Electronic differences between coordinating functionalities of chiral phosphine-phosphites and effects in catalytic enantioselective hydrogenation

Article abstract of DOI:10.1021/om020140c

A convenient synthesis of new chiral phosphine-phosphites (P-OP) has been described. The versatility of the synthetic protocol developed has allowed the preparation of ligands with different phosphine fragments and the choice of the stereogenic element location. Analyses of the values of ¹J_PSe of the corresponding diselenides are in accord with the expected lower σ-donor ability of the phosphite fragment, with respect to the phosphine group, and with an increase of phosphine basicity after substitution of phenyl substituants by methyl groups. Inspection of v(CO) values on a series of complexes RhCl(CO)(P-OP) demonstrated a variable π-aceptor ability of the phosphite group, compensating for the change of basicity of the phosphine functionality, as well as having a rather reduced electron density at the metal center compared with diphosphine analogues. The distinct nature of the phosphorus functionalities has also been evidenced in rhodium-catalyzed enantioselective hydrogenation of methyl Z-α-acetamido-cinnamate (MAC). Thus, the coordination mode of the substrate is governed by the chiral ligand, directing the olefinic bond to a cis position with respect to the phosphite group, as demonstrated by NMR studies performed with [Rh(P-OP)(MAC)]⁺ complexes. In consequence, the phosphite group has a greater impact on the enantioselectivity of the product. However, the optical purity of the process also depends on the nature of the phosphine group, and hence, an appropriate election of both phosphorus functionalities is required for the attainment of excellent enantioselectivities (99% ee).

Full text of DOI:10.1021/om020140c

Organometallics 2002, 21, 4611-4621
4611
Electr on ic Differ en ces betw een Coor d in a tin g
F u n ction a lities of Ch ir a l P h osp h in e-P h osp h ites a n d
Effects in Ca ta lytic En a n tioselective Hyd r ogen a tion
Andre´s Sua´rez,† Miguel A. Me´ndez-Rojas,‡ and Antonio Pizzano*,†
Instituto de Investigaciones Quı´micas, Consejo Superior de Investigaciones
Cientı´ficas-Universidad de Sevilla, c/ Ame´rico Vespucio s/ n, Isla de la Cartuja,
41092 Sevilla, Spain, and Centro de Investigaciones Qu´ımicas, Universidad Auto´noma
del Estado de Hidalgo, Pachuca 42073, Hidalgo, Mexico
Received February 20, 2002
A convenient synthesis of new chiral phosphine-phosphites (P-OP) has been described.
The versatility of the synthetic protocol developed has allowed the preparation of ligands
with different phosphine fragments and the choice of the stereogenic element location.
Analyses of the values of 1J PSe of the corresponding diselenides are in accord with the expected
lower σ-donor ability of the phosphite fragment, with respect to the phosphine group, and
with an increase of phosphine basicity after substitution of phenyl substituents by methyl
groups. Inspection of υ(CO) values on a series of complexes RhCl(CO)(P-OP) demonstrated
a variable π-aceptor ability of the phosphite group, compensating for the change of basicity
of the phosphine functionality, as well as having a rather reduced electron density at the
metal center compared with diphosphine analogues. The distinct nature of the phosphorus
functionalities has also been evidenced in rhodium-catalyzed enantioselective hydrogenation
of methyl Z-R-acetamido-cinnamate (MAC). Thus, the coordination mode of the substrate is
governed by the chiral ligand, directing the olefinic bond to a cis position with respect to the
phosphite group, as demonstrated by NMR studies performed with [Rh(P-OP)(MAC)]+
complexes. In consequence, the phosphite group has a greater impact on the enantioselectivity
of the product. However, the optical purity of the process also depends on the nature of the
phosphine group, and hence, an appropriate election of both phosphorus functionalities is
required for the attainment of excellent enantioselectivities (99% ee).
In tr od u ction
ligands of the type P-N, P-O, P-P′, or P-S has been
applied in different transformations.4 In particular, we
have been interested in the synthesis of chiral phos-
phine-phosphites and their application in asymmetric
catalysis. More interestingly, ligands of this kind have
produced a benchmark in the study of asymmetric
hydroformylation reactions giving outstanding enantio-
selectivities associated with a remarkable substrate
scope.5 These bifunctional molecules once again deserve
attention and have provided satisfactory results in
several asymmetric catalytic processes such as hydro-
genation and allylic substitution.6
Chelating ligands with two different coordinating
functionalities (L-L′) are receiving a growing interest
in asymmetric catalysis.1 When compared with broadly
used C2 symmetric derivatives (L-L), the former can
introduce a bigger number of diastereomeric transition
states making the stereocontrol of the process more
difficult.2 However, this complicated picture can be
greatly simplified if the bidentate ligand influences the
coordination mode of the prochiral substrate, dictating
the coordination position where the stereogenic center
will be generated.3 Therefore, the bigger complexity
introduced by C1 symmetric ligands should be advanta-
geous, since they offer the possibility of optimizing
fragments L and L′, thus rendering chiral environments
inaccessible for C2 symmetric ligands.
In this contribution we report the synthesis of new
phosphine-phosphites based on a rather rigid achiral
o-oxyphenyl bridge (A).7 The synthetic protocol de-
With respect to processes catalyzed by medium- and
late-transition metals, the large diversity of bifunctional
* To whom correspondence should be addressed. E-mail: pizzano@
cica.es.
† Instituto de Investigaciones Qu´ımicas.
‡ Centro de Investigaciones Qu´ımicas.
(1) Comprehensive Asymmetric Catalysis; J acobsen, E. N., Pfaltz,
A., Yamamoto, H., Eds; Springer, Berlin, Germany, 1999.
(2) Whitesell, J . K. Chem. Rev. 1989, 89, 1581.
(3) For successful applications of this concept see: (a) Evans, D. A.;
Campos, K. R.; Tedrow, J . S.; Michael, F. E.; Cagne, M. R. J . Am. Chem.
Soc. 2000, 122, 7905. (b) Pre`tot, R.; Pfaltz, A. Angew. Chem., Int. Ed.
1998, 37, 323.
scribed herein allows the preparation of a series of
ligands which can differ in the nature of the phosphine
fragment, as well as in the position of the stereogenic
element(s), covering a wide structural diversity for
catalyst optimization.8 In addition, we have performed
10.1021/om020140c CCC: $22.00 © 2002 American Chemical Society
Publication on Web 10/05/2002

4612 Organometallics, Vol. 21, No. 22, 2002
Sua´rez et al.
Sch em e 1
Sch em e 2
studies regarding the electronic characteristics of these
ligands and about their performance in rhodium-
catalyzed olefin asymmetric hydrogenation. These re-
sults have been complemented by information about
ligand structure, determined by an X-ray crystallogra-
phy study, as well as with NMR investigations regard-
ing the coordination mode of the prochiral olefin toward
the metal-ligand moiety.
Resu lts a n d Discu ssion
Liga n d Syn th esis. We initially pursued a versatile
synthesis of phosphine-phosphites of type A that can
allow us their application in asymmetric catalysis
following a modular approach. As shown below, we have
approached the synthesis of phosphine-phosphites A
by the condensation of the appropriate phenol-phos-
phine B9,10 and chlorophosphite C. Reagents B can be
conveniently obtained by demethylation of o-anisyl
phosphines with BBr3.11 Thus, treatment of phosphine
1a (Scheme 1) with 2.3 molar equiv of BBr3, followed
by reaction with MeOH, produced the phosphonium
bromide 2a . This reaction sequence also cleanly yielded
the corresponding methyl (2b) and isopropyl (2c) de-
rivatives.
Subsequent treatment of phosphonium salts 2a -c
with NEt3 produced a quantitative deprotonation to
afford the corresponding phenol-phosphines 3. How-
ever, the intermediacy of the phosphonium salts is of
practical convenience, since alkyl derivatives 2b and 2c
are less air-sensitive solids than their phosphine coun-
terparts. For this reason we have used them as stock
reagents to generate in situ the corresponding phos-
phines.
A touchstone to the procedure depicted in Scheme 1
comes from the investigation of the reaction sequence
on an enantiopure starting material. Noteworthy, treat-
ment of (S)-PAMP12 produced the corresponding phenol-
phosphine (S)-3d without racemization. This was con-
firmed by a comparative analysis, performed by 31P{1H}
NMR spectroscopy, of the phosphine-phosphites ob-
tained by condensation between (S)-3d and (rac)-3d
with chlorophosphite (S)-4a (see Experimental Section).
The expected retention of phosphorus configuration was
ensured by the preparation of the borane adduct of (S)-
3d ,10 previously described in the literature, and exami-
nation of its optical rotation.
Finally, a simple condensation between phenol phos-
phines 3 and an appropriate chlorophosphite 4 led to
the desired phosphine-phosphites in good yields (5,
Scheme 2). For this purpose we intentionally chose both
enantiomers of 3,3′-di-tert-butyl-5,5′,6,6′-tetramethyl-
2,2′-bisphenoxyphosphorus chloride (4a ),13 as well as
conformationally flexible 3,3′,5,5′-tetra-tert-butyl-2,2′-
bisphenoxyphosphorus chloride (4b),14 to cover a variety
of situations with respect to the position of the chiral
element(s) in the ligand: at the phosphite, at the
phosphine, or at both fragments.
