Meta-substituted thienyl benzenes: A comparative Synthetic, structural and computational study

Article abstract of DOI:10.1021/jo801771d

A selection of metal-catalyzed C-C bond-forming strategies has been evaluated in the synthesis of a series of meta-substituted thienylbenzenes, (T_n)₂C₆H₄ and (T_n) ₃C₆H₃ (n = 1, 2; T₁ = 2-thienyl; T₂ = 2, 2′-bithien-5-yl). Kumada coupling reactions catalyzed by PdCl₂(dppf) between the appropriate thienyl Grignard and either 1, 3- or 1, 3, 5-bromo- or iodobenzenes were found to be the most reliable in terms of convenience, selectivity and yield (dppf = 1, 1′- bis(diphenylphosphino)ferrocene). These conditions also allowed the optimized syntheses of mixed (thienyl)(halo)benzenes, (T_n)C₆H ₃X₂ and (T_n)₂C₆H ₃X(X = Br, I); the latter bromides could be further elaborated in subsequent Stille, Sonogashira, or Kumada reactions to furnish bis(thienyl) compounds bearing electron-donating or -withdrawing groups in the third meta position, (T_n)₂C₆H₃R (R= Ph, p-MeOC₆H₄, p-FC₆H₄; n = 1, 2) and (T₁)₂C₆H₃R′ (R′ = Me, C=CPh, Fc; Fc = ferrocenyl). The relative effects of R, R′, and n were evaluated by electronic spectroscopy, cyclic voltammetry and calculation. The absorption and emission characteristics and calculated ionization potentials and HOMO-LUMO gaps of these compounds were strongly dependent on n and largely insensitive to R/R′. These measured and calculated properties were also found to be largely invariant with respect to the degree of substitution about the central ring in the meta-substituted benzenes (T_n) _mX_3-mC₆H₃ (m = 1-3; n = 1, 2; X = Br, H), although in the case of n = 1, there was a smooth, albeit small, increase in the emission maximum with increasing thienyl substitution. These findings essentially confirmed earlier theoretical predictions that the thienyl "arms" of meta-substituted phenyl-cored dendrimers were predominantly responsible for absorption and that excitons were localized to the "arms" without any electronic coupling between them, but also introduced the caveat that there was a minimum of two contiguous thiophene rings required for strict exciton localization to the arms. The oxidation potentials of the compounds in solution ranged from 0.9 to 1.6 V and were not rationally dependent on the degree of substitution or the nature of R/R′.

Full text of DOI:10.1021/jo801771d

Meta-Substituted Thienyl Benzenes: A Comparative Synthetic,
Structural and Computational Study
Angelica L. P. Cornacchio,† Jacquelyn T. Price,† Michael C. Jennings,† Robert McDonald,‡
Viktor N. Staroverov,† and Nathan D. Jones*,†
Department of Chemistry, The UniVersity of Western Ontario, London, Ontario, Canada N6A 5B7
njones26@uwo.ca
ReceiVed August 7, 2008
A selection of metal-catalyzed C-C bond-forming strategies has been evaluated in the synthesis of a series
of meta-substituted thienylbenzenes, (Tn)2C6H4 and (Tn)3C6H3 (n ) 1, 2; T1 ) 2-thienyl; T2 ) 2,2′-bithien-
5-yl). Kumada coupling reactions catalyzed by PdCl2(dppf) between the appropriate thienyl Grignard and
either 1,3- or 1,3,5-bromo- or iodobenzenes were found to be the most reliable in terms of convenience,
selectivity and yield (dppf ) 1,1′-bis(diphenylphosphino)ferrocene). These conditions also allowed the optimized
syntheses of mixed (thienyl)(halo)benzenes, (Tn)C6H3X2 and (Tn)2C6H3X (X ) Br, I); the latter bromides
could be further elaborated in subsequent Stille, Sonogashira, or Kumada reactions to furnish bis(thienyl)
compounds bearing electron-donating or -withdrawing groups in the third meta position, (Tn)2C6H3R (R )
Ph, p-MeOC6H4, p-FC6H4; n ) 1, 2) and (T1)2C6H3R′ (R′ ) Me, C≡CPh, Fc; Fc ) ferrocenyl). The relative
effects of R, R′, and n were evaluated by electronic spectroscopy, cyclic voltammetry and calculation. The
absorption and emission characteristics and calculated ionization potentials and HOMO-LUMO gaps of these
compounds were strongly dependent on n and largely insensitive to R/R′. These measured and calculated
properties were also found to be largely invariant with respect to the degree of substitution about the central
ring in the meta-substituted benzenes (Tn)mX3-mC6H3 (m ) 1-3; n ) 1, 2; X ) Br, H), although in the case
of n ) 1, there was a smooth, albeit small, increase in the emission maximum with increasing thienyl
substitution. These findings essentially confirmed earlier theoretical predictions that the thienyl “arms” of
meta-substituted phenyl-cored dendrimers were predominantly responsible for absorption and that excitons
were localized to the “arms” without any electronic coupling between them, but also introduced the caveat
that there was a minimum of two contiguous thiophene rings required for strict exciton localization to the
arms. The oxidation potentials of the compounds in solution ranged from 0.9 to 1.6 V and were not rationally
dependent on the degree of substitution or the nature of R/R′.
Introduction
the last several years, both as monomers in cross-linked semicon-
ducting polymers3-6 and as the components of conjugated
dendrimers.1,2,7-9 These and closely related materials have potential
Branched oligothiophenes based on central phenyl (or, oc-
casionally, thienyl1-3) rings (Chart 1) have come to the fore over
‡ Department of Chemistry, The University of Alberta, Edmonton, Alberta,
Canada T6G 2G2.
† The University of Western Ontario.
530 J. Org. Chem. 2009, 74, 530–544
10.1021/jo801771d CCC: $40.75 2009 American Chemical Society
Published on Web 12/16/2008

Meta-Substituted Thienyl Benzenes
CHART 1. Examples of Branched Oligothiophenes Based On a Central Benzene Ring
application in a wide range of devices, for example, field-effect
transistors,10 organic light-emitting diodes (OLEDs; C),11 photo-
voltaics (B,9 D12,13) and other light-harvesting applications (E),14
that exploit the unique optical and electronic properties of olig-
othiophenes (UV-visible absorption and fluorescence, semicon-
ductivity, etc.), and the physical advantages conferred by their
dendrimeric nature (determinate syntheses, flexibility, processability
from solution, etc.). In addition, these several physical character-
istics can be “tuned.” For example, the absorption and emission
maxima of A were red-shifted by increasing the degree of
substitution around the core,15 as were those of B by increasing
the number of generations in the dendrimer; 9 and the incoporation
of more rigid benz-fused and truxene cores in D13 and E,14
respectively, led to enhanced conjugation and even greater batho-
chromic shifts.
In these and simpler systems, the meta-disubstituted thienyl-
benzene motif has played an important role. It has been
(1) Xia, C. J.; Fan, X.; Locklin, J.; Advincula, R. C.; Gies, A.; Nonidez, W.
J. Am. Chem. Soc. 2004, 126, 8735.
(2) Xia, C. J.; Fan, X. W.; Locklin, J.; Advincula, R. C. Org. Lett. 2002, 4,
2067.
CHART 2. Relevant Thienylbenzenes
(3) Sun, X.; Liu, Y.; Chen, S.; Qiu, W.; Yu, G.; Ma, Y.; Qi, T.; Zhang, H.;
Xu, X.; Zhu, D. AdV. Funct. Mater. 2006, 16, 917.
(4) Cherioux, F.; Guyard, L.; Audebert, P. Chem. Commun. 1998, 2225.
(5) Cherioux, F.; Guyard, L. AdV. Funct. Mater. 2001, 11, 305.
(6) Belot, C.; Filiatre, C.; Guyard, L.; Foissy, A.; Knorr, M. Electrochem.
Commun. 2005, 7, 1439.
(7) Ko¨se, M. E.; Mitchell, W. J.; Kopidakis, N.; Chang, C. H.; Shaheen,
S. E.; Kim, K.; Rumbles, G. J. Am. Chem. Soc. 2007, 129, 14257.
(8) Kopidakis, N.; Mitchell, W. J.; van de Lagemaat, J.; Ginley, D. S.;
Rumbles, G.; Shaheen, S. E.; Rance, W. L. Appl. Phys. Lett. 2006, 89, 103524.
(9) Mitchell, W. J.; Kopidakis, N.; Rumbles, G.; Ginley, D. S.; Shaheen,
S. E. J. Mater. Chem. 2005, 15, 4518.
(10) Ponomarenko, S. A.; Kirchmeyer, S.; Elschner, A.; Huisman, B.-H.;
Karbach, A.; Drechsler, D. AdV. Funct. Mater. 2003, 13, 591.
(11) Justin Thomas, K. R.; Lin, J. T.; Tao, Y.-T.; Ko, C.-W. Chem. Mater.
2002, 14, 1354.
(12) de Bettignies, R.; Nicolas, Y.; Blanchard, P.; Levillain, E.; Nunzi, J.-
M.; Roncali, J. AdV. Mater. 2003, 15, 1939.
J. Org. Chem. Vol. 74, No. 2, 2009 531

