Synthesis, electrochemistry, and reactivity of half-sandwich ruthenium complexes bearing metallocene-based bisphosphines

Article abstract of DOI:10.1021/om900062a

The bimetallic complexes CpRu(P-P)X [Cp = n⁵-C₅H ₅; X = Cl, H; P-P = dppf (1,1′-bis(diphenylphosphino)ferrocene) , dppr (1,1′-bis(diphenylphosphino)ruthenocene), dppo (1,1′- bis(diphenylphosphino)-osmocene), dippf (1,1′-bi

Full text of DOI:10.1021/om900062a

3804 Organometallics 2009, 28, 3804–3814
DOI: 10.1021/om900062a
Synthesis, Electrochemistry, and Reactivity of Half-Sandwich Ruthenium
Complexes Bearing Metallocene-Based Bisphosphines
Anthony P. Shaw, Jack R. Norton,* and Daniela Buccella
Department of Chemistry, Columbia University, New York, New York 10027
Lauren A. Sites,† Shannon S. Kleinbach, Daniel A. Jarem,‡
Katherine M. Bocage,§ and Chip Nataro*
Department of Chemistry, Lafayette College, Easton, Pennsylvania 18042. †Current address: Department
of Chemistry, Pennsylvania State University, University Park, PA 16802. ‡Current address: Department
of Chemistry, Brown University, Providence, RI 02912. §Current address: Church & Dwight Co., Inc.,
Princeton, NJ 08540
Received January 27, 2009
The bimetallic complexes CpRu(P-P)X [Cp=η5-C5H5; X=Cl, H; P-P=dppf (1,10-bis(diphenylphos-
phino)ferrocene), dppr (1,10-bis(diphenylphosphino)ruthenocene), dppo (1,10-bis(diphenylphosphino)-osm-
ocene), dippf (1,10-bis(diisopropylphosphino)ferrocene), dcpf (1,10-bis(dicyclohexylphosphino)ferrocene)],
Cp*Ru(P-P)X [Cp*=η5-C5Me5; X=Cl, H; P-P=dppf, dippf, dppomf (1,10-bis(diphenylphosphino)
octamethylferrocene), dppc (1,10-bis(diphenylphosphino)cobaltocene)], [Cp*Ru(P-P)X]+ (X=H, CCPh;
P-P=dppc+), and [Cp*Ru(P-P)L]2+ (L=CH3CN, t-BuCN; P-P=dppc+) have been synthesized.
Most of the chloride and hydride complexes have been studied by cyclic voltammetry. The X-ray structures
of [Cp*Ru(dppc)CH3CN][PF6]2 and [Cp*Ru(dppc)CCPh][PF6] have been determined. Protonation of
[Cp*Ru(dppc)CCPh]+ gives the vinylidene complex [Cp*Ru(dppc)CCHPh]2+. The Co(III/II) potential of
the dppc+ ligand undergoes a cathodic shift upon coordination in [Cp*Ru-(dppc)H]+ and an anodic shift
upon coordination in [Cp*Ru(dppc)CH3CN]2+. The 1H NMR spectrum of Cp*Ru(dppc)H is consistent
with its formulation as a Co(II)/Ru(II) complex. As gauged by their reactivity toward iminium cations, the
hydride complexes are poor hydride donors; proton and electron transfer are dominant. CpRu(dippf)H and
CpRu(dcpf)H deprotonate iminium cations with acidic R-hydrogens. Cp*Ru(dppc)H is oxidized by the
N-(benzylidene)pyrrolidinium cation, giving [Cp*Ru-(dppc)H]+ and the vicinal diamine 1,2-bis(N-pyrro-
lidino)-1,2-diphenylethane. Most of the hydride complexes give trans-dihydride cations upon protona-
tion; an exception is [Cp*Ru(dppc)H]+, which forms a dihydrogen complex [Cp*Ru(dppc)(H2)]2+ with
surprising kinetic stability. This dihydrogen complex is more acidic and less thermodynamically stable than
its dihydride isomer. The H2 ligand in [Cp*Ru(dppc)-(H2)]2+ is readily replaced by nitriles; the reaction with
t-BuCN occurs by a dissociative mechanism.
Introduction
recently been reported that dtbpf (1,10-bis(di-tert-butylpho-
sphino)ferrocene) is superior to dppf for the Pd-catalyzed
R-arylation of ketones and for certain Pd-catalyzed Suzuki
couplings.6,7 Similarly, the ligands Fe(η5-C5H4P(o-(i-Pr)-
C6H4)2)2 and Fe(η5-C5H4P(o-MeOC6H4)2)2 give higher cat-
alytic activities than dppf in some Pd-catalyzed Suzuki and
Buchwald-Hartwig reactions.8 The chemistry of Ru- and
Os-based dppf analogues is just beginning to be explored.
The X-ray structures of Pd(dppr)Cl2 and Pd(dppo)Cl2 have
recently been reported,9,10 as has their activity in several
Suzuki and Buchwald-Hartwig reactions.8
Since 1,10-bis(diphenylphosphino)ferrocene (dppf) was
reported in 1971,1 this metallocene-based bisphosphine has
attracted much attention. The dppf ligand is valued for the
unique steric and electronic properties it bestows upon its
complexes, particularly those of Pd.2 In many cases, the use
of dppf improves the yields of Pd-catalyzed Suzuki, Heck,
and Buchwald-Hartwig coupling reactions.2-4 The use of
dppf analogues appears to have a promising future.5 It has
*Corresponding authors. E-mail: jrn11@columbia.edu.
(1) Bishop, J. J.; Davison, A.; Katcher, M. L.; Lichtenberg, D. W.;
Merrill, R. E.; Smart, J. C. J. Organomet. Chem. 1971, 27, 241–249.
(2) Chien, S. W.; Hor, T. S. A. In Ferrocenes: Ligands, Materials, and
Biomolecules; Stepnicka, P., Ed.; John Wiley & Sons, Ltd.: West Sussex,
England, 2008, pp 33-116.
(3) Colacot, T. J. Platinum Met. Rev. 2001, 45, 22–30.
(4) Hartwig, J. F. Angew. Chem., Int. Ed. 1998, 37, 2047–2067.
(5) Colacot, T. J.; Parisel, S. In Ferrocenes: Ligands, Materials, and
Biomolecules; Stepnicka, P., Ed.; John Wiley & Sons, Ltd.: West Sussex,
England, 2008, pp 117-140.
(6) Grasa, G. A.; Colacot, T. J. Org. Lett. 2007, 9, 5489–5492.
(7) Colacot, T. J.; Shea, H. A. Org. Lett. 2004, 6, 3731–3734.
(8) Gusev, O. V.; Peganova, T. A.; Kalsin, A. M.; Vologdin, N. V.;
Petrovskii, P. V.; Lyssenko, K. A.; Tsvetkov, A. V.; Beletskaya, I. P.
Organometallics 2006, 25, 2750–2760.
(9) Nataro, C.; Campbell, A. N.; Ferguson, M. A.; Incarvito, C. D.;
Rheingold, A. L. J. Organomet. Chem. 2003, 673, 47–55.
(10) Martinak, S. L.; Sites, L. A.; Kolb, S. J.; Bocage, K. M.;
McNamara, W. R.; Rheingold, A. L.; Golen, J. A.; Nataro, C.
J. Organomet. Chem. 2006, 691, 3627–3632.
r
pubs.acs.org/Organometallics
Published on Web 06/03/2009
2009 American Chemical Society

Article
The 1,10-bis(diphenylphosphino)cobaltocenium cation
Organometallics, Vol. 28, No. 13, 2009 3805
Scheme 1. Synthesis of CpRu Chlorides and Hydrides
(dppc+), isoelectronic with dppf, was first reported in
1978.11 The neutral paramagnetic compound (dppc) was later
isolated and characterized by X-ray crystallography.12 The
dppcligand is nearly identical in size and shape todppf13 but is
much more easily oxidized.9,12,14 For this reason the “redox
switch” properties of dppc+/dppc have been investigated:
Wrighton and co-workers reported that [Re-(CO)4dppc]2+
reacts with azide 5400 times faster than the reduced form, [Re
(CO)4dppc]+.14 Soon after, they reported that [Rh(dppc)
(acetone)n]2+ is an active hydrosilylation catalyst, while [Rh
(dppc)(acetone)n]+ is an active hydrogenation catalyst.15
Given such interesting activity, it is surprising that so few
dppc+/dppc complexes have been reported.
Combining our interests in half-sandwich ruthenium com-
plexes (J.R.N.) and metallocene-based ligands (C.N.), we
report here the synthesis, properties, and electrochemistry of
(Cp/Cp*)Ru complexes bearing several metallocene-based
bisphosphines.16 In view of the interest in one of our labora-
tories (J.R.N.) in hydride transfer reactions,17-19 we have
tested the ability of the hydride complexes to transfer H- to
iminium cations but have found that they are ineffective in
that role. However, electrochemical studies of the hydride
and chloride complexes have allowed us to compare the
electronic influence of the various bisphosphine ligands.
As few dppc+ or dppc complexes have been reported, we
have made them with nitrile, phenylacetylide, phenylacety-
lene, hydride, and dihydrogen ligands. (The only previously
reported half-sandwich Ru complex containing dppc+ or
dppc is [Cp*Ru(dppc)Cl][PF6].20) We have examined the
redox chemistry of the acetonitrile and hydride complexes.
We also report the X-ray structures of the acetonitrile and
phenylacetylide complexes.
Scheme 2. Synthesis of Cp*Ru Chlorides and Hydrides
Results and Discussion
Synthesis of the Complexes. Like many complexes of the
type CpRu(P-P)Cl, the Ru chlorides 1a-1e can be prepared
by treating CpRu(PPh3)2Cl with the corresponding bispho-
sphine in refluxing benzene (Scheme 1). This method worked
well even with the bulky dippf and dcpf ligands, but it failed
with the exceptionally hindered dtbpf ligand; in the latter
case the retention of PPh3 was thermodynamically favored.
In general, Ru hydrides can be synthesized by treating the
chlorides with NaOMe in refluxing MeOH. After preparing
(11) Rudie, A. W.; Lichtenberg, D. W.; Katcher, M. L.; Davison, A.
Inorg. Chem. 1978, 17, 2859–2863.
(12) DuBois, D. L.; Eigenbrot, C. W.; Miedaner, A.; Smart, J. C.;
Haltiwanger, R. C. Organometallics 1986, 5, 1405–1411.
ꢀ
(13) Casellato, U.; Ajo, D.; Valle, G.; Corain, B.; Longato, B.;
Graziani, R. J. Crystallogr. Spectrosc. Res. 1988, 18, 583–590.
(14) Lorkovic, I. M.; Wrighton, M. S.; Davis, W. M. J. Am. Chem.
Soc. 1994, 116, 6220–6228.
(15) Lorkovic, I. M.; Duff, R. R.; Wrighton, M. S. J. Am. Chem. Soc.
1995, 117, 3617–3618.
(16) A recent review on (Cp/Cp*)Ru complexes containing dppf: Lu,
X. L.; Yu, X.; Lou, J.-D. Synth. React. Inorg. Met.-Org. Nano-Met.
Chem. 2006, 36, 733–745.
