A theorectical design of performant chlorinated benzothiadiazole-based polymers as donor for organic photovoltaic devices

Organic Electronics 61 (2018) 46–55

Contents lists available at ScienceDirect

Organic Electronics journal homepage: www.elsevier.com/locate/orgel

A theorectical design of performant chlorinated benzothiadiazole-based polymers as donor for organic photovoltaic devices

T

Zhi-Wen Zhaoa, Qing-Qing Pana, Yun Genga,∗∗, Shui-Xing Wub, Min Zhanga, Liang Zhaoa, Zhong-Min Sua,∗ a b

Institute of Functional Material Chemistry, Faculty of Chemistry, Northeast Normal University, Chang Chun, 130024, Jilin, PR China School of Chemistry and Chemistry Engineering, Hainan Normal University, Haikou, 571158, PR China

A R T I C LE I N FO

A B S T R A C T

Keywords: Organic photovoltaic voltages (OPVs) Donor Chlorination effect Theoretical design Interfacial process

Chlorination is often adopted in the synthesis of donor materials considering the obvious improvement on efficiency of organic photovoltaic (OPV) device. Thus the chlorination effect was firstly probed based on density functional theory (DFT) calculation, which shows that the chlorinated benzothiadiazole connected with two thiophene rings (BTClTT) fragment although has larger optical bandgap and relatively blue-shifted spectrum, its deeper frontier molecular orbital (FMO) energy level, larger charge transfer distance (DCT) and much more transferred charges (CT), obvious negative natural population analysis (NPA) charges, larger twisted dihedral angles and smaller value of BLA indicate its favorable properties for the application in OPVs. Then, we designed a series of donor materials 2–7 adopting two strategies based on the experimental polymer 1 which contains BTClTT fragment. The calculated physical parameters characterizing OPV performance such as open-circuit voltage (VOC) and absorption-properties manifest that the design through side chains modifications for 2–5 may have slight influence on their cell performance while the backbone replacements for 6 and 7 exhibit prominent enhancement whether on VOC and photon-absorption property or on the charge separation ability. Therefore, our calculations suggest feasible strategies of designing donor materials for the chlorinated BT-based donor materials in OPVs. We hope our work can provide some guidelines for the future study on chlorinated donor materials.

1. Introduction Organic photovoltaic voltages (OPVs) constructed with polymer as donor (D) and PC71BM as acceptor (A) with a number of merits including low cost, transparency, flexibility, light-weight and large-area have attracted mounting interests, as power conversion efficiency (PCE) of OPVs has made some improvements in recent years [1–7]. And considerable efforts have been put on enhancing the PCE of OPVs [8–13]. Importantly, molecular design and optimization of both D and A are widely adopted in recent advances [14,15], among which introducing substituent in donor polymers has aroused much attention [16,17]. Particularly, fluorinations have been intensively used in donor materials to improve performance of OPVs and made a bit of achievements in PCE [18–21], yet fluorinated molecules with higher prices but lower yields in the reactions limit their commercial manufacture in a degree. Recently, some scientists have turned their eyes to chlorinated molecules. Early in 2009, Bao et al. pointed out that chlorinated semiconductors have similar or superior properties in electron mobility

∗

Corresponding author. Corresponding author. E-mail addresses: [email protected] (Y. Geng), [email protected] (Z.-M. Su).

∗∗

https://doi.org/10.1016/j.orgel.2018.06.047 Received 11 April 2018; Received in revised form 25 June 2018; Accepted 28 June 2018 Available online 30 June 2018 1566-1199/ © 2018 Published by Elsevier B.V.

and ambient stability than fluorinated ones. Confirmed by their calculations and measurements, chlorinated molecules tend to have lower lowest unoccupied molecular orbital (LUMO) energy level than fluorinated molecules [22]. From then on, chlorinated molecules have been gradually investigated on OPVs. Jian Pei et al. synthesized molecules Cl-IIDT and F-IIDT as D to construct PC71BM–based OPVs, which present the PCE of 4.60% and 1.19%, respectively. They reported that chlorinated molecules with better performance than fluorinated molecules is ascribed to their larger torsional angles which could reduce the crystallization tendency [23]. In 2015, Hao-Li Zhang and coworkers demonstrated that the fluorinated and chlorinated donors present similar frontier orbital energy levels and optical absorption [24]. Recently, Feng He and coworkers reported a PCE of 9.11% for OPVs constructed with chlorinated polymers and PC71BM [25]. And then they designed and synthesized another chlorinated benzothiadiazolebased polymer reaching a PCE of 8.21%. They pointed out that chlorination could lower the highest occupied molecular orbital (HOMO) energy level and subtly tuned the optical bandgap, which meant the

Organic Electronics 61 (2018) 46–55

Z.-W. Zhao et al.

investigated polymers were optimized under PBE0 functional and 6311G (d,p) basis set with no imaginary frequencies allowing for that PBE0 functional is proper to thiophene derivatives in some researches [45,46]. For an accurate description of optical properties, functionals B3LYP, PBE0, CAM-B3LYP, BHandHLYP and M06-2X were employed and the result computed at TD-PBE0/6-311G (d,p) level is closer to the experimental data [25], which are listed in the supporting information. Thus, all of the optical properties were calculated at this level and absorption spectra were simulated by GaussSum 3.0 software [47]. As for interface models, fifteen starting structures, in which the top panel of donor was plotted with three selected positions (above D unit, DA unit or A unit) for acceptor, and there are five D/A stacking patterns for each position as shown in Fig. 1 (taking the position above DA unit as example) and Fig. S1 in supporting information, were chosen with initial distance between donor and PC71BM as 3.5 Å and further optimized at B3LYP/6-311G (d,p) level. While for more information about initial models and the optimized interface models would be detailed in section 3.2.3. Based on the optimized structures, we calculated the counterpoise-corrected total interaction energies, which demonstrate that the Face-onD-Conf-3D configuration is favored as it shows strongest interaction energy among these fifteen configurations. Thus, the following investigated systems adopted this configuration for further calculations. Moreover, the excited-state properties of the interface model were calculated at TD-CAM-B3LYP/6-311G (d,p) level, in view of its better description on the charge transfer states verified by many theoretical works [48,49]. All of the calculations mentioned above were performed in the Gaussian 09 program [50]. Beisdes, We identified charge transfer (CT) states by Multiwfn software [51] based on TD-DFT calculations for interface model. Meanwhile, the charge transfer rate kinter-CT and charge recombination rate kinter-CR were assessed by Marcus charge transfer theory [52], which is detailed in supporting information.

