A built-in electric field induced by ferroelectrics increases halogen-free organic solar cell efficiency in various device types

Journal Pre-proof A Built-in Electric Field Induced by Ferroelectrics Increases Halogen-Free Organic Solar Cell Efficiency in Various Device Types Tanya Kumari, Sungwoo Jung, Yongjoon Cho, Hwang-Pill Kim, Jae Won Lee, Jiyeon Oh, Jungho Lee, Sang Myeon Lee, Mingyu Jeong, Jeong Min Baik, Wook Jo, Changduk Yang PII:

S2211-2855(19)31034-1

DOI:

https://doi.org/10.1016/j.nanoen.2019.104327

Reference:

NANOEN 104327

To appear in:

Nano Energy

Received Date: 14 November 2019 Revised Date:

24 November 2019

Accepted Date: 24 November 2019

Please cite this article as: T. Kumari, S. Jung, Y. Cho, H.-P. Kim, J.W. Lee, J. Oh, J. Lee, S.M. Lee, M. Jeong, J.M. Baik, W. Jo, C. Yang, A Built-in Electric Field Induced by Ferroelectrics Increases HalogenFree Organic Solar Cell Efficiency in Various Device Types, Nano Energy, https://doi.org/10.1016/ j.nanoen.2019.104327. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Elsevier Ltd. All rights reserved.

Article type: Full Paper

A Built-in Electric Field Induced by Ferroelectrics Increases Halogen-Free Organic Solar Cell Efficiency in Various Device Types Tanya Kumari,a† Sungwoo Jung, a† Yongjoon Cho,a Hwang-Pill Kim,b Jae Won Lee,c Jiyeon Oh,a Jungho Lee,a Sang Myeon Lee,a Mingyu Jeong,a Jeong Min Baik,c Wook Jo,b Changduk Yanga* a

Department of Energy Engineering, School of Energy and Chemical Engineering, Perovtronics Research Center,

Low Dimensional Carbon Materials Center, Ulsan National Institute of Science and Technology (UNIST), 50 UNIST-gil, Ulju-gun, Ulsan 44919, Republic of Korea. b

School of Materials Science and Engineering, Ulsan National Institute of Science and Technology (UNIST), 50

UNIST-gil, Ulju-gun, Ulsan 44919, Republic of Korea. c

School of Materials Science and Engineering, KIST-UNIST Ulsan Center for Convergent Materials, Ulsan

National Institute of Science and Technology (UNIST), 50 UNIST-gil, Ulju-gun, Ulsan 44919, Republic of Korea. E-mail: [email protected]

Keywords: electric field, grafted copolymers, ferroelectric additives, organic solar cells, environment-friendly processing Graphical Abstract

1

Article type: Full Paper

A Built-in Electric Field Induced by Ferroelectrics Increases Halogen-Free Organic Solar Cell Efficiency in Various Device Types Tanya Kumari,a† Sungwoo Jung, a† Yongjoon Cho,a Hwang-Pill Kim,b Jae Won Lee,c Jiyeon Oh,a Jungho Lee,a Sang Myeon Lee,a Mingyu Jeong,a Jeong Min Baik,c Wook Jo,b Changduk Yanga* a

Department of Energy Engineering, School of Energy and Chemical Engineering, Perovtronics Research Center, Low Dimensional Carbon Materials Center, Ulsan National Institute of Science and Technology (UNIST), 50 UNIST-gil, Ulju-gun, Ulsan 44919, Republic of Korea. b School of Materials Science and Engineering, Ulsan National Institute of Science and Technology (UNIST), 50 UNIST-gil, Ulju-gun, Ulsan 44919, Republic of Korea. c School of Materials Science and Engineering, KIST-UNIST Ulsan Center for Convergent Materials, Ulsan National Institute of Science and Technology (UNIST), 50 UNIST-gil, Ulju-gun, Ulsan 44919, Republic of Korea. † Tanya Kumari and Sungwoo Jung contributed equally to this work. E-mail: [email protected]

Abstract In principle, an electric field via ferroelectric materials can affect the photovoltaic properties, although there is not yet a complete mechanistic understanding. Herein, a built-in electric field without a poling processing step was established by introducing developed PVDF-based ferroelectric additives within active-layer matrices of organic solar cells (OSCs). Upon the existence of the ferroelectric polarization induced by the ferroelectric additives in o-xylene/Nmethylpyrrolidone pair featuring halogen-free processing system, high efficiencies of 11.02% and 11.76% are achieved in fullerene and non-fullerene acceptor bulk-heterojunction OSCs, respectively. A comparative study exploring the role of the ferroelectric polarization surrounding the active-layer matrix was also performed using structural, electrical, and morphological techniques, to shed light on the underlying ferroelectric polarization effects on OSCs. 1

Furthermore, the use of the ferroelectric additive is extended to p-n like bilayer OSC to access a rich understanding of the complex enhancement mechanisms afforded by it, demonstrating a highly efficient (11.83%) bilayer device. The above results are fairly comparable to the highest value reported for the recently developed state-of-the-art OSCs processed from halogen-free systems. The use of the ferroelectric additives in the halogen-free system is promising in related organic-semiconductor fields for reasons extending beyond the enhancement of efficiency and the environment-friendly manufacturing.

Keywords: electric field, grafted copolymers, ferroelectric additives, organic solar cells, environment-friendly processing

Graphical Abstract

Research highlights •

A local built-in electric field induces in the active layer by incorporating ferroelectric additives.

•

The Ferroelectric polarization induced by cosolvent recrystallization without a poling process, resulting in enhanced photovoltaic property is demonstrated.

•

High efficiencies of 11.02% and 11.76% are achieved in fullerene and non-fullerene acceptor test-bed BHJ OSCs, respectively. 2

•

For the first time, the studies of the ferroelectric additive effect on the ‘bilayer systems’ for verifying of the working mechanisms behind ferroelectric induced electric fields in OSCs is demonstrate with 11.83% efficiency.

1. Introduction Ferroelectric materials that feature spontaneous electric polarization−commonly referred to as ferroelectricity−that can be switched by external electric fields, have remained a fascinating area of study.[1-7] In particular, the discovery of the ferroelectric phenomenon in organic polymers in the early 1970s has generated tremendous research interest in these soft materials with intriguing physical properties, and has enabled a plethora of applications ranging from sensors and actuators to acoustic imaging and information technology devices.[3,8-15] Inspired by the everincreasing demand for advanced energy technologies, there have been recent attempts to utilise the built-in electric field generated by the electric polarization of ferroelectric polymers to improve the power-conversion efficiency (PCE) in organic solar cells (OSCs).[3,13,15-18] Huang et al. incorporated a thin interfacial layer made of poled ferroelectric poly(vinylidene fluoridetrifluoroethylene) (PVDF–TrFE) at the electrode interface to generate an extra electric field in the active layer, and demonstrated an increase in the PCE of the OSCs after the poling of the PVDF–TrFE layer following the application of a large voltage pulse on the electrode.[18] More recently, Chaudhary et al. enhanced the charge collection efficiency and achieved an internal quantum efficiency of 100% by mixing a small amount of PVDF–TrFE polymer into the active layer that led to an improved device performance.[13] Nonetheless, the former method has limitations when applied to many recently developed state-of-the-art, low-band-gap polymers. This is because the Langmuir−Blodgett (LB) deposition used to fabricate the PVDF–TrFE mono3

