Effect of chelate ring size on the rate of hydride transfer from CpRu(P-P)H (P-P = chelating diphosphine) to an iminium cation

Article abstract of DOI:10.1021/om0303404

The rate constants for hydride transfer from CpRu(P-P)H (P-P = bis(diphenylphosphino)-methane (dppm), bis(diphenylphosphino)ethane (dppe), bis(diphenylphosphino)benzene (dpbz), or bis(diphenylphosphino)propane (dppp)) to 1-(1-phenylethylidene)pyrrolidiniu

Full text of DOI:10.1021/om0303404

4084
Organometallics 2003, 22, 4084-4089
Effect of Ch ela te Rin g Size on th e Ra te of Hyd r id e
Tr a n sfer fr om Cp Ru (P -P )H (P -P ) Ch ela tin g
Dip h osp h in e) to a n Im in iu m Ca tion
Hairong Guan, Masanori Iimura, Matthew P. Magee, J ack R. Norton,* and
Kevin E. J anak
Department of Chemistry, Columbia University, New York, New York 10027
Received May 8, 2003
The rate constants for hydride transfer from CpRu(P-P)H (P-P ) bis(diphenylphosphino)-
methane (dppm), bis(diphenylphosphino)ethane (dppe), bis(diphenylphosphino)benzene
(dpbz), or bis(diphenylphosphino)propane (dppp)) to 1-(1-phenylethylidene)pyrrolidinium
tetrafluoroborate have been measured. The bite angles of the hydride complexes CpRu-
(dppm)H, CpRu(dppe)H, CpRu(dpbz)H, and CpRu(dppb)H (dppb ) bis(diphenylphosphino)-
butane) have been determined by X-ray diffraction. Hydride transfer is faster when the
chelate ring of the diphosphine is smaller (i.e., CpRu(dppm)H > CpRu(dppe)H ≈ CpRu-
(dpbz)H > CpRu(dppp)H . CpRu(dppb)H). Boiling CpRu(PPh3)2Cl with dpbz in benzene or
toluene results in the formation of [CpRu(PPh3)(η2-dpbz)]Cl as well as CpRu(dpbz)Cl.
Sch em e 1
In tr od u ction
We have previously reported1 the ruthenium-cata-
lyzed hydrogenation of iminium cations (including pro-
tonated imines) by an ionic mechanism, involving
heterolytic cleavage of H2 and sequential transfer of H-
and H+ to the substrates. When chiral diphosphines are
employed, CdN bonds can be hydrogenated enantiose-
lectively with ee’s up to 60%. As the literature contains
few catalysts for the asymmetric hydrogenation of CdN
bonds,2,3 we have studied the mechanism of ours; the
transfer of H- from CpRu(P-P)H to the iminium cation
(Scheme 1) has proven to be turnover-limiting and
enantioselectivity-determining.1
It is possible to stop the reaction after a single H-
transfer if it is carried out in the absence of H2 and the
presence of a coordinating solvent (usually CH3CN) (eq
1). To screen phosphines for catalytic use, we have
measured the rate at which their CpRu(P-P)H com-
plexes undergo this “stoichiometric” hydride transfer.
When such an experiment (eq 2) showed CpRu((S,S)-
DIOP)H to be inactive at hydride transfer to 1-(1-
phenylethylidene)pyrrolidinium tetrafluoroborate (1),1
we suspected an effect of the size of the chelate ring.
(The CpRu hydrides of other chelating diphosphines
with one alkyl and two aryl substituents on each
phosphorus transfer H- to 1.1)
co-workers.4 The Brookhart group has shown that the
kinetic barriers for the migratory insertion of complexes
[Ph2P(CH2)nPPh2]Pd(CH3)(CO)+ (n ) 2-4) decrease
with an increase in the P-Pd-P bond angle.5 Vilar et
al. have recently reported a similar influence of chelate
ring size on the insertion of isocyanides into the Pd-C
bond of (P-P)Pd(CH3)Cl complexes.6
The effect of chelate ring size on metal-catalyzed C-C
bond forming reactions, such as hydroformylation, al-
lylic substitution, hydrocyanation, and the amination
of aryl halides, has been reviewed by van Leeuwen and
(1) Magee, M. P.; Norton, J . R. J . Am. Chem. Soc. 2001, 123, 1778-
1779.
(2) (a) Willoughby, C. A.; Buchwald, S. L. J . Am. Chem. Soc. 1994,
116, 8952-8965. (b) Broger, E. A.; Burkart, W.; Hennig, M.; Scalone,
M.; Schmid, R. Tedrahedron: Asymmetry 1998, 9, 4043-4054. (c) Xiao,
D.; Zhang, X. Angew. Chem., Int. Ed. 2001, 40, 3425-3428.
(3) Review on hydrogenation of imines: Blaser, H.-U.; Spindler, F.
Hydrogenation of Imine Groups. In Comprehensive Asymmetric Ca-
talysis; J acobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer-
Verlag: New York, 1999; Vol. 1, pp 247-265.
