Addition of electrophiles to unsymmetric alkenes. Effects of β oxygen substituents on the stereo- and regio-chemistry of positive halogen addition

Article abstract of DOI:10.1039/a801001c

The stereo- and regio-chemistry of two step additions of positive halogen electrophiles to 5-methylene-2-phenyl-1,3-dioxane, 1-tert-butyl-4-methylenecyclohexane and 3-methoxy-2-(methoxymethyl)prop-1-ene have been determined. The direction of initial electrophile attack is in line with the frontier orbital and electrostatic considerations described by us previously. The regiochemistry of addition is strongly affected by hyperconjugative effects, acting between the intermediate epihalonium ion and the β C-X bonds. Where the β C-X bonds bear a fixed periplanar relation to the epihalonium ion and X is more electronegative than hydrogen, anti-Markownikoff addition is strongly promoted and becomes exclusive when two such β C-X bonds are present. If the β C-X bond is free to rotate away from periplanarity, then β C-H bonds will adopt the geometry required for hyperconjugation and Markownikoff regiochemistry will be favoured. The results are consistent with ab initio theoretical calculations and can be rationalised using a simple electrostatic model.

Full text of DOI:10.1039/a801001c

View Article Online / Journal Homepage / Table of Contents for this issue
Addition of electrophiles to unsymmetric alkenes. Effects of
â oxygen substituents on the stereo- and regio-chemistry of positive
halogen addition
(The late) John Hudeca and John W. Liebeschuetzb
a Department of Chemistry, University of Southampton, Southampton, UK SO17 1BJ
b Proteus Molecular Design Ltd., Beechfield House, Lyme Green Business Park, Macclesfield,
UK SK11 0JL
The stereo- and regio-chemistry of two step additions of positive halogen electrophiles to 5-methylene-
2-phenyl-1,3-dioxane, 1-tert-butyl-4-methylenecyclohexane and 3-methoxy-2-(methoxymethyl)prop-1-ene
have been determined. The direction of initial electrophile attack is in line with the frontier orbital and
electrostatic considerations described by us previously. The regiochemistry of addition is strongly
affected by hyperconjugative effects, acting between the intermediate epihalonium ion and the â C᎐X
bonds. Where the â C᎐X bonds bear a fixed periplanar relation to the epihalonium ion and X is more
electronegative than hydrogen, anti-Markownikoff addition is strongly promoted and becomes exclusive
when two such â C᎐X bonds are present. If the â C᎐X bond is free to rotate away from periplanarity, then
â C᎐H bonds will adopt the geometry required for hyperconjugation and Markownikoff regiochemistry
will be favoured. The results are consistent with ab initio theoretical calculations and can be rationalised
using a simple electrostatic model.
Our recent investigation into the stereochemistry of the add-
ition of electrophiles to 1,1-disubstituted alkenes such as the
5-methylenedioxane 1 and the methylenecyclohexane 2 has
Results
The dioxane 1 was prepared by the method of Weiss et al.3 as
described previously, via the intermediate 2,2-bis(hydroxy-
methyl)norbornene 4. Bis-methylation of 4, followed by flash
vacuum pyrolysis in a quartz tube cleanly afforded 3 (Scheme
1). The methylenecyclohexane 2 was prepared according to the
method of Corey and Suggs.4
OMe
O
But
Ph
O
OMe
1
2
3
(i)
(iii)
OMe
revealed that the hyperconjugative interaction of the correctly
aligned β C᎐O, C᎐C and C᎐H bonds with the π system has a
strong influence on the stereoreactivity of the system towards
those electrophiles which add in a one step process.1 We now
report on the stereo- and, in particular, the regio-chemistry of
additions to 1 and 2 of electrophiles which add in a two step
manner.
89%
(ii)
+
OH
OMe
OMe
OMe
OH
4
5
3
Scheme 1 Reagents and conditions: (i) NaH, THF, (ii) MeI, (iii) 500 ЊC,
1 torr
It is to be expected that the Markownikoff rule will be obeyed
in additions to the cyclohexane 2. The expected result for add-
ition to the dioxane, 1, is less clear cut however. The two oxy-
gens are, together, likely to exert a strong inductive effect which
may act to destabilise any positive charge on the tertiary carbon
of the double bond, during the formation and subsequent
attack of a cationic intermediate. Alternatively the oxygen lone
pairs could play some important role, perhaps by forming an
oxygen bridged cationic intermediate. A third factor, of particu-
lar interest to us, was whether the β C᎐O bonds had some influ-
ence, mediated by hyperconjugation, that could affect the
regiochemistry of addition. Our previous results had suggested
that such an influence might be strong.1 In addition theoretical
calculations carried out some time ago by Radom et al. had
predicted that a β C᎐O bond aligned with the p orbital of an
adjacent carbocation would be poorer at stabilizing the cation
hyperconjugatively than a β C᎐H or β C᎐C bond.2 However,
few experimental studies have unequivocally demonstrated this
to be the case in practice.
The reagents employed and the results of the addition reac-
tions are given in Tables 1 and 2. All the reagents chosen are
sources of positive halogen and the reaction, in each case, is
expected to proceed via an epihalonium ion. To ensure the
validity of the results, it was essential that alternative radical
mediated pathways be eliminated and also that the presence of
any acidic by-products be excluded to prevent epimerisation of
the dioxane ring in the products from 1. To this end, all the
reactions were carried out in the dark. Hypobromination and
hypoiodination were performed in the presence of a sodium
acetate or sodium carbonate buffer. Propylene oxide was
employed to remove hydrohalogen acid in the bromination and
iodochlorination reactions. Its use was essential in the reactions
of 1, for in its absence very different product ratios were
observed. Thus bromination gave both possible dibromides and
iodochlorination gave a mixture of all four possible chloro-
iodides, each of which was isolated. Additionally, the bromin-
ation was repeated in acetonitrile, a polar solvent, in the pres-
ence of a radical inhibitor (methylene blue) with the same
results as those observed in CCl4. The product mixtures were
separated by flash column chromatography on silica and the
individual components isolated and characterised. Product
ratios were generally determined for the crude product mixture
by measuring the integrated areas of characteristic peaks in the
Our primary aim was to identify any such hyperconjugative
effect. It was necessary therefore to prepare a model compound
analagous to 1 which would have no conformational restric-
tions on the geometry of the β substituents with respect to
the double bond. 3-Methoxy-2-(methoxymethyl)prop-1-ene 3,
was accordingly chosen.