Cornacchio et al.
CHART 3. Compounds Made in this Study
incoporated by way of monomers F16-18 and H18 (Chart 2) into
conjugated polymers, and has also found application in singlet
oxygen-generating photosensitizing agents (F and G, n ) 0).19
It is now well established that the physical properties of
thiophenes (and of their respective polymers) can be tuned by
the introduction of electron-donating or -withdrawing groups
to the ring.20-23 Studies of a series 3-(p-R-phenyl)thiophene
“monomers” K (Chart 2, R ) -H, -CMe3, -Me, -OMe, -COMe,
-CO2Et, -F, -Cl, -Br, -CF3, -SO2Me) by Guerrero et al. have
demonstrated that the oxidation potentials of both the monomers
and resulting polymers correlated well with the Hammett sigma
constants of the substituents.23 As expected, the addition of an
electron-donating substituent (σp < 0) to the para-position on
the phenyl ring shifted the redox potentials of the thienyl
compounds cathodically, whereas addition of electron-with-
drawing substituents (σp > 0) shifted them anodically.
SCHEME 1. Reported Metal-Catalyzed Syntheses of 1 (M
) MgBr, ZnCl) and 3 (M ) SnBu3, ZnCl, B(OH)2)
benzenes (1-4, Chart 3). Although a selection of these9,27-29
(and other noncatalytic)4-6,30-32 methods have been used in
the past, we present improved syntheses for these known
compounds as well as new rational methods to make the
valuable mixed (halo)(thienyl)benzene compounds I and J in
synthetically useful yields (Chart 2, X ) Br, I; J: n ) 0 (5Br,
5I); n ) 1 (6Br, 6I). I: n ) 0 (7Br, 7I); n ) 1 (8Br, 8I)). We
demonstrate that the halogen handle in group I may be used to
give a family of elaborated compounds (9-17, Chart 3) whose
properties depend in principle on the nature of the R1 group in
the third meta-position. We use density functional theory
methods to calculate the gas-phase first ionization potential and
HOMO-LUMO gap for all compounds and correlate these data
with the nature of the R1 group, the degree of substitution about
the central phenyl ring, and the number of contiguous thienyl
rings. In addition, we present the molecular structures of 4 and
12, including an analysis of their packing in the solid state.
It has also been recognized that the incorporation of aryl
substituents into poly(thiophenes) is potentially advantageous
for two reasons: not only do the aryl rings provide potential
anchoring sites for onward elaboration, but also they introduce
stereoelectronic effects that alter the physical and conducting
properties of the polymer.20,24
Despite the considerable interest in molecules of this type,
however, convenient syntheses are not always available, and
progress in the field has been limited by practical difficulties.
Although overall yields in some cases are respectable given the
synthetic challenge, for example, about 15% over 5-7 steps
for B,9 the reality is often daunting, for example, about 3% over
9 (n ) 1) or 11 (n ) 2) steps for D.13 Even apparently simple
thienylbenzene compounds are often difficult to access. Reported
syntheses are typically plagued by the problems generally
associated with Pd-catalyzed coupling of aromatic rings: pre-
functionalization of reagents, long reaction times, arduous
separations of partially substituted products,5 and low isolated
yields. The eminently valuable meta-substituted benzenes that
bear both thiophene and halogen groups (I and J, Chart 2),
which would allow both electrochemical and chemical elabora-
tion, respectively, are known primarily as unintended byproducts
of other reactions.25 Targeted approaches to these compounds
would furnish new monomers for electropolymerized conducting
materials, but remain undeveloped. Moreover, solid-state X-ray
crystallographic data for the product monomers are lacking.26
In this paper, we combine the aspects of meta-substitution,
aryl incorporation, and inclusion of both electron-donating and
-withdrawing groups, in a synthetic, structural, and computa-
tional study of meta-substituted thienylbenzenes. We systemati-
cally compare Suzuki, Stille, Kumada, and Negishi coupling
approaches to the syntheses of both di- and trisubstituted thienyl
(13) Nicolas, Y.; Blanchard, P.; Levillain, E.; Allain, M.; Mercier, N.; Roncali,
J. Org. Lett. 2004, 6, 273.
(14) Wang, J.-L.; Luo, J.; Liu, L.-H.; Zhou, Q.-F.; Ma, Y.; Pei, J. Org. Lett.
2006, 8, 2281.
(15) Pappenfus, T. M.; Mann, K. R. Org. Lett. 2002, 4, 3043.
(16) Pelter, A.; Maud, J. M.; Jenkins, I.; Sadeka, C.; Coles, G. Tetrahedron
Lett. 1989, 30, 3461.
(17) Sato, T.; Hori, K.; Tanaka, K. J. Mater. Chem. 1998, 8, 589.
(18) Song, C.; Swager, T. M. Macromolecules 2005, 38, 4569.
(19) Marles, R. J.; Compadre, R. L.; Compadre, C. M.; Soucy-Breau, C.;
Redmond, R. W.; Duval, F.; Mehta, B.; Morand, P.; Scaiano, J. C.; Arnason,
J. T. Pestic. Biochem. Physiol. 1991, 41, 89.
(20) Waltman, R. J.; Bargon, J. Can. J. Chem. 1986, 64, 76.
(21) Sarker, H.; Gofer, Y.; Killian, J. G.; Poehler, T. O.; Searson, P. C. Synth.
Met. 1997, 88, 179.
(22) Alhalasah, W.; Holze, R. Microchim. Acta 2007, 156, 133.
(23) Guerrero, D. J.; Ren, X.; Ferraris, J. P. Chem. Mater. 1994, 6, 1437.
(24) Andersson, M. R.; Berggren, M.; Inganaes, O.; Gustafsson, G.;
Gustafsson-Carlberg, J. C.; Selse, D.; Hjertberg, T.; Wennerstroem, O. Macro-
molecules 1995, 28, 7525.
(25) Bras, J.; Guillerez, S.; Pépin-Donat, B. Chem. Mater. 2000, 12, 2372.
(26) Thallapally, P. K.; Chakraborty, K.; Carrell, H. L.; Kotha, S.; Desiraju,
G. R. Tetrahedron 2000, 56, 6721.
(27) Sato, T.; Hori, K.; Fujitsuka, M.; Watanabe, A.; Ito, O.; Tanaka, K.
J. Chem. Soc., Faraday Trans. 1998, 94, 2355.
(28) Pelter, A.; Jenkins, I.; Jones, D. E. Tetrahedron 1997, 53, 10357.
(29) Obara, S.; Tada, K. EP 2004/153942.
(30) Rebourt, E.; Pépin-Donat, B.; Dinh, E. Polymer 1995, 36, 399.
(31) Kim, T. Y.; Kim, H. S.; Lee, K. Y.; Kim, J. N. Bull. Korean Chem.
Soc. 1999, 20, 1255.
(32) Kim, T. Y.; Kim, H. S.; Lee, K. Y.; Kim, J. N. Bull. Korean Chem.
Soc. 2000, 21, 521.
532 J. Org. Chem. Vol. 74, No. 2, 2009

