Reactions of allyloxy(methoxy)carbene in solution. Carbene rearrangement and Claisen rearrangement of the carbene dimer

Article abstract of DOI:10.1016/j.tet.2005.04.024

Allyloxy(methoxy)carbene, with and without deuterium in the α-position of the allyloxy group, was generated in benzene at 50 and at 110°C. At the higher temperature, the carbene fragmented to allyl and methoxycarbonyl radicals that subsequently coupled. At the lower temperature, most of the carbene dimerised. The structure of the major product and the distribution of deuterium indicated that the dimer underwent Claisen rearrangement at 50°C to methyl 2-allyloxy-2-methoxy-4-pentenoate. Facile rearrangement of the dimer was supported by the results of a computation which placed the barrier at about 18 kcal mol^-1.

Full text of DOI:10.1016/j.tet.2005.04.024

Tetrahedron 61 (2005) 5788–5796
Reactions of allyloxy(methoxy)carbene in solution. Carbene
rearrangement and Claisen rearrangement of the carbene dimer
b
Damian Plazuk,a John Warkentinb, and Nick Henry Werstiuk
*
aDepartment of Organic Chemistry, University of Lodz, Lodz, Poland
bDepartment of Chemistry, McMaster University, Hamilton, 1280 Main St. W., Hamilton, Ont., Canada L8S 4M1
Received 5 January 2005; revised 14 April 2005; accepted 14 April 2005
Abstract—Allyloxy(methoxy)carbene, with and without deuterium in the a-position of the allyloxy group, was generated in benzene at 50
and at 110 8C. At the higher temperature, the carbene fragmented to allyl and methoxycarbonyl radicals that subsequently coupled. At the
lower temperature, most of the carbene dimerised. The structure of the major product and the distribution of deuterium indicated that the
dimer underwent Claisen rearrangement at 50 8C to methyl 2-allyloxy-2-methoxy-4-pentenoate. Facile rearrangement of the dimer was
supported by the results of a computation which placed the barrier at about 18 kcal molK1
.
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
(4)/radical pair (5) mechanism best accounted for their
results.
Both [1,2]- and [2,3]-sigmatropic rearrangement of allylic,
heteroatom carbenes, as well as radical-pair formation, are
known. In 1977 and 1978, Iwamura and co-workers1
ascribed rearrangement of aryl(4,4-dimethylallyloxy)car-
benes (1) to a [1,2] shift, Scheme 1. In 1972, Baldwin and
Walker2 showed that S-methyl-(3,3-dimethylallyl)carbene
(2) undergoes a clean [2,3] sigmatropic rearrangement to a
dithiocarboxylic acid methyl ester at 65 8C, Scheme 1. In
the same year, Hoffmann and co-workers3 published a
detailed study of the gas phase thermolysis of 3a and 3b, at
250 8C, Scheme 2. They concluded that a sequential carbene
We had shown4 that two phenyl-substituted allyloxy-
(methoxy)carbenes (e.g., 6) fragment to radical pairs in
benzene at 110 8C, Scheme 3. Carbene 6, and the isomeric
species with Ph at the terminus of the double bond, favored
the product from apparent 1,2-migration. Although it was
evident that some, but not all, of the product could have
come from concerted 1,2-rearrangement, it was unclear
whether the product distribution reflected some sigmatropic
rearrangement or whether it could also be explained in terms
of a concurrent radical-pair mechanism, with a cage effect
operating in solution. Very rapid radical-pair coupling, in
systems that do not involve a molecule, such as N2, between
the initially-formed radicals, might occur without complete
rotational equilibration, favoring re-attachment at the allylic
site originally bonded to oxygen. Such coupling would be
indistinguishable from concerted [1,2]-migration. [1,2]-
Migration is theoretically unfavorable in comparison with
[2,3]-sigmatropic rearrangement, but it can eclipse the latter
if steric demand is high.1
In the gas phase study of Hoffmann and co-workers, a cage
effect could not operate and it was pointed out that the high
temperature (250 8C) favored the radical-pair mechanism on
the grounds of entropy.3
Scheme 1.
Keywords: Allyloxy(methoxy)carbene; Carbene dimer; Claisen;
Rearrangement.
We decided to examine the rearrangement of allyloxy-
(methoxy)carbene with precursors 7a–d (Scheme 4), from
which 4a and 4b can be generated in solution at 110 8C (7a,
*
Corresponding author. Tel.: C1 905 525 9140x23478; fax: C1 905 522
2509; e-mail: warkent@mcmaster.ca
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.04.024

D. Plazuk et al. / Tetrahedron 61 (2005) 5788–5796
5789
Scheme 2.
Scheme 3.
Scheme 5.
benzene and in carefully dried benzene, in a sealed tube to
determine the effects of adventitious water. tert-Butyl
alcohol, or the stable free radical TEMPO, were added in
separate experiments to ascertain whether both a carbene
and radicals could be trapped. After 24 h, the solutions were
analyzed by gas chromatography.
3. Thermolysis of 7a in ordinary benzene
Scheme 4.
A solution of 7a in dry benzene at 110 8C for 24 h gave one
major product and several minor products noticeable by GC.
