Comparative properties of coordinated H2 and H2S at a ruthenium(ii) centre

Article abstract of DOI:10.1039/c3dt50314c

Thermodynamic data for the reversible formation of cis-RuCl ₂(P-N)(PPh₃)(η²-H₂) (2a) from trans-RuCl₂(P-N)(PPh₃) in C₆D₆ are determined by variable temperature ³¹P{¹H} and ¹H NMR spectroscopy; P-N = o-diphenylphosphino-N,N′- dimethylaniline. Values of ΔH°= -26 ± 4 kJ mol^-1, ΔS°= -40 ± 15 J mol^-1 K^-1, and ΔG°(at 25°C) = -13.8 ± 0.2 kJ mol^-1 are compared with recently reported data for the corresponding H₂S adduct (4a), where the exothermicity is greater by ～20 kJ mol^-1, but this is counteracted by a more unfavourable entropy change, and overall the K and ΔG°values at 25°C are close. For loss of H₂ from 2a in the solid state, whose X-ray structure is presented, ΔH°is 50 ± 3 kJ mol^-1 as measured by Differential Scanning Calorimetry. The pK_a values of the coordinated H₂ (～11) and H ₂S (～14) are estimated by reactions of 2a and 4a with proton sponge (1,8-bis(dimethylamino)naphthalene) in CD₂Cl₂ at 20°C; the mono-hydrido and -mercapto products are identified in situ. A corresponding H₂O adduct is not deprotonated under the same conditions. Related dihydrido, mercapto and hydroxy species are formed by in situ reactions of 1a with NaH, NaSH, and NaOH, respectively. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c3dt50314c

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Comparative properties of coordinated H2 and H2S at a
ruthenium(II) centre†
Cite this: Dalton Trans., 2013, 42, 7614
Erin S. F. Ma, Dona C. Mudalige, Brian O. Patrick and Brian R. James*
Thermodynamic data for the reversible formation of cis-RuCl2(P–N)(PPh3)(η2-H2) (2a) from trans-RuCl2-
(P–N)(PPh3) in C6D6 are determined by variable temperature 31P{1H} and 1H NMR spectroscopy; P–N =
o-diphenylphosphino-N,N’-dimethylaniline. Values of ΔH° = −26 4 kJ mol−1, ΔS° = −40 15 J mol−1
K−1, and ΔG° (at 25 °C) = −13.8 0.2 kJ mol−1 are compared with recently reported data for the corres-
ponding H2S adduct (4a), where the exothermicity is greater by ∼20 kJ mol−1, but this is counteracted by
a more unfavourable entropy change, and overall the K and ΔG° values at 25 °C are close. For loss of H2
from 2a in the solid state, whose X-ray structure is presented, ΔH° is 50 3 kJ mol−1 as measured by
Differential Scanning Calorimetry. The pKa values of the coordinated H2 (∼11) and H2S (∼14) are esti-
mated by reactions of 2a and 4a with proton sponge (1,8-bis(dimethylamino)naphthalene) in CD2Cl2 at
20 °C; the mono-hydrido and -mercapto products are identified in situ. A corresponding H2O adduct is
not deprotonated under the same conditions. Related dihydrido, mercapto and hydroxy species are
formed by in situ reactions of 1a with NaH, NaSH, and NaOH, respectively.
Received 30th January 2013,
Accepted 8th March 2013
DOI: 10.1039/c3dt50314c
www.rsc.org/dalton
properties of H2 and H2S coordinated at the same transition
metal site.
Introduction
We reported recently on the reversible binding (in solution
and the solid state) of H2S, MeSH, and EtSH,1 and the oxygen-
containing analogues H2O, MeOH and EtOH,2 to five-coordi-
nate complexes of the type trans-RuX2(P–N)(PR3) complex,
where X is a halogen, R is aryl, and P–N = o-diphenylphos-
phino-N,N′-dimethylaniline. The products of the S-ligand
systems are cis-isomers, whereas the O-donor ligands generate
trans-products (Chart 1). The findings, together with earlier
work on the corresponding binding of H2, N2, and N2O (which
also form cis-adducts), reveal the quite remarkable reactivity of
such trans-RuX2(P–N)(PR3) species, and allow for comparisons
of bond strengths within the coordinating moiety as well as
trans-effects and influences of the some of the donor ligands.2
We now report a crystal structure of cis-RuCl2(P–N)(PPh3)-
(η2-H2), some thermodynamic data on the H2-binding, and
acidity of the coordinated H2, and make comparisons with cor-
responding data for the analogous H2S system. The studies
lead to syntheses of hydrido and mercapto species, and allow
for the first reported, direct comparison between the
Experimental section
Unless stated otherwise, all manipulations were performed
under an oxygen-free, Ar atmosphere at room temperature (r.t.,
∼22 °C) using standard Schlenk techniques. Commercially
available compounds were supplied by Aldrich or Fisher, and
were used as received unless stated otherwise. Proton sponge
(PS
= 1,8-bis(dimethylamino)naphthalene), obtained from
Aldrich, was purified by passing an n-pentane solution of the
amine through an alumina column, and evaporating the
eluant to yield a white solid. Anal. Calcd C14H18N2: C, 78.46;
H, 8.47; N, 13.07. Found: C, 78.5; H, 8.4; N, 12.8. 1H NMR
(CD2Cl2): δ 7.4–6.9 (6H, m, Ph), 2.80 (6H, s, N(CH3)2), the data
agreeing with literature values.3 RuCl2(PPh3)3,4 trans-RuCl2-
(P–N)(PPh3) (1a) and the p-tolyl analogue (1b),1,5 RuBr2(P–N)-
(PPh3),6 and RuCl2(P–N)(PPh3)(SH2) (4a)6 were prepared by
the reported methods; the precursor RuCl3·xH2O was donated
by Colonial Metals, Inc. Complexes containing PPh3 are
labelled a, and the corresponding P(p-tolyl)3 compounds are
labelled b.
Department of Chemistry, University of British Columbia, Vancouver,
British Columbia, Canada, V6T 1Z. E-mail: brj@chem.ubc.ca
Spectral and analytical grade solvents, and deuterated sol-
vents, were purified and stored as described recently.2 Purified
Ar (H.P.) and H2 (Research, extra dry) were Union Carbide
Canada products, and anhydrous H2S was obtained from
Matheson Gas Co; the Ar was dried by passage through CaSO4
†Electronic supplementary information (ESI) available: 1H NMR data for deter-
mination of K, and associated van’t Hoff plot; DSC curve for 2a. CCDC 921295.
