Platinum-catalyzed asymmetric alkylation of a secondary phosphine: Mechanism and origin of enantioselectivity

Article abstract of DOI:10.1021/om061116s

The catalyst precursor Pt((R,R)-Me-Duphos)(Ph)(Cl) (1) mediated asymmetric alkylation of the secondary phosphine PHMe(Is) (2; Is = 2,4,6-(i-Pr) ₃C₆H₂) with benzyl bromide in the presence of the base NaOSiMe₃ to yield enantioenriched PMeIs(CH₂Ph) (3). A mechanism for the catalysis has been proposed, on the basis of studies of the individual stoichiometric steps. The terminal phosphido complex Pt((R,R)-Me-Duphos)(Ph)(PMeIs) (4) was formed by proton transfer from 2 to the silanolate ligand in Pt((R,R)-Me-Duphos)(Ph)(OSiMe₃) (5), which was generated from 1 (or from Pt((R,R)-Me-Duphos)(Ph)(Br) (7)) and NaOSiMe ₃. The silanolate complex 5 reacted with water to yield Pt((R,R)-Me-Duphos)(Ph)(OH) (6); both 6 and 7 were crystallographically characterized. The stoichiometric reaction of 4 and benzyl bromide in toluene gave the bromide 7 and 3. In more polar solvents these compounds were in equilibrium with the cation [Pt((R,R)-Me-Duphos)(Ph)(PMeIs(CH ₂Ph))][Br] (8-Br), the major Pt complex present during catalysis, which was isolated as the BF₄ salt. Treatment of 8-BF₄ with 2 and NaOSiMe₃ yielded phosphine 3 and regenerated phosphido complex 4. This reaction does not appear to proceed via phosphine ligand substitution on 8-BF₄ to yield the secondary phosphine complex cation [Pt((R,R)-Me-Duphos)(Ph)(PHMe-(Is))][BF₄] (12). Instead, treatment of 8-BF₄ with NaOSiMe₃ gave phosphine 3 and 5, which then reacted with 2 to yield 4. The crystal structure of the major diastereomer of 8-BF₄ showed that the major enantiomer of 3 formed by catalyst precursor 1 had an R_P absolute configuration. Low-temperature NMR studies on the major diastereomer of phosphido complex 4 were consistent with the R_P solid-state structure. Thus, the major enantiomer of phosphine 3 appeared to be formed from the major diastereomer of intermediate 4, and enantioselectivity was determined mainly by the thermodynamic preference for one of the rapidly interconverting diastereomers of 4, although their relative rates of alkylation were also important (Curtin-Hammett kinetics).

Full text of DOI:10.1021/om061116s

1788
Organometallics 2007, 26, 1788-1800
Platinum-Catalyzed Asymmetric Alkylation of a Secondary
Phosphine: Mechanism and Origin of Enantioselectivity
Corina Scriban,† David S. Glueck,*,† James A. Golen,‡ and Arnold L. Rheingold‡
Departments of Chemistry, 6128 Burke Laboratory, Dartmouth College, HanoVer, New Hampshire 03755,
and UniVersity of California, San Diego, 9500 Gilman DriVe, La Jolla, California 92093
ReceiVed December 8, 2006
The catalyst precursor Pt((R,R)-Me-Duphos)(Ph)(Cl) (1) mediated asymmetric alkylation of the secondary
phosphine PHMe(Is) (2; Is ) 2,4,6-(i-Pr)3C6H2) with benzyl bromide in the presence of the base NaOSiMe3
to yield enantioenriched PMeIs(CH2Ph) (3). A mechanism for the catalysis has been proposed, on the
basis of studies of the individual stoichiometric steps. The terminal phosphido complex Pt((R,R)-Me-
Duphos)(Ph)(PMeIs) (4) was formed by proton transfer from 2 to the silanolate ligand in Pt((R,R)-Me-
Duphos)(Ph)(OSiMe3) (5), which was generated from 1 (or from Pt((R,R)-Me-Duphos)(Ph)(Br) (7)) and
NaOSiMe3. The silanolate complex 5 reacted with water to yield Pt((R,R)-Me-Duphos)(Ph)(OH) (6);
both 6 and 7 were crystallographically characterized. The stoichiometric reaction of 4 and benzyl bromide
in toluene gave the bromide 7 and 3. In more polar solvents these compounds were in equilibrium with
the cation [Pt((R,R)-Me-Duphos)(Ph)(PMeIs(CH2Ph))][Br] (8-Br), the major Pt complex present during
catalysis, which was isolated as the BF4 salt. Treatment of 8-BF4 with 2 and NaOSiMe3 yielded phosphine
3 and regenerated phosphido complex 4. This reaction does not appear to proceed via phosphine ligand
substitution on 8-BF4 to yield the secondary phosphine complex cation [Pt((R,R)-Me-Duphos)(Ph)(PHMe-
(Is))][BF4] (12). Instead, treatment of 8-BF4 with NaOSiMe3 gave phosphine 3 and 5, which then reacted
with 2 to yield 4. The crystal structure of the major diastereomer of 8-BF4 showed that the major enantiomer
of 3 formed by catalyst precursor 1 had an RP absolute configuration. Low-temperature NMR studies on
the major diastereomer of phosphido complex 4 were consistent with the RP solid-state structure. Thus,
the major enantiomer of phosphine 3 appeared to be formed from the major diastereomer of intermediate
4, and enantioselectivity was determined mainly by the thermodynamic preference for one of the rapidly
interconverting diastereomers of 4, although their relative rates of alkylation were also important (Curtin-
Hammett kinetics).
Scheme 1. Phosphorus Inversion and Alkylation in
Diastereomeric Metal Phosphido Complexesa
Introduction
We recently reported a new method for asymmetric synthesis
of P-stereogenic phosphines: Pt-catalyzed asymmetric alkylation
of racemic secondary phosphines.1 Independently, Bergman and
co-workers developed a similar Ru-catalyzed process.2 Our work
was guided by the hypothesis shown in Scheme 1. Terminal
metal phosphido complexes undergo rapid phosphorus inver-
sion.3 With a chiral ancillary ligand, the diastereomers AR and
AS of a P-stereogenic phosphido complex (*[M]-PRR′) often
interconvert quickly on the NMR time scale;4 they are nucleo-
philic5 and react with electrophiles, such as alkyl halides, with
different rate constants kR and kS to make new P-C bonds. If
the interconversion of AR and AS is faster than their alkylation
(k1, k-1 . kR, kS), then the product ratio P ) [R]/[S] will be
given by P ) Keq[kR/kS] (Curtin-Hammett kinetics).6,7
a M* ) chiral metal-ligand fragment, R′′X ) alkyl halide.
For enantioselective catalysis to succeed, the background
reaction of a secondary phosphine with an electrophile must be
much slower than the metal-catalyzed reaction, and the chiral
ancillary ligand must resist displacement from the metal by the
excess phosphine substrate and products. To meet these require-
ments, we used the base NaOSiMe3 to minimize the concentra-
* To whom correspondence should be addressed. E-mail: Glueck@
Dartmouth.Edu.
† Dartmouth College.
‡ University of California, San Diego.
(1) Scriban, C.; Glueck, D. S. J. Am. Chem. Soc. 2006, 128, 2788-
2789.
(5) (a) Malisch, W.; Maisch, R.; Colquhoun, I. J.; McFarlane, W. J.
Organomet. Chem. 1981, 220, C1-C6. (b) Bohle, D. S.; Jones, T. C.;
Rickard, C. E. F.; Roper, W. R. Organometallics 1986, 5, 1612-1619. (c)
Crisp, G. T.; Salem, G.; Wild, S. B.; Stephens, F. S. Organometallics 1989,
8, 2360-2367. (d) Buhro, W. E.; Zwick, B. D.; Georgiou, S.; Hutchinson,
J. P.; Gladysz, J. A. J. Am. Chem. Soc. 1988, 110, 2427-2439.
(6) Seeman, J. R. Chem. ReV. 1983, 83, 83-134.
(2) Chan, V. S.; Stewart, I. C.; Bergman, R. G.; Toste, F. D. J. Am.
Chem. Soc. 2006, 128, 2786-2787.
(3) Rogers, J. R.; Wagner, T. P. S.; Marynick, D. S. Inorg. Chem. 1994,
33, 3104-3110.
(4) (a) Wicht, D. K.; Kovacik, I.; Glueck, D. S.; Liable-Sands, L. M.;
Incarvito, C. D.; Rheingold, A. L. Organometallics 1999, 18, 5141-5151.
(b) Moncarz, J. R.; Laritcheva, N. F.; Glueck, D. S. J. Am. Chem. Soc.
2002, 124, 13356-13357.
(7) For an example of Scheme 1 in the alkylation of a chiral Fe-
phosphido complex, see ref 5c.
10.1021/om061116s CCC: $37.00 © 2007 American Chemical Society
Publication on Web 02/20/2007

Pt-Catalyzed Alkylation of a Secondary Phosphine
Organometallics, Vol. 26, No. 7, 2007 1789
Scheme 2a
a Is ) 2,4,6-(i-Pr)3C6H2.
Scheme 3a
Figure 1. ORTEP diagram of Pt((R,R)-Me-Duphos)(Ph)(OH) (6),
showing one of the two independent molecules in the asymmetric
unit.
Table 1. Crystallographic Data for the Complexes
Pt((R,R)-Me-Duphos)(Ph)(X) (X ) OH (6), Br (7),
PMeIs(CH2Ph) (8-BF4))a
6
7
8-BF4
formula
formula wt
space group
a, Å
b, Å
c, Å
C24H34OP2Pt
595.54
P21
8.434(2)
13.588(4)
20.825(6)
90
101.266(6)
90
2340.7(11)
4
1.690
6.143
208(2)
4.48
10.02
C24H33BrP2Pt
658.44
P212121
10.701(2)
13.890(3)
16.873(3)
90
90
90
2508.0(8)
4
1.744
C47H66BF4P3Pt
1005.81
P212121
12.9514(16)
17.462(2)
21.291(3)
90
a [Pt] ) Pt((R,R)-Me-Duphos).
tion of the phosphido anion [PRR′]- and the rigid bidentate
chiral alkylphosphine (R,R)-Me-Duphos to ensure tight binding
to the metal. We chose the secondary phosphine substrate
PHMe(Is) (2; Is ) 2,4,6-(i-Pr)3C6H2),8 with a large Is
and a small Me substituent, in hopes of controlling the
equilibrium ratio of phosphido diastereomers (Keq; Scheme 1).
Indeed, Pt-catalyzed alkylation of 2 was enantioselective (77%
ee; Scheme 2).1
R, deg
â, deg
90
90
γ, deg
V, Å3
4814.9(10)
4
1.388
3.059
213(2)
2.27
Z
D(calcd), g/cm3
µ(Mo KR), mm-1
temp, K
R(F), %b
Rw(F2), %b
7.323
213(2)
2.36
4.94
4.94
To test the hypothesis of Scheme 1, we studied the mechanism
of this catalytic reaction in detail by investigating the stoichio-
metric steps leading to P-C bond formation, the possibility of
catalyst inhibition by strong binding of the product phosphine,
and the origin of enantioselectivity.
a Complexes 6 and 7 are neutral; 8 is a cation (BF4 anion). b Quantity
minimized: Rw(F2) ) ∑[w(Fo - Fc )2]/∑[(wFo )2]1/2; R ) ∑∆/∑(Fo), ∆
2
2
2
) |(Fo - Fc)|, w ) 1/[σ2(Fo ) + (aP)2 + bP], P ) [2Fc2 + Max(Fo ,0)]/3.
2
2
A Bruker CCD diffractometer was used in all cases.
Information).10 Hydroxide 6 was structurally characterized by
X-ray crystallography (see Figure 1, Table 1, and the Supporting
Information for details).
Results and Discussion
1. Synthesis and Structure of Potential Intermediates in
Catalysis. (a) Formation of Phosphido Complex 4. The
catalyst precursor Pt((R,R)-Me-Duphos)(Ph)(PMeIs) (4)1 was
prepared by treating Pt((R,R)-Me-Duphos)(Ph)(Cl) (1)9 with
PHMe(Is)/NaOSiMe3 (Scheme 3); air-stable 1 was a more
convenient and general precatalyst. Reaction of 1 with NaO-
SiMe3 generated the silanolate Pt((R,R)-Me-Duphos)(Ph)(O-
SiMe3) (5), which was characterized spectroscopically. We could
not isolate 5 in pure form because of its ready reaction with
traces of water to yield the hydroxide complex Pt((R,R)-Me-
Duphos)(Ph)(OH) (6), which was prepared independently from
chloride 1 and NaOH, or from the methoxide Pt((R,R)-Me-
Duphos)(Ph)(OMe) and water (Scheme 3 and the Supporting
The silanolate 5 reacted quickly with PHMe(Is) (Scheme 3)
to yield the phosphido complex 4, which was expected to exist
as a mixture of two diastereomers, rapidly interconverting by
P inversion. At room temperature, only one set of NMR signals
for this complex was observed, consistent with the predicted
behavior. As discussed in more detail elsewhere, four diaster-
eomers of 4 were observed by low-temperature 31P NMR
spectroscopy in a 98:1:1:2 ratio at -50 °C.11 The “extra” signals
appeared to include some rotamers of the expected two
“invertomers”. We could not directly measure the barrier to
inversion in 4, but barriers in the closely related complexes Pt-
((R,R)-i-Pr-Duphos)(Ph)(PMeIs) and Pt((R,R)-Me-Duphos)(X)-
(10) For related platinum alkoxides and hydroxides and their intercon-
version, see: (a) Appleton, T. G.; Bennett, M. A. Inorg. Chem. 1978, 17,
738-747. (b) Bryndza, H. E.; Calabrese, J. C.; Wreford, S. S. Organome-
tallics 1984, 3, 1603-1604. (c) Bryndza, H. E.; Fong, L. K.; Paciello, R.
A.; Tam, W.; Bercaw, J. E. J. Am. Chem. Soc. 1987, 109, 1444-1456.
(11) Scriban, C.; Glueck, D. S.; DiPasquale, A. G.; Rheingold, A. L.
Organometallics 2006, 25, 5435-5448.
(8) Brauer, D. J.; Bitterer, F.; Dorrenbach, F.; Hessler, G.; Stelzer, O.;
Kruger, C.; Lutz, F. Z. Naturforsch., B 1996, 51, 1183-1196.
(9) Brunker, T. J.; Blank, N. F.; Moncarz, J. R.; Scriban, C.; Anderson,
B. J.; Glueck, D. S.; Zakharov, L. N.; Golen, J. A.; Sommer, R. D.; Incarvito,
C. D.; Rheingold, A. L. Organometallics 2005, 24, 2730-2746.

