Substituents effects in POP pincer complexes of ruthenium

Article abstract of DOI:10.1021/om050053v

Reactions of [RuCl₂(p-cymene)]₂ with ( ^tBu₂PCH₂CH₂)₂O (POP- ^tBu) and (ⁱPr₂PCH₂CH ₂)₂O (POP-ⁱPr) afforded RuCl ₂(POP-^tBu) (1) and [Ru₂(μ-Cl) ₃(POP-ⁱPr)₂]Cl (2·Cl), respectively. The POP ligand is coordinated in a mer fashion in complex 1, whose crystal structure revealed a γ-agostic C-H...Ru interaction of one ^tBu group. Spectroscopic evidence indicated that this agostic interaction is retained in 1 in solution. A related compound, [Ru(N ₂)Cl(POP-^tBu)]BPh₄ (4), which also showed agostic bonding of a ^tBu group, was obtained by substitution of N₂ for Cl^- in 1, in the presence of NaBPh₄. Compound 2·Cl readily underwent ion exchange with LiBPh₄ or LiPF₆ to give 2·BPh₄ or 2·PF₆ salts, respectively. A crystallographic analysis of 2·PF₆ established a co-facial bioctahedral geometry of the [Ru₂(μ-Cl) ₃(POP-ⁱPr)₂]⁺ cation containing two POP ligands coordinated in a fac fashion. Reactions of 1 and 2 with H ₂ afforded the dihydrogen complexes cis,trans-Ru(H ₂)Cl₂(POP-^tBu) (3) and cis,cis-Ru(H ₂)Cl₂(POP-ⁱPr) (5), respectively. The H-H bond distances are very similar in both compounds, r(H-H) = 1.0 ± 0.1 A, based on the T_1min and J_HD data and results of DFT calculations. Reaction of 2 with N₂ gave the dinitrogen complex cis,cis-Ru(N₂)Cl₂(POP-ⁱPr) (6), but solutions of 1 under a nitrogen atmosphere showed no evidence of an analogous compound. The different steric requirements of the phosphorus substituents of the POP ligands were identified as the source of the differences in the coordination properties of the POP-^tBu and POP-ⁱPr complexes 1-6.

Full text of DOI:10.1021/om050053v

2492
Organometallics 2005, 24, 2492-2501
Substituents Effects in POP Pincer Complexes of
Ruthenium
Quinn Major,† Alan J. Lough,‡ and Dmitry G. Gusev*,†
Department of Chemistry, Wilfrid Laurier University, Waterloo, Ontario, Canada N2L 3C5,
and Department of Chemistry, University of Toronto, Toronto, Ontario, Canada M5S 3H6
Received January 25, 2005
Reactions of [RuCl2(p-cymene)]2 with (tBu2PCH2CH2)2O (POP-tBu) and (iPr2PCH2CH2)2O
(POP-iPr) afforded RuCl2(POP-tBu) (1) and [Ru2(µ-Cl)3(POP-iPr)2]Cl (2‚Cl), respectively. The
POP ligand is coordinated in a mer fashion in complex 1, whose crystal structure revealed
t
a γ-agostic C-H‚‚‚Ru interaction of one Bu group. Spectroscopic evidence indicated that
this agostic interaction is retained in 1 in solution. A related compound, [Ru(N2)Cl(POP-
tBu)]BPh4 (4), which also showed agostic bonding of a tBu group, was obtained by substitution
of N2 for Cl- in 1, in the presence of NaBPh4. Compound 2‚Cl readily underwent ion exchange
with LiBPh4 or LiPF6 to give 2‚BPh4 or 2‚PF6 salts, respectively. A crystallographic analysis
of 2‚PF6 established a co-facial bioctahedral geometry of the [Ru2(µ-Cl)3(POP-iPr)2]+ cation
containing two POP ligands coordinated in a fac fashion. Reactions of 1 and 2 with H2 afforded
the dihydrogen complexes cis,trans-Ru(H2)Cl2(POP-tBu) (3) and cis,cis-Ru(H2)Cl2(POP-iPr)
(5), respectively. The H-H bond distances are very similar in both compounds, r(H-H) )
1.0 ( 0.1 Å, based on the T1min and JHD data and results of DFT calculations. Reaction of 2
with N2 gave the dinitrogen complex cis,cis-Ru(N2)Cl2(POP-iPr) (6), but solutions of 1 under
a nitrogen atmosphere showed no evidence of an analogous compound. The different steric
requirements of the phosphorus substituents of the POP ligands were identified as the source
of the differences in the coordination properties of the POP-tBu and POP-iPr complexes 1-6.
Chart 1
Introduction
The importance of pincer ligands and their metal
complexes to organic synthesis and catalysis has been
recently underlined by the publication of two reviews.1
Chart 1 renders a generalized structural drawing of a
pincer complex featuring the characteristic mer-DGD
tridentate ligand set. The reactivity of such systems is
substantially altered by ligand modification, particularly
by variations in the pincer ligand backbone, donor
atoms, and their substituents. The donor atoms have a
strong effect on the electron density at the metal center,
whereas their substituents may create moderate to
strong steric congestion around the metal to control its
accessibility. Thus, fine-tuning of the reactivity of the
pincer complexes for catalysis and organic synthesis can
be accomplished through systematic manipulations of
the DGD set.
An interesting type of pincer ligand is one incorporat-
ing an oxygen and two phosphorus donor centers in a
POP fashion. In combination with a late transition
metal this would result in formation of a M(POP)
fragment with a weak OfM bond. This feature can
enhance bonding of other groups to the metal center at
the coordination site trans to the oxygen atom. The
simplest POP pincer ligand geometry comprises a di-
Chart 2
ethyl ether backbone connecting two dialkyl or diaryl
phosphine units (POP-R) as shown in Chart 2. A
number of phosphorus substituents have been incorpo-
rated in this ligand including R ) Ph, tBu, Et, Cy, and
the more exotic 3-pyridyl and 4-((diethylamino)methyl)-
phenyl groups. However, only a small number of POP-
based pincer complexes have been reported,2 and only
two, ReCl2(tN)(POP-Ph) and [Rh(CO)(POP-Ph)]+,2a,i
have been structurally characterized and their chem-
istry is underdeveloped. In the present work, our main
objective was to obtain first POP pincer complexes of
ruthenium and to study the effect of phosphine substi-
tution on the reactivity and structure of such species.
To this end, we used the known bis(2-(di-tert-butylphos-
phino)ethyl) ether ligand (POP-tBu) and prepared a new
bis(2-(diisopropylphosphino)ethyl) ether ligand (POP-
iPr) and isolated their ruthenium complexes from reac-
* To whom correspondence should be addressed. E-mail: dgoussev@
wlu.ca.
† Wilfrid Laurier University.
‡ University of Toronto.
(1) (a) Singleton, J. T. Tetrahedron 2003, 59, 1837. (b) van der Boom,
M. E.; Milstein, D. Chem. Rev. 2003, 103, 1759.
10.1021/om050053v CCC: $30.25 © 2005 American Chemical Society
Publication on Web 04/12/2005

POP Pincer Complexes of Ruthenium
Organometallics, Vol. 24, No. 10, 2005 2493
Scheme 1
tions with [RuCl2(p-cymene)]2. The structures of the
products and their reactivity toward hydrogen and
nitrogen have been investigated. The experimental
findings are presented and discussed in the following
sections.
solution. POP-iPr was isolated by distillation as a
pyrophoric colorless oil. The POP-tBu and POP-iPr
ligands were characterized by multinuclear NMR spec-
troscopy. The spectra of the former were very similar
to those reported in the literature.2g The latter ligand
gave spectra with signals of expected chemical shifts,
patterns, and coupling constants (see Experimental
Section). A property worth noting in the POP-iPr ligand
is the diastereotopic nature of the methyl groups in each
iPr substituent. In the NMR spectra, this gave rise to
two distinct methyl resonances for the four chemically
Results
Syntheses and Characterization of Bis(2-(di-
alkylphosphino)ethyl) Ether Ligands. The prepara-
tion of bis(2-(di-tert-butylphosphino)ethyl) ether (POP-
tBu) was carried out following the literature method.2g
The synthetic methodology used to prepare POP-iPr was
that employed in the syntheses of other diphosphino-
ethyl ether ligands2j and proceeded by reacting bis(2-
chloroethyl) ether with 2 equiv of lithium diisopropyl-
phosphide, which was made in situ from commercially
available chlorodiisopropylphosphine and lithium metal.
i
equivalent Pr groups.
Syntheses of Ruthenium Complexes. An outline
of synthetic routes developed in this investigation is
presented in Scheme 1. Addition of POP-tBu and POP-
iPr to ruthenium was achieved by substitution of p-
cymene in [RuCl2(p-cymene)]2. Heating solutions of
[RuCl2(p-cymene)]2 with 2 equiv of POP-tBu or POP-
iPr afforded the chlorides RuCl2(POP-tBu) (1) and
[(POP-iPr)Ru(µ-Cl3)Ru(POP-iPr)]Cl (2‚Cl), respectively.
Good yields of crystalline material were obtained in both
cases. Compound 2‚Cl readily underwent anionic sub-
stitution with LiBPh4 or LiPF6 in methanol. The ex-
change reactions afforded the corresponding salts [(POP-
iPr)Ru(µ-Cl3)Ru(POP-iPr)]BPh4 (2‚BPh4) and [(POP-
iPr)Ru(µ-Cl3)Ru(POP-iPr)]PF6 (2‚PF6), respectively.
