Evaluation of semiconducting molecular thin films solution-processed via the photoprecursor approach: The case of hexyl-substituted thienoanthracenes

Article abstract of DOI:10.1039/c5tc00794a

Organic electronic devices are expected to be easily scalable and highly cost-effective, presuming the good solution processability of high-performance organic semiconductors. However, there are cases where an organic compound with promising semiconducting properties lacks adequate processability and does not form well-performing thin films through conventional solution-based deposition techniques. The photoprecursor approach, in which a soluble photoprecursor is solution-deposited on a substrate and then converted to a target material by in situ photoreaction, can be an effective means to evade such a problem. Herein, we describe a comparative evaluation of thin films deposited by three different methods; i.e., vacuum deposition, photoprecursor approach, and direct spin coating. Two highly crystalline molecular semiconductors, hexyl-substituted anthra[1,2-b:4,3-b′:5,6-b′′:8,7-b′′′]tetrathiophene (C₆-ATT) and anthra[1,2-b:5,6-b′]dithiophene (or bent anthradithiophene, C₆-BADT), are employed in this study along with the corresponding newly synthesized α-diketone-type photoprecursors. In the case of C₆-ATT, thin films prepared through the photoprecursor approach are as good as those obtained by vacuum deposition in terms of surface smoothness and space-charge-limited-current (SCLC) mobility, while direct spin coating affords highly inhomogeneous films. For C₆-BADT, on the other hand, employment of the photoprecursor approach is not as effective, albeit it is still advantageous as compared to direct spin coating. These results highlight the power and limitations of the photoprecursor approach, and will serve as a basis for the preparation of practically useful organic devices through this unique approach.

Full text of DOI:10.1039/c5tc00794a

Journal of
Materials Chemistry C
View Article Online
View Journal
PAPER
Evaluation of semiconducting molecular thin films
solution-processed via the photoprecursor
approach: the case of hexyl-substituted
thienoanthracenes†
Cite this:DOI: 10.1039/c5tc00794a
Cassandre Quinton,a Mitsuharu Suzuki,*a Yoshitaka Kaneshige,a Yuki Tatenaka,ab
Chiho Katagiri,c Yuji Yamaguchi,c Daiki Kuzuhara,a Naoki Aratani,a
Ken-ichi Nakayamacd and Hiroko Yamada*ad
Organic electronic devices are expected to be easily scalable and highly cost-effective, presuming the
good solution processability of high-performance organic semiconductors. However, there are cases
where an organic compound with promising semiconducting properties lacks adequate processability
and does not form well-performing thin films through conventional solution-based deposition techniques.
The photoprecursor approach, in which a soluble photoprecursor is solution-deposited on a substrate and
then converted to a target material by in situ photoreaction, can be an effective means to evade such a
problem. Herein, we describe a comparative evaluation of thin films deposited by three different methods;
i.e., vacuum deposition, photoprecursor approach, and direct spin coating. Two highly crystalline mole-
cular semiconductors, hexyl-substituted anthra[1,2-b:4,3-b0:5,6-b00:8,7-b0 0 0]tetrathiophene (C6-ATT) and
anthra[1,2-b:5,6-b0]dithiophene (or bent anthradithiophene, C6-BADT), are employed in this study along
with the corresponding newly synthesized a-diketone-type photoprecursors. In the case of C6-ATT, thin
films prepared through the photoprecursor approach are as good as those obtained by vacuum deposition
in terms of surface smoothness and space-charge-limited-current (SCLC) mobility, while direct spin
coating affords highly inhomogeneous films. For C6-BADT, on the other hand, employment of the
photoprecursor approach is not as effective, albeit it is still advantageous as compared to direct spin
coating. These results highlight the power and limitations of the photoprecursor approach, and will serve as
a basis for the preparation of practically useful organic devices through this unique approach.
Received 21st March 2015,
Accepted 11th May 2015
DOI: 10.1039/c5tc00794a
www.rsc.org/MaterialsC
promising semiconducting properties are not well compatible with
conventional direct solution deposition, typically because of too
1. Introduction
Solution processability is an important consideration in designing strong self-interactions. Such interactions often lead to undesirable
organic semiconductors, as facile and large-scale preparation of formation of large aggregates, or even complete insolubility in
high-quality active layers is the key to achieving cost-effective customary solvents for intended solution-deposition processes.
organic devices.1–8 However, some molecular compounds with Unsubstituted acenes9 and tetrabenzoporphyrin10–14 represent this
class of compounds, and their use in solution-processed devices
a Graduate School of Materials Science, Nara Institute of Science and Technology
has been largely limited owing to the lack of general approach to
obtain smooth, homogeneous films that can be favourably used in,
for example, organic solar cells or light-emitting diodes.15
(NAIST), 8916-5 Takayama-cho, Ikoma, Nara 630-0192, Japan.
E-mail: hyamada@ms.naist.jp, msuzuki@ms.naist.jp
b Dojindo Laboratories, 2025-5 Tabaru, Mashiki-machi, Kumamoto 861-2202, Japan
c Department of Organic Device Engineering, Yamagata University, 4-3-16 Jonan,
Yonezawa, Yamagata 992-8510, Japan
The precursor approach, in which a well-soluble precursor
compound is solution-deposited and then converted in situ
to a target semiconductor, has been proposed to evade the above-
mentioned obstacle. Among various types of precursor compounds,
a-diketone-type photoprecursors of acenes are characterized by the
extremely mild reaction conditions for their post-deposition conver-
sion, where only visible-light irradiation is required.16 Several
a-diketone-type precursors have been employed in the preparation
of organic field-effect transistors and photovoltaic cells comprising
d Core Research for Evolutional Science and Technology (CREST), Japan Science and
Technology Agency (JST), 4-1-8 Honcho, Kawaguchi, Saitama 332-0012, Japan
† Electronic supplementary information (ESI) available: Photoreactions in solution,
photoelectron spectroscopy in air, cyclic voltammetry, DFT calculations, simulated
powder X-ray diffraction parameters, space-charge-limited-current measurements,
NMR spectra of photoprecursors 1 and 2. CCDC 1052010, 1052256 and 1052257. For
ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/
c5tc00794a
This journal is ©The Royal Society of Chemistry 2015
J. Mater. Chem. C

View Article Online
Paper
Journal of Materials Chemistry C
Fig. 1 Structures of C6-ATT, C6-BADT, and their a-diketone-type photo-
precursors 1 and 2.
hardly soluble acene compounds.17–21 In the case of pentacene
and 2,6-di(2-thienyl)anthracene, this approach can afford high-
quality semiconducting thin films that are comparable to the
corresponding vacuum-deposited films in terms of charge-carrier
mobility.19,21
Meanwhile, to our knowledge, there is not yet a report on direct
comparison between the photoprecursor approach and conven-
tional solution processes. Such a comparison will answer the
question of whether the photoprecursor approach is merely a
means to enable indirect solution deposition of insoluble mate-
rials, or holds any substantial difference from the direct solution
deposition regarding the quality of resulting thin films. With
these in mind, we have prepared a-diketone-type photoprecursors of
two known soluble p-type semiconductors, tetrahexylanthra[1,2-
b:4,3-b0:5,6-b00:8,7-b00 0]tetrathiophene (C6-ATT) and dihexylanthra-
[1,2-b:5,6-b0]dithiophene (C6-BADT) (Fig. 1). Derivatives of ATT and
BADT have been the subject of interest in several studies on charge-
carrier transport characteristics.22–28 These thieno-annulated anthra-
cene systems have also been examined as building units of polymeric
semiconductors.27,29
In the following sections, we describe the synthesis and reactivity
of the a-diketone-type photoprecursors of C6-ATT and C6-BADT
(compounds 1 and 2, respectively), as well as the difference between
the thin films deposited through vacuum deposition, the photo-
precursor approach, or direct spin coating. It is shown that, in the
case of C6-ATT, the photoprecursor approach can provide thin films
whose structural and (opto)electronic characteristics are similar to
those of the vacuum deposited counterparts, while direct spin
coating affords considerably lower quality films having large grains
and cracks. The other compound, C6-BADT, is indicated to have
stronger self-interactions compared to C6-ATT, and employment of
the photoprecursor approach is not sufficient to obtain smooth films
under the examined conditions. Nevertheless, the C6-BADT films
prepared by the photoprecursor approach are still of higher quality
than the directly spin-coated samples when the same solvent system
and spin-coating conditions are employed.
