Synthesis and thermal rearrangements of 4-allyl-4-arylcyclobutenones

Article abstract of DOI:10.1021/jo990052a

The synthesis and thermolysis of 4-allyl-4-arylcyclobutenones are detailed. Allylcyclobutenones were prepared through the trapping of the cyclobutenonium cation generated by treatment of 4-hydroxycyclobutenones with Lewis acids. A study of the nucleophilic trapping revealed the following mechanistic highlights: the most likely intermediate is the cyclobutenonium cation, the regioselectivity of the allylation reaction is dictated by attack at the site most suited to stabilize the cation, and the rate of allylation is dependent on carbocation formation. Thermolysis of 4-allyl-4- arylcyclobutenones made possible a study of the competitive [2 + 2] cycloaddition and 6π electrocyclization. A judicious choice of substituents allows some control of the selectivity between the two pathways. The substituents control this selectivity through a combination of imposed electronics in the dienylketene intermediate, the relative ketenophilicity of the 4-position substituent, and the relative contortion substituents in the 2- and 3-positions impose on the transition state of the respective ring- closure processes. The scope of the cyclobutenonium cation trapping with silylated carbon nucleophiles was explored, producing 4-allenylcyclobutenones from the use of propargylsilane and 4-spirocyclobutenones from triisopropylsilane. Thermolysis studies of these precursors are also detailed.

Full text of DOI:10.1021/jo990052a

4030
J . Org. Chem. 1999, 64, 4030-4041
Syn th esis a n d Th er m a l Rea r r a n gem en ts of
4-Allyl-4-a r ylcyclobu ten on es
Ralf Tiedemann, Philip Turnbull, and Harold W. Moore*
Department of Chemistry, University of California, Irvine, California 92697-2025
Received J anuary 11, 1999
The synthesis and thermolysis of 4-allyl-4-arylcyclobutenones are detailed. Allylcyclobutenones were
prepared through the trapping of the cyclobutenonium cation generated by treatment of 4-hy-
droxycyclobutenones with Lewis acids. A study of the nucleophilic trapping revealed the following
mechanistic highlights: the most likely intermediate is the cyclobutenonium cation, the regiose-
lectivity of the allylation reaction is dictated by attack at the site most suited to stabilize the cation,
and the rate of allylation is dependent on carbocation formation. Thermolysis of 4-allyl-4-
arylcyclobutenones made possible a study of the competitive [2 + 2] cycloaddition and 6π
electrocyclization. A judicious choice of substituents allows some control of the selectivity between
the two pathways. The substituents control this selectivity through a combination of imposed
electronics in the dienylketene intermediate, the relative ketenophilicity of the 4-position substitu-
ent, and the relative contortion substituents in the 2- and 3-positions impose on the transition
state of the respective ring-closure processes. The scope of the cyclobutenonium cation trapping
with silylated carbon nucleophiles was explored, producing 4-allenylcyclobutenones from the use
of propargylsilane and 4-spirocyclobutenones from triisopropylsilane. Thermolysis studies of these
precursors are also detailed.
Sch em e 1
In tr od u ction
Reported here is a study concerning the synthesis and
thermal rearrangements of cyclobutenones bearing two
different ketenophilic substituents at the 4-position.
Specifically, the synthesis and thermolysis of 4-allyl-4-
arylcyclobutenones and selected related compounds are
presented. Although much is known of the synthetic
utility and mechanisms of rearrangements of cyclobuten-
ones, until now no systematic study has appeared in
which two groups are positioned to competitively inter-
cept the corresponding vinylketene intermediates.1 The
generalized results arising from this study are outlined
in Scheme 1. Thermolyses (refluxing toluene) of cy-
clobutenones 1 induce a rapid and reversible electrocyclic
ring opening to the corresponding vinylketenes 2 and 4.
In the rate-limiting steps, the former undergoes an
intramolecular [2 + 2] cycloaddition to give bicyclo[3.2.0]-
heptenones 3, and the latter proceeds to allylated naph-
thols 5 via 6π electrocyclic ring closure followed by
tautomerization.1-3 Both products are formed from the
examples presented here. However, because the inter-
mediate ketenes equilibrate under the reaction condi-
tions, a judicious choice of substituents allows some
control of the selectivity of the slower product-forming
ring-closure steps. Details of this investigation are pre-
sented below.
Syn th esis of 4-Allyl-4-a r ylcyclobu ten on es a n d
Rela ted Com p ou n d s. Syntheses of the required 4-allyl-
4-arylcyclobutenones stem from our continuing study of
the synthetic utility of cyclobutenonium ions, which are
readily generated from appropriately substituted cy-
clobutenones.4 In this regard, we previously reported that
4-aryl-4-hydroxycyclobutenones give high yields of the
4-protio derivatives when treated with BF3‚Et2O in the
presence of trialkylsilanes.5 For example, 6 gave 8 in 94%
yield where cation 7 is envisaged as a reaction intermedi-
ate (Scheme 2). When 8 was subjected to thermolysis
(1) For selected reviews on the synthetic utility of cyclobutenones
see: (a) Moore, H. W.; Yerxa, B. R. Advances in Strain in Organic
Chemistry; J AI Press: Greenwich, 1995; Vol. 4, pp 81-162. (b) Moore,
H. W.; Yerxa, B. R. Chemtracts 1992, 5, 273. (c) Liebeskind, L. S.
Tetrahedron 1989, 45, 3053. (d) Bellus, D.; Ernst, B. Angew. Chem.
1988, 100, 820. (e) Moore, H. W.; Decker, O. H. W. Chem. Rev. 1986,
86, 821.
(2) Xu, S. L.; Moore, H. W. J . Org. Chem. 1989, 54, 6018. Xu, S. L.;
Xia, H.; Moore, H. W. J . Org. Chem. 1991, 56, 6094.
(3) For an excellent review on intramolecular ketene/alkene cy-
cloadditions see: Snider, B. B. Chem. Rev. 1989, 88, 793.
(4) Xu, S. L.; Moore, H. W. J . Org. Chem. 1989, 54, 4024.
(5) Turnbull, P.; Moore, H. W. J . Org. Chem. 1995, 60, 644. Turnbull,
P.; Heileman, M. J .; Moore, H. W. J . Org. Chem. 1996, 61, 2584.
(6) For examples see: (a) Yamamoto, Y.; Noda, M.; Ohno, M.;
Eguchi, S. J . Org. Chem. 1997, 62, 1292. (b) Yamamoto, Y.; Ohno, M.;
Eguchi, S. Bull. Chem. Soc. J pn. 1996, 69, 1353. (c) Yamamoto, Y.;
Nunokawa, K.; Okamoto, K.; Ohno, M.; Eguchi, S. Synthesis 1995, 571.
(d) Yamamoto, Y.; Ohno, M.; Eguchi, S. Tetrahedron 1994, 50, 7783.
(e) Ohno, M.; Yamamoto, Y.; Eguchi, S. Syntlett 1998, 1167.
10.1021/jo990052a CCC: $18.00 © 1999 American Chemical Society
Published on Web 05/12/1999

Synthesis and Thermolysis of 4-Allyl-4-arylcyclobutenones
J . Org. Chem., Vol. 64, No. 11, 1999 4031
Sch em e 2
Sch em e 4
Sch em e 3
Sch em e 5
a class of compounds readily prepared from squaric acid
(Schemes 4-7).1,8 As exemplified in Scheme 4, methylene
chloride solutions of 4-aryl-4-hydroxycyclobutenones 17a-l
and allyltrimethylsilane were slowly treated with BF3‚
Et2O at -5 °C and then allowed to warm to ambient
temperature. These reaction conditions are envisaged to
induce the formation of cyclobutenonium ions 18, which
are intercepted by allyltrimethylsilane to give 19, the
ultimate precursors to 20. Under similar conditions,
17b,e,f,j gave adducts 21a -d in good to excellent yields
upon treatment with 2-methyl-3-trimethylsilylpropene
(Scheme 5).
(refluxing toluene) the ring-expanded naphthol 9 was
obtained in 87% yield. In a related and more extensive
study, Eguchi and co-workers6 generated analogous cy-
clobutenonium ions from cyclobutenedione monoketals7
and utilized these in a number of potentially useful
transformations, selected examples of which are provided
in Scheme 3. Specifically, various allyltrimethylsilanes
were observed to react with 3,4,4-triethoxy-2-methylcy-
clobutenone (10) in the presence of Lewis acids to give
allylated products such as 11-13 and 15. Under similar
conditions, 1-trimethylsilyl-2-phenylethyne gave the
4-alkynyl-4-ethoxycyclobutenone 16, and ditrimethylsi-
lylethyne gave the rearranged cyclopentene-1,3-dione 14.
Syntheses of the 4-allyl-4-arylcyclobutenones required
for the present study start with 4-hydroxycyclobutenones,
Even though cyclobutenonium ions such as 18 are
resonance hybrids and, thus, allylation could conceivably
take place at either position 2 or 4, the major products,
(8) For lead references concerning the synthesis of substituted
cyclobutenones see: Schmidt, A. H.; Ried, W. Synthesis 1978, 1. Knorr,
H.; Ried, W. Synthesis 1978, 649. Schmidt, A. H.; Ried, W. Synthesis
1978, 869. Reed, M. W.; Pollart, D. J .; Perri, S. T.; Foland, L. D.; Moore,
H. W. J . Org. Chem. 1988, 53, 2477. Liebeskind, L. S.; Fengl, R. W.;
Wirtz, K. R.; Shawe, T. T. J . Org. Chem. 1988, 53, 2482. Liebeskind,
L. S.; Fengl, R. W. J . Org. Chem. 1990, 5359. Liebeskind, L. S.; Wirtz,
K. R. J . Org. Chem. 1990, 5350. Liebeskind, L. S.; Wang, J . Tetrahe-
dron Lett. 1990, 4293.
(7) Gayo, L.; Moore, H. W. J . Org. Chem. 1992, 57, 6896.

