Structural-property relationship in pyrazino[2,3-g]quinoxaline derivatives: Morphology, photophysical, and waveguide properties

Article abstract of DOI:10.1021/cm100798q

The objective of this contribution is to achieve crystalline one-dimensional organic micro/nanostructures by self-assembling for active waveguides and other optoelectronic nanodevices. To improve solid emitting performance for highly efficient waveguide f

Full text of DOI:10.1021/cm100798q

Chem. Mater. 2010, 22, 3735–3745 3735
DOI:10.1021/cm100798q
Structural-Property Relationship in Pyrazino[2,3-g]quinoxaline
Derivatives: Morphology, Photophysical, and Waveguide Properties
Xiao Wang, Yan Zhou, Ting Lei, Nan Hu, Er-Qiang Chen,* and Jian Pei*
Beijing National Laboratory for Molecular Sciences, Key Laboratories of Bioorganic Chemistry and
Molecular Engineering and of Polymer Chemistry and Physics of Ministry of Education, College of
Chemistry and Molecular Engineering, Peking University, Beijing 100871, China
Received March 18, 2010. Revised Manuscript Received May 7, 2010
The objective of this contribution is to achieve crystalline one-dimensional organic micro/
nanostructures by self-assembling for active waveguides and other optoelectronic nanodevices. To
improve solid emitting performance for highly efficient waveguide fibers, our strategy here is to adopt
nonplanar conjugated segments to minimize unfavorably intermolecular fluorescent quenching in
solid state; meanwhile, long alkyl chains are added as structural group that controls molecular
packing. With this intention, a series of molecules based on pyrazino[2,3-g]quinoxaline (PyQ) units
were developed. The effects of the molecular geometry and symmetry on their optical properties,
crystal dimensions, liquid crystals, and active waveguide properties were studied in detail to under-
stand the structure-property relationship of organic conjugated molecules, which is of vital impor-
tance to rational design for organic self-assemblies toward optoelectronic applications. On a single
molecular level, the molecular structures and their optical properties were correlated by computa-
tional methods. The combined results of morphology study, single crystal structure and two-
dimensional wide-angle X-ray diffraction (2D-WAXD) indicated that the loss of the different
molecular symmetry element affected the crystal dimensions. The microfibers of molecules 1 and 2
with their desirable crystal dimensions and surfaces were proved to be low-loss waveguide materials
(0.02-0.05 dB/μm). Hence, we suggest that nonplanar aromatic core modified with proper side
group may represent a promising approach to achieve crystalline optical waveguides. These
structure-property investigations provide a deeper understanding for the structure and intermole-
cular interactions of such molecular aggregates.
Introduction
optoelectronics, such as organic field-effect transistor,8
chemical sensors,9 optical waveguides and lasers.10
Organic materials as one-dimensional (1D) active wave-
guide with high efficiency are required to process good
fluorescent emission, proper crystal dimensions, and smooth
crystal surface.10 However, for conventional organic semi-
conductors with flat aromatic cores, there is a dilemma.
On the one hand, these molecules tend to form face-to-
face stacking due to considerable π-π interactions; how-
ever, such cofacial arrangement as H-aggregates does not
Because of their unique optical and electronic properties,
organic conjugated molecules have been widely applied in
dye industry,1 fundamental electron/energy transfer studies,2
biological fluorescent probes,3 dye sensitized solar cells,4
and information storage techniques (CD-R, DVD-R).1,5
Recently, with the rapid developments of organic electro-
nics6 and nano-optics,7 the micro/nano structure of such
molecule aggregates have attracted considerable atten-
tions because of their promising application in miniaturized
(8) (a) Zhou, Y.; Liu, W.-J.; Ma, Y.; Wang, H.; Qi, L.; Cao, Y.; Wang,
J.; Pei, J. J. Am. Chem. Soc. 2007, 129, 12386–12387. (b) Briseno,
A. L.; Mannsfeld, S. C. B.; Lu, X.; Xiong, Y.; Jenekhe, S. A.; Bao, Z.;
Xia, Y. Nano Lett. 2007, 7, 668–675.
(9) (a) Naddo, T; Che, Y.; Zhang, W.; Balakrishnan, K.; Yang, X.;
Yen, M.; Zhao, J.; Moore, J. S.; Zang, L. J. Am. Chem. Soc. 2007,
129, 6978–6979. (b) Chen, Y.; Yang, X.; Loser, S.; Zang, L. Nano Lett.
2008, 8, 2219–2223.
(10) (a) Takazawa, K.; Kitahama, Y.; Kimura, Y.; Kido, G. Nano Lett.
2005, 5, 1293–1296. (b) Takazawa, K. J. Phys. Chem. C 2007, 111,
8671–8676. (c) Lebedenko, A. N.; Guralchuk, G. Y.; Sorokin, A. V.;
Yefimova, S. L.; Malyukin, Y. V. J. Phys. Chem. B 2006, 110, 17772–
17775. (d) Zhao, Y. S.; Xu, J.; Peng, A.; Fu, H.; Ma, Y.; Jiang, L.; Yao, J.
Angew. Chem., Int. Ed. 2008, 47, 7301–7305. (f) Zhao, Y. S.; Peng,
A.; Fu, H.; Ma, Y.; Yao, J. Adv. Mater. 2008, 20, 1661–1665. (g) Liao,
Q.; Fu, H.; Yao, J. Adv. Mater. 2009, 21, 4153–4157. (h) Che, Y.; Yang,
X.; Balakrishnan, K.; Zuo, J.; Zang, L. Chem. Mater. 2009, 21, 2930–
2934.
*Corresponding author. E-mail: jianpei@pku.edu.cn.
(1) Zollinger, H. Color Chemistry: Syntheses, Properties and Applica-
tions of Organic Dyes and Pigments, rev. ed.; Wiley-VCH: Z€urich,
Switzerland, 2003; Chapter 14.
(2) (a) Miller, J. R.; Calcaterra, L. T.; Closs, G. L. J. Am. Chem. Soc.
1984, 106, 3047–3049. (b) Marcus, R. A. Pure Appl. Chem. 1997, 69,
13–29.
(3) Zhang, J.; Campbell, R. E.; Ting, A. Y.; Tsien, R. Y. Nat. Rev. Mol.
Cell Biol. 2002, 3, 906–918.
€
(4) (a) O’ Regan, B.; Gratzel, M. Nature 1991, 353, 737–740. (b) Lee, K.;
Park, S. W.; Ko, M. J.; Kim, K.; Park, N.-G. Nat. Mater. 2009, 8, 665–
671.
(5) Shang, Y.; Wen, Y.; Li, S.; Du, S.; He, X.; Cai, L.; Li, Y.; Yang, L.;
Gao, H.; Song, Y. J. Am. Chem. Soc. 2007, 129, 11674–11675.
(6) Organic Electronics: Materials, Processing, Devices and Applica-
tions; So, F., Ed.; CRC PressL Boca Raton, FL, 2010.