For ESI and crystallographic data in CIF or other electronic format see DOI:
10.1039/c3dt50314c
7614 | Dalton Trans., 2013, 42, 7614–7621
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Chart 1 Reactions of trans-RuX2(P–N)(PPh3) with various L ligands.
columns, and H2 was passed through an Engelhard Deoxo cata-
lytic hydrogen purifier to remove trace O2.
3a: 31P{1H} NMR (CD2Cl2): δ 82.74 (d, P–N), 67.39 (d, PPh3);
1
2JPP = 33.2. H NMR: δ 8.1–6.5 (29H, m, Ph), 3.49 (3H, s, NMe),
Details on the measurements of NMR and IR spectra, and 2.90 (3H, s, NMe), −27.2 (1H, br s, Ru–H). At −80 °C, the −27.2
procedure used for differential scanning calorimetry (DSC) peak was resolved into a pseudo triplet (2JHP = 28 Hz). IR:
were also provided recently:2 J values are reported in hertz νRu–H 2079 w.
(Hz), and s = singlet, d = doublet, m = multiplet, br = broad.
3b: 31P{1H} NMR (CD2Cl2): δ 83.44 (d, P–N), 64.58 (d, PPh3);
1
The T1 relaxation times for all the phosphorus nuclei were in 2JPP = 34.0. H NMR: δ 8.2–6.6 (29H, m, Ph), 3.50 (3H, s, NMe),
the range 1.7–2.3 s, and a pulse sequence of 13 s was used in 2.90 (3H, s, NMe), 2.25 (Me of p-tolyl), −27.6 (1H, br s, Ru–H);
order to obtain reliable integration values. Microanalyses were as noted for 3a, the hydride resonance became a pseudo-
performed in this department on
instrument.
a
Carlo Erba 1106 triplet at low temperature. IR: νRu–H 2079 w.
In situ syntheses of Ru(L)X(P–N)(PPh3) (X = Cl, Br) and
RuL2(P–N)(PPh3) (L = H, SH, OH)
cis-RuCl2(P–N)(PPh3)(η2-H2) (2a)
These O2-sensitive species were prepared in NMR tubes
equipped with poly(tetrafluoroethylene) J. Young valves.
The RuCl2(P–N)(PPh3) precursor (1a) was prepared in situ by
stirring a solution of RuCl2(PPh3)3 (85.1 mg, 0.09 mmol) and
P–N (29.2 mg, 0.09 mmol) in acetone (10 mL) at 50 °C for
30 min. H2 was then bubbled slowly through the solution for
Ru(H)Cl(P–N)(PPh3) (3a)
2 h during which time the dark green colour of 1a changed to A solution of RuCl2(P–N)(PPh3) (1a, 10 mg, 0.014 mmol) and
orange. The mixture was stirred for 48 h when a yellow precipi- PS (3 mg, 0.014 mmol) in CD2Cl2 (∼1 mL) on exposure to
tate that formed was collected and stored under Ar; the solid 1 atm H2 turned orange instantaneously to form 3a that was
was susceptible to loss of H2 with re-formation of the green stable at r.t. The NMR data are essentially the same as those
precursor. Yield: 35 mg, 52%. Anal. Calcd C38H37NCl2P2Ru: C, listed above for isolated 3a.
61.54; H, 5.03; N, 1.89. Found: C, 61.47; H, 4.89; N, 1.75.
2
31P{1H} NMR (CDCl3): δ 49.30 (d, P–N), 45.49 (d, PPh3); JPP
=
Ru(SH)Cl(P–N)(PPh3) (5a)
1
26.8. H NMR: δ 8.4–6.4 (29H, m, Ph), 3.68 (3H, s, NMe), 3.17
(3H, s, NMe), −10.90 (2H, br s, Ru(η2-H2)). IR (KBr): νRu–H2
2149 cm−1. Yellow, block crystals of 2a were obtained from a
saturated acetone solution of the complex left standing under
1 atm H2 for 2 days.
(a) To an NMR-tube containing 1a (10 mg, 0.014 mmol) and
NaSH·xH2O (5 mg, ∼0.05 mmol), d6-acetone (0.75 mL) was
vacuum transferred with the aid of liquid N2. The resulting
orange solution was stored at −78 °C (dry ice/acetone), and
NMR spectra were measured at this temperature. 31P{1H}
2
1
NMR: δ 55.26 (d, P–N), 46.33 (d, PPh3); JPP = 30.88. H NMR:
δ 8.1–6.2 (29H, m, Ph), 3.27 (3H, s, NMe), 3.18 (3H, s, NMe),
−2.08 (1H, s, SH). This species decomposed above −30 °C.
(b) To an NMR tube containing 1a (10 mg, 0.014 mmol)
and PS (3 mg, 0.014 mmol), CD2Cl2 (0.75 mL) was vacuum
transferred. Exposure of this solution to 1 atm H2S formed an
orange solution and 5a is observed at −78 °C. 31P{1H} NMR:
Ru(H)Cl(P–N)(PR3); R = Ph (3a), p-tolyl (3b)
A benzene solution (2 mL) of 1a/b (0.05 mmol) and PS (40 mg,
0.20 mmol) was stirred under 1 atm H2 for 10 h at r.t.; the ori-
ginal green solution rapidly turned orange and became turbid
due to precipitation of PSH+Cl−, which was identified by its
1H NMR spectrum in CD2Cl2 solution.3 The mixture was fil-
tered through Celite to remove the PSH+Cl−, and subsequent
addition of hexanes to the filtrate precipitated a yellow product
(3a/b) that was collected, washed with hexanes (3 × 15 mL),
and vacuum dried. Yield: ∼30 mg (78%). The air-sensitive pro-
ducts were handled under Ar in a glove-box, but acceptable
2
δ 54.52 (d, P–N), 46.06 (d, PPh3); JPP = 30.1. 1H resonances
could not be assigned because of overlapping NMe2 signals
from RuCl2(P–N)(PPh3)(SH2), 6 (see below), PS, and PSH+.
