Effects of phosphine ligand chelation on the reactivity of monomeric parent amido ruthenium complexes: Synthesis and reactivity of such a complex bearing monodentate ligands

Article abstract of DOI:10.1021/ja027620f

The parent amido complex cis-(PMe₃)₄Ru(H)(NH₂) (2) has been prepared via the deprotonation of [cis-(PMe₃)₄Ru(H)(NH₃)⁺] [BPh₄^-]. The amido complex is a somew

Full text of DOI:10.1021/ja027620f

Published on Web 11/15/2002
Effects of Phosphine Ligand Chelation on the Reactivity of
Monomeric Parent Amido Ruthenium Complexes: Synthesis
and Reactivity of Such a Complex Bearing Monodentate
Ligands
Andrew W. Holland and Robert G. Bergman*
Contribution from the Department of Chemistry and Center for New Directions in Synthesis,
UniVersity of California, Berkeley, California 94720
Received July 9, 2002
Abstract: The parent amido complex cis-(PMe3)4Ru(H)(NH2) (2) has been prepared via the deprotonation
of [cis-(PMe3)4Ru(H)(NH3)+][BPh4-]. The amido complex is a somewhat weaker base than the DMPE
analogue trans-(DMPE)2Ru(H)(NH2) but is still basic enough to quantitatively deprotonate fluorene and
reversibly deprotonate 1,3-cyclohexadiene and toluene. Complex 2 exhibits very labile phosphine ligands,
two of which can be replaced by DMPE to yield the mixed complex cis-(PMe3)2(DMPE)Ru(H)(NH2). Because
of the ligand lability, 2 also undergoes hydrogenolysis and rapid exchange with labeled NH3. The amide
complex reacts with alkyl halides to yield E2 and SN2 products, along with ruthenium hydrido halide
complexes including the ruthenium fluoride cis-(PMe3)4Ru(H)(F). Ruthenium hydrido ammonia halide ion
pair intermediates [cis-(PMe3)4Ru(H)(NH3)+][X-] are observed in some deprotonation and E2 reactions,
and measurement of the equilibrium constants for NH3 displacement from these complexes suggests that
they benefit from significant hydrogen bonding between X- and NH3 groups. Cumulenes also react with
complex 2 to afford the products of insertion into an NH bond. The rates of neither these NH insertion
reactions nor the reversible deprotonation reactions show any dependence on the concentration of PMe3
present, suggesting that these reactions take place directly at the NH2 group and do not involve
precoordination of substrate to the metal center.
Introduction
late metal amide complexes still remain rare relative to their
alkoxide counterparts,5,16 and the chemistry of the most funda-
mental amide complexes, parent amido species (M-NH2), is
particularly underdeveloped.17-20 This is a consequence of both
the rarity of practical NH2- sources and the relative instability
of most reported parent amido complexes.
Our group recently reported the synthesis of trans-(DMPE)2Ru-
(H)(NH2) (1) (DMPE ) 1,2-bisdimethylphosphinoethane), the
first isolable monomeric structurally characterized late-metal
parent amido complex.21-23 This complex proved to be remark-
ably basic, and the pKa of the corresponding ammonia complex
was estimated to be approximately 23-24 (as measured in
THF).23 Complex 1 quantitatively deprotonates weak carbon
acids such as fluorene and undergoes H/D exchange with weaker
acids (including toluene) via reversible proton transfer.23
Late metal complexes featuring alkoxide and amide ligands
have long been implicated as intermediates in important catalytic
reactions.1-6 Amide complexes7 in particular are postulated to
be formed as intermediates in the catalytic cycles of important
synthetic processes including hydroamination,8-10 asymmetric
hydrogenation,11 and C-N coupling reactions.12-15 Nonetheless,
* Address correspondence to this author. E-mail: bergman@
cchem.berkeley.edu.
(1) Sawyer, D. T. Oxygen Chemistry; Oxford University Press: New York,
1991.
(2) Cornils, B.; Herrmann, W. A. Applied Homogeneous Catalysis with
Organometallic Compounds; VCH Publishers: New York, 1996; Vol. 1.
(3) Cotton, F. A.; Wilkinson, G.; Murillo, C. A.; Bochmann, M. AdVanced
Inorganic Chemistry, 6th ed; John Wiley & Sons: New York, 1999.
(4) Pez, G. P.; Galle, J. E. Pure Appl. Chem. 1985, 57, 1917.
(5) Bryndza, H. E.; Tam, W. Chem. ReV. 1988, 88, 1163.
(6) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles and
Applications of Organotransition Metal Chemistry; University Science
Books: Mill Valley, CA, 1987.
(7) Kempe, R. Angew. Chem., Int. Ed. 2000, 39, 468.
(8) Muller, T. E.; Beller, M. Chem. ReV. 1998, 98, 675.
(9) Muller, T. E.; Grosche, M.; Herdtweck, E.; Pleier, A.-K.; Walter, E.; Yan,
Y.-K. Organometallics 2000, 19, 170.
(10) Gasc, M. B.; Lattes, A.; Perie, J. J. Tetrahedron 1983, 39, 703.
(11) Noyori, R.; Hashiguchi, S. Acc. Chem. Res. 1997, 30, 97.
(12) Hartwig, J. F.; Kawatsura, M.; Hauck, S. I.; Shaughnessy, K. H.; Alcazar-
Roman, L. M. J. Org. Chem. 1999, 64, 5575.
(13) Hartwig, J. F.; Richards, S.; Baranano, D.; Paul, F. J. Am. Chem. Soc.
1996, 118, 3626.
(16) Fulton, J. R.; Holland, A. W.; Fox, D. J.; Bergman, R. G. Acc. Chem. Res.
2002, 35, 44.
(17) Dewey, M. A.; Knight, D. A.; Gladysz, J. A. Chem. Ber. 1992, 125, 815.
(18) Joslin, F. L.; Johnson, M. P.; Mague, J. T.; Roundhill, D. M. Organome-
tallics 1991, 10, 41.
(19) Tahmassebi, S. K.; McNeil, W. S.; Mayer, J. M. Organometallics 1997,
16, 5342.
(20) Jayaprakash, K. N.; Conner, D.; Gunnoe, T. B. Organometallics 2001, 20,
5254.
(21) Kaplan, A. W.; Ritter, J. C. M.; Bergman, R. G. J. Am. Chem. Soc. 1998,
120, 6828.
(22) Fulton, J. R.; Bouwkamp, M. W.; Bergman, R. G. J. Am. Chem. Soc. 2000,
(14) Wolfe, J. P.; Buchwald, S. L. J. Am. Chem. Soc. 1997, 119, 6054.
(15) Wagaw, S.; Rennels, R. A.; Buchwald, S. L. J. Am. Chem. Soc. 1997, 119,
8451.
122, 8799.
(23) Fulton, J. R.; Sklenak, S.; Boukamp, M.; Bergman, R. G. J. Am. Chem.
Soc. 2002, 124, 4722.
9
14684
J. AM. CHEM. SOC. 2002, 124, 14684-14695
10.1021/ja027620f CCC: $22.00 © 2002 American Chemical Society

Reactivity of Parent Amido Ruthenium Complexes
A R T I C L E S
One unfortunate feature of the DMPE ligands of complex 1
is that their dechelation is not detectable by typical kinetic
experiments, and thus, the role of phosphine dissociation in the
reactivity of 1 is beyond the reach of conventional experiments.
Additionally, the chelation of DMPE strongly disfavors phos-
phine dissociation and, thus, retards any reactivity that does
require an open site at the metal center.3,24 When given the
frequency with which organometallic complexes require such
sites for both stoichiometric and catalytic reactions,3,24 this is
likely to interfere with attempts to extend the reactivity of amide
complexes such as 1 beyond reactions that occur directly at the
NH2 group.
decomposes to multiple products over the course of several days.
The complex decomposes quickly in air and reacts rapidly with
traces of water to liberate ammonia and form the known
hydroxide complex cis-(PMe3)4Ru(H)(OH)26 (5) (eq 2). In the
presence of catalytic quantities of some acids, including tBuOH,
Ph2NH, and molecular sieves, complex 2 decomposes to cleanly
form the known cyclometalated complex cis-(PMe3)3Ru(CH2-
PMe2)(H)27 (6) and liberate ammonia (eq 3). In the cases of
We have thus sought to prepare an analogue of 1 featuring
at least one monodentate phosphine ligand. Our hope was that
such a complex, while perhaps less stable than 1, would exhibit
novel reactivity and lend itself to a mechanistic investigation
more readily. To this end, we have now synthesized the complex
cis-(PMe3)4Ru(H)(NH2) (2), and we report herein on its reactiv-
ity.
tBuOH and Ph2NH, this transformation has been observed to
proceed via an ion pair intermediate [cis-(PMe3)4Ru(H)(NH3)+]-
[A-] which forms immediately and is converted to 6 in 2-4
days (at ambient temperature).
Complex 2 was prepared in the hope that its PMe3 ligands
would prove sufficiently labile to allow for the investigation
and extension of the chemistry exhibited by DMPE analogue
1, and evidence suggests that at least some of the PMe3 ligands
are indeed quite labile. Reaction of 2 with either CO or tBuNC
results in the liberation of PMe3 and the formation of multiple
ruthenium hydride products featuring PMe3 displacement both
trans and/or cis to the hydride. Treatment of 2 with 2 equiv of
DMPE results in the displacement of only two PMe3 ligands to
form a new amide complex (PMe3)2(DMPE)Ru(H)(NH2) (7)
(eq 4). The 31P NMR spectrum of 7 features two resonances in
Results
Synthesis and Characterization of cis-(PMe3)4Ru(H)(NH2).
The ammonia complex [cis-(PMe3)4Ru(H)(NH3)+][BPh4-] (3)
has been prepared previously and found to form the dihydride
cis-(PMe3)4Ru(H)2 (4) upon treatment with hydride bases.25
Treatment of 3 with KN(SiMe3)2, however, cleanly affords
amide complex cis-(PMe3)4Ru(H)(NH2) (2) (eq 1). Complex 2
the DMPE region (∼δ 60-30 ppm) and two in the PMe3 region
(∼δ 20 to -10 ppm), including an upfield DMPE signal
corresponding to a phosphorus atom positioned trans to a
hydride (δ 30.03 ppm) and a PMe3 signal (δ 3.67 ppm) very
similar to that observed for the phosphorus atom trans to the
NH2 group in complex 2 (δ 3.12 ppm). The other two
phosphorus atoms are disposed trans to each other and are, thus,
strongly coupled (J ) 323 Hz).
Acid-Base Reactivity. The basicity of complex 2 was
probed by exploring its reactivity toward sterically encumbered
weak acids. Complex 2 was found to quantitatively deprotonate
fluorene to form the fluorenide salt 8 (eq 5), but no reaction
is produced in 95% yield as a crude yellow-tan oil of 95% purity,
and subsequent crystallization of the highly soluble complex
from pentane affords pure 2 as off-white crystals in 60% yield.
The cis geometry of 2 is indicated by the presence of resonances
1
for three chemically inequivalent PMe3 ligands in the H, 13C,
and 31P NMR spectra. The RuH signal (δ -8.49 ppm) also
indicates a cis geometry, featuring one large coupling (J ) 99
Hz) to a phosphorus atom trans to the hydride and a doublet of
triplets pattern arising from two smaller couplings (J ) 29, 24
Hz) to the three cis disposed phosphorus groups. The NH2
protons appear in the 1H NMR spectrum as a single, sharp peak
at δ -2.39 ppm, in roughly the same region as the corresponding
signal associated with DMPE analogue 1 (δ -3.42 ppm).21 The
shape and chemical shift of this peak are strongly dependent
on the purity of the sample; it broadens and generally shifts
downfield in the presence of impurities.
When pure, amide complex 2 is stable indefinitely in the solid
state at room temperature and stable in solution in halide free,
nonacidic solvents below 75 °C. At this temperature, it slowly
was observed between 2 and triphenylmethane (which is
reversibly deprotonated by 1). Like its DMPE analogue,
fluorenide salt 8 did not undergo NH3 displacement, even upon
(24) Crabtree, R. H. The Organometallic Chemistry of Transition Metals, 2nd
ed.; John Wiley & Sons: New York, 1994.
(25) Rappert, T.; Yamamoto, A. Organometallics 1994, 13, 4984.
9
J. AM. CHEM. SOC. VOL. 124, NO. 49, 2002 14685

