Reactions of nucleophilic carbenes with enols

Article abstract of DOI:10.1039/a700169j

The formal insertion of dimethoxycarbene (1a) into the acidic C-H bond of pentane-2,4-dione (9a), methyl acetoacetate (9b), 3-methylpentane-2,4-dione (9c) and 1,3-diphenylpropane-1,3-dione (9d) is reported as well as the insertion of 3-benzoyloxazolidin-2

Full text of DOI:10.1039/a700169j

View Article Online / Journal Homepage / Table of Contents for this issue
Reactions of nucleophilic carbenes with enols
Philippe Couture, David L. Pole and John Warkentin*
Department of Chemistry, McMaster University, Hamilton, Ontario, Canada, L8S 4M1
The formal insertion of dimethoxycarbene (1a) into the acidic C᎐H bond of pentane-2,4-dione (9a),
methyl acetoacetate (9b), 3-methylpentane-2,4-dione (9c) and 1,3-diphenylpropane-1,3-dione (9d)
is reported as well as the insertion of 3-benzoyloxazolidin-2-ylidene (1b) into 9c. The â-dicarbonyl
compounds 9 are known to be equilibrated with their corresponding enols in benzene solution and
the insertions appear to proceed by proton abstraction from the enol tautomers of 9 to generate enolate
anions and either a dimethoxymethyl cation (from 1a) or a 3-benzoyloxazolidin-2-ium cation (from 1b).
Collapse of these ion pairs at the carbon atom of the enolate yields the major product. Formal insertion
of 1a into the O᎐H bond of the enol tautomer of anthrone (12) is also reported.
Investigations of the reactions of carbenes with alcohols and
phenols have resulted in a better understanding of the mechan-
isms of insertion of carbenes into the O᎐H bond. There is
good evidence for a mechanism in which nucleophilic carbenes
such as dimethoxycarbene (1a, Scheme 1) abstract the O᎐H
Reports of reactions of nucleophilic carbenes with com-
pounds having a high enol content have not appeared. Depro-
tonation of an enol by a nucleophilic carbene should yield an
enolate and a stabilized carbocation (e.g. dimethoxymethyl
cation in the case of 1a). Such an ion pair could collapse to
afford either the product of O-alkylation or that of C-alkyl-
ation. Which process predominates is not known because
carbene generated ion pairs in which the anionic portion is an
enolate have apparently not been reported.
1
Results and discussion
Scheme 1
3
General thermolysis of 2,2-dimethoxy-5,5-dimethyl-∆ -1,3,4-
proton of an alcohol to generate an ion pair (2) that collapses to
yield an orthoformate (3). For example, absolute rate constants
for the reaction of dimethoxycarbene with a variety of alcohols
6
oxadiazoline (8a) and 3,4,9-triaza-9-benzoyl-2,2-dimethyl-1,6-
7
dioxaspiro[4.4]non-3-ene (8b) (Scheme 3) resulted in the gener-
2
are correlated to the ionization constants of the alcohols. A
large primary deuterium kinetic isotope effect (k /k = 3.3) has
H
D
been measured for the insertion of dimethoxycarbene into the
3
O᎐H bond of methanol. In addition, the cationic inter-
mediates resulting from protonation of several nucleophilic
carbenes have been detected spectroscopically by means of laser
4
,5
flash photolysis. These results are all in keeping with the
mechanism in Scheme 1 for the insertion of dimethoxycarbene
into an O᎐H bond.
Proton transfer to the nucleophilic carbene generates the
cation–anion pair (2) in intimate contact. These carbene-
generated contact ion pairs have been shown to lead to scram-
bling when the cationic portion is delocalized. Thus, photolysis
of the phenylallene 4 results in rearrangement to the vinyl-
carbene 5 which has been shown to abstract a proton from
Scheme 3
6
,7
ation of stabilized carbenes (1a and 1b). Typically, a solu-
tion of an oxadiazoline (0.05 ) in dry benzene containing one
equivalent of a β-dicarbonyl compound (9a–d ) was heated at
10 ЊC for 20 h in the case of 8a or at 90 ЊC for 24 h for 8b. The
major products from dimethoxycarbene (10, Scheme 4) result
from carbene insertion into a C᎐H bond of the keto tautomers
1
methanol to yield the allyl cation–methoxide ion pair 6 (Scheme
5
2
). The methoxide ion bonds to either C-1 or C-3 of the allyl
Scheme 2
Scheme 4
cation to generate the isomeric ethers 7a or 7b in a 1:2.8 ratio.
In addition, the allyl cation has been observed as a transient
during laser flash photolysis and it has been characterized by
of 9a–d (see Table 1). In the cases of 9a and 9b, the reaction
mixture also contained the products 11a and 11b which arose
from elimination of methanol from 10a and 10b, respectively.
Product 11b was a mixture of E and Z isomers in roughly equal
5
means of extensive deuterium labelling experiments.
J. Chem. Soc., Perkin Trans. 2, 1997
1565

View Article Online
Table 1 Yields of products of dimethoxycarbene trapping (10a–d,
1a–b and 13) with activated methylene compounds 9a–d and 12
1
a
Methylene compound
Product(s)
Yield
1
11a
0a
17
8
9
9
a
Scheme 5
1
1
0b
29
14
b
1b (E, Z isomers)
were informative. Methyl cyanoacetate (9e), malononitrile (9f)
and dimethyl malonate (9g) all proved to be ineffective agents
for trapping of 1a. Only trace amounts of likely products of
trapping could be detected in the GC–MSD traces of the crude
reaction mixtures. Addition of pyridine in various amounts had
little effect on the trapping efficiency of 9e.
1
1
0c
0d
26
56
9
9
c
d
Since carbonyl compounds 9 exist in equilibrium with the
corresponding enols, there are two pathways, both involving
proton abstraction, to achieve overall C᎐H insertion (Scheme
1
2
13
30
6
). As discussed above, there is a body of evidence to suggest
a
Yield, by NMR spectroscopy, is based on oxadiazoline.
proportions. The reaction mixture from trapping with 9c did
3
not contain such products because the methyl group (R = CH )
3
blocked elimination. Evidence for formation of 11d from the
reaction of 1 with 9d was not found in our analysis of the
reaction mixture. For the traps 9a and 9b, the ratios of primary
products (10) to products of methanol loss (11) were 2.2:1 and
4
.7:1, respectively. Yields of carbene-derived products based
on oxadiazoline, determined by NMR spectroscopic analysis of
the reaction mixtures containing an internal standard, are given
in Table 1.
