Synthesis and self-assembly of triphenylene-containing conjugated macrocycles

Article abstract of DOI:10.1039/c3ra23421e

Two imine-based shape-persistent macrocycles containing triphenylene building blocks, along with their noncyclic model compounds, have been synthesized. Their stable conformations are predicted by quantum mechanical calculations and confirmed by 2D NOESY NMR measurements. Compared to the noncyclic model compounds, the macrocycles (3 and 6) show a higher propensity for solution self-assembly due to their stronger π-π stacking interactions, and form micro-objects which exhibit an enhanced conductivity after doping. While both macrocycles have identical molecular geometries and very similar structures, their solution self-assembly leads to micro-objects with very different shapes. The two macrocycles also show different pH-dependent optical properties.

Full text of DOI:10.1039/c3ra23421e

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Synthesis and self-assembly of triphenylene-containing
conjugated macrocycles3
Cite this: RSC Advances, 2013, 3,
6008
Tanmoy Dutta,a Yanke Che,b Haizhen Zhong,c John H. Laity,d Vladimir Dusevich,e
James B. Murowchick,f Ling Zangb and Zhonghua Peng*a
Two imine-based shape-persistent macrocycles containing triphenylene building blocks, along with their
noncyclic model compounds, have been synthesized. Their stable conformations are predicted by quantum
mechanical calculations and confirmed by 2D NOESY NMR measurements. Compared to the noncyclic
model compounds, the macrocycles (3 and 6) show a higher propensity for solution self-assembly due to
their stronger p–p stacking interactions, and form micro-objects which exhibit an enhanced conductivity
after doping. While both macrocycles have identical molecular geometries and very similar structures, their
solution self-assembly leads to micro-objects with very different shapes. The two macrocycles also show
different pH-dependent optical properties.
Received 20th December 2012,
Accepted 11th February 2013
DOI: 10.1039/c3ra23421e
www.rsc.org/advances
triphenylenes etc., have been incorporated into SPMs,6 some of
which have been found to self-assemble into nanofibers with
Introduction
sizable conductivity along their long axis.7 Although these
studies are exemplary, SPMs containing PCAs are still rare,
presumably due to the synthetic challenges associated with the
preparation of size-selective macrocycles. We have been
interested in preparing conjugated systems containing PCAs,
including conjugated macrocycles,8 dendrimers9 and poly-
mers.10 In this communication, we report the synthesis and
self-assembly of two triphenylene-containing conjugated
macrocycles. The two macrocycles, differing only in the
positions of the imine N atoms, show dramatically different
optical properties and self-assemble into different microstruc-
tures.
Shape-persistent macrocycles (SPMs) based on p-conjugated
systems appeal to scientists from both fundamental research
aspects and applied technology fronts.1 Driven mostly by p–p
stacking interactions, these electrically active macrocycles can
self-assemble to form molecular crystals with wide channels.2
Among the reported SPMs, those containing polycyclic
aromatic systems (PCAs) are particularly interesting due to
their high propensity for strong p–p stacking aggregation and
the associated appealing electronic and optical properties of
these stacked PCAs.3 For example, p-stacked columns of some
PCAs exhibit strong electron delocalization and a high charge-
mobility.4 PCA-containing SPMs can self-assemble to form
multiple PCA-stacked columns which can facilitate charge
transport, a property crucial to some molecular electronic
devices such as light-emitting diodes, field-effect transistors,
solar cells and sensors.4,5 Indeed, PCAs such as carbazoles,
Results and discussion
Synthesis
aDepartment of Chemistry, University of Missouri-Kansas City, Kansas City, MO
64110, United States. E-mail: pengz@umkc.edu; Fax: +1-816-235-2290;
Tel: +1-816-235-2288
Scheme 1 shows the structures and syntheses of the two
triphenylene-containing macrocyles 3 and 6. Except for the
location difference of the four imine nitrogen atoms, the two
macrocycles have an identical structure. The triphenylene
derivative 1 was synthesized in good yield via a previously
published procedure.11 The two bromo groups in 1 were
bDepartment of Materials Science & Engineering, University of Utah, Salt Lake City,
UT 84108, United States
cDepartment of Chemistry, University of Nebraska-Omaha, Omaha, NE, United States
dDepartment of Cell Biology and Biophysics, University of Missouri-Kansas City,
Kansas City, MO 64110, United States
eDepartment of Oral Biology, University of Missouri-Kansas City, Kansas City, MO
converted to aldehydes via
a lithium–halogen exchange
followed by quenching with excess DMF. The synthesis of
the model compound 3m was afforded via a condensation
reaction of 2 with freshly distilled aniline in the presence of
molecular sieves and a Dean–Stark trap. The same reaction
conditions were utilized to synthesize macrocycle 3 via the
reaction of 2 with m-diaminobenzene.
