Radical-chain reductive alkylation of electron-rich alkenes mediated by silanes in the presence of thiols as polarity-reversal catalysts

Article abstract of DOI:10.1039/a903961i

In the presence of a thiol catalyst, triphenylsilane mediates the reductive alkylation of electron-rich terminal alkenes R¹R²C=CH₂ by organic halides R³Hal via electrophilic carbon-centred radicals R^3?

Full text of DOI:10.1039/a903961i

View Article Online / Journal Homepage / Table of Contents for this issue
Radical-chain reductive alkylation of electron-rich alkenes
mediated by silanes in the presence of thiols as polarity-reversal
catalysts
Hai-Shan Dang,a Mark R. J. Elsegood,†b Kyoung-Mahn Kima and Brian P. Roberts*a
a Christopher Ingold Laboratories, Department of Chemistry, University College London,
20 Gordon Street, London UK WC1H 0AJ
b Department of Chemistry, University of Newcastle-upon-Tyne, Newcastle-upon-Tyne,
UK NE1 7RU
Received 18th May 1999, Accepted 10th June 1999
In the presence of a thiol catalyst, triphenylsilane mediates the reductive alkylation of electron-rich terminal alkenes
1
2
3
3
ؒ
R R C᎐CH by organic halides R Hal via electrophilic carbon-centred radicals R . Reactions were carried out in
᎐
2
benzene or dioxane solvent using di-tert-butyl hyponitrite (at 60 ЊC) or dilauroyl peroxide (at 80 ЊC) as initiators and
good yields of the adducts R1R2CHCH2R3 were obtained with either methyl thioglycolate or triphenylsilanethiol as
catalysts (5–10 mol% based on alkene). In the presence of the thiol, the slow direct abstraction of hydrogen from
the silane by the nucleophilic adduct radical R1R2CCH R3 is replaced by a cycle of more rapid polarity-matched
˙
2
reactions in which hydrogen-atom transfer to the adduct radical from the thiol is followed by abstraction of hydrogen
from the silane by the derived thiyl radical, to regenerate the catalyst. In the absence of thiol, negligible yields of
reductive alkylation products were obtained. The homochiral thiols, 1-thio-β--mannopyranose tetraacetate and
1-thio-β--glucopyranose tetraacetate, and the tetrapivalate and tetrabenzoate analogues of the latter were effective
catalysts and reductive carboxyalkylation products with enantiomeric excesses up to 72% were obtained from
prochiral alkenes. Homochiral samples of two of these adducts were obtained by recrystallisation and their
absolute configurations were determined by X-ray diffraction.
In recent years, free-radical chemistry has provided a number
of useful methods for the selective formation of carbon–carbon
bonds.1 The key step in these reactions generally involves the
inter- or intra-molecular addition of a carbon-centred radical
to a multiply-bonded carbon acceptor, often followed by atom-
or group-transfer to the resulting adduct radical to give the final
product and a new radical which goes on to propagate a chain
process. The reductive alkylation of electron-poor alkenes,
using alkyl halides or pseudohalides (RX) in the presence
of tributyltin hydride, constitutes an important example
of such a C–C bond-forming reaction.2 The propagation
stage of this radical-chain process is shown in Scheme 1
arising from the toxicity of organotin compounds and the
difficulty of completely removing tin residues from the final
product.4 Furthermore, because of the high rate at which alkyl
radicals abstract hydrogen from the tin hydride, it is also usually
necessary to keep the concentration of the latter low relative to
the concentration of the alkene, in order to suppress premature
ؒ
trapping of R by the tin hydride to give RH in competition
ؒ
with addition of R to the alkene to produce 1.
We have shown that, in conjunction with a thiol catalyst,
simple trialkyl- or triaryl-silanes can serve as effective replace-
ments for trialkyltin hydrides for the reduction of alkyl halides,
sulfides and xanthates to hydrocarbons.5 In the absence of
thiol catalyst, the direct abstraction of electron-rich hydrogen
from silicon by a nucleophilic alkyl radical [reaction (1)] is rel-
EWG
Bu3Sn•
RX
R
R•
+
+
R′3SiH
XSH
RH
+
+
R′3Si•
XS•
(1)
(2)
(3)
R•
RH
Bu3SnH
Bu3SnX
XS•
+
R′3SiH
XSH
+
R′3Si•
EWG
R•
atively sluggish, because of adverse polar effects, and is usually
too slow to maintain the chain. In the presence of thiol, the
direct hydrogen transfer is replaced by the catalytic cycle as
shown in reactions (2) and (3), both of which benefit from
favourable polar effects and are relatively rapid, because the
thiyl radical is electrophilic and the sulfhydryl hydrogen atom
is electron-poor.ठThe thiol acts as a protic polarity-reversal
catalyst for reaction (1).6 Similarly, because of catalysis of
reactions of type (1), radical-chain hydrosilylation of alkenes
by simple silanes is also catalysed by thiols.8,9
R
EWG
1
Scheme 1
(EWG = electron-withdrawing group) and both addition of a
nucleophilic alkyl radical to the alkene and abstraction of
hydrogen from the tin hydride by the electrophilic adduct rad-
ical 1 are facilitated by the favourable polar effects which
operate in the respective transition states.3
In common with other reactions that are mediated by trialk-
yltin hydrides, this reductive alkylation suffers from drawbacks
‡ It should be noted that factors other than polar effects must be con-
sidered in order to understand the relative rates of hydrogen-atom
transfer to alkyl radicals from silanes and from thiols.6,7
† Correspondence concerning the X-ray crystallography should be
directed to this author.
§ For discussion of the reversibility of reaction (3), see refs. 6 and 7.
J. Chem. Soc., Perkin Trans. 1, 1999, 2061–2068
2061

View Article Online
Table 1 Reductive alkylation of alkenes using organic halides in the
We reasoned that favourable polar effects should also make
the silane–thiol couple an effective mediator of the reductive
alkylation of an electron-rich alkene by an organic halide RHal
which gives rise to a relatively electrophilic radical R . The
proposed propagation cycle is shown in Scheme 2 (EDG =
presence of triphenylsilane, catalysed by thiol and initiated by TBHNa
in dioxane at 60 ЊC
ؒ
Adduct yield
Entry
Alkene
R3Hal
Thiola
Adduct
(%)b
RHal
R′3SiHal
1
2
3
4
5
6
7
8
9
10
11
12
13
14d
15d
16d
3a
3a
3a
3a
3a
3a
3a
3b
3b
3c
3d
3e
3f
4a
4a
4b
4b
4c
4d
4e
4a
4b
4a
4a
4a
4a
4a
4a
4g
MTG
TPST
MTG
TPST
TPST
TPST
MTG
TPST
TPST
TPST
TPST
TPST
TPST
MTG
TPST
MTG
5aa
5aa
5ab
5ab
5ac
5ad
5ae
5ba
5bb
5ca
5da
5ea
5fa
5aa
5aa
5af
78
88c
75
72
78
83
76
86
75
85
78
60
63
89
92
48
R′3Si •
R•
EDG
XSH
R′3SiH
EDG
XS•
R
XSH
2
EDG
R
3a
3a
3a
Scheme 2
a Each 5 mol% based on alkene. b Isolated yields based on alkene. c The
yield was similar in benzene solvent. Only a trace of adduct was formed
in the absence of thiol. d The reaction was carried out in benzene solv-
ent under gentle reflux, using dilauroyl peroxide (5 mol%) as initiator.
electron-donating group). Addition of the electrophilic carbon
radical to the electron-rich alkene will be rapid and will
necessarily yield a nucleophilic adduct radical 2, which in turn
rapidly abstracts the electron-deficient hydrogen from sulfur in
the thiol catalyst. Such a chain process is complementary to the
tin hydride-mediated reaction shown in Scheme 1, which is par-
ticularly suited to the reductive alkylation of electron-deficient
alkenes. A preliminary report of the realisation of reactions
based on Scheme 2 has appeared10 and, in the present paper, we
give a full account of this work and its extension to enantio-
selective reductive alkylation catalysed by homochiral thiols.
(5)] evidently proceeds by the radical-chain mechanism shown
in Scheme 2.
A good yield of the adduct 5ab was obtained when the chlo-
romalonate was replaced by methyl bromoacetate, using either
MTG or TPST as catalyst (entries 3 and 4), but with methyl
chloroacetate under the same conditions the yield of 5ab was
reduced to 40% and a large amount of the silane adduct 6 was
Results and discussion
MeO2C
MeO2C
OAc
Initial reactions were carried out at 60 ЊC and were initiated by
thermal decomposition of di-tert-butyl hyponitrite (TBHN,
t1/2 = ca. 55 min),11,12 which produces tert-butoxyl radicals [eqn.
(4)] that go on to abstract hydrogen from the silane and/or the
Ph3Si
O
O
O
O
6
7
8
Ph3Si
ButON=NOBut
2 ButO•
N2
(4)
O
O
9
R1
R1
Ph3SiH
–Ph3SiHal
R3Hal
(5)
R3
also isolated. Evidently, the triphenylsilyl radical adds to
isopropenyl acetate and abstracts the halogen from the chloro-
acetate at comparable rates, whilst with the more reactive
bromoacetate, halogen abstraction becomes the major pathway.