(4) (a) van Leeuwen, P. W. N. M.; Kamer, P. C. J .; Reek, J . N. H.;
Dierkes, P. Chem. Rev. 2000, 100, 2741-2769. (b) Kamer, P. C. J .;
van Leeuwen, P. W. N. M.; Reek, J . N. H. Acc. Chem. Res. 2001, 34,
895-904.
(5) Ledford, J .; Shultz, C. S.; Gates, D. P.; White, P. S.; DeSimone,
J . M.; Brookhart, M. Organometallics 2001, 20, 5266-5276.
10.1021/om0303404 CCC: $25.00 © 2003 American Chemical Society
Publication on Web 08/23/2003

Effect of Chelate Ring Size
Organometallics, Vol. 22, No. 20, 2003 4085
However, little is known about the effect of chelate
ring size on hydride transfer, although much work has
been done on other factors influencing the rate of such
transfers from transition-metal hydrides.7 DuBois and
co-workers have determined the equilibrium constants
for several H- transfer reactions of [HM(diphosphine)2]-
PF6 (M ) Pt, Ni) complexes and have found that a
smaller chelate ring favors hydride transfer thermody-
namically.8 Knowledge of the kinetic effects of chelate
ring size is still lacking.
In this paper we report the kinetics of hydride
transfer from CpRu(P-P)H, with bidentate phosphines
of different chelate ring sizes, to the iminium cation 1.
We also report the structures of these complexes and
their influence on the rates of hydride transfer.
Resu lts a n d Discu ssion
F igu r e 1. Structure of [CpRu(PPh3)(η2-dpbz)]Cl (4) (20%
probability level). Selected bond lengths (Å) and angles
(deg): Ru-P1 2.3478(6), Ru-P2 2.2787(7), Ru-P3 2.3136(7),
P1-Ru-P2 97.26(2), P1-Ru-P3 97.58(2), P3-Ru-P2
82.87(2).
P r ep a r a tion of Cp Ru (P -P )H. The literature con-
tains two general methods of preparing CpRu(P-P)H:
(1) the synthesis of CpRu(P-P)Cl and its reduction by
CH3ONa,9 and (2) the displacement of carbonyl ligands
from CpRu(CO)2H by diphosphines.10 Method 1 is more
convenient because the chlorides CpRu(P-P)Cl (3) are
air stable; however, method 2 gives better results when
the distance between the P atoms is large and bridging
is a possibility. We thus prepared CpRu(dppm)H (2a ),
CpRu(dppe)H (2b), and CpRu(dpbz)H (2c) by method
1; CpRu(dppp)H (2d ) and CpRu(dppb)H (2e) by method
2.
12
and [CpRu(PPh3)(η2-dppe)]BF4 (prepared by treating
CpRu(dppm)Cl or CpRu(dppe)Cl with PPh3 and NH4-
BF4 in CH3OH).
Most 3 are readily prepared by displacing PPh3 from
CpRu(PPh3)2Cl with the appropriate diphosphine. How-
ever, the preparation of CpRu(dpbz)Cl (3c) by this
method in boiling benzene (eq 3) gives an interesting
byproduct: a light green precipitate, 4. (The precipitate
4 is not an intermediate in the formation of 3c: 4
remains after 3 days in boiling benzene, or when the
A dimeric or oligommeric structure for 4 in solution
cannot be ruled out, although we are not aware of any
case where dpbz has served as a bridging ligand. We
have been able to grow single crystals of 4, by slow
diffusion of hexanes into a concentrated CH2Cl2 solution.
X-ray diffraction shows (Figure 1)13 a monomeric struc-
ture in the solid state.
1
higher-boiling toluene is used as solvent.) The H NMR
of 4 shows, in addition to the Cp singlet (δ 4.71), 39
protons in the aromatic region; the 31P{1H} NMR
reveals a triplet at δ 43.39 and a doublet at δ 72.05,
with J ) 34.0 Hz. Thus 4 is [CpRu(PPh3)(η2-dpbz)]Cl,
analogous to the known [CpRu(PPh3)(η2-dppm)]BF4
11,12
The ruthenium in 4 sits in a very crowded environ-
ment, with both the PPh3 and the dpbz coordinated. The
dissociation of the Cl- in eq 3 parallels its dissociation
in polar solvents, such as CH3OH, C2H5OH, and DMSO.
Treichel and co-workers have measured the rates of
halide ion solvolysis of CpRuL2Cl (L ) PMe3, PPhMe2,
PPh2Me, PPh2(OMe), P(OEt)3, PMe(OMe)2; L2 ) dppe)
in DMSO and CH3CN.14 (They found that a less
electron-donating phosphine retards Cl- dissociation,
while silver, ammonium, or sodium salts facilitate it.)
However, in nonpolar solvents such as benzene, toluene,
or decalin, phosphine dissociation from CpRu(P-P)Cl
(6) Owen, G. R.; Vilar, R.; White, A. J . P.; Williams, D. J . Organo-
metallics 2002, 21, 4799-4807.
(7) (a) Labinger, J . A. Nucleophilic Reactivity of Transition Metal
Hydrides. In Transition Metal Hydrides; Dedieu, A., Ed.; VCH: New
York, 1992; pp 361-379. (b) Labinger, J . A.; Komadina, K. H. J .
