Synthesis, structure, aggregation-induced emission, self-assembly, and electron mobility of 2,5-bis(triphenylsilylethynyl)-3,4-diphenylsiloles

Article abstract of DOI:10.1002/chem.201003382

2,5-Bis(triphenylsilylethynyl)-3,4-diphenylsiloles with different 1,1-substituents [XYSi(CPh)₂(C-C≡C-SiPh₃) ₂] (Ph=phenyl) were synthesized in high yields by the Sonogashira coupling of 2,5-dibromo-3,4-diphenylsiloles with triphenylsilylacetylene, and two of these were characterized crystallographically. Crystal structures and theoretical calculations showed that the new silole molecules had higher conjugation than 2,5-diarylsiloles. They possessed low HOMO and LUMO energy levels due to the electron-withdrawing effect of the triphenylsilylethynyl groups. Cyclic voltammetry analysis revealed low electron affinities, which were comparable to those of perfluoroarylsiloles. B3LYP/6-31* calculations demonstrated that the new siloles possessed large reorganization energies for electron and hole transfers and high electron mobilities. A mobility of up to 1.2 × 10^-5 cm²V^-1s^-1 was obtained by the transient electroluminescence method, which was about fivefold higher than that of tris(8-hydroxyquinolinato)aluminum_, a widely used electron-transport material, under the same conditions. All of the silole molecules possessed high thermal stability. Although, their solutions were weakly emissive, their nanoparticle suspensions and thin films emitted intense blue-green light upon photoexcitation, demonstrating a novel feature of aggregation-induced emission (AIE). Polarized emissions were observed in the silole crystals. The addition of solvents, which did not dissolve the silole molecules, into silole-containing solutions caused self-assembly of the molecules, which produced macroscopic fibrils with strong light emissions. Copyright

Full text of DOI:10.1002/chem.201003382

DOI: 10.1002/chem.201003382
Synthesis, Structure, Aggregation-Induced Emission, Self-Assembly, and
Electron Mobility of 2,5-Bis(triphenylsilylethynyl)-3,4-diphenylsiloles
Zujin Zhao,[a, b] Dandan Liu,[c] Faisal Mahtab,[a] Linyuan Xin,[c] Zhongfang Shen,[c]
Yong Yu,[a] Carrie Y. K. Chan,[a] Ping Lu,[c] Jacky W. Y. Lam,[a] Herman H. Y. Sung,[a]
Ian D. Williams,[a] Bing Yang,[c] Yuguang Ma,[c] and Ben Zhong Tang*[a, d]
Abstract: 2,5-Bis(triphenylsilylethyn-
yl)-3,4-diphenylsiloles with different
ꢀ ꢁ ꢀ
were comparable to those of perfluoro-
arylsiloles. B3LYP/6-31* calculations
demonstrated that the new siloles pos-
sessed large reorganization energies for
electron and hole transfers and high
electron mobilities. A mobility of up to
1.2ꢀ10ꢀ5 cm2 Vꢀ1 sꢀ1 was obtained by
same conditions. All of the silole mole-
cules possessed high thermal stability.
Although, their solutions were weakly
emissive, their nanoparticle suspensions
and thin films emitted intense blue-
green light upon photoexcitation, dem-
onstrating a novel feature of aggrega-
tion-induced emission (AIE). Polarized
emissions were observed in the silole
crystals. The addition of solvents, which
did not dissolve the silole molecules,
into silole-containing solutions caused
self-assembly of the molecules, which
produced macroscopic fibrils with
strong light emissions.
1,1-substituents [XYSiACTHNUTRGNEUNG(CPh)2(C C C
SiPh3)2] (Ph=phenyl) were synthesized
in high yields by the Sonogashira cou-
pling of 2,5-dibromo-3,4-diphenylsiloles
with triphenylsilylacetylene, and two of
these were characterized crystallo-
graphically. Crystal structures and the-
oretical calculations showed that the
new silole molecules had higher conju-
gation than 2,5-diarylsiloles. They pos-
sessed low HOMO and LUMO energy
levels due to the electron-withdrawing
effect of the triphenylsilylethynyl
groups. Cyclic voltammetry analysis re-
vealed low electron affinities, which
the
transient
electroluminescence
method, which was about fivefold
higher than that of tris(8-hydroxyqui-
nolinato)aluminum, a widely used elec-
tron-transport material, under the
Keywords: electron transport
·
electronic structure · fluorescence ·
self-assembly · silole
Introduction
occupied molecular orbital (LUMO) energy level.[1] As a
result, siloles exhibit high electron affinity and fast electron
mobility.[2] Recently, great efforts have been made to func-
tionalize siloles and utilize them as electron-transport mate-
rials,[3] light emitters,[4] building blocks for conducting poly-
mers,[5] fluorescence sensors,[6] solar cells,[7] electrochemilu-
minescence,[8] and so forth. Meanwhile, many theoretical
studies have been conducted in order to better understand
the substitution effect on their electronic structures and op-
toelectronic properties.[9]
Siloles (silacyclopentadienes) are silicon-containing five-
membered cyclic dienes that have attracted extensive atten-
tion due to their unique electronic properties and potential
high-technological applications. They are considered novel
s*–p* conjugated materials, in which the s* orbital of the
silicon–carbon bond effectively interacts with the p* orbital
of the butadiene fragment, leading to a low-lying lowest un-
One-dimensional (1D) self-assembly of functional materi-
als has attracted increasing interest for the fabrication of
nanoscale electronic and optoelectronic devices, in which
the device miniaturization requires small channel materi-
als.[10] In particular, macroscopic self-assembled materials
constructed from small molecules often show more unique
and/or enhanced properties than their molecular compo-
nents.[11] The driving force of the supramolecular self-assem-
bly can be hydrogen bonding and/or other weak interac-
tions, such as p–p stacking, van der Waals forces, and elec-
trostatic forces. Recent investigations suggest that p–p
stacking between planar, rigid, organic small molecules and
oligomers is an effective way for aromatic molecules to as-
semble into materials with a 1D columnar nanostructure.
Such processes have increased the charge carrier mobility of
the organic semiconductors due to strong electronic cou-
[a] Dr. Z. Zhao, F. Mahtab, Y. Yu, C. Y. K. Chan, Dr. J. W. Y. Lam,
Dr. H. H. Y. Sung, Prof. Dr. I. D. Williams, Prof. Dr. B. Z. Tang
Department of Chemistry
The Hong Kong University of Science & Technology
Clear Water Bay, Kowloon, Hong Kong (P. R. China)
Fax : (+852)2358-1594
E-mail: tangbenz@ust.hk
[b] Dr. Z. Zhao
College of Material, Chemistry and Chemical Engineering
Hangzhou Normal University, Hangzhou 310036 (P. R. China)
[c] D. Liu, L. Xin, Z. Shen, Dr. P. Lu, Dr. B. Yang, Prof. Dr. Y. Ma
State Key Laboratory of Supramolecular Structure and Materials
Jilin University, Changchun 130012 (P. R. China)
[d] Prof. Dr. B. Z. Tang
Department of Polymer Science and Engineering
Zhejiang University, Hangzhou 310027 (P. R. China)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201003382.
5998
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 5998 – 6008

FULL PAPER
pling between the molecules and polarized fluorescence aris-
ing from anisotropy in different directions.[12] However, it is
generally known that chromophore aggregation can quench
light emission due to increased excited state relaxation
through nonradiative pathways.[13] Thus, the synthesis of flu-
orophores with strong emissions in the aggregate states
through 1D self-assembly is intriguing, although such mate-
rials are seldom reported. Recently, Park prepared several
organogels by the self-assembly of low-molecular-weight flu-
orophores and found that the emission is induced or en-
hanced along the stacking direction.[14] This may open a new
avenue for the development of columnar self-assembling
materials with efficient light emissions.
During our search for highly luminescent materials, we
were attracted by a group of propeller-shaped molecules
called siloles. The structures of some molecules are shown in
Scheme 1. These molecules are nonluminescent in solution
Scheme 2. Synthesis of 2,5-bis(triphenylsilylethynyl)-3,4-diphenylsiloles.
Np=naphthyl, TMEDA=tetramethylethylenediamine. NBS=N-bromo-
succinimide.
of solvents, which do not dissolve the silole molecules, into
solutions of 1c caused self-assembly of the molecules and
produced fluorescent fibrils with macroscopic sizes.
Results and Discussion
Scheme 1. Chemical structures of 2,3,4,5-tetraphenylsiloles.
