Revised Mechanisms for Aldehyde Disproportionation and the Related Reactions of the Shvo Catalyst

Article abstract of DOI:10.1021/acscatal.8b01153

It is widely believed that the Shvo catalyst (1) dissociates to form two active species in solution: the 18-electron hydride RuH(CO)₂[η⁵-C₅(OH)Ph₄] (2) and the naked 16-electron complex Ru(CO)₂[η⁴-C₅(=O)Ph₄] (3). This combined experimental/computational study demonstrates that a sustained presence of 3 is not viable in the reactions of alcohols and organic carbonyls; thus, 3 is better treated as nonexistent under the typical catalytic conditions. We propose a modified view where the key catalytic species are the hydride 2 and the 18-electron metal alkoxide intermediate Ru(OR)(CO)₂[η⁵-C₅(OH)Ph₄] existing in equilibrium with the corresponding alcohol complex. An X-ray crystallographic study of 2 revealed an interesting dihydrogen-bonded dimer structure in the solid state. The mechanistic ideas of this paper explain the highly efficient Tishchenko-like aldehyde disproportionation reaction with the Shvo catalyst. Additionally, our observations explain why 1 is inefficient for hydrogenation of ethyl acetate and for the acceptorless dehydrogenative coupling of ethanol. Our findings provide practical guidance for future catalyst design on the basis of the Shvo ruthenium dimer prototype.

Full text of DOI:10.1021/acscatal.8b01153

Subscriber access provided by Lancaster University Library
Article
Revised mechanisms for aldehyde disproportionation
and the related reactions of the Shvo catalyst
Dmitry G. Gusev, and Denis M. Spasyuk
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.8b01153 • Publication Date (Web): 12 Jun 2018
Downloaded from http://pubs.acs.org on June 13, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
Revised
mechanisms
for
aldehyde
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
disproportionation and the related reactions of the
Shvo catalyst.
a
b
Dmitry G. Gusev,* Denis M. Spasyuk
a
Department of Chemistry and Biochemistry, Wilfrid Laurier University, Waterloo ON N2L 3C5
b
Canada. Canadian Light Source, 44 Innovation Blvd., Saskatoon SK S7N 2V3 Canada
ABSTRACT. It is widely believed that the Shvo catalyst (1) dissociates to form two active
5
species in solution: the 18ꢀelectron hydride RuH(CO) [η ꢀC (OH)Ph ] (2) and a naked 16ꢀ
2
5
4
4
electron complex Ru(CO) [η ꢀC (=O)Ph ] (3). This combined experimental/computational study
2
5
4
demonstrates that a sustained presence of 3 is not viable in the reactions of alcohols and organic
carbonyls, thus 3 is better treated as nonꢀexistent under the typical catalytic conditions. We
propose a modified view (illustrated by the TOC graphic) where the key catalytic species are the
5
hydride 2 and an 18ꢀelectron metal alkoxide intermediate Ru(OR)(CO) [η ꢀC (OH)Ph ] existing
2
5
4
in equilibrium with the corresponding alcohol complex. An Xꢀray crystallographic study of 2
revealed an interesting dihydrogenꢀbonded dimer structure in the solid state. The mechanistic
ideas of this paper explain the highly efficient Tishchenkoꢀlike aldehyde disproportionation
reaction with the Shvo catalyst. Additionally, our observations explain why 1 is inefficient for
hydrogenation of ethyl acetate and for the acceptorless dehydrogenative coupling of ethanol. Our
ACS Paragon Plus Environment
1

ACS Catalysis
Page 2 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
findings provide practical guidance for future catalyst design based on the Shvo ruthenium dimer
prototype.
KEYWORDS: Shvo catalyst, catalytic mechanisms, catalytic hydrogenation, catalytic
dehydrogenation, aldehyde disproportionation.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Introduction
1
ꢀ3
4ꢀ7
The discovery and extensive subsequent investigations of the Shvo catalyst led to the
8ꢀ13
development of the original ideas
of metalꢀligand cooperation in outerꢀsphere catalytic
reactions. These ideas envisaged a process where the catalyst cycles between an 18ꢀelectron
metal hydride and a coordinatively unsaturated (16ꢀelectron) intermediate according to Scheme
1
. Thus, the diruthenium complex 1 is commonly viewed as a precursor to complexes 2 (18ꢀ
electron) and 3 (16ꢀelectron, Scheme 2) under the catalytic conditions, where the 18ꢀelectron
species hydrogenates organic carbonyls (aldehydes and ketones), whereas the 16ꢀelectron species
4
ꢀ7
can act as a dehydrogenation catalyst.
Scheme 1. The original interpretation of the bifunctional outerꢀsphere catalytic hydrogenation
Scheme 2. Formation of 2 and 3 from the Shvo catalyst
ACS Paragon Plus Environment
2