layers is incompatible with their process owing to the high-temperature annealing (>130°C) needed to convert the high-quality ferroelectric phase in the PVDF–TrFE LB film. For such polymers, the implementation of high-temperature annealing immediately after casting can induce micrometre-sized phase segregation, thus dramatically reducing the donor–acceptor interfacial area and the device performance. For the latter approach—despite the improved PCE as a result of the local built-in electric field within the unpoled PVDF–TrFE mixed active layer—the positively poled OSC did not result in an additional performance enhancement; instead, the negative poling adversely degraded the device performance. Thus, the role of PVDF–TrFE and the working mechanisms behind ferroelectric-induced electric fields in OSCs are still not completely understood. In addition, both studies mostly focused on a traditional poly(3-hexylthiophene):[6,6]-phenyl-C61-butyric acid methyl ester (P3HT:PC61BM) system referred to as the ‘fruit fly’ platform of OSCs. As a consequence, despite the synergistic effects of electric fields induced by the electric polarization on photovoltaic properties, the achieved maximal PCEs were less than 5%. Poly(vinylidene fluoride), PVDF-based polymers are semicrystalline polymers with polymorphs referred to as α-, β-, γ-, and δ-phases (directly related to the ferroelectric properties), which are sensitive to external stimuli, such as stress, strain, temperature, electric field, or chemical substances.[4,5,7,14,19-24] Recently, it was reported that the crystal structures and phase transitions of these polymers can be easily tuned by grafting one polymer onto PVDF-based polymers.[3,8,20,25-27] To advance the understanding of the ferroelectric photovoltaic mechanism, the effects of a family of PVDF-based grafted ferroelectric polymers, PVDF, poly(vinylidene fluoride)-graft-poly(tert-butyl acrylate) (PVDF–g–PBA), PVDF–TrFE, and poly(vinylidene fluoride-trifluoroethylene)-graft-poly(tert-butyl acrylate) (PVDF–TrFE–g–PBA)) as additives 4

are systematically examined on the photovoltaic properties of fullerene and non-fullerene acceptor

bulk-heterojunction

(BHJ)

test-bed

OSC

systems

(poly[4,8-bis(5-(2-

ethylhexyl)thiophen-2-yl)benzo[1,2-b;4,5-b']dithiophene-2,6-diyl-alt-(4-(2-ethylhexyl)-3fluorothieno[3,4-b]thiophene-)-2-carboxylate-2-6-diyl] (PTB7-Th):[6,6]-phenyl-C71-butyric acid methyl ester (PC71BM) blend and (poly[(2,6-(4,8-bis(5-(2-ethylhexyl-3-fluoro)thiophen-2-yl)benzo[1,2-b:4,5-b']dithiophene))-alt-(5,5-(1',3'-di-2-thienyl-5',7'-bis(2-ethylhexyl)benzo[1',2'c:4',5'-c']dithiophene-4,8-dione)) (commonly called as PM6):3,9-bis(2-methylene-((3-(1,1dicyanomethylene)-6,7-difluoro)-indanone))-5,5,11,11-tetrakis(4-hexylphenyl)-dithieno[2,3d:2',3'-d']-s-indaceno[1,2-b:5,6-b']dithiophene (IT-4F) blend,[28-32] respectively. On the basis of high-performance

BHJ

OSCs

achieved

based

on

the

combination

of

o-xylene/N-

methylpyrrolidone (NMP) featuring halogen-free solvent/additive system previously reported from our group,[33] the o-xylene/NMP pair was used in this study as the processing solvent/additive for fabricating the environmentally less toxic solvent processed OSCs. Note that the solubility of the ferroelectric polymers in NMP is satisfactory but poor in o-xylene. Therefore, the halogen-free o-xylene/NMP pair drives the recrystallization of the ferroelectric additives into the active-layer matrices, generates built-in electric fields, and yields significant PCE enhancements of 11.02% and 11.76% for fullerene and non-fullerene acceptor OSCs, respectively. Furthermore, in order to provide the critical role of the ferroelectric additive in each donor and acceptor layer, we also performed the studies of the electric field’s effect on the p-n like PM6/IT-4F non-fullerene acceptor bilayer system and achieved PCE up to 11.83% in the bilayer device with the best-performing ferroelectric additive. The PCE values reported in this study, to the best of our knowledge, are one of the highest performing OSCs processed from halogen-free systems. The ferroelectric polarization induced by the simple addition of the 5

ferroelectric additives tool requires no additional fabrication steps, and is readily applicable to various organic optoelectronic devices in an environment-friendly way.

2. Results and Discussion For simplicity, the ferroelectric polymers PVDF, PVDF–g–PBA, PVDF–TrFE, and PVDF– TrFE–g–PBA, are henceforth denoted as P1, P2, P3, and P4, respectively, and the chemical structures are shown in Fig. 1a. P1 and P3 were commercially available materials, and P2 and P4 were prepared from grafting PBA onto the corresponding main backbones via atom transfer radical polymerization (ATRP). The details of the synthesis and characterisation (see Supplementary Information, SI Fig. S1), including the grafting ratios, are provided in the Experimental section of SI. The relative dielectric properties of the ferroelectric polymers were determined using a frequency-dependent capacitance measurement over the frequency range of 103 to 105 Hz. The relative dielectric constants (εr) were calculated from the equation (1):

εr = Cd/ε0A

(1)

Where C is the capacitance, d is the thickness of the tested film, ε0 is the permittivity of free space, and A is the area. Fig. 1b shows the frequency-dependent dielectric constant plots for the ferroelectric polymers. The dielectric constant values (P1–P4) adhere to the sequence P1 < P3 < P4 < P2 in the testing of the frequency ranges, thus verifying that the grafting of PBA in the PVDF-based backbones increases the dielectric constant values. Therefore, it is anticipated that when optimal amounts of P2 or P4 surround the host active components, efficient charge transport properties would be facilitated, thereby ultimately improving the performance of the OSCs.

6

Fig. 1. Molecular structures, dielectric strength, and theoretically simulated electrostatic surface potential. (a) Chemical structures of P1, P2, P3, and P4. (b) Dielectric constant of PVDF-based ferroelectric polymer’s thin-film capacitors. (c) Electrostatic charge distribution and dipole direction (by blue arrow) with magnitude for structural models of PVDF-based ferroelectric polymers (VDF)4, (VDF)3-(VDF-BA)1, (VDF)3-(TrFE)1, and (VDF)3-(TrFE-BA)1 in all-trans (tttt) conformation. The scale bar shows the colors scheme for the ESP. Red indicates electron rich (partially negative charge) region; yellow indicates slightly electron rich region; green indicates neutral region; light blue indicates slightly electron deficient region; and blue indicates electron deficient (partially positive charge) region, respectively.

The thermal transition behaviours of the ferroelectric polymers were evaluated via differential scanning calorimetry (DSC) (Fig. S2, SI). In the heating and cooling processes, the PVDF-based polymers (P1 and P2) underwent endothermic melting at ~170°C and exothermic crystallization transition at ~135°C, while in the case of the P(VDF–TrFE)-based polymers (P3 and P4), there were two endothermic peaks with two corresponding endothermic events. The low-temperature event was associated with a ferroelectric-to-paraelectric Curie transition, while the hightemperature event was ascribed to the melting phase transition.[34-40] Note that changes in the transition temperatures and relative integrating peaks exist in P2 and P4 compared to the 7

corresponding parent polymers P1 and P3. This indicates that the incorporation of PBA into the backbones influences the polymorphic phases and crystallinity by tuning their structural conformations. Accordingly, these account for the varied ferroelectric effects on the photovoltaic behaviours, as discussed in the following photovoltaic performance section. The intrinsic phase identification of the ferroelectric polymers was further studied by powder Xray diffraction (XRD), whereby the broad diffractions were analysed by the de-convolution of the fitted peaks (Figs. S3a and S3b, SI). P1 yields characteristic peaks at 17.77°, 18.36°, 19.96°, and 26.59°, that correspond to the (100), (020), (110), and (021) reflections of the ferroelectric αphase, whereas two peaks at 19.96° and 20.15°, observed in P3, are associated with the (110) and (200) reflection of the β-phase.[4,7,41,42] Interestingly, the peaks featuring both the α- and β-phases coexist in the grafted ferroelectric polymers (P2 and P4), and elucidate the changed crystalline structure via the introduction of the PBA to the PVDF-based backbones. Molecular simulations on model compounds with the four repeating units for each polymer were performed based on the density functional theory (DFT) at the BLYP/6–31G level. The structural models, (VDF)4, (VDF)3–(VDF-BA)1, (VDF)3–(TrFE)1, and (VDF)3–(TrFE-BA)1, were used for the calculations, whereby only the conformation with the all-trans, planar, zigzag β-phase was considered because it is responsible for the ferroelectric property.[20,21,23,43] The electrostatic potential distribution and orientation of the electric dipole at the molecular level are shown in Fig. 1c (see SI Fig. S4 for other conformational cases). The DFT calculations indicate that the presence of the PBA unit in the backbones is responsible for the variations in charge polarization, and for the direction and magnitude of the dipole moments (µ). These outcomes correlate well with the findings in the DSC data listed above.

