Exploring the stereoselectivity in the peterson reaction of several 2- substituted 1-azabicyclo[2.2.2]octan-3-ones

Article abstract of DOI:10.1016/S0040-4020(00)00015-6

The Peterson reaction between a series of 2-substituted 1- azabicyclo[2.2.2]octan-3-ones and 3-methyl-5-trimethylsilanylmethyl isoxazole (1) was explored. (Z)-Stereoselectivity was obtained with all 2-substituents, ranging from 65:35 for benzyl to 95:5 for phenylsulphanyl. Different bases were investigated, revealing that an organolithium base is necessary for the reaction to occur. A transition state is suggested, involving a four-membered ring in which the lithium ion is covalently bound α to the silyl-group and chelated to the carbonyl oxygen. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4020(00)00015-6

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 1139–1146
Exploring the Stereoselectivity in the Peterson Reaction of Several
2-Substituted 1-Azabicyclo[2.2.2]octan-3-ones
Janne E. Tønder,a Mikael Begtrup,b John Bondo Hansena and Preben H. Olesena,
*
a
˚
Health Care Discovery, Novo Nordisk A/S, Novo Nordisk Park, DK-2760 Maløv, Denmark
bDepartment of Medicinal Chemistry, The Royal Danish School of Pharmacy, Universitetsparken 2, DK-2100 Copenhagen, Denmark
Received 11 October 1999; revised 29 November 1999; accepted 16 December 1999
Abstract—The Peterson reaction between a series of 2-substituted 1-azabicyclo[2.2.2]octan-3-ones and 3-methyl-5-trimethylsilanylmethyl
isoxazole (1) was explored. (Z)-Stereoselectivity was obtained with all 2-substituents, ranging from 65:35 for benzyl to 95:5 for phenyl-
sulphanyl. Different bases were investigated, revealing that an organolithium base is necessary for the reaction to occur. A transition state is
suggested, involving a four-membered ring in which the lithium ion is covalently bound a to the silyl-group and chelated to the carbonyl
oxygen. ᭧ 2000 Elsevier Science Ltd. All rights reserved.
Introduction
The stereoselectivity of the Peterson reaction may originate
in the initial C–C bond forming step to give a b-hydroxy-
silane or in the subsequent elimination of hexaalkyl disil-
oxane to give the alkene. When the intermediate
b-hydroxysilane can be isolated it has been established
that acidic conditions lead to anti-elimination, whereas
alkaline conditions lead to syn-elimination.3–7
In the search for compounds with agonistic effect at the
nicotinic acetylcholine receptors, a series of (isoxazole)-
methylene-1-azabicyclic compounds were recently
prepared.1 All compounds shared a double bond, connecting
an azabicyclic back bone to a substituted isoxazole and for
all pairs of (Z/E)-isomers, the (Z)-form possessed the
highest affinity; (Z)-3-(3-methyl-isoxazol-5-yl)methylene-
1-azabicyclo[2.2.2]octane (6g) being one of the most potent
compounds.1
In the case of stabilised a-silyl carbanions the elimination
step is too fast to allow for isolation of the intermediate
b-hydroxysilanes. Hence, the stereocontrol must reside in
the C–C bond forming step. The factors influencing the
stereoselectivity of the C–C bond formation have been
examined in a number of studies,8–11 and it is generally
accepted, that (Z)-alkenes are favoured. However, even
though different transition states have been suggested, no
overall mechanistic explanation has been given so far. In
In order to develop a (Z)-stereoselective synthesis for these
compounds the Peterson reaction between the easily available
3-methyl-5-trimethylsilanylmethyl isoxazole2 (1) and deri-
vatives of 1-azabicyclo[2.2.2]octan-3-one (2) has now been
examined (Scheme 1).
Scheme 1. (a) Organometallic base, 3-methyl-5-trimethylsilanylmethyl isoxazole (1) in THF, Ϫ78ЊC, N2; 6a!6g: SnBu3H, AIBN in methanol, 15 min at
120ЊC.
Keywords: stereocontrol; olefination; bicyclic heterocyclic compounds; isoxazoles.
* Corresponding author. Tel.: ϩ45-4443-4886; fax: ϩ45-4443-4547; e-mail: phol@novo.dk
0040–4020/00/$ - see front matter ᭧ 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)00015-6

1140
J. E. Tønder et al. / Tetrahedron 56 (2000) 1139–1146
disulphide is resistant to ketone enolates,15 DMPU was
added to enhance the reactivity.
Three 1-azabicyclo[2.2.2]octan-3-ones with bulky alkyl-
substituents in the 2-position were prepared following a
modified literature procedure for the synthesis of 5d
(Scheme 3).16 1-Azabicyclo[2.2.2]octan-3-one (2) was
treated with potassium hydroxide in the presence of an alde-
hyde, giving rise to the expected aldol condensation. Due to
their bulkiness, pivalic aldehyde and 2-ethylbutyric alde-
hyde reacted very slowly to give 4e,f and yields were only
36 and 44%. Nevertheless, the subsequent hydrogenation to
give 5e,f was very fast and proceeded in 76 and 96% yield.
Scheme 2. (a) a,b (i) LDA in THF, 0ЊC, N2, (ii) Electrophile, N2, c (i) LDA
in THF, 0ЊC, N2, (ii) DMPU, 0ЊC, N2, (iii) Electrophile, N2.
one case10 a six-membered transition state involving lithium
has been suggested. In cases where chelation control is
absent an approach-control model was suggested.11,12
Apparently, the stereoselectivity is highly dependent on
the reaction conditions, thereby allowing optimisation of
the nature of starting materials and reaction conditions.
The Peterson reaction of 3a–c and 5d–f with 3-methyl-5-
trimethylsilanylmethyl isoxazole (1) was performed in THF
using LDA as the base producing 6a–f and 7a–f in 32–92%
yield (Scheme 1). The isomers 6 and 7 were separated by
column chromatography and purified by crystallisation of
their derived oxalates.
In the present work, we report an investigation of some of
the factors influencing the (Z)-stereoselectivity of the
Peterson reaction, by the reaction between a-substituted
1-azabicyclo[2.2.2]octan-3-ones and 3-methyl-5-trimethyl-
silanylmethyl isoxazole (1). We suggest a four-membered
transition state, which can explain the observed variations in
stereoselectivity. Finally the synthesis of (Z)-3-(3-methyl-
isoxazol-5-yl)methylene-1-azabicyclo[2.2.2]octane (6g) via
(RS)-2-phenylsulphanyl-1-azabicyclo[2.2.2]octan-3-one (3a)
was attempted.
