Coordination of (β-N-sulfonylaminoalkyl)phosphines and their analogous arsines to PdII and PtII. Application of the Pd-complexes as chiral catalysts in asymmetric hydrosilylation of 1,3-dienes

Article abstract of DOI:10.1039/b101464l

Optically active ligands 2 and 3 were synthesised from phenylglycine and the coordination of ligand 2 to PdCl₂(CH₃CN)₂ and PtCl₂(C₆H₅CN)₂ was studied with ¹H, ³¹P and ¹⁵N-¹H-HMQC NMR spectroscopy. The results indicated P,O rather than P,N bidentate coordination. Both ligands were tested in the palladium catalysed hydrosilylation of cyclic 1,3-dienes. Asymmetric induction of up to 84% (≤25% yield), which is the hitherto highest reported ee value for asymmetric hydrosilylation of 1,3-dienes, was achieved as measured on the allylic alcohols formed after ethanolysis-oxidation of the initially formed allylic silanes.

Full text of DOI:10.1039/b101464l

View Article Online / Journal Homepage / Table of Contents for this issue
Coordination of (ꢀ-N-sulfonylaminoalkyl)phosphines and
II
II
their analogous arsines to Pd and Pt . Application of the
Pd-complexes as chiral catalysts in asymmetric hydrosilylation
of 1,3-dienes†
1
Magnus Gustafsson, Karl-Erik Bergqvist and Torbjörn Frejd*
Organic Chemistry, Centre for Chemistry and Chemical Engineering, Lund University,
P. O. Box 124, S-22100 Lund, Sweden. E-mail: torbjorn.frejd@orgk1.lu.se
Received (in Cambridge, UK) 14th February 2001, Accepted 20th April 2001
First published as an Advance Article on the web 24th May 2001
Optically active ligands 2 and 3 were synthesised from phenylglycine and the coordination of ligand 2 to
PdCl2(CH3CN)2 and PtCl2(C6H5CN)2 was studied with 1H, 31P and 15N–1H-HMQC NMR spectroscopy. The
results indicated P,O rather than P,N bidentate coordination. Both ligands were tested in the palladium catalysed
hydrosilylation of cyclic 1,3-dienes. Asymmetric induction of up to 84% (≤25% yield), which is the hitherto highest
reported ee value for asymmetric hydrosilylation of 1,3-dienes, was achieved as measured on the allylic alcohols
formed after ethanolysis–oxidation of the initially formed allylic silanes.
Introduction
In connection with projects in natural product synthesis1,2 we
became interested in the synthesis and use of allylic silanes
generated via 1,4-addition of silanes to 1,3-dienes. For unsym-
metrical 1,3-dienes both regiochemistry and stereochemistry
must be controlled. Although asymmetric allylic silanes in
which the silicon is attached to a stereogenic carbon center are
useful building blocks in organic synthesis they are rarely found
in the literature. Despite many formal similarities between the
catalytic version of hydrogenation and hydrosilylation, the latter
reaction has received much less attention. Little is known about
the mechanism and how the stereochemical information is trans-
ferred from the ligand to the substrate, although suggestions
regarding the mechanism in palladium catalysed hydrosilyl-
ation of dienes have been made.3–5 Bidentate bisphosphine
ligands such as BINAP have been shown to be less efficient in
the palladium catalysed asymmetric hydrosilylation due to the
formation of an inactive PdL intermediate.6 Instead, different
monodentate or potentially heterobidentate ligands have been
used such as 2-methoxy-2Ј-diphenylphosphino-1,1Ј-binaphthyl
(MOP),7 2-methoxy-2Ј-diphenylphosphino-1,1Ј-biphenanth-
renyl (MOP-phen),4 (β-N-sulfonylaminoalkyl)phosphine,8 phos-
phetanes and [2-(diphenylphosphino)ferrocenyl]ethylamine
(PPFA) (Fig. 1).3,9,10
Hayashi et al. have systematically studied the asymmetric
hydrosilylation of dienes, but hitherto the yields and degree of
asymmetric induction have not been satisfactory.10–12 In 1990
and 1995 Achiwa et al. reported that different (β-N-sulfonyl-
aminoalkyl)phosphine ligands derived from amino acids
showed promising activity in the hydrosilylation and the Heck
reaction.8,13 Inspired by these results and our interest in the
synthesis and applications of new amino acids14–17 we investi-
gated these aspects further. In this paper we wish to report our
results on the coordination of some new (β-N-sulfonylamino-
alkyl)phosphines to Pd and Pt, as studied by NMR-
spectroscopy, and the efficiency of the Pd-complexes of the
sulfamido phosphines and their arsenic analogues as catalysts
in the asymmetric hydrosilylation of cyclic dienes.
