7-Azaindolyl- and indolyl-functionalized starburst molecules with a 1,3,5-triazine or a benzene core - Syntheses and luminescence

Article abstract of DOI:10.1139/V06-022

The syntheses of several new 7-azaindolyl-, indolyl-, and 3-methylindolyl-functionalized starburst molecules that contain either a 1,3,5-triazine, a benzene, a 2,4,6-triphenyl-2,4,6-triazine, a 1,3,5-triphenylbenzene, a 2,4,6-tris(biphenyl)-1,3,5-triazine, or a 1,3,5-tris(biphenyl)benzene core have been achieved. The synthetic methods used for these new compounds involve mostly Ullmann condensation and Suzuki coupling reactions. The thermal properties of the new molecules have been found to be highly dependent on the molecular weight and the central core. Glass transition temperatures greater than 150 °C were observed for the large starburst molecules based on the 2,4,6-tris(biphenyl)-1,3,5-triazine or the 1,3,5-tris(biphenyl)benzene core. The new starburst molecules are luminescent in the violet-blue region in solution and in the solid state. The emission energy and the quantum efficiency were found to be highly dependent on the central core and the functional group. In general, the triazine-based molecules were found to have a smaller HOMO-LUMO energy gap than the benzene-based analogues. The triphenyltriazine- or triphenylbenzene-based molecules were found to display a higher emission quantum efficiency than the large tris(biphenyl)triazine- or tris(biphenyl)benzene-based molecules.

Full text of DOI:10.1139/V06-022

477
7-Azaindolyl- and indolyl-functionalized starburst
molecules with a 1,3,5-triazine or a benzene core —
Syntheses and luminescence1
Wei-Hua Huang, Wen-Li Jia, and Suning Wang
Abstract: The syntheses of several new 7-azaindolyl-, indolyl-, and 3-methylindolyl-functionalized starburst molecules
that contain either a 1,3,5-triazine, a benzene, a 2,4,6-triphenyl-2,4,6-triazine, a 1,3,5-triphenylbenzene, a 2,4,6-
tris(biphenyl)-1,3,5-triazine, or a 1,3,5-tris(biphenyl)benzene core have been achieved. The synthetic methods used for
these new compounds involve mostly Ullmann condensation and Suzuki coupling reactions. The thermal properties of
the new molecules have been found to be highly dependent on the molecular weight and the central core. Glass transi-
tion temperatures greater than 150 °C were observed for the large starburst molecules based on the 2,4,6-tris(biphenyl)-
1,3,5-triazine or the 1,3,5-tris(biphenyl)benzene core. The new starburst molecules are luminescent in the violet-blue
region in solution and in the solid state. The emission energy and the quantum efficiency were found to be highly de-
pendent on the central core and the functional group. In general, the triazine-based molecules were found to have a
smaller HOMO–LUMO energy gap than the benzene-based analogues. The triphenyltriazine- or triphenylbenzene-based
molecules were found to display a higher emission quantum efficiency than the large tris(biphenyl)triazine- or
tris(biphenyl)benzene-based molecules.
Key words: 7-azaindolyl, indolyl derivative, triazine and benzene cores, starburst molecule, blue luminescence.
Résumé : On a réalisé la synthèse de plusieurs nouvelles molécules à base d’indolyle, de 3-méthylindolye et de
7-azaindolyle fonctionnalisés qui donnent l’impression d’une explosion d’étoiles et qui contiennent comme partie cen-
trale une 1,3,5-triazine, un benzène, un 2,4,6-triphényl-2,4,6-triazine, un 1,3,5-triphénylbenzène, un 2,4,6-tris(biphényl)-
1,3,5-triazine ou un 1,3,5-tris(biphényl)benzène. Les méthodes de synthèse utilisées pour ces nouveaux produits impli-
quent principalement des réactions de condensation de Ullmann et de couplage de Suzuki. On a trouvé que les proprié-
tés des nouvelles molécules dépendent fortement du poids moléculaire et de la nature de la partie centrale. On a
observé des températures de transition de verre supérieures à 150 °C pour les grosses molécules qui donnent
l’impression d’une explosion d’étoiles à base d’une partie centrale de 2,4,6-tris(biphényl)-1,3,5-triazine ou de 1,3,5-
tris(biphényl)benzène. Tant en solution qu’à l’état solide, les nouvelles molécules qui donnent l’impression d’une explo-
sion d’étoiles sont luminescentes dans la région du violet-bleu. On a trouvé que l’énergie d’émission et l’efficacité
quantique dépendent fortement de la nature de la partie centrale et du groupe fonctionnel. On a trouvé que, en général,
l’écart entre les énergies des orbitales moléculaires hautes occupées (HO) et basses vacantes (BV) des molécules à base
de triazine sont plus faibles que celles de leurs analogues à base de benzène. On a observé que les efficacités quanti-
ques d’émission des molécules à base de triphényltriazine ou de triphénylbenzène sont plus grandes que celles des mo-
lécules à base de tris(biphényl)triazine ou de tris(biphényl)benzène.
Mots clés : 7-azaindolyle, dérivé de l’indolyle, partie centrale à base de triazine et de benzène, molécule qui donne
l’impression d’une explosion d’étoiles, luminescence bleue.
[Traduit par la Rédaction] Huang et al. 485
Introduction
7-azaindolyl-functionalized aromatic molecules can be used
as effective blue emitters in OLEDs (2). In addition, the 2,2′-
dipyridylamino and 7-azaindolyl functionalities allow the at-
tachment of metal ions to the aromatic molecules for emis-
sion color tuning and supramolecular assembly (3). Several
series of starburst molecules with a C3 symmetry that are
functionalized by 2,2′-dipyridylamino groups on a 1,3,5-
triazine, benzene, triphenyltriazine, or triphenylbenzene core
Starburst or star-shaped molecules have been shown to
have the tendency to form amorphous films with a relatively
high Tg, a property that is highly desired for applications of
organic materials in photonic and optoelectronic devices
such as organic light emitting diodes (OLEDs) (1). Our
group has recently demonstrated that 2,2′-dipyridylamino- or
Received 1 June 2005. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on 31 March 2006.
W.-H. Huang, W.-L. Jia, and S. Wang.2 Department of Chemistry, Queen University, Kingston, ON K7L 3N6, Canada.