Fig. 2 (a) Synthetic route to a-diketone-type photoprecursor 1 through
C6-ATT. (b) Thermal ellipsoid representation of compound 1 in the single-
crystalline state. Ellipsoids are at 50% probability. For clarity, only the first
carbon atom is shown for each hexyl substituent, and the hydrogen atoms
are omitted.
3 with iron(III) chloride,25,30 employing slightly modified conditions
from the reported ones to obtain an improved yield of 86%. C6-ATT is
well soluble in common organic solvents (e.g., 410 mg mLꢀ1
in chloroform), allowing straightforward purification by column
chromatography. The Diels–Alder reaction of C6-ATT with vinylene
carbonate was conducted by heating at 180 1C in an autoclave for
3 days to afford carbonate 4 in 76% yield. Compound 4 was
hydrolyzed under basic conditions in 98% yield, and then
subjected to Swern oxidation to afford the target a-diketone 1
as a yellow powder in an isolation yield of 28% after purification
by silica gel column chromatography and recrystallization.
The synthesis of photoprecursor 2 was performed as sum-
marized in Fig. 3. The indium-catalyzed cyclization developed
by Fu¨rstner et al.31 afforded C6-BADT in 90% yield, which is as
soluble as C6-ATT. This protocol gave higher yields and repro-
ducibility in our hands, and allowed easier scale up compared
to the known photochemical reaction employed in the cyclization
of 2,5-bis(2-thienyl)-1,4-divinylbenzene.27 Similar to compound 1,
the target a-diketone-type photoprecursor 2 was synthesized from
C6-BADT via the Diels–Alder addition with vinylene carbonate,
hydrolysis under basic conditions, and Swern oxidation using
trifluoroacetic anhydride (TFAA) as an activator of dimethyl
sulfoxide (DMSO). After purification by silica gel column chroma-
2. Results and discussion
2.1. Synthesis
The synthetic pathway to photoprecursor 1 is outlined in Fig. 2. tography and recrystallization, compound 2 was isolated as a
C6-ATT was synthesized by the known oxidation of tetrathienylbenzene yellow powder in 57% yield.
J. Mater. Chem. C
This journal is ©The Royal Society of Chemistry 2015

View Article Online
Journal of Materials Chemistry C
Paper
Fig. 3 (a) Synthetic route to a-diketone-type precursor 2 through C6-BADT.
(b) Thermal ellipsoid representation of compound 2 in the single-crystalline
state. Ellipsoids are at 50% probability. For clarity, only the first carbon atom is
shown for each hexyl substituent, and hydrogen atoms are omitted.
Hence, the syntheses of new photoprecursors 1 and 2 were
completed in few steps from the corresponding thienoanthra-
cenes C6-ATT and C6-BADT, respectively. The final products and
intermediates were unambiguously identified by 1H and 13C
NMR, and mass spectroscopic measurements. In addition, the
structures of 1 and 2 were confirmed by single-crystal X-ray
analysis as shown in Fig. 2b and 3b (see also Fig. S6 in ESI†).
C6-ATT, C6-BADT, compounds 1 and 2 were all purified by
recrystallization, and their purities were checked to be 499%
by high-performance liquid chromatography prior to use in the
preparation of thin films.
Fig. 4 Change in (a) NMR and (b) UV–vis absorption during the conver-
sion of 1 to C6-ATT under photoirradiation. The NMR spectra were taken in
CDCl3, while UV-vis spectra were measured in toluene. See ESI† for
Experimental details.
The reaction quantum yields (Fr) were determined based on
the change in UV–vis absorption at the initial stage of each
photoreaction (see the Experimental section for the method).
The results show that increased thieno-annulation leads to
lower efficiency of the photoreaction; namely, the Fr values in
toluene are 0.37 for the parent anthracene diketone (ADK),
0.25 for the doubly thieno-annulated derivative 2, and 0.21
for the quadruply thieno-annulated derivative 1 (Table 1). At
the same time, the Fr values of precursors 1 and 2 are still
considerably higher than those of tetracene diketone (TDK,
Fr E 0.1) and pentacene diketone (PDK, Fr = 0.027). Here, Fr
rapidly decreases with linear extension of the acene framework
from ADK to TDK and to PDK, while the nonlinear thieno-
annulation from ADK to 1 and to 2 results in relatively minor
decrease in Fr.
2.2. Photoreactivity
a-Diketone-type photoprecursors 1 and 2 are smoothly con-
verted to the corresponding anthracene derivatives C6-ATT and
C6-BADT, respectively, upon photoirradiation in solution. The
quantitative one-to-one conversion was confirmed by monitoring
the reactions with NMR and UV–vis absorption spectroscopy. As
shown in Fig. 4a, precursor 1 in CDCl3 shows a 1H NMR peak at
5.50 ppm corresponding to the bridgehead protons (Ha), which
decreases and eventually disappears over the curse of the photo-
reaction. Concomitantly, a peak emerges at 8.67 ppm, which can
be ascribed to the anthracene protons (Hb) of C6-ATT. The
reaction completed without formation of any detectable side
products in these measurements. The change in photoabsorption
is associated with a clear isosbestic point at 439 nm (Fig. 4b),
indicating the constant reaction stoichiometry and the absence of
a secondary reaction. The quantitative conversion of 2 was also
shown by the same experiments (Fig. S1 in ESI†).
Based on the above observations, photoreaction in the solid
state was expected not to be a problem for 1 and 2, considering
that PDK, which has a much lower Fr than 1 and 2, can be
photoconverted to pentacene both in the thin-film and single-
crystalline states.19,32 The solid-state conversion of precursors 1
This journal is ©The Royal Society of Chemistry 2015
J. Mater. Chem. C

View Article Online
Paper
Journal of Materials Chemistry C
Table 1 Reaction quantum yields for the photo-induced decarbonylation
of a-diketone-type photoprecursors in toluenea
Photoprecursor
1
2
ADK
0.37
TDK
PDK
Fr
0.21
0.25
B0.10b
0.027
a
Value for ADK was determined by the absolute method, and others
b
were determined by the relative method. Rough estimation because
of significant overlap of absorption peaks between the reactant
and product. Experimental details are described in the Experimental
section.
Fig. 6 UV-vis absorption spectra of thin films deposited by the three
different methods: (a) C6-ATT (b) C6-BADT.
absorption peak of which shifts by ca. 0.10 eV from 411 to
425 nm associated with considerable broadening of all the peaks.
This indicates perturbation of molecular electronic structure in
Fig. 5 Solid-state IR spectra before and after the photolysis: (a) a-diketone
1 to C6-ATT (b) a-diketone 2 to C6-BADT. The films were prepared by spin
coating of a 3 mg mLꢀ1 solution of 1 or 2 in chloroform. Irradiation the thin films through nonspecific intermolecular interactions
conditions: 470 nm, 200 mW cmꢀ2, 30 min.