Fig. 1 Some monodentate and potentially heterobidentate ligands
previously used in asymmetric hydrosilylations of alkenes and dienes.
Results and discussion
Ligands 2 and 3 were synthesised in two steps from (S)-phenyl-
glycinol (Scheme 1). Reaction of the aminoalcohol with
methanesulfonyl chloride and triethylamine in dichloro-
methane afforded 1 in 88% yield as a white solid. Treatment of
1 with LiPPh2 (generated from diphenylphosphine and butyl-
lithium)13 or LiAsPh2 (generated from triphenylarsine and
lithium metal)18 at Ϫ30 ЊC gave ligands 2 and 3 in 40% and 50%
yield, respectively. In general the yield of ligand 3 was higher,
probably owing to the better solubility of ligand 3 compared to
ligand 2, which facilitated the isolation. Two equivalents of
LiPPh2 or LiAsPh2, respectively, were needed to get a reason-
able yield due to competitive deprotonation of the acidic sulfon-
amide proton. It is not likely that aziridine 4 (synthesized from
1 by treatment with NaH in THF) was an intermediate in the
ligand synthesis, because when 4 was reacted with LiPPh2 in
THF a low yield of 2 and an unknown product were formed.
It is known from the literature that a monoalkylated
aminophosphine ligand, (p-CH3C6H4)2PCH2CH(iPr)NHCH2-
(p-OCH3C6H4), derived from (S)-valine coordinates to Pd and
† Electronic supplementary information (ESI) available: NMR spectra.
See http://www.rsc.org/suppdata/p1/b1/b101464l
1452
J. Chem. Soc., Perkin Trans. 1, 2001, 1452–1457
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b101464l

View Article Online
Fig. 2 31P-NMR spectra of different mixtures of PdCl2(CH3CN)2 and
ligand 2.
Scheme 1 Reagents and conditions: i) MeSO2Cl, Et3N, CH2Cl2; ii)
LiAsPh2, THF, Ϫ30 ЊC; iii) LiPPh2, THF, Ϫ30 ЊC; iv) NaH, THF, rt.
1
equivalents of the ligand per metal atom. Both the H-NMR
spectrum and 31P-NMR spectrum showed in this case that the
ratio (the sum of the integrals of the resonances at 21.6 and 9.7
ppm compared to that of Ϫ22.8 ppm) between complex and
free ligand was almost 1 : 1.
Pt in a bidentate fashion.19 When the nitrogen atom binds to
the metal a more or less labile stereogenic center is created at
nitrogen.20 This transfer of chirality from the backbone of the
ligand closer to the metal center can be useful in asymmetric
catalysis as shown by Anderson et al.21 In the palladium
catalysed allylic substitution these authors showed that an
introduction of a potential stereogenic center on nitrogen in
P,N chelating ligands affected the absolute configuration of the
product. To our knowledge the coordination of the sulfon-
amide group to transition metals in organic solvents has been
investigated to a lesser extent, although a few reports regarding
N-sufonylamino acids and their coordination to metals have
been published.22–25
Pregosin et al. have shown that in sp2-hybridised nitrogen
ligands the 15N-shift moves 40 ppm downfield upon coordin-
ation to a transition metal, in this case platinum.26,27 We there-
fore recorded a 15N–1H-HMQC spectrum of the sample
containing 3–4 equiv. of the ligand to PdCl2(CH3CN)2, which
revealed three nitrogen–proton correlations well in agreement
with the 31P-NMR data mentioned. 15N-resonances from the
free ligand appeared at 84.5 ppm (J(N,H) = 87 Hz) and from the
complex at 80.6 ppm (J(N,H) = 89 Hz) and 87.4 ppm (J(N,H) = 88.0
Hz) (Fig. 3, spectrum b). The small shift difference between free
ligand and the complexed ligands indicated that the nitrogens
may not be directly coordinated to Pd in these complexes.
For comparison the corresponding Pt-complexation was
examined. Thus, when one equivalent of PtCl2(PhCN)2 and one
equivalent of ligand 2 were mixed and stirred in CH2Cl2 at rt for
one hour a complex was formed showing only one 31P signal at
Since all attempts to grow crystals suitable for X-ray crystal-
lographic structure determination failed, our ligand–metal
complexes were investigated by NMR-spectroscopy.
