Ion pairing effects on the regioselectivity of arylic versus benzylic C-O bond reductive cleavage: synthetic applications

Article abstract of DOI:10.1016/j.tet.2007.09.010

The regioselectivity of the reductive cleavage of 3,4,5-trimethoxybenzyl methyl ether strongly depends on the alkali metal employed as a reducing agent and solvent effects. Reactions run using Na as a reducing agent led to aromatic C(4)-O bond cleavage, whilst reductions run in the presence of Na/15-crown-5, or using Li as a reducing agent, led to highly regioselective benzylic C-O bond cleavage. This regioselectivity turnaround is discussed in terms of major solvent effects affecting the fragmentation paths of a common reaction intermediate. Synthetic applications of these findings led to the synthesis of biologically active compounds, like 2,5-dialkyl-substituted resorcinols, or 1-(3,4,5-trimethoxyphenyl)-2-arylethanes structurally related to combretastatin.

Full text of DOI:10.1016/j.tet.2007.09.010

Tetrahedron 63 (2007) 11998–12006
Ion pairing effects on the regioselectivity of arylic versus benzylic
C–O bond reductive cleavage: synthetic applications
*
Ugo Azzena, Giovanna Dettori, Ilaria Mascia, Luisa Pisano and Mario Pittalis
Dipartimento di Chimica, Universitaꢀ di Sassari, via Vienna 2, I-07100 Sassari, Italy
Received 15 June 2007; revised 16 August 2007; accepted 6 September 2007
Available online 11 September 2007
Dedicated to Professor Domenico Spinelli on the occasion of his 75th birthday
Abstract—The regioselectivity of the reductive cleavage of 3,4,5-trimethoxybenzyl methyl ether strongly depends on the alkali metal
employed as a reducing agent and solvent effects. Reactions run using Na as a reducing agent led to aromatic C(4)–O bond cleavage, whilst
reductions run in the presence of Na/15-crown-5, or using Li as a reducing agent, led to highly regioselective benzylic C–O bond cleavage.
This regioselectivity turnaround is discussed in terms of major solvent effects affecting the fragmentation paths of a common reaction inter-
mediate. Synthetic applications of these findings led to the synthesis of biologically active compounds, like 2,5-dialkyl-substituted resor-
cinols, or 1-(3,4,5-trimethoxyphenyl)-2-arylethanes structurally related to combretastatin.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
Known results on this topic draw attention to the possibility
of achieving this goal by an appropriate choice of reaction
conditions. Indeed, literature reports show that reductive
cleavage reactions of aromatic12 and benzylic13 ethers pro-
ceed via fragmentation of intermediate radical anions, and
point out the importance of counterion and solvent effects
on the regioselectivity of the cleavage step.12–17 Accord-
ingly, besides developing new synthetic applications of the
reductive metallation procedure, this work is devoted to
highlight the factors governing the regioselectivity of the re-
ductive cleavage reaction under investigation.
The regioselective generation of functionalized organo-
metallic reagents is a topic of current interest in synthetic
organic chemistry.1
Following our interest in the generation of polar organome-
tallic reagents by the reductive cleavage of aromatic2 and
benzylic ethers,3 as well as the employment of the resulting
carbanions in the synthesis of biologically active com-
pounds,4,5 we investigated the effect of different alkali
metals, namely Na and Li, on the reductive metallation of
3,4,5-trimethoxybenzyl methyl ether (1).
2. Results and discussion
Indeed, ether 1 can be considered as a cheap starting material
allowing either the synthesis of 2,5-dialkylresorcinols,
a class of naturally occurring compounds endowed with cy-
totoxic6 and antibiotic7–10 activities, or of 1,2,3-trimethoxy-
arenes, a class of compounds endowed, inter alia, with
interesting anticancer properties.11
2.1. Synthesis of 3,4,5-trimethoxybenzyl methyl ether, 1,
and reductive cleavage reactions
3,4,5-Trimethoxybenzyl methyl ether (1) was prepared in
88% isolated yield by the reaction of 3,4,5-trimethoxybenzyl
alcohol with NaH in dry tetrahydrofuran (THF), followed by
reaction of the resulting alkoxide with CH3I.
The key steps of these transformations rely on the set up of
highly regioselective procedures, allowing the reductive
metallation of ether 1, either at the aromatic C(4)–O bond,
or at the benzylic C–O bond, as depicted in Scheme 1.
Reductive metallations of ether 1 were carried out under Ar,
with an excess of Na or Li metal in dry THF, at temperatures
ranging from rt to ꢀ50 ꢁC. Na chunks and Li wire (0.32 mm)
were freshly cut under dry 2,4,4-trimethylpentane (isooc-
tane) immediately before use. Reductive lithiations were
run in the presence of 10 (mol %) of naphthalene (C10H8).
Selected reactions were quenched with D2O to provide
Keywords: Alkali metals; Metallation; Cleavage reactions; Regioselectivity;
Ionic pairs; Natural products.
* Corresponding author. Tel.: +39 079229549; fax: +39 079229559; e-mail:
ugo@uniss.it
0040–4020/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2007.09.010

U. Azzena et al. / Tetrahedron 63 (2007) 11998–12006
11999
Alk
Alk
1. aromatic
reductive
metalation
1. benzylic
reductive
metalation
H3CO
OCH3
H3CO
OCH3
2. AlkX
2. AlkX
OCH3
OCH3
H3CO
CH2Alk
H3CO
OCH3
OCH3
1. benzylic
reductive
metalation
H3CO
H3CO
1
2. EX
CH2E
Scheme 1. Reductive metallation routes to 2,5-dialkylresorcinols and 1,2,3-trimethoxyarenes.
evidence for the formation of intermediate organometals
(Scheme 2 and Table 1).
Reduction of ether 1 with 1.5 equiv of Na metal at rt for 16 h,
afforded a reaction mixture containing, besides 3,5-dime-
thoxybenzyl methyl ether (2a), minor amounts of 1-(3,5-
dimethoxyphenyl)ethanol (5) (Table 1, Entry 1).
OCH3
H3CO
OCH3
1. M, THF
Formation of this by-product can be rationalized in several
ways, all of which, however, require the intermediate forma-
tion of (at least) one a-methoxy-substituted benzylic carban-
ion, as depicted in Scheme 3. Na-mediated reductive
metallation of starting material leads to the formation of
2,6-dimethoxy-4-methoxymethylphenyl sodium (I), proba-
bly undergoing acid–base equilibration with the parent com-
pound, to afford the a-methoxy-substituted benzylic
carbanion II.
