B-H bond activation using an electrophilic metal complex: Insights into the reaction pathway

Article abstract of DOI:10.1021/ic300390s

A highly electrophilic ruthenium center in the [RuCl(dppe) ₂][OTf] complex brings about the activation of the B-H bond in ammonia borane (H₃N·BH₃, AB) and dimethylamine borane (Me₂HN·BH₃, DMAB). At room temperature, the reaction between [RuCl(dppe)₂][OTf] and AB or DMAB results in trans-[RuH(η²-H₂)(dppe)₂][OTf], trans-[RuCl(η²-H₂)(dppe)₂][OTf], and trans-[RuH(Cl)(dppe)₂], as noted in the NMR spectra. Mixing the ruthenium complex and AB or DMAB at low temperature (198/193 K) followed by NMR spectral measurements as the reaction mixture was warmed up to room temperature allowed the observation of various species formed enroute to the final products that were obtained at room temperature. On the basis of the variable-temperature multinuclear NMR spectroscopic studies of these two reactions, the mechanistic insights for B-H bond activation were obtained. In both cases, the reaction proceeds via an η¹-B-H moiety bound to the metal center. The detailed mechanistic pathways of these two reactions as studied by NMR spectroscopy are described.

Full text of DOI:10.1021/ic300390s

Article
pubs.acs.org/IC
B−H Bond Activation Using an Electrophilic Metal Complex: Insights
into the Reaction Pathway
Rahul Kumar and Balaji R. Jagirdar*
Department of Inorganic & Physical Chemistry, Indian Institute of Science, Bangalore 560012, India
S
* Supporting Information
ABSTRACT: A highly electrophilic ruthenium center in the [RuCl(dppe)2]-
[OTf] complex brings about the activation of the B−H bond in ammonia
borane (H3N·BH3, AB) and dimethylamine borane (Me2HN·BH3, DMAB). At
room temperature, the reaction between [RuCl(dppe)2][OTf] and AB or
DMAB results in trans-[RuH(η2-H2)(dppe)2][OTf], trans-[RuCl(η2-H2)-
(dppe)2][OTf], and trans-[RuH(Cl)(dppe)2], as noted in the NMR spectra.
Mixing the ruthenium complex and AB or DMAB at low temperature (198/193
K) followed by NMR spectral measurements as the reaction mixture was
warmed up to room temperature allowed the observation of various species
formed enroute to the final products that were obtained at room temperature.
On the basis of the variable-temperature multinuclear NMR spectroscopic
studies of these two reactions, the mechanistic insights for B−H bond activation
were obtained. In both cases, the reaction proceeds via an η1-B−H moiety
bound to the metal center. The detailed mechanistic pathways of these two reactions as studied by NMR spectroscopy are
described.
of the B−H binding to the metal center and its subsequent
activation but also from its potential as a hydrogen source and
storage material.13,14 Although several systems are known to
bring about dehydrogenation of ammonia borane and the
substituted amine boranes the intricate mechanisms of these
reactions involving the B−H bond activation and the ability of
the metal complexes to retain a B−N moiety in the reaction
sequence are yet to be fully elucidated.13,14 Recently, Alcaraz
and Sabo-Etienne reviewed the progress made in the
coordination chemistry of ammonia borane and the substituted
amine boranes.13 In addition, reports on σ complexes of amine
boranes, coordinated aminoborane to the metal center, and
ammonia borane dehydrogenation using transition-metal
catalysts have also appeared.15 In our effort to study the
binding and activation of the B−H bond to electrophilic metal
centers, we recently undertook the task of studying the
reactivity of ammonia borane and the related amine boranes
toward the five-coordinated ruthenium complex [RuCl-
(dppe)2][OTf] [1; dppe = 1,2-bis(diphenylphosphino)ethane,
INTRODUCTION
■
The activation of the very inert C−H bond in simple alkanes,
e.g., CH4, C2H6, cyclohexane, etc., and its subsequent
functionalization using transition-metal complexes have re-
mained challenging problems since the late 1950s.1 The weak
nature of the M−H−C interaction makes it very difficult to
study the binding and activation of the C−H bonds in alkanes
under ambient conditions. The binding and cleavage of the X−
H (X = H, Si, B) bonds of other molecules such as dihydrogen,
silanes, and borane−Lewis base adducts have close resem-
blances with one another and are quite significant because they
serve as models for the activation and functionalization of the
C−H bond in methane and other alkanes. Whereas a large
number of the so-called σ complexes bearing H2 and Si−H2−6
as a ligand/ligand fragment have been reported, the synthesis
and characterization of coordination compounds with saturated
hydrocarbons as ligands have been particularly challenging.7−10
On the other hand, four-coordinated boranes are isoelectronic
with methane; the B−H bond in these species binds with metal
centers in a stable M−H−B interaction and could serve as good
models for the C−H binding and activation.
−
Ph2PCH2CH2PPh2; OTf = CF3SO3 ]. Herein, we describe the
results of the mechanistic studies of those reactions.
Hartwig et al. demonstrated the η2 binding of a three-
coordinated borane, which paved the way to exploration of the
coordination chemistry of three-coordinated borane com-
plexes.11 On the other hand, the chemistry of the four-
coordinated boron species has been undergoing rapid develop-
ment with the introduction of the Lewis base−borane
complexes of the M(CO)5 (M = Cr, Mo, W) fragment by
Shimoi et al.12 In this context, ammonia borane has been the
focus of recent research interest not only from the standpoint
EXPERIMENTAL SECTION
■
General Procedures. All manipulations were carried out under a
dry and oxygen-free N2 atmosphere using standard Schlenk and inert-
atmosphere techniques unless otherwise specified. Reagent-grade
solvents were dried and distilled under a N2 atmosphere from sodium
benzophenone [hexane, petroleum ether, tetrahydrofuran (THF),
Received: February 21, 2012
Published: December 17, 2012
© 2012 American Chemical Society
28
dx.doi.org/10.1021/ic300390s | Inorg. Chem. 2013, 52, 28−36

Inorganic Chemistry
Article
Figure 1. 1H NMR spectral stack plot with temperature showing intermediates 3a and 3b and complexes 4−6 in the reaction of complex 1 with AB
in CD2Cl2.
diethyl ether] just before use. Dichloromethane was first dried and
distilled using P2O5 and then dried and distilled over CaH2.