4032 J . Org. Chem., Vol. 64, No. 11, 1999
Tiedemann et al.
Sch em e 6
Sch em e 7
Structure assignments of the products resulting from
these studies are based upon characteristic spectral
properties, as well as a complete X-ray crystal structure
of 28. Knowing the structure of 28 allows the unambigu-
ous assignment of the structure of its regioisomer 29.
Once again, the major isomer, 28, is the one arising from
allylation at the benzylic site better able to stabilize the
cationic charge in the intermediate. By analogy, the
structure of 25 is assigned, i.e., allylation takes place at
the benzylic position as opposed to the propargylic site.
Further structural information is gleaned through analy-
sis of the thermolysis products and will be discussed
later.
Significant compromised selectivity was observed for
the allylation of the 3-alkyl-4-hexynyl-4-hydroxy-2-phen-
ylcyclobutenones 30a and 30b (Scheme 7). The former
gave 31a and 32a in a respective ratio of 1:1.3, and the
latter gave 31b and 32b in a ratio of 1.1:1. For these
cases, formation of both regioisomers was not anticipated
because, as previously described, the closely related
4-hexynyl-4-hydroxy-2-phenyl-3-isopropoxycyclobuten-
one 23 gave nearly a quantitative yield of only 25
(Scheme 6). Clearly, the 3-alkyl groups of 30a ,b, as
compared to the 3-alkoxy group in 23, cause a significant
reduction in the allylation selectivity. This is likely due
to greater steric congestion for attack at the benzylic
position in, for example, carbenium ion intermediate 33
as compared to 34 because alkyl groups are sterically
with few exceptions, are those resulting from attack of
the silane at the position bearing the aryl group, i.e., the
site better able to support the positive charge.4,5 For
example, allylation of differentially substituted hydroxy-
cyclobutenones 17a -c gave the corresponding 4-allyl-4-
aryl-2-hexynyl-3-isopropoxycyclobutenones 20a-c in yields
ranging from 95% to 96%. Analogously, 17h -j gave only
3-alkoxy-2-alkyl-4-allyl-4-phenylcyclobutenones 20h -j in
70-98% yields.
Further details concerning the mechanism of the
allylation reactions were obtained when the regioisomeric
cyclobutenones 22 and 23 were subjected to the allylation
conditions (Scheme 6). Both were observed to give a
nearly quantitative yield of a single allylated cyclobuten-
one 25, a result in agreement with a mechanism involv-
ing the common intermediacy of cyclobutenonium ion 24.
The same conclusion was drawn from a related study
starting with the regioisomeric diarylcyclobutenones 26
and 27, which were observed to give a mixture of the
allylated regioisomers 28 and 29 in ratios (averaged) of
4.0:1.0 and 3.5:1.0, respectively, as determined by 1H
NMR analysis of the crude products for two different
runs.
more encumbering than alkoxy substituents (e.g., AMe
)
1.74 kcal/mol and At-OBu ) 0.75 kcal/mol).9 Apparently,
this difference is great enough to allow the seemingly less
stable propargylic site in the carbocations arising from
31a ,b to favorably compete with the benzylic position for
allylation, a situation not observed for 3-alkoxy analogues
22 and 23.
(9) Eliel, E.; Wilen, S. H.; Mander, L. N. Stereochemistry of Carbon
Compounds; J ohn Wiley and Sons: New York, 1994; pp 695-696.

Synthesis and Thermolysis of 4-Allyl-4-arylcyclobutenones
J . Org. Chem., Vol. 64, No. 11, 1999 4033
Sch em e 8
Sch em e 9
Further mechanistic information is gleaned from the
qualitative relative allylation rates listed in Schemes
4-7. Reaction rates generally correlate with carbocation
stability and thus suggest ionization as the rate-limiting
step of the reaction. The following specific examples are
illustrative of this point: (1) 4-Hydroxy-3-methoxy-2,4-
diphenylcyclobutenone (17d ) reacts nearly twice as fast
as its 3-isopropoxy analogue 17e. The smaller methoxy
group imparts less steric constraints, and thus both
phenyl groups can participate more fully in cation
stabilization, i.e., coplanarity of both phenyl groups is
conceivably allowed for 18d but not for 18e. (2) 4-Hy-
droxy-3-isopropoxy-4-(4-methoxyphenyl)-2-phenylcy-
clobutenone (26) reacts approximately twice as fast as
its regioisomer 27. Here again, coplanarity of both aryl
groups is apparently partially impeded during the ioniza-
tion step as a result of steric interaction of the isopropyl
group with the 2-aryl substituent. Thus, formation of the
cation from 26 gains more assistance from the 4-meth-
oxyphenyl group than does the ionization of 27. (3)
3-Isopropoxy-2,4-di(4-methoxyphenyl)cyclobutenone (17f)
reacts approximately 24 times faster than the 2,4-
diphenyl- (17e) or 2,4-di(3-methoxyphenyl)- (17g) ana-
logues. (4) 2-Hexynyl-3-isopropoxy-4-(4-methoxyphenyl)-
cyclobutenone (17b) and 2-hexynyl-3-isopropoxy-4-(3,4-
dimethoxyphenyl)-cyclobutenone (17c) react 8-12 times
faster than the 4-(3-methoxyphenyl) analogue 17a .
Similar relative rate comparisons are summarized in
Scheme 8, in which four selected cyclobutenones (17e,
17h , 22, and 17l) representing the general structural
types are presented along with their respective qualita-
tive relative allylation rates. Steric inhibition for copla-
narity of the two phenyl groups during the ionization step
would account for 17e having the slowest rate. This is,
however, of particular note in comparison to 17l, which
reacts more than 500 times faster even though nonco-
planarity of the phenyl groups should be even more
pronounced because the isobutyl group is sterically more
encumbering than an isopropoxy group. Apparently, the
controlling factor is that cyclobutenonium ions such as
35 bearing σ-donating 3-alkyl groups are more stable
than the corresponding examples bearing σ-withdrawing
3-alkoxy groups, e.g., 36. This is, of course, expected if
structures 35 and 36 best represent the hybrid structure,
because the cationic center would not experience direct
π-donation by the cross-conjugated alkoxy group. How-
ever, if pseudoaromatic cyclobutadienium dication struc-
tures such as 37 made major contributions to the hybrid,
direct π-donation of the alkoxy group would be important
and 36 would be expected to be formed at a faster rate
than 35. The fact that this is not the case argues against
significant contributions from structures such as 37.
These data favor a mechanism of allylation in which
the rate-limiting step is carbocation formation. Indeed,
this possibility was confirmed in one case by a rate/
concentration study that showed the reaction rate to be
independent of the allylsilane concentration. Specifically,
the rate of allylation of 17e was observed to be unaffected
when the concentration of allylsilane was varied. As
summarized in Scheme 9, the ratio of starting cy-
clobutenone 17e to product 20e at ambient temperature
remained essentially constant (1H NMR analysis) over a
period of 48 h, even though the equivalents of allyltri-
methylsilane and BF3‚Et2O were varied from 5 to 20. On
the basis of these results, it is assumed that all other
examples behave analogously.
In summary, 4-aryl-4-hydroxycyclobutenones serve as
valuable precursors to allylated derivatives. The data
provided above not only point to the synthetic utility of
this reaction but also support a mechanism in which
Lewis acid induced ionization of the 4-hydroxy group in
4-aryl-4-hydroxycyclobutenones is the rate-limiting step.
Allylation then arises via the kinetically faster trapping
of the cyclobutenonium ion intermediate. In the absence
of steric encumbrances, allylation of such hybrid cations
ions takes place at the site better able to stabilize a
positive charge.
F u r th er Stu d ies Con cer n in g th e Syn th etic Scop e
Danheiser, Dixon, and Gleason reported the reaction
of enones such as 38 with allyltriisopropylsilane in the
presence of TiCl4 to give cyclopentenones 40 via intramo-
lecular trapping of the cationic center in 39 by the enol
moiety (Scheme 10).10 It was reasoned that under similar
conditions the bulky silicon-stabilized cation resulting
from the reaction of 4-aryl-4-hydroxycyclobutenones with
allyltriisopropylsilane would be intercepted by the proxi-
mal aromatic group, to give spiro-fused cyclobutenones.
This reaction was initially studied by treating cyclobuten-
one 17c with allyltriisopropylsilane and BF3‚Et2O. Under
these conditions, a 90% yield of a 1:1.1 mixture of,
respectively, the spiroannulated cyclobutenone 42 and
the allylated product 20c was realized. This ratio changed
to 1:1.5 when TiCl4 was employed. The most efficient
(10) Danheiser, R. L.; Dixon, B. R.; Gleason, R. W. J . Org. Chem.
1992, 57, 6094.

4034 J . Org. Chem., Vol. 64, No. 11, 1999
Tiedemann et al.
Sch em e 10
Sch em e 11
expansion to synthetically useful ortho-quinone methide
intermediates.12,13
Th er m olysis Stu d ies
As noted earlier (Scheme 1), 4-allyl-4-arylcyclobuten-
ones 1 would be expected to undergo thermal electrocyclic
ring opening to the corresponding vinylketenes 2 and/or
4. On the basis of torquoselectivity arguments, one would
expect 2 to be preferentially formed, i.e., the outward
rotation of the aryl group is favored over an alkyl group.14
This could then lead to the prediction that intramolecular
[2 + 2] cycloaddition to give bicyclo[3.2.0]heptenones
would be favored over the electrocyclic ring closure to
naphthols. However, as it was previously shown that
vinylketenes can equilibrate with their precursor cyclo-
butenones, it is difficult to predict product distribution
in the absence of experimental data.5 To probe this in
detail, the thermolysis of a number of the cyclobutenones
synthesized herein were investigated, and the data are
provided below.
In general, it was found that 4-allyl-4-arylcyclobuten-
ones undergo facile rearrangements in refluxing toluene
to give mixtures of the corresponding naphthols and
bicyclo[3.2.0]heptenones. For example, 25 and 20b-l
gave good to excellent yields (83-97%) of mixtures of the
corresponding naphthols 45 and bicyclo[3.2.0]heptenones
46 (Scheme 12). Among these examples slight, but
significant, selectivities in product formation were ob-
served. Specifically, 25, 20b, and 20c constitute a simi-
larly substituted series in which thermolysis results in
the preferential formation of the naphthol 45a -c over
the bicyclo[3.2.0]heptenone 46a -c in respective ratios
varying from 1.5:1 to 3:1. This preference is reversed for
those cyclobutenones bearing 2-aryl and 3-alkoxy groups,
i.e., 20d -f gave 45d -f and 46d -f in ratios varying from
1:2 to 1:2.6. Interestingly, the naphthol:bicyclo[3.2.0]-
heptenone ratio reverts to one favoring the naphthol
(1.3:1 to 2:1) for 20k ,l, both of which are 3-alkyl-2-
phenylcyclobutenones as compared to the 3-alkoxy ana-
logues.
spiroannulation was realized when AlCl3 was employed,
resulting in an 85% yield of 42 and 20c in a ratio of 4:1.
The spiroannulation appears to be limited to cyclobuten-
ones bearing electron-rich aryl groups at position 4. For
example, when 17b and 22 were treated with allyl-
triisopropylsilane/AlCl3, the products detected were the
4-allyl derivatives 20b and 25, respectively, in yields
ranging from 40% to 73%.
Finally, the results of a limited study of the Lewis acid
promoted reactions of propargyltrimethylsilane with
4-aryl-4-hydroxycyclobutenones are summarized in
Scheme 11. This study was directed to the synthesis of
4-allenylcyclobutenones and is based on the well-estab-
lished route to allenes arising from electrophilic substitu-
tion of propargylsilanes.11 The reaction proceeds as
expected starting with 4-hydroxycyclobutenones 22, 17c,k,
and 43 to give the corresponding 4-allenylcyclobutenones
44a -d , members of a class of compounds of interest
because they have been shown to undergo thermal ring
Conclusions drawn from these data include the follow-
ing: 1) Naphthol formation increases as the ketene
(13) For reviews on the chemistry of quinone methides see: Veciana,
J .; Martinez, A. D.; Armet, O. Rev. Chem. Intermed. 1988, 10(1), 35-
70. Volod′kin, A. A.; Ershov, V. V. Usp. Khim. 1988, 57(4), 595-624.
Also see the following and references therein: Angle, S. R.; Arnaiz, D.
O. J . Org. Chem. 1990, 55, 3708-10. Angle, S. R.; Turnbull, K. D. J .
Am. Chem. Soc. 1990, 112, 3698-700. Angle, S. R.; Yang, W. J . Am.
Chem. Soc. 1990, 112, 4524-8. Angle, S. R.; Louie, M. S.; Mattson, H.
L.; Yang, W. Tetrahedron Lett. 1989, 30, 1193-6. Angle, S. R.;
Turnbull, K. D. J . Am. Chem. Soc. 1989, 111, 1136-8.
(11) For reviews see: (a) Fleming, I. In Comprehensive Organic
Synthesis; Heathcock, C. H., Ed.; Pergamon Press: Oxford, 1991; Vol.
2, pp 563-593. (b) Schinzer, D. Synthesis 1988, 263.
(14) Houk, K. N.; Rondan, N. G. J . Am. Chem. Soc. 1985, 107, 2099.
Niwayama, S.; Kallel, E. A.; Sheu, C.; Houk, K. N. J . Org. Chem. 1996,
61, 2517.
(12) Taing, M.; Moore, H. W. J . Org. Chem. 1996, 61, 329.

