X=Y-ZH systems as potential 1,3-dipoles. Part 56: Cascade 1,3-azaprotio cyclotransfer-cycloaddition reactions between aldoximes and divinyl ketone: The effect of oxime E/Z isomerism on cycloaddition stereoselectivity

Article abstract of DOI:10.1016/S0040-4020(02)00835-9

The cascade 1,3-azaprotiocyclotransfer (1,3-APT)-1,3-dipolar-cycloaddition (1,3-DC) reaction between aldoximes and divinyl ketone affords mixtures of exo and endo-isomers of substituted 1-aza-7-oxabicyclo[3.2.1]octan-4-ones, the ratio of which is dependent on the E/Z geometry of the starting oxime and its ability to isomerise under the thermal reaction conditions.

Full text of DOI:10.1016/S0040-4020(02)00835-9

Tetrahedron 58 (2002) 7715–7725
X5Y–ZH systems as potential 1,3-dipoles. Part 56:
Cascade 1,3-azaprotio cyclotransfer–cycloaddition reactions
between aldoximes and divinyl ketone: the effect of oxime E/Z
isomerism on cycloaddition stereoselectivityq
b
Mark Blackwell,a Peter J. Dunn,b Alison B. Graham,a Ronald Grigg,a Paul Higginson,
Imaad S. Sabaa and Mark Thornton-Petta
,
*
aMolecular Innovation, Diversity and Automated Synthesis (MIDAS) Centre, School of Chemistry, The University of Leeds,
Woodhouse Lane, Leeds LS2 9JT, UK
bPfizer Global Research and Development (UK), Sandwich, Kent CT13 9NJ, UK
Received 22 April 2002; revised 18 June 2002; accepted 11 July 2002
Abstract—The cascade 1,3-azaprotiocyclotransfer (1,3-APT)-1,3-dipolar-cycloaddition (1,3-DC) reaction between aldoximes and divinyl
ketone affords mixtures of exo and endo-isomers of substituted 1-aza-7-oxabicyclo[3.2.1]octan-4-ones, the ratio of which is dependent on the
E/Z geometry of the starting oxime and its ability to isomerise under the thermal reaction conditions. q 2002 Elsevier Science Ltd. All rights
reserved.
In the preceeding paper we discussed the general features of
cascade 1,3-APT-1,3-DC processes involving ketoximes
and divinyl ketone.1 In this paper we report full details of
related studies with aldoximes and divinyl ketone and the
role of oxime E/Z-stereochemistry in these processes.
the rate of E-oxime 1,3-APT is sluggish (k2qk4). We expect
that k3, k5qk6, k26 since the intramolecular cycloaddition is
expected to be rapid and the barrier to nitrone isomerisation
is high (Ea¼25–36 kcal mol21). Thus we have assumed that
k6 and k26 are essentially zero. Isomerisation of endo-4a to
exo-4a, which would occur via a retro-cycloaddition–E/
Z-nitrone isomerisation–cycloaddition sequence, was found
to be slow† showing that the exo/endo-ratio, under our
normal conditions (818C, 48 h) is the outcome of an
essentially kinetically controlled process (k23, k25 very
small). We propose that the outcome of the cascade is
controlled by the relative rates of E/Z oxime isomerism
(k1 and k21) and 1,3-APT (k2 and k4) if 1,3-APT is
essentially irreversible at 818C5 (k22, k24<0).
The reaction between an aldoxime 1 and divinyl ketone 2
represents a Class 2a process where both the azaprotiophile
(nitrone generating functionality) and dipolarophile are
located in the same molecule (Scheme 1).2 In the case of
ketoximes, such processes lead to the formation of 1-aza-7-
oxabicyclo[3.2.1]octan-4-ones 3 and 1-aza-8-oxabicyclo-
[3.2.1]octan-4-ones 4. Control of the cascade outcome is
achieved by a judicious choice of experimental conditions.3
In non-symmetrical ketoxime cases diastereomeric products
are also possible at C(8) in 3 and C(7) in 4.
It is proposed that path a and path b (Scheme 1) have
different rate determining steps. In path a, oxime isomeri-
sation may become the rate determining step when the
aldoxime R substituent is large. In such cases the steric
compression between R and the hydroxyl moiety in the
Z-oxime retards the E/Z-isomerism. In contrast, 1,3-APT
may be rate determining in path b because the R group
impedes the reactivity of the lone pair. The ratio of exo and
endo-products is determined by the relative rates of these
steps which are, in turn, determined by the nature of R.
In contrast, under thermal conditions (818C, MeCN)
aldoximes (R or R0¼H) afford mixtures of exo-4 and
endo-4 and 3 is rarely detected.4 The mechanism of the
cascade is outlined in Scheme 1. Thermal E/Z intercon-
version of oximes is expected to be rapid. Due to steric
blockade of the aldoxime lone pair by the R moiety in E-1,
q For part 55, see Ref. 1.
†
Keywords: aliphatic oximes; isomerism; cycloaddition.
*
A 3:1 mixture of endo-4a and exo-4a was heated at 818C in CD3CN to
afford, after 84 h, a 1:1 mixture and after 136 h a 1:2 mixture of endo-
4a/exo-4a.
Corresponding author. Tel./fax: þ44-113-2336501;
e-mail: r.grigg@chem.leeds.ac.uk
0040–4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(02)00835-9

7716
M. Blackwell et al. / Tetrahedron 58 (2002) 7715–7725
Scheme 1.
Table 1. Reactions between aldoximes and 2 carried out in a STEM block
R0
Oxime
R
Temperature (8C)a
Conversion (%)b
Ratiob exo-4/endo-4
Yield (%)c
1a
1b
1c
1d
1e
1a
1b
1c
1d
1e
1a
1a
1e
Me
Et
i-Pr
t-Bu
Bn
Me
Et
i-Pr
t-Bu
Bn
H
H
H
H
H
H
H
H
H
H
H
H
H
80
80
80
80
80
95
95
95
95
95
120d
150d
120d
.95
.95
90
3.4:1
2.0:1
1.5:1
.20:1
3.0:1
5.6:1
2.7:1
2.0:1
.20:1
4.3:1
8.0:1
5.0:1
8.0:1
59
41
34
11
65
59
53
54
37
57
65
70
81
25
.95
.95
.95
90
45
.95
.95
.95
.95
Me
Me
Bn
Reaction conditions: 1 (1 mol equiv.), 2 (1.2 mol equiv.), acetonitrile (15 mL mmol21 oxime), 48 h.
a
Temperature of STEM block.
1
Measured from the H NMR spectra of the crude products.
Combined isolated yield.
Reaction carried out in a Schlenk tube, estimated reaction temperature, temperature of oil bath is 130 or 1608C.
b
c
d
in a slow process (25% conversion at 808C, 45% conversion
at 958C). The stereochemistry of exo-4d was determined by
X-ray crystallography of its oxime 5 (Fig. 1) whilst the
A survey of aldoximes 1a–e showed that the diastereo-
selectivity of the reaction is dependent on the size of the
aldoxime R group (Table 1). Predominant formation of
exo-4 (e.g. for aldoximes 1a, 1e, Table 1) indicates that
oxime E/Z-isomerisation is faster than E-oxime 1,3-APT.
These reactions were carried out in a STEM block utilising
the SK233 Automated Workstation. Increasing the size
of the R group (aldoximes 1a–d) resulted in a decrease in
exo-selectivity. Oxime 1d affords exo-4d as the sole product
Figure 1. X-Ray crystal structure of 5 (the oxime of exo-4d).

M. Blackwell et al. / Tetrahedron 58 (2002) 7715–7725
Table 2. Reactions between aldoximes 1a–e and 2
7717
R0
Oxime
R
Conversion (%)a Ratiob exo-4/endo-4 Yield (%)a
1a
1b
1c
1d
1d
1e
Me
Et
i-Pr
t-Bu
t-Bu
Bn
H
H
H
H
H
H
.95
.95
90
3.4:1
2.0:1
1.5:1
.20:1
.20:1
3.0:1
59
50
38
11
59
60
25
.95c
.95
Reaction conditions: 1 (1 mol equiv.), 2 (1.2 mol equiv.), acetonitrile
(15 mL mmol21 oxime), 48 h.
a
Combined isolated yield.
b
c
1
Measured from the H NMR spectra of the crude products.
