 alsosummarizes the average and the upper limit of the character-istics of the devices. When the 3 vol % PDMS precursorsolution is added, the device shows the best performance; theaverage power conversion efficiency (PCE) increased from6.59% to 7.28% while enhancing FF from 63.13% to 68.39%.The best of these devices exhibits a Voc of 0.73 V, a Jsc of 14.87mA/cm2, a FF of 70.04%, and a PCE of 7.68%. From the devicecharacteristics, we confirmed that the main factor that enhancesthe device performance is an increase in FF. The FF of devicewith added PDMS precursor exhibited a value of 68% on

average and an upper limit of 70%. There is negligible deviationin the external quantum efficiency (EQE) intensity betweendevices with only DIO and with DIO and PDMS precursor asshown in Figure 2b, which indicates that the absorption and thecharge extraction abilities of device are not influenced by theinsulating PDMS precursor additive. This result is in goodagreement with the value obtained from the J−V curve showingvalues of 14.68 mA/cm2 and 14.64 mA/cm2 for devices withonly DIO and with DIO and PDMS precursor, respectively.In order to understand why the device performance

improved, transmission electron microscopy (TEM) measure-ments were carried out to investigate the morphology of theactive layer as shown in Figure 3. Figure 3a and Figure 3b show

Figure 2. (a) J−V curves and (b) external quantum efficiencies (EQE) of the PSCs based on PBDTTT-CF:PC71BM with DIO and with DIO andPDMS precursor additives.

Table 1. Device Performance of PSCs with Various Amounts of PDMS Precursor

condition Voc (V) Jsc (mA/cm2) FF (%) PCE (%)

ref 0.72 (0.73) 14.38 (14.13) 63.13 (65.78) 6.59 (6.80)a

1% PDMS precursor 0.71 14.94 64.03 6.883% PDMS precursor 0.74 (0.73) 14.34 (14.87) 68.39 (70.04) 7.28 (7.62)a

5% PDMS precursor 0.73 14.15 68.09 7.087% PDMS precursor 0.74 14.29 63.79 6.81

aThe best PCE.

Figure 3. The first row contains the TEM images of PBDTTT-CF:PC71BM (a) without DIO and PDMS precursor, (b) with DIO and without 3%PDMS precursor, (c) without 5% DIO and with PDMS precursor, and (d) with 5% DIO and with 3% PDMS precursor. The second row shows theTEM images of PBDTTT-CF (e) without DIO and PDMS precursor, (f) with DIO and without 3% PDMS precursor, (g) without 5% DIO and withPDMS precursor, and (h) with 5% DIO and with 3% PDMS precursor.

ACS Applied Materials & Interfaces Research Article

DOI: 10.1021/acsami.6b10715ACS Appl. Mater. Interfaces 2016, 8, 30372−30378

30374

the blended film of PBDTTT-CF:PC71BM without additivesand with only the DIO additive. As shown in Figure 3a,PC71BM has large aggregates with grains of around 200−300nm in diameter, which suppressed charge separation in thedonor and acceptor interface. The large domains of PC71BMdisappeared, and the donor and acceptor are evenly blendedwith added DIO as shown in Figure 3b. When PDMS precursoris added, the size and shape of PC71BM domains do notchange, and continuous bold lines appear behind the grains ofPC71BM indicating that the PDMS precursor influenced thedonor polymer more selectively as shown in Figure 3c andFigure 3d. To further clarify the impact of PDMS precursor onthe donor polymer, we examined the TEM images of donorpolymer without the PC71BM acceptor. Figure 3e shows acontrast difference dependent on the charge density despitehaving a film of pure PBDTTT-CF with darker networks ofpolymer, suggesting a highly aggregated polymer.26,27 Thenetworks observed in the pure PBDTTT-CF film are untangledby the addition of DIO as shown in Figure 3e and Figure 3f.The polymer networks become more obvious and continuouswith the PDMS precursor additive and do not disappear withthe addition of DIO as shown in Figure 3g and Figure 3h.Surface phase images of the PBDTTT-CF obtained fromatomic force microscopy (AFM) are consistent with themorphologies observed in TEM images as shown in FigureS1 in Supporting Information. Unlike finely dispersedmorphology of film with only DIO, continuously connectednetwork morphology is observed in film with DIO and PDMSprecursor as shown in Figure S1, parts b and d. However, wecould not observe distinct difference between films with andwithout PDMS precursor from GIWAXS and UV−vis spectraas shown in Figure S2, which is similar to results from previouspapers. In the GIWAXS analysis of PTB base polymer:PC71BMfilms with and without DIO, the orientation and crystallinityhave almost same property despite the significant change ofBHJ morphology.28,29 Likewise we anticipate the PDMSprecursor modified the BHJ morphology inducing aggregationof polymer; however it did not affect the crystallinity andorientation morphology. To distinguish the effect of PDMSprecursor on the donor and acceptor, TEM images of thePC71BM acceptor without the donor polymer were measured asshown in Figure S3. Compared to the donor polymer film withPDMS precursor, the networks are not observed in thePC71BM film with PDMS precursor. From these results, weconfirmed that DIO suppresses the aggregation of PC71BM andPBDTTT-CF whereas PDMS precursor induces the formationcontinuous networks in the donor polymer. The optimizedmorphology of the active layer should have good inter-penetration of the donor and acceptor materials for efficientexciton separation and an effective continuous pathway forcharge transport. The addition of binary additives bettermatches the above criteria, and as a result, it generatedimproved FF and PCE.To investigate the presence and the distribution of PDMS

