Effect of intermolecular interaction with phenothiazine core on inverted organic photovoltaics by using different acceptor moiety

Article abstract of DOI:10.1016/j.dyepig.2017.07.027

Six novel small molecules containing phenothiazine as the core moiety of p-type materials were synthesized in combination with PC₆₁BM or PC₇₁BM as n-type materials for bulk heterojunction organic photovoltaic (OPV) cells. Phenothiazine was used as donor building blocks to construct acceptor–donor–acceptor (A–D–A) and A–D–D–A small molecule. We investigated the relationship between the molecular structures, optoelectronic and thermal properties, and their blend film morphologies for OPV applications. We further studied their photophysical properties using a time-dependent density functional theory model and the B3LYP functional. A solution-processed OPV employing PT2-DE combined with the electron acceptor PC₆₁BM and additive DIO achieved optimal device performance, with a short-circuit current of 6.3 mA cm^?2, an open-circuit voltage of 770 mV, a fill factor of 0.55, and a corresponding overall conversion efficiency of 2.64%. To the best of our knowledge, this is the best-reported performance of a phenothiazine derivative for OPV application.

Full text of DOI:10.1016/j.dyepig.2017.07.027

Accepted Manuscript
Effect of intermolecular interaction with phenothiazine core on inverted organic
photovoltaics by using different acceptor moiety
Yu-Lung Weng, Yi-Chan Li, Chih-Ping Chen, Yuan Jay Chang
PII:
S0143-7208(17)31188-9
10.1016/j.dyepig.2017.07.027
DYPI 6116
DOI:
Reference:
To appear in: Dyes and Pigments
Received Date: 23 May 2017
Revised Date: 16 June 2017
Accepted Date: 14 July 2017
Please cite this article as: Weng Y-L, Li Y-C, Chen C-P, Chang YJ, Effect of intermolecular interaction
with phenothiazine core on inverted organic photovoltaics by using different acceptor moiety, Dyes and
Pigments (2017), doi: 10.1016/j.dyepig.2017.07.027.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Effect of intermolecular interaction with phenothiazine core on
inverted organic photovoltaics by using different acceptor moiety
Yu-LungWenga,†,Yi-Chan Lib,†, Chih-Ping Chenb, ,andYuan JayChanga,
aDepartment of Chemistry, Tunghai University, NO. 1727, Sec.4, Taiwan Boulevard, Xitun District, Taichung
40704, Taiwan
b Department of Materials Engineering, Ming Chi University of Technology, 84 Gungjuan Rd., Taishan Dist.
New Taipei City 24301, Taiwan
Abstract: Six novel small molecules containing phenothiazine as the core moiety of p-type materials were
synthesized in combination with PC61BM or PC71BM as n-type materials for bulk heterojunction organic
photovoltaic (OPV) cells. Phenothiazine was used as donor building blocks to construct
acceptor–donor–acceptor (A–D–A) and A–D–D–A small molecule. We investigated the relationship between
the molecular structures, optoelectronic and thermal properties, and their blend film morphologies for OPV
applications. We further studied their photophysical properties using a time-dependent density functional theory
model and the B3LYP functional. A solution-processed OPV employing PT2-DE combined with the electron
acceptor PC61BM and additive DIO achieved optimal device performance, with a short-circuit current of 6.3
mA·cm−2, an open-circuit voltage of 770 mV, a fill factor of 0.55, and a corresponding overall conversion
efficiency of 2.64%. To the best of our knowledge, this is the best-reported performance of a phenothiazine
derivative for OPVapplication.
Keywords. Inverted organic photovoltaics (OPVs); solution-processed bulk heterojunction (BHJ);
small-molecule organic photovoltaics (SMOPVs);A–D–A;A–D–D–A.
1

ACCEPTED MANUSCRIPT
*Corresponding author.
E-mail addresses: cpchen@mail.mcut.edu.tw (C.-P. Chen); jaychang@thu.edu.tw (Y. J. Chang)
†These authors contributed equallyto this work
1. Introduction
In recent years, because of the depletion of fossil fuels and increased greenhouse gas emissions that have
resulted in the problem of global warming, the investigation and development of alternative renewable and
environmentally friendly energy sources has becomeincreasingly important. In particular, solar energyhas been
acknowledged as a potential alternative energy resource. Organic photovoltaics (OPVs) convert sunlight
directly into electricity. OPV solar cells have attracted considerable attention because of their portability, easy
fabrication, energy level tunable by functional groups in the molecular structure, compatibility with flexible
substrates, and low-cost roll-to-roll processing [1–3].
A general preconception regarding OPVs is that the use of a conjugated polymer as a p-type material and
fullerene derivatives as a n-type material can achieve high power conversion efficiency (PCE). With the
advancement of device engineering and molecular design, polymer-based solar cells have been demonstrated in
excess of 12% by using a solution-processed bulk heterojunction (BHJ) [4–6]. Polymeric materials can be
roughly classified into two broad types: homopolymer and donor–acceptor copolymers. Donor–acceptor
copolymers exhibit a wide absorption region and favorable film morphologies, resulting in high device
performance [7, 8]. A comparison of polymeric materials reveals that small molecules (SMs) possess
predominant advantages such as well-defined molecular structures and higher reliability, higher reproducibility,
and easier energy level tuning and device fabrication than large molecules. The use of solution processing for
device fabrication is generally considered to enable large production through inject printing or roll-to-roll
2

ACCEPTED MANUSCRIPT
coating, which are more cost-effective than vacuum deposition. Therefore, several studies have investigated
small-molecule OPVs (SMOPVs), which have PCEs of more than 10%, by using solution-processed
fabrication techniques [9–11]. The high device performance achieved using SMOPVs suggests a new direction
for innovative ideas regarding the design of novel SMs with tailor-made structures and designated physical
characteristics. Novel small-molecule-based p-typematerials in OPVs have been used to develop a widevariety
of promising organic materials with different donor–acceptor moiety combinations such as symmetrical
acceptor–donor–acceptor (A−D−A) [12–14] and donor–acceptor–donor (D−A−D) [15–17] combinations.
Systems with PCEs similar to those of polymeric systems have been developed [9]. A variety of photovoltaic
materials with prominent chemical frameworks that were produced using solution-processed deposition have
been reported; these include oligothiophene [18, 19], fused polycycle [20–22], diketopyrrollopyrrole [23, 24],
phathalocyanine [25, 26], and isoindigo [27, 28].
Weak intermolecular interactions between SMs of p-type materials can hamper charge transport, resulting in
a low fill factor (FF)[29]. Intermolecular interaction is a key factor that determines the stacking structures of a
p-type material; thus, it significantly affects electrical properties and device performance, and it can result in the
enhancement of FF and PCE. Extending π-stacking moieties and increasing coplanarity enhances the
intermolecular interactions between p-type materials. Although several studies have investigated the influence
of π–π stacking interactions between SMs of p-type material on the performance of OPVs, few studies have
focused on other intermolecular interactions such as hydrogen bonding or dipole–dipole interaction [17, 30].
Phenothiazine has been shown to be a good electron-donating moiety, and it is used in organic light-emitting
diodes [31] and dye-sensitized solar cells [32–34].However, few studies have successfullydemonstrated the use
of phenothiazine as a building block for OPVs [34, 35]. Maglione et al. used phenothiazine as donor blocks and
constructed a D–A–D material for OPV application that had a PCE of 0.7%, the highest reported for pheno-
3

ACCEPTED MANUSCRIPT
thiazine derivatives [36]. In the present study, we used phenothiazine as an electron-donating moiety, coupled
with three acceptors applied to A–D–A and A–D–D–A dipolar systems as p-type materials. We fabricated an
inverted OPV device with indium tin oxide (ITO)/ZnO/active layers/MoO3/Ag using solution-processed
deposition; the OPV device employing material PT2-DE combined with electron acceptor PC61BM and
1,8-diiodooctane (DIO) achieved a high PCE of 2.64%. To the best of our knowledge, this is the highest ever
PCE reported for phenothiazine-based SMs for OPV application. The high PCE can be attributed to the high Voc
of 0.77 V.
2. Results and discussion
2.1 Synthesis ofA–D–AandA–D–D–Amolecules
The structures of phenothiazine-based SMs and their synthetic sequences are presented in Figure 1 and
Scheme 1. Molecule 1 was synthesized first; one or two bromo substituents were subsequently added through a
bromination reaction to yield 2 or 4. The number of core structure units was increased by adding another
phenothiazine moiety through a Stille coupling reaction; thereafter, bromination reaction was performed to yield
6. Athiophene unit, used to extendπ-conjugation, was connected to2 and 6 through a Stille coupling reaction to
yield 3 and 7 [37]. Thesix p-type materials were obtained through Knoevenagel condensation to build three
types of acceptor unit by using malononitrile, ethyl 2-cyanoacetate, and 2-cyano-n-hexylacetamide [38]. All the
new compound structures were characterized using spectroscopy.
2.2 Thermal stability
4