2.8 mol equiv. of 2 used, reaction time 96 h—see Section 3.
Figure 2. X-Ray crystal structure of exo-4e.
conversion of 1d by portionwise addition of 2 (2.8 mol
equiv.) over 96 h, affording exo-4d in 59% yield (Table 2).
stereochemistry of the remaining isomers was established
by nOe studies. In exo-4 irradiation of H7 results in
enhancement at H60, whereas in endo-4 irradiation of H7
results in enhancement at H6. The signals corresponding to
H6 and H60 are easily identified by their coupling constants
Increasing the temperature of the STEM block from 80 to
958C resulted in an increase in the propensity of aliphatic
aldoximes to form exo-4 (Table 1). The results obtained
using conventional laboratory reactions are reported in
Table 2. The STEM block in conjunction with the
ReactArray condenser system mimics a closed system to
some extent since the degree of exo-selectivity is signifi-
cantly higher at 958C. We were able to utilise this effect in
the case of phenyl acetaldehyde oxime (1e), attaining an
exo-selectivity of 8:1 (81%) when the reaction temperature
was 1208C (closed system) (Table 1). Similarly, 1a afforded
an 8:1 ratio of exo-4b to endo-4b (65%). However
increasing the temperature of the system further (1508C)
resulted in reduced levels of exo-selectivity (5:1). A 3:1
mixture of exo-4a and endo-4a was recovered (89%)
unchanged, after heating in a closed system at 1208C for
1 h indicating that a cycloreversion–cycloaddition pathway
does not account for the increased exo-selectivity. Increased
temperatures will facilitate higher rates of E/Z-oxime
equilibration, E-1,3-APT (k4, Scheme 1) and Z-1,3-APT
(k2, Scheme 1) but not all to the same degree. It is proposed
that the effect is most pronounced on the E/Z oxime
isomerisation step (k1/k21, Scheme 1), such that Z-1,3-APT
(k2, Scheme 1) becomes the rate determining step under
‘high temperature’ conditions.
0
to H5; JH6 –H5 ca. 8 Hz and JH6–H5¼0 Hz (dihedral angle
H6–H5¼908). The structure of exo-4e was confirmed by
X-ray crystallography (Fig. 2). Overlapping signals in the
1H NMR spectra of exo-4h and exo-4i hampered nOe
studies; the stereochemistry was assigned by comparison of
0
the values of JH6 –H7 (9.0 Hz in exo-4h and 9.1 Hz in exo-4i)
0
with that obtained for exo-4k (JH6 –H7¼8.9 Hz). The latter
structure was established by X-ray crystallography (Fig. 3)
An increase in the size of the aldoxime R group is expected
to decrease both the rate of E-oxime isomerisation (Scheme
1, k1) and E-oxime 1,3-APT (k4) but we propose that the
steric effect is more pronounced on the oxime isomerisation
step due to the developing steric compression between the
R group and the hydroxyl moiety in Z-1. The E-oxime
1,3-APT step can occur provided the oxime R substituent
contains an a-hydrogen (as in 1a–c and 1e). This allows
conformations where a C–H bond, as opposed to a C–R
bond, impedes the reactivity of the lone pair. As the size
of R increases, the increasing suppression of the E-oxime
1,3-APT step is reflected in lower yields and conversions
(Table 1). When the oxime R substituent contains no
a-hydrogens (1d) severe impediment of the N-lone pair
reactivity is unavoidable and 1,3-APT proceeds via the
Z-oxime (path a), affording exo-4d only (Scheme 1). The
slow rate of oxime isomerisation for 1d is reflected in
the low conversion (25% after 48 h in STEM block at 808C)
and yield (11%). We were able to obtain complete
Unlike aliphatic oximes, E-benzaldoxime 1f does not
isomerise at 818C. However, both E and Z-isomers can be
prepared and stored separately.6 The observation that E-1f
forms endo-4f exclusively while Z-1f affords exo-4f
exclusively (Table 3) supports the mechanism outlined in
Scheme 1. The application of microwave activation to the
reaction of Z-1f has been briefly studied. As expected
microwave heating to 1508C decreases the reaction from
48 h to 15 min. whilst retaining the excellent diastereo-
selectivity observed at the lower temperature (Table 4).
However, the regioselectivity suffers somewhat in that a
significant amount (14%) of exo-3f is also produced
showing that the Z-nitrone conversion to 3 becomes
activated at these higher temperatures.
We also carried out the process with both E- and Z-para-
substituted benzaldoximes (Table 3). Table 3 shows that
Z-benzaldoxime and Z-benzaldoximes with a para-electron
withdrawing group (CF3, CN, NO2) (Table 3, entries 3, 5, 7
Figure 3. X-Ray crystal structure of exo-4k.

7718
M. Blackwell et al. / Tetrahedron 58 (2002) 7715–7725
Table 3. Reactions between aromatic aldoximes and 2
Entry
Isomer
pKa of OH9
Conversion (%)a
Ratio exo-4/endo-4a
Yield (%)b
1
2
3
4
5
6
7
8
Hc
H
E-1fc
E-1f
Z-1f
E-1g
Z-1g
E-1h
Z-1h
E-1i
Z-1i
E-1j
Z-1j
E-1k
Z-1k
E-1l
Z-1l
11.37
11.37
11.37
10.81
10.81
10.68
10.68
10.57
10.57
11.49
11.49
11.05
11.05
11.18
11.18
11.72
10
55
1:.20
1:.20
.20:1
–
.20:1
–
–
18
79
–
76
–
H
.95
,5
.95
,5
90
,5
55
80
.95
.95
.95
.95
.95
80
CF3
CF3
CN
CN
NO2
NO2
MeO
MeO
Cl
.20:1
–
44
–
9
.20:1
4:1d
7:1
10:1e
8:1
8:1
12:1
1:2
19
55
78
61
76
72
70
50
10
11
12
13
14
15
16
Cl
F
F
2, 4-diMeO
E-1m
Reaction conditions: 1 (1 mol equiv.), 2 (1.2 mol equiv.), acetonitrile (15 mL mmol21 oxime), reflux, 48 h.
a
b
c
d
e
1
Measured from the H NMR spectra of the crude products.
Combined isolated yield.
Reaction carried out in a STEM block at 808C.
30% of exo-3j was obtained.
10% of exo-3k was obtained.
and 9) give endo-4 products stereoselectively with little, if
any, detectable exo-4 isomer. Thus in these cases E/Z-oxime
isomerisation is considerably slower than the 1,3-APT-1,3-
DC sequence. E/Z-Isomerisation becomes more competitive
with the 1,3-APT-1,3-DC sequence in Z-benzaldoximes
with a para-donor group (MeO) or -I, p-donor groups (Cl,
F) (Table 3, entries 11, 13 and 15). The strongest p-donor
(MeO) furnishes a 7:1 mixture of exo-4 and endo-4 (entry
11) whilst the most electronegative halogen (F) exhibits less
E/Z-isomerisation (entry 15) (12:1 mixture of exo-4 and
endo-4) than para-chloro benzaldoxime (entry 13, 8:1
mixture of exo-4 and endo-4). As expected Z-benzaldox-
imes with the strongest electron withdrawing groups (CN,
NO2) give lower conversions (entries 7 and 9) mirroring the
trend in pKa7 and emphasizing the nucleophilic component
in the 1,3-APT process.
nucleophilicity and the steric blockade of the nitrogen lone
pair by the syn-aryl group in the 1,3-APT step i.e. this key
initial step (Scheme 1, path b) is switched off. However, the
E-benzaldoximes bearing electron donor groups afford
mixtures of exo and endo-4 with the former predominating
(Table 3, entries 10, 12 and 14) except for the E-2,4-di-
methoxybenzaldoxime in which endo-4 is the predominant
isomer (Table 3, entry 16). In these cases the electron donor
substituents promote both 1,3-APT in the sterically hindered
E-oximes and E/Z-isomerisation in proportion to their donor
capability. It should be remembered that aryl groups are
stereochemically anisotropic and can project an edge or
face of the aryl ring towards the nitrogen lone pair in the
E-benzaldoximes. Note that we have previously reported
MNDO calculations which indicate that C–N bond
formation is in advance of C–H bond formation in the
1,3-APT transition state.8
In the reaction of E-oximes we observe that those bearing
electron withdrawing groups are essentially configuration-
ally stable under our reaction conditions with ,5%
conversion observed (Table 3, entries 4, 6 and 8). The low
conversion illustrates the combined effect of low nitrogen
In summary, we have shown that the ratio of exo/endo
isomers in tandem 1,3-azaprotio cyclotransfer–cyclo-
addition reactions between divinyl ketone and aldoximes
is dependent on the relative rates of Z/E oxime isomerisation
and 1,3-APT.
