Comparison of the catalytic activity of [(η5-C 5H5)Ru(2,2′-bipyridine)(L)]OTf versus [(η5-C5H5)Ru(6,6′-diamino-2,2′- bipyridine)(L)]OTf (L = labile ligand) in the hydrogenation of cyclohexanone. Evidence for the presence of a metal-ligand bifunctional mechanism under acidic conditions

Article abstract of DOI:10.1021/om400871v

The two title complexes as well as the dimeric complex [Ru(II) (η⁵-C₅H₅)(6,6′-diamino-2,2′- bipyridine)]₂(OTf)₂ have been synthesized and characterized by NMR and single-crystal X-ray crystallography. The direct structural comparison of the 2,2′-bipyridine and 6,6′-diamino-2, 2′-bipyridine complexes suggests that the electronic and steric environments of the ruthenium centers in both complexes are essentially equivalent, providing for a unique opportunity to probe the influence of the noncoordinated amine substituent on the relative reactivity and catalytic activity of the complexes. Opposite to what would be anticipated on the basis of steric effects, the bulkier amine-substituted ligand results in a catalyst showing substantially higher activity in the hydrogenation of cyclohexanone in acidic medium, which is attributed to the operation of a metal-ligand bifunctional hydrogenation mechanism mediated by the amine substituents in their protonated form acting as proton shuttles.

Full text of DOI:10.1021/om400871v

Article
pubs.acs.org/Organometallics
Comparison of the Catalytic Activity of [(η5‑C5H5)Ru(2,2′-
bipyridine)(L)]OTf versus [(η5‑C5H5)Ru(6,6′-diamino-2,2′-
bipyridine)(L)]OTf (L = labile ligand) in the Hydrogenation of
Cyclohexanone. Evidence for the Presence of a Metal−Ligand
Bifunctional Mechanism under Acidic Conditions
Domenico DiMondo,† Michelle E. Thibault,† James Britten,‡ and Marcel Schlaf*,†
†The Guelph-Waterloo Centre for Graduate Work in Chemistry, (GWC)2, Department of Chemistry, University of Guelph, Guelph,
Ontario, Canada N1G 2W1
‡Department of Chemistry McMaster University, Hamilton, Ontario, Canada L8S 4M1
S
* Supporting Information
ABSTRACT: The two title complexes as well as the dimeric complex [Ru(II)(η5-C5H5)(6,6′-diamino-2,2′-bipyridine)]2(OTf)2
have been synthesized and characterized by NMR and single-crystal X-ray crystallography. The direct structural comparison of
the 2,2′-bipyridine and 6,6′-diamino-2,2′-bipyridine complexes suggests that the electronic and steric environments of the
ruthenium centers in both complexes are essentially equivalent, providing for a unique opportunity to probe the influence of the
noncoordinated amine substituent on the relative reactivity and catalytic activity of the complexes. Opposite to what would be
anticipated on the basis of steric effects, the bulkier amine-substituted ligand results in a catalyst showing substantially higher
activity in the hydrogenation of cyclohexanone in acidic medium, which is attributed to the operation of a metal−ligand
bifunctional hydrogenation mechanism mediated by the amine substituents in their protonated form acting as proton shuttles.
was the Shvo system.8−16 The mechanism of the Shvo-type
INTRODUCTION
■
systems has been extensively studied and discussed by both
The concept of a metal−ligand bifunctional mediated
heterolytic activation of H2(g) by ruthenium, iridium, and
more recently iron-based complexes has led to the development
of a wide variety of extremely efficient hydrogenation and
Backvall17−21 and Casey,22−32 with the latter extending it to an
̈
analogous iron-based system.33,34 More typical of metal−ligand
bifunctional systems are complexes in which the donor atom
acting as the proton shuttle is either a directly coordinated or
transfer hydrogenation catalysts with wide applicability to C
O and CN bond containing substrates.1−5 This concept, in
pendant amine function such as those developed by
Noyori2,35−38 and Morris39−42 for the asymmetric (transfer-)
which a donor atomtypically but not always an amine
functionalityacts as proton shuttle in a proton (“H+”)
transfer from the donor atom and hydride (“H−”) transfer
from the metal to the unsaturated substrate,6 has also been
termed outer-sphere hydrogenation catalysis.7 A key feature of this
process is thatin contrast to classical hydrogenation catalysts
such as the Wilkinson catalystthe substrate does not directly
coordinate to the metal.
hydrogenation of ketones and a pincer-type complex43 by
Milstein that is active in the hydrogenation of esters.44 Further
examples of and mechanistic studies on these types of metal−
ligand bifunctional catalysts were reported by Gusev,45,46
Saudan,47 and Bergens.48−51 In these three examples it is
further postulated that the coordinated nitrogen donor atom
Arguably the first reported example of a (transfer-)
Received: August 29, 2013
hydrogenation catalyst operating by an outer-sphere mechanism
© XXXX American Chemical Society
A
dx.doi.org/10.1021/om400871v | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 1. Complexes with Noncoordinated Pedant Amines in Their Ligand Framework Capable of Heterolytically Activating
Dihydrogen Gas: (a) Iridium System Reported by Crabtree et al.;55−57 (b) Ruthenium Cyclopentadienyl Complex with a
Tethered Pendant Amine Function Described by Lau54
Chart 1. Structures of Previously Synthesized and Tested Alcohol Deoxygenation Catalysts Based on the η6/η5-Half-Sandwich-
bipy/phen Ruthenium Complex Motif
can act as a noninnocent ligand by switching from an amine to an
amide, for example, either by deprotonation (for the aliphatic
amine ligands) or by dearomatization of the pincer backbone.52
Typically these systems operate under basic or neutral
conditions.53
Two examples of complexes with pendant amine functions
capable of heterolytic activation of dihydrogen are shown in
Scheme 1. While, to our knowledge, the iridium complex
shown on the left has not actually been employed as a
hydrogenation catalyst, it also exhibits cooperative binding of
ketones via the N−H group and the metal, a feature that may
be considered an intermediate or resting state of structurally
similar systems that are active catalysts. The ruthenium
cyclopentadienyl complex shown on the right is active in the
hydrogenation of CO2 to formic acid and was reported by
Lau.54
Further examples of catalytically active complexes with
noncoordinated pendant N−H or O−H functions on the
ligand backbone that operate under basic or base-activated
conditions and for which a metal−ligand bifunctional hydrogen
activation was inferred by spectroscopic, isotope tracing, or
theoretical means were reported by Sola and Oro,58 Thiel,59
Papish,60 Szymczak,61 and Ikariya.62
In the context of our ongoing research program to develop
high-temperature, acid- and water-stable homogeneous hydro-
genation catalysts with potential applications to the selective or
total deoxygenation of biomass-derived sugars, sugar alcohols,
and their condensates,63,64 we recently described the use of the
half-sandwich complexes [(η6-arene)Ru(X)(N∩N)](OTf)n (X
= H2O, H, η6-arene = p-Me-iPr-C6H4, C6Me6, N∩N = 2,2′-
bipyridine (bipy), 1,10-phenanthroline (phen), 6,6′-diamino-
bipy (dabipy), 2,9-diamino-phen (daphen), n = 1, 2) and
[Cp(*)Ru(L)(N∩N)](OTf) (L = H2O or CH3CN, N∩N =
bipy or phen) as catalysts for the deoxygenation of the sugar
alcohol model system 1,2-hexanediol to 1-hexanol and
hexane.65 The structures of the catalysts employed are shown
in the top row of Chart 1. In this study we postulated that,
induced by the operation of a metal−ligand bifunctional
mechanism, the catalysts with the 6,6′-diamino-bipy (dabipy)
or 2,9-diamino-phen (daphen) ligands should exhibit a higher
catalytic activity.
However, the experimental results for the η6-arene systems
clearly established that (a) for these complexes those bearing
the pendant amine function show a lower catalytic activity than
the parent bipy and phen complexes in the hydrogenation of
carbonyl substrates to alcohols and (b) in the complexes [(η6-
arene)Ru(X)(N∩N)](OTf)n (X = H2O, Cl, N∩N = dabipy,
daphen) the amino substituent on the ligand cannot be
protonated in aqueous acidic medium whose acidity isdue to
solvent levelinglimited by that of the acidity of H3O+ (pKa =
−1.74).