ACCEPTED MANUSCRIPT
Favorable thermal stability is essential for p-typeorganic materials because photovoltaic devices areoperated
under sunlight in a relatively high-temperature environment. The thermal properties of the PT-series were
investigated using thermal gravimetric analysis (Figure 2 and Table 1). The thermal stability of the SMs
decreased in the order PT2-series > PT1-series for the core moiety and DCA > DAM > DE for the acceptor
unit. The decomposition temperature Td (corresponding to 5% weight loss) of these compounds, which was
greater than 350 °C, indicated the good thermal stability of these phenothiazine-based SMs. PT1-DCN and
PT2-DCN exhibited relatively superior thermal stability with decomposition temperatures of 395 and 404 °C
under N2 conditions
2.3 Optoelectronic properties of the molecules
The optical absorption spectra of all PT-series materials in methylene chloride solution and those of their thin
films are presented in Figures 3 and S13, and their photochemical and electrochemical parameters are listed in
Table 1. The spectra of all p-type materials exhibit two broad and high-intensity absorption peaks in the range
360–563 nm. Similar absorption was discovered for the materials in solution as for those in film form; the
short-wavelength region of 360–391 nm is attributed to localized π-π* and n-π* transitions, whereas the
long-wavelength region of 486–563 nm is due to intramolecular charge-transfer (ICT) transitions. The ICT
transitions decreased in the order PT1-series > PT2-series for the core moiety and DCN > DE > DAM for the
acceptor unit. We assumed that the dicyanovinylene group possesses a stronger electron-withdrawing ability
than the ester and amide moieties. The solid state of ICT transition in thin film form showed a bathochromatic
shift of approximately 6–31 nm with respect to that in DCM solution, which indicated intermolecular
J-aggregation on the surface of the substance. Thespectra of the compounds PT1-DCN and PT2-DCN have a
large absorption region and large full width at half maximum of the ICT band.
The ionic potentials (IPs) corresponding to the highest occupied molecular orbitals (HOMO) of the molecules
were measured using cyclic voltammetry (Figures 3, S18, and S19). The electron affinity levels of the
molecules were estimated from the HOMO levels and zero–zero bandgaps at the onset of the absorption spectra
(Table 1). Clearly, in the dipolar system of the PT-series p-type materials, the electron-withdrawing moiety
5

ACCEPTED MANUSCRIPT
strongly affects the IPs and electron delocalization. The structure skeleton contains the dicyanovinylene group
because the electron-withdrawing groups of PT1-DCN and PT2-DCN were shown to have larger IP energy
levels than the ethyl cyanoacrylate and cyano-N-hexylacrylamide groups. The compounds PT1-DCN and
PT2-DCN also had a narrow HOMO–lowest unoccupied molecular orbital (LUMO) energy gap, which is
consistent with the wide absorption ranges and more red-shifted ICT absorption bands illustrated in Figure 3.
For proper operation of an OPV containing it, the compound PT1-DCN had the deepest HOMO energy level
and its small bandgap should give corresponding devices a high Voc and Jsc.
2.4 Computational analysis
The electronic characteristics of PT-series SMs were investigated using theoretical models. Full geometrical
optimizations were performed using the B3LYP/6-31G*hybrid functional, which was implemented in Q-Chem
3.0 [39]. The optimized molecular geometry of the PT-series SMs is shown in Figures 4 and S17. Clearly, the
phenothiazine core unit of the adjacent thiophene rings was slightly twisted; however, the orientation of the
adjacent phenothiazine in the PT2-series was twisted by 36°–38° because of steric hindrance. This slight
distortion in SM conformation did not inhibit electron transport between the electron donor and electron
acceptor moieties. As depicted in Figures 4 and S16, the electron densities of the HOMOs were mainly
distributed around the phenothiazine donor sites, whereas those of the LUMOs were distributed around the
acceptor moieties. Accordingly, HOMO to LUMO transitions contributed more than 92% to the first-singlet
excited state (S1 state) and oscillator strength (f) beyond 0.6 in all PT-series materials.
The bar chart in Figure 3c indicates that PT1-DCN had a higher HOMO energy level than the other
compounds; the trends PT1-DCN > PT1-DE ≈ PT1-DAM and PT2-DCN > PT2-DE ≈ PT2-DAM are
consistent with the computational results. The energy gap of the p-type materials could be verified from the
dipole moment: PT1-DCN (8.5314 D) and PT2-DCN (11.0752 D) had a larger dipole moment than the others
(Table S1); the trends PT1-DCN > PT1-DE > PT1-DAM and PT2-DCN > PT2-DE > PT2-DAM are
consistent with the absorption results both for theSMs in methlyene chloride and in film form. The large dipole
moment of the molecules can be attributed to the strong electron-donating moiety of dicyanovinylene in
6

ACCEPTED MANUSCRIPT
PT1-DCN and PT2-DCN. The p-type material PT1-DCN had the deepest HOMO energy level (−5.51 eV)
and the largest gap (p-typeHOMO−n-typeLUMO) to the LUMO energy level of the n-type materials PC61BM or
PC71BM; this result is consistent with the experiment data obtained through CV (−5.19 eV). We expected that
the PT1-DCN would have a high Voc in a corresponding OPV device, and this was verified by fabricating
devices using solution-processed deposition.
2.5 Photovoltaic performance of OPVs
We fabricated OPV devices with a layer structure of ITO [Sanyo, Japan (8 Ω square−1)]/ZnO/active
layers/MoO3/Ag and an active area of 10 mm2. The detailed fabrication conditions are provided in the
supporting information (SI). The ZnO layer (40 nm thick) was prepared from a 0.5 M zinc acetate precursor
solution in 2-methoxyethanol and deposited on top of the ITO [40, 41]. The ZnO layer was annealed in air at
160 °C for 30 mins prior to use. The active layer solution of PT-series SMs containing PC61BM or PC71BM
(1:1.5) was stirred in CHCl3 with or without DIO overnight, filtered through apolytetrafluoroethylene (0.2 µm)
filter, and then spin-coated onto the ZnO layer. Incident photocurrent conversion efficiency (IPCE) and
photocurrent–voltage (J–V) curves were measured for all molecules under AM 1.5 solar light (100 mW·cm−2);
Jsc, Voc, FF, and PCE are presented in Tables 2 and 3, and the J–V curves and IPCEs of the constructed OPV
devices are shown in Figures 5 and 6. The PCE of PT1-DCN/PC61BM, PT1-DE/PC61BM, and
PT1-DAM/PC61BM was 0.85%, 0.52%, and 0.51%, respectively. The slightly higher PCE of the PT1-DCN
device can be attributed to its higherJsc, which is consistent with its lower energy bandgap. Figure 5b illustrates
the IPCE response of the devices. For the PT1-DE and PT1-DAM devices, the IPCE response occurred at
wavelengths 400–700 nm, whereas that for the PT1-DCN device was in the range 400−750 nm. These
intensities and IPCE responses are consistent with the corresponding Jsc and UV-vis spectra. When a small
amount of additive (DIO) was added, we observed a significant improvement in the PCE of the
PT1-DCN/PC61BM devices. This phenomenon was not observed in thePT1-DE and PT1-DAM devices. The
PCE of the PT1-DCN/PC61BM device improved from 0.85% to 1.97% because of the significant increase in
Jsc and FF. The PCE further improved to 2.42% when we replaced the PC61BM with PC71BM. Figure 5d
7