A R T I C L E S
Holland and Bergman
Scheme 1
prolonged heating. To directly compare the relative basicities
of 2 and the trans-DMPE analogue 1, ammonia complex 8 was
treated with 1. This resulted in an immediate quantitative proton
transfer to yield 2 and DMPE ammonia complex 9 (eq 6). Salt
3 displayed similar proton transfer reactivity.
Figure 1. Pseudo-first-order plot for kinetics of the isomerization of 1,4-
cyclohexadiene (0.15 M) by 2 (0.038 M ) 25 mol %) in the presence of
PMe3 (0.005 M, ]; or 0.11 M, 9) in C6D6 at 22 °C.
equilibration takes place with a half-life of approximately 90
min when a catalyst loading of 25% 2 is used. The kinetics of
Like DMPE analogue 1, amide complex 2 quantitatively
deprotonates p-cresol (TolOH) upon mixing to yield the
ammonia-aryloxide ion pair 10 (Scheme 1). This species exists
in slow equilibrium with the inner-sphere cresolate 1128 (Keq
≈
400 M-1), in which the ammonia has been displaced by the
aryloxide. Efforts to investigate the effect of [PMe3] on the rate
of this displacement reaction were complicated by the reversible
displacement of ammonia by PMe3 to yield the pentaphosphino
hydride cation 12. Species 12 is characterized by a doublet of
this process were investigated in the presence of both 0.1 and
3.0 equiv of PMe3, and the rate law was found to be first order
in [substrate] and independent of [PMe3] (Figure 1). This
suggests that phosphine dissociation is not required and implies
a mechanism involving reversible, direct, intermolecular depro-
tonation of diene by 2 (Scheme 2).
1
quintets in the RuH region of its H NMR spectrum and the
presence of both a doublet and quintet in the 31P NMR
spectrum.29 This equilibrium, and the significance of its position,
is discussed further below. Complex 2 also reacts with aniline
to yield the protonolysis product cis-(PMe3)4Ru(H)(NHPh)28
(13) and NH3 (eq 7), although in this case the ion pair
intermediate was not observed.
Scheme 2
Treatment of 2 with ND3 results in a rapid decrease in the
intensity of the RuNH2 signal and appearance of NHnD3-n in
the 1H NMR spectrum of the mixture (eq 10). As in the case of
In contrast to 1, amide complex 2 does not react significantly
with neat toluene-d8 at room temperature, but slow H/D
exchange between the benzylic position of the solvent and
complex 2 does occur at 45 °C (eq 8). The RuNH2 protons are
exchanged first (90% deuteration in 2 d), and deuteration of
the phosphine ligands is observed thereafter (90% deuteration
in 5 d). Addition of PMe3 to this reaction mixture has no
significant effect on the rate of initial decay of the RuNH2
resonance, although deuteration of the free phosphine is
eventually observed. Complex 2 also catalyzes the equilibration
of 1,3-cyclohexadiene with its 1,4 isomer (eq 9), and this
H/D exchange with toluene-d8, exchange into the phosphine
ligands eventually occurs thereafter. When ND3 is removed after
deuteration of the RuNH2 group (∼10 min), the amide group is
almost entirely reprotonated and the phosphine ligands become
partially deuterated over the course of the next 24 h. The initial
exchange between ND3 and the NH2 group occurs much more
readily than that between ND3 and 1, suggesting that a different
mechanism may be operative for the less basic complex
(exchange processes involving amide complexes 1 and 2 are
compared in Scheme 3). Treatment of 2 with 15NH3 (δ -0.18
ppm (d, J ) 60.8 Hz)) results in rapid liberation of 14NH3 (δ
-0.18 ppm (t, J ) 42.8 Hz)) (eq 11), and the mass spectrum of
the isolated product cis-(PMe3)4Ru(H)(15NH2) (2-15N) confirms
(26) Burn, M. J.; Fickes, M. G.; Hartwig, J. F.; Hollander, F. J.; Bergman, R.
G. J. Am. Chem. Soc. 1993, 115, 5875.
(27) Werner, H.; Werner, R. J. Organomet. Chem. 1981, 209(3), C60.
(28) Hartwig, J. F.; Andersen, R. A.; Bergman, R. G. Organometallics 1991,
10, 1875.
(29) A very similar complex has been reported previously: Burn, M. J.;
Bergman, R. G. J. Organomet. Chem. 1994, 472, 43.
9
14686 J. AM. CHEM. SOC. VOL. 124, NO. 49, 2002

Reactivity of Parent Amido Ruthenium Complexes
A R T I C L E S
Scheme 3
it to be enriched in 15N (m/z ) 424). This exchange, which
does not take place at all with DMPE analogue 1, occurs too
rapidly at room temperature to allow convenient exploration of
the phosphine dependence of the reaction rate.
Amide complex 2 reacts slowly with D2 to yield NH2D and
P4RuHD (4-d1) (eq 12) rather than H/D exchange products
analogous to those observed with 1. This reaction is slowed
dramatically in the presence of 2 equiv of PMe3, but H/D
exchange into the RuNH2 of the starting material still is not
Figure 2. ORTEP diagram of 21 (thermal ellipsoids at 50% probability).
Hydrogen atoms, second molecule of unit cell, and solvent have been
omitted for clarity. Selected bond distances (Å) and angles (deg): Ru-
N1, 2.180(6); Ru1-P1, 2.349(2); Ru1-P2, 2.283(2); Ru1-P3, 2.339(2);
Ru1-P4, 2.346(3); C1-N1, 1.341(9); C1-N2, 1.408(10); C1-N3, 1.313-
(10); Ru1-N1-C1, 133.1(6); N1-C1-N2, 114.4(7); N2-C1-N3, 117.6-
(7); P2-Ru1-N1, 172.8(2); P1-Ru1-P3, 162.59(9); P4-Ru1-N1, 89.9-
(2); Ru1-N1-C1-N2, 11(1).
1
observed by H NMR spectroscopy.
complex. While traces of ethylene and ammonia bromide
complex 16 are observed in this reaction, the major products
are ethylamine (20) and bromide complex 19 (eq 16). Even
when the reaction is monitored within minutes of mixing, no
intermediate ethylamine-bound cation is observed.
While complex 2 does not appear to behave strictly as a
Brønsted base in its reactions with ND3 and D2, its considerable
basicity is reflected in its reactivity toward various alkyl halides.
Complex 2 reacts with 1,1,1-trifluoroethane, 1-chloro-2-phen-
ylethane, and 2,3-dibromobutane to form protonated organo-
metallic products [cis-(PMe3)4Ru(H)(NH3)+][X-] (X ) F, 14;
X ) Cl, 15; X ) Br, 16) featuring the halides as outer-sphere
anions (eq 13-15). Over several hours at room temperature,
Treatment of complex 2 with di-p-tolyl carbodiimide results
in an overall insertion of the carbodiimide into the NH bond of
the amide complex to yield guanidinate 21 (eq 17). The 1H NMR
spectrum of the product is characterized by the presence of two
tolyl groups (δ 2.35, 2.21 ppm) and a single RuNH proton at δ
9.02 ppm. (The second NH proton is not observed even at
reduced (-80 °C) or elevated (75 °C) temperatures, presumably
because of quadrupole broadening). The 31P NMR spectrum
features a downfield resonance at δ 5.37 ppm, which is
consistent with the presence of an anionic nitrogen group trans
to the corresponding phosphine. The structure of complex 21
was confirmed by an X-ray diffraction study, and the ORTEP
diagram and significant bond lengths and angles are shown in
Figure 2. In an effort to determine whether the insertion shown
in eq 17 takes place via precoordination of the carbodiimide
(following phosphine loss) or by a direct nucleophilic attack of
the NH2 group on the substrate, we attempted to study the
dependence of the rate of the reaction on [PMe3]. Unfortunately,
even in the presence of 10 equiv of PMe3, the reaction proceeded
too rapidly at -80 °C for its rate to be measured. Dicyclohexyl
carbodiimide reacts with amide complex 2 in a similar manner,
yielding N-H insertion product 22 (eq 17). In this case, the
the ammonia is displaced to form the corresponding ruthenium
hydrido fluoride (17), chloride (18), or bromide (19) complex.
The organic dehydrohalogenation products styrene and 2-bro-
mobutene are cleanly generated in these reactions, but 1,1-
difluoroethylene has not been observed. The stereochemistry
of the elimination was investigated using rac- and meso-2,3-
dibromobutane. Dehydrobromination of the rac isomer yielded
the Z-2-bromobutane with 98% selectivity, whereas the meso
compound yielded 97% of the E isomer (eq 15).
Nucleophilic Reactivity. The reaction of amide complex 2
with ethyl bromide highlights the nucleophilic reactivity of the
9
J. AM. CHEM. SOC. VOL. 124, NO. 49, 2002 14687

