[CpRu((R)-Binop-F)(H2O)][SbF6], a New Fluxional Chiral Lewis Acid Catalyst: Synthesis, Dynamic NMR, Asymmetric Catalysis, and Theoretical Studies

Article abstract of DOI:10.1021/ja0374123

The C₂-symmetric electron-poor ligand (R)-BINOP-F (4) was prepared by reaction of (R)-BINOL with bis(pentafluorophenyl)-phosphorus bromide in the presence of triethylamine. The iodo complex [CpRu-((R)-BINOP-F)(I)] ((R)-6) was obtained by substitution of two carbonyl ligands by (R)-4 in the in situ-prepared [CpRu(CO)₂H] complex followed by reaction with iodoform. Complex 6 was reacted with [Ag(SbF ₆)] in acetone to yield [CpRu((R)-BINOP-F)(acetone)][SbF ₆] ((R)-7). X-ray structures were obtained for both (R)-6 and (R)-7. The chiral one-point binding Lewis acid [CpRu((R)-BINOP-F)][SbF₆] derived from either (R)-7 or the corresponding aquo complex (R)-8 activates methacrolein and catalyzes the Diels-Alder reaction with cyclopentadiene to give the [4 + 2] cycloadduct with an exo/endo ratio of 99:1 and an ee of 92% of the exo product. Addition occurs predominantly to the methacrolein C _α-Re face. In solution, water in (R)-8 exchanges readily. Moreover, a second exchange process renders the diastereotopic BINOP-F phosphorus atoms equivalent. These processes were studied by the application of variable-temperature ¹H, ³¹p, and ¹⁷O NMR spectroscopy, variable-pressure ³¹p and ¹⁷O NMR spectroscopy, and, using a simpler model complex, density functional theory (DFT) calculations. The results point to a dissociative mechanism of the aquo ligand and a pendular motion of the BINOP-F ligand. NMR experiments show an energy barrier of 50.7 kJ mol^-1 (12.2 kcal mol^-1) for the inversion of the pseudo-chirality at the ruthenium center.

Full text of DOI:10.1021/ja0374123

Published on Web 03/25/2004
[CpRu((R)-Binop-F)(H2O)][SbF6], a New Fluxional Chiral Lewis
Acid Catalyst: Synthesis, Dynamic NMR, Asymmetric
Catalysis, and Theoretical Studies
,#
Vale´rie Alezra,† Ge´rald Bernardinelli,§ Cle´mence Corminboeuf,*,| Urban Frey,*,
E. Peter Ku¨ndig,*,† Andre´ E. Merbach, Christophe M. Saudan,† Florian Viton,† and
Jacques Weber|
Contribution from the Department of Organic Chemistry and Department of Physical Chemistry
and Laboratory of X-ray Crystallography, UniVersity of GeneVa, CH-1211 GeneVa 4,
Switzerland, and Institute of Chemical Sciences and Engineering,
Swiss Federal Institute of Technology, CH-1015 Lausanne, Switzerland
Received July 20, 2003; E-mail: peter.kundig@chiorg.unige.ch; urban.frey@hevs.ch; clemence.corminboeuf@chiphy.unige.ch
Abstract: The C2-symmetric electron-poor ligand (R)-BINOP-F (4) was prepared by reaction of (R)-BINOL
with bis(pentafluorophenyl)-phosphorus bromide in the presence of triethylamine. The iodo complex [CpRu-
((R)-BINOP-F)(I)] ((R)-6) was obtained by substitution of two carbonyl ligands by (R)-4 in the in situ-prepared
[CpRu(CO)2H] complex followed by reaction with iodoform. Complex 6 was reacted with [Ag(SbF6)] in
acetone to yield [CpRu((R)-BINOP-F)(acetone)][SbF6] ((R)-7). X-ray structures were obtained for both (R)-6
and (R)-7. The chiral one-point binding Lewis acid [CpRu((R)-BINOP-F)][SbF6] derived from either (R)-7
or the corresponding aquo complex (R)-8 activates methacrolein and catalyzes the Diels-Alder reaction
with cyclopentadiene to give the [4 + 2] cycloadduct with an exo/endo ratio of 99:1 and an ee of 92% of
the exo product. Addition occurs predominantly to the methacrolein CR-Re face. In solution, water in (R)-8
exchanges readily. Moreover, a second exchange process renders the diastereotopic BINOP-F phosphorus
1
atoms equivalent. These processes were studied by the application of variable-temperature H, 31P, and
17O NMR spectroscopy, variable-pressure 31P and17O NMR spectroscopy, and, using a simpler model
complex, density functional theory (DFT) calculations. The results point to a dissociative mechanism of the
aquo ligand and a pendular motion of the BINOP-F ligand. NMR experiments show an energy barrier of
50.7 kJ mol-1 (12.2 kcal mol-1) for the inversion of the pseudo-chirality at the ruthenium center.
Introduction
perfluoroaryl phosphinite ligands used in 1-3 create the chiral
environment around the coordination site of the enal and offset
the donor properties of the cyclopentadienyl- and indenyl-
ligands.
Lewis acids play a key role as catalysts in organic synthesis,
and applications advance at a rapid pace.1 Our work in this area
has focused on the development of stable, chiral, single point
coordination transition metal Lewis acids that incorporate C2-
symmetric electron-poor bidentate ligands. The readily prepared,
mild Fe and Ru Lewis acids 1-3 activate enals and effectively
catalyze their asymmetric Diels-Alder reactions with dienes
and 1,3-dipolar cycloaddition reactions with nitrones.2,3 The
* To whom correspondence should be addressed: E.P.K. (synthesis,
catalysis), U.F. (NMR), C.C. (calculations).
† Department of Organic Chemistry, 30 Quai Ernest Ansermet.
§ Laboratory of X-ray Crystallography, 24 Quai Ernest Ansermet.
| Department of Physical Chemistry, 30 Quai Ernest Ansermet.
Institute of Molecular and Biological Chemistry, EPFL-BCH.
# Present address: University of Applied Sciences Valais (HEVs),
Chemistry Department, Rte du Rawyl 64, CH-1950 Sion, Switzerland.
(1) Handbook of Lewis AcidssApplication in Organic Synthesis; Yamamoto,
H., Ed.; Wiley-VCH: Weinheim, 2000.
Bidentate C2-symmetric chiral ligands have been widely used
in asymmetric catalysis.4 Chiral bidentate ligands without C2-
(2) (a) Ku¨ndig, E. P.; Bourdin, B.; Bernardinelli, G. Angew. Chem., Int. Ed.
Engl. 1994, 33, 1856. (b) Bruin, M. E.; Ku¨ndig, E. P. Chem. Commun.
1998, 2635. (c) Ku¨ndig, E. P.; Saudan, C. M.; Bernardinelli, G. Angew.
Chem., Int. Ed. 1999, 38, 1220. (d) Ku¨ndig, E. P.; Saudan, C. M.; Viton,
F. AdV. Synth. Catal. 2001, 343, 51. (e) Ku¨ndig, E. P.; Saudan, C. M.;
Alezra, V.; Viton, F.; Bernardinelli, G. Angew. Chem., Int. Ed. 2001, 40,
4481.
(3) Viton, F.; Bernardinelli, G.; Ku¨ndig, E. P. J. Am. Chem. Soc. 2002, 124,
4968.
(4) (a) Whitesell, J. K. Chem. ReV. 1989, 89, 1581. (b) Brunner, H.; Zettlmeier,
W. Handbook of EnantioselectiVe Catalysis; Verlag Chemie: Weinheim,
1993; Vol. 2. (c) Noyori, R. Asymmetric Catalysis; John Wiley: New York,
1994.
9
10.1021/ja0374123 CCC: $27.50 © 2004 American Chemical Society
J. AM. CHEM. SOC. 2004, 126, 4843-4853
4843

A R T I C L E S
Alezra et al.