Meta-Substituted Thienyl Benzenes
SCHEME 2. Palladium-Catalyzed Kumada Couplings of Trihalobenzenes and 2-Thienylmagnesium Bromide To Give 3, 5
and 7
SCHEME 3. Palladium-Catalyzed Synthesis of 4 by Suzuki
Coupling
is also commercially available, but significantly more expen-
sive.) The Kumada reactions featured only slight excesses of
the thienyl starting material (<2 mol equiv of Grignard reagent
per halide) and low catalyst loadings (3 mol %) and were
conducted under mild synthetic conditions (rt for 48 h). The
Grignard reagent, 2-thienylmagnesium bromide, was easily
generated in situ from commercially available 2-bromothiophene.
Although the synthesis of 1 by Negishi cross-coupling was
comparable to our Kumada reaction (81 vs 80%, respectively),
the necessity for prelithiation in the generation of the 2-thie-
nylzinc chloride starting material in the former case made it
less desirable. The Suzuki cross-coupling reaction reported in
the patent literature29 gave the highest reported yield of 3 (90%)
in the shortest reaction time, but in our hands synthesis of the
requisite boronic acid precursor by the reported method was
unreliable. We had similar experience with the preparation of
2,2′-dithienyl-5′-boronic acid for attempted Suzuki synthesis of
4 (vide infra). Therefore, the Kumada reaction was once again
preferable based on the relative ease of in situ preparation of
the Grignard reagent.
Compound 3 has also been made using various cyclotrim-
erization reactions,4-6,30-32 but the overall yields (starting from
thiophene) were significantly lower than that of our Kumada
reaction (e.g., 86 vs 56% for trimerization of 2-acetylth-
iophene5). In addition, this approach was limited to production
of trisubstituted species, while stepwise metal-catalyzed coupling
routes gave access compounds of lower symmetry (vide infra).
Our Pd-catalyzed synthesis of 3 gave predominantly the
intended product, but also trace quantities of 1,3-diiodo-5-(2′-
thienyl)benzene (5I) and 1-iodo-3,5-bis(2′-thienyl)benzene (7I;
Scheme 2; X ) I). Compounds like 7I, which have previously
been considered only as unwanted byproduct, and for which
dedicated syntheses have not been developed, are potentially
very useful in their own rights because they offer both chemical
and electrochemical handles for onward manipulation. There-
fore, the Kumada reaction was optimized in order to favor the
production of the disubstituted product. We reasoned that 1,3,5-
tribromobenzene would be a better starting material than 1,3,5-
triiodobenzene for the production of mixed (halo)(thienyl)ben-
zenes because the more sluggish reaction of the arylbromide
would make controlled partial substitution more feasible. Thus,
7Br was produced in the highest yield (46%) when a 2:1 ratio
of 2-thienylmagnesium bromide and 1,3,5-tribromobenzene was
used, and the reaction mixture was heated gently overnight (60
°C) in the presence of PdCl2(dppf) catalyst. Only traces of 3
and a small quantity of 5Br (10%) were formed in addition to
the intended 7Br (Scheme 2). We also assayed under identical
conditions Pd2(dba)3 ·CHCl3/4PPh3, PdCl2(PPh3)2, and Pd(PPh3)4
as precatalysts for this transformation; these produced 7Br in
56, 49, and 54% yields, respectively. However, they also gave
Results and Discussion
Syntheses of 1 and 3. The syntheses of 1,3-bis- (1)27,28 and
1,3,5-tris(2′-thienyl)benzene (3),9,28,29 have been reported previ-
ously; a variety of metal-catalyzed coupling reactions and, in
the case of 3, cyclotrimerization reactions4-6,30-32 have been
used. In the synthesis of 1, both nickel-catalyzed Negishi
(65%)28 and Kumada (81%)27 couplings have been used
(isolated yields in parentheses), while 3 has been made by Pd-
catalyzed Stille (47%),9 Negishi (60%),28 and Suzuki (90%)29
reactions. In every case but that of the Kumada reaction, the
zinc, tin, or boron-containing precursors were relatively straight-
forward to prepare but required a prelithiation step (e.g.,
2-tributylstannylthiophene was generated from 2-lithiothiophene
and tributylstannyl chloride). All of these couplings used either
1,3-di- or 1,3,5-tribromobenzene starting materials (Scheme 1).
In our synthetic survey, we found that the most reliable
method was instead the Kumada cross-coupling of either 1,3-
di- or 1,3,5-triiodobenzene and 2-thienylmagnesium bromide
using PdCl2(dppf) (dppf ) 1,1′-bis(diphenylphosphino)fer-
rocene) as catalyst, which gave 1 or 3 in 80 or 86% yield,
respectively. We chose PdCl2(dppf) as our key catalyst because
it was the most discriminating between iodo- and bromoben-
zenes of all the Pd complexes we surveyed and it, therefore,
permitted excellent control over product distribution based on
the choice of halide in the starting material. Thus, in contrast
to the excellent cross-coupling involving 1,3,5-triiodobenzene,
the PdCl2(dppf)-catalyzed reaction between 2-thienyl Gringard
and tribromobenzene gave predominantly 1-bromo-3,5-bis(2′-
thienyl)benzene (7Br) and only traces of 3 (Scheme 2).
In the synthesis of 3, we used triiodo- instead of tribromoben-
zene to encourage complete substitution; this did not introduce
significant complications because 1,3,5-triiodobenzene was
readily accessible from its commercially available bromo
analogue.33 (At the time of this writing, 1,3,5-triiodobenzene
(33) Gan, Z.; Roy, R. Can. J. Chem. 2002, 80, 908.
J. Org. Chem. Vol. 74, No. 2, 2009 533

Cornacchio et al.
SCHEME 4. Palladium-Catalyzed Stille Couplings of Trihalobenzenes and 5′-Tributylstannyl-2,2′-bithiophene To Give 4, 6, 8,
and Unwanted Tetrathiophene Side-Product
significant quantities of the unwanted trisubstituted product 3
(34, 44, and 16%, respectively). Therefore, the PdCl2(dppf)
precatalyst was deemed better for the selective production of
meta-disubstituted thienylbenzenes.
Although ours is the first metal-catalyzed synthesis of 4, the
compound has also been made using cyclotrimerization
reactions,5,34 but reported yields were only moderate (56-65%).
As expected, the Stille coupling synthesis of 4 gave a mixture
of products from which 1,3-diiodo-5-{5′-(2′,2′′-bithienyl)}benzene
(6I), and 1-iodo-3,5-bis{5′-(2′,2′′-bithienyl)}benzene (8I) were
also isolated in trace quantities (Scheme 4; X ) I). A significant
side-product was tetrathiophene. We once again altered the
synthesis to favor production of either 6 and 8: When a 2:1
ratio of stannylthiophene:bromobenzene was heated to 60 °C
for 48 h in the presence of PdCl2(dppf) (3 mol %), the desired
products were obtained in poor yield (6Br, 10% and 8Br, 27%);
under these conditions, 4 was not formed, and the production
of tetrathiophene was limited to 20%. The low isolated yields
of 4, 6Br, and 8Br in these reactions were in part due to difficulty
in separating them from tetrathiophene, which was prone to
streaking during purification by column chromatography.
Syntheses of 9-17. To explore the effects of electron-
donating and -withdrawing groups on the physical properties
of the meta-substituted thienylbenzenes, we made a series of
new compounds (Chart 3) with the general formula 1,3-bis(2′-
thienyl)-5-R-benzene, where R ) Me (9), Ph (10), p-C6H4OMe
(11), p-C6H4F (12), CCPh (13), and Fc (14; Fc ) ferrocenyl).
We also made a smaller family of the analogous bithienyl-
substituted compounds, 1,3-bis({5′-(2′,2′′-bithienyl)})-5-R-
benzene, where R ) Ph (15), p-C6H4OMe (16), p-C6H4F (17).
Coupling of 7Br or 8Br with the appropriate Grignard reagent
in the presence of PdCl2(dppf) as catalyst gave compounds
10-12 and 15-17 in modest to good yields (35-84%; Scheme
5). Slight excesses of the Grignard reagent (1.5 mol equiv) and
low catalyst loading (3 mol %) were used.
Syntheses of 2 and 4. Due to the tedious synthesis of the
required 5′-bromo-2,2′-bithiophene precursor for the Kumada
syntheses of 1,3-bis- (2) and 1,3,5-tris{5′-(2′,2′-bithienyl)ben-
zene (4), we opted instead to take a Stille cross-coupling
approach. Reactions between either 1,3-di- or 1,3,5-triiodoben-
zene and 5′-tributylstannyl-2,2′-bithiophene using PdCl2(dppf)
as the catalyst gave 2 and 4 in 83 or 18% yield, respectively; a
similar synthesis by Song and Swager using PdCl2(PPh3)2 as
catalyst gave 2 in 86% yield. Another reported synthesis of 2
involved the Ni-catalyzed Kumada coupling of 2-thienylmag-
nesium bromide with 1,3-bis{5′′-bromo-5′-(2′,2′′-bithienyl)ben-
zene, but this gave only a 9% yield.
As was the case in the synthesis of 1 and 3, these reactions
used slight excesses of the bithienyl starting materials (1.5 mol
equiv per halide) and low catalyst loadings (3 mol %). Although
the stannyl reagent was straightforward to synthesize by reaction
of 2,2′-bithiophene with n-BuLi followed by addition of
n-Bu3SnCl, purification of the tin-containing product by vacuum
distillation required high temperature (∼450 °C) and resulted
in only moderate yields (76%). Because the isolated yield of 4
was low, we investigated other possible synthetic routes.
Inspired by the reported synthesis of 3 by Obara and Tada,29
we applied the Suzuki cross-coupling reaction to the synthesis
of 4. In this reaction, 2,2′-dithienyl-5-boronic acid was coupled
with 1,3,5-triiodobenzene in the presence of Pd(OAc)2 (5 mol
%) and PPh3 to afford 4 in high yield (90%, Scheme 3). The
disadvantage to this approach was that the synthesis of the
boronic acid starting material proved to be very unreliable; on
some occasions, no product was formed. Moreover, there was
no reported method of purification of the acid. The crude mixture
was therefore used in the Suzuki coupling reaction; this gave a
reaction mixture that was difficult to purify using flash chro-
matography. Although the Stille reaction was very low yielding
by constrast, the metallated coupling reagent, 5′-tributylstannyl-
2,2′-bithiophene, could be obtained relatively easily.
The synthesis of 9 by Kumada coupling of MeMgBr and 7Br
in the presence of PdCl2(dppf) gave three products: the intended
compound 9, 1, and the homocoupled compound 18 (Scheme
6). Compound 1 presumably resulted from the hydrolysis of
the Grignard reagent derived from 7Br, probably during workup,
and could not be separated from unreacted 7Br by column
chromatography. Compound 18, isolated in 25% yield, was
(34) Kotha, S.; Kashinath, D.; Lahiri, K.; Sunoj, R. B. Eur. J. Org. Chem.
2004, 4003.
534 J. Org. Chem. Vol. 74, No. 2, 2009

