Amido/phosphine pincer hydrides of ruthenium

Article abstract of DOI:10.1039/b206202j

The chemistry of the ligand (R₂PCH₂SiMe₂)₂N^- (R = cyclohexyl and ^tBu), PNP-R , on ruthenium is developed, including RuH(PNP-Cy)(PPh₃) and (HPNP-R)RuH₃Cl. The latter contains a protonated nitrogen (i.e., amine as a donor to Ru) and one H₂ ligand (X-ray structure for R = ^tBu). This compound can be dehydrohalogenated to give (PNP-Cy)RuH₃, which undergoes H/D exchange of D₂ into its cyclohexyl rings, and is itself dehydrogenated by excess H₂C=CHR to give [Cy₂PCH₂SiMe₂NSiMe₂CH ₂PCy(C₆H₈)] Ru, which contains a triply dehydrogenated cyclohexyl ring π-allyl bonded to Ru. (PNP-Cy)RuH₃ reacts with dihydrofurans to give the heteroatom-stabilized carbene complex (PNP-Cy)RuH[=CO(CH₂)₃].

Full text of DOI:10.1039/b206202j

View Article Online / Journal Homepage / Table of Contents for this issue
Amido/phosphine pincer hydrides of ruthenium
Lori A. Watson, Joseph N. Coalter III, Oleg Ozerov, Maren Pink, John C. Huffman and
Kenneth G. Caulton*
Department of Chemistry and Molecular Structure Center, Indiana University, Bloomington,
IN 47405, USA. E-mail: caulton@indiana.edu
Received (in New Haven, CT, USA) 26th June 2002, Accepted 15th September 2002
First published as an Advance Article on the web 17th December 2002
t
The chemistry of the ligand (R2PCH2SiMe2)2Nꢀ (R ¼ cyclohexyl and Bu), ‘‘PNP-R’’, on ruthenium
is developed, including RuH(PNP-Cy)(PPh3) and (HPNP-R)RuH3Cl. The latter contains a protonated
nitrogen (i.e., amine as a donor to Ru) and one H2 ligand (X-ray structure for R ¼ tBu). This compound
can be dehydrohalogenated to give (PNP-Cy)RuH3 , which undergoes H/D exchange of D2 into its cyclohexyl
=
rings, and is itself dehydrogenated by excess H2C CHR to give [Cy2PCH2SiMe2NSiMe2CH2PCy(C6H8)] Ru,
which contains a triply dehydrogenated cyclohexyl ring p-allyl bonded to Ru. (PNP-Cy)RuH3 reacts with
=
dihydrofurans to give the heteroatom-stabilized carbene complex (PNP-Cy)RuH[ CO(CH2)3].
The many pincer ligands I that have been reported recently1–10
fall into two general categories, those with a neutral donor G
(e.g., pyridine-based), and those with an anionic donor G (e.g.,
phenyl-based). The neutral donor Do can be phosphorous or
nitrogen, and this DoRR0 group can have controllable electro-
nic and steric (including chiral) features, leading to a versatile
set of pincer ligands.5,11,12 Depending on the nature of the
‘‘arm’’ that links G to Do , the donor can be at the amine or
imine oxidation level. We have been attracted to the pincer
ligands II pioneered by Fryzuk because the group G is anionic
and, unlike phenyl, bears a lone pair.13
We report here our efforts to develop ruthenium polyhydride
chemistry with the PNP ligand carrying primarily cyclohexyl
substituents, since ligand steric bulk has been proven effective
in preventing reagent deactivation via dimerization.
Results and discussion
Preparation of the PNP-R ligands
One additional advantage of the Fryzuk ligand class is the pos-
sibility of systematic modification of phosphine alkyl groups; a
variety of these have been prepared.1 The synthesis of the
PNP-R ligands in this study followed a modified preparation
in which the desired phosphines, HPR2 , are deprotonated at
ꢀ78 ꢁC in THF to yield the lithium phosphide in situ; this
was then reacted with the silylamide to form the desired
LiPNP-R salt. Recrystallization from ether gives 60–75% yield
of the corresponding etherate (Scheme 1).
Both the N-protonated and N-TMS-protected PNP-R
ligands can also be synthesized. The protonated ligand,
HPNP-R, has been found to be a useful source of the ligand
in this study. Made from treatment of the Li–PNP salt with
1 M HCl in ether at 0 ꢁC, the HPNP-R ligand (a clear oil) is
typically used as a solution in benzene. The TMS-protected
version of the PNP ligand can be prepared by treatment of
an ether solution of the corresponding Li salt with TMS–
OTf. The TMS-protected ligand was initially prepared so as
to minimize protonation of the amide in subsequent synthesis
steps.
The ability of an amide N to participate in p-donation to the
metal is something we have developed14 as a way to access,
under mild conditions (e.g., 20 ꢁC), unsaturated (poly)hydride
molecules; the ligand p-lone pair can donate to an otherwise
unsaturated metal, making it metastable (persistant), but
nevertheless leaving it operationally unsaturated. Amides of
the late transition metals with 16 valence electrons are quite
prone to b-hydrogen migration to give a hydride and an imine,
III; the presence of silicon on nitrogen in the Fryzuk ligand
helps to prevent such a degradation, albeit at the price of a
somewhat diminished nitrogen nucleophilicity.
Scheme 1
DOI: 10.1039/b206202j
New J. Chem., 2003, 27, 263–273
263
This journal is # The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2003

View Article Online
structure, IV.
The hydride, while not located in the crystal structure refine-
ment, was located by DFT calculations,15 which placed the
˚
hydride at 1.562 A from the Ru metal center (Fig. 2). Good
Scheme 2
agreement with the above crystal structure data was achieved
with all other bond lengths and angles in the optimized structure
(Table 2). The calculated N–Ru–PH3 angle of 158.5ꢁ compares
favorably to the experimentally determined angle of 164.2ꢁ.
These reactions and accompanying crystal structure show
that the PNP ligand is a suitable ligand for the ruthenium sys-
tem and can form unsaturated hydride complexes. The next
goal was the synthesis of a polyhydride complex that could
participate in C–H activation reactions.
Synthesis and characterization of RuH(PNP-R)(PR03)
The reaction of the lithium salt of the PNP-R ligand with Ru
hydrido-chlorides forms the hydride-phosphine complex
RuH(PNP-R)(PR03) (Scheme 2).
Both the PPh3 and PiPr3 derivatives can be synthesized; the
former from reaction of a toluene solution of RuHCl(PPh3)3
with the LiPNP-R salt, and the latter from [RuHCl(PR03)2]2 .
Reaction of LiPNP-Ph or Cy with the ruthenium hydrides at
room temperature gives quantitative conversion to the appro-
priate RuH(PNP-R)(PR03). In each case, a hydride signal was
observed as a doublet of triplets, showing splitting from both
the pincer ligand and non-chelating phosphine. Two diastereo-
topic Si–Me signals were also observed, each integrating to six
hydrogens. The 31P{H} NMR exhibits a doublet, due to the
interaction of PR2 with PR03 , and a downfield triplet from
PR03 coupling to PNP-PR2 . Synthetic scale purification of
RuH(PNP-R)(PPh3) was performed by removal of the liber-
ated phosphine via sublimation.
Synthesis and characterization of (HPNP-Cy)RuH3Cl
A
successful entry into polyhydride PNP-Ru chemistry
involves the synthesis and isolation of the pincer-protonated
(HPNP-Cy)RuH3Cl as a precursor to a 16 eꢀ Ru species.
(HPNP-Cy)RuH3Cl can be made [eqn. (1)] in poor (isolated)
yield from the corresponding protonated HPNP-Cy ligand
and RuH3Cl(PCy3)2 , liberating 2 equiv of PCy3 .
HPNP-Cy þ RuH3ClðPCy3Þ2
C6H6
ðHPNP-CyÞRuH Cl þ 2PCy ð1Þ
ꢀꢀꢀꢀ!
3
3
Solid-state structure of Ru(H)(PNP-Cy)(PPh3)
Synthesis of (HPNP-Cy)RuH3Cl is also possible from a variety
of common Ru starting materials, including [(p-cymene)RuCl2]2 ,
[(COD)RuCl2]n or [(C6H6)RuCl2]2 , by the use of the LiPNP-Cy
salt under an atmosphere of hydrogen [eqn. (2)]. Stirring this
solution overnight in THF at room temperature produces
(HPNP-Cy)RuH3Cl; this 18 eꢀ complex has proven relatively
easy to isolate in good yield by recrystallization from pentane.
A single crystal suitable for X-ray diffraction studies was
obtained from slow evaporation of a toluene solution of
RuH(PNP-Cy)(PPh3). The molecular structure and selected
atom labelling are illustrated in Fig. 1. Details of the structural
determination are presented in Tables 1 and 2.
The molecular structure of RuH(PNP-Cy)(PPh3) shows the
expected coordination geometry of two trans PR2 groups [with
a P1–Ru–P2 angle of 161.71(5)ꢁ] mutually cis to the amide
0:5½p-cymeneÞRuCl2l2
or
˚
nitrogen. The amide nitrogen lies 2.145(4) A away from the
Ru center and shows no pyramidalization (sum of angles
around the nitrogen center is 359.96ꢁ). The Ru–P (PNP) bond
1=n½CODꢂRuCl2ꢂn þ LiPNP-Cy
THF
ðHPNP-CyÞRuH Cl
ꢀꢀꢀꢀ!
3
H2
˚
lengths are 2.3764(15) and 2.3538(14) A for P1 and P2 , respec-
or
tively, slightly shorter than the corresponding Ru–P distances
˚
[2.3892(8) and 2.3998(7) A] in the previously characterized
structure of RuCl(C6H4PPh2)[NH(SiMe2CH2PPh2)2].13 PPh3
0:5½ðC6H6ÞRuCl2ꢂ2
ð2Þ
˚
has a Ru–P bond length of 2.235 A with distances and angles
Table 1 Crystal structure parameters
in the structure suggesting that no agostic Cy or Ph is present.