2a21 and 2d in this way (method A), we found that these
hydrides are also available in good yield by stirring the
respective chlorides with excess NaBH4 in alcohol at room
temperature (method B). We subsequently prepared 2b, 2c,
and 2e by method B.
(17) Guan, H.; Iimura, M.; Magee, M. P.; Norton, J. R.; Zhu, G.
J. Am. Chem. Soc. 2005, 127, 7805–7814.
(18) Guan, H.; Saddoughi, S. A.; Shaw, A. P.; Norton, J. R.
Organometallics 2005, 24, 6358–6364.
(19) Shaw, A. P.; Ryland, B. L.; Franklin, M. J.; Norton, J. R.; Chen,
J. Y.-C.; Hall, M. L. J. Org. Chem. 2008, 73, 9668–9674.
(20) The X-ray structure of 1j[PF6] has been reported: Wu, X.-H.;
Ren, Y.; Yu, G.-A.; Liu, S.-H. Acta Crystallogr. 2006, E62, m426–m428.
(21) 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.
Hembre reported the synthesis of Cp*Ru(dppf)Cl/H
(1f/2f) by treating dppf with refluxing Cp*Ru(PPh3)2Cl/
benzene followed by refluxing NaOMe/MeOH.22 The first
part of this sequence (method C, Scheme 2) worked for 1i but
not for 1g or 1h; the latter two complexes were prepared from
(22) Hembre, R. T.; McQueen, J. S.; Day, V. W. J. Am. Chem. Soc.
1996, 118, 798–803.

3806 Organometallics, Vol. 28, No. 13, 2009
Shaw et al.
Figure 1. Upfield 1H NMR spectrum of 2i (400 MHz, toluene-d8). The small peak at δ -50.4 is due to some residual Cp2Co.
Scheme 3. Synthesis of the dppc and dppc+ Complexes
Cp*Ru(1,5-COD)Cl (method D) since 1,5-cyclooctadiene is
easily displaced from Ru(II).23,24 The Cp*Ru hydrides
(except for 2i, see below) were prepared by the usual treat-
ment of the chlorides with refluxing NaOMe/MeOH.
The dppc and dppc+ complexes synthesized in this work
are shown in Scheme 3. The Co(II) in 1i is readily oxidized to
Co(III) in the presence of an acid and atmospheric oxygen. A
convenient acid is NH4PF6. It is convenient to manipulate
these complexes with the ligand in the oxidized state since
they are then generally air-stable and amenable to flash
chromatography. Heating a CH3CN solution of 1j[PF6]20
with excess NH4PF6 gave the characteristically crimson 6j-
[PF6]2, which, after purification by flash chromatography,
was used as a precursor to the other complexes.
is a triplet due to coupling with the two 31P nuclei of the
bisphosphine ligand (δ -12.28, JP-H=34.7 Hz). The reduc-
tion of 2j+ with Cp2Co gave the paramagnetic hydride
complex 2i; a portion of its 1H NMR spectrum is shown in
Figure 1. As expected for a Co(II) complex, the resonances
with the greatest isotropic shifts are those of nuclei closest to
the Co(II) atom: the cyclopentadienyl hydrogens of the dppc
ligand are found at δ -57.8, -52.2, -27.3, and -24.5. The
hydride resonance appears at δ -15.76 as a broad (about
50 Hz at half-height) singlet. It is noteworthy that no 31P-1H
coupling is visible (recall the large ∼35 Hz splitting for the
hydride triplet of 2j+). Such apparent decoupling, known as
T1 spin decoupling, was quantitatively described nearly
35 years ago,25 but relatively few unambiguous examples
have since appeared in the literature. The hydride resonance
of 2j+/2i can be described as the A subsystem of an AX2
system, where X2 are in close proximity to a metal. When the
metal (in this case Co) is paramagnetic, X2 (the two 31P) relax
quickly; this causes additional broadening in the resonance
of A (the hydride ligand) as well as a decrease in the apparent
X2-A (31P-1H) coupling constant. The two effects result in
a singlet for the hydride resonance of 2i, although a Ni(II)
After failing to prepare pure samples of 2i or 2j[PF6] in
reasonable yield from 1i, 1j[PF6], or 6j[PF6]2 by the usual
treatment with refluxing NaOMe/MeOH, we found the
reaction of 6j[PF6]2 with H2 and NEt3 in MeOH to be a
viable route to 2j[PF6]. As with the other hydride comp-
1
lexes in this work, the H NMR hydride resonance of 2j+
(23) Albers, M. O.; Robinson, D. J.; Shaver, A.; Singleton, E.
Organometallics 1986, 5, 2199–2205.
(24) Beck, C. U.; Field, L. D.; Hambley, T. W.; Humphrey, P. A.;
Masters, A. F.; Turner, P. J. Organomet. Chem. 1998, 565, 283–296.
(25) Navon, G.; Polak, M. Chem. Phys. Lett. 1974, 25, 239–242.

Article
Organometallics, Vol. 28, No. 13, 2009 3807
Figure 3. Molecular structure of 7j[PF6] Et2O (20% probabil-
3
Figure 2. Molecular structure of 6j[PF6]2 Et2O CH3CN (20%
3
3
ity level). Counterion, hydrogen atoms, and solvent molecule
are omitted for clarity. Selected distances (A) and angles (deg):
Ru-C21 2.012(4), Ru-P1 2.2801(9), Ru-P2 2.2737(10), C21-
C22 1.212(5), C22-C23 1.434(5), Ru-C21-C22 175.3(3),
C21-C22-C23 166.4(4), P1-Ru-P2 98.23(3), P1-Ru-C21
86.34(10), P2-Ru-C21 87.98(11), Ru-P1-C11 121.83(11),
Ru-P2-C16 121.70(12).
probability level). Counterions, hydrogen atoms, and disor-
dered solvent molecules are omitted for clarity. Selected dis-
tances (A) and angles (deg): Ru-N 2.063(5), Ru-P1 2.3314(13),
Ru-P2 2.3198(14), N-C21 1.112(6), Ru-N-C21 166.3(4),
P1-Ru-P2 96.87(4), P1-Ru-N 91.15(12), P2-Ru-N 90.42
(12), Ru-P1-C11 120.51(15), Ru-P2-C16 119.92(16).
complex has recently been reported where a reduced X-A
coupling constant (in that case 19F-13C) was observable.26
To further study the reactivity of 6j2+, we attempted to
prepare a vinylidene complex by refluxing a CHCl3 solution
of 6j[PF6]2 and phenylacetylene. Instead, the green alkynyl
complex 7j[PF6] was obtained in poor yield. An acceptable
yield of 7j[PF6] was achieved by heating a slurry of 6j[PF6]2
with phenylacetylene and NEt3 in toluene. Treating 7j[PF6]
Table 1. Summary of Crystallographic Data
6j[PF6]2 Et2O CH3CN
7j[PF6] Et2O
3
3
3
empirical formula
fw
C52H59CoF12N2OP4Ru C56H58CoF6OP3Ru
1239.89
orange/plate
0.53 ꢀ 0.12 ꢀ 0.01
125(2)
1113.93
cryst color/shape
cryst size (mm)
temp (K)
cryst syst
space group
a (A)
green/plate
0.43 ꢀ 0.31 ꢀ 0.01
125(2)
monoclinic
P21/c
monoclinic
P21/c
with HBF4 OMe2 in CDCl3 gave a yellow solution of the
The Ru-bound 13C resonance is a weak
3
vinylidene 8j2+ 27
.
19.3450(18)
13.5915(12)
22.247(2)
90
15.8294(19)
15.4514(18)
20.606(2)
90
b (A)
c (A)
R (deg)
β (deg)
downfield triplet (δ 357.58, JP-C=15.7 Hz) and confirms the
vinylidene structure.28
X-ray Structures. The structures of 6j[PF6]2 and 7j[PF6]
were determined by single-crystal X-ray diffraction
(Figures 2, 3, Table 1). The only dppc+ complex of Ru with
a previously published X-ray structure is 1j[PF6].20 Gan and
Hor identified several structural parameters specific to dppf
and its analogues,29 some of which are listed for 1j[PF6],
6j[PF6]2, and 7j[PF6] in Table 2. The coordination of the
dppc+ ligand is similar in all three cases. In each structure,
the Cp rings are nearly eclipsed, as indicated by the low τ
values. The P atoms are slightly displaced out of the corre-
sponding Cp planes, away from the Co. The Cp rings are not
coplanar, as indicated by the nonzero θ and the small
deviations of Cpcent-Co-Cpcent from linearity. The ligand
bite angles (P1-Ru-P2) range from 96.7° to 98.2°. These
structures are similar to those of related Ru(II) dppf com-
plexes. The dppf bite angles of several (η5-C5R5)Ru(dppf)X
114.8370(10)
90
92.002(2)
90
γ (deg)
V (A3)
5308.4(8)
4
1.551
5036.8(10)
4
1.469
0.71073
0.786
78 581
15 606
15 606/0/614
1.048
Z
dcalc (g/cm3)
λ(Mo KR) (A)
0.71073
0.800
53 864
μ (mm-1
)
data collected
unique data
data/restraints/params 13 140/51/624
13 140
GOF on F2
R1, wR2 [I > 2σ(I)]
R1, wR2 (all data)
1.015
0.0601, 0.0994
0.1521, 0.1094
0.0494, 0.1016
0.1218, 0.1331
Table 2. Structural Parameters of the dppc+ Ligand
1j[PF6]20
6j[PF6]2
7j[PF6]
Ru Co (A)
P1 P2 (A)
4.438
3.458
4.371
3.480
4.384
3.443
-0.030
-0.101
2.53
3 3 3
3 3 3
δ
δ
P1 (A)a
P2 (A)a
-0.083
-0.081
3.63
-0.080
-0.019
1.19
::
::
(26) Kohn, R. D.; Haufe, M.; Kociok-Kohn, G. J. Am. Chem. Soc.
2006, 128, 10682–10683.
τ (deg)b
(27) Ru vinylidenes are catalytic precursors or intermediates in many
processes; a good review is: Bruneau, C. In Topics in Organometallic
Chemistry; Bruneau, C., Dixneuf, P. H., Eds.; Springer-Verlag: Berlin,
2004; Vol. 11, pp 125-153.
θ (deg)c
2.66
4.03
2.37
Cpcent-Co-Cpcent (deg)
P1-Co-P2 (deg)
P1-Ru-P2 (deg)
177.76
58.96
96.69(3)
176.76
59.76
96.87(4)
178.00
58.87
98.23(3)
(28) The dppf analogue of 8j2+ has been reported, and its Ru-bound
13C resonance is a triplet at δ 356.5 (JP-C=17.0 Hz): Sato, M.; Sekino,
M. J. Organomet. Chem. 1993, 444, 185–190.
a The distance between the P atom and the Cp plane, negative means
away from Co. b The torsion angle defined by CA-Cpcent(A)-Cpcent(B)
CB where CA and CB are the P-bound C atoms of the Cp rings.
-
(29) Gan, K.-S.; Hor, T. S. A. In Ferrocenes: Homogeneous Catalysis,
Organic Synthesis, Materials Science; Togni, A., Hayashi, T., Eds.; VCH
Publishers: New York, 1995; pp 3-104.
c The angle between the two Cp planes.