involving multiple modes and degrees. The corresponding red-
shift is smaller for C6-ATT, being only 0.04 eV from 431 to 437 nm
with relatively minor broadening of the peaks. Consequently, the
and 2 to C6-ATT and C6-BADT, respectively, was monitored by
absorption onset of C6-BADT is essentially the same as that of
IR spectroscopy as shown in Fig. 5. The carbonyl stretches of
C6-ATT in the thin-film state. These observations indicate that
the a-diketone unit are observable around 1730 cmꢀ1 for thin-
C6-BADT has somewhat stronger intermolecular interactions
film samples, which disappear upon irradiation of 470 nm light
than C6-ATT in the solid state. At the same time, no effective,
specific intermolecular interaction that leads to the emergence of
a new peak is indicated in these spectra even for C6-BADT. This is
in contrast to the case of pentacene deposited by the photo-
at 200 mW cmꢀ2 for 30 min under a nitrogen atmosphere,
indicating completion of the photo-induced decarbonylation in
the films. Note that these conditions are the same as those
employed for the photoconversion of the a-diketone-type pre-
precursor approach, which shows long-wavelength absorption
cursor of linear anthradithiophene,18 while a longer reaction
and fluorescent peaks that can be attributed to the formation
time is needed for PDK.19
of p-stacks in the solid state.33
The highest occupied molecular orbital (HOMO) energies
of C6-ATT and C6-BADT in the thin-film state were estimated
2.3. Optoelectronic properties of thin films
UV–vis absorption characteristics are essentially the same for by photo-electron spectroscopy (PES) in air.34 Here again, the
the three differently prepared films in each case of C6-ATT and difference in the deposition process does not lead to major
C6-BADT. Specifically, the absorption spectra of C6-ATT films changes; namely, the observed HOMO energies are from ꢀ5.52
consist of five major peaks at around 330, 344, 380, 411, and to ꢀ5.47 eV for C6-ATT and from ꢀ5.43 to ꢀ5.39 eV for C6-BADT
437 nm associated with an absorption onset at 445 nm (2.79 eV) (Table 2 and Fig. S2 and S3 in ESI†). The HOMO levels were also
ox
regardless of the deposition process (Fig. 6a).
estimated from the onset of the first oxidation potential (E
)
onset
Similarly, C6-BADT films show absorption peaks at about in cyclic voltammograms (CVs) according to the known empirical
ox
ox
295, 390, and 425 nm associated with an absorption onset at equation HOMO = ꢀE
ꢀ4.8 eV, where E
is referenced to
onset
onset
around 445 nm with little variations among the three differ- the ferrocene/ferrocenium standard.35 The obtained values are
ently prepared samples (Fig. 6b).
ꢀ5.25 and ꢀ5.40 eV for C6-ATT and C6-BADT, respectively (Fig. S4
Both compounds show redshift of the photoabsorption in ESI†). Thus, C6-BADT has a lower HOMO level than C6-ATT in
peaks in the solid state as compared to in solution. The degree solution according to the CVs, which parallels the results of
of redshift is larger for C6-BADT, the longest-wavelength density functional theory (DFT) calculations (Fig. S5 in ESI†).
J. Mater. Chem. C
This journal is ©The Royal Society of Chemistry 2015

View Article Online
Journal of Materials Chemistry C
Paper
Table 2 Optical band gap energies (Eg,opt) and estimated frontier-orbital
energies of C6-ATT and C6-BADT in the thin-film state
Deposition method Eg,opt [eV] HOMOb [eV] LUMOc [eV]
a
C6-ATT
Vacuum deposition 2.79
Photoprecursor 2.79
Direct spin coating 2.79
ꢀ5.52
ꢀ5.48
ꢀ5.47
ꢀ2.73
ꢀ2.69
ꢀ2.68
C6-BADT Vacuum deposition 2.81
Photoprecursor 2.79
Direct spin coating 2.79
ꢀ5.40
ꢀ5.39
ꢀ5.43
ꢀ2.59
ꢀ2.60
ꢀ2.64
a
b
Calculated from the absorption onsets. Estimated by photo-electron
c
spectroscopy in air. Calculated as HOMO + Eg,opt
.
On the other hand, the PES measurements indicate that the
HOMO level is higher for C6-BADT in thin films. This discrepancy
is assumed to be caused by the difference in intermolecular
interactions in the condensed phase, as observed in the UV–vis
spectra of the thin films. The LUMO levels were also estimated
based on the equation LUMO = HOMO + Eg,opt, where Eg,opt stands
for the optical band gap obtained from the UV–vis absorption
spectra of thin films. The values are summarized in Table 2.
2.4. Structural characteristics of thin films
Thin films of C6-ATT and C6-BADT were prepared on ITO/glass
substrates with a MoO3 buffer layer, which were also used for
the space-charge-limited-current (SCLC) measurements described
in the next section. Vacuum deposition was performed at a rate of
1.8 nm minꢀ1 and a pressure of 1.2 ꢁ 10ꢀ4 Pa. Deposition by the
photoprecursor approach was conducted by spin coating of a
Fig. 7 AFM images (a–c) and out-of-plane XRD patterns (d–f) of thin films
of C6-ATT. Thin films are prepared by vacuum deposition (a and d), the
photoprecursor approach (b and e), or direct spin coating (c and f).
10 mg mLꢀ1 solution of photoprecursor 1 or 2 in chloroform at Fig. S6a in ESI†), indicating an edge-on mode arrangement of
800 rpm, followed by irradiation of 470 nm light at 200 mW cmꢀ2 molecules towards the substrate. The thin-film deposited via the
for 30 min in a nitrogen atomosphere. Direct solution deposition photoprecursor approach has a smoother surface having an RMS
was first tried by spin coating of a 10 mg mLꢀ1 solution of C6-ATT value of 0.69 nm, and the difference between the highest and
or C6-BADT in chloroform at 800 rpm as in the case of the lowest points is only about 5 nm in the observed area (Fig. 7b).
photoprecursor approach; however, highly inhomogeneous films The XRD pattern is similar to that of the vacuum deposited film,
with visually observable cracks were obtained in these cases. The but with weaker peak intensities so that the second peak cannot
unfavourable morphology can be ascribed to the high tendency of be distinguish from the noise (Fig. 7e).
C6-ATT and C6-BADT to aggregate from the chloroform solutions,
The surface morphology of the film directly spin-coated
and in turn, it is a clear manifestation of the power of the photo- from a chloroform–chlorobenzene (1 : 1) solution is signifi-
precursor approach to afford smoother films based on such cantly different from the two described above—the film con-
materials. The quality of directly spin-coated films was improved tains large grains, whose lateral dimensions reach a few mm or
when a 1 : 1 mixture of chloroform and chlorobenzene was used as even larger (Fig. 7c). The height difference within the observed
solvent, so that the resulting films look homogeneous to the eye. area is over 50 nm, which is similar to the average thickness
The microscopic surface morphology of the resulting films of the film (ca. 50 nm). This high roughness prevents the film
was investigated by tapping-mode atomic-force microscopy (AFM), from being employed in the SCLC measurement, since the
and the crystallinity of the films was examined by out-of-plane effective film thickness and interface area are hard to define.
X-ray diffraction (XRD) analysis. Fig. 7 shows the results for three It would also be unfavourable to employ such a highly inhomo-
differently prepared samples of C6-ATT. The vacuum-deposited geneous film in electronic devices. The X-ray diffraction from
film has a relatively smooth surface with a root-mean square the directly spin-coated film is significantly stronger compared
(RMS) roughness of 1.94 nm (Fig. 7a). It also has a moderate to the films prepared by vacuum deposition and the photo-
crystallinity showing a sharp diffraction peak at 2y = 4.621 (d = precursor approach (Fig. 7d–f). This indicates that large crystal-
19.1 Å) associated with a weak diffraction at 2y = 9.251 (d = 9.56 Å) line domains are formed in the direct spin-coating conditions,
(Fig. 7d). Although these peaks do not match with the diffraction which correlates well with the observation of significantly large
pattern simulated from the single-crystal X-ray diffraction data domains by AFM.