Thinking that some of those could be caused by adventitious
water, we ran the reaction in ordinary benzene which gave,
in addition to a solvent peak, six fractions large enough to be
collected by GC. Yields from reactions in ordinary benzene
are not relevant because such reactions were used only to
identify the minor components that were always detectable
by GC. Fraction one, not collected, contained benzene and,
presumably, acetone and other volatile products. Fractions
2–7 were methyl 3-butenoate (9), methyl 2-propenyl
carbonate (10), allyl dimethyl orthoformate (11), allyl
methyl 2-propenyl orthoformate (12), diallyl methyl
orthoformate (13) and methyl benzoate (14). Yields in
Scheme 6 pertain to reaction in dry benzene (distilled from
CaH2); conditions for detecting but not isolating 11 and 13.
7b) or 50 8C (7c, 7d). Sigmatropic rearrangement, as
observed by Baldwin and Walker,2 might well become a
competitive (or exclusive) process for 4, especially at the
lower temperature, given that fragmentation should be less
likely than was the case with the phenyl-substituted
allyloxycarbenes, such as 6, which were studied
previously.4
Results of that study, as well as Claisen rearrangement of
1,2-diallyloxy-1,2-dimethoxyethene under relatively mild
conditions, are reported.
2. Methods, results, and discussion
2-Allyloxy-2-methoxy-5,5-dimethyl-D3-1,3,4-oxadiazoline
(7a) and the dideuterio analogue (7b), as well as the
corresponding 5-methyl-5-(p-methoxyphenyl) compounds,
7c and 7d were prepared by the route5 shown in Scheme 5.
Compound 7a was thermolysed at 110 8C, in ordinary
Probable origins of products 9–14, orthoformates 11 and 13
coming from capture of the carbene by adventitious water,
are shown in Scheme 7. Formation of carbonates, such as
10, is a known process in the thermolysis of oxadiazolines

5790
D. Plazuk et al. / Tetrahedron 61 (2005) 5788–5796
Scheme 6.
Scheme 7.
analogous to 7a6 and the [1–4] sigmatropic shift to afford 12
is a known reaction of carbonyl ylides, particularly in
acetoxy analogues.7,8 Methyl benzoate (14) is formed by
attack of the methoxycarbonyl radical on benzene.
Thermolysis of 7a in the presence of tert-butyl alcohol gave
allyl tert-butyl methyl orthoformate (15), in support of the
intermediacy of carbene 4a, Scheme 8.
Scheme 8.
Scheme 9.

D. Plazuk et al. / Tetrahedron 61 (2005) 5788–5796
5791
Scheme 10.
Thermolysis of 7a in the presence of TEMPO and collection
by GC afforded the known adducts of the methoxy-
carbonyl4b,9 and allyl10 radicals, 16 and 17, respectively,
Scheme 9. Yields of 16 and 17 were 13 and 18%,
respectively. Methyl 3-butenoate (9, 17%) was also
obtained.
We interpret the high temperature results as follows. First,
we can dismiss compounds 10–14, Scheme 6, because they
are really minor when dry benzene is used and we know
their origins (Scheme 7). Second, we know that allyloxy-
(methoxy)carbene is formed from 7a,b because that carbene
can be captured by tert-butyl alcohol. Third, a part of the
carbene from 7a undergoes scission to a radical pair if it is
not trapped quickly, as indicated by the finding of methyl
benzoate (Scheme 6) and the TEMPO adducts of the
methoxycarbonyl and allyl radicals (Scheme 9). The low
yields of the TEMPO adducts suggests that methyl
3-butenoate (9) arises largely by sigmatropic or 1,2-
migration in the carbene. Efficient, in cage radical coupling,
relative to radical separation, cannot be ruled out because
TEMPO would not trap radicals that are not separated by
diffusion.
Thermolysis of 7b at 110 8C gave 18 (1.3 parts) and 19
(1 part), Scheme 10. With added TEMPO, adducts of
deuterated allyl and TEMPO (12%, 1 part a- to 1.2 parts
g-deuterated) were obtained as well as deuteriated methyl
butenoates (27%, 1.7 parts a- to 1 part g-deuteriated).
Thermolysis of 7b in the presence of tert-butyl alcohol gave
21 (60%), in which all of the deuterium was alpha.
In contrast, at 50 8C, 7c gave methyl-3-butenoate (9) (35%)
and 22 (54%), while in the presence of tert-butyl alcohol it
afforded 15 (60%), Scheme 11. Under the same conditions,
7d gave 18, 19, and 24, Scheme 12. In the presence of
TEMPO, 7c and 7d afforded only trace amounts of adducts
16 and 17 (presumably D-labeled from 7d).
Thermolysis of 7b in the presence of TEMPO (Scheme 10,
top) gave a 1.7:1 ratio (a- to g-deuteriated) of ester products
rather than the 1:1 ratio observed by Hoffmann et al.3 in
their gas-phase study. This result implies that in concerted
processes [1,2]- is favored over [2,3]-rearrangement. The
allyl fragments captured by TEMPO are nearly equally
deuteriated at the a- and g-positions, as expected. Those
adducts are expected to form in 1:1 ratio and we cannot
explain the apparent 1.2:1 ratio obtained from the 2H NMR
spectra.
In the absence of TEMPO, where radical processes compete
(Scheme 10, center) the ester ratio was down from 1.7:1 to
1.3:1, implying that the portion of ester from radical
coupling is formed in a ratio closer to the 1:1 ratio expected
from radicals separated by diffusion. If we assume that
Scheme 11.
Scheme 12.

