NMR studies of chiral P,S-chelate platinum, rhodium, and iridium complexes and the x-ray structure of a palladium (II) allyl derivative

Article abstract of DOI:10.1021/om960823q

Several Rh(I), Ir(III), and Pt(II) complexes of the chiral P,S-bidentate ligand 2 have been prepared and characterized. Detailed two-dimensional NMR studies show that (i) the boat-type chelate ring and the stereogenic sulfur center can invert rapidly at a

Full text of DOI:10.1021/om960823q

Organometallics 1997, 16, 579-590
579
NMR Stu d ies of Ch ir a l P ,S-Ch ela te P la tin u m , Rh od iu m ,
a n d Ir id iu m Com p lexes a n d th e X-r a y Str u ctu r e of a
P a lla d iu m (II) Allyl Der iva tive
Alberto Albinati,† J u¨rgen Eckert,‡ Paul Pregosin,*,§ Heinz Ru¨egger,§
Renzo Salzmann,§ and Corinna Sto¨ssel§
Laboratory of Inorganic Chemistry, ETH-Zentrum, CH-8092 Zu¨rich, Switzerland,
Institute of Pharmaceutical Chemistry, University of Milan, I-20131 Italy,
and Los Alamos National Laboratory, Los Alamos, New Mexico
Received September 25, 1996X
Several Rh(I), Ir(III), and Pt(II) complexes of the chiral P,S-bidentate ligand 2 have been
prepared and characterized. Detailed two-dimensional NMR studies show that (i) the boat-
type chelate ring and the stereogenic sulfur center can invert rapidly at ambient temperature
and (ii) the sulfur donor may dissociate, essentially destroying the chiral pocket. The solid-
state structure of [Pt(η3-C3H5)(2)]PF6 (3) has been determined and the sulfur substituent
shown to have an axial orientation. The six-membered chelate ring takes up a boatlike
conformation. As shown by an X-ray diffraction study for 3, and via incoherent inelastic
neutron scattering (IINS) measurements for the Pd analog, 4, the OH group is remote from
the metal atom.
In tr od u ction
in which the S-donor arises from a thioglucose, e.g. 1,
can have some utility.
Chiral chelating ligands are now widely employed in
enantioselective homogeneous catalysis.1 Initially,
diphosphine ligands predominated;2 however, in recent
studies mixed donor ligands have been shown to be
advantageous.3-7 Specifically, the development of P,O-
and P,N-donor ligands has led to high observed enan-
tiomeric excesses in allylic alkylation and Heck chem-
istry.8 We have recently shown9 that a P,S combination,
† University of Milan.
‡ Los Alamos National Laboratory.
§ ETH Zu¨rich.
X Abstract published in Advance ACS Abstracts, February 1, 1997.
(1) Trost, B. M.; van Vranken, D. L. Chem. Rev. 1996, 96, 395. Togni,
A.; Venanzi, L. M. Angew. Chem. 1994, 33, 497. Tanner, D. Angew.
Chem., Int. Ed. Engl. 1994, 106, 625. Hayashi, T. In Catalytic
Asymmetry Synthesis; Ojima, I., Ed.; VCH: Deerfield Beach, FL, 1993;
p 325.
We have also studied ligand 2 in connection with the
Pd-catalyzed allylic alkylation reaction.1 Ligand 2 was
(2) Mackenzie, P. B.; Whelan, J .; Bosnich, B. J . Am. Chem. Soc.
1985, 107, 2046. Auburn, P. R.; Mackenzie, P. B.; Bosnich, B. J . Am.
Chem. Soc. 1985, 107, 2033. Trost, B. M.; Breit, B.; Peukert, S.;
Zambrano, J .; Ziller, J . W. Angew. Chem. 1995, 107, 2577. Hayashi,
T.; Yamamoto, A. Tetrahedron Lett. 1988, 29, 669. Hayashi, T.;
Yamamoto, A.; Ito, Y.; Nishioka, E.; Miura, H.; Yanagi, K. J . Am. Chem.
Soc. 1989, 111, 6301. Togni, A.; Breutel, C.; Soares, M.; Zanetti, N.;
Gerfin, T.; Gramlich, V.; Spindler, F.; Rihs, G. Inorg. Chim. Acta 1994,
222, 213. Togni, A.; Breutel, C.; Schnyder, A.; Spindler, F.; Landert,
H.; Tijani, A. J . Am. Chem. Soc. 1994, 116, 4062. Togni, A.; Barbaro,
P. Organometallics 1995, 14, 3570.
(3) Brown, J . M.; Hulmes, D. I.; Guiry, P. J . Tetrahedron 1994, 50,
4493. Brown, J . M.; Lloyd-J ones, G. C. J . Am. Chem. Soc. 1994, 116,
866.
(4) Rieck, H.; Helmchen, G. Angew. Chem. 1995, 107, 2881. Knu¨hl,
G.; Sennhenn, P.; Helmchen, G. J . Chem. Soc., Chem. Commun. 1995,
1845. Sprinz, J .; Kiefer, M.; Helmchen, G.; Reggelein, M.; Huttner,
G.; Zsolnai, I. Tetrahedron Lett. 1994, 35, 1523. Sprinz, J .; Helmchen,
G. Tetrahedron Lett. 1993, 1769.
(5) Lloyd-J ones, G. C.; Pfaltz, A. Angew. Chem. 1996, 107, 534.
Pfaltz, A. Acc. Chem. Res. 1993, 26, 339. von Matt, P.; Pfaltz, A. Angew.
Chem., Int. Ed. Engl. 1993, 32, 566. Leutenegger, U.; Umbricht, C.;
Fahrni, P.; von Matt, P.; Pfaltz, A. Tetrahedron 1992, 48, 2143. Mueller,
D.; Umbricht, B.; Weber, A.; Pfaltz, A. Helv. Chim. Acta 1991, 74, 232.
(6) Frost, C. G.; Williams, M. J . Synlett 1994, 551. Dawson, G.; Frost,
C. G.; Williams, J . M. J . Tetrahedron Lett. 1993, 34, 3149.
(7) Togni, A.; Burckhardt, U.; Gramlich, V.; Pregosin, P.; Salzmann,
R. J . Am. Chem. Soc. 1996, 118, 1031. Burckhardt, U.; Hintermann,
L.; Schnyder, A.; Togni, A. Organometallics 1995, 14, 5415. Abbenhuis,
H. C. L.; Burckhardt, U.; Gramlich, V.; Togni, A.; Albinati, A.; Mueller,
B. Organometallics 1994, 13, 4481.
potentially interesting in that (i) its chirality is not
associated with the phosphine backbone, (ii) it is a
mixed-donor bidentate species, and (iii) there might be
some interesting secondary chemistry associated with
the OH function. Unfortunately, in the allylation reac-
tion, the observed enantiomeric excesses were disap-
pointing.10 Since the successful design of future cata-
lysts depends on avoiding auxiliaries prone to “err”, it
seemed important to trace the problems in the chem-
istry of this ligand. We report here on some Pt(II), Rh-
(I), and Ir(III) complexes of 2 and suggest that the lack
of selectivity associated with 2 arises from the presence
(8) Loiseleur, O.; Meier, P.; Pfaltz, A. Angew. Chem. 1996, 108, 218.
(9) Albinati, A.; Pregosin, P. S.; Wick, K. Organometallics 1996, 15,
2419.
(10) Herrmann, J .; Pregosin, P. S.; Salzmann, R.; Albinati, A.
Organometallics 1995, 14, 3311.
S0276-7333(96)00823-0 CCC: $14.00 © 1997 American Chemical Society