Synthesis: To enlarge the library of silole molecules, we pre-
pared three 2,5-bis(triphenylsilylethynyl)-3,4-diphenylsiloles
with different 1,1-substituents according to Scheme 2. The
one-pot synthesis of 2,5-extended siloles from 2,5-dilithiosi-
loles generally results in many side reactions, making the pu-
rification of the desired products a painful task. However,
Sonogashira coupling of aromatic halides with alkynes is
more efficient and much “cleaner” and has been widely
used to construct conjugated monomers and oligomers.[19]
Thus, we prepared the 2,5-dibromosiloles (4a–4c) inter-
mediates according to the literature methods,[16c,19] by bromi-
nation of 2,5-dilithiosiloles with N-bromosuccinimide
(NBS). Single crystals of 4a and 4c were grown from their
methanol/dichloromethane mixtures. The ORTEP plots of
4a and 4c are shown in Figure S1 in the Supporting Infor-
mation and their crystal data are summarized in Table S1.
Coupling of 4a–4c with triphenylsilylacetylene in the pres-
but emit intensely in the solid-phase.[15] We coined this
novel phenomenon as “aggregation-induced emission”
(AIE) because the nonemissive silole molecules are induced
to emit intensely by aggregate formation. Mechanistic stud-
ies by different approaches reveal that the AIE effect is at-
tributed to the restriction of intramolecular rotation
(RIR).[16,17] Utilizing the AIE characteristics, we have fabri-
cated light-emitting diodes (LEDs) using siloles as active
layers and obtained outstanding device performances. An
external quantum efficiency of 8% was achieved in an or-
ganic light-emitting diode (OLED) device of MPPS
(Scheme 1;
MPPS=1-methyl-1,2,3,4,5-pentaphenylsilole),
approaching the limit of the possible.[18] Technological appli-
cations of silole molecules as environmental sensors and bio-
logical probes were also explored.
Siloles are unique materials, and, therefore, preparation
of new siloles is of academic and practical value. Studying
their structure–property relationship will give valuable infor-
mation on the synthesis of new molecules with novel molec-
ular structures and properties. Due to its linear geometry
and rich chemistry, triple bonds have been used as function-
al bridging groups to link aromatic rings together. In this
work, we designed and synthesized new silole molecules by
attaching triphenylsilyl groups to the silole core at the
2,5-positions through triple-bond functionalities (Scheme 2).
The substitution effect at the 2,5-positions on their crystal
and electronic structures were investigated. Thanks to the
triphenylsilylethynyl groups, 1a–1c showed higher thermal
stability and electron mobility than HPS and MPPS
(Scheme 1; HPS=1,1,2,3,4,5-hexaphenylsilole). The addition
ence of [PdACHTNUTRGNEG(UN PPh3)4] (PPh3 =triphenylphosphine) and CuI
furnished the desired products 1a–1c in high yields. The
new silole molecules were soluble in common organic sol-
vents such as THF, dichloromethane, chloroform, and tolu-
ene, but were insoluble in ethanol, methanol, and water.
Crystal structures: The effect of bulkiness of the triphenyl-
silyl groups on the molecule structures of 1a–1c was investi-
gated by single-crystal X-ray diffraction. The crystals of si-
loles 1a and 1b were grown from solutions of the complexes
in methanol or n-hexane/dichloromethane mixtures as
yellow blocks and yellow needles, respectively. Crystals of
1c were, however, not obtained under similar conditions.
Suitable specimens for crystallography were mounted in air
Chem. Eur. J. 2011, 17, 5998 – 6008
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5999

B. Z. Tang et al.
on glass fibers and their X-ray diffraction intensity data
were collected on a diffractometer. The ORTEP drawings of
1a and 1b are shown in Figure 1 and their crystallographic
data are summarized in Table S2 in the Supporting Informa-
tion. The geometric parameters for the crystal structures of
both siloles are listed in Table 1. Siloles 1a and 1b crystal-
lized in the monoclinic system with space groups of P21/n
and P21/c, respectively. Both molecules exhibited an almost
planar silacyclopentadiene ring with low torsion angles
down to 1.18. Similar observations were also found in our
previous investigations on siloles with the same substituents
at the 1,1-positions.[17] Both 1a and 1b showed highly twist-
ed conformations of the phenyl groups at the ring carbons
of the silole skeletons. The phenyl substituents of 1a and 1b
were twisted out of the plane of silacyclopentadiene, with
dihedral angles to the phenyl planes of ꢀ124 and ꢀ58.78 at
the C2 ring carbon atoms and ꢀ142.9 and ꢀ608 at the C3
position, respectively. The extent of the twisting was, howev-
er, small, when compared with HPS and MPPS at the same
positions. It is worth noting that the phenyl blades of the tri-
phenylsilyl groups of adjacent molecules of 1a are arranged
in a parallel fashion with a interplanar distance of 3.518 ꢂ
(Figure 1c). Such spacing im-
plies p–p stacking and interac-
tions,[9e,20] which enables tight
packing of the molecules in the
three-dimensional crystal lat-
tice.
ꢀ
The endocyclic C C single
ꢀ
bond lengths (e.g., C2 C3=
1.508 ꢂ) of 1a were much
longer than the endocyclic C=C
double bonds (e.g., C1=C2=
ꢀ
1.361 ꢂ) and exocyclic C C
ꢀ
single bonds (C2 C11=1.486
ꢀ
and C1 C5=1.419 ꢂ). The en-
docyclic Si C single bond
lengths were also longer than
ꢀ
ꢀ
the exocyclic Si C single bonds.
Silole 1b also exhibited similar
Figure 1. a) and b) ORTEP drawings of 1a and 1b, respectively. c) p–p interactions between phenyl groups of
adjacent molecules of 1a. d) B3LYP/6-31G* calculated molecular orbital amplitude plots of HOMO and
LUMO energy levels of 1a.
ꢀ
behavior. The Si C single
bonds of 1a were somewhat
stretched due to the steric
Table 1. Experimental and calculated structural parameters for 1a and
effect of its two bulky phenyl groups, whereas in 1b, this
effect was lessened because it had only one phenyl group at
the 1-position. The bond angles of the silole rings of 1a and
1b varied slightly, with a typical pattern of a relatively small
angle about the silicon C1-Si1-C4 (90.78 for 1a and 90.18 for
1b), with standard angles of 109.4–110.28 at C1 and C4, and
wider angles of 114.5–115.58 between C2 and C3. The bond
angles at sp carbons (C6-C5-C1 and C5-C6-Si2, for example)
were in the range of 172.6–177.18, and those between the sp
carbons and the silole ring sp2 carbons (C2-C1-C5) were
1b.
1a
1b
X-ray
1.868(13)
1.871(13)
1.361(17)
1.508(18)
1.861(14)
calcd
1.889
X-ray
calcd
1.890
1.887
1.376
1.497
1.884
1.881[b]
1.405
1.486
1.227
bond lengths [ꢂ]
endocyclic Si1 C1
endocyclic Si1 C4
endocyclic C1 C2
endocyclic C2 C3
exocyclic Si1 C31
exocyclic Si1 C
exocyclic C1 C5
exocyclic C2 C11
exocyclic C5 C6
bond angles[8]
C1-Si1-C4
C2-C1-Si1
C3-C4-Si1
C1-C2-C3
C4-C3-C2
C1-C2-C11
C2-C1-C5
ꢀ
1.864(3)
1.867(3)
1.356(4)
1.502(4)
1.852(3)
ꢀ
1.892
1.377
1.496
1.892
1.878[a]
1.406
1.486
1.227
ꢀ
ꢀ
ꢀ
1.862(14)[a]
1.419(17)
1.486(17)
1.208(18)
1.844(3)[b]
1.426(4)
1.489(4)
1.205(4)
ꢀ
ꢀ
ꢀ
ꢁ
125.08 in 1a and 126.18 in 1b, which indicated that the C C
triple bonds were well conjugated with the silole ring.
ꢀ
90.7(6)
109.4(9)
109.9(9)
115.5(11)
114.5(11)
121.3(11)
125.0(12)
175.6(14)
175.7(13)
90.532
109.066
109.181
115.733
115.441
122.413
126.596
174.922
178.448
90.1(14)
110.2(2)
109.9(2)
114.7(3)
114.9(3)
123.2(3)
126.1(3)
172.6(3)
177.1(3)
90.569
109.229
109.105
115.420
115.628
122.483
127.239
174.336
178.802
Electronic structures: Silole absorbs more into the red
region than cyclopentadiene.[21] Tamao investigated the
structure of silole and compared it with cyclopentadiene by
HF/6-31G* calculations to evaluate the role of the silicon
atom in the electronic properties of the silole molecule. Re-
sults showed that silole had low-lying HOMO and LUMO
energy levels, which were about 0.4 and 1.3 eV lower than
those of cyclopentadiene, respectively. PM3 calculations re-
vealed that the low-lying LUMO energy level arose from
s*–p* conjugation between s* orbitals of the exocyclic s
bonds on silicon and p* orbitals of the butadiene skeleton.
