Optical properties of novel 2,3-dicyano-5-methyl-6H-1,4-diazepine dyes in the solid state

Article abstract of DOI:10.1246/bcsj.78.1167

Novel nonplanar fluorescent dyes, 2,3-dicyano-7-methyl-6H-1,4-diazepines, were synthesized. The fluorescence intensity of 6-substituted 2,3-dicyano-5-[4-(diethylamino)styryl]-7-methyl-6H-1,4-diazepines in vapor-deposited film was on the order of the substituent at the 6-position: n-Bu, Et > t-Bu, H. That of 2,3-dicyano-5-[4-(dialkylamino)-styryl]-6-ethyl- 7-methyl-6H-1,4-diazepines in the solid state was on the order of the alkyl group: CH₂(3,5-(di-t-Bu)C₆H₃) > Bn > Et. Thus, the fluorescence intensity in the solid state basically increased with the bulkiness of the substituents on the chromophoric system. X-ray structure analysis clearly showed that the substituent at the 6-position and the dialkylamino moiety should inhibit intermolecular interactions between the chromophores so as to enhance the fluorescence intensity in the solid state.

Full text of DOI:10.1246/bcsj.78.1167

Ó 2005 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 78, 1167–1173 (2005) 1167
Optical Properties of Novel 2,3-Dicyano-5-methyl-6H-1,4-diazepine
Dyes in the Solid State
ꢀ
Emi Horiguchi, Shinya Matsumoto,1 Kazumasa Funabiki, and Masaki Matsui
Department of Materials Science and Technology, Faculty of Engineering, Gifu University, Yanagido, Gifu 501-1193
1Department of Environmental Sciences, Faculty of Education and Human Sciences,
Yokohama National University, 79-2 Tokiwadai, Hodogaya-ku, Yokohama 240-8501
Received November 26, 2004; E-mail: matsui@apchem.gifu-u.ac.jp
Novel nonplanar fluorescent dyes, 2,3-dicyano-7-methyl-6H-1,4-diazepines, were synthesized. The fluorescence
intensity of 6-substituted 2,3-dicyano-5-[4-(diethylamino)styryl]-7-methyl-6H-1,4-diazepines in vapor-deposited film
was on the order of the substituent at the 6-position: n-Bu, Et > t-Bu, H. That of 2,3-dicyano-5-[4-(dialkylamino)-
styryl]-6-ethyl-7-methyl-6H-1,4-diazepines in the solid state was on the order of the alkyl group: CH2(3,5-(di-t-
Bu)C6H3) > Bn > Et. Thus, the fluorescence intensity in the solid state basically increased with the bulkiness of the
substituents on the chromophoric system. X-ray structure analysis clearly showed that the substituent at the 6-position
and the dialkylamino moiety should inhibit intermolecular interactions between the chromophores so as to enhance the
fluorescence intensity in the solid state.
Organic electroluminescent (EL) devices have been inten-
sively studied for use in upcoming flat panel displays (FPD),
since they provide superior display characteristics, e.g. vivid
image production based on a high contrast ratio, a fast imaging
response, and low electricity consumption, compared with typ-
ical FPDs, such as a liquid-crystal display and a plasma dis-
play.1 Two types of organic materials have been applied to
EL devices as an emitter: low-molecular-weight organic dyes
and fluorescent polymers. The conventional EL devices using
organic dyes as an emitter consist of electron-transport materi-
als, emitters, hole-transport materials, and electrodes. Three
kinds of emitters for the primary colors (red, green, and blue)
are required for full-colored EL devices. Among them, red
emitters have usually been prepared with dopants.1 Only fewer
kinds of red emitters without dopants have been known: tetra-
cenes,2 pyrenes,3 perylenes,4 arylaminofumaronitriles,5 male-
imides,6 and azomethine dyes.7 When red emitters are used
without dopants, the close stacking of molecules in the solid
state should bring about a decrease in the EL intensity due
to intermolecular interactions. The introduction of bulky sub-
stituents into a fluorophore was considered to prevent this in-
terference to enhance the emission intensity. A great number
of materials researches have devoted efforts to develop EL
characteristics; in contrast, there are only several reports on
fundamental studies of the correlation among molecular struc-
ture, solid-state structure and fluorescent properties in the solid
state.8 Thus, the molecular design for solid-state fluorescent
materials is still uncertain compared to the remarkable prog-
ress in device technologies. We have proposed a novel dicya-
nodiazepine chromophore as a nonplanar fluorophore.9 The
methylene moiety of 2,3-dicyano-6H-1,4-diazepines was
found to project out of the conjugated planar ꢀ-plane. The cor-
relation between the crystal structure of 6-substituted 2,3-di-
cyano-5-[4-(diethylamino)styryl]-7-methyl-6H-1,4-diazepines
and the fluorescence intensity in the solid state was briefly re-
ported in a previous paper.10 The introduction of bulky sub-
stituents at the 6-position was found to prohibit cofacial mo-
lecular stacking in the solid state, so that the fluorescence
was not quenched, even in vapor-deposited films. In this paper,
substituent effects in 2,3-dicyano-6H-1,4-diazepine dyes on
the optical properties both in solution and in the solid state
are described in detail. Special attention is paid to the correla-
tion between the fluorescence intensity in the solid state and
molecular stacking.
Results and Discussion
Synthesis of 2,3-Dicyano-5-methyl-6H-1,4-diazepine
Dyes. The synthesis of 2,3-dicyano-5-methyl-6H-1,4-diaze-
pine dyes 7–9 is shown in Scheme 1. Diaminomaleonitrile
(1, DAMN) reacted with 3-substituted pentane-2,4-diones 2
to give 6-substituted 2,3-dicyano-5,7-dimethyl-6H-1,4-diaze-
pines 3, followed by condensation with aromatic aldehydes
4–6 to provide 7–9 in low-to-moderate yields.