Compound 2‚BPh4 precipitated immediately from the
methanol solution, while 2‚PF6 crystallized slowly over
several hours.
ClPiPr2 + 2Li f LiPiPr2 + LiCl
2LiPiPr2 + (ClCH2CH2)2O f
(iPr2PCH2CH2)2O + 2LiCl
Both reactions were monitored by 31P{1H} NMR and
indicated essentially quantitative product formation in
(2) (a) Bolzati, C.; Boschi, A.; Uccelli, L.; Tisato, F.; Refosco, F.;
Cagnolini, A.; Duatti, A.; Prakash, S.; Bandoli, G.; Vittadin, A. J. Am.
Chem. Soc. 2002, 124, 11468. (b) Buhling, A.; Kamer, P. C. J.; van
Leeuwen, P. W. N. M.; Elgersma, J. W.; Goubitz, K.; Fraanje, J.
Organometallics 1997, 16, 3027. (c) Vogl, E. M.; Bruckmann, J.;
Kessler, M.; Kru¨ger, C.; Haenel, M. W. Chem. Ber./Recl. 1997, 130,
1315. (d) Steffey, B. D.; Miedaner, A.; Maciejewski-Framer, M. L.;
Bernatis, P. R.; Herring, A. M.; Allured, V. S.; Carperos, V.; DuBois,
D. L. Organometallics 1994, 13, 4844. (e) George, T. A.; Jackson, M.
A.; Kaol, B. B. Polyhedron 1991, 10, 467. (f) George, T. A.; Jackson,
M. A. Inorg. Chem. 1988, 27, 924. (g) Timmer, K.; Thewissen, D. H.
M. W.; Marsman, J. W. Recl. Trav. Chim. Pays-Bas 1988, 107, 248.
(h) Thewissen, D. H. M. W.; Timmer, K.; Noltes, J. G.; Marsman, J.
W.; Laine, R. M. Inorg. Chim. Acta 1985, 97, 143. (i) Alcock, N. W.;
Brown, J. M.; Jeffery, J. C. J. Chem. Soc., Dalton Trans. 1976, 583. (j)
Green, P. T.; Sacconi, L. J. Chem. Soc. A 1970, 866. (j) Sacconi, L.;
Gelsomini, J. Inorg. Chem. 1968, 7, 291.
No N2 coordination to ruthenium was observed in
solutions of 1; however H2 addition to 1 resulted in
formation of the dihydrogen complex cis,trans-Ru(H2)-
Cl2(POP-tBu) (3). The cationic dinitrogen complex [Ru-
(N2)Cl(POP-tBu)]BPh4 (4) was obtained when 1 was
reacted with NaBPh4 in CH2Cl2 under an atmosphere
of nitrogen. An important structural feature of com-
plexes 1 and 4 that is not represented in the simplified
drawings of Scheme 1 is C-H‚‚‚Ru agostic bonding
t
involving a Bu group of the POP-tBu ligand.

2494 Organometallics, Vol. 24, No. 10, 2005
Major et al.
Scheme 2
Scheme 3
Figure 1. ORTEP and atom-labeling scheme for 1 with
the ellipsoids at 30%. Most of the hydrogen atoms are
omitted for clarity. Selected bond distances (Å) and angles
(deg): Ru-Cl1 2.3505(8), Ru-Cl2 2.3901(7), Ru-O 2.123-
(2), Ru‚‚‚H15b 2.23, P1-Ru-P2 161.1(1), P1-Ru-O 82.0-
(1), P2-Ru-O 80.7(1), Cl1-Ru-Cl2 93.0(1), Cl1-Ru-P1
95.2(1), Cl1-Ru-P2 93.7(1), Cl1-Ru-O 93.6(1), Cl2-Ru-
P2 98.7(1), Cl2-Ru-P1 97.5(1), Cl2-Ru-O 173.5(1).
tion and chemical shifts indicated an assignment of 4
t
× Bu, 2 × PCH2, and 2 × OCH2, respectively. The
spectra were not consistent with the chiral solid-state
structure of 1, which possesses four unique tBu groups
and eight unique protons of the CH2 groups. A combina-
tion of four independent processes can explain the room-
temperature spectra. One is the inversion of the struc-
ture described in Scheme 2 involving chloride ligand
exchange that makes the POP plane an effective mo-
lecular symmetry plane. Another is the reversible
cleavage of the C-H‚‚‚Ru agostic bonding, as shown in
Scheme 3, that results in a time-averaged O-Cl-Cl
Heating solutions of 2‚Cl under H2 or N2 afforded the
dihydrogen or dinitrogen complexes, cis,cis-Ru(H2)-
Cl2(POP-iPr) (5) or cis,cis-Ru(N2)Cl2(POP-iPr) (6), re-
spectively. Interconversion of 5 and 6 could be easily
accomplished by degassing and addition of N2 (5 f 6)
or H2 (6 f 5) to a solution of the appropriate starting
compound. In solutions of 3 and 5 under an atmosphere
of D2, isotopic substitution and formation of the Ru(HD)-
Cl2(POP) isotopomers was observed by NMR spectros-
copy within minutes at 50 °C. A prolonged exposure to
D2 (>24 h) resulted in the Ru(D2)Cl2(POP) isotopomers.
Characterization of Complexes. Complexes 1,
2‚PF6, 4, 5, and 6 were crystallized, and their molecular
structures were determined by X-ray crystallography.
All compounds 1-6 were characterized by multinuclear
NMR spectroscopy and elemental analysis.
t
mirror plane. The third process is rotation of the Bu
groups, averaging the methyl groups. The fourth process
involves the interconversion of the agostic C-H bond
with two pendent C-H bonds within the agostic methyl
group, i.e., methyl rotation, averaging all three hydro-
gens.
Variable-temperature NMR spectra revealed these
processes in a stepwise manner. At -30 °C, the chloride
ligand exchange in Scheme 2 became slow. At this
1
temperature the H NMR spectrum showed decoales-
cence of the broad tBu signal into two virtual triplets (δ
0.50, 1.40) and decoalescence of the PCH2 and OCH2
signals into two multiplets, each with an integration of
4 × 2H. Further lowering the temperature to -110 °C
Crystal Structure of RuCl2(POP-tBu) (1). The
structure of 1, shown in Figure 1, confirms the triden-
tate coordination of the mer-POP pincer ligand. The
geometry around ruthenium can be considered ap-
proximately octahedral with one coordination site oc-
t
caused decoalescence of the Bu resonances into four
signals (δ 1.72, 1.58, 1.20, and -2.28) of approximate
intensity 6H:12H:12H:6H. The signals were broad and
the three downfield signals showed some overlap, but
clearly they were due to methyl protons and indicated
t
cupied by an agostic C-H bond of a Bu group. The
γ-agostic C15-H15b‚‚‚Ru interaction is indicated by the
geometric parameters of the POP ligand. The Ru-P2-
C13 ) 101.7(1)° and P2-C13-C15 ) 99.8(2)° angles are
significantly smaller than the corresponding Ru-P1-
C5 ) 116.6(1)° and P1-C5-C7 ) 106.9(2)° angles,
respectively. The Ru-P2 ) 2.320(1) Å bond is 0.085 Å
shorter than the Ru-P1 ) 2.405(1) Å bond due to a
contraction in the Ru-P2-C13-C15-H15b cycle. The
Ru-C15 separation of 2.845 Å may seem long for an
agostic interaction, but precedence has been established
and longer C-H‚‚‚Ru interactions have been observed
between a tBu group of a PtBu2Me ligand and a Ru atom
in [Ru{CHdC(SiMe3)(Ph)}(CO)(PtBu2Me)]+ [Ru‚‚‚C-
(agostic) ) 3.049 Å]3 and [Ru{(Me3Si)HCdC-CHdCH-
(SiMe3)}(CO)(PtBu2Me)]+ [Ru‚‚‚C(agostic) ) 2.943 Å].4
Spectroscopic Characterization of 1 and NMR
Evidence for Agostic Interaction. The ambient-
temperature 1H NMR spectrum of 1 in CD2Cl2 displayed
three broad signals (δ 0.99, 2.29, 4.48), whose integra-
t
slow rotation of the Bu groups at -110 °C. Specific
t
assignments for the three downfield Bu resonances
cannot not be made, but the upfield chemical shift of
the -2.28 ppm resonance is in agreement with the
C-H‚‚‚Ru agostic interaction present in solution as well
as in the crystal. The integration of 6H for this signal
is consistent with two dynamic processes still occurring
rapidly at the low temperature according to Scheme 3
and involving two methyl groups associated with C7 and
C15 in Figure 1: (i) exchange between the agostic CH3
group and the pendant CH3 group across the O-Cl-Cl
plane and (ii) rotation of the two methyl groups.
Spectroscopic Characterization and Solid-State
Molecular Structure of 2. Compound 2‚Cl proved to
be insoluble in many organic solvents of varying polarity
(hexane, benzene, tetrahydrofuran, and acetone), but it
was moderately soluble in CH2Cl2 and very soluble in
methanol, ethanol, and 2-propanol. These solubility
properties provided the first evidence of the ionic nature
of the compound. The NMR spectroscopic analysis
indicated an asymmetric environment for the POP
(3) Huang, D.; Folting, K.; Caulton, K. G. J. Am. Chem. Soc. 1999,
121, 10318.