J. Chem. Soc., Perkin Trans. 2, 1998
1139

View Article Online
Table 1 Products from addition of halohydrins to 5-methylene-2-phenyl-1,3-dioxane (1), 4-tert-butyl-1-methylenecyclohexane (2) and 3-methoxy-
2-(methoxymethyl)prop-1-ene (3)
Products
Markownikoff
(%)
Anti-
markownikoff (%)
Entry
1
Substrate
Reagent
Structure
exo
endo
exo
endo
1
N-Bromoacetamide, aq. ButOH, NaOAc
6 R = H
7 R = Ac
9 X = Y = Br
8 R = H
10 R = Ac
11 X = Y = I
14, 15 X = Br
14 X = Br
16, 17 X = I
18, 19
0
0
71
14
5
51
36
13
0
0
2
1
N-Iodoacetamide, aq. ButOH, NaOAc
0
0
0
0
3
4
5
2
2
2
N-Bromoacetamide, aq. ButOH,
92
>90
>90
8a
N-Bromoacetamide, aq. ButOH, NaOAc
N-Iodoacetamide, aq. ButOH, NaOAc
6
7
2
3
N-Iodoacetamide, aq. ButOH, Na2CO3
N-Bromoacetamide, aq. ButOH, NaOAc
16, 17 X = I
12, 13
84
16
0
0
76
24
a Data from ref. 6.
Table 2 Products from addition of halogens to 5-methylene-2-phenyl-1,3-dioxane (1), 4-tert-butyl-1-methylenecyclohexane (2) and 3-methoxy-2-
(methylenemethoxy)prop-1-ene (3)
Products
Markownikoff
(%)
Anti-
markownikoff (%)
Entry
Substrate
Reagent
Structure
endo
exo
exo
endo
1
2
3
4
1
1
1
1
Br2 in CCl4
9, 20 X = Y = Br
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
16
100
100
100
84
0
0
Br2 in CCl4 ϩ propylene oxide
Br2 in MeCN ϩ propylene oxide
Br2 in MeCN ϩ propylene oxide
ϩ methylene blue
0
5
6
7
8
9
1
1
2
2
2
ICl4 in CCl4
ICl in CCl4 ϩ propylene oxide
Br2 in CCl4
Br2 in CCl4 ϩ propylene oxide
ICl in CCl4 ϩ propylene oxide
21, 22, 23, 24 X = I, Y = Cl
27, 28 X = Y = Br
7
0
34
34
20
0
66
66
3
100
N/A
N/A
70
0
N/A
N/A
29 X = I, Y = Cl
30 X = Y = Cl
25, 26 X = I, Y = Cl
60
40
68
0
N/A
32
10
3
ICl in CCl4 ϩ propylene oxide
1H NMR spectrum (normally the peak due to the exocyclic
methylene group).
of the methylenecyclohexane 2 was similarly evident from the
1H and 13C NMR spectra and the stereochemistry of the prod-
ucts was established by conversion to the corresponding
epoxides, 35 and 36, which are already known.5 The result of
hypobromination is in agreement with that reported previ-
ously.6 Hypobromination of 3 gave both possible adducts,
whose regiochemistry was defined by examination of their 13C
NMR spectra.
The regiochemistry of hypobromination and hypoiodination
of the dioxane 1 could be ascertained from the 1H NMR spec-
trum of the single halohydrin isolated from each. A 6.5 Hz
coupling between the hydroxy H and the exocyclic methylene
hydrogens was clearly visible indicating that only the anti-
Markownikoff adduct had formed in each case. The stereo-
chemistry of the addition was determined by treating the
isolated halohydrin (6 or 8) with KOH in methanol. Both halo-
hydrins afforded exclusively the same epoxide 31 (Scheme 2),
The isolation of all the possible isomers from the addition of
bromine and iodine chloride to 1, in the absence of propylene
oxide, made the task of determining their stereo- and regio-
chemistry much easier. Comparison of the 13C NMR shifts for
the primary and tertiary iodine bearing carbons established the
regiochemistry of the iodochlorination products. Their stereo-
chemistry was assigned by making use of the results of Eliel
and Enanoza,7 who showed that for 5-substituted-1,3-dioxanes,
a 5-equatorial methyl(ene) resonates substantially upfield of a
5-axial methyl(ene) in the 1H NMR spectrum. This is a reversal
of the situation observed for methylcyclohexanes. To confirm
these assignments, samples of the pure compounds 9 and 23
were treated with acid. Epimerisation of the acetal group
occurred rapidly to give a mixture of stereoisomers in a ratio
similar to that obtained in the addition reactions carried out in
the absence of propylene oxide. These experiments showed that
the more thermodynamically stable epimer is that with the
5-methylene group axial rather than that with the 5-methylene
group equatorial, which is the isomer formed as the major
product in the addition reactions. For the case of 1, this
O
O
KOH in MeOH
HO
O
Ph
Ph
O
O
X
X = Br, I
31
Scheme 2
which had been characterised previously,1 indicating that the
initial electrophilic attack by positive halogen had occurred on
the exo (axial) face of the double bond. Both hypohalogen-
ations gave a number of minor products (7, 9, 10 and 11), which
result from capture of the intermediate epihalonium ion, on the
route to the halohydrin, by acetate or halide. Acetylation of the
halohydrins 6 and 8 gave rise to 7 and 10 respectively confirm-
ing their structures and stereochemistry.
The regiochemistry of hypobromination and hypoiodination
1140
J. Chem. Soc., Perkin Trans. 2, 1998

View Article Online
hexanes.10,11 In neither the bromination nor the iodochlorin-
ation additions to 2 and 3 were the product ratios affected when
propylene oxide was included in the reaction mixture.
O
O
O
Ph
Ph
Ph
O
OR
O
OR
O
Y
Br
I
X
6 R = H
7 R = Ac
9
X = Y =Br
8
R = H
11 X =Y = I
10 R = Ac
Discussion
23 X = I, Y = Cl
Our results indicate that whilst the methylenecyclohexane
2 undergoes exclusive Markownikoff addition as expected,
the methylenedioxane 1 shows, conversely, complete anti-
Markownikoff addition with positive halogen reagents. Fur-
ther, assuming that anti opening of the epihalonium ion occurs,
it is also clear that initial halogen attack on 1 must be over-
whelmingly from the axial face, i.e. exo. Similar exo attack is
also favoured for the hypohalogenation of the methylene-
cyclohexane 2, although here some product does arise as a
result of initial attack from the endo face. Molecular bromine
adds to 2 with a preference for the exo face of 2:1, whereas for
the methylenedioxane 1, initial addition is still exclusively exo.
