Zeolite β induced rearrangement of allyl benzyl ethers. 6. Variation of the aromatic part and synthesis of dihydronaphthalene derivatives

Article abstract of DOI:10.1021/jo980698n

The zeolite β induced rearrangement of substituted allyl benzyl ethers to give 4-arylbutanals was investigated with respect to the substituents in the aromatic ring. In some cases the resulting aldehydes cyclized spontaneously to give dihydronaphthalene derivatives. The rearrangement and also the ring closure to dihydronaphthalenes failed or gave poor yields in cases where too weak electron-donating substituents were present in the aromatic ring. Also the dimensions of the pore size of the zeolite in relation to the transition state of the cyclization seemed to be of importance. Replacement of the benzylic part with other structures potentially capable of stabilizing cationic centers have hitherto not resulted in successful rearrangements.

Full text of DOI:10.1021/jo980698n

54
J . Org. Chem. 1999, 64, 54-59
Zeolite â In d u ced Rea r r a n gem en t of Allyl Ben zyl Eth er s. 6.
Va r ia tion of th e Ar om a tic P a r t a n d Syn th esis of
Dih yd r on a p h th a len e Der iva tives†
J ohan Wennerberg, Fredrik Ek, Anna Hansson, and Torbjo¨rn Frejd*
Organic Chemistry 1, Department of Chemistry, Lund University, P.O. Box 124, S-221 00 Lund, Sweden
Received April 15, 1998
The zeolite â induced rearrangement of substituted allyl benzyl ethers to give 4-arylbutanals was
investigated with respect to the substituents in the aromatic ring. In some cases the resulting
aldehydes cyclized spontaneously to give dihydronaphthalene derivatives. The rearrangement and
also the ring closure to dihydronaphthalenes failed or gave poor yields in cases where too weak
electron-donating substituents were present in the aromatic ring. Also the dimensions of the pore
size of the zeolite in relation to the transition state of the cyclization seemed to be of importance.
Replacement of the benzylic part with other structures potentially capable of stabilizing cationic
centers have hitherto not resulted in successful rearrangements.
Sch em e 1. L ) BF 3‚OEt2 or Zeolite â
In tr od u ction
We recently found that substituted alkoxybenzyl allyl
ethers 1 rearranged to give 4-arylbutanals 2 on treatment
with zeolite â or BF3‚OEt2 via a combined 1,4-rearrange-
ment/1,2-migration sequence (Scheme 1).1 Here the over-
all reaction is referred to as the BenzAll rearrangement
(benzyl allyl ether rearrangement). As an early example,
a natural product was synthesized very efficiently using
this rearrangement as a key step,2 and we have also
reported on the mechanism,3 the reaction conditions,4 and
variations of the allylic moiety.5 We now present results
for substrates carrying other aromatic parts than p-
methoxyphenyl as part of an ongoing study of the scope
and limitations of the BenzAll rearrangement.
Resu lts a n d Discu ssion
the low-yielding 1e. The aldehydes resulting from the
disubstituted derivatives 1e-g cyclodehydrated sponta-
neously in the presence of the zeolite to give the dihy-
dronaphthalenes 2e-g.6 In the case of 1e, the rearrange-
ment was not as favored as one might have expected. Due
to steric crowding the lone pairs of the 2-methoxy group
is probably forced out of conjugation with the π-system
of the phenyl ring; a case of steric inhibition of reso-
nance.7-9 However, as soon as some of the corresponding
aldehyde was formed, it immediately cyclodehydrated to
give the dihydronaphthalene 2e. The cyclization step was
of course facilitated by the presence of the 3-methoxy
group. Even if the 2-methoxy group of 1i also would be
inefficient in delocalization of charge due to steric inhibi-
tion, the less-crowded 4-methoxy group could serve this
purpose, thus promoting the rearrangement to give 2i
in good yield (entry 7).
The reactions were performed in a very simple and
practical way. Mixtures of the substrates 1 and zeolite â
in dichloromethane were stirred at +20 °C for 12 h,
followed by filtration and evaporation of the solvent.
Thus, extraction is not necessary in contrast to methods
using Lewis acids in solution. As seen in Table 1, a
number of substrates rearranged to give the correspond-
ing 4-arylbutanals and in some cases the aldehydes
cyclodehydrated in situ to give dihydronaphthalenes.
