Thiophene-fused ladder boroles with high antiaromaticity

Article abstract of DOI:10.1021/ja2019977

A series of polycyclic thiophene-fused boroles were synthesized on the basis of stepwise substitution reactions from thienylboronic ester precursors. In these ladder-type π-conjugated systems, the thiophene-fused structure enhances the antiaromaticity of the borole ring. This trend is opposite to the conventional understanding that the arene-fused structure decreases the antiaromaticity of the 4π-electron ring skeletons. The ladder boroles exhibited characteristic properties such as long-wavelength absorptions and low reduction potentials.

Full text of DOI:10.1021/ja2019977

COMMUNICATION
pubs.acs.org/JACS
Thiophene-Fused Ladder Boroles with High Antiaromaticity
Azusa Iida and Shigehiro Yamaguchi*
Department of Chemistry, Graduate School of Science, Nagoya University and JST-CREST, Furo, Chikusa, Nagoya 464-8602, Japan
S Supporting Information
b
ABSTRACT: A series of polycyclic thiophene-fused bor-
oles were synthesized on the basis of stepwise substitution
reactions from thienylboronic ester precursors. In these
ladder-type π-conjugated systems, the thiophene-fused
structure enhances the antiaromaticity of the borole ring.
This trend is opposite to the conventional understanding
that the arene-fused structure decreases the antiaromaticity
of the 4π-electron ring skeletons. The ladder boroles
exhibited characteristic properties such as long-wavelength
absorptions and low reduction potentials.
orole is isoelectronic to a 4π-electron cyclopentadienyl
cation and thus represents an important building unit to
B
Figure 1. Ladder boroles and related compounds.
investigate π-conjugated systems with antiaromatic characteristics.1,2
Theoretical studies have suggested unusual electronic features of
borole-based π-conjugated systems.3 Whereas the first borole deri-
vative, 1,2,3,4,5-pentaphenylborole, was reported by Eisch and
co-workers in 1969,4,5 only recently has its crystal structure been
determined by X-ray crystallography.6,7 Since then, significant atten-
tion has been focused on this chemistry. During the past few years,
various fascinating borole derivatives have been developed,8,9 such as
boroleꢀmetal complexes,8b carbene-coordinated boroles,8c and
highly Lewis acidic boroles.9
Scheme 1. Synthesis of Dithienoborole
One crucial issue in the molecular design of new borole-based
materials is how to stabilize the borole skeleton. General
strategies are to construct the ring-fused structure in order to
decrease the antiaromaticity and to introduce a bulky aryl group
onto the boron atom for kinetic stabilization (Figure 1). On the
basis of these considerations, several stable dibenzoboroles 1
with potential utilities for various applications, such as fluores-
cent sensors10a and electron-transporting materials,10b were
synthesized. However, examples of the ring-fused ladder-type
borole derivatives are still limited, and their chemistry has not yet
matured. In particular, little is known about heteroarene-fused
borole derivatives.11 It can be envisioned that fusing of the
heteroaryl rings, such as the electron-donating thiophene, should
significantly perturb the electronic structure of the electron-
accepting borole skeleton.
We now report synthetic methodologies for a series of thio-
phene-fused ladder boroles 2ꢀ4 and their unusual antiaromatic
characteristics compared to the dibenzoborole 1a (R = iPr, Ar = H)
or the parent borole 5. Recently, analogous ladder heteroles with
various main-group elements,12 such as compounds 6ꢀ8 contain-
ing Si,13 P,14 and S,15 have been reported as promising emitting
materials and stable transistor materials. Comparisons with these
other heteroles in terms of photophysical and electrochemical
properties will also provide a clear-cut view of the uniqueness of
the ladder boroles.
