Development of a new class of chiral phosphorus ligands: P-chirogenic diaminophosphine oxides. A unique source of enantioselection in Pd-catalyzed asymmetric construction of quaternary carbons

Article abstract of DOI:10.1021/jo050800y

We have recently developed a new class of chiral phosphorus ligands: P-chirogenic diaminophosphine oxides. These pentavalent phosphorus compounds have been successfully applied to Pd-catalyzed asymmetric construction of tertiary and quaternary carbons. The actual ligand structure was the trivalent phosphorus species 17, which was generated in situ by BSA-induced P(V) to P(III) transformation of 6, the preligand. Detailed mechanistic studies, including asymmetric amplification and initial rate kinetics, revealed that complex 18 [Pd-17 (1:2) complex] was the active catalyst. The important function of the nitrogen atom on the sidearm in the ligands was also clarified. The source of enantioselection in the construction of asymmetric quaternary carbons was the secondary ligand substrate interaction mediated by N-Zn coordination.

Full text of DOI:10.1021/jo050800y

Development of a New Class of Chiral Phosphorus Ligands:
P-Chirogenic Diaminophosphine Oxides. A Unique Source of
Enantioselection in Pd-Catalyzed Asymmetric Construction of
Quaternary Carbons
Tetsuhiro Nemoto, Takamasa Masuda, Takayoshi Matsumoto, and Yasumasa Hamada*
Graduate School of Pharmaceutical Sciences, Chiba University, Yayoi-cho, Inage-ku,
Chiba 263-8522, Japan
hamada@p.chiba-u.ac.jp
Received April 19, 2005
We have recently developed a new class of chiral phosphorus ligands: P-chirogenic diaminophos-
phine oxides. These pentavalent phosphorus compounds have been successfully applied to
Pd-catalyzed asymmetric construction of tertiary and quaternary carbons. The actual ligand
structure was the trivalent phosphorus species 17, which was generated in situ by BSA-induced
P(V) to P(III) transformation of 6, the preligand. Detailed mechanistic studies, including asymmetric
amplification and initial rate kinetics, revealed that complex 18 [Pd-17 (1:2) complex] was the
active catalyst. The important function of the nitrogen atom on the sidearm in the ligands was
also clarified. The source of enantioselection in the construction of asymmetric quaternary carbons
was the secondary ligand substrate interaction mediated by N-Zn coordination.
SCHEME 1. Tautomerization of Secondary
Phosphine Oxides and Phosphonic Acid
Derivatives
Introduction
Considerable effort has been devoted to the design and
synthesis of various types of chiral ligands for transition
metal-mediated asymmetric reactions.1 From a practical
point of view, stable, inexpensive, and easily accessible
ligands are highly desirable. The stability of phosphorus
atoms toward air oxidation is the simplest but most
crucial index for evaluating the practicality of phosphorus
ligands. Secondary phosphine oxides and phosphonic acid
derivatives are air- and moisture-stable pentavalent
phosphorus compounds. These compounds exist in equi-
librium between pentavalent forms (RR′P(dO)H) and
trivalent tautomeric forms (RR′POH). During equilibri-
um, stereochemical arrangement around the phosphorus
center is retained (Scheme 1).2 These properties allow for
the use of secondary phosphine oxides as air- and
moisture-stable ligands in several reactions such as Pt-
catalyzed hydroformylation,3a Pt-catalyzed hydrolysis3b
and amidation3c of nitriles, Pd-catalyzed aromatic sub-
stitution reactions,4a and Pd-catalyzed cross-coupling
reactions.4b-h In those reactions, tautomeric phosphinous
acid coordinates to the metal center through a phospho-
rus atom.
Chiral ligands with a stereogenic center on the phos-
phorus atom are expected to create an effective asym-
(3) (a) van Leeuen, P. W. N. M.; Roobeek, C. F. In Homogeneous
Transition Metal Catalyzed Reactions; Advances in Chemistry Series
230; American Chemical Society: Washington, DC, 1992; p 367. (b)
Ghaffar, T.; Parkins, A. W. Tetrahedron Lett. 1995, 36, 8657. (c) Cobley,
C. J.; van den Heuvel, M.; Abbadi, A.; de Vries, J. G. Tetrahedron Lett.
2000, 41, 2467.
(1) For review: Ojima, I. Catalytic Asymmetric Synthesis II; Wiley-
VCH: New York, 2000.
(2) Barton, D.; Ollis, W. D. In Comprehensive Organic Chemistry;
Sutherland, I. O. Ed.; Pergammon Press: Oxford, 1979; Vol. 2, Part
10.
10.1021/jo050800y CCC: $30.25 © 2005 American Chemical Society
Published on Web 07/29/2005
7172
J. Org. Chem. 2005, 70, 7172-7178

P-Chirogenic Diaminophosphine Oxides
metric environment around the central metal.5 Although
simple P-chirogenic secondary phosphine oxides have
been prepared by classical resolution6a or phosphanyl-
thioic acid-mediated three-step resolution,6b their ap-
plication to asymmetric catalysis has long been neglected.