Dichloromethane-d2 (CD2Cl2) was purchased from Cambridge
Isotope Ltd., USA, and used as received. Methanol was dried and
distilled using MgI2, whereas acetone was dried and distilled over
K2CO3.
Ammonia borane (H3N·BH3, AB) and H3N·BD3 (ABD) were
synthesized according to literature procedures.16 The amine borane
adduct, dimethylamine borane (Me2HN·BH3, DMAB), was purchased
from Sigma-Aldrich and was purified by sublimation prior to use. The
[RuCl(dppe)2][OTf] complex 1 was prepared using the reported
method.17
1
The H, 31P{1H}, 11B, and 19F NMR spectral data were obtained
using an Avance Bruker 400 MHz instrument. The 31P{1H} NMR
spectra were recorded relative to 85% H3PO4 (aqueous solution) as an
external standard and the 19F NMR spectra relative to CFCl3. Variable-
temperature (VT) NMR experiments were carried out in flame-sealed
1
(in vacuo) NMR tubes. The H and 31P spin−lattice relaxation time
(T1) measurements were carried out using the inversion−recovery
method.18
Reaction of Complex 1 with H3N·BH3, AB (1:1). In a thoroughly
dried Schlenk NMR tube having a Teflon stopcock, complex 1 (22 mg,
0.02 mmol) and AB (0.7 mg, 0.02 mmol) were loaded and the tube
was evacuated and then refilled with N2 gas. It was then immersed in a
liquid N2 bath, and CD2Cl2 (0.7 mL) was added quickly under a
Figure 2. 31P{1H} NMR spectral stack plot with temperature showing
intermediate 3b and complexes 4−6 in the reaction of complex 1 with
AB in CD2Cl2.
positive flow of N2 gas such that the solvent froze before coming into
contact with the solid components at the bottom of the tube. The
stopcock was quickly replaced, and the tube was once again evacuated
and held under vacuum for about 20−30 min with the contents in the
frozen state. The stopcock was closed tightly, and the tube was flame-
sealed with the contents still in the frozen state. It was then inserted
Table 1. Relative Percentage of Complexes 4−6 after the
a
Reaction of Complex 1 with AB and DMAB at 298 K in
1
into the NMR probe, which was precooled to 198 K, and the H,
Stoichiometric Ratio 1:1
31P{1H}, and 11B NMR spectra were recorded. The reaction
progressed as the temperature was raised. Figures 1 and 2 show the
reagent
4
5
6
1
VT H and 31P{1H} NMR spectral stack plots, respectively, of this
AB
96
44
trace
28
4
reaction. The same reaction was also carried out at room temperature,
and the products were characterized using NMR spectroscopy. The
relative percentages of the products obtained in this reaction at room
temperature are summarized in Table 1. The NMR spectra have been
deposited in the Supporting Information. At room temperature, the
intense-dark red-colored solution of the reaction mixture turned pale
yellow instantaneously accompanied by vigorous gas evolution, which
ceased after a few minutes, and an off-white-colored precipitate also
formed.
Reaction of Complex 1 with Me2HN·BH3, DMAB (1:1). This
reaction was conducted in a manner similar to that of complex 1 with
AB starting with 22 mg (0.02 mmol) of complex 1 and 1.3 mg (0.02
mmol) of DMAB. The sealed NMR tube with the frozen contents was
DMAB
28
a
Integrations of the peaks were done after ensuring that no further
change in the peak intensities took place, which signaled that the
reaction had gone to completion.
inserted into the NMR probe precooled to 193 K, and the 1H,
31P{1H}, and 11B NMR spectra were recorded as the reaction
progressed upon an increase in the temperature. Figures 3−5 show the
1
1H NMR spectral stack plot, H−1H COSY spectrum, and 31P{1H}
NMR spectral stack plot, respectively, of this reaction. The same
reaction was carried out at room temperature, and in this case as well,
the products were identical with those of the reaction of complex 1
29
dx.doi.org/10.1021/ic300390s | Inorg. Chem. 2013, 52, 28−36

Inorganic Chemistry
Article
1
Figure 3. H NMR spectral stack plot showing intermediates 2′ and
3b and complexes 4−7 in the reaction of complex 1 with DMAB in
CD2Cl2.
with AB. The relative percentages of the products obtained in this
reaction are summarized in Table 1.
Reaction of Complex 1 with H3N·BD3 (1:1). In a 5 mm NMR
tube, complex 1 (11 mg, 0.01 mmol) and H3N·BD3 (0.4 mg, 0.01
mmol) were loaded, and the tube was capped with a septum. Then
CD2Cl2 (0.7 mL) was added under a flow of N2 gas, and the 1H NMR
spectrum of the reaction mixture was recorded immediately upon
mixing the contents of the tube to observe the formation of
isotopomers (spectra have been deposited in the Supporting
Information).
Figure 5. 31P{1H} NMR spectral stack plot with temperature showing
intermediates 2′ and 3b and complexes 4−7 in the reaction of
complex 1 with DMAB in CD2Cl2.
Reaction of Complex 1 with Excess AB and DMAB. These
reactions were carried out in a manner similar to that of complex 1
with AB and DMAB (1:1 stoichiometric amounts). Complex 1 (11
mg, 0.01 mmol) was treated with AB (3.2 mg, 0.1 mmol) and DMAB
(6.0 mg, 0.1 mmol) in separate NMR tubes in a CD2Cl2 solvent. The
sealed NMR tube with the frozen contents was inserted into the NMR
probe precooled to 193 K, and the 1H, 31P{1H}, and 11B NMR spectra
were recorded as the reaction progressed upon an increase in the
complex 1 with AB) was added to 50−60 mL of ammoniacal
dichloromethane (CH2Cl2 dissolved in liquid ammonia, 243 K) in a
Schlenk flask and stirred for ∼30 min. Then, the volume was reduced
to 10 mL. After that, ∼15 mL of hexane was added, which resulted in
the formation of an off-white precipitate. It was washed several times
with hexane and dissolved in dichloromethane, and the 1H and
31P{1H} NMR spectra were recorded (see the Supporting Information
for the spectra).