580 Organometallics, Vol. 16, No. 4, 1997
Albinati et al.
Ch a r t 1
F igu r e 1. ORTEP view of the cation of 3.
Ta ble 1. Selected Bon d Len gth s (Å) a n d Bon d
An gles (d eg) for [M(η3-C3H5)(2)]+
3 (M ) Pt)
4 (M ) Pd)
M-P(1)
2.264(1)
2.318(1)
2.151(6)
2.169(8)
2.195(5)
1.37(1)
2.287(2)
2.350(2)
2.145(9)
2.13(1)
2.198(7)
1.34(1)
1.41(1)
M-S
M-C(1)
M-C(2)
M-C(3)
C(1)-C(2)
C(2)-C(3)
C(1)-C(2a)
C(2a)-C(3)
S-C(4)
1.40(1)
1.38(2)
1.40(1)
1.839(6)
1.821(5)
1.815(5)
1.815(5)
1.830(6)
1.8559(9)
1.826(6)
1.891(7)
1.829(7)
1.841(8)
S-C(5)
P(1)-C(111)
P(1)-C(121)
P(1)-C(131)
P(1)-M-S
95.00(5)
167.0(2)
130.9(2)
130.4(2)
98.7(2)
95.41(9)
167.2(2)
132.2(2)
S-M-C(1)
S-M-C(2)
S-M-C(2a)
S-M-C(3)
99.4(3)
97.2(2)
165.1(3)
103.3(3)
124.6(9)
P(1)-M-C(1)
P(1)-M-C(3)
C(1)-M-C(3)
C(1)-C(2)-C(3)
C(1)-C(2a)-C(3)
C(4)-S-C(5)
M-S-C(4)
97.9(2)
165.7(2)
68.3(2)
123.8(7)
133(2)
98.3(3)
104.5(2)
111.7(2)
103.3(3)
112.0(2)
M-S-C(5)
of stereoisomers due to the S atom, together with chelate
ring inversion and, possibly, the propensity for this
sulfur donor to dissociate.
Section). The immediate coordination sphere consists
of the η3-C3H5 allyl, the sulfur atom, and the phosphorus
donor. The various bonding distances are normal: Pt-
P(1), 2.264(1) Å; Pt-S, 2.318(1) Å, Pt-C(1), 2.151(6) Å;
Pt-C(2), 2.169(8) Å; Pt-C(3), 2.195(5) Å. The coordina-
tion angles of interest for the major (exo) isomer are as
follows: P(1)-Pt-S, 95.00(5)°; P(1)-Pt-C(1), 97.9(2)°;
P(1)-Pt-C(3), 165.7(2)°; S-Pt-C(1), 167.0(2)°; S-Pt-
C(3), 98.7(2)°. Within the allyl group one finds the
following: C(1)-C(2), 1.36(1) Å; C(2)-C(3), 1.40(1) Å;
C(1)-C(2)-C(3), 123.8(7)°. Generally, the Pt-S, Pt-
P, and Pt-C(allyl) separations are in agreement with
the known chelating P,S11-13 and allyl14-20 structural
literature. Moreover, these distance and angle data for
3 are in good agreement with those for the previously
reported palladium analog 4 (see Table 1). Interest-
Resu lts
The new complexes were prepared as shown in Chart
1, using classical starting materials. Since the solution
measurements were often complicated, we begin with
the solid-state structure for 3.
Solid -Sta te Str u ctu r e of [P t(η3-C3H5)(2)]P F 6 (3).
The solid-state structure of 3 was determined by X-ray
diffraction at 173 K, and an ORTEP view of the cation
is given in Figure 1. Table 1 shows a selection of bond
lengths and bond angles for 3, whereas Table 2 gives
experimental details of the structure determination.
There are two different orientations of the η3-C3H5 group
with respect to the coordination plane, with that shown
in the ORTEP having the exo structure (endo and exo
refer to the orientation of the central allyl C-H bond
relative to the norborneol S substituent). In the solid
state the exo/endo ratio is ca. 3:1 (see Experimental
(11) Abel, E.; Evans, D. G.; Koe, J . R.; Sik, V.; Hursthouse, M. B.;
Bates, P. A. J . Chem. Soc., Dalton Trans. 1989, 2315. Abel, E.; Domer,
J . C.; Ellis, D.; Orrell, K. G.; Sik, V.; Hursthouse, M. B.; Mazid, M. H.
J . Chem. Soc., Dalton Trans. 1992, 1073.

Chiral P,S-Chelate Pt, Rh, and Ir Complexes
Organometallics, Vol. 16, No. 4, 1997 581
Ta ble 2. Exp er im en ta l Da ta for th e X-r a y
Diffr a ction Stu d y of [P t(η3-C3H5)(2)]P F 6 (3)
chem formula
mol wt
T, K
C32H38F6OP2PtS
841.7
173
space group
a, Å
P212121 (No. 18)
10.815(2)
b, Å
16.121(4)
c, Å
18.736(2)
V, Å3
3267(1)
Z
4
F(calcd), g cm-3
µ, cm-1
λ, Å
1.711
45.549
0.710 69 (graphite
monochromated, Mo KR)
2.5 < θ < 27.0
3954
θ range, deg
no. of indep data collected
no. of obsd rflns (no)
transmissn coeff
Ra
3583 (|Fo| > 5.0σ(|F|))
0.9998-0.9008
0.023
a
Rw
0.034
a
R ) ∑(|Fo - (1/k)Fc|)/∑|Fo| and Rw ) [∑w(Fo - (1/k)Fc)2/
∑w|Fo| ]1/2, where w ) [σ2(Fo)]-1 and σ(Fo) ) [σ2(Fo2) + f4(Fo2)]1/2
/
2
F igu r e 2. Space-filling model giving a view of the cation
of 3, revealing the open space around the sulfur atom
(shaded).
2Fo.
ingly, for 4, the Pd-P and Pd-S separations 2.287(2) and
2.350(2) Å are longer than those for 3, the Pt analog.
The distances and angles for the minor (endo) isomer
do not differ significantly.
is on the same side as C(111); C(5), at +1.67Å from the
plane, completes the definition of the chiral array. The
+1.67Å distance supports a pseudoaxial arrangement
of this sulfur norborneol substituent.
The O-H group has been localized and refined. The
oxygen atom is 3.300(4) Å from the metal, and impor-
tantly, H(O1) is 3.56(6) Å from the metal and is
consequently directed away from the platinum atom.
There is no obvious hydrogen bonding between the OH
The angles C(4)-S-C(5) (98.3(3)°), Pt-S-C(4) (104.5-
(2)°), and Pt-S-C(5) (111.7(2)°) suggest that the ste-
reochemically interesting lone pair on sulfur lies be-
tween the allyl and the exo norborneol, pointing away
from the oxygen. The top area (see Figure 2) is rela-
tively open and accommodates the sulfur lone pair. The
remaining distances and angles are relatively routine.
Summarizing the structure of the chiral pocket: the
two phenyl groups assume pseudoaxial and pseu-
doequatorial positions. The norborneol on sulfur is
pseudoaxial, thus leaving a relatively large open area
for the sulfur lone pair. This tendency for S-axial
substituents has been noted previously.21 The OH is
directed away from the metal.
-
group and either the PF6 anion or the sulfur donor;
however, from the packing distances one observes
somewhat short F(6)-H(O1) (2.70(7) Å) and F(4)-H(O1)
(3.03(7) Å) separations. In addition, the F(3)-O(1)
contact is 2.808(6) Å, which is less than the sum of the
van der Waals radii for O and F.21 We note that the O
atom is sitting approximately in a pseudo fifth coordina-
tion position of the Pt atom.
The allyl plane makes an angle of 113° with the plane
defined by the metal, P, and S atoms. This is normal
for η3-C3H5 ligands and is in agreement with the 110°
value observed in 4. The chiral P,S ligand offers an
interesting chiral pocket with respect to the allyl ligand,
and Figure 2 shows a space-filling model for the cation.
The three allyl carbons are almost symmetrically placed
with respect to the coordination plane (C(1), +0.15Å;
C(2), -0.13Å; C(3), +0.15Å). The positions of the ipso
carbons of the three P-phenyl groups (C(111), -1.67Å;
C(121), +0.96Å; C(131), 0.66Å) suggest that C(111) is
axial and C(121) is equatorial. The carbon bound to
sulfur, C(4), is -0.72Å from the coordination plane and
NMR Stu d ies on [P t(η3-C3H5)(2)]P F 6 (3). The 31P
and 195Pt NMR spectra of 3, recorded at 323 K, show
1
two components, in almost equal amounts, with J (Pt,P)
values of 4063 and 4059 Hz. For the discussion which
follows it is useful to note the nomenclature
(12) Chivers, T.; Edwards, M.; Meetsma, A.; van der Grampel, J .;
van der Lee, A. Inorg. Chem. 1992, 31, 2156.
(13) Capdevila, M.; Clegg, W.; Gonzalez-Duarte, P.; Harris, B.; Mira,
I.; Sola, J . Taylor, I. C. J . Chem. Soc., Dalton Trans. 1992, 2817.
(14) Musco, A.; Pontellini, R.; Grassi, M.; Sironi, A.; Meille, S. V.;
Ru¨egger, H.; Ammann, C.; Pregosin, P. S. Organometallics 1988, 7,
2130.
(15) Breutel, C.; Pregosin, P. S.; Salzmann, R.; Togni, A. J . Am.
Chem. Soc. 1994, 116, 4067. Albinati, A.; Kunz, R. W.; Aumann, C.;
Pregosin, P. S. Organometallics 1991, 10, 1800.
and remember that the allyl can exist in exo and endo
forms.
Moreover, we will make extensive use of the dihedral
angle dependence of the three vicinal coupling con-
stants22 shown (view down the Pt-S bond):
(16) Fernandez-Galan, R.; Manzano, B. R.; Otero, A.; Lanfranchi,
M.; Pellinghelli, A. Inorg. Chem. 1994, 33, 2309.
(17) Knierzinger, A.; Schoholzer, P. Helv. Chim. Acta 1992, 75, 1211.
(18) Ozawa, F.; Son, T.; Ebina, S.; Osakada, K.; Yamamoto, A.
Organometallics 1992, 11, 171.
(19) Smith, A. E. Acta Crystallogr. 1965, 18, 331.
(20) Grassi, M.; Meille, S. V.; Musco, A.; Pontellini, R.; Sironi, A. J .
Chem. Soc., Dalton Trans. 1990, 251; 1989, 615.
(21) Bondi, A. J . Phys. Chem. 1964, 68, 44.

