consisting of central benzodithiophene (BDT) donor core coupled with terminal 1,3-indandione acceptor units, which are monosubstituted with various halogen groups

by Heavy Halogen Substitution onto π -Conjugated Small Molecules

Scheme 4-1) consisting of central benzodithiophene (BDT) donor core coupled with terminal 1,3-indandione acceptor units, which are monosubstituted with various halogen groups

including fluorine, chlorine, bromine, and iodine (2–5), through 3-hexylbithiophen. The unsubstituted parent molecule 1 is an analogue of the high performance SM donor, BDT-2T-ID, that is recently developed in our group. The effect of the halogen substitution varying from fluorine to iodine on the photophysical properties, charge transport characteristics, and photovoltaic performances was systematically investigated. The structural homogeneity of the present SMs 1–5 is regarded as an ideal platform for the strict comparison and the reliable analysis. BHJ-OSCs based on the halogenated compounds 2–5 as donors and [6,6]-phenyl-C71 -butyric acid methyl ester (PC71BM) as acceptor exhibited superior photovoltaic performance

owing to higher fill factors (FFs) than those of the unsubstituted compound 1. The highest PCE of up to 9.2% was achieved in OSCs based on iodo-functionalized 5 without any processing additives nor additional treatments of the active layer. Through intensive morphology analysis for 1–5:PC71BM blend films, the difference in device performance was attributed to domain sizes of phase-separated blend films, which were further associated with interfacial free energy at a heterointerface between the donors and the acceptor.

Scheme 4-1. Synthetic routes for 1–5

Material Synthesis and Characterization

The five SMs 1–5 were synthesized via two-fold Stille cross-coupling reactions between central BDT and 5'-bromo-3,4'-dihexyl-[2,2'-bithiophene]-5-carbaldehyde using Pd(PPh3)4 as the catalyst, followed by Knoevenagel condensation with corresponding (halogenated) 1,3-indandione (Scheme 4-1). This synthetic route allows reactive halogen groups to avoid reacting with highly active palladium catalyst in the Stille coupling step. Despite elongated rigid π-conjugated backbone, all SMs showed good solubility in common organic solvents such as chloroform and chlorobenzene owing to long and branched alkyl chains, ensuring their solution processability. The chemical structures of 1–5 were confirmed by ¹H and ¹³C NMR spectroscopy, matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectrometry, and elemental analysis. ¹H NMR spectra for halogenated compounds 2–5 presented two isomers of E- and Z-compounds around double bond linkages connecting hexylthiophene and halogenated 1,3-indandione. We conducted further evaluations for these compounds without separation of these isomers.

Thermogravimetric analysis (TG) analysis demonstrated excellent thermal stability under N2 with 10% weight loss temperature (Td) of 394, 386, 384, 372, and 353°C for compounds 1–

5, respectively (Figure 4-1a). The depression of Td along with increasing the period number of halogens would be caused by weak C–halogen bonds. Differential scanning calorimetry (DSC) analysis (Figure 4-1b-f) revealed that compounds 2–5 possess around 20⁰C higher melting

points (221–227°C) than that of compound 1 (204°C), implying stronger intermolecular interaction for the halogenated materials.

Optical Properties and Electronic Structures.

Density functional theory (DFT) calculations at B3LYP/3-21G* level were performed to gain insight the fundamental aspects of molecular geometry and electronic structures of 1–5.

As depicted in Figure 4-2, the distributions of both the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) on 1–5 were similarly delocalized over the BDT skeletons regardless of the difference in halogens. However, energy levels of both HOMO and LUMO simultaneously decreased by 0.06–0.14 eV when the terminal hydrogens of 1 were replaced with halogen groups, presumably because of inductive effect of halogen groups. The calculated first singlet excited state (S1) was dominated by HOMO→

LUMO transition with almost same oscillator strength (f) among the five molecules, while the transition energy for 2–5 were slightly decreased in comparison with that for 1, implying small contribution from halogen atomic orbitals.

Figure 4-1. (a) TG and (b–f) DSC thermograms of (b) 1, (c) 2, (d) 3, (e) 4, and (f) 5 at a scanning rate of 10 ^°C min⁻¹ under N2.

Figure 4-2. Frontier molecular orbital distributions, energy levels for 1–5 calculated at the B3LYP/3-21G* level.

The alkyl chains were replaced by methyl groups to minimize the calculation cost. The arrows indicate the transition to the first singlet excited state (S1) with corresponding oscillator strength (f).

Figure 4-3a,b shows the UV–vis absorption spectra of 1–5 in chloroform solutions and as solid thin films and their photophysical parameters are listed in Table 4-1. Diluted solutions of 1–5 exhibited similar absorption spectra but with slightly red-shifted peaks as increase in halogen period, which are consistent with the result of DFT calculations. In the solid thin films of 1–5, new red-shifted intense absorption bands appeared at approximately 660 nm with onsets at longer than 730 nm (Figure 4-3), which can be attributed to the formation of strong intermolecular interactions in the condensed solid states. The almost same absorption spectra and absorption coefficient α in the thin film state suggest that difference in photovoltaic performance of OSCs does not originate from the photoabsorption properties (discussed later).

Figure 4-3. UV–vis absorption spectra of 1–5 in (a) chloroform solutions and (b) solid thin films. (c) Photoelectron yield spectra of 1–5 in thin films measured in air.

The HOMO energy levels (i.e., ionization potentials) of 1–5 in the solid state were determined using photoelectron yield spectroscopy in air as shown in Figure 4-3 and listed in Table 4-1. The HOMO energy levels of the halogenated SMs 2−5 (from –5.21 to –5.25 eV) were found to be lower than that of the parent compound 1 (–5.10 eV). Likewise, the LUMO energy levels of these compounds exhibited a similar trend within the range from −3.52 to −3.55 eV for 2–5 and −3.42 eV for 1. These results are consistent with the trend estimation from the DFT calculations. Overall, the LUMO energy levels of these five molecules were sufficiently higher than that of the electron acceptor PC71BM (ca. −4.3 eV), implying that effective photo-induced charge separation is possible and they can be utilized as donor materials in OSCs.

Photovoltaic Properties.

