Structure of an unexpected trimer from the reaction of ageratochromene II with aluminum chloride

Article abstract of DOI:10.1002/mrc.1023

A new trimer from the reaction of ageratochromene [1] (6,7-dimethoxy-2,2-dimethyl-1-benzopyran) with anhydrous aluminum chloride was shown to be 3,4-dihydro-6,7-dimethoxy-2,2-dimethyl-3-(6′,7′ -dimethoxy-2′,2′-dimethyl-2H-1-benzopyran-4′-yl)-4-(3″, 4″-dih

Full text of DOI:10.1002/mrc.1023

MAGNETIC RESONANCE IN CHEMISTRY
Magn. Reson. Chem. 2002; 40: 458–460
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/mrc.1023
Spectral Assignments and Reference Data
Structure of an unexpected trimer from the
reaction of ageratochromene II with
aluminum chloride
O
O
R
OMe
OMe
MeO
MeO
1 R = H
Changhu Chu,1∗ Jiehan Hu,1 Tao XU,2 Hongbin Xiao1 and
Xinmiao Liang1
2 R = OMe
MeO
O
MeO
3
O
1
Dalian Institute of Chemical Physics, Chinese Academy of Sciences,
MeO
HB
HC
Zhongshan Road 161, Dalian 116011, China
Institute of Tropical and Subtropical Ecology, South China Agricultural
University, Guangzhou 510642, China
HD
HE
2
HA
MeO
Received 6 September 2001; revised 17 January 2002; accepted 21 January 2002
MeO
4
OCH3
13'
A new trimer from the reaction of ageratochromene [1]
(6,7-dimethoxy-2,2-dimethyl-1-benzopyran) with anhy-
drous aluminum chloride was shown to be 3,4-dihydro-
6,7-dimethoxy-2,2-dimethyl-3-(6ꢀ ,7ꢀ-dimethoxy-2ꢀ,2ꢀ-di-
methyl-2H-1-benzopyran-4ꢀ-yl)-4-(3ꢀꢀ,4ꢀꢀ-dihydro-6ꢀꢀ,
7ꢀꢀ-dimethoxy-2ꢀꢀ,2ꢀꢀ-dimethyl-2H-1-benzopyran-3ꢀꢀ -yl)-
2H-1-benzopyran. Its structure was confirmed by NMR
(1H, 13C, DEPT-135. COSY, HMBC, HSQC, TOCSY and
NOESY), IR, mass spectra and elemental analysis. Copy-
right  2002 John Wiley & Sons, Ltd.
6'
11
CH3
8
OCH3
7'
O
1
14
5'
2
7
9
14'
H3CO
CH
12 3
4'
10'
10
3
8'
H3CO
13
6
4
9'
5
Ha
Hb
Hc
Hd
3''
O
1'
He
Hf
3'
4''
5''
CH312"
2'
H3C
13"
H3CO
10''
CH3
11''
H3C
11''
6''
7''
12'
O
9''
1''
KEYWORDS: NMR; 1H NMR; 13C NMR; trimer; ageratochromene
II; assignment
14"
H3CO
8''
5
Scheme 1
INTRODUCTION
chemical shift of the olefinic proton in 5 can be estimated using the
empirical equation11 υH D 5.25 C 0.45 ꢀ 0.25 ꢀ 0.07 D 5.38, which is
identical with the experiment value υ 5.38).
The DEPT-135 spectrum of the trimer displayed six positive
signals between υ 21.46 and 28.81 arising from the six methyl groups.
The HSQC and HMBC correlation exhibited the connections of C2,
C20 and C200 to the attached methyl groups.
The DEPT-135 spectrum showed a negative signal at υ 24.13 ppm,
which corresponds to C400. From the HSQC correlation from C400 to
Hd (υ 2.73) and He (υ 2.50) and the COSY correlation from Hd to He,
the connection of Hd and He to C400 was deduced.
The DEPT spectrum displayed three saturated CH (C3, C4 and
C300) groups at υ 47.56, 40.63 and 43.49. From the HSQC correlation
from C3 (υ 47.56) to Ha (υ 3.17), from C4 (υ 40.63) to Hb (υ 3.35) and
from C300 (υ 43.49) to Hc (υ 2.02), the connections of Ha to C3, Hb to
C4 and Hc to C300 were obtained.
