Synthesis of isomerically pure anti-anthradithiophene derivatives

Article abstract of DOI:10.1021/ol202089t

The regiospecific total synthesis and characterization of anti-isomers of 2,8-dialkylanthradithiophenes are described. The "anti" structure of the ADT derivatives is demonstrated by ¹³C NMR as well as single crystal X-ray diffraction.

Full text of DOI:10.1021/ol202089t

ORGANIC
LETTERS
2011
Vol. 13, No. 19
5208–5211
Synthesis of Isomerically Pure
anti-Anthradithiophene Derivatives
†
Benoıt Tylleman, Christophe M. L. Vande Velde,†,‡ Jean-Yves Balandier,† Sara Stas,†
ˆ
Sergey Sergeyev,†,§ and Yves Henri Geerts*,†
ꢀ
Universite Libre de Bruxelles (ULB), Faculte des Sciences, Laboratoire de Chimie des
Polymeres, CP 206/01, Boulevard du Triomphe, 1050 Bruxelles, Belgium, Karel de Grote
ꢀ
ꢁ
University College, Department of Applied Engineering, Salesianenlaan 30, 2660
Antwerp, Belgium, and University of Antwerp, Department of Chemistry,
Groenenborgerlaan 171, 2020 Antwerp, Belgium
ygeerts@ulb.ac.be
Received August 2, 2011
ABSTRACT
The regiospecific total synthesis and characterization of anti-isomers of 2,8-dialkylanthradithiophenes are described. The “anti” structure of the
ADT derivatives is demonstrated by 13C NMR as well as single crystal X-ray diffraction.
For decades, organic electronics have formed a hot topic
in materials science.1 A wide range of conjugated polymers
is currently used asorganic semiconductors.2 However, the
drawback of using polymers for optoelectronic applica-
tions is their low purity and difficult characterization
inherent to their mass distribution. Consequently, the use
of “small” organic molecules in electronic devices is an
interesting alternative, since theyare easier to produce with
a high purity degree. In addition, they order much better in
the solid state. Among the large variety of small molecules,
linear fused (hetero)acenes have been preferentially used
thanks to their high performances in electronic devices.1f,g
For example, pentacene has already demonstrated a charge
carrier mobility (μ) of 1.5 up to 5 cm2/V s in field-effect
3
transistors (FET).3 However, it is known for being poorly
stable toward photoinduced degradation, especially in
solution,4 which is a weak point for solution processed
organic electronic devices. Analogues of fused linear acenes
including heteroatoms in their aromatic core, such as
anthradithiophene (ADT) derivatives, have been described
and have shown (i) a better stability toward photo-oxida-
tion than their corresponding acenes and (ii) high charge
carrier mobilities of about 0.4 to 6.0 cm2/V s.5 The use of
3
(3) (a) Lin, Y. Y.; Gundlach, D. J.; Nelson, S.; Jackson, T. N. IEEE
Trans. Electron Devices 1997, 44, 1325. (b) Kelly, T. W.; Muyres, D. V.;
Baude, P. F.; Smith, T. P.; Jones, T. D. Mater. Res. Soc. Symp. Proc.
2003, 2, 678.
(4) Ono, K.; Totani, H.; Hiei, T.; Yoshino, A.; Saito, K.; Eguchi, K.;
Tomura, M.; Nishida, J. I.; Yamashita, Y. Tetrahedron 2007, 63, 9699.
(5) (a) Jurchescu, O. D.; Hamadani, B. H.; Xiong, H. D.; Park, S. K.;
Subramanian, S.; Zimmerman, N. M.; Anthony, J. E.; Jackson, T. N.;
Gundlach, D. J. Appl. Phys. Lett. 2008, 92, 132103. (b) Jurchescu, O. D.;
Subramanian, S.; Kline, R. J.; Hudson, S. D.; Anthony, J. E.; Jackson,
T. N.; Gundlach, D. J. Chem. Mater. 2008, 20, 6733.
(6) (a) Conrad, B. R.; Chan, C. K.; Loth, M. A.; Parkin, S. R.; Zhang,
X.; DeLongchamp, D. M.; Anthony, J. E.; Gundlach, D. J. Appl. Phys.
