Substituent effects on the reaction mode between 2-hydroxybenzyl alcohol derivatives and MEM chloride: Synthesis and mechanistic aspects of seven- and ten-membered benzo-fused O,O-acetals

Article abstract of DOI:10.1016/j.tet.2004.09.077

The synthesis of (RS)-2- or (RS)-3-methoxy-2,3-dihydro-5H-1,4- benzodioxepins and (RS)-5- or (RS)-3-methoxy-2,3,5,6-tetrahydro-8H-benzo-[1,4,7] -trioxecins has been developed. The mechanism of such a reaction via the boron trifluoride etherate-promoted transformation of 2-(methoxyethoxymethoxy) benzyloxyacetaldehyde dimethyl acetals or 2-(methoxyethoxymethoxymethyl) phenyloxyacetaldehyde dimethyl acetals has been proposed. Transannular versions of the reaction results in the facile ring contraction of 12-membered intermediates to the 10- and to 7-membered benzene-fused O,O-acetals. The characterization of the by-products strongly supports the mechanisms proposed. Graphical Abstract.

Full text of DOI:10.1016/j.tet.2004.09.077

Tetrahedron 60 (2004) 11453–11464
Substituent effects on the reaction mode between 2-hydroxybenzyl
alcohol derivatives and MEM chloride: synthesis and mechanistic
aspects of seven- and ten-membered benzo-fused O,O-acetals
a
Estrella Saniger, Monica Dıaz-Gavilan, Beatriz Delgado, Duane Choquesillo,
a
a
b
´
Josefa M. Gonzalez-Perez, Stefania Aiello, Miguel A. Gallo, Antonio Espinosa
´
´
b
a
a
a
´
´
´
a,
*
´
and Joaquın M. Campos
a
´ ´ ´
Departamento de Quımica Farmaceutica y Organica, Facultad de Farmacia, c/ Campus de Cartuja s/n, 18071 Granada, Spain
b
´
´
Departamento de Quımica Inorganica, Facultad de Farmacia, c/ Campus de Cartuja s/n, 18071 Granada, Spain
Received 23 April 2004; revised 16 September 2004; accepted 17 September 2004
Available online 12 October 2004
Abstract—The synthesis of (RS)-2- or (RS)-3-methoxy-2,3-dihydro-5H-1,4-benzodioxepins and (RS)-5- or (RS)-3-methoxy-2,3,5,6-
tetrahydro-8H-benzo-[1,4,7]-trioxecins has been developed. The mechanism of such a reaction via the boron trifluoride etherate-promoted
transformation of 2-(methoxyethoxymethoxy)benzyloxyacetaldehyde dimethyl acetals or 2-(methoxyethoxymethoxymethyl)phenyloxy-
acetaldehyde dimethyl acetals has been proposed. Transannular versions of the reaction results in the facile ring contraction of 12-membered
intermediates to the 10- and to 7-membered benzene-fused O,O-acetals. The characterization of the by-products strongly supports the
mechanisms proposed.
q 2004 Elsevier Ltd. All rights reserved.
1. Introduction
isomeric seven-membered benzo-fused O,O-acetals, and
isomeric ten-membered benzo-fused analogues.
Selective protection and deprotection of functional groups is
one of the major issues in multistep synthetic strategies of
organic compounds. In particular, hydroxyl groups are
targets for selective protection, because selectively access-
ible OH-groups are often required for the following
reaction. Many OH-protecting groups are known and the
ability to protect a primary hydroxyl group in the presence
of a secondary one was found with a variety of protecting
reagents.1,2 It has lately been shown that hydroxyalkyl
phenols undergo selective protection either at the hydroxyl
or at the phenol group by simply choosing the protecting
reagent under essentially the same reaction conditions.3
A literature survey revealed no reports on the regioselective
protection of 2-hydroxybenzyl alcohol derivatives as a
function of the electronic nature of the substituents at
positions 3 or 5 of the aromatic ring. Accordingly, we
decided to fill this gap in scientific literature and, at the same
time, to use this synthetic tool for the preparation of
We have recently reported the synthesis and biological
activities of 2,3-dihydro-5H-1,4-benzodioxepin derivatives
condensed with the 5-fluorouracil (5-FU) moiety on position
3 (1).4 For these compounds the starting materials were the
2,3-dihydro-5H-1,4-benzodioxepin synthons 2a–c. We
embarked on a programme of synthesis and study of the
biological properties of 2,3-dihydro-5H-1,4-benzodioxepin
fragments that have the pyrimidine base linked in all the
possible sites of the seven-membered ring, and directed our
efforts in a second phase to the preparation of the cyclic
O,O-acetal 3a,d,e, with the acetalic methoxy group on
position 2. The mechanistic aspects of the reaction between
the acyclic O,O-acetals 4a–g or the cyclic ones 2a–c and
5-fluorouracil (5-FU) have been reported.5 In the course of
our present studies, the benzo-fused ten-membered
O,O-acetals 5a,d,e were also obtained. Here we report the
three-step synthesis of 3a,d,e and 5a,d,e, together with their
mechanisms. When the 2-hydroxybenzyl alcohol has a
5-OMe group or a 3-OMe substituent, the final compounds
are the seven-membered O,O-acetals 2b,c, together with 6c.
The importance of the ten-membered O,O-acetals 5a,d,e
and 6c lies in the following: (a) These unreported structures
could be the starting synthons for the preparation of the
Keywords: Acetals; Benzodioxepins; Medium-ring heterocycles;
Mechanisms.
* Corresponding author. Tel.: C34 958 243849; fax: C34 958 243845;
e-mail: jmcampos@ugr.es
0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2004.09.077

11454
E. Saniger et al. / Tetrahedron 60 (2004) 11453–11464
Chart 1.
corresponding ten-membered O,N-acetals that, in a similar
way to what was reported for the fourteen-membered bis(5-
FU O,N-acetals), could exhibit notable biological activities
against breast cancer cells; and (b) their formation sheds
light on the mechanism of reaction in which the neighbour-
ing-group participation plays a pivotal role (Chart 1).
compound 6a was obtained using conditions (a) (See
Section 4).
Both MEM ethers [2-(methoxy-2-ethoxymethoxymethyl)
phenol and 8a] possess similar polarities (very close Rf, 0.3
and 0.2, respectively, using diethyl ether/hexane: 3/1 as
eluant) and spectroscopic properties. Both compounds show
the same molecular-ion peak of MC (calculated for
C11H16O4Na (MCNa)C 235.0946, found 235.0946) in
their high resolution liquid secondary ion mass spectrum
(HR LSIMS) spectra, confirming that both have incorpor-
ated the MEM moiety into their structures. We thought that
in the corresponding 1H NMR spectra the chemical shift of
the –O–CH2–O– group could serve as a probe to decide the
identity of both isomers: in compound 8a such a group
should appear at a lower field (d 5.34 ppm) than in
compound 2-(methoxy-2-ethoxymethoxymethyl)phenol (d
4.85 ppm), due to the electron-withdrawing effect origi-
nated by the phenoxy moiety. Once the structure of 8a had
been demonstrated we decided to extend the reaction
starting with 2-hydroxybenzyl alcohols with different
substituents on the aromatic ring (8d,e). The synthesis of
the cyclic O,O-acetals was carried out in a three-step
process: (a) the formation of MEM ethers 8a,d,e using
MEMCl (1.5 equiv), K2CO3 (1.1 equiv), the salicyl alcohols
(1 equiv) in acetone as solvent at 0 8C, under an inert
atmosphere; (b) preparation of the intermediate acyclic
O,O-acetals 9a,d,e by alkylation of the benzylic hydroxy
group with bromoacetaldehyde dimethyl acetal, using
2. Results and discussion
2.1. Reaction between 2-hydroxybenzyl alcohols 7a,d,e
and 2-methoxyethoxymethyl chloride
Before carrying out the synthesis of 3a,d,e it is necessary to
protect the phenolic hydroxy group of the 2-hydroxybenzyl
alcohol 7a. Among other functionalities, the 2-methoxy-
ethoxylmethyl (MEM) group was developed as a protective
group of alcohols6 and phenols.7 Nevertheless, this
protective group does not present enough selectivity and
also leads to the blocking of the benzylic alcohol.
Accordingly, the protection reaction with MEMCl has
been carried out under several conditions, with the object of
improving its modest selectivity in favour of 8a and to the
detriment of 2-(methoxyethoxymethoxymethyl)phenol, by
using several bases and solvents. Such a study was
performed on 2-hydroxybenzyl alcohol (salicyl alcohol)
7a. We have studied three experimental conditions: (a)
acetone and potassium carbonate; (b) sodium hydride and
tetrahydrofuran (THF); and (c) diisopropylethylamine
(DIPEA) and methylene chloride. The better yield in
sodium hydride as
a
base and anhydrous
Scheme 1. Reagents: (i) K2CO3, anhydrous acetone, MEMCl; (ii) BrCH2CH(OMe)2, NaH, anhydrous DMF; (iii) BF3$OEt2 in anhydrous Et2O.