(4) See for instance; P-N ligands: Alcock, N.; Hulmes, D. I.; Brown,
J . M. J . Chem. Soc., Chem. Commun. 1995, 395. Wimmer, P.; Wildham,
M. Tetrahedron: Asymmetry 1995, 6, 657. P-O ligands: Nandy, M.;
J in, J .; RajanBabu, T. V. J . Am. Chem. Soc. 1999, 121, 9899. Uozumi,
Y.; Tanahashi, A.; Lee, S.-Y.; Hayashi, T. J . Org. Chem. 1993, 58, 1945.
P-S ligands: Hauptman, E.; Fagan, P. J .; Marshall, W. Organome-
tallics 1999, 18, 2061. P-P′ ligands: Ireland, T.; Grossheimann, G.;
Wieser-J eunesse, C.; Knochel, P. Angew. Chem., Int. Ed. 1999, 38,
3212. Ohashi, A.; Imamoto, T. Tetrahedron Lett. 2001, 42, 1099.
Tanaka, K.; Fu, G. C. J . Org. Chem. 2001, 66, 8177.
(5) (a) Nozaki, K.; Sakai, N.; Nanno, T.; Higashijima, T.; Mano, S.;
Horiuchi, T.; Takaya, H. J . Am. Chem. Soc. 1997, 119, 4413. (b)
Horiuchi, T.; Ohta, T.; Shirakawa, E.; Nozaki, K.; Takaya, H. J . Org.
Chem. 1997, 62, 4285.
(6) (a) Deerenberg, S.; Schrekker, H. S.; Strijdonck, G. P. F.; Kamer,
P. C. J .; van Leeuwen, P. W. N. M. J . Org. Chem. 2000, 65, 4810. (b)
Deerenberg, S.; Kamer, P. C. J .; van Leeuwen, P. W. N. M. Organo-
metallics 2000, 19, 2065. (c) Deerenberg, S.; Pa`mies, O.; Die´guez, M.;
Claver, C.; Kamer, P. C. J .; van Leeuwen, P. W. N. M. J . Org. Chem.
2001, 66, 7626. (d) Pa`mies, O.; Die´guez, M.; Net, G.; Ruiz, A.; Claver,
C. J . Org. Chem. 2001, 66, 8364.
(7) For other derivatives based on this backbone see: (a) Baker, M.
J .; Pringle, P. J . Chem. Soc., Chem. Commun. 1993, 314. (b) Kranich,
R.; Eis, K.; Geis, O.; Mu¨hle, S.; Bats, J . W.; Schmalz, H.-G. Chem. Eur.
J . 2000, 6, 2874.
(8) This synthetic protocol has been described in a preliminary
form: Sua´rez, A.; Pizzano, A. Tetrahedron: Asymmetry 2001, 12, 2501.
(9) Several methods have been used for the synthesis of diverse
phenol phosphines. For representative procedures see: (a) Rauchfuss,
T. B. Inorg. Chem. 1977, 16, 2966. (b) Heinicke, J .; Kadyrov, R.;
Kindermann, M. K.; Koesling, M.; J ones, P. G. Chem. Ber. 1996, 129,
1547. (c) Schmutzler, R.; Schomburg, D.; Bartsch, R.; Stelzer, O. Z.
Naturforsch. 1984, 39, 1177.
(10) Only recently, J uge´ and co-workers have described the first
enantioselective synthesis of a P-stereogenic 2-hydroxyaryl phosphine
((R)-o-anisyl-2-hydroxynaphthylphenyl phosphine) using an alternative
procedure; however, this method requires difficultly accessible enan-
tiopure phosphinite boranes: Moulin, D.; Bago, S.; Bauduin, C.; Darcel,
C.; J uge´, S. Tetrahedron: Asymmetry 2000, 11, 3939.
(11) This sequence of reactions has been used previously in the
synthesis of some quinolyl phosphines: Sembiring, S. B.; Colbran, S.
B.; Craig, D. C. Inorg. Chem. 1995, 34, 761.
(12) J uge´, S.; Ste´phan, M.; Laffitte, J . A.; Genet, J . P. Tetrahedron
Lett. 1990, 31, 6357.

Studies of New Chiral Phosphine-Phosphites
Organometallics, Vol. 21, No. 22, 2002 4613
Sch em e 3
Electr on ic Ch a r a cter istics of Liga n d s 5. The most
interesting feature of molecules 5 is the presence of two
phosphorus functionalities with rather different elec-
tronic properties. When coordinated to a metal center,
a good σ-donor abilitiy can be expected for the phosphine
group, whereas a phosphite fragment should be a poorer
σ-donor and a better π-acceptor. A simple way of
evaluating the σ-donor ability of either a phosphine or
a phosphite functionality is by measuring the magnitude
1
of J SeP in the 77Se isotopomer of the corresponding
phosphine-selenide or seleno-phosphate, respectively.15
To estimate the donor abilities of both phosphorus
functionalities in compounds 5 we prepared the corre-
sponding diselenides 7 by simply reacting phosphine-
phosphites 5 with elemental selenium in toluene at 100
°C (Scheme 3). Interestingly, this reaction proceeded in
a stepwise manner through the intermediacy of phos-
phine selenides 6, and while the phosphine fragment
showed high reactivity (for instance t1/2 is ca. 0.5 h at
room temperature for 5b, Figure 1), the phosphite
functionality was rather unreactive, needing prolonged
heating times (2-8 days) to complete the reaction.
Values of 31P-77Se coupling constants observed in com-
pounds 7 (CDCl3) are compiled in Table 1.
F igu r e 1. 31P{1H} NMR (C6D5CD3, 162 MHz) of the
reaction between 5b and Se: (a) 30 min, room temperature;
(b) 18 h, 100 °C; and (c) 3 d, 100 °C.
Ta ble 1. 31P -77Se Cou p lin g Con sta n ts (Hz) in
Com p ou n d s 7
1
entry
compd
1J PSe(phosphine)
1J PSe(phosphite)
From the magnitude of J PSe in compounds 7 it is
possible to draw some interesting conclusions. First, the
phosphine fragments of ligands 5 exhibit the expected
donor ability trend, and are ordered on a decreasing
basicity (including in parentheses the coupling constant
in Hz of the phosphine-selenide fragment of 7, entries
1-2 and 4-6) as follows: 5b (703) > 5d (716), 5e (711),
5f (716) > 5a (732),15 i.e., by successive substitution of
methyl by phenyl groups. It should also be noticed that
the value obtained for the isopropyl derivative 7c (719
Hz, entry 3) is higher than expected, since this ligand
should be the more basic at the phosphine site.16
1
2
3
4
5
6
7a
7b
7c
7d
7e
7f
732
703
719
716
711
716
1058
1053
1053
1053
1061
1055
Sch em e 4
On the other hand, the selenophosphate fragment of
compounds 7 showed, as expected, a markedly higher
value of 1J PSe than the phosphine-selenide functional-
ity. All compounds exhibited coupling constants in the
range 1053-1061 Hz, therefore there are no significant
differences between them, although they are clearly
higher (ca. 30-40 Hz) than values exhibited by acyclic
triaryl phosphites.15c This further reduced basicity can
be attributed to the geometrical constraint produced by
the cyclic nature of the phosphite fragment.17
To ascertain the electronic properties of ligands 5 we
have prepared a series of compounds RhCl(CO)(5),18 as
summarized in Scheme 4. The resulting complexes 8
exhibit spectroscopic data in accord with the proposed
formulation. The low value of the coupling constant
(13) The two enantiomers of chlorophosphite 4a can be readily
prepared from the corresponding optically active bisphenol. These diols
are easily available on multigram scale, see: Alexander, J .; Schrock,
R. R.; Davis, W. M.; Hultzsch, K. C.; Hoveyda, A. H.; Houser, J . H.
Organometallics 2000, 19, 3700.
(14) Buisman, G. J . H.; Kamer, P. C. J .; van Leeuwen, P. W. N. M.
Tetrahedron: Asymmetry 1993, 4, 1625.
1
J PRh exhibited by the phosphine fragment (ca. 120 Hz)
is indicative of a trans coordination between this
functionality and the CO ligand,19 further confirmed by
an X-ray study (see below).
All compounds 8 exhibit very similar υ(CO) values,
in the proximity of 2050 cm-1. These are higher by ca.
(15) (a) Allen, D. W.; Taylor, B. F. J . Chem. Soc., Dalton Trans. 1982,
51. (b) Socol, S. M.; Verkade, J . G. Inorg. Chem. 1984, 23, 3487. (c)
Barnard, T. S.; Mason, M. R. Organometallics 2001, 20, 206.
(16) We do not have a definite explanation for this deviation.
However, it could be due to a distortion of the molecule caused by steric
(17) Kroshefsky, R. D.; Verkade, J . G. Inorg. Chem. 1979, 18, 469.
(18) The magnitude of this parameter has been amply used as a
measure of the electron density at a metal center; for a recent
hindrance. In this respect it can be recalled that highly basic But
-
3
PdSe also exhibited a lower coupling constant value than expected
(712 Hz, ref 15).
application see: Serron, S.; Huang, J .; Nolan, S. P. Organometallics
1998, 17, 534.