Page 3 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
Complex 2 could not be crystallized; however, the available NMR data are consistent with the
1
4
proposed structure. On the other hand, 3 has never been observed. A dimer of 3, complex 4 of
15
Scheme 3, was isolated and structurally characterized.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 3. The structure of diruthenium complex 4
Viewing 3 as a key catalytic intermediate has significant implications for catalyst design and
for the development of catalytic processes. For example, Menashe and Shvo reported a highly
efficient method for the synthesis of symmetrical esters from aldehydes in the presence of a
1
6
catalytic amount of the corresponding alcohols, according to Scheme 4. In comparison with the
1
7
related Tishchenko reaction, the Shvo process is about two orders of magnitude more efficient
and afforded unprecedented turnover numbers (TON), up to 20,000. It was postulated that the
catalytic amount of alcohol in Scheme 4 was required to produce the hemiacetal intermediate, to
16
be dehydrogenated by 3 according to Scheme 5.
Scheme 4. Catalytic ester synthesis with 1
ACS Paragon Plus Environment
3

ACS Catalysis
Page 4 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 5. The mechanism of ester formation with 1, proposed by Menashe and Shvo
We undertook this present study to get a better understanding of the catalytic properties and
behavior of complexes 1 ꢀ 3, and to gain more information about their speciation in solution in
the presence of several representative species (H , PPh , pyridine, ethanol, acetaldehyde).
2
3
Among the principal results presented and discussed below, we have demonstrated that 1 is
surprisingly efficient for the catalytic ester synthesis of Scheme 4 without added alcohol. We
explain this observation with the help of a modified view of the bifunctional reactions where the
catalyst cycles between the 18ꢀelectron metal hydride 2 and an 18ꢀelectron metal alkoxide
intermediate. We also, for the first time, crystallized and structurally characterized complex 2
that forms an unusual dimer supported by two dihydrogen bonds. Finally, a DFT theoretical
study of complexes 2 and 3 was carried out that provided new mechanistic insights, particularly
into why the Shvo catalyst is inactive for ester hydrogenation and for the acceptorless
dehydrogenative coupling of primary alcohols.
Experimental observations
1
8
Isolation and structure of 2. Complex 2 forms from 1 under H . In THF (12 h at 100 °C,
2
1
13
1
p(H ) = 50 bar), this reaction afforded a clear solution that displayed H and C{ H} NMR
2
spectra consistent with the reported data for 2 (these and other key NMR spectra are provided
ACS Paragon Plus Environment
4

Page 5 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
with the SI). Solvent removal followed by trituration of the residue with hexane afforded a
yellow solid (1.41 g, 77%) of 2 containing traces of THF (7.5 mol%) and hexane (4 mol%). This
material was insoluble in alkanes (hexane, iꢀoctane); in other solvents (CD Cl , toluene, THF,
2
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Et O, EtOAc, EtOH) the solubility of 2 appeared unlimited. NMR spectra of the freshly prepared
2
solutions of 2 under Ar displayed no decomposition.
Slow diffusion of hexane into a solution of 50 mg of 2 in 0.3 mL of toluene produced a
mixture of darkꢀred crystals of 1 and yellow crystals of 2, analyzed by Xꢀray crystallography.
The structure of 1 obtained in our work displays only marginal differences compared with the
1
9
known structure of the Shvo catalyst (see SI for details), whereas complex 2 forms an unusual
C symmetrical dimer where the ruthenium fragments are connected by two intermolecular
i
RuHꢁꢁꢁHO dihydrogen bonds as shown in Figure 1. The intermolecular RuHꢁꢁꢁHO distances are
1
.97 Å in the Xꢀray structure, or 1.75 Å when the crystallographic RuꢀH and OꢀH bond distances
of 1.54 and 0.76 Å are corrected to the more realistic values of 1.61 and 0.97 Å predicted by
2
0
DFT calculations for 2 (vide infra). Other structural parameters of 2 are regular; the principal
bond distances and angles are listed in the caption of Figure 1. It has been known that 2 is
hydrogenꢀbonded in solution, however, the proposed structural interpretation invoked a regular
2
1
intermolecular OHꢁꢁꢁOH hydrogen bond between the molecules.
ACS Paragon Plus Environment
5