Organomet. Chem. 1978, 155, C25-C28. (c) Kinney, R. J .; J ones, W.
D.; Bergman, R. G. J . Am. Chem. 1978, 100, 7902-7915. (d) Kao, S.
C.; Gaus, P. L.; Youngdahl, K.; Darensbourg, M. Y. Organometallics
1984, 3, 1601-1603. (e) Kao, S. C.; Spillett, C. T.; Ash, C.; Lusk, R.;
Park, Y. K.; Darensbourg, M. Y. Organometallics 1985, 4, 83-91. (f)
Gaus, P. L.; Kao, S. C.; Youngdahl, K.; Darensbourg, M. Y. J . Am.
Chem. Soc. 1985, 107, 2428-2434. (g) Martin, B. D.; Warner, K. E.;
Norton, J . R. J . Am. Chem. Soc. 1986, 108, 33-39. (h) Smith, K.-T.;
Norton, J . R.; Tilset, M. Organometallics 1996, 15, 4515-4520. (i)
Hembre, R. T.; McQueen, S. J . Am. Chem. Soc. 1994, 116, 2141-2142.
(j) Hembre, R. T.; McQueen, J . S. Angew. Chem., Int. Ed. Engl. 1997,
36, 65-67. (k) Cheng, T.-Y.; Bullock, R. M. Organometallics 1995, 14,
4031-4033. (l) Cheng, T.-Y.; Brunschwig, B. S.; Bullock, R. M. J . Am.
Chem. Soc. 1998, 120, 13121-13137. (m) Cheng, T.-Y.; Bullock, R. M.
Organometallics 2002, 21, 2325-2331. (n) Sarker, N.; Bruno, J . W. J .
Am. Chem. Soc. 1999, 121, 2174-2180.
(12) Ashby, G. S.; Bruce, M. I.; Tomkins, I. B.; Wallis, R. C. Aust. J .
Chem. 1979, 32, 1003-1016.
(13) The solvent molecule is severely disordered, and a suitable
model cannot be constructed for the solvate. The program SQUEEZE
(see van der Sluis, P.; Spek, A. L. Acta Crystallogr. 1990, A46, 194-
201) was used to refine the solvent molecule as a diffuse contribution
to overall scattering without specific atom positions; however, the
empirical formula, density, and absorption coefficient reflect the full
formula.
(8) Berning, D. E.; Noll, B. C.; DuBois, D. L. J . Am. Chem. Soc. 1999,
121, 11432-11447.
(9) Bruce, M. I.; Humphrey, M. G.; Swincer, A. G.; Wallis, R. C. Aust.
J . Chem. 1984, 37, 1747-1755.
(10) White, C.; Cesarotti, E. J . Organomet. Chem. 1985, 287, 123-
125.
(14) (a) Treichel, P. M.; Komar, D. A.; Vincenti, P. J . Inorg. Chim.
Acta 1984, 88, 151-152. (b) Treichel, P. M.; Vincenti, P. J . Inorg. Chem.
1985, 24, 228-230.
(11) Bruce, M. I.; Humphrey, M. G.; Patrick, J . M.; White, A. H.
Aust. J . Chem. 1983, 36, 2065-2072.

4086 Organometallics, Vol. 22, No. 20, 2003
Guan et al.
is generally more facile,15 so the Cl- dissociation in eq
3 is surprising.
Most of the hydrides 2 proved to be stable in haloge-
nated solvents, but 2d , regardless of the batch or the
synthetic methods used to prepare it, gave CpRu(dppp)-
Cl within minutes in CD2Cl2.16,17 This type of halogena-
tion is known to be a radical process18 and potentially
retarded by radical inhibitors, so we examined the
stability of 2d in the presence of several such inhibitors.
After 1 day with a suspension of hydroquinone (only
slightly soluble) in CD2Cl2, about 85% of the 2d re-
mained, but 4-tert-butylcatechol, which has better solu-
bility in halogenated solvents, reacted immediately with
2d (to give complex products that were not identified).
However, TEMPO (2,2,6,6-tetramethyl-1-piperidinyloxy)
proved unreactive toward 2d and efficient at preventing
its reaction with CD2Cl2: no significant decomposition
of 2d in CD2Cl2 solution was observed after 30 h in the
presence of 1% (by weight) TEMPO.
F igu r e 2. Plot of kobs vs [1] for hydride transfer from 2b
to the iminium cation 1 at 300 K with MeCN present (b).
(The value of kobs without MeCN present is also displayed
in this figure (0).)