8

The polarization behaviours of each ferroelectric polymer film fabricated from the NMP solution were characterised by the electrical polarization versus the electric field (P–E) loops using a triangular voltage waveform at 1 Hz. Fig. S5 (insets depict an optical visualisation of the films) shows that all the ferroelectric polymer films exhibit lossy dielectric behaviours, thus implying that the current P–E responses do not reflect any possible presence of inherent ferroelectricity without the external stimuli, as mentioned in the Introduction section.[3-5,7,14,24,44,45]

Fig. 2. Ferroelectricity of active layer-based thin-film capacitors using PVDF-based ferroelectric polymers as an additive in PTB7–Th:PC71BM host matrix and morphology illustration. Hysteresis loops without poling measured at various locations for (a) NMP only, 9

(b) P1, (c) P2, (d) P3, (e) P4. Insets are the corresponding optical microscopy images for the illustration of reason behind different hysteresis loops in the case of active layers with additives. Scatter plots of (f) remanent polarization, Pr (µC cm-2), and (g) the corresponding coercive field, Ec (kV cm-1) in each case. (h) Illustration of the active layer morphology in absence/presence of ferroelectric additive.

However, in the presence of additional o-xylene used as a cosolvent to NMP, the P–E hysteresis loops of all the tested systems develop into well-saturated loops, as is evident from the shape profiles shown in Fig. 2 and SI Fig. S7. This suggests that the probed spots have adequate resistances to sustain an adequately high electric field to switch the ferroelectric domains. The P–E loops of the PTB7-Th:PC71BM blend film based on the o-xylene/NMP pair without and with the optimal amount of ferroelectric additives prepared based on the same optimized OSC condition, were measured at various locations (Fig. 2). The details regarding the device optimization, including the donor:acceptor ratios and amounts of NMP and ferroelectric additives, are discussed in the Photovoltaic Performance and the Experimental section which follow. There were no ferroelectric characteristics associated with the active-layer film that was processed with o-xylene/NMP. By contrast, a typical ferroelectric behaviour was evoked that yielded rectangular-shaped hysteretic loops from all the active-layer films casted upon the addition of ferroelectric additives. This apparent development of ferroelectricity by the oxylene/NMP pair is obviously because the o-xylene serves as a good solvent for active components, while NMP acts as a good solvent for ferroelectric polymers. These results demonstrate that the formation of the β-phase crystals of the ferroelectric additives is enhanced with the use of the mixed o-xylene/NMP pair as a cosolvent recrystallization process. In turn, this effect can strongly invigorate their ferroelectric nature, even without poling treatments, in contrast to previous works.[13,15,18] Fluorescence microscopy images of the ferroelectric activelayer films processed with the use of additives and o-xylene/NMP showed direct evidence of 10

partially formed crystalline chunks of the ferroelectric additives via recrystallization within the active-layer matrix (the insets of Fig. 2 and of Fig. S6 in SI). We note that the optical images are actually enlarged to clearly show the crystalline chunks in our naked eyes. One can expect that the ferroelectricity induced by the sporadically and locally formed permanent polarization at the nanoscale is attributed to the efficient separation of the holes and electrons within the activelayer matrix, as schematically illustrated in Fig. 2h. The extracted values of the remnant polarization (Pr) and coercive field (Ec) of the processed films with the use of ferroelectric additives are summarised and compared in Figs. 2f and 2g. The active-layer film processed with P2 yielded higher Pr values (>200 µC cm-2) yet lower Ec values compared to those obtained from processed films with other additives. Accordingly, this finding has been considered as an indication of a more ordered microstructure with a high β-phase content. The measurements were obtained from the spots at which PTB7-Th:PC71BM (conductors) and ferroelectric additives (insulator) co-existed. Therefore, the obtained Pr values are the results of conductive and ferroelectric phases. Additionally, the P–E loops of the pure ferroelectric polymer films cast from the o-xylene/NMP pair were also measured at various locations (see SI Fig. S7). Strong ferroelectric responses are observed in all four cases. A maximum Pr value of ~250 µC cm-2 was obtained for P2 that is consistent with the trend of the ferroelectric behaviours associated with the additives processed active-layer films, as previously discussed. The fluorescence microscopy images of all the pure ferroelectric polymers from o-xylene/NMP featured similar bulk crystals (insets of SI Fig. S7 and Fig. S8), thus strongly justifying the o-xylene/NMP cosolvent recrystallization process.

11

Fig. 3. Effect of ferroelectric additives on the PTB7–Th:PC71BM OSCs and charge recombination dynamics study. (a) J–V characteristics (inset: J–V characteristics in the dark). (b) Histograms of power conversion efficiencies (PCE) for 10 devices in each case. (c) The corresponding EQE curves. (d) Dependence of open-circuit voltage (VOC), (e) current density (JSC) on light intensity and (f) photocurrent measurement of devices with and without PVDFbased ferroelectric polymers as an additive under AM 1.5G irradiation at 100 mW cm-2.

The photovoltaic properties of OSCs were investigated in a conventional architecture of indium tin oxide (ITO)/poly(3,4-ethylenedioxythiophene):poly(styrene sulfonate) (PEDOT:PSS)/active layer/Al based on a PTB7-Th:PC71BM active layer using an o-xylene/NMP pair, where the PTB7-Th to PC71BM ratio was maintained to 1:1.5 in this study. The best current-density– voltage (J−V) characteristics and the histogram of the PCEs for a population of 10 devices with each optimized ferroelectric additive are displayed in Figs. 3a, b respectively, and the performance parameters are summarized in Table 1. The as-cast PTB7-Th:PC71BM control device yielded a low PCE of 4.67% (average PCE = 4.23%), with a short-circuit current density 12

(JSC) of 12.52 mA/cm2, an open-circuit voltage (VOC) of 0.795 V, and a fill factor (FF) of 46.8%. However, the JSC and FF were significantly improved following the addition of an optimal amount of NMP (3.0 vol%), which resulted in a maximum PCE value of 10.20% (average PCE = 9.96%). Note that the VOC values remained at similar levels in the range of 0.795–0.799 V in the cases of the two optimized devices without and with the NMP. The photovoltaic performances were then evaluated at various weight percentages of the ferroelectric additives (0.5−4.0 wt% with respect to the total active-layer concentration (35 mg mL-1)). These additives were dissolved in the NMP prior to the addition to the o-xylene active-layer solution. The performance characteristics as a function of NMP/ferroelectric additive concentration are presented in Figs. S9–S13 (SI). Additionally, the relevant parameters and histograms for all the applied NMP/ferroelectric additives are summarized the in the SI Tables S1–S5 and Fig. S14. The optimized photovoltaic performances were obtained with the use of 1.5 wt% in all the studied cases except for P3, which was 2.0 wt%. With the exception of the device which was processed with P1 (PCE = 10.10%)—in which the PCE was slightly decreased compared to the optimized device which was processed with NMP—increases in device performances were observed upon the addition of the ferroelectric additives. Specifically, all the parameters, including JSC, VOC, and FF, were improved in the case of the device which was processed with P2. This resulted into champion PCE value of 11.02% was obtained (JSC = 20.30 mA cm−2, VOC = 0.812 V, and FF = 66.50%), which is by far the highest PCE reported to date for PTB7 derivative and fullerene combination-based OSCs using halogen-free solvent/additive process (see Fig. 5c in a later section). The JSC and FF improvements in the devices with the ferroelectric additives can be explained by the enhanced local built-in electric field of ferroelectric dipoles embedded within the active layer, as observed from the aforementioned polarization study. On the other hand, the 13