3808 Organometallics, Vol. 28, No. 13, 2009
Shaw et al.
Table 4. Ru(III/II) Potentials of the Hydride Complexesa
Table 3. Potentials of the Chloride Complexesa
b
complex
E1
E2
complex
potential
parameter
CpRu(dppf)Cl (1a)c
CpRu(dppr)Cl (1b)e
CpRu(dppo)Cl (1c)e
CpRu(dippf)Cl (1d)c
CpRu(dcpf)Cl (1e)e
Cp*Ru(dppf)Cl (1f)c
Cp*Ru(dppomf)Cl (1g)c
Cp*Ru(dippf)Cl (1h)c
+0.06
+0.08
+0.09
-0.07
-0.07
-0.13
-0.35
-0.34
+0.48d
CpRu(dppf)H (2a)
CpRu(dppr)H (2b)
CpRu(dppo)H (2c)
CpRu(dippf)H (2d)
CpRu(dcpf)H (2e)
Cp*Ru(dppf)H (2f)
Cp*Ru(dppomf)H (2g)
Cp*Ru(dippf)H (2h)
[Cp*Ru(dppc)H]+ (2j+
-0.35b
-0.41c
-0.20c
-0.61d
-0.65e
-0.63d
-0.72d
-0.90d
-0.19f
E1/2
Epa
Epa
E1/2
E1/2
E1/2
E1/2
E1/2
Epa
+0.46d
+0.45d
+0.44d
+0.20d
+0.37d
)
a E1/2 (V vs FcH+/FcH) for reversible cyclic voltammograms in
[Bu4N]PF6/CH2Cl2. b Ru(III/II). c 50 mV/s. d Fe(III/II). e 100 mV/s.
a Potentials (V vs FcH+/FcH) for reversible cyclic voltammograms in
[Bu4N]PF6/CH2Cl2 unless otherwise noted. b 50 mV/s, ia/ic = 1.17.
c 100 mV/s, irreversible. d 50 mV/s. e 100 mV/s. f 100 mV/s, [Bu4N]PF6/
CH3CN, irreversible.
and [(η5-C5R5)Ru(dppf)L]+ are in the range 95-99°.21,22,30
The Cp rings of the dppf ligands in these complexes are
usually eclipsed; notable exceptions are CpRu(dppf)H (2a)
and Cp*Ru(dppf)H (2f).21,22
donor ligand than chloride,38,39 so the Ru(III/II) potentials
of the hydrides are 0.4 to 0.6 V less positive than those of the
corresponding chlorides. Methyl substitution on the Ru-
bound Cp ring decreases the potentials substantially (com-
pare 2a/2f and 2d/2h). Hembre initially assigned the first
oxidation of Cp*Ru(dppf)H (2f) to Ru(III/II), and this was
later confirmed by the EPR and spectroelectrochemical
studies of Kaim.22,35
The potentials of the hydride complexes are more sensitive
to the nature of the PR2 groups than those of the chlorides.
The dippf and dcpf chlorides 1d and 1e are oxidized at
potentials only 0.13 V less positive than CpRu(dppf)Cl
(1a). The analogous hydride complexes 2d and 2e are oxi-
dized at potentials 0.26 and 0.30 V less positive than CpRu
(dppf)H (2a). In contrast, alkyl substitution on the ferrocenyl
rings affects the chlorides more than the hydrides. The
difference between the Ru(III/II) potentials of the dppf
and dppomf chlorides 1f and 1g is 0.22 V; the potentials of
2f and 2g differ by only 0.09 V.
The oxidations of 18-electron transition metal hydrides
are often irreversible because the resulting radical cations are
about 20 pKa units more acidic than the parent hydrides.40-43
Deprotonation of MH•+ is a common (but not the only)
follow-up reaction.44 Ferrocenyl bisphosphines stabilize the
Ru(III) hydride radical cations, making most of the oxida-
tions in Table 4 reversible. For example, the oxidation of
CpRu(dppe)H (dppe=1,2-bis(diphenylphosphino)ethane) in
[Bu4N]PF6/CH2Cl2 is irreversible,19 while that of CpRu-
(dppf)H (2a) under the same conditions is chemically rever-
sible. The dppr and dppo ligands apparently do not have the
same stabilizing effect and the oxidations of 2b and 2c are
irreversible.
Electrochemistry. The anodic electrochemistry of free
dppf,9,12,31 dppomf,31 dippf,32 dppr,9,31 dppo,10,31 and dppc12,14
has been reported. There are many reports of the elec-
trochemical oxidation of complexes containing these and
other metallocene-based bisphosphines.9,10,12,14,31-35 The
oxidation of (Cp/Cp*)Ru complexes containing such
ligands has not been studied extensively; the few chloride
and hydride complexes that have been examined include
CpRu(dppf)Cl (1a),22,36 Cp*Ru(dppf)Cl (1f),22 CpRu-
(dppf)H (2a),19,22 Cp*Ru(dppf)H (2f),19,22,35 and the car-
boxy- and ester-substituted complexes (CpCO2H)Ru(dppf)-
Cl and (CpCO2t-Bu)Ru(dppf)Cl.36
The Ru chlorides bearing ferrocenyl bisphosphines in
Table 3 exhibit two reversible one-electron oxidations. For
1a and 1f, Hembre assigned the first oxidation (E1) to Ru(III/
II) and the second (E2) to Fe(III/II).22 The E1 potential
decreases by 0.19 V when the Cp ligand in 1a is exchanged
for a Cp* in 1f, but E2 decreases only by 0.04 V. The
E2 potentials of 1a, 1d-f, and 1h are all within a 0.11 V
range, but the E2 of 1g, containing the methyl-substituted
dppomf ligand, is 0.17 V less than the lowest E2 of the other
complexes. These facts support the aforementioned assign-
ments of E1 and E2.
The Ru and Os metallocenes are considerably more diffi-
cult to oxidize than their Fe analogues.10 The dppr and dppo
complexes 1b and 1c undergo only one oxidation within the
potential window; presumably it is the Cp-bearing Ru(III/
II). The identity of the metal in the bisphosphine ligand
appears to have little influence over the E1 potentials of 1a-
1c.37 Instead, the E1 of (η5-C5R5)Ru(P-P)Cl is most affected
by substituents at the P atoms and the Ru-bound Cp ring.
Ru(III) radical cations result from the oxidation of the
hydride complexes (Table 4). A hydride is generally a better
Steric bulk in the vicinity of the hydride ligand slows the
rate of follow-up reactions involving that ligand. Examples
include (Cp/Cp*)Ru(dppe)H and (Cp/Cp*)Ru(dppf)H
(2a/2f): in both cases the Cp* complex displays the more
reversible oxidation.19 The complexes in Table 4 exhibit
(30) Lu, X. L.; Vittal, J. J.; Tiekink, E. R. T.; Tan, G. K.; Kuan, S. L.;
Goh, L. Y.; Hor, T. S. A. J. Organomet. Chem. 2004, 689, 1978–1990.
(31) (a) Gusev, O. V.; Peterleitner, M. G.; Kal’sin, A. M.; Vologdin,
N. V. Elektrokhimiya 2003, 39, 1444–1451. (b) English translation:
Russ. J. Electrochem. 2003, 39, 1293-1299.
(32) Ong, J. H. L.; Nataro, C.; Golen, J. A.; Rheingold, A. L.
Organometallics 2003, 22, 5027–5032.
(33) Blanco, F. N.; Hagopian, L. E.; McNamara, W. R.; Golen, J. A.;
Rheingold, A. L.; Nataro, C. Organometallics 2006, 25, 4292–4300.
(34) Ohs, A. C.; Rheingold, A. L.; Shaw, M. J.; Nataro, C. Organo-
metallics 2004, 23, 4655–4660.
(38) Tilset, M.; Fjeldahl, I.; Hamon, J.-R.; Hamon, P.; Toupet, L.;
Saillard, J.-Y.; Costuas, K.; Haynes, A. J. Am. Chem. Soc. 2001, 123,
9984–10000.
(39) Rottink, M. K.; Angelici, R. J. J. Am. Chem. Soc. 1993, 115,
7267–7274.
(40) Skagestad, V.; Tilset, M. J. Am. Chem. Soc. 1993, 115,
5077–5083.
(41) Tilset, M. J. Am. Chem. Soc. 1992, 114, 2740–2741.
(42) Ryan, O. B.; Tilset, M.; Parker, V. D. Organometallics 1991, 10,
298–304.
(35) Sixt, T.; Fiedler, J.; Kaim, W. Inorg. Chem. Commun. 2000, 3,
80–82.
(36) Shaw, A. P.; Guan, H.; Norton, J. R. J. Organomet. Chem. 2008,
693, 1382–1388.
(37) It is known that dppr is not significantly more basic than dppf:
O’Connor, A. R.; Nataro, C. Organometallics 2004, 23, 615–618.
(43) Ryan, O. B.; Tilset, M.; Parker, V. D. J. Am. Chem. Soc. 1990,
112, 2618–2626.
(44) CpRu(P-P)H•+ are known to disproportionate: Smith, K.-T.;
Romming, C.; Tilset, M. J. Am. Chem. Soc. 1993, 115, 8681–8689.
e

Article
Organometallics, Vol. 28, No. 13, 2009 3809
Table 5. Co(III/II) Potentials
transfer hydride to 10.45,46 However, in that study the rates
were greatly affected by the bisphosphine bite angle (with
smaller bite angles resulting in faster rates). The accessibility
of the hydride ligand is of primary importance. Guan noted
that CpRu(dppb)H (P1-Ru-P2=94°)47,46 does not react
with 10, so it is noteworthy that 2a, with a substantially larger
bite angle (P1-Ru-P2=99°),21 does (Table 6). The dppr
and dppo ligands in 2b and 2c are expected to have even
larger bite angles than the dppf in 2a. In the series Pd(P-P)
Cl2 the bite angles of dppf, dppr, and dppo are 97.98(4)°,
100.02(17)°, and 100.82(9)° respectively.9,10,48 The low re-
activities of 2b and 2c in Table 6 are likely the result of steric
hindrance. Another example demonstrating the importance
of steric factors is the reactivity of (Cp/Cp*)Ru(dppf)H
(2a/2f) with N-(benzylidene)pyrrolidinium cation (11); the
Cp complex transfers its hydride, while the more electron-
rich (but also more hindered) Cp* complex does not.19
Ru hydrides that are easily oxidized are also basic. For
example, CpRu(dppe)H (Epa=-0.16 V) transfers H- to 10,
while Cp*Ru(dppe)H (E1/2=-0.51 V) deprotonates 10.17,19
Both CpRu(dippf)H (2d) and CpRu(dcpf)H (2e) deproto-
nate 9 and 10 (eq 1).
complex
electrolyte
potentiala
Cp2Co
[Cp*Ru(dppc)H]+ (2j+
dppc
[Et4N]BF4/CH3CN
[Bu4N]PF6/CH3CN
[Et4N]BF4/CH3CN
[Bu4N]PF6/CH3CN
[Bu4N]BF4/THF
-1.33b
-1.27c
-1.11b
-1.05d
-0.98b
-0.91e
-0.71e
)
[Cp*Ru(dppc)CH3CN]2+ (6j2+
Mo(CO)4(dppc)
)
fac-[Re(CO)3(dppc)Br]+
[Re(CO)4(dppc)]2+
[Bu4N]PF6/CH3CN
[Bu4N]PF6/CH3CN
a E1/2 (V vs FcH+/FcH) for reversible cyclic voltammograms unless
otherwise noted. b Ref 12, 50 mV/s. c 200 mV/s. d 200 mV/s, ia/ic = 0.85.
e Ref 14.