Such effects give siloles unusual optical properties.[22] Tamao
C6-C5-C1
C5-C6-Si2
torsion angle [8]
Si1-C1-C2-C3
C1-C2-C3-C4
C3-C2-C11-C16
C2-C3-C21-C22
1.1(14)
0.01(16)
1.599
ꢀ2.512
5.1(4)
ꢀ2.6(4)
1.706
ꢀ2.631
ꢀ124.0(14)
ꢀ129.831
ꢀ131.868
ꢀ58.7(4)
ꢀ60.0(4)
ꢀ48.292
ꢀ47.938
ꢀ142.9(13)
ꢀ
ꢀ
[a] Si1 C41. [b] Si1 C30.
6000
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 5998 – 6008

Siloles
FULL PAPER
also investigated the properties of siloles with different sub-
stituents at the ring silicon atom[23] and 2,5-positions of the
silole ring.[9f] Both modifications were found to have strong
influences on the electronic structures and energy levels of
the silole molecules.
groups worked better than the trialkylsilylethynyl groups at
lowering the energy levels of the silole molecules. Although,
compared with the HOMO, the LUMO might play a more
important role in determining the electronic properties of si-
loles. The lower LUMO energy level of 1a than HPS may
be one of the key reasons for the difference in electron-
transport properties. The calculated band gap of 1a and 1b
were ꢂ0.6 eV lower than HPS (3.7 eV).
To evaluate the substituent effect on the electronic struc-
tures of 2,5-bis(triphenylsilylethynyl)-3,4-diphenylsiloles, we
carried out theoretical calculations of 1a and 1b. The
ground-state geometries were optimized using density func-
tional theory (DFT) with B3LYP hybrid functional at the
basis set level of 6-31G*, and the unrestricted formalism
(UB3LYP) was adopted for the ion-state geometries. All of
the calculations were performed using the Gaussian 03 pack-
age in a power leader workstation. The calculated values for
the pertinent geometrical parameters were comparable to
the experimental values determined from X-ray crystallogra-
phy (Table 1). The optimized bond lengths and angles were
slightly longer, but were still in good agreement with the ex-
perimental values. However, the torsion angles showed a
large variation. This was in some senses expected because
the packing effect was not included in the calculation, which
offered the silole molecules a more open and less spherically
compressed molecular structure. The torsion angles of 1a
and 1b at the 3,4-positions were almost the same, revealing
that the substituents at the 1,1-positions had no pronounced
effects on the configurations of the phenyl groups at the 3,4-
positions. The optimized structures of 1a and 1b also
showed that their phenyl groups at the 3,4-positions were
less twisted and the silole rings were more planar than HPS
and MPPS.[17] This may be due to the triple bond functional
bridge groups, which showed steric hindrance between the
silole cores and the triphenylsilyl groups.[16c,24]
Redox activity: Hole and electron injections in electronic
devices strongly influence the device efficiency. It is impor-
tant to assess hole and electron injection barriers at the in-
terfaces of various materials in designing device configura-
tions. Such parameters can be estimated by comparing the
ionization potentials (IPs) and electronic affinities (EAs) of
the molecular components with each other and/or with
Fermi energies of inorganic electrode materials.[25] These
data can be obtained through electrochemical measure-
ments.
We investigated the redox behaviors of 1a–1c in the solu-
tion state by cyclic voltammetry using ferrocene as the inter-
nal reference. As illustrated in Figure 2, all of the 2,5-bis(tri-
The HOMOs and LUMOs of 1a and 1b displayed almost
identical surfaces. Figure 1d depicts the HOMO and LUMO
of 1a as an example. The HOMO and LUMO are mainly
dominated by orbitals originating from the silole ring and
ꢁ
ꢁ
two C C triple bonds at the 2,5-postions. The two C C
triple bonds had significant contributions to both HOMOs
and LUMOs due to efficient orbital overlapping. On the
other hand, the phenyl groups at the 3,4-positons participat-
ed and contributed little to the energy levels. A contribution
was not found for the phenyl groups attached to the Si2 and
Si3 atoms. Owing to the s*–p* conjugation, the LUMOs
Figure 2. Cyclic voltammograms of 1a–1c in dichloromethane containing
0.1m Bu4NPF6.
ꢀ
had significant orbital density at the two exocyclic Si1 C
bonds, which were identical to those of HPS, MPPS, and
other silole derivatives.[24]
phenylsilylethynyl)-3,4-diphenylsiloles exhibited only one ir-
reversible reduction peak at about ꢀ1.75 V. Similar to other
silole derivatives,[4b] an oxidation peak was not found. This
means that hole injection is more difficult than electron in-
jection in these siloles. According to references [4b,9c,d,26],
the solid-state EA of a molecular species can be estimated
by comparing its electrochemical potential for the
M0/ꢀ couple with that for a commonly used electron-trans-
port material (e.g., tris(8-hydroxyquinolinato)aluminum
(Alq3)), for which the solid-state EA and reduction potential
have been determined. Thus, the EAs of 1a–1c can be cal-
culated by the following equation [Eq. (1)].
The calculated HOMO energy levels for 1a and 1b were
nearly identical (ꢀ5.42 eV for 1a and ꢀ5.44 eV for 1b).
These values were lower than those of 1,1-dimethyl-2,3,4,5-
tetraphenylsilole (DMTPS, ꢀ5.29 eV), 2,5-dipyridylsilole
(PySPy, ꢀ5.28 eV),[9b] and 2,5-bis(trialkylsilylethynyl)-3,4-di-
phenylsiloles (from ꢀ5.28 to ꢀ5.39).[16c] The LUMO energy
levels for 1a and 1b were ꢀ2.29 eV and were much lower
than DMTPS (ꢀ1.59 eV), PySPy (ꢀ1.82 eV), and other
silole derivates determined using the same method,[9b,c,f] indi-
cating that the new silole derivatives were more electron af-
finitive. This also suggested that the triphenylsilylethynyl
Chem. Eur. J. 2011, 17, 5998 – 6008
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6001

B. Z. Tang et al.
loles,[9b,d,18] the geometric modifications for 1a and 1b upon
carrier-injection were primarily localized within the central
portion of the molecular system, the silole ring with two
triple bonds. When an electron was injected (anion), the
ð1Þ
EAðMÞꢀEAðAlq3Þ ¼ E=
1
ðAlq30=ꢀÞꢀE=
1
2
ðM0=ꢀÞ
2
Taking the values of E
1/2A
ACHTUNGTRENNUNG
ꢀ
exocyclic Si1 C bond lengths were increased, and the ring
lar to Ere (ꢀ1.75 V), the EAs of 1a–1c are in the range from
ꢀ2.55 to ꢀ3.05 eV. These values are ꢂ0.5 eV lower than
HPS (ꢀ1.95 to ꢀ2.45 eV),[9c] and close to those of 2,5-diper-
fluoroarylsiloles (ꢀ2.58 to ꢀ2.93 eV).[4b] This showed that
the triphenylsilylethynyl substituents have a similar effect to
the perfluoroaryl groups (e.g., C6F5, C12F9) for electron in-
jection. The values were also ꢂ0.5 eV lower than that of
Alq3 and were comparable to n-type conducting materials
with the lowest EA energies.[29] The low EA values of 1a–1c
might thus allow the molecules to find an array of applica-
tions in electronics. One aspect should be clarified that the
EAs of the silole molecules estimated by the above proce-
dures were only rough values. The main reason was that the
reduction potential value was irreversible and was not equal
to the real half-wave reduction potential. Nevertheless, the
comparisons were performed based on the same method
and data reported in the literature.
ꢀ
Si1 C bond lengths shortened. The cis-butadiene portion of
the central ring underwent a significant decrease in the
bond length alternation as the C C bonds adjusted to the
presence of the extra electron. Meanwhile, the total lengths
of the exocyclic C C single bonds and the C C triple bonds
decreased significantly to stabilize the additional charge.
The addition of a hole into 1a and 1b shortened the exocy-
clic Si1 C bond, but lengthened the ring Si1 C bond. Simi-
lar to the anion case, there was a significant change in the
ꢀ
ꢀ
ꢁ
ꢀ
ꢀ
ꢀ
bond-length alternation pattern of the C C bonds in the
central ring and the C C triple bonds at the 2,5-positions.