1This article is part of a Special Issue dedicated to Professor Walter A. Szarek.
2Corresponding author (e-mail: wangs@chem.queensu.ca).
Can. J. Chem. 84: 477–485 (2006)
doi:10.1139/V06-022
© 2006 NRC Canada

478
Can. J. Chem. Vol. 84, 2006
were reported previously by our group (3). These molecules
indeed displayed good film-forming properties and were
shown to be effective as emitters for blue OLEDs (2). Our
current work focuses on the development of starburst mole-
cules that are functionalized by the 7-azaindolyl group. In
contrast to the 2,2′-dipyridylamino group, which can act as
either a chelate ligand or a terminal ligand to metal ions, the
7-azaindolyl group can only bind to the metal center as a ter-
minal ligand, which offers new binding modes for coordina-
tion compounds and supramolecular assembly (4). In our
previous work, three starburst molecules containing the
7-azaindolyl group were investigated (1,3,5-tris(N-7-
azaindolyl)benzene, 1,2,4,5-tetrakis(N-7-azaindolyl)benzene
(4), and hexakis(p-N-7-azaindolylphenyl)benzene (3e)). Al-
though these three compounds were found to form unusual
complexes with a variety of metal ions and may be used as
blue emitters in OLEDs (2b), they have the tendency to re-
main crystalline, which limits their applications in OLEDs.
Our current efforts therefore focus on the synthesis of large
starburst molecules containing the 7-azaindolyl or the
indolyl group with the aim to enhance the Tg of the amor-
phous films. In this report, the syntheses, characterization,
and thermal and luminescent properties of several new
starburst molecules are presented.
Synthesis of 2,4,6-tris(1-N-7-azaindolyl)-1,3,5-triazine (1)
To a vigorously stirred solution of 7-azaindole (0.400 g,
3.4 mmol) in dry THF (15 mL), n-butyllithium solution in
hexanes (1.6 mol/L, 2.2 mL, 3.5 mmol) was added slowly at
–78 °C. After stirring for 30 min at –78 °C, a cyanuric chlo-
ride (0.186 g, 1.0 mmol) solution in THF (10 mL) was
slowly added to the mixture. A yellow solid quickly ap-
peared. After stirring for 1 h at –78 °C, the mixture was
warmed to room temperature and stirred for 12 h. The mix-
ture was then concentrated under vacuum. Water (40 mL)
was added and the crude product was extracted with CH2Cl2
(3 × 40 mL). The organic layer was dried over MgSO4 and
then concentrated under vacuum to give a yellow solid.
Recrystallization from CH2Cl2 and hexanes produced a
white solid (0.350 g, yield 82%, mp 289 °C). 1H NMR
(400 MHz, CDCl3, 298 K, ppm) δ: 8.70 (d, J = 4.8 Hz, 3H,
7-aza), 8.69 (d, J = 4.0 Hz, 3H, 7-aza), 8.00 (d, J = 7.8 Hz,
3H, 7-aza), 7.31 (dd, J1 = 7.8 Hz, J2 = 4.8 Hz, 3H, 7-aza),
6.80 (d, J = 4.0 Hz, 3H, 7-aza). 13C NMR (400 MHz,
CDCl3, 298 K, ppm) δ: 163.5, 148.7, 145.1, 129.3, 127.1,
124.6, 118.9, 106.4. HRMS (ESI, m/z) calcd. for C24H16N9:
430.1523; found: 430.1513.
Synthesis of 1,3,5-tris[p-(1-N-7-azaindolyl)phenyl]ben-
zene (2)
A mixture of 1,3,5-tris(p-bromophenyl)benzene (1.100 g,
2.02 mmol), 7-azaindole (1.071 g, 9.08 mmol), CuI
(0.115 g, 6.05 mmol), Cs2CO3 (4.160 g, 12.8 mmol), 1,10-
phenanthroline (0.218 g, 1.21 mmol), and DMF (5 mL) was
heated to 150 °C. After stirring for 48 h at 150 °C, the mix-
ture was cooled to room temperature and was dissolved in
CH2Cl2 (50 mL) and water (50 mL). The water layer was
separated and was extracted with CH2Cl2 (2 × 50 mL). The
combined organic layer was dried over MgSO4, and the sol-
vents were evaporated under reduced pressure to give a
brown solid that was purified by a chromatographic column
using THF–hexanes (1:2.5) as the eluent to obtain a white
solid. Recrystallization from THF and hexanes yielded a col-
orless compound (0.842 g, yield 64%, mp 287 °C). 1H NMR
(300 MHz, CDCl3, 298 K, ppm) δ: 8.45 (d, J = 4.7 Hz, 3H,
7-aza), 8.03 (d, J = 7.8 Hz, 3H, 7-aza), 7.96 (d, J = 8.8 Hz,
6H, ph), 7.92 (s, 3H, ph), 7.90 (d, J = 8.8 Hz, 6H, ph), 7.63
(d, J = 3.6 Hz, 3H, 7-aza), 7.20 (dd, J1 = 7.8 Hz, J2 =
4.7 Hz, 3H, 7-aza), 6.71 (d, J = 3.6 Hz, 3H, 7-aza). 13C
NMR (300 MHz, CDCl3, 298 K, ppm) δ: 148.3, 144.4,
142.5, 139.6, 138.7, 129.8, 128.9, 128.4, 125.7, 125.0,
122.3, 117.5, 102.6. HRMS (ESI, m/z) calcd. for C45H31N6:
655.2604; found: 655.2605.
Experimental
All reactions were carried out under a nitrogen atmo-
sphere and monitored by thin layer chromatography (TLC).
All starting materials were purchased from the Sigma-
Aldrich Chemical Company and used without further purifi-
cation. NMR spectra were recorded on Bruker 300, 400, or
600 MHz spectrometers. High-resolution mass spectra were
recorded on an Applied Biosystems/MDS Sciex QSTAR XL
instrument with an Agilent HP1100 Cap-LC system by elec-
tron spray ionization (ESI) and a Waters/Micromass GC–
TOF instrument with an Agilent HP6890 GC system by
electron impact (EI). Elemental analyses were performed by
Canadian Microanalytical Service Ltd., Delta, British
Columbia. TLC was carried out on SiO2. Column chromato-
graphic separation was carried out on silica (silica gel, 70–
230 mesh 60). Melting points were measured on a Fisher-
Johns melting point apparatus. Excitation and emission spec-
tra were recorded on a Photon Technologies International
QuantaMaster Model C-60 spectrometer at room tempera-
ture. DSCs were performed on a PerkinElmer Pyris DSC 6.