Meta-Substituted Thienyl Benzenes
SCHEME 5. Kumada Syntheses of Compounds 10-12 and 15-17
SCHEME 6. Palladium-Catalyzed Reaction between Methyl Grignard and 7Br
thought to be the product of Kumada coupling between
compound 7Br and its corresponding Grignard reagent. An
attempt to make 18 in a rational method from 2 equiv of 7Br
and 1 equiv of Mg turnings failed in the absence of added
MeMgBr. This implied that formation of a Grignard reagent
from 7Br was not favorable, but that transmetalation from
MeMgBrwasfacile.OtherKumadareactionswereuncomplicated.
Compound 13 was made in moderate yield (68%) using a
Sonogashira coupling reaction between 7Br and phenylacetylene.
A slightly higher catalyst loading was required in this reaction
(5 mol % Pd(PPh3)2Cl2 and 10 mol % CuI) than in the Kumada
reactions.
There is much current interest in incorporating redox-stable
metallic components into organic conducting polymers for the
purpose of generating materials with interesting structural and
electrical properties,35 and also for memory applications.36
Therefore, we set about making a thiophenyl monomer that
included ferrocene. Compound 14 was successfully made albeit
in low yield via the Stille cross-coupling of tributylstannylfer-
rocene and 7Br using PdCl2(dppf) as catalyst.
Solid-State Characterization of 4 and 12. ORTEP repre-
sentations of the molecular structures and packing diagrams of
FIGURE 1. ORTEP representation (20% ellipsoids, H-atoms removed
compounds 4 and 12 are given in Figures 1-4. The crystal
for clarity) of the molecular structure of one of the crystallographically
independent forms of 4 with the central phenyl ring in the plane of the
page (a) and nearly perpendicular to the page (b). Carbon atoms are
shown in gray and sulfur in yellow.
structure of 3 has been reported previously.37
Compound 4 crystallized in the chiral space group P212121
(No. 19) and had eight molecules per unit cell; these fell into
two crystallographically independent groups of four. One of the
two configurations (absent from Figure 1) showed disorder in
one of its three terminal thienyl rings. Although the relative
orientation of the thienyl rings in each of the bithienyl
substituents was head-to-tail, as expected, the molecule did not
adopt pseudo-C3 symmetry in the solid state. Instead, it assumed
an asymmetric “scorpion-like” configuration, shown most clearly
in Figure 1a, with one of the substituents serving as the “tail”
and the other two as the “claws”. As shown in Figure 1b, 4
was discoidal, but not planar in the solid state, although two of
the bithienyl substituents featured essentially coplanar thienyl
rings; the dihedral angles between the rings in these substituents
were approximately 1.6 and 12.5 °, while that between the rings
in the remaining substituent was 32.1 °. The “inner” thienyl
rings made torsion angles with the central phenyl ring of about
24.1, 26.5, and 14.3 °. Bond distances and angles were generally
unremarkable.
(35) Pannell, K. H.; Wang, F.; Sharma, H. K.; Cervantes-Lee, F. Polyhedron
2000, 291.
(36) Choi, T.-L.; Lee, K.-H.; Joo, W.-J.; Lee, S.; Lee, T.-W.; Chae, M. Y.
J. Am. Chem. Soc. 2007, 129, 9842.
(37) Guzei, I. A.; Chang, H.-Y.; Choi, M.-G. Z. Kristallogr.-New Cryst. Struct.
2000, 215, 491.
J. Org. Chem. Vol. 74, No. 2, 2009 535

Cornacchio et al.
FIGURE 2. Packing view of the solid state structure of 4 viewed along the b-axis for two contiguous unit cells in (a) capped-stick and (b) space-
filling representations with the thienyl ring disorder and H-atoms removed for clarity. The unit cell is indicated in (b) with the a-axis is in red and
the c-axis in blue. Carbon atoms are shown in gray and sulfur in yellow.
FIGURE 4. Packing view of the solid state structure of 12 viewed
along the b-axis for two contiguous unit cells in (a) capped-stick and
(b) space-filling representations with the thienyl ring disorder and
H-atoms removed for clarity. The unit cell is indicated in (b) with the
a-axis is in red, the c-axis in blue. Carbon atoms are shown in gray,
sulfur in yellow, and fluorine in green.
FIGURE 3. ORTEP representation (20% ellipsoids) of the molecular
structure of 12 with the central phenyl ring in the plane of the page (a)
Physical Characterization by Electronic Spectroscopy,
and nearly perpendicular to the page (b). Disorder in the thienyl rings
Cyclic Voltammetry, and DFT Calculations. Pertinent physi-
and H-atoms have been removed for clarity. Carbon atoms are shown
cal data for compounds 1-4 and 9-17 are collected in Table
in gray, sulfur in yellow, and fluorine in green.
1. Gas-phase first ionization potentials (IP), molecular
orbitals, and HOMO-LUMO gaps were calculated by density
Compound 4 packed (Figure 2) as lamellae of canted discs,
functional theory (DFT) using the B3PW91 functional38 and
with the interlamella distance, as defined by the separation
the 6-31G* basis set as described in the Experimental
between equivalent phenyl carbon atoms, being 4.84 Å.
Methods section. B3PW91 was chosen because it is one of
In the solid-state structure of 12 (Figure 3), both thienyl rings
were disordered; these were modeled as 0.81:0.19 occupancies
of positions corresponding to approximate 180 ° rotations about
the thienyl-phenyl bonds. Once again, the compound was not
the best standard functionals for calculating ionization
potentials.39
To determine the nature of the chromophore in these
compounds, we compared their electronic spectra to those
planar: the torsion angles between the thienyl and phenyl rings
of related species containing phenyl and thienyl rings. As
were approximately 23.7 and 31.4 °. The torsion angle between
the two phenyl rings was about 38.2 °. Figure 4 shows a view
was the case for the known compounds, all oligothiophenes
of the unit cell along the b-axis in (a) stick and (b) spacefilling
representations. Like 4, 12 was packed in layers, with the
(38) Becke, A. D. J. Chem. Phys. 1993, 98, 5638.
(39) Staroverov, V. N.; Scuseria, G. E.; Tao, J.; Perdew, J. P. J. Chem. Phys.
distance between layers being 5.07 Å.
2003, 119, 12129; 2004, 121, 11507(E).
536 J. Org. Chem. Vol. 74, No. 2, 2009

Meta-Substituted Thienyl Benzenes
TABLE 1. Collected Calculated and Measured Properties of Meta Di- and Trisubstituted Thienylbenzenes
a Gas-phase values calculated at the B3PW91/6-31G* level.
FIGURE 5. Representations of the highest occupied (HOMO) and lowest unoccupied (LUMO) molecular orbitals of 1 and 3.
reported in this article, except 13, exhibited broad, featureless
absorption bands, which were assigned to π-π* transitions.
The energies of the absorption maxima were inversely
proportional to conjugation length, as expected, owing to
enhanced spatial distribution of the singlet excited-state in
more highly conjugated systems. Thus, absorption maxima
fell cleanly into two ranges based on the number of
contiguous thienyl rings: those of the “monothienyl” com-
pounds (1, 3, 9-14) ranged from 284-292 nm and were
significantly red-shifted with respect to thiophene (T, 231
nm),40 but fell within the span marked by 2-phenylthiophene
(PT, 282 nm) and 2,2′-bithiophene (BT, 303 nm),40 while
those of the “bithienyl” compounds (2, 4, 15-17) ranged
from 351-357 nm and were instead closest to that of 2,2′:
5′,2′′-terthiophene (TT, 354 nm).40 The same analysis could
be applied to the molar extinction coefficients (ε) that, with
J. Org. Chem. Vol. 74, No. 2, 2009 537