64110, United States
fDepartment of Geosciences, University of Missouri-Kansas City, Kansas City, MO
64110, United States
Electronic supplementary information (ESI) available: supplementary data
3
associated with this article can be found in the online version. The data includes
1H and 13C NMR, MALDI-TOF spectra, 2D NOESY spectra, X-ray diffraction
patterns of the self-assembled structures and QM calculations. See DOI: 10.1039/
c3ra23421e
6008 | RSC Adv., 2013, 3, 6008–6015
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
Scheme 1 The synthesis of macrocycles 3 and 6 and their model compounds 3m and 6m.
To synthesize macrocycle 6, compound 1 was first converted
to compound 4 through a Pd-catalyzed coupling reaction.12
The acid-catalyzed hydrolysis of 4 yielded a triphenylene
derivative 5, with two amino functional groups in the salt
form. The condensation reaction of 5 with benzaldehyde or
isophthalaldehyde afforded the model compound 6m and the
macrocycle 6, respectively. Both macrocyles 3 and 6 are soluble
in common organic solvents and their structures are con-
firmed by 1H NMR, MALDI-TOF and elemental analysis (see
contact with H7 (7.33 ppm) and H9 (4.21 ppm) but a negligible
NOE contact with H5, indicating that 3 adopts mostly the 3o
conformation.13 On the other hand, the 2D 1H–1H NOESY
spectrum of 6 (Fig. 2b) shows that the imine hydrogen H2 has
NOE contacts with H5 and H1, but not H7, indicating that,
contrary to 3, the imine H’s in 6 are located inside the ring
cavity. In other words, 6 adopts a 6i conformation. It is noted
that the 3o and 6i conformations have an identical molecular
shape, although their imine H atoms point in opposite
directions.
experimental section and ESI for detailed characterization
3
data).
Quantum mechanics calculations
Conformation analysis
To provide a reasonable explanation of why one isomer is
preferred over another, we carried out quantum mechanics
(QM) calculations for compounds 3 and 6. To simplify the
calculations, we truncated –OC12H25 to –OC2H5. This simpli-
fication is justifiable because (a) the superimposition of 3 and
6 with –OC12H25 and their counterparts with –OC2H5 showed
that the core macrocycle structures were almost identical (Fig.
An imine bond may exist as a cis or trans isomer. However, due
to geometrical constraints, it is likely that the imine bonds in 3
and 6 adopt only the trans conformation. Depending on the
orientation of the imine-H’s, the macrocycle 3 (or 6) may adopt
3i (or 6i, all imine H’s pointing towards the ring cavity), 3o (or
6o, all imine H’s pointing away from the ring cavity), or 3io (or
6io two inside and two outside) structures, as shown in Fig. 1.
These configurations can interconvert through rotations along
the two parallel single bonds that are indicted by arrows in
Fig. 1. Whether an imine H is located inside or outside the ring
cavity can be determined by its physical proximity to H7 or H5.
To determine which isomer compounds 3 and 6 might
adopt, we carried out 2D 1H–1H NOESY experiments. As shown
in Fig. 2a, the imine hydrogen H2 (9.34 ppm) has a strong NOE
S20 ), and (b) the change was far from the imine hydrogen and
3
therefore should have little effect on the core geometry.
DFT calculations with the basis set B3LYP/6-31G**//B3LYP/
STO-3G show that 3o is indeed the most stable conformer
(Table 1). The other two isomers of 3 (3io and 3i) are at least 11
kcal mol21 higher in energy than 3o, suggesting that at room
temperature, 3o would be the dominant conformer. 6i is also
calculated to be the most stable conformer. However, the
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 6008–6015 | 6009

View Article Online
Paper
RSC Advances
Fig. 1 Various conformers of macrocycles 3 and 6 and their interconversion.
energetic difference between conformers of 6 is much smaller
compared to those of 3. The experimental observations of 6i
indicate that the inter-conversion between the conformers is
kinetically hindered. The high rotational barrier is likely due
to 1) the rigidity of the macrocyclic skeleton, 2) the require-
ment of simultaneous rotation along two bonds, and 3) the
partial double bond character of the rotating C–C and C–N
bonds (indicated by arrows in Fig. 1) due to extended
p-conjugation. Therefore, 3o and 6i are the preferred products
during the imine/macrocycle formation process. The selectiv-
ity may be explained as shown in Fig. 3. The protonated
hemiaminal intermediates leading to 3o and 6i are likely to be
thermodynamically more stable than other intermediates due
to the existence of hydrogen bonding.
Optical properties
Both 3 and 6 are soluble in THF under ambient conditions,
but are insoluble in methanol and hexane. In THF, macrocycle
3 shows three absorption bands at 298, 360 and 414 nm, and a
shoulder band at around 390 nm (Fig. 4). Compared to the
model compound 3m, which also shows these bands, and a 5–
10 nm bathochromic shift is observed for all bands.
Macrocycle 6 and its model compound 6m, on the other
hand, show only two broad absorption bands at around 280
nm and 365 nm for 6 and 275 nm and 355 nm for 6m. While
the absorption tails for both 3 and 6 extend to around 440 nm,
macrocycle 3 has a much sharper band edge and better
resolved vibronic features.