E. Saniger et al. / Tetrahedron 60 (2004) 11453–11464
11455
dimethylformamide (DMF) as solvent; and (c) the cleavage
of the MEM moiety and subsequent cyclization to yield the
target molecules 3a,d,e. In the original paper, which
introduced the MEM group as a protective group for the
hydroxyl function,6 the advantages of using anhydrous
ZnBr2 or TiCl4 over other Lewis acids were highlighted. We
have reported the BF3$OEt2-mediated seven-membered
cyclization of acyclic O,O-acetals4,8,9 and accordingly, we
supposed that the use of such a catalyst could lead to the
target molecules 3a,d,e in a one-step/pot reaction, as a
consequence of the simultaneous deblocking/cyclization
process. The experimental results confirmed our hypothesis
but, in addition to the expected benzofused seven-
membered O,O-acetals 3a,d,e, the ten-membered O,O-
acetals 5a,d,e were also produced (Scheme 1).
the corresponding values of 9a,d,e cannot be explained by
the field/inductive effects of the aromatic methoxy frag-
ments because the distance between the two centres
involved is very high in both cases. However, such a
chemical shift difference could be justified should the
oxygen atom of the benzylic alcohol be alkylated by the
MEM moiety, instead of the oxygen atom of the phenol
group. Should this be the case, the sequence of reactions
(Scheme 2) would lead to the previously reported seven-
membered O,O-acetals 2b,c (with the acetalic –OMe
fragment in position 3), together with 6c in the case of
starting from 7c. Scheme 2 depicts the synthetic route
followed whose difference with respect to Scheme 1 is
based on the different alkylation site by MEMCl.
Another key point is the chemical shift of the benzylic
carbon atoms of both target molecules 3a,d,e and 2b,c. For
compounds 2b,c, such carbons are located g (an 1,3-
relationship) in relation to the acetalic methoxy groups, their
13C chemical shifts being very sensitive to the steric
compression. As a rule, it is found that the 13C NMR
chemical shifts of carbon atoms in spatially crowded alkyl
groups are more upfield than similar carbon atoms in
unperturbed systems. Therefore, such an effect is negligible
for compounds 3a,d,e because the proximity relationship
between both groups is even higher (delta or an 1,4-
relationship). Table 1 shows the 13C chemical shifts of the
corresponding seven-membered moieties of the cyclic
O,O-acetals.
In order to confirm the structures of the compounds, we
focused our attention on the NMR chemical shift of the
benzylic carbon atoms and found that in the case of 8a,d,e,
the range covers a narrow interval of z1 ppm (in CDCl3): d
61.58 ppm (9a), d 60.62 ppm (9d), and d 60.41 ppm (9e).
2.2. Reaction between 2-hydroxybenzyl alcohols 7b,c and
2-methoxyethoxymethyl chloride
Nevertheless, when we tried to extend this series of
reactions with the aim of obtaining 8b,c, starting from the
salicyl alcohols 7b,c, their 13C NMR chemical behaviour
was not compatible with such structures on the basis of the
chemical shifts of the benzylic carbon atoms, that is, d
66.80 ppm when the benzene ring had a 5-OMe group or d
64.55 ppm when the aromatic substituent was the 3-OMe
moiety. These two low field chemical shifts, in relation to
In spite of the accurate 13C NMR reasoning carried out to
prove the structures of 2b,c, the confirmation of such
compounds needed to be corroborated because this point is
Scheme 2. Reagents: (i) K2CO3, anhydrous acetone, MEMCl; (ii) BrCH2CH(OMe)2, NaH, anhydrous DMF; (iii) BF3$OEt2 in anhydrous Et2O.
Table 1. 13C NMR chemical shiftsa (ppm) for the 2,3-dihydro-5H-1,4-dioxepin moiety in 3a,d,e and 2b,c for CDCl3 solutions
3a
3d
3e
2bb
2cb
C-2
C-3
C-5
C-10
C-11
103.99
74.86
72.87
154.30
133.26
103.99
74.85
72.37
152.85
134.80
104.04
74.85
72.32
153.45
135.27
73.00
101.54
63.23
152.91
131.15
72.37
101.25
62.85
147.90
130.49
a Each reading was quoted to the nearest 0.05 ppm.
b See Ref. 4.

11456
E. Saniger et al. / Tetrahedron 60 (2004) 11453–11464
Figure 1. Molecular structure of (RS)-1-(7-methoxy-2,3-dihydro-5H-1,4-benzodioxepin-3-yl)-5-fluorouracil (ORTEP drawing at 50% probability).
critical for the confirmation of the alkylation site of 7b by
MEMCl. There is always the chance that the structure of 2b
with the acetalic –OMe group at position 3 could have been
mistaken for the corresponding analogue having the acetalic
–OMe group at position 2 (the hypothetical molecule 3b)
because their 1H and 13C NMR data are very close.
Accordingly, we decided to unequivocally elucidate the
structure of the acetal (2b or 3b) by its reaction with
5-fluorouracil, 1,1,1,3,3,3-hexamethyldisilazane (HMDS)
and trimethylchlorosilane (TCS), under acid catalysis
(SnCl4) in acetonitrile during 144 h. Such a process led to
(RS)-1-(7-methoxy-2,3-dihydro-5H-1,4-benzodioxepin-3-
yl)-5-fluorouracil,4 whose structure was unambiguously
determined by X-ray crystallography (Fig. 1). Therefore,
the regioselective protection of the primary hydroxy
group of the corresponding salicyl alcohol was finally
proved by a synthetic method, which made secure our
previous structural assignments.
the breaking of the methoxyethoxymethyl moiety, then the
nucleophilic attack of the phenoxy group to the acetalic
functionality with the concomitant cyclization process
should give rise to the seven-membered acetal 3a,d,e; and
(b) formation of the ten-membered acetal 5a,d,e is not so
obvious: the terminal methyl ether and the internal
methylene-oxy group of the MEM fragment should be
eliminated before or after the corresponding cyclization step
takes place. Such processes are likely to occur through
concerted processes and rearrangements on common
intermediates. It must be emphasized that outside the
protective group arena, MEMCl has been used to alkylate
enolates11 and aryllithium reagents in the presence of
Ph2TlBr.12 MEM ethers have also proved to be a good
one-carbon source for the preparation of isochromans13 and
seven- and eight-membered oxacyclic rings.14
Scheme 3 shows a possible mechanism for the formation of
both cyclic O,O-acetals. First of all, the complexation of the
ethereal oxygen atom of the methoxy group of the MEM
moiety takes place with the concomitant O-5† participation
of the ethereal phenoxy atom and formation of a 1,3-
dioxolane-1-ylium cation. The intermediate 13a,d,e might
undergo s-bond rotation about the CPh–O bond, and then its
highly electrophilic carbon atom of the methylenedioxy
fragment could be attacked by one of the acetalic –OMe
groups. This would give rise to the 12-membered transition
state 14a,d,e, which could suffer a reduction of the ring size
to the 10-membered intermediate 15a,d,e by means of an
intramolecular reaction and the later leaving of the
methoxymethanol fragment. An O-5 participation of the
oxygen atom at position 1 and the acetalic carbon of 15a,d,e
gives rise to a ring contraction leading to 3a,d,e through the
intermediacy of the seven-membered oxonium ion 16a,d,e.
The explanation of the different chemical behaviour (see
Schemes 1 and 2) is very simple: the acidity of phenolic
compounds is modulated by electronic effects. ortho and
para electron-donating groups in relation to the phenol
group decrease acidity, whilst electron-withdrawing groups
at the same position act in the opposite manner. As a result
of both resonance and field/inductive effects, charge
concentration leads to lesser stability of phenoxy anions
and to a decrease in acidity.10 Accordingly, the electronic
properties of the ortho and para substituents to the hydroxy
phenoxy group modifies the selectivity of the alkylation site
by MEMCl.
2.3. Mechanistic aspects of the synthesis of (RS)-2-
methoxy-2,3-dihydro-5H-1,4-benzodioxepins 3a,d,e and
(RS)-5-methoxy-2,3,5,6-tetrahydro-8H-benzo-[1,4,7]-
trioxecins 5a,d,e
On one hand, it could have been supposed that, rather
than the formation of 16a,d,e through the intermediates
12a,d,e–15a,d,e, the synthesis of 3a,d,e could be considered
more directly and simply from the open acetals 9a,d,e by
nucleophilic attack of the phenoxy oxygen to the acetalic
This process is effected by the reaction of 9a,d,e (1 equiv) in
tetrahydrofuran (THF) at 0 8C under an inert atmosphere
with 0.5 equiv of BF3$OEt2. If the structures of the starting
material 9a,d,e and of the final compounds 3a,d,e and 5a,d,e
are compared, one comes to the conclusion that the MEM
moiety of 9a,d,e should suffer two different cleavage
processes from a formal point of view: (a) on one hand,
† When describing nucleophilic participation it is frequently convenient to
use the symbol G-n, where G is the participating group and n the size of
the ring that is formed in the transition state.

