Unexpected formation of ruthenium(II) hydrides from a reactive dianiline precursor and 1,2-(Ph2P)2-1,2-closo-C2B 10H10

Article abstract of DOI:10.1021/ic701871t

Reaction of the new precursor cis,trans-Ru(cod)(anln)₂Cl ₂ with the diphosphine 1,2-bis(diphenylphosphino)-1,2-dicarba-closo- dodecaborane (o-dppc) unexpectedly results in two new ruthenium(II) hydrides, trans-Ru(o-dppc)₂(H)Cl and the neutral, five-coordinate complex Ru(o-dppc)(nido-dppc)(H), depending upon the reaction conditions [anln is aniline and nido-dppc is 7,8-(Ph₂P)₂C₂BgH ₁₀^-]. Chloride abstraction from trans-Ru(o-dppc) ₂(H)Cl leads to another five-coordinate hydride, [Ru(o-dppc) ₂(H)]⁺, which is isolated as either a triflate or hexafluorophosphate salt. On the basis of labeling and reactivity studies, the source of the hydride appears to be the cod ligand.

Full text of DOI:10.1021/ic701871t

Inorg. Chem. 2008, 47, 1871-1873
Unexpected Formation of Ruthenium(II) Hydrides from a Reactive
Dianiline Precursor and 1,2-(Ph2P)2-1,2-closo-C2B10H10
Jeramie J. Adams, Andrew S. Del Negro, Navamoney Arulsamy, and B. Patrick Sullivan*
Department of Chemistry, UniVersity of Wyoming, Laramie, Wyoming 82071-3838
Received September 26, 2007
Reaction of the new precursor cis,trans-Ru(cod)(anln)2Cl2 with the
diphosphine 1,2-bis(diphenylphosphino)-1,2-dicarba-closo-dodeca-
borane (o-dppc) unexpectedly results in two new ruthenium(II)
hydrides, trans-Ru(o-dppc)2(H)Cl and the neutral, five-coordinate
complex Ru(o-dppc)(nido-dppc)(H), depending upon the reaction
conditions [anln is aniline and nido-dppc is 7,8-(Ph2P)2C2B9H10-].
Chloride abstraction from trans-Ru(o-dppc)2(H)Cl leads to another
five-coordinate hydride, [Ru(o-dppc)2(H)]+, which is isolated as
either a triflate or hexafluorophosphate salt. On the basis of labeling
and reactivity studies, the source of the hydride appears to be the
cod ligand.
2 is also a precursor for the formation of new 16-electron,
five-coordinate hydrides by both adventitious and designed
means. Collectively, these reactions constitute new routes
to ruthenium(II) hydride complexes.3 In addition, the com-
plexes described here are the first examples containing two
o-dppc ligands at one metal center.
The preparation of complex 1 is modeled on that of the
ditoluidine complex, which, in our hands, is highly insoluble
in common organic solvents, limiting its viability as a
reactive precursor.4 Complex 1 is easily and inexpensively
prepared in high yields (ca. 95%) from the cod polymer di-
µ-chloro-[RuCl2(η4-C8H12)]x (x > 2).5 It should be noted that
a similar starting material, cis,trans-Ru(cod)(CH3CN)2Cl2,
is prepared6,7 in an analogous fashion, which when reacted
with o-dppc gives different products (see later). Crystal
structure data for complex 1 are given along with selected
bond lengths and angles in Figure 1. The structure shows a
distorted octahedral environment with a drastically bent
Cl-Ru-Cl angle at 156.54(4)°, a common distortion found
for ligands axial to the equatorial plane containing cod in
ruthenium(II) complexes.7,8
During the course of our search for new precursors for
the preparation of oligometallic molecular assemblies, we
have discovered an extremely reactive ruthenium(II) dianiline
complex, cis,trans-Ru(cod)(anln)2Cl2 (1; anln is aniline),
which by virtue of its stereochemistry and the nature of the
ligands can exhibit selective, stepwise substitutions at pairs
of cis-coordination sites. In complex 1, the labile anln ligands
are capable of being substituted under mild conditions
because of the hard—soft nature of their interaction, the
chloride ligands can undergo ligand metathesis via salt
formation, and the strongly bound cod ligand can be replaced
by selective photochemical substitution. Potential applications
of this and related complexes range from the preparation of
isomerically pure octahedral molecular squares, and linear
oligomeric complexes,1 to new bis-chelate complexes of the
type trans-Ru(chelate)2Cl2 or trans-Ru(chelate)(chelate′)Cl2.