ACCEPTED MANUSCRIPT
illustrates the IPCE of PT1-DCN/PC61BM in the presence of the 1.5 vol% of DIO; the IPCE was increased in
the range 400–750 nm compared with the normal device. The IPCE increase from 650 to 750 nm suggested
efficient π−π molecular packing in the PT1-DCN SMs. Comparing the PC61BM device with the PC71BM
device revealed a significant improvement in IPCE from 400 to 600 nm, primarily because of the light
harvesting of PC71BM at visible wavelengths. The PCEs of PT2-DCN/PC61BM, PT2-DE/PC61BM, and
PT2-DAM/PC61BM were 0.36%, 0.39%, and 0.12%, respectively. We observed the highest Jsc for the
PT2-DCN device; this value is consistent with the lower energy bandgap (from UV-vis spectrum) and IPCE
result (Figures 3 and 6b). When we added a small amount of additive (DIO), we observed a significant
improvement in the PCE of the PT2-DE/PC61BM device. This phenomenon was not observed for the
PT2-DCN and PT2-DAM devices. The PCE of the PT2-DE/PC61BM device improved from 0.36% to 2.64%
in the presence of DIO because of the significant increase in Jsc and FF. We observed a similar PCE of 2.31%
for the PT2-DE/PC71BM device. These performance differences were due to the different solubility of the
PT-series materials in the DIO.
To determine the changes in film morphology when using DIO as an additive, we performed
tapping-mode atomic force microscopy (AFM). Figures 7a−c present the AFM images of the PT1-DCN BHJ
films. From the topographical images, we determined the arithmetical mean surface roughness (Ra) of the
PT1-DCN/PC61BM, PT1-DCN/PC61BM(DIO), and PT1-DCN/PC71BM (DIO) blend films to be 0.83, 3.97,
and 5.51 nm, respectively. The surface of PT1-DCN/PC61BM was featureless and exhibited a very small
surface roughness, which may be due to the well-mixed PT1-DCN and PC61BM. The high surface roughnesses
obtained may have been due to the crystallization of PT1-DCN molecules. This phenomenon was significant
for the PT2-DE blend film. The Ra of PT2-DE/PC61BM and PT2-DE/PC71BM was 0.318 and 0.274 nm,
respectively, and increased to 18.1 and 19.6 nm in the presence of DIO. A grain size greater than 1 µm was
determined for the PT2-DE-rich domains. Typically, a well-defined blend morphology is required for a highly
efficient OPVdevice; the domain sizes of the donorand acceptor need to be segregated on the nanoscale. In this
study, we observed microscale segregation of the PT2-DE:fullerene blend film with 1.5 vol% of DIO, which
resulted in high OPV performance. This phenomenon is not common; we speculate that the carrier mobility in
8

ACCEPTED MANUSCRIPT
the normal PT2-DE devices (w/o DIO) is low, despite the nanoscale blend film morphology. The incorporation
of DIO may significantly increase the carrier mobility; we obtained a high PCE even for microscale phase
segregation.
To understand the cause of the differences in device performance, we performed hole-only space charge
limited
current
(SCLC)
measurements
for
devices
having
the
configuration
ITO/PEDOT:PSS/PT-SMs/MoO3/Ag [40, 42]. The SCLC hole mobility of PT1-DE, PT1-DAM, and
PT1-DCN was 7.5 × 10−10, 9.5 × 10−11, and 3.9 × 10−9 m2 V−1 s−1, respectively. The highest SCLC hole mobility
was obtained for the PT1-DCN devices; this result is consistent with their higher PCE performance. To
investigate the cause of the increase in device PCEin the presence of DIO, we further fabricated SCLC devices
based on the PT1-DCN blend films. The hole mobility of PT1-DCN:PC61BM and PT1-DCN:PC71BM was
1.3 × 10−10 and 3.2 × 10−9 m2 V−1 s−1, respectively. In the presence of DIO, we observed increases in the SCLC
hole mobility for PT1-DCN:PC61BM (2.5 × 10−10 m2 V−1 s−1) and PT1-DCN:PC71BM (6.5 × 10−9 m2 V−1 s−1).
These results suggest that DIO modified the blend film morphology, increasing hole transport efficiency; thus,
we observed an enhancement in the PCE of the devices. The SCLC mobility of the pristine PT2-DCN,
PT2-DE, and PT2-DAM was 3.3 × 10−10, 2.3 × 10−10, and 2.2 × 10−10m2 V−1 s−1, respectively. We observed
similar mobility for materials that exhibited similar OPV device performance. The hole mobility of devices
based on PT2-DE blend films was 4.6 × 10−11 and 1.1 × 10−11 m2 V−1 s−1 for the PC61BM- and PC71BM-based
blends, respectively. Again, we observed an increase in the hole mobility after the addition of DIO; the mobility
of devices increased to 4.4 × 10−9 and 7.1 × 10−10 m2 V−1 s−1 for the PC61BM and PC71BM blends, respectively.
The main factor that determined the carrier mobility was presumably the differences in the morphologies of the
active layers; in this case, these differences weredue to superior molecular packing (Figure 7). To confirm this,
we performed grazing incidence X-ray diffraction (XRD) analysis and studied the UV-vis spectra of the related
blend films. The XRD patterns of PT2-DE:PC61BM are shown in Figure 8a. The blend film without DIO
exhibited an amorphous characteristic and a broader peak at the 2θ between 15° and 25°. We observed an
increase in the structural ordering of the PT2-DE:PC61BM blend in the presence of DIO. The PT2-DE
molecules showed a strong (100) reflection peak at 2θ = 4.4° corresponding to ad [100]-spacing of 20.0 Å.
9

ACCEPTED MANUSCRIPT
This value indicated the efficient π−π packing of the PT2-DE molecules when the blend film was formed from
the DIO-containing solution. This distance covers the range between the coplanar of the phenothiazine main
chains of PT2-DE separated by the flexible side chains. Figure 8b presents the UV-vis spectra of the blend
films. A red-shift of 150 nm was observed in the spectra of the blend films when DIO had been added,
indicating highly organized PT2-DE molecules in the solid state. Large-scale phase-segregation morphology
was observed for the PT2-DE−DIO blend films because of the crystalline-grain-sized increase in the PT2-DE
molecules; therefore, we obtained an improvement of greater than one order of magnitude in the SCLC hole
mobility. As a result, a 5-fold increase in the PCE of the PT2-DE-based OPV devices was determined. We
observed similar phenomena for the PT1-DCN blend film. In the presence of DIO, the PT1-DCN film
exhibited a strong (100) reflection peak at 2θ = 5°, a (200) reflection peak at 2θ = 10°, and a (300) reflection
peak at 15°, indicating the ordering structure in the blend and again a red-shift in the UV-vis spectrum. The
performance of the PT1-DCN–based device increased because of a significant improvement in the mobility. In
the future, we expect to further increase the performance of phenothiazine-based materials through optimized
device engineering. Determining the order of molecular orientation (from variations in materials and blend
system to variations in device engineering) can be challenging; our XRD data confirmed the enhancement in
the molecular order for the blend films in the presence of DIO and we observed improvement in the device
performance (PCE and hole mobility). Further understanding the molecular arrangement and self-assembly
behavior of these phenothiazine-based materials has been carried out in the laboratory. By effective controlling
the molecular packing of these materials will havethe opportunityto further improvethe device performance.
3. Conclusion
In summary, we demonstrated that PT-series p-type materials utilize phenothiazine moieties as
electron-donating (D) groups with three different acceptors (dicyanovinyl, ethyl 2-cyanoacrylate, and
2-cyano-N-hexylacrylamide) by employing A–D–A and A–D–D–A dipolar systems as the p-type materials in
photovoltaic solar cells. By substituting different molecules as the A unit, we obtained molecules with various
10

ACCEPTED MANUSCRIPT
optoelectronic properties.
A well-defined blend morphology is required for a highly efficient OPV device; the domain sizes of the
donor and acceptor need to be segregated on the nanoscale. In this study, we observed microscale segregation of
the PT2-DE:fullerene blend film with 1.5 vol% of DIO, which resulted in high OPV performance.
Incorporating PT2-DE/PC61BM in an OPV cell with the additive DIO resulted in a PCE of 2.64% and Voc was
0.77 V, Jsc was 6.3 mAcm−2, and the FF was 54.7%; the PCE was thus increased 5-fold when DIO was added.
This improvement was due to the superior π−π molecular packing in the PT2-DE (as determined using XRD)
and resulted in a one-order-of-magnitude increase in hole mobility. We also observed a similar PCE of 2.31%
for the PT2-DE/PC71BM device. Because of differences in the solubility of the PT materials in DIO, the
performance of the devices varied.
4. Experimental Section
4.1 General Information
All reactions and manipulations were performed in a nitrogen atmosphere. Solvents were distilled freshly
according to standard procedures. 1H and 13C NMR spectra were recorded on a Bruker AVIII HD 400 MHz
spectrometer, and CDCl3 was used as solvents. Chemical shifts were reported on the δ scale downfield from the
tetramethylsilane peak. Absorption spectra were recorded on a SHIMADZU UV-1800 and Jasco V-630
spectrophotometer, and emission spectra were obtained using a Hitachi F-4500 spectrofluorimeter. Emission
spectra in solutions were measured using a spectral grade solvent and right-angle detection. Redox potentials
were measured using cyclic voltammetry on a CHI 620 analyser. All measurements were performed in THF
solution containing 0.1 M tetrabutylammonium hexaflourophosphate as a supporting electrolyte in ambient
conditions after purging the solution with N2 for 10 min. The conventional three-electrode configuration,
consisting of a glassy carbon working electrode, a platinum counter electrode, and an Ag/Ag+ reference
11