1790 Organometallics, Vol. 26, No. 7, 2007
Scriban et al.
(RP)-4 corresponds to the solid-state structure (Figure 2). The
opposite stereochemistry ((SP)-4) was observed in the crystal
structure of 4-i-Pr.11 Finally, rotation of (SP)-4 about the Pt-P
bond would yield the diastereomer (SP)-4′, which is related to
the solid-state structures of the cations [Pt((R,R)-Me-Duphos)-
(Ph)(PMeIs(CH2Ph))][BF4] (8; see below) and [Pt((R,R)-Me-
Duphos)(Ph)(PHMe(Is))][BF4] (12).11
NOEs between the PMeIs group and the Pt((R,R)-Me-
Duphos)(Ph) fragment were used to differentiate these structures.
First, several P-Me to Pt-Ph NOEs were observed (Figure
1
3), but since the PMe (C40) and Duphos Me (C11) H NMR
signals overlapped, assignment was not straightforward. How-
ever, estimating the H-H distances (Pt-Ph to P-Me and Pt-
Ph to Duphos Me) from the crystal structure of (RP)-4 enabled
assignment of NOEs to one or both of the methyl groups
(Table 2, Figure 3).
These observations, as well as NOEs observed between Is
hydrogens and the Pt-Ph group and substituents on both
phospholane rings, were consistent with the solid-state structure
and the (RP)-PMeIs absolute configuration (Supporting Informa-
tion). Ideally, determination of the solution structures of both
diastereomers of 4, and comparison of the data, would enable
definitive assignment of the absolute configuration. However,
because of the thermodynamic bias for one diastereomer of 4,
presumed to have the RP configuration, detailed NMR studies
of the minor isomer ((SP)-4 or (SP)-4′) were not possible.13
The crystal structure of (SP)-Pt((R,R)-i-Pr-Duphos)(Ph)(P-
MeIs) (4-i-Pr) provided a model for (SP)-4.11 The Pt(Ph)(PMeIs)
fragments in the solid-state structures of 4 and 4-i-Pr adopted
similar conformations. As a result of the opposite absolute
configurations, however, the PMeIs group was found on
different sides of the Pt-Ph ring in 4 and 4-i-Pr (Figure 4).
The position of the P-Me groups (C40 and C1, respectively)
changed little in these structures; the estimated P-Me to Pt-
Ph H-H distances calculated using 4-i-Pr as a model for (SP)-4
(as in Table 2; see the Supporting Information, Table S3) were
also consistent with the observed NOEs.
Figure 2. ORTEP diagram of Pt((R,R)-Me-Duphos)(Ph)(PMeIs)
(4).11
Chart 1. Possible Phosphido Configurations in the Major
Diastereomer of 4, with Solid-State Models
(PMeIs) (X ) Cl, I) were about 10-12 kcal/mol,11 similar to
those in related phosphido complexes cited in refs 3 and 11.
In the solid state only one diastereomer of 4, with an (RP)-
PMeIs group, was observed in the crystal studied (Figure 2).11
The related complexes Pt((R,R)-Me-Duphos)(X)(PMeIs) (X )
I, Cl, PMeIs) also crystallized with the RP configuration, while
Pt((R,R)-i-Pr-Duphos)(Ph)(PMeIs), whose Duphos ligand has
an absolute configuration opposite that of the (R,R)-Me-Duphos
one, crystallized as the SP diastereomer.11 The simplest explana-
tion of these observations is that the RP configuration was
thermodynamically favorable for the (R,R)-Me-Duphos com-
plexes and the major diastereomer was observed in each case.
However, it is also possible that this diastereomer was favored
in the solid state or that the crystals chosen for structure
determination were not representative of the bulk.
However, the positions of the P-Is groups differed signifi-
cantly in 4 and 4-i-Pr (Figure 4). Although the argument
requires several assumptions, a combination of the estimated
H-H distances between the o-i-Pr methine hydrogens H37 (in
4) and H8 (in 4-i-Pr) and the Pt-Ph ring (Table 3) and the
NOE observations may be used to differentiate structures 4 and
4-i-Pr and the associated P stereochemistries.
Therefore, we used low-temperature NMR spectroscopy to
investigate the absolute configuration of the major diastereomer
of 4 in solution; details are given in the Experimental Section
and the Supporting Information.12 Rotation about the Pt-C(Ph)
bond was slow under these conditions, and the known (R,R)-
Me-Duphos stereochemistry was used to determine the position
of the Pt-Ph group, which appeared to be perpendicular to the
The observed NOEs for 4 from H37 to H6 and H5 (Figure
4) were consistent with the estimated solid-state distances,
highlighted in boldface type in Table 3. The hypothetical (SP)-4
(if 4-i-Pr is a good model for it) would be expected to show
NOEs from H37 to H2, H3, and H6 but not to H4 and H5 (Table
3, highlighted in boldface type for H8, the analogue of H37 in
4-i-Pr). Although it is dangerous to make arguments on the
basis of negative observations,14 the lack of NOEs between H37
and H2 or H3, combined with the observed H37-H5 NOE, is
consistent with (RP)-4 and not with (SP)-4 (as modeled by (SP)-
4-i-Pr).
1
square plane, as in the solid state (Figure 2). H-1H NOEs
between the cis (P1) phospholane substituents and the Pt-Ph
hydrogens were then used to establish the relative position of
the Pt-Ph group with respect to the square plane (H2 and H3
above, H5 and H6 below).
This information on the structure and conformation of the
Pt((R,R)-Me-Duphos)(Ph) fragment was then used to investigate
the absolute configuration of the phosphido ligand. Chart 1
shows three different ways to arrange the Me, Is, and Pt
substituents and the lone pair about the pyramidal P center.
It is difficult to rule out another possible structure ((SP)-4′),
derived from (RP)-4 by interchanging the P-Me and the lone
pair, which would require minimal motion and be consistent
with observed NOEs to the Is group (Figure 4). However,
(13) As described in more detail above and in ref 11, four diastereomers
of 4 were observed at low temperature; we were only able to study the
NMR spectra of the major one in detail.
(14) Neuhaus, D.; Williamson, M. P. The Nuclear OVerhauser Effect in
Structural and Conformational Analysis. VCH: New York, 1989; p 358.
(12) For a related report of NOESY data for Pd(Me-Duphos) aryl
complexes, see: Drago, D.; Pregosin, P. S. Organometallics 2002, 21,
1208-1215.

Pt-Catalyzed Alkylation of a Secondary Phosphine
Organometallics, Vol. 26, No. 7, 2007 1791
Figure 3. NOEs between the C40 P-Me group, the Duphos C11 Me group, and the Pt-Ph hydrogens and resulting assignments. For
clarity, NOEs from the C11 Me group to the Pt-Ph hydrogens are not shown in the left-hand panel (see the related discussion in the text).
Arrows in the right-hand panel (a portion of the ORTEP diagram in Figure 2) show NOEs from both the P-Me (C40, solid arrows) and
the C11 Me (dashed arrows) to H2 and H3. As discussed above, however, we cannot tell which of the two methyl groups makes a bigger
contribution to these NOEs.
Table 2. Calculated Average Distances (Å) between the
Pt-Ph Hydrogens and the P-Me (C40) and Duphos Me
(C11) Hydrogens in the Structure of 4 and the Resulting
Assignment of the Observed Pt-Ph/Me NOEs
diastereomer makes it impossible to definitively rule out an SP
configuration for 4. With this caveat in mind, we assume that
the major diastereomer of 4 is RP in the mechanistic analysis
below.
(b) Alkylation of Phosphido Complex 4. Formation of 3,
7, and 8. The stoichiometric reaction of 4 with benzyl bromide
in toluene gave the bromide complex Pt((R,R)-Me-Duphos)-
(Ph)(Br) (7) and the phosphine PMeIs(CH2Ph) (3) (Scheme 4).
Consistent with the ideas of Scheme 1 and the observations in
catalysis, isolated phosphine 3 was enantioenriched (70% ee at
21 °C, 79% ee at -5 °C).
d(H-H)
Pt-Ph H
P-Me (C40) (av)
CHMe (C11) (av)
NOE assignt
H2a (o)
H3a (m)
H4a (p)
H5a (m)
H6a (o)
4.099
4.851
5.053
4.400
3.532
4.001
5.583
7.168
7.547
6.495
both
both
P-Me
P-Me
P-Me
The structure of bromide complex 7 was analogous to that
of the previously reported chloride 1 and hydroxide 6 (see Figure
6, Tables 1 and 4, and the Supporting Information).9 The
identical Pt-P (trans to halide) distances of 2.2028(11) and
2.2012(7) Å for 7 and 1, respectively, showed that Br and Cl
have the same trans influence in these complexes. This is
consistent with 31P NMR data in toluene; JPt-P values for the
Duphos P trans to halide were 3892 and 3893 Hz for 7 and 1.15
The Pt-P (trans to OH) distance in 6 of 2.195(2) Å (average
value for the two molecules in the unit cell), along with the
JPt-P value (3218 Hz), showed that the trans influence of
hydroxide is similar to that of these halides.
When a mixture of bromide 7 and phosphine 3 was dissolved
in CD2Cl2, an equilibrium with the cation [Pt((R,R)-Me-
Duphos)(Ph)(PMeIs(CH2Ph))][Br] (8-Br) was observed (Scheme
4). A similar mixture was obtained when phosphido complex 4
was generated from Pt((R,R)-Me-Duphos)(Ph)(Cl) (1) with
PHMe(Is)/NaOSiMe3 in situ and then alkylated. These results
are consistent with formation of the cation 8-Br at equilibrium
favored by a more polar medium.16
comparison to the crystal structures of analogous complexes
provided some evidence against this structure. Table 3 includes
estimated H-C distances from the Pt-Ph hydrogens to the P-C
carbons in 4 and 4-i-Pr. The position of Is ipso carbon C2 in
4-i-Pr is a model for the P-Me carbon in the hypothetical (SP)-
4′ (Figure 4). The significantly longer estimated distances to it,
in comparison to those to the real Me group (C1), are potentially
inconsistent with the NOE observations. A similar trend is
apparent for the C40-H(Ph) and C25-H(Ph) distances in 4
(Table 3).
Similarly, Figure 5 shows portions of the solid-state structures
of 4 and its protonated and alkylated analogues [Pt((R,R)-Me-
Duphos)(Ph)(PHMe(Is))][BF4] (12)11 and [Pt((R,R)-Me-Duphos)-
(Ph)(PMeIs(CH2Ph))][BF4] (8; see below). The P-H hydrogen
(H3) in 12 provided another model for the position of the P-Me
in the hypothetical (SP)-4′. The calculated position of H3 was
further from the Pt-Ph ring than was the P-Me carbon (C40;
see Table S5 in the Supporting Information), potentially
inconsistent with the observed NOEs of Figure 3. Rotation about
the Pt-P bond in 8-BF4 places the benzyl group (C34) in a
position distinct from those of other P substituents in 4, 4-i-Pr,
and 12. Here too C34 was further from the Pt-Ph ring than the
P-Me carbon (C41; see Table S6 in the Supporting Informa-
tion).
Treatment of 1 with AgBF4 in acetonitrile yielded [Pt((R,R)-
Me-Duphos)(Ph)(NCMe)][BF4], and subsequent reaction with
PMeIsCH2Ph (3) gave 8-BF4, isolated as a mixture of two
diastereomers. Reaction of 8-BF4 with excess tetraoctylammo-
nium bromide (NOct4Br) in toluene gave phosphine 3 and
bromide 7, consistent with the proposed equilibrium (Scheme
Therefore, the NOEs shown in Figures 3 and 4 and the
Supporting Information are most simply explained by assuming
that the major diastereomer of complex 4 contained a (RP)-
PMeIs group in solution, as observed in the solid state. Although
this conclusion is consistent with both the solution NMR data
and comparison to the crystal structure of 4 and several of its
analogues, the absence of NMR data for the (inaccessible) minor
(15) Appleton, T. G.; Clark, H. C.; Manzer, L. E. Coord. Chem. ReV.
1973, 10, 335-422.
(16) This is consistent with the equilibrium observed in ref 4b between
Pd((R,R)-Me-Duphos)(Ph)(I), PHMe(Is) and the cation [Pd((R,R)-Me-
Duphos)(Ph)(PHMe(Is))][I], which favored the neutral iodo complex.