A R T I C L E S
Holland and Bergman
Table 1. Equilibrium Constants for Displacement of NH3 by PMe3
in Eq 19
entry
anion (compd)
K
eq
1
2
3
4
5
6
7
OTol- (10)
OTf- (24)
C13H9- (8)
BF4- (25)
BPh4- (3)
BArf- (26)
F- (14)
0.45 ( 0.05
1.0 ( 0.1
1.4 ( 0.1
4.5 ( 0.1
16 ( 1
19 ( 1
a
a No product.
positions of related equilibria (eq 19) should be sensitive to the
hydrogen bonding capabilities of A- if such interactions are
significant in these species. The position of this equilibrium was
Figure 3. Pseudo-first-order plot for kinetics of the reaction of 2 (0.032
M) with dicyclohexyl carbodiimide (0.30 M) in the presence of PMe3 (0.003
M, ]; or 0.16 M, 9) in THF-d8 at -29 °C.
investigated for ion pairs [(PMe3)4RuH(NH3)+][A-] where A-
-
-
-
) BPh4 (3), C13H9 (8), OTol- (10), OTf- (24), BF4 (25),
and BArf- (BArf ) B(3-5-C6H3(CF3)2)4) (26), and the resulting
equilibrium constants are shown in Table 1. The data show that
the ammonia complex is more favored for anions with localized
charges and steric properties suitable for hydrogen bonding.
Unfortunately, ammonia halide complexes 14-16 collapsed
irreversibly to hydrido halide complexes 17-19 before equi-
librium with PMe3 was attained.
Discussion
Synthesis and Stability. While amide complex 2 can be
prepared by NaNH2/NH3(l) methodology analogous to that used
in the synthesis of DMPE analogue 1,21 we found the depro-
tonation route shown in eq 1 to be more effective. Deprotonation
by hydride and alkyllithium bases has been used to prepare
parent amido complexes previously,18,20 but these bases were
ineffective in performing this deprotonation cleanly. The steri-
cally hindered base KN(SiMe3)2, however, affords product 2
in excellent purity and high crude yield, presumably because
of the inability of the base to displace the ammonia ligand.
Unfortunately, the extremely high solubility of 2 even in pentane
(∼0.5 g/mL) renders its isolation somewhat difficult, limiting
the yield of its synthesis.
Figure 4. Pseudo-first-order plot for kinetics of the reaction of 2 (0.040
M) with dicyclohexyl carbodiimide (0.40 M, ]; or 0.79 M, 9) in THF-d8
at -31 °C.
reaction proceeds sufficiently slowly that it can be monitored
at -30 °C. The pseudo-first-order plots of two parallel experi-
ments using 0.1 and 5.0 equiv of PMe3 are shown in Figure 3;
they show that the concentration of PMe3 is not a factor in the
rate of the reaction. In a separate experiment employing various
concentrations of carbodiimide (Figure 4), the reaction was
found to be cleanly first order in the carbodiimide.
The N-H insertion reactivity of complex 2 is not limited to
heterocumulenes, as diphenyl allene also reacts rapidly with 2
1
to yield product 23 (eq 18). Again, both the H and 31P NMR
Unlike DMPE analogue 1, complex 2 assumes a cis geometry.
This geometry is common to almost all (PMe3)4Ru(H)(X)
complexes and is presumably due largely to the steric preference
against crowding the relatively large PMe3 groups into an
equatorial plane. In this case, the cis geometry also allows the
alignment of the strongly π-donating NH2 ligand trans to a PMe3
ligand (which is weakly π-acidic) rather than a hydride (which
is not π-accepting). The adoption of the trans geometry in the
DMPE system is probably a result of steric influences imposed
by the chelating ligands and may be either a thermodynamic or
kinetic phenomenon.
Complex 2 is reasonably stable at moderate temperatures,
although it is very air sensitive (turning brown rapidly upon
exposure) and reacts with traces of water to liberate ammonia
and produce the corresponding hydroxide complex 5. The amide
group is reactive toward protonolysis by stronger acids such as
cresol as well, although this takes place by an initial formation
of an ion pair followed by subsequent ammonia displacement,
spectra of this product are consistent with its formulation as an
N-bound N-H insertion product.
Investigation of Hydrogen Bonding in Ion Pairs. As
described earlier, ammonia cresolate complex 10 reacts revers-
ibly with PMe3 to yield the pentaphosphino ruthenium hydride
cresolate 12 (Scheme 1). This equilibrium provided an op-
portunity to investigate the energetic significance of hydrogen
bonding interactions in [(PMe3)4RuH(NH3)+][A-] products
resulting from proton transfer to 2. To the extent that hydrogen
bonding is possible in ion pairs such as 10, we suspected that
such interactions would favor the ammonia complexes relative
to phosphine displacement products such as 12. Thus, the
9
14688 J. AM. CHEM. SOC. VOL. 124, NO. 49, 2002

Reactivity of Parent Amido Ruthenium Complexes
A R T I C L E S
Scheme 4
Scheme 5
Scheme 6
as is observed for DMPE analogue 1. It is not clear why the
protonolysis by water is particularly rapid and does not involve
an observable ion pair intermediate. In cases where ammonia
displacement by the anion is sterically prohibitive, as with
tBuOH and Ph2NH, these ion pairs sometimes decompose by
ammonia displacement and PMe3 cyclometalation to yield
complex 6. The cyclometalation could take place via deproto-
nation of the P-CH3 group by the strongly basic anion (we
have independently observed related base-induced cyclometa-
lation in a similar ruthenium ammonia complex30) or by initial
dissociation of either NH3 or PMe3; further experiments may
well distinguish among these mechanisms. The conversion of
2 to 6 and NH3 occurs even when catalytic amounts of acid are
used, and the addition of 4 atm of NH3 to the reaction mixture
does not regenerate amide 2. This suggests that the amide
complex is less stable than the cyclometalation products.
Basicity and Exchange Processes. All evidence suggests that
2 is moderately less basic than its DMPE analogue 1. Species
2 undergoes H/D exchange with toluene and cyclohexadienes
more slowly than does 1, and direct comparison of the two
species results in quantitative deprotonation of ammonia
complex 8 by 1 (pKa ) 23-24 in THF)23 (eq 8). Although 2 is
sufficiently basic to cleanly deprotonate fluorene (pKa ) 22.9
in THF)31 to yield ion pair 8, this result gives an overestimate
of the pKa of 8 because of ion pairing effects that we have
discussed at length elsewhere.23 The most reasonable estimate
of the pKa of 8 is probably 20-21 (in THF) on the basis of the
two results above. The attenuated basicity of 2 relative to DMPE
analogue 1 can be attributed largely to its cis geometry and the
resulting stabilization of its NH2 lone pair by its interaction with
the weakly π-accepting P-C σ* orbitals of the trans PMe3
ligand. The presence of π-acidic substituents or ligands has
previously been observed to reduce the basicity of amide
complexes,20 and a ligand in the trans orientation would be
expected to have the greatest impact in this regard.3 Geometric
constraints imposed by chelating ligands have previously been
observed to enhance the basicity of transition metal centers.32
Another striking difference between complexes 1 and 2 is
the lability of the PMe3 ligands in 2 compared with the tightly
bound chelating DMPE ligands of 1. While 1 exchanges only
slowly with ND3 and does not exchange with 15NH3 at all, 2
exchanges rapidly with both substrates at room temperature.
When given the lower basicity of 2, we presume that its
exchange reactions with labeled ammonia take place via
phosphine displacement by the incoming NH3 group, rather than
by its initial deprotonation (Scheme 4). Additionally, complex
2 reacts with D2 to yield the hydrido deuteride 4-d1 and NDH2
rather than H/D exchange products (like those observed in the
analogous reaction of 1), and this reaction is retarded by the
addition of phosphine. This suggests the mechanism shown in
Scheme 5, which has previously been proposed for the hydro-
genolysis of related amide and alkoxide complexes,28 in which
phosphine dissociation is a key step.
The lability of the phosphine ligands in 2 is not surprising.
Not only are they monodentate ligands in contrast to the
chelating DMPE groups in 1, but the PMe3 ligand located trans
to the hydride is labilized by both the strong trans effect of the
hydride ligand and the strong cis effect of the π-donating NH2
group.3 The reaction of complex 2 with DMPE highlights the
regiochemistry of phosphine dissociation (product 7 appears to
be the kinetic product) as a new complex substituted trans to
the hydride and cis to the NH2 group is formed (eq 4). This
substitution reaction also suggests that complex 2 may indeed
be a useful synthon in the preparation of new amide complexes
featuring a variety of ligands, and we have begun to explore
this possibility.
Role of Phosphine Dissociation. While amide complex 2
has relatively labile phosphine ligands, this feature does not
appear to play a significant role in many of the basic and
nucleophilic reactions the complex undergoes. H/D exchange
reactions involving complex 2 are kinetically complicated,
because of the fact that exchange with the NH2 group is only
moderately faster than the exchange between the NH2 protons
and those of the PMe3 ligands. Additionally, efforts to inves-
tigate the effect of added phosphine on H/D exchange were
complicated by an apparent exchange between free phosphine
and partially deuterated bound phosphine. These exchange
processes result in the overall catalytic H/D exchange of ND3
with PMe3, and a representative cycle for this deuterium transfer
is illustrated in Scheme 6. The isomerization of 1,3-cyclohexa-
diene to the 1,4 isomer (eq 9) provides a more useful probe
into the rate of proton transfer and occurs at a rate qualitatively
similar to that for H/D exchange into the diene. The isomer-
ization proceeds at a rate independent of phosphine concentra-
tion (Figure 1), indicating that the process takes place by direct,
(30) Holland, A. W.; Bergman, R. G. Organometallics 2002, 21, 2149-2152.
(31) Streitwieser, A.; Wang, D. Z.; Stratakis, M.; Facchetti, A.; Gareyev, R.;
Abbotto, A.; Krom, J. A.; Kilway, K. V. Can. J. Chem. 1998, 76, 765.
(32) Angelici, R. J. Acc. Chem. Res. 1995, 28, 51.
9
J. AM. CHEM. SOC. VOL. 124, NO. 49, 2002 14689