Scheme 1
symmetry are more recent but powerful tools for asymmetric
catalysis.5 In CpML1L2L3 and analogous arene complexes, the
pseudotetrahedral metal is a stereogenic center. The use of chiral,
dissymmetric bidentate L1-L2 ligands then leads to diastere-
oisomers that differ in the configuration at the metal center.6,7
Even when isolated, single diastereoisomers of these optically
active complexes are configurationally unstable at the metal
center under catalysis conditions and epimerize easily to afford
mixtures of diastereomeric complexes.8 Some of them are
nevertheless efficient catalysts for a number of asymmetric
transformations because for steric and/or electronic reasons only
one of the diastereoisomeric complexes is an efficient catalyst.7d-f
The pseudotetrahedral geometry makes this compound class
particularly suitable for the study of configurational stability
and configurational preference of the metal center and its role
in enantioselective catalysis.9 The ultimate way to investigate
racemization of a chiral metal center consists of the use of
asymmetric complexes devoid of ligand-based chirality.9,10 Over
the years, a number of groups have focused on applications in
organic transformations of such enantiomerically pure piano-
stool complexes. To the best of our knowledge, complexes with
metal-based rather than ligand-based chirality have not been
successfully used in asymmetric catalysis. Elegant and pioneer-
ing stoichiometric applications stem from the groups of Brun-
ner,11 Davies,12 Faller,13 and Gladysz,14 among others. For
asymmetric catalysis, configurational stability of the catalyst is
of essence, as racemization would result in rapid erosion of
enantiomeric purity of the product. Investigation of the race-
mization of optically active complexes devoid of ligand-based
chirality is inherently limited to chiroptical measurements.9c,d,10b
Computational studies have focused on the role of the ligands
and the electronic configuration at the metal.15 Quantitative
experimental data on the process of inversion at the metal is
still scarce however.8,9c,d In this article we report on a new
BINOL-based Ru-Lewis acid catalyst, which catalyzes the
asymmetric Diels-Alder reaction of methacrolein and cyclo-
pentadiene.16 Its NMR characteristics suggested the complex
to be suitable to probe the mechanism of the inversion of the
pyramidal geometry at the metal center and establish an energy
diagram for this process. The mechanistic tools used in this study
(5) For reviews, see: (a) Inoguchi, K.; Sakuraba, S.; Achiwa, K. Synlett 1992,
169. (b) Frost, C. G.; Howarth, J.; Williams, J. M. J. Tetrahedron:
Asymmetry 1992, 3, 1089. (c) Reiser, O. Angew. Chem., Int. Ed. Engl. 1993,
32, 547. For representative examples, see: (d) Jacobsen, E. N.; Zhang,
W.; Gu¨ler, M. L. J. Am. Chem. Soc. 1991, 113, 6703. (e) Nishiyama, H.;
Yamaguchi, S.; Kondo, M.; Itoh, K. J. Org. Chem. 1992, 57, 4306. (f)
Rieck, H.; Helmchen, G. Angew. Chem., Int. Ed. Engl. 1995, 34, 2687. (g)
Schnyder, A.; Hintermann, L.; Togni, A. Angew. Chem., Int. Ed. Engl.
1995, 34, 931. (h) Rajanbabu, T. V.; Casalnuovo, A. L. J. Am. Chem. Soc.
1996, 118, 6325. (i) Faller, J. W.; Lloret-Fillol, J.; Parr, J. New J. Chem.
2002, 26, 883.
include variable-temperature H, 31P, and 17O NMR spectros-
1
copy, variable-pressure 31P and 17O NMR spectroscopy,17 and,
using a simpler model complex, density functional theory (DFT)
calculations
(6) For recent examples, see: (a) Attar, S.; Nelson, J. H.; Fischer, J.; de Cian,
A.; Sutter, J.-P.; Pfeffer, M. Organometallics 1995, 14, 4559. (b) Brunner,
H.; Oeschey, R.; Nuber, B. Organometallics 1996, 15, 3616. (c) Carmona,
D.; Lahoz, F. J.; Elipe, S.; Oro, L. A.; Lamata, M. P.; Viguri, F.; Mir, C.;
Cativiela, C.; Lopez-Ram de Viu, M. P. Organometallics 1998, 17, 2986.
(d) Carmona, D.; Vega, C.; Lahoz, F. J.; Elipe, S.; Oro, L. A.; Lamata, M.
P.; Viguri, F.; Garcia-Correas, R.; Cativiela, C.; Lopez-Ramde Viu, M. P.
Organometallics 1999, 18, 3364. (e) Therrien, B.; Koenig, A.; Ward, T.
R. Organometallics 1999, 18, 1565. (f) Therrien, B.; Ward, T. R. Angew.
Chem., Int. Ed. 1999, 38, 405. (g) Gul, N.; Nelson, J. H. Organometallics
1999, 18, 709. (h) Faller, J. W.; Parr, J. Organometallics 2000, 19, 3556.
(i) Drommi, D.; Faraone, F.; Francio, G.; Belletti, D.; Graiff, C.; Tiripicchio,
A. Organometallics 2002, 21, 761.
Results and Discussion
Synthesis and Characterization of [CpRu((R)-BINOP-F)I]
Complex. The C2-symmetric electron-poor ligand (R)-BINOP-F
4 was prepared by reaction of (R)-BINOL with bis(pentafluo-
rophenyl)-phosphorus bromide in the presence of triethyl-
amine.18 The hydride complex (R)-5 was obtained by substitu-
tion of two carbonyl ligands by (R)-4 in the in situ-prepared
[CpRu(CO)2H] complex (Scheme 1).19,20
(7) For recent examples, see: (a) Carmona, D.; Elipe, S.; Lahoz, F. J.; Oro, L.
A.; Cativiela, C.; Pilar Lopez-Ram de Viu, M.; Pilar Lamata, M.; Vega,
F.; Viguri, C. Chem. Commun. 1997, 2351. (b) Davenport, A. J.; Davies,
D. L.; Fawcett, J.; Garratt, S. A.; Russell, D. R. Chem. Commun. 1999,
2331. (c) Davenport, A. J.; Davies, D. L.; Fawcett, J.; Garratt, S. A.; Russell,
D. R. J. Chem. Soc., Dalton Trans. 2000, 4432. (d) Faller, J. W.; Grimmond,
B. J.; D’Alliessi, D. G. J. Am. Chem. Soc. 2001, 123, 2525. (e) Faller, J.
W.; Parr, J. Organometallics 2001, 20, 697. (f) Faller, J. W.; Grimmond,
B. J. Organometallics 2001, 20, 2454. (g) Faller, J. W.; Parr, J.; Lavoie,
A. R. New J. Chem. 2003, 27, 899.
(14) (a) Dalton, D. M.; Fernandez, J. M.; Emerson, K.; Larsen, R. D.; Arif, A.
M.; Gladysz, J. A. J. Am. Chem. Soc. 1990, 112, 9198. (b) Dalton, D. M.;
Garner, C. M.; Fernandez, J. M.; Gladysz, J. A. J. Org. Chem. 1991, 56,
6823. (c) Klein, D. P.; Gladysz, J. A. J. Am. Chem. Soc. 1992, 114, 8710.
(d) Gladysz, J. A.; Boone, B. J. Angew. Chem., Int. Ed. Engl. 1997, 36,
550.
(15) (a) Ward, T. R.; Schafer, O.; Daul, C.; Hofmann, P. Organometallics 1997,
16, 3207. (b) Costuas, K.; Saillard, J.-Y. Organometallics 1999, 18, 2505.
(c) Smith, K. M.; Poli, R.; Legzdins, P. Chem. Commun. 1998, 1903. (d)
Yang, Y.; Asplund, M. C.; Kotz, K. T.; Wilkens, M. J.; Frei, H.; Harris,
C. B. J. Am. Chem. Soc. 1998, 120, 10154.
(8) Brunner, H.; Zwack, T. Organometallics 2000, 19, 2423.
(9) (a) Brunner, H. AdV. Organomet. Chem. 1980, 18, 151. (b) Consiglio, G.;
Morandini, F. Chem. ReV. 1987, 87, 761. (c) Brunner, H. Angew. Chem.,
Int. Ed. 1999, 38, 1194. (d) Brunner, H. Eur. J. Inorg. Chem. 2001, 905.
(10) (a) Brunner, H. Angew. Chem., Int. Ed. Engl. 1969, 8, 38. (b) Brunner, H.
Acc. Chem. Res. 1979, 12, 250.
(16) (a) Kagan, H. B.; Riant, O. Chem. ReV. 1992, 92, 1007. (b) Pindur, U.;
Lutz, G.; Otto, C. Chem. ReV. 1993, 93, 741. (c) Oh, T.; Reilly, M. Org.
Prep. Proced. Int. 1994, 26, 129. (d) Dias, L. C. J. Braz. Chem. Soc. 1997,
8, 289. (e) Corey, E. J.; Guzman-Perez, A. Angew. Chem., Int. Ed. 1998,
37, 389. (f) Saudan, C.; Ku¨ndig, E. P. In Handbook of Lewis Acids;
Yamamoto, H., Ed.; VCH-Wiley: Weinheim, 2000; pp 597-652. (g)
Nicolaou, K. C.; Snyder, S. A.; Montagnon, T.; Vassilikogiannakis, G.
Angew. Chem., Int. Ed. 2002, 41, 1668. (h) Corey, E. J. Angew. Chem.,
Int. Ed. 2002, 41, 1650.
(11) (a) Brunner, H.; Aclasis, J.; Langer, M.; Steger, W. Angew. Chem., Int.
Ed. Engl. 1974, 13, 810. (b) Brunner, H.; Fisch, K.; Jones, P. G.; Salbeck,
J. Angew. Chem., Int. Ed. Engl. 1989, 28, 1521.
(12) (a) Davies, S. G. Pure Appl. Chem. 1988, 60, 13. (b) Davies, S. G.
Aldrichim. Acta 1990, 23, 31.