Reductive carboxyalkylation of isopropenyl acetate using
diethyl 2-bromo-2-methylmalonate afforded the adduct 5ac in
good yield (entry 5), showing that the more sterically-hindered
R2
R2
3a R1 = OAc, R2 = Me
b R1 = OAc, R2 = But
4a (MeO2C)2CHCl
b MeO2CCH2Br
5
c R1 = OSiMe2But, R2 = Me c (EtO2C)2CMeBr
d R1 = OBu, R2 = H
d 1-AdO2CCH2Br
e 1-AdC(O)CH2Br
f PhSO2CH2Br
e R1 = Pentyl, R2 = Me
f R1 = CH2OAc, R2 = Me
and less electrophilic radical MeC(CO Et) adds effectively to
˙
2
2
the alkene under the reaction conditions.13
thiol to afford chain-carrying silyl or thiyl radicals. When a
dioxane solution containing isopropenyl acetate 3a (2.50
mmol), triphenylsilane (3.25 mmol), dimethyl chloromalonate
4a (3.75 mmol) and TBHN (0.125 mmol) was heated under
argon for 2 h, examination of the reaction mixture by 1H NMR
spectroscopy showed that <1% of the adduct 5aa¶ had been
formed. However, when the experiment was repeated in the
presence of methyl thioglycolate (MeO2CCH2SH, MTG, 0.125
mmol, 5 mol% based on alkene) under otherwise identical con-
ditions, the adduct 5aa was isolated in 78% yield. A somewhat
higher yield was obtained in the presence of triphenylsi-
lanethiol (TPST, 5 mol%) as catalyst (Table 1, entries 1 and 2)
and this thiol was used as the achiral catalyst in most sub-
sequent experiments. Thus, the reductive alkylation of 3a [eqn.
Triphenylbromosilane is formed as a by-product in all reac-
tions that involve organic bromides as the source of radical
addenda. This bromosilane is a moderately strong Lewis acid,
and is also readily hydrolysed to give HBr, and thus care must
be exercised when working with compounds that are sensitive to
such acids. Similar, though less severe problems could arise
from the presence of triphenylchlorosilane. Precautions should
be taken to exclude moisture during the reactions and the
standard work-up procedure involves neutralisation of the
reaction mixture with aqueous sodium bicarbonate at an early
stage.
Similar reductive alkylation reactions were carried out with
the acyclic terminal alkenes 3b–f and the results are summarised
in Table 1 (entries 8–13); in no case was a significant amount
of product formed in the absence of thiol catalyst. In order
to make the procedure more convenient, reactions were also
carried out at ca. 80 ЊC in gently-refluxing benzene solvent,
¶ The adduct 5aa arises from 3a and 4a, the adduct 5ab arises from 3a
and 4b and so on.
2062
J. Chem. Soc., Perkin Trans. 1, 1999, 2061–2068

View Article Online
using commercially available dilauroyl peroxide (DLP, 5 mol%
based on alkene) as initiator. The yield of 5aa obtained from
isopropenyl acetate and dimethyl chloromalonate was similar to
that obtained at 60 ЊC using TBHN as initiator (entries 14 and
15). Reductive sulfonylmethylation using PhSO2CH2Br proved
somewhat problematic and major amounts of the simple reduc-
tion product, methyl phenyl sulfone, were obtained under the
usual reaction conditions. However, a moderate yield of the
adduct 5af was obtained when the thiol concentration was kept
low by slowly adding a solution containing MTG and DLP (5
mol% of each) to the reaction mixture (entry 16).
chiral at the radical centre and the thiol is homochiral, then this
step should be enantioselective. We have shown previously9e
that thiol-catalysed hydrosilylation can similarly be rendered
enantioselective by the use of homochiral thiols as catalysts
and, for example, the silane adducts 12a and b have been
obtained in moderate to high enantiomeric purity by this route.
For the present work, the carbohydrate-derived thiols 15–18
AcO
AcO
OAc
O
O
AcO
AcO
AcO
AcO
SH
SH
SH
SH
Tris(trimethylsilyl)silane (TTMSS) is a much more reactive
hydrogen atom donor than simple trialkyl- or triaryl-silanes
and has been used successfully as an effective replacement for
tributyltin hydride in reductive alkylation reactions and in
several other types of radical-chain processes.4,14 It was thus
important to compare the efficiency of TTMSS with that of the
triphenylsilane–thiol couple used in the present work. When the
reaction between isopropenyl acetate and dimethyl chloroma-
lonate was carried out under the conditions of the first entry in
Table 1, except that the triphenylsilane was replaced by
TTMSS, the reductive alkylation product 5aa was formed in
OAc
15
16
PivO
BzO
O
O
PivO
PivO
BzO
BzO
OPiv
OBz
17
18
were chosen as homochiral catalysts, but it seems likely that
improved asymmetric induction could be obtained using more
specifically designed thiol catalysts. A selection of reductive
alkylation reactions of 10 and 13 was carried out using DLP as
initiator in refluxing benzene and the results are summarised
in Table 2; 10 mol% of thiol and initiator were used in these
experiments. Chemical yields were generally good, in some
cases better than those of the racemic products obtained using
MTG or TPST as catalysts under the same conditions, and the
enantiomeric excess (ee) was determined either by chiral-
stationary-phase HPLC or by 1H NMR spectroscopy using
(ϩ)-tris[3-(heptafluoropropylhydroxymethylene)camphorato]-
europium() [Eu(hfc)3] as a homochiral shift reagent. The
enantiomeric purities of the adducts 11ai and 11bg could be
upgraded by recrystallisation and the absolute configurations
of the homochiral compounds were determined by X-ray
diffraction; the structures are shown in Figs. 1 and 2.
1
31% yield, as judged by H NMR spectroscopy. Under these
conditions, dimethyl malonate was also formed in 55% yield, as
a result of relatively rapid trapping by the silane of the electro-
philic radical (MeO C) CH prior to its addition to the alkene.
˙
2
2
Reductive alkylation of the methylenelactones 7 and 10a and
b proceeded smoothly at 60 ЊC (TBHN initiator) or at 80 ЊC
(DLP initiator), using either MTG or TPST (5–10 mol%) as
catalyst, although the latter usually gave somewhat higher
yields. Again, essentially no products were obtained in the
absence of thiol. The silane adduct 9 was obtained as a by-
product (15% yield) in the reaction of 7 with dimethyl chloro-
malonate, as a result of trapping of the triphenylsilyl radical by
the alkene in competition with the abstraction of chlorine. Simi-
larly, minor and variable amounts of 12a and b were obtained
from reactions of the methylenelactones 10. The isolated yields
of racemic adducts 8 and 11 are given in the Experimental
section.
Both reductive alkylation products have the S-configuration
at C-6 (Scheme 3). The adduct 12a, obtained by thiol-promoted
Reductive alkylation of the unsaturated cyclic carbonate 13
to give 14a and b proceeded smoothly at 80 ЊC (DLP initiator)
Me
Me
Ph
Ph
R
R
EtO2C
EtO2C
EtO2C
R
R
But
Ph3SiH
(6)
R3Hal
R3
O
O
O
O
–Ph3SiHal
O
O
O
O
O
(S)-11ai
(S)-11bg
Me
Me
Ph3Si
10a R = Me
b R = Ph
4a (MeO2C)2CHCl
b MeO2CCH2Br
d 1-AdO2CCH2Br
e 1-AdC(O)CH2Br
f PhSO2CH2Br
g ButC(O)CH2Br
h (ButO2C)2CHCl
i (EtO2C)3CCl
11
O
O
(R)-12a
X*SH
approach B
j ButO2CCH2Br
Small
O
Small
O
R
R
O R3
O
Medium
R3
Me
Me
Me
Me
Large
R
R
Medium
R
Large
O
O
R
R3
approach A
Ph3Si
O
O
O
O
O
O
X*SH
14a R3 = (MeO2C)2CH
b R3 = PhSO2CH2
Scheme 3
12a R = Me
b R = Ph
13
addition of triphenylsilane to the methylenelactone 10a, has
been shown to have the R-configuration at C-6, when either 15
or 16 was used as catalyst.9e Thus, in all three cases the products
are formed by selective transfer of hydrogen from the β--
pyranose thiols to topologically-similar faces of the adduct rad-
ical intermediate, as shown in Scheme 3. The adduct radical
should exist as a rapidly-equilibrating pair of enantiomers and,
using TPST as catalyst, showing that this sensitive functionality
can tolerate the presence of Ph3SiBr under the reaction
conditions.