Heating scans were performed at 10° min–1. Cooling scans
were performed at 20° min–1. Cyclic voltammetry was per-
formed on a BAS CV-50W analyzer with scan rates of
100 mV s–1. The electrolytic cell used was a conventional
three-compartment cell with a Pt working electrode, Pt aux-
iliary electrode, and Ag/AgCl reference electrode. All exper-
iments were performed at room temperature in CH2Cl2 or
THF with 0.10 mol/L tetrabutylammonium hexafluorophos-
phate as the supporting electrolyte. The ferrocenium/
ferrocene couple was used as the internal standard. 1,3,5-
Tris(p-bromophenyl)benzene and 2,4,6-tri(p-bromophenyl)-
1,3,5-triazine were synthesized by a modified literature
method (2a). Melting points were determined on a Fisher-
Johns melting point apparatus (maximum temperature
330 °C) and were confirmed by DSC measurements.
Synthesis of 2,4,6-tris[p-(1-N-7-azaindolyl)phenyl]-1,3,5-
triazine (3)
A
mixture of 2,4,6-tri(p-bromophenyl)-1,3,5-triazine
(0.789 g, 1.46 mmol), 7-azaindole (0.776 g, 6.58 mmol),
CuI (0.083 g, 0.44 mmol), Cs2CO3 (3.00 g, 9.2 mmol), 1,10-
phenanthroline (0.158 g, 0.87 mmol), and DMF (4 mL) was
heated to 150 °C and stirred for 48 h at 150 °C. After cool-
ing to room temperature, the mixture was dissolved in
CH2Cl2 (100 mL) and water (100 mL). Two phases were
separated and the water layer was extracted with CH2Cl2
(3 × 50 mL). All organic layers were combined and the sol-
vents were evaporated to give a dry brown solid. The solid
© 2006 NRC Canada

Huang et al.
479
was purified by column chromatography using THF–hex-
anes (2:5) first and then pure THF as the eluent to obtain a
crude yellow product. The crude product was decolored with
activated carbon in hot DMF and recrystallized in hot DMF
three times. A pale yellow solid (0.464 g, yield 48%, mp
8.1 Hz, J2 = 8.0 Hz, 2H, indolyl), 7.24 (dd, J1 = 8.0 Hz, J2 =
7.6 Hz, 2H, indolyl), 7.18 (s, 2H, indolyl), 2.41 (s, 6H, Me).
Synthesis of intermediate 1,3,5-tris[(p-boronic
acid)phenyl]benzene (b)
1
333 °C) was obtained. H NMR (600 MHz, CDCl3, 298 K,
To a vigorously stirred solution of 1,3,5-tris(p-bromo-
phenyl)benzene (2.72 g, 4.98 mmol) in dry THF (100 mL),
butyllithium solution in hexanes (1.6 mol/L, 15.6 mL,
24.9 mmol) was added slowly at –78 °C. After stirring for
1 h at –78 °C, triisopropyl borate (6.3 mL, 27.4 mmol) was
added slowly. The solution was stirred for 1 h at –78 °C then
warmed to room temperature and stirred for 12 h. The reac-
tion was quenched by adding satd. aq. NH4Cl (30 mL). The
water phases were separated and extracted with CH2Cl2 (3 ×
20 mL). All of the organic layers were combined and con-
centrated to give a white solid. Recrystallization from meth-
anol and hexanes yielded a white compound (1.31 g, yield
64%, mp > 300 °C). 1H NMR (400 MHz, acetone-d6, 298 K,
ppm) δ: 8.05 (d, J = 8.2 Hz, 6H, ph), 8.00 (s, 3H, ph), 7.88
(d, J = 8.2 Hz, 6H, ph), 7.27 (s, 6H, OH).
ppm) δ: 8.97 (d, J = 6.8 Hz, 6H, ph), 8.44 (d, J = 4.6 Hz,
3H, 7-aza), 8.10 (d, J = 6.8 Hz, 6H, ph), 8.01 (d, J = 7.8 Hz,
3H, 7-aza), 7.67 (d, J = 3.6 Hz, 3H, 7-aza), 7.19 (dd, J1 =
7.8 Hz, J2 = 4.6 Hz, 3H, 7-aza), 6.71 (d, J = 3.6 Hz, 3H, 7-
aza). 13C NMR (600 MHz, CDCl3, 298 K, ppm) δ: 170.8,
147.5, 143.6, 142.0, 133.5, 130.1, 129.1, 122.9, 127.1,
122.0, 117.0, 102.5. Elemental anal. calcd. for C42H27N9: C
76.71, H 4.10, N 19.18; found: C 76.64, H 4.04, N 19.21.