(7) Principles of Nano-Optics; Novotny, L.; Hecht, B., Eds.; Cambridge
University Press: Cambridge, U.K., 2006.
r
2010 American Chemical Society
Published on Web 05/21/2010
pubs.acs.org/cm

3736 Chem. Mater., Vol. 22, No. 12, 2010
Wang et al.
Chart 1. Molecular Structures of 1-5
favor high fluorescence quantum yield in solid state.11 On
the other hand, highly fluorescent J-aggregates12 are
seldom observed, mainly because of their relatively small
stokes shift, which leads to significant self-absorption lost
during light propagation.10 Though the H-aggregates of
dyes have been reported as active waveguide fibers,10a
how to weaken the unfavorably intermolecular π-π inter-
actions within stacks is worth exploring to further im-
provethe waveguide performance of molecular assemblies.
To enhance fluorescent emission in solid state, we con-
sider enlarging the intermolecular distance of adjacent
aromatic rings by introducing steric hindrance; hence a
nonplanar conjugated core was adopted to minimize the
unfavorable intermolecular fluorescent quenching.13
With this intention, a series of molecules based on tetra-
substituted pyrazino[2,3-g]quinoxaline (PyQ) units were
developed, and the torsion of the adjacently attached
aromatic rings embodied the desired steric hindrance.
With its electron affinity and ease in syntheses, the PyQ
unit is widely used in electroluminescent polymer mate-
rials,14a,b coupling of large aromatic cores (porphyrin14c
and hexabenzocoronene14d as representatives) and opti-
cal electronic sensors of inorganic anions.14e Previous
experiences also suggest that it is an effective way to engi-
neer molecular aggregates by changing side groups.6a,10h,15
Therefore, to promote the self-assembling process,
we also introduced long alkyl chains as the structural
group.
In this contribution, molecules 1-4 are designed with
the same electron-deficient pyrazino[2,3-g]quinoxaline
ring in the center and different electron-rich aromatic
rings as periphery adjusting site for color and hindrance.
Their structures are illustrated in Chart 1. PyQ is a highly
symmetric core (D2h), and both the molecules with main-
tained symmetry (1 and 2) and those with lowered sym-
metry introduced by periphery substitutes (3 and 4) were
explored to demonstrate the favorable direction for the
chemical modification of PyQ moiety. Though molecular
assemblies are a combined result of molecular structure
and growth conditions, understanding structure-property
relationships of organic conjugated molecule is a prere-
quisite for controllable self-assembling process and is
crucial to rational materials design. In addition, in solid
state, the interactions between organic molecules are
relatively weak compared with ionic bond and covalent
bond, and the packing of molecular crystals is therefore
generally complex yet molecular-shape-determined to
some extent.16 Hence, investigation on the geometric
aspects of organic molecules are also worthwhile for
functionalizing molecular assemblies. The effects of the
molecular geometry and symmetry on their optical prop-
erties, crystal dimensions, liquid crystals (LC), and active
waveguide properties are investigated in detail in the
Results and Discussion section. The molecular structures
and their photophyscial properties are correlated by
computational methods on a single molecular level. Mor-
phology and crystallography studies are then presented to
build structure-property relationship in solid state. Liquid
crystal behaviors are also discussed to provide deeper
understanding of intermolecular interactions of mole-
cules 1-4. Finally, the waveguide behaviors of fibers of
1 and 2 are characterized to demonstrate the potential use
of our materials as active waveguide fibers.
(11) Jenekhe, S. A.; Osaheni, J. A. Science 1994, 265, 765.
(12) (a) J-Aggregates; Kobayashi, T, Ed.; World Science: Singapore, 1996.
(b) Kaiser, T. E.; Stepanenko, V.; W€urthner, F. J. Am. Chem. Soc. 2009,
131, 6719–6732.
(13) de Halleux, V.; Calbert, J.-P.; Brocorens, P.; Cornil, J.; Declercq,
ꢀ
J.-P.; Bredas, J.-L.; Geerts, Y. Adv. Funct. Mater. 2004, 14, 649–
659.
(14) (a) Cui, Y.; Zhang, X.; Jenekhe, S. A. Macromolecules 1999, 32,
3824–2826. (b) Zhu, Y.; Yen, C. T.; Jenekhe, S. A.; Chen, W. C.
Macromol. Rapid Commun. 2004, 25, 1829–1834. (c) Crossley, M. J.;
Thordarson, P. Angew. Chem., Int. Ed. 2002, 41, 1709–1013.
(d) Fogel, Y.; Kastler, M.; Wang, Z.; Andrienko, D.; Bodwell, G. J.;
M€ullen, K. J. Am. Chem. Soc. 2007, 129, 11743–11749. (e) Anzenbacher,
Results and Discussion
Synthesis. Scheme 1 illustrates the synthetic approaches
to compounds 1-4. This set of molecules were synthesized
by the condensation of bezene-1,2,4,5-tetraamine and
ꢀ
P.; Palacios, M. A.; Jursíkova, K.; Marquez, M. Org. Lett. 2005, 7, 5027–
5030.
(15) (a) Zang, L.; Che, Y.; Moore, J. S. Acc. Chem. Res. 2008, 41, 1596–
1608. (b) Wang, J.; Yan, J.; Li, Z.; Han, J.; Ma, Y.; Bian, J.; Pei, J.
Chem.;Eur. J. 2008, 14, 7760–7764. (c) Dautel, O. J.; Guillaume, W.;
Almairac, R.; Flot, D.; Hirsch, L.; Lere-Porte, J.; Parneix, J.; Serein-
Spirau, F.; Vignau, L.; Moreau, J. J. E. J. Am. Chem. Soc. 2006, 128,
4892–4901.
(16) Evans, R. C.; An Introduction to Crystal Chemistry, 2nd ed.;
Cambridge University Press: Cambridge, U.K., 1964; Chapter 14,
pp 352-356.

Article
Chem. Mater., Vol. 22, No. 12, 2010 3737
Scheme 1. Synthesis Routes of Compounds 1-4
1,2-diketones. 2-Bromo-5-dodecylthiophene reacted with
metal magnesium toconvert toa Grignard reagent, follow-
ed by coupling with oxalyl chloride and lithium diaryl-
cuprate to afford 6 in 40% yield. The oxidation of diaryl-
ethyne, which was prepared through a Sonogashira reaction
between 1-ethynylbenzenes (7 or 8) and 1-iodobenzenes
(9 or 10), gave compounds 11 or 12 in moderate yields.
Finally, the condensation of bezene-1,2,4,5-tetraamine
and 1,2-diketones (6, 13, or 14) afforded 1, 2, and 4,
respectively. Differing from the literature reports on
o-quinone,17 the yield of cross-condensation product 3
was quite low (13%) no matter which diketone (6 or 13)
was added first. This may result from a lowered intramo-
lecular ring-closing rate caused by steric hindrance of
two adjacent aromatic rings. Another notable result is
that only anti product was obtained in the synthesis of 4,
which also indicated a long period of dynamic-chemical-
bonding process before intramolecular ring closure.
