Versatile chiral bidentate ligands derived from α-amino acids: Synthetic applications and mechanistic considerations in the palladium- mediated asymmetric allylic substitutions

Article abstract of DOI:10.1021/jo991615l

A new class of chiral amidine-phosphine hybrid ligands 7a,b, which are readily accessible from the corresponding α-amino acids, were developed. A versatility for construction of new ligands is desirable, by which a variety of reactions and substrates become applicable. Indeed, a variety of modifications, such as exchange reactions to other amino groups in the amidine skeleton and the production of other types of ligands, are possible using the precursor compounds of 7a. Thus, novel chiral ligands 7c,d, 8, 11, and 13, which provide sterically and electronically different chiral circumstances, were prepared and used for the palladium-mediated asymmetric allylic substitutions of both acyclic and cyclic compounds. In these reactions, high levels of asymmetric induction were achieved for both substrates. A marked advancement of reactivity and enantioselectivity in palladium-catalyzed asymmetric allylations of 1,3-diphenylpropen-2-yl pivalate 14a was attained by examination of electronic substituent effects in a new series of chiral P-N and S-N hybrid ligands 8 and 11. Mechanistic views concerning the enantiodiscriminating step were demonstrated, in which a good correlation between a novel Pr/Mr concept and the absolute configuration of allylation products are discussed for the prediction of enantioselecting direction. The use of ketene silyl acetals as nucleophiles was investigated and compared with the corresponding harder anionic carbon nucleophiles. The former nucleophiles afforded higher enantioselectivity in asymmetric allylic transformations of 14a.

Full text of DOI:10.1021/jo991615l

VOLUME 65, NUMBER 14
J ULY 14, 2000
© Copyright 2000 by the American Chemical Society
Articles
Ver sa tile Ch ir a l Bid en ta te Liga n d s Der ived fr om r-Am in o Acid s:
Syn th etic Ap p lica tion s a n d Mech a n istic Con sid er a tion s in th e
P a lla d iu m -Med ia ted Asym m etr ic Allylic Su bstitu tion s
Akihito Saitoh, Kazuo Achiwa, Kiyoshi Tanaka, and Toshiaki Morimoto*
School of Pharmaceutical Sciences, University of Shizuoka, 52-1 Yada, Shizuoka-shi 422-8526, J apan
Received October 15, 1999
A new class of chiral amidine-phosphine hybrid ligands 7a ,b, which are readily accessible from
the corresponding R-amino acids, were developed. A versatility for construction of new ligands is
desirable, by which a variety of reactions and substrates become applicable. Indeed, a variety of
modifications, such as exchange reactions to other amino groups in the amidine skeleton and the
production of other types of ligands, are possible using the precursor compounds of 7a . Thus, novel
chiral ligands 7c,d , 8, 11, and 13, which provide sterically and electronically different chiral
circumstances, were prepared and used for the palladium-mediated asymmetric allylic substitutions
of both acyclic and cyclic compounds. In these reactions, high levels of asymmetric induction were
achieved for both substrates. A marked advancement of reactivity and enantioselectivity in
palladium-catalyzed asymmetric allylations of 1,3-diphenylpropen-2-yl pivalate 14a was attained
by examination of electronic substituent effects in a new series of chiral P-N and S-N hybrid
ligands 8 and 11. Mechanistic views concerning the enantiodiscriminating step were demonstrated,
in which a good correlation between a novel Pr/ Mr concept and the absolute configuration of
allylation products are discussed for the prediction of enantioselecting direction. The use of ketene
silyl acetals as nucleophiles was investigated and compared with the corresponding harder anionic
carbon nucleophiles. The former nucleophiles afforded higher enantioselectivity in asymmetric allylic
transformations of 14a .
In tr od u ction
methodology for useful synthetic applications such as new
ring formation via intramolecular cyclization and natural
product synthesis.2 For the asymmetric version, develop-
Palladium-mediated allylic transformations of allyl
compounds, alkenes, and conjugated dienes have greatly
contributed to the field of synthetic organic chemistry as
an effective tool for the construction of C-C and C-X (X
) H, heteroatoms) bonds.1 Extensive investigations on
this transformation in terms of reaction types, substrates,
and nucleophiles resulted in the establishment of the
(2) Recent reviews including palladium-catalyzed allylic substitu-
tions; see: (a) Wills, M.; Tye, H. J . Chem. Soc., Perkin Trans. 1 1999,
1109. (b) Wills, M. J . Chem. Soc., Perkin Trans. 1 1998, 3101. (c) Trost,
B. M.; Van Vranken, D. L. Chem. Rev. 1996, 96, 395. (d) Williams, J .
M. J . Synlett 1996, 705. (e) Tsuji, J . Palladium Reagents and Catalysts,
Innovations in Organic synthesis; J ohn Wiley & Sons: New York, 1995;
p 290. (f) Hayashi, T. Catalytic Asymmetric Synthesis; Ojima, I., Ed.;
VCH: Weinheim, 1993. (g) Pfaltz, A. Acc. Chem. Res. 1993, 26, 339.
(h) Reiser, O. Angew. Chem., Int. Ed. Engl. 1993, 32, 547. (i) Frost, C.
G.; Howarth, J .; Williams, J . M. J . Tetrahedron: Asymmetry 1992, 3,
1089.
(1) For pioneering studies, for example, see: (a) Tsuji, J .; Takahashi,
H.; Morikawa, M. Tetrahedron Lett. 1965, 4387. (b) Atkins, K. E.;
Walker, W. E.; Manyik, R. M. Tetrahedron Lett. 1970, 3821. (c) Trost,
B. M.; Strege, P. E. J . Am. Chem. Soc. 1977, 99, 1650.
10.1021/jo991615l CCC: $19.00 © 2000 American Chemical Society
Published on Web 06/09/2000

4228 J . Org. Chem., Vol. 65, No. 14, 2000
Saitoh et al.
Sch em e 1
Sch em e 2
ment of chiral ligands that can create effective chiral
circumstances has opened a way to enantioselective
recognition during the bond formation. A number of
useful chiral ligands have been prepared to date, by
which high levels of catalytic performance have been
demonstrated.2
In this area, we have developed new types of chiral
ligands derived from easily available R-amino acids in
both natural and unnatural types.3 A new P-N hybrid
ligand consisting of phosphine and amidine skeletons 7a
(VALAP) was developed at the initial stage of our
investigation. A tight Pd-N chelation could be expected
by the more electron-rich imino nitrogen due to the
amidine structure in VALAP. Since versatility for a
number of reaction types and substrates is an important
factor as a design concept, a variety of modifications such
as exchange reactions to other amino groups in the
amidine skeleton and the production of other types of
ligands using the synthetic precursor of VALAP were
carried out. The present paper reports the development
of a new class of chiral P-N and S-N hybrid ligands
7b-d , 8 and 11 and a diphosphine type auxiliary 13
based on VALAP. Their application to palladium-cata-
lyzed asymmetric allylic alkylations and mechanistic
considerations of enantiodiscriminating steps are also
described. To extend the scope of enantioselective allylic
substitutions, ketene silyl acetals (KSA) were used as a
nucleophile, and their synthetic utility is discussed by
comparison with “harder” anionic nucleophiles.
Syn th esis of New Ch ir al P -N,S-N Hybr id Ligan ds.
The phosphine-amidine bidentate ligand 7a (VALAP)
was developed at the early stage of our investigation.3
VALAP was easily accessible from a commercially avail-
able R-amino acid, L-valine (1a ), based on the procedure
described for the preparation of a chiral diphosphine
ligand, BPPM4 (Scheme 1).5 Similarly, the amidine 7b
was prepared from L-tert-leucine (1b). L-tert-Leucinol
(2b), prepared by the reduction of 1b, was converted
quantitatively into 3b by protection with di-tert-butyl
dicarbonate. The sulfonyloxybutane 4b was derived from
3b using p-toluenesulfonyl chloride. A diphenylphosphino
group was introduced by the reaction of 4b with potas-
sium diphenylphosphide in THF, giving diphenylphos-
Sch em e 3
phinobutane 5b, which was converted into 6b by depro-
tection of the amino moiety using trifluoroacetic acid.
Finally, the reaction of 6b and N,N-dimethylformamide
dimethylacetal gave (S)-N′-[1-[(diphenylphosphino)methyl]-
2,2-dimethylpropyl]-N,N-dimethylmethanimidamide (7b).
To examine the possibility of sterical and electronical
refinement in the structure of ligands as well as intro-
duction of an additional chiral center at the amidine part
of VALAP, replacement of the dimethylamino group by
other secondary amines was first examined. VALAP was
simply treated with pyrrolidine and piperidine at the
corresponding reflux temperatures to give the modified
amidine ligands 7c,d (Scheme 2). The chiral P-N hybrid
ligands 8 consisting of diphenylphosphino and phenyl
imino groups with different substituents at the para
position were also developed.6 The ligands 8 were readily
prepared in one step by the reaction of 6a with the
corresponding aldehydes (Scheme 3).
In contrast to a number of studies on the P-N hybrid
chiral auxiliaries, sulfur-containing hybrid ligands have
rarely been used for the transition metal-catalyzed
reactions except for a few examples,7 probably due to
their catalyst poisoning property. We previously reported
that high enantioselectivity up to 96% ee was achieved
by using the S-N hybrid ligands bearing chiral imida-
zolines in palladium-catalyzed asymmetric allylic sub-
stitutions of 1,3-diphenylallyl substrate with dimethyl
malonate in the presence of N,O-bis(trimethylsilyl)-
acetamide (BSA)-AcOLi. (S)-N′-[1-[(Phenylthio)methyl]-
2-methylpropyl]-N,N-dimethylmethanimidamide (11a ), a
structure similar to that of VALAP, was prepared through
(3) Saitoh, A.; Morimoto, T.; Achiwa, K. Tetrahedron: Asymmetry
1997, 8, 3567. Recently, we abbreviated the amidine-phosphine hybrid
ligand 7a derived from L-valine as “VALAP”.
(4) Achiwa, K. J . Am. Chem. Soc. 1976, 98, 8265.
(5) A similar procedure to 10a was found in a previous publication
by Hiroi et al.; see: Hiroi, K.; Haraguchi, M.; Masuda, Y.; Abe, J . Chem.
Lett. 1992, 2409.
(6) Saitoh, A.; Misawa, M.; Morimoto, T. Synlett 1999, 483.