varied VOCs are most likely correlated with the tuned lowest-occupied molecular orbital (LUMO) offset by the ferroelectric additives implanted in the interface of nano p-n BHJs.[46-50] Based on the J−V curves obtained in dark conditions (inset of Fig. 3a), it was found that both as-cast control and the devices processed with P1 displayed a higher current density at reverse bias that led to lower diode rectification ratios compared to other systems. This finding reflects the relatively lower FFs of the two devices. Table 1. Summary of device parameters for BHJ OSCs with and without PVDF-based ferroelectric additives under AM 1.5G irradiation at 100 mW cm-2. Concentration Additive JSC (mA cm−2) VOC (V) FF (%) PCE (%) (vol%/wt%) 4.23 ± 0.44 As-casta 11.84 ± 0.68 0.790 ± 0.005 45.3 ± 1.5 (4.67) 9.96 ± 0.24 NMP onlya 3.0 vol% 19.13 ± 0.30 0.796 ± 0.003 65.4 ± 0.3 (10.20) 9.79 ± 0.31 P1a 1.5 wt% 19.83 ± 0.36 0.803 ± 0.001 61.5 ± 0.7 (10.10) 10.72 ± 0.30 P2a 1.5 wt% 20.01 ± 0.29 0.810 ± 0.002 66.1 ± 0.4 (11.02) 10.20 ± 0.30 P3a 2.0 wt% 19.75 ± 0.32 0.794 ± 0.001 65.0 ± 0.8 (10.50) 10.58 ± 0.25 P4a 1.5 wt% 19.78 ± 0.31 0.805 ± 0.002 66.4 ± 0.4 (10.83) 8.84 ± 0.20 b As-cast 17.01 ± 0.15 0.829 ± 0.004 62.1 ± 1.2 (9.04) 11.61 ± 0.15 b P2 1.0 wt% 18.94 ± 0.15 0.829 ± 0.002 73.9 ± 0.3 (11.76) *Data corresponds to the average value of 10 devices and deviation from its maximum. Data in parenthesis corresponds to the maximum values. aPTB7-Th:PC71BM and bPM6:IT-4F based OSC devices.

The optical effects are ruled out as possible mechanisms for the enhancement of performance given the generation of nearly identical optical spectra from the PTB7-Th:PC71BM host system upon the addition of ferroelectric additives (SI Figs. S15–S16). Compared to other systems, an aggressively large aggregation was observed in the case of P1, as evidenced by the fluorescence 14

microscopy images shown in the SI Fig. S6. This effect constitutes a rational reason for the observation of the relatively inferior performance of devices processed with P1. In addition, the external quantum efficiencies of all the devices displayed broad photo-responses in the range from 300 to 800 nm that matched closely the trends of the measured JSC values (Fig. 3c). To understand the charge recombination behaviour as a function of the ferroelectric additives used, the light intensity dependence of the J−V characteristics was measured for the optimized devices with and without ferroelectric additives (Figs. 3d and 3e). In principle, the relationship between the JSC values and the incident light intensity (Plight) values adhered to a power-law dependence, JSC ∝ Plightα, whereas the VOC values depended on the natural logarithm of Plight with a slope of nkT/q, where k = Boltzmann’s constant, T = temperature, q = elementary charge, and n is the ideality factor.[17,51-54] All the devices yielded increased and similar values for α (0.96 − 0.99) and n (1.66 − 2.11), thus suggesting that geminate or trap-assisted Shockley–Read–Hall recombination is the predominant mechanism for the loss of free-charge carriers. Note that the system processed with P2 yielded a value for α which was close to unity, and had the smallest n slope. This verified the small losses associated with all the types of recombination kinetics. The variations of the photo-generated current density (Jph) as a function of the effective voltage (Veff) were also plotted to evaluate the exciton dissociation and charge collection in the devices (Fig. 3f).[28,55-57] The Jph reached the saturation value (Jsat) at high Veff values (>0.3 V) for all the devices except for the as-cast control device. Under the short-circuit condition, the exciton dissociation probability (Pdiss = Jph/Jsat, where Jsat is the saturation photo-current density) of the as-cast control device was 88%, while distinctly higher Pdiss values of the order of ~97% were achieved in the cases of other devices (Table S6, SI). Taken together, these results suggest that more efficient exciton dissociations and charge collections occurred in the devices processed 15

with a) the use of NMP only, and b) ferroelectric additives. Potentially, these could contribute to their higher JSC values. The hole and electron mobilities of the blend films were also determined (Fig. S17 and Table S6, SI) by fitting the dark current to the model of a single carrier space-charge limited curve (SCLC), as described by the Mott–Gurney equation.[58-61] For the as-cast film, the hole/electron mobilities were calculated to be 8.96 × 10–5/9.56 × 10–5 cm2 V−1 s−1, whereas in the case of the film processed only with NMP, the corresponding values were respectively improved to 1.93 × 10– 4

/2.24 × 10–4 cm2 V−1 s−1. Upon the addition of the ferroelectric additives, both hole/electron

mobilities were further improved, especially for the film processed with P2 that had the highest hole/electron mobilities (2.94 × 10–4/3.93 × 10–4 cm2 V−1 s−1). This outcome could partially account for the lowest recombination loss as well as for the enhanced JSC and FF values, as discussed earlier. The changed trends in the charge dynamics and transport properties may not be sufficient in explaining the enhanced device performance upon the addition of a small amount of the ferroelectric additives because the ferroelectric dipoles sporadically form only small spots within the large active area, as depicted in the microscopy images in Fig. 2.

16

Fig. 4. Morphology and packing orientation in blend films. HR-TEM, AFM (height) images and GIXD patterns of PTB7–Th:PC71BM blend films with and without PVDF-based ferroelectric additives. (a), (g) As-cast, (b), (h) NMP only, (c), (i) P1, (d), (j) P2, (e), (k) P3, and (f), (l) P4.

The microstructure and surface morphology of the blend films were thoroughly investigated with the use of high-resolution transmission electron microscopy (HR–TEM), tapping-mode atomic force microscopy (AFM), and grazing incident wide-angle X-ray diffraction (GIXD). The HR– TEM images of the blend films (Fig. 4) with NMP only and those with ferroelectric additives show homogeneous morphologies with fine phase-separated domains compared to the as-cast blend film which contains oversized dark regions corresponding to PC71BM-rich domains. In 17

addition, based on the AFM images (see Fig. 4 and SI Fig. S18), the surface of the as-cast film was relatively coarser with a large root-mean-square (RMS) surface roughness of 8.04 nm, whereas the blend films with NMP only and ferroelectric additives exhibited a smooth surface and uniform morphology with small RMS values in the range of 1.04–2.13 nm. The AFM data agreed well with the TEM results. Shown in Fig. 4 and Fig. S19 (see SI) are GIXD images and line-cut profiles of the blend films, respectively. Fig. S20 (see SI) shows GIXD images of pure additive films. All the blend films exhibited preferentially face-on orientations with respect to the substrates, as evidenced by a) the intensified out-of-plane π−π stacking (010) peak at q ≈ 1.85 Å1

, and b) with a defused ring-like (100) peak at q ≈ 0.31 Å-1 (Table S7, SI). Additionally, the

locations of the diffraction peaks were nearly identical in all the cases, thus providing additional evidence that the improved OSC performance exhibited in the case of ferroelectric additives resulted from the ferroelectric dipoles rather than from the molecular packing and orientation changes.

Fig. 5. Device Performance. (a) J–V characteristics of PM6:IT-4F based non-fullerene BHJ OSCs with NMP only and P2 as ferroelectric additive under AM 1.5G irradiation at 100 mW cm-2. (b) Radar 18

plot of the device performances in representative bilayer OSCs with P2 in each layer for comparative study of its role in charge transport. (c) Comparison of the device performances for representative halogen-free processed OSCs recently reported in various literatures.