A recent increase in attention to the potential of second-
ary phosphine oxides in transition metal catalysis has
inspired the creation of a new research field: asymmetric
catalysis with P-chirogenic phosphine oxides.7 The first
application of P-chirogenic secondary phosphine oxide to
Ir-catalyzed asymmetric hydrogenation was recently
reported.8 In their ligand preparation, however, optical
resolution by preparative chiral HPLC is necessary to
obtain the chiral secondary phosphine oxides. Thus, the
development of an efficient strategy to synthesize various
P-chirogenic secondary phosphine oxides or their equiva-
lent in an optically pure form is in high demand. We
hypothesized that this preparative drawback could be
overcome by designing new ligands based on a chiral
diamine framework. We recently reported a new class of
chiral phosphorus ligands: aspartic acid-derived P-
chirogenic diaminophosphine oxides (DIAPHOXs), which
were applied to the catalytic asymmetric synthesis of
quaternary carbon centers through Pd-catalyzed asym-
metric allylic substitution.9 While the construction of
asymmetric quaternary carbons was successful, the
source of enantioselection and the details of the reaction
mechanism were obscure. In this article, we report the
full details of a Pd-catalyzed asymmetric construction of
quaternary carbons using P-chirogenic diaminophosphine
oxides. Detailed mechanistic studies of two types of
mechanistically different asymmetric allylic substitution
reactions clarified the multifunctional properties of the
developed ligands in the asymmetric catalysis.
SCHEME 2. Strategy for the Synthesis of
P-Chirogenic Diaminophosphine Oxides
SCHEME 3. Synthesis of (S,RP)-Ph-DIAPHOX 6a
a Reagents and conditions: (a) aniline, DMF, rt, 1 h, then
aniline, WSCI, rt, 24 h. (b) Pd-C (2 mol %), H2, 2-propanol-DMF,
rt, 6 h. (c) Benzoyl chloride, NEt3, THF, rt, 1 h. (d) LiAlH4, THF,
reflux, 13 h. (e) PCl3, NEt3, toluene, -78 °C to room temperature,
16 h. (f) SiO2, H2O, AcOEt, rt, 18 h.
(4) (a) Li, G. Y. Angew. Chem., Int. Ed. 2001, 40, 1513. (b) Li, G. Y.
J. Org. Chem. 2002, 67, 3643. (c) Yang, W.; Wang, Y.; Corte, J. R. Org.
Lett. 2003, 5, 3131. (d) Wolf, C.; Lerebours, R. J. Org. Chem. 2003, 68,
7077. (e) Wolf, C.; Lerebours, R. J. Org. Chem. 2003, 68, 7551. (f) Wolf,
C.; Lerebours, R. Tanzini, E. H. Synthesis 2003, 2069. (g) Wolf, C.;
Lerebours, R. Org Lett. 2004, 6, 1147. (h) Ackermann, L.; Born, R.
Angew. Chem., Int. Ed. 2005, 44, 2444.
Results and Discussion
Design, Synthesis, and Application of Aspartic
Acid-Derived P-Chirogenic Diaminophosphine
Oxides (DIAPHOXs). Cyclic diaminophosphine oxides
with a stereogenic center on the phosphorus atom can
be prepared from asymmetric diamines. Separation of the
diastereomeric mixture, however, is necessary to obtain
optically pure P-chirogenic diaminophosphine oxide. Our
strategy to introduce chirality into the phosphorus atom
is outlined in Scheme 2. Triaminophosphines are reactive
to water under acidic conditions via an SN2-type process,
affording the corresponding diaminophosphine oxides
through P(III) to P(V) tautomerization.2 Therefore, we
expected that diastereoselective formation of P-chirogenic
triaminophosphine starting from optically active branched
triamines, followed by the introduction of oxygen func-
tionality on the phosphorus atom, would be an efficient
synthetic route.
Our ligand was readily synthesized from the known
acid anhydride 1 (Scheme 3). Nucleophilic opening of 1,
followed by condensation with aniline, yielded the cor-
responding dianilide 2. Removal of a Z group followed
by amide formation with benzoyl chloride afforded tri-
amide 3. After reduction of all of the amide groups, the
obtained triamine 4 was reacted with phosphorus trichlo-
ride to afford the corresponding triaminophosphine 5,
(5) For recent representative examples of P-chirogenic phosphorus
ligands: (a) Imamoto, T.; Watanabe, J.; Wada, Y.; Masuda, H.;
Yamada, H.; Tsuruta, H.; Matsukawa, S.; Yamaguchi, K. J. Am. Chem.
Soc. 1998, 120, 1635. (b) Yamanoi, Y.; Imamoto, T. J. Org. Chem. 1999,
64, 2988. (c) Tsuruta, H.; Imamoto, T. Tetrahedron: Asymmetry 1999,
10, 877. (d) Miura, T.; Imamoto, T. Tetrahedron Lett. 1999, 40, 4833.
(e) Ohashi, A.; Imamoto, T. Org. Lett. 2001, 3, 373. (f) Tsuruta, H.;
Imamoto, T. Synlett 2001, 999. (g) Tang, W.; Zhang, X. Angew Chem.,
Int. Ed. 2002, 41, 1612. (h) Oohara, N.; Katagiri, K.; Imamoto, T.
Tetrahedron: Asymmetry 2003, 14, 2127. (i) Tang, W.; Wang, W.; Chi,
Y.; Zhang, X. Angew. Chem., Int. Ed. 2003, 42, 3509. (j) Hoge, G. J.
Am. Chem. Soc. 2003, 125, 10219. (k) Hoge, G.; Wu, H.-P.; Kissel, W.
S.; Pflum, D. A.; Greene, D. J.; Bao, J. J. Am. Chem. Soc. 2004, 126,
5966. (l) Hoge, G. J. Am. Chem. Soc. 2004, 126, 9920. (m) Reetz, M.
T.; Ma, J.-A.; Goddard, R. Angew Chem. Int. Ed. 2005, 44, 412.
(6) (a) Drabowicz, J.; Lyzwa, P.; Omelanczuk, J.; Pietrusiewicz, K.