Lett. 2010, 97, 133306. (b) Laquindanum, J. G.; Katz, H. E.; Lovinger,
A. J. J. Am. Chem. Soc. 1998, 120, 664. (c) Jaquith, M. J.; Anthony, J. E.;
Marohn, J. A. J. Mater. Chem. 2009, 19, 6116. (d) Chen, M. C.; Kim, C.;
Chen, S.-Y.; Chiang, Y.-J.; Chung, M. C.; Facchetti, A.; Marks, T. J.
J. Mater. Chem. 2008, 18, 1029. (e) Lloyd, M. T.; Mayer, A. C.;
Subramian, S.; Mourey, D. A.; Herman, D. J.; Bapat, A. V.; Anthony,
J. E.; Malliaras, G. G. J. Am. Chem. Soc. 2007, 129, 9144.
†
ꢀ
Universite Libre de Bruxelles.
‡ Karel de Grote University College.
§ University of Antwerp.
(1) For example: (a) Mas-Torrent, M.; Rovira, C. Chem. Soc. Rev.
2008, 37, 827. (b) Allard, S.; Forstern, M.; Souharce, B.; Thiem, H.;
Scherf, U. Angew. Chem., Int. Ed. 2008, 47, 4070. (c) Facchetti, A.
Materials Today 2007, 10, 28. (d) Coakley, K. M.; McGehee, M. D.
Chem. Mater. 2004, 16, 4533. (e) Murphy, A. R.; Frechet, J. M. J. Chem.
Rev. 2007, 107, 1066. (f) Anthony, J. E. Chem. Rev. 2006, 106, 5028. (g)
Coropceanu, V.; Kwon, O.; Wex, B.; Kaafarani, B. R.; Gruhn, N. E.;
Durivage, J. C.; Neckers, D. C.; Bredas, J. L. Chem.;Eur. J. 2006, 12,
2073. (h) Sergeyev, S.; Pisula, W.; Geerts, Y. H. Chem. Soc. Rev. 2007,
36, 1902. (i) Coropceanu, V.; Cornil, J.; da Silva Filho, D. A.; Olivier, Y.;
ꢀ
ꢀ
ꢀ
Silbey, R.; Bredas, J.-L. Chem. Rev. 2007, 107, 926. (j) Roncali, J.;
Leriche, P.; Cravino, A. Adv. Mater. 2007, 19, 2045.
(2) (a) Dennler, G.; Scharber, M. C.; Brabec, C. J. Adv. Mater. 2009,
21, 1323. (b) Guenes, S.; Neugebauer, H.; Sariciftci, N. S. Chem. Rev.
2007, 107, 1324.
r
10.1021/ol202089t
Published on Web 09/14/2011
2011 American Chemical Society

Scheme 1. Synthesis of ADT anti-1a,b and Quinones mix-12a,b
ADT in FETs has been extensively explored, but the reported
studies have always relied on the use of a mixture of syn-/
anti-ADT isomers (Figure 1).5b,6 Takimiya et al. have re-
cently proposed the synthesis of isomeric syn- and anti-
linear naphtodithiophene (NDT) derivatives (Figure 1).
mixture of ADT amides and their use in polymer solar
cells.8 They have shown that the efficiency is higher for
bulk-heterojunctions based on poly(3-hexylthiophene)
solar cells incorporating the syn-ADT amide isomer. In
this context, we investigated the total synthesis of isomeri-
cally pure anti-ADT derivatives. Alkyl chains were at-
tached on positions 2 and 8 of the ADT skeleton in order
to fulfill solubility requirements to perform structural
characterizations.