Table 2. Summary of device parameters for PM6:IT-4F based bilayer OSCs with and without PVDF-based ferroelectric additive (P2, 1 wt%) under AM 1.5G irradiation at 100 mW cm-2. Device Additive J (mA cm−2) VOC (V) FF (%) PCE (%) Structure (Donor: Acceptor) SC 10.60 ± 0.15 Bilayer 1a X:X 17.93 ± 0.30 0.828 ± 0.001 70.0 ± 1.1 (10.75) 11.06 ± 0.23 a Bilayer 2 O:X 18.41 ± 0.04 0.827 ± 0.001 72.6 ± 1.2 (11.29) 11.30 ± 0.17 Bilayer 3a X:O 18.47 ± 0.16 0.832 ± 0.001 73.5 ± 0.4 (11.47) 11.62 ± 0.21 Bilayer 4a O:O 18.93 ± 0.28 0.828 ± 0.002 74.0 ± 0.3 (11.83) *Data corresponds to the average value of 10 devices and deviation from its maximum. Data in parenthesis corresponds to the maximum values. aDevice structures for each bilayer system with/without P2 are illustrated in Fig. 5b.

In the final part of this study, the use of P2 ferroelectric additive was extended to PM6:IT-4F non-fullerene acceptor BHJ OSC, to further demonstrate the positive effect of the ferroelectric polarization that leads to the enhanced OSC performance (SI, Table S8). The use of 1.0 wt% optimal amount of P2 on PM6:IT-4F film improved overall PCE of 11.76% (JSC = 19.09 mA cm−2, VOC = 0.830 V, and FF = 74.20%) (Table 1 and Fig. 5a), compared to the control device without P2 (Table 1). In fact, the exact origin for the enhanced photovoltaic parameters (JSC, FF, and VOC) in any given BHJ type OSC is comparably complex since they highly depend on many factors such as the phase separation, interfacial area, and morphology of the donor and acceptor blend. Therefore, to clearly verify the role of ferroelectric additive in each donor and acceptor component, we also carried out the study of the P2’s effect on the p-n like PM6/IT-4F bilayer system; in concrete terms, putting its optimal amount into PM6 only layer, IT-4F only layer, and both layers of the bilayer devices, respectively (see Fig. 5b and Fig. S21 and Table 2). The 19

fabrication details of the BHJ and bilayer devices are provided in the Experimental section. The control bilayer device without P2 offered a PCE of 11.14% (JSC = 18.52 mA cm−2, VOC = 0.830 V, and FF = 72.5 %). As what was observed in the above studies, the JSC values increased upon the addition of P2 in any layers, due to the efficient charge transport property caused by the extra built-in electric field. In addition, the FF values are also enhanced in the bilayer devices with P2, especially for its addition into either IT-4F only layer or both layers rather than PM6 only layer. In principle, a charge transport process is predominately present in the acceptor regions, while the donor regions have the main role of exciton generation and diffusion at the interface. Thus, introducing the ferroelectric additive in the acceptor layer makes a more effective charge transport process, contributing to the observed higher FFs. Therefore, the best-performing bilayer OSC (11.83%) was obtained from the both layers with P2. Another interesting point is that the VOC values of the tested all bilayer devices are quite similar, due to the absence of nano p-n BHJs as unlike as the all BHJ devices above. Fig. 5c, summarized in Table S9, compares the highperformance OSCs processed from halogen-free processing systems in various types of the device structures, not only highlighting that our results are one of the highest performances reported thus far, but also demonstrating that the ferroelectric additive forms a versatile method of experimental control for improving the performances of various types of OSCs in an environment-friendly way.

3. Experimental Section Solar cell fabrication: PTB7-Th:PC71BM (1:1.5 ratio) and based devices were fabricated in the conventional device structure of glass/ITO/PEDOT:PSS/active layer/Al. On the pre-cleaned ITO substrate, a PEDOT:PSS layer (Clevios P VP AI. 4083) was spin-coated at 4000 rpm for 60 s 20

and then thermally annealed at 140 °C for 20 min. The substrates were immediately transferred to glovebox and the active layer was spin-coated yielding thickness in the range of 160-200 nm (see SI Fig. S22). It should be noted the thickness mention here is excluding thickness of crystalline ferroelectric polymer additives (its thickness is mentioned in hysteresis part of SI). The films were subsequently treated by vacuum annealing for 15 minutes. Finally, a 100 nm thick layer of Al was deposited using a shadow mask (device area: 0.13 cm2) under high vacuum (< 3 × 10−6 Pa). For active layer preparation, PTB7-Th:PC71BM (35 mg ml-1) were dissolved in anhydrous o-xylene and stirred at 60 °C for overnight. P1, P2, P3, and P4 were dissolved separately in NMP and stirred at 60 °C for overnight. Next day, different wt. % of additives were added separately in the active layer solution and again stirred at 60 °C for 2 hrs. For PM6:IT4F (1:1 ratio) based binary devices were fabricated in the conventional device structure of glass/ITO/PEDOT:PSS/active layer/PDINO/Al. Total active layer solution concentration of 20 mg mL-1 was dissolved in anhydrous o-xylene and stirred at 45 °C for overnight. Next day, different wt. % of P2 additive was added separately in the active layer solution and again stirred at 80 °C for less than 2 hrs. The active layer was spin-coated at 2500 rpm for 40 s and subsequently thermally annealed at 100 °C for 10 min. Thereafter, perylene-diimide (PDINO) layer (0.75 mg mL-1 in methanol) as electron transporting layer (ETL) is spin coated at 3000 rpm for 40s. For bilayer devices, conventional device architecture is adopted in the configuration of glass/ITO/PEDOT:PSS/PM6/IT4F/PDINO/Al. PM6 (20 mg mL-1) and IT4F (28 mg mL-1) was dissolved separately in anhydrous o-xylene and stirred at 45 °C for overnight. Next day, 1 wt. % of P2 additive were added in the solutions accordingly for fabricating four bilayer (B1, B2, B3, and, B4) systems and again stirred at 80 °C (< 2 hrs). For active layer, PM6 was first spin-coated 21

at 2200 rpm for 40 s and subsequently vacuum annealed for 15 minutes to remove any residual solvent. This solidifies the donor layer which significantly prevents its washing when the acceptor layer (IT-4F based solution) is spin-coated on it at 4000 rpm for 40 s. The devices are subsequently at 100 °C for 10 min and thereafter, PDINO layer (0.75 mg mL-1 in methanol) as ETL is spin coated at 3000 rpm for 40s.

4. Conclusions In summary, high-performance OSCs were successfully demonstrated with a built-in local electric field induced by a simple addition of ferroelectric additives (P1, P2, P3, and P4) in both fullerene and non-fullerene acceptor-based active layers. The recrystallization of the ferroelectric additives with an o-xylene/NMP pair featuring halogen-free system into the active-layer matrix was a driving force for vitalising their ferroelectric polarizations without poling treatments, as evidenced by the advent of the rectangular-shaped P–E hysteretic loops following the use of this cosolvent combination. Outstanding PCEs as high as 11.02% and 11.76% were obtained from the fullerene and non-fullerene acceptor BHJ OSCs processed with P2, respectively, which was accounted for by the enhanced permanent polarizations that originated from the highly ordered,

β-phase microstructure. The in-depth structural, electrical, and morphological analyses performed in this study indicated that the ferroelectric dipoles locally embedded in the activelayer matrix—rather than the morphological changes—can ultimately affect the device performance. Besides, to further clarify the mechanistic understanding of how it impacts individual donor and acceptor component in OSCs, we extended the use of P2 into p-n like PM6/IT-4F bilayer system and achieved11.83% bilayer device. The PCEs achieved in this study are fairly comparable to the highest value reported for any types of halogen-free processed 22

OSCs. This study can facilitate the practical applications of many ferroelectric materials for their use as additives in various optoelectronic applications as an effective strategy for realizing desired properties including environment-friendly OSC technologies, even as the considerations expend beyond that of raw efficiency. Supplementary Information Supplementary Information is available from the Elsevier or from the author.