Synthesis of 1,3,5-tris{p-[N-(3-methyl-indolyl)]phenyl}-
benzene (4)
A mixture of 1,3,5-tris(p-bromophenyl)benzene (1.00 g,
1.84 mmol), 3-methylindole (1.09 g, 8.29 mmol), CuI
(0.11 g, 0.55 mmol), Cs2CO3 (3.78 g, 11.6 mmol), 1,10-
phenanthroline (0.20 g, 1.10 mmol), and DMF (5 mL) was
heated to 150 °C. After stirring for 48 h at 150 °C, the mix-
ture was cooled to room temperature and dissolved in
CH2Cl2 (20 mL) and water (20 mL). Two phases were sepa-
rated and the water layer was extracted with CH2Cl2 (3 ×
20 mL). All organic layers were combined, dried over
MgSO4, and concentrated to give a brown solid. The crude
product was purified twice by a chromatographic column us-
ing chloroform–hexanes (at a 1:1.7 concentration initially
and secondly, a 2:1 concentration) as the eluent to obtain a
white solid. Recrystallization from CH2Cl2 and hexanes
yielded the colorless compound 4 (0.65 g, yield 51%, mp
Synthesis of intermediate 1-bromo-3,5-bis(N-
indolyl)benzene (c)
A mixture of 1,3,5-tribromobenzene (3.780 g, 12 mmol),
indole (2.808 g, 24 mmol), CuSO4·5H2O (0.150 g, 0.6 mmol),
and K2CO3 (3.312 g, 24 mmol) was heated to 195 °C and
stirred for 16 h at 195 °C. After cooling to room tempera-
ture, the mixture was dissolved in CH2Cl2 (100 mL) and wa-
ter (100 mL). Two phases were separated and the water layer
was extracted with CH2Cl2 (3 × 30 mL). All of the organic
layers were combined, dried over MgSO4, and concentrated
to give a brown solid. The crude product was purified by a
chromatographic column using ethyl acetate – hexanes
(1:20) as the eluent to obtain a colorless solid (0.580 g, yield
13%, mp 117 to 118 °C). 1H NMR (400 MHz, CDCl3,
298 K, ppm) δ: 7.65 (d, J = 7.8 Hz, 2H, indolyl), 7.64 (d, J =
8.1 Hz, 2H, indolyl), 7.61 (s, 1H, ph), 7.54 (s, 2H, ph), 7.29
(d, J = 3.3 Hz, 2H, indolyl), 7.21 (dd, J1 = 8.1 Hz, J2 =
7.2 Hz, 2H, indolyl), 7.14 (dd, J1 = 8.0 Hz, J2 = 7.8 Hz, 2H,
indolyl), 6.66 (d, J = 3.3 Hz, 2H, indolyl).
1
275 °C). H NMR (400 MHz, CDCl3, 298 K, ppm) δ: 7.93
(s, 3H, ph), 7.90 (d, J = 8.2 Hz, 6H, phenyl), 7.69 (d, J =
7.2 Hz, 3H, indolyl), 7.68 (d, J = 7.6 Hz, 3H, indolyl), 7.66
(d, J = 7.2 Hz, 6H, phenyl), 7.31 (dd, J1 = 7.8 Hz, J2 =
7.2 Hz, 3H, indolyl), 7.25 (s, 3H, indolyl), 7.24 (dd, J1 =
7.8 Hz, J2 = 7.6 Hz, 3H, indolyl), 2.46 (s, 9H, Me). 13C
NMR (400 MHz, CDCl3, 298 K, ppm) δ: 141.79 (ph), 139.6
(ph), 138.4 (ph), 135.9, 129.9, 128.4, 125.3, 124.9, 124.2,
122.5, 119.9, 119.3, 113.2, 110.4, 9.6. HRMS (EI, m/z)
calcd. for C51H39N3: 693.3144; found: 693.3135.
Synthesis of intermediate 3,5-bis(1-N-indolyl)-1-
phenylboronic acid (d)
A solution of tert-butyllithium (1.6 mol/L in hexanes,
4.85 mL, 7.76 mmol) was added slowly to a THF (50 mL)
solution of 1-bromo-3,5-bis(1-N-indolyl)benzene (2.00 g,
5.17 mmol) at –78 °C. After stirring for 1 h at –78 °C,
triisopropyl borate (2.02 mL, 8.79 mmol) was added slowly.
The solution was stirred for 1 h at –78 °C then slowly
warmed to room temperature and stirred for 12 h. The reac-
tion was quenched by adding satd. aq. NH4Cl solution
(20 mL). The water phases were separated and extracted
with CH2Cl2 (3 × 30 mL). All organic layers were combined
and then concentrated to give a brown solid. The crude prod-
uct was purified by a chromatographic column using THF–
hexanes (2:5) as the eluent. Recrystallization from CH2Cl2
and hexanes yielded a colorless solid (1.15 g, yield 63%). 1H
NMR (400 MHz, actone-d6, 298 K, ppm) δ: 8.10 (s, 2H, ph),
7.86 (s, 1H, ph), 7.75 (d, J = 7.7 Hz, 2H, indolyl), 7.73 (d,
J = 3.3 Hz, 2H, indolyl), 7.70 (d, J = 7.8 Hz, 2H, indolyl),
7.54 (s, 2H, ph), 7.66 (s, 2H, OH), 7.26 (dd, J1 = 8.0 Hz,
Synthesis of intermediate 1-bromo-3,5-bis[N-(3-methyl-
indolyl)]benzene (a)
A
mixture of 1,3,5-tribromobenzene (2.160 g,
6.86 mmol), 3-methyl indole (1.807 g, 13.7 mmol),
CuSO4·5H2O (0.086 g, 0.05 mmol), K2CO3 (1.89 g,
13.7 mmol), and dodecane (4 mL) was heated to 190 °C and
stirred for 10 h at 195 °C. After cooling to room tempera-
ture, the mixture was dissolved in CH2Cl2 (50 mL) and wa-
ter (50 mL). Two phases were separated and the water layer
was extracted with CH2Cl2 (3 × 20 mL). All of the organic
layers were combined, dried over MgSO4, and concentrated
to give a brown solid. The crude product was purified by a
chromatographic column using CH2Cl2–hexanes (1:4) as the
eluent to obtain a colorless solid (0.628 g, yield 22%, mp
184 to 185 °C). 1H NMR (400 MHz, CDCl3, 298 K, ppm) δ:
7.63 (d, J = 7.8 Hz, 2H, indolyl), 7.64 (d, J = 8.1 Hz, 2H,
indolyl), 7.61 (s, 1H, ph), 7.59 (s, 2H, ph), 7.31 (dd, J1 =
© 2006 NRC Canada

480
Can. J. Chem. Vol. 84, 2006
J2 = 7.8 Hz, 2H, indolyl), 7.17 (dd, J1 = 8.0 Hz, J2 = 7.7 Hz,
2H, indolyl), 6.76 (d, J = 3.3 Hz, 2H, indolyl).
acid (0.930 g, 2.64 mmol), Pd(PPh3)4 (0.04 g, 0.03 mmol),
Na2CO3 solution (2 N, 30 mL), toluene (50 mL), and ethanol
(20 mL) was refluxed for 40 h. A white precipitate formed.