2. H2O or D2O
H3CO
1
H/D
OCH3
OCH3
H3CO
OCH3
H3CO
H3CO
OCH3
CH2H/D
3a
CH3
H3CO
2a
4
Once benzylic organometal II is formed, it undergoes a 1,2-
Wittig rearrangement to the corresponding alkoxide III.18
Scheme 3 further describes a likely transformation of inter-
mediate III into 1-(3,5-dimethoxyphenyl)ethanol (5), via re-
ductive dealkoxylation and protonation. Other possibilities
(not reported in Scheme 3) could involve, e.g., acid–base
equilibration between ether 2a and organometallic interme-
diate I (or IV) leading to a new a-methoxy-substituted ben-
zylic carbanion.
Scheme 2. Reductive cleavage of 3,4,5-trimethoxybenzyl methyl ether (1).
M¼Na or Li.
Depending upon the reaction conditions, we recovered reac-
tion mixtures containing different amounts of 3,5-dimethoxy-
benzyl methyl ether (2a), 3,4,5-trimethoxytoluene (3a) and
3,5-dimethoxytoluene (4). Although reductive metallations
of 1,2,3-trimethoxyarenes are usually accompanied by the
formation of minor amounts (5–15%) of phenolic com-
pounds,2 we did not take into consideration the formation
of these by-products. It is however worth noting that recov-
ered reaction products always account for more than 90% of
the starting material.
Lowering the temperature to 0 ꢁC led to exclusive formation
of ether 2a, although a relatively low amount of deuterium
was incorporated onto the aromatic ring upon D2O quench-
ing (Table 1, Entry 2). After several attempts, we obtained
a better result by running the reaction at ꢀ20 ꢁC in the pres-
ence of 5 equiv of Na metal (Table 1, Entry 3). It is worth
noting that a reductive metallation run in the presence of
a catalytic amount (10 mol %) of naphthalene (C10H8), did
not improve this result, nor affect the regioselectivity of
the reductive cleavage (not reported in Table 1).
Table 1. Reductive cleavage of ether 1
Entry Metal (equiv) T (ꢁC) t (h)
Product distribution (%)a
2a (%D)b 3a (%D)b
4
1
2
3
4
5
6
7
8
Na (1.5)
Na (1.5)
Na (5.0)
Li (2.5)
Li (2.5)
Li (2.5)
Na (5)d
Li (2.5)e
20
0
16
16
14
5
80c
>95 (66)
>95 (78)
—
—
—
—
—
—
At variance with these results, metallations run in the pres-
ence of Li metal (and a catalytic amount of C10H8), led to
preferential or exclusive cleavage of the benzylic carbon–
oxygen bond. Indeed, reactions run at temperatures ranging
from 0 to ꢀ20 ꢁC afforded reaction mixtures containing,
besides 3,4,5-trimethoxytoluene (2a), minor amounts of
3,5-dimethoxytoluene (4), i.e., the product of a double dealk-
oxylation reaction (Table 1, Entries 4 and 5), whilst lower-
ing the reaction temperature to ꢀ50 ꢁC allowed the highly
regioselective generation of 3,4,5-trimethoxybenzyllithium
(Table 1, Entry 6).
ꢀ20
0
—
—
—
—
—
78
85
22
15
ꢀ20
ꢀ50
ꢀ20
0
5
16
16
12
>95 (77)
>95
>95
—
<5
<5
As determined by 1H NMR of crude reaction mixture.
a
As determined by 1H NMR spectroscopy by monitoring the percentage of
deuteration after D2O quenching.
Alcohol 5 (20%) was also detected.
b
c
d
e
In the presence of 1 equiv of 15-crown-5.
In the presence of 1 equiv of 12-crown-4.

12000
U. Azzena et al. / Tetrahedron 63 (2007) 11998–12006
OCH3
OCH3
Na
H3CO
H3CO
OCH3
2 Na
H3CO
H3CO
H
H3CO
OCH3
I
1
H3CO
2a
OCH3
OCH3
Na
OCH3
H3CO
H3CO
OCH3
H3CO
OCH3
2 Na
H3C
ONa
H3C
ONa
H3CO
Na
III
IV
II
quenching
H3CO
OCH3
H3C
OH
5
Scheme 3. Reductive demethoxylation of ether 1, and formation of alcohol 5.
Although the same benzyllithium derivative can be gener-
ated by the reductive lithiation of 3,4,5-trimethoxybenzyl-
bromide,19 it is worth noting that the present methodology
presents two distinct benefits. Indeed, our procedure allows
the generation of stable solutions of this useful organometal,
which can be taken as an advantage from a synthetic point of
view (see below), and avoids formation of the corresponding
dimeric product, i.e., 1,2-di-(3,4,5-trimethoxyphenyl)-
ethane, thus rendering this approach more economical
from an atomic point of view.
To challenge this assumption, ether 1 was reacted with the
alkali metals in the presence of crown ethers, i.e., under con-
ditions favouring the formation of solvent separated ionic
pairs.12
Reduction with Na and 15-crown-5, run at ꢀ20 ꢁC, afforded
3,4,5-trimethoxytoluene (3a), contaminated by a minor
amount (<5%) of compound 4 (Table 1, Entry 7), and an
analogous result was obtained employing Li metal and 12-
crown-4 at 0 ꢁC (Table 1, Entry 8). A comparison with results
obtained in the absence of crown ethers (Table 1, Entries 1–3
for reductions with Na and Table 1, Entries 4–6 for reductions
with Li) clearly supports the hypothesis of a major solvent
effect on the regioselectivity of the cleavage reaction.
To rationalize the above observed regioselectivity, we pro-
pose that aromatic C–O cleavage of ether 1, which takes
place employing Na as a reducing agent, occurs via poorly
solvated or tight ionic pairs, whilst the competitive benzylic
C–O cleavage, favoured by the employment of Li metal at
relatively low temperatures, preferentially occurs via solvent
separated ionic pairs.
Additional evidence was obtained by running a few reactions
on the dimethyl acetal of 3,4,5-trimethoxybenzaldehyde (6),
a structurally related substrate (Scheme 4 and Table 2). In
OCH3
H3CO
OCH3
M, THF
H3CO
OCH3
6
OCH3
OCH3
OCH3
OCH3
H3CO
OCH3
H3CO
OCH3
H3CO
H3CO
H3CO
OCH3
+
+
+
+
CH3
CH3
H3CO
OCH3
OCH3
OCH3
2a
7
3a
4
1
Scheme 4. Reductive cleavage of the dimethyl acetal of 3,4,5-trimethoxybenzaldehyde (6). M¼Na or Li.

U. Azzena et al. / Tetrahedron 63 (2007) 11998–12006
12001
Table 2. Reductive cleavage of acetal 6
an excess of n-hexyl bromide, followed by vigourous stirring
of the reaction mixture during 22 h, and aqueous work up.
According to this procedure, we recovered 1,3-dimethoxy-
2-hexyl-5-methoxymethylbenzene (2b), in satisfactory iso-
lated yield (65%). The employment of stoichiometric
amounts of the electrophile and/or lower reaction times, re-
sulted in significantly lower yields.