Acknowledgements C. Y. conceptualized the project. T. K. carried out the experiments and performed the device fabrication and other experiments, as well as analysed the data. S. J. synthesized graft copolymers. T. K. and S. J. contributed equally to this work. H. –P. K. performed hysteresis measurement. J. W. L measured dielectric constant. J. Lee synthesized IT-4F. S. M. Lee and M. Jeong synthesized PM6. T. K. and C. Y. wrote the manuscript. This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (2018R1A2A1A05077194), Center for Advanced Soft-Electronics funded by the Ministry of Science and ICT as Global Frontier Project (2012M3A6A5055225), Wearable Platform Materials Technology Center (WMC; 2016R1A5A1009926) funded by the National Research Foundation of Korea (NRF) Grant by the Korean Government (MSIT), and the Research Project Funded by Ulsan City (1.190099) of UNIST (Ulsan National Institute of Science & Technology). GIXD measurements at PLS-II 6D UNIST-PAL beamline and 9A beamline were supported in part by MEST, POSTECH, and UNIST UCRF. The authors declare no competing financial interests. Correspondence and requests for materials should be addressed to C.Y.

23

References [1]

[2]

[3] [4]

[5]

[6]

[7]

[8]

[9]

[10]

[11] [12] [13]

[14]

[15]

Karak, S., Page, Z. A., Tinkham, J. S., Lahti, P. M., Emrick, T., Duzhko, V. V. Raising efficiency of organic solar cells with electrotropic additives. Appl. Phys. Lett. 106 (10) (2015) 103303. Li, M., Wondergem, H. J., Spijkman, M.-J., Asadi, K., Katsouras, I., Blom, P. W. M., de Leeuw, D. M., Revisiting the δ-phase of poly(vinylidene fluoride) for solution-processed ferroelectric thin films. Nat. Mater. 12 (5) (2013) 433. Li, Q. & Wang, Q., Ferroelectric Polymers and Their Energy-Related Applications. Macromol. Chem. Phys. 217 (11) (2016) 1228-1244. Ma, W., Zhang, J., Chen, S. & Wang, X., Crystalline Phase Formation of Poly(vinylidene fluoride) from Tetrahydrofuran/N,N‐dimethylformamide Mixed Solutions. J. Macromol. Sci. B. 47 (3) (2008) 434-449. Manna, S. & Nandi, A. K., Piezoelectric β Polymorph in Poly(vinylidene fluoride)Functionalized Multiwalled Carbon Nanotube Nanocomposite Films. J. Phys. Chem. C 111 (40) (2007) 14670-14680. Murayama, N., Oikawa, T., Katto, T. & Nakamura, K. i., Persistent polarization in poly(vinylidene fluoride). II. Piezoelectricity of poly(vinylidene fluoride) thermoelectrets. J. Polym. Sci. Poly. Phys. Ed. 13 (5) (1975) 1033-1047. Wolff, S., Jirasek, F., Beuermann, S. & Türk, M., Crystal phase transformation of α into β phase poly(vinylidene fluoride) via particle formation caused by rapid expansion of supercritical solutions. RSC Adv. 5 (82) (2015) 66644-66649. Lee, J. W., Cho, H. J., Chun, J., Kim, K. N., Kim, S., Ahn, C. W., Kim, I. W., Kim, J.-Y., Kim, S.-W., Yang, C., Baik, J. M., Robust nanogenerators based on graft copolymers via control of dielectrics for remarkable output power enhancement. Sci. Adv. 3 (5) (2017) 1602902. Cao, Y., Figueroa, J., Li, W., Chen, Z., Wang, Z.L. and Sepúlveda, N., Understanding the dynamic response in ferroelectret nanogenerators to enable self-powered tactile systems and human-controlled micro-robots. Nano Energy 63 (2019) 103852. Cao, Y., Li, W., Figueroa, J., Wang, T., Torres, D., Wang, C., Wang, Z.L. and Sepúlveda, N., Impact-activated programming of electro-mechanical resonators through ferroelectret nanogenerator (FENG) and vanadium dioxide. Nano energy, 43 (2018) 278-284. McFee, J. H., Bergman, J. G. & Crane, G. R., Pyroelectric and nonlinear optical properties of poled polyvinylidene fluoride films. Ferroelectrics 3 (1) (1972) 305-313. Nakamura, K. i. & Wada, Y., Piezoelectricity, pyroelectricity, and the electrostriction constant of poly(vinylidene fluoride). J. Polym. Sci., Part A-2 9 (1) (1971) 161-173. Nalwa, K. S., Carr, J. A., Mahadevapuram, R. C., Kodali, H. K., Bose, S., Chen, Y., Petrich, J. W., Ganapathysubramanian, B., Chaudhary, S., Enhanced charge separation in organic photovoltaic films doped with ferroelectric dipoles. Energy Environ. Sci. 5 (5) (2012) 7042-7049. Yu, S., Zheng, W., Yu, W., Zhang, Y., Jiang, Q., Zhao, Z., Formation Mechanism of βPhase in PVDF/CNT Composite Prepared by the Sonication Method. Macromolecules 42 (22) (2009) 8870-8874. Yuan, Y., Xiao, Z., Yang, B. & Huang, J., Arising applications of ferroelectric materials in photovoltaic devices. J. Mater. Chem. A 2 (17) (2014) 6027-6041. 24

[16] [17]

[18]

[19]

[20]

[21]

[22] [23] [24] [25]

[26]

[27]

[28]

[29]

[30]

Chen, X., Han, X. & Shen, Q.-D., PVDF-Based Ferroelectric Polymers in Modern Flexible Electronics. Adv. Electron. Mater. 3 (5) (2017) 1600460. 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 high efficiency and photovoltage. Nat. Photonics 9 (2015) 174. Yuan, Y., Reece, T. J., Sharma, P., Poddar, S., Ducharme, S., Gruverman, A., Yang, Y., Huang, J., Efficiency enhancement in organic solar cells with ferroelectric polymers. Nat. Mater. 10 (2011) 296. Abdalla, S., Obaid, A. & Al-Marzouki, F. M., Preparation and characterization of poly(vinylidene fluoride): A high dielectric performance nano-composite for electrical storage. Results Phys. 6 (2016) 617-626. Bohlén, M. & Bolton, K., Conformational studies of poly(vinylidene fluoride), poly(trifluoroethylene) and poly(vinylidene fluoride-co-trifluoroethylene) using density functional theory. Phys. Chem. Chem. Phys. 16 (25) (2014) 12929-12939. Buensuceso, J. G. G., Cuan, A., Castillo, R. V. & Cortes-Romero, C. M., DFT study of PVDF as polymer composite precursor in 12th Congreso Internacional de Ingeniería (CONIIN), 1-6 (IEEE) (2016). Lovinger, A. J., Unit cell of the γ phase of poly(vinylidene fluoride). Macromolecules 14 (2) (1981) 322-325. Pei, Y. & Zeng, X. C., Elastic properties of poly(vinyldene fluoride) (PVDF) crystals: A density functional theory study. J. Appl. Phys. 109 (9) (2011) 093514. Zhang, Q. M., Bharti, V. & Kavarnos, G., Poly(Vinylidene Fluoride) (PVDF) and its Copolymers in Encyclopedia of Smart Materials, Wiley: New York: (2002) 807– 825. Guan, F., Wang, J., Yang, L., Tseng, J.-K., Han, K., Wang, Q., Zhu, L., ConfinementInduced High-Field Antiferroelectric-like Behavior in a Poly(vinylidene fluoride-cotrifluoroethylene-co-chlorotrifluoroethylene)-graft-polystyrene Graft Copolymer. Macromolecules 44 (7) (2011) 2190-2199. Guan, F., Yang, L., Wang, J., Guan, B., Han, K., Wang, Q., Zhu, L., Confined Ferroelectric Properties in Poly(Vinylidene Fluoride-co-Chlorotrifluoroethylene)-graftPolystyrene Graft Copolymers for Electric Energy Storage Applications. Adv. Funct. Mater. 21 (16) (2011) 3176-3188. Lee, J. P., Lee, J. W., Yoon, B.-K., Hwang, H. J., Jung, S., Kim, K. A., Choi, D., Yang, C., Baik, J. M., Boosting the energy conversion efficiency of a combined triboelectric nanogenerator-capacitor. Nano Energy 56 (2019) 571-580. Zhang, M., Guo, X., Ma, W., Ade, H. & Hou, J., A Large‐Bandgap Conjugated Polymer for Versatile Photovoltaic Applications with High Performance. Adv. Mater. 27 (31) (2015) 4655-4660. Zhao, W., Li, S., Yao, H., Zhang, S., Zhang, Y., Yang, B. & Hou, J., Molecular optimization enables over 13% efficiency in organic solar cells. J. Am. Chem. Soc. 139 (21) (2017) 7148-7151. Fan, Q., Su, W., Wang, Y., Guo, B., Jiang, Y., Guo, X., Liu, F., Russell, T.P., Zhang, M. & Li, Y., Synergistic effect of fluorination on both donor and acceptor materials for high performance non-fullerene polymer solar cells with 13.5% efficiency. Sci. China Chem. 61 (5) (2018) 531-537.