The photovoltaic performances were evaluated by fabricating BHJ-OSCs using 1–5 as donor materials and PC71BM as an acceptor material with an inverted device structure of ITO/ZnO (30 nm)/PFN-Br/donor:PC71BM (~80 nm)/MoO3 (10 nm)/Ag (100 nm). The ZnO

4.8 5.2 5.6 6.0

0 20

40 1

2 3 4 5 Yield0.5 (cps0.5 )

Energy (eV) 0

5 10

300 400 500 600 700 800 900 0

10 20

1 2 3 4

4-1-14-1 ε (10Mcm)α (10cm) 5

Wavelength (nm) (a)

(b)

(c)

Table 4-1. Photophysical Properties for 1–5 compound

λmax (nm) HOMO^c

(eV)

LUMO^d

(eV) Egd (eV) solution^a thin film^b

1 542 659 −5.10 −3.42 1.68

2 548 559 −5.21 −3.50 1.71

3 554 665 −5.22 −3.53 1.69

4 554 665 −5.24 −3.55 1.69

5 556 666 −5.25 −3.55 1.70

aSolution UV−vis spectra measured in chloroform solutions (10⁻⁵ M) at room temperature. ^bThin films (ca.

100 nm) spin-coated from chloroform solution onto quartz slides. ^cDetermined by photoelectron yield spectroscopy for each spin-coated thin film in air. ^dLUMO = HOMO + Eg, in which the optical energy gap, Eg, was derived from the absorption onset of the thin film.

electron extraction layer was deposited on ITO by using a sol–gel method, followed by spin-coating of the PFN-Br layer for efficient carrier extraction.⁴⁰ The active layer was spin-coated on the PFN-Br layer from a blended solution of the donor and PC71BM (2:1, w/w) in chloroform.

MoO3 and Ag, as a hole extraction layer and an anode, respectively, were subsequently vacuum-deposited on top of the active layer to complete the inverted device configuration. Notably, the optimal active layers of all devices were obtained without any additives and post-treatment, which were facilitated for future industrial manufacturing.

Figure 4-4a,b shows the current density–voltage (J–V) characteristics under simulated AM 1.5G illumination at 100 mWcm⁻² and the incident photon-to-current conversion efficiency (IPCE) spectra under monochromatic light, respectively, for the optimized BHJ OSCs. The corresponding photovoltaic parameters are summarized in Table 4-2. Compared with unhalogenated 1, all of halogenated SMs 2–5 performed very high FF of around 68% and superior photovoltaic performances in spite of lower short-circuit current density (Jsc) values for 2–4 in OSCs. The best PCE of 9.2% was achieved in the 5-based devices owing to the high FF of 68% along with nearly identical Jsc of 14.2 mA/cm² and open-circuit voltage (Voc) of 0.94 V to those of the 1-based devices. As far as I know, this efficiency is one of the highest for those additive-free and post-treatment free SM-OSCs.¹ Interestingly, although the fluorination of the SM 1 decreased Jsc in the resultant OSCs, the Jsc values gradually recovered as the periods of halogen increase, and finally the devices bearing iodine-functionalized 5 reached higher Jsc than that of the 1-based devices. IPCE spectra in Figure 4-4b indicate that the difference in the Jsc

originate not from the absorption wavelength but conversion efficiency in each wavelength. To rationalize the difference in Jsc, the nano-scale morphology of the blend films were examined in the later part. Additionally, the halogenated SMs demonstrated 1.5 times higher FF than that of non-halogenated compound 1 in OSCs, indicating the efficient carrier extraction in the active layers. Thus, we examined the carrier recombination behavior in donor:PC71BM blend films at the short-circuit condition by measuring the dependence of Jsc on incident light intensity.

Figure 4-4. (a) J–V characteristics under one sun illumination (100 mWcm⁻², AM 1.5G) (inset: dark J–V curves) and (b) IPCE spectra for optimized BHJ-OSCs based on as-deposited 1–5:PC71BM (2:1, w/w) blend films. Light intensity dependence of (c) Jsc and (d) Voc for the optimized BHJ-OSCs. Solid lines represent linear fits to the respective data.

In general, relationship between Jsc and incident light intensity (P) is described by the following power-law equation:^41–44

Y_Z[ ∝ ]^{^} (1)

where α is a parameter indicating the extent of bimolecular recombination. A value of α = 1 denotes the absence of bimolecular recombination loss in the OSCs. By fitting the double logarithmic plots of Jsc versus P using the equation, α values of 0.94, 0.95, 0.96, 0.96, and 0.96 were obtained for the OSCs adopting 1−5, respectively (Figure 4-4c). Thus, all of the device demonstrated efficient charge sweep-out under short-circuit conditions (no external bias), with

300 400 500 600 700 800 900 0

20 40 60 80 100

IPCE (%)

Wavelength (nm) 1 2 3 4 5

-0.4 0.0 0.4 0.8 1.2

-20 -15 -10 -5 0 5 10

Current density (mA/cm2 )

Voltage [V]

1 2 3 4 5 -0.8-0.4 0.0 0.4 0.8 1.2

10^-6 10^-4 10^-2 10⁰

1 10 100

0.1 1

10 1 (α = 0.94) 2 (α = 0.95) 3 (α = 0.96) 4 (α = 0.96) 5 (α = 0.96)

Jsc (mA/cm2 )

Light Intensity (mW/cm²) 0.75 e⁰ e¹ e² e³ e⁴ e⁵ 0.80

0.85 0.90

0.95 1 (1.09 kT/q) 2 (1.33 kT/q) 3 (1.41 kT/q) 4 (1.35 kT/q) 5 (1.43 kT/q)

Voc (V)

Light intensity (mW/cm²)

(a) (b)

Table 4-2. Optimized Photovoltaic Parameters for OSCs Based on BHJ Blends of 1–5:PC71BM^a

donor thickness

(nm) Jscb

(mA cm^–2) J^c

(mA cm^–2) Vocb

(V) FF^b

(%) PCE^b,d

(%) μD/Ae (cm²

V^-1s^-1) μDe (cm² V^-1s^-1) 1 77 14.0 ± 0.2 16.0 0.94 ± 0.01 44 ± 1 5.8 ± 1 (5.9) 3.5 × 10⁻³ 9.5 × 10⁻² 2 77 11.6 ± 0.1 12.8 0.95 ± 0.01 67 ± 1 7.4 ± 1 (7.5) 3.0 × 10⁻² 4.3 × 10⁻² 3 80 11.4 ± 0.2 12.4 0.94 ± 0.01 68 ± 1 7.2 ± 1 (7.4) 4.6 × 10⁻² 4.6 × 10⁻² 4 74 12.5 ± 0.1 13.3 0.94 ± 0.01 68 ± 1 8.0 ± 1 (8.2) 4.1 × 10⁻² 4.0 × 10⁻² 5 78 14.2 ±0.1 15.1 0.94 ± 0.01 68 ± 1 9.0 ± 1 (9.2) 5.1 × 10⁻² 5.1 × 10⁻² aDevice structure: ITO/ZnO (30 nm)/PFN-Br/donor:PC71BM (80 nm)/MoOx (10 nm)/Ag (100 nm). ^bAverage value obtained from four individual devices. ^cCalculated by integrating the IPCE spectra. ^dAveraged PCE calculated as PCE = (Jsc × Voc × FF)/P0, where P0 is the incident light intensity (100 mW cm⁻²). ^eHole mobilities for the donor:PC71BM (2:1, w/w) blend film (μD/A) and pristine donor neat film (μD) determined by the SCLC technique.