582 Organometallics, Vol. 16, No. 4, 1997
Albinati et al.
Ta ble 3. Selected 1H a n d 31P NMR Da ta for
The 195Pt,1H interaction will help us to understand
the ring conformation, whereas the 31P,13C and 13C,1H
values can be diagnostic with respect to the orientation
of the sulfur substituent.
[P t(η3-C3H5)(2)]+ (Ca tion of 3)a
group or position
3a
3b
3c
3d
Allyl
The 1D-1H and 2D-TOCSY NMR spectra at 323 K
reveal two different η3-allyl and norborneol moieties and
somewhat broadened resonances arising from the me-
thylene protons H10 and H11. Surprisingly, these mea-
surements reveal vanishingly small differences in the
1H chemical shifts between analogous allyl proton
resonances. This situation is incompatible with the
formulation of a static chiral array as found in the X-ray
structure of 3. The heteronuclear coupling constants
3J (Pt,H) for the two H11 protons show averaged values
of ca. 75(5) Hz. If the chelate ring conformation were
fixed, one would expect very different vicinal couplings,
on the basis of the known dihedral angle dependence
1s
1a
2
4.83
2.95
4.44
3.28
4.75
3.30
4.61
2.82
5.22-5.28
3.77 and 3.63
2.88 and 2.72
3s
3a
Norborneol Fragment
OH
2
8
9
10
10′
2.20
4.04
0.99
0.74
2.39
3.42
2.15
4.11
0.98
0.77
2.76
3.09
2.25
4.11
0.96
0.71
2.31
3.13
2.16
4.05
1.00
0.78
2.95
3.08
Chelate
P
14.7b
4100
7.44
14.1
4018
7.57
14.1b
4064
13.4b
3992
J (Pt,P)
13
3
(Karplus relation) for J (Pt,H), e.g., as indicated (view
down the S-C(11) bond:
a
In CD2Cl2 solution at 233 K. δ values are in ppm and J values
in Hz. Isomers 3a and 3b are those shown in the upper half of
Chart 2, whereas isomers 3c and 3d are in the lower half of Chart
b
2. Assignments could be reversed.
The two allyl species are in relatively slow exchange
at 323 K, on the basis of phase-sensitive NOESY results,
and we assign these two species to the exo and endo
orientations of the allyl ligand.23,24
A series of low-temperature measurements between
183 and 293 K proved informative; unfortunately, at no
single temperature was there ever a completely sharp
1H spectrum. The best compromise was achieved at 233
1
K. At this temperature, from both the 31P and H data,
there are clearly four isomers in solution, with similar
1
but not quite equal populations. The various allyl H
NMR signals for 3, together with several important
protons from the ligand such as the methyl and the
hydroxy groups, can be localized and assigned, using a
series of 2D NMR methods (see Table 3).
F igu r e 3. Section of the 1H 2-D exchange spectrum
showing the four allyl syn protons (each trans to phospho-
rus) in the four isomers of 3 (500 MHz, 233 K, CD2Cl2).
These four protons take part in two selective exchange
processes. There is an unresolved, very broad line not
involved in the exchange.
1
1
At 233 K the allyl protons Hs and Ha are both
relatively sharp and well-resolved, whereas the protons
H3 are relatively broad. Figure 3 shows the section of
the 2-D proton exchange spectrum containing the syn
In ela stic In coh er en t Neu tr on Sca tter in g (IINS)
Stu d ies on [P d (η3-C3H5)(2)]CF3SO3. Role of th e OH
Gr ou p . We observed earlier10 that the Pd analog of 3,
[Pd(η3-C3H5)(2)]CF3SO3 (4), shows selective proton NOE’s
from the OH group such that one might think that the
hydroxy group was rigidly in place above the Pd atom.
Some form of interaction with the metal center could
not be excluded. Since the OH group was not localized10
in the X-ray diffraction study for 4, the point remained
open.
1
allyl protons trans to the P-donor Hs at 233 K. There
are two selective exchange processes among the four
isomers. Although they are not reproduced, the allyl
1
protons Ha and the norbornyl methyl signals exhibit
the same type of selectivity.
At this point it seemed that the two new isomers
might arise from the stereogenic sulfur atom, some
preferred ring conformation, or perhaps the involvement
of the OH group.
In view of the well-known sensitivity25 of inelastic
incoherent neutron scattering (IINS) to proton motion,
we have employed this technique to settle this question
for 4. In order to isolate the vibrational modes involving
the hydroxy moiety, a sample difference technique was
used (see Experimental Section). The resulting differ-
ence spectrum shows the vibrational modes of the “O-
H” fragment, as well as some other modes arising from
large-amplitude displacements of this group, e.g. certain
ring deformations.
(22) See: Erickson, L. E.; Sarneski, J . E.; Reilley, C. N. Inorg. Chem.
1975, 14, 3007. Erickson, L. E.; Sarneski, J . E.; Reilley, C. N. Inorg.
Chem. 1978, 17, 1701. Yano, S.; Tukada, T.; Yoshikawa, S. Inorg.
Chem. 1978, 17, 2520 and references therein (for Karplus type
relationships involving 195Pt with 13C and 1H). For the Karplus relation
and 3J (31P,X), see also: Methods in Stereochemical Analysis; Verkade,
J . G., Quin, L. D., Eds.; VCH: Deerfield Beach, FL, 1987; Vol. 8.
(23) Faller, J . W. In Determination of Organic Structures by Physical
Methods; Nachod, F. C., Zuckerman, J . J ., Eds.; Academic Press: New
York, 1973; Vol. 5, p 75. Vrieze, K. In Dynamic Nuclear Magnetic
Resonance Spectroscopy; J ackman, L. M., Cotton, F. A., Eds.; Academic
Press: New York, 1975.
(24) Faller, J .; Thomsen, M. E. J . Am. Chem. Soc. 1969, 91, 6871.
Faller, J .; Incorvia, M. J .; Thomsen, M. E. J . Am. Chem. Soc. 1969,
91, 518.
(25) Bacon, G. E. Neutron Scattering in Chemistry; Butterworths:
London, England, 1977.