Synthesis and Thermolysis of 4-Allyl-4-arylcyclobutenones
J . Org. Chem., Vol. 64, No. 11, 1999 4035
Sch em e 12
Sch em e 13
Sch em e 14
becomes more electrophilic. Thus, assuming the reso-
nance structures in Scheme 12 adequately represent the
ketene,15 the alkynyl ketenes 47 should be more electro-
philic than the corresponding aryl ketenes, i.e., anion
delocalization into the alkynyl groups would be favored.
Additionally, the 2-aryl-1-methoxy-1,5-pentadienyl group
would be a stronger electron-withdrawing group (EWG)
than the 1-methyl analogue and thereby would increase
the electrophilic nature of the respective ketene arising
from 3-alkoxy- vs 3-alkylcyclobutenones. (2) Naphthol
formation is also increased as the 4-aryl group increases
in nucleophilic character. This is particularly true if the
π-donating methoxy group is in position 3 of the aryl
substituent, i.e., R3 ) OMe. In this regard, it is notewor-
thy that 20c gives a 3:1 ratio of naphthol 45c to bicyclo-
[3.2.0]heptenone 46c. Even higher selectivity was ob-
served for the thermolysis of 4-allyl-2-hexynyl-3-isopro-
poxy-4-(3-methoxyphenyl)-cyclobutenone 20a (Scheme
13). Here the ratio of naphthols 48 and 49 to bicyclo-
[3.2.0]heptenone 50 was 3.2:1.4:1. It is of interest to note
that both regioisomeric naphthols 48 and 49 were formed,
the former from attack of the ketene para to the methoxy
group and the latter from ortho attack.
and bicyclo[3.2.0]heptenones. The former gave the re-
spective products (51:55) in a ratio of 1:1.6, and the latter
gave a ratio (52:56) of 1:1 (Scheme 14). It was necessary
to determine these ratios by 1H NMR of the crude product
mixture because the naphthols readily undergo oxidation
to the corresponding hydroperoxides 53 and 54 (24%)
upon attempted purification.
Thermolysis (toluene, 110 °C) of 2-alkylcyclobutenones
20h ,i also gave mixtures of the corresponding naphthols
(15) For a discussion of ketene stability and reactivity see: (a) Gong,
L.; McAllister, M. A.; Tidwell, T. T. J . Am. Chem. Soc. 1991, 113, 6021.
(b) Tidwell, T. T. Ketenes; J ohn Wiley and Sons: New York, 1995.

4036 J . Org. Chem., Vol. 64, No. 11, 1999
Tiedemann et al.
Sch em e 15
Sch em e 16
The highest selectivity was observed when 4-allyl-2-
tert-butyl-3-methoxy-4-phenylcyclobutenone 20j was sub-
jected to the above thermolysis conditions. Here, an 82%
yield of a mixture of the keto tautomer of the naphthol
57 (13%) and the bicyclo[3.2.0]heptenone 58 (69%) was
realized. In this case, the bulky tert-butyl group appar-
ently disfavors the planar or helical transition state
associated with the electrocyclic ring closure as compared
to the nonplanar transition state associated with the
orthogonal arrangement required for an assumed con-
certed intramolecular [2 + 2] ketene alkene cycloaddition.
That the keto tautomer 57 is isolated rather than the
aromatic naphthol is of interest and is apparently due
to less steric strain associated with having the bulky tert-
butyl group attached to an sp3 hybridized carbon atom.
A modification of the 4-allyl group was observed to
have some influence on the selectivity of the thermal ring
expansion. Specifically, 21a -d all bear a 4-(2-methyl-
allyl) group. The methyl group is so placed that it would
stabilize presumed cationic or radical character in the
transition state for a [2 + 2] cycloaddition but should
have little influence on the competitive electrocyclic ring-
closure process.3 Thus, one would anticipate an increase
in the amount of the bicyclo[3.2.0]heptenones relative to
the naphthols in the thermolysis of 21a -d as compared
to their respective desmethyl analogues, 20b,e,f,j. This
was not observed for 21a , which again would result in
the more electrophilic vinylketene, i.e., the ratio of 59a
to 60a was 1.4:1 as compared to 1.5:1 for 20b (Scheme
15). However, a clear increase in the expected selectivity
was observed for the remaining examples: 21b gave 59b
and 60b in a respective ratio of 1:4 (cf. 1:2.1 for 20e); for
21c the ratio 59c to 60c was 1:4.3 (cf. 1:2 for 20f); and
for 21j the only product detected was the bicyclo[3.2.0]-
heptenone 60d .
Sch em e 17
naphthols or to reactive o-quinone methide intermedi-
ates.13 To gain some insight into these possibilities, 44d
was subjected to thermolysis in refluxing acetonitrile
containing methanol (Scheme 16). In this case, the only
product isolated was the phenol 62 (55%), a product that
must arise through methanol trapping of the o-quinone
methide 61.
The above thermolyses employ cyclobutenones bearing
aryl and allyl groups at position 4. In general, such
compounds lead to competitive reactions in which the
intermediate vinylketenes undergo electrocyclic ring
closure to provide the corresponding naphthol or proceed
to bicyclo[3.2.0]heptenones via intramolecular [2 + 2]
ketene/alkene cycloadditions. An example in which the
competition is between an alkynyl group and an allyl
group is outlined in Scheme 17. Here, 32a was subjected
to thermolysis in refluxing toluene, and the only detected
product was the bicyclo[3.2.0]heptenone 63 (89%), the
product arising from the intramolecular cycloaddition of
Previously, we reported that 4-allenyl-4-hydroxycy-
clobutenones undergo facile thermal ring expansion to
o-quinone methide intermediates, which can be trapped
via intra- or intermolecular processes.12,13 The synthetic
and pharmacological potential of this ring expansion is
significantly expanded because previously unknown 4-al-
lenyl-4-arylcyclobutenones are now readily available as
outlined in Scheme 11. Such compounds would, of course,
also have the potential for competitive ring expansion to