potential application of chlorinated polymers in OPVs [26]. In these excellent chlorinated polymers mentioned above, the chlorinated benzothiadiazole was always adopted, indicating that it may be an essential building block in this kind of donor materials. Since chlorinated polymers exhibit advanced performance in OPVs and the chlorinated benzothiadiazole is one essential unit in these donors, what are the advantages of the chlorinated benzothiadiazole compared to benzothiadiazole, and how can we improve the capability of this kind of donors containing chlorinated benzothiadiazole unit in OPVs? These questions arouse our interest in exploring their performance in OPV. A systemic theoretical investigation is necessary, since numerous theoretical works focusing on OPVs have been carried out and already demonstrated to be helpful to understanding the mechanism and improving efficiency in recent years [27–30]. For instance, recent studies have explored interfacial mechanism [31–33], charge transport process [34,35] and donor/acceptor (D/A) arrangement [36,37]. Also, other works have concentrated on designing donor and acceptor materials [38–40], investigating on electron interfacial process and so forth [41,42]. Therefore, it is meaningful to find the advantage of chlorinated benzothiadiazole in OPVs and design more efficient donors based on this special unit from theoretical perspective. In this work, we firstly probe the chlorination effect in benzothiadiazole (BTT) fragment on the photoelectric properties in detail to explore the importance of chlorination in further donor-material design, since OPVs constructed with Benzothiadiazole (BT)-based polymers and fullerenes are widely reported. Then based on one chlorinated polymer 1 constructed with benzo [1,2-b:4,5-b]dithiophene(BDT) unit and chlorinated benzothiadiazole fragment BTTCl which was reported a PCE of 9.11% for OPVs with PC71BM as acceptor, we designed four polymers depicted in Scheme 1 trying to obtain more efficient donor materials. However, the tiny improvement in cell performance indicates our design strategy only through changing side units of BDT is not very effective. Thus we further took the replace of the whole BDT unit by thiophene analogue cyanomethylene-CPDT for 6 and cyclopenta [2,1b:3,4-b0]dithiophen-4-one for 7 with the aim of finding available design for donor. It is gratifying that the designed molecules 6 and 7 presented in Scheme 1 realize a better balance between the open-circuit voltage (VOC) and short circuit density (JSC) and thus improve the ability of charge separation significantly. Therefore, our calculations indicate that the strategy of designing donor materials is feasible and chlorinated BT-based donor materials could be optimized or designed to improve performance of OPVs. We hope our work can provide some guidelines for the future study on chlorinated donor materials.

3. Results and discussions 3.1. A systemic theoretical investigation on the effect of chlorination in BTTT unit

2. Computational methods

As mentioned above, the chlorinated benzothiadiazole is an essential building block in these excellent chlorinated polymers. A detailed investigation about it will favor our understanding of the chlorination and further effective design of this kind of materials. Therefore, we firstly analyzed the fragment BTClTT contained in chlorinated polymer 1 and compared it with non-chlorinated fragment BTTT to gain insight into the chlorinated effect in this kind of donor materials.

For chlorinated polymers 1–7, all of the branched chains were replaced by alkyl to save computational cost [43,44]. Besides, frontier molecular orbital (FMO) energy levels were calculated as our previous work [45]. All of the ground-state structures of building blocks and

3.1.1. Geometric and electronic properties DFT and TD-DFT methods were applied to analyze structure-property relationship to gain insight into photoelectric properties of benzothiadiazole building blocks presented in Scheme 2. The dihedral

Scheme 1. Chemical structures of investigated polymers 1–7. 47

Organic Electronics 61 (2018) 46–55

Z.-W. Zhao et al.

Fig. 1. Schematic representation of initial configurations for D/A complex. The top panel of donor was plotted with three selected positions (above D unit, DA unit or A unit) for acceptor. The underneath presents five D/A stacking patterns for one selected position (above DA unit).

angels and bond-length alternation (BLA = 1/8(C2-C3+C4-C5+C6C7+C8-C9+C9-C10 + C10-C5+C8-C11 + C12-C13)-1/6(C1-C2+C3C4+C5-C6+C7-C8+C11-C12 + C13-C14) across the fragment backbone listed in Table 1 illustrate the difference of geometric structures between blocks BTTT and BTClTT. The calculated results indicate that BTTT has a relatively symmetrical structure with almost the same dihedral angles θ1 and θ2; however, BTClTT breaks this symmetrical structure demonstrated by the apparent difference between θ1 and θ2, which is caused by the larger size of chlorine atoms with steric hindrance. The smaller BLA in BTClTT than that of BTTT imparts that the bond lengths in the former tend to be average and thus chlorination can make the double bond longer and the single bond shorter, which can be manifested by the longer bond length of C5-C6 and C7-C8 and shorter C6-C7 bond length in BTClTT than those in BTTT. Furthermore, as mentioned in the introduction section, it has been reported that larger twisted dihedral angles in chlorinated molecules could reduce crystallization tendency [23]. Therefore, the larger twisted dihedral angles in BTClTT may favor the reduction of its crystallization. Natural atomic orbital (NAO), natural population analysis (NPA) and electrostatic surface potentials (ESP) were carried out in the Multiwfn software [51]. BTClTT has an obvious localization of the electrostatic potential plotted

Scheme 2. Benzothiadiazole building blocks BTTT and BTClTT (chlorine is connected to atom 6).

Table 1 Calculated dihedral angels and bond-length alternation (BLA) of BTTT and BTClTT.

θ1 θ2 BLA

BTClTT

BTTT

38.4142 8.3017 0.0205

0.0005 0.0066 0.0600

Fig. 2. Electrostatic surface potentials (isovalue surface, 0.0004 au) of investigated fragments BTTT and BTClTT. The green color represents the areas of low potential while the red color depicts the areas of high potential. Intermediary colors indicate the areas of intermediary electrostatic potentials. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.) 48

Organic Electronics 61 (2018) 46–55

Z.-W. Zhao et al.