NMR spectroscopy
The complex formed on mixing one equivalent of PdCl2-
(CH3CN)2 and one equivalent of ligand 2 in CDCl3 showed
31P-NMR signals at 23.2, 21.6 and 9.7 ppm which integrated as
0.31, 0.42 and 0.27, respectively, at T = 290 K (Fig. 2, spectrum
I). No free ligand (Ϫ22.8 ppm) was observed. When the sample
was heated at 320 K for 30 min an apparently irreversible pro-
cess took place. The peak at 23.2 ppm increased at the expense
of the other two and the ratios were now 0.57 to 0.20 to 0.23,
respectively. Further heating resulted in one dominant peak at
23.2 ppm (Fig. 2, spectrum II). In order to gain some inform-
ation about the composition of the complexes (monomeric,
dimeric, etc.) a few control experiments were performed. When
PdCl2(CH3CN)2 was used in excess (2 equivalents) a new signal
appeared at 23.7 ppm. The integrals were 0.82 (23.7 ϩ 3.2 ppm)
to 0.13 (21.6 ppm) to 0.05 (9.7 ppm) (Fig. 2, spectrum III).
When ligand 2 was in excess (2 equivalents) the resonances at 23
ppm decreased and the peaks at 21.6 and 9.7 ppm increased.
When further ligand was added (4 equivalents) the resonance at
23 ppm disappeared. The signals at 21.6 ppm (0.24) and 9.7
ppm (0.33) and free ligand (not shown) at Ϫ22.8 ppm (0.43)
remained (Fig. 2, spectrum IV). When a racemate of the ligand
was added to PdCl2(CH3CN)2 in a 1 : 1 ratio the signal at 9.7
ppm was split into two, indicating the presence of diastereo-
meric complexes, possibly containing two ligands in the same
complex. The signal at 21.6 ppm was not a doublet but was
broadened. A further indication that the complexes with reson-
ances at 9.7 and 21.6 ppm may carry two ligands e.g. PdL2 was
indicated by the integrals in the NMR sample containing 4
1
5.7 ppm. The J(Pt,P) was 2526 Hz in good agreement with
known trans Pt-bisphosphine complexes.28 The 15N-shift, 79.7
ppm (J(N,H) = 88 Hz), of our complex was close to the shift of
the free ligand (84.5 ppm) indicating also in this case that the
ligand probably did not coordinate to the metal via its nitrogen
atom. The 1H-set for the Pt-complex was very similar to that of
the Pd-complex having 31P at 9.7 ppm.
In the light of these results, we suggest that when one equiv-
alent of PdCl2(CH3CN)2 and one equivalent of ligand 2 were
mixed, three complexes were formed; two PdL2 complexes (21.6
ppm and 9.7 ppm) and one PdL complex (23 ppm) (Fig. 4). In
the two PdL2 complexes the ligands should coordinate in a
monodentate fashion via the phosphorus atom. Compared with
the platinum complex (trans-PtL2) it seems reasonable to
assume that the PdL2 at 9.7 ppm is the trans complex, while the
21.6 ppm signal belongs to the cis complex. The complexes
having resonances at 23 ppm may coordinate in a bidentate
fashion via P and O of the ligand. A P,N bidentate coordination
is not likely due to the fact that the 15N-shift differences for the
PdL complex and free ligand was only 1–2 ppm (Fig. 3, spec-
trum a and b). The complex showing a 31P-signal at 23.7 ppm is
probably a kinetically formed complex, which on heating
converts to a new complex having a 31P-signal at 23.2 ppm.
Another observation worth mentioning is that the signals
in both H and 31P NMR spectra are quite broad in the PdL
complex compared to the PdL2 complexes. If the PdL complex
1
J. Chem. Soc., Perkin Trans. 1, 2001, 1452–1457
1453

View Article Online
Fig. 4 Tentative coordination modes of complexes formed on mixing
PdCl2(CH3CN)2 and ligand 2.
Fig. 3 15N–1H-HMQC spectra of the complexes corresponding to
spectra II (a) and IV (b) in Fig. 2.