Apparently, ruthenium(II) aniline complexes are scarce
despite their obvious appeal as precursors containing good
leaving groups.2
(3) For a summary of ruthenium(II) hydride preparative routes, see:
Clapham, S. E.; Hadzovic, A.; Morris, R. H. Coord. Chem. ReV. 2004,
248, 2201.
(4) Abel, E. W.; Bennett, M. A.; Wilkinson, G. J. Chem. Soc. 1959, 3178–
3182.
(5) cis,trans-Ru. (cod)(anln)2Cl2 (1). The polymer di-µ-chloro-[RuCl2(η4-
C8H12)]x (x > 2; 5.00 g, ∼17.8 mmol) and anln (12 mL, 126.1 mmol)
were placed in a 100-mL round-bottomed flask equipped with a
magnet. The reaction was placed in an oil bath at 90 °C with stirring
for 1 h. The reaction was cooled to room temperature, and the product
was precipitated with diethyl ether. The product was then redissolved
in chloroform (150 mL) and filtered by suction filtration to remove a
small amount of insoluble material. The volume of the filtrate was
reduced by rotary evaporation to 15 mL, a microcrystalline cream-
colored precipitate formed during the volume reduction, and the
product was further precipitated by adding diethyl ether. The cream-
colored product was collected by suction filtration, washed several
times with diethyl ether, and dried under vacuum (6.99 g, 94% yield).
Suitable crystals for single-crystal X-ray diffraction were grown by
slow evaporation of the aniline diethyl ether filtrate, which was placed
in a freezer at –10 °C. Anal. Calcd: C, 51.62; H, 5.62; N, 6.01. Found:
C, 51.50; H, 5.61; N, 5.99.1H NMR (chloroform-d): δ 7.332 (d, 4H),
7.259 (t, 4H), 7.128 (t, 2H), 4.560 (s, 4H), 3.705 (s, 4H), 2.563 (m,
4H), 1.955 (d, 4H).
In this Communication, we present details of the reaction
of complex 1 with the chelate 1,2-bis(diphenylphosphino)-
1,2-dicarba-closo-dodecaborane
[1,2-(Ph2P)2-1,2-closo-
C2B10H10; o-dppc], which surprisingly results in the formation
of the hydride complex trans-Ru(o-dppc)2(H)Cl (2). Complex
* To whom correspondence should be addressed. E-mail:
bpat@uwyo.edu.
(1) Woessner, S. M.; Helms, J. B.; Lantzky, K. M.; Sullivan, B. P. Inorg.
Chem. 1999, 38, 4378–4379.
(6) Albers, M. O.; Ashworth, T. V.; Oosthuizen, H. E.; Singleton, E. Inorg.
Synth. 1989, 26, 68–77.
(2) Clark, T.; Cochrane, J.; Colson, S. F.; Malik, K. S.; Robinson, S. D.;
Steed, J. W. Polyhedron 2001, 20, 1875–1880.
(7) Widegren, J. A.; Weiner, H.; Miller, S. M.; Finke, R. G. J. Organomet.
Chem. 2000, 610, 112–117.