While the two monomethoxy-substituted substrates 1a
and 1b carrying the methoxy substituent in the 2-position
rearranged to give aldehydes 2a and 2b in good yields
(entry 1), the 3-methoxy-substituted substrate 1q (Figure
1), and the unsubstituted benzyl ether itself did not
rearrange.1
Also the di- and trisubstituted methoxy derivatives
shown in entries 2-8 rearranged efficiently except for
(6) In connection with the synthesis of dihydronaphthalene deriva-
tives, we observed that some samples were oxidized to naphthalene
derivatives during storage in CDCl3 solution. This is probably due to
air dissolved in the solvent, since this phenomenon was not observed
with neat crystalline substances, although they were still somewhat
sensitive to oxygen.
(7) Wepster, B. M. Recl. Trav. Chim. Pays-Bas 1952, 71, 1159-1178.
(8) Wheland, G. W. The Theory of Resonance; Wiley & Sons: New
York, 1944.
† For Part 5, see ref 5.
(1) Wennerberg, J .; Eklund, L.; Polla, M.; Frejd, T. Chem. Commun.
1997, 445.
(2) Wennerberg, J .; Frejd, T. Nat. Prod. Lett. 1997, 10, 1-6.
(3) Wennerberg, J .; Frejd, T. Acta Chem. Scand. 1998, 52, 95-99.
(4) Wennerberg, J .; Olofsson, C.; Frejd, T. Acta Chem. Scand. 1998,
52, 232-235.
(5) Wennerberg, J .; Olofsson, C.; Frejd, T. J . Org. Chem. 1998, 63,
3595-3598.
(9) Exner, O.; Folli, U.; Marcaccioli, S.; Vivarelli, P. J . Chem. Soc.,
Perkin Trans. 2 1983, 757-760.
10.1021/jo980698n CCC: $18.00 © 1999 American Chemical Society
Published on Web 12/12/1998

Zeolite â Induced Rearrangement
J . Org. Chem., Vol. 64, No. 1, 1999 55
Ta ble 1. Resu lts of th e Zeolite â Ca ta lyzed Rea r r a n gem en t of Su bstitu ted Ben zyl Eth er s 1a -la
a
Conditions: 1 mmol scale, 100 mg of zeolite â, 12 h, rt.
The naphthalene derivative 1k gave a useful yield of
2k (entry 9) but cyclization leading to a seven-membered
ring was not observed under the mild conditions used. A
slow reaction of the thiophene ether 1l was noticed at
room temperature and in refluxing dichloromethane a
low yield of the benzo[b]thiophene derivative 2l was
formed. A more detailed investigation of other thiophene
cases is in progress in our laboratory. It should be
mentioned that the corresponding furan derivative, 1u
(Figure 1), was too labile to give any identifiable products
under the reaction conditions used.
by the substituents in the aromatic ring. Thus, the
methoxy groups in the 2- and 4-positions were ideally
positioned for the rearrangement unless steric inhibition
of resonance prevented efficient delocalization of the
positive charge. Most likely this explains why 1m had
to be heated (110 °C) in order to react, but then gave rise
to three compounds. The first formed aldehyde probably
cyclodehydrated under these condidtions to give naph-
thalene derivative 2m together with 3 and 4 in a 3:2:2
ratio, respectively (Scheme 2). The latter two compounds
may have been formed via disproportionation of 2m due
to the higher reaction temperature. Steric inhibition of
resonance may also explain why 1v did not rearrange
(Figure 1) and that 2i did not cyclize (entry 7). Molecular
From the results shown in Table 1 it is apparent that
the rearrangement required that the positive charge (full
or partial) on the benzylic carbon could be delocalized

56 J . Org. Chem., Vol. 64, No. 1, 1999
Wennerberg et al.
Sch em e 3
However, to determine if the cyclization may take place
inside the pores of the zeolite, further experimentation
is necessary.
The BF3‚OEt2-catalyzed trichloroacetimidate method
is frequently used for PMB-protection of alcohols12 and
was also very suitable for the synthesis of the allyl benzyl
ethers used in this investigation. Since we earlier showed
that besides zeolites,13 BF3‚OEt2 also catalyzed the Benz-
All rearrangement,1,3 it occurred to us that 4-arylbutanals
and perhaps also the dihydronaphthalenes could be
formed in situ already in the benzylation step. When this
methodology was applied to the coupling of 3,4-dimethoxy-
benzyl alcohol with 2-methyl-2-propenol, the dihydronaph-
thalene derivative 2g was indeed formed directly in the
reaction mixture (Scheme 3). Even if the reaction was
rather low yielding, it still represents an example of an
overall 4-step in situ production of dihydronaphthalenes
from the allyl alcohol/trichloroacetimidate mixture. At-
tempts to increase the yield by using a higher concentra-
tion of the Lewis acid were unsuccessful due to decom-
position of the imidate.