5792
D. Plazuk et al. / Tetrahedron 61 (2005) 5788–5796
Scheme 13.
Scheme 14.
TEMPO captures essentially all of the diffusion-separated
radicals, then 1.7:1 represents the ratio of the methyl
4-butenoates arising from in-cage coupling, or the net effect
of 1,2-migration superimposed on in-cage coupling. A cage
effect, favoring coupling of the methoxycarbonyl- and the
allylic radical at the allyl site originally bonded to the
carbene oxygen, had been observed before,4 in a system
more likely than 7b to react by the radical pair mechanism.
The analysis cannot be more detailed because we do not
know what fraction of the radicals is not scavenged by
TEMPO but leads to methyl 3-butenoate without ever
becoming free. Alternatively, there could be some 1,2-
migration in the carbene, in competition with radical-pair
formation. What is certain is that there is a radical pair
component and that sigmatropic rearrangement, as observed
by Baldwin2 for a thio analogue, can be ruled out as the
major mechanism because it would have led primarily to 19.
These results led us to look for a new mechanism for the
formation of 22. One could arrive at that structure by means
of a Claisen rearrangement of the dimer of 4a, which is an
allyl vinyl ether twice over, Scheme 14. The dimer was not
observed, but that is not surprising in view of the large
substituent effects on rearrangement rates. For example, an
alkoxy substituent at C-2 can lead to facile rearrangement at
about 35 8C.12 Coupling of two a-deuteriated carbenes,
followed by Claisen rearrangement, would lead to 24 with
a-deuterium in the allyloxy group and g-deuterium in the
allyl group, as shown in Scheme 14. That was shown to be
the case with the 1H NMR spectrum. Whereas the 1H NMR
spectrum of 22 has the a-CH2 signal of the OCH2CH]CH2
group at 3.96 and 4.06d, the spectrum of 24 did have any
absorption at that position, indicating that the allyloxy group
in 24 was OCD2CH]CH2. Moreover, whereas 22 has the
terminal vinyl signals of the CCH2CH]CH2 group at 5.12
and 5.14d, that of 24 did not absorb in that region, indicating
that its C-allyl group was CH2CH]CD2. The rest of the
signals from 24 were in agreement with that structure,
taking into account the effects of deuterium. These results
are in keeping with coupling of the carbene to generate a
tetraalkoxyethene and sigmatropic Claisen rearrangement
of the latter. Coupling of dialkoxycarbenes that do not
rearrange fast has been observed before13 and all that is
required in this case is that both concerted rearrangement
and fragmentation to radical pairs be slow at 50 8C.
At the lower temperature available with 7c, fragmentation
of carbene 4a to radicals was barely detectable, as indicated
by the negligible yields of TEMPO adducts. The carbene
was still formed, as indicated by the formation of 9 as well
as the trapping with tert-butyl alcohol, Scheme 11. We were
surprised to find 22, a strikingly-new product that we had
never seen at the higher temperature required to thermolyze
7a. A simple rationale for compound 22, based on the
precedented ring expansion when strained carbonyl com-
pounds are treated with dimethoxycarbene,11 is shown in
Scheme 13. That result might follow if fragmentation (and
rearrangement) of the carbene were slow at 50 8C,
permitting it to accumulate sufficiently to attack 9 by
addition to carbonyl carbon. Although the mechanism had
attractive features, it would be surprising if the nucleophilic
carbene 4a were to add efficiently to the carbonyl group of
9. This simple explanation lost credence when it was found
that inclusion of ester 9 at the outset actually decreased the
yield of 22. Moreover, inclusion of ethyl phenylacetate did
not lead to even a trace of the product of carbene attack on
that ester, Scheme 13.
4. Computational studies
The Claisen rearrangement of carbene dimer 25H to 22 was
modeled with Gaussian 0314a at the DFT level of theory
using the Becke3PW91 functional.14b,c The optimized
equilibrium geometry of 25H and transition state 25H-TS
for its rearrangement were obtained at the B3PW91/6-31C
G(d) level using the default convergence criteria.
Frequency calculations were performed to characterize 25H
and 25H-TS. The latter, as expected, exhibited one

D. Plazuk et al. / Tetrahedron 61 (2005) 5788–5796
5793
Figure 1. Geometrical structures of (a) (E)-1,2-diallyloxy-1,2-dimethoxyethene (25H) and (b) transition state (25H-TS) for [3,3] Claisen rearrangement of (E)-
25H with selected inter-nuclear distances in angstroms.
Table 1. Thermochemical data for [3,3] Claisen rearrangement of (E)-1,2-diallyloxy-1,2-dimethoxyethene at the B3PW91/6-31CG(d) level
Eo (EelecCZPE)a
Ha,b (ECRT)
Ga,c
a
Compound
Eelec
25H
25H-TS
K691.213958
K691.184465
18.50
K690.964018
K690.936315
17.38
K690.945705
K690.918880
16.84
K691.012480
K690.981539
19.41
Barrier (kcal molK1
)
a In hartrees.
b Corrected to 298.15 K. EZEoCEvibCErotCEtrans
c Corrected to 298.15 K. GZHKTS.