ACCEPTED MANUSCRIPT
electrode calibrated with ferrocene/ ferrocenium (Fc/Fc+) as an internal reference, was employed. Mass spectra
were recorded on a JEOL JMS-700 double-focusing mass spectrometer. The surface morphology of the
materials was characterized by the non-contact mode atomic force microscopy (Bruker Dimension Edge
microscope operated in the dynamic force mode at ambient temperature.). Thermogravimetric analysis (TGA;
Pyris 1 TGA Lab system) measurements were carried out with heating rate of 10 °C/min under a nitrogen gas
atmosphere.
The chemicals N-Bromosuccinimide (NBS), trans-Dichlorobis(triphenylphosphine) palladium(II)
(PdCl2(PPh3)2), Dichloromethane, tetrakis(triphenylphosphine)palladium(0) (P d(PPh3)4), n-Butyllithium (1.6
M in hexane), Triisopropyl borate N,N’-Dimethylformamide, Malononitrile, Ethyl cyanoacetate,
2-Cyano-N-hexylacetamide, Ammonium acetate, Tetrahydrofuran, Toluene, and Piperidine were purchased
from Acros, Alfa, Merck, Lancaster, TCI, Sigma-Aldrich, and Showa, separately, and purified when necessary.
Chromatographic separations were performed using silica gel from Merck (Kieselgel Si 60; 40–63µm).
4.2 Synthesis of PT-series p-type materials
4.2.1 3,7-Dibromo-10-(4-(dodecyloxy)phenyl)-10H-phenothiazine (2).
A mixture of 1 (5.0 g, 10.8 mmol) in dry CH2Cl2 was placed in a three-necked flask under a nitrogen
atmosphere. Following this, added dropwise NBS (3.84 g, 21.6 mmol) in CH2Cl2 to this reaction and stirred 6
hours room temperature. The mixture was quenched by adding DI water, and the mixture extracted with
CH2Cl2. The organic layer was dried over anhydrous MgSO4 and evaporated under vacuum. The product was
purified by silica gel column chromatograph with hexane as eluent. Yellow solid of 2 was obtained in 97%
yield (6.5 g, 10.5 mmol). 1H NMR (400 MHz, CDCl3): δ 7.22 (d, 2H, J = 8.8 Hz), 7.09 (d, 2H, J = 8.8 Hz), 7.07
(d, 2H, J = 2.2 Hz), 6.90 (dd, 2H, J = 8.8, 2.2 Hz), 6.01 (d, 2H, J = 8.8 Hz), 4.02 (t, 2H, J = 6.5 Hz), 1.80-1.86
(m, 2H), 1.46-1.51 (m, 2H), 1.27-1.36 (m, 16H), 0.88 (t, 3H, J = 6.8 Hz). 13C NMR (100 MHz, CDCl3): δ 159.2,
12

ACCEPTED MANUSCRIPT
143.5, 132.2, 131.7, 129.7, 128.7, 121.0, 116.9, 116.7, 114.5, 68.4, 31.9, 29.7, 29.6, 29.6, 29.5, 29.4, 29.3, 29.2,
26.1, 26.7, 14.1. MS (FAB, 70 eV): m/z (relative intensity) 615 (M+, 100); HRMS calcd for C30H35NOS79Br2:
615.0806, found 615.0818.
4.2.2 5,5'-(10-(4-(Dodecyloxy)phenyl)-10H-phenothiazine-3,7-diyl)bis(thio phene-2-carbaldehyde) (3).
To a three-necked flask containing a mixture of 2 (5.0 g, 8.1 mmol), PdCl2(PPh3)2 (190 mg, 0.25 mmol),
(5-(1,3-dioxolan-2-yl)thiophen-2-yl) tributylstannane (4.0 g, 9.0 mmol) was added DMF (50 mL). The reaction
mixture was stirred at 90 °C for 24 h. After cooling, the reaction was quenched by adding methanol and KF(aq)
(saturated 45 mL) Evaporation of the solvent gave a crude product. The crude mixture was added Acetic acid/
.
THF/ de-ionized water (40/20/10 mL) at 60 °C. After3 hours, the reaction was quenched by adding de-ionized
water and brine. The mixture was extracted with CH2Cl2 and the organic layer dried overanhydrous MgSO4.
Evaporation of the solvent gave the crude, which was purified by silica gel with CH2Cl2 as eluent. Red solid of
3 was obtained in 63.6% (3.5 g, 5.1 mmol). 1H NMR (400 MHz, CDCl3): δ 9.75 (s, 2H), 7.58 (d, 2H, J = 3.8
Hz), 7.12-7.18 (m, 6H), 7.08 (d, 2H, J = 8.72 Hz), 7.01 (d, 2H, J = 8.8 Hz), 6.08 (d, 2H, J = 8.6 Hz), 4.00 (t, 2H,
J = 6.36 Hz), 1.79-1.83 (m, 2H), 1.45-1.49 (m, 2H), 1.28-1.31 (m, 16H), 0.87 (t, 3H, J = 6.36). 13C NMR (100
MHz, CDCl3): δ 182.2, 159.1, 152.7, 144.1, 141.3, 137.4, 131.6, 131.3, 127.3, 125.0, 123.7, 122.8, 119.5, 116.6,
115.9, 68.2, 31.7, 29.5, 29.4, 29.3, 29.1, 29.0, 25.9, 22.5, 14.0. MS (FAB, 70 eV): m/z (relative intensity) 679
(M+, 100); HRMS calcd forC40H41NO3S3: 679.2249, found 679.2245.
4.2.3 3-Bromo-10-(4-(dodecyloxy)phenyl)-10H-phenothiazine (4).
Compound 4 was synthesized according to the same procedure as that of 2. Yellow solid was obtained in
46% yield. 1H NMR (400 MHz, CDCl3): δ 7.24 (d, 2H, J = 8.12 Hz), 7.07-7.11 (m, 3H), 6.96 (dd, 1H,J = 7.24,
13

ACCEPTED MANUSCRIPT
1.94 Hz), 6.89 (dd, 1H, J = 8.8, 2.2 Hz), 6.76-6.85 (m, 2H), 6.18 (dd, 1H, J = 7.96, 1.4 Hz), 6.02 (d, 2H, J = 8.8
Hz), 4.02 (t, 2H, J = 6.5 Hz), 1.80-1.87 (m, 2H), 1.46-1.52 (m, 2H), 1.28-1.31 (m, 16H), 0.89 (t, 3H, J = 6.9 Hz).
13C NMR (100 MHz, CDCl3): δ 159.0, 144.3, 143.9, 132.6, 132.0, 129.4, 128.6, 127.1, 126.6, 122.5, 121.8,
118.8, 116.8, 116.8, 116.5, 115.7, 114.168.4, 31.9, 29.7, 229.6, 29.5, 29.4, 29.4, 29.3, 26.1, 22.7, 14.1. MS
(FAB, 70 eV): m/z (relative intensity) 537 (M+, 100); HRMS calcd for C30H36NOS79Br: 537.1701, found
537.1691.
4.2.4 10,10'-Bis(4-(dodecyloxy)phenyl)-10H,10'H-3,3'-biphenothiazine (5).
To a three-necked round-bottom flask containing 4 (2.5 g, 4.6 mmol) was added dropwise BuLi (5.0 mL,
8.0 mmol, 1.6 M in hexane) in dry THF at -78 ºC, after then the solution was brought up to 0ºC and wasstirred
by a magnetic bar for 30 minutes. The solution was cooled again to -78 ºC and to it was added dropwise
triisopropyl borate (2.2 mL, 8.0 mmol). The reaction was warmed up gradually to room temperature and was
stirred overnight. To the reaction was then added excess amount of 10% HCl(aq)(30 mL), while the mixture was
stirredfor another 1 h. The reaction was quenched by pouring into distilled water, followed by extraction with
ethyl acetate. The organic layer was dried overanhydrous MgSO4. Evaporation of the solvent gave a crude
product, which was immediately subjected to the next reaction. It was mixed with 4 (2.5 g, 4.6 mmol),
K2CO3(aq) (2.76 g, 2 mmol) in 10 mL H2O, and Pd(PPh3)4(190 mg, 0.16 mmol) in dry toluene/THF (2/1). The
mixture was heated to 60 °C for 12 h. After cooling, the products were extracted with ethyl acetate and the
organic layer dried overanhydrous MgSO4. The crude product was driedin vacuo, and was purifies by silica gel
column chromatograph eluted with CH2Cl2/hexane (1/8). Yellow solid of 5 was obtained in 72% yield (3.1 g,
3.3 mmol). 1H NMR (400 MHz, CDCl3): δ 7.29 (d, 4H, J = 8.72 Hz), 7.11-7.09 (m, 6H), 6.99 (d, 6H, J = 7.0
Hz), 6.84-6.81 (m, 4H), 6.19 (d, 4H, J = 7.9 Hz), 4.03 (t, 4H, J = 6.5 Hz), 1.87-1.80 (m, 4H), 1.52-1.46 (m, 4H,),
14