Table 6. Stoichiometric Hydride Transfer Reactionsa
complex
iminium
reaction (%)b
CpRu(dppf)H (2a)
CpRu(dppr)H (2b)
CpRu(dppo)H (2c)
CpRu(dppf)H (2a)
CpRu(dppr)H (2b)
CpRu(dppo)H (2c)
R = Ph (10)
R = Ph (10)
R = Ph (10)
R = Me (9)
R = Me (9)
R = Me (9)
18
10
9
6
<3
<3
a General reaction conditions: 0.01 mmol of hydride complex,
0.01 mmol of iminium salt, 1 μL of CH3CN, and 1 mL of CD2Cl2.
b Percent loss of hydride complex after 3 h, determined by 1H NMR.
The N-(benzylidene)pyrrolidinium cation (11) does not
have any acidic protons, so we tested its reactivity toward
some of our hydride complexes. Like 2f, the dppc+ hydride
2j+ does not react with 11. More electron-rich hydrides such
as 2g and 2h are slowly oxidized by 11, but there is no evidence
of H- transfer.49 The Co(II) in 2i is immediately oxidized by
one electron in the presence of 11, cleanly giving 2j[BF4] and
the vicinal diamine 13 (eq 2). Cp2Co is also oxidized by 11,
giving [Cp2Co][BF4] and 13 (see Experimental Section). It is
known that vicinal diamines are available from the SmI2-
promoted reductive coupling of iminium cations.50
additional anodic waves, typically several hundred mV
positive of the initial oxidations. However, these waves are
irreversible and we cannot reliably assign them.
Wrighton and DuBois found that coordination of dppc to
Re- or Mo-carbonyl fragments causes an anodic shift of 100
to 400 mV in the Co(III/II) potential.12,14 A similar shift
occurs with dppf: the Fe(III/II) potentials of group 6 dppf
carbonyls are more positive than that of free dppf.34 Such
anodic shifts are expected upon coordination of dppc or dppf
to an electron-poor transition metal, as electron density is
removed from the ligand. We studied the reduction of 2j+
and 6j2+ (Table 5) in CH3CN, the same solvent used by
DuBois and Wrighton for most of their studies. The Co(III/
II) potential of the dppc+ ligand undergoes a surprisingly
large cathodic shift upon coordination in 2j+. Even in 6j2+
,
where Ru is cationic, the Co(III/II) potential is only 60 mV
more positive than that of free dppc. Clearly, the Cp*Ru-
(CH3CN)+ and Cp*RuH fragments are quite electron-rich.
Reactivity of the Hydrides with Iminium Cations. We have
previously shown that some CpRu(P-P)H can catalyze the
ionic hydrogenation of iminium cations.17 The rate-deter-
mining step of such hydrogenations is H- transfer from the
Ru hydride to the iminium cation. The Ru hydrides in this
work are generally poor hydride donors (Table 6) and there-
fore would not be suitable catalysts. The dppf complex 2a
was the most effective hydride donor, although only 18% of
this hydride reacted with 10 over the course of 3 h.
The isopropyl-substituted iminium cation 12 does not
oxidize Cp2Co, so we were confident that it would not
oxidize the Co(II) in 2i (see the potentials in Table 5). While
this prediction was correct, 2i did not transfer H- to 12 and
no reaction was observed (eq 3).
Electron-rich Ru hydride complexes are not necessarily
good hydride donors. Guan found an inverse correlation
between the ease of oxidation of CpRu(P-P)H (P-P =
dppm, dppe, dppp) and the rate at which these complexes
(47) dppb=1,4-bis(diphenylphosphino)butane.
(48) Butler, I. R.; Cullen, W. R.; Kim, T.-J.; Rettig, S. J.; Trotter, J.
Organometallics 1985, 4, 972–980.
(49) As expected from their potentials, Cp*Ru(dippf)H reacts more
quickly than Cp*Ru(dppomf)H. There is no evidence of hydride trans-
fer. Instead, the dihydride cations (a decomposition product of the
hydride radical cations) slowly appear over the course of several hours;
see ref 44.
(45) dppm=1,1-bis(diphenylphosphino)methane; dppe=1,2-bis(di-
phenylphosphino)ethane; dppp=1,3-bis(diphenylphosphino)propane.
(46) Guan, H.; Iimura, M.; Magee, M. P.; Norton, J. R.; Janak, K. E.
Organometallics 2003, 22, 4084–4089.
ꢀ
(50) Aurrecoechea, J. M.; Fernandez-Acebes, A. Tetrahedron Lett.
1992, 33, 4763–4766.

3810 Organometallics, Vol. 28, No. 13, 2009
Shaw et al.
dppomf methyl groups of 4g+). This argument has
previously been made for 4f+ and trans-(CpCO2H)Ru-
(dppf)(H)2+ 36,59
.
Surprisingly, the protonation of [Cp*Ru(dppc)H][PF6]
(2j[PF6]) gave a dihydrogen complex that was stable for
many hours at room temperature (Scheme 3). The [Cp*Ru-
(dppc)(H2)]2+ (3j2+) isomerized over the course of 48 h to
the trans-dihydride 4j2+. As with the other dihydride com-
1
plexes, the H NMR spectrum of 4j2+ has only two reso-
nances for the Cp hydrogens of the dppc+ ligand. In the
1H NMR spectrum of 3j2+ these hydrogens give four
resonances, a result of the lower symmetry of this H2 com-
plex. Generally, isomeric dihydrogen and dihydride com-
plexes may be distinguished by comparing the T1(min) of
their 1H NMR hydride resonances.60,61 The T1(min) for
Cp*Ru(dppf)(H2)+ is 11.5 ms (218.5 K, 300 MHz), while
that for 4f+ is much longer, 151 ms (195.2 K, 300 MHz).19
A similar difference distinguishes 3j2+ and 4j2+: the T1(min)
for the dihydrogen complex is 9.4 ms, while that for the
dihydride is 260 ms (Figure 4).
Figure 4. T1 of the hydride resonances of 3j2+ and 4j2+ at various
temperatures (300 MHz, CD2Cl2).
Scheme 4. Ru Dihydride Complexes with Fe-Based Ligands
The H2 in 3j2+ is readily replaced by nitriles. The proto-
nation of 2j+ with HBF4 OMe2 in CD3CN cleanly gives the
3
solvent complex [Cp*Ru(dppc)CD3CN]2+ (and presumably
H2, eq 5). The reaction of 3j2+ with excess t-BuCN in CD2Cl2
gives [Cp*Ru(dppc)t-BuCN]2+. The rate of this reaction
does not depend on [t-BuCN] and is therefore dissociative;
the half-life of 3j2+ is about 6 min (eq 6). Even though it
is quite labile, H2 is not lost in the absence of a replace-
ment ligand (recall the slow conversion of 3j2+ to 4j2+ in
Scheme 3).
Protonation of the Hydride Complexes. The protonation of
a transition metal hydride generally gives a dihydrogen
complex as the kinetic product.51 In the case of (Cp/Cp*)
Ru(L2)H, the resulting cationic dihydrogen complexes
usually isomerize, either partially or completely, to the
trans-dihydride isomers (eq 4).52-58 For example, the proto-
nation of 2f gives Cp*Ru(dppf)(H2)+, which rapidly iso-
merizes to trans-Cp*Ru(dppf)(H)+2 (4f+) above 0 °C.19
HBF4 OMe2
3
2 þ
2jþ ꢁꢁꢁꢁꢁꢁ ½Cp RuðdppcÞCD CNꢃ þ H2 ð5Þ
ꢂ
f
3
CD3CN
xs: t-BuCN; CD2Cl2
2 þ
3j2 þ
ꢁꢁꢁꢁꢁꢁꢁꢁꢁꢁꢁ
ꢂ
½Cp RuðdppcÞt-BuCNꢃ þ H2
f
Treating CD2Cl2 solutions of the hydrides with HBF4 OMe2
3
k ¼2:0ð1Þꢀ10-3 s-1 at 19:5 °C
(or acidic iminium cations in some cases, eq 1) at room
temperaturecleanlygivestrans-dihydridecations(Scheme4);
presumably, these result from the rapid isomerization of
intermediate dihydrogen complexes.
ð6Þ
The dihydride complex 4j2+ is much less acidic than its
dihydrogen isomer 3j2+. In CD2Cl2, 4j2+ is only partially
deprotonated by excess NEt3, while 3j2+ is quantitatively
deprotonated by 2,6-lutidine, indicating that the pKa values
of the two isomers differ by more than 7 units.62 Because the
deprotonations of isomeric M(H2)+ and M(H)2+ cations give
the same monohydride complex, their pKa values are related
by the [M(H)+2 ]/[M(H2)+] equilibrium constant (eqs 7, 8).58
The large difference in acidity between 3j2+ and 4j2+ is in
agreement with the fact that the equilibrium between the two
The 1H NMR spectra of 4+ establish the trans geometry of
the two hydride ligands. The high symmetry of this arrange-
ment results in only two resonances for the Cp hydrogens of
the bisphosphine ligands (and only two resonances for the
(51) Papish, E. T.; Magee, M. P.; Norton, J. R. In Recent Advances in
Hydride Chemistry; Poli, R., Ed.; Elsevier: New York, 2001; pp 39-74.
(52) Chinn, M. S.; Heinekey, D. M. J. Am. Chem. Soc. 1990, 112,
5166–5175.
(53) Law, J. K.; Mellows, H.; Heinekey, D. M. J. Am. Chem. Soc.
2002, 124, 1024–1030.
lies heavily in favor of 4j2+
.
(54) Belkova, N. V.; Dub, P. A.; Baya, M.; Houghton, J. Inorg. Chim.
Acta 2007, 360, 149–162.
(59) Hembre, R. T.; McQueen, J. S. Angew. Chem., Int. Ed. Engl.
1997, 36, 65–67.
ꢀ
(55) Jimenez-Tenorio, M.; Puerta, M. C.; Valerga, P. J. Organomet.
Chem. 2000, 609, 161–168.
(60) Luo, X. L.; Crabtree, R. H. Inorg. Chem. 1990, 29, 2788–2791.
(61) Morris, R. H. Coord. Chem. Rev. 2008, 252, 2381–2394.
(62) Our estimate for the pKa of 4j2+ is 21. In CH3CN, the pKa of
Et3NH+ is 18.82, while that of 2,6-lutidinium is 14.13: Kaljurand, I.;
ꢀ
(56) Jimenez-Tenorio, M.; Palacios, M. D.; Puerta, M. C.; Valerga, P.
Inorg. Chem. 2007, 46, 1001–1012.
(57) Conroy-Lewis, F. M.; Simpson, S. J. J. Chem. Soc., Chem.