Ru(SH)Br(P–N)(PPh3) (5a′)
elemental analysis was obtained only for 3b. Anal. Calcd The species was made in situ by procedure (a) described above
C41H42NClP2Ru: C, 65.90; H, 5.66; N, 1.87; Cl, 4.74. Found: C, for 5a, but using RuBr2(P–N)(PPh3) (10 mg, 0.012 mmol) as
65.88; H, 5.91; N, 1.86, Cl, 4.98.
precursor. 31P{1H} NMR at −78 °C: δ 56.62 (d, P–N), 46.16
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 7614–7621 | 7615

View Article Online
Paper
Dalton Transactions
2
(d, PPh3); JPP = 30.5. 1H NMR: δ 8.1–6.2 (29H, m, Ph), 3.56
(3H, s, NMe), 3.17 (3H, s, NMe), −1.63 (1H, s, Ru–SH).
Ru(SH)2(P–N)(PPh3) (6)
Like formation of 5a/a′ (see above), 6 was formed when either
RuX2(P–N)(PPh3), X = Cl (1a) or Br, was reacted in d6-acetone
with excess NaSH·xH2O, or in CD2Cl2 with 1 atm H2S in the
presence of PS (3 equiv.). The resulting yellow solution was
stable at −78 °C, but at r.t. decomposed to uncharacterized
brown species after ∼10 min. 31P{1H} NMR: δ 84.06 (d, P–N),
2
1
59.53 (d, PPh3); JPP = 33.8. H NMR: δ 8.1–6.4 (29H, m, Ph),
3.20 (6H, s, NMe2), 0.80 (2H, s, Ru–(SH)2).
Ru(OH)Cl(P–N)(PPh3)
The species was formed after dissolving 1a (10 mg,
0.014 mmol) and NaOH (∼5 equiv.) in d6-acetone and heating
the solution at 60 °C for 2 h. NMR spectra of this orange solu-
tion were measured at r.t. 31P{1H} NMR: δ 64.09 (d, P–N), 50.76
Fig. 1 ORTEP diagram of cis-RuCl2(P–N)(PPh3)(η2-H2) (2a) with 50% probability
thermal ellipsoids.
2
1
(d, PPh3); JPP = 43.0 Hz. H NMR: δ 8.9–6.6 (29H, m, Ph), 3.04
(3H, s, NMe), 2.69 (3H, s, NMe).
Table 1 Selected bond lengths (Å) angles (°) for cis-RuCl2(P–N)(PPh3)(η2-H2)
(2a) with estimated standard deviations in parentheses. The H1 site represents
the double-occupancy η2-H2 refined isotropically
Ru(OH)Br(P–N)(PPh3)
The species was prepared as described for the chloro(hydroxy)
species, but using RuBr2(P–N)(PPh3) (10 mg, 0.12 mmol) as
precursor. 31P{1H} NMR at r.t.: δ 65.95 (d, P–N), 51.23
Bond lengths (Å)
2
(d, PPh3); JPP = 41.2. 1H NMR: δ 8.2–6.4 (29H, m, Ph), 3.22
Ru–H(1)
Ru–P(1)
Ru–P(2)
1.60(2)
Ru–N(1)
Ru–Cl(1)
Ru–Cl(2)
2.306(2)
2.4543(7)
2.4090(6)
2.2884(7)
2.3098(6)
(3H, s, NMe), 2.72 (3H, s, NMe).
Bond angles (°)
P(1)–Ru–N(1)
P(1)–Ru–Cl(1)
P(1)–Ru–Cl(2)
P(1)–Ru–P(2)
P(2)–Ru–N(1)
P(2)–Ru–Cl(1)
P(2)–Ru–Cl(2)
Cl(1)–Ru–Cl(2)
Ru(OH)2(P–N)(PPh3)
80.34(6)
172.22(2)
88.52(2)
105.27(3)
172.78(6)
82.34(3)
97.79(2)
88.86(2)
Cl(1)–Ru–N(1)
Cl(2)–Ru–N(1)
P(1)–Ru–H(1)
P(2)–Ru–H(1)
N(1)–Ru–H(1)
Cl(1)–Ru–H(1)
Cl(2)–Ru–H(1)
92.20(6)
86.78(6)
93.6(8)
87.3(8)
87.8(8)
88.3(8)
173.8(8)
This species was observed when d6-acetone solutions of 1a or
the dibromo analogue and NaOH (∼5 equiv.) were reacted for
5 h at 60 °C, when the initially formed orange solution became
orange-brown. 31P{1H} NMR at r.t.: δ 79.11 (d, P–N), 73.44
2
(d, PPh3); JPP = 67.4. 1H NMR: δ 8.2–6.4 (29H, m, Ph), 2.60
(3H, s, NMe), 2.28 (3H, s, NMe).
found in a very reasonable position, and was assigned as two
H-atoms. All other H-atoms were placed in calculated posi-
tions. The ORTEP plot and selected bond lengths and angles
of 2a are shown in Fig. 1 and Table 1, while the full experimen-
tal parameters and details of the structure are given in CIF
format in the ESI.†
Ru(H)2(P–N)(PPh3)
This orange species was formed on reacting 1a or the dibromo
species with NaH (∼5 equiv.) in d6-acetone at 60 °C for
∼15 min. NMR spectra were measured at r.t. 31P{1H} NMR:
δ 61.64 (d, P–N), 50.44 (d, PPh3); JPP = 24.71. 1H NMR:
2
δ 8.1–6.1 (29H, m, Ph), 2.51 (6H, s, NMe2), −21.16 (2H, d of d,
Ru–(H)2, 2JHP = 32.7, 29.1).