The relative configuration of the (Z)- and (E)-compounds
was determined on the basis of 1H NMR data. After assign-
ing the individual protons from COSY experiments, NOE-
effects were used for determination of the spatial arrange-
ment around the double bond. The protons marked HA–HD
(Scheme 4) were employed in this determination.
In a (Z)-compound (i.e. 6a) NOE-effects were observed
between HA and HD and between HB and HC. Accordingly,
in an (E)-compound (i.e. 7a) NOE-effects were observed
between HA and HC and between HB and HD. In the one-
dimensional 1H NMR spectrum it proved possible to distin-
guish between (Z)- and (E)-compounds by means of the
chemical shifts of HA, HC and HD. The largest difference
in chemical shifts between (E)- and (Z)-forms was seen for
Results and Discussion
(Z)-3-(3-Methyl-isoxazol-5-yl)methylene-1-azabicyclo[2.2.2]-
octane (6g) was first synthesised directly from 3-methyl-5-
trimethylsilanylmethyl isoxazole (1) and 1-azabicyclo-
[2.2.2]octan-3-one (2). The result was a mixture of (Z)-
and (E)-isomers in the ratio 60:40. Due to difficulties
Scheme 3. (a) R2CHO, KOH in methanol. (b) H2, 10% Pd/C in methanol.
experienced in the separation of the two isomers, the
isolated yield of 6g was only 28%.13 In an attempt to
increase the (Z)-stereoselectivity, a large number of sub-
stituents were incorporated in the carbonyl a-position of
(2). The 2-substituents were selected on basis of steric and
electrostatic properties and both p-systems (SPh, SePh, Bn),
electronegative atoms (SPh, SePh, SMe) and aliphatic
chains (CH2C(CH3)3, CH2CH(CH2CH3)2) were introduced.
The substituents were introduced by generation of the
enolate of 1-azabicyclo[2.2.2]octan-3-one (2) by treatment
with LDA,14 followed by addition of electrophile. This gave
the 2-substituted 1-azabicyclo[2.2.2]octan-3-ones (3a–c) in
32–68% isolated yield (Scheme 2). When the electrophile
was diphenyl disulphide or dimethyl disulphide, the best
yields were obtained using 2.4 equiv. of both LDA and
electrophile.15 Since the stronger S–S bond of dimethyl
Scheme 4. Protons employed in the structure determination.

J. E. Tønder et al. / Tetrahedron 56 (2000) 1139–1146
1141
Table 1. Variation of base and 2-substituent
resonance structure is unable to be part of a six-
membered ring.
Ketone R1
Base
Yield (%) (ZϩE) (Z:E) (%)
213
3a
3a
3a
3a
3b
3c
5d
5e
5f
H
SPh
SPh
SPh
SPh
SePh
SMe
LDA
LDA
n-BuLi
KN(SiMe3)2
KH
LDA
LDA
LDA
LDA
83
72
73
0
60:40
95:5
80:20
–
Presumably the lithium ion is involved in a four-membered
cyclic transition state with lithium covalently bound to the
silyl substituted carbon atom and simultaneously coordi-
nated to the carbonyl oxygen atom (Scheme 5). When race-
mic ketone was used as the starting material, two
enantiomeric transition states are possible. However, as
they both lead to the same alkene, only one enantiomer is
considered in the following. It is assumed that the Liϩ–Hϩ
interchange in the a-silyllithium species is fast, enabling
fast equilibration to the low energy enantiomer.
0
–
85
32
79
92
81
75:25
88:12
65:35
80:20
75:25
CH2Ph
CH2C(CH3)3
CH2CH(CH2CH3)2 LDA
dHC. The signal was shifted almost 1 ppm downfield in the
(E)-form where the isoxazole is situated closely to HC.
The R-substituent of the ketone probably impedes attack by
the a-silyllithium species from the face where it is situated.
According to CPK models the trimethylsilyl (SiMe3) group
is considerably larger than the isoxazole. Therefore, the
SiMe3 group will be oriented opposite to the carbonyl
compound as in transition state A and B (Scheme 5). CPK
models also reveal that in the silyllithoxy intermediate, the
face with the bridgehead proton (HC) takes up more space
than HD and R. Thus, transition state B leading to the (Z)-
alkene is more favoured than transition state A.
GC-analysis of the crude product mixture showed that in all
cases the (Z)-product was the predominant isomer (Table 1).
Changing from LDA to n-BuLi lowered the stereoselec-
tivity, but still the (Z)-isomer was the major product. In
17
contrast to LDA ꢀpKA ˆ 36† and n-BuLi ꢀpKA ˆ 50†;17
17
neither potassium bis(trimethylsilyl)amide ꢀpKA ˆ 26†
17
nor potassium hydride ꢀpKA ˆ 35† lead to alkene for-
mation.
According to this model, the relative amount of (Z)-alkene is
expected to drop when the size of R is increased. In fact, this
trend was observed when the size of R was increased from a
sulphur substituent to more bulky selenium or CH2 sub-
stituents. According to MMFF calculations, the volume of
the CH2 group is larger than that of the sulphur atom in the
direction of the hydrogen atoms and it will therefore have a
larger steric effect on the approaching a-silyllithium species
than the sulphur atom.
This indicates that the lithium ion is involved in the C–C
bond formation step of the Peterson reaction studied. The
apparent chelation control exerted by the lithium ion makes
it impossible to use the approach-control model, as it was
derived for cases in which chelation control is absent.11,12 In
the model proposed by Bell and coworkers10 the stabiliser of
the a-silyl-anion is a nitrogen atom in the 2-oxazolyl group,
which enables delocalisation of the negative charge into the
oxazole ring. The resulting resonance structure is suggested
to be a part of a six-membered transition state, which then
determines the stereochemical outcome of the reaction.
Such an electron delocalisation is also possible when
the stabilising group is 5-isoxazolyl, but the resulting
In the case of RˆH, one would expect the reaction to be
highly (Z)-stereoselective. However, it is possible, that the
trimethylsilyl-group can interchange with the isoxazole-
group creating another transition state (C), now that the
Scheme 5. (i) C–C bond formation; (ii) syn-elimination.