precursor in the device, X-ray photoelectron spectroscopy(XPS) was employed.30 Table 2 shows the distribution ofatomic concentrations in the top and bottom regions of thefilms with only DIO and with DIO and PDMS precursor. Weidentified that PDMS precursor is a nonvolatile additive and ismainly distributed in the top region of the film. The Si atomconcentration of the film without PDMS precursor is zero,while in the film with PDMS precursor, the concentration of Siatoms is 6.14% in the top region and 1.48% in the bottom

region. Most of the blended PDMS precursor migrates to airleading to vertical phase separation of the active layer. Thevertical phase separation could be induced by the surfaceenergy difference of the PDMS precursor with the BHJconsisting of the donor and acceptor materials.31−33 PDMSprecursor has a relatively low surface energy, which is morehydrophobic compared to the donor and acceptor materials.17

The film with PDMS precursor has a higher contact anglecompared to the film without PDMS precursor as shown inFigure S4, which means that the surface becomes morehydrophobic with the addition of PDMS precursor. During spincoating, the blended PDMS precursor becomes excluded fromthe polymer due to this surface energy difference and facilitatesthe refinement of the polymer, leading to continuous polymernetworks through polymer aggregation. In addition, Moule etal. reported that the addition of poor solvent in P3HT:PCBMsolution causes the P3HT aggregate and partial crystallinity.6 Itis possible that the PDMS precursor can act as poor solvent toPBDTTT:CF:PC71BM and induce it to increase aggregation.To investigate the effect of an insulating additive on the

charge transport, we calculated the charge mobility of the filmswith only DIO and with DIO and PDMS precursor using spacecharge limited current (SCLC) measurements as shown inFigure 4.34 The values of the electron mobility are 1.42 × 10−4

cm2 V−1 s−1 and 1.62 × 10−4 cm2 V−1 s−1, and the values of thehole mobility are 2.65 × 10−4 cm2 V−1 s−1 and 2.67 × 10−4 cm2

V−1 s−1 for devices with only DIO and with DIO and PDMSprecursor, respectively. There are no significant changes to theelectron and the hole mobility despite the blending of theinsulating additive into the active layer. These results show thatthe blending in an insulating additive of adequate amount doesnot disturb the electrical properties of the device.In order to further understand the characteristics of the

device fabricated using binary additives, the electron−holerecombination mechanism needs to be investigated. Thephotocurrent density can be expressed as follows:

=J qL G I P I V( ) ( , )ph c (1)

where q is elemental charge, L is the thickness of active layer,G(I) is the generation rate of bound electron−hole pairs perunit volume, and Pc(I,V) is the charge collection proba-bility.21,35 Figure 5a and Figure 5b show a set of normalizedphotocurrents characteristic of devices with only DIO and withDIO and PDMS precursor under different irradiationintensities, which shows Pc(I,V) on voltage and light intensities.From the J−V curves shown in Figure 5a and Figure 5b, it canbe presumed that Jph is saturated at −0.5 V, Pc(I,V) = 1 becauseall the charge carriers are swept out with this large reverse bias.The Pc(I,V) is strongly dependent on light intensity, indicatingthat geminate and nongeminate recombination dominate in the