Commun. 1987, 1675–1676.
(58) Jia, G.; Morris, R. H. J. Am. Chem. Soc. 1991, 113, 875–883.
::
::
::
Kutt, A.; Soovali, L.; Rodima, T.; Maemets, V.; Leito, I.; Koppel, I. A.
J. Org. Chem. 2005, 70, 1019–1028.

Article
Organometallics, Vol. 28, No. 13, 2009 3811
Keq
compounds). Experiments on the PAR 263A used a glassy
carbon working electrode, an Ag/AgCl reference, and a plati-
num wire auxiliary electrode. Experiments on the BAS CV-50W
used a platinum working electrode, an Ag/Ag+ reference (Ag
wire, 0.01 M AgNO3 + 0.10 M [Bu4N]PF6 in CH3CN), and a
platinum wire auxiliary electrode. The supporting electrolyte for
the sample solutions was 0.10 M [Bu4N]PF6 in CH3CN (2j[PF6],
6j[PF6]2) or CH2Cl2 (all other compounds). All samples were
prepared under an N2/Ar atmosphere and further purged with
Ar before measurement. Analyte concentrations were 0.001 M.
All potentials are reported in volts (V) vs FcH+/FcH.
MðH2Þþ h MðHÞþ2
ð7Þ
ð8Þ
pKeq ¼ pKaðMðH2ÞþÞ ꢁ pKaðMðHÞ þÞ
2
Conclusion
These complexes extend the series (Cp/Cp*)Ru(P-P)L+
and (Cp/Cp*)Ru(P-P)X. As shown by the X-ray structures,
the coordination of dppc+ in 6j2+ and 7j+ is similar to that
of dppf in related complexes. Of the complexes examined by
CV, all but the dppc ones are initially oxidized at Ru. (This is
in contrast to Pd(P-P)Cl2 (P-P=dppf or variants), which
undergo initial oxidation at the ligand.31) Ferrocenyl bispho-
sphines stabilize Ru(III) radical cations, and the Ru(III/II)
potentials reflect the donor ability of the neutral bispho-
sphine ligands. Steric factors aside, the reactivity of the Ru
hydrides toward iminium cations is correlated with the Ru-
(III/II) potentials. The easily oxidized Ru hydrides in this
study are more likely to serve as bases or single-electron
reductants than as hydride sources. Upon protonation at
room temperature, most of the Ru hydrides give trans-
dihydride cations, presumably via cationic dihydrogen
complexes. The dihydrogen complex [Cp*Ru(dppc)(H2)]2+
(3j2+) has considerably more kinetic stability than its ferro-
cenyl analogue. The H2 ligand in 3j2+ is readily replaced by
nitriles; the reaction with excess t-BuCN occurs by a dis-
sociative mechanism. The difference in acidity between 3j2+
and its dihydride isomer 4j2+ reflects the fact that 4j2+ is
more stable thermodynamically.
CpRu(dppf)H (2a). A solution of CpRu(dppf)Cl (1a, 0.056 g,
0.074 mmol) and NaBH4 (0.025 g, 0.661 mmol) in 10 mL of
EtOH was stirred at room temperature for 4 h. Additional
NaBH4 (0.010 g, 0.264 mmol) was added, and the solution
was stirred for another 2 h. The yellow product was collected
by filtration, washed with EtOH, and dried under vacuum
(0.032 g, 0.044 mmol, 60% yield).
trans-[CpRu(dppf)(H)2]+ (4a+) was prepared in situ by the
protonation of CpRu(dppf)H (2a) with HBF4 OMe2 in CD2Cl2
3
at room temperature. 1H NMR (400 MHz, CD2Cl2): δ -7.63 (t,
Ru(H)2, JP-H=23.0 Hz, 2H), 4.37 (s, br, dppf Cp, 4H), 4.43 (s,
br, dppf Cp, 4H), 4.84 (s, RuCp, 5H), 7.56 (m, Ar, 12H), 7.74 (m,
Ar, 8H). 31P{1H} NMR (161.9 MHz, CD2Cl2): δ 59.93. T1 of the
dihydride resonance (400 MHz, CD2Cl2, ambient temperature):
0.721 s.
CpRu(dppr)Cl (1b). A solution of CpRu(PPh3)2Cl (0.116 g,
0.159 mmol) and dppr (0.113 g, 0.189 mmol) in 25 mL of
benzene was refluxed for 36 h. Evaporation of the solvent gave
an orange powder that was washed with warm hexanes (5 mL ꢀ 3)
and dried under vacuum (0.107 g, 0.134 mmol, 84% yield). 1H
NMR (400 MHz, CDCl3): δ 3.95 (s, RuCp, 5H), 4.43 (m, dppr
Cp, 8H), 7.25-7.43 (m, Ar, 20H). 31P{1H} NMR (161.9 MHz,
CDCl3): δ 42.7. Anal. Calcd for C39H33ClP2Ru2: C, 58.46; H,
4.15. Found: C, 58.34; H, 4.35.
CpRu(dppr)H (2b). A solution of 1b (0.160 g, 0.199 mmol) and
NaBH4 (0.076 g, 2.01 mmol) in 25 mL of EtOH was stirred at
room temperature for 4 h. Additional NaBH4 (0.040 g, 1.06
mmol) was added and the solution stirred for another 2 h. The
yellow product was collected by filtration, washed with EtOH,
and dried under vacuum (0.128 g, 0.166 mmol, 84% yield). 1H
NMR (400 MHz, C6D6): δ -11.96 (t, RuH, JP-H=35.2 Hz, 1H),
3.98 (s, RuCp, 5H), 4.45-4.77 (m, dppr Cp, 8H), 7.27-7.40 (m,
Ar, 10H), 7.65-7.83 (m, Ar, 10H). 31P{1H} NMR (161.9 MHz,
C6D6): δ 63.2. FAB+ MS, m-nitrobenzyl alcohol matrix (m-
NBA): m/z 766.97 [M - 1]+. The spectrum agreed with the
calculated isotopic distribution.
Experimental Section
General Procedures. All air-sensitive compounds were pre-
pared and handled under an N2/Ar atmosphere using standard
Schlenk and inert-atmosphere box techniques. CDCl3, CD2Cl2,
and CD3CN were degassed and stored over 3 A molecular
sieves. THF-d8, toluene-d8, and C6D6 were dried over sodium/
benzophenone and purified by vacuum transfer. Hexanes, Et2O,
and CH2Cl2 were dried and deoxygenated by two successive
columns (activated alumina, Q5 copper catalyst). Anhydrous
CH3CN (Sigma-Aldrich) was stored over 3 A molecular sieves.
Benzene and THF were distilled from sodium/benzophenone
under an N2 atmosphere. Sodium cyclopentadienide,63 dppr,64
dppo,65 dcpf,66 CpRu(PPh3)2Cl,67 Cp*Ru(PPh3)2Cl,68 CpRu
(dppf)Cl (1a),21 CpRu(dppf)H (2a),21 Cp*Ru(dppf)Cl (1f),22
Cp*Ru(dppf)H (2f),22 trans-[Cp*Ru(dppf)(H)2]+ (4f+),19 and
the iminium salts 9-1269 were prepared by literature methods.
Electrochemical Procedures. Cyclic voltammetry was per-
formed with a PAR 263A potentiostat/galvanostat (1b, 2b,
1c, 2c, 1e, 2e) or a BAS CV-50W potentiostat (all other
CpRu(dppo)Cl (1c). A solution of CpRu(PPh3)2Cl (0.165 g,
0.227 mmol) and dppo (0.156 g, 0.227 mmol) in 25 mL of
benzene was refluxed for 24 h. Evaporation of the solvent gave
an orange powder, which was washed with warm hexanes (5 mL
ꢀ 3) and dried under vacuum (0.128 g, 0.144 mmol, 63%). 1H
NMR (400 MHz, CDCl3): δ 3.95 (s, RuCp, 5H), 4.71 (br, s, dppo
Cp, 2H), 4.86 (br, s, dppo Cp, 2H), 5.01 (br, s, dppo Cp, 2H),
5.56 (br, s, dppo Cp, 2H), 7.22-7.39 (m, Ar, 20H). 31P{1H}
NMR (161.9 MHz, CDCl3):
δ 46.6. Anal. Calcd for
C39H33ClOsP2Ru: C, 52.61; H, 3.74. Found: C, 52.41; H, 3.36.
CpRu(dppo)H (2c). A solution of 1c (0.113 g, 0.127 mmol) and
NaBH4 (0.050 g, 1.32 mmol) in 30 mL of 1-butanol/EtOH (1:1)
was stirred at room temperature for 6 h. Additional NaBH4
(0.050 g, 1.32 mmol) was added and the solution stirred for
another 18 h. The yellow product was collected by filtration,
washed with EtOH, and dried under vacuum (0.075 g, 0.088
(63) Panda, T. K.; Gamer, M. T.; Roesky, P. W. Organometallics
2003, 22, 877–878.
(64) Li, S.; Wei, B.; Low, P. M. N.; Lee, H. K.; Hor, T. S. A.; Xue, F.;
Mak, T. C. W. J. Chem. Soc., Dalton Trans. 1997, 1289–1293.
(65) Gusev, O. V.; Kalsin, A. M.; Petrovskii, P. V.; Lyssenko, K. A.;
Oprunenko, Y. F.; Bianchini, C.; Meli, A.; Oberhauser, W. Organome-
tallics 2003, 22, 913–915.
(66) Kim, T. J.; Kim, Y. H.; Kim, H. S.; Shim, S. C.; Kwak, Y. W.;
Cha, J. S.; Lee, H. S.; Uhm, J. K.; Byun, S. I. Bull. Korean Chem. Soc.
1992, 13, 588–592.
(67) Bruce, M. I.; Windsor, N. J. Aust. J. Chem. 1977, 30, 1601–1604.
(68) Morandini, F.; Dondana, A.; Munari, I.; Pilloni, G.; Consiglio,
G.; Sironi, A.; Moret, M. Inorg. Chim. Acta 1998, 282, 163–172.
(69) Leonard, N. J.; Paukstelis, J. V. J. Org. Chem. 1963, 28,
3021–3024.
1
mmol, 69% yield). H NMR (400 MHz, C6D6): δ -12.08 (t,
RuH, JP-H=35.9 Hz, 1H), 3.97 (s, RuCp, 5H), 4.72 (br, s, dppo
Cp, 2H), 4.76 (br, s, dppo Cp, 2H), 4.82 (br, s, dppo Cp, 2H),
4.88 (br, s, dppo Cp, 2H), 7.30 (m, Ar, 10H), 7.62 (m, Ar, 5H),
7.78 (m, Ar, 5H). 31P{1H} NMR (161.9 MHz, C6D6): δ 66.9.
FAB+ MS (m-NBA): m/z 855.00 [M - 1]+. The spectrum
agreed with the calculated isotopic distribution.

3812 Organometallics, Vol. 28, No. 13, 2009
Shaw et al.