ACCEPTED MANUSCRIPT
1.40-1.28 (m, 32H), 0.91 (t, 6H, J = 6.4 Hz); 13C NMR (100 MHz, CDCl3): δ 158.9, 143.6, 134.0, 132.9, 132.1,
126.8, 126.6, 124.5, 124.1, 122.2, 120.0, 119.2, 116.4, 115.8, 115.6, 77.3, 77.0, 76.7, 68.3, 31.9, 29.7, 29.6, 29.6,
29.4, 29.3, 29.3, 26.1, 22.7, 14.1. MS (FAB, 70 eV): m/z (relative intensity) 916 (M+, 100); HRMS calcd for
C60H72N2O2S2: 916.5035, found 916.5020.
4.2.5 7,7'-Dibromo-10,10'-bis(4-(dodecyloxy)phenyl)-10H,10'H-3,3'-biphenothiazine (6).
Compound 6 was synthesized according to the same procedure asthat of 2. Yellow solid of6 was obtained
in 86% yield. 1H NMR (400 MHz, CDCl3): δ 7.22-7.25 (m, 4H), 7.05-7.10 (m, 8H), 6.88(dd, 4H, J = 2.06 Hz,
8.78 Hz), 6.15 (d, 2H, J = 8.5 Hz), 6.01 (d, 2H, J = 8.6 Hz), 4.02 (t, 4H, J = 6.4 Hz), 1.81-1.85 (m, 4H),
1.46-1.51 (m, 4H), 1.28-1.38 (m, 32H), 0.89 (t, 6H, J = 6.6 Hz). 13C NMR (100 MHz, CDCl3): δ 159.0, 143.6,
1432, 134.0, 132.5, 131.9, 129.4, 128.6, 124.8, 124.1, 121.4, 119.2, 116.8, 116.5, 115.9 ,114.2, 68.4, 31.9, 29.7,
29.6, 29.6, 29.5, 29.4, 29.3, 29.3, 26.1, 22.7, 14.1. MS (FAB, 70 eV): m/z (relative intensity) 1072 (M+, 100);
HRMS calcd for C60H70Br2N2O2S2: 1072.3245, found 1072.3219.
4.2.6 5,5'-(10,10'-Bis(4-(dodecyloxy)phenyl)-10H,10'H-[3,3'-bipheno-thiazine]-7,7'-diyl)bis(thiophene-
2-carbaldehyde) (7).
Compound 7 was synthesized according to the same procedure asthat of 3. Yellow solid of7 was obtained
in 42% yield. 1H NMR (400 MHz, CDCl3): δ 9.82 (s, 2H),7.65 (dd, 2 H, J = 0.78 Hz, 3.89 Hz) 7.24-7.27 (m,
6H), 7.21 (s, 1H), 7.20 (s, 1H,), 7.08-7.12 (m, 8H), 6.89 (d, 2H, J = 7.84 Hz), 6.16 (dd, 4H, J = 4.06Hz, 8.34
Hz), 4.03 (t, 4H, J = 6.4 Hz), 1.87-1.81 (m, 4H), 1.55-1.47 (m, 4H), 1.4-1.28 (m 32H), 0.89 (t, 6H, J = 6.4 Hz);
13C NMR (100 MHz, CDCl3): δ 182.4, 159.1, 153.3 ,145.1, 142.6, 141.4, 137.5, 134.2, 131.7, 127.1, 125.1,
124.7, 124.0, 122.8, 120.0, 119.2, 116.6, 116.1, 115.7, 68.4, 31.9, 29.6, 29.6, 29.6, 29.4, 29.3, 29.2, 26.1, 22.7.,
15

ACCEPTED MANUSCRIPT
14.1; MS (FAB, 70 eV): m/z (relative intensity) 1136 (M+, 100); HRMS calcd for C70H76N2O4S4: 1136.4688,
found 1136.4645.
4.2.7 2,2'-(((10-(4-(Dodecyloxy)phenyl)-10H-phenothiazine-3,7-diyl) bis(thiophene-5,2-diyl))bis-
(methanylylidene))dimalononitrile (PT1-DCN).
A mixture of 3 (0.5 g, 0.7 mmol), NH4Oac (0.1 g, 13 mmol), and malononitrile (0.2 g, 3.0 mmol) were
placed in a three-necked flask containing dry toluene (20 mL) under N2. The resulted mixture was heated to
reflux for 24 hours. After cooling, the reaction was quenched by pouring into water and it was extracted with
CH2Cl2. The combined organic layer was dried over anhydrous MgSO4 and evaporated under vacuum. The
mixture was purified by silica gel column chromatograph with CH2Cl2 as eluent. The black solid of PT1-DCN
was collected in 81% yield (0.47 g, 0.6 mmol). 1H NMR (400 MHz, CDCl3): δ 8.25(s, 2H), 7.68 (d, 2 H, J =
4.12 Hz) 7.24-7.28 (m, 2H), 7.14 (d, 4H, 8.6 Hz), 7.13 (d, 4H, J = 8.8 Hz), 6.16 (d, 2H, J = 8.6Hz), 4.05 (t, 2H,
J = 6.5 Hz), 1.82-1.89 (m, 2H), 1.28-1.38 (m 22H), 089 (t, 6H, J = 6.6 Hz); 13C NMR (100 MHz, CDCl3): δ
159.5, 155.2, 150.2, 144.9, 140.1, 133.5, 131.5, 131.3, 127, 125.7, 124.2, 123.6, 119.9, 116.9, 116.3, 114.3,
113.5, 75.8, 68.5, 31.9, 29.6, 29.6, 29.6, 29.4, 29.3, 29.2, 26.0, 22.7, 14.1; MS (FAB, 70 eV): m/z (relative
intensity) 775 (M+, 100); HRMS calcd for C46H41N5OS3: 775.2473, found 775.2473.
4.2.8 Diethyl 3,3'-((10-(4-(dodecyloxy)phenyl)-10H-phenothiazine-3,7-diyl)bis(thiophene-5,2-diyl))(2E,2'E)-
bis(2-cyanoacrylate) (PT1-DE).
Compound PT1-DE was synthesized according to the same procedure as that of PT1-DCN. Red solid of
PT1-DE was obtained in 83% yield. 1H NMR (400 MHz, CDCl3): δ 8.25 (s, 2H),7.68 (d, 2 H, 4.12 Hz),
7.24-7.28 (m, 6H), 7.13 (dd, 4H, J = 8.76, 1.96 Hz), 6.16 (d, 2H, J = 8.6Hz), 4.35 (q, 4H, J = 7.15 Hz), 4.05 (t,
16