1
temperature. The H, 31P{1H}, and 11B NMR spectral stack plots are
given in Supporting Information. The same reactions were carried out
at room temperature as well (for spectra, see the Supporting
Information).
RESULTS AND DISCUSSION
■
Reaction of Complex 1 with AB (1:1). The [RuCl-
(dppe)2][PF6] complex was reported by Morris and his co-
workers.19 In spite of being a coordinatively unsaturated 16-
electron species, it is air-stable in solution and reacts with small
molecules such as H2, CO, and CH3CN to afford the respective
18-electron derivatives. This compound also brings about the
catalytic hydrogenation of trimethylsilyl enol ethers to
cyclohexanone and Me3SiH.20 The reaction involves heterolytic
cleavage of the bound H2 ligand as a key step, wherein a H+ is
transferred to the O atom of the enol ether. We chose to use
Reaction of DMAB with HOTf (1:1): Characterization of
Me2HN·BH2(OTf). In a regular NMR tube, DMAB (6 mg, 0.1 mmol)
was dissolved in 0.4 mL of dried and degassed dichloromethane and
then HOTf (9 μL, 0.1 mmol) was added followed by another 0.2 mL
of CH2Cl2. Vigorous bubbling was noted. The 11B NMR spectrum was
recorded immediately after the reaction, which shows
Me2HN·BH2(OTf) formation along with a small amount of unreacted
DMAB (see the Supporting Information).
Synthesis of trans-[Ru(H)(NH3)(dppe)2][OTf] and Its Charac-
terization by NMR Spectroscopy. A reaction mixture of complexes
4−6 and trans-[Ru(H)(NH3)(dppe)2][OTf] (from the reaction of
Figure 4. 1H−1H COSY spectrum showing the off-diagonal cross peak for the σ-borane intermediate (2′) in the reaction of complex 1 with DMAB
at 193 K in CD2Cl2. Figure at the right after zooming.
30
dx.doi.org/10.1021/ic300390s | Inorg. Chem. 2013, 52, 28−36

Inorganic Chemistry
Article
[RuCl(dppe)2][OTf] to explore its reactivity with AB and
DMAB (Scheme 1). AB and DMAB are simple Lewis acid−
electron-rich ligand could approach the metal center trans to
the Cl and in between the two equatorial P atoms, resulting in a
six-coordinated species having a trans geometry. A second
possibility could also be considered, wherein a B−H bond will
approach the Ru center trans to one of the equatorial P atoms
and adjacent to the Cl atom. These two possibilities result in
the intermediate species 2a and 2b (unobserved; see Scheme
2). These unobserved intermediates undergo rapid B−H bond
cleavage reactions. In the case of intermediate 2a, the
[H3N·BH2]+ species that results upon B−H bond cleavage
abstracts the Cl−. On the other hand, intermediate 2b
undergoes B−H and Ru−Cl metathesis. In both the cases
(Scheme 2, step ii), H3N·BH2Cl is eliminated. Elimination of
H3N·BH2Cl from the intermediate 2a results in a five-
coordinated species, [RuH(dppe)2][OTf] (3a), adopting a
square-pyramidal geometry. The quintet pattern [J(H,Pcis) =
18.0 Hz] at δ −13.9 ppm in the 1H NMR spectrum is
consistent with the square-pyramidal geometry of species 3a.
This species appears at 198 K. The metathesis reaction results
in the formation of yet another five-coordinated species,
[RuH(dppe)2][OTf] (3b), adopting a trigonal-bipyramidal
geometry. The species 3b exhibits a doublet of a quartet at δ
Scheme 1. Reaction of Complex 1 with Amine Boranes
base adducts in which the BH H atoms are hydridic and the
NH H atoms are acidic in nature. The reaction of 1 with AB or
DMAB at room temperature resulted in the formation of trans-
[RuH(η2-H2)(dppe)2][OTf] (4), trans-[RuCl(η2-H2)(dppe)2]-
[OTf] (5), and trans-[RuH(Cl)(dppe)2] (6) complexes and
H3N·BH2Cl21 or Me2HN·BH2Cl,22 as evidenced by NMR
spectroscopy (Scheme 1; see the Supporting Information for
spectra).
The spectral properties of 4−6 matched those of the
reported ones.19,23 The relative percentages of 4−6 formed in
these reactions are summarized in Table 1. Although AB is not
completely soluble in CD2Cl2, yet it reacts vigorously with
[RuCl(dppe)2][OTf] accompanied by H2 gas evolution. The
dark red solution of 1 in CD2Cl2 instantaneously faded to pale
yellow, and an off-white precipitate also formed. Because the
NMR spectra showed the presence of only B−H bond-cleaved
products and no intermediate species, it suggests that the
reaction proceeds very rapidly at room temperature.
1
−7.1 ppm in the H NMR spectrum at a temperature (208 K)
higher than that of 3a. The doublet pattern is due to trans H−P
coupling on the order of 95.4 Hz, and the quartet [J(H,Pcis) =
20.6 Hz] results from the coupling of the hydride H atom with
the three cis-P atoms. Perutz and his co-workers reported a
quintet pattern at δ −11.9 ppm [J(H,Pcis) = 20.0 Hz] and a
1
singlet at δ 52.6 ppm in the H and 31P{1H} NMR spectra,
respectively, in THF-d8 for complex 3a.24 Complex 3b is
hitherto unknown. As the temperature of the reaction mixture
is raised, 3a undergoes isomerization to 3b (Scheme 2, step iii),
Therefore, we carried out the same reaction at low
temperature and monitored the progress of the reaction using
VT NMR spectroscopy in order to gain insight into the
1
which is evident from the disappearance of the H NMR
spectral signal of 3a and the increase in the signal intensity of
3b at 208 K (Figure 1). This is a minor pathway by which 3b is
produced in addition to the main route via the metathetical
reaction in 2b. The B−H bond cleavage product, H3N·BH2Cl,
was noted in the 11B NMR spectrum as a triplet at δ −9.0 ppm
(see the Supporting Information).21 On the basis of the prior
1
mechanistic aspects. The VT H and 31P{1H} NMR spectral
stack plots of this reaction are shown in Figures 1 and 2,
respectively, and the proposed mechanism is given in Scheme 2.