.
imaginary frequency (K432.0 cmK1). Moreover, animation
of the vibration showed that it connected the reactant to the
product. The conformations of (E)-25H and 25H-TS (the
transition state for the [3,3] sigmatropic rearrangement) that
were used in calculating the barrier are displayed in Figure 1
and the thermochemical data are listed in Table 1. The
Claisen transition state 25H-TS is chair-like, but it is quite
6. Experimental
6.1. General
Allyl compounds of the type XCH2CH]CHcisHtrans, in
general, have low field H NMR spectra in which Hcis and
1
H
trans appear as doublets. At high field or in expanded-scale
˚
unsymmetrical. The C–C bond (2.438 A) being formed is
low field spectra, those doublets are split further, to show
geminal coupling, for example. Many of the spectra below
are simplified, to have the appearance of low field spectra
although a signal described as a doublet can actually look
more complex if it is expanded.
˚
significantly longer than the C–O bond (1.870 A) being
broken. That barrier is in the region of 17–18 kcal molK1
and is in good accord with the facile reaction observed
experimentally; formation and rearrangement of the dimer
in benzene was complete (NMR) at 50 8C in 24 h.
1H NMR spectra were obtained from solutions in C6D6, or in
CDCl3, with Bruker spectrometers operating at 200 or
600 MHz, unless otherwise indicated, and are referenced to
the signal at 7.16 attributed to C6HD5 or to CHCl3 at
7.26 ppm. 2H NMR spectra were obtained at 92.1 MHz and
are referenced to natural abundance deuterium in the
benzene or chloroform solvent. 13C NMR spectra, run at
50.9 or 150.9 MHz, in C6D6 or in CDCl3, are referenced to
the center line of the solvent triplet at 128.1 or 77.2 ppm,
respectively. Mass spectra were run with a Micromass/
Waters GCT time of flight instrument. 2,2-Dialkoxyoxa-
diazolines generally do not afford molecular ions in the
mass spectrometer under any conditions and it was not
possible to get high resolution data for compounds 7a–7d.
By NMR spectroscopy, they were O95% pure.
5. Conclusions
The reactions exhibited by allyloxy(methoxy)carbene in
solution are very temperature dependent. At 110 8C, a
fraction of the carbene fragments to a radical pair that
couples to afford methyl 3-butenoate. Alpha deuteriated
carbene (MeOCOCD2CH]CH2) rearranges with a
preference for methyl 2,2-dideuterio-3-butenoate,
suggesting incomplete rotational equilibration (a cage
effect) in the radical pair and/or some competition from a
1,2-rearrangment.
At 50 8C, fragmentation to a radical pair is not important (an
entropic effect) and it appears that unimolecular reactions
are slow enough to permit the carbene to dimerise. A major
product (22) is one attributed to Claisen rearrangement of
GC separations were performed with a 60!4.0 mm
(internal) column containing 10% OV 101. The helium
flow rate was 40 mL minK1 and the temperature program
was 60 8C for 3 min, then 5 8C per minute to 200 8C.
Infrared spectra were taken with neat samples on a Bruker
Tensor 27 FTIR instrument equipped with a Harrick ATR
accessory and a ZnSe crystal.
the carbene dimer (25). The low barrier (17–18 kcal molK1
)
computed at the DFT level provides support for the
conclusion that 22 is formed via Claisen rearrangement of
the carbene dimer 25.

5794
D. Plazuk et al. / Tetrahedron 61 (2005) 5788–5796
6.1.1. 2-Acetoxy-2-methoxy-5,5-dimethyl-D3-1,3,4-oxa-
diazoline (8a). This compound has been reported.16a
Compound 7c (pale yellow oil, major diastereomer, ca.
75%). H NMR (200 MHz, C6D6) d: 1.66 (s, 3H), 3.20 (s,
1
3H), 3.23 (s, 3H), 4.37 (t, JZ5.2 Hz, 2H), 4.95 (d, JZ
10.4 Hz, 1H), 5.24 (d, JZ18.0 Hz, 1H), 5.84 (m, 1H), 6.72
(d, JZ8.8 Hz, 2H), 7.53 (d, JZ8.8 Hz, 2H); 13C NMR
(150.9 MHz, CDCl3) d: 26.4, 52.2, 55.2, 65.9, 113.9, 117.3,
121.3, 126.5, 130.5, 133.6, 138.2, 159.8; minor dia-
6.1.2. 2-Acetoxy-2-methoxy-5-(4-methoxyphenyl)-5-
methyl-D3-1,3,4-oxadiazoline (8b) (single isomer,
purified by column chromatography). IR (cmK1): 1747;
1H NMR (600 MHz, CDCl3) d: 1.61 (s, 3H), 2.07 (s, 3H),
3.32 (s, 3H), 3.40 (s, 3H), 6.80 (d, JZ7.2 Hz, 2H), 7.61 (d,
JZ7.2 Hz, 2H); 13C NMR (50.9 MHz, C6D6) d: 20.8, 25.0,
52.6, 54.8, 114.3, 124.5, 127.0, 128.9, 131.5, 160.3, 166.2.
1
stereomer (ca. 25%); H NMR (200 MHz, CDCl3) d: 3.39
(s, 3H), 4.19 (t, JZ6.1 Hz, 2H), 4.90 (d, JZ10.5 Hz, 1H),
5.17 (m, 1H), 5.76 (m, 1H); 13C NMR (150.9 MHz, CDCl3)
d: 26.2, 65.8, 117.0, 126.6, 130.3, 130.5, 133.2, 133.3 (4
signals not resolved).