25

[31]

[32]

[33]

[34] [35] [36]

[37]

[38]

[39] [40]

[41] [42]

[43] [44] [45] [46]

[47]

Yuan, J., Zhang, Y., Zhou, L., Zhang, G., Yip, H.L., Lau, T.K., Lu, X., Zhu, C., Peng, H., Johnson, P.A. & Leclerc, M., Single-junction organic solar cell with over 15% efficiency using fused-ring acceptor with electron-deficient core. Joule 3 (4) (2019) 1140-1151. Gao, B., Yao, H., Hou, J., Yu, R., Hong, L., Xu, Y. & Hou, J., Multi-component nonfullerene acceptors with tunable bandgap structures for efficient organic solar cells. J. Mater. Chem. A 6 (46) (2018) 23644-23649. Kumari, T., Lee, S. M. & Yang, C., Cubic-Like Bimolecular Crystal Evolution and over 12% Efficiency in Halogen-Free Ternary Solar Cells. Adv. Funct. Mater. 28 (19) (2018) 1707278. Chen, Y., Chen, X., Zhou, D., Shen, Q.-D. & Hu, W., Low-temperature crystallization of P(VDF-TrFE-CFE) studied by Flash DSC. Polymer 84 (2016) 319-327. Feng, T., Xie, D., Zang, Y., Wu, X., Ren, T., Pan, W., Temperature Control of P(VDFTrFE) Copolymer Thin Films. Integr. Ferroelectrics 141 (1) (2013) 187-194. Fu, Z., Xia, W., Chen, W., Weng, J., Zhang, J., Zhang, J., Jiang, Y., Zhu, G., Improved Thermal Stability of Ferroelectric Phase in Epitaxially Grown P(VDF-TrFE) Thin Films. Macromolecules 49 (10) (2016) 3818-3825. Lang, M. H. & Zhang, J., Morphology and properties of poly(vinylidene fluoride) (PVDF)/ethylene–vinyl acetate copolymer (EVA) blends. Plast. Rubb. Compos. 43 (1) (2014) 8-14. Le, M. Q., Capsal, J.-F., Galineau, J., Ganet, F., Yin, X., Yang, M., Chateaux, J.-F., Renaud, L., Malhaire, C., Cottinet, P.-J., Liang, R., All-organic electrostrictive polymer composites with low driving electrical voltages for micro-fluidic pump applications. Sci. Rep. 5 (2015) 11814. Teyssedre, G., Bernes, A. & Lacabanne, C., DSC and TSC study of a VDF/TrFE copolymer. Thermochimica Acta 226 (1993) 65-75. Teyssèdre, G. & Lacabanne, C. Study of the thermal and dielectric behavior of P(VDFTrFE) copolymers in relation with their electroactive properties. Ferroelectrics 171 (1) (1995) 125-144. Feng, T., Xie, D., Zang, Y., Wu, X., Ren, T., Pan, W., Temperature Control of P (VDFTrFE) Copolymer Thin Films. Integr. Ferroelectrics 141 (1) (2013) 187-194. Kim, D.-W., Kim, J.-H., Kim, J.-N., Park, H.-J., Jeon, H.-S., Park, B.-E., Characterization of Metal-Ferroelectric-Semiconductor Structure using Ferroelectric Polymer Polyvinylidene Fluoride-Trifluoroethylene (PVDF-TrFE)(51/49). Integr. Ferroelectrics 98 (1) (2008) 121-127. Sarkar, R. & Kundu, T. K., Density functional theory studies on PVDF/ionic liquid composite systems. J. Chem. Sci 130 (8) (2018) 115. Loidl, A., Krohns, S., Hemberger, J. & Lunkenheimer, P., Bananas go paraelectric. J. Phys.: Condens. Matter 20 (19) (2008) 191001. Scott, J. F., Ferroelectrics go bananas. J. Phys.: Condens. Matter 20 (2) (2008) 021001. Yang, B., Yuan, Y., Sharma, P., Poddar, S., Korlacki, R., Ducharme, S., Gruverman, A., Saraf, R. & Huang, J., Tuning the energy level offset between donor and acceptor with ferroelectric dipole layers for increased efficiency in bilayer organic photovoltaic cells. Adv. Mater. 24 (11) (2012) 1455-1460. Zojer, E., Taucher, T.C. & Hofmann, O.T., The Impact of Dipolar Layers on the Electronic Properties of Organic/Inorganic Hybrid Interfaces. Adv. Mater. Interfaces (2019) 1900581. 26

[48] [49]

[50] [51]

[52]

[53]

[54]

[55]

[56]

[57]

[58]

[59]

[60]

[61]

Lim, K.G., Ahn, S. & Lee, T.W., Energy level alignment of dipolar interface layer in organic and hybrid perovskite solar cells. J. Mater. Chem. C, 6 (12) (2018) 2915-2924. Zhu, L., Wang, L., Pan, C., Chen, L., Xue, F., Chen, B., Yang, L., Su, L. & Wang, Z.L., Enhancing the efficiency of silicon-based solar cells by the piezo-phototronic effect. ACS Nano, 11 (2) (2017) 1894-1900. Wen, X., Wu, W. and Wang, Z.L., Effective piezo-phototronic enhancement of solar cell performance by tuning material properties. Nano Energy, 2 (6) (2013) 1093-1100. Gao, L., Zhang, Z.-G., Xue, L., Min, J., Zhang, J., Wei, Z., Li, Y., All-Polymer Solar Cells Based on Absorption-Complementary Polymer Donor and Acceptor with High Power Conversion Efficiency of 8.27%. Adv. Mater. 28 (9) (2016) 1884-1890. Kyaw, A. K. K., Wang, D. H., Gupta, V., Leong, W. L., Ke, L., Bazan, G. C., Heeger, A. J., Intensity Dependence of Current–Voltage Characteristics and Recombination in HighEfficiency Solution-Processed Small-Molecule Solar Cells. ACS Nano 7 (5) (2013) 4569-4577. Riedel, I., Parisi, J., Dyakonov, V., Lutsen, L., Vanderzande, D., Hummelen, J. C., Effect of Temperature and Illumination on the Electrical Characteristics of Polymer–Fullerene Bulk-Heterojunction Solar Cells. Adv. Funct. Mater. 14 (1) (2004) 38-44. Schilinsky, P., Waldauf, C. & Brabec, C. J., Recombination and loss analysis in polythiophene based bulk heterojunction photodetectors. Appl. Phys. Lett. 81 (20) (2002) 3885-3887. Guo, B., Li, W., Guo, X., Meng, X., Ma, W., Zhang, M., Li, Y., High Efficiency Nonfullerene Polymer Solar Cells with Thick Active Layer and Large Area. Adv. Mater. 29 (36) (2017) 1702291. Nian, L., Gao, K., Liu, F., Kan, Y., Jiang, X., Liu, L., Xie, Z., Peng, X., Russell, T. P., Ma, Y., 11% Efficient Ternary Organic Solar Cells with High Composition Tolerance via Integrated Near-IR Sensitization and Interface Engineering. Adv. Mater. 28 (37) (2016) 8184-8190. Jiang, C., Jing, L., Huang, X., Liu, M., Du, C., Liu, T., Pu, X., Hu, W. and Wang, Z.L., Enhanced solar cell conversion efficiency of InGaN/GaN multiple quantum wells by piezo-phototronic effect. ACS Nano, 11 (9) (2017) 9405-9412. Armin, A., Juska, G., Ullah, M., Velusamy, M., Burn, P. L., Meredith, P., Pivrikas, A., Balanced Carrier Mobilities: Not a Necessary Condition for High-Efficiency Thin Organic Solar Cells as Determined by MIS-CELIV. Adv. Energy Mater. 4 (4) (2014) 1300954. Goh, C., Kline, R. J., McGehee, M. D., Kadnikova, E. N. & Fréchet, J. M. J., Molecularweight-dependent mobilities in regioregular poly(3-hexyl-thiophene) diodes. Appl. Phys. Lett. 86 (12) (2005) 122110. Kumari, T., Lee, S. M., Kang, S.-H., Chen, S. & Yang, C., Ternary solar cells with a mixed face-on and edge-on orientation enable an unprecedented efficiency of 12.1%. Energy Environ. Sci. 10 (1) (2017) 258-265. Kumari, T., Lee, S. M., Lee, K. C., Cho, Y. & Yang, C., Harmonious Compatibility Dominates Influence of Side-Chain Engineering on Morphology and Performance of Ternary Solar Cells. Adv. Energy Mater. 8 (22) (2018) 1800616.