The ageratochromenes I (1) and II (2) have been shown to induce
precociousmetamorphosiswhenapplied to larval stagesofinsects.1,2
Both the synthesis of dimers (3 and 4) by dimeric reaction and
isolation of dimer (3) of ageratochromene II from the essential oil
of Ageratum conyzoides have been reported.3,4 Fraga and co-workers
reported the dimeric reaction of ageratochromene II in 1983; different
Lewis acids such as iron(III) chloride, zinc bromide and silver nitrate
supported in silica gel were used as dimeric agents to obtain these
dimers.5–9
Studies of some dimeric and trimeric ageratochromene I analogs
have been reported.10 However, the synthesis of trimer 5 has not
been reported previously and may prove to be of potential interest
as an antijuvenile hormone. Furthermore, neither the reaction of
ageratochromene II (2) with anhydrous aluminum chloride nor a
detailed structural description of this type of compounds has been
reported. We describe here the reaction of ageratochromene II with
aluminum chloride and discuss the structure of the trimer 5 in detail.
By analysis of the COSY and NOESY correlation from Ha to H
and from Hc to Hd and He, and the NOESY correlation from Hbc
to Ha and Hb, the connections among C3, C4, C300 and C400 can be
described as shown.
DISCUSSION
These connections coincide with the TOCSY correlation from Hc
to Hb and it precludes the structures A and B, because Hc cannot
arise from a benzyl proton. Hence 5 is the only reasonable structure
which is consistent with all spectral data.
In addition to the dimers 3 and 4, an unexpected solid was isolated
from the reaction of aluminum with ageratochromene II (2). The 1H
and 13C NMR spectra data in CDCl3 of 5 are shown in Table I.
This solid is a trimer of ageratochromene II based on the
following facts: (1) the molecular formula of the solid was elucidated
to be C39H48O9 from its mass spectrum (EI, 70 eV, MC 660) and
elemental analyses; (2) the 1H NMR spectrum showed the presence
of six aromatic protons (sharp singlets at υ 6.90, 6.87, 6.47, 6.34 and
6.29). The DEPT spectrum showed only one negative signal at υ
24.13, which can be attributed to a CH2 group. Based on the facts
mentioned above, this solid may be one of the structures A, B, C and
5 (Scheme 2).
Based on the TOCSY correlation from Hf to H120 (υ 1.19), it was
deduced that C20 (υ 77.00) is adjacent to the olefinic link. Similarly,
from the TOCSY correlation from Ha to H11 (υ 1.53) and H12 (υ 1.26),
it can be deduced that C3 is close to C2. It is also seem likely that C200
is adjacent to C300.
The DEPT spectrum showed that these unsaturated CH groups
give 13C signals at υ 126.24, 113.32, 111.54, 107.35, 101.55, 101.32 and
101.19. From the HSQC correlation υ 126.24 to olefinic proton Hf,
the signal at υ 126.24 was assigned to the olefinic carbon C30. Based
on the empirical formula11 for 13C chemical shifts in substituted
benzenes, the signals at υ 113.32, 107.35 and 111.54 were assigned to
C5, C50 and C500 and the signals at υ 101.32, 101.55 and 101.19 were
assigned to C8, C80 and C800, respectively. The differences among
C5, C50 and C500 are due to their adjacent groups (CH, —C CH—
and CH2, respectively). Additionally, empirical estimates for the
chemical, shifts indicate that C5 is at highest frequency (υ 113.0302),
C50 is at lowest frequency (υ 111.54) and the middle one is C5 (υ
107.35). The TOCSY correlation from H5 to H8 and from H50 to H80
The 1H NMR spectra of this solid showed a singlet at υ 5.38(1H).
This is not consistent with structure C, since the olefinic proton
should appear at a frequency higher than υ 6.00. The chemical shift
of the olefinic proton Hf in 5 is close to the experimental data (the
ŁCorrespondence to: Changhu Chu, Dalian Institute of Chemical Physics,
Chinese Academy of Sciences. Zhongshan Road 161, Dalian 116011, China.
E-mail: changhuchu@yahoo.com
Copyright  2002 John Wiley & Sons, Ltd.