Table 4. Microwave induced reactions between (Z )-benzaldoxime and 2
Time
(min)
Conversion
a,b
exo-4f/endo-4f
exo-4f/exo-3fa
Yield
(%)
%
5
88
93
.95
.95
.95
16:1
6:1
6:1
5.5:1
5.5:1
6:1
–
–
10
15
30
60
.20:1
.20:1
.20:1
.20:1
66c
–
1. Experimental
–
Reaction conditions: 1f (2 mmol),
2
(2.4 mmol), acetonitrile
Melting points were determined on a Koffler hot-stage
apparatus and are uncorrected. Mass spectral data were
obtained on an Autospec instrument at 70 eV. Nuclear
magnetic resonance spectra were recorded on Brucker AM
(2.5 mL mmol21 oxime), reactions carried out in a sealed tube.
a
b
c
1
Measured from the H NMR spectra of the crude products.
In all cases trace amount (,5%) of benzaldehyde are formed.
Combined yield.

M. Blackwell et al. / Tetrahedron 58 (2002) 7715–7725
7719
250, Brucker DPX 300 and Brucker DRX 500 machines.
Chemical shifts are given in parts per million (d ) downfield
from tetramethylsilane (TMS) as internal standard. All
spectra were recorded in deuteriochloroform. The following
abbreviations are used: s¼singlet, d¼doublet, t¼triplet,
q¼quartet, m¼multiplet, dd¼double doublet, bs¼broad
singlet, Cq¼quaternary carbon. Microanalyses were
obtained using a Carbo Erba MOD11016 instrument.
Infra-red data were recorded on films, prepared by
evaporation of a dichloromethane solution on a Spectra-
Tech Inc. Trough Plate Ark on a Nicolet Magna-IR 560
spectrometer. HPLC analysis was performed on Gilson
instruments. X-Ray analysis was performed on a Stoe
STADI 4-circle machine or a Nonius Kappa CCD area-
detector diffractometer. The term ‘petroleum ether’ refers to
the fraction of petroleum ether with boiling point between
40 and 608C. Column chromatography was performed using
flash silica gel (Merck 9385).
NCH); m/z (FAB) 190 (Mþ1, 100); n (cm21) 3179 (OH)
and 1653 (CvN).
1.2.2. (Z)-4-Nitrobenzaldoxime (Z-1i). (E )-4-Nitrobenz-
aldoxime (5.00 g, 30 mmol) was treated with concentrated
hydrochloric acid (15 mL) according to the general pro-
cedure. Subsequent work up afforded the product (4.25 g,
85%) which crystallised from acetone–pentane as yellow
needles, mp 175–1778C. Found: C, 50.6; H, 3.95; N, 17.1;
C7H6N2O3 requires C, 50.6; H, 3.65; N, 16.85%; dH
(250 MHz; d6-DMSO) 12.22 (s, 1H, NOH), 8.18 (m, 4H,
4£ArH) and 7.59 (s, 1H, NCH); m/z (EI; %) 166 (Mþ, 48),
76 (67), 75 (73), 74 (44), 65 (95), 63 (48), 51 (52), 50 (100)
and 39 (79); n (cm21) 3168 (OH), 1643 (CvN), 1519 (NO)
and 1342 (NO).
1.2.3. (Z )-4-Methoxybenzaldoxime Z-1j. (E )-4-Methoxy-
benzaldoxime (2.11 g, 14 mmol) was treated with concen-
trated hydrochloric acid (7 mL) according to the general
procedure. The crude product was obtained as a colourless
solid (1.61 g, 75%) which crystallised from dichloro-
methane–petroleum ether as colourless needles (0.72 g,
34%), mp 128–1308C. Found C, 63.6; H, 6.05; N, 9.25;
C8H9NO2 requires, C, 63.6; H, 6.0; N, 9.5%; dH (250 MHz)
9.80 (bs,1H, NOH), 7.95 (m, 2H, 2£ArH), 7.33 (s, 1H,
NCH), 6.95 (m, 2H, 2£ArH) and 3.85 (s, 3H, Me); m/z
Assignment of the geometry of aldoxime isomers is based
on the chemical shift of the imine proton which is
deshielded when it is cis to the hydroxy moiety. In
the absence of any literature data, or comparison with
the opposite isomer, oximes were assumed to be the
(E)-isomers.
Oximes 1a, E-1f, E-1i, E-1l and E-1m were obtained from
commercial sources. Oximes 1b–e, E-1f–h, E-1j, E-1k,
Z-1f, Z-1h and Z-1j were prepared according to the general
procedure and gave identical spectroscopic data to that
reported in the literature.10 Divinyl ketone was prepared
according to the literature procedure.11
(FAB; %) 152 (Mþ1, 100) and 151 (Mþ, 53); n cm21
(film) 3175 (OH), and 1601 (CvN).
)
1.2.4. (Z)-4-Chlorobenzaldoxime (Z-1k). (E)-4-Chloro-
benzaldoxime (5.00 g, 32 mmol) was treated with concen-
trated hydrochloric acid (15 mL) according to the general
procedure. Subsequent work up afforded the product
(4.00 g, 80%) which crystallised from dichloromethane–
pentane as colourless needles, mp 132–1338C. Found:
C, 53.75; H, 4.15; N, 8.7; Cl, 22.5; C7H6NOCl requires
C, 54.05; H, 3.9; N, 9.0; Cl, 22.8%; dH (250 MHz;
d6-DMSO) 11.85 (s, 1H, NOH), 8.01 (d, 2H, J¼8.6 Hz,
2£ArH), 7.51 (d, 2H, J¼8.6 Hz, 2£ArH) and 7.46
(s, 1H, NCH); m/z (FAB; %): 158 37(Mþ1, 35), 156
35(Mþ1, 100) and 149 (59); n (cm21) 3159 (OH) and 1653
(CvN).
1.1. General procedure for the preparation of aldoximes
The appropriate ketone was added to a solution of
hydroxylamine hydrochloride and sodium acetate in 2:1
v/v acetonitrile–water and the resulting mixture stirred at
ambient temperature for 18 h. The mixture was then
extracted with chloroform and the combined organic
extracts were washed with water, dried (MgSO4), filtered
and the filtrate concentrated in vacuo. The residue was
crystallised or distilled as appropriate.
1.2.5. (Z)-4-Fluorobenzaldoxime (Z-1l). (E)-4-Fluoro-
benzaldoxime (5.00 g, 36 mmol) was treated with concen-
trated hydrochloric acid (17 mL) according to the general
procedure. The product (4.45 g, 89%) crystallised from
dichloromethane–pentane as colourless needles, mp 130–
1328C. Found: C, 60.45; H, 4.35; N, 10.05; C7H6NOF
requires C, 60.65; H, 4.35; N, 10.05; dH (250 MHz,
d6-DMSO) 11.71 (s, 1H, NOH), 8.07 (m, 2H, 2£ArH),
7.44 (s, 1H, NCH) and 7.28 (m, 2H, 2£ArH); m/z (EI; %)
139 (Mþ, 61), 121 (33), 96 (78), 95 (79), 94 (35), 83 (32), 75
(100), 57 (36) and 50 (36); n (cm21) 3190 (OH) and 1653
(CvN).
1.2. General procedure for Z-aryl aldoximes
The appropriate (E )-oxime was added to an excess of
concentrated hydrochloric acid with cooling. The mixture
was then cooled further to 58C and the resulting precipitate
filtered and slowly added to a stirred solution of saturated
aqueous sodium bicarbonate and the resulting slurry
warmed for a few minutes. The precipitate was collected
by filtration and crystallised.
