We have synthesized and theoretically calculated 5-methylisoindolo[2,1-a]quinoline derivatives as novel near-infrared absorption dyes via a ruthenium-catalyzed one-pot metathesis/oxidation/1,3-dipolar cycloaddition protocol. The reactivity in 1,3-dipolar cycloaddition was governed by the electronic effect of aromatic ring substituents. Substrates with an electron-withdrawing group on the aromatic ring afforded higher yields. The maximal absorption wavelength of 3,5-dimethyl-11-phenylisoindolo[2,1-a]quinoline-7,10-dione and 11-(4-methoxyphenyl)-5-methylisoindolo[2,1-a]quinoline-7,10-dione in MeOH increased to 736 and 737 nm, although that of 3a was 727 nm.
We have synthesized and theoretically calculated 5-methylisoindolo[2,1-a]quinoline derivatives as novel near-infrared absorption dyes via a ruthenium-catalyzed one-pot metathesis/oxidation/1,3-dipolar cycloaddition protocol. The reactivity in 1,3-dipolar cycloaddition was governed by the electronic effect of aromatic ring substituents. Substrates with an electron-withdrawing group on the aromatic ring afforded higher yields. The maximal absorption wavelength of 3,5-dimethyl-11-phenylisoindolo[2,1-a]quinoline-7,10-dione and 11-(4-methoxyphenyl)-5-methylisoindolo[2,1-a]quinoline-7,10-dione in MeOH increased to 736 and 737 nm, although that of 3a was 727 nm.
New applications of
near-infrared (NIR) dyes have sparked recent
interest in these substances with strong NIR absorption, proving an
important and useful phenomenon. While fluorescence is particularly
applicable to labeling in microscopy,[1] organic
molecules with eminent NIR absorption have attracted interest owing
to the wide range of applications for their optical and electronic
properties.[2,3] Nowadays, nonmetallic organic dyes have
many potential advantages, such as being lightweight, low-cost, and
easily processed with colorful and transparent features. These classical
metal-free NIR organic dyes[4] include squaraines,[5] cyanines,[6] and the
particularly well-known rhodamines,[7] boron-dipyrromethenes[8] and diketopyrrolopyrroles[9] (Figure ). These
structures have received increasing attention from both industry and
academic research. However, the range of the reported NIR dye molecular
skeletons is limited, making the development of novel NIR dye core
structures highly desirable. Although NIR chromophores constructed
from novel nitrogen-containing polyheterocycles[10,11e] have been discovered recently, methods for obtaining these desired
NIR dyes have yet to be established. One such approach to obtaining
the desired NIR dyes is through the discovery of novel chromophores.
Figure 1
Metal-free
NIR absorption chromophores.
Metal-free
NIR absorption chromophores.In our search for a novel and efficient ruthenium-catalyzed
one-pot
reaction involving both metathesis and nonmetathesis reactions,[11,12] we previously developed a one-pot ring-closing metathesis (RCM)/oxidation/1,3-dipolar
cycloaddition protocol to prepare various isoindolo[2,1-a]quinolines 3 from N-allyl-2-alkenylaniline
derivatives 1 (Scheme ).[13] The key intermediate
in this reaction is likely to be azomethine ylide I derived
from 1,2-dihydroquinoline 2.
The isoindolo[2,1-a]quinoline motif (Figure ) is attractive because
of its biological properties.[14] For instance,
5,11-dioxoisoindolo[2,1-a]quinolines are analogs
of berberine alkaloids that have been shown to have an inhibitory
effect against N2-induced hypoxia[14a] and inhibit human topoisomerase II.[14b] Meanwhile, 7,10-dioxoisoindolo[2,1-a]quinolines,
first synthesized by our group,[13] are functional
dyes. In particular, we found that compound 3 showed
a variety of colors when exchanging the R1 substituent
(R1 = Ph, Me), including blue, yellow, and red (Scheme and Figure ), despite the core structure
of 3 being the same as that of the 7,10-dioxoisoindolo[2,1-a]quinoline system.
Figure 2
Structures of isoindolo[2,1-a]quinolines.
Figure 3
Colors of compounds 3a–c (500 μM in toluene).