Thus, in this series, the molecules with symmetric center
are more favorable as organic materials from the stand-
point of the ease of synthesis. The new compounds were
readily soluble in common organic solvents, such as
toluene, THF, DMSO, and CH2Cl2. The structures and
purity of all new compounds were fully characterized and
Figure 1. Absorption (solid line) and PL (dashed line) spectra of molecule
1-4 in dichloromethane solution (1 ꢀ 10-5 M). All PL spectra were collec-
ted under excitation at the maximum absorption peak of the correspond
compounds.
about 464 nm, 449 nm, and 327 nm. However, in compa-
rison with molecule 1, for 2, all the three peaks red-shift
20-50 nm, indicating that the effective conjugation length is
increased after introducing the thiophene units in place of
the phenyl rings in 1. This may attributed to that the adja-
cent five-member rings have smaller steric hindrance than
the six member rings and thus the whole molecule of 2 is
better coplanar. It is also observed that the three wave-
lengths of the absorption maximum (λmax) for 3 are bet-
ween those of 1 and 2. Also, for 1, 2, and 3, their absorp-
tion and PL features are quite similar in terms of vibration
structures (absorption peak in 400-500 nm and emission
peak in 500-550 nm), full width at half-maximum (fwhm,
34-49 nm) and Stokes shifts (28-38 nm). Molecule 4
shows the absorption features very similar to 1. The inves-
tigation of the absorption features of molecules 1-4
indicate the conjugated nature of molecule 1-4, though
the steric hindrance between adjacent periphery rings has
undermined the coplanarity of the whole molecule. As shown
in Figure S1 in the Supporting Information, computational
1
verified by H and 13C NMR, elemental analysis, and
HRMS.
Photophysical Properties. The absorption and photo-
luminescence (PL) spectra of molecules 1-4 were mea-
sured in dilute CH2Cl2 solution as presented in Figure 1.
Table 1 summarizes all photophysical properties of mole-
cules 1-4 in dilute solution. Molecules 1, 2, and 3 exhi-
bited the similar absorption features with two absorption
bands. For molecule 1, the absorption maxima peak at
(17) Sendt, K.; Johnston, L. A.; Hough, W. A.; Crossley, M. J.; Hush,
N. S.; Reimers, J. R. J. Am. Chem. Soc. 2002, 124, 9299–9309.

3738 Chem. Mater., Vol. 22, No. 12, 2010
Wang et al.
Table 1. Summary of optical properties, crystal dimensions, and thermo Transitions of 1-4
optical properties
1 ꢀ 10-5 M in DCM
λ
abs (nm) (lgε)
λPL (fwhm) Stokes shift transition dipole
moment ( μe) (D)
microfiber
dimensions
thermo transition (°C) and
d-spacings (A)
a
˚
molecules
(dm3 mol-1 cm-1
)
(nm)
(nm)
1
464 (4.98), 449 (4.98),
327 (5.08)
502 (47)
39
13
length, 100-1000 μm;
width, 1-5 μm
heating, Cr (27.4)-103-Sm
(35.8)-172-I; cooling:
I -166-Sm (34.8)-73-Cr1
(34.7)-42-Cr2 (34.0)-18-Cr3
heating, Cr-97-I; cooling,
I-57-Cr
2
3
4
515 (4.26), 489 (4.26),
346 (4.36)
488 (4.76), 471 (4.78),
335 (4.90)
450 (4.46), 309 (4.61)
543 (49)
522 (34)
560 (93)
28
34
5
length, 100-1000 μm;
width, 0.5-3 μm
length, 5-10 μm; width,
1-2 μm
length, >100 μm; width, heating: Cr (36.7)-48-Sm
100 nm
heating, Cr-81-I; cooling,
I-76-Cr
110
25
(39.6)-78-N-114-I;
cooling, I-102-N-55-Sm
(40.0)-37-Cr (36.7)
a Cr stands for crystal; Sm stands for smectic phase, detailed discussion see the text; N stands for nematic phase; I stands for isotropic liquid.
Figure 2. Graphical representation of calculated molecular orbitals of molecule 1, 2, 4 and the core PyQ. The long alkyl chains were shortened to ethyl or
butyl for brevity. A represents that the corresponding orbital is antisymmetrical to symceter; S represents symmetrical to symceter; Dashed lines indicate
correlations between the MOs of 1, 2, and 4, those of periphery electron rich rings and PyQ according to orbital symmetry and the numbers of nodes. PhO
stands for ethoxylbenzene, Th for 2-ethylthiophene, Ph for benzene, PhO2 for 1,2-dibutoxylbenzene, respectively.
results, which were consistent with the experimental
absorption spectra, further confirmed the electronic com-
munication between periphery rings and the core by visu-
alizing highly delocalized HOMOs. Figure 2 shows simu-
lated molecular orbital (MO) diagram of molecules 1, 2,
and 4, in which 2 had more delocalized MOs than 1.
On the basis of their conjugated nature, the donor and
acceptor interactions of 1-4 could be qualitatively ana-
lyzed. In a grossly simplified consideration, MOs of mole-
cules 1-4 were regarded as a result of linear combination
of MOs of PyQ and those of periphery aromatic rings.18
As shown in Figure 2 and Table S1 in the Supporting
Information, the separately calculated results of the MOs
of the core and the periphery substitutes indicate that the
HOMOs of electron-rich substitutes are about 0.8 eV
higher than that of PyQ, whose LUMO is around 3 eV
lower than those of electron rich rings. Then, the previous
MOs of each segment are identified from new synthesized
molecules by symmetry and the numbers of nodes.19 For
example, the previous HOMO of PyQ is symmetrical to
symceter (a g orbital, labeled by S) with three nodes, and
the two occupied MOs in newly synthesized molecules
with the same g symmetry and three nodes in the core are
therefore correlated with the HOMO of PyQ. The quali-
tatively correlating diagram (as illustrated in Figure 2)
shows that the HOMOs of periphery rings have contrib-
uted most of HOMO, HOMO-1, and HOMO-2 because
of the energy level proximity and quantitative abundance,
whereas the core has made up most of the new LUMO
because of its much lower LUMO energy level. Consider-
ing that the HOMO is a potential electron donor, whereas
the LUMO is a potential electron acceptor, this set of
molecules 1-4 could also be classified as D-A systems.
In addition to the common photophysical features of
molecules 1-4, comparing the different donor segments
is helpful to future rational design of optical material. A
notable phenomenon is that the extinction coefficient of
1 (lgε=4.98) is five times large as that of 2 (lgε=4.26) (see
Table 1). This difference could partly be attributable to
(18) Combs, L. L. Int. J. Quantum Chem. 1977, 11, 1001–1004.
(19) (a) Lin, H.-C.; Jin, B.-Y. J. Phys. Chem. A 2008, 112, 2948–2954.
