Novel sulfinyl imine ligands for asymmetric catalysis

Article abstract of DOI:10.1021/ol027468m

(Matrix presented) A novel class of P,N-sulfinyl imine ligands has been prepared that incorporates chirality solely at sulfur. The Pd complex of ligand 14 catalyzes the allylic alkylation reaction with high enantioselectivity (94%), and the first crystal structure of a Pd-bound sulfinyl imine provides insight into binding mode and origins of stereoselectivity.

Full text of DOI:10.1021/ol027468m

ORGANIC
LETTERS
2003
Vol. 5, No. 4
545-548
Novel Sulfinyl Imine Ligands for
Asymmetric Catalysis
Laurie B. Schenkel and Jonathan A. Ellman*
Center for New Directions in Organic Synthesis, Department of Chemistry,
UniVersity of California, Berkeley, California 94720
jellman@uclink.berkeley.edu
Received December 13, 2002
ABSTRACT
A novel class of P,N-sulfinyl imine ligands has been prepared that incorporates chirality solely at sulfur. The Pd complex of ligand 14 catalyzes
the allylic alkylation reaction with high enantioselectivity (94%), and the first crystal structure of a Pd-bound sulfinyl imine provides insight
into binding mode and origins of stereoselectivity.
Over the past several years, many useful chiral ligand classes
have been developed due to the increasing importance of
asymmetric catalysis in organic synthesis.1 The vast majority
of these ligands incorporate chirality at a carbon center, and
very few ligands that rely solely on heteroatom chirality have
provided promising levels of asymmetric induction.2 Re-
cently, we reported on the synthesis and utility of bis-
(sulfinyl)imidoamidine (SIAM) ligands, whose Cu(II) com-
plexes provide extremely high levels of enantio- and
diastereoselectivity in the Diels-Alder reaction.3 Chirality
in these ligands is derived solely from the chiral sulfur center
of tert-butanesulfinamide, a commercially available com-
pound that can readily be transformed into sulfinyl imines
through condensation with aldehydes and ketones.4 Sulfinyl
imines are attractive as a versatile ligand class due to their
ease of synthesis, which enables straightforward modification
of their steric and electronic properties. Furthermore, there
is potential for metal coordination proximal to the chiral
center through the N, O, or S atoms.3,5 Herein, we report on
the utility of a new class of sulfinyl imine ligands 1 (Figure
1), which provides high levels of enantioselectivity in
Figure 1. Sulfinyl imine and phosphinooxazoline ligand scaffolds.
palladium-catalyzed allylic alkylation. In addition, we report
the first X-ray crystal structure of a Pd-bound sulfinyl imine
ligand, which provides insight into the binding mode and
origins of selectivity.
In designing chiral, sulfinamide-based ligands for pal-
ladium catalysis, we focused on the phosphinooxazoline
(1) Catalytic Asymmetric Synthesis, 2nd ed.; Ojima, I., Ed; Wiley-
VCH: New York, 2000.
(2) For leading references: (a) Hiroi, K.; Suzuki, Y.; Kawagishi, R.
Tetrahedron Lett. 1999, 40, 715. (b) Bolm, C.; Simic´, O. J. Am. Chem.
Soc. 2001, 123, 3820. (c) Harmata, M.; Ghosh, S. Org. Lett. 2001, 3, 3321.
(d) Hiroi, K.; Watanabe, K.; Abe, I.; Koseki, M. Tetrahedron Lett. 2001,
42, 7617. (e) Bolm, C.; Simic´, O.; Martin, M. Synlett. 2001, 12, 1878.
(3) Owens, T.; Hollander, F.; Oliver, A.; Ellman, J. J. Am. Chem. Soc.
2001, 123, 1539.
(4) Liu, G.; Cogan, D.; Owens, T.; Tang, T.; Ellman, J. J. Org. Chem.
1999, 64, 1278.
(5) Souers, A.; Owens, T.; Oliver, A.; Hollander, F.; Ellman, J. Inorg.
Chem. 2001, 40, 5299.
10.1021/ol027468m CCC: $25.00 © 2003 American Chemical Society
Published on Web 01/21/2003