However, the effect of electron addition on the lengths of
ꢁ
ꢀ
ꢁ
exocyclic Si1 C single and C C triple bonds was more pro-
found than that of hole. With regard to the bond angle, the
anion had larger endocyclic C-Si-C and smaller exocyclic
C-Si-C bond angles than the neutral molecules. The cation
exhibited the opposite behavior. Both the cation and the
anion exhibited smaller torsion angles of the phenyl groups
at the 3,4-positions and shorter bridge bond lengths than
those of the neutral molecule, indicating that both the
cation and the anion possess less twisted structures with
higher conjugation. Overall, upon charge injection into 1a
and 1b, geometry relaxation took place mainly in the silole
Cation and anion properties: To investigate if 1a–1c can be
employed as charge transport materials in organic electronic
devices, geometry optimizations for both the radical anion
and radical cation states (i.e., a neutral molecule in the pres-
ence of an extra electron and a hole) were performed.
Table 2 lists the calculated variations of the structural pa-
rameters in both states. Similar to HPS, MPPS, and other si-
ꢁ
ring and the adjacent C C triple bonds at the 2,5-positions.
Reorganization energy and carrier mobility: Many reports
have showed that siloles have higher electron mobility than
Alq3.[2,17] To prove that it also holds for our new silole mole-
cules, we calculated the barriers for electron and hole trans-
fers in the radical anion and cation species and the carrier
mobility of 1a and 1b. To describe the charge transport in
the crystal state, we considered an incoherent hopping
model, in which the charge could only transfer between the
adjacent molecules. Viewing each hopping event as a nona-
diabatic electron transfer reaction, we used standard Marcus
theory[30] to express the rate of charge motion between the
neighboring molecules (W) in terms of the reorganization
energy (l) and the coupling matrix element (V). Assuming
that the temperature is sufficiently high that vibrational
modes can be treated classically, we obtained Equation (2),
in which kB is the Boltzmann constant and T is the tempera-
ture (298 K). Given the hopping rate between the two
neighbors, the diffusion coefficient can be evaluated accord-
ing to Equation (3), in which n is the dimensionality, Wi is
the hopping rate due to charge carrier to the ith neighbor, r
is the distance to neighbor i, and P is the relative probability
for charge carrier to a particular ith neighbor [Eq. (4)]. In
Equation (2), when n=1 the diffusion coefficient is in a spe-
cial direction, and 2n=6, summing over all possible hops,
leads to the averaged diffusion coefficient. The drift mobili-
ty of hopping, m, is then evaluated from the Einstein relation
[Eq. (5)], in which e is the electronic charge. Only when the
Table 2. Calculated variations in the structural parameters of 1a and 1b
in the cationic and anionic states.
1a
1b
cation
anion
cation
0.0126
0.0127
0.0383
anion
bond length [ꢂ]
ꢀ
endocyclic Si1 C1
0.0175
0.0118
0.0365
ꢀ0.0225
ꢀ0.0235
0.0478
ꢀ0.0203
ꢀ0.0197
0.0468
ꢀ
endocyclic Si1 C4
ꢀ
endocyclic C1 C2
ꢀ
endocyclic C2 C3
ꢀ0.0424
ꢀ0.0135
ꢀ0.0111[a]
ꢀ0.0233
ꢀ0.0083
0.0085
ꢀ0.058
ꢀ0.0417
ꢀ0.0067
ꢀ0.013[b]
ꢀ0.0238
ꢀ0.0095
0.0087
ꢀ0.0579
ꢀ
exocyclic Si1 C31
0.0195
0.0127
0.0185[a]
ꢀ0.0173
ꢀ0.0026
0.0117
0.0213[b]
ꢀ0.0182
ꢀ0.0022
0.0121
ꢀ
exocyclic Si1 C
ꢀ
exocyclic C1 C5
ꢀ
exocyclic C2 C11
ꢀ
exocyclic C5 C6
bond angle [8]
C1-Si1-C4
ꢀ2.3755
ꢀ1.3237
1.7738
ꢀ0.6057
ꢀ0.5196
ꢀ0.7156
ꢀ1.2841
ꢀ0.2354
0.3524
1.2721
ꢀ0.6754
ꢀ0.9592
0.2127
ꢀ2.2204
ꢀ0.1491
1.6951
ꢀ0.5767
ꢀ0.5402
ꢀ0.797
ꢀ1.3907
0.1739
1.0497
ꢀ1.3005
ꢀ0.7944
0.3714
C2-C1-Si1
C3-C4-Si1
C1-C2-C3
C4-C3-C2
C1-C2-C11
C2-C1-C5
C6-C5-C1
0.3711
0.2063
ꢀ1.5236
ꢀ1.6266
0.4943
0.2542
0.9031
0.9883
1.1654
0.3652
C5-C6-Si2
ꢀ1.9362
torsion angle [8]
Si1-C1-C2-C3
C1-C2-C3-C4
C3-C2-C11-C16
C2-C3-C21-C22
ꢀ1.3237
1.1139
1.9801
2.7244
ꢀ0.6754
0.6332
2.5573
2.7539
ꢀ0.1491
0.4602
3.963
ꢀ1.3005
1.1041
2.6495
2.8894
2.7568
ꢀ
ꢀ
[a] Si1 C41. [b] Si1 C30.
6002
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 5998 – 6008

Siloles
FULL PAPER
intermolecular transfer integral (V) and the reorganization
energy (l) have been calculated, can the average drift mobi-
lity be evaluated using Equations (2)–(5). A direct dimer
Hamiltonian method was adopted to calculate the intermo-
lecular transfer integral (V). It can be written as given in
and 1b, respectively, which were higher than the previously
reported values for HPS, MPPS, and other silole deriva-
tives.[9b] The energies for hole transfer (l+) were 0.486 and
0.519 eV, respectively. The l energy of 1a was much higher
ꢀ
than HPS (0.49 eV), fluorine-containing HPS (0.56 eV),[9c]
and other pyridine-substituted siloles (ꢂ0.50 eV),[9b] and
was approximately twice the value for Alq3 (0.276 eV) cal-
culated at the same level.[31] The l+ was also appreciably
high, being four to five times larger than pentacene[32] and
more than twofold higher than that of N,N’-diphenyl-N,N’-
0,site 1
0,site 2
Equation (6) , in which hꢀ
j and jꢀ
i rep-
HOMO=LUMO
HOMO=LUMO
resents the HOMO/LUMO wave functions of two isolated
molecules at different sites. F0 is the Fock matrix construct-
ed from the density matrix of the noninteracting dimer,
which can be evaluated as F=SCeCꢀ1, in which S is the
overlap matrix of the dimer, and the Kohn–Sham orbital co-
efficients C and energy Eigen value e are obtained by one-
step diagonalization without iteration.
bis(3-methylphenyl)-1,1’-biphenyl-4,4’-diamine,
a
widely
used hole transport material.[33] The geometric modifications
that occurred during the reduction and oxidation of the new
2,5-bis(triphenylsilylethynyl)-3,4-diphenylsiloles were pri-
marily localized at the central portion of the molecules, par-
ꢀ
ꢁ
ꢀ
ꢁ
1
=
V2
ꢀh
p
l
2
ð2Þ
ð3Þ
W ¼
D ¼
exp ꢀ
ticularly
in
the
silole
ring,
two
4kBT
lkBT
ꢁ
ꢀ
C C triple bonds, and exocyclic Si1 C bonds. This localiza-
tion was mainly responsible for the relatively large reorgani-
zation energy of 1a and 1b.