Synthesis and Thermolysis of 4-Allyl-4-arylcyclobutenones
J . Org. Chem., Vol. 64, No. 11, 1999 4037
Sch em e 18
(electrocyclic ring-closure mode). Although both reaction
pathways are operative, selectivity can be slightly en-
hanced by a judicious choice of substituents.
Exp er im en ta l Section
Gen er a l Meth od s a n d Ma ter ia ls. All reactions were
carried out in flame-dried glassware under a positive pressure
of dry nitrogen. Tetrahydrofuran (THF) was dried by passing
it through two 4 × 36 in.2 columns of anhydrous neutral A-2
alumina to remove H2O. Toluene and CH2Cl2 were distilled
from calcium hydride. All solids were recrystallized from CH2-
Cl2/pentane. Removal of solvents was accomplished on a rotary
evaporator at approximately 20 Torr. All reactions were
followed by TLC using E. Merck silica gel 60 F-254 coated on
glass. Flash column chromatography was performed using E.
Merck silica gel (230-400 mesh). Radial chromatography was
performed on rotors coated with Merck 7749 silica gel. 1H NMR
(300 and 500 MHz) and 13C NMR (75 and 125 MHz) spectra
were recorded in CDCl3 as solvent; coupling constants are
given in hertz. Melting points are uncorrected.
Rep r esen ta tive P r oced u r e for th e Allyla tion of 4-Hy-
dr oxycyclobu ten on es. 4-Allyl-2-(1-h exyn yl)-3-isopr opoxy-
4-(3-m eth oxyph en yl)-2-cyclobu ten on e (20a). 2-(1-Hexynyl)-
4-hydroxy-3-isopropoxy-4-(3-methoxyphenyl)-2-cyclobuten-
one (17a ) (0.164 g, 0.5 mmol, 1 equiv) was dissolved in dry
CH2Cl2 (8 mL) under a positive pressure of N2. The resulting
solution was treated with allyltrimethylsilane (0.40 mL, 2.5
mmol, 5.0 equiv) and cooled to ca. -5 °C prior to dropwise
addition of BF3‚OEt2 (0.31 mL, 2.5 mmol, 5.0 equiv). The ice
bath was removed after 15 min, and after an additional 1.75
h, the reaction mixture was quenched with saturated aqueous
sodium bicarbonate (3 mL). Further dilution with CH2Cl2 (10
mL) was followed by extraction. The organic layer was washed
with saturated aqueous sodium chloride (5 mL), and the
combined aqueous layers were extracted with CH2Cl2 (2 × 10
mL). The combined organic portions were dried (MgSO4) and
concentrated to give a yellow oil. Flash silica gel chromato-
graphy (9:1 hexanes/EtOAc) provided 0.167 g (95%) of 20a as
a yellow oil: 1H NMR (500 MHz, CDCl3) δ 7.22 (t, J ) 8.0,
1H), 7.01-6.97 (m, 2H), 6.79-6.77 (m, 1H), 5.80-5.71 (m, 1H),
5.28 (septet, J ) 6.1, 1H), 5.09 (d, J ) 18.1, 1H), 5.08 (d, J )
9.5, 1H), 3.78 (s, 3H), 2.71 (dd, J ) 14.3, J ) 6.8, 1H), 2.60
(dd, J ) 14.3, J ) 8.0, 1H), 2.30 (t, J ) 7.0, 2H), 1.53-1.44
(m, 8H), 1.38 (hextet, J ) 7.3, 2H), 0.89 (t, J ) 7.3, 3H); 13C
NMR (125 MHz, CDCl3) δ 187.5, 181.2, 159.5, 140.4, 132.7,
129.4, 118.6, 118.1, 112.6, 111.4, 102.5, 92.3, 78.2, 69.8, 67.7,
55.0, 38.5, 30.4, 22.3, 22.3, 21.8, 19.0, 13.4; IR (neat) 1767,
1606 cm-1; HRMS (CI) calcd for C23H28O3 (M+) 352.2038, found
352.2037.
2-(1-Hexyn yl)-3-isop r op oxy-4-(4-m eth oxyp h en yl)-4-(2-
m eth yla llyl)-2-cyclobu ten on e (21a ). Prepared in the same
manner as described for 20a . A CH2Cl2 solution of 2-(1-
hexynyl)-4-hydroxy-3-isopropoxy-4-(4-methoxyphenyl)-2-cy-
clobutenone (17b) (0.164 g, 0.5 mmol, 1 equiv) was treated with
(2-methylallyl)trimethylsilane (0.320 g, 2.5 mmol, 5 equiv)
followed by BF3‚OEt2 (0.31 mL, 2.5 mmol, 5 equiv) and stirred
at room temperature for 2.5 h. Workup and purification by
flash silica gel chromatography (12:1 hexanes/EtOAc) gave
0.179 g (98%) of 21a as a pale yellow oil: 1H NMR (500 MHz,
CDCl3) δ 7.35 (d, J ) 8.5, 2H), 6.83 (d, J ) 8.5, 2H), 5.27
(septet, J ) 6.1, 1H), 4.85 (s, 1H), 4.72 (s, 1H), 3.77 (s, 3H),
2.65 (d, J ) 14.0, 1H), 2.60 (d, J ) 14.0, 1H), 2.31 (t, J ) 7.0,
2H), 1.71 (s, 3H), 1.52-1.46 (m, 2H), 1.47 (d, J ) 6.1, 6H),
1.38 (hextet, J ) 7.3, 2H), 0.89 (t, J ) 7.3, 3H); 13C NMR (125
MHz, CDCl3) δ 188.4, 181.6, 158.6, 141.1, 131.5, 126.9, 115.4,
113.7, 102.4, 92.2, 78.2, 69.4, 67.9, 55.2, 42.6, 30.5, 23.7, 22.4,
22.3, 21.9, 19.1, 13.5; IR (neat) 1766, 1606 cm-1; HRMS (CI)
calcd for C24H30O3 (M+) 366.2195, found 366.2192.
vinylketene 65. Apparently, ring closure of ketene 64 to
the diradical 66, a known precursor to the quinone 67,
is kinetically unfavorable in comparison to the intramo-
lecular cycloaddition.1a
Finally, a transformation of potential synthetic impor-
tance as a route to phenalenones, useful precursors to a
variety of natural products and drug templates, is
outlined in Scheme 18.16,17 Here the mixture of diaster-
eomeric spirocyclobutenones 42 was subjected to the
standard thermolysis conditions (toluene, 110 °C), result-
ing in the formation of phenalene 68 in 60% isolated
yield. Oxidation using 2.1 equiv of DDQ afforded phe-
nalenone 69 in 50% yield. Significantly higher yields of
this product were realized when the initial thermolysate
was treated directly with the oxidizing agent to give 69
in 72% overall yield.
Con clu sion s
In conclusion, we wish to note the following significant
points arising from the study outlined herein: (1) The
thermal rearrangements of appropriately substituted
cycobutenones provides one of the most versatile routes
to highly condensed ring systems.1 The most obvious
limitation to this methodology is the availability of the
starting cyclobutenones. The work outlined here helps
to minimize this problem and adds to the rapidly growing
number of viable synthetic routes to such compounds.
Specifically, readily available 4-aryl-4-hydroxycyclobuten-
ones are efficiently allylated upon treatment with BF3‚
Et2O and allyltrialkylsilanes. (2) The selectivity of the
allylation is dependent upon the structure of the inter-
mediate cyclobutenonium cation. In general, allylation
takes place predominantly at that site better able to
stabilize the positive charge. (3) The 4-aryl-4-allylcy-
clobutenones available by this route undergo electrocyclic
ring opening to the corresponding intermediate vinyl-
ketenes. These, in turn, proceed to bicyclo[3.2.0]hepten-
ones ([2 + 2] cycloaddition mode) or to 4-allylnaphthols
(16) For a synthesis of a phenalenone-based natural product see:
Danheiser, R. L.; Helgason, A. L. J . Am. Chem. Soc. 1994, 116, 9471.
(17) For phenalenone-based drug templates see: Von Voigtlander,
P. F.; Athaus, J . S.; Ochoa, M. C.; Neff, G. L. Drug Dev. Res. 1989, 17,
71. Tang, A. H.; Franklin, S. R.; Code, R. A.; Athaus, J . S.; Von
Voigtlander, P. F.; Darlington, W. H.; Szmuszkaovicz, J . Drug Dev.
Res. 1990, 21, 53.
4-Allyl-2-(1-h exyn yl)-3-isop r op oxy-4-p h en yl-2-cyclo-
bu ten on e (25) fr om 22. Prepared as described for 20a . A CH2-
Cl2 solution of 2-(1-hexynyl)-4-hydroxy-3-isopropoxy-4-phenyl-
2-cyclobutenone (22) (0.503 g, 1.69 mmol, 1 equiv) was treated
with allyltrimethylsilane (1.34 mL, 8.44 mmol, 5 equiv)