(4) Huang, D.; Oliva´n, M.; Huffman, J. C.; Eisenstein, O.; Caulton,
K. G. Organometallics 1998, 17, 4700.

POP Pincer Complexes of Ruthenium
Organometallics, Vol. 24, No. 10, 2005 2495
Figure 3. ORTEP view of the cation in 2‚PF6 showing the
C2 symmetry axis. Carbon and hydrogen atoms are omitted
for clarity.
Figure 2. ORTEP and atom-labeling scheme of 2‚PF6 with
thermal ellipsoids at 30%. The hydrogen atoms are omitted
for clarity. Selected bond distances (Å) and angles (deg):
Ru1-P1 2.284(2), Ru1-P2 2.278(2), Ru2-P3 2.262(2),
Ru2-P4 2.289(2), Ru1-O1 2.136(6), Ru2-O2 2.147(6),
Cl1-Ru1-Cl2 81.4(1), Cl1-Ru2-Cl2 78.2(1), Cl1-Ru1-
Cl3 78.0(1), Cl1-Ru2-Cl3 81.2(1), Cl1-Ru1-P1 90.7(1),
Cl1-Ru2-P3 92.0(1), Cl1-Ru1-O1 91.3(2), Cl1-Ru2-O2
91.8(2), Cl2-Ru1-Cl3 82.6(1), Cl2-Ru2-Cl3 82.3(1), Cl2-
Ru1-P1 102.1(1), Cl2-Ru1-P2 102.2(1), Cl2-Ru2-P4
88.3(1), Cl2-Ru2-O2 90.0(2), Cl3-Ru1-P1 167.1(1), Cl3-
Ru2-P3 101.4(1), Cl3-Ru1-P2 88.7(1), Cl3-Ru2-P4 102.8-
(1), Cl3-Ru1-O1 89.7(2), Cl3-Ru2-O2 170.6(2), P1-
Ru1-P2 101.8(1), P3-Ru2-P4 101.0(1), P1-Ru1-O1
84.3(2), P3-Ru2-O2 85.1(2), P2-Ru1-O1 83.4(2), P4-
Ru2-O2 82.5(2).
atoms: Ru1 T Ru2, Cl2 T Cl3, P1 T P3, P2 T P4, O1
T O2 (Figure 3). Thus, in solution the chemical equiva-
lence of the two POP ligands is expected. The asym-
metry within each POP ligand, as indicated by the NMR
spectra, is also evident from the structure in Figure 3.
The question of whether Ru-Ru bonding is present
in 2 can be answered by analysis of the Ru-Ru distance.
For direct Ru-Ru bonding, the usual range of distances
is 2.28-2.95 Å.5 Elongated Ru-Ru bonds have been
reported to be in the range 2.9-3.1 Å,6 and some very
long second-row transition metal-metal bonds have
been reported to be as long as 3.2 Å.7 The Ru-Ru
separation in 2, 3.29 Å, is too long to support any
significant metal-metal bonding interaction. It is simi-
lar to the Ru-Ru separation (3.44-3.35 Å) in related
complexes with a Ru(II)(µ-Cl)3Ru(II) core.8 Theoretically,
there should be no metal-metal bonding in a 36-
electron, saturated binuclear complex.
1
ligand. The H NMR showed three regions of overlap-
ping signals between δ 1.35-1.70, 1.94-2.13, and 3.14-
3.4 with integrations of 26H, 4H, and 6H, respectively.
The 31P{1H} NMR showed two signals, both doublets
2
of equal coupling, JPP ) 29.6 Hz, consistent with two
Spectroscopic Characterization of the Dihydro-
inequivalent PiPr2 groups coordinated in a cis fashion
to the same ruthenium atom. The 13C{1H} NMR showed
eight and four signals for the primary and tertiary
1
gen Complexes 3 and 5. The H NMR spectra of 3
and 5 exhibited one resonance of intensity 2H in the
hydride region (δ -10.1, 3; -14.7, 5). The resonance of
3 was observed as a triplet with 2JHP ) 9 Hz, while that
of 5 was a broad singlet (w1/2 ) 7 Hz). In the region of
the phosphine substituents, tBu and iPr, two resonances
were observed which showed the virtual coupling phe-
nomenon characteristic of trans P-Ru-P bonding and
indicating the meridional coordination mode of the POP
ligands. This was supported by a single resonance in
i
carbons of the Pr groups, respectively, indicating a
i
unique chemical environment for each Pr group. The
inequivalence of the two phosphorus groups also re-
sulted in the chemical inequivalence of the two OCH2
and PCH2 carbons of the POP ligand. Further NMR
analysis of 2 with HETCOR and DEPT allowed detailed
assignments to be made (see Experimental Section) but
could not further elaborate the structure of the complex.
Attempts to crystallize 2‚Cl were unsuccessful; how-
ever, suitable crystals of 2‚PF6 were obtained and
subjected to X-ray analysis. The cation in 2‚PF6 has a
core structure containing two Ru(II) centers and an
overall co-facially bioctahedral geometry (Figure 2). The
molecule has one symmetrically bridging (Ru1-Cl1 )
2.488(2) Å, Ru2-Cl1 ) 2.485(2) Å) and two asymmetri-
cally bridging chloride ligands (Ru1-Cl2 ) 2.389(2) Å,
Ru2-Cl2 ) 2.560(2) Å and Ru1-Cl3 ) 2.546(2) Å, Ru2-
Cl3 ) 2.385(2) Å). For the asymmetric chlorides, the
shorter bond is trans to oxygen and the longer is trans
to phosphorus. For example, Cl2 is trans to P3 through
Ru2 (Cl2-Ru2-P3 ) 168.9(1)°) and trans to O1 through
Ru1 (Cl2-Ru1-O1 ) 170.3(1)°). Similar bond angles
are observed for Cl3, but Cl1 is trans to two phosphorus
atoms (Cl1-Ru1-P2 ) 165.8(1)° and Cl1-Ru2-P4 )
165.3(1)°); consequently the two Ru-Cl1 separations are
identical. These differences result from the much weaker
trans influence of the coordinated oxygen donor com-
pared to that of the phosphorus. Ignoring the iPr groups,
2 has a C2 axis extending through Cl1 and the midpoint
of the Ru-Ru vector. The C2 operation interchanges the
1
the 31P{1H} NMR spectra of both complexes. The H
spectrum of 3 indicated two chemical environments for
t
the Bu groups, while in 5 (with diastereotopic methyl
1
groups) the H spectrum pointed to equivalence of all
four iPr groups. These observations suggested effective
Cs and C2v symmetries for the structures of 3 and 5,
respectively, according to Scheme 1. The 1H NMR data
for the PCH2 and OCH2 groups, along with the 13C{1H}
NMR data, were consistent with these conclusions.
Estimates of the H-H Distance in 3 and 5 in
Solution (T1min and JHD). To further characterize the
dihydrogen ligands in 3 and 5, we obtained estimates
(5) (a) Dekleva, T. W.; Thorburn, I. S.; James, B. R. Inorg. Chim.
Acta 1985, 100, 49. (b) Mattson, B. M.; Heiman, J. R.; Pignolet, L. H.
Inorg. Chem. 1976, 15, 564. (c) Schumann, H.; Opitz, J.; Pickardt, J.
J. Organomet. Chem. 1977, 128, 253. (d) Jones, R. A.; Wilkinson, G.;
Coloquohoun, I. J.; McFarlane, W.; Galas, A. M. R.; Hursthouse, M.
B. J. Chem. Soc., Dalton Trans. 1980, 2480.
(6) (a) Keijsper, J.; Polm, L. H.; van Koten, G.; Vrieze, K.; Seignette,
P. F. A. B.; Stam, C. H. Inorg. Chem. 1985, 24, 518. (b) Carty, A. J.;
MacLaughlin, S. A.; van Wagner, J.; Taylor, N. J. Organometallics
1982, 1, 1013.
(7) Dahl, L. F.; Wampler, D. L. J. Am. Chem. Soc. 1959, 81, 3150.
(8) Seddon, E. A.; Seddon, K. R. The Chemistry of Ruthenium;
Elsevier: Amsterdam, 1984; pp 310, 487-515.

2496 Organometallics, Vol. 24, No. 10, 2005
Major et al.
Figure 4. Crystal structure of 5 with the ellipsoids at 30%.
Most of the hydrogen atoms are removed for clarity.
Selected bond distances (Å) and angles (deg): Ru-P1
2.350(1), Ru-Cl1 2.4093(9), Ru-Cl2 2.4076(8), Ru-O
2.167(2), Ru‚‚‚H1ru, H2ru 1.67(2), H1ru-H2ru 0.90, P1-
Ru-P2 163.46(4), P1-Ru-O 81.74(7), Cl1-Ru-Cl2 177.08-
(3), Cl1-Ru-P1 91.60(3), Cl1-Ru-O 89.87(6), Cl2-Ru-
P1 88.04(3), Cl2-Ru-O 87.21(6).
Figure 5. Theoretical structures of complexes 3 and 5.