Optical Properties of 2,3-Dicyano-5-methyl-6H-1,4-di-
azepine Dyes in Solution and in the Solid State.
The
UV–vis absorption and fluorescence spectra of 7a–d, 8, and
9 in chloroform are shown in Figs. 1 and 2, respectively. Their
characteristics are summarized in Table 1. No significant dif-
ference was observed in the absorption maxima (ꢁmax) and flu-
orescence maxima (Fmax), since the electronic structure in the
visible region is only slightly influenced by substituents R1 and
R2. The dyes 7a–c, 8, and 9, except for 7d, showed a ꢁmax of
around 490–498 nm with a molar absorption coefficients (") of
40100–43300 dm3 molꢁ1 cmꢁ1. Their Fmax values were ob-
served at around 586–608 nm. In the case of 7d, ꢁmax and
Fmax exhibit a small bathochromic shift compared with those
of other derivatives. This result is attributed to the difference
in the dominant stereo-isomer between 7d and the other deriv-
Published on the web June 6, 2005; DOI 10.1246/bcsj.78.1167

1168 Bull. Chem. Soc. Jpn., 78, No. 6 (2005)
Optical Properties of Diazepine Dyes
atives, as discussed in a later section. However, the fluores-
cence intensity is quite different. A quantum yield for 7a
was calculated to be 0.03. The relative fluorescence intensity
(RFI) was in the following order: 7a ꢂ 9, 7b, 7c > 8, 7d. This
could be attributed to the introduction of a flexible substituents
that can accelerate the internal conversion so as to reduce the
fluorescence intensity in solution.10
Figures 3(A) and (B) depict the UV–vis absorption and flu-
orescence spectra of 7–9 in vapor-deposited thin films, respec-
tively. Their characteristics are also listed in Table 1. The ꢁmax
values were observed at around 478–551 nm, being relatively
bathochromic compared with those in chloroform. The Fmax
values were observed in the range of 637–682 nm, being more
bathochromic compared with those in chloroform. Even
though the thickness (40–57 nm) and the absorbance (0.47–
0.55) of the films were almost similar, the fluorescence inten-
sity changed considerably: 9 > 8, 7c > 7b ꢂ 7d, 7a, being of
different order from that in solution. This change in the fluo-
rescence intensity should be affected by intermolecular inter-
actions in the solid state. Therefore, an attempt to analyze their
crystal structure was made in order to investigate the differ-
ence in the fluorescence intensity between in solution and in
the solid state.
Stacking of Molecules in Crystals. A crystal-structure
analysis was carried out for 7a–c and 8. Unfortunately, no suit-
able single crystal was obtained for 7d and 9. A detailed ex-
amination of the crystal structures made it clear that there
are three kinds of characteristic molecular pairs, as shown in
Figs. 4–6: stacking pairs between phenylene rings, between di-
cyanodiazepine moieties, and between whole chromophores.
Figure 4 shows the stacking pairs between the phenylene
Me
O
R1
O
Me
R1
Me
N
NC
NC
NH2
NH2
NC
NC
2
(CO2H)2, Benzene
N
Me
NR22
3
1
OHC
NR22
N
NC
NC
4-6
Piperidine, Benzene
R1
N
Me
7-9
R1: 2a, 3a = H, 2b, 3b = Et, 2c, 3c = n-Bu, 2d, 3d = t-Bu
R2: 4 = Et, 5 = Bn, 6 = CH2(3,5-di-t-Bu)C6H3
7a: R1 = H, R2 = Et, 7b: R1 = Et, R2 = Et, 7c: R1 = n-Bu,
R2 = Et, 7d: R1= t-Bu, R2 = Et, 8: R1 = Et, R2 = Bn,
9: R1 = Et, R2 = CH2(3,5-di-t-Bu)C6H3
Scheme 1.
1000
800
600
400
200
0
0.6
0.5
0.4
0.3
0.6
0.5
0.4
0.3
0.2
0.1
1000
800
600
400
200
0
UV-vis
FL
UV-vis
7b
FL
7a
7c
7b
7a
7d
9
8
7b
7c
7b
9
0.2
0.1
0
8
7d
0
300
400
500
600
700
800
300
400
500
600
700
800
Wavelength / nm
Wavelength / nm
Fig. 2. UV–vis absorption and fluorescence spectra of 7b,
8, and 9 (1 ꢃ 10ꢁ5 mol dmꢁ3) excited with the absorption
maximum in chloroform.
Fig. 1. UV–vis absorption and fluorescence spectra of 7a–d
(1 ꢃ 10ꢁ5 mol dmꢁ3) excited with the absorption maxi-
mum in chloroform.
Table 1. UV–vis Absorption and Fluorescence Properties of 7–9
aÞ
CHCl3
bÞ
Film
bÞ
Compd
R1
R2
ꢁmax ("max
/nm
)
Fmax
/nm
ꢁmax Fmax
/nm
Thickness
/nm
RFIcÞ
SSdÞ
RFIeÞ
SSdÞ
/nm
7a
7b
7c
7d
8
H
Et
n-Bu
t-Bu
Et
Et
Et
Et
Et
Bn
490 (40400)
498 (41700)
498 (43300)
514 (46800)
479 (40100)
490 (41900)
608
596
595
632
586
587
100fÞ
36
35
10
19
39
118
98
97
92
107
97
499
530
551
526
493
478
663
676
664
682
640
637
100
679
860
119
849
1132
164
146
113
156
147
159
44
42
40
57
49
40
9
Et
CH2(3,5-di-t-Bu)C6H3
a) Measured at the concentration of 1:0 ꢃ 10ꢁ5 mol dmꢁ3 at 25 ꢄC. b) Excited with the absorption maximum. c) Relative fluorescence
intensity was determined by considering the fluorescence intensity of 7a in chloroform as 100. d) Stokes Shift. e) Relative fluores-
cence intensity was determined by considering the fluorescence intensity of 7a on film as 100. f) Quantum yield was calculated to be
0.03.