10.1021/ic701871t CCC: $40.75
Published on Web 01/25/2008
2008 American Chemical Society
Inorganic Chemistry, Vol. 47, No. 6, 2008 1871

COMMUNICATION
refluxed in the higher-boiling solvent o-dichlorobenzene for
3 h, complex 1 is converted into the neutral five-coordinate
complex Ru(o-dppc)(nido-dppc)(H) (3), where nido-dppc is
a monoanionic, partially degraded form of o-dppc that results
from the removal of a single BH vertex and protonation of
-
the open face, resulting in a nido-7,8-(Ph2P)2C2B9H10
ligand.10
31P NMR studies convincingly show that slow conversion
of complex 2 to 3 occurs under the more forcing conditions.
The formation of nido-7,8-(Ph2P)2C2B9H10- from o-dppc in
the presence of a Lewis base has been demonstrated by
Teixidor et al. and is enhanced by metal coordination.11,12
It is likely that the free anln produced upon substitution is a
contributing factor to cage degradation. Complex 3 showed
a loss of phosphorus equivalency, indicated by the two
31P{1H} NMR signals at 98.1 and 60.2 ppm that were split
with a trans-phosphine coupling constant of 208 Hz. The
Figure 1. ORTEP representation of 1. Hydrogen atoms and an anln of
crystallization are omitted for clarity. Thermal ellipsoids are drawn at the
30% probability level. Selected bond lengths (Å) are as follows: Ru(1)-C(1),
2.189(3); Ru(1)-C(2), 2.193(4); Ru(1)-C(6), 2.193(4); Ru(1)-C(5),
2.198(4); Ru(1)-N(1), 2.201(3); Ru(1)-N(2), 2.209(3); Ru(1)-Cl(2),
2.4250(11); Ru(1)-Cl(1), 2.4343(12). Selected bond angles (deg) are as
follows: N(1)-Ru(1)-N(2), 84.98(11); Cl(2)-Ru(1)-Cl(1), 156.54(4).
1
hydride was spectroscopically characterized by H NMR
spectroscopy as a seven-line multiplet at –30.2 ppm arising
from the fortuitous overlap of a triplet of triplets due to one
of the dppc-type ligands having a JPH value that is double
the other and exhibiting coupling constants of 24 and 12
(9) trans-Ru(o-dppc)2(H)Cl (2). Complex 1 (0.200 g, 0.429 mmol) and
o-dppc (0.441 g, 0.861 mmol) were added to a 50-mL round-bottomed
flask. This flask was evacuated, and about 7 mL of dry o-xylene was
added via vacuum transfer. The reaction was refluxed with magnetic
stirring for 3 h under a N2 atmosphere. The brown suspension turned
to a yellow precipitate. The suspension was then cooled to room
temperature, and the yellow product was collected via suction filtration
(0.228 g, 46% yield). Suitable crystals for single-crystal X-ray
diffraction were grown by slow evaporation of a saturated solution in
chloroform. Anal. Calcd for {Ru(o-dppc)2(H)Cl ·C8H10}: C, 56.58; H,
5.62. Found: C, 56.62; H, 5.56.1H NMR (chloroform-d): δ 7.705 (s,
8H, broad), 7.631 (d, 8H, broad), 7.259 (t, 4H), 7.197 (t, 4H), 6.960
(m, 16H), –17.372 (qnt, 1H, hydride, JPH ) 21 Hz).31P{1H} NMR
(chloroform-d): δ 89.6 (s).11B NMR (acetone-d6): δ –1.01 (s, 6B,
broad), –7.78 (s, 4B, broad). The complex was also made by adding
0.150 g (0.126 mmol) of Ru(PPh3)3(H)Cl with 0.167 g (0.325 mmol)
of dppc to a 25-mL flask containing 7 mL of o-xylene. The reaction
was refluxed with stirring for 3 h, which precipitated the yellow
product. The product was collected via suction filtration and rinsed
with diethyl ether (0.130 g, 88% yield).1H and31P{1H} NMR spectra
were in agreement with the preceding values. Note: this complex
incorporates solvents depending on conditions and subsequent recrys-
tallizations, which are observed in1H NMR spectra.