ACCEPTED MANUSCRIPT
2H, J = 6.5 Hz), 1.82-1.89 (m, 2H), 1.28-1.38 (m 22H), 089 (t, 6H, J = 6.6 Hz); 13C NMR (100 MHz, CDCl3):
δ 163.0, 159.4, 153.3, 146.3, 144.4, 139.2, 134.1, 1317., 131.4, 127.3, 125.3, 124.0, 123.3, 119.8, 116.8, 116.1,
97.3, 68.4, 62.3, 31.9, 29.6, 29.6, 29.9, 29.4, 29.3, 29.2, 26.0, 22.6, 14.2, 14.1. MS (FAB, 70 eV): m/z (relative
intensity) 869 (M+, 100); HRMS calcd for C50H51N3O5S3: 869.2991, found 869.2986.
4.2.9 (2E,2'E)-3,3'-((10-(4-(Dodecyloxy)phenyl)-10H-phenothiazine-3,7-diyl)bis(thiophene-5,2-diyl))bis-
(2-cyano-N-hexylacrylamide) (PT1-DAM).
Compound PT1-DAM was synthesized according to the same procedure as that of PT1-DCN. Red solid
of PT1-DAM was obtained in 62% yield. 1H NMR (400 MHz, CDCl3): δ 8.23 (s, 2H), 7.60 (d, 2 H, J = 4.08
Hz), 7.25-7.27 (m, 4H), 7.22 (d, 2 H, J = 4.04 Hz), 7.10-7.14 (m, 4H), 6.23 (t, 2H,J = 5.68 Hz), 6.15 (d, 2H, J =
8.6 Hz), 4.04 (t, 2H, J = 6.48 Hz), 3.40 (q, 4H, J = 6.92 Hz), 1.81-1.88 (m, 2H), 1.47-1.62 (m, 6H), 1.28-1.38
(m, 28H), 0.86-0.91 (m, 9H); 13C NMR (100 MHz, CDCl3): δ 160.6, 159.3, 151.9, 144.4, 144.4, 144.3, 138.4,
134.6, 131.8, 1315, 127.5, 125.3, 124.0, 123.2, 119.8, 117.5, 116.7, 116.1, 98.8, 68.4, 40.5, 31.9, 31.4, 29.6,
29.6, 29.6, 29.4, 29.3, 29.2, 26.5, 26.0, 22.6, 22.5, 14.1, 13.9; MS (FAB, 70 eV): m/z (relative intensity) 979
(M+, 100); HRMS calcd forC58H69N5O3S3: 979.4563, found 979.4561.
4.2.10 2,2'-(((10,10'-Bis(4-(dodecyloxy)phenyl)-10H,10'H-[3,3'-bipheno-thiazine]-7,7'-diyl)bis(thiophene-5,2-
diyl))bis(methanylylidene)) dimalononitrile (PT2-DCN).
Compound PT2-DCN was synthesized according to the same procedure asthat of PT1-DCN. Black solid
of PT2-DCN was obtained in 70% yield. 1H NMR (400 MHz, CDCl3): δ 7.71 (s, 2H), 7.63 (d, 2 H, J = 4.2 Hz),
7.24-7.26 (m, 8H), 7.12 (d, 6H, J = 8.8 Hz), 7.07 (d, 2H, J = 2.0Hz), 6.92 (dd, 2H, J = 8.6, 2.4 Hz), 6.16 (dd, 4H,
J = 8.6, 2.4Hz), 4.04 (t, 4H, J = 6.4 Hz), 1.81-1.88 (m, 4H), 1.47-1.54 (m, 4H), 1.28-1.38 (m, 32H), 0.89-0.86 (t,
17

ACCEPTED MANUSCRIPT
6H, J = 6.8 Hz); 13C NMR (100 MHz, CDCl3): δ 159.3, 155.8, 150.2, 145.9, 142.4, 140.2, 134.4, 133.2, 132.0,
131.6, 126.2, 125.5, 124.8, 124.2, 124.1, 123.3, 120.3, 119.1, 116.7, 116.3, 115.8, 114.4, 113.6, 77.3, 77.2, 77.0,
76.7, 75.4, 68.4, 31.9, 29.7, 29.6, 29.6, 29.6, 29.4, 29.3, 29.2, 26.1, 22.7, 14.1. MS (FAB, 70 eV): m/z (relative
intensity) 1232 (M+, 100); HRMS calcd for C76H76N6O2S4: 1232.4913, found 1232.4927.
4.2.11 Diethyl 3,3'-((10,10'-bis(4-(dodecyloxy)phenyl)-10H,10'H-[3,3'-biphenothiazine]-7,7'-
diyl)bis(thiophene-5,2-diyl))(2E,2'E)-bis(2-cyanoacrylate) (PT2-DE).
Compound PT2-DE was synthesized according to the same procedure as that of PT1-DCN. Black solid
of PT2-DE was obtained in 56% yield. 1H NMR (400 MHz, CDCl3): δ 8.24 (s, 2H), 7.68 (d, 2H, J = 4.08 Hz),
7.23-7.28 (m, 8H), 7.08-7.14 (m, 8H), 6.93 (dd, 2H, J = 2.0 Hz, 8.6 Hz), 6.17 (dd, 4H, J = 8.6, 3.24Hz), 4.35 (q,
4H, J = 7.08 Hz), 4.04 (t, 4H, J = 6.44 Hz), 1.82-1.89 (m, 4H), 1.47-1.53 (m, 4H), 1.28-1.40 (m, 38H), 089 (t,
6H, J = 6.96 Hz); 13C NMR (100 MHz, CDCl3): δ 163.1, 159.2, 153.8, 146.4, 145.4, 142.6, 139.3, 134.3, 134.0,
132.2, 131.7, 126.8, 125.3, 124.8, 124.1, 123.1, 120.2, 119.2, 116.7, 116.2, 115.7, 97.1, 68.4, 62.3, 31.9, 29.7,
29.6, 29.6, 29.4, 29.3, 29.3, 26.1, 22.7, 14.2, 14.1. MS (FAB, 70 eV): m/z (relative intensity) 1326 (M+, 100);
HRMS calcd for C80H86N4O6S4: 1326.5430, found 1326.5415.
4.2.12 (2E,2'E)-3,3'-((10,10'-bis(4-(dodecyloxy)phenyl)-10H,10'H-[3,3'-biphenothiazine]-7,7'-
diyl)bis(thiophene-5,2-diyl))bis(2-cyano-N-hexylacrylamide) (PT2-DAM).
Compound PT2-DAM was synthesized according to the same procedure as that of PT1-DCN. Red solid
of PT2-DAM was obtained in 83% yield. 1H NMR (400 MHz, CDCl3): δ 8.31(s, 2H), 7.60 (d, 2 H, J = 4.08
Hz), 7.26-7.27 (m, 4H), 7.19 (d, 2H, J = 4.0 Hz), 7.08-7.13 (m, 8H), 6.92 (dd, 2H, J = 8.6, 1.72 Hz), 6.22 (t, 2H,
J = 5.52 Hz), 6.16 (dd, 4H, J = 8.6, 3.96 Hz), 4.04 (t, 4H, J = 6.36 Hz), 3.40 (q, 4H, J = 6.72 Hz), 1.81-1.88 (m,
18

ACCEPTED MANUSCRIPT
4H), 1.47-1.55 (m, 8H), 1.28-1.38 (m, 44H), 0.86-0.88 (m, 12H); 13C NMR (100 MHz, CDCl3): δ 160.7, 159.2,
152.4, 145.2, 144.4, 142.6, 138.4, 134.4, 134.3, 132.3, 131.8, 127.0, 125.1, 124.7, 124.1, 124.1, 123.0, 120.1,
119.2, 117.6, 116.6, 116.2, 115.7, 98.6, 68.4, 40.6, 31.9, 31.4, 29.7, 29.6, 29.6, 29.4, 29.3, 29.3, 26.5, 16.1, 22.7,
22.5, 14.1, 14.0; MS (FAB, 70 eV): m/z (relative intensity) 1436 (M+, 100); HRMS calcd for C88H104N6O4S4:
1436.7002, found 1436.6996.
4.3 Quantum Chemistry Computations
All organic molecules were optimized by using B3LYP/6-31G* hybrid functional. For excited states, the
time-dependent density functional theory (TDDFT) and the B3LYP functional were used. All analyses were
performed using the Q-Chem 3.0 software. The frontier orbital plots of the highest and lowest occupied
molecular orbitals (hereafter abbreviated as HOMO and LUMO, respectively) were drawn using GaussView
04.
Acknowledgements
This work was financial supported by the Ministry of Science and Technology of Taiwan (MOST
103-2113-M-029-007-MY2, 105-2119-M-029-001-), Tunghai University, Ming Chi University of Technology
and Academia Sinica are gratefully acknowledged. Special thanks to Professor C.-P. Hsu at the Institute of
ChemistrySinica ofTaiwan for her assistance on quantum.
Appendix. Supplementary data
1
13
The H and C NMR spectra of all compounds, absorption spectra in methylene chloride solution, the
absorption spectra and emission spectra on thinfilm, TDDFT calculated orbitals, Oxidative voltammograms
and AFM images. Supplementary data associated with this article can be found in the online version, at doi:
19

