Oxidative coupling reactions mediated by MoCl5 leading to 2,2′-cyclolignans: The specific role of HCl

Article abstract of DOI:10.1002/ejoc.200300263

The scope and performance of oxidative coupling reaction using MoCl ₅ can be significantly improved by employing Lewis-acidic additives such as TiCl₄, SnCl₄, or SiC14. Since the by-product, HCl, plays a particular function

Full text of DOI:10.1002/ejoc.200300263

FULL PAPER
Oxidative Coupling Reactions Mediated by MoCl5 Leading to
2,2Ј-Cyclolignans: The Specific Role of HCl
Beate Kramer,[a] Roland Fröhlich,[a] and Siegfried R. Waldvogel*[a]
Keywords: Biaryls / C-C coupling / Cyclization / Lewis acids / Molybdenum
The scope and performance of oxidative coupling reaction
using MoCl5 can be significantly improved by employing
Lewis-acidic additives such as TiCl4, SnCl4, or SiCl4. Since
the by-product, HCl, plays a particular function as an inhib-
itor for MoCl5 by forming chloro complexes, exploiting typ-
ical chloride scavengers, like silver salts or molecular sieves,
is also successful.
subjected to the dehydrodimerization reaction. For the oxid-
ative coupling reaction, a para-substitution pattern — with a
donor group on the aryl system — is pivotal for allowing ac-
cess to the 2,2Ј-cyclolignans 5a,e−g.
A variety of differently substituted 1,4-diarylbutanes 4a−g,
which were easily synthesized by a modular approach, were
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2003)
Introduction
ation involving a number of differently substituted 1,4-diar-
ylbutanes. Furthermore, we investigated additives that we
expected to support the molybdenum pentachloride me-
diated oxidative coupling reaction.
Eight-membered biaryl compounds, such as the 2,2Ј-
cyclolignans, are widely occurring natural products. The
most prominent method for the construction of these moi-
eties employs an oxidative coupling reaction. Recently, we
described a total synthesis of (ϩ)-detigloyloxysteganolide Results and Discussion
in which the key step was accomplished by a molybdenum
pentachloride mediated oxidative coupling reaction.[1] For
The synthesis of the precursors for the oxidative cycliza-
the selective eight-membered ring formation in the 2,2Ј-
cyclolignan synthesis, we believe that a preceding preor-
ganization occurs of the electron-rich aryl components with
the Lewis-acidic molybdenum pentachloride. While varia-
tions in the substitution pattern of the butylidene bridge
support this hypothesis, the exact nature of the pre-complex
is still unclear. It turns out that the dehydrodimerization
reaction involving molybdenum pentachloride prefers a 1,2-
disubstitution pattern of donor substituents on the aryl
components.[2] In addition, the oxidative coupling reaction
is not limited to methoxy-substituted arenes. Recently, we
developed several protective groups for the phenolic oxygen
atom that are compatible with the reaction conditions.[3]
Furthermore, this methodology has been exploited success-
fully for the construction of particular scaffolds that are
usually only accessible by multi-step procedures.[4Ϫ6]
tion to 2,2Ј-cyclolignans started with the readily available
3,4-dimethoxycinnamic acid, which provided, after several
standard transformations, the iodo derivative 1[7,8] that was
converted into the phosphonium salt 2. The straightforward
synthetic sequence, which involved a Wittig olefination fol-
lowed by a hydrogenation with a heterogeneous catalyst,
afforded the 1,4-diarylbutanes 4aϪg in good yields
(Scheme 1, Table 1). The formed triphenylphosphane oxide
and the accompanying salts were removed yielding a crude
mixture of olefinic isomers that were subjected directly to
the catalytic reduction. The compounds 4aϪg were ob-
tained in an analytically pure state. The modular approach
to the 1,4-diarylbutanes gave ready access to a variety of
differently substituted starting materials for the oxidative
cyclization reaction.