Cornacchio et al.
FIGURE 6. Representations of the highest occupied (HOMO) and lowest unoccupied (LUMO) molecular orbitals of 2 and 4.
the exception of 11 and 13 (vide infra), were significantly
larger for the bithienyl (5.37-6.92 × 104 M-1 cm-1) than
for the monothienyl compounds (2.88-5.13 × 104 M-1
cm-1).
and “tails” of these species were electronically distinct and,
therefore, that the nature of the third substituent in all cases
had little effect on optical properties.
Empirical studies by Sato et al.27 have shown that the
approximate conjugation length as represented by the number
of contiguous double bonds (m) in a series of ortho-, meta-,
and para-bis(bithienyl) benzenes may be estimated from the
absorption energy E (in eV) using eq 1
The emission maxima of 1, 3, and 9-14 (361-381 nm)
approximated that of BT (362 nm),40 while those of 2, 4, and
15-17 (408-423 nm) were similar to those of 5-phenyl-2,2′-
bithiophene (PBT, 413 nm)27 and TT (426 nm).40 Unlike the
absorption spectra, the rt solution-phase emission spectra of all
compounds, and those of related compounds in the literature,27
exhibited structured bands. The difference has been attributed
to the transition from a nonrigid, nonplanar ground-state to a
more rigid, planar excited state.27 This is readily rationalized
by reference to calculated representations of the HOMO and
LUMO of compounds 1 and 3, and 2 and 4, which are shown
in Figures 5 and 6, respectively. Clearly evident from these
diagrams is that the HOMO and LUMO have aromatic and
quinoid character, respectively. Thus, population of the LUMO
by excitation results in the development of partial double bond
character, and therefore increased rigidity and planarity, between
the thienyl and phenyl rings. A similar analysis of 2 by
semiempirical methods has been reported.27
E ) (7.32 ⁄ m) + 2.26
(1)
Table 1 includes a column giving m for each of the
compounds. The electronic data and derived approximate
conjugation lengths indicated some electronic coupling
between the thienyl and phenyl rings in all of the compounds,
but also showed that the coupling did not extend from one
thienyl ring to another through the central phenyl ring in this
family of compounds. These empirical findings essentially
confirmed the detailed ab initio calculations of Ko¨se and co-
workers,7 which showed that the thienyl “arms” of related
dendrimers were predominantly responsible for absorption,
and that meta substitution about the phenyl ring localized
excitons to the arms without any electronic coupling between
them. A qualitatively similar result would be derived simply
by drawing charge-separated quinoidal resonance forms for
the compounds, which would demonstrate “on the back of
the envelope” that naive arrow pushing should not permit
delocalization between thienyl rings in different meta posi-
tions (Chart 4).
For 1 and 2, both MOs were distributed over the entire
molecule, but there was a nodal plane that perpendicularly
bisected the phenyl ring. In 3 and 4, the HOMO did not
encompass one of the three substituents as extensively as it did
the other two, and the LUMO was strictly confined to only two
of them.
A hypothesis consistent with these data was that the optical
spectra should depend primarily on the number of contiguous
thienyl rings, and not on the degree of substitution around
the central phenyl ring. Given that we had developed methods
to make mono-, bi- and trithienyl benzenes selectively, we
were able to test this directly. Figure 9 correlates the
excitation and emission maxima for monothienyl (1, 3, 5Br,
7Br, PT), and bithienyl (2, 4, 6Br, 8Br, PBT) compounds with
the degree of meta-substitution about the central phenyl ring.
The HOMO and LUMO of analogous mono- (10-12) and
bithienyl (15-17) compounds bearing phenyl, para-methox-
yphenyl, and para-fluorophenyl substituents are shown in
Figures 7 and 8, respectively. With the exception of 11, these
compounds shared characteristic orbital features with 1 and 2,
respectively, meaning that both the “arms” as well as the “heads”
(40) Becker, R. S.; Seixas de Melo, J.; Mac¸anita, A. L.; Elisei, F. J. Phys.
Chem. 1996, 100, 18683.
538 J. Org. Chem. Vol. 74, No. 2, 2009

Meta-Substituted Thienyl Benzenes
FIGURE 7. Representations of the highest occupied (HOMO) and lowest unoccupied (LUMO) molecular orbitals of 10-12.
Clearly evident from this representation of the data is that
there was only a slight variation in the optical properties with
the degree of substitution about the ring. It was also
interesting to note that the brominated compounds 5Br-8Br
had absorbance (287-354 nm) and emission (354-408 nm)
spectra that were very similar to those of their bromine-free
analogues, 1-4 (284-357 and 361-423, respectively), which
emphasized the insensitivity of the electronic spectroscopy
not only to the degree, but also to the nature of the meta
substitution in this family of thienyl benzenes.
Only the emission maxima of the monothienyl compounds
1, 3, 5Br, 7Br, and PT demonstrated a smooth, albeit weak,
dependence on the degree of substitution. Therefore, it was
possible that a minimum of two contiguous thienyl rings was
required for absorption by, and strict localization of the
exciton on, electronically independent “arms” in meta-
substituted thienyl benzenes. When arms were restricted to
only a single thienyl ring, weak electronic coupling between
them became apparent.
For all compounds, the calculated gas-phase first ionization
potentials (IP) were significantly lower than those of
thiophene (8.55 eV calculated; 8.86 eV observed41) and
benzene (8.88 eV calculated; 9.24 eV observed41), which was
expected on the basis of extended conjugation. As for the
electronic spectra, the calculated IPs fell cleanly into two
ranges based on the number of contiguous thienyl rings: for
the monothienyl compounds (1, 3, and 10-12), they ranged
from 6.87-7.12 eV, while for the bithienyl compounds (2,
4, and 15-17), IP were significantly lower and spanned a
(41) Lide, D. R. CRC Handbook of Chemistry and Physics, 88th ed.; CRC
Press: Boca Raton, FL, 2008.
J. Org. Chem. Vol. 74, No. 2, 2009 539

Cornacchio et al.
FIGURE 8. Representations of the highest occupied (HOMO) and lowest unoccupied (LUMO) molecular orbitals of 15-17.
CHART 4. Quinoidal Resonance Forms for 4 Showing Localization to Only One of the Bithienyl “Arms”
mere 0.09 eV (6.33-6.42 eV). The HOMO-LUMO gaps
for the monothienyl compounds showed only a very slight
dependence on the nature of the third meta-substituent and
fell in the narrow range 4.53-4.58 eV. The corresponding
gaps of the bithienyl compounds were essentially independent
of the substituent (3.70-3.72 eV) but were significantly lower
than those of the monothienyl compounds by about 0.8 eV
because of extended conjugation in these systems.
A pair of compounds had remarkable properties: (i) Substan-
tial contribution to the HOMO by the OMe group in 11 (Figure
540 J. Org. Chem. Vol. 74, No. 2, 2009

Meta-Substituted Thienyl Benzenes
forms of 1-4 and 9-17 are not free solvated cations (see Table
S5 in the SI). There also appeared not to be a rational Hammett-
type connection between either the electronic nature of the
substituent in the third meta-position or between the degree of
meta-substitution and the measured oxidation potential. Instead,
the number of contiguous thienyl rings was once again the prime
determinant. Further work is required to rationalize the observed
oxidation potentials.
Conclusions
Kumada coupling reactions catalyzed by PdCl2(dppf) between
the appropriate thienyl Grignard and either 1,3- or 1,3,5-bromo-
or iodobenzenes are the most reliable of the typical metal-
catalyzed approaches for the production of meta-substituted
thienylbenzenes. These reactions are generally higher yielding
than the catalyzed routes previously reported in the literature,
and are more convenient because they circumvent the need for
prelithiation. In addition, they are more flexible in that they also
permit the controlled generation of mixed (halo)(thienyl)ben-
zenes, which are valuable compounds possessing both electrical
and chemical “handles” and which cannot be accessed by
cyclotrimerization reactions.
FIGURE 9. Variation of absorption (0) and emission (]) maxima with
degree of meta-substitution about a central phenyl ring for compounds
incorporating thienyl groups with one (n ) 1; 1, 3, 5Br, 7Br, and PT)
and two (n ) 2; 2, 4, 6Br, 8Br, and PBT) contiguous rings. Absorption
data for PBT and emission data for PT were not available. Lines are
meant only to guide the eye.
UV-visible absorption and fluorescence data show clearly
that the electronic character of these compounds is determined
primarily by the number of contiguous thienyl rings, not by the
degree of meta-substitution about the central phenyl ring nor
by the nature of the third substituent in the case of compounds
9-12 and 15-17. This finding essentially confirms ab initio
calculations of ref 7, which show that the thienyl “arms” of
phenyl-cored dendrimers are predominantly responsible for
absorption and that meta-substitution about the phenyl ring
localizes excitons to the arms without any electronic coupling
between them. However, a smooth increase in the emission
maximum of the “monothienyl” compounds with increasing
meta-substitution about the central phenyl ring demonstrates that
a minimum of two contiguous rings is necessary for strict
localization of the excitions to the arms.
Within the field of conducting polymers, meta-substitution
is generally avoided because it blocks conjugation. In some
cases, it has been introduced into poly(phenylene)s specifically
to interrupt conjugation in a rational manner.42,43 Our findings
for these model systems (or monomers) are in accord with this
sentiment. However, we note that their oxidation potentials can
be varied (albeit unpredictably, rather than by rational “tuning”)
over a large and useful range by introduction of different groups
meta to the thienyl substituent. In addition, the steric qualities
of the groups may give rise to enhanced physical properties of
7) had the effect of lowering the IP and HOMO-LUMO gap
of this compound and raising its measured molar extinction
coefficient with respect to other members of the monothienyl
family. (ii) The C≡C “bridge” allowed significant electronic
interaction between the head and tail of 13 (Figure 10) and gave
it a lower HOMO-LUMO gap, a more complex absorption
spectrum, a bathochromically shifted emission, and a much
larger molar extinction coefficient than either the biphenyl-
(10-12) or phenyl-based compounds (1 and 3). In 13, the
HOMO essentially encompassed the diphenylacetlyene “core”
of the molecule, while the HOMO-1 orbital (-5.9 kJ mol-1
)
had essentially the same character as the HOMO in the other
monothienyl compounds.
Table 1 also lists the solution oxidation potentials, ap-
proximated by anodic peak potentials, for compounds 1-4 and
9-17 as measured by cyclic voltammetry. Values for the
monothienyl species ranged from 1.13-1.60 V, while those of
the bithienyl compounds were significantly less positive and
spanned a narrower range (0.90-1.27 V); these values were in
accord with those of related compounds reported in the
literature.5,16-18 We have calculated the Gibbs free energies of
oxidation for 1-4 and 9-17 in CH2Cl2 solution, but found that
these values (expressed in eV) were much greater than the
measured oxidation potentials, apparently because the oxidized
FIGURE 10. The HOMO, HOMO-1, and LUMO of compound 13.
J. Org. Chem. Vol. 74, No. 2, 2009 541