E. Horiguchi et al.
Bull. Chem. Soc. Jpn., 78, No. 6 (2005) 1169
0.6
0.5
0.4
0.3
0.2
7d
8
9
7a
7b
(A)
7c
0.1
0
300
400
500
600
700
800
Wavelength / nm
7a
7b
2500
2000
1500
1000
500
(B)
9
8
7c
7b
7d
7a
0
300 400
500
600
700
800
Wavelength / nm
7c
8
Fig. 3. (A) UV–vis absorption and (B) fluorescence spectra
of 7–9 excited with the absorption maximum in vapor-de-
posited thin film.
Fig. 5. Stacking pairs of 7a, 7b, 7c, and 8 between dicyano-
diazepine moieties.
8, respectively. In compounds 7a, 7b, and 7c, the longest inter-
planar distance was observed in the case of 7b. This is ascribed
to a different conformation of the diethylamino group com-
pared to those of 7a and 7c. The ethyl groups of 7a and 7c
point to the same direction; whereas those of 7b are directed
towards opposite directions. Because phenylene rings of 7c
are less overlapped than those of 7a and 7b, the repulsive in-
teractions between the planes would be small to give the short-
est interplanar distance. On the other hand, the distance in 8 is
found to be considerably large. This must be due to the intro-
duction of bulky substituents on the amino group.
7a
7b
Figure 5 illustrates the stacking pairs between the dicyano-
diazepine moieties. The interplanar distance between planes
composed of the dicyanoethene moiety in the dicyanodiaze-
ꢀ
pine moieties were close: 3.389, 3.325, and 3.400 A for 7a,
7b, and 7c, respectively. The relative geometry for 8 was ob-
ꢀ
tained to be 3.564 A by calculating the interplanar distance be-
tween a plane composed of a part of the diazepine ring and the
ethene moiety of the styryl group. No significant difference
was observed among them, as compared with the case in the
stacking pairs between phenylene rings. Close intermolecular
atomic contacts were found between the nitrogen atoms of
the diazepine ring and the carbon atoms in the dicyanoethene
moiety in 7a and 7b. The cyano groups were closely stacked to
each other in 7c. These results indicate that the electrostatic in-
teractions should dominate the formation of this kind of stack-
ing pairs. On the other hand, cyano groups showed no influ-
ence on the stacking pair in 8, and there existed no significant
intermolecular atomic contact. Hence, van der Waals interac-
tions would be important in this case.
7c
8
Fig. 4. Stacking pairs of 7a, 7b, 7c, and 8 between phenyl-
ene rings.
rings. Two molecules are stacked in a tail-to-tail fashion. The
interplanar distances between the phenylene rings were calcu-
ꢀ
lated to be 3.636, 3.832, 3.415, and 5.235 A for 7a, 7b, 7c, and
A molecular pair stacked in a head-to-tail fashion was also

1170 Bull. Chem. Soc. Jpn., 78, No. 6 (2005)
Optical Properties of Diazepine Dyes
recognized in 7a, as shown in Fig. 6. The chromophores were
found to be completely overlapped and to be closely stacked
ꢀ
on each other. The interplanar distance was 3.257 A. This type
of stacking pair, as shown in Fig. 6. The complete overlap be-
tween ꢀ-conjugation systems in a crystalline state is consid-
ered to be a key structure for evaluating the fluorescence prop-
erties in the solid state. Regulation of this type of stacking
should be an effective strategy for enhancing the fluorescence
intensity in the solid state.
of pair was not observed for other derivatives. Because the
methylene groups at the 6-position in the diazepine ring are di-
rected to another molecule, the alkyl groups at the 6-position
must inhibit this type of stacking. The fluorescence intensity
of 7a in vapor-deposited films is considerably weak compared
to those of the others. From a structural point of view, this mo-
lecular pair is regarded as being a key structure for fluores-
cence intensity in the solid state. The stacking between ꢀ-con-
jugation systems, i.e. chromophores, is known to play a signif-
icant role in the fluorescence properties in the solid state.8
However, the actual influence of the interactions between
molecules on the fluorescence properties has not been fully un-
derstood. On the other hand, the electronic states of a parallel
dimer and H-aggregates have been successfully interpreted by
a theoretical treatment based on quantum chemistry.11 The par-
allel dimer and the related assemblies are generally non-fluo-
rescent. In the case of 7a, the excitation energies in the solid
state would be quenched by a cofacial stacking pair to show
weak fluorescence. Dye 7d also exhibited very weak fluores-
cence in vapor-deposited films. One of the reason for this is
the relatively weak fluorescence of 7d in a solution. However,
there also exists the possibility that 7d may form a similar type
Spectral Shift in Absorption. In the present dyes, sub-
stituents R1 and R2 are not directly correlated with the ꢀ-con-
jugation system. Therefore, the electronic structures of a mole-
cule are almost similar, except for 7d, because of the differ-
ence in an isomeric structure, as described in the next section.
The similarity of the electronic structures was confirmed by
semi-empirical molecular orbital calculations, as shown in
Table 2.12 No remarkable differences in ꢁmax, f, and the
HOMO and LUMO energy levels among 7a–d, 8, and 9 were
calculated. Thus, the relative difference in the fluorescence in-
tensity is considered to originate from the introduction of sub-
stituents. On the other hand, there exists a slight spectral shift
in the absorption spectra from solution to the solid state. One
of the reasons is an energy shift based on non-resonance inter-
actions.11 We also recognized that the effect of molecular de-
formation due to crystallization should not be neglected. This
effect is one of the well-known intermolecular interactions
showing an influence on the electronic states in the solid state.