Complex 1 adopts a trigonal-bipyramidal geometry in
solution and is stable in CH2Cl2 for several days. An
a
Scheme 2. Proposed Mechanism of the Reaction of Complex 1 with AB
a
Step vii: transformation of 5 or 6 to 4 takes place in the presence of 6 or 5, respectively Step viii: transformation of 4 to 6 takes place in the
presence of Cl−.
31
dx.doi.org/10.1021/ic300390s | Inorg. Chem. 2013, 52, 28−36

Inorganic Chemistry
Article
theoretical work of Baker et al.,25,26 we propose that this species
reacts with unreacted AB, resulting in the dehydrogenation of
AB in sequential fashion, and leads to formation of the BNHx
polymer via an unobserved oligomeric species
[BH2ClNH2BH2NH3]. We also cannot rule out other pathways
for the dehydrogenation of AB such as concerted B−H and N−
H activation15t or stepwise B−H and N−H activation at the
metal center.15g The H2 generated via these reactions has two
possible substrates to react with: (i) reaction with 3a and 3b to
afford 4 and (ii) reaction with the starting complex [RuCl-
(dppe)2][OTf], resulting in the corresponding dihydrogen
complex 5 (see Scheme 2, step v). We noted the formation of
both 4 and 5 at 218 K (Figures 1 and 2 and Scheme 2).
The peaks corresponding to 1 broaden and then resharpen as
the temperature is increased. Mezzetti et al. studied the
dynamic behavior of an analogous species, [RuCl(dcpe)2]-
[BPh4] [dcpe = 1,2-bis(dicyclohexyl)phosphinoethane,
(C6H11)2PCH2CH2P(C6H11)2].28 Our observations with regard
to VT NMR spectral behavior for complex 1 were similar to
those of the [RuCl(dcpe)2][BPh4] complex. Mezzetti et al. did
not elaborate on these observations. We propose that the
observed low-temperature NMR spectral behavior of complex 1
could be attributed to the stereochemical rigidity resulting from
a favored weak interaction in a position cis to chloride (in an
octahedral geometry) of triflate (OTf) counterion or the ortho
protons of the phenyl group of the dppe ligand with the metal
center (see the VT 1H, 31P, and 19F NMR spectral stack plots in
the Supporting Information). We noted broadening and an
−
−
Complexes similar to 4 and 5 with BPh4 and PF6
counterions, respectively, were reported by Morris et al.19,23
1
Complex 4 shows a quintet at δ −10.5 ppm (Ru−H) and a
upfield shifting of one of the phenyl proton signals in the H
1
broad singlet at δ −4.8 ppm (Ru−η2-H2) in the H NMR
NMR spectra, broadening and downfield shifting of the 31P
NMR spectral signals, and upfield shifting of the 19F NMR
spectral signal at low temperatures (233−223 K). These
observations suggest that the ortho protons of the dppe phenyl
group and triflate (OTf) counterion compete with one another
to bind to the vacant site on the metal center. These weak
interactions result in stereochemical rigidity at low temperature.
Upon warming, this rigidity breaks down and the 31P NMR
peaks sharpen. We also examined the VT NMR spectral
behavior of complex 1 without added AB. In this case as well,
the spectral properties were similar to those in the reaction of 1
with AB. This rules out any contribution from AB to this
dynamic process (see the Supporting Information). Further-
more, Morris and co-workers reported similar downfield shifts
of the 31P NMR spectral signals upon cooling for a similar
complex [RuCl(dppe)2][PF6] alone.19
spectrum and a sharp singlet at δ 68.0 ppm in the 31P{1H}
NMR spectrum; on the other hand, complex 5 shows a broad
singlet at δ −12.3 ppm in the 1H NMR and a sharp singlet at δ
51.5 ppm in the 31P{1H} NMR spectra.19,23 Upon warming of
the reaction mixture further, the concentrations of 4 and 5
increased, as was evident from an increase in their signal
1
intensities in the H NMR spectra (Figure 1). It is interesting
to note that a singlet appears transiently at δ 4.6 ppm in the 1H
NMR spectrum, which could be ascribed to free H2 before it
reacts with the respective substrates. Whereas compound 4
persists upon warming the reaction mixture to room temper-
ature, the bound H2 in compound 5 undergoes heterolysis to
afford the hydride derivative 6 (Scheme 2, step vi). Heterolysis
of the bound H2 in 5 in the presence of a base leads to the
formation of 6, which shows a quintet at δ −18.9 ppm in the
1H NMR spectrum and a sharp singlet at δ 62.3 ppm in the
31P{1H} NMR spectrum. The spectral stack plots (Figures 1
and 2) suggest that complexes 5 and 6 transform into complex
4 to a certain extent; however, the pathway that they take is not
clear. This was reported earlier by Morris and co-workers.19
Complex 4 in the presence of Cl− ion could transform into
complex 6, as reported by Morris and co-workers;19 it is not
clear whether the same is happening in our case because there
is another Cl− acceptor in the reaction solution, namely,
[H3N·BH2]+. On the basis of 31P{1H} inversion−recovery and
spin-saturation transfer experiments on the reaction mixture,
we noted that the transformation of 5 or 6 to 4 takes place in
the presence of 6 or 5, respectively, with slow reaction rates
(Scheme 2, step vii, and the Supporting Information).