Cornacchio et al.
resultant polymers. Finally, compounds 3 and 45 lead in principle
to highly cross-linked polymers that should have interesting
properties. These things in combination make polymers of the
compounds reported in this paper worthwhile targets that we
are currently pursuing.
1,3,5-Tris(2′-thienyl)benzene (3). This compound and 1 were
made similarly. Thus, the reaction of a THF (100 mL) solution of
1,3,5-triiodobenzene (1.34 g, 2.94 mmol) and solid PdCl2(dppf)
(0.064 g, 0.088 mmol) with a suspension of 2-thienylmagnesium
bromide (0.117 M, 150 mL, 18 mmol) in the same solvent gave,
after workup and purification by flash column chromatography (20:1
hexanes/CHCl3, Rf ) 0.3), the desired product as a white crystalline
solid: mp (lit.) 160-162 °C (156-158 °C). Yield: 0.820 g (86%).
HRMS C18H12S3 calcd (found): 324.0101 (324.0112). The spec-
troscopic data for this compound were the same as those reported
in the literature.28
Experimental Section
Calculations. All calculations of the first ionization potentials
(IP) and HOMO-LUMO gaps were carried out using the Gaussian
03 program.44 Each IP was evaluated as the difference between
the zero-point-energy (ZPE)-corrected total energies of the molec-
ular cation and the corresponding neutral species. The geometries
of the neutral molecules and molecular cations were optimized at
the B3PW91/6-31G* level without symmetry constraints. The zero-
point energies were also evaluated at the B3PW91/6-31G* level
using a scale factor of 0.9573 recommended by Scott and Radom.45
Compounds 1-4 and 9-17 had multiple local minima (confor-
mations), but because the IP and HOMO-LUMO gaps were much
larger than energy differences between the various conformations,
it was not necessary to use the most stable conformations to evaluate
an IP. The optimized Cartesian coordinates of the conformations
used in this work are given in the Supporting Information. Harmonic
vibrational analysis of the conformers showed that none of the 13
neutral species and 13 cations had imaginary frequencies, and
therefore, these structures represent true minima on the potential
energy surfaces.
1,3,5-Tris{5′-(2′,2′′-bithienyl)}benzene (4). By Stille coupling.
This was similar to the synthesis of 2. Thus, the reaction of 1,3,5-
triiodobenzene (0.184 g, 0.402 mmol) and 5′-tributylstannyl-2,2′-
bithiophene (1.10 g, 2.42 mmol) in refluxing toluene (40 mL)
catalyzed by PdCl2(dppf) (0.062 g, 0.048 mmol) gave after workup
and purification using flash column chromatography (neat hexanes,
Rf ) 0.1) the desired product as a yellow powder. Yield: 0.92 g
(18%). By Suzuki coupling. In a 250 mL three-neck flask equipped
with a condenser, Pd(OAc)2 (0.014 g, 0.063 mmol), PPh3 (0.050
g, 0.190 mmol), and an aqueous solution of Na2CO3 (1 M, 3.6 mL)
were added to a pale yellow solution of 2,2′-bithienyl-5′-boronic
acid (0.381 g, 1.81 mmol) and 1,3,5-triiodobenzene (0.207 g, 0.453
mmol) in THF (100 mL). The resulting green solution was brought
to reflux for 3 d. The brown solution was removed from heat, cooled
under a slow flow of N2 and quenched on ice with H2O (100 mL).
Dichloromethane (50 mL) was added and the aqueous layer was
extracted with the same solvent (3 × 50 mL). The organic layers
were combined and dried over MgSO4 and the solvent was
evaporated to give a light brown solid from which the product was
isolated by chromatography as above. Yield: 0.234 g (90%). Mp
(lit.) 210-214 °C (136 °C). HRMS C30H18S6 calcd (found):
569.9733 (569.9740). 13C{1H} NMR: δ 122.1, 124.1, 124.8, 124.8,
128.2, 135.7, 137.5, 137.7, 142.2. 1H NMR spectroscopic data for
this compound were the same as those reported in the literature.28
Crystals of this compound suitable for X-ray diffraction analysis
were grown at 4 °C from CH2Cl2/hexane over a period of 2 weeks.
1,3-Dibromo-5-(2′-thienyl)benzene (5Br). This compound, a
white crystalline solid (mp 56-59 °C), was isolated using column
chromatography (neat hexanes, Rf ) 0.6) as a byproduct of the
synthesis of 7Br (below). Yield: 0.38 g (10%). HRMS C10H6Br2S
calcd (found): 315.8557 (315.8567). 1H NMR: δ 7.09 (dd, 1H, 3JHH
) 3.8, 3JHH ) 5.0), 7.31 (pdd, 1H), 7.35 (pdd, 1H), 7.56 (pt, 1H),
7.67 (pd, 2H). 13C{1H} NMR: δ 123.6, 124.9, 126.7, 127.7, 128.5,
132.8, 138.0, 141.2. This compound has been prepared previously
by the Suzuki reaction of 2-thienylboronic acid and 1,3,5-tribro-
mobenzene.46
The B3PW91/6-31G* ionization potentials of benzene (calcu-
lated, 8.88 eV; experimental, 9.24 eV41) and thiophene (calculated,
8.55 eV; experimental, 8.86 eV41) were about 0.3 eV lower than
the experimental values. This suggests that the calculated IPs of
compounds 1-4 and 9-17 may also be underestimated.
1,3-bis(2′-thienyl)benzene (1). In a 500 mL three-neck flask
equipped with a bubbler, an orange suspension of 1,3-diiodobenzene
(1.50 g, 4.55 mmol) and PdCl2(dppf) (0.099 g, 0.137 mmol) in
THF (15 mL) was added dropwise to an ice cold suspension of
2-thienylmagnesium bromide (0.17 M, 160 mL, 27 mmol) in the
same solvent to give a yellow slurry. This mixture was stirred
magnetically at r.t. for 2 d before being hydrolyzed on ice with 1
M HCl (200 mL), followed by H2O (100 mL). The aqueous layer
was extracted with Et2O (3 × 100 mL). The combined organic
fractions were washed with saturated NaHCO3 (50 mL), followed
by H2O (50 mL) before being dried over MgSO4 and reduced to a
dark brown liquid by rotatory evaporation. The desired product, a
white microcrystalline solid, was obtained by flash column chro-
matography using petroleum ether as eluent (Rf ) 0.4): mp (lit.)
83-85 °C (85-86 °C). Yield: 0.880 g (73%). HRMS C14H10S2
calcd (found): 242.0224 (242.0223). 13C{1H} NMR: δ 123.7, 123.8,
125.4, 128.3, 129.7, 135.3, 144.2 (one quaternary carbon atom not
observed). 1H NMR data for this compound were the same as those
reported in the literature.28
1,3-Bis{5′-(2′,2′′-bithienyl)}benzene (2). In a 250 mL three-neck
flask equipped with a condenser, a yellow solution of 5′-tributyl-
stannyl-2,2′-bithiophene (2.08 g, 4.58 mmol) in toluene (5 mL) was
added to an orange suspension containing 1,3-diiodobenzene (0.504
g, 1.53 mmol) and PdCl2(dppf) (0.100 g, 0.777 mmol) in the same
solvent (50 mL). This gave a pale brown suspension, which was
brought to reflux for 3 d. The dark brown reaction mixture was
then cooled under a slow flow of Ar and hydrolyzed with H2O (50
mL). The aqueous layer was extracted with Et2O (3 × 50 mL).
The combined organic fractions were washed with an aqueous
solution of KF (3 × 50 mL), dried over MgSO4, and evaporated to
dryness to give a yellow solid. The desired product, a yellow
powder, was obtained by flash column chromatography using
petroleum ether as eluent (Rf ) 0.2): mp 168-172 °C. Yield: 0.528
g (71%). HRMS C22H14S4 calcd (found): 405.9978 (405.9968). The
NMR spectroscopic data for this compound were the same as those
reported in the literature.9
1,3-Dibromo-5-{5′-(2′,2′′-bithienyl)}benzene (6Br). This com-
pound, a pale yellow solid (mp 139-143 °C), was isolated using
column chromatography (neat hexanes, Rf ) 0.4), as a byproduct
of the synthesis of 8Br (below). Yield: 0.038 g (4%). HRMS
C10H6Br2S2 calcd (found): 397.8434 (397.8433). 13C{1H} NMR: δ
123.6, 124.4, 124.9, 125.2, 125.6, 127.3, 128.2, 132.8, 137.6 (4
quaternary carbon atoms not observed). 1H NMR spectroscopic data
for this compound were the same as those reported in the patent
literature.29
1-Bromo-3,5-bis(2′-thienyl)benzene (7Br). This compound and
1 were made similarly, with the modification that the reaction
mixture was heated to 60 °C overnight. Thus, a reaction between
1,3,5-tribromobenzene (3.90 g, 12 mmol), PdCl2(dppf) (0.09 g, 0.13
mmol), and 2-thienylmagnesium bromide (0.08 M, 330 mL, 26
(42) Reddinger, J. L.; Reynolds, J. R. Macromolecules 1997, 30, 479.
(43) Liao, L.; Pang, Y.; Ding, L.; Karasz, F. E. Macromolecules 2001, 34,
7300.
(44) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B. et al.; Gaussian 03, ReVision
c.03; Gaussian, Inc.: Wallingford, CT, 2004.
(45) Scott, A. P.; Radom, L. J. J. Phys. Chem. 1996, 100, 16502.
(46) Xia, H.; Yang, X.-Y.; Shan, Z.-X.; Zhao, D.-J. Hecheng Huaxue 2000,
8, 335.
542 J. Org. Chem. Vol. 74, No. 2, 2009