1792 Organometallics, Vol. 26, No. 7, 2007
Scriban et al.
Figure 4. Partial ORTEP diagrams (p-i-Pr groups omitted) for the complexes (RP)-Pt((R,R)-Me-Duphos)(Ph)(PMeIs) (4) and (SP)-Pt-
((R,R)-i-Pr-Duphos)(Ph)(PMeIs) (4-i-Pr, one of the two independent molecules in the unit cell), illustrating close contacts between ortho
i-Pr CH groups H37 (in 4) and H8 (in 4-i-Pr) and the Pt-Ph ring. The arrows on the left show NOEs observed between i-Pr CH H37 and
Pt-Ph hydrogens H5 and H6.
Table 3. Selected Estimated Solid-State H-H and H-C Distances (Å) in the Complexes (RP)-Pt((R,R)-Me-Duphos)(Ph)(PMeIs)
(4) and (SP)-Pt((R,R)-i-Pr-Duphos)(Ph)(PMeIs) (4-i-Pr)a
Pt-Ph (4)
d(H-H) (H37)
d(C-H) (C40, Me)
d(C-H) (C25, Is)
Pt-Ph (4-i-Pr)
d(H-H) (H8)
d(C-H) (C1, Me)
d(C-H) (C2, Is)
H2a (o)
H3a (m)
H4a (p)
H5a (m)
H6a (o)
4.895
6.166
6.039
4.495
2.393
3.913
4.840
5.140
4.478
3.453
5.507
7.165
7.655
6.609
4.738
H22 (o)
H21 (m)
H20 (p)
H19 (m)
H18 (o)
2.245
4.290
5.860
6.037
4.816
3.607
4.669
5.190
4.764
3.754
4.643
6.515
7.598
7.178
5.546
a Distances were estimated using calculated hydrogen atom positions; average values for the two independent molecules are reported for 4-i-Pr. Distances
given in boldface type for 4 correspond to observed NOEs, while distances given in boldface type for 4-i-Pr correspond to predicted NOEs.
Figure 5. Partial ORTEP diagrams of the complexes Pt((R,R)-Me-Duphos)(Ph)(PMeIs) (4), [Pt((R,R)-Me-Duphos)(Ph)(PHMe(Is))][BF4]
(12-BF4), and [Pt((R,R)-Me-Duphos)(Ph)(PMeIs(CH2Ph))][BF4] (8-BF4).11
Scheme 4a
4). The crystal structure of the major diastereomer of 8-BF4
(Figure 7, Table 1, and the Supporting Information) and the 1H
NMR spectrum of the single crystal used for structure deter-
mination showed that its coordinated phosphine had an SP
absolute configuration. This corresponds to the RP configuration
for the major enantiomer of phosphine 3, without the Pt
substituent.
Comparison of this structure with that of 4 and its protonated
derivative [Pt((R,R)-Me-Duphos)(Ph)(PHMe(Is))][BF4] (12) pro-
vided information on the structural consequences of protonation/
alkylation at P (Table 5).11 The sum of angles at P3 increased
on quaternization, from 320.6(2)° in the phosphido species 4
(pyramidal geometry) to 342.47(19)° in the secondary phosphine
a [Pt] ) Pt((R,R)-Me-Duphos).
complex 12 (reduced steric demand of a proton in comparison
to the P lone pair). The average P3 angle in the alkylated cation
8-BF4 was 109.43(15)°, the ideal value for tetrahedral geometry.
Presumably P3 is more tetrahedral in 8-BF4 than in 12 because
the benzyl group is more sterically demanding than the proton.
Even with the increased steric bulk around P, the Pt-P3 distance

Pt-Catalyzed Alkylation of a Secondary Phosphine
Organometallics, Vol. 26, No. 7, 2007 1793
Table 5. Selected Bond Lengths (Å) and Angles (deg) for the
Complexes Pt((R,R)-Me-Duphos)(Ph)(PMeIs) (4),
[Pt((R,R)-Me-Duphos)(Ph)(PHMe(Is))][BF4]‚CH2Cl2
(12‚CH2Cl2), and
[Pt((R,R)-Me-Duphos)(Ph)(PMeIs(CH2Ph))][BF4] (8-BF4)
4
12‚CH2Cl2
8-BF4
Pt-P1
Pt-P2
Pt-P3
Pt-X
2.2849(14)
2.3028(13)
2.3761(13)
2.070(5)
2.2772(13)
2.3096(11)
2.3126(13)
2.082(4)
2.2853(9)
2.3085(7)
2.3338(8)
2.093(3)
P1-Pt-P2
P1-Pt-P3
P1-Pt-Ph
P2-Pt-P3
P2-Pt-Ph
P3-Pt-Ph
P3 angles
85.66(5)
85.52(4)
84.93(3)
171.14(3)
88.47(9)
103.09(3)
172.94(9)
83.35(8)
115.60(14)
99.07(15)
102.62(14)
103.65(11)
113.59(10)
122.04(10)b
b
171.47(5)
89.83(14)
97.37(5)
173.81(15)
86.51(14)
98.8(2)
177.11(4)
90.21(13)
96.82(4)
172.02(16)
87.26(13)
103.9(13)
120.38(15)
118.19(19)
109.95(15)
111.89(19)
Figure 6. ORTEP diagram of Pt((R,R)-Me-Duphos)(Ph)(Br) (7).
Table 4. Selected Bond Lengths (Å) and Angles (deg) in the
Complexes Pt((R,R)-Me-Duphos)(Ph)(X) (X ) Cl (1), Br (7),
OH (6))
∑ P3 angles
320.6(2)
342.47(19)
a Labeling of the phosphorus atoms: P3 ) phosphido in 4, secondary
phosphine in 12‚CH2Cl2, tertiary phosphine in 8-BF4; P1 ) Duphos trans
to P3; P2 ) Duphos cis to P3. Note that the labeling shown in the ORTEP
diagram of the complex 8-BF4 is slightly different. Data for complexes 4
and 12‚CH2Cl2 are taken from ref 11. b The complex 8-BF4 is unique in
that bond angles involving all four P substituents could be measured (for
the other complexes only three were available). The average angle at P3
was 109.43(15)°.
7
19
6b
Pt-C
Pt-P2
Pt-P1
Pt-X
2.065(4)
2.074(3)
2.054(10)
2.195(2)
2.289(3)
2.055(6)
2.2028(11)
2.2755(11)
2.4900(5)
2.2012(7)
2.2774(7)
2.3721(7)
C-Pt-P2
C-Pt-P1
P2-Pt-P1
C-Pt-X
P2-Pt-X
P1-Pt-X
93.51(12)
171.26(13)
87.00(4)
89.30(11)
175.14(3)
90.84(3)
93.67(8)
170.79(9)
86.99(3)
89.47(8)
175.22(2)
90.50(3)
93.3(3)
177.4(3)
86.44(9)
87.8(3)
177.4(2)
92.6(2)
Table 6. Selected 31P NMR Data (in Hz) and Bond Lengths
(Å) for the Terminal and Alkylated/Protonated Phosphido
a
Group in the Complexes 4, 8-BF4, and 12‚CH2Cl2
J
Pt-P (trans
to PR2)
complex
Pt-P1 (Å)
a Labeling: P1 is trans to Ph, P2 is trans to X. b Two independent
molecules in the unit cell; average value. Note that the P atom numbering
in Figure 1 is reversed from the convention in this table.
Pt((R,R)-Me-Duphos)(Ph)(PMeIs) (4)
[Pt((R,R)-Me-Duphos)(Ph)(PMeIs-
(CH2Ph))][BF4] (8-BF4)
[Pt((R,R)-Me-Duphos)(Ph)(PHMe(Is))]-
[BF4] (12)
1621
2.2849(14)
2.2853(9)
2443, 2459b
2623, 2618b
2.2772(13)
a Labeling of the phosphorus atoms is as in Table 5: P3 ) phosphido in
4, secondary phosphine in 12‚CH2Cl2, tertiary phosphine in 8-BF4; P1 )
Duphos trans to P3; P2 ) Duphos cis to P3. Note that the labeling shown
in the ORTEP diagram of complex 8-BF4 is slightly different. Solvent:
CD2Cl2 for cations 8 and 12, C6D6 (room temperature) for 4. Data for 4
and 8 is from ref 11. b Data are given for the two diastereomers of 8-BF4
and 12.
From the Pt-P1 (trans to P3) bond lengths, which are the
same for 4 and 8-BF4 but a bit shorter for 12, the apparent trans
influence trend would be PMeIs ≈ PMeIs(CH2Ph) > PHMe-
(Is).15 The 31P NMR data (Table 6) showed that JPt-P for the
phosphine trans to the phosphido group increased upon alky-
lation/protonation, suggesting a PMeIs > PMeIs(CH2Ph) >
PHMe(Is) trans influence series for these ligands.11
2. Mechanism and Origin of Enantioselectivity. We
investigated the mechanisms of the stoichiometric reactions
which interconvert these potential intermediates to learn more
about the overall mechanism of catalysis.
(a) Formation of Phosphido Complex 4. We reported
previously that treatment of the palladium iodo complex Pd-
((R,R)-Me-Duphos)(Ph)(I) (1-Pd-I) with PHMe(Is)/NaOSiMe3
led to 4-Pd via deprotonation of the intermediate cationic
secondary phosphine complex [Pd((R,R)-Me-Duphos)(Ph)(PH-
Me(Is))][I];16 no reaction between NaOSiMe3 and PHMe(Is),
Figure 7. ORTEP diagram of [Pt((R,R)-Me-Duphos)(Ph)(PMeIs-
(CH2Ph))][BF4] (8-BF4). The BF4 anion and disorder involving C
atoms 14, 15, 17, 28, 29, 30, and 38 are omitted.
in phosphido complex 4 (2.3761(13) Å) shortened to 2.3126-
(13) Å on protonation (12) and to 2.3338(8) Å on alkylation
(8-BF4), perhaps because quaternization eliminated repulsive
Pt-P π-interactions.17 The Pt-Ph bonds became slightly longer
on quaternization, changing from 2.070(5) Å in 4 to 2.082(4)
Å in 12 and 2.093(3) Å in 8-BF4; the shorter Pt-P3 bond might
result in greater steric repulsion.
(17) (a) Caulton, K. G. New J. Chem. 1994, 18, 25-41. (b) Holland, P.
L.; Andersen, R. A.; Bergman, R. G. Comments Inorg. Chem. 1999, 21,
115-129.