1.2.1. (Z)-4-Trifluoromethylbenzaldoxime (Z-1g). (E)-4-
Trifluoromethylbenzaldoxime (3.00 g, 16 mmol) was treated
with concentrated hydrochloric acid (7 mL) according to the
general procedure. Subsequent work up afforded the product
(2.17 g, 72%) which crystallised from diethyl ether as
colourless needles, mp 115–1178C. Found C, 51.1; H, 3.3;
N, 7.4; C8H6F3NO requires C, 50.8; H, 3.2; N, 7.4%; dH
(250 MHz, d6-DMSO) 12.05 (s, 1H, NOH), 8.18 (d, 2H,
J¼8.2 Hz), 7.82 (d, 2H, J¼8.3 Hz, 2£ArH) and 7.60 (s, 1H,
1.3. General procedure for the 1,3-APT-1,3-DC cascades
A solution of divinyl ketone (108 mL, 1.2 mmol) and
oxime (1 mmol) in acetonitrile (15 mL) was heated at reflux
with stirring for 48 h. After cooling the mixture was
concentrated in vacuo and the residue purified by column
chromatography.

7720
M. Blackwell et al. / Tetrahedron 58 (2002) 7715–7725
1.3.1. 7-exo-Methyl-8-oxa-1-azabicyclo[3.2.1]octan-4-one
(exo-4a) and 7-endo-methyl-8-oxa-1-azabicyclo[3.2.1]octan-
4-one (endo-4a). Acetaldoxime (0.059 g, 1 mmol) and
divinyl ketone (108 mL, 1.2 mmol) were reacted according
to the general procedure for 48 h. Subsequent work up
afforded the crude product as a brown gum, comprising of
a 3.4:1 mixture of exo-4a and endo-4a. Flash chroma-
tography, eluting with 1:4 v/v ethyl acetate–diethyl ether
afforded the same isomeric mixture in a pure form (0.083 g,
59%). Partial separation was achieved by preparative
HPLC: Chiralpak AD 250£20 mm2; wavelength: 254 nm;
work up afforded the crude product as a brown gum
comprising of a 2:1 mixture of exo-4b and endo-4b. Flash
chromatography, eluting with diethyl ether afforded exo-4b
(0.050 g, 32%) and a 5:1 mixture of endo-4b and exo-4b
(0.027 g, 17%).
exo-4b. The product was obtained as a pale yellow oil.
Found C, 62.2; H, 8.65; N, 8.75; C8H13NO2 requires C,
61.9; H, 8.4; N, 9.0%. dH (500 MHz) 4.39 (d, 1H, J¼8.0 Hz,
H5), 3.73 (ddd, 1H, J¼14.1, 10.7, 5.9 Hz, H20), 3.24 (dd,
1H, J¼14.1, 8.4 Hz, H2), 3.19 (m, 1H, H7), 2.53 (ddd, 1H,
J¼16.8, 10.7, 8.4 Hz, H3), 2.39 (dd, 1H, J¼13.1, 8.2 Hz,
H6), 2.31 (dd, 1H, J¼016.8, 5.9 Hz, H30), 2.14 (ddd, 1H, J¼
13.1, 8.0, 4.7 Hz, H6 ), 1.71–1.45 (dm, 2H, CH2CH3) and
1.00 (t, 3H, J¼7.3 Hz, CH2CH3); m/z (EI; %) 155 (Mþ, 92),
84 (76), 70 (100), 56 (82), 42 (78), 41 (74) and 39 (32); n
(cm21) 1730 (CvO).
mobile phase: 4:1 hexane–ethanol; flow rate: 1 mL min21
;
exo-4a tR 14.48 and 15.56 min (enantiomers); endo-4a tR
19.55 and 21.89 min (enantiomers).
exo-4a. Data were obtained from an enriched 10:1 mixture
of exo-4a and endo-4a, obtained as a pale yellow oil. Found:
C, 59.55; H, 7.9; N, 9.7; C7H11NO2 requires: C, 59.55; H,
7.85; N, 9.9%; dH (500 MHz) 4.42 (d, 1H, J¼8.1 Hz, H5),
3.71 (ddd, 1H, J¼14.3, 10.7, 6.2 Hz, H20), 3.52 (m, 1H, H7),
3.28 (ddt, 1H, J¼14.3, 8.4, 1.1 Hz, H2), 2.52 (dddd, 1H,
J¼16.7, 10.7, 8.4, 0.5 Hz, H3), 2.44 (ddd, 1H, J¼13.2, 6.4,
1.2 Hz, H6), 2.31 (ddt, 1H, J¼16.7, 6.2,0 1.2 Hz, H30), 2.12
(dddd, 1H, J¼13.2, 8.1, 4.3, 1.0 Hz, H6 ) and 1.32 (d, 3H,
J¼6.6 Hz, CH3); m/z (EI; %) 141 (Mþ, 70), 98 (23), 70 (29),
56 (100), 55 (56), 43 (41), 42 (71) and 41 (37); n (cm21
1725 (CvO).
)
endo-4b. Data were obtained from an enriched 5:1 mixture
of endo-4b and exo-4b, obtained as a pale yellow oil. HRMS
found 155.0936; C8H13NO2 requires 155.0946; dH
(500 MHz) 4.37 (m, 1H, H5), 3.69 (ddd, 1H, J¼14.6,
10.6, 6.4 Hz, H20), 3.53 (m, 1H, H7), 3.44 (ddt, 1H, J¼14.6,
8.60, 1.0 Hz, H2), 2.69 (dtd, 1H, J¼12.9, 9.6, 1.1 Hz,
H6 ), 2.41 (m, 1H, H3), 2.28 (distorted dd, 1H, J¼16.9,
6.4 Hz, H30), 1.82 (m, 1H, CH HCH3), 1.67 (ddd, 1H,
J¼12.9, 7.1, 2.9 Hz, H6), 1.60 (m, 1H, CHHCH3) and 1.10
(t, 3H, J¼7.4 Hz, CH3); m/z (EI; %): 155 (Mþ, 81), 84 (61),
73 (93), 70 (98), 56 (100), 55 (90), 43 (42), 42 (95) and 41
(78).
1.3.3. 7-exo-Isopropyl-8-oxa-1-azabicyclo[3.2.1]octan-4-
one (exo-4c) and 7-endo-isopropyl-8-oxa-1-azabicyclo-
[3.2.1]octan-4-one (endo-4c). Isobutyraldoxime (0.087 g,
1 mmol) and divinyl ketone (108 mL, 1.2 mmol) were
reacted according to the general procedure for 48 h.
Subsequent work-up afforded the crude product as a
brown gum, comprising of a 1.5:1 mixture of exo-4c and
endo-4c and unreacted oxime (10%). Flash chroma-
tography, eluting with diethyl ether, afforded exo-4c
(0.034 g, 20%) and endo-4c (0.030 g, 18%).
endo-4a. Data were obtained as an enriched 3.3:1 mixture of
endo-4a and exo-4a, obtained as a pale yellow oil. Found: C,
59.55; H, 7.9; N, 9.7; C7H11NO2 requires: 59.55; H, 7.85; N,
9.9%; dH (500 MHz) 4.37 (dd, 1H, J¼8.7, 1.6 Hz, H5), 3.77
(dtd, 1H, J¼9.6, 7.0, 6.9 Hz, H7), 3.70 (ddd, 1H, J¼14.7,
10.7, 6.4 Hz, H20), 3.49 (ddt, 1H, J¼14.7, 8.5, 1.0 Hz, H2),
2.75 (dtd, 1H, J¼12.9, 9.6, 1.1 Hz, H60), 2.46 (m, 1H, H3),
2.28 (dd, 1H, J¼16.9, 6.4 Hz, H30), 1.69 (ddd, 1H, J¼12.9,
6.9, 1.6 Hz, H6) and 1.40 (d, 3H, J¼7.0 Hz, CH3); m/z (EI;
%) 141 (Mþ, 66), 83 (32), 70 (37), 56 (100), 55 (78), 43
(54), 42 (93) and 41 (44); n (cm21) 1718 (CvO).