(Based on the assumption that, as with the other reagents,
bromine adds with anti-Markownikoff regiochemistry.) In
comparing this aspect of the positive halogen additions to 1
and 2, we note that the behaviour is similar to that observed
with diimide, which reacts stereospecifically exo with 5-
methylenedioxanes, but almost unselectively towards methyl-
enecyclohexane.1 For the reaction of diimide, this was most
satisfactorily accounted for by invoking the pronounced differ-
ence in electrostatic potential field between the two faces of the
double bond which calculations had shown for 1. This differ-
ence gives a substantial bias for exo (axial) attack by an electro-
phile with an appreciable dipole, which, with the exception of
bromine, is the case for the electrophiles that we have examined
here. It has been argued previously that even uncharged but
polarisable electrophiles, such as molecular bromine, can be
directed by the electrostatic potential field because the field
assists separation of charge within the electrophile,12 or may act
to ‘guide in’ the electrophile.13
In considering the regiochemistry of addition of the unsym-
metrical reagents to the two cyclic compounds, we note the
remarkable propensity for the dioxane 1 to undergo anti-
Markownikoff addition. This cannot be explained by invok-
ing only an inductive effect, as otherwise 3-methoxy-2-
(methoxymethyl)prop-1-ene 3 should also show exclusive anti-
Markownikoff addition. However this molecule has a small
preference for Markownikoff addition. An alternative rational-
isation might be that described by de la Mare to account for
the slight preference for anti-Markownikoff addition of
hypochlorous acid to 3-chloro- and 3-bromo-propene.14 This
involves neighbouring group participation by the 3-halogen to
form the double epihalonium ion shown in Scheme 3, followed
by attack, preferred at the terminal sites, to afford 32 as the
major product. For the propenes, this explanation is supported
OH
Br
But
OH
MeO
MeO
12
Br
MeO
OH
MeO
X
13
14 X = Br
16 X = I
But
But
But
OAc
X
I
OH
OAc
I
15 X = Br
17 X = I
19
18
O
O
X
Y
O
Ph
O
O
Ph
Y
X
Ph
O
X
Y
20 X = Y = Br
24 X = I, Y = Cl
22 X = I, Y = Cl
21 X = I, Y = Cl
Cl
I
MeO
I
MeO
Cl
MeO
MeO
25
26
But
But
Br
O
Br
27
35
But
Y
X
29 X = I, Y = Cl
30 X = Y = Cl
But
But
Br
O
Br
28
36
result shows conclusively that the addition reactions are under
kinetic rather than thermodynamic control. We draw the infer-
ence that it is likely that this conclusion can be extrapolated to
the products formed in the halohydrin additions. Hence we
conclude that, under our experimental conditions, all the add-
itions to the dioxane 1 are under kinetic control. It seems likely
that this is also true for the other two substrates, 2 and 3. (The
thermodynamic preference for the 5-methylene group axial over
equatorial in dioxanes has been noted previously by Eliel and
Kaloustian.8)
X
X = Br,Cl
HOCl
X
+
The 13C NMR spectra enabled easy assignment of the regio-
chemistry of the iodochlorination products from the open
chain ether 3. For the methylenecyclohexane 2, iodochlorin-
ation gave two rather unstable products. The major one, 29, was
an iodochloride which could be assigned Markownikoff regio-
chemistry on the basis of its 13C NMR spectrum. The minor
isomer, 30, was a dichloride. The propensity for the formation
of dichlorides during ICl addition has been noted before.9
Unfortunately, no reliable assignment of stereochemistry could
be made for either of these two compounds. Bromination of 2
gave a mixture of two dibromides. Their stereochemistry was
determined by comparison of their 13C NMR and IR spectra
with those of previously reported bromomethylenecyclo-
Cl
X
Cl
OH
OH
X
X
Cl
OH
Cl
32
33
34
Scheme 3
J. Chem. Soc., Perkin Trans. 2, 1998
1141

View Article Online
by the formation of products in which the halogen initially at
C-3 is found at C-2. However, in the reactions of 1 and 3,
neighbouring group participation by β-oxygen is unlikely to be
occurring because, despite a careful search, in neither case
could any similar rearrangement products be found (Scheme 4).
+
+
O
O
Ph
Ph
O
O
X
X
C
D
H
+
H
OMe
H
OMe
NBA
H
+
MeO
MeO
Br
OH
X
MeO
aq. ButOH
H
OMe
OMe
OMe
X
H
3
E
F
Scheme 4
Scheme 6
The striking difference between the behaviour of the rigid
methylenedioxane 1 and the flexible 3-methoxy-2-(methoxy-
methyl)prop-1-ane 3 points strongly to an effect dependent
on the alignment of the β C᎐O bonds with the intermediate
epihalonium ion. We suggest, therefore, that a hyperconjugative
interaction between the β C᎐O bond and the epihalonium ion,
analogous to that discussed in our previous paper,1 is respon-
sible for this difference.
The situation may be compared to those examined in the
pioneering ab initio calculations of Radom and co-workers on
the stability of carbonium ions β-substituted with a substituent
X, where the stability of the carbocation depends both on the
nature of X and the conformation of the C᎐X bond relative to
the p-orbital on the cationic centre.2,15 The calculations show
that where the substituent X is more electropositive than hydro-
gen, the cation is best stabilised in the aligned conformation A
shown in Scheme 5, so that extensive σ–π electron donation can
occur. On the other hand, when X is more electronegative than
hydrogen, the most stable conformation is that shown in B
(Scheme 5) with the C᎐X bond in the plane perpendicular to the
Oδ–
X+
δ–
δ+
δ+
δ–
δ+
β
α
γ
Fig. 1 Induced dipoles in the epihalonium ion resulting from axial
attack on 1
interactions might affect nucleophilic attack on the epi-
halonium ion formed by initial addition to the exocyclic bond
of the dioxane 1. The Cβ᎐Cγ bond is close to antiperiplanar to
the O᎐Cα bond. Therefore the dipole along the O᎐Cα bond is
liable to induce a corresponding dipole along the Cβ᎐Cγ bond.
This is illustrated in Fig. 1. The arrows are intended to illustrate
this charge shift and not the absolute charge on each atom
which is likely to be positive for both Cβ and Cγ. The transfer of
charge from Cβ to Cγ will make Cγ more electropositive than Cβ
and hence liable to nucleophilic attack. This Cβ to Cγ charge
transfer, mediated by hyperconjugation, will occur most
strongly if the O᎐Cα bond is periplanar to the bonds of the
epihalonium ring. If the O᎐Cα bond holds no fixed dihedral
angle to the epihalonium ring, as in the intermediates from
addition to 3, then much less anti-Markownikoff addition
would be expected.
H
+
+
X
H
H
H
X
H
H
H
H
Conclusions
A
B
For single step electrophilic additions to C᎐C double-bonds,
we have already shown that the stereochemistry is profoundly
influenced by the hyperconjugative interactions between the π
system and a C᎐X bond situated β with respect to it.1 We have
here demonstrated that, for two step additions of positive halo-
gen electrophiles, these same hyperconjugative effects, acting
between an intermediate epihalonium ion and a β C᎐X bond,
have a very significant influence on the regiochemistry of the
process. Where X is more electronegative than hydrogen then,
provided that the β C᎐X bond bears a fixed periplanar relation
to the epihalonium ion, anti-Markownikoff addition is strongly
promoted to the extent that it becomes exclusive when two such
β C᎐X bonds are present. If the β C᎐X bond is free to rotate
away from periplanarity, then it will do so, allowing alternative
β C᎐H bonds to achieve the geometry required for hyper-
conjugation and hence favouring the Markownikoff regio-
chemistry. These results are consistent with ab initio theoretical
calculations14 and can be rationalised using a simple electro-
static model.