dipole moment change (Δμ) were carried out in the Multiwfn software [51]. Seen from Table 2, BTClTT has larger CT and DCT, which is favorable for intramolecular exciton separation [53]. Besides, it has been reported that a large dipole moment change (Δμ) from S0→Sn could facilitate exciton dissociation and generation [54]. In our calculation, Δμ of BTClTT tends to be larger in the strongest absorption state, suggesting easier charge separation. As mentioned above, chlorine broke the symmetry of the charge distribution as positive and negative charge easier to separate, which explains the increase number of transferred charges. In summary, photo-absorption property of BTClTT is not favorable for a donor material requiring larger oscillator strength and red-shifted spectrum; however, it demonstrates more apparent charge transfer character, which is beneficial for charge transfer. In addition, the higher dipole moment change for the S0→S5 transition involved the strongest-absorption state is more conducive to the improvement of performance. As a result, the chlorinated benzothiadiazole fragment BTClTT has some essential properties including smaller value of BLA, deeper FMO energy levels, asymmetric charge distribution and more apparent charge transfer character, which are favorable for the application in OPVs. Recent works adopting BTClTT-based molecules as donor materials may take advantage of these potential properties for BTClTT. Our theoretical analysis above also aroused some interests in designing more promising chlorinated donor materials.

in Fig. 2. Specifically, the negative part is smaller, the positive part is larger and electrostatic potential of thiophene ring close to chlorine is more negative in block BTClTT. It means that chlorination breaks the symmetry of the positive and negative charge distributions, demonstrating a more obvious charge distribution around the chlorine, which leads to be easier to connect to electrons or electron withdrawals from the closer thiophene ring. NPA analysis shown in supporting information was employed to supplement the charge distribution. As such, BTTT has a similar natural charge numbers of two thiophene rings, and BTClTT has different natural charges of two thiophene rings. Natural charges in the thiophene ring close to chlorine are relatively small. It indicates that chlorine substitution affects the distribution of charge in the molecule, which is the same as the ESP result. Therefore, the different charge and ESP distribution of BTClTT demonstrates that chlorine can withdraw electron from thiophene ring close to it and break the symmetrical distribution of delocalized π orbitals, which may lead to more transferred charges through photo excitation. In addition, BTClTT shows deeper FMO energy levels compared with BTTT plotted in supporting information, which is originated from the contribution of central BTCl to FMO energy levels. Through NAO analysis, the central BTCl block has an increased contribution to LUMO energy level. To be precise, BT block in BTTT has 80.37% contribution to LUMO energy level while BTCl block in BTClTT has 83.74% contribution to LUMO energy level. In summary, the results reveal that chlorination could change structure both in dihedral angels and bond length, leading to asymmetric charge distribution and deeper FMO energy levels which have close relation with performance in OPV devices.

3.2. The designed donor polymers based on polymer 1 To design superior chlorinated polymers, we adopted two strategies based on chlorinated polymer 1. First, we designed molecules 2–5 through side chains modifications on the benzo[1,2-b:4,5-b]dithiophene(BDT). Then, backbone replacement of BDT for molecules 6 and 7 were carried out to reach prominent enhancement of efficiency. Some important parameters including VOC, driving forces ΔEL-L, optical properties and charge separation efficiency at D/A interface were evaluated to judge our design strategies.

3.1.2. Optical and excitonic properties Photo-absorption properties are crucial to donor materials [46], thus probing the effect of chlorination on the optical properties is necessary. Apparently, the absorption spectrum of BTClTT is blue-shifted compared with that of BTTT presented in the supporting information, which is in accordance with the larger bandgap of BTClTT. As analyzed above, it is originated from the less delocalization of π electrons across the conjugated backbones induced by chlorine. Seen from Table 2 and supporting information which present the main excitation transitions and relevant configurations, the strongest absorption comes from S0→ S5 transition for both BTTT and BTClTT. In this transition, BTClTT has a more obvious charge transfer from thiophene rings to central BTCl block with larger Δr depicted in charge density difference (CDD) maps (Fig. 3), which suggests charge-transfer length during electron excitation, i.e., the smaller the Δr is, the more likely the transition is local excitation (LE) rather than charge-transfer excitation (CT) or Rydberg excitation. The evaluation of Δr value, charge transfer distance (DCT) (defined as the spatial distance between the barycenters of the electron density increment), transferred charges (CT) (generated from integration of electron density of increment or depletion over all space) and

3.2.1. FMO, VOC and driving forces (ΔEL-L) FMO energy levels of donor materials are important to cell performance [55]. In particularly, a better match between FMO energy levels of donor and acceptor is a prerequisite for efficient charge transfer at the interface. Fortunately, all of the driving forces ΔEL-L listed in Table 3 are large enough for efficient charge separation. FMO energy levels of investigated polymers were all calculated at the PBE0/6-311G (d,p) level, which is widely used in predicting FMO values of thiophene derivatives. Especially, LUMO energy levels were calculated by adding the first singlet excited energy computed at the TD-PBE0/6-311G (d,p) level to the HOMO energy level to obtain a more accurate data as our previous work [45]. Fig. 4 depicts the FMO distributions of molecules 1–7. For molecules 1–5, the orbital distributions of HOMO are delocalized on the whole conjugated backbone while LUMO are partially on the acceptor unit, and as for molecules 6 and 7, the orbital distributions (both HOMO and LUMO) are delocalized nearly on the whole conjugated backbones. Thus, the LUMO energy levels have little changes for molecules 2–5; however, the designed molecules 6 and 7 have an obvious change in LUMO energy levels. Besides, the designed molecules 2–5 have similar optical bandgaps (EgOPT) with polymer 1, and molecules 6–7 have smaller EgOPT. As we all know, open-circuit voltages (Voc) is one of the important parameters of OPV devices [56], which could be expressed by formula (1) [46]:

Table 2 Calculated excitation energies E (nm), oscillator strengths f, transferred charge (CT) and charge transfer distance (DCT), dipole moment change (Δμ) and major contributions of fragments BTTT and BTClTT.

BTTT

BTClTT

a

E (nm)

f

CT (q)

DCT (Å)

Δμ (D)

Contributiona

S1 S5

487.92 307.00

0.3268 0.5436

0.574 0.385

1.311 0.870

3.617 1.609

S6

286.06

0.1052

0.254

0.537

0.655

S1 S5

461.49 300.04

0.2953 0.3341

0.6030 0.3850

1.233 1.664

3.569 3.080

S6

293.40

0.3071

0.3200

1.029

1.583

H → L (98%) H → L+1 (84%) H-3 → L (14%) H-4 → L (60%) H → L+2 (32%) H → L (98%) H → L+1 (65%) H-4 → L (26%) H-4 → L (56%) H → L+1 (30%) H→ L+2 (9%)

Voc =

1 [EIP (D) − EEA (A)] − 0.3V e

(1)

Here e is the elementary charge, 0.3 is an empirical value [45], and EIP(D) and EEA(A) are approximately estimated by the HOMO energy level of donor and the LUMO energy level of acceptor PC71BM,

“H” denotes HOMO and “L” denotes LUMO. 49

Organic Electronics 61 (2018) 46–55

Z.-W. Zhao et al.