Scheme 2 Reagents: i) HSiMeCl2, PdL 2 mol%; ii) EtOH, Et3N; iii)
coordinates in a bidentate fashion via the phosphorus and the
nitrogen atoms, a five-membered ring would be formed, while
coordination via the phosphorus and oxygen atoms would form
a seven-membered ring. The broad signals and lack of well
defined coupling patterns may be further evidence in favour of
the seven-membered P,O coordination mode. One would expect
a five-membered arrangement to have a more fixed structure
leading to sharper lines in its NMR-spectrum than a seven-
membered one.
KF, H2O2.
halogenated silanes were favoured due to the reported too weak
coordination of trialkylsilanes to the Pd center to allow good
yields.31
Our first choice was HSiMe2Cl, which is easily handled and
not very sensitive towards moisture and air. When we used this
silane together with PdCl2(CH3CN)2 and ligand 2 as catalyst in
the hydrosilylation of cyclohexa-1,3-diene the yield was rather
low, an average of 15% and maximum 25%, but the asymmetric
induction was quite efficient resulting in the highest ee reported
(84%) in asymmetric hydrosilylation of 1,3-dienes (entry 1).
The low yield was probably due to the observed precipitation of
Pd metal. In entries 2 and 3 a dihalogenated silane, HSiMeCl2,
was used. In these cases precipitation of Pd metal was not
noticed. The yield was now over 80%, but the degree of asym-
metric induction was lower than in entry 1, 62% ee at 20 ЊC and
72% ee at 0 ЊC. When the arsenic ligand 3 was used the ee was
even lower, 40% (entry 5). No asymmetric induction was
achieved when HSiCl3 was used, entry 4. It has been reported
that phenyl groups in the hydrosilane may play an important
role in the induction.6 We therefore tried HSiPhCl2 but the ee
now dropped to 34%, entry 6.
Hydrosilylation
Asymmetric 1,4-hydrosilylation of dienes followed by oxidation
may provide an interesting route to optically active allylic alco-
hols since it has been shown that oxidation of optically active
silanes gives alcohols with complete retention of configur-
ation.29,30 A few such examples have been demonstrated
although not with more substituted dienes.3,4,6,8,10,13 During our
studies in natural product synthesis we needed optically active
cyclic allylic alcohols and we therefore reasoned that the men-
tioned route via allylic silanes could be useful. Moreover, the
prognosis of chromatographic determination of the enantio-
meric composition was judged better for the alcohols than for
the corresponding silanes. Thus, we chose to measure the ees of
the alcohols. Scheme 2 and Table 1 summarises the palladium
catalysed hydrosilylation of cyclohexa-1,3-diene 5, 1-butylcyclo-
hexa-1,3-diene 8, 2-methylcyclohexa-1,3-diene 11 and cyclo-
pentadiene 14. Several types of hydrosilanes were used but
A considerable adverse effect on both the ee and yield
resulted when 1-butylcyclohexa-1,3-diene was used. In fact,
when ligand 2 was applied no conversion of the starting
material was achieved even after 72 h (entry 7). On the other
1454
J. Chem. Soc., Perkin Trans. 1, 2001, 1452–1457

View Article Online
Table 1 Results of the hydrosilylation–ethanolysis–oxidation sequence of dienes
Diene/
Silane
Yields of
alkoxysilanes (%)
Ee (%), c absolute
Entry
ligand a
conditions b
configuration d of alcohols
1
2
3
4
5
6
7
8
9
5/2
5/2
HSiMe2Cl
HSiMeCl2/rt
HSiMeCl2
HSiCl3
HSiMeCl2
HSiPhCl2
HSiMeCl2/50 h
HSiMeCl2/50 h
HSiMeCl2
HSiMeCl2
HSiMeCl2
HSiMeCl2
6 <25
6 >80
6 >80
6 >80
6 >80
6 60
7 84 (S)
7 62 (S)
7 72 (S)
7 0
5/2
5/2
5/3
7 40 (S)
7 34 (S)
10 —
5/2
8/2
9 —
8/3
9 70
10 14 e
11/2
11/3
14/2
14/3
12 >85
12 >85
15 >85
15 >85
13 73 (S)
13 53 (S)
16 37 (S)
16 28(S)
10
11
12
a Ligand in the Pd-complex used as the catalyst 2 mol%. b Reaction time: 40 h, 0 ЊC unless otherwise indicated. c Determined by GC analysis on an
alpha-DEX column. d Determined by comparison of optical rotation data. e Not determined.
hand and quite unexpectedly, with the use of the Pd-complex of
ligand 3 as catalyst, a good yield of 9 was obtained (70%),
although the ee was only 14% (entry 8). This product seems
to have been formed via a 1,2-addition, which is rather unusual
for a Pd-catalysed hydrosilylation of 1,3-dienes where syn
1,4-addition is the normal outcome.4,31 An alternative route
would be a 1,4-addition followed by an isomerisation, which at
this stage cannot be excluded.