X
1
2n
r2WiPi
i
The calculated hole mobility of 1a and 1b was as low as
3.30ꢀ10ꢀ7 and 3.08ꢀ10ꢀ7 cm2 Vꢀ1 sꢀ1, respectively. However,
they exhibited reasonably high electron mobility. The value
for 1a was 2.92ꢀ10ꢀ5 cm2 Vꢀ1 sꢀ1, which was about tenfold
higher than that for 1b (3.84ꢀ10ꢀ6 cm2 Vꢀ1 sꢀ1). The much
higher electron mobility of 1a might be attributed to its
more symmetrical structure, which encouraged a more com-
pact stacking of molecules. This resulted in shorter intermo-
lecular distances and, hence, promoted larger electronic cou-
pling. Moreover, the p–p intermolecular interactions be-
tween the phenyl rings of the triphenylsilyl groups in the
crystals of 1a might be helpful to increase the electron hop-
ping rate,[9b,e] which thus gave the molecule a higher electron
mobility than those of 1,1,3,4-tetraphenylsioles with small-
sized trialkylsilylethynyl substituents at the 2,5-positions
(1.33ꢀ10ꢀ5–4.06ꢀ10ꢀ6 cm2 Vꢀ1 sꢀ1).[24a]
Wi
P
Pi ¼
m ¼
ð4Þ
Wi
i
e
D
ð5Þ
ð6Þ
kBT
0,site 1
HOMO=LUMO
0, site 2
HOMO=LUMO
Vþ=ꢀ ¼ hꢀ
jF0jꢀ
i
The reorganization energy (l) can be separated into the
sum of two primary components: 1) the medium reorganiza-
tion energy that arises from the modifications to the polari-
zation of the surrounding medium due to the presence of
excess charge (ls) and 2) the intramolecular reorganization
energy (li), which corresponds to the sum of geometry relax-
ation energies upon going from the neutral-state geometry
to the charged-state geometry and vice versa. Hence, l for
Thermal stability: The thermal properties of 1a–1c were ex-
amined by differential scanning calorimetry (DSC) and ther-
mogravimetric analysis (TGA) under nitrogen at a heating
rate of 20Kminꢀ1. Table 4 summarizes the results and their
TGA thermograms are given in Figure S4 in the Supporting
Information. For comparison, the thermograms of HPS,
MPPS, and DMTPS are also provided in the same figure.[17]
The melting points of 1a–1c were 251, 248, and 1948C, re-
hole and electron transfers is given by Equation (7), in
þ=ꢀ
which E
and Eng are the energies of the charged-state
ng
þ=ꢀ
and neutral-state with the neutral-state geometry, and E
cg
and Ecg are the energies of the charged-state and neutral-
state with the charged-state geometry. Considering the negli-
gible medium reorganization energy in the solid state, we
calculated the intramolecular contribution to the total reor-
ganization energies for electron and hole transfers, and the
results are compared in Table 3
and the charge-hopping path-
ways are illustrated in Figur-
es S2 and S3 in the Supporting
Information.
Table 3. Calculated transport characteristics of 1a and 1b.
Reorganization
energy [eV]
Pathway Dimer
Transfer
Drift mobility
[cm2 Vꢀ1 sꢀ1
m+ m
ꢀ
distance [ꢂ] integral [eV]
]
l+
l
V+
Vꢀ
ꢀ
I
9.761
4.92ꢀ10ꢀ5 2.42ꢀ10ꢀ3
1.17ꢀ10ꢀ4 2.20ꢀ10ꢀ4
2.48ꢀ10ꢀ5 3.17ꢀ10ꢀ4
1.35ꢀ10ꢀ4 2.64ꢀ10ꢀ5
4.30ꢀ10ꢀ6 9.30ꢀ10ꢀ5
1.43ꢀ10ꢀ5 1.34ꢀ10ꢀ5
2.06ꢀ10ꢀ4 8.20ꢀ10ꢀ4
2.35ꢀ10ꢀ5 1.10ꢀ10ꢀ4
þ=ꢀ
þ=ꢀ
II 17.672
III
IV
I
II 18.875
III
IV 18.063
l
¼ ðE ꢀE Þ þ ðEcgꢀEngÞ
3.30ꢀ10ꢀ7 2.92ꢀ10ꢀ5
3.08ꢀ10ꢀ7 3.84ꢀ10ꢀ6
þ=ꢀ
ng
cg
1a 0.486
0.574
0.541
9.020
8.650
8.402
ð7Þ
The total reorganization en-
ergies for electron transfer (l )
were 0.574 and 0.541 eV for 1a
1b 0.519
8.811
ꢀ
Chem. Eur. J. 2011, 17, 5998 – 6008
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6003

B. Z. Tang et al.
Table 4. Optical properties of 1 in solution (Soln),[a] crystalline (Cryst),
the silole ring, which hampered their electronic communica-
tion and resulted in larger band gaps.
and amorphous (Amor) states and their thermal properties.[b]
labs [nm]
lem [nm]
FF
Eg
Tm/Td
A
E
[eV] [oC]
Light emission: Similar to the previously reported siloles,
the wet spots of 1a–1c on TLC plates were nonluminescent
under the illumination of a UV lamp, but became visible
when the solvent evaporated. Their dilute solutions in THF
(10 mm) exhibited weak photoluminescence (PL) at 483–
490 nm when photoexcited at 370 nm. The fluorescence
quantum yields (FF) of 1a–1c in THF were measured to be
2.3, 1.0, and 0.7%, respectively (Table 4). Previous studies
showed that the active intramolecular rotation of the phenyl
rings around the axes of the single bond linked to the silole
ring had annihilated the excitons, thus making the silole
molecules weakly emissive in the dilute solutions.[9a,f,18] Al-
though the triphenylsilyl groups were bulky, it seemed that
they could not efficiently restrict the intramolecular rota-
tions of 1a–1c, resulting in low FF values in solution. How-
ever, their thin films were highly emissive at the same exci-
tation wavelength. The FF values of their thin solid films
were 18.1, 28.3, and 34.6% (Table 4), respectively, revealing
that their emissions, similar to other silole derivatives, were
induced by aggregate formation or in other words, they ex-
hibited a novel characteristic of AIE.
To confirm whether 1a–1c were really AIE active, we
added different amounts of water, in which the silole mole-
cules are insoluble, to their solutions in THF and studied the
PL change. The final concentration of the mixtures was kept
at the same value of 10 mm to allow standard comparison.
Figure 4a shows the PL spectra of 1a in THF/water mix-
tures with different water fractions. The PL intensity re-
mained virtually unchanged when up to ꢂ60% of water was
added. Afterward, it swiftly increased. Compound 1a is in-
soluble in water; therefore, its molecules must have aggre-
gated in the mixtures with high water contents. The popula-
tion of the aggregates and hence the PL intensity continued
to increase with subsequent increases in the water content.
Clearly, 1a was AIE active. Similar behaviors were also ob-
served in 1b and 1c (Figure 4b).
1a 405 (1.23)
1b 400 (1.24)
1c 397 (1.12)
490 487
486 484
491
489
490
18.1 (2.3) 2.75 251/412
28.3 (1.0) 2.77 248/401
34.6 (0.7) 2.79 194/396
483
–
[a] Dilute solutions in THF (10 mm). [b] Abbreviations: labs =absorption
maximum, e=molar absorption, lem =emission maximum, FF =fluores-
cence quantum yield of the film determined by integrating sphere (data
given in the parentheses were determined in THF solutions using 9,10-di-
phenylanthracene with a FF value of 90% in cyclohexane as standard),
Eg =energy band gap determined from the UV/Vis spectra, Tm =melting
point, Td =decomposition onset temperature.
spectively. They demonstrated high thermal stability with
onset decomposition temperatures at 412, 401, and 3968C
for 1a–1c, respectively. Such values were much higher than
those of HPS, MPPS, and DMTPS, probably due to the ther-
mally stable, bulky triphenylsilylethynyl groups, which gave
the present molecules high resistance to thermolysis.
UV/Vis absorption: Figure 3 shows the UV/Vis absorption
spectra of 1a–1c in THF. All of the molecules exhibited an
absorption band at approximately 260 nm arising from the
phenyl groups of the triphenylsilyl units as well as ones at-
tached to the silole rings. The conjugated silole core ab-
sorbed at longer wavelengths and showed a blueshift from
405 nm in 1a to 397 nm in 1c. Similar phenomenon have
been observed by Tamao[23] and us[15b,16c] and were attributed
to the change in the inductive effect by varying the substitu-
ents at the 1,1-positions. The absorption maxima of 1a–1c
showed a ꢂ40 nm redshift from those of HPS, MPPS, and
DMTPS (366, 363, and 361 nm, respectively)[17] with higher
molar absorptions. The band gaps estimated from the onsets
of their absorption spectra were, however, smaller. These re-
sults agreed well with the data obtained from the theoretical
calculations that 1a–1c possessed higher conjugation. Due
to steric hindrance, the phenyl groups at the 2,5-positions of
HPS, MPPS, and DMTPS, were twisted out of the plane of
To prove that aggregates of 1a–1c were formed when
large amounts of water was present in their THF solutions,
we carried out particle size analysis. The aggregates of 1a
were nanodimensional and their average diameters de-
creased from 441 to 42 nm when the water fraction in the
mixture was increased from 70 to 99.5% (Figure S5a in the
Supporting Information). The particle size was further con-
firmed by transmission electron microscopy (TEM). An ex-
ample of a TEM image of nanoaggregates of 1a in a THF/
water mixture with 90% water is given in Figure S5. Similar
phenomena were observed in our previous work.[34] In mix-
tures with higher water contents, the molecules might ag-
glomerate quickly to form smaller aggregates, while at lower
water contents, the molecules of 1a might slowly assemble
in an ordered fashion to form larger clusters.