1142
J. E. Tønder et al. / Tetrahedron 56 (2000) 1139–1146
Scheme 6.
total steric effect of HD and R is much less around the
2-position in the ketone. As illustrated in Scheme 6 tran-
sition state C leads to formation of the (E)-alkene.
involving moisture-sensitive reagents were performed
under N2 using standard septum techniques. The (Z/E)-
ratios of the crude products were measured by gas chroma-
tography (Chrompack CP9001).
Based on this reasoning, a strategy for the (Z)-stereoselec-
tive synthesis of 6g was designed using the phenylsulphanyl
group to improve stereocontrol (Scheme 1). The synthesis of
6a in gram scale via 3a proceeded in 24% overall yield
using recrystallisation for purification in both steps. The
final removal of the assisting phenylsulphanyl group by
reduction with tributyltin hydride turned out to be difficult,
as the yield of 6g from 6a was only 12%.
3-Methyl-5-trimethylsilanylmethyl isoxazole (1)2, (RS)-2-
benzylidene-1-azabicyclo[2.2.2]octan-3-one (4d),16 and
(RS)-2-benzyl-1-azabicyclo[2.2.2]octan-3-one (5d)16 were
synthesised following literature procedures.
Method A
(RS)-2-Phenylsulphanyl-1-azabicyclo[2.2.2]octan-3-one
(3a). To a solution of diisopropylamine (6.30 mL, 0.048
mol) in THF (150 mL) at 0ЊC n-butyllithium (2.50 M in
hexanes, 19.2 mL, 0.048 mol) was added dropwise over a
period of 5 min. The resulting yellow solution was stirred at
0ЊC for 30 min, whereupon 1-azabicyclo[2.2.2]octan-3-one
(2) (2.50 g, 0.020 mol) dissolved in THF (10 mL) was
added. After 30 min at 0ЊC, diphenyl disulphide (10.5 g,
0.048 mol) was added as a solution in THF (25 mL). Stirring
the reaction mixture at 20ЊC for 2 h was followed by
quenching with water (200 mL). The alkaline water phase
(pHˆ11) was extracted with dichloromethane (4×150 mL),
and the combined organic extracts were dried (MgSO4), and
concentrated in vacuo to give 11.4 g of a yellow oil. Column
chromatography (eluent: ethyl acetate/dichloromethane
1:10) gave 2.34 g (50%) of the title product as yellowish
crystals.
Conclusion
It has proved possible to achieve (Z)-stereoselectivity by
insertion of a bulky substituent a to the carbonyl group.
The observed variations in stereoselectivity could be
explained by steric interactions in a transition state in
which a lithium ion is taking part.
The (Z)-stereoselectivity of the Peterson reaction between 2-
substituted 1-azabicyclo[2.2.2]octan-3-ones and 3-methyl-5-
trimethylsilanylmethyl isoxazole (1) decreases with the size
of the substituent a to the carbonyl group. This can be
rationalised if the C–C bond formation is the rate limiting
step passing a four-membered cyclic transition state with
lithium covalently bound to the silyl substituted carbon
atom and simultaneously coordinated to the carbonyl
oxygen atom.
Mp 116–118ЊC. MS m/z (%): 234 (Mϩϩ1, 3), 205 (100),
172 (27), 128 (33), 110 (42), 96 (19). IR (KBr, cmϪ1): 3422,
1
2944, 2856, 1728, 1472, 1072, 739. H NMR (300 MHz,
Experimental
CDCl3) d 7.55 (d, 2H, Jˆ7 Hz, Hmeta); 7.32–7.29 (m, 3H,
H
orthoϩHpara); 4.76 (s, 1H, NCHSCy); 3.72–3.60 (m, 1H,
General methods and materials
NCH2); 3.32–3.18 (m, 1H, NCH2); 3.10–2.95 (m, 1H,
NCH2); 2.94–2.80 (m, 1H, NCH2); 2.57 (m, 1H, CH);
2.03 (t, 4H, Jˆ7 Hz, CH2). Anal. (C13H15NOS): Calcd C,
66.92; H, 6.48; N, 6.00. Found C, 67.04; H, 6.68; N, 5.98.
1H NMR spectra were recorded at 300 MHz on a Bruker
AC-300 MHz FT-NMR instrument, and COESY, NOESY
and ROESY spectra were recorded at 400 MHz on a Bruker
AC-400 MHz FT-NMR instrument. Mass spectra were
recorded on a Finnigan 5100 mass spectrometer, and mel-
ting points (uncorrected) were determined on a Buchi capil-
lary melting point apparatus. IR were recorded with a
Perkin–Elmer 1600 spectrometer. Column chromatography
was performed on silica gel 60 (70–230 mesh, ASTM,
Merck). Elemental analyses were performed by Novo
Nordisk Microanalytical Laboratory, Denmark, and were
within ^0.4% of the calculated values. TLC was monitored
via UV or potassium permanganate spray. All reactions
The following compounds were prepared according to
Method A with the variations stated below.
(RS)-2-Phenylselanyl-1-azabicyclo[2.2.2]octan-3-one (3b).
In 68% yield starting from LDA (1.2 equiv.), 1-azabicy-
clo[2.2.2]octan-3-one (2) (1.0 equiv.), and phenylselenenyl
chloride (1.05 equiv.). Phenylselenenyl chloride was added
at Ϫ78ЊC, and the reaction mixture stirred at Ϫ78ЊC for
30 min, before it was allowed to warm up to 20ЊC.

J. E. Tønder et al. / Tetrahedron 56 (2000) 1139–1146
1143
Mp 90–92ЊC. MS m/z (%): 280 (Mϩ, 1), 253 (100), 176
Jˆ3 Hz, CH); 2.02–1.83 (m, 4H, CH2); 1.68 (dd, 1H,
Jˆ14 Hz, 4 Hz, R–CH2); 1.47 (dd, 1H, Jˆ14 Hz, 8 Hz,
R–CH2); 0.98 (s, 9H, CH3). Anal. (C12H21NO): Calcd C,
73.80; H, 10.84; N, 7.17. Found C, 73.77; H, 11.13; N, 7.19.
(49), 172 (74), 145 (86) 96 (47). IR (KBr, cmϪ1): 3424,
1
2948, 2870, 1722, 1477, 1070, 736. H NMR (300 MHz,
CDCl3) d 7.64 (m, 2H, Hmeta); 7.26 (m, 3H, HorthoϩHpara);
5.02 (s, 1H, NCHSeCy); 3.68–3.49 (m, 1H, NCH2); 3.35–
3.16 (m, 1H, NCH2); 3.09–2.78 (m, 2H, NCH2); 2.50 (quin-
tet, 1H, Jˆ3 Hz, CH); 2.07–1.87 (m, 4H, CH2). Anal.