X-ray crystallographic analysis
Results and discussion
Structure of cis-RuCl2(P–N)(PPh3)(η2-H2) (2a)
X-ray analysis of 2a was carried out at 173 K on a Siemans
SMART Platform CCD area diffractometer with graphite mono-
chromated MoKα radiation (0.71073 Å). Some crystallographic Complex 2b, the P(p-tolyl)3 analogue of 2a, was previously syn-
data for the yellow, monoclinic crystals are: 8555 total reflec- thesized in situ by exposure of RuCl2(P–N)(P(p-tolyl)3) to 1 atm
tions, 4935 unique (Rint = 0.0167), 4816 observed [I > 2σ(I)], R1 H2 in CDCl3 at r.t., and was characterized as a molecular-H2
= 0.0208; wR2 = 0.0532; GOF = 1.026; residual density = 0.379 e complex by NMR studies.5 These included: (a) determination
Å−3. The structure was solved by direct methods and refined by over the temperature range 202–293 K of T1, the spin-lattice
full-matrix least squares (SHELXL-97).7 All non H-atoms were relaxation time of the hydrogen nuclei, which showed a
refined anisotropically. The η2-H2 was somewhat problematic minimum value of 13.4 0.2 ms corresponding to an intra-
to refine: the H-atoms were not found in the difference map, molecular H–H bond distance of 0.87 0.03 Å;5,8 and (b) iden-
but a difference peak attributed to the H–H sigma bond was tification of the η2-HD isotopomer as a 1 : 1 : 1 triplet (1JHD
=
7616 | Dalton Trans., 2013, 42, 7614–7621
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
2
30 Hz) of 1 : 2 : 1 triplets (cis, JHP = 8.5 Hz).5 Species 2a could Thermodynamic studies 2a in solution and in the solid state
also be made similarly in situ in CD3Cl from RuCl2(P–N)(PPh3)
When 2a is dissolved in solution at r.t., an equilibrium is
1
(1a) and H2, and the r.t., high-field H signal, a broad singlet
established by rapid dissociation of some H2 to partially form
at δ −11.02, was the same as that reported for the p-tolyl ana-
logue.5 The r.t., 31P{1H} data for 2a were also very similar to
RuCl2(P–N)(PPh3)(1a). 31P{1H} and 1H NMR spectral data in
C6D6 at ∼20 °C (Fig. 2 and 3, respectively) clearly illustrate the
equilibrium. The PA and PX resonances of 1a1,5 and 2a are
seen in Fig. 2 (the difference in the line-widths of the 1a
signals results from the presence of trace water and has been
discussed in our recent paper2). Fig. 3 shows the single NMe2
resonance for 1a1,5 and the two signals for the inequivalent
methyls of the NMe2 in 2a. The resonances of 2a in C6D6 are
slightly shifted from those in CDCl3 given in the Experimental:
δPA and δPX are now seen at 47.14 and 45.33 (2JPP = 26.5 Hz),
respectively, the δ(Me) signals at 3.78 and 2.79, and the δ (H2)
signal at −10.6. The equilibrium constant for formation of 2a
from 1a as shown in eqn (1) was determined by vacuum trans-
ferring C6D6 into an NMR tube containing a solid sample of 1a
(0.0163 M); after the solution was warmed to r.t., H2 was
added at 1 atm, and the sample shaken for ∼2 min to attain
equilibrium before being placed again under 1 atm H2 at a
series of temperatures (20.9 to 60.6 °C).
those reported for 2b: AX doublets at δ 48.04 and 42.14 (2JPP
=
27 Hz),5 the lower and higher field signals being respectively
2
those of the P–N and P(p-tolyl)3 ligands.2 The JPP value is in
the same 29–30 Hz region reported for the corresponding H2S,
MeSH and EtSH adducts, which also form as cis-Cl2 species
from RuCl2(P–N)(PPh3) (Chart 1).1,6
That 2a is an η2-H2 complex is supported further following
its isolation as a yellow solid from a reaction of RuCl2(PPh3)3
and P–N in acetone under 1 atm H2 (see Experimental
section), and its solid state properties. Microanalysis agrees
with the formulation of 2a, which is stable as a solid under H2,
but slowly loses H2 under Ar and in air with decomposition.
The IR spectrum shows a medium intensity, νRu–H2 band at
2149 cm−1; νH–H is not observed and such bands are weak and
rarely located (typically in the 2400 to 2700 cm−1 range).8 X-ray
analysis of crystals grown from a saturated acetone solution 2a
under H2 reveals a distorted octahedral structure (Fig. 1 and
Table 1) with ‘normal’ bond distances of RuII to the P-, Cl-,
and N-atoms; in the structure, P(1) and P(2) correspond
respectively to the PA and PX NMR labels. The bond lengths
and angles are within 0.02 Å and 2–8° of those of the analo-
gous H2S and thiol complexes.1,6 This, coupled with the fact
that the Ru–PA bond length and δPA of 2a fit well to a linear
correlation between these two parameters found for all the
RuCl2(P–N)(PR3)L complexes (R = Ph, p-tolyl; L = H2S, thiols,
and H2O),2 further confirms that 2a is a six-coordinate RuII
species formed via coordination of H2 at the vacant position of
five-coordinate 1a; i.e. 2a is a η2-H2 complex and not a RuIV-
dihydrido species. The relatively short Ru–H(1) distance of
1.60 Å seems consistent with reasonable stability with respect
to loss of H2 in the solid state. For example, this length is
shorter than the 1.81 Å of Ru–H2 in the H2-labile complex
trans-[RuH(η2-H2)(diphos)2][BPh4],9 and is longer than this dis-
tance (∼1.45 Å) within the non-labile complex (isoPFA)(η2-H2)-
Ru(μ-Cl)2(μ-H)RuH(PPh3)2, where isoPFA is 1-[α-N,N′-dimethyl-
aminoethyl]-2-(diphenylphosphino)ferrocene.10 Of note, as in
2a, the η2-H2 in the isoPFA species is cis to both a chloride and
a P–N ligand, and trans to the second chloride, whereas a
difference lies in the occupancy of the remaining cis position:
a PPh3 in 2a and a hydride in the isoPFA complex. The Ru–H
(1) distance in 2a is close to those reported for RuII-hydride
species such as RuH(Cl)(PPh3)3 (1.70 Å),11 RuH(dmpe)2-
(naphthyl) (1.67 Å),12 RuH(PPh3)3(O2CCH3) (1.68 Å),13 RuH(Cl)-
(diop)2 (1.65 Å),14 trans-[RuH(η2-H2)(diphos)2][BPh4] (1.64 Å),9
and RuH(SC6H4pCH3)(CO)2(PPh3)2 (1.58 Å);15 dmpe = 1,2-bis-
(dimethylphosphino)ethane, and diop = 4,5-bis((diphenylphos-
phino)methyl)-2,2-dimethyl-1,3-dioxolane). The NMR, IR, and
structural data for 2a are incompatible with a RuIV(H)2 formu-
lation. Some reversible solution thermodynamic and DSC data
for 2a (see below) are also consistent with the RuII(η2-H2)
formulation.
K
!
trans-RuCl2ðPꢀNÞðPPh3Þ ð 1aÞ þ H2
ð1Þ
cis-RuCl2ðPꢀNÞðPPh3ÞðH2Þ ð 2aÞ
Fig. 2 31P{1H} NMR spectrum showing the 1a/2a equilibrium in C6D6 at 20 °C.