ligand class 2 (Figure 1) due to the success of these and
other related P,N ligands in palladium-catalyzed transforma-
tions.6 The sulfinyl imine scaffold was thus designed to
incorporate a chelating phosphine and sp2 nitrogen in
positions relative to the chiral center that are analogous to
those in the phosphinooxazoline scaffold. Ligand 3 was
selected for initial optimization of the allylic alkylation
conditions. Preparation of 3 was completed in a single step
via Ti-mediated condensation of tert-butanesulfinamide with
commercially available 2-(diphenylphosphino)benzaldehyde
(Scheme 1). A range of solvents and Pd reagents was then
for high enantioselectivity (entry 7). This interesting result
when an excess of ligand is present could potentially be due
to displacement of the chiral sulfinyl imine moiety from
palladium by the phosphine of another ligand molecule to
give a less stereoselective catalyst.
Having identified optimal reaction conditions for 3, the
effects of varying both R1 and R2 of the ligand structure
(Figure 1) were investigated. With 4, the reaction proceeded
slowly and the product formed was racemic (Table 1, entry
8), suggesting that either the greater steric bulk or the
increased electron-donating ability of the tert-butyl group
relative to the p-tolyl group is required for catalytic activity
and stereoselectivity. Ketimine ligand 7 (Scheme 2) was next
Scheme 1. One-Step Synthesis of Sulfinyl Imine Ligands
Scheme 2. Synthesis of Ketimine Ligand 7
explored in the allylic alkylation of 1,3-diphenylpropenyl
acetate using ligand 3 (Table 1). While moderate enantiose-
Table 1. Optimization of the Allylic Alkylationa
prepared in three steps from 2-(diphenylphosphino)benzoic
acid via the Weinreb amide intermediate 5. Under the
optimized conditions for the allylic alkylation reaction, 7
provided a modest increase in reaction rate but with a
significant reduction in stereoselectivity (Table 1, entry 9).
To elucidate the way in which 3 imparts stereoselectivity,
an X-ray crystal structure was obtained of the π-allyl Pd(II)
complex of 3, which corresponds to the Pd(II) intermediate
in the catalytic cycle (Figure 2). This structure confirms that
ligand 3 binds palladium through phosphorus and nitrogen
to form a six-membered ring chelate. Similar to previously
reported π-allyl Pd crystal structures, the palladium in
complex 8 has a slightly distorted square-planar geometry,7
and the disorder of the allyl unit reflects a known η3-η1-
η3 isomerization.8
Notably, the bond lengths between palladium and the
carbons of the allyl unit exhibit a trans influence similar to
that reported for other P,N ligands (Table 2).9 It has been
shown that mixed chelate ligands can electronically induce
asymmetry due to the lengthening, and thus increased
reactivity, of the Pd-allyl carbon bond trans to the better
π-acceptor. Nucleophilic attack is known to occur prefer-
entry
L*
solventb
Pd source
timec (h)
% eed
1
2
3
4
5
6
7
8
9
3
3
3
3
3
3
3
4
7
CH3CN
CH3CN
THF
C6H6
C6H6
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
[Pd(allyl)Cl]2
Pd2dba3‚CHCl3
Pd2dba3‚CHCl3
[Pd(allyl)Cl]2
Pd2dba3‚CHCl3
[Pd(allyl)Cl]2
[Pd(allyl)Cl]2
[Pd(allyl)Cl]2
[Pd(allyl)Cl]2
6
1
5
43
53
67
86
88
93
41f
0
26
>33e
8.5
20
>25e
5
56
a Reactions run with 30 mol % L* and 1:1.3 L*/Pd. b Reactions run at
[0.07] in substrate. c Time required for disappearance of starting material
by TLC analysis. d Determined by chiral HPLC (Chiralpak AD). e Reaction
was not complete in time indicated. f Reaction run with 20 mol % 3 and
2:1 3/Pd.
lectivities were observed in coordinating solvents (entries
1-3), a significant improvement in selectivity occurred upon
switching to benzene (entries 4 and 5). We were pleased to
find that in methylene chloride (entry 6), the Pd complex of
3 generated from [Pd(allyl)Cl]2 catalyzed the allylic alkyl-
ation reaction with 93% ee and in high conversion. Surpris-
ingly, an excess of palladium relative to ligand was required
(7) (a) Sprinz, J.; Kiefer, M.; Helmchen, G. Tetrahedron Lett. 1994, 35,
1523. (b) Albinati, A.; Kunz, R.; Amman, C.; Pregosin, P. Organometallics
1991, 10, 1800 and references therein.
(8) Von Matt, P.; Lloyd-Jones, G.; Minidis, A.; Pfaltz, A.; Macko, L.;
Neuburger, M.; Zehnder, M.; Ruegger, H.; Pregosin, P. HelV. Chim. Acta
1995, 78, 265.
(9) Frost, C.; Howarth, J.; Williams, J. Tetrahdron: Asymmetry, 1992,
3, 1089.
(6) (a) Helmchen, G.; Pfaltz, A. Acc. Chem. Res. 2000, 33, 336. (b) Hou,
D.-R.; Reibenspies, J.; Burgess, K. J. Org. Chem. 2001, 66, 206. (c)
Gilbertson, S.; Xie, D.; Fu, Z. J. Org. Chem. 2001, 66, 7240. (d) Saitoh,
A.; Morimoto, T.; Achiwa, K. Tetrahedron: Asymmetry, 1997, 8, 3567.
546
Org. Lett., Vol. 5, No. 4, 2003