F igu r e 1. Examples of allylic ethers for which the BenzAll
rearrangement was unsuccessful.
Sch em e 2
mechanics calculations (MM3(92))10 clearly indicated that
the lone pairs of the 3-methoxy group of 2,3,4-tri-
methoxytoluene, a model of 2i, did not overlap with the
π-system of the benzene ring in the lowest energy
conformations.
(11) We believe that the structures 2d ′, 2h ′, and 2f ′ of the initial
products of the carbocyclization could be used as reasonable transition-
state models for estimating the sizes of the transition states of the
cyclizations. Thus, these structures were used as input structures for
the molecular mechanics calculations of the low-energy conformers
using CS Chem3D Pro version 3.2. In the case of 2d ′ the lowest energy
conformation obtained was the one having the methoxy methyls
essentially in plane with the aromatic ring. The distance between the
atomic nulei of the outermost hydrogens of the methoxy groups was
8.9 Å. For 2h ′ the corresponding treatment gave a distance between
the bromine and the methyl hydrogens of the para-positioned methoxy
group of 7.8 Å, and for 2f ′ the smallest distance 5.6 Å was between
the benzylic methylene hydrogens and the OH group. It should be
pointed out that all structures occupy larger volumes when the van
der Waals radii are included. Thus, for 2f ′ it was estimated that the
molecule would be accommodated in a pore size of 7.4 × 7.4 Å. In all
three cases the trans orientation of the hydroxyl group and the methyl
group of the nonaromatic ring was used.
The easy formation of the dihydronaphthalenes may
be explained by the fact that all substrates carried a
methoxy group in the 3-position in relation to the benzylic
carbon. In the cyclization step this methoxy group will
be positioned in the para position to the ring carbon
undergoing electrophilic attack. The 2,4-dimethoxy sub-
strate 1c (entry 2) cannot easily undergo ring closure
since both methoxy groups would be in the meta position
to the cyclization site. Interestingly, the 2,5-dimethoxy
compound 1d (entry 3) and the bromo compound 1h
(entry 6) were very well arranged for a tandem rear-
rangement-cyclodehydration reaction, but still gave very
little of the cyclization products corresponding to alde-
hydes 2d and 2h . In these cases the dimensions of the
transition states leading to ring closure (estimated to be
>7.8 Å for both cases) would be too large to be accom-
modated in the pores of the zeolite (7.4 Å). This was,
however, not the case for 1f and 1g having the substit-
uents in the 3- and 4-positions. Here the transition states
for the cyclizations would be smaller (<7.4 Å) and would
therefore better fit the size of the pores, resulting in a
more efficient cyclization.11 Soluble Lewis acids such as
BF3‚OEt2 did indeed induce cyclization of 2d as judged
by NMR analysis of a purified sample, but the reaction
mixture contained several other unknown compounds.
(12) Audia, J . E.; Boisvert, L.; Patten, A. D.; Villalobos, A.; Dan-
ishefsky, S. J . J . Org. Chem. 1989, 54, 3738-3740.
(13) Zeolites having pore sizes similar to those of zeolite â (7.4 ×
7.4 Å) such as zeolite Y (Y-25, 7.4 × 7.4 Å) and mordenite (M-82, 6.5
× 7.0 Å) may also be used, but zeolites with smaller pore sizes such as
molecular sieves (4 Å) and zeolite ZSM-5 (5.1 × 5.4 Å) did not give
any reaction. More details of the reaction conditions may be found in
ref 4.
(10) Software: MacMimic. InStar Software AB, Ideon Research
Park, S-223 70 Lund, Sweden.

Zeolite â Induced Rearrangement
J . Org. Chem., Vol. 64, No. 1, 1999 57
Sch em e 4
appropriate allylic halides and benzylic alcohols except
for 1m , the synthesis of which was best performed from
the corresponding benzylic halide and allylic alcohol.
In conclusion, it seems necessary that strongly electron-
donating substituents must be present in the aromatic
ring in order to achieve high yields of the 4-arylbutanals
in the BenzAll rearrangement. This requirement may be
less emphasized once the conditions have been further
developed. Strong activation of the aromatic ring toward
electrophilic aromatic substitution leads to ring closure
to give dihydronaphthalenes under the reaction condi-
tions. In those cases where dihydronaphthalenes were
not spontaneously formed, the initially formed aldehydes
may be cyclized e.g. by the use of PPA as described
earlier.1,5 We also point out that the experimental proce-
dure is very simple and should be adaptable to any scale.