The final product-forming step in the thiol-catalysed reduc-
tive alkylation process (Scheme 2) is hydrogen-atom transfer
from the thiol to the adduct radical 2 and, if the latter is pro-
J. Chem. Soc., Perkin Trans. 1, 1999, 2061–2068
2063

View Article Online
Table 2 Enantioselective reductive alkylation of alkenes mediated by triphenylsilane in the presence of homochiral thiol catalysts initiated by DLPa
in benzene at ca. 80 ЊC
Product and isolated
yield (%)b
R3Hal
Thiola
Product ee (%)
[α]D21
1 c
Entry
Alkene
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
10a
10a
10a
10a
10a
10b
10b
10b
10b
10b
10b
10b
10b
10b
10b
10b
10b
10b
13
4a
4b
4f
15
15
15
15
16
15
16
15
16
15
15
15
15
16
17
18
15
15
15
11aa (62)
11ab (85)
11af (72)
11ai (67)
11ai (65)
11ba (90)
11ba (92)
11bb (75)
11bb (80)
11bd (57)
11be (52)
11bf (58)
11bg (70)
11bg (74)
11bg (71)
11bg (75)
11bh (70)
11bj (70)
14a (74)
28
19
7
Ϫ14.6 (2.24)
Ϫ18.5 (1.74)
Ϫ3.5 (2.54)
—
Ϫ26.9 (1.71)
Ϫ117.0 (2.15)
Ϫ112.5 (2.23)
Ϫ95.4 (1.70)
—
Ϫ130.5 (1.65)
Ϫ98.4 (2.24)
Ϫ115.2 (1.78)
Ϫ143.6 (1.62)
Ϫ141.0 (1.42)
Ϫ138.1 (1.75)
Ϫ159.6 (1.56)
Ϫ83.2 (1.45)
Ϫ117.5 (1.74)
Ϫ18.5 (2.84)
4i
46
72
54
52
41
45
52
41
53
54
53
52
60
33
50
8
4i
4a
4a
4b
4b
4d
4e
4f
4g
4g
4g
4g
4h
4j
4a
a Each 10 mol% based on alkene. b Isolated yields based on alkene. c CHCl3 solvent, c (g/100 cm3) shown in parentheses
Fig.
1
Structure of (S)-(Ϫ)-6-[2,2,2-tris(ethoxycarbonyl)ethyl]-5,5-
dimethyltetrahydropyran-2-one, (S)-11ai, determined by X-ray crystal-
lography. Selected geometrical parameters (bond lengths in Å, bond
angles in degrees): C2–C3 1.533(4), C3–C4 1.481(4), C4–C5 1.496(4),
C5–O1 1.350(3), C1–O1 1.465(3), C1–C2 1.528(3), C5–O2 1.200(3),
C1–C8 1.526(3), C2–C7 1.526(3), C2–C6 1.528(4), C2–C1–O1
111.72(16), C2–C1–C8 114.79(18), C8–C1–O1 104.30(17), C1–O1–C5
121.15(18), C4-C5–O1 119.2(2).
Fig.
2
Structure of (S)-(Ϫ)-6-(4,4-dimethyl-3-oxopentyl)-5,5-
diphenyltetrahydropyran-2-one, (S)-11bg, determined by X-ray crystal-
lography. Selected geometrical parameters (bond lengths in Å, bond
angles in degrees): C1–C2 1.530(2), C2–C3 1.521(3), C3–C4 1.496(3),
C4–O1 1.347(3), C5–O1 1.460(2), C1–C5 1.548(2), C4–O2 1.192(3),
C5–C6 1.527(3), C1–C13 1.530(2), C1–C19 1.542(3), C1–C5–O1
112.38(13), C1–C5–C6 115.45(15), C6–C5–O1 106.97(14), C5–O1–C4
123.63(15), O1–C4–C3 119.52(17).
assuming that approach of the thiol takes place preferentially in
a direction anti to the substituent R3, the diastereoisomeric
transition state resulting from approach A is evidently of lower
energy than that resulting from approach B for hydrogen-atom
transfer from the homochiral catalysts 15 and 16. Approach A
involves attack of the thiol at the face where the substituents at
the radical centre are arranged in the clockwise order large,
medium and small.
Comparison of the pairs of entries 1 and 6, 2 and 8, and 3
and 12 shows that the ees of adducts obtained from 10b are
appreciably greater than those obtained from 10a, presumably
as a result of the increased bulk of the large substituent in the
intermediate radicals derived from the former alkene. For
reductive carboxyalkylation of 10a using triethyl chlorometh-
anetricarboxylate 4i, the ee obtained using the mannopyranose
thiol 16 as catalyst was appreciably higher than that obtained
using the glucopyranose thiol 15 (entries 4 and 5). Unfortun-
ately, the 5,5-diphenyl analogue 10b failed to undergo reductive
carboxyalkylation with 4i. In other reductive alkylation reac-
tions of 10b, no significant differences in the ees of the products
were found between the four thiol catalysts 15–18 (entries 6 and
7, 8 and 9, and 13 and 16). Reductive carboxyalkylation of 10b
using dimethyl chloromalonate gave a product of higher ee than
that obtained from di-tert-butyl chloromalonate (entries 6 and
17), perhaps because the larger bulk of the medium-sized sub-
stituent in the intermediate adduct radical derived from the
latter makes this group more similar in size to the large substitu-
ent, thus reducing the steric chirality at the radical centre. In
contrast, the product derived from tert-butyl bromoacetate
showed a somewhat higher ee than that obtained from methyl
bromoacetate (entries 8 and 18).
Enantioselective reductive carboxylation of 10a could also be
mediated by tributyltin hydride. Thus, when a dioxane solution
containing the tin hydride (1.3 equiv.) and TBHN (5 mol%
based on alkene) was added slowly during 2 h to a dioxane
solution containing the methylenelactone 10a (1 equiv.), dime-
thyl chloromalonate (1.3 equiv.), the thiol 15 (1 mol%) and
TBHN (1 mol%) at 60 ЊC, followed by further heating at 60 ЊC
2064
J. Chem. Soc., Perkin Trans. 1, 1999, 2061–2068

View Article Online
for 1 h, the adduct 11aa was isolated in 74% yield and showed
an ee of 26%. The adduct with the same ee was obtained in 68%
yield when the experiment was repeated using triphenylsilane in
a one-pot reaction in which all the silane was present initially. In
the Bu3SnH-mediated reaction, the chemical yield and adduct
ee both drop if the tin hydride is added rapidly, because the
latter then becomes a competitive achiral donor of hydrogen to
the adduct radical 2, as well as prematurely trapping the elec-
trophilic malonyl radical. When the reaction was repeated with
slow addition of TTMSS in place of the tin hydride, the adduct
11aa was isolated in 67% yield and showed an ee of 26%.
When all the TTMSS was added at the start of the reaction, the
yield of adduct was very low and dimethyl malonate was the
major product. These results demonstrate that thiols can act
as polarity-reversal catalysts for the abstraction of hydrogen by
nucleophilic radicals from both tributyltin hydride and from
TTMSS. Similarly, the thiols 15 and 16 have been shown to
catalyse the enantioselective hydrosilylation of 10a by TTMSS,
confirming that these thiols are acting here as hydrogen donors
to the silyl-radical adduct of the alkene, under the conditions
used.9e
We conclude that the triphenylsilane–thiol couple is an effect-
ive mediator of the reductive alkylation of electron-rich alkenes
by halides that yield electrophilic carbon radicals; slow addition
techniques are not necessary because this silane is a poorer
hydrogen-atom donor, towards the electrophilic radical
addenda involved, than tributyltin hydride or tris(trimethylsi-
lyl)silane. Enantioselective reductive alkylation of the prochiral
alkenes, mediated by triphenylsilane in the presence of homo-
chiral thiol catalysts, gave products of moderate enantiomeric
purity. However, the SH groups in the four carbohydrate-
derived thiols 15–18 are not in particularly chiral environments
and there would appear to be considerable scope for the
rational design of homochiral thiols that would be more dis-
criminating donors of hydrogen to the different enantiotopic
faces of the intermediate prochiral adduct radical. Further
work is clearly needed to identify structural factors in both the
thiol catalyst and the prochiral adduct radical that lead to high
enantioselectivity in the hydrogen-atom transfer reaction.
chloride, using the method described by Mendenhall.12 Triphe-
nylsilanethiol, methyl thioglycolate and 2,3,4,6-tetra-O-acetyl-
1-sulfanyl-β--glucopyranose were obtained commercially
(Aldrich) and were used as received. 2,3,4,6-Tetra-O-acetyl-1-
sulfanyl-β--mannopyranose 16 and 2,3,4,6-tetra-O-pivaloyl-
1-sulfanyl-β--glucopyranose 17 were prepared as described
previously.9e The thiol 18 was prepared by the same general
procedure
starting
from
2,3,4,6-tetra-O-benzoyl-α--
glucopyranosyl bromide.15 The enol acetate 3b,16 the silyl enol
ether 3c,17 the methylenelactones 7,18 10a9e and 10b9e,19 and the
cyclic carbonate 1320 were prepared according to published
methods, as was triethyl chloromethanetricarboxylate 4i.21
Admantyl bromoacetate 4d. Bromoacetyl bromide (6.7 g, 33
mmol) was added dropwise to a stirred solution of admantan-
1-ol (5.0 g, 33 mmol) and N,N-dimethylaniline (4.4 g, 36 mmol)
in dry diethyl ether (25 cm3), with cooling in an ice–water bath.