Entry Metal (equiv) T (ꢁC) t (h)
Product distribution (%)a
1
2a
3a
4
7
1
2
3
Na (1.5)
Na (5.0)c
Li (2.5)
20
ꢀ20
ꢀ50
18
18
16
—
18
22
—
19
20
—
—
7
—
2
3
>95b
5d
24e
As determined by 1H NMR of crude reaction mixture.
According to Ref. 20.
In the presence of 1 equiv of 15-crown-5.
Compound 6 (56%) was also recovered.
Compound 6 (24%) was also recovered.
a
b
c
d
e
Under similar reaction conditions, quenching the intermedi-
ate organometal with n-BuBr led to the recovery of 1,3-
dimethoxy-2-butyl-5-methoxymethylbenzene (2c), in 70%
isolated yield.
The introduction of an alkyl chain in the benzylic position of
ethers 2b and 2c was accomplished via a highly regioselec-
tive reductive benzylic metallation, run with an excess of Li
metal in the presence of a catalytic amount of C10H8.
Accordingly, reductive lithiation of n-hexyl-derivative 2b,
followed by quenching with H2O, allowed the almost quan-
titative recovery of 1,3-dimethoxy-2-hexyl-5-methylben-
zene (8a). Under similar conditions, quenching of the
reduction mixture with ethyl bromide, allowed the recovery
of the corresponding alkylated derivative 8b (90%, as deter-
mined by 1H NMR), in satisfactory yield.
close analogy with ether 1, reduction of compound 6 with
Na metal in THF affords the dimethyl acetal of 3,5-dime-
thoxybenzaldehyde (7) as the only reaction product (Table
2, Entry 1).20
Further reductions of 6 were run either with Na metal
and 15-crown-5 at ꢀ20 ꢁC (Table 2, Entry 2), or with Li
metal and a catalytic amount of C10H8 at ꢀ50 ꢁC (Table 2,
Entry 3).
In these cases, we recovered relatively complex reaction
mixtures, due to the fact that benzylic dealkoxylation can oc-
cur twice; it is however clear that both reactions led to the
formation of major amounts of products of reductive ben-
zylic C–O cleavages, further highlighting that competition
between aromatic and benzylic delkoxylation is affected
by solvating effects.
Application of a similar procedure to the n-butyl-derivative
2c, followed by quenching with n-BuBr, allowed the recov-
ery of 1,3-dimethoxy-2-butyl-5-pentylbenzene (8c), in satis-
factory yield (82%, as determined by 1H NMR).
Crude dimethyl ethers 8a–c were not isolated, but directly
hydrolyzed to the corresponding resorcinols, by the reac-
tion with BBr3 in CH2Cl2.23 According to this procedure,
resorcinols 9a–c were recovered in satisfactory overall
yields.
2.2. Synthetic applications
2.2.1. Synthesis of 2,5-dialkylresorcinols. We next investi-
gated the reactivity of 2,6-dimethoxy-4-methoxymethylphe-
nylsodium, I, towards alkyl halides,21 with the aim of
developing an efficient approach to the synthesis of several
biologically active 2,5-dialkylresorcinols, namely 2-hexyl-
5-methylresorcinol (9a),22 DB-2073 (9b),8,9 and stemphol
(9c)8,9 (Scheme 5).
2.2.2. Synthesis of combretastatin analogues. Due to the
biological significance of several natural products possess-
ing a 1,2,3-trimethoxybenzene ring,11 we next investigated
the reactivity towards electrophilic reagents of the arylme-
thyllithum obtained by the reductive lithiation of ether 1,
with the aim to synthesize a series of 1-(3,4,5-trimethoxy-
phenyl)-2-arylethanes, structurally related to biologically
active combretastatins.24
OCH3
OCH3
Alk
H3CO
H3CO
OCH3
1. Li, C10H8 (10%)
2. EX
1. Na
Accordingly, a new series of reductive lithiations of ether 1
were carried out with an excess of Li wire and in the pres-
ence of a catalytic amount of C10H8, at ꢀ50 ꢁC, followed
by addition of an electrophilic reagent and, finally, aqueous
work up (Scheme 6 and Table 3).
2. AlkBr
H3CO
H3CO
1
2b, Alk = C6H13, 65%
2c, Alk = C4H9, 70%
Alk
Alk
H3CO
OCH3
HO
OH
BBr3
OCH3
OCH3
OCH3
OCH3
E
H3CO
H3CO
E
1. Li, C10H8 (10%)
2. EX
8a, Alk = C6H13, E = H
8b, Alk = C6H13, E = C2H5
8c, Alk = E = C4H9
9a, Alk = C6H13, E = H, 85%
9b, Alk = C6H13, E = C2H5, 85%
9c, Alk = E = C4H9, 70%
H3CO
E
Scheme 5. Synthesis of 2,5-dialkylresorcinols 9a–c.
1
3b-f
Scheme 6. Synthesis of 1-(3,4,5-trimethoxyphenyl)-2-arylethanes 3b–f.
Compound 3b: E¼PhCH2; 3c: E¼4-(CH3O)C6H4CH2; 3d: E¼PhCHOH;
3e: E¼3-(EOMO)-4-(CH3O)C6H3CHOH; 3f: E¼2-(EOMO)-3-(CH3O)-
C6H3CHOH; EOM¼CH3CH2OCH2O, ethoxymethyl.
Accordingly, the reaction mixture obtained by the reductive
cleavage of ether 1 with Na metal in THF was reacted with

12002
U. Azzena et al. / Tetrahedron 63 (2007) 11998–12006
Table 3. Reductive lithiation of ether 1, and reaction with electrophilesa
70% yield. Under similar conditions, but employing THF/
H2O as a solvent, we obtained isocombretastatin A (10b),
a known inhibitor of the growth of the P388 lymphocytic
leukaemia cell line, in 68% isolated yield.19 Finally, reflux-
ing acetal 3e with 2.5 M HCl in dioxane/H2O afforded trans
alkene 10c (trans-combretastatin A-4) in 82% isolated yield.
Although less cytotoxic than its Z stereoisomer,25 this E iso-
mer of combretastatin A-4 exhibited significant antibiotic
activity.26
Entry EX
Product, E
Yieldb (%)
73
1
2
3
4
5
PhCH2Cl 3b, PhCH2
4-(CH3O)C6H4CH2Cl 3c, 4-(CH3O)C6H4CH2 50
PhCHO
3d, PhCHOH
3e, Ar1CHOH
3f, Ar2CHOH
68
65
61
Ar1CHOc
Ar2CHOd
a
All reactions were run in the presence of 2.5 equiv of Li and 10 mol % of
C10H8.
b
c
d
Yields determined on isolated products.
Ar1¼3-(EOMO)-4-(CH3O)C6H3.
Ar2¼2-(EOMO)-3-(CH3O)C6H3.