The crude samples from trapping of dimethoxycarbene with
9
a–c (1 equiv.) were analyzed directly by GC and GC–MSD.
Reaction products were then concentrated by removal of the
solvent, along with much of the excess β-dicarbonyl compound,
under vacuum. The product from dimethoxycarbene trapping
1
by 9d was identified by H NMR spectroscopy on the reaction
mixture which still contained leftover trap. Nearly total conver-
sion of 9d to 10d was achieved by conducting a thermolysis
with a 2.5-fold excess of oxadiazoline 8a.
Mixtures resulting from trapping of 1a by 9a and 9b could be
converted entirely to the products of methanol loss (11a and
Scheme 6
1
1b) by adding a catalytic amount of pyridine (5 µl in 25 ml)
and reheating the mixture to 110 ЊC for 20 h (unoptimized con-
ditions). However, addition of pyridine prior to thermolysis (so
that it was present during carbene generation) resulted in low-
ered yields of carbene-derived products. The cause is unlikely to
that the O᎐H insertion reactions of dialkoxycarbenes are step-
wise in nature and progress through an ion pair intermediate.
However, proton abstraction from either the keto form or the
enol form of 9a, for example, leads to the same ion pair inter-
mediate 15. Although collapse of the ion pair could lead to a
structure of either type 10 or 16, only 10 and its analogues were
observed from reaction of 1a and 9.
The observation of predominant C-alkylation is in keeping
with the greater thermodynamic stability of 10 compared to
16. C-Alkylation products are generally more stable than O-
alkylation products, largely because of the greater bond energy
8
be competitive trapping of the carbene with pyridine, as trap-
ping of a nucleophilic carbene by pyridine is expected to be
quite slow. Rather, pyridine results in full elimination of meth-
anol, with the methanol produced intercepting some ion pairs
(or dimethoxycarbene), thus lowering yields of products of
type 10. For instance, trimethyl orthoformate was observed in
1
about 43% yield ( H NMR, p-xylene as reference) from a reac-
tion involving pentane-2,4-dione 9a (0.29 ) and oxadiazoline
of a C᎐O double bond compared to a C᎐C double bond (a differ-
᎐ ᎐
Ϫ1
2
8
a (0.24 ) in [ H ]benzene even in the absence of pyridine. This
ence of ca. 100 kJ mol ). Whether or not C-alkylation is pre-
6
explanation does not account for the poor yield of 10c from
ferred kinetically, as observed for soft electrophiles such as alkyl
iodides, cannot be determined from these data. If O-alkylation
9
3
-methylpentane-1,3-dione 9c, however.
Similar methods were used to examine the trapping of 1a
is kinetically preferred (the more likely alternative) then the
observation of exclusive C-alkylation must mean that 16 is
unstable to the reaction conditions. Return to the ion pair 15,
because the enolate anion of β-dicarbonyl systems is a good
leaving group, would ultimately afford the thermodynamically
favoured C-alkylation product (10).
with anthrone 12. Solutions of 0.05  oxadiazoline, 1 equiv. of
anthrone 12 and a catalytic amount of pyridine were heated
at 110 ЊC for 20 h. The reaction mixture contained a single
anthrone-derived product 13, formally generated from insertion
of dimethoxycarbene into the O᎐H bond of the enol tautomer,
anthranol 14 (Scheme 5). In that case addition of pyridine prior
to thermolysis was effective in increasing the yield of 13 (from
Fortunately, the type of tautomer involved in the reaction
with the carbene could be deduced. Tautomerization equi-
<
10% in the absence of pyridine), presumably by catalyzing
librium constants in benzene and pK s in water of some of the
a
the keto–enol equilibration, without the complication from
formation of methanol.
Trapping studies with other activated methylene compounds
carbonyl compounds are presented in Table 2. For 9a–d, the
enolic tautomer is known to be plentiful in non-polar solvents
such as benzene. We confirmed the presence of the enolic forms
1
566
J. Chem. Soc., Perkin Trans. 2, 1997

View Article Online
Table 2 Physical properties of carbonyl compounds 9a–f
proton from 10 to afford a new ion pair with the potential for
loss of methoxide, affording 11. Methoxide could then become
a chain carrier, either by direct proton abstraction from 10 or
its enol, or through formation of the enolate from 9. Thus the
anions could also serve to catalyze the tautomerization of keto-
9 to enol-9. Dimethoxycarbene itself could also deprotonate the
enol of 10, thus leading to overall removal of methanol. In
keeping with the observed results, this mechanism predicts that
once the production of carbene has ended, and the source of
the ion pairs is gone, the conversion of 10 to 11 ends. The
expected trimethyl orthoformate, from capture of dimethoxy-
carbene or of dimethoxymethyl cation by methanol, was a
major product in the one case where that product was sought
(above).
a
f
Compound
KT
pKa
b
g
9
9
9
9
9
9
2
a
b
c
d
e
f
14.7
9
11
b
g
0.26c
0.67
—
—
9
c
22.8
Ϫ10 d
h
2.6 × 10
n/a
0.0025
h
13
e
1
—
a
b
Equilibrium constant for tautomerization; K = [enol]/[keto]. In
T
2
c
d
[
H ]benzene, from ref. 10. In CDCl3 at 40 ЊC, from ref. 11. In
6
e 2
dioxane–water for the ethyl ester from ref. 12. In [ H ]toluene, from
8
f
g
h
ref. 10. pK in H O. From ref. 10. From ref. 13.
a
2
Exclusive O-alkylation of anthrone 12 is opposite to the
behaviour of the other active methylene compounds. The lower
enol content of 12 is presumably sufficient to trap the carbene
which is generated over the course of 20 h. Although the prod-
uct 13 was formed in benzene alone, the addition of a drop of
pyridine resulted in improved yields. This is likely to be the
result of an acceleration of the known slow rate of tautomeriz-
ation of anthrone in non-polar solvents, thereby replenishing
the enol as it is consumed by the carbene. The predominance of
O-alkylation could again be the result of either kinetic or
thermodynamic control but in this case both thermodynamics
and kinetics are likely to favour O-alkylation. Although the
product could isomerize by reforming the ion pair from which it
was born, the enolate of anthrone 12 is probably less stable than
that of pentane-2,4-dione in benzene solution.
1
2
of 9a–d by H NMR spectroscopy of [ H ]benzene solutions,
6
which had been freshly prepared at approximately the therm-
olysis concentration from the pure carbonyl compounds, and
then dried with MgSO . As expected, the hydrogen bonded pro-
ton of each enol appeared at high frequency as a singlet
9a, 86%, δ 16.17; 9b, 16%, δ 12.67; 9c, 48%, δ 17.10; 9d, 100%,
4
(
1
5
δ 17.72) and concentrations were relatively close to the equili-
brium values in Table 2.