Fig. 3. CDD maps of BTTT and BTClTT, where green color represents the decrease of electron density and red color represents the increase of electron density. (Δr (Å) index is used to measure charge-transfer length during electron excitation, the smaller the Δr is, the more likely the transition is local excitation (LE) rather than charge-transfer excitation (CT) or Rydberg excitation). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

maximum absorptions in the moderate-energy region come from S0→S2 and S0→S12 corresponding to the configurations H→L+1, H→L+2 and H→L+4. And for polymer 7, the maximum absorption in the moderateenergy region corresponds to S0→S4 transition with configurations H→ L+1, H→L+2 and H→L+3. Therefore, these results indicate that our design strategy through side chains modification on BDT for molecules 2–5 have little effect on optical properties, while the backbone replacement of BDT for molecules 6 and 7 may be feasible for designing better donor materials in light of the red-shifted and broadened photoabsorption spectra compared with polymer 1. Accordingly, by employing different design strategies, we could find superior candidates of chlorinated donor materials for OPVs. In addition, considering that hole mobility of polymer is also an important parameter related to the efficiency of organic solar cell [58], we also give a qualitative comparison of the hole transport ability among polymers 1, 2 and 7 with strongest absorption among the designs of side chains modification and backbone replacement, respectively, mainly focusing on their hole reorganization energy, which is one of key factors determining hole transport ability according to Marcus equation [52]. The method to estimate hole reorganization energy is listed in supporting information. The results show that polymers 1 and 2 have close hole reorganization energies and polymer 7 exhibits slightly larger value, suggesting that polymers 1 and 2 may have similar hole transport ability while polymer 7 may possesses lower hole mobility than them in light of Marcus equation. However, we should note that another key parameter, namely electronic coupling between neighboring fragments in one polymer or neighboring polymers, could not be evaluated here since we can't make clear the packing of polymers.

Table 3 HOMO and LUMO energy values (eV) of 1–7, bandgaps EgOPT (eV), open circuit voltages Voc (V) and energetic driving forces ΔEL-L (eV) for polymers 1–7.

1 2 3 4 5 6 7

HOMO

LUMO

EgOPT

Voc (exp.)

ΔEL-L

−5.35 −5.38 −5.38 −5.32 −5.33 −5.38 −5.45

−3.37 −3.41 −3.39 −3.36 −3.35 −3.62 −3.61

1.98 1.97 1.99 1.96 1.98 1.76 1.84

0.75 (0.76) 0.78 0.78 0.72 0.73 0.78 0.85

0.93 0.89 0.91 0.94 0.95 0.68 0.69

respectively. Listed in Table 3, the calculated Voc of 1 (0.75 V) is highly consistent with the experimental one (0.76 V), which proved the rationality of our method. Besides, it manifests that the Voc of designed polymers 2, 3, 6 and 7 are higher than polymer 1, which is ascribed to the decline of their HOMO energy levels. Accordingly, the strategy of designing donor materials through branched chains modification and backbone replacement could be feasible, especially for the molecule 7 through backbone replacement, which has a more prominent enhancement on Voc. 3.2.2. Photo-absorption properties and hole transporting properties A better match with solar spectrum is beneficial for donor materials to improve short-circuit current (JSC) [57]. We have studied the optical properties of chlorinated BTClTT and get an in-depth understanding on the favorable effect of chlorination. Here, the photo-absorption properties of polymer 1 and designed polymers (2–7) which contain BTClTT fragment were also focused on aiming at judging whether they have ability to obtain abundant photon-absorption. The absorption spectra and the corresponding properties of polymers 1–7 are shown in Fig. 5 and supporting information. From Fig. 5, we can easily find that the designed polymers 2–5 have similar absorption spectra, while molecules 6 and 7 have obvious red-shifted absorption spectra. Thus for molecules 6 and 7, their red-shifted spectra may be beneficial for improving JSC. According to supporting information, all the maximum absorptions of polymers 1–7 are found to be originated from transition S0→S1 with main configuration H→L in the low-energy region. In the moderate-energy region, the maximum absorptions come from the S0→ S4 transition and correspond to the configurations of H-2→L, H-3→L +1 and H-1→L+1 transitions for polymers 1–5. As for polymer 6, the

3.2.3. The optimization of interface models The interface models have an impact on charge-transfer electronic states. However, the stacking patterns remain to be explored [54,59,60]. We simulated fifteen polymer-PC71BM models plotted in Fig. 6: i) PC71BM appears on top of the central BDT unit (referred to as the “Face-onD-Conf-1” configuration), thiophene between BDT unit and chlorinated BT unit (Face-on DA–Conf-1 configuration), or the central chlorinated BT unit (Face-on A-Conf-1 configuration); ii) PC71BM appears on top of the central BDT unit with benzene ring of PC71BM paralleled with BDT orientation (Face-onD-Conf-2D configuration) and BT orientation (Face-onD-Conf-2A configuration), thiophene between BDT unit and chlorinated BT unit with benzene ring of PC71BM 50

Organic Electronics 61 (2018) 46–55

Z.-W. Zhao et al.

Fig. 4. FMOs distribution of molecules 1–7.

paralleled to BDT orientation (Face-onDA-Conf-2D configuration) and BT orientation (Face-onDA-Conf-2A configuration), or the central chlorinated BT unit with benzene ring of PC71BM paralleled to BDT orientation (Face-on A-Conf-2D configuration) and BT orientation (Face-on A-Conf-2A configuration); iii) PC71BM lies just above the central BDT unit with branched chains of PC71BM towards BT orientation (Face-onD-Conf-3A configuration) and BDT orientation (FaceonD-Conf-3D configuration), thiophene between BDT unit and chlorinated BT unit with branched chains of PC71BM paralleled with BT orientation (Face-onDA-Conf-3A configuration) and BDT orientation (Face-onDA-Conf-3D configuration), or the central chlorinated BT unit with branched chains of PC71BM paralleled to BT orientation (Face-on A-Conf-3A configuration) and BDT orientation (Face-on A-Conf-3D configuration). The calculated counterpoise-corrected interaction energies reveal that the configuration of Face-onD-Conf-3D is preferable since the lowest counterpoise-corrected interaction energy occurs at this configuration suggesting the stablest stacking pattern among them. Thus, our following computation is based on this favorable interface packing pattern. Interestingly, series Conf-3 tend to be more preferable with larger interaction energies, which is beyond our traditional cognition that the configuration with the benzene ring of PC71BM from phenyl butyric acid methyl ester functional group paralleled to donor is