6.1.3. Synthesis of 7a and 7b. To a solution of crude 8a
(7.32 g, ca. 4.7 g of pure 8a) and 3.7 g of allyl alcohol (3.8 g
of 1,1-dideuterioallyl alcohol) in methylene chloride
(100 mL) was added p-toluenesulfonic acid (ca. 100 mg).
The solution was stirred at room temperature for 3 days
before it was washed twice with 50 mL of saturated
NaHCO3 solution and dried over MgSO4. Evaporation of
the solvent and column chromatography on SiO2 with
hexane (9 parts)/ethyl acetate (1 part) gave 7a (7b) as
colorless oils in 50–55% yield.
Compound 7d (pale yellow oil, major diastereomer, ca.
72%). 1H NMR (600 MHz, CDCl3) d: 1.79 (s, 3H), 3.37 (s,
3H), 3.81 (s, 3H), 5.22 (d, JZ10.8 Hz, 1H), 5.35 (d, JZ
17.4 Hz, 1H), 6.91 (d, JZ8.8 Hz, 2H), 7.49 (d, JZ8.8 Hz,
2H); 2H NMR (92.1 MHz, CHCl3) d : 4.38; 13C NMR
(150.9 MHz, CDCl3) d: 26.4, 52.2, 55.2, 113.9, 117.3,
121.3, 126.5, 130.5, 133.6, 138.2, 159.8; minor dia-
1
stereomer (ca. 28%): H NMR (600 MHz, CDCl3) d: 1.69
(s, 3H), 3.60 (s, 3H), 5.13 (d, JZ10.8 Hz,1H), 5.21 (d, JZ
2
17.4 Hz, 1H), 5.83 (m, 1H) (2 signals not resolved); H
NMR (92.1 MHz, CHCl3) d: 4.12; 13C NMR (150.9 MHz,
CDCl3) d: 26.2, 67.8, 113.9, 117.0, 126.6, 130.3, 133.2 (5
signals not resolved).
6.1.4. Synthesis of 7c and 7d. To a solution of Pb(OAc)4
(12.4 g, 28 mmol) in methylene chloride (50 mL) was
added, slowly at 0 8C, a solution of the methoxycarbonyl-
hydrazone of 4-methoxyacetophenone (7.2 g, 28 mmol) in
methylene chloride (30 mL).15 The mixture was stirred at
0 8C for 7 h before it was washed with saturated NaHCO3
solution and dried over MgSO4. Evaporation of the solvent
at 5–10 8C left a crude product that was a mixture of 8b (ca
37%) and the acyclic isomer13b,16 8c. To a solution of the
mixture (6.0 g, 21 mmol) in 50 mL of methylene chloride
was added 1.0 g of allyl alcohol, or 1,1-dideuterioallyl
alcohol.17 After 4 h of stirring at room temperature the
solution was washed twice with 20 mL of saturated
NaHCO3 solution and dried with MgSO4 before the solvent
was evaporated at 5–10 8C. Column chromatography on
150 mL of SiO2, with pentane (85 parts)/diethyl ether (15
parts) gave 7c (7d) as pale yellow oils in about 49% yield.
The diastereomers were not separated and the spectra that
follow are those of the mixtures. It was possible to see the
signals from the major diastereomer easily, but not all of the
signals from the minor diastereomer were cleanly resolved.
Isomer ratios were estimated from resolved signals.
Compound 8d (yellow oil). IR (cmK1) 1765; 1H NMR
(200 MHz, C6D6) d: 1.73 (s, 3H), 2.01 (s, 3H), 3.21 (s, 3H),
3.22 (s, 3H), 6.72 (d, JZ8.8 Hz, 2H), 7.48 (d, JZ8.8 Hz,
2H); 13C NMR (150.9 MHz, C6D6) d: 21.9, 24.2, 55.1, 55.4,
102.0, 114.2, 127.3, 130.8, 159.9, 162.0, 168.6.
6.1.5. Thermolysis of 7a in dry benzene. A solution of 7a
(239 mg, 1.5 mmol) in ordinary benzene (1 mL) at 110 8C
for 24 h gave six products that were collected directly by
GC. Yields were estimated from calibration graphs prepared
by injection of benzene solutions of 9, 10 and 11 and each
yield was estimated from 2 or more runs.
6.1.6. Methyl 3-butenoate (9).18 Yield 60%. 1H NMR
(200 MHz, C6D6) d: 2.77 (d of t, 2H); 3.27 (s, 3H), 4.86–
4.98 (m, 2H), 5.78–5.97 (m, 1H). MS (EI) m/z: 100 (MC,
80%), 59 (96%), 41 (100%). The NMR spectrum matched
that of an authentic sample of the ester, prepared by treating
3-butenonitrile with methanol.
1
Compound 7a. H NMR (200 MHz, C6D6) d: 1.54 (s, 6H),
6.1.7. Methyl-3-butenoate-d2. 1H NMR (200 MHz, C6D6)
d: 2.78 (d, JZ7.0 Hz, 2H), 3.27 (s, 3H) 5.79–5.93 (m, 1H);
2H NMR (92.1 MHz, C6H6) d: 4.92; HRMS (CI, NH3) m/z:
calcd for C5H7D2O2 (MCH)C, 103.0728 found 103.0724.