4614 Organometallics, Vol. 21, No. 22, 2002
Sua´rez et al.
roughly divided in two halves by the coordination plane.
This causes one But group to be placed below the
coordination plane toward the CO ligand, and the other
above the plane directed to the backbone. The confor-
mation of the latter, with the phenylene group oriented
down the coordination plane, is similar to that observed
in other related complexes.7b,23 This arrangement should
minimize the interaction between the backbone and the
phosphite moiety (structure D), whereas the alternative
conformation with the backbone aromatic ring placed
above the coordination plane would be hampered by its
proximity to the upper But group (E). As a consequence
F igu r e 2. ORTEP drawing of 8a with the atom-labeling
scheme. Thermal ellipsoids are drawn at the 30% prob-
ability level (hydrogen atoms and solvent molecule are
omitted for simplicity).
Ta ble 2. Selected Bon d Len gth s (Å) a n d An gles
(d eg) for 8c
C(43)-Rh(1)
Cl(1)-Rh(1)
P(1)-Rh(1)
P(2)-Rh(1)
1.902(5)
2.3651(12)
2.3234(12)
2.1484(12)
P(2)-Rh(1)-P(1)
C(43)-Rh(1)-Cl(1)
P(2)-Rh(1)-Cl(1)
P(1)-Rh(1)-Cl(1)
88.58(4)
89.02(15)
176.25(5)
91.73(4)
50 cm-1 than those reported for the diphosphine ana-
logues RhCl(CO)(R2PCH2CH2PR2),20,21 clearly indicating
that ligands 5 give rise to less electron-rich compounds
than the chiral diphosphines most often used in asym-
metric catalysis.
of the backbone conformation, a chiral, propeller-like24
distribution of the phenyl groups of the phosphine
fragment is produced, such that the substituent defined
by C(7) occupies an edge-oriented, pseudoaxial position,
while the phenyl defined by C(1) is placed in a pseudo-
equatorial position with a face orientation.
In good agreement with the assumption of a preferred
conformation, a 2D 1H-1H NOESY experiment per-
formed with compound 8c showed an intense NOE
cross-peak between the aromatic proton placed ortho
with respect to the phosphine group and only one of the
methine protons of the Pri groups (Figure 3). In addition,
31P{1H} NMR studies performed at variable tempera-
ture with the latter compound displayed the existence
of only one species in solution in the range of 300-180
K.
Str u ctu r e of Coor d in a ted P h osp h in e-P h osp h i-
tes. To obtain information about the structure of a
coordinated phosphine-phosphite 5 we have studied by
X-ray crystallography compound 8a . Figure 2 shows an
ORTEP view of the molecular structure of the complex,
while Table 2 collects the values of some selected bond
distances and angles. This complex shows square-planar
coordination, with the carbonyl trans to the phosphine
fragment. A clear difference between the distances of
Rh-P bonds, that corresponding to the phosphite being
shorter, is observed (2.15 vs 2.32 Å). A similar observa-
tion has been described for other phosphine-phosphite
derivatives. Also notably, the rigidity of the backbone
caused by the 1,2-phenylene ring imposes a rather small
bite angle (ca. 89°) compared with values obtained with
a phosphine-phosphite with an aliphatic bridge (ca.
97°)6a or with BINAPHOS (ca. 96°).22
Most noteworthy is the structure of the coordinated
phosphine-phosphite. Thus, the phosphite fragment is
1
(19) The magnitude of J PRh has been widely used as a structural
F igu r e 3.
tool in the analysis of rhodium complexes. The magnitude of this
parameter depends markedly on the ligand situated in the trans
position with respect to the phosphine/phosphite ligand, being larger
for a phosphorus trans to chloride than to a carbonyl ligand. See for
instance: Naaktgeboren, A. J .; Nolte, R. J . M.; Drenth, W. J . Am.
Chem. Soc. 1980, 102, 3350.
Beh a vior of Liga n d s 5 in th e Asym m etr ic Hy-
d r ogen a tion of Meth yl Z-r-a ceta m id o-cin n a m a te.
(20) Sanger, A. R. J . Chem. Soc., Dalton Trans. 1977, 121.
(21) Del Zotto, A.; Costella, L.; Mezzetti, A.; Rigo, P. J . Organomet.
Chem. 1991, 414, 109.
(22) Nozaki, K.; Sato, N.; Tonomura, Y.; Yasutomi, M.; Takaya, H.;
Hiyama, T.; Matsubara, T.; Koga, N. J . Am. Chem. Soc. 1997, 119,
12779.
(23) Sembring, S. B.; Colbran, S. B.; Craig, D. C.; Sander, M. L. J .
Chem. Soc., Dalton Trans. 1995, 3731.
(24) This distribution is observed in a great number of bis aryl
phosphines with a chiral backbone, see for instance: Noyori, R.
Asymmetric Catalysis in Organic Synthesis; J ohn Wiley and Sons: New
York, 1994.

Studies of New Chiral Phosphine-Phosphites
Organometallics, Vol. 21, No. 22, 2002 4615
1
Sch em e 5
the phosphine-phosphite ligand in the H NMR spec-
trum as well as broad resonances attributable to coor-
dinated DME. Subsequent addition of 1 equiv of MAC
produced the corresponding complex [Rh(5a )(MAC)]+
(10a ), along with minimal amounts of the starting
material and other unidentified species. However, for
a cleaner generation of 10a it is more convenient to start
with norbornadiene derivative [(NBD)Rh(5a )]BF4.27 Fol-
lowing this sequence of reactions we also prepared the
isopropyl derivative [Rh(5c)(MAC)]+ (10c).
Considering the C1 symmetric nature of ligands 5
there are four possible diastereomers of composition
[Rh(P-OP)(MAC)]+: two with the olefin cis to the
phosphite (structures F and G) and two cis to the
phosphine (H and J ). We have initially investigated the
(a ) Str u ctu r e of [Rh (5)(MAC)]+ Com p lexes. We
have next examined the coordination mode of methyl
Z-R-acetamidocinnamate (MAC) in compounds of com-
position [Rh(P-OP)(MAC)]+. To prepare the enamide
adducts, we first synthesized compounds of composition
[(COD)Rh(P-OP)]BF4 (9). These complexes were simply
obtained by addition of an stoichiometric amount of
ligand 5 to complex [(COD)2Rh]BF4. All compounds 9
showed at room temperature NMR spectra composed
of one set of resonances. To investigate the possible
importance of dynamic processes such as backbone
mobility, we additionally studied compounds 9a and 9d
by 31P{1H} NMR spectroscopy in the 298-180 K tem-
perature range. The former exhibited in the entire
temperature interval the existence of only one com-
pound in solution. Note that this observation parallels
the behavior exhibited by 8c, and evinces the rigidity
of the coordinated phosphine-phosphite. In contrast,
compound 9d displayed a fluxional character, and while
at 298 K it showed two doublet doublets, lowering the
temperature produced a broadening of both resonances,
each of which coalesced at 230 K, and finally split at
lower temperatures to give two doublet doublets indica-
tive of two species in solution. The difference in behavior
between 9a and 9d should be a consequence of the
flexible phosphite fragment existing in the latter com-
pound, and therefore, the dynamic behavior of 9d can
be adscribed to the interconversion between the corre-
sponding atropisomers,25 which in view of the interac-
tion between the phenylene bridge and the phosphite
moiety mentioned above will probably be accompanied
by a backbone conformational change.
coordination mode of the enamide ligand by analyzing
in more detail the structures of compounds 10 by means
of complementary NMR experiments. Thus, the 2D 31P-
1H correlation experiment showed a coupling between
the olefinic proton (assigned in turn by a 2D 13C-1H
correlation experiment) and the phosphorus of the
phosphite, but not with that of the phosphine (Figure
4). This can be compared with the analogue complex
containing diphosphine BisP*, studied by Gridnev and
Imamoto.28 In the latter case, the corresponding proton
couples only with the phosphorus nucleus cis to the
olefin group. In the particular case of compound 10c,
We initially attempted the generation of MAC adducts
by the direct reaction between [(COD)Rh(5a )]BF4 and
the enamide (Scheme 5). Interestingly, no interaction
was observed between this compound and a large excess
of the enamide (ca. 100 molar equiv) after 3 d at room
temperature. As an alternative, we also studied the
hydrogenation of compound 9a . Thus, exposure of a
solution of 9a in 1,2-dimethoxyethane (DME) to 4 atm
of hydrogen produced the apparition of a new compound
characterized in the 31P{1H} NMR by two doublet
3
2
coupling constants J HP ) 4.8 Hz and J HRh ) 2.8 Hz
could be obtained by comparison of the 1H NMR
spectrum with the phosphorus selectively decoupled
experiments. These values are practically identical with
those observed in the mentioned BisP* derivative. In
2
doublets at δ 135.4 (1J RhP ) 348 Hz, J PP ) 90 Hz) and
δ 37.4 (1J RhP ) 189 Hz), along with minor impurities.26
The main species showed the characteristic signals of
(25) Sua´rez, A.; Pizzano, A.; Ferna´ndez, I.; Khiar, N. Tetrahedron:
Asymmetry 2001, 12, 633.
(26) We have noted a rather low reactivity of 9a toward hydrogen
compared with diphosphine-based catalyst precursors. This observation
is in good agreement with the behavior described for other phosphine-
phosphite derivatives (ref 6d) and could be adscribed to the rather low
donating ability of ligands 5.