the stablest one. Therefore, our calculations on the interface models demonstrate an interesting discovery, which may enlighten the future explore on interfacial model. 3.2.4. Electron interfacial process For a typical OPV device, interfacial process plays a key role in determining the PCE of one OPV device. In this process, charge transfer and recombination compete with each other [61,62]. In general, a higher charge transfer rate kinter-CT and lower charge recombination rate kinter-CR are expected for a good OPV device as the charge transfer/ recombination rates affect JSC directly [63]. To save computational cost, we just calculated polymers 1, 2 with highest VOC among designed molecules through side chains modification and 7 with obvious enhancement on VOC for electron interfacial process. Considering that the side chains of polymers may present a remarkable impact on the D/A molecular packings and thus the electronic structures of the D/A interfaces, we take the experimental molecule (1) as example to build donor/PC71BM interfaces with and without side chains in donor (named style 2 and style 1, respectively) to give a simple comparison. The initial D/A interfaces without optimization are exhibited in Fig. S2 and the calculated parameters pertaining to interfacial process are collected in Table S1 in the supporting information. The results indicate 51

Organic Electronics 61 (2018) 46–55

Z.-W. Zhao et al.

Fig. 5. Absorption spectra of polymers 1–7.

These CT states where holes are only localized on the donor and electrons are localized on PC71BM are plotted in Fig. 7. Seen from CDD maps, we inferred that S12, S21, S22 and S24 are major CT states for 1/ PC71BM, S12, S21, S25 and S29 are major CT states for 2/PC71BM, and S15, S22, S25 and S26 are major CT states for 7/PC71BM. According to the Marcus formula in the supporting information, rate k is relevant to

that interfacial model without chains could give a reasonable description for interfacial process since the side chains in donor has a little influence on the charge transfer/recombination rates and the final effective charge dissociation. Normally, charge-transfer (CT) states were distinguished by CDD maps generated by Multiwfn 3.3.9 software [51]. We just considered several major CT states in the first thirty states.

Fig. 6. Simulated interface models with counterpoise-corrected interaction energies of 1/PC71BM. 52

Organic Electronics 61 (2018) 46–55

Z.-W. Zhao et al.

Fig. 7. CDD maps of systems 1/PC71BM, 2/PC71BM and 7/PC71BM, where green color represents the decrease of electron density and red color represents the increase of electron density. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.) Table 4 Calculated electronic couplings VDA-CR (eV) and oscillator strengths f for CT states at 1/PC71BM, 2/PC71BM and 7/PC71BM interface. 1/PC71BM

2/PC71BM

7/PC71BM

CT States

VDA-CR

f

CT States

VDA-CR

f

CT States

VDA-CR

f

CT1 CT2 CT3 CT4

−0.0559 0.0913 −0.2349 −0.0345

0.0002 0.0009 0.0042 0.0002

CT1 CT2 CT3 CT4

−0.0554 0.0979 0.0265 0.0530

0.0013 0.0066 0.0002 0.0014

CT1 CT2 CT3 CT4

0.1453 −0.4162 −0.1813 −0.3325

0.0012 0.0030 0.0026 0.0021

Table 5 Calculated internal reorganization energy λint (eV), total reorganization energy λ (eV), Gibbs free energy change ΔG (eV) including ΔGinter-CT and ΔGinter-CR, the charge transfer kinter-CT (s−1) and the charge recombination kinter-CR (s−1) of 1/PC71BM, 2/PC71BM and 7/PC71BM heterojunctions. λi-CT 1/PC71BM 2/PC71BM 7/PC71BM

0.29 0.26 0.49

λi-CR 0.11 0.09 0.15

λCT 0.50 0.46 0.70

λCR

ΔGinter-CT

ΔGinter-CR

0.31 0.30 0.36

−0.93 −0.89 −0.69

−1.05 −1.08 −1.15

several parameters such as the inner reorganization energy λint (eV) and external reorganization energy λext (eV), the value of electronic coupling (VDA) and Gibbs free energy change ΔG (eV). The reorganization energy λ as an exponential part could affect the rate severely. The inner reorganization energy λint is originated from geometrical changes of donor and acceptor and the external reorganization energy λext is a parameter pertaining to ambient, which is difficult to calculate precisely. The reorganization energy λ is computed by adding λint to λext and in the order of 7/PC71BM > 1/PC71BM > 2/PC71BM. The square of VDA is proportional to rate k according to Marcus formula in the supporting information. The calculated VDA of CT states listed in Table 4 present that the designed 7/PC71BM have obvious larger VDA

kinter-CT

kinter-CR 13

4.50 × 10 8.18 × 1012 6.75 × 1015

1.16 × 108 8.98 × 105 3.94 × 108

values, which may lead to a relative bigger kinter-CR than 1/PC71BM. Moreover, the charge transfer/recombination processes are exothermic reactions (ΔG < 0) [64]. In general, the rate kinter-CT will increase along with the decrease of absolute value of ΔGinter-CT and the kinter-CR will decrease with the increase of absolute value of ΔGinter-CR when the absolute value of ΔG is larger than λ. The calculated absolute value of ΔGinter-CT is in the order of 1/PC71BM > 2/PC71BM > 7/PC71BM, whereas the order of absolute value of ΔGinter-CR is in the opposite. Therefore, the kinter-CT may increase and the kinter-CR may decrease from 1/PC71BM to 2/PC71BM and 7/PC71BM in the view of ΔG; however, 2/ PC71BM has a relatively smaller kinter-CT and 7/PC71BM has a relatively larger kinter-CR than 1/PC71BM. All of these parameters are presented in 53

Organic Electronics 61 (2018) 46–55

Z.-W. Zhao et al.

Table 5. Our results indicate that the designed system 2/PC71BM presents obvious smaller value of kinter-CR and 7/PC71BM has an apparent higher value of kinter-CT compared with 1/PC71BM, which is favorable for OPV performance. This kind of advantage comes from the decreased value of reorganization energy λ for system 2/PC71BM and the increased value of VDA for 7/PC71BM. Also, the change of tendency on parameter ΔG makes a contribution to the increasement of the rat kinterCT for the designed systems. The results also demonstrate that the designed systems 2/PC71BM and 7/PC71BM can be better at electron interfacial process than the reported 1/PC71BM, again manifesting the possibility of designing superior chlorinated BT-based donor materials.