1794 Organometallics, Vol. 26, No. 7, 2007
Scriban et al.
Scheme 5. Potential Mechanisms for Formation of
Phosphido Complexes 4 and 4-Pda
Scheme 9. Possible Mechanisms for Formation of 4 from
8-Bra
a [M] ) M((R,R)-Me-Duphos), M ) Pd, Pt.
Scheme 6a
a [Pt] ) Pt((R,R)-Me-Duphos).
Scheme 10a
a [Pt] ) Pt((R,R)-Me-Duphos).19b
Scheme 7a
a [Pt] ) Pt((R,R)-Me-Duphos).
Scheme 11a
a [Pt] ) Pt(dppe).19b
Scheme 8a
a [Pt] ) Pt((R,R)-Me-Duphos).
a [Pt] ) Pt((R,R)-Me-Duphos).
(b) Alkylation of Phosphido Complex 4. We hypothesized,
as in Scheme 1, that P-C bond formation in 4 occurred by
direct SN2 attack of the Pt-phosphido group on benzyl bromide
to yield the cation 8-Br (path a, Scheme 6), followed (in toluene)
by displacement of the coordinated phosphine 3 by the bromide
counterion. Alternatively, oxidative addition of benzyl bromide
to 4 (path b, Scheme 6) might yield a Pt(IV) intermediate, from
which selective P-C reductive elimination would give the
products 7 and 3. Although oxidative addition of benzyl bromide
to the Pt(II) phosphine complex Pt(PMe2Ph)2Me2 is known,19
P-C reductive elimination from Pt(IV) does not appear to have
been reported.20
or between NaOSiMe3 and 1-Pd-I was observed by NMR
spectroscopy (Scheme 5).4b In contrast, the Pt chloride 1 did
not appear to react with PHMe(Is). Instead, treatment of 1 with
PHMe(Is) and NaOSiMe3 gave 4; this reaction seemed to
proceed via formation of the silanolate 5, followed by proton
transfer from the secondary phosphine to yield 4 (Schemes 3
and 5). The latter step presumably occurred by coordination of
phosphine 2 to Pt in 5, which would make the P-H proton
more acidic; a four-center transition state for Pt-P and O-H
bond formation is plausible, or the ion pair [Pt((R,R)-Me-
Duphos)(Ph)(PHMe(Is))][OSiMe3], formed via displacement of
the silanolate ion by 2, may be involved.11,18 Since both the
metal and the leaving group were changed, we cannot tell which
factor controlled the difference in the mechanism of M-P bond
formation.
To examine these possibilities in a model system, we treated
Pt(dppe)(Me)(PPh2) with 13C-labeled methyl iodide (*MeI;
(19) (a) Brown, M. P.; Puddephatt, R. J.; Upton, C. E. E.; Lavington, S.
W. J. Chem. Soc., Dalton Trans. 1974, 1613-1618. (b) The stereochemistry
of the proposed Pt(IV) complexes in Schemes 6 and 7 was drawn arbitrarily
by analogy to refs 19a and 22; for more information on the stereochemistry
of related complexes, see also: Brown, M. P.; Puddephatt, R. J.; Upton, C.
E. E. J. Chem. Soc., Dalton Trans. 1974, 2457-2465. Goldberg, K. I.;
Yan, J.; Breitung, E. M. J. Am. Chem. Soc. 1995, 117, 6889-6896.
(20) For studies of C-O reductive elimination from analogous Pt(IV)
complexes, see: Williams, B. S.; Goldberg, K. I. J. Am. Chem. Soc. 2001,
123, 2576-2587.
(18) For formation of Pt-phosphido complexes by treatment of a Pt-
methoxide complex with primary or secondary phosphines, see: (a) Wicht,
D. K.; Paisner, S. N.; Lew, B. M.; Glueck, D. S.; Yap, G. P. A.; Liable-
Sands, L. M.; Rheingold, A. L.; Haar, C. M.; Nolan, S. P. Organometallics
1998, 17, 652-660. (b) Kourkine, I. V.; Sargent, M. D.; Glueck, D. S.
Organometallics 1998, 17, 125-127.

Pt-Catalyzed Alkylation of a Secondary Phosphine
Organometallics, Vol. 26, No. 7, 2007 1795
1
Chart 2. H and 13C (Boldface Italics) NMR Assignments for the Major Diastereomer of 4 (THF-d8, -60 °C)
a Which Duphos quaternary C atom is adjacent to P1 (or P2) could not be determined; similarly, the o-Duphos Ar CH signals overlapped. b Only
two of the three expected 13C NMR signals for the Is o- and p-quaternary C atoms were observed; we cannot tell the difference between the ortho
1
and para resonances. c We cannot tell which Is m-C is on which side of the ring because of the coincident H NMR shifts of H30 and H35. d We
could not differentiate the CH2 groups in the phospholane or unambiguously assign their 13C signals. e The assignment of the o-CH (i-Pr) 1H NMR
1
peaks shown is based on their similar chemical shifts and NOE data. f Because of H NMR peak overlap of the Me (i-Pr) signals, definitive
assignment of the corresponding 13C NMR signals was not possible. The assignments shown are consistent with correlations between the i-Pr CH
and Me signals, but unambiguous assignment of individual peaks to ortho or para positions was not possible.
Scheme 7).18,21 SN2 alkylation of the phosphido group (path a)
would yield the cation [Pt(dppe)(Me)(PPh2*Me)][I] (9); by
analogy to the chemistry of Schemes 4 and 6, it might be in
equilibrium with neutral Pt(dppe)(Me)(I) (10) and PPh2*Me,
with the carbon label only in the P-*Me group. After oxidative
addition of *MeI (path b), however, reductive elimination from
the Pt(IV) complex Pt(dppe)(Me)(*Me)(PPh2)(I) (11) would
yield 10 and PPh2Me, with the labeled methyl group scrambled
between Pt and P sites. The known reaction of MeI with Pt-
(dppe)Me2, which gave the stable Pt(IV) complex Pt(dppe)-
Me3I, suggests that this pathway is plausible.22
unfavorable equilibrium, in which 8-Br was favored over 12-
Br, although this seems unlikely on steric grounds. This
possibility was tested by treating 12-BF4 with PMeIs(CH2Ph),
but no reaction occurred. We cannot rule out a slowly
established equilibrium between cations 8-Br and 12-Br, leading
to product 4, but the two other pathways in Scheme 9 appear
to be more likely.
(b) From Scheme 4, the cation 8-Br was in equilibrium with
phosphine 3 and bromide 7. Stoichiometric treatment of 7 with
NaOSiMe3/PHMe(Is) yielded the phosphido complex 4 via
silanolate 5 (Scheme 9). Catalytic turnover might then plausibly
occur by a combination of these two processes. The position of
the 8-Br/3 + 7 equilibrium during catalysis would depend on
the concentration of the bromide anion and that of the tertiary
phosphine 3. NaBr precipitated during the reaction while more
phosphine 3 formed, which should favor 8-Br. Further, the
reduced concentration of NaOSiMe3 and PHMe(Is) as the
reaction proceeds should slow down the formation of 4 from 7,
reducing the rate of catalyst turnover by this pathway.
(c) In the third possible pathway, the cation 8-Br reacted with
NaOSiMe3 to yield silanolate 5, which then deprotonated PHMe-
(Is), regenerating the Pt-phosphido complex 4. These reactions
were observed directly with 8-BF4. On treatment with NaO-
SiMe3 it was slowly partially converted to 5 in an apparent
equilibrium. Silanolate complex 5 then reacted quickly with
secondary phosphine 2 to yield phosphido complex 4.
From these observations of the stoichiometric steps in
catalysis, pathways (b) and (c) are most likely responsible for
catalytic turnover. The proposed mechanism is summarized in
Scheme 10.
The experiment of Scheme 7 provided unambiguous evidence
for the SN2 mechanism. Treatment of Pt(dppe)(Me)(PPh2) with
*MeI in toluene-d8 caused immediate precipitation of cation 9,
1
which was identified by 31P, 13C, and H NMR spectroscopy
and by comparison to the reported 31P NMR spectrum of the
PF6 salt;23 the 13C label was found exclusively in the P site. By
analogy, we assume that alkylation of phosphido complex 4
also proceeded by a nucleophilic pathway.
(c) Product Formation and Catalyst Regeneration. Cation
8-Br was the major Pt complex observed by 31P NMR
spectroscopy during catalysis,24 suggesting that the rate-
determining step involved substitution of the coordinated tertiary
phosphine 3 with the phosphido ligand. Stoichiometric treatment
of 8-BF4 with PHMe(Is) and NaOSiMe3 gave 3 and regenerated
the phosphido complex 4 (Scheme 8).
We considered three possible mechanisms for this step under
catalytic conditions (Scheme 9). (a) The simplest would involve
ligand substitution, in which PHMe(Is) displaced the bulkier
tertiary phosphine to yield [Pt((R,R)-Me-Duphos)(Ph)(PHMe-
(Is))][Br] (12-Br). Independently prepared11 12-BF4 reacted
quickly with NaOSiMe3 to give 4; thus, 12-Br would be a
competent intermediate.
Origin of Enantioselectivity. As in Scheme 1, we hypoth-
esized that P inversion in the phosphido intermediate 4 was
much faster than alkylation. Under these Curtin-Hammett
conditions, is the major product enantiomer formed from the
major or the minor Pt-phosphido diastereomer?25,26 P-alkylation
at three-coordinate organophosphorus centers as well as in
However, the proposed ligand substitution via reaction of
8-BF4 with PHMe(Is) was very slow. This might be due to an
(21) Scriban, C.; Wicht, D. K.; Glueck, D. S.; Zakharov, L. N.; Golen,
J. A.; Rheingold, A. L. Organometallics 2006, 25, 3370-3378.
(22) Appleton, T. G.; Bennett, M. A.; Tomkins, I. B. J. Chem. Soc.,
Dalton Trans. 1976, 439-446.
(23) Goel, A. B.; Goel, S. Inorg. Chim. Acta 1982, 59, 237-240.
(24) Speciation appeared to depend on concentration and percent
conversion during catalysis. In addition to 8-Br, a small amount of
phosphido complex 4 was also observed later in the reaction.
(25) Product racemization via pyramidal inversion should be slow under
these conditions (Mislow, K. Trans. N. Y. Acad. Sci. 1973, 35, 227-242).
(26) (a) Halpern, J. Science 1982, 217, 401-407. (b) Landis, C. R.;
Halpern, J. J. Am. Chem. Soc. 1987, 109, 1746-1754. (c) Schmidt, T.;
Baumann, W.; Drexler, H.-J.; Arrieta, A.; Heller, D.; Buschmann, H.
Organometallics 2005, 24, 3842-3848.