Ta ble 1. Ra tes of Hyd r id e Tr a n sfer fr om
Cp Ru (P -P )H to th e Im in iu m Ca tion 1 a t
300 K in CD2Cl2
The fact that TEMPO does not itself abstract H• from
2d warrants comment. TEMPO abstracts H• from group
14 organometallic hydrides (e.g., Bu3SnH) at 80 °C,19
forming the hydroxylamine 5 (eq 4). As the O-H bond
dissociation energy (BDE) of 5 in benzene is 69 kcal/
mol20 and the Ru-H BDE value for 2d can be estimated
at 65 cal/mol,21 H• abstraction from 2d by TEMPO
should be thermodynamically favorable. The reason it
does not occur at an appreciable rate must be kinetic:
the cyclopentadienyl ring and dppp ligand in 2d must
make the hydride ligand inaccessible to a bulky radical
like TEMPO.
chelating
ligand
bite angle P-Ru-P
rate constant k
(M-1 s-1
complex
(deg)
)
2a
2b
2c
2d
2e
dppm
dppe
dpbz
dppp
dppb
72.01(3)
84.497(16)
84.22(4)
91.9(1)b
6.4(3) × 10-1a
2.0(1) × 10-2
2.1(2) × 10-2
2.8(2) × 10-3
no reaction
93.96(3)
(<4 × 10-6
)
a
Average of two kinetic experiments with different concentra-
b
tions of 1 and 2a . See ref 31.
is second-order overall (eq 6). The rate constant k for
2b was calculated from the slope of Figure 2 and
inserted in Table 1.
Kin etics of Hyd r id e Tr a n sfer fr om th e Hyd r id e
Com p lexes 2 to th e Im in iu m Ca tion 1. When H-
transfer to 1 from any of the hydride complexes 2a -d
is carried out in the presence of CH3CN, the cation 6
results (eq 5). Monitoring the disappearance of the Cp
1
signal of the hydride 2b by H NMR in the presence of
excess 1 (10 equiv) gave a pseudo-first-order rate
constant kobs. Variation of [1] from 0.10 to 0.26 M (Table
S-1, Figure 2) showed that kobs for 2b was linear with
[1], implying that the hydride transfer reaction in eq 5
Without CH3CN kobs does not change appreciably
(Figure 2), although the amine 7 becomes coordinated
instead of the CH3CN. The rate-determining step in
(15) Davies, S. G.; McNally, J . P.; Smallridge, A. J . Adv. Organomet.
Chem. 1990, 30, 1-76, and references therein.
(16) The fact that the dppp hydride 2d is more reactive toward CD2-
Cl2 than 2a and 2b is probably due to the more negative potential of
2d ,17 which enables it to initiate a radical chain more easily by electron
transfer to the solvent.
(17) J ia, G.; Morris, R. H. J . Am. Chem. Soc. 1991, 113, 875-883.
(18) (a) Shackleton, T. A.; Mackie, S. C.; Fergusson, S. B.; J ohnson,
L. J .; Baird, M. C. Organometallics 1990, 9, 2248-2253. (b) Brown, T.
L. In Organometallic Radical Processes; Trogler, W. C., Ed.; J ournal
of Organometallic Chemistry Library 22; Elsevier: New York, 1990;
pp 67-107, and references therein.
22
reaction 5 is clearly H- transfer, with CH3CN serving
as an efficient trap. Sixteen-electron half-sandwich
ruthenium diphosphine complexes, with and without
agostic interactions, have been isolated and structurally
characterized. They are particularly stable when a Cp*
ring and bulky phosphine ligands such as (iPr)2PCH2-
CH2P(iPr)2 or two PMe(iPr)2 are both present,23 but can
also be made when there is a less bulky diphosphine.24
(Sixteen-electron complexes with less bulky diphos-
phines are more reactive.)
(19) Lucarini, M.; Marchesi, E.; Pedulli, G. F.; Chatgilialoglu, C. J .
Org. Chem. 1998, 63, 1687-1693.
(20) Mahoney, L. R.; Mendenhall, G. D.; Ingold, K. U. J . Am. Chem.
Soc. 1973, 95, 8610-8614.
(21) The BDE of CpRu(CO)2H has been reported as 65 kcal/mol, and
phosphine substitution usually has a negligible effect on M-H ener-
gies. See: (a) Tilset, M.; Parker, V. D. J . Am. Chem. Soc. 1989, 111,
6711-6717; correction, J . Am. Chem. Soc. 1990, 112, 2843. (b)
Kristja´nsdo´ttir, S. S.; Norton, J . R. Acidity of Hydrido Transition Metal
Complexes in Solution. In Transition Metal Hydrides; Dedieu, A., Ed.;
VCH: New York, 1992; pp 309-359.
The kobs value for 2c was obtained by monitoring the
appearance of the coordinated CH3CN signal of 6c,
(22) Iimura, M.; Magee, M. P.; Norton, J . R. Unpublished work.
(23) J ime´nez-Tenorio, M.; Mereiter, K.; Puerta, M. C.; Valerga, P.
J . Am. Chem. Soc. 2000, 122, 11230-11231.

Effect of Chelate Ring Size
Organometallics, Vol. 22, No. 20, 2003 4087
F igu r e 3. Plot of ln{([RuH]0/[RuH]t) + 1} vs time t for
hydride transfer from 2a to the iminium cation 1 at 300 K
in CD2Cl2 with MeCN present. Initially [2a ] ) 7.5 × 10-3
M, [1] ) 1.5 × 10-2 M.