The most straightforward synthetic route to the dithienobor-
ole skeleton may be the reaction of dimetalated bithiophene with
ArBX2.11 However, all our attempts were unsuccessful. There-
fore, we explored an alternative route, stepwise substitution
reactions on the boron atom, as shown in Scheme 1. Thus, using
30-bromo-2,20-bithienyl-3-boronic ester 9 as a key precursor, the
2,4,6-triisopropylphenyl (Tip) group was first introduced onto
the boron atom by the reaction with TipMgBr in refluxing THF.
As the Tip group is bulky enough to protect the boron center,
lithiation with tBuLi then proceeded selectively at the 30 position
of the bithiophene skeleton. The generated anion subsequently
underwent intramolecular nucleophilic substitution on the boron
atom to produce the target compound 2. After silica gel column
chromatography under an argon atmosphere, recrystallization
from pentane gave the target compound in 37% yield.
Received: March 4, 2011
Published: April 19, 2011
r
2011 American Chemical Society
6952
dx.doi.org/10.1021/ja2019977 J. Am. Chem. Soc. 2011, 133, 6952–6955
|

Journal of the American Chemical Society
COMMUNICATION
Scheme 2. Synthesis of Ladder Boroles
Figure 3. Plot of the 1H NMR chemical shifts of the ortho-iPr methyne
protons for 1aꢀ4 as a function of the NICS(1)zz values of their borole
rings. The calculations were conducted at the B3LYP/6-311þG** level.
comparable to those of the non-fused tetraphenylated borole
derivatives.7 Worth noting is the extent of the bond alternation in
the butadiene moiety. While the C1ꢀC2 and C3ꢀC4 bond
lengths are 1.377(3) and 1.376(3) Å, respectively, the C2ꢀC3
bond length is 1.478(3) Å. The extent of this bond alternation is
rather small compared to those of the non-fused borole
derivatives7 and also that of the parent borole 5 optimized at
the B3LYP/6-31G* level. The small bond alternation in the
borole ring is also observed for the other ladder compounds 2
and 3. This structural feature should affect their electronic
structures (vide infra).
We initially envisioned that the electron-donating thio-
phene would decrease the Lewis acidity of the boron atom
and thereby stabilize the borole ring. Against our expectation,
however, all the thiophene-fused ladder molecules were air-
and moisture-sensitive and, therefore, needed handling under
an argon atmosphere using dry solvents for purification. This
is in contrast to the fact that the dibenzoborole 1a can be
isolated by silica gel column chromatography in the open
atmosphere. This difference demonstrates the higher reactiv-
ity of the thiophene-fused borole skeletons. To elucidate this
difference, we investigated the effect of the fused thiophene
ring on the antiaromaticity of the borole ring. As a measure of
the antiaromaticity, we evaluated the nucleus-independent
chemical shift (NICS) values by the DFT calculations at the
B3LYP/6-311þG** level,17 using the geometries derived
from the crystal structures.18 All the thiophene-fused boroles
have more positive NICS(1)zz values than that (þ29.4 ppm)
of the parent borole 5. This is in contrast to the fact that the
dibenzoborole 1a (þ24.5) has a less positive NICS(1)zz value
than 5. The NICS(1)zz values increase in the order of 3
(þ30.1) < 2 (þ40.3) < 4 (þ45.3).19
Figure 2. ORTEP drawing of 4 (50% probability for thermal
ellipsoids). Hydrogen atoms are omitted for clarity.
The combination of the present stepwise substitution
reaction with a domino-type intramolecular double cycli-
zation16 enabled us to synthesize more extended ladder
boroles, as shown in Scheme 2. The diarylacetylenes 10 and
11 bearing boronic ester groups were employed as precursors.
Reactions of these compounds with TipMgBr selectively
produced the corresponding dimeric products 12 and 13,
which were isolated by column chromatography in 52% and
23% yields, respectively. Successive treatments of these
compounds with tBuLi and a toluene solution of elemental
sulfur produced the thiolate anion intermediates, which
underwent 5-endo-dig cyclization followed by the intramole-
cular substitution at the boron atom. Overall, the borole and
thiophene ring skeletons were constructed in a fused fashion
in one step. Based on this procedure, a four-ring-fused borole
3 and a five-ring-fused borole 4 were obtained in 39% and
24% yields, respectively.