(Table S3 in ESI†), the d-spacing calculated for the first diffraction
Fig. 8 summarizes the results for C6-BADT. The vacuum-
is close to the lateral molecular dimension of C6-ATT (B21 Å, deposited film of C6-BADT has larger grains compared to the
This journal is ©The Royal Society of Chemistry 2015
J. Mater. Chem. C

View Article Online
Paper
Journal of Materials Chemistry C
corresponding C6-ATT film, indicating a higher tendency of investigation has revealed that the film deposited by the photo-
the former compound to aggregate than the latter. This is precursor approach is similar to the directly spin-coated films
consistent with the results of UV–vis and ionization potential in terms of surface roughness in the case of C6-BADT. The RMS
measurements described earlier in Section 2.3. On the other value for the directly spin-coated film is 16.4 nm, and the
hand, the surface roughness is similar for the two cases; difference between the highest and lowest points is over
specifically, the RMS values are 1.94 and 1.66 nm for the 100 nm (Fig. 8c). On the other hand, the grain size is much
vacuum deposited C6-ATT and C6-BADT, respectively (Fig. 7a larger for direct spin coating, probably because of the use of a
and 8a). The XRD measurement gave four distinct peaks at higher boiling-point solvent (chlorobenzene) and thus slower
2y = 4.91, 9.90, 14.7, 19.81 corresponding to the d-spacings of growth of crystallites. In accordance with this, the intensity of
18.0, 8.93, 6.03, and 4.48 Å, respectively (Fig. 8d). The first peak X-ray diffraction is much stronger for the directly spin-coated
is not very far from the (001) diffraction of the previously film (Fig. 8f).
reported single-crystal structure (c = 22.2 Å), indicating an
These results demonstrate the potential and limitations of
end-on-mode arrangement of molecules against the sub- the photoprecursor approach. It can be stated that the solution
strate.27 The deviation of about 4 Å may be explained by a tilt processability of highly crystalline materials is generally
of molecules in relation to the c* axis in the thin-film state improved by employing the photoprecursor approach. This
compared to the a (901) or b (911) of the single-crystal unit cell, was manifested by the difference in the quality of C6-ATT and
or difference in the orientation of alkyl chains.
C6-BADT films when chloroform was employed as a solo
The XRD pattern of the C6-BADT film deposited by the photo- solvent—direct spin coating led to cracked films whereas
precursor approach is similar to that of the vacuum-deposited the films were visually homogeneous when deposited by the
film, having a strong diffraction at 2y = 4.891 associated with photoprecursor approach. Indeed, the film quality in terms of
weak diffractions at 9.82, 14.7, and 19.91, although the third peak microscopic surface morphology and crystallinity is as high as
is hard to distinguish from the noise (Fig. 8e). On the other hand, the vacuum-deposited film in the case of C6-ATT. For C6-BADT,
the AFM images show different features between those two films; on the other hand, the quality of the thin film prepared by
namely, the film prepared from the photoprecursor 2 has larger, the photoprecursor approach is considerably lower than the
but structurally less defined domains (Fig. 8b). The RMS value is vacuum-deposited sample. The intrinsic aggregation property
as large as 19.2 nm, with the difference between the highest and of C6-BADT seems to have a more profound effect than that of
lowest points exceeding 100 nm within the scanned area. Further C6-ATT, and is hard to overcome just by employing the photo-
precursor approach at least under the examined conditions.
2.5. Space-charge-limited-current measurements
The charge-carrier mobility in C6-ATT deposited by the photo-
precursor approach was measured by the SCLC technique.
While the field-effect mobility probes the property of materials
only at the interface with gate dielectrics, the SCLC method
evaluates the bulk mobility, and thus would be more suitable
when comparing bulk qualities of thin films.
Hole-only devices with the structure of [ITO/MoO3/C6-ATT/
MoO3/Al] were fabricated for the measurements. As shown in
Fig. S7b of ESI,† a carrier mobility of 2.07 ꢁ 10ꢀ5 cm2 Vꢀ1 sꢀ1 was
obtained when the active layer was deposited by the photo-
precursor approach. This value is in the same range as that
obtained for a vacuum deposited film (1.79 ꢁ 10ꢀ5 cm2 Vꢀ1 sꢀ1
,
Fig. S7a in ESI†), which is reasonable considering that these two
deposition methods afford thin films having similar crystallinity
and morphology as described in Section 2.4. The directly spin-
coated film of C6-ATT is too rough to obtain a reliable mobility by
the SCLC method, because its effective thickness is hard to
define. For the same reason, SCLC mobilities were not measured
for the solution-processed films of C6-BADT.
3. Conclusions
We have synthesized new a-diketone-type photoprecursors 1
and 2, and evaluated the thin films of C6-ATT and C6-BADT
deposited by the photoprecursor approach in comparison with
Fig. 8 AFM images (a–c) and out-of-plane XRD patterns (d–f) of thin films
of C6-BADT. The samples are prepared by vacuum deposition (a and d),
the photoprecursor approach (b and e), or direct spin coating (c and f).
J. Mater. Chem. C
This journal is ©The Royal Society of Chemistry 2015

View Article Online
Journal of Materials Chemistry C
Paper
those prepared by vacuum deposition and direct spin coating. and 13C{1H} NMR spectra were recorded on a JEOL JNM-AL300 or
The syntheses were completed in good overall yields from the a JNM-ECP400 spectrometer using tetramethylsilane or solvent
known starting materials, and the target photoprecursors were resonances as internal standards. Mass spectra were recorded on
unambiguously identified by spectroscopic analyses and single- a JEOL AccuTOF/JMS-T100LC (ESI), JEOL JMS-700 (EI), or a
crystal X-ray diffractometry. The comparative evaluation revealed Bruker Autoflex II (MALDI-TOF) instrument. Infrared spectra
that, in the case of C6-ATT, the photoprecursor approach provided were recorded on a JASCO FT/IR-4200 spectrometer with an
thin films that have considerably higher quality than the directly ATR PRO450-S attachment. Melting points were measured on a
spin-coated samples, and are comparable to the vacuum depos- Stanford Research System OptiMelt MPA100 apparatus.
ited counterparts in terms of surface morphology and SCLC
1,2,4,5-Tetrakis(5-hexylthiophen-2-yl)benzene (3). To
a
mobility. In contrast, the photoprecursor approach had only solution of 2-hexylthiophene (1.3 mL, 7.2 mmol, 6.4 equiv.) in
limited success under the examined processing conditions in anhydrous THF (100 mL) under argon was added a solution of
the case of C6-BADT, which seemed to have a stronger tendency n-butyllithium (1.6 M in hexanes, 5.0 mL, 8.0 mmol, 7.1 equiv.)
to self-aggregate in comparison with C6-ATT according to the at ꢀ78 1C. The mixture was stirred at the same temperature for
spectroscopic and microscopic analyses.
1 h, and a solution of ZnCl2 (1.3 g, 9.5 mmol, 8.5 equiv.) in THF
These results show that the photoprecursor approach is more (50 mL) was added to it. The mixture was stirred for 1 h at room
than just a way to enable the use of solution-based techniques temperature, to which a solution of 1,4-dibromo-2,5-diiodo-
for deposition of insoluble (or hardly soluble) materials. Rather, benzene (0.55 g, 1.1 mmol) and Pd(PPh3)4 (29 mg, 25 mmol,
this approach also extends the range of applicable solution- 2.2 mol%) in THF (30 mL) was added. The mixture was further
processing conditions for soluble semiconducting materials. As stirred at 80 1C for 12 hours, before cooling to room tempera-
observed for C6-ATT, the photoprecursor approach can afford ture and the solvent was removed under reduced pressure. The
thin films of much higher quality than direct solution deposition mixture was then partitioned between CH2Cl2 and sat. NH4Cl.
without an extensive search for processing conditions. Even in The organic layer was collected and the aqueous layer was
the more difficult case of C6-BADT, the film quality is higher for extracted with CH2Cl2. The combined organic layers were dried
the photoprecursor method than direct spin coating when the over anhydrous Na2SO4, filtered, and evaporated. The resulting
same solvent (chloroform) was employed. It would be worth crude product was purified by flash silica gel column chroma-
noting here that the a-diketone-type photoprecursors 1 and 2 tography (hexanes) to give a pale yellow solid (0.60 g, 0.81 mmol,
are well soluble in common organic solvents and not strongly 72%). The product was identified by comparing its 1H NMR
aggregating, so that their thin films are highly homogeneous spectrum with the literature data.25 1H NMR (300 MHz, CDCl3):
without forming large grains in a wide variety of spin-coating d 7.57 (s, 2H), 6.76 (d, J = 3.5 Hz, 4H), 6.64 (d, J = 3.7 Hz, 4H),
conditions (different solvent systems, solution concentrations, 2.76 (t, J = 7.5 Hz, 8H), 1.66–1.61 (m, 8H), 1.38–1.26 (m, 24H),
spin rates, etc.) (Fig. S8 in ESI†). This considerably simplifies the 0.88 (distorted t, J = 7.0 Hz, 12H).