(C13H15NOSe): Calcd C, 55.72; H, 5.40; N, 5.00. Found
C, 55.69; H, 5.43; N, 4.95.
The following compounds were prepared according to
Method B with the variations stated below.
2-(2-Ethyl-butylidene)-1-azabicyclo[2.2.2]octan-3-one (4f).
In 44% yield starting from 1-azabicyclo[2.2.2]octan-3-one
(2) and 2-ethylbutyric aldehyde, the reaction mixture was
stirred at 60ЊC for 7 days. The resulting colourless oil was
crystallised as the oxalate salt.
(RS)-2-Methylsulphanyl-1-azabicyclo[2.2.2]octan-3-one
(3c). In 32% yield starting from LDA (2.4 equiv.), 1-azabi-
cyclo[2.2.2]octan-3-one (2) (1.0 equiv.), and dimethyl
disulphide (2.4 equiv. DMPU (5.0 equiv.) was added and
the reaction mixture stirred for 10 min before the addition
of dimethyl disulphide. The title compound decomposed
when subjected to column chromatography. Thus, the
crude product was partly purified by crystallisation with
oxalic acid from acetone, followed by a recrystallisation
from acetone. The product (white crystals) was used without
further purification.
Mp 96–97ЊC. MS m/z (%): 207 (Mϩ, 44), 179 (17), 164
(37), 150 (100), 136 (40), 122 (76). IR (KBr, cmϪ1):
3453, 2964, 2862, 1732, 1661. 1H NMR (300 MHz,
CDCl3) d 6.15 (s, 1H, CHy (Z)); 6.11 (s, 1H, CHy (E));
3.19–3.05 (m, 2H, NCH2); 2.94–2.81 (m, 2H, NCH2);
2.72–2.58 (m, 1H, CHCy); 2.58 (quintet, 1H, Jˆ3 Hz,
CH); 1.96 (dt, 4H, Jˆ8 Hz, 3 Hz, q-CH2); 1.60–1.44 (m,
2H, R–CH2); 1.37–1.22 (m, 2H, CH2); 0.85 (t, 6H, Jˆ7 Hz,
CH3). Anal. (C13H21NO, C2H2O4): Calcd C, 60.59; H, 7.80;
N, 4.71. Found C, 60.66; H, 7.81; N, 4.70.
MS m/z (%): 171 (Mϩ, 1), 143 (87), 128 (100), 110 (12), 96
(15). 1H NMR (300 MHz, CDCl3) d 4.32 (s, 1H, NCHSCy);
3.58–3.45 (m, 1H, NCH2); 3.31–3.19 (m, 1H, NCH2);
3.03–2.92 (m, 1H, NCH2); 2.85–2.72 (m, 1H, NCH2);
2.48 (quintet, 1H, Jˆ3 Hz, CH); 2.19 (s, 3H, CH3); 2.05–
1.90 (m, 4H, CH2).
(RS)-2-(2-Ethyl-butyl)-1-azabicyclo[2.2.2]octan-3-one 5f.
In 76% yield starting from 2-(2,2-dimethyl-propylidene)-1-
azabicyclo[2.2.2]octan-3-one (4f), the colourless oil was
crystallised as the oxalate salt.
Method B (Aldol condensation followed by hydro-
genation)
Mp 151–153ЊC. MS m/z (%): 209 (Mϩ, 1), 181 (42), 152
(100), 124 (11), 111 (56), 110 (42), 96 (28). IR (KBr, cmϪ1):
3436, 2964, 2877, 1734, 1636. 1H NMR (300 MHz, CDCl3)
d 3.21–2.98 (m, 3H, NCH2ϩNCHCy); 2.98–2.75 (m, 2H,
NCH2); 2.40 (quintet, 1H, Jˆ3 Hz, CH); 2.03–1.84 (m, 4H,
q-CH2); 1.72–1.59 (m, 1H, R–CH); 1.57–1.18 (m, 6H, R–
CH2); 0.88 (t, 3H, Jˆ7 Hz, CH3); 0.85 (t, 6H, Jˆ7 Hz, CH3).
Anal. (C13H23NO, C2H2O4): Calcd C, 60.18; H, 8.42; N,
4.68. Found C, 60.21; H, 8.63; N, 4.70.
2-(2,2-Dimethyl-propylidene)-1-azabicyclo[2.2.2]octan-3-
one (4e). KOH (1.46 g, 0.026 mol) was added to a solution
of 1-azabicyclo[2.2.2]octan-3-one (2) (11.0 g, 0.087 mol) in
methanol (75 mL). When all KOH had dissolved, pivalic
aldehyde (9.90 mL, 0.091 mol) was added and the reaction
mixture was stirred at 60ЊC for 12 days. Evaporation
followed by filtration through silica gel using ethyl acetate
as eluent gave 6.09 g (36%) of 4e as yellow crystals.
Method C (Peterson reaction)
Mp 67–69ЊC. MS m/z (%): 193 (Mϩ, 63), 178 (6), 165
(100), 150 (66), 137 (30), 122 (100). IR (KBr, cmϪ1):
(Z)-(RS)-3-(3-Methyl-isoxazol-5-ylmethylene)-2-phenyl-
sulphanyl-1-azabicyclo[2.2.2]octane (6a) and (E)-(RS)-3-
(3-methyl-isoxazol-5-ylmethylene)-2-phenylsulphanyl-1-
azabicyclo[2.2.2]octane (7a). To a solution of diisopropyl-
amine (0.56 mL, 0.0043 mol) in THF (15 mL) at 0ЊC,
n-butyllithium (2.50 M in hexanes, 1.80 mL, 0.0043 mol)
was added dropwise over a period of 5 min. The resulting
yellow solution was stirred at 0ЊC for 30 min, after which
3-methyl-5-trimethylsilanylmethyl isoxazole (1) (0.609 g,
0.0036 mol) was slowly added. After 15 min at 0ЊC the
temperature was lowered to Ϫ78ЊC and stirring was contin-
ued for another 15 min. Over a period of 10 min (RS)-2-
phenylsulphanyl-1-azabicyclo[2.2.2]octan-3-one (3a) (0.699
g, 0.0030 mol) was added as a solution in THF (35 mL).