F igu r e 4.
(27) Borner, A.; Heller, D. Tetrahedron Lett. 2001, 42, 223.
(28) Gridnev, I. D.; Higashi, N.; Asakura, K.; Imamoto, T. J . Am.
Chem. Soc. 2000, 122, 7183.

4616 Organometallics, Vol. 21, No. 22, 2002
Sua´rez et al.
F igu r e 5. 31P{1H} NMR (CD2Cl2, 162 MHz) variable-temperature spectra of 10a .
addition, compounds 10a and 10c showed intense NOE
cross-peaks in the 2D 1H-1H NOESY experiment
between the olefinic proton and the lower field But
group. Overall, these data strongly support structures
F and G, which possess the olefin group coordinated cis
with respect to the phosphite, for compounds 10.29
Multinuclear NMR spectra obtained with compounds
10 indicated the existence in solution of one set of
resonances according to either the presence of one
compound or a fast interchange between several species
in the NMR scale time. To discriminate between these
possibilities we next examined compounds 10 by vari-
able-temperature 31P{1H} NMR. Thus, compound 10a
showed one sharp set of signals from room temperature
to 240 K (Figure 5). If additionally cooled, the phosphite
signal broadened, coalesced at 210 K, and finally split
into two doublet doublets with approximately 2:1 rela-
tive intensity. In turn, the phosphine signal only
broadened slightly and the presence of the minor species
was only detected at 180 K, as a doublet doublet
overlapped with the resonance corresponding to the
major species. In the case of 10c coalescence was
observed at 200 K, while at the lowest temperature
investigated (180 K) it was not reached in the slow
exchange regime. As in the former case, broadening was
much higher in the phosphite signal. Taking into
consideration the similarity in chemical shift and cou-
pling constants between the two species in the low
exchange spectrum, it is reasonable to assume that both
compounds have the MAC molecule in the same coor-
dination mode (i.e. olefin cis to the phosphite). Then,
two distinct possibilities can be considered for the
dynamic behavior observed: (i) exchange between F and
G diastereomers differing in the olefin coordinated face
and (ii) interchange between conformers which differ
in their backbone structure. An estimation of ca. 9 kcal
mol-1 for the activation barrier of the dynamic process
observed in 10a coincides with that obtained for the
intramolecular olefin face exchange in similar com-
pounds.28,30 From these data, as well from the confor-
mational preference observed in 8c and 9a , we incline
toward the first possibility to explain the dynamic
behavior exhibited by compounds 10.
(b) P er for m a n ce of Liga n d s 5 in th e Asym m etr ic
Hyd r ogen a tion of MAC. Taking advantage of our
ligand design, which enables us to choose the position
of the stereogenic element of the ligand, we next
analyzed the performance of ligands 5 in MAC asym-
metric hydrogenation (Table 3). From the coordination
mode described above, a higher influence of the phos-
phite fragment on the enantioselectivity of the process
can be expected. Indeed, the configuration of the bi-
phenyl moiety controls the enantiomer produced. Thus,
catalyst precursor 9a with S configuration at the
phosphite produced the R enantiomer of the product,
while 9f produced the opposite, both with excellent ee
values (99%). These results are in good accord with a
recent revisitation of the quadrant rule, since an S
(29) Recently, Claver and van Leeuwen have proposed the same
coordination mode for a derivative of a phosphine-phosphite: ref 6c.
(30) Kadyrov, R.; Freier, T.; Heller, D.; Michalik, M.; Selke, R. Chem.
Commun. 1995, 1745.

Studies of New Chiral Phosphine-Phosphites
Organometallics, Vol. 21, No. 22, 2002 4617
Ta ble 3. Resu lts of th e Asym m etr ic
Hyd r ogen a tion sa
entry
cat. precursor
conversion
% ee (conf)
1
2
3
4
5
6
9a
9b
9c
9d
9e
9f
100
100
20
45
90
99.5 (R)
56.1 (R)
90.6 (R)
22.0 (S)
70.3 (R)
99.0 (S)
100
a
Reactions were carried out at room temperature under an
initial H2 pressure of 4 atm and 0.2 M dichloromethane solutions
of substrate, using the appropriate catalyst precursor for 16 h
(entries 1-3) or 24 h (entries 4-6). Conversion was determined
by 1H NMR and ee by chiral GC.
phosphite would be expected to exert the higher steric
impediment in the lower right quadrant.28 Then, the
result obtained with catalyst precursor 9d can be
ascribed to the coexistence of species with the two
configurations of the phosphite fragment,31 since the
conformational flexibility of this group has been ob-
served in the catalyst precursor. Interestingly, this
dynamic process can be responsible for the lower activity
displayed by catalyst originating from 9d . Thus, taken
as a reference a ligand with S configuration, a change
of configuration of the phosphite fragment in species F
would lead to a less reactive F ′ derivative (Figure 6).32
Otherwise the low activity displayed by the isopropyl
derivative 9c can be simply attributed to the steric
encumbrance imposed by the PPri group.33
2
The achievement of a high enantioselectivity needs,
in addition, an appropriate phosphine group. Deriva-
tives 9a , 9c, and 9f achieved good to excellent enantio-
selectivities (90-99% ee), while compounds 9b and 9e
produced lower optical purities (56 and 70% ee, respec-
tively). Although at the actual level of understanding
of this catalytic system it is difficult to give a definite
explanation for this secondary ligand influence, a
reasonable interpretation can be suggested upon obser-
vation of the structures of compounds 10. From com-
parison of the picture and the hydrogenation results, it
can be concluded that catalysts with a Ph or Pri group
in the R1 position lead to the more enantioselective
catalysts, while catalyst precursors with a smaller Me
group produced lower ee values. Therefore, it is reason-
able to assume that hydrogen addition to diastereomer
G should be made more difficult by a bulky R1 substitu-
ent and consequently the production of the minor S
enantiomer, thus enhancing the enantioselectivity (Fig-
ure 7).34 Note that the more enantioselective catalysts
aproximately reproduce the arrangement depicted by
C2 symmetric diphosphines, with alternate quadrants
offering the larger steric impediment. Therefore, in the
case of our ligands, although an excellent enantioselec-
F igu r e 6.
F igu r e 7.
tivity can be obtained with only a chiral phosphite
group, its influence is also extended to the phosphine
group, hence, the high enantioselectivity is a conse-
quence of a synergistic action between both coordinating
fragments.
(31) Reetz, M. T.; Neugebaer, T. Angew. Chem., Int. Ed. Engl. 1999,
38, 179.
(32) The structure of dihydrides K and K′ has been proponed by
similarity with those described in ref 28.
(33) Burk, M. J .; Kalberg, C. S.; Pizzano, A. J . Am. Chem. Soc. 1998,
120, 4345.
(34) In Figure 7 the positions of R1 and R2 have been assumed for
S phosphite configuration. Then, the figure would be applicable for
an enantiomer of 5f, in which the Ph group would occupy the position
denoted by R1.
Con clu sion s
A convenient synthesis of new modularly designed
chiral phosphine-phosphites, which differ in the nature

4618 Organometallics, Vol. 21, No. 22, 2002
Sua´rez et al.
in CH2Cl2 and precipitated with Et2O, yielding a white solid
of the phosphine group and in the position of the
stereogenic element, has been described. An investiga-
tion of the electronic properties of the two phosphorus
functionalities of compounds 5 indicates substantial
differences between them. Thus, studies of 1J PSe values
of the corresponding diselenides in connection with IR
analysis of compounds Rh(Cl)(CO)(P-OP) are in ac-
cordance with a rather reduced σ-donor ability and a
strong π-aceptor character of the phosphite fragment.
Moreover, this functionality experiences a variable back-
donation from the metal, thus compensating the differ-
ent basicity of the phosphine fragment along the series
of ligands. Overall, these phosphine-phosphites are
considerably less donating than diphosphines.
1
3
(3.1 g, 80%). H NMR (CDCl3, 300 MHz): δ 1.20 (dd, J HP
)
3
3
19.4 Hz, J HH ) 7.2 Hz, 6H, 2 CHMe2), 1.35 (dd, J HP ) 19.8
Hz, 3J HH ) 7.1 Hz, 6H, 2 CHMe2), 3.01 (m, 2H, 2 CHMe2), 6.87
(tdd, 3J HH ) 7.5 Hz, 4J HP ) 2.3 Hz, 4J HH ) 1.0 Hz, 1H, H arom),
1
3
7.36 (dt, J HP ) 468 Hz, J HH ) 7.0 Hz, 1H, P-H), 7.41 (m,
1H, H arom), 7.58 (ddd, 3J HP ) 14.8 Hz, 3J HH ) 7.6 Hz, 4J HH
)
1.3 Hz, 1H, H arom), 7.73 (m, 1H, H arom), 10.39 (s, 1H, O-H).
31P{1H} NMR (CDCl3, 121 MHz): δ 33.1. 13C{1H} NMR (CDCl3,
75 MHz): δ 17.7 (d, J CP ) 2 Hz, 2 CHMe2), 18.1 (2 CHMe2),
21.2 (d, J CP ) 44 Hz, 2 CHMe2), 98.9 (d, J CP ) 79 Hz, Cq arom),
118.1 (d, J CP ) 7 Hz, CH arom), 121.0 (d, J CP ) 12 Hz, CH
arom), 135.9 (d, J CP ) 7 Hz, CH arom), 137.3 (CH arom), 161.7
(Cq arom). Anal. Calcd for C12H20POBr: C, 49.5; H, 6.9.