Chiral P,S-Chelate Pt, Rh, and Ir Complexes
Organometallics, Vol. 16, No. 4, 1997 583
13C1H long-range correlations, together with NOESY
two-dimensional experiments. NMR data for these are
summarized in Table 5. The solution structure of the
major isomer 5a can be established on the basis of (i)
H-H distance constraints, as evidenced by two-dimen-
sional NOE spectroscopy, and (ii) the dihedral angle
dependence of the vicinal and long-range 13C1H, 31P1H,
and 195Pt1H coupling constants.
Specifically, the conformation of the P,S chelate six-
membered ring is determined by a strong NOE between
eq
the aryl proton H13 and the chelate ring proton H11
,
plus a weak interaction between the phosphine phenyl
ortho protons resonating at δ 7.77 and H11ax (see Figure
5a). The relative volumes of the appropriate NOESY
cross-peaks are consistent with the distances observed
in the crystal structure of 3, i.e., 2.25 and 3.00 Å,
respectively, suggesting that a static boat-type ring
conformation similar to that in 3 is preserved in solution
for 5a . The values 3J (195Pt,1H) for the nonequivalent
ring methylene signals strongly support this assign-
1
F igu r e 4. Section of the H 2-D exchange spectrum for 3
showing the OH region of the spectrum. The OH reso-
nances at high frequency are in selective exchange with
each other and in nonselective exchange with traces of
water (low frequency) in the solvent. The splitting on the
OH signals stems from CHOH (500 MHz, 233 K, CD2Cl2).
eq
ment, as H11 shows a large coupling of 116 Hz,
ax
whereas the geminal H11 shows a moderate coupling
The IINS difference spectrum has a prominent peak
at 645 cm-1 and several overlapping bands between 700
of 43 Hz. These values are in agreement with the
expected Karplus-like behavior for J (Pt,H), noted above,
3
and 1100 cm-1, as well as a broad band at ca. 1560 cm-1
.
and the proposed boat-type chelate ring conformation.
4
Since the IINS band intensities are largely determined
by the displacement amplitudes of the H atoms, and as
these are larger for deformation modes, the peak at 645
cm-1 may readily be identified as the out-of-plane
bending mode γ(OH) of an essentially noninteracting,
non-hydrogen-bonded OH group, on the basis of its
intensity.26 The in-plane bending δ(OH), found as a
broad band near 1560 cm-1, is generally not affected
significantly by weak interactions. Other vibrational
modes involving hydroxyl displacements are found at
810, 965, and 1130 cm-1 and may be described as
C-C-O stretches.27
Moreover, the J (P,H) long-range coupling constant for
eq
H11 (4.3 Hz) is similar in magnitude to that for H13
(4.6 Hz), indicative of the quasi-planar arrangement for
both of these spins. We show below a fragment of the
major isomer of 5, with key protons indicated:
Consequently, we conclude from the low-frequency γ-
(OH) data in this complex that the OH group is not
interacting with the metal through hydrogen bonding.
Given this result, we reconsidered the solution exchange
data for the platinum complex 3.
Figure 4 shows the section of the solution exchange
spectrum for the four nonequivalent OH protons for the
isomers of 3, at 233 K. Each of these protons is coupled
to the adjacent methine CH proton. Two of these OH
signals, at lower frequency, overlap somewhat; never-
theless, it is readily seen that all four are in slow
exchange within themselves as well as with the small
amount of free H2O present. Consequently, the OH
group plays no special role in this platinum compound,
in agreement with the IINS study for the Pd analog.
We assume that the relatively slow exchange of the OH
group (also observable at ambient temperature) is due
to a weak interaction with the anion.
The sulfur norborneol substituent is axial, as is
evident from NOE spectroscopy; e.g., H11 is always
ax
distant from the norborneol moiety (see Figure 5b). This
configuration at sulfur is further supported by the
3
absence of a J (31P,13C) coupling for C10 and the presence
of a cross-peak in the CH long-range correlation,
ax
indicating a substantial 3J value between C10 and H11
.
With respect to the former, it should be noted that the
corresponding dihedral angle, C10-S-Pt-P, from the
X-ray study of 3 is 81.3°, which is close to the expected
minimum in the Karplus relation, thereby suggesting
the following fragment for 5a (view down the Pt-S
bond):
P tCl2(2) (5). This relatively simple complex of 2 was
chosen to study the chelate ring conformation. Complex
5 exists in solution as two stereoisomers which equili-
1
brate slowly at ambient temperature. The H and 13C
resonances can be fully assigned for the major isomer
(93%), and in part for the minor isomer (7%), on the
basis of a combination of 195Pt1H, 31P1H, 13C1H, and
(26) Noval, A. Struct. Bonding 1974, 17, 177.
(27) Daimay Lin-Vien. The Handbook of infrared and Raman
characteristic frequencies of organic molecules; Academic Press: Bos-
ton, 1991.