E. Saniger et al. / Tetrahedron 60 (2004) 11453–11464
11457
Scheme 3. Reagents: (i) BF3$OEt2, THF; (ii) H2O.
functionality, after complexation by BF3 of one of the
acetalic oxygens. Then the intermediate analogous to
16a,d,e should arise, but in this case substituted on the
oxonium oxygen by a 2-methoxyethoxymethyl group.
Cleavage of this group should also deliver 3a,d,e. Never-
theless, the proof of the presence of the by-product
2-(methoxymethoxy)ethanol15 (See Section 4), formed
through 12a,d,e–15a,d,e, and the absence of methoxy-
ethoxymethanol, arising directly from 9a,d,e, allow us to
settle the proposed mechanism. On the other hand, it has
been checked that the seven-membered rings 3a,d,e (major
products of the rearrangements) do not arise from the ten-
membered rings 5a,d,e, upon treatment of the latter with
boron trifluoride diethyl etherate under the conditions of the
rearrangement.
could explain the course of the cyclization/contraction
reaction. We believe that the mechanism of the transform-
ation 11b/2b is best represented as in Scheme 4. The
aromatic –OMe substituent has an influence on the course of
the reaction: the phenolic oxygen atom (O-1), whose
nucleophilicity may be strongly influenced by the electronic
character of the 4-OMe moiety, should intervene as a
neighbouring group. We have previously reported a similar
feature.4 According to this hypothesis, the intermediate 17
suffers the neighbouring group attack to give the oxyranium
ion 18, much more reactive than its predecessor. After a
s-bond rotation through the C–OC bond of this highly
reactive species, the acetalic-like carbon atom could be
attacked by either of the three oxygen atoms of the adjacent
lateral chain [routes (a), (b), or (c) and (d)]. Through any of
the twelve- or nine-membered intermediates (19 and 20,
respectively), the final destiny is the seven-membered
intermediate 21, which after work up leads to 2b. The
characterization of the by-product 2-(methoxymethoxy)-
ethanol justifies the proposed mechanism (See Section 4).
2.4. Mechanistic aspects on the synthesis of (RS)-3-
methoxy-2,3-dihydro-5H-1,4-benzodioxepins 2b,c and
(RS)-3-methoxy-2,3,5,6-tetrahydro-8H-benzo-[1,4,7]-
trioxecins 6c
The most important feature of this mechanism is the
electrophilic character of the acetalic carbon atom. Never-
theless, the course of the reaction that leads to 2c and 6c is
When the starting materials are 11b and 11c, both the nature
and the yields of the final compounds, are determining
factors to shed light on the two different mechanisms that

11458
E. Saniger et al. / Tetrahedron 60 (2004) 11453–11464
Scheme 4. Reagents: (i) BF3$OEt2, THF; (ii) H2O.
different (Scheme 5). In this case and via a different
mechanism, closely related to the one shown in Scheme 3,
the acetalic –OMe group acts as a nucleophile and the
MEM-derived chain as a good electrophile through the 1,3-
dioxolane-1-ylium cation 23. Again, the proof of the
presence of 2-(methoxymethoxy)ethanol15 and methoxy-
methanol16 strongly supports the mechanism.
(27) is highly unstable due to the closeness of both positive
charges and accordingly very unlikely.
In short, when in the doubly protected salicyl alcohol the
substituent R1, para in relation to the phenolic oxygen atom,
is electronical neutral (H), electron-withdrawing (Cl, Br) or
electron-releasing groups the phenolic O-linked moiety acts
as an electrophile and the alcoholic O-linked fragment acts
as a nucleophile (Schemes 3 and 4). Nevertheless, the
differences in nucleophilicity and electrophilicity of such
groups are so subtle that the presence of an electron-
releasing group ortho in relation to the phenolic O-linked
fragment can invert the reactivity of both lateral chains: that
is to say, the unstability of the intermediate 27 makes the
upper O-phenolic fragment to act as electrophile and,
accordingly the lower alcoholic O-linked moiety to work as
nucleophile (Scheme 5). Such behaviour can be confirmed
after the structural proofs of the by-products methoxy-
methanol and methoxymethoxyethanol.
An important question that needs to be answered is the
following: Why this different behaviour if in 11b and in 11c
the aromatic –OMe groups are para and ortho, respectively,
in relation to the phenolic oxygen atom that carries the
acetaldehyde dimethyl acetal moiety? Although the elec-
tronic effects of the –OMe group in both positions are
composed of field/inductive and resonance effects, the latter
is far more important and, in principle, the mechanisms of
the transformations 11b/2b (Scheme 4) and 11c/2cC6c
(Scheme 5) should have been the same. Should this be the
case, Chart 2 shows the two key intermediates, one of them

E. Saniger et al. / Tetrahedron 60 (2004) 11453–11464
11459
Scheme 5. Reagents: (i) BF3$OEt2, THF; (ii) H2O.
same signals of 2a–c (d 3.56–3.62 ppm). Regarding the 13C
NMR spectra, the acetalic C-2 atoms of 3d,e (d 101.32–
101.40 ppm) are upfield in relation to the same signals of
2a–c (d 103.99–104.04 ppm). This tendency is also
observed with C-3 [3d,e (d 72.43–72.80 ppm) and 2a–c
(dz74.8 ppm)] and C-5 [3d,e (d 62.93–63.05 ppm) and
2a–c (d 72.87–72.32 ppm)].
The 10-membered cycloacetals 5a,d,e show the following
characteristics: (a) The resonance of the acetalic proton H-5
appears between d 4.77–4.81 ppm as dd with coupling
constants of 1.7 and 6.6 Hz; (b) the signals of H-8 resonate
as dd at a d value between 4.56–4.65 ppm with JZ1.2–
2.6 Hz and Jgemz13.3–14.1 Hz; (c) the protons of the
methylene groups H-2, H-3 and H-6 appear as multiplets;
and (c) the singlet of the acetalic –OMe group presents a
chemical shift close in all cases to d 3.4 ppm. The most
interesting aspects of the 13C NMR spectra of the 10-
membered moiety of 5a,d,e are the following: (a) The
acetalic C-5 atom is the most deshielded one and appears at
d 100.47–103.22 ppm, followed by C-6 at d 72.19–
74.92 ppm, C-2 (d 71.85–72.72 ppm), C-8 (d 67.20–
71.58 ppm) and C-3 (d 63.05–68.25 ppm); (b) the acetalic
–OMe moiety appears close to d 59.10 ppm, except in the
case of 6c that resonates at d 56.14 ppm; (c) the chemical
shift values of 6c appear generally slightly upfield with
regard to the rest of the compounds 5a,d,e.
Chart 2.
2.5. Structural characteristics of the seven-membered
O,O-acetals (3a,d,e and 2b,c), and the ten-membered
ones (5a,d,e and 6c)
The structures of all derivatives were ascertained by their
spectroscopic data (1H, 13C NMR, MS) and elemental
analyses. In compounds 3a,d,e the acetalic hydrogen atom
(H-2) appears between d 4.65–4.89 ppm as double of
doublets (dd), whilst the H-3 atoms resonate between d
4.00–4.22 ppm as dd with a Jgem in the range of 12.5–
13.00 Hz. The H-5 atoms are in all cases diastereotopic and
compounds 3d,e show a Jgemz14 Hz. Nevertheless, in the
case of 2a–c their chemical shifts are nearly equivalent
giving an aspect of a doublet (d) with a small coupling
constant (JZ1.50–2.10 Hz). It is noteworthy that the
chemical shifts of the acetalic –OMe group of compounds
3d,e are located upfield (d 3.46–3.47 ppm) in relation to the