Figure 2. ORTEP representation of 2 with thermal ellipsoids drawn at the
50% probability level. Hydrogen atoms except those of the hydride ligand
are omitted for clarity. Selected bond lengths (Å) are as follows: Ru(1)-P(2),
2.3410(4); Ru(1)-P(1), 2.3444(4); Ru(1)-P(3), 2.3465(4); Ru(1)-P(4),
2.3560(4); Ru(1)-Cl(1), 2.5142(4); Ru(1)-H(1), 1.51(2). Selected bond
angles (deg) are as follows: P(2)-Ru(1)-P(1), 86.842(15); P(3)-Ru(1)-P(4),
88.515(15); Cl(1)-Ru(1)-H(1), 177.3(8).
(10) Ru(o-dppc)(nido-dppc)(H) (3). Complex 1 (0.200 g, 0.429 mmol)
and o-dppc (0.441 g, 0.861 mmol) were added to a 50-mL round-
bottomed flask containing about 6 mL of o-dichlorobenzene. The
reaction was refluxed with magnetic stirring for 3 h. The brown
suspension turned to a dark-colored solution. The solution was then
cooled to room temperature. Upon cooling, the product precipitated
as a yellow microcrystalline solid. The crude product was collected
via suction filtration and was subsequently purified by dissolving it in
a large volume of methylene chloride and then reducing the volume
via rotary evaporation, allowing the flask to cool upon solvent
evacuation. Upon reduction of the volume, the pure product precipi-
tated out as a yellow microcrystalline solid, which was collected via
suction filtration (0.245 g, 51% yield). Anal. Calcd: C, 55.99; H, 5.42.
Found: C, 56.24; H, 5.60.1H NMR (chloroform-d): δ 8.083 (t, 5H),
7.837 (t, 3H) 7.745 (t, 5H), 7.079 (m, 13H), 6.803 (t, 6H), 6.519 (m,
Initial studies of the reactivity of complex 1 with chelating
diphosphines (PP), for example, 1,2-bis(diphenylphospino)-
methane and cis-1,2-bis(diphenylphospino)ethylene, showed
that within minutes in dichloromethane at room temperature
trans-Ru(PP)2Cl2 complexes are formed quantitatively. The
attempt to expand this reactivity to the o-dppc ligand resulted
in no reaction under the same conditions; however, when
heated to reflux in chlorobenzene or xylene, a yellow
microcrystalline precipitate is formed. This complex was not
the expected trans-dichloride but rather the hydride complex
2, which subsequently was structurally characterized (Figure
2; along with selected bond lengths and angles).
1
8H), –2.349 (broad, 1H, BHB), –30.211 (tt, 1H, hydride, JPH ) 24
Hz, JPH2 ) 12 Hz).31P{1H} NMR (chloroform-d): δ 98.1 (d, JPP-trans
) 207 Hz), 60.2 (d, JPP-trans ) 209 Hz).11B NMR (chloroform-d): δ
–1.99 (s, 2B, broad), –7.71 (s, 3B, broad), –12.42 (s, 4B, broad), –15.49
(m, 8B, broad), –27.32 (s, 1B, broad), –34.90 (s, 1B, broad). Complex
3 is stable in air as a solid and in solution for about 2 days; after that,
the complex undergoes decomposition to form an uncharacterized
white species.
Complex 2 can be used further to make two 16-electron,
five-coordinate complexes. When the reaction is carried out
to make complex 2 under more forcing conditions such as
longer reflux times (12 h, xylenes) or the reaction is simply
(11) Teixidor, F.; Vinas, C.; Mar Abad, M.; Nunez, R.; Kivekas, R.;
Sillanpaa, R. J. Organomet. Chem. 1995, 503, 193.
(8) Ashworth, T. V.; Liles, D. C.; Robinson, D. J.; Singleton, E. S. Afr.
J. Chem. 1987, 40, 183–188.
(12) Teixidor, F.; Vinas, C.; Mar Abad, M.; Kivekas, R.; Sillanpaa, R. J.