The oxidative coupling reaction employing molybdenum
pentachloride as the sole reagent gave the desired eight-
membered ring system in significant amounts only with
substrate 4e; the substitution pattern was similar to that
used in previous experiments.[1] Applying this methodology
to other substrates resulted in almost no conversion. Only
4a was converted into the 2,2Ј-cyclolignan 5a in moderate
yield (Scheme 2), which might be explained by the high oxi-
dation potential of the anisole moiety. In all cases, elevated
Since the naturally occurring 2,2Ј-cyclolignans exhibit
highly diverse substitution patterns on the biaryl moiety, we
focused our attention on the scope of the dehydrodimeriz-
[a]
Organisch-Chemisches Institut, Westfälische Wilhelms-
Universität Münster,
Corrensstrasse 40, 48149 Münster, Germany
Fax: (internat.) ϩ 49-251/8339772
E-mail: waldvog@uni-muenster.de
Eur. J. Org. Chem. 2003, 3549Ϫ3554
DOI: 10.1002/ejoc.200300263
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3549

B. Kramer, R. Fröhlich, S. R. Waldvogel
FULL PAPER
screening experiments.[12] The Lewis acidity of MoCl5 is not
known precisely, since common NMR spectroscopy tech-
niques are not compatible with its paramagnetic
nature.[13Ϫ15] Based on the Lewis acid catalyzed reactions,
MoCl5 is considered as active as TiCl4.[16] Initially, we an-
ticipated that a strong Lewis acid would activate the re-
maining molybdenum chloride clusters and, therefore, me-
diate the oxidative coupling reactions. The tested additives
are known not to react with MoCl5 at ambient tempera-
ture.[17] When NaCl is treated with a mixture of AlCl3 and
MoCl5, the more Lewis-acidic aluminum center assumes
the chloride anion, to generate the tetrachloroaluminate
anion, while the molybdenum chloride remains un-
changed.[18] During the oxidative coupling reaction, hydro-
gen chloride is formed as a by-product that might deactivate
the molybdenum reagent by forming chloro complexes with
MoCl5. The [MoCl6]Ϫ system is known, but it is considered
to be a complex having low stability.[19,20]
Scheme 1. a) PPh3, DMF, reflux; b) i. KOtBu, THF, 0 °C, ii. ben-
Using very strong Lewis acids, like SbCl5, resulted in an
immediate decomposition and chlorination of all substrates,
forming a sluggish reaction mixture. Employing AlCl3 led
to a significant increase in reactivity, except for the sub-
strate 4a, which resulted in a lower yield. Using a mixture
containing boron trifluorideϪdiethyl ether gave very similar
results to those of AlCl3. A vast improvement was obtained
by the addition of TiCl4 to the MoCl5-mediated oxidative
coupling reactions, which led to ameliorated yields of 2,2Ј-
cyclolignans. Alternative additives for substrates that are es-
pecially electron-rich, such as 4g, are SiCl4 and SnCl4, the
use of which result in cleaner reaction mixtures and excel-
lent yields. Less electron-rich substrates, like 4a and 4f, suf-
fer, however, even under these conditions, from low conver-
sions. It became clear that the excess of hydrogen chlo-
ride had to be eliminated from the reaction mixture. Com-
mon additives based on amines resulted in no reactivity at
all, since either the amines would coordinate directly to the
molybdenum center and decrease its electrophilic be-
havior[21,22] or the molybdenum center was directly reduced
because of the low oxidation potentials of amines. Addition
of inert amines, such as 2,2,6,6-tetramethylpiperidine, dis-
favored the dehydrodimerization reaction in all cases. Only
a low conversion was observed when the most successful
substrate, 4g, was allowed to react in the presence either of
the co-reagent 2,6-lutidine or 2,6-di-tert-butylpyridine.
Switching to typical chloride scavengers, like silver salts, im-
proved the transformation and gave good yields. Using mo-
lecular sieves as hydrogen chloride trapping agents had a
similar effect, and had the advantages of a simple workup
zaldehyde derivative; c) Pd/C, H2, THF, room temp.
Table 1. Synthesis of 1,4-diarylbutanes
Entry Compound R1
R2
R3
R4
Yield [%][a]
1
2
3
4
5
6
7
4a[9]
4b
4c
4d
4e
H
OCH3
H
OCH3
H
OCH3 OCH3
OCH3
H
OCH3
H
H
H
H
H
H
H
85
79
OCH3 78
OCH3 77
H
H
H
OϪCH2ϪO
89
83
87
4f
4g
H
OCH3 OCH3 OCH3
[a]
Refers to analytically pure material obtained over two steps
from 2.
temperatures, prolonged reaction times and higher amounts
of MoCl5 did not improve the outcome of the reaction, and
led mostly to degradation of the substrates.[10]
Scheme 2. a) MoCl5, additive, CH2Cl2, 0 °C
Lewis acids are usually added to enhance the power of and cost effectiveness. Performing the MoCl5-mediated oxi-
oxidizing agents,[11] because they affect the removal of a dative coupling reaction in the presence of 0.5 equiv. of
ligand or substituent from the oxidatively active material. LiCl resulted in a decrease in reactivity, which is demon-
The resulting electrophilic species is then more capable of strated by the very low conversions, even with substrate 4e.
mediating the redox process. Therefore, we screened several This finding clearly supports the particular inhibitor func-
Lewis-acidic additives in the oxidative coupling transform- tion of chloride anions in the MoCl5-mediated oxidative
ations of 4aϪg. We used equimolar amounts of additives coupling reaction (Figure 1).
relative to MoCl5 and monitored the reactions by GC. The
The detailed screening experiments reveal that the by-
best reagent combination was then used for the synthesis product, hydrogen chloride, plays a significant role in the
on a larger scale and displayed a good correlation with the reactivity of the dehydrodimerization reaction performed by
3550
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2003, 3549Ϫ3554

Oxidative Coupling Reactions Mediated by MoCl5 Leading to 2,2Ј-Cyclolignans
FULL PAPER
Figure 1. Screening for additives in the MoCl5-mediated oxidative
coupling reaction
MoCl5. For a successful oxidative coupling reaction leading
to the desired 2,2Ј-cyclolignans, a para-substitution pattern
with donor functions is pivotal. Despite involving two elec-
tron-rich arene systems, 4c and 4d were not suitable sub-
strates for the oxidative cyclization reaction. More import-
antly, an activated para position on the arene gives rise to
the formation of the eight-membered ring system. Other
oxidative coupling products, such as dimers or six-mem-
bered rings, were not found in the reaction mixtures. The
2,2Ј-cyclolignan 5a had properties identical with those re-
ported in the literature.[9] The other isolated cyclolignans,
5eϪ5g, were fully characterized. All NMR spectroscopic
signals were clearly assigned by NOE and COSY experi-
ments, which are described in detail for 5f, that consistently
support the exclusive formation of the eight-membered ring
systems. This feature was verified by X-ray crystallographic
analyses of single crystals of 5f and 5g that revealed, in both
cases, the same twist-boat-chair conformation,[23] which is
considered to be the preferred conformation for 2,2Ј-cyclo-
lignans.[24] The biaryl moieties are twisted by 64° in 5f and
60° in 5g (Figure 2).