Additionally, we independently established the transformation
of 4 to 6 in the presence of Cl− ion and also the transformation
of 5 and 6 to 4 (see the Supporting Information). Thus,
complexes 4 and 6 undergo interconversion through different
pathways. The interconversion of 6 and 4 becomes facile with
an increase in the temperature, which leads to the expected
gradual drop in the 31P T1 values. However, we observed 31P
T1(min) at 263 K for 4 and 6 (see the Supporting
Information), which suggests their slower interconversion and
tumbling rate below 263 K and faster above it in comparison to
their respective Larmor frequencies.27 The appearance of 4−6
upon warming of the sample in this experiment to room
At 198 K, the 31P{1H} NMR spectrum showed the presence
of the starting material in major quantity. In addition, a very
small quantity of 3a (δ 53.0 ppm), which was evident by the
weak intensity of the signal, was also noted. Upon warming to
218 K, three apparent multiplets at δ 79.5, 65.3, and 64.4 ppm
for 3b appeared. Raising the temperature of the reaction
mixture further resulted in the evolution of 4−6. Compounds 4
and 5 arise from the binding of H2 produced from the reaction
of H3N·BH2Cl and AB, which is an intermolecular pathway
involving B−H as well as N−H activation in stepwise fashion
through on- and off-metal processes, respectively. Dehydrogen-
ation of amine boranes going through on- and off-metal
mechanisms has been described in detail recently by Sabo-
Etienne et al.15m Complexes 4−6 adopt a trans geometry with
respect to the dppe ligand, making all of the P atoms equivalent
to each other; hence, only singlets were noted in the 31P{1H}
NMR spectrum for these complexes at δ 68.0, 51.5, and 62.3
ppm, respectively.19,23 Free H2 generated in the reaction gets
consumed by its two major scavengers, complexes 3a/3b and
the starting complex 1, as noted above (Scheme 2, step v). All
of these complexes react almost simultaneously with H2, giving
rise to complexes 4 and 5 (Scheme 2, step v). These reactions
proceed until all of the H2 liberated is consumed. The fate of
the proton equivalent that results upon heterolysis of the bound
H2 in complex 5 is not known. In order to test whether any of
the steps shown in Scheme 2 are reversible, we warmed the
sample and recooled it several times. We found no evidence of
any of the reactions being reversible. We also carried out the
reaction of [RuCl(dppe)2][OTf] with AB in molar ratios of 1:5
and 1:10. In these experiments, we noted proportionately larger
amounts of H3N·BH2Cl being formed, which ultimately gets
1
temperature in the H NMR spectrum is consistent with the
observations made in the room temperature NMR spectral
1
data. The spectral changes observed in the H NMR spectra
with temperature matched those of the 31P{1H} NMR spectral
data shown in Figure 2.
32
dx.doi.org/10.1021/ic300390s | Inorg. Chem. 2013, 52, 28−36

Inorganic Chemistry
Article
a
Scheme 3. Proposed Mechanism of the Reaction of Complex 1 with DMAB
a
Step vi: transformation of 5 or 6 to 4 takes place in the presence of 6 or 5, respectively. Step vii: transformation of 4 to 6 takes place in the presence
of Cl−.
transformed into the BNHx polymer (see the Supporting
Information).We also observed the formation of μ-amino-
diborane (μ-NH2B2H5), which is characterized by a triplet of
doublets (1:2:1) signal at δ −26.0 ppm in the 11B NMR
spectrum. This species was also established using 1H−11B
HETCOR spectroscopy (see the Supporting Information).
Stephens et al. provided spectroscopic characterization for μ-
aminodiborane as well as the pathway of its formation,25 and,
recently, this compound has also been characterized by X-ray
crystallography by Shore et al.29 The formation of complex 3b
via Ru−Cl and B−H bond metathesis is an example of σ-
complex-assisted metathesis (σ-CAM), which has recently been
highlighted by Perutz and Sabo-Etienne.30 The σ-CAM
reactions are attracting the interest of organometallic chemists
mainly because they could serve as models for metal-complex-
catalyzed carbon−heteroatom bond formation in organic
synthesis through C−H bond activation.
Herein, the metathesis reaction is facilitated by the affinity of
boron toward chlorine as well as having a triflate counterion in
complex 1, which is noncoordinating (or weakly coordinating)
in nature, resulting in the formation of a B−Cl rather than a B−
OTf bond; H3N·BH2Cl was noted in the NMR spectra rather
than H3N·BH2(OTf) (see the Supporting Information). In
order to establish that B−H bond activation had taken place,
we carried out the reaction of complex 1 with NH3·BD3 and
noted the formation of HD isotopomers of complex 4. A triplet
featuring a 1:1:1 intensity pattern instead of a broad signal for
(Ru-η2-HD) at δ −4.8 ppm in the 1H NMR spectrum indicates
the presence of isotopomers (see the Supporting Informa-
tion).23
same products, i.e., 4−6, as in the case of the reaction of AB
with complex 1, and the 11B NMR spectrum showed
Me2HN·BH2Cl.22 Once again in this case we carried out VT
NMR spectral measurements to elucidate the reaction pathway
(see Scheme 3). The VT NMR spectral studies revealed some
interesting features of the reaction, different from those with
AB, with a notable one being observation of a σ-borane species
as a key intermediate at 193 K. The B−H bond of DMAB
approaches the Ru center trans to one of the equatorial P atoms
and adjacent to the Cl atom of complex 1, resulting in a σ-
borane intermediate 2′, wherein the B−H bond bound to the
metal is cis to Cl and trans to one of the P atoms of the dppe
1
ligand (Scheme 3, step i). The H NMR spectrum of the
reaction mixture at 193 K exhibits a broad singlet at δ −12.1
ppm (Figure 3). We also recorded the 1H−1H COSY spectrum
at 193 K (Figure 4), which revealed that the signal at δ −12.1
ppm correlates with one off-diagonal cross peak at δ ∼1.5 (br)
ppm. The integration ratio of these two peaks is ∼1:2,
indicating that the high-field signal corresponds to the B−H H
atom bound to the metal and the low-field one corresponds to
two terminal B−H H atoms in an almost frozen state.
Furthermore, the off-diagonal cross peak shows two spots
with the center of the contours that are apart by ∼48.0 Hz,
suggesting that the B−H bond bound to the metal center is
considerably stretched and that it is nearly a hydride. The
J(B,H) of 60.0 Hz found for σ-borane complexes by Shimoi et
al. has been ascribed to a moderately stretched B−H bond.12
The 31P{1H} NMR spectrum of 2′ is comprised of three signals
at δ 39.1 (m), 41.4 (m), and 48.9 (m) ppm, indicating the
presence of three kinds of phosphorus environments (Figure 5
and Scheme 3). The broad feature of the spectral signals at 193
K precluded characterization of the σ-borane intermediate 2′ by
11B NMR spectroscopy.