Macrocycle 3 and 3m are colourless in THF, but give an
orange–red solution when dissolved in CDCl3. It is suspected
that the residual acid in CDCl3 is responsible for this colour
change. To confirm the acid effect, to the THF solution of 3 (or
3m) is added TsOH. Indeed a broad absorption extending up
to 600 nm is observed for both 3 and 3m (Fig. 5). After
neutralization with Et3N, the red-shifted band disappears and
Table 1 Ground state energies of various conformers of 3 and 6
Conformer
Energy (hartrees)
DEa (kcal mol21
)
3i
23293.264344
23293.283739
23293.301365
23293.28764
23293.286299
23293.283896
23.23
11.06
0.00
0.00
0.84
2.35
3io
3o
6i
6io
6o
Fig. 2 Selected regions from a 2D 600 MHz 1H NMR watergate NOESY spectra
of 3 (a) and 6 (b), showing strong NOE contacts, recorded in THF-d8 at 298 K
with a 400 ms NOE mixing time.
a
Energy difference relative to the most stable conformer.
6010 | RSC Adv., 2013, 3, 6008–6015
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
Fig. 3 Protonated hemiaminal intermediates leading to different macrocyclic
conformers.
the original spectral feature in the 300–420 nm range is
recovered. It is noted that, while only a slight decrease in
absorbance in the 300–420 nm range occurs for 3 after the
acidification/neutralization cycle, the solution of 3m shows a
significant loss in absorption in the 300–420 nm range, which
is due to imine hydrolysis under acidic conditions. When the
same experiment is carried out with macrocycle 6 or 6m, a red-
shifted band is not observed when acid is added. Instead, the
longest absorption band (around 360 nm) completely dis-
appeared after the addition of acid and is not recovered after
subsequent neutralization, indicating that the imine bonds in
6m and 6 have completely hydrolysed. This shows that the
Fig. 5 The UV-Vis absorption spectra of 6, 6m, 3 and 3m in original THF solution
(black), after addition of TsOH (red), and after subsequent neutralization with
Et3N (blue).
imine bonds in macrocycle 3 have a higher hydrolytic stability
over those in the model compound 3m and even more so than
those in 6 and 6m. The drastically different optical properties
of the two macrocycles under acidic conditions reflect their
structural differences. Under acidic conditions, the imine
nitrogen is protonated, which leaves the imine carbon with a
partial positively charged. The adjacent electron-rich triphe-
nylene p-system in
3 (and 3m) helps to stabilize the
carbocation through a charge transfer resonance effect (Fig.
S13 ). Such a charge transfer transition may account for the
3
significantly red-shifted absorption of 3 and 3m under acidic
conditions and may also explain the better hydrolytic stability
of the imine bonds in 3 and 3m. For 6 and 6m, imine
protonation leads to a benzyl-type cation which does not have
red-shifted charge transfer transitions and thus no red-shifted
absorption band and exhibits a short life. The better hydrolytic
stability of 3 over 3m may be due to the rigid macrocycle
Fig. 4 The UV-Vis absorption spectra of 3m and 3 (top), 6m and 6 (bottom) in
THF.
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 6008–6015 | 6011

View Article Online
Paper
RSC Advances
Fig. 7 I–V curves measured over the microfibers of 3 before (1) and after (2)
being exposed to saturated iodine vapor for 5 min. The microfibers were drop
cast onto a gold electrode-pair (3 mm gap, 14 mm long) fabricated on the surface
of oxidized silicon.
Fig. 6 SEM images of self-assembled objects from macrocycles 3 (a), 6 (b), and
their model compounds 3m (c), 6m (d).
structure which blocks the nucelophilic attack of water on the
carbocation.
However, the microfibers demonstrated an increase in current
of about one order of magnitude after exposure to saturated
iodine vapor for 5 min. The enhancement in the conductivity
implies not only a certain level of charge carrier doping, but
more importantly efficient charge delocalization along the p–p
stacking of the building block molecules (here presumably the
long axis of the nanofiber).14 To achieve a high conductivity
demands both a high concentration of carriers and continuous
transport pathways. The p–p stacking induced charge trans-
port has been extensively proven for various p-conjugated
molecules using both experimental and theoretical investiga-
tions.15 Moreover, strong p–p stacking within the microfibers
was also evident by the observation that the nanofibers as
fabricated were completely nonfluorescent, i.e., no emission
was detected (using either a fluorometer or fluorescence
microscope) even under high intensity of UV irradiation.