Structures of isoindolo[2,1-a]quinolines.Colors of compounds 3a–c (500 μM in toluene).Blue compound 3a (Figure ), which bears a methyl substituent at the
5-position of isoindolo[2,1-a]quinoline, clearly
showed absorption peaks in the NIR region with a maximum at 673–720
nm (Figure ). In contrast,
the maximal absorption wavelengths of yellow compound 3b and red compound 3c containing phenyl substituents
at the 5-position absorbed at 447–452 nm and 477–494
nm, respectively.
Figure 4
Absorption spectra of compounds 3a–c (200–900
nm; 50 μM in CHCl3).
Absorption spectra of compounds 3a–c (200–900
nm; 50 μM in CHCl3).Considering that the chemical yield of 5-methylisoindolo[2,1-a]quinoline 3a from corresponding precursor 1a was low, and that 3a was a novel metal-free
NIR absorption dye, we decided to optimize the synthetic conditions
to afford 3a and synthesize 5-methylisoindolo[2,1-a]quinoline derivatives 3 with different substituents
on the aromatic ring to better understand the relationship between
the structure and NIR absorption. We also performed theoretical calculation
of the newly synthesized 5-methylisoindolo[2,1-a]quinoline
derivatives 3 (Scheme ).
Scheme 2
This Work: Synthesis, Absorbance Properties, and Theoretical
Calculation
of 5-Methylisoindolo[2,1-a]quinolines 3
Results and Discussion
N-Allyl-N-benzylaniline derivatives 1 were prepared systematically and efficiently. Our tandem
catalysis strategy was first investigated using N-allyl-N-benzylaniline derivatives 1a, 1,4-benzoquinone, and ruthenium carbene catalysts. We first optimized
the reaction conditions of the one-pot reaction, as shown in Table . Entries 2–4
show that a higher loading of the Grubbs II catalyst (20 mol %) was
not effective, with a 1 mol % loading proving sufficient for this
one-pot reaction. Furthermore, benzene was successfully substituted
with less-toxic toluene as the solvent (entry 5), affording 3a in 18% yield in a shorter reaction time of 10 min (entry
6). We next optimized the reaction concentration. When the one-pot
reaction was performed in a 0.1 M solution, 3a was obtained
in 5% yield with a cross metathesis product obtained in 28% yield
(entry 7). The chemical yield of 3a was not improved
under glovebox conditions (entry 8). We next applied these optimal
conditions (entry 6) to the synthesis of substituted derivatives of
5-methylisoindolo[2,1-a]quinoline 3.
Table 1
Optimization of the One-Pot RCM/Oxidation/1,3-Dipolar
Cycloaddition Protocol
RCM
1,3-dipolar cycloaddition
isolated yield
entry
Grubbs II (mol %)
solvent
temp. (°C)
conc. (M)
time (min)
temp. (°C)
conc. (M)
time (min)
(%, two steps)
1
10
benzene
reflux
0.01
30
reflux
0.01
60
15
2
20
benzene
reflux
0.01
30
reflux
0.01
60
11
3
5
benzene
reflux
0.01
30
reflux
0.01
60
16
4
1
benzene
reflux
0.01
30
reflux
0.01
60
18
5
1
toluene
80
0.01
30
80
0.01
60
18
6
1
toluene
80
0.01
10
80
0.01
30
18
7a
1
toluene
80
0.1
10
80
0.1
30
5
8b
1
toluene
80
0.01
10
80
0.01
30
15
Cross metathesis product was also
isolated in 28% yield.
Reaction
was performed in a glovebox.
Cross metathesis product was also
isolated in 28% yield.Reaction
was performed in a glovebox.Experiments performed to establish the substrate scope are summarized
in Scheme . The RCM
of 1 proceeded quantitatively, as determined by thin
layer chromatography (TLC) analysis, regardless of the substituent
attached to the aromatic ring. However, 1,3-dipolar cycloaddition
and oxidation were affected by the substituent on the aromatic ring.
Therefore, the electronic effects of substituents on the aromatic
ring were responsible for the substrate reactivity toward 1,3-dipolar
cycloaddition. Higher yields of one-pot reaction products 3d, 3e, 3f, 3g, and 3h were obtained using substrates with an electron-withdrawing group
on the aromatic ring. 3,5-Dimethylisoindolo[2,1-a]quinoline (3j) was synthesized in 17% yield. Although
generation of 3i and 3k was observed by
TLC, these products were unstable and decomposed during purification.