Organomet. Chem. 1996, 509, 139.
1872 Inorganic Chemistry, Vol. 47, No. 6, 2008

COMMUNICATION
Hz. The BHB bridging proton was located as a broad signal
at –2.3 ppm.12 There was no evidence in the H NMR
1
spectrum for an agositic B-H-Ru interaction in solution
involving the nido-dppc ligand; however, this does not
preclude its existence in the solid state.13 The 11B NMR
signals also reflected the loss in symmetry of the carborane
cages. The typical o-dppc 4:6 splitting pattern is overlaid
with the typical nido-dppc 2:3:2:1:1 splitting in the range
from –2 to –35 ppm. This neutral, coordinatively unsaturated
complex is the first to contain both an o-dppc ligand and a
nido-dppc ligand and may have application as an active
catalyst for Kharasch additions.14,15
A second five-coordinate hydride was prepared in a fashion
similar to approaches taken by others,16–18 using chloride
abstraction with either AgOTf or TlPF6 to generate the
monocationic five-coordinate complex trans-[Ru(o-
dppc)2(H)]+ in high yield (>90%).19 Structural data for[Ru(o-
dppc)2(H)][PF6] (4) are displayed in Figure 3 along with
selected bond lengths and angles. Unlike the case of complex
2, the hydride ligand could not be located. Here, the halves
of crystallographically unique cations are present in the
asymmetric unit, with the metal centers being located on
2-fold symmetry axes, although they share nearly identical
structural features. The four ligating phosphorus atoms and
the central ruthenium atom are coplanar, indicating possible
disorder of the hydride ligand above and below the RuP4
plane. Perhaps as a consequence of this, the hydride could
not be located in the Fourier maps. The five-coordinate
cationic species is air-stable for several days in CH2Cl2 and
is apparently more robust than similar complexes with other
Figure 3. ORTEP representation of the cation in trans-4 with thermal
ellipsoids drawn at the 50% probability level. Hydrogen atoms are omitted
for clarity. Selected bond lengths (Å) are as follows: Ru1-P1, 2.3515(7);
Ru1-P1A, 2.3515(7); Ru1-P1C, 2.3423(7); Ru1-P1B, 2.3423(7). Selected
bond angles (deg) are as follows: P1C-Ru1-P1B, 180; P1C-Ru1-P1B,
92.28; P1B-Ru1-P1A, 87.72; P1C-Ru1-P1, 87.72; P1C-Ru1-P1A,
92.26; P1-Ru1-P1A, 180.
diphosphine ligands.20 Sixteen-electron metal hydride com-
plexes have traditionally been used for hydrogenation of
alkynes and asymmetric transfer hydrogenation.20,21 Other
five-coordinate ruthenium(II) hydrides are useful for the
coordination of small molecules (most notably dihydrogen)
and also have been found to initiate C-F bond activation of
perfluoroalkanes.22
A series of studies implicate the cod ligand as the source
of the hydride. First, a labeling study using cis,trans-
Ru(cod)(anln-d7)2Cl2 showed the formation of the hydride 2
but no incorporation of deuterium. Second, the reaction of
o-dppc with a non-cod-containing ruthenium(II) source,
Ru(PPh3)3Cl2, resulted in an insoluble orange precipitate but
no hydride complexes. Third, intentional addition of a small
amount of water or D2O to the reaction in xylene did not
change the course of the reaction or its yield. Fourth, the
reaction of dppc with cis,trans-Ru(cod)(CH3CN)2Cl2 gave a
10% yield of complex 2, with the remaining ruthenium
appearing as a pale-orange insoluble complex. Last, an
analysis of the filtrate from complex 1 synthesis using 31P
(13) Nunez, R.; Vinas, C.; Teixidor, F.; Mar Abad, M. Appl. Organomet.
Chem. 2003, 17, 509–517.