F igu r e 4. Molecular structure of CpRu(dppm)H (2a ) (20%
probability level). Selected bond lengths (Å) and angle
(deg): Ru-P1 2.2435(9), Ru-P2 2.2399(9), P1-Ru-P2
72.01(3).
because the Cp of 6c has a 1H NMR chemical shift
(δ 4.635) very close to that of the Cp of 2c (δ 4.639).
Variation of [1] from 0.10 to 0.25 M (Table S-2, Figure
S-2) also showed a linear relationship between kobs and
[1], confirming that the hydride transfer is a second-
order reaction. The rate constant k for 2c has been
calculated from the slope of Figure S-2 and is shown in
Table 1.
thermore, the y-intercept (0.744 ( 0.072) is consistent
with ln 2 in eq 8, again confirming that H- transfer from
2 to 1 is a second-order reaction.27 The rate constant k
for 2a has been derived from the slope (k[2a ]0) and is
shown in Table 1.
Str u ctu r es of Hyd r id e Com p lexes Cp Ru (P -P )H
(2). The literature contains surprisingly little crystal-
lographic information about half-sandwich ruthenium
phosphine hydrides. Structures had been reported for
the bisphosphine complexes CpRu(PPh3)2H28 and CpRu-
(PMe3)2H29 and for the chelating diphosphine complexes
CpRu(dppf)H (dppf ) bis(diphenylphosphino)ferrocene)30
and 2d .31
Crystals of 2a were obtained during a study of the
mechanism of ionic hydrogenation.32 Crystals of 2b, 2c,
and 2e were grown by adding CH3OH to a saturated
benzene solution of the hydrides. Thermal ellipsoid plots
of the solid-state structures of 2a , 2b, 2c, and 2e are
shown in Figures 4-7. (Figure 6 shows only one of the
two independent molecules of 2c in a unit cell.33)
The coordination spheres of Ru in all these hydride
complexes adopt pseudo-octahedral geometries. (All the
hydride ligands were located except the one in 2e.) The
Ru-P distances of these hydrides (from 2.21 Å in 2c to
2.25 Å in 2e) are similar to those found in the structures
previously reported. The only significant differences are
in the P-Ru-P angles, which vary from 72.01(3)° in
2a to 93.96(3)° in 2e. Relevant bite angles are shown
in Table 1.
Measuring the kinetics of H- transfer to 1 from 2d
required the addition of TEMPO to suppress H/Cl
exchange between 2d and CD2Cl2. (A control experiment
with the stable hydride 2b showed that the presence of
TEMPO had no effect on the rate of hydride transfer.25)
The kinetics behavior of 2d was similar to that of 2b
(Table S-3, Figure S-3). The rate constant k for 2d was
calculated from the slope of Figure S-3 and is shown in
Table 1.
The transfer of hydride to 1 from 2a was so fast that
it was difficult to measure kobs under pseudo-first-order
conditions (the reaction of 2a with 1 was complete
within minutes). The rate constant was measured using
a 1:2 ratio of [2a ] to [1]. Under these conditions, a
second-order reaction between 2a and 1 will behave
according to eq 7,26 where [RuH]t is the concentration
of 2a at time t, [RuH]0 is the initial concentration of
2a , and k is the second-order rate constant.
ln{1 + (∆0[RuH]0/[2a ]0[RuH]t)} )
ln([1]0/[2a ]0) + k∆0t (7)
Rela tion sh ip betw een Ch ela te Bite An gle a n d
Hyd r id e Tr a n sfer Ra te. As Table 1 shows, k decreases
∆0 ) [1]0 - [2a ]0
When [1]0 ) 2[2a ]0, ∆0 ) [2a ]0, eq 7 can be rewritten as
(27) Two experiments with different initial concentrations gave the
eq 8.
same second-order rate constants within experimental error: k2
)
6.5(2) × 10-1 s-1 M-1 when [2a ] ) 7.5 × 10-3 M, [1] ) 1.5 × 10-2 M;
k2 ) 6.3(3) × 10-1 s-1 M-1 when [2a ] ) 1.0 × 10-2 M, [1] ) 2.0 ×
10-2 M.
ln{1 + ([RuH]0/[RuH]t)} ) ln 2 + k[2a ]0t
(8)
(28) Smith, K. T.; Rømming, C.; Tilset, M. J . Am. Chem. Soc. 1993,
115, 8681-8689.
(29) (a) Lemke, F. R.; Brammer, L. Organometallics 1995, 14, 3980-
3987. (b) Brammer, L.; Klooster, W. T.; Lemke, F. R. Organometallics
1996, 15, 1721-1727.
Experimental data confirm a linear relationship
between ln{1 + [RuH]0/[RuH]t} and t (Figure 3). Fur-
(24) Aneetha, H.; J ime´nez-Tenorio, M.; Puerta, M. C.; Valerga, P.;
Sapunov, V. N.; Schimid, R.; Kirchner, K.; Mereiter, K. Organometallics
2002, 21, 5334-5346. Extensive earlier theoretical work on such
systems is cited therein.
(30) Bruce, M. I.; Butler, I. R.; Cullen, W. R.; Koutsantonis, G. A.;
Snow, M. R.; Tiekink, E. R. T. Aust. J . Chem. 1988, 41, 963-969.
