The highly enantioselective transformation of silylketenes into α-silylthioesters catalysed by cinchona alkaloids

Article abstract of DOI:10.1016/S0040-4039(01)00304-5

The reaction of silylketenes with thiophenol, mediated by cinchona alkaloid catalysts, proceeds to give α-silylthioester products in good chemical yield and with enantiomeric excess values in the range 79-93%. The absolute configuration of one of the thioester products was determined by X-ray diffraction.

Full text of DOI:10.1016/S0040-4039(01)00304-5

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 2877–2881
The highly enantioselective transformation of silylketenes into
a-silylthioesters catalysed by cinchona alkaloids
Alexander J. Blake,a Christopher L. Friend,a Robert J. Outram,a Nigel S. Simpkinsa,* and
Andrew J. Whiteheadb
aSchool of Chemistry, University of Nottingham, University Park, Nottingham NG7 2RD, UK
bGlaxoWellcome, Gunnels Wood Rd, Stevenage, Hertfordshire SG1 2NY, UK
Received 11 January 2001; accepted 20 February 2001
Abstract—The reaction of silylketenes with thiophenol, mediated by cinchona alkaloid catalysts, proceeds to give a-silylthioester
products in good chemical yield and with enantiomeric excess values in the range 79–93%. The absolute configuration of one of
the thioester products was determined by X-ray diffraction. © 2001 Elsevier Science Ltd. All rights reserved.
Forty years ago the research group of Pracejus reported
a remarkable transformation in which simple ketenes,
such as phenyl methyl ketene 1, react with alcohols in
the presence of sub-stoichiometric quantities of cin-
chona alkaloid catalysts, such as quinine derivative 2,
to give ester products 4 in up to 76% ee, Scheme 1.1
tioselective (typically 80–90% ee) transformation of
silylketenes into chiral thioester products.
On recognising the potential of the Pracejus asymmetric
transformation we were immediately attracted to the
possibility of applying the reaction to silylketenes.4
These compounds are readily prepared, are reasonably
stable and might lead to interesting chiral silylesters
when transformed according to Scheme 1. Following
established protocols, alkylation of the parent a-silyl-
acids 6 (R=Me, Ph), gave a range of derivatives 7,
which could then be converted into the required
silylketenes 8 (Table 1).5
Despite the obvious potential of this reaction for the
asymmetric synthesis of carboxylic acid derivatives
rather little directly related research has been carried
out subsequently.2 Quite recently this process was re-
examined by Fu and co-workers, employing the syn-
thetic azaferrocene derivative 5 as catalyst in place of
the alkaloid, and selectivities of up to 80% ee were
obtained.3
These preparations were not optimised and it is thought
that the yields of many ketenes could be improved. In
our hands the alkylations of trimethylsilylacetic acid 6
(R=Me) were more facile than those of the more
In this communication we describe our independent
studies in this area, which have uncovered a novel
variant of this reaction which enables the highly enan-
_
+
NHR3*
O
O
CH2OTBS
N
O
H
N
C
2 (1 mol%)
Ph
Ph
Ph
Fe
MeO
OMe
OMe
Me
Me
Me
Me
MeOH, -110 oC
OBz
Me
Me
H
Me
H
Me
N
4
1
5
2
3
Scheme 1.
* Corresponding author.
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)00304-5