Reaction of Complex 1 with DMAB (1:1). DMAB with
two methyl groups on the N atom is sterically encumbered and
at the same time possesses an electron-rich B−H bond by
virtue of the two methyl substituents. It is slightly less polar
than AB and soluble to a greater extent in organic solvents, in
particular CH2Cl2 rather than AB. The reaction of DMAB with
complex 1 at room temperature resulted in the formation of the
We also measured the spin−lattice relaxation time, T1, for the
1H NMR spectral signal at δ −12.1 ppm at 193 K and obtained
a value of ∼116 ms, indicating its hydridic nature (see the
Supporting Information).18 The intermediate 2′ undergoes
33
dx.doi.org/10.1021/ic300390s | Inorg. Chem. 2013, 52, 28−36

Inorganic Chemistry
Article
three different reactions upon an increase in the temperature of
the sample, (a) B−H bond cleavage, (b) Ru−Cl and B−H
bond metathesis (via σ-CAM mechanism), and (c) B−N bond
cleavage with Ru−Cl and B−H bond metathesis (Scheme 3,
step ii), and the evolution of their respective products, 6, 3b
(trigonal-bipyramidal geometry), and trans-[RuH(NHMe2)-
1
(dppe)2][OTf] (7) was noted in the H NMR spectrum at
203 K (Figure 3). As in the case of the reaction of complex 1
with AB, we warmed and recooled the sample in the reaction of
complex 1 with DMAB several times as well to determine if any
of the steps were indeed reversible. We, however, found no
evidence that pointed to any of the reactions being reversible.
On the other hand, from the data that we have obtained for this
reaction, it is not possible for us to decipher whether step ii
shown in Scheme 3 proceeds in a stepwise or concerted
fashion. Therefore, neither of these pathways can be ruled out.
Compound 6 is also formed via the intermediacy of 5, which
undergoes heterolytic cleavage of the bound H2 ligand. The
origin of 5 is the reaction of the starting complex 1 with H2.
The relatively unstable 3b picks up H2 produced from the
reaction of Me2HN·BH2Cl or Me2HN·BH2(OTf) (see the
Supporting Information) and the unreacted DMAB to form 4,
which was also noted in the 1H NMR spectrum at 203 K. Upon
an increase in the sample temperature, the concentration of 4
builds up at the expense of 3b (Figures 3 and 5). On the other
hand, Ru−Cl and B−H bond metathesis with concerted B−N
bond cleavage (Scheme 3, step ii) leads to the formation of
cationic complex 7, which exhibits a quintet at δ −17.3 ppm
Figure 6. 11B NMR spectral stack plot with temperature showing
Me2HN·BH2(OTf) and Me2HN·BH2Cl in the reaction of complex 1
with DMAB in CD2Cl2.
of three kinds of phosphorus environments, whereas those of
complexes 4−6 along with the intermediate 3b appeared when
the temperature was raised (Figure 5 and Scheme 3).
Surprisingly, in this case, we did not observe complex 3a in
1
the H NMR as well as in the 31P{1H} NMR spectra. We
1
attempted to characterize 2′ further by H−31P correlation
spectroscopy at 193 K. However, no correlation spot was
found, which could be due to the low concentration of the
intermediate 2′ in the solution (see the Supporting
Information).
Reaction of 1 with Excess AB and DMAB. Reaction of
complex 1 with excess AB and DMAB were performed at both
room and variable temperature. The reaction in each case
proceeded with vigorous gas evolution and resulted in species
that were similar to those obtained in the stoichiometric (1:1)
reactions and were identified and characterized; in addition, the
reactions, however, were rendered complex by the formation of
multiple species, which could not be identified and
unambiguously characterized. The reaction of complex 1 with
excess AB at room temperature afforded complexes 4, 6, and
trans-[RuH(NH3)(dppe)2][OTf] and an unidentified species
based on 1H and 31P{1H} NMR spectroscopy (see the
Supporting Information). However, the 11B NMR spectrum
at room temperature revealed the presence of the BNHx
polymer, μ-aminodiborane, and H3N·BH2Cl (see the Support-
ing Information).
1
and a singlet at δ 65.0 ppm in the H and 31P{1H} NMR
spectra, respectively. This species is stable up to a temperature
of 253 K, above which it almost completely transforms to 4
(see Figure 3). On the basis of the 31P{1H} inversion−recovery
experiment (see the Supporting Information) on the reaction
mixture at 203 K, we noted that 4 and 7 interconvert by slow
reaction rates.
Recently, Whittlesey and co-workers observed the σ-borane
complex [Ru(xantphos)(PPh3)(H3B·NH3)H)][BArF ] [xant-
4
phos = 4,5-bis(diphenylphosphino)-9,9′-dimethylxanthene;
BArF = [B{3,5-(CF3)2C6H3}4]−] in the reaction of [Ru-
4
(xantphos)(PPh3)(OH2)H)][BArF ] with AB in which AB
4
coordinates through the B−H bond to ruthenium. After 72 h,
they found a mixture of complexes [Ru(xantphos)(PPh3)(η2-
H2)H)][BArF ] and [Ru(xantphos)(PPh3)(NH3)H)][BArF ].
We also carried out a VT NMR (1H, 31P{1H}, and 11B)
spectral study of the reaction of complex 1 with excess AB. In
this reaction, we noted the evolution of complexes 4−6 and
trans-[RuH(NH3)(dppe)2][OTf] through the intermediacy of
3a and 3b. The spectral features of the trans-[RuH(NH3)-
(dppe)2][OTf] complex obtained as part of this mixture
matched those of an independently prepared sample (see the
Supporting Information). In addition, an unidentified species
4
4
The latter species forms when the σ-borane complex undergoes
B−N bond cleavage assisted by H2O in the dehydrogenation
process of AB followed by the coordination of NH3 to
ruthenium.31 In an independent reaction of complex 1 and AB
having a mixture of complexes 4 and 6, with aqueous NH3, a
1
quintet at δ −17.3 ppm was noted in the H NMR spectrum,
indicating the formation of trans-[RuH(NH3)(dppe)2][OTf] as
the exclusive product. In a separate reaction of complex 4 with
AB, again trans-[RuH(NH3)(dppe)2][OTf] was obtained as the
sole product (for details, see the Supporting Information). The
VT 11B NMR spectral stack plot revealed the presence of
Me2HN·BH2(OTf), whose intensity fades upon an increase in
the sample temperature with the concomitant evolution of
Me2HN·BH2Cl (Figure 6). As in the case of the reaction of 1
with AB, the species that persist upon warming the sample to
room temperature in the reaction of 1 with DMAB are 4−6.