ꢁ
ꢁ
(b) Lois, S.; Flores, J.-C.; Lere-Porte, J.-P.; Serein-Spirau, F.; Moreau,
ꢁ
J. J. E.; Miqueu, K.; Sotiropoulos, J.-M.; Baylere, P.; Tillard, M.; Belin,
C. Eur. J. Org. Chem. 2007, 4019–4031.

Article
Chem. Mater., Vol. 22, No. 12, 2010 3739
Figure 3. (a, e, h, k) SEM images of the morphologies of 1-4, respectively (insets of a and e present flat and smooth end surfaces); (b, f, i, l) fluorescent
microscope photographs of the morphologies 1-4 that were grown under the conditions same as that of the SEM images; (c, g) fluorescent microscope
photographs of the wires of 1 and 2, respectively, taken under spot laser lighting, displaying the optical waveguide property; (d, j, m) polarized optical
microscope (POM) photographs of 1 (88 °C, cooling), 3 (84 °C, heating), and 4 (125 °C, heating), respectively.
the greater transition dipole moment of 1 (13 D) than that
of 2 (5 D), because the absorption strength (oscillator
strength f ) is in direct proportion to the square of transi-
tion dipole moment ( μe).20 However, 4 could not be com-
pared in the same way for its molecular symmetry differ-
ent from 1 and 2. To enable calculation by Lippert-
Mataga equation,21 we had technically made an assump-
tion that the flexible alkyl chains were negligible to the
effective cavity radius of these molecules (see Figure S3 in
the Supporting Information). Compound 5 (Chart 1) was
synthesized as a model to test this assumption. The result
that the difference of the μe values of 1 (13 D) and 5 (14 D)
is insignificant, leading to the validity of this simplified
method. Another notable phenomenon is that 4 losses
its vibration structure in both absorption and emission
spectra in comparison with 1-3, and has a much broad-
ened emission peak with a large Stokes shift. The dis-
appearance of vibration structure of 4 could be ascribed
to the reduction of its molecular symmetry, because the
molecular vibrations are coupled with electron transi-
tions. However, the broadening of emission spectra of 4
could not simply explained by the vibration pattern.
Actually, the HOMO-1 and LUMO of 4 are both u MOs
(antisymmetrical to symcenter, labeled by A), while
according to Laporte selective rule the transition u MO
and u MO is forbidden.22 Hence, there is no radiative
relaxation pathway from LUMO to HOMO-1. Also
taking its large μe (25 D) into consideration, an internal
charge transfer (ICT) state is suggested to exist in 40s
excited state,23 which becomes dominant as solvent
polarity increases.
Morphology and Structural Analysis. As discussed in the
previous session, the molecular symmetry plays an impor-
tant role in single molecular optical behaviors: the absence
of mirror planes (4 compared with 1 and 2) is more influ-
ential to the shape of molecular optical spectra than the
absence of symmetrical center (3). At the level of molecular
aggregate or crystal, the effect of molecular symmetry is
more pronounced. As shown in Figure 3 and Table 1, all
micronano structures of 1-4 grown from solution are
fibrous. It is very interesting that the absence of symmetrical
center leads to shortened fiber length (10 μm in Figure 3h
compared with >100 μm in panels a and e in Figure 3);
while the absence of mirror planes leads to narrowed fiber
width (100 nm in Figure 3k compared with 1-5 μm in
Figures 3a and 3e). To relate the molecular structure to the
macroscopic morphology, we obtained the single-crystal
structure of 1. Figure 4 shows that packing of alkyl chains is
more crucial in determining crystal structure than the
stacking of π planes because of the following observations:
(1) all alkyl chains are completely parallel and overlapped,
which maximized the interactions between alkyl chains; (2)
˚
no typical distance of π-π stacking (3.4 A) is found in the
single-crystal structure. Hence, the reason that molecular
symmetry affects macroscopic morphology is highly related
to the difference of arrangement of alkyl chains.
To further illustrate the molecular packing in the nano-
ofibers of 1, two-dimensional wide-angle X-ray diffraction
(2D-WAXD) experiments were conducted. 2D-WAXD
(20) Zollinger, H.; Color Chemistry: Syntheses, Properties and Applica-
tions of Organic Dyes and Pigments, rev. ed.; Wiley-VCH: Z€urich,
Switzerland, 2003; Chapter 2, pp 31-32.
(21) Lakowicz, J. R.; Principles of Fluorescence Spectroscopy, Springer
Science and Business Media, 2006, New York, p 208.
(23) Singh, M. K.; Pal, H.; Bhasikuttan, A. C.; Spare, A. V. Photochem.
Photobiol. 1998, 68, 32–38.
(22) Laporte, O.; Meggers, W. F. J. Opt. Soc. Am. 1925, 11, 459–460.

3740 Chem. Mater., Vol. 22, No. 12, 2010
Wang et al.
Figure 4. Single-crystal structure of 1: (a) geometry of a single molecule in the crystal, in which the two alkyl chains in diagonal are all-trans conformation,
˚
˚
˚
whereas the other two adopt the same direction with all-trans chains by one gauche turn; (b) single-crystal structure of 1, a = 9.21 A, b = 14.6 A, c = 27.96 A,
R = 79.02°, β = 84.70°, γ = 80.96°; the shading circle represents an aromatic column; the aromatic columns and the alkyl chains form a layered structure.
technique has been widely used in the structure elucida-
tions of the crystal or liquid crystal phase of polymers24a,b
and small organic molecules.24c To prepare appropriate
sample for 2D-WAXD (Figure 5a), a solvent-evaporation-
assisted dipping method for the growth of nanowire
arrays was adopted.25 Molecule 1 was first dissolved in
a solvent mixture of chloroform and ethanol. Because
chloroform is more volatile than ethanol and the total
volume of the solution of 1 was decreasing with time, the
nanowires of 1 precipitated and arrayed on the surface
of aluminum foil which was partially immersed in the
solution.25b Figure 5c of a SEM image shows that the
arrays of nanowires, though not perfect, do align parallel
with each other in certain domains. The 2D-WAXD
patterns were obtained under both vertical and parallel
geometries (Figure 5d,f) and the corresponding integral
XRD spectra were indexed by using the cell parameters of
the single crystal of 1 (Figure 5e,g). The peaks from 18 to
25° in Figure 5e are much stronger than that of Figure 5g,
which indicates that the normal directions of the lattice
planes corresponding to 18-25° are in parallel, at least
almost in parallel, with the direction of nanofibers. The
spots in 2D-WAXD pattern obtained from vertical geo-
metry are correlated with the diffraction of certain lattice
planes of the crystal of 1 (Figure 6, also see Figure S5 and
Table S2 in the Supporting Information).26 The (001)
diffraction appears on the equator, corresponding to a
˚
d-spacing of 27.4 A, which represents the distance bet-
ween two adjacent columns of π planes across the alkyl
chain layer. The (210) plane is near vertical to the plane
˚
(001), of which a d-spacing of 4.5 A is the distance of the
adjacent π planes within the column. The planes of (223),
(213) and are parallel to the direction of alkyl chains, and
˚
thus the d-spacings of 4.0 and 3.8 A stand for the distances
of the adjacent alkyl chains within the alkyl chain layer. It
is notable that the distances of lattice planes related to the
˚
alkyl tails of 1 (4.0 and 3.8 A) are almost the same as that
of polyethylene chain packed in orthorhombic crystal
lattice,27 which further suggests that the driving force of
(26) Both the distance and angle and of certain lattice plane were
calculated and compared in this identification process to ensure
correctness of matching. It is worth noting that the diffraction
(24) (a) Kasai, N.; Kakudo, M. X-Ray Diffraction by Macromolecules;
Springer Series in Chemical Physics; Springer, Heideburge, 2005; Vol.