However, the molecules demonstrated detectable emission
when dissolved in a solution. The non-emitting phase is due to
the forbidden low-energy excitonic transition as previously
observed for other p-molecules.16
Solution self-assembly
When a poor solvent such as hexane or methanol is slowly
diffused into the THF solution of 3 or 6, self-assembly occurs,
leading to opaque solutions in less than one day. SEM images
(Fig. 6 and Fig. S14 ) show that 3 self-assembles into rigid rods
3
that are approximately 50 mm long and have an outer diameter
of y1 mm. The XRD of the fibers shows a distinct diffraction
peak at 3.63 Å (Fig. S17 ), which is consistent with the p–p
3
stacking distances. Macrocycle 6, on the other hand, self-
assembles into plates/panels with dimensions of around 100
mm 6 10 mm, which stack into bundles. The panels have a
pointed triangular edge at the two longitudinal ends. The XRD
pattern of the self-assembled plates (Fig. S18 ) shows a
3
diffraction peak at 3.42 Å, indicating a significantly shorter
p–p stacking distance in 6 which is close to the graphite layer
spacing of 3.35 Å. Macrocycle 6 also shows a better long range
order. However, when hexane is diffused into a THF solution
of 3m or 6m, no aggregates are visible even after 10 days. If
methanol is used as the poor solvent, self-assembly does occur
after four days and the SEM images show the formation of 2D
Conclusions
ribbons (Fig. 6c and 6d). The XRD of 3m (Fig. S19 ) shows a
3
much large p–p stacking distance of 4.17 Å. The faster self-
assembly process of the macrocycles reflects their stronger p–p
stacking interactions brought about by the higher number of
aromatic p-systems and their relatively flat structure.
In summary, two shape-persistent triphenylene-based macro-
cycles have been synthesized. Both quantum mechanical
calculations and 2D NMR measurements show that 3 adopts
the 3o conformation and 6 adopts the 6i conformation. The
interconversion of the conformers is restricted due to the rigid
shape-persistent ring structure. Under acidic conditions, both
3 and 3m show a significantly red-shifted absorption band
which is attributed to a charge transfer transition mediated by
the triphenylene p system. 6 and 6m, on the other hand, lack
such a charge transfer transition and hydrolyse rather quickly
under acidic conditions. Compared to the model compounds,
the macrocycles exhibit a stronger propensity for solution self-
assembly. While 3 self-assembles into nano and microrods, 6
p-Stacking among triphenylene rings is believed to be
responsible for the facile self-assembly of both macrocycles
and the model compounds. The triphenylene stack may bring
charge mobility to the assembled micro-objects. To test this
hypothesis, the conductivity of the pristine microfibers of 3
with or without doping is measured. As shown in Fig. 7, the
pristine fibers exhibited a minimal current of only about 1 pA
under a bias of 10 V. This is consistent with the high purity of
the microfibers thus fabricated. Such poor conductivity is
common for single crystalline organic semiconductors.
6012 | RSC Adv., 2013, 3, 6008–6015
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
aggregates into panels with a significantly higher crystalline
order. The sizable electrical conductivity after doping and the
complete fluorescence quenching of the self-assembled micro-
objects indicate that strong p–p stacking among the tripheny-
lene rings is responsible for their facile self-assembly process.
Compound 2
BuLi (1.18 mL, 2.5M in hexane, 2.96 mmol) was added
dropwise to a chilled solution of 1 (1.05 g, 1.28 mmol) in dry
THF (90 mL) at 220 uC. The reaction mixture was stirred at
220 uC for 1 h. Then anhydrous DMF (0.397 mL, 5.15 mmol)
was added dropwise. The reaction mixture was allowed to
warm to room temperature overnight. After being quenched
with 100 mL water, the solution was extracted with dichlor-
omethane. The organic layer was washed twice with water,
dried over MgSO4, filtered and concentrated under reduced
pressure. The crude product was purified by chromatography
on a silica gel column using dichloromethane : hexanes (1 : 1
v/v) as the eluent, affording 2 as a yellow solid (0.71 g, 77%). 1H
NMR (400 MHz, CDCl3) d: 10.59 (s, 2H), 9.01 (s, 2H), 7.81
(s,2H), 7.73 (s, 2H), 4.24 (t, J = 8.0 Hz, 4H), 4.13 (s, 6H), 1.95 (m,
4H), 1.25 (m, 36H), 0.86 (t, J = 8.0 Hz, 6H); 13C NMR (100 MHz,
CDCl3) d: 189.5, 159.4, 150.8, 134.7, 125.1, 124.8, 124.1, 122.7,
107.6, 103.7, 69.5, 55.7, 31.9, 29.7, 29.5, 29.3, 26.2, 22.7, 14.1;
anal. calcd for C46H64O6: C, 77.49; H, 9.05. Found: C, 76.78; H,
8.90.