Guideline of the Education Department of Jilin Prov. China. Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx. doi.org/10.1016/j.orgel.2018.06.047. References [1] M.A. Green, K. Emery, Y. Hishikawa, W. Warta, E.D. Dunlop, Solar cell efficiency tables (version 47), Prog. Photovoltaics Res. Appl. 24 (2016) 3–11. [2] G.J. Hedley, A. Ruseckas, I.D. Samuel, Light harvesting for organic photovoltaics, Chem. Rev. 117 (2017) 796–837. [3] Y. Qin, Y. Chen, Y. Cui, S. Zhang, H. Yao, J. Huang, W. Li, Z. Zheng, J. Hou, Achieving 12.8% efficiency by simultaneously improving open-circuit voltage and short-circuit current density in tandem organic solar cells, Adv. Mater. 29 (2017) 1606340. [4] S. Chen, Y. Liu, L. Zhang, P.C.Y. Chow, Z. Wang, G. Zhang, W. Ma, H. Yan, A widebandgap donor polymer for highly efficient non-fullerene organic solar cells with a small voltage loss, J. Am. Chem. Soc. 139 (2017) 6298–6301. [5] G. Li, W.-H. Chang, Y. Yang, Low-bandgap conjugated polymers enabling solutionprocessable tandem solar cells, Nature Reviews Materials 2 (2017) 17043. [6] W. Zhao, S. Li, H. Yao, S. Zhang, Y. Zhang, B. Yang, J. Hou, Molecular optimization enables over 13% efficiency in organic solar cells, J. Am. Chem. Soc. 139 (2017) 7148–7151. [7] Y. Cui, H. Yao, B. Gao, Y. Qin, S. Zhang, B. Yang, C. He, B. Xu, J. Hou, Fine-tuned photoactive and interconnection layers for achieving over 13% efficiency in a fullerene-free tandem organic solar cell, J. Am. Chem. Soc. 139 (2017) 7302–7309. [8] Z. Yin, J. Wei, Q. Zheng, Interfacial materials for organic solar cells: recent advances and perspectives, Advanced science 3 (2016) 1500362. [9] R. Yu, S. Zhang, H. Yao, B. Guo, S. Li, H. Zhang, M. Zhang, J. Hou, Two wellmiscible acceptors work as one for efficient fullerene-free organic solar cells, Adv. Mater. 29 (2017) 1700437. [10] J. Huang, H. Wang, K. Yan, X. Zhang, H. Chen, C.Z. Li, J. Yu, Highly efficient organic solar cells consisting of double bulk heterojunction layers, Adv. Mater. 29 (2017) 1606729. [11] T. Erdmann, S. Fabiano, B. Milian-Medina, D. Hanifi, Z. Chen, M. Berggren, J. Gierschner, A. Salleo, A. Kiriy, B. Voit, A. Facchetti, Naphthalenediimide polymers with finely tuned in-chain pi-conjugation: electronic structure, film microstructure, and charge transport properties, Adv. Mater. 28 (2016) 9169–9174. [12] Y. Cai, L. Huo, Y. Sun, Recent advances in wide-bandgap photovoltaic polymers, Adv. Mater. 29 (2017) 1605437. [13] B. Guo, W. Li, X. Guo, X. Meng, W. Ma, M. Zhang, Y. Li, High efficiency nonfullerene polymer solar cells with thick active layer and large area, Adv. Mater. 29 (2017) 1702291. [14] L. Lu, T. Zheng, Q. Wu, A.M. Schneider, D. Zhao, L. Yu, Recent advances in bulk heterojunction polymer solar cells, Chem. Rev. 115 (2015) 12666–12731. [15] P. Zhou, Y. Yang, X. Chen, Z.-G. Zhang, Y. Li, Design of a thiophene-fused benzotriazole unit as an electron acceptor to build d–a copolymers for polymer solar cells, J. Mater. Chem. C 5 (2017) 2951–2957. [16] J. Ku, Y. Gim, Y. Lansac, Y.H. Jang, N-alkylthienopyrroledione versus benzothiadiazole pulling units in push-pull copolymers used for photovoltaic applications: density functional theory study, Phys. Chem. Chem. Phys.: Phys. Chem. Chem. Phys. 18 (2016) 1017–1024. [17] S. Xiao, Q. Zhang, W. You, Molecular engineering of conjugated polymers for solar cells: an updated report, Adv. Mater. 29 (2017) 1601391. [18] Z. Zheng, O.M. Awartani, B. Gautam, D. Liu, Y. Qin, W. Li, A. Bataller, K. Gundogdu, H. Ade, J. Hou, Efficient charge transfer and fine-tuned energy level alignment in a thf-processed fullerene-free organic solar cell with 11.3% efficiency, Adv. Mater. 29 (2017) 1604241. [19] M. Wang, D. Cai, Z. Yin, S.C. Chen, C.F. Du, Q. Zheng, Asymmetric-indenothiophene-based copolymers for bulk heterojunction solar cells with 9.14% efficiency, Adv. Mater. 28 (2016) 3359–3365. [20] Y. Liu, W. Zhao, Y. Wu, J. Zhang, G. Li, W. Li, W. Ma, J. Hou, Z. Bo, Enhancing the power conversion efficiency of polymer solar cells to 9.26% by a synergistic effect of fluoro and carboxylate substitution, J. Mater. Chem. 4 (2016) 8097–8104. [21] Z. Zheng, S. Zhang, J. Zhang, Y. Qin, W. Li, R. Yu, Z. Wei, J. Hou, Over 11% efficiency in tandem polymer solar cells featured by a low-band-gap polymer with finetuned properties, Adv. Mater. 28 (2016) 5133–5138. [22] M.L. Tang, J.H. Oh, A.D. Reichardt, Z. Bao, Chlorination: a general route toward electron transport in organic semiconductors, J. Am. Chem. Soc. 131 (2009) 3733–3740. [23] Y.-Q. Zheng, Z. Wang, J.-H. Dou, S.-D. Zhang, X.-Y. Luo, Z.-F. Yao, J.-Y. Wang, J. Pei, Effect of halogenation in isoindigo-based polymers on the phase separation and molecular orientation of bulk heterojunction solar cells, Macromolecules 48 (2015) 5570–5577. [24] S.X. Sun, Y. Huo, M.M. Li, X. Hu, H.J. Zhang, Y.W. Zhang, Y.D. Zhang, X.L. Chen, Z.F. Shi, X. Gong, Y. Chen, H.L. Zhang, Understanding the halogenation effects in diketopyrrolopyrrole-based small molecule photovoltaics, ACS Appl. Mater. Interfaces 7 (2015) 19914–19922. [25] D. Mo, H. Wang, H. Chen, S. Qu, P. Chao, Z. Yang, L. Tian, Y.-A. Su, Y. Gao, B. Yang, W. Chen, F. He, Chlorination of low-band-gap polymers: toward high-performance polymer solar cells, Chem. Mater. 29 (2017) 2819–2830.