After the addition was complete, the mixture was stirred at
room temperature for 4 h and the precipitated hydrobromide
salt was removed by filtration. The filtrate was diluted with
diethyl ether (50 cm3), washed with 5% aqueous HCl (3 × 5
cm3), then with saturated brine and dried over MgSO4. The
solvent was removed under reduced pressure and the residue
was purified by chromatography on silica gel, using light
petroleum–diethyl ether (10:1) as eluent, to give the product
(8.5 g, 94%) as an oil. Recrystallisation from light petroleum
gave 4d as a solid, mp 35 ЊC; δH 1.66 (6 H, br s, 3 CH2 of Ad),
2.11 (6 H, br s, 3 CH2 of Ad), 2.18 (3 H, br s, 3 CH of Ad), 3.74
(2 H, s, CH2Br); δC 27.7, 30.8, 36.0, 41.0, 82.9, 165.8; νmax (mull)
1736, 1273, 1105, 1054, 968 (Found: C, 52.5; H, 6.1. C12H17BrO2
requires C, 52.76; H, 6.27%).
Di-tert-butyl chloromalonate 4h. A solution of butyllithium
in hexane (1.6 mol dmϪ3, 27.0 cm3, 43 mmol) was added drop-
wise to a stirred solution of di-tert-butyl malonate (8.7 g, 40
mmol) in dry tetrahydrofuran (80 cm3) cooled in a solid CO2–
ethanol bath. The resulting mixture was stirred at Ϫ78 ЊC for 30
min and then liquid chlorine (ca. 1.5 cm3, ca. 60 mmol) was
allowed to evaporate and the gas was carried, in a slow stream
of nitrogen, into the stirred solution. The mixture was allowed
to warm to room temperature during 1 h and then stirred for a
further 1 h. The solvents were removed by evaporation under
reduced pressure and the residue was diluted with diethyl ether
(100 cm3), washed with 5% aqueous NaHCO3 (15 cm3) and
dried over MgSO4. The ether was removed under reduced pres-
sure and the residue was distilled to give 4h (4.5 g, 55%) as a
clear oil, bp 59–60 ЊC/0.1 mm Hg; δH 1.49 (18 H, 6 Me), 4.65
(1 H, CHCl); δC 27.7, 44.3, 83.9, 163.5; νmax (film) 2982, 1754,
1368, 1140, 844 (Found: C, 52.8; H, 7.5. C11H19ClO4 requires C,
52.70; H, 7.64%).
Experimental
NMR spectra were recorded using a Varian VXR-400 instru-
1
ment (400 MHz for H). The solvent was CDCl3 and chemical
shifts are reported relative to Me4Si; J values are quoted in Hz.
Infrared spectra were recorded for liquid films or Nujol mulls
using a Perkin-Elmer 1600 series FTIR spectrometer; the units
are cmϪ1. Column chromatography and TLC were carried out
using Merck Kieselgel 60 (230–400 mesh) and Kieselgel 60 F254
aluminium-backed pre-coated plates, respectively. Determin-
ation of enantiomeric excess by HPLC was carried out using a
Chiralcel-OD column (4.6 mm × 250 mm; Daicel Chemical
Industries Ltd.) in conjunction with hexane–isopropyl alcohol
eluent (flow rate 1 cm3 minϪ1). The proportion of alcohol in the
mobile phase and retention times of the two enantiomers are
given in the text; UV detection was at 254 nm (unless stated
otherwise) and in all cases the major enantiomer was eluted first.
Determination of enantiomeric excess by 1H NMR analysis
was carried out using (ϩ)-tris[3-(heptafluoropropylhydroxy-
methylene)camphorato]europium() [Eu(hfc)3] as shift re-
agent. Optical rotations were measured on an AA Series Polaar
2000 polarimeter (Optical Activity Ltd.) using a 1 dm cell and
Other materials were commercially available and were used
as received, except bromomethyl phenyl sulfone 4f (Aldrich)
which was purified by column chromatography (diethyl ether–
light petroleum 2:1 eluent), followed by recrystallisation from
light petroleum–diethyl ether.
Representative procedure for thiol-catalysed reductive alkylation
mediated by triphenylsilane
A solution in dry dioxane (4 cm3) containing isopropenyl acet-
ate (3a, 0.250 g, 2.50 mmol), triphenylsilane (0.846 g, 3.25
mmol), dimethyl chloromalonate (4a, 0.625 g, 3.75 mmol),
TBHN (22 mg) and triphenylsilanethiol (37 mg) was stirred and
heated at 60 ЊC for 2 h under an atmosphere of dry argon. The
solvent was removed by evaporation under reduced pressure,
the residue was dissolved in diethyl ether (10 cm3), and the solu-
tion was washed with 5 % aqueous NaHCO3 (2 × 10 cm3), then
with saturated brine (10 cm3) and dried (MgSO4). After evapor-
ation of the ether, light petroleum (5 cm3) was added and the
slurry was filtered to remove most of the triphenylsilanol, which
are given in units of 10Ϫ1 deg cm2 gϪ1
.
All manipulations of air-sensitive substances were carried
out under an atmosphere of dry argon or nitrogen. Light
petroleum refers to the fraction of distillation range 40–60 ЊC.
Dioxane and benzene were dried by heating under reflux over
calcium hydride and were distilled and stored under argon.
TBHN was prepared by the reaction of sodium hyponitrite
with tert-butyl bromide in diethyl ether, in the presence of zinc
J. Chem. Soc., Perkin Trans. 1, 1999, 2061–2068
2065

View Article Online
was washed on the sinter with a little light petroleum. After
removal of the solvent from the filtrate, the residue was purified
by column chromatography (light petroleum–diethyl ether 95:5
to 5:1 gradient elution) to give the adduct 5aa (0.51 g, 88%) as a
clear oil.
Benzene functioned equally well as the solvent and, when
DLP was used as initiator, all reductive alkylation reactions
were carried out at ca. 80 ЊC under gentle reflux in this solvent.
The characteristics of the racemic adducts 5, 6, 8, and 9 are
given below.
4.72 (1 H, dd, J 11.0 and 2.2, OCH); δC 20.9, 24.9, 25.9, 31.0,
34.6, 51.6, 79.9, 171.1, 173.7; νmax (film) 2966, 1739, 1435, 1371,
1242, 1167, 1021, 960, 878, 707 (Found: C, 61.1; H, 9.2.
C11H20O4 requires C, 61.09; H, 9.32%).
Dimethyl (2-tert-butyldimethylsiloxypropyl)malonate 5ca. Oil;
δH 0.01 (3 H, s, SiMe), 0.04 (3 H, s, SiMe), 0.87 (9 H, s, But),
1.15 (3 H, d, J 6.1, MeCH), 1.93 (1 H, ddd, J 13.8, 8.9 and 4.8,
CHAHB), 2.07 (1 H, ddd, J 13.8, 9.6 and 3.7, CHAHB), 3.62
(1 H, dd, J 9.6 and 4.8, CH), 3.73 (3 H, s, OMeA), 3.74 (3 H, s,
OMeB), 3.83 (1 H, m, OCH); δC Ϫ5.1, Ϫ4.3, 19.0, 23.9, 25.8,
38.3, 48.2, 52.4, 52.6, 66.0, 170.0, 170.3; νmax (film) 2956, 2858,
1739, 1438, 1342, 1253, 1152, 1001, 837, 777 (Found: C, 56.6;
H, 7.4. C13H22O6Si requires C, 56.35; H, 7.43%).
Dimethyl (2-acetoxypropyl)malonate 5aa. Oil; δH 1.24 (3 H, d,
J 6.2, Me), 2.00 (3 H, s, Ac), 2.35 (2 H, m, CH2), 3.46 (1 H, dd,
J 8.6 and 6.1, CH), 3.72(7) (3 H, s, OMeA), 3.73(2) (3 H, s,
OMeB), 4.89 (1 H, m, OCH); δC 20.1, 21.1, 34.7, 48.4, 52.7(5),
52.7(7), 68.7, 169.3, 169.5, 170.4; νmax (film) 2956, 2361, 1738,
1438, 1375, 1244, 1158, 1066, 953 (Found: C, 51.8; H, 6.8.
C10H16O6 requires C, 51.72; H, 6.94%).
Dimethyl 3-oxaheptylmalonate 5da. Oil; δH 0.89 (3 H, t, J 7.3,
Me), 1.43 (2 H, m, CH2), 1.50 (2 H, m, CH2), 2.16 (2 H, m,
CH2), 3.36 (2 H, t, J 6.5, CH2O), 3.43 (2 H, t, J 6.0, CH2O), 3.56
(1 H, t, J 7.3, CH), 3.72 (6 H, s, 2 OMe); δC 13.9, 19.2, 29.0,
31.7, 48.8, 52.5, 53.9, 67.7, 70.7, 169.8; νmax (film) 2959, 2870,
1736, 1437, 1160, 1116, 1017, 707 (Found: C, 56.7; H, 8.7.