ADT anti-1a,b derivatives were prepared as presented in
Scheme 1. The synthetic strategy was inspired by the work
of Wex et al. in which the preparation of a single isomer of
thienobisbenzothiophene is described.9
Starting materials 4a,b and 6 were prepared following
synthetic procedures previously described: synthons 4a,b
were obtained in quantitative yields by bromination of
2-alkylthiophenes 3a,b using N-bromosuccinimide10 and
2,5-dimethoxyterephthalaldehyde (6) was synthesized in
63% yield by lithiation of compound 5 with n-BuLi
followed by reaction of the corresponding dilithiated
species with DMF.11 Intermediates 7a,b were generated
in situ from 4a,b via a halogen dance reaction using LDA12
and were subsequently quenched with compound 6 to
Figure 1. Structures of anti-/syn-isomers of anthradithiophene
(ADT) and naphtodithiophene (NDT) derivatives.
They have demonstrated that the centrosymmetric (= anti-)
isomer exhibits a higher charge carrier mobility μ, up to
1.50 cm2/V s, than its axisymmetric counterpart (= syn-)
3
isomer that reaches at best μ = 0.06 cm2/V s.7 Lately,
Anthony et al. have reported the synthesis and the separa-
tion by fractional recrystallization of the syn-/anti-isomer
3
(9) (a) Wex, B.; Kaafarani, B. R.; Kirschbaum, K.; Neckers, D. C. J.
Org. Chem. 2005, 70, 4502. (b) Wex, B.; Kaafarani, B. R.; Neckers, D. C.
J. Org. Chem. 2004, 69, 2197.
(10) (a) Leroy, J.; Levin, J.; Sergeyev, S.; Geerts, Y. H. Chem. Lett.
2006, 35, 166. (b) Leroy, J.; Boucher, N.; Sergeyev, S.; Sferrazza, M.;
Geerts, Y. H. Eur. J. Org. Chem. 2007, 8, 1256.
(7) Shinamura, S.; Osaka, I.; Miyazaki, E.; Nakao, A.; Yamagishi,
M.; Takeya, J.; Takimiya, K. J. Am. Chem. Soc. 2011, 133, 5024.
(8) Li, Z.; Lim, Y. F.; Kim, J. B.; Parkin, S. R.; Loo, Y. L.; Malliaras,
G. G.; Anthony, J. E. Chem. Commun. 2011, 47, 7617.
Org. Lett., Vol. 13, No. 19, 2011
5209

provide the diols 8a,b. These were then reduced without
further purification into compounds 9a (49% from 6) and
9b (66% from 6) using NaBH3CN and ZnI2.9 The exact
positions of the bromine atoms were confirmed by HMBC
2D-NMR (Figure SI6) as well as single crystal XRD
structure determination (Figure SI1). Compounds 10a,b
were obtained in good yields (10a: 93% and 10b: 94%)
by formylation of dibrominated derivatives 9a,b using
n-BuLi, followed by reaction of the organolithium inter-
mediates with DMF. The next step was performed using
Amberlyst acidic resin as a cyclization/aromatization
agent.9 Surprisingly, the initially expected anti-11a,b ADT
derivatives were not achieved from the cyclization step
since anti-12a,b quinones were always isolated. Several
experimental conditions, listed in Table 1, have been tested
on the hexyl derivative 10a in order to optimize the
reaction yield. The first conditions applied were the
ones described by Wex et al. (entry 1). In our case, no
reaction was observed since the starting compound 10a
was quantitatively recovered even after 64 h of reaction
time. The use of toluene instead of benzene (entry 2) led
to the formation of the quinone anti-12a in 23% yield.