1796 Organometallics, Vol. 26, No. 7, 2007
Scriban et al.
Reagents were from commercial suppliers, except for these
compounds, which were made by the literature procedures: Pt-
((R,R)-Me-Duphos)(Ph)(Cl),9 PMeIs(CH2Ph),1 PHMe(Is),8 Pt((R,R)-
Me-Duphos)(Ph)(PMeIs) and [Pt((R,R)-Me-Duphos)(Ph)(PHMeIs)]-
[BF4],1,11 and Pt(dppe)(Me)(PPh2).21
metal-phosphido complexes proceeds with retention of
configuration;27,5c thus, alkylation of (RP)-4 should yield (RP)-3
(SP when it remains complexed to Pt, in the cation 8-BF4). This
matched the experimental observations.28
Therefore, the major product ((RP)-3) appears to be formed
from the major intermediate ((RP)-4). In contrast to the classic
major product from minor intermediate result in asymmetric
hydrogenation,26 these results suggest that enantioselectivity was
determined mainly by the thermodynamic preference for one
of the rapidly interconverting diastereomers of 4, although their
relative rates of alkylation were also important (Curtin-
Hammett kinetics; Scheme 11).
Generation of Pt((R,R)-Me-Duphos)(Ph)(OSiMe3) (5). To a
stirred slurry of Pt((R,R)-Me-Duphos)(Ph)(Cl) (1; 123 mg, 0.2
mmol) in toluene (0.2 mL) was added a slurry of NaOSiMe3 (22
mg, 0.2 mmol) in 0.3 mL of toluene. The slurry immediately
dissolved; the solution was transferred into an NMR tube and
monitored by 31P NMR spectroscopy. After 3 days the mixture
consisted of Pt((R,R)-Me-Duphos)(Ph)(OSiMe3) (5; 67%), Pt((R,R)-
Me-Duphos)(Ph)(OH) (6; 15%), and unreacted Pt((R,R)-Me-Du-
phos)(Ph)(Cl) (1; 18%). The reaction mixture was added to
NaOSiMe3 (7 mg, 0.06 mmol), and according to the 31P NMR
spectrum the only species present were Pt((R,R)-Me-Duphos)(Ph)-
(OSiMe3) (5; major) and Pt((R,R)-Me-Duphos)(Ph)(OH) (6). The
reaction mixture was filtered through Celite, and the solvent was
removed under vacuum. The 31P NMR spectrum (CD2Cl2) of the
solution indicated a mixture of Pt((R,R)-Me-Duphos)(Ph)(OSiMe3)
Conclusions
We have presented evidence for the mechanism (Scheme 10)
of Pt-catalyzed asymmetric alkylation of a secondary phosphine.
Consistent with the hypothesis of Scheme 1, the phosphido
complex 4 existed as a mixture of diastereomers, which
interconverted via rapid P inversion. Nucleophilic attack on
benzyl bromide yielded the product phosphine 3, which
remained coordinated to Pt in the major resting state 8; thus, as
expected, product inhibition slowed catalytic turnover. The two
possible routes we identified for turnover (Scheme 9) suggest
ways to address this problem in this system and related catalyses.
First, a smaller base such as methoxide might favor nucleophilic
displacement of the tertiary phosphine (so might a less bulky
phosphine, although that might reduce the enantioselectivity).
The potential bromide route via 7 suggests that having more
bromide available (with a polar solvent or added bromide) might
also increase the rate.
1
(86%) and Pt((R,R)-Me-Duphos)(Ph)(OH) (14%). The H NMR
spectrum (CD2Cl2) also showed some impurities; therefore, the
solvent was removed under reduced pressure and the white residue
was washed with toluene (2 portions of 0.2 mL) and dried under
vacuum, yielding white crystals. The parent ion was not observed
by ESMS: m/z 578.1 (Pt((R,R)-Me-Duphos)(Ph)+), 619.2, 651.2
(Pt((R,R)-Me-Duphos)(Ph)(SiMe3)+).
Pt((R,R)-Me-Duphos)(Ph)(OSiMe3) (5; 90% of the Mixture).
31P{1H} NMR (C6D6): δ 67.1 (d, J ) 4, JPt-P ) 1702), 46.6 (d, J
1
) 4, JPt-P ) 3619). The following H NMR (C6D6) spectrum is
reported as a mixture of 5 (90%) and hydroxide 6 (10%) unless
1
otherwise indicated. H NMR (C6D6): δ 8.03-7.94 (m, 2H, Ar),
Since the major diastereomer of phosphido intermediate 4
appeared to yield the major enantiomer of the product phosphine
3, catalytic enantioselectivity (Scheme 11) resulted mainly from
a thermodynamic preference for one of the diastereomers of 4.
Tuning sterics and electronics of the secondary phosphine
substrate and the catalyst might then provide a way to maximize
Keq for rational control of enantioselection. However, Scheme
11 and the experimental observation that product ee in the
alkylation of PHMe(Is) (2) depended on the benzyl bromide
substrate1 emphasize that the rate constants kS and kR were also
important in enantioselection.
7.37-7.31 (m, 2H, Ar), 7.22-6.94 (m, 5H, Ar), 3.44-3.37 (m,
1H, 6), 3.35-3.26 (m, 1H, 5), 3.16-3.07 (m, 1H), 2.40-2.25 (m,
1H), 2.21-1.86 (m, 5H), 1.68 (dd, J ) 19, 7, 3H, Me, 6), 1.61
(dd, J ) 18, 7, 3H, Me, 5), 1.56-1.34 (m, 2H), 1.24 (dd, J ) 18,
8, 3H, Me), 1.06-0.87 (m, 2H), 0.72 (dd, J ) 15, 8, 3H, Me),
0.65 (dd, J ) 16, 7, 3H, Me), 0.33 (9H). 13C{1H} NMR (C6D6):
δ 164.4 (dd, J ) 126, 8, quat, Pt-Ph), 147.0 (dd, J ) 47, 38, quat
Ar), 143.1 (dd, J ) 35, 24, quat Ar), 139.7 (broad, Ph), 133.2 (d,
J ) 12, Ar), 132.7 (dd, J ) 15, 3, Ar), 130.8-130.6 (m, Ar), 128.0
(d, J ) 7), 126.0, 123.9 (Ar), 40.51 (d, J ) 26), 40.48 (d, J ) 39),
37.5, 36.9 (d, J ) 4), 36.7, 35.5 (d, J ) 6), 33.70 (d, J ) 38),
33.69 (d, J ) 25), 17.5 (d, J ) 10, Me), 16.8 (d, J ) 6, Me), 14.44
(d, J ) 5, Me), 14.43 (d, J ) 5, Me), 6.0 (OSiMe3).
Experimental Section
Pt((R,R)-Me-Duphos)(Ph)(OH) (6). Method I. To a stirred
slurry of Pt((R,R)-Me-Duphos)(Ph)(Cl) (1; 61.4 mg, 0.1 mmol) in
MeOH (10 mL) was added NaOH (90 mg, 2.3 mmol). After 10
min the solvent was removed under reduced pressure and toluene
(10 mL) was added to the white residue. The resulting white slurry
was filtered through Celite, and the filtrate was concentrated under
vacuum. The 31P NMR spectrum of the filtrate indicated a mixture
of 6 (∼35%) and Pt((R,R)-Me-Duphos)(Ph)(OMe) (∼65%, see the
Supporting Information). The solvent was removed under vacuum,
and the white powder was redissolved in THF. Water (50 µL, 2.8
mmol) was added via a microliter syringe. After 10 min the solvent
was removed under vacuum and toluene (5 mL) was added to the
white powder. The 31P NMR spectrum of the solution indicated
only 6. The toluene was removed under vacuum, yielding 39 mg
(66%) of white crystals.
Unless otherwise noted, all reactions and manipulations were
performed in dry glassware under a nitrogen atmosphere at 20 °C
in a drybox or using standard Schlenk techniques. Petroleum
ether (bp 38-53 °C), ether, THF, toluene, and CH2Cl2 were
dried using columns of activated alumina.29 NMR spectra were
1
recorded using Varian 300 and 500 MHz spectrometers. H and
13C NMR chemical shifts are reported vs Me4Si and were
determined by reference to the residual 1H and 13C solvent
peaks. 31P NMR chemical shifts are reported vs H3PO4 (85%) used
as an external reference. Coupling constants are reported in Hz as
absolute values. Unless indicated, peaks in NMR spectra are
singlets. Elemental analyses were provided by Schwarzkopf Mi-
croanalytical Laboratory or Quantitative Technologies Inc. Mass
spectra were recorded at the University of Illinois Urbana-
Champaign.
Method II. To a stirred slurry of 1 (123 mg, 0.2 mmol) in THF
(10 mL) was added H2O (0.5 mL) and NaOH (80 mg, 2.0 mmol).
After 1 h the solvent was removed under reduced pressure and
toluene (10 mL) was added to the white residue. The resulting white
slurry was filtered through Celite, and the filtrate was concentrated
under vacuum. The 31P NMR spectrum of the filtrate indicated a
mixture of 6 (∼90%) and unreacted 1 (∼10%). THF (10 mL) was
(27) Quin, L. D. A Guide to Organophosphorus Chemistry, Wiley-
Interscience: New York, 2000; p 301.
(28) Note that the RP product was also favored in the alkylation of PHPh-
(Cy) (Cy ) cyclohexyl) to give PPh(Cy)(CH2Ph) using the same catalyst
precursor (1).1
(29) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518-1520.