(31) Litster, S. A.; Redhouse, A. D.; Simpson, S. J . Acta Crystallogr.
1992, C48, 1661-1663.
(25) kobs ) 3.6(1) × 10-3 s-1 when [2b] ) 1.5 × 10-2 M, [1] ) 1.5 ×
10-1 M; kobs ) 3.5(2) × 10-2 s-1 when [2b] ) 1.5 × 10-2 M, [1] ) 1.5 ×
10-1 M, with 1% (by weight) TEMPO.
(32) Single crystals of 2a for structure analysis were grown from a
saturated solution of 2a , a slight excess of 1 (1.5 equiv), and an excess
of free amine 7 (ca. 7 equiv) at -35 °C.22
(26) Espenson, J . H. Chemical Kinetics and Reaction Mechanisms,
2nd ed.; McGraw-Hill: New York, 1995; p 25, eq 2-34.
(33) The aromatic ring in the backbone of the other molecule is
disordered.

4088 Organometallics, Vol. 22, No. 20, 2003
Guan et al.
increasing steric repulsion from the longer “backbone”
between the phosphorus atoms.
There is, however, good evidence for electronic control
of H- transfer rates as a function of the other ligands
present. Bullock and co-workers have reported that,
when Ph3C+BF4- is the hydride acceptor, the rate of H-
transfer is significantly enhanced when CO is replaced
by a tertiary phosphine: trans-CpMo(CO)2(PMe3)H is
104 times as reactive as CpMo(CO)3H.7k,l
Tabulated parameters (EL ) 0.43 V for dppm, 0.36 V
for dppe) for estimating the potentials of octahedral
complexes34 imply a more positive potential for 2a than
for 2b, suggesting that the energy of the highest
occupied Ru orbital (the HOMO) increases in the order
2a < 2b < 2d < 2e. Indeed, CV measurements (anodic
peak potentials of irreversible oxidation waves) on 2a ,
2b, and 2d show E ) -0.04 V for 2a , -0.09 V for 2b,
and -0.22 V for 2d ,17 in agreement with the predicted
order of the HOMO for 2a , 2b, and 2d .16
F igu r e 5. Molecular structure of CpRu(dppe)H (2b) (20%
probability level). Selected bond lengths (Å) and angle
(deg): Ru-P1 2.2365(4), Ru-P2 2.2322(4), P1-Ru-P2
84.497(16).
However, the hydride donor ability of a hydride
complex is more closely related to the energy of the
orbital left behind after hydride transfer than it is to
the energy of its own HOMO. For example, the hydride
donor abilities of (P-P)2MH+ (M ) Ni, Pt) toward a
generic substrate S+ (eq 9) correlate with the reduction
potentials (II/I) of (P-P)2M2+;8 these potentials measure
the energy of the acceptor orbital (the LUMO) of the
coordinatively unsaturated, 16-electron 8. Larger che-
late bites distort 8 from planarity, lowering the energy
of its LUMO and making the equilibrium constant for
eq 9 less favorable.8
F igu r e 6. Molecular structure of CpRu(dpbz)H (2c) (20%
probability level). Selected bond lengths (Å) and angles
(deg): Ru1-P1 2.2138(9), Ru1-P2 2.2147(10), P1-C61
1.834(4), P2-C62 1.841(4), C61-C62 1.382(5), P1-
Ru1-P2 84.22(4), Ru1-P1-C61 110.29(12), P1-C61-C62
115.9(3), C61-C62-P2 115.3(3), C62-P2-Ru1 110.61(13).
(P-P)2MH+ + S+ h (P-P)2M2+ + S-H
M ) Ni, Pt
(9)
8
The rate of hydride transfer from 2 to 1 is indepen-
dent of [CH3CN], so its rate-determining step is prob-
ably the forward reaction in eq 10, forming the coordi-
natively unsaturated, 16-electron, intermediates 9. Such
complexes prefer a pyramidal geometry, with a LUMO
that will be raised in energy by the additional bending
imposed by a smaller chelate ring24 (and by the stabiliz-
ing agostic interactions23 that such a ring may make
possible). The energy of the LUMO of 9 will thus, as
with 8, decrease as the size of its chelate ring increases.
CpRu(P-P)H + S+ h CpRu(P-P)+ + S-H (10)
9
Exp er im en ta l Section
F igu r e 7. Molecular structure of CpRu(dppb)H (2e) (20%
probability level). Selected bond lengths (Å) and angle
(deg): Ru-P1 2.2452(8), Ru-P2 2.2463(8), P1-Ru-P2
93.96(3).