optimization of solution-deposition conditions, and is highly
2,5,9,12-Tetrahexylanthra[1,2-b:4,3-b0:5,6-b00:8,7-b0 0 0]tetra-
advantageous especially when the choice of solvent is limited—a thiophene (C6-ATT). A solution of iron(III) chloride (0.60 g,
typical example is the solution processing of multi-layer thin 3.7 mmol, 8.1 equiv.) in nitromethane (6 mL) was added
films by employing orthogonal solvent systems. In addition, the dropwise to a solution of 3 (0.34 g, 0.46 mmol) in anhydrous
photoprecursor approach allows us to control the film morpho- CH2Cl2 (900 mL) at 0 1C. After 15 min, methanol (150 mL)
logy through adjusting the photoreaction conditions including was added and the mixture was stirred for 1 h. The product
the light intensity and duration, or by the use of small amounts was collected by filtration and rinsed with methanol. The
of solvent additives.19 Further investigation along these lines is crude product was purified by flash silica gel column chromato-
underway in our group.
graphy (hexanes/CHCl3, 3 :1) to give a yellow solid (0.29 g,
0.39 mmol, 86%). The product was identified by comparing
its 1H NMR spectrum with the literature data.25 1H NMR
(300 MHz, CDCl3): d 8.67 (s, 2H), 7.35 (s, 4H), 3.04 (distorted t,
J = 7.5 Hz, 8H), 1.88–1.82 (m, 8H), 1.54–1.34 (m, 24H), 0.92
(t, J = 7.0 Hz, 12H).
4. Experimental
4.1. Synthesis and structural characterization of compounds
General. Anhydrous THF (super plus grade, stabilizer free),
2,5,9,12-Tetrahexyl-7,14,15,19-tetrahydro-7,14-[4,5]epidioxolo-
toluene (super plus grade), and CH2Cl2 (super grade) were anthra[1,2-b:4,3-b0:5,6-b00:8,7-b0 00]tetrathio-phen-17-one (4). C6-ATT
obtained from Kanto Chemical and used as received. Anhydrous (1.0 g, 1.4 mmol) and vinylene carbonate (8.0 mL, 13 mmol,
DMSO (super dehydrated grade) was obtained from Wako 93 equiv.) in anhydrous toluene (50 mL) were placed in an
Chemical, and used as received. N,N-Diisopropylethylamine was autoclave and stirred at 180 1C for 3 days. The reaction mixture
distilled and stored over KOH. All the other solvents and chemicals was cooled to room temperature and the solvent was evapo-
were of the reagent grade and obtained commercially and used rated, to which methanol was added. The product was isolated
without further purification. Room temperature, or rt, means 20– by filtration and further purified by flash silica gel column
25 1C. Thin layer chromatography and flash column chromato- chromatography (hexanes; hexanes/CHCl3, 2 : 1) to give a pale
graphy were performed on Merck Silica Gel 60 F254 (Art. 5554) and brown solid (0.86 g, 1.0 mmol, 76%). 1H NMR (300 MHz,
1
Kanto Silica Gel 60N (spherical, 40–50 mm), respectively. H NMR CDCl3): d 7.29 (s, 2H), 7.28 (s, 2H), 5.31–5.30 (m, 2H), 5.11–5.10
This journal is ©The Royal Society of Chemistry 2015
J. Mater. Chem. C

View Article Online
Paper
Journal of Materials Chemistry C
(m, 2H), 2.95 (m, J = 7.0 Hz, 8H), 1.78 (m, 8H), 1.43–1.25 (m, 24H), were washed with brine, dried over Na2SO4, filtered, and
0.90 (distorted t, J = 7.0 Hz, 12H); 13C {1H} NMR (75 MHz, CDCl3): evaporated. The resulting crude product was purified by flash
d 147.4, 146.8, 134.3, 134.2, 134.2, 133.3, 131.7, 127.4, 126.1, 119.7, column chromatography on silica gel containing 10 wt.% of
119.6, 76.5, 45.2, 31.6, 31.6, 31.4, 31.3, 31.1, 28.9, 28.8, 22.6, 14.1; anhydrous K2CO3 (CH2Cl2/hexanes, 1 : 2) to give 2,5-dibromo-
HRMS (ESI): m/z calcd for C49H60O3S4 ([M + Na]+) 847.3323, found 1,4-bis(5-hexyl-2-thienyl)benzene as a colorless oil (5.7 g,
847.3323; m.p.: 163 1C.
10 mmol, 81%). 1H NMR (300 MHz, CDCl3): d 7.76 (s, 2H),
2,5,9,12-Tetrahexyl-7,14-dihydro-7,14-ethanoanthra[1,2-b:4,3- 7.18 (d, J = 3.7 Hz, 2H), 6.78 (d, J = 3.5 Hz, 2H), 2.84 (t, J = 7.5 Hz,
b0:5,6-b00:8,7-b0 0 0]tetrathiophene-15,16-dione (1). To a solution of 4H), 1.72 (quint, J = 7.5 Hz, 4H), 1.50–1.24 (m, 12H), 0.88
4 M NaOH (10 mL) were added THF (10 mL) and the vinylene (distorted t, J = 6.8 Hz, 6H); 13C {1H} NMR (100 MHz, CDCl3):
carbonate adduct 4 (0.66 g, 0.80 mmol) under a N2 atmosphere. d 148.0, 137.1, 135.8, 135.6, 128.2, 124.3, 120.9, 31.7, 30.2, 29.0,
The mixture was stirred at reflux for 2 h. After cooling to room 22.7, 14.2; HRMS (EI): m/z calcd for C26H32Br2S2 (M+) 566.0312,
temperature, the mixture was acidified with 1 M HCl. After found 566.0305.
addition of water, the mixture was extracted with chloroform,
To a solution of 2,5-dibromo-1,4-bis(5-hexyl-2-thienyl)-benzene
and the combined organic layer was washed with water and (2.1 g, 3.7 mmol), PdCl2(PPh3)2 (0.35 g, 0.50 mmol, 13 mol%) and
brine, dried over Na2SO4, filtered, and evaporated. The resulting CuI (0.20 g, 1.1 mmol, 28 mol%) in dry THF (70 mL) under an
crude product was purified by flash silica gel column chromato- argon atmosphere were added N,N-diisopropylamine (7.5 mL,
graphy (hexanes/EtOAc, 3 :1) to give the corresponding diol as a 54 mmol, 14 equiv.) and trimethylsilylacetylene (2 mL, 14 mmol,
1
pale yellow solid (0.63 g, 0.79 mmol, 98%). H NMR (300 MHz, 3.8 equiv.) at room temperature. The resultant mixture was stirred
CDCl3): d 7.29 (s, 2H), 7.28 (s, 2H), 5.08 (m, 2H), 4.31 (m, 2H), at reflux for 2 days, and then allowed to cool to room temperature
2.98–2.93 (t, J = 7.5 Hz, 8H), 2.29 (m, 2H), 1.81–1.77 (m, 8H), 1.42– before being treated with sat. NH4Cl. The resulting mixture was
1.31 (m, 24H), 0.90 (distorted t, J = 7.2 Hz, 12H); 13C{1H} NMR extracted with dichloromethane. The combined organic phases
(75 MHz, CDCl3): d 146.5, 146.4, 135.1, 133.4, 133.0, 129.3, 128.3, were washed with brine, dried over Na2SO4, filtered, and evapo-
119.5, 119.5, 68.7, 48.9, 31.6, 31.4, 31.4, 31.1, 31.1, 28.9, 28.8, 22.6, rated. The crude product was purified by column chromatography
14.1; HRMS (ESI): m/z calcd for C48H62O2S4 ([M + Na]+) 821.3503, on neutral activated alumina (hexanes) to give the target com-
found 821.3539; m.p.: 146 1C.