The products were identified by various NMR spectroscopies
and mass spectrometry. In the 11B NMR spectra, broad signals
were observed at 62ꢀ65 ppm for 2ꢀ4, which are characteristic of
a borole boron atom.6,7 The structures of all the compounds were
finally confirmed by X-ray crystallographic analysis (see the
Supporting Information). As a representative example, an OR-
TEP drawing of the longest ladder borole 4 is shown in Figure 2.
Compound 4 has a nearly coplanar π-conjugated framework
with small dihedral angles between the central borole and two
terminal benzene rings of 2.00(14)° and 4.25(15)°. The Tip
group is arranged perpendicular to the borole ring. As for the
geometry of the borole ring, the BꢀC bond lengths (1.574(3)
and 1.576(3) Å) and the CꢀBꢀC bond angle (102.84(15)°) are
The trend in the antiaromaticity for the ladder boroles is
consistent with the experimental observations. According to the
crystal structures of the ladder boroles 1aꢀ4, the methyne protons
of the ortho-isopropyl groups in the Tip group locate above and
below the borole boron atom. In the 1H NMR spectra, the chemical
shifts (δ) of the methyne protons should be perturbed by the
paratropic ring-current effects of the borole ring. Indeed, the
chemical shifts of the methyne protons exhibited downfield shifts
in the order of 1a < 3 < 2 < 4; more importantly, the plot of the δ
values versus the calculated NICS(1)zz values showed a linear
relationship (Figure 3). All these results imply that the thiophene-
fused structure enhances the antiaromaticity of the borole ring.
Notably, this is opposite to the conventional understanding that
fusion of the aromatic arene rings can decrease the antiaromatic
character of the 4π electron skeletons.20
6953
dx.doi.org/10.1021/ja2019977 |J. Am. Chem. Soc. 2011, 133, 6952–6955

Journal of the American Chemical Society
COMMUNICATION
Figure 5. KohnꢀSham HOMO and LUMO energy levels for a series of
the ladder boroles 1aꢀ4 and related compounds 6ꢀ8, calculated at the
B3LYP/6-31G* level of theory.
Figure 4. UVꢀvis absorption spectra of 2 (solid line), 3 (dashed line),
and 4 (dotted line) in CH2Cl2, together with that of 1a (dashed dotted
line). Inset: photographs of the CH2Cl2 solutions of 1aꢀ4.
longest absorption maximum (λmax) at 420 nm as a weak shoulder
band, the dithieno-fused analogue 2 showed a purple color, and its
longest λmax was 552 nm, which is more than 130 nm red-shifted
compared to that of 1a. The molar absorption coefficient was also
increased by replacing the benzene ring with a thiophene ring.
Notably, the four-ring-fused 3 has a shorter λmax than 2, despite its
longer π-conjugated framework, indicative of the unique electronic
structure of the “dithieno-fused” borole skeleton. In fact, the more
extended ladder molecule 4, with the dithienoborole skeleton, has
the longest λmax of 600 nm and showed a blue color. It is note-
worthy that this λmax value is much longer even compared to those
of the other heterole analogues with identical π-conjugated lengths:
6, 382 nm;13 7, 424 nm;14 8, 344 nm.15b
In the fluorescence spectra, while the dibenzoborole 1a showed
a green fluorescence at 522 nm with a quantum yield of 0.48
(CH2Cl2), all the thiophene-fused boroles 2ꢀ4 were non-emissive.
This difference suggests the different nature of the excited states
between 1a and 2ꢀ4.