M.; Mikolajajczyk, M. Tetrahedron: Asymmetry 1999, 10, 2757. (b)
Haynes, R. K.; Au-Yeung, Y.-L.; Chen, W.-K.; Lam, W.-L.; Li, Z.-Y.;
Yeung, L.-L.; Chen, A. S. C.; Li, P.; Koen, M.; Mitchell, C. R.; Vonwiller,
S. C. Eur. J. Org. Chem. 2000, 3205.
(7) Recent progress in transition metal catalysis using chiral and
achiral phosphine oxides is highlighted in the following reference;
see: Dubrovina, N. V.; Bo¨rner, A. Angew. Chem., Int. Ed. 2004, 43,
5883.
(8) Jiang, X.; Minnaard, A. J.; Hessen, B.; Feringa, B. L.; Duchateau,
A. L. L.; Andrien, J. G. O.; Booger, J. A. F.; de Vries, J. G. Org. Lett.
2003, 5, 1503.
(9) Nemoto, T.; Matsumoto, T.; Masuda, T.; Hitomi, T.; Hatano, K.;
Hamada, Y. J. Am. Chem. Soc. 2004, 126, 3690.
J. Org. Chem, Vol. 70, No. 18, 2005 7173

Nemoto et al.
TABLE 1. Scope and Limitations
which was converted into diaminophosphine oxide 6
[(S,RP)-Ph-DIAPHOX] by treatment with silica gel in wet
ethyl acetate.10 Direct purification of the crude triami-
nophosphine residue with silica gel column chromatog-
raphy could be utilized as a more convenient alternative
method, affording 6 in comparable yield.
The obtained diaminophosphine oxide 6 was applied
to catalytic enantioselective construction of a quaternary
carbon center.11 Pd-catalyzed asymmetric allylic substi-
tution using prochiral nucleophiles is one of the most
straightforward approaches toward this end. Few suc-
cessful reactions of this type have been reported since
the 1980s.12-14 When asymmetric allylic substitution of
cinnamyl acetate 7a with ethyl 2-oxocyclohexane car-
boxylate 8a was performed using 2.5 mol % (η3-C3H5-
PdCl)2 and 10 mol % 6, no reaction occurred in the
presence of typical bases (NaH, LDA, DBU, 1,1,3,3-
tetramethylguanidine, and i-Pr2NEt). Interestingly, the
reaction proceeded in the presence of N,O-bis(trimeth-
ylsilyl)acetamide (BSA)15 to afford 9a in 53% ee, even
though the yield was only 10%. Encouraged by this result,
we investigated the effect of the addition of acetate salt.
Detailed screening revealed that Zn(OAc)2 was the best
additive for asymmetric induction.16
The scope and limitation of different substrates were
further examined using these efficient conditions (Table
1). When 0.5-5 mol % of the catalyst was used, asym-
metric substitution of allyl acetates 7a-h using prochiral
nucleophiles with a six-membered ring (runs 1-5, 10-
18) proceeded at room temperature to provide the cor-
responding products with modest to high enantioselec-
tivity (63-94% ee). This reaction system was also effective
for nucleophiles with five-, seven-, and eight-membered
rings (runs 6-9) and for acyclic nucleophiles (rus 19-
21), resulting in the formation of quaternary carbon
centers with modest to good enantioselectivity (72-85%
ee).17
Mechanistic Studies of the Developed Catalytic
Asymmetric Reaction: Application to Pd-Catalyzed
Asymmetric Allylic Alkylation via an Attack at
Enantiotopic Termini of the meso-π-Allyl Complex.
Enantioselective nucleophilic attack of a stabilized
prochiral anion on π-allylpalladium is not easy to control
by the chiral ligand on the palladium atom, because the
incoming nucleophile resides at the side opposite to the
stereogenic center in the transition state. Very interest-
ingly, this difficult process proceeded with high reactivity
a Isolated yield. b Determined by HPLC analysis. c Performed
with 2 mol % of the Pd catalyst. d Performed with 1 mol % of the
Pd catalyst. e Performed with 0.5 mol % of the Pd catalyst.
f Performed with 5 mol % of the Pd catalyst. g Xylenes were used
as a solvent. h Performed with 12.5 mol % of 6. i Reactions were
carried out at 0 °C in the absence of Zn(OAc)2. j Product with a
trans-olefin was obtained exclusively.
and stereoselectivity under unique reaction conditions:
(1) BSA is the only base that promoted the reaction; (2)
Zn(OAc)2 is the most effective additive for realizing the
highly asymmetric induction.
We reasoned that application to a mechanistically
different but well-established reaction would be helpful
to clarify the reaction mechanism of the developed
process. Thus, Pd-catalyzed asymmetric allylic alkylation
of 10 with dimethyl malonate was selected for the
monitoring reaction, because this asymmetric allylic
alkylation is recognized as a benchmark system (Table
2). When the reaction was performed using 1 mol % (η3-
C3H5PdCl)2 and 4 mol % 6, the reaction proceeded
smoothly in the presence of Zn(OAc)2, affording the
corresponding product 11 in 88% yield and in 96% ee with
the (S)-configuration. Higher selectivity was obtained
(10) 6 is obtained as an air- and moisture-stable solid. The absolute
configuration was unequivocally determined by X-ray crystal structure
analysis. For detailed information of the X-ray analysis, see the
Supporting Information of ref 9.
(11) For recent reviews, see: (a) Corey, E. J.; Guzman-Perez, A.
Angew. Chem., Int. Ed. 1998, 37, 388. (b) Christoffers, J.; Mann, A.