The PNP pincer ligand is approximately coplanar with the
ruthenium metal center; however, the triphenylphosphine
ligand is bent away from that plane, suggesting a Y-shaped
Chemical formula
Ru(H)(PPh3)
(N(SiMe2CH2PCy2)2)
C48H76NP3RuSi2
917.30
RuH3Cl[NH(SiMe2-
CH2PtBu2)2]
C22H76NP3RuSi2
589.32
Empirical formula
Molecular weight
Crystal system
Space group
T/K
Triclinic
P1 bar
113
Orthorhombic
Pbca
113(2)
m/mmꢀ1
0.509
0.786
3
˚
U/A
2399.75
11.083(1)
12.643(1)
18.442(1)
88.64(1)
88.80(1)
68.28(1)
2
6170.1(2)
14.8336(3)
13.0180(3)
31.9522(7)
90
˚
a/A
˚
b/A
˚
c/A
a/ꢁ
b/ꢁ
90
g/ꢁ
90
8
19135
Z
Total reflections collected
Unique reflections
Rint
11017
10301
14501
0.050
9648
0.067
6091
Observed reflections
[I > 2.3s(I)]
R(F) (all data)
Rw(F) (all data)
0.046
0.037
0.0305
0.0776
Fig. 1 Crystal structure determination of (PNP-Cy)RuH(PCy3).
264
New J. Chem., 2003, 27, 263–273

View Article Online
Table 2 Comparison of calculated and experimental values for
(PNP)RuHL (L ¼ H or Ph)
(PNP)RuH(PH3)
Calcd (B3PW91)
(PNP)RuH(PPh3)
Exptal
Ru–N
N–Si
Si–C
2.118
1.734
1.876
1.850
2.326
2.275
1.562
113.0
164.2
2.15
1.71
1.87
1.83
C–P
P–Ru
2.37
2.24
Ru–PR3
Ru–H
N–Ru–H
N–Ru–PR3
N/A
N/A
158.5
Fig. 2 Geometry optimized structure of (PNP)RuH(PH3) (see Table
2 for parameters).
(HPNP-Cy)RuH3Cl exhibits one hydride resonance at
ꢀ12.46 ppm (t, J ¼ 14.2 Hz). This hydride resonance remains
a sharp triplet to ꢀ20 ꢁC, where it begins to broaden; at
ꢀ80 ꢁC, the resonance is a broad singlet. A T1(min) was found
at 59(1) ms (C7D8 , 400 MHz, ꢀ30 ꢁC), suggesting a trihydride
structure with relatively small RH–H and cH–Ru–H or an
˚
is rather long at 2.5263(3) A. Normally, M–P and M–Cl dis-
tances to Ru are essentially equal, but here the Ru–Cl distance
˚
is longer by 0.16 A. A similar intramolecular hydrogen bond
between an NH of an HPNP ligand and a metal-bound halide
was observed in Ir complexes of protonated PNP ligand.17,18
t
RuH(H2) structure with a long H–H bond. The Bu analog
(see below) helps resolve this uncertainty. The N–H resonance
of the ligand amine is observed as a singlet at 3.10 ppm. While
the cyclohexyl region of the spectrum is quite crowded, the
cyclohexyl resonances integrate to approximately 44H.
Reactivity of (HPNP-Cy)RuH3Cl
Attempts to form (HPNP-Cy)RuH4 from (HPNP-Cy)RuH3Cl
by using various hydride transfer reagents [NaBH4 , LiAlH4 ,
Cp2ZrHCl, Et3SiH, Me2PhSiH, (tBu)3SiH] were unsuccessful;
when reacting at all, only intractable mixtures of products
were formed. Neither could the lone chloride ligand be
replaced by a more weakly binding anion, using such reagents
as AgOTf. No exchange with the hydrogen ligands was
observed when (HPNP-Cy)RuH3Cl was allowed to react with
1 atm of D2 , even at elevated temperatures (60 ꢁC for 20 h in
C6D6).
However, some ligand replacement reactions were moder-
ately successful. (HPNP-Cy)RuHCl(PiPr3) can be formed in
35% yield by reaction of (HPNP-Cy)RuH3Cl with a stoichio-
metric amount of PiPr3 for 5 days at 60 ꢁC. This slow rate is
apparently the result of the complex being saturated. No
change in yield or identity of products formed is observed with
the addition of excess phosphine. The synthesis of (HPNP-
Cy)RuHCl(PiPr3) was confirmed by the independent reaction
of [RuHCl(PiPr3)2]2 with HPNP-Cy, which produced (HPNP-
Cy)RuHCl(PiPr3) in quantitative yield. The N–H signal of the
Synthesis and structure of (HPNP-tBu)RuH3Cl
Although (PNP-Cy)Li served to introduce the PNP-Cy ligand
onto Ru via a number of common Ru starting materials (vide
supra), attempts to use (PNP-tBu)Li analogously were unsuc-
cessful. Reactions of (PNP-tBu)Li with RuHCl(PPh3)3 ,
[RuHCl(PR3)2]2 and [(arene)RuCl2]2 only resulted in low
( < 50%) conversion to the desired products and were plagued
by side reactions. Encouraged by our success in surmounting
similar problems in the introduction of the PNP ligand onto
Re via the utilization of Mg derivatives of PNP,16 we decided
to try this approach here. The reaction between
(PNP-tBu)MgCl(dioxane) and [(p-cymene)RuCl2]2 in C6D6 ,
followed by exposure to H2 atmosphere, cleanly produces
(HPNP-tBu)RuH3Cl (95% purity by NMR). Solid
(HPNP-tBu)RuH3Cl was isolated in the form of X-ray quality
crystals in 61% yield. The RuH3 spin system gives rise to a sin-
gle resonance in the 1H NMR spectrum at ꢀ12.96 ppm (t,
JHP ¼ 15 Hz) and selective decoupling of only the alkyl hydro-
gens gives rise to a quartet (from three H on Ru) in the 31P
NMR spectrum for the equivalent P nuclei of the HPNP
ligand. The environment around Ru in the solid state structure
(Fig. 3 and Table 3) can be described as approximately octahe-
dral. The results of the X-ray diffraction study are consistent
with a dihydrogen ligand occupying the position trans to the
NH ligand and a hydride ligand trans to Cl. The compression
of the P–Ru–P angle to 163.725(11)ꢁ from the idealized octahe-
dral value of 180ꢁ can be attributed to the pincer ligand con-
straints. In spite of such constraints, the Ru–N distance is
nearly as long as the distance from Ru to the much larger
atom, phosphorus. The chloride ligand is also somewhat dis-
placed from an idealized octahedral position towards the
NH functionality (cN1–Ru–Cl1 ¼ 82.6ꢁ), presumably due
to the N–Hꢃ ꢃ ꢃCl hydrogen bonding. This hydrogen bonding
is also evident in the unusually small cRu–N–H of
91.0(16)ꢁ, which has the effect of shortening this very nonlinear
(i.e., unfavorable) hydrogen bond. The dihedral angle H1N–
N1–Ru1–Cl1 is 23ꢁ. The participation of Cl in hydrogen bond-
ing is likely facilitated by the trans influence of the hydride
ligand weakening the Ru–Cl bond; indeed, the Ru–Cl distance
Fig. 3 X-Ray crystal structure of (HPNP-tBu)RuH(H2)Cl.
New J. Chem., 2003, 27, 263–273
265

View Article Online
t
Table 3 Selected distances (A) and angles (deg) for [HPNP- Bu]-
RuH(H2)Cl
complex, RuH(Cl)(H2)(PCy3)2 .21
˚
Ru(1)–N(1)
Ru(1)–P(2)
2.3115(10)
2.3520(3)
2.3636(3)
2.5263(3)
1.48(2)
Ru(1)–H(3R)
Si(1)–N(1)
Si(2)–N(1)
1.52(3)
1.7919(11)
1.7672(11)
0.79(2)
Ru(1)–P(1)
Ru(1)–Cl(1)
N(1)–H(1N)
H(2R)–H(3R)
Ru(1)–H(1R)
Ru(1)–H(2R)
N(1)–Ru(1)–P(2)
N(1)–Ru(1)–P(1)
P(2)–Ru(1)–P(1)
N(1)–Ru(1)–Cl(1)
P(2)–Ru(1)–Cl(1)
P(1)–Ru(1)–Cl(1)
1.12(3)
1.61(2)
87.89(3)
90.27(3)
163.725(11) Si(2)–N(1)–Si(1)
P(1)–Ru(1)–H(1R) 81.9(9)
Cl(1)–Ru(1)–H(1R) 165.4(8)
DFT calculation of these RuH3 complexes15 revealed that
the geometry optimizes to a (PNP)RuH(H2) structure (Fig. 4,
structure VI above but with all Me and Cy replaced by H),
regardless of the initial geometry (trihydride or hydride/dihy-
drogen) employed. The Ru(IV) form, (PNP)Ru(H)3 , is not a
minimum on the potential energy surface. The Ru species is
124.06(6)
107.21(5)
112.25(5)
111.6(16)
105.7(16)
82.56(3)
98.691(11)
97.098(11)
Si(2)–N(1)–Ru(1)
Si(1)–N(1)–Ru(1)
Si(2)–N(1)–H(1N)
Si(1)–N(1)–H(1N)
N(1)–Ru(1)–H(1R) 82.9(8)
P(2)–Ru(1)–H(1R) 81.8(9)
Ru(1)–N(1)–H(1N) 91.0(16)
˚
calculated to have an H–H distance of 0.95 A, lengthened con-
˚
siderably from the value calculated in free H2 (0.74 A), due to
back donation into s*(H–H) enhanced by the p-donor amide
ligand trans to itself. RuH bond lengths to H2 hydrogens are
ꢄ0.1 A longer than to Ru–H (1.66 A vs. 1.56 A), consistent
with neutron diffraction structural data on MH(H2) com-
pounds.22
amine is observed slightly upfield of (HPNP-Cy)RuH3Cl, at
3.08 ppm, and two diastereotopic Si–Me signals are observed
at 0.35 and 0.13 ppm. As in the case of (PNP-Cy)RuH(PiPr3),
the 31P{1H} NMR spectrum is characterized by two signals, a
doublet, due to PCy2/PiPr3 coupling, and a downfield triplet
from PiPr3 .
˚
˚
˚
Reaction of (HPNP-Cy)RuH3Cl with 1 atm of CO in C6D6
immediately results in the formation of a yellow solution (from
the red-brown of the starting materials) and the evolution of
H2 (as seen by 1H NMR). Within 1 h the reaction is complete,
forming primarily (HPNP-Cy)RuH(CO)Cl, which is character-
ized by two diastereotopic Si–methyl signals at 0.43 and 0.37
ppm, integrating to 6 hydrogens each, as well as a triplet in
the hydride region at ꢀ5.47 ppm (JP–H ¼ 19 Hz). The 31P
NMR resonance can be observed as a singlet at 53.7 ppm.