The ZPE-corrected energies are relative to the more stable
isomer in each pair.
with the differences within 0.03 Å and 2° for the bond
distances and angles, respectively. Isomer 5′, possessing
H2 trans to a chloride, is 9.2 kcal/mol less stable than
5′′. The higher stability of 5′′ over 5′ is probably due to
somewhat stronger bonding of the H2 trans to a weak
donor oxygen atom in 5′′, which manifests in the slightly
longer H-H distance of 0.925 Å in 5′′ versus 0.907 Å in
5′. It is however difficult to say whether this effect is
due to stronger σ-bonding (H2fRu) or better back-
bonding (H2rRu), or a combination of both.
of the H-H distances by measuring the T1min (minimum
spin-lattice relaxation times) of the H2 resonances
along with the JHD (H-D coupling constants) of the
monodeuterated isotopomers Ru(HD)Cl2(POP-tBu) (3-
d) and Ru(HD)Cl2(POP-iPr) (5-d). The T1min times for 3
and 5 were found to be 16.1 and 14.6 ms, respectively,
at -55 °C (300 MHz, toluene-d8), Using the methodology
of Halpern and co-workers we calculated the H-H
separations of 1.13 Å in 3 and 1.11 Å in 5.9 No correction
for rapid H2 spinning has been applied in these calcula-
tions because of the computational evidence (see below)
indicating that the H2 ligands of 3 and 5 are undergoing
2-fold reorientation that should have no effect on the
T1.10 The JHD couplings in 3-d and 5-d are 27.0 and 27.3
Hz, respectively, and these values correspond to H-H
separations of 1.00 Å in 3 and 0.99 Å in 5.11
Solid-State Molecular Structure of 5. The dihy-
drogen complex 5 was crystallized and characterized by
X-ray diffraction. The crystallographic results are pre-
sented in Figure 4. The crystal structure is in agreement
with the structure deduced from the NMR data in
solution, where 5 is expected to have an effective C2v
symmetry due to a combination of rapid rotation of the
iPr groups and conformational nonrigidity of the POP
ligand backbone. Although complex 3 was also crystal-
lized, crystallographic determination of its molecular
structure was unsuccessful due to disorder problems.
Theoretical Structures of 3 and 5. To better
understand structural preferences in the system of
complexes 3 and 5, we carried out a series of DFT
calculations and optimized four geometries presented
in Figure 5 possessing the dihydrogen ligand trans to
chloride (3′ and 5′) or trans to oxygen (3′′ and 5′′). For
Ru(H2)Cl2(POP-iPr), the more stable isomer 5′′ has the
structure closely resembling the crystal structure of 5,
In the system of Ru(H2)Cl2(POP-tBu) the structural
preferences are reversed and the Cs symmetrical isomer
3′ is 7.3 kcal/mol more stable than the approximately
C2v symmetrical 3′′, in agreement with the solution
NMR data for 3. The main difference between 3 and 5
is the size of the groups on phosphorus, and the trans-
dichloride structure 3′′ is disfavored for steric reasons.
The steric problems in 3′′ become obvious at once when
the C-H hydrogens of the iPr groups in 5′′ are replaced
by methyls, because this brings two CH3 groups within
1.40 Å of Cl2. To avoid such close contacts, the PtBu2
groups had to move during the DFT optimization of 3′′,
which resulted in a distorted POP ligand backbone with
the P1-C1 and P2-C4 bonds rotated out of the POP
plane. We have earlier commented12 on the structure
of related ruthenium complexes derived from 1,5-bis-
(di-tert-butylphosphino)pentane and explained their
preference for square-pyramidal geometry by the steric
t
requirements of the Bu2 groups, two of which occupy
one coordination site, limiting the coordination number
to five. A similar situation is seen in 1, where the
structure is furthermore stabilized by agostic bonding.
In 3′ and 5′′, the dihydrogen ligand is aligned along
the O-Cl2 and P1-P2 axes, respectively. For these two
complexes we studied the process of H2 rotation around
the Ru-H2 bond. Single transition-state structures were
found, 3′ts and 5′′ts, with the relative energies of 0.9
and 0.7 kcal/mol above 3′ and 5′′, respectively. These
energy barriers are very low and, in solution, the H2
reorientation should be fast on the time scale of molec-
ular tumbling. In both transition-state geometries the
coordinated H2 is rotated by 90° relative to the ground-
state orientation; that is, the H-H is coplanar with P1
(9) Desrosiers, P. J.; Cai, L.; Lin, Z.; Richards, R.; Halpern, J. J.
Am. Chem. Soc. 1991, 113, 4173.
(10) (a) Zilm, K. W.; Millar, J. M. Adv. Magn. Opt. Reson. 1990, 15,
163. (b) Gusev, D. G.; Kuhlman, R. L.; Renkema, K. B.; Eisenstein,
O.; Caulton, K. G. Inorg. Chem. 1996, 35, 6775. (c) Facey, G. A.; Fong,
T. P.; Gusev, D.; Macdonald, P. M.; Morris, R. H.; Schlaf, M.; Xu, W.
Can. J. Chem. 1999, 77, 1899.
(11) (a) Gusev, D. G. J. Am. Chem. Soc. 2004, 126, 14249. (b)
Calculated using eq 3 from ref 11a.
(12) Gusev, D. G.; Lough, A. J. Organometallics 2002, 21, 5091.

POP Pincer Complexes of Ruthenium
Organometallics, Vol. 24, No. 10, 2005 2497
Figure 6. ORTEP and atom-labeling scheme of the cation
in 4 with the ellipsoids at 30%. Most of the hydrogen atoms
are omitted for clarity. Selected bond distances (Å) and
angles (deg): Ru-Cl 2.337(1), Ru-O 2.117(3), Ru-N1
1.946(3), N1-N2 1.054(5), Ru‚‚‚H19b 2.08, P1-Ru-P2
162.81(4), P1-Ru-O 81.87(7), P2-Ru-O 82.39(7), Cl-
Ru-N1 88.2(1), Cl-Ru-P1 95.21(4), Cl-Ru-P2 91.62(4),
Cl-Ru-O 89.78(8), N1-Ru-P1 97.0(1), N1-Ru-P2 99.0-
(1), N1-Ru-O 177.6(1), Ru-N1-N2 175.2(4).
Figure 7. ORTEP and atom-labeling scheme of 6 with the
ellipsoids at 30%. The hydrogen atoms are omitted for
clarity. Selected bond distances (Å) and angles (deg): Ru-
Cl1 2.395(1), Ru-O 2.147(3), Ru-Cl2 2.420(1), Ru-N1
1.893(3), Ru-P1 2.368(1), P1-Ru-P2 163.8(1), Cl1-Ru-
Cl2 176.8(1), O-Ru-N1 179.7(1), Cl1-Ru-P1 89.8(1),
Cl1-Ru-N1 90.6(1), Cl1-Ru-O 89.2(8), Cl2-Ru-P1 90.4-
(1), Cl2-Ru-N1 92.5(1), Cl2-Ru-O 87.7(1), P1-Ru-O
81.8(1), P1-Ru-N1 98.4(1), N2-N1-Ru, 179.3(3).
and P2 in 3′ts and is coplanar with Cl1 and Cl2 in 5′′ts.
Thus, the intramolecular reorientation of the dihydro-
gen ligand around the Ru-H2 bond comprises 2-fold
jumps rather than a continuous spinning.
) 117.6(1)° and P1-C9-C10 ) 107.8(3)° angles, re-
spectively. The Ru-P2 ) 2.340(1) Å bond is 0.093 Å
shorter than the Ru-P1 ) 2.433(1) Å bond due to a
contraction in the Ru-P2-C17-C19-H19b cycle. After
comparing these data with the corresponding geo-
metrical parameters in 1, it appears that the degree of
contraction is slightly greater in 4. The agostic Ru-C19
separation is also slightly shorter in 4: 2.740 Å versus
the 2.845 Å Ru-C15 distance in 1. This C-H‚‚‚Ru
agostic bonding persists in 4 in solution. Above -60 °C,
The calculated r(H-H) ) 0.93 Å is the same in 3′ and
5′′ and is shorter than the experimental values of 1.00
Å in 3 and 0.99 Å in 5 derived from the JHD couplings,
which in turn are shorter than the distances determined
from the T1min relaxation times. These differences are
natural because the theoretical H-H distances cor-
respond to electronic minimums, whereas the values
obtained by experimental means can be to a different
degree affected by anharmonic vibrations of the hydro-
gen atoms13 and, for dihydrogen complexes, systemati-
cally appear longer than the distances from DFT
calculations.11
Solid-State Molecular Structure and Spectro-
scopic Characterization of 4 and 6. These two
species provide another example of the dramatic differ-
ence the substituents on phosphorus make for the
stability and structure of the POP pincer complexes.
After establishing in the preceding section that the
vacant site in the pyramidal cis-RuCl2(POP-tBu) frag-
ment is unfavorable for steric reasons, the lack of
dinitrogen coordination to 1, and so the instability of
cis,trans-Ru(N2)Cl2(POP-tBu), come as no surprise. It is
interesting, however, that a dinitrogen complex could
be made from 1 in the presence of NaBPh4, via substi-
tution of a chloride ligand. An elemental analysis of the
product crystallized from dichloromethane confirmed
the composition, [Ru(N2)Cl(POP-tBu)]BPh4‚CH2Cl2 (4).
A strong peak was observed in the IR spectrum of 4 due
to the NtN stretch at 2143 cm-1. Finally, a crystal
structure of 4 was determined, which provided impor-
tant structural details.