Versatile Chiral Bidentate Ligands
J . Org. Chem., Vol. 65, No. 14, 2000 4229
Sch em e 4
Sch em e 6
an extension of the ligand system based on VALAP.9 The
newly designed ligand 13 was prepared in one step by
treatment of 6a with phthaloyl dichloride (Scheme 6).
Asym m etr ic Allylic Su bstitu tion s of Acyclic Su b-
str a tes. Bidentate ligands of phosphorus-nitrogen3,6,10
and sulfur-nitrogen7 mixed-donors have recently been
investigated for which high levels of enantioselection
were demonstrated in palladium-catalyzed asymmetric
allylic substitutions. In the case of ligands possessing two
different donor centers, it is generally accepted that
nucleophilic attack to π-allyl complexes occurs predomi-
nantly at the allyl terminus trans to better π-acceptors,
which must be the phosphino groups in the P-N hybrid
ligands.7d,11 Such a course of the nucleophilic attack
provides an advantage to the design of mixed-donor
ligands. Among the hybrid ligands, chiral phosphanyl-
dihydrooxazoles, independently developed by several
groups,12 have attracted much attention. The phospha-
nyldihydrooxazoles, so-called oxazoline ligands, are gen-
erally effective for asymmetric allylic replacements of
acyclic compounds and Heck-type reactions.13 Recently,
it was also reported that a phosphine-oxazoline ligand
with fine-tuning by a cymantrene unit are efficient
ligands for asymmetric transformation of cyclic sub-
strates, which has been regarded as one of the most
difficult transformations to overcome.14
Sch em e 5
Using the prepared ligands of hybrid type, asymmetric
allylic substitutions of 1,3-diphenylpropen-2-yl pivalate
(14a ) and its acetate analogue (reaction 1 in Scheme 7)
were first examined. Since the optimized reaction condi-
tions are different in each chiral ligands, effects of
10a 7g from L-valinol (2a ). Since thiols bearing phenyl
groups with various substituents at the para position
were widely available, other ligands 11b,c were also
synthesized via two steps after the reaction of 4a with
the corresponding sodium sulfides (Scheme 4).
(9) (a) Saitoh, A.; Misawa, M.; Morimoto, T. Tetrahedron: Asymmetry
1999, 10, 1025. (b) Saitoh, A.; Uda, T.; Morimoto, T. Tetrahedron:
Asymmetry 1999, 10, 4501.
As one of the structural variation of the amidine-based
ligands, diamidine-type auxiliaries 12a ,b were then
prepared by treatment with N,N-dimethylformamide
dimethylacetal from (1R,2R)-1,2-diaminocyclohexane
and (1R,2R)-1,2-diphenylethylenediamine, respectively
(Scheme 5).
(10) For selected recent articles for P-N hybrid ligands, see: (a)
Zhang W.; Yoneda, Y.; Kida, T.; Nakatsuji, Y.; Ikeda, I. Tetrahedron:
Asymmetry 1998, 9, 3371. (b) Imai, Y.; Zhang, W.; Kida, T.; Nakatsuji,
Y.; Ikeda, I. Tetrahedron Lett. 1998, 39, 4343. (c) Cahill, J . P.; Gulry,
P. J . Tetrahedron: Asymmetry 1998, 9, 4301. (d) Widhalm, M.; Mereiter,
K.; Bourghida, M. Tetrahedron: Asymmetry 1998, 9, 2983. (e) Burgess,
K.; Porte, A. M. Tetrahedron: Asymmetry 1998, 9, 2465. (f) Ogasawara,
M.; Yoshida, K.; Kamei, H.; Kato, K.; Uozumi, Y.; Hayashi, T.
Tetrahedron: Asymmetry 1998, 9, 1779. (g) Bourghida, M.; Widhalm,
M. Terahedron: Asymmetry 1998, 9, 1073. (h) Saitoh, A.; Achiwa, K.;
Morimoto, T. Tetrahedron: Asymmetry 1998, 9, 741. (i) Kunz, H.;
Glaser, B. Synlett 1998, 53. (j) Ahn, K. H.; Cho, C.-W.; Park, J .; Lee,
S. Tetrahedron: Asymmetry 1997, 8, 1179. (k) Brunel, J . M.; Constan-
tieux, T.; Labande, A.; Lubatti, F.; Buono, G. Terahedron Lett. 1997,
38, 5971. (l) Sudo, A., Saigo, K. J . Org. Chem. 1997, 62, 5508. (m) Mino,
T.; Imiya, W.; Yamashita, M. Synlett 1997, 583. (n) Gilbertson, S. R.;
Chang, C.-W. T. Chem. Commun. 1997, 915.
(11) (a) Sprinz, J .; Kiefer, M.; Helmchen, G.; Reggelin, M.; Huttner,
G.; Walter, O.; Zsolnai, L. Tetrahedron Lett. 1994, 35, 1523. (b) Dawson,
G. J .; Williams, J . M. J .; Coote, S. J . Tetrahedron: Asymmetry 1995, 6,
2535.
(12) (a) von Matt, P.; Pfaltz, A. Angew. Chem., Int. Ed. Engl. 1993,
32, 566. (b) Sprinz, J .; Helmchen, G. Tetrahedron Lett. 1993, 34, 1769.
(c) Dawson, G. J .; Frost, C. G.; Williams, J . M. J .; Coote, S. J .
Tetrahedron Lett. 1993, 34, 3149.
Optically active diphosphine ligands with C2-symmetry
have been proven to create a chiral environment ef-
fectively for both substrates and nucleophiles.8 Thus, we
designed a new C2-symmetric diphosphine ligand 13 as
(7) (a) Frost, C. G.; Williams, J . M. J . Tetrahedron Lett. 1993, 34,
2015. (b) Frost, C. G.; Williams, J . M. J . Tetrahedron: Asymmetry 1993,
4, 1785. (c) Dawson, G. J .; Frost, C. G.; Martin, C. J .; Williams, J . M.
J .; Coote, S. J . Tetrahedron Lett. 1993, 34, 7793. (d) Allen, J . V.; Coote,
S. J .; Dawson, G. J .; Frost, C. G.; Martin, C. J .; Williams, J . M. J . J .
Chem. Soc., Perkin Trans. 1 1994, 2065. (e) Morimoto, T.; Tachibana,
K.; Achiwa, K. Synlett 1997, 783. (f) Chesney, A.; Bryce, M. R.; Chubb,
R. W. J .; Batsanov, A. S.; Howard, J . A. K. Tetrahedron: Asymmetry
1997, 8, 3185. (g) Anderson, J . C.; J ames, D. S.; Mathias, J . P.
Tetrahedron: Asymmetry 1998, 9, 753. (h) Adams, H.; Anderson, J . C.;
Cubbon, R.; J ames, D. S.; Mathias, J . P. J . Org. Chem. 1999, 64, 8256.
(i) Boog-Wick, K.; Pregosin, P.; Trabesinger G. Organometallics 1998,
17, 3254.
(13) (a) Loiseleur, O.; Meier, P.; Pfaltz, A. Angew. Chem., Int. Ed.
Engl. 1996, 35, 200. (b) Loiseleur, O.; Hayashi, M.; Schmees, N.; Pfaltz,
A. Synthesis 1997, 1338.
(14) Kudis, S.; Helmchen, G. Angew. Chem., Int. Ed. Engl. 1998,
37, 3047.
(8) Whitesell, J . K. Chem. Rev. 1989, 89, 1581.

4230 J . Org. Chem., Vol. 65, No. 14, 2000
Saitoh et al.
Sch em e 7
1, entry 14). Among the oxazoline ligands, replacement
of an isopropyl group by a tert-butyl group caused
opposite effects of improvement or reduction in the
enantioselectivity, depending upon kinds of ligands
employed.7a-c,10a,12c For instance, oxazolinylferrocene
ligands of the P-N hybrid type with an isopropyl group
on the oxazoline ring demonstrated higher ee than those
of a tert-butyl analogue.10a The ligands having pyrrolidine
or piperidine units in the amidine skeleton showed
essentially the same result as that of VALAP (Table 1,
entries 15 and 16). The effects of amino groups in the
amidine skeleton were ambiguous in the present type of
reaction, for example, chiral diamidine ligands 12a ,b
were not so effective (Table 1, entries 17 and 18). A search
for more potential asymmetric reactions with respect to
this type of ligands is in progress.
The ligands of type 7 were investigated in the reactions
with a sterically less demanding acyclic substrate, 3-pen-
ten-2-yl pivalate 14b (reaction 2 in Scheme 7). As
summarized in Table 2, the use of VALAP resulted in
the production of (R)-15b in 60% ee (Table 2, entry 2), a
comparable level of asymmetric induction as that with a
phosphine-oxazoline ligand.12a Since it has been reported
that improved enantiomeric excess was obtained by
employing the ratio of ligand to metal below unity,10e the
run using the ratio below 1 was attempted, but, no
significant improvement was observed with the present
ligand (Table 2, entry 3). Despite relatively low catalytic
activity, it was suggested that the ligand 7b having a
tert-butyl moiety was somewhat effective compared with
VALAP in terms of enantioselectivity.
As mentioned before, it has been generally accepted
that the nucleophilic attack proceeds predominantly at
the allyl terminus trans to a better π-acceptor, which is
supposed to be the diphenylphosphino group in the
current ligand.7d,11 In the present systems, the nucleo-
philic replacement would proceed through the complex
18a with an M-shaped allyl moiety so as to provide the
R configuration predominantly (Scheme 8).
Electr on ic Su bstitu en t Effects of Hybr id Liga n d s
8 a n d 11 on Asym m et r ic Allyla t ion s of Acyclic
Su bstr a tes. To get a better understanding of the design
of hybrid ligands in asymmetric allylic alkylations, we
next examined the electronic effects. Electronic tuning
of ligands has been one of our interests in catalytic
asymmetric hydrogenations.18 It is well documented that
substituents on the ligands often play key roles electroni-
cally in various types of transition metal-catalyzed asym-
metric reactions.19 Several investigations on the elec-
tronic effects of ligands have been carried out in
solvents, bases, catalyst amounts, reaction temperature,
and reaction time were first investigated in some detail
for VALAP and its related ligands 7b-d on the basis of
our previously obtained results3 (Table 1). In the present
runs, the absolute configuration of the major enantiomer
was determined to be R in all cases by the optical rotation
and the comparison with HPLC peaks. In terms of base,
some difference in the asymmetric induction between the
use of dimethyl malonate/BSA and dimethyl sodioma-
lonate (entries 10 and 12) was observed. The previous
work with (S)-BINAP also revealed the significant reduc-
tion in the enantiomeric excess in the substitution with
dimethyl sodiomalonate.2c,15 In recent studies, most of
substitution reactions were conducted according to Trost’s
procedure16 using BSA in the presence of metal acetate
salts. Under the conditions using dichloromethane (Table
1, entry 1), the alkylated product 15a was obtained with
an excellent level of enantiomeric excess in quantitative
yield.17 Similarly, using the acetate analogue of 14a , an
almost identical level of asymmetric induction (86%, 93%
ee) was achieved under the analogous conditions. A
remarkable variation of the catalytic activity by the
identity of solvents was observed in the reactions using
0.025 equiv of [Pd(η3-C3H5)Cl]2 (Table 1, entries 1-5). The
drop of catalytic activity in THF may be assignable to
an interaction of the solvent with a key π-allyl complex
resulting in reduction of the activity of the cationic
palladium complex and/or the coordination ability of the
bidentate ligand. In contrast, the excellent catalytic
performance was maintained in halogenated solvents
even when the amount of catalyst was reduced to 0.01
equiv of the palladium dimer (Table 1, entries 6 and 7).
The useful catalytic activity was also demonstrated in
the reaction with a considerably small amount of catalyst,
0.005 equiv of [Pd(η3-C3H5)Cl]2 (Table 1, entry 8). From
these experiments, dichloromethane was selected as the
solvent for VALAP and related ligands. Actually, the
reaction in dichloromethane proceeded smoothly within
1 h (Table 1, entry 10). Essentially the same result was
obtained regardless of the countercation of the acetate
salts (Table 1, entry 9). Lowering the reaction tempera-
ture to -30 °C caused slow of the reaction rate without
any improvement in ee (Table 1, entries 11 and 13). In
addition, the catalytic behavior with the other closely
related ligands 7b-d were examined in reaction 1 (Table
1, entries 14-16). The ligand 7b bearing a large sub-
stituent on the chiral center led to lower catalytic
performance in comparison with that of VALAP (Table
(18) For representative articles, see: (a) Takahashi, H.; Hattori, M.;
Chiba, M.; Morimoto, T.; Achiwa, K. Tetrahedron Lett. 1986, 27, 4477.
(b) Morimoto, T.; Takahashi, H.; Fujii, K.; Chiba, M.; Achiwa, K. Chem.
Lett. 1986, 2061. (c) Morimoto, T.; Chiba, M.; Achiwa, K. Tetrahedron
Lett. 1988, 29, 4755. (d) Inoguchi, K.; Morimoto, T.; Achiwa, K. J .
Organomet. Chem. 1989, 370, C9. (e) Takahashi, H.; Achiwa, K. Chem.
Lett. 1989, 305. (f) Takahashi, H.; Morimoto, T.; Achiwa, K. J . Synth.
Org. Chem., J pn. 1990, 48, 29. (g) Inoguchi, K.; Sakuraba, S.; Achiwa,
K. Synlett 1992, 169.
(19) For instance, for asymmetric hydrogenations, see: (a) Rajan-
Babu, T. V.; Ayers, T. A.; Casalnuovo, A. L. J . Am. Chem. Soc. 1994,
116, 4101. (b) RajanBabu, T. V.; Radetich, B.; You K. K.; Ayers, T. A.;
Casalnuovo, A. L.; Calabrese, J . C. J . Org. Chem. 1999, 64, 3429.
Asymmetric epoxidations, see: (c) J acobsen, E. N.; Zhang, W.; Gu¨ler,
M. L. J . Am. Chem. Soc. 1991, 113, 6703. (d) Sasaki, H.; Irie, R.;
Katsuki, T. Synlett 1994, 356. Asymmetric hydrosilylations, see: (e)
Nishiyama, H.; Yamaguchi, S.; Kondo, M.; Itoh, K. J . Org. Chem. 1992,
57, 4306. Asymmetric hydrocyanations, see: (f) RajanBabu, T. V.;
Casalnuovo, A. L. J . Am. Chem. Soc. 1992, 114, 6265. (g) RajanBabu,
T. V.; Casalnuovo, A. L. J . Am. Chem. Soc. 1996, 118, 6325.
(15) (a) Brown, J . M.; Hulmes, D. I.; Guiry, P. J . Tetrahedron 1994,
50, 4493. (b) Yamaguchi, M.; Shima, T.; Yamagishi, T.; Hida, M.
Tetrahedron: Asymmetry 1991, 2, 663.
(16) Trost, B. M.; Murphy, D. J . Organometallics 1985, 4, 1143.
(17) Commercially available dichloromethane with water content
less than 30 ppm (WAKO Chemical) was employed for the present
experiments.