Naphthylamine derivative 1l was also converted to pentacyclic
compound 3l in 44% yield. 1,4-naphthoquinone was also
used as a 1,3-dipolarophile in the reaction, affording pentacyclic
compound 3m. We also synthesized 3n and 3o from the corresponding precursor in 6 and 12% yields, respectively.
Scheme 3
Scope of One-Pot RCM/Oxidation/1,3-Dipolar Cycloaddition Protocol
The all-obtained isoindolo[2,1-a]quinolines 3 were blue-colored solids. The
color of the solution of compound 3 in CHCl3 is shown in the Supporting Information S1 (Figure S1). Compounds 3d–3h containing electron-withdrawing
groups formed blue-colored solutions
in CHCl3. Meanwhile, solutions of methyl-substituted tetracyclic
compound (3j) and benzo[g]isoindolo[2,1-a]quinoline (3l) were light-green to green
in color. The solution of 3m in CHCl3, synthesized
from 1f and 1,4-naphtoquinone, had a deep blue color.
The color of a solution of 3n and 3o in
CHCl3 was blue.We then investigated the absorption
profiles of compounds 3 in CHCl3 (Figure ) and in MeOH (Figure ). The maximal absorption
wavelengths (λex) and molar absorption coefficients
(ε) of compounds 3 are shown in Table . Except for 3l, all compounds 3 showed similar absorption spectra
in the visible region in CHCl3. The maximal absorption
wavelength of 3d, with
a chloride substituent at the 4-position of the isoindolo[2,1-a]quinoline, was extended in the same way as that of 3a in both solvents. However, other chloro-substituted compounds 3e–3g had lower maximal absorption wavelengths and
molar absorption coefficients than 3a in all solvents.
A bromo-substituent (3h) affected the absorption characteristics
in a similar fashion to the chloro-substituent (3e).
In contrast, the maximal absorption wavelength of 3j,
with a methyl substituent at position 3 of the isoindolo[2,1-a]quinoline, was red-shifted compared to that of 3e. Particularly in MeOH, the maximal absorption wavelength of 3j was extended up to 736 nm. The weaker absorption of benzo[g]quinoline derivative 3l in CHCl3 was observed at a shorter wavelength. The maximal absorption wavelength
of 3n, with a p-methoxyphenyl substituent
at the 11-position of the isoindolo[2,1-a]quinoline,
was red-shifted compared to that of 3a. Particularly
in MeOH, the maximal absorption wavelength of 3n was
extended up to 737 nm. In contrast, the maximal absorption wavelength
of 3o, with a p-chlorophenyl substituent
at the 11-position of the isoindolo[2,1-a]quinoline,
was shorter than that of 3a. An electron-donating group
on the aromatic ring or heterocycle affected the maximum absorbance
wavelength longer. Molar absorption coefficients were very low, that
is, <10 000.
Figure 5
Absorption spectra of compounds 3 (250–1100
nm; 50 μM in CHCl3).
Figure 6
Absorption spectra of compounds 3 (250–1100
nm; 50 μM in MeOH).
Table 2
Maximal Absorption Wavelengths and
Molar Absorption Coefficients of Compounds 3 (250–1100
nm; 50 μM in CHCl3 or MeOH)
CHCl3
MeOH
R1
R2
λex (nm)
abs.
ε
λex (nm)
abs.
ε
3a
H
H
722
0.376
7530
726
0.301
6020
3d
4-Cl
H
722
0.389
7770
727
0.258
5160
3e
3-Cl
H
688
0.250
5010
698
0.258
5160
3f
2-Cl
H
691
0.288
5760
688
0.239
3780
3g
1-Cl
H
690
0.287
5740
673
0.161
3230
3h
3-Br
H
685
0.301
6010
697
0.237
4750
3j
3-Me
H
726
0.298
5960
736
0.255
5100
3l
715
0.134
2690
715
0.134
2690
3n
H
OMe
724
0.292
5833
737
0.119
2379
3o
H
Cl
714
0.099
1971
717
0.132
2638
Absorption spectra of compounds 3 (250–1100
nm; 50 μM in CHCl3).Absorption spectra of compounds 3 (250–1100
nm; 50 μM in MeOH).In addition, we calculated all the above compounds 3a, 3d–h, 3j, 3l, 3m, and 3p (Figure )[13] in order to
understand a structure and an absorption–wavelength relationships
using time-dependent density functional theory (TD-DFT). All calculations
were performed using Gaussian 09 (Rev. C) software. The ground-state
geometries were optimized using the DFT method with B3LYP/6-31G(d).