4038 J . Org. Chem., Vol. 64, No. 11, 1999
Tiedemann et al.
followed by BF3‚OEt2 (1.04 mL, 8.44 mmol, 5 equiv) and stirred
at room temperature for 2 h. Workup and purification by flash
silica gel chromatography (9:1 hexanes/EtOAc) gave 0.513 g
(94%) of 25 as a pale yellow oil: 1H NMR (500 MHz, CDCl3) δ
7.43-7.41 (m, 2H), 7.33-7.30 (m, 2H), 7.26-7.24 (m, 1H),
5.81-5.72 (m, 1H), 5.30 (septet, J ) 6.5, 1H), 5.12-5.08 (m,
2H), 2.75-2.71 (m, 1H), 2.64-2.60 (m, 1H), 2.31 (t, J ) 7.0,
2H), 1.52-1.47 (m, 8H), 1.39 (hextet, J ) 7.3, 2H), 0.89 (t, J
) 7.3, 3H); 13C NMR (125 MHz, CDCl3) δ 187.8, 181.4, 139.0,
132.8, 128.4, 127.1, 125.8, 118.7, 102.4, 92.3, 78.3, 69.8, 67.8,
38.6, 30.5, 22.4, 22.3, 21.9, 19.0, 13.5; IR (neat) 1766, 1608
cm-1; HRMS (EI) calcd for C22H26O2 (M+) 322.1933, found
322.1936.
98.2, 73.2, 67.9, 37.6, 30.4, 22.0, 19.3, 14.5, 13.5; IR (neat) 2221,
1766, 1641, 1619 cm-1; HRMS (CI) calcd for C20H23O (MH+)
279.1749, found 279.1753.
4-Allyl-2-(1-h exyn yl)-3-isobu tyl-4-p h en yl-2-cyclobu ten -
on e (31b) a n d 4-Allyl-4-(1-h exyn yl)-3-isobu tyl-2-p h en yl-
2-cyclobu ten on e (32b). Prepared as described for 20a . A
CH2Cl2 solution of 4-(1-hexynyl)-4-hydroxy-3-isobutyl-2-phen-
yl-2-cyclobutenone (30b) (0.142 g, 0.48 mmol, 1 equiv) was
treated with allyltrimethylsilane (0.38 mL, 2.4 mmol, 5 equiv)
followed by BF3‚OEt2 (0.3 mL, 2.4 mmol, 5 equiv) and stirred
at -5 °C for 5 min. Workup was followed by 1H NMR analysis
which showed a 1.1:1 ratio of products favoring 31b over 32b.
Radial chromatography (100:1 hexanes/EtOAc) gave as first
fraction 0.050 g (33%) of 32b as a yellow oil: 1H NMR (500
MHz, CDCl3) δ 7.68 (d, J ) 7.5, 2H), 7.38 (t, J ) 7.4, 2H), 7.32
(t, J ) 7.1, 1H), 5.94-5.86 (m, 1H), 5.14 (d, J ) 16.9, 1H),
5.10 (d, J ) 10.1, 1H), 2.81 (dd, J ) 14.0, J ) 8.7, 1H), 2.66-
2.61 (m, 3H), 2.39 (septet, J ) 6.6, 1H), 2.22 (t, J ) 7.0, 2H),
1.47 (quintet, J ) 7.0, 2H), 1.39 (hextet, J ) 7.0, 2H), 1.07 (d,
J ) 6.6, 6H), 0.89 (t, J ) 7.1, 3H); 13C NMR (125 MHz, CDCl3)
δ 188.6, 175.6, 144.3, 133.4, 129.7, 128.7, 128.5, 127.3, 118.1,
87.6, 77.5, 63.7, 39.4, 38.7, 30.7, 27.2, 23.7, 23.0, 21.9, 18.6,
13.5; IR (neat) 1764, 1634, 1598 cm-1; HRMS (CI) calcd for
4-Allyl-3-isop r op oxy-4-(4-m eth oxyp h en yl)-2-p h en yl-2-
cyclobu ten on e (28) a n d 4-Allyl-3-isop r op oxy-2-(4-m eth -
oxyp h en yl)-4-p h en yl-2-cyclobu ten on e (29) fr om 26. Pre-
pared as described for 20a . A CH2Cl2 solution of 4-hydroxy-
3-isopropoxy-4-(4-methoxyphenyl)-2-phenyl-2-cyclobutenone (26)
(0.324 g, 1.0 mmol, 1 equiv) was treated with allyltrimethyl-
silane (0.80 mL, 5 mmol, 5 equiv) followed by BF3‚OEt2 (0.61
mL, 5 mmol, 5 equiv) and stirred at room temperature for 6
1
h. Workup was followed by H NMR analysis which showed a
4.3:1 ratio of products favoring 28 over 29. This ratio was
observed to be 3.7:1 in another run, giving an average of 4.0:
1. Radial chromatography (30:1 hexanes/EtOAc) gave as first
fraction 0.064 g (18%) of 29 as a white crystalline solid: mp
108 °C; 1H NMR (500 MHz, CDCl3) δ 7.62 (d, J ) 8.6, 2H),
7.47 (d, J ) 7.7, 2H), 7.37 (t, J ) 7.7, 2H), 7.28 (d, J ) 7.3,
1H), 6.90 (d, J ) 8.6, 2H), 5.97-5.88 (m, 1H), 5.23 (d, J )
16.8, 1H), 5.15 (d, J ) 10.1, 1H), 4.58 (septet, J ) 6.1, 1H),
3.82 (s, 3H), 3.13 (dd, J ) 14.5, J ) 6.5, 1H), 2.84 (dd, J )
14.5, J ) 7.7, 1H), 1.39 (d, J ) 6.1, 3H), 1.12 (d, J ) 6.1, 3H);
13C NMR (125 MHz, CDCl3) δ 186.5, 175.9, 158.8, 138.7, 133.3,
128.7, 128.6, 127.4, 126.2, 121.9, 121.3, 118.7, 113.8, 78.0, 71.5,
55.3, 36.3, 23.0, 22.7; IR (neat) 1749, 1632, 1601 cm-1; HRMS
(CI) calcd for C23H24O3 (M+) 348.1725, found 348.1728.
Second fraction, 0.278 g (80%) of 28 as a pale yellow
crystalline solid: mp 115 °C; 1H NMR (500 MHz, CDCl3) δ
7.72 (d, J ) 7.6, 2H), 7.41 (d, J ) 8.7, 2H), 7.38 (t, J ) 7.7,
2H), 7.27 (t, J ) 7.4, 1H), 6.92 (d, J ) 8.7, 2H), 5.99-5.91 (m,
1H), 5.26 (d, J ) 16.8, 1H), 5.17 (d, J ) 10.2, 1H), 4.60 (septet,
J ) 6.1, 1H), 3.80 (s, 3H), 3.15 (dd, J ) 14.6, J ) 6.5, 1H),
2.86 (dd, J ) 14.6, J ) 7.7, 1H), 1.41 (d, J ) 6.1, 3H), 1.13 (d,
J ) 6.1, 3H); 13C NMR (125 MHz, CDCl3) δ 186.7, 177.5, 158.8,
133.2, 130.2, 129.1, 128.3, 127.3, 127.2, 126.9, 121.2, 118.7,
114.0, 78.2, 71.1, 55.1, 36.0, 22.9, 22.6; IR (neat) 1747, 1642,
1632, 1597 cm-1; HRMS (CI) calcd for C23H24O3 (M+) 348.1725,
found 348.1726.
C
23H28O (M+) 320.2140, found 320.2142.
Second fraction, 0.059 g (38%) of 31b as a yellow oil: 1H
NMR (500 MHz, CDCl3) δ 7.35-7.23 (m, 5H), 5.78-5.70 (m,
1H), 5.12 (d, J ) 17.1, 1H), 5.08 (d, J ) 10.1, 1H), 2.86 (dd, J
) 14.5, J ) 7.0, 1H), 2.75 (dd, J ) 14.5, J ) 7.5, 1H), 2.43 (d,
J ) 7.1, 2H); 2.38 (t, J ) 7.1, 2H), 2.34 (septet, J ) 6.7, 1H),
1.54 (quintet, J ) 7.0, 2H), 1.43 (hextet, J ) 7.2, 2H), 1.00 (d,
J ) 6.7, 3H), 0.99 (d, J ) 6.7, 3H), 0.92 (t, J ) 7.3, 3H); 13C
NMR (125 MHz, CDCl3) δ 190.4, 184.0, 138.8, 133.3, 129.8,
128.6, 127.2, 126.2, 118.6, 97.8, 73.0, 68.8, 39.3, 37.2, 30.3, 25.8,
23.2, 23.2, 21.9, 19.3, 13.5; IR (neat) 2224, 1768, 1642, 1609
cm-1; HRMS (CI) calcd for C23H28O (M+) 320.2140, found
320.2138.
4-Allyl-3-isopr opoxy-2,4-diph en yl-2-cyclobu ten on e (20e).
Kin etic Stu d ies. 4-Hydroxy-3-isopropoxy-2,4-diphenyl-2-cy-
clobutenone (17e) (0.147 g, 0.5 mmol, 1 equiv) was dissolved
in dry CH2Cl2 (8 mL) under a positive pressure of N2. The
resulting solution was treated with allyltrimethylsilane (X
equiv, X ) 5, 20) and cooled to ca. -5 °C prior to dropwise
addition of BF3‚Et2O (Y equiv, Y ) 5, 20). The ice bath was
removed after 15 min. After 30 min and 1, 4, 10, 24, and 48 h,
aliquots were taken out of the reaction mixture. Workup of
the aliquots was followed by 1H NMR analysis (see Scheme
9).
Sp ir o[3-(1-h exyn yl)-4-isop r op oxy-2-oxo-3-cyclobu ten e-
1,1′(2′H)-3′,4′-d ih yd r o-6′,7′-d im eth oxyn a p h th a len e] (42)
a n d 4-Allyl-2-(1-h exyn yl)-3-isop r op oxy-4-(3,4-d im eth oxy-
p h en yl)-2-cyclobu ten on e (20c). Prepared in the same man-
ner as described for 20a . A CH2Cl2 solution of 2-(1-hexynyl)-
4-hydroxy-3-isopropoxy-4-(3,4-dimethoxyphenyl)-2-cyclobuten-
one (17c) (0.179 g, 0.5 mmol, 1 equiv) was treated with
allyltriisopropylsilane (0.36 mL, 1.5 mmol, 3 equiv) followed
by AlCl3 (200 mg, 1.5 mmol, 3 equiv) and stirred at -5 °C for
10 min. Workup gave a yellow oil. 1H NMR analysis of the
crude product showed a 3.7:1 ratio of products favoring 42 over
20c. Purification by flash silica gel chromatography (4:1
hexanes/EtOAc) gave as first fraction 0.181 g (67%) of 42 as a
pale yellow amorphous solid (mixture of diastereomers 1.2:
1*): 1H NMR (500 MHz, CDCl3) δ 6.69, 6.65* (each s, 1H),
6.60*, 6.58 (each s, 1H), 5.33-5.20 (m, 1H), 3.83, 3.82, 3.80
(each s, 6H), 2.86-2.71 (m, 2H), 2.362, 2.360* (each t, J ) 7.0,
2H), 2.21-1.86 (m, 2H), 1.58-1.36 (m, 10H), 1.22-1.08 (m,
22H), 0.92 (t, J ) 7.3, 3H); 13C NMR (125 MHz, CDCl3) δ 190.4,
188.7, 184.9, 182.4, 148.4, 148.3, 147.4, 147.3, 132.9, 132.4,
123.9, 123.6, 111.8, 111.5, 109.4, 108.9, 103.5, 92.6, 78.1, 77.7,
68.2, 68.0, 66.6, 66.3, 56.1, 55.9, 55.7, 55.7, 31.9, 31.8, 31.7,
31.5, 30.5, 22.1, 22.1, 21.9, 21.9, 19.1, 19.1, 19.0, 17.5, 17.3,
13.5, 10.9, 10.8; IR (neat) 1766, 1606 cm-1; HRMS (EI) calcd
for C33H50O4Si (M+) 538.3478, found 538.3487.
4-Allyl-2-(1-h exyn yl)-3-m eth yl-4-p h en yl-2-cyclobu ten -
on e (31a ) a n d 4-Allyl-4-(1-h exyn yl)-3-m eth yl-2-p h en yl-2-
cyclobu ten on e (32a ). Prepared as described for 20a . A
CH2Cl2 solution of 4-(1-hexynyl)-4-hydroxy-3-methyl-2-phenyl-
2-cyclobutenone (30a ) (0.254 g, 1.0 mmol, 1 equiv) was treated
with allyltrimethylsilane (0.80 mL, 5.0 mmol, 5 equiv) followed
by BF3‚OEt2 (0.61 mL, 5.0 mmol, 5 equiv) and stirred at -5
°C for 5 min. Workup was followed by 1H NMR analysis which
showed a 1.3:1 ratio of products favoring 32a over 31a . Radial
chromatography (200:1 hexanes/EtOAc) gave as first fraction
0.149 g (54%) of 32a as a yellow oil: 1H NMR (500 MHz,
CDCl3) δ 7.66 (d, J ) 7.5, 2H), 7.39 (t, J ) 7.5, 2H), 7.32 (t, J
) 7.5, 1H), 5.91-5.83 (m, 1H), 5.14 (dd, J ) 16.8, J ) 1.0,
1H), 5.09 (d, J ) 10.3, 1H), 2.60 (d, J ) 7.2, 2H), 2.43 (s, 3H),
2.23 (t, J ) 7.0, 2H), 1.51-1.45 (m, 2H), 1.40 (hextet, J ) 7.3,
2H), 0.90 (t, J ) 7.3, 3H); 13C NMR (125 MHz, CDCl3) δ 188.0,
172.5, 143.5, 133.4, 129.8, 128.7, 128.6, 127.2, 118.2, 87.8, 76.8,
64.3, 39.0, 30.9, 21.9, 18.6, 13.8, 13.6; IR (neat) 1764, 1639,
1599 cm-1; HRMS (CI) calcd for C20H22O (M+) 278.1671, found
278.1671.
Second fraction, 0.112 g (40%) of 31a as a yellow oil: 1H
NMR (500 MHz, CDCl3) δ 7.35-7.24 (m, 5H), 5.77-5.68 (m,
1H), 5.12 (d, J ) 17.0, 1H), 5.08 (d, J ) 10.1, 1H), 2.85 (dd, J
) 14.6, J ) 6.9, 1H), 2.74 (dd, J ) 14.6, J ) 7.4, 1H), 2.38 (t,
J ) 7.1, 2H), 2.28 (s, 3H), 1.58-1.52 (m, 2H), 1.43 (hextet, J
) 7.4, 2H), 0.92 (t, J ) 7.4, 3H); 13C NMR (125 MHz, CDCl3)
δ 189.8, 181.3, 139.0, 133.3, 130.9, 128.6, 127.2, 126.1, 118.5,
Second fraction, 34 mg (18%) of 20c as a yellow oil.
Rep r esen ta tive P r oced u r e for th e Allen yla tion of
4-Hyd r oxycyclobu ten on es. 4-Allen yl-2-(1-h exyn yl)-3-iso-