Reactivity of (PNP-Cy)RuH3
Reactivity of (PNP-Cy)RuH3 toward H2. H2 (1 atm) adds
reversibly (but incompletely) to form (PNP-Cy)RuH5 at room
temperature [Fig. 5 and eqn. (4)]. The hydrides within the
resulting (PNP-Cy)RuH5 are never fully decoalesced (ki for
intramolecular site exchange is very large), even at ꢀ95 ꢁC,
while signals for H2 and for (PNP-Cy)RuH3 are resolved on
1
the H NMR time scale (kf is small) at ꢀ95 ꢁC [eqn. (4)].
kf
)
Formation of an unsaturated polyhydride
*
ðPNP-CyÞRuH3 þ H2
ðPNP-CyÞRuH5
ð4Þ
|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}
|fflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflffl}
(PNP-Cy)RuH3 can be prepared in poor (isolated) yield by the
reaction of RuH3Cl(PCy3)2 with the corresponding LiPNP-Cy
salt. Because of the difficulty in separating free phosphine
from the reaction mixture, it was thought that reaction of
(HPNP)RuH3Cl with agents that would affect removal of
HCl was a more promising synthetic route. (HPNP-Cy)-
RuH3Cl reacts quantitatively with lithium 2,2,6,6-tetramethyl-
piperidide (LiTMP) or Me3SiCH2Li to form (PNP-Cy)RuH3 ,
LiCl, and the protonated base [eqn. (3)]. Recrystallization from
pentane gives a 45% yield of the unsaturated Ru species.
1H ð20 ꢁCÞ coalesced at d ꢀ10:72
ki;1H ð20 ꢁCÞ dꢀ9:06
At 20 ꢁC in d8-toluene, the (PNP-Cy)RuH3 resonance,
usually observed at ꢀ15.04 ppm, is coalesced with the dis-
solved hydrogen, forming a broad singlet at ꢀ10.72 ppm with
1 atm H2 added (the position of the signal is dependent on the
pressure of hydrogen added and the temperature of the
ðHPNP-CyÞRuH3Cl þ LiTMP
ꢀLiCl
ðPNP-CyÞRuH þ TMP-H
ꢀꢀꢀꢀꢀ!
3
ðHPNP-CyÞRuH3Cl þ Me3SiCH2Li
ꢀLiCl
ðPNP-CyÞRuH þ SiðCH Þ
ð3Þ
ꢀꢀꢀꢀꢀ!
3
3
4
A triplet at ꢀ15.04 ppm is seen for (PNP-Cy)RuH3 in the
hydride region of the 1H NMR spectrum if a spectrum is taken
immediately upon dissolution [allowing (PNP-Cy)RuH3 to
stand in deuterated solvent causes the signal to broaden from
H–D exchange, as detailed in the C–H activation section later
in this paper]. The Si(Me)2 groups are equivalent at all avail-
able temperatures, and a 31P singlet is observed at 55.4 ppm.
Even at ꢀ90 ꢁC, only one signal is observed. A T1(min) of
45(2) ms (C7D8 , 300 MHz, ꢀ60 ꢁC) was measured. As in the
case of (HPNP-Cy)RuH3Cl, this does not unequivocally con-
firm a structure, but is consistent with an averaged T1 from
a hydride/dihydrogen system or with three independent
hydrides placed in close proximity to each other (short
R
H–H).19,20
Thus, two structures consistent with the spectral data can be
considered, V and VI. The redox isomer containing Ru(II) has
been seen for the corresponding chloro-bisphosphine Ru
Fig. 4 DFT
(PNP)RuHn .
calculated,
geometry-optimized
structures
of
266
New J. Chem., 2003, 27, 263–273

View Article Online
of sp3 carbons, by an olefin acting as a hydrogen acceptor.
ð5Þ
In the reaction mixture, the Z3-cyclohexenyl(cyclohexyl)-
phosphine complex is characterized by an AB pattern at
31P{1H} d 104.51 and 32.58 (JP–P ¼ 303 Hz), as well as four
SiCH3chemical shifts. The proposed structure is based on such
reactions with other cyclohexyl-substituted phosphines.
The addition of [NEt3H]Cl minimizes the formation of the
Z3-metallated cyclohexyl phosphine ring, allowing, for exam-
ple, in the case of the dihydrofuran, progression to the carbene
F [eqn. (6)].
Fig. 5 Variable temperature NMR spectra of the (PNP-Cy)RuH3/
(PNP-Cy)RuH5 equilibrium.
sample). With a reduction in temperature, the increased mole
fraction of (PNP)RuH5 increases the signal intensity at
ꢀ9.06 ppm. As the mole fraction of (PNP)RuH3 and H2
ð6Þ
decrease,
a
substantial shift and broadening of the
(PNP)RuH3/H2 coalesced signal is seen, broadening almost
completely at ꢀ20 ꢁC (ꢀ12.62 ppm). No additional signal
beside that assigned to (PNP)RuH5 is observed at ꢀ95 ꢁC
(Fig. 5), where the equilibrium in eqn. (4) is shifted nearly fully
to the right. Returning the sample to room temperature results
in complete reversal of the temperature-dependent shifts to
their original values. Evacuation of the sample, followed by
redissolution in deuterated solvent confirms that only
(PNP)RuH3 (and its H–D exchange products) are present.
Performing a similar experiment as above in C7H8 and fol-
lowing the reaction by 2H NMR also results in the appearance
of a coalesced (PNP-Cy)RuH3 and D2 signal at ꢀ12.39 ppm
(0.9 atm of D2 added at ambient laboratory temperature).
By 31P NMR, the reaction mixture contained 34 mole percent
of the (PNP-Cy)Ru(H/D)5 product, while the remainder was
(PNP-Cy)Ru(H/D)3 . The addition of D2 was again reversible;
removal of the solvent in vacuo and redissolution of the
remaining reddish-brown solid in C6D6 showed only the pre-
One possible mechanism that accounts for the observed inter-
mediates and the influence of the addition of [NEt3H]Cl upon
the product distribution is detailed in Scheme 3.
The formation of A can be observed as the first signals
resulting from the reaction of any amount of 2,3-dihydrofuran
with (PNP-Cy)RuH3 . Within 5 min of their mixing, there is a
complete disappearance of the signals due to (PNP)RuH3 (i.e.,
the hydride at ꢀ15.03 ppm and the accompanying 31P signal)
and the appearance of only one hydride resonance, a triplet
at ꢀ17.6 ppm. A 31P {1H} singlet and other 1H resonances
are also shifted from those of the starting material, although
the multiplicities and relative separations are nearly identical.
A small amount of free THF is also seen. This complex is
persistent, existing in the reaction mixture at room temper-
ature as long as there is still some unreacted olefin.
By following this reaction by NMR at room temperature,
the production of additional THF can be observed. If H2 is
added at this point, all excess 2,3-dihydrofuran is converted
to tetrahydrofuran. Some (approximately 10% of the reaction
mixture after 3 h) of the Z3-metallated cyclohexyl product
(identified by the AB 31P{H} NMR pattern) is also formed.
After approximately 5 h, the first evidence for the carbene
is seen (hydride triplet at ꢀ16.86 ppm). After allowing the
reaction to proceed at room temperature for 2.5 days, an
equilibrium mixture consisting of 20–30% of the Z3-metallated
product, 20–30% of the carbene, and 40–60% of complex A is
obtained. The addition of a catalytic amount of [NEt3H]Cl at
this point decreases the amount of the Z3-cyclometallated
product in solution; thus, B and C are in equilibrium
under H2 .
1
sence of (PNP-Cy)Ru(H/D)3 by H and 31P NMR.
As mentioned above, the hydride ligands within (PNP)RuH5
are not decoalesced even at low temperature; therefore, an
experimental determination of the hydride/dihydrogen nature
of these ligands is not possible. A DFT study of the
(PNP)RuH5 complex,15 using the same model as described
above for the (PNP)RuH3 calculation, found that (PNP)RuH5
converges to a hydride/bis-dihydrogen structure, with the H2
˚
trans to N having a longer H/H distance (0.92 A) than that
˚
trans to hydride (0.80 A), consistent with differential back-
donation into each of the s*(H–H) caused by the trans ligand,
˚
H or N (Fig. 4). The Ru–H(hydride) distance (1.58 A) is
shorter than those to H2 , and a shorter H–H distance (H4
Heating the reaction mixture without the addition of
[NEt3H]Cl results in the conversion of A to the carbene spe-
cies, with little change in the amount of Z3 product observed.
After 15 h at 60 ꢁC, about 80% of the reaction mixture has con-
verted to the carbene, with the remainder being primarily the
Z3-cyclometallated product. If [NEt3H]Cl is added at any
point, nearly quantitative conversion to the carbene is seen.
Addition of [NEt3H]Cl to a solution of (PNP-Cy)RuH3 and
subsequent addition of the other olefins under consideration
˚
to H5), 0.80 A, correlates with a longer Ru–H distance, 1.88
˚
˚
A (cf. ꢄ1.68 A to H2 and H3). The two H2 molecules are
orthogonal, which is a symptom of their interaction with dif-
ferent dp orbitals for back-donation.
Reactivity of (PNP-Cy)RuH3 toward olefins. The reaction
(typically 5–10 h at room temperature followed by 10–15 h
at 60 ꢁC in C6D6) of (PNP-Cy)RuH3 with a variety of olefins
(typically in a 1:4 mole ratio) results in the formation of an
Z3-cyclohexenyl ring [eqn. (5)], in addition to small amounts
of bound or isomerized (to a carbene in the case of dihydro-
furan) olefin as well as equivalent amounts of hydrogenated
olefin. These reactions are thus net dehydrogenations, even
t
(styrene, Bu-ethylene) results simply in the olefin adduct of
the unsaturated (PNP-Cy)RuH3 complex, an addition that is
reversible upon removal of volatiles in vacuo. Each olefin com-
plex shows a 31P{1H} NMR AB pattern, due to the prochiral
character of the olefin.
Identical products are observed [eqn. (7)] from the reaction
of (HPNP-Cy)RuH3Cl and NEt3 , which allows the in situ
New J. Chem., 2003, 27, 263–273
267

View Article Online
Scheme 3
dehydrohalogenation of the 18 eꢀ species and the formation of
a reactive 16 eꢀ complex, presumably (PNP-Cy)RuH3 .
processes.24–26 This was first observed as H/D exchange in
C6D6 . While a triplet at ꢀ15.04 ppm is seen for (PNP-
1
Cy)RuH3 in the hydride region of the H NMR spectrum if
a spectrum is taken immediately upon dissolution, allowing
(PNP-Cy)RuH3 to stand in deuterated solvent causes the
signal to broaden from H/D exchange (Fig. 6). To slow the
H/D exchange at room temperature, a 1:1 mixture of C6H6/
C6D6 was employed. With phosphorus decoupling, two signals
can be resolved in as little as 15 min in contact with C6D6 .