Harmonic frequency calculations were carried out to confirm that these
geometries are stable with no imaginary frequencies. To obtain excited
state energies, we calculated vertical excitation energies by the
TD-DFT method with TD-B3LYP/6-31G(d). We compared experimental and
calculated excitation energies in the electron volt unit. Although
the calculated wavelength underestimated the experimental peak absorption
wavelength, we found the linearity between the calculated and the
experimental maximum absorption wavelength.
Figure 7
Structure of 3p. Absorption wavelength of 3p is 663 nm.
Structure of 3p. Absorption wavelength of 3p is 663 nm.Figure shows the
correlation between the calculated and the experimental absorption
wavelength in CHCl3. In Figure , the line is determined by the least squares
method using the regression equation. y = 1.0x + b. The differences in calculated and
experimental energies (1.7 vs 2.6 eV) might come from the shape and
dyes’ structures and charge-transfer band. According to the
regression analysis (Figure ), the compounds outside the 95% confidence interval were 3a, 3j, 3g, and 3p.
Especially, 3g seems to be an outlier. The difference
of 3g from other compounds is that R1 is Cl-substituent.
This big difference of experimental wavelength of 3g and
calculation might be caused by the repulsion between the Cl and phenyl
group which can produce the distortion of the chromophore structure.
These calculated results show that the TD-DFT approach is an effective
method to understand the relationships between the structure and the
absorption property.
Figure 8
Correlation between experimental and calculated wavelength
excitation
energies.
Correlation between experimental and calculated wavelength
excitation
energies.We investigated the absorption
profiles of 3j in various
solvents in Figure . A higher maximal absorption wavelength was observed for polar solvents,
as shown in Table , such as alcohol or DMSO. In dioxane, the maximal absorption wavelength
was about 50 nm lower than that in MeOH. The molar absorption coefficient
was stronger in CHCl3 and toluene than in other solvents.
Figure 9
Absorption
spectra of 3j in various solvents (250–1100
nm; 50 μM).
Table 3
Maximal
Absorption Wavelengths and
Molar Absorption Coefficients of 3j in Various Solvents
(250–1100 nm; 50 μM)
λex (nm)
abs.
ε
MeOH
736
0.255
5100
EtOH
733
0.250
5010
DMSO
732
0.247
4940
CHCl3
726
0.298
5960
DMF
722
0.239
4770
CH3CN
709
0.231
4620
acetone
707
0.247
4940
AcOEt
697
0.239
4770
toluene
693
0.262
5250
dioxane
689
0.243
4850
Absorption
spectra of 3j in various solvents (250–1100
nm; 50 μM).Finally,
we also synthesized 6-methylisoindolo[2,1-a]quinoline 5, a novel isomer of the methyl substituent,
from N-allyl-N-(2-methylallyl)-2-vinylaniline 4 via our one-pot strategy in 53% yield (Scheme ). The structure of 5 was determined by single-crystal X-ray diffraction (Figure ). According to X-ray diffraction,
there is no steric repulsion between the methyl substituent at position
6 and the carbonyl group at position 6 of isoindolo[2,1-a]quinoline. Thus, isoindolo[2,1-a]quinoline chromophore
of 5 has a planar structure.
Scheme 4
Synthesis of 6-methylisoindolo[2,1-a]quinoline 5
Figure 10
X-ray structure of 6-methylisoindolo[2,1-a]quinoline 5.