584 Organometallics, Vol. 16, No. 4, 1997
Albinati et al.
Ta ble 4. 1H a n d 13C NMR Da ta for th e Com p lex
[P tCl2(2)]a
major isomerb
minor isomerc
group or δ-
position (1H) (Pt,H) (P,H)
J - J - δ-
J -
δ-
J -
J -
δ-
(
13C) (P,C) (1H) (Pt,H) (P,H)
(
13C)
OH
1
2
3.92
1.78
54.0d
75.7
39.6
52.6
76.6
4.04
1.81
1.81
1.76
1.81
1.19
1.42
1.61
4.08
3
4
5
45.2
27.4
46.4
27.2
1.12
1.49
1.95
6
29.9
7
8
9
10
48.8d
19.9
20.5
36.4
49.0
20.4
20.8
1.01
0.71
3.53
2.26
1.0
0.88
2.94
3.35
38
∼2
∼7
41
50
11
4.13 116
4.3
36.4 10.5 3.82
3.74
43
4.45 107
3.9
12
137.2 10.3
13
14
15
16
7.37
7.63
7.52
6.93
4.6 133.8
-
-
8.2 7.51
3.0
9.4
5.9 6.87
132.2e
130.5
11.6 135.2
11.4
17
Phax
Phax
123.8 60.5
i
126.7 72.9
12.8 135.2 11.4 7.55
129.7f 11.9
o
7.77
7.54
7.65
m
Pha
p
i
Phax
Pheq
Pheq
Pheq
Pheq
132.9
2.9
124.8 64.0
o
m
p
7.49
7.54
7.63
134.8 11.0 7.71
128.8f 12.3
132.3e
2.8
a
In CDCl3 at 295 K. δ values are in ppm and J values in Hz.
δ(31P) 2.3, J (Pt,P) ) 3712, δ(195Pt) -4166. c δ(31P) 2.9, J (Pt,P) )
b
3672, δ(195Pt) -4136. d-f Assignments could be reversed.
same rigor. However, invaluable information is ob-
tained from (i) the HMQC 195Pt1H correlation and (ii) a
1H1H exchange experiment. The former reveals that the
3J (195Pt1H) coupling constants are similar in 5a and 5b
(see Table 4), whereas the latter shows there is rela-
tively slow ring inversion on the NMR time scale.
Figure 6 shows the 195Pt satellites for H10 and H11 and,
in addition, indicates that the exchange is crosswise,
eq
3
i.e., the equatorial proton H11 in 5a (large J (Pt,H))
moves into an axial position in 5b (small 3J (Pt,H)), and
that the axial spin H11 in 5a exchanges with proton
H11 in 5b. Figure 7 shows the appropriate exchange
ax
eq
spectrum. Consequently, isomer 5b has its P,S chelate
ring inverted. Although not proven conclusively, it is
likely that the configuration at the stereogenic sulfur
center inverts simultaneously. In 5b , the value
3J (31P,13C) to C10, the exo methylene carbon, is not
resolved. This is in keeping with an axial position for
C10. In an equatorial position the dihedral angle would
be close to 150°, thereby affording a relatively large
3J (31P,13C).
The overall shape of the minor isomer has very much
the same form as the major isomer. In fact, the
structure for 5b could be thought of as resulting from a
reflection of the structure of 5a and “changing” the
hydroxy group from position 2 to 6. As a consequence,
the “chiral pockets” of the two isomers are almost, but
not quite, enantiomorphic. The source of the difference
in ring-inversion energies between the allyl and dichloro
complexes is not clear; however, for the latter, models
suggest that the hydroxy proton could hydrogen-bond
to a proximate Cl-, thus adding some additional rigidity
to 5.
F igu r e 5. Sections of the NOESY spectrum of 5 showing
(a, top) the selective cross-peak between H11ax, at ca. 3.74
ppm, and the ortho protons of the P-phenylax, group, at ca.
7.7 ppm (lower left), plus the selective cross-peak due to
the H13/H11 NOE (center-bottom), thereby helping to de-
termine the ring conformation, and (b, bottom) the absence
of NOE between H11 and the norbornyl protons (dotted
box), thus showing the axial nature of the S-substituent
(400 MHz).
ax
These results are all consistent with the structural
fragment shown previously.
Due to its small abundance, it is not possible to
deduce the structure of the minor isomer 5b with the

Chiral P,S-Chelate Pt, Rh, and Ir Complexes
Organometallics, Vol. 16, No. 4, 1997 585
Ch a r t 2
Ch a r t 3
F igu r e 6. 195Pt satellites from the HMQC spectra for
5a (lower trace) and 5b (upper trace). The arrows
indicate which signals exchange with one another. It
is obvious that there are both relatively large and rela-
tively small 3J (195Pt,1H) values; e.g., in the lower trace,
see the first two anti-phase doublets relative to the second
two anti-phase doublets. Note that the high-frequency
proton in 5a with the relatively large 3J coupling is
exchanging with a proton in 5b with a much smaller 3J
coupling.
103Rh coupling. The 1H and the 13C NMR spectra
showed sharp resonances for both the diolefin and the
coordinated ligand 2, and these could be assigned on
the basis of homo- and heteronuclear correlation experi-
ments. Details of these NMR assignments are given in
Table 5.
These results are consistent with either the presence
of one single stereoisomer or, alternatively, with two
isomeric forms in fast exchange on the NMR time scale.
That the latter is indeed the case is evident from the
following observations.
(i) The diastereotopic protons H11 are isochronous in
CDCl3 solutions, and the pairs of olefinic protons on the
same double bond have almost the same chemical shifts.
This points to a situation where dynamic averaging
leads to the presence of a pseudo plane of symmetry
coinciding with the coordination plane.
(ii) In CD2Cl2 solution, where the ring H11 protons
possess slightly different chemical shifts, 31P1H and
103Rh1H correlation experiments (see Figure 8) reveal
that both protons are almost equally coupled to P and
Rh. A detailed analysis, shown in Figure 9, affords
F igu r e 7. Section of the NOESY spectrum of 5 showing
the exchange between the various H10 and H11 protons
(indicated by the dotted boxes).
Given these data, it is now possible to propose
structures for the four allyl derivatives 3 (Chart 2), as
well as for the two isomers of 5 (Chart 3).
NMR St u d ies on [R h (1,5-COD)(2)]CF 3SO3 (6).
The ambient-temperature 31P NMR spectrum for the
1,5-COD complex 6 showed one sharp doublet due to
3
4J (P,H) values of 3.3 and 2.6 Hz and J (Rh,H) values of
2.3 and 1.9 Hz for the H11 protons resonating at δ 3.65
and 3.58, respectively. This is clearly in contradiction
to the situation encountered for 5 but reminiscent of
the observations for 3. On the basis of the spread of

586 Organometallics, Vol. 16, No. 4, 1997
Albinati et al.
Ta ble 5. Selected NMR Da ta a for 6
position
1H
13C
OH
2
3.77
4.06
1.83
1.83
1.72
1.75
1.13
1.57
1.40
1.05
0.78
2.39
3.09
3.58
3.58
5.63
5.63
3.88
3.95
75.1
41.4
41.4
45.5
27.1
27.1
31.7
31.7
20.3
20.3
36.0
36.0
3
3′
4
5
5′
6
6′
8
9
10
10′
11
11′
18
19
20/21
21/21
35.2, J (P,C) ) 13
103.8
105.2, J (Rh,C) ) 9, J (P,C) )9
87.4, J (Rh,C) ) 11
88.0, J (Rh,C) ) 11
a
Endo protons are primed; CDCl3 solution. Chemical shifts are
F igu r e 9. Detailed analysis for the two H11 protons in 6
(derived from Figure 8) showing the magnitudes of (a, top)
the rhodium-proton spin-spin and (b, bottom) the phos-
phorus-proton spin-spin interactions.
in ppm and J values in Hz.
ratio is in good agreement also with the averaged
4J (P,H11) long-range coupling values.
(iii) In the NOESY spectrum (see Figure 10) one
observes close contacts between the ortho- protons of
both phenyl rings and the H11 protons, again inconsis-
tent with a single chelate ring conformation but in
agreement with the fast inversion of the boat-type
chelate ring. In addition, H13 shows twice as many
contacts as would be expected from the results for 5a ,
but this is in agreement with the presence of a second
isomer and fast interconversion of the two forms. It is
of interest to note that the ratio of cross-peak volumes
for several protons is consistently ca. 3:2.
Low-temperature NMR studies showed the presence
of two separate isomers in a ratio of ca. 2:1 at 173 K. At
this temperature all the 1H resonances are relatively
broad and exchange processes are still dominating the
two-dimensional NOE spectrum. There are indications
that P-phenyl rotation is hindered, thus complicating
the interpretation.
The ambient-temperature phase-sensitive 1H-NOESY
spectrum for 6 reveals that the olefinic protons, shown
as “a” and “b”, take part in a selective slow exchange
with “d” and “c”, respectively.28 However, the spins a
and b do not exchange with one another. We observed
similar behavior previously29 in the chiral cationic
rhodium complexes [Rh(P,N)(diene)]+ (diene ) 1,5-COD,
F igu r e 8. 31P,1H and 103Rh,1H correlations for the two H11
protons in 6. The splitting in the vertical direction repre-
sents 1J (103Rh,31P) in both cases. Note the rather similar
chemical shifts for these protons.
(28) Two COD protons trans to the S-donor are well resolved at 500
MHz; however, the two olefinic protons trans to the P-donor overlap.
Consequently, we cannot be completely certain about the selectivity
of the exchange with respect to the overlapping signals. A process in
which the the two olefinic protons trans to the S-donor exchange
selectively, but those trans to the P-donor exchange nonselectively,
thevicinal couplings 3J (Pt,H11) in 3 and 5 and the
assumption that the analogous 3J (Rh,H11) values are
proportionally scaled, one estimates a ratio of ca. 3:2
for the two stereoisomers 6a and 6b, respectively. This
would require
mechanism.
a rather complicated, and as yet unprecedented,