80. (b) Yin, X.; Ye, C.; Ma, X.; Chen, E.; Qi, X.; Duan, X.; Wan, X.;
Cheng, S. Z. D.; Zhou, Q. J. Am. Chem. Soc. 2003, 125, 6854–6855.
˚
speckles in high θ region (with the lattice distance of 3-5 A) are
actually corresponding to several lattice planes instead of a single
one, thus the lattice index from 15 to 25° in panels e and g in
Figure 5 and lattice planes displayed in Figure 6 (except 001) are
only representatives of all lattice planes that contribute to certain
diffraction speckle. For detailed calculations and comparisons, see
Table S2 in the Supporting Information.
€
(c) Fischbach, I.; Pakula, T.; Minkin, P.; Fechtenkotter, A.; M€ullen, K.;
Spiess, H. W. J. Phys. Chem. B 2002, 106, 6408–6418.
(25) (a) Kaslter, M.; Pisula, W.; Wasserfallen, D.; Pakula, T.; Mullen,
€
K. J. Am. Chem. Soc. 2005, 127, 4286–4296. (b) Zhang, C.; Zhang,
X.; Zhang, X.; Fan, X.; Jie, J.; Chang, J. C.; Lee, C.-S.; Zhang, W.; Lee,
S.-T. Adv. Mater. 2008, 20, 1716–1720.
(27) Yin, Li.; Chen, J.; Yang, X.; Zhou, E. Polymer 2003, 44, 6489–6493.

Article
Chem. Mater., Vol. 22, No. 12, 2010 3741
Figure 5. (a,b) Preparation method for 2D-WAXD sample. A solution of 1 in chloroform/ethanol =1/ 4 (V/V) (5 mg/mL) in a 10 mL bottle was set in air
for 2 days. When the good solvent of chloroform was evaporated, the nanofibers of 1 were deposited on the surface of a aluminum foil with the fiber axis
preferentially perpendicular to the solution surface. (c) SEM images of nanowire arrays on the aluminum foil. (d, f): diffraction geometris of 2D-XRD
experiments (2θ of 1.2-28.0°, recorded, step 0.02°) with X-ray incident beam vertical to and parallel to the nanowires, respectively. (e, g) Integral
2D-WAXD patterns of d and f, respectively; the peaks are indexed on the basis of the single-crystal cell parameters of 1.
Figure 6. Indexing 2D-WAXD patterns on the basis of the single crystal structure of 1; the yellow ribbon represents a microfiber of 1 with the a* axis
parallel to the long axis of the fiber.
the packing of 1 in crystal state is largely related to the
interaction among the alkyl chains. As mentioned above,
the (001) diffraction is located on the equator, indicating
that the c* is perpendicular to the nanofiber axis which is
along the meridian. Hence, the end facet of nanofibers
of 1 should be almost parallel to the π plane of 1. This

3742 Chem. Mater., Vol. 22, No. 12, 2010
Wang et al.
Figure 7. (a, b) WAXD profiles of 1 at 135 °C and 4 at 60 °C, respectively. The intersections are the corresponding indexes.
conclusion is also in agreement with the structure analysis
24c,25a
have a higher position order than the molecules forming
simple SmA or SmC with a 1D positional order.30 The
€
based on 2D-WAXD patterns given by Mullen et al.
˚
Because the molecular weights (1185-1126 D) and
the mass percents of alkyl chains (57-60%) of 1-4 are
almost the same, the conclusion drawn from the crystal
structure of 1 should be applicable to the structures of --4
to some extent. However, we note that the angles of five-
member rings with respect to the PyQ core in 3,are diffe-
rent from those of six-member rings in 1. As a result, it is
unfavorable for 3 to achieve all-parallel conformation
for alkyl chains. Though alkyl chains are flexible in some
way, minor angle deviation may lead to systematically
deformation resulting in a fusiform morphology. For 4,
because of the nearer position of two adjacent alkyl
chains, they would pack with each other in the solid state,
and thus one pair of intermolecular interaction is replaced
by one pair of intramolecular interaction in comparison
with 1. One result of this change is that intermolecular
packing between two adjacent molecules along molecular
long axis (they clipped with each other in 1) would be
weakened, which may be a reasonable explanation for
the narrowed fiber width of 4. Another result is that melt
point of 4 (114 °C) is much lower than that of 1 (172 °C,
Table 1).
Liquid Crystals. The formation of mesophases of 1 and
4 were also observed.28 1 displayed a typical mosaic tex-
ture of smectic (Sm) phase29 at above 103 °C (Figure 3d).
Combining the WAXD result recorded at 135 °C (Figure 7a),
we consider that the LC phase of 1 has a layered structure.
Molecule 3 did not have a stable liquid crystal phase
region, but its focal conic texture, which is typical in SmA
phases,29a was observed when the sample was prepared by
drop cast. For 4, identifiable texture of LC phase was not
observed; but WAXD result shows that it actually has a
layered structure at elevated temperatures (Figure 7b).
For both 1 and 4, it is worth noting that a diffraction
which can be assigned to (010) diffraction appear in the
low-angle region (Figure 7a,b), suggesting that 1 and 4
d-spacing of the (010) plane of ∼10 A is fairly equal to the
width of the molecules, whereas the d-spacing of the (001)
is close to the molecular length calculated with an as-
sumption that the alkyl tails are in an all-trans conforma-
tion. However, although the layered structures render the
(00l) diffractions with l up to 5 for 1 and for 4, only (010)
diffractions are observed for these two molecules. There-
fore, 1 and 4 do not possess a perfect 2D positional order
that can be detected in a typical columnar LC phase. Two
factors are considered responsible to the formation of
these apparently transition LC states that have the order
higher than Sm phase but lower than columnar phase.