Polarized emission is generally originated from an or-
dered alignment of molecules in the solid state. However,
the strong p–p interactions in the aggregated and crystal
Figure 3. UV/Vis absorption spectra of 1a–1c in THF solutions.
6004
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 5998 – 6008

Siloles
FULL PAPER
crystalline CdSe nanorods, highly fluorescent inorganic sem-
iconductors.[35]
Self-assembly: Although we failed to obtain crystals of 1c,
we found that it exhibited a facile one-dimensional self-as-
sembling ability. When a poor solvent, such as methanol,
was added slowly into its dichloromethane solution
(1 mgmLꢀ1) in a glass tube, thin 1D nanorods started to
form at the solvent interface. Figures 6A and 6B show the
scanning electron microscopy (SEM) images of the fibrils
obtained from the methanol/dichloromethane mixture. The
fibrils were elongated in shape with diameters of ꢂ50 nm
and lengths of several tens of micrometers (Figure 6A).
After standing at ambient conditions for several days, more
nanocables were formed and their association afforded fi-
brils with micrometer sizes (Figure 6B). Bundles of 1c were
formed by further gathering of microrods. It is worth noting
that parallel bands normal to the director of the microrods
were clearly seen, suggesting that the microfibrils were con-
structed by side-by-side self-assembling of the nanorods.
Analysis by TEM showed that the ropes possessed a uni-
form inner structure, as shown in Figure S6 in the Support-
ing Information.
After solvent evaporation, large gatherings of microrods
were obtained at the bottom of the tube. The majority of
the rods were arranged uniformly in a parallel fashion.
Analysis by powder X-ray diffraction showed that these mi-
crorods were crystalline in nature (Figure S7 in the Support-
ing Information). Irrespective of the size, all of the cables
emitted intensive green light upon photoexcitation, especial-
ly at the cross sections (Figure 6C). It is truly amazing that
the silole molecules can spontaneously self-assemble into
materials with well-organized morphological structures by
simple evaporation methods. Together with their strong light
emission, such materials are promising candidates in the fab-
rication of miniature electronic and photonic devices.
Figure 4. a) PL spectra of 1a in THF/water mixtures with different water
fractions; concentration: 10mm; excitation wavelength: 370 nm. b) Plots
of (I/I0)ꢀ1 versus water fractions in THF/water mixtures, in which I0 was
the PL intensity in pure THF.
states of most conventional luminophores significantly
quench their PL. Due to their AIE characteristics, the ag-
gregates and solid films of 1a–1c can emit intense light
upon photoexcitation. It would
be even better if the silole mol-
ecules could exhibit polarized
emission. Therefore, we investi-
gated the PL of crystals of 1a
and 1b in vertical and horizon-
tal directions. Interestingly, the
crystals of 1a and 1b showed
emission anisotropy with strong
PL intensity in the vertical di-
rection (Figures 5a and 5b).
The reproducible polarization
ratios (Iv/Ih)
were 4.49 and 4.75, and the de-
grees
of
polarization
[P=
N
ACHTUNGTRENNUNG(Iv+Ih)] were calcu-
lated to be 0.64 and 0.65, re-
spectively. These values, al-
though preliminary, were quite
Figure 5. Polarized emissions from crystals of a) 1a and b) 1b and c) microrods of 1c. Abbreviations: Iv =PL
intensity in the vertical direction, Ih =PL intensity in the horizontal direction, P=degree of polarization calcu-
high and close to that of single lated by (IvꢀIh)/
G
ACHTUGNRTNE(NUGN Iv+2Ih).
Chem. Eur. J. 2011, 17, 5998 – 6008
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6005

B. Z. Tang et al.
Figure 7. Voltage dependence on the transient EL of 1a with a device
configuration of ITO/NPB/1a/Al with the applied rectangular voltage
pulse.
Figure 6. A) and B) SEM micrographs of fibrils of 1c obtained from
methanol/dichloromethane mixtures; C) and D) fluorescent micrographs
of microrods of 1c.
Figure 8 shows the change in the electron mobility of 1a
and 1b against the square root of the bias field. The data for
HPS and MPPS are also included in Figure 8 for compari-
son. All of the siloles showed higher mobilities at higher
electric fields. This kind of field-dependence phenomenon is
typical in organic EL materials. The electron mobility of 1b
was 5.0ꢀ10ꢀ6–1.2ꢀ10ꢀ5 cm2 Vꢀ1 sꢀ1 and 1a was 3.1ꢀ10ꢀ6–
6.3ꢀ10ꢀ6 cm2 Vꢀ1 sꢀ1. The mobility of 1a and 1b, particularly
1b, was steadily increased with increasing field strength. The
value for 1c was comparatively low (2.0ꢀ10ꢀ6–5.2ꢀ
10ꢀ6 cm2 Vꢀ1 sꢀ1), which is only about half of that for 1b. The
electron mobility of 1a and 1b was superior to HPS,
MPPS,[17] and Alq3, a commonly used electron-transport ma-
terial with an electron mobility of 2.3ꢀ10ꢀ6 cm2 Vꢀ1 sꢀ1
under the same conditions.[37] Recent reports showed that
the excellent electron-transport properties of tetraphenylsi-
We measured the PL of these microrods to get some pre-
liminary information about their internal structures. The
emission peak was located at 488 nm, similar to the emission
in THF. The ordered stacking of nanorods into fibrils with
macroscopic sizes should make the molecular conformation
more rigid and restrict the rotation of the phenyl rings.
Thus, it was anticipated that the cables would efficiently
emit and show polarized emission. The absolute FF value of
the microrods was 26.8%, which was more than twofold
higher than that of the solid film and comparable to some of
the highly emissive organic nanowires previously report-
ed.[36] The polarized PL spectrum of the microrods of 1c is
shown in Figure 5c. The emission polarization in the micro-
rods was 2.2. Taking into the account that the microrods
were arranged in random directions, we believed this result
demonstrates a reasonably high degree of molecular order.
Electron mobility: Our previous studies revealed that HPS
and MPPS possessed higher electron mobility than Alq3.[17]
Calculation showed that 1a–1c were more conjugated with
lower LUMO energy levels and larger reorganization ener-
gies than HPS and MPPS; therefore, they might exhibit
even higher electron-transport abilities. Thus, we investigat-
ed the electron mobility of 1a–1c using the transient elec-
troluminescence (EL) method. Figure 7 shows the voltage
dependence on the transient EL of ITO/NPB)/1a/Al (ITO=
indium tin oxide, NPB=N,Nꢃ-bis(1-naphthyl)-N,Nꢃ-diphe-
nylbenzidine) device with the applied rectangular voltage
pulse. With an increase in the applied voltage, the delay
time (td) became shortened and a steep rise in the EL was
observed. The td value was lowered because the charge car-
rier took less time to travel across the organic layers. This
also meant that the carrier mobility of the organic molecules
was enhanced with increasing electrical field.
Figure 8. Plot of electron mobility of 1a and 1b versus the square root of
the bias field. Data for HPS and MPPS are shown for comparison.
6006
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 5998 – 6008

Siloles
FULL PAPER
loles stem from their large electron affinities, which origi-
nate from the s*–p* conjugation, and the stability of the re-
sultant anionic species.[2a] These factors might also be possi-
ble reasons for the high mobility of 1a and 1b. Noteworthy
is that the experimental values for all of the siloles as well
as Alq3 are quite different from their calculated values,
which demonstrates that besides molecular reorganization
energy, the intermolecular electronic coupling, packing pat-
tern, and morphology also play important roles in determin-
ing the electron mobility of a material.[9c] Although the re-
sults obtained above are preliminary, it shows that the new
2,5-bis(triphenylsilylethynyl)-3,4-diphenylsiloles are promis-
ing electron-transport materials for electronic applications.
Acknowledgements
We thank the Research Grants Council of Hong Kong (603509, 601608,
CUHK2/CRF/08, and HKUST2/CRF/10), the Innovation and Technology
Commission (ITP/008/09NP and ITS/168/09), the University Grants Com-
mittee of Hong Kong (AoE/P-03/08), and the National Science Founda-
tion of China (20974028). B.Z.T. is grateful to Cao Gaungbiao Founda-
tion of Zhejiang University.
[1] a) K. Tamao, S. Yamaguchi, M. Shiro, J. Am. Chem. Soc. 1994, 116,
11715; b) K. Tamao, S. Yamaguchi, Y. Ito, Y. Matsuzaki, T. Yamabe,
M. Fukushima, Macromolecules 1995, 28, 8668; c) K. Tamao, S.
Ohno, S. Yamaguchi, Chem. Commun. 1996, 1873; d) S. Yamaguchi,
K. Tamao, J. Chem. Soc. Dalton Trans. 1998, 3693; e) L. H. Chan,
R. H. Lee, C. F. Hsieh, H. C. Yeh, C. T. Chen, J. Am. Chem. Soc.
2002, 124, 6469; f) L. H. Chan, H. C. Yeh, C. T. Chen, Adv. Mater.