C11H20O5 requires C, 56.88; H, 8.68%).
Methyl 4-acetoxypentanoate 5ab. Oil; δH 1.21 (3 H, d, J 6.3,
Me), 1.87 [2 H, m, CH2C(O)], 2.00 (3 H, s, Ac), 2.34 (2 H, m,
CH2CO), 3.65 (3 H, s, OMe), 4.89 (1 H, m, OCH); δC 19.8, 21.2,
30.1, 30.8, 51.6, 69.9, 170.6, 173.4; νmax (film) 2980, 2953, 1738,
1439, 1375, 1245, 1077, 964, 707 (Found: C, 55.4; H, 8.0.
C8H14O4 requires C, 55.16; H, 8.10%).
Dimethyl 2-methylheptylmalonate 5ea. Oil; δH 0.87 (3 H, t,
J 7.0, Me), 0.88 (3 H, d, J 6.5, MeCH), 1.0–1.5 (9 H, 4 CH2 and
1 CH), 1.67 (1 H, ddd, J 14.5, 7.9 and 6.6, CH CHAHB), 1.96
(1 H, ddd, J 14.5, 8.6 and 5.5, CH CHAHB), 3.47 [1 H, dd, J 8.6
and 6.6, CH(CO2Me)2], 3.72(5) (3 H, s, OMe), 3.72(9) (3 H, s,
OMe); δC 14.1, 19.2, 22.6, 26.3, 30.8, 32.0. 35.9, 36.6, 49.8, 52.4
52.5, 170.1, 170.2; νmax (film) 2956, 2857, 1739, 1437, 1331,
1247, 1200, 1154, 1015, 976 (Found: C, 63.8; H, 10.0. C13H24O4
requires C, 63.91; H, 9.90%).
Diethyl (2-acetoxypropyl)methylmalonate 5ac. Oil; δH 1.21–
1.28 (9 H, m, 3 Me), 1.40 (3 H, s, Me), 1.95 (3 H, s, Ac), 2.10
(1 H, dd, J 15.0 and 3.0, CH2A), 2.33 (1 H, dd, J 15.0 and 10.5,
B
CH2 ), 4.15 (4 H, m, OCH2), 5.03 (1 H, m, OCH); δC 14.0, 19.4,
21.0, 40.7, 52.0, 61.1, 61.4, 67.4, 170.3, 171.8, 172.1); νmax (film)
2984, 2940, 1736, 1454, 1376, 1296, 1242, 1116, 1020, 951, 862
(Found: C, 57.2; H, 8.2. C13H22O6 requires C, 56.92; H, 8.08%).
Dimethyl 3-acetoxy-2-methylpropylmalonate 5fa. Oil; δH 0.95
(3 H, d, J 6.4, Me), 1.79 (2 H, m, CH2), 2.04 (1 H, m, CH), 2.05
(3 H, s, Ac), 3.50 [1 H, dd, J 8.4 and 6.8, CHC(O)], 3.73 ( 6H, s,
2 MeO), 3.91 (2 H, ddd, J 13.7, 11.0 and 5.5, OCH2); δC 16.5,
20.8, 30.6, 32.6, 49.4, 52.6 (2 C), 68.6, 169.6, 169.8, 171.0; νmax
(film) 2959, 1738, 1439, 1369, 1242, 1157, 1038, 911 (Found: C,
53.7; H, 7.4. C11H18O6 requires C, 53.65; H, 7.37%).
Admantyl 4-acetoxypentanoate 5ad. Oil; δH 1.21 (3 H, d,
J 6.3, Me), 1.65 ( 6 H, br s, 3 CH2 of Ad), 1.82 (2 H, m,
CH2CH2CH), 2.01 (3 H, s, Ac), 2.08 (6 H, br s, 3 CH2 of Ad),
2.13 (3 H, br s, 3 CH of Ad), 2.24 (2 H, m, OCCH2), 4.90 (1 H,
m, CHOAc); δC 19.9, 21.3, 30.8, 31.0, 31.7, 36.2, 41.3, 70.1,
80.5, 170.6, 172.0; νmax (film) 2914, 2856, 1735, 1452, 1373,
1243, 1181, 1056, 966, 869 (Found: C, 69.6; H, 9.0. C17H26O4
requires C, 69.36; H, 8.90%).
(2-Acetoxypropyl)triphenylsilane 6. Mp 61–62 ЊC; δH 1.21 (3
H, d, J 6.1, MeCH), 1.68 (3 H, s, Ac), 1.73 (1 H, dd, J 14.9 and
6.4, SiCHA), 1.98 (1 H, dd, J 14.9 and 8.0, SiCHB), 5.16 (1 H, m,
CHOAc), 7.31–7.62 (15 H, m, 3 Ph); δC 20.9, 21.9, 23.5, 69.2,
127.9, 129.5, 134.6, 135.6, 170.3; νmax (mull) 1720, 1247, 1109,
948 (Found: C, 76.6; H, 6.6. C23H24O2Si requires C, 76.62; H,
6.71%).
1-(4-Acetoxy-1-oxopentyl)admantane 5ae. Oil; δH 1.21 (3 H,
d, J 6.3, Me), 1.69–1.90 (17 H, complex Ad and CH2), 2.02
(3 H, s, Ac), 2.47 (2 H, m, OCCH2), 4.87 (1 H, m, CHOAc);
δC 20.1, 21.4, 27.9, 29.7, 31.9, 36.5, 38.3, 46.3, 70.5, 170.7, 214.6;
νmax (film) 2908, 2852, 1736, 1698, 1451, 1373, 1244, 1013, 952,
707 (Found: C, 73.5; H, 9.3. C17H26O3 requires C, 73.35; H,
9.41%).
5-[2,2-Bis(methoxycarbonyl)ethyl]tetrahydrofuran-2-one
8.
Oil; yield 72%; δH 1.90 (1 H, m, CHCHAHBCH), 2.15 (1 H, ddd,
J 15.1, 9.7 and 5.3, CHCHAHBCH), 2.23 (2 H, m, 4-H), 2.53
(2 H, m, 3-H), 3.66 [1 H, dd, J 9.4 and 5.3, CHC(O)], 3.73 (3 H,
s, OMeA), 3.75 (3 H, s, OMeB), 4.52 (1 H, m, OCH); δC 27.9,
28.4, 34.7, 48.2, 52.8, 77.8, 169.1, 173.3; νmax (film) 2957, 1735,
1740, 1438, 1180, 1043, 926, 652 (Found: C, 52.3; H, 6.2.
C10H14O6 requires C, 52.17; H, 6.13%).
3-Acetoxybutyl phenyl sulfone 5af. Oil; δH 1.21 (3 H, d, J 6.0,
Me), 1.98 (2 H, m, CH2), 2.00 (3 H, s, Ac), 3.13 (2 H, m,
CH2SO2), 4.92 (1 H, m, CHOAc), 7.59 (2 H, m, Ph), 7.91 (1 H,
m, Ph), 7.93 (2 H, m, Ph); δC 19.9, 21.2, 28.8, 52.8, 68.8, 128.1,
129.4, 133.9, 138.9, 170.2; νmax (film) 1736, 1447, 1374, 1307,
1243, 1148, 1088 (Found: C, 56.2; H, 6.5; S, 12.3. C12H16O4S
requires C, 56.23; H, 6.29; S, 12.51%).
5-Triphenylsilylmethyltetrahydrofuran-2-one 9. Mp 96–98 ЊC;
δH 1.65 (1 H, m, 4-HA), 1.80 (1 H, dd, J 14.4 and 9.5, SiCHA),
1.97 (1 H, m, 3-HA), 2.23 (1 H, dd, J 14.4 and 5.0, SiCHB), 2.40
(2 H, m, 4-HB and 3-HB ), 4.72 (1 H, m, OCH), 7.3–7.6 (15 H,
m, 3 Ph); δC 21.5, 29.5, 30.8, 79.5, 128.1, 129.9, 133.7, 135.6,
176.8; νmax (mull) 1736, 1105, 968, 725 (Found: C, 77.2; H, 6.3.
C23H22O2Si requires C, 77.06; H, 6.19%).
Dimethyl 2-acetoxy-3,3-dimethylbutylmalonate 5ba. Oil; δH
0.90 ( 9 H, s, But), 1.98 (1 H, ddd, J 14.4, 11.1 and 4.8, CHAHB),
2.05 (3 H, s, Ac), 2.28 (1 H, ddd, J 14.4, 9.9 and 2.0, CHAHB),
3.33 [1 H, dd, J 9.9 and 4.8, CHC(O)], 3.72 (3 H, s, OMeA), 3.74
(3 H, s, OMeB), 4.73 (1 H, dd, J 11.1 and 2.0, AcOCH); δC 20.8,
25.7, 29.2, 34.7, 48.7, 52.7, 78.2, 169.8, 171.0; νmax (film) 2961,
1738, 1437, 1372, 1243, 1155, 1022, 958, 892 (Found: C, 56.9;
H, 8.0. C13H22O6 requires C, 56.92; H, 8.08%).