2878
A. J. Blake et al. / Tetrahedron Letters 42 (2001) 2877–2881
Table 1. Preparation of silylketenes
O
O
O
RMe2Si
LDA (2 equiv.)
R'CH2Hal
RMe2Si
C
method A or B*
RMe2Si
OH
OH
C
R'
R'
8
6
7
Entry
R
R%
7 (%)
8 (%)
1
2
3
4
5
6
Me
Me
Me
Me
Ph
Ph
7a 93
7b 88
7c 72
7d 90
7e 71
7f 41
8a 45 (A)
8b 48 (A) 65 (B)
8c 29 (A)
8d 27 (A)
8e 67 (B)
2-Naphthyl
nPr
PhCH2
Ph
(CH2)2CHꢀCH2
Ph
8f 72 (B)
Method A: (i) (COCl)2, DMF, CH2Cl2 (ii) Br2 (iii) Zn, THF. Method B: (COCl)2, Et3N.
hindered dimethylphenylsilyl derivative 6 (R=Ph). For
the conversion of acids 7 into the corresponding kete-
nes we employed either dehalogenation (method A) or
dehydrochlorination (method B).6
methanol and alkaloid catalyst at the low temperature
usually employed. After some experimentation we
turned instead to PhSH as a more nucleophilic partner
for the ketene and found that smooth addition occurred
to give a-silylthioester 9a (Scheme 2).7
Initial attempts to apply the Pracejus reaction to a
typical silylketene 8a were very disappointing, with very
little reaction being observed in the presence of
To our delight, the use of 2.5–10% of quinine derivative
2 gave (−)-9a in yields of ca. 95% and ee values of
O
O
MeO
H
OBz
catalyst (see text)
Me3Si
Ph
C
C
Ph
Me3Si
SPh
N
PhH, PhSH, -78 oC
H
H
N
8a
9a
10
Scheme 2.
Table 2. Asymmetric synthesis of a-silylthioesters
O
O
RMe2Si
R'
C
catalyst 2 or 10
C
R'
SPh
PhH, PhSH, -78 oC
RMe2Si
H
8
9
Entry
Ketene
Catalyst
Product (%)
eea
1
2
3
4
5
6
7
8
9
8a
8a
8b
8b
8c
8d
8e
8e
8f
2
10
2
10
2
2
2
10
2
10
(−)-9a (97)
(+)-9a (99)
(−)-9b (99)
(+)-9b (99)
(−)-9c (99)
(−)-9d (86)
(+)-9e (94)
(−)-9e (84)
(+)-9f (86)
(−)-9f (84)
91
91
93
94
89
89
82
79
84b
82b
10
8f
a Determined by HPLC, see: Ref. 10.
b Approximate values due to incomplete enantiomer separation.