A methyl substituent in the 2-position as for 2-methyl-
cyclohexa-1,3-diene was better tolerated. In this case a 1,4-
addition occurred and both the yield and degree of induction
were relatively high, over 80% yield and 73% ee. Cyclopentadi-
ene gave much lower ees, 37% and 28% for 2 and 3, respectively,
than with the similar ligand used by Achiwa et al., see Fig. 1.8,13
Other metal complexes of ligands 2 and 3 based on Rh2Cl2-
(COD)2, PtCl2Ph(CN)2 and NiCl2 have so far given products
with ees close to 0% in the reactions of cyclohexadiene,
although the yields were fairly good, 50–70%.
mm. After elution, the TLC plates were visualized with UV
light followed by spraying with a solution of p-methoxy-
benzaldehyde (26 mL), glacial acetic acid (11 mL) concentrated
sulfuric acid (35 mL), and 95% ethanol (960 mL) followed by
heating. All solvents were dried over 4 Å MS for 24 h prior to
use, unless otherwise mentioned. The molecular sieves were
activated at 400 ЊC for 6 h and then allowed to cool under
argon. Cyclopentadiene 14 was prepared via thermal cracking
of the corresponding dimer. 2-Methylcyclohexa-1,3-diene 1132
and 1-butylcyclohexa-1,3-diene 833 were synthesised as
described by Bäckvall et al. All other reagents were used as
received.
(S)-2-Methylsulfonylamino-2-phenyl-1-methylsulfonyloxyethane
1
A solution of methanesulfonyl chloride (2.92 g, 25.5 mmol) in
CH2Cl2 (10 mL) was added to a mixture of (S)-(ϩ)-2-phenyl-
glycinol (1.0 g, 7.3 mmol, 98%) and triethylamine (2.65 g,
26.6 mmol) in CH2Cl2 (50 ml) at 0 ЊC under an argon atmos-
phere. The resulting solution was stirred at rt for 15 h where-
after saturated NaHCO3 (30 ml) was added and the reaction
mixture was worked up as follows: extraction with ethyl acetate
(3 × 50 ml), washing of the collected organic extracts with brine
(2 × 30 ml) followed by drying (Na2SO4), filtration and removal
of the solvent under reduced pressure. The residue, a yellow oil
was column chromatographed (SiO2, heptane–ethyl acetate
1 : 1) to give the title compound (1.89 g, 88%) as a white solid;
mp 105–107 ЊC; [α]Dϩ47.5 (c 0.3, CHCl3); δH 2.76 (3 H, s), 3.02
(3 H, s), 4.46–4.34 (2 H, m), 4.91–4.81 (1 H, m), 5.64 (1 H, d,
J 7.2), 7.46–7.33 (5 H, m); δC 37.9, 42.0, 56.9, 71.0, 127.0, 129.1,
129.3, 136.3; HRMS (FABϩ): (M ϩ H); Found: m/z, 294.0481.
C10H15O5NS2 requires m/z, 294.0470.
Conclusion
The (β-N-sulfonylaminoalkyl)phosphine ligand 2 coordinated
to Pd in several coordination modes: both by monodentate
phosphorus coordination and by bidentate coordination via the
phosphorus and oxygen atoms. The mixture of complexes
formed by combining ligand 2 and PdCl2(CH3CN)2 in the ratio
1 : 1 was used as catalyst in the asymmetric hydrosilylation of
1,3-dienes to give optically active allylic silanes, which were
oxidised to give the corresponding optically active allylic
alcohols. The presence of an aromatic substituent at the 2-
position of the ligand backbone improved the ee compared
with other ligand structures reported in the literature.
Exchanging the phosphorus for arsenic in the ligands had a
positive effect only for one substrate, entry 8. Further investi-
gations regarding the hydrosilylation and its mechanistic
aspects will be reported in due course.
General procedure for the synthesis of ligands 2 and 3
A solution of 1 (one equivalent) in THF (5 ml mmolϪ1) was
added to a solution of either LiPPh213 or LiAsPh218 (two equiv-
alents) in THF (10 ml mmolϪ1) at Ϫ30 ЊC under an argon
atmosphere. The resulting mixture was stirred at Ϫ30 ЊC for
20 h whereafter saturated NaHCO3 was added followed by
extraction with degassed toluene (or with ethyl acetate in the
synthesis of ligand 2 due to the low solubility of 2 in toluene).