CpRu(dippf)Cl (1d). A solution of CpRu(PPh3)2Cl (0.45 g,
0.62 mmol) and dippf (0.27 g, 0.64 mmol) in 30 mL of benzene
was refluxed for 5 h. The reaction mixture was concentrated in
vacuo, and 20 mL of hexanes was added. Cooling this mixture in
an EtOH/CO2 bath gave an orange precipitate, which was
filtered cold, washed with hexanes (5 mL ꢀ 2) at room tempera-
ture, and dried under vacuum (0.26 g, 0.42 mmol, 67% yield). 1H
NMR (300 MHz, CD2Cl2): δ 1.07-1.39 (m, dippf CH3, 24H),
2.42-2.56 (m, dippf CH, 2H), 2.64-2.78 (m, dippf CH, 2H),
4.09 (s, dippf Cp, 2H), 4.21 (s, dippf Cp, 4H), 4.67 (s, RuCp, 5H),
4.90 (s, dippf Cp, 2H). 31P{1H} NMR (121.5 MHz, CD2Cl2):
δ 52.46. Anal. Calcd for C27H41ClFeP2Ru: C, 52.31; H, 6.67.
Found: C, 52.49; H, 6.72.
Lithium tetramethylcyclopentadienide (2.85 g, 22.24 mmol) was
dissolved in 80 mL of dry THF and cooled in a water-ice bath. To
this solution was added dropwise Ph2PCl (4.0 mL, 22.24 mmol),
the ice bath removed, and the resulting mixture stirred for 70 min.
The mixture was cooled again in a water-ice bath, and n-BuLi
(13.9 mL, 1.6 M in hexanes) was added dropwise. The ice bath was
removed, and the mixture was stirred for 45 min. The mixture was
cooled again in a water-ice bath, and a slurry of FeCl2 (1.41 g,
11.12 mmol) in 40 mL of THF was added via cannula. The ice bath
was removed, and the mixture was stirred for 45 min, then added to
a separatory funnel containing 100 mL of Et2O and 300 mL of
water. The fractions were separated, and the aqueous fraction was
washed with 50 mL of Et2O. The organic fractions were combined,
washed with 100 mL of water, and dried over MgSO4. Evaporation
of two-thirds of the solvent gave an orange precipitate, which was
filtered and dried under vacuum (3.30 g, 4.95 mmol, 45% yield
based on FeCl2). 1H NMR (400 MHz, CD2Cl2): δ 1.32 (s, CH3,
12H), 1.77 (s, CH3, 12H), 7.20-7.30 (m, Ar, 12H), 7.34-7.41
(m, Ar, 8H). 31P{1H} NMR (161.9 MHz, CD2Cl2): δ -20.07.
FAB+ MS (m-NBA): m/z 666.15 [M]+. The spectrum agreed with
the calculated isotopic distribution.
Cp*Ru(dppomf)Cl (1g). A solution of Cp*Ru(1,5-COD)Cl
(0.38 g, 1.00 mmol) and dppomf (0.67 g, 1.00 mmol) in 60 mL of
EtOH was refluxed for 90 min. The reaction mixture was cooled
to room temperature, and the orange-red precipitate was fil-
tered, washed with EtOH and hexanes, and dried under vacuum
(0.63 g, 0.67 mmol, 67% yield). 1H NMR (400 MHz, CD2Cl2):
δ 0.81 (s, dppomf CH3, 6H), 1.07 (s, dppomf CH3, 6H), 1.15 (s,
RuCp*, 15H), 1.36 (s, dppomf CH3, 6H), 1.41 (s, dppomf CH3,
6H), 7.17-7.22 (m, Ar, 2H), 7.35 (br, Ar, 8H), 7.42-7.55 (m, Ar,
4H), 7.82 (br, Ar, 4H), 8.67-8.74 (m, Ar, 2H). 31P{1H} NMR
(161.9 MHz, CD2Cl2): δ 49.86. Anal. Calcd for C52H59ClFe-
P2Ru: C, 66.56; H, 6.34. Found: C, 66.32; H, 6.36.
CpRu(dippf)H (2d). A solution of 1d (0.23 g, 0.37 mmol) and
NaOMe (0.23 g, 4.26 mmol) in 25 mL of MeOH was refluxed for
90 min. The reaction mixture was cooled to room temperature,
and the yellow precipitate was filtered, washed with MeOH
(10 and 5 mL), and dried under vacuum (0.20 g, 0.33 mmol, 90%
yield). The product may also be prepared by stirring 1d with
1
excess NaBH4 in EtOH. H NMR (300 MHz, CD2Cl2): δ -
13.44 (t, RuH, JP-H=37.2 Hz, 1H), 0.92-1.30 (m, dippf CH3,
24H), 1.80-1.92 (m, dippf CH, 2H), 2.03-2.17 (m, dippf CH,
2H), 4.16 (m, dippf Cp, 6H), 4.32 (s, dippf Cp, 2H), 4.71
(s, RuCp, 5H). 31P{1H} NMR (121.5 MHz, CD2Cl2): AB
pattern, δ 77.72 (Δν=0.20 ppm, JP-P =9.8 Hz). FAB+ MS
(m-NBA): m/z 584.98 [M - 1]+. The spectrum agreed with the
calculated isotopic distribution.
trans-[CpRu(dippf)(H)2]+ (4d+) was prepared in situ by the
protonation of 2d with HBF4 OMe2 in CD2Cl2 at room tem-
3
1
perature. H NMR (300 MHz, CD2Cl2): δ -9.12 (t, Ru(H)2,
J
P-H=22.1 Hz, 2H), 1.15-1.32 (m, dippf CH3, 24H), 2.16-2.31
(m, dippf CH, 4H), 4.31 (s, dippf Cp, 4H), 4.49 (s, dippf Cp, 4H),
5.46 (s, RuCp, 5H). 31P{1H} NMR (121.5 MHz, CD2Cl2):
δ 79.17. T1 of the dihydride resonance (400 MHz, CD2Cl2,
ambient temperature): 0.895 s.
CpRu(dcpf)Cl (1e). A solution of CpRu(PPh3)2Cl (0.381 g,
0.525 mmol) and dcpf (0.306 g, 0.529 mmol) in 50 mL of benzene
was refluxed for 24 h. Evaporation of the solvent gave an orange
solid, which was washed with EtOH (5 mL ꢀ 3) and dried under
vacuum (0.142 g, 0.182 mmol, 35% yield). 1H NMR (400 MHz,
CDCl3): δ 1.08-2.12 (m, Cy, 44H), 4.04 (m, dcpf Cp, 8H), 4.62
(s, RuCp, 5H). 31P{1H} NMR (161.9 MHz, CDCl3): δ 47.6.
Anal. Calcd for C39H57ClFeP2Ru: C, 60.04; H, 7.36. Found:
C, 60.19; H, 7.18.
Cp*Ru(dppomf)H (2g). A solution of 1g (0.18 g, 0.19 mmol)
and NaOMe (0.18 g, 3.33 mmol) in 15 mL of MeOH was
refluxed for 45 min. The reaction mixture was cooled to room
temperature, and the yellow precipitate was filtered, washed
with MeOH (5 mL ꢀ 2), and dried under vacuum (0.16 g, 0.17
1
mmol, 90% yield). H NMR (300 MHz, CD2Cl2): δ -13.02
(t, RuH, JP-H=36.9 Hz, 1H), 0.88 (s, dppomf CH3, 6H), 1.06
(s, dppomf CH3, 6H), 1.33 (s, dppomf CH3, 6H), 1.40 (s, RuCp*,
15H and dppomf CH3, 6H), 7.25-7.40 (m, Ar, 12H), 8.09 (m,
Ar, 4H), 8.29 (m, Ar, 4H). 31P{1H} NMR (121.5 MHz, CD2Cl2):
δ 67.69 (m). FAB+MS (m-NBA): m/z 903.9 [M - 1]+. The
spectrum agreed with the calculated isotopic distribution.
trans-[Cp*Ru(dppomf)(H)2]+ (4g+) was prepared in situ by
CpRu(dcpf)H (2e). A solution of 1e (0.136 g, 0.175 mmol) and
NaBH4 (0.066 g, 1.75 mmol) in 25 mL of EtOH was stirred
at room temperature for 4 h. Additional NaBH4 (0.01 g,
0.26 mmol) was added and the solution stirred for another
2 h. The yellow product was collected by filtration, washed with
the protonation of 2g with HBF4 OMe2 in CD2Cl2 at room
3
temperature. 1H NMR (300 MHz, CD2Cl2): δ -7.30 (t, Ru
(H)2, JP-H =27.5 Hz, 2H), 1.29 (s, dppomf CH3, 12H), 1.42
(s, RuCp*, 15H), 1.51 (s, dppomf CH3, 12H), 7.45-7.56 (m, Ar,
12H), 7.80-7.89 (m, Ar, 8H). 31P{1H} NMR (121.5 MHz,
CD2Cl2): δ 63.00 (m).
EtOH, and dried under vacuum (0.111 g, 0.149 mmol, 85%
=
1
yield). H NMR (400 MHz, C6D6): δ -13.56 (t, RuH, JP-H
37.4 Hz, 1H), 1.03-1.37 (m, Cy, 22H), 1.60-2.13 (m, Cy, 22H),
4.10 (m, dcpf Cp, 8H), 4.66 (s, RuCp, 5H). 31P{1H} NMR (161.9
MHz, C6D6): δ 69.6. FAB+ MS (m-NBA): m/z 745.18 [M - 1]+.
The spectrum agreed with the calculated isotopic distribution.
trans-[CpRu(dcpf)(H)2]+ (4e+) was prepared in situ by the
Cp*Ru(dippf)Cl (1h). A solution of Cp*Ru(1,5-COD)Cl (0.31
g, 0.82 mmol) and dippf (0.34 g, 0.82 mmol) in 40 mL of EtOH
was refluxed for 1 h. The reaction mixture was concentrated in
vacuo until a red precipitate formed. The precipitate was
filtered, washed with EtOH (10 mL) and hexanes (5 mL), and
dried under vacuum (0.28 g, 0.41 mmol, 50% yield). 1H NMR
(400 MHz, CD2Cl2): δ 0.88 (br, dippf CH3, 6H), 1.16-1.26
(m, dippf CH3, 12H), 1.34 (br, dippf CH3, 6H), 1.62 (s, RuCp*,
15H), 2.32-2.53 (br, dippf CH, 4H), 4.02-4.20 (br, dippf Cp,
6H), 4.91 (s, dippf Cp, 2H). 31P{1H} NMR (161.9 MHz,
CD2Cl2): δ 44.77. Anal. Calcd for C32H51ClFeP2Ru: C, 55.70;
H, 7.45. Found: C, 55.90; H, 7.57.
protonation of 2e with HBF4 OMe2 in CD2Cl2 at room tem-
3
1
perature. H NMR (300 MHz, CD2Cl2): δ -9.15 (t, Ru(H)2,
J
P-H = 22.5 Hz, 2H), 1.30 (m, Cy, 20H), 1.79 (m, Cy, 24H),
4.24 (s, dcpf Cp, 4H), 4.47 (s, dcpf Cp, 4H), 5.40 (s, RuCp, 5H).
31P{1H} NMR (121.5 MHz, CD2Cl2): 69.0. T1 of the dihydride
resonance (300 MHz, CD2Cl2, ambient temperature): 0.589 s.