Table 1. Photophysical and Electrochemical Data for the
Ladder Boroles and Related Compounds
UVꢀvis absorptiona
reduction potentialb
compd
λabs/nm
log ε
E1/2/Vc
Epc/Vd
1a
2
420 (sh)
552
2.33
3.05
2.93
3.04
ꢀ2.11
ꢀ1.98
ꢀ1.96
ꢀ1.72
ꢀ3.05
ꢀ2.79
ꢀ2.89
ꢀ2.61
3
469
4
600
a In CH2Cl2. b Determined by cyclic voltammetry under the following
conditions: sample 1 mM; Bu4NþPF6ꢀ (0.1 M) in THF; scan rate 100
mVsꢀ1. c First reduction potential vs Fc/Fcþ. d Second reduction peak
potential vs Fc/Fcþ.
This unusual effect of the thiophene-fused structure may be
rationalized by considering the geometrical features of the thio-
phene-fused borole skeletons. We conducted NICS calculations on
the parent boroles 50 and 500, whose geometries were directly
derived from the crystal structures of 2 and 4, respectively, without
including the thiophene moieties (see the Supporting Infor-
mation). Notably, these boroles have much more positive NICS-
(1)zz values, þ43.3 and þ40.3 ppm, than that of the optimized 5,
despite the absence of an electronic influence from the thiophene
rings. These results suggest that the borole geometry in the
thiophene-fused skeleton should be the origin of the enhanced
antiaromaticity. A major difference in the geometries between the
boroles 50 and 500 and the optimized 5 is the extent of bond
alternation. While in general the borole ring tends to have a large
bond alternation in the butadiene moiety in order to minimize the
antiaromatic character,7 the butadiene moieties in 50 and 500 have
shorter CꢀC single bonds as well as longer CdC double bonds
compared to those in 5, resulting in the smaller extent of the bond
alternation, as described in the crystal structure section. This may
be responsible for the enhanced antiaromatictiy of the thiophene-
fused borole skeletons.
To gain insight into the electronic structures of the ladder
boroles, we compared the frontier orbital energy levels of 1aꢀ4
and 6ꢀ8, as shown in Figure 5. There are some notable points.
(1) While the dithieno-fused borole 2 has a slightly lower-lying
LUMO than that of the dibenzo-fused analogue 1a, the HOMO
of 2 is substantially higher, by 0.61 eV, compared to that of 1a.
This difference in the HOMO level mainly contributes to the
significantly red-shifted λmax of the dithienoborole. (2) Whereas
the antiaromaticity increases in the order of 1a < 3 < 2 (vide
supra), the LUMO levels of 2 and 3 are lower than that of 1a but
are comparable to each other. This comparison implies that the
LUMO level and antiaromaticity are not directly related. (3)
Among a series of bis(benzothieno)heteroles, the ladder borole 4
has the lowest-lying LUMO. While incorporation of the electron-
withdrawing PdO moiety is recognized as an effective strategy to
attain a low LUMO level,14,21 this also results in a decreased
HOMO level. In contrast, incorporation of the boron atom
decreases the LUMO level while maintaining the high HOMO
level. The lower LUMO level of 4 compared to the PdO-
containing 7 demonstrates the significant effect of the orbital
interaction in the borole ring. As a consequence, 4 has the longest
λmax among the series of bis(benzothieno)heteroles. Time-
dependent DFT calculations confirmed that the major contribu-
tions to the longest absorption band for all these compounds are
the HOMOꢀLUMO transitions.
We have two interests regarding the electronic properties. One is
how the thiophene-fused structure affects the photophysical and
electrochemical properties compared to the benzene-fused struc-
ture. The other is how the ladder boroles are different from the
other ladder heteroles. In the UVꢀvis absorption spectra in
CH2Cl2, the thiophene-fused ladder boroles indeed showed char-
acteristic absorption bands, as shown in Figure 4. Their data are
summarized in Table 1. While the dibenzoborole 1a showed the
To experimentally confirm the electronic features discussed
in the previous section, we measured the cyclic voltammetry
6954
dx.doi.org/10.1021/ja2019977 |J. Am. Chem. Soc. 2011, 133, 6952–6955

Journal of the American Chemical Society
COMMUNICATION
for 1aꢀ4. Their electrochemical data are also summarized in
Table 1. All the ladder boroles showed two-step reduction
processes with the first reversible redox waves and the second
irreversible reduction wave. Notably, the first reduction potential
of 4 (E1/2 = ꢀ1.72 V vs Fc/Fcþ) is more positive than that of the
PdO-containing analogue 7 (E1/2 = ꢀ1.98 V vs Fc/Fcþ).22 This
difference demonstrates the uniqueness of the borole ring as an
electron-accepting building unit.