Chiral P,S-Chelate Pt, Rh, and Ir Complexes
Organometallics, Vol. 16, No. 4, 1997 587
Simple ring and/or sulfur inversion, as suggested for
3 and 5a , would change the local environments of the
COD-olefinic protons, e.g. a and b, but would not ration-
alize the observed exchange. We cannot exclude diolefin
“rotation” as a mechanism but consider this unlikely,
given that several bonds must be broken simulta-
neously. More importantly, our results for 6 raise the
question of metal-sulfur bond breaking, an additional
potential contributor to poor enantioselectivity.
[Ir H2(2){P (p-XC6H4)3)}2]BF 4 (8). Relative to the
chiral pocket of coordinated 2, a square-planar coordi-
nation sphere offers a steric environment different from
that found in a six-coordinate octahedral compound.
Consequently, we chose to study the cations [IrH2(2)-
{P(p-XC6H4)3)}2]+, as related dihydrides have some
catalytic relevance.30
Reaction of the known cationic hydrogenation precur-
sor [IrH2(acetone)2{P(p-XC6H4)3)}2]+ with 2 gave the
octahedral complexes 8. For X ) Cl (8a ), F (8b), and H
(8c), the complexes were readily obtained in pure form;
however, for X ) OMe (8d ), it was not possible to obtain
an analytically pure material.
The p-Cl compound 8a was analyzed in detail, al-
though preliminary NMR results for the fluoro and
hydrogen analogs suggest these are qualitatively identi-
cal. The ambient-temperature 31P NMR spectrum for
8a revealed a broad signal of relative intensity 2 at δ
5.1 and a sharp triplet of intensity 1 at δ -8.4.
Empirically, we find that these two regions are typical
for the 31P spins of the iridium-coordinated P(p-XC6H4)3
and 2, respectively. At 243 K the 31P spectrum shows
more than 16 broad but resolvable resonances between
δ +15 and -15. Many of these resonances show small
F igu r e 10. Section of the NOESY spectrum of 6 showing
the “doubled” NOEs (dotted boxes), in the ratio ca. 3:2. The
same aromatic proton is showing NOEs to very differently
placed methyl groups and the two H10 protons (vertical
boxes).
2
splittings due to modest J (P,P)cis values.
Close inspection of the conventional 31P spectrum at
243 K shows that several of the 31P signals arise as a
consequence of trans AB subspectra from ABX spin
systems, with typical31 2J (P,P)trans values in excess of
300 Hz. Assuming, for I, that there are slow dynamics
norbornadiene) with the P,N-ligand 7. We suggested29
that the rhodium diolefin complexes of 7 isomerized via
dissociation of the stereogenic secondary nitrogen atom,
followed by rearrangement of the three-coordinate “T”-
shaped complex and then re-formation of the four-
coordinate species. We believe that an analogous
mechanism is operating in 6, i.e.
similar to those described above for 3 and 5, the two
monodentate arylphosphines (L) need not be equivalent;
i.e., isomer I would readily afford two ABX 31P spectra.
However, these two diastereomeric isomers are insuf-
ficient to account for the more than 16 well-separated
31P resonances. The data can be rationalized by as-
suming the presence of geometric isomer II, and this
assumption was shown to be correct via a 31P1H cor-
relation. Several of the 31P resonances which revealed
large 2J (P,P)trans values of >300 Hz correlated to protons
of the P,S ligand, i.e., the 31P spin of 2 was trans to an
(30) Shapley, J . R.; Schrock, R. R.; Osborn, J . A. J . Am. Chem. Soc.
1969, 91, 2816. Schrock, R. S.; Osborn, J . A. J . Am. Chem. Soc. 1971,
93, 3089.
(31) Pregosin, P. S.; Kunz, R. W. In NMR Basic Principles and
Progress; Springer Verlag: Heidelberg, Germany, 1979; Vol. 16.
Pregosin, P. S. In Methods in Stereochemical Analysis; Verkade, J . G.,
Quin, L. D., Eds.; Deerfield Beach, FL, 1987; Vol. 8.
(29) Berger, H.; Nesper, R.; Pregosin, P. S.; Ru¨egger, H.; Wo¨rle, M.
Helv. Chim. Acta 1993, 76, 1520.

588 Organometallics, Vol. 16, No. 4, 1997
Albinati et al.
Ta ble 6. 1H Hyd r id ea a n d 31P NMR Da ta for th e
Isom er s I a n d II in th e (p-ClC6H4)P An a log of 8
(CD2Cl2, 243 K)
II
I
isomer
δ(Htrans
2J (P,H)trans 125
)
-12.32 -12.13 -12.15 -12.34 -11.33 -12.02
121 132 125 106 125
δ(Hcis
)
17.37
-10.9 -10.5
-16.82 -16.26 -16.31 -15.96 -15.64
δ(P)b
-7.6
-5.3
-5.9
J (P,P)trans 320
308
4.9
313
12.6
292
-4.8
δ(P)c
3.9
8.4
296
8.0
J (P,P)trans 315
δ(P)c
272
10.3
278
5.7
J (P,P)trans
300
a
The descriptors cis and trans for the hydride proton are
relative to the 31P spin. All δ values are in ppm and J values in
Hz. In the chelate. c (p-ClC6H4)3P.
b
plays a role in determining the relative stabilities of
these isomers. There are some indications, from the
relative line widths in the 1H and 31P spectra at low
temperature, that the isomers I may be somewhat more
dynamic than those for II.33 In any case, in this
octahedral environment, coordinated 2 still inverts the
chelate ring and, presumably, the sulfur atom quite
readily.
1
F igu r e 11. Section of the H 2-D exchange spectrum for
Discu ssion
7a showing the hydride resonances trans to the sulfur
donor. There are three selective exchange processes: one
of type I and two of type II. The third (central) signal stems
from two overlapping resonances (500 MHz, 243 K, CD2-
Cl2).
The Pd(II), Pt(II), Rh(I), and Ir(III) structural chem-
istry associated with ligand 2 involves a series of
equilibrating diastereomers concerned partially with the
stereogenic sulfur center. Coupled to this inversion is
a simultaneous change in the chelate ring conformation.
In this connection, similarities in the 1H chemical shifts
of the diastereotopic protons H11 and the lack of
L ligand. In the isomers II, both the metal and the
sulfur are stereogenic centers, thus increasing the
number of possible diastereomers to four. Careful
3
selectivity in the values J (M,1H) (M ) 195Pt, 103Rh) have
1
consideration of the hydride section of the H spectra
proven to be useful tools for the recognition of the rapid
chelate ring flip. This interconversion can also be noted
in that there is, often, an unexpected “doubling” of
NOE’s; i.e., a proton seems to be in contact with
spatially remote sites.
readily revealed six complexes. Together with their 31P
data, these hydride signals could be assigned to two type
I and four type II diastereomers.
1
Figure 11 shows a section of the H exchange spec-
trum, at 243 K, in the region of the hydride proton trans
to the sulfur donor. If one accepts that the third
(middle) resonance contains two overlapping signals
(this is obvious from the exchange pattern), then one
can see three selective exchange processes. Once again,
we suggest that these dynamics are associated with the
ring and stereogenic sulfur inversions: one pair for the
major geometric isomer I (the two highest frequency
signals in Figure 10) and two sets for geometric isomer
II (the four lowest frequency signals). We give some
Unfortunately, these equilibria can be further com-
plicated by the presence of geometric isomers (exo/endo
isomers in the Pd allyl species or cis/trans isomers in
the Ir dihydride). For a given complex, it is frequently
the case that the various stereo- and/or geometric
isomers have similar populations; i.e., the ligands
remaining in the coordination sphere do not strongly
differentiate (either electronically or sterically) between
the various possible isomers. There are indications,
specifically from the rhodium chemistry, that this chiral
P,S thioether chelate may dissociate the sulfur donor.
As 2 has a pendant hydroxy group, it was conceivable
that this donor might influence the structural chemis-
try. However, for the Pd(II) and Pt(II) allyl complexes
this seems not to be the case, as demonstrated by IINS
and X-ray diffraction methods, respectively. It is pos-
sible that the hydroxy group interacts with one of the
chloride ligands in 5, thus slowing the ring inversion
somewhat.
1
NMR details for these H hydride and 31P resonances
in Table 6.
We do not know the mechanism of the isomerization
from I to II. Complexes I and II are not in equilibrium
at 243 K (see Figure 11) but do exchange rapidly at
ambient temperature, as noted above. The exchange
process may involve ligand dissociation to afford a five-
coordinate intermediate, followed by rearrangement and
ligand recoordination. We cannot exclude isomerization
via a molecular hydrogen complex. The relative abun-
dances of I and II were found to depend on the
arylphosphine with the ratios I:II being ca. 2:1 for X )
Cl and ca. 9:1 for X ) H (the latter complex measured
at 253 K32), suggesting that the donor capability of L
On the basis of crystallographic and MM2* results,
it has been suggested34,35 that a rigid chiral pocket is
(33) From the temperature dependence of the resonances in the
hydride region, it seems clear that I is more dynamic than II; however,
the complexity of the “normal” proton region does not allow us to
speculate as to why this might be so.
(32) For X ) OMe, the isomers I make up ca. 85% of the sum of I
plus II; however, as noted, the sample is not pure.
(34) Barbaro, P.; Currao, A.; Herrmann, J .; Nesper, R.; Pregosin,
P. S.; Salzmann, R. Organometallics 1996, 15, 5160.