The VT 31P{1H} NMR spectral studies are in conformity with
1
that shows a broad peak at δ −2.6 ppm in the H NMR
spectrum was also noted. This signal appeared at 293 K. The
31P{1H} NMR spectrum corroborates with the 1H NMR
spectral features with regard to the formation of complexes 4−6
and trans-[RuH(NH3)(dppe)2][OTf]. The 11B NMR spectra,
however, showed the presence of H3N·BH2Cl21 and another
unidentified species.
On the other hand, complex 1 reacts with excess DMAB at
room temperature to give complexes 4 and 6 and some other
1
unidentified species, as evidenced by H and 31P{1H} NMR
1
those of H NMR spectral studies. Spectral signals were noted
spectroscopy (Supporting Information). The 11B NMR
spectrum showed the presence of the cyclic dimer
(Me2NBH2)2,15a,32 showing a triplet at δ ∼5.0 ppm [J(B,H)
for the σ-borane intermediate (2′) at 193 K, three signals at δ
39.1 (m), 41.4 (m), and 48.9 (m) ppm, indicating the presence
34
dx.doi.org/10.1021/ic300390s | Inorg. Chem. 2013, 52, 28−36

Inorganic Chemistry
Article
(10) Bernskoetter, W. H.; Schauer, C. K.; Goldberg, K. I.; Brookhart,
M. Science 2009, 326, 553.
(11) Hartwig, J. F.; Muhoro, C. N.; He, X.; Eisenstein, O.; Bosque,
≈ 112 Hz], a broad triplet of doublets (1:2:1) at δ −17.8 ppm
assignable to μ-aminodiborane (H2B(μ-H,NMe2)BH2),33
a
triplet at δ ∼37 ppm corresponding to H2BNMe2,32d and a
few other unknown species. Cyclic dimer (Me2NBH2)2
formation could be an off-metal as well as an on-metal
process.15m The VT 1H and 31P{1H} NMR spectral stack plots
of the reaction of complex 1 with excess DMAB revealed the
presence of complexes 2′, 3b, 4, 6, and 7. The VT 11B NMR
spectral stack plot showed the formation of Me2HN·BH2(OTf)
at 223 K, and as the temperature of the sample was raised,
Me2HN·BH2Cl22 was also noted. The spectral assignment of
Me2HN·BH2(OTf) was confirmed using a sample prepared
independently by reacting DMAB and HOTf (see the
Supporting Information).
Thus, reactions of complex 1 with excess AB and DMAB
gave rise to products that were common to those of reactions
with stoichiometric amounts. At the same time, few other
products were also formed that could not be definitively
characterized, as a result of which the mechanistic pathways in
these cases could not be deduced.
R.; Maseras, F. J. Am. Chem. Soc. 1996, 118, 10936.
(12) Shimoi, M.; Nagai, S.; Ichikawa, M.; Kawano, Y.; Katoh, K.;
Uruichi, M.; Ogino, H. J. Am. Chem. Soc. 1999, 121, 11704.
(13) Alcaraz, G.; Sabo-Etienne, S. Angew. Chem., Int. Ed. 2010, 49,
7170.
(14) Staubitz, A.; Robertson, A. P. M.; Manners, I. Chem. Rev. 2010,
110, 4079.
(15) (a) Jaska, C. A.; Temple, K.; Lough, A. J.; Manners, I. J. Am.
Chem. Soc. 2003, 125, 9424. (b) Jaska, C. A.; Manners, I. J. Am. Chem.
Soc. 2004, 126, 1334. (c) Shrestha, R. P.; Diyabalanage, H. V. K.;
Semelsberger, T. A.; Ott, K. C.; Burrell, A. K. Int. J. Hydrogen Energy
2009, 34, 2616. (d) Denney, M. C.; Pons, V.; Hebden, T. J.; Heinekey,
D. M.; Goldberg, K. I. J. Am. Chem. Soc. 2006, 128, 12048.
(e) Blaquiere, N.; Diallo-Garcia, S.; Gorelsky, S. I.; Black, D. A.;
Fagnou, K. J. Am. Chem. Soc. 2008, 130, 14034. (f) Kaß, M.; Friedrich,
̈
A.; Drees, M.; Schneider, S. Angew. Chem., Int. Ed. 2009, 48, 905.
(g) Keaton, R. J.; Blacquiere, J. M.; Baker, R. T. J. Am. Chem. Soc.
2007, 129, 1844. (h) Conley, B. L.; Williams, T. J. Chem. Commun.
2010, 46, 4815. (i) Chapman, A. M.; Haddow, M. F.; Wass, D. F. J.
Am. Chem. Soc. 2011, 133, 8826. (j) Wright, W. R. H.; Berkeley, E. R.;
Alden, L. R.; Baker, R. T.; Sneddon, L. G. Chem. Commun. 2011, 47,
3177. (k) Kim, S.-K.; Han, W.-S.; Kim, T.-J.; Kim, T.-Y.; Nam, S. W.;
Mitoraj, M.; Pieko, L.; Michalak, A.; Hwang, S.-J.; Kang, S. O. J. Am.
Chem. Soc. 2010, 132, 9954. (l) Conley, B. L.; Guess, D.; Williams, T.
J. J. Am. Chem. Soc. 2011, 133, 14212. (m) Stevens, C. J.; Dallanegra,
R.; Chaplin, A. B.; Weller, A. S.; Macgregor, S. A.; Ward, B.; McKay,
D.; Alcaraz, G.; Sabo-Etienne, S. Chem.Eur. J. 2011, 17, 3011.
(n) Chaplin, A. B.; Weller, A. S. Inorg. Chem. 2010, 49, 1111.
(o) Chaplin, A. B.; Weller, A. S. Angew. Chem., Int. Ed. 2010, 49, 581.
(p) Johnson, H. C.; Robertson, A. P. M.; Chaplin, A. B.; Sewell, L. J.;
Thompson, A. L.; Haddow, M. F.; Manners, I.; Weller, A. S. J. Am.