Figure 2. Structures of 5f (top) and 5g (bottom) in the crystal
dative coupling reaction fails completely in the absence of
a para-substitution pattern.
Experimental Section
General Remarks: Tetrahydrofuran and dichloromethane were
dried with sodium wire and CaH2, respectively. Subsequent distil-
lation and handling was carried out under argon. Column chroma-
tography was performed on silica gel with a particle size of 63Ϫ200
µm from Merck KGaA (Darmstadt, Germany). All benzaldehyde
derivatives were used directly as purchased from Aldrich. Melting
points were determined with an MFB 595 Gallenkamp apparatus
(UK) and are uncorrected. Microanalysis was performed using a
Vario EL III (Elementar-Analysensysteme, Hanau, Germany).
NMR spectra were recorded with a Bruker ARX 300, AM 360 or
AMX 400 spectrometer using TMS as an internal standard. EI
Mass spectra were obtained using an MAT8200 system (Finnigan-
MAT, Bremen, Germany); ESI mass spectra were obtained using a
Quattro LCS (Waters-Micromass, Manchester, UK). Exact mass
was determined using
Manchester, UK).
a GC-TOF spectrometer (Micromass,
Conclusion
[3-(3,4-Dimethoxyphenyl)propyl]triphenylphosphonium Iodide (2): 4-
(3Ј-Iodopropyl)-1,2-dimethoxybenzene (1)[7,8] (10.0 g, 0.033 mol)
was combined with triphenylphosphane (8.8 g, 0.033 mol) in DMF
(50 mL) and the reaction mixture was heated under reflux over a
period of 4 h. Removal of the solvent by short-path distillation and
subsequent recrystallization provided pure colorless 2 (14.9 g,
84%). M.p. 161Ϫ162 °C (from MeOH/Et2O). 1H NMR (CDCl3,
300 MHz): δ ϭ 1.86Ϫ1.97 (m, 2 H, CH2ϪCH2ϪP), 2.93 (t, 3JH,H ϭ
The by-product, hydrogen chloride, limits the perform-
ance of MoCl5-mediated oxidative coupling reactions. Ad-
ditives consisting of simple Lewis acids, such as TiCl4,
SnCl4, or SiCl4, improve the transformation remarkably.
Furthermore, chloride scavengers like Agϩ or molecular si-
eves also lead to a significant amelioration in reactivity,
which is of particular synthetic interest. The appropriate
substitution pattern of the donor groups on the arene enti-
7 Hz,
2
H, CH2ϪCH2ϪCH2), 3.59Ϫ3.68 (m,
2
H,
arylϪCH2ϪCH2), 3.80 (s, 6 H, OCH3), 6.71Ϫ6.82 (m, 3 H, Harom.),
ties is decisive in the formation of 2,2Ј-cyclolignans: the oxi- 7.60Ϫ7.78 (m, 15 H, P(C6H5)3] ppm. 13C NMR (CDCl3, 75 MHz):
Eur. J. Org. Chem. 2003, 3549Ϫ3554
www.eurjoc.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3551

B. Kramer, R. Fröhlich, S. R. Waldvogel
FULL PAPER
2
δ
ϭ
21.7 (d, JC,P
ϭ
2
51 Hz, CH2ϪCH2ϪCH2), 24.6 148.7 (C-3Ј, C-4Ј), 160.6, 160.8 (C-3ЈЈ, C-5ЈЈ) ppm. MS (EI, 70 eV):
(arylϪCH2ϪCH2), 34.9 (d, JC,P ϭ 15 Hz, CH2ϪCH2ϪP), 55.9, m/z (%) ϭ 330.1 (100) [M]ϩ, 151 (70) [veratryl]ϩ. C20H26O4 (330.1):
56.3 (OCH3), 111.4 (C-3), 117.4 (C-6), 118.6 (C-5), 120.7 (C-4), calcd. C 72.70, H 7.93; found C 72.30, H 7.77.
130.3, 130.5, 132.5, 133.5, 133.6, 135.0 [P(C6H5)3], 147.5 (C-1),
1,2-Dimethoxy-4-[4-(2,5-dimethoxyphenyl)butyl]benzene (4d): Col-
149.0 (C-2) ppm. 31P NMR (CDCl3, 121 MHz): δ ϭ 25.0 ppm. MS
orless oil (3.05 g, 77%). 1H NMR (CDCl3, 400 MHz): δ ϭ
(ESI): m/z ϭ 441.3 [M Ϫ I]ϩ. C29H30IO2P (586.1): calcd. C 61.28,
3
3
1.73Ϫ1.84 (m, 4 H, 2-, 3-H), 2.72, 2.74 (tt, JH,H ϭ 7, JH,H
ϭ
H 5.32; found C 61.23, H 5.16.