Organic molecules are generally deformed by crystallization to
change the electronic states of molecules. Many examples of
color polymorphism have been successfully explained by this
effect.13 In the present study, this effect was found to contrib-
ute to a slight bathochromic shift from solution to the solid
state, as shown in Table 3. The other well-known intermolec-
ular interaction in the solid state is a resonance interaction in
the excited states (exciton interaction).11 In the present dyes,
however, the nearest-neighboring molecules are aligned in
such a way that the transition dipole moment in the visible re-
gion is arranged in an out-of-phase orientation. This state is
symmetrically forbidden to give no optical transition. Thus,
the two effects, that is the non-resonance interactions and the
effect of deformation, can contribute to the spectral shift in
these dyes from solution to the solid state.
Isomeric Structures of Dyes. The axial- and equatorial-
isomers can exist for 7b, 7c, 7d, 8, and 9. Because the axial-
Fig. 6. Stacking pair of 7a between whole chromophores.
Table 2. Calculated UV–vis Absorption Band for the Optimized Structure of Possible Isomers for 7a–d, 8, and 9aÞ
Compd
7acÞ
Conformation
—
ꢁ
max/nm
fbÞ
HOMO/eV
LUMO/eV
344.17
1.16
ꢁ7:5029
ꢁ0:8308
7bcÞ
7bdÞ
equatorial
axial
349.04
362.26
1.22
1.18
ꢁ7:3927
ꢁ0:7811
ꢁ7:4256
ꢁ0:8995
7ccÞ
7cdÞ
equatorial
axial
348.36
360.15
1.21
1.16
ꢁ7:4134
ꢁ7:4331
ꢁ0:7446
ꢁ0:8839
7ddÞ
7dcÞ
equatorial
axial
347.75
361.94
1.17
1.10
ꢁ7:4399
ꢁ0:7735
ꢁ7:4232
ꢁ0:9001
8cÞ
8dÞ
equatorial
axial
348.74
360.67
1.42
1.37
ꢁ7:3464
ꢁ7:3599
ꢁ0:7706
ꢁ0:9073
9cÞ
9dÞ
equatorial
axial
349.79
360.87
1.42
1.37
ꢁ7:2917
ꢁ0:7418
ꢁ7:3092
ꢁ0:8777
a) Calculated for R-isomers. b) Oscillator strength. c) Preferentially produced. d) Not detected.

E. Horiguchi et al.
Bull. Chem. Soc. Jpn., 78, No. 6 (2005) 1171
Table 3. Calculated UV–vis Absorption Band for 7a–c and 8 Using the Fractional Coordinate Sets
of X-ray Analysis
Compd
Conformation
ꢁ
max/nm
faÞ
HOMO/eV
LUMO/eV
7a
7b
7c
8
equatorial
equatorial
equatorial
equatorial
356.62
357.73
370.52
360.98
1.31
1.22
1.30
1.52
ꢁ7:3027
ꢁ7:2184
ꢁ7:1333
ꢁ7:1231
ꢁ0:7988
ꢁ0:6505
ꢁ0:7101
ꢁ0:6422
a) Oscillator strength.
1H NMR spectra of 7d showed that only axial-7d were formed.
This result can be attributed to the bulky equatorial t-butyl
group adjacent to the methyl groups to inhibit the aldol reac-
tion to preferentially give axial-7d. The molecular-orbital cal-
culations for the equatorial- and axial-isomers revealed that
the UV–vis absorption spectra of the axial-isomers exhibited
a bathochromic shift of about 10–20 nm, compared with those
of the related equatorial-isomers, as shown in Table 2. Thus,
the bathochromicity of ꢁmax and Fmax in 7d, in which only
the axial-isomers were produced, compared with those of 7b,
7c, 8, and 9, in which the equatorial-isomers were preferential-
ly provided, is attributed to stabilization of the LUMO energy
level.
R1= Et, n-Bu, t-Bu
R1= t-Bu
R1
H
Me
Me
NC
NC
R1
N
N
NC
NC
H
N
N
Me
Me
equatorial 3
axial 3
R1= Et, n-Bu, t-Bu
H
Ar
H
Me
NC
NC
R1
N
N
NC
N
R1
Conclusion
NC
N
The fluorescence intensity of novel nonplanar 2,3-dicyano-
5-[4-(dialkylamino)styryl]-7-methyl-6H-1,4-diazepines in the
solid state increased by introducing bulky substituents at the
6-position and the dialkylamino moiety, respectively, due to
the prevention of a complete overlap between the chromo-
phores. This structural feature is regarded as being the key
structure for solid state fluorescence, and thus a modification
of the molecular structure based on the present strategy should
be effective to develop of emitters for non-doped EL displays.
Me
Ar
R-equatorial 7-9
S-equatorial 7-9
R1= t-Bu
R1
Ar
R1
Me
NC
NC
H
N
N
NC
NC
H
N
N
Me
Ar
R-axial 7-9
Experimental
S-axial 7-9
Instruments.
Melting points were measured with a
Yanagimoto MP-S2 micro-melting-point apparatus. EIMS spectra
were recorded on a Shimadzu QP-1000 spectrometer. NMR spec-
tra were obtained by a Varian Inova 400 spectrometer. Elemental
analysis was performed with a Yanaco MT-6 CHN corder. UV–
vis absorption and fluorescence spectra were taken on Hitachi
U-3500 and F-4500 spectrophotometers, respectively. The fluores-
cence quantum yield was determined using quinine sulfate in 0.1
mol dmꢁ3 sulfuric acid as a reference (ꢁf ¼ 0:55, ꢁem ¼ 366
nm). Vapor-deposited films were prepared by a conventional vac-
uum deposition equipment (ULVAC VPC-60) on a slide glass.
The film thickness was measured with a Kosaka Laboratory
Ltd., Surfcorder SE-2300 Instrument.
Materials. DAMN (1) was purchased from Sigma-Aldrich.
4-(Diethylamino)benzaldehyde (4) were purchased from Tokyo
Kasei Co., Ltd. 3-Ethylpentane-2,4-dione (2b) was purchased
from Wako Pure Chemical Industries, Ltd. 1-Bromomethyl-3,5-
di-t-butylbenzene,15 3-butylpentane-2,4-dione (2c),16 3-t-butyl-
2,4-pentanedione (2d),17 and 2,3-dicyano-6H-1,4-diazepine
(3a)14 were prepared as described in the literature.
Fig. 7. Isomeric structures.
and equatorial-protons at the 6-position in diazepine ring can
1
be distinguished by H NMR spectroscopy, the isomers have
been identified.14 Furthermore, since these compounds have
an asymmetric carbon at the 6-position, the R- and S-isomers,
whose UV–vis absorption and fluorescence spectra are identi-
cal, can also exist. Therefore, four kinds of isomers, as shown
in Fig. 7, can exist for 7b, 7c, 7d, 8, and 9. The 1H NMR spec-
tra of 3b and 3c showed that only the equatorial-3 were
formed. The ethyl and butyl substituted equatorial-3 are more
stable than the axial-3 due to a steric repulsion between the al-
1
kyl group and the diazahexene moiety. Actually, the H NMR
spectra of 7b, 7c, 8, and 9 in CDCl3 indicated that only the
equatorial-isomers existed in solution. The formations of R-
and S-equatorial 7b, 7c, and 8 were confirmed by X-ray struc-
ture analysis. Meanwhile, the 1H NMR spectrum of 3d in
CDCl3 showed that both the equatorial- and axial-3d were
formed in a molar ratio of 3 to 2. The bulky t-butyl group
can show a steric repulsion for the methyl groups at the 5-
and 7-positions to produce both the equatorial- and axial-3d.
Then, equatorial- and axial-3d can form R-equatorial, S-equa-
torial, R-axial, and S-axial-7d. However, interestingly, the
Synthesis of N,N-dibenzyl- and N,N-bis(3,5-di-t-butyl-
benzyl)anilines. To a DMF solution (10 cm3) of aniline (465
mg, 5 mmol) were added triethylamine (1.31 g, 13 mmol) and
benzyl bromide or 3,5-di-t-butylbenzyl bromide (13 mmol). The
mixture was stirred at room temperature for 5 h. After the reaction

1172 Bull. Chem. Soc. Jpn., 78, No. 6 (2005)
Optical Properties of Diazepine Dyes
was completed, the mixture was poured into water, extracted with
ethyl acetate, and dried over sodium sulfate. After removing the
solvent, the product was isolated by column chromatography
(SiO2, hexane:ethyl acetate = 20:1). The physical and spectral
data are given below.
N,N-Dibenzylaniline: Yield 36%. oil. 1H NMR (CDCl3) ꢂ
4.64 (s, 4H), 6.70 (t, J ¼ 8:1 Hz, 1H), 6.74 (d, J ¼ 8:1 Hz, 2H),
7.17 (t, J ¼ 8:1 Hz, 2H), 7.24–7.32 (m, 10H). EIMS (70 eV)
m=z (rel intensity) 273 (Mþ; 34), 196 (16), 182 (19), 91 (100),
77 (25), 65 (21).
Synthesis of 2,3-Dicyano-5-methyl-6H-1,4-diazepine Dyes
7–9. To a benzene solution (15 cm3) of 6-substituted 2,3-dicya-
no-5,7-dimethyl-6H-1,4-diazepines 3 (1 mmol) were added piperi-
dine (5 drops) and 4-(dialkylamino)benzaldehydes 4–6 (1 mmol).
The mixture was stirred with a Dean–Stark trap (7a, 7b, and 7c;
reflux, 6 h, 7d; 50 ꢄC to reflux, 66 h, 8 and 9; 60 ꢄC to reflux,
7 h). After removing the solvent, the product was isolated by col-
umn chromatography (SiO2, 7a and 7b; chloroform:ethyl ace-
tate = 10:1, 7c and 8; toluene:ethyl acetate = 10:1, 7d; benzene:
ethyl acetate = 15:1, 9; toluene:ethyl acetate = 30:1) and recrys-
tallized (7a; benzene, 7b and 8; toluene, 7c; cyclohexane, 7d;
ethanol–water, 9; hexane). The physical and spectral data are
shown below.
2,3-Dicyano-5-[4-(diethylamino)styryl]-7-methyl-6H-1,4-di-
azepine (7a): Yield 50%. mp >300 ꢄC. 1H NMR (CDCl3) ꢂ 1.21
(t, J ¼ 7:2 Hz, 6H), 1.59 (s, 3H), 1.83 (br s, 1H), 3.43 (q, J ¼ 7:2
Hz, 4H), 4.57 (br s, 1H), 6.67 (d, J ¼ 15:9 Hz, 1H), 6.68 (d, J ¼
8:7 Hz, 2H), 7.44 (d, J ¼ 15:9 Hz, 1H), 7.45 (d, J ¼ 8:7 Hz, 2H).
EIMS (70 eV) m=z (rel intensity) 331 (Mþ; 57), 317 (85), 316
(100). Anal. Found: C, 72.52; H, 6.53; N, 21.38%. Calcd for
C20H21N5: C, 72.48; H, 6.39; N, 21.13%.
N,N-Bis(3,5-di-t-butylbenzyl)aniline:
Yield 25%. oil.
1H NMR (CDCl3) ꢂ 1.22–1.33 (m, 36H), 4.60 (s, 4H), 6.69 (t, J ¼
8:0 Hz, 1H), 6.81 (d, J ¼ 8:0 Hz, 2H), 7.10 (d, J ¼ 1:8 Hz, 4H),
7.17 (t, J ¼ 8:0 Hz, 2H), 7.29 (t, J ¼ 1:8 Hz, 2H). EIMS (70 eV)
m=z (rel intensity) 497 (Mþ; 34), 294 (98), 203 (58), 148 (43), 133
(49), 57 (100).
Synthesis of 6-Substituted 2,3-Dicyano-5,7-dimethyl-6H-1,4-
diazepines 3. To a benzene solution (20 cm3) of DAMN (1, 1.08
g, 10 mmol) were added oxalic acid (50 mg, 0.6 mmol) and pen-
tane-2,4-diones (2, 10 mmol). The mixture was refluxed for 5–31
h using a Dean–Stark trap. After removing the solvent, the product
was isolated by column chromatography (SiO2, 3b; ethyl acetate,
3c; ethyl acetate:hexane = 5:1, 3d; dichloromethane) and recrys-
tallized from benzene. The physical and spectral data are given
below.
2,3-Dicyano-5-[4-(diethylamino)styryl]-6-ethyl-7-methyl-6H-
Yield 23%. mp 196–198 ꢄC. 1H NMR
1,4-diazepine (7b):
(CDCl3) ꢂ 1.17 (t, J ¼ 7:4 Hz, 3H), 1.21 (t, J ¼ 7:1 Hz, 6H),
1.39 (t, J ¼ 7:4 Hz, 1H), 2.05 (s, 3H), 2.31–2.42 (m, 2H), 3.43
(q, J ¼ 7:1 Hz, 4H), 6.31 (d, J ¼ 15:0 Hz, 1H), 6.64 (d, J ¼
8:8 Hz, 2H), 7.41 (d, J ¼ 8:8 Hz, 2H), 7.68 (d, J ¼ 15:0 Hz,
1H). EIMS (70 eV) m=z (rel intensity) 359 (Mþ; 81), 344 (100).
Anal. Found: C, 73.58; H, 7.06; N, 19.27%. Calcd for C22H25N5:
C, 73.51; H, 7.01; N, 19.48%.
2,3-Dicyano-6-ethyl-5,7-dimethyl-6H-1,4-diazepine
(3b):
Yield 51%. mp 180–182 ꢄC. 1H NMR (CDCl3) ꢂ 1.51 (t, J ¼
7:4 Hz, 3H), 1.27 (t, J ¼ 7:4 Hz, 1H), 2.15 (s, 6H), 2.21–2.36
(m, 2H). EIMS (70 eV) m=z (rel intensity) 200 (Mþ; 77), 185
(100).
6-Butyl-2,3-dicyano-5,7-dimethyl-6H-1,4-diazepine
(3c):
6-Butyl-2,3-dicyano-5-[4-(diethylamino)styryl]-7-methyl-6H-
Yield 16%. mp 132–134 ꢄC. 1H NMR
Yield 20%. mp 122–124 ꢄC. 1H NMR (CDCl3) ꢂ 0.99 (t, J ¼
7:0 Hz, 3H), 1.32 (t, J ¼ 7:5 Hz, 1H), 1.42–1.46 (m, 4H), 2.13
(s, 6H), 2.22 (q, J ¼ 7:5 Hz, 2H). EIMS (70 eV) m=z (rel intensi-
ty) 228 (Mþ; 51), 186 (44), 185 (83), 172 (52), 171 (50), 55 (100).
6-t-Butyl-2,3-dicyano-5,7-dimethyl-6H-1,4-diazepine (3d):
1,4-diazepine (7c):
(CDCl3) ꢂ 1.00 (t, J ¼ 7:0 Hz, 3H), 1.21 (t, J ¼ 7:0 Hz, 6H),
1.27 (br s, 1H), 1.42–1.48 (m, 4H), 2.04 (s, 3H), 2.21–2.33 (m,
2H), 3.42 (q, J ¼ 7:0 Hz, 4H), 6.30 (d, J ¼ 15:0 Hz, 1H), 6.64
(d, J ¼ 8:8 Hz, 2H), 7.42 (d, J ¼ 8:8 Hz, 2H), 7.67 (d, J ¼
15:0 Hz, 1H). EIMS (70 eV) m=z (rel intensity) 387 (Mþ; 52),
372 (100). Anal. Found: C, 74.44; H, 7.56; N, 18.08%. Calcd
for C24H29N5: C, 74.38; H, 7.54; N, 18.07%.
6-t-Butyl-2,3-dicyano-5-[4-(diethylamino)styryl]-7-methyl-
6H-1,4-diazepine (7d): Yield 16%. mp 87–90 ꢄC. 1H NMR
(CDCl3) ꢂ 0.99 (s, 9H), 1.21 (t, J ¼ 7:1 Hz, 6H), 2.27 (s, 3H),
3.43 (q, J ¼ 7:1 Hz, 4H), 4.35 (s, 1H), 6.59 (d, J ¼ 15:4 Hz,
1H), 6.65 (d, J ¼ 9:0 Hz, 2H), 7.44 (d, J ¼ 9:0 Hz, 2H), 7.50
(d, J ¼ 15:4 Hz, 1H). EIMS (70 eV) m=z (rel intensity) 387
(Mþ; 76), 372 (82), 330 (100), 289 (31), 245 (34), 217 (35). Anal.
Found: C, 74.99; H, 7.68; N, 17.74%. Calcd for C24H29N5: C,
74.38; H, 7.54; N, 18.07%.
2,3-Dicyano-5-[4-(dibenzylamino)styryl]-6-ethyl-7-methyl-
6H-1,4-diazepine (8): Yield 39%. mp 200–202 ꢄC. 1H NMR
(CDCl3) ꢂ 1.15 (t, J ¼ 7:3 Hz, 3H), 1.36 (t, J ¼ 7:3 Hz, 1H),
2.03 (s, 3H), 2.30–2.37 (m, 2H), 4.73 (s, 4H), 6.32 (d, J ¼ 15:0
Hz, 1H), 6.73 (d, J ¼ 9:1 Hz, 2H), 7.20–7.37 (m, 10H), 7.38 (d,
J ¼ 9:1 Hz, 2H), 7.65 (d, J ¼ 15:0 Hz, 1H). EIMS (70 eV) m=z
(rel intensity) 483 (Mþ; 18), 392 (6), 91 (100). Anal. Found: C,
79.66; H, 6.33; N, 14.13%. Calcd for C32H29N5: C, 79.47; H,
6.04; N, 14.48%.
Yield 8%. mp 160–162 ꢄC. 1H NMR (CDCl3) ꢂ 0.84 (s, 1H),
ꢀ
ꢀ
ꢀ
ꢀꢀ ꢀꢀ
1.32 (s, 9H), 2.25 (s, 6H), 0.98 (s, 9H),
ꢀꢀ
2.31 (s, 6H),
4.08 (s, 1H) . EIMS (70 eV) m=z (rel intensity) 228 (Mþ; 12),
ꢀ
ꢀꢀ
213 (12), 172 (76), 57 (100), 41 (82). and
show the peaks
attributed to the equatorial- and the axial-isomers, respectively.
Synthesis of 4-(Dialkylamino)benzaldehydes 5 and 6. To a
DMF solution (2 cm3) of dialkylanilines (2 mmol) was added
dropwise a DMF solution (1 cm3) of phosphorus trichloride
ꢄ
(367 mg, 2.2 mmol), and allowed to stand at 0 C. The mixture
was stirred for 15 h at room temperature, then heated to 60 ꢄC
until the reaction was complete. The reaction mixture was poured
into water, alkalized, and extracted with dichloromethane. After
evaporating the solvent, the product was isolated by column chro-
matography (SiO2, chloroform). The physical and spectral data
are given below.
4-(Dibenzylamino)benzaldehyde (5): Yield 66%. mp 95–96
ꢄC. 1H NMR (CDCl3) ꢂ 4.75 (s, 4H), 6.79 (d, J ¼ 8:8 Hz, 2H),
7.21–7.38 (m, 10H), 7.69 (d, J ¼ 8:8 Hz, 2H), 9.73 (s, 1H). EIMS
(70 eV) m=z (rel intensity) 301 (Mþ; 10), 210 (6), 91 (100), 65
(15).
4-[N,N-Bis(3,5-di-t-butylbenzyl)amino]benzaldehyde
(6):
ꢄ
Yield 69%. mp 130–133 C. 1H NMR (CDCl3) ꢂ 1.22–1.38 (m,
36H), 4.71 (s, 4H), 6.84 (d, J ¼ 8:9 Hz, 2H), 7.02 (d, J ¼ 1:8
Hz, 4H), 7.34 (t, J ¼ 1:8 Hz, 2H), 7.71 (d, J ¼ 8:9 Hz, 2H),
9.34 (s, 1H). EIMS (70 eV) m=z (rel intensity) 525 (Mþ; 40),
322 (70), 203 (54), 148 (70), 133 (53), 57 (100).
5-{4-[Bis(3,5-di-t-butylbenzyl)amino]styryl}-2,3-dicyano-6-
ethyl-7-methyl-6H-1,4-diazepine (9): Yield 27%. mp 133–135
1
ꢄC. H NMR (CDCl3) ꢂ 1.16 (t, J ¼ 7:4 Hz, 3H), 1.27 (s, 36H),
1.38 (t, J ¼ 7:4 Hz, 1H), 2.05 (s, 3H), 2.29–2.43 (m, 2H), 4.69
(s, 4H), 6.34 (d, J ¼ 15:0 Hz, 1H), 6.79 (d, J ¼ 8:8 Hz, 2H),

E. Horiguchi et al.
Bull. Chem. Soc. Jpn., 78, No. 6 (2005) 1173
7.02 (s, 4H), 7.33 (s, 2H), 7.40 (d, J ¼ 8:8 Hz, 2H), 7.69 (d, J ¼
15:0 Hz, 1H). EIMS (70 eV) m=z (rel intensity) 504
(Mþ ꢁ (C4H9)2C6H3CH2; 44), 203 (85), 148 (17), 133 (35), 91
(17), 57 (100). Anal. Found: C, 81.43; H, 8.68; N, 9.82%. Calcd
for C48H61N5: C, 81.42; H, 8.68; N, 9.89%.
X-ray Crystallography. Single crystals for dyes 7a, 7b, 7c,
and 8 were obtained by a solvent-diffusion method using hexane
and chloroform. In all dyes, racemic crystals were obtained. The
details of the X-ray analysis for 7a, 7b, and 7c are reported in
Ref. 10. The diffraction data for 8 were collected by a RIGAKU
Raxis RAPID-F imaging-plate area detector using graphite mono-
Pigm., 64, 45 (2005).
a) H. Langhals, T. Potrawa, H. Noth, and G. Linti, Angew.
8
Chem., Int. Ed. Engl., 28, 478 (1989). b) K. Shirai, M. Matsuoka,
and K. Fukunishi, Dyes Pigm., 42, 95 (1999). c) M. Brinkmann,
G. Gadret, M. Muccini, C. Taliani, N. Masciocci, and A. Sironi,
J. Am. Chem. Soc., 122, 5147 (2000). d) M. Levitus, G. Zepeda,
H. Dang, C. Godinesz, T.-A. V. Khuong, K. Schemieder, and
M. A. Garcis-Garibay, J. Org. Chem., 66, 3188 (2001). e) K.
Hirano, S. Minakata, and M. Komatsu, Bull. Chem. Soc. Jpn.,
74, 1567 (2001). f) K. Yoshida, Y. Ooyama, H. Miyazaki, and
S. Watanabe, J. Chem. Soc., Perkin Trans. 2, 2002, 700. g) K.
Yoshida, Y. Ooyama, S. Tanikawa, and S. Watanabe, J. Chem.
Soc., Perkin Trans. 2, 2002, 708. h) Z. Fei, N. Kocher, J.
Mohrschladt, H. Ihmels, and D. Stalke, Angew. Chem., Int. Ed.,
42, 783 (2003). i) K. Shirai, M. Matsuoka, S. Matsumoto, and
M. Shiro, Dyes Pigm., 56, 83 (2003). j) C.-T. Chen, C.-L. Chlang,
Y.-C. Lin, L.-H. Chan, C.-H. Huang, Z.-W. Tsai, and C.-T. Chen,
Org. Lett., 5, 1261 (2003). k) Y. Sonoda, Y. Kawanishi, T. Ikeda,
M. Goto, S. Hayashi, Y. Yoshida, N. Tanigaki, and K. Yase,
J. Phys. Chem. B, 107, 3376 (2003).
ꢀ
chromated Cu Kꢃ radiation (ꢁ ¼ 1:54178 A). Crystal data for 8:
ꢂ
C32H29N5, Mw ¼ 483:61, triclinic, P1, Z ¼ 2, a ¼ 10:04ð3Þ,
ꢄ
ꢄ
ꢀ
b ¼ 10:30ð6Þ, c ¼ 13:79ð5Þ A, ꢃ ¼ 92:3ð2Þ , ꢄ ¼ 99:3ð1Þ , ꢅ ¼
104:3ð2Þꢄ, Dcalcd ¼ 1:181 g cmꢁ3, T ¼ 296 K, F000 ¼ 512, ꢆ ¼
5:54 cmꢁ1, 11307 reflections were collected, 4363 unique
(Rint ¼ 0:042). The structure was solved by a direct method
(SIR88)18 and refined by fill-matrix least-squares calculations.
3268 observed (I > 2ꢇðIÞ), 364 parameters, R1 ¼ 0:064, wR2 ¼
0:090, refinement on F, H atoms were located on the calculated
position and not refined. All calculations were performed using
the Crystal Structure 3.10 program package.19 Crystallographic
data have been deposited at the CCDC, 12 Union Road,
Cambridge CB2 1EZ, UK, and copies can be obtained on request,
free of charge, by quoting the deposition numbers for 7a
(CCDC222683), 7b (CCDC222682), 7c (CCDC222681), and 8
(CCDC230160).
9
a) E. Horiguchi, K. Shirai, M. Matsuoka, and M. Matsui,
Dyes Pigm., 53, 45 (2002). b) E. Horiguchi, K. Shirai, J.-Y. Jung,
M. Furusho, K. Takagi, and M. Matsuoka, Dyes Pigm., 50, 99
(2001).
10 E. Horiguchi, S. Matsumoto, K. Funabiki, and M. Matsui,
Chem. Lett., 33, 170 (2004).
11 M. Kasha, ‘‘Spectroscopy of the Excited State,’’ ed by B.
D. Bartolo, Plenum Press, New York (1976), p. 337.
12 Win MOPAC ver. 3.0 package, Fujitsu, Chiba, Japan.
13 J. Bernstein, ‘‘Polymorphism in Molecular Crystals,’’
Oxford University Press, New York (2002), and references cited
therein.
14 R. W. Begland, D. R. Hartter, F. N. Joes, D. J. Sam, W. A.
Sheppared, O. W. Webster, and F. J. Weigert, J. Org. Chem., 39,
2341 (1974).
References
1
J. Shinar, ‘‘Organic Light-Emitting Devices,’’ Springer-
Verlag, New York, Berlin, Heidelberg (2004).
S. A. Odom, S. R. Parkin, and J. E. Anthony, Org. Lett., 23,
4245 (2003).
2
3 Y. Aso, T. Okai, Y. Kawaguchi, and T. Otsubo, Chem.
Lett., 2001, 420.
15 K. R. Kopecky and M. Yeung, Can. J. Chem., 66, 374
(1988).
4
a) S. Toguchi, Y. Morioka, H. Ishikawa, A. Oda, and E.
Hasegawa, Synth. Met., 111/112, 57 (2000). b) M. Yoshida, A.
Fujii, Y. Ohmori, and K. Yoshino, Jpn. J. Appl. Phys., 35, L397
(1996).
5 H.-C. Yeh, S.-J. Yeh, and C.-T. Chen, Chem. Commun.,
2003, 2632.
16 N. Ono, T. Yoshimura, T. Saito, R. Tamura, R. Tanigawa,
and A. Kaji, Bull. Chem. Soc. Jpn., 52, 1716 (1979).
17 W. Jiao and T. D. Lash, J. Org. Chem., 68, 3896 (2003).
18 M. C. Burla, M. Camalli, G. Cascarano, C. Giacovazzo,
G. Polidori, R. Spagna, and D. Viterbo, J. Appl. Crystallogr.,
22, 389 (1989).
6 H.-C. Yeh, L.-H. Chan, W.-C. Wu, and C.-T. Chen, J.
Mater. Chem., 14, 1293 (2004).
19 Crystal Structure 3.10: Crystal Structure Analysis Package,
Rigaku and Rigaku/MSC (2000–2002).
7
S. H. Kim, S. H. Yoon, S. H. Kim, and E. M. Han, Dyes