Synthesis and optical properties of perylene diimide derivatives with triphenylene-based dendrons linked at the bay positions through a conjugated ethynyl linkage

Article abstract of DOI:10.1016/j.tet.2012.02.008

A number of groups including trimethylsilyl, phenyl, triphenylene, and triphenylene-based dendron have been linked to the bay positions of a perylene diimide (PDI) core through an ethynyl bridge. The photophysical properties of the resulting bay-substitut

Full text of DOI:10.1016/j.tet.2012.02.008

Tetrahedron 68 (2012) 2806e2818
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Synthesis and optical properties of perylene diimide derivatives with
triphenylene-based dendrons linked at the bay positions through a conjugated
ethynyl linkage
Mahuya Bagui a, Tanmoy Dutta a, Haizhen Zhong b, Shaohua Li a, Sanjiban Chakraborty a,
,
Andrew Keightley c, Zhonghua Peng a
*
a Department of Chemistry, University of Missouri-Kansas City, Kansas City, MO 64110, USA
b Department of Chemistry, University of Nebraska-Omaha, Omaha, NE 68162, USA
c School of Biological Sciences, University of Missouri-Kansas City, Kansas City, MO 64110, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A number of groups including trimethylsilyl, phenyl, triphenylene, and triphenylene-based dendron have
been linked to the bay positions of a perylene diimide (PDI) core through an ethynyl bridge. The pho-
tophysical properties of the resulting bay-substituted PDI derivatives have been carefully studied in
different solvents and as thin films. Without any capping group, the two ethynyl bay-substituted PDI
derivates PAT and PRT both aggregate strongly even in dilute solutions but in different perylene-perylene
Received 20 October 2011
Received in revised form 3 February 2012
Accepted 6 February 2012
Available online 10 February 2012
pep stacking modes; PRT aggregates through slipped (or longitudinal) stacking while PAT self-
Keywords:
assembles by rotational (or cross) stacking. With capping groups, the perylene core stacking is com-
pletely blocked for PATS in both solution and solid film. For PRTS, the slipped stacking is observed only
for its film sample, while for PTB, association only occurs after excitation (excimer formation). When
triphenylene or triphenylene-based G1 dendron is attached to the acetylene bridge, the resulting do-
noreacceptor systems (PTG0 and PTG1) exhibit strong electronic coupling between the dendritic donors
and the PDI acceptor, leading to significantly red-shifted absorption bands. The conjugated linkage also
facilitates photoinduced electron transfer from the triphenylene or triphenylene dendron to the PDI core,
effectively quenching fluorescence emissions of both the donor and the acceptor. The significantly red-
shifted absorption bands and the efficient photoinduced electron transfer observed on PTG0 and PTG1
indicate that these new PDI derivatives may find applications in solar cells.
Perylenediimide
Triphenylene
Charge transfer
Dendrimer
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
existence of a node plane at the imido position.18 Substituents at
the bay positions,19e25 particularly those with
p-conjugated link-
Perylene diimide (PDI) derivatives are one of the most exten-
sively studied n-type organic semiconductors, not only due to their
strong electron accepting1e3 and fast electron transporting prop-
erties,4,5 but also due to their remarkable chemical and thermal
stability.6e8 An attractive feature of the PDI core is that structural
modifications at either its bay positions and/or the imide positions
can be readily accomplished, enabling the optimization of its
properties.9 One property often targeted for improvement is its
absorption wavelength range. For applications in certain molecular
electronic devices such as sensors,10e13 photovoltaic cells,12,14e18 it
is often desirable to extend its absorption to longer wavelength
regions. For this purpose, substitution at the bay positions is fa-
vored since substituents at the imide positions have negligible
ground state electronic interaction with the PDI core due to the
ages,26e30 however could effectively modulate electronic states of
the PDI core.
Among various conjugated linkages, the acetylene bridge is
found to be quite effective in modulating the PDI core’s bandgap, as
the carbonecarbon triple bond linkage imposes low steric hin-
drance when tethered to the bay positions and thus is favorable for
creating extended conjugated scaffold.31 The extension of
p-con-
jugation from the PDI core through the acetylene bridge to attached
electron donors, such as porphyrin,32 phthalocyanine,33 and aryl-
amine,34 leads to ultrafast photoinduced intramolecular charge
separation. While a number of
p-conjugated systems have been
anchored to the bay positions of the PDI core through the acetylene
linkage, there is still lack of systematic study in regard to the cor-
relation between the extent of extended
p-conjugation and the
chromophore’s bandgap. The aggregation behavior of such
p
-ex-
tended PDI systems in solid states is also not well understood. In
this contribution, we report the synthesis of a set of PDI derivatives
* Corresponding author. E-mail address: pengz@umkc.edu (Z. Peng).
0040-4020/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2012.02.008

M. Bagui et al. / Tetrahedron 68 (2012) 2806e2818
2807
with
p
-conjugation progressively extended at PDI’s bay positions. A
mixture containing 1,6- and 1,7-regioisomers was synthesized
according to a literature procedure.35 Without separating the two
regioisomers due to their poor solubility in common organic sol-
vents, the mixture was reacted with hexylamine in 1:1 water/1-
butanol solvent mixture, converting the dianhydride to a diimide.
The isolation of the 1,7-isomer from the 1,6-isomer using conven-
tional column chromatography or recrystallization techniques us-
ing common organic solvents was unsuccessful. The mixture was
thus coupled with triisopropylsilyl (TIPS) acetylene to yield PRTS
initially as a 1,6- and 1,7-regioisomeric mixture. At this stage, the
pure 1,7-isomer was isolated from the 1,6-isomer by column
chromatography. The bulky TIPS group appears to limit PDI core
aggregation and makes the separation of the regioisomers easier.
Previously it has been reported that the triphenylsilyl group can
also be used for the straightforward separation of the 1,7-isomer
from the 1,6-isomer by column chromatography.32 It should be
noted that the trimethylsilyl-analog of PRTS does not allow iso-
meric separation using similar separation techniques. Clearly the
bulkiness of ethynyl’s capping groups is critical for isomer separa-
tion. The TIPS protected 1,7-diethynyl PDI (PRTS) was desilylated
systematic photophysical studies on both their solutions and solid
films have been carried out. It is shown that substitution at the bay
area not only affects the p-extension, but also has a profound effect
on the mode of PDI core aggregation.
2. Results and discussions
2.1. Synthesis of PDI derivatives
Chart 1 list the structures of PDI derivatives subjected to this
study. Compounds PRT and PRTS have a linear hexyl chain at the
imido position while the rest of the compounds possess a 2,6-
diisopropylphenyl group at the imido position. All seven com-
pounds have an ethynyl group at the 1, 7-bay positions of the PDI
core. The PDI’s p-system is extended through the acetylene bridge
to benzene (PTB), triphenylene (PTG0), and a triphenylene dendron
(PTG1).
The synthetic approaches to the above PDI derivatives are
shown in Schemes 1 and 2. A dibrominated perylene dianhydride
Chart 1. List of structures of bay-substituted PDI derivatives.
Scheme 1. Synthesis of PRTS and PRT.