Together these observations excluded the operation of a
metal−ligand bifunctional mechanism for these η6-arene-based
systems in aqueous acidic medium with the lower catalytic
activity of the dabipy- and daphen-based catalysts then
attributed to the steric demand of the amino functions on
the ligand, which hinders substrate access to the metal center.
Also these catalysts are deactivated by loss of their arene ligand
at temperatures above 125 °C.
Postulating that the corresponding Cp(*)-based systems
would result in more stable and active catalysts based on the
stronger bond between the metal and anionic Cp(*) ligand, we
subsequently synthesized and tested the complexes shown in
the second row of Chart 1.66 The Cp(*) complexes are stable
B
dx.doi.org/10.1021/om400871v | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
and active up to 200 °C in aqueous acidic medium and effect a
total deoxygenation of glycerol to propene. For the syntheti-
cally more easily accessed Cp* systems, we were however not
able to prepare the complexes with the pendant amine [Ru(η5-
C5Me5)(N∩N)(L)](OTf) (N∩N = dabipy, daphen, L = H2O,
CH3CN). Simple molecular models indicate that this is due to
the steric interaction of the methyl groups of the Cp* ring with
the amine functions of the dabipy or daphen ligand preventing
its coordination, and thus all attempts to isolate a defined
reaction product failed.
(CH3CN)3](PF6) using modified literature protocols.66,67 As
the hexafluorophosphate salt is not hydrolytically stable and
therefore unsuitable for catalysis in aqueous acidic medium as
intended here, the procedure was modified to generate the
−
corresponding triflate (−OTf = CF3SO3 ) complexes [Ru(η5-
C5H5)(bipy)(CH3CN)](OTf) (1) and the new complex
[Ru(η5-C5H5)(dabipy)(CH3CN)](OTf) (2). The synthetic
sequence is summarized in Scheme 2.68 As previously reported
complex 1 was isolated as a crystalline solid in 93% yield, and
its single-crystal X-ray structure obtained.66 The dabipy
1
complex was unexpectedly isolated as a ∼1:1 mixture (by H
We have now developed a synthetic protocol for the
synthesis of the Cp complex [Ru(η5-C5H5)(N∩N)(NCCH3)]-
(OTf) (N∩N = dabipy (2)) shown on the right of Chart 2.
NMR, 70% isolated yield) of the anticipated complex 2 and its
dimer [Ru(η5-C5H5)(dabipy)]2(OTf)2 (3), in which the labile
acetonitrile ligand has been displaced by an intermolecular
interaction with one of the free amine substituents. Thus 2
could not be obtained as a pure solid, but is generated as the
single species (by NMR, vide infra for a detailed discussion of
the NMR spectra of 2 and 3) by dissolving the mixture of 2 and
3 or pure 3 in CH3CN. The isolation of pure dimer [Ru(II)(η5-
C5H5)(dabipy)]2(OTf)2 (3) from the dimer/monomer mixture
was achieved by heating the mixture of 2 or 3 under dynamic
oil pump vacuum (<200 mTorr) at 110 °C for 48 h. Crystals of
3 suitable for single-crystal X-ray analysis were obtained by slow
diffusion of THF into a saturated solution of 2 in acetonitrile,
which together with the above procedure suggests that in the
solid state dimer 3 rather than monomer 2 is the
thermodynamic sink of the system.
Chart 2. Structures of the Two Complexes [Ru(η5-
C5H5)(N−N)(CH3CN)](OTf) [N−N = 2,2′-Bipyridine (1)
and N−N = 6,6′-Diamino-2,2′-bipyridine (2)] Evaluated in
Direct Comparison as Catalysts in This Study
X-ray Crystallographic Data and Structural Comparisons.
Figure 1 shows ORTEPs of [Ru(II)(η5-C5H5)(bipy)-
(CH3CN)](OTf) (1) and [Ru(II)(η5-C5H5)(dabipy)]2(OTf)2
(3).
The availability and close structural and electronic
equivalence of the pair of catalysts 1 and 2 then provide for
what we believe to be a unique and unprecedented opportunity
to test for the effect of a pendant amine functionality on catalyst
activity in the hydrogenation of CO bonds under acidic
conditions. Using cyclohexanone as a model substrate we here
present the comparative evaluation and reactivity study of these
two catalysts that we interpret to provide evidence for the
enhancement of the relative catalytic activity of 2 over 1 by the
action of the pendant amine group as a proton shuttle in a
metal−ligand bifunctional mechanism.
With the carbon atom of the Cp ring that defines the plane
CCp−Ru bisecting the Npy−Ru−Npy angle as a reference point
and denoted C1 in both structures, Table 1 gives the
crystallographic information and directly compares the
characteristic bond length and angles around the ruthenium
centers. The Ru−N distances to the pyridine ring nitrogens in 3
are only slightly longer than in 1, corresponding to a slightly
smaller bond angle N11−Ru−N22; however the symmetry
breakage in 3 caused by the bridging amine substituent (N44)
has no effect; that is, at the 3σ level the bond lengths Ru−N11
and Ru−N22 are still equivalent. Also the distance from the
ruthenium center to the centroid of the Cp ring in the piano-
stool arrangement is the same in both structures. The angle
RESULTS
■
Catalyst Synthesis. The synthesis of the catalysts was
carried out via reaction of the chelates with [Ru(η5-C5H5)-
Scheme 2. Reaction Pathway for the Synthesis of [Ru(II)(η5-C5H5)(N−N)(CH3CN)](OTf) Complexes (N−N = 2,2′-
Bipyridine, 6,6′-Diamino-2,2′-bipyridine)
C
dx.doi.org/10.1021/om400871v | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 1. ORTEPs of [Ru(II)(η5-C5H5)(bipy)(CH3CN)](OTf) (1) and [Ru(η5-C5H5)(dabipy)]2(OTf)2 (3) as determined by single-crystal X-ray
crystallography. Ellipsoids are drawn at the 50% probability level. Hydrogens, the OTf− counterion, and one equivalent of THF contained in the unit
cell of 3 are omitted for clarity.
shielding around the nitrogen through donation of electron
density to ruthenium. On the basis of the similarity in shifts
between the monomer and dimer for the aromatic region in the
spectrum of 2 and its coupling pattern, the triplet at δ = 7.52
ppm was assigned to d. The remaining protons were then
mapped out using 2D COSY and HSQC experiments and
assigned as shown in Figure 2.69
Table 1. Crystallographic Information and Comparison of
the Characteristic Bond Lengths and Angles in [Ru(II)(η5-
C5H5)(bipy)(CH3CN)](OTf) (1) and [Ru(II)(η5-
C5H5)(dabipy)]2(OTf)2 (3)
a
parameter
1
3
d(Ru−N11) [Å]
d(Ru−N22) [Å]
d(Ru−Cpcentroid) [Å]
d(Ru−N6) [Å]
d(Ru−N44) [Å]
2.095(3)
2.094(3)
1.791(3)
2.071(3)
n/a
2.123(5)
2.126(4)
1.790(3)
n/a
Reactivity of [Ru(II)(η5-C5H5)(bipy)(CH3CN)](OTf) (1) with
HOTf Acid. Previous work on the catalytic activity of 1 had
shown that the presence of triflic acid in hydrogenation
reactions was necessary to avoid deactivation and decom-
position of 1.66 This is caused by the coordination of
ethylamine generated by hydrogenation of the acetonitrile
ligand, which is prevented by its protonation in acidic medium.
In the course of the current study (i.e., during the actual
hydrogenation study, vide infra), we observed that the addition
of acid to carbonyl solutions of 1, 2, and 3 further stabilized
these solutions against decomposition in air (as evidenced by a
color change from dark red to brown or black) for extended
periods of time. Therefore reactions of 1 with triflic acid in both
acetonitrile-d3 and acetone-d6 were studied in detail.