After 30 min, 31P decoupling of the hydride signal reveals a
downfield 1:1:1 triplet (with JH–D ¼ 5 Hz) due to the forma-
tion of (PNP-Cy)RuH2D as well a singlet still assignable to
(PNP-Ru)H3 . These signals overlap, forming a triplet with a
downfield shoulder with JH–P ¼ 12.8 Hz if phosphorus decou-
pling is not employed. Complete disappearance of a hydride
signal (and thus complete conversion to (PNP-Cy)RuD3) in
ð7Þ
This transient unsaturated complex was trapped by the addi-
tion of excess 2,3-dihydrofuran to a benzene solution of equi-
molar (HPNP-Cy)RuH3Cl and NEt3 . After 2 days at 60 ꢁC
complete conversion to the carbene complex F could be seen
[eqn. (7)], together with 1 equiv of tetrahydrofuran. The hydride
resonance of F is found as a triplet at ꢀ16.82 ppm, with the
31P{H} NMR signal appearing as a singlet at 41.2 ppm. Addi-
tional hydrogen signals corresponding to the hydrogens on the
heterocycle appear as triplets at 3.90 and 3.27 ppm, and as a
broad triplet at 1.97 ppm. A carbon signal at 297.8 ppm, a
broad singlet due to unresolved coupling to P, further confirms
the assignment of this species to a carbene moiety. Such reactiv-
ity has also been seen with the unsaturated Ru complex
RuHCl(PiPr3)2 .23 The slow reactivity of (HPNP-Cy)RuH3Cl
with 2,3-dihydrofuran without base must be due to its
saturated character, with no good leaving group available.
Similar to (PNP-Cy)RuH3 above, this in situ formed com-
plex can also isomerize 2,5-dihydrofuran to the identical car-
bene [eqn. (8)]. With the addition of 2,5-dihydrofuran and
1
the H NMR occurs after 8 h in pure C6D6 .
Interestingly, such H/D exchange also occurs with aliphatic
solvents. In cyclohexane-d12 , for example, (PNP-Cy)RuH3
gives a triplet at d ꢀ15.49 (with JP–H ¼ 12.8 Hz) upon initial
dissolution. Within 2 h, broadening of the resonance is seen,
indicating a significant amount of H/D exchange. Analysis
2
of this sample by H NMR (using solvent suppression techni-
ques to minimize the interference of C6D12) reveals a deuter-
ium signal at ꢀ15.4 ppm, which corresponds to at least the
partial formation of (PNP-Cy)Ru(H/D)3 (Fig. 7).
The C–H bonds in the cyclohexyl rings in (PNP-Cy)RuH3
also undego H/D exchange. By dissolving (PNP-Cy)RuD3
[made either from allowing (PNP-Cy)RuH3 to stir overnight
in C6D6 or from reacting (PNP-Cy)RuH3 with D2 in deuter-
ated solvent, then removing volatiles in vacuo] in toluene-d8
at room temperature, the presence of 2H as a broad resonance
at d 1.83 in the cyclohexyl region of the spectrum was
observed. Following the D/H exchange of (PNP—Cy)RuD3
=
NEt3 , (HPNP-Cy)RuH3Cl forms (PNP-Cy)Ru(H)( COC3H6)
quantitatively in 5 days at 60 ꢁC, again producing 1 equiv of
THF. In both cases, an insoluble material ([NEt3H]Cl) precipi-
tates from the benzene solution.
2
with toluene-H8 by H NMR, the percent deuteration at the
hydride site decreased more rapidly than that observed at the
cyclohexyl phosphine sites.
ð8Þ
Conclusions
The pincer-ligated compounds reported here can be compared
to those with chloride in place of amide; (PNP-Cy)RuH(H2)
then can be compared with RuH(H2)Cl(PCy3)2 . Both of
Reactivity of (PNP-Cy)RuH3 toward C–H bonds. (PNP-
Cy)RuH3 also participates in C(sp3)–H activation
268
New J. Chem., 2003, 27, 263–273

View Article Online
Experimental
General considerations
All manipulations were performed using standard Schlenk
techniques or in an argon-filled glovebox unless otherwise
noted. Solvents were distilled from Na/benzophenone,
˚
CaH2 , or 4 A molecular sieves, degassed prior to use, and
stored in air-tight vessels. RuCl2(PPh3)3 ,27 RuHCl(PPh3)3ꢃ
C7H8 ,28 RuCl2(PCy3)2(CHCHCMe2),29 [RuHCl(PiPr3)2]2 ,23
HN(SiMe2CH2PPh2)2 ,30 LiN(SiMe2CH2PPh2)2 ,30 HPtBu2
31
and RuH3Cl(PCy3)2 were prepared according to published
21
procedures. All other reagents were used as received from com-
mercial vendors. 1H NMR chemical shifts are reported in ppm
relative to protio impurities in the deuterated solvents. 31P
spectra are referenced to an external standard of 85% H3PO4at
0 ppm. NMR spectra were recorded with a Varian Gemini
2000 (300 MHz 1H; 121 MHz 31P; 75 MHz 13C), a Varian
Unity Inova (400 MHz 1H; 162 MHz 31P; 101 MHz 13C), or a
Fig. 6 Time-dependent 400 MHz 1H NMR spectra (20 ꢁC) of (PNP-
Cy)RuH3 in a 1:1 C6H6–C6D6 solvent mixture.
1
Varian Unity Inova (500 MHz H, 126 MHz 13C) instrument.
these compounds share the hydride/dihydrogen ground state
structure and thus both show a preference for Ru(II) and so
an aversion to a higher metal oxidation state. The greater p-
donor power of amide vs. chloride then fails to reduce H2 to
2 Hꢀ1. On the other hand, the greater donor power of the
amide gives intact (PNP-Cy)RuH(H2) enough p-basicity at
Ru to improve the thermodynamics of binding additional
H2 , to give (PNP-Cy)RuH5 . Perhaps even more demon-
strative of the p-basicity (i.e., reducing power) of (PNP-
Cy)RuH(H2) is its ability to react, by (endothermic, but
thermally accessible) oxidative addition, with arenes, with its
own cyclohexyl C–H, and with free cyclohexane H–C(sp3)
bonds, all evidenced by H/D exchange. The endothermic char-
acter of all these except the intramolecular version is character-
istic of a 4d metal, and might be reversed for the 5d analog, Os,
because of the generally stronger M–H and M–C bonds for 5d
vs. 4d metals.
Synthetic access here to (PNP-Cy)RuH(H2) involves dehy-
drochlorination: removal of H from N and Cl from Ru in
(HPNP-Cy)RuH3Cl. While this reaction is successful (with a
strong base), and probably benefits thermodynamically from
donation of the resulting amide nitrogen lone pair to an
otherwise unsaturated Ru, this and other reactions reported
here are slower than desired. The activation energies impli-
cated are probably due at least in part to steric effects, and thus
the four cyclohexyl substituents in most of the molecules
reported here may represent ‘‘overprotection’’ in PNP-Cy,
which may be ameliorated by changing to smaller phosphine
substituents.
Syntheses
(PNPtBu)MgCl(dioxane). tBu2PH (6.33 mL, 34.2 mmol) was
dissolved in 100 mL of 1:1 toluene–THF mixture. n-BuLi (21.4
mL of 1.6 M in hexanes, 34.2 mmol) was added to this solu-
t
tion, and the yellow color of Bu2PLi appeared. The mixture
was stirred for 15 min and then HN(SiMe2CH2Cl)2 (2.49
mL, 11.4 mmol) was added and the color dissipated. n-BuLi
(7.1 mL of 1.6 M in hexanes, 11.4 mmol) was added to this
solution. The mixture was stirred for 15 min and then HN(Si-
Me2CH2Cl)2 (0.83 mL, 3.8 mmol) was added. n-BuLi (2.4 mL
of 1.6 M in hexanes, 3.8 mmol) was added to this solution. The
resulting mixture was stirred for 15 min and then HN(Si-
Me2CH2Cl)2 (0.28 mL, 1.27 mmol) was added. The resulting
mixture was stirred for 15 min more and then the solvent
removed in vacuo. The residue was extracted with pentane, fil-
tered and stripped to dryness. The remaining white solid was
treated with 25 mL of THF and anhydrous MgCl2 (1.90 g,
20 mmol) was added. This was stirred for 24 h, then treated
with 4 mL of 1,4-dioxane, stirred for 4 h and filtered. The fil-
trate was stripped to dryness, redissolved in Et2O–dioxane,
and stripped again. The residue was extracted with ether, fil-
tered, and the volume of the filtrate was reduced to ca. 10
mL. This was treated with 60 mL of pentane and placed in a
freezer (ꢀ30 ꢁC) for 24 h. The fluffy white solid was filtered
off, washed with cold pentane and dried in vacuo to give the
first crop of the product (4.61 g, 45%). The combined washings
from the last step were reduced in volume to ca. 10 mL and
after 24 h at ꢀ30 ꢁC the second crop of the product (0.99 g,
11%) was collected. Total yield: 5.60
g
(56%).
(PNP-tBu)MgCl(dioxane): 1H NMR (C6D6): d 3.37 (s, 8H,
dioxane), 1.14 (d, 13 Hz, 36H, CMe3), 0.58 (d, 10 Hz, 4H,
P–CH2–Si), 0.44 (s, 12H, Si–CH3). 31P{1H} NMR (C6D6): d
17.9 (s).
(HPNP-tBu)RuH3Cl. (PNP-tBu)MgCl(dioxane) (24.2 mg,
40.5 mmol) and [(cymene)RuCl2] (12.4 mg, 40.5 mmol) were
mixed in a J. Young tube in 0.6 mL C6D6 . After shaking for
1
h, the NMR spectrum revealed the formation of
(PNP-tBu)RuCl and an equivalent amount of free cymene.
This suspension was treated with 2 mL of pentane, filtered
and stripped to a yellow oil. This oil was dissolved in 0.6 mL
C6D6 in a J. Young tube and then degassed by two freeze-
pump-thaw cycles. The tube was back-filled with H2 at
1 atm. This caused a change of color to blue-green in the time
of mixing. After ca. 20 min under an H2 atmosphere, the color
again became yellow and the NMR indicated the formation of
(HPNP-tBu)RuH3Cl ( > 90% purity). The volatiles were
removed in vacuo in a small flask and the resultant yellow oil
2
Fig. 7 Time-dependent 1H and H NMR spectra (20 ꢁC) of (PNP-
Cy)RuH3 in C6D12
.