2808
M. Bagui et al. / Tetrahedron 68 (2012) 2806e2818
Scheme 2. Synthesis of PDI derivatives PATS, PAT, PTB, PTG0, PTG1.
with tetrabutylammonium fluoride, leading to compound PRT in
45% yield.
assigned to their respective structures (See Figs. S1eS3 in Supple-
mentary data). Compared to the 1H NMR spectrum of PBr (Fig. S3),
one observes that when the Br group of PBr is replaced with an
acetylene bridge, proton H1 (see Fig. 1 for proton labeling) shifts
downfield by more than 0.7 ppm while protons H2 and H3 shows
negligible changes in their chemical shifts. The significant downfield
shift of H1 is likely due to the deshielding effect of the adjacent
When 2,6-diisopropylphenyl, instead of the n-hexyl group, was
introduced to the imido positions, the pure 1,7-dibromo-PDI de-
rivative (PBr) can be isolated by simple recrystallization (Scheme
2).36 The pure PBr was subjected to the Sonogashira coupling re-
action with trimethylsilylacetylene to yield compound PATS in 65%
yields, which was then converted to PAT by routine desilylation.
Our initial attempt toward PTG0 and PTG1 synthesis involves
the reaction of PAT or PATS with triflate-functionalized tripheny-
lene (G0OTf) and triphenylene-based dendron G1OTf (replacing
the ethynyl group of G1T with a triflate gives G1OTf) since both
G0Tf and G1OTf had previously been synthesized during our re-
search on triphenylene-based dendrons.37 Unfortunately, Pd-
catalyzed coupling reactions between PAT and G0OTf, PATS and
G0OTf were both unsuccessful, presumably due to the low re-
activity of the triphenylene-attached triflate group. The synthetic
focus was then shifted to utilize PBr as the coupling reagent. For
this purpose, ethynyl-functionalized triphenylene (G0T) and
triphenylene-based G1 dendron (G1T) were synthesized from
G0OTf and G1Tf, respectively.37 Direct Sonogashira coupling be-
tween PBr and ethynylbenzene, G0T, or G1T successfully yielded
PTB, PTG0, and PTG1, respectively. PTG0 and PTG1 are only par-
tially soluble in some organic solvents, such as chloroform, THF, and
DMF, even at elevated temperatures. The poor solubility of the
product made the purification process challenging and led to the
low yields of isolated pure products.
acetylene
p electrons. The capping group at the end of the acetylene
bridge, whether a TMS in PATS, TIPS in PRTS or phenyl in PTB, ap-
pears to have minimal effect on the chemical shifts of the protons on
the PDI core.
When the TIPS group of PRTS is removed, the solubility of the
resulting compound PRT is significantly lowered. The 1H NMR
spectrum of PRT shows broad signals and is concentration de-
pendent. As shown in Fig. 1, when the concentration increases, the
aromatic proton signals are broadened and shift upfield. While
three aromatic proton signals are clear and well separated, ex-
tremely broad humps in both the aromatic and aliphatic regions are
discernible. Even for a highly diluted solution where the three ar-
omatic signals are sharp and their multiplicities are clear, the broad
hump covering the entire aromatic region is still noticeable. These
results indicate that for PRT there is strong inter-perylene
pep
stacking, even in highly diluted conditions.
Unlike PRT, the 1H NMR spectrum of PAT does not show the
broad humps, but rather two sets of perylene proton signals
(Fig. S4), one sharp and one broad. The broad signal of H1 is shifted
upfield while those of H2 and H3 are shifted downfield, when
compared to their respective sharp signals. The protons on the
phenyl ring, however, give only sharp, multiplicity-clear signals, i.e.,
no broad signals. These results indicate that the perylene core of
2.2. Structural characterizations
Among the seven PDI derivatives, PRTS, PATS, and PTB are sol-
uble in common organic solvents and their 1H NMR spectra show
well-resolved sharp signals, which can all be unambiguously
PAT
due to the bulkiness of the 2,6-diisopropylphenyl rings, which
prevent parallel inter-perylene -stacking. When two perylene
p stack as well, but perhaps through a rotational offset, likely
p

M. Bagui et al. / Tetrahedron 68 (2012) 2806e2818
2809
Fig. 1. 1H NMR spectra of PRTS and PRT in CDCl3.
rings cross-stack (i.e., 90ꢀ rotational offset), the
p-electrons of one
2.3. Photophysical properties
perylene ring will have a shielding effect on H1 but deshielding
effect on H2 and H3, resulting in upfield shift for H1 but downfield
shift for H2 and H3. The lack of broad hump signals perhaps in-
The optical properties of all seven PDI derivatives, together with
PBr, G0T, and G1T, were studied in dilute chloroform solutions,
hexane solutions, and as solid films. The results are discussed in the
following sections.
dicates that the
p-stacking of PAT is limited to small aggregates (i.e.,
lower oligomers).
PTG0, albeit with two hexyloxy side chains on each tripheny-
lene ring, shows poor solubility in all solvents tested, even at el-
evated temperatures. The solubility is so poor that its 1H NMR
spectrum is dominated by solvent signals (Fig. S5). While one can
identify characteristic proton signals from both the perylene core
and the attached triphenylene units after expanding the spectrum,
the messy spectrum provides no proof of either the structure or
the purity of the product. Thin-layer chromatography (TLC),
however, confirms one single pure compound and MALDI-TOF
analysis shows only two peaks at 1672.96 and 1694.94 (Fig. S6),
which correspond to Mþ (1672.13) and MþNaþ (1695.12), re-
spectively. Elemental analysis corroborates the purity and struc-
ture of PTG0.
2.3.1. Steady-state absorption. Perylene and perylene diimide
compounds are well known organic dyes, which show intense
absorption in the visible spectral range. Without exception, the
seven PDI derivatives show rich absorption bands in the visible
range as well. As shown in Fig. 2, when the bay substitution changes
When a bulkier triphenylene-based G1 dendron is attached to
the acetylene bridge, the resulting PTG1 exhibits better solubility in
common organic solvents than PTG0. The 1H NMR spectrum of
PTG1 (Fig. S7) in CDCl3 shows broad but clear signals. Although
signals associated with the G1 dendron dominate the spectrum,
one can still clearly identify peaks attributed to the perylene core. It
is interesting to note that perylene’s H1 protons in PTG1 are further
downfield shifted from 10.2 ppm observed for PATS, PRTS, and PTB
(Fig. S3) to 10.6 ppm (Fig. S7), presumably reflecting a stronger
p-
conjugation which likely increases the -electron density of the
p
perylene core due to the electron-donating nature of the G1 den-
dron. The structure and purity of PTG1 is again confirmed by TLC
(one single spot) and MALDI-TOF analysis, which shows one single
peak in the 500e5000 range at 3702.70 (Fig. S6) corresponding to
MþAgþ (3702.55).
Fig. 2. UV/vis absorption spectra of PBr, PATS, PTB, PTG0, and PTG1 in dilute chlo-
roform solutions.