Scheme 5
p-orbital and in which the C᎐H bonds can now act to stabilise
the electron deficiency.
In addition to this entirely hyperconjugative effect, Radom
and his colleagues were also able to demonstrate that there is an
underlying inductive effect that, when X is more electronegative
than H, destabilises the cation regardless of which conform-
ation (A or B) it adopts. The inductive effect is still present in γ
substituted cations, albeit somewhat reduced.
Our regiochemical results are very much as would be pre-
dicted on the basis of Radom’s calculations if a simple carbo-
cation mechanism were assumed. The equilibrium between the
primary (D) and tertiary (C) cations formed by addition to the
methylenedioxane 1 might well favour the primary cation
(Scheme 6), for the tertiary one will be adversely affected by the
C᎐O bonds. The tertiary carbocation formed by addition to the
open chain 3 can, on the other hand, achieve the more stabilised
conformation E predicted by Radom.
The fact that any anti-Markownikoff addition occurs at all
with 3 could then be explained as a consequence of inductive
destabilisation of the tertiary cation E by the β oxygens, with a
resultant change in its energy to a level similar to that of the
primary cation F.
Experimental†
All solvents were dried and distilled before use. Flash column
chromatography was carried out using a hand pump as air pres-
However, none of the reactions discussed here are thought to
proceed via a simple carbocation and therefore the direct appli-
cation of the results of these calculations, though attractive, is
not clear cut. We would suggest an electrostatic picture as a
plausible alternative way of visualising how hyperconjugative
† The products have been named using E and Z terminology to describe
the positions of the two substituents relative to the plane of the dioxane
ring. Z indicates that the two substituents are located on the same side
of the ring and E indicates that they are on opposite faces of the ring.
1142
J. Chem. Soc., Perkin Trans. 2, 1998

View Article Online
sure source and MN Kieselgel 60 (230–400) as the solid phase.
Melting points were determined on a Gallenkamp solid block
melting point apparatus and are uncorrected. Elemental
analyses were performed at the microanalytical laboratory,
University College, London. IR spectra were recorded on a
Perkin-Elmer 298 Infrared Spectrophotometer, spectra were
recorded in solution cells of 0.1 mm thickness (unless otherwise
stated) in chloroform solvent. NMR spectra were recorded on a
Varian Associates XL-100-12 with internal lock.
δH(100 MHz, CDCl3) 1.17 (s, 3H), 4.088 (d, J 12, 2H), 4.257
(d, J 12, 2H), 4.40 (s, 2H), 5.53 (s, 1H), 7.48 (m, 5H).
(Z)-5-Bromo-5-bromomethyl-2-phenyl-1,3-dioxane, 9. Slow
crystallising material. ν/cmϪ1 2860, 1604, 1453, 1390, 1105, 980,
910; δH(100 MHz, CDCl3) 3.81(s, 2H), 4.24 (s, 4H), 5.53 (s, 1H),
7.43 (m, 5H); (C6D6) 3.30 (s, 2H), 3.643 (d, J 12.5, 2H), 3.84
(d, J 12.5, 2H), 7.125 (s, 1H), 7.19 (m, 3H), 7.6 (m, 2H); δC(25.18
MHz, CDCl3) 34.8, 59.4, 74.5, 101.6, 126.0, 128.5, 129.4, 137.0,
m/z 333–338 (Mϩ).
Preparation of 5,5-bis(methoxymethyl)bicyclo[2.2.1]hept-2-ene,
5
Treatment of (Z)-5-bromo-5-hydroxymethyl-2-phenyl-1,3-
dioxane, 6, with methanolic potassium hydroxide
A solution of 5,5-bis(hydroxymethyl)bicyclo[2.2.1]hept-2-ene,
4, (16 g), in tetrahydrofuran (100 ml) was added dropwise, with
stirring, to a slurry of sodium hydride (8.9 g, 60% in mineral oil,
prewashed with pentane) in tetrahydrofuran (50 ml), under
nitrogen. The mixture was heated to reflux for 60 min. and
allowed to cool. Iodomethane (13.86 ml) was added dropwise
and the mixture left overnight before being poured into brine.
The organic solvent was removed under reduced pressure
and the residue extracted with chloroform. The extracts were
combined and washed with water. After drying, filtration and
removal of solvent, an oil (17.7 g) was collected. Distillation
yielded 5 as a colourless oil (14.6 g, 78%), bp 104.5–105.5 ЊC
(22 mmHg). ν/cmϪ1 2970, 2880, 1570, 1430, 1333, 1197, 1110,
975, 717; δH(100 MHz, CDCl3) 0.75 (dd, J 12, 6, 1H), 1.5 (m,
3H), 2.62 (m, 1H), 2.78 (m, 1H), 3.05 (d, J 8, 1H), 3.20, (d, J 8,
1H), 3.26 (s, 3H), 3.36 (s, 3H), 3.35 (d, J 8, 1H), 3.5 (d, J 8, 1H),
6.15 (m, 2H).
Bromohydrin 6 (214 mg, 0.784 mmol) was dissolved in meth-
anol (2 ml) and a solution (0.82 ) of potassium hydroxide in
methanol (1.85 ml) was added. After 7 days the solvent was
removed under reduced pressure and the residue taken up
in diethyl ether. This was washed (H2O), dried and filtered.
Evaporation of the solvent afforded material from which
the exo-epoxide 311 could be isolated by preparative TLC
(silica, 90:10 CHCl3–EtOAc). No endo-epoxide could be
detected.
Addition of hypobromous acid to 1-tert-butyl-4-methylenecyclo-
hexane
1-tert-Butyl-4-methylenecyclohexane, 2, (168 mg, 1.1 mmol)
and N-bromoacetamide (0.69 g, 5 mmol) were added, with
stirring, to a mixture of sodium acetate (0.625 g, 7.6 mmol) in
30:70 water–tert-butyl alcohol. The reaction was left in the
dark for 48 h. The organic solvent was evaporated off under
reduced pressure and the residue extracted with diethyl ether.
The combined extracts were washed (1  NaOH, H2O), dried,
filtered and the solvent evaporated under reduced pressure
to afford primarily a single compound (200 mg) by NMR
spectroscopy. Preparative TLC (silica, CHCl3) yielded a colour-
less crystalline solid (130 mg, 47%).
(E)-4-tert-Butyl-1-bromomethylcyclohexan-1-ol, 14. Recry-
stallised from hexane, mp 71–71 ЊC. ν/cmϪ1 3570, 2950, 2870,
1600, 1450, 1365, 1050, 950; δH(100 MHz, CDCl3) 0.89 (s, 9H),
1.0–2.0 (m, 9H), 2.15 (br s, 1H, disappears on D2O shake),
3.33 (s, 2H).