86 a lower extent of bimolecular recombination.

In addition to the Jsc dependence on P, variation of Voc as a function of P is known to represent the ratio of bimolecular recombination to monomolecular recombination in OSCs at near open-circuit condition, in which all photogenerated charge carriers should recombine within the device (i.e., J = 0). When bimolecular recombination is the sole loss mechanism, the dependence of Voc on P can be described as^42–45

_`ab

c −^de_c ln f ^ghi_i^j_j_o^%kl^mnp (2)

where Egap is the energy difference between the HOMO of the donor and the LUMO of the acceptor, q is the elementary charge, k is the Boltzmann constant, T is temperature, PD is the dissociation probability of electron−hole pairs, γ is the Langevin recombination constant, NC is the density of states, and G is the generation rate of bound electron−hole pairs. Since G is solely and directly proportional to the intensity of the incident light, the slope of Voc versus the natural logarithm of P provides the ratio of bimolecular and trap-assisted monomolecular recombination. In principle, a slope equal to kT/q indicate that bimolecular recombination is dominant. In contrast, if the degree of bimolecular recombination is comparable to monomolecular recombination, a slope of ~2kT/q can be observed.^42–45 As shown in Figure 4-4d, the slope of fitted lines for the OSCs composed of 2–5 fell in the range of 1.33–1.43kT/q, while the 1-based devices exhibited the slope of 1.09kT/q. These results indicate that bimolecular recombination is comparably less to monomolecular recombination in the devices of 2–5, indicating fast carrier extraction from the active layer even around V ≅ Voc, where effective internal electric field is very low.

Charge Transport Properties.

To further examine the behavior of the charge carriers, we applied the space-charge limited current (SCLC) technique in neat and blend films with a device structure of ITO/PEDOT:PSS (30 nm)/donor (128~180 nm, for neat films) or donor:PC71BM (2:1, w/w; 158–173 nm, for blend films)/MoO3 (10 nm)/Ag(100 nm).^46–48 The dark J–V characteristics are shown in Figure 4-5, and the estimated hole mobilities are integrated in Table 4-1.

In neat films all SMs exhibited similarly high hole mobilities of 9.5 × 10⁻², 3.0 × 10⁻², 4.6

× 10⁻², 4.1 × 10⁻², 5.1 × 10⁻² cm²V⁻¹s⁻¹ for 1–5, respectively. In the blend state 1:PC71BM showed one order of magnitude lower hole mobility of 3.5 × 10⁻³ cm²V⁻¹s⁻¹than that of neat state, which arise from interrupted crystallization of 1 by addition of PC71BM. By contrast, the depression of hole mobility was not observed in 2–5:PC71BM blend films compared with those

in neat films. Therefore, the improved FFs for the devices based on 2–5 are attributed to these high carrier mobilities. These results indicate that the introduction of halogen groups into a molecular skeleton is an effective way to sustain fast charge transport properties even in the blended state with PC71BM, and to achieve high FFs in the OSCs.

Figure 4-5. Double logarithmic plots of dark J–V curves of hole-only devices based on (a) 1–5 neat films and (b) 1–5:PC71BM blend films. The light-blue solid lines represent the best fits to the SCLC model: the slope of log(J) vs log(V) is ∼2.

Film Morphology and Nanostructure Characterization.

Grazing incidence X-ray diffraction (GIXD) was conducted to characterize the self-organized structure and morphology for neat films of 1–5 and blend films of 1–5:PC71BM. In Figure 4-6, (100) reflections originating from lamellar structures of 1–5 were observed at q = 0.320–0.343 nm⁻¹ in both of neat and blend films. Using the Scherrer equation,⁴⁹ we estimated the mean crystallite size (or crystal coherence lengths) from (100) reflections as denoted L in Figure 4-6. In comparison with the neat film state, the mixing of PC71BM to 1 decreased crystallite size from 30 nm for neat state to 18 nm for blend state while those of 2–5 were retained to around 13 nm even in the blended state. These results are consistent with the tendency of hole mobility that decrease after mixing with PC71BM for 1 and the others not so.

Hence, it is suggested that crystallization of the halogenated compounds 2–5 are not interrupted by the mixing with PC71BM.

0.1 1 10

10^-3 10^-2 10^-1 10⁰ 10¹ 10² 10³ 10⁴

J (mA/cm2 )

V (V) 1 2 3 4 5

SCLC fitting

0.1 1 10

10^-3 10^-2 10^-1 10⁰ 10¹ 10² 10³ 10⁴

V (V)

1:PC₇₁BM 2:PC₇₁BM 3:PC₇₁BM 4:PC₇₁BM 5:PC₇₁BM SCLC fitting

(a) (b)

Figure 4-6. GIXD patterns of (a) 1–5 neat films and (b) 1–5:PC71BM blend films. The L values represent the average crystallite size calculated from Scherrer equation.

Furthermore, we investigated the interior morphologies of the blend films composed of 1–

5 and PC71BM with transmission electron microscopy (TEM), as shown in Figure 4-7a–e. The bright and dark regions in the images correspond to the donor-rich and PC71BM-rich domains, respectively, because PC71BM has higher electron density than donor SMs. For quantitative comparison of the degree of phase segregation among the blend films, we deduced the domain sizes of quasi-periodic phase-separated structures by using the power spectral densities (PSDs, Figure 4-7f) obtained from two-dimensional fast Fourier transform (2D-FFT) images of the TEM images,^50,51 and denoted them as D on bottom-right of each image. The estimated domain sizes were found to be in the order of 1:PC71BM (D = 11 nm) < 5:PC71BM (D = 32 nm) <

4:PC71BM (D = 44 nm) < 3:PC71BM (D = 55 nm) ≅ 2:PC71BM (D = 56 nm). Since exciton diffusion length is typically limited less than 10 nm in organic semiconducting materials, smaller domain sizes for the 1:PC71BM and 5:PC71BM blend films are favorable for efficient charge generation than those of 3:PC71BM and 2:PC71BM systems, which agree with high Jsc

in the corresponding OSCs. These result suggest that the incorporation of iodine atoms into the BDT-2T-ID skeleton (1) relatively keeps its high miscibility with PC71BM, leading to finely phase-separated structure in the blend films.