11460
E. Saniger et al. / Tetrahedron 60 (2004) 11453–11464
3. Conclusion
(methoxyethoxymethoxymethyl)phenol and 2-(methoxy-
ethoxymethoxy)benzyl alcohol 8a. K2CO3 (5.6 g,
40.8 mmol) were added to a solution of 2-hydroxy-
benzyl alcohol 7a (5.6 g, 45.2 mmol) in anhydrous
acetone (65 mL), and the suspension was left at room
temperature under stirring for 30 min. After this time,
the temperature of the suspension had to fall to 0 8C
before the addition of MEMCl (6.85 mL, 60 mmol) and
the suspension was left under stirring at 0 8C for 6 h.
K2CO3 was filtrated and the resulting solution was
concentrated in vacuo. The resulting residue was
purified by flash chromatography (diethyl ether/hexane:
1/2) and the following two fractions were obtained: the
first one was identified as 2-(methoxyethoxymethoxy-
methyl)phenol (2.58 g, 27% yield) and the second one
identified as 8a (4.31 g, 45% yield). When other conditions
were used, the corresponding yields were the following:
NaH and THF [2-(methoxyethoxymethoxymethyl)phenol
17%, and 8a 17%], and DIPEA and CH2Cl2 [2-(methoxy-
ethoxymethoxymethyl)phenol 50%, and 8a 29%]. When
5-chloro-2-hydroxybenzyl alcohol 7d, and 5-bromo-2-
hydroxybenzyl alcohol 7e were used, the 2-[(2-methoxy-
ethoxymethoxy)]benzyl alcohols 8d,e were the only
compounds obtained, the corresponding 2-(methoxyethoxy-
methoxymethyl)phenols were not detected. When 5-
methoxy-2-hydroxybenzyl alcohol 7b, and 3-methoxy-2-
hydroxybenzyl alcohol 7c were used, the 2-(methoxy-
2-ethoxymethoxymethyl)phenols 10b,c were the only
compounds obtained.
Three main conclusions can be drawn from our results: (1) It
has been found that the substituents on 2-hydroxybenzylic
alcohols affect the protection mode with MEMCl of the two
different hydroxyl groups. The 5-methoxy O-alcoholic-
MEM-protected phenol structure was demonstrated on the
1
following basis: (a) by H and 13C NMR assignments, and
(b) by an X-ray crystallographic determination of (RS)-1-(7-
methoxy-2,3-dihydro-5H-1,4-benzodioxepin-3-yl)-5-
fluorouracil, which unambiguously proved the nature of the
starting material. (2) The mild reaction conditions can be of
particular interest for the preparation of seven- and ten-
membered benzo-fused acetals, which are otherwise
difficult to prepare, although the latter ones are obtained
with low yields. (3) The formation of the ten-membered
O,O-acetals 5a,d,e and 6b,c and characterization of the by-
products throw light on the course of the BF3$OEt2-
promoted reaction on 9a,d,e and 11b,c, respectively.
4. Experimental
All moisture-sensitive reactions were performed in flame-
dried glassware equipped with rubber septa under a positive
pressure of dry argon. Organic extracts were dried over
MgSO4. Thin layer chromatography was performed on
Merck Kieselgel 60 F254, the spots being developed at the
1
UV light. H and 13C NMR spectra were recorded on a
Bruker at 300.13 and 400.1 MHz, and at 75.78 and
100.03 MHz, respectively in CDCl3 solutions. Chemical
shifts were measured in d and referenced to CDCl3
(7.25 ppm for 1H NMR and 77.20 ppm for 13C NMR).
The accurate mass determination was carried out in an
AutoSpec-Q mass spectrometer arranged in an EBE
geometry (Micromass Instruments, Manchester, UK) and
equipped with a FAB (LSIMS) source. The instrument was
operated at 8 kV of accelerating voltage and CsC cations
were used as primary ions. The GC/MS was carried out on a
Platfom II mass spectrometer (Micromass Instruments,
Manchester, UK) coupled with a Carlo Erba gas chromato-
graph (ThermoInstruments, CA, USA) and equipped with
an EI source at 70 eV. The analysis was performed on a HP-
5MS capillary column (30 m!0.25 mm!0.25 mm) in a
splitless mode, inserted directly into the ion source. The
temperature programme was the following: 60 8C,
10 8C/min to 300 8C, then isothermal for 10 min. The
carrier gas was helium with a flow rate of 1 mL/min.
Solvents were obtained dry as follows: tetrahydrofuran
(THF) was distilled from benzophenone ketyl, CH2Cl2 was
refluxed over, and distilled from CaH2 and then stored over
4.1.1. 2-(Methoxyethoxymethoxymethyl)phenol. Com-
pond 7a was used as starting material. Yield: 27%. Rf
(diethyl ether/hexane: 3/1): 0.3. 1H NMR (300 MHz) d 7.22
(dt, 1H, H-5; JZ1.7, 7.7 Hz); 7.06 (dd, 1H, H-6; JZ1.5,
7.4 Hz); 6.87 (m, 2H, H-3 and H-4); 4.85 (s, 2H, OCH2O);
4.76 (s, 2H, PhCH2O); 3.65 (m, 4H, OCH2CH2O); 3.41
(s, 3H, OMe). HR LSIMS calcd for C11H16O4Na (MCNa)C
235.0946, found 235.0946. Anal. for C11H16O4: Calcd: C
62.25; H 7.60. Found: C 62.48; H 8.00.
4.1.2. 2-(Methoxyethoxymethoxy)benzyl alcohol 8a.
Yield: 45%. Rf (diethyl ether/hexane: 3/1): 0.2. H NMR
1
(300 MHz,) d 7.33 (dd, 1H, H-3 or H-6, JZ1.8, 7.4 Hz);
7.26 (dt, 1H, H-5 or H-4, JZ1.8, 7.9 Hz); 7.14 (dd, 1H, H-6
or H-3, JZ1.0, 7.9 Hz); 7.02 (dt, 1H, H-4 or H-5, JZ1.0,
7.4 Hz); 5.34 (s, 2H, OCH2O); 4.72 (s, 2H, CH2OH); 3.87
(m, 2H,0AA0 part of the ethylenedioxy fragment); 3.56 (m,
2H, BB part of the ethylenedioxy fragment); 3.37 (s, 3H,
OMe). 13C NMR (75 MHz) d 155.07 (C-2); 130.00 (C-1);
128.98, 128.93 (C-6 and C-4); 121.96 (C-5); 114.12 (C-3);
93.57 (OCH2O); 71.62 (CH2OMe); 68.04 (CH2CH2OMe);
61.58 (CH2OH); 59.04 (OMe). HR LSIMS calcd for
C11H16O4Na (MCNa)C 235.0946, found 235.0946. Anal.
for C11H16O4: Calcd: C 62.25; H 7.60. Found: C 62.48; H
8.00.
˚
molecular sieves (3 A), CH3OH from Mg. 2-Hydroxybenzyl
alcohols were kept at 40 8C and 0.1 mm Hg for 48 h.
BF3$OEt2 was distilled prior to use, in an all-glass apparatus
with calcium hydroxide to remove volatile acids and to
reduce bumping.
4.1.3. [5-Chloro-2-(methoxyethoxymethoxy)]benzyl
alcohol 8d. Compond 7d was used as starting material.
4.1. Reaction between 3- or 5-substituted-2-hydroxy-
benzyl alcohols and 2-methoxyethoxymethyl chloride
(MEMCl)
1
Yield: 38%. Rf (diethyl ether/hexane: 4/1): 0.3. H NMR
(300 MHz) d 7.28 (d, 1H, H-6, JZ2.6 Hz); 7.15 (dd, 1H,
H-4, JZ2.6, 8.7 Hz); 7.01 (d, 1H, H-3, JZ8.7 Hz); 5.24
(s, 2H, OCH2O); 4.60 (d, 2H, CH2OH, JZ6.1 Hz); 3.77 (m,
2H, AA0 part of the ethylenedioxy fragment); 3.50 (m, 2H,
The general procedure is exemplified with the case
of 2-hydroxybenzyl alcohol. Synthesis of 2-