Enantioselective reactions
Enantioselective reductive alkylation reactions were carried out
in refluxing benzene, using the general method described above,
with DLP as the initiator (10 mol% based on alkene) and one of
the homochiral thiols 15–18 as catalyst (10 mol%). The optical
Methyl 4-acetoxy-5,5-dimethylhexanoate 5bb. Oil; δH 0.91
(9 H, s, But), 1.75 (1 H, m, CHAHB), 1.95 (1 H, m, CHAHB), 2.06
(3 H, s, Ac), 2.28 (2 H, t, J 7.9, CH2CO), 3.68 (3 H, s, OMe),
2066
J. Chem. Soc., Perkin Trans. 1, 1999, 2061–2068

View Article Online
rotations and the corresponding ees of the products are given in
Table 2; the characteristics of the racemic adducts, obtained
using TPST (10 mol%) as catalyst, are given below.
and CHBCO2), 2.85 (1 H, m, 3-HB), 3.63 ( 3 H, s, OMe), 5.21
(1 H, ddd, J 11.3, 2.3 and 2.0, 6-H), 7.0–7.4 (10 H, m, 2 Ph); δC
26.9, 27.4, 27.6, 30.3, 47.8, 51.7, 83.4, 126.8, 127.0, 127.1, 127.4,
128.7, 128.8, 143.8, 143.9, 169.6, 173.4; νmax (mull) 1720, 1560,
1287, 1053, 950, 755 (Found: C, 74.4; H, 6.4. C21H22O4 requires
C, 74.54; H, 6.55%). The ee was determined by 1H NMR
analysis.
5,5-Dimethyl-6-[2,2-bis(methoxycarbonyl)ethyl]tetrahydro-
pyran-2-one 11aa. Oil; yield 62%; δH 0.95 (3 H, s, CMeA), 1.01 (3
H, s, CMeB), 1.60 (2H, m, 4-CH2), 1.98 (1 H, m, OCHCHA),
2.25 (1 H, m, OCHCHB), 2.52 (2 H, m, 3-H), 3.73 (6 H, s, 2
OMe), 3.74 [1 H, dd, J 10.4 and 4.0, CHC(O)], 4.00 (1H, dd,
J 11.2, 1.6, 6-H); δC 19.3, 26.2, 27.3, 29.6, 31.9, 34.3, 47.8, 52.7,
52.8, 84.6, 169.4, 169.7, 170.7; νmax (film) 1779, 1737, 1438,
1350, 1260, 1162 (Found: C, 57.0; H, 7.5. C13H20O6 requires C,
57.33; H, 7.40%). The ee was determined by 1H NMR analysis.
6-[2-(1-Adamantyloxycarbonyl)ethyl]-5,5-diphenyltetrahydro-
pyran-2-one 11bd. Mp 163–164 ЊC; yield 52%; δH 1.38 ( 1 H, m,
OCHCHAHB), 1.81 ( 1 H, m, OCHCHAHB), 2.04 (6 H, br d, J
2.8, 3 CH2 of Ad), 2.10 (1 H, m, 4-HA), 2.13 (3 H, br s, 3 CH of
Ad), 2.32 ( 1 H, m, 4-HB), 2.48 [2 H, m, CH2C(O)], 2.59 ( 1 H,
ddd, J 18.5, 6.1 and 1.8, 3-HA), 2.89 (1 H, m, 3-HB), 5.20 (1 H,
m, 6-H), 7.10–7.40 (10 H, m, 2 Ph); δC 26.9, 27.5, 27.6, 30.8,
31.6, 36.1, 41.3, 47.9, 80.6, 83.4, 126.8, 126.9, 127.2, 127.5,
128.7, 128.8, 143.9, 144.0, 167.7, 171.9; νmax (mull) 1699, 1652,
1176, 1057, 701 (Found: C, 78.4; H, 7.5. C30H34O4 requires C,
78.57; H, 7.47%). The ee was determined by HPLC (eluent 15%
isopropyl alcohol; tR 10.4 and 13.0 min).
5,5-Dimethyl-6-[2-(methoxycarbonyl)ethyl]tetrahydropyran-
2-one 11ab. Oil; yield 87%; δH 0.93 (3 H, s, MeA), 1.00 (3 H, s,
MeB), 1.60 (1 H, m, OCHCHAHB), 1.69 (2 H, m, 4-H), 1.92
(1 H, m, OCHCHAHB), 2.40–2.70 [4 H, m, 3-H and CH2C(O)],
3.65 (3 H, s, OMe), 3.97 (1 H, dd, J 11.2 and 1.6, 6-H); δC 19.4,
25.2, 26.4, 27.4, 30.0, 32.0, 34.4, 51.7, 86.4, 171.3, 173.7; νmax
(film) 1737, 1440, 1352, 1168, 1056 (Found: C, 61.6; H, 8.4.
C11H18O4 requires C, 61.66; H, 8.47%). The ee was determined
by 1H NMR analysis.
6-(1-Adamantyl-1-oxo-propyl)-5,5-diphenyltetrahydropyran-
2-one 11be. Mp 145–147 ЊC; yield 57%; δH 1.38 (1 H, m, OCH-
CHAHB), 1.80 (1 H, m, OCHCHAHB), 1.60–2.02 (15 H, m, Ad),
2.11 (1H, ddd, J 19.1, 11.8 and 7.0, 4-HA), 2.45 (1 H, m, 4-HB ),
2.57 [2 H, m, CH2C(O)], 2.73 (1 H, ddd, J 19.1, 7.8 and 5.4, 3-
HA), 2.89 (1 H, 3-HB), 5.16 (1 H, m, 6-H), 7.10–7.40 (10 H, m, 2
Ph); δC 26.2, 26.9, 27.5, 27.9, 32.3, 36.5, 38.2, 46.2, 47.9, 83.5,
126.8, 126.9, 127.2, 127.5, 128.7, 128.8, 144.0, 144.1, 170.0,
214.9; νmax (mull) 1736, 1055, 701 (Found: C, 81.3; H, 7.7.
C30H34O3 requires C, 81.41; H, 7.74%). The ee was determined
by HPLC (eluent 15% isopropyl alcohol; tR 11.5 and 25.4 min).
6-(2-Phenylsulfonylethyl)-5,5-dimethyltetrahydropyran-2-one
11af. Mp 127–128 ЊC; yield 72%; δH 0.93 (3 H, s, Me), 1.01 (3 H,
s, Me), 1.61 (1 H, m, OCHCHAHB), 1.70 (1 H, m, OCH-
CHAHB), 1.91 (1 H, m, 4-HA), 2.15 (1 H, m, 3-HA), 2.51 ( 2 H,
4-HB and 3-HB), 3.14 (1 H, ddd, J 14.0, 9.8 and 5.6,
SO2CHAHB), 3.41 (1 H, ddd, J 14.0, 9.8 and 5.2, SO2CHAHB),
4.07 (1 H, dd, J 11.2 and 2.0, 6-H), 7.58 (2 H, m, aromatic), 7.67
(1 H, m, aromatic), 7.91 (2 H, m, aromatic); δC 19.0, 23.0, 26.2,
27.2, 32.0, 34.3, 52.9, 85.4, 127.8, 129.3, 133.8, 139.1, 170.7;
νmax (mull) 1732, 1211 1050, 753 (Found: C, 60.8; H, 6.7.
C15H20O4S requires C, 60.79; H, 6.80%). The ee was determined
by HPLC (eluent 20% isopropyl alcohol; tR 6.7 and 8.9 min).
6-[2-(Phenylsulfonyl)ethyl]-5,5-diphenyltetrahydropyran-2-
one 11bf. Mp 192–194 ЊC; yield 70%; δH 1.62 ( 1 H, m, OCH-
CHAHB), 2.00 (1 H, m, OCHCHAHB), 2.13 ( 1 H, ddd, J 18.7,
11.6 and 7.2, 4-HA), 2.46 (1 H, m, 4-HB), 2.55 (1 H, ddd, J 18.7,
5.6 and 2.5, 3-HA), 2.75 (1 H, m, 3-HB), 3.15 (1 H, ddd, J 14.1,
9.1 and 6.3, SO2CHA), 3.32 (1 H, ddd, J 14.1, 9.1 and 5.3,
SO2CHB), 5.24 (1 H, ddd, J 11.2, 2.0 and 1.9, 6-H), 7.0–7.7 (15
H, m, 3 Ph); δC 25.8, 27.0, 27.4, 47.8, 52.7, 82.4, 127.0, 127.2,
127.3 (2C), 127.8, 128.8(5), 128.9(1), 129.2, 133.7, 138.7, 143.1,
143.4, 169.1; νmax (mull) 1739, 1053, 930, 745, 702 (Found: C,
71.2; H, 5.7. C25H24O4S requires C, 71.41; H, 5.75%). The ee
was determined by HPLC (eluent 20% isopropyl alcohol; tR
11.3 and 14.4 min).