Table 2. Atomic Concentrations in the Top and BottomRegions of Films with DIO and with DIO and PDMSPrecursor

with DIOwith DIO and PDMS

precursor

atom top bottom top bottom

F 1.31 1.07 0.93 0.98O 11.27 12.99 11.11 7.21C 83.35 82.12 78.35 86.27S 4.06 3.81 3.47 4.07Si 0 0 6.14 1.48

ACS Applied Materials & Interfaces Research Article

DOI: 10.1021/acsami.6b10715ACS Appl. Mater. Interfaces 2016, 8, 30372−30378

30375

device with only DIO as shown in Figure 5a. Conversely, thePc(I,V) is independent of light intensity from −0.5 to 0.4 V,indicating that the efficiency loss in this range is dominated bygeminate recombination. A distinct discontinuity is observedfrom 0.4 to Voc, indicating that nongeminate recombinationdominates in this region for a device with DIO and PDMSprecursor, suggesting that nongeminate recombination isreduced by the PDMS precursor from −0.5 to 0.4 V. Moreover,under the open-circuit condition, the recombination mecha-nism of BHJ PSCs can be explored in detail because allphotogenerated carriers recombine at this point. When thedominant loss mechanism is nongeminate recombination, Voccan be simply described as

= +V kT q P( / ) ln( ) constantoc light (2)

where the slope of Voc vs ln(Plight) will be kT/q. In contrast,when the geminate recombination is dominant, the slope willincrease to 2kT/q. The slopes of Voc vs ln(I) are 1.27kT/q forthe device with DIO and 1.50kT/q for the device with DIO andPDMS precursor as shown in Figure 5c. From these results, wewould expect that the devices have nongeminate and geminatedrecombination; however, the geminate or trap-assistedrecombination decreases with the addition of PDMS precursorto active layer.36 The continuous network of the donor polymerreduces the trap sites by forming a favorable percolationpathway for charge transport. Also, a great gap filling propertyof PDMS precursor due to its molecular flexibility might havesuppressed the formation of trap sites and facilitate the chargetransport.16,17 These data clearly show that the enhanced deviceperformance with PDMS precursor is caused by reduced chargerecombination.

■ CONCLUSIONSIn this work, we improved the performance of a BHJ PSCdevice using the binary additives DIO and PDMS precursor.The blended binary additives each have a different influence onthe morphology of the BHJ film. DIO suppresses the formationof large domains of PC71BM with fine untangled morphology,while the PDMS precursor induces the formation of thecontinuous networks in donor polymer selectively upon phaseseparation. From the electron and hole mobility measured bySCLC, we confirmed that the blended PDMS precursor doesnot disturb the electronic properties of the device and that themobility of the hole and the electron are well balanced. Inaddition, measurements of the dependence of photovoltaic

Figure 4. SCLC of devices with DIO and with DIO and PDMSprecursor as an (a) electron only device and as a (b) hole only device.

Figure 5. J−V characteristics of devices (a) with DIO and (b) withDIO and PDMS precursor under various light intensities and (c) Vocversus the light intensity of devices with DIO and with DIO andPDMS precursor.

ACS Applied Materials & Interfaces Research Article

DOI: 10.1021/acsami.6b10715ACS Appl. Mater. Interfaces 2016, 8, 30372−30378

30376

parameters on light intensity revealed that nongeminate andgeminate recombination decreases for the device with binaryadditives compared to a device using only one additive. Furtheroptimization of the film morphology and the reduction ofcharge recombination via binary additives enhance the deviceperformance from 6.59% to 7.28%, and the best device showedup to 7.62% with a high FF. These results demonstrate that theintroduction of binary additives could be a promising methodto further optimize the morphology of the BHJ and furtherenhance device performance.

■ ASSOCIATED CONTENT

*S Supporting InformationThe Supporting Information is available free of charge on theACS Publications website at DOI: 10.1021/acsami.6b10715.

AFM images of PBDTTT-CF films depending on variousadditives, TEM imgages of PCBM films depending onvarious additives, GIWAXS images and UV−vis spectraof PBDTTT-CF:PC71BM with DIO and with DIO andPDMS precursor, contacat angles of PDMS precursorand PBDTTT-CF:PC71BM with only DIO and with DIOand PDMS precursor (PDF)

■ AUTHOR INFORMATION

Corresponding Author*E-mail: kimdy@gist.ac.kr.