(13) (a) Schilling, B. E. R.; Hoffmann, R.; Faller, J. W. J. Am. Chem. Soc.
1979, 101, 592. (b) Faller, J. W.; Linebarrier, D. L. Organometallics 1990,
9, 3182. (c) Faller, J. W.; Ma, Y. Organometallics 1992, 11, 2726.
(17) Helm, L.; Merbach, A. E. J. Chem. Soc., Dalton Trans. 2002, 633.
9
4844 J. AM. CHEM. SOC. VOL. 126, NO. 15, 2004

[CpRu((R)-Binop-F)(H O)][SbF ]
A R T I C L E S
2
6
Figure 1. ORTEP view of the crystal structure of [CpRu((R)-BINOP-F)I]
(R)-6 with 40% probability ellipsoids. Selected bond lengths (Å) and bond
Figure 2. ORTEP view of the crystal structure of [CpRu((R)-BINOP-F)-
(acetone)][SbF6] (R)-7 with 30% probability ellipsoids. Selected bond
lengths (Å) and bond angles (deg): Ru-P ) 2.306(3), 2.297(3); Ru-O )
2.162(8); Cp(mean plane)‚‚‚Ru ) 1.872(1); P-Ru-P ) 100.3(1); O-Ru-P
) 81.2(2), 105.0(2).
angles (deg): Ru-P ) 2.298(3), 2.271(4); Ru-I ) 2.735(1); Cp(mean plane)
‚
‚‚Ru ) 1.851(1); P-Ru-P ) 99.0(1); I-Ru-P ) 84.0(1), 107.2(1).
The hydride-labilizing effect was crucial to the success of
CO substitution.21 Attempts to carry out this ligand exchange
in the analogous chloro complex were not successful. The robust
air-stable hydrido complex (R)-5 was isolated from the crude
mixture by column chromatography (aluminum oxide, neutral).
Its 1H NMR spectrum in CD2Cl2 showed a doublet of doublets
at δ ) -11.82 ppm for the Ru-H proton.22 Subsequent treatment
with iodoform in refluxing acetone afforded the air-stable iodo
complex (R)-6.23 Crystals suitable for X-ray diffraction analysis
were obtained from a CH2Cl2/MeOH solution (Figure 1).
Scheme 2
Synthesis, Characterization, and Catalytic Behavior of
[CpRu((R)-BINOP-F)L][SbF6] (L ) Acetone, H2O) Com-
plexes. Reaction of a red-orange solution of (R)-6 in acetone/
CH2Cl2 with AgSbF6 generated the yellow cationic complex
(R)-7, which was isolated by precipitation with Et2O after
concentration of the acetone/CH2Cl2 mixture (Scheme 2).
Scheme 3
Crystals suitable for X-ray diffraction analysis were obtained
by slow diffusion of Et2O into a solution of (R)-7 in acetone
(Figure 2).
The Diels-Alder reaction of methacrolein and cyclopenta-
diene (1 mol % catalyst, CH2Cl2, -20 °C) was used as a test
reaction. Reaction occurred at a similar rate as that of the
previously reported complexes [CpRu(BIPHOP-F)L][SbF6] (L
) acetone, H2O) 2.2c,d The cycloadduct 9 was isolated in 92%
yield, with a exo:endo diastereomeric ratio of 99:1, an enanti-
oselectivity of 92% ee (exo) and 2S absolute configuration
(Scheme 3). The aquo complex (R)-8 was formed upon addition
of water to a solution of (R)-7 in acetone.24 The aquo ligand is
labile, and this provides another entry to the active catalytic
species, the results obtained in the test Diels-Alder reaction
being identical in terms of activity and selectivity to those
obtained with the acetone complex (R)-7. Ligand 4 forms a nine-
membered metallacycle upon coordination to Ru, whereas
BIPHOP-F (e.g., in complex 2) forms a seven-membered
metallacycle. The effect on the shape of the chiral site of the
catalyst is minor though, and it looks very similar to that of 2.
The enantioselectivity and the sense of asymmetric induction
are identical in both cases.
(18) Ku¨ndig, E. P.; Dupre´, C.; Bourdin, B.; Cunningham, Jr., A. F.; Pons, D.
HelV. Chim. Acta 1994, 77, 421.
(19) For a listing of synthetic routes to [CpRuLL′X] complexes, see: Shen, J.;
Stevens, E. D.; Nolan, S. P. Organometallics 1998, 17, 3000.
(20) Bruce, M. I.; Jensen, C. M.; Jones, N. L. Inorg. Synth. 1990, 28, 216.
(21) Baldovino, C.; Cesarotti, E.; Prati, L. Gazz. Chim. Ital. 1992, 122, 475
and references therein.
Upon complexation to the metal, the C2-symmetry of the free
ligand (R)-4 is lost. Indeed, in the pseudotetrahedral complexes
(22) (η6-Arene)RuHL2 complexes: (a) Deeming, A. J.; Speel, D. M.; Stchedroff,
M. Organometallics 1997, 16, 6004. (b) Wiles, J. A.; Bergens, S. H.
Organometallics 1998, 17, 2228. (c) Geldbach, T. J.; Pregosin, P. HelV.
Chim. Acta 2002, 85, 3937. (d) Geldbach, T. J.; Pregosin, P.; Albinati, A.
J. Chem. Soc., Dalton Trans. 2002, 2419.
(24) Aquo complexes of organometallic Lewis acids are frequently observed.
Some examples are given in refs 6c,d,f and 7a-c,e,f and in the following:
(a) Takahashi, Y.; Akita, M.; Hikichi, S.; Morooka, Y. Inorg. Chem. 1998,
37, 3186. (b) Takahashi, Y.; Hikichi, S.; Akita, M.; Moro-oka, Y. Chem.
Commun. 1999, 1491.
(23) Bruce, M. I.; Humphrey, M. G.; Swincer, A. G.; Wallis, R. C. Aust. J.
Chem. 1984, 37, 1747.
9
J. AM. CHEM. SOC. VOL. 126, NO. 15, 2004 4845

A R T I C L E S
Alezra et al.
Figure 5. Inversion of (R)-8 via a η2-η1-η2 hemidissociation mechanism
between two identical structures of the complexes and an
inversion at the chirotopic metal center.25 Literature precedent
includes the indenyl ruthenium complex 32e and arene ruthe-
nium26 and Cp* rhodium27 complexes bearing C2-symmetric
bisoxazoline ligands. For more detailed mechanistic investiga-
tion we chose the aquo complex (R)-8 rather than the acetone
complex (R)-7, which was always contaminated by small
amounts of 8 due to the presence of residual traces of water.
Mechanism. The temperature-dependence of the NMR
spectra of (R)-8 is attributed to the interconversion of the two
identical structures A and G (Figures 5 to 8). If we limit the
analysis to processes that involve either 18- or 16-electron
intermediates, a first possible mechanism for this inversion at
the chirotopic metal center is dissociation/recoordination of one
of the P termini of the chiral chelating ligand as shown in Figure
5.
Alternatively, and much more likely in view of the behavior
of 8 in catalysis, interconversion may take place via a sequence
that involves dissociation of the water molecule, inversion of
the nine-membered metallacycle in [CpRu((R)-BINOP-F)]+, and
reassociation of the H2O ligand. In turn, this process may occur
either via a dissociative D mechanism (Figure 6) or a dissocia-
tive interchange Id (Figure 7). In addition, this swing motion
may also proceed through an associative H2O exchange with
inversion at ruthenium and haptotropic η5 / η3 rearrangement
of the Cp ligand (Figure 8).
Figure 3. Variable-temperature 31P NMR stack plot of [CpRu((R)-BINOP-
F)(H2O)][SbF6] (R)-8 ([(R)-7] ) 0.05 mol l-1, [H2O] ) 1.20 mol l-1
acetone-d6).
,
Figure 4. Variable-temperature 1H NMR stack plot of [CpRu((R)-BINOP-
F)(H2O)][SbF6] (R)-8 ([(R)-7] ) 0.05 mol l-1, [H2O] ) 1.20 mol l-1
acetone-d6).
,
[CpRu(PP)X], where PP is a C2-symmetric bidentate ligand,
the two diastereotopic phosphorus give a characteristic AB
quartet. This AB pattern is actually observed in 31P NMR
(acetone-d6) of the hydrido and iodo complexes (R)-5 and (R)-
6. In contrast, the 31P NMR spectra (acetone-d6) of the cationic
complexes (R)-7 and (R)-8 at room temperature show a singlet
for the two diastereotopic P nuclei of the bidentate ligand. The
presence of a singlet resulting from coincidental signal overlap
was excluded by the observation of the usual AB quartet at
-40 °C for (R)-7 and (R)-8 (Figure 3). The previously reported
indenyl complexes 3 showed similar 31P NMR behavior,2e very
different from the cyclopentadienyl analogues 2, which gave
an AB quartet (acetone-d6, room temperature).2c
An analysis of the mechanistic schemes in Figures 6-8 was
carried out by a combination of variable-temperature and
-pressure 31P and 17O NMR experiments and DFT calculations.