ACS Catalysis
Page 6 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. The structure of 2 with the thermal ellipsoids at 50%; most hydrogen atoms are
omitted for clarity. Selected bond distances (Å) and angles (°): Ru1ꢀC1 2.294(2), Ru1ꢀC2
2
1
0
.256(2), Ru1ꢀC3 2.253(2), Ru1ꢀC4 2.266(2), Ru1ꢀC5 2.257(1), Ru1ꢀC30 1.876(2), Ru1ꢀC31
.881(2), Ru1ꢀH1Ru 1.54 (3), C30ꢀO2 1.143(2), C31ꢀO3 1.139(2), C1ꢀO1 1.348(2), O1ꢀH1O
.76(3), C1ꢀC2 1.430(2), C1ꢀC5 1.420(2), C2ꢀC3 1.457(2), C3ꢀC4 1.426(2), C4ꢀC5 1.452(2),
Ru1ꢀX 1.907, Ru1ꢀRu1 6.004(1), XꢀRu1ꢀC30 128.7, XꢀRu1ꢀC31 131.3, XꢀRu1ꢀH1Ru 92.5, C30ꢀ
Ru1ꢀC31 92.5(1), C30ꢀRu1ꢀH1Ru 88(1), C31ꢀRu1ꢀH1Ru 81(1), where X is the centroid of the
C ring. The RuHꢁꢁꢁHO dihydrogen bond distance of 1.75 Å was obtained after changing the
5
crystallographic RuꢀH and OꢀH distances to 1.611 and 0.974 Å, respectively, found in the DFT
optimized structure of 2.
Reactions of 1 and 2. Complex 2 very slowly eliminates H to give 1 in solution, and vice
2
9
,22
versa.
Reaching an equilibrium, 1 + H ⇌ 2 2, takes ca. 3 d at 60 °C under 1 atm H in
2
2
1
tolueneꢀd , regardless of whether starting from 2 or from 1. Integration of the H NMR
8
2
ꢀ
resonances at 4.51 (H ), ꢀ9.42 (2), and ꢀ17.83 ppm (1) gave the equilibrium constant K = [2] [1]
2
1
ꢀ1
[
H ] = 0.57 (from 2) and 0.58 (from 1), and ꢂG = 0.4 kcal/mol (total metal concentration,
2 333
ACS Paragon Plus Environment
6

Page 7 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
[
Ru] = 0.01 M). An alternative calculation using [H ] = p(H ) = 1 atm gave K = 0.0024 (from 1),
2
2
ꢀ1
ꢀ1
0
.0026 molꢁL ꢁatm (from 2), and ꢂG = 4.0 kcal/mol. This latter way of deriving K is more
333
prone to errors since it requires accurate [1], [2], and p(H ) values. However, the difference in
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ꢂG333 (3.6 kcal/mol) seems to be mostly due to the entropy of gaseous H being greater than the
2
ꢀ1 ꢀ1
23
entropy of the dissolved H by 12 calꢁmol ꢁK in toluene (TꢂS = 4 kcal/mol at 333 K). Further
2
measurements of the equilibrium constants gave the data of Figure 2 and the estimates of the
ꢀ1
ꢀ1
reaction enthalpy and entropy: ꢂH = 5.9 ± 0.5 kcal/mol, ꢂS = 16.5 ± 1 calꢁmol ꢁK (per mole of
1
).
1
0
.5
.5
0.0028
0.0029
0.0030
/T
0.0031
1
2
ꢀ
Figure 2. A plot of Rꢁln(K) vs. 1/T for the equilibrium 1 + H ⇌ 2 2. Data used: K = [2] [1]
2
1
ꢀ1
[H ] = 0.36 at 43.5°, 0.58 at 60°, 0.83 at 74.5°, and 1.07 at 85 °C.
2
ACS Paragon Plus Environment
7