After cooling to room temperature, CH2Cl2 (100 mL) and
water (100 mL) were added to the mixture. The water layer
was extracted with CH2Cl2 (3 × 30 mL). All of the organic
layers were combined and concentrated to give a white
solid. The crude product was dried and purified by a chro-
matographic column using THF–hexanes (2:5) as the eluent
to obtain a yellow solid. Recrystallization from THF and
hexanes initially, then from CH2Cl2 and hexanes three times
yielded the colorless compound 7 (0.561 g, yield 61%, mp >
Synthesis of 1,3,5-tris[3,5-bis(N-indolyl)-4′-
biphenyl]benzene (5)
A mixture of 1,3,5-tris[(p-boronic acid)phenyl]benzene
(0.192 g, 0.439 mmol), 1-bromo-3,5-bis(N-indolyl)benzene
(0.595 g, 1.537 mmol), Pd(PPh3)4 (0.020 g, 0.018 mmol),
Na2CO3 solution (2 N, 30 mL), toluene (50 mL), and ethanol
(20 mL) was refluxed for 40 h. A white precipitate formed.
After cooling to room temperature, the mixture was ex-
tracted with CH2Cl2 (3 × 50 mL). All of the organic layers
were combined and concentrated to give a colorless solid.
The crude product was dried and purified by a chromato-
graphic column using THF–hexanes (2:5) as the eluent to
obtain a pale white solid. Recrystallization from CH2Cl2 and
hexanes twice yielded compound 5 (0.47 g, yield 88%, mp >
1
330 °C). H NMR (600 MHz, CDCl3, 298 K, ppm) δ: 8.94
(d, J = 8.5 Hz, 6H, ph), 7.92 (d, J = 8.5 Hz, 6H, ph), 7.85 (s,
6H, ph), 7.74 (d, J = 7.8 Hz, 6H, indolyl), 7.72 (d, J =
8.3 Hz, 6H, indolyl), 7.71 (s, 3H, ph), 7.30 (dd, J1 = 8.3 Hz,
J2 = 8.0 Hz, 6H, indolyl), 7.48 (d, J = 3.2 Hz, 6H, indolyl),
7.22 (dd, J1 = 8.0 Hz, J2 = 7.8 Hz, 6H, indolyl), 6.78 (d, J =
3.2 Hz, 6H, indolyl). 13C NMR (600 MHz, CDCl3, 298 K,
ppm) δ: 171.1, 143.4, 143.3, 141.7, 136.0, 135.5, 129.9,
129.6, 127.5, 127.3, 122.7, 121.2, 120.7, 120.5, 118.9,
110.2, 104.4. Elemental anal. calcd. for C87H57N9: C 85.09,
H 4.65, N 10.27; found: C 85.16, H 4.76, N 9.94.
1
330 °C). H NMR (400 MHz, CDCl3, 298 K, ppm) δ: 7.96
(s, 3H, ph), 7.91 (d, J = 8.2 Hz, 6H, ph), 7.85 (d, J = 8.2 Hz,
6H, ph), 7.84 (s, 6H, ph), 7.76 (d, J = 7.5 Hz, 6H, indolyl),
7.75 (d, J = 8.1 Hz, 6H, indolyl), 7.70 (s, 3H, ph) 7.50 (d,
J = 3.2 Hz, 6H, indolyl), 7.32 (dd, J1 = 8.1 Hz, J2 = 7.2 Hz,
6H, indolyl), 7.22 (dd, J1 = 7.5 Hz, J2 = 7.2 Hz, 6H,
indolyl), 6.80 (d, J = 3.2 Hz, 6H, indolyl). 13C NMR
(400 MHz, CDCl3, 298 K, ppm) δ: 143.7, 141.8, 141.6,
141.0, 138.8, 135.7, 129.5, 128.1, 127.8, 127.7, 125.3,
122.8, 121.4, 120.8, 120.5, 118.6, 110.4, 104.4. Elemental
anal. calcd. for C90H60N4: C 88.24, H 4.90, N 6.86; found: C
87.87, H 4.96, N 6.78.
Results and discussion
Synthesis and characterization
The new starburst molecules were obtained by using ei-
ther the Ullmann condensation methods (5) or the Suzuki
coupling methods (6) as depicted in Schemes 1–3.
Synthesis of 1,3,5-tris{3,5-bis[N-(3-methyl-indolyl)]-4′-
biphenyl}benzene (6)
Compound 1 was synthesized by the reaction of 7-
azaindolyl lithium with cyanuric chloride in high yield
(82%). Compounds 2–4 were prepared in moderate yields
(48%–64%) from the reaction of 7-azaindole or 3-
methylindole with 1,3,5-tris(p-bromophenyl)benzene or
2,4,6-tris(p-bromophenyl)-1,3,5-triazine using the modified
Ullmann condensation procedure (5) where CuI and 1,10-
phenanthroline acted as the catalysts and Cs2CO3 as the HBr
scavenger. The intermediates, 1,3,5-tris(p-bromophenyl)ben-
zene and 2,4,6-tris(p-bromophenyl)-1,3,5-triazine for com-
pounds 2–4, were synthesized by using modified literature
methods (2). The syntheses of compounds 5 and 6 were
achieved in good or moderate yields by Pd-catalyzed
Suzuki–Miyaura cross couplings of 1,3,5-tris[(p-boronic
acid)phenyl]benzene (intermediate b) with the correspond-
ing disubstituted bromobenzene (intermediates a or c), as
shown in Scheme 2. The intermediates a and c were synthe-
sized by Ullmann condensation reactions between 2 equiv.
of indole or 3-methylindole and 1,3,5-tribromobenzene (2).
