Enantioselective catalytic allylation of arylmethylketones using tetraallyltin and tin(IV) chloride mixtures

Article abstract of DOI:10.1016/j.jorganchem.2005.11.017

Reaction of SnCl₄ with Sn(CH₂CHCH₂) ₄ in the presence of water and monothiobinaphthol [2,2′-C ₂₀H₁₂(OH)(SH)] allows the formation of a selective catalyst for the allylation of ketones (up t

Full text of DOI:10.1016/j.jorganchem.2005.11.017

Journal of Organometallic Chemistry 691 (2006) 1515–1519
www.elsevier.com/locate/jorganchem
Note
Enantioselective catalytic allylation of arylmethylketones using
tetraallyltin and tin(IV) chloride mixtures
Oscar Prieto, Simon Woodward *
School of Chemistry, The University of Nottingham, University Park, Nottingham NG7 2 RD, United Kingdom
Received 9 November 2005; accepted 9 November 2005
Available online 19 December 2005
Abstract
Reaction of SnCl with Sn(CH CH@CH ) in the presence of water and monothiobinaphthol [2,2 -C H (OH)(SH)] allows the
0
4
2
2 4
20 12
formation of a selective catalyst for the allylation of ketones (up to 94% e.e.).
Ó 2005 Elsevier B.V. All rights reserved.
Keywords: Tin; Asymmetric catalysis; Ketones; Allylation; Autocatalysis
1
. Introduction
pre-catalyst (4a) requires four synthetic steps. We wondered
if we could attain similar reactivity based on in situ forma-
tion of (4b) improving the utility of this approach.
In 2002 we identified an unusual asymmetric reaction
where, in the presence of catalytic amounts of the chiral
ligand MTBH samples of tetraallyltin of 98% chemical
purity ketones were allylated up to nine times more selec-
2
2. Results and discussion
tively than with completely pure Sn(CH CH@CH )
2
2 4
Encouraged by the work of Baba, who added SnCl to
(
>99% purity gives <35% ee for PhC(O)Me) (Scheme 1)
2
modify the reactivity of Bu Sn(CH CH@CHPh) [5], we rea-
[
1]. Subsequent careful work revealed the selective catalyst
3
2
soned that addition of small amounts SnCl to pure (>99%)
is formed from trace amounts of EtSnCl(CH CH@CH )
4
2
2 2
commercially available tetraallyltin should result in the for-
(
3) derived from incomplete reaction of the SnCl and the
4
mation of (4b) through the intermediacy of ClSn(CH CH@
presence of EtBr (used for Grignard preparation) [2]. Com-
pound (3) undergoes facile allyl hydrolysis with trace water
to provide the crystallographically characterized bis(dially-
ldiethyldichlorodistanoxane) (4a). Reaction of (4a) with
2
CH ) . From our previous studies, we knew that active
2 3
catalysts are formed even if only trace amounts of water
are present in such mixtures, however, the issue of stereose-
lectivity from such species was unknown. Therefore, tetraal-
lyltin (1.07 equiv. based on ketone used), traces of water
(
Sa)-1 led to a very selective catalyst which we propose oper-
ates via the working model (5) (Scheme 1). Alternative pro-
posals via mononuclear species, such as (MTB)Sn(allyl)L
species are possible but, in our hands, control reactions with
such species are inactive or unselective. Our system is
unique, not only for its complex reactivity, but because it
is the only tin-based catalyst showing high enantioselectiv-
ity in this process, all other work concentration on tita-
nium-based catalysts [3,4]. Unfortunately, preparation of
(
0.41 equiv.) and chiral ligand (S )-MTBH (20 mol%) were
a 2
equilibrated with 1–3 mol% SnCl (based on ketone used) in
4
toluene at 50 °C to form active catalysts. Neat PhC(O)Me
100 mol%) was added to these solutions at room tempera-
ture and the dependence of product e.e. versus yield by GC
(
over 16 h. With 2–3 mol% of the SnCl promoter selective
4
catalysts were formed but at 1 mol% the insufficient chiral
catalyst was produced to compete with racemic background
*
autocatalysis from the Sn(allyl) [OC-Me(Ph)(allyl)]
2–3) kinetic products [2].
