Anxin Sun , Jingui Xu , Guanhua Zong Zuo Xiao , Yong Hua , Bin Zhang , and Liming Ding

1Yunnan Key Laboratory for Micro/Nano Materials& Technology, School of MaterialsandEnergy, YunnanUniversity, Kunming650091, China

2Center for Excellencein Nanoscience (CAS), Key Laboratory of Nanosystem and Hierarchical Fabrication (CAS), National Center for Nanoscience and Technology,Beijing100190, China

3Schoolof MaterialsScienceand Engineering, Changzhou University, Changzhou 213164, China

Wide-bandgap copolymer donors with fused-ring acceptor units (FAUs) present excellent performance in nonfullereneorganicsolarcells due to theircomplementary light-absorption with nonfullerene acceptors, deep the highest occupiedmolecularorbital (HOMO) levels and high hole mobilities[1−4]. A bunch of FAU-based copolymer donors were developed in recent years, such as PM6[5], PM7[6], PBQx-TCl[7], PTQ10[8], PBQ6[9], P2F-EHp[10], D16[11], L1-S[12], D18[13,14]and D18-Cl[15,16]. They delivered >16% power conversion efficiencies (PCEs) in solar cells. To develop good FAUs is the key towardefficient FAU-based copolymerdonors.A good FAU generally hasastrong electron-withdrawingcharacter that leads toalowHOMOlevel and a high open-circuit voltage(Voc), and a relatively large molecular plane that facilitates polymer stacking and enhances hole mobility. Recently, we developed copolymer donors D18 and D18-Cl by using dithieno[3',2':3,4;2'',3'':5,6]benzo[1,2-c][1,2,5]thiadiazole (DTBT)unit[13](Fig.1(a)). Thanks tothe strong electron-withdrawing propertyandthe rigidandextendedmolecular plane of DTBT, D18 and D18-Cl deliver outstanding PCEsupto 18.69%[13−16]. The success of D18 polymers stimulated us to design more high-performance copolymer donors with novel FAUs. In this work, we designed a wide-bandgap copolymer donor P1 by using a fused-ring imide building block, 5-methyl-4H-dithieno[3,2-e:2',3'-g]isoindole-4,6(5H)-dione(MDTID). Compared with the thiadiazole moietyin DTBT, the imide moiety in MDTID ismore electron-withdrawing. The density functional theory (DFT) calculations show that MDTID has deeper HOMO and the lowest unoccupied molecular orbital (LUMO) levels than DTBT, suggesting the stronger electronaccepting capability of MDTID (Fig. 1(a)). DFT calculations alsoindicatethatMDTID leads toa deeper HOMO for P1 than that of D18, thus benefitingVoc(Fig. S1).

Thesynthetic routeforP1 is showninFig.1(b). Stille coupling of 3,4-dibromo-1-methyl-1H-pyrrole-2,5-dione and tributyl(thiophen-3-yl)stannane gave compound 1 in 83%yield. Scholl reaction[17]of compound 1 with FeCl3afforded MDTIDin43%yield. Bromination of MDTIDwith NBS gave MDTID-Br in 90% yield. Stille coupling of MDTID-Br and tributyl(4-(2-butyloctyl)thiophen-2-yl)stannane gave compound 2 in41% yield.Brominationofcompound 2 with NBS gave monomer M1 in 81% yield. Finally, copolymerization of M1 with (4,8-bis(5-(2-ethylhexyl)-4-fluorothiophen-2-yl)benzo[1,2-b:4,5-b']dithiophene-2,6-diyl)bis(trimethylstannane) gave P1 in 75% yield. The number-average molecular weight (Mn)for P1 is 69.7 kDa,andthepolydispersity index(PDI) is1.73.P1 is soluble in chloroform and chlorobenzene.