2810
M. Bagui et al. / Tetrahedron 68 (2012) 2806e2818
from Br to TMS-ethynyl, phenylethynyl, triphenylene-ethynyl, and
G1-ethynyl, all absorption bands continuously shift to longer
wavelengths. The longest wavelength absorption bands of PBr,
PATS, PTB, PTG0, and PTG1 are 528, 550, 566, 614, and 626 nm,
earlier, the PDI core in PAT likely adopts cross (rotational) pep
stacking. Theoretical calculation has shown that such rotational
aggregates can indeed lead to ‘a single broadened peak that spans
a 4000 cmꢁ1 range, devoid of structure’.34
respectively, indicating that the PDI core’s
extended through the acetylene bridge to the attached
the -conjugation expands, the extended -system gradually loses
some identity of the PDI core, which is reflected in their absorption
spectra where fine structures are diminished.
In addition to absorption bands in the visible range, PTG0 and
PTG1 also show even stronger absorptions in the 300e450 nm
range. PDI core shows only a very weak absorption band at around
400 nm (S0eS4 transition), while G0T exhibits a weak flattened
hump from around 320 nme370 nm. The intense peak at 358 nm of
PTG0 is clearly a new transition not seen in the absorption spec-
trum of a 2:1 G0T/PTB physical mixture (Fig. S8). This transition is
p
system is increasingly
p-units. As
2.3.2. Steady-state fluorescence. In dilute chloroform solutions, PDI
derivatives PBr, PATS, PRTS, and PTB are all highly fluorescent with
unity quantum yield. PAT, however, has a fluorescence quantum
yield of only 0.16 (lex¼512 nm) while PRT is non-fluorescent. The
significant fluorescence quenching of PAT and PRT is consistent
with their fore-stated aggregation behavior in solutions.
p
p
PTG0 and PTG1 are also poor fluorescence emitter. Although
conjugation extends over both the PDI core and the triphenylene
G0 or G1 dendron, there appears to be two types of extended
p-
p
-
systems in both PTG0 and PTG1, one roots from the PDI core, which
absorbs in the visible range and the other originates from the G0 or
G1 dendron which contributes to the absorption bands in the
300e450 nm range. Thus, depending on the excitation wavelength,
PTG0 and PTG1 may have two different excitons, which emit light
of different wavelengths, if excitonic energy and/or electron
transfer is not so efficient between them. It turns out though that
both PTG0 and PTG1 in chloroform show very weak fluorescence
emissions no matter what the excitation wavelength is. If G0 or G1
likely a
inating from the triphenylene ring. Due to covalent attachment to
the PDI core, the normally symmetry forbidden * transition of
the triphenylene’s system becomes symmetry allowed after
pep* transition involving a p-system extending and orig-
pep
p
extending beyond the acetylene bridge, leading to an intense ab-
sorption band at 358 nm. For PTG1, the intense absorption band at
382 nm can be assigned to the absorption of the G1 dendron as G1T
shows similar absorption bands in the 320e430 nm range (Fig. S9).
Unlike PATS and PRTS, as well as PBr and PTB, all of which show
typical PDI core absorptions with resolved multiple bands, the
absorption spectra of PAT and PRT in dilute chloroform solutions, as
shown in Fig. 3, show one broad structureless band in the visible
range and a long tail, features typical of ground state aggregates.
The ratio of the absorptivity at 550 nm (the 0e0 band) to that at
500 nm (the 0e1 band) is 2.7 for both PATS and PRTS, but only 1.1
for PAT and 0.8 for PRT; the significant decrease in this value is
again indicative of strong aggregation.38 While both PAT and PRT
appear to be prone to solution aggregation, there are noticeable
differences between their absorption behaviors. First of all, the
absorption ratio of I550/I500 is lower for PRT so much so that the
lmax of PRT is now around 500 nm. The solution spectrum of PRT
closely matches that of its film spectrum in the visible range, clearly
indicating strong aggregation of PRT even in a dilute solution,
which is consistent with NMR results. The film spectrum of PAT, on
the other hand, is broadened to longer wavelengths by about
15 nm, indicating PAT aggregation in solutions is not as strong as
that of PRT, again consistent with NMR results. As mentioned
is excited (lex<400 nm), both PTG0 and PTG1 emits weakly around
440 nm (Fig. 4a) with quantum yields less than 0.01. To confirm that
such emissions are not due to any possible residual unreacted G0T
or G1T, the fluorescence and excitation spectra of PTG1 and PTG0
are compared with those of G1T and G0T, respectively, in Fig. 4. G1T
is moderately fluorescent with an emission maximum at 427 nm
and a quantum yield of 0.3. The emission wavelength of PTG1
(442 nm) is red-shifted by 15 nm compared to that of G1T. The
excitation spectra of PTG1 and G1T also differ significantly. The
most striking difference is that the excitation spectrum of PTG1
shows a sharp and strong peak at 416 nm where G1T does not
absorb (Fig. 4b). When the fluorescence and excitation spectra of
PTG0 are compared with those of G0T, the differences are even
more dramatic: PTG0 shows 25 nm red-shifted emissions and red-
shifted excitation bands at 365 and 393 nm, which are missing in
the excitation spectrum of G0T (Fig. 4b). These results exclude the
observed weak PTG0 and PTG1 emissions as those of unreacted G0T
or G1T. The fact that the strongest excitation for PTG1 or PTG0
occurs at wavelengths where G1T or G0T does not even absorb
indicates that the exciton responsible for the 440 nm emission is
due to a
p-system, which extends beyond the G1 or G0 dendron,
the acetylene bridge and into at-least part of the PDI core.
If the PDI is excited, PTG1 in chloroform emits at 648 nm while
PTG0 shows no fluorescence. In hexane, though PTG0 does show
weak PDI fluorescence with a maximum emission wavelength at
607 nm and a quantum yield of 0.03. Details on solvent effects are
discussed later. Similar to their absorption spectra shown in Fig. 2,
continuous red-shift in the maximum emission wavelength is also
observed, as clearly shown in Fig. 5. The maximum emission
wavelengths for PBr, PATS, PTB, PTG0, and PTG1 are 545, 563, 590,
607 (in hexane), and 648 nm, respectively.
2.3.3. Solvent effect and thin film optical properties. The absorption
spectra of all seven PDI derivatives have been studied in hexane and
as thin films as well. Comparison of their optical properties in dif-
ferent solvents and as thin films provides some insight into their
self-assembly behavior.
First, the absorption, fluorescence emission, and excitation
spectra of PATS and PRTS are compared. As shown in Fig. 6, PATS
and PRTS show nearly identical solution absorption and fluores-
cence emission behavior: same absorption and emission wave-
lengths, same extent of blue shift when the solvent is changed from
chloroform to hexane. The absorption and emission spectra of their
Fig. 3. UV/vis absorption spectra of PAT (black) and PRT (red) in dilute chloroform
(dark line) and as thin films (light line).

M. Bagui et al. / Tetrahedron 68 (2012) 2806e2818
2811
Fig. 4. Fluorescence emission (a) and excitation (b) spectra of PTG1, G1T, PTG0, and G0T.
(Fig. S10). These data suggest that the perylene-perylene
pep
stacking in solid films is lacking or very limited for PATS but strong
for PRTS. The limited aggregation of PATS may be resulted from the
complete blocking effect of the TMS groups at the bay and the
diisopropylphenyl units at the imido ends. PRTS, on the other hand,
has only the bay positions blocked by the silyl groups while its two
imido ends are relatively open due to linear alky substitutions.
Thus, slipped stacking, forming J-aggregates, is possible, which may
account for the much red-shifted absorptions and emissions. It is
worth noting that PATS and PRTS bear different silyl groups. As
mentioned earlier, the TMS analog of PRTS could not be obtained in
pure isomeric form. Although one can synthesize the TIPS analog of
PATS, that compound is expected to show little or no aggregation,
just like PATS since TIPS is an even bulkier substituent.
Like PATS and PRTS, PTB is highly fluorescent in both chloro-
form and hexane. The optical properties of PTB also show similar
solvent dependence: when the solvent is changed from chloroform
to hexane, the maximum absorption wavelength lmax, maximum
emission wavelength lem, and maximum excitation wavelength lex
are all blue-shifted by about 14 nm. The emissions of the PTB film
however are quite unique where two distinct peaks, one at 605 nm
and the other at 675 nm are observed as clearly shown in Fig. 7. The
excitation spectra of these two emission peaks are nearly identical
and closely match that of PTB’s chloroform solution (Fig. 7, center).
The absorption spectra of the film and the chloroform solution also
resemble each other except for around 10 nm red-shift in the ab-
sorption band edge. Taking these data together, one may conclude
that the 605 nm emission is due to monomeric emission while the
675 nm emission is due to aggregate emission. The similar excita-
tion spectra of 675 nm emission to that of 605 nm emission, cou-
pled with the lack of new absorption peaks in the film absorption
spectrum indicate that the 675 nm emission band may be of exci-
mer nature (dynamic aggregates), which is different from PRTS
where aggregation occurs at the ground state. It is possible that at
ground state the acetylene-linked phenyl ring is twisted in respect
Fig. 5. Normalized fluorescence emission spectra of PDI derivatives in dilute chloro-
form solutions ( ex>400 nm) except for PTG0 whose PDI emission was measured in
hexane. The inset shows the emission spectra of PTG0 and PTG1 in chloroform when
l
excited at 350 nm.
films, however, are very different. While the film of PATS shows
identical absorption wavelengths to those of its chloroform solu-
tion; the absorption spectrum of PRTS film has stark difference
from that of its chloroform solution (Fig. 6a). Most notably, a rela-
tively strong absorption tail extending beyond 700 nm is observed.
The 0e0 transition seen as the strongest absorption band in both
chloroform and hexane solutions disappeared while the 0e1
transition becomes the strongest one in the visible region. The
fluorescence emission difference between PATS and PRTS is also
obvious: PRTS emits at much longer wavelength (lem¼650 nm)
than that of PATS (lem¼600 nm). The excitation spectrum of PRTS
film is also significantly red-shifted compared to that of PATS
to the perylene ring, which prevents peryleneeperylene
pep
stacking. In the excited state, one of the phenyl rings and the per-
ylene core may be planar (due to the charge transfer nature of the

2812
M. Bagui et al. / Tetrahedron 68 (2012) 2806e2818
Fig. 6. UV/vis absorption (a) and fluorescence emission (b) spectra of PATS (top) and PRTS (bottom) in chloroform (black), hexane (blue) and as thin films (red).
exciton), which is conducive to excimer formation. Clearly, sub-
stitution at the bay area not only affects the
p-extension, but also
has a profound effect on the mode of PDI core aggregation.
Unlike PDI derivatives discussed so far (PBr, PATS, PRTS, PTB),
all of which show blue shift in absorption and fluorescence emis-
sion when the solvent is changed from chloroform to hexane, PTG0
however shows red-shift in the absorption spectrum as shown in
Fig. 8a. The absorption spectrum of PTG0 in hexane closely matches
that of PTG0 film, particularly for the absorption band edge.
Fig. 8. UV/vis absorption (a) and fluorescence emission (b) of PTG0 in chloroform
(black), hexane (blue) and, as thin films (red). The excitation wavelength is 370 nm
(except for the inset).
Fig. 7. Fluorescence emission, excitation, and absorption spectra of PTB in chloroform
(red), hexane (blue), and as thin film (red).