New J. Chem., 2003, 27, 263–273
269

View Article Online
was layered with ca. 0.3 mL of pentane. After standing for
24 h at ambient temperature large, X-ray quality crystals of
the product formed. Yield: 14.5 mg (61%). (HPNP-tBu)-
RuH3Cl: 1H NMR (C6D6): d 3.15 (br, 1H, NH), 1.42 (vt, 5
Hz, 18H, CMe3), 1.15 (vt, 5 Hz, 18H, CMe3), 0.41 (s, 6H,
Si–CH3), 0.23 (s, 6H, Si–CH3), ꢀ12.96 (t, 15 Hz, 3H, RuH).
31P{1H} NMR (C6D6): d 75.4 (s).
JP–H ¼ 4 Hz, 4H, CH2), 1.10 [d, J ¼ 10.8 Hz, 36H, P(tBu)2].
31P{1H} NMR (162 MHz, C6D6 , 20 ꢁC): d 19.0 (s).
HN(SiMe2CH2PCy2)2.
LiN(SiMe2CH2PCy2)2ꢃ0.75Et2O
(850 mg, 1.38 mmol) was dissolved in 20 mL ether and cooled
to 0 ꢁC in an ice bath. via syringe, 1.4 mL (1.4 mmol) of 1 M
HCl in ether was added, and the reaction was stirred for 15
min before warming to room temperature and stirring an addi-
tional 45 min. The solution was filtered through a fine frit and
the LiCl residue was extracted with 10 mL ether. The com-
bined ether extracts were concentrated to 3 mL in vacuo, but
cooling overnight at ꢀ70 ꢁC produced no precipitate. The
remaining volatiles were removed to a liquid N2 trap to yield
the title compound as a viscous oil. Yield: 525 mg (95%).
For convenience, the reagent was used as a 0.44 M solution
.
LiN(SiMe2CH2PCy2)2 0.75Et2O. This procedure is a slight
modification of Fryzuk’s preparations for analogous com-
pounds LiN(SiMe2CH2PR2), where R ¼ Me, iPr, tBu.32 nBuLi
(17 mL of 2.5 M in hexanes, 42.5 mmol) was added to a solu-
tion of 7.50 g (37.8 mmol) HPCy2 in 75 mL hexane at room
temperature. The white slurry was allowed to stir for 5 days,
the supernatant was decanted via cannula, and the product
was washed with 75 mL pentane. Drying in vacuo yielded
7.70 g of LiPCy2 (quantitative). The lithio salt (7.70 g, 37.7
mmol) was slurried in 30 mL toluene, diluted with 100 mL
THF, and cooled to 0 ꢁC. Over a period of 15 min, a solution
of 2.90 g (12.6 mmol) 1,3-bis(chloromethyl)-1,1,3,3-tetra-
methyldisilazane [HN(SiMe2CH2Cl)2] in 10 mL THF was
added dropwise via syringe. The mixture was allowed to warm
to room temperature and stirred 30 min before removal of the
volatiles to a liquid N2 trap. The residue was extracted with
pentane (2 ꢅ 75 mL), filtered, and reduced to dryness in vacuo.
Attempts to isolate product from the viscous yellow oil by
crystallization from hexane or pentane failed, yielding only a
trace of LiCl precipitate. The LiCl was separated via cannula,
and the mother liquor volatiles were removed to a liquid N2
trap. Dissolving the residue in a minimum of ether and cooling
to ꢀ70 ꢁC for 5 days produced a white crystalline solid, which
was washed with 10 mL cold ether (ꢀ70 ꢁC) and dried in vacuo
to yield 4.50 g (58%) of the title compound as a 4:3 diethyl
1
in C6D6 unless generated in situ. H NMR (400 MHz, C6D6 ,
20 ꢁC; NH proton not observed): d 0.35 (s, 12H, SiMe2), 0.60
(d, JP–H ¼ 4 Hz, 4H, CH2), 1.1–1.3, 1.5, 1.6, 1.7–1.9 [br m,
44H, P(C6H11)2]. P{1H} NMR (162 MHz, C6D6 , 20 ꢁC): d
31
13
ꢀ12.6 (s). C{1H} NMR (101 MHz, C6D6 , 20 ꢁC): d 2.9 (d,
JP–C ¼ 5 Hz, SiMe2), 9.1 (d, JP–C ¼ 37 Hz, CH2), 27.0 [s,
P(4-C6H11)2], 27.7 [s, P(3/5-C6H11)2], 27.8 [s, P(3/5-C6H11)2],
29.5 [d, JP–C ¼ 11 Hz, P(2/6-C6H11)2], 30.3 [d, JP–C ¼ 14
Hz, P(2/6-C6H11)2], 35.1 [d, JP–C ¼ 17 Hz, P(1-C6H11)2].
HN(SiMe2CH2PtBu2)2 can be prepared identically, beginning
from the LiPNP-tBu salt; an 82% isolated yield is obtained
1
of the very thick clear oil. H NMR (400 MHz, C6D6 , 20 ꢁC;
NH proton not observed): d 0.30 (s, 12H, SiMe2), 0.53 (d,
JP–H ¼ 3.6 Hz, 4H, CH2), 1.09 [d, J ¼ 11.2 Hz, 36H,
P(tBu)2]. P{1H} NMR (162 MHz, C6D6 , 20 ꢁC): d 18.8 (s).
31
RuH(PNP-Cy)(PPh3). RuHCl(PPh3)3ꢃC7H8 (750 mg, 0.74
mmol) and 454 mg (0.74 mmol) LiN(SiMe2CH2P-
Cy2)2ꢃ0.75Et2O were added to a Schlenk flask and stirred in
30 mL toluene for 48 h at room temperature. The red solution
was filtered and the volatiles were removed to a liquid N2 trap.
The resulting red solid was powdered in a mortar and pestle
and the free PPh3 liberated in the reaction was removed by
sublimation (60 ꢁC, 0.003 torr, 1 week). Isolated yield: 620
1
etherate (1H NMR integration). H NMR (400 MHz, C6D6 ,
20 ꢁC): d 0.50 (s, 12H, SiMe2), 0.78 (d, JP–H ¼ 4 Hz, 4H,
3
CH2), 1.12 [t, 4.5H, JH–H ¼ 7 Hz, O(CH2CH3)2], 1.26 [br
m, 20H, P(C6H11)2], 1.64 [br t, 8H, JH–H ¼ 12 Hz,
P(C6H11)2], 1.77 [br s, 8H, P(C6H11)2], 1.90 [br s, 8H,
3
P(C6H11)2], 3.30 [q, 3H, JH–H ¼ 7 Hz, O(CH2CH3)2].
1
mg (92%). H NMR (400 MHz, C6D6 , 20 ꢁC): d ꢀ26.47 (dt,
31P{1H} NMR (162 MHz, C6D6 , 20 ꢁC): d ꢀ9.3 (very br s).
13C{1H} NMR (101 MHz, C6D6 , 20 ꢁC): d 7.3 (s, SiMe2),
11.4 (d, JP–C ¼ 24 Hz, CH2), 15.4 [s, O(CH2CH3)2], 26.9 [s,
P(4-C6H11)2], 27.9 [d, JP–C ¼ 9 Hz, P(3/5-C6H11)2], 28.0 [d,
JP–C ¼ 9 Hz, P(3/5-C6H11)2], 30.3 [d, JP–C ¼ 9 Hz, P(2/6-
C6H11)2], 30.6 [d, JP–C ¼ 9 Hz, P(2/6-C6H11)2], 34.4 [br s,
P(1-C6H11)2], 65.8 [s, O(CH2CH3)2]. Notes: (1) The lithium
phosphide salt need not be isolated, but can also be prepared
and used in situ (THF, ꢀ78 ꢁC addition of nBuLi, then 3 h
of stirring at room temperature). (2) Over several months,
the lattice-bound ether is lost (1H NMR, C6D6) to yield a
material less soluble in aliphatic solvents, though reactivity is
unaffected.
2
2
0
JP –H ¼ 44 Hz, JP–H ¼ 19 Hz, 1H, RuH), 0.59 (s, 6H,
PNP-SiMe2), 0.61 (s, 6H, PNP-SiMe2), 0.7–1.8 (m, 46H,
PNP-C6H11 overlapping with PNP-CH2), 2.21 (br d, 2H,
J ¼ 8Hz, PNP-CH2), 7.03 [apparent q, JH–H ¼ 6 Hz, 3H,
P(p-C6H5)3], 7.09 [apparent t, JH–H ¼ 8 Hz, 6H, P(m-
C6H5)3], 7.84 [apparent t, JH–H ¼ JP–H ¼ 9 Hz, 6H, P(o-
C6H5)3]. P{1H} NMR (162 MHz, C6D6 , 20 ꢁC): d 48.8 (d,
31
0
0
JP–P ¼ 26 Hz, 2P, PNP-PCy2), 73.3 (t, JP –P ¼ 26 Hz, 1P,
PPh3). C{1H} NMR (101 MHz, C6D6 , 20 ꢁC): d 6.1 (s,
13
PNP-SiMe2), 7.4 (s, PNP-SiMe2), 10.7 (s, PNP-CH2), 26.7 [s,
P(C6H11)2], 27.2 [s, P(C6H11)2], 27.5 [vt, JP–C ¼ 7 Hz, P(2/6-
C6H11)2], 27.6 [vt, JP–C ¼ 5 Hz, P(2/6-C6H11)2], 27.95 [vt,
JP–C ¼ 2 Hz, P(2/6-C6H11)2], 28.02 [vt, JP–C ¼ 4 Hz, P(2/6-
C6H11)2], 29.2 [s, P(C6H11)2], 29.5 [s, P(C6H11)2], 30.6 [s,
P(C6H11)2], 31.9 [s, P(C6H11)2], 34.7 [vt, JP–C ¼ 8 Hz, P(4-
C6H11)2], 39.0 [vt, JP–C ¼ 11 Hz, P(4-C6H11)2], 127.4 [d,
JP–C ¼ 8 Hz, P(m-C6H6)3], 128.5 [s, P(p-C6H6)3], 135.0 [d,
JP–C ¼ 10 Hz, P(o-C6H6)3], 142.7 [d, JP–C ¼ 33 Hz,
P(i-C6H6)3].
n
LiN(SiMe2CH2PtBu2)2 0.75Et2O. BuLi (14.5 mL of 2.0 M
.
in pentane, 1.06 equiv.) was added to a solution of 4.0 g (27.4
mmol) HPtBu2 in 40 mL THF dropwise over 20 min at ꢀ78 ꢁC.