Found: C, 49.4; H, 6.7.
P r ep a r a tion of P h en ol-P h osp h in es 3. Compounds 3
were synthesized by deprotonation of phosphonium bromides
2. Below the preparation of compound 3c is described.
Diisop r op yl-1-(2-h yd r oxyp h en yl)p h osp h in e (3c). Phos-
phonium salt 2c (0.69 g, 2.4 mmol) was suspended in toluene
(25 mL) and NEt3 (0.8 mL, 5.5 mmol) added. The resulting
mixture was stirred vigorously for 2 h and filtered. Evapora-
tion of the solvent produced a colorless oil (0.50 g, 99%). IR
(Nujol mull, cm-1): 3380 (s, υ(OH)). 1H NMR (CDCl3, 500
An investigation of the structure of the coordinated
phosphine-phosphite indicates a suitable scaffold for
asymmetric catalytic hydrogenation due to the arrange-
ment of the phosphite fragment. In addition, interaction
between the phosphite and the rather rigid backbone
led to a preferred conformation of the latter.
Due to the substantial electronic differences between
the coordinating functionalities, MAC prefers to coor-
dinate its olefinic bond cis to the phosphite, as evidenced
by detailed NMR studies. The proximity between the
phosphite and the coordinated olefin causes the former
to direct the enantioselectivity of the process. However,
the phosphine group exerts a secondary action and, for
the attainment of a product of high optical purity, it is
necessary to use an adequate phosphine group. Inter-
estingly, this double effect can be accomplished with
ligands which are only chiral at the phosphite, thus
simplifying the catalyst design, as the influence of the
phosphite is extended to the phosphine group.
3
3
MHz): δ 0.92 (dd, J HP ) 12.4 Hz, J HH ) 6.9 Hz, 6H, 2
3
3
CHMe2), 1.11 (dd, J HP ) 16.6 Hz, J HH ) 7.0 Hz, 6H, 2
CHMe2), 2.14 (hd, 3J HH ) 7.0 Hz, 2J HP ) 3.7 Hz, 2H, 2 CHMe2),
6.87-6.92 (m, 2H, 2 H arom), 6.98 (br s, 1H, O-H), 7.23-
7.26 (m, 2H, 2 H arom). 31P{1H} NMR (CDCl3, 202 MHz): δ
-25.3. 13C{1H} NMR (CDCl3, 126 MHz): δ 18.7 (d, J CP ) 7
Hz, 2 CHMe2), 20.0 (d, J CP ) 18 Hz, 2 CHMe2), 22.9 (d, J CP
)
6 Hz, 2 CHMe2), 114.9 (CH arom), 118.1 (d, J CP ) 8 Hz, Cq
arom), 119.9 (CH arom), 131.0 (CH arom), 132.6 (CH arom),
161.1 (d, J CP ) 19 Hz, Cq arom). HRMS (EI): m/z 211.1247,
[M + H]+ (exact mass calculated for C12H20OP: 211.1252).
(S )-3,3′-D i-t er t -b u t y l-5,5′,6,6′-t e t r a m e t h y l-2,2′-b is -
p h en oxyp h osp h or u s Ch lor id e ((S)-4a ). (S)-3,3′-di-tert-
butyl-5,5′,6,6′-tetramethyl-2,2′-bisphenol (4.6 g, 13.0 mmol)
was azeotropically dried with toluene (2 × 30 mL), dissolved
in toluene (100 mL), and added dropwise over a mixture of
PCl3 (1.1 mL, 13.0 mmol) and pyridine (2.3 mL, 28.6 mmol)
dissolved in 60 mL of toluene. The resulting suspension was
stirred for 14 h, filtered, and evaporated to dryness to yield
(S)-4a as a white solid (5.0 g, 92%). [R]20D 508 (c 1.0, THF). 1H
NMR (CD2Cl2, 300 MHz): δ 1.44 (s, 9H, CMe3), 1.45 (s, 9H,
CMe3), 1.82 (s, 3H, Ar-Me), 1.88 (s, 3H, Ar-Me), 2.29 (s, 3H,
Ar-Me), 2.30 (s, 3H, Ar-Me), 7.24 (s, 1H, H arom), 7.27 (s, 1H,
H arom). 31P{1H} NMR (CD2Cl2, 121 MHz): δ 163.1. 13C{1H}
NMR (CD2Cl2, 75 MHz): δ 16.5 (Ar-Me), 16.8 (Ar-Me), 20.5 (2
Ar-Me), 31.2 (d, J CP ) 5 Hz, CMe3), 32.3 (CMe3), 34.9 (CMe3),
35.3 (CMe3), 128.7 (CH arom), 129.3 (CH arom), 130.6 (Cq
arom), 131.9 (d, J CP ) 6 Hz, Cq arom), 133.5 (Cq arom), 134.5
(Cq arom), 135.0 (Cq arom), 136.0 (Cq arom), 137.9 (Cq arom),
138.7 (Cq arom), 144.0 (Cq arom), 145.6 (Cq arom). HRMS
(EI): m/z 418.1829, [M]+ (exact mass calculated for C24H32O2-
PCl: 418.1828).
Exp er im en ta l Section
Gen er a l P r oced u r es. All reactions and manipulations
were performed under nitrogen or argon, either in a Braun
Labmaster 100 glovebox or with standard Schlenk-type tech-
niques. All solvents were distilled under nitrogen with the
following desiccants: sodium-benzophenone-ketyl for benzene,
diethyl ether (Et2O,) and tetrahydrofuran (THF); sodium for
petroleum ether and toluene; CaH2 for dichloromethane (CH2-
Cl2); and NaOMe for methanol (MeOH). NMR spectra were
obtained on Bruker DPX-300, DRX-400, or DRX-500 spec-
trometers. 31P{1H} NMR shifts were referenced to external
85% H3PO4, while 13C{1H} and 1H shifts were referenced to
the residual signals of deuterated solvents. All data are
reported in ppm downfield from Me4Si. GC analyses were
performed by using a Hewlett-Packard Model HP 6890 chro-
matograph. HRMS data were obtained on a J EOL J MS-SX
102A mass spectrometer. Elemental analyses were run by the
Analytical Service of the Instituto de Investigaciones Qu´ımicas.
Optical rotations were measured on a Perkin-Elmer Model 341
polarimeter.
Syn th esis of P h osp h in e-p h osp h ites (5). Ligands 5 were
prepared by condensation between a phenol phosphine (3) and
an equimolar amount of the appropriate chlorophosphite (4)
in the presence of NEt3. A representative procedure is de-
scribed for 5c.
Syn th esis of P h osp h on iu m Br om id es (2). These com-
pounds were prepared by demethylation of anisyl phosphines
1, as described below for derivative 2c.