4. Conclusions In this work, we theoretically analyzed chlorinated BT-based molecules using DFT, TD-DFT methods and Marcus theory. Firstly, we analyzed the structure-property relationship of chlorinated fragment BTClTT to gain insight into optoelectronic properties of chlorinated molecules. Secondly, we chose polymer 1 constructed with BTClTT block and a common donor unit BDT in order to get a complete understanding of chlorination effect on donor materials. Thirdly, we investigated fifteen possible interface models of polymer 1 and PC71BM to find preferable stacking patterns. Finally, we designed polymers 2–5 through side chains modification and polymers 6 and 7 through backbone replacement based on polymer 1. The photoelectric properties and some parameters pertaining to electron interfacial process were calculated and analyzed. The main conclusions we can draw are as follows: i) Chlorination effect mainly plays role in breaking symmetrical structure, affecting the charge distribution, lowering FMO energy levels, decreasing π electron delocalization, and increasing optical band gap, transferred charges, charge transfer distance, and dipole moment change from ground state to excited state. ii) The molecular design through backbone replacement can effectively improve the efficiency of charge separation at the interface as 7/PC71BM has higher kinter-CT, which is consequently expected to be candidate system in OPV device. While the molecule optimization through side chains modification seems to have slight influence on physical parameters including VOC and absorption-properties towards OPV performance; however it could lower kinter-CR in interfacial process. These results indicate that OPV performance could be enhanced by designing and optimizing chlorinated donor materials. iii) As for the interface model, we found a more stable configuration Face-on D-Conf-3D, which has the largest interaction energies among fifteen configurations. The stable configuration of this kind of stacking pattern may be due to the interaction between the branched chains of donor material and PC71BM, which reminds us to notice the especial interface model in such donor material with branched thiophene rings that could interact with acceptor PC71BM. Our work provides useful information in analyzing chlorination effect and demonstrates two efficient strategies of designing BTClTTbased donor molecules to realize enhancement of performance for OPV devices. Particularly, we found especial intermolecular packing pattern in the polymer-PC71BM, which inspires the future prediction of D/A interface models. We hope our work may be a guideline for experimental design and synthesis. Acknowledgements The authors gratefully acknowledge financial support from National Natural Science Foundation of China (21673036, 21771035, 21663011), the Fundamental Research Funds for the Central Universities (2412018ZD006) and Thirteen Five-Year Sci-tech Research 54

Organic Electronics 61 (2018) 46–55

Z.-W. Zhao et al.

[26] Z. Hu, H. Chen, J. Qu, X. Zhong, P. Chao, M. Xie, W. Lu, A. Liu, L. Tian, Y.-A. Su, W. Chen, F. He, Design and synthesis of chlorinated benzothiadiazole-based polymers for efficient solar energy conversion, ACS Energy Letters 2 (2017) 753–758. [27] J.-L. Brédas, J.E. Norton, J. Cornil, V. Coropceanu, Molecular understanding of organic solar cells: the challenges, Acc. Chem. Res. 42 (2013) 1691–1699. [28] A.J. Heeger, 25th anniversary article: bulk heterojunction solar cells: understanding the mechanism of operation, Adv. Mater. 26 (2014) 10–28. [29] A.A. Bakulin, S.D. Dimitrov, A. Rao, P.C.Y. Chow, C.B. Nielsen, B.C. Schroeder, I. McCulloch, H.J. Bakker, J.R. Durrant, R.H. Friend, Charge-transfer state Dynamics Following hole and electron transfer in organic photovoltaic devices, J. Phys. Chem. Lett. 4 (2013) 209–215. [30] H. Tamura, I. Burghardt, Ultrafast charge separation in organic PhotovoltaicsEnhanced by charge delocalization and vibronically hot exciton dissociation, J. Am. Chem. Soc. 135 (2013) 16364–16367. [31] S.M. Ryno, Y.-T. Fu, C. Risko, J.-L. Brédas, Polarization energies at organic–organic interfaces: impact on the charge separation barrier at donor–acceptor interfaces in organic solar cells, ACS Appl. Mater. Interfaces 8 (2016) 15524–15534. [32] S. Ono, K. Ohno, Combined impact of entropy and Carrier delocalization on charge transfer exciton dissociation at the donor-acceptor interface, Phys. Rev. B 94 (2016) 075305. [33] A.A. Bakulin, A. Rao, V.G. Pavelyev, P.H.M. van Loosdrecht, M.S. Pshenichnikov, D. Niedzialek, J. Cornil, D. Beljonne, R.H. Friend, The role of driving energy and delocalized states for charge separation in organic semiconductors, Science 335 (2012) 1340–1344. [34] P. Li, Y. Cui, C. Song, H. Zhang, Effects of sulfur oxidation on the electronic and charge transport properties of fused oligothiophene derivatives, J. Phys. Chem. C 120 (2016) 14484–14494. [35] D. Alberga, I. Ciofini, G.F. Mangiatordi, A. Pedone, G. Lattanzi, J. Roncali, C. Adamo, Effects of substituents on transport properties of molecular materials for organic solar cells: a theoretical investigation, Chem. Mater. 29 (2017) 673–681. [36] X.-K. Chen, M.K. Ravva, H. Li, S.M. Ryno, J.-L. Brédas, Effect of molecular packing and charge delocalization on the nonradiative recombination of charge-transfer states in organic solar cells, Adv. Energy Mater 6 (2016) 1601325. [37] G. Han, Y. Guo, X. Song, Y. Wang, Y. Yi, Terminal π–π stacking determines threedimensional molecular packing and isotropic charge transport in an a–Π–a electron acceptor for non-fullerene organic solar cells, J. Mater. Chem. C 5 (2017) 4852–4857. [38] S. Biswas, A. Pramanik, S. Pal, P. Sarkar, A theoretical perspective on the photovoltaic performance of S,N-Heteroacenes: an even–odd effect on the charge separation dynamics, J. Phys. Chem. C 121 (2017) 2574–2587. [39] Z.W. Zhao, Q.Q. Pan, S.B. Li, Y.A. Duan, Y. Geng, M. Zhang, Z.M. Su, A theoretical exploration of the effect of fluorine and cyano substitutions in diketopyrrolopyrrolebased polymer donor for organic solar cells, J. Mol. Graph. Model. 77 (2017) 9–16. [40] Q.-Q. Pan, S.-B. Li, Y. Wu, Y. Geng, M. Zhang, Z.-M. Su, Exploring more effective polymer donors for the famous non-fullerene acceptor itic in organic solar cells by increasing electron-withdrawing ability, Org. Electron. (2017), http://dx.doi.org/ 10.1016/j.orgel.2017.11.002. [41] G. Han, Y. Yi, Z. Shuai, From molecular packing structures to electronic processes: theoretical simulations for organic solar cells, Advanced Energy Materials (2018) 1702743. [42] M.H. Lee, J. Aragó, A. Troisi, Charge dynamics in organic photovoltaic materials: interplay between quantum diffusion and quantum relaxation, J. Phys. Chem. C 119 (2015) 14989–14998. [43] X. Liu, C. Huang, M. Li, Theoretical investigation on porphyrin-based small molecules as donor materials for photovoltaic applications, J. Phys. Chem. C 120 (2016) 27148–27158. [44] Q.-Q. Pan, S.-B. Li, Y. Wu, G.-Y. Sun, Y. Geng, Z.-M. Su, A comparative study of a fluorene-based non-fullerene electron acceptor and PC61BM in an organic solar cell at a quantum chemical level, RSC Adv. 6 (2016) 81164–81173. [45] S.B. Li, Y.A. Duan, Y. Geng, H.B. Li, J.Z. Zhang, H.L. Xu, M. Zhang, Z.M. Su, A designed bithiopheneimide-based conjugated polymer for organic photovoltaic with ultrafast charge transfer at donor/PC(71)BM interface: theoretical study and characterization, Phys. Chem. Chem. Phys.: Phys. Chem. Chem. Phys. 16 (2014) 25799–25808. [46] S.-B. Li, Y.-A. Duan, Y. Geng, H.-Z. Gao, Y.-Q. Qiu, Z.-M. Su, Theoretical design and characterization of pyridalthiadiazole-based chromophores with fast charge