0.5 1.0

5, L = 12 nm 4, L = 12 nm 3, L = 13 nm 2, L = 14 nm

Intensity (a.u.)

q (Å^-1)

1, L = 30 nm

0.5 1.0

5:PC₇₁BM, L = 13 nm 4:PC₇₁BM, L = 14 nm 3:PC₇₁BM, L = 14 nm 2:PC₇₁BM, L = 14 nm

q (Å^-1)

1:PC₇₁BM, L = 18 nm

(a) (b)

Figure 4-7. TEM images of BHJ active layers composed of (a) 1:PC71BM (b) 2:PC71BM (c) 3:PC71BM (d) 4:PC71BM and (e) 5:PC71BM. The D values represent the average domain sizes. (f) PSD profiles of the blend films obtained from 2D-FFT analysis of the corresponding TEM images.

Interfacial Free Energy and Miscibility.

For discussion of the intermolecular interactions and miscibility between donor and acceptor materials, we evaluated surface free energies of neat films of 1–5 and PC71BM with the contact angle measurement. According to the Owens–Wendt theory,⁵² the dispersive and polar components of surface free energy can be calculated from contact angles between spin-coated films and multiple probe solvent droplets by plotting the following equation:^52–54

k_r^stsau gv w%

Nxk_r^y xz_J⁾+ xz_J⁵x_k^k^r^b

ry (3)

where γ is the surface free energy, θ is the contact angle, subscripts S and L refer to the solid and liquid surface, and superscripts d and p refer to the dispersive and polar contributions, respectively (γLtotal is the sum of the dispersive and polar components). Figure 4-8a represents a plot of eq. 3 using the obtained contact angles and the literature values of surface free energy components of probe liquid. To eliminate experimental errors and uncertainty of reported surface free energy of probe solvents, we adopted several solvents such as water, glycerol, ethylene glycol, N,N-dimethylformamide, and octane. The linear fit of the plot allows accurate determination of the dispersive and polar components of the organic films, as summarized in Table 4-3. As a result, PC71BM exhibited the highest γ^d among the six materials, which result in total surface free energy of PC71BM similar as the literature.^25,55–58 For the halogenated compounds 2–5, there is a specific tendency that γ^d gradually increase and approach to that of PC71BM with increasing the halogen period number while γ^p are almost constant. Since γ^d

represents the contribution from the van der Waals force, this tendency would be attributed to expanded molecular orbitals on heavy halogens.

From these results, we calculated interfacial free energies at the heterointerface between 1–

5 and PC71BM, which play an important role for charge separation in OSCs. Interfacial free energy γD/A was calculated with the following equation:^25,58,59

z_{/| z_{ ^-+ z_| ^-− 2 xz_{⁾z_|⁾+ xz_{⁵z_|⁵% (4)

where subscripts D and A denote SM donors 1–5 and acceptor PC71BM, respectively. The calculated interfacial free energy of each material are summarized in Table 4-3. For 2–5 the interfacial free energy against PC71BM, which represents an excess energy at the heterointerface, decreased in a periodic order of halogens, implying better miscibility of the heavy-halogen functionalized SMs with PC71BM and facile interface formation during spin-coating process. Here, we plotted domain sizes of 2–5:PC71BM blend films against γD/A for each combination, and presented in Figure 4-8b. An obvious correlation was found between domain size and γD/A. In addition, an evident relationship is also observed between the Jsc values of the resulting devices and the γD/A. These results suggest the new halogenation strategy that small difference in surface free energy components can enhance exciton dissociation probability and

Figure 4-8. (a) Owens–Wendt plots for neat films of 1–5 and PC71BM according to eq. 3. (b) Relationship of domain size of 1–5:PC71BM blend films and Jsc of corresponding OSCs against interfacial free energy (γDA).

Table 4-3. Calculated surface energy components for 1–5 and PC71BM and interfacial free energy.

compound γ^d γ^p γD/Aa

1 20.6 0.76 1.80

2 26.8 0.24 0.65

3 27.6 0.18 0.60

4 28.6 0.17 0.51

5 28.5 0.27 0.42

PC71BM 34.6 0.77 –

aInterfacial free energy between each donor and PC71BM calculated using eq. 4

91 thus improve Jsc in the actual OSCs.

Experimental

4-8-1. Instrumentation

1H and ¹³C NMR spectra were recorded on a Bruker Avance III 400 spectrometer. Chemical shifts of ¹H and ¹³C NMR signals were quoted to tetramethylsilane (δ = 0.00) and CDCl3 (δ = 77.0) as internal standards. Matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectra were collected on a Bruker Daltonics Autoflex III spectrometer using dithranol as the matrix. UV/Vis absorption spectra were measured with a JASCO V-670Y spectrometer. The HOMO energy levels were determined using a Riken-Keiki AC-2 ultraviolet photoelectron yield spectrometer. The LUMO energy levels were estimated by subtracting the optical energy gap (Eg) from the measured HOMO energy levels; the Eg values were determined from the onset position of the absorption spectra. Differential scanning calorimetry (DSC) was performed using a Hitachi X-DSC7000. Transmission electron microscope images were obtained using a JEOL JEM-2010. Grazing incidence X-ray diffraction (GIXD) patterns were obtained using a Rigaku SmartLab with Ni-filtered CuKα radiation. All quantum chemical calculations were performed using the Gaussian 16 program package. Geometry optimization was carried out using the B3LYP functional with the 3-21G* basis set. Low-lying excited singlet states were computed using the optimized structures with time-dependent density functional theory (TD-DFT) at the same level. Contact angle measurement was conducted using Kyowa Interface Science Co. DMs-401.

4-8-2. Materials

Commonly available reagents and solvents were used without further purification unless otherwise noted. All of the reactions were performed under a nitrogen atmosphere in dry solvents using standard Schlenk techniques. PFN-Br was purchased from 1-material.

Ethanolamine and 2-methoxyethanol were purchased from Sigma-Aldrich. Zinc acetate was purchased from Wako chemical. PC71BM was purchased from Frontier Carbon Corporation.

PEDOT:PSS (Clevios AI 4083) was purchased from Heraeus. (4,8-bis(5-(2-butyloctyl)thiophen-2-yl)benzo[1,2-b:4,5-b']dithiophene-2,6-diyl)bis(trimethylstannane) was purchased from Suna Tech Inc.

4-8-3. OSC device fabrication and evaluation.