3. Conclusions
Our results describe the first example of a switch in regiose-
lectivity for the reductive metallation of an aromatic poly-
ether, allowing the generation of different organometallics
from a common and easily accessible starting material.
Taking advantage of the possibility to generate stable solu-
tions of 3,4,5-trimethoxybenzyllithium, we were able to
trap this intermediate with alkyl halides (PhCH2Cl and 4-
(CH3O)C6H4CH2Cl, Table 3, Entries 1 and 2), affording the
dibenzyl derivatives 3b and 3c, in satisfactory isolated yields.
The key step of these protocols is the reductive metallation
of different ethereal C–O bonds, and the regioselectivity of
this step is efficiently controlled by the choice of reaction
conditions, i.e., by taking advantage of the different behav-
iour of a common intermediate (the p* radical anion) under
different reaction conditions.
It is worth noting that 1-(4-methoxyphenyl)-2-(3,4,5-trime-
thoxyphenyl)ethane (3c), originally developed as a synthetic
analogue of combretastatin A-4, is known to possess cyto-
toxic activity and to inhibit the polymerization of tubulin.25
Good results were obtained trapping the same intermediate
with benzaldehydes (PhCHO, 3-(EOMO)-4-(CH3O)C6H3-
CHO and 2-(EOMO)-3-(CH3O)C6H3CHO, Table 3, Entries
3–5), allowing the synthesis of the corresponding alcohols,
3d–f.
To rationalize this finding it is necessary to analyze the com-
petitive fragmentation processes of the intermediate p-radi-
cal anion, both from a thermodynamic and a kinetic point if
view.
According to the literature, dissociating p-radical anions can
be divided into two different structural groups, depending on
whether the possibility of overlap between the p-network
and the s* of the scissile bond exists (cleavage of a benzylic
C–O bond) or not (cleavage of an aromatic C–O bond).27
Finally, acidic elaboration of acetal 3e allowed the synthesis
of some natural and synthetic combretastatins, depending
upon reaction conditions (Scheme 7).
OCH3
From a thermodynamic point of view, benzylic C–O bond
cleavage (with a BDE likely lower than 83–76 kcal/mol)
represents the favoured reaction path, in comparison with
aromatic C–O bond cleavage (with a BDE of approx. 101–
98 kcal/mol).28,29
H3CO
OCH3
i or ii or iii
OH
OEOM
OCH3
From a kinetic point of view, every fragmentation path re-
quires a certain degree of intramolecular electron-transfer
from the p* radical anion to the scissile bond (s* radical an-
ion), and this electron density redistribution poses intrinsic
barriers to the overall fragmentation process.12,13,27,30–32
However, cleavage of an arylic C–O bond, orthogonal to the
p-network, should involve a large intramolecular p*–s*
electron-transfer, as opposed to benzylic C–O bond fragmen-
tation, which appears as an easy reaction, due to possible
overlapping between the p* and the s* systems.12,27
3e
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
H3CO
H3CO
H3CO
OCH3
OH
or
or
OH
OCH3
10a
OH
OCH3
OH
OCH3
Therefore, it appears that benzylic C–O bond fragmentation
should be the thermodynamically and kinetically favoured
reaction pathway for our intermediate.
10b
10c
Scheme 7. Further elaboration of acetal 3e: (i) 0.6 M HCl in THF/H2O,
0 ꢁC, 2 h, 10a, 70%; (ii) 0.6 M HCl in CH3OH, 0 ꢁC, 2 h, 10b, 67%; (iii)
2.5 M HCl in dioxane/H2O, reflux, 1 h, 10c, 82%.
To rationalize the result obtained with Na metal, and in
agreement with already described ionic pair effects,12,15–17
we propose that a strong interaction between cation and rad-
ical anion, as in a tight ionic pair, strongly affects the elec-
tronic distribution of the intermediate and, by lowering the
energetic barrier relative to the intramolecular p*–s*
Indeed, hydrolysis of the starting material with 0.6 M HCl in
CH3OH at 0 ꢁC during 2 h, afforded the methyl ether 10a in

U. Azzena et al. / Tetrahedron 63 (2007) 11998–12006
12003
electron-transfer, makes the aromatic C–O bond cleavage
a kinetically favoured reaction path.
with reflux condenser and magnetic stirrer, washed with dry
THF (3ꢂ10 mL), and suspended in dry THF (40 mL). The
mixture was chilled to 0 ꢁC and a solution of the appropriate
methoxybenzaldehyde (8.6 g, 41 mmol) dissolved in THF
(20 mL) was added dropwise. The resulting mixture was
stirred for 1.5 h at rt. To this reaction mixture, chilled to
0 ꢁC, a solution of EOMCl (4.6 g, 4.5 mL, 49 mmol), dis-
solved in 15 mL of THF, was added dropwise. After stirring
for 24 h at rt, the mixture was quenched by slow dropwise
addition of H2O (20 mL), and the resulting mixture was
extracted with AcOEt (3ꢂ20 mL). The organic phase was
washed with brine (10 mL), dried (K2CO3) and evaporated.
Crude products were purified and characterized as follows.
Accordingly, our reactions resemble the behaviour observed
in the reductive cleavage of alkyl aryl ethers, where compe-
tition between aromatic C–O bond versus aliphatic C–O
bond cleavage of an intermediate radical anion is driven to-
wards dealkylation (the thermodynamically favoured reac-
tion path) by solvent separated ionic pairs, whilst formation
of less solvated or tight ionic pairs favours the dealkoxylation
reaction (kinetically controlled reaction path).12
Practical application of these findings led to the development
of efficient synthetic approaches to different classes of bio-
logically active carboaromatics, thus disclosing a hitherto
unexplored regioselectivity feature of reductive metallation
reactions.
4.3.1. 3-(Ethoxymethoxy)-4-methoxybenzaldehyde. Puri-
fied by flash chromatography (AcOEt/petroleum ether/
NEt3¼1:1:0.1), pale yellow oil, 53% yield (4.57 g). Found:
C, 62.75; H, 6.78. C11H14O4 requires C, 62.85; H, 6.71; Rf
0.63 (AcOEt/petroleum ether/NEt3¼1:1:0.1;); IR (liquid
4. Experimental section
4.1. General
film): n¼1686 cmꢀ1
;
dH¼1.24 (3H, t, J¼7.2 Hz,
CH3CH2O), 3.78 (2H, q, J¼6.9 Hz, OCH2CH3), 3.97 (3H,
s, OCH3), 5.34 (2H, s, OCH2O), 7.01 (1H, d, J¼8.4 Hz,
ArH), 7.55 (1H, dd, J¼1.8, 8.4 Hz, ArH), 7.69 (1H, d,
J¼1.8 Hz, ArH), 9.86 (1H, s, CHO); dC¼15.0, 56.1, 64.6,
93.9, 110.9, 115.2, 126.5, 130.0, 147.0, 154.9, 190.8.