6-[2,2,2-Tris(ethoxycarbonyl)ethyl]-5,5-dimethyltetrahydro-
pyran-2-one 11ai. Oil; which partially solidified on standing at
room temperature, yield 57%; δH 0.96 (3H, s, CMeA), 1.06 (3H,
s, CMeB), 1.26 (9 H, t, J 7.2, OCH2CH3), 1.60 (1H, ddd, J 13.6,
7.2 and 5.6, 4-HA), 1.70 (1 H, m, 4-HB), 2.15 (1 H, dd, J 14.8
and 1.6, OCHCHA), 2.28 (1 H, dd, J 14.8 and 10.0, OCHCHB),
2.51 (2 H, m, 3-H), 4.26 (6H, q, J 7.2, OCH2CH3), 4.55 (1H, dd,
J 10.0 and 1.6, 6-H); δC 13.9, 19.7, 26.3, 27.3, 32.3, 34.2(8),
34.3(4), 62.4, 63.0, 82.5, 166.7, 170.4; νmax (film) 1740, 1469,
1368, 1221, 1075 (Found: C, 58.2; H, 7.6, C18H28O8 requires
C, 58.05; H, 7.58%). The ee was determined by 1H NMR
analysis. Careful repeated recrystallisation from hexane–
dichloromethane gave enantiopure (S)-(Ϫ)-11ai, [α]D22 = Ϫ37.5
(c 1.44, CHCl3), mp 63–64 ЊC.
6-(4,4-Dimethyl-3-oxopentyl)-5,5-diphenyltetrahydropyran-2-
one 11bg. Mp 112 ЊC; yield 78%; δH 1.11 (9 H, s, But), 1.39 (1 H,
dddd, J 14.9, 12.4, 7.6 and 2.5, OCHCHAHB), 1.80 (1 H, m,
OCHCHBHA), 2.11 (1 H, ddd, J 19.1, 12.1 and 7.3, 4-HA), 2.46
(1 H, m, 4-HB), 2.60 [2 H, m, CH2C(O)], 2.77 ( 1 H, ddd, J 18.5,
7.9 and 5.3, 3-HA), 2.88 (1 H, m, 3-HB), 5.17 ( 1 H, m, 6-H),
7.10–7.40 (10 H, m, 2 Ph); δC 26.5, 26.9, 27.5(1), 27.5(3), 32.7,
44.0, 47.9, 83.5, 126.8, 127.0, 127.2, 127.5, 128.7, 128.8, 143.9,
144.0, 169.9, 215.2; νmax (mull) 1735, 1699, 1651, 1182, 1056,
761 (Found: C, 78.8; H, 7.8. C24H28O3 requires C, 79.09; H,
6-[(2,2-Bis(methoxycarbonyl)ethyl]-5,5-diphenyltetrahydro-
pyran-2-one 11ba. Mp 130–131 ЊC; yield 92%; δH 1.68 (1 H, ddd,
J 18.2, 7.9 and 6.7, OCHCHAHB), 2.10 (1 H, ddd, J 19.1, 11.8
and 7.0, 4-HA), 2.22 (1 H, ddd, J 18.2, 11.5 and 6.6, OCH-
CHAHB), 2.46 (1 H, m, 4-HB), 2.59 (1 H, ddd, J 18.8, 6.1,and
1.8, 3-HA), 2.87 (1 H, m, 3-HB), 3.68 (3 H, s, OMe), 3.71 (1 H,
dd, J 7.5 and 6.7, CH), 3.73 (3 H, s, OMe), 5.30 ( 1 H, ddd,
J 9.3, 2.2 and 2.0, 6-H), 7.10–7.40 (10 H, m, Ph); δC 26.7, 27.3,
31.8, 47.6, 48.3, 52.7, 52.9, 81.9, 126.9, 127.1, 127.2, 127.3,
128.8 (2 C), 143.4, 143.6, 169.1, 169.7; νmax (mull) 1746, 1597,
1268, 1180, 1057, 933, 760 (Found: C, 69.5; H, 6.0. C23H24O6
1
7.74%). The ee was determined by H NMR analysis. Careful
repeated recrystallisation from hexane–dichloromethane gave
enantiopure (S)-(Ϫ)-11bg, [α]D20 = Ϫ266.0 (c 1.35, CHCl3), mp
170–172 ЊC.
6-[2,2-Bis(tert-butoxycarbonyl)ethyl]-5,5-diphenyltetrahydro-
pyran-2-one 11bh. Mp 84–85 ЊC; yield 62%; δH 1.34 (9 H, s, But),
1.42 (9 H, s, But), 1.61 (1 H, ddd, J 14.6, 9.5 and 2.4, OCH-
CHAHB), 2.11 (2 H, m, 4-HA and OCHCHAHB), 2.47 (1 H, m,
4-HB), 2.60 (1H, m, 3-HA), 2.87 (1 H, m, 3-HB), 3.53 [1 H, dd,
J 9.5 and 4.8, CHC(O)], 5.26 (1 H, m, 6-H), 7.10–7.50 (10 H, 2
Ph); δC 26.9, 27.4, 27.8, 27.9, 31.4, 47.8, 50.2, 81.8, 83.5, 126.8,
127.0, 127.2, 127.4, 128.8(0), 128.8(2), 143.6, 143.8, 168.2,
168.4, 169.4; νmax (mull) 1730, 1699, 1651, 1272, 758 (Found:
1
requires C, 69.68; H, 6.10%). The ee was determined by H
NMR analysis.
6-[2-(Methoxycarbonyl)ethyl]-5,5-diphenyltetrahydropyran-2-
one 11bb. Mp 143–144 ЊC; yield 75%; δH 1.43 (1 H, dddd, J 14.6,
12.1, 7.6 and 2.6, OCHCHAHB), 1.89 (1 H, dddd, J 14.6, 11.3,
6.7 and 4.6, OCHCHAHB), 2.12 (1 H, ddd, J 18.7, 11.4 and 6.7,
4-HA), 2.44 (2 H, m, 3-HA and CHACO2), 2.56 (2 H, m, 4-HB
J. Chem. Soc., Perkin Trans. 1, 1999, 2061–2068
2067

View Article Online
C, 72.4; H, 7.3. C29H36O6 requires C, 72.48; H, 7.55%). The ee
was determined by 1H NMR analysis.
µ(Cu-Kα) = 0.586 mmϪ1, transmission range 0.664 to 0.525,
T = 200 K (the structure undergoes a phase transition at
around 175 K as the temperature is lowered giving a unit cell with
three times the volume and three molecules in the asymmetric
unit), 3660 reflections measured to 2θmax = 135Њ, 3467 unique
(Rint = 0.0436). Final wR = 0.1269 for all data, conventional
R = 0.0450 [for 3435 reflections with F2 > 2σ(F2)]. The absolute
6-[(2-tert-Butoxycarbonyl)ethyl]-5,5-diphenyltetrahydro-
pyran-2-one 11bj. Mp 110–112 ЊC; yield 83%; δH 1.34 (9 H, s,
But), 1.40 (1 H, m, OCHCHAHB), 1.82 (1 H, m, OCHCHAHB),
2.12 [1 H, ddd, J 18.9, 11.6 and 6.8, CHAC(O)], 2.32 [1 H, m,
CHBC(O)], 2.40–2.52 ( 2 H, m, 3-HA and 4-HA), 2.59 (1 H, ddd,
J 18.7, 6.1 and 2.0, 4-HB), 2.86 (1 H, m, 3-HB), 5.19 (1 H, ddd,
J 11.2, 2.3 and 2.0, 6-H) 7.10–7.35 (10 H, m, 2 Ph); δC 26.9,
27.4, 27.6, 28.1, 31.6, 47.9, 80.5, 83.4, 126.8, 127.0, 127.2, 127.5,
128.8 (2C), 144.0 (2C), 169.8, 172.2; νmax (mull) 1721, 1278,
1060, 954, 755 (Found: C, 75.5; H, 7.3. C24H28O4 requires C,
75.76; H, 7.42%). The ee was determined by 1H NMR analysis.
configuration was determined with
a Flack parameter
x = 0.1(3). The final electron density map was featureless.
Acknowledgements
We are grateful to the EPSRC for supporting this work and we
thank Dr S. V. Kelkar for preparing the thiol 18. We would also
like to thank Professor W. Clegg for use of the crystallographic
facilities in Newcastle.