Mechanism Attribution. (a) 17O NMR; Water Exchange.
The 17O NMR spectra of a solution of 0.05 mol L-1 (R)-8 and
0.5 mol L-1 38% 17O enriched H2O in acetone-d6 at 273 K
shows a signal at 0 ppm for the free water (reference) and a
small and broad signal at -160 ppm corresponding to the
coordinated water molecule. The water exchange on (R)-8 in
acetone (eq 1) was followed by 17O NMR spectroscopy, and a
line shape analysis of the resulting NMR spectra afforded the
detailed kinetic parameters. The derived kinetic parameters for
the variable-temperature and -pressure studies of water exchange
on (R)-8 are shown in Table 1. The water exchange rate constant
1H and 13C NMR spectra of complexes (R)-7 and (R)-8 in
acetone-d6 at room temperature showed a simple system of
signals for the coordinated (R)-BINOP-F ligand (10 carbon
signals and 6 proton signals for the binaphthyl backbone),
similar to those observed for the noncoordinated (R)-BINOP-F
(4). In comparison, complexes (R)-5 and (R)-6, as expected,
display different signals for all 20 carbons and 12 protons of
the binaphthyl backbone. On cooling (R)-8 to -40 °C, the
resonances assigned to the (R)-BINOP-F ligand split into two
(kex) observed in acetone solutions ([(R)-8] ) 0.025 mol L-1
)
1
sets and became sharper (Figure 4). In addition, the H NMR
(25) (a) Mislow, K. J. Am. Chem. Soc. 1984, 106, 3319. (b) Eliel, E. L.; Wilen,
S.; Mander, L. N. Stereochemistry of Organic Compounds; Wiley & Sons:
New York, 1994; pp 53 and 1195.
(26) (a) Asano, H.; Katayama, K.; Kurosawa, H. Inorg. Chem. 1996, 35, 5760.
(b) Kurosawa, H.; Asano, H.; Miyaki, Y. Inorg. Chim. Acta 1998, 270,
87. (c) Faller, J. W.; Lavoie, A. J. Organomet. Chem. 2001, 630, 17. (d)
Davies, D. L.; Fawcett, J.; Garratt, S. A.; Russell, D. R. Organometallics
2001, 20, 3029.
(27) Davies, D. L.; Fawcett, J.; Garratt, S. A.; Russell, D. R. J. Organomet.
Chem. 2002, 662, 43.
spectrum of the aquo complex (R)-8 in acetone-d6 at -40 °C
showed two singlets at 5.50 and 3.91 ppm for the coordinated
and free water, respectively. At room temperature, a single,
averaged signal at 3.18 ppm is apparent.
The above observations point to a fluxional process that
renders the P atoms equivalent and results in a rapid exchange
9
4846 J. AM. CHEM. SOC. VOL. 126, NO. 15, 2004

[CpRu((R)-Binop-F)(H O)][SbF ]
A R T I C L E S
2
6
Table 1. Derived Kinetic Parameters for the Variable-Temperature
and -Pressure Studies of Water Exchange and Swing Motion on
(R)-8 in Acetone
water exchange
swing motion
k298a (s-1
∆Hq (kJ mol-1
)
29 200 ( 400
50.57 ( 1.4
10.15 ( 4.9
4.8 ( 0.6
8370 ( 630
79.14 ( 1.8
95.60 ( 6.5
11.6 ( 0.3
)
∆Sq (J mol-1 K-1
)
)
)
∆Vqb (cm3 mol-1
∆Gq298 (kJ mol-1
47.54 ( 0.03
50.65 ( 0.40
a kex (water exchange); ksw (swing motion). b At 293.4 K.
dissociative mode of activation, and the modest value of ∆Vq
does support an Id mechanism for the water molecule exchange.
*k
h
[{Ru}(OH2)] + H2O ex[{Ru}(*OH )] + H O
(1)
2
2
(R)-8
(R)-8
Parameters for water and acetonitrile exchange on a number
of ruthenium(II) complexes are given in Table 2. Both H2O
and MeCN are greatly labilized by nonleaving π-bonding
ligands. The kinetic trans effect increases the lability by 6 ×
1010-fold along the sequence [Ru(MeCN)6]2+ < [Ru(η6-C6H6)-
(MeCN)3]2+ < [Ru(η5-C5H5)(MeCN)3]+ (Table 2, entries 1-3)
and is a consequence of a corresponding decrease in ∆Hq. It is
also accompanied by an increasingly strong d-activation mode
as indicated by the growing ∆Vq. A similar trend is also
observed for the aquo complexes (Table 2, entries 4-6). We
therefore can explain the high rate constant, as well as the
strengthened d-activation mode on (R)-8, by the strong labilizing
effect of the η5-C5H5 ligand (Table 2, entries 3 and 6).
(b) 31P NMR; Inversion at Ru. The kinetic parameters for
the swing motion (eq 2, Figures 5-8) were determined using
variable-temperature and variable-pressure 31P NMR spectros-
copy.
Figure 6. Inversion of (R)-8 via a two-step dissociative D mechanism.
Detailed kinetic parameters of the 31P NMR study are shown
in Table 1. No dependence of the swing motion rate constant
(ksw) with respect to added HSbF6 was observed. The positive
volume and entropy of activation are in accordance with a
dissociative mode of activation (Id or D mechanisms). Consider-
ing the high ∆Vq value along with a positive ∆Sq value, it is
tempting to assign this to a η2-η1-η2 hemidissociation process
of the (R)-BINOP-F ligand. However, neither synthetic proper-
ties nor NMR measurements substantiate this interpretation.
From the observation of different inversion barriers in three
Cp*Rh(bisoxazoline) complexes, Davies et al. also concluded
that hemidissociation of one oxazoline moiety was an unlikely
process.27 While we cannot completely exclude this possibility,
a more likely event is the dissociation of water that triggers the
movement of inversion. In addition, failure to observe similar
fluxional behaviors in the hydrido and iodo analogues (R)-5
and (R)-6 support this interpretation.26,31
Figure 7. Inversion of (R)-8 via a dissociative interchange Id mechanism.
Figure 8. Inversion of (R)-8 via an associative mechanism with haptotropic
η5 / η3 rearrangement of the Cp ligand.
showed no water concentration dependence ([H2O] ) 0.56-
2.25 mol L-1) at 291 K (see Supporting Information). Also, no
dependence with added HSbF6 was observed (see Supporting
Information). The positive volume of activation, as well as the
first-order rate constant, first order in complex concentration
and zero order in water concentration, are in accordance with a
(28) Rapaport, I.; Helm, L.; Merbach, A. E.; Bernhard, P.; Ludi, A. Inorg. Chem.
1988, 27, 873.
(29) Luginbu¨hl, W.; Zbinden, P.; Pittet, P.-A.; Armbruster, T.; Bu¨rgi, H.-B.;
Merbach, A. E.; Ludi, A. Inorg. Chem. 1991, 30, 2350.
9
J. AM. CHEM. SOC. VOL. 126, NO. 15, 2004 4847

A R T I C L E S
Alezra et al.
Table 2. Parameters for Solvent Exchange on Ruthenium(II) Complexes
a
q
q
q
3
entry
complex
kex298 (s-1
)
∆H (kJ mol-1
)
∆S (J mol-1 K-1
)
∆V (cm mol-1
)
ref
1
2
3
4
5
6
[Ru(CH3CN)6]2+
8.9 × 10-11
4.1 × 10-5
5.6
140.3
102.5
86.5
87.8
75.9
50.6
33.3
15.0
59.6
16.1
29.9
10.2
0.4
2.4
11.1
-0.4
1.5
28
29
29
28
30
this study
[Ru(η6-C6H6)(CH3CN)3]2+
[Ru(η5-C5H5)(CH3CN)3]+
[Ru(H2O)6]2+
1.8 × 10-2
[Ru(η6-C6H6)(H2O)3]2+
(R)-8
11.5
2.9 × 104
4.8
a Rate constant for the exchange of a particular coordinated solvent molecule.