pound as a brown solid (1.7 g, 2.8 mmol, 76%). 1H NMR
Under an argon atmosphere, TFAA (2.8 mL, 33 mmol, 24 (300 MHz, CDCl3): d 7.68 (s, 2H), 7.54 (d, J = 3.7 Hz, 2H), 6.75
equiv.) was added to a stirred solution of anhydrous DMSO (d, J = 3.5, 2H), 2.83 (t, J = 7.1 Hz, 4H), 1.71 (quint, J = 7.1 Hz,
(2.8 mL, 39 mmol, 48 equiv.) in anhydrous CH2Cl2 (8.0 mL) at 4H), 1.42–1.27 (m, 12H), 0.90 (distorted t, J = 6.9 Hz, 6H), 0.26
ꢀ60 1C. After stirring at ꢀ60 1C for 30 min, the diol obtained as (s, 18H); 13C {1H} NMR (75 MHz, CDCl3): d 147.2, 138.1, 134.4,
above (0.65 g, 0.81 mmol) in CH2Cl2 (4.0 mL) was added. The 134.0, 127.1, 124.5, 120.1, 104.7, 101.3, 31.7, 31.6, 30.3, 28.9,
reaction was stirred at ꢀ60 1C for 1.5 h, and then triethylamine 22.6, 14.1, ꢀ0.4; HRMS (ESI): calcd for C36H50Si2S2 ([M + Na]+)
(8 mL, 57 mmol, 71 equiv.) was added to the reaction. After 625.2790, found 625.2768.
stirring for 1 h, the reaction was allowed to warm up to room
2,8-Dihexylanthra[1,2-b:5,6-b0]dithiophene (C6-BADT). To a
temperature and diluted with CH2Cl2. The organic phase was solution of 5 (1.8 g, 2.9 mmol) in dry THF (40 mL) was added
isolated and sequentially washed with 1 M HCl, water and dropwise a solution of tetrabutylammonium fluoride (1.0 M,
brine, and then dried over Na2SO4, filtrated, and evaporated. 10 mL, 10 mmol, 3.4 equiv.) at 0 1C, and the resulting solution
The residue was purified by flash silica gel column chromato- was stirred at room temperature for 1 h. The solvent was then
graphy (hexanes/CH2Cl2, 1 : 1). The isolated product was further evaporated, and the crude product was purified by column
purified by recrystallization from hot hexanes to give an orange chromatography on silica gel (hexanes) to give the desilylated
solid (0.18 g, 0.23 mmol, 28%). 1H NMR (300 MHz, CDCl3): intermediate as a yellow solid (0.98 g, 2.1 mmol, 72%). 1H NMR
d 7.34 (s, 4H), 5.50 (s, 2H), 2.99 (t, J = 7.6 Hz, 8H), 1.80 (quint, (300 MHz, CDCl3): d 7.72 (s, 2H), 7.52 (d, J = 3.7 Hz, 2H), 6.77
J = 7.5 Hz, 8H), 1.42–1.32 (m, 24H), 0.90 (distorted t, J = 7.1 Hz, (d, J = 3.7 Hz, 2H), 3.37 (s, 2H), 2.84 (t, J = 7.1 Hz, 4H), 1.76–1.66
12H); 13C {1H} NMR (75 MHz, CDCl3): d 180.8, 148.4, 135.8, (m, 4H), 1.41–1.28 (m, 12H), 0.90 (distorted t, J = 7.1 Hz, 6H);
133.4, 124.5, 119.8, 56.3, 31.6, 31.4, 31.1, 28.8, 22.6, 14.1; HRMS 13C {1H} NMR (75 MHz, CDCl3): d 147.5, 137.7, 134.8, 127.3,
(ESI): m/z calcd for C48H58O2S4 ([M + Na]+) 817.3217, found 124.8, 119.7, 83.4, 83.0, 31.6, 30.2, 28.9, 22.6, 14.1; HRMS
817.3221; FT-IR (ATR): nꢀmax (cmꢀ1) 2955, 2927, 2855, 1730, (MALDI-TOF): calcd for
C
30H34S2 (M+) 458.2090, found
1538, 1464, 1380, 826.
458.2096; m.p.: 66 1C.
InCl3ꢂ4H2O (41 mg, 0.14 mmol, 13 mol%) was dehydrated by
2,5-Bis(5-hexylthiophen-2-yl)-1,4-bis(trimethylsilyl-ethynyl)-
benzene (5). A solution of 2-tributylstannyl-5-hexylthiophene heating in a Schlenk flask under vacuum, to which the inter-
(15 g, 33 mmol, 2.7 equiv.), 1,4-dibromo-2,5-diiodobenzene mediate obtained as above (0.50 g, 1.1 mmol) and xylenes
(6.0 g, 12 mmol), and Pd(PPh3)4 (1.4 g, 1.2 mmol, 10 mol%) in (30 mL) were added under an argon atmosphere. The resultant
dry DMF (30 mL) under an argon atmosphere was stirred at mixture was stirred at 120 1C for 12 h and then allowed to cool
80 1C for 3 days, then allowed to cool to room temperature. to room temperature before being filtrated through Celite. The
The mixture was partitioned between diethyl ether and water. filtrate was evaporated and the crude product was purified by
The organic phase was isolated and the aqueous phase was flash silica gel column chromatography (CH2Cl2/hexanes, 1 : 4)
extracted with diethyl ether. The combined organic phases to give C6-BADT as a yellow solid (0.45 g, 0.98 mmol, 90%).
J. Mater. Chem. C
This journal is ©The Royal Society of Chemistry 2015

View Article Online
Journal of Materials Chemistry C
Paper
The product was identified by comparing its 1H NMR spectrum sequentially with 1 M HCl, H2O and brine, dried over Na2SO4,
with the literature data.28 1H NMR (300 MHz, CDCl3): d 8.59 filtered, and evaporated. The crude product was purified by flash
(s, 2H), 7.85 (d, J = 7.9 Hz, 2H), 7.70 (d, J = 7.9 Hz, 2H), 7.16 silica gel column chromatography (CH2Cl2). The isolated material
(s, 2H), 3.01 (t, J = 7.5 Hz, 4H), 1.82 (quint, J = 7.3 Hz, 4H), 1.46– was further purified by recrystallization from toluene and hexanes
1.27 (m, 12H), 0.91 (distorted t, J = 7.0 Hz, 6H); m.p.: 111 1C to give the pure target compound as an orange solid (0.23 g,
(lit. 117–119).
0.44 mmol, 57%). 1H NMR (300 MHz, CDCl3): d 7.67 (d, J = 8.1 Hz,
2,8-Dihexyl-6,12,13,17-tetrahydro-6,12-[4,5]epidioxolo-anthra- 2H), 7.47 (d, J = 8.1 Hz, 2H), 7.06 (s, 2H), 5.27 (s, 2H), 2.93 (t, J =
[1,2-b:5,6-b0]dithiophen-15-one (6). C6-BADT (0.45 g, 0.98 mmol) 7.5 Hz, 4H), 1.76 (quint, J = 7.3 Hz, 4H), 1.42–1.28 (m, 12H), 0.89
and vinylene carbonate (2.5 mL, 39 mmol, 40 equiv.) in anhy- (distorted t, J = 7.0 Hz, 6H); 13C {1H} NMR (75 MHz, CDCl3): d 182.3,
drous toluene (30 mL) were added to an autoclave and stirred at 148.4, 141.9, 137.0, 129.4, 129.2, 123.8, 122.6, 121.5, 58.1, 31.6,
180 1C for 3 days. The reaction mixture was cooled to room 31.1, 31.0, 28.8, 22.6, 14.1; HRMS (ESI): m/z calcd for C32H34O2S2
temperature and the solvent was evaporated to afford an oil, ([M + Na]+) 537.1898, found 537.1911; IR (ATR): nꢀmax (cmꢀ1) 2953,
which solidified by the addition of methanol (5 mL). After 2926, 2855, 1748, 1732, 1466, 1444, 1407, 1114, 1048, 837, 725.
filtration, the crude product was purified by flash silica gel
Single-crystal X-ray diffraction measurements. The crystallo-
column chromatography (hexanes) to give the vinylene carbo- graphic data of compound C6-ATT were recorded on a Bruker-
nate adduct as a pale yellow solid (0.41 g, 0.74 mmol, 76%). APEXII X-ray diffractometer equipped with a large area CCD
1H NMR (300 MHz, CDCl3): d 7.53 (d, J = 7.9 Hz, 1H), 7.52 (d, J = detector by using graphite monochromated Mo Ka radiation
7.9 Hz, 1H), 7.40 (d, J = 7.9 Hz, 2H), 7.01 (s, 1H), 6.99 (s, 1H), (l = 0.71073 Å). The crystallographic data of compounds 1 and 2
5.03–4.98 (m, 4H), 2.93–2.86 (m, 4H), 1.75 (quint, J = 7.2 Hz, 4H), were recorded on a Rigaku R-AXIS RAPID/S imaging plate
1.47–1.26 (m, 12H), 0.89 (distorted t, J = 6.4 Hz, 6H); 13C {1H} diffractometer with graphite monochromated Mo Ka radiation.
