Chiral mixed phosphorus/sulfur ligands for palladium-catalyzed allylic alkylations and aminations

Full text of DOI:10.1021/jo990344b

2994
J . Org. Chem. 1999, 64, 2994-2995
Ta ble 1. Allylic Alk yla tion of 5 w ith Rep r esen ta tive
Nu cleop h iles (Eq 1)a
Ch ir a l Mixed P h osp h or u s/Su lfu r Liga n d s for
P a lla d iu m -Ca ta lyzed Allylic Alk yla tion s a n d
Am in a tion s
David A. Evans,* Kevin R. Campos, J ason S. Tedrow,
Forrest E. Michael, and Michel R. Gagne´
Department of Chemistry and Chemical Biology,
Harvard University, Cambridge, Massachusetts 02138
CH2(CO2Me)2, BSAb ee,
BnNH2c ee,
% (yield 5b, %)
L*
% (yield 5a , %)
Received February 25, 1999
1a
2a
3a
91 (93)
98 (97)
94 (95)
99 (96)
95 (97)
95 (95)
The incorporation of C2 symmetry into chiral ligand design
is a well-recognized strategy for restricting the number of
diastereomeric transition states in metal-catalyzed enantio-
selective processes.1 Equally powerful stereochemical re-
strictions may also be realized with chiral ligands lacking
C2 symmetry through the use of electronic effects such as
the trans influence.2 Such effects are a natural consequence
of the use of chiral bidentate ligands equipped with strong
and weak donor heteroatom pairs (e.g., PR3/NR3 or PR3/SR2).
Such electronic effects have the potential to influence both
the stability and reactivity of the intervening diastereomeric
reaction intermediates in the catalytic cycle. While mixed
phosphorus/nitrogen bidentate ligands incorporating this
construct have been applied in enantioselective palladium-
catalyzed nucleophilic alkylation of allylic esters,3 chiral
thioether-containing donor ligands have been less well
developed.4 As seen in structure A, thioether complexation
creates an S-chiral sulfur center; however, a potential
liability associated with these ligands is the relatively low
barrier to sulfur inversion (15-20 kcal/mol) for transition
metal-coordinated thioethers.5 In this paper, we report a new
class of mixed phosphorus/sulfur ligands 1-3 that incorpo-
rates a metal-bound thioether as a chiral control element
in asymmetric catalysis. The utility of these ligands is
illustrated in the palladium-catalyzed allylic alkylation6 with
enol-malonate and amine nucleophiles.
1b
2b
3b
28 (91)
30 (94)
69 (92)
78 (90)
66 (95)
89 (93)
a
Reactions were run in CH2Cl2 at -20 °C using 2 mol % Pd
and 2.8 mol % L*. Enantiomeric purity determined by chiral HPLC
b
analysis (Daicel Chiralcel AD). 3 equiv of malonate and BSA and
cat. KOAc were used relative to substrate. c 2 equiv of BnNH2 used
relative to 4.
independently varied to generate a large ligand family
containing sterically and electronically differentiated ana-
logues. The diarylphosphinite moiety was selected for the P
terminus by virtue of its ease of incorporation and its
documented utility as a ligand component.7 Diarylphosphin-
ites 18 and 29 were identified as valuable ligands after a
survey of both thioether and diarylphosphinite ligand com-
ponents. For example, in test reactions of the Pd-catalyzed
alkylation of 1,3-diphenylpropenyl acetate (4) with dimethyl
malonate and bis(trimethylsilyl)acetamide (BSA),10 ligands
1a and 2a afforded product 5a in good yields and enanti-
oselectivities (91 and 98% ee, respectively, eq 1, Table 1).
For the sulfur donor moiety, two trends were noted for the
alkylation process with malonate nucleophile. First, in-
creased steric hindrance was found to directly correlate with
increased enantioselection with the S-tert-butyl substituent
being optimal. Second, alkyl substituents proved to be
superior to their aryl counterparts. For the diarylphosphinite
moiety, neither electron-withdrawing nor electron-donating
substituents proved to be superior to phenyl.9
Ligand 3, readily synthesized in enantiomerically pure
form in two steps from cyclohexene oxide and tert-butyl-
mercaptan using methodology recently reported by Shiba-
saki,11 was considered as a structural analogue of 2. The
corresponding malonate alkylation with ligand 3a afforded
product 5a in 94% ee (Table 1). The data in Table 1 also
demonstrate that all three ligands promote allylic amination
with benzylamine in 95-99% ee. The comparative alkylation
reactions of the R-naphthyl ligand series 1b-3b is also
Ligands 1-3 are composed of three subunits that include
the Ar2P- and RS- heteroatom fragments and the intercon-
necting skeletal backbone. Each of these fragments may be
(1) Whitesell, J . K. Chem. Rev. 1989, 89, 1581-1590.
(2) (a) Appleton, T. G.; Clark, H. C.; Manzer, L. E. Coord. Chem. Rev.