4,4-Dimethyl-5-[2,2-bis(methoxycarbonyl)ethyl]-1,3-dioxolan-
2-one 14a. Oil; yield 71%; δH 1.40 (3 H, s, Me), 1.50 (3 H, s,
Me), 2.15 (1 H, ddd, J 17.3, 11.0 and 4.8, CHAHB), 2.20 (1 H,
ddd, J 17.3, 9.8 and 2.6, CHAHB), 3.65 [1 H, dd, J 9.8 and 4.8,
CHC(O)], 3.76 (3 H, s, OMe), 3.78 (3 H, s, OMe), 4.31 (1 H, dd,
J 11.0 and 2.6, 5-H); δC 21.1, 26.0, 28.7, 47.8, 52.9(6), 53.0(2),
82.3, 83.8, 153.3, 168.7, 168.8; νmax (film) 2957, 1794, 1732,
1437, 1346, 1047, 776, 763 (Found: C, 48.5; H, 6.5. C10H16O7
requires C, 48.39; H, 6.50%). The ee was determined by HPLC
(eluent 15% isopropyl alcohol; tR 5.4 and 7.5 min with detection
at 224 nm).
References
1 (a) B. Giese, Radicals in Organic Synthesis: Formation of Carbon–
Carbon Bonds, Pergamon Press, Oxford, 1986; (b) M. Ramaiah,
Tetrahedron, 1987, 43, 3541; (c) D. P. Curran, Synthesis, 1988, 417,
489; (d) C. P. Jasperse, D. P. Curran and T. L. Fevig, Chem. Rev.,
1991, 91, 1237; (e) W. B. Motherwell and D. Crich, Free Radical
Chain Reactions in Organic Synthesis, Academic Press, London,
1992; ( f ) D. P. Curran, N. A. Porter and B. Giese, Stereochemistry
of Radical Reactions, VCH, Weinheim, 1996.
2 B. Giese. J. A. Gonzáles-Gómez and T. Witzel, Angew. Chem., Int.
Ed. Engl., 1984, 23, 69.
3 C. Walling, Free Radicals in Solution, Wiley, New York, 1957.
4 P. A. Baguley and J. C. Walton. Angew. Chem., Int. Ed. Engl., 1998,
37, 3073.
4,4-Dimethyl-5-(2-phenylsulfonylethyl)-1,3-dioxolan-2-one
14b. Oil; yield 55%; δH 1.32 (3 H, s, MeA), 1.44 (3 H, s, MeB),
1.99–2.07 (2 H, m, SO2CH2), 3.11–3.20 (1 H, m, CH2A), 3.22–
B
3.31 (1 H, m, CH2 ), 4.35 (1 H, dd, J 2.8 and 10.8, OCH), 7.53–
5 (a) R. P. Allen, B. P. Roberts and C. R. Willis, J. Chem. Soc., Chem.
Commun., 1989, 1387; (b) J. N. Kirwan, B. P. Roberts and C. R.
Willis, Tetrahedron Lett., 1990, 31, 5093; (c) S. J. Cole, J. N. Kirwan,
B. P. Roberts and C. R. Willis, J. Chem. Soc., Perkin Trans. 1, 1991,
103.
7.87 (5 H, m, Ph); δC 21.2, 23.0, 26.2, 52.6, 82.9, 84.0, 127.9,
129.6, 134.3, 138.7, 153.3; νmax (film) 1790, 1448, 1308, 1279,
1147 (Found: C, 55.2; H, 5.6; S, 11.3. C13H16O5S requires C,
54.92; H, 5.67; S, 11.28%).
6 B. P. Roberts, Chem. Soc. Rev., 1999, 28, 25.
7 (a) B. P. Roberts and A. J. Steel, J. Chem. Soc., Perkin Trans. 2, 1994,
2155; (b) B. P. Roberts, J. Chem. Soc., Perkin Trans. 2, 1996, 2719;
(c) A. A. Zavitsas, J. Chem. Soc., Perkin Trans. 2, 1998, 499; (d)
C. H. Schiesser and M. A. Skidmore, J. Chem. Soc., Perkin Trans. 2,
1998, 2329.
8 W. Smadja, M. Zahouily, M. Journet and M. Malacria, Tetrahedron
Lett., 1991, 32, 3683.
9 (a) H.-S. Dang and B. P. Roberts, Tetrahedron Lett., 1995, 36, 3731;
(b) H.-S. Dang and B. P. Roberts, Chem. Commun., 1996, 2201;
(c) H.-S. Dang and B. P. Roberts, J. Chem. Soc., Perkin Trans. 1,
1998, 67; (d) Y. Cai and B. P. Roberts, J. Chem. Soc., Perkin
Trans. 1, 1998, 467; (e) M. B. Haque, B. P. Roberts and D. A. Tocher,
J. Chem. Soc., Perkin Trans. 1, 1998, 2881.
10 H.-S. Dang, K.-M. Kim and B. P. Roberts, Chem. Commun., 1998,
1413.
11 H. Kiefer and T. G. Traylor, Tetrahedron Lett., 1966, 6163.
12 G. D. Mendenhall, Tetrahedron Lett., 1983, 24, 451.
13 V. Diart and B. P. Roberts, J. Chem. Soc., Perkin Trans. 2, 1992, 1761.
14 C. Chatgilialoglu, Acc. Chem. Res., 1992, 25, 188; Chem. Rev., 1995,
95, 1229.
X-Ray crystallography
All measurements were made on a Stoe-Siemens AED 2 dif-
fractometer with Cu-Kα radiation (λ = 1.54184 Å). For (S)-
11ai, a Bede Scientific Microsource X-ray tube was employed
using an 8 µm filter to remove Cu-Kβ radiation. For (S)-11bg,
the source was a conventional sealed X-ray tube and the radi-
ation was monochromated by a graphite crystal. All machine
control calculations were performed with standard Stoe DIF4
software. Intensities were measured with ω/θ scans and on-line
profile fitting.22 Data were obtained at low temperature using a
Cryostream cooler23 and were corrected for Lorentz and polar-
isation effects, crystal decay, and (by ψ-scans) for absorption.
The structures were solved by direct methods and refined by
full-matrix-least-squares on F2.24 CCDC reference number 207/
343. See http://www.rsc.org/suppdata/p1/1999/2061 for crystal-
lographic files in .cif format.
15 S. V. Kelkar and B. P. Roberts, unpublished results.
16 H. O. House, D. S. Crumrine, A. Y. Teranishi and H. D. Olmstead,
J. Am. Chem. Soc., 1973, 95, 3310.
17 K. Mikami, S. Matsumoto, A. Ishida, S. Takamuku, T. Suenobu
and S. Fukuzumi, J. Am. Chem. Soc., 1995, 117, 11134.
18 (a) E. J. Cragoe, A. M. Pietruszkiewicz and C. M. Robb, J. Org.
Chem., 1958, 23, 971; (b) E. J. Cragoe and A. M. Pietruszkiewicz,
J. Org. Chem., 1957, 22, 1338.
19 D. M. T. Chen, T. B. Marder, D. Milstein and N. J. Taylor, J. Am.
Chem. Soc., 1987, 109, 6385.
20 (a) J. M. Joumier, C. Bruneau and P. H. Dixneuf, J. Chem. Soc.,
Perkin Trans. 1, 1991, 3271; (b) P. L. Gendre, F. Jerome, C. Bruneau
and P. H. Dixneuf, Chem. Commun., 1998, 533.
Crystal data for (S)-(؊)-6-[2,2,2-tris(ethoxycarbonyl)ethyl]-
5,5-dimethyltetrahydropyran-2-one 11ai. C18H28O8, M = 372.40,
monoclinic, a = 10.0392(8), b = 7.6254(6), c = 13.2278(11) Å,
β = 107.559(7)Њ, V = 965.44(13) Å3 [from 2θ values of 42 reflec-
tions measured at ω (43Њ < 2θ < 50Њ)], space group P21, Z = 2,
µ(Cu-Kα) = 0.843 mmϪ1, transmission range 0.891 to 0.708,
T = 160 K, 3630 reflections measured to 2θmax = 135Њ, 2956
unique (Rint = 0.0267). Final wR = 0.1101 for all data, conven-
tional R = 0.0409 [for 2898 reflections with F2 > 2σ(F2)]. The
absolute configuration was determined with a Flack parameter
x = 0.1(2). The final electron density map was featureless.
21 F. Adickes, W. Brunnert and O. Luker, J. Prakt. Chem., 1931, 130,
166.
22 W. Clegg, Acta Crystallogr., Sect. A, 1981, 37, 22.
23 J. Cosier and A. M. Glazer, J. Appl. Crystallogr. 1986, 19, 105.
24 G. M. Sheldrick, SHELXTL User Manual, Version 5, Bruker AXS
Inc., Madison, WI, 1994.
Crystal data for (S)-(؊)-6-(4,4-dimethyl-3-oxopentyl)-5,5-
diphenyltetrahydropyran-2-one 11bg. C24H28O3, M = 364.46,
monoclinic, a = 9.8331(10), b = 10.3392(10), c = 10.5216(12) Å,
β = 99.957(12)Њ, V = 1053.58(19) Å3 [from 2θ values of 43 reflec-
tions measured at ω (45Њ < 2θ < 50Њ)], space group P21, Z = 2,
Paper 9/03961I
2068
J. Chem. Soc., Perkin Trans. 1, 1999, 2061–2068