7 Hz, 4 H, 1-, 4-H), 3.87, 3.89 (s, 6 H, 2ЈЈ-, 5ЈЈ-OCH3), 3.97, 3.98
(s, 6 H, 3Ј-, 4Ј-OCH3), 6.79Ϫ6.91 (m, 6 H, Harom.) ppm. 13C NMR
(CDCl3, 100.5 MHz): δ ϭ 29.4 (C-4), 30.0 (C-2), 31.3 (C-3), 35.3
(C-1), 55.5, 55.6, 55.7, 55.8 (OCH3), 110.6 (C-4ЈЈ), 111.2, 111.3,
111.9, 114.7 (C-2Ј, C-5Ј, C-3ЈЈ, C-6ЈЈ), 120.2 (C-6Ј), 132.3 (C-1ЈЈ),
135.4 (C-1Ј), 147.0, 148.8 (C-3Ј, C-4Ј), 151.7, 153.4 (C-2ЈЈ, C-5ЈЈ)
ppm. MS (EI, 70 eV): m/z (%) ϭ 330.1 (100) [M]ϩ, 151.1 (64)
[veratryl]ϩ. C20H26O4 (330.1): calcd. C 72.70, H 7.93; found C
72.47, H 7.99.
General Procedure for Preparation of 1,4-Diarylbutanes 4aϪg: A
chilled suspension of the phosphonium salt 2 (12 mmol) in dry
THF (25 mL) was treated with potassium tert-butoxide (12 mmol)
at 0 °C. The orange suspension was stirred at 0 °C for 2 h before
the substituted benzaldehyde (12 mmol) was added. After stirring
overnight, the turbid reaction mixture was filtered and the tri-
phenylphosphane oxide washed with diethyl ether. The solvents
were evaporated in vacuo. The residue was fractionated between
diethyl ether (3 ϫ 150 mL) and brine (100 mL). The combined or-
ganic phases were dried (MgSO4), filtered, and concentrated under
reduced pressure and then purified by column chromatography on
silica (cyclohexane/ethyl acetate). Subsequent hydrogenation of the
furnished substrates provided colorless oils or solids in pure form.
The substrates were dissolved in THF (25 mL) and Pd/C (10% Pd,
0.2 g) was suspended in the solution. These mixtures were stirred
at room temperature under H2 for 12 h. The catalyst was filtered
off and the solvent was evaporated. The different 1,4-diarylbutane
derivatives 4aϪg were synthesized in overall yields of 77Ϫ89%. For
systematic numbering of the carbon atoms see Scheme 3.
5-[4-(3,4-Dimethoxyphenyl)butyl]benzo[1,3]dioxole (4e): Colorless
crystals (3.36 g, 89%). M.p. 64 °C (from EtOH). 1H NMR (CDCl3,
300 MHz): δ ϭ 1.59Ϫ1.64 (m, 4 H, 2-, 3-H), 2.52Ϫ2.58 (m, 4 H,
1-, 4-H), 3.83, 3.85 (s, 6 H, OCH3), 5.89 (s, 2 H, OϪCH2ϪO),
6.59Ϫ6.79 (m, 6 H, Harom.) ppm. 13C NMR (CDCl3, 75 MHz): δ ϭ
31.0 (C-4), 31.2 (C-2), 35.3 (C-3), 35.4 (C-1), 55.8, 55.9 (OCH3),
100.6 (O-CH2-O), 108.0, 108.8 (C-2ЈЈ, C-5ЈЈ), 111.3, 111.8 (C-2Ј,
C-5Ј), 120.2, 121.0 (C-6Ј, C-6ЈЈ), 135.2 (C-1ЈЈ), 136.4 (C-1Ј), 145.5,
147.1 (C-3Ј, C-4Ј), 148.5, 148.8 (C-2ЈЈ, C-3ЈЈ) ppm. MS (EI, 70 eV):
m/z (%) ϭ 314.2 (100) [M]ϩ, 151.1 (99) [veratryl]ϩ, 135.1 (48)
[OϪCH2ϪOϪC6H3ϪCH2]ϩ. C19H22O4 (314.2): calcd. C 72.59, H
7.05; found C 72.63, H 7.19.
1,2-Dimethoxy-4-[4-(2,3-dimethoxyphenyl)butyl]benzene (4f): Color-
1
less oil (3.29 g, 83%). H NMR (CDCl3, 400 MHz): δ ϭ 1.66, 1.68
3
3
3
(tt, JH,H ϭ 5, JH,H ϭ 5 Hz, 4 H, 2-, 3-H), 2.61 (t, JH,H ϭ 7 Hz,
3
2 H, 1-H), 2.68 (t, JH,H ϭ 7 Hz, 2 H, 4-H), 3.82, 3.85, 3.86, 3.88
(s, 12 H, OCH3), 6.72Ϫ6.80 (m, 5 H, 2Ј-, 5Ј-, 6Ј-, 5ЈЈ-, 6ЈЈ-H),
6.96Ϫ7.00 (m, 1 H, 4ЈЈ-H) ppm. 13C NMR (CDCl3, 100.5 MHz):
δ ϭ 29.6 (C-4), 30.2 (C-2), 31.4 (C-3), 35.3 (C-1), 55.6, 55.7, 55.8,
55.9 (OCH3), 110.1 (C-4ЈЈ), 111.3, 111.9 (C-2Ј, C-5Ј), 120.2 (C-6Ј),
121.8, 123.6 (C-5ЈЈ, C-6ЈЈ), 135.3 (C-1ЈЈ), 136.3 (C-1Ј), 147.0, 147.1
(C-3Ј, C-4Ј), 148.8, 152.7 (C-2ЈЈ, C-3ЈЈ) ppm. MS (EI, 70 eV): m/z
(%) ϭ 330.2 (100) [M]ϩ, 151.1 (71) [veratryl]ϩ. C20H26O4 (330.1):
calcd. C 72.70, H 7.93; found C 72.53, H 8.19.