ACS Catalysis
Page 8 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
13
Figure 3. 400 MHz H{ C} NMR spectra of the mixture of 1 and 2 (total [Ru] = 0.01 M) fully
equilibrated at 60 °C under 1 atm H in tolueneꢀd with and without added ethanol. Three solvent
2
8
resonances are seen at 6.98, 7.02, and 7.10 ppm.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
ꢀ
The equilibrium 1 + H ⇌ 2 2 is affected by ethanol. The equilibrium constant K = [2] [1]
2
1
ꢀ1
[
H ] increased threefold, from 0.58 to 1.81, at 60 °C with [EtOH]/[Ru] = 22 (0.22 M EtOH in
2
tolueneꢀd , under 1 atm H ). Furthermore, two oꢀCH resonances of 2 shifted in the proton
8
2
spectrum (see Figure 3) as well as the RuH resonance (from ꢀ9.42 to ꢀ9.34 ppm), whereas dimer
1
was little affected. These observations indicate that 2 is stabilized by hydrogen bonding with
5
ethanol, like the related hydride RuH(CO) [η ꢀC (OH)Ph Tol ] that exists as a hydrogenꢀbonded
2
5
2
2
2
1
dimer in tolueneꢀd , however becomes monomeric in this solvent in the presence of ethanol.
8
Most, if not all, reported studies with the Shvo catalyst have been conducted at ≥ 60 °C. To
investigate whether 1 is inert at room temperature, dimer 1 (0.005 M) was reacted with PPh or
3
C D N (0.3 M each) in tolueneꢀd . Slow reactions ensued, e.g. affording (in 3 h) 11% conversion
5
5
8
4
4
of 1 toward Ru(PPh )(CO) [η ꢀC (=O)Ph ] (5) and 15% conversion toward Ru(C D N)(CO) [η ꢀ
3
2
5
4
5
5
2
24
C (=O)Ph ] (6); both products formed along with equal amounts of 2. The known complexes
5
4
1
5,25
26
1
5
and 6 were conveniently monitored by following their distinct oꢀCH H resonances at
3
1
7.80 and 8.10 ppm, respectively. The P resonance of 5 appeared at 38 ppm. These experiments
reveal that the Shvo catalyst 1 is reactive with 2ꢀelectron donors at room temperature. In the next
part, we present further observations of room temperature reactions of 1 with acetaldehyde and
4
ethanol, presumably involving formation of the unstable intermediates Ru(CH CHO)(CO) [η ꢀ
3
2
4
C (=O)Ph ] (7) and Ru(EtOH)(CO) [η ꢀC (=O)Ph ] (8) together with 2.
5
4
2
5
4
ACS Paragon Plus Environment
8

Page 9 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
Catalytic reactions. We first tested reactions of Scheme 6aꢀb: hydrogenation of ethyl acetate
and acceptorless dehydrogenative coupling of ethanol. Only traces of the products were
observed: ca. 2 turnovers of ethyl acetate to ethanol and 12 turnovers of ethanol to ethyl acetate.
We also tested the dehydrogenative coupling of ethanol to ethyl acetate with added acetone,
according to Scheme 6c. This reaction gave 11% ethanol conversion and produced a 1:2 mixture
of ethyl acetate and 2ꢀpropanol as expected. A trace amount of acetaldehyde was also seen as the
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
characteristic quartet at 9.73 ppm (J = 2.8 Hz) in the H NMR spectrum of the reaction solution.
Scheme 6. Catalytic reactions of ethanol and ethyl acetate
1
In related experiments, we established that 2 was stable in neat ethyl acetate and gave H NMR
spectra that did not change even after heating the sample at 80 °C for 20 min, in agreement with
the observed lack of catalytic hydrogenation activity. Also, complex 2 dissolved without
decomposition in neat ethanol at 24 °C. Upon heating to 80 °C, some hydrogen loss from 2 and
precipitation of 1 (insoluble in ethanol) were apparent; however, even after 1 h at 80 °C, ca. 60%
of the initial amount of 2 persisted in solution and no resonances other than those of 2 and
ethanol were evident in the NMR spectra. Our observations indicate that extrusion of H gas
2
from ethanol solutions of 2 is too slow for efficient acceptorless dehydrogenation.
With these observations in hand, we proceeded to the principal part of this investigation: the
Shvo ester synthesis of Scheme 4. The original publication stated that in the presence of 1, the
ACS Paragon Plus Environment
9

ACS Catalysis
Page 10 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
reaction “was found to be an efficient process, provided a catalytic amount of alcohol is present
1
6
in the reaction solution”. Since no data is available on the ester formation without added
alcohol, we decided to investigate a typical process ꢀ the catalytic synthesis of ethyl acetate with
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
and 2 in a 1:1 mixture of anhydrous acetaldehyde and toluene (total 0.2 moles), using the metal
ꢀ5
4
concentration [Ru] = 10 M and the substrateꢀtoꢀruthenium ratio [CH CHO]/[Ru] = 10 .
3
In these experiments, the Shvo catalyst 1 afforded TON = 3995 in 24 h at 60 °C. Complex 1
also produced ethyl acetate at 24 °C, albeit at a slower rate, reaching TON = 2468, 4962, and
7
1
033 in 7, 14, and 21 days, respectively. This room temperature reaction accelerated from TOF =
ꢀ1
ꢀ1
56 d after 24 h, to 242 d in 2 days. The initial rate of ethyl acetate formation was faster with
2 as the catalyst, affording 294 turnovers after 24 h at 24 °C (TON = 5186 after 15 d). Addition
of 5 mol% ethanol (0.3 M) further accelerated the production of ethyl acetate with 1 to give TON
=
4379 in 24 h at room temperature. The recorded values are accurate to ±2%.
A minor byproduct was noted forming along with ethyl acetate in all experiments, e.g., giving
TON = ca. 200 after 2 weeks with both 1 and 2 at 24 °C. While attempting isolation of this
compound, we distilled a fraction containing the byproduct in toluene that could be
unambiguously characterized by NMR spectroscopy. The NMR data is summarized in Scheme 7
2
7ꢀ28
and match that of a known compound, 1ꢀethoxyethyl acetate.
ACS Paragon Plus Environment
10