Fig. 3 1H NMR spectrum showing the 1a/2a equilibrium in C6D6 at 20 °C;
inset shows Ru(η2-H2) signal; phenyl region omitted; free H2 (δ 4.44) not seen
because of low [H2].
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 7614–7621 | 7617

View Article Online
Paper
As the temperature is raised, the integrations of the NMe2 and mol−1 1
Dalton Transactions
,
which again reveals the greater exothermicity for the
Ru–H2 proton signals of 2a decrease while those of 1a (δ 3.07, reverse coordination process. In both the H2 and H2S systems,
NMe2) increase; this shows qualitatively that formation of 2a is the product from the DSC experiments is cis-RuCl2(P–N)(PPh3)
exothermic. The free H2 in solution is not observed in Fig. 3 and not the trans-Cl2-isomer reactant used in deriving the solu-
because of its low concentration but, under 1 atm of the gas, tion thermodynamic data, as in eqn (1). The formation of the
the free H2 in solution is readily detected at δ 4.44. The cis-Cl2-isomer and its rapid conversion to the trans-Cl2-isomer
measured integrated peak areas for an NMe signal of 2a in solution were established in a DSC experiment with the H2S
(labelled α), the NMe2 signal of 1a (β), and the free H2 (ω), system, which led to a ΔH° value of about −39 kJ mol−1 for the
allow for the calculation of K (= [2a]/[1a][H2]) using the follow- cis to trans conversion;1 the corresponding value estimated
ing expressions:
from the H2-system is about −(50–26), i.e. −24 kJ mol−1, which
is considered reasonable agreement. Of note, ΔH° values in
the range of 48–15 kJ mol−1 have been reported for the solid-
phase isomerizations of the chloride ligands within the six-
coordinate RuCl2(CO)(PR)3 complexes, where R = Ph2Me,
PhMe2, and Me3,18 and so the ΔH° values derived from our
cis-RuCl2(P–N)(PPh3) systems seem reasonable.
½2aꢁ α=3
½2aꢁ α=3
¼
½Ruꢁtotal
1 þ x
x ¼
¼
; y ¼
; ½1aꢁ ¼
½2aꢁ
;
½1aꢁ β=6
½H2ꢁ ω=2
½2aꢁ ¼ ½Ruꢁtotal ꢀ ½1aꢁ; ½H2ꢁ ¼
y
1
Table S1† shows values for the H NMR integrations, x, y,
[1a], [2a] and [H2], at various temperatures, and the corres-
ponding K values; a van’t Hoff plot is shown in Fig. S1.† At
25 °C, K = 261 20 M−1, with associated ΔH°, ΔS°, and ΔG° (at
25 °C) values of −26 4 kJ mol−1, −40 15 J mol−1 K−1 and
−13.8 0.2 kJ mol−1, respectively. The exothermicity and nega-
tive entropy, as found for the corresponding coordination of
H2S and thiols in C6D6,1 are consistent with an equilibrium of
the type shown in eqn (1); ignoring solvent effects the forward
reaction is expected to be entropically unfavourable and thus
has to be exothermic. The data for the H2S binding are K =
153 5 M−1, ΔH° = −46 4 kJ mol−1, ΔS° = −112 14 J mol−1
Reactions of cis-RuCl2(P–N)(PPh3)(L) (L = H2 and H2S) with
proton sponge (PS)
The reaction of 2b, the P(p-tolyl)3 analogue of 2a, with PS, a
strong base and usually considered weakly coordinating
(because of steric effects),3,19,20 was briefly mentioned in a
communication from our group,5 and this established the
chemistry outlined in eqn (3) where the phosphine ligands are
omitted. The chlorohydrido product Ru(H)Cl(P–N)(P(p-tolyl)3)
(3b) was isolated, but few details were provided;5 the Exper-
imental section now gives details on the isolation of both 3a
and 3b, and the in situ synthesis of 3a. Bothspecies are very
air-sensitive, but are well characterized spectroscopically; a satis-
factory elemental analysis was only obtained for 3b. For 3a
and 3b, the hydride is seen in the IR, and at −80 °C the broad,
K
−1, and ΔG° = −12.5 0.1 kJ mol−1; thus, compared to H2-
binding, the exothermicity for H2S binding is greater but this
is counteracted by a more unfavourable entropy change, and
overall the K (and ΔG° values) are reasonably close. The more
favourable entropy change for the H2 system possibly results
from the small size of the H2 molecule. Of note, the ΔH° and
ΔS° values for the H2-binding are close to the respective values
of about −32 kJ mol−1 and −97 J mol−1 K−1 reported for the
similar binding of H2 to the five-coordinate Ru(H)Cl(CO)-
(PiPr3)2 complex in toluene.16 Related is the ΔH° value of
−60 kJ mol−1 estimated for the forward reaction of eqn (2)
where P–P = Ph2P(CH2)4PPh2, in which H2 again binds at a for-
mally five-coordinate dimeric, RuII species in CH2Cl2;17 in this
system, the van’t Hoff plot actually yielded anoverall ΔH° value
of about zero but consideration of the endothermic process
(∼60 kJ mol−1) for the terminal to bridging chloride rearrange-
ment led to the estimated ΔH° value for the coordination of
H2. A positive ΔS° value also suggested that the apparent five-
coordinate reactant might coordinate CH2Cl2 at each Ru
centre,17 and so a direct comparison of the ΔH° values for the
P–P and P–N systems is probably unrealistic.
2
high-field 1H signal is seen as a pseudo-triplet with a JHP
value indicating cis-phosphine hydride coupling.21 The 31P{1H}
data are close to those of the corresponding precursor trans-
Cl2 complexes (Chart 1),1,5 and the structures of 3a/b are likely
to be square pyramidal, likely those of 1a/b with a Cl− replaced
by H−, or with the hydride trans to the vacant site.
PS;Keq
H2
ꢀ!
2
þ
ꢀ
ꢀꢀꢀ!