1
the H NMR spectra of 4 showed two 1:1 resonances
due to tBu groups at δ 1.44 and 0.40. The latter became
very broad at -60 °C and decoalesced at -80 °C to give
one broad peak of intensity 12H between δ 1.1-1.5 and
another of intensity 6H at δ -1.9 ppm. At -100 °C, the
tBu region of the proton spectrum of 4 looked very
similar to that of 1. Obviously the dynamic process
proposed in Scheme 3 operates in both compounds at
-100 °C.
We next turn our attention to the dinitrogen complex
Ru(N2)Cl2(POP-iPr) (6), whose formation is in accord
with the existence of the analogous dihydrogen complex
Ru(H2)Cl2(POP-iPr) (5). Notwithstanding the absence of
a hydride resonance, the 1H and 13C{1H} NMR spectra
of 6 were very similar to those of 5. In addition, the IR
spectrum of 6 showed a very strong ν(NtN) absorbance
at 2119 cm-1, a frequency lower than ν(NtN) ) 2143
cm-1 for the cationic 4, as expected. These results
indicated a structure depicted in Scheme 1 similar to
that observed for 5 except with a N2 ligand in place of
the H2 ligand. Confirmation was provided by an X-ray
crystallographic analysis, the results of which are
presented in Figure 7.
The structure of 6 is octahedral with the POP ligand
in a pincer coordination mode. It shows two trans
chlorides and a η1-N2 ligand trans to the oxygen atom.
There is some disorder in one iPr group (C16 and C16A),
but otherwise the structure is well defined. The most
notable feature of 6 is the dinitrogen ligand. The N-N
separation is 1.101(5) Å and is not statistically different
from the corresponding separation in free N2 (1.0975
Å). Thus the dinitrogen ligand in 6 is only marginally
activated.14 Other known pincer complexes of ruthenium
show weak activation of coordinated dinitrogen and
The structure of the cation in 4 is shown in Figure 6
and closely resembles that of the parent compound 1 in
Figure 1. Except for the dinitrogen ligand that replaced
a chloride, 4 appears identical with 1 and likewise
t
features agostic bonding of a Bu group. This γ-agostic
C19-H19b‚‚‚Ru interaction makes 4 asymmetric. The
Ru-P2-C17 ) 98.0(1)° and P2-C17-C19 ) 99.7(3)°
angles are smaller than the corresponding Ru-P1-C9
(13) Heinekey, D. M.; Lledo´s, A.; Lluch, J. M. Chem. Soc. Rev. 2004,
33, 175.

2498 Organometallics, Vol. 24, No. 10, 2005
Major et al.
Scheme 4
Chart 3
is smaller than chloride. A qualitative estimate of the
relative sizes of η2-H2 and chloride ligands may be
obtained if it is assumed that the steric profile of each
ligand is circular from the perspective of the metal
center and radiates out from the center of the ligand to
the limit of the van der Waals radius. The size of a
chloride is then 1.8 Å, i.e., its van der Waals radius.17
The size of the η2-H2 ligands in 3 and 5 is estimated to
be 1.7 Å, which is arrived at by adding one-half the H-H
bond distance (0.5 Å) to the van der Waals radius of
hydrogen (1.2 Å).17 By these estimates, even along its
longest axis the H2 ligand is smaller than chloride. The
effective size of coordinated dihydrogen is even smaller
in the direction perpendicular to the H-H bond, and
the ligand can rotate away from bulky groups. The
second factor that influences the amount of steric
have a propensity to undergo ligand substitution at the
N2 site.15,16 The N2 ligand in 6 also displays this
reactivity and is easily displaced by H2. After 1 h at
room temperature a benzene solution of 6, which was
placed under a H2 atmosphere, began to show the
1
characteristic resonances of 5 in the H and 31P{1H}
NMR spectra. Full conversion of 6 to 5 was accom-
plished by repeated degas and H2 backfill cycles. This
conversion is reversible, i.e., 5 to 6, by several degas
and N2 backfill cycles (Scheme 1).
t
interaction with the Bu substituents in 3 and 5 is the
bond length. In these complexes, the Ru-H and Ru-
Cl distances compare as ca. 1.63 and 2.42 Å, respec-
tively. This places the H2 ligand much closer to ruthe-
nium and away from the tBu substituents, as illustrated
in Chart 3.
Discussion
The dinitrogen complexes obtained in this investiga-
tion further demonstrated the influence of the bulky tBu
substituents. Compound 2 reacted with N2 to give 6, yet
1 did not react with N2 to form an analogous compound,
nor did 3 substitute N2 for H2. An argument similar to
that presented above can rationalize these observations.
The Ru-N1 bond in 6 is 1.89 Å and the Ru-N2 distance
is 2.99 Å. This would place an N2 ligand, if it were
coordinated at the axial site in 1, in an area occupied
Several times in this investigation we have seen that
the chemistry of POP complexes is strongly influenced
by the steric demand of the substituents on phosphorus.
The reaction of [RuCl2(p-cymene)]2 and POP-tBu yielded
the mer-POP complex 1, while the reaction of POP-iPr
afforded the dimer 2 with two fac-POP ligands. Forma-
tion of 1 and 2 might proceed by a similar mechanism
depicted in Scheme 4. In the first step, 2 equiv of POP
displace p-cymene in [RuCl2(p-cymene)]2 to form a short-
lived intermediate, Ru2(µ-Cl)2Cl2(POP)2 (A). In the next
step, Cl- dissociates from A to afford [Ru2(µ-Cl)3(POP)2]-
Cl (B), which was isolated in the case of the POP-iPr
ligand. The last step involves the reaction of B with H2
or N2 to form C. In the case of POP-tBu, the larger tBu
substituents strongly destabilize the hypothetical dimers
A and B, which dissociate to give 1.
Another demonstration of the role of phosphine sub-
stituents in the chemistry of POP pincer complexes is
the difference in the structure of 3 and 5. In 5, the H2
ligand is trans to the oxygen in the equatorial plane
(defined as the POP plane), while in 3, the H2 ligand
occupies the most crowded axial site. Two properties of
H2 ligand, compared to chloride, make it the more
favorable ligand for a crowded environment. First, H2
t
by the Bu groups, much like the chloride in Chart 3.
The isolation of the cationic dinitrogen species [Ru-
(N2)Cl(POP-tBu)]+ (4) via substitution of a chloride in
1 decisively demonstrated that the lack of N2 coordina-
tion to 1 could not be due to an electronic effect.
There is another important structural aspect of the
chemistry of the pincer ligands that should be discussed.
Milstein and co-workers have recently reported a reac-
tion of RuCl2(PPh3)3 and PNP-tBu (PNP-tBu ) 2,6-bis-
(di-tert-butylphosphinomethyl)pyridine) in THF that
afforded a mixture of RuCl2(N2)(PNP-tBu) and a dini-
trogen-bridged dimer complex, Ru2Cl4(µ-N2)(PNP-tBu)2,
shown in Scheme 5.18 The two products existed in
equilibrium dependent on the concentration of the
complex and the amount of N2 in the system. Under a
nitrogen atmosphere, the equilibrium favored the dimer
complex by a factor of 10:1. Under an argon purge,
nearly all of the monomer was transformed into the
dimer.
(14) (a) MacKay, B. A.; Fryzuk, M. D. Chem. Rev. 2004, 104, 385.
(b) Fryzuk, M. D.; Johnson, S. A. Coord. Chem. Rev. 2000, 200-202,
379.
(15) Gusev, D. G.; Dolgushin, F. M.; Antipin, M. Y. Organometallics
2000, 19, 3429.
(16) (a) del R´ıo, I.; Back, S.; Hannu, M. S.; Rheinwald, G.; Lang,
H.; van Koten, G. Inorg. Chim. Acta 2000, 300-302, 1094. (b) del R´ıo,
I.; Gossage, R. A.; Hannu, M. S.; Lutz, M.; Spek, A. L.; van Koten, G.
Organometallics 1999, 18, 1097.
(17) Pauling, L. The Nature of the Chemical Bond, 3rd ed.; Cornell
University Press: Ithaca, 1960.
(18) Zhang, J.; Gandelman, M.; Shimon, L. J. W.; Rozenberg, H.;
Milstein, D. Organometallics 2004, 23, 4026.

POP Pincer Complexes of Ruthenium
Organometallics, Vol. 24, No. 10, 2005 2499
effectively more bulky ligand than the related PNP-t-
Bu system.
Scheme 5
Experimental Section
All manipulations were preformed under nitrogen in a
drybox or under argon using standard Schlenk techniques. All
solvents were purchased anhydrous and were stored and
dispensed in a drybox. Deuterated solvents were deoxygenated
and dried by standard methods.19 NMR spectra were recorded
on a Varian Unity Inova 300 NMR spectrometer. Spin-lattice
relaxation times T1 were determined as a function of temper-
ature by the inversion-recovery method. Infrared spectra were
recorded on a Perkin-Elmer Spectrum BX FT-IR spectrometer.
Preparation of [RuCl2(p-cymene)]2 was preformed as described
in the literature.20 Bis(2-(di-tert-butylphosphino)ethyl) ether
was prepared according to literature methods.2g Gaseous D2
(99.8%), lithium metal pellets, chlorodiisopropylphosphine, and
bis(2-chloroethyl) ether were purchased from Aldrich and used
without further purification.