Prepatterned ITO-coated glass substrates were cleaned sequentially by sonicating in detergent

solution, deionized water, acetone, and isopropyl alcohol for 10 min each, and were then subjected to UV/ozone treatment for 15 min. A thin layer (≈ 30 nm) of ZnO was prepared by spin-coating (at 5000 rpm for 60 s) a precursor solution of zinc acetate (1.00 g) and ethanolamine (0.28 g) in 2-methoxyethanol (10 mL) through a 0.20 µm polyethylene membrane filter, followed by baking at 200 °C for 10 min under air. PFN-Br thin layer was deposited by spin-coating at 5000 rpm for 60 s from a solution of PFN-Br in methanol (1.0 mg mL⁻¹) through a 0.45 µm poly(tetrafluoroethylene) membrane filter. The photoactive layer was then deposited by spin-coating from a chloroform solution containing a donor molecule (9.3 mg mL^-1) and PC71BM (4.6 mg mL^-1), after passing through a 0.45 µm poly(tetrafluoroethylene) membrane filter. The thickness of the photoactive layer was around 80 nm, measured with a profilometer.

Finally, 10-nm-thick MoO3 and 100-nm-thick Ag layers were thermally evaporated on top of the active layer under high vacuum less than 5.0 × 10^-4 Pa in an E-200 vacuum evaporation system (ALS Technology), through a shadow mask defining an active area for each device of 0.04 cm². The current density-voltage (J–V) and IPCE measurements for the fabricated OSCs were conducted on a computer-controlled Keithley 2400 source measure unit in air under simulated AM 1.5G solar illumination at 100 mW cm^-2 (1 sun) using a Xe lamp-based Bunko-keiki SRO-25GD solar simulator and IPCE measurement system, calibrated with a standard Si photodiode. Jsc and Voc dependence on light intensity were measured in the same equipment with a reduced AM 1.5G illumination through ND filters, calibrated with the standard Si photodiode.

4-8-4. Carrier Mobility Measurements.

The mobility measurements on the pristine neat films and BHJ blend films were carried out using the following diode structure: ITO/PEDOT:PSS (30 nm)/1–5 or 1–5:PC71BM layer/MoOx

(10 nm)/Ag (100 nm) for hole mobility; dark J−V curves were recorded over a voltage range of 0−10 V. The charge carrier mobilities were calculated using the SCLC model: J = (9/8)ε0εrμ(V²/L³), where ε0 is the permittivity of free space (8.85 × 10⁻¹⁴ C V⁻¹ cm⁻¹), εr is the relative dielectric constant of the transport medium (assumed to be 3.0), μ is the hole mobility, and L is the thickness of the active layer.

4-8-5. Thin-Film Characterization.

GIXD experiments were conducted using Ni-filtered CuKα radiation for the samples prepared on Si substrates in the same manner as actual devices at a fixed incidence angle of 0.1°. TEM measurements were performed using a JEM-2010 transmission electron microscope (JEOL) at

an accelerating voltage of 120 kV. The spin-coated thin films on mica were peeled from the substrates by soaking in water and then transferred onto copper grids for the TEM observations.

Contact angle measurement was carried out using 1 μL of probe solvent for the samples prepared on quartz substrates in the same condition as the optimal device fabrication. Contact angles were estimated with curve fitting as perfect circle method for the edge of liquid droplets after 1 s from formation of droplet on the films. The contact angles were averaged over five different measurement.

94 4-8-6. Material Synthesis

S S C6H13

C₆H₁₃ Br

Pd(PPh3)4 DMF S

S S

Sn Sn

C₄H₉ C6H13

C4H9 C₆H₁₃

S S

C₄H₉ C6H13

S S

C6H13

C₆H₁₃ O S

C₆H₁₃ C6H13 O

O O

CHCl3/ DBU

S S

C₄H₉ C₆H₁₃

C4H9 C₆H₁₃

S S

C₆H₁₃

C6H13 O

O S

C₆H₁₃ C₆H₁₃

O O

Br S

S S

C4H9 C₆H₁₃

C₄H₉ C₆H₁₃

S S

C₆H₁₃

C₆H₁₃ O

O S

C₆H₁₃ C6H13

O O

S S

C₄H₉ C6H13

C₄H₉ C₆H₁₃

S S

C₆H₁₃

C6H13 O

O S

C6H13 C₆H₁₃

O O

S S

C₄H₉ C₆H₁₃

C4H9 C₆H₁₃

S S

C₆H₁₃

C6H13 O

O S

C6H13 C₆H₁₃

O O

Br O O O

Ac2O / Et3N, r.t.

O O O 1)

2) HCl/H2O

O O

F O O O

Ac2O / Et3N, r.t.

O O O 1)

2) HCl/H2O

O O

Cl O O O

Ac2O / Et3N, r.t.

O O O 1)

2) HCl/H2O

O O

I O O O

I O

O I I

OH OH 1) KMnO4

pyridine/H2O reflux 2) HCl

Ac2O ^{1) NEt}r.t.³^/Ac²^O 2) HCl/H2O O O O

S S

C₄H₉ C₆H₁₃

C4H9 C₆H₁₃

S S

C₆H₁₃

C6H13 O

O S

C₆H₁₃ C₆H₁₃

O O

CHCl3/ DBU

O O

CHCl3/ DBU

O O

CHCl3/ DBU

O O

CHCl3/ DBU

6 7

8 9 10 11

Scheme 2 Synthesis routes for molecules 1–5.

5-fluoro-1H-indene-1,3(2H)-dione (6). To a stirred solution of 5-fluorophthalic anhydride (6.98 g, 42.0 mmol) in acetic anhydride (16.8 mL) and triethylamine (9.1 mL) was added ethyl acetoacetate (6.01 g, 46.2 mmol). The mixture was stirred overnight at room temperature, followed by adding ice (20 mL) and concentrated HCl (48.4 mL). The mixture was stirred at 80 °C for 20 min in air. After cooled to room temperature, the mixture was extracted with chloroform. The organic layer was dried over anhydrous Na2SO4. After filtration and evaporation, the product was purified by silica gel column chromatography (eluent:

chloroform), recrystallized from chloroform/hexane, and dried under vacuum to afford 6 as a pale yellow solid (yield = 3.63 g, 53%). ¹H NMR (400 MHz, CDCl3) δ 8.03–8.00 (m, 1H), 7.61–7.59 (m, 1H), 7.53 (td, J = 8.5, 2.3 Hz, 1H), 3.28 (s, 2H). MS (MALDI-TOF) m/z: [M−H]⁻, 163.02; Found, 162.38.