As shown in Table 1, the rate constant of the water exchange
process kex is about three times higher than the one of the (R)-
BINOP-F swing motion, ksw. This may reflect the two different
possibilities of a water molecule to be exchanged in a dissocia-
tive mechanism. Indeed, the entering water molecule may
exchange either from the same side or from the opposite side
of the outgoing ligand, only the second possibility implying a
swing motion of the bidentate ligand. To complete experimental
observations and provide further information on the exact
dissociative process, D and Id mechanisms were subjected to
computational analysis. Note that the associative mechanisms
have not been considered in our calculations, since the activation
parameters obtained previously do not support such a mecha-
nism. Moreover, it has been shown previously that the approach
of an additional ligand to a coordinatively saturated Ru(II)
complex is electrostatically disfavored.32
(c) Calculations. Geometry optimization and electronic
structure calculations were performed with the Gaussian98 and
Gaussian03 programs.33 The DZVP34 basis set were employed
with density functional theory using the PW91 functional.35
To reduce computation time, most of the calculations were
performed on a truncated complex, in which the ligand BINOP-F
was replaced by H2PO(C6H4)2OPH2.
lecular mechanics (QM/MM) approaches seem to be well suited
for performing explicit solvation simulations.37,38 For the
calculation of solvent exchange or racemization processes, the
solvent is frequently neglected39-42 or treated implicitly using
a continuum model.43,44 In this study, solvent effects were taken
into account statically by means of polarized continuum model
(PCM) calculations45 using a self-consistent field technique as
implemented in Gaussian03.33b,45d In this model a quantum
mechanically treated solute molecule is placed within the so-
called solute cavity. The shape and dimensions of the cavity
are determined by the molecular structure. The cavity is
surrounded by a continuum dielectric polarized by the solute
molecule. In turn, the dielectric polarization generates an
electrostatic field at the solute molecule, modifying its electron
density. Free energies of solvation were calculated with acetone
as a solvent following the experimental conditions (ꢀ ) 20.7).
The geometries were kept optimized as for the gas-phase species
(single-point calculations). Due to the rigidity of the structures
involved, optimization within PCM changes the relative energies
and bond lengths by less than 0.5 kcal mol-1 and 0.5 Å,
respectively, as it was tested for two geometries (see Table S-11
and S-13 in Supporting Information).
While this simplification appears severe, it is expected to
provide helpful insight into the mechanism operative in other
[CpRu(PP)L]+-based compounds such as (R)-8. Of course, when
considering the energy barrier, we will have to keep in mind
the electronic effect caused by the π-acid BINOP-F ligand. As
a further approximation, the counterion as well as the explicit
quantum-mechanical treatment of solvent were neglected, as it
would have been too time-consuming with today’s computa-
tional possibilities.36 However, quantum mechanics and mo-
Minimum energy optimizations of the gas-phase structures
were based on complete geometry optimizations. Transition state
calculations were obtained following the correct normal mode
manually or using the transition-state optimization type as
implemented in the Gaussian98 program.33a Frequency calcula-
(30) Stebler-Ro¨thlisberger, M.; Hummel, W.; Pittet, P.-A.; Bu¨rgi, H.-B.; Ludi,
A.; Merbach, A. E. Inorg. Chem. 1988, 27, 1358.
(31) For related discussions, see for example: (a) Slugovc, C.; Simanko, W.;
Mereiter, K.; Schmid, R.; Kirchner, K.; Xiao, L.; Weissensteiner, W.
Organometallics 1999, 18, 3865. (b) Faller, J. W.; Grimmond, B. J.; Curtis,
M. Organometallics 2000, 19, 5174.
(32) De Vito, D.; Sidorenkova, H.; Rotzinger, F. P.; Weber, J.; Merbach, A. E.;
Inorg. Chem. 2000, 39, 5547.
(37) (a) Gao, J. In ReViews in Computational Chemistry; Lipkowitz, K. B., Boyd,
D. B., Eds.; VCH: New York, 1996; Vol. 7. (b) Gao, J.; Xia, X. Science
1992, 258, 631. (c) Gao, J.; Xia, X. J. Phys. Chem. 1992, 96, 537.
(38) Woo, T. K.; Blo¨chl, P. E.; Ziegler, T. J. Mol. Struct. (THEOCHEM) 2000,
506, 313.
(39) Rotzinger, F. P. J. Am. Chem. Soc. 1996, 118, 6760.
(40) Kowall, Th.; Caravan, P.; Bourgeois, H.; Helm, L.; Rotzinger, F. P.;
Merbach, A. E. J. Am. Chem. Soc. 1998, 120, 6569.
(41) Hartmann, M.; Clark, T.; van Eldik, R. J. Phys. Chem. A 1999, 103, 9899.
(42) Robitzer, M.; Ritleng, V.; Sirlin, C.; Dedieu, A.; Pfeffer, M. C. R. Chimie
2002, 5, 467.
(43) Cramer, C. J.; Truhlar, D. G. In ReViews in Computational Chemistry;
Lipkowitz, K. B., Boyd, D. B., Eds.; VCH: New York, 1995; Vol. 6.
(44) Tomasi, J. Chem. ReV. 1994, 94, 2027.
(45) (a) Miertus, S.; Tomasi, J. Chem. Phys. 1982, 65, 239. (b) Miertus, S.;
Scorcco, E.; Tomasi, J. J. Chem. Phys. 1981, 55, 117. (c) Mennucci, B.;
Cance`s, E.; Tomasi, J. J. Phys. Chem. B 1997, 101, 10506. (d) Cossi, M.;
Scalmani, G.; Rega, N.; Barone, V. J. Chem. Phys. 2002, 117, 43. (e)
Scalmani, G.; Barone, V.; Kudin, K. N.; Pomelli, C. S.; Scuseria, G. E.;
Frisch, M. J. Theo. Chem. Acc. 2003, submitted.
(33) (a) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A.
D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi,
M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.;
Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.;
Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.;
Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,
P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-
Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe,
M.; Gill, P. M. W.; Johnson, B. G.; Chen, W.; Wong, M. W.; Andres, J.
L.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98, revision
A.8; Gaussian, Inc.: Pittsburgh, PA, 1998. (b) Gaussian 03, revision A.1;
Gaussian, Inc., Pittsburgh, PA, 2003.
(34) Godbout, N.; Salahub, D. R.; Andzelm, J.; Wimmer, E. Can. J. Chem.
1992, 70, 560.
(35) Perdew; Wang, Y. Electronic Structure of Solids ’91; 1991; p 11.
(36) Erras-Hanauer, H.; Clark, T.; van Eldik, R. Coord. Chem. ReV. 2003, 238,
233.
9
4848 J. AM. CHEM. SOC. VOL. 126, NO. 15, 2004

[CpRu((R)-Binop-F)(H O)][SbF ]
A R T I C L E S
2
6
Table 3. Geometrical Parameters of (R)-8 at the PW91/3-21G* Level and Truncated A-D and H at the PW91/DZVP Level Compared to
the X-ray Structure of (R)-7
truncated PW91/DZVP
entire PW91/3-21G*
A
B
C
D
H
(R)-8
(R)-7 X-ray
R(Ru-P) Å
2.30
2.33
2.32
1.68
1.85
2.37
1.67
1.82
2.36
1.68
1.81
3.81/3.85
107.2
126.3
99.2
2.28
1.66
1.92
2.30
1.61
1.87
R(P-O) Å
1.69
1.88
2.28
99.2
122.4
98.4
1.70
1.85
3.25
99.3
123.5
98.7
R(Cpcentro¨ıd-Ru) Å
R(Ru-L) Å
2.18
2.16
θ(P-Ru-P)
99.9
124.8
99.0
102.6
128.7
98.8
100.1
-179.0
98.5
102.1
121.0
98.8
100.3
123.5
98.7
θ(P-Ru-Cpcentro¨ıd
)
θ(H-P-H) or θ(C-P-C)
θ(H-P-O) or θ(C-P-O)
æ (Cpcentro¨ıd-Ru-P-P)
æ (H-P-H-O)
98.5
98.9
99.8
99.7
97.6
97.0
-138.5
-141.2
-144.2
-175.0
-137.9
-136.6
96.7
96.5
98.1
99.9
97.1
98.8
or æ (C-P-C-O)
Table 4. Energies, Enthalpies, and Free Energies in the Gas Phase for the Racemization of A at the PW91/DZVP Level and PCM
Estimates of the Free Energy of Solvation in Acetonea,b
D
Id
∆A−B
∆A−(C + H O)
∆A−(D + H O)
∆(A + H O)−H
2
2
2
gas phase
∆(E + ZPE)
+58.26
+57.59
+57.09
+107.79
+102.27
+52.07
+110.00
+114.05
+66.80
+52.07
+55.41
+68.57
∆Hgas (298.15 K)
∆Ggas (298.15 K)
liquid phase ꢀ ) 20.7
∆∆Gsolv
+7.61
-7.71
+56.99
-14.10
-7.71
+30.26
-11.09
-7.71
+48.00
+16.26
-7.71
+77.12
RTln(1/22.4)
∆Gsol (298.15 K)
a At 298.15 K. All energies are in kJ mol-1. ∆Hgas and ∆Ggas are values from gas-phase calculations with ∆Hgas ) ∆(Htrans + Ηrot + Ηvib + ΗZPE + Eele
)
and ∆Ggas ) ∆Hgas - T∆Sgas.(see ref 46 for more details). ∆Gsol ) ∆Ggas + ∆∆Gsolv + RTln(1/22.4), note that “reduced” solvation energy is used, in which
the difference between the molecular motion contributions calculated in vacuo and in solution is neglected. ∆∆Gsolv is given by the PCM calculation.
b Vibrational entropy does not include any low-frequency modes corresponding to internal rotation.46,54
tions were performed for all minima and transition states in the
gas phase. Each transition state was characterized by a single
imaginary vibrational frequency, whereas reactants and products
had no imaginary frequency. Moreover, frequency calculations
were also performed to provide an estimate of the thermal
correction to the activation enthalpy and entropy within the
standard harmonics oscillator/rigid rotator/ideal gas approxima-
tion.46,47 The calculation of vibrational frequencies, already very
demanding for the isolated complex, is even more tedious within
the PCM approach and has not been achieved. Therefore, for
the thermochemical corrections of the liquid phase, the gas-
phase data were taken as the first approximation, but it is
common practice to neglect the changes in thermal motion and
to use a “reduced” solvation free energy.45c,48 A listing of
Cartesian coordinates and total energies of all the reactants,
intermediates, and transition states can be found in Supporting
Information (see Table S-11-16).