Boiling and melting points are uncorrected; the air bath tem-
peratures on bulb-to-bulb distillation are given as boiling
points. Starting materials were of the highest commercial
quality and were purified by distillation or recrystallization
immediately prior to use. Commercially available Li wire
(Ø 3.2 mm) was of 99% purity (high Na content), and Na
metal (chunks) was of 99% purity. D2O was of 99.8% isoto-
pic purity. THF was distilled from Na/K alloy under N2 im-
4.3.2. 2-(Ethoxymethoxy)-3-methoxybenzaldehyde. Puri-
fied by flash chromatography (AcOEt/petroleum ether/
NEt3¼3:7:0.1), pale yellow oil, 89% yield (7.67 g). Found:
C, 62.73; H, 6.80. C11H14O4 requires C, 62.85; H, 6.71; Rf
0.60 (AcOEt/petroleum ether/NEt3¼3:7:0.1); IR (liquid
1
mediately prior to use. H NMR spectra were recorded at
film): n¼1692 cmꢀ1
;
dH¼1.22 (3H, t, J¼7.2 Hz,
300 MHz and 13C NMR spectra were recorded at 75 MHz
in CDCl3 with a Varian VXR 300 with SiMe4 as internal
standard. Deuterium incorporation was calculated by moni-
toring the 1H NMR spectra of crude reaction mixtures, and
comparing the integration of the signal corresponding to pro-
tons in the arylmethyl position with that of known signals.
The resonances of the CH2D proton is shifted 0.04 ppm (d)
upfield relative to the resonance of the corresponding CH3
proton; the resonance of the arylmethyl CH2D carbon
appears as a triplet (J¼18 Hz) shifted 0.2 ppm (d) upfield
relative to the corresponding arylmethyl CH3 carbon. IR
spectra were recorded with a Jasco FT/IR-480 Plus. Flash
chromatography was performed on Merck silica gel 60
(40–63 mm), and TLC analyses on Macherey-Nagel silica
gel pre-coated plastic sheets (0.20 mm). Elemental analyses
were performed by the microanalytical laboratory of the
CH3CH2O), 3.81 (2H, q, J¼7.2 Hz, OCH2CH3), 3.89 (3H,
s, OCH3), 5.28 (2H, s, OCH2O), 7.14–7.18 (2H, m,
2ꢂArH), 7.44 (1H, dd, J¼3.6, 6.3 Hz, ArH), 10.46 (1H, s,
CHO); dC¼15.0, 56.10, 65.9, 97.8, 117.7, 119.0, 124.4,
130.4, 149.3, 152.4, 190.5.
4.4. Reduction of 3,4,5-trimethoxybenzyl methyl ether
(1), with Na metal, and reaction with electrophiles.
General procedure
Two to three pieces of freshly cut Na metal (0.82 g, 5 equiv)
wereplacedunder Arin a 100 mL two-necked flask, equipped
with reflux condenser and magnetic stirrer, and suspended in
dry THF (10 mL) at ꢀ20 ꢁC. 3,4,5-Trimethoxybenzyl methyl
ether, 1 (1.50 g, 7.1 mmol), dissolved in dry THF (25 mL),
was added dropwise, each metal piece was scratched with
a spatula, and the reaction mixture was vigourously stirred
at ꢀ20 ꢁC for 14 h.
ꢀ
Dipartimento di Chimica, Universita di Sassari.
4.2. Starting materials
When necessary (Table 1, Entry 7), 15-crown-5 (1.56 g,
7.1 mmol) was dissolved in dry THF and added together
with the substrate.
3,4,5-Trimethoxybenzyl methyl ether, 1, was prepared and
characterized according to a literature procedure.33 3,4,5-
Trimethoxybenzaldehyde dimethyl acetal, 6, was purchased
from Aldrich. Other products were prepared and character-
ized as follows.
D2O quenching was performed by dropwise addition of the
electrophile (1 mL), dissolved in THF (2 mL), to the reduc-
tion mixture chilled at ꢀ20 ꢁC. The cold bath was removed,
the mixture stirred at rt for several minutes, and finally
quenched by slow dropwise addition of H2O (10 mL) (cau-
tion!). The mixture was extracted with Et2O (3ꢂ10 mL), the
organic phases were collected, washed with brine (10 mL),
dried (Na2SO4) and the solvent was evaporated.
4.3. Preparation of EOM protected benzaldehydes.
General procedure
NaH(1.96 g of a 60%dispersion in mineral oil, 49 mmol) was
placed under dry N2 in a 250 mL two-necked flask equipped

12004
U. Azzena et al. / Tetrahedron 63 (2007) 11998–12006
Quenching with alkyl halides was performed by the
dropwise addition of electrophile (5 equiv), dissolved in
THF (10 mL), to the reduction mixture chilled at ꢀ20 ꢁC.
The resulting mixture was vigourously stirred and slowly al-
lowed to reach rt during 22 h, and worked up as described
above.
4.5. Reductive metallation of ethers 1, 2b and 2c, with Li
metal, and reaction with electrophiles. General
procedure
Li (0.21 g, 30 mg atoms, 2.5 equiv) was placed under Ar in
a 100 mL two-necked flask equipped with reflux condenser
and magnetic stirrer, and suspended in dry THF (30 mL). A
catalytic amount of C10H8 (80 mg, 0.6 mmol, 10 mol %)
was added to the suspended metal. Each metal piece was
scratched with a spatula, and the resulting mixture was
stirred at rt until a dark green colour appeared. The mixture
was chilled at the reported temperature (Tables 1 and 2) and
a solution of the appropriate ether (6.0 mmol) dissolved in
THF (20 mL) was added dropwise. When necessary (Table
1, Entry 8), 12-crown-4 (1.06 g, 6.0 mmol) was dissolved
in dry THF and added together with the substrate.
Quenching with ClCOOCH3 (2 equiv) or HCOOCH3
(0.45 equiv) was performed by adding the electrophile drop-
wise, dissolved in THF (10 mL), to the reduction mixture
chilled at ꢀ20 ꢁC. The resulting mixture was vigourously
stirred at the same temperature during 3 h, and worked up
as described above.
Reaction products were purified by flash chromatography.
Compounds 2a,3 and 534 were characterized by comparison
with literature data, other products were characterized as
follows.
Reaction mixtures were stirred for the reported time, and a so-
lution of the appropriate electrophile (6.6 mmol, 1.1 equiv),
dissolved in THF (10 mL), was added dropwise. After stirring
for 20 min, the mixtures were quenched by slow dropwise ad-
dition of a 1:1 H2O/THF mixture (15 mL) (caution!), the cold
bath removed, and the resulting mixtures extracted with Et2O
(3ꢂ10 mL). Organic phases were collected, washed with
brine (10 mL), dried (Na2SO4) and the solvent evaporated.