Diisop r op yl-1-(2-h yd r oxyp h en yl)p h osp h on iu m Br o-
m id e (2c). To a stirred solution of o-anisyldiisopropylphos-
phine (3.0 g, 13.4 mmol) in CH2Cl2 (100 mL) cooled at -78 °C
was added BBr3 (3.0 mL, 31.7 mmol) via syringe. The resulting
solution was left to warm to room temperature, stirred for 12
h, and carefully evaporated to dryness. A portion of toluene
was added to the residue and then evaporated to secure the
evaporation of volatiles. The resulting solid was treated with
MeOH (20 mL) at 0 °C and stirred for 48 h at room temper-
ature. Volatiles were evaporated and the resulting oil dissolved
Com p ou n d 5c. Phosphonium bromide 2c (0.69 g, 2.3 mmol)
was suspended in toluene (25 mL) and NEt3 (0.8 mL, 5.7 mmol)
was added. The resulting suspension was stirred for 2 h and
filtered. The solution obtained was evaporated to dryness and
subjected to further drying by addition of portions of toluene
(2 × 20 mL). The resulting oil was dissolved in toluene (50
mL) and the solution added dropwise over a mixture of (S)-4a
(0.91 g, 2.3 mmol) and NEt3 (0.4 mL, 2.8 mmol) dissolved in
toluene (60 mL). The reaction was stirred for 48 h and filtered
and volatiles were removed, resulting a colorless oil that was

Studies of New Chiral Phosphine-Phosphites
Organometallics, Vol. 21, No. 22, 2002 4619
redissolved in Et2O and passed through a short pad of neutral
alumina, yielding compound 5c as a white foamy solid (1.0 g,
mixture was stirred for 30 min, and then CO was bubbled
through the mixture for 15 min, changing the color of the
solution from orange to yellow. The mixture was evaporated
to dryness and the resulting solid was recrystallized from a
mixture of hexanes/Et2O (3:1) giving 8c as a yellow powder
(0.086 g, 63%). Attempts to obtain satisfactory elemental
analysis for this compound have proven fruitless. IR (Nujol
20
1
71%). [R]D 347 (c 1.0, THF). H NMR (CD2Cl2, 300 MHz): δ
3
3
0.79 (dd, J HP ) 11.2 Hz, J HH ) 7.0 Hz, 3H, CHMe2), 0.83
(dd, 3J HP ) 13.0 Hz, 3J HH ) 7.0 Hz, 3H, CHMe2), 1.02 (dd, 3J HP
3
3
) 10.8 Hz, J HH ) 7.0 Hz, 3H, CHMe2), 1.07 (dd, J HP ) 10.9
3
Hz, J HH ) 7.0 Hz, 3H, CHMe2), 1.31 (s, 9H, CMe3), 1.40 (s,
1
9H, CMe3), 1.83 (s, 3H, Ar-Me), 1.88 (s, 3H, Ar-Me), 1.98 (h,
mull, cm-1): 2042 (s, υ(CO)). H NMR (CD2Cl2, 400 MHz): δ
3J HH ) 7.0 Hz, 1H, CHMe2), 1.98 (hd, J HH ) 7.0 Hz, J HP
)
1.16 (dd, J HP ) 14.3 Hz, J HH ) 7.0 Hz, 3H, CHMe2), 1.30
3
2
3
3
1.6 Hz, 1H, CHMe2), 2.27 (s, 3H, Ar-Me), 2.32 (s, 3H, Ar-Me),
6.60 (m, 1H, H arom), 7.05 (dt, J HH ) 7.3 Hz, J HH ) 0.9 Hz,
1H, H arom), 7.14 (m, 1H, H arom), 7.16 (s, 1H, H arom), 7.23
(dd, 3J HP ) 16.8 Hz, 3J HH ) 7.2 Hz, 3H, CHMe2), 1.34 (dd, 3J HP
) 16.3 Hz, J HH ) 7.2 Hz, 3H, CHMe2), 1.37 (s, 9H, CMe3),
1.46 (s, 9H, CMe3), 1.54 (dd, J HP ) 15.8 Hz, J HH ) 7.1 Hz,
3H, CHMe2), 1.83 (s, 6H, 2 Ar-Me), 2.26 (s, 3H, Ar-Me), 2.31
3
4
3
3
3
3
4
4
(s, 1H, H arom), 7.42 (ddd, J HH ) 7.1 Hz, J HP ) 4.7 Hz, J HH
) 1.8 Hz, 1H, H arom). 31P{1H} NMR (CD2Cl2, 121 MHz): δ
-1.0 (br d, P-C), 129.3 (d, 4J PP ) 32 Hz, P-O). 13C{1H} NMR
(CD2Cl2, 126 MHz): δ 16.4 (Ar-Me), 16.7 (Ar-Me), 19.7 (d, J CP
) 9 Hz, CHMe2), 20.4 (2 Ar-Me), 20.5 (d, J CP ) 21 Hz, CHMe2),
20.6 (d, J CP ) 14 Hz, CHMe2), 20.7 (d, J CP ) 19 Hz, CHMe2),
23.3 (d, J CP ) 12 Hz, CHMe2), 24.5 (d, J CP ) 14 Hz, CHMe2),
31.2 (d, J CP ) 5 Hz, CMe3), 31.5 (CMe3), 34.8 (CMe3), 35.0
(CMe3), 121.3 (d, J CP ) 13 Hz, CH arom), 123.5 (d, J CP ) 3
Hz, CH arom), 128.1 (CH arom), 128.3 (Cq arom), 128.5 (CH
arom), 129.9 (CH arom), 131.1 (Cq arom), 132.3 (2 Cq arom),
133.1 (Cq arom), 134.7 (Cq arom), 135.3 (d, J CP ) 9 Hz, CH
arom), 135.4 (Cq arom), 137.8 (Cq arom), 138.6 (Cq arom), 145.0
(Cq arom), 145.4 (d, J CP ) 6 Hz, Cq arom), 156.1 (d, J CP ) 14
Hz, Cq arom). HRMS (EI): m/z 592.3240, [M]+ (exact mass
calculated for C36H50O3P2: 592.3235).
Diselen id es 7. Derivatives 7 were prepared by heating a
solution in toluene of the corresponding phosphine-phosphite
in contact with elemental selenium as described below for 7c.
Com p ou n d 7c. Over a solution of 5c (0.020 g, 0.03 mmol)
in toluene (0.5 mL) was added elemental selenium (0.011 g,
0.14 mmol) and the mixture was heated at 100 °C for 8 days.
The resulting mixture was filtered through Celite and evapo-
rated to dryness, resulting an oil that was additionally dried
by adding a portion of benzene. Evaporation of volatiles
3
2
(s, 3H, Ar-Me), 2.64 (dh, J HH ) 7.0 Hz, J HP ) 7.0 Hz, 1H,
CHMe2), 3.06 (dh, 2J HP ) 12.8 Hz, 3J HH ) 7.0 Hz, 1H, CHMe2),
3
4
4
6.74 (ddt, J HH ) 7.1 Hz, J HP ) 4.3 Hz, J HH ) 1.0 Hz, 1H, H
arom), 7.23 (s, 1H, H arom), 7.25 (m, 1H, H arom), 7.27 (s,
3
1H, H arom), 7.41 (t, J HH ) 7.4 Hz, 1H, H arom), 7.56 (ddd,
3J HP ) 7.9 Hz, J HH )6.5 Hz, J HH )1.6 Hz, 1H, H arom). 31P-
3
4
{1H} NMR (CD2Cl2, 162 MHz): δ 16.2 (dd, J PRh ) 122 Hz,
1
P-C), 133.5 (dd, 1J PRh ) 278 Hz, 2J PP ) 62 Hz, P-O). 13C{1H}
NMR (CD2Cl2, 126 MHz): δ 16.5 (Ar-Me), 16.6 (Ar-Me), 18.3
(CHMe2), 19.1 (d, J CP ) 3 Hz, CHMe2), 19.5 (CHMe2), 20.2 (Ar-
Me), 20.4 (Ar-Me), 20.6 (d, J CP ) 4 Hz, CHMe2), 24.1 (d, J CP
)
24 Hz, CHMe2), 27.3 (d, J CP ) 23 Hz, CHMe2), 31.6 (CMe3),
32.9 (CMe3), 35.1 (CMe3), 35.5 (CMe3), 115.6 (dd, J CP ) 35, 10
Hz, Cq arom), 122.7 (CH arom), 125.0 (d, J CP ) 4 Hz, CH arom),
128.8 (CH arom), 129.5 (CH arom), 130.0 (d, J CP ) 2 Hz, Cq
arom), 130.7 (Cq arom), 131.9 (CH arom), 132.8 (CH arom),
133.3 (Cq arom), 134.0 (Cq arom), 135.1 (Cq arom), 135.9 (Cq
arom), 137.6 (Cq arom), 138.2 (d, J CP ) 4 Hz, Cq arom), 143.6
(d, J CP ) 7 Hz, Cq arom), 146.3 (d, J CP ) 14 Hz, Cq arom),
156.3 (dd, J CP ) 10, 3 Hz, Cq arom), 182.9 (ddd, J CP ) 110, 13
Hz, J CRh ) 63 Hz, CO).
[(COD)Rh (P -OP )]BF 4 Com p lexes (9). Compounds 9
were prepared by reaction of [(COD)2Rh]BF4 with a stoichio-
metric amount of the appropriate phosphine-phosphite as
described for 9c.