Experimental section
Solvents were distilled from appropriate drying agents under
inert conditions prior to use. Unless otherwise stated, all other
chemicals were purchased from commercial sources and used
without further purification. All reactions were carried out
under nitrogen using standard Schlenk techniques. 1H NMR
and 13C NMR spectra were collected using a Varian INOVA 400
MHz spectrometer. 2D NOESY NMR spectra were collected
using a 14.1T Varian INOVA spectrometer equipped with a
cryogenically cooled probe. The NOESY spectra were collected
using the Watergate NOESY pulse sequence with a mixing time
of 400 ms, 16 transients and ni = 1024. Chemical shifts were
referenced to residual protio-solvent signals. UV-Vis absorp-
tion spectra (1026 M THF solution) were measured using a
Hewlett–Packard 8452A diode array spectrophotometer. A
Voyager DE Pro (Perceptive Biosystems/ABI) MALDI-TOF mass
spectrometer was used for mass spectra measurements,
operating in the reflector mode. Compounds were spotted
for MALDI-TOF MS analysis in a matrix containing dithranol
freshly dissolved in distilled THF. A field-emission scanning
electron microscope (SEM) XL30 (FEI Company, Hillsboro, OR)
was used with an accelerating voltage of 15.0 kV. Specimens
were prepared by placing drops of suspension either on a
double sided carbon tape or on a mica substrate. Specimens
were used with or without sputter-coated Au–Pd. Self-
assembled aggregates were obtained by the diffusion of
hexane or methanol vapor through the THF solution of the
respective compounds. XRD measurements were performed
using a Rigaku Miniflex automated diffractometer with Cu-Ka
radiation. Electrode fabrication and I–V measurements were
done following the same procedures as previously described.14
Compound 3m
2 (0.12 g, 0.17 mmol), molecular sieves and freshly distilled
aniline (76 mL, 0.84 mmol) were added to a flask equipped with
a Dean–Stark trap and reflux condenser under a nitrogen
atmosphere. 10 mL anhydrous toluene and 50 mL anhydrous
ethanol were added and the reaction mixture was refluxed for
2
days. After cooling to room temperature, the yellow
precipitate was collected by filtration. The precipitate was
washed multiple times with methanol and then hexane to
yield 3m as a yellow solid (0.09 g, 62%). H NMR (400 MHz,
1
THF-d8) d: 9.45 (s, 2H), 9.07 (s, 2H), 8.04 (s, 2H), 7.99 (s, 2H),
7.37 (br, 4H), 7.28 (br, 4H), 7.18 (br, 2H), 4.28 (t, J = 4.8 Hz,
4H), 4.16 (s, 6H), 1.93 (m, 4H), 1.31 (m, 36H), 0.89 (t, J = 5.0 Hz,
6H); 13C NMR (100 MHz, THF-d8) d: 159.2, 156.3, 154.4, 151.7,
129.9, 126.4, 125.9, 125.4, 124.3, 123.5, 122.1, 108.6, 104.7,
70.1, 56.2, 33.1, 30.9, 30.7, 30.6, 23.8,14.6; MALDI-TOF MS:
calcd for C58H74N2O4: 863.22, found: 863.22; anal. calcd for
C58H74N2O4: C, 80.70; H, 8.64; N, 3.25. Found: C, 80.20; H,
8.83; N, 3.17.
Compound 3
Compound 1
To a dry flask equipped with a Dean–Stark trap, 2 (0.1 g, 0.14
mmol), m-phenylenediamine (0.015 g, 0.14 mmol), 10 mL
anhydrous toluene, 50 mL anhydrous ethanol and molecular
sieves were added. The reaction mixture was refluxed for 4
days. After cooling to room temperature, the yellow precipitate
was collected by filtration. The precipitate was washed multi-
ple times with methanol followed by hexane to yield 3 as a
To a solution of 2,3-bis(dodecyloxy)-6,11-dimethoxytripheny-
lene (2.05 g, 3.12 mmol) in 100 mL dichloromethane was
added dropwise along with bromine (0.32 mL, 6.24 mmol) in
10 mL dichloromethane. The reaction mixture was stirred at
room temperature for 2 h and was then poured into water. The
organic layer was washed with 5% sodium thiosulphate
solution and water and dried over MgSO4. After removal of
the solvent, the crude product was purified by recrystallization
from dichloromethane/methanol to yield 1 as a yellow solid
(2.0 g, 79%). 1H NMR (400 MHz, CDCl3) d: 8.42 (s, 2H), 7.81 (s,
2H), 7.67 (s, 2H), 4.24 (t, J = 6.2 Hz, 4H), 4.09 (s, 6H), 1.94 (m,
4H), 1.25 (m, 36H), 0.86 (t, J = 7.2 Hz, 6H); 13C NMR (100 MHz,
CDCl3) d: 154.0, 149.8, 129.0, 127.4, 123.7, 122.8, 111.6, 106.8,
103.7, 69.5, 56.2, 29.7 (2), 29.6, 29.4 (2), 26.2, 22.7, 14.1; anal.
calcd for C44H62Br2O4: C, 64.86; H, 7.67. Found: C, 64.78; H,
7.52.