Breher, F.; Chiu, C.-W.; Gamon, D.; Nied, D.; Radacki, K. Angew. Chem.
Int. Ed. 2010, 49, 8975.
(9) (a) Fan, C.; Piers, W. E.; Parvez, M. Angew. Chem. Int. Ed. 2009,
48, 2955. (b) Fan, C.; Mercier, L. G.; Piers, W. E.; Tuononen, H. M.;
Parvez, M. J. Am. Chem. Soc. 2010, 132, 9604. (c) Fan, C.; Piers, W. E.;
Parvez, M.; McDonald, R. Organometallics 2010, 29, 5132.
(10) (a) Yamaguchi, S.; Shirasaka, T.; Akiyama, S.; Tamao, K. J. Am.
Chem. Soc. 2002, 124, 8816. (b) Wakamiya, A.; Mishima, K.; Ekawa, K.;
Yamaguchi, S. Chem. Commun. 2008, 579.
In summary, we established a synthetic route to the thiophene-
fused ladder boroles based on the stepwise substitution of the
thienylboronic esters. The boroles 2ꢀ4 are air- and moisture-
sensitive due to the enhanced antiaromaticity of the borole ring
with the thiophene-fused skeleton. The characteristic absorption
and electrochemical properties of the ladder boroles, significantly
different from those of the other known heterole analogues,
demonstrate the uniqueness of the borole skeleton as a building
unit. These results provide important fundamental knowledge on
the extended π-electron systems with high antiaromatic character-
istics. Further study on the structureꢀproperties relationships for a
variety of heteroarene-fused boroles is now in progress.
(11) The synthesis of a dithienoborole derivative was reported pre-
viously, but its photophysical properties are largely different from our
results: Kim, S.; Song, K.; Kang, S. O.; Ko, J. Chem. Commun. 2004, 68.
(12) Fukazawa, A.; Yamaguchi, S. Chem. Asian J. 2009, 4, 1386.
(13) Ohshita, J.; Lee, K.-H.; Kimura, K.; Kunai, A. Organometallics
2004, 23, 5622.
(14) Dienes, Y.; Eggenstein, M.; Kꢀarpꢀati, T.; Sutherland, T. C.;
Nyulꢀaszi, L.; Baumgartner, T. Chem. Eur. J. 2008, 14, 9878.
(15) (a) Schroth, W.; Hintzsche, E.; Viola, H.; Winkler, R.; Klose,
H.; Boese, R.; Kempe, R.; Sieler, J. Chem. Ber. 1994, 127, 401. (b)
Okamoto, T.; Kudoh, K.; Wakamiya, A.; Yamaguchi, S. Org. Lett. 2005,
7, 5301. (c) Gao, J.; Li, R.; Li, L.; Meng, Q.; Jiang, H.; Li, H.; Hu, W. Adv.
Mater. 2007, 19, 3008.
(16) Mouri, K.; Wakamiya, A.; Yamada, H.; Kajiwara, T.; Yamaguchi,
S. Org. Lett. 2007, 9, 93.
’ ASSOCIATED CONTENT
(17) (a) Schleyer, P. v. R.; Maerker, C.; Dransfeld, A.; Jiao, H.;
van Eikema Hommes, N. J. R. J. Am. Chem. Soc. 1996, 118, 6317.
(b) Corminboeuf, C.; Heine, T.; Seifert, G.; Schleyer, P. v. R.; Weber, J.