5-chloro-1H-indene-1,3(2H)-dione (7). This compound was prepared by a similar method to that of 6, using 5-chlorophthalic anhydride (7.30 g, 40.0 mmol) and ethyl acetoacetate (5.73 g, 44.0 mmol). The product was purified by silica gel column chromatography (eluent:

chloroform), recrystallized from chloroform/hexane, and dried under vacuum to afford 7 as a pale yellow solid (yield = 5.14 g, 71%). ¹H NMR (400 MHz, CDCl3) δ 7.94 (s, 1H), 7.93 (d, J

= 9.8 Hz 1H), 7.80 (dd, J = 8.2, 1.9 Hz, 1H), 3.27 (s, 2H). MS (MALDI-TOF) m/z: [M−H]⁻, 178.99; Found, 178.55.

5-bromo-1H-indene-1,3(2H)-dione (8). This compound was prepared by a similar method to that of 6, using 5-bromophthalic anhydride (6.81 g, 30.0 mmol) and ethyl acetoacetate (4.29 g, 33.0 mmol). The product was purified by silica gel column chromatography (eluent:

chloroform), recrystallized from chloroform/hexane, and dried under vacuum to afford 8 as a pale yellow solid (yield = 5.06 g, 75%). ¹H NMR (400 MHz, CDCl3) δ 8.12 (d, J = 1.8 Hz, 1H), 7.96 (dd, J = 8.0, 1.8 Hz, 1H), 7.85 (d, J = 8.0 Hz, 1H), 3.26 (s, 2H). MS (MALDI-TOF) m/z:

96 [M−H]⁻, 222.94; Found,.222.62.

4-iodophthalic acid (9). To a stirred solution of 4-iodo-1,2-dimethylbenzene (3.48 g, 15.0 mmol) in pyridine (31.5 mL) and distilled water (66 mL) was added KMnO4 (23.7 g, 150 mmol).

The mixture was stirred under reflux for 24 h. The hot reaction mixture was filtered and the residual solid was washed with KOH aqueous solution (1 M, 100 mL). After added concentrated HCl until the filtrate became pH = 1, organic layer was extracted with ethyl acetate.

The organic layer was dried over anhydrous Na2SO4. After filtration and evaporation, the product 9 was obtained as a white solid (yield = 3.58 g, 82%). ¹H NMR (400 MHz, CDCl3) δ 8.22 (d, J = 1.8 Hz, 1H), 7.99 (dd, J = 8.3, 1.8 Hz, 1H), 7.64 (d, J = 8.0 Hz, 1H). MS (MALDI-TOF) m/z: [M−H]⁻, 290.92; Found, 290.81.

5-iodophthalic anhydride (10). The mixture of 9 (3.42 g, 11.7 mmol) in acetic anhydride (50 mL) was stirred overnight at 100 °C. After removal of solvent, the pure product 10 was obtained as a yellow solid (yield = 2.94 g, 92%) without any other purification. ¹H NMR (400 MHz, CDCl3) δ 8.38 (d, J = 1.0 Hz, 1H), 8.26 (dd, J = 8.0, 1.5 Hz, 1H), 7.73 (d, J = 8.3 Hz, 1H). MS (MALDI-TOF) m/z: [M−H]⁻, 272.91; Found, 272.50.

5-iodo-1H-indene-1,3(2H)-dione (11). This compound was prepared by a similar method to that of 6, using 10 (1.51 g, 5.50 mmol) and ethyl acetoacetate (0.787 g, 6.05 mmol). The product was purified by silica gel column chromatography (eluent: chloroform), recrystallized from chloroform/hexane, and dried under vacuum to afford 15 as a pale yellow solid (yield = 1.06 g, 71%). ¹H NMR (400 MHz, CDCl3) δ 8.36 (t, J = 0.8 Hz, 1H), 8.18 (dd, J = 8.0, 1.5 Hz, 1H), 7.70 (dd, J = 8.0, 0.5 Hz, 1H), 3.23 (s, 2H). MS (MALDI-TOF) m/z: [M−H]⁻, 270.93; Found, 269.84.

5',5'''-(4,8-bis(5-(2-butyloctyl)thiophen-2-yl)benzo[1,2-b:4,5-b']dithiophene-2,6-diyl)bis(3,4'-dihexyl-[2,2'-bithiophene]-5-carbaldehyde) (12) To a stirred solution of (4,8-

bis(5-(2-butyloctyl)thiophen-2-yl)benzo[1,2-b:4,5-b']dithiophene-2,6-diyl)bis(trimethylstannane) (1.83 g, 1.80 mmol) and 5'-bromo-3,4'-dihexyl-[2,2'-bithiophene]-5-carbaldehyde (1.67 g, 3.78 mmol) in dry DMF (20 mL) were added Pd(PPh3)4 (83 mg, 0.072 mmol). The solution was N2-bubbled for 5 min and then stirred overnight at 80 °C. After cooling to room temperature, the reaction mixture was poured into methanol and the precipitate was collected by filtration. The product was purified by silica gel column chromatography (eluent: chloroform), recrystallized from chloroform/methanol, and dried under vacuum to afford 12 as a red solid (yield = 2.20 g, 87%). ¹H NMR (400 MHz, CDCl3) δ 9.83 (s, 2H), 7.70 (s, 2H), 7.59 (s, 2H), 7.35 (d, J = 3.5 Hz, 2H), 7.14 (s, 2H), 6.91 (d, J = 3.5 Hz, 2H), 2.88–2.80 (m, 12H), 1.72–1.65 (m, 10H), 1.38–1.28 (m, 56H), 0.92–0.84 (m, 24H). MS (MALDI-TOF) m/z: [M+H]⁺, 1411.67; Found, 1411.14.

2,2'-(((4,8-bis(5-(2-butyloctyl)thiophen-2-yl)benzo[1,2-b:4,5-b']dithiophene-2,6-diyl)bis(3,4'-dihexyl-[2,2'-bithiophene]-5',5-diyl))bis(methaneylylidene))bis(1H -indene-1,3(2H)-dione) (1). To a stirred solution of 12 (212 mg, 0.150 mmol) and 1,3-indandione (46 mg, 0.31 mmol) in chloroform (3 mL) was added 1,8-diazabicyclo[5.4.0]undec-7-ene (1 drop).

The mixture was stirred overnight at 60 °C. After cooling to room temperature, the reaction mixture was directly purified by silica gel column chromatography (eluent: chloroform), recrystallized from chloroform/methanol, and dried under vacuum to afford 1 as a black solid.