This was followed by stirring at Ϫ78ЊC for 5 h, after
which the reaction was quenched with 1 N HCl (10 mL)
and left stirring at 20ЊC overnight. Ethyl acetate (20 mL)
was added to the yellow solution, the layers were separated,
and the organic phase extracted with 2×25 mL 1N HCl. The
combined aqueous extracts were made alkaline (K2CO3)
and extracted with dichloromethane (4×25 mL). Drying
1
3398, 2961, 2868, 1706, 1637, 1252. H NMR (300 MHz,
CDCl3) d 6.41 (s, 1H, CHy); 3.11–3.00 (m, 2H, NCH2);
2.92–2.80 (m, 2H, NCH2); 2.51 (quintet, 1H, Jˆ3 Hz, CH);
1.94 (dt, 4H, Jˆ7 Hz, 3 Hz, CH2); 1.20 (s, 9H, CH3). Anal.
(C12H19NO) Calcd C, 74.57; H, 9.91; N, 7.25. Found C,
74.62; H, 10.18; N, 7.27.
(RS)-2-(2,2-Dimethyl-propyl)-1-azabicyclo[2.2.2]octan-3-
one (5e). A suspension of 2-(2,2-dimethyl-propylidene)-1-
azabicyclo[2.2.2]octan-3-one (4e) (3.00 g, 0.0155 mol) in
methanol (250 mL) was hydrogenated in a Parr shaker at
35 psi over 10% Pd/C. After the hydrogen consumption was
complete (2 h), the mixture was filtered through Celite and
concentrated in vacuo to give 2.92 g (96%) of 5e as colour-
less crystals.
Mp 62–64ЊC. MS m/z (%): 195 (Mϩ, 8), 193 (58), 165 (88),
150 (63), 137 (33), 122 (100). IR (KBr, cmϪ1): 3413, 2956,
1
2870, 1718. H NMR (300 MHz, CDCl3) d 3.16–2.90 (m,
4H, NCH2); 2.85–2.72 (m, 1H, NCHCy); 2.39 (quintet, 1H,

1144
J. E. Tønder et al. / Tetrahedron 56 (2000) 1139–1146
(MgSO4) and concentration in vacuo gave 0.960 g of a
brown oil. Gas chromatography of the crude product
showed, that the (Z/E) ratio was 95:5%. Column chroma-
tography (eluent: methanol/dichloromethane 1:15) gave
0.57 g (61%) (Z)-(RS)-3-(3-methyl-isoxazol-5-ylmethy-
lene)-2-phenylsulphanyl-1-azabicyclo[2.2.2]octan-3-one (6a)
as a colourless oil and 0.10 g (11%) as a mixture of 6a and
7a. 6a was crystallised as the oxalate salt.
7b. Mp 134–136ЊC. MS m/z (%): 360 (Mϩϩ1, 12), 279 (5),
203 (100), 162 (81), 148 (15), 134 (41). IR (KBr, cmϪ1):
1
3435, 2959, 2544, 1719, 1635, 1566, 1201, 999, 690. H
NMR (300 MHz, CDCl3) d 7.68–7.59 (m, 2H, Hmeta);
7.35–7.23 (m, 3H, HorthoϩHpara); 6.56 (d, 1H, Jˆ3 Hz,
CHy); 5.93 (s, 1H, isox-H); 5.48 (s, 1H, NCHSeCy);
3.75–3.58 (m, 1H, NCH2); 3.50 (quintet, 1H, Jˆ4 Hz,
CH); 3.32–3.12 (m, 1H, NCH2); 3.01–2.72 (m, 2H,
NCH2); 2.28 (s, 3H, CH3); 1.89–1.62 (m, 4H, CH2). Anal.
(C18H20N2OSe, C2H2O4): Calcd C, 53.46; H, 4.93; N, 6.23.
Found C, 53.16; H, 5.02; N, 5.98.
6a. Mp 167–168ЊC. MS m/z (%): 312 (Mϩ, 35), 279 (9), 203
(100), 162 (70), 148 (12), 134 (32). IR (KBr, cmϪ1): 3433,
1
2953, 2813, 1735, 1699, 1569, 1207, 1001, 698. H NMR
(300 MHz, CDCl3) d 7.60 (d, 2H, Jˆ7 Hz, Hmeta); 7.38–
7.20 (m, 3H, HorthoϩHpara); 6.31 (d, 1H, Jˆ1 Hz, CHy);
6.13 (s, 1H, isox-H); 5.37 (s, 1H, NCHSCy); 3.92–3.79
(m, 1H, NCH2); 3.28–3.13 (m, 1H, NCH2); 2.84–2.69 (m,
2H, NCH2); 2.57 (m, 1H, CH); 2.28 (s, 3H, CH3); 1.92–1.63
(m, 4H, CH2). Anal. (C18H20N2OS, 1.5 C2H2O4): Calcd C,
56.37; H, 5.18; N, 6.26. Found C, 56.52; H, 5.31; N, 6.26.
(Z)-(RS)-3-(3-Methyl-isoxazol-5-ylmethylene)-2-methyl-
sulphanyl-1-azabicyclo[2.2.2]octane (6c) and (E)-(RS)-3-
(3-methyl-isoxazol-5-ylmethylene)-2-methylsulphanyl-1-
azabicyclo[2.2.2]octane (7c). In 75% total yield starting
from 3-methyl-5-trimethylsilanylmethyl isoxazole (1) and
(RS)-2-methylsulphanyl-1-azabicyclo[2.2.2]octan-3-one (3c).
Gas chromatography of the crude product showed that the
(Z/E) ratio was 88:12%. 7c could not be isolated as a pure
isomer, and in the 1H NMR spectrum of the product mixture
the overlap of the aliphatic peaks was too pronounced to
enable a proper assignment. 6c (colourless oil) was crystal-
lised as the oxalate salt.
1
7a. H NMR (300 MHz, CDCl3) d 7.53 (m, 2H, Hmeta);
7.35–7.18 (m, 3H, HorthoϩHpara); 6.61 (d, 1H, Jˆ3 Hz,
CHy); 5.97 (s, 1H, isox-H); 5.18 (s, 1H, NCHSCy);
3.75–3.54 (m, 1H, NCH2); 3.54 (quintet, 1H, Jˆ3 Hz,
CH); 3.28–3.10 (m, 1H, NCH2); 3.05–2.71 (m, 2H,
NCH2); 2.29 (s, 3H, CH3); 1.86–1.58 (m, 4H, CH2).