Page 11 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
Scheme 7. The byproduct of ethyl acetate synthesis from acetaldehyde (NMR data in toluene)
To investigate reasons behind the acceleration of the synthesis of ethyl acetate by added
ethanol we prepared two 0.005 M tolueneꢀd solutions of 1, one with ethanol (0.33 M, EtOH/1 =
8
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
66) and the other in neat tolueneꢀd , as a reference. While the latter produced the expected H
8
NMR spectrum of 1 (δ RuH ꢀ17.83), formation of a mixture of 1 and 2 (δ RuH ꢀ17.87, ꢀ9.34) was
apparent in the sample with added ethanol. Resonances of acetaldehyde were also seen at 9.13
3
1
(q) and 1.42 ppm (d, J = 2.8 Hz). Figure 4 illustrates the H NMR spectra.
1
13
Figure 4. Partial 400 MHz H{ C} NMR spectra of 1 (top trace) in tolueneꢀd and of a mixture
8
of 1 and 2 formed in tolueneꢀd with ethanol after 4 h at 24 °C.
8
The 2/1 ratio increased as 0.011 (0.5 h), 0.027 (1 h), 0.052 (2 h), 0.059 (3 h), 0.063 (4 h), and
further to 0.099 after 3 d at 24 °C. A trace of ethyl acetate formed in the sample after 3 d. Our
observations indicate that an equilibrium established in tolueneꢀd : 1 + EtOH ⇌ 2 2 + CH CHO.
8
3
Estimates of the equilibrium constant after 4 h and 3 d gave ꢂG = 11.0 and 10.1 kcal/mol,
respectively. The latter value might be underestimated because of the secondary changes in
solution – formation of ethyl acetate (an exothermic process, from acetaldehyde).
ACS Paragon Plus Environment
11

ACS Catalysis
Page 12 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
We note a relationship between the formation of 2 from 1 under H , and formation of 2 from 1
2
and ethanol. Whereas molecular hydrogen is directly available in the former reaction, dimer 1
dehydrogenates ethanol in the latter. The Gibbs energy of ethanol dehydrogenation can be
estimated as ꢂG = 10.1 kcal/mol at 25 °C, based on the known enthalpy of reduction of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
29
acetaldehyde (ꢀ19.4 kcal/mol, in triglyme) combined with the standard molar entropies of the
3
0
species. Thus, formation of 2 from 1 and ethanol is more endergonic compared to the reaction
of 1 and H by ca. 10 kcal/mol because of the extra energy needed for the extrusion of H from
2
2
ethanol.
In summary, efficient catalytic conversion of acetaldehyde to ethyl acetate with 1 does not
require added ethanol. The acceleration of the catalytic synthesis of ethyl acetate with added
ethanol appears to be linked with the increase of [2] formed from 1 and ethanol. The lack of
1
5
observable 3 or 4 in Figure 4 (bottom trace) agrees with the previously documented
“instantaneous, quantitative, and selective formation” of 1 upon addition of ethanol to a toluene
solution of 4. In practical terms, 3 is not compatible with ethanol; therefore, assuming a sustained
presence of 3 in catalytic solutions containing alcohols is unrealistic.
DFT computational data
Reactions of complexes 1 - 3. We next turn to DFT methods to learn more on the stability and
reactions of complexes 2 and 3. We first probed the stability of 3 in the presence of acetaldehyde
and ethanol. Considering that the Shvo catalyst is typically used at elevated temperatures, the
reaction Gibbs energies of Chart 1 were calculated at 60 °C. The observations of Chart 1 are
straightforward: there is a strong thermodynamic preference of 3 to dimerize. Formation of the
acetaldehyde and ethanol complexes 7 and 8 is less favorable, however, even the least exoergic
ACS Paragon Plus Environment
12