First, the orthogonal interactions of alkyl chains and
π planes make the cores and coronas tend to segregate,
thus a layered structure has formed like some surfactants
with along alkyl chains.31 Second, the lubrication pro-
videdby flexiblechains areimportant fordiscotic mesogens
to form column phase;30 however, the large aspect ratio
(5:1)32 of molecule 1 and molecule 4 leads to an extremely
anisotropic distribution of alky chains and results in a 2D
continuous layer of aromatic cores (based on single-crystal
structure of 1, Figure 4). As a result, the fact that the
aromatic core columns were not well-separated by flexible
alkyl chains has weakened the columnar feature, though
position order exists to some extent within each layer. It is
also notable that the d-spacing (Table 1) of the layered
˚
structure of 4 (39.4-40.0 A) is larger than that of 1
(34.8-35.8 A). As the lengths of 1 and 4 along molecular
˚
long axis are the same, this observation further supports the
weakened attachments between layers of 4 in compa-
rison with 1. Moreover, because the shape of 4 is rather
calamitic (rod-like) than sanidic (lath-like, and the shape
of 1-3 are more sanidic), 4 also has a nematic phase at
above 78 °C (Table 1). Here, though the phase identifica-
tion is preliminary, we can still conclude that the differ-
ent LC behaviors of 1 and 4 are largely dependent on the
(28) The absence of stable mesophase of molecules containing thio-
phene segments (2 and 3) is probably due to the lack of polar groups
between the aromatic core and these alkyl chains, see: Tschierske,
C. J. Mater. Chem. 1998, 8, 1485–1508.
(29) (a) Dierking, I.; Textures of Liquid Crystals; Wiley-VCH: Weinheim,
Germany, 2003; p 136. (b) Liu, P.; Wu, Y.; Pan, H.; Li, Y.; Gardner, S.;
Ong, B. S.; Zhu, S. Chem. Mater. 2009, 13, 2727–2732.
(30) De Genes, P.-G.; Prost, J. The Physcis of Liquid Crystals; Oxford
University Press: Oxford, U.K., 1995; pp 2-6.
(31) Chae, P. S.; Guzel, I. A.; Gellman, S. H. J. Am. Chem. Soc. 2010,
132, 1953–1959.
(32) The aspect ratio is measured according to the single-crystal struc-
ture of 1; the ratio of the length of molecular long axis and that of
molecular short axis is 4.8:1.

Article
Chem. Mater., Vol. 22, No. 12, 2010 3743
Figure 8. Waveguide characterizations of the microfibers of 1 and 2. (a, b) Fluorescent microscope photographs of the microfibers of 1 and 2 (gray value
images and painted with false color); (c) optical losses. The ratio of intensity Itip/I body against distance X between the excited spot and the emitting tips. The
curveswere fittedbyanexponentialdecayfunctionItip/Ibody =Aexp(-RX), whereAandR arefittingparametersandRrepresentsthe opticalloss(dB/μm).
intermolecular forces and the molecular shape (aspect
ratio).
of the amount of light escaping from the excitation spot
and that of light propagating along the fiber.33 The other
fitting parameter R presents the optical loss for a guide
length of 1 μm. The results of R are 0.024-0.039 dB/μm of
1 and 0.014-0.043 dB/μm of 2 (Figure 8) are comparable
to those of low-loss inorganic nanowires (0.026 dB/μm)34a-c
and are lower than those of previously reported organic
waveguide materials, such as dye aggregates (0.5 dB/μm)10b
and polyfluoroene (0.48 dB/μm).34d If the optical losses
were caused by reabsorption, R should be proportional to
extinction coefficient ε.10b Divided by the body concen-
tration 2.2 mol/L of 1 (estimated from the cell volume of
Low-Loss Waveguide Fibers. Another notable phenom-
enon is that 1-4 have remained fairly high fluorescence in
the solid state (Figure 3 and Figure S4 in the Supporting
Information). Especially, the fibers of 1 and 2 with their
high crystallinity (for 2, see XRD and TEM results in
Figure S8 in the Supporting Information) have active
waveguide property (Figure 3c,g). 3 and 4 could not be
applied as 1D waveguide materials, because that the nano-
structure of 3 absent a distinct end face and the fiber width
of 4 is quite limited.10a In this investigation, the intensity
at body excited site (Ibody) and emitting tip (Itip) were
recorded and the optical loss coefficient (R) was calcu-
lated by single exponential fitting [Ibody/Itip = Aexp-
(-RX), where X is the distance between the exciting site
and emitting tip]. Because the total gray values were
employed to analyze the optical loss, the obtained R is
an average evaluation of the waveguide efficiency of 1 and
2 through their full fluorescent region. The Ibody was
adopted as reference to reduce the fluctuation of the
overlap area of laser spot and fiber surface. The physical
meaning of fitting parameter A is suggested to be the ratio
3
˚
the single crystal of 1 as 3639 A ), the R of 1 could be
converted to 180 L mol-1 cm-1, which is equal to 0.2% of
the ε of 10s absoption peak at 464 nm. Hence, the overlap
of the absorption and fluorescent spectra of 1 is negligi-
ble. Despite the reduced reabsorption, there are other two
important factors that may contribute to the excellent
optical waveguide behaviors of 1 and 2: (1) both the fibers
of 1 and 2 have smooth surface and distinctly flat end
(34) (a) Law, M.; Sirbuly, D. J.; Johnson, J. C.; Goldberger, J.;
Saykally, R. J.; Yang, P. Science 2004, 305, 1269–1273. (b) Pan,
A.; Liu, D.; Liu, R.; Wang, F.; Zhu, X.; Zou, B. Small 2005, 1, 980–983.
(c) Barrelet, C. J.; Graytak, A. B.; Lieber, C. M. Nano Lett. 2004, 4,
1981–1985. (d) O' Carroll, D.; Lieberwirth, I.; Redmond, G. Small
2007, 3, 1178–1183.
(35) The emitting spectra of 1-4, see Figure S4 in the Supporting
Information: only slight shifts of the absorption and emission
maxima of 1, 3, and 4 were observed in going from solution to
solid; the emerging vibrational structure of the absorption spectra
of 1-4 were explained by the restriction of rotation freedom of the
periphery rings; the bathochromic shift of the fluorescent band of 2
is ascribed to the conformational change of thiophene rings in solid
state.
(33) The discussion on the physical meaning of A could not be found in
previous literature to the best of our effort, though this fitting
method was widely used, see ref 10. For one-dimensional wave-
guide, the propagating direction possesses one dimension of the
three dimensions of space, hence theoretical value of A would be
1/2 (= (1/3)/(2/3)) if the fluorescent light at excitation spot is
isotropic. According to our result, the experimental value of A is
indeed around 0.5, and hence the physical meaning of A is reason-
able.

3744 Chem. Mater., Vol. 22, No. 12, 2010
Wang et al.
facets which have minimized the optical losses caused by
scattering; (2) the steric hindrance between nonplanar
1 ꢀ 10-4 M solutions of 1-4 in tolene and ethanol were placed in
a sealed bottle at room temperature and the microfibers of 1-4
gradually precipitated during 1 week. Absorption spectra were
recorded on PerkinElmer Lambda 35 UV-vis spectrometer. PL
spectra were carried out on F4500 FL Spectrophotometer. Dif-
ferential scanning calorimetry (DSC) analyses were performed
on a METTER TOLEDO Instrument DSC822e calorimeter.