M. Bagui et al. / Tetrahedron 68 (2012) 2806e2818
2813
Apparently, PTG0 molecules aggregate in hexane solutions. In-
terestingly, PTG0 in hexane is far more fluorescent than in chlo-
roform. As shown in Fig. 8b, when excited at 370 nm (G0T
excitation), PTG0 in hexane not only give structured emissions in
the 360e500 nm range, but also a longer wavelength emission at
605 nm. As mentioned earlier, the short wavelength emissions are
shift. When the emission wavelength is set at 605 nm, the excitation
spectrum shows a strong excitation peak at 574 nm. It is noted that
PTG0 in hexane exhibits an absorption band at 635 nm, which is at
a longer wavelength than the observed emission band. Since the film
has a similar absorption band, one can conclude that the 635 nm
absorption is due to non-emissive aggregates. In hexane, both the
monomeric PTG0 molecules and PTG0 aggregates appear to be
present. It is reasonable to assume that the maximum absorption
wavelength of the monomeric species in hexane is around 580 nm
attributed to the extended
attached G0T while the longer wavelength emission is assigned to
the -system extending out from the PDI core. When the PDI core is
p-system originating from perylene-
p
excited, the longer wavelength emission is again observed. The
fluorescence quantum yields of PTG0 in hexane are 0.01 at 370 nm
excitation and 0.03 at 570 nm excitation. It is worth noting that
PTG0 in chloroform gives no PDI emissions, no matter what exci-
tation wavelength is used and PTG0 film is non-fluorescent.
To identity the emitting species of PTG0 in hexane, fluorescence
excitation spectra at emitting wavelengths of 440 nm and 605 nm
were recorded and compared with its absorption spectrum. As
shown in Fig. 9, the excitation spectrum with lem¼440 nm shows
well-structured peaks and a steep spectral band edge. The spectrum
resembles the one in chloroform (Fig. 4b) except for w8 nm blue
(lmax of PTB in hexane is 554 nm). Given the resolved structured
emission, the similar excitation spectra in hexane to those in chlo-
roform, the apparent matching of the excitation spectra to the ab-
sorption of the non-aggregated monomeric species, one can
conclude that the emissions observed for PTG0 in hexane are due to
non-aggregated PTG0 molecules. Some PTG0 aggregates do exist.
Such aggregates are non-emissive and are responsible for the red-
shifted absorptions. Although PTG0 exhibits low fluorescence
quantum yields in both solvents, its hexane solution is clearly more
fluorescent, presumably due to hexane’s low polarity.
PTG1 shows similar absorption behavior to PTG0: red-shifted
absorptions in hexane, which resemble those of its film (Fig. S12).
Different from PTG0 though which shows stronger emission in
hexane than in chloroform, PTG1 is non-fluorescent either in
hexane or as thin films.
2.3.4. Molecular dynamics (MD) simulations. To understand the
aggregation behaviors of these PDI derivatives, molecular dynamics
simulations were carried out using Sander in AMBER10.0 software
package39 with the AMBER 99SB force field40 and chloroform as the
solvent to obtain the energy-minimized dimer structures. The
starting structure of the PRT dimer was built and optimized with
the MMFF force field using MOE program41 based on a structurally
similar dimer obtained in the Cambridge Structural Database (CSD
entry 4312157).42 The PAT dimer and the PTB dimer were built
based on the PRT dimer and were minimized with the MMFF force
field. Each system (PRT dimer, PAT dimer, and PTB dimer) was
soaked in a rectangular box of explicit chloroform molecules ex-
ꢀ
tended 10 A away in all directions from any solute atom. Each
system was minimized for 1000 steps to reduce steric clash, fol-
lowed by 30 ps of ramping up the temperature from 10 K to 300 K,
and a 100 ps equilibration at 300 K. The equilibrated system was
continued with a production run of 2000 ps, collecting 2000
snapshots with the time step of 2 fs. The 2000th ps snapshot
structure for the PRT dimer was shown in Fig. 10A. The 2000th ps
Fig. 9. Excitation spectra of PTG0 in hexane with the emission wavelength set at
440 nm (red) or 605 nm (black), in comparison with its absorption spectrum (dotted
line).
Fig. 10. Structures of PRT dimer (A), PAT dimer (B), and PTB dimer (C) generated by molecular dynamics simulation. The PRT dimer was taken from the 2000th ps snapshot while
the PAT and PTB trajectories were taken from the 500th ps. Hydrogen atoms were not shown for clarity.

2814
M. Bagui et al. / Tetrahedron 68 (2012) 2806e2818
snapshot structures for PAT and PTB dimers however were disso-
ciated. We traced back the MD trajectories and found that the 500th
ps snapshot structures were dimers. These two structures were
minimized and reported in Fig. 10B and C for the PAT dimer and the
PTB dimer, respectively. From MD simulations, one can draw three
energetically favorable, the frontier orbitals of PTGn were evaluated
using cyclic voltammetry measurements. The cyclovoltammograms
of PTB, PTG0, G0T, and G1T scanned in chloroform solutions using
a 1 mm2 Pt disk as the working electrode, Ag/AgNO3 as the refer-
ence electrode and a Pt wire as the counter electrode are shown in
Fig. 12.
conclusions: (1) the PDI core of PAT forms a
pep stacking with
a much larger rotational offset than that of the PRT dimer. (2) The
PRT dimer structure is much more stable than the PAT dimer
structure; while the PRT dimer structure was well preserved at the
2000th ps snapshot, the PAT dimer started to dissociate from the
600th ps, and at the 800th ps the PAT dimer completely dissociated.
(3) There is no PDIePDI core stacking in the PTB dimer as the two
PDI units are nearly perpendicular to each other. These results are
consistent with their fore-discussed NMR data and optical
properties.
MD simulations were also carried out on PTG0 and PTG1. We
have previously carried out MD simulations on two ether-linked
analogs POG0 and POG1.36 The structures of PTG0 and PTG1 were
built based on those of POG0 and POG1, respectively, followed by
optimization using the MMFF force field.43 As shown in Fig.11, while
the G0 or G1 unit on one side of the PDI core adopts a conformation
almost perpendicular tothe PDI core, the triphenylene moietyon the
other side of the PDI core has a much smaller twisting angle in re-
PDI core is fairly electron deficient and thus easy to reduce and
difficult to oxidize. G0T and G1T, on the other hand, are electron
rich and easy to oxidize. During cathodic scan, PTB and PTG0 both
show two reversible reduction waves (Fig. 12), which can be at-
tributed to the reduction of the
p-acceptor system. The first re-
duction potential of PTG0 is very close to that of PTB (ꢁ1.04 V for
PTB and ꢁ1.03 V for PTG0, vs Cp2Fe). The stronger electron-do-
nating nature of triphenylene’s
p-system (than phenyl’s) would
lower PDI’s reduction potential (more negative potential) while the
more extended -conjugation in the PTG0 system would raises
p
PDI’s reduction potential (less negative). The two opposite effects
apparently cancel out. Based on the above potentials, the LUMO of
the
p-acceptor system for PTB and PTG0 can be calculated to be
ꢁ3.76 and ꢁ3.77 eV, respectively. On the anodic scan, PTG0 shows
two semi-reversible oxidation waves as well; at least the first can
be attributed to the oxidation of the
p-donor system. From the
onset potential of the first oxidation process (0.82 V vs Ag/Agþ or
0.38 V vs Cp2Fe), the HOMO of the PTG0, which is associated with
spect to the PDI core, enabling extended p-conjugation between the
PDI core and at least one side of the triphenylene dendrons, which
accounts for their red-shifted absorptions.
the
p
-donor component is estimated to be ꢁ5.18 eV. The absorption
spectra of PTG0 give bandgap energies of the two
p
-systems as 2.95
Fig. 11. Energy minimized structures of PTG0 (A) and PTG1 (B) generated from MD simulations.
2.3.5. Electrochemistry and photoinduced charge transfer. As dis-
cussed earlier, both PTG0 and PTG1 in chloroform are very weak
fluorescent compounds with quantum yields less than 0.01. Con-
sidering the fact that PTB has near unity fluorescence quantum
yield and G1T is also moderately fluorescent (0.30 quantum yield),
fluorescence quenching is clearly efficient in both PTG0 and PTG1.
and 2.02 eV, respectively. Coupled with CV data, the frontier or-
bitals for the donor (G0T-associated) and the acceptor (PDI asso-
ciated) components are obtained as shown in Fig. 13. Due to
insufficient sample, we were unable to obtain the cyclo-
voltammogram of PTG1. As a rough approximation the HOMO of
PTG1 was estimated from the oxidation potential of G1T (0.72 V vs
Ag/Agþ or 0.28 V vs Cp2Fe), which yields a HOMO level of ꢁ5.08 eV.
While there are p-conjugation extending to both the PDI core and
the attached triphenylene dendron, UV/vis absorption spectra of
Similarly, the LUMO of the p-acceptor system in PTG1 can be
PTG0 and PTG1 clearly indicate the existence of two extended
roughly estimated to be the same as that of PTG0. Combining the
electronic bandgaps (2.85 and 1.95 eV) obtained in the absorption
spectrum of PTG1, the frontier orbitals of the donor component and
the acceptor component are also estimated and are shown in
Fig. 13.
For both PTG0 and PTG1, when the donor component is excited,
one sees a facile electron transfer from excited donor to the LUMO
of the acceptor, leading to a non-emissive charge-separated state.
chromophoric
extending to part of G0 or G1 structure, which absorbs in the visible
range and can be labeled as the -acceptor system, and the other
from the G0 or G1 extending toward the PDI core, which absorbs in
the 300e450 nm range and can be called the -donor system. To
p-systems, one originating from the PDI core and
p
p
confirm whether or not a charge transfer process, following either
the donor-system excitation or the acceptor system excitation, is