These data favour a mechanism in which proton abstraction
from the enol is the source of the ion pair intermediate. In the
cases of active methylene compounds with at least one ketone
functional group (9a–d), the enol content (Table 2) is sufficient
for the well precedented abstraction of a proton from an O᎐H
group. Analogous abstraction from carbon is unlikely because
the pK values of methyl cyanoacetate and of pentane-2,4-dione
are similar in aqueous solution, and yet the former is a poor
The carbene 1b was intercepted efficiently by the enol 9c, to
afford 17c in 72% yield (Scheme 7). That acetal-like compound
a
trap for carbene 1a, whilst the latter is very effective. Since the
carbenes reactivity toward hydroxy protons has been correlated
2
to the pK values of corresponding acids, one might expect
a
similar trapping efficiencies for methyl cyanoacetate and
pentane-2,4-dione if abstraction from the keto tautomer
occurred. Although values for the ionization constants of these
two β-dicarbonyl compounds in benzene are not available, and
are likely to be significantly different from those in water, the
order of carbon acid acidity is likely to remain the same. On
the other hand, pentane-2,4-dione exists largely as the enol
tautomer while the concentration of the enol tautomer of
methyl cyanoacetate is vanishingly small, presumably because
the enol cannot be stabilized by intramolecular H-bonding to
the cyano group. Thus the enol concentration at equilibrium is
more important for successful trapping of 1a than the pK of
a
the carbon acid.
The mechanism of elimination of methanol from 9a and 9b is
of some interest even though the products (11) are known. The
synthesis of compounds such as 11 from the action of trimethyl
orthoformate and acetic anhydride on β-dicarbonyl com-
Scheme 7
could be purified by careful chromatography, presumably by
virtue of the benzoyl substituent. The enol of pentane-2,4-
dione (9a) was probably effective also in trapping 1b, but in that
case the major initial product (17a, assumed but not character-
ized) hydrolyzed upon exposure to the atmosphere, affording N-
(2-hydroxyethyl)benzamide (58%). In the case of 9a, two inter-
esting minor products were also identified as 18 and 19, both
isolated in 2% yield. Compound 18 is clearly the product of bis-
alkylation of 9a, and that process presumably has to occur in
the order: C-alkylation, re-enolization and O-alkylation, as O-
alkylation first can only result in bis-alkylation by return to the
ion pair and subsequent C-alkylation. The E geometry was
assigned on the basis of both a lesser steric interaction and the
assumption that the enolic precursor was H-bonded to the car-
bonyl oxygen. The origin of 19 is very likely the initial ion pair.
In addition to collapsing to 17, that species could find an alter-
native path, namely benzoyl group transfer from the cation to
the oxygen of the anion. Although benzoyl group transfer to
carbon of the enolate may compete, the anticipated product
was not found, nor was the expected co-product, oxazoline,
1
4
pounds has been known for over a century. The mechanism of
the formation of these compounds in the reaction mixtures
studied here was probed with 9a as a model. A GC trace of the
reaction mixture from thermolysis of 8 in the presence of 9a,
showed that 10a and 11a were present in a 1:1 ratio. Reheating
of that mixture at 130 ЊC for 20 h did not cause a significant
change in the GC trace. Thus, 11 did not arise from 10 as a
result of the thermal instability of the latter. Complete conver-
sion of 10a and 10b to 11a and 11b could be achieved by reheat-
ing the sample in the presence of pyridine. Thus the 1:1 ratio of
1
0a to 11a is not the equilibrium value and 11a must come from
a process related to the presence of the carbene intermediate.
Our interpretation of these results invokes occasional proton
transfers between an ion pair intermediate 15 and a β-
dicarbonyl compound 9 or the corresponding initial product
1
0. Although the lifetime of an ion pair in benzene is expected
to be short, the lifetime of 15 is probably extended through its
regeneration from its reservoir 16. Thus, the enolate anion of
the ion pair (or a solvent separated analogue) could abstract a
J. Chem. Soc., Perkin Trans. 2, 1997
1567

View Article Online
identified. It is likely that a product analogous to 18 was among
the many components of the complex mixtures that generally
resulted from trapping of 1a.
3-(Dimethoxymethyl)pentane-2,4-dione (10a). νmax(gas phase)/
Ϫ1
cm 3005 (w), 2961 (m), 2944 (m), 2844 (w), 1718 (s), 1444 (w),
1363 (m), 1284 (w), 1199 (m), 1120 (s), 1084 (s); δ (300 MHz)
H
3
In summary, reactions of nucleophilic carbenes with some
active methylene compounds indicate that such carbenes can
abstract a proton from an enol to generate a contact ion pair in
which the anion is an enolate. The ion pair collapses to form the
eventual product of formal carbene insertion into the C᎐H
bond of the keto tautomer in the cases of 9a–d or into the O᎐H
bond of the enol tautomer of 12. Elimination of methanol from
insertion products arising from 1a and 9a (or 9b) appears to be
carbene initiated.
2.17 (6H, s, COCH ), 3.31 (6H, s, OCH ), 4.01 [1H, d, J 8.4,
3
3
3
(CH CO) CH], 4.94 [1H, d, J 8.4, (CH O) CH]; δ (75.5 MHz)
3
2
3
2
C
30.2 (CH CO), 54.5 (OCH ), 71.5 [(COCH ) CH], 103.1
3
3
3 2
ϩ
[CH(OCH ) ], 200.7 (C᎐O); m/z (EI) (%) 143 [M Ϫ OCH ] (4),
3
2
3
ϩ
131 [M Ϫ CH CO] (28), 127 (12), 101 (51), 85 (100), 83 (14),
75 [(CH O) CH] (82), 69 (10), 55 (10), 47 (25), 43 (94).
3
ϩ
3
2
3-(Methoxymethylene)pentane-2,4-dione
phase)/cm 3017 (w), 2950 (w), 2856 (w), 1700 (s), 1610 (s),
1444 (w), 1369 (w), 1278 (s), 1191 (w), 1133 (s), 1066 (w), 995
(11a).