The optical, electrochemical and hole-transportingproperties of P1 were investigated. In solution, P1 shows an absorption band at 400–620 nm, with a peak at 536 nm (Fig. 1(c)).Forthe film, thispeak shifts to523 nm. The 13 nm-blueshift suggests the H-aggregation of P1 in film[18]. The absorption onset ofP1 filmis 601 nm,correspondingto an optical bandgap (Egopt) of 2.06 eV. The absorption spectra for acceptors N3[19]andIT-4F[20]are also shown inFig. 1(c). They arecomplementary with that of P1. The HOMO and LUMO energy levels of P1 were estimated from cyclic voltammetry (CV) measurements (Fig. S11). An energyleveldiagramis given inFig.S12. P1 has a HOMO of –5.54 eV and a LUMO of –2.78 eV.From CV, theHOMOof P1isdeeper than thatof D18[10],similar to the result of DFT calculation (Fig. S1). The hole mobility(μh) of P1 was measured by using space-charge limited current (SCLC)method(Fig. S13)[21−25].Pure P1 film presented aμhof 7.10 × 10–4cm2/(V·s). The goodμhof P1 could be due to the extended molecular plane of MDTID, which facilitates polymer packing and enhances charge transport.

Fig. 1. (a) The structures of D18 and P1, and DFT-predicted molecular geometries, HOMO and LUMO of DTBT and MDTID units. (b) The synthetic route for P1. (c) Absorption spectra for P1 in CHCl3, and P1, N3 and IT-4F films. (d) J–V curves for P1:N3 and P1:IT-4F solar cells. (e) EQE spectra for P1:N3 and P1:IT-4F solar cells.

Solar cells were made with a structure of ITO/PEDOT:PSS/active layer/PDIN/Ag. TheD/Aratio, active layerthickness and additive content were optimized (Tables S1–S6).J–Vcurves and external quantum efficiency (EQE) spectra for the best cells are presented in Figs. 1(d) and 1(e), respectively. Performancedata are listedin Table 1. The best P1:N3 cell gave a PCE of 14.52%, with aVocof 0.90 V, aJscof 24.52 mA/cm2and an FF of 65.8%, while the best P1:IT-4F cell gave a PCE of 12.46%, with aVocof 0.95 V,aJscof 20.31mA/cm2andanFF of 64.6%. Both cells gave highVoc(> 0.9 V) due to the deep HOMOofP1. TheVocofP1:IT-4Fcells is0.05 V higher than that of P1:N3 cells due to the higher LUMO level of IT-4F (Fig.S12). Compared with P1:IT-4F cells, P1:N3 cells gave much higherJscdue to thebroaderlight-absorption of N3 than that of IT-4F. P1:N3 cells presented higher EQE at 760–960 nm(Fig. 1(e)). The integrated photocurrent densities from EQE spectra are 23.34 and 19.83 mA/cm2for P1:N3 and P1:IT-4F cells, respectively, consisting withJsc. From SCLC measurements (Figs. S14 and S15, Table S7), we found theμhand electron mobility (μe) are both higher in P1:N3 film than those in P1:IT-4F film,suggesting moreefficientcharge transport in the former cells. Theμh/μeare 1.48 and2.06 for P1:N3 and P1:IT-4Fcells,respectively,suggestingchargetransportis morebalancedin P1:N3 cells,thus explaining the higherFF.Thebimolecularrecombination was studiedby plottingJscagainst lightintensity[26−29]. Theαvalueis closer to 1 for P1:N3cells, suggesting lessbimolecularrecombination (Fig.S16). We studied the morphology of blend films by using atomic force microscope (AFM) (Fig. S17). The root-mean-square roughnesses forP1:N3 andP1:IT-4F films are0.80 and 1.76 nm, respectively. Typicalnano-fiberswereobserved. The P1:N3 film might have more favorablephase separation.

Table 1. Performance data for P1:N3 and P1:IT-4F solar cells.

In short, we designed acopolymerdonorby using a fused-ring imide unit MDTID. The polymerdelivered highVoc(> 0.9V) anddecent PCEs(up to 14.52%) inorganic solar cells. This work indicates that MDTID is a promising building block.

Acknowledgements

We thank the National Key Research and Development Program of China (2017YFA0206600) and the National Natural ScienceFoundation ofChina (51773045, 21772030,51922032 and 21961160720).

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1088/1674-4926/42/10/100502.