M. Bagui et al. / Tetrahedron 68 (2012) 2806e2818
2815
Fig. 12. Cyclic voltammograms of PTB, PTG0 on cathodic scan (left), and PTG0, G0T, G1T on anodic scan (right). Graphs are vertically offset for clarity. Under identical conditions, the
ferrocene/ferrocenium couple has a half wave potential of 0.36 V.
Fig. 13. Frontier orbitals of the donor and the acceptor components in PTG0 and PTG1.
One competing pathway is the deactivation of the donor exciton
more polar solvent is conducive to charge separation, it is likely that
the photoinduced electron transfer of PTG0 in hexane is much less
efficient than in chloroform, which accounts for its higher PDI
fluorescence in hexane.
€
through Forster energy transfer, leading to acceptor excitation,
which can be subsequently quenched by charge transfer. When the
acceptor component is excited, there is a facile hole-transfer
pathway, which moves the hole from the HOMO of the acceptor
to the HOMO of the donor, leading to the same non-emissive
charge-separated state. Thus, whether the donor or the acceptor
is excited, photoinduced electron (or hole) transfer follows, which
quenches both the donor and the acceptor fluorescence and leads
to a non-emissive photoinduced charge-separated state. Since
photoinduced electron transfer is solvent polarity dependent and
3. Conclusions
A number of groups including trimethylsilyl, phenyl, tripheny-
lene, and triphenylene-based G1 dendron have been linked to the
bay positions of a perylene diimide (PDI) core through an ethynyl
linkage. The photophysical properties of the resulting bay-

2816
M. Bagui et al. / Tetrahedron 68 (2012) 2806e2818
substituted PDI derivatives have been carefully studied in different
solvents and as thin films. Without any capping group, the two
ethynyl bay-substituted PDI derivatives PAT and PRT both aggre-
gate strongly even in dilute solutions but in different per-
the reference energy levels of ferrocene (4.8 eV below the vacuum
level) according to EHOMO=ELUMO ¼ 4:8 þ ðE1=2 ꢁ E1FC=2Þ eV below
the vacuum level.
yleneeperylene
p
e
p
stacking motif; PRT aggregates through
4.1.1. PRTS. Regioisomeric mixture of PRBr was synthesized
according to a literature procedure.44 A mixture of 1,7- and 1,6-
PRBr (0.11 g, 0.16 mmol), triisopropylsilylacetylene (0.18 mL,
0.81 mmol), Pd(PPh3)2Cl2 (0.0068 g, 0.0096 mmol), CuI (0.0045 g,
0.0236 mmol), triethylamine (20 mL) was stirred under nitrogen at
90 ꢀC for 24 h and was then poured into dichloromethane. The
resulting mixture was passed through Celite. The filtrate was
washed with aqueous HCl (1 N) and then with water (two times).
The organic layer was separated and dried over MgSO4. The crude
product obtained by solvent evaporation was purified by chroma-
tography on silica gel using 50% toluene in hexane as the eluent to
give the product as bright red solid (0.113 g, 78%, mp 256.3 ꢀC). 1H
slipped (or longitudinal) stacking while PAT self-assembles by ro-
tational (or cross) stacking. With capping groups, the PDI core
stacking is completely blocked for PATS in both solution and solid
film. For PRTS, the slipped stacking is observed only for its film
sample, while for PTB, association only occurs after excitation
(excimer formation). When triphenylene or triphenylene-based G1
dendron is attached to the acetylene bridge, the resulting
donoreacceptor systems exhibit strong electronic coupling be-
tween the dendritic donors and the PDI acceptor, leading to sig-
nificantly red-shifted absorption bands. The conjugated linkage
also facilitates photoinduced electron transfer from the tripheny-
lene or triphenylene dendron to the PDI core, effectively quenching
both the donor and acceptor fluorescence emissions. The signifi-
cantly red-shifted absorption bands and the efficient photoinduced
electron transfer observed on PTG0 and PTG1 indicate that these
new PDI derivatives may find applications in solar cells.
NMR (400 MHz, CDCl3):
d
10.21 (d, J¼8.0 Hz, 2H), 8.71 (s, 2H), 8.51
(d, J¼8.0 Hz, 2H), 4.18 (t, J¼4.0 Hz, 4H), 1.75 (m, 4H), 1.45 (m, 12H),
1.21 (m), 1.19 (s, 36H), 0.89 (t, J¼4.0 Hz, 6H). 13C NMR (400 MHz,
CDCl3):
d 163.0, 138.4, 134.0, 130.4, 127.6, 127.3, 123.0, 121.9, 119.9,
107.4, 103.8, 40.7, 31.5, 28.0, 26.8, 22.5, 18.8, 14.1, 11.3. Anal. Calcd for
C58H74N2O4Si2 C, 75.77%, H, 8.11%, N, 3.05%. Found: C, 76.13%, H,
8.04%, N, 2.95%.
4. Experimental section
4.1. General
4.1.2. PRT. To a solution of PRTS (0.076 g, 0.083 mmol) in THF
(10 mL), a sample of TBAF (0.056 g, 0.214 mmol) was added drop-
wise. The mixture was stirred at room temperature for 30 min. The
crude product was obtained by solvent evaporation from reaction
mixture. The resulting solid was washed with methanol to give
a blackish red solid (0.023 g, 45%, no melting transition observed
before crosslinking onsets around 150 ꢀC). 1H NMR (400 MHz,
All reagents and solvents were obtained from either Aldrich or
Fisher and were used as received unless otherwise stated. Anhy-
drous THF were distilled over sodium/benzophenone. Triethyl-
amine was distilled from calcium hydride prior to use. All air- and
moisture-sensitive reactions were carried out in oven-dried glass-
ware under a nitrogen atmosphere unless stated otherwise.
CDCl3):
2H), 4.18 (br, 4H), 3.79 (s, 2H), 1.90e0.6 (br). 13C NMR (400 MHz,
CDCl3): 162.5 (br), 128.1 (br), 58.7, 41.1 (br), 31.5, 26.7, 22.5, 14.0.
d
10.00 (d, J¼8.0 Hz, 2H), 8.79 (s, 2H),
d
8.68 (d, J¼8.0 Hz,
1H and 13C NMR spectra were recorded on a Varian Unity
400 MHz NMR spectrometer. A Voyager DE Pro (Perceptive Bio-
systems/ABI) MALDI-TOF mass spectrometer was used for mass
measurement, operating in reflector mode. A mixture of silver tri-
fluoroacetate/dithranol (1,8-dihydroxyanthrone; 1:25 w/w) was
used as the matrix. UVevis absorption spectra were recorded in
1 cm path length UV-grade spectrophotometric cells using a Hew-
lettePackard 8452A diode array spectrophotometer. The fluores-
cence emission and excitation spectra were measured using
a Shimadzu RF-5301PC spectro-fluorophotometer. All samples
were deoxygenated by bubbling N2 gas through the sample im-
mediately prior to the fluorescence measurements. Fluorescence
quantum yields were determined by comparing the integrated
fluorescence spectra of the compounds with the integrated fluo-
rescence spectrum of the selected standard and correcting for the
different refractive indices of the sample and the standard. For G0
and G1 emissions, quinine sulfate in 1 N H2SO4 (Ffl¼0.55) was used
as the standard, while cresyl violet perchlorate in methanol
d
Anal. Calcd for C40H34N2O4 C, 79.19%, H, 5.65%, N, 4.62%. Found: C,
77.95%, H, 6.02%, N, 4.43%.
4.1.3. PATS. PBr was synthesized according to published literature
procedures.45 A mixture of PBr (0.20 g, 0.23 mmol), trimethylsily-
lacetylene (0.17 mL, 1.17 mmol), Pd(PPh3)2Cl2 (0.0117 g,
0.0167 mmol), CuI (0.004 g, 0.021 mmol), triethylamine (30 mL)
was stirred under nitrogen at 60 ꢀC for 19 h and was then poured
into dichloromethane. The resulting mixture was passed through
Celite. The filtrate was washed with aqueous HCl (1 N) and then
water (two times). The organic layer was separated and dried over
MgSO4. The crude product obtained by solvent evaporation was
purified by chromatography on silica gel using 50% CH2Cl2 in hex-
ane as the eluent to give the product as bright red solid (0.13 g, 62%,
no melting transition observed before decomposition starts around
300 ꢀC). 1H NMR (400 MHz, CDCl3):
d
10.28 (dd, J¼8.0 Hz, 2H), 8.92
(Ffl¼0.54) was used as the standard for determining the fluores-
(d, J¼8.0 Hz, 2H),
d
8.76 (dd, J¼8.0 Hz, 2H), 7.50 (t, J¼4.0 Hz, 2H),
cence quantum yield of the PDI core emissions. The refractive index
of the solvent was used to approximate the sample refractive index,
and the refractive index of water was used for the standard (both at
25 ꢀC). Cyclic voltammetry (CV) studies were carried out in chlo-
roform at room temperature under argon protection using a BAS
Epsilon EC electrochemical station employing a 1 mm2 Pt disk as
the working electrode, Ag/AgNO3 as the reference electrode and
a Pt wire as the counter electrode. 0.1 M tetrabutylammonium
hexafluorophosphate was the supporting electrolyte, and the scan
rate was 50 mV/s. Each measurement was calibrated with ferrocene
7.36 (d, J¼8.0 Hz, 4H), 2.73 (m, 4H), 1.18 (m, 24H), 0.37 (s, 18H). 13C
NMR (400 MHz, CDCl3):
d 163.5, 163.2, 163.0, 162.8, 145.7, 145.6,
138.9,138.6,135.4,134.8,134.5,133.9,131.1,130.7,130.5,130.3,130.1,
129.9,129.8,129.7,128.4,127.9,127.6,124.1,123.6,123.3,122.1,121.8,
120.7, 120.2, 106.5, 106.3, 105.5, 105.4, 29.2, 24.0, 0.4. Anal. Calcd for
C58H58N2O4Si2 C, 77.12%, H, 6.47%, N, 3.10%. Found: C, 77.83%, H,
6.64%, N, 2.87%.
4.1.4. PAT. To a solution of PATS (0.029 g, 0.032 mmol) in THF
(5 mL), a sample of TBAF in 1 M THF (0.1 mL, 0.1 mmol) was added
dropwise. The mixture was stirred at room temperature for 15 min.
The crude product was obtained by solvent evaporation from re-
action mixture. The resulting solid was washed with methanol to
give a blackish red solid (0.0125 g, 52%, no melting transition ob-
served before crosslinking onsets around 150 ꢀC). 1H NMR
(Fc) as the internal standard (with the measured
FC
E
¼ 0:355V vs Ag=AgNO3). In the case of reversible curves, each
1=2
anodic and corresponding cathodic potential was averaged as
red=ox
1=2
E
¼ 1=2ðEpc þ EpaÞ to obtain oxidation and reduction poten-
tials. HOMO and LUMO energy levels were estimated on the basis of