The intermediates 1,3,5-tris[(p-boronic acid)phenyl]benzene
and 3,5-bis(1-N-indolyl)-1-phenylboronic acid were obtained
by the reaction of B(O-i-Pr)3 with 1,3,5-tris(4-lithium-
phenyl]benzene and 3,5-bis(1-N-indolyl)phenyllithium, re-
spectively. Compound 7 was obtained in good yield by
Suzuki coupling between 2,4,6-tris(p-bromophenyl)-1,3,5-
triazine and the intermediate d (Scheme 3). The switching
from the boric acid triphenyltriazine core c as the one used
in the syntheses of 5 and 6 to the brominated triphenyl-
triazine core and the change of the bromo-substituted inter-
mediate a to the boric acid intermediate d were necessary to
ensure the successful synthesis of compound 7 (when the
A mixture of 1,3,5-tris[(p-boronic acid)phenyl]benzene
(0.150 g, 0.343 mmol), 1-bromo-3,5-bis[N-(3-methyl-
indolyl)]benzene (0.498 g, 1.20 mmol), Pd(PPh3)4 (0.016 g,
0.014 mmol), Na2CO3 solution (2 N, 30 mL), toluene
(50 mL), and ethanol (20 mL) was refluxed for 40 h. A
white precipitate formed. After cooling to room temperature,
CH2Cl2 (50 mL) and water (50 mL) were added to the mix-
ture. The water layer was separated and extracted with CH2Cl2
(2 × 50 mL). The combined organic layers were dried over
MgSO4, and the solvents were evaporated under reduced
pressure. Purification of the crude product by column (THF–
hexanes, 1:16) afforded a white solid. Recrystallization from
CH2Cl2 and hexanes yielded compound 6 (0.254 g, yield
57%, mp 318 °C). 1H NMR (600 MHz, CDCl3, 298 K, ppm)
δ: 7.93 (s, 3H, ph), 7.87 (d, J = 8.3 Hz, 6H, ph), 7.82 (d, J =
8.3 Hz, 6H, ph), 7.74 (s, 6H, ph), 7.71 (d, J = 8.2 Hz, 6H,
indolyl), 7.66 (d, J = 7.7 Hz, 6H, indolyl), 7.61 (s, 3H, ph),
7.28 (dd, J1 = 7.7 Hz, J2 = 7.2 Hz, 6H, indolyl), 7.26 (s, 6H,
indolyl), 7.22 (dd, J1 = 8.0 Hz, J2 = 7.2 Hz, 6H, indolyl),
2.42 (s, 18H, Me). 13C NMR (600 MHz, CDCl3, 298 K,
ppm) δ: 143.5, 141.8, 141.7, 140.8, 139.1, 135.8, 130.0,
128.0, 127.7, 127.5, 125.2, 122.8, 120.2, 119.6, 119.4, 117.8,
113.6, 110.4, 9.6. Elemental anal. calcd. for C96H72N6: C
88.07, H 5.50, N 6.42; found: C 88.26, H 5.21, N 6.18.
Synthesis of 2,4,6-tris[3,5-bis(1-N-indolyl)-4′-biphenyl]-
1,3,5-triazine (7)
A
mixture of 2,4,6-tris(p-bromophenyl)-1,3,5-triazine
(0.412 g, 0.75 mmol), 3,5-bis(N-indolyl)-1-phenylboronic
© 2006 NRC Canada

Huang et al.
481
Scheme 1.
Cl
N
N
N
N
N
N
Cl
N
Cl
BuLi
N
N
N
THF,-78 oC
NLi
N
H
o
N
N
N
-78 C
N
1, 82%
R
R
Y
Br
N
N
H
Y
CuI, Cs2CO3,
1,10-Phenanthroline
DMF, 150 oC, 48 h
X
X
X
X
X
X
Y
Br
Br
N
N
R
Y
R
X = CH, Y = N, R = H, 2, 64%
X = Y = N, R = H, 3, 48%
X = Y = CH, R = Me, 4, 51%
bromide and the boronic acid starting materials were re-
versed as for 5 and 6, no compound 7 was isolated).
ucts, we also attempted the syntheses of 3-methylindolyl de-
rivatives for the larger cores. Two 3-methylindolyl-
substituted compounds (4 and 6) were obtained successfully.
Compound 6 was found to be more soluble than the indolyl
analogue 5 in organic solvents. Compounds 1–7 were char-
Based on the functional group, compounds 1–7 can be di-
vided into three categories: 7-azaindolyl-substituted com-
pounds (1–3), indolyl-substituted compounds (5 and 7), and
3-methyl-substituted compounds (4 and 6). Because of their
potential use in coordination and supramolecular assembly,
our initial synthetic effort focused on 7-azaindolyl deriva-
tives with the triazine core (the benzene analogue of 1 was
reported earlier by our group), the triphenyl triazine–
triphenylbenzene core, and the tris(biphenyl)triazine–
tris(biphenyl)benzene core. Although the smaller molecules
(1–3) with 7-azaindolyl groups were obtained successfully,
we were not able to isolate 7-azaindolyl products that con-
tained the larger triphenyltriazine–triphenylbenzene or the
tris(biphenyl)triazine–tris(biphenyl)benzene core. The prob-
lem appeared to be mostly caused by the poor solubility of
the products in organic solvents (e.g., THF, CH3CN, CH2Cl2,
and DMF), which made the purification and (or) isolation
and characterization difficult. The indolyl group is an ana-
logue of 7-azaindolyl, but its derivatives are in general more
soluble in organic solvents than the corresponding 7-
azaindolyl derivatives. For the larger cores, we therefore
attempted the syntheses of indolyl derivatives. The
hexaindolyl-substituted compounds 5 and 7 were obtained
successfully. To further enhance the solubility of the prod-
1
acterized by H, 13C NMR, HRMS, or elemental analyses
(for compounds whose HRMS data could not be obtained
owing to either being too high in molecular weight or too in-
soluble in the solvent matrix).
Thermal properties
The glass-forming properties and phase transition of com-
pounds 1–7 were studied by DSC experiments. All seven
compounds were heated under argon with a heating rate of
10 °C min–1. After heating, all compounds were cooled to
50 °C at a cooling rate of 20 °C min–1. The second heating
was carried out with a heating rate of 10 °C min–1. Well-
defined melting points were observed for all compounds ex-
cept compounds 5 and 7 in the DSC diagrams. These melt-
ing point values are consistent with those obtained by using
a melting point apparatus. The high melting points (275 to
>330 °C, Table 1) displayed by compounds 1–7 support that
they have a high thermal stability. Compounds 2 and 3 are 7-
azaindolyl derivative molecules with similar structures, but a
different central core (benzene for 2 and triazine for 3). In-
terestingly, however, the melting point of compound 3 is
© 2006 NRC Canada

482
Can. J. Chem. Vol. 84, 2006
Scheme 2.