Except for being aromatized to naphthalenes, dihy-
dronaphthalenes may be used as substrates for asym-
metric epoxidation,20 dihydroxylation,21and Heck reac-
tions.22 Moreover, substituted naphthalenes and hydro-
naphthalenes are common features of numerous natural
products and pharmaceuticals,23-31 and much attention
has been paid to improve and develop methods for their
synthesis.32 Since a vast number of allylic halides and
benzylic alcohols are easily available, the BenzAll rear-
rangement constitutes an interesting alternative for the
combinatorial synthesis of naphthalenic libraries.
To further test the scope of the BenzAll rearrangement
a number of allyl benzyl ethers carrying groups other
than methoxy in the para position were tested (Scheme
4). Thus, benzyloxy derivative 1n gave aldehyde 2n in
good yield (75%). On the other hand, the TBDMS-
protected derivative 1r (Figure 1) was unreactive; the
starting material was recovered in high yield. Again,
steric inhibition of the delocalization of the positive
charge may be the reason for the unreactivity. In this
case the free electron pairs of oxygen may be prevented
to participate effectively in delocalization of charge
because the steric bulk of the TBDMS group forces the
lone pair electrons of oxygen out of conjugation with the
aromatic π-system. In addition, the lone-pair electrons
may be less available for conjugation due to the neigh-
boring silicon effect. It was shown several years ago that
Lewis acids such as SnCl4 did not form complexes with
silyl ethers,14,15 which recently was explained to be due
to the lower coefficients of the HOMO at oxygen in silyl
ethers as compared to that of corresponding alkyl ethers.16
Surprisingly, hydroxy compound 1s, synthesized by
deprotection of 1r using Bu4NF, was unreactive. A
possible explanation could be protonation or Lewis acid
coordination of the hydroxy group, which would coun-
teract the charge delocalization. Both the p-methyl
derivative 1o and the p-methylthio derivative 1p gave
low yields of the corresponding 4-arylbutanals 2o and 2p ,
respectively, despite prolonged reaction times and in-
crease of the reaction temperature to 40 °C. As expected
the nitro compound 1t failed to react. From these
experiments it is obvious that to be efficient the BenzAll
rearrangement needs strongly electron-donating substit-
uents ortho or para to the benzylic carbon.
Finally, some nonbenzylic derivatives carrying groups
known to be able to stabilize cations were investigated.
Cyclopropyl derivative 617 was not stable under the
present reaction conditions (zeolite â); a multitude of
products were formed as detected by TLC. The galactosyl
derivative 7,18 on the other hand, was essentially unre-
active; just a small loss of benzaldehyde was noticed. This
is in contrast to the results of Ley et al.19 who reported
that simple THP-allyl ethers (the THP unit may be
regarded as a simplified carbohydrate) rearranged in the
presence of SnCl4. However, in our case the carbohydrate
carries several coordination sites for Lewis acids or
protons and is also more bulky, perhaps too bulky to fit
into the pores of the zeolite.
Exp er im en ta l Section
Gen er a l. GC analyses were performed on a DBwax (J &W
Scientific) capillary column 30 m × 0.25 mm i.d., 0.25 µm
stationary phase. NMR spectra were recorded at 300 MHz
using CDCl3 as internal standard. Mass spectra were recorded
in the following modes: EI (70 eV) using both direct inlet and
inlet via a gas chromatograph equipped with a DB Wax column
as above, CI (CH4, 200 eV) and FAB. Chromatographic
separations were performed on Matrex Amicon normal phase
silica gel 60 (0.035-0.070 mm). Thin-layer chromatography
were performed on Merck precoated TLC plates with Silica
gel 60 F-254, 0.25 mm. After eluation, the TLC plates were
visualized with UV light or sprayed with a solution of p-
methoxybenzaldehyde (26 mL), glacial acetic acid (11 mL),
concentrated sulfuric acid (35 mL), and 95% ethanol (960 mL)
followed by heating. Chemicals were reagent grade. Zeolite â
was a gift from EKA Chemicals AB, Bohus, Sweden, and was
activated at 400 °C for 3 h before use. Zeolite â is also
commercially available.33 THF was distilled under N2 from
sodium benzophenone ketyl, and CH2Cl2 was distilled from
P2O5 prior to use.
(20) Zhang, W.; Loebach, J . L.; Wilson, S. R.; J acobsen, E. N. J . Am.
Chem. Soc. 1990, 112, 2801-2803.