νmax(gas
Ϫ1
(
s), 958 (w); δ 2.24 (3H, s, COCH ), 2.30 (3H, s, COCH ), 3.96
H
3
3
(
6
3H, s, OCH ), 7.53 (1H, s, vinyl); δ 29.1 (CH ), 32.0 (CH ),
3.5 (OCH ), 122.2 (vinyl), 166.9 (vinyl), 197.0 (C᎐O), 197.8
3
ϩ ϩ
3 C 3 3
Experimental
Typical thermolyses were performed with an oxadiazoline
concentration between 0.03 and 0.10  in dry benzene and one
equivalent of trap. Samples were sealed into a resealable therm-
olysis tube and heated in a constant temperature oil bath at
(C᎐O); m/z (EI) (%) 142 [M] (5), 127 [M Ϫ CH3] (61), 85
(100), 69 (5), 43 (55); m/z (CI, NH ) (%) 160 [M ϩ NH ] (10),
143 [M ϩ H] (100).
᎐
ϩ
3
4
ϩ
Conversion of 10a to 11a with pyridine
1
10 ЊC for 20 h, unless otherwise indicated. For reactions of
Dimethoxydimethyloxadiazoline 8a (0.10 g, 0.6 mmol) and
pentane-2,4-dione (9a) (0.07 g, 0.7 mmol) were dissolved in 12.5
ml of benzene and sealed in a thermolysis tube. After heating at
dimethoxydimethyloxadiazoline 8a, the crude product mixtures
were analyzed by GC–MSD and GC–FTIR. The acetals 10 and
the enol ethers 11 are not stable to chromatography. Excess trap
in the reaction mixture, in the cases of 9a–c, could be removed
1
10 ЊC for 20 h, 0.33 g of pyridine was added to the thermolysis
1
tube and the mixture was heated again for 20 h. Conversion of
the initial 10a to 11a was found to be complete after this time.
The only product detectable by GC–MSD was 11a; 0.03 g was
isolated. Addition of the same quantity of pyridine before
thermolysis of the oxadiazoline gave products of carbene trap-
ping in trace amounts only.
under vacuum to yield samples that could be analyzed by H
1
3
and C NMR spectroscopy without the spectra being obscured
by excess trap. In the case of 9d and 12, thermolyses of a sample
with 2–3:1 ratio of oxadiazoline:trap resulted in products
which were fairly free of excess trap. Yields reported in Table 1
were determined from an aliquot of the reaction mixture to
which a non-volatile standard was added. The benzene was
removed and the residue was dissolved in CDCl and analyzed
by integration of the H NMR signals. The incomplete mass
balance for reactions of 8a is the result of formation of tri-
methyl orthoformate (where appropriate) and other uncharac-
terized volatile products. For 8b, minor thermolysis pathways
Thermolysis of dimethoxydimethyloxadiazoline (8a) in presence
of methyl acetoacetate (9b)
A solution containing 0.14 g (0.8 mmol) of dimethoxydimethyl-
oxadiazoline 8a and 0.09 g (0.8 mmol) of methyl acetoacetate
3
1
(9b) in 25.0 ml of benzene was sealed into a thermolysis tube
and heated at 110 ЊC for 20 h. Removal of the solvent and
excess 9b yielded 0.08 g of a mixture of 10b and 11b. Analysis
by H NMR spectroscopy showed a mixture of 10b and 11b in a
4
4
troscopy (p-bromoacetophenone as internal standard). Spectral
data for 11b were in agreement with those reported in the
literature.
7
competing with formation of 1b have been described.
In general, the products 10 and 11 could not be stored as neat
liquids due to their high susceptibility to hydrolysis from
atmospheric moisture. In the cases where isolation of the prod-
ucts of trapping was not successful, spectroscopic analyses ( H,
C, MS, FTIR) were performed on the crude reaction mixtures.
Although 10a–d and 13 are new compounds, their sensitivity to
moisture (i.e. the compounds hydrolyzed on silica) meant that
only 10c could be isolated in sufficient purity for elemental
1
.7:1 ratio. The total yield of carbene derived products was
1
3%, based on oxadiazoline and determined by H NMR spec-
1
1
3
16
Methyl 2-(dimethoxymethyl)-3-oxobutanoate (10b). νmax(gas
Ϫ1
phase)/cm 3004 (w), 2962 (m), 2845 (w), 1765 (s), 1737 (s),
1
analysis.
441 (w), 1364 (m), 1299 (w), 1195 (s), 1138 (m), 1117 (s), 1094
1
Unless otherwise stated for the particular compound,
and C NMR spectra were obtained in CDCl solvent with a
Bruker AC-200 NMR spectrometer operating at 200 and 50.3
H
(
(
s), 989 (w); δ 2.23 (3H, s, CH CO), 3.35 (3H, s, OCH ), 3.38
3H, s, OCH ), 3.69 (3H, s, CO CH ), 3.84 [1H, d, J 8,
H
3
3
1
3
3
3
3
2
3
3
CH(CO) ], 4.91 [1H, d, J 8, CH(OCH ) ]; δ 30.3 (CH CO),
5
1
2
3
2
C
3
MHz, respectively, with residual CHCl (δ 7.24) and the centre
3
H
2.4 (CO CH ), 55.0 (OCH ), 55.2 (OCH ), 63.2 (COCHCO),
2 3 3 3
line of the CDCl triplet (δ 77.0) as the internal standards;
3
C
03.2 [C(OCH )], 166.7 (CO CH ), 199.8 (COCH ); m/z (EI)
3
2
3
3
J values are given in Hz. Gas phase FTIR: Hewlett-Packard
HP-5890 gas chromatograph connected to a Bio-Rad FTS-40
FTIR with a Bio-Rad GC/C 32 GC interface. MS(EI): HP-5890
Series II gas chromatograph with a 5971A mass selective
detector or a VG Analytical ZAB-E double focusing mass
spectrometer. MS(CI): VG Analytical ZAB-E double focusing
mass spectrometer.
ϩ
(%) 147 [M Ϫ CH CO] (<1), 143 (10), 131 (14), 117 (30), 115
3
(6), 101 (5), 85 (100), 75 (85), 69 (14), 59 (11), 47 (17), 43 (37).
(
E/Z )-Methyl 2-(methoxymethylene)-3-oxobutanoate (11b).