This material was further purified by recycling preparative GPC (eluent: chloroform) prior to use (yield = 229 mg, 91%). ¹H NMR (400 MHz, CDCl3) δ 7.98–7.94 (m, 4H), 7.88 (s, 2H), 7.79–7.76 (m, 6H), 7.72 (s, 2H), 7.38 (d, J = 3.5 Hz, 2H), 7.32 (s, 2H), 6.94 (d, J = 3.5 Hz, 2H),

2.90–2.84 (m, 12H), 1.76–1.67 (m, 10H), 1.44–1.28 (m, 56H), 0.93–0.84 (m, 24H). ¹³C NMR (126 MHz, CDCl3) δ 190.53, 189.91, 146.20, 145.57, 145.02, 142.20, 142.06, 141.08, 140.68, 139.30, 137.16, 136.94, 136.72, 135.86, 135.11, 134.90, 134.79, 134.48, 133.80, 131.14, 128.06, 125.72, 123.96, 123.69, 123.10, 122.92, 122.21, 40.20, 34.91, 33.62, 33.20, 32.11, 31.84, 31.82, 30.74, 30.18, 29.90, 29.83, 29.52, 29.36, 29.07, 26.84, 23.24, 22.88, 22.85, 22.80, 14.38, 14.33, 14.31, 14.29. MS (MALDI-TOF) m/z: [M+H]⁺, 1667.72; Found, 1667.22. Anal.

calcd (%) for C102H122O4S8: C 73.42, H 7.37; found: C 73.53, H 7.42.

2,2'-(((4,8-bis(5-(2-butyloctyl)thiophen-2-yl)benzo[1,2-b:4,5-b']dithiophene-2,6-diyl)bis(3,4'-dihexyl-[2,2'-bithiophene]-5',5-diyl))bis(methaneylylidene))bis(5-fluoro-1H -indene-1,3(2H)-dione) (2). This compound was prepared by a similar method to that of 1, using 12 (706 mg, 0.500 mmol) and 6 (181 mg, 1.10 mmol). The product was purified by silica gel column chromatography (eluent: chloroform), recrystallized from chloroform/methanol, and dried under vacuum to afford 2 as a black solid. This material was further purified by recycling preparative GPC (eluent: chloroform) prior to use (yield = 592 mg, 69%). ¹H NMR (400 MHz, CDCl3) δ 7.98–7.94 (m, 2H), 7.86 (d, J = 4.0 Hz, 2H), 7.79 (d, J = 4.3 Hz, 2H), 7.72 (s, 2H), 7.59–7.56 (m, 2H), 7.42 (tt, J = 8.5, 2.6 Hz, 2H), 7.38 (d, J = 3.5 Hz, 2H), 7.32 (s, 2H), 6.94 (d, J = 3.3 Hz, 2H), 2.90–2.84 (m, 12H), 1.76–1.67 (m, 10H), 1.44–1.28 (m, 56H), 0.93–0.84 (m, 24H). ¹³C NMR (126 MHz, CDCl3) δ 189.23, 188.97, 188.49, 188.42, 166.15, 166.02, 146.23, 145.96, 145.83, 145.65, 145.48, 142.12, 141.23, 141.20, 139.32, 138.30, 137.17, 136.91, 136.76, 136.69, 136.23, 136.15, 134.70, 134.56, 134.38, 134.02, 131.26, 128.08, 125.73, 123.77, 123.71, 122.52, 122.33, 122.26, 122.16, 110.04, 109.93, 109.86, 109.75, 40.21, 34.91, 33.63, 33.20, 32.11, 31.85, 31.82, 30.74, 30.16, 29.90, 29.83, 29.53, 29.36, 29.07, 26.85, 23.25, 22.88, 22.85, 22.80, 14.38, 14.33, 14.31, 14.29. MS (MALDI-TOF) m/z: [M+H]⁺, 1703.70;

Found, 1703.58. Anal. calcd (%) for C102H120F2O4S8: C 71.87, H 7.10; found: C 71.80, H 7.13.

2,2'-(((4,8-bis(5-(2-butyloctyl)thiophen-2-yl)benzo[1,2-b:4,5-b']dithiophene-2,6-diyl)bis(3,4'-dihexyl-[2,2'-bithiophene]-5',5-diyl))bis(methaneylylidene))bis(5-chloro-1H -indene-1,3(2H)-dione) (3). This compound was prepared by a similar method to that of 1, using 12 (706 mg, 0.500 mmol) and 6 (199 mg, 1.10 mmol). The product was purified by silica gel column chromatography (eluent: chloroform), recrystallized from chloroform/methanol, and dried under vacuum to afford 3 as a black solid. This material was further purified by recycling preparative GPC (eluent: chloroform) prior to use (yield = 307 mg, 35%). ¹H NMR (400 MHz, CDCl3) δ 7.89–7.85 (m, 6H), 7.79 (d, J = 4.3 Hz, 2H), 7.72–7.68 (m, 4H), 7.38 (d, J = 3.3 Hz, 2H), 7.31 (s, 2H), 6.94 (d, J = 3.3 Hz, 2H), 2.90–2.83 (m, 12H), 1.76–1.67 (m, 10H), 1.44–1.28 (m, 56H), 0.93–0.84 (m, 24H). ¹³C NMR (126 MHz, CDCl3) δ 189.25, 189.17, 188.62, 188.40, 146.22, 146.03, 145.95, 145.85, 145.78, 143.47, 142.09, 141.98, 141.72, 141.55, 141.25, 141.22, 140.19, 139.26, 138.68, 137.16, 136.91, 136.67, 136.49, 135.07, 134.89, 134.69, 134.64, 134.32, 134.11, 131.31, 128.09, 125.72, 124.37, 124.20, 123.68, 123.40, 123.21, 123.08, 122.20, 40.21, 34.92, 33.64, 33.21, 32.12, 31.85, 31.82, 30.72, 30.13, 29.92, 29.88, 29.55, 29.37, 29.08, 26.86, 23.26, 22.89, 22.87, 22.82, 14.39, 14.34, 14.32. MS (MALDI-TOF) m/z: [M+H]⁺, 1735.64; Found, 1736.04. Anal. calcd (%) for C102H120Cl2O4S8: C 70.51, H 6.96;

found: C 70.29, H 6.96.

100

2H), 7.38 (d, J = 3.5 Hz, 2H), 7.32 (s, 2H), 6.94 (d, J = 3.5 Hz, 2H), 2.90–2.83 (m, 12H), 1.76–

1.67 (m, 10H), 1.44–1.28 (m, 55H), 0.93–0.84 (m, 24H). ¹³C NMR (126 MHz, CDCl3) δ 189.43, 189.07, 188.80, 188.29, 146.22, 146.06, 145.97, 145.90, 145.84, 143.41, 142.09, 141.93, 141.26, 141.23, 140.56, 139.26, 139.05, 137.96, 137.78, 137.16, 136.91, 136.67, 136.56, 134.70, 134.67, 134.31, 134.13, 131.32, 130.28, 130.08, 128.10, 126.26, 126.14, 125.72, 124.45, 124.30, 123.68, 123.21, 122.21, 40.21, 34.93, 33.64, 33.21, 32.12, 31.85, 31.82, 30.71, 30.12, 29.92, 29.88, 29.56, 29.37, 29.08, 26.86, 23.26, 22.89, 22.87, 22.82, 14.39, 14.34, 14.32.