exo-4c. Obtained as a pale yellow oil. HRMS: Found
169.1106; C9H15NO2 requires 169.1103; dH (500 MH0z)
4.38 (m, 1H, H5), 3.71 (ddd, 1H, J¼14.0, 10.7, 5.8 Hz, H2 ),
3.20 (ddt, 1H, J¼14.0, 8.4, 1.1 Hz, H2), 2.99 (m, 1H, H7),
2.53 (ddd, 1H, J¼16.9,0 10.7, 8.4 Hz, H3), 2.320 (ddt, 1H,
J¼16.9, 5.8, 1.1 Hz, H3 ), 2.30–2.27 (m, 2H, H6 and H6),
1.72 (m, 1H, CH(CH3)2), 1.02 (d, 3H, J¼6.6 Hz, CH3) and
0.93 (d, 3H, J¼6.7 Hz, CH3); m/z (EI; %) 169 (Mþ, 54), 84
1.3.2. 7-exo-Ethyl-8-oxa-1-azabicyclo[3.2.1]octan-4-one
(exo-4b) and 7-endo-ethyl-8-oxa-1-azabicyclo[3.2.1]-
octan-4-one (endo-4b). Propionaldoxime (0.073 g,
1 mmol) and divinyl ketone (108 mL, 1 mmol) were reacted
according to the general procedure for 48 h. Subsequent

M. Blackwell et al. / Tetrahedron 58 (2002) 7715–7725
7721
(100), 70 (55), 56 (77), 55 (93), 43 (62), 42 (99) and 40 (56);
n (cm21) 1730.
1.3.5. (4-E)-7-tert-Butyl-8-oxa-1-azabicyclo[3.2.1]octane-
4-one oxime (5). 7-exo-tert-Butyl-8-oxa-1-azabicyclo-
[3.2.1]octan-4-one (0.095 g, 0.52 mmol) (4d) and hydroxyl-
amine hydrochloride (0.040 g, 0.57 mmol) were reacted
according to the general procedure for 18 h. Subsequent
work up afforded the crude product as an amorphous yellow
solid (0.082 g, 80%) that crystallised from petroleum ether
as pale yellow plates, mp 131–1338C. Found: C, 60.85; H,
9.15; N, 14.25; C10H18N2O2 requires: C, 60.6; H, 9.15; N,
14.15%; dH (500 MHz) 8.39 (s, 1H, NOH), 4.66 (d, 1H,
J¼7.3 Hz, H5), 3.52 (m, 1H, H20), 2.95 (m, 3H, H7, H2 and
H3), 2.28 (m, 2H, H30 and H60), 2.13 (dd, 1H, J¼12.9,
8.6 Hz, H6) and 0.91 (s, 9H, 3£CH3); m/z (EI; %) 198 (Mþ,
100), 141 (73), 125 (31), 114 (36), 110 (31), 98 (75), 96
(35), 70 (54), 69 (52), 57 (67), 55 (43), 54 (47), 43 (32), 42
(96) and 39 (34); n (cm21) 3238 (OH) and 1653 (CvN).
1.3.6. 7-exo-Benzyl-8-oxa-1-azabicyclo[3.2.1]octan-4-one
(exo-4e). (Z )-Phenylacetaldoxime (0.135 g, 1 mmol) and
divinyl ketone (108 mL, 1.2 mmol) were reacted according
to the general procedure for 48 h. Subsequent work-up
afforded the crude product as a yellow solid, comprising of a
3:1 mixture of exo-4e and endo-4e. Flash chromatography,
eluting with 4:1 v/v diethyl ether–pentane, afforded the
same isomeric mixture in a pure form (0.130 g, 60%).
Crystallisation from cyclohexane afforded the pure 7-exo
isomer as colourless needles, mp 119–1218C. Found C,
72.05; H, 6.75; N, 6.4; C13H15NO2 requires C, 71.85; H,
6.95; N, 6.45%; dH (300 MHz) 7.34–7.21 (m, 5H, 5£ArH),
4.43 (d, 1H, J¼6.9 Hz, H5), 3.71 (ddd, 1H, J¼14.1, 10.2,
5.7 Hz, H20), 3.63–3.54 (m, 1H, H7), 3.19 (dd, 1H, J¼14.1,
8.4 Hz, H2), 3.09 (dd, 1H, J¼13.6, 7.6 Hz, PhCH H), 2.69
(dd, 1H, J¼13.6, 6.9 Hz, PhCHH ), 2.54–2.42 (m, 1H, H3)
and 2.36–2.20 (m, 3H, H30, H60 and H6); m/z (%) 218
(Mþ1, 7), 217 (Mþ, 46), 146 (9), 126 (53), 91 (100), 84
(73), 70 (54) and 56 (53); n (cm21) (DCM solution) 1725.
endo-4c. Obtained as a pale yellow oil. HRMS: Found
169.1106; C9H15NO2 requires 169.1103; dH (500 MHz):
4.37 (d, 1H, J¼8.8 Hz, H5), 3.70 (ddd, 1H, J¼14.6, 10.6,
6.4 Hz, H20), 3.49 (dd, 1H, J¼14.6, 8.5 Hz, H2), 3.26 (ddd,
1H, J¼11.2, 9.3, 7.7 Hz, H7), 2.65 (m, 1H, H60), 2.46 (m,
1H, H3), 2.28 (dd, 1H, J¼16.8, 6.4 Hz, H30), 1.74–1.70 (m,
2H, H6 and CH(CH3)2), 1.14 (d, 3H, J¼6.3 Hz, CH3) and
1.07 (d, 3H, J¼6.6 Hz, CH3); m/z (EI; %) 169 (Mþ, 34), 878
(50), 83 (41), 56 (37), 55 (100), 43 (42), 42 (45) and 41 (42);
n (cm21) 1730 (CvO).
1.3.7. 7-exo-Phenyl-8-oxa-1-azabicyclo[3.2.1]octan-4-
one (exo-4f). (a) (Z)-Benzaldoxime (0.242 g, 2 mmol) and
divinyl ketone (215 mL, 2.4 mmol) were reacted according
to the general procedure for 48 h. Subsequent work-up
afforded the crude product as a pale brown solid. Flash
chromatography, eluting with 9:1 v/v diethyl ether–pentane
afforded the product (0.320 g, 79%) which crystallised from
toluene as colourless plates, mp 110–1128C. Found C, 70.7;
H, 6.25; N, 7.15; C12H13NO2 requires C, 70.9; H, 6.45; N,
6.9; d (400 MHz) 7.40–7.24 (m, 5H, 5£PhH), 4.57 (d, 1H,
J¼8.0 Hz, H5), 4.46 (dd, 1H, J¼8.9, 5.1 Hz, H7), 3.84 (ddd,
1H, J¼14.2, 10.7, 6.0 Hz, H20), 3.42 (dd, 1H, J¼14.2,
1.3.4. 7-exo-tert-Butyl-8-oxa-1-azabicyclo[3.2.1]octan-4-
one (exo-4d). Divinyl ketone (125 mL, 1.4 mmol) was
added to a stirred solution of 1d (0.101 g, 1 mmol) in dry
acetonitrile (15 mL) and the resulting mixture was heated
and stirred at reflux. After 24 and 48 h further additions of
divinyl ketone (63 mL, 0.7 mmol) were made. After a total
reaction time of 96 h the reaction mixture was allowed to
cool. Subsequent work up afforded the crude product as a
brown solid. Flash chromatography, eluting with diethyl
ether, afforded the product as a colourless amorphous solid
(0.108 g, 59%) that crystallised from pentane as colourless
needles, mp 55–608C. Found C, 65.55; H, 9.35; N, 7.35;
C10H17NO2 requires C, 65.45; H, 9.20; N, 7.60%; dH
(500 MHz) 4.38 (d, 1H, J¼7.9 Hz, H5), 3.73 (ddd, 1H, J¼
14.0, 10.5, 5.9 Hz, H20), 3.18 (dd, 1H, J¼14.0, 8.3 Hz, H2),
2.99 (dd, 1H, J¼8.6, 6.1 Hz, H7),02.54 (ddd, 1H, J¼16.8,
10.5, 5.9 Hz, H3), 2.35 (m, 2H, H6 and H30), 2.17 (dd, 1H,
J¼13.5, 8.6 Hz, H6) and 0.94 (s, 9H, 3£Me); m/z (EI; %)
183 (Mþ, 45), 126 (43), 84 (100), 70 (83), 57 (67), 56 (51),
42 (77), 43 (53) and 41 (90); n (cm21) 1728 (CvO).