Preparation of 2-methoxy-3-(methoxymethyl)prop-1-ene, 3
Vapourised 5,5-bis(methoxymethyl)bicyclo[2.2.1]hept-2-ene, 5,
(7 g) was passed through a quartz pyrolysis tube (50 × 1.5 cm),
packed with quartz glass fragments and wound with nichrome
resistance wire (5 Ω mϪ1), with a pitch of 4 mm, on top of
alumina cement (3 mm thick) heated to 500 ЊC, at a pressure of
1 mmHg. The collected pyrolysate was fractionally distilled
under nitrogen to afford 3 as a mobile oil (3.97 g, 89%), bp 132–
134 ЊC (760 mmHg). ν/cmϪ1 2930, 1658, 1450, 1191, 923; δH(100
MHz, CDCl3) 3.34 (s, 6H), 3.92 (d, J 1, 4H), 5.17 (s, 2H);
δC(25.18 MHz, CDCl3) 58.0, 73.4, 113.7, 143.
Addition of hypobromous acid to 3-methoxy-2-methoxymethyl-
prop-1-ene
Addition of hypobromous acid to 5-methylene-2-phenyl-1,3-
dioxane
3-Methoxy-2-methoxymethylprop-1-ene, 3, (780 mg, 6.7 mmol)
and N-bromoacetamide (2.74 g, 19.9 mmol) were added, with
stirring, to a mixture of sodium acetate (3.7 g, 45 mmol) in
30:70 water–tert-butyl alcohol. The reaction was left in the
dark for 1 h. The bulk of the organic solvent was evaporated off
under reduced pressure and the residue extracted with diethyl
ether. The combined extracts were washed (1  NaOH, H2O),
dried, filtered and the solvent was removed under reduced
pressure to afford an oil (0.90 g, 63%). Flash chromato-
graphy (silica, CH2Cl2–CHCl3) afforded the two predominant
constituents.
5-Methylene-2-phenyl-1,3-dioxane (1.05 g, 5.97 mmol) and N-
bromoacetamide (4.26 g, 28 mmol) were added, with stirring, to
a mixture of sodium acetate (3.75 g, 46 mmol) in 50:50 water–
tert-butyl alcohol. The reaction was left in the dark for 16 h,
after which the organic solvent was evaporated off under
reduced pressure and 1  sodium hydroxide added (20 ml).
The mixture was extracted with diethyl ether and the extracts
combined, washed, dried, filtered and the solvent evaporated
to afford a brown oil (1.35 g, 83%). The mixture was separated
by flash column chromatography on silica using a solvent
gradient from 50:50 toluene–CH2Cl2 to 80:20 CH2Cl2–
EtOAc. Three major components were identified: (Z)-5-
2-Bromomethyl-1,3-dimethoxypropan-2-ol, 12. ν/cmϪ1 3550,
2930, 1447, 1111, 972; δH(100 MHz, CDCl3) 2.90 (br s, 1H),
3.41 (s, 6H), 3.47 (s, 4H), 3.52 (s, 2H); δC(25.18 MHz, CDCl3)
36.2, 59.5, 73.1, 73.9; m/z (CI with isobutane) 213.0177,
215.0146 (calc. for C6H1379BrO3ϩ ϩ H: 213.0127; for
C6H1381BrO3ϩ ϩ H: 215.0106).
bromo-5-hydroxymethyl-2-phenyl-1,3-dioxane
(845
mg),
(Z)-5-acetoxymethyl-5-bromo-2-phenyl-1,3-dioxane (186 mg)
and (Z)-5-bromo-5-bromomethyl-2-phenyl-1,3-dioxane (69
mg).
(Z)-5-Bromo-5-hydroxmethyl-2-phenyl-1,3-dioxane, 6.
Recrystallised from cyclohexane, mp 95–96 ЊC. ν/cmϪ1 3620,
3580, 2840, 1600, 1450, 1380, 1140, 1090, 905; δH(100 MHz,
CDCl3) 2.25 (t, J 6.5, 1H, disappears on D2O shake), 3.78
(d, J 6.5, 2H), 4.03 (d, J 12, 2H), 4.29 (d, J 12, 2H), 5.51 (s, 1H),
7.48 (m, 5H); δC(25.18 MHz, CDCl3) 64.5, 65.9, 73.4, 101.4,
126.2, 128.2, 129.2, 137.1; m/z 274.009 (calc. for C11H13O379Brϩ:
274.003).
2-Bromo-3-methoxy-2-methoxymethylpropan-1-ol, 13. ν/cmϪ1
3300 (br), 2920, 1110; δH(100 MHz, CDCl3) 2.90 (1H, br s),
3.40 (8H, s), 3.73 (2H, s); δC(25.18 MHz, CDCl3) 59.5, 66.9,
69.4, 75.3; m/z (CI with isobutane) 213.0125 (calc. for
C6H1379BrO3ϩ ϩ H: 213.0127).
Addition of hypoiodous acid to 5-methylene-2-phenyl-1,3-
dioxane
(Z)-5-Acetoxymethyl-5-bromo-2-phenyl-1,3-dioxane, 7. Slow
5-Methylene-2-phenyl-1,3-dioxane, 1, (862 mg, 4.9 mmol) and
N-iodosuccinimide (3.38 g, 21.3 mmol) were added, with stir-
crystallising material. ν/cmϪ1 2850, 2760, 1750, 1605, 1455;
J. Chem. Soc., Perkin Trans. 2, 1998
1143

View Article Online
ring, to a mixture of sodium acetate (3.64 g, 44 mmol) in 50:50
water–tert-butyl alcohol (100 ml). The reaction was left in the
dark for 5 days. The organic solvent was evaporated off under
reduced pressure in a foil enclosed flask. The aqueous residue
was saturated with potassium carbonate and extracted with
diethyl ether. The combined extracts were washed (0.2 
Na2S2O3, H2O), dried, filtered and the solvent evaporated under
reduced pressure; to afford a brown oil (934 mg). This was sep-
arated by flash column chromatography, (CH2Cl2–CHCl–
EtOAc gradient); into three components: (Z)-5-iodo-5-iodo-
methyl-2-phenyl-1,3-dioxane (93 mg), (Z)-5-acetoxymethyl-
5-iodo-2-phenyl-1,3-dioxane (268 mg) and (Z)-5-iodo-5-
hydroxymethyl-2-phenyl-1,3-dioxane (377 mg).