Meta-Substituted Thienyl Benzenes
mmol) in dry THF (100 mL) gave, following workup and
chromatography (neat hexanes, Rf ) 0.2), of the desired product
as a white solid (mp 73-74 °C). Yield 1.85 g (46%). HRMS
125.6, 128.4, 129.1, 135.9, 136.9, 141.9, 144.0, 161.7, 164.2.
Colorless crystals suitable for analysis by X-ray diffraction were
grown from a concentrated CH2Cl2 solution by slow diffusion of
hexanes.
1
C14H9BrS2 calcd (found): 319.9329 (319.9332). H NMR: δ 7.11
(dd, 2H, 3JHH ) 3.6, 3JHH ) 4.8), 7.35 (m, 4H), 7.66 (pd, 2H), 7.73
(pt, 1H). 13C{1H} NMR: δ 122.4, 123.6, 124.5, 126.1, 127.8, 128.4,
137.0, 142.5.
1-(Phenylacetylenyl)-3,5-bis(2′′-thienyl)benzene (13). In a 50
mL three-neck flask equipped with a condenser and bubbler, 7Br
(0.253 g, 0.786 mmol) was dissolved in a 1:1 mixture of of THF/
Et3N (10 mL). The catalyst mixture, Pd(PPh3)2Cl2 (0.028 g, 0.039
mmol) and CuI (0.015 g, 0.008 mmol), was added to this solution
to give an orange suspension. Phenylacetylene (0.120 g, 1.18 mmol)
was added dropwise over 5 min to give a dark brown mixture. Once
the addition was complete, the mixture was heated to 60 °C for
24 h. The dark brown suspension was cooled, CH2Cl2 was added,
and the mixture was filtered through Celite. The filtrate was reduced
to give a black viscous liquid. The product, a white solid (mp
118-120 °C), was isolated by flash column chromatography (neat
1-Bromo-3,5-bis{5′-(2′,2′′-bithienyl)}benzene (8Br). This com-
pound and 2 were made similarly. Thus, reaction of 5′-tributyl-
stannyl-2,2′-bithiophene (3.26 g, 7.17 mmol) and 1,3,5-tribro-
mobenzene (1.12 g, 3.58 mmol) in toluene (75 mL) catalyzed by
PdCl2(dppf) (0.079 g, 0.11 mmol) gave after workup and purifica-
tion by flash column chromatography (neat hexanes, Rf ) 0.1) the
desired product as a yellow solid (mp 175-179 °C). Yield: 0.302
g (27%). HRMS C22H13BrS4 calcd (found): 483.9083 (483.9082).
3
1H NMR: δ 7.05 (dd, 2H, JHH ) 3.6,3JHH ) 5.2), 7.18 (d, 2H,
3
3JHH ) 3.8), 7.24 (m, 4H), 7.29 (d, 2H, JHH ) 3.8), 7.64, (pd,
hexanes, Rf ) 0.3). Yield: 0.227 g (68%). HRMS C22H14S2 calcd
1
(found): 426.0199 (426.0189). H NMR: δ 7.13 (dd, 2H,3JHH
)
2H), 7.69 (pt, 1H). 13C{1H} NMR: δ 101.5, 110.0, 124.3, 124.8,
125.1, 125.2, 127.4, 128.2, 132.4 (3 quaternary carbon atoms not
observed).
3.6, 3JHH ) 5.0), 7.35 (dd, 2H, 3JHH ) 5.0), 7.41 (m, 5H), 7.63 (m,
2H), 7.70 (pd, 2H), 7.80 (pt, 1H). 13C{1H} NMR: δ 88.9, 90.3,
123.2, 123.5, 124.1, 124.8, 125.7, 128.1, 128.4, 128.6, 128.8, 132.0,
135.5, 143.3.
1-Methyl-3,5-bis(2′-thienyl)benzene (9). A solution of meth-
ylmagnesium bromide (3.0 M, 0.62 mL, 1.9 mmol) in Et2O and
solid PdCl2(dppf) (0.027 g, 0.037 mmol) were added to an ice cold
solution of 7Br in the same solvent (100 mL) to give an orange
solution. After heating to 60 °C overnight, standard aqueous workup
and purification by flash column chromatography (neat hexanes,
Rf ) 0.2) gave the desired product as a white solid (mp 55-57
°C). Yield: 0.0824 g (52%). HRMS C15H12S2 calcd (found):
256.0380 (256.0376). 1H NMR: δ 2.44 (s, 3H), 7.11 (dd, 2H, 3JHH
) 3.6, 3JHH ) 5.2), 7.31 (d, 2H, 3JHH ) 5.2), 7.37 (m, 4H), 7.67 (s,
1H). 13C{1H} NMR: δ 21.7, 121.1, 123.61, 125.2, 126.2, 128.2,
135.2, 139.4, 144.3.
1-Ferrocenyl-3,5-bis(2′-thienyl)benzene (14). In a 100 mL
three-neck flask, a dark red solution of tributylstannylferrocene
(0.612 g, 1.3 mmol) in THF (2 mL) was added to a colorless
solution of 7Br (0.278 g, 0.865 mmol) in THF (30 mL) to give an
orange mixture. Following the addition of PdCl2(dppf) (0.018 g,
0.026 mmol), the mixture was brought to reflux for 2 d. The
resulting dark brown solution was cooled under a stream of Ar.
The reaction was quenched with H2O (10 mL), solid KF (2 equiv)
was added, and the mixture was stirred for 2 h. Ether (15 mL) was
then added and the aqueous layer was extracted with the same
solvent (2 × 15 mL). The combined organic fractions were washed
with H2O (10 mL), dried over MgSO4, and reduced to give a dark
red liquid. The crude product was purified by flash column
chromatography (20:1 hexanes/CH2Cl2) to obtain the desired
product as a dark orange liquid. Yield: 0.033 g (8%). HRMS
C24H18FeS2 calcd (found): 426.0199 (426.0189). 1H NMR: δ 4.10
(s, 5H), 4.37 (pt, 2H), 4.72 (pt, 2H), 7.14 (dd, 2H, 3JHH ) 3.6, 3JHH
1-Phenyl-3,5-bis(2′-thienyl)benzene (10). This compound and
9 were made similarly. Thus, reaction of a solution of phenylmag-
nesium bromide in Et2O (3.0 M, 0.78 mL, 2.3 mmol), PdCl2(dppf)
(0.034 g, 0.047 mmol), and 7Br (0.500 g, 1.55 mmol) in THF (100
mL) gave the desired product as a white solid (mp 133-135 °C),
following workup and purification by flash column chromatography
(neat hexanes, Rf ) 0.1). Yield: 0.334 g (68%). HRMS C20H14S2
calcd (found): 318.0537 (318.0536). 1H NMR: δ 7.13 (dd, 2H, 3JHH
) 3.6, 3JHH ) 4.8), 7.34 (dd, 2H, 3JHH ) 4.8, 4JHH ) 1.0), 7.43 (m,
3
4
3
) 5.2), 7.34 (dd, 2H, JHH ) 5.2, JHH ) 1.2), 7.41 (dd, 2H, JHH
) 3.6, 4JHH ) 1.2), 7.63 (pd, 2H), 7.67 (pt, 1H). 13C{1H} NMR: δ
67.1, 69.4, 70.0, 85.0, 121.7, 123.3, 123.7, 125.3, 128.3, 135.3,
141.2, 144.3 (two quaternary carbon atoms not observed).
3
3
4
3H), 7.50 (t, 2H, JHH ) 7.6), 7.67 (dd, 2H, JHH ) 7.6, JHH
)
1.6), 7.74 (pd, 2H), 7.83 (pt, 1H). 13C{1H} NMR: δ 122.8, 123.9,
124.36, 125.5, 127.5, 128.0, 128.3, 129.1, 135.8, 140.9, 142.9,
144.1.
1-Phenyl-3,5-bis{5′-(2′,2′′-bithienyl)}benzene (15). This com-
pound and 10 were made similarly. Thus, reaction of phenylmag-
nesium bromide in Et2O (3.0 M, 0.173 mL, 0.519 mmol),
PdCl2(dppf) (0.008 g, 0.010 mmol) and 8Br (0.100 g, 0.346 mmol)
in THF (20 mL) gave, after workup and purification by flash column
chromatography (neat hexanes, Rf ) 0.1), the desired product as a
pale yellow solid (mp. 134-138 °C). Yield: 0.109 g (68%). HRMS
C28H18S4 calcd (found): 482.0291 (482.0285). 1H NMR: δ 7.05 (dd,
1-(4′-Methoxyphenyl)-3,5-bis(2′′-thienyl)benzene (11). This
compound and 9 were made similarly. Thus, reaction between 7Br
(0.400 g, 1.24 mmol), 4-methoxyphenylmagnesium bromide in THF
(0.5 M, 3.72 mL, 1.87 mmol) and solid PdCl2(dppf) (0.027 g, 0.037
mmol) in THF (100 mL) gave, following workup and purification
by flash column chromatography (5:2 hexanes/CH2Cl2, Rf ) 0.4),
the desired product as a white solid (mp 121-123 °C). Yield: 0.225
3
3
3
2H, JHH ) 5.2, JHH ) 3.6), 7.20 (d, 2H, JHH ) 3.6), 7.24 (m,
4H), 7.35 (d, 2H, 3JHH ) 3.6), 7.42 (t, 1H, 3JHH ) 7.2), 7.50 (t, 2H,
3JHH ) 7.2), 7.67 (d, 2H, 3JHH ) 7.2), 7.71 (pd, 2H), 7.79 (pt, 1H).
13C{1H} NMR: δ 121.9, 124.0, 124.1, 124.6, 124.7, 124.8, 127.5,
128.1, 128.2, 129.1, 135.5, 137.4, 137.5, 140.7, 142.6, 143.1.
g (52%). HRMS C21H16OS2 calcd (found): 348.0643 (348.0644).
3
1H NMR: δ 3.88 (s, 3H, H), 7.03 (m, 2H), 7.13 (dd, 2H, JHH
)
5.2,3JHH ) 3.6), 7.34 (dd, 2H, JHH ) 1.2, JHH ) 5.2), 7.42 (dd,
2H, 3JHH ) 1.2,3JHH ) 3.6), 7.61 (m, 2H), 7.69 (pd, 2H), 7.78 (pt,
1H). 13C{1H} NMR: δ 55.6, 114.5, 122.2, 123.8, 124.0, 125.4,
128.3, 128.6, 133.3, 135.7, 142.5, 144.2, 159.8.
3
3
1-(4′-Methoxyphenyl)-3,5-bis{5′′-(2′′,2′′′-bithienyl)}benzene (16).
This compound and 11 were made similarly. Thus, reaction of 8Br
(0.146 g, 0.5 mmol), 4-methoxyphenylmagnesium bromide in THF
(0.5 M, 1.52 mL, 0.758 mmol), and PdCl2(dppf) (0.011 g, 0.015
mmol) in THF (20 mL) gave, following workup and purification
by flash column chromatography (3:1 hexanes/CH2Cl2, Rf ) 0.2),
the desired product as a pale yellow solid (mp 144-147 °C). Yield:
0.090 g (35%). HRMS C29H20OS4 calcd (found): 512.0397
1-(4′-Fluorophenyl)-3,5-bis(2′′-thienyl)benzene (12). This com-
pound and 9 were made similarly. Thus, the reaction of 4-fluo-
rophenylmagnesium bromide in THF (0.5 M, 1.89 mL, 1.87 mmol),
PdCl2(dppf) (0.027 g, 0.037 mmol), and 7Br (0.400 g, 1.24 mmol)
in THF (100 mL) gave, following workup and purification by flash
column chromatography (20:1 hexanes/CH2Cl2, Rf ) 0.2), the
desired product as a white solid (mp 95-98 °C). Yield: 0.350 g
1
(512.0390). H NMR: δ 3.89 (s, 3H), 7.04 (m, 4H), 7.19 (d, 2H,
1
(84%). HRMS C20H13FS2 calcd (found): 336.0443 (336.0437). H
3JHH ) 3.8), 7.24 (m, 4H), 7.33 (d, 2H, 3JHH ) 3.8), 7.61 (m, 2H),
7.67 (pd, 2H), 7.74 (pt, 1H). 13C{1H} NMR: δ 55.6, 110.0, 114.6,
121.4, 123.6, 124.0, 124.5, 124.7, 124.8, 128.1, 128.5, 133.1, 135.4,
137.4, 137.6, 142.6, 142.7.
NMR: δ 7.16 (m, 4H), 7.35 (dd, 2H, 3JHH ) 5.2, 4JHH ) 1.2), 7.43
(dd, 2H, 3JHH ) 3.6, 4JHH ) 1.2), 7.63 (m, 2H), 7.67 (pd, 2H), 7.82
1
(pt, 1H). 13C{1H} NMR: δ 116.0 (d, JCF ) 85.6), 122.7, 124.1,
J. Org. Chem. Vol. 74, No. 2, 2009 543