Chart 4. View along the N-Ru and O-Ru Axis in
RuCl2(POP-tBu) (A) and RuCl2(PNP-tBu) (B)
Preparation of Lithium Diisopropylphosphide. A tet-
i
rahydrofuran solution (20 mL) of Pr2PCl (10.0 g, 65.6 mmol)
was added dropwise over a period of approximately 20 min to
a stirred suspension of lithium metal pellets (1.03 g, 149 mmol)
in tetrahydrofuran (50 mL). The mixture was stirred for
approximately 48 h. The solution was filtered to remove excess
lithium, and the product was used without further purification.
Preparation of Bis(2-(diisopropylphosphino)ethyl)
Ether (POP-iPr). A tetrahydrofuran (15 mL) solution of
(ClCH2CH2)2O (4.69 g, 32.8 mmol) was added to a cooled and
stirred tetrahydrofuran (70 mL) solution of LiPiPr2 (8.14 g,
65.6 mmol) dropwise over a period of 30 min. The temperature
of the mixture was kept between -20 and -30 °C. The solution
was stirred at ambient temperature for 1 h. The solvent was
removed and the residue dried in vacuo for 30 min. Hexane
(40 mL) was added to the residue, and the mixture was washed
with water (3 × 20 mL). The product was purified by fractional
distillation under reduced pressure (0.01 mmHg). Bis(2-
(diisopropylphosphino)ethyl) ether was obtained as a pyro-
phoric colorless oil. Yield: 5.26 g (52%). 1H NMR (C6D6): δ
0.99 (m, 24H, PC(CH3)2), 1.57 (dqq, 2JHP ) 2.0 Hz, 2JHH ) 7.0
Chart 5. Side Views of RuCl2(POP-tBu) (A) and
RuCl2(PNP-tBu) (B)
It is interesting that these complexes are stable and
yet the analogous POP complexes, RuCl2(N2)(POP-tBu)
or Ru2Cl4(µ-N2)(POP-tBu)2, are not. Conversely, it is
curious that the POP-tBu pincer compound 1 was
isolated and yet Milstein et al. did not observe an
analogous compound RuCl2(PNP-tBu). The differing
factor between these complexes is the pincer ligand
backbone. In 1, the ether linkage is well aligned along
the POP plane (structure A in Chart 4). In complexes
3-6 the fused five-membered rings of the coordinated
POP ligands are coplanar. On the contrary, the solid-
state molecular structure of Ru2Cl4(µ-N2)(PNP-tBu)2 (B
in Chart 4)18 shows the pyridine ring rotated out of the
PNP plane by about 26° and has one CH2 group above
and the other below the equatorial plane of the mol-
ecule. It appears that these differences can be due to
conformational preferences of the ligands. An inspection
of 124 crystal structures in the Cambridge Structural
Database of complexes with PNP and PCP pincer
ligands containing an aromatic ring in the backbone
found that two-thirds of them have the aromatic ring
rotated with respect to the ligand plane.
2
Hz, JHH ) 7.0 Hz, 4H, PCH), 1.72 (m, 4H, PCH2), 3.63 (m,
4H, OCH2). 31P{1H} NMR (C6D6): δ -1.3 (s). 13C{1H} NMR
(C6D6): δ 19.2 (d, 2JCP ) 9.7 Hz, PCCH3), 20.6 (d, 1JCP ) 17.0
1
2
Hz, PCCH3′), 23.5 (d, JCP ) 19.8 Hz, PCH2), 24.0 (d, JCP
)
2
13.6 Hz, PCH), 71.1 (d, JCP ) 31.7 Hz, OCH2).
Preparation of RuCl2{(tBu2PCH2CH2)2O} (1). A stirred
mixture of [RuCl2(p-cymene)]2 (626 mg, 1.02 mmol) and (tBu2-
PCH2CH2)2O (775 mg, 2.14 mmol) in 15 mL of toluene was
heated to 90 °C for 23 h. The mixture was cooled to ambient
temperature for a period of 1 h. The solid product was filtered,
washed with toluene (3 × 3 mL), and dried in vacuo. Yield:
785 mg (72%). Anal. Calcd for C20H44Cl2OP2Ru: C, 44.94; H,
1
8.29. Found: C, 44.52; H, 8.12. H NMR (CD2Cl2, 243 K): δ
0.50 (vt, vJ ) 11.4 Hz, 18H, PC(CH3)3), 1.40 (vt, vJ ) 12.9 Hz,
18H, PC(CH3)3), 2.19 (m, 2H, PCH2), 2.37 (m, 2H, PCH2), 3.92
(m, 2H, OCH), 4.31 (m, 2H, OCH). 31P{1H} NMR (CD2Cl2, 243
K): δ 17.7 (s). 13C{1H} NMR (CD2Cl2, 243 K): δ 20.6 (vt, vJ )
11.5 Hz, PCH2), 27.7 (s, PC(CH3)3), 29.0 (s, PC(CH3)3), 35.2
(vt, vJ ) 10.0 Hz, PC(CH3)3), 38.0 (vt, vJ ) 14.6 Hz, PC(CH3)3),
In the case of complexes 1 and 3-6, the PR2 groups
are in an eclipsed conformation as in A in Chart 5. In
1, this arrangement places two tBu substituents in the
vacant coordination site and makes it unfavorable for
coordination of even small ligands such as N2 or Cl. The
twisted backbone of the PNP-tBu system causes a
gauche conformation of the two PtBu2 groups, as sche-
v
76.9 (vt, J ) 13.9 Hz, OCH2).
RuCl2{(tBu2PCH2CH2)2O} is a green crystalline solid. It has
good solubility in dichloromethane but is much less soluble in
tetrahydrofuran, benzene, or toluene. In solid form the com-
pound is stable in air for 24 h or more, but the compound is
very air sensitive in solution.
t
matically shown in B in Chart 5. In this case, the Bu2
(19) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals, 3rd ed.; Pergamon: New York, 1988.
(20) Bennett, M. A.; Huang, T. N.; Matheson, T. W.; Smith, A. K.
Inorg. Synth. 1981, 21, 74.
groups are more evenly spread between the ligands that
eases the repulsion. It can be said that the POP-tBu
ligand has conformational properties that make it an

2500 Organometallics, Vol. 24, No. 10, 2005
Major et al.
Preparation of [Ru2(µ-Cl)3{(iPr2PCH2CH2)2O}2]Cl (2).
A stirred mixture of [RuCl2(p-cymene)]2 (655 mg, 1.07 mmol)
and (iPr2PCH2CH2)2O (659 mg, 2.15 mmol) in 12 mL of
methanol was heated to 65 °C for 48 h. The solvent was
removed in vacuo, and 12 mL of toluene was added to the
residue. On standing for several hours an orange powder
precipitated from solution. The solid was filtered, washed with
toluene (3 × 3 mL), and dried in vacuo. The supernatant was
placed in a -28 °C freezer overnight, and a second crop of
2JCP ) 5.8 Hz, PCCH3), 20.85 (d, 2JCP ) 3.2 Hz, PCCH3), 20.94
2
2
(d, JCP ) 1.5 Hz, PCCH3), 21.15 (s, PCCH3), 21.39 (d, JCP
)
2.6 Hz, PCCH3), 25.99 (d, 1JCP ) 24.1 Hz, PCH2), 27.95 (d, 1JCP
1
) 23.9 Hz, PCH), 29.81 (d, JCP ) 25.0 Hz, PCH2), 29.81 (d,
1
1JCP ) 21.0 Hz, PCH), 30.48 (d, JCP ) 19.0 Hz, PCH), 32.49
1
(d, JCP ) 23.1 Hz, PCH), 77.22 (s, OCH2), 79.00 (s, OCH2).
[Ru2(µ-Cl)3{(iPr2PCH2CH2)2O}2}]PF6 is a red crystalline
material that has good solubility in tetrahydrofuran and
acetone but only fair solubility in methanol. It is insoluble in
isooctane and benzene.
1
crystals was collected. Yield: 708 mg (70%). H NMR (CD3-
Preparation of Ru(H2)Cl2{(tBu2PCH2CH2)2O} (3). A
suspension of 1 (103 mg, 0.193 mmol) in 4 mL of benzene was
degassed and placed under an atmosphere of H2 (∼3 psi). The
suspension was heated to 50 °C for 2 h, then allowed to cool
to ambient temperature for 1 h. The solid product was collected
by filtration, washed with hexane (2 × 3 mL), and dried in
vacuo. Yield: 86 mg (83%). Anal. Calcd for C20H46Cl2OP2Ru:
OD): δ 1.35-1.70 (m, 52H, 48H from PCCH3 and 4H from
PCH2), 1.94 (m, 2H, PCH2), 2.13 (m, 6H, 4H from PCH and
2H from PCH2), 3.1-3.4 (m, 6H, 4H from PCH and 2H from
OCH2), 3.5-3.8 (m, 4H, OCH2), 3.95 (m, 2H, OCH2). 31P{1H}
NMR (CD3OD): δ 64.8 (d, 2JPP ) 29.6 Hz), 72.5 (d, 2JPP ) 29.6
2
Hz). 13C{1H} NMR (CD3OD): δ 19.16 (d, JCP ) 8.4 Hz,
PCCH3), 19.80 (d, 2JCP ) 5.5 Hz, PCCH3), 20.60 (d, 2JCP ) 2.4
Hz, PCCH3), 20.79 (d, 2JCP ) 5.2 Hz, PCCH3), 21.06 (s, PCCH3),
21.16 (s, PCCH3), 21.29 (s, PCCH3), 20.46 (s, PCCH3), 26.61
1
C, 44.78; H, 8.64. Found: C, 44.99; H, 8.39. H NMR (C6D6):
2
δ -10.1 (t, JHP ) 8.8 Hz, 2H, Ru(H2)), 1.01 (m, 2H, PCH2),
1
1
v
(d, JCP ) 23.9 Hz, PCH2), 28.37 (d, JCP ) 23.9 Hz, PCH),
29.99 (d, 1JCP ) 21.9 Hz, PCH), 30.08 (d, 1JCP ) 16.1 Hz, PCH2),
31.11 (d, 1JCP ) 19.0 Hz, PCH), 33.06 (d, 1JCP ) 23.3 Hz, PCH),
77.68 (s, OCH2), 79.29 (s, OCH2).