(Z)-5-Iodo-5-hydroxymethyl-2-phenyl-1,3-dioxane, 8. Recry-
stallised as colourless needles from cyclohexane. ν/cmϪ1 3560,
2870, 1455, 1000, 700; δH(100 MHz, CDCl3) 2.25 (t, J 7, 1H,
disappears on D2O shake), 3.72 (d, J 7, 2H), 3.75 (d, J 13,
2H), 4.29 (d, J 13, 2H), 5.48 (s, 1H), 7.45 (m, 5H); δC(25.18
MHz, CDCl3) 52.3, 68.0, 75.9, 102.1, 126.6, 128.5, 129.4, 137.3;
m/z 320 (Mϩ).
3.50 (s, 2H); δC(25.18 MHz, CDCl3) 23.8, 24.5, 27.6, 32.2, 37.5,
47.5, 69.9; m/z 278 (Mϩ Ϫ 17).
(Z)-1-Acetoxy-4-tert-butyl-1-iodomethylcyclohexane,
18.
ν/cmϪ1 2960, 1731, 1370, 1254, 1022; δH(60 MHz, CDCl3) 0.8 (s,
9H), 1.0–2.0 (m, 7H), 2.0 (s, 3H), 2.5 (m, 2H), 3.9 (s, 2H).
Treatment of crude hypoiodous acid adduct of 1-tert-butyl-4-
methylenecyclohexane with methanolic sodium hydroxide
Crude product from the previous reaction (55 mg) was dis-
solved in methanol (6 ml) containing sodium hydroxide (60 mg)
and the mixture stirred for 3 h. Solvent was evaporated off
and the residue taken up in diethyl ether. This was washed,
dried, filtered and the diethyl ether removed to yield an oil
(20 mg, 70%) consisting of exo-and endo-epoxides 35 and 36
in the ratio 84:16. The spectral data were identical with those
previously reported;4 δH(100 MHz, CDCl3) of epoxide mixture
0.9 (s, 9H), 1.2–2.2 (m, 9H), 2.40 (s, 1.6H), 2.46 (s, 0.4H).
Repetition of this process with pure iodohydrin 16 afforded
solely epoxide 35.
(Z)-5-Acetoxymethyl-5-iodo-2-phenyl-1,3-dioxane, 10. ν/cmϪ1
3000, 1710, 1600, 1270, 1040; δH(60 MHz, CDCl3) 3.9 (d, J 13,
2H), 4.3 (d, J 13, 2H), 4.45 (s, 2H), 5.5 (s, 1H), 7.5 (m, 5H).
(Z)-5-Iodo-5-iodomethyl-2-phenyl-1,3-dioxane, 11. Obtained
as an oil that coloured violet on standing, too unstable to char-
acterize with certainty; δH(60 MHz, CDCl3) 4.2 (s, 2H), 4.5
(s, 4H), 5.6 (s, 1H), 7.4 (s, 5H).
Bromination of model compounds: general procedure
A solution of bromine (0.95 equiv. Br2 w.r.t. olefin) in carbon
tetrachloride, or benzene, was gradually added to a stirred
solution of the alkene in the dark under nitrogen, with or with-
out co-reagents. When the solution had almost completely de-
colourised, it was washed (sat. NaHCO3, Na2S2O3, H2O), dried,
filtered and the solvent evaporated under reduced pressure to
give crude product mixture.
Treatment of crude hypoiodous acid adduct of 5-methylene-2-
phenyl-1,3-dioxane with methanolic potassium hydroxide
Crude product from the previous reaction (174 mg) was dis-
solved in methanol (5 ml). A solution of methanolic potassium
hydroxide was added (8 ml, 0.27 ) and the mixture stirred for 2
days. The solvent was evaporated off and the residue taken up
in diethyl ether. The diethyl ether extracts were washed (H2O),
dried, filtered and the solvent removed under reduced pressure
to afford an oil (92 mg, 96%) that consisted of two components,
separable by preparative TLC. One was exo-epoxide 31, the
other had spectral properties consistent with it being the ring
opened product of methoxide attack on 31. No endo-epoxide
was found.
Bromination of 5-methylene-2-phenyl-1,3-dioxane in carbon
tetrachloride without propylene oxide
Compound 1 (1.035 g, 5.9 mmol), and bromine [19.1 ml of
0.273  soln. in CCl4, (5.214 mmol)] were reacted together as
above in carbon tetrachloride (50 ml). The reaction took 1 h
and, after work up, yielded a brown oil (1.783 g, 90%). Separ-
ation using flash column chromatography (silica, toluene–
CHCl3) gave two isomeric fractions; the major: (E)-5-bromo-
5-bromomethyl-2-phenyl-1,3-dioxane, 20, and the minor: (Z)-
5-bromo-5-bromomethyl-2-phenyl-1,3-dioxane, 9, previously
described. Small quantities of decomposition products, includ-
ing benzaldehyde, were also obtained.
(E)-5-Bromo-5-bromomethyl-2-phenyl-1,3-dioxane, 20.
Recrystallised from cyclohexane–hexane, mp 42.5–43.5 ЊC.
(Found: C, 39.1; H, 3.55; Br, 47.7. C11H12Br2O2 requires: C,
39.3; H, 3.60; Br, 47.6%); ν/cmϪ1 2870, 1600, 1450, 1390, 1100,
1020, 980, 930; δH(100 MHz, CDCl3) 4.165 (2H, s), 4.18 (d,
J 10.25, 2H), 4.40 (d, J 10.25, 2H), 5.46 (s, 1H), 7.58 (m, 5H);
δC(25.18 MHz, CDCl3) 38.7, 57.2, 74.1, 102.2, 125.9, 128.3,
129.3, 136.7.
Addition of hypoiodous acid to 1-tert-butyl-4-methylenecyclo-
hexane
1-tert-Butyl-4-methylenecyclohexane, 2 (0.50 g, 3.25 mmol),
and sodium bicarbonate (3.125 g) were dissolved in 50:50
water–tert-butyl alcohol. N-iodosuccinimide (3.75 g, 16.7
mmol) was added and the mixture stirred for 15 min. The bulk
of the organic solvent was evaporated off under reduced pres-
sure in a foil enclosed flask. The residue was taken up in diethyl
ether and washed (0.2  Na2S2O3, H2O) and the washings back
extracted. The diethyl ether extracts were combined and the
solvent removed under reduced pressure to yield an oil (608 mg,
63%). This was separated by flash column chromatography [sil-
ica, 10:90 diethyl ether–light petroleum (bp 40–60 ЊC)] into two
major fractions. The first fraction (219 mg) largely consisted of
epoxides formed in situ by base catalysed ring closure of the
initially formed iodohydrins. The second fraction (152 mg)
was identified as iodohydrin 16, (E)-4-tert-butyl-1-iodomethyl-
cyclohexan-1-ol. Unidentified by-products (95 mg) were also
isolated.
The reaction was repeated using sodium acetate instead of
sodium bicarbonate as buffer. This led to the isolation of two
major fractions, one 16, the other iodoacetate (Z)-1-acetoxy-4-
tert-butyl-1-iodomethylcyclohexane, 18.