7722
M. Blackwell et al. / Tetrahedron 58 (2002) 7715–7725
8.4 Hz, H2), 2.84 (dd, 1H, J¼13.3, 8.9 Hz, H6), 2.71–2.62
(m, 1H, H3), 2.57 (ddd, 1H, J¼103.3, 8.0, 5.1 Hz, H60) and
2.42 (dd, 1H, J¼16.9, 6.0 Hz, H3 ); m/z (%) 204 (Mþ1, 9),
203 (Mþ, 62), 186 (14), 175 (41), 132 (20), 118 (47), 105
(100), 91 (67), 77 (50) and 55 (40); n (cm21) (DCM
solution) 1730.
H7), 3.86 (ddd, 1H, J¼14.4, 10.7, 6.1 Hz, H20), 3.42 (dd,
1H, J¼14.4, 8.5 Hz, H2), 2.87 (ddd, 1H, J¼13.5, 9.0,
1.1 Hz, H6), 2.66 (m, 1H, H3), 2.52 (dddd, 1H, J¼13.5, 8.0,
4.9, 1.0 Hz, H60) and 2.44 (dd, 1H, J¼17.0, 6.1 Hz, H30);
m/z (FAB; %) 272 (Mþ1, 100) and 136 (35); n (cm21) 1733
(CvO).
(b) Microwave procedure: divinyl ketone 2 (215 mL,
2 mmol) was added to a solution of (Z)-benzaldoxime
(0.254 g, 2 mmol) in acetonitrile (5 mL). The reaction
mixture was stirred and heated at 1508C in a CEM Discover
instrument for the appropriate time before being allowed to
cool to room temperature. The reaction mixture was then
evaporated to dryness to afford a pale yellow oil consisting
of a 6:1 mixture of exo-4 and exo-3, which crystallised from
n-heptane as colourless plates which comprised a 24:1
mixture of exo-4/exo-3.
1.3.8. 7-endo-Phenyl-8-oxa-1-azabicyclo[3.2.1]octan-4-
one (endo-4f). (E )-Benzaldoxime (0.121 g, 1 mmol) and
divinyl ketone (108 mL, 1.2 mmol) were reacted according
to the general procedure for 48 h. Subsequent work-up
afforded the crude product as a brown oil, comprising of
endo-157a (55%) and unreacted (E )-oxime (45%). Flash
chromatography, eluting with 9:1 v/v diethyl ether–pentane
afforded the product as a colourless solid (0.037 g, 18%).
Found C, 71.1; H, 6.75; N, 6.65; C12H13NO2 requires C,
70.9; H, 6.45; N, 6.9; d (400 MHz) 7.50–7.36 (m, 5H,
5£ArH), 5.01–4.97 (m, 1H, H7), 4.61 (d,0 1H, J¼8.5 Hz,
H5), 3.56 (ddd, 1H, J¼14.4, 9.6, 7.0 Hz, H2 ),03.01 (dd, 1H,
J¼14.4, 8.3 Hz, H2), 2.93–2.85 (m, 1H, H6 ), 2.52 (ddd,
1H, J¼13.1, 7.5, 1.7 Hz, H6), 2.35–2.26 (m, 1H, H3) and
2.20 (dd, 1H, J¼16.9, 7.0 Hz, H30); m/z (%) 204 (Mþ1, 9),
203 (Mþ, 60), 186 (27), 175 (43), 132 (30), 104 (100), 91
(79), 77 (62) and 55 (69); n (cm21) (DCM solution) 1730.
1.3.10. 4-(4-Oxo-8-oxa-1-azabicyclo[3.2.1]oct-7yl)benzo-
nitrile (exo-4h). (Z)-4-Cyanobenzaldoxime (0.292 g,
2 mmol) and divinyl ketone (216 mL, 2.4 mmol) were
reacted according to the general procedure for 48 h.
Subsequent work up afforded the crude product as a viscous
orange oil comprising of exo-4h and unreacted (E )-oxime
(10%). Flash chromatography, eluting with 4:1 v/v ethyl
acetate–diethyl ether, afforded the product (0.201 g,
44%) which crystallised from dichloromethane–pentane
as colourless needles, mp 1488C. Found C, 68.2; H, 5.35; N,
12.3; C13H12N2O2 requires: C, 68.4; H, 5.3; N, 12.3%; dH
(300 MHz) 7.65 (d, 2H, J¼8.4 Hz, 2£ArH) 7.53 (d, 2H,
J¼8.4 Hz, 2£ArH), 4.57 (d, 1H, J¼8.7 Hz, H5), 4.53 (dd,
1H, J¼9.0, 4.9 Hz, H7), 3.87 (ddd, 1H, J¼14.3, 10.7,
6.0 Hz, H20), 3.43 (ddt, 1H, J¼14.3, 8.3, 1.1 Hz, H2), 2.90
(ddd, 1H, J¼13.5, 9.0, 1.1 Hz, H6), 2.67 (dddd, 1H, J¼17.1,
10.7, 8.3, 2.1 Hz, H3) and 2.48 (m, 2H, H30 and H60); m/z
(EI; %) 228 (Mþ, 27), 143 (68), 130 (84), 129 (81), 1116
(46), 83 (30), 55 (100), 43 (75) and 41 (64); n (cm21) 2227
(CN) and 1729 (CvO).
1.3.11. 7-exo-(4-Nitrophenyl)-8-oxa-1-azabicyclo[3.2.1]-
octan-4-one (exo-4i). (Z)-4-Nitrobenzaldoxime (0.332 g,
2 mmol) and divinyl ketone (215 mL, 2.4 mmol) were
reacted according to the general procedure for 48 h.
Subsequent work up afforded the crude product as a yellow
crystalline solid, comprising of exo-4i and unreacted
(E )-oxime (45%). Flash column chromatography, eluting
with diethyl ether, afforded the product (0.095 g, 19%)
which crystallised from dichloromethane–pentane as pale
yellow rods, mp 162–1648C. Found C, 57.9: H, 4.8; N,
11.35; C12H12N2O4 requires C, 58.1; H, 4.85; N, 11.3%; dH
(300 MHz) 8.21 (d, 2H, J¼8.8 Hz, 2£ArH), 7.59 (d, 2H,
J¼8.8 Hz, 2£ArH), 4.58 (m, 2H, H7 and H5), 3.89 (ddd,
1H, J¼14.2, 11.0, 6.1 Hz, H20), 3.44 (dd, 1H, J¼14.2,
8.3 Hz, H2), 2.92 (dd, 1H, J¼13.5, 9.1 Hz, H6), 2.68 (dddd,
1H, J¼17.3, 11.0, 8.3, 2.3 Hz, H3), and 2.49 (m, 2H, H30
and H60); m/z (%) 248 (Mþ, 19), 205 (39), 150 (59), 149
(74), 130 (33), 119 (33), 117 (59), 104 (100), 102 (49), 91
(38), 90 (47), 89 (40), 78 (31), 77 (97), 76 (34), 75 (34), 63
(39), 55 (78), 51 (57), 50 (49) and 42 (53); n (cm21) 1730
(CvO), 1515 (NO) and 1345 (NO).
1.3.9. 7-exo-(4-Trifluoromethylphenyl)-8-oxa-1-azabicyclo-
[3.2.1]octan-4-one(exo-4g). (Z )-4-Trifluoromethylbenzal-
doxime (0.400 g, 2.1 mmol) and divinyl ketone (226 mL,
2.5 mmol) were reacted according to the general procedure
for 48 h. Subsequent work up afforded the crude product as
a viscous orange oil comprising of exo-4g only. Flash
chromatography, eluting with 3:2 v/v ethyl acetate–pentane
afforded the product (0.438 g, 76%) which crystallised
from dichloromethane–petroleum ether as colourless rods,
mp 116–1188C. Found C, 57.35; H, 4.55; N, 5.05;
C13H12F3NO2 requires C, 57.55; H, 4.45; N, 5.15%. dH
(500 MHz) 7.62 (m, 2H, 2£ArH), 7.52 (m, 2H, 2£ArH),
4.56 (d, 1H, J¼8.0 Hz, H5), 4.52 (dd, 1H, J¼9.0, 4.9 Hz,

M. Blackwell et al. / Tetrahedron 58 (2002) 7715–7725
7723
1.3.12. 7-exo-(4-Methoxyphenyl)-8-oxa-1-azabicyclo[3.2.1]-
octan-4-one (exo-4j). (Z)-4-Methoxybenzaldehyde oxime
(0.302 g, 2 mmol) and divinyl ketone (215 mL, 2.4 mmol)
were reacted according to the general procedure for 48 h.