459
Spectral Assignments and Reference Data
Table 1. 13C and 1H NMR spectral data for 5
Position
13C
1H
Position
13C
1H
Position
13C
1H
2
3
4
76.19
47.56
40.63
20
30
40
77.00
126.24
133.01
200
300
400
75.34
43.49
24.13
3.17 (1H, d, 8.8, Ha)
3.40 (1H, d, 8.8, Hb)
5.38 (1H, s, Hf)
2.02 (1H, br, Hc)
2.73 (1H, dd, 8.4, Hd)
2.50 (1H, dd, 4, He)
6.29 (1H, s)
5
6
113.32
142.35
147.51
101.32
148.13
115.50
26.08
6.90 (1H, s)
6.34 (1H, s)
50
60
70
80
90
100
110
120
130
140
107.35
142.98
147.95
101.55
148.13
112.19
28.81
6.87 (1H, s)
6.47 (1H, s)
500
600
700
800
900
1000
1100
1200
1300
1400
111.54
142.98
146.51
101.19
149.93
113.42
26.92
7
8
6.29 (1H, s)
9
10
11
12
13
14
1.54 (3H, s)
1.26 (3H, m)
3.61 (3H, s)
3.77 (3H, s)
1.29 (3H, s)
1.19 (3H, s)
3.86 (3H, s)
3.86 (3H, s)
1.38 (3H, s)
1.43 (3H, s)
3.70 (3H, s)
3.79 (3H, s)
28.50
21.46
27.47
55.84
55.61
56.18
55.71
56.90
55.71
The HMBC correlation from C40 to Hf, H50 and Ha showed that
this signal at υ 133.01 arises from C40. From the empirical chemical
shifts of C6, C60 and C600, and the HMBC correlation, the C6, C60 and
C600 resonances were assigned.
Based on the DEPT spectrum and the HMBC correlation from C
to H and the estimated chemical shifts of C7, C70 and C700 from the
empirical formula,11 these signals at υ 147.51, 147.95 and 146.51 are
attributed to C7, C70 and C700, respectively. In the same way, it was
deduced that C9, C90 and C900 correspond to the signals at υ 148.13,
148.13 and 149.93, respectively.
O
H
H
MeO
MeO
H
O
H
H
H
O
MeO
OMe
MeO
OMe
A
OMe
OMe
O
Hd
He
Hc
C
Hb
C
Ha
C
MeO
MeO
C
H
H
H
MeO
MeO
O
H
H
H
EXPERIMENTAL
O
B
Melting-points were determined on a Kofler melting-point apparatus
and are uncorrected. Elemental analyses were carried out on a
Yanaco CHN Corder MT-3 analyzer. IR spectra were obtained in
KBr discs on a Nicolet FT-IR 170SX spectrometer. Mass spectra were
measured on an HP-5988A spectrometer (EI at 70 eV). 1H NMR
spectra ꢀCDCl3ꢁ were recorded on a JEOL FX-200X, spectrometer:
13C and all two-dimensional NMR spectra were recorded on a
Bruker-DRX400, instrument in 5 mm tubes at room temperature,
OMe
OMe
H
O
MeO
MeO
with TMS as an internal standard at a concentration of 40 mg mlꢀ1
.
O
H
H
H
H
The pulse conditions were pulse width H 10 µs and C 14 µs,
°
flip angle 90 , recycling delay (RD), 1.5 s and number of data
H
points 219.
MeO
MeO
O
Reaction of ageratochromene II with anhydrous aluminum
chloride
A mixture of ageratochromene II (2.2 g) and anhydrous aluminum
chloride (4 g) in untreated diethyl ether (50 ml) was stirred at
C
°
Scheme 2
ꢀ5 C for 15–20 min. The reaction mixture was diluted with 50 ml of
ice–water and filtered, and the water layer was extracted with diethyl
ether (3 ð 50 ml). The organic layer was collected, washed well with
brine, dried with anhydrous sodium sulfate and chromatographed
on silica gel using light petroleum–diethyl ether (4 : 1 to 1 : 1) as
eluent. Pure dimers 3 and 4 were obtained by recrystallization from
methanol and pure trimer 5 was obtained by recrystallization from
showed that C5 and C8, C50 and C80, and C500 and C800 are in three
different aromatic rings. The signals at υ 101.32, 101.55 and 101.19
were assigned to C8, C80 and C800, respectively.