(E)-4-tert-Butyl-1-iodomethylcyclohexan-1-ol, 16. Recrystal-
lised from hexane, mp 99.5–100.5 ЊC (dec) (Found: C, 45.7; H,
7.35; I, 41.7. C11H21IO requires: C, 44.6; H, 7.15; I, 42.8%);
ν/cmϪ1 3560, 2910, 2870, 1455, 1368, 1190, 1038; δH(100 MHz,
CDCl3) 0.88 (s, 9H), 1.3–1.9 (m, 6H), 1.96 (m, 3H), 2.64 (s, 1H),
Bromination of 5-methylene-2-phenyl-1,3-dioxane in carbon
tetrachloride with propylene oxide
Compound 1 (107 mg, 0.61 mmol), propylene oxide (53 mg)
and bromine were reacted together as above in carbon tetra-
chloride (35 ml). Work up afforded the Z dibromide, 9 (162 mg,
79%) only.
Bromination of 5-methylene-2-phenyl-1,3-dioxane in acetonitrile
with propylene oxide
Compound 1 (110 mg, 0.62 mmol), propylene oxide (53 mg)
and bromine [Br2 soln. in benzene (0.6 ml, 0.95 equiv.)] were
reacted together in acetonitrile (2 ml). Work up afforded an oil
(160 mg) that was estimated, by 1H NMR spectroscopy, to con-
tain 9 (68%), unreacted starting material (18%) and unidentified
material (14%) (possibly Ritter type adducts).
Bromination of 5-methylene-2-phenyl-1,3-dioxane in acetonitrile
with propylene oxide and methylene blue
Compound 1 (103 mg, 0.585 mmol), propylene oxide (50 mg),
1144
J. Chem. Soc., Perkin Trans. 2, 1998

View Article Online
ϩ
:
methylene blue (10 mg) and bromine (0.95 equiv.) were reacted
together in acetonitrile (2 ml) for 3 h. Evaporation of the
solvent afforded an oil that was examined by 1H NMR
spectroscopy. Only one dibromide, 9, was identified as being
present, along with unreacted starting material.
(calc. for C11H1237ClIO2ϩ: 339.95, and for C11H1235ClIO2
337.96).
Iodochlorination of 5-methylene-2-phenyl-1,3-dioxane in
carbon tetrachloride with propylene oxide
Compound 1 (115 mg, 0.65 mmol) and iodine monochloride
(0.77 mmol) were reacted together in carbon tetrachloride (10
ml) with propylene oxide added (20 mg) and gave iodochloride
23 only (>91%). No decomposition products were noted.
Bromination of 1-tert-butyl-4-methylenecyclohexane
Compound 2 (0.9 g, 6.2 mmol) and bromine (20.1 ml of a 0.273
 soln. in CCl4) were reacted together as above in carbon tetra-
chloride (50 ml). After 18 h, the reaction was worked up to
afford an oil (1.9 g) which was separated by flash column
chromatography [silica, light petroleum (bp 60–80 ЊC)] to afford
the dibromides (Z)- and (E)-1-bromo-1-bromomethyl-4-tert-
butylcyclohexane 27 and 28. Repetition of the experiment with
propylene oxide (50 mg) present made no difference to the final
isomer ratio.
Iodochlorination of 1-tert-butyl-4-methylenecyclohexane
Compound 2 (0.337 g, 2.19 mmol) and iodine monochloride
[2.19 mmol in CCl4 (11 ml)] were reacted together for 10 min.
Work up yielded an oil (548 mg, 80%). The two predominant
products were separated using flash column chromatography to
afford two unstable oils identified as 1-chloro-1-iodomethyl-4-
tert-butylcyclohexane, 29, and 1-chloro-1-chloromethyl-4-tert-
butylcyclohexane, 30. Repetition of the experiment with the
addition of propylene oxide (75 mg) gave an identical yield and
product ratio.
(Z)-1-Bromo-1-bromomethyl-4-tert-butylcyclohexane,
27.
Colourless oil. ν/cmϪ1 2980, 1440, 1366, 1222, 1091, 845, 788,
668, 626, 603; δH(100 MHz, CDCl3) 0.93 (s, 9H), 1.5–2.0 (m,
9H), 3.94 (s, 2H); δC(25.18 MHz, CDCl3) 23.8, 27.5, 32.4, 38.4,
44.4, 46.4, 72.5.
1-Chloro-1-iodomethyl-4-tert-butylcyclohexane, 29. ν/cmϪ1
2860, 1480, 1374, 1098, 1017, 888; δH(100 MHz, CDCl3) 0.89
(s, 9H), 1.2 (m, 5H), 1.7 (m, 2H), 2.3 (m, 2H), 3.70 (s, 2H);
δC(25.18 MHz, CDCl3) 17.5, 24.8, 27.5, 32.2, 40.7, 46.9, 70.0;
m/z 187 (Mϩ Ϫ I).
(E)-1-Bromo-1-bromomethyl-4-tert-butylcyclohexane,
28.
Colourless oil. ν/cmϪ1 2980, 1450, 1365, 1010, 895, 685; δH(100
MHz, CDCl3) 0.88 (s, 9H), 1.0–2.6 (m, 9H), 3.90 (s, 2H);
δC(25.18 MHz, CDCl3) 25.2, 27.4, 32.3, 41.2, 41.9, 46.8, 67.4;
m/z 311 (Mϩ Ϫ H), 231 (Mϩ Ϫ Br).
1-Chloro-1-chloromethyl-4-tert-butylcyclohexane, 30. ν/cmϪ1
2960, 1448, 1373, 1098, 983, 690, 618; δH(100 MHz, CDCl3)
0.89 (s, 9H), 1.2 (m, 5H), 1.5 (m, 4H), 1.9 (m, 4H), 4.175 (s, 2H);
δC(25.18 MHz, CDCl3) 27.2, 27.5, 32.3, 39.7, 46.9, 57.4, 62.0.
Iodochlorination of model compounds: general procedure
A solution of iodine monochloride in carbon tetrachloride (1.0
equivs. ICl w.r.t. olefin) was added to a stirred solution of the
unsaturated substrate in the dark under nitrogen, with or with-
out co-reagents. The solution generally changed colour from
brown to violet over the space of 1 h. It was washed (0.2 
Na2S2O3, H2O), dried, filtered and the solvent evaporated under
reduced pressure to give crude product mixture.
Iodochlorination of 3-methoxy-2-methoxymethylprop-1-ene
Iodochlorination of 3 (0.31 g, 2.58 mmol) with ICl (0.95 equiv.)
in CCl4 (20 ml) for 80 min yielded, after work up, an oil (600
mg, 84%). The two predominant components were separated by
flash column chromatography (silica, CH2Cl2–CCl4) to afford 2-
chloro-2-iodomethyl-1,3-dimethoxypropane, 25, and 2-chloro-
methyl-1,3-dimethoxy-2-iodopropane, 26, as colourless oils.