E. Saniger et al. / Tetrahedron 60 (2004) 11453–11464
11461
BB0 part of the ethylenedioxy fragment); 3.31 (s, 3H, OMe).
13C NMR (75 MHz) d 153.29 (C-2); 131.85 (C-5); 128.36
(C-4); 128.32 (C-6); 126.86 (C-1); 115.33 (C-3); 93.66
(OCH2O); 71.56 (CH2OMe); 68.06 (CH2CH2OMe); 60.62
(CH2OH); 59.02 (OMe). HR LSIMS calcd for C11H15O4-
NaCl (MCNa)C 269.0556, found 269.0553. Anal. for
C11H15O4Cl: Calcd: C 53.56; H 6.13. Found: C 53.85; H
5.83.
(12.5 mL) while cooling in an ice bath and the resulting
mixture was left under stirring at room temperature for 2 h.
After this time, bromoacetaldehyde dimenthyl acetal
(1.1 mL, 9.4 mmol) was added while cooling with ice, and
the resulting mixture was left under stirring at room
temperature at 50 8C for 5 h. The reaction mixture was
distilled under diminished pressure and the resulting residue
was diluted with water, the pH was adjusted to 2–3 by
adding an aqueous solution of HCl 2N and was then
extracted (EtOAc). The extract was washed with brine and
dried with Na2SO4, filtrated, and concentrated in vacuo. The
resulting residue was purified by flash chromatography
(diethyl ether/hexane: 1/1.5) and 9a was obtained (0.5 g,
35%). Rf (diethyl ether/hexane: 3/1): 0.38. 1H NMR
(300 MHz) d 7.41 (dd, 1H, HAr; JZ1.6, 7.4 Hz); 7.25 (dt,
1H, HAr; JZ1.6, 7.4 Hz); 7.15 (d, 1H, HAr; JZ1.0, 7.4 Hz);
7.02 (dt, 1H, HAr; JZ1.0, 7.4 Hz); 5.41 (s, 2H, OCH2O);
4.65 (s, 2H, C-1-CH2O); 4.57 (t, CH(OMe)2); JZ5.2 Hz);
3.86 (m, 2H, AA0 part of the ethylenedioxy fragment); 3.56
(m, 4H, BB0 part of the ethylenedioxy fragment and
O–CH2–CH); 3.41 (s, 6H, (OMe)2); 3.39 (s, 3H, OMe).
13C NMR (75 MHz) d 154.88 (C-2); 129.18, 128.84
(C-6 and C-4); 127.00 (C-1); 121.73 (C-5); 114.04 (C-3);
102.74 (CH(OMe)2); 95.46 (OCH2O); 71.61, 69.93, 68.28,
67.71 (CH2OMe), (CH2CH2OMe), (C-1-CH2O), (O–CH2–
CH); 59.05 (OMe); 53.85 (OMe)2. HR LSIMS calcd for
C15H23O6Na (MCNaK1)C 299.1494, found 299.1494.
Anal. for C15H24O6: Calcd: C 59.98; H 8.05. Found: C
59.95; H 8.23.
4.1.4. [5-Bromo-2-(methoxyethoxymethoxy)]benzyl
alcohol 8e. Compond 7e was used as starting material.
1
Yield: 47%. Rf (diethyl ether/hexane: 2/1): 0.26. H NMR
(400 MHz) d 7.42 (d, 1H, H-6, JZ2.5 Hz); 7.28 (dd, 1H,
H-4, JZ2.5, 8.7 Hz); 6.95 (d, 1H, H-3, JZ8.7 Hz); 5.23 (s,
2H, OCH2O); 4.59 (d, 1H, 2H, CH2OH, JZ3.5 Hz); 3.76
(m, 2H,0AA0 part of the ethylenedioxy fragment); 3.49 (m,
2H, BB part of the ethylenedioxy fragment); 3.30 (s, 3H,
OMe). 13C NMR (100 MHz) d 153.76 (C-2); 132.36 (C-1);
131.23, 131.16 (C-4, C-6); 115.77 (C-3); 114.30 (C-5);
93.61 (OCH2O); 71.55 (CH2OMe); 68.05 (CH2CH2OMe);
60.41 (CH2OH); 58.95 (OMe). HR LSIMS calcd for
C11H15O4NaBr (MCNa)C 313.0051, found 313.0049.
Anal. for C11H15O4Br: Calcd: C 45.38; H 5.19. Found: C
44.99; H 4.84.
4.1.5. [2-(Methoxyethoxymethoxymethyl)]-4-methoxy-
phenol 10b. Compond 7b17 was used as starting material.
Yield: 57%. Rf (diethyl ether/hexane: 2/1): 0.4. H NMR
1
(300 MHz) d 6.75 (d, 1H, H-3, JZ7.5 Hz); 6.70 (d, 1H, H-6,
JZ3.0 Hz); 6.68 (dd, 1H, H-4, JZ3.0, 7.5 Hz); 4.75 (s, 2H,
OCH2O); 4.65 (s, 2H, CH2OH); 3.69 (s, 3H, C-5-OMe);
3.69 (m, 2H, AA0 part of the ethylenedioxy fragment); 3.52
(m, 2H, BB0 part of the ethylenedioxy fragment); 3.35 (s,
3H, OMe). 13C NMR (75 MHz) d 153.02 (C-5); 149.28
(C-2); 123.32 (C-1); 117.01 (C-3); 114.70, 114.51 (C-4,
C-6); 94.44 (OCH2O); 71.60 (CH2OCH3); 67.13 (CH2CH2-
OMe); 66.80 (CH2OH); 58.85 (OMe); 55.61 (C-5-OMe).
HR LSIMS calcd for C12H18O5 (MCNa)C 265.1051, found
265.1050. Anal. for C12H18O5: Calcd: C 59.49; H 7.49.
Found: C 59.76; H 7.22.
4.2.1. [5-Chloro-2-(methoxyethoxymethoxy)benzyloxy-
acetaldehyde dimethyl acetal 9d. Compond 8d was used
as starting material. Yield: 39%. Rf (ethyl acetate/hexane:
1
4/1): 0.54. H NMR (400 MHz) d 7.37 (d, 1H, H-6, JZ
2.6 Hz); 7.16 (dd, 1H, H-4, JZ2.6, 8.7 Hz); 7.06 (d, 1H,
H-3, JZ8.76 Hz); 5.25 (s, 2H, OCH2O); 4.57 (s, 2H, C-1-
CH2O); 4.55 (t, 1H, H-10, JZ5.2 Hz); 3.79 (m, 2H, H-20);
3.54 (m, 4H, AA0BB0 system); 3.40 (s, 3H, OMe); 3.36 (s,
6H, (OMe)2). 13C NMR (100 MHz) d 153.11 (C-2);
129.08 (C-5); 128.47 (C-4); 128.22 (C-6); 126.82 (C-1);
115.25 (C-3); 102.74 (CH(OMe)2); 93.60 (OCH2O);
71.53, 70.30, 67.79, 67.71 (O–CH2–CH) (CH2OMe),
(C-1-CH2O), (CH2CH2OMe); 58.99 (OMe); 53.92
(OMe)2. HR LSIMS calcd for C15H23O6NaCl (MCNa)C
357.1080, found 357.1078. Anal. for C15H23O6Cl: Calcd: C
53.81; H 6.92. Found: C 53.48; H 7.23.
4.1.6. [2-(Methoxyethoxymethoxymethyl)]-6-methoxy-
phenol 10c. Compond 7c was used as starting material.
1
Yield: 47%. Rf (diethyl ether/hexane: 4/1): 0.42. H NMR
(400 MHz) d 6.9 (dd, 1H, H-5, JZ4.3, 5.2 Hz); 6.25 (d, 2H,
H-4, H-6, JZ5.2 Hz); 6.17 (s, 1H, OH); 4.81 (s, 2H,
OCH2O); 4.70 (s, 2H, CH2OH); 3.85 (s, 3H, C-3-OMe);
3.75 (m, 2H, AA0 part of the ethylenedioxy fragment); 3.56
(m, 2H, BB0 part of the ethylenedioxy fragment); 3.39 (s,
3H, OMe). 13C NMR (100 MHz) d 146.66 (C-3); 144.02
(C-2); 123.48 (C-1); 121.73 (C-5); 119.36 (C-6); 110.47
(C-4); 94.76 (OCH2O); 71.70 (CH2OMe); 66.78 (CH2-
CH2OMe); 64.55 (CH2OH); 58.85 (CH2OMe); 55.94 (C-3-
OMe). HR LSIMS calcd for C12H18O5Na (MCNa)C
265.1051, found 265.1050. Anal. for C12H18O5: Calcd: C
59.49; H 7.49. Found: C 59.22; H 7.76.
4.2.2. [5-Bromo-2-(methoxyethoxymethoxy)benzyloxy-
acetaldehyde dimethyl acetal 9e. Compond 8e was used
as starting material. Yield: 42%. Rf (ethyl acetate/hexane:
1
4/1): 0.38. H NMR (400 MHz) d 7.47 (d, 1H, H-6, JZ
2.5 Hz); 7.27 (dd, 1H, H-4, JZ2.5, 8.7 Hz); 6.97 (d, 1H,
H-3, JZ8.7 Hz); 5.21 (s, 2H, OCH2O); 4.53 (s, 2H, C-1-
CH2O); 4.51 (t, 1H, H-10, JZ5.2 Hz); 3.75 (m, 2H, H-20);
3.50 (m, 4H, AA0BB0 system); 3.35 (s, 3H, OMe); 3.31 (s,
6H, (OMe)2). 13C NMR (100 MHz) d 153.61 (C-2); 131.34
(C-4);131.17(C-6);129.51(C-1);115.67(C-3); 114.22(C-5);
102.73 (CH(OMe)2); 93.51 (OCH2O); 71.50, 70.31, 67.78,
67.67 (O–CH2–CH), (CH2OMe), (C-1-CH2O), (CH2CH2-
OMe); 58.94 (OMe); 53.88 (OMe)2. HR LSIMS calcd for
C15H23O6NaBr (MCNa)C 401.0575 C15H23O6NaBr (MC
Na)C 401.0575, found 401.0583. Anal. for C15H23O6Br:
Calcd: C 47.51; H 6.11. Found: C 47.9; H 5.76.
4.2. Synthesis of 2-(methoxyethoxymethoxy)benzyloxy-
acetaldehyde dimethyl acetals 9a,d,e
The general procedure is exemplified with the case of 9a: 8a
(1 g, 4.7 mmol) was added to a suspension of NaH (0.42 g of
a 80% dispersion in mineral oil) in anhydrous DMF