Scheme 3. Systematic numbering of the carbon atoms in the 1,4-
diarylbutanes and 2,2Ј-cyclolignans
1,2-Dimethoxy-4-[4-(2-methoxyphenyl)butyl]benzene (4b): Colorless
[4-(3,4-Dimethoxyphenyl)butyl]-2,3,4-trimethoxybenzene (4g): Col-
orless oil (3.76 g, 87%). 1H NMR (CDCl3, 300 MHz): δ ϭ
1.57Ϫ1.69 (m, 4 H, 2-, 3-H), 2.55Ϫ2.61 (m, 4 H, 1-, 4-H), 3.83,
3.84, 3.85, 3.86, 3.88 (s, 15 H, OCH3), 6.60 (s, 1 H, 4Ј-H),
6.69Ϫ6.83 (m, 2 H, Harom.), 6.76Ϫ6.81 (m, 2 H, Harom.) ppm. 13C
NMR (CDCl3, 75 MHz): δ ϭ 29.1 (C-1), 29.5 (C-3), 30.5 (C-2),
35.4 (C-4), 55.8, 55.9, 56.0, 60.7, 60.9 (OCH3), 107.3 (C-5Ј), 111.4,
112.0 (C-2ЈЈ, C-5ЈЈ), 120.3 (C-1Ј), 123.7 (C-6Ј, C-6ЈЈ), 135.5 (C-1ЈЈ),
142.4 (C-3Ј), 151.9 (C-2Ј, C-4Ј, C-3ЈЈ, C-4ЈЈ) ppm. MS (EI, 70 eV):
m/z (%) ϭ 360.2 (100) [M]ϩ, 181.1 (90) [(OCH)3C6H2CH2]ϩ, 151.1
(99) [veratryl]ϩ. C21H28O5 (360.2): calcd. C 69.98, H 7.83; found C
69.82, H 8.17.
crystals (2.84 g, 79%). M.p. 49 °C (from MeOH). 1H NMR
3
(CDCl3, 400 MHz): δ ϭ 1.66 (m, 4 H, 2-, 3-H), 2.62 (t, JH,H
ϭ
3
7 Hz, 2 H, 1-H), 2.68 (t, JH,H ϭ 7 Hz, 2 H, 4-H), 3.83, 3.87, 3.89
(s, 9 H, OCH3), 6.74Ϫ6.92 (m, 5 H, 2Ј-, 5Ј-, 6Ј-, 3ЈЈ-, 5ЈЈ-H),
7.14Ϫ7.21 (m,
2
H, 4ЈЈ-, 6ЈЈ-H) ppm. 13C NMR (CDCl3,
100.5 MHz): δ ϭ 29.4 (C-4), 29.9 (C-2), 31.4 (C-3), 35.3 (C-1), 55.1,
55.8, 55.9 (OCH3), 110.2 (C-3ЈЈ), 111.2 (C-2Ј), 111.8 (C-5Ј), 120.1
(C-5ЈЈ), 120.3 (C-6Ј), 126.8 (C-1ЈЈ), 129.7 (C-4ЈЈ), 130.9 (C-6ЈЈ),
135.5 (C-1Ј), 147.0, 148.7 (C-3Ј, C-4Ј), 157.4 (C-2ЈЈ) ppm. MS (EI,
70 eV): m/z (%) ϭ 300.1 (100) [M]ϩ, 151.1 (81) [veratryl]ϩ, 121.0
(14) [OCH3ϪC6H4ϪCH2]ϩ. C19H24O3 (300.1): calcd. C 75.97, H
8.05; found C 75.72, H 8.05.