1
yellow solid (0.09 g, 40%). H NMR (400 MHz, THF-d8) d: 9.77
(s, 4H), 9.34 (s, 4H), 8.08 (s, 4H), 8.02 (s, 4H), 7.9 (br, 2H), 7.47
(br, 2H), 7.33 (br, 4H), 4.29 (t, J = 6.2 Hz, 8H), 4.21 (s, 12H),
1.94 (m, 8H), 1.32 (m, 72H), 0.9 (t, J = 5.8 Hz, 12H); 13C NMR
(100 MHz, THF-d8) d: 159.3, 154.8, 154.7, 153.5, 151.9, 133.7,
126.2, 125.8, 124.6, 109.3, 104.8, 70.4, 57.5, 30.9, 30.8, 30.7,
30.5, 27.4, 23.7,14.5; MALDI-TOF MS: calcd for C104H136N4O8:
1570.21, found: 1570.21; anal. calcd for C104H136N4O8: C,
79.55; H, 8.73; N, 3.57. Found: C, 79.11; H, 8.84; N, 3.39.
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 6008–6015 | 6013

View Article Online
Paper
RSC Advances
Compound 4
Compound 6
A toluene solution (50 mL) of tris(dibenzylideneacetone)
dipalladium(0) (0.14 g, 0.153 mmol) and rac-BINAP (0.19 g,
0.305 mmol) was stirred at 110 uC for 30 min under argon.
After cooling to room temperature, benzophenone imine (0.26
mL, 1.55 mmol), sodium tert-butoxide (0.15 g, 1.55 mmol) and
1 (0.5 g, 0.614 mmol) were added to the reaction mixture and
stirred at 110 uC overnight. The mixture was cooled and
evaporated to dryness. The crude product was purified by
column chromatography on a silica gel with ethyl acetate :
hexane (1 : 3 v/v) as the eluent, affording 4 as a yellow solid
To a suspension of 5 (0.1 g, 0.131 mmol) in anhydrous ethanol
(50 mL) and anhydrous toluene (10 mL), triethylamine (0.09
mL, 0.647 mmol) was added. The mixture was stirred at room
temperature for 15 min under
a nitrogen atmosphere.
Molecular sieves and isophthalaldehyde (0.0176 g, 0.131
mmol) were then added and a Dean–Stark trap and reflux
condenser were attached. The reaction mixture was refluxed
for 4 days under a nitrogen atmosphere. After cooling to room
temperature, the yellow precipitate was collected by filtration.
The precipitate was washed multiple times with methanol and
1
1
(0.61 g, 97%). H NMR (400 MHz, CDCl3) d: 7.8–7.78 (m, 4H),
then hexane to yield 6 as a yellow solid (0.1 g, 48%). H NMR
7.75 (s, 2H), 7.57 (s, 2H), 7.52 (s, 2H), 7.48 – 7.38 (m, 6H), 7.15
(m, 10H), 4.16 (t, J = 6.6 Hz, 4H), 3.84 (s, 6H), 1.89 (m, 4H), 1.24
(m, 36H), 0.85 (t, J = 6.8 Hz, 6H); 13C NMR (100 MHz, CDCl3) d:
170.3, 149.1, 149.0, 140.9, 139.5, 136.8, 130.7, 129.5, 128.8,
128.5, 128.1, 127.7 (2), 125.6, 123.8, 123.4, 115.0, 107.3, 103.7,
69.7, 55.6, 31.9, 29.7 (3), 29.5, 29.4 (2), 26.2, 22.7, 14.1; anal.
calcd for C70H82N2O4: C, 82.80; H, 8.14; N, 2.76. Found: C,
82.57; H, 8.00; N, 2.69.
(400 MHz, THF-d8) d: 9.01 (s, 4H), 8.65 (s, 2H), 8.41 (s, 4H),
8.19 (d, J = 7.6 Hz, 4H), 8.04 (s, 4H), 8.01 (s, 4H), 7.6 (t, J = 7.8
Hz, 2H), 4.27 (t, J = 6.2 Hz, 8H), 4.13 (s, 12H), 1.94 (m, 8H), 1.32
(m, 72H), 0.9 (t, J = 6 Hz, 12H); 13C NMR (100 MHz, THF-d8) d:,
161.4, 153.3, 151.2, 142.2, 139.0, 131.6, 131.1, 129.9, 129.3,
125.5, 125.0, 117.4, 109.0, 106.5, 70.4, 56.8, 33.0, 30.85, 30.7,
30.5, 27.4, 23.7, 14.5; MALDI-TOF MS: calcd for C104H136N4O8:
1570.21, found: 1570.21; anal. calcd for C104H136N4O8: C,
79.55; H, 8.73; N, 3.57. Found: C, 78.27; H, 8.45; N, 3.42.