Angew. Chem., Int. Ed. 2001, 40, 4591. (c) Douglas, C. J.; Overmann,
L. E. Proc. Natl. Acad. Sci. U.S.A. 2004, 101 5363.
(12) For examples using â-keto esters as the prochiral nucleophile,
see: (a) Trost, B. M.; Radinov, R.; Grenzer, E. M. J. Am. Chem. Soc.
1997, 119, 7879. (b) Kuwano, R.; Ito, Y. J. Am. Chem. Soc. 1999, 121,
3236. (c) Brunel, J. M.; Tenaglia, A.; Buono, G. Tetrahedron: Asym-
metry 2000, 11, 3585. See also: (d) Sawamura M.; Nakayama, Y.; Tang,
W.-M.; Ito, Y. J. Org. Chem. 1996, 61, 9090. (e) Sawamura, M.; Sudoh,
M.; Ito, Y. J. Am. Chem. Soc. 1996, 118, 3309.
(13) For examples using 1,3-diketones as the prochiral nucleophile,
see: (a) Hayashi, T.; Kanehira, K.; Tsuchiya, H, Kumada, M. J. Chem.
Soc., Chem. Commun. 1982, 1162. (b) Hayashi, T.; Kanehira, K.;
Hagihara, T.; Kumada, M. J. Org. Chem. 1988, 53, 113. (c) Sawamura,
M.; Nagata, H.; Sakamoto, H.; Ito, Y. J. Am. Chem. Soc. 1992, 114,
2586. (d) Kuwano, R.; Uchida, K.; Ito, Y. Org. Lett. 2003, 5, 2177.
(14) For examples using R-monosubstituted ketones as the prochiral
nucleophile, see: (a) Trost, B. M.; Schroeder, G. M. J. Am. Chem Soc.
1999, 121, 6759. (b) You, S.-L.; Hou, X.-L.; Dai, L.-X.; Zhu, X.-Z. Org.
Lett. 2001, 3, 149. (c) Trost, B. M.; Schroeder, G. M.; Kristensen, J.
Angew. Chem., Int. Ed. 2002, 41, 3492. (d) Behenna, D. C.; Stoltz, B.
M. J. Am. Chem. Soc. 2004, 126, 15044. (e) Trost, B. M.; Frederiksen,
M. U. Angew. Chem., Int. Ed. 2005, 44, 308. (f) Trost, B. M.; Xu, J. J.
Am. Chem. Soc. 2005, 127, 2846.
(15) For review, see: El Gihani, M. T.; Heaney, H. Synthesis 1998,
357.
(16) See Supporting Information for details.
(17) In the case of acyclic nucleophiles, enantioselectivity was not
affected by the addition of acetate salt. This result indicates that the
source of enantioselection is different from that of cyclic nucleophiles.
For detailed data, see Supporting Information.
7174 J. Org. Chem., Vol. 70, No. 18, 2005

P-Chirogenic Diaminophosphine Oxides
TABLE 2. Pd-Catalyzed Asymmetric Allylic Alkylation
of 10 with Dimethyl Malonate
run
additive
Zn(OAc)2
solvent
yieldb
eec
FIGURE 1. Observed species in FAB MS analysis.
1
2
toluene
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
88%
98%
80%
92%
95%
93%
86%
77%
77%
94%
96% ee
98% ee
95% ee
98% ee
97% ee
97% ee
97% ee
99% ee
95% ee
99% ee
SCHEME 4. Reaction Mechanism of BSA-Induced
P(V) to P(III) Transformation of 12
Zn(OAc)2
LiOAc
3
4
NaOAc
5
KOAc
6
CsOAc
7
Mg(OAc)2‚4H2O
In(OAc)3
n-Bu4NOAc
8
9
10
a Absolute configuration was determined by comparing the
measured optical rotation with the reported optical rotation.
b Isolated yield. c Determined by HPLC analysis.
when CH2Cl2 was used as the solvent. Further investiga-
tions revealed that, in contrast to the case of the
quaternary carbon center synthesis, the addition of
several acetate salts did not affect the enantioselectivity,
and the best result was obtained in the absence of acetate
salt (run 10).