Pt-Catalyzed Alkylation of a Secondary Phosphine
Organometallics, Vol. 26, No. 7, 2007 1797
added to the white powder, followed by H2O (0.5 mL) and NaOH
(80 mg, 2.0 mmol). After 12 h, the solvent was removed under
reduced pressure and toluene (10 mL) was added to the white
residue. The resulting white slurry was filtered through Celite, and
the filtrate was concentrated under vacuum. The 31P NMR spectrum
of the filtrate indicated a mixture of 6 (∼95%) and unreacted 1
(∼5%). The solvent was removed under vacuum, yielding 96 mg
(81%) of white powder.
PHMe(Is) was observed, 5 mg of PHMe(Is) (0.02 mmol) was added.
After 15 min phosphido complex 4 and phosphine 3 were the major
components of the mixture. Unreacted PHMe(Is) was also present.
A small amount of cation 8 could also be detected. After 45 min
the ratio 3:2 was 7.9:1. The mixture was unchanged the next day.
Stoichiometric Reaction of Pt((R,R)-Me-Duphos)(Ph)(PMeIs)
(4) with Benzyl Bromide. Synthesis of Pt((R,R)-Me-Duphos)-
(Ph)(Br) (7). A solution of Pt((R,R)-Me-Duphos)(Ph)(PMeIs) (4;
124.2 mg, 0.15 mmol) in toluene (5 mL) was transferred into an
NMR tube, which was fitted with a septum. The tube was cooled
in a NaCl/ice bath at -5 °C. Benzyl bromide (26 mg, 18 µL, 0.15
mmol) was added via a microliter syringe, and the reaction mixture
was kept at -15 °C for 10 h, when the reaction was completed,
according to 31P NMR. Complex 7 and PMeIs(CH2Ph) (3) were
the only components of the mixture observed in toluene. The solvent
was removed under vacuum, and 5 mL of petroleum ether was
added to the white solid. The solid was allowed to settle, and the
liquid was removed with a pipet. The white solid was further
washed with three 5 mL portions of petroleum ether, dried under
vacuum, and redissolved in CD2Cl2. Complex 7 (41%) and PMeIs-
(CH2Ph) (36%) were observed as the major components of the
mixture, along with 23% of 8-Br, which was independently
synthesized as the BF4 salt (see below). The solvent was removed
under vacuum, and 5 mL of a 9:1 mixture of petroleum ether and
THF was added to the residue. The resulting white precipitate was
washed with five 5 mL portions of 9:1 petroleum ether-THF. The
washes were collected, and the tertiary phosphine was isolated by
filtration on a silica column (5 cm height, 0.6 cm diameter), using
a 9:1 petroleum ether-THF mixture as eluent, followed by solvent
removal under vacuum, yielding 38 mg of a colorless oil (75%
yield). The white precipitate (7) was dried under vacuum and
recrystallized from THF/petroleum ether at -10 °C, yielding 80
mg (82%) of white crystals suitable for X-ray crystallography.
Similar experiments in which the ee of 3 was measured are
described in the Supporting Information.
Anal. Calcd for C24H34OP2Pt: C, 48.40; H, 5.75. Calcd for
C24H34OP2Pt‚0.05C24H33ClP2Pt: C, 48.33; H, 5.74. Found: C,
48.87; H, 5.76. The parent ion was not observed by ESMS: m/z
578.2 (Pt((R,R)-Me-Duphos)(Ph)+), 619.2, 791.4, 1182.5. 31P{1H}
NMR (C6D6): δ 67.6 (d, J ) 5, JPt-P ) 1803), 48.7 (d, J ) 5,
JPt-P ) 3218). 1H NMR (C6D6): δ 7.99 (t, J ) 13, JPt-P ) 43, 2H,
Ar), 7.35-7.32 (m, 2H, Ar), 7.22-7.19 (m, 1H, Ar), 7.14-7.08
(m, 2H, Ar), 7.01-6.97 (m, 2H, Ar), 3.49-3.43 (m, 1H), 3.16-
3.07 (m, 1H), 2.40-2.32 (m, 1H), 2.19-2.10 (m, 2H), 2.03-1.91
(m, 2H), 1.82 (broad, 1H), 1.68 (dd, J ) 19, 7, 3H, Me), 1.63-
1.32 (m, 2H), 1.31-1.21 (m, 1H), 1.21 (dd, J ) 18, 7, 3H, Me),
1.11-0.92 (m, 2H), 0.75 (dd, J ) 15, 7, 3H, Me), 0.68 (dd, J )
16, 8, 3H, Me). 13C{1H} NMR (C6D6): δ 166.8 (dd, J ) 121, 7,
quat, Pt-Ph), 147.1 (dd, J ) 45, 38, quat Ar), 143.5 (dd, J ) 35,
25, quat Ar), 138.5 (JPt-C ) 22, Ph), 133.3 (d, J ) 13, Ar), 132.8
(dd, J ) 14, 3, Ar), 130.8-130.6 (m, Ar), 124.2 (Ar), 40.2 (d, J )
27), 39.8 (d, J ) 36), 37.5, 36.71, 36.67 (d, J ) 4), 35.6 (d, J )
6), 34.3 (d, J ) 25), 34.0 (d, J ) 36), 18.1 (d, J ) 11, Me), 17.3
(d, J ) 6, Me), 14.49 (d, J ) 4, Me), 14.47 (d, J ) 3, Me). Note:
hydroxide complex 6 took up CO2 from the air to yield (Pt((R,R)-
Me-Duphos)(Ph))2(µ-CO3), which was characterized crystallo-
graphically (see the Supporting Information).
Stoichiometric Reaction of Pt((R,R)-Me-Duphos)(Ph)(Cl)
(1) with NaOSiMe3/PHMe(Is). Generation of Pt((R,R)-Me-
Duphos)(Ph)(PMeIs) (4) in Situ. A slurry of NaOSiMe3 (11 mg,
0.1 mmol) in 0.5 mL of toluene was added to 1 (61.4 mg, 0.1
mmol). The resulting colorless solution was transferred into
an NMR tube. The components of the mixture, observed by 31P
NMR spectroscopy, were chloride 1 and silanolate 5 (3.5:1). A
very small amount of hydroxide 6 could also be detected. The
solution was added to PHMe(Is) (2; 25 mg, 0.1 mmol); the
mixture turned yellow immediately. Complex 4 was the major
component along with a small amount of 5, indicating that a
substoichiometric amount of secondary phosphine was added. The
mixture was added to 6 mg of PHMe(Is) (0.024 mmol); then, 4
was the only Pt species observed, along with a small excess of
PHMe(Is).
Anal. Calcd for BrC24H33P2Pt: C, 43.78; H, 5.05. Found: C,
43.85; H, 4.97. 31P{1H} NMR (CD2Cl2): δ 66.0 (d, J ) 6, JPt-P
)
1647), 56.8 (d, J ) 6, JPt-P ) 3960). 1H NMR (CD2Cl2): δ 7.78-
7.73 (m, 1H), 7.66-7.56 (m, 3H), 7.47 (t, J ) 7, JPt-H ) 36, 2H,
Ar), 7.10 (td, J ) 8, 2, 2H), 6.88 (t, J ) 7, 1H), 3.49-3.40 (m,
1H), 3.32-3.16 (m, 1H), 2.86-2.74 (m, 1H), 2.74-2.53 (m, 1H),
2.44-2.24 (m, 2H), 2.13-1.97 (m, 2H), 1.94-1.82 (m, 1H), 1.82-
1.72 (m, 1H), 1.64-1.54 (m, 1H), 1.45 (dd, J ) 19, 7, 3H, Me),
1.19 (dd, J ) 19, 7, 3H, Me), 0.94-0.87 (m, 6H, Me), 0.84-0.74
(m, 1H). 13C{1H} NMR (CD2Cl2): δ 138.3 (Ph), 132.9 (d, J ) 13,
Ar), 132.6 (m, Ar), 131.6-131.4 (m, Ar), 127.7 (d, J ) 7, Ph),
122.9 (Ph), 41.7 (d, J ) 7), 41.5, 37.6, 37.0 (d, J ) 3), 36.5, 35.8
(d, J ) 29), 35.2 (d, J ) 6), 33.2 (d, J ) 38), 17.2 (d, J ) 9, Me),
16.0 (d, J ) 5, Me), 14.3 (d, J ) 19, 2Me). The quaternary aryl
carbons were not observed.
Reaction of Pt((R,R)-Me-Duphos)(Ph)(PMeIs) (4) Generated
in Situ with 1 Equiv of PhCH2Br Followed by Treatment with
NaOSiMe3/PHMe(Is). To the above mixture was added benzyl
bromide (17 mg, 12 µL, 0.1 mmol) via a microliter syringe. The
mixture became light yellow in less than 30 min. Only Pt((R,R)-
Me-Duphos)(Ph)(Br) (7) and PMeIsCH2Ph (3) were observed by
31P NMR spectroscopy, but the light yellow color was consistent
with the presence of a small amount of the phosphido complex
(see below). This reaction mixture was added to NaOSiMe3 (11
mg, 0.1 mmol); it remained yellow. After 1 h, complexes 7 and
Pt((R,R)-Me-Duphos)(Ph)(OSiMe3) (5) (12:1) were the major Pt
species observed, along with 3. Small amounts of 4, cation 8, and
hydroxide 6 could also be detected. The presence of 4 was probably
due to addition of a substoichiometric amount of PhCH2Br in the
previous step; therefore, after 1 day 2 µL of PhCH2Br (3 mg, 0.017
mmol) was added. The color faded in ∼5 min. Complexes 7, 5,
and 6 (ratio 12.8:17.1:1), along with a small amount of 8, and
phosphine 3 were observed.
Stoichiometric Reaction of Pt((R,R)-Me-Duphos)(Ph)(PMeIs)
(4) with Benzyl Bromide, Followed by Treatment with PHMe-
(Is)/NaOSiMe3. An orange solution of 4 (33 mg, 0.04 mmol) in
toluene (5 mL) was transferred into an NMR tube, which was
fitted with a septum. Benzyl bromide (7 mg, 5 µL, 0.04 mmol)
was added via a microliter syringe. After 10 min the color bleached
and the reaction was almost complete, according to 31P NMR
spectroscopy. Bromide 7 and PMeIs(CH2Ph) (3) were observed,
plus a trace of 4.
The reaction mixture was added to 1 equiv of PHMe(Is) (10
mg, 0.04 mmol). Initially, the solution turned yellow, but after ∼30
s the color bleached again. The 31P NMR spectrum showed 7, 3,
and 2 (the ratio of tertiary to secondary phosphine was ∼1:1). The
mixture was added to 1 equiv of NaOSiMe3 (5 mg, 0.04 mmol).
The mixture turned orange, and a small amount of precipitate was
observed. This mixture contained 4 (∼60% of the Pt species) and
3, along with 7 and a small amount of 8-Br (∼10% of the Pt
The solution was added to PHMe(Is) (25 mg, 0.1 mmol). The
mixture turned yellow immediately. After 5 min 31P NMR monitor-
ing showed the presence of 4, 7, and 3, along with a small amount
of 8. After 15 min some silanolate 5 was also detected. Since no