Versatile Chiral Bidentate Ligands
J . Org. Chem., Vol. 65, No. 14, 2000 4231
Ta ble 1. P a lla d iu m -Ca ta lyzed Asym m etr ic Allylic Tr a n sfor m a tion s of 14a Usin g VALAP a n d Its An a logu es a s Liga n d s
in Rea ction 1a
molar equiv of
entry
ligand
solvent
CH2Cl2
ClCH2CH2Cl
toluene
THF
CH3CN
CH2Cl2
ClCH2CH2Cl
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
THF
[Pd(η3-C3H5)Cl]2
base
reaction temp. (°C)
reaction time (h)
yieldb (%)
eec (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
VALAP
VALAP
VALAP
VALAP
VALAP
VALAP
VALAP
VALAP
VALAP
VALAP
VALAP
VALAP
VALAP
7b
0.025
0.025
0.025
0.025
0.025
0.01
BSA-AcOLi
BSA-AcOLi
BSA-AcOLi
BSA-AcOLi
BSA-AcOLi
BSA-AcOLi
BSA-AcOLi
BSA-AcOLi
BSA-AcONa
BSA-AcOLi
BSA-AcOLi
NaH
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
-30
rt
-30
rt
rt
rt
rt
24
24
24
24
24
24
24
24
24
1
98
97
61
53
11
93
90
74
96
88
28
86
26
67
94
92
8
95
94
91
85
94
95
93
93
93
91
94
71
89
88
91
92
46
33
0.01
0.005
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.025
24
1
THF
NaH
24
24
24
24
24
24
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
BSA-AcOLi
BSA-AcOLi
BSA-AcOLi
BSA-AcOLi
BSA-AcOLi
7c
7d
12a
12b
rt
13
a
b
Molar ratio: [Pd(η3-C3H5)Cl]2/ligand/14a /dimethyl malonate/base ) 0.5-2.5/2-10/100/300/300. Isolated yield by preparation TLC
on silica gel (toluene/AcOEt ) 20:1). c The enantiometric excess was determined by HPLC with a chiral column, Daicel Chiralpack AD
(n-hex/IPA ) 20:1).
Ta ble 2. Asym m etr ic Allylic Su bstitu tion s of 14b in
Rea ction 2a
Ta ble 3. Effect of Su bstitu en ts R on Liga n d s 8 a n d 11 in
Rea ction 1a
molar equiv molar ratio of yieldb
eec
(%)
molar equiv of
entry
ligand
of Pd dimer
ligand/Pd
(%)
entry ligand
R
[Pd(η-C3H5)Cl]2 yieldb (%) eec (%)
1
2
3
4
5
VALAP
VALAP
VALAP
7b
0.05
2
2
0.9
2
2
36
27
29
4
59
60
58
67
54
1
2
3
4
5
6
7
8
9
8a
8b
8c
8d
8e
8f
8f
11a
11b
11c
H
0.025
0.025
0.025
0.025
0.025
0.025
0.005
0.025
0.025
0.025
57
42
46
76
88
99
94
75
45
93
52
19
38
74
85
92
89
86
91
84
0.025
0.025
0.025
0.025
CO2Me
CF3
CH3
OMe
NMe2
NMe2
H
7c
8
a
Molar ratio: [Pd(η3-C3H5)Cl]2/ligand/14b/dimethyl malonate/
BSA/AcOLi ) 2.5-5/4.5-20/100/300/300/5. 3-Penten-2-yl pivalate
was derived from commercially available trans 3-penen-2-ol (Al-
F
OMe
b
drich). The conversion yield was determined by GLC due to its
10
volatality. c The enantiomeric excess was obtained using GLC with
a chiral column, CP-Chiralsil-DEX CB.
a
Molar ratio: [Pd(η3-C3H5)Cl]2/ligand/14a /dimethyl malonate/
b
BSA/lithium acetate ) 0.5-2.5/2-10/100/300/300/5. Isolated
yield by preparative TLC on silica gel. c The enantiometric excess
was determined by HPLC with a chiral column, Daicel Chiralpack
AD.
Sch em e 8
8a < 8d < 8e < 8f).6 Among the chiral ligands tested, 8f
bearing a more electron-donating dimethylamino group
at the para position on the phenyl group showed a
dramatic improvement of catalytic activity together with
excellent levels of asymmetric induction, giving 15a
quantitatively (Table 3, entry 6). The allylic substitution
proceeded smoothly in the reduction of the catalyst
amount to a 0.005 molar equiv of [Pd(η3-C3H5)Cl]2 while
still keeping a high level of the enantiomeric excess
(Table 3, entry 7). It was proven that the potential
activity of 8f is superior to that of VALAP. This improved
effect was also demonstrated in reaction 3.
In contrast to the favorable effect with electron-
donating substituents, introduction of electron-withdraw-
ing groups into the ligands afforded detrimental effects.
Although the steric influence of the substituents of the
ligands should be considered, a net role of the electronic
effects by substituents was clearly demonstrated in
entries 3 and 4 where the ligands 8c with a p-CF3 group
and 8d bearing a p-CH3 substituent were used. In this
way, the clear evidence of the electronic influence in P-N
hybrid ligands on palladium-catalyzed asymmetric allylic
transformations could be demonstrated.
asymmetric allylic substitutions.20 Thus, variations of
substituents in a series of ligands 8 and 11 could lead to
alternation in the catalytic activity, which provides useful
information about development of ligands for different
types of catalytic asymmetric transformations. The sub-
stitution effect on the catalytic activity was examined by
using 14a , and the results are tabulated. As can be seen
from Table 3, clear electronic effects of the substituents
in ligands 8 were observed with respect to both the
enantioselectivity and the catalytic activity (8b < 8c <
(20) (a) Allen, J . V.; Bower, J . F.; Williams, J . M. J . Tetrahedron:
Asymmetry 1994, 5, 1895. (b) Nomura, N.; Mermet-Bouvier, Y. C.;
RajanBabu, T. V. Synlett 1996, 745. (c) Porte, A. M.; Reibenspies, J .;
Burgess, K. J . Am. Chem. Soc. 1998, 120, 9180.
It is known that para-substituted styrene derivatives
with electron-donating groups such as NMe2 and OMe

4232 J . Org. Chem., Vol. 65, No. 14, 2000
Saitoh et al.
Ta ble 4. Asym m etr ic Allylic Su bstitu tion s of 16 in
Rea ction 3a
68 and 69% ee, respectively (Table 4, entries 2 and 4).
Among a series of the ligands 8 examined, only 8f
demonstrated significantly efficient catalytic activity, by
which the electronic effect of substituents was again
suggested.
molar equiv of
Pd dimer
entry
ligand
yieldb (%)
eec (%)
1
2
3
4
5
VALAP
0.05
0.05
0.05
0.05
0.025
trace
24
11
25
64
62
68
57
69
>99
7b
7c
8f
Chiral diphosphine ligands with C2-symmetry elements
possessing a large bite angle induced remarkably high
enantioselectivity for sterically less demanding cycloal-
kenyl substrates.22a,c Thus, the newly designed ligand 13
was applied to reaction 3. As expected, 13 exhibited high
level of asymmetric induction, to give (S)-17 over 99%
ee in 64% yield (Table 4, entry 5),9 where the versatility
of our ligands for the variation of reaction types and
substrates was successfully demonstrated. Like the
transformation of the acyclic substrates by VALAP, better
results were obtained by the use of dichloromethane as
a solvent compared with the use of THF. In the latter
case, a decrease in the catalytic activity (99% ee, 11%
yield) was observed.
Asym m etr ic Allylic Tr an sfor m ation s Usin g Keten e
Silyl Aceta ls a s a Nu cleop h ile. Ordinary anionic
species generated from malonates bearing a active meth-
ylene group, so-called stabilized soft nucleophiles, have
been extensively investigated in the asymmetric allylic
alkylations.2 In contrast, unstabilized “harder” nucleo-
philes have attracted little attention, probably due to
their poor ability of asymmetric induction.2c,24 With the
aim of expanding the scope of asymmetric allylic trans-
formations, we focused on ketene silyl acetals (KSA) 19
as alternative nucleophiles to “harder” anionic species.
Since ketene silyl acetals can be readily accessible from
esters having R-hydrogen atoms, it is worthy to examine
the possibility of KSA as nucleophiles.25
Thus, the KSA 1926 was used for asymmetric allylic
substitutions, represented by reaction 4 using 14a as the
substrate, in place of corresponding harder nucleophiles
such as anionic species of monoesters.10h The asymmetric
substitution by 19 proceeded under usual reaction condi-
tions at room temperature, where excellent levels of
catalytic activity were observed (Table 5). The absolute
configuration of 20a was determined by HPLC analysis
comparing with an authentic sample of S-configuration
derived from (R)-15a .10h The reaction conditions discussed
in Table 1 were also applicable to the present reaction,
in which good yield and enantioselectivity in the produc-
tion of 20 (Table 5, entries 1 and 9) were achieved using
dichloromethane as a solvent compared with THF. Ad-
ditionally, the analogous ligands 7b-d exhibited com-
parable levels of asymmetric induction to VALAP, al-
though the catalytic activity was somewhat lowered
(Table 5, entries 4-6). Reduction in the yield was also
observed with KSA 19b having cyclohexyl group at R1
13
Molar ratio: [Pd(η3-C3H5)Cl]2/ligand/16/dimethyl malonate/
a
b
BSA/AcOLi ) 2.5-5/6-12/100/300/300/5. Isolated yield by pre-
parative TLC on silica gel (toluene/AcOEt ) 20/1). c Enantiometric
excess for 17 was determined by GLC with a CP-Chirasil-DEX
CB column.
stabilize cationic palladium(II)-styrene complexes.21 The
electronic effect caused by substituent of ligands in
palladium complexes could affect the bond strength
between metal-ligand in terms of more effective σ-dona-
tion of ligand to palladium than π-back-bonding from the
metal. The predominant formation of reactive cationic
palladium(II) complex prior to a neutral one in the
equilibrium between two intermediate complexes is an
essential factor for proceeding a catalytic cycle.2i However
more cationic palladium(II) complexes activated by elec-
tron-withdrawing substituents tend to interact with
anionic species such as the pivalate anion produced in
the catalytic cycle, to provide neutral complexes. Thus,
the free cationic palladium(II) complex stabilized by
electron-donating groups might play a key role for the
dramatic advancement of catalytic activity.
Variations on catalytic activities in the runs using S-N
hybrid ligands 11 might be explained by the stability of
the cationic palladium(II) complex in a manner similar
to that for the ligands 8. For instance, introduction of
an electron-donating methoxy group would lead to an
improvement of catalytic activity by stabilization of the
cationic complex (Table 3, entry 10). In contrast to 11c,
the use of 11b with an electron-withdrawing group
exhibited opposite effects on both the catalytic activity
and the enantioselectivity.
Asym m etr ic Allylic Su bstitu tion s of Cyclic Allyl
Su bstr a tes. For the purpose of widespread applicability
of the palladium-mediated substitution reaction, we next
investigated asymmetric allylic transformations of steri-
cally less demanding cyclic substrates. Among the sub-
strates, asymmetric induction onto the cyclic allyl sys-
tems has been regarded as one of the most troublesome
processes. Consequently, only a few examples of highly
enantioselective induction over 90% ee have been re-
ported to date.14,22 In the asymmetric allylic transforma-
tions of cyclohexen-2-yl pivalate 16 catalyzed by the Pd-
VALAP complex (reaction 3, Table 4), (S)-17 was obtained
in 62% ee (Table 4, entry 1), which was almost the same
value of asymmetric induction as that with chiral phos-
phanyldihydrooxazoles.23 The ligands 7b and 8f were
found to be more effective in reaction 3, giving (S)-17 in
(24) (a) Fiaud, J .-C.; Aribi-Zouioueche, L. J . Organomet. Chem. 1985,
295, 383. (b) Fiaud, J .-C.; Legros, J .-Y. Tetrahedron Lett. 1991, 32,
5089. (c) Fotiadu, F.; Cros, P.; Faure, B.; Buono, G. Tetrahedron Lett.
1990, 31, 77.
(25) Palladium-catalyzed allylic substitutions using KSA and achiral
ligands, see: (a) Tsuji, J .; Minami, I.; Shimizu, I. Chem. Lett. 1983,
1325. (b) Tsuji, J .; Minami, I.; Shimizu, I. Tetrahedron Lett. 1983, 24,
1793. (c) Shimizu, I.; Minami, I.; Tsuji, J . Tetrahedron Lett. 1983, 24,
1797. (d) Tsuji, J .; Takahashi, K.; Minami, I.; Shimizu, I. Tetrahedron
Lett. 1984, 25, 4783. (e) Carfagna, C.; Mariani, L.; Musco, A.; Sallese,
G. J . Org. Chem. 1991, 56, 3924. (f) Carfagna, C.; Galarini, R.; Musco,
A.; Santi, R. J . Mol. Catal. 1992, 72, 19. For iron complexes for the
reaction of allyl acetate with KSA, see: (g) Enders, D.; Frank, U.; Fey,
P.; J andeleit, B.; Lohray, B. B. J . Org. Chem. 1996, 61, 147. Molyb-
denum- and tungsten-catalyzed allylic substitutions, see: Malkov, A.
V.; Baxendale, I. R.; Dvora´k, D.; Mansfield, D. J .; Kocovsky, P. J . Org.
Chem. 1999, 64, 2737.
(21) Kurosawa, H.; Majima, T.; Asada, N. J . Am. Chem. Soc. 1980,
102, 6996.
(22) (a) Trost, B. M.; Bunt, R. C. J . Am. Chem. Soc. 1994, 116, 4089.
(b) Knu¨hl, G.; Sennhenn, P.; Helmchen, G. J . Chem. Soc., Chem
Commun. 1995, 1845. (c) Dierkes, P.; Ramdeehul, S.; Barloy, L.; Cian,
A. D.; Fischer, J .; Kamer, P. C. J .; van Leeuwen, P. W. N. M.; Osborn,
J . A. Angew. Chem., Int. Ed. Engl. 1998, 37, 3116.
(23) For asymmetric allylations of cycloalkenyl substrates by phos-
phanyldihydrooxazoles, see: (a) Sennhenn, P.; Gabler, B.; Helmchen,
G. Tetrahedron Lett. 1994, 35, 8595. (b) Gilbertson, S. R.; Chang, C.-
W. T. J . Org. Chem. 1998, 63, 8424. By other P-N hybrid ligands, see
refs 10g and 15a.