however, on the structure of the nitrogen moiety on the
sidearm. Thus, we attempted to clarify the actual ligand
structure using fast atom bombardment mass spectrom-
etry (FAB MS).16 FAB MS analysis of the residue
obtained from a CH2Cl2 solution of 6 and BSA (30 equiv
with respect to 6) exhibited a peak at 464 m/z. The
detected peak corresponded to the molecular weight of
15 [Figure 1a]. In addition, FAB MS analysis of the
residue obtained from a CH2Cl2 solution of (η3-C3H5-
PdCl)2, 6, and BSA in a ratio of 1:2:60 (Pd:6 ) 1:1)
exhibited a peak at 610 m/z. The detected peak cor-
responded to the molecular weight of the cationic complex
16 [Figure 1b].19 These data clearly suggested that the
nitrogen atom on the sidearm is not silylated, even in
the presence of excess BSA.20 On the basis of these
results, we concluded that diaminophosphine oxide 6, the
preligand, reacts with BSA in the reaction mixture,
affording the trivalent phosphorus compound 17, which
is proposed to be the actual ligand structure (Scheme 5).21
Considering the structure of the actual ligand 17, as
well as the cationic palladium complex 16 detected by
FAB MS analysis, it is possible that 17 chelates to the
Pd metal in a bidentate manner through both the
phosphorus atom and the nitrogen atom. This led us to
question whether this Pd-ligand ratio (Pd:6 ) 1:2) was
BSA-Induced P(V) to P(III) Transformation of
Diaminophosphine Oxides. 31P NMR experiments
reported in the preceding communication indicated that
BSA is related to the generation of a trivalent phosphorus
species [chemical shift: 111.0 ppm (CDCl3)] from 6
[chemical shift: 12.5 ppm (CDCl3)].9 To obtain funda-
mental information about the mechanism of BSA-induced
P(V) to P(III) transformation of diaminophosphine oxides,
as well as to determine the actual structure of the
trivalent phosphorus species derived from 6, we first
performed NMR experiments using simple diaminophos-
phine oxide 12.16 When 12 (31P NMR chemical shift: 6.7
ppm) was reacted with BSA in CDCl3, the formation of a
trivalent phosphorus species (31P NMR chemical shift:
101.0 ppm) was observed similar to the case of 6. BSA is
a powerful reagent for the introduction of trimethylsilyl
groups to heteroatoms.15 No reaction occurred, however,
with use of chlorotrimethylsilane instead of BSA, and the
predominant peak was observed at the upfield side [31P
NMR chemical shift: -12.5 ppm (CDCl3)]. The same
transformation was achieved in combination with chlo-
rotrimethylsilane and triethylamine. The fact that no
P(V) to P(III) transformation was observed in the pres-
ence of only 12 and triethylamine indicates that tri-
methylsilylation of 12 is essential for the base-induced
P(V) to P(III) transformation. These experiments pro-
vided useful information to explain the reaction mecha-
nism of BSA-induced P(V) to P(III) transformation
(Scheme 4). The reaction is triggered by silylation of
diaminophosphine oxide 12. The resulting cationic prod-
uct 13 reacts with a BSA-derived anion to provide the
trivalent phosphorus species 14.18
(18) Structure of 14 was confirmed by comparing the NMR data of
an authentic sample, which was prepared according to the reported
method using N,N-diethyltrimethylsilylamine. For reference, see:
Plinta, H.-J.; Neda, I.; Fischer, A.; Jones, P. G.; Schmutzler, R. Chem.
Ber. 1995, 128, 695.
(19) 31P NMR analysis of a CDCl3 solution of (η3-C3H5PdCl)2, 6, and
BSA in a ratio of 1:2:60 (Pd:6 ) 1:1) exhibited two singlet signals at
100.0 and 101.7 ppm. Although the corresponding structures have not
been clarified, partial information has been obtained from the analyti-
cal experiments using 25 instead of 6. See Supporting Information for
details.
(20) No trimethylsilylation of N-methylaniline occurred in the
presence of excess BSA. This fact strongly supports our conclusion.
(21) For recent examples of P-chirogenic diamidophosphite ligands,
see: (a) Brunel, J. M.; Constantieux, T.; Buono, G. J. Org. Chem. 1999,
64, 8940. (b) Delapierre, G.; Brunel, J. M.; Constantieux, T.; Buono,
G. Tetrahedron: Asymmetry 2001, 12, 1345. (c) Tsarev, V. N.; Lyubi-
mov, S. E.; Shiryaev, A. A.; Zheglov, S. V.; Bondarev, O. G.; Davankov,
V. A.; Kabro, A. A.; Moiseev, S. K.; Kalinin, V. N.; Gavrilov, K. N. Eur.
J. Org. Chem. 2004, 2214.
Active Catalyst Structure and Reaction Mecha-
nism. Preliminary experiments using 12 were very
informative for determining the structure of the diami-
dophosphite moiety of the trivalent phosphorus species
derived from 6. There is no experimental information,
J. Org. Chem, Vol. 70, No. 18, 2005 7175

Nemoto et al.
SCHEME 6. Proposed Reaction Mechanism
SCHEME 5. Actual Ligand Structure
TABLE 3. Effect of Pd-Ligand Ratio
Pd-ligand
run substrate
ratio
time
yieldb
eec
1
2
3
4
5
6
10
10
10
7a
7a
7a
1:2
1:1.5
1:1
1:2
1:1.5
1:1
0.5 h 84%
99% ee type A
98% ee
0.5 h 57%
0.5 h no reaction
2 h
2 h
2 h
89%
93% ee type B
93% ee
34%
no reaction
a Reaction conditions. Runs 1-3: dimethyl malonate (3.0 equiv),
BSA (3.0 equiv), CH2Cl2 (0.2 M), rt. Runs 4-6: 8a (1.5 equiv),
BSA (4.0 equiv), Zn(OAc)2 (10 mol %), toluene (0.15 M), rt.
b Isolated yield. c Determined by HPLC analysis.
molecule of Zn(OAc)2, yielding product 9a with high
enantiomeric excess [Scheme 6b].24
optimal. To resolve this issue, we investigated the Pd-
ligand ratio using two representative reactions: 10 to 11
(reaction type A) and 7a to 9a (reaction type B) (Table
3). When the catalyst was prepared in different Pd-
ligand ratios, there was a remarkable decrease in reac-
tivity without any loss of enantioselectivity, as compared
to the case of the best ratio. In particular, no reaction
occurred using the catalyst prepared from the Pd source
and 6 in a ratio of 1:1. In addition, there were positive
nonlinear effects22 in both types of reactions.16 These
results led us to hypothesize that the Pd-17 (1:2)
complex exists in the reaction mixture and might function
as the active species.