3.25 (s, 3H), 4.22–4.28 (m, 2H), 4.99 (d, JZ10.4 Hz, 1H),
5.14 (d, JZ17.2 Hz, 1H), 5.70– 5.89 (m, 8 lines, 1H); 13C
NMR (50.9 MHz, C6D6) d: 23.9, 24.0, 51.7, 65.7, 116.6,
119.1, 134.1, 137.8; 13C NMR (50.9 MHz, CDCl3) d: 24.0,
24.1, 52.0, 65.7, 117.2, 119.3, 133.4, 137.0. Similar
oxadiazolines have absorptions at about 119 (C5) and at
137–138 (C2) ppm in their 13C NMR (CDCl3) spectra.13c
6.1.8. Allyl methyl carbonate (10).19 Colorless liquid.
Yield (5.7%). H NMR (600 MHz, C6D6) d: 3.31 (s, 3H);
1
4.37 (d, JZ6.0 Hz, 2H); 4.91 (d, JcisZ10.8 Hz, 1H), 5.08
(d, JtransZ17.4 Hz, 1H), 5.64 (m, 1H).
Compound 7b. 1H NMR (600 MHz, CDCl3) d: 1.55 (s, 3H),
1.56 (s, 3H), 3.25 (s, 3H), 5.18 (d, JZ10.4 Hz, 1H), 5.30 (d,
JZ17.2 Hz, 1H), 5.91 (m, 1H); 2H NMR (92.1 MHz,
CHCl3) d: 4.21 (s), 4.28 (s); 13C NMR (50.9 MHz, CDCl3)
d: 24.1, 52.0, 117.5, 118.9 (C5), 119.3, 133.3, 137.0 (C2)
(a signal from OCH2, which would be expected at ca.
65 ppm, was absent); MS (ESI) (m/z): 211.1 (MCNa)C,
187.1 (MKH)C, 157.0 (MKOMe)C.
6.1.9. Allyl dimethyl orthoformate (11). Colorless liquid.
1
Yield (!3%). H NMR (200 MHz, C6D6) d: 3.14 (s, 6H),
3.99 (d, 2H); 4.96 (s, orthoformyl H, partly superimposed on
5.00 (d, JZ9.5 Hz, composite 2H), 5.23 (d, JZ17 Hz, 1H),
5.77 (m, 1H). 13C NMR (150.9 MHz, C6D6) d: 51.0, 65.0,
113.9, 116.1, 134.9; MS m/z: calcd for C6H12O3 132.0786

D. Plazuk et al. / Tetrahedron 61 (2005) 5788–5796
5795
1
found 132.0779. The structure was confirmed with the H
NMR spectrum of an authentic sample, prepared by reaction
of trimethyl orthoformate (32 g, 0.30 mol) with allyl alcohol
(8.7 g, 0.15 mol) in the presence of a catalytic amount of
p-toluenesulfonic acid. The mixture was refluxed for 60 h,
before it was washed with aqueous NaHCO3 and extracted
with methylene chloride. The allyl dimethyl orthoformate
(4 g) was isolated by fractional distillation.
(5.0 mL) containing TEMPO (451 mg, 2.94 mmol) gave 16
(13%) and 17 (18%) that were separated and collected by
GC. Compound 9 was also obtained (15%) as well as
carbonate 10 (2.5%).
1-[(Methoxycarbonyl)oxy]-2,2,6,6-tetramethylpiperidine
(16). This compound gave the H NMR spectrum reported
1
in the literature.4b
2,2,6,6-Tetramethyl-1-(2-propenyloxypiperidine) (17). The
1H NMR spectrum of this compound matched that
reported.10
6.1.10. Allyl methyl 2-propenyl orthoformate (12).
Colorless liquid. Yield (1–2%). 1H NMR (600 MHz,
C6D6) d: 1.72 (s, 3H); 3.18 (s, 3H); 4.04 (br s, 3H,
diastereotopic OCH2 plus 1H of C]CH2); 4.32 and 4.33 (d,
JZ3.3 Hz, 1H, other H of C]CH2); 4.98 (d, JZ10.6 Hz,
1H), 5.22 (d, JZ17 Hz, 1H), 5.57 (s, 1H), 5.81 (m, 1H).
Gradient HSQC and HMBC spectra were in agreement with
the assignment. 13C NMR (150.9 MHz, C6H6) d: 20.0, 49.6,
64.0, 86.6, 110.5, 116.6, 133.8, 151.8; HRMS (CI, NH3)
m/z: calcd for C8H13O3 (MKH)C 157.0865 found
157.0837.
6.2.4. Thermolysis of 7b in benzene. A solution of 7b
(243 mg, 1.29 mmol) in 4.4 mL of benzene was heated at
110 8C for 24 h. The workup was the same as those
described above for 7a.
6.2.5. Thermolysis of 7b in benzene containing TEMPO.
A solution of 7b (212 mg, 1.12 mmol) and TEMPO
(399 mg, 2.62 mmol) in benzene (3.7 mL) was heated at
110 8C for 24 h. Workup was analogous to that described for
7a.
6.1.11. Diallyl methyl orthoformate (13). Colorless liquid.
Yield (2%). H NMR (200 MHz, C6D6) d: 3.16 (s, 3H);
1
3.99–4.03 (m, 4H), 5.00 (d, JZ10.4 Hz, 2H), 5.10 (s, 1H),
5.24 (d, JZ17 Hz, 2H) 5.73–5.89 (m, 2H); 13C NMR
(150.9 MHz, C6D6) d: 51.0, 64.3, 112.3, 115.3, 134.1;
HRMS (CI, NH3) m/z: calcd for C8H15O3 (MCH)C
159.1021 found 159.1021.