RuCl2 ꢀ ðη -H2ÞRuCl2 ꢀꢀꢀ HRuCl þ PSH Cl
ð3Þ
3a=b
1a=b
2a=b
Measurement of the Keq shown in eqn (3) allows for esti-
mation of the acidity (i.e. the pKa value) of the coordinated
η2-H2 through the relationship pKa = pKeq + pKPSH+, where the
last term is the pKa value for the conjugate acid of PS, which in
H2O is 12.1.3 Although a direct comparison of acidity cannot
be made between pKa values in organic and aqueous solutions,
they are related linearly; e.g., the expression pKa(H2O) =
pKa(MeCN) − 7.5 has been established within a series of
hydrido complexes.22 Evaluation of Keq was carried out for 2a
in CD2Cl2 because this solvent dissolves all the species of eqn
(3). Thus, a mixture of 1a and 0.75 to 3.0 equiv. PS in CD2Cl2
½RuClðP–PÞðμ-ClÞꢁ2 þ H2 ⇄ ðη2-H2ÞðP–PÞRuðμ-ClÞ3RuClðP–PÞ
ð2Þ
The relatively labile nature of the η2-H2 in complex 2a in the was exposed to 1 atm H2 at 20 °C when rapid equilibration
solid state is quantified by a DSC experiment, where ΔH° for occurs to give a yellow-brown solution. NMR spectra (e.g.,
the loss of H2 was found to be 50 3 kJ mol−1 (Fig. S2†); the Fig. 4 and 5) reveal that 1a, 2a, and 3a, are in equilibrium, the
corresponding value for the removal of H2S was 85
2 kJ species all being stable under the experimental conditions,
7618 | Dalton Trans., 2013, 42, 7614–7621
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
and Keq is readily obtained from the expression [3a][PSH+]/[2a]-
[PS], in which only concentration/integrations ratios are
needed (e.g., from Fig. 5). The data give Keq = 15 5 and thus
the pKa of 2a is 10.9 0.1, which is within the wide range of
values reported for other RuII(η2-H2) complexes such as trans-
[RuX(H2)(P–P)]+ (X = Cl, H; P–P = a diphosphine) and several
Ru(H2)Cp-containing complexes.8d,e,23 Our prime goal was to
compare the pKa value of 2a with that of the corresponding
H2S species, cis-RuCl2(P–N)(PPh3)(SH2), because there are no
reports on the acidity of coordinated H2S. Hence, attempts
were made to again use the reaction with PS as in eqn (3), but
instability of the mercapto products complicated the studies,
which are described below.
Exposure of a CD2Cl2 solution of 1a to 1 atm H2S at r.t. gen-
erates immediately the well characterized, yellow cis-RuCl2-
(P–N)(PPh3)(SH2) complex (4a, see Fig. 6a).6 In the presence of
3 equiv. of PS, a new yellow species (6) is seen in the 31P{1H}
spectrum and is identified by an AX pattern at δ 82.25 and
57.88 (2JPP = 34.0 Hz) (Fig. 6b); however, within 10 min, 6
decomposes to a dark brown solution with precipitation of
Fig. 6 31P{1H} NMR spectra (CD2Cl2) for five different experiments with [Ru]total
= 0015 M: RuCl2(P–N)(PPh3)(SH2) (4a) at 20 °C (a) and RuCl2(P–N)(PPh3) (1a) +
3PS + 1 atm H2S at 20 °C (b), −25 °C (c), −60 °C (d), −70 °C (e); 5a and 6 slowly
decompose even at low temperatures.
1
some white PSH+Cl− (identified by its H spectrum in CDCl3,
which more readily dissolves PS/PSH+ mixtures than CD2Cl2).
A brown solid isolated from the filtrate gave no 31P{1H} NMR
signals and broad 1H peaks in the phenyl and δ 3.8–1.5
regions, suggestive of a paramagnetic RuIII species. These
observations perhaps resemble the decomposition of
[RuII(NH3)5(SH2)]2+ to [RuIII(NH3)5(SH)]2+ and H2,24 although
no H2 was detected in our system. No reaction was observed
between H2S and PS in CD2Cl2, and thus PS clearly does
abstract a proton from the coordinated H2S which is thus
more acidic than free H2S (pKa ∼ 7 in H2O at ∼20 °C);25
however, the Ru–SH species formed undergo decomposition
even at low temperatures (see below).
Fig. 4 31P{1H} NMR spectrum (20 °C, CD2Cl2) of the in situ reaction of RuCl2-
(P–N)(PPh3) (1a) with 1.5 equiv. PS under 1 atm H2.
The unstable 6 was identified as Ru(SH)2(P–N)(PPh3) by its
in situ formation at r.t. from reaction of either RuX2(P–N)-
(PPh3) (X = Cl, 1a; Br, 1b) with excess NaSH·xH2O in
d6-acetone, both halide ligands of each species being replaced
by SH−. The 31P{1H} data are essentially the same as those
1
seen in CD2Cl2 (Fig. 6b), and the SH protons are seen as a H
singlet at δ 0.80; the NMe2 protons are thus equivalent, imply-
ing that 6 has a symmetrical, square-pyramidal structure
similar to that of 1a. Attempts to isolate 6 at −78 °C, where it
was stable, from the 1a/H2S/PS reaction in CD2Cl2 were unsuc-
cessful: addition of hexanes did precipitate a yellow-brown
solid but, on being filtered off at about −20 °C, it decomposed
to the dark brown powder. The 31P{1H} NMR spectra of the
same in situ reaction at ≤−30 °C (Fig. 6c–e) show the presence
Fig. 5 1H NMR spectrum (20 °C, CD2Cl2) spectrum in the δ 2.7 to 3.7 region of
the in situ reaction of RuCl2(P–N)(PPh3) (1a) with 1.5 equiv. PS under 1 atm H2.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 7614–7621 | 7619

View Article Online
Paper
Dalton Transactions
Chart 2 Plausible pathway for reaction of cis-RuCl2(P–N)(PPh3)(SH2) with PS.
of 4a, 6, and a third species (5a). This was identified as Ru- molecules. This implies that the strong π-donor properties of
(SH)Cl(P–N)(PPh3) by NMR data from its in situ formation at RuII (d6 low spin) are of minor importance compared to elec-
−78 °C via a 1 : 1 reaction of 1a and PS under 1 atm H2S in tron donation to the metal from the respective bonding orbi-
CD2Cl2, or a 1a/NaSH reaction in d6-acetone: an AX 31P{1H} tals of the coordinated H2 (the H–H σ-bond) and the S-atom of
pattern and, in the 1H NMR, a Ru–SH singlet at −2.08 and two the H2S; the donation leads to increased acidity of the H–H
NMe resonances. The corresponding bromo-mercapto species bond more than donation from the S-atom does for an S–H
5a′ could also be formed in situ; both 5a (see Chart 2) and 5a′ bond. The only other reported pKa value for coordinated H2S is
are likely square-pyramidal species in CD2Cl2, but the in situ that of ∼4 estimated for [Ru(NH3)5(SH2)]2+ in aqueous
data given in the Experimental section are in d6-acetone, media,24 where the cationic nature (vs. a neutral species such
where six-coordinate, solvated species are likely present. Of as 4a) would certainly lead to higher acidity.