To elucidate the detailed reaction profiles, we per-
formed kinetic experiments.16 The reaction rate of allylic
alkylation of 10 with dimethyl malonate (reaction type
A) had first-order dependence on the Pd catalyst, zero-
order dependence on 10, and first-order dependence on
dimethyl malonate. Similarly, the reaction rate of allylic
substitution of 7a with 8a (reaction type B) had first-
order dependence on the Pd catalyst, zero-order depen-
dence on 7a, first-order dependence on 8a, and first-order
dependence on Zn(OAc)2. These reaction profiles clearly
support our expectation that two molecules of the ligand
chelate to the Pd metal and result in a positive nonlinear
effect. Consequently, we propose the following reaction
mechanism. Complexation of the Pd source and 6 first
occurred in the presence of BSA, affording the Pd(0)
complex 18 (Pd:17 ) 1:2), which is proposed to be the
common active species for both types of reactions [Scheme
6a]. Kinetic experiments also indicate that, in the case
of reaction type B, the rate-determining nucleophilic
attack of enol silyl ether 2023 to a cationic π-allylpalla-
dium complex 19 is achieved in cooperation with a single
Source of Enantioselection in the Asymmetric
Catalysis. The sources of enantioselection are quite
different between the two types of reactions examined:
attack at enantiotopic termini of the meso-π-allyl complex
(reaction type A) and differentiation of the prochiral
nucleophile face (reaction type B).25 Despite the mecha-
nistic difference, a high level of enantioselection was
realized in both cases by the same Pd complex 18. The
fact that the cooperation of Zn(OAc)2 is indispensable to
the high stereoselectivity in reaction type B suggested
that nitrogen atoms on the sidearms in 19 fix the
prochiral nucleophile in the appropriate position through
a secondary ligand substrate interaction13a-c,26 mediated
by Zn metal. To examine this possibility, we first inves-
tigated the effect of ligand modifications (Table 4). When
structurally modified diaminophosphine oxides 22-2527
were used as the ligand, there were no significant
changes in the stereoselectivity in the case of reaction
(23) In general, reaction of BSA with â-keto esters proceeds readily
to afford the corrsponding trimethylsilyl enol ethers (see ref 15). Enol
silyl ether 20 was observed in 1H NMR, when 8a was treated with 1
equiv of BSA in toluene-d8. When 20 itself was used as the substrate,
however, a remarkable decrease in the reactivity was observed, despite
the ee value remaining unchanged [16 h, 13% yield, 93% ee, condi-
tions: Pd catalyst (2 mol %), 6 (4 mol %), Zn(OAc)2 (10 mol %), 20 (1.5
equiv), BSA (2.5 equiv), toluene, rt]. For discussions about this result,
see Supporting Information.
(24) In the early 1980s, Negishi and co-workers reported their
pioneering work on the Pd-catalyzed allylation of Zn enolates, where
enolate nucleophiles attacked the π-allylpalladium species at the side
opposite to the Pd metal. [For references, see: (a) Negishi, E.;
Matsushita, H.; Chatterjee, S.; John, R. A. J. Org. Chem. 1982, 47,
3188. (b) Negishi, E.; John, R. A. J. Org. Chem. 1983, 48, 4098.] The
obtained kinetic profiles (first-order dependence on 8a and first-order
dependence on Zn(OAc)2), however, clearly indicate that Zn enolates
are not generated in our reaction system.
(25) For reviews, see: (a) Trost, B. M. Chem. Rev. 1996, 96, 395. (b)
Trost, B. M.; Crawley, M. L. Chem. Rev. 2003, 103, 2921.
(26) For reviews, see: (a) Sawamura, M.; Ito, Y. Chem. Rev. 1992,
92, 857. (b) Bo¨rner, A. Eur. J. Inorg. Chem. 2001, 327.
(22) For review, see: Girard, C.; Kagan, H. B. Angew. Chem., Int.
Ed. 1998, 37, 2923.
7176 J. Org. Chem., Vol. 70, No. 18, 2005

P-Chirogenic Diaminophosphine Oxides
TABLE 4. Effect of Ligand Modifications
TABLE 5. Effect of Ligand Structure
run substrate ligand time
yieldb
eec
run substrate ligand time
yieldb
eec
1
2
10
10
10
10
10
10
7a
7a
7a
7a
7a
7a
6
0.5 h 84%
99% ee
98% ee
94% ee
97% ee
97% ee
-80% eed
93% ee
79% ee
64% ee
84% ee
63% ee
type A
type B
1
2
3
4
5
6
7
8
10
10
10
10
7a
7a
7a
7a
6
24 h 94%
99% ee type A
87% ee
22
23
24
25
26
6
0.5 h 98%
0.5 h 82%
0.5 h 50%
0.5 h 85%
0.5 h 96%
27
28
29
6
24 h 27%
3
24 h no reaction
24 h no reaction
16 h 99%
16 h no reaction
16 h no reaction
16 h no reaction
4
5
93% ee type B
6
27
28
29
7
2 h
2 h
2 h
2 h
2 h
2 h
89%
8
22
23
24
25
26
37%
9
55%
10
11
12
90%
29%
no reaction
aReaction conditions. Runs 1-4: dimethyl malonate (3.0 equiv),
BSA (3.0 equiv), CH2Cl2 (0.2 M), rt. Runs 5-8: 8a (1.5 equiv),
BSA (4.0 equiv), Zn(OAc)2 (10 mol %), toluene (0.15 M), rt.
b Isolated yield. c Determined by HPLC analysis.
TABLE 6. Solvent Effect
run substrate solvent time
yieldb
CH2Cl2 24 h 94%
THF 24 h 49%
eec
aReaction condition. Runs 1-6: dimethyl malonate (3.0 equiv),
BSA (3.0 equiv), CH2Cl2 (0.2 M), rt. Runs 7-12: 8a (1.5 equiv),
BSA (4.0 equiv), Zn(OAc)2 (10 mol %), toluene (0.15 M), rt.
b Isolated yield. c Determined by HPLC analysis. d Negative value
means the opposite enantiomer.