6.2.6. Dideuterioallyl adducts of TEMPO. The mixture of
a,a- and g,g-dideuterioallyl adducts was isolated by GC; 1H
NMR (200 MHz, CDCl3) d: 1.11 (s, 6H), 1.16 (s, 6H), 1.49
(m, 6H), 4.28 (d, JZ5.3 Hz, 0.71H, therefore 1.29D), 5.12
(d, JZ10.5 Hz) and 5.27 (d, JZ17.2 Hz), composite area
0.90H, therefore 1.1D), 5.88 (m, 1H). The integrations give
the isomer ratio (a,a: g,g)Z1.17:1. MS (ESI) m/z: 200.1
(MCH)C.
6.1.12. Methyl benzoate (14). Colorless liquid. Yield (1%).
1H NMR (200 MHz, C6D6) d: 3.47 (s, 3H), 7.06 (m, partly
obscured by solvent signal), 8.09–8.14 (dd, 2H). Its identity
was confirmed by running the 1H NMR spectrum of
authentic methyl benzoate in C6D6.
6.2.7. Thermolysis of 7c. Thermolysis of 7c at 50 8C and
workup as described for 7a gave methyl 3-butenoate (35%)
and compound 22, 54%. In the presence of dry methyl-3-
butenoate (conditions like those for 7a above) the yield of
22 dropped. Addition of dry ethyl phenylacetate to a
benzene solution of 7c did not produce any benzyl-
substituted analogue of 22.
6.2. Standard procedures for thermolysis of 7a–7e
An oxadiazoline (7a,b) (1.4–1.5 mmol) in dry benzene
(4.8 mL) was heated at 110 8C for 24 h in a sealed tube
alone or with either TEMPO or tert-BuOH added. The
products were separated directly by means of GC. Yields
were estimated from calibration graphs and each yield was
determined from two or more runs.
Methyl 2-allyloxy-2-methoxy-4-pentenoate (22). Colorless
1
oil, IR (cmK1) 1735; H NMR (600 MHz, CDCl3) d: 2.67
(dt, JZ7.2, 1.2 Hz, 2H, H-3), 3.32 (s, 3H, H-10), 3.77 (s,
3H, H-9), 3.96 (ddt, JZ12.6, 6.0, 1.2 Hz, 1H, H-6), 4.06
(ddt, JZ12.6, 6.0, 1.2 Hz, 1H, H-6), 5.12 (s, 1H, H-5, cis),
5.14 (dd, JZ8.4, 1.2 Hz, 1H, H-5, trans), 5.18 (dd, J Z10.8,
1.2 Hz, 1H, H-8, cis), 5.32 (dd, JZ17.4, 1.5 Hz, 1H, H-8
trans), 5.65–5.71 (m, 1H, H-4), 5.91–5.97 (m, 1H, H-7); 13C
NMR (150.9 MHz, CDCl3) d: 39.1, 50.0, 52.5, 64.0, 102.3,
117.2, 119.3, 130.9, 134.0, 169.3; HRMS (EI) m/z: calcd
for C9H13O3 (MKOMe)C 169.0865, found 169.0855; calcd
for C7H11O4 (MKCH2CH]CH2)C 159.0657, found
159.0640; calcd for C8H13O2 (MKMeOCO)C 141.0916,
found 141.0902.
6.2.1. Thermolysis of 7a in dry benzene. Thermolysis of
7a in benzene that was freshly distilled from CaH2 gave 7
(60%). The other products (Scheme 6) gave only small
displacements from the baseline of the GC trace.
6.2.2. Thermolysis of 7a in benzene containing tert-butyl
alcohol. Thermolysis of 7a (261 mg, 1.40 mmol) in benzene
(5.0 mL) containing 424 mg (5.73 mmol) of tert-butyl
alcohol gave 15 as a colorless oil, collected by means of
GC. Yield 47%. 1H NMR (600 MHz, C6D6) d: 1.15 (s, 9H),
3.22 (s, 3H), 4.08 (m, 2H), 5.03 (d, JZ10.2 Hz, 1H), 5.28
(d, JZ17.4 Hz, 1H), 5.33 (s, 1H), 5.88 (m, 1H). 13C NMR
(150.9 MHz, C6D6) d: 28.8, 49.8, 64.0, 73.8, 109.5, 115.7,
135.4. HRMS (EI) m/z: calcd for C8H15O2 (MKOMe)C
143.1072, found 143.1072. Carbonate 10 (6.6%) was also
obtained.
6.2.8. Thermolysis of 7d. Thermolysis of 7d at 50 8C in dry
benzene and workup as described for 7a gave methyl
2-allyloxy-2-methoxy-pentenoate (50%,
deuteriated, numbering system in Scheme 14).
partially
1
6.2.3. Thermolysis of 7a in benzene containing TEMPO.
Thermolysis of 7a (254 mg, 1.37 mmol) in dry benzene
Partially deuteriated 22 Colorless oil, IR (cmK1) 1735; H
NMR (600 MHz, CDCl3) d: 2.66 (m, 2H, H-3), 3.31 (s, 3H,

5796
D. Plazuk et al. / Tetrahedron 61 (2005) 5788–5796
Tetrahedron Lett. 1993, 34, 7549. (c) El-Saidi, M.; Kassam,
K.; Pole, D. L.; Tadey, T.; Warkentin, J. J. Am. Chem. Soc.