A R T I C L E S
Holland and Bergman
Scheme 7
amination of alkyl halides using a catalytic amount of complex
2 are underway.
Hydrogen Bonding. While the displacement of ammonia
from ion pairs [(PMe3)4RuH(NH3)+][A-] by PMe3 prevented
us from conveniently exploring the significance of phosphine
dissociation in these reactions, it also offered an opportunity to
explore the extent of hydrogen bonding in these systems. The
crystal structures and NMR behavior of some ion pairs involving
DMPE analogue 1 showed evidence of strong hydrogen bonding
interactions,23 and such interactions would be important to the
energetics involved in deprotonation reactions involving amides
such as 1 and 2 and in the subsequent displacement reactions
of initial proton transfer products. To explore the role of
solution-phase hydrogen bonding in these systems, we inves-
tigated the effect of the anion on the position of the equilibrium
shown in eq 19 and found that anions capable of functioning
as hydrogen bonding acceptors favored the corresponding
ammonia complex over the pentaphosphino displacement prod-
uct. Cresolate favored the ammonia complex most strongly, and
triflate, while a very weak base, favored the ammonia complex
more strongly than did the sterically hindered and delocalized
fluorenide anion. These results demonstrate that hydrogen
bonding has a significant effect on the energetics of this system
in solution. Unfortunately, the N-H stretching vibrations in the
IR spectra of these complexes were too weak and complicated
to be of use in investigating hydrogen bonding in these ion pairs.
reversible proton transfer to 2 (Scheme 2) rather than any
pathway requiring coordination of the substrate to the metal
center.
The rate of insertion of dicyclohexyl carbodiimide into an
NH bond of complex 2 is also independent of [PMe3] (Figure
3). This reaction is rapid at room temperature despite the steric
demands of the cyclohexyl groups, and the corresponding
reaction of di-p-tolyl carbodiimide is rapid even at -80 °C.
This is consistent with a mechanism involving direct nucleo-
philic attack by the NH2 group on the electrophilic carbon
(Scheme 7), which proceeds more rapidly at the more electro-
philic (and less sterically congested) carbon in the aryl
substituted carbodiimide. This mechanism is also consistent with
the formation of an N-H insertion product rather than the
product of M-N insertion that would be anticipated to result
from intramolecular NH2 migration (Scheme 7, bottom path).
While these two isomers could potentially interconvert by an
intramolecular proton transfer and displacement, such a reaction
would be expected to be slow at low temperature in the presence
of excess phosphine. Similar regiochemistry consistent with
N-H insertion is observed in the reaction of diphenyl acetylene
with DMPE analogue 1.23
Alkyl Halide Reactions. Unlike complex 1, amide 2 reacts
cleanly with a variety of alkyl halides to yield either E2 or SN2
products. In the case of the E2 reactions, complex 2 behaves in
a manner expected for a more conventional base such as KOtBu.
An investigation of the stereochemistry of E2 eliminations
involving 2 conducted using the rac and meso isomers of 2,3-
dibromobutane showed that both isomers reacted to give almost
exclusively the anti elimination product. In this regard, complex
2 behaves no differently than a more conventional base.33 These
reactions do, however, also yield organometallic products of
the type [(PMe3)4RuH(NH3)+][X-], and ultimately their dis-
placement products (PMe3)4Ru(H)(X). In the case of the reaction
between 2 and trifluoroethane, this provides a novel route to a
relatively rare late metal fluoride species, 17 (eq 13). In this
reaction, no olefin or other simple organic product is observed;
we attribute this to some combination of the volatility and
reactivity of 1,1-difluoroethylene. We have independently
confirmed that 1,1-difluoroethylene reacts with amide complex
2 to yield multiple products.
Various scales have been developed for evaluating the extent
to which “noncoordinating” anions donate electron density to
metal centers,35-39 but fewer directly compare these anions in
terms of their abilities as hydrogen bonding acceptors. Such
evaluation is clearly important in any reactions featuring such
anions and weakly acidic metal species such as aquo or amine
complexes. We investigated the relative hydrogen bonding
abilities of BF4-, BPh4-, and BArf- using the equilibrium in
eq 19 and found BF4 to engage in significantly more H bonding
than the other two anions and the fluorinated borate BArf- to
engage in the least. All three complexes favored the pentaphos-
phino displacement product much more than did triflate or
fluorenide anions.
Conclusion
In summary, we have found that the amide complex (PMe3)4-
Ru(H)(NH2) (2) can be conveniently prepared by deprotonation.
The complex is slightly less basic than DMPE analogue 1,
although it is still sufficiently basic to deprotonate fluorene,
isomerize cyclohexadiene, and serve as a potent base for E2
eliminations. Complex 2 is also nucleophilic, reacting with ethyl
bromide to yield ethylamine and with carbodiimides and
diphenylallene to yield N-H insertion products, even at
temperatures well below room temperature. Neither the basic
nor nucleophilic reactivity of the complex involves phosphine
dissociation, and it appears instead to originate directly at the
ruthenium-bound NH2 group. Efforts to exploit the reactivity
The SN2 reaction of complex 2 with ethyl bromide shows
-
that it is possible to use complex 2 to deliver an NH2 unit to
an organic substrate. This is typically a difficult transformation
because of the unavailablility of soluble NH2- sources and the
propensity for over-alkylation in the reactions of alkyl halides
with ammonia.34 Efforts to exploit this reactivity to effect the
(34) March, J. AdVanced Organic Chemistry; 4th ed.; John Wiley and Sons:
New York, 1995.
(35) Chen, E. Y.-X.; Marks, T. J. Chem. ReV. 2000, 100, 1391.
(36) Reed, C. A. Acc. Chem. Res. 1998, 31, 133.
(37) Siedle, A. R.; Hangii, B.; Newmark, R. A.; Mann, K. R.; Wilson, T.
Makromol. Chem., Macromol. Symp. 1995, 89, 299.
(38) Strauss, S. H. Chem. ReV. 1993, 93, 927.
(33) Bordell, F. G.; Landis, P. S. J. Am. Chem. Soc. 1957, 79, 1593.
(39) Bochmann, M. Angew. Chem., Int. Ed. Engl. 1992, 31, 1181.
9
14690 J. AM. CHEM. SOC. VOL. 124, NO. 49, 2002

Reactivity of Parent Amido Ruthenium Complexes
A R T I C L E S
recrystallized from pentane at -35 °C to yield 6 (65 mg, 65% yield)
as a white solid. The spectroscopic features of this complex matched
those reported in the literature.27
of this complex as an H/D exchange catalyst and soluble NH2
source are underway.
Experimental Section
cis-(PMe3)2(DMPE)Ru(H)(NH2) (7). Amide complex 2 (276 mg,
654 µmol) was dissolved in benzene (5 mL), and DMPE (100 mg, 666
µmol) was added. The solution was allowed to stand at room
temperature for 24 h, after which the volatile materials were removed
in vacuo and the oily residue was redissolved in benzene (5 mL). After
an additional 24 h, the volatile materials were again removed, yielding
General Procedures and Materials. General procedures have been
described elsewhere.40 In NMR assignments, phosphine groups are
identified by a subscript indicating the element to which they are
oriented trans. 1,3,5-trimethoxybenzene (Aldrich, recrystallized from
pentane) was employed as an internal standard in NMR tube reactions.
Gas chromatography (GC) was performed using a Hewlett-Packard
model 5890 series II gas chromatograph equipped with a 30-m HP-1
column. Chromatographic separations were achieved using isothermal
conditions (50 °C) and a constant flow rate of 1 mL/min. The X-ray
structure determination was performed by Dr. Fred Hollander and Dr.
Allen Oliver at the UCB CHEXRAY facility.
Unless otherwise indicated, all reagents were purchased from
commercial suppliers and used as received. Solvent purification methods
have been described elsewhere.40,41 All alkyl halides were degassed
and dried over 4-Å molecular sieves for 2 d and then distilled prior to
use. Aniline and cresol were dried over 4-Å molecular sieves. Fluorene
was recrystallized from pentane prior to use. Trimethylphosphine
(Aldrich) was vacuum transferred from sodium metal prior to use.
Complexes 1,21 3,25 and 9,23 diphenyl allene,42 NaBArf,43 and dibro-
mobutanes44 were prepared according to published procedures.
cis-(PMe3)4Ru(H)(NH2) (2). In two vials, ammonia complex 3 (2.16
g, 2.91 mmol) and KN(SiMe3)2 (0.585 g, 2.94 mmol) were each
dissolved separately in THF (10 mL each). A stirbar was added to the
solution of the ammonia complex, and the base solution was then added
dropwise to the stirred solution of 3. This resulted in the formation of
flocculent white material and a slight yellowing of the previously pale
tan solution. The mixture was stirred for 1 h and then filtered through
Celite. After the yellow filtrate was collected, its volatile materials were
removed in vacuo. The yellow residue was extracted with pentane (2
× 10 mL), and the extract was concentrated to 3 mL in vacuo. This
solution was chilled to -35 °C, and three batches of off-white crystals
1
95% pure 7 according to H and 31P NMR spectra. The oily residue
was extracted with pentane (5 mL), and the extract was concentrated
to 0.3 mL in vacuo and crystallized at -35 °C to yield 7 (103 mg,
1
38% yield) as an off-white oily solid after 2 d. H NMR (C6D6): δ
1.71, (3H, d, J ) 8.6 Hz, P(CH3)), 1.40 (4H, m, P(CH2)2P), 1.33 (9H,
dd, J ) 6.3, 1.7 Hz, P(CH3)3), 1.20 (6H, d, J ) 5.9 Hz, 2 × P(CH3)),
1.12 (3H, d, J ) 7.8 Hz, P(CH3)), 1.02 (9H, d, J ) 7.72 Hz, P(CH3)3),
-2.51 (2H, s, RuNH2), -7.51 (1H, dq, J ) 99.6, 24.8 Hz, RuH) ppm.
31P{1H} NMR (C6D6): δ 44.11 (ddd, J ) 323, 27, 12 Hz, P(CH2)2PP),
30.03 (q, J ) 18 Hz, P(CH2)2PH), 3.67 (br, (CH3)3PN), 0.04 (dt, J )
323, 18 Hz, (CH3)3PP) ppm. 13C{1H} NMR (C6D6): δ 34.3 (m, P(CH3)),
28.6 (m, P(CH3)3), 27.6 (m, P(CH3)3), 22.6 (m, P(CH3)), 21.0 (m,
P(CH3)), 15.8 (m, P(CH3)) ppm. IR (C6H6, cm-1): 3332 (w), 3261
(w), 2967 (s), 2901 (s), 1794 (s), 1420 (s), 1295 (s), 1273 (s), 1039
(s), 924 (s). MS m/z (EI): 421 (M+), 404 (M - NH3). Anal. Calcd for
C12H37NP4Ru: C, 34.28; H, 8.80; N, 3.33. Found: C, 34.23; H, 9.12;
N, 3.01.
[cis-(PMe3)4Ru(H)(NH3)+][C13H9-] (8). Amide complex 2 (432 mg,
1.02 mmol) was dissolved in pentane (10 mL), and a pentane solution
of fluorene (183 mg, 1.10 mmol) was added dropwise with swirling.
Orange powder formed during the addition, and this was collected by
filtration and recrystallized from THF (10 mL) layered with pentane
(10 mL) at -35 °C. After 24 h, red-orange crystals of 8 (530 mg, 89%
yield) were collected. 1H NMR (THF-d8): δ 7.85 (d, J ) 7.6 Hz, 2H,
Ar), 7.26 (d, J ) 8.0 Hz, 2H, Ar), 6.78 (t, J ) 7.6 Hz, 2H, Ar), 6.40
(t, J ) 7.2 Hz, 2H, Ar), 5.95 (s, 1H, Ar), 1.17 (9H, d, J ) 7.6 Hz,
PN(CH3)3), 1.08 (18H, t, J ) 2.8 Hz, PP(CH3)3), 0.96 (9H, d, J ) 5.6
Hz, PH(CH3)3), -0.24 (s, 3H, RuNH3), -10.00 (dtd, J ) 92, 32, 20
Hz, 1H, RuH) ppm. 31P{1H} NMR (THF-d8): δ 13.57 (dt, J ) 37, 23
Hz, PN), -4.73 (dd, J ) 37, 23 Hz, PP), -15.72 (q, J ) 23 Hz, PH)
ppm. 13C{1H} NMR (THF-d8): δ 138.1 (s, Ar), 123.3 (s, Ar), 119.9
(s, Ar), 119.5 (s, Ar), 117.0 (s, Ar), 108.8 (s, Ar), 83.7 (s, Ar), 25.9 (d,
J ) 23.9 Hz, PN(CH3)3), 22.7 (td, J ) 13.5, 3.4 Hz, PP(CH3)3), 21.7 (d,
J ) 17.9 Hz, PH(CH3)3). IR (THF, cm-1): 3397 (w), 3340 (w), 1799
(s), 1658 (m), 1596 (s). MS m/z (FAB): 424 (M - C13H9). Anal. Calcd
for C25H49NP4Ru: C, 51.01; H, 8.39; N, 2.38. Found: C, 51.02; H,
8.14; N, 2.29.
1
of 2 (0.738 g, 60% yield) were collected over the course of 8 d. H
NMR (C6D6): δ 1.37 (18H, t, J ) 2.6 Hz, PP(CH3)3), 1.14 (9H, d, J )
6.4 Hz, PN(CH3)3), 1.12 (9H, d, J ) 5.0 Hz, PH(CH3)3), -2.39 (2H, s,
RuNH2), -8.49 (1H, dtd, J ) 99.2, 29.6, 24.4 Hz, RuH) ppm. 31P{1H}
NMR (C6D6): δ 3.12 (td, J ) 26, 21 Hz, PN), -1.50 (t, J ) 26 Hz,
PP), -11.04 (q, J ) 22 Hz, PH) ppm. 13C{1H} NMR (C6D6): δ 28.8
(dd, J ) 20, 2 Hz, PN(CH3)3), 23.4 (td, J ) 14, 4 Hz, PP(CH3)3), 22.3
(dq, J ) 14, 2 Hz, PH(CH3)3) ppm. IR (C6H6, cm-1): 3364 (w), 2964
(s), 2903 (s), 1808 (s), 1418 (s), 1292 (s), 1275 (s), 937 (s). MS m/z
(EI): 423 (M), 406 (M - NH3), 347 (M - PMe3). Anal. Calcd for
C12H39NP4Ru: C, 34.12; H, 9.30; N, 3.31. Found: C, 34.02; H, 9.10;
N, 2.98.
Reaction of 8 with 1. Complex 1 (12 mg, 29 µmol) and internal
standard (1 mg, 6 µmol) were dissolved in THF-d8 (0.5 mL) and
transferred to an NMR tube, and an initial 1H NMR spectrum was
acquired. Complex 8 (12 mg, 29 µmol) was then added to the sample,
and new 31P and 1H NMR spectra were acquired within 10 min. These
showed complete conversion to complexes 2 (1H NMR (THF-d8): δ
-8.86 (dq, RuH) ppm; 31P{1H} NMR (THF-d8) δ 3.65 (td, PN) ppm)
and 9 (1H NMR (THF-d8) δ -20.32 (dq, RuH) ppm; 31P{1H} NMR
(THF-d8) δ 43.08 ppm).45
cis-(PMe3)4Ru(H)(OH) (5). Complex 2 (25 mg, 59 µmol) was
dissolved in C6D6 (1 mL), and the solution was transferred to an NMR
tube which was capped with a septum. Degassed H2O (1.0 µL, 56 µmol)
was then added by syringe, and after 5 min, H and 31P NMR spectra
1
were acquired showing quantitative conversion of 2 to hydroxide
complex 5. The spectroscopic features of this complex matched those
reported in the literature.26
cis-(PMe3)3Ru(CH2PMe2)(H) (6). Complex 2 (104 mg, 246 µmol)
was dissolved in THF (5 mL), and Ph2NH (14 mg, 83 µmol) was added.
The solution was heated to 45 °C for 4 h, after which time the volatile
materials were removed in vacuo to yield a white residue. This was
[cis-(PMe3)4Ru(H)(NH3)+][OTol-] (10). Complex 2 (186 mg, 443
µmol) was dissolved in pentane (2 mL), and p-cresol (48 mg, 0.44
µmol) was added as a pentane solution (1 mL). The solution was then
chilled to -35 °C, resulting in the formation of white precipitate 10
(210 mg, 90% yield). Elemental analysis was consistent with this
material being pure, although in solution phase 10 was always
contaminated with displacement product 11. Addition of 1 atm of NH3
(40) Alaimo, P. J.; Arndtsen, B. A.; Bergman, R. G. Organometallics 2000, 19,
2130.
(41) Alaimo, P. J.; Peters, D. W.; Arnold, J.; Bergman, R. G. J. Chem. Educ.
2001, 78, 64.
1
(42) Elsevier, C. J.; Stehouwer, P. M.; Westmijze, H.; Vermeer, P. J. Org. Chem.
to these solutions suppressed this contamination. H NMR (THF-d8):
1983, 48, 1103.
(43) Brookhart, M.; Grant, B.; Volpe, A. F. Organometallics 1992, 11, 3920.
(44) Young, W. G.; Dillon, R. J.; Lucas, H. J. J. Am. Chem. Soc. 1929, 51,
2528.
(45) The identification of this compound was confirmed by comparison to an
authentic sample.
9
J. AM. CHEM. SOC. VOL. 124, NO. 49, 2002 14691