1
2
3
4
5
6
7
8
10
10
10
10
7a
7a
7a
7a
99% ee type A
96% ee
96% ee
96% ee
60% ee type B
78% ee
93% ee
toluene 24 h 67%
CH3CN 24 h 98%
CH2Cl2 16 h 23%
THF
16 h 41%
toluene 16 h 99%
CH3CN 16 h no reaction
type A (runs 1-5). In contrast, remarkable changes were
observed in the case of reaction type B. Electronic and
steric changes on the nitrogen atom (runs 8, 9), as well
as changes in the sidearm length (run 10), influenced
catalytic activity, resulting in decreased reactivity and
selectivity. Particularly noteworthy is that there was a
significant decrease in the enantioselectivity, when 25,
a ligand without a nitrogen atom on the sidearm, was
used (run 11). With the use of 26,27 a ligand possessing
the (S)-configuration on the phosphorus atom, opposite
stereoselection was observed in reaction type A (run 6).
In addition, no reaction occurred when 26 was used in
reaction type B (run 12). These results clearly indicate
the importance of chirality on the phosphorus atom. The
effect of the ligand structure was also investigated using
the known chiral diaminophosphine oxides 27,28a 28,28a
and 2928b (Table 5). The structure of the chiral diamino-
phosphine oxide dramatically affected the catalytic activ-
ity, suggesting that ligand structure with a sidearm is
important for achieving the asymmetric catalysis. More-
over, the striking difference in the solvent effect indicates
that the polarity of the reaction media dramatically
affected the transition state structure incorporated with
Zn(OAc)2 (Table 6). These experimental findings led us
to conclude that the secondary ligand substrate interac-
a Reaction conditions. Runs 1-4: dimethyl malonate (3.0 equiv),
BSA (3.0 equiv), CH2Cl2 (0.25 M), rt. Runs 5-8: 8a (1.5 equiv),
BSA (4.0 equiv), Zn(OAc)2 (10 mol %), toluene (0.15 M), rt.
b Isolated yield. c Determined by HPLC analysis.
tion mediated by N-Zn coordination has a crucial role
in the enantiofacial recognition of prochiral nucleophiles
in reaction type B.29
To obtain more detailed structural information for
discussion on the sources of enantioselection, we tried
crystallizing the catalyst species. Unfortunately, a crystal
for X-ray analysis could not be obtained. Although the
complete transition state is not clear, the mechanistic
findings, as well as the absolute configuration of the
products, suggest that the enantiofacial discrimination
of prochiral nucleophiles in reaction type B would be
explained by the working model shown in Figure 2. The
observed absolute configuration of 9a and other products
indicate that the C-C bond formation occurs on the Si-
face of the enolate nucleophile. The fact that nucleophiles
(29) It is known that coordination affinity of amines for Zn(II) species
is dependent on the steric requirement of the amine rather than the
amine basicity. In particular, tertiary amines such as triethylamine
have extremely low coordination affinity compared with those of
primary and secondary amines. [For references, see: (a) Higgins, G.
M. C.; Saville, B. J. Chem. Soc. 1963, 2812. (b) Coates, E.; Rigg, B.
Saville, B.; Skelton, D. J. Chem. Soc. 1965, 5613.] This information
provides us with a reasonable explanation for the observed stereose-
lectivity in the reaction using 23 (Table 4, run 9), where the sidearm
with a tertiary amine would be recognized as one of the steric elements,
analogous to that of 25.
(27) Diaminophosphine oxide 24 was prepared from (S)-glutamic
acid; diaminophosphine oxides 25 and 26 were prepared from (S)-
homophenylalanine. See Supporting Information for details.
(28) (a) Koeller, K. J.; Spilling, C. D. Tetrahedron Lett. 1991, 32,
6297. (b) Devitt, P. G.; Kee, T. P. Tetrahedron 1995, 51, 10987. (c) See
also, Mareno, G. E.; Quintero, L.; Bernes, S.; de Parrodi, C. A.
Tetrahedron Lett. 2004, 45, 4245.
J. Org. Chem, Vol. 70, No. 18, 2005 7177

Nemoto et al.
ammonium salt, the organic phase was concentrated in vacuo.
The obtained residue was purified by flash column chroma-
tography (SiO2, hexane/ethyl acetate 5/1 to 1/1) to afford 6 (1.26
g, 51%): mp 131-132 °C; IR (KBr) ν 3370, 2927, 1601, 1524,
1502, 1327, 1303, 1219, 1170, 1127, 1034, 950, 751, 730, 700
cm-1 1H NMR (CDCl3) δ 1.67-1.76 (m, 1H), 2.00-2.08 (m,
;
1H), 3.04-3.18 (m, 2H), 3.31-3.37 (m, 1H), 3.45 (s, 1H), 3.65
(m, 1H), 3.76-3.82 (m, 1H), 4.20 (dd, J ) 13.2 Hz (PNCH),
15.2 Hz (HCH), 1H), 4.44 (dd, J ) 10.4 Hz (PNCH), 15.2 Hz
(HCH), 1H), 6.50-6.53 (m, 2H), 6.71-6.75 (m, 1H), 6.99-7.04
(m, 1H), 7.15-7.19 (m, 4H), 7.29-7.47 (m, 7H) 7.70 (d, J )
628 Hz (OdPH), 1H); 13C NMR (CDCl3) δ 32.7 (d, J ) 3.9 Hz),
39.6, 46.5 (d, J ) 6.1 Hz), 48.9 (d, J ) 10.4 Hz), 52.1 (d, J )
7.2 Hz), 112.8 (x 2), 115.9 (d, J ) 5.3 Hz) (× 2), 117.9, 121.9,
127.8, 128.6 (× 2), 128.7 (× 2), 129.3 (× 2), 129.5 (× 2),136.7,
141.4 (d, J ) 7.7 Hz), 147.6; 31P NMR (CDCl3): δ 12.5; EI-
LRMS m/z 391 (M+); FAB-LRMS m/z 392 (MH+); [R]18D -38.6
(c 0.5, CHCl3, >99% ee); FAB HRMS calcd for C23H27N3OP:
392.1845 (MH+), found 392.1854 (MH+). Anal. Calcd for
C23H26N3OP: C, 70.57; H, 6.69; N, 10.73. Found, C, 70.27; H,
6.64; N, 10.66. The enantiomeric excess was determined by
HPLC analysis (DAICEL CHIRALCEL OD-H, 2-propanol/
hexane 1/1, flow rate 0.4 mL/min, tR 49.5 min (ligand prepared
from D-Asp) and 54.2 min (ligand prepared from L-Asp),
detection at 254 nm).