Reaction in p-xylene, which possesses a higher boiling
point than benzene and toluene, was found to be
unsuccessful since degradation of the reaction mixture
was observed (entry 3). Using a different cyclization/
aromatization agent from Amberlyst-15, i.e. p-tolue-
nesulfonic acid (p-TSA)13a (entry 4) or Amberlyst-3613b
(entry 5), resulted in the degradation of the reaction
mixture (with p-TSA) and in recovery of the starting
material (with Amberlyst-36). It is rational to sup-
pose that the cyclization/aromatization process might
involve charged intermediates. Thus, an effect on
the reaction is expected by changing the solvent polar-
ity. Indeed, the reaction in nonpolar decane (entry 6)
led to the formation of the quinone anti-12a in 6%
yield whereas the use of polar chlorobenzene (entry 7)
provided the compound anti-12a in 38% yield. The two
conditions which gave the best results (entries 2 and 7)
were repeated on the derivative 10b bearing 3,7-di-
methyloctyl chains instead of hexyl chains. Similar
observations were made since quinone anti-12b was
obtained in 22% and 37% yield, using toluene and
chlorobenzene as reaction solvents, respectively.
Table 1. Optimization of the Cyclization Step Conditions on 10a
C/A
temp
time
(h)
yield of
entry
agent
solvent
(°C)
anti-12a (%)
1
2
3
4
5
6
7
A15
A15
A15
p-TSA
A36
A15
A15
benzene
toluene
p-xylene
toluene
touene
80
110
140
110
110
130
130
64
64
64
64
64
64
29
no reaction
23
degradation
degradation
no reaction
6
decane
Cl-benzene
38
mixtures of quinones mix-12a,b, in 73% (mix-12a) and in
58% (mix-12b) yields, was carried out by condensation of
5-alkylthiophene-2,3-carboxaldehyde 13a,b with cyclohex-
ane-1,4-dione 14 in the presence of 5% KOHaq solution
(Scheme 1).6b
1H NMR and UVÀvisible measurements did not afford
any information on the synthesis of isomerically pure qui-
1
nones anti-12a,b since their H NMR and absorption
spectra are identical to those of mix-12a,b (Figures SI3,5,-
7,9,23À26). On the contrary, the presence of single anti-
quinone isomers was established using 13C NMR spectro-
scopy where slight changes were observed for quaternary
carbons CB, CD, CI, and CG located in the aromatic region
(Figure 2). For the pure isomers anti-12a,b these carbon
atoms appear as single peaks at δ = 129.0, 130.5 ppm for
CB/I and at δ = 144.4, 144.6 ppm for CD/G whereas twin
peaks are observed in the 13C NMR spectra of mix-12a,b
indicating the presence of an isomer mixture (Figure 2).
This observation shows that single anti-isomers were
formed for compounds anti-12a,b, and it was also con-
firmed by X-ray diffraction measurement performed on
a crystalline splinter of anti-12a (crystals were grown by
slow evaporation of CH2Cl2 solution at room tempera-
ture, Figure 3). A slight amount of disorder (8.2%) is
present in the orientation of the thiophene rings, as it is
often the case for thiophene systems with unsubstituted
position 3.14 The disorder is also identical on either side
of the molecule, in agreement with the crystallographic
symmetry.
13C NMR spectroscopy indicates that the product con-
tains only the anti-12a,b isomer and the formation of the
syn-product is impossible from the geometryof the starting
compound 9a,b, which was independently confirmed
(Figures SI1,6). Therefore, the observed disorder is not a
result of the presence of a mixture of syn- and anti-isomers,
but 8.2% of the molecules are simply flipped over in the
lattice. The impact of the molecular shape on entering it
into the lattice in the wrong way is minimal. The effect is
also minimal from the energy point of view: the sulfur side
At first glance, the yields of 22À37% obtained for the
synthesis of products anti-12a,b seem rather low. However,
taking into account that four reactions are involved, cycli-
zation, aromatization, cleavage of the methoxy groups,
and oxidation (Figure SI31), the observed yields appear
appreciable. In order to confirm the selective formation
of isomerically pure quinones anti-12a,b by spectro-
scopic characterizations, the synthesis of syn-/anti-isomers
(11) Kuhnert, N.; Rossignolo, G. M.; Lopez-Periago, A. Org.
Biomol. Chem. 2003, 1, 1157.