Subsequent work up afforded the crude product as a dark
brown solid, comprising of a 7:1 mixture of exo-4j and
endo-4j. Flash chromatography, eluting with diethyl ether,
afforded the same isomeric mixture (0.182 g, 78%) in a pure
form. Fractional crystallisation from dichloromethane–
pentane afforded a pure sample of exo-4j as colourless
plates, mp 1118C. Found: C, 66.7; H, 6.4; N, 6.0;
C13H15NO3 requires: C, 66.9; H, 6.5; N, 6.0%; dH (500
MHz) 7.30 (d, 2H, J¼8.7 Hz, 2£ArH), 6.78 (d, 2H,
J¼8.7 Hz, 2£ArH), 4.55 (d, 1H, J¼8.0 Hz, H5), 4.41 (dd,
1H, J¼8.8, 5.0 Hz, H7), 3.81 (ddd, 1H, J¼14.2, 10.7,
6.0 Hz, H20), 3.79 (s, 3H, MeO), 3.40 (dd, 1H, J¼14.2,
8.4 Hz, H2), 2.79 (ddd, 1H, J¼13.4, 8.8, 1.1 Hz, H6), 2.64
(ddd, 1H, J¼16.9, 10.7, 8.4 Hz, H3), 2.54 (dddd, 1H,
J¼13.4, 8.0, 5.0, 0.9 Hz, H60) and 2.40 (dd, 1H, J¼16.9,
6.0 Hz, H30); m/z (FAB; %): 234 (Mþ1, 100) and 233 (Mþ,
58); n (cm21) 1727 (CvO).
exo-4l (0.155 g, 70%) which crystallised from dichloro-
methane–pentane as colourless plates, mp 91–938C. Found
C, 65.1; H, 5.55; N, 6.45; C12H12NO2F requires C, 65.15; H,
5.45; N, 6.35%; dH (500 MHz) 7.36 (m, 2H, 2£ArH), 7.02
(m, 2H, 2£ArH), 4.56 (d, 1H, J¼8.0 Hz, H5), 4.44 (dd, 1H,
J¼8.9, 5.0 Hz, H7), 3.83 (ddd, 1H, J¼14.2, 10.7, 6.0 Hz,
H20), 3.40 (dd, 1H, J¼14.2, 8.4 Hz, H2), 2.82 (ddd, 1H,
J¼13.5, 8.9, 1.1 Hz, H6), 2.65 (m, 1H, H3), 2.52 (m, 1H,
H60) and 2.42 (ddt, 1H, J¼17.0, 6.0, 1.2 Hz, H30); m/z (EI;
%); 221 (Mþ, 10), 193 (17), 136 (36), 123 (62), 122 (95),
109 (82), 103 (42), 84 (41), 75 (40), 55 (68), 49 (100) and 42
(66); n (cm21) 1730 (CvO).
1.3.15. 7-endo-(2,4-Dimethoxyphenyl)-8-oxa-1-azabicyclo-
[3.2.1]octan-4-one (endo-4m) and 7-exo-(2,4-dimethoxy-
phenyl)-8-oxa-1-azabicyclo[3.2.1]octan-4-one (exo-4m).
(E )-2,4-Dimethoxybenzaldoxime (0.905 g, 5 mmol) and
divinyl ketone (538 mL, 6 mmol) were reacted according
to the general procedure for 48 h. Subsequent work up
afforded a pale brown solid comprising of a 2:1 mixture of
endo-4m and exo-4m and unreacted oxime (20%). Flash
chromatography, eluting with 4:1 v/v diethyl ether–pen-
tane, afforded endo-4m as a pale yellow oil (0.400 g, 30%)
and exo-4m as a pale yellow oil (0.258 g, 20%).
1.3.13. 7-exo-(4-Chlorophenyl)-8-oxa-1-azabicyclo[3.2.1]-
octan-4-one (exo-4k). (E )-4-Chlorobenzaldoxime (0.311 g,
2 mmol) and divinyl ketone (215 mL, 2.4 mmol) were
reacted according to the general procedure for 48 h.
Subsequent work up afforded a viscous brown oil,
comprising of a 10:1:1 mixture of exo-4k, endo-4k and
exo-3k. Flash chromatography, eluting with diethyl ether,
afforded exo-4k as a colourless solid (0.289 g, 61%) that
crystallised from dichloromethane–pentane as colourless
prisms, mp 129–1308C. Found C, 60.5; H, 5.1; N, 5.8; Cl,
14.85; C12H12ClNO2 requires C, 60.6; H, 5.1; N, 5.9; Cl,
14.9%; dH (300 MHz) 7.31 (m, 4H, 4£ArH), 4.55 (d, 1H,
J¼8.0 Hz, H5), 4.44 (dd, 1H, J¼8.9, 5.0 Hz, H7), 3.82 (ddd,
1H, J¼14.2, 10.7, 6.0 Hz, H20), 3.40 (dd, 1H, J¼14.2,
8.3 Hz, H2), 2.83 (ddd, 1H, J¼13.5, 8.9, 1.1 Hz, H6), 2.64
(ddd, 1H, J¼16.9, 10.7, 8.03 Hz, H3), 2.50 (dddd, 1H,
J¼13.5, 8.0, 5.0, 0.8 Hz, H6 ) and 2.42 (ddt, 1H, J¼16.9,
6.0, 1.2 Hz, H30); m/z (EI; %) 239 37(Mþ, 5), 237 35(Mþ,
31), 152 (51), 140 (41), 138 (100), 131 (32), 125 (51), 103
(62), 77 (35) and 55 (43); n (cm21) 1731 (CvO).
exo-4m. Crystallisation from diethyl ether afforded the
product as colourless needles, mp 92–948C. Found: C,
63.55; H, 6.55; N, 5.35; C14H17NO4 requires: C, 63.85; N,
6.5; N, 5.3%; dH (500 MHz) 7.51 (d, 1H, J¼8.5 Hz, ArH),
6.50 (dd, 1H, J¼8.5, 2.4 Hz, ArH), 6.44 (d, 1H, J¼2.4 Hz,
ArH), 4.72 (dd, 1H, J¼8.7, 5.0 Hz, H7), 4.47 (d, 1H,
J¼8.0 Hz, H5), 3.81 (s, 3H, MeO), 3.80 (s, 3H, MeO), 3.78
(m, 1H, H20), 3.37 (dd, 1H, J¼14.1, 8.5 Hz, H2), 2.80 (dd,
1H, J¼13.5, 8.7, H6), 2.70 (distorted dd, 1H, J¼16.9,
8.5 Hz, H3) and 2.36 (m, 2H, H30 and H60); m/z (FAB; %):
264 (Mþ1, 100) and 263 (Mþ, 66); n (cm21) 1728 (CvO).
1.3.14. 7-exo-(4-Fluorophenyl)-8-oxa-1-azabicyclo[3.2.1]-
octan-4-one (exo-4l). (Z )-4-Fluorobenzaldoxime (0.278 g,
2 mmol) and divinyl ketone (215 mL, 2.4 mmol) were
reacted according to the general procedure for 48 h.