A R T I C L E S
Holland and Bergman
6.50 (2H, d, J ) 7.5 Hz, Ar), 5.93 (2H, d, J ) 7.5 Hz, Ar), 2.40 (3H,
s, RuNH3), 2.10 (3H, s, ArCH3), 1.47 (18H, t, J ) 3.0 Hz, PP(CH3)3),
1.46 (9H, d, J ) 6.6 Hz, PN(CH3)3), 1.39 (9H, d, J ) 8.2 Hz, PH(CH3)3),
-9.46 (dtd, J ) 92, 32, 20 Hz, 1H, RuH) ppm. 31P{1H} NMR (THF-
d8): δ 14.36 (dt, J ) 38, 24 Hz, PN), -3.90 (dd, J ) 38, 24 Hz, PP),
-14.32 (q, J ) 24 Hz, PH) ppm. 13C{1H} NMR (THF-d8): δ 169.7 (s,
Ar), 129.9 (s, Ar), 120.5 (s, Ar), 117.6 (s, Ar), 26.3 (d, J ) 26 Hz,
PN(CH3)3), 23.3 (td, J ) 14, 3 Hz, PP(CH3)3), 22.4 (d, J ) 18 Hz,
PH(CH3)3), 21.8 (s, ArCH3) ppm. IR (THF, cm-1): 3401 (w), 3342
(w), 1798 (s), 1655 (m), 1597 (s). MS m/z (FAB, sulfolane): 424 (M
- OTol). Anal. Calcd for C19H47NOP4Ru: C, 43.01; H, 8.93; N, 2.64.
Found: C, 43.06; H, 8.76; N, 2.30.
cis-(PMe3)4Ru(H)(OTol) (11). Complex 2 (204 mg, 483 µmol) and
cresol (53 mg, 491 µmol) were dissolved in C6H6 (5 mL), and the
solution and a stirbar were loaded into a glass vessel equipped with a
Teflon stopcock. The solution was degassed and stirred at room
temperature for 24 h. After this time, the volatile materials were
removed in vacuo, and the remaining off-white solid was recrystallized
from pentane (2 mL) at -35 °C for 48 h to yield 11 as white crystals
(174 mg, 70% yield). The spectroscopic features of this complex
matched those reported in the literature.28
of the RuNH2 signal at -2.39 ppm and an NHnD3-n signal at δ -0.17
ppm, accounting for the missing intensity. No other peaks were
significantly changed. After 3 h, only 20% of the original RuNH2 signal
remained, and the P(CH3)3 signals were broadened and less intense.
The 31P NMR showed broadening in the P(CH3)3 signals as well. After
24 h, the P(CH3)3 signals had further decreased in intensity, while the
RuNH2 signal had actually increased slightly. After 72 h, both P(CH3)3
and RuNH2 signals appeared at ∼50% of their original intensities, and
the RuH signal remained unchanged. The solution was evaporated to
dryness in vacuo, and a mass spectrum (EI) of the resulting off-white
oily solid showed a broad envelope centered at m/z ) 441, correspond-
2
ing to ∼50% deuteration. The H{1H} NMR (C6H6) of this material
was consistent with deuteration at all positions except RuH: δ 1.4
(18D), 1.1 (18D), -2.4 (2D) ppm.
cis-(PMe3)4Ru(H)(15NH2) (2-15N). Complex 2 (21 mg, 50 µmol) was
dissolved in C6D6 (300 µL) in an NMR tube equipped with a Teflon
stopcock, the solution was degassed, and 15NH3 (6.6 mL × 141 Torr,
50 µmol) was added by vacuum transfer from a known volume bulb.
The 1H NMR spectrum of the solution after 5 min showed a 1:1 mixture
of free 15NH3 (δ -0.17 (d, J ) 60.8 Hz) ppm) and 14NH3 (δ -0.17 (t,
J ) 42.8 Hz) ppm) and slight broadening in the NH2 signal (δ -2.40
ppm). The solution was briefly degassed under vacuum, and additional
15NH3 (6.6 mL × 412 Torr, 150 µmol) was added. This addition resulted
in the liberation of additional 14NH3 and further broadening of the
cis-(PMe3)4Ru(H)(NHPh) (13). Complex 2 (142 mg, 338 µmol) and
aniline (33 mg, 350 µmol) were dissolved in C6H6 (5 mL), and the
solution was stirred at room temperature for 24 h. After this time, the
volatile materials were removed in vacuo, and the remaining off-white
solid was recrystallized from pentane (2 mL) at -35 °C for 48 h to
yield 13 (101 mg, 59% yield) as a yellow-white solid. The spectroscopic
features of this complex matched those reported in the literature.28
H/D Exchange between 2 and Toluene-d8. Complex 2 (18 mg, 43
µmol) and internal standard were dissolved in toluene-d8 (300 µL),
and the mixture was divided evenly between two NMR tubes. Both
solutions were degassed, and PMe3 (66 mL × 12 Torr, 43 µmol) was
added to one tube by vacuum transfer from a known volume bulb.
Both tubes were flame sealed under vacuum, heated to 45 °C, and
monitored periodically by 1H and 31P NMR spectroscopy. Slow
decreases in the intensities of the P(CH3)3 and RuNH2 resonances were
observed in both samples and took place at approximately the same
initial rate in the two samples. In the sample containing PMe3,
isotopomers PMe3-dn (δ -57.58 - (0.33)n ppm) were observed in the
31P NMR spectrum, and an increasing degree of deuterium incorporation
in this group was observed as the reaction progressed. Deuteration of
the RuH was not observed. After 3 d, the samples were evaporated to
dryness in vacuo. C6H6 (300 µL) was added, and deuterium incorpora-
1
RuNH2 signal in the H NMR spectrum. The volatile materials were
then removed, and 2-15N (20 mg, 96% yield) was collected as a pale
yellow oil. EIMS m/z: ) 424 (M+).
Reaction of 2 with D2. Amide complex 2 (19 mg, 45 µmol) was
dissolved in THF-d8 (1 mL), and this solution was divided evenly
between two NMR tubes, each equipped with a Teflon stopcock. The
tubes were then degassed on the vacuum line, and PMe3 (29 mL × 12
Torr, 19 µmol) was added to one of the samples by vacuum transfer
from a known volume bulb. Both tubes were then charged with D2
(754 Torr) and closed. The samples were shaken and kept at 22 °C
and monitored by 1H NMR spectroscopy. After 2 h, the solution
containing no PMe3 had undergone 40% conversion to P4RuHD28 (1H
NMR (THF-d8) δ 1.31 (36H, br, P(CH3)3), -10.13 (1H, m, RuH); 2H-
{1H} NMR (THF-d8) δ - 10.1 (br); 31P{1H} NMR (THF-d8) δ 0.7 (t),
-6.7 (t) ppm) and NH2D (δ 0.34 (2H, t) ppm), and the sample
containing PMe3 had undergone <5% conversion to the same products.
After 8 h, the respective extents of conversion in the two tubes were
85 and 5%, respectively. In neither case was any deuteration of
remaining 2 or additional deuteration of P4RuHD observed.
[cis-(PMe3)4Ru(H)(NH3)+][F-] (14). Complex 2 (253 mg, 602
µmol) was dissolved in benzene (5 mL), and the solution and a stirbar
were loaded into a glass vessel equipped with a Teflon stopcock. The
solution was degassed, and 1,1,1-trifluoroethane (66 mL × 170 Torr,
613 µmol) was added via vacuum transfer from a known volume bulb.
The stopcock was then closed, and the solution was stirred at room
temperature for 48 h, during which time the color changed from pale
yellow to bright orange. The sample was evaporated to dryness in vacuo,
then dissolved in THF (2 mL), layered with pentane (10 mL), and
chilled to -35 °C. Complex 14 (156 mg, 59% yield) was collected as
an off-white solid after 24 h. 1H NMR (THF-d8): δ 3.94 (3H, s,
RuNH3), 1.53 (18H, s, PP(CH3)3), 1.49 (9H, d, J ) 6.2 Hz, PN(CH3)3),
1.32 (9H, d, J ) 8.2 Hz, PH(CH3)3), -9.77 (dtd, J ) 89.2, 32.4, 20.0
Hz, 1H, RuH) ppm. 19F NMR (THF-d8): δ -82.8 (br) ppm.31P{1H}
NMR (THF-d8): δ 13.90 (dt, J ) 36, 28 Hz, PN), -2.55 (dd, J ) 36,
23 Hz, PP), -12.37 (q, J ) 24 Hz, PH) ppm. 13C{1H} NMR (THF-d8):
δ 26.8 (dq, J ) 25, 3 Hz, PN(CH3)3), 24.0 (td, J ) 15, 3 Hz, PP(CH3)3),
23.2 (d, J ) 17 Hz, PH(CH3)3) ppm. IR (THF, cm-1): 3395 (w), 3334
(w), 1799 (s). Anal. Calcd for C12H40FNP4Ru: C, 32.57; H, 9.11; N,
3.17. Found: C, 32.20; H, 8.84; N, 2.89.
2
tion was confirmed by H{1H} NMR spectroscopy: δ 1.4 (18D), 1.1
(18D), -2.4 (2D) ppm.
Isomerization of 1,3-Cyclohexadiene by 2. A solution of 1,3-
cyclohexadiene (7.0 mg, 88 µmol), 2 (9.2 mg, 22 µmol), and internal
standard (3.0 mg, 18 µmol) in C6D6 (580 µL) was quickly prepared
and divided evenly between two NMR tubes. The tubes were fitted to
Cajon adapters and frozen at -196 °C within 2 min of mixing the
diene and 2. PMe3 (6.6 mL × 3.2 Torr, 2.6 µmol; 6.6 mL × 90 Torr,
32 µmol) was then added to each tube via vacuum transfer from a
known volume bulb, and the tubes were sealed under vacuum. The
samples were then kept in a water bath at 22 °C, and the ratio of 1,3-
cyclohexadiene (δ 5.86 (2H, d), 5.67 (2H, d), 1.95 (3H, s) ppm) and
1,4 cyclohexadiene (δ 5.60 (4H, s), 2.51 (3H, s) ppm)45 was monitored
1
by H NMR over the course of several hours. No loss of diene or 2
was observed during this time. The results are shown in Figure 1.
cis-(PMe3)4Ru(H)(NH2)-dn via ND3. Amide complex 2 (10 mg, 24
µmol) and internal standard were dissolved in C6D6, (0.5 mL) and this
solution was loaded into an NMR tube equipped with a Teflon stopcock.
After an initial 1H NMR spectrum was acquired, the solution was
degassed on the vacuum line, ND3 (6.6 mL × 720 Torr, 260 µmol)
was added by vacuum transfer from a known volume bulb, and the
Reaction of 2 with PhCH2CH2Cl. Complex 2 (23 mg, 55 µmol)
was dissolved in THF-d8 (300 µL), 1-phenyl-2-chloroethane (7.7 mg,
55 µmol) was added, and the solution was loaded into an NMR tube.
1
tube was closed. A H NMR spectrum of the resulting solution was
acquired within 5 min and showed a 30% decrease in the integration
9
14692 J. AM. CHEM. SOC. VOL. 124, NO. 49, 2002