2.239(5)
109.4(3)
76.10(2)
n/a
∠(Ru−N44−C21) [deg]
∠(N11−Ru−N22) [deg]
∠(N11−Ru−N6) [deg]
∠(N22−Ru−N6) [deg]
∠(N11−Ru−N44) [deg]
∠(N22−Ru−N44) [deg]
∠(C1−Ru−N11) [deg]
∠(C1−Ru−N22) [deg]
∠(C1−Ru−N6) [deg]
∠(C1−Ru−N44) [deg]
n/a
76.36(10)
87.39(10)
85.76(11)
n/a
n/a
87.9(2)
84.7(2)
103.1(2)
105.2(2)
n/a
n/a
110.33(12)
108.88(13)
158.95(12)
n/a
166.5(2)
As previously reported.66
a
Reactions in Acetonitrile-d3. After addition of 1
1
equivalent of acid, the H NMR spectrum of 1 (Figure 3)
showed the formation of new well-defined species. By
comparison with the initial spectrum of 1 in acetonitrile, the
most noticeable change is the loss of symmetry in the Cp
ligand, suggesting protonation to the neutral η4-cyclopenta-
diene (CpH). However, the 1H NMR of a Ru(η4-C5H6) species
should contain only two alkene peaks due to its symmetry and
two peaks for the respective exo and endo protons of the
Ru−N44−C21 is the ideal tetrahedral angle indicating that the
bridging interaction in the dimer does not suffer from steric
strain. Together these structural parameters suggest that the
steric and electronic environments of the ruthenium center are
very similar if not in fact almost identical in both complexes.
1
NMR Spectra. The H NMR spectra of complex 1 have
been previously reported.66 Figure 2 shows the 1H NMR
spectra of 2 in CD3CN and 3 in the noncoordinating solvent
CD2Cl2, respectively. The acetonitrile complex 2 is quantita-
tively regenerated from 3 by dissolving the latter in CD3CN.
The signals for the Cp protons in 2 appear at 4.36 and at
4.52 ppm in 3, compared to 4.41 ppm in 1. In complex 2 the
peak for the free amine substituent appears at 6.08 ppm. For
complex 3 the peak for the free pendant amine (b) is present at
δ = 6.32 ppm, while the protons on the coordinated amine (i)
are diastereotopic and appear at δ = 8.33 and 8.65 ppm. The
latter are seen at higher chemical shifts due to the loss of
−CH2− group.70 This does not agree with the H NMR
1
spectrum of the new species, which displays four distinct CH
chemical shifts for the CpH ligand. In addition, the 13C NMR
spectra show 15 inequivalent carbons, i.e., total loss of
symmetry in the complex. Analysis of the 13C, 1H, and
HSQC spectra, instead, suggested the Ru(η2-C5H6) species
shown in Figure 3.71
Further evidence for the formation of the [Ru(II)(η2-
C5H6)(bipy)(CH3CN)2](OTf)2 complex comes from a
comparison to the known literature η2-Cp ruthenium complex
D
dx.doi.org/10.1021/om400871v | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 2. 1H NMR spectrum (400 MHz) of 2 formed by dissolving either the 2/3 mixture or 3 in CD3CN (top) and the spectrum of 3 in CD2Cl2
(bottom) as a noncoordinating solvent.
Ru(η2-C5H6)(NH3)5](OTf)2, which shows similar chemical
shifts and coupling constants for the Cp protons, in particular
the diastereotopic CH2 exo/endo protons of the CH2 group of
the protonated ring.72 Once the acid is added to the solution of
1, and the new species Ru(II)(η2-C5H6)(bipy)(CH3CN)2]-
(OTf) is generated, it survives in solution for approximately 6 h
at ambient temperature in air before decomposing with
formation of Ru(0). On the basis of these results, the
established stabilizing effect of added acid when using 1 as
the procatalyst in hydrogenation reactions appears inconsistent,
but can be rationalized by the decomposition pathway
proposed in Scheme 3. Acetonitrile is hydrogenated to
ethylamine, which then can coordinate to the ruthenium
center. This lowers the pKa of the amine proton and positions it
to be transferred to the Cp ring, ultimately causing loss of the
Cp ligand and decomposition of the catalyst. In the presence of
acid the ethylamine generated is protonated to ethyl
ammonium and can therefore not act as a ligand toward the
ruthenium center, thus blocking this pathway under catalysis
reaction conditions.
E
dx.doi.org/10.1021/om400871v | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
1
Figure 3. H NMR spectrum of 1 in acetonitrile-d3 after addition of 1 equivalent of triflic acid.
Scheme 3. Possible Decomposition Pathway of 1 Observed in the Absence of Stabilizing Acid
Reactions in Acetone-d6. Addition of 3 equivalents of
triflic acid to a solution of 1 in acetone-d6 resulted in a
deepening of the red color of the solution and the changes of
which displays 10 separate aromatic carbon peaks for the bipy
ligand of the minor species. The loss of symmetry in the bipy
ligand of 1 must then result from an exchange of the labile
acetonitrile ligand for an asymmetric ligand, which in turn can
originate only from the acetone solvent, but due to its
symmetry, acetone itself cannot be the ligand in question. In
order to elucidate the nature of this ligand, we carried out a
control NMR experiment, which replicated the concentration
of HOTf in acetone-d6 in the absence of 1.73 The 13C NMR
1
the H NMR spectrum appearance shown in Figure 4. On the
basis of comparison with the reference spectrum of 1 (bottom
of Figure 4) the new minor species is a modification of 1 with
loss of symmetry in the bipy but not the Cp ligand.
Further evidence for the loss of symmetry around the bipy
ligand is seen in the 13C jmod spectrum of the acidic solution,
F
dx.doi.org/10.1021/om400871v | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
1
1
Figure 4. (1) H NMR spectrum of 1 in acetone-d6. (2) H NMR spectrum of 1 in acetone-d6 after addition of triflic acid.
spectra from this experiment show the presence of two
additional carbonyl resonances at 210.4 and 199.0 ppm along
with several other peaks that are consistent with the presence of
the acetone condensation products diacetone alcohol and
mesityl oxide, the latter giving a characteristic resonance at 6.12
ppm in the 1H NMR. These resonances are also present in the
13C NMR of the sample corresponding to Figure 4 (spectrum
2, top). Either of these condensation products could act as a
weakly coordinating ligand, explaining both the observed
symmetry breakage in the bipy resonances of 1 and the overall
stabilization of the complex in acetone. The analogous effect
then takes place in the actual substrate cyclohexanone, for
which the corresponding condensates have been directly
observed in the reaction mixture (vide infra: Scheme 4 and
Table 2).
Scheme 4. Metal/Acid-Catalyzed Hydrogenation/
Dehydration/Dimerization Reaction Cascade Leading from
Cyclohexanone to the Products Listed in Table 2
Reactivity of [Ru(II)(η5-C5H5)(dabipy)(CH3CN)](OTf) (2)
with HOTf Acid. In contrast to 1 complexes 2 and 3 do not
require the presence of HOTf to be stabilized against
decomposition in solution under reducing conditions. This
behavior must be due to the presence of the ortho-amine
substituents, which either prevent the coordination of ethyl-
amine by steric repulsion or change the reactivity of the protons
on the coordinating amine function through hydrogen bonding,
which then modifies the overall reactivity such that the
decomposition pathway proposed in Scheme 3 no longer
operates.
The ability of the pendant amine in complex 2 to be
protonated in the reaction solution is a necessary, but not
sufficient condition for it to operate as a metal−ligand
bifunctional catalyst. Addition of 0.5 equivalent of triflic acid
to a solution of complex 2 in acetonitrile-d3 resulted in the
formation of a new species, as revealed by the comparison of
the spectra shown in Figure 5. The spectrum after addition of
acid shows two sets of Cp and aromatic CH peaks, with 2
remaining as the major species and 2-NH3 as the minor
species, with the loss of the amine peak attributed to rapid
proton exchange causing line-broadening into the baseline.
+
G
dx.doi.org/10.1021/om400871v | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
base [A−], an acid titration experiment was carried out tracking
the relative concentrations of the species involved by 1H NMR.