(14) Simal, F.; Sebille, S.; Demonceau, A.; Noels, A. F.; Nunez, R.; Abad,
M.; Teixidor, F.; Vinas, C. Tetrahedron Lett. 2000, 41, 5347–5351.
(15) Demonceau, A.; Simal, F.; Noels, A. F.; Vinas, C.; Nunez, R.; Teixidor,
F.; Vinas, C. Tetrahedron Lett. 1997, 38, 7879–7882.
(16) Winter, R.; Hornung, F. M. Inorg. Chem. 1997, 36, 6197–6204.
(17) Saburi, M.; Fujii, T.; Shibusawa, T.; Masui, D.; Ishii, Y. Chem. Lett.
1999, 1343–1344.
(18) Martelletti, A.; Gramlich, V.; Zurcher, F.; Mezzetti, A. New J. Chem.
1999, 199–206.
(19) trans-[Ru(o-dppc)2(H)][OTf]. Complex 2 (0.200 g, 0.172 mmol) and
AgOTf (0.049 g, 0.191 mmol) were added to a 50-mL round-bottomed
flask containing about 25 mL of 1,2-dichloroethane. The reaction was
covered with aluminum foil and refluxed with magnetic stirring for
3 h. The yellow suspension turned to a red solution. The solution was
then cooled to room temperature, the AgCl was filtered from the
supernatant, and the solvent was removed by rotary evaporation. A
minimum amount of methylene chloride was added to dissolve the
product, and diethyl ether was added to precipitate an orange solid,
which was collected via suction filtration and dried under vacuum
(0.206 g, 94% yield). Suitable crystals for single-crystal X-ray
diffraction were grown by slow evaporation of a saturated solution in
methylene chloride. Anal. Calcd: C, 49.88; H, 4.82. Found: C, 50.00;
H, 4.90.1H NMR (acetone-d6): δ 7.896 (s, 8H, broad), 7.543 (m, 8H),
7.405 (m, 16H), 7.049 (s, 8H, broad), –27.031 (qnt, 1H, hydride, JPH
) 19 Hz).31P{1H} NMR (acetone-d6): δ 88.0 (s).11B NMR (acetone-
d6): δ –1.01 (s, 6B, broad), –7.78 (s, 4B, broad). ESI-MS. Calcd: m/z
1127.5. Found: m/z 1127.7. trans-[Ru(o-dppc)2(H)][PF6] (4). This
complex was prepared and purified in the same manner as that of
trans-[Ru. (dppc)2(H)][OTf] using 0.066 g (0.189 mmol) of TlPF6
instead of AgOTf (0.198 g, 91% yield). Anal. Calcd: C, 49.09; H,
4.83. Found: C, 48.83; H, 4.82.1H NMR (acetone-d6): δ 7.892 (s, 8H,
broad), 7.545 (m, 8H), 7.401 (m, 16H), 7.033 (s, 8H, broad), –26.969
(qnt, 1H, hydride, JPH ) 19 Hz).31P{1H} NMR (acetone-d6): δ 88.0
(s).
1
and H NMR spectroscopy revealed only unreacted and
coordinated o-dppc ligands, thereby arguing against ligand
degradation as a substantial source of the hydride. Given
the unusual nature of this chemistry, we are exploring the
reaction of complex 1 with other electron-poor phosphines.
Acknowledgment. B.P.S. thanks the School of Energy
Resources at the University of Wyoming and the National
Science Foundation (Grant NSFFLOC4835) for support.
IC701871T
(20) Saburi, M.; Ohnuki, M.; Ogasawara, M.; Takahashi, T.; Uchida, Y.
Tetrahedron Lett. 1992, 33, 5783–5786.
(21) Albers, M. O.; Singleton, E.; Viney, M. M. J. Mol. Catal. 1985, 33,
77–82.
(22) Kirkham, M. S.; Mhaon, M. F.; Whittlesy, K. K. Chem. Commun.
2001, 813–814.
Inorganic Chemistry, Vol. 47, No. 6, 2008 1873