Synthesis and Thermolysis of 4-Allyl-4-arylcyclobutenones
J . Org. Chem., Vol. 64, No. 11, 1999 4039
p r op oxy-4-p h en yl-2-cyclobu ten on e (44a ). 2-(1-Hexynyl)-
4-hydroxy-3-isopropoxy-4-phenyl-2-cyclobutenone (22) (0.107
g, 0.36 mmol, 1 equiv) was dissolved in dry CH2Cl2 (5 mL)
under a positive pressure of N2. The resulting solution was
treated with propargyltrimethylsilane (0.16 mL, 1.08 mmol,
3.0 equiv) and cooled to ca. -5 °C prior to the dropwise addition
of BF3‚OEt2 (0.13 mL, 1.08 mmol, 3.0 equiv). The ice bath was
removed after 15 min, and after an additional 2.75 h, the
reaction mixture was quenched with saturated aqueous so-
dium bicarbonate (10 mL). Further dilution with CH2Cl2 (5
mL) was followed by extraction. The organic layer was washed
with saturated aqueous sodium chloride (5 mL), and the
combined aqueous layers were extracted with CH2Cl2 (2 × 10
mL). The combined organic portions were dried (MgSO4) and
concentrated to give a brown oil. Flash silica gel chromato-
graphy (19:1 hexanes/EtOAc) provided 0.080 g (70%) of 44a
as a pale yellow oil: 1H NMR (300 MHz, CDCl3) δ 7.44-7.41
(m, 2H), 7.35-7.28 (m, 3H), 5.48 (t, J ) 6.5, 1H), 5.30 (septet,
J ) 6.5, 1H), 4.86 (d, J ) 6.5, 2H), 2.33 (t, J ) 7.0, 2H), 1.54-
1.39 (m overlapping with d, J ) 6.0, 10H), 0.91 (t, J ) 7.0,
3H); 13C NMR (75 MHz, CDCl3) δ 208.4, 185.8, 180.3, 137.4,
128.4, 127.5, 126.4, 102.9, 93.0, 90.5, 78.6, 78.2, 67.8, 30.5, 22.3,
21.9, 19.1, 13.5; IR (neat) 1946, 1768, 1607 cm-1; HRMS (EI)
calcd for C22H24O2 (M+) 320.1776, found 320.1774.
22.1, 19.6, 13.6; IR (neat) 3485, 1626, 1580 cm-1; HRMS (CI)
calcd for C23H28O3 (M+) 352.2038, found 352.2031.
Second fraction, 0.028 g (17%) of 50 as a yellow oil: 1H NMR
(500 MHz, CDCl3) δ 7.46 (s, 1H), 7.24-7.22 (m, 2H), 6.79-
6.76 (m, 1H), 4.84 (septet, J ) 6.0, 1H), 3.82 (s, 3H), 3.39 (dd,
J ) 18.1, J ) 9.0, 1H), 3.26 (dd, J ) 15.5, J ) 7.9, 1H), 3.07
(dd, J ) 18.1, J ) 6.4, 1H), 2.89-2.83 (m, 1H), 2.69 (d, J )
15.5, 1H), 2.26 (t, J ) 7.0, 2H), 1.53-1.36 (m, 4H), 1.39 (d, J
) 6.0, 3H), 1.17 (d, J ) 6.0, 3H), 0.90 (t, J ) 7.2, 3H); 13C
NMR (125 MHz, CDCl3) δ 201.1, 159.3, 147.4, 136.8, 128.9,
119.4, 117.0, 112.6, 112.2, 90.0, 74.0, 73.7, 73.1, 55.1, 52.6, 37.1,
32.1, 30.7, 23.1, 22.9, 21.9, 18.8, 13.6; IR (neat) 1783, 1623,
1599 cm-1; HRMS (CI) calcd for C23H28O3 (M+) 352.2038, found
352.2029.
Third fraction, 0.040 g (25%) of 49 as a yellow oil: 1H NMR
(500 MHz, CDCl3) δ 10.08 (s, 1H); 7.48 (d, J ) 8.6, 1H), 7.30
(t, J ) 8.2, 1H), 6.73 (d, J ) 7.8, 1H), 5.95 (ddt, J ) 17.4, J )
9.9, J ) 5.6, 1H), 5.00 (dd, J ) 9.9, J ) 1.5, 1H), 4.93 (dd, J )
17.4, J ) 1.5, 1H), 4.88 (septet, J ) 6.1, 1H), 4.03 (s, 3H), 3.77
(dt, J ) 5.6, J ) 1.5, 2H), 2.58 (t, J ) 7.2, 2H), 1.70-1.64 (m,
2H), 1.53 (hextet, J ) 7.3, 2H), 1.32 (d, J ) 6.1, 6H), 0.96 (t,
J ) 7.3, 3H); 13C NMR (125 MHz, CDCl3) δ 156.3, 156.0, 154.5,
137.2, 134.5, 126.3, 118.8, 117.8, 115.2, 111.9, 103.4, 102.7,
99.1, 75.4, 73.9, 56.3, 31.1, 30.1, 22.5, 22.2, 19.8, 13.7; IR (neat)
3364, 1621, 1606, 1581 cm-1; HRMS (CI) calcd for C23H28O3
(M+) 352.2038, found 352.2036.
Rep r esen ta tive P r oced u r e for th e Th er m olysis of
4-Allylcyclob u t en on es. 4-Allyl-2-(1-h exyn yl)-3-isop r o-
p oxyn a p h t h ol (45a ) a n d 5-(1-Hexyn yl)-4-isop r op oxy-3-
p h en ylbicyclo[3.2.0]h ep t-3-en -6-on e (46a ). A solution of 25
(0.251 g, 0.78 mmol) in toluene (16 mL, 0.05 M) was heated to
reflux under a blanket of N2 for 3 h. Removal of the volatile
4-Allyl-2-n -bu tyl-3-isop r op oxyn a p h th ol (51), 4-Allyl-2-
n -b u t yl-4-h yd r op er oxy-3-isop r op oxy-4H -n a p h t h a len -1-
on e (53), a n d 5-n -Bu tyl-4-isop r op oxy-3-p h en ylbicyclo-
[3.2.0]h ep t-3-en -6-on e (55). A solution of 20h (0.186 g, 0.62
mmol) in toluene (12 mL, 0.05 M) was heated to reflux under
a blanket of N2 for 2 h. Removal of the volatile components
1
components was followed by H NMR analysis which showed
a 2:1 ratio of products favoring the naphthol over the bicyclo-
heptenone. Flash silica gel chromatography (50:1 hexanes/
EtOAc) gave as first fraction 0.127 g (51%) of naphthol 45a
as an off-white oil: 1H NMR (500 MHz, CDCl3) δ 8.19 (dd, J
) 8.5, J ) 0.5, 1H), 7.86 (d, J ) 8.5, 1H), 7.48 (ddd, J ) 8.0,
J ) 6.5, J ) 1.0, 1H), 7.40 (ddd, J ) 8.0, J ) 6.5, J ) 1.0, 1H),
6.41 (s, 1H), 6.41-5.96 (m, 1H), 5.04-4.95 (m, 2H), 4.77
(septet, J ) 6.0, 1H), 3.82-3.81 (m, 2H), 2.59 (t, J ) 7.0, 2H),
1.68 (quintet, J ) 7.5, 2H), 1.54 (hextet, J ) 7.5, 2H), 1.35 (d,
J ) 6.0, 6H), 0.99 (t, J ) 7.5, 3H); 13C NMR (125 MHz, CDCl3)
δ 153.9, 152.0, 137.2, 132.9, 127.3, 124.5, 123.9, 122.6, 120.5,
118.6, 115.3, 101.9, 100.7, 75.8, 72.9, 30.8, 29.7, 22.6, 22.1, 19.6,
13.6; IR (neat) 3487, 1620, 1574 cm-1; HRMS (CI) calcd for
1
was followed by H NMR analysis which showed a 1.6:1 ratio
of products favoring the bicycloheptenone 55 over the naphthol
51. Flash silica gel chromatography (19:1 hexanes/EtOAc) gave
as first fraction 0.104 g (56%) of bicycloheptenone 55 as a water
white oil: 1H NMR (500 MHz, CDCl3) δ 7.77-7.74 (m, 2H),
7.35-7.32 (m, 2H), 7.22-7.19 (m, 1H), 4.51 (septet, J ) 6.0,
1H), 3.20 (dd, J ) 18.0, J ) 9.0, 1H), 3.09 (dd, J ) 16.0, J )
8.0, 1H), 2.97 (dd, J ) 17.5, J ) 6.0, 1H), 2.75 (d, J ) 16.0,
1H), 2.59-2.55 (m, 1H), 1.83-1.80 (m, 1H), 1.75-1.70 (m, 1H),
1.36-1.22 (m, 7H), 1.08 (d, J ) 6.0, 3H), 0.89 (t, J ) 7.0, 3H);
13C NMR (125 MHz, CDCl3) δ 209.1, 148.7, 135.8, 128.1, 126.9,
126.5, 119.3, 83.3, 71.8, 51.5, 37.5, 29.4, 28.2, 26.4, 23.03, 22.98,
22.3, 13.9; IR (neat) 1767, 1622 cm-1; HRMS (EI) calcd for
C
22H26O2 (M+) 322.1933, found 322.1929.
Second fraction, 0.093 g (37%) of 46a as a white crystalline
C
20H26O2 (M+) 298.1933, found 298.1938.
solid: mp 52-53 °C; 1H NMR (500 MHz, CDCl3) δ 7.75-7.70
(m, 2H), 7.39-7.31 (m, 2H), 7.24-7.19 (m, 1H), 4.83 (septet,
J ) 6.0, 1H), 3.39 (dd, J ) 18.0, J ) 9.0, 1H), 3.28 (dd, J )
15.5, J ) 7.5, 1H), 3.08 (dd, J ) 18.5, J ) 6.5, 1H), 2.89-2.84
(m, 1H), 2.70 (d, J ) 15.5, 1H), 2.27 (t, J ) 7.0, 2H), 1.51
(quintet, J ) 7.0, 2H), 1.42 (hextet, J ) 7.0, 2H), 1.39 (d, J )
6.0, 3H), 1.17 (d, J ) 6.0, 3H), 0.91 (t, J ) 7.0, 3H); 13C NMR
(125 MHz, CDCl3) δ 201.1, 147.2, 135.5, 128.0, 126.8, 126.5,
117.0, 89.9, 74.0, 73.8, 73.1, 52.5, 37.0, 32.2, 30.7, 23.0, 22.8,
21.9, 18.7, 13.5; IR (neat) 1779, 1620 cm-1; HRMS (CI) calcd
for C22H26O2 (M+) 322.1933, found 322.1926.
Second fraction, 53. Isolated from attempts to purify the
corresponding naphthol 51 as a white crystalline solid: mp
84-86 °C; 1H NMR (500 MHz, CDCl3) δ 8.11 (d, J ) 8.0, 1H),
7.88 (s, 1H), 7.64-7.62 (m, 2H), 7.47-7.43 (m, 1H), 5.15
(septet, J ) 6.0, 1H), 5.12-5.05 (m, 1H), 4.83-4.80 (m, 2H),
2.90 (dd, J ) 12.5, J ) 7.5, 1H), 2.59-2.49 (m, 3H), 1.47 (m,
10H), 0.92 (t, J ) 7.5, 3H); 13C NMR (125 MHz, CDCl3) δ 185.4,
164.3, 140.7, 132.7, 132.5, 130.2, 129.5, 128.5, 126.6, 124.2,
120.0, 86.2, 75.5, 42.3, 30.8, 24.0, 23.44, 23.41, 23.0, 13.9; IR
(neat) 3316, 1631, 1599, 1572 cm-1; HRMS (CI) calcd for
C
20H27O4 (MH+) 331.1909, found 331.1910.
4-Allyl-2-(1-h e xyn yl)-3-isop r op oxy-6-m e t h oxyn a p h -
th ol (48), 4-Allyl-2-(1-h exyn yl)-3-isop r op oxy-8-m eth oxy-
n a ph th ol (49), a n d 5-(1-Hexyn yl)-4-isopr op oxy-3-(3-m eth -
oxyp h en yl)-bicyclo[3.2.0]h ep t-3-en -6-on e (50). A solution
of 20a (0.162 g, 0.46 mmol) in toluene (9.2 mL, 0.05 M) was
heated to reflux under a blanket of N2 for 3 h. Removal of the
4-Allyl-2-sec-bu tyl-3-isop r op oxyn a p h th ol (52), 4-Allyl-
2-sec-bu tyl-4-h yd r op er oxy-3-isop r op oxy-4H-n a p h th a len -
1-on e (54), an d 5-sec-Bu tyl-4-isopr opoxy-3-ph en ylbicyclo-
[3.2.0]h ep t-3-en -6-on e (56). A solution of 20i (0.166 g, 0.56
mmol) in toluene (11.1 mL, 0.05 M) was heated to reflux under
a blanket of N2 for 5 h. Removal of the volatile components
was followed by 1H NMR analysis which showed a 1:1 ratio of
52 and 56. Radial chromatography (50:1 hexanes/EtOAc) gave
as first fraction 0.073 g (44%) of bicycloheptenone 56 as a pale
yellow oil (mixture of diastereomers 1.3:1): 1H NMR (500 MHz,
CDCl3) δ 7.77 (d, J ) 7.4, 2H), 7.34 (t, J ) 7.8, 2H), 7.21 (t, J
) 7.4, 1H), 4.62-4.54 (m, 1H), 3.14, 3.13 (each dd, J ) 18.1
and J ) 9.0, J ) 18.0 and J ) 9.0, 1H), 3.01 (dd, J ) 15.7, J
) 8.1, 1H), 2.89, 2.86 (each dd, J ) 18.1 and J ) 5.9, J ) 18.0
and J ) 5.9, 1H), 2.75 (d, J ) 15.8, 1H), 2.71-2.65 (m, 1H),
2.06-1.98 (m, 1H), 1.62-1.50 (m, 1H), 1.34 (d, J ) 6.2, 3H),
1.08, 1.07 (each d, J ) 5.7, J ) 5.1, 3H), 1.00-0.93 (m, 4H),
0.89, 0.88 (each t, J ) 7.2, J ) 6.6, 3H); 13C NMR (125 MHz,
1
volatile components was followed by H NMR analysis which
showed
a
3.2:1.4:1 ratio (48:49:50). Radial chromato-
graphy (30:1 hexanes/EtOAc) gave as first fraction 0.088 g
(54%) of naphthol 48 as a yellow oil: 1H NMR (500 MHz,
CDCl3) δ 8.11 (d, J ) 9.1, 1H), 7.18 (s, 1H), 7.06 (d, J ) 9.1,
1H), 6.38 (s, 1H), 5.98 (ddt, J ) 16.6, J ) 10.6, J ) 5.6, 1H),
5.06 (d, J ) 10.6, 1H), 5.04 (d, J ) 16.6, 1H), 4.79 (septet, J )
6.1, 1H), 3.89 (s, 3H), 3.79 (d, J ) 5.6, 2H), 2.58 (t, J ) 7.1,
2H), 1.67 (quintet, J ) 7.4, 2H), 1.53 (hextet, J ) 7.3, 2H),
1.35 (d, J ) 6.1, 6H), 0.98 (t, J ) 7.2, 3H); 13C NMR (125 MHz,
CDCl3) δ 158.8, 153.9, 152.8, 137.2, 134.6, 124.3, 117.5, 115.7,
115.6, 115.2, 103.8, 101.2, 98.5, 75.7, 72.9, 55.1, 30.8, 30.0, 22.5,