Phys. Chem. Chem. Phys. 2004, 6, 273. (c) Fallah-Bagher-Shaldael, H.;
Wannere, C. S.; Corminboeuf, C.; Puchta, R.; Schleyer, P. v. R. Org. Lett.
2006, 8, 863.
S
Supporting Information. Experimental procedures and
b
spectral data for all new compounds, crystallographic data and
ORTEP drawings of 1aꢀ4, and results of theoretical calculations.
This material is available free of charge via the Internet at http://
pubs.acs.org.
(18) The NICS values based on the optimized structures for the
corresponding model compounds were also consistent with the values
obtained for the crystal structures (Supporting Information).
(19) The NICS(1)zz values of the thiophene rings in 2, 3, and 4
are ꢀ13.1, ꢀ12.3, and ꢀ13.7 ppm, respectively.
’ AUTHOR INFORMATION
Corresponding Author
yamaguchi.shigehiro@b.mbox.nagoya-u.ac.jp
(20) Allen, A. D.; Tidwell, T. T. Chem. Rev. 2001, 101, 1333.
(21) (a) Baumgartner, T.; Rꢀeau, R. Chem. Rev. 2006, 106, 4681.
(b) Fukazawa, A.; Hara, M.; Okamoto, T.; Son, E.-C.; Xu, C.; Tamao, K.;
Yamaguchi, S. Org. Lett. 2008, 10, 913.
’ ACKNOWLEDGMENT
(22) This value was determined under conditions identical with
those for the measurement of 4 and is consistent with the value originally
reported in the literature (E1/2 = ꢀ1.45 V vs Ag/AgCl).14
This work was partly supported by a Grant-in-Aid (No. 19675001)
from the Ministry of Education, Culture, Sports, Science, and
Technology, Japan, and CREST, JST. A.I. acknowledges the JSPS
research fellowship for young scientists.
’ REFERENCES
(1) Huynh, K.; Vignolle, J.; Tilley, T. D. Angew. Chem. Int. Ed. 2009,
48, 2835.
(2) (a) Minsky, A.; Meyer, A. Y.; Rabinovitz, M. Tetrahedron 1985,
41, 785. (b) Jusꢀelius, J.; Sundholm, D. Phys. Chem. Chem. Phys. 2008,
10, 6630.
(3) (a) Salzner, U.; Lagowski, J. B.; Pickup, P. G.; Poirier, R. A. Synth.
Met. 1998, 96, 177. (b) Cao, H.; Ma, J.; Zhang, G.; Jiang, Y. Macro-
molecules 2005, 38, 1123.
(4) Eisch, J. J.; Hota, N. K.; Kozima, S. J. Am. Chem. Soc. 1969,
91, 4575.
(5) (a) Eisch, J. J.; Galle, J. E.; Kozima, S. J. Am. Chem. Soc. 1986,
108, 379. (b) Eisch, J. J.; Galle, J. E.; Shafii, B.; Rheingold, A. L.
Organometallics 1990, 9, 2342. (c) Fagan, P. J.; Nugent, W. A.; Calabrese,
J. C. J. Am. Chem. Soc. 1994, 116, 1880.
(6) Braunschweig, H.; Fernꢀandez, I.; Frenking, G.; Kupfer, T. Angew.
Chem. Int. Ed. 2008, 47, 1951.
(7) So, C.-W.; Watanabe, D.; Wakamiya, A.; Yamaguchi, S. Organo-
metallics 2008, 27, 3496.
(8) (a) Braunschweig, H.; Kupfer, T. Chem. Commun. 2008, 4487.
(b) Braunschweig, H.; Chiu, C.-W.; Radacki, K.; Brenner, P. Chem.
Commun. 2010, 46, 916. (c) Braunschweig, H.; Chiu, C.-W.; Radacki, K.;
Kupfer, T. Angew. Chem. Int. Ed. 2010, 49, 2041. (d) Braunschweig, H.;
6955
dx.doi.org/10.1021/ja2019977 |J. Am. Chem. Soc. 2011, 133, 6952–6955