4.4.1. 1,3-Dimethoxy-2-hexyl-5-methoxymethylbenzene
(2b). Purified by flash chromatography (AcOEt/petroleum
ether¼1:4), colourless oil, 65% yield (1.25 g). Found: C,
71.98; H, 9.96. C16H26O3 requires C, 72.14; H 9.84; Rf
0.54 (AcOEt/petroleum ether¼1:4); IR (liquid film):
n¼1608, 1589 cmꢀ1; dH¼0.88 (3H, t, J¼6.6 Hz, CH2CH3),
1.21–1.48 (8H, m, 4ꢂCH2), 2.60 (2H, t, J¼6.6 Hz,
ArCH2CH2), 3.40 (3H, s, CH3OCH2), 3.81 (6H, s,
2ꢂArOCH3), 4.42 (2H, s, ArCH2O), 6.52 (2H, s, 2ꢂArH);
dC¼14.1, 22.7, 22.8, 29.2, 29.5, 31.8, 55.7, 58.1, 75.2,
103.1, 119.0, 136.7, 158.2.
Compounds 3a and 4 were characterized by comparison
with commercially available samples. Compounds 3b35
and 3c25 were characterized by comparison with literature
data. Compounds 8a–c were not purified, but directly sub-
mitted to the next reaction step. Other compounds were pu-
rified and characterized as follows.
4.4.2. 1,3-Dimethoxy-2-butyl-5-methoxymethylbenzene
(2c). Purified by flash chromatography (AcOEt/petroleum
ether¼1:4), colourless oil, 70% yield (1.18 g). Found: C,
70.49; H, 9.38. C14H22O3 requires C, 70.56; H, 9.30; Rf
0.51 (AcOEt/petroleum ether¼1:4); IR (liquid film):
n¼1695, 1589 cmꢀ1; dH¼0.91 (3H, t, J¼7.2 Hz, CH2CH3),
1.25–1.51 (4H, m, 2ꢂCH2), 2.61 (2H, t, J¼7.2 Hz,
ArCH2CH2), 3.40 (3H, s, CH3OCH2), 3.81 (6H, s,
2ꢂArOCH3), 4.42 (2H, s, ArCH2O), 6.52 (2H, s, 2ꢂArH);
dC¼14.1, 22.6, 22.8, 31.5, 55.7, 58.1, 75.2, 103.1, 118.8,
136.6, 158.2.
4.5.1. 1-Phenyl-2-(3,4,5-trimethoxyphenyl)ethanol (3d).
Purified by flash chromatography (AcOEt/petroleum
ether¼1:1), pale yellow oil, 68% yield (1.18 g). Found: C,
70.76; H, 7.12. C17H20O4 requires C, 70.81; H, 6.99; Rf
0.45 (AcOEt/petroleum ether¼1:1); IR (liquid film):
n¼3464 cmꢀ1; dH¼2.02 (1H, d, J¼3.0 Hz, OH), 2.92 (1H,
dd, J¼8.1, 13.8 Hz, CHaHbAr), 3.00 (1H, dd, J¼4.8,
13.5 Hz, CHbHaAr), 3.81 (6H, s, 2ꢂArOCH3), 3.83 (3H, s,
ArOCH3), 4.87–4.92 (1H, m, ArCHOH), 6.38 (2H, s,
2ꢂArH), 7.30–7.38 (5H, m, 5ꢂArH); dC¼46.4, 56.0, 60.8,
75.1, 106.3, 125.9, 127.6, 128.4, 133.5, 136.5, 143.7, 153.1.
4.4.3. 2,6-Dimethoxy-4-methoxymethylbenzoic acid
methyl ester (2d). Purified by flash chromatography
(AcOEt/petroleum ether¼2:3), colourless oil, 72% yield
(1.23 g). Found: C, 59.82; H, 6.78. C12H16O5 requires C,
59.99; H, 6.71; Rf 0.41 (AcOEt/petroleum ether¼2:3); IR
(liquid film): n¼1735 cmꢀ1; dH¼3.39 (3H, s, CH3OCH2),
3.82 (6H, s, 2ꢂArOCH3), 3.91 (3H, s, COOCH3), 4.44
(2H, s, CH2), 6.54 (2H, s, 2ꢂArH); dC¼52.4, 56.0, 58.2,
74.4, 102.7, 142.1, 157.4, 167.0.
4.5.2. 1-(3-(Ethoxymethoxy)-4-methoxyphenyl)-2-(3,4,5-
trimethoxyphenyl)ethanol (3e). Purified by flash chroma-
tography (AcOEt/petroleum ether/NEt3¼5:5:0.1), pale
yellow oil, 65% yield (1.53 g). Found: C, 64.35; H, 7.21.
C21H28O7 requires C, 64.27; H, 7.19; Rf 0.33 (AcOEt/petro-
leum ether/NEt3¼5:5:0.1); IR (liquid film): n¼3500 cmꢀ1
;
dH¼1.24 (3H, t, J¼7.2 Hz, CH3CH2O), 1.92 (1H, br s,
OH), 2.90–2.95 (2H, m, ArCH2), 3.76–3.86 (11H, m,
3ꢂArOCH3+OCH2CH3), 3.88 (3H, s, ArOCH3), 4.86–4.94
(1H, m, ArCHO), 5.30 (2H, s, OCH2O), 6.41 (s, 2H, ArH),
6.88 (1H, d, J¼8.1 Hz, ArH), 7.00 (1H, dd, J¼2.1, 8.4 Hz,
ArH), 7.21 (1H, d, J¼2.1 Hz, ArH); dC¼15.0, 46.3, 55.9,
56.0, 60.8, 64.3, 74.8, 94.0, 106.2, 111.4, 114.1, 119.7,
133.7, 136.5, 146.5, 149.1, 153.1.
4.4.4. Bis-(2,6-dimethoxy-4-methoxymethylphenyl)me-
thanol (2e). Purified by flash chromatography (AcOEt/pe-
troleum ether¼7:3), white solid, 55% yield (1.53 g), mp
158–160 ꢁC (EtOH). Found: C, 64.18; H, 7.25. C21H28O7 re-
quires C, 64.27; H, 7.19; Rf 0.42 (AcOEt/petroleum
ether¼7:3); IR (Nujol): n¼3564, 3525 cmꢀ1; dH¼3.37
(6H, s, 2ꢂCH3OCH2), 3.76 (12H, s, 4ꢂArOCH3), 4.39
(4H, s, 2ꢂCH2), 5.58 (1H, d, J¼10.2 Hz, CH), 6.62 (1H,
d, J¼10.2 Hz, OH), 6.65 (4H, s, 4ꢂArH); dC¼55.8, 58.0,
64.3, 74.8, 103.6, 119.5, 138.0, 158.2.