Ϫ1
νmax(gas phase)/cm 3025 (w), 2955 (m), 2856 (w), 1736 (s),
708 (s), 1615 (s), 1441 (m), 1374 (m), 1287 (s), 1267 (s), 1196
1
(s), 1140 (s), 1079 (m), 997 (m); δH 2.25 (3H, s, CH CO), 2.30
3
(3H, s, CH CO), 3.67 (3H, s, OCH ), 3.73 (3H, s, OCH ), 3.92
3
3
3
(
(
5
3H, s, CO CH ), 3.94 (3H, s, CO CH ), 7.50 (1H, s, vinyl), 7.51
2 3 2 3
Thermolysis of dimethoxydimethyloxadiazoline (8a) in the
presence of pentane-2,4-dione (9a)
A solution containing 8a (0.13 g, 0.8 mmol) and 9a (0.09 g, 0.9
mmol) in 25 ml of benzene was sealed into a thermolysis tube
and heated at 110 ЊC for 20 h. The solvent and excess 9a were
1H, s, vinyl); δ 28.5 (COCH ), 31.5 (COCH ), 51.6 (CO CH ),
C 3 3 2 3
1.7 (CO CH ), 63.5 (OCH ), 63.6 (OCH ), 113.2 (vinyl), 114.0
2 3 3 3
(vinyl), 165.4 (C᎐O), 165.6 (vinyl), 165.8 (C᎐O), 166.7 (vinyl),
᎐
᎐
ϩ
1
95.0 (C᎐O), 196.5 (C᎐O); m/z (EI) (%) 158 [M] (4), 143
᎐ ᎐
ϩ ϩ
[
(
M Ϫ CH ] (58), 127 [M Ϫ OCH ] (17), 85 (43), 75 (100), 69
3 3
ϩ
1
removed under reduced pressure to give 0.05 g of a mixture. H
15), 43 (37); m/z (CI, NH ) (%) 176 [M ϩ NH ] (10), 159
3 4
ϩ
NMR analysis showed 10a and 11a in a 2.3:1 ratio. The total
yield of carbene derived products was 25% by NMR spec-
troscopy, based on oxadiazoline (p-bromoacetophenone as
internal standard). Spectral characteristics of 11a were in keep-
[
M ϩ H] (100).
Conversion of 10b to 11b with pyridine
Dimethoxydimethyloxadiazoline 8a (0.10 g, 0.6 mmol) and
methyl acetoacetate (9b) (0.08 g, 0.7 mmol) were dissolved in
1
6
ing with those reported in the literature.
1
568
J. Chem. Soc., Perkin Trans. 2, 1997

View Article Online
1
2.5 ml of benzene and the solution was sealed into a therm-
Synthesis of spiro oxadiazoline 8b
olysis tube. After heating at 110 ЊC for 20 h, 0.33 g of pyridine
was added to the thermolysis tube and the mixture was heated
again for 20 h. Conversion of 10b to 11b was found to be com-
plete. The only product detectable by GC–MSD was 11b, iso-
lated in 0.06 g yield. Addition of pyridine before thermolysis
gave products of carbene trapping in trace amounts only.
As briefly described in ref. 7, the synthesis of 8b involved the
transamination of acetone semicarbazone with ethanolamine
to acetone 4-(2-hydroxyethyl)semicarbazone. Oxidation of the
latter with PhI(OAc) gave a spiro oxadiazoline unsubstituted at
2
N, which was then benzoylated with PhCOCl to afford 8b.
Acetone 4-(2-hydroxyethyl)semicarbazone. The procedure for
the preparation of this semicarbazone is based on that of
1
7
Thermolysis of dimethoxydimethyloxadiazoline (8a) in presence
of 3-methylpentane-2,4-dione (9c)
Iwao. A solution of the semicarbazone of acetone (19.6 g,
0.170 mol) and ethanolamine (11.1 g, 0.182 mol) was refluxed in
toluene (55 ml), under dry nitrogen, for 7 h. The solution was
allowed to cool to room temperature and precipitation of the
product occurred within a few hours. The product was then
collected by filtration and washed well with diethyl ether to give
A solution containing 0.20 g (1.2 mmol) of dimethoxydimethyl-
oxadiazoline 8a and 0.14 g (1.2 mmol) of 3-methylpentane-2,4-
dione 9c in 25.0 ml of benzene was sealed into a thermolysis
tube and heated at 110 ЊC for 20 h. Removal of the volatiles
under vacuum yielded 10c. The yield of 10c (26%, based on
1
7
22.0 g (81%) of a pale yellow solid, mp 89–94 ЊC (lit., 94–
1
Ϫ1
oxadiazoline) was determined by H NMR spectroscopy
96 ЊC). νmax(KBr)/cm 3379 (s), 3317 (s), 3210 (s), 3070 (m),
(
p-dimethoxybenzene as internal standard).
2927 (m), 2869 (w), 1652 (s), 1554 (s), 1458 (w), 1427 (m),
1365 (m), 1337 (m), 1281 (w), 1254 (m), 1212 (m), 1157 (w),
1132 (m), 1088 (w), 1068 (m), 980 (w), 884 (w), 836 (w), 773
(w), 722 (w); δH 1.84 (3H, s, CH ), 1.97 (3H, s, CH ), 3.39–
3
-Methyl-3-(dimethoxymethyl)pentane-2,4-dione (10c). νmax-
Ϫ1
(gas phase)/cm 3002 (m), 2959 (b), 2943 (b), 2840 (m), 1712
(
s), 1451 (b), 1360 (m), 1193 (m), 1110 (s), 1091 (s), 963 (w); δH
3
3
1
.38 (3H, s, CH ), 2.09 [6H, s, (CH CO) C], 3.49 [6H, s,
3.53 (2H, m, NCH ), 3.69–3.82 (2H, m, OCH ), 6.56 [1H, br
3
3
2
2
2
(
(
2
CH O) C], 4.90 [1H, s, CH(OCH ) ]; δ 13.3 (CH ), 27.3
s, C(O)NHC], 8.02 [1H, br s, C(O)NHN] (OH caused a slight
distortion in the baseline between 3 and 4 ppm); δC 16.36
(CH ), 25.20 (CH ), 42.85 (NCH ), 63.35 (OCH ), 147.85
3
2
3
2
C
3
CH CO), 58.6 [C(OCH ) ], 71.8 [CH(CH )], 107.8 [C(OCH ) ],
3 3 2 3 3 2
04.7 (CH CO); m/z (EI) (%) 145 (13), 115 (36), 114 (15), 99
3
3
3
2
2
(100), 97 (8), 75 (57), 47 (17), 43 (76) (Found: C, 57.3; H, 8.6%.
(C᎐N), 157.88 (C᎐O).
᎐ ᎐
Calc. for C H O : C, 57.4; H, 8.5%).
3,4,9-Triaza-2,2-dimethyl-1,6-dioxaspiro[4.4]non-3-ene.