1,10-Bis(diphenylphosphino)octamethylferrocene (dppomf). The
following method was adapted from published procedures.70-72
Cp*Ru(dippf)H (2h). A solution of 1h (0.20 g, 0.29 mmol) and
NaOMe (0.20 g, 3.70 mmol) in 20 mL of MeOH was refluxed
for 30 min. The reaction mixture was cooled to room tempera-
ture, and the yellow precipitate was filtered, washed with
MeOH (3 mL ꢀ 2), and dried under vacuum (0.18 g, 0.28 mmol,
(70) Ninoreille, S.; Broussier, R.; Amardeil, R.; Kubicki, M. M.;
Gautheron, B. Bull. Soc. Chim. Fr. 1995, 132, 128–138.
(71) Trouve, G.; Broussier, R.; Gautheron, B.; Kubicki, M. M. Acta
Crystallogr. 1991, C47, 1966–1967.
(72) Szymoniak, J.; Dormond, J.; Moıse, A. J. Org. Chem. 1990, 55,
1429–1432.
¨
96% yield). 1H NMR (300 MHz, CD2Cl2):
δ -14.37

Article
Organometallics, Vol. 28, No. 13, 2009 3813
(t, RuH, JP-H=36.8 Hz, 1H), 0.91-1.19 (m, dippf CH3, 18H),
1.23-1.37 (m, dippf CH3, 6H), 1.87 (br, RuCp*, 15H and dippf
CH, 2H), 2.08 (br, dippf CH, 2H), 4.12 (s, dippf Cp, 6H), 4.23 (s,
dippf Cp, 2H). 31P{1H} NMR (121.5 MHz, CD2Cl2): δ 70.31.
FAB+ MS (m-NBA): m/z 655.95 [M - 1]+. The spectrum
agreed with the calculated isotopic distribution.
solutions. However, the resulting crystals contain uncoordi-
nated Et2O and CH3CN and must be dried under vacuum for
an extended period to remove these solvents.
[Cp*Ru(dppc)H][PF6] (2j[PF6]). A solution of 6j[PF6]2 (0.112 g,
0.10 mmol) and NEt3 (0.28 mL, 2.0 mmol) was stirred in 10 mL of
MeOH under 80 psi of H2 for 24 h. The solvent was evaporated, and
the residue was dissolved in CH2Cl2 and loaded on a silica column
(230-400 mesh, 6 cm ꢀ 2 cm diameter). The product was eluted
with a 1:1 mixture of Et2O/CH2Cl2. The solvent was evaporated,
and the residue was dissolved in a small amount of CH3CN. The
addition of Et2O gave a brown precipitate that was filtered, washed
with 5 mL of Et2O, and dried under vacuum (0.040 g, 0.043 mmol,
43%). On a larger scale, 6j[PF6]2 (0.90 g, 0.80 mmol) and NEt3
(2.3 mL, 16.5 mmol) were stirred in 15 mL of MeOH under 80 psi of
H2 for 16 h. The solvent was evaporated, and the residue was
dissolved in CH2Cl2 and passed through two successive silica
columns (230-400 mesh, 12 cm ꢀ 2 cm diameter, 1:1 Et2O/
CH2Cl2). Precipitation, washing, and drying of the product (as
described above) gave a brown solid (0.33 g, 0.35 mmol, 44%). 1H
NMR (300 MHz, CD3CN): δ-12.28 (t, RuH, JP-H=34.7 Hz, 1H),
1.21 (s, RuCp*, 15H), 5.18 (s, dppc+ Cp, 2H), 5.28 (s, dppc+ Cp,
2H), 5.44 (s, dppc+ Cp, 2H), 5.91 (s, dppc+ Cp, 2H), 7.43 (m, Ar,
6H), 7.58 (m, Ar, 6H), 7.73 (br, Ar, 4H), 7.88 (br, Ar, 4H). 31P{1H}
trans-[Cp*Ru(dippf)(H)2]+ (4h+) was prepared in situ by the
protonation of 2h with HBF4 OMe2 in CD2Cl2 at room tem-
3
1
perature. H NMR (300 MHz, CD2Cl2): δ -8.84 (t, Ru(H)2,
J
P-H = 26.7 Hz, 2H), 1.09-1.33 (m, dippf CH3, 24H), 1.97
(s, RuCp*, 15H), 2.15-2.29 (m, dippf CH, 4H), 4.21 (s, dippf
Cp, 4H), 4.44 (s, dippf Cp, 4H). 31P{1H} NMR (121.5 MHz,
CD2Cl2): δ 69.86.
1,10-Bis(diphenylphosphino)cobaltocene (dppc). The following
method was adapted from published procedures.12,73 Sodium
cyclopentadienide (1.96 g, 22.24 mmol) was dissolved in 80 mL
of dry THF and cooled in an EtOH/CO2 bath. To this solution
was added dropwise Ph2PCl (4.0 mL, 22.24 mmol), the cold bath
was removed, and the resulting mixture was stirred for 70 min.
The mixture was cooled again in an EtOH/CO2 bath, and
n-BuLi (13.9 mL, 1.6 M in hexanes) was added dropwise. The
cold bath was removed, and the mixture was stirred for 45 min.
The mixture was cooled again in an EtOH/CO2 bath, and a
slurry of CoCl2 (1.44 g, 11.12 mmol) in 20 mL of THF was added
via cannula. The cold bath was removed, and the mixture was
stirred for 15 h, after which the solvent was evaporated to give a
brown residue. The residue was extracted with Et2O (150 mL
and 3 ꢀ 100 mL), and the resulting brown solution was
concentrated in vacuo to a volume of 20 mL. Addition of
50 mL of hexanes and evaporation of the solvent gave the
product, a dark brown powder (4.39 g, 7.87 mmol, 71% yield
based on CoCl2). 1H NMR (300 MHz, THF-d8): δ -54.8
(br, Cp, 4H), -22.7 (br, Cp, 4H), 6.38 (br, Ar, 8H), 7.38
(t, Ar, 4H), 8.09 (br, Ar, 8H). 31P{1H} NMR (121.5 MHz,
THF-d8): δ -158 (br). FAB+ MS (m-NBA): m/z 556.86 [M]+.
The spectrum agreed with the calculated isotopic distribution.
Cp*Ru(dppc)Cl (1i). A solution of Cp*Ru(PPh3)2Cl (0.50 g,
0.63 mmol) and dppc (0.35 g, 0.63 mmol) in 20 mL of benzene
was refluxed for 1 h. The solvent was evaporated, and the
resulting brown solid was washed with hexanes twice and dried
under vacuum (0.39 g, 0.47 mmol, 75% yield). FAB+MS
(m-NBA): m/z 829.21 [M]+, 794.20 [Cp*Ru(dppc)]+. The spec-
trum agreed with the calculated isotopic distribution.
NMR (121.5 MHz, CD3CN): δ -144.46 (septet, PF-6 , JF-P
=
706 Hz), 61.78 (s, dppc+). FAB+MS (m-NBA): m/z 795.32 [Cp*Ru
(dppc)H]+. The spectrum agreed with the calculated isotopic
distribution.
[Cp*Ru(dppc)(H2)]2+ (3j2+). The title dihydrogen complex
was prepared in situ by the protonation of 2j[PF6] with HBF4
OMe2 in CD2Cl2 at room temperature. H NMR (300 MHz,
3
1
RT, CD2Cl2): δ -7.32 (s, br, Ru(H2), 2H), 1.36 (s, RuCp*, 15H),
5.63 (s, dppc+ Cp, 2H), 5.67 (s, dppc+ Cp, 2H), 5.81 (s, dppc+
Cp, 2H), 6.48 (s, dppc+ Cp, 2H), 7.35-7.44 (m, Ar, 4H), 7.54
(m, Ar, 10H), 7.68-7.85 (m, Ar, 6H). 31P{1H} NMR (121.5
MHz, RT, CD2Cl2): δ 51.98 (s, dppc+). T1 of the dihydrogen
resonance (300 MHz in CD2Cl2): 16.3(4) ms, 187.5 K; 10.9(2)
ms, 207.5 K; 9.4(2) ms, 228.4 K; 10.2(1) ms, 248.4 K; 12.0(1) ms,
269.4 K.
trans-[Cp*Ru(dppc)(H)2]2+ (4j2+) was prepared by leaving a
solution of 3j2+ at room temperature for 48 h; 1H and 31P{1H}
NMR indicated nearly complete conversion to 4j2+. 1H NMR
(300 MHz, RT, CD2Cl2): δ -7.52 (t, Ru(H)2, JP-H=26.2 Hz,
2H), 1.31 (s, RuCp*, 15H), 5.73 (s, dppc+ Cp, 4H), 5.75 (s,
dppc+ Cp, 4H), 7.67-7.79 (m, Ar, 12H), 7.87-7.97 (m, Ar, 8H).
31P{1H} NMR (121.5 MHz, RT, CD2Cl2): δ 56.81 (s, dppc+).
T1 of the dihydride resonance (300 MHz in CD2Cl2): 0.49(2) s,
187.5 K; 0.301(3) s, 207.5 K; 0.260(1) s, 228.4 K; 0.267(4) s, 248.4
K; 0.306(4) s, 269.4 K; 0.410(6) s, 297.6 K.
[Cp*Ru(dppc)CH3CN][PF6]2 (6j[PF6]2). A solution of 1i
(0.25 g, 0.30 mmol) and NH4PF6 (0.49 g, 3.0 mmol) in 40 mL
of CH3CN, stirred rapidly under air, became dark green over the
course of 10 min. This solution was refluxed under air for 30 min
and became red. The red solution was passed through a short
silica column (230-400 mesh, 10 cm ꢀ 2 cm diameter), and then
the solvent was removed under vacuum. Extraction of the
resulting residue with CH2Cl2 (20 and 5 mL) gave a translucent
red solution. Concentration in vacuo and addition of hexanes
gave a brick red precipitate, which was filtered, washed with a
2:1 mixture of hexanes/CH2Cl2 (2 ꢀ 15 mL) and hexanes (5 mL),
and dried under vacuum (0.15 g, 0.13 mmol, 44% yield). On
a larger scale, starting with 4.58 mmol of Cp*Ru(PPh3)2Cl, the
Cp*Ru(dppc)H (2i). Cp2Co (0.02-0.03 mmol) and 2j[PF6]
(ca. 0.01 mmol) were added to 500 μL of CH3CN; vigorous
shaking gave a dark precipitate. The solvent was carefully
decanted from the product, which was then washed with
CH3CN (2 ꢀ 300 μL) and dried under Ar. This dark solid, the
paramagnetic 2i, readily dissolves in toluene to give a red
1
solution. H NMR (400 MHz, toluene-d8): δ -57.8 (br, dppc
Cp, 2H), -52.2 (br, dppc Cp, 2H), -27.3 (br, dppc Cp, 2H),
-24.5 (br, dppc Cp, 2H), -15.76 (br, RuH, 1H), 2.63 (s, RuCp*,
15H), 6.16 (m, Ar, 2H), 6.76 (br, Ar, 4H), 6.84 (m, Ar, 2H), 7.56
(br, Ar, 4H), 7.70 (br, Ar, 4H), 7.95 (br, Ar, 4H); a small amount
of residual Cp2Co (δ -50.4) was present. Solutions of 2i are
conveniently generated by reducing 2j[PF6] with Cp2Co in THF-
d8; [Cp2Co][PF6] (yellow) precipitates, while 2i (red) is soluble.