ACCEPTED MANUSCRIPT
xxx.
References
[1] Pastorelli F, Schmidt TM, Hösel M, Søndergaard RR, Jørgensen M, Krebs FC. The Organic Power
Transistor: Roll-to-Roll Manufacture, Thermal Behavior, and Power Handling When Driving Printed
Electronics.Advanced Engineering Materials. 2016;18:51-5.
[2] Emmott CJM, Moia D, Sandwell P, Ekins-Daukes N, Hösel M, Lukoschek L, et al. In-situ, long-term
operational stability of organic photovoltaics foroff-grid applications in Africa. Solar Energy Materials and
Solar Cells. 2016;149:284-93.
[3] Wang K, Liu C, Meng T, Yi C, Gong X. Inverted organic photovoltaic cells. Chem Soc Rev.
2016;45:2937-75.
[4] Yi C, Hu X, Gong X, Elzatahry A. Interfacial engineering for high performance organic photovoltaics.
Materials Today. 2016;19:169-77.
[5] Li S, Ye L, Zhao W, Zhang S, Mukherjee S, Ade H, et al. Energy-Level Modulation of Small-Molecule
ElectronAcceptors toAchieve over 12% Efficiencyin Polymer Solar Cells.Adv Mater. 2016;28:9423-9.
[6] Zhao J, Li Y, Yang G, Jiang K, Lin H, Ade H, etal. Efficient organic solar cells processed from hydrocarbon
solvents. Nature Energy. 2016;1:15027.
[7] Lu L, Zheng T, Wu Q, Schneider AM, Zhao D, Yu L. Recent Advances in Bulk Heterojunction Polymer
Solar Cells. Chem Rev. 2015;115:12666-731.
[8] Zhang S, Ye L, Hou J. Breaking the 10% Efficiency Barrier in Organic Photovoltaics: Morphology and
Device Optimization ofWell-Known PBDTTTPolymers.Advanced EnergyMaterials. 2016;6:1502529.
[9] Collins SD, Ran NA, Heiber MC, Nguyen T-Q. Small is Powerful: Recent Progress in Solution-Processed
Small Molecule Solar Cells.Advanced EnergyMaterials. 2017:1602242-n/a.
[10] Kan B, Li M, Zhang Q, Liu F, Wan X, Wang Y, et al. ASeries of Simple Oligomer-like Small Molecules
Based on Oligothiophenes for Solution-Processed Solar Cells with High Efficiency. J Am Chem Soc.
2015;137:3886-93.
20

ACCEPTED MANUSCRIPT
[11] Zhang Q, Kan B, Liu F, Long G, Wan X, Chen X,et al. Small-molecule solar cells with efficiencyover 9%.
Nat Photon. 2015;9:35-41.
[12] Sim J, Lee H, Song K, Biswas S, Sharma A, Sharma GD, et al. Solution processed bulk heterojunction
solar cells based on A-D-Asmall molecules with a dihydroindoloindole (DINI) central donor and different
acceptor end groups. Journal of Materials ChemistryC. 2016;4:3508-16.
[13] Chung C-L, Chen C-Y, Kang H-W, Lin H-W, Tsai W-L, Hsu C-C, et al. A–D–A type organic donors
employing coplanar heterocyclic cores for efficientsmall molecule organic solar cells. Organic Electronics.
2016;28:229-38.
[14] Vegiraju S, Hsieh C-M, Huang D-Y, Chen Y-C, Priyanka P, Ni J-S, et al. Synthesis and characterization of
solution-processable diketopyrrolopyrrole (DPP) and tetrathienothiophene (TTA)-based small molecules
for organic thin film transistors and organic photovoltaic cells. Dyes and Pigments. 2016;133:280-91.
[15] Li Z, Dong Q, Li Y, Xu B, Deng M, Pei J, et al. Design and synthesis of solution processable small
molecules towards high photovoltaic performance. Journal of Materials Chemistry. 2011;21:2159-68.
[16] Zitzler-Kunkel A, Lenze MR, Schnier T, Meerholz K, Würthner F. Comparative Studies on Optical, Redox,
and Photovoltaic Properties of a Series of D–A–D and Analogous D–A Chromophores. Advanced
Functional Materials. 2014;24:4645-53.
[17] Aytun T, Barreda L, Ruiz-Carretero A, Lehrman JA, Stupp SI. Improving Solar Cell Efficiency through
Hydrogen Bonding: A Method for Tuning Active Layer Morphology. Chemistry of Materials.
2015;27:1201-9.
[18] Li Z, He G, Wan X, Liu Y, Zhou J, Long G, et al. Solution Processable Rhodanine-Based Small Molecule
Organic Photovoltaic Cells with a Power Conversion Efficiency of 6.1%. Advanced Energy Materials.
2012;2:74-7.
[19] Long G, Wan X, Kan B, Liu Y, He G, Li Z, et al. Investigation of Quinquethiophene Derivatives with
Different End Groups for High Open Circuit Voltage Solar Cells. Advanced Energy Materials.
2013;3:639-46.
[20] Winzenberg KN, Kemppinen P, Fanchini G, Bown M, Collis GE, Forsyth CM, et al.
21

ACCEPTED MANUSCRIPT
Dibenzo[b,def]chrysene Derivatives: Solution-Processable Small Molecules that Deliver High
Power-Conversion Efficiencies in Bulk Heterojunction Solar Cells. Chemistry of Materials.
2009;21:5701-3.
[21] Lloyd MT, Mayer AC, Subramanian S, Mourey DA, Herman DJ, Bapat AV, et al. Efficient
Solution-Processed Photovoltaic Cells Based on an Anthradithiophene/Fullerene Blend. J Am Chem Soc.
2007;129:9144-9.
[22] Wong WWH, Singh TB, Vak D, Pisula W, Yan C, Feng X, et al. Solution Processable Fluorenyl
Hexa-peri-hexabenzocoronenes in Organic Field-Effect Transistors and Solar Cells. Advanced Functional
Materials. 2010;20:927-38.
[23] Tamayo AB, Walker B, Nguyen* T-Q. A Low Band Gap, Solution Processable Oligothiophene with a
Diketopyrrolopyrrole Core for Use in Organic Solar Cells. The Journal of Physical Chemistry C.
2008;112:11545-51.
[24] Kumar CHP, Ganesh K, Suresh T, Sharma A, Bhanuprakash K, Sharma GD, et al. Influence of thermal
and solvent annealing on the morphology and photovoltaic performance of solution processed, D-A-D
type small molecule-based bulk heterojunction solarcells. RSCAdvances. 2015;5:93579-90.
[25] Dittmer JJ, Lazzaroni R, Leclère P, Moretti P,Granström M, Petritsch K, et al. Crystal network formation in
organic solar cells. Solar Energy Materials and Solar Cells. 2000;61:53-61.
[26] Mayukh M, Macech MR, Placencia D, Cao Y, Armstrong NR, McGrath DV. Solution Processed Titanyl
Phthalocyanines as Donors in Solar Cells: Photoresponse to 1000 nm. ACS Applied Materials & Interfaces.
2015;7:23912-9.
[27] Elsawy W, Lee C-L, Cho S, Oh S-H, Moon S-H, Elbarbary A, et al. Isoindigo-based small molecules for
high-performance solution-processed organic photovoltaic devices: the electron donating effect of the
donor group on photo-physical properties and deviceperformance. Physical Chemistry Chemical Physics.
2013;15:15193-203.
[28] Stalder R, Mei J, Graham KR, Estrada LA, Reynolds JR. Isoindigo, a Versatile Electron-Deficient Unit For
High-Performance OrganicElectronics. ChemistryofMaterials. 2014;26:664-78.
22