Compound 5
QM energy calculations
A 2.0 M HCl solution (0.7 mL, 1.4 mmol) was added dropwise
to a THF solution (10 mL) of 4 (0.6 g, 0.591 mmol) at room
temperature. The reaction mixture was stirred at room
temperature for 1 h. The resulting precipitate was filtered
and washed with hexane (10 mL 6 3), dried under vacuum to
yield 5 as a light yellow solid (0.4 g, 89%).1H NMR (400 MHz,
DMSO-d6) d: 8.09 (s, 2H), 8.04 (s, 2H), 8.01 (s, 2H), 4.26 (t, J =
6.2 Hz, 4H), 4.12 (s, 6H), 3.58 (br, 4H), 1.83 (m, 4H), 1.25 (m,
36H), 0.85 (t, J = 6.8 Hz, 6H); 13C NMR (100 MHz, DMSO-d6) d:
149.2, 149.0, 128.3, 123.1, 122.2, 112.5, 112.3, 107.7, 104.7,
68.9, 56.1, 31.0, 28.8, 28.7, 28.6, 28.3, 25.5, 21.7, 17.4, 13.5.
The model structures (ethoxy analogs) of 3i, 3o, 3io, 6i, 6o, and
6io were built in Maestro and were minimized with the OPLS
¨
force field using the MacroModel software in the Schrodinger
software suite.17 The DFT B3LYP18 functional method with the
6-31G** basis set was employed to calculate the QM energy
¨
using the Jaguar package in the Schrodinger software. The
accuracy level of SCF was selected to its highest level (fully
analytic). Due to the large number of atoms in the model
compounds, the STO-3G level of theory was used for
minimization, followed by single point energy calculations.
Compound 6m
To a suspension of 5 (0.09 g, 0.012 mmol) in anhydrous
ethanol (12 mL), triethylamine (0.08 mL, 0.573 mmol) was
added. The resulting mixture was stirred at room temperature
for 15 min under a nitrogen atmosphere. Molecular sieves and
benzaldehyde (0.06 mL, 0.588 mmol) in anhydrous ethanol (2
mL) were then added. With a Dean–Stark trap and reflux
condenser attached, the reaction mixture was refluxed for 2
days under nitrogen atmosphere. After cooling to room
temperature, the yellow precipitate was collected by filtration.
The precipitate was washed multiple times with methanol and
Acknowledgements
We thank the National Science Foundation (DMR 0804158)
and the Army Research Office (W911NF-10-1-0476) for sup-
porting this work.
Notes and references
1 (a) W. Zhang and J. S. Moore, Angew. Chem., Int. Ed., 2006,
45, 4416; (b) M. J. MacLachlan, Pure Appl. Chem., 2006, 78,
873; (c) C. Ma, A. Lo, A. Abdolmaleki and M. J. MacLachlan,
Org. Lett., 2004, 6, 3841; (d) C. S. Hartley and J. S. Moore, J.
Am. Chem. Soc., 2007, 129, 11682; (e) S. Klyatskaya,
1
then hexane to yield 6m as a yellow solid (0.077 g, 75%). H
NMR (400 MHz, THF-d8) d: 8.71 (s, 2H), 8.22 (s, 2H), 8.05 (s,
2H), 7.99 (br, 6H), 7.46 (br, 6H), 4.27 (t, J = 6 Hz, 4H), 4.06 (s,
6H), 1.92 (m, 4H), 1.31 (m, 36H), 0.9 (t, J = 6.0 Hz, 6H); 13C
NMR (100 MHz, THF-d8) d:, 161.9, 153.0, 150.9, 143.3, 138.3,
131.9, 129.8, 129.5, 128. 8, 125.3, 124.8, 115.8, 108.4, 105.8,
70.1, 56.5, 33.1, 31.0, 30.9, 30.9, 30.8, 30.7, 30.6, 27.5, 23.8,
14.6; MALDI-TOF MS: calcd for C58H74N2O4: 863.22, found:
863.22; anal. calcd for C58H74N2O4: C, 80.70; H, 8.64; N, 3.25.
Found: C, 80.13; H, 8.81; N, 3.18.
¨
D. Dingenouts, C. Rosenauer, B. Mu¨ller and S. Hoger, J.
Am. Chem. Soc., 2006, 128, 3150; (f) S. Hoger, Pure Appl.
¨
Chem., 2010, 82, 821; (g) R. R. Tykwinski, M. Gholami,
S. Eisler, Y. Zhao, F. Melin and L. Echegoyen, Pure Appl.
Chem., 2008, 80, 621.
2 (a) D. Zhao and J. S. Moore, J. Org. Chem., 2002, 67, 3548;
(b) S. H. Seo, T. V. Jones, H. Seyler, J. O. Peters, T. H. Kim, J.
Y. Chang and G. N. Tew, J. Am. Chem. Soc., 2006, 128, 9264;
(c) W. Pisula, M. Kastler, C. Yang, V. Enkelmann and
6014 | RSC Adv., 2013, 3, 6008–6015
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
K. Mu¨llen, Chem.–Asian J., 2007, 2, 51; (d) S. Chen, Q. Yan,
T. Li and D. Zhao, Org. Lett., 2010, 12, 4784.
7 Y. Che, X. Yang, Z. Zhang, J. Zuo, J. S. Moore and L. Zang,
Chem. Commun., 2010, 46, 4127.
¨
3 (a) M. D. Watson, A. Fechtenkotter and K. Mu¨llen, Chem.