6c. Mp 158–159ЊC. MS m/z (%): 250 (Mϩ, 5), 204 (100),
203 (46), 162 (17), 134 (18). IR (KBr, cmϪ1): 3433, 3145,
1
The following compounds were prepared according to
Method C with the variations stated below.
2935, 2559, 1628, 1566, 1409, 1197, 1177, 720. H NMR
(400 MHz, MeOD) d 6.59 (d, 1H, Jˆ3 Hz, CHy); 6.38 (s,
1H, isox-H); 5.98 (s, 1H, NCHSCy); 4.21–4.12 (m, 1H,
NCH2); 3.75–3.66 (m, 1H, NCH2); 3.45–3.33 (m, 2H,
NCH2); 2.90 (m, 1H, CH); 2.50 (s, 3H, SCH3); 2.32 (s,
3H, CH3); 2.25–2.11 (m, 2H, CH2); 2.07–1.92 (m, 2H,
CH2). Anal. (C13H18N2OS, C2H2O4): Calcd C, 52.93; H,
5.92; N, 8.23. Found C, 52.99; H, 5.95; N, 8.16.
6a and 7a using n-butyllithium. n-Butyllithium (2.50 M in
hexanes, 1.3 molar equivalents) was added at Ϫ78ЊC to a
solution of 3-methyl-5-trimethylsilanylmethyl isoxazole (1)
in THF. The reaction mixture was stirred for 30 min at
Ϫ78ЊC, before addition of (RS)-2-phenylsulphanyl-1-azabi-
cyclo[2.2.2]octan-3-one (3a). Gas chromatography of the
crude product showed, that the (Z/E) ratio was 80:20%.
(Z)-(RS)-3-(3-Methyl-isoxazol-5-ylmethylene)-2-benzyl-
1-azabicyclo[2.2.2]octane (6d) and (E)-(RS)-3-(3-methyl-
isoxazol-5-ylmethylene)-2-benzyl-1-azabicyclo[2.2.2]oc-
tane (7d). In 79% total yield starting from 3-methyl-5-
trimethylsilanylmethyl isoxazole (1) and (RS)-2-benzyl-1-
azabicyclo[2.2.2]octan-3-one (5d). Upon addition of
ketone, the reaction mixture was stirred at Ϫ78ЊC for
90 min, followed by 90 min at 20ЊC. Gas chroma-
tography of the crude product showed that the (Z/E)
ratio was 65:35%. Column chromatography (eluent:
ethyl acetate/methanol/ammonium hydroxide 25% in
water 94:6:0.5) gave 850 mg 6d and 440 mg 7d as
colourless oils, which were crystallised with oxalic
acid from acetone.
(Z)-(RS)-3-(3-Methyl-isoxazol-5-ylmethylene)-2-phenyl-
selanyl-1-azabicyclo[2.2.2]octane (6b) and (E)-(RS)-3-(3-
methyl-isoxazol-5-ylmethylene)-2-phenylselanyl-1-azabi-
cyclo[2.2.2]octane (7b). In 85% total yield starting from
3-methyl-5-trimethylsilanylmethyl isoxazole (1) and (RS)-
2-phenylselanyl-1-azabicyclo[2.2.2]octan-3-one (3b). 1H
NMR of the crude product showed, that the (Z/E) ratio
was 84:16%. Gas chromatography was not possible, as the
products decomposed in the oven. Column chromatography
(eluent: ethyl acetate) gave 46 mg 7b as a colourless oil,
141 mg 6b as a colourless oil, and 607 mg of a 6b/7b
mixture. 6b was crystallised as the hydrochloride salt, and
7b was crystallised as the oxalate salt.
6d. Mp 139–140ЊC. MS m/z (%): 294 (Mϩ, 39), 203 (100),
172 (25), 162 (14), 134 (12). IR (KBr, cmϪ1): 3438, 2961,
2579, 1743, 1651, 1570, 1194, 996, 701. 1H NMR
(400 MHz, MeOD) d 7.57 (d, 2H, Jˆ7 Hz, Hortho); 7.40 (t,
2H, Jˆ7 Hz, Hmeta) 7.32 (quartet, 1H, Jˆ7 Hz, Hpara); 6.59
(s, 1H, CHy); 6.34 (s, 1H, isox-H); 5.22 (m, 1H, NCHCy);
3.89–3.80 (m, 1H, NCH2); 3.49–3.21 (m, 5H,
NCH2ϩCH2Ph); 3.09 (dd, 1H, Jgemˆ18 Hz, Jvicˆ14 Hz,
CH2Ph); 2.95 (m, 1H, CH); 2.31 (s, 3H, CH3); 2.25–2.15
(m, 3H, CH2); 2.08–1.92 (m, 1H, CH2). Anal. (C19H22N2O,
1.25 C2H2O4): Calcd C, 63.46; H, 6.07; N, 6.88. Found C,
63.45; H, 6.39; N, 6.86.
6b. Mp 206–208ЊC. MS m/z (%): 360 (Mϩϩ1, 12), 279 (5),
203 (100), 162 (81), 148 (15), 134 (41). IR (KBr, cmϪ1):
3432, 2956, 2877, 2473, 1748, 1636, 1568, 1419, 1396, 999,
1
692. H NMR (300 MHz, CDCl3) d 7.75–7.64 (m, 2H,
Hmeta); 7.32–7.20 (m, 3H, HorthoϩHpara); 6.21 (d, 1H,
Jˆ3 Hz, CHy); 6.10 (s, 1H, isox-H); 5.62 (s, 1H, NCHSCy);
3.98–3.80 (m, 1H, NCH2); 3.32–3.13 (m, 1H, NCH2);
2.94–2.68 (m, 2H, NCH2); 2.52 (m, 1H, CH); 2.28 (s, 3H,
CH3); 1.90–1.61 (m, 4H, CH2). Anal. (C18H20N2OSe, HCl,
0.25 H2O): Calcd C, 54.01; H, 5.41; N, 7.00. Found C,
54.13; H, 5.41; N, 6.97.

J. E. Tønder et al. / Tetrahedron 56 (2000) 1139–1146
1145
7d. Mp 136–137ЊC. MS m/z (%): 294 (Mϩ, 39), 203 (100),
82 mg pure 7f as a colourless oil, which was then crystal-
lised with oxalic acid from acetone.