1.21 (vt, J ) 12.3 Hz, 18H, PC(CH3)3), 1.60 (m, 2H, PCH2),
v
1.68 (vt, J ) 12.9 Hz, 18H, PC(CH3)3), 2.59 (m, 2H, OCH2),
3.52 (m, 2H, OCH2). 31P{1H} NMR (C6D6, 298 K): δ 61.4 (s).
13C{1H} NMR (C6D6): δ 24.4 (vt, vJ ) 10.6 Hz, PCH2), 25.5 (s,
PCH2), 30.6 (vt, vJ ) 4.6 Hz, PC(CH3)3), 31.6 (vt, vJ ) 4.3 Hz,
PC(CH3)3), 35.0 (vt, vJ ) 15.5 Hz, PC(CH3)3), 39.7(vt, vJ ) 9.5
Hz, PC(CH3)3), 75,32 (s, OCH2).
[Ru2(µ-Cl)3{(iPr2PCH2CH2)2O}2}]Cl is an orange powder that
has good solubility in methanol, ethanol, and 2-propanol, fair
solubility in 3-methyl-1-butanol, but is insoluble in acetone,
tetrahydrofuran, 2-methyl-2-propanol, 2-methyl-2-butanol,
dichloromethane, and nonpolar solvents such as hexane,
benzene, and toluene. The compound is stable in air for several
days in solid form but air sensitive in solution.
Ru(η2-H2)Cl2{(tBu2PCH2CH2)2O} is an orange powder that
is sparingly soluble in tetrahydrofuran, benzene, or toluene.
It is insoluble in hexane and like solvents.
Preparation of Ru(HD)Cl2{(tBu2PCH2CH2)2O} (3-d). A
C6D6 (0.65 mL) solution of 3 (15 mg) was placed in a Wilmad
NMR tube with a J-Young valve. The solution was subjected
to three freeze-pump-thaw cycles and backfilled with gaseous
D2. The solution was vigorously shaken for 15 min, then heated
Preparation of [Ru2(µ-Cl)3{(iPr2PCH2CH2)2O}2}][BPh4]
(2‚BPh4). A 2.5 mL methanol solution of LiBPh4‚3CH3OCH2-
CH2OCH3 (125 mg, 0.210 mmol) was added dropwise to a
stirred methanol (5 mL) solution of [Ru2(µ-Cl)3{(iPr2PCH2-
CH2)2O}2}]Cl (201 mg, 0.210 mmol). An orange precipitate
immediately formed. The mixture was stirred for an additional
10 min. The precipitate was collected via filtration and washed
with methanol (3 × 3 mL). Yield: 223 mg (86%). Anal. Calcd
for C56H92BCl3O2P4Ru2: C, 54.22; H, 7.47. Found: C, 53.92;
H, 7.42. 1H NMR (thf-d8): δ 1.26-1.64 (m, 54H), 2.01 (m, 6H),
2.92-3.53 (m, 10H), 3.71-3.85 (m, 2H), 6.71 (m, 4H, C6H5),
6.85 (m, 8H, C6H5), 7.28 (m, 8H, C6H5). 31P{1H} NMR (thf-d8):
1
to 50 °C for a few minutes. The H NMR spectrum showed a
mixture of 3 and Ru(HD)Cl2{(tBu2PCH2CH2)2O} in a ratio of
2:3, respectively. Allowing the solution to stand overnight at
ambient temperature resulted in near complete transformation
to Ru(D2)Cl2{(tBu2PCH2CH2)2O}.
Preparation of [Ru(N2)Cl{(tBu2PCH2CH2)2O}][BPh4]‚
CH2Cl2 (4). Dichloromethane (3 mL) was added to a mixture
of RuCl2{(tBu2PCH2CH2)2O} (104 mg, 0.195 mmol) and NaB-
Ph4 (67 mg, 0.196). The mixture was vigorously stirred for 3
h. After crystallization of NaCl (-30 °C, overnight) the mixture
was filtered. The volume of the filtrate was reduced to
approximately 1 mL, and hexane (0.5 mL) was added. The
solution was placed in a -30 °C freezer overnight. A dark
crystalline material was collected by filtration and washed
with hexane (3 × 1.5 mL). Yield: 148 mg (90%). Anal. Calcd
for C45H66BCl3N2OP2Ru: C, 58.04; H, 7.14; N, 3.01. Found:
C, 58.31; H, 7.37; N, 2.99. IR: ν(NtN) 2143 cm-1 (KBr and
Nujol). 1H NMR (CD2Cl2, 297 K): δ 0.48, 1.48 (br s, PC(CH3)3),
1.90 (m, 2H, PCH2), 2.00 (m, 2H, PCH2), 3.71 (m, 2H, OCH),
3.91 (m, 2H, OCH), 6.96, 7.16, 7.58 (m, 20H, BPh4-). 31P{1H}
NMR (CD2Cl2, 297 K): δ 54.6 (s). 1H NMR (CD2Cl2, 273 K): δ
2
2
δ 64.7 (d, JPP ) 29.1 Hz), 73.5 (d, JPP ) 29.1 Hz). 13C{1H}
2
NMR (thf-d8): δ 19.14 (d, JCP ) 8.4 Hz, PCCH3), 19.81 (d,
2JCP ) 7.2 Hz, PCCH3), 20.52 (d, 2JCP ) 3.7 Hz, PCCH3), 20.62
2
2
(d, JCP ) 6.0 Hz, PCCH3), 20.98 (d, JCP ) 3.2 Hz, PCCH3),
2
21.03 (d, JCP ) 1.7 Hz, PCCH3), 21.24 (s, PCCH3), 21.47 (s,
PCCH3), 26.22 (d, 1JCP ) 23.6 Hz, PCH2), 28.17 (d, 1JCP ) 23.6
1
1
Hz, PCH), 29.97 (d, JCP ) 21.3 Hz, PCH), 30.00 (d, JCP
)
24.4 Hz, PCH2), 30.79 (d, 1JCP ) 18.4 Hz, PCH), 32.86 (d, 1JCP
) 23.0 Hz, PCH), 77.13 (s, OCH2), 78.96 (s, OCH2), 122.03 (s,
B(C6H5)4-), 125.87 (m, JCB ) 2.9 Hz, B(C6H5)4-), 137.35 (m,
JCB ) 1.4 Hz, B(C6H5)4-), 165.35 (q, JCB ) 49.5 Hz, B(C6H5)4-).
[Ru2(µ-Cl)3{(iPr2PCH2CH2)2O}2}BPh4 is an orange powder
that shows good and fair solubility in CH2Cl2 and tetrahydro-
furan, respectively. It is insoluble in methanol, isooctane, and
benzene.
v
v
0.44 (vt, J ) 6.5 Hz,18H, PC(CH3)3), 1.46 (vt, J ) 7.4 Hz,
18H, PC(CH3)3), 1.84 (m, 2H, PCH2), 2.01 (m, 2H, PCH2), 3.63
(m, 2H, OCH), 3.86 (m, 2H, OCH), 6.96, 7.16, 7.58 (m, 20H,
BPh4-). 31P{1H} NMR (CD2Cl2, 273 K): δ 53.8 (s). 1H NMR
(CD2Cl2, 173 K): δ -2.05 (br s, 6H, PC(CH3)), 1.13, 1.36, 1.54
(br s, 34H, PC(CH3)3, PCH2), 3.31 (br, 2H, OCH2), 3.59 (br,
2H, OCH2), 6.82, 7.01, 7.39 (m, 20H, BPh4-). 31P{1H} NMR
(CD2Cl2, 173 K): δ 51.0 (s).
Preparation of [Ru2(µ-Cl)3{(iPr2PCH2CH2)2O}2}][PF6]
(2‚PF6). A 6 mL methanol solution of LiPF6 (71 mg, 0.47 mmol)
was added dropwise to a stirred methanol (4 mL) solution of
[Ru2(µ-Cl)3{(iPr2PCH2CH2)2O}2}]Cl (102 mg, 0.107 mmol). The
solution was stirred for 30 min. The solution was place in a
-28 °C freezer for 1-2 days. Red crystals were collected by
decantation and washed with isooctane (2 × 3 mL). Yield:
73%. Anal. Calcd for C32H72Cl3F6O2P5Ru2: C, 36.05; H, 6.81.