Page 13 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
ꢀ
reaction, toward 7, has ꢂG = ꢀ9.9 kcal/mol and the associated equilibrium constant K = [7][3]
1
ꢀ1
6
[
acetaldehyde] > 10 . It is not surprising that complex 3 has never been observed.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Chart 1. Calculated MN15ꢀL Gibbs energies of formation of 4, 7, and 8 from 3 at 60 °C in
toluene continuum; all values are per mole of 3
Scheme 8 depicts the mechanism of acetaldehyde conversion to ethyl acetate with complex 2:
30
2
CH CHO → EtOAc, ꢂH° = ꢀ22.7, ꢂG° = ꢀ18.2 kcal/mol.
3
The reverse (counterclockwise) process of Scheme 8 is the ester hydrogenolysis:
30
EtOAc + H (1 atm) → EtOH + CH CHO, ꢂH° = +2.3, ꢂG° = +7.1 kcal/mol.
2
3
Correspondingly, two sets of Gibbs energies are given in the scheme: one for the forward
reaction of ester synthesis and the other (in parentheses) for the reverse reaction. The
interpretation of Scheme 8 is based on the mechanistic ideas proposed by Grützmacher for a
31
rhodium catalyst, and it is guided by the results of our recent work on aldehyde
3
2ꢀ33
disproportionation catalyzed by pincer complexes of osmium.
Related mechanisms were also
3
4ꢀ35
considered by Hasanayn for the Milstein catalyst.
The ester formation of Scheme 8 starts by the familiar concerted H transfer from 2 to
2
acetaldehyde via TS1 to give Int 1, leading to the ethanol complex 8 (ꢂG = ꢀ3.7 kcal/mol). The
related ethoxide Int 2 is less stable than 8 by 8.6 kcal/mol; this preference for Ru(0) in 8 vs.
Ru(II) in Int 2 seems unusual in ruthenium chemistry where Ru(II) compounds abound. The
ethoxide Int 2 further reacts with acetaldehyde in an outerꢀsphere fashion in steps III ꢀ IV to give
ACS Paragon Plus Environment
13

ACS Catalysis
Page 14 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a relatively stable hemiacetal intermediate Int 3. After internal rotation of the hemiacetal, the
final product is formed by H transfer in steps V – VI to regenerate catalyst 2.
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
The Gibbs energies (kcal/mol) are given for the clockwise process, 2 + acetaldehyde, and (in
parentheses) for the reverse reaction: 2 + ethyl acetate. The origin of the energies (in toluene
continuum, at 298.15 K) is 2 with acetaldehyde, or 2 with ethyl acetate, for the forward and the
reverse reactions, respectively; mass balance is ensured throughout.
a
Scheme 8. Ethyl acetate formation/hydrogenolysis with 2.
The bottleneck of the catalytic ester synthesis without added ethanol must be the low
concentration of 2 in solution, most likely produced by the substitution: 1 + acetaldehyde ⇌ 2 +
‡
7
. Formation of ethyl acetate is relatively fast only because of the low barriers ꢂG = 12.1
‡
‡
‡
kcal/mol (ꢂH = 5.9 kcal/mol) and ꢂG = 13.9 kcal/mol (ꢂH = 5.5 kcal/mol) from 2 to Int 1 and
from 8 to TS2, respectively. Another potential problem is the possibility of ethanol displacement
from 8 by acetaldehyde to give 7. This reaction is endergonic (see the energies of Chart 1),
4
however when [acetaldehyde]/[Ru] = 10 , termination of the cycle by ethanol displacement from
ACS Paragon Plus Environment
14

Page 15 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
8
is a certain concern. The observed 28ꢀfold acceleration of the roomꢀtemperature synthesis of
ethyl acetate with 5 mol% EtOH can be rationalized as a combined effect of the increased
concentration of the catalyst 2 produced from 1 with ethanol, plus certain stabilization of 8 by
the added ethanol.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The ideas of Scheme 8 also explain formation of the 1ꢀethoxyethyl acetate byproduct. Int 3 can
rearrange to give a 1ꢀethoxyethoxide Ru(II) intermediate (analogous to the ethoxide Int 2)
followed by an outerꢀsphere addition of acetaldehyde via a transition state similar to TS2 to give
a 1ꢀ(1ꢀethoxyethoxy)ethanol Ru(0) intermediate. Dehydrogenation of the latter should produce
1
ꢀethoxyethyl acetate. Since the principal reaction (toward ethyl acetate) must be fast via the
lowꢀenergy TS3, the lifetime of Int 3 is most likely short, thus the 1ꢀ(1ꢀethoxyethoxy)ethanol
Ru(0) intermediate forms infrequently. Along this line, polymerization of aldehydes might be a
feasible process with metal alkoxide catalysts not capable of efficient hydride abstraction.
The reverse (counterclockwise) process of Scheme 8 is the hydrogenolysis of ethyl acetate to
ethanol and acetaldehyde. This organic reaction is endergonic; therefore, an efficient catalytic
reduction of ethyl acetate to give two equivalents of ethanol requires rapid consumption of the
intermediate acetaldehyde. Acetaldehyde hydrogenation should not be a problem with 2, thus the
reason 2 is inefficient for the reduction of ethyl acetate is most likely because of the slow
formation and/or splitting of the intermediate hemiacetal. The hemiacetal intermediate Int 3 is
not prohibitively high at 12.3 kcal/mol; however, the reaction encounters the highꢀenergy TS2
next, at 28.7 kcal/mol. The alternative elimination of 1ꢀethoxyethanol form Int 3 to give 3 is 27.6
kcal/mol uphill from ethyl acetate with 2.
When analyzing possible reasons for the highꢀenergy TS2 in the hydrogenation of ethyl
acetate, it is worth having a look at the products of reaction step III (from TS2): Int 2 and
ACS Paragon Plus Environment
15