The yellow solution was allowed to stir for 2 h, then recooled
to ꢀ78 ꢁC. Over a period of 1.5 h, a solution of 2.107 g (9.1
mmol)
1,3-bis(chloromethyl)-1,1,3,3-tetramethyldisilazane
[HN(SiMe2CH2Cl)2] in 25 mL THF was added dropwise via
addition funnel. The mixture was allowed to warm to room
temperature and stirred 30 min before removal of the volatiles
to a liquid N2 trap. The residue was extracted with pentane
(2 ꢅ 75 mL), filtered, and reduced to dryness in vacuo. Dissol-
ving the residue in a minimum of ether and cooling to ꢀ70 ꢁC
for 3 days produced a white crystalline solid, which was
washed with 10 mL cold ether (ꢀ70 ꢁC) and dried in vacuo
to yield 2.24 g (56%) of the title compound. 1H NMR
(400 MHz, C6D6 , 20 ꢁC): d 0.33 (s, 12H, SiMe2), 0.56 (d,
RuH(PNP-Ph)(PPh3). RuHCl(PPh3)3ꢃC7H8 (750 mg, 0.74
mmol) and 396 mg (0.74 mmol) LiN(SiMe2CH2PPh2)2 were
added to a Schlenk flask and stirred in 50 mL toluene for 48
h at room temperature. The red solution was filtered and the
volatiles were removed to a liquid N2 trap. The resulting red
solid was powdered in a mortar and pestle and the free PPh3
liberated in the reaction was removed by sublimation (60 ꢁC,
0.003 torr, 1 week). Isolated yield: 600 mg (91%). 1H NMR
2
(300 MHz, C6D6 , 20 ꢁC): d ꢀ20.45 (dt, JP –H ¼ 42 Hz,
0
2JP–H ¼ 20 Hz, 1H, RuH), 0.01 (s, 6H, PNP-SiMe2), 0.54 (s,
270
New J. Chem., 2003, 27, 263–273

View Article Online
2
6H, PNP-SiMe2), 1.79 (2nd order dvt, JH–H ¼ 13 Hz, JP–H
¼
C6H11), 30.0 (s, PNP-C6H11), 30.3 (s, PNP-C6H11), 31.4
(s, PNP-C6H11), 37.7 (s, PNP-C6H11), 39.0 (s, PNP-C6H11).
4 Hz, 2H, PNP-CH2), 1.89 (2nd order dvt, JH–H ¼ 13 Hz,
JP–H ¼ 5 Hz, 2H, PNP-CH2), 6.8–7.7, 8.1 (m, 35H, PNP-
C6H5 overlapping with PPh3-C6H5). 31P{1H} NMR (121
2
(HPNP-Cy)RuHCl (PiPr3). (HPNP-Cy)RuH3Cl (24.3 mg,
0.0363 mmol) was dissolved in approximately 0.5 mL C7D8
and placed in an NMR tube. PiPr3 (7.1 mL, 1.02 equiv) was
added via syringe. No reaction was observed by NMR after
18 h at room temperature. Heating the solution for an addi-
tional 5 days at 60 ꢁC resulted in the formation of (HPNP-
Cy)RuHCl(PiPr3) (35% yield), among other decomposition
MHz, C6D6 , 20 ꢁC): d 40.7 (d, JP–P ¼ 19 Hz, 2P, PNP-
0
0
PPh2), 85.7 (t, JP –P ¼ 19 Hz, 1P, PPh3).
RuH(PNP-Cy)(PiPr3). [RuHCl(PiPr3)2]2 (10 mg, 0.011
mmol) and 13.5 mg (0.022 mmol) LiN(SiMe2CH2P-
Cy2)2ꢃ0.75Et2O were combined in 0.5 mL C6D6 and added to
an NMR tube. H and 31P{1H} spectra recorded 20 min later
products. H NMR (400 MHz, C7D8 , 20 ꢁC): d ꢀ12.50 (br s,
1
1
revealed quantitative conversion to the title compound with
liberation of 1 equiv of free PiPr3 . 1H NMR (400 MHz,
1H, RuH), 0.13 (s, 6H, PNP-SiMe2), 0.35 (s, 6H, PNP-SiMe2),
3
1.30 [br d, JP–H ¼ 9 Hz, 18H, P(CHMe2)3], 1.0–2.2 [m, 43H,
2
2
C6D6 , 20 ꢁC): d ꢀ23.52 (dt, JP –H ¼ 48 Hz, JP–H ¼ 20 Hz,
PNP-C6H11 overlapping with PiPr3-CHMe2], 2.24 [br d,
J(H)P–H ¼ 14 Hz, 2H, PNP-CH2 or PNP-C6H11(methine)],
2.30 [br d, J(H)P–H ¼ 12 Hz, 2H, PNP-CH2 or PNP-
C6H11(methine)], 2.53 [br d, J(H)P–H ¼ 24 Hz, 2H, PNP-CH2
or PNP-C6H11(methine)], 2.66 [br d, J(H)P–H ¼ 13 Hz, 2H,
PNP-CH2 or PNP-C6H11(methine)], 3.08 (br s, 1H, HPNP).
0
1H, RuH), 0.47 (s, 6H, PNP-SiMe2), 0.51 (s, 6H, PNP-SiMe2),
3
3
d 1.29 [dd, JP–H ¼ 12 Hz, JH–H ¼ 8 Hz, 18H, P(CHMe2)3],
1.0–2.2 (m, 43H, PNP-C6H11 overlapping with PiPr3-CHMe2),
2.00 [br d, J(H)P–H ¼ 13 Hz, 2H, PNP-CH2 or PNP-
C6H11(methine)], 2.13 [br d, J(H)P–H ¼ 12 Hz, 2H, PNP-CH2
or PNP-C6H11(methine)], 2.36 [br d, J(H)P–H ¼ 12 Hz, 2H,
PNP-CH2 or PNP-C6H11(methine)], 2.52 [br d, J(H)P–H ¼ 12
Hz, 2H, PNP-CH2 or PNP-C6H11(methine)]. 31P{1H} NMR
31
1
P{ H} NMR (162 MHz, C7D8 , 20 ꢁC): d 51.8 (d, JP–P
¼
0
21 Hz, 2P, PNP-PCy2), 73.4 (t, JP –P ¼ 19 Hz, 1P, PiPr3).
0
(162 MHz, C6D6 , 20 ꢁC): d 41.4 (d, JP–P ¼ 22 Hz, 2P, PNP-
Reaction of (HPNP-Cy)RuH(CO)Cl. (HPNP-Cy)RuH3Cl
(10 mg, 0.0148 mmol) was dissolved in approximately 0.5
mL C6D6 and placed in an NMR tube. This solution was
placed under 1 atm of CO by standard gas line techniques.
Within 5 min, the red-brown solution had become bright yel-
low. (HPNP-Cy)RuH(CO)Cl was identified by NMR. 1H
0
PCy2), 91.1 (t, JP –P ¼ 19 Hz, 1P, PiPr3).
0
RuH(PNP-Ph)(PiPr3). [RuHCl(PiPr3)2]2 (10 mg, 0.011
mmol) and 11.7 mg (0.022 mmol) LiN(SiMe2CH2PPh2)2 were
combined in 0.5 mL C6D6 and added to an NMR tube. 1H and
31P{1H} spectra recorded 20 min later revealed 85% conver-
sion to the title compound with liberation of 1 equiv of free
2
NMR (400 MHz, C6D6 , 20 ꢁC): d ꢀ5.47 (t, JP–H ¼ 19 Hz,
1H, RuH), 0.43 (s, 6H, PNP-SiMe2), 0.37 (s, 6H, PNP-SiMe2),
0.9–2.1 (m, 48H, PNP-C6H11 , PNP-CH2). 31P{1H} NMR (162
MHz, C6D6 , 20 ꢁC): d 53.7 (s).
PiPr3 . H NMR (400 MHz, C6D6 , 20 ꢁC): d ꢀ22.74 (dt,
1
2
2
0
JP –H ¼ 46 Hz, JP–H ¼ 20 Hz, 1H, RuH), ꢀ0.24 (s, 6H,
3
PNP-SiMe2), 0.46 (s, 6H, PNP-SiMe2), 0.92 [dd, JP–H ¼ 12
3
Hz, JH–H ¼ 7 Hz, 18H, P(CHMe2)3], 1.37 [m, 3H, P(CH-
Ru(PNP-Cy)H3. (HPNP-Cy)RuH3Cl (199.9 mg, 0.2989
mmol) was dissolved in approximately 15 mL of C6H6 .
LiTMP (74.1 mg, 1.68 equiv, in 5 mL C6H6) was added drop-
wise over 20 min at 0 ꢁC to the red-brown solution and allowed
to stir at room temperature for 20 min. The solvent was
removed in vacuo, and the resulting reddish-brown solid was
extracted with pentane and filtered through Celite, yielding
an off-white solid (LiCl) and a red filtrate. The solution was
concentrated in vacuo and cooled to ꢀ40 ꢁC; (PNP-Cy)RuH3
was obtained as a reddish powder (isolated yield 46%; yield
of the crude product 71%). Following the identical procedure
with (Me)3SiCH2Li gives comparable yields and product pur-
ity. 1H NMR (400 MHz, C6D6 , 20 ꢁC): d ꢀ15.03 (t,
2JP–H ¼ 13.0 Hz, 3H, RuH3), 0.49 (s, 12H, PNP-SiMe2),
0.9–1.89 (m, 48H, PNP-C6H11 , PNP-CH2). 31P{1H} NMR
(162 MHz, C6D6 , 20 ꢁC): d 55.48 (s). Attempts to obtain a
13C{1H} NMR spectrum were unsuccessful, due to decomposi-
tion of (PNP)RuH3 over several hours in solution.
Me2)3], 1.41 (br vt, JP–H ¼ 4 Hz, 2H, PNP-CH2), 1.94 (br vt,
JP–H ¼ 3 Hz, 2H, PNP-CH2), 6.9–7.2, 7.5, 8.2 (m, 20H,
PNP-C6H5). P{1H} NMR (162 MHz, C6D6 , 20 ꢁC): d
31
0
0
39.9 (d, JP–P ¼ 24 Hz, 2P, PNP-PPh2), 93.1 (t, JP –P ¼ 19
Hz, 1P, PiPr3).
Ru(HPNP-Cy)H3Cl.