Cornacchio et al.
1-(4′-Fluorophenyl)-3,5-bis{5′′-(2′′,2′′′-bithienyl)}benzene (17).
This compound and 12 were made similarly. Thus, reaction of 8Br
(0.100 g, 0.346 mmol), 4-fluorophenylmagnesium bromide in THF
(0.5 M, 1.04 mL, 0.519 mmol), and PdCl2(dppf) (0.008 g, 0.010
mmol) in THF (20 mL) gave, following workup and purification
by flash column chromatography (10:1 hexanes/CH2Cl2, Rf ) 0.3),
the desired product as a pale yellow solid (mp 154-157 °C). Yield:
0.123 g (71%). HRMS C28H17FS4 calcd (found): 500.0197 (500.0208).
hexanes, Rf ) 0.1). Yield: 0.112 g (25%). HRMS C28H18S4 calcd
1
3
(found): 482.0291 (482.0290). H NMR: δ 7.15 (dd, 4H, H, JHH
3
3
4
) 5.0, JHH ) 3.6), 7.36 (dd, 4H, JHH ) 5.0, JHH ) 1.0), 7.45
3
4
(dd, 4H, JHH ) 3.6, JHH ) 1.0), 7.77 (pd, 4H), 7.90 (pt, 2H).
13C{1H} NMR: δ 123.2, 124.1, 124.4, 125.7, 128.4, 135.9, 142.4,
143.9.
Acknowledgment. We thank the Natural Sciences and
Engineering Research Council of Canada (NSERC) for a
Discovery Grant to N.D.J. and a Post-Graduate Master’s
Scholarship (PGS-M) to A.L.P.C. We acknowledge also the
contribution of the Province of Ontario for an Ontario Graduate
Scholarship (OGS) to A.L.P.C. and an Early Researcher Award
(ERA) to N.D.J. The assistance of Dr. Zhifeng Ding (UWO) in
the acquisition of CV data is also gratefully recognized.
3
3
1H NMR: δ 7.05 (dd, 2H, JHH ) 5.2, JHH ) 3.6), 7.19 (d, 4H),
3
7.26 (m, 4H), 7.34 (d, 2H, JHH ) 3.6), 7.64 (m, 4H), 7.77, (pt,
1H). 13C{1H} NMR: δ 116.0 (d, 1JCF ) 85.6), 121.9, 123.8, 124.1,
124.7, 124.8, 128.2, 129.1, 129.2, 135.6, 136.8, 137.5, 137.6, 142.1,
142.4.
1,1′,3,3′-Tetra(2′′-thienyl)-5,5′-biphenyl (18). This compound
was isolated as a byproduct of the synthesis of 9. A dedicated
synthesis was as follows: 7Br (0.600 g, 1.86 mmol) was dissolved
in THF (20 mL), and solid Pd(dppf)Cl2 (0.014 g, 0.019 mmol) was
added to give an orange mixture. Magnesium turnings (0.022 g,
0.931 mmol) and a few crystals of I2 were then added. After holding
the mixture at reflux overnight, MeI (100 µL, 0.309 mmol) was
added, which caused the Mg turnings to dissolve within minutes.
The reaction mixture was heated to reflux for 18 h before being
quenched on ice with H2O (10 mL). Workup was the same as
outlined for 1. The desired product, a white solid (mp 148-155
°C), was obtained following flash column chromatography (neat
Supporting Information Available: General experimental
considerations, details and additional results of density functional
calculations, including Cartesian coordinates and energies for
1
1-4 and 9-17, H and 13C NMR spectra, summary table of
crystallographic data for compounds 4 and 12, and full data in
CIF. This material is available free of charge via the Internet at
http://pubs.acs.org.
JO801771D
544 J. Org. Chem. Vol. 74, No. 2, 2009