20
rendered compound 7c quantitatively as a white solid. [R]D
162 (c 1.0, THF). 1H NMR (CDCl3, 500 MHz): δ 0.59 (dd, 3J HP
[(COD)R h (5c)]BF 4 (9c). Over
a stirred solution of
3
3
) 19.2 Hz, J HH ) 6.9 Hz, 3H, CHMe2), 0.74 (dd, J HP ) 19.3
[(COD)2Rh]BF4 (0.100 g, 0.25 mmol) in CH2Cl2 (5 mL) was
added slowly 5c (0.153 g, 0.26 mmol) dissolved in CH2Cl2 (10
mL). The resulting orange solution was stirred for 20 min, and
then concentrated to a fourth of the volume. The mixture was
filtered and precipitated with Et2O (45 mL). The resulting solid
was washed with Et2O (2 × 10 mL) and recrystallized from a
mixture of CH2Cl2/Et2O (1:3), yielding 9c as an orange-yellow
Hz, 3J HH ) 6.9 Hz, 3H, CHMe2), 1.06 (dd, 3J HP ) 18.5 Hz, 3J HH
3
3
) 6.7 Hz, 3H, CHMe2), 1.20 (dd, J HP ) 18.7 Hz, J HH ) 6.7
Hz, 3H, CHMe2), 1.31 (s, 9H, CMe3), 1.43 (s, 9H, CMe3), 1.79
(s, 3H, Ar-Me), 1.90 (s, 3H, Ar-Me), 2.26 (s, 3H, Ar-Me), 2.32
2
3
(s, 3H, Ar-Me), 2.55 (dh, J HP ) 6.5 Hz, J HH ) 6.5 Hz, 1H,
2
3
CHMe2), 3.11 (dh, J HP ) 6.7 Hz, J HH ) 6.7 Hz, 1H, CHMe2),
6.84 (br s, 1H, H arom), 7.18 (s, 1H, H arom), 7.21 (m, 2H, 2
H arom), 7.27 (s, 1H, H arom), 8.47 (m, 1H, H arom). 31P{1H}
NMR (CDCl3, 202 MHz): δ 52.7 (1J PSe ) 1053 Hz, P-O), 75.7
(1J PSe ) 719 Hz, P-C). 13C{1H} NMR (CDCl3, 126 MHz): δ
16.5 (Ar-Me), 16.6 (Ar-Me), 18.4 (CHMe2), 18.6 (CHMe2), 18.8
(CHMe2), 19.0 (CHMe2), 20.4 (Ar-Me), 20.5 (Ar-Me), 26.7 (d,
J CP ) 43 Hz, CHMe2), 27.7 (d, J CP ) 43 Hz, CHMe2), 31.1
(CMe3), 32.9 (CMe3), 34.8 (CMe3), 35.4 (CMe3), 119.7 (dd, J CP
) 56, 9 Hz, Cq arom), 119.9 (CH arom), 124.9 (d, J CP ) 11 Hz,
CH arom), 129.0 (Cq arom), 129.1 (CH arom), 129.5 (Cq arom),
130.7 (CH arom), 132.5 (CH arom), 134.0 (Cq arom), 134.6 (Cq
arom), 135.6 (2 Cq arom), 137.6 (d, J CP ) 4 Hz, Cq arom), 138.2
(d, J CP ) 5 Hz, Cq arom), 139.9 (d, J CP ) 10 Hz, CH arom),
143.1 (d, J CP ) 9 Hz, Cq arom), 145.8 (d, J CP ) 16 Hz, Cq arom),
151.3 (d, J CP ) 12 Hz, Cq arom). HRMS (FAB): m/z 753.1646,
1
solid (0.152 g, 68%). H NMR (CD2Cl2, 500 MHz): δ 0.99 (dd,
3J HP ) 13.4 Hz, 3J HH ) 6.8 Hz, 3H, CHMe2), 1.24 (s, 9H, CMe3),
3
3
1.30 (dd, J HP ) 17.3 Hz, J HH ) 6.8 Hz, 3H, CHMe2), 1.38
3
3
(dd, J HP ) 14.6 Hz, J HH ) 6.7 Hz, 3H, CHMe2), 1.50 (s, 9H,
3
3
CMe3), 1.58 (dd, J HP ) 18.2 Hz, J HH ) 7.1 Hz, 3H, CHMe2),
1.76 (s, 3H, Ar-Me), 1.84 (s, 3H, Ar-Me), 2.11-2.65 (m, 9H, 4
CH2 COD and CHMe2), 2.30 (s, 3H, Ar-Me), 2.32 (s, 3H,
3
Ar-Me), 3.08 (h, J HH ) 7.1 Hz, 1H, CHMe2), 3.93 (br m, 1H,
dCH COD), 5.65 (br m, 1H, dCH COD), 5.90 (br m, 1H, dCH
COD), 6.19 (br m, 1H, dCH COD), 6.70 (m, 1H, H arom), 7.24
(s, 1H, H arom), 7.33 (s, 1H, H arom), 7.38 (m, 1H, H arom),
7.47 (m, 2H, 2 H arom). 31P{1H} NMR (CD2Cl2, 202 MHz): δ
1
1
14.9 (dd, J PRh ) 137 Hz, P-C), 129.7 (dd, J PRh ) 265 Hz,
2J PP ) 45 Hz, P-O). 13C{1H} NMR (CD2Cl2, 126 MHz): δ 16.4
(Ar-Me), 16.6 (Ar-Me), 17.4 (d, J CP ) 2 Hz, CHMe2), 17.6 (d,
J CP ) 5 Hz, CHMe2), 19.7 (d, J CP ) 4 Hz, CHMe2), 20.2 (Ar-
[M
753.1644).
+
H]+ (exact mass calculated for C36H51O3P2Se2:
Me), 20.4 (Ar-Me), 22.5 (d, J CP ) 4 Hz, CHMe2), 23.9 (d, J CP
)
Rh Cl(CO)(P -OP ) Com p lexes (8). Chlorocarbonyls were
prepared by reacting [RhCl(C2H4)2]2 with the appropriate
phosphine-phosphite followed by treatment of the resulting
mixture with carbon monoxide. Below is described a repre-
sentative preparation.
Rh Cl(CO)(5c) (8c). A solution of 5c (0.107 g, 0.18 mmol)
in THF (5 mL) was added slowly over a stirred solution of [Rh-
(µ-Cl)(C2H4)2]2 (0.035 g, 0.09 mmol) in THF (5 mL). The
22 Hz, CHMe2), 28.8 (CH2 COD), 29.2 (CH2 COD), 29.9 (d, J CP
) 23 Hz, CHMe2), 31.8 (CMe3 and CH2 COD), 32.1 (CH2 COD),
32.3 (CMe3), 35.0 (CMe3), 35.2 (CMe3), 91.3 (dd, J CP ) 9 Hz,
J CRh ) 7 Hz, dCH COD), 103.1 (t, J CP ) 7 Hz, J CRh ) 7 Hz,
dCH COD), 106.6 (dd, J CP ) 11 Hz, J CRh ) 5 Hz, dCH COD),
108.6 (dd, J CP ) 14 Hz, J CRh) 5 Hz, dCH COD), 114.5 (dd,
J CP ) 36, 10 Hz, Cq arom), 123.6 (CH arom), 126.4 (d, J CP ) 3
Hz, CH arom), 129.0 (CH arom), 129.2 (Cq arom), 129.6 (CH

4620 Organometallics, Vol. 21, No. 22, 2002
Sua´rez et al.
Ta ble 4. Cr ysta l Da ta a n d Da ta Collection
P a r a m eter s for 8a
arom and Cq arom), 131.7 (CH arom), 133.6 (CH arom), 134.5
(Cq arom), 134.7 (Cq arom), 135.7 (Cq arom), 136.6 (Cq arom),
137.5 (2 Cq arom), 144.0 (d, J CP ) 7 Hz, Cq arom), 144.7 (d,
J CP ) 14 Hz, Cq arom), 155.1 (dd, J CP ) 8, 5 Hz, Cq arom).
Anal. Calcd for C44H62O3BF4P2Rh‚0.5CH2Cl2: C, 57.3; H, 6.8.
Found: C, 57.4; H, 6.7.
[Rh (5a )(MAC)]BF 4 (10a ). 10a was obtained as an orange
solid as described for 10c. 1H NMR (CD2Cl2, 500 MHz): δ 1.25
(s, 9H, CMe3), 1.42 (s, 9H, CMe3), 1.82 (s, 3H, Ar-Me), 1.87 (s,
3H, Ar-Me), 2.06 (s, 3H, C(O)Me), 2.27 (s, 3H, Ar-Me), 2.37 (s,
empirical formula
fw
C
47H56ClO5P2Rh
901.22
temp
299(2) K
wavelength (Mo KR)
0.71073 Å
cryst syst
orthorhombic
P2(1)2(1)2(1)
a ) 10.0304(2) Å
b ) 13.3552(3) Å
c ) 34.9411(9) Å
R ) â ) γ ) 90°
4680.6(2) Å3
4
space group
unit cell dimens
3H, Ar-Me), 3.05 (s, 3H, OMe), 6.55 (dd, 3J HH ) 7.8 Hz, 4J HP
)
3
3
4
5.6 Hz, 1H, Hd), 6.72 (ddd, J HP ) 9.4 Hz, J HH ) 8.0 Hz, J HH
vol
Z
3
3
) 1.4 Hz, 1H, Ha), 6.81 (dd, J HP ) 12.8 Hz, J HH ) 7.5 Hz,
2H, 2 Ho′), 7.10 (m, 3H, 2 Hm′ and CdCH), 7.14 (m, 1H, Hb),
7.19 (s, 1H, H arom), 7.25 (m, 4H, 2 Hm and 2 Ho′′), 7.32 (s,
1H, H arom), 7.40 (m, 1H, Hp′), 7.53 (m, 6H, Hc, Hp, 2 Hï and
2 Hm′′), 7.64 (m, 1H, Hp′′), 8.90 (br s, 1H, NH). 31P{1H} NMR
calcd density
abs coeff
F(000)
1.279 g cm-3
0.532 mm-1
1880
cryst size
θ range
no. of rflns (collected/unique)
R(int)
0.22 × 0.35 × 0.35 mm3
1.17 < θ < 23.29°
24482/6735
0.0485
1
(CD2Cl2, 121 MHz): δ 6.0 (dd, J PRh ) 152 Hz, P-C), 134.3
1
2
(dd, J PRh ) 263 Hz, J PP ) 69 Hz, P-O). 13C{1H} NMR (CD2-
Cl2, 75 MHz): δ 16.7 (2 Ar-Me), 20.2 (d, J CRh ) 5 Hz, C(O)Me),
20.4 (Ar-Me), 20.5 (Ar-Me), 31.3 (CMe3), 31.7 (CMe3), 34.8
(CMe3), 35.2 (CMe3), 52.4 (OMe), 85.8 (d, J CP ) 11 Hz, Cd
CH), 87.4 (dd, J CP ) 21 Hz, J CRh ) 7 Hz, CdCH), 115.7 (dd,
J CP ) 55, 8 Hz, Cq arom), 122.3 (CH arom), 126.1 (d, J CP ) 7
Hz, CH arom), 127.5 (d, J CP ) 54 Hz, Cq), 127.9 (d, J CP ) 49
Hz, Cq), 128.7 (2 CH arom), 128.8 (2 CH arom), 129.0 (CH
arom), 129.3 (2 CH arom), 129.4 (2 CH arom), 129.4 (2 CH