General Procedure for the Oxidative Coupling Reaction: A solution
of substrates 4aϪg (5 mmol) in dry CH2Cl2 (20 mL) was placed in
1,2-Dimethoxy-4-[4-(3,5-dimethoxyphenyl)butyl]benzene (4c): Col-
orless crystals (3.09 g, 78%). M.p.: 48Ϫ50 °C (from MeOH). 1H a well-dried flask. After the temperature had been adjusted to 0
3
3
NMR (CDCl3, 400 MHz): δ ϭ 1.64 (quint, JH,H ϭ 3, JH,H
ϭ
°C, MoCl5 (10.5 mmol) and the corresponding additive
(10.5 mmol) were added at once. The reaction mixture was stirred
7 Hz, 4 H, 2-, 3-H), 2.57 (t, 3JH,H ϭ 7 Hz, 4 H, 1-, 4-H), 3.74, 3.75,
3.83, 3.85 (s, 12 H, OCH3), 6.28Ϫ6.33 (m, 3 H, 2ЈЈ-, 4ЈЈ-, 6ЈЈ-H), for a period of 30Ϫ60 min and then it was quenched with saturated
6.68Ϫ6.78 (m, 3 H, 2Ј-, 5Ј-, 6Ј-H) ppm. 13C NMR (CDCl3,
aqueous NaHCO3 (50 mL). The mixture was extracted with EtOAc
100.5 MHz): δ ϭ 30.6 (C-2), 31.1 (C-3), 35.2, 36.0 (C-1, C-4), 55.0, (3 ϫ 100 mL). The combined organic fractions were washed with
55.1, 55.7, 55.8 (OCH3), 97.5 (C-4ЈЈ), 106.4 (C-2ЈЈ, C-6ЈЈ), 111.2, water (100 mL) and brine (100 mL), dried (MgSO4), and then con-
111.8 (C-2Ј, C-5Ј), 120.1 (C-6Ј), 135.1 (C-1Ј), 144.9 (C-1ЈЈ), 147.0, centrated to provided a crude product that was purified either by
3552
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2003, 3549Ϫ3554

Oxidative Coupling Reactions Mediated by MoCl5 Leading to 2,2Ј-Cyclolignans
FULL PAPER
column chromatography (SiO2; cyclohexane/ethyl acetate; 5a, 5e,
5f) or by recrystallization (5g). For systematic numbering of the
[1]
B. Kramer, A. Averhoff, S. R. Waldvogel, Angew. Chem. 2002,
114, 3103Ϫ3104; Angew. Chem. Int. Ed. 2002, 41, 2981Ϫ2982.
S. R. Waldvogel, Synlett 2002, 622Ϫ624.
B. Kramer, R. Fröhlich, K. Bergander, S. R. Waldvogel, Syn-
thesis 2003, 91Ϫ96.
S. R. Waldvogel, E. Aits, C. Holst, R. Fröhlich, Chem. Com-
mun. 2002, 1278Ϫ1279.
S. R. Waldvogel, R. Fröhlich, C. A. Schalley, Angew. Chem.
2000, 112, 2580Ϫ2583; Angew. Chem. Int. Ed. 2000, 39,
2472Ϫ2475.
S. R. Waldvogel, A. R. Wartini, P. H. Rasmussen, J. Rebek, Jr.,
Tetrahedron Lett. 1999, 3515Ϫ3517 and literature cited therein.
M. Arisawa, N. G. Ramesh, M. Nakajima, H. Tohma, Y. Kita,
J. Org. Chem. 2001, 66, 59Ϫ65.
carbon atoms see Scheme 3.
[2]
[3]
2,3-Dimethoxy-5,6,7,8-tetrahydrobenzo[3,4]cycloocta[1,2-f][1,3]-
benzodioxole (5e): Colorless oil (0.60 g, 38%). 1H NMR (CDCl3,
[4]
[5]
3
300 MHz): δ ϭ 1.45 (t, JH,H ϭ 9 Hz, 2 H, 8-H), 1.99Ϫ2.18 (m, 4
H, 6-, 7-H), 2.57Ϫ2.67 (m, 2 H, 5-H), 3.87, 3.92 (s, 6 H, OCH3),
5.94, 5.95 (s, 2 H, CH2), 6.71, 6.72, 6.73, 6.74 (s, 4 H, Harom.) ppm.
13C NMR (CDCl3, 75 MHz): δ ϭ 29.2, 29.3 (C-5, C-8), 32.4, 32.4
(C-6, C-7), 55.9, 56.0 (OCH3), 100.8 (O-CH2-O), 109.0, 109.1 (C-
1, C-12), 112.2, 112.5 (C-4, C-9), 132.5, 135.5 (C-12a, C-12b),
135.0, 136.3 (C-4a, C-8a), 145.3, 146.7, 147.0, 148.5 (C-2, C-3, C-
9, C-10) ppm. MS (EI, 70 eV): m/z (%) ϭ 312.2 (100) [M]ϩ. HRMS:
m/z calcd. for C21H26O5 312.1362; found 312.1343.
[6]
[7]
[8]
T. B. T. Lam, K. Iiyama, B. A. Stone, Phytochemistry 1992,
31, 1179Ϫ1184.
V. D. Parker, A. Ronlan, J. Am. Chem. Soc. 1975, 97,
4714Ϫ4721.
[9]
2,3,9,10-Tetramethoxy-5,6,7,8-tetrahydrodibenzo[a,c]cyclooctene
(5f): Colorless crystals (0.67 g, 41%). M.p. 133Ϫ134 °C (from
´
1
3
[10]
MeOH). H NMR (CDCl3, 400 MHz): δ ϭ 1.47 (t, JH,H ϭ 9 Hz,
A variety of chlorinated products appear after long reaction
times (Ͼ 2 h) or under too harsh conditions. The components
were identified by GC-MS methods.
G. Lessene, K. S. Feldman, in Modern Arene Chemistry (Ed.:
D. Astruc), Wiley-VCH, Weinheim, 2002, pp. 479Ϫ535.
The isolated yields after crystallization (except 5e) are 5a 35%,
5e 38%, 5f 41%, and 5g 62%.