A. J. Blake et al. / Tetrahedron Letters 42 (2001) 2877–2881
2879
+
N
PhSH
_
CH2Ar
OCOPh
O
H
R3Si
MeO
N
Figure 2. Model for the asymmetric protonation of intermedi-
ate 12 (vinyl group of quinuclidine system omitted for clarity).
Figure 1. X-Ray structure of (+)-9b.
alkaloid to generate an ammonium enolate 12 that
protonates selectively and then undergoes attack by
thiolate to give the product 14 and regenerate the
alkaloid catalyst.
90–93%. The catalyst loading could be reduced to 1%,
but the reaction then took longer to proceed to comple-
tion and there was a drop in the selectivity to around
80% ee. Commercially available (DHQ)2PHAL gave
similar results to 2,8 and the use of the pseudoenan-
tiomeric quinidine-derived catalyst 10 resulted in the
formation of (+)-9a in similarly high yield and similar
ee of 84%. Lowering the reaction temperature to ca.
−105°C or running the reaction at −45°C resulted in
lower ee values, with some by-product formation being
observed at the higher temperature.
Whichever mechanism operates, it is assumed that one
isomer of the intermediate enolate is formed by attack
on the ketene anti to the largest group, RL in ketene 11,
which we assume is the bulky silicon group in our
examples. In the case of the Pracejus mechanism it is
especially difficult to account for the observed selectiv-
ity because of the unknown spatial relationship of the
ion-pair prior to proton transfer. In the case of the
mechanism outlined in Scheme 3 the stereochemical
outcome of the quinidine catalysed reaction leading to
(S)-9b can be reasonably accounted for based on the
model of Wynberg for the related ketene-chloral
cycloaddition (Fig. 2).12
Following this initial exploration we applied the opti-
mal conditions to the ketenes prepared earlier and
obtained the results shown in Table 2.9
The reactions are very clean, and give high yields of
thioester products in good enantiomeric excess. In the
case of the trimethylsilyl derivatives the quinine catalyst
2 gave (−)-products and the quinidine catalyst (+)-prod-
ucts, with the reverse being apparent for the dimethyl-
silyl compounds. In the case of (+)-9b we were able to
crystallise the product and obtain an enantiomerically
pure sample suitable for X-ray analysis (Fig. 1).11
In order to establish some preliminary idea of the utility
of the novel chiral silicon-containing products we next
carried out the range of transformations illustrated in
Scheme 4.
Reaction of the trimethylsilylthioester 9a with
organocuprate reagents generated from the correspond-
ing organolithium and CuBr–SMe2 gave the expected
ketones 15 in good to excellent yield.13 This type of
ketone has been demonstrated by Enders and co-work-
ers to be extremely useful in stereocontrolled aldol
reactions.14 Reduction of dimethylphenylsilyl derivative
(−)-9e, thought to have the (S)-configuration shown,15
with DIBAL gave alcohol 16 which, when treated
under the conditions described by Fleming, was
smoothly transformed into the diester 18.16
The structure clearly shows the (S)-configuration for
this product and, given the trends in the sign of [h]D
values and HPLC properties, we believe that the other
members of the trimethylsilyl series of products are also
of the same configuration.
The results can be rationalised on the basis of the
mechanism originally proposed by Pracejus, i.e.
analogous to that shown in Scheme 1, which involves
initial addition of thiol to the ketene to generate a
thioester enolate that undergoes stereoselective proton
transfer from the chiral ammonium counterion to the
enolate carbon centre. Alternatively, it is possible that
the initial attack on the ketene 11 is by the cinchona
These transformations would be expected to occur with
little or no loss of stereochemical integrity, although at
present we have not proved the levels of enrichment of
the products.
_
O
O
O
_
O
NR3*
PhSH
PhS
+
C
RL
+
RL
RL
RL
+
NR3*
NR3*
SPh
NR3*
RS
12
H
RS
14
RS
H
RS
13
11
Scheme 3.