9
16
4
1
8
Iodobenzene diacetate (7.47 g, 23 mmol), dissolved in
dichloromethane (150 ml), was added dropwise over 1–2 h, to a
stirred solution of acetone 4-(2-hydroxyethyl)semicarbazone
(3.50 g, 22 mmol) in dichloromethane (200 ml) at 0 ЊC, under
nitrogen. After addition was complete, the solution was kept in
the ice bath for another 2 h, and was then allowed to warm to
room temperature overnight. After evaporating approximately
one half of the solvent in vacuo, the solution was washed, first
Thermolysis of dimethoxydimethyloxadiazoline (8a) in presence
of 1,3-diphenylpropane-1,3-dione (9d)
A solution of dimethoxydimethyloxadiazoline (8a, 0.12 g, 0.8
mmol) and 1,3-diphenylpropane-1,3-dione (9d, 0.19 g, 0.8
mmol) in 25.0 ml of benzene was sealed into a resealable
thermolysis tube and heated at 110 ЊC for 20 h. Removal of the
solvent yielded 0.24 g of a mixture of product 10d in 56% yield
(p-bromoacetophenone as internal standard) and excess 9d.
with ice-cold aqueous NaHCO (5%) and then with ice-cold
3
Pure 10d was obtained from another thermolysis in which an
excess of oxadiazoline was used in order to convert 9d more
fully to 10d. A solution of 8a (0.15 g, 1.0 mmol) and 9d (0.08 g,
brine, before it was dried over MgSO . Evaporation of the sol-
4
vent afforded an oil (ca. 6 g), consisting of the oxadiazoline and
iodobenzene, which was generally used without purification in
the next step.
0
.4 mmol) in 25.0 ml of benzene, treated as described above,
gave 0.14 g (50%) of 10d.
-(Dimethoxymethyl)-1,3-diphenylpropane-1,3-dione (10d). δ
3,4,9-Triaza-9-benzoyl-2,2-dimethyl-1,6-dioxaspiro[4.4]non-
3-ene (8b). The crude mixture containing the NH oxadiazoline
was dissolved in dichloromethane (25 ml), and pyridine (1.10 g,
13.9 mmol) was added. Benzoyl chloride (1.86 g, 13.2 mmol)
was added dropwise to this stirred solution over ca. 10 min.
After stirring at room temperature for another 30 min, the solu-
tion was diluted with dichloromethane and washed successively
with water, aqueous NaHCO3 (5%) and brine. The organic
2
H
3
3
3
.4 [6H, s, C(OCH ) ], 5.32 (1H, d, J 7.9), 5.71 (1H, d, J 7.9),
3 2
7
.35 (4H, m, aromatic), 7.48 (2H, m, aromatic), 7.94 (4H, m,
aromatic); δ 56.7, 61.1, 106.2, 128.1, 128.6, 133.3, 136.5, 192.0;
C
ϩ
ϩ
m/z (EI) (%) 267 [M Ϫ OCH ] (5), 193 [M Ϫ C H CO] (40),
3
6
5
ϩ
ϩ
ϩ
1
05 [C H CO] (60), 77 [C H ] (25), 75 [(CH O) CH] (100).
6 5 6 5 3 2
Thermolysis of dimethoxydimethyloxadiazoline (8a) in presence
of anthrone 12
phase was dried over MgSO , the solvent was evaporated, and
4
hexane was added to the crude product to induce crystalliz-
ation. Recrystallization from methanol–ethyl ether yielded
1.41 g (25%, from the semicarbazone) of 8b as a white solid, mp
A solution containing 0.12 g (0.8 mmol) of dimethoxydimethyl-
oxadiazoline 8a and 0.15 g (0.8 mmol) of anthrone 12 in 25.0 ml
of benzene containing one drop of pyridine was sealed into
a resealable thermolysis tube and heated at 110 ЊC for 20 h.
Removal of the solvent yielded 0.19 g of a mixture of 13 and
excess 12. The yield (30%, based on oxadiazoline) was deter-
mined by NMR spectroscopy after the addition of an internal
standard. A similar thermolysis in which the oxadiazoline (0.15
g, 1.0 mmol) was in excess over 12 (0.11 g, 0.6 mmol) in 25.0 ml
of benzene containing two drops of pyridine afforded 0.17 g of
Ϫ1
108–110 ЊC (dec.). νmax(KBr)/cm 2994 (w), 2902 (w), 1655 (s),
1578 (w), 1453 (w), 1389 (s), 1272 (m), 1222 (m), 1162 (m), 1115
(m), 1081 (w), 1046 (w), 1011 (w), 973 (w), 921 (m), 856 (w), 817
(w), 798 (w), 722 (m); δ (300 MHz) (broad peaks caused by slow
H
exchange) 1.20–1.80 (6H, br m, 2 × CH ), 3.85–4.00 (1H, m,
3
NCH ), 4.05–4.20 (1H, m, NCH ), 4.25–4.45 (2H, m, OCH ),
2
2
2
7.35–7.60 (5H, m, aromatic); δ (75 MHz) 21.40 (CH ), 25.19
C
3
(CH ), 47.35 (NCH ), 65.40 (OCH ), 120.41 [C(CH ) ], 127.26
3
2
2
3 2
1
3 (89%, based on anthrone) which was pure enough to charac-
(aromatic C2, C6), 128.36 (aromatic C3, C5), 130.94 (aromatic
terize by NMR spectroscopy although it still contained a
detectable amount of anthrone.
C4), 133.22 (spiro C), 135.40 (aromatic C1), 169.00 (C᎐O); m/z
ϩ
ϩ
ؒ
(EI) (%) 233 [M Ϫ N ] (0.9), 192 [M Ϫ N Ϫ C H ] (5), 175
2
᝽
2
3
5
ϩ
9
-Anthracenyl dimethyl orthoformate (13). δH 3.54 [6H, s,
[M Ϫ N Ϫ C H O] (C H NO ) (27), 174 (30), 147 (15), 146
2 3 6 10 9 2
ϩ
(10), 145 (12), 131 (43), 117 (20), 105 [C H CO ] (100), 103 (21),
88 (25), 77 [C H ] (35), 51 (21).†
᝽
C(OCH ) ], 5.74 [1H, s, CH(OCH ) ], 7.45 (2H, m, aromatic)
3
2
3
2
6
5
ϩ
7
1
.5 (2H, m, aromatic), 7.95 (2H, m, aromatic), 8.23 (1H, s,
6 5
0-H), 8.50 (2H, m, aromatic); δC 51.3 [C(OCH ) ], 116.4
3
2
[
(
C(OCH ) ], 122.9, 123.1, 125.2, 125.3, 128.0, 132.0, 146.4; m/z
3 2
ϩ ϩ
EI) (%) 268 [M] (2), 237 [M Ϫ OCH ] (9), 221 (5), 193 (38),
3
†
Mass spectrometric experiments on this oxadiazoline involved LRP
1
65 (71), 164 (18), 163 (31), 139 (12), 115 (5), 75 (100), 63 (5), 47
and HRP electron impact (EI), collisional activation (CA) spectrometry
and chemical ionization (CI) using ammonia.