M. Bagui et al. / Tetrahedron 68 (2012) 2806e2818
2817
(400 MHz, CDCl3):
d
10.14e9.92 (br, 2H), 9.05 (t, J¼2.0 Hz, 2H),
[M]þ 1672.96, [MþNa]þ 1694.94. Anal. Calcd for C114H114N2O10 C,
d
8.97e8.80 (br, 2H), 7.50 (m, 2H), 7.34 (m, 4H), 3.48 (br), 2.71 (br),
81.88%, H, 6.87%, N, 1.68%. Found: C, 81.37%, H, 7.02%, N, 1.76%.
1.17 (m). HRMS-EI, calculated for C52H42N2O4 758.31, found
758.3145.
4.1.8. PTG1. A sample of triethylamine (1 mL) was added to the
mixture of G1T (0.035 g, 0.0242 mmol), PBr (0.010 g, 0.0115 mmol),
Pd(PPh3)2Cl2 (0.001 g, 0.0014 mmol), CuI (0.0007 g, 0.004 mmol),
and DMF (3 mL) at room temperature. The resulting mixture was
stirred under nitrogen at 90 ꢀC for 24 h. The product was pre-
cipitated out from the reaction mixture and collected by filtration.
The resulting solid was purified by chromatography eluting with
THF to give the pure product as dark green solid (0.0086 g, 21%, no
melting transition observed before decomposition). The pure
product had limited solubility in common organic solvents. 1H NMR
4.1.5. G1T. A mixture of G1OTf (0.169 g, 0.1077 mmol), trime-
thylsilylacetylene (0.05 mL, 0.345 mmol), Pd(PPh3)2Cl2 (0.0035 g,
0.0049 mmol), CuI (0.0035 g, 0.0184 mmol), triethylamine
(0.07 mL), and anhydrous DMF (10 mL) was stirred under nitrogen
at 60 ꢀC for 21 h. The product was precipitated out from the re-
action mixture and collected by filtration. The resulting solid was
washed with methanol to give G1-TMS as a greenish yellow solid
(0.113 g, 69%). The desilylation was carried out by adding a sample
of TBAF in 1 M THF (0.08 mL, 0.08 mmol) to a THF (10 mL) solution
of the above greenish yellow solid (0.093 g, 0.0613 mmol). The
mixture was stirred at room temperature for 45 min and was then
poured into water. The organic phase was extracted with methy-
lene chloride. The organic extracts were washed with water, dried
over anhydrous MgSO4, and evaporated in vacuo to give a crude
product, which was purified by recrystallization, using CH2Cl2 and
methanol solvent mixture to afford G1T as a greenish yellow solid
(0.0832 g, 94%. No melting transition observed before crosslinking
(400 MHz, CDCl3):
d 10.54 (br), 8.67 (br), 8.33 (br), 7.82 (br), 7.51
(br), 7.32 (br), 7.1 (br), 4.29 (br), 4.09 (br), 2.81 (br), 1.81 (br), 1.63
(br), 1.49 (br), 1.31 (br), 0.87 (br). MALDI-TOF mass analysis: cal-
culated for C246H258N2O22 3594.68, found [MþAg]þ 3702.70. Anal.
Calcd for C246H258N2O22 C, 82.19%, H, 7.23%, N, 0.78%. Found: C,
81.46%, H, 7.49%, N, 0.83%.
Acknowledgements
onsets around 165 ꢀC). 1H NMR (400 MHz, CDCl3):
d 8.64 (s, 1H),
We thank the National Science Foundation (DMR 0804158) and
the Army Research Office (W911NF-10-1-0476) for the support of
this work.
8.57 (d, J¼4 Hz, 2H), 8.53 (s, 1H), 8.25 (m, 4H), 8.18 (s, 1H), 7.77 (d,
J¼8 Hz, 2H), 7.66 (s,1H), 7.63 (s,1H), 7.59 (s,1H), 7.54 (s,1H), 7.48 (d,
J¼8 Hz, 2H), 7.41 (s, 1H), 7.12 (m, 4H), 4.24 (m, 21H), 3.52 (s, 1H),
2.04e0.71 (m, 66H). 13C NMR (400 MHz, CDCl3):
d 159.0, 158.6,
Supplementary data
158.1, 157.9, 157.8, 157.5, 131.6, 131.3, 131.2, 130.0, 129.6, 129.4, 129.3,
129.2, 129.1, 128.7, 128.5, 124.6, 124.3, 124.1, 124.0, 123.7, 123.5,
122.8, 122.1, 115.8, 115.6, 115.5, 115, 113.5, 113.4, 111.9, 107.4, 106.1,
106.0, 105.1, 104.0, 103.5, 91.5, 90.9, 90.8, 82.0, 80.4, 69.2, 69.0, 68.3,
68.2, 56.1, 31.7, 31.6, 29.5, 29.4, 25.9, 25.8, 22.7, 22.5, 14.1, 14.0. Anal.
Calcd for C99H110O9 C, 82.35%, H, 7.68%. Found: C, 82.07%, H, 7.72%.
1H and 13C NMR spectra of the compounds, MALDI-TOF mass
spectra, molecular dynamics (MD) simulations, additional optical
spectra (absorption, fluorescence emission, excitation). Supple-
mentary data associated with this article can be found, in the online
version, at doi:10.1016/j.tet.2012.02.008.
4.1.6. PTB. A sample of diisopropyl amine (4 mL) was added to the
mixture of PBr (0.02 g, 0.023 mmol), ethynylbenzene (0.007 g,
0.0685), Pd(PPh3)2Cl2 (0.001 g, 0.0014 mmol), CuI (0.0005 g,
0.0026 mmol) at room temperature. The resulting mixture was
stirred under nitrogen at 75 ꢀC for overnight. The product was
precipitated out from the reaction mixture and collected by filtra-
tion. The resulting solid was washed with methanol to give PTB as
References and notes