1973, 10, 335-422. (b) Murray, S.; Hartley, F. Chem. Rev. 1981, 81, 365-
414.
(3) Chiral P,N ligands: (a) Pfaltz, A. Acta Chim. Scand. 1996, 50, 189-
194 and references therein. (b) Kudis, S.: Helmchen, G. Angew. Chem., Int.
Ed. Engl. 1998, 37, 3047-3050. (c) Dawson, G. J .; Frost, G.; Williams, J .
M. J . Tetrahedron Lett.. 1993, 34, 3149-3150.
(7) (a) Trost, B. M.; Murphy, D. J . Organometallics 1985, 4, 1143-1145.
(b) Nomura, N.; Mermet-Bouvier, Y. C.; RajanBabu, T. V. Synlett 1996, 745-
746 and refs. cited therein. (c) Seebach, D.; Devaquet, E.; Ernst, A.;
Hayakawa, M.; Kuhnle, F.; Schweizer, W. B.; Weber, B. Helv. Chim. Acta
1995, 78, 1636-1650.
(4) Chiral N,S ligands: (a) Morimoto, T.; Tachibana, K.; Achiwa, K.
Synlett 1997, 783-785. (b) Anderson, J . C.; J ames, D. S.; Mathias, J . P.
Tetrahedron: Asymmetry 1998, 9, 753-756. (c) Sprinz, J .; Keifer, M.;
Helmchen, G.; Regglein, M.; Huttner, G.; Walter, O.; Zsolanai, L. Tetrahe-
dron Lett. 1994, 10, 1523-1526. (d) Allen, J .; Bower, J .; Williams, J .
Tetrahedron: Asymmetry 1994, 5, 1895-1898. (e) Boog-Wick, K.; Pregosin,
P.; Trabesinger, G. Organometallics 1998, 17, 3254-3264. Chiral P,S
ligands: (f) Albinati, A.; Pregosin, P.; Wick, K. Organometallics 1996, 15,
2419-2421 and references therein. (g) Hiroi, K.; Suzuki, Y. Tetrahedron
Lett. 1998, 39, 6499-6502. (h) Hauptman, E.; Shapiro, R.; Marshall, W.
Organometallics 1998, 17, 4976-4982.
(5) Abel, E.; Bhargava, S. K.; Orrell, K. G. Prog. Inorg. Chem. 1984, 32,
1-118. Abel, E.; Dormer, J .; Ellis, D.; Orrell, K. G.; Sik, V.; Hursthouse,
M. B.; Mazid, M. A. J . Chem. Soc., Dalton Trans. 1992, 1073-1080.
(6) For a general review of the asymmetric transition metal-catalyzed
allylic alkylation, see: Trost, B. M.; Van Vranken, D. L. Chem. Rev. 1996,
96, 395-422 and references therein.
(8) A range of thioether substituent analogues of ligand 1a were
evaluated in reactions between
4 and malonate/BSA. Aryl and alkyl
substituents investigated: 3,5-Me2Ph (63% ee), Bn (89% ee), Cy (91% ee),
tert-butyl (91% ee). See the Supporting Information for the ligand synthesis.
(9) A range of thioether substituent analogues of ligand 2a were
evaluated in reactions between 4 and malonate/BSA but none were superior
to tert-butyl. Aryl and alkyl substituents investigated: 3,5-Me2Ph (85% ee),
Bn (75% ee), Cy (81% ee), and tert-butyl (98% ee). A range of phosphinate
aryl substituent analogues of ligand
2 were evaluated but none were
superior to phenyl. Aryl substituents investigated: 3,5-Me2Ph (80% ee), 3,5-
(CF3)2Ph (93% ee), 4-MeOPh (82% ee), 4-FPh (93% ee), 2-MeOPh (29% ee),
Cy (47% ee), and R-naphthyl (30% ee). See the Supporting Information for
the ligand synthesis.
(10) Trost, B. M.; Murphy, D. J . Organometallics 1985, 4, 1143-1145.
(11) Iida, T.; Yamamoto, N.; Sasai, H.; Shibasaki, M. J . Am. Chem. Soc.
1997, 119, 4783-4784.
10.1021/jo990344b CCC: $18.00 © 1999 American Chemical Society
Published on Web 04/02/1999

Communications
J . Org. Chem., Vol. 64, No. 9, 1999 2995
-
F igu r e 1. X-ray structures of 13 and 14. SbF6 counterions from each structure omitted for clarity.
Ta ble 2. Alk yla tion of Cyclic Allylic Ester s (Eqs 2, 3)a
ligand for 1,3-diphenylpropenyl acetate (4) displacements
(95-98% ee, Table 1), affected the 7b f 8b transformation
in only 38% ee.
Evidence that the sulfur is functioning as a coordinating
ligand in these reactions is supported by X-ray structures
of the [Pd(2a )(π-1,3-diphenylallyl)](SbF6) complex 1314 and
the [Pd(1a )-(cyclohexenyl)](SbF6) complex 1415 (Figure 1).