NMR (75 MHz, CDCl3): d 154.3, 147.5, 147.0, 140.6, 137.5, 136.5, The structures were refined with the SHELXL-97. Crystallo-
132.3, 132.2, 131.1, 130.7, 123.2, 122.2, 121.8, 121.7, 121.4, 121.3, graphic data for C6-ATT: C46H58S4, Mw = 739.16, monoclinic,
76.2, 46.5, 46.4, 31.6, 31.1, 31.1, 31.0, 28.9, 28.8, 22.6, 14.1; space group P21/n (No. 14), a = 12.604(2) Å, b = 4.8761(9) Å,
HRMS (ESI): m/z calcd for C33H36O3S2 ([M + Na]+) 567.2004, c = 32.657(6) Å, b = 100.676(4)1, V = 1972.3(6) Å3, rcalcd
=
found 567.2004; m.p.: 83 1C.
1.245 g cmꢀ3, T = 90(2) K, Z = 2, R1 = 0.0574 [I 4 2.0s(I)],
2,8-Dihexyl-6,12-dihydro-6,12-ethanoanthra[1,2-b:5,6-b0]dithio- Rw = 0.1098 (all data), GOF = 1.013. Crystallographic data for 1:
%
phene-13,14-dione (2). To a solution of 4 M NaOH (15 mL) were
C
48H58O2S4ꢂ0.5(C6H14), Mw = 838.27, triclinic, space group P1
added THF (30 mL) and compound 6 (0.32 g, 0.59 mmol) under a (No. 2), a = 10.9591(3) Å, b = 11.8130(3) Å, c = 19.6453(5) Å,
nitrogen atmosphere, and the mixture was stirred at reflux for 2 h. a = 104.6590(10)1, b = 96.3200(10)1, g = 103.8200(10)1, V =
After cooling to room temperature, the mixture was neutralized 2349.57(11) Å3, rcalcd = 1.185 g cmꢀ3, T = 193(2) K, Z = 2, R1 =
with 1 M HCl. After the addition of water, the mixture was 0.0605 [I 4 2.0s(I)], Rw = 0.1930 (all data), GOF = 1.102.
extracted with chloroform, and the combined organic layers Crystallographic data for 2: C32H34O2S2, Mw = 514.74, monoclinic,
were washed with brine, dried over Na2SO4, filtered, and space group P2/c (No. 13), a = 12.4700(5) Å, b = 13.7514(5) Å,
evaporated. The crude product was purified by flash silica gel c = 16.2964(6) Å, b = 105.0607(8)1, V = 2698.51(16) Å3, rcalcd
column chromatography (hexanes/EtOAc, 4 : 1) to give the 1.267 g cmꢀ3, T = 103(2) K, Z = 4, R1 = 0.0589 [I 4 2.0s(I)], Rw
=
=
corresponding diol as a pale yellow solid (0.21 g, 0.41 mmol, 0.1559 (all data), GOF = 1.050. CCDC 1052010 (C6-ATT), 1052256
69%). 1H NMR (300 MHz, CDCl3): d 7.50 (d, J = 7.7 Hz, 1H), 7.45 (1) and 1052257 (2).
(d, J = 7.9 Hz, 1H), 7.42 (d, J = 8.1 Hz, 1H), 7.36 (d, J = 7.9 Hz,
1H), 7.00 (s, 1H), 6.98 (s, 1H), 4.79 (d, J = 2.6 Hz, 1H), 4.74 (d, J =
2.6 Hz, 1H) 4.23–4.13 (m, 2H), 2.89 (t, J = 7.6 Hz, 4H), 2.16 (two
overlapping d, 2H), 1.74 (quint, J = 7.7 Hz, 4H), 1.41–1.25
(m, 12H), 0.89 (distorted t, J = 7.0 Hz, 6H); 13C {1H} NMR
(75 MHz, CDCl3): d 146.6, 140.0, 138.0, 123.4, 121.6, 121.3,
121.2, 120.8, 120.7, 69.0, 68.0, 50.2, 49.9, 31.6, 31.2, 31.1, 31.0,
31.0, 28.9, 28.8, 22.6, 14.0; HRMS (ESI): m/z calcd for
4.2. Photoreaction quantum yield determination
A chemical actinometer (K3[Fe(C2O4)3])36 was used for quantum
yield determination of the photochemical reaction of ADK in
toluene. A square quartz cuvette (10 mm I.D.) that contained a
deaerated solution (3.0 mL) of ADK was irradiated with 468 nm
light through a monochromator (Ritsu MC-10N) using a 500 W
xenon lamp (Ushio XB-50102AA-A). The photochemical reaction
was monitored by using a JASCO UV/vis/NIR V-570 spectro-
photometer. The quantum yields were determined from the
increase in the absorbance of anthracene at 378 nm in the
beginning of the reaction. The Fr values of other compounds
(TDK, PDK, 1, and 2) were measured by the relative method
using the value for ADK (Fr = 0.37) as the standard.
C
32H38O2S2 ([M + Na]+) 541.2211, found 541.2220; m.p.: 136 1C.
TFAA (2.0 mL, 14 mmol, 18 equiv.) was added to a stirred
solution of DMSO (2.0 mL, 28 mmol, 36 equiv.) in anhydrous
CH2Cl2 (7.3 mL) at ꢀ60 1C under an argon atmosphere. After
stirring at the same temperature for 30 min, the diol obtained
as described above (400 mg, 0.77 mmol) in CH2Cl2 (4.5 mL) was
added to the reaction. The reaction was further stirred at
ꢀ60 1C for 1.5 h, and then triethylamine (5.5 mL, 39 mmol,
51 equiv.) was added to the reaction. After stirring for 1 h,
4.3. Film preparation and characterization
the reaction was allowed to warm up to room temperature, Three methods were employed to prepare the films of C6-ATT
before being diluted with CH2Cl2. The organic phase was washed and C6-BADT: (a) vacuum deposition of C6-ATT or C6-BADT with
This journal is ©The Royal Society of Chemistry 2015
J. Mater. Chem. C

View Article Online
Paper
Journal of Materials Chemistry C
a deposition rate of 1.8 nm minꢀ1; (b) the photopre-
cursor approach in which a solution of 1 or 2 in chloroform
(10 mg mLꢀ1) was spin-coated at 800 rpm for 30 s and the
resulting film was irradiated with a blue LED (l E 470 nm) at
200 mW cmꢀ2 for 30 min in a nitrogen-filled glovebox;
(c) direct spin coating of a solution of C6-ATT or C6-BADT in
chlorobenzene/chloroform (1 : 1 by volume, 10 mg mLꢀ1) at
800 rpm for 30 s. The thicknesses of the resulting films were
measured by Bruker DektakXT. UV-vis absorption spectra of
thin films on glass substrates were recorded in air using a
JASCO V-650 spectrometer. Ionization energies were measured
on a Bunko Keiki AC-3 photoelectron spectroscopy instru-
ment. The surface morphology of organic films was observed
using an SII SPA400/SPI3800N atomic force microscope (AFM)
in tapping mode with a silicon probe at a resonant frequency
of 138 kHz and a force constant of 16 N mꢀ1 (SII, SI-DF20). The
bulk structure of thin-films was evaluated by out-of-plane
X-ray diffraction (XRD) measurements using a Rigaku Smart-
Lab diffractometer equipped with a rotating anode (Cu Ka
radiation, l = 1.5418 Å). Hole mobility characteristics were
measured by the space-charge-limited-current (SCLC) method
using an Agilent Technologies 4155C Semiconductor Para-
meter Analyzer operated by Interactive Characterization Soft-
ware (ICS), version 3.6.0.