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTS

This work was supported by the National Research Foundationof Korea (NRF) grant funded by the Korea government (GrantNRF-2015R1A2A1A10054466) and the Korea Institute ofEnergy Technology Evaluation and Planning (KETEP) and theMinistry of Trade, Industry & Energy (MOTIE) of theRepublic of Korea (Grant 20153010012110). This work wassupported by the “GRI (GIST Research Institute)” Projectthrough a grant provided by GIST in 2016.

■ REFERENCES(1) Deibel, C.; Strobel, T.; Dyakonov, V. Rol of the Charge TransferState in Organic Donor-Acceptor Solar Cells. Adv. Mater. 2010, 22,4097−4111.(2) Jo, J.; Na, S.-I.; Kim, S.-S.; Lee, T.-W.; Chung, Y.; Kang, S.-J.; Vak,D.; Kim, D.-Y. Three-Dimensional Bulk Heterojunction Morphologyfor Achieving Internal Quantum Efficiency in Polymer Solar Cells.Adv. Funct. Mater. 2009, 19, 2398−2406.(3) Kozub, D. R.; Vakhshouri, K.; Kesava, S. V.; Wang, C.; Hexemer,A.; Gomez, E. D. Direct measurements of exciton diffusion lengthlimitations on organic solar cell performance. Chem. Commun. 2012,48, 5859−5861.(4) Ma, W.; Yang, C.; Gong, X.; Lee, K.; Heeger, A. J. ThermallyStable, Efficienct Polymer Solar Cells with Nanoscale Control of theInterpenetrating Network Morphology. Adv. Funct. Mater. 2005, 15,1617−1622.(5) Zhang, F.; Jespersen, K. G.; Bjoerstroem, C.; Svensson, M.;Andersson, M. R.; Sundstroem, V.; Magnusson, K.; Moons, E.;Yartsev, A.; Inganaes, O. Influence of Solvent Mixing on theMorphology and Performance of Solar Cells Based on PolyfluoreneCopolymer/Fullerene Blends. Adv. Funct. Mater. 2006, 16, 667−674.(6) Moule, A. J.; Meerholz, K. Controlling Morphology in Polymer-Fullerene Mixtures. Adv. Mater. 2008, 20, 240−245.