1H NMR (300 MHz, THF-d8): δ -54.9 (br, dppc Cp), -45.4
(br, dppc Cp), -28.0 (br, dppc Cp), -23.0 (br, dppc Cp), -15.56
(br, RuH, 1H), 2.29 (s, RuCp*, 15H), 6.47 (m, Ar, 2H), 6.66
(br, Ar, 4H), 7.01 (m, Ar, 2H), 7.75 (m, Ar, 12H).
1
title compound was prepared in 43% overall yield. H NMR
(400 MHz, CD3CN): δ 1.08 (s, RuCp*, 15H), 2.89 (s, CH3CN),
5.53 (s, dppc+ Cp, 2H), 5.63 (s, dppc+ Cp, 2H), 5.66 (s, dppc+
Cp, 2H), 5.70 (s, dppc+ Cp, 2H), 7.40-7.50 (br, Ar, 4H), 7.50-
7.66 (m, Ar, 14H), 7.71-7.79 (m, Ar, 2H). The coordinated
CH3CN exchanges with CD3CN over the course of several
hours at room temperature. 31P{1H} NMR (121.5 MHz,
CD3CN): δ -144.46 (septet, PF-6 , JF-P =706 Hz), 40.67 (s,
dppc+). Anal. Calcd for C46H46NCoF12P4Ru: C, 49.12; H, 4.12;
N, 1.25. Found: C, 48.86; H, 4.20; N, 1.32. The product is readily
crystallized by the addition of Et2O to concentrated CH3CN
[Cp*Ru(dppc)CCPh][PF6] (7j[PF6]). A rapidly stirred slurry of
6j[PF6]2 (0.113 g, 0.100 mmol), NEt3 (3.5 mL), phenylacetylene
(0.5 mL), and toluene (5.0 mL) was heated to 105 °C for 2 h.
The volatiles were evaporated and the residue, dissolved in a
(73) Brasse, C. C.; Englert, U.; Salzer, A.; Waffenschmidt, H.;
Wasserscheid, P. Organometallics 2000, 19, 3818–3823.

3814 Organometallics, Vol. 28, No. 13, 2009
Shaw et al.
small amount of CH2Cl2, was loaded on a silica column
(230-400 mesh, 4 cm ꢀ 1 cm diameter). A green band was
spread to the bottom with CH2Cl2 and then eluted with 99:1
CH2Cl2/CH3CN. Evaporation of the solvent gave a green
residue. The addition of Et2O to a concentrated CH2Cl2 solution
gave a blue precipitate that became green once dried under
vacuum (0.077 g, 0.074 mmol, 74% yield). 1H NMR (300 MHz,
CDCl3): δ 1.20 (s, RuCp*, 15H), 5.45 (s, dppc+ Cp, 2H), 5.56 (s,
dppc+ Cp, 4H), 6.55 (s, dppc+ Cp, 2H), 7.20 (m, Ar, 1H), 7.32-
7.63 (m, Ar, 18H), 7.59 (m, Ar, 2H), 8.02 (m, Ar, 4H). 31P{1H}
NMR (121.5 MHz, CDCl3): δ -144.33 (septet, PF-6 , JF-P=713
Hz), 46.39 (s, dppc+). Anal. Calcd for C52H48CoF6P3Ru: C,
60.06; H, 4.65. Found: C, 60.35; H, 4.72.
Kinetics of the Reaction between 3j2+ and t-BuCN. A stock
solution of 3j2+ was prepared by treating 0.06 mmol of 2j[PF6]
with an equimolar amount of HBF4 OMe2 in 2.4 mL of
3
CD2Cl2. The stock solution was stored at -20 °C when not in
use. For each experiment, an NMR tube was loaded with 400 μL
of stock solution, additional CD2Cl2, and t-BuCN. The addi-
tional CD2Cl2 was such that the total volume of the solution was
always 600 μL. After the addition of t-BuCN, the NMR tube
was shaken and kept in an EtOH/CO2 bath until being inserted
into the NMR probe. The ratio of the 3j2+ peak at δ 6.48 to the
solvent peak at δ 5.32 was monitored at 19.5 °C. The rate did not
depend on [t-BuCN]. The average of five experiments (ranging
from 13.6 to 31.7 equiv of t-BuCN) gave a first-order rate
[Cp*Ru(dppc)CCHPh]2+ (8j2+). Treating 7j[PF6] with HBF4
constant of 2.0(1) ꢀ 10-3 s-1 for the disappearance of 3j2+
.
3
1
OMe2 in CDCl3 gave a yellow solution. H NMR (300 MHz,
The product of the reaction is [Cp*Ru(dppc)t-BuCN]2+ 1H
.
CDCl3): δ 1.34 (s, RuCp*, 15H), 5.65 (s, dppc+ Cp, 2H), 5.87 (s,
RuCCHPh, 1H), 6.08 (s, dppc+ Cp, 4H), 6.14 (s, dppc+ Cp,
2H), 7.22 (m, Ar, 4H), 7.31-7.55 (m, Ar, 15H), 7.65 (m, Ar, 4H),
7.78 (m, Ar, 2H). 31P{1H} NMR (121.5 MHz, CDCl3): δ 47.80
(s, dppc+). 13C{1H} NMR (75 MHz, CDCl3) distinctive
resonance: δ 357.58 (t, RuCCHPh, JP-C=15.7 Hz).
NMR (300 MHz, CD2Cl2): δ 1.09 (s, RuCp*, 15H), 1.72 (s,
t-BuCN, 9H), 5.63 (s, br, dppc+ Cp, 4H), 5.69 (s, dppc+ Cp,
2H), 5.98 (s, dppc+ Cp, 2H), 7.43-7.79 (m, Ar, 20H). 31P{1H}
NMR (121.5 MHz, CD2Cl2): δ 40.45 (s, dppc+).
X-ray Structure Determination. Data were collected on a
Bruker Apex II diffractometer. The structures were solved
using direct methods and standard difference map tech-
niques and refined by full-matrix least-squares procedures using
SHELXTL version 5.10 software. Hydrogen atoms on carbon
were included in calculated positions. Vapor diffusion of Et2O
into a concentrated CH3CN solution of 6j[PF6]2 gave crystals of
6j[PF6]2 Et2O CH3CN suitable for X-ray analysis; the disor-
N-(Isopropylidene)pyrrolidinium Tetrafluoroborate (9). 1H NMR
(400 MHz, CDCl3): δ 2.22 (m, β-CH2, 4H), 2.51 (s, CH3, 6H),
3.95 (m, R-CH2, 4H). FAB+MS (m-NBA): m/zfor C7H14N+ calcd
112.1126; found 112.1120.
N-(1-Methylbenzylidene)pyrrolidinium Tetrafluoroborate (10).
1H NMR (400 MHz, CDCl3): δ 2.09 (m, β-CH2, 2H), 2.31 (m, β-
CH2, 2H), 2.78 (s, CH3, 3H), 3.89 (m, R-CH2, 2H), 4.21 (m, R-CH2,
2H), 7.46-7.62 (m, Ar, 5H). FAB+MS (m-NBA): m/z for
C12H16N+ calcd 174.1283; found 174.1276.
3
3
dered solvent molecules were treated with SQUEEZE in PLA-
1
TON.75 A H NMR spectrum in CD2Cl2 confirmed that the
crystals contained Et2O, CH3CN, and 6j[PF6]2 in a 1:1:1 ratio.
Layering Et2O onto a concentrated CH2Cl2 solution of 7j[PF6]
gave crystals of 7j[PF6] Et2O suitable for X-ray analysis;
N-(Benzylidene)pyrrolidinium Tetrafluoroborate (11). 1H NMR
(300 MHz, CD2Cl2): δ 2.13-2.42 (m, β-CH2, 4H), 4.23 (m, R-CH2,
2H), 4.43 (m, R-CH2, 2H), 7.68 (t, m-Ar, J=7.7 Hz, 2H), 7.82
(t, p-Ar, J=7.4 Hz, 1H), 7.95 (d, o-Ar, J=7.7 Hz, 2H), 9.06 (s, CH,
1H). FAB+MS (m-NBA): m/z for C11H14N+ calcd 160.1126;
found 160.1130.
3
the uncoordinated Et2O was not disordered. CCDC 709698
and 709699 contain the supplementary crystallographic data
for 6j[PF6]2 Et2O CH3CN and 7j[PF6] Et2O. These data
3
3
3
can be obtained free of charge from The Cambridge Crys-
tallographic Data Centre via http://www.ccdc.cam.ac.uk/
data_request/cif.
N-(Isobutylidene)pyrrolidinium Tetrafluoroborate (12). 1H NMR
(300 MHz, CDCl3): δ 1.32 (d, CH3, J=6.7 Hz, 6H), 2.23 (m, β-
CH2, 4H), 2.89 (m, CH, 1H), 3.98 (m, R-CH2, 2H), 4.26 (m, R-CH2,
2H), 8.35 (d, CH, J=9.4 Hz, 1H). FAB+MS (m-NBA): m/z for
C8H16N+ calcd 126.1283; found 126.1279.
Acknowledgment. We thank the National Science
Foundation (CHE-0749537) for partial support of this
research (A.P.S. and J.R.N.). We thank Prof. G. Parkin
for assistance with the X-ray crystallography (D.B.) and
the National Science Foundation for funding the acquisi-
tion of an X-ray diffractometer (CHE-0619638). We
thank the donors of the Petroleum Research Fund,
administered by the American Chemical Society, for
partial support of this research (L.A.S., S.S.K., D.A.J.,
K.M.B., and C.N.).
1,2-Bis(N-pyrrolidino)-1,2-diphenylethane (13). The two dia-
stereomers of 13 may be distinguished by 1H NMR in CDCl3 but
not in THF-d8. In CDCl3, the methine proton resonances of the
diastereomers appear at δ 3.93 and 3.87, while in THF-d8 these
resonances overlap at δ 3.95. The 1H NMR spectrum of (S,S)-13
in CDCl3 has been reported.74 Mixing 11 with 2i in THF-d8 gave
a solution of 13 and 2j[BF4]. In another experiment, mixing 11
with Cp2Co in THF-d8 gave [Cp2Co][BF4] as a yellow precipi-
tate and 13 in solution. An aliquot of the solution was used for
FAB+MS (m-NBA): m/z for 13(H)+ calcd 321.2331; found
321.2318. The remainder of the solution was decanted and
1
Supporting Information Available: Crystallographic data
(CIF files) for 6j[PF6]2 Et2O CH3CN and 7j[PF6] Et2O. This
evaporated, and the residue was dissolved in CDCl3. The H
NMR spectrum showed a 4:5 ratio of the δ 3.93 to 3.87 peaks,
indicating a 4:5 diastereomer ratio.
3
3
3
material is available free of charge via the Internet at http://
pubs.acs.org.
(74) Johansson, M. J.; Schwartz, L.; Amedjkouh, M.; Kann, N.
Tetrahedron: Asymmetry 2004, 15, 3531–3538.
(75) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7–13.