1798 Organometallics, Vol. 26, No. 7, 2007
Scriban et al.
species). This is consistent with the presence of a slight excess of
benzyl bromide, which was shown to react with 4 under similar
conditions to yield 7 and cation 8.
solvent was removed with a pipet, and the precipitate was dried
under vacuum, yielding 90 mg (95%) of white powder. (Note: if
a trace of silver salts remained in the nitrile complex precursor
(see above), then a small amount of what appears to be the silver-
phosphine complex [Ag(PMeIs(CH2Ph))]+ was also formed (31P-
Generation of Pt((R,R)-Me-Duphos)(Ph)(PMeIs) (4) and Its
Reactions with Benzyl Bromide and PHMe(Is)/NaOSiMe3. A
solution of PHMe(Is) (10 mg, 0.04 mmol) in 0.2 mL of toluene
was added to a suspension of NaOSiMe3 (5 mg, 0.04 mmol) in 0.2
mL of toluene. The resulting colorless solution was added to a
suspension of Pt((R,R)-Me-Duphos)(Ph)(Cl) (1; 25 mg, 0.04 mmol)
in 0.1 mL of toluene. The mixture turned yellow immediately,
indicating the formation of 4. The mixture was transferred into an
NMR tube, which was fitted with a septum. Benzyl bromide (7
mg, 5 µL, 0.04 mmol) was added via a microliter syringe. The
color bleached in less than 5 min, and the major components of
the mixture were 7 and 3, plus small amounts of 8-Br and 4,
presumably because a substoichiometric amount of benzyl bromide
was added.
{1H} NMR (CD2Cl2): δ -17.4, J
) 528, 616).) Recrys-
107/109Ag-P
tallization from CH2Cl2/Et2O gave crystals suitable for X-ray
crystallography and elemental analyses. Anal. Calcd for C47H66P3-
PtBF4: C, 56.12; H, 6.61. Found: C, 55.74; H, 6.70. HRMS: m/z
calcd for C47H66P3Pt+ (M+), 917.4004; found, 917.4011.
31P NMR spectroscopy showed a mixture of the two diastereo-
mers a and b (the ratio of a to b was ∼10:1). The following NMR
spectra are reported for the major diastereomer a, unless otherwise
indicated. 31P{1H} NMR (CD2Cl2, 21 °C): diastereomer a, δ 64.7
(broad, JPt-P ) 1705), 58.2 (dd, J ) 370, 6, JPt-P ) 2443), 0.1 (d,
J ) 370, JPt-P ) 2500); diastereomer b, δ 58.25 (dd, J ) 366, 8,
1
JPt-P ) 2459), 59.2 (broad, JPt-P ≈ 1730), -5.7 (d, J ) 366). H
NMR (CD2Cl2, 21 °C): δ 7.99-7.95 (m, 1H), 7.81-7.75 (m, 1H),
7.75-7.66 (m, 3H), 7.34 (m, JPt-P ) 35, 1H), 7.26 (broad, 1H),
7.18 (t, J ) 8, 1H), 7.16-7.11 (m, 1H), 7.07 (t, J ) 8, 2H), 6.85
(t, J ) 7, 1H), 6.77 (broad, 1H), 6.65 (broad, 1H), 6.45 (1H, Ar),
6.44 (1H, Ar), 4.22 (broad, 1H), 3.94 (broad, 1H, P-CH2), 3.22
(broad d, J ) 12, 1H, P-CH2), 3.20-3.06 (overlapping m, 2H),
3.00-2.83 (m, 2H), 2.71-2.53 (broad m, 2H), 2.29-2.16 (broad
m, 1H), 2.14-1.88 (m, 5H), 1.85-1.74 (m, 2H), 1.55-1.47 (broad
m, 9H, CH3), 1.292 (d, J ) 7, 3H, CH3, Is), 1.288 (d, J ) 7, 3H,
CH3, Is), 1.19 (dd, J ) 19, 8, 6H, CH3), 1.28-1.17 (m, 3H, P-CH3,
overlapping the other peaks), 0.80 (dd, J ) 16, 7, 3H, CH3), 0.36
(broad, 3H, CH3), 0.12 (broad, 3H, CH3). 13C{1H} NMR (CD2Cl2,
21 °C): δ 155.2 (m, quat), 152.7 (m, quat), 151.0-150.7 (m, quat),
141.6-141.3 (m, quat), 138.8-138.4 (m, quat), 134.3-133.9 (m,
Ar), 133.6 (d, J ) 5, Ar), 133.5 (d, J ) 5, Ar), 133.2 (broad, Ar),
133.1-132.8 (m, Ar), 132.6 (broad, Ar), 130.1 (m, Ar), 128.6 (m,
Ar), 128.0-127.6 (broad m, Ar), 124.8 (broad, Ar), 123.7 (m, Ar),
123.4 (m, Ar), 44.8-44.1 (m), 42.9-42.3 (m), 41.1-40.6 (m), 39.7
(dd, J ) 26, 3), 37.9-37.6 (broad m), 37.1 (broad), 36.1, 35.0-
34.8 (m), 34.4 (d, J ) 7) overlapping with 34.6-34.0 (broad), 33.0
(dd, J ≈ 35, 5, CH), 29.6 (broad), 27.9 (broad), 25.6, 25.1-24.8
(broad m), 23.9 (d, J ) 4) overlapping with 24.1-23.6 (broad m),
18.9-18.6 (m), 15.5-15.1 (m), 14.8 (d, J ) 6), 14.6, 13.8-13.2
(m). 31P{1H} NMR (CD2Cl2, -20 °C): diastereomer a, δ 64.4 (dd,
J ) 14, 6, JPt-P ) 1695), 57.9 (dd, J ) 368, 6, JPt-P ) 2434), 0.9
(dd, J ) 368, 14, JPt-P ) 2490); diastereomer b, δ 58.6 (dd, J )
17, 7, JPt-P ) 1731), 58.1 (dd, J ) 364, 7, JPt-P ) 2448), -5.7 (d,
J ) 364, JPt-P ) 2563, overlapping the satellite peaks of
NaOSiMe3 (5 mg, 0.04 mmol) and PHMe(Is) (10 mg, 0.04 mmol)
were added. The mixture turned yellow immediately, and 31P NMR
spectroscopy showed that 4 and 3 were the main components of
the mixture, along with a small amount of 8-Br and 2 (the ratio of
3 to 2 was 1.9:1). Evidently a slight excess of the secondary
phosphine was added.
[Pt((R,R)-Me-Duphos)(Ph)(NCMe)][BF4]. To a stirred solution
of Pt((R,R)-Me-Duphos)(Ph)(Cl) (1; 92.1 mg, 0.15 mmol) in MeCN
(15 mL) was added AgBF4 (29.2 mg, 0.15 mmol). An off-white
precipitate, which turned black over time, formed immediately. The
mixture was filtered through Celite, yielding a pale brown solution.
The solvent was removed under vacuum, and petroleum ether (5
mL) was added to the pale brown residue. The solvent was removed
with a pipet, and the pale brown precipitate was dried under
vacuum, yielding 105 mg (99%) of off-brown powder. The solid
was redissolved in CH2Cl2 and filtered through Celite, yielding a
colorless solution. The solvent was removed under vacuum, and
the white residue was washed with 5 mL of petroleum ether and
dried under vacuum, yielding an off-white powder. This complex
was not obtained in analytically pure form, presumably because of
the presence of trace silver salts (brown color), but it could be used
successfully (either as generated, or after isolation) to prepare the
cation 8-BF4 by displacement of MeCN with the phosphine PMeIs-
(CH2Ph).
Anal. Calcd for C26H36NP2PtBF4: C, 44.21; H, 5.14; N, 1.98.
Calcd for C26H36NP2PtBF4‚0.05AgCl: C, 43.76; H, 5.09; N, 1.96.
Found: C, 43.55; H, 5.08; N, 1.57. HRMS: m/z calcd for (C26H36-
NP2Pt)+ (M+), 618.1950; found, 618.1949. 31P{1H} NMR (CD2-
1
diastereomer a). H NMR (CD2Cl2, -20 °C): δ 7.96-7.93 (m,
Cl2): δ 71.4 (d, J ) 4, JPt-P ) 1656), 50.5 (d, J ) 4, JPt-P
)
1
1H), 7.80-7.74 (m, 1H), 7.74-7.63 (m, 3H), 7.39-7.27 (m, 1H),
7.24 (m, 1H), 7.16 (t, J ) 8, 1H), 7.14-7.07 (m, 1H), 7.07-6.96
(m, 2H), 6.83 (t, J ) 8, 1H), 6.73-6.67 (broad m, 1H), 6.65-6.60
(broad m, 1H), 6.39 (1H, Ar), 6.38 (1H, Ar), 4.24-4.14 (m, 1H),
3.90 (dd, J ) 14, 7, 1H, P-CH2), 3.16 (dm, J ) 14, 1H, P-CH2),
3.18-3.00 (overlapping m, 2H), 3.00-2.75 (m, 2H), 2.70-2.47
(m, 2H), 2.24-2.15 (m, 1H), 2.08-1.82 (m, 5H), 1.81-1.70 (m,
2H), 1.56 (d, J ) 7, 3H, CH3, Is), 1.51 (dd, J ) 19, 7, 3H, CH3),
1.46 (d, J ) 6, 3H, CH3, Is), 1.262 (d, J ) 7, 3H, CH3, Is), 1.255
(d, J ) 7, 3H, CH3, Is), 1.19 (dd, J ) 16, 7, 3H, CH3), 1.17 (dd,
J ) 19, 7, 3H, CH3), 1.23-1.09 (m, 3H, P-CH3, obscured by the
other peaks), 0.77 (dd, J ) 16, 7, 3H, CH3), 0.33 (d, J ) 7, 3H,
CH3, Is), 0.02 (d, J ) 7, 3H, CH3, Is).
Reaction of 8-BF4 with Excess [NOct4][Br]. A solution of
[NOct4][Br] (41 mg, 0.075 mmol, 5 equiv) in 0.3 mL of toluene
was added to a slurry of 8-BF4 (15 mg, 0.015 mmol, highly
diastereomerically enriched) in toluene (0.2 mL). The mixture
became a colorless homogeneous solution immediately. It was
transferred into an NMR tube and monitored by 31P NMR
spectroscopy. After 30 min, the bromide 7 and PMeIs(CH2Ph) (3)
were the only components of the mixture.
4085). H NMR (CD2Cl2): δ 7.84-7.78 (m, 1H), 7.75-7.68 (m,
3H), 7.50-7.46 (m, JPt-H ) 35, 2H), 7.24-7.20 (m, 2H), 7.04 (t,
J ) 8, 1H), 3.24-3.12 (m, 1H), 3.02-2.89 (m, 2H), 2.76-2.64
(m, 1H), 2.57-2.42 (m, 2H), 2.44 (3H, NCCH3), 2.14-2.04 (m,
1H), 2.02-1.89 (m, 1H), 1.88-1.77 (m, 1H), 1.72-1.57 (m, 2H),
1.41 (dd, J ) 19, 7, 3H, CH3), 1.21 (dd, J ) 19, 7, 3H, CH3),
0.98-0.89 (m, 6H, 2CH3), 0.82-0.70 (m, 1H). 13C{1H} NMR
(CD2Cl2): δ 155.6 (dd, J ) 101, 8, quat), 141.5 (dd, J ) 54, 36,
quat), 138.5 (dd, J ) 42, 21, quat), 137.4 (Ar), 133.6 (d, J ) 11,
Ar), 133.0 (d, J ) 5, Ar), 132.9-132.8 (m, Ar), 128.5 (d, J ) 6,
JPt-C ) 42, Ar), 124.6 (Ar), 123.9 (d, J ) 6, CN), 41.8 (d, J ) 42,
CH), 40.7 (d, J ) 29, CH), 37.1 (overlapping d, J ) 28, CH +
CH2), 36.8 (d, J ) 5, CH2), 36.4 (CH2), 35.0 (d, J ) 6, CH2), 33.7
(d, J ) 40, CH), 17.7 (d, J ) 8, CH3), 16.0 (d, J ) 3, JPt-C ) 40,
CH3), 14.2 (d, J ) 3, CH3), 14.0 (d, J ) 2, CH3), 3.7 (NCCH3).
[Pt((R,R)-Me-Duphos)(Ph)(PMeIs(CH2Ph))][BF4] (8-BF4). To
a stirred solution of [Pt((R,R)-Me-Duphos)(Ph)(NCMe)][BF4] (66.4
mg, 0.94 mmol) in MeCN (15 mL) was added PMeIs(CH2Ph) (31
mg, 0.94 mmol, prepared catalytically at -5 °C, 82% ee). The
solvent was removed under vacuum, and petroleum ether (5 mL)
was added to the white residue. A white precipitate formed. The