8 D. Wang, J. Hsu, M. Bagui, V. Dusevich, Y. Wang, Y. Liu, A.
J. Holder and Z. Peng, Tetrahedron Lett., 2009, 50, 2147.
9 (a) M. Bagui, J. S. Melinger, S. Chakraborty, A. Keightley
and Z. Peng, Tetrahedron, 2009, 65, 1247; (b) M. Bagui,
T. Dutta, S. Chakraborty, J. S. Melinger, H. Zhong,
A. Keightley and Z. Peng, J. Phys. Chem. A, 2011, 115, 1579.
10 (a) C. Chou, D. Wang, M. Bagui, J. Hsu, S. Chakraborty and
Z. Peng, J. Lumin., 2010, 130, 986; (b) C. Chou, D. Wang,
J. Hsu, Y. Liu and Z. Peng, Synth. Met., 2009, 159, 1657.
11 J. Kang, D. Wang, M. Bagui, S. Chakroborty and Z. Peng,
Lett. Org. Chem., 2006, 3, 674.
Rev., 2001, 101, 1267; (b) S. Grimme, Angew. Chem., Int. Ed.,
2008, 47, 3430; (c) D. Adam, P. Schuhmacher, J. Simmerer,
L. Haussling, K. Siemensmeyer, K. H. Etzbachi,
H. Ringsdorf and D. Haarer, Nature, 1994, 371, 141.
4 (a) W. Pisula, X. Feng and K. Mu¨llen, Chem. Mater., 2011,
23, 554; (b) B. R. Kaafarani, Chem. Mater., 2011, 23, 378.
5 (a) T. Naddo, Y. Che, W. Zhang, K. Balakrishnan, X. Yang,
J. Zhao, J. S. Moore and L. Zang, J. Am. Chem. Soc., 2007,
129, 6978; (b) L. Zang, Y. Che and J. S. Moore, Acc. Chem.
Res., 2008, 41, 1596; (c) W. Zhao, Q. Tang, H. S. Chan, J. Xu,
K. Y. Lo and Q. Miao, Chem. Commun., 2008, 4324–4326; (d)
S. Sergeyev, W. Pisula and H. Geerts, Chem. Soc. Rev., 2007,
36, 1902–1929; (e) B. Walker, C. Kim and T. -Q. Nguyen,
Chem. Mater., 2011, 23, 470–482.
12 C. Long, J. Kim, T. Ishizuka, Y. Honsho, A. Saeki, S. Seki,
H. Ihee and D. Jiang, J. Am. Chem. Soc., 2009, 131, 7287.
13 Very weak NOE contacts between H2, H1 are also observed,
indicating the existence of some minor 3i conformer (Fig.
¨
6 (a) S.-H. Jung, W. Pisula, A. Rouhanipour, H. J. Rader,
S5 ).
3
J. Jacob and K. Mu¨llen, Angew. Chem., Int. Ed., 2006, 45,
4685; (b) W. Huang, M. Wang, C. Du, Y. Chen, R. Qin, L. Su,
C. Zhang, Z. Liu, C. Li and Z. Bo, Chem.–Eur. J., 2011, 17,
440; (c) W. Zhang, H.-M. Cho and J. S. Moore, Org. Synth.,
2007, 84, 177; (d) W. Zhang and J. S. Moore, J. Am. Chem.
14 Y. Che, X. Yang, G. Liu, C. Yu, H. Ji, J. Zuo, J. Zhao and
L. Zang, J. Am. Chem. Soc., 2010, 132, 5743.
´
15 J.-L. Bredas, D. Beljonne, V. Coropceanu and J. Cornil,
Chem. Rev., 2004, 104, 4971.
16 J. Wu, W. Pisula and K. Mu¨llen, Chem. Rev., 2007, 107, 718.
17 Protein Preparation Wizard, Maestro, MacroModel, Phase,
¨
Soc., 2004, 126, 12796; (e) H. Wettach, S. Hoger,
¨
D. Chaudhuri, J. M. Lupton, F. Liu, E. M. Lupton,
S. Tretiak, G. Wang, M. Li, S. De Feyter, S. Fischer and
Induced Fit, Jaguar, and Glide; Schrodinger, LLC: Portland,
OR, 2009.
¨
S. Forster, J. Mater. Chem., 2011, 21, 1404; (f) B. N. Boden,
18 (a) B. G. Johnson, P. M. W. Gill and J. A. Pople, J. Chem.
Phys., 1993, 98, 5612; (b) C. Lee, W. Yang and R. G. Parr,
Phys. Rev. B, 1988, 37, 785; (c) A. D. Becke, J. Chem. Phys.,
1993, 98, 1372.
A. Abdolmaleki, C. T.-Z. Ma and M. J. MacLachlan, Can. J.
Chem., 2008, 86, 50; (g) J. Kaleta and C. Mazal, Org. Lett.,
2011, 13, 1326.
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 6008–6015 | 6015