Polarized microscopic (POM) pictures were recorded on DMLP
Polar Microscope with Heat Stage produced by Leica. The SEM
images were recorded on a S4800 cold field-emission scanning
microscope. 1D WAXD was performed on a Philips X’pert Pro
with an X’Celerator detector and 2D WAXD experiments were
carried out on a Bruker D8 DISCOVER with a GADDS
detector. The X-ray wavelength was 0.154 nm.
˚
π planes minimizes effective π-π interaction (4.5 A), thus
the optical features in single molecular level remains in the
solid state.35 Similar conclusion has been drawn by Geerts
et al. in the study of the high fluorescent solids of tetra-
phenylpyrene derivatives.13
Conclusion
In summary, we have developed pyrazino[2,3-g]quin-
oxaline derivatives 1-4 for high efficiency active wave-
guide fibers. The effects of the molecular geometry and
symmetry on their optical properties, crystal dimensions,
and liquid crystal behaviors were investigated by a com-
bined application of optical spectroscopy, microscopy,
single crystal analysis, and 1D- and 2D-WAXD techni-
ques. At a single molecular level, the geometrical diffe-
rence of steric hindrance between 5-member aromatic
ring and 6-member aromatic ring leads to different extent
of conjugation. The calculated orbital diagram under the
principle of linear combination of molecular orbitals re-
veals the D-A feature of this set of molecules. The effects
of molecular symmetry of 1-4 not only are expressed in
electron-vibration spectra, but also are crucial to the
aggregates morphology. The absence of symmetrical
center in molecule is more unfavorable to fibrous growth
than the absence of mirror planes, because in this series
the alkyl chains tend to array in parallel and their inter-
actions are influential to the arrangements of such mole-
cules with nonplanar aromatic core. The liquid crystal
behaviors of 1 and 4 further confirm the role which the
alkyl chains play in the molecular packing: (1) they pro-
vide the microsegregation force; (2) their positions strongly
influence the molecular anisotropy and the strength of
intermolecular interactions. The structure-property in-
vestigations have provided a deeper understanding for the
structure and intermolecular interactions of such mole-
cular aggregates. In addition, our analyses on molecular
arrangements in the microfibers by combining single-
crystal and 2D-WAXD techniques may be a general method
for elucidating fiber-shape organic assembles.
Computational Detail. Atomic structures of 1, 2, 4, and other
molecular segments were optimized with density functional
theory (DFT) calculations using the B3LYP/6-31G (d). Mole-
cular orbital shapes and energies discussed in the text are those
calculated at the optimized geometries. Orbital pictures were
plotted with Gaussview4.1. Electronic excitation energies and
oscillation strengths were computed for the 60 lowest singlet
excited electronic states of 1 and 2 with timedependent (TD)
DFT calculations. In plotting computed electronic spectra, a
Lorentzian line width of 0.15 eV was superimposed to each
computed intensity. All quantum-chemical calculations were
performed with the Gaussian03 package.
Waveguide Characterization. The laser was provided by Co-
herent CUBE-405-50C and introduced into the Olympus BX51
microscope by a single mode optical fiber. The laser was focused
to 1 um in diameter through a 100X object len (NA=0.8). The
microfibers of 1 and 2 were dispersed in ethanol and then
deposited onto glass. Single wire with width of about 1 μm
was excited under dark field with laser spot through microscope
and recorded with a DVC-1412 M CCD camera. The data were
analyzed with Image J software and the relative gray values of
I(X)/Io were plotted against distance X to obtain the optical loss
value R.
Compound 6 was prepared according to Babudri et al.36 and
compound 11-14 was prepared according to Foster et al.37
General Procedure for Compounds 1-5. A mixture of diketone
(0.2 mmol), benzene-1,2,4,5-tetraamine tetrachloride (0.1 mmol),
and sodium acetate (0.4 mmol) in ethanol (10 mL) were refluxed
overnight under nitrogen. Ethanol was removed from the resulting
mixture under reduced pressure, and the residue was redissolved in
dichloromethane. After washing by water for two times, the com-
bined organic phase was dried and the solvent was removed under
reduced pressure. The residue was further purified by column
chromatography (silica gel, petroleum ether: dichloromethane =3:
1 to 1: 1) to give the desired products.
The crystalline fibrous aggregates of 1 and 2 were then
proved to be low-loss active waveguide fibers (0.02-
0.05 dB/μm), which suggests the success of our strategy
of integrating the nonplanar aromatic core to enlarge the
2,3,7,8-Tetrakis(4-(dodecyloxy)phenyl)pyrazino[2,3-g]quinoxaline
1
(1). Yield: 58%, as yellow solid. H NMR (CDCl3, 300 MHz,
˚
distance of molecular stacks (4.5 A) and the alkyl chain as
structural group to promote self-assembly. Hence, this
strategy may be a promising approach to achieve crystal-
line fluorescent materials for waveguides and lasers.
ppm): δ 8.88 (2H, s), 7.57-7.60 (8H, d, J=9.0 Hz), 6.88-6.91
(8H, d, J = 9.0 Hz), 3.98-4.02 (8H, t, J = 6.6 Hz), 1.76-1.85
(8H, m), 1.24-1.34 (72H, m), 0.86-0.91 (12H, t, J = 6.6 Hz).
13C NMR (CDCl3, 50 MHz, ppm): δ = 159.8, 154.2, 139.9,
131.4-130.9., 113.9-114.1, 67.7, 31.6, 29.8-28.9, 25.7, 22.4,
13.8. Elemental anal. Calcd for C82H118N4O4: C, 80.47; H, 9.72;
N, 4.58. Found: C, 79.86; H, 9.58; N, 4.63. MALDI-ICR-FTMS
m/z: calcd, 1223.923; found, 1223.928 [MHþ].
Experimental Section
General Methods. Commercial chemicals were purchased and
1
used as received. H and 13C NMR spectra were recorded on
Varian Mercury (200 MHz or 300 MHz) at room temperature in
CDCl3. All chemical shifts were reported in parts per million
(ppm). MS spectra were recorded on Fourier Transform Ion
Cyclotron Resonance Mass Spectrometer (Bruker APEX IV).
The microfibers of 1-4 were obtained by vapor diffusion method:
(36) Babudri, F.; Fiandanese, V.; Marchese, G.; Punzi, A. Tetrahedron
Lett. 1995, 36, 7305–7408.
(37) Foster, E. J.; Babuin, J.; Ngugen, N.; Williams, V. E. Chem.
Commun. 2004, 2052–2053.