note, RuII-hydrido(mercapto) and -bis(mercapto) complexes,
The complex RuCl2(P–N)(PPh3) (1a) surprisingly catalyses
thermally stable at r.t., have been made from reactions of H2S the protonation of PS using H2S as the proton source. For
with Ru(H)2(Ph2PCH2PPh2)2, where the H2S provides a proton example, an in situ, r.t. reaction of 1a with excess H2S
to remove the metal-hydride as H2.26 The similarly stable Ru- (100 equiv. added by a syringe) and PS (10 equiv.) in CD2Cl2
(SH)2(CO)2(PPh3)2 has been made from reaction of H2S with converted in 1 h all the PS (δH 2.80, NMe2) to PSH+ (δH 3.38,
Ru(H)2(CO)2(PPh3)2, as well as via a two-stage oxidative NMe2), the process being accompanied by decomposition of
addition–protonation reaction of Ru(CO)2(PPh3)3 with H2S.26
1a and 4a. The counter-anions available for the 10 equiv. of
In the P–N system, at the conditions used to obtain the data PSH+ must be Cl− and SH−, the latter presumably being
of Fig. 6, the H2S complex is initially fully formed in a rever- formed via H2S that becomes coordinated at some repeatedly
sible process;1 thus 5a, which is formed by proton abstraction generated vacant site; the catalyst could be a species derived
from the Ru–H2S moiety by PS (Chart 2), must be an inter- from the decomposition product(s).
mediate en route to 6. Warming the low temperature solutions
The cis-RuCl2(P–N)(PPh3)(RSH) complexes, where R = Me,
to 20 °C (Fig. 6) gives complete formation of 6, further sup- Et,1 did not react with PS in CD2Cl2, consistent with lower
porting the 5a to 6 conversion. How this occurs is unclear, but acidities of the thiols (pKa ∼ 10.5)27 vs. that of H2S; in prin-
initial coordination of H2S followed by a further proton ciple, a stronger base such as an alkoxide could deprotonate
abstraction seems most plausible (Chart 2). A variable temp- the thiol complexes, and the resulting thiolate species are
erature NMR study shows that the formation of 5a is reversible likely to be more stable than the corresponding mercapto
whereas formation of 6 is not. The integration ratio of the 31P species. The pKa of [Ru(NH3)5(EtSH)]2+ in aqueous media has
{1H} resonances of 4a and 5a decreases as the temperature is been estimated to be ∼9, compared to the value of ∼4 for the
raised from −70 to −50 °C, but this is reversed when the H2S adduct.24
system is re-cooled to −70 °C; at 20 °C, 6 is fully formed, and
The aqua complex RuCl2(P–N)(PPh3)(H2O)2 in CD2Cl2 was
remains so when the temperature is decreased. Efforts to also unreactive toward PS, again presumably because of the
measure equilibrium concentrations of 4a and 5a en route to much lower acidity of water (vs. H2S), but there is a difference
determining the pKa of 4a (as done for the H2 complex – see in that the H2O is trans to phosphorus while the H2S is trans to
above) were complicated because 5a (and 6) species are chloride (Chart 1). Species thought to be the mono(hydroxy)
unstable even at −70 to −50 °C within the in situ experiments species Ru(OH)X(P–N)(PPh3) (X = Cl, Br) and dihydroxy species
of the type shown in Fig. 6. Discrepancies between 31P{1H} Ru(OH)2(P–N)(PPh3) could be made in situ, however, in
NMR integrations were evident for repeated experiments, even d6-acetone, respectively, from 1a and from 1a or 1b using
with long acquisition times, and there was also uncertainty in NaOH, analogous to formation of the mercapto species using
the solubilities of the species at the low temperatures. Never- NaSH. The mono(hydroxyl) species have NMR data similar to
theless, from the data at −25 and −60 °C shown in Fig. 6c and those of the corresponding mono(mercapto) species 5a/5a′ and
6d, the respective Keq values for the 4a + PS ⇄ 5a + PSH+Cl− are thought to have the same structures (see above). The NMR
equilibrium are ∼1.2 × 10−2 and ∼6 × 10−2, giving pKa values data of the dihydroxy species contrast with those of the bis-
of 4a of ∼14.2 to 13.5 in CD2Cl2 (see Appendix 1†); the value is (mercapto) analogue 6 in having two NMe resonances, a PPh3
2
thus expected to be somewhat higher at 20 °C, where the pKa signal further downfield by ∼15 ppm, and a JPP value about
value of ∼11 was determined for the H2 complex. Thus the twice as large; a structure with cis-OH groups and trans-P
acidity of coordinated H2 is greater than that of coordinated atoms (cf. Chart 1) is indicated. Use of NaH (vs. NaOH) under
H2S, in contrast to the comparative acidities of the free similar in situ conditions generated Ru(H)2(P–N)(PPh3). The
7620 | Dalton Trans., 2013, 42, 7614–7621
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
hydrides are seen at high-field as a doublet of doublets with
2JHP values indicative of cis-phosphine hydride coupling21
within a trans-dihydride structure; such species are well-known
in RuII chemistry,28 including some with hydride shifts in the
same −20 ppm region.28b
1990, 23, 95; (d) P. G. Jessop and R. H. Morris, Coord. Chem.
Rev., 1992, 121, 155; (e) D. M. Heinekey and W. J. Oldham
Jr., Chem. Rev., 1993, 93, 913; (f) M. A. Esteruelas and
L. A. Oro, Chem. Rev., 1998, 98, 577.
9 A. Albinati, W. T. Klooster, T. F. Koetzle, J. B. Fortin,
J. S. Ricci, J. Eckert, T. P. Fong, A. J. Lough, R. H. Morris
and A. P. Golombek, Inorg. Chim. Acta, 1997, 259, 351.
10 C. Hampton, W. R. Cullen and B. R. James, J. Am. Chem.
Soc., 1988, 110, 6918.