27

Vitae

Dr. Tanya Kumari received her Ph.D. from Ulsan National Institute of Science and Technology (UNIST, South Korea) in Energy Engineering in 2019 under the supervision of Prof. Changduk Yang. She is currently working as a postdoctoral research associate in UNIST. Her main research focus is on device physics of organic solar cells for the development of stretchable devices with high performance.

Sungwoo Jung is in the combined master's and doctorate course for Energy Engineering in Ulsan National Institute of Science and Technology (UNIST). He received BS in 2017 in Energy Engineering from UNIST. He joined Professor Changduk Yang's group in 2017 and his main research focus is on synthesizing organic materials which will be applied for optoelectronic devices and triboelectric nanogenerator.

28

Yongjoon Cho is in the combined master’s and doctorate course for department of energy engineering in Ulsan National Institute of Science and Technology (UNIST). He received his BS in 2016 in Energy Conversion & Storage and Nanochemistry from UNIST. He joined Professor Changduk Yang’s group in 2016 and his main research interest is focusing on synthesizing novel organic materials for organic photovoltaics, organic field-effect transistors and organic photocatalysts.

Hwang-Pill Kim is currently the combined M.S./Ph.D. candidate in the School of Materials Science and Engineering at Ulsan National Institute of Science and Technology (UNIST), Ulsan, Republic of Korea. He got his B.S degree in Material Science and Engineering in 2015 from UNIST. His research field is about the engineering of ferroelectric domains in piezoelectric single crystals under the supervision of Prof. Wook Jo.

Jae Won Lee received his Ph.D. from Ulsan National Institute of Science and Technology (UNIST) in School of Materials Science and Engineering in 2019. He is currently working as a post-doctor fellow in UNIST. His research focuses on development of triboelectric generators by 29

using composite and interlayer design based polymer for sustainable energy conversions, high output performance device, and fundamental study.

Jiyeon Oh is in the combined master's and doctorate course for Energy Engineering in Ulsan National Institute of Science and Technology (UNIST, south Korea). Before coming to UNIST, she received BS in 2016 in Advanced Materials Chemistry at Korea University. She joined Professor Changduk Yangs group in 2017. Her main research is device physics of organic solar cells.

Jungho Lee is in the combined master's and doctorate course for Energy Engineering in Ulsan National Institute of Science and Technology (UNIST). He received his BS in 2013 in Energy Conversion & Storage and Chemical Engineering from UNIST. He then joined Prof. Changduk Yang's group since 2013 and his main research focus is on development of novel organic materials and device physics in organic solar cells.

30

Sang Myeon Lee is enrolled in the combined master's and doctorate course in Department of Energy Engineering at Ulsan National Institute of Science and Technology (UNIST, South Korea). He received his B.S. in 2014 in Department of Energy Engineering from UNIST. In the same year, he has joined Prof. Changduk Yang's group, and his main research focus is on the synthesis of conjugated organic semiconductors for electronic applications.

Mingyu Jeong received his B.S. degree in Nano-Bioscience and Chemical Engineering in 2015 from Ulsan National Institute of Science (UNIST). Currently, he is in the combined master's and Ph.D program in Prof. Changduk Yang’s group and his main research focus is synthesis of organic materials for organic photovoltaics and perovskite solar cells.

Dr. Jeong Min Baik is Associate Professor in School of Materials Science and Engineering, Ulsan National Institute of Science and Technology (UNIST). He received his Ph.D. from Pohang University in Department of Materials Science and Engineering in 2006. His recent research interest is focused on the synthesis of nanomaterials and nanostructures such as nanoparticles, nanowires, nanolayers, and nanopores for the applications of Energy-Conversion Devices and Nano-photonic Devices. Particular interests are concerned with the development of Piezoelectric/Triboelectric Nanogenerators and Artificial Photosynthesis.

31

Dr. Wook Jo is a Professor at the School of Materials Science and Engineering and the head of Jülich-UNIST Joint Leading Institute for Advanced Energy Research, Ulsan National Institute of Science and Technology (UNIST), Ulsan, Korea. Prior to joining the faculty of UNIST in 2014, he had served as a group leader for the Processing of Ferroelectrics lab, Technische Universität Darmstadt, Germany as of 2007. His recent research focuses on the functional properties of ferroelectric materials with a special emphasis on the lead-free piezoceramics and relaxor ferroelectrics as well as processing and characterization of room-temperature ferromagneticallycoupled ferroelectrics.

Prof. Changduk Yang obtained his Ph.D. degree from the Max Planck Institute for Polymer Research (Germany) in 2006 under Prof. Klaus Müllen. He finished his postdoctoral training in 2009 at the University of California Santa Barbara (U.S.) with Prof. Fred Wudl. In March 2009, he has joined UNIST. He serves on the editorial board for Polymer (South Korea) and works as a fulltime professor at UNIST. His research focus is on the development of organic semiconducting materials and multifunctional molecule-related self-assembly for various applications, including organic optoelectronics, triboelectric generators, and secondary batteries.

32

Article type: Full Paper

A Built-in Electric Field Induced by Ferroelectrics Increases Halogen-Free Organic Solar Cell Efficiency in Various Device Types Tanya Kumari,a† Sungwoo Jung, a† Yongjoon Cho,a Hwang-Pill Kim,b Jae Won Lee,c Jiyeon Oh,a Jungho Lee,a Sang Myeon Lee,a Mingyu Jeong,a Jeong Min Baik,c Wook Jo,b Changduk Yanga* a

Department of Energy Engineering, School of Energy and Chemical Engineering, Perovtronics Research Center,

Low Dimensional Carbon Materials Center, Ulsan National Institute of Science and Technology (UNIST), 50 UNIST-gil, Ulju-gun, Ulsan 44919, Republic of Korea. b

School of Materials Science and Engineering, Ulsan National Institute of Science and Technology (UNIST), 50

UNIST-gil, Ulju-gun, Ulsan 44919, Republic of Korea. c

School of Materials Science and Engineering, KIST-UNIST Ulsan Center for Convergent Materials, Ulsan

National Institute of Science and Technology (UNIST), 50 UNIST-gil, Ulju-gun, Ulsan 44919, Republic of Korea. E-mail: [email protected]

Funding This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (2018R1A2A1A05077194), Center for Advanced SoftElectronics funded by the Ministry of Science and ICT as Global Frontier Project (2012M3A6A5055225),

Wearable

Platform

Materials

Technology

Center

(WMC;

2016R1A5A1009926) funded by the National Research Foundation of Korea (NRF) Grant by the Korean Government (MSIT), and the Research Project Funded by Ulsan City (1.190099) of UNIST (Ulsan National Institute of Science & Technology). GIXD measurements at PLSII 6D UNIST-PAL beamline and 9A beamline were supported in part by MEST, POSTECH, and UNIST UCRF. The authors declare no competing financial interests.

1