4.5.3. 1-(2-(Ethoxymethoxy)-3-methoxyphenyl)-2-(3,4,5-
trimethoxyphenyl)ethanol (3f). Purified by flash chromato-
graphy (AcOEt/petroleum ether/NEt3¼5:5:0.1), pale yellow

U. Azzena et al. / Tetrahedron 63 (2007) 11998–12006
12005
oil, which solidifies upon standing, 61% yield (1.44 g).
Found: C, 64.19; H, 7.11. C21H28O7 requires C, 64.27; H,
7.19; Rf 0.61 (AcOEt/petroleum ether/NEt3¼5:5:0.1); IR
(liquid film): n¼3481 cmꢀ1; dH¼1.24 (3H, t, J¼6.9 Hz,
CH3CH2O), 1.66 (1H, br s, OH), 2.99 (1H, dd, J¼13.8,
8.4 Hz, ArCHaHb), 3.10 (1H, dd, J¼4.8, 13.8 Hz, ArCHbHa),
3.79–3.85 (11H, m, 3ꢂArOCH3+OCH2CH3), 3.86 (3H,
s, ArOCH3), 5.14–5.26 (m, 1H, ArCHO), 6.45 (2H, s,
2ꢂArH), 6.87 (1H, dd, J¼1.8, 8.1 Hz, ArH), 6.99 (1H,
dd, J¼8.1, 1.8 Hz, ArH), 7.08 (1H, t, J¼7.8 Hz, ArH);
dC¼15.1, 44.0, 55.8, 56.0, 60.8, 65.8, 70.4, 97.8, 106.2,
111.4, 118.9, 124.5, 134.5, 136.3, 137.7, 143.4, 151.8, 153.0.
55.9, 56.5, 60.8, 84.6, 106.3, 110.2, 112.8, 118.7, 134.2,
134.8, 145.6, 146.0, 152.7.
4.8. Synthesis of 5-[1-hydroxy-2-(3,4,5-trimethoxyphe-
nyl)ethyl]-2-methoxyphenol (isocombretastatin A, 10b)
Acetal 3e (0.57 g, 1.5 mmol) was added under Ar to a stirred
0.6 M solution of HCl in 5:1 THF/H2O (15 mL), chilled to
0 ꢁC. The mixture was stirred at 0 ꢁC until complete disap-
pearance of the starting material (2–3 h), as determined by
TLC, then worked up as described for 10a. The crude prod-
uct was purified by flash chromatography (AcOEt/petroleum
ether/Et3N¼7:3:0.1; Rf 0.54) to afford 10b as a colourless
solid (0.33 g, 1.0 mmol, 67%), which was characterized by
comparison with literature data.19,37
4.6. Hydrolysis of 1,3-dimethoxy-2,5-dialkylbenzenes
8a–c. General procedure
A solution of the 1,3-dimethoxy substituted derivative
(5.1 mmol) was dissolved in 80 mL of dry CH2Cl2 in
a 250 mL two-necked flask, equipped with reflux condenser
and magnetic stirrer, under dry N2. The mixture was chilled
to ꢀ80 ꢁC, and a 1 M solution of BBr3 in CH2Cl2 (2.2 equiv,
10.4 mL) dissolved in CH2Cl2 (20 mL) was added dropwise.
The reaction was allowed to warm to rt during 14 h. The
mixture was quenched by slow dropwise addition of H2O
(10 mL) (caution!), the organic phase separated and the
aqueous phase extracted with CH2Cl2 (3ꢂ10 mL). Organic
phases were collected, washed with 1 N HCl (2ꢂ10 mL),
H2O (2ꢂ10 mL) and brine (2ꢂ10 mL). The combined or-
ganic extracts were dried (Na2SO4) and the solvent was
evaporated.
4.9. Synthesis of (E)-1-(3-hydroxy-4-methoxyphenyl)-
2-(3,4,5-trimethoxyphenyl)ethene (E-combretastatin
A-4, 10c)
Acetal 3e (0.48 g, 1.2 mmol) was added under Ar to a 2.5 M
solution of HCl in 3:1 dioxane/H2O (12 mL), stirred under
reflux during 1 h, and worked up as described for 10a. The
crude product was purified by flash chromatography
(AcOEt/petroleum ether/Et3N¼7:3:0.1; Rf 0.67) to afford
10c as a colourless solid (0.31 g, 1.0 mmol, 82%), which
was characterized by comparison with literature data.38
Acknowledgements
Crude reaction mixtures were purified by flash chromato-
graphy (AcOEt/petroleum ether¼1:4). Compounds 9a,22
9b,36 and 9c10 were characterized by comparison with liter-
ature data.
ꢀ
Financial support from the Universita di Sassari (Fondo di
Ateneo) is gratefully acknowledged.
References and notes
4.7. Synthesis of 5-[1-methoxy-2-(3,4,5-trimethoxyphe-
nyl)ethyl]-2-methoxyphenol (10a)
1. For a recent collection of papers on this topic, see: Tetrahedron,
Symposium-in-Print no. 111, Najera, C.; Yus, M. Guest Eds.
Tetrahedron 2005, 61, 3139–3450.
2. Azzena, U.; Melloni, G.; Pisano, L. J. Chem. Soc., Perkin
Trans. 1 1995, 261–266 and references cited therein.
3. Azzena, U.; Pilo, L.; Sechi, A. Tetrahedron 1998, 54, 12389–
12398 and references cited therein.
4. Azzena, U.; Idini, M. V.; Pilo, L. Synth. Commun. 2003, 33,
1309–1317 and references cited therein.
5. Azzena, U.; Dettori, G.; Idini, M. V.; Pisano, L.; Sechi, G. Appl.
Organomet. Chem. 2003, 17, 851–855 and references cited
therein.
Acetal 3e (0.57 g, 1.5 mmol) was added under Ar to a stirred
0.6 M solution of HCl in CH3OH (15 mL) [obtained by add-
ing AcCl (0.75 mL) to the CH3OH (15 mL)], chilled to 0 ꢁC.
The mixture was stirred at 0 ꢁC until complete disappear-
ance of the starting material (2–3 h), as determined by
TLC. The mixture was diluted with H2O (20 mL), and the
solvent evaporated under reduced pressure. The resulting
mixture was extracted with AcOEt (4ꢂ10 mL), the organic
phases were collected, washed with brine (1ꢂ20 mL), H2O
(1ꢂ20 mL) dried (Na2SO4) and evaporated. The crude prod-
uct was purified by flash chromatography (AcOEt/petroleum
ether¼7:3; Rf 0.54) to afford 10a as a pale yellow oil (0.37 g,
1.1 mmol, 70%), which was characterized as follows.
6. Arisawa, M.; Ohmura, K.; Kobayashi, A.; Morita, N. Chem.
Pharm. Bull. 1989, 37, 2431–2434.