Gen er a l P r oced u r es. All air-sensitive compounds were
prepared and handled under
a N2/Ar atmosphere using
standard Schlenk and inert-atmosphere box techniques. Tolu-
ene and hexanes were deoxygenated and dried over two
successive activated alumina columns under argon.35 Benzene
was distilled from Na and benzophenone under a nitrogen
atmosphere. CD2Cl2 was dried over CaH2, degassed by three
freeze-pump-thaw cycles, and then purified by vacuum
transfer at room temperature. CpRu(PPh3)2Cl,36 CpRu(dppm)H
(2a ),9 CpRu(dppe)H (2b),9 CpRu(dppp)H (2d ),10 CpRu(dppb)H
(2e),10 and 1-(1-phenylethylidene)pyrrolidinium tetrafluorobo-
rate37 (1) were prepared as described in the literature.
by at least an order of magnitude for each additional
carbon in the backbone of the chelating diphosphine
ligand. One possible explanation is steric: more space
is available for the approach of the iminium cation when
the chelate ring is smaller. A wider bite angle (P-Ru-
P) should retard hydride transfer. Table 1 shows that
there is little increase in bite angle between hydride 2d
(91.9(1)°) and hydride 2e (93.96(3)°), probably because
there is little room for expansion beyond the 90° of an
octahedral geometry. However, the rate constant for
hydride transfer (k) declines significantly (2.8(2) × 10-3
M-1 s-1 for 2d , no reaction for 2e), perhaps because of
(34) Lever, A. B. P. Inorg. Chem. 1990, 29, 1271-1285.
(35) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J . Organometallics 1996, 15, 1518-1520.
(36) Bruce, M. I.; Windsor, N. J . Aust. J . Chem. 1977, 30, 1601-
1604.

Effect of Chelate Ring Size
Organometallics, Vol. 22, No. 20, 2003 4089
Ta ble 2. Su m m a r y of Cr ysta llogr a p h ic Da ta
4‚C6H14
2a ‚CH2Cl2
2b‚C6H6
2c
2e(-H)
empirical formula
fw
C
59H58ClP3Ru
C31H30Cl2P2Ru
636.46
C37H36P2Ru
643.67
C35H30P2Ru
613.60
C33H33P2Ru
592.60
996.48
temp, K
cryst syst
space group
a, Å
b, Å
c, Å
293(2)
monoclinic
P21/c
9.7432(6)
23.3215(16)
20.5948(14)
90
94.0850(10)
90
293(2)
monoclinic
P21/c
9.7901(5)
17.0869(10)
17.2365(10)
90
95.2370(10)
90
243(2)
monoclinic
P21/c
11.7095(6)
18.9662(9)
14.3963(7)
90
107.3910(10)
90
293(2)
monoclinic
P21/c
11.0831(7)
27.3950(17)
18.6843(11)
90
92.2170(10)
90
293(2)
monoclinic
P21/c
10.6447(5)
19.8601(9)
13.3136(6)
90
98.7860(10)
90
R, deg
â, deg
γ, deg
V, Å3
4667.8(5)
4
1.418
2871.3(3)
4
1.472
3051.0(3)
4
1.401
5668.7(6)
8
1.438
2781.5(2)
4
1.415
Z
d
calc, g/cm3
λ(Mo KR), Å
0.71073
0.537
31 301
10 391
10391/0/524
1.028
0.71073
0.862
19 076
6469
6469/0/330
1.028
0.71073
0.643
20 068
6851
6851/0/366
1.029
0.71073
0.689
33 512
12 872
12 872/0/694
1.011
0.71073
0.699
18 736
6319
6319/0/326
1.030
µ, mm-1
no. of data collected
no. of unique data
no. of data/restraints/params
GOP on F2
R1, wR2 (I > 2σ(I))
R1, wR2 (all data)
0.0366, 0.0724
0.0667, 0.0791
0.0449, 0.1176
0.0694, 0.1285
0.0250, 0.0633
0.0291, 0.0652
0.0531, 0.1347
0.0671, 0.1437
0.0358, 0.0937
0.0548, 0.1017
Cp R u (η2-d p b z)Cl (3c) a n d [Cp R u (P P h 3)(η2-d p b z)]Cl
(4). Under a nitrogen atmosphere 300 mL of benzene was
added to a mixture of CpRu(PPh3)2Cl (1.162 g,1.6 mmol) and
dpbz (734 mg, 1.64 mmol), giving an orange-red solution. While
the solution was boiled for 24 h, a light green precipitate
formed, which was removed by filtration after the mixture was
cooled to room temperature. The volume of the orange filtrate
was reduced to 40 mL, then hexanes were added to cause
precipitation. Solvent was removed by cannula, and the orange
solid was washed with hexanes three times, then dried under
vacuum to give 591 mg of the orange powder 3c (57% yield).
1H NMR (400 MHz, CDCl3): δ 4.62 (s, Cp, 5H), 6.92-6.96 (m,
Ar, 4H), 7.09-7.13 (m, Ar, 4H), 7.22-7.27 (m, Ar, 2H), 7.36-
7.42 (m, Ar, 8H), 7.67-7.69 (m, Ar, 2H), 7.90-7.94 (m, Ar,
4H). 31P NMR (121.1 MHz, CDCl3): δ 75.21 (s). Its LRMS
(FAB+) showed a parent ion peak at m/e 648.1 (102Ru), with
the appropriate isotopic distribution. Anal. Calcd for C35H29P2-
RuCl: C, 64.87; H, 4.51; Cl, 5.47. Found: C, 64.75; H, 4.70;
Cl, 5.47. The light green solid after first filtration was washed
with bezene twice and hexanes three times, then dried under
vacuum to afford 400 mg of 4 as a light green powder (27%
NMR Mea su r em en ts of Ra te Con sta n ts for Hyd r id e
Tr a n sfer fr om th e Hyd r id e Com p lexes 2a -d to th e
Im in iu m Ca t ion 1. In a typical experiment, 2b (11.3 mg,
20 µmol) and 1 (52.2 mg, 200µmol) were placed in a vial in an
inert atmosphere box. CD2Cl2 (1.0 mL) containing CH3CN
(4.2 µL, 80 µmol) was added by a gastight syringe, and the
solution was immediately transferred to a J . Young NMR tube.