(21) Kolb, H. C.; Vannieuwenhze, M. S.; Sharpless, K. B. Chem. Rev.
1994, 94, 2483-2547.
(22) Cabri, W.; Candiani, I. Acc. Chem. Res. 1995, 28, 2-7.
(23) J ardine, I. Podophyllotoxins in anticancer agents based on
natural product model; Academic Press: New York, 1980.
(24) Look, S. A.; Fenical, W.; J acobs, R. S.; Clardy, J . Proc. Natl.
Acad. Sci. U.S.A. 1986, 83, 6238-6240.
(25) Look, S. A.; Fenical, W.; Matsumoto, G. K.; Clardy, J . J . Org.
Chem. 1986, 51, 5140-5145.
The syntheses of the allyl ether starting materials were
performed by Williamson ether syntheses between the
(26) Chan, A. S. C. Chemtech 1993, 46-51.
(27) Rucker, M.; Bru¨ckner, R. Synlett 1997, 1187-1189 and refer-
ences therein.
(14) Kahn, S. D.; Keck, G. E.; Hehre, W. J . Tetrahedron Lett. 1987,
28, 279-280.
(28) Suzuki, K. In Classics in Total Synthesis. Targets, Strategies,
Methods; Nicolaou, K. C., Sorensen, E. J ., Eds.; VCH: Weinheim, 1996,
pp 509-521 and references therein.
(15) Keck, G. E.; Castellino, S. Tetrahedron Lett. 1987, 28, 281-
284.
(29) Ward, R. S. Nat. Prod. Rep. 1993, 10, 1-28.
(30) Ward, R. S. Nat. Prod. Rep. 1995, 12, 183-205.
(31) Ward, R. S. Nat. Prod. Rep. 1997, 14, 43-74.
(32) Katritzky, A. R.; Zhang, G.; Xie, L. J . Org. Chem. 1997, 62,
721-725 and references therein.
(16) Shambayati, S.; Blake, J . F.; Wierschke, S. G.; J orgensen, W.
L.; Schreiber, S. L. J . Am. Chem. Soc. 1990, 112, 697-703.
(17) Schlosser, M.; Strunk, S. Tetrahedron 1989, 45, 2649-2664.
(18) Frejd, T. Unpublished results.
(19) Buffet, M. F.; Dixon, D. J .; Edwards, G. L.; Ley, S.; Tate, E.
Synlett 1997, 1055-1056.
(33) PQ Corporation, Zeolites and Catalysts Division, P.O. Box 840,
Valley Forge, PA 19482.

58 J . Org. Chem., Vol. 64, No. 1, 1999
Wennerberg et al.
Gen er a l P r oced u r e for th e Syn th esis of Allylic Eth er s
fr om Ben zylic Alcoh ols. A solution of the appropriate
benzylic alcohol (10 mmol) in THF (5 mL) was slowly added
to a suspension of NaH (440 mg, 11 mmol, 60% in mineral oil)
in DMF (5 mL) at 0 °C under Ar, followed by 30 min stirring.
3-Bromo-2-methylpropene or 3-bromopropene (10.3 mmol) in
THF (2 mL) was then added, and the mixture was stirred for
2 h, whereafter water (1 mL) and ether (10 mL) were added.
The organic phase of the resulting two-phase mixture was
washed with water (5 mL) and brine (5 mL) followed by drying
of the organic extract (MgSO4) and concentration under
reduced pressure.
3-(2-Meth oxyben zyloxy)-2-m eth ylp r op en e (1b). Chro-
matography (heptane-EtOAc, 10:1) gave 1.52 g (79%) of 1b:
1H NMR (CDCl3) δ 7.43, (m, 1 H), 7.29, (m, 1 H), 6.99, (m, 1
H), 6.89, (m, 1 H), 5.06, (m, 1 H), 4.96, (m, 1 H), 4.57, (s, 2 H),
4.01, (s, 2 H), 3.85, (s, 3 H), 1.81, (s, 3 H); 13C NMR (CDCl3) δ
157.1, 142.5, 128.8, 128.5, 126.9, 120.4, 112.1, 110.2, 74.4, 66.7,
55.3, 19.5. HRMS calcd for C12H16O2 192.1151, found 192.1151.
2-Meth yl-3-(3,4-m eth ylen edioxyben zyloxy)pr open e (1g).