11462
E. Saniger et al. / Tetrahedron 60 (2004) 11453–11464
4.3. Synthesis of (RS)-2-methoxy-2,3-dihydro-5H-1,4-
benzodioxepins 2a,d,e and (RS)-5-methoxy-2,3,5,6-
tetrahydro-8H-benzo-[1,4,7]-trioxecin 5a,d,e
HR LSIMS calcd for C10H11O3NaCl (MCNa)C 237.0294,
found 237.0299. Anal. for C10H11O3Cl: Calcd: C 55.96; H
5.17. Found: C 56.22; H 4.87.
The general procedure is exemplified with the case of 2a
and 5a: 9a (1.5 g, 4.9 mmol) was dissolved in anhydrous
THF (15 mL), and BF3$OEt2 (0.3 mL) was added at 0 8C
and the mixture was kept at this temperature under stirring
for 24 h. After this time the solution was washed with an
aqueous solution of K2CO3 (10%) and extracted with
CH2Cl2. The organic layer was dried (Na2SO4), filtered, and
concentrated. The residue was purified by flash chromato-
graphy with diethyl ether/hexane (1/3) yielding 0.36 g of 3a
(41%), and 0.12 g of 5a (6%). After concentrating the
organic layer, the residue was subjected to a simple
combination of directly-coupled gas chromatography-mass
spectrometry (GC-MS): tR of methoxymethanol: 4.14 min;
tR of methoxymethoxyethanol: 6.76 min. The identity of
methoxymethanol14 and methoxymethoxyethanol15 were
confirmed by comparison of their retention times with pure
authentic samples.
Compound 5d. Yield: 22%. Rf (ethyl acetate/hexane: 4/1):
0.6. H NMR (400 MHz) d 7.23 (dd, 1H, H-11, JZ2.5,
1
8.4 Hz); 7.15 (d, 1H, H-9, JZ2.5 Hz); 7.05 (d, 1H, H-12,
JZ8.4 Hz); 4.8 (dd, 1H, H-5, JZ1.6, 6.6 Hz); 4.59 (dd, 2H,
H-8, JZ1.2, 13.6 Hz); 4.11 (m, 2H, H-6, H-2); 3.8 (m, 2H,
H-6, H-2); 3.6 (m, 2H, H-3, H-3); 3.39 (s, 3H, OMe). 13C
NMR (100 MHz) d 153.00 (C-13); 134.97 (C-10); 129.22
(C-11); 128.92 (C-9); 128.85 (C-14); 122.87 (C-12); 103.18
(C-5); 74.91 (C-6); 72.22 (C-2); 71.54 (C-8); 68.20 (C-3);
59.15 (OMe). HR LSIMS calcd for C12H15O4NaCl (MC
Na)C 281.0556, found 281.0563. Anal. for C12H15O4Cl:
Calcd: C 55.71; H 5.84. Found: C 55.83; H 6.03.
4.3.2. (RS)-7-Bromo-2-methoxy-2,3-dihydro-5H-1,4-
benzodioxepin 3e and (RS)-10-bromo-5-methoxy-
2,3,5,6-tetrahydro-8H-benzo-[1,4,7]-trioxecin 5e. Com-
pound 9e was used as a starting material. 3e: Yield: 27%. Rf
(diethyl ether/hexane: 2/1): 0.6. 1H NMR (400 MHz) d 7.35
(dd, 1H, H-8, JZ2.4, 8.4 Hz); 7.27 (d, 1H, H-6, JZ2.4 Hz);
6.99 (d, 1H, H-9, JZ8.4 Hz); 4.65 (dd, 1H, H-2, JZ1.6,
6.1 Hz); 4.57 (d, 2H, H-5, JZ2.1 Hz); 4.00 (dd, 1H, H-3,
JZ1.6, 12.6 Hz); 3.76 (dd, 1H, H-3, JZ6.1, 12.6 Hz); 3.56
(s, 3H, OMe). 13C NMR (100 MHz) d 153.45 (C-10); 135.27
(C-11); 132.25 (C-8); 131.76 (C-6); 123.32 (C-9); 116.46
(C-7); 104.04 (C-2); 74.85 (C-3); 72.32 (C-5); 56.63 (OMe).
HR LSIMS calcd for C10H12O3Br (MCC1) 258.9969,
found 258.9976. Anal. for C10H11O3Br: Calcd: C 46.36; H
4.28. Found: C 45.97; H 4.11.
Compound 3a. Rf (diethyl ether/hexane: 2/1): 0.61. 1H NMR
(300 MHz) d 7.31 (dt, 1H, HAr, JZ1.7, 7.7 Hz); 7.20 (dd,
1H, HAr, JZ1.7, 7.7 Hz); 7.17 (dd, 1H, HAr, JZ1.3,
7.7 Hz); 7.09 (dt, 1H, HAr, JZ1.3, 7.7 Hz); 4.70 (dd, 1H,
H-2, JZ1.6, 6.3 Hz); 4.68 (d, 2H, H-5, 1.6); 4.07 (dd,
1H, H-3, JZ1.6, 12.5 Hz); 3.81 (dd, 1H, H-3, JZ6.3,
12.5 Hz); 3.62 (s, 3H, OMe). 13C NMR (75 MHz) d
154.30 (C-10); 133.26 (C-11); 129.49, 129.04 (C-6 and
C-8); 124.00 (C-9); 121.48 (C-7); 103.99 (C-2); 74.86
(C-3); 72.87 (C-5); 56.50 (OMe). HR LSIMS calcd for
C10H12O3Na (MCNa)C 203.0694, found 203.0683.
Anal. for C10H12O3: Calcd: C 66.65; H 6.71. Found: C
66.19; H 6.85.
Compound 5e. Yield: 18%. Rf (diethyl ether/hexane: 2/1):
1
0.46. H NMR (400 MHz) d 7.34 (dd, 1H, H-11, JZ2.4,
8.4 Hz); 7.26 (d, 1H, H-9, JZ2.4 Hz); 6.97 (d, 1H, H-12,
JZ8.4 Hz); 4.77 (dd, 1H, H-5, JZ1.7, 6.6 Hz); 4.56 (dd,
2H, H-8, JZ1.5, 14.1 Hz); 4.07 (m, 2H, H-6, H-2); 3.76 (m,
2H, H-6, H-2); 3.56 (m, 2H, H-3, H-3); 3.35 (s, 3H, OMe).
13C NMR (100 MHz) d 153.60 (C-13); 135.45 (C-10);
132.28 (C-11); 131.82 (C-9); 123.35 (C-12); 116.53 (C-14);
103.22 (C-5); 74.92 (C-6); 72.19 (C-2); 71.61 (C-8); 68.25
(C-3); 59.17 (OMe). HR LSIMS calcd for C12H15O4NaBr
(MCNa)C 325.0051, found 325.0053. Anal. for
C12H15O4Br: Calcd: C 47.54; H 4.99. Found: C 47.81; H
5.13.
Compound 5a. Rf (diethyl ether/hexane: 4/1): 0.39. 1H NMR
(400 MHz) d 7.29 (dt, 1H, H-11 or H-10, JZ1.7, 7.8 Hz);
7.17 (dd, 1H, H-9 or H-12, JZ1.7, 7.4 Hz); 7.13 (dd, 1H,
H-12 or H-9, JZ1.1, 7.8 Hz); 7.07 (dt, 1H, H-10 or
H-11, JZ1.1, 7.4 Hz); 4.81 (dd, 1H, H-5, JZ1.7, 6.8 Hz);
4.65 (dd, 2H, H-8, JZ2.6, 13.3 Hz); 4.14 (m, 2H, H-6, H-2);
3.81 (m, 2H, H-6, H-2); 3.61 (m, 2H, H-3, H-3); 3.39 (s, 3H,
OMe). 13C NMR (100 MHz) d 154.44 (C-13); 133.43
(C-14); 129.52 (C-11); 129.05 (C-9); 124.05 (C-10); 121.40
(C-12); 103.16 (C-5); 74.91 (C-6); 72.72 (C-2); 71.58 (C-8);
68.08 (C-3); 59.12 (OMe). HR LSIMS calcd for
C12H16O4Na (MCNa)C 247.0946, found 247.0945. Anal.
for C12H16O4$0.1H2O: Calcd: C 64.27; H 7.19. Found: C
64.00; H 7.20.
4.4. Synthesis of 2-(methoxyethoxymethoxymethyl)-
phenyloxyacetaldehyde dimethyl acetals 11b and 11c
The procedure was similar to the one used in 4.3, but
changing 9a by 10b, and 9a by 10c.
4.3.1. (RS)-7-Chloro-2-methoxy-2,3-dihydro-5H-1,4-
benzodioxepin 3d and (RS)-10-chloro-5-methoxy-
2,3,5,6-tetrahydro-8H-benzo-[1,4,7]-trioxecin 5d. Com-
pound 9d was used as a starting material. 3d: Yield: 45%. Rf
(ethyl acetate/hexane: 4/1): 0.7. 1H NMR (400 MHz) d 7.23
(dd, 1H, H-8, JZ2.5, 8.4 Hz); 7.15 (d, 1H, H-6, JZ2.5 Hz);
7.07 (d, 1H, H-9, JZ8.4 Hz); 4.68 (dd, 1H, H-2, JZ1.5,
6.0 Hz); 4.60 (d, 2H, H-5, JZ1.7 Hz); 4.03 (dd, 1H, H-3,
JZ1.5, 12.6 Hz); 3.80 (dd, 1H, H-3, JZ6.0, 12.6 Hz); 3.59
(s, 3H, OMe). 13C NMR (100 MHz) d 152.85 (C-10); 134.80
(C-11); 134.79 (C-7); 129.20 (C-8); 128.83 (C-6); 122.94
(C-9); 103.99 (C-2); 74.85 (C-3); 72.37 (C-5); 56.61 (OMe).
4.4.1. [4-Methoxy-2-(methoxyethoxymethoxymethyl)]-
phenyloxyacetaldehyde dimethyl acetal 11b. Compound
10b was used as starting material. Yield: 26%. Rf (ethyl
1
acetate/hexane: 2/1): 0.4. H NMR (300 MHz) d 6.95 (d,
1H, H-3, JZ2.8 Hz); 6.76 (t, 1H, H-6, JZ8.8 Hz); 6.73 (dd,
1H, H-5, JZ2.8, 8.8 Hz); 4.82 (s, 2H, OCH2O); 4.66 (t, 1H,
H-10, JZ5.2 Hz); 4.63 (s, 2H, C-2-CH2O); 3.94 (d, 2H,
H-20, JZ5.2 Hz); 3.73 (s, 1H, C-4-OMe); 3.73 (m, 2H, AA0
part of the ethylenedioxy fragment); 3.54 (m, 2H, BB0 part
of the ethylenedioxy fragment); 3.41 (s, 6H, (OMe)2); 3.37