ACCEPTED MANUSCRIPT
[29] Lee OP, Yiu AT, Beaujuge PM, Woo CH, Holcombe TW, Millstone JE, Douglas JD, Chen MS, Fréchet
JMJ.Adv Mater. 2011;23:5359-63.
[30] Kim K-H, Yu H, Kang H, Kang DJ, Cho C-H, Cho H-H, et al. Influence of intermolecular interactions of
electron donating small molecules on their molecular packing and performance in organic electronic
devices. Journal of Materials ChemistryA. 2013;1:14538-47.
[31]AhnY, Jang DE, ChaY-B, Kim M,Ahn K-H, KimYC. Bull. Korean Chem. Soc. 2013;34:107-11.
[32] Marszalek M, Nagane S, Ichake A, Humphry-Baker R, Paul V, Zakeeruddin SM, et al. Tuning spectral
properties of phenothiazine based donor-[small pi]-acceptor dyes for efficient dye-sensitized solar cells.
Journal of Materials Chemistry. 2012;22:889-94.
[33] Chang YJ, Chou P-T, Lin Y-Z, Watanabe M, Yang C-J, Chin T-M, et al. Organic dyes containing
oligo-phenothiazine for dye-sensitized solar cells.Journal of Materials Chemistry. 2012;22(40):21704-12.
[34] Luo J-S, Wan Z-Q, Jia C-Y. Recent advances in phenothiazine-based dyes for dye-sensitized solar cells.
Chinese Chemical Letters. 2016;27:1304-18.
[35] Li K-C, Hsu Y-C, Lin J-Ts, Yang C-C, Wei K-H,Lin H-C. Novel narrow-band-gap conjugated copolymers
containing phenothiazine-arylcyanovinyl units for organic photovoltaic cell applications. Journal of
Polymer Science PartA: Polymer Chemistry. 2008;46:4285-304.
[36] Maglione C, Carella A, Centore R, Chávez P, Lévêque P, Fall S, et al. Novel low bandgap phenothiazine
functionalized DPP derivatives prepared by direct heteroarylation: Application in bulk heterojunction
organic solar cells. Dyes and Pigments. 2017;141:169-78.
[37] Stille JK. The Palladium-Catalyzed Cross-Coupling Reactions of Organotin Reagents with Organic
Electrophiles [New Synthetic Methods (58)]. Angewandte Chemie International Edition in English.
1986;25:508-24.
[38] Knoevenagel E. Ber. Dtsch. Chem. Ges. 1898;31:2596-2619.
[39] Shao Y, Molnar LF, Jung Y, Kussmann J, Ochsenfeld C, Brown ST, et al. Advances in methods and
algorithms in a modern quantum chemistry program package. Physical Chemistry Chemical Physics.
2006;8(27):3172-91.
23

ACCEPTED MANUSCRIPT
[40] Juang T-Y, Hsu Y-C, Jiang B-H, Chen C-P. Highly Efficient Inverted Organic Photovoltaics Containing
Aliphatic Hyperbranched Polymers as Cathode Modified Layers. Macromolecules. 2016;49(20):7837-43.
[41] Chang C-C, Jiang B-H, Li Y-C, Chen C-P, Cheng C-H. Indacenodithiophene-based conjugated polymers
incorporating alkylthiophene side chains: Improvement of organic solar cell performance. Dyes and
Pigments. 2017;142:139-46.
[42] Jiang B-H, Peng Y-J, Chen C-P. Simple Structured Polyetheramines, Jeffamines, as Efficient Cathode
Interfacial Layers for Organic Photovoltaics Providing Power Conversion Efficiencies up to 9.1%. Journal
of Materials ChemistryA. 2017.
[b]
[c]
HOMO/LUMO[a] Band gap[a]
(eV)
f [a]
λmax(sol)
λmax(film) (nm) Td
IP[d]
Eg[e] EA[f]
P-type
(℃)
(eV) (eV) (eV)
material
-5.51/-3.02
-5.24/-2.63
-5.15/-2.45
-5.13/-2.81
-4.94/-2.47
-4.88/-2.29
2.49
2.61
2.70
2.32
2.47
2.59
0.8344 386, 537
0.8797 378, 511
0.9192 372, 496
0.6098 380, 532
0.6752 367, 502
0.6488 360, 486
379, 549
374, 521
369, 502
391, 563
375, 525
366, 505
395 -5.19 1.93 -3.26
355 -5.15 2.00 -3.15
371 -5.13 2.08 -3.05
404 -5.04 1.91 -3.13
373 -4.99 2.03 -2.96
386 -5.00 2.11 -2.89
PT1-DCN
PT1-DE
PT1-DAM
PT2-DCN
PT2-DE
PT2-DAM
Table 1. Photochemical and electrochemical parameters of donor materials.
[a] TDDFT/B3LYP calculated values. [b] Absorptionin DCM and film. [c] Td point obtained from TGA. [d] Ionic potentials weremeasured using cyclic voltammetry. [e]Eg
calculatedbytheonsetofabsorptionspectra.[f]EAcalculatedbyIP–Eg.
24

ACCEPTED MANUSCRIPT
Active layer
(weight ratio)
Voc
Jsc
FF
PCE
RS
RSH
(V)
(mA·cm-2)
(%)
(%)[c]
(Ω cm2) (kΩ cm2)
PT1-DCN/PC61BM[a]
0.75
0.77
0.78
0.72
0.78
0.69
0.80
3.6
6.0
7.8
2.2
2.3
2.2
1.6
30.8
42.7
39.8
32.5
33.3
33.0
36.0
0.85
1.97
2.42
0.52
0.59
0.51
0.48
75
35
284
452
248
434
515
511
769
PT1-DCN/PC61BM[b]
PT1-DCN/PC71BM[b]
PT1-DE/PC61BM[a]
PT1-DE/PC61BM[b]
PT1-DAM/PC61BM[a]
PT1-DAM/PC61BM[b]
33
167
186
172
167
[a]p-type/n-typeratioof1/1.5.[b]withDIOadditive(1.5%).
Table 2 Photovoltaic parameters of devices made PT1-series p-type materials with PC61BM and PC71BM as
active layer.
25

ACCEPTED MANUSCRIPT
Table 3 Photovoltaic parameters ofPT2-derived devices.
Active layer
Voc
Jsc
FF
PCE
(%)[c]
RS
RSH
(weight ratio)
(V)
(mA·cm-2)
(%)
(Ω cm2) (kΩ cm2)
PT2-DCN/PC61BM[a]
PT2-DCN/PC61BM[b]
PT2-DE/PC61BM[a]
PT2-DE/PC61BM[b]
PT2-DE/PC71BM[b]
PT2-DAM/PC61BM[a]
PT2-DAM/PC61BM[b]
0.43
0.42
0.81
0.77
0.73
0.39
0.33
2.3
2.4
1.5
6.3
6.3
1.0
0.5
39.6
40.0
32.0
54.7
50.4
33.5
33.0
0.36
0.40
0.39
2.64
2.31
0.12
0.05
87
52
394
350
797
1082
704
655
111
223
3.9
4.3
233
431
[a]p-type/n-typeratioof1/1.5.[b]withDIOadditive(1.5%).
26

ACCEPTED MANUSCRIPT
Figure Captions
Scheme 1.
Synthesis
route
of
PT-series
SM.
i)
NBS,
CH2Cl2;
ii)
(5-(1,3-dioxolan-2-yl)thiophen-2-yl)tributyl-stannane, PdCl2(PPh3)2, DMF, 110 °C; iii) a. n-BuLi/ triisopropyl
borate, -78oC; followed by 3M HCl.; b. Pd(PPh3)4, compound 4, 2M K2CO3, toluene/THF=1.5/1; iv)
.
Malononitrile or Ethyl cyanoacetate or 2-Cyano-N-hexylacetamide, CH3COONH4, THF, reflux/ Piperidine,
Toluene, reflux.
Fig. 1. The structures ofPT-series for OPVs p-type materials.
Fig. 2. Thermogravimetric analysis ofPT-series with a heating rate 10 min-1 under N2.
Fig. 3. Absorption spectra of molecules in (a) DCM solution; (b) thin films, and (c) energy diagrams of these
p-type materials and other materials use in this study.
Fig. 4. Computed optimum structure and frontier molecular orbitals of PT1-DCN and PT2-DE calculated with
TDDFT (B3LYP/6-31G*) in vacuum.
Fig. 5. J-V curves and EQE responses of thePT1-based OPVs devices.
Fig. 6. J-V curves and EQE of the SMOPVs devices ofPT2-series with PC61BM and PC71BM as active layer.
Fig. 7. AFM (a) PT1-DCN/PC61BM; (b) PT1-DCN/PC61BM/DIO; (c) PT1-DCN/PC71BM/DIO; (d)
PT2-DE/PC61BM; (e) PT2-DE/PC61BM/DIO; (f) PT2-DE/PC71BM; (g) PT2-DE/PC71BM/DIO. (2μm x 2
μm)
Fig. 8. a) XRD patterns and b) UV-vis spectra of thePT2-DE and PT1-DCN based blend films in the presence
of and without the DIO.
27

ACCEPTED MANUSCRIPT