arom), 129.5 (2 CH arom), 129.8 (Cq arom), 129.9 (2 CH arom),
131.4 (Cq arom), 132.0 (Cq arom), 133.2 (Cq arom), 133.7 (CH
arom), 133.8 (CH arom), 133.9 (2 CH arom), 134.2 (Cq arom),
134.3 (Cq arom), 136.1 (Cq arom), 136.3 (Cq arom), 137.2 (Cq
arom), 145.2 (d, J CP ) 8 Hz, Cq arom), 146.1 (d, J CP ) 13 Hz,
abs corr
empirical
no. of data/restraints/params
6735/0/517
1.035
GooF on F2
R[I > 2σ(I)]
R1 ) 0.0336
wR2 ) 0.0738
0.392/-0.310 e Å-3
largest diff. peak and hole
) 4 Hz, CHMe2), 17.5 (d, J CP ) 2 Hz, CHMe2), 19.3 (d, J CP
)
3 Hz, CHMe2), 19.4 (d, J CP ) 3 Hz, CHMe2), 20.1 (d, J CRh ) 5
Hz, C(O)Me), 20.4 (Ar-Me), 20.5 (Ar-Me), 23.5 (d, J CP ) 24 Hz,
CHMe2), 29.4 (d, J CP ) 24 Hz, CHMe2), 31.7 (CMe3), 32.0
(CMe3), 35.0 (CMe3), 35.3 (CMe3), 52.3 (OMe), 82.7 (d, J CP
)
13 Hz, CdCH), 84.8 (dd, J CP ) 23 Hz, J CRh ) 8 Hz, CdCH),
114.1 (dd, J CP ) 38, 6 Hz, Cq arom), 122.5 (CH arom), 125.8
(d, J CP ) 5 Hz, CH arom), 128.6 (CH arom), 129.0 (CH arom),
129.5 (CH arom), 129.5 (2 CH arom), 129.6 (2 CH arom), 130.0
(Cq arom), 131.2 (CH arom), 132.0 (Cq arom), 133.6 (CH arom),
133.9 (Cq arom), 134.2 (Cq arom), 136.0 (Cq arom), 136.4 (d,
J CP ) 4 Hz, Cq arom), 136.6 (Cq arom), 137.2 (Cq arom), 137.6
(d, J CP ) 2 Hz, Cq arom), 145.5 (d, J CP ) 9 Hz, Cq arom), 146.0
(d, J CP ) 14 Hz, Cq arom), 155.8 (t, J CP ) 8 Hz, Cq arom), 166.2
(d, J CP ) 5 Hz, C(O)Me), 185.8 (dd, J CRh ) 7, J CP ) 5 Hz, CO2-
Me).
Cq arom), 155.3 (dd, J CP ) 10, 2 Hz, Cq arom), 165.8 (d, J CP
5 Hz, C(O)Me), 184.6 (br s, CO2Me).
)
[Rh (5c)(MAC)]BF 4 (10c). A solution of [(COD)Rh(5c)]BF4
(0.045 g, 0.05 mmol) and norbornadiene (0.50 g, 5.4 mmol) in
CH2Cl2 (3 mL) was stirred overnight. The resulting mixture
was evaporated under reduced pressure and the residue
washed with Et2O (2 × 5 mL) and redissolved in CH2Cl2, and
the resulting solution was filtered through Celite. The obtained
solution was evaporated and dissolved in DME (5 mL), and
the solution was hydrogenated in a Fischer-Porter reaction
vessel under 4 atm for 5 h. To the resulting solution was added
MAC (0.011 g, 0.05 mmol) and the mixture obtained was
stirred overnight. The solution was evaporated to dryness and
the resulting solid washed with petroleum ether, dissolved in
CH2Cl2, and filtered through Celite. Evaporation of the solvent
produced compound 10c as an orange powder. 1H NMR (CD2-
Cl2, 500 MHz): δ 0.63 (dd, 3J HP ) 18.5 Hz, 3J HH ) 7.0 Hz, 3H,
CHMe2), 0.78 (dd, 3J HP ) 15.0 Hz, 3J HH ) 6.9 Hz, 3H, CHMe2),
Gen er a l P r oced u r e for th e En a n tioselective Hyd r o-
gen a tion s. In a glovebox, a Fischer-Porter reactor (80 mL)
was charged with a solution of methyl Z-R-acetamido-cin-
namate (MAC) (0.23 g, 1.1 mmol) and 6a (0.002 g, 0.0022
mmol) in CH2Cl2 (5 mL). The vessel was brought outside the
glovebox, submitted to vacuum-hydrogen cycles, and finally
pressurized to 4 atm. The reaction mixture was stirred for 16
h, then the reactor was depressurized and the mixture
obtained was evaporated to dryness, redissolved in an ethyl
acetate-hexanes (1:1) mixture, and passed through a short
1
pad of silica. The resulting residue was analyzed by H NMR
3
3
1.29 (s, 9H, CMe3), 1.32 (dd, J HP ) 17.8 Hz, J HH ) 7.2 Hz,
to determine conversion and by chiral GC for enantiomeric
excess as follows: N-acetyl phenylalanine methyl ester
(Chrompack Chirasil L-Val, 80 °C (3 min), then 3 °C/min up
to 190 °C, 15.0 psi He) (R) t1 ) 32.14 min, (S) t2 ) 32.76 min.
X-r a y Str u ctu r e Deter m in a tion . Single crystals of 8a
were grown from diethyl ether solution at -30°C. X-ray data
were collected on a Bruker SMART 5000 CCD-based diffrac-
tometer, using ω-scans and a rotating anode with Mo KR
radiation (λ ) 0.71073 Å).35 The frames were integrated with
3
3H, CHMe2), 1.49 (dd, J HH ) 7.1 Hz, 3H, CHMe2), 1.51 (s,
9H, CMe3), 1.82 (s, 6H, 2 Ar-Me), 2.03 (br m, 1H, CHMe2), 2.25
(s, 3H, Ar-Me), 2.30 (s, 3H, C(O)Me), 2.35 (s, 3H, Ar-Me), 2.89
3
(br m, 1H, CHMe2), 3.01 (s, 3H, OMe), 6.20 (dd, J HH ) 8.0
4
3
2
Hz, J HP ) 4.4 Hz, 1H, Hd), 6.77 (dd, J HP ) 4.8 Hz, J HRh
)
2.8 Hz, 1H, CdCH), 7.16 (s, 1H, H arom), 7.27 (m, 1H, Hc),
7.30 (s, 1H, H arom), 7.34-7.40 (m, 3H, Hb and 2 Hm), 7.44-
7.49 (m, 4H, Ha, 2 Hï and Hp), 8.62 (br s, 1H, NH). 31P{1H}
1
NMR (CD2Cl2, 121 MHz): δ 19.2 (dd, J PRh ) 151 Hz, P-C),
136.6 (dd, J PRh ) 259 Hz, J PP ) 63 Hz, P-O). 13C{1H} NMR
1
2
(35) BrukerAXS, SMART, Area Detector Control Software; Bruker
Analytical X-ray Systems: Madison, WI, 1995.
(CD2Cl2, 75 MHz): δ 16.6 (Ar-Me), 16.7 (Ar-Me), 16.8 (d, J CP

Studies of New Chiral Phosphine-Phosphites
Organometallics, Vol. 21, No. 22, 2002 4621
the SAINT software package,36 using a narrow-frame algo-
rithm, and the structure was solved by direct methods,
completed by subsequent difference Fourier synthesis, and
refined by full-matrix least-squares procedures with SHELXTL
5.1.37 The structure was checked with PLATON.38 Non-H
atoms were refined with anisotropic thermal parameters.
Hydrogen atoms were treated as idealized contributions.
Pertinent crystallographic data are collected in Table 4.
Ack n ow led gm en t. We thank Drs. M. L. Poveda, N.
Khiar, M. Davies, M. Paneque, and V. Salazar and Prof.
E. Carmona for hepful comments. We acknowledge
financial support from Ministerio de Ciencia y Tecno-
log´ıa (Grant No. BQU2000-1169). M.A.M-R. also thanks
CONACYT-Me´xico and Universidad Auto´noma del Es-
tado de Hidalgo for a Ca´tedra de Consolidacio´n Insti-
tucional (No. 010316, CIQ-UAEH).
(36) BrukerAXS, SAINT, Integration Software; Bruker Analytical
X-ray Systems: Madison, WI, 1995.
(37) Sheldrick, G. M. SHELXTL 5.1, Program for Structure Solution
and Least Squares Refinement; University of Go¨ttingen: Go¨ttingen,
Germany, 1998.
(38) Spek, A. L. Acta Crystallogr. 1990, A46, C34. Speck, A. L.
PLATON, A Multipurpose Crystallographic Tool; Utrecht University:
Utrecht, The Netherlands, 2001.
Su p p or tin g In for m a tion Ava ila ble: Full experimental
and characterization details for the compounds described and
representative NMR spectra and crystallographic information
for compound 8a . This material is available free of charge via
the Internet at http://pubs.acs.org.
OM020140C