Reactivity of Parent Amido Ruthenium Complexes
A R T I C L E S
1
The H NMR spectrum of this solution after 30 min showed equal
) 34, 26 Hz, PP), -6.78 (p, J ) 34 Hz, PH) ppm. 13C{1H} NMR (THF-
d8): δ 27.5 (d, J ) 25 Hz, P(CH3)3), 22.4 (td, J ) 15, 4 Hz, P(CH3)3),
20.0 (dq, J ) 15, 2 Hz, P(CH3)3) ppm. 19F NMR (THF-d8): δ -396.1
(dq, J ) 150, 35 Hz). IR (KBr): 1803 (s) cm-1. MS m/z (EI): 426
(M+), 350 (M - PMe3). Anal. Calcd for C12H37FP4Ru: C, 33.88; H,
8.77. Found: C, 34.00; H, 8.94.
amounts of styrene (1H NMR (THF-d8) δ 7.3-7 (5H, m, Ph), 6.56
(1H, dd, PhCHdCH2), 5.59 (d, PhCHdCHH), 5.05 (1H, d, PhCHd
CHH)) and an organometallic complex with spectroscopic character-
istics consistent with an ammonia cation complex 15. Efforts to isolate
complex 15 from this product mixture were unsuccessful. Over the
course of 24 h, the resonances assigned to 15 decreased in intensity
and were replaced by those for NH3 displacement product 18. The
volatile materials were then removed in vacuo, and the residue was
recrystallized from pentane (0.5 mL) at -35 °C for 24 h to yield 18
cis-(PMe3)4Ru(H)Br (19). Complex 16 (70 mg, 140 µmol) was
dissolved in C6D6 (0.5 mL), and the solution was loaded into an NMR
tube. The tube was affixed to a Cajon adaptor, degassed, and flame
sealed under vacuum. The sample was heated for 24 h at 45 °C, yielding
19 and NH3 as the only products observed by NMR spectroscopy. The
volatile materials were removed in vacuo, and the residue was
recrystallized from THF (0.5 mL) and pentane (3 mL) to yield 19 (55
1
(14 mg, 55% yield). 15: H NMR (THF-d8) δ 2.87 (s, 3H, RuNH3),
1.52 (18H, t, J ) 3.1 Hz, P(CH3)3), 1.49 (9H, d, J ) 6.2 Hz, P(CH3)3),
1.35 (9H, d, J ) 7.6 Hz, P(CH3)3), -9.63 (1H, dtd, J ) 92.1, 31.8,
20.8 Hz, RuH) ppm; 31P{1H} NMR (THF-d8) δ 14.91 (td, J ) 37, 24
Hz, PN), -2.85 (dd, J ) 36, 23 Hz, PP), -13.75 (q, J ) 23 Hz, PH)
ppm. 18: 1H NMR (C6D6) 1.49 (18H, t), 1.24 (9H, d), 1.09 (9H, d),
-8.51 (1H, dq) ppm; 31P{1H} NMR (THF-d8) δ 16.65 (td), -4.85 (dd),
-16.61 (q) ppm; Lit.46 1H NMR (C6D6) 1.48 (18H, t), 1.24 (9H, d),
1.09 (9H, d), -8.50 (1H, dq) ppm; 31P{1H} NMR (THF-d8) δ 16.7
(td), -4.9 (dd), -16.6 (q) ppm.
1
mg, 81% yield) as yellow crystals. H NMR (C6D6): δ 1.48 (18H, t,
J ) 2.7 Hz, P(CH3)3), 1.19 (9H, d, J ) 5.5 Hz, P(CH3)3), 1.04 (9H, d,
J ) 7.8 Hz, P(CH3)3), -8.90 (1H, dq, J ) 103.2, 29.6 Hz, RuH) ppm.
31P{1H} NMR (C6D6): δ 17.95 (td, J ) 34, 19 Hz, PBr), -7.02 (dd, J
) 34, 26 Hz, PP), -19.55 (q, J ) 23 Hz, PH) ppm. 13C{1H} NMR
(C6D6): δ 28.2 (dq, J ) 24, 3 Hz, P(CH3)3), 24.8 (td, J ) 14, 3 Hz,
P(CH3)3), 22.9 (dq, J ) 17, 2 Hz, P(CH3)3) ppm. IR (KBr): 2994 (s),
1805 (s) cm-1. MS m/z (EI): 487 (M+). Anal. Calcd for C12H37BrP4-
Ru: C, 29.63; H, 7.67. Found: C, 29.88; H, 7.74.
[cis-(PMe3)4Ru(H)(NH3)+][Br-] (16). Complex 2 (105 mg, 249
µmol) and rac-2,3-dibromobutane (56 mg, 259 µmol) were each
dissolved in pentane (1 mL each), and the bromobutane solution was
then added dropwise to 2. The resulting solution was chilled to -35
°C for 4 h, yielding 16 (82 mg, 65% yield) as a white solid. 1H NMR
(THF-d8): δ 2.57 (3H, s, RuNH3), 1.53 (18H, t, J ) 3.1 Hz, P(CH3)3),
1.48 (9H, d, J ) 6.4 Hz, P(CH3)3), 1.36 (9H, d, J ) 7.4 Hz, P(CH3)3),
-9.60 (dtd, J ) 92, 32, 20 Hz, 1H, RuH) ppm. 31P{1H} NMR (THF-
d8): δ 15.23 (dt, J ) 37, 24 Hz, PN), -2.65 (dd, J ) 36, 23 Hz, PP),
-13.98 (q, J ) 23 Hz, PH) ppm. 13C{1H} NMR (THF-d8): δ 26.2 (d,
J ) 26 Hz, P(CH3)3), 23.2 (td, J ) 14, 3 Hz, P(CH3)3), 22.2 (d, J ) 17
Reaction of 2 with CH3CH2Br. Complex 2 (40 mg, 100 µmol)
and ethyl bromide (12 mg, 110 µmol) were each dissolved separately
in THF-d8 (0.25 mL). The solution of ethyl bromide was then added
dropwise to 2, and the resulting solution was loaded into an NMR tube
1
capped with a Teflon stopcock. The H and 13C NMR spectra of this
solution showed ethylamine (20) (1H NMR (THF-d8) δ 2.63 (2H, q, J
) 7.0 Hz, CH3CH2NH2), 1.37 (2H, br, CH3CH2NH2), 1.01 (3H, t, J )
7.0 Hz, CH3CH2NH2) ppm; 13C{1H} NMR (THF-d8) δ 37.9 (s, CH3CH2-
NH2), 19.9 (s, CH3CH2NH2) ppm)45 and ethylene (δ 5.37 ppm) in a
10:1 molar ratio, as well as a trace of ammonia. Complex 19 and a
small amount of 16 were the only organometallic products observed.
Ethylamine was readily separated from the organometallic products
by vacuum transfer of the volatile materials.
Hz, P(CH3)3) ppm. IR (THF): 3400 (w), 3341 (w), 11819 (s) cm-1
.
MS m/z (FAB, sulfolane): 424 (M - Br). Anal. Calcd for C12H40-
BrNP4Ru: C, 28.63; H, 8.01; N, 2.78. Found: C, 28.39; H, 8.16; N,
2.64.
Dehydrobromination of meso- and rac-2,3-Dibromobutane by 2.
In a typical experiment, amide complex 2 (8.4 mg, 20 µmol) and rac-
2,3-dibromobutane (4.4 mg, 20 µmol) were each dissolved separately
in C6D6 (0.25 mL), and then the base solution was added dropwise to
the stirred dibromobutane solution. The resulting solution was trans-
cis-(PMe3)4Ru(H)(NHC(NTol)(NHTol)) (21). Complex 2 (245 mg,
580 µmol) was dissolved in toluene (3 mL), and di-p-tolyl carbodiimide
(135 mg, 608 µmol) was added. The solution was allowed to stand at
room temperature for 3 h, layered with pentane (10 mL), and chilled
to -35 °C for 48 h to yield 21 (231 mg, 62% yield) as off-white
crystals. 1H NMR (C6D6): δ 9.02 (1H, s, RuNH), 8.38 (2H, d, J ) 8.4
Hz, Ar), 7.30 (2H, d, J ) 8.1 Hz, Ar), 7.24 (2H, d, J ) 8.1 Hz, Ar),
7.16 (2H, d, J ) 8.4 Hz, Ar), 2.35 (3H, s, ArCH3), 2.21 (3H, s, ArCH3),
1.10 (18H, t, J ) 5.6 Hz, PP(CH3)3), 0.92 (9H, d, J ) 7.4 Hz, PN(CH3)3),
0.74 (9H, d, J ) 7.2 Hz, PH(CH3)3), -8.57 (1H, dq, J ) 90.0, 27.2 Hz,
RuH) ppm. 31P{1H} NMR (C6D6): δ 5.37 (td, J ) 29.2, 22.3 Hz, PN),
-2.81 (t, J ) 29.2 Hz, PP), -11.27 (q, J ) 24.3 Hz, PH) ppm. 13C-
{1H} NMR (C6D6): δ 158.0 (s, CN3), 154.3 (s, Ar), 143.0 (s, Ar),
130.1 (s, Ar), 129.7 (s, Ar), 127.2 (s, Ar), 127.1 (s, Ar), 126.1 (s, Ar),
117.7 (s, Ar), 27.3 (d, J ) 24.6 Hz, PN(CH3)3), 23.2 (td, J ) 13.4, 2.4
Hz, PP(CH3)3), 21.8 (d, J ) 17.0 Hz, PH(CH3)3), 21.6 (s, ArCH3), 21.3
(s, ArCH3) ppm. IR (KBr): 3232 (w), 3216 (w), 2998 (m), 2971 (m),
2907 (m), 1803 (s), 1614 (s), 1591 (s), 1512 (s), 1428 (m), 1313 (m)
945 (s), 855 (m), 712 (m). MS m/z (EI): 645 (M+), 569 (M - PMe3).
Anal. Calcd for C27H53N3P4Ru: C, 50.30; H, 8.29; N, 6.52. Found: C,
50.34; H, 8.15; N, 6.41.
1
ferred into an NMR tube and the H NMR spectrum of the mixture
showed Z-2-bromobutene (δ 5.21 (1H, q, J ) 7.3 Hz), 1.97 (3H, s),
0.84 (3H, d, J ) 7.3 Hz) ppm)45 to be the only significant product.
The experiment was repeated in o-xylene and filtered through a plug
of silica, and GC analysis of the product showed Z (tr ) 1.50 min) and
E (tr ) 1.40 min) isomers in a 98:2 ratio. The analogous experiments
using meso-2,3-dibromobutane produced the E isomer according to 1H
NMR spectroscopy (δ 5.71 (1H, q, J ) 6.0 Hz), 1.85 (3H, s), 1.12
(3H, d, J ) 6.0 Hz) ppm)45 and the Z and E isomers in a 3:97 ratio by
GC analysis. In both NMR experiments, 16 was the only initial
organometallic product observed, and it was replaced by 19 and NH3
over the course of 48 h.
cis-(PMe3)4Ru(H)F (17). Complex 14 (50 mg, 110 µmol) was
dissolved in C6D6 (0.5 mL), and the solution was loaded into an NMR
tube. The tube was affixed to a Cajon adaptor, degassed, and flame
sealed under vacuum. The sample was heated for 24 h at 45 °C, yielding
17 and NH3 as the only products observed by NMR spectroscopy. The
volatile materials were removed in vacuo, and the residue was
recrystallized from THF (0.5 mL) and pentane (5 mL) to yield 17 (35
mg, 72% yield) as yellow crystals.1H NMR (C6D6): δ 1.38 (18H, br,
P(CH3)3), 1.19 (9H, d, J ) 6.4 Hz, P(CH3)3), 0.98 (9H, d, J ) 7.6 Hz,
P(CH3)3), -7.61 (1H, dq, J ) 104.0, 29.6 Hz, RuH) ppm. 31P{1H}
NMR (THF-d8): δ 18.86 (dtd, J ) 150, 34, 18 Hz, PF), -0.29 (dt, J
X-ray Diffraction Study of 21. Colorless plates of 21 were grown
from THF/pentane at -30 °C. A fragment measuring 0.35 × 0.30 ×
0.09 mm3 was mounted on a glass fiber using Paratone N hydrocarbon
oil and studied using a SMART CCD area detector with graphite
monochromated Mo KR radiation. The structure was solved by heavy-
atom Patterson methods and expanded using Fourier techniques. Some
nonhydrogen atoms were refined anisotropically, while the rest were
refined isotropically. The hydride hydrogens and the hydrogens on the
nitrogens were located as peaks in reasonable locations on a difference
(46) Jones, R. A.; Real, F. M.; Wilkinson, G.; Galas, A. M. R.; Hursthouse, M.
B.; Malik, K. M. A. J. Chem. Soc., Dalton Trans. 1980, 511.
9
J. AM. CHEM. SOC. VOL. 124, NO. 49, 2002 14693