As predicted, the coordinated thioether ligand in both
structures is oriented trans to the isopropyl group to
minimize nonbonding interactions. In addition, the adjacent
methyl substituent increases the steric demands of the
isopropyl moiety by orienting it in the direction of the bound
thioether. Noteworthy differences in the two structures may
be found in the ring conformations of the bound ligands.
While a twist-boat conformation is observed in complex 13,
the chelate ring conformation in 14 is more chairlike. These
conformational differences appear to be coupled to the
conformation of the Ph2P moiety where the phenyl edge/face
relationships are clearly different in the two complexes. The
crystal structures also reveal the relative electronic impact
of the heteroatom phosphinite and thioether donors. For
example, the Pd-C1 bond trans to the phosphinite is longer
than the Pd-C2 bond trans to the thioether, emphasizing
the stronger trans influence of the phosphinite moiety.2 On
the basis of the orientation of the π-allyl ligand in the crystal
structure, attack of the nucleophile trans to the phosphinite
in the illustrated crystal geometries predicts the stereo-
chemistry that is observed for all reactions.15 Further studies
in this area are ongoing.
d
CH2(CO2Me)2, BSAb ee,c
% (yield, %)
BnNH2 ee,e
substrate
% (yield, %)
7a
7b
7c
94 (94) 8a
94 (91) 8b
96 (98) 8c
94 (95) 11
91 (93) 9a
91 (97) 9b
97 (95) 9c
94 (99) 12
10
See Table 1 for footnotes a,b. Determined by 1H NMR chiral
a
b
d
shift with Eu(hfc)3 in C6D6. 2 equiv of BnNH2 was used relative
to substrate. e Determined by achiral HPLC analysis of the cor-
responding (S)-Mosher amide.
provided. These data establish that the diphenylphosphinyl
moiety is superior to its R-naphthyl counterpart for allylic
acetate 4. This trend is to be contrasted to the alkylation
results for cyclic allylic acetates where ligand 1b is the
ligand of choice (cf. Table 2).
Chiral ligands that effectively promote the enantioselec-
tive alkylation of cyclic allylic esters have different structural
requirements than their acyclic counterparts.12 Accordingly,
we surveyed these substrates with this ligand family (Table
2, eqs 2 and 3). From the ligand screen with cycloalkenyl
acetates 7a -c and malonate, ligand architecture 1 surfaced
as the optimal ligand backbone with bis(R-naphthyl)phos-
phinite 1b being superior (7b f 8b, 94% ee) to its phenyl
counterpart 1a (7b f 8b, 90% ee). Heteroatom analogues
such as 10 are also effective alkylation substrates (10 f 11,
94% ee). Benzylamine may also be employed as an effective
nucleophile, affording products 9a -c and 12 in equivalent
enantioselectivities and yields. As an illustration of the
importance of ligand architecture, 2a , while an excellent
Ack n ow led gm en t. We gratefully acknowledge the
NSF (CHE-9633582), NIH(GM-33328), Pfizer, Merck, and
DuPont for research support.
Su p p or tin g In for m a tion Ava ila ble: Experimental proce-
dures, spectral data, and enantiomeric purity assays for all
compounds.
J O990344B
(13) Crystals of 13 (C37H44POSPdSbF6) were grown from a warm solution
of 7 in methanol to yield yellow prisms. The compound crystallizes in the
orthorhombic crystal system, space group P212121; a ) 19.597(2) Å, b )
20.706(3) Å, c ) 9.7944(13) Å, R ) â ) γ ) 90°; V ) 3974.4(9) Å3; Z ) 4; R
) 0.0386, GoF ) 0.974.
(14) Crystals of 14 (C28H40POSPdSbF6) were grown from a warm solution
of 13 in methanol to yield yellow prisms. The compound crystallizes in the
orthorhombic crystal system, space group P212121; a ) 10.228(5) Å, b )
17.727(8) Å, c ) 18.088(7) Å, R ) â ) γ ) 90°; V ) 3279 (2) Å3; Z ) 4; R )
0.0433, GoF ) 1.282.
(15) The direction of attack trans to the stronger π-acceptor has been
previously documented by others: ref 3. See also: Ward, T. R. Organo-
metallics 1996, 15, 2836-2838.
(12) Trost, B. M.; Bunt, R. C. J . Am. Chem. Soc. 1994, 116, 4089-4090.
Sennhenn, P.; Gabler, B.; Helmchen, G. Tetrahedron Lett. 1994, 35, 8595-
8598. Knuhl, G.; Sennhenn, P.; Helmchen, G. J . Chem. Soc., Chem.
Commun. 1995, 1845-1846. Kudis, S.; Helmchen, G. Angew. Chem., Int.
Ed. Engl. 1998, 37, 3047-3050.