Repetition of the experiment with propylene oxide (150 mg)
added made no difference to the final isomer ratio.
Iodochlorination of 5-methylene-2-phenyl-1,3-dioxane in carbon
tetrachloride without propylene oxide
2-Chloro-2-iodomethyl-1,3-dimethoxypropane, 25. ν/cmϪ1
2995, 2925, 2895, 2720, 1452, 1111; δH(100 MHz, CDCl3) 3.45
(s, 6H), 3.64 (s, 2H), 3.68 (s, 4H); δC(25.18 MHz, CDCl3) 10.7,
59.5, 69.6, 75.7; m/z (CI with isobutane) 278.968 (calc. for
C6H1235ClIO3ϩ: 278.965).
1 (1.00 g, 5.68 mmol) and iodine monochloride (1 equiv.) were
reacted together in carbon tetrachloride (70 ml) as above for
40 min yielding an oily material (1.7 g, 88%). Flash column
chromatography [silica, light petroleum (bp 60–80 ЊC)–diethyl
ether] afforded four fractions. The first fraction (1.00 g of an oil)
was identified as a mixture of the adducts (E)-5-chloro-5-
iodomethyl-2-phenyl-1,3-dioxane, 22, and (E)-5-chloromethyl-
5-iodo-2-phenyl-1,3-dioxane, 24. The second fraction (0.2 g)
consisted largely of benzaldehyde. The third (40 mg of
yellow crystals) and the fourth (87 mg) were identified as
(Z)-5-chloromethyl-5-iodo-2-phenyl-1,3-dioxane, 23 and (Z)-5-
chloro-5-iodomethyl-2-phenyl-1,3-dioxane, 21.
2-Chloromethyl-1,3-dimethoxy-2-iodopropane, 26. ν/cmϪ1
3000, 2930, 2895, 2735, 1467, 1450, 1114, 700; δH(100 MHz,
CDCl3) 3.46 (s, 6H), 3.60 (s, 2H), 3.99 (s, 2H); δC(25.18
MHz, CDCl3) 49.5, 52.1, 59.3, 76.4; m/z (CI with NH3) 279
(Mϩ Ϫ 1).
Acknowledgements
(Z)-5-Chloro-5-iodomethyl-2-phenyl-1,3-dioxane, 21. δH(100
MHz, CDCl3) 3.46 (s, 2H), 4.17 (s, 4H), 5.48 (s, 1H), 7.4
(m, 5H); δC(25.18 MHz, CDCl3) 8.3, 62.6, 75.2, 101.6, 126.4,
128.5, 129.4, 136.9; m/z 337–340 (Mϩ).
I am heavily indebted to Dr Ian D. R. Stevens of the Depart-
ment of Chemistry, The University of Southampton, for
the major part he played in the preparation of this manuscript
and for the helpful guidance he was able to give. This he did
gladly for the sake of the memory of my supervisor, his good
friend and greatly missed colleague Dr John Hudec. Very
sadly Ian died suddenly shortly before this manuscript went
to press. We thank the SERC for the award of a maintenance
grant (J. W. L.).
(E)-5-Chloro-5-iodomethyl-2-phenyl-1,3-dioxane, 22. δH(100
MHz, CDCl3) 3.87 (s, 2H), 3.91 (d, J 10.5, 2H), 4.26 (d, J 10.5,
2H), 5.32 (s, 1H), 7.3 (m, 5H); δC(25.18 MHz, CDCl3)
13.8, 60.4, 74.1, 101.8, 125.7, 128.1, 129.0, 136.5; m/z 337–340
(Mϩ).
(Z)-5-Chloromethyl-5-iodo-2-phenyl-1,3-dioxane, 23. ν/cmϪ1
3015, 2865, 1462, 1271, 1173, 1110, 995, 706; δH(100 MHz,
CDCl3) 4.05 (d, J 12, 2H), 4.09 (s, 2H), 4.24 (d, J 12, 2H), 5.52
References
(s, 1H), 7.4 (m, 5H); δC(25.18 MHz, CDCl3) 45.3, 49.4, 75.8,
101.9, 128.5, 129.4, 137.1; m/z 337.97 (calc. for C11H1235ClIO2
ϩ
:
1 J. Hudec, J. Huke and J. W. Liebeschuetz, J. Chem. Soc., Perkin
Trans.2, 1998, preceeding paper.
2 L. Radom, J. A. Pople and P. v. R. Schleyer, J. Am. Chem. Soc, 1972,
94, 5935.
3 F. Weiss, A. Isard and R. Bensa, Bull. Soc. Chim. Fr., 1965,
1355.
4 E. J. Corey and J. W. Suggs, Tetrahedron Lett., 1975, 44, 3775.
337.96).
(E)-5-Chloromethyl-5-iodo-2-phenyl-1,3-dioxane, 24. δH(100
MHz, CDCl3) 4.12 (s, 2H), 4.25 (d, J 11, 2H), 4.38 (d, J 11,
2H), 5.43 (s, 1H), 7.3 (m, 5H); δC(25.18 MHz, CDCl3) 44.2,
50.8, 75.0, 101.8, 125.7, 128.1, 129.0, 136.7; m/z 339.95, 337.96
J. Chem. Soc., Perkin Trans. 2, 1998
1145

View Article Online
5 A. Sevin and J.-M. Cense, Bull. Soc. Chim. Fr., 1974, 963.
6 R. Carson and R. Ardon, J. Org. Chem, 1971, 36, 216.
7 E. L. Eliel and R. M. Enanoza, J. Am. Chem. Soc, 1972, 94,
8072.
8 E. L. Eliel and M. K. Kaloustian, J. Chem. Soc., Chem. Commun.,
1970, 290.
13 Y.-D. Wu, Y. Li, J. Na and K. N. Houk, J. Org. Chem, 1993, 58,
4625.
14 P. B. D. de la Mare, Electrophilic Halogenation, Cambridge
University Press, Cambridge, 1976.
15 R. Hoffmann, L. Radom, J. A. Pople, P. v. R. Schleyer, W. J. Hehre
and L. Salem, J. Am. Chem. Soc, 1972, 94, 6221.
9 G. H. Schmid and J. W. Gordon, Can. J. Chem., 1984, 62, 2526.
10 J. F. King and M. J. Coppen, Can. J. Chem., 1971, 49, 3714.
11 P. Klaboe, J. J. Lothe and K. Lunde, Acta Chem. Scand., 1956, 10,
1465.
12 G. Bellucci, G. Berti, R. Bianchini, G. Ingrosso and E. Mastrorilli,
J. Org. Chem., 1978, 43, 422.
Paper 8/01001C
Received 4th February 1998
Accepted 13th February 1998
1146
J. Chem. Soc., Perkin Trans. 2, 1998