From this study the pKa of the protonated amine was
determined to be ∼0.55, establishing that 2 is in fact protonated
by H3O+ in aqueous acidic medium.75
Table 2. Product Distribution and Mass Balance of the
Cylcohexanone Hydrogenation Reaction Catalyzed by 1 and
3 at t = 10 h
a
[Ru(II)(η5-C5H5)(dabipy)(CH3CN)](OTf) (2) vs [Ru(II)(η5-
C5H5)(dabipy)]2(OTf)2 (3) as the Procatalyst. The fact that
complex 2 cannot be isolated as a pure solid forces the use of
dimer 3 as the procatalyst for the catalytic reactions, meaning
that the two procatalysts (1 and 3) to be compared would differ
in more than only the intended difference, i.e., the presence of
the amine substituents on the dabipy ligand in 2. Establishing
the principal validity of a direct comparison of complexes 1 and
3 as procatalysts therefore required further considerations, in
particular whether or not the dimer 3 is broken up into
monomeric species when dissolved in the carbonyl substrates.
Figure 6 (a, bottom) shows the 1H NMR spectrum of the same
sample of 3 shown in Figure 2 after evaporation to dryness of
the CD2Cl2 solution in vacuo76 and redissolution in acetone-d6
resulting in the formation of several unidentified new
complexes.
As Figure 6b shows, in acetone-d6 2 is quantitatively
regenerated from 3 in situ upon addition of 2 equivalents of
acetonitrile; that is, under these conditions a direct comparison
of 1 and 2 is assured in solution. In order to explore whether
there is any difference between using 3 with and without the
presence of 2 equivalents of acetonitrile as the procatalyst in the
carbonyl hydrogenations, both these reactions were realized
using cyclohexanone as the test substrate.
Comparative Catalytic Hydrogenation of Cyclohexanone.
The relative catalytic activity of the complexes 1, 2 (obtained in
situ from the solid dimer 3 and 2 equivalents of acetonitrile),
and 3 was compared through hydrogenation reactions of neat
cyclohexanone at 175 °C, 900 psi H2(g) pressure (cold), at an
equal concentration of catalysts ([carbonyl substrate]/[ruthe-
nium] = 2000:1) and auxiliary acid (2 equivalents of triflic acid/
ruthenium). The requirement for the presence of acid when
using 1 as the procatalyst also determined the selection of
cyclohexanone, as it is comparatively stable against aldol
condensation reactions in the presence of acid at room
temperature. This is a critical criterion since it allowed for
the preparation of a stable reaction solution without side
reactions of the substrate under ambient conditions. Attempts
to screen other more reactive carbonyl compounds, such as
aliphatic aldehydes, led to immediate condensation reactions
after addition of triflic acid. Decane was added as an internal
standard for the quantitative analysis of the reaction mixture by
GC calibrated against authentic compounds, where readily
available (cyclohexanone, cyclohexanol, cyclohexane). Other
products were identified by GC-MS by matching their M+
peaks and their fragmentation pattern against the MS databases
supplied with the instrument software and quantified by
estimating their FID response factors normalized to their
carbon number by using the effective carbon number concept,77
which resulted in overall mass balances of 97−98%. The
progress of the hydrogenation reactions was monitored through
the pressure drop in the reactor at temperature with pressure
readings taken every hour for the first 7 h and a final reading
after 10 h.
a
For dimeric products normalized to cyclohexanone equivalents.
These results suggest that the pKa of catalyst 2 is in fact high
enough to allow the pendant amine to act as a proton shuttle.
As stated above, this is in contrast to the analogous dicationic
η6-arene systems described by us earlier, whose amine function
could not be protonated under similar reaction conditions.65
The increase in the pKa of 2 stems from the anionic nature of
the Cp ligand, which creates a [Ru]+ complex containing less
positive charge than the structurally analogous [Ru]2+ η6-arene
systems, allowing the protonation of the amine ligand in the Cp
systems. Furthermore, the ability to protonate the pendant
amineagain generating a di- or (by double protonation of
each amine function) possibly a tricationic speciesmay also
contribute to the added stability observed for acidic solutions of
2 in air.
pKa of [Ru(η5-C5H5)(dabipy)(CH3CN)](OTf) in Aqueous
Solution. Triflic acid is the strongest hydrolytically stable
nonoxidizing acid known and is fully dissociated in polar
solvents such as water and DMSO.74 In aqueous solution, the
concentration of H3O+ (pKa = −1.74) formed is therefore equal
to the concentration of triflic acid added; that is, [H3O+] =
[HOTf], and the acid strength is solvent leveled to that of
Figure 7 shows the difference in the rate of H2 uptake
(equivalent to the pressure drop in the reactor) for the
reactions using 1, 2 (generated in situ from 3 by addition of 2
equivalents of acetonitrile), and 3 as the procatalyst for the
H3O+. Treating the protonated species 2-NH3 as a weak acid
[HA], and the unprotonated species 2 as its conjugate weak
+
H
dx.doi.org/10.1021/om400871v | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 5. (1) 1H NMR spectrum of 2 in acetonitrile-d3. (2) 1H NMR spectrum of 2 in acetonitrile-d3 after addition of 0.5 equivalent of triflic acid.
1
Figure 6. H NMR spectra of 3 in acetone-d6 (a) and after addition of 2 equivalents of acetonitrile (b).
I
dx.doi.org/10.1021/om400871v | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 7. Hydrogen pressure drop (in psi) at 175 °C as a function of time in the hydrogenation of cyclohexanone by 1, 3, and 3 + 2 equivalents of
CH3CN.
hydrogenation of cyclohexanone. Table 2 lists the products and
relative percentage yields (by GC/GC-MS) of the observed
products and gives the overall mass balance of the reactions.
The dominant reaction product in all cases is the initial product
cyclohexanol. Due to the acidic nature of the reaction mixture,
caused by both the addition of free triflic acid and the
postulated heterolytic activation of hydrogen gas (see
Discussion section), cyclohexane is also formed by loss of
water followed by hydrogenation; that is, the catalysts are also
capable of hydrogenating carbon−carbon double bonds. At the
same time the direct hydrogenation of cyclohexanone competes
with acid-catalyzed aldol condensation side reactions, leading to
the set of dimeric products listed in Table 2. A comparison of
the relative product yields listed in Table 2 shows that the more
active the catalyst, the fewer the dimeric products that are
formed, which are however also further dehydrated and
hydrogenated. Therefore a maximum amount of the initial
aldol condensation and dehydration product 2-cyclohexenylcy-
clohexanone (26%) remains with the least active catalyst (1),
while 2 and 3 generate more deoxygenated and hydrogenated
products by both the direct cyclohexanone → cyclohexanol →
cyclohexene → cyclohexane route and the indirect route
starting from the aldol condensation product. Scheme 4
illustrates the pathways leading to the set of products listed
in Table 2. The generation of water by this reaction cascade
also means that the strongest acid present in the reaction
mixture is H3O+, which on the basis of the pKa of the amine
substituent is acidic enough to protonate the pendant amine
function in the dabipy complexes.
the procatalyst show a faster rate of hydrogen uptake and
ultimately much higher total consumption of hydrogen than the
reaction catalyzed by 1. Table 3 lists the TON realized for all
Table 3. TON and Relative Initial Rates Achieved by
Catalysts 1, 2 (3 + 2 equiv of CH3CN), and 3
initial rate of H2 uptake as total TON (at t = 10 h) [equiv
H2 uptake
[Δpsi/min]
pressure drop at of H2 consumed/Ru calcd
catalyst
175 °C [psi]
from GC yields]
1
0.97
1.6
336
474
1220
2 (3 + 2
equiv of
CH3CN)
1900 60 ( 7%)
3
3.26
540
1840 60 ( 7%)
three catalysts as defined by the equivalents of hydrogen
consumed (calculated from the amounts of hydrogenated
products quantified by GC) and the relative initial rates of
hydrogen uptake in Δpsi/min.78 The reaction with 2 as the
procatalyst, i.e., 3 in the presence of 2 equivalents of
acetonitrile, is slower and shows a 12% lower pressure drop
of hydrogen (474 vs 540 psi at 10 h), corresponding to a lower
conversion to hydrogenated products, which does however not
correlate with the TON calculated from the GC yields, which in
fact suggests a slightly higher yield of hydrogenated products
for the reaction catalyzed by procatalyst 2. These differences
can be attributed to the approximations made in the
determination of the GC yields of the secondary products
listed in Table 2, which by comparison to the accurately
measured pressure drops ( 0.13% psi by digital pressure
transducer) must have total associated errors of ca. 7%.79 The
final pressure drop for the reaction catalyzed by 1 was 336 psi,
i.e., 29% or 38% less than that for procatalyst 2 or 3,
respectively. The presence of acetonitrile thus does have an
impact on the observed activity of 3; however in both reactions
it still substantially exceeds that of the nonaminated parent
complex 1, and no starting material remains for the two
reactions catalyzed by 2 or 3.