Found: C, 35.81; H, 6.63. 1H NMR (acetone-d6): δ 1.32-1.86
(m, 51H), 1.92-2.29 (m, 9H), 3.12-4.02 (m, 12H). 31P{1H}
NMR (acetone-d6): δ -143.1 (septet, 1JPF ) 5.8 Hz, PF6-), 64.4
[RuCl(N2){(tBu2PCH2CH2)2O}][BPh4] is an air-sensitive dark
colored crystalline material that is soluble in CH2Cl2 and
tetrahydrofuran and insoluble in hexane.
Preparation of Ru(H2)Cl2{(iPr2PCH2CH2)2O} (5). A
2-methyl-2-butanol (10 mL) solution of [RuCl2(p-cymene)]2 (329
mg, 0.537 mmol) and (iPr2PCH2CH2)2O (342 mg, 1.12 mmol)
was degassed and placed under an atmosphere of H2 (∼3 psi).
The mixture was heated to 90 °C for 26 h. Deep red crystals
2
2
(d, JPP ) 29.6 Hz), 73.1 (d, JPP ) 29.6 Hz). 13C{1H} NMR
(acetone-d6): δ 19.09 (d, 2JCP ) 8.3 Hz, PCCH3), 19.76 (d, 2JCP
) 7.8 Hz, PCCH3), 20.42 (d, JCP ) 3.8 Hz, PCCH3), 20.58 (d,
2

POP Pincer Complexes of Ruthenium
Organometallics, Vol. 24, No. 10, 2005 2501
formed after cooling to ambient temperature overnight (∼16
h). The crystals were collected by filtration and washed with
isooctane (3 × 3 mL). Recrystallization of the supernatant
afforded a second crop of product. Yield: 308 mg (60%). Anal.
Calcd for C16H38Cl2OP2Ru: C, 40.00; H, 7.97. Found: C, 40.22;
structures were solved and refined using SHELXTL V6.123 for
full-matrix least-squares refinements that were based on F2.
The H atoms were placed in calculated positions and included
in the structure refinement in a riding motion approximation.
The hydride atoms in 6 were refined independently, and their
thermal parameters were tied to that of the Ru atom such that
Uiso(H) ) 1.5Ueq(Ru). Crystallographic data for complexes 1,
2‚PF6, 4, 5, and 6 are provided in CIF format with the
Supporting Information.
Computational Details. The calculations were done with
Gaussian 03 (Revision B05) and GaussView (version 3.09)
programs.24 All geometries were fully optimized without
symmetry or internal coordinate constraints using the
mPW1PW91 functional, which included modified Perdew-
Wang exchange and Perdew-Wang 91 correlation.25 The
nature of the stationary points 3′, 3′′, 5′, 5′′, 3′ts, and 5′′ts
was verified by frequency calculations, which were used to
calculate ZPE without scaling. The basis set employed in the
calculations included SDD + ECP for Ru, 6-31G(p) for the
coordinated H2, 6-31G for all CH3 groups, and 6-31G(d) for the
rest of the atoms.26
1
H, 7.84. H NMR (C6D6): δ -14.7 (s, 2H, Ru(H2)), 1.23 (d vt,
3JHH ) 6.7 Hz, vJ ) 13.4 Hz, 12H, PC(CH3)2), 1.42 (d vt, 3JHH
v
) 7.8 Hz, J ) 15.5 Hz, 12H, PC(CH3)2), 1.86 (m, 4H, PCH2),
3.00 (m, 4H, PCH), 3.51 (m, 4H, OCH2). 31P{1H} NMR (C6D6):
δ 64.7 (s). 13C{1H} NMR (C6D6): δ 19.2 (s, PC(CH3)2), 20.9 (vt,
v
vJ ) 1.8 Hz, PC(CH3)2), 21.97 (vt, J ) 11.5 Hz, PCH), 26.48
v
v
(vt, J ) 7.5 Hz, PCH2), 73.5 (vt, J ) 2.4 Hz, OCH2).
Ru(H2)Cl2{(iPr2PCH2CH2)2O} is a red crystalline material
that is soluble in benzene, toluene, and to a lesser extent
2-methyl-2-butanol. It has poor solubility in isooctane. It is
air stable for several days in solid form.
Preparation of Ru(HD)Cl2{(iPr2PCH2CH2)2O} (5-d). A
C6D6 (0.65 mL) solution of 5 (11 mg) was placed in a Wilmad
NMR tube with a J-Young valve. The solution was subjected
to three freeze-pump-thaw cycles and backfilled with gaseous
D2. The solution was heated to 50 °C for a few minutes, then
vigorously shaken for 15 min. The 1H NMR spectrum showed
a mixture of 5 and Ru(HD)Cl2{(iPr2PCH2CH2)2O} in a ratio of
3:7, respectively. Allowing the solution to stand overnight at
ambient temperature resulted in near complete transformation
to Ru(D2)Cl2{(iPr2PCH2CH2)2O}.
Acknowledgment. We gratefully acknowledge the
Natural Sciences and Engineering Research Council of
Canada (NSERC) and the Ontario Government for
funding.
Preparation of RuCl2(N2){(iPr2PCH2CH2)2O} (6). A
stirred 2-methyl-2-butanol (12 mL) solution of [Ru2(µ-Cl)3{(iPr2-
PCH2CH2)2O}2}]Cl (206 mg, 0.215 mmol) was heated to 75 °C
for 2 h under a N2 atmosphere. The solvent was removed in
vacuo, and the products crystallized from a toluene/isooctane
mixture at -28 °C. Yield: 180 mg (83%). Anal. Calcd for
C16H36Cl2N2OP2Ru: C, 37.95; H, 7.17. Found: C, 38.20; H,
7.12. IR: ν(NtN) 2119.0 cm-1 (Nujol), 2114.6 cm-1 (KBr). 1H
Supporting Information Available: Atomic coordinates
of the calculated complexes and crystallographic data for
complexes 1, 2‚PF6, 4, 5, and 6. This material is available free
of charge via the Internet at http://pubs.acs.org.
OM050053V
(23) Sheldrick, G. M. SHELXTL/PC, Version 6.1 Windows NT
Version; Bruker AXS Inc.: Madison, WI, 2001.
3
v
NMR (C6D6): δ 1.23 (d vt, JHH ) 7.2 Hz, J ) 13.3 Hz, 12H,
(24) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.,
Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Salvador, P.; Dannenberg, J. J.; Malick, D. K.; Rabuck,
A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.;
Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.;
Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-
Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill,
P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez,
C.; Head-Gordon, M. E.; Replogle, S.; Pople, J. A. Gaussian 03 (Revision
B05) and GaussView (version 3.09); Gaussian, Inc.: Pittsburgh, PA,
2001.
3
v
PC(CH3)2), 1.43 (d vt, JHH ) 7.5 Hz, J ) 15.3 Hz, 12H, PC-
(CH3)2), 1.62 (m, 4H, PCH2), 3.08 (m, 4H, PCH), 3.26 (m, 4H,
OCH2). 31P{1H} NMR (C6D6): δ 48.3 (s). 13C{1H} NMR (C6D6):
δ 19.04 (s, PC(CH3)2), 20.45 (vt, vJ ) 2.9 Hz, PC(CH3)2), 22.91
(vt, vJ ) 21.6 Hz, PCH), 25.75 (vt, vJ ) 16.1 Hz, PCH2), 74.00
v
(vt, J ) 4.3 Hz, OCH2).
RuCl2(N2){(iPr2PCH2CH2)2O} is a yellow crystalline material
that shows good soluble in benzene, toluene, and CH2Cl2 but
only fair solubility in 2-methyl-2-butanol. The compound shows
poor solubility in hexane and isooctane. The compound is air
stable in solid form for several days.
X-ray Crystallographic Analyses. For all compounds,
data were collected on a Nonius Kappa-CCD diffractometer
using monochromated Mo KR (wavelength ) 0.71073 Å)
radiation. Each data set was measured using a combination
of φ scans and ω scans with κ offsets, to fill the Ewald sphere.
The data were processed using the Denzo-SMN package.21
Absorption corrections were carried out using SORTAV.22 The
(25) (a) Adamo, C.; Barone V. J. Chem. Phys. 1998, 108, 664. (b)
Perdew, J. P.; Burke, K.; Wang, Y., Phys. Rev. B 1996, 54, 16533. (c)
Burke K.; Perdew, J. P.; Wang, Y. In Electronic Density Functional
Theory: Recent Progress and New Directions; Dobson, J. F., Vignale,
G., Das, M. P., Eds.; Plenum: New York, 1998.
(26) For more information about the basis sets implemented in
Gaussian 03 and detailed references see: Frish, A.; Frish, M. J.;
Trucks, G. W. Gaussian 03 User’s Reference; Gaussian, Inc.: Pitts-
burgh, PA, 2003. The basis sets are also available from the Extensible
Computational Chemistry Environment Basis Set Database
(www.emsl.pnl.gov/forms/basisform.html), which is developed and
distributed by the Molecular Science Computing Facility, Environ-
mental and Molecular Sciences Laboratory, which is part of the Pacific
Northwest Laboratory, P.O. Box 999, Richland, WA 99352.
(21) Otwinowski, Z.; Minor, W. In Methods in Enzymology: Mac-
romolecular Crystallography; Carter, C. W., Sweet, R. M., Eds.;
Academic Press: London, 1997; Vol. 276, part A, pp 307-326.
(22) Blessing, R. H. Acta Crystallogr. 1995, A51, 33.