1. Gawrys, P.; Boudinet, D.; Zagorska, M.; Djurado, D.; Verilhac, J.; Horowitz, G.;
Pecaud, J.; Pouget, S.; Pron, A. Synth. Met. 2009, 159, 1478.
2. (a) Weil, T.; Reuther, E.; Mullen, K. Angew. Chem., Int. Ed. 2002, 41, 1900; (b)
Serin, J. M.; Brousmiche, D. W.; Frechet, J. M. J. Chem. Commun. 2002, 2605; (c)
Sautter, A.; Kaletas, B. K.; Schmid, D. G.; Dobrawa, R.; Zimine, M.; Jung, G.; van
Stokkum, I. H. M.; Cola, L. D.; Williams, R. M.; Wurther, F. J. Am. Chem. Soc. 2005,
127, 6719.
3. Lee, S. K.; Zu, Y.; Herrmann, A.; Geerts, Y.; Mullen, K.; Bard, A. J. J. Am. Chem. Soc.
1999, 121, 3513.
4. Yan, H.; Chen, Z.; Zheng, Y.; Newman, C.; Quinn, J. R.; Dotz, F.; Kastler, M.;
Facchetti, A. Nature 2009, 457, 679.
5. (a) Jones, B. A.; Ahrens, M. J.; Yoon, M. H.; Facchetti, A.; Marks, T. J.; Wasie-
lewski, M. R. Angew. Chem., Int. Ed. 2004, 43, 6363; (b) An, Z.; Yu, J.; Jones, S. C.;
Barlow, S.; Yoo, S.; Domercq, B.; Prins, P.; Siebbeles, L. D. A.; Kippelen, B.;
Marder, S. R. Adv. Mater. 2005, 17, 2580.
6. Law, K. Y. Chem. Rev. 1993, 93, 449.
7. O’Neil, M. P.; Niemczyk, M. P.; Svec, W. A.; Gosztola, D.; Gaines, G. L., III; Wa-
€
ꢁ
€
a dark red solid (0.028 g, 74%). 1H NMR (400 MHz, CDCl3):
(dd, J¼8.0 Hz, 2H), 9.02 (d, J¼8.0 Hz, 2H), 8.87 (dd, J¼8.0 Hz, 2H),
7.67 (m, 4H), 7.47 (m, 8H), 7.37(d, J¼8.0 Hz, 4H), 2.77 (m, 4H), 1.20
(m, 24H). 13C NMR (400 MHz, CDCl3):
177.8, 163.6, 145.6, 138.2,
d 10.22
€
d
€
d
134.8, 134.4, 131.9,131.2, 131.0, 130.5, 129.8, 128.9, 128.4, 128.1,127.7,
127.3, 124.2, 123.1, 122.2, 122.0, 121.2, 120.5, 98.8, 90.6, 68.0, 47.9,
29.3, 25.6, 24.0, 19.3. Anal. Calcd for C64H50N2O4 C, 84.37%, H, 5.53%,
N, 3.07%. Found: C, 84.06%, H, 5.62%, N, 3.13%.
sielewski, M. R. Science 1992, 257, 63.
8. Prathapan, S.; Yang, S. I.; Seth, J.; Miller, M. A.; Bocian, D. F.; Holten, D.; Lindsey,
J. S. J. Phys. Chem. B 2001, 105, 8237.
4.1.7. PTG0. A sample of triethylamine (3 mL) was added to the
mixture of PBr (0.044 g, 0.0507 mmol), G0T (0.0521 g,
0.1036 mmol), Pd(PPh3)2Cl2 (0.002 g, 0.0028 mmol), CuI (0.001 g,
0.0053 mmol), and DMF (3 mL) at room temperature. The resulting
mixture was stirred under nitrogen at 90 ꢀC for 24 h. The crude
product obtained by solvent evaporation from reaction mixture,
was dissolved in hot chloroform and was absorbed on silica gel and
loaded on a column. The column was eluted with CH2Cl2 followed
by ethyl acetate to remove impurities. The crude product absorbed
on silica gel was then subjected to Soxhlet extraction for 3 days
using CH2Cl2 to yield PTG0 as a black solid (0.029 g, 35%, no melting
transition observed before decomposition). The pure product had
limited solubility in common organic solvents. 1H NMR (400 MHz,
9. Huang, C.; Barlow, S.; Marder, S. R. J. Org. Chem. 2011, 76, 2386.
10. (a) He, X.; Liu, H.; Li, Y.; Wang, S.; Wang, N.; Xiao, J.; Xu, X.; Zhu, D. Adv. Mater.
2005, 17, 2811; (b) Wang, B.; Yu, C. Angew. Chem., Int. Ed. 2010, 49, 1485.
11. Che, Y.; Yang, X.; Liu, G.; Yu, C.; Ji, H.; Zuo, J.; Zhao, J.; Zang, L. J. Am. Chem. Soc.
2010, 132, 5743.
12. Che, Y.; Datar, A.; Yang, X.; Naddo, T.; Zhao, J.; Zang, L. J. Am. Chem. Soc. 2007,
129, 6354.
13. Huang, Y.; Fu, L.; Zou, W.; Zhang, F.; Wei, Z. J. Phys. Chem. C 2011, 115, 10399.
14. (a) Dittmer, J. J.; Marseglia, E. A.; Friend, R. H. Adv. Mater 2000, 12, l270; (b)
Roncali, J.; Frre, P.; Blanchard, P.; Bettignies, R.; Turbiez de, M.; Roquet, S.;
Leriche, P.; Nicolas, Y. Thin Solid Films 2006, 5l1/512, 567; (c) Li, J.; Dierschke, F.;
Wu, J.; Grimsdale, A. C.; Mllen, K. J. Mater. Chem. 2006, 16, 96; (d) Jonkheijm, P.;
Stutzmann, N.; Chen, Z.; de Leeuw, D. M.; Meijer, E. W.; Schenning, A. P. H. J.;
Wrthner, F. J. Am. Chem. Soc. 2006, 128, 9535; (e) Neuteboom, E. E.; Meskers, S.
C. J.; van Hal, P. A.; van Duren, J. K. J.; Meijer, E. W.; Janssen, R. A. J.; Dupin, H.;
ꢁ
Pourtois, G.; Cornil, J.; Lazzaroni, R.; Bredas, J.-L.; Beljonne, D. J. Am. Chem. Soc.
CDCl3): d 11.27 (br), 9.30 (br), 9.26 (br), 9.06 (br), 7.22 (br), 7.16 (br),
6.75 (br), 6.65 (br), 4.27 (br), 1.22 (br), 1.13 (br), 0.83 (br). MALDI-
TOF mass analysis: calculated for C114H114N2O10 1672.13, found
2003, 125, 8625.
15. Li, C.; Liu, M.; Pschirer, N. G.; Baumgarten, M.; Mullen, K. Chem. Rev. 2010, 110,
6817.