Article
2,3,7,8-Tetrakis(5-dodecylthiophen-2-yl)pyrazino[2,3-g]quino-
Chem. Mater., Vol. 22, No. 12, 2010 3745
7.59-7.61 (8 H, d, Ar-H, J = 6.6 Hz), 6.91-6.93 (8 H, d,
Ar-H, J = 7.5 Hz), 3.86 (12 H, s, CH3). ESI-ICR-FTMS m/z:
calcd, 607.234; found, 607.234 [MHþ].
xaline (2). Yield: 70%, as reddish orange solid. 1H NMR
(CDCl3, 300 MHz, ppm): δ 8.63 (2H, s), 7.24-7.25 (4H, d, J =
3.6 Hz), 6.72-6.73 (4H, d, J=3.6 Hz), 2.86-2.91 (8H, t, J=7.5
Hz), 1.73-1.78 (8H, m), 1.27-1.41 (72H, m), 0.86-0.90 (12H, t,
J=6.6 Hz). 13C NMR (CDCl3, 50 MHz, ppm): δ=151.0, 147.7,
139.4, 138.7, 130.1, 124.5, 31.6, 31.3, 30.1, 29.6-28.9, 22.4, 13.8.
Elemental anal. Calcd for C74H110N4S4: C, 75.07; H, 9.36; N,
4.73; S, 10.83. Found: C, 74.85; H, 9.37; N, 4.66; S, 10.56. MALDI-
ICR-FTMS m/z: calcd, 1183.769; found, 1183.767 [MHþ].
2,3-Bis(4-(dodecyloxy)phenyl)-7,8-bis(5-dodecylthiophen-2-yl)-
pyrazino[2,3-g]quinoxaline (3). Yield: 13%, as yellowish orange
solid. 1H NMR (CDCl3, 300 MHz, ppm): δ 8.75 (2H, s),
7.55-7.58 (4 H, d, J = 8.7 Hz), 7.23-7.24 (2H, d, J =3.0 Hz),
6.87-6.90 (4H, d, J = 8.7 Hz), 6.73-6.74 (2H, d, J = 3.6 Hz),
3.98-4.02 (4H, t, J = 6.3 Hz), 2.86-2.88 (4H, t, J = 7.8 Hz),
1.73-1.80 (8H, m), 1.21-1.44 (72H, m), 0.86-0.96 (12H, t, J=
6.6 Hz). 13C NMR (CDCl3, 50 MHz, ppm): δ = 159.9, 150.9,
130.9-131.1, 130.1, 127.1, 124.54, 113.9-114.0, 110.0, 67.8,
31.2-31.6, 30.1, 29.3-28.2, 25.7, 22.4, 13.8. Elemental anal.
Calcd for C78H114N4O2S2: C, 77.82; H, 9.54; N, 4.65; S, 5.33.
Found: C, 77.57; H, 9.38; N, 4.54; S, 5.39. MALDI-ICR-FTMS
m/z: valcd, 1203.846; found, 1203.850 [MHþ].
1,2-Bis(5-dodecylthiophen-2-yl)ethane-1,2-dione (6). Yield: 40%,
1
a yellow solid. H NMR (CDCl3, 300 MHz, ppm): δ 7.87-7.88
(2H, d, J=3.9 Hz), 6.89-6.90 (2H, d, J=3.9 Hz), 2.87-2.92 (4H,
t, J=7.5 Hz), 1.71-1.75 (4H, m), 1.27-1.34 (36H, m), 0.87-0.91
(6H, t, J = 6.6 Hz). Elemental anal. Calcd for C34H54O2S2: C,
73.06; H, 9.74; S, 11.47. Found: C, 73.34; H, 9.88; S, 10.86. ESI-
ICR-FTMS m/z: calcd, 559.364; found, 559.364 [MHþ].
1-(3,4-Bis(dodecyloxy)phenyl)-2-phenylethane-1,2-dione (14).
Yield: 78%, a white solid. 1H NMR (CDCl3, 300 MHz, ppm): δ
7.95-7.98 (2H, m), 7.60-7.63 (1H, m), 7.57-7.58 (1H, d, J =
2.1 Hz), 7.48-7.53 (2H, d, J=8.1 Hz), 7.40-7.42 (1H, dd, J1=
1.8 Hz, J2 = 8.1 Hz), 7.26-7.27(1H, d, J = 2.7 Hz), 6.84-6.87
(1H, d, J = 8.7 Hz), 4.04-4.08 (4H, t, J = 5.4 Hz), 1.83-1.88
(4H, m), 1.30-1.47 (36H, m), 0.85-0.90 (6H, t, J = 6.0 Hz).
Elemental anal. Calcd for C38H58O4: C, 78.85; H, 10.10. Found:
C, 78.87; H, 10.03. ESI-ICR-FTMS m/z: calcd, 579.441; found,
579.440 [MHþ].
Acknowledgment. This work was supported by the Major
State Basic Research Development Program (2006CB921602
and 2009CB623601) from the Ministry of Science and Techno-
logy, and National Natural Science Foundation of China. We
thank Dr. Haibin Song for single crystal elucidation and Lin
Ding for SEM experiments.
2,7-Bis(3,4-bis(dodecyloxy)phenyl)-3,8-diphenylpyrazino[2,3-g]-
quinoxaline (4). Yield: 69%. 1H NMR (CDCl3, 300 MHz, ppm):
δ 8.98 (2H, s), 7.63-7.66 (4H, m), 7.41-7.43 (6H, m), 7.27-7.30
(2H, dd, J1=1.8 Hz, J2=8.1 Hz), 7.05-7.07 (2H, d, J=1.8 Hz),
6.88-6.91 (2H, d, J = 8.4 Hz), 4.02-4.07 (4H, t, J = 6.6 Hz),
3.75-3.79 (4H, t, J = 6.6 Hz), 1.83-1.88 (4H, m), 1.70-1.75
(4H, m), 1.22-1.49 (72H, m), 0.87-0.91 (12H, t, J = 6.9 Hz).
13C NMR (CDCl3, 50 MHz, ppm): δ = 155.4-155.5, 154.8,
150.5, 148.5, 143.7, 140.7-140.4, 139.5, 131.0, 129.9, 129.4,
128.5, 123.5, 116.9, 115.5, 97.6, 95.5, 88.7, 69.2, 69.1, 44.2-
43.7, 32.4-32.1, 29.9-29.1, 26.2, 22.9, 14.4. Elemental anal.
Calcd for C82H118N4O4: C, 80.47; H, 9.72; N, 4.58. Found: C,
80.46; H, 9.70; N, 4.56. MALDI-ICR-FTMS m/z: calcd,
1223.923; found, 1223.917 [MHþ].
Note Added after ASAP Publication. There was a spelling
error in an author’s name in the version published ASAP May 21,
2010; the corrected version was published May 25, 2010.
Supporting Information Available: Data of calculated energy
level of molecular orbitals, UV-vis and fluorescence spectra in
1
different solvent and in solid, DSC spectra, HNMR spectra,
and 13CNMR spectra (PDF); information of single-crystal
structure (CIF file). This material is available free of charge
via the Internet at http://pubs.acs.org.
2,3,7,8-Tetrakis(4-methylphenyl)pyrazino[2,3-g]quinoxaline (5).
1H NMR (CDCl3, 300 MHz, ppm): δ 8.89 (2H, s, Ar-H),