2880
A. J. Blake et al. / Tetrahedron Letters 42 (2001) 2877–2881
O
O
R2CuLi, Et2O
Ph
Me3Si
Ph
Me3Si
R
SPh
H
H
9a
15 (60-90%)
R = Me, Bu, Ph
O
Hg(OAc)2, AcO2H
DIBAL-H, THF
65%
Ph
Ph
PhMe2Si
Ph
SPh
OH
OAc
AcOH, RT
70%
PhMe2Si
AcO
H
H
H
(-)-9e
16
18
Scheme 4.
In conclusion, we have demonstrated a novel, and
highly enantioselective transformation of silylketenes
that has potential for the synthesis of both silicon
containing compounds, such as allylsilanes, and silicon-
free products such as diols (cf. 18). Further work is
underway to establish the scope of the new process with
other types of ketene.17
relatively unstable derivatives are free of amine, but gives
somewhat modest yields. However, we then discovered
that the more stable dimethylphenylsilyl series were very
easily prepared in much better yield by simple dehydro-
halogenation of the intermediate acid chlorides.
7. For related work, see: Fehr, C.; Stempf, I.; Galindo, J.
Angew Chem., Int. Ed. Engl. 1993, 32, 1044.
8. (DHQ)2PHAL=dihydroquinine
1,4-phthalazinediyl
diether, available from the Aldrich Chemical Company.
9. General experimental procedure:
Acknowledgements
Thiophenol (1 equiv.) was added dropwise to a solution
of the ketene (1 equiv.) and alkaloid catalyst (0.1 equiv.)
in toluene at −78°C. After 1 h the solvent was evaporated
and the residue dissolved in a small amount of pentane
and filtered to remove insoluble alkaloid catalyst. The
crude thioester products obtained this way were esti-
mated to be >95% pure by 1H and 13C NMR analysis and
were normally used without purification. The
dimethylphenylsilyl series of thioester products are stable
enough to be further purified by flash column chromatog-
raphy without serious losses if necessary.
We thank the Engineering and Physical Sciences
Research Council (EPSRC), and GlaxoWellcome,
Stevenage, for support of C.L.F. and R.J.O. We also
thank Mr. Ivan Perez of the University of Oviedo,
Spain, for exploration of the transformation of 9e into
18 shown in Scheme 4.
References
10. Determination of ee values was carried out using a
Chirapak OD column using small amounts of iPrOH
(typically 0.25%) in hexane as eluant.
11. Crystallographic data have been deposited with the Cam-
bridge Crystallographic Data Centre: CCDC 155471. The
absolute configuration shown follows from refinement of
a Flack parameter [value 0.0(2)], see: Flack, H. D. Acta
Crystallogr., Sect. A 1983, 39, 876.
12. Wynberg, H. In Topics in Stereochemistry; Eliel, E. L.;
Wilen, S. H.; Allinger, N. L., Eds.; John Wiley: New
York, 1986; Vol. 16, p. 87.
13. Anderson, R. J.; Henrick, C. A.; Rosenblum, L. D. J.
Am. Chem. Soc. 1974, 96, 3654.
1. (a) Pracejus, H. Justus Liebigs Ann. Chem. 1960, 634, 9;
(b) For a review, see: Buschmann, H.; Scharf, H.-D.;
Hoffmann, N.; Esser, P. Angew. Chem., Int. Ed. Engl.
1991, 30, 477.
2. For related works, see: (a) Calter, M. A. J. Org. Chem.
1996, 61, 8006; (b) Huang, W.; Moore, P. A.; Rafai Far,
A.; Tidwell, T. T. J. Org. Chem. 2000, 65, 3877.
3. Hodous, B. L.; Ruble, J. C.; Fu, G. C. J. Am. Chem. Soc.
1999, 121, 2637.
4. For reviews of silylketenes, see: (a) Pommier, A.; Kocien-
ski, P.; Pons, J.-M. J. Chem. Soc., Perkin Trans. 1 1998,
2105; (b) Tidwell, T. T. Ketenes; John Wiley & Sons,
New York, 1995; (c) Shiori, T.; Takaoka, K.; Aoyama, T.
J. Heterocycl. Chem. 1999, 1555.
14. Enders, D.; Adam, J.; Klein, D.; Otten, T. Synlett 2000,
1371.
5. Weiguny, J.; Scha¨fer, H. J. Liebigs Ann. Chem. 1994, 235.
6. We initially favoured a method (method A) employed for
the synthesis of phenyltrimethylsilylketene, which
involves a-bromination of an intermediate acid chloride
and then dehalogenation with activated zinc, see: Brady,
W. T.; Cheng, T. C. J. Organomet. Chem. 1977, 137, 287.
This approach is useful for ensuring that volatile or
15. We were able to convert alcohol 16 into an allylsilane,
prepared previously by Enders,18 by Swern oxidation
followed by methylenation, although in very low (ca.
10%) yield. The [h]D (−22.5, c=0.32, CHCl3) for the
small amount (8 mg) of allylsilane prepared was in accord
with the stereochemistry shown for 16. Confirmation of
this result must await further work.

A. J. Blake et al. / Tetrahedron Letters 42 (2001) 2877–2881
2881
16. Fleming, I.; Sanderson, P. E. J. Tetrahedron Lett. 1987,
28, 4229.
17. The prospect of employing reactive ketenes, generated in
situ is realistic, see: Taggi, A. E.; Hafez, A. M.; Wack,
H.; Young, B.; Drury, III, W. J.; Lectka, T. J. Am.
Chem. Soc. 2000, 122, 7831.
18. (a) Bhushan, V.; Lohray, B. B.; Enders, D. Tetrahedron
Lett. 1993, 34, 5067; (b) Enders, D.; Lohray, B. B.;
Burkamp, F.; Bhushan, V.; Hett, R. Liebigs Ann. 1996,
189.
.
.