The first 1H NMR spectrum was recorded within 5 min, and
single-pulse spectra were taken every 120 s for three to five
half-lives. The height of the Cp resonance of the hydride
(δ 4.66) was compared to that of residual CDHCl2 (δ 5.32).
Zero-filling was used to ensure adequate digital resolution. The
probe temperature (300 K) was calibrated using anhydrous
methanol.38
For 2a the procedure was similar except that 2 equiv of 1
was used. For 2d , 1% (by weight) TEMPO was dissolved in
CD2Cl2 before CD2Cl2 was added to the starting materials. For
2c hexamethylcyclotrisiloxane was used as internal standard,
and the reaction was monitored by observing the growth of
the signal of coordinated CH3CN (δ 0.99) in the product [CpRu-
(dpbz)(CH3CN)]BF4 (6c). For 2e <5% reaction was observed
after 24 h with excess 1 (0.15 M), implying kobs < 6 × 10-7 s-1
1
yield). H NMR (400 MHz, CDCl3): δ 4.71 (s, Cp, 5H), 6.33-
6.38 (m, Ar, 4H), 6.40-6.50 (m, Ar, 2H), 6.80-6.90 (m, Ar,
2H), 6.95-7.05 (m, Ar, 12H), 7.07-7.12 (m, Ar, 4H), 7.19-
7.26 (m, Ar, 7H), 7.30-7.34 (m, Ar, 4H), 7.66-7.67 (m, Ar,
4H). 31P NMR (121.1 MHz, CDCl3): δ 43.39 (t, J P-P ) 34.0
Hz), 72.05 (d, J P-P ) 34.0 Hz). Its LRMS(FAB+) showed m/e
and k < 4 × 10-6 M-1 s-1
.
X-r a y Str u ctu r e Deter m in a tion s. Crystal data collection
and refinement parameters are summarized in Table 2. Data
were collected on a Bruker P4 diffractometer equipped with a
SMART CCD detector. The structures were solved using direct
methods and standard difference map techniques and refined
by full-matrix least-squares procedures using SHELXTL.
Hydrogen atoms on carbon were included in calculated posi-
tions.
875.1 for [CpRu(PPh3)(η2-dpbz)]+
(
102Ru) and m/e 613.1 for
102Ru) with the appropriate isotopic distri-
[CpRu(η2-dpbz)]+
(
bution. Anal. Calcd for C53H44P3RuCl: C, 69.92; H, 4.87; Cl,
3.89. Found: C, 70.10; H, 4.97; Cl, 3.60.
Cp Ru (η2-d p bz)H (2c). Sodium (110 mg, 4.78 mmol) was
added to a suspension of 575 mg of CpRu(dpbz)Cl (3c) (0.66
mmol) in 30 mL of CH3OH. Boiling the resulting mixture for
2 h gave a yellow suspension, from which (after cooling to room
temperature) the solvent was removed by cannula. The solid
was washed with CH3OH three times, then dried under
vacuum to give 242 mg of yellow powder 2c (48% yield). 1H
NMR (400 MHz, CD2Cl2): δ -13.36 (t, RuH, J P-H ) 33.2 Hz,
1H), 4.63 (s, Cp, 5H), 7.16-7.27 (m, Ar, 18H), 7.32-7.34 (m,
Ar, 2H), 7.52-7.56 (m, Ar, 4H). 31P NMR (CD2Cl2): δ 90.26
(s). Its LRMS (FAB+) showed a parent ion peak at m/e 613.6
Ack n ow led gm en t. This research was supported by
NSF grant CHE-0211310. We thank Dr. D. L. DuBois
and Dr. R. M. Bullock for helpful discussions, and Prof.
G. Parkin for assistance with the X-ray structure
determinations.
Su p p or tin g In for m a tion Ava ila ble: Complete details of
the crystallographic study (PDF and CIF), kinetic data, and
plots of these data. This material is available free of charge
via the Internet at http://pubs.acs.org.
(
102Ru), with the appropriate isotopic distribution. Anal. Calcd
for C35H30P2Ru: C, 68.51; H, 4.93. Found: C, 67.82; H, 4.77.
OM0303404
(37) Leonard, N. J .; Paukstelis, J . V. J . Org. Chem. 1963, 28, 3021-
3024.
(38) (a) van Geet, A. L. Anal. Chem. 1970, 42, 679-680. (b) Raidford,
D. S.; Fisk, C. L.; Becker, E. D. Anal. Chem. 1979, 51, 2050-2051.