The collected organic extracts were washed with water and
brine followed by drying (Na2SO4), filtration and removal of
the solvent under reduced pressure. The residue, a slightly
yellow oil, was column chromatographed (SiO2, toluene–ethyl
acetate 4 : 1) to give ligands 2 and 3 as white solids. All oper-
ations, including the chromatography, were performed under a
nitrogen gas atmosphere and using degassed solvents and solu-
tions in order to avoid oxidation. Instead of elemental analyses
we used HRMS due to the high risk of oxidation of the ligands.
Experimental
GC analyses were performed with either an alpha- or beta-
DEX column (Supelco, 30 m × 0.25 mm id × 0.25 µm film
thickness). NMR spectra were recorded at 400 MHz (Bruker
DRX NMR spectrometer), unless otherwise stated. Chemical
1
shift data are in ppm, referenced to external TMS for H and
13C, H3PO4 for 31P and CH3NO2 for 15N. The coupling constants
were measured in Hertz. Optical rotations were measured with
a Perkin-Elmer 241 LC polarimeter at 23 ЊC and are given in
10Ϫ1 deg cm2
g
Ϫ1. Preparative chromatographic separations
were performed on Matrex Amicon normal phase silica gel 60
(0.035–0.070 mm). Thin-layer chromatography was performed
on Merck precoated TLC plates with Silica gel 60 F-254, 0.25
J. Chem. Soc., Perkin Trans. 1, 2001, 1452–1457
1455

View Article Online
[(2S)-2-(Methylsulfonylamino)-2-phenylethyl]diphenylphos-
phine (؉)-2. Yield 40%. δH 2.56 (3 H, s), 2.58 (1 H, ddd, J 13.4,
6.6 and 1.0), 2.68 (1 H, ddd, J 13.9, 8.6 and 1.2), 4.53 (1 H, m),
5.20 (1 H, d, J 5.4), 7.50–7.15 (15 H, m); δC 38.0 (d, J 14.5),
42.0, 56.4 (d, J 17.8), 126.9, 128.4, 128.9 (d, J 7), 129.2, 129.4
(d, J 7), 133.0 (d, J 15.1), 133.1 (d, J 15.1), 141.5 (d, J 5);
δP Ϫ22.8; mp 163–165 ЊC; [α]D ϩ18.6 (c 0.22, CHCl3); HRMS
(FABϩ) Found: m/z, 383.1113. C21H22O2NSP requires m/z,
383.1109.
optical rotation data known from the literature. Absolute
configuration of 10 has not yet been determined.
3-(Diethoxymethylsilyl)cyclohexene 6, cyclohex-2-en-1-ol 7,
2-methylcyclohex-2-en-1-ol 13, 3-(diethoxymethylsilyl)cyclo-
pentene 15, and cyclopent-2-en-1-ol 16 had NMR spectra
in agreement with those reported.13,35,36 1-Butyl-4-(diethoxy-
methylsilyl)cyclohexene 9 was oxidized directly without work-
up to the corresponding alcohol 10.
4-Butylcyclohex-3-en-1-ol 10. [α]D ϩ12.5 (c 0.12, CDCl3);
δH 0.91 (1 H, t, J 7.2), 2.0–1.3 (13 H, m), 4.19 (1 H, m), 5.50
(1 H, m); δC 14.2, 19.3, 22.6, 28.7, 29.9, 32.2, 37.5, 66.1, 123.7,
142.9; HRMS (FABϩ) Found: m/z, 154.1369. C10H18O requires
m/z, 154.1358.
[(2R)-2-(Methylsulfonylamino)-2-phenylethyl]diphenylphos-
phine (Ϫ)-2. The ligand was prepared in 40% yield according to
the general procedure above except using (R)-1 synthesised
from (R)-phenylglycinol and had the same spectroscopic
data as ligand (ϩ)-2. The ligand was used in the preparation of
a racemic mixture of ligand 2. [α]D Ϫ18.4 (c 0.27, CHCl3).