Br
Br
R
K2CO3, CuSO4
195 °C
N
N
R
R
+
N
H
Br
Br
R = Me, (a), 22%
R = H, (c), 13%
Br
B(OH)2
B(OiPr)3
-78 °C
BuLi, THF
-78 °C
H2O
Br
Br
(HO)2B
B(OH)2
b, 64%
R
R
N
N
Pd(PPh3)4, Na2CO3
Toluene, EtOH, reflux 40 h
a (or c)
+
b
R
R
N
N
N
N
R
R
R = H, 5, 88%
R = Me, 6, 57%
about 44 °C higher than that of 2. The same trend was also
observed previously in 2,2′-dipyridylamino-substituted
triphenyltriazine–triphenylbenzene or tris(biphenyl)triazine–
tris(biphenyl)benzene compounds, where the triazine-based
compounds consistently have a much higher melting point
than the corresponding benzene-based analogue (2, 7). This
perhaps can be attributed to the fact that the electronegative
nitrogen atoms in the triazine core makes the molecule more
polar, hence the presence of stronger intermolecular interac-
tions in the crystal lattice, compared with the benzene core
analogue.
187 °C) in the second heating–cooling curve of the DSC ex-
periments (Table 1). No glass transition temperatures were
observed for compounds 1–4 in either the first or the second
heating cycles, an indication that these smaller molecules
have the tendency to remain crystalline. The DSC diagram
(second and third heating and cooling cycles) of compound
6 is shown in Fig. 1 as a representative example. In addition
to Tg and Tm, a well-defined crystallization temperature (Tc)
was also observed for 6. Compounds 5 and 7 are another
pair of benzene core – triazine core analogues that contain
six indolyl groups. Although no melting points were ob-
served for either compound in the DSC experiments, the
DSC data again support that triazine compound 7 has a
Glass transition temperatures were observed for com-
pounds 5 (Tg ~ 168 °C), 6 (Tg ~ 188 °C), and 7 (Tg
~
© 2006 NRC Canada

Huang et al.
483
Scheme 3.
Br
B(OH)2
N
B(OiPr)3
-78 °C
BuLi, THF
-78 °C
H2O
N
N
N
c
d, 63%
N
N
Pd(PPh3)4, Na2CO3
toluene, EtOH
Br
N
N
N
N
N
N
N
Br
Br
N
N
N
7, 61%
Fig. 1. DSC diagrams of compound 6 (second and third) heating
(10 °C min–1) and cooling (20 °C min–1) (the first heating and
cooling diagram is identical to these).
Table 1. Thermal properties of compounds 1–7.
Compound Tg (°C)
Tc (°C) Tm (°C)a
289
287
333
275
>330
1
2
3
4
5
6
7
28
Second heating/cooling
Third heating/cooling
Tm
24
20
16
12
160 (heating) / 168 (cooling)
183 (heating) / 188 (cooling) 265
175 (heating) / 186 (cooling)
318
>330
Tg
aObtained by a melting point apparatus.
much higher Tg than its benzene analogue 5. The inclusion
of the methyl groups in 6 also appears to enhance the glass
transition temperature of the molecule as shown by the
higher Tg of 6, compared with that of 5. The exceptionally
high Tg values of compounds 5–7 indicate that these large
molecules are capable of forming stable amorphous phases,
a property highly desired for applications in organic
photonic devices such as organic light emitting diodes. The
remarkable high thermal stability of compounds 5–7 can be
attributed to their high molecular weights and their highly
branched structures, in comparison with compounds 1–4.
Tc
50
100
150
200
250
T (oC)
300
350
400
For all seven compounds, no reliable reduction peaks were
detected within the solvent limit (THF was used as the sol-
vent for reduction potential measurements). For compounds
2 and 4, well-defined oxidation peaks were observed in
CH2Cl2. For the remaining compounds, no reliable oxidation
peaks were detected in CH2Cl2. The failure to observe oxi-
dation potentials for some of the compounds can be partially
attributed to the poor solubility of the materials, especially
the large molecules (5–7). The first oxidation potential of 2
and 4 were found to be 1.57 and 1.14 V (vs. Ag/AgCl), re-
Electronic and luminescent properties
Compounds 1–7 have intense absorption bands in the
UV–vis region that can be attributed to π–π* transitions. At-
tempts were made to measure the oxidation and reduction
potentials of compounds 1–7 by using cyclic voltammetry.
© 2006 NRC Canada

484
Can. J. Chem. Vol. 84, 2006
Fig. 2. Cyclic voltammetry diagrams of 2 (top) and 4 (bottom)
Fig. 3. The excitation and emission spectra of 6 in CH2Cl2.
in CH2Cl2.
1.2
Em: 427 nm
Ex: 312 nm
1.0
0.8
0.6
0.4
0.2
0
200
300
400
500
600
Wavelength (nm)
data are summarized in Table 2. Compared to the
corresponding free 7-azaindole, indole, or 3-methylindole
molecules, the absorption and emission spectra of com-
pounds 1–7 are all red-shifted, which is clearly caused by
the conjugation of the 7-azaindolyl, indolyl, or 3-
methylindolyl chromophore with the aromatic central core.
One obvious trend is the difference between the triazine and
benzene core analogues (2 and 3 and 5 and 7). As shown by
the data in Table 2, the triazine derivative consistently has
smaller absorption–excitation and emission energies (i.e.,
the red shift of the absorption and emission band), compared
with the corresponding benzene derivative. The same trend
was also observed in 2,2′-dipyridylamino-functionalized
triazine and benzene derivatives (2, 7). The methyl-
substituted compound 6 has an emission energy that is about
25 nm red-shifted, compared with the nonsubstituted indolyl
analogue 5, which can be attributed to the electron-donating
nature of the methyl group that effectively destablizes the
HOMO level (π) and stablizes the LUMO level (π*), hence
decreasing the band gap. Similar emission energy red shift
induced by electron-donating groups has been observed in
BPh2(2-Py-indolyl-X) compounds, where X is a substituent
on the indolyl ring (11). The quantum efficiencies of com-
pounds 1–7 vary considerably. Molecules (2–4) that have ei-
ther a triphenyltriazine or triphenybenzene core display the
highest emission quantum efficiencies. The quantum effi-
ciencies of the large molecules (5–7) are substantially lower.
For applications as emitters in organic light emitting diodes,
it is required for the molecule to have both a high Tg and a
high emission quantum efficiency. The high Tg is to ensure
the morphological stability of the organic layer. Taking both
factors into consideration, compounds 5 and 6 are the most
promising as emitters in OLEDs, which are being examined
by our group, and the results will be disclosed in due course.