MS (MALDI-TOF) m/z: [M+H]⁺, 1823.54; Found, 1822.94. Anal. calcd (%) for C102H120Br2O4S8: C 67.08, H 6.62; found: C 67.13, H 6.62.

= 5.8 Hz, 2H), 2.90–2.84 (m, 12H), 1.73 (dd, J = 15.1, 7.0 Hz, 10H), 1.44–1.28 (m, 56H), 0.93–

0.84 (m, 24H). ¹³C NMR (126 MHz, CDCl3) δ 189.76, 189.11, 189.04, 188.26, 146.22, 146.06, 145.96, 145.88, 145.84, 143.88, 143.69, 143.07, 142.09, 141.59, 141.27, 141.11, 139.58, 139.26, 137.16, 136.92, 136.68, 136.59, 136.54, 134.75, 134.33, 134.14, 132.27, 132.20, 131.34, 128.10, 125.73, 124.31, 124.16, 123.68, 122.97, 122.21, 102.88, 102.62, 40.21, 34.92, 33.63, 33.20, 32.12, 31.85, 31.82, 30.71, 30.12, 29.92, 29.88, 29.55, 29.36, 29.07, 26.85, 23.26, 22.89, 22.87, 22.81, 22.70, 14.39, 14.32, 14.30. MS (MALDI-TOF) m/z: [M+H]⁺, 1920.51;

Found, 1920.97. Anal. calcd (%) for C102H120I2O4S8: C 63.80, H 6.30; found: C 63.80, H 6.24.

101 Conclusion

A series of OSC donor SMs, 1–5, based on BDT-2T-ID analogue modified with various halogen groups were synthesized to reveal the halogenation effect on photovoltaic performances. BHJ-OSCs based on the halogenated SMs 2–5 as donors exhibited superior photovoltaic performance owing to 1.5 times higher FFs than those of the unsubstituted compound 1, resulting in the highest PCE of 9.2% for iodized SM 5 without any processing additives or additional treatments of the active layer. The comprehensive analysis for crystallinity and charge-transport properties of 1–5 films with and without existence of PC71BM revealed that the halogenated SMs 2–5 retained relatively higher crystallinity and hole mobility even in the blended state compared with 1. Moreover, morphological analysis and interfacial free energy measurement proved that better miscibility and smaller domain sizes for 4:PC71BM and 5:PC71BM than those for 2:PC71BM and 3:PC71BM, which originate from small interfacial free energy between heavy-halogen-modified SMs and PC71BM. This study provides a valuable insight into the structure–property relationship for photovoltaic SMs, and a molecular design strategy for further high-performance photovoltaic SMs and devices.

References

(1) Tang, M. L.; Bao, Z. Halogenated Materials as Organic Semiconductors. Chem. Mater. 2011, 23, 446–455.

(2) Yao, Z.; Wang, J.; Pei, J. Control of π–π Stacking via Crystal Engineering in Organic Conjugated Small Molecule Crystals. Cryst. Growth Des. 2018, 18, 7–15.

(3) Sakamoto, Y.; Suzuki, T.; Kobayashi, M.; Gao, Y.; Fukai, Y.; Inoue, Y.; Sato, F.; Tokito, S.

Perfluoropentacene:  High-Performance p−n Junctions and Complementary Circuits with Pentacene. J. Am.

Chem. Soc. 2004, 126, 8138–8140.

(4) Ponzini, F.; Zagha, R.; Hardcastle, K.; Siegel, J. S. Phenyl/Pentafluorophenyl Interactions and the Generation of Ordered Mixed Crystals: Sym-Triphenethynylbenzene and Sym-Tris(Perfluorophenethynyl)Benzene.

Angew. Chem. Int. Ed. 2000, 39, 2323–2325.

(5) Cho, D. M.; Parkin, S. R.; Watson, M. D. Partial Fluorination Overcomes Herringbone Crystal Packing in Small Polycyclic Aromatics. Org. Lett. 2005, 7, 1067–1068.

(6) Dou, J.; Zheng, Y.; Yao, Z.; Yu, Z.; Lei, T.; Shen, X.; Luo, X.; Sun, J.; Zhang, S.; Ding, Y.; Han, G.; Yi, Y.;

Wang, J.; Pei, J. Fine-Tuning of Crystal Packing and Charge Transport Properties of BDOPV Derivatives through Fluorine Substitution. J. Am. Chem. Soc. 2015, 137, 15947–15956.

(7) Wang, Y.; Parkin, S. R.; Gierschner, J.; Watson, M. D. Highly Fluorinated Benzobisbenzothiophenes. Org.

Lett. 2008, 10, 3307–3310.

(8) Sakamoto, Y.; Komatsu, S.; Suzuki, T. Tetradecafluorosexithiophene:  The First Perfluorinated Oligothiophene. J. Am. Chem. Soc. 2001, 123, 4643–4644.

(9) Dou, J.; Zheng, Y.; Yao, Z.; Lei, T.; Shen, X.; Luo, X.; Yu, Z.; Zhang, S.; Han, G.; Wang, Z.; Yi, Y.; Wang, J.-Y.; Pei, J. A Cofacially Stacked Electron-Deficient Small Molecule with a High Electron Mobility of over 10 cm² V⁻¹ s⁻¹ in Air. Adv. Mater. 2015, 27, 8051–8055.

(10) Hu, H.; Jiang, K.; Yang, G.; Liu, J.; Li, Z.; Lin, H. Terthiophene-Based D–A Polymer with an Asymmetric Arrangement of Alkyl Chains That Enables Efficient Polymer Solar Cells. J. Am. Chem. Soc. 2015, 137, 14149–14157.

(11) Chen, H.-Y.; Hou, J.; Zhang, S.; Liang, Y.; Yang, G.; Yang, Y.; Yu, L.; Wu, Y.; Li, G. Polymer Solar Cells with Enhanced Open-Circuit Voltage and Efficiency. Nat. Photon. 2009, 3, 649–653.

(12) Liang, Y.; Xu, Z.; Xia, J.; Tsai, S.; Wu, Y.; Li, G.; Ray, C.; Yu, L. For the Bright Future–Bulk Heterojunction Polymer Solar Cells with Power Conversion Efficiency of 7.4%. Adv. Mater. 2010, 22, E135–E138.

ドキュメント内 Development of Narrow-Bandgap π-Conjugated Small Molecules and Their Application in Solution-Processed Organic Solar Cells (ページ 81-106)