172 (25), 162 (14), 134 (12). IR (KBr, cmϪ1): 3433, 2959,
1
2569, 1720, 1638, 1572, 1417, 1311, 1207, 700. H NMR
(300 MHz, CDCl3) d 7.36–7.18 (m, 5H, Ph); 5.86 (s, 1H,
CHy); 5.82 (s, 1H, isox-H); 3.72 (t, 1H, Jˆ9 Hz, NCHCy);
3.45 (m, 1H, CH); 3.28–3.15 (m, 1H, NCH2); 3.12–2.90 (m,
3H, NCH2ϩCH2Ph); 2.86–2.72 (m, 2H, NCH2); 2.28 (s, 3H,
CH3); 1.86–1.62 (m, 4H, CH2). Anal. (C19H22N2O, C2H2O4,
0.5 C3H6O): Calcd C, 65.36; H, 6.58; N, 6.78. Found C,
65.49; H, 6.43; N, 6.43.
6f. Mp 169–171ЊC. MS m/z (%): 288 (Mϩ, 44), 259 (27),
217 (10), 204 (66), 176 (28). IR (KBr, cmϪ1): 3434, 2959,
2874, 2566, 1729, 1627, 1571, 1449, 1445, 1378, 1201, 996,
715, 701. 1H NMR (300 MHz, DMSO) d 6.42 (s, 1H, CHy);
6.33 (s, 1H, isox-H); 4.73 (m, 1H, NCHCy); 3.44–3.10 (m,
4H, NCH2); 2.79 (m, 1H, CH); 2.22 (s, 3H, CH3); 2.08–1.08
(m, 11H, CH2ϩR–CH2ϩR–CH); 0.85 (t, 3H, Jˆ7 Hz, R–
CH2); 0.80 (t, 3H, Jˆ7 Hz, R–CH2). Anal. (C18H28N2O,
C2H2O4): Calcd C, 63.47; H, 7.99; N, 7.40. Found C,
63.50; H, 8.05; N, 7.36.
(Z)-(RS)-3-(3-Methyl-isoxazol-5-ylmethylene)-2-(2,2-di-
methyl-propyl)-1-azabicyclo[2.2.2]octane (6e) and (E)-
(RS)-3-(3-Methyl-isoxazol-5-ylmethylene)-2-(2,2-dimethyl-
propyl)-1-azabicyclo[2.2.2]octane (7e). In 92% total yield
starting from 3-methyl-5-trimethylsilanylmethyl isoxazole
(1) and (RS)-2-(2,2-dimethylpropyl)-1-azabicyclo[2.2.2]-
octan-3-one (5e). Upon addition of ketone, the reaction
mixture was stirred at Ϫ78ЊC for 90 min, followed by
90 min at 20ЊC. Gas chromatography of the crude product
showed that the (Z/E) ratio was 80:20%. Flash chroma-
tography (eluent: ethyl acetate) gave 690 mg 6e, 58 mg 7e
and 513 mg of a 6e/7e mixture. 6e and 7e were crystallised
with oxalic acid from acetone to give white powders.
7f. Mp 131–133ЊC. MS m/z (%): 288 (Mϩ, 44), 259 (27),
217 (10), 204 (66), 176 (28). IR (KBr, cmϪ1): 3431, 2957,
1
2875, 1739, 1663, 1599, 1414, 1370, 1200, 700. H NMR
(300 MHz, CDCl3) d 6.03 (s, 1H, CHy); 5.89 (s, 1H, isox-
H); 3.46 (m, 1H, NCHCy); 3.41 (m, 1H, CH); 3.16–2.94
(m, 2H, NCH2); 2.89–2.70 (m, 2H, NCH2); 2.28 (s, 3H,
CH3); 1.81–1.19 (m, 11H, CH2ϩR–CH2ϩR–CH); 0.93–
0.83 (m, 6H, R–CH2). Anal. (C18H28N2O, C2H2O4): Calcd
C, 63.47; H, 7.99; N, 7.40. Found C, 63.55; H, 8.04; N, 7.30.
(Z)-3-(3-Methyl-isoxazol-5-yl)methylene-1-azabicyclo-
[2.2.2]octane (6g). SnBu3H (1.3 mL, 4.8 mmol) and a,a0-
azoisobutyronitrile (AIBN) (0.20 g, 1.2 mmol) was added
to a stirred solution of (Z)-(RS)-3-(3-methyl-isoxazol-5-
ylmethylene)-2-phenylsulphanyl-1-azabicyclo[2.2.2]octane
(6a) (0.76 g, 2.4 mmol) in dry toluene (40 mL) at 120ЊC.
Upon stirring for 15 min at 120ЊC, TLC (methanol/dichloro-
methane 1:9) showed that all starting material had reacted.
The reaction mixture was poured into 1 N HCl (50 mL),
whereupon the layers were separated. Extraction with
further 2×50 mL 1 N HCl gave an aqueous phase, which
was basified (K2CO3, pHϷ11) and extracted with
4×50 mL dichloromethane. The organic phase was dried
(MgSO4) and evaporated. Column chromatography (eluent:
dichloromethane/methanol/ammonium hydroxide 25% in
water 94:6:0.5) of the crude oil gave 61 mg (12%) of 6g
as a colourless oil. The spectral data of 6g were identical to
those previously published.1
6e. Mp 192–193ЊC. MS m/z (%): 274 (Mϩ, 66), 259 (39),
217 (85), 203 (24), 189 (48). IR (KBr, cmϪ1): 3431, 3047,
1
2954, 1713, 1695, 1649, 1570, 1415, 1190, 697. H NMR
(400 MHz, CDCl3) d 6.03 (s, 1H, CHy); 5.88 (s, 1H, isox-
H); 4.14 (m, 1H, NCHCy); 3.26–3.18 (m, 1H, NCH2);
3.18–2.99 (m, 1H, NCH2); 2.76–2.60 (m, 2H, NCH2);
2.42 (m, 1H, CH); 2.28 (s, 3H, CH3); 1.84–1.54 (m, 5H,
CH2ϩR–CH2); 1.46 (dd, 1H, Jˆ15 Hz, 2 Hz, R–CH2); 1.06
(s, 9H, R–CH3). Anal. (C17H26N2O, C2H2O4): Calcd C,
62.62; H, 7.74; N, 7.69. Found C, 62.75; H, 7.89; N,
7.65.