Although no Tg was detected, compound 2 has a fairly large
band gap and a deep HOMO level, which makes it poten-
tially useful as either a host material for phosphorescent
dopants or as a hole-blocking layer.
spectively (Fig. 2). The HOMO energy levels for 2 and 4
were calculated to be –5.9 and –5.4 eV, respectively, by us-
ing the oxidation potential and the ferrocene HOMO energy
level of –4.8 eV (relative to the vacuum level) as the stan-
dard (8). Clearly, the 7-azaindolyl derivative 2 has a much
deeper HOMO level than that of the 3-methyl indolyl deriv-
ative 4, which can be attributed to the extra electronegative
nitrogen atom on the 7-azaindolyl ring that stabilizes the
HOMO level, relative to that of the indolyl compound. By
using the optical band gap obtained from the UV–vis ab-
sorption spectra (absorption edge), the LUMO energy levels
of 2 and 4 were calculated to be –2.2 and –2.1 eV, respec-
tively. The HOMO energy levels of 2 and 4 are considerably
deeper than that of the well-known hole-transport material
(9) NPB and the LUMO levels are much higher than those
of the well-known electron-transport materials (Alq3 and
phenanthroline derivatives) (10). As a result, these com-
pounds are not good candidates to act as hole- or electron-
transport layers because of their relatively deep HOMO and
high LUMO. However, they could be used as blue emitters
or hosting materials in OLEDs.
Compounds 1–7 emit a blue or violet color in solution and
the solid state when irradiated by UV light. The luminescent
spectra of 1–7 are typical of fluorescence as shown by the
spectra of 6 in Fig. 3. The absorption data and luminescent
© 2006 NRC Canada

Huang et al.
485
Table 2. Absorption and luminescent properties of compounds 1–7.
Solid λ (nm)
Solutiona λ (nm)
UV–vis absorption (solution)a λ (nm),
Quantum yieldb
(%) (solution)a
Compound
ε ((mol/L)–1 cm–1)
Ex.
Em.
Ex.
Em.
2
35
47
60
17
14
6
1
2
3
4
5
6
7
270 (78 400), 294 (shoulder, 41 900)
230 (46 800), 284 (97 600)
314
343
400
357
350
343
393
361, 396
427
450
379
403
318
307
344
332
318
312
331
423
365
408
400
403
427
481
234 (10 700), 280 (82 400), 344 (79 600)
232 (79 400), 278 (67 900), 320 (94 300)
232 (89 400), 270 (167 000), 302 (208 000)
232 (129 000), 272 (175 000), 306 (221 000)
232 (89 100), 266 (127 000), 306 (169 000)
401
454
aAll data were collected in CH2Cl2 solution at 298 K, except the data for S3 were collected in THF solution at 298 K. Concentration: [M] ≈ 10–5 mol/L.
bThe quantum yields were obtained by using anthracene or 9,10-diphenylanthracene as a reference.
(e) W.L. Jia, R.Y. Wang, D.T. Song, S. Ball, A. McLean, and
S. Wang. Chem. Eur. J. 11, 832 (2005).
Acknowledgement
4. (a) D.T. Song and S. Wang. Comments Inorg. Chem. 25, 1
(2004) and refs. therein; (b) D.T. Song and S. Wang. Eur. J.
Inorg. Chem. 3773 (2003).
We thank the Natural Sciences and Engineering Research
Council of Canada (NSERC) for financial support.
5. A. Klapars, J.C. Antilla, X. Huang, and S.L. Buchwald, J. Am.
Chem. Soc. 123, 7727 (2001).
References
6. (a). N. Muyaura. Adv. Met.–Org. Chem. 6, 187 (1998), and
refs. therein; (b) A. Suzuki. J. Organomet. Chem. 576, 147
(1999), and refs. therein.
7. Q.D. Liu, W.L. Jia, G. Wu, and S. Wang. Organometallics, 22,
3781 (2003).
8. J. Pommerehne, H. Vestweber, W. Guss, R.F. Mahrt, H.
Bassler, M. Porsch, and J. Daub. Adv. Mater. 7, 551 (1995).
9. (a) T. Tsutsui. Mater. Res. Soc. Bull. 22, 39 (1997); (b) W.L.
Jia, X.D. Feng, D.R. Bai, Z.H. Lu, and S. Wang. Chem. Mater.
17, 164 (2005).
10. (a) C.W. Tang and S.A. VanSlyke. Appl. Phys. Lett. 51, 913
(1987); (b) L.S. Hung and C.H. Chen. Mater. Sci. Eng. R. 39,
143 (2002); (c) S. Naka, H. Okada, H. Onnagawa, and T.
Tsutsui. Appl. Phys. Lett. 76, 197 (2000).
1. (a) Y. Shirota. J. Mater. Chem. 10, 25 (2000); (b) E. Ueta, H.
Nakano, and Y. Shirota. Chem. Lett. 2396 (1994); (c) I.Y. Wu,
J.T. Lin, Y.T. Tao, and E. Balasubramaniam. Adv. Mater. 12,
668 (2000); (d) S. Berleb, W. Brutting, M. Schwoerer, R.
Wehrmann, and A. Elschner. Adv. Mater. 6, 677 (1994).
2. (a) J. Pang, Y. Tao, S. Freiberg, X.P. Yang, M. D’Iorio, and S.
Wang. J. Mater. Chem. 12, 206 (2002); (b) Q. Wu, J.A.
Lavigne, Y. Tao, M. D’Iorio, and S. Wang. Chem. Mater. 13,
71 (2001).
3. (a) Q.G. Wu, A. Hook, and S. Wang. Angew. Chem. Int. Ed.
39, 3933 (2000); (b) J. Pang, E.J.-P. Marcotte, C. Seward, R.S.
Brown, and S. Wang. Angew. Chem. Int. Ed. 40, 4042 (2001);
(c) J. Pang, C. Seward, and S. Wang. Eur. J. Inorg. Chem. 6,
1390 (2002); (d) C. Seward, W.L. Jia, R.Y. Wang, G.D.
Enright, and S. Wang. Angew. Chem. Int. Ed. 43, 2933 (2004);
11. Q.D. Liu, M.S. Mudadu, R. Thummel, Y. Tao, and S. Wang.
Adv. Funct. Mater. 15, 143 (2005).
© 2006 NRC Canada