Chem. Soc. 2011, 133, 11076. (q) Douglas, T. M.; Chaplin, A. B.;
Weller, A. S. J. Am. Chem. Soc. 2008, 130, 14432. (r) Sewell, L. J.;
Lloyd Jones, G. C.; Weller, A. S. J. Am. Chem. Soc. 2012, 134, 3598.
(s) Alcaraz, G.; Vendier, L.; Colt, E.; Sabo-Etienne, S. Angew. Chem.,
Int. Ed. 2010, 49, 918. (t) Paul, A.; Musgrave, C. B. Angew. Chem., Int.
Ed. 2007, 46, 8153.
CONCLUSIONS
■
The ruthenium complex [RuCl(dppe)2][OTf] brings about B−
H bond activation and cleavage in AB and DMAB via the
intermediacy of [RuCl(η1-HBH2·NH3)(dppe)2][OTf] and
[RuCl(η1-HBH2·NMe2H)(dppe)2][OTf], respectively. Because
the ruthenium complex 1 was found to be quite reactive toward
AB and DMAB, multinuclear VT NMR spectral studies were
carried out to decipher the various intermediate species that
were formed enroute to the final products. On the basis of
these studies, the reaction pathway in each case was deduced
and proposed.
ASSOCIATED CONTENT
* Supporting Information
■
S
NMR spectral data of the reactions and spin−lattice relaxation
time (T1, ms) measurements data. This material is available free
of charge via the Internet at http://pubs.acs.org.
(16) Ramachandran, P. V.; Gagare, P. D. Inorg. Chem. 2007, 46, 7810.
́
(17) Fox, A. M.; Harris, E. J.; Heider, S.; Perez-Gregorio, V.;
Zakrzewska, E. M.; Farmer, D. J.; Yufit, D. S.; Howard, J. A. K.; Low, P.
J. J. Organomet. Chem. 2009, 694, 2350.
(18) Hamilton, D. G.; Crabtree, R. H. J. Am. Chem. Soc. 1988, 110,
4126.
(19) Chin, B.; Lough, A. J.; Morris, R. H.; Schweitzer, C. T.;
D’Agostino, C. Inorg. Chem. 1994, 33, 6278.
AUTHOR INFORMATION
Corresponding Author
*E-mail: jagirdar@ipc.iisc.ernet.in.
■
Notes
The authors declare no competing financial interest.
(20) Nishibayashi, Y.; Takei, I.; Hidai, M. Angew. Chem., Int. Ed.
1999, 38, 3047.
(21) Hu, M. G.; Geanangel, R. A. Inorg. Chem. 1979, 12, 3297.
ACKNOWLEDGMENTS
■
(22) Noth, H.; Bayer, H. Chem. Ber. 1960, 93, 2251.
̈
We are grateful to the Department of Science & Technology,
India (Science & Engineering Research Board, India), for
financial support.
(23) Bautista, M. T.; Cappellani, E. P.; Drouin, S. D.; Morris, R. H.;
Schweitzer, C. T.; Sella, A.; Zubkowski, J. J. Am. Chem. Soc. 1991, 113,
4876.
(24) Jasim, N. A.; Perutz, R. N.; Foxon, S. P.; Walton, P. H. J. Chem.
Soc., Dalton Trans. 2001, 1676.
REFERENCES
■
(25) Stephens, F. H.; Baker, R. T.; Matus, M. H.; Grant, D. J.; Dixon,
D. A. Angew. Chem., Int. Ed. 2007, 46, 746.
(26) Stephens, F. H.; Pons, V.; Baker, R. T. Dalton Trans. 2007,
2613.
(27) Atkins, P.; De Paula, J. Atkins’ Physical Chemistry, 8th ed.;
Oxford University Press: Oxford, U.K., 2006.
(28) Mezzetti, A.; Zotto, A. D.; Rigo, P. J. Chem. Soc., Dalton Trans.
1989, 1045.
(1) Labinger, J. A.; Bercaw, J. E. Nature 2002, 417, 507.
(2) Kubas, G. J. Chem. Rev. 2007, 107, 4152.
(3) Kubas, G. J. Metal Dihydrogen and Sigma-Bond Complexes:
Structure, Theory, and Reactivity; Kluwer Academic: New York, 2001.
(4) Kubas, G. J. Adv. Inorg. Chem. 2004, 56, 127.
(5) Sabo-Etienne, S.; Chaudret, B. Coord. Chem. Rev. 2008, 252,
2395.
(6) Corey, J. Y. Chem. Rev. 2011, 111, 863.
(29) Chen, X.; Zhao, J.; Shore, S. G. J. Am. Chem. Soc. 2010, 132,
10658.
(30) Perutz, R. N.; Sabo-Etienne, S. Angew. Chem., Int. Ed. 2007, 46,
2578.
(7) Hall, C.; Perutz, R. N. Chem. Rev. 1996, 96, 3125.
(8) Bercaw, J. E.; Labinger, J. A. Proc. Natl. Acad. Sci. U.S.A. 2007,
104, 6899.
(9) Cowan, A. J.; George, M. W. Coord. Chem. Rev. 2008, 252, 2504.
35
dx.doi.org/10.1021/ic300390s | Inorg. Chem. 2013, 52, 28−36

Inorganic Chemistry
Article
(31) Ledger, A. E. W.; Ellul, C. E.; Mahon, M. F.; Williams, J. M. J.;
Whittlesey, M. K. Chem.Eur. J. 2011, 17, 8704.
(32) (a) Jaska, C. A.; Temple, K.; Lough, A. J.; Manners, I. Chem.
Commun. 2001, 962. (b) Burg, A. B.; Randolph, C. L., Jr. J. Am. Chem.
Soc. 1949, 71, 3451. (c) Haubold, W.; Schaeffer, R. Chem. Ber. 1971,
104, 513. (d) Noth, H.; Vahrenkamp, H. Chem. Ber. 1967, 100, 3353.
̈
(33) Phillips, W. D.; Miller, H. C.; Muetterties, E. L. J. Am. Chem. Soc.
1959, 81, 4496.
36
dx.doi.org/10.1021/ic300390s | Inorg. Chem. 2013, 52, 28−36