ACS Catalysis
Page 16 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
acetaldehyde. At this point, the reaction path is still at 23.5 kcal/mol and this relatively high
energy is due to the unusual feature of the Shvo system where the Ru(II) ethoxide is unstable. In
contrast, the Noyoriꢀtype complexes (i.e., those possessing a primary or secondary amino group
on ruthenium) form Ru(II) alkoxides; for example, we isolated and structurally characterized a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
36
highly efficient ester hydrogenation catalyst, RuH(OEt)(PPh )[HN(CH CH SEt) ]. One can
3
2
2
2
speculate that if the ethoxide Int 2 had the stability of the ethanol complex 8, then TS2 could
have been stabilized proportionally, down to near 20 kcal/mol, and the ester reduction might
have been possible with the Shvo catalyst.
a
The Gibbs energies are per mole of 2ꢁEtOH, calculated with the MN15ꢀL functional at 298.15
K in ethanol continuum, with H dissolved in ethanol.
2
a
Scheme 9. Acceptorless ethanol dehydrogenation with 2.
It is further instructive to look at the acceptorless dehydrogenation of ethanol of Scheme 9
proceeding by protonation of the hydride by ethanol, followed by H elimination and catalyst
2
regeneration via ethanol dehydrogenation. The calculated barriers are relatively high (22.3 – 24.6
kcal/mol). We know that the reaction of Scheme 9 should be feasible from 8 to 2ꢁEtOH and
acetaldehyde, based on the observation of a successful transferꢀdehydrogenation of ethanol in the
ACS Paragon Plus Environment
16

Page 17 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
‡
presence of acetone. Johnson and Bäckvall reported the rateꢀlimiting barrier ꢂG = 22.2 kcal/mol
3
7
for the dehydrogenation of 1ꢀ(4ꢀfluorophenyl)ethanol with 1 at 70 °C.
The overall organic transformation EtOH → H + CH CHO is endergonic, therefore the
2
3
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
catalytic dehydrogenation requires removal of H from the reaction mixture. A salient feature of
2
Scheme 9 is the absence of a favorable (i.e., accompanied by a negative Gibbs energy)
intermediate. The most stable species on the reaction path (8) is 6.3 kcal/mol above 2ꢁEtOH,
therefore the concentration of H in the reaction solution must be extremely low. It therefore
2
seems that the acceptorless dehydrogenation of ethanol with 2 might be severely hindered due to
slow diffusion of the trace amounts of H from the liquid into the gas phase. In this regard, we
2
may again recall the SNSꢀcatalyst RuCl (PPh )[HN(CH CH SEt) ] that forms
2
3
2
2
2
36
RuH(OEt)(PPh )[HN(CH CH SEt) ] when treated with NaOEt in ethanol. This catalytic
3
2
2
2
36
system gave TON = 8900 for the conversion of ethanol to ethyl acetate in 24 h, under reflux. In
5
conclusion, the instability of the ruthenium ethoxide Ru(OEt)(CO) [η ꢀC (OH)Ph ] (Int 2 of
2
5
4
Scheme 8) seems to be a problem in the catalytic reactions with 1 or 2.
Discussion
Discoveries of the Shvo and Noyori catalysts at the end of the past century inspired the
development of the original ideas of metalꢀligand cooperation in the catalytic outerꢀsphere
38
hydrogenation reactions. As was mentioned in the Introduction, these ideas envisaged a process
where the catalyst cycles between an 18ꢀelectron metal hydride and a coordinatively unsaturated
(16ꢀelectron) intermediate according to Scheme 1. The mechanism of Scheme 1 captured much
attention perhaps not in the least because of the unconventional nature of the reaction where the
substrate does not have to bind to the metal for catalysis. The elegant simplicity of a concerted
transfer of a hydride and a proton in Scheme 1 has been appealing to chemists. Intuitively, it
ACS Paragon Plus Environment
17