7e. Mp 164–165ЊC. MS m/z (%): 274 (Mϩ, 66), 259 (39),
217 (85), 203 (24), 189 (48). IR (KBr, cmϪ1): 3429, 3023,
1
2949, 2878, 1745, 1662, 1588, 1413, 1376, 1199, 698. H
NMR (400 MHz, CDCl3) d 6.03 (s, 1H, CHy); 5.88 (s, 1H,
isox-H); 3.91 (m, 1H, NCHCy); 3.37 (quintet, 1H, Jˆ7 Hz,
CH); 3.10–2.98 (m, 2H, NCH2); 2.92–2.82 (m, 1H, NCH2);
2.77–2.67 (m, 1H, NCH2); 2.28 (s, 3H, CH3); 1.79–1.57 (m,
5H, CH2ϩR–CH2); 1.46–1.39 (m, 1H, R–CH2); 1.00 (s,
9H, R–CH3). Anal. (C17H26N2O, C2H2O4, 0.5 C3H6O, 0.5
H2O): Calcd C, 61.87; H, 7.98; N, 7.04. Found C, 61.94; H,
8.15; N, 6.70.
References
1. Tønder, J. E.; Hansen, J. B.; Begtrup, M.; Pettersson, I.;
Rimvall, K.; Christensen, B.; Ehrbar, U.; Olesen, P. H. Improving
the Nicotinic Pharmacophore with a Series of (Isoxazole)Methy-
lene-1-Azacyclic Compounds: Synthesis, Structure–Activity
Relationship, and Molecular Modeling. J. Med. Chem. 1999, 42,
4970–4980.
2. Nesi, R.; Ricci, A.; Taddei, M.; Tedeschi, P. Group IVB Organo-
metallic Derivatives: Synthesis and 13C NMR Spectra of New Silyl
and Stannyl Isoxazoles. J. Organomet. Chem. 1980, 195, 275–283.
3. Ager, D. J. The Peterson Reaction. Synthesis 1984, 384–398.
4. Barrett, A. G. M.; Hill, J. M.; Wallace, E. M.; Flygare, J. A.
Recent Studies on the Peterson Olefination Reaction. Synlett 1991,
764–770.
(Z)-(RS)-3-(3-Methyl-isoxazol-5-ylmethylene)-2-(2-ethyl-
butyl)-1-azabicyclo[2.2.2]octane (6f) and (E)-(RS)-3-(3-
methyl-isoxazol-5-ylmethylene)-2-(2-ethyl-butyl)-1-azabi-
cyclo[2.2.2]octane (7f). In 83% total yield starting from
3-methyl-5-trimethylsilanylmethyl isoxazole (1) and (RS)-
2-(2-ethyl-butyl)-1-azabicyclo[2.2.2]octan-3-one (5f). Upon
addition of ketone, the reaction mixture was stirred at
Ϫ78ЊC for 90 min, followed by 90 min at 20ЊC. Gas chro-
matography of the crude product showed that the (Z/E) ratio
was 75:25%. The oily crude product was crystallised with
oxalic acid from acetone, and recrystallised in boiling ace-
tone to give 439 mg of 6f as white crystals. The remaining
mixture of 6f and 7f was purified by column chroma-
tography (eluent: methanol/dichloromethane 1:10) to give
5. Hudrlik, P. F.; Agwaramgbo, E. L. O.; Hudrlik, A. M. Concer-
ning the Mechanism of the Peterson Olefination Reaction. J. Org.
Chem. 1989, 54, 5613–5618.
6. Kawashima, T.; Okazaki, R. Synthesis and Reactions of the

1146
J. E. Tønder et al. / Tetrahedron 56 (2000) 1139–1146
Intermediates of the Wittig Peterson, and their Related Reactions.
Synlett 1996, 600–608.
7. Ager, D. J. The Peterson Olefination Reaction; In Organic
Reactions, Wiley: New York, 1990, pp 1–223.
8. Grigorieva, N. Y.; Pinsker, O. A.; Buevich, A. V.; Moiseenkov,
A. M. Z-Stereoselective Peterson Olefination of Ketones. Russ.
Chem. B 1995, 44, 492–499.
9. Grigorieva, N. Y.; Pinsker, O. A.; Moiseenkov, A. M. Stereo-
control in the Peterson Olefination of Ketones. Mendeleev
Commun. 1994, 129–130.
Reaction Conditions, and a New Model for the Addition of Carba-
nions to Carbonyl Derivatives in the Absence of Chelation Control.
J. Chem. Soc., Perkin Trans. 2 1986, 593–597.
13. Olesen, P. H.; Swedberg, M. D. B.; Eskesen, K.; Judge, M. E.;
Egebjerg, J.; Tønder, J. E.; Rasmussen, T.; Sheardown, M. J.;
Rimvall, K. Identification of Novel (Isoxazole)Methylene-1-
Azabicyclic Compounds With High Affinity for the Central Nico-
tinic Cholinergic Receptor. Bioorg. Med. Chem. Lett. 1997, 7,
1963–1968.
14. Stotter, P. L.; Friedman, M. D. Model Studies of a Concep-
tually New Approach to the Total Synthesis of Quinine. J. Org.
Chem. 1985, 50, 29–31.
15. Seebach, D.; Teschner, M. Herstellung (-Thiolierter Carbonyl-
verbindungen. Chem. Ber. 1976, 109, 1601–1616.
10. Bell, D.; Crowe, E. A.; Dixon, N. J.; Geen, G. R.; Mann, I. S.;
Shipton, M. R. Stereoselectivity in the Peterson Reaction—Appli-
cation to the Synthesis of BRL-49467. Tetrahedron 1994, 50,
6643–6652.
11. Bassindale, A. R.; Ellis, R. J.; Taylor, P. G. The Role of
Silicon–Oxygen Interactions in Determining the Stereochemical
Outcome of the Peterson Reaction. J. Chem. Res., Synop. 1996,
34–35.
16. Kolb, H. C.; Andersson, P. G.; Sharpless, K. B. Toward an
Understanding of the High Enantioselectivity in the Osmium-
Catalyzed Asymmetric Dihydroxylation (AD) 1. Kinetics. J. Am.
Chem. Soc. 1994, 116, 1278–1291.
12. Bassindale, A. R.; Ellis, R. J.; Lau, J. C. Y.; Taylor, P. G. The
Mechanism of the Peterson Reaction. Part 2. The Effect of
17. Leonard, J.; Lygo, B.; Procter, G. 79. Advanced Practical
Organic Chemistry; 2nd ed.; Blackie: London, 1995, p 2.