(n =
4ꢀn
Corresponding author.
E-mail address: simon.woodward@nottingham.ac.uk (S. Woodward).
n
0
022-328X/$ - see front matter Ó 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.11.017

1516
O. Prieto, S. Woodward / Journal of Organometallic Chemistry 691 (2006) 1515–1519
Table 1
Influence of different additives on Sn(CH
promoted acetophenone (1 Ar = Ph) allylation
2
CH@CH
2
)
4
/SnCl
4
a 2
/(S )-MTBH
a
Entry
Additive
Time/h
a
Product (S )-2 (Ar = Ph)
b
b,c
Yield/%
e.e./%
a
1
2
3
4
5
6
7
8
9
0
1
2
3
4
a
None
4
2
17
17
2
75
51
90
75
>95
60
63
90
70
>95
90
86
72
50
18
8
12
80
47
26
2
i
d,e
d,e
d,f
d,e
f
Pr OH
Pr OH
Pr OH
PhOH
PhOH
i
i
2
2
g
g
(ꢀ)-Menthol
(ꢀ)-Menthol
17
17
71
17
36
18
18
t
d,e
(Bu O)
4
Sn
d,e
1
1
1
1
1
ClSiMe
MS 4 A
Tetraallylsilane
3
e
˚
40
68
0
g,h
28
10
0
g
(OEt)
3
Si(allyl)
g
Cl Si(allyl)
3
–
Reagent mixture: Sn(CH
2 2 4 4 2 a 2
CH@CH ) /SnCl /H O/(S )-MTBH /PhC-
(
O)Me 1.07:0.03:0.41:0.20:1.00 equiv. The e.e. value maximizedat 86% (75%
yield 2 at 6 h) but was 78% at the maximum conversion attained (95%).
b
Determined by GLC, using undecane as internal standard. using a
c-CD column (see Section 4).
c
Absolute configuration: (S)-2.
Scheme 1.
d
1
equiv. was used (w.r.t. 1).
e
f
Without water.
Water was added (0.41 equiv. w.r.t. 1).
The conversion dependent enantioselectivity of Fig. 1
g
5
a 2 2
mol% of tetraallyltin w.r.t. 1; 20 mol% (S )-MTBH /Sn(CH -
suggested that formation of the selective catalyst is slow
and therefore the presence of additional additives to
further improve the process was sought (Table 1). As iso-
propanol had proved a valuable additive in WalshÕs tita-
nium-based chemistry [4] it was tried here but produced
no significant improvement in the rate or selectivity in
either the presence or absence of water (runs 2–4). Con-
versely, phenol promoted a dramatic increase in the reac-
tion rate but at the loss of any significant stereoselection
ChH@CH
2 4
) .
Reaction run using 5 mol% tetraallyltin.
OPh). This hypothesis is supported by use of (ꢀ)-menthol
which leads to a significant improvement in the stereoselec-
tivity (run 7) at low substrate conversion suggesting it is
incorporated as an O-menthyl group in (5). Unfortunately,
the falling e.e. with increased conversion (run 8) strongly
suggests the menthyl is lost from the catalyst the exchange
(
run 6). We hypothesized that this might be to its exchange
with (allkyl)
SnOCMePh(allyl) product. The need for
3
into the selective transition state (5) as phenoxide (Y =
water in the system was confirmed as alternative catalyst
9
8
8
7
7
6
6
5
5
0
5
0
5
0
5
0
5
0
0
10
20
30
40
50
60
70
80
90
100
Yield
4 2
Fig. 1. Influence of SnCl on MTBH -catalysed asymmetric allylation of acetophenone.