6 X.-H. Zhu, J. Peng, Y. Cao and J. Roncali, Chem. Soc. Rev.,
2011, 40, 3509–3524.
7 C. Zhong, C. Duan, F. Huang, H. Wu and Y. Cao, Chem.
Mater., 2011, 23, 326–340.
8 A. Facchetti, Chem. Mater., 2011, 23, 733–758.
9 M. Watanabe, K.-Y. Chen, Y. J. Chang and T. J. Chow,
Acc. Chem. Res., 2013, 46, 1606–1615.
10 S.-D. Jeong, B. Min, S. Y. Cho, C. Lee, B. K. Park, K.-S. An and
J. Lim, J. Org. Chem., 2012, 77, 8329–8331.
11 T. Okujima, Y. Hashimoto, G. Jin, H. Yamada, H. Uno and
N. Ono, Tetrahedron, 2008, 64, 2405–2411.
12 S. Aramaki, Y. Sakai and N. Ono, Appl. Phys. Lett., 2004, 84,
2085–2087.
13 H. Uno, T. Ishikawa, T. Hoshi and N. Ono, Tetrahedron Lett.,
2003, 44, 5163–5165.
14 S. Ito, T. Murashima, N. Ono and H. Uno, Chem. Commun.,
1998, 1661–1662.
15 One of the few exceptions is the use of benzoporphyrin in
solution-processed photovoltaic active layers by employing
its soluble precursor. Y. Matsuo, Y. Sato, T. Niinomi, I. Soga,
H. Tanaka and E. Nakamura, J. Am. Chem. Soc., 2009, 131,
16048–16050.
16 M. Suzuki, T. Aotake, Y. Yamaguchi, N. Noguchi, H. Nakano,
K. Nakayama and H. Yamada, J. Photochem. Photobiol., C,
2014, 18, 50–70.
17 Y. Yamaguchi, M. Suzuki, T. Motoyama, S. Sugii, C. Katagiri,
K. Takahira, S. Ikeda, H. Yamada and K. Nakayama, Sci.
Rep., 2014, 4, 7151.
18 H. Yamada, Y. Yamaguchi, R. Katoh, T. Motoyama,
T. Aotake, D. Kuzuhara, M. Suzuki, T. Okujima, H. Uno,
N. Aratani and K. Nakayama, Chem. Commun., 2013, 49,
11638–11640.
19 K. Nakayama, C. Ohashi, Y. Oikawa, T. Motoyama and
H. Yamada, J. Mater. Chem. C, 2013, 1, 6244–6251.
20 T. Motoyama, T. Kiyota, H. Yamada and K. Nakayama,
Sol. Energy Mater. Sol. Cells, 2013, 114, 156–160.
21 H. Yamada, C. Ohashi, T. Aotake, S. Katsuta, Y.
Honsho, H. Kawano, T. Okujima, H. Uno, N. Ono, S.
Seki and K. Nakayama, Chem. Commun., 2012, 48,
11136–11138.
22 A. A. Leitch, K. A. Stobo, B. Hussain, M. Ghoussoub,
S. Ebrahimi-Takalloo, P. Servati, I. Korobkov and J. L.
Brusso, Eur. J. Org. Chem., 2013, 5854–5863.
23 C. Fu, F. Rosei and D. F. Perepichka, ACS Nano, 2012, 6,
7973–7980.
Acknowledgements
This work was partly supported by Grants-in-Aid for Scientific
Research (KAKENHI) No. 23685030, 25107519 (‘AnApple’),
25288092, 25288113, 25620061, 26105004, 26288038,
26600004, and 26620167 from the Japan Society for the
Promotion of Science (JSPS), the Murata Science Foundation,
the PRESTO program by Japan Science and Technology
Agency (JST), Leading Industrial Technology Creation Project
by New Energy and Industrial Technology Development
Organization (NEDO), and the program for promoting the
enhancement of research universities in NAIST supported
by the Ministry of Education, Culture, Sports, Science and
Technology (MEXT). C.Q. thanks the JSPS for a post-doctoral
fellowship. We thank Ms Y. Nishikawa and Mr S. Katao
(NAIST) for help in mass spectrometry measurements and
single-crystal X-ray diffraction analyses, respectively.
Notes and references
24 J. L. Brusso, O. D. Hirst, A. Dadvand, S. Ganesan, F. Cicoira,
C. M. Robertson, R. T. Oakley, F. Rosei and D. F. Perepichka,
Chem. Mater., 2008, 20, 2484–2494.
1 C. Sekine, Y. Tsubata, T. Yamada, M. Kitano and S. Doi, Sci.
Technol. Adv. Mater., 2014, 15, 034203.
2 B. Kang, W. H. Lee and K. Cho, ACS Appl. Mater. Interfaces, 25 W.-J. Liu, Y. Zhou, Y. Ma, Y. Cao, J. Wang and J. Pei, Org.
2013, 5, 2302–2315.
Lett., 2007, 9, 4187–4190.
3 C. L. Chochos, N. Tagmatarchis and V. G. Gregoriou, 26 Y. Yi, L. Zhu and J.-L. Bredas, J. Phys. Chem. C, 2012, 116,
RSC Adv., 2013, 3, 7160–7181. 5215–5224.
4 P. F. Moonen, I. Yakimets and J. Huskens, Adv. Mater., 2012, 27 A. Pietrangelo, B. O. Patrick, M. J. MacLachlan and M. O.
24, 5526–5541. Wolf, J. Org. Chem., 2009, 74, 4918–4926.
5 S. Kola, J. Sinha and H. E. Katz, J. Polym. Sci., Part B: Polym. 28 A. Pietrangelo, M. J. MacLachlan, M. O. Wolf and B. O.
´
Phys., 2012, 50, 1090–1120.
Patrick, Org. Lett., 2007, 9, 3571–3573.
J. Mater. Chem. C
This journal is ©The Royal Society of Chemistry 2015

View Article Online
Journal of Materials Chemistry C
Paper
29 F. He, W. Wang, W. Chen, T. Xu, S. B. Darling, J. Strzalka, 33 S. Katsuta, H. Saeki, K. Tanaka, Y. Murai, D. Kuzuhara,
Y. Liu and L. Yu, J. Am. Chem. Soc., 2011, 133, 3284–3287.
30 Y. Zhou, W.-J. Liu, W. Zhang, X.-Y. Cao, Q.-F. Zhou, Y. Ma
and J. Pei, J. Org. Chem., 2006, 71, 6822–6828.
31 V. Mamane, P. Hannen and A. Fu¨rstner, Chem. – Eur. J.,
2004, 10, 4556–4575.
M. Misaki, N. Aratani, S. Masuo, Y. Ueda and H. Yamada,
J. Mater. Chem. C, 2014, 2, 986–993.
34 M. Kochi, Y. Harada, T. Hirooka and H. Inokuchi, Bull.
Chem. Soc. Jpn., 1970, 43, 2690–2702.
35 Y. Liu, M. S. Liu and A. K.-Y. Jen, Acta Polym., 1999, 50, 105–108.
32 S. Masuo, K. Tanaka, M. Oe and H. Yamada, Phys. Chem. 36 C. G. Hatchard and C. A. Parker, Proc. R. Soc. London, Ser. A,
Chem. Phys., 2014, 16, 13483–13488.
1956, 235, 518–536.
This journal is ©The Royal Society of Chemistry 2015
J. Mater. Chem. C