The complexity of the overall hydrogenation/dehydration/
dimerization reaction cascade under acidic conditions, which
results in the parallel and simultaneous operation of multiple
hydrogenation pathways, as well as the requirement to conduct
the reactions in a 316SS autoclave at elevated temperature and
pressure, makes an actual kinetic analysis of the hydrogen
uptake data very challenging. However, taking the rate of
pressure drop and total amounts of hydrogen consumed for all
reactions as a quantitative measure of relative activity, it is
evident from the plots in Figure 7 that both reactions using 3 as
J
dx.doi.org/10.1021/om400871v | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
DISCUSSION
Scheme 5. Proposed Mechanism for the Metal−Ligand
Bifunctional Hydrogenation of Carbonyls by 2-H2
■
In contrast to the dicationic η6-arene 6,6′-diamino-bipyridine
and 2,9-diamino-phenanthroline complexes previously de-
scribed,65 the monocationic η5-Cp systems presented here
can be protonated on their amine function in aqueous acidic
medium and show an enhanced catalytic activity even though
the steric bulk of the amino substituents wouldon the basis
of our observations with the η6-arene systemsdemand the
opposite effect. All other parameters being equal, this therefore
points to a specific role of the amine substituentsin particular
in their protonated formin the rate-determining step of the
hydrogenation reaction. While the experimental conditions
required (T = 175 °C, p(H2) ≥ 900 psi, 316 SS reactor)
prevent a direct observation of metal hydride species in situ or
the catalyst system as a whole in operando, in analogy to the
structurally and electronically similar catalysts [(η6-arene)Ru-
(X)(N∩N)](OTf)n and more specifically [(Cp*Ru(CO)2)2(μ-
H)](OTf) and the related thermally unstable acidic dihydrogen
complex [Cp*Ru(CO)2(η-H2)](OTf)2,80−83 we postulate that
all three procatalysts 1, 2, and 3 in solution form complexes of
the type [CpRu(N∩N)(L)]+ with L being a labile ligand
(substrate, water, product alcohol, etc.) that at elevated
temperature and under hydrogen pressure can be displaced
by dihydrogen to form the corresponding dihydrogen complex
[CpRu(N∩N)(η2-H2)]+ as a reactive intermediate. This
complex then heterolytically activates hydrogen gas to give
the hydride complex [CpRu(N∩N)(H)] and a proton, which
for the reactive intermediates derived from 2 or 3 can be taken
up by triflate, water, solvent, substrate, product, or the pendant
amine substituents to give [CpRu(6-amino-6′-ammonium-2,2′-
bipyridine)(H)]+ (2-H2) as the active catalyst. The observed
enhanced catalytic activity of 2/3 vs 1 then also demands that
the rate-determining step must be associated with an
elementary reaction involving the protonated amine function.
This could be in either the heterolytic activation of dihydrogen,
the transfer of H+/H− to the unsaturated substrate, possibly in a
concerted fashion, or the release of an alkoxide ligand by
protonation from the ammonium substituent. In any of these
scenarios the close spatial proximity of the amine function to
the dihydrogen, hydride, or alkoxide ligand on ruthenium,
respectively, is then key to this phenomenon. In summary, this
suggests the existence of a metal−ligand bifunctional hydro-
genation mechanism as proposed in Scheme 5 as the cause of
the relatively higher catalytic activity of 2/3 vs 1. The higher
activity of 3 vs 2, i.e., without the addition of 2 equivalents of
acetonitrile, which under the reaction conditions are converted
bifunctional hydrogenation catalysts the system described here
operates in acidic rather than basic reaction medium, which we
believe to be the first observation of such an effect under such
conditions. The catalyst tolerates water and comparatively high
temperatures, which may allow the application of this type of
system to the catalytic deoxygenation of sugar alcohol-derived
substrates by iterative acid-catalyzed dehydration and metal-
catalyzed reaction cascades in a single reactor.
EXPERIMENTAL SECTION
■
All manipulations were conducted under an atmosphere of argon using
standard Schlenk-line techniques or in a glovebox, using freshly
distilled solvents or degassed water saturated with argon. All other
reagents were available commercially and were used without further
purification. NMR spectra were recorded on Bruker 300 MHz, 400
MHz, or 600 MHz spectrometers. Spectra were calibrated to the
residual protonated solvent signal, and chemical shifts are reported
relative to tetramethyl silane (=0 ppm). Mass spectroscopic analyses
were performed by the Biological Mass Spectrometry Facility,
University of Guelph, Guelph, ON, Canada. X-ray crystallography
was performed by Jim Britten, Department of Chemistry, McMaster
University, Hamilton, ON, Canada. Industrial-grade hydrogen gas was
used for all hydrogenation experiments. An Autoclave Engineers (AE)
mini-reactor with a 50 mL stainless steel (316SS) reactor vessel fitted
with a Setra C 206 digital pressure transducer (accuracy 0.13%) was
employed for all hydrogenations. Control reactions in the absence of
added catalyst did not give any substrate conversion under the reaction
conditions employed.84 Cyclohexanone was purified by passing it
through a short column of neutral alumina (Brockmann Activity I)
immediately before use as the substrate. GC analyses were performed
on a Varian 3800 using a 30 m × 0.25 mm × 0.025 mm high-polarity
Stabilwax-DA column. Quantitative analysis of the reaction solutions
was performed through a three-level calibration against authentic
samples using decane as an internal standard for cyclohexanone,
cyclohexanol, and cyclohexane. Quantification for all other products
identified in the GC traces was approximated using the GC-FID
response factor for the parent carbonyl.77 GC-MS analyses were
conducted using a Varian 3800 GC/Saturn 2000 MS running EI
default mode using a Stabilwax-DA column.
66
+
to CH3CH2NH3 , is a function of the overall acidity of the
reaction medium, as in this case the added HOTf is not
consumed by the protonation of CH3CH2NH2, but instead
pushes the acid−base equilibrium further toward the
protonation of the pendant amine functions, underlining the
importance of a proton transfer in the catalytic cycle.
CONCLUSION
■
The introduction of pendant amine functions in the ortho-
positions of the bipy ligands in the complex [(η5-C5H5)Ru-
(2,2′-bipyridine)(CH3CN)]OTf to give [(η5-C5H5)Ru(6,6′-
diamino-2,2′-bipyridine)(CH3CN)]OTf results in an enhance-
ment of its catalytic activity in the hydrogenation of
cyclohexanone, which in the catalytic cycle must be rooted in
an active proton-transfer role by the amine functions in their
protonated form. In contrast to the majority of metal−ligand
6,6′-Diamino-2,2′-bipyridine was prepared via 6,6′-dibromo-2,2′-
bipyridine following our own optimized protocol.85
Safety Note and Considerations Regarding the Synthesis of
the 6,6′-Dibromo-2,2′-bipyridine Intermediate. 6,6′-Dibromo-
2,2′-bipyridine was prepared following the literature procedure
K
dx.doi.org/10.1021/om400871v | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 6. Synthesis of 6,6′-Diamino-2,2′-bipyridine
described by Holm and co-workers shown in Scheme 6, which
employs an Ullman-type coupling of 2,6-dibromopyridine by n-BuLi
and CuCl2.86 n-BuLi is present in very slight excess to selectively
lithiate one of the bromide sites and promote generation of a pyridyl-
copper(I) species, which can then undergo the oxidative addition/
reductive elimination needed to form the coupled product.