(7) Hu, X.; Wang, M.; Huang, F.; Gong, X.; Cao, Y. 23% EnhancedEfficiency of Polymer Solar Cells Processed with 1-Chloronaphthaleneas the Solvent Additive. Synth. Met. 2013, 164, 1−5.(8) Liao, H.-C.; Ho, C.-C.; Chang, C.-Y.; Jao, M.-H.; Darling, S. B.;Su, W.-F. Additives for Morphology Control in Hihg-EfficiencyOrganic Solar Cells. Mater. Today 2013, 16, 326−336.(9) Guo, X.; Cui, C.; Zhang, M.; Huo, L.; Huang, Y.; Hou, J.; Li, Y.High Efficiency Polymer Solar Cells based on Poly(3hexylthiophene)/indene-C70 Bisadduct with Solvent Additive. Energy Environ. Sci. 2012,5, 7943−7943.(10) Chen, H.-Y.; Hou, J.; Zhang, S.; Liang, Y.; Yang, G.; Yang, Y.;Yu, L.; Wu, Y.; Li, G. Polymer Solar Cells with Enhanced Open-CircuitVoltage and Efficiency. Nat. Photonics 2009, 3, 649−653.(11) He, Z.; Xiao, B.; Liu, F.; Wu, H.; Yang, Y.; Xiao, S.; Wang, C.;Russell, T. P.; Cao, Y. Single-Junction Polymer Solar Cells with HighEfficiency and Photovoltage. Nat. Photonics 2015, 9, 174−179.(12) Guo, S.; Herzig, E. M.; Naumann, A.; Tainter, G.; Perlich, J.;Muller-Buschbaum, P. Influence of Solvent and Solvent Additive onthe Morphology of PTB7 Films Probed via X-ray Scattering. J. Phys.Chem. B 2014, 118, 344−350.(13) Alekseev, A.; Hedley, G. J.; Al-Afeef, A.; Ageev, O. A.; Samuel, I.D.W. Morpholgy and Local Electrical Properties of PTB7:PC71BMBlends. J. Mater. Chem. A 2015, 3, 8706−8714.(14) Graham, K. R.; Wieruszewski, P. M.; Stalder, R.; Hartel, M. J.;Mei, J.; So, F.; Reynolds, J. R. Improved Performance of MolecularBulk-Heterojucntion Photovoltaic Cells through Predictable Selectionof Solvent Additives. Adv. Funct. Mater. 2012, 22, 4801−4813.(15) Graham, K. R.; Mei, J.; Stalder, R.; Shim, J. W.; Cheun, H.;Steffy, F.; So, F.; Kippelen, B.; Reynolds, J. R. Polydimethylsiloxane asa Macromolecular Additive for Enhanced Performance of MolecularBulk Heterojuction Organic Solar Celss. ACS Appl. Mater. Interfaces2011, 3, 1210−1215.(16) Farahat, M. E.; Wei, H.-Y.; Ibrahem, M. B.; Boopathi, K. M.;Wei, K.-H.; Chu, C.-W. A Dual-Fucntional Additive Improves thePerformance of Molecular Bulk Heterojunction Photovoltaic Cells.RSC Adv. 2014, 4, 9401−9411.(17) Wu, F.-C.; Hsu, S.-W.; Cheng, H.-L.; Chou, W.-Y.; Tang, F.-C.Effect of Soft Insulating Polymer Doping on the PhotovoltaicProperties of Polymer-Fullerene Blend Solar Cells. J. Phys. Chem. C2013, 117, 8691−8696.(18) Kim, Y. J.; Ahn, S.; Wang, D. H.; Park, C. E. A MechanisticUnderstanding of a Binary Additive System to Synergistically BoostEfficiency in All-Polymer Solar Cells. Sci. Rep. 2015, 5, 18024.(19) Cheng, P.; Ye, L.; Zhao, X.; Hou, J.; Li, Y.; Zhan, X. Binayadditives Synergistically Boost the Efficiency of All-Polymer Solar Cellsup to 3.45%. Energy Environ. Sci. 2014, 7, 1351−1356.(20) Aïch, B. R.; Lu, J.; Beaupre, S.; Leclerc, M.; Tao, Y. Control ofthe active layer nanomorphology by using co-additives towards high-performance bulk heterojuction solar cless. Org. Electron. 2012, 13,1736−1741.(21) Liu, C.; Hu, X.; Zhong, C.; Huang, M.; Wang, K.; Zhang, Z.;Gong, X.; Cao, Y.; Heeger, A. J. The Infulence of Binary ProcessingAdditives on the Performance of Polymer Solar Cells. Nanoscale 2014,6, 14297−14304.(22) Liu, Y.; Chen, C.-C.; Hong, Z.; Gao, J.; Yang, Y. M.; Zhou, H.;Dou, L.; Li, G.; Yang, Y. Solution-Processed Small-Molecule SolarCells: Breaking the 10% Power Conversion Efficiency. Sci. Rep. 2013,DOI: 10.1038/srep03356.(23) Zhou, J.; Zuo, Y.; Wan, X.; Long, G.; Zhang, Q.; Ni, W.; Liu, Y.;Li, Z.; He, G.; Li, C.; Kan, B.; Li, M.; Chen, Y. Solution-Processed andHigh-Peroformance Organic Solar Cells Using Small Molecules with aBenzodithiophene Unit. J. Am. Chem. Soc. 2013, 135, 8484−8487.(24) Zhou, J.; Wan, X.; Liu, Y.; Zuo, Y.; Li, Z.; He, G.; Long, G.; Ni,W.; Li, C.; Su, X.; Chen, Y. Small Molecules Based on Benzo[1,2-b:4,5-b′]dithiophene Unit for High-Performance Solution-ProcessedOrganic Solar Cells. J. Am. Chem. Soc. 2012, 134, 16345−16351.(25) Choi, D.; Kim, H.; Persson, N.; Chu, P.-H.; Chang, M.; Kang, J.-H.; Graham, S.; Reichmanis, E. Elastomer-Polymer Semiconductor