E. Saniger et al. / Tetrahedron 60 (2004) 11453–11464
11463
(s, 3H, OMe). 13C NMR (75 MHz) d 154.05 (C-4); 150.20
(C-1); 128.22 (C-2); 114.75 (C-6); 113.16 (C-5); 113.12
(C-3); 102.37 (CH(OMe)2); 95.29 (OCH2O); 71.82, 68.97,
66.86, 64.55 (O-CH2–CH), (CH2OMe), (C-2-CH2O), (CH2-
CH2OMe); 59.03 (OMe); 55.72 (C-4-OMe); 54.17 (OMe)2.
HR LSIMS calcd for C16H26O7 (MCNa)C 353.1576, found
353.1577. Anal. for C16H26O7: Calcd: C 58.17; H 7.93.
Found: C 57.82; H 8.27.
13C NMR (75 MHz) d 150.47 (C-12); 147.87 (C-13); 130.12
(C-14); 122.48 (C-10); 120.18 (C-9); 111.38 (C-11); 100.47
(C-5); 72.19 (C-6); 71.85 (C-2); 67.20 (C-8); 63.05 (C-3);
59.16 (C-12-OMe); 56.14 (C-5-OMe). HR LSIMS calcd for
C13H18O5Na (MCNa)C 277.1052, found 277.1055. Anal.
for C13H18O5: Calcd: C 61.4; H 7.13. Found: C 61.02; H
6.88.
4.6. X-ray crystallographic study of (RS)-1-(7-methoxy-
2,3-dihydro-5H-1,4-benzodioxepin-3-yl)-5-fluorouracil
4.4.2. [6-Methoxy-2-(methoxyethoxymethoxymethyl)-
phenyloxyacetaldehyde dimethyl acetal 11c. Compond
10c was used as starting material. Yield: 33%. Rf (diethyl
ether/hexane: 2/1): 0.35. 1H NMR (300 MHz) d 7.00 (d, 1H,
H-4, JZ7.7 Hz); 6.96 (dd, 1H, H-3 or H-5, JZ1.9, 7.7 Hz);
6.82 (dd, 1H, H-5 or H-3, JZ1.9, 7.7 Hz); 4.80 (s, 2H,
OCH2O); 4.70 (t, 1H, H-10, JZ5.4 Hz); 4.67 (s, 2H, C-2-
CH2O); 4.01 (d, 2H, H-20, JZ5.4 Hz); 3.81 (s, 3H, C-6-
OMe); 3.72 (m, 2H, AA0 part of the ethylenedioxy
fragment); 3.55 (m, 2H, BB0 part of the ethylenedioxy
fragment); 3.40 (s, 6H, (OMe)2); 3.37 (s, 3H, OMe). 13C
NMR (75 MHz) d 152.31 (C-6); 145.84 (C-1); 131.84 (C-2);
124.06 (C-4); 121.24 (C-3); 111.95 (C-5); 102.47
(CH(OMe)2); 95.15 (OCH2O); 71.87, 71.82, 66.87, 64.50
(O-CH2–CH), (CH2OMe), (C-1-CH2O), (CH2CH2OMe);
59.05 (OMe); 55.77 (C-6-OMe); 53.85 (OMe)2. HR LSIMS
calcd for C16H26O7Na (MCNa)C 353.1576, found
353.1573. Anal. for C16H26O7: Calcd: C 58.17; H 7.93.
Found: C 57.85; H 8.02.
A colourless crystal was mounted on a glass fibre and used
for data collection. Crystal data were collected at 298(2) K,
using a Bruker SMART CCD 1000 diffractometer. Graphite
˚
monochromated Mo Ka radiation (lZ0.71073 A) was used
throughout. The data were processed with SAINT18 and
corrected for absorption using SADABS (transmissions
factors: 0.976–0.971).19 The structure was solved by direct
methods using the programme SHELXS-9720 and refined by
full-matrix least-squares techniques against F2 using
SHELXL-97.21 Positional and anisotropic atomic displace-
ment parameters were refined for all nonhydrogen atoms.
Hydrogen atoms were located in difference maps and
included as fixed contributions riding on attached atoms
with isotropic thermal parameters 1.2 times those of their
carrier atoms. Criteria of a satisfactory complete analysis
were the ratios of rms shift to standard deviation less than
0.001 and no significant features in final difference maps.
Atomic scattering factors from ‘International Tables for
Crystallography’.22 Molecular graphics and geometrical
calculatioons from PLATON23 and SHELTXL.24 Relevant
crystal data: formula C14H13FN2O5, formula weight 308.26,
TZ298(2) K, crystal system triclinic, space group P-1, unit
cell dimensions aZ6.491(2), bZ7.588(3) and cZ
4.5. Synthesis of (RS)-3-methoxy-2,3-dihydro-5H-1,4-
benzodioxepins 2b,c and (RS)-3,12-dimethoxy-2,3,5,6-
tetrahydro-8H-benzo-[1,4,7]-trioxecin 6c
The procedure is similar to the one used in 4.3, but changing
9a by 11b, and 9a by 11c.
˚
14.037(2) A, and aZ89.41(2)8, bZ87.70(2) and gZ
73.72(2), ZZ2, DZ1.544 Mg mK3
,
m(Mo Ka)Z
0.127 mmK1, measured/unique reflections 7805/3007
[R(int) 0.0191], refined parameters 200, final R1 (IO
2s(I))Z0.0408 and wR2Z0.1119, and GOFZ1.053.
CCDC reference number 236010. Copies of the data can
be obtained, free of charge, on application to CCDC, 12
Union Road, Cambridge, CB2 1EZ, UK [fax: C44 1223
336033 or e-mail: deposit@ccd.cam.ac.uk].
4.5.1. (RS)-2,7-Dimethoxy-2,3-dihydro-5H-1,4-benzo-
dioxepin 2b. Compond 11b was used as a starting material,
yielding 40% of 2b, whose spectroscopical characteristics
were identical to the ones previously described.4 The
organic residue was subjected to a simple combination of
directly-coupled gas chromatography-mass spectrometry
(GC-MS): tR of methoxymethoxyethanol: 6.76 min.
4.5.2. (RS)-3,9-Dimethoxy-2,3-dihydro-5H-1,4-benzo-
dioxepin 2c and (RS)-3,12-dimethoxy-2,3,5,6-tetra-
hydro-8H-benzo-[1,4,7]-trioxecin 6c. Compond 11c was
used as a starting material, yielding 70% of 2c and 15% of
6c. The spectroscopic characteristics of 2c were identical to
the ones previously described.5 The aqueous layer was
freeze-dried and the residue was subjected to a simple
combination of directly-coupled gas chromatography-mass
spectrometry (GC-MS): tR of methoxymethanol: 4.14 min;
tR of methoxymethoxyethanol: 6.76 min.
Acknowledgements
This study was supported by the Instituto Carlos III (Fondo
de Investigaciones Sanitarias, Projects PI03225 and
PI021029). E. S. was supported by a fellowship from the
´
Ministerio de Educacion, Cultura y Deporte. The referees’
criticisms are greatly appreciated.
Compound 6c. Yield: 15%. Rf (2/1, diethyl ether/hexane):
0.27. 1H NMR (300 MHz) d 6.88 (d, 1H, H-10, JZ7.6 Hz);
6.80 (d, 1H, H-9, JZ1.5 Hz); 6.64 (dd, 1H, H-11, JZ1.5,
7.6 Hz); 5.24 (d, 1H, H-8, JZ14.5 Hz); 5.05 (dd, 1H, H-5,
JZ3.4, 7.6 Hz); 4.38 (dd, 1H, H-8, JZ14.5 Hz); 4.29 (dd,
1H, H-6, JZ3.4, 13.1 Hz); 4.13 (dd, 1H, H-6, JZ7.6,
13.1 Hz); 3.95 (m, 1H, H-2); 3.83 (s, 3H, C-12-OMe); 3.69
(m, 1H, H-2); 3.56 (m, 2H, H-3, H-3); 3.38 (s, 3H, OMe).
References and notes
1. Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic
Synthesis, 3rd ed.; Wiley: New York, 1999.
2. Pearson, A.; Roush, W. R. Handbook of Reagents for Organic
Synthesis: Activating and Agents and Protecting Groups;
Wiley: Chichester, 1999.