The primary hazard of this reaction results from the addition of O2
to the reaction. The oxygen treatment was thought to be necessary for
the oxidative coupling of the pyridylcopper(I) species. However, any
excess n-BuLi remaining after lithiation in the presence of the solvent
diethyl ether and oxygen can promote formation of peroxides and/or
hydroxyperoxides. Though the reaction is performed between −50 and
−80 °C to the avoid formation of such byproducts, if peroxides do
form, the reaction becomes shock sensitive, and under an atmosphere
of O2 and ether vapor, the risk of explosion is high. Explosions have
occurred during either the bubbling of O2 or the addition of HCl after
the reaction in both our laboratory and elsewhere.87 It is therefore our
opinion and advice that any future synthesis of 6,6-dibromo-2,2′-
bipyridine should be attempted using only the alternative procedure
laid out by Uchida and Gao (Scheme 7).88,89
Hz, 2H). 13C NMR (150 MHz, acetonitrile-d3): δ 70.8 (CH), 110.4
(CH), 111.9 (CH), 138.9 (CH), 156.8 (C), 162.1 (C). MS (ESI+
TOF): m/e calcd for C15H15N4101Ru 353.3623, found 353.0293
[Cp101Ru(dabipy)]+.
Preparation of Pure [Ru(η5-C5H5)(dabipy)]2(OTf)2 (3) as a Solid. A
0.67 g (1.2 mmol) sample of the mixture was added to a 10 mL
Schlenk flask under an inert atmosphere. The red solid was then
heated to 110 °C under dynamic vacuum for 48 h. The vessel was
cooled to room temperature, resulting in a color change of the solid to
a more yellow tint. Yield: 0.6 g (94%). 1H NMR (600 MHz,
dichloromethane-d2): δ 4.52 (s, 5H), 6.14 (dd, J = 7.8, 3 Hz, 1H), 6.33
(br s, 2H), 6.83 (d, J = 8.4 Hz, 1H), 6.96 (d, J = 7.2 Hz, 1H), 7 (m,
2H), 7.53 (t, J = 7.8 Hz, 1H), 8.33 (d, J = 9 Hz, 1H), 8.64 (d, J = 8.4
Hz, 1H). 13C NMR (600 MHz, dichloromethane-d2): δ 68.62 (CH),
111.19 (CH), 111.93 (CH), 113.61 (CH), 115.61 (CH), 137.69
(CH), 138.37 (CH), 154.19 (C), 155.53 (C), 158.04 (C), 161.11 (C).
Single crystals of complex 3 suitable for X-ray diffraction were grown
from a concentrated acetonitrile solution by slow diffusion of
tetrahydrofuran as the cosolvent.
+
[Ru(η5-C5H5)(C10H8N3(NH3 ))(CH3CN)](OTf)2. A 0.035 g (64 μmol)
portion of 2 was dissolved in 1 mL of acetonitrile-d3. Then 3 μL (32
μmol) of trifluoromethanesulfonic acid was added to the solution and
transferred to a J. Young NMR tube. The solution was shaken for 1
min. The product was obtained in situ as a mixture of the protonated
Scheme 7. Alternative Synthesis of 6,6′-Dibromo-2,2′-
bipyridine
+
1
complex and the starting material. The broad R-NH3 peak in the H
NMR is not present due to rapid exchange of the proton (400 MHz,
acetonitrile-d3): δ 4.26 (s, 5H), 7.19 (d, J = 9.2, 2H), 7.25 (d, J = 7.6,
2H), 7.99 (dd, J = 7.2, 7.6, 2H).
pKa of [Ru(η5-C5H5)(dabipy)(CH3CN)](OTf). A 0.045 (82 μmol)
amount of 2 was dissolved in a 5 mL solution of D2O containing 15%
CD3CN by volume. Then 14 μL (164 μmol) of trifluoromethane-
sulfonic acid was added to the solution and transferred to a J. Young
According to the literature, this reaction also gives 6,6′-dibromo-
2,2′-bipyridine in yields comparable to Holm’s procedure, avoiding use
of oxygen gas as an oxidant. In addition, the procedure does not
involve a workup that requires the reaction solution to be diluted with
hydrochloric acid.
1
NMR tube. The solution was shaken for 1 min, and H NMR spectra
were acquired. The volume of trifluoromethanesulfonic acid in the
Ruthenocene [(η5-C5H5)2Ru] and [Ru(η5-C5H5)(C10H8)][OTf]
were prepared following the procedure by Kundig and Monnier.67
[Ru(η5-C5H5)(CH3CN)3][OTf] was prepared as a dark orange,
viscous oil by modifying their procedure using a dedicated liquid/
liquid extractor. This procedure as well as the synthesis of [Ru(η5-
C5H5)(bipy)(CH3CN)][OTf] (1) was previously described by us.66
[Ru(η5-C5H5)(dabipy)(CH3CN)](OTf) (2). A 100 mL Schlenk flask
was charged with 765 mg (1.75 mmol) of [Ru(η5-C5H5)(CH3CN)3]-
[OTf] and the oil dissolved in 5 mL of degassed CH2Cl2, under inert
conditions. A separate 50 mL Schlenk was charged with 335 mg (1.7
mmol) of 6,6′-diamino-2,2′-bipyridine and the ligand dissolved in 20
mL of degassed CH2Cl2/CH3CN (9:1). The ligand solution was then
added by syringe to the solution of [Ru(η5-C5H5)(CH3CN)3][OTf].
The reaction was heated to reflux for 1 h while stirring. The light red
solution was cooled to room temperature, and the solvent removed to
yield the crude product as a dark red precipitate. The precipitate was
redissolved in cold CH2Cl2 (100 mL), and any unreacted starting
material filtered off. The solvent was removed by dynamic vacuum,
yielding a dark red powder containing a mixture of the monomer (2)
and dimer (3) complexes in approximately 1:1 ratio. Yield: 0.67 g
sample was then increased by an additional of 14 μL, for a total of 28
μL (328 μmol), and the H NMR spectrum was acquired again. This
1
process was repeated with the sample for 10, 20, 30, 40, 50, 60, and 66
molar equivalents of acid with respect to catalyst. At 66 equivalents
+
only the protonated species [Ru(η5-C5H5)(C10H8N3(NH3 ))-
1
(CH3CN)](OTf)2 was present by H NMR. Spectra were calibrated
to the residual protonated solvent signal for acetonitrile-d3. The broad
NH3 peak in the 1H NMR is not present due to rapid exchange of the
proton (300 MHz, acetonitrile-d3): δ 1.95 (s, 1H), 4.22 (s, 5H), 7.08
(dd, J = 12, 12 Hz, 2H), 7.16 (dd, J = 9.6, 10 Hz, 2H), 7.92 (dd, J = 10,
10 Hz, 2H).
[Ru(η2-C5H6)(bipy)(CH3CN)2](OTf)2. A 0.015 g (30 μmol) sample of
1 was dissolved in 1 mL of acetonitrile-d3. Then 8 μL (87 μmol) of
trifluoromethanesulfonic acid was added to the solution and
transferred to a J. Young NMR tube. The solution was shaken for 1
min or until it became brown in color. [Ru(η2-C5H6)(bipy)-
(CD3CN)2](OTf)2 was obtained in situ in quantitative yield as
determined by 1H NMR. 1H NMR (400 MHz, acetonitrile-d3): δ 1.95
(s, 3H), 2.35 (d, J = 24 Hz, 1H), 2.97 (d, J = 24 Hz, 1H), 5.15 (m,
1H), 5.67 (d, J = 4 Hz, 1H), 5.93 (m, 2H, overlay of two doublets),
7.68 (m, 2H, overlay of two triplets), 8.14 (m, 2H, overlay of two
triplets), 8.36 (m, 2H, overlay of two doublets), 8.83 (m, 2H, overlay
of two doublets). 13C NMR (100 MHz, acetonitrile-d3): δ 41.96
(CH2), 94.81 (CH), 103.66 (CH), 124.87 (CH), 125.04 (CH), 128.15
(CH), 128.25 (CH), 132.13 (CH), 139.93 (CH), 140.02 (CH) 140.20
(CH), 154.22 (CH), 154.92 (CH), 157.83 (C), 157.97 (C).