Chiral P,S-Chelate Pt, Rh, and Ir Complexes
Organometallics, Vol. 16, No. 4, 1997 589
the real and imaginary parts of the anomalous dispersion, were
taken from the literature.38
The handedness of the structure was tested by refining both
enantiomorphs; the coordinates giving the significantly39 lower
Rw factor were used.
Upon convergence the final Fourier difference map showed
no significant peaks. All calculations were carried out by using
the Enraf-Nonius MOLEN crystallographic programs.36
In ela stic Neu tr on Sca tter in g Exp er im en ts (IINS). The
IINS data were collected at the Manuel Lujan Neutron
Scattering Center (MLNSC) of the Los Alamos National
Laboratory.
In order to be able to distinguish the O-H vibrational modes
from all the other vibrations involving hydrogen atoms of the
ligands, a “sample difference” technique40 was used. The
method is based on the fact that the incoherent neutron
scattering cross sections of hydrogen and deuterium are very
different: 79.91(4) and 2.04(3) b, respectively.41 Thus, vibra-
tional modes involving deuterium atoms are difficult to detect
by IINS in presence of many hydrogens, and the difference
between two experimental IINS spectra, i.e. those of the
compound containing the “O-D” and “O-H” groups, respec-
tively, should leave only the peaks involving the modes of
O-H, provided that any possible coupling of these modes to
other molecular modes is negligible. However, very high
counting rates are necessary, in order to obtain reasonable
statistical precision on the difference data set. The filter
difference spectrometer (FDS) at LANSCE42 is particularly
well-suited for this type of measurement.
In the FDS spectrometer, a pulsed “white” beam of neutrons
is scattered by the sample and energy analyzed by a cooled Be
filter, placed between the sample and the detectors. Of the
scattered neutrons, only those with energies falling within the
band-pass (5.2 meV, λ ≈ 4 Å) of the filter analyzer will reach
the detectors; the neutron final energies can thus be deter-
mined. The “time of flight” of the neutrons reaching the
detectors determines the incident energies of the neutrons. The
FDS spectrometer can cover an energy range from 250 to 4000
cm-1 with a resolution of 2-10% of the incident neutron
energy.
an important quality for a successful auxiliary in
connection with high enantioselectvity. For all of the
various metal complexes in this study, in both square-
1
planar and octahedral environments, 2-D H exchange
experiments reveal that coordinated 2 is sufficiently
mobile, such that one cannot speak of a rigid chiral
pocket. Chart 2 shows the two structures found for the
dichloride complex 5. Although the configuration at the
stereogenic sulfur atom is reversed, a coordinated
prochiral substrate would find little that would distin-
guish these two. All of our structural data, when taken
together with the observed dynamics, clearly indicate
that this type of chiral chelate is not likely to become
(and has not yet been) a successful auxiliary.
Exp er im en ta l Section
Platinum metals were obtained from J ohnson-Matthey.
FAB mass spectra as well as microanalytical measurements
were performed in the analytical laboratories of the ETH
Zurich. Solution NMR spectra were measured using Bruker
AC-250 DRX 400 and AMX-500 MHz spectrometers. Refer-
encing is to H3PO4 (31P), TMS (1H, 13C), Na2PtCl6 (195Pt) and
¥ ) 3.16 MHz (103Rh). Two-dimensional NMR spectra were
measured, as described by us previously.7,10,15 Mixing times
in the exchange spectra were 0.5-1.0 s.
Ligand 2 was prepared as described.10
Cr ysta llogr a p h y. A CAD4 diffractometer was used for the
unit cell and space group determination and for the data
collection. A suitable crystal was mounted on a glass fiber
and cooled to -100 °C by using an Enraf-Nonius FR558SH
nitrogen gas-stream cryostat.
Unit cell dimensions were obtained by a least-squares fit of
the 2θ values of 25 high-order reflections (9.69 e θ e 18.65°).
Selected crystallographic and other relevant data are listed
in Table 2 and Supplementary Table S1 (Supporting Informa-
tion).
Data were measured with variable scan speed to ensure
constant statistical precision on the collected intensities.
Three standard reflections were used to check the stability of
the crystal and of the experimental conditions and measured
every 1 h; no significant variation was detected. Data were
corrected for Lorentz and polarization factors and for decay
using the data reduction programs of the MOLEN crystal-
lographic package.36 An empirical absorption correction was
also applied (azimuthal (Ψ) scans of five reflections having ø
> 88°).37 The standard deviations on intensities were calcu-
lated in terms of statistics alone, while those on Fo were
calculated as shown in Table 2.
The structure was solved by a combination of direct and
Fourier methods and refined by full-matrix least squares.
Moreover, from the Fourier difference maps two orientations
for the allyl moiety were clearly visible. Therefore, the central
carbon atom was split into two positions (C(2), C((2a)) and
refined; the occupancy factors are 0.75 and 0.25, respectively.
Anisotropic displacement parameters for all atoms were used
during the final refinement, except for (C(2a)) which was
treated isotropically.
Approximately 0.8 g of the compound [Pd(η3-C3H5)(2)]CF3-
SO3 (4) and of the deuterium analogue were sealed in a
cylindrical Al sample holder and maintained at ∼10 K during
the data collection. Total counting times were approximately
14 h for each sample.
P t(η3-C3H5)(2)]P F 6 (3). The tetramer {PtCl(C3H5)}4 (48.3
mg, 0.0444 mmol) and ligand 2 (81.9 mg, 0.1778 mmol) were
suspended in 3 mL of acetone and stirred for 24 h at room
temperature. Filtration of the resulting suspension through
Celite was followed by removal of the solvent in vacuo.
Addition of ether to the resulting oil produces 128 mg of a solid.
This solid was collected, dissolved in 5 mL of acetone, and
treated with ca. 1.2 equiv of KPF6. After the mixture was
stirred overnight, the acetone was removed and a small
amount of CH2Cl2 added. The solids which formed (KCl, KPF6)
were filtered, and after removal of the solvent, the crude
product was then recrystallized from CH2Cl2/ether. Yield:
97.3 mg (65.0%). Anal. Calcd (found) for C32H38F6OP2PtS
(MW ) 841.7): C, 45.66 (45.49); H, 4.55 (4.56). IR (CsI; ν,
The hydrogen atom bonded to atom O(1) was located on a
Fourier difference map and refined, while for the remaining
hydrogen atoms, only their contribution, in idealized positions
(C-H ) 0.95 Å, B ) 1.5B(carbon) Å2), was taken into account
but not refined. The function minimized was [∑w(|Fo| -
cm-1): 841 (s, P-F). MS (FAB+; m/e): 696.5 (M+ - PF6,
-
100%). 9F NMR (CD2Cl2; δ): -73.46, 1J (P,F) ) 712 Hz, PF6
.
1
31P NMR (δ): 14.0, J (Pt,P) ) 4026 Hz.
1/k|Fc|)2]) with w ) [σ2(Fo)]-1
. No extinction correction was
deemed necessary. The scattering factors used, corrected for
(38) International Tables for X-ray Crystallography; Kynoch: Bir-
mingham, England, 1974; Vol. 4.
(39) Hamilton, W. C. Acta Crystallogr. 1965, 17, 502.
(40) Eckert, J . Physica 1986, 136B, 150.
(35) Seebach, D.; Devaquet, E.; Ernst, A.; Hayakawa, M.; Ku¨hnle,
F. N. M.; Schweizer, W. B.; Weber, B. Helv. Chim. Acta 1995, 78, 1636.
(36) MOLEN: Enraf-Nonius Structure Determination Package;
Enraf-Nonius, Delft, The Netherlands, 1990.
(37) North, A. C. T.; Phillips, D. C.; Mathews, F. S. Acta Crystallogr.,
Sect. A 1968, 24, 351.
(41) Sears, V. F. Thermal Neutron Scattering Lengths and Cross-
Sections for Condensed Matter Research; Atomic Energy of Canada
Ltd., Chalk River Nuclear Laboratories: Chalk River, Ontario, Canada
1984.
(42) Taylor, A. D.; Wood, E. J .; Goldstone, J . A.; Eckert, J . J . Nucl.
Instrum. Methods Phys. Res. 1984, A221, 408.