Pt-Catalyzed Alkylation of a Secondary Phosphine
Organometallics, Vol. 26, No. 7, 2007 1799
Formation of [Pt(dppe)(Me)(PPh2*Me)][I] from the Reaction
of Pt(dppe)(Me)(PPh2) and 13C-Labeled *MeI. A yellow solution
of Pt(dppe)(Me)(PPh2) (39.7 mg, 0.05 mmol) in toluene-d8 (0.5
mL) was transferred to an NMR tube, which was fitted with a
septum. 13CH3I (3.1 µL, 7.1 mg, 0.05 mmol) was added via a
microliter syringe. A white precipitate formed immediately, and
the solution turned colorless. 31P NMR spectroscopy (toluene-d8)
NMR signals. A very small amount of PMeIs(CH2Ph) (3) was
observed after 2 days. The solution, which contained mostly PHMe-
(Is), was decanted, and the solid was dried under vacuum, yielding
∼8 mg of white powder, which was dissolved in 0.5 mL of CH2-
Cl2. The solution was transferred into an NMR tube; it consisted
of unreacted 8-BF4 and PHMe(Is), according to 31P NMR spec-
troscopy. No reaction was observed after 4 days. PHMe(Is) (5 mg,
0.02 mmol) was added, and the solvent was removed under vacuum,
yielding a viscous colorless residue, to which a slurry of NaOSiMe3
(2 mg, 0.02 mmol) in toluene (0.5 mL) was added. The mixture
turned yellow in ∼30 s. After 1 h phosphido 4 was the only Pt
species observed, along with 3 and unreacted 2, indicating that
excess PHMe(Is) (2) was present.
1
did not show any signal, and H and 13C NMR spectra showed
only signals due to toluene-d8. The solvent was removed with a
pipet, and the white precipitate was washed with toluene (0.5 mL)
and dried under vacuum to give 45 mg (97% yield) of white powder,
which was characterized by multinuclear NMR spectroscopy.
31P{1H} NMR (CD2Cl2): δ 54.1 (ddd, J ) 384, 6, JP-C ) 3,
JPt-P ) 2690), 49.7 (ddd, J ) 19, 6, JP-C ) 4, JPt-P ) 1776), 8.2
(ddd, J ) 384, 19, JP-C ) 36, JPt-P ) 2726), consistent with the
literature data for the PF6 salt in CH2Cl2/C6D6.23 1H NMR (CD2-
Cl2): δ 7.70-7.56 (m, 10H, Ar), 7.56-7.51 (m, 2H, Ar), 7.51-
7.40 (m, 10H, Ar), 7.36-7.29 (m, 4H, Ar), 7.29-7.21 (m, 4H,
Ar), 2.39 (apparent dt, J ) 19, 11, 3H, P-*Me), 2.08-1.97 (m,
2H), 1.82-1.71 (m, 2H), 0.48 (apparent q, J ) 7, JPt-H ) 59, 3H,
Stoichiometric Reaction of [Pt((R,R)-Me-Duphos)(Ph)(PHMe-
(Is))][BF4] (12-BF4) with PMeIs(CH2Ph). A solution of PMeIs-
(CH2Ph) (9 mg, 0.025 mmol) in 0.3 mL of toluene was added to a
slurry of 12-BF4 (23 mg, 0.025 mmol) in toluene (0.2 mL). The
mixture was monitored by 31P NMR spectroscopy; the cation was
soluble enough to give weak NMR signals. After 45 min the only
Pt species observed was 12-BF4. A broad peak was also observed
for PMeIs(CH2Ph). The mixture remained unchanged after two
weeks.
Pt-Me). 13C{1H} NMR (CD2Cl2): δ 133.7 (d, J ) 11, JPt-C
)
22, Ar), 133.5 (d, J ) 12, JPt-C ) 14, Ar), 132.58 (d, J ) 11, JPt-C
) 20, Ar) overlapping with 132.58 (d, J ) 2, Ar), 132.3 (d, J ) 2,
Ar), 131.4 (d, J ) 2, Ar), 130.6 (d, J ) 55, quat Ar), 129.78 (d, J
) 11, Ar), 129.75 (d, J ) 10, Ar), 129.1 (d, J ) 11, Ar), 128.2 (d,
J ) 49, quat Ar), 127.1 (d, J ) 55, quat Ar), 32.0-31.5 (m, CH2),
29.3-28.7 (m, CH2), 13.8 (very intense apparent dt, J ) 36, 3,
JPt-C ) 29, P-*Me), 1.00 (dm, J ) 73, Pt-Me).
31P NMR Monitoring of the Catalytic Reaction of Benzyl
Bromide with PHMe(Is). To PHMe(Is) (150 mg, 0.6 mmol) in
0.2 mL of toluene was added NaOSiMe3 (67.3 mg, 0.6 mmol)
suspended in 0.2 mL of toluene. The mixture was added to catalyst
precursor 1 (18 mg, 0.03 mmol, 5 mol %) in 0.1 mL of toluene.
The reaction mixture was transferred to an NMR tube. Benzyl
bromide (71 µL, 103 mg, 0.6 mmol) was added via microliter
syringe, and the reaction mixture was monitored by 31P NMR
spectroscopy. After 5 min, and again after 1.5 h (∼10% conversion),
8-Br was observed as the resting state of the catalyst. An increasing
amount of white precipitate was observed as the reaction progressed.
The reaction was complete after 18 h, when 31P NMR monitoring
showed the presence of 7 along with several other unidentified
species. In a similar experiment on half the scale, the same resting
state was also observed later in the reaction (∼30% conversion).
Stoichiometric Reaction of [Pt((R,R)-Me-Duphos)(Ph)(PMeIs-
(CH2Ph))][BF4] (8-BF4) with NaOSiMe3/PHMe(Is). A solution
of 8-BF4 (20 mg, 0.02 mmol, containing ∼1% of P(O)MeIs(CH2-
Ph) impurity) in THF (0.2 mL) was added to NaOSiMe3 (2 mg,
0.02 mmol). Formation of PMeIs(CH2Ph) was observed as early
as 30 min by 31P NMR. After 1 day the amount of PMeIs(CH2Ph)
increased, and silanolate 5 and hydroxide 6 were observed, along
with unreacted 8-BF4 (6.8:1:11.4 ratio). The solution was light
purple. After 3 days the ratio became 3:1:1.9. The mixture was
added to PHMe(Is) (5 mg, 0.02 mmol). The mixture changed to
yellow in ∼30 s. After 15 min 4 was already observed, and the
ratio of PMeIsCH2Ph to PHMe(Is) was 1.8:1. After 1 h the ratio
became 2:1, and peaks belonging to both 4 and unreacted 8-BF4
could be observed. A small amount of [Pt((R,R)-Me-Duphos)(Ph)-
(PHMe(Is))][BF4] (12-BF4) could be detected, indicating that a
substoichiometric amount of base was added. After 3 h the ratio of
3 to 2 became 2.8:1, and 4 and 12 were the major Pt species. A
very small amount of unreacted 8-BF4 was also observed; it was
consumed after 1 day, when the ratio of 3 to 2 was 4:1.
Stoichiometric Reaction of [Pt((R,R)-Me-Duphos)(Ph)(PMeIs-
(CH2Ph))][BF4] (8-BF4) with NaOSiMe3 in Toluene. A slurry of
highly diastereomerically enriched 8-BF4 (20 mg, 0.02 mmol) in
toluene (0.4 mL) was added to NaOSiMe3 (2 mg, 0.02 mmol) in
0.1 mL of toluene. Formation of PMeIs(CH2Ph) was observed as
early as 45 min. After 2 days solid 8-BF4 was still present, but the
amount of PMeIs(CH2Ph) increased, and silanolate 5 was observed,
along with unreacted 8-BF4 (ratio 1.9:1).
[Pt((R,R)-Me-Duphos)(Ph)(PMeIs(CH2Ph))][Br] (8-Br). 31P-
{1H} NMR (toluene): diastereomer a, δ 64.5 (JPt-P ) 1692), 58.1
(d, J ) 369, JPt-P ) 2498), -0.2 (d, J ) 369, JPt-P ) 2507);
diastereomer b, δ 58.5 (broad), 58.2 (d, J ) 369, JPt-P ) 2436),
-6.1 (d, J ) 369).
Stoichiometric Reaction of Pt((R,R)-Me-Duphos)(Ph)(Br) (7)
with PHMe(Is), PMeIs(CH2Ph), and NaOSiMe3. A solution of
PHMe(Is) (10 mg, 0.04 mmol) in 0.3 mL of toluene was added to
a solution of 7 (26.3 mg, 0.04 mmol) in toluene (0.2 mL). The
mixture was transferred into an NMR tube and monitored by 31P
NMR spectroscopy. No reaction was observed after 1 h. The
solution was added to PMeIs(CH2Ph) (13.6 mg, 0.04 mmol).
Monitoring by 31P NMR spectroscopy showed no reaction, even
after 1 day. The mixture was then added to NaOSiMe3 (4.5 mg,
0.04 mmol). The solution turned yellow immediately. After 5 min
the only Pt species observed by 31P NMR spectroscopy was 4. The
ratio of 2 to 3 was ∼1:2.6, indicating that there was an excess of
PHMe(Is).
NMR Structure Determination for Pt((R,R)-Me-Duphos)(Ph)-
(PMeIs) (4). As described above, one- and two-dimensional NMR
1
spectroscopy was used to assign the H and 13C NMR spectra of
the major diastereomer of 4 in THF-d8 at -60 °C. Chart 2 shows
the assignments; note that the assignment of some signals remains
ambiguous, as explained in the footnote. A list of these spectro-
scopic data, including coupling constants, with atom labels as in
the solid-state structure (Figure 2) is also included. For details of
the NOESY analysis, see the Supporting Information.
1H NMR (-60 °C, THF-d8; the assignments are as in Table S1
and Chart 2, but see the text and the chart footnote for comments
on ambiguous assignments): δ 7.87-7.86 (m, 1H, Ar, Duphos,
H15), 7.78-7.72 (m, 2H, overlapping Ar, Duphos H16 and Ph ortho
H6), 7.56-7.55 (m, 2H, Ar, Duphos, H14/H17), 7.25 (broad t, J
) 7, 1H, Ph ortho, H2), 7.07 (broad t, J ) 8, 1H, Ph meta, H5),
7.00 (broad t, J ) 8, 1H, Ph meta, H3), 6.99 (2H, Is meta, H30/
H35), 6.78 (broad t, J ) 8, 1H, Ph para, H4), 4.92-4.87 (m, 1H,
CH, Is, H27), 4.72-4.70 (m, 1H, CH, Is, H37), 2.94-2.81 (m,
Stoichiometric Reaction of [Pt((R,R)-Me-Duphos)(Ph)(PMeIs-
(CH2Ph))][BF4] (8-BF4) with PHMe(Is)/NaOSiMe3. A solution
of PHMe(Is) (5 mg, 0.02 mmol) in 0.3 mL of toluene was added
to a slurry of 8-BF4 (20 mg, 0.02 mmol, highly diastereomerically
enriched) in toluene (0.2 mL). The mixture was monitored by 31P
NMR spectroscopy; the cation was soluble enough to give weak

1800 Organometallics, Vol. 26, No. 7, 2007
Scriban et al.
3H, overlapping 2CH Duphos (H7/H10) and CH Is (H32)), 2.44-
2.39 (m, 1H, CH Duphos, H19), 1.99-1.77 (m, 4H, 2CH2), 1.64-
1.53 (m, 4H, CH2), 1.49 (dd, J ) 17, 7, 3H, CH3 Duphos, H20),
1.39 (dd, J ) 17, 10, 3H, CH3 Duphos, H12), 1.36-1.29 (m, 7H,
overlapping 2CH3 Is (H29/H38) and 1CH Duphos (H23)), 1.23
(broad d, J ) 7, 9H, CH3 Is, H39/H33/H34), 1.03 (d, J ) 7, 3H,
CH3 Is, H28), 0.77-0.73 (m, 6H, overlapping CH3 Duphos (H11)
and P-Me (H40), JPt-H ) 57), 0.61 (dd, J ) 14, 7, 3H, CH3
Duphos, H24). 13C{1H} NMR (-60 °C, THF-d8): δ 163.0 (d, J )
95, JPt-C ) 803, Pt-C (quat Ph, C1)), 155.6 (quat Is, C31), 154.0
(d, J ) 26, P-C (quat Is, C25)), 148.1 (2 quat Is, C26/C36), 147.2-
146.6 (m, quat Duphos, C18), 144.1-143.3 (m, quat Duphos, C13),
140.2 (d, J ) 25, Ph ortho, C6), 137.9 (Ph ortho, C2), 135.2-
134.6 (m, Ar, Duphos, C16), 134.5-134.0 (m, Ar, Duphos, C15),
131.9 (d, J ) 25, Ar, Duphos, C17), 131.3 (d, J ) 25, Ar, Duphos,
C14), 128.3 (d, J ) 30, Ph meta, C5), 128.1 (d, J ) 32, Ph meta,
C3), 122.1 (d, J ) 16, Ph para, C4), 121.8 (d, J ) 10, Is meta,
C30), 120.0 (m, Is meta, C35), 45.1-44.7 (m, CH Duphos, C19),
42.4 (d, J ) 32, CH Duphos, C10), 38.1 (broad, CH2), 36.9-35.7
(broad, overlapping CH2), 35.5 (d, J ) 32, CH, Is, C32), 35.1 (d,
J ) 23, CH, Is, C37), 33.4 (d, J ) 34, CH, Is, C27), 32.9 (d, J )
33, CH, Duphos, C7), 32.6 (d, J ) 23, CH, Duphos, C23), 28.3
(CH3, Is, C28), 28.1 (CH3, Is, C39), 27.3 (CH3, Is, C38), 25.1 (CH3,
Is, C33), 24.9 (CH3, Is, C34), 22.3 (CH3, Is, C29), 18.8-18.2 (m,
CH3, Duphos, C20), 16.3-15.9 (m, CH3, Duphos, C12), 15.5 (d, J
) 15, CH3, Duphos, C24), 15.1 (d, J ) 13, CH3, Duphos, C11),
12.4 (broad, P-CH3, C40).
Acknowledgment. We thank the National Science Founda-
tion for support and Karen Goldberg (University of Washington)
for a helpful suggestion.
Note Added after ASAP Publication. In the version of this
paper published on the Web on February 20, 2007, the chemical
formulas on the last line of the right-hand column on the second
page of the paper were incorrect, due to a production error. The
version that now appears is correct.
Supporting Information Available: Text, figures, and tables
giving more experimental and characterization data, details of the
X-ray structure determinations, and additional information on the
NMR structure determination of complex 4 and CIF files giving
crystal data. This information is available free of charge via the
Internet at http://pubs.acs.org.
OM061116S