ACS Applied Materials & Interfaces Research Article

DOI: 10.1021/acsami.6b10715ACS Appl. Mater. Interfaces 2016, 8, 30372−30378

30377

Blends for High-Performance Stretchable Charge Transport Networks.Chem. Mater. 2016, 28, 1196−1204.(26) Kawashima, K.; Tamai, Y.; Ohkita, H.; Osaka, I.; Takimiya, K.High-Efficiency Polymer Solar Cells with Small Photon Energy Loss.Nat. Commun. 2015, 6, 10085.(27) Foster, S.; Deledalle, F.; Mitani, A.; Kimura, T.; Kim, K.-B.;Okachi, T.; Kirchartz, T.; Oguma, J.; Miyake, K.; Durrant, J. R.; Doi,S.; Nelson, J. Electron Collection as a Limit to Polymer:PCBM SolarCell Efficiency: Effect of Blend Microstructure on Carrier Mobility andDevice Performance in PTB7:PCBM. Adv. Energy Mater. 2014, 4,1400311.(28) Hammond, M. R.; Kline, R. J.; Herzing, A. A.; Richter, L. J.;Germack, D. S.; Ro, H.-W.; Soles, C. L.; Fischer, D. A.; Xu, T.; Yu, L.;Toney, M. F.; DeLongchamp, D. M. Molecular Order in High-Efficiency Polymer/Fullerene Bulk Heterojunction Solar Cells. ACSNano 2011, 5, 8248−8257.(29) Das, S.; Keum, J. K.; Browning, J. F.; Gu, G.; Yang, B.; Dyck, O.;Do, C.; Chen, W.; Chen, J.; Ivanov, I. N.; Hong, K.; Rondinone, A. J.;Joshi, P. C.; Geohegan, D. B.; Duscher, G.; Xiao, K. Correlating HighPower Conversion Efficiency of PTB7:PC71BM Inverted OrganicSolar Cells with Nanoscale Structures. Nanoscale 2015, 7, 15576−15583.(30) Xu, Z.; Chen, L. M.; Yang, G.; Huang, C.-H.; Hou, J.; Wu, Y.; Li,G.; Hsu, C.-S.; Yang, Y. Vertical Phase Separation in Poly(3-hexylthiophene):Fullerene Derivative Blends and its Advantage forInverted Structure Solar Cells. Adv. Funct. Mater. 2009, 19, 1227−1234.(31) Qiu, L.; Lim, J. A.; Wang, X.; Lee, W. H.; Hwang, M.; Cho, K.Versatile Use of Vertical-Phase-Separatio-Induced Bilayer Structures inOrganic Thin-Film Transistors. Adv. Mater. 2008, 20, 1141−1145.(32) Jung, Y.-S.; Yeo, J.-S.; Yu, B.-K.; Kim, D.-Y. Spontaneous PhaseSeparation of a Zinc Oxide Interfacial Layer in Bulk HeterojunctionOrganic Photovoltaics. Sol. Energy Mater. Sol. Cells 2015, 134, 291−297.(33) Arias, A. C.; Endicott, F.; Street, R. A. Surface-Induced Self-Encapsulation of Polyemr Thin-Film Transistors. Adv. Mater. 2006, 18,2900−2904.(34) Dutta, G. K.; Kim, T.; Choi, H.; Lee, J.; Kim, D. S.; Kim, J. Y.;Yang, C. Synthesis of Fluorinated Analogues of a Practical PolymerTQ for Improved Open-Circuit Voltages in Polymer Solar Cells.Polym. Chem. 2014, 5, 2540−2547.(35) Cowan, S. R.; Roy, A.; Heeger, A. J. Recombination in Polymer-Fullerene Bulk Heterojunction Solar Cells. Phys. Rev. B: Condens.Matter Mater. Phys. 2010, 82, 245207.(36) Yeo, J.-S.; Yun, J.-M.; Kang, M.; Khim, D.; Lee, S.-H.; Kim, S.-S.;Na, S.-I.; Kim, D.-Y. An Approach for and Advanced Anode InterfacialLayer with Electron-Blocking Ability to Achieve High-EfficiencyOrganic Photovoltaics. ACS Appl. Mater. Interfaces 2014, 6, 19613−19620.

ACS Applied Materials & Interfaces Research Article

DOI: 10.1021/acsami.6b10715ACS Appl. Mater. Interfaces 2016, 8, 30372−30378

30378