Chromatography (heptane-EtOAc, 8:2) gave 1.6 g (80%) of
1g: 1H NMR (CDCl3) δ 6.87, (s, 1 H), 6.79, (m, 2 H), 5.96, (s,
2 H), 5.00, (s, 1 H), 4.94, (s, 1 H), 4.39, (s, 2 H), 3.91, (s, 2 H),
1.78, (s, 3 H); 13C NMR (CDCl3) δ 147.7, 147.0, 142.2, 132.3,
121.3, 112.4, 108.5, 108.0, 101.0, 73.9, 71.7, 19.6; HRMS calcd
for C12H14O3 206.0943, found 206.0944.
2-Me t h oxy-1-[((2-m e t h yla llyl)oxy)m e t h yl]n a p h t h a -
len e (1k ). Chromatography (heptane-EtOAc, 8:2) gave 1.81
g (71%) of 1k : 1H NMR (CDCl3) δ 8.15, (d, J ) 8.7 Hz, 1 H),
7.83, (m, 2 H), 7.54-7.25, (m, 3 H), 5.05, (s, 1 H), 5.02, (s, 2
H), 4.93, (s, 1 H), 4.02, (s, 2 H), 3.97, (s, 3 H), 1.78, (s, 3 H);
13C NMR (CDCl3) δ 155.6, 142.7, 142.7, 133.9, 130.2, 129.2,
128.2, 126.7, 123.9, 123.6, 113.4, 112.3, 74.2, 62.0, 56.7, 19.7;
HRMS calcd for C16H18O2 242.1306, found 242.1310.
2-Meth yl-5-[((2-m eth yla llyl)oxy)m eth yl]th iop h en e (1l).
Chromatography (heptane-EtOAc, 8:2) gave 1.62 g (80%) of
1l: 1H NMR (CDCl3) δ 6.79, (d, J ) 3.6 Hz, 1 H), 6.61, (d, J )
3.6 Hz, 1 H), 5.01, (s, 1 H), 4.97, (s, 1 H), 4.53, (s, 2 H), 3.92,
(s, 2 H), 2.48, (s, 3 H), 1.78, (s, 3 H); 13C NMR (CDCl3) δ 142.0,
140.5, 138.8, 126.5, 124.6, 112.6, 73.5, 66.4, 19.6, 15.4; HRMS
calcd for C10H15OS (M + H) 183.0844, found 183.0843.
2-[((2-Meth yla llyl)oxy)m eth yl]fu r a n (1u ). Chromatog-
raphy (heptane-EtOAc, 8:2) gave 0.90 g (59%) of 1u : 1H NMR
(CDCl3) δ 7.41, (s, 1 H), 6.34, (m, 2 H), 4.99, (s, 1 H), 4.92, (s,
1 H), 4.43, (s, 2 H), 3.93, (s, 2 H), 1.75, (s, 3 H); 13C NMR
(CDCl3) δ 151.9, 142.7, 141.9, 112.7, 110.1, 109.2, 73.9, 63.6,
19.5; HRMS calcd for C9H13O2 (M + H) 153.0915, found
153.0922.
3-(3,4,5-Tr im eth oxyben zyloxy)-2-m eth ylp r op en e (1m ).
A solution of 2-methyl-2-propenol (720 mg, 10 mmol) in THF
(5 mL) was slowly added to a suspension of NaH (440 mg, 11
mmol, 60% in mineral oil) in DMF (5 mL) at 0 °C under Ar
followed by 30 min of stirring. 3,4,5-Trimethoxybenzyl chloride
(2.23 g, 10.3 mmol) in THF (2 mL) was then added, and the
mixture was then stirred for 2 h. Workup was as described
above. Chromatography of the residue (heptane-EtOAc 1:1)
gave the title compound (2.17 g, 86%): 1H NMR (CDCl3) δ 6.57,
(s, 2 H), 4.99, (m, 1 H), 4.92, (m, 1 H), 4.41, (s, 2 H), 3.92, (m,
2 H), 3.84, (s, 6 H), 3.82, (s, 3 H), 1.76, (s, 3 H); 13C NMR
(CDCl3) δ 153.2, 142.1, 134.1, 112.5, 105.7, 104.5, 74.1, 71.9,
60.8, 56.1, 19.6; HRMS calcd for C14H20O4 252.1362, found
252.1364.