X-ray structure of 6-methylisoindolo[2,1-a]quinoline 5.A solution color
of the 6-methylisoindolo[2,1-a]quinoline 5 in MeOH is red, although that of 5-methylisoindolo[2,1-a]quinoline 3a is blue (Figure ). As both compounds 3a and 5 had the same core isoindolo[2,1-a] system,
we considered that the methyl substituent’s electronic effects
to the pi-conjugate system were dramatically changed dependent on
the methyl substituent’s position on the pi-conjugate system.
We then investigated the absorption profiles of compounds 3a and 5 in MeOH (Figure ). The maximal absorption wavelength of compound 5 in MeOH was 474 nm. Compound 5 has no absorption
in the NIR region. A molar absorption coefficient of 5 was 6730 M–1 cm–1, which is
higher than that of 3a. From these results, we surmised
that the most important thing for absorbing light in the NIR region
was the methyl substituent at position 5 of the isoindolo[2,1-a]quinoline chromophore.[15]
Figure 11
Colors of
compound 3a and 5 (50 μM
in MeOH).
Figure 12
Absorption spectra of 3a and 5 in MeOH
(250–1100 nm; 50 μM).
Colors of
compound 3a and 5 (50 μM
in MeOH).Absorption spectra of 3a and 5 in MeOH
(250–1100 nm; 50 μM).In summary, we synthesized isoindolo[2,1-a]quinoline
derivatives 3d–3h, 3j, 3l, 3n, and 3o as novel NIR absorption dyes
and found that, among them, 3j and 3n were
the red-shifted NIR dye in MeOH. We also found that the maximal absorption
wavelength of isoindolo[2,1-a]quinolines 3 was affected by solvent polarity and that TD-DFT was effective to
understand the structure and the absorption–wavelength relationships
of 3.
Experimental Section
General Information
Chemicals and solvents were either
purchased from commercial suppliers or purified by standard techniques.
All reactions were performed under a N2 atmosphere unless
otherwise noted. For TLC, silica gel plates Merck 60 F254 were used. 1H NMR were recorded at 300, 400, and 500
MHz. 13C NMR spectra were recorded at 101 and 126 MHz.
Chemical shifts are given in ppm relative to tetramethylsilane (TMS)
and the coupling constants J are given in Hz. The
spectra were recorded in CDCl3 as the solvent at room temperature
unless otherwise noted. TMS served as the internal standard (δ
= 0 ppm) for 1H NMR and CDCl3 was used as the
internal standard (δ = 77.0 ppm) for 13C NMR. Column
chromatography was performed with silica gel 60N (spherical, neutral,
63–210 μm, Kanto Chemical Co., Inc.), flash silica gel
60 (spherical, acid, 40–50 μm, Kanto Chemical Co., Inc.),
and flash silica gel 60N (spherical, neutral, 40–50 μm,
Kanto Chemical Co., Inc.) unless otherwise noted. Melting points were
determined on the heated plate and are uncorrected. HRMS (m/z) was measured using a MALDI (matrix-assisted
laser desorption/ionization)-TOF(time-of-flight) spectrometer unless
otherwise noted. 3a, 3b, and 3c are known compounds.[13]
General Procedure
for Preparation of N-Allyl-N-benzyl-o-isopropenylaniline 1
Procedure A for Synthesizing N-Benzyl-o-isopropenylaniline (Reductive
Amination)[16]
To a solution of o-isopropenylaniline
derivatives (1.0 equiv), PhCHO (1.2 equiv) and AcOH (1.2 equiv) in
toluene (0.1 M) were added NaBH(OAc)3 (2.0 equiv) and the
reaction mixture was stirred at rt for a while and then at refluxing
temperature for 12 h. After the reaction mixture was cooled to 0 °C,
saturated aqueous Na2CO3 was added to the reaction
mixture. The organic compounds were extracted with AcOEt. The organic
phases were washed with brine, dried over Na2SO4, and concentrated in vacuo. The obtained residue was prepurified
by filtration through a short-pad silica gel with n-hexane/AcOEt = 24:1 as an eluent to give a crude compound containing N-benzyl-o-isopropenylaniline.