1992, 114, 8751. (d) Moss, R. A.; Wlostowski, M.; Shen, S.;
Krogh-Jespersen, K.; Matro, A. J. Am. Chem. Soc. 1988, 110,
4443. (e) Scheeren, J. W.; Staps, R. J. F. M.; Nivard, R. J. F.
Recl. Trav. Chim. Pays-Bas 1973, 92, 11. (f) Hoffmann, R. W.;
H-10), 3.77 (s, 3H, H-9), 5.19 (d, JZ10.2 Hz, 1H, H-8 cis),
5.33 (d, JZ16.8 Hz, 1H, H-8 trans), 5.67 (bs, 1H, H-4), 5.92
(dd, JZ17.4, 10.2 Hz, 1H, H-7); H-5 and H-6 signals were
not observed, indicating that those sites were deuteriated;
13C NMR (150.9 MHz, CDCl3) d: 38.9 (C-3), 50.0 (C-9),
52.5 (C-10), 63.6 (quint, JZ23.7 Hz, C-6), 102.3 (C-2),
117.4 (C-8), 118.7 (quint, JZ23.7 Hz, C-5), 130.7 (C-4),
133.8 (C-7), 169.3 (C-1); HRMS (EI) m/z: calcd for
C10H13D4O4 (MCH)C 205.1378 found 205.1371.
¨
Hauser, H. Tetrahedron Lett. 1964, 197.
14. (a) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G.
E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.;
Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar,
S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.;
Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada,
M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida,
M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.;
Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala,
P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg,
J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain,
M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghava-
chari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson,
B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.
Gaussian, Inc.: Pittsburgh, PA, 2003. (b) Lee, C.; Yang, W.;
Parr, R. G. Phys. Rev. B 1998, 37, 785. (c) Becke, A. D.
J. Chem. Phys. 1993, 98, 5648.
Acknowledgements
We thank Dr. Arkadiusz Klys for the correlation spectra and
NSERC for financial support.
References and notes
1. (a) Iwamura, H.; Tukada, H. Tetrahedron Lett. 1978, 3451. (b)
Iwamura, H.; Iwai, M.; Kihara, H. Chem. Lett. 1977, 881.
2. Baldwin, J. E.; Walker, J. A. J. Chem. Soc., Chem. Commun.
1972, 354.
3. Hoffmann, R. W.; Hirsch, R.; Fleming, R.; Reetz, M. T. Chem.
Ber. 1972, 105, 3532.
4. (a) Venneri, P. C.; Warkentin, J. J. Am. Chem. Soc. 1998, 120,
11182. Addendum: J. Am. Chem. Soc. 1999, 121, 490. (b)
Merkley, N.; Venneri, P. C.; Warkentin, J. Can. J. Chem. 2001,
79, 312.
15. An analogous procedure for the generation of dimethoxy
carbene at 40–50 8C has been reported. Zhou, H.; Warkentin, J.
Org. Lett. 2005, 7, 487.
5. Kassam, K.; Pole, D. L.; El-Saidi, M.; Warkentin, J. J. Am.
Chem. Soc. 1994, 116, 1161.
6. Danheiser, R. L.; Bronson, J. J.; Okano, K. J. Am. Chem. Soc.
1985, 107, 4579.
16. (a) Czardybon, W.; Klys, A.; Warkentin, J.; Werstiuk, N. H.
Can. J. Chem. 2003, 81, 1438. (b) Lu, X.; Reid, D. L.;
Warkentin, J. Can. J. Chem. 2001, 79, 319. (c) Er, H.-T.; Pole,
D. L.; Warkentin, J. Can. J. Chem. 1996, 74, 1480. (d) Kassam,
K.; Pole, D. L.; El-Saidi, M.; Warkentin J. Am. Chem. Soc.
1994, 116, 1161.
7. Majchrzak, M. W.; Warkentin, J. Can. J. Chem. 1989, 67,
1753.
´
8. Hitchcock, A. P.; Zweep, S.; Steel, T.; Bekhazi, M.;
Warkentin, J. Can. J. Chem. 1982, 60, 2914.
9. Merkley, N.; El-Saidi, M.; Warkentin, J. Can. J. Chem. 2000,
78, 356.
17. Faller, J. W.; Blankenship, C.; Whitmore, B.; Sena, S. Inorg.
Chem. 1985, 24, 4483.
10. Marque, S.; Le Mercier, P. C.; Tordo, P.; Fischer, H.
Macromolecules 2000, 33, 4403.
18. Baird, W. C., Jr.; Hartgerink, R. L.; Surridge, J. H. J. Org.
Chem. 1985, 50, 4601.
11. Venneri, P. C.; Warkentin, J. Can. J. Chem. 2000, 78, 1194.
12. Ziegler, F. E. Chem. Rev. 1988, 88, 1423.
19. (a) Shaikh, A. A. G.; Sivaram, S. Ind. Eng. Chem. Res. 1992,
31, 1167. (b) Woo, E. P.; Cheng, F. C. W. J. Org. Chem. 1986,
51, 3704.
13. (a) Couture, P.; El-Saidi, M.; Warkentin, J. Can. J. Chem.
1997, 75, 326. (b) Ge, C.-S.; Jefferson, E. A.; Moss, R. A.