11464
E. Saniger et al. / Tetrahedron 60 (2004) 11453–11464
3. Sefkow, M.; Kaatz, H. Tetrahedron Lett. 1999, 40, 6561–6562.
´
4. Saniger, E.; Campos, J.; Entrena, A.; Marchal, J. A.; Suarez, I.;
14. Blumenkopf, T. A.; Look, G. C.; Overman, L. E. J. Am. Chem.
Soc. 1990, 112, 4399–4403.
´
Aranega, A.; Choquesillo, D.; Niclos, J.; Gallo, M.A.;
´
15. Bartels, B.; Garcıa-Yebra, C.; Rominger, F.; Helmchen, G.
Eur. J. Inorg. Chem. 2002, 2569–2586.
´
´
Espinosa, A. Tetrahedron 2003, 59, 5457–5467.
16. Heaton, B. T.; Jacob, C.; Monks, G. L.; Hursthouse, M. B.;
Ghatak, I.; Somerville, R. G.; Heggie, W.; Page, P. R.; Villax, I.
J. Chem. Soc., Dalton Trans. 1996, 61–67.
5. Saniger, E.; Campos, J.; Entrena, A.; Marchal, J. A.; Boulaiz,
´
´
H.; Aranega, A.; Gallo, M.A.; Espinosa, A. Tetrahedron 2003,
59, 8017–8026.
17. Hammond, M. L.; Zambias, R. A.; Chang, M. N.; Jensen,
N. P.; McDonald, J.; Thompson, K.; Boulton, D. A.; Kopka,
I. E.; Hand, K. M.; Opas, E. E.; Luell, S.; Bach, T.; Davies, P.;
MacIntyre, D. E.; Bonney, R. J.; Humes, J. L. J. Med. Chem.
1990, 33, 908–918.
6. Corey, E. J.; Gras, J.-L.; Ulrich, P. Tetrahedron Lett. 1976, 11,
809–812.
7. Sato, T.; Otera, J.; Nozaki, H. J. Org. Chem. 1990, 55,
4770–4772.
´
8. Espinosa, A.; Entrena, A.; Gallo, M.A.; Campos, J.;
´ ´
Domınguez, J. F.; Camacho, E.; Sanchez, I. J. Org. Chem.
18. Bruker. SMART and SAINT. Area Detector Control and
Integration Software; Bruker Analytical X ray Instruments
Inc.: Madison, WI, USA, 1997.
1990, 55, 6018–6023.
´
9. Campos, J.; Pineda, M. J.; Gomez, J. A.; Entrena, A.; Trujillo,
´
19. Sheldrick, G. M., SADABS. Program for Empirical Absorption
´
M.A.; Gallo, M.A.; Espinosa, A. Tetrahedron 1996, 52,
8907–8924.
¨
Correction of Area Detector Data; University of Gottingen:
Germany, 1997.
10. Smith, M. B.; March, J. March’s Advanced Organic
Chemistry. Reactions, Mechanisms, and Structure; Wiley:
New York, 2001; p 345.
20. Sheldrick, G. M. Acta Cryst. 1990, A46, 467–473.
21. Sheldrick, G. M. SHELXL-97. Program for the Refinement of
¨
Crystal Structures; University of Gottingen: Germany, 1997.
22. Wilson, A. J. C.. In International Tables for Crystallography,
Vol. C; Kluwer: Dordrecht, 1995.
11. Hlasta, D. J.; Casey, F. B.; Ferguson, E. W.; Gangell, S. J.;
Heimann, M. R.; Jaeger, E. P.; Kullnig, R. K.; Gordon, R. J.
J. Med. Chem. 1991, 34, 1560–1570.
23. Spek, A. L. PLATON. A Multipurpose Crystallographic Tool;
Utrecht University: Utrecht, 2003.
´
12. Marko, I. E.; Kantam, M. L. Tetrahedron Lett. 1991, 32,
2255–2258.
24. Bruker. SHELXTL: Integrated System for the Determination of
Crystal Structures; Bruker AXS Inc.: Madison, WI, USA,
2000.
13. Mohler, D. L.; Thompson, D. W. Tetrahedron Lett. 1987, 28,
2567–2570.