A R T I C L E S
Holland and Bergman
1
Fourier map and included without refinement. Other hydrogen atoms
were also included but not refined.
as an orange-red solid. H NMR (C6D6): δ 7.53 (2H, d, J ) 7.8 Hz,
Ar), 7.34 (2H, t, J ) 7.4 Hz, Ar), 7.24 (2H, t, J ) 7.3 Hz, Ar), 7.15
(2H, d, J ) 7.3 Hz, Ar), 7.03 (1H, t, J ) 7.3 Hz, Ar), 6.85 (1H, t, J )
7.5 Hz, Ar), 6.21 (1H, s, CdCH(Ph)), 4.11 (2H, s, CH2(Ph)), 1.72 (1H,
d, J ) 5.7 Hz, RuNH), 1.18 (18H, t, J ) 2.7 Hz, PP(CH3)3), 0.99 (9H,
d, J ) 6.9 Hz, PN(CH3)3), 0.76 (9H, d, J ) 4.8 Hz, PH(CH3)3), -7.78
(1H, dq, J ) 100.0, 28.0 Hz, RuH) ppm. 31P{1H} NMR (C6D6): δ
5.45 (td, J ) 30.8, 17.8 Hz, PN), -1.57 (t, J ) 30.8 Hz, PP), -14.68
(q, J ) 21.6 Hz, PH) ppm. 13C{1H} NMR (C6D6): δ 157.8 (s, CNC2),
154.3 (s, Ar), 146.5 (s, Ar), 141.8 (s, Ar), 131.1 (s, Ar), 128.8 (s, Ar),
128.6 (s, Ar), 126.2 (s, Ar), 125.4 (s, Ar), 117.4 (s, CdCH(Ph)), 43.5
(s, CH2Ph), 27.7 (d, J ) 24.0 Hz, PN(CH3)3), 23.7 (td, J ) 14.6, 4.0
Hz, PH(CH3)3), 21.9 (d, J ) 15.0 Hz, PH(CH3)3) ppm. IR (KBr, cm-1):
3321 (w), 1631 (w), 1805 (s). MS m/z (EI): 629 (M+). Anal. Calcd
for C27H49NP4Ru: C, 52.75; H, 8.36; N, 2.28. Found: C, 52.55; H,
8.34; N, 2.15.
cis-[(PMe3)4Ru(H)(NH3)+][OTf-] (24). Complex 2 (89 mg, 210
µmol) was dissolved in pentane (5 mL), and HOTf (30 mg, 200 µmol)
was added as a pentane solution (2 mL). This resulted in precipitation
of an off-white solid, which was collected by filtration and recrystallized
from THF (3 mL) layered with pentane (10 mL) at -35 °C. After 48
h, 24 (49 mg, 41% yield) was collected as off-white crystals. 1H NMR
(THF-d8): δ 2.01 (3H, s, RuNH3), 1.47 (18H, t, J ) 3.0 Hz, PP(CH3)3),
1.46 (9H, d, J ) 6.6 Hz, PN(CH3)3), 1.39 (9H, d, J ) 8.2 Hz, PH(CH3)3),
-9.46 (dtd, J ) 92, 32, 20 Hz, 1H, RuH) ppm. 31P{1H} NMR (THF-
d8): δ 15.00 (dt, J ) 35, 24 Hz, PN), -2.44 (dd, J ) 35, 24 Hz, PP),
-14.47 (q, J ) 24 Hz, PH) ppm. 19F NMR (THF-d8): δ -75.29 ppm.
13C{1H} NMR (THF-d8): δ 26.2 (d, J ) 27 Hz, PN(CH3)3), 23.2 (td,
J ) 14, 3 Hz, PP(CH3)3), 22.3 (d, J ) 18 Hz, PH(CH3)3). IR (THF,
cm-1): 3401 (w), 1821 (s). Anal. Calcd for C13H40F3NO3P4RuS: C,
27.27; H, 7.04; N, 2.45. Found: C, 27.41; H, 7.25; N, 2.30.
The compound crystallizes in space group P1h (#2) with four
molecules in the triclinic unit cell together with two molecules of
solvent. The theoretical formula consists of one molecule of the complex
and half a molecule of THF. However, the solvent in this crystal appears
to be a mixture of THF and pentane in an approximately 70:30 ratio.
Thus, the calculated cell contents and the model cell contents are slightly
different. One PMe3 ligand was torsionally disordered, but all atoms
refined normally other than those carbons and the atoms in the THF/
pentane solvent region.
The two molecules of 21 in the unit cell are chemically identical
and very similar (but not identical) in terms of their conformations
and bond distances. The orientations of the imine-bound tolyl group
and the PMe3 group trans to the nitrogen are slightly different in the
two molecules. Selected bond lengths and angles for one molecule are
given in Figure 1; all crystallographic data are included in the
Supporting Information. A short summary of crystallographic data
follows. Space group P1h (#2), a ) 11.8161(2) Å, b ) 14.1196(4) Å,
c ) 22.7184(6) Å, V ) 3573.1(2) Å3, Z ) 4, Fcalc ) 1.26 g/cm3, µ(Mo
KR) ) 6.41 cm-1, no. of unique reflections ) 11 447, no. of reflections
with I > 2.50σ(I) ) 6091, R ) 5.2%.
cis-(PMe3)4Ru(H)(NHC(NCy)(NHCy) (22). Complex 2 (145 mg,
343 µmol) was dissolved in diethyl ether (10 mL), and dicyclohexyl
carbodiimide (78 mg, 380 µmol) was added. The solution was allowed
to stand at room temperature for 3 h and was then concentrated to 2
mL in vacuo and chilled to -35 °C. White crystals of 22 (123 mg,
54% yield) were collected after 24 h. 1H NMR (C6D6): δ 5.35 (1H, d,
J ) 7.2 Hz, RuNH), 4.37 (1H, m, Cy), 3.29 (1H, m, Cy), 2.51 (2H, m,
Cy), 2.26 (2H, m, Cy), 2.10 (2H, m, Cy), 1.9-1.3 (14H, m, Cy), 1.25
(18H, t, J ) 2.8 Hz, PP(CH3)3), 1.01 (9H, d, J ) 5.1 Hz, PN(CH3)3),
0.98 (9H, d, J ) 6.9 Hz, PH(CH3)3), -8.46 (1H, dq, J ) 120.8, 26.4
Hz, RuH) ppm. 31P{1H} NMR (THF-d8): δ 5.33 (td, J ) 30.8, 20.4,
PN), -0.77 (t, J ) 30.8 Hz, PP), -14.57 (q, J ) 25.1 Hz, PH) ppm.
13C{1H} NMR (C6D6): δ 159.6 (s, CN3), 57.4 (CdNC), 49.7 (s, NC),
38.2 (s, Cy), 36.2 (s, Cy), 28.0 (s, Cy), 27.6 (d, J ) 28.2 Hz, PN(CH3)3),
27.6 (s, Cy), 27.3 (s, Cy), 26.7 (s, Cy), 23.6 (td, J ) 13.2, 3.6 Hz,
PP(CH3)3), 21.1 (d, J ) 15.9 Hz, PH(CH3)3) ppm. IR (KBr, cm-1): 3267
(w), 2969 (m), 2905 (m), 1807 (s), 1514 (m), 1313 (m), 944 (s), 858
(m), 710 (m). MS m/z (EI): 629 (M+), 553 (M - PMe3). Anal. Calcd
for C25H61N3P4Ru: C, 47.76; H, 9.78; N, 6.68. Found: C, 47.80; H,
9.93; N, 6.80.
Kinetic Studies of the Formation of 22. In a typical experiment,
complex 2 (9.6 mg, 23 µmol) and internal standard were dissolved in
THF-d8 (0.4 mL), and 0.20 mL of this solution was added to an NMR
tube. The tube was affixed to a Cajon adaptor, degassed, and charged
with PMe3 (6.6 mL × 3.0 Torr, 1.1 µmol) via vacuum transfer from a
known volume bulb. The tube was then chilled to -78 °C and opened
to a positive pressure of N2. The stopcock of the Cajon adapter was
then removed, and the dicyclohexyl carbodiimide was added slowly
as a THF-d8 solution (50 µL × 0.44 M, 22 µmol). The carbodiimide
solution was allowed to run down the walls of the tube so that it would
cool before reaching the solution of 2. The walls of the tube were then
rinsed with THF-d8 (50 µL). The stopcock was replaced, the sample
was frozen at -196 °C, and the tube was flame sealed under vacuum.
The sample was then thawed at -78 °C and introduced to a NMR
spectrometer whose probe was precooled to -29 °C. Kinetic data were
acquired at this temperature by monitoring the change in the product
resonance at δ 5.11 ppm (RuNH) and total hydride resonances (two
overlapping overlapping signals, δ 8-9 ppm) relative to internal
standard. The results are reported in Figures 3 and 4.
cis-[(PMe3)4Ru(H)(NH3)+][BF4-] (25). Complex 2 (123 mg, 291
µmol) was dissolved in pentane (3 mL), and HBF4‚Et2O (45 mg, 280
µmol) was added as a pentane solution (1 mL). This resulted in
precipitation of yellow oil, which was collected and recrystallized from
THF (1 mL) layered with pentane (5 mL) at -35 °C. After 48 h, 25
1
(72 mg, 43% yield) was collected as a yellow-beige solid. H NMR
(THF-d8): δ 1.91 (s, 3H, RuNH3), 1.56 (18H, t, J ) 3.2 Hz, PP(CH3)3),
1.46 (9H, d, J ) 6.2 Hz, PN(CH3)3), 1.37 (9H, d, J ) 7.6 Hz, PH(CH3)3),
-9.49 (dtd, J ) 94, 32, 20 Hz, 1H, RuH) ppm. 31P{1H} NMR (THF-
d8): δ 15.00 (dt, J ) 36, 25 Hz, PN), -2.45 (dd, J ) 36, 25 Hz, PP),
-14.55 (q, J ) 25 Hz, PH) ppm. 19F NMR (THF-d8): δ 148.8 (br)
ppm. 13C{1H} NMR (THF-d8): δ 26.4 (d, J ) 26 Hz, PN(CH3)3), 23.1
(td, J ) 15, 3 Hz, PP(CH3)3), 22.3 (d, J ) 18 Hz, PH(CH3)3). IR (THF,
cm-1): 3398 (w), 1818 (s). Anal. Calcd for C12H40BF4NP4Ru: C, 28.24;
H, 7.90; N, 2.75. Found: C, 28.12; H, 7.69; N, 2.69.
cis-[(PMe3)4Ru(H)(NH3)+][BArf-] (26). Complex 3 (143 mg, 193
µmol) was suspended in ether (5 mL), and NaBArf (177 mg, 200 µmol)
was added. The resulting mixture was stirred for 6 h and then filtered
through Celite. The filtrate was layered with pentane (10 mL) and
chilled to -35 °C for 12h, and 26 (133 mg, 54% yield) was collected
as a white solid. 1H NMR (THF-d8): δ 7.80 (8H, br, BArf), 7.59 (4H,
br, BArf), 1.87 (3H, s, RuNH3), 1.57 (18H, t, J ) 3.0 Hz, P(CH3)3),
1.46 (9H, d, J ) 6.0 Hz, P(CH3)3), 1.38 (9H, d, J ) 7.4 Hz, P(CH3)3),
-9.41 (dtd, J ) 92, 33, 21 Hz, 1H, RuH) ppm. 31P{1H} NMR (THF-
d8): δ 15.00 (dt, J ) 36, 25 Hz, PN), -2.42 (dd, J ) 36, 25 Hz, PP),
-14.60 (q, J ) 25 Hz, PH) ppm. 19F NMR (THF-d8): δ -59.61 ppm.
13C{1H} NMR (THF-d8): δ 26.3 (d, J ) 26 Hz, PN(CH3)3), 23.2 (td,
J ) 15, 3 Hz, PP(CH3)3), 22.1 (d, J ) 18 Hz, PH(CH3)3). IR (THF,
cm-1): 3351 (w), 3194 (w), 1822 (s), 1610 (m). Anal. Calcd for
C44H52F24BNP4Ru: C, 41.07; H, 4.07; N, 1.09. Found: C, 41.09; H,
3.86; N, 0.98.
cis-(PMe3)4Ru(H)(NHC(CHPh)(CH2Ph)) (23). Complex 2 (65 mg,
154 µmol) was dissolved in toluene (4 mL), and diphenylallene (29
mg, 155 µmol) was added. The resulting bright red solution was allowed
to stand at room temperature for 3 h and was then layered with pentane
(10 mL) and chilled to -35 °C for 72 h to yield 23 (46 mg, 49% yield)
cis-[(PMe3)4Ru(H)(PMe3)+][BPh4-] (27). Complex 3 (240 mg, 323
µmol) was dissolved in THF (3 mL), and the solution was loaded into
a glass vessel equipped with a Teflon stopcock. The solution was
degassed, and PMe3 (66 mL × 100 Torr, 360 µmol)was added via
9
14694 J. AM. CHEM. SOC. VOL. 124, NO. 49, 2002