Geometrical Parameters. Previous benchmark calculations
of A have been performed on the full structure with the smaller
3-21G* basis set to enable a comparison of the most relevant
geometrical parameters, in particular those located in the vicinity
of the coordinated sites. Some of the most significant geo-
metrical parameters were compared with the X-ray structure of
(R)-7 (see Figure 2), and the data is given in Table 3. Fair
agreement was found for most bond distances and angles. The
calculated bond and torsion angles are within 3° of the X-ray
structures of (R)-7. The Ru-P and Cp-Ru bond lengths are
within ( 0.05 Å of the experimental value but larger deviations
are observed for the P-O and Ru-L bond lengths that are
slightly overestimated by the DFT calculations (obviously, the
Ru-L bond length cannot be compared with the X-ray structure
(L ) I), but Ru-I is expected to be larger than Ru-O).
Nevertheless the DFT-geometries should give a reasonable
picture of the system under investigation here.
The calculated energy barriers for the Id and D mechanisms
are displayed in Table 4 and shown in Figures 9 and 10.
Gas-Phase Calculations. D Mechanism. In the D pathway
(Figure 6), departure of a water molecule in A first leads to the
chiral 16-electron species C. Inversion of the pyramidal
geometry of C may then occur via a pendulum movement of
the bidentate ligand, through the planar-at-the-metal compound
D. Alternatively, this movement can also be described as a
rocking motion of the Cp ligand with respect to the Ru-P-P
plane.
In the theoretical study of the D mechanism (singlet electronic
pathway), two transition states have been located (see Tables 4
and 5 and Figure 9). The first transition state results from the
dissociation of the water molecule (B) and is characterized by
an imaginary vibrational mode involving almost exclusively the
Ru(II)-OH2 bond (see Table 5). The second transition state
(D) arises from the rocking motion of the Cp ligand as depicted
in Table 5 and has a pseudo-C2-symmetry (æ (Cp-Ru-P-P)
) 179.0°).49 Furthermore, the lengthening of the Ru-P bonds
in D (see Table 3) compared with the pyramidal structures A
and C is consistent with the enhanced back-bonding upon
pyramidalization.15a In the gas-phase, when neglecting the
(46) Ochterksi, J. W. Thermochemistry in Gaussian; Gaussian, Inc.; Pittsburgh,
PA, 2002.
(47) McQuarrie, D. A. Statistical Thermodynamics; Harper and Row: New York,
1973.
(48) Senn, H. M.; Margl, P. M.; Schmid, R.; Ziegler, T.; Blo¨ch P. E. J. Chem.
Phys. 2003, 118, 1089.
9
J. AM. CHEM. SOC. VOL. 126, NO. 15, 2004 4849

A R T I C L E S
Alezra et al.
Figure 9. Free energy scheme of the dissociative mechanism.
Figure 10. Free energy scheme of the Id mechanism.
thermal and entropic contributions, the intermediate C is at a
higher energy than the transition state B (see Table 4).
Stabilization of C is thus strongly linked to the larger transla-
tional entropy resulting from the increase in the number of
particles (one reactant, two products), as is characteristic for
dissociative reactions (see terms T*Strans for A, C, and H2O in
Table S-11, 13, and 16 in Supporting Information).
Comparison to literature data42 of an analogous interconver-
sion for the complex [(η6-C6H6)Ru(C6H4-2-CH2NMe2)(NCMe])]+-
-
PF6 is of interest. The dissociation of the ligand was not
investigated in that case, but approximations were similar to
ones we have made (truncated species, solvent and counterion
neglected). The relative energy between the minima correspond-
ing to (A) and intermediate (C) as well as between their
intermediate (C) and transistion state (D) are, respectively,
within 6.3 and 3.9 kJ mol-1 of our gas-phase values.
(49) Triplet D has also been investigated. The energy required to change the
spin state cannot be directly evaluated since it is required that both states
possess similar geometries and energies for the spin crossing to occur. In
this study, the relative energy of C and D as triplet species was used to
estimate the inversion barrier. This turns out to be considerably larger than
in the singlet electronic pathway (63.1 vs 14.6 kJ mol-1 for Eele).
Id Mechanism. In case of the Id mechanism, a single
transition state (H, Figure 10) has been located. The imaginary
9
4850 J. AM. CHEM. SOC. VOL. 126, NO. 15, 2004

[CpRu((R)-Binop-F)(H O)][SbF ]
A R T I C L E S
2
6
Table 5. Imaginary Mode of the Calculated Transition States
mode involves a concerted motion of the leaving water molecule
with the rocking motion of the Cp ligand (see Table 5). The
transition state was found to deviate from the expected pseudo-
C2-symmetry (see Table 3, æ (Cpcentroid-Ru-P-P) ) -175.0°
and ∆(Ru-O) ) 0.04 Å). The latter, which has also been located
(5 kJ higher in energy), has two coupled imaginary frequencies
and hence is a hyper saddle point having no relevance. As shown
in Tables 3 and 5, the bond lengths of the incoming and leaving
water ligands are 3.81 and 3.85 Å, respectively. The Ru-O
bond of the leaving water molecule is elongated by 1.6 Å
compared to the equilibrium structure, indicative of the dis-
sociative nature of the interchange. When considering the
electronic term only, the energy barrier is lower than that found
for the D mechanism (+52.07 kJ mol-1). However, for the
reasons already mentioned previously, translational entropy
disfavors the transition state H and actually lies 11.48 kJ mol-1
higher in energy than B and 1.77 kJ mol-1 higher than D.
Effect of Solvent. So far, the ∆G(298.15 K) values have been
discussed for the gas-phase (∆Ggas) and not the “liquid”-phase
electronic density. As mentioned before, the effects of the
solvent (acetone) have been investigated here using a continuum
model, and the calculation of Gibbs free energies of reaction in
solution (∆Gsolv) has been estimated according to eq 350-52
Information) indicated only very small changs (<2 kJ mol-1
<0.5 Å, <1°).
,
As shown in Table 4, the largest values of ∆∆Gsolv are
observed between A and C (14.10 kJ mol-1) and A and D (11.09
kJ mol-1). Its origin lies in the solvation energy of the separated
molecules and the complex. The separation of one charged
molecule (A) into another charged molecule (C) and a water
molecule is strongly favored in polar solvent because of the
latter’s dipole moment. As a consequence, when solvent effects
are included, the free energy of both C and D is lowered, while
the energy of B remains fairly unchanged. For similar reasons,
the energy of transition state H increases (+16.26 kJ mol-1).
The effect of an acetone environment thus distinctly favors the
D mechanism over the Id mechanism.
It is clear that because of the approximations made,55 the
computed values cannot be directly correlated with the experi-
mental value ∆Gq298displayed in Table 1 (+50.65 ( 0.40 kJ
mol-1), but it is likely, nevertheless, that a D mechanism is
energetically favored at 298 K.
Concluding Remarks
The combined NMR and theoretical studies support a
dissociative mechanism for the inversion of the ruthenium center
geometry of [Cp((R)-BINOP-F)(H2O)][SbF6] (R)-8 in acetone.
Previous theoretical studies15a on half-sandwich complexes
∆Gsol ) ∆Ggas + ∆∆Gsolv + RTln(1/22.4)
(3)
[CpML1L2L3] gave inversion barriers lower than 15 kcal mol-1
.
where ∆∆Gsolv is the difference in free energy of solvation
between the different structures (∆Gsolv) computed using PCM
techniques.45 Note that a correction term51,53 (RTln(1/22.4) )
-7.71 kJ mol-1) is necessary to account for the change of
reference state from gas phase (ideal gas) to solution. Of course,
continuum models are by nature unable to account explicitly
for specific solute-solvent interactions, but they have been quite
successful in describing the general aspects of solvation.44 In
the calculations reported here, gas-phase geometries were used
for the PCM calculations because structure optimization (carried
out for some structures; see Tables S-11 and S-13 in Supporting
The data on complex (R)-8 do not challenge this finding, but it
must be emphasized that the theoretical estimate of free energy
values should be taken with caution. Provided that the mech-
anism that leads to this “inversion” at the chirotopic metal center
does not involve partial dissociation of the chiral ligand, it is
not expected to lead to an erosion of product enantioselectivity
in catalysis. An identical stereochemical environment exists in
the catalyst-substrate complex at all times. Conversely, in
analogous enantiomerically pure complexes devoid of ligand-
based chirality, this inversion leads to racemization at the metal.