2818
M. Bagui et al. / Tetrahedron 68 (2012) 2806e2818
€
€
€
16. Schmidt-Mende, L.; Fechtenkotter, A.; Mullen, K.; Moons, E.; Friend, R. H.;
MacKenzie, J. D. Science 2001, 293, 1119.
30. (a) Fabian, N.; Wojciech, P.; Sibylle, M.; Christopher, K.; Mullen, K. Chem. Mater.
2006, 18, 3715; (b) Yan, Q.; Zhao, D. Org. Lett. 2009, 11, 3426.
17. Zhan, X.; Tan, Z. A.; Domercq, B.; An, Z.; Zhang, X.; Barlow, S.; Li, Y.; Zhu, D.;
31. (a) Odom, S. A.; Richard, F. K.; Ohira, S.; Trenton, R.; Ensley, T. R.; Huang, C.;
Lazaro, A.; Padilha, L. A.; Webster, S.; Coropceanu, V.; Barlow, S.; Hagan, D. J.;
Stryland, E. W. V.; Brdas, J.-L.; Anderson, H. A.; Wasielewski, M. R.; Marder,
Kippelen, B.; Marder, S. R. J. Am. Chem. Soc. 2007, 129, 7246.
€
18. Wurthner, F. Chem. Commun. 2004, 1564.
19. Zang, L.; Che, Y.; Moore, J. S. Acc. Chem. Res. 2008, 41, 1596.
20. Wasielewski, M. R. J. Org. Chem. 2006, 71, 5051.
S. R. J. Phys. Chem. A 2009, 113, 10826; (b) Kelley, R. F.; Won, S. S.;
Rybtchinski, B.; Sinks, E. L.; Wasielewski, M. R. J. Am. Chem. Soc. 2007, 129,
21. Li, Y.; Tan, L.; Wang, Z.; Qian, H.; Shi, Y.; Hu, W. Org. Lett. 2008, 10, 529.
22. Zhan, C. L.; Li, A. D. Q. Curr. Org. Chem. 2011, 15, 1314.
3173.
ꢁ
32. Jimenez, A. J.; Fabian, S.; Rodríguez-Morgade, M. S.; Ohkubo, K.; Fukuzumi, S.;
23. Ke, D.; Zhan, C.; Xu, S.; Ding, X.; Peng, A.; Sun, J.; He, S.; Li, A. D. Q.; Yao, J. J. Am.
Chem. Soc. 2011, ASAP.
Guldi, D. M.; Torres, T. Org. Lett. 2007, 9, 2481.
33. An, Z.; Susan, A. O.; Kelley, R. F.; Huang, C.; Zhang, X.; Barlow, S.; Padilha, L. A.;
Fu, J.; Webster, S.; Hagan, D. J.; Stryland, E. W. V. S.; Wasielewski, M. R.; Marder,
S. R. J. Phys. Chem. A 2009, 113, 5585.
34. Clark, A. E.; Qin, C.; Li, A. D. Q. J. Am. Chem. Soc. 2007, 129, 7586.
35. Wuerthner, F.; Stepanenko, V.; Chen, Z.; Saha-Moeller, C. R.; Kocher, N.; Stalke,
D. J. Org. Chem. 2004, 69, 7933.
24. (a) Chen, Z.; Debije, M. G.; Debaerdemaeker, T.; Osswald, P.; Wrthner, F.
ChemPhysChem 2004, 5, 137; (b) Wang, W.; Shaller, A. D.; Li, A. D. Q. J. Am. Chem.
Soc. 2008, 130, 8271; (c) Schmidt, R.; Oh, J. H.; Sun, Y.-S.; Deppisch, M.; Krause,
A.-M.; Radacki, K.; Braunschweig, H.; Knemann, M.; Erk, P.; Bao, Z.; Wrthner, F.
J. Am. Chem. Soc. 2009, 131, 6215; (d) Tang, M. L.; Oh, J. H.; Reichardt, A. D.; Bao,
Z. J. Am. Chem. Soc. 2009, 131, 3733; (e) Schmidt, R.; Ling, M. M.; Oh, J. H.;
Winkler, M.; Konemann, M.; Bao, Z.; Wurthner, F. Adv. Mater. 2007, 19, 3692; (f)
Delgado, M. C. R.; Kim, E.-G.; da Silva Filho, D. A.; Bredas, J.-L. J. Am. Chem. Soc.
2010, 132, 3375; (g) Weitz, R. T.; Amsharov, K.; Zschieschang, U.; Villas, E. B.;
Goswami, D. K.; Burghard, M.; Dosch, H.; Jansen, M.; Kern, K.; Klauk, H. J. Am.
Chem. Soc. 2008, 130, 4637; (h) Molinari, A. S.; Alves, H.; Chen, Z.; Facchetti, A.;
Morpurgo, A. F. J. Am. Chem. Soc. 2009, 131, 2462; (i) Yoon, M. H.; Facchetti, A.;
Stern, C. E.; Marks, T. J. J. Am. Chem. Soc. 2006, 128, 5792.
25. Dubey, R. K.; Efimov, A.; Lemmetyinen, H. Chem. Mater. 2010, 23, 778.
26. (a) Chao, C.-C.; Leung, M.-k.; Su, Y. O.; Chiu, K.-Y.; Lin, T.-H.; Shieh, S.-J.; Lin, S.-C.
J. Org. Chem. 2005, 70, 4323; (b) Chen, S.; Liu, Y.; Qiu, W.; Sun, X.; Ma, Y.; Zhu, D.
Chem. Mater. 2005, 17, 2208; (c) Huang, J.; Fu, H.; Wu, Y.; Chen, S.; Shen, F.;
Zhao, X.; Liu, Y.; Yao, J. J. Phys. Chem. C 2008, 112, 2689; (d) Huang, J.; Wu, Y.; Fu,
H.; Zhan, X.; Yao, J.; Barlow, S.; Marder, S. R. J. Phys. Chem. A 2009, 113, 5039; (e)
Ahrens, M. J.; Sinks, L. E.; Rybtchinski, B.; Liu, W.; Jones, B. A.; Giaimo, J. M.;
Gusev, A. V.; Goshe, A. J.; Tiede, D. M.; Wasielewski, M. R. J. Am. Chem. Soc. 2004,
126, 8284.
36. Bagui, M.; Dutta, T.; Chakraborty, S.; Melinger, J. S.; Zhong, H.; Keightley, A.;
Peng, Z. J. Phys. Chem. A 2011, 115, 1579.
37. Bagui, M.; Melinger, J.; Chakraborty, S.; Keightley, A. J.; Peng, Z. Tetrahedron
2009, 65, 1247.
ꢁ
38. Gomez, R.; Veldman, D.; Blanco, R.; Seoane, C.; Segura, J. L.; Janssen, R. A.
Macromolecules 2007, 40, 2760.
39. Case, D. A.; Darden, D. A.; Cheatham, T. E., III; Simmerling, C. L.; Wang, J.; Duke,
R. E.; Luo, R.; Crowley, M.; Walker, R. C.; Zhang, W.; Merz, K. M.; Wang, B.;
Hayik, S.; Roitberg, A.; Seabra, G.; Kolossv_ary, I.; Wong, K. F.; Paesani, F.; Va-
nicek, J.; Wu, X.; Brozell, S. R.; Steinbrecher, T.; Gohlke, H.; Yang, L.; Mongan, J.;
Hornak, V.; Kollman, P. A. AMBER10; University of California: San Francisco,
2008.
40. Hornak, V.; Abel, R.; Okur, A.; Strockbine, B.; Roitberg, A.; Simmerling, C. L.
Proteins 2006, 65, 712.
41. The Molecular Operating Environment (MOE); Chemical Computing Group:
Montreal, Quebeck, Canada, 2010.
42. The Cambridge Structural Database. http://www.ccdc.cam.ac.uk.
43. Zhong, H.; Carlson, H. A. Proteins 2005, 58, 222.
27. Qian, H.; Negri, F.; Wang, C.; Wang, Z. J. Am. Chem. Soc. 2008, 130, 17970.
28. Popere, B. C.; Della Pelle, A. M.; Thayumanavan, S. Macromolecules 2011, 44,
4767.
ꢁ
44. Serin, J. M.; Brousmiche, D. W.; Frechet, J. M. J. J. Am. Chem. Soc. 2002, 124,
11848.
29. Balaji, G.; Kale, T. S.; Keerthi, A.; Della Pelle, A. M.; Thayumanavan, S.; Valiya-
veettil, S. Org. Lett. 2010, 13, 18.
45. Liu, Y.; Li, Y.; Jiang, L.; Gan, H.; Liu, H.; Li, Y.; Zhuang, J.; Lu, F.; Zhu, D. J. Org.
Chem. 2004, 69, 9049.