Procedure for the Hydrogenation of Cyclohexanone in a 50 mL
316 SS Autoclave. A 1 mmol amount of decane as the internal
standard (ISTD) was dispensed into a 10.0 mL volumetric flask.
Considering likely volume contractions upon mixing, the volume of
1
(70%). H NMR (400 MHz, dichloromethane-d2): dimer δ 4.52 (s,
5H), 6.14 (dd, J = 6.8, 2.8, 1H), 6.33 (br s, 2H), 6.83 (d, overlay of
two doublets, 1H), 6.96 (d, J = 8 Hz, 1H), 7 (m, 2H), 7.52 (t, overlay
of two triplets, 1H), 8.33 (d, J = 8.4 Hz, 1H), 8.64 (d, J = 9.3 1H);
monomer δ 1.974 (s, 3H), 4.41 (s, 5H), 6 (br s, 4H), 6.81 (d, overlay
of two doublets, 2H), 7.29 (d, J = 7.6 Hz, 2H), 7.57 (t, overlay of two
triplets, 2H). Dissolving this mixture in CD3CN generates the
monomer 2 cleanly as determined by 1H NMR (300 MHz,
acetonitrile-d3): δ 1.95 (s, 3H), 4.36 (s, 5H), 6.08 (br s, 4H), 6.79
(dd, J = 8.1, 1.2 Hz, 2H), 7.37 (dd, J = 7.5, 0.9 Hz, 2H), 7.58 (t, J = 7.7
L
dx.doi.org/10.1021/om400871v | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
the ISTD (195 μL) was ignored for further calculations. The
remaining volume was then filled with cyclohexanone (10 mL =
96.56 mmol). Trifluoromethanesulfonic acid (2 molar equivalents with
respect to catalyst, 8.5 μL) was added, and the solution mixed well.
The catalyst (0.05 mol % of substrate, ∼25 mg) was weighed and
combined with the carbonyl solution in the mini-reactor vessel and
stirred in the sealed reactor for several minutes before removing 0.5
mL for initial GC analysis and resealing the reactor. The reactor was
then evacuated for 2 min using a water aspirator, pressurized to 900 psi
with H2 gas, and allowed to equilibrate for 2 min. The evacuation/
pressurization cycle was repeated twice more. Stirring was set at about
350 rpm, and the reactor was heated to 175 °C. Pressure readings were
taken at 0, 1, 2, 3, 4, 5, 6, 7, and 10 h from reaching the set operating
temperature. After 10 h the reactor heating was turned off and the
reactor placed in an ice bath for 30 min to condense any volatile
products. The reactor was vented and opened, and a final sample taken
for GC/GC-MS analysis.
(18) Ell, A. H.; Johnson, J. B.; Backvall, J.-E. Chem. Commun. 2003,
1652.
(19) Johnson, J. B.; Backvall, J.-E. J. Org. Chem. 2003, 68, 7681.
̈
(20) Samec, J. S. M.; Ell, A. H.; Aberg, J. B.; Privalov, T.; Eriksson, L.;
Backvall, J. E. J. Am. Chem. Soc. 2006, 128, 14293.
(21) Privalov, T.; Samec, J. S. M.; Backvall, J.-E. Organometallics
̈
2007, 26, 2840.
(22) Casey, C. P.; Singer, S. W.; Powell, D. R. Can. J. Chem. 2001, 79,
1002.
(23) Casey, C. P.; Singer, S. W.; Powell, D. R.; Hayashi, R. K.;
Kavana, M. J. Am. Chem. Soc. 2001, 123, 1090.
(24) Casey, C. P.; Vos, T. E.; Singer, S. W.; Guzei, I. A.
Organometallics 2002, 21, 5038.
(25) Casey, C. P.; Johnson, J. B. J. Org. Chem. 2003, 68, 1998.
(26) Casey, C. P.; Bikzhanova, G. A.; Cui, Q.; Guzei, I. A. J. Am.
Chem. Soc. 2005, 127, 14062.
(27) Casey, C. P.; Johnson, J. B. Can. J. Chem. 2005, 83, 1339.
(28) Casey, C. P.; Johnson, J. B. J. Am. Chem. Soc. 2005, 127, 1883.
(29) Casey, C. P.; Johnson, J. B.; Singer, S. W.; Cui, Q. J. Am. Chem.
Soc. 2005, 127, 3100.
(30) Casey, C. P.; Clark, T. B.; Guzei, I. A. J. Am. Chem. Soc. 2007,
129, 11821.
(31) Casey, C. P.; Beetner, S. E.; Johnson, J. B. J. Am. Chem. Soc.
2008, 130, 2285.
(32) Casey, C. P.; Johnson, J. B.; Singer, S. W.; Cui, Q. J. Am. Chem.
ASSOCIATED CONTENT
* Supporting Information
Details of the pKa determination of 2-NH3 . Collection of
■
S
+
compound data sheets with images of NMR and MS spectra.
Details of the reaction of 1 with HOTf in CD3CN and acetone-
d6. This material is available free of charge via the Internet at
http://pubs.acs.org.
Soc. 2008, 130, 1110.
(33) Casey, C. P.; Guan, H. J. Am. Chem. Soc. 2007, 129, 5816.
(34) Casey, C. P.; Guan, H. J. Am. Chem. Soc. 2009, 131, 2499.
(35) Noyori, R. Angew. Chem., Int. Ed. 2002, 41, 2008.
(36) Noyori, R.; Ohkuma, T. Angew. Chem., Int. Ed. 2001, 40, 40.
(37) Hashiguchi, S.; Fujii, A.; Haack, K.-J.; Matsumura, K.; Ikariya,
T.; Noyori, R. Angew. Chem., Int. Ed. Engl. 1997, 36, 288.
(38) Haack, K.-J.; Hashiguchi, S.; Fujii, A.; Ikariya, T.; Noyori, R.
Angew. Chem., Int. Ed. Engl. 1997, 36, 285.
AUTHOR INFORMATION
Corresponding Author
*E-mail: mschlaf@uoguelph.ca.
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
(39) Abdur-Rashid, K.; Faatz, M.; Lough, A. J.; Morris, R. H. J. Am.
Chem. Soc. 2001, 123, 7473.
This work was supported by the Natural Science and
Engineering Research Council of Canada (NSERC), the
Canadian Foundation for Innovation (CFI), Natural Resources
Canada (NRCAN), and the Ontario Ministry of Agriculture
and Food (OMAF). M.E.T. thanks NSERC and the Ontario
Graduate Scholarship program for graduate fellowships.
(40) Abdur-Rashid, K.; Clapham, S. E.; Hadzovic, A.; Harvey, J. N.;
Lough, A. J.; Morris, R. H. J. Am. Chem. Soc. 2002, 124, 15104.
(41) Rautenstrauch, V.; Hoang-Cong, X.; Churlaud, R.; Abdur-
Rashid, K.; Morris, R. H. Chem.Eur. J. 2003, 9, 4954.
(42) Abbel, R.; Abdur-Rashid, K.; Faatz, M.; Hadzovic, A.; Lough, A.
J.; Morris, R. H. J. Am. Chem. Soc. 2005, 127, 1870.
(43) van Koten, G.; Gebbink, R. J. M. K. Dalton Trans. 2011, 40,
8731.
(44) Zhang, J.; Leitus, G.; Ben-David, Y.; Milstein, D. Angew. Chem.,
Int. Ed. 2006, 45, 1113.
(45) Spasyuk, D.; Smith, S.; Gusev, D. G. Angew.Chem., Int. Ed. 2012,
51, 2772.
(46) Spasyuk, D.; Smith, S.; Gusev, D. G. Angew. Chem., Int. Ed.
REFERENCES
■
(1) Noyori, R.; Hashigushi, S. Acc. Chem. Res. 1997, 30, 97.
(2) Noyori, R.; Yamakawa, M.; Hashiguchi, S. J. Org. Chem. 2001, 66,
7931.