590 Organometallics, Vol. 16, No. 4, 1997
Albinati et al.
A crystal suitable for X-ray diffraction was obtained by slow
diffusion of ether into a CH2Cl2 solution of 3.
P tCl2(2). The complex was prepared from PtCl2(CH3CN)2
and 2 and was obtained as a colorless solid. Anal. Calcd
(found) for C29H33Cl2OPPtS (MW ) 726.6): C, 47.94 (47.66);
H, 4.58(4.58). NMR data are given in Table 4.
[Ir H2(2)(P P h 3)2]BF 4: yield 70 mg, 70.3%. Anal. Calcd
(found) for C65H65BF4IrOSP3 (MW ) 1266): C, 61.66 (60.89);
H, 5.17 (5.45). MS (FAB+; m/ e): 1179.2 (M+ - BF4, 29.9%).
[Ir H2(2)(p-P (C6H4Cl)3)2]BF 4: yield 105 mg, 91%. Anal.
Calcd (found) for C65H59BCl6F4IrOSP (MW ) 1473): C, 53.00
(52.37); H, 4.04 (4.15). MS (FAB+; m/e): 1385 (M+ - BF4,
4.8%).
[Rh (1,5-COD)(2)]CF 3SO3 (6). 1,5-COD (25 µL, 0.2039
mmol) was added to a solution of [RhCl(1,5-COD)]2 (50.0 mg,
0.1014 mmol) and AgSO3CF3 (52.4 mg, 0.2039 mmol) in 2 mL
of CH2Cl2 using a microsyringe. Stirring for 1 h was followed
by addition of ligand 2 (94.0 mg, 0.2041 mmol) also in 2 mL of
CH2Cl2. The previously orange suspension changed color to
yellow-brown. Stirring for an additional 45 min was followed
by filtration through Celite. Concentration to ca. 0.5 mL and
addition of ether induces precipitation of the product as a
golden powder which was dried in vacuo to afford 148 mg (89%
yield of 5. Anal. Calcd (found) for C38H45O4F3PS2Rh (MW )
838.4): C, 54.41 (54.25); H, 5.65 (5.43). MS (FAB+ m/e): 671.9
(M+ - [SO3CF3-], 30%). 31P NMR (δ): 24.0, d, 1J (Rh,P) )
145.0 Hz. IR (CsI; cm-1): 3434.0 (br, OH bands); 2925 (s),
1283 (s), 1255 (s), 1156 (s), 1027 (s) 696 (m), 637 (s), 529 (m),
519 (s). 1H and 13C NMR data are given in Table 5.
Gen er a l P r ep a r a tion of [Ir H2(2){P (p-XC6H4)3)}2]BF 4 (8;
X ) Cl, F , H, OMe). [IrH2(acetone)2{P(p-XC6H4)3)}2]BF4
(0.0783 mmol) and 2 (0.0783 mmol) were dissolved in 10 mL
of CH2Cl2 and stirred for 40 min. Concentration to ca. 1 mL
was followed by addition of ether to afford a colored powder
in 70-91% yield. The dihydride starting materials were
obtained by passing hydrogen through cold acetone solutions
(6 mL) of the bisphosphine complexes [Ir(1,5-COD){P(p-
XC6H4)3)}2]BF4 for 45 min. Addition of 50 mL of ether induces
precipitation of the dihydride product.
[Ir H2(2)(p-P (C6H4F )3)2]BF 4: yield 76 mg, 71%. Anal.
Calcd (found) for C65H59BF10IrOSP3‚H2O (MW ) 1392): C,
56.03 (55.10); H, 4.38 (4.29). MS (FAB+; m/e+): 1287.4 (M+
BF4, 7.9%).
-
Ack n ow led gm en t. P.P. thanks the Swiss National
Science Foundation and the ETH Zurich for financial
support, as well as J ohnson Matthey for the loan of
precious metals. This work has benefited from the use
of the Manuel Lujan J r. Neutron Scattering Center, a
national facility funded as such by the Department of
Energy, Office of Basic Energy Sciences. A.A. acknowl-
edges support from the CNR.
Su p p or tin g In for m a tion Ava ila ble: For 3a , a figure
giving an additional view, experimental data for the X-ray
diffraction study (Table S1), final positional and isotropic
equivalent displacement parameters (Table S2), calculated
positional parameters for the hydrogen atoms (Table S3),
anisotropic displacement parameters and (Table S4), and an
extended list of bond distances, bond angles, and torsion angles
for 3a (Table S5) (15 pages). Ordering information is given
on any current masthead page.
OM960823Q