Conclusions
Thermodynamic data are determined for the reversible
binding of H2 (which forms a molecular-H2 species) to the
five-coordinate RuCl2(P–N)(PPh3) complex (1a), where P–N =
o-diphenylphosphino-N,N′-dimethylaniline, and are compared
to data for the H2S adduct; the solution equilibrium constants
at 25 °C are similar, the equality resulting from a less exother-
mic process for H2-binding being balanced by a less negative
ΔS° value. In CD2Cl2 solution, the acidity (measured by reac-
tions with proton sponge) of the coordinated H2 is greater
than that of H2S, the reverse order of the solution acidities of
the free molecules. The proton loss from the H2 and H2S gen-
erates, respectively, hydrido and mercapto derivatives that can
also be made via chloride-substitution reactions of 1a with
sodium hydride and hydrosulfide salts; use of NaOH similarly
gives hydroxy derivatives, which were not formed via attempted
analogous deprotonation of coordinated H2O.
11 A. C. Skapski and P. G. H. Troughton, Chem. Commun.,
1968, 1230.
12 U. A. Gregory, S. D. Ibekwe, B. T. Kilbourn and
D. R. Russell, J. Chem. Soc. A, 1971, 1118.
13 A. C. Skapski and F. A. Stephens, J. Chem. Soc., Dalton
Trans., 1974, 390.
14 R. G. Ball and J. Trotter, Inorg. Chem., 1981, 20, 261.
15 P. G. Jessop, S. J. Rettig, C.-L. Lee and B. R. James, Inorg.
Chem., 1991, 30, 4617.
16 D. G. Gusev, A. B. Vymenits and V. I. Bakhmutov, Inorg.
Chem., 1992, 31, 1.
17 D. E. K.-Y. Chau and B. R. James, Inorg. Chim. Acta, 1995,
240, 419.
18 D. W. Krassowski, K. Reimer, H. E. LeMay Jr. and
J. H. Nelson, Inorg. Chem., 1988, 27, 4307.
19 (a) R. W. Alder, Chem. Rev., 1989, 89, 1215;
(b) B. Brzezinski, E. Grech, Z. Malarski and L. Sobczyk,
J. Chem. Soc., Perkin Trans. 2, 1991, 1267.
20 (a) S. N. Gamage, R. H. Morris, S. J. Rettig, D. C. Thackray,
I. S. Thornburn and B. R. James, J. Chem. Soc., Chem.
Commun., 1987, 894; (b) N. J. Farrer, R. McDonald and
J. S. McIndoe, Dalton Trans., 2006, 4570.
21 (a) H. D. Kaesz and R. B. Saillant, Chem. Rev., 1972, 72,
231; (b) B. N. Chaudret, D. J. Cole-Hamilton, R. S. Nohr and
G. Wilkinson, J. Chem. Soc., Dalton Trans., 1977, 1546.
22 S. S. Kristjansdottir and J. R. Norton, in Transition Metal
Hydrides: Recent Advances in Theory and Experiment, ed.
A. Dedieu, VCH, New York, 1991, ch. 10.
Acknowledgements
We thank the Natural Sciences and Engineering Council of
Canada for funding, Dr Victor G. Young, Jr (X-ray Crystallo-
graphic Laboratory at the University of Minnesota) for solving
the structure of 2a, and Colonial Metals Inc. for a loan of
RuCl3·xH2O.
References
1 E. S. F. Ma, S. J. Rettig, B. O. Patrick and B. R. James, Inorg. 23 (a) R. H. Morris, Inorg. Chem., 1992, 31, 1471;
Chem., 2012, 51, 5427.
2 E. S. F. Ma, B. O. Patrick and B. R. James, Dalton Trans.,
2013, 42, 4291.
(b) E. P. Cappellani, S. D. Drouin, G. Jia, P. A. Maltby,
R. H. Morris and C. T. Schweitzer, J. Am. Chem. Soc., 1994, 116,
3375; (c) B. Chin, A. J. Lough, R. H. Morris, C. T. Schweitzer
and C. D’Agostino, Inorg. Chem., 1994, 33, 6278.
3 (a) R. W. Alder, P. S. Bowman, W. R. S. Steele and
D. R. Winterman, Chem. Commun., 1968, 723; (b) H. A. Staab 24 C. G. Kuehn and H. Taube, J. Am. Chem. Soc., 1976, 98, 689.
and T. Saupe, Angew. Chem., Int. Ed., 1988, 27, 865.
4 P. S. Hallman, T. A. Stephenson and G. Wilkinson, Inorg.
Syn., 1970, 12, 237.
25 D. D. Perrin, Ionization Constants of Inorganic Acids and
Bases in Aqueous Solution, Pergamon Press, Oxford, 2nd
edn, 1982.
5 D. C. Mudalige, S. J. Rettig, B. R. James and W. R. Cullen, 26 P. G. Jessop, C.-L. Lee, G. Raster, B. R. James, C. J. L. Lock
J. Chem. Soc., Chem. Commun., 1993, 830. and R. Faggiani, Inorg. Chem., 1992, 31, 4601.
6 E. S. Ma, S. J. Rettig and B. R. James, Chem. Commun., 27 (a) M. M. Kreevoy, E. T. Harper, F. E. Stary, E. A. Katz and
1999, 2463.
J. H. Sellstedt, J. Org. Chem., 1964, 29, 1641; (b) J. P. Danehy
and C. J. Noel, J. Am. Chem. Soc., 1960, 82, 2511.
28 (a) K. Abdur-Rashid, S. E. Clapham, A. Hadzovic,
J. N. Harvey, A. J. Lough and R. H. Morris, J. Am. Chem.
Soc., 2002, 124, 15104; (b) R. F. R. Jazzar, P. H. Bhatia,
M. F. Mahon and M. K. Whittlesey, Organometallics, 2003,
22, 670.
7 G. M. Sheldrick, SHELKL-97, Bruker-AXS, Madison, WI,
1997.
8 (a) D. G. Hamilton and R. H. Crabtree, J. Am. Chem. Soc.,
1988, 110, 4126; (b) M. T. Bautista, K. A. Earl, P. A. Maltby,
R. H. Morris, C. T. Schweitzer and A. Sella, J. Am. Chem.
Soc., 1988, 110, 7031; (c) R. H. Crabtree, Acc. Chem. Res.,
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 7614–7621 | 7621