Subsequent work up afforded an orange solid comprising
of a 12:1 mixture of exo-4l and endo-4l. Flash chroma-
tography, eluting with diethyl ether afforded the product

7724
M. Blackwell et al. / Tetrahedron 58 (2002) 7715–7725
endo-4m. Crystallisation from diethyl ether afforded the
product as pale yellow rods, mp 99–1018C. Found C, 63.85;
H, 6.4; N, 5.45; C14H17NO4 requires: C, 63.85; N, 6.5; N,
5.3%; dH (500 MHz; CDCl3) 7.30 (d, 1H, J¼8.4 Hz, ArH),
6.51 (m, 2H, 2£ArH), 5.03 (dd, 1H, J¼9.9, 7.6 Hz, H7),
4.52 (d, 1H, J¼7.9 Hz, H5), 3.87 (s, 3H, MeO), 3.840(s, 3H,
MeO), 3.52 (ddd, 1H, J¼14.4, 9.1 and 7.3 Hz, H2 ), 2.88
(dd, 1H, J¼14.4, 8.5 Hz, H2), 2.76 (dtd, 1H, J¼13.5, 9.9,
1.0 Hz, H60), 2.50 (dd, 1H, J¼13.5, 7.6, 1.9 Hz, H6), 2.40
(ddd, 1H, J¼17.4, 0 9.1, 8.5 Hz, H3) and 2.23 (dd, 1H,
J¼17.4, 7.3 Hz, H3 ); m/z (FAB; %) 264 (Mþ1, 100) and
263 (Mþ, 29); n (cm21) (film) 1729 (CvO).
Structure refinement. Number of parameters¼146, good-
ness of fit, s¼1.059; wR2¼0.1038, R1¼0.0404.
Crystal data for 5. C9H14N2O3: M¼198.22, monoclinic,
space group C2/c, a¼17.3718(3), b¼11.7339(3), c¼
3
˚
˚
11.2632(2) A, b¼122.979(1)8, U¼1925.94(7) A , Z¼8,
Dc¼1.37 g cm23, m¼0.103 mm21, F(000)¼848, T¼190 K.
Data collection. 3.50,u, 30.08; 2263 unique data were
collected (Rint¼0.039); 1967 reflections with Fo.4.0s(Fo).
Structure refinement. Number of parameters¼138, good-
ness of fit, s¼1.081; wR2¼0.1163, R1¼0.0353.
Crystallographic data (excluding structure factors) for the
structures in this paper have been deposited with the
Cambridge Crystallographic Data Centre as supplementary
publication numbers CCDC182398, 182399 and 182400.
Copies of the data can be obtained, free of charge, on
application to CCDC, 12 Union Road, Cambridge,
CB2 1EZ, UK (fax: þ44-1223-336033 or e-mail: deposit@
ccdc.cam.ac.uk).
Acknowledgments
We thank EPSRC and Pfizer (UK) for CASE studentships
(to I. S. S. and A. G.), the University of Leeds for support
and the National Mass Spectrometry Service for mass
spectra.
1.4. Single-crystal X-ray analyses
Crystallographic data for all three compounds were
measured on a Nonius KappaCCD area-detector diffracto-
meter using a mixture of area detector v and f-scans and
References
˚
Mo Ka radiation (l¼0.71073 A). All three structures were
solved by direct methods using SHELXS-8612 and were
refined by full-matrix least-squares (based on F 2) using
SHELXL-97.13 All non-hydrogen atoms were refined with
anisotropic displacement parameters whilst hydrogen atoms
were constrained to predicted positions using a riding
model. The residuals wR2 and R1, given below, are
1. Part 55. Dunn, P. J.; Graham, A. B.; Grigg, R.; Saba, I. S.;
Thornton-Pett, M. Tetrahedron, 2002, 58, 7701–7713.
2. (a) Grigg, R.; Markandu, J.; Surendrakumar, S.; Thornton-
Pett, M.; Warnock, W. J. Tetrahedron 1991, 47, 4007–4030.
(b) Grigg, R.; Heaney, F.; Surendrakumar, S.; Warnock, W. J.
Tetrahedron 1991, 47, 4477–4494. (c) Armstrong, P.; Grigg,
R.; Heaney, F.; Surendrakumar, S.; Warnock, W. J.
Tetrahedron 1991, 47, 4495–4518. (d) Grigg, R.; Dorrity,
M.; Heaney, F.; Malone, J. F.; Rajviroonjit, S.; Sridharan, V.;
Surendrakumar, S. Tetrahedron 1991, 47, 8297–8322.
(e) Grigg, R.; Markandu, J.; Perrior, T.; Surendrakumar, S.;
Warnock, W. J. Tetrahedron 1992, 48, 6929–6952.
3. Saba, I. S.; Frederickson, M.; Grigg, R.; Dunn, P. J.; Levett,
P. C. Tetrahedron Lett. 1997, 38, 6099–6102.
P
P
defined as wR2 ¼ ð ½wðFo2 2 Fc2Þ2ꢀ= ½wFo2ꢀ2Þ1=2 and R1 ¼
P
P
kFol 2 lFck= lFol:
Crystal data for 4e. C10H18N2O2, M¼198.26, monoclinic,
space group C2/c, a¼34.309(3), b¼5.8733(5), c¼
3
˚
˚
10.7105(5) A, b¼96.918(4)8, U¼2142.5(3) A , Z¼8,
Dc¼1.23 g cm23, m¼0.086 mm21, F(000)¼864, T¼150 K.
Data collection. 3.52,u,25.998; 2001 unique data were
collected (Rint¼0.044); 1602 reflections with Fo.4.0s(Fo).
4. Preliminary communication: Dunn, P. J.; Graham, A. B.;
Grigg, R.; Saba, I. S.; Higginson, P. J. Chem. Soc., Chem.
Commun. 2000, 20, 2033–2034.
Structure refinement. Number of parameters¼132, good-
ness of fit, s¼1.030; wR2¼0.1196, R1¼0.0431.
5. Boyd, D. R.; Neill, D. C. J. Chem. Soc., Perkin Trans. 1 1977,
11, 1308–1313. Boyd, D. R.; Neill, D. C.; Stubbs, M. E.
J. Chem. Soc., Perkin Trans. 1 1974, 39, 3447–3449.
Sommermeyer, K.; Seifert, W.; Wilker, W. Tetrahedron
Lett. 1974, 15, 1821–1824.
Crystal data for 4k. C12H12ClNO2: M¼237.68, orthor-
hombic, space group P212121, a¼9.1352(6), b¼10.5378(8),
3
23
˚
˚
c¼11.1591(11) A, U¼1074.2(2) A , Z¼4, Dc¼47 g cm
,
6. Oeda, H. Bull. Chem. Soc. Jpn 1934, 9, 8–12.
7. Cordes, E. H.; Jencks, W. P. J. Am. Chem. Soc. 1963, 85,
2843–2848.
m¼0.338 mm21, F(000)¼496, T¼150 K.
Data collection. 2.50,u,30.08; 2051 unique data were
collected (Rint¼0.053); 1973 reflections with Fo.4.0s(Fo).
8. Grigg, R.; Gunaratne, H. Q. N.; Kemp, J. J. Chem. Soc., Perkin
Trans. 1 1984, 41–46.

M. Blackwell et al. / Tetrahedron 58 (2002) 7715–7725
7725
9. The pKa’s were determined using the ACD/I-Lab service,
version 4.5.
Chung, J. C.; Costello, T. D.; Valvis, I.; Ma, P.; Kauffman, S.;
Ward, R. J. Org. Chem. 1997, 62, 2466–2470. E-1j: Majer,
J. R.; Azzous, A. S. P. J. Chem. Soc., Faraday Trans. 1983, 79,
675–688. E-1k: Kawase, M.; Kikugawa, Y. J. Chem. Soc.,
Perkin Trans. 1 1979, 3, 643–645.
10. 1b: Vogel, A. I.; Cresswell, W. T.; Jeffrey, G. H.; Leicester, J.
J. Am. Chem. Soc. 1952, 1, 514–541. 1c: Johnson, D. J. Am.
Chem. Soc. 1939, 3194. 1d: DeJohngh, D. C.; Lin, D. C. K.;
LeClaire-Lanteignew, P.; Gravel, D. Can. J. Chem. 1975, 53,
3175–3187. 1e: Dornow, A.; Mueller, A. Chem. Ber. 1960,
93, 32–40. E-1g: Barfknecht, C. F.; Westby, T. R. J. Med.
Chem. 1967, 10, 1192–1193. E-1h and Z-1h: Zhang, L. H.;
11. Jones, N.; Taylor, H. T. J. Chem. Soc. 1961, 1345–1347.
12. Sheldrick, G. M. Acta Crystallogr. 1990, A46, 467–473.
13. Sheldrick, G. M. SHELXl-97: Program for Refinement of
¨
Crystal Structures; University of Gottingen, Germany, 1993.