Method
1.
LiN(SiMe2CH2P-
Cy2)2ꢃ0.75Et2O (1.0032 g, 1.87 mmol) and 470.1 mg (1.002
equivalents) of ](C6H6)RuCl2]2 were combined in 125 mL of
THF at room temperature to form a reddish brown slurry.
The head space of the 300 mL flask was evacuated and refilled
with H2 (1 atm). The reaction mixture was allowed to stir for
3 h before the flask was refilled with H2 (1 atm). Stirring over-
night yielded a reddish homogeneous solution. Removal of the
solvent in vacuo gave a red-brown solid that was then extracted
with toluene. The solvent was removed from the toluene fil-
trate and the resulting red solid washed with cold pentane
and dried under vacuum for 3 h (587.4 mg, 93.7%). [(p-cyme-
ne)RuCl2]2 and [(COD)RuCl2]n can be used under identical
conditions to give comparable yields.
Ru(PNP-Cy)H5. (PNP-Cy)RuH3 (9.6 mg, 0.0152 mmol) was
dissolved in approximately 0.5 mL C6D6 and placed in a gas
tight NMR tube. The solution was degassed and the head
space gasses removed. One atmosphere (20 ꢁC) of H2 was
added via standard gas line techniques. After 10 min,1H
NMR revealed a new broad singlet at ꢀ8.90 ppm, and a shift-
ing and broadening of the RuH3 signal, formerly found at
ꢀ15.0 ppm, to approximately ꢀ11 ppm (exact value is deter-
mined by the amount of hydrogen present in the system).
Similarly, a new 31P signal, which can be assigned to (PNP-
Cy)RuH5 , was present at 62.5 ppm, corresponding to approxi-
mately 10% of the reaction mixture at 20 ꢁC. Removal of all
volatiles and redissolution in C6D6 gave 100% (PNP-
Cy)RuH3 , as shown by a broad singlet (formerly a triplet,
but broadened by the H–D solvent exchange described else-
where) at ꢀ14.98 ppm and a singlet in the 31P at 55.3 ppm.
Variable temperature experiments did not decoalesce the
RuH5 hydrogens, though a reduction in temperature to less
Method 2. RuH3Cl(PCy3)2 (10.0 mg, 0.0143 mmol) was dis-
solved in 0.5 mL C6D6 in an NMR tube and 32.5 mL of
HPNP-Cy (0.44 M in C6D6) was added via syringe. Quantita-
tive conversion to (HPNP-Cy)RuH3Cl was seen within 1 h;
attempts to scale up this reaction led to diminished yields
and problems with product isolation. 1H NMR (400 MHz,
2
C6D6 , 20 ꢁC): d ꢀ12.46 (t, JP–H ¼ 13.6 Hz, 3H, RuH3),
0.223, 0.216, (singlets, 12H total, PNP-SiMe2), 0.9–1.89 (m,
44H, PNP-C6H11), 0.96 (dt, JHH ¼ 14 Hz, JPH ¼ 4Hz, 2H,
PNP-CH2), 1.20 (dt, JHH ¼ 14 Hz, JPH ¼ 4Hz, 2H, PNP-
CH2), 3.10 [s, 1H, HN-(SiMe2CH2PCy2)2RuH3Cl]. 31P{1H}
NMR (162 MHz, C6D6 , 20 ꢁC): d 48.01 (s). 13C{1H, 31P}
NMR (101 MHz, C6D6 , 20 ꢁC): d 12.0 (PNP-SiMe2), 3.1 (s,
PNP-SiMe2), 15.4 (s, PNP-CH2), 26.7 (s, PNP-C6H11), 26.9
(s, PNP-C6H11), 27.5 (s, PNP-C6H11), 28.0 (s, PNP-C6H11),
28.1 (s, PNP-C6H11), 28.5 (s, PNP-C6H11), 29.6 (s, PNP-
New J. Chem., 2003, 27, 263–273
271

View Article Online
than ꢀ20 ꢁC allowed the nearly quantitative production of
and 32.58 (AB pattern, JP–P ¼ 303 Hz). No hydrides are
0
(PNP-Cy)RuH5 from (PNP-Cy)RuH3 .
observed.
Reaction of Ru(PNP-Cy)H3 with CH2CHC(CH3)3. (PNP-
Cy)RuH3 (7.4 mg, 0.0117 mmol)was dissolved in 0.5 mL of
C6H6 . tBu-ethylene, CH2CHC(CH3)3 , (7.5 mL, 4.97 equiv)
was added via syringe. After 15 min at room temperature,
two products [bound olefin and an Z3-cyclohexenyl(cyclo-
hexyl) phosphine complex in an approximately 0.8:1 ratio]
were observed, along with some liberated neohexane. After
15 h at 60 ꢁC, 95% conversion to the Z3-cyclohexenyl(cyclo-
hexyl) phosphine complex was seen. 1H NMR (400 MHz,
RuH(PNP-Cy)(CO(CH2)3). Method 1. (HPNP-Cy)RuH3Cl
(10 mg, 0.015 mmol) was dissolved in 0.5 mL C6D6 ; 2.2 mL
(1.02 equiv) of NEt3 and 6.0 mL of C4H6O (2,3-dihydrofuran,
5.15 equiv) were added via syringe. The solution was trans-
ferred to an NMR tube. NMR spectra taken through 12 h at
room temperature showed no change in the observed spectra.
After 2 h of heating at 60 ꢁC, 14% conversion to the title pro-
duct was seen; this increased to 95+% after 2 days of heating at
1
60 ꢁC. H NMR (400 MHz, C6D6 , 20 ꢁC): d ꢀ16.82 (t,
C6D6 , 20 ꢁC) of the olefin complex: d ꢀ19.97 (br t, JP–H
¼
2JP–H ¼ 22 Hz, 1H, RuH), 0.43 (s, 6H, PNP-SiMe2), 0.28 (s,
6H, PNP-SiMe2), 0.80 (dt, JHH ¼ 9 Hz, JPH ¼ 4Hz, 2H,
PNP-CH2), 1.1–2.2 (m, PNP-C6H11 , PNP-CH2), 1.97 (br t,
2H), 3.27 (t, J ¼ 7.2 Hz), 3.90 (t, J ¼ 5.6 Hz). 31P{1H}
NMR (162 MHz, C6D6 , 20 ꢁC): d 41.23 (s). 13C{1H, 31P}
NMR (101 MHz, C6D6 , 20 ꢁC): d 2.9 (s, PNP-SiMe2), 3.3 (s,
19 Hz, 3H, RuH3), 0.41 (s, 3H, PNP-SiMe2), 0.36 (s, 3H,
PNP-SiMe2), 0.31 (s, 3H, PNP-SiMe2), 0.28 (s, 3H, PNP-
SiMe2), 0.9–1.94 (m, 48H, PNP-C6H11 , PNP-CH2), bound
tBu ethylene resonances are located under free Bu ethylene
t
and evidenced by broadening only. 31P{1H} NMR (162
MHz, C6D6 , 20 ꢁC) of the olefin complex: d 63.1 and 36.0
=
PNP-SiMe2), 12.7 (s, PNP-CH2), 24.1 (s, Ru COCH2CH2-
(AB pattern, JP–P ¼ 326 Hz). 1H NMR and 31P{1H} NMR
0
CH2), 24.9 (s, PNP-C6H11), 25.7 (s, PNP-C6H11), 26.1 (s,
PNP-C6H11), 26.8 (s, PNP-C6H11), 27.1 (s, PNP-C6H11), 28.0
(s, PNP-C6H11), . 28.2 (s, PNP-C6H11), 28.3 (s, PNP-C6H11),
29.6 (s, PNP-C6H11), 30.2 (s, PNP-C6H11), 36.7 (s, PNP-
of the Z3 complex are the same as reported above.
C–H/D exchange
=
C6H11), 37.8 (s, PNP-C6H11), 53.2 (s, Ru COCH2CH2CH2),
In a typical experiment, 10 mg of (PNP-Cy)RuH3 was dis-
solved in the appropriate solvent in a gas-tight NMR tube;
25 mg of the Ru complex was used for H spectra. In order
to clearly follow H/D exchange in benzene at room tempera-
ture and within reasonable time intervals, a 1:1 mixture of
C6H6/C6D6 was employed; positions of deuteration were
determined by 1H and 2H NMR. In cyclohexane-d12 , no
resolved coupling for (PNP-Cy)RuH2D was seen in the 1H
NMR; in all cases, a broad singlet results after the specified
time periods.
=
=
75.9 (s, Ru COCH2CH2CH2), 297.8 (s, Ru COCH2CH2-
CH2). C{1H} NMR (101 MHz, C6D6 , 20 ꢁC, selected reso-
13
2
=
nance): d 297.8 (t, JPC ¼ 7.4 Hz, Ru COCH2CH2CH2).
Method 2. (HPNP-Cy)RuH3Cl (10.3 mg, 0.0158 mmol) was
dissolved in 0.5 mL C6D6 ; 2.3 mL (1.07 equiv) of NEt3 and
5.0 mL of C4H6O (2,5-dihydrofuran, 4.18 equiv) were added
via syringe. The solution was transferred to an NMR tube.
NMR spectra taken through 12 h at room temperature showed
no change. After 4 days heating at 60 ꢁC, 95+% conversion to
indicated product was observed.
Method 3. (PNP-Cy)RuH3 (11.1 mg, 0.01756 mmol) was dis-
solved in 0.5 mL C6D6 ; 3.0 mg (1.24 equiv) of NEt3ꢃHCl
and 6.0 mL of C4H6O (2,3-dihydrofuran, 4.52 equiv) were
added via syringe. The solution was transferred to an NMR
tube. After 2 h at 25 ꢁC, NMR spectra showed 60% of
(HPNP-Cy)RuH3Cl, 30% the end carbine (characterization
above), and 10% (PNP-Cy)RuH3 . After 4 h of heating at 60
ꢁC, 95% conversion to the hydrido carbene was seen.
Method 4. (HPNP-Cy)RuH3Cl (10.3 mg, 0.01582 mmol) was
dissolved in 0.5 mL C6D6 ; 2.2 .mL (0.99 equiv) of NEt3 and
5.0 mL of C4H6O (2,5-dihydrofuran, 4.18 equiv) were added
via syringe. The solution was transferred to an NMR tube.
After 5 days at 60 ꢁC, NMR spectra showed quantitative con-
version to the hydrido carbene.