Procedure
B for Synthesis of N-Benzyl-o-isopropenylaniline
(Alkylation)
A solution of o-isopropenylaniline
derivatives (1.0 equiv) in THF (0.1
M) was cooled to −78 °C and was stirred. To a stirred
mixture, n-BuLi (2.6 M in n-hexane,
1.0 equiv) was added dropwise and the whole mixture was stirred for
15 min. To a stirred mixture, BnBr (1.0 equiv) was added dropwise
and the stirred mixture was warmed to rt. After 1 h, the reaction
was quenched by the addition of water. The organic compounds were
extracted with AcOEt. The organic phases were washed with brine, dried
over Na2SO4, and concentrated in vacuo. The
obtained residue was prepurified by filtration through a short-pad
silica gel with n-hexane/AcOEt = 24:1 as an eluent
to give a crude compound containing N-benzyl-o-isopropenylaniline.
Synthesis of N-Allyl-N-benzyl-o-isopropenylaniline
(Allylation)
A solution of
crude N-benzyl-o-isopropenylaniline
in 1,4-dioxane (0.5 M) was cooled to 0 °C and was stirred. To
the stirred mixture, NaH (60% in mineral oil, 2.0 equiv) was added
and the whole mixture was stirred for 15 min. To a stirred mixture,
allyl bromide (2.0 equiv) and tetrabutyl ammonium fluoride (50 mol
%) were added and the reaction mixture was stirred at 100 °C
for 12 h. After the reaction mixture was cooled to 0 °C, saturated
K2CO3 was added to the reaction mixture in MeOH.
The organic compounds were extracted with AcOEt. The organic phases
were washed with water, brine, dried over Na2SO4, and concentrated in vacuo. The obtained residue was subjected to
column chromatography (neutral acid flash silica gel, n-hexane → n-hexane/AcOEt = 49:1) to give
a colorless or brown oil.
Following the general procedure A and allylation, 1n (1.15 g, 3.85 mmol, 64% in 2 steps) was obtained as a yellow
oil from commercial isopropenylaniline (0.85 mL, 6.00 mmol).1H NMR (CDCl3, 400 MHz): δ 7.23 (2H,
d, J = 8.3 Hz), 7.16 (2H, m), 7.12 (2H, d, J = 8.7 Hz), 6.97 (1H, dd, 8.0 Hz, 6.9 Hz), 6.87 Hz (1H,
d, J = 6.9 Hz), 5.81–5.71 (1H, m), 5.12–5.06
(4H, m), 4.16 (2H, s), 3.58 (2H, d, J = 6.4 Hz),
2.22 (3H, s); 13C{1H} NMR (CDCl3,
126 Hz): δ 147.8, 147.6, 138.6, 136.8, 134.6, 132.5, 130.4,
130.2, 128.2, 127.5, 122.5, 121.1, 117.7, 114.8, 55.3, 54.5, 22.4;
HRMS (MALDI): calcd for C19H21NCl, 298.1357
[(M + H)+]; found, 298.1356.
General Procedure for Preparation
of Isoindolo[2,1-a]quinoline 3
To a solution of N-allyl-2-alkenylaniline derivatives 1 (1.0 equiv) in
toluene (0.01 M) was added Grubbs II (1 mol %) and the reaction mixture
was stirred at 80 °C for 10 min. After completion of the RCM
(monitored by TLC), quinone (10 equiv) was added to the reaction mixture.
The reaction mixture was stirred at 80 °C for another 30 min
and then cooled to room temperature. To the reaction mixture saturated
aqueous NH4Cl was added. The organic compounds were extracted
with AcOEt. The organic phases were washed with brine, dried over
Na2SO4, and concentrated in vacuo. The obtained
residue was subjected to column chromatography (neutral flash silica
gel, n-hexane/AcOEt = 9:1 → 3:1; 2nd time, n-hexane/toluene/AcOEt = 6:18:1) to give 3.