FIGURE 2. Working model. L* ) 17. The trimethylsiloxy
group on the phosphorus atom has been omitted for clarity.
with a smaller methyl ester gave higher enantioselec-
tivities suggests that the ester moiety of the nucleophiles
is located inside the chiral pocket. This spatial arrange-
ment is stabilized through the Zn-mediated secondary
ligand substrate interaction. Effective fixation of the
three reactants results in not only an increase in the
enantiofacial discrimination of nucleophiles but also an
enhancement of the reactivity. The clarified multifunc-
tional properties of the P-chirogenic diaminophosphine
oxides will be extended to other catalytic asymmetric
reactions.
General Procedure for the Pd-Catalyzed Asymmetric
Construction of Quaternary Carbons Using (S,RP)-Ph-
DIAPHOX (Table 3, Run 1). To a stirred mixture of cinnamyl
acetate (7a) (83.3 µL, 0.5 mmol), keto ester (8a) (120 µL, 0.75
mmol), (η3-C3H5PdCl)2 (1.8 mg, 0.0025 mmol), 6 (7.8 mg, 0.02
mmol), and Zn(OAc)2 (9.2 mg, 0.05 mmol) in toluene (3.33 mL)
was added BSA (0.494 mL, 2.0 mmol) at room temperature.
After being stirred for 16 h, the reaction was concentrated
under reduced pressure, and the obtained residue was purified
by flash column chromatography (SiO2, hexane/ethyl acetate
30/1 to 15/1) to give (1S)-2-oxo-1-(3-phenyl-allyl)-cyclohexan-
ecarboxylic acid ethyl ester (9a) as a colorless oil (143 mg, 99%,
93% ee). [R]18D -80.9 (c 0.85, CHCl3, 93% ee). The enantiomeric
excess was determined by HPLC analysis (DAICEL CHIRAL-
CEL OD-H, 2-propanol/hexane 5/95, flow rate 0.4 mL/min, tR
14.5 min [(R)-isomer] and 16.8 min [(S)-isomer], detection at
254 nm).
Experimental Section
Experimental Procedure for the Transformation of
Triamine to P-Chirogenic Diaminophosphine Oxides.
Synthesis of (S,RP)-Ph-DIAPHOX (6) from Triamine 4:
Method A. To a stirred solution of 4 (1.99 g, 5.77 mmol) and
triethylamine (2.90 mL, 20.8 mmol) in toluene (29 mL) at -78
°C was add dropwise phosphorus trichloride (0.553 mL, 6.35
mmol) over 3 min. The reaction temperature was gradually
warmed to room temperature, and then the mixture was kept
stirring for 12 h. The reaction mixture was diluted with CHCl3,
and the organic layer was washed with H2O, and then dried
over Na2SO4. After concentration in vacuo, the obtained
residue was dissolved into ethyl acetate, and then SiO2 (50
mg/mmol substrate) and H2O (0.13 mL, 7.21 mmol) were
added. After being stirred for 20 h at room temperature under
air, the reaction was diluted with CHCl3 and then filtered
through a short pad of silica gel, and the silica residue was
washed with CHCl3 and ethyl acetate intensively. After
concentration of the organic washings in vacuo, the obtained
yellow solid was washed with ethyl acetate and then filtered
to afford 6 as a white solid. Additionally, purification of the
yellow residue by flash column chromatography (SiO2, hexane/
ethyl acetate 5/1 to 1/1) afforded 6 (total yield: 1.36 g, 60%).
Method B. To a stirred solution of 4 (2.20 g, 6.35 mmol)
and triethylamine (3.44 mL, 24.7 mmol) in toluene (42 mL) at
-78 °C was add dropwise phosphorus trichloride (0.610 mL,
6.98 mmol) over 5 min. The reaction temperature was gradu-
ally warmed to room temperature, and then the mixture was
kept stirring for 12 h. After filtration of the generated
Acknowledgment. This work was supported by a
Grant-in Aid for Encouragement of Young Scientists (B)
from the Ministry of Education, Culture, Sports, Sci-
ence, and Technology, Japan, and the Banyu Award in
Synthetic Organic Chemistry, Japan.
Supporting Information Available: Experimental pro-
cedures, compound characterization, data for the nonlinear
effect and the kinetic experiments, and other data and dis-
cussions. This material is available free of charge via the
Internet at http://pubs.acs.org.
JO050800Y
7178 J. Org. Chem., Vol. 70, No. 18, 2005