O. Prieto, S. Woodward / Journal of Organometallic Chemistry 691 (2006) 1515–1519
1517
Table 2
Allylation of (1) using Sn(CH
lane as standard; J values are given in Hz. analysis by
GC was performed on a Varian 3380 gas chromato-
graph. Analysis by HPLC was performed using a Hew-
lett-Packard 1100LC machine. Toluene solutions of
a
2
CH@CH
2
)
4
/SnCl
4
/(S
a
)-MTBH
2
Entry
Ar
Time/h
Product (S )-2
a
(
Ar = Ph)
b
b,c
Yield/%
e.e./%
SnCl solutions were freshly prepared using commercial
4
1
2
3
4
5
6
7
8
9
Ph
4-MeC H
6 4
4
3
3
3
3
3
3
3
3
75
47
53
70
68
85
<2
45
8
86
66
55
81
76
94
–
80
3
76
SnCl4 (>98%) and toluene distilled from sodium wire.
The allylated products were identified by comparison
with authentic samples [2].
d
(E)-PhCH@CH
4-ClC
4-BrC
4-NO
4-(MeO)C
2-C10
Propiophenone
6
H
4
6 4
H
4.2. Representative catalytic allylation for acetophenone,
2 6 4
C H
6
H
4
(S)-(ꢀ)-2-phenyl-pent-4-en-2-ol
H
8
Solid (S )-(+)-MTBH (24.0 mg, 0.08 mmol) in a
a
2
10
(S)-(ꢀ)-2-(3-bromophenyl)-
70
Schlenk tube was treated with H O (3 ll, 0.16 mmol)
2
pent-4-en-2-ol
and commercial available tetraallyltin (103 ll, 0.431
a
Reagent mixture: Sn(CH
2
CH@CH
2
)
4
/SnCl
4
/H
2
O/(S
a
)-MTBH
2
/Ph-
mmol) followed by the addition of a 2 or 3% SnCl solu-
4
C(O)Me 1.07:0.03:0.41:0.20:1.00 equiv. The e.e. value maximized at 86%
tion in toluene (8.8 or 13.2 lmol; 1.0 or 1.5 ml) under an
atmosphere of argon. The resulting mixture was stirred at
ambient temperature for 5 min then heated at 52 °C for
(
(
75% yield 2 at 4 h) but was 78% at the maximum conversion attained
95%).
b
Determined by GLC, using undecane as internal standard. using a
c-CD column (see Section 4).
2
h. The mixture was allowed to cool to room tempera-
c
Absolute configuration (S).
Absolute configuration not determined.
ture then acetophenone (47 ll, 0.4 mmol) was and unde-
cane (10 ll) were added. The mixture was stirred up to
16 h. Every sample taken from the reaction mixture was
concentrated and the residue chromatographed on silica
d
preparations using dehydrating agents all led to poor selec-
tivities (runs 9–11). Finally, the possibility of using allylsi-
lane as the terminal allyl source was investigated but rapid
transmetallation to tin could not be attained (runs 12) and
the reactions performed poorly in its absence (runs 13–14).
The optimal reaction procedure was applied to a range
of arylmethyl ketones to define the scope and utility of
the new process (Table 2). Optimal enantioselectivities were
attained after 3–4 h in all cases. Running the reactions for
longer periods resulted in lower e.e. values due to compe-
tion from background reactions. Best results were attained
with an electron deficient ketone (run 6). Electron rich 4-
(hexane then Et O). Yield was determined by GC up to
2
1
90%, e.e. up to 86%. H NMR (CDCl , 400 MHz): d
3
1.57 (s, 3H), 2.12 (s, 1H), 2.53 (dd, J = 13.7 and 8.3 Hz,
plus unresolved couplings, 1H), 2.72 (ddt, J = 13.7, 6.4
and 1.1 Hz, 1H), 5.12–5.14 (m, 1H), 5.15–5.18 (m, 1H),
5.64 (dddd, J = 14.7, 10.2, 8.3 and 6.4 Hz, 1H), 7.26 (tt,
J = 7.3 and 1.3 Hz 1H), 7.34–7.38 (m, 2H), 7.44–7.47
(m, 2H). These data were concordant with published val-
ues [2]. The enantiomeric excesses were determined by GC
(6-Me-2,3-pe-c-CD, 100 °C isothermal). Retention times
20.1. (R), 21.4 (S).