7. Eisholy, H. N.; Turner, C. E.; Clark, A. M.; Eisholy, M. A.
J. Pharm. Sci. 1982, 71, 1319–1323.
8. Kanda, N.; Ishizaki, N.; Inoue, N.; Oshima, M.; Handa, A.;
Kitahara, T. J. Antibiot. 1975, 28, 935–942.
9. Achenbach, H.; Kohl, W.; Kunze, B. Chem. Ber. 1979, 112,
1841–1848.
10. Marumo, S.; Hattori, H.; Katayama, M. Agric. Biol. Chem.
1985, 49, 1521–1522.
Purified by flash chromatography (AcOEt/petroleum
ether¼7:3). Found: C, 65.58; H, 7.11. C19H24O6 requires
C, 65.50; H, 6.94; Rf 0.67 (AcOEt/petroleum ether¼7:3);
IR (liquid film): n¼3418 cmꢀ1; dH¼2.79 (1H, dd, J¼6.0,
13.8 Hz, ArCHaHb), 3.03 (1H, dd, J¼13.8, 7.5 Hz, ArCHbHa),
3.20 (3H, s, CH3OCH2), 3.78 (6H, s, 2ꢂArOCH3), 3.81
(3H, s, ArOCH3), 3.89 (3H, s, ArOCH3), 4.20 (1H, dd,
J¼6.0, 7.2 Hz, ArCHO), 5.62 (1H, s, OH), 6.31 (2H, s,
2ꢂArH), 6.68 (1H, dd, J¼8.4, 2.1 Hz, ArH), 6.79 (1H, d,
J¼8.1 Hz, ArH), 6.90 (1H, d, J¼1.8 Hz, ArH); dC¼44.9,
11. Srivastava, V.; Negi, A. S.; Kumar, J. K.; Gupta, M. M.;
Khanuja, S. P. S. Bioorg. Med. Chem. 2005, 13, 5892–5908
and references cited therein.

12006
U. Azzena et al. / Tetrahedron 63 (2007) 11998–12006
12. Casado, F.; Pisano, L.; Farriol, M.; Gallardo, I.; Marquet, J.;
Melloni, G. J. Org. Chem. 2000, 65, 322–331 and references
cited therein.
24. Pettit, G. R.; Toki, B. E.; Herald, D. L.; Boyd, M. R.; Hamel, E.;
Pettit, R. K.; Chapuis, J. C. J. Med. Chem. 1999, 42, 1459–1465
and references cited therein.
ꢁ
13. Pisano, L.; Farriol, M.; Asensio, X.; Gallardo, I.; Gonzalez-
25. Cushman, M.; Nagarathnam, D.; Gopal, D.; Chakraborti,
A. K.; Lin, C. M.; Hamel, E. J. Med. Chem. 1991, 34, 2579–
2588 and references cited therein.
Lafont, A.; Lluch, J. M.; Marquet, J. J. Am. Chem. Soc. 2002,
124, 4708–4715 and references cited therein.
14. Maerker, A. Angew. Chem., Int. Ed. Engl. 1987, 26, 972–989
and references cited therein.
26. Pettit, G. R.; Lippert, J. W., III; Herald, D. L.; Hamel, E.; Pettit,
R. K. J. Nat. Prod. 2000, 63, 969–974.
15. Lazana, M. C. R. L. R.; Franco, M. L. T. M. B.; Herold, B. J.
J. Am. Chem. Soc. 1989, 111, 8640–8646.
27. Maslak, P.; Narvaez, J. N.; Kula, J.; Malinski, D. S. J. Org.
Chem. 1990, 55, 4550–4559.
16. Azzena, U.; Denurra, T.; Melloni, G.; Fenude, E.; Rassu, G.
J. Org. Chem. 1992, 57, 1444–1448.
28. Blanksby, S. J.; Ellison, G. B. Acc. Chem. Res. 2003, 36, 255–
263.
17. Azzena, U.; Casado, F.; Fois, P.; Gallardo, I.; Pisano, L.;
Marquet, J.; Melloni, G. J. Chem. Soc., Perkin Trans. 2 1996,
2563–2565.
29. Handbook of Chemistry and Physics, 66th ed.; CRC: Boca
Raton, FL, 1985–1986; F-193.
30. Pierini, A. B.; Duca, J. S., Jr. J. Chem. Soc., Perkin Trans. 2
1995, 1821–1828.
31. (a) Maslak, P.; Guthrie, R. D. J. Am. Chem. Soc. 1986, 108,
2628–2636; (b) Maslak, P.; Guthrie, R. D. J. Am. Chem. Soc.
1986, 108, 2637–2640.
€
18. Schollkopf, U. Angew. Chem., Int. Ed. Engl. 1970, 9, 763–773
and references cited therein.
19. Singh, S. B.; Pettit, G. R. Synth. Commun. 1987, 17, 877–
892.
ꢁ
20. Azzena, U.; Melloni, G.; Piroddi, A. M.; Azara, E.; Contini, S.;
Fenude, E. J. Org. Chem. 1992, 57, 3101–3106.
21. The reactivity of arylsodium I towards carbonyl compounds
was briefly investigated. Quenching the reduction mixture
with ClCOOCH3 (2 equiv, ꢀ20 ꢁC, 3 h) led to the recovery
of 2,6-dimethoxy-4-methoxymethylbenzoic acid methyl ester,
2d, in 72% isolated yield. Under similar conditions, quenching
the organometal with HCOOCH3 (0.45 equiv), led to the for-
mation of bis-(2,6-dimethoxy-4-methoxymethylphenyl)me-
thanol, 2e, in 55% isolated yield. See Section 4.
22. Pohanka, A.; Levenfors, J.; Broberg, A. J. Nat. Prod. 2006, 69,
654–657.
32. Asensio, X.; Gonzalez-Lafont, A.; Marquet, J.; Lluch, J. M.
J. Org. Chem. 2005, 70, 540–548.
33. Hayashi, K.; Tokura, K.; Okabe, K.; Yamamoto, K.; Tawara, K.
Chem. Pharm. Bull. 1982, 30, 2860–2869.
34. Cristol, S. J.; Ali, M. B.; Sankar, I. V. J. Am. Chem. Soc. 1989,
111, 8207–8211.
35. Asakawa, Y.; Hashimoto, T.; Takikawa, K.; Tori, M.; Ogawa, S.
Phytochemistry 1991, 30, 235–251.
36. Sankawa, U.; Shimada, H.; Yamasaki, K. Chem. Pharm. Bull.
1981, 29, 3601–3605.
37. Niveta, J.; Krishnamurty, H. G. Indian J. Chem., Sect. B 2000,
39, 817–821.
23. Achenbach, H.; Kohl, W.; Reichenbach, H. Chem. Ber. 1976,
109, 2490–2502.
38. Orsini, F.; Pellizzoni, F.; Bellini, B.; Migliarini, G. Carbohydr.
Res. 1997, 301, 95–109.