The chemical shifts of C10, C100 and C1000 should be at the lowest
frequency among all aromatic quaternary carbon atoms. The DEPT
spectrum and empirical chemical shift showed that the signals at
υ 115.50, 112.19 and 113.42 corresponded to C10. C100 and C1000,
respectively.
light petroleum–diethyl ether (2 : 1).
5
Dimer 3, m.p. 155 C (lit., 154 C); yield, 12%; 1H NMR
°
°
(CDCl3 –TMS, 200 MHz), υ 1.24, 1.34, 1.40 and 1.49 (s, each 3H,
Copyright  2002 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2002; 40: 458–460

460
Spectral Assignments and Reference Data
CH3), 1.85 (m, 2H, 3-H), 3.48–3.60, (t, 1H, J D 7 Hz, 4-H), 3.75, 3.77
OCH3
and 3.90 (s, each 3H, MeO), 6.00 (s, 1H, 40-H), 6.36, 6.39 and 6.41 (s,
each 1H, ArH).
OCH3
O
CH3
CH3
5
1
°
°
Dimer 4, m.p. 203–204 C (lit., 198–199 C); yield, 72%; H NMR
(CDCl3 –TMS, 200 MHz), υ 1.24, 1.34, 1.40 and 1.47 (s, each 3H, CH3),
1.3–1.6 (m, 1H, He), 2.10–2.28 (m, 2H, H and H ), 3.08–3.20 (m, 1H,
Hc), 4.45 (d, 1H, J D 7 Hz, Ha), 6.21, 6.34band 7.2d8 (s, each 1H, ArH).
H3CO
H3CO
°
Trimer 5, m.p. 88–89 C; yield, 14%; (found, C 70.87, H 7.37;
Ha
Hf
C39H48O9 requires C 70.89, H 7.32%); MS (EI, 70 eV), 660 (MC, 12),
645 (3), 440 (22), 439 (26), 425 (14), 409 (10), 383 (15), 323 (25), 260
(18), 259 (100), 221 (18), 219 (18), 212 (17), 207 (10), 205 (25), 195 (20),
Hb
Hd
He
O
Hc
H3C
CH3
CH3
167 (39); IR, 2920–2970, 1640, 1520, 1200, 1010 cmꢀ1
.
CH3
O
H3CO
H3CO
REFERENCES
HMBC Correlations of 5
1. Bowers WS, Ohta T, Cleere JS, Marsella PA. Science 1976; 193:
542.
2. Burt ME, Kuhr RJ, Bowers WS. Pestic. Biochem. Physiol. 1978; 9:
300.
3. Kasturi TR, Thomas M, Abraham EM. Indian J. Chem. 1973; 11:
91.
4. Kasturi TR, Abraham EM. Tetrahedron Lett. 1967; 27: 2573.
5. Kasturi TR, Abraham EM, Brown P. J. Chem. Soc., Perkin Trans. 1
1973; 2468.
OCH3
H
OCH3
CH3
CH3
O
H3CO
H3CO
H
Ha
H
6. Fraga BM, Garcia VP, Gonzalez AG, Hernandez MG, Hanson JR,
Hitchcock PB. J. Chem. Soc., Perkin Trans. 1 1983; 1(11): 2687.
7. Fraga BM, Garcia VP. J. Chem. Res. (S) 1986; 3: 94.
8. Fraga BM, Garcia VP. J. Org. Chem. 1987; 52: 5032.
9. Fraga BM, Cabrea I, Garcia VP. Heterocycles 1999; 51: 2747.
10. Vyas AV, Mulchandani NB. J. Chem. Res. (S) 1987; 4: 98.
11. Hu JH. Nuclear Magnetic Resonance Spectroscopy. Hydrocarbon
Processing Press: Beijing, China, 1988; 65–68 (in Chinese).
H
d H
bHf
O
He
Hc
CH3
CH3
H3C
CH3
O
H3CO
H3CO
NOESY Correlation of timer 5
Copyright  2002 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2002; 40: 458–460