Table 3. Allylic Alkylation with 3 at 5 Mol % Catalyst
Loadinga
entry
concentration (M)
timeb (h)
% eec
1
2
3
4
5
0.07
0.10
0.20
0.40
0.03
>30
>30
>30
>30
>168
86
86
71
19
92
a Reactions run with 5 mol % 3 and 1:1.3 3/Pd ratio. b Reaction was not
complete in time indicated. c Determined by chiral HPLC.
to significantly increase steric hindrance about the phos-
phorus. Preparation of these second-generation ligands began
with formation of the acetal of 2-bromobenzaldehde (Scheme
3). Treatment with magnesium or t-BuLi, followed by
Figure 2. X-ray crystal structure of π-allyl Pd-bound 3. Hydrogens
and the counterion have been omitted for clarity.
Scheme 3. Preparation of Ligands 13 and 14a
entially at the allyl carbon trans to the phosphorus atom in
related systems.9
Table 2. Selected Bond Distances and Angles in Complex 8
bond
distance (Å)
angle
deg
Pd-N
Pd-P
Pd-C1
Pd-C2
Pd-C3
Pd-C4
2.10
2.27
2.21
2.14
2.12
2.21
N-Pd-P
88.4
100.1
68.1
P-Pd-C1
C1-Pd-C3
C3-Pd-N
103.1
b
c
a R ) Cy, t-BuLi. R ) o-tol, Mg. Crude material was carried
on to the next step without purification.
While ligand 3 performed well in the Pd-catalyzed allylic
alkylation at 30 mol % catalyst loading (Table 1), attempts
to reduce the catalyst loading to 5 mol % were disappointing.
Under the optimized conditions, the reaction rate was
drastically reduced (Table 3, entry 1). Furthermore, the
stereoselectivity showed a profound concentration depen-
dence, with higher concentrations resulting in a dramatic
reduction in enantioselectivity (entries 2-4). At low con-
centration (entry 5), the rate of reaction was unacceptably
slow.
Having already investigated changes in the ligand scaffold
at both the chiral sulfur center and the imine carbon, we
chose to evaluate ligands with substituents on phosphorus
with different steric and electronic properties in order to
improve the performance of the catalyst system. The dicyclo-
hexylphosphine 13 would allow evaluation of a dialkylphos-
phine ligand, while the di(o-tolyl)phosphine 14 was selected
addition to the diaryl- or dialkylphosphine chloride, provided
compounds 9 and 10. The acetals were deprotected to give
aldehydes 11 and 12, which were subsequently condensed
with tert-butanesulfinamide to provide the desired ligands
13 and 14 in good yields.
Table 4 summarizes the activity of 13 and 14 in the allylic
alkylation reaction. At 30 mol % catalyst loading, the Pd
complex of 13 was inferior in both rate and enantioselectivity
(entry 1) when compared with ligand 3 (entry 6, Table 1).
In contrast, at 30 mol % catalyst loading the Pd complex of
14 was both more active and stereoselective (entry 2).
Notably, additional experiments demonstrated that both high
catalytic activity and stereoselectivity were maintained at
various ligand/Pd ratios (entries 3 and 4) and at high and
low concentrations (entries 5-7). Furthermore, low catalyst
loading gave complete conversion in 6 h to provide the allylic
alkylation product in 95% isolated yield and 94% ee (entry
Org. Lett., Vol. 5, No. 4, 2003
547

In conclusion, a new sulfinyl imine ligand class has been
developed with the design and synthesis of phosphine-
containing scaffolds. The Pd complex of ligand 14 catalyzes
the allylic alkylation of 1,3-diphenylpropenyl acetate with
high levels of enantioselectivity at low catalyst loading. The
first X-ray crystal structure of a π-allyl Pd-bound sulfinyl
imine confirms that these are P,N ligands and suggests that
high enantioselectivity is achieved via a trans influence. The
application of this new ligand class to additonal Pd- and other
transition metal-catalyzed reactions is forthcoming.
Table 4. Evaluation of 13 and 14 in the Allylic Alkylation
timeb
entry L* mol % L* L*:Pd concentrationa (M) (h) % eec
1
2
3
4
5
6
7
13
14
14
14
14
14
14
30
30
30
30
5
1:1.3
1:1.3
1.5:1
1:1
1:1
1:1
1:1
0.10
0.07
0.07
0.07
0.07
0.10
0.20
24
2
1
1
6
49
96
96
96
94
93
93
Acknowledgment. This work was supported by the
National Science Foundation (CHE0139468). We also thank
Dr. Allen G. Oliver and Dr. Frederick J. Hollander of the
Berkeley CHEXray facility for solving the X-ray crystal
structure. The Center for New Directions in Organic Syn-
thesis is supported by Bristol-Myers-Squibb as sponsoring
member.
5
5
4
4
a Substrate concentration. b Time required for disappearance of starting
material by TLC analysis. c Determined by chiral HPLC.
5). The observed selectivity approaches that of the most
selective of the phosphinooxazoline catalysts (99% ee),10 and
we anticipate that further optimization of the phosphine
sulfinyl imine ligand scaffold will result in improved
selectivity.
Supporting Information Available: Crystallographic
data, full experimental details, and characterization for all
new compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
(10) von Matt, P.; Pfaltz, A. Angew. Chem., Int. Ed. Engl. 1993, 32, 2,
4, 566.
OL027468M
548
Org. Lett., Vol. 5, No. 4, 2003