Reactivity of Parent Amido Ruthenium Complexes
A R T I C L E S
vacuum transfer from a known volume bulb. Complex 27 (184 mg,
70% yield) precipitated as a white solid. 1H NMR (THF-d8): 7.27 (8H,
br, Ar), 6.85 (8H, t, J ) 7.3 Hz, Ar), 6.72 (4H, t, J ) 7.3 Hz, Ar), 1.51
(36H, br, PP(CH3)3), 1.35 (9H, d, J ) 6.3 Hz, PH(CH3)3), -11.36 (dquin,
J ) 72, 24 Hz, 1H, RuH) ppm. 31P{1H} NMR (THF-d8): δ -9.11 (d,
J ) 27 Hz, PP), -22.22 (quin, J ) 27 Hz, PH) ppm. 13C{1H} NMR
(CD2Cl2): δ 164.6 (q, JBC ) 100 Hz), 136.4 (s, Ar), 126.2 (s, Ar),
122.3 (s, Ar), 26.8 (m PP(CH3)3), 26.2 (d, J ) 20 Hz, PH(CH3)3). IR
(THF, cm-1): 1841 (s). Anal. Calcd for C39H66BP5Ru: C, 58.43; H,
8.30. Found: C, 58.24; H, 8.41.
Measurement of Keq for Equilibrium between [(PMe3)4Ru(H)-
(NH3)+][A-] and [(PMe3)4Ru(H)(PMe3)+][A-]. In a typical experi-
ment, 3 (11 mg, 15 µmol) and internal standard were dissolved in THF-
d8 (300 µL) and the solution was loaded into an NMR tube equipped
with a Teflon stopcock. The solution was then degassed, and PMe3
(6.6 mL × 34 Torr, 13 µmol) was added via vacuum transfer from a
known volume bulb. The tube was then closed and left at room
temperature, and the concentrations of 3 (δ -9.48 (1H, dtd) ppm), 27
(δ -11.33 (1H, dquin) ppm), NH3 (δ -0.17 (3H, t) ppm), and PMe3
(δ 0.94 (9H, s) ppm) were monitored by 1H NMR at 20 °C. When the
relative concentrations stopped changing, they were measured and used
to calculate Keq (15.9). Additional NH3 (6.6 mL × 30 Torr, 12 µmol)
was added to the tube via vacuum transfer from a known volume bulb.
The tube was again monitored by 1H NMR, and the Keq (16.1) recorded
after equilibrium was achieved.
Acknowledgment. The authors thank the NSF for funding
this work (Grant No. CHE-0094349), Dr. Fred Hollander and
Dr. Allen Oliver for performing the X-ray diffraction study, and
Jennifer R. Krumper, Andrew P. Duncan, and Daniel J. Fox
for generous donations of NaBArf, diphenyl allene, and 1,
respectively.
Supporting Information Available: Full details of the X-ray
diffraction study of 21. This material is available free of charge
via the Internet at http://pubs.acs.org.
JA027620F
9
J. AM. CHEM. SOC. VOL. 124, NO. 49, 2002 14695