R. F. Childs, D. L. Mulholland, A. Nixon, Can. J. Chem. 1982,
60, 801Ϫ808.
J. F. Deters, P. A. Mccusker, R. C. Pilger, Jr., J. Am. Chem.
Soc. 1968, 90, 4583Ϫ4585.
3
2 H, 5-, 8-H), 1.84 (t, JH,H ϭ 12 Hz, 1 H, 6-H), 1.99Ϫ2.14 (m, 3
3
3
H, 6-, 7-H), 2.63 (dd, JH,H ϭ 8, JH,H ϭ 13 Hz, 1 H, 5-H), 3.08
[11]
[12]
[13]
[14]
[15]
[16]
3
3
(dd, JH,H ϭ 8, JH,H ϭ 13 Hz, 1 H, 8-H), 3.87, 3.89, 3.91, 3.92 (s,
3
12 H, OCH3), 6.74, 6.76 (s, 2 H, 1-, 4-H), 6.83, 6.97 (d, JH,H
ϭ
9 Hz, 2 H, 11-, 12-H) ppm. 13C NMR (CDCl3, 100.5 MHz): δ ϭ
25.8 (C-8), 28.5 (C-5), 29.4 (C-6), 32.6 (C-7), 55.7, 55.9, 56.0 (C-2-,
C-3-, C-10-OCH3), 60.4 (C-9-OCH3), 109.5 (C-11), 112.2, 112.4
(C-1, C-4), 124.0 (C-12), 132.4 (C-8a), 134.4 (C-12a), 135.2 (C-
12b), 137.0 (C-4a), 146.4, 146.6 (C-2, C-3), 148.4 (C-10), 151.9 (C-
9) ppm. 1D NOE (360 MHz, CDCl3): Irradiation at δ ϭ 1.47 (5-
H) ppm, NOE signal at δ ϭ 1.99Ϫ2.14 (6-H), 6.76 (4-H) ppm;
irradiation at δ ϭ 1.47 (8-H) ppm, NOE signal at δ ϭ 3.89 (C-9-
OCH3) ppm; irradiation at δ ϭ 1.84 (6-H) ppm, NOE signal at δ ϭ
3.08 (8-H) ppm; irradiation at δ ϭ 1.99Ϫ2.14 (6-H) ppm, NOE
signal at δ ϭ 1.47 (5-H), 2.63 (5-H) ppm; irradiation at δ ϭ
1.99Ϫ2.14 (7-H) ppm, NOE signal at δ ϭ 3.89 (C-9-OCH3) ppm;
irradiation at δ ϭ 3.08 (8-H) ppm, NOE signal at δ ϭ 1.84 (6-H),
3.89 (C-9-OCH3) ppm. MS (EI, 70 eV): m/z (%) ϭ 328.2 (100)
[M]ϩ. C20H24O4 (328.2): calcd. C 73.15, H 7.37; found C 72.78,
H 7.41.
W. Klemm, H. Steinberg, Z. Anorg. Allgem. Chem. 1936, 227,
193Ϫ213.
S. Kobayashi, T. Busujima, S. Nagayama, Chem. Eur. J. 2000,
6, 3491Ϫ3494.
N. D. Chikanov, Russ. J. Inorg. Chem. 1981, 26, 769.
B. G. Korshunov, E. D. Lapkina, Russ. J. Inorg. Chem. 1963,
8, 1354Ϫ1356.
S. M. Horner, S. Y. Tyree, Jr., Inorg. Chem. 1963, 2, 568Ϫ571.
V. Gutmann, F. Mairinger, Monatsh. Chem. 1958, 89, 724Ϫ730.
D. A. Edwards, G. W. A. Fowles, J. Chem. Soc. 1961, 24Ϫ28.
W. M. Carmichael, D. A. Edwards, R. A. Walton, J. Chem.
Soc., A 1966, 97Ϫ100.