While the prospect of using “chiral at the metal” complexes as
catalysts is a very attractive proposition, this may only be
(50) Aquino, A. J. A.; Tunega, D.; Haberhauer, G.; Gerzabek, M. H.; Lischka,
H. J. Phys. Chem. A 2002, 106, 1862.
(51) Tunega, D.; Haberhauer, G.; Gerzabek, M.; Lischka, H. J. Phys. Chem. A
2000, 104, 6824.
(54) Ayala, P. Y.; Schlegel, H. B. J. Chem. Phys. 1998, 108, 2315.
(55) Structure truncation apart, entropy contributions are based on the assumption
that the contributions to the binding in the gas phase and in solution is the
same. Stranks, D. R. Pure Appl. Chem. 1974, 38, 303.
(52) Colominas, C.; Teixido, J.; Cemeli, J.; Luque, F. J.; Orozco, M. J. Phys.
Chem. B 1998, 102, 2269.
(53) Cieplak, P.; Kollman, P. A. J. Am. Chem. Soc. 1988, 110, 3734.
9
J. AM. CHEM. SOC. VOL. 126, NO. 15, 2004 4851

A R T I C L E S
Alezra et al.
) 2.3 Hz), 133.4, 133.2, 131.3, 131.0, 130.7, 130.6, 127.9, 127.6, 127.5,
127.4, 126.2, 126.0, 125.3, 124.9, 124.3 (d, J(C,P) ) 3.4 Hz), 123.4-
123.2 (m), 123.0-122.9 (m), 86.5. 31P NMR (CD2Cl2, 202 MHz, rt):
δ 115.6 (qAB, ∆νAB ) 2404 Hz, JAB ) 50 Hz).
realizable with complexes that link the auxiliary arene or Cp
ligand with the other ligands and hence suppress racemization.6f
Experimental Section
[CpRu((R)-BINOP-F)I] (R)-6. A yellow suspension of (R)-5 (1.48
g, 1.25 mmol) and iodoform (4.92 g, 12.50 mmol) in acetone (12.50
mL) was refluxed for 28 h. The solution was cooled to room
temperature, and the solvent was evaporated; the orange residue was
purified by chromatography (silica) with a mixture of pentane and
CH2Cl2 (10:1 to 1:1). The orange band was collected, and the volatiles
were removed in vacuo. The orange residue was then recrystallized
from CH2Cl2/MeOH to give the iodo complex (R)-6 (1.19 g, 72%) as
General Methods. All synthetic manipulations were carried out
using standard Schlenk techniques under an inert atmosphere. All
solvents were dried and distilled before use according to standard
1
laboratory procedures. H, 31P, and 13C NMR spectra were recorded
on Bruker AMX400 and Bruker Avance 500 spectrometers. Iodoform
and AgSbF6 were purchased from Fluka. Ru3(CO)12 was prepared
according to the literature.20 The Diels-Alder reaction of methacrolein
with cyclopentadiene catalyzed by (R)-7 and (R)-8 was realized
following previously described procedures.2
1
orange crystals. H NMR (CD2Cl2, 500 MHz, rt): δ 8.09 (d, J ) 9.1
Hz, 1H), 7.85 (d, J ) 9.1 Hz, 1H), 7.81 (d, J ) 9.1 Hz, 1H), 7.70 (d,
J ) 7.9 Hz, 1H), 7.67 (d, J ) 8.2 Hz, 1H), 7.45 (d, J ) 9.1 Hz, 1H),
7.37 (ddd, J ) 7.9, 6.6, 1.3 Hz, 1H), 7.34 (ddd, J ) 8.2, 7.0, 1.3 Hz,
1H; C10H12), 7.15 (ddd, J ) 8.2, 6.6, 1.3 Hz, 1H), 7.08 (ddd, J ) 8.2,
7.0, 1.3 Hz, 1H), 6.63 (d, J ) 8.5 Hz, 1H), 6.45 (d, J ) 8.5 Hz, 1H),
4.72 (s, 5H). 13C NMR (CD2Cl2, 125 MHz, rt): δ 150.9 (d, J(C,P) )
4.0 Hz), 149.3 (d, J(C,P) ) 4.0 Hz), 133.7, 133.4, 131.3, 131.2, 130.7,
130.4, 128.0, 127.9, 127.7, 127.2, 126.5, 126.1, 125.4, 125.2, 125.0,
122.6 (d, J(C,P) ) 3.4 Hz), 122.5 (d, J(C,P) ) 2.3 Hz), 122.1, 88.1.
31P NMR (CD2Cl2, 202 MHz, rt): δ 113.8 (qAB, ∆νAB ) 1258 Hz, JAB
) 66 Hz).
All 31P and 17O NMR spectra for the determination of the kinetic
parameters were recorded on a Bruker ARX400 spectrometer operating
at 162 and 54 MHz, respectively. The temperature was determined by
substituting the sample with a Pt-100 resistor.56 The variable-pressure
measurements to determine the activation volumes were performed
between 0.1 and 200 MPa using a homemade high-pressure NMR
probe.57 The observed rate constants were obtained by NMR line-shape
analysis using the program NMRICMA.58 The analyses of the rate
constants using the required equations were accomplished with a
nonlinear least-squares fitting program. All experimental data are given
in Supporting Information. The reported errors represent standard
deviations.
[CpRu((R)-BINOP-F)(Acetone)][SbF6] (R)-7.
A solution of
AgSbF6 (952 mg, 2.71 mmol) in dried CH2Cl2 (27 mL) was added at
room temperature to (R)-6 (784 mg, 0.60 mmol) in a mixture of dried
acetone (10 mL) and CH2Cl2 (13 mL). The mixture was stirred at room
temperature for 30 min. After evaporation of the solvent under vacuum,
the residue was dissolved in a 1:1 mixture of acetone and CH2Cl2 and
filtered through Celite. After evaporation of volatiles under vacuum,
the product was purified by dissolution in acetone and precipitation by
addition of Et2O to give compound (R)-7 as an orange solid (847 mg,
95%). 1H NMR (acetone-d6, 400 MHz, 40 °C): δ 8.13 (d, J ) 9.1 Hz,
2H), 7.87 (d, J ) 8.1 Hz, 2H), 7.73 (d, J ) 9.1 Hz, 2H), 7.47 (dd, J1
) 8.1 Hz, J2 ) 7.3 Hz, 2H), 7.26 (dd, J1 ) 8.1 Hz, J2 ) 7.3 Hz, 2H),
6.58 (d, J ) 8.1 Hz, 2H), 5.07 (s, 5H). 1H NMR (acetone-d6, 400 MHz,
-40 °C): δ 8.18 (d, J ) 9.6 Hz, 2H), 8.00-7.94 (m, 2H), 7.77-7.70
(m, 2H), 7.54-7.46 (m, 2H), 7.35-7.23 (m, 2H), 6.57 (d, J ) 8.6 Hz,
1H), 6.42 (d, J ) 8.3 Hz, 1H), 5.14 (s, 5H). 13C NMR (acetone-d6,
100 MHz, 40 °C): δ 150.8, 133.8, 133.0, 131.9, 128.8, 128.8, 127.4,
125.5, 123.0, 122.2, 87.2. 31P NMR (acetone-d6, 162 MHz, 40 °C): δ
123.8. 31P NMR (acetone-d6, 162 MHz, -40 °C): δ 124.3 (qAB, ∆νAB
) 2931 Hz, JAB ) 64 Hz).