Versatile Chiral Bidentate Ligands
J . Org. Chem., Vol. 65, No. 14, 2000 4233
Ta ble 5. Asym m etr ic Allylic Su bstitu tion s Usin g Keten e
Silyl Aceta lsa
Ta ble 6. Asym m etr ic Allylic Su bstitu tion s w ith Ester
En ola tes 21a
molar equiv of
yieldb eec
reaction
entry
ligand
KSA [Pd(η3-C3H5)Cl]2 solvent
(%)
(%)
temp reaction yieldb eec
entry enolate
M
solvent
(°C)
time (h) (%) (%)
1
2
3
4
5
6
7
8
9
VALAP
VALAP
VALAP
7b
7c
7d
(S)-BINAP 19a
(S)-BINAP 19a
VALAP
VALAP
7b
19a
19a
19a
19a
19a
19a
0.025
0.01
CH2Cl2
CH2Cl2
THF
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
THF
CH2Cl2
THF
THF
CH2Cl2
CH2Cl2
THF
78
43
13
17
30
45
89
6
90
87
74
81
89
87
65
83
90
77
79
1
2
3
4
5
6
7
8
9
21a
21a
21a
21b
21a
21a
21a
21a
21
Li
Li
Li
Li
THF
THF
THF
THF
rt
rt
-20
rt
rt
rt
-20
rt
rt
rt
rt
rt
24
1
24
24
24
1
24
24
24
24
24
24
24
89
75
73
92
88
76
11
27
41
19
36
2
59
59
59
57
12
13
14
14
25
50
23
29
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.05
n-Bu4N THF
n-Bu4N THF
n-Bu4N THF
n-Bu4N DME
n-Bu4N toluene
n-Bu4N CH2Cl2
n-Hex4N THF
n-Hex4N toluene
n-Hex4N CH2Cl2
19b
19b
19b
19c
19d
19d
42
17
11
10
11
10
11
12
13
21a
21a
21a
21a
12d VALAP
13
14
VALAP
VALAP
0.025
0.025
93
25
93
78
rt
trace
a
Molar ratio: [Pd(η3-C3H5)Cl]2/ligand/14a /19 ) 1-5/4-20/100/
a
Molar ratio: [Pd(η3-C3H5)Cl]2/ligand/14a /21 ) 2.5/10/100/300.
b
Isolated yield by preparative TLC on silica gel (toluene). c The
enantiometric excess was determined by using HPLC with a chiral
column, Daicel Chiralpak AD (n-hex/IPA ) 40:1).
b
300. Isolated yield by preparative TLC on silica gel (toluene).
c The enantiometric excess was determined by using HPLC with
d
a chiral column. Daicel Chiralpak AD (n-hex/IPA ) 40:1). Reac-
tion time: 85 h.
replacement with the stabilized anionic nucleophile
yielded 15a moderately. Although the same result was
obtained with 8e, ligand 8f with fine electronic tuning
was the sole ligand that exerted significant catalytic
activity (32%, 67% ee) in the allylic replacement by 19a .
The change in the catalytic activity could support the
mechanistic consideration based on the electronic sub-
stituent effect of ligands. As mentioned above, it can be
considered that stabilization of the cationic palladium-
(II) complex is a key factor to conduct the reaction,
especially in the case of less reactive nucleophiles.
Recently, the quantitative investigation on the reactivity
between cationic metal electrophiles and noncharged
nucleophiles such as KSA and silyl enol ethers was
reported, which suggests the reactivity of KSA toward
the cationic palladium complex lies near a borderline.27
Hence, the degree of stabilization of the cationic palla-
dium(II) complex caused by variations of electronic effects
of the ligands would result in significant change of
catalytic activity.
and R2 (Table 5, entries 9-11), where some repulsive
steric interaction between the ligand and the KSA might
operate. The generality of this reaction was exhibited by
runs using (S)-BINAP (Table 5, entries 7 and 8). Finally,
no reactions with 19c occurred even when the amount
of the catalyst and the reaction time were increased. This
result should originate from the lower nucleophilicity of
19c compared with 19a ,b.
The reactive anionic species might be produced directly
from dimethyl malonate in the well-known procedure16
for the generation of the soft anionic nucleophile in the
system of dimethyl malonate, BSA, and acetate salts.
When dimethyl malonate was treated with BSA in the
presence of AcOLi, formation of carbomethoxy ketene
methyltrimethylsilyl acetal 19d could not be detected by
1H NMR analysis.10h This observation implies that BSA
does not work as a silylating agent leading to 19d , but
acts as a base. The reaction probably proceeds by the
nucleophilic attack of KSA on the π-allyl complex through
an interaction of π-orbitals between the π-allyl and the
carbon-carbon double bond of KSA. Another possible
path is the nucleophilic attack of pivalate anion to the
silyl group. The latter possibility might be adopted to the
reactions with 19d because satisfactory conversion was
realized regardless of its lower nucleophilicity due to the
electron-withdrawing group.
The soft-hard principle is an interesting subject
related to the influence upon the course of nucleophilic
displacement reaction. For comparison to the reaction
with KSA itself, the reaction with the “harder” ester
enolates 21 generated from the reaction of KSA with
methyllithium and tetraalkylammonium halides was
next examined.
The capability of the imine-phosphine hybrid ligands
8 for the substitution with 19a was then examined. The
reaction using 8a with no substituents on the phenyl ring
afforded only a trace amount of 20a , whereas the
The results are summarized in Table 6, where a
considerably different reactivity was demonstrated. The
reaction with 21a in THF resulted in lower enantiose-
lectivity and higher reactivity compared with that using
19a itself (Table 6, entries 1, 2, 5, and 6). The same
behavior was observed in the use of the lithium enolate
21b (Table 6, entry 4). Again, these results support that
the reaction with KSA proceeds through the interaction
between π-allylpalladium complex and KSA itself. In the
reactions of ester enolates bearing tetraalkylammonium
groups as a cationic part, a clear effect of both the size of
ion pair and solvents on the reaction was observed. Trost
(26) Ketene methyltrimethylsilyl acetal 19c was prepared as a 7:3
mixture with a C-silylated compound, methyltrimethylsilyl acetate.
Carbomethoxy ketene methyltrimethylsilyl acetal 19d tends to be
hydrolyzed easily. Therefore, the hydrolyzed product dimethyl mal-
onate was removed by distillation just before conducting the asym-
metric reaction. For articles referring to the preparation of ketene silyl
acetals, see: (a) Ainsworth, C.; Chen, F.; Kuo, Y.-N. J . Organomet.
Chem. 1972, 46, 59. (b) Rathke, M. W.; Sullivan, D. F. Synth. Commun.
1973, 3, 67. (c) Kita, Y.; Haruta, J .; Segawa, J .; Tamura, Y. Tetrahedron
Lett. 1979, 4311.