[17]
[18]
[19]
[20]
[21]
[22]
[23]
X-ray crystal structure analysis of 5g: Empirical formula
2,3,9,10,11-Pentamethoxy-5,6,7,8-tetrahydrodibenzo[a,c]cyclooctene
(5g): Colorless crystals (1.11 g, 62%). M.p. 156Ϫ157 °C (from
C21H26O5, M ϭ 358.42, colorless crystal 0.20 ϫ 0.20 ϫ
˚
0.15 mm, a ϭ 17.975(1), b ϭ 6.902(1), c ϭ 30.994(1) A, β ϭ
1
103.04(1)°, V ϭ 3746.1(6) A , ρcalcd. ϭ 1.271 g cmϪ3, µ ϭ 0.90
3
˚
MeOH). H NMR (CDCl3, 300 MHz): δ ϭ 1.37Ϫ1.53 (m, 2 H, 5-,
cmϪ1, empirical absorption correction (0.982 Յ T Յ 0.987),
3
3
8-H), 1.77 (dd, JH,H ϭ 11, JH,H ϭ 13 Hz, 1 H, 6-H), 1.97Ϫ2.04
˚
(m, 2 H, 6-, 7-H), 2.15 (dd, 3JH,H ϭ 11, 3JH,H ϭ 13 Hz, 1 H, 7-H),
Z ϭ 8, monoclinic, space group C2/c (no. 15), λ ϭ 0.7103 A,
3
3
3
T ϭ 198 K, ω and scans, 14391 reflections collected (Ϯh, Ϯk,
2.64 (dd, JH,H ϭ 8, JH,H ϭ 13 Hz, 1 H, 5-H), 2.98 (dd, JH,H
ϭ
Ϯl), [(sinθ)/λ] ϭ 0.66 AϪ1, 4458 independent (Rint ϭ 0.054)
˚
3
8, JH,H ϭ 13 Hz, 1 H, 8-H), 3.87, 3.89 (s, 6 H, C-2-, C-3-OCH3),
3.92, 3.93, 3.94 (s, 9 H, C-9-, C-10-, C-11-OCH3), 6.58 (s, 1 H, 12-
H), 6.75, 6.76 (s, 2 H, 1-, 4-H) ppm. 13C NMR (CDCl3, 75 MHz):
δ ϭ 25.4 (C-8), 28.4 (C-5), 29.2 (C-6), 32.5 (C-7), 55.9, 56.0 (C-2-,
C-3-OCH3), 60.6, 60.7, 60.8 (C-9-, C-10-, C-11-OCH3), 107.9 (C-
12), 112.0, 112.1 (C-1, C-4), 129.0 (C-8a), 132.4 (C-12b), 135.1,
136.0 (C-4a, C-12a), 141.4 (C-10), 146.5, 148.6 (C-2, C-3), 150.8
(C-11), 151.1 (C-9) ppm. MS (EI, 70 eV): m/z (%) ϭ 358.2 (100)
[M]ϩ, 343.2 (14) [M Ϫ CH3]ϩ. C21H26O5 (358.2): calcd. C 70.37,
H 7.31; found C 69.98, H 7.31.
and 2627 observed reflections [I Ն 2σ(I)], 240 refined param-
eters, R ϭ 0.051, wR2 ϭ 0.102, max. residual electron density
0.23 (Ϫ0.2) e·AϪ3, hydrogen atoms calculated and refined as
˚
riding atoms. X-ray crystal structure analysis of 5f: Empirical
formula C20H24O4, M ϭ 328.39, colorless crystal 0.30 ϫ 0.25
˚
ϫ 0.10 mm, a ϭ 8.986(1), b ϭ 9.183(1), c ϭ 10.853(1) A, α ϭ
3
˚
79.12(1), β ϭ 74.66(1), γ ϭ 84.62(1)°, V ϭ 847.2(2) A , ρcalcd. ϭ
1.287 g cmϪ3, µ ϭ 0.89 cmϪ1, empirical absorption correction
¯
(0.974 Յ T Յ 0.991), Z ϭ 2, triclinic, space group P1 (no. 2),
˚
λ ϭ 0.71073 A, T ϭ 198 K, ω and scans, 9075 reflections
collected (Ϯh, Ϯk, Ϯl), [(sinθ)/λ] ϭ 0.66 AϪ1, 3993 indepen-
˚
dent (Rint ϭ 0.041) and 2926 observed reflections [I Ն 2σ(I)],
221 refined parameters, R ϭ 0.047, wR2 ϭ 0.113, max. residual
electron density 0.23 (Ϫ0.28) e·AϪ3, hydrogen atoms calculated
˚
Acknowledgments
and refined as riding atoms. Data sets were collected using a
Nonius KappaCCD diffractometer, equipped with a Nonius
FR591 rotating anode generator. Programs used: data collec-
tion COLLECT (Nonius B. V., 1998), data reduction Denzo-
SMN (see: Z. Otwinowski, W. Minor, Methods Enzymol. 1997,
276, 307Ϫ326), absorption correction SORTAV (R. H. Bless-
This work was supported by the Deutsche Forschungsgemeinschaft
(DFG) and the European Graduate College ‘‘Template-Directed
Chemical Synthesis.’’ The gift of MoCl5 by H. C. Starck (Goslar,
Germany) was very beneficial.
Eur. J. Org. Chem. 2003, 3549Ϫ3554
www.eurjoc.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3553

B. Kramer, R. Fröhlich, S. R. Waldvogel
FULL PAPER
ing, Acta Crystallogr., Sect. A 1995, 51, 33Ϫ37; R. H. Blessing,
J. Appl. Crystallogr. 1997, 30, 421Ϫ426), structure solution
SHELXS-97 (G. M. Sheldrick, Acta Crystallogr. 1990, A46,
467Ϫ473), structure refinement SHELXL-97 (G. M. Sheldrick,
Universität Göttingen, 1997), graphics SCHAKAL (E. Keller,
Universität Freiburg, 1997). CCDC-208900 and -208901 con-
tain the supplementary crystallographic data for this
paper. These data can be obtained free of charge at
www.ccdc.cam.ac.uk/conts/retrieving.html [or from the
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; Fax: (internat.) ϩ 44-1223/336-033;
E-mail: deposit@ccdc.cam.ac.uk].
[24]
R. Dhal, Y. Landais, A. Lebrun, V. Lenain, J.-P. Robin, Tetra-
hedron 1994, 50, 1153Ϫ1164.
Received April 30, 2003
3554
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2003, 3549Ϫ3554