[47]

[48] [49]

[50]

[51] [52] [53]

[54]

[55]

[56] [57]

[58]

[59]

[60]

[61]

[62] [63]

[64]

55

transfer at donor/acceptor interface toward small molecule organic photovoltaics, RSC Adv. 5 (2015) 29401–29411. N.M. O'Boyle, A.L. Tenderholt, K.M. Langner, Cclib: a library for package-independent computational chemistry algorithms, J. Comput. Chem. 29 (2008) 839–845. T. Liu, A. Troisi, Absolute rate of charge separation and recombination in a molecular model of the P3HT/Pcbm interface, J. Phys. Chem. C 115 (2011) 2406–2415. Y. Li, T. Pullerits, M. Zhao, M. Sun, Theoretical characterization of the pc60bm:pddtt model for an organic solar cell, J. Phys. Chem. C 115 (2011) 21865–21873. Gaussian 09, Revision D.01,M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, and D. J. Fox, Gaussian, Inc., Wallingford CT, (2013). T. Lu, F. Chen, Multiwfn: a multifunctional wavefunction analyzer, J. Comput. Chem. 33 (2012) 580–592. R.A. Marcus, Electron transfer reactions in chemistry: theory and experiment (nobel lecture), Angew Chem. Int. Ed. Engl. 32 (1993) 1111–1121. Y. Cui, P. Li, C. Song, H. Zhang, Terminal modulation of D−Π–a small molecule for organic photovoltaic materials: a theoretical molecular design, J. Phys. Chem. C 120 (2016) 28939–28950. M. Qiu, D. Zhu, L. Yan, N. Wang, L. Han, X. Bao, Z. Du, Y. Niu, R. Yang, Strategy to manipulate molecular orientation and charge mobility in d–a type conjugated polymer through rational fluorination for improvements of photovoltaic performances, J. Phys. Chem. C 120 (2016) 22757–22765. M.C. Scharber, D. Mühlbacher, M. Koppe, P. Denk, C. Waldauf, A.J. Heeger, C.J. Brabec, Design rules for donors in bulk-heterojunction solar cells—towards 10 % energy-conversion efficiency, Adv. Mater. 18 (2006) 789–794. N.K. Elumalai, A. Uddin, Open circuit voltage of organic solar cells: an in-depth review, Energy Environ. Sci. 9 (2016) 391–410. L. Zhang, W. Shen, R. He, X. Liu, X. Tang, Y. Yang, M. Li, Fine structural tuning of diketopyrrolopyrrole-cored donor materials for small molecule-fullerene organic solar cells: a theoretical study, Org. Electron. 32 (2016) 134–144. J. Lee, A.R. Han, J. Kim, Y. Kim, J.H. Oh, C. Yang, Solution-processable ambipolar diketopyrrolopyrrole-selenophene polymer with unprecedentedly high hole and electron mobilities, J. Am. Chem. Soc. 134 (2012) 20713–20721. M.K. Ravva, T. Wang, J.-L. Brédas, Nature of the binding interactions between conjugated polymer chains and fullerenes in bulk heterojunction organic solar cells, Chem. Mater. 28 (2016) 8181–8189. X. Shen, G. Han, D. Fan, Y. Xie, Y. Yi, Hot charge-transfer states determine exciton dissociation in the Dtdctb/C60complex for organic solar cells: a theoretical insight, J. Phys. Chem. C 119 (2015) 11320–11326. Q.-Q. Pan, S.-B. Li, Y. Wu, J. Zhang, H.-B. Li, Y. Geng, M. Zhang, Z.-M. Su, Theoretical design of three-dimensional non-fullerene acceptor materials based on an arylenediimide unit towards high efficiency organic solar cells, N. J. Chem. 41 (2017) 3857–3864. C.M. Proctor, M. Kuik, T.-Q. Nguyen, Charge Carrier recombination in organic solar cells, Prog. Polym. Sci. 38 (2013) 1941–1960. M.A. Faist, T. Kirchartz, W. Gong, R.S. Ashraf, I. McCulloch, J.C. de Mello, N.J. Ekins-Daukes, D.D. Bradley, J. Nelson, Competition between the charge transfer state and the singlet states of donor or acceptor limiting the efficiency in polymer:fullerene solar cells, J. Am. Chem. Soc. 134 (2012) 685–692. Y. Yi, V. Coropceanu, J.-L. Brédas, A comparative theoretical study of exciton-dissociation and charge-recombination processes in oligothiophene/fullerene and oligothiophene/perylenediimide complexes for organic solar cells, J. Mater. Chem. 21 (2011) 1479.