ACS Catalysis
Page 18 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
“
makes sense” for the product alcohol to form accompanied by a 16ꢀelectron catalytic
intermediate. The flaw in this thinking is apparent in the case of the Shvo catalyst when we recall
that 3 rapidly reacts with ethanol. Thus, e.g., acetaldehyde hydrogenation cannot produce ethanol
alongside the naked 16ꢀelectron 3.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Since 2005, the ideas of Scheme 1 have undergone gradual revisions for the Noyoriꢀtype
3
9ꢀ43
catalysts possessing a protic NꢀH group on the metal. The experimental studies by Bergens,
4
4
Baratta and coꢀworkers demonstrated that the Noyoriꢀtype hydrogenation catalysts undergo
insertion with ketones in stoichiometric reactions, leading to the corresponding alkoxides. DFT
calculations incorporating solvent effects established that the Noyoriꢀtype catalysts act via
−
+
31,45ꢀ50
discrete H /H transfer steps, via a zwitterionic alkoxide intermediate.
These
computational studies also suggested that the heterolytic splitting of H occurs between the metal
2
4
7ꢀ51
and the oxygen of the alkoxide ion.
Throughout the catalytic reaction, the protic NꢀH group
is engaged via NꢀH···O hydrogen bonding, providing important stabilization to the transition
−
+
states encountered in the H /H transfer steps.
A modern view of the outerꢀsphere catalytic hydrogenation can be summarized with the help
of Scheme 10 (top). Several features of this scheme are distinctly different from the original
interpretation of Scheme 1. The cooperating protic ligand [L]ꢀH is a catalyst accessory that first
facilitates the hydrideꢀtoꢀcarbonyl transfer via hydrogen bonding to the carbonyl oxygen, [L]ꢀ
H···O, then stabilizes the alkoxide during the heterolytic activation of H . The metal alkoxide
2
intermediate of Scheme 10 is in equilibrium with the corresponding alcohol complex. This
equilibrium may be important for the intramolecular rearrangement of the alkoxide ligand, as
3
3
31
discussed in our recent paper and by others. The reverse of the hydrogenation process is the
acceptorless dehydrogenation of an alcohol in Scheme 10 (bottom) that starts by the familiar
ACS Paragon Plus Environment
18

Page 19 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
protonation of the metal hydride by the alcohol, leading to formation and elimination of H2.
When applied to the Shvo catalyst, the ideas of Scheme 10 can be expressed with the help of
Scheme 11. We note two features of the latter that are special: (a) the H transfer is concerted
2
4
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
with the Shvo catalyst and (b) the Ru(II) intermediates are relatively unstable vs. Ru(0).
a
R and R' can be a hydrogen atom or a hydrocarbon group. Dotted lines to the protic (red)
hydrogen atoms indicate hydrogen bonds.
a
Scheme 10. A modern interpretation of the outerꢀsphere catalytic (de)hydrogenation
Ph
Ph
Ph
O
Ph
Ph
Ph
Ph
OH
R'
Ru
H
O
O
H
OC
OC
Ph
Ru
R
H
R
H
H2
OC
OC
R
O
OH
R'
H2
R
R
R
Ph
Ph
Ph
O
O
Ph
Ph
Ph
Ph
Ph
Ph
O
Ph
O
O
H
Ph
Ru
R
R
R'
R'
Ph
Ru
H
Ru
H
O
OC
OC
OC
OC
OC
OC
H
O
H
R
R
R
R
ACS Paragon Plus Environment
19

ACS Catalysis
Page 20 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Scheme 11. (De)Hydrogenation reactions of the Shvo catalyst
Choosing between the mechanisms of Schemes 1 and 10ꢀ11 is not only a problem of theory;
there are practical implications of making the right choice. The importance of viewing catalysts
as cycling between the 18ꢀelectron hydride and the alkoxide structures of Scheme 10/11 becomes
apparent when the scope of the bifunctional outerꢀsphere catalytic reactions is extended to
include the mechanistically related ester hydrogenation and the catalytic coupling of aldehydes
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
2
leading to esters and (with primary amines ) to amides. These reactions can be consistently
rationalized only with the alkoxides as catalytic intermediates, such as in Scheme 8. As we have
proved here, acetaldehyde disproportionation with the Shvo catalyst does not require the free 1ꢀ
ethoxyethanol intermediate in the catalytic solution and the reaction naturally does not need the
1
6ꢀelectron dehydrogenation