3-(Diethoxymethylsilyl)-2-methylcyclohexene 12. δH 0.13 (s,
3 H), 1.22 (t, 3 H, J 7.0), 1.23 (t, 3 H, J 7.0), 1.99 (br m, 2 H),
1.76 (m, 3 H), 3.79 (q, 2 H, J 7.0), 3.80 (q, 2 H, J 7.0), 5.36 (m,
1 H); δC Ϫ5.0, 18.6, 21.6, 24.2, 24.9, 25.4, 29.6, 58.4, 120.7,
134.5; HRMS (FABϩ) Found: m/z, 228.1544. C12H24O2Si
requires m/z, 228.1546.
[(2S)-2-(Methylsulfonylamino)-2-phenylethyl]diphenylarsine
3. Yield 50%. δH 2.50 (1 H, dd, J 12.7 and 7.4), 2.56 (3 H, s),
2.61 (1 H, dd, J 12.7 and 7.9), 4.66 (1 H, q, J 7.4), 4.97 (1 H, d,
J 7.1), 7.55–7.24 (15 H, m); δC 37.4, 42.0, 57.2, 126.7, 128.5,
128.8, 128.9, 129.0, 129.1, 129.2, 133.1 and 133.3; mp 145–
147 ЊC; [α]D ϩ19.7 (c 0.25, CHCl3); HRMS (FABϩ) Found:
m/z, 427.05823. C21H22O2NSAs requires m/z, 427.05872.
Acknowledgements
We thank the Swedish Natural Science Research Council,
the Crafoord Foundation, the Knut and Alice Wallenberg
Foundation, and the Royal Physiographic Society in Lund for
financial support.
(R)-N-(Methylsulfonyl)-2-phenylaziridine 4
NaH (30 mg, 0.75 mmol, 60% in mineral oil) was added to a
solution of 1 (0.20 g, 0.68 mmol) in THF (10 ml) at rt. The
reaction mixture was heated at 50 ЊC for 4 h. Aqueous work-up
followed by filtration through a short plug of silica afforded
0.11 g, 82% of 4 as a slightly yellow solid. The compound had
the same 1H and 13C spectroscopic data as previously
reported.34 Mp 43–45 ЊC; [α]D Ϫ206.2 (c 0.29, CHCl3); HRMS
(FABϩ) Found: m/z, 198.0581. C9H12O2NS (M ϩ H) requires
m/z, 198.0589.
References
1 L. Pettersson and T. Frejd, J. Chem. Soc., Chem. Commun., 1993,
1823.
2 L. Pettersson, G. Magnusson and T. Frejd, Acta Chem. Scand., 1993,
47, 196.
3 A. Marinetti, Tetrahedron Lett., 1994, 35, 5861.
4 K. Kitayama, H. Tsuji, Y. Uozumi and T. Hayashi, Tetrahedron
Lett., 1996, 37, 4169.
5 G. Pioda and A. Togni, Tetrahedron: Asymmetry, 1998, 9, 3903.
6 H. Ohmura, H. Matsuhashi, M. Tanaka, M. Kuroboshi, T. Hiyama,
Y. Hatanaka and K.-i. Goda, J. Organomet. Chem., 1995, 499,
167.
7 Y. Uozumi and T. Hayashi, J. Am. Chem. Soc., 1991, 113, 9887.
8 T. Okada, T. Morimoto and K. Achiwa, Chem. Lett., 1990, 999.
9 A. Marinetti and L. Ricard, Organometallics, 1994, 13, 3956.
10 T. Hayashi, Y. Matsumoto, I. Morikawa and Y. Ito, Tetrahedron:
Asymmetry, 1990, 1, 151.
Preparation of the catalyst
PdCl2(CH3CN)2 (one equivalent) and the appropriate ligand
(one equivalent) were mixed in benzene and heated at 50 ЊC for
1 h under an argon atmosphere. The solvent was evaporated
under reduced pressure and the resulting yellow–orange powder
was used as catalyst. The catalyst was not rigorously protected
from contact with air.
11 T. Hayashi and Y. Uozumi, Pure Appl. Chem., 1992, 64, 1911.
12 T. Hayashi, Acta. Chem. Scand., 1996, 50, 259.
13 S. Sakuraba, T. Okada, T. Morimoto and K. Achiwa, Chem. Pharm.
Bull., 1995, 43, 927.
14 B. Basu and T. Frejd, Acta. Chem. Scand., 1996, 50, 316.
15 B. Basu, S. K. Chattopadhyay, A. Ritzén and T. Frejd, Tetrahedron:
Asymmetry, 1997, 8, 1841.
16 A. Ritzén, B. Basu, S. K. Chattopadhyay, F. Dossa and T. Frejd,
Tetrahedron: Asymmetry, 1998, 9, 503.