4234 J . Org. Chem., Vol. 65, No. 14, 2000
Saitoh et al.
and co-workers reported the same trend of reaction
influenced by counterions in the reactions of cyclopenten-
2-yl acetate.2c,22a As expected from the previous study,22a
the use of dichloromethane as a solvent caused an
improvement in ee, probably due to the tightening ion
pair or dimerization of enolates (Table 6, entry 10).28
However, the degree of asymmetric induction was mod-
erate compared with those listed in Table 6. In this way,
an effectiveness of the approach employed KSA itself as
a nucleophile was reconfirmed.
Mech a n istic View s Con cer n in g th e En a n tiod is-
cr im in a tion in Allylic Su bstitu tion s. Mechanistic
studies on the stereochemistry and reaction course of
palladium-catalyzed allylic substitutions by soft nucleo-
philes have been well investigated.29 The transformation
consists of two sequential inversion processes, in which
a π-allylpalladium(II) complex is first formed with the
inversion through the substitution of the leaving group
by a palladium(0) complex, followed by the nucleophilic
attack to the allylic position from the site opposite to the
metal. Despite the well-established catalytic cycle, reac-
tion pathway, and stereochemistry, a key step for enan-
tiodetermination is still ambiguous for these transfor-
mations. Recently, it has been reported that a late
transition state (product-like) is a key step in the
enantiodiscrimination.15a,22c,30 Thus, after nucleophilic
attack, a rotation occurs to give the Pd(0)-olefin complex,
from which the products are released. Osborn and co-
workers proposed a theory that the nucleophilic attack
to the π-allyl complex and the subsequent rotation to the
Pd-olefin complex proceed so as to avoid unfavorable
steric repulsion.22c
Generally, for applying to various types of chiral
ligands, the rotational mechanism for the enantiodeter-
mining step was taken into consideration by the combi-
nation with the P/M chirality concept,31 where the
positioning array of four phenyl rings of diphosphine
ligands, established by X-ray analysis, closely correlates
with the absolute configuration of products in asymmetric
hydrogenations.9 Herein, we illustrate Pr(plus region)/
Mr(minus region) chirality model to classify various
chiral ligands and visualize the bulky quadrant in Figure
1. The gray circles are in accordance with the sterically
hindered region assignable to quasiequatorial phenyl or
bulky substituents.
F igu r e 1. Pr/ Pr chirality model.
Next, the correlation between the Pr/Mr chirality of
ligands and the absolute configuration of products was
investigated. Commercially available ligands and re-
cently developed chiral auxiliaries are selected and
classified as listed in Table 7, which includes new ligands
in this study. (S)-BINAP, (S,S)-Chiraphos, and (S,S)-
Norphos with Mr chirality32 afforded (R)-15a in reaction
1,15b demonstrating a good correlation with the Pr/Mr
chirality (Table 7, entries 4-6). (S)-17 was produced
using (S)-BINAP under the same conditions as those of
entry 1 in Table 4 (Table 7, entry 14). Chiral oxazoline
ligands 22a with Pr chirality11a provided (S)-15a (Table
7, entry 3),12 and (R)-17 was obtained by 22b (Table 7,
entry 12).14,23a A C2-symmetric diamino ligand, (-)-R-
isosparteine 23, which is considered to have Mr chirality
from its molecular structure, gave the products with the
expected configurations (Table 7, entries 7 and 15).33 (R,
R)-Duthixantphospholane 24 with Mr chirality also
exhibited reasonable results for both acyclic and cyclic
substrates as discussed in the preceding paper by Osborn
and co-workers (Table 7, entries 8 and 16).22c The use of
VALAP and 8f led to the production of (R)-15a and (S)-
17, which can correlate with Mr chirality of ligands
(entries 1, 2, 9, and 10). Actually, from the study on the
structure of a rhodium complex bearing a similar type
of compound, it was suggested that the equatorial phenyl
(32) Crystal structures of palladium-BINAP complexes have been
elucidated by several groups, in which equatorial phenyl substituents
on each phosphorus atom of (S)-BINAP are found to occupy the
positions assignable to Mr chirality; see: (a) Ozawa, F.; Kubo, A.;
Matsumoto, Y.; Hayashi, T.; Nishioka, E.; Yanagi, K.; Moriguchi, K.
Organometallics 1993, 12, 4188. (b) Pregosin, P. S.; Ru¨egger, H.;
Salzmann, R.; Albinati, A.; Lianza, F.; Kunz, R. W. Organometallics
1994, 13, 83. (c) Pregosin, P. S.; Ru¨egger, H.; Salzmann, R.; Albinati,
A.; Lianza, F.; Kunz, R. W. Organometallics 1994, 13, 5040. (d)
Yamaguchi, M.; Yabuki, M.; Yamagishi, T.; Sakai, K.; Tsubomura, T.
Chem. Lett. 1996, 241. (e) Kuwano, R.; Ito, Y. J . Am. Chem. Soc. 1999,
121, 3236. In ref 32b,c, structural studies for palladium-(S,S)-
Chiraphos complexes were presented using multidimensional NMR
spectroscopy and MM2 calculations. On the basis of these data, Mr
chirality of (S,S)-Chiraphos was pointed out although less equatorial
and axial character was demonstrated compared with the palladium-
BINAP complex. Moreover, the X-ray crystal structure of rhodium
complex with (R,R)-Chiraphos suggests that the ligand has two
equatrial phenyl groups at the Pr position; see: Halpern, J . In
Asymmetric Synthesis Vol. 5; Morruson, J . D., Ed.; Academic Press:
New York, 1985. It is reasonable that (S,S)-Norphos is assigned to have
Mr chirality from a similarity of molecular structure to (S,S)-Chira-
phos. The results in rhodium-catalyzed asymmetric hydrogenation of
N-benzoylhydrazone using (R,R)-Norphos also supported the assign-
ment; see: Yamazaki, A.; Achiwa, I.; Horikawa, K.; Tsurubo, M.;
Achiwa, K. Synlett 1997, 455.
(27) Kuhn, O.; Mayr, H. Angew. Chem., Int. Ed. Engl. 1999, 38, 343.
(28) It has been demonstrated in the study for X-ray crystal
structures of tetraalkylammonium enolates derived from malonates
that the enolates present as dimeric structure even if the crystal was
formed in DMSO; see: Reetz, M. T.; Hu¨tte, S.; Goddard, R. J . Am.
Chem. Soc. 1993, 115, 9339.
(29) (a) Trost, B. M.; Weber, L.; Strege, P. E.; Fullerton, T. J .;
Dietsche, T. J . Am. Chem. Soc. 1978, 100, 3435. (b) Trost, B. M.;
Verhoeven, T. R. J . Am. Chem. Soc. 1978, 100, 3416. (c) Trost, B. M.;
Verhoeven, T. R. J . Am. Chem Soc. 1980, 102, 4730. (d) Hayashi, T.;
Hagihara, T.; Konishi, M.; Kumada, M. J . Am. Chem. Soc. 1983, 105,
7767. (e) A° kermark, B.; Ba¨ckvall, J . E.; Lowenborg, A.; Zetterberg, K.
J . Organomet. Chem. 1979, 166, C33. (f) A° kermark, B.; J utand, A. J .
Organomet. Chem. 1981, 217, C41.
(30) (a) von Matt, P.; Lloyd-J ones, G.; Minidis, A. B. E.; Pfaltz, A.;
Macko, L.; Neuburger, M.; Zehnder, M.; Regger, H.; Pregosin, P. S.
Helv. Chim. Acta 1995, 78, 265. (b) Steinhagen, H.; Reggelin, M.;
Helmchen, G. Angew. Chem., Int. Ed. Engl. 1997, 36, 2108.
(31) (a) Sakuraba, S.; Morimoto, T.; Achiwa, K. Tetrahedron: Asym-
metry 1991, 2, 597. The P/M chirality concept was further generalized
by Pz(plus zone)/Mz(minus zone) chirality concept in which the course
of reaction in asymmetric hydrogenations using chiral bisphospholane
ligands could be well explained. (b) Achiwa, K. Principles and Ap-
plications of Chiral Technology; Industrial Publishing & Consulting:
J apan, 1999. (c) Fujie, N.; Matsui, M.; Achiwa, K. Chem. Pharm. Bull.
1999, 47, 436.
(33) Kang, J .; Cho, W. O.; Cho, H. G. Tetrahedron: Asymmetry 1994,
5, 1347.

Versatile Chiral Bidentate Ligands
J . Org. Chem., Vol. 65, No. 14, 2000 4235
Ta ble 7. Rela tion sh ip betw een P r /Mr Ch ir a lity of
Liga n d s a n d Absolu te Con figu r a tion of P r od u cts in
Asym m etr ic Allyla tion s
directed to one side. Actually, an example assignable to
the consideration was recently demonstrated in the
imine-sulfide hybrid ligand.7h In this case, it was clearly
shown by the X-ray analysis that the o-chlorophenyl
group on the imino nitrogen is sterically more effective
than the phenyl substituent on the sulfur atom. Although
the bond length between the palladium metal and the
allyl carbon trans to the sulfide is longer than that trans
to the imino nitrogen, the nucleophilic attack may
procced at the position trans to the nitrogen by the
predominant steric control compared with the electronic
effect.
Pr/ Mr
configuration
(% ee)
entry
ligand
VALAP
8f
22a
(S)-BINAP
(S,S)-Ch ir a p h os
(S,S)-Nor p h os
23
24
VALAP
chirality substratea
ref
Adaptability of the consideration mentioned above give
helpful information for design of effective ligands as well
as prediction of the stereochemistry of products in this
type of transformations.
1
2
3
4
5
6
7
8
9
Mr
Mr
Pr
A
A
A
A
A
A
A
A
B
B
B
B
B
B
B
R (95)
R (92)
S (94-98.5) 12
3
6
Mr
Mr
Mr
Mr
Mr
Mr
Mr
Mr
Pr
R (30)
R (90)
R (81)
R (88)
R (97)
S (62)
S (69)
S (>99)
R (51)
S (34)
S (62)
S (93)
15b
15b
15b
33
22c
current
6
Con clu sion
It is meaningful to consider that the versatility on
designing chiral ligands, which can easily lead to a
variety of structural modification, is applicable to differ-
ent types of reactions and substrates. With this concept,
a new class of chiral amidine-phosphine hybrid ligands
7 were first developed by using chiral source of R-amino
acids, L-valine and L-tert-leucine. A variety of chiral P-N
or S-N hybrid ligands 8 and 11 with electronic tuning
and a new C2-symmetric diphosphine ligand 13 were
readily prepared from the synthetic precursor of 7a
(VALAP). Replacement of the dimethylamino group by
secondary amines such as pyrrolidine and piperidine in
the amidine skeleton was successfully demonstrated. In
asymmetric allylic substitutions of 1,3-diphenylpropen-
2-yl pivalate (14a ) with the anion of dimethyl malonate,
high enantioselectivity of up to 95% ee was achieved
using newly developed hybrid ligands. Though the hybrid
ligands were less effective to palladium-catalyzed allylic
alkylations of sterically less demanding cyclohexen-2-yl
pivalate (17), extremely high levels of asymmetric induc-
tion of over 99% ee were attained by the use of diphos-
phine ligand 13. In this way, the versatility of the new
ligands was demonstrated.
The electronic influence of ligands on the asymmetric
allylations with 8 was investigated and clearly demon-
strated that the use of chiral ligand 8f having a more
electron-donating dimethylamino group at the para posi-
tion on the phenyl group led to a dramatic improvement
of catalytic activity together with an excellent level of
asymmetric induction.
With the aim of expanding the synthetic scope of
asymmetric allylic substitutions, ketene silyl acetals were
used as nucleophiles that afforded excellent levels of
enantioselective alkylations. This approach can extend
not only the utility of asymmetric alkylations but also
applicability of the less stabilized “harder” nucleophiles
such as anionic species of monoesters in organic synthe-
sis.
10 8f
11 13
12 22b
13 (S)-BINAP
14 23
9
23a
current
33
Mr
Mr
Mr
15 24
22c
a
A: 1,3-diphenylpropen-2-yl acetate or pivalate. B: cyclohexen-
2-yl acetate or pivalate.
group belongs to Mr chirality.34 The diphosphine ligand
13, which affords (S)-17, is predicted to have Mr chirality
from the view for VALAP.
The good correlation described above can be explained
by mechanistic considerations for an enantiodiscriminat-
ing step in C2-symmetric and heterohybrid ligands as
depicted in Figure 2a,b. For the C2-symmetric ligands, a
possible conformation of π-allyl palladium complexes is
limited due to the symmetrical element of ligands. Thus,
the direction of nucleophilic attack could be controlled
to avoid the steric repulsion in the late transition state.
In two possible π-allyl complexes of the hybrid ligands,
the nucleophilic attack proceeds at the allyl terminus
trans to the better π-accepting phosphino group. The
predominant π-allyl complex, which leads to a sterically
favorable palladium-olefin complex in the late transition
state, would exclusively receive the nucleophilic attack.
On the basis of this consideration, it is predicted that
the ligand type described in Figure 2c may give a
significant effect on the high level of asymmetric induc-
tion for cyclic allyl compounds. The phosphine-oxazoline
hybrid ligand modified by the cymantrene unit,14 in
which Pr chirality on the nitrogen and Mr chirality on
the phosphorus were demonstrated in the X-ray analysis,
could be assigned to the case of Figure 2c. The use of the
oxazoline ligand resulted in high levels of asymmetric
induction for cycloalkenyl substrates.
As depicted in Figure 2d for the heterohybrid type with
the predominant steric effect on the nitrogen site, a
possibility through the nucleophilic attack to the allyl
terminus trans to the nitrogen may be considered in the
case that the steric effect on the nitrogen site is large
and the equilibrium between two possible complexes is
In a mechanistic consideration of asymmetric allylic
transformations of both acyclic and cyclic allyl substrates,
we considered Pr(plus region)/Mr(minus region) chirality
model with respect to the rotational mechanism for the
stereodetermining step for applying to various types of
chiral ligands, by which a good correlation between the
Pr/Mr chirality of ligands and the direction of enantio-
discrimination at the late transition state was demon-
strated.
(34) The X-ray crystal structure of a rhodium complex with a P-N
hybrid ligand based on L-valine consisting of bis (4-methylphenyl)-
phosphino and 4-methoxybenzylamino groups indicates that the ligand
has Mr chirality; see: Berger, H.; Nesper, R.; Pregosin, P. S.; Ru¨egger,
H.; Wo¨rle, M. Helv. Chim. Acta 1993, 76, 1520.