[CpRu((R)-BINOP-F)(H2O)][SbF6] (R)-8. H2O (9 µL, 0.5 mmol)
was added to an orange solution of (R)-7 (125 mg, 0.085 mmol) in
CH2Cl2 (1 mL) at room temperature. The solution was stirred for 5
min, and the volatiles were removed in vacuo. The solid residue was
dissolved in CH2Cl2 (1 mL); water (9 µL, 0.5 mmol) was added, and
the solution was stirred for 5 min before the volatiles were evaporated
under vacuum. Another cycle of dissolution-addition of water-
stirring-evaporation gave the aquo complex (R)-8 as a yellow solid
(120 mg, 98%). 1H NMR (acetone-d6, 400 MHz, 25 °C): δ 8.10 (d, J
) 9.1 Hz, 2H), 7.86 (d, J ) 8.1 Hz, 2H), 7.73 (d, J ) 9.1 Hz, 2H),
7.47 (dd, J1 ) 8.1 Hz, J2 ) 7.2 Hz, 2H), 7.26 (dd, J1 ) 8.1 Hz, J2 )
7.2 Hz, 2H), 6.60 (d, J ) 8.1 Hz, 2H), 5.04 (s, 5H). 1H NMR (acetone-
d6, 400 MHz, -40 °C): δ 8.16 (d, J ) 9.1 Hz, 2H), 7.94-7.90 (m,
2H), 7.73-7.69 (m, 2H), 7.51-7.45 (m, 2H), 7.30-7.22 (m, 2H), 6.61
(d, J ) 8.6 Hz, 1H), 6.45 (d, J ) 8.6 Hz, 1H), 5.71 (bs, 2H), 5.06 (s,
5H). 13C NMR (acetone-d6, 100 MHz, 40 °C): δ 149.8, 132.9, 131.8,
130.8, 127.7, 126.4, 124.5, 121.8, 121.6, 86.0. 13C NMR (100 MHz,
acetone-d6, -40 °C): δ 149.8, 148.1, 132.0, 131.4, 131.0, 130.8, 130.2,
129.6, 127.3, 127.0, 126.9, 125.7, 125.5, 123.7, 123.4, 121.6, 121.3,
119.7, 86.5. 31P NMR (acetone-d6, 162 MHz, 25 °C): δ 123.3. 31P
NMR (acetone-d6, 162 MHz, -40 °C): δ 122.9 (qAB, ∆νAB ) 2439
Hz, JAB ) 61 Hz).
(R)-BINOP-F) (R)-4. A solution of (C6F5)2PBr (10.01 g, 22.50
mmol) in anhydrous Et2O (6 mL) at room temperature was added
dropwise to a solution of (R)-BINOL (3.22 g, 11.25 mmol) and Et3N
(3.14 mL, 22.53 mmol) in anhydrous Et2O (100 mL) at 10 °C under
an N2 atmosphere. The resulting white suspension was stirred at 10 °C
for 2 h and then at room temperature for 16 h. The mixture was filtered
over Celite under N2, and the Celite was rinsed with Et2O. The solvent
was then evaporated, and the white residue was further purified by
recrystallization from hexane (60 mL) to give (R)-BINOP-F as a white
1
solid (8.99 g, 79%). H NMR (500 MHz, CD2Cl2, rt, SiMe4): δ )
7.92 (d, J ) 9.2 Hz, 2H), 7.81 (d, J ) 8.2 Hz, 2H), 7.48 (d, J ) 8.8
Hz, 2H), 7.41 (ddd, J ) 8.2, 6.9, 1.3 Hz, 2H), 7.25 (ddd, J ) 8.2, 7.0,
1.3 Hz, 2H), 7.00 (d, J ) 8.5 Hz, 2H). 13C NMR (125 MHz, CD2Cl2,
rt, SiMe4): δ ) 152.7 (d, J(C,P) ) 16.1 Hz), 133.6, 131.0, 128.1, 127.5,
125.8, 125.7, 122.5 (d, J(C,P) ) 9.2 Hz), 119.9 (d, J(C,P) ) 9.2 Hz).
19F NMR (470 MHz, CD2Cl2, rt, C6F6): δ ) 31.17-30.95 (m, 4F),
30.59-30.38 (m, 4F), 14.15 (tm, J ) 19.8 Hz, 2F), 13.20 (tm, J )
19.9 Hz, 2F), 2.81-2.63 (m, 4F), 2.34-2.17 (m, 4F). 31P NMR (202
MHz, CD2Cl2, rt, H3PO4): δ ) 89.0 (qm, J(P,F) ) 32 Hz). [R]D (20
°C, c 0.95, hexane): +88.7
[CpRu((R)-BINOP-F)H] (R)-5. Cylopentadiene (4.40 mL, 53.60
mmol) was added to an orange solution of Ru3(CO)12 (426 mg, 0.67
mmol) in refluxing anhydrous heptane (120 mL), and the mixture was
heated at reflux for 2 h under an N2 atmosphere. The resulting yellow
solution was then cooled to 80 °C; (R)-BINOP-F (R)-4 was added,
and the mixture was refluxed for 17 h. The solution was cooled to
room temperature, and the solvent was evaporated; the yellow residue
was purified by chromatography (aluminum oxide, neutral) with hexane
and then toluene. The yellow band was collected, and the volatiles were
removed in vacuo, yielding the hydrido complex (R)-5 (2.06 g, 81%)
as a yellow solid. 1H NMR (CD2Cl2, 500 MHz, rt): δ 7.87 (d, J ) 9.1
Hz, 1H), 7.79 (d, J ) 9.1 Hz, 1H), 7.77 (d, J ) 9.1 Hz, 1H), 7.67-
7.61 (m, 3H), 7.38-7.31 (m, 2H), 7.20-7.16 (m, 1H), 7.15-7.10 (m,
1H), 6.81 (d, J ) 8.5 Hz, 1H), 6.63 (d, J ) 8.5 Hz, 1H), 4.39 (s, 5H),
-11.82 (ddd, J(H,P) ) 41.2, 38.2 Hz, J(H,F) ) 3 Hz, 1H). 13C NMR
(CD2Cl2, 125 MHz, rt): δ 152.2 (d, J(C,P) ) 2.3 Hz), 150.0 (d, J(C,P)
(56) Amann, C.; Meyer, P.; Merbach, A. E. J. Magn. Reson. 1982, 46, 319.
(57) Cusanelli, A.; Nicula-Dadci, L.; Frey, U.; Merbach, A. E. Inorg. Chem.
1997, 36, 2211.
(58) Helm, L.; Borel, A.; NMRICMA; Lausanne University: Lausanne, Swit-
zerland, 1998.
9
4852 J. AM. CHEM. SOC. VOL. 126, NO. 15, 2004

[CpRu((R)-Binop-F)(H O)][SbF ]
A R T I C L E S
2
6
X-ray Crystallographic Analysis. Cell dimensions and intensities
were measured at 200 K on a Stoe STADI4 diffractometer with
graphite-monochromated Cu KR radiation (λ ) 1.5418 Å) for [CpRu-
((R)-BINOP-F)I] (R)-6 and on a Stoe IPDS diffractometer with graphite-
monochromated Mo KR radiation (λ ) 0.71073 Å) for [CpRu((R)-
BINOP-F)(acetone)][SbF6] (R)-7. Data were corrected for Lorentz and
polarization effects and for absorption. The structures were solved by
direct methods (SIR97),59 and all other calculations were performed
with XTAL system60 and ORTEP61 programs. [CpRu((R)-BINOP-F)I]
(acetone)][SbF6]. [Et2O] (R)-7: C56H33F26O4P2RuSb; M ) 1548.6, dx
) 1.643 g cm-3, orthorhombic, P212121, Z ) 4, a ) 13.4024(7), b )
19.3137(10), c ) 24.1839(15) Å, U ) 6260.0(7) Å3; 78 798 measured
reflections, 12 186 unique reflections of which 7103 were observables
(|Fo| > 4σ (Fo)). Full-matrix least-squares refinement based on F gave
final values R ) 0.042, ωR ) 0.047, and Flack parameter x ) -0.01(3).
Hydrogen atoms of the complex were placed in calculated positions.
Both Et2O solvent molecules are disordered and refined with population
parameters of 0.5 and restraints on bond lengths and bond angles.
(R)-6: C50H19Cl2F20IO2P2Ru, M ) 1392.5, dx ) 1.901 g cm-3
,
Acknowledgment. Financial support for this work by the
Swiss National Science Foundation is gratefully acknowledged.
We thank the Ministe`re des Affaires Etrange`res of France for a
Lavoisier grant to V.A.
orthorhombic, P212121, Z ) 4, a ) 13.038(1), b ) 16.812(1), c )
22.199(2) Å, U ) 4865.9(2) Å3; 6880 measured reflections, 5979 unique
reflections of wich 5272 were observables (|Fo| > 4σ (Fo)). Full-matrix
least-squares refinement based on F gave final values R ) 0.043, ωR
) 0.042, and Flack parameter x ) 0.00(1). Hydrogen atoms of the
complex were placed in calculated positions. [CpRu((R)-BINOP-F)-
Supporting Information Available: Listing of experimental
data obtained from variable-temperature, -pressure, and -con-
centration 17O and 31P NMR experiments and X-ray crystal-
lographic files (CIF) for (R)-6 and (R)-7. This material is
available free of charge via the Internet at http://pubs.acs.org.
(59) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G.; Giacovazzo, C.;
Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna, R. J. Appl. Cryst.
1999, 32, 115.
(60) Hall, S. R., Flack, H. D., Stewart, J. M., Eds. XTAL3.2 User’s Manual;
Universities of Western Australia and Maryland, 1992.
(61) Johnson, C. K. ORTEP II; Oakridge National Laboratory: Oak Ridge, TN,
1976; Report ORNL-5138.
JA0374123
9
J. AM. CHEM. SOC. VOL. 126, NO. 15, 2004 4853