(3) Clapham, S. E.; Hadzovic, A.; Morris, R. H. Coord. Chem. Rev.
2004, 248, 2201.
(4) Ito, M.; Ikariya, T. Chem. Commun. 2007, 5134.
(5) Morris, R. H. Chem. Soc. Rev. 2009, 38, 2282.
(6) Crabtree, R. H. New J. Chem. 2011, 35, 18.
(7) Eisenstein, O.; Crabtree, R. H. New J. Chem. 2013, 37, 21.
(8) Shvo, Y.; Blum, Y.; Reshef, D.; Menzin, M. J. Organomet. Chem.
1982, 226, C21.
(9) Blum, Y.; Shvo, Y. Isr. J. Chem. 1984, 24, 144.
(10) Blum, Y.; Shvo, Y. J. Organomet. Chem. 1984, 263, 93.
(11) Blum, Y.; Czarkie, D.; Rahamin, Y.; Shvo, Y. Organometallics
1985, 4, 1459.
2013, 52, 2538.
(47) Saudan, L. A.; Saudan, C. M.; Debieux, C.; Wyss, P. Angew.
Chem., Int. Ed. 2007, 46, 7473.
(48) Hamilton, R. J.; Leong, C. G.; Bigam, G.; Miskolzie, M.;
Bergens, S. H. J. Am. Chem. Soc. 2005, 127, 4152.
(49) Hamilton, R. J.; Bergens, S. H. J. Am. Chem. Soc. 2006, 128,
13700.
(50) Hamilton, R. J.; Bergens, S. H. J. Am. Chem. Soc. 2008, 130,
11979.
(51) John, J. M.; Takebayashi, S.; Dabral, N.; Miskolzie, M.; Bergens,
S. H. J. Am. Chem. Soc. 2013, 135, 8578.
(52) Gunanathan, C.; Milstein, D. Acc. Chem. Res. 2011, 44, 588.
(53) Grotjahn, D. B. Dalton Trans. 2008, 6497.
(54) Chu, H. S.; Lau, C. P.; Wong, K. Y.; Wong, W. T.
Organometallics 1998, 17, 2768.
(55) Lee, D. H.; Patel, B. P.; Clot, E.; Eisenstein, O.; Crabtree, R. H.
Chem. Commun. 1999, 297.
(56) Gruet, K.; Crabtree, R. H.; Lee, D. H.; Liable-Sands, L.;
Rheingold, A. L. Organometallics 2000, 19, 2228.
(12) Blum, Y.; Shvo, Y. J. Organomet. Chem. 1985, 282, C7.
(13) Shvo, Y.; Czarkie, D.; Rahamin, Y. J. Am. Chem. Soc. 1986, 108,
7400.
(14) Abed, M.; Goldberg, I.; Stein, Z.; Shvo, Y. Organometallics 1988,
7, 2054.
(15) Mays, M. J.; Morris, M. J.; Raithby, P. R.; Shvo, Y.; Czarkie, D.
Organometallics 1989, 8, 1162.
(16) Menashe, N.; Shvo, Y. Organometallics 1991, 10, 3885.
́
(17) Csjernyik, G.; Ell, A. H.; Fadini, L.; Pugin, B.; Backvall, J.-E. J.
̈
Org. Chem. 2002, 67, 1657.
M
dx.doi.org/10.1021/om400871v | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(57) Gruet, K.; Clot, E.; Eisenstein, O.; Lee, D. H.; Patel, B.;
Macchioni, A.; Crabtree, R. H. New J. Chem. 2003, 27, 80.
(58) Martin, M.; Sola, E.; Tejero, S.; Andres, J. L.; Oro, L. A.
Chem.Eur. J. 2006, 12, 4043.
(59) Jozak, T.; Zabel, D.; Schubert, A.; Sun, Y.; Thiel, W. R. Eur. J.
Inorg. Chem. 2010, 5135.
(60) Nieto, I.; Livings, M. S.; Sacci, J. B., III; Reuther, L. E.; Zeller,
M.; Papish, E. T. Organometallics 2011, 30, 6339.
(61) Moore, C. M.; Szymczak, N. K. Chem. Commun. 2013, 49, 400.
(62) Watanabe, M.; Kashiwame, Y.; Kuwata, S.; Ikariya, T. Eur. J.
Inorg. Chem. 2012, 504.
(63) Schlaf, M. J. Chem. Soc., Dalton Trans. 2006, 4645.
(64) Taher, D.; Thibault, M. E.; Mondo, D. D.; Jennings, M.; Schlaf,
M. Chem.Eur. J. 2009, 10132.
(65) Dykeman, R. R.; Luska, K. L.; Thibault, M. E.; Jones, M. D.;
Schlaf, M.; Khanfar, M.; Taylor, N. J.; Britten, J. F.; Harrington, L. J.
Mol. Catal. A: Chem. 2007, 277, 233.
(66) Thibault, M. E.; DiMondo, D. V.; Jennings, M.; Abdelnur, P. V.;
Eberlin, M. N.; Schlaf, M. Green Chem. 2011, 13, 357.
(67) Kundig, E. P.; Monnier, F. R. Adv. Synth. Catal. 2004, 346, 901.
(68) A very detailed account of the experiments leading to the
optimized procedures presented here can be found in: Di Mondo, D.
V. Heterogeneous and Homogeneous Hydrogenation Catalysts for the
Selective Deoygenation of Biomass Feedstocks. M.Sc. Thesis,
University of Guelph, 2011.
(69) See Supporting Information for images of the 2D and 13C NMR
spectra and compound data sheets for 1, 2, and 3.
(70) Davies, S. G.; Simpson, S. J.; Felkin, H.; Fillebeen-Khan, T.
Organometallics 1983, 2, 539.
(71) See Supporting Information for images and interpretation of
these spectra.
(72) Spera, M. L.; Chin, R. M.; Winemiller, M. D.; Lopez, K. W.;
Sabat, M.; Harman, W. D. Organometallics 1996, 15, 5447.
1
(73) See Supporting Information for images of spectra the H and
13C jmod and control experiment spectra obtained.
(74) Howells, R. D.; Crown, J. D. M. Chem. Rev. 1977, 77, 69.
(75) The details of the experiment and data analysis can be found in
the Supporting Information.
(76) Note the residual solvent peak at 5.32 ppm.
(77) Scanlon, J. T.; Willis, D. E. J. Chromatogr. Sci. 1985, 23, 333.
(78) On the time scale of the experiments and with the technically
achievable heating rates of the reactor no induction period was
observed for any catalyst.
(79) Calculated by assuming 1840 ≈ 1900 and taking the average of
the measured pressure drops (507 psi) as the reference point.
(80) Chinn, M. S.; Heinekey, D. M.; Payne, N. G.; Selfield, C. D.
Organometallics 1989, 8, 1824.
(81) Fagan, P. J.; Voges, M. H.; Bullock, R. M. Organometallics 2010,
29, 1045.
(82) Schlaf, M.; Ghosh, P.; Fagan, P. J.; Hauptman, E.; Bullock, R. M.
Adv. Synth. Catal. 2009, 351, 789.
(83) Schlaf, M.; Gosh, P.; Fagan, P. J.; Hauptman, E.; Bullock, R. M.
Angew. Chem., Int. Ed. 2001, 40, 3887.
(84) Di Mondo, D.; Ashok, D.; Waldie, F.; Schrier, N.; Morrison, M.;
Schlaf, M. ACS Catal. 2011, 355.
(85) Thibault, M. E.; Luska, K. L.; Schlaf, M. Synthesis 2007, 791.
(86) Parks, J. E.; Wagner, B. E.; Holm, R. H. J. Organomet. Chem.
1973, 56, 53.
(87) Nagata, T. Chem. Eng. News 2000, 78, 2.
(88) Uchida, Y.; Echikawa, N.; Oae, S. Heteroat. Chem. 1994, 5, 409.
(89) Bai, X. L.; Liu, X. D.; Wang, M.; Kang, C. Q.; Gao, L. X.
Synthesis 2005, 458.
N
dx.doi.org/10.1021/om400871v | Organometallics XXXX, XXX, XXX−XXX