4236 J . Org. Chem., Vol. 65, No. 14, 2000
Saitoh et al.
F igu r e 2. Preferable paths predicted by Pr/ Mr chirality model
In conclusion, we believe the newly developed ligand
systems in this study can provide us useful information
about the design of chiral ligands. Studies on applying
other type of reactions and substrates are in progress.
alcohol matrix, and the measurement was performed
using xenone atoms. High-resolution mass spectra (HRMS)
were obtained at the Institute for Chemical Research,
Kyoto University. Gas chromatographic analysis was
performed using an SE-30 capillary column (0.25 mm ×
50 m) for a general analysis or Chirasil-DEX CB(0.25 mm
× 50 m) for determining the enantiomeric excess of
optically active products. High-pressure liquid chroma-
tography (HPLC) with a Daicel Chiralpak AD column
was employed to determine the enatiomeric excess.
Melting points are uncorrected. Column chromatographic
isolation was conducted using silica gel 60 (70-230 mesh,
Merck). Silica gel 60 F254 aluminum plates (Merck) were
Exp er im en ta l Section
1
Gen er a l P r oced u r es. H, 13C and 31P NMR spectra
were recorded in CDCl3 solution at 270, 67.8, and 161.7
MHz, respectively. Chemical shift values are expressed
in ppm relative to tetramethylsilane for 1H, CDCl3 for
13C, and phosphoric acid for 31P. IR spectra were obtained
using an FT-IR instrument. For fast-atom bombardment
(FAB)-MS, the samples were dissolved in 3-nitrobenzyl

Versatile Chiral Bidentate Ligands
J . Org. Chem., Vol. 65, No. 14, 2000 4237
1
employed as a thin-layer chromatography (TLC). Silica
gel 60 F254 (0.5 and 2 mm, Merck) was used for
preparative TLC.
mp 79-81 °C; H NMR (270 MHz, CDCl3) δ: 0.85 (s,
9H), 1.42 (s, 9H), 1.89-1.99, 2.34-2.40 (m, 2H), 3.44-
3.56 (m, 1H), 4.29 (d, 1H, J ) 10.8 Hz), 7.28-7.45 (m,
10H); IR (KBr) 3410, 1698, 1514 cm-1; FAB-MS m/z 386
(MH+).
Rea gen ts. In general, all organic reagents were used
as received. Dimethyl malonate was distilled in vacuo
and dried over molecular sieves. THF was distilled over
sodium metal/benzophenone ketyl and used as a peroxide-
free solvent. For the catalytic reactions, dehydrated
dichloromethane, toluene, and acetonitrile were pur-
chased and degassed before exposure to the reaction.
Dehydrated and degassed 1,2-dichloroethane and DME
were dried over activated molecular sieves. Pyridine was
dried by storing in the presence of sodium hydroxide.
Tetrabutylammonium fluoride and terahexylammonium
bromide were dried in an oil bath at 70-80 °C under
reduced pressure (4 mmHg) prior to use.
(S )-1-(Dip h e n ylp h osp h in o)-3-m e t h yl-2-b u t a n -
a m in e (6b). To a stirred solution of 5b (250 mg, 0.649
mmol) in CH2Cl2 (25 mL) was added trifluoroacetic acid
(2.4 mL, 30.9 mmol) in an ice-water bath under argon.
The solution was stirred at low temperature for 1 h and
room temperature overnight. The resulting solution was
quenched with water. The water layer neutralized with
aqueous NaOH was extracted with CH2Cl2. The combined
organic layers were washed with saturated aqueous
NaHCO3 and brine. From the solution dried over MgSO4,
the solvent was removed under vacuum, giving 6b as a
viscous liquid (158 mg, 85%): [R]28D +141.5 (c 0.6, CHCl3);
1H NMR (270 MHz, CDCl3) δ 0.87 (s, 9H), 1.71-1.82 (m,
1H), 2.41-2.50 (m, 2H), 7.29-7.52 (m, 10H); IR (neat)
3382, 1591 cm-1; FAB-MS m/z 285 (MH+).
Gen er a l P r oced u r e for 7a ,b. The preparation of 3a -
6a was documented in the preceding article.9b
(S)-[1-(H yd r oxym et h yl)-2,2-d im et h ylp r op yl]ca r -
ba m ic Acid 1,1-Dim eth yleth yl Ester (3b). To a mix-
ture of L-tert-leucinol (2b) (1 g, 8.55 mmol) and Et3N (1.02
g, 10.1 mmol) in CH2Cl2 (30 mL) was added di-tert-butyl
dicarbonate (2.21 g, 10.1 mmol) in an ice-cold bath at 0
°C, followed by stirring at 0 °C for 1 h and at room
temperature for 8 h. The reaction mixture was washed
with brine and dried over MgSO4. Volatile fractions were
(S)-N′-[1-[(Dip h en ylp h osp h in o)m eth yl]-2-m eth yl-
p r op yl]-N,N-d im et h ylm et h a n im id a m id e, VALAP
(7a ). A mixture of 6a (1.90 g, 7 mmol) and N,N-
dimethylformamide dimethylacetal (4.49 g, 5 mL, 37.7
mmol) was stirred at room temperature for 3 h. The
reaction was monitored by gas chromatography. After
removal of the volatile fractions in vacuo, VALAP 7a
was obtained in a virtually quantitative yield. The
product was purified as a colorless viscous liquid by silica
gel column chromatography eluted by toluene including
0.5 vol % of Et3N, followed by checking the purity by
removed in vacuo, affording 3a as a white solid (1.72 g,
1
93%): [R]23 +8.5 (c 0.4, MeOH); mp 102-103 °C; H
D
NMR (270 MHz, CDCl3) δ 0.94 (s, 9H), 1.46 (s, 9H), 2.17
(brs, 1H), 3.49 (d, 2H, J ) 7.6 Hz), 3.80-3.87 (m, 1H),
4.62 (brs, 1H); 13C NMR (68 MHz, CDCl3) δ 26.8, 28.4,
33.6, 61.0, 63.2, 79.6; IR (KBr) 3376, 1680, 1562 cm-1
;
using gas chromatography before use: [R]28 +34.1 (c
0.2, CHCl3); H NMR (270 MHz, CDCl3) δ 0.83 (d, 3H,
D
1
FAB-MS m/z 218 (MH+).
J ) 6.6 Hz), 0.87 (d, 3H, J ) 6.6 Hz), 1.73-1.83 (m,
1H), 2.37 (d, 2H, J ) 3.5 Hz), 2.67 (s, 6H), 2.69-2.79 (m,
1H), 6.98 (s, 1H), 7.25-7.44 (m, 10 H); 31P{1H} NMR
(161.7 MHz, CDCl3, H3PO4) δ -17.89; IR (neat) 1651
cm-1; HRMS (EI) calcd for C20H27N2P 326.1914, found
326.1905.
(S)-[-2,2-Dim eth yl-1-[[[(4-m eth ylp h en yl)su lfon yl]-
oxy]m eth yl]p r op yl]ca r ba m ic Acid 1,1-Dim eth yleth -
yl Ester (4b). p-Toluenesulfonyl chloride (1.55 g, 8.12
mmol) dissolved in pyridine (5 mL) was added dropwise
to 3b (1.64 g, 7.56 mmol) in pyridine (15 mL) at -35 °C
under argon. The resulting solution was stirred at low
temperature for 18 h and in an ice-water bath for 1 h.
The excess pyridine was quenched with 10% HCl aqueous
solution. The product thus obtained was extracted into
AcOEt. The extraction was successively washed with
saturated aqueous NaHCO3 and brine. The reaction
mixture was dried over MgSO4 and concentrated under
vacuum. The product 4b was isolated as a white wax
using silica gel column chromatography eluted by a 20:1
mixture of toluene and AcOEt (1.54 g, 55%): [R]D26 -1.04
(c 0.6, MeOH); 1H NMR (270 MHz, CDCl3) δ 0.90 (s, 9H),
1.42 (s, 9H), 3.57-3.64 (m, 1H), 4.04-4.17 (m, 2H), 4.62
(brd, 1H, J ) 9.6 Hz), 7.35 (d, 2H, J ) 8.2 Hz), 7.79 (d,
2H, J ) 8.3 Hz); 13C NMR (68 MHz, CDCl3) δ 21.6, 26.7,
28.2, 34.1, 56.8, 69.5, 79.4, 127.9, 129.8, 132.6, 144.9,
155.6; IR (KBr) 3299, 1721, 1537, 1358, 1177 cm-1; FAB-
MS m/z 372 (MH+).
(S )-N ′-[1-[(Dip h e n ylp h osp h in o)m e t h yl]-2,2-d i-
m eth ylpr opyl]-N,N-dim eth ylm eth an im idam ide (7b).
The same procedure as that for 7a was employed. Viscous
liquid: yield quant; [R]27 +63.1 (c 0.6, CHCl3); 1H NMR
D
(270 MHz, CDCl3) δ 0.86 (s, 9H), 2.32 (d, 2H, J ) 6.9
Hz), 2.57-2.64 (m, 1H), 2.68 (s, 6H), 6.92(s, 1H), 7.29-
7.45 (m, 10H); 31P{1H} NMR (161.7 MHz, CDCl3, H3PO4)
δ -16.81; IR (KBr) 1649 cm-1; HRMS (EI) calcd for
C21H29N2P 340.2054, found 340.2086.
Gen er a l P r oced u r e for th e Su bstitu tion Rea c-
tion s of th e Dim eth yla m in o Gr ou p of VALAP w ith
Secon d a r y Am in es. The replacement reaction to 7c is
typical.
(S)-[1-[(Dip h en ylp h osp h in o)m eth yl]-2-m eth ylp r o-
p yl]-N-[(1-p yr r olid in o)m eth ylen e]a m in e (7c). A mix-
ture of VALAP 7a (250 mg, 0.767 mmol) and pyrrolidine
(1.70 g, 2 mL, 24.0 mmol) was degassed and heated in
the presence of 10-camphorsulfonic acid (10 mg, 0.0430
mmol) at reflux under argon. The reaction was periodi-
cally monitored by gas chromatography. The reaction
mixture was stirred overnight and cooled to room tem-
perature. The resulting solution was diluted with toluene,
followed by washing the organic phase with saturated
aqueous NaHCO3 and brine. The organic phase was dried
over MgSO4, and the solvent was evaporated. The residue
was purified using silica gel column chromatography
eluted by toluene including 0.5 vol % of Et3N, followed
(S)-[1-[(Dip h en ylp h osp h in o)m et h yl]-2,2-d im et h -
ylp r op yl]ca r ba m ic Acid 1,1-Dim eth yleth yl Ester
(5b). Potassium diphenylphosphide 0.5 M THF solution
(7.9 mL, 3.95 mmol) was added to 4b (700 mg, 1.89 mmol)
in THF (35 mL) at -35 °C under argon. After being
stirred for 10 h while cooling was continued, the solution
was allowed to warm to room temperature and filtered
through Celite. After evaporation of the solvent, 5b was
isolated as a white solid (315 mg, 43%) by silica gel
column chromatography using toluene containing 0.5
vol % of Et3N as the eluent: [R]27 +43.2 (c 0.6, CHCl3);
D