7
(30).
J. Chem. Soc., Perkin Trans. 2, 1997
1569

View Article Online
Thermolysis of 8b with 3-methylpentane-2,4-dione (9c)
A solution containing 8b (0.13 g, 0.50 mmol) was thermolyzed
with 9c (0.069 g, 0.61 mmol) in 10 ml of benzene. The major
product, 17c, was obtained as a colourless oil in 72% yield after
purification by radial chromatography (Chromatotron appar-
atus, silica gel-coated glass plate) with 10–30% ethyl acetate in
hexane as elution solvent.
matic C3, C5), 133.81 (aromatic C1), 158.14, 163.50 [OC᎐O,
᎐
ϩ
O᎐C–C(H)᎐C], 195.56 (CH C᎐O); m/z (EI) (%) 204 [M] (<1),
᎐
᎐
3
ϩ
161 (1), 105 [PhCO] (100), 77 (48), 51 (18), 43 (17). m/z (CI,
NH ) (%) 222 [M ϩ NH ] (13), 205 [M ϩ H] (13), 105
[PhCO] (100).
ϩ
ϩ
3
3
ϩ
Preparation of authentic (Z)-1-methyl-3-oxo-1-enyl benzoate (19)
Benzoyl chloride (0.74 g, 5.3 mmol) was added to a solution of
pentane-2,4-dione 9a (0.50 g, 5.0 mmol) and pyridine (0.42 g,
3
-[2-(3-Benzoyl)oxazolidinyl]-3-methylpentane-2,4-dione
Ϫ1
(
17c). νmax(CCl )/cm 2991 (w), 2950 (w), 2885 (w), 1800 (w),
4
5
.3 mmol) in 10 ml of CH Cl . After stirring at room temp-
1
(
9
705 (s), 1650 (s), 1581 (w), 1497 (w), 1449 (w), 1416 (m), 1377
m), 1356 (m), 1296 (w), 1203 (m), 1157 (w), 1081 (m), 1029 (w),
58 (w), 909 (w); δ (300 MHz) 1.44 (3H, s, CH ), 2.24 (3H, s,
2 2
erature overnight, the solution was diluted with 10 ml of
CH Cl and washed with water (2 × 10 ml), aq. NaHCO (5%,
2
2
3
H
3
2
× 10 ml), and water (10 ml) again, before it was dried over
COCH ), 2.29 (s, 3H, COCH ), 3.58–3.66 (2H, m, NCH ),
3
3
2
MgSO . Purification of 19 was done by chromatography (10–
3
.67–3.81 (1H, m, OCH ), 4.04–4.19 (1H, m, OCH ), 6.28 [1H,
4
2
2
1
13
2
0% ethyl acetate in hexane); H and C spectra were identical
to those of 19 isolated from thermolysis of 8b in the presence of
a.
s, oxazolidine C(2)H], 7.35–7.57 (5H, m, aromatic); δ (75.5
C
MHz) 13.22 (CH ), 27.72 (COCH ), 28.12 (COCH ), 48.90
3
3
3
9
(
NCH ), 66.55 (OCH ), 71.19 {[CH C(O)] C}, 90.75 (oxazol-
2 2 3 2
idine C2), 127.43 (aromatic C2, C6), 128.41 (aromatic C3, C5),
1
2
31.08 (aromatic C4), 135.18 (aromatic C1), 169.78 (NC᎐O),
Acknowledgements
᎐
03.25 (C᎐O), 204.26 (C᎐O), diastereotopicity not detected in
᎐
᎐
The authors thank the Natural Sciences and Engineering
Research Council of Canada for funding of this research.
P. C. and D. L. P. were the recipients of Ontario Government
and NSERC Scholarships, respectively.
the NMR spectra; m/z (EI) (%) 246 (7), 176 (benzoyloxazol-
ϩ
idinium cation) (16), 148 (7), 105 [PhCO] (100), 77 (41), 51
ϩ
(
10), 47 (27); m/z (CI, NH ) (%) 290 [M ϩ H] (7), 176 (benz-
3
oyloxazolidium cation) (100).
References
Thermolysis of 8b with pentane-2,4-dione (9a)
A solution of 8b (0.26 g, 1.0 mmol) was heated with 9a (0.11 g,
1
2
W. Kirmse, in Advances in Carbene Chemistry, ed. U. H. Brinker,
JAI Press, 1994, pp. 1–57.
X.-M. Du, H. Fan, J. L. Goodman, M. A. Kesselmayer, K. Krogh-
Jespersen, J. A. LaVilla, R. A. Moss, S. Shen and R. S. Sheridan,
J. Am. Chem. Soc., 1990, 112, 1920.
1
.1 mmol) in 20 ml of benzene. The thermolysis products were
separated by radial chromatography, by elution with ethyl
acetate–hexane mixtures or 8% MeOH in CHCl . The fractions
3
had to be rechromatographed several times in order to achieve a
good separation.
3 R. A. Moss, S. Shen and M. Wlostowski, Tetrahedron Lett., 1988,
29, 6417.
4
J. J. Zupancic, P. B. Grasse, S. C. Lapin and G. B. Schuster,
Tetrahedron, 1985, 41, 1471; J. E. Chateauneuf, J. Chem. Soc., Chem.
Commun., 1991, 1437; W. Kirmse, J. Kilian and S. Steenken, J. Am.
Chem. Soc., 1990, 112, 6399.
N-(2-Hydroxyethyl)benzamide. White solid (58% yield), mp
1
9
Ϫ1
6
1
0.5–61.5 ЊC (lit., 60 ЊC); νmax(KBr)/cm 3100–3600 (br),
637 (s), 1542 (s), 1491 (w), 1310 (m), 1067 (m), 712 (m); δ 3.09
H
(
1H, br s, OH), 3.56–3.66 (2H, m, NCH ), 3.75–3.84 (2H, m,
2
5 W. Kirmse, I. K. Strehlke and S. Steenken, J. Am. Chem. Soc., 1995,
117, 7007.