(
MeO)C H C(O)Me was inactive in this chemistry.
6 4
3
. Conclusions
4.3. (S)-(ꢀ)-2-(4-bromophenyl)-pent-en-2-ol
The possibility of using a simple achiral additive to
Prepared by the allylation of 4-bromoacetophenone.
Yield (after 3 h) 68% (76% e.e.); H NMR (CDCl3,
1
strongly promote the stereoseletivity observed in tin-based
asymmetric ketone allylation has been demonstrated. Dra-
matically improved e.e. values are attained without the
need to prepare all the individual components of the mix-
tures of organotin species required for this chemistry.
The presence of background racemic allylation, promoted
by the kinetic product of the reaction dictates that the
ligand accelerated catalysis brought about by the chiral
ligand has to be great. A search is continuing to identify
400 MHz): d 1.53 (s, 3H), 2.12 (s, 1H), 2.49 (ddt,
J = 13.7, 8.2 and 0.8 Hz, 1H), 2.64 (ddt, J = 13.7, 6.6
and 1.0 Hz, 1H), 5.10–5.13 (m, 1H), 5.15–5.18 (m, 1H),
5.60 (dddd, J = 14.7, 9.7, 8.2 and 6.6 Hz, 1H), 7.31 (dd,
J = 8.7 and 2.0 Hz 2H), 7.46 (dd, J = 8.7 and 2.0 Hz,
2H). These data were concordant with published values
[2]. The enantiomeric excesses were determined by GC (6-
ꢀ
1
Me-2,3-pe-c-CD, 90 °C initial, 4 deg min
Retention times 42.0 (R), 43.1 (S).
to 125 °C).
a ligand with superior properties to MTBH as in this case
2
a technically very simple process would be attained.
4.4. (S)-(ꢀ)-(4-tolyl)-pent-4-en-2-ol
4
. Experimental
Prepared by the allylation of 4-methylacetophenone.
1
4
.1. General
Yield (after 3 h) 47% (66% e.e.); H NMR (CDCl3,
00 MHz): d 1.55 (s, 3H), 2.04 (s, 1H), 2.36 (s, 3H), 2.50
4
Proton spectra were recorded on Bruker (AM 400)
spectrometer at ambient temperature using tetramethylsi-
(dd, J = 13.7 and 8.2 Hz, plus unresolved couplings, 1H),
2.69 (dd, J = 13.7 and 6.4 Hz, plus unresolved couplings,

1518
O. Prieto, S. Woodward / Journal of Organometallic Chemistry 691 (2006) 1515–1519
1
H) 5.11–5.13 (m, 1H), 5.14–5.17 (m, 1H), 5.64 (dddd,
J = 14.7, 10.1, 8.2 and 6.4 Hz, 1H), 7.17 (d, J = 8.2 Hz,
H), 7.34 (d, J = 8.2 Hz, 2H). These data were concordant
and 1.0 Hz, 1H), 5.10–5.15 (m, 1H), 5.15–5.16 (m, 1H),
5.64 (dddd, J = 15.8, 11.4, 8.1 and 6.6 Hz, 1H), 7.21 (t,
J = 7.9 Hz, 1H), 7.36 (dd, J = 7.9, 1.9 and 1.0 Hz, 1H),
overlapped by 7.37 (ddd, J = 7.9, 1.9 and 1.1 Hz, 1H),
7.63 (t, J = 1.9 Hz, 1H) These data were concordant with
published values [2]. The enantiomeric excesses were deter-
mined by GC (Cyclodex-B, 120 °C isothermal). Retention
times 130.1 (S), 135.6 (R).
2
with published values [2]. The enantiomeric excesses were
determined by GC (6-Me-2,3-pe-c-CD, 90 °C initial for
ꢀ
1
5
2
min, 3 deg min to 110 °C). Retention times 26.2 (R),
7.2 (S).
4
.5. 3-Methyl-1-phenyl-hexa-1,5-dien-3-ol
4.9. (S)-(ꢀ)-2-(2-naphthyl)-pent-4-en-2-ol
Prepared by the allylation of benzylidene acetone.