3-(4-Hyd r oxyben zyloxy)-2-m eth ylp r op en e (1s). A solu-
tion of tetrabutylammonium fluoride (TBAF) (2.5 mL, 2.5
mmol, 1 M in THF) was added dropwise to 1r (440 mg, 1.5
mmol) in THF (30 mL) at 0 °C. The mixture was then kept at
room temperature for 4 h, whereafter it was concentrated
under reduced pressure. Ether (20 mL) was added to the
residue, and the resulting solution was washed with aqueous
saturated NaHCO3 (10 mL). The organic phase was dried
(MgSO4) and concentrated under reduced pressure. Chroma-
tography (heptane-EtOAc, 1:1) gave 255 mg (96%) of the title
compound: 1H NMR (CDCl3) δ 7.28, (d, J ) 8.6 Hz, 2 H), 6.89,
(d, J ) 8.6 Hz, 2 H), 5.08, (s, 1 H), 4.98, (s, 1 H), 4.58, (s, 2 H),
4.42, (s, 2 H), 3.92, (m 1 H), 1.80, (s, 3 H); 13C NMR (CDCl3) δ
158.4, 140.8, 133.2, 128.6, 114.8, 112.7, 71.8, 67.7, 65.0, 19.4;
HRMS calcd for C11H14O2 178.0994, found 178.0992.
Gen er a l P r oced u r e for th e Rea r r a n gem en t. A mixture
of the appropriate allyl benzyl ether (1.0 mmol) in CH2Cl2 (2
mL), or the solvent indicated, and zeolite â (100 mg), activated
at 400 °C for 3 h, was stirred under argon for 12 h at room
temperature, unless other times and temperatures are indi-
cated. The resulting colored mixture was then filtered through
Celite, the filter cake was washed with CH2Cl2 (2 × 5 mL),
and the combined organic extracts were concentrated under
reduced pressure.
4-(2-Meth oxyp h en yl)bu ta n a l (2a ). Chromatography (hep-
tane-EtOAc, 95:5) gave 122 mg (69%) of 2a : 1H NMR (CDCl3)
δ 9.76, (t, J ) 1.8 Hz, 1 H), 7.22-7.10, (m, 2 H), 6.91-6.84,
(m, 2 H), 3.82, (s, 3 H), 2.68, (t, J ) 7.5 Hz, 2 H), 2.44, (dt, J
) 7.4 Hz, J ) 1.8 Hz, 2 H), 1.94, (pent, J ) 7.4 Hz, 2 H); 13C
NMR (CDCl3) δ 202.9, 157.4, 130.0, 129.6, 127.4, 120.4, 110.2,
55.2, 43.3, 29.5, 22.3; IR (film) cm-1 2830, 1720; HRMS calcd
for C11H15O2 (M + H) 179.1072, found 179.1055.
1,2-Dih yd r o-6,7-d im eth oxy-3-m eth yln a p h th a len e (2f).
Chromatography (heptane-EtOAc, 10:1) gave 124 mg (61%)
of 2f: 1H NMR (CDCl3) δ 6.66, (s, 1 H), 6.55, (s, 1 H), 6.13, (s,
1 H), 3.87, (s, 3 H), 3.86, (s, 3 H), 2.75, (t, J ) 8.2 Hz, 2 H),
2.21, (t, J ) 8.3 Hz, 2 H), 1.89, (s, 3 H); 13C NMR (CDCl3) δ
147.3, 147.0, 136.1, 127.9, 126.3, 122.0, 111.3, 109.1, 56.1, 28.9,
27.8, 23.4; HRMS calcd for C13H16O2 204.1151, found 204.1147.
1,2-Dih yd r o-3-m e t h yl-6,7-m e t h yle n e d ioxyn a p h t h a -
len e (2g). Chromatography (heptane-EtOAc, 95:5) gave 118
mg (63%) of 2g: 1H NMR (CDCl3) δ 6.61 (s, 1 H), 6.52, (1 s, 1
H), 6.11, (s, 1 H), 5.89, (s, 2 H), 2.72, (t, J ) 8.2 Hz, 2 H), 2.19,
(t, J ) 8.2 Hz, 2 H), 1.89, (s, 3 H); 13C NMR (CDCl3) δ 145.9,
145.3, 136.2, 129.0, 127.6, 122.4, 108.3, 106.0, 100.5, 28.9, 28.3,
23.3; HRMS calcd for C12H12O2 188.0837, found 188.0840.
4-(2-Meth oxy-1-n a p h th yl)-2-m eth ylbu ta n a l (2k ). Chro-
matography (heptane-EtOAc, 8:2) gave 150 mg (62%) of 2k :
1H NMR (CDCl3) δ 9.71, (s, 1 H), 7.94, (d, J ) 8.7 Hz, 1 H),
7.79, (m, 2 H), 7.52, (t, 1 H), 7.37, (t, 1 H), 7.29, (d, 1 H), 3.94,
(s, 3 H), 3.13, (t, 2 H), 2.48, (m, 1 H), 2.06, (m,1 H), 1.70, (m,
1 H), 1.12, (d, 3 H); 13C NMR (CDCl3) δ 205.2, 154.3, 132.8,
129.3, 129.1, 128.6, 127.9, 126.5, 123.3, 122.9, 113.1, 56.4, 46.3,
30.7, 22.3, 13.6; IR (film) cm-1 2940, 1720; HRMS calcd for
C
16H18O2 242.1306, found 242.1306.