4040 J . Org. Chem., Vol. 64, No. 11, 1999
Tiedemann et al.
CDCl3) δ 209.8, 209.5, 148.3, 148.1, 135.81, 135.77, 128.0,
127.0, 126.9, 126.4, 120.2, 120.1, 88.24, 88.15, 71.6, 51.8, 51.4,
37.8, 37.6, 34.7, 34.1, 25.1, 25.0, 24.6, 23.9, 23.0, 22.2, 14.3,
13.3, 12.2 11.8; IR (neat) 1769, 1622, 1598 cm-1; HRMS (CI)
calcd for C20H26O2 (M+) 298.1933, found 298.1936.
(t, J ) 7.2, 3H); 13C NMR (125 MHz, CDCl3) δ 202.2, 158.0,
145.6, 128.3, 128.1, 117.3, 113.4, 91.3, 72.5, 72.1, 58.4, 55.2,
44.8, 36.1, 30.8, 23.0, 22.7, 22.2, 21.8, 18.7, 13.5; IR (neat) 1779,
1632, 1607 cm-1; HRMS (CI) calcd for C24H30O3 (M+) 366.2195,
found 366.2193.
Second fraction, 0.043 g (24%) of 54. Isolated from attempts
to purify the corresponding naphthol 52 as a yellow crystalline
solid (mixture of diastereomers 1.8:1): mp 73-74 °C; 1H NMR
(500 MHz, CDCl3) δ 8.09-8.04 (m, 1H), 7.95, 7.85 (each br s,
1H), 7.63-7.58 (m, 2H), 7.46-7.40 (m, 1H), 5.28-5.05 (m, 2H),
4.86-4.79 (m, 2H), 3.22-3.11 (m, 1H), 2.90, 2.88 (each dd, J
) 12.3, J ) 7.6, 1H), 2.59, 2.57 (each dd, J ) 12.8, J ) 7.1,
1H), 1.86-1.58 (m, 2H), 1.46 (d, J ) 6.1, 3H), 1.43, 1.42 (each
d, J ) 6.2, 3H), 1.27, 1.22 (each d, J ) 7.1, J ) 7.0, 3H), 0.86,
0.81 (each t, J ) 7.5, 3H); 13C NMR (125 MHz, CDCl3) δ 185.3,
164.4, 163.9, 140.5, 140.4, 134.1, 134.0, 133.1, 132.9, 132.6,
129.6, 129.5, 128.5, 128.4, 126.53, 126.46, 124.1, 123.9, 120.1,
120.0, 86.3, 75.7, 75.5, 42.7, 42.4, 33.0, 32.9, 29.7, 27.6, 27.1,
23.6, 23.4, 23.3, 18.6, 18.5, 13.0, 12.9; IR (neat) 3348, 1644,
1606, 1574 cm-1; HRMS (CI) calcd for C20H27O4 (MH+)
331.1909, found 331.1908.
4-Allyl-2-ter t-bu tyl-3-m eth oxy-2H-n aph th alen -1-on e (57)
a n d 5-ter t-Bu tyl-4-m eth oxy-3-p h en ylbicyclo[3.2.0]h ep t-
3-en -6-on e (58). A solution of 20j (0.216 g, 0.80 mmol) in
toluene (16 mL, 0.05 M) was heated to reflux under a blanket
of N2 for 14 d. Removal of the volatile components was followed
by 1H NMR analysis which showed a 6:1 ratio of products
favoring the bicycloheptenone over the naphthalenone. Radial
chromatography (100:1 hexanes/EtOAc) gave as first fraction
0.149 g (69%) of bicycloheptenone 58 as a yellow oil: 1H NMR
(500 MHz, CDCl3) δ 7.64-7.62 (m, 2H), 7.37 (t, J ) 7.8, 2H),
7.25 (t, J ) 7.6, 1H), 3.59 (s, 3H), 3.14 (dd, J ) 17.9, J ) 9.1,
1H), 3.00 (dd, J ) 16.1, J ) 7.8, 1H), 2.86 (dd, J ) 18.1, J )
5.8, 1H), 2.73-2.67 (m, 2H), 1.12 (s, 9H); 13C NMR (125 MHz,
CDCl3) δ 209.5, 152.1, 135.4, 128.2, 127.3, 127.0, 122.9, 88.8,
58.4, 51.1, 37.1, 34.6, 26.4, 26.3; IR (neat) 1763, 1618, 1596
cm-1; HRMS (CI) calcd for C18H22O2 (M+) 270.1620, found
270.1623.
3-Meth oxy-6-m eth oxym eth yl-2,4-(3,4-dim eth oxyph en yl)-
p h en ol (62). 4-Allenyl-3-methoxy-2,4-(3,4-dimethoxyphenyl)-
2-cyclobutenone (44d ) (0.103 g, 0.25 mmol, 1 equiv) was
dissolved in acetonitrile (6 mL, 0.03 M) and treated with
anhydrous methanol (1.02 mL, 25.2 mmol, 100 equiv). The
resulting solution was heated to gentle reflux for 2 h. Removal
of the volatile components provided an orange oil which was
purified by flash silica gel chromatography (19:1 CH2Cl2/
EtOAc) to furnish 0.061 g (55%) of 62 as a white crystalline
1
solid: mp 158-159 °C; H NMR (500 MHz, CDCl3) δ 7.14-
7.13 (m, 2H), 7.07 (dd, J ) 8.0, J ) 2.0, 1H), 7.03-6.99 (m,
3H), 6.91 (d, J ) 8.0, 1H), 6.79 (s, 1H), 4.64 (s, 2H), 3.94 (s,
3H), 3.92 (s, 3H), 3.903 (s, 3H), 3.900 (s, 3H), 3.48 (s, 3H), 3.21
(s, 3H); 13C NMR (125 MHz, CDCl3) δ 155.6, 152.6, 148.9,
148.5, 148.4, 147.9, 131.3, 129.8, 126.5, 125.7, 123.2, 122.7,
121.2, 118.7, 113.6, 112.4, 111.1, 110.9, 72.4, 60.5, 58.4, 55.94,
55.90, 55.8; IR (neat) 3436, 1592 cm-1; HRMS (EI) calcd for
C
25H28O7 (M+) 440.1835, found 440.1830.
3-(1-H exyn yl)-4-m et h yl-5-p h en ylb icyclo[3.2.0]h ep t -3-
en -6-on e (63). A solution of 32a (0.147 g, 0.53 mmol) in
toluene (11 mL, 0.05 M) was heated to reflux under a blanket
of N2 for 3 h. Removal of the volatile components followed by
radial chromatography (20:1 hexanes/EtOAc) furnished 0.131
g (89%) of 63 as a water white oil: 1H NMR (500 MHz, CDCl3)
δ 7.34-7.23 (m, 5H), 3.35-3.28 (m, 1H), 3.20-3.15 (m, 1H),
2.92-2.85 (m, 2H), 2.58 (d, J ) 16.7, 1H), 2.40 (t, J ) 7.0,
2H), 1.67 (s, 3H), 1.56 (quintet, J ) 7.0, 2H), 1.46 (hextet, J )
7.4, 2H), 0.95 (t, J ) 7.2, 3H); 13C NMR (125 MHz, CDCl3) δ
206.1, 142.8, 138.2, 128.4, 127.0, 125.8, 120.8, 95.6, 87.7, 76.7,
51.2, 43.2, 34.2, 30.8, 21.9, 19.2, 13.6, 12.7; IR (neat) 2217,
1770, 1602 cm-1; HRMS (EI) calcd for C20H22O (M+) 278.1671,
found 278.1684.
Second fraction, 0.029 g (13%) of 57 as a yellow oil: 1H NMR
(500 MHz, CDCl3) δ 7.77 (dd, J ) 7.8, J ) 1.4, 1H), 7.48 (dt,
J ) 7.8, J ) 1.4, 1H), 7.33 (d, J ) 7.6, 1H), 7.21 (dt, J ) 7.6,
J ) 1.0, 1H), 5.92 (ddt, J ) 17.1, J ) 10.1, J ) 5.6, 1H), 5.13