1
Yield (after 3 h) 53% (55% e.e.); H NMR (CDCl3,
4
Prepared by the allylation of 2-acetonaphthone. Yield
1
00 MHz): d 1.38 (s, 3H), 1.81 (s, 1H), 2.36 (ddt,
J = 13.6, 8.1 and 0.8, 3H), 2.44 (ddt, J = 13.6, 6.7 and
.1 Hz, 1H), 5.13–5.16 (m, 1H) 5.17–5.18 (m, 1H), 5.84
dddd, J = 16.4, 10.8, 8.1 and 6.7, 1H), 6.29 (d,
J = 16.1 Hz, 1H), 6.59 (d, J = 16.1 Hz, 1H), 7.22 (tt,
J = 7.3 and 1.4 Hz, 1H), 7.28–7.33 (m, 2H), 7.36–7.39
m, 2H). These data were concordant with published val-
ues [2]. The enantiomeric excesses were determined by GC
2,6-Me-3-pe-c-CD, 70 °C initial for 25 min, 1 deg min
to 120 °C). Retention times 110.3, 113.2.
(after 3 h) 45% (80% e.e.); H NMR (CDCl , 400 MHz):
3
d 1.66 (s, 3H), 2.23 (s, 1H), 2.61 (dd, J = 13.8, 8.4 Hz, plus
unresolved couplings 1H), 2.82 (dd, J = 13.8, 6.4 Hz, plus
unresolved couplings 1H), 5.14 (d, J = 10.1 Hz, plus unre-
solved couplings, 1H), 5.18 (d, J = 14.7 Hz, pus unresolved
couplings, 1H), 5.64 (dddd, J = 14.7, 10.1, 8.4 and 6.4 Hz,
1H), 7.45–7.52 (m, 2H), 7.56 (dd, J = 8.6 and 1.9 Hz, 1H),
7.82–7.88 (m, 3H), 7.94 (d, J = 1.7 Hz, plus unresolved
long range couplings 1H). These data were concordant
with published values [2]. The enantiomeric excesses were
determined by Diacel HPLC (OD column, 100:0–98:2 hex-
1
(
(
ꢀ
1
(
ꢀ1
4
.6. (S)-(ꢀ)-2-(4-chlorophenyl)-pent-4-en-2-ol
Prepared by the allylation of 4-chloroacetophenone.
ane:IPA over 35 min; flow rate 0.5 ml min ). Retention
times 53.4 (S), 65.2 (R).
1
Yield (after 3 h) 70% (81% e.e.); H NMR (CDCl3,
00 MHz): d 1.54 (s, 3H), 2.05 (s, 1H), 2.49 (ddt,
4.10. (S)-(ꢀ)-2-(4-nitrophenyl)-pent-en-2-ol
4
J = 13.7, 8.3 and 0.8 Hz, 1H), 2.65 (ddt, J = 13.7, 6.5
and 1.1 Hz, 1H), 5.10–5.14 (m, 1H), 5.15–5.18 (m, 1H),
Prepared by the allylation of 4-nitroacetophenone. Yield
1
(after 3 h) 85% (94% e.e.); H NMR (CDCl , 400 MHz): d
3
5
.60 (dddd, J = 14.8, 9.6, 8.3 and 6.6 Hz, 1H), 7.31 (d,
1.58 (s, 3H), 2.22 (s, 1H), 2.54 (dd, J = 13.8 and 8.1 Hz,
plus unresolved couplings, 1H), 2.68 (dd, J = 13.8 and
6.6 Hz, plus unresolved couplings 1H), 5.13–5.16 (m,
1H), 5.17–5.18 (m, 1H), 5.59 (dddd, J = 16.3, 10.8, 8.1
and 6.6 Hz, 1H), 7.62 (dd, J = 9.0 and 2.0 Hz, 2H), 8.19
(dd, J = 9.0 and 2.0 Hz, 2H). These data were concordant
with published values [2]. The enantiomeric excesses were
determined by Diacel HPLC (AD column, 95:5 hex-
J = 8.8 Hz, 2H), 7.38 (d, J = 8.8 Hz, 2H). These data were
concordant with published values [2]. The enantiomeric
excesses were determined by GC (6-Me-2,3-pe-c-CD,
ꢀ
1
9
0 °C initial, 3 deg min to 120 °C). Retention times 32.9
(
R), 34.2 (S).