17 A. Ritzén, B. Basu, A. Wållberg and T. Frejd, Tetrahedron:
Asymmetry, 1998, 9, 3491.
18 A. M. Aguiar and T. G. Archibald, J. Org. Chem., 1967, 32, 2627.
19 A. Albinata, F. Lianza, H. Berger, P. S. Pregosin, H. Rüegger and
R. W. Kunz, Inorg. Chem., 1993, 32, 478.
20 V. V. Dunina, E. B. GolovanЈ, N. S. Gulyukina, Y. K. Grishin and
I. P. Beletskaya, Russ. Chem. Bull., 1997, 46, 1331.
21 J. C. Anderson, R. J. Cubbon and J. D. Harling, Tetrahedron:
Asymmetry, 1999, 10, 2829.
22 L. Menabue, M. Saladini and M. Sola, Inorg. Chem., 1990, 29, 1293.
23 L. Menabue and M. Saladini, Inorg. Chem., 1991, 30, 1651.
24 G. Battistuzzi, G. Gavioli, M. Borsari, L. Menabue, M. Saladini and
M. Sola, J. Chem. Soc., Dalton Trans., 1994, 279.
25 G. Battistuzzi, M. Borsari, L. Menabue, M. Saladini and M. Sola,
Inorg. Chim. Acta., 1998, 273, 397.
General procedure for hydrosilylation of substrates 5, 8, 11 and
14 followed by ethanolysis and oxidation
Under an atmosphere of argon the catalyst (0.02 equivalents)
was dissolved in a minimum amount of CH2Cl2. The substrate
(one equivalent) was added and the mixture was cooled to 0 ЊC.
HSiCl2Me (1.5 equivalents) was then added via a syringe and
the mixture was allowed to stir for 40 h at 0 ЊC. The excess of
solvent and hydrosilane was evaporated under reduced pressure
and the resulting crude product was added to a solution of
Et3N in EtOH at 0 ЊC and stirred for 2 h. Pentane was added to
the resulting mixture which was filtered through a small plug of
silica to remove the precipitate. The filtrate was concentrated
under reduced pressure to give the corresponding alkoxysilanes
(for yields, see Table 1).
Oxidation of the alkoxysilanes was performed using a stand-
ard method: KHCO3 (one equivalent), KF (one equivalent) and
H2O2 (one equivalent) were added to the alkoxysilane in
MeOH–THF 1 : 1 at room temperature and then the mixture
was heated at reflux for 5 h. Aqueous work-up and filtration
through a short plug of silica afforded the corresponding alco-
hols in 40–70% yield. The alcohols 7, 10, and 13 were analysed
with respect to the ees using GLC analyses (alpha-DEX
column). The ee of alcohol 16 and the absolute configuration
of alcohols 7, 13 and 16 were determined by comparison of
26 H. Motschi and P. S. Pregosin, Inorg. Chim. Acta., 1980, 40,
141.
27 P. S. Pregosin, R. Rüedi and C. Anklin, Magn. Reson. Chem., 1986,
24, 255.
28 F. H. Allen and A. Pidcock, J. Chem. Soc. (A), 1968, 2700.
1456
J. Chem. Soc., Perkin Trans. 1, 2001, 1452–1457

View Article Online
29 K. Tamao, T. Kakui, M. Akita, T. Iwahara, R. Kanatani, J. Yoshida
and M. Kumada, Tetrahedron, 1983, 39, 983.
30 G. R. Jones and Y. Landais, Tetrahedron, 1996, 52, 7599.
31 B. Marciniec, Comprehensive Handbook on Hydrosilylation,
Pergamon Press, Oxford, 1992, pp. 110–114.
32 J.-E. Bäckvall, S. E. Byström and R. E. Nordberg, J. Org. Chem.,
1984, 49, 4619.
33 M. Sellén, J.-E. Bäckvall and P. Helquist, J. Org. Chem., 1991, 56,
835.
34 P. Müller, C. Baud and Y. Jacquier, Can. J. Chem., 1998, 76,
738.
35 T. Hayashi, K. Kabeta, T. Yamamoto, K. Tamao and M. Kumada,
Tetrahedron Lett., 1983, 24, 5661.
36 O. Fujimura and R. H. Grubbs, J. Org. Chem., 1998, 63, 824.
J. Chem. Soc., Perkin Trans. 1, 2001, 1452–1457
1457