(dq, J ) 17.1, J ) 1.6, 1H), 5.04 (dq, J ) 10.1, J ) 1.6, 1H),
3.63 (s, 3H), 3.45 (ddt, J ) 15.9, J ) 5.4, J ) 2.0, 1H), 3.34-
3.28 (m, 1H), 3.20 (s, 1H), 0.97 (s, 9H); 13C NMR (125 MHz,
CDCl3) δ 201.9, 155.5, 139.8, 136.2, 133.8, 131.3, 126.1, 125.8,
124.1, 118.6, 115.5, 58.8, 58.0, 37.9, 29.1, 28.4; IR (neat) 1682,
1644, 1634, 1598 cm-1; HRMS (EI) calcd for C18H22O2 (M+)
270.1620, found 270.1620.
2-(1-H e xyn yl)-3-isop r op oxy-7-m e t h oxy-4-(2-m e t h yl-
a llyl)-n a p h th ol (59a ) a n d 5-(1-Hexyn yl)-4-isop r op oxy-3-
(4-m et h oxyp h en yl)-1-m et h yl-b icyclo[3.2.0]h ep t -3-en -6-
on e (60a ). A solution of 21a (0.175 g, 0.48 mmol) in toluene
(9.5 mL, 0.05 M) was heated to reflux under a blanket of N2
for 3 h. Removal of the volatile components was followed by
1H NMR analysis which showed a 1.4:1 ratio of products
favoring the naphthol over the bicycloheptenone. Radial chro-
matography (100:1 pentane/EtOAc) gave as first fraction 0.103
g (59%) of naphthol 59a as a yellow oil: 1H NMR (500 MHz,
CDCl3) δ 7.70 (d, J ) 9.3, 1H), 7.47 (d, J ) 2.5, 1H), 7.13 (dd,
J ) 9.2, J ) 2.5, 1H), 6.37 (s, 1H), 4.76 (s, 1H), 4.68 (septet, J
) 6.1, 1H), 4.36 (s, 1H), 3.93 (s, 3H), 3.69 (s, 2H), 2.59 (t, J )
7.1, 2H), 1.82 (s, 3H), 1.68 (quintet, J ) 7.3, 2H), 1.54 (hextet,
J ) 7.4, 2H), 1.34 (d, J ) 6.2, 6H), 0.99 (t, J ) 7.3, 3H); 13C
NMR (125 MHz, CDCl3) δ 156.3, 152.6, 150.6, 144.7, 128.5,
126.4, 121.2, 119.5, 118.8, 111.1, 101.8, 101.3, 100.7, 75.7, 73.1,
55.3, 33.8, 30.8, 23.2, 22.5, 22.1, 19.7, 13.6; IR (neat) 3486,
1599, 1575 cm-1; HRMS (CI) calcd for C24H30O3 (MH+)
366.2195, found 366.2191.
5-(1-H exyn yl)-2,3-d ih yd r o-6-h yd r oxy-4-isop r op oxy-2-
tr iisop r op ylsilyl-7,8-d im eth oxy-1H-p h en a len e (68). A so-
lution of 42 (0.067 g, 0.13 mmol) in toluene (6 mL, 0.02 M)
was heated to reflux under a blanket of N2 for 45 min. Removal
of the solvent was followed by flash silica gel chromatography
(9:1 hexanes/EtOAc) which furnished 0.057 g (60%) of 68 as a
water white oil: 1H NMR (500 MHz, CDCl3) δ 10.29 (s, 1H),
6.96 (s, 1H), 4.73 (septet, J ) 6.0, 1H), 4.04 (s, 3H), 3.95 (s,
3H), 3.49-3.45 (m, 1H), 3.09-3.01 (m, 2H), 2.69-2.63 (m, 1H),
2.59 (t, J ) 7.0, 2H), 1.67 (quintet, J ) 7.5, 2H), 1.55-1.46
(m, 3H), 1.33 (d, J ) 6.0, 6H), 1.31-1.18 (m, 3H), 1.15 (d, J )
7.0, 18H), 0.96 (t, J ) 7.5, 3H); 13C NMR (125 MHz, CDCl3) δ
154.2, 150.1, 146.0, 140.7, 134.5, 126.1, 119.1, 114.7, 113.6,
103.1, 99.1, 75.2, 73.9, 62.1, 56.8, 34.1, 31.1, 26.7, 22.7, 22.5,
22.2, 19.9, 19.2, 13.7, 11.0; IR (neat) 3297, 1606 cm-1; HRMS
(EI) calcd for C33H50O4Si (M+) 538.3478, found 538.3464.
2-(1-Hexyn yl)-3-isop r op oxy-5-t r iisop r op ylsilyl-8,9-d i-
m eth oxy-p h en a len on e (69). A solution of 42 (0.045 g, 0.08
mmol) in toluene (3.3 mL, 0.5 M) was heated to reflux under
a blanket of N2 for 45 min. After cooling to room temperature,
the pale yellow reaction mixture was treated with 2,3-dichloro-
5,6-dicyano-1,4-benzoquinone (0.040 g, 0.18 mmol). After 5
min, the crude, red reaction mixture was filtered through a
silica plug and concentrated to afford a dark yellow solid. Flash
silica gel chromatography (4:1 hexanes/EtOAc) furnished 0.032
g (72%) of 69 as a yellow crystalline solid: mp 168-170 °C;
1H NMR (500 MHz, CDCl3) δ 8.28 (d, J ) 1.0, 1H), 8.02 (d, J
) 1.0, 1H), 7.51 (s, 1H), 4.52 (septet, J ) 6.0, 1H), 4.09 (s,
3H), 4.04 (s, 3H), 2.58 (t, J ) 7.0, 2H), 1.66 (quintet, J ) 7.0,
2H), 1.56-1.48 (m, 5H), 1.43 (d, J ) 6.5, 6H), 1.15 (d, J ) 7.0,
18H), 0.95 (t, J ) 7.0, 3H); 13C NMR (125 MHz, CDCl3) δ 183.0,
166.0, 156.0, 153.4, 138.3, 132.3, 131.2, 128.4, 125.1, 122.5,
119.8, 113.7, 111.7, 101.7, 76.0, 73.7, 61.4, 56.2, 30.9, 22.9, 22.2,
19.9, 18.6, 13.7, 10.8; IR (neat) 1631, 1577, 1562 cm-1; HRMS
(CI) calcd for C33H47O4Si (MH+) 535.3243, found 535.3233.
Second fraction, 0.071 g (40%) of 60a as a yellow oil: 1H
NMR (500 MHz, CDCl3) δ 7.67 (d, J ) 8.8, 2H), 6.87 (d, J )
8.8, 2H), 4.80 (septet, J ) 6.0, 1H), 3.81 (s, 3H), 3.29 (d, J )
17.7, 1H), 2.91 (d, J ) 17.7, 1H), 2.88 (s, 2H), 2.29 (t, J ) 7.0,
2H), 1.51 (quintet, J ) 7.0, 2H), 1.42 (hextet, J ) 7.2, 2H),
1.40 (s, 3H), 1.38 (d, J ) 6.0, 3H), 1.14 (d, J ) 6.0, 3H), 0.90

Synthesis and Thermolysis of 4-Allyl-4-arylcyclobutenones
J . Org. Chem., Vol. 64, No. 11, 1999 4041
Su p p or tin g In for m a tion Ava ila ble: Procedures and
additional compound characterization data, X-ray data for
compound 28, copies of 1H and 13C NMR spectra. This material
is available free of charge via the Internet at http://pubs.acs.org.
Ack n ow led gm en t. The authors thank the National
Institutes of Health (GM-36312) for financial support.
We are also grateful to Professor Everly B. Fleischer
for determining the X-ray crystallographic structure of
compound 28.
J O990052A