OCH ), 6.84 (1H, br s, NH), 7.30–7.57 (3H, m, aromatic meta,
2
para), 7.73–7.81 (2H, m, aromatic ortho); δC 42.74 (NCH ),
6 M. El-Saidi, K. Kassam, D. L. Pole, T. Tadey and J. Warkentin,
J. Am. Chem. Soc., 1992, 114, 8751; T. Wong, J. Warkentin and
J. K. Terlouw, Int. J. Mass Spectrom. Ion Processes, 1992, 115, 33;
L. Isaacs and F. Diederich, Helv. Chim. Acta, 1993, 76, 2454;
W. W. Win, M. Kao, M. Eiermann, J. J. McNamara, F. Wudl,
D. L. Pole, K. Kassam and J. Warkentin, J. Org. Chem., 1994, 59,
5871; D. L. Pole and J. Warkentin, Liebigs Ann., 1995, 1907;
A. de Meijere, S. I. Kozhushkov, D. S. Yufit, R. Boese, T. Haumann,
D. L. Pole, P. K. Sharma and J. Warkentin, Liebigs Ann., 1996, 601.
2
6
1.83 (OCH ), 126.93 (aromatic C2, C6), 128.47 (aromatic C3,
2
C5), 131.57 (aromatic C4), 134.00 (aromatic C1), 168.68 (C᎐O);
᎐
ϩ
m/z (EI) (%) 165 [M ] (2), 147 (12), 134 (10), 122 (19), 105
PhCO] (100), 84 (16), 77 (61), 51 (28). m/z (CI, NH ) (%) 166
ϩ
[
[
3
ϩ
M ϩ H] (100).
Compound 18. Isolated as a single diastereomer, in 2% yield,
Ϫ1
after up to six separations by chromatography. νmax(CCl )/cm
4
7
P. Couture, J. K. Terlouw and J. Warkentin, J. Am. Chem. Soc., 1996,
18, 4214.
1
669 (s), 1636 (m), 1617 (m), 1523 (m), 1486 (w), 1385 (m), 1288
1
(w), 1262 (m), 1149 (m), 1094 (m), 1024 (m), 965 (w), 909 (s),
8
J. E. Jackson, N. Soundararajan, M. S. Platz and M. T. H. Liu,
8
3
49 (w); δ (300 MHz) 2.25 (3H, s, CH ), 2.31 (3H, s, CH ),
H 3 3
J. Am. Chem. Soc., 1988, 110, 5595.
9 I. Fleming, Frontier Orbitals and Organic Chemical Reactions, John
Wiley and Sons, Chichester, 1976.
0 S. G. Mills and P. Beak, J. Org. Chem., 1985, 50, 1216.
1 M. Bassetti, G. Cerichelli and B. Floris, Tetrahedron, 1988, 44, 2997.
2 A. R. Eberlin and D. L. H. Williams, J. Chem. Soc., Perkin Trans. 2,
.60–4.30 (8H, m, NCH , OCH ), 5.25 [1H, br s, 2-alkyloxazol-
2
2
idine C(2)H], 7.20–7.65 [9H, m, 2-alkoxyoxazolidine C(2)H,
1
1
1
aromatic], 7.85–7.94 (2H, m, aromatic); δ (75.5 MHz) 15.14
C
(
C᎐CCH ), 29.18 (COCH ), 40.11 (NCH ), 47.40 (NCH ),
᎐
3 3 2 2
6
5.60 (OCH ), 69.81 (OCH ), 82.46 (2-alkyloxazolidine C2),
2 2
1
996, 1043.
1
13.75 (2-alkoxyoxazolidine C2), 116.84 (COC᎐C), 127.05
᎐
13 H. O. House, Modern Synthetic Reactions, Benjamin, New York,
1965, p. 164.
14 L. Claisen, Annalen, 1897, 297, 1.
5 H. Baba and T. Takemura, Tetrahedron, 1968, 24, 4779; T. Takemura
and H. Baba, Tetrahedron, 1968, 24, 5311; F. M. Menger and
R. R. Williams, J. Org. Chem., 1974, 39, 2131; J. C. Scaiano,
C. W. B. Lee, Y. L. Chow and G. E. Buono-Core, J. Photochem.,
(
aromatic C2, C6, 2 coincident signals from analogous car-
bons), 128.12 (aromatic C3, C5), 128.48 (aromatic C3, C5),
31.13 (aromatic C4), 131.39 (aromatic C4), 134.39 (aromatic
C1), 135.31 (aromatic C1), 167.35, 167.68, 168.76 (COC᎐C,
1
1
᎐
2
× NCO), 193.78 (CH CO); m/z (EI) (%) 303 (3), 286 (5), 260
3
ϩ
(3), 244 (4), 214 (2), 198 (3), 180 (5), 148 (66), 105 [PhCO]
1
982, 20, 327.
ϩ
(100), 77 (43); m/z (CI, NH ) (%) 468 [M ϩ NH ] (4), 451
16 L. Crombie, D. E. Games and A. W. G. James, J. Chem. Soc., Perkin
Trans. 1, 1978, 464.
17 J. Iwao, Pharm. Bull., 1956, 4, 247.
18 J. G. Sharefkin and H. Saltzman, Org. Synth., 1963, 43, 62.
9 K. Arai, S. Tamura, T. Masumizu, K.-I. Kawai and S. Nakajima,
3
4
ϩ
[
M ϩ H] (14), 286 (100).
Z)-1-Methyl-3-oxo-1-enyl benzoate (19). Colourless oil iso-
2
0
(
Ϫ1
lated in 2% yield. νmax(CCl )/cm 1741 (s), 1706 (w), 1673 (m),
4
1
1
637 (w), 1451 (w), 1426 (w), 1378 (w), 1359 (w), 1262 (s), 1195
Can. J. Chem., 1990, 68, 903.
20 A. Fürstner and D. N. Jumbam, Tetrahedron, 1993, 48, 5991.
(
(
(
m), 1177 (w), 1153 (m), 1082 (m), 1065 (m), 1024 (w); δ 2.18
H
3H, s, CH ), 2.22 (3H, s, CH ), 5.92 (1H, s, vinyl), 7.45–7.70
3
3
3H, m, aromatic meta, para), 8.08–8.16 (2H, m, aromatic
Paper 7/00169J
ortho); δ 21.51 (C᎐CCH ), 31.03 (COCH ), 117.40 (COC᎐C),
Received 2nd January 1997
Accepted 26th March 1997
C
3
3
1
28.67 (aromatic C2, C6), 129.02 (aromatic C4), 130.20 (aro-
1
570
J. Chem. Soc., Perkin Trans. 2, 1997