4
.7. 3-Phenyl-hex-5-en-3-ol
ꢀ1
ane:IPA; flow rate 0.5 ml min ). Retention times 15.9
Prepared by the allylation of propiophenone. Yield
(S), 25.2 (R).
1
(
after 3 h) 8% (3% e.e.); H NMR (CDCl , 400 MHz): d
3
0
2
.78 (t, J = 7.4 Hz, 3H), 1.80–1.91 (m, 2H), 2.04 (s, 1H),
.51 (ddt, J = 13.8, 8.6 and 0.8 Hz, 1H), 2.74 (ddt,
4.11. 2-(4-Methoxyphenyl)-pent-4-en-2-ol
J = 13.8, 6.1, 1.4 Hz, 1H), 5.09–5.17 (m, 2H), 5.59 (dddd,
J = 14.8, 10.1, 8.6, and 6.1 Hz, 1H), 7.24 (tt, J = 7.2 and
Prepared by the allylation of 4-methoxyacetophenone.
Yield (after 3 h) <2% (e.e. no determined); H NMR
1
1
.4 Hz, 1H), 7.32–7.37 (m, 2H), 7.39–7.42 (m, 2H). These
(CDCl , 400 MHz): d 1.53 (s, 3H), 1.99 (s, 1H), 2.48 (dd,
3
data were concordant with published values [2]. The enan-
J = 13.7 and 8.1 Hz, plus unresolved couplings 1H), 2.66
(dd, J = 13.7 and 6.6 Hz, plus unresolved couplings 1H),
3.81 (s, 3H) 5.10 (apparent s, 1H), 5.14 (m, 1H), 5.63 (dddd,
J = 14.7, 10.4, 8.1 and 6.6 Hz, 1H), 6.87 (dd, J = 8.8 and
tiomeric excesses were determined by GC (6-Me-2,3-pe-c-
ꢀ1
CD, 90 °C initial, 2 deg min to 100 °C). Retention times
2
7.7, 28.6.
2.1 Hz, 2H), 7.36 (dd, J = 8.8 and 2.1 Hz, 2H). These data
4
.8. (S)-(ꢀ)-2-(3-bromophenyl)-pent-4-en-2-ol
were concordant with published values [2].
Prepared by the allylation of 3-bromoacetophenone.
Yield (after 3 h) 70% (73% e.e.); H NMR (CDCl3,
Acknowledgements
1
4
00 MHz): d 1.54 (s, 3H), 2.02 (s, 1H), 2.50 (ddt,
We thank EPSRC for an Award (GR/R69051/01) and
The European Science Foundation for support through
J = 13.8, 8.1 and 0.9 Hz, 1H), 2.65 (ddt, J = 13.8, 6.6

O. Prieto, S. Woodward / Journal of Organometallic Chemistry 691 (2006) 1515–1519
1519
the COST-D24 Programme. We thank a referee for point-
ing out the possibility of catalysis by (MTB)Sn(allyl)R
species.
[2] A. Cunningham, V. Mokal-Parekh, C. Wilson, S. Woodward, Org.
Biol. Chem. (2004) 728–741.
[
[
[
3] S. Kii, K. Maruoka, Chirality 15 (2003) 68–70;
H. Hanawa, S. Kii, K. Maruoka, Adv. Synth. Catal. 343 (2001) 57–60.
4] J.G. Kim, K.M. Waltz, I.F. Garcia, D. Kwiatkowski, P.J. Walsh, J.
Am. Chem. Soc. 126 (2004) 12580–12585.
5] M. Yasuda, K. Hirata, M. Nishino, A. Yamamoto, A. Baba, J. Am.
Chem. Soc. 124 (2002) 13442–13447.
References
[
1] A. Cunningham, S. Woodward, Synlett (2002) 43–44.