(14) As an example, we searched the Crystal Structure Database
(v5.31, with four updates) for unsubstituted benzo[b]thiophenes in the
3-position. Molecules substituted asymmetrically on the benzo-
[b]thiophene benzene ring were removed from the resulting set. This
yields 76 benzo[b]thiophenes which can in principle display this type of
disorder. Of these 76, 43 are disordered (57%).
€
(12) Frohlich, J.; Hametner, C.; Kalt, W. Monatsh. Chem. 1996, 127,
325.
(13) (a) p-TSA = lower molecular weight analogue of Amberlyst
resins; (b) Capacity of Amberlyst-15 = 4.70 equiv/kg; Amberlyst-36 =
5.40 equiv/kg.
5210
Org. Lett., Vol. 13, No. 19, 2011

Figure 2. Comparison of selected window of 13C NMR spectra of (I) mix-12a, (II) anti-12a, (III) mix-12b, and (IV) anti-12b.
δ = 8.72 and 8.58 ppm for anti-12b) and Hc (δ = 7.23 and
7.21 ppm for anti-12a,b respectively) are shifted to higher
fields (δ(Ha,b) = 7.79 and 7.68 ppm; δ(Hc) = 6.99 ppm)
for anti-1a,b. The synthesis of ADT anti-1a,b was also
confirmed by (i) mass spectrometry by the observation of
m/z ratios at 458.2108 for 1a and at 570.3368 for 1b
(Figures SI29,30) and (ii) optical measurements since
absorption spectra of compounds anti-1a,b present typical
ADT spectral profiles showing three waves beyond 410
nm: 424, 452, and 483 nm for 1a and 425, 452, and 483 nm
for 1b (Figures SI27,28).6d
Figure 3. Single crystal XRD image of anti-12a. Displacement
ellipsoids at the 50% level (The structure has overlap between 3
In conclusion, we have reported the first total synthesis
of the 5 rings in a stairway-like arrangement. The intermolecular
˚
distance amounts to 3.56 A. Neighboring ADT are shifted up or
down by half of this intermolecular distance, leading to stag-
of pure anti-isomers of anthradithiophene derivatives
using a six-step synthetic pathway. The anti-configuration
of the ADT compounds has been unambiguously proven
by 13C NMR and X-ray diffraction experiments on their
quinone precursors. This synthetic work, which can be
regarded as a first milestone in ADT chemistry, offers an
opportunity for further physical studies on the property
performance of anti-ADT compared to the syn-/anti-ADT
mixture in electronic devices such as field effect transistors.
gered columns. The hexyl chains can be found in layers between
the anthradithiophene layers). The .cif file is available as Sup-
porting Information.
of the ring exchanges an edge-to-edge CHÀS interaction in
the major orientation for interactions with the hexyl chains
in the minor orientation.
Finally, quinones anti-12a,b were reduced into ADT
anti-1a,b in 30% yields using LiAlH4. Although both
products present poor solubility in common organic sol-
vents, their 1H NMR spectra have been recorded (Figures
SI21,22). Comparison of the 1H NMR spectra of ADT
anti-1a,b with those of their quinone precursors anti-12a,b
(Figures SI3,5) shows the presence of an extra signal in the
aromatic region at δ = 6.14 ppm which is assigned to the
protons in positions 5,11of the ADT core. After reduction,
signals for protons Ha,b (δ = 8.76 and 8.62 ppm for anti-12a;
Acknowledgment. We thank Pascal Gerbaux (UMons)
for MS measurements, Michel Luhmer (ULB) for NMR
experiments, and Matthias Zeller (YSU) for XRD data
collection. B.T. is a FRIA research fellow.
Supporting Information Available. Experimental pro-
cedures, optical and NMR spectra and XRD data in CIF
format for 9a and anti-12a. This material is free of charge
via the Internet at http://pubs.acs.org.
Org. Lett., Vol. 13, No. 19, 2011
5211