2,5-Dim eth yl-6,7-d ih yd r oben zo[b]th iop h en e (2l). Chro-
matography (heptane-EtOAc, 95:5) gave 17 mg (10%) of 2l:
1H NMR (CDCl3) δ 6.43, (s, 1 H), 6.09, (s, 1 H), 2.80, (t, J )
9.0 Hz, 2 H), 2.41, (s, 3 H), 2.31, (t, J ) 9.0 Hz, 2 H), 1.85, (s,
3 H); 13C NMR (CDCl3) δ 142.8, 135.8, 133.5, 129.4, 123.2,
117.7, 29.8, 23.2, 23.1, 15.2; HRMS calcd for C10H13S (M + H)
165.0738, found 165.0738.
1,2-Dih ydr o-5,6,7-tr im eth oxy-3-m eth yln aph th alen e (2m ),
1,2,3-Tr im eth oxy-7-m eth yln a p h th a len e (3), a n d 6,7,8-
Tr im eth oxy-2-m eth yltetr a lin (4). A mixture of allyl benzyl
ether 1m (252 mg, 1.0 mmol) in toluene (2 mL) and zeolite â
(100 mg) was stirred in a sealed tube at 110 °C for 18 h. The
workup was as described in the general procedure for the
rearrangement. Chromatography of the residue (heptane-
EtOAc, 9:1) gave a 3:2:2 mixture of 2m , 3, and 4, which were
identified by GC-HRMS.
1,2-Dih yd r o-3-m e t h yl-6,7-m e t h yle n e d ioxyn a p h t h a -
len e (2g) via BF 3‚Et2O Tr ea tm en t of 5. A solution of 3,4-
methylenedioxybenzyl alcohol (1.14 g, 7.50 mmol) in ether (10
mL) was added to a suspension of NaH (30 mg, 0.75 mmol,
60% in mineral oil) in ether (20 mL). The mixture was stirred
for 30 min and was then cooled to 0 °C. Trichloroacetonitrile
(0.75 mL, 7.5 mmol) was added, and the mixture was allowed
to reach room temperature over 4 h. The solvent was evapo-
rated under reduced pressure, and the orange residue was
dissolved in heptane (15 mL) containing MeOH (32 µL)
followed by filtration through Celite. Concentration of the
filtrate under reduced pressure gave crude 5 as a yellow syrup,
which was dissolved in cyclohexane (15 mL). To this solution
was added 2-methyl-2-propenol (360 mg, 5.00 mmol) in CH2-
Cl2 (7.5 mL). After the solution was cooled to 0 °C, BF3‚Et2O
(15 µL) was added followed by 12 h of stirring at room
temperature. The precipitate was then removed by filtration

Zeolite â Induced Rearrangement
J . Org. Chem., Vol. 64, No. 1, 1999 59
through Celite, and the filtrum was washed with a 1:2 mixture
of CH2Cl2 and cyclohexane (2 × 15 mL). The combined organic
phases were washed with saturated aqueous NaHCO3 (20 mL),
dried (MgSO4), and concentrated under reduced pressure.
Chromatography of the residue (heptane-EtOAc, 8:2) gave 2g
(188 mg, 20%) and 1g (484 mg, 47%). TLC Rf (heptane-EtOAc,
8:2) ) 0.62 for 2g and 0.23 for 1g.
Su p p or tin g In for m a tion Ava ila ble: 13C NMR spectra for
all compounds analyzed by HRMS, data for compounds 1a ,
1c-f, 1h -j, 1n , 1o-r , 1t, 1v, 2b-e, 2h -j, and 2n -p , and
stereoview of the three MM3(92)-minimized low-energy con-
formations of 2,3,4-trimethoxytoluene (43 pages). This material
is contained in libraries on microfiche, immediately follows this
article in the microfilm version of the journal, and can be
ordered from the ACS; see any current masthead page for
ordering information.
Ack n ow led gm en t. We thank the Swedish Natural
Science Research Council, The Knut and Alice Wallen-
berg Foundation, and The Crafoord Foundation for
economic support.
J O980698N