4238 J . Org. Chem., Vol. 65, No. 14, 2000
Saitoh et al.
by checking the purity by using gas chromatography
CDCl3) δ 0.90 (d, 3H, J ) 6.3 Hz), 0.92 (d, 3H, J ) 6.3
Hz), 1.91-2.05 (m, 1H), 2.48 (d, 2H, J ) 6.6 Hz), 2.95-
3.05 (m, 1H), 3.83 (s, 3H), 6.87 (d, 2H, J ) 8.9 Hz), 7.22-
7.56 (m, 1H), 7.95 (s, 1H); IR (neat) 1644, 1252, 1032
cm-1; HRMS (EI) calcd for C25H28ONP 389.1910, found
389.1929.
before the use. The modified amidine 7c was obtained
as a viscous liquid (154 mg, 57%): [R]23 +45.8 (c 0.6,
D
1
CHCl3); H NMR (270 MHz, CDCl3) δ 0.82 (d, 3H, J )
6.6 Hz), 0.89 (d, 3H, J ) 6.6 Hz), 1.75-1.80 (m, 4H), 2.40
(d, 2H, J ) 5.6 Hz), 2.62-2.73 (m, 1H), 3.09-3.20 (m,
4H), 7.21 (s, 1H), 7.27-7.46 (m, 10 H); IR (neat) 1644
cm-1; HRMS (EI) calcd for C22H29N2P 352.2071, found
352.2093.
(S)-[1-[(Dip h en ylp h osp h in o)m eth yl]-2-m eth ylp r o-
p yl)]-N-(4-(d im eth yla m in o)ben zylid en e)a m in e (8f):
white solid; [R]25 +217.1 (c ) 1.0, CHCl3); mp 57-58
D
1
(S)-[1-[(Dip h en ylp h osp h in o)m eth yl]-2-m eth ylp r o-
°C; H NMR (270 MHz, CDCl3) δ 0.88 (d, 3H, J ) 6.4
p yl]-N-[(1-p ip er id in o)m eth ylen e]a m in e (7d ): viscous
Hz), 0.91 (d, 6H, J ) 6.4 Hz), 1.91-2.03 (m, 1H), 2.47 (d,
2H, J ) 8.2 Hz), 2.86-2.93 (m, 1H), 3.00 (s, 6H), 6.66 (d,
2H, J ) 8.9 Hz), 7.23-7.51 (m, 12 H), 7.88 (s, 1H); 31P-
{1H} NMR (161.7 MHz, CDCl3, H3PO4) δ -19.55; IR
(KBr) 1638 cm-1; HRMS(EI) calcd for C26H31N2P 402.2227,
found 402.2233. Anal. Calcd for C26H31N2P: C, 77.58; H,
7.76; N, 6.96. Found: C, 77.63; H, 7.71; N, 6.91.
1
liquid; yield 30%; [R]25 +54.9 (c 0.4, CHCl3); H NMR
D
(270 MHz, CDCl3) δ 0.82 (d, 3H, J ) 6.6 Hz), 0.86 (d,
3H, J ) 6.6 Hz), 1.44-1.56 (m, 6H), 1.69-1.82 (m, 1H),
2.36 (d, 2H, J ) 6.9 Hz), 2.63-2.74 (m, 1H), 3.00-3.17
(m, 4H), 6.96 (s, 1H), 7.25-7.45 (m, 10 H); IR (neat) 1644
cm-1; HRMS (EI) calcd for C23H31N2P 366.2227, found
366.2246.
(S)-N′-[1-[(P h en ylt h io)m et h yl]-2-m et h ylp r op yl]-
N,N-d im eth ylm eth a n im id a m id e (11a ). A solution of
(S)-3-methyl-1-(phenylthio)-2-butanamine (10a ) (212 mg,
1.09 mmol) and N,N-dimethylformamide dimethylacetal
(2.69 g, 3 mL, 22.6 mmol) was stirred at room tempera-
ture for 5 h. The reaction was monitored by gas chroma-
tography. Removal of the excess acetal gave 11a quan-
titatively. The product was purified by silica gel column
chromatography (5:1 mixture of toluene and AcOEt),
followed by checking the purity by using gas chromatog-
P r oced u r e for Im in e-P h osp h in e Hybr id Liga n d s
(8). (S)-1-(Diphenylphosphino)-3-methyl-2-butanamine
(6a ) (400 mL, 1.48 mmol) was reacted with the corre-
sponding benzaldehydes (1.48 mmol) in 3 mL of toluene
at room temperature under an argon atmosphere over-
1
night. The solvent was stripped off in vacuo. H NMR
spectra of the products showed the quantitative conver-
sion of 6a to the desirable imine-phosphine hybrid
ligands 8. The hybrid ligands except for 8b were distilled
in vacuo using Kugelrohr before the application to
asymmetric reactions.
raphy before the use: [R]27 +76.9 (c 1.0, CHCl3); 1H
D
NMR (270 MHz, CDCl3) δ 0.87 (d, 3H, J ) 6.9 Hz), 0.90
(d, 3H, J ) 6.9 Hz), 1.74-1.88 (m, 1H), 2.74-2.80 (m,
1H), 2.79 (s, 6H), 3.01 (dd, 1H, J BX ) 9.2 Hz, J AB ) 12.5
Hz), 3.21 (dd, 1H, J AX ) 3.6 Hz, J AB ) 12.5 Hz), 7.15 (s,
1H), 7.09-7.33 (m, 4H), 7.95 (s, 1H); 13C NMR (68 MHz,
CDCl3) δ 18.6, 19.9, 33.2, 39.2, 71.2, 125.0, 128.4, 128.5,
137.8, 154.3; IR (neat) 1649 cm-1; HRMS (EI) calcd for
C14H22N2S 250.1506, found 250.1506.
(S)-N-Ben zylid en e-[1-[(d ip h en ylp h osp h in o)m eth -
yl]-2-m eth ylp r op yl]a m in e (8a ): viscous liquid; [R]25
D
+113.4 (c 1.0, CHCl3); 1H NMR (270 MHz, CDCl3) δ 0.90
(d, 3H, J ) 4.3 Hz), 0.93 (d, 3H, J ) 4.3 Hz), 1.93-2.05
(m, 1H), 2.49 (d, 2H, J ) 6.6 Hz), 2.67 (s, 6H), 2.97-3.07
(m, 1H), 7.21-7.62 (m, 15H), 8.03 (s, 1H); IR (neat) 1644
cm-1; HRMS (EI) calcd for C24H26NP 359.1805, found
359.1791.
Gen er a l P r oced u r e for P r ep a r a tion of 11b,c fr om
4b,c. A representative procedure from 4b,c to 11b,c was
described by the preparation of 11b.
(S )-N -(4-C a r b o m e t h o x y b e n z y li d e n e )-[1-[(d i -
p h en ylp h osp h in o)m et h yl]-2-m et h ylp r op yl]a m in e
(8b): viscous liquid; [R]24D +128.4 (c 1.0, CHCl3); 1H NMR
(270 MHz, CDCl3) δ 0.91 (d, 3H, J ) 7.3 Hz), 0.94 (d,
3H, J ) 7.3 Hz), 1.94-2.06 (m, 1H), 2.50 (d, 2H, J ) 8.2
Hz), 3.03-3.13 (m, 1H), 3.93 (s, 3H), 7.19-7.46 (m, 10H),
7.63 (d, 2H, J ) 8.3 Hz), 8.01 (d, 2H, 8.3 Hz), 8.07 (s,
1H); IR (neat) 1725, 1642 cm-1; HRMS (EI) calcd for
C26H28O2NP 417.1859, found 417.1832.
(S)-[1-[(4-F lu or op h en ylth io)m eth yl]-2-m eth ylp r o-
p yl]ca r ba m ic Acid 1,1-Dim eth yleth yl Ester (9b). A
solution of p-fluorothiophenol (700 mg, 5.47 mmol) in
THF (3 mL) was added to a suspension of NaH (150 mg,
6.25 mmol, prewashed with n-pentane) in THF (15 mL)
in an ice-water bath, followed by stirring for 1 h. To the
solution was added 4a (900 mg, 2.53 mmol) in THF (5
mL), and the reaction mixture was stirred at room
temperature overnight. Celite filtration and concentra-
tion of the filtrate followed by silica gel column chroma-
tography eluted by toluene/AcOEt (20:1) afforded 9b (546
(S)-[1-[(Dip h en ylp h osp h in o)m eth yl]-2-m eth ylp r o-
p yl)]-N-(4-tr iflu or om eth ylben zylid en e)a m in e (8c):
viscous liquid; [R]24 +95.7 (c 1.0, CHCl3); 1H NMR (270
D
MHz, CDCl3) δ 0.91 (d, 3H, J ) 6.9 Hz), 0.93 (d, 3H, J )
6.9 Hz), 1.94-2.06 (m, 1H), 2.50 (d, 2H, J ) 8.3 Hz),
3.03-3.14 (m, 1H), 7.20-7.47 (m, 10H), 7.32 (d, 2H, 8.2
Hz), 7.67 (d, 2H, 8.6 Hz), 8.06 (s, 1H); IR (KBr) 1647,
1325 cm-1; HRMS (EI) calcd for C25H25NF3P 427.1679,
found 427.1687.
mg, 69%): white solid; [R]24 +32.6 (c 0.6, CHCl3); mp
D
84-85 °C; 1H NMR (270 MHz, CDCl3) δ 0.89 (d, 3H, J )
6.6 Hz), 0.91 (d, 3H, J ) 6.6 Hz), 1.43 (s, 9H), 1.83-1.96
(m, 1H), 2.99 (d, 2H, J ) 5.9 Hz), 3.56-3.67 (m, 1H), 4.52
(brd, 1H, J ) 10.8 Hz), 6.95-7.04 (m, 2H), 7.36-7.42 (m,
2H); 13C NMR (68 MHz, CDCl3) δ 17.7, 19.4, 28.4, 30.8,
38.9, 55.1, 79.2, 115.9, 116.2, 131.2, 132.8, 132.9, 155.6,
163.7; IR (KBr) 3304, 1674, 1539, 1175 cm-1; FAB-MS
m/z 314 (MH+).
(S)-[1-[(Dip h en ylp h osp h in o)m eth yl]-2-m eth ylp r o-
p yl)]-N-(4-m eth ylben zylid en e)a m in e (8d): viscous liq-
1
uid; [R]23 +105.7 (c 1.0, CHCl3); H NMR (270 MHz,
D
CDCl3) δ 0.89 (d, 3H, J ) 5.6 Hz), 0.92 (d, 3H, J ) 5.6
Hz), 1.92-2.04 (m, 1H), 2.36 (s, 3H), 2.48 (d, 2H, J ) 7.3
Hz), 2.93-3.03 (m, 1H), 7.14-7.51 (m, 14H), 7.98 (s, 1H);
IR (neat) 1645 cm-1; HRMS (EI) calcd for C25H28NP
373.1961, found 373.1972.
(S )-1-(4-F lu o r o p h e n y lt h io )-3-m e t h y l-2-b u t a n -
a m in e (10b). To a solution of 9b (515 mg, 1.65 mmol) in
CH2Cl2 (40 mL) was added trifluoroacetic acid (9.12 g,
6.2 mL, 51.9 mmol) in an ice-water bath. After the
mixture was stirred at room temperature overnight,
water was added. The water phase neutralized with a
NaOH aqueous solution and then extracted with CH2-
(S)-[1-[(Dip h en ylp h osp h in o)m eth yl]-2-m eth ylp r o-
p yl)]-N-(4-m eth oxyben zylid en e)a m in e (8e): viscous
1
liquid; [R]26 +103.9 (c 0.6, CHCl3); H NMR (270 MHz,
D