2001, 13, 1637.
Conclusion
[2] a) H. Murata, G. G. Malliaras, M. Uchida, Y. Shen, Z. H. Kafafi,
Chem. Phys. Lett. 2001, 339, 161; b) A. J. Mꢄkinen, M. Uchida, Z. H.
Kafafi, J. Appl. Phys. 2004, 95, 2832.
In this paper, new 2,5-bis(triphenylsilylethynyl)-3,4-diphe-
nylsiloles were synthesized in high yields by Sonogashira
coupling of 2,5-dibromo-3,4-diphenylsiloles with triphenylsi-
lylacetylene. Single-crystal X-ray diffraction and theoretical
calculations showed that when compared with 2,5-diarylsi-
loles, the present molecules were less twisted because the
steric hindrance between the triphenylsilyl units and silole
cores was lessened by the acetylene linkages. The electron-
withdrawing effect of the triphenylsilylethynyl groups gave
1a–1c low HOMO and LUMO energy levels. Analysis by
cyclic voltammetry revealed that the silole molecules exhib-
ited low electron affinities, which were comparable to those
of perfluoroarylsiloles. Calculations of the reorganization
energies for electron-transfers showed that the 2,5-bis(tri-
phenylsilylethynyl)-3,4-diphenylsiloles had higher electron
mobilities than HPS, MPPS, and Alq3. Values up to 1.2ꢀ
10ꢀ5 cm2 Vꢀ1 sꢀ1 were measured by the transient EL method.
All of the siloles were thermally stable, losing little of their
weight when heated up to 4008C. Although their solutions
were weakly emissive, their nanoaggregates in poor solvents
and thin films emitted intense blue-green light upon photo-
excitation. The bulky triphenylsilylethynyl groups had effi-
ciently suppressed the close packing of the silole molecules
in the condensed phase and, hence, prevented a redshift in
the emission spectra as observed in 2,5-bis(trialkysilylethy-
nyl)siloles.[16c] Polarized emission was observed in the crys-
talline aggregates of 1a–1c. The addition of poor solvents to
their THF solutions self-assembled their molecules, evapora-
tion of which generated macroscopic fibrils with efficient
emissions. With these novel properties, similar to the previ-
ously prepared AIE molecules such as HPS and MPPS, the
new silole molecules are anticipated to find an array of ap-
plications in electronics and optics.
[3] a) K. Tamao, M. Uchida, T. Izumizawa, K. Fukukawa, S. Yamaguchi,
J. Am. Chem. Soc. 1996, 118, 11974; b) M. Uchida, T. Izumizawa, T.
Nakano, S. Yamaguchi, K. Tamao, K. Furukawa, Chem. Mater. 2001,
13, 2680; c) H. Murata, Z. H. Kafafi, M. Uchida, Appl. Phys. Lett.
2002, 80, 189.
[4] a) Y. Liu, Z. Chen, J. Chen, F. Wang, Y. Cao, Polym. Bull. 2007, 59,
31; b) K. Geramita, J. McBee, Y. Shen, N. Radu, T. D. Tilley, Chem.
Mater. 2006, 18, 3261; c) H.-J. Son, W.-S. Han, J.-Y. Chun, C.-J. Lee,
J.-I. Han, J. Ko, S. O. Kang, Organometallics 2007, 26, 519; d) B. Z.
Tang, X. Zhan, G. Yu, P. P. S. Lee, Y. Liu, D. Zhu, J. Mater. Chem.
2001, 11, 2974; e) Z. Zhao, S. Chen, J. W. Y. Lam, . K. W. Jim,
C. Y. K. Chan, Z. Wang, P. Lu, H. S. Kwok, Y. Ma, B. Z. Tang, J.
Phys. Chem. C 2010, 114, 7963.
[5] a) H. Usta, G. Lu, A. Facchetti, T. J. Marks, J. Am. Chem. Soc. 2006,
128, 9034; b) K. Tamao, S. Yamaguchi, M. Shiozaki, Y. Nakagawa,
Y. Ito, J. Am. Chem. Soc. 1992, 114, 5867; c) U. Salzner, J. B. Lagow-
ski, P. G. Pickup, R. A. Poirier, Synth. Met. 1998, 96, 177.
[6] a) Y. Dong, J. W. Y. Lam, A. Qin, Z. Li, J. Liu, J. Sun, Y. Dong,
B. Z. Tang, Chem. Phys. Lett. 2007, 446, 124; b) Z. Li, Y. Dong,
J. W. Y. Lam, J. Sun, A. Qin, M. Hꢄußler, Y. Dong, H. H. Y. Sung,
I. D. Williams, H. S. Kwok, B. Z. Tang, Adv. Funct. Mater. 2009, 19,
905; c) Y. Dong, J. W. Y. Lam, A. Qin, Z. Li, J. Sun, Y. Dong, B. Z.
Tang, J. Inorg. Organomet. Polym. 2007, 17, 673; d) J. C. Sanchez,
A. G. DiPasquale, A. L. Rheingold, W. C. Trogler, Chem. Mater.
2007, 19, 6459; e) H. Sohn, R. M. Calhoun, M. J. Sailor, W. C. Tro-
gler, Angew. Chem. 2001, 113, 2162; Angew. Chem. Int. Ed. 2001, 40,
2104; f) L. Peng, M. Wang, G. Zhang, D. Zhang, D. Zhu, Org. Lett.
2009, 11, 1943.
[7] a) B. Mi, Y. Dong, Z. Li, J. W. Y. Lam, M. Hꢄußler, H. H. Y. Sung,
H. S. Kwok, Y. Dong, I. D. Williams, Y. Liu, Y. Luo, Z. Shuai, D.
Zhu, B. Z. Tang, Chem. Commun. 2005, 3583; b) F. Wang, J. Luo, K.
Yang, J. Chen, F. Huang, Y. Cao, Macromolecules 2005, 38, 2253.
[8] a) C. Booker, X. Wang, S. Haroun, J. Zhou, M. Jennings, B. L. Pa-
genkopf, Z. Ding, Angew. Chem. 2008, 120, 7845; Angew. Chem. Int.
Ed. 2008, 47, 7731; b) M. M. Sartin, A. J. Boydston, B. L. Pagenkopf,
A. J. Bard, J. Am. Chem. Soc. 2006, 128, 10163.
[9] a) S. Yamaguchi, Y. Itami, K. Tamao, Organometallics 1998, 17,
4910; b) C. Risko, G. P. Kushto, Z. H. Kafafi, J.-L. Brꢅdas, J. Chem.
Phys. 2004, 121, 9031; c) X. Zhan, C. Risko, F. Amy, C. Chan, W.
Zhao, S. Barlow, A. Kahn, J. L. Brꢅdas, S. R. Marder, J. Am. Chem.
Soc. 2005, 127, 9021; d) X. Zhan, C. Risko, A. Korlyukov, F. Sena,
T. V. Timofeeva, M. Y. Antipin, S. Barlow, J.-L. Brꢅdas, S. R.
Marder, J. Mater. Chem. 2006, 16, 3814; e) N. Huby, L. Hirsch, L.
Aubouy, P. Gerbier, A. Van Der Lee, F. Amy, A. Kahn, Phys. Rev. B
2007, 75, 115416; f) S. Yamaguchi, T. Endo, M. Uchida, T. Izumiza-
wa, K. Furukawa, K. Tamao, Chem. Eur. J. 2000, 6, 1683.
Experimental Section
[10] a) T. Shimizu, M. Masuda, H. Minamikawa, Chem. Rev. 2005, 105,
1401; b) P. X. Gao, Y. Ding, W. Mai, W. L. Hughes, C. S. Lao, Z. L.
For experimental details, please see the Supporting Information.
Chem. Eur. J. 2011, 17, 5998 – 6008
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6007

B. Z. Tang et al.
Wang, Science 2005, 309, 1700; c) X. Y. Kong, Y. Ding, R. S. Yang,
Z. L. Wang, Science 2004, 303, 1348; d) D. Y. Yan, Y. F. Zhou, J.
Hou, Science 2004, 303, 65.
[22] S. Yamaguchi, K. Tamao, Bull. Chem. Soc. Jpn. 1996, 69, 2327.
[23] S. Yamaguchi, R. Z. Jin, K. Tamao, J. Organomet. Chem. 1998, 559,
73.