X-Ray structure determinations
RuH(PPh3)[N(SiMe2CH2PCy2)2]. The orange crystal of
RuH(PPh3)[N(SiMe2CH2PCy2)2], grown from a saturated
toluene solution by slow evaporation, was affixed to a glass
fiber using silicone grease. The sample was then transferred
from the glove bag to the goniostat where it was cooled to
113 K using a gas-flow cooling system of local design. The data
were collectedonaBrukerSMART 6000diffractometerat113K
using 5 s frames with an omega scan of 0.30 degrees. Data were
corrected for Lorentz and polarization effects and equivalent
reflections averaged using the Bruker SAINT software as well
as utility programs from the XTEL library. The structure was
readily solved using SHELXTL and Fourier techniques. With
the exception of the hydride hydrogen, all hydrogen atoms
were readily located in a difference Fourier phased on the
non-hydrogen atoms. All hydrogen atoms located were
allowed to vary isotropically in the final cycles of refinement.
A careful examination of the final difference Fourier map did
not locate any peaks that could be readily identified as the
metal hydride position. A final difference Fourier map was
Reaction of Ru(PNP-Cy)H3 with C6H5CHCH2. (PNP-
Cy)RuH3 (10 mg, 0.0161 mmol)was dissolved in 0.5 mL of
C6H6 . Styrene, C6H5CHCH2 , (5.5 mL, 2.98 equiv) was added
via syringe. After 3 h at room temperature, two products
[bound olefin and an Z3-cyclohexenyl(cyclohexyl) phosphine
complex in an approximately 3:1 ratio] were observed. After
15 h at 60 ꢁC, 95% conversion to the Z3-cyclohexenyl(cyclo-
hexyl) phosphine complex was seen. 1H NMR (400 MHz,
C6D6 , 20 ꢁC) of the olefin complex: d ꢀ22.72 (dd, JP–H ¼ 24
ꢀ3
˚
featureless, the largest peak being 1.32 e A at the metal site.
(HPNP-tBu)RuH3Cl. A yellow crystal, grown from C6D6
and ether by layering, was cut to the approximate dimensions
0.30 ꢅ 0.30 ꢅ 0.30 mm3 and was placed onto the tip of a 0.1
mm diameter glass capillary and mounted on a SMART6000
(Bruker) at 113(2) K. A preliminary set of cell constants was
calculated from reflections harvested from three sets of 20
frames. These initial sets of frames were oriented such that
orthogonal wedges of reciprocal space were surveyed. This
produced initial orientation matrices determined from 460
reflections. The data collection was carried out using Mo Ka
radiation (graphite monochromator) with a frame time of 10
s and a detector distance of 5.01 cm. A randomly oriented
0
Hz, JP –H ¼ 20 Hz, 3H, RuH3), 0.59 (s, 3H, PNP-SiMe2),
0.48 (s, 3H, PNP-SiMe2), 0.39 (s, 3H, PNP-SiMe2), 0.19 (s,
3H, PNP-SiMe2), 0.9–1.89 (m, 48H, PNP-C6H11 , PNP-CH2),
bound styrene resonances are located under free styrene and
evidenced by broadening only. 31P{1H} NMR (162 MHz,
C6D6 , 20 ꢁC) of the olefin complex: d 44.6 and 35.0 (AB
1
pattern, JP–P ¼ 319 Hz). H NMR (400 MHz, C6D6 , 20 ꢁC)
0
of the Z3 complex: d 0.76 (s, 3H, PNP-SiMe2), 0.53 (s, 3H,
PNP-SiMe2), 0.42 (s, 3H, PNP-SiMe2), 0.23 (s, 3H, PNP-
SiMe2), 0.9–2.40 (m, 47H, PNP-C6H11 , PNP-CH2). 31P{1H}
NMR (162 MHz, C6D6 , 20 ꢁC) of the Z3 complex: d 104.51
272
New J. Chem., 2003, 27, 263–273

View Article Online
8
9
G. J. P. Britovsek, V. C. Gibson, S. K. Spitzmesser, K. P.
Tellmann, A. J. P. White and D. J. Williams, J. Chem. Soc.,
Dalton Trans., 2002, 1159.
region of reciprocal space was surveyed to the extent of 1.5
˚
spheres and to a resolution of 0.51 A. Five major sections of
frames were collected with 0.30ꢁ steps in o at five different f
settings and a detector position of ꢀ43ꢁ in 2y. An additional
set of 50 frames was collected in order to model decay. The
intensity data were corrected for absorption and decay
(SADABS).33 Final cell constants were calculated from the
xyz centroids of 9648 strong reflections from the actual data
collection after integration (SAINT).34 The space group Pbca
was determined based on systematic absences and intensity sta-
tistics. The structure was solved using SIR-92335 and refined
with SHELXL-97.36 A direct-methods solution was calculated,
which provided most non-hydrogen atoms from the E-map.
Full-matrix least squares/difference Fourier cycles were per-
formed ,which located the remaining non-hydrogen atoms.
All non-hydrogen atoms were refined with anisotropic displa-
cement parameters. All hydrogen atoms were placed in ideal
positions and refined as riding atoms with individual isotropic
displacement parameters except for the hydrogen atoms
bonded to Ru and N, which were refined for all parameters.
The final full-matrix least-squares refinement converged to
R1 ¼ 0.0305 and wR2 ¼ 0.0776 (F2, all data). The remaining
electron density is located around the metal and the chlorine
atom.
M. D. Fryzuk and P. A. MacNeil, Organometallics, 1983, 2, 355.
10 C. Gemel, K. Folting and K. G. Caulton, Inorg. Chem., 2000,
39, 1593.
11 W.-W. Xu, G. P. Rosini, K. Krogh-Jespersen, A. S. Goldman,
M. Gupta, C. M. Jensen and W. C. Kaska, Chem. Commun.,
1997, 23, 2273.
12 J. C. Grimm, C. Nachtigal, H.-G. Mack, W. C. Kaska and H. A.
Mayer, Inorg. Chem. Commun., 2000, 3, 511.
13 M. D. Fryzuk, C. D. Montgomery and S. J. Rettig, Organometal-
lics, 1991, 10, 467.
14 K. G. Caulton, New J. Chem., 1994, 18, 25.
15 L. A. Watson and K. G. Caulton, Mol. Phys., 2002, 100, 385.
16 O. V. Ozerov, H. F. Gerard, L. A. Watson, J. C. Huffman and
K. G. Caulton, Inorg. Chem., 2002, 41, 5615.
17 M. D. Fryzuk, P. A. MacNeil and S. J. Rettig, J. Am. Chem. Soc.,
1987, 109, 2803.
18 (a) M. D. Fryzuk, P. A. MacNeil and S. J. Rettig, J. Am. Chem.
Soc., 1987, 109, 2803; (b) M. D. Fryzuk, P. A. MacNeil, S. J.
Rettig and M. Stepan, Acta Crystallogr., Sect. C, 1996, 52, 1115.
19 P. G. Jessop and R. H. Morris, Coord. Chem. Rev., 1992, 121, 155.
20 K. A. Earl, G. Jia, P. A. Maltby and R. A. Morris, J. Am. Chem.
Soc., 1991, 113, 3027.
21 M. L. Christ, S. S. Sabo-Etienne and B. Chaudret, Organometal-
lics, 1994, 13, 3800.
22 R. Bau and M. H. Orabnis, Inorg. Chim. Acta, 1997, 259, 27.
23 J. N. Coalter III, J. C. Bollinger, J. C. Huffman, U. Werner-
Zwanziger, K. G. Caulton, E. R. Davidson, H. Gerard, E. Clot
and O. Eisenstein, New J. Chem., 2000, 24, 9.
CCDC reference numbers 197075–6. See http://
www.rsc.org/suppdata/nj/b2/b206202j/ for crystallographic
files in CIF or other electronic format.
24 P. Dani, M. A. M. Toorneman, G. P. M. van Klink and G. van
Koten, Organometallics, 2000, 19, 5287.
25 Y. Guari, S. Sabo-Etienne and B. Chaudret, Eur. J. Inorg. Chem.,
1999, 7, 1047.
26 M. Albrecht and G. van Koten, Angew. Chem., Int. Ed., 2001, 40,
3750.
27 P. S. Hallman, T. A. Stephenson and G. Wilkinson, Inorg. Synth.,
1970, 12, 237.
Acknowledgements
This work was supported by the U.S. Department of Energy.
LAW was supported by an NSF graduate fellowship.
28 R. A. Schunn and E. R. Wonchoba, Inorg. Synth., 1971, 13, 131.
29 T. E. Wilhelm, T. R. Belderrain, S. N. Brown and R. H. Grubbs,
Organometallics, 1997, 16, 3867.
30 M. D. Fryzuk, P. A. MacNeil, S. J. Rettig, A. S. Secco and
J. Trotter, Organometallics, 1982, 1, 918.
31 D. G. Gusev, M. Madott, F. M. Dolgushin, K. A. Lyssenko and
M. Y. Antipin, Organometallics, 2000, 19, 1734.
32 M. D. Fryzuk, A. Carter and A. Westerhaus, Inorg. Chem., 1985,
24, 642.
33 An empirical correction for absorption anisotropy: R. Blessing,
Acta Crystallogr., Sect. A, 1995, 51, 33.
34 SAINT 6.1, Bruker Analytical X-Ray Systems, Madison, WI,
USA.
35 SIR92: A. Altomare, G. Cascarno, C. Giacovazzo and A.
Gualardi, J. Appl. Crystallogr., 1993, 26, 343.
36 SHELXTL-Plus V5.10, Bruker Analytical X-Ray Systems,
Madison, WI, USA.
References
1
2
B. L. Shaw, J. Am. Chem. Soc., 1975, 97, 3856.
I. Del Rio, R. A. Gossage, M. S. Hannu, M. Lutz, A. L. Spek and
G. Van Koten, Can. J. Chem., 2000, 78, 1620.
R. M. Gauvin, H. Rozenberg, L. J. W. Shimon and D. Milstein,
Organometallics, 2001, 20, 1719.
3
4
5
6
7
B. Cetinkaya, E. Cetinkaya, M. Brookhart and P. S. White,
J. Mol. Catal. A: Chem., 1999, 142, 101.
M. W. Haenel, S. Oevers, K. Angermund, W. C. Kaska, H. Fan
and M. B. Hall, Angew. Chem., 2001, 40, 3596.
M. Gupta, W. C. Kaska and C. M. Jensen, Chem. Commun., 1997,
461.
R. G. Cavell, R. P. Kamalesh-Babu and K. Aparna, J.Organomet.
Chem., 2001, 617, 158.
New J. Chem., 2003, 27, 263–273
273