Versatile Chiral Bidentate Ligands
J . Org. Chem., Vol. 65, No. 14, 2000 4239
Cl2. The combined organic layers were successively
washed with saturated aqueous NaHCO3 and brine,
followed by drying over anhydrous MgSO4. The solvent
was evaporated to afford the desired amine 10b as a
[1R ,2R -[1r(E ),2r(E )]]-N ,N ′-Bis(d im e t h yla m in o-
m eth ylen e)-1,2-cycloh exa n ed ia m in e (12a ). A mixture
of (1R,2R)-1,2-diaminocyclohexane (500 mg, 4.38 mmol)
and N,N-dimethylformamide dimethylacetal (4.49 g, 5
mL, 37.7 mmol) was stirred at room temperature for 3
h. The reaction was followed by gas chromatography. The
volatile fractions were stripped off in vacuo, affording 12a
quantitatively. The product was purified as a transparent
transparent liquid (313 mg, 89%): [R]23 +85.3 (c 0.6,
D
1
CHCl3); H NMR (270 MHz, CDCl3) δ 0.91 (d, 3H, J )
6.9 Hz), 0.93 (d, 3H, J ) 6.9 Hz), 1.65-1.77 (m, 1H), 2.68
(d, 2H, J ) 9.2 Hz), 3.06-3.16 (m, 1H), 6.98 (d, 1H, J )
8.6 Hz), 7.01 (d, 1H, J ) 8.6 Hz), 7.35 (d, 1H, J ) 5.3
Hz), 7.38 (d, 1H, J ) 5.3 Hz); 13C NMR (68 MHz, CDCl3)
δ 17.5, 19.2, 32.9, 41.5, 55.3, 115.8, 116.1, 131.1, 132.3,
132.4, 159.9; IR (neat) 3376, 1589, 1227 cm-1; FAB-MS
m/z 214 (MH+).
liquid using Kugelrohr (150-160 °C/4-5 mmHg): [R]25
D
-192.4 (c 1.0, CHCl3); 1H NMR (270 MHz, CDCl3) δ 1.33-
1.78 (m, 10H), 2.76 (s, 12H), 7.20 (s, 2H); 13C NMR (68
MHz, CDCl3) δ 25.5, 34.9, 37.4, 70.1, 155.4; IR (neat) 1651
cm-1; FAB-MS m/z 225 (MH+); HRMS (EI) calcd for
C12H24N4 224.2003, found 224.1984.
(S)-N′-[1-[(4-F lu or op h en ylth io)m eth yl]-2-m eth yl-
pr opyl]-N,N-dim eth ylm eth an im idam ide (11b). A mix-
ture consisting of 10b (300 mg, 1.51 mmol) and N,N-
dimethylacetamide dimethylacetal (2.69 g, 3 mL, 22.6
mmol) was stirred at room temperature for 5 h. The
reaction was monitored by gas chromatography. After
removal of the volatile fractions in vacuo, the desired
product 11b was obtained as a viscous liquid quantita-
tively. The product was exposed to silica gel column
chromatography (toluene with 0.5 vol % of Et3N) before
[1R ,2R -[1r(E ),2r(E )]]-N ,N ′-Bis(d im e t h yla m in o-
m eth ylen e)-1,2-d ip h en yleth ylen ed ia m in e (12b). A
solution of (1R,2R)-1,2-diphenylethylenediamine (100 mg,
0.471 mmol) and N,N-dimethylformamide dimethylacetal
(1.79 g, 2 mL, 15.1 mmol) was stirred in a flask immersed
in an oil bath at 50 °C for 2 h. The reaction was monitored
by gas chromatography. After removal of the volatile
fractions, the diamidine 12b was produced as white solid
quantitatively. The product was recrystallized as a white
solid from a mixture of n-hexane and AcOEt. Although
the 13C NMR spectrum of the product suggests the
presence of a minor rotational isomer, the solids were
used for the catalytic reaction as it stands: [R]26D -133.8
(c 1.05, CHCl3); mp 99-100 °C; 1H NMR (270 MHz,
CDCl3) δ 2.82 (s, 12H), 4.15 (s, 2H), 7.03-7.14 (m, 10H),
7.27 (s, 2H); 13C NMR (68 MHz, CDCl3) δ 37.0, 77.5,
125.9, 126.5, 127.4, 128.1, 128.6, 143.1, 144.0, 153.5,
155.3; IR (KBr) 1645 cm-1; FAB-MS m/z 323 (MH+);
HRMS (EI) calcd for C20H26N4 322.2160, found 322.2135.
Gen er a l P r oced u r e for Rea ction s 1-3. The reac-
tion types 1-3 are typified by the reaction using 14a as
a substrate. A solution of [Pd(η3-C3H5)Cl]2 (4 mg, 0.0109
mmol) and VALAP (14.2 mg, 0.0436 mmol) in CH2Cl2 (1
mL, dry and oxygen free) was stirred at room tempera-
ture under argon, followed by addition of 14a (128 mg,
0.436 mmol) in CH2Cl2 (1 mL). To the mixture was added
a nucleophilic solution prepared in another flask by
mixing dimethyl malonate (173 mg, 1.31 mmol) and BSA
(266 mg, 1.31 mmol) in the presence of lithium acetate
(1.4 mg, 0.0218 mmol) in the solvent (1 mL). The
reactions were monitored by TLC. The solution was
stirred under the required conditions, and the volatile
fractions were removed in vacuo. The residue was puri-
fied by preparative TLC (toluene/EtOAc ) 20/1).
Rep r esen ta tive P r oced u r e for Rea ction 4. The
reaction catalyzed by the Pd-VALAP complex using 19a
as ketene silyl acetals is described as a representative
procedure for the reaction type 4. The palladium chloride
dimer, [Pd(η3-C3H5)Cl]2 (4 mg, 0.0109 mmol) and VALAP
(14.2 mg, 0.0436 mmol) were combined in CH2Cl2 (1 mL,
dry and degassed) under argon and stirred at room
temperature for 15 min. To the solution was added the
allylic pivalate 14a (128 mg, 0.0436 mmol) in CH2Cl2 (1
mL), followed by stirring for 30 min. The resulting
catalyst solution was treated with 19a (228 mg, 1.31
mmol) in CH2Cl2 (2 mL). After the mixture was stirred
under the required conditions, the volatile fractions were
stripped off immediately. The residue was purified by
preparative TLC using toluene as the eluent.
the asymmetric reactions: [R]25 +62.7 (c 1.0, CHCl3);
D
1H NMR (270 MHz, CDCl3) δ 0.85 (d, 3H, J ) 6.6 Hz),
0.88 (d, 3H, J ) 6.6 Hz), 1.73-1.86 (m, 1H), 2.68-2.75
(m, 1H), 2.79 (s, 6H), 2.99 (dd, 1H, J BX ) 8.9 Hz, J AB
)
12.5 Hz), 3.15 (dd, 1H, J AX ) 3.6 Hz, J AB ) 12.5 Hz), 6.94
(d, 2H, J ) 8.6 Hz), 6.96 (d, 2H, J ) 8.6 Hz), 7.15 (s,
1H), 7.29 (d, 2H, J ) 5.3 Hz), 7.33 (d, 2H, J ) 5.3 Hz);
13C NMR (68 MHz, CDCl3) δ 18.7, 19.9, 33.3, 37.1, 40.6,
71.5, 115.4, 115.8, 131.2, 131.4, 132.7, 154.4, 159.4; IR
(neat) 1649, 1225 cm-1; HRMS (EI) calcd for C14H21N2-
FS 268.1411, found 268.1425.
(S)-[1-[(4-Met h oxyp h en ylt h io)m et h yl]-2-m et h yl-
p r op yl]ca r ba m ic a cid 1,1-d em eth yleth yl ester
(9c): white solid; yield 70%; mp 74-75 °C; [R]25 +34.3
D
(c 1.0, CHCl3); 1H NMR (270 MHz, CDCl3) δ 0.87 (d, 3H,
J ) 6.6 Hz), 0.89 (d, 3H, J ) 6.6 Hz), 1.43 (s, 9H), 1.83-
1.96 (m, 1H), 2.95 (d, 2H, J ) 5.3 Hz), 3.52-3.63 (m, 1H),
3.79 (s, 3H), 4.54 (brd, 1H, J ) 5.4 Hz), 6.84 (d, 2H, J )
8.9 Hz), 7.39 (d, 2H, J ) 8.9 Hz); 13C NMR (68 MHz,
CDCl3) δ 17.7, 19.3, 28.3, 30.8, 39.5, 55.2, 78.9, 114.5,
126.3, 133.5, 155.6, 158.9; IR (KBr) 3382, 1684, 1524,
1246, 1028 cm-1; FAB-MS m/z 326 (MH+).
(S)-1-(4-Met h oxyp h en ylt h io)-3-m et h yl-2-b u t a n -
a m in e (10c): viscous liquid; yield 92%; [R]25 +85.8 (c
D
1.0, CHCl3); 1H NMR (270 MHz, CDCl3) δ 0.89 (d, 3H, J
) 6.6 Hz), 0.91 (d, 3H, J ) 6.6 Hz), 1.63-1.75 (m, 1H),
2.62 (d, 2H, J ) 7.9 Hz), 3.01-3.10 (m, 1H), 3.80 (s, 3H),
6.84 (d, 1H, J ) 8.9 Hz), 7.36 (d, 1H, J ) 8.9 Hz); 13C
NMR (68 MHz, CDCl3) δ 17.5, 19.2, 32.8, 42.2, 55.3,
114.5, 126.2, 133.2, 158.9; IR (neat) 3364, 1593, 1244,
1032 cm-1; FAB-MS m/z 226 (MH+).
(S)-N′-[1-[(4-Meth oxyp h en ylth io)m eth yl]-2-m eth -
ylp r op yl]-N,N-d im eth ylm eth a n im id a m id e (11c): vis-
1
cous liquid; yield quant; [R]24 +87.8 (c 1.0, CHCl3); H
D
NMR (270 MHz, CDCl3) δ 0.84 (d, 3H, J ) 6.6 Hz), 0.86
(d, 3H, J ) 6.6 Hz), 1.71-1.84 (m, 1H), 2.66-2.73 (m,
1H), 2.81 (s, 6H), 2.94 (dd, 1H, J BX ) 8.6 Hz, J AB ) 12.5
Hz), 3.12 (dd, 1H, J AX ) 3.6 Hz, J AB ) 12.5 Hz), 3.79 (s,
3H), 6.82 (d, 2H, J ) 8.9 Hz), 7.15 (s, 1H), 7.31 (d, 2H, J
) 8.9 Hz), 7.95 (s, 1H); 13C NMR (68 MHz, CDCl3) δ 18.7,
19.9, 33.1, 37.0, 41.3, 55.1, 71.3, 114.2, 127.9, 132.1,
154.3, 158.2; IR (neat) 1649, 1244, 1034 cm-1; HRMS (EI)
calcd for C15H24ON2S 280.1611, found 280.1585.
(S)-r,r-Dim eth yl-â-(2-p h en yleth en yl)ben zen ep r o-
p a n oic a cid m eth yl ester (20a ): [R]26 -43.1 (c 0.7,
D
EtOH) for 20a with 90% ee; 1H NMR (270 MHz, CDCl3)

4240 J . Org. Chem., Vol. 65, No. 14, 2000
Saitoh et al.
δ 1.18 (s, 3H), 1.23 (s, 3H), 3.60 (s, 3H), 3.75 (d, 1H, J )
9.6 Hz), 6.45 (d, 1H, J ) 15.8 Hz), 6.60 (dd, 1H, J ) 9.6,
16.2 Hz), 7.21-7.41 (m, 10H); 13C NMR (68 MHz, CDCl3)
δ 22.5, 23.2, 47.3, 51.6, 56.9126.3, 126.7, 127.3, 127.7,
128.0, 128.4, 128.6, 128.9, 129.1, 132.5, 137.3, 140.418,
177.3; IR (neat) 1730 cm-1; MS m/z 294 (M+), 263, 233,
205, 193, 116; HRMS (EI) calcd for C20H22O2 294.1621,
found 294.1594.
quenched with water and extracted with Et2O. After the
organic phase was dried over MgSO4, the solution was
concentrated. The residue was purified by a preparative
TLC using toluene as the eluent.
Rep r esen ta tive P r oced u r e for Rea ction 5 Usin g
KSA/Tetr a a lk yla m m on iu m Ha lid es. To a catalyst
solution consisting of [Pd(η3-C3H5)Cl]2 (4 mg, 0.0109
mmol) and VALAP (14.2 mg, 0.0436 mmol) in THF (1
mL, dry and degassed) was added the allylic pivalate 14a
(128 mg, 0.0436 mmol) in THF (1 mL), followed by
stirring at room temperature for 30 min. Under the
required reaction temperature, tetrabutylammonium
fluoride (341 mg, 1.31 mmol) in THF (1 mL) and 19a (228
mg, 1.31 mmol) in THF (1 mL) were added dropwise to
the catalyst solution continuously. The reaction mixture
was worked up in a similar manner as the reaction using
KSA/MeLi.
(S)-1-(1,3-Dip h en yl-2-p r op en yl)cycloh exa n eca r -
boxylic a cid m eth yl ester (20b): [R]25 -27.4 (c 0.3,
D
EtOH) for 20b with 90% ee; 1H NMR (270 MHz, CDCl3)
δ 1.22-1.33, 1.52-1.63 (m, 8H), 2.10-2.24 (m, 2H), 3.50
(d, 1H, J ) 10.8 Hz), 3.57 (s, 3H), 6.41 (d, 1H, J ) 16.2
Hz), 6.65 (dd, 1H, J ) 9.9, 15.5 Hz), 7.14-7.37 (m, 10H);
13C NMR (68 MHz, CDCl3) δ 23.5, 23.6, 25.6, 32.0, 32.4,
51.2, 52.6, 58.8, 126.3, 126.7, 127.2, 127.7, 128.0, 128.4,
128.7, 128.9, 129.0, 132.2, 137.4, 140.4, 175.7; IR (neat)
1726 cm-1; MS m/z 334 (M+), 303, 275, 193, 116; HRMS
(EI) calcd for C23H26O2 334.1934, found 334.1930.
Ack n ow led gm en t. This work was partly supported
by a Grant-in-Aid for Scientific Research from the
Ministry of Education, Science and Culture of J apan
(No. 11672111). We wish to thank Mrs. Tomoko Terada,
Institute for Chemical Research, Kyoto University, for
the measurement of HRMS. Dr. Mitsuo Uchida at
Analytical Center, University of Shizuoka, is gratefully
acknowledged for the measurement of FAB-MS.
Rep r esen ta tive P r oced u r e for Rea ction 5 Usin g
KSA/MeLi. The palladium chloride dimer [Pd(η3-C3H5)-
Cl]2 (4 mg, 0.0109 mmol) was mixed with VALAP (14.2
mg, 0.0436 mmol) in THF (1 mL, dry and degassed) at
room temperature for 15 min. The allyl pivalate 14a (128
mg, 0.0436 mmol) dissolved in THF (1 mL) was added
dropwise to the catalyst solution, followed by stirring for
30 min. In another two-necked flask, 1.14 M methyl-
lithium Et2O solution (1.15 mL, 1.31 mmol) was added
to 19a (228 mg, 1.31 mmol) in THF (2 mL) on an acetone
bath at -78 °C. After being stirred for 1 h, the nucleo-
phile solution was treated with the catalyst solution
under each reaction condition. The reaction mixture was
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra of
compounds 3a -7a , 3b-7b, 7c,d , 8a -8f, 11a , 9b-11b, 9c-
11c, 12a ,b, and 20a ,b. This material is available free of charge
via the Internet at http://pubs.acs.org.
J O991615L