[11] a) F. J. M. Hoeben, P. Jonkheijm, E. W. Meijer, A. P. H. J. Schenning,
Chem. Rev. 2005, 105, 1491; b) Z. Zhao, J.-H. Li, X. Chen, P. Lu, Y.
Yang, Org. Lett. 2008, 10, 3041; c) X.-Q. Li, X. Zhang, S. Ghosh, F.
Werthner, Chem. Eur. J. 2008, 14, 8074.
[24] a) Z. Zhao, P. Lu, J. W. Y. Lam, B. Yang, Y. Ma, B. Z. Tang, Sci.
China Ser. B 2010, 53, 2311; b) J. Ohshita, N. Mimura, H. Arase, M.
Nodono, Macromolecules 1998, 31, 7985.
[25] A. J. Mꢄkinen, M. Uchida, Z. H. Kafafi, J. Appl. Phys. 2004, 95,
2832.
[12] L. Zang, Y. Che, J. S. Moore, Acc. Chem. Res. 2008, 41, 1596.
[13] a) X. Zhang, B. Sun, J. Phys. Chem. B 2007, 111, 10881; b) C. A.
Breen, T. Deng, T. Breiner, E. L. Thomas, T. M. Swager, J. Am.
Chem. Soc. 2003, 125, 9942; c) A. Datar, K. Balakrishnan, X. Yang,
X. Zuo, J. Huang, R. Oitker, M. Yen, J. Zhao, D. M. Tiede, L. Zang,
J. Phys. Chem. B 2006, 110, 12327; d) L. Yang, H. Peng, K. Huang,
J. T. Mague, H. Li, Y. Lu, Adv. Funct. Mater. 2008, 18, 1526.
[14] a) J. W. Chung, B.-K. An, S. Y. Park, Chem. Mater. 2008, 20, 6750;
b) J. Seo, J. W. Chung, E.-H. Jo, S. Y. Park, Chem. Commun. 2008,
2794; c) X. Tong, Y. Zhao, B.-K. An, S. Y. Park, Adv. Funct. Mater.
2006, 16, 1799; d) J. W. Chung, B.-K. An, J. W. Kim, J.-J. Kim, S. Y.
Park, Chem. Commun. 2008, 2998; e) T. H. Kim, M. S. Choi, B.-H.
Sohn, S. Y. Park, W. S. Lyood, T. S. Lee, Chem. Commun. 2008,
2364.
[15] a) J. Luo, Z. Xie, J. W. Y. Lam, L. Cheng, H. Chen, C. Qiu, H. S.
Kwok, X. Zhan, Y. Liu, D. Zhu, B. Z. Tang, Chem. Commun. 2001,
1740; b) J. Chen, C. C. W. Lam, J. W. Y. Lam, Y. Dong, S. M. F. Lo,
I. D. Williams, D. Zhu, B. Z. Tang, Chem. Mater. 2003, 15, 1535.
[16] a) Z. Li, Y. Dong, B. Mi, Y. Tang, M. Hꢄussler, H. Tong, P. Dong,
J. W. Y. Lam, Y. Ren, H. H. Y. Sun, K. Wong, P. Gao, I. D. Williams,
H. S. Kwok, B. Z. Tang, J. Phys. Chem. B 2005, 109, 10061; b) Y. Q.
Dong, J. W. Y. Lam, Z. Li, H. Tong, Y. P. Dong, X. D. Feng, B. Z.
Tang, J. Inorg. Organomet. Polym. Mater. 2005, 15, 287; c) Z. Zhao,
Z. Wang, P. Lu, C. Y. K. Chan, D. Liu, J. W. Y. Lam, H. H. Y. Sung,
I. D. Williams, Y. Ma, B. Z. Tang, Angew. Chem. 2009, 121, 7744;
Angew. Chem. Int. Ed. 2009, 48, 7608.
[26] J. Pommerehne, H. Vestweber, W. Guss, R. F. Mahrt, H. Bassler, M.
Porsch, J. Daub, Adv. Mater. 1995, 7, 551.
[27] J. D. Anderson, E. M. McDonald, P. A. Lee, M. L. Anderson, E. L.
Ritchie, H. K. Hall, T. Hopkins, E. A. Mash, J. Wang, A. Padias, S.
Thayumanavan, S. Barlow, S. R. Marder, G. E. Jabbour, S. Shaheen,
B. Kippelen, N. Peyghambarian, R. M. Wightman, N. R. Armstrong,
J. Am. Chem. Soc. 1998, 120, 9646.
[28] a) W. Y. Gao, A. Kahn, J. Appl. Phys. 2003, 94, 359; b) D. Cahen, A.
Kahn, Adv. Mater. 2003, 15, 271.
[29] a) A. Facchetti, M. H. Yoon, C. L. Stern, H. E. Katz, T. Marks,
Angew. Chem. 2003, 115, 4030; Angew. Chem. Int. Ed. 2003, 42,
3900; b) B. L. Lee, T. Yamamoto, Macromolecules 1999, 32, 1375;
c) Y. Zhu, M. M. Alam, S. A. Jenekhe, Macromolecules 2003, 36,
8958.
[30] a) R. A. Marcus, J. Chem. Phys. 1956, 24, 966; b) R. A. Marcus, Rev.
Mod. Phys. 1993, 65, 599; c) R. A. Marcus, H. Eyring, Annu. Rev.
Phys. Chem. 1964, 15, 155.
[31] B. C. Lin, C. P. Cheng, Z.-Q. You, C.-P. Hsu, J. Am. Chem. Soc.
2005, 127, 66.
[32] N. E. Gruhn, D. A. da Silva, T. G. Bill, M. Malagoli, V. Coropceanu,
A. Kahn, J. L. Brꢅdas, J. Am. Chem. Soc. 2002, 124, 7918.
[33] a) M. Malagoli, J. L. Breꢅdas, Chem. Phys. Lett. 2000, 327, 13;
b) B. C. Lin, C. P. Cheng, Z. P. M. Lao, J. Phys. Chem. A 2003, 107,
5241.
[34] Y. Yu, Y. Hong, C. Feng, J. Liu, J. W. Y. Lam, M. Faisal, K. M. Ng,
K. Q. Luo, B. Z. Tang, Sci. China Ser. B 2008, 52, 15.
[35] J. Hu, L.-S. Li, W. Yang, L. Manna, L.-W. Wang, A. P. Alivisatos,
Science 2001, 292, 2060.
[17] G. Yu, S. Yin, Y. Liu, J. Chen, X. Xu, X. Sun, D. Ma, X. Zhan, Q.
Peng, Z. Shuai, B. Z. Tang, D. Zhu, W. Fang, Y. Luo, J. Am. Chem.
Soc. 2005, 127, 6335.
[18] H. Y. Chen, J. W. Lam, J. D. Luo, Y. L. Ho, B. Z. Tang, D. Zhu, M.
Wong, H. S. Kwok, Appl. Phys. Lett. 2002, 81, 574.
[36] a) S. S. Babu, V. K. Praveen, S. Prasanthkumar, A. Ajayaghosh,
Chem. Eur. J. 2008, 14, 9577; b) T. Naddo, Y. Che, W. Zhang, K. Ba-
lakrishnan, X. Yang, M. Yen, J. Zhao, J. S. Moore, L. Zang, J. Am.
Chem. Soc. 2007, 129, 6978; c) J. Luo, T. Lei, L. Wang, Y. Ma, Y.
Cao, J. Wang, J. Pei, J. Am. Chem. Soc. 2009, 131, 2076.
[37] D. G. Ma, G. Wang, Y. F. Hu, Y. G. Zhang, L. X. Wang, X. B. Jing,
F. S. Wang, Appl. Phys. Lett. 2003, 82, 1296.
[19] a) A. J. Boydston, Y. Yin, B. L. Pagenkopf, J. Am. Chem. Soc. 2004,
126, 10350; b) A. J. Boydston, Y. Yin, B. L. Pagenkopf, J. Am.
Chem. Soc. 2004, 126, 3724; c) A. J. Boydston, B. L. Pagenkopf,
Angew. Chem. 2004, 116, 6496; Angew. Chem. Int. Ed. 2004, 43,
6336.
[20] a) C. Janiak, J. Chem. Soc. Dalton Trans. 2000, 3885; b) A. Qin,
C. K. W. Jim, Y. Tang, J. W. Y. Lam, J. Liu, F. Mahtab, P. Gao, B. Z.
Tang, J. Phys. Chem. B 2008, 112, 9281.
[21] V. N. Khabashesku, V. Balaji, S. E. Boganov, O. M. Nefedov, J.
Michl, J. Am. Chem. Soc. 1994, 116, 320.
Received: November 24, 2010
Published online: April 13, 2011
6008
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 5998 – 6008