To a solution of N-benzyl-2-vinylaniline[13] (88.0 mg, 0.598 mmol) in DMF (1.2 mL) was added
3-chloro-2-methyl-1-propene (117 μL, 1.20 mmol), K2CO3 (207 mg, 1.50 mmol) and TBAI (221 mg, 0.598 mmol)
and the whole was stirred at 120 °C overnight. After the reaction
mixture was cooled to rt, to the reaction mixture was added sat. K2CO3 in MeOH. The organic compounds were extracted
with AcOEt. The organic phases were washed with water, brine, dried
over Na2SO4 and concentrated in vacuo. The obtained
residue was subjected to preparative gel permeation liquid chromatography
(CHCl3) to give 4 (42.1 mg, 0.160 mmol, 27%)
as a colorless oil.1H NMR (CDCl3, 500
MHz): δ 7.50 (1H, dd, J = 7.8, 1.4 Hz), 7.31–7.15
(6H, m), 7.12 (1H, ddd, J = 7.8, 7.8, 1.4 Hz), 7.00
(1H, dd, J = 7.8, 7.8 Hz), 6.89 (1H, dd, J = 7.8, 1.4 Hz), 5.69 (1H, dd, J = 17.9,
1.4 Hz), 5.26 (1H, dd, J = 11.0, 1.4 Hz), 4.93 (1H,
s), 4.88 (1H, s), 4.14 (2H, s), 3.51 (2H, s), 1.72 (3H, s); 13C{1H} NMR (CDCl3, 126 MHz, 60 °C): δ
149.2, 142.8, 138.4, 134.9, 133.3, 128.7, 128.1, 128.0, 127.9, 126.9,
126.8, 121.9, 113.4, 113.2, 59.2, 57.5, 20.8; HRMS (MALDI): calcd
for C19H22N, 264.1747 [(M + H)+];
found, 264.1745.
6-Methyl-isoindolo[2,1-a]quinoline-7,10-dione
(5)
To a solution of N-benzyl-N-(2-methylallyl)aniline derivatives 4 (27.7
mg, 0.100 mmol) in toluene (10 mL), Grubbs II (8.5 mg, 0.010 mmol)
was added and the reaction mixture was stirred at 80 °C for 10
min. After completion of the RCM (monitored by TLC), 1,4-benzoquinone
(108 mg, 1.00 mmol) was added to the reaction mixture. The reaction
mixture was stirred at 80 °C for another 30 min and then cooled
to room temperature. Saturated aqueous NH4Cl was added
to the reaction mixture. The organic compounds were extracted with
AcOEt. The organic phases were washed with brine, dried over Na2SO4, and concentrated in vacuo. The obtained residue
was subjected to column chromatography (neutral flash silica gel, n-hexane/AcOEt = 9:1 → 3:1) to give 5 (17.8 mg, 0.0527 mmol, 53%) as a red solid.1H
NMR (CDCl3, 500 MHz): δ 7.39–7.31 (5H, m),
7.22 (2H, d, J = 7.4 Hz), 7.06 (1H, d, J = 7.4 Hz), 6.96 (1H, d, J = 10.3 Hz), 6.72 (1H,
d, J = 10.3 Hz), 6.62 (1H, d, J =
8.0 Hz), 6.37 (1H, d, J = 7.4 Hz), 5.03 (2H, s),
2.11 (3H, s); 13C{1H} NMR (CDCl3, 126 MHz): δ 183.3, 181.2, 142.5, 138.2, 135.4, 134.1, 132.9,
132.8, 130.2, 129.6, 129.4, 128.9, 128.2, 127.5, 126.8, 126.5, 125.8,
121.8, 118.8, 113.0, 24.1; HRMS (MALDI): calcd for C23H15NO2, 337.1097 [M+]; found, 337.1098;
mp 212.0–213.0 °C (recrystallized from n-hexane/AcOEt, red column).
Experiments for Photochemical
Properties Measurements[17]
A 10
mM DMSO stock solution of each
compound was prepared. Each spectrum was measured by using diluted
stock solution of the desired solvent. A quartz cuvette containing
the prepared solution was held on a cell holder placed in the light
path of a monochromator. UV/vis/NIR spectra were recorded in the range
from 250 to 1100 nm. The set-up parameters were as follows: UV/vis
bandwidth: 2 nm, NIR bandwidth: 8 nm, UV/vis response: 0.06 s, NIR
response: 0.06 s, path length: 1 cm, scan speed: 1000 nm/min, and
grating and detector switch-over: 850 nm.
Figure 1
Scheme 1
Figure 2
Figure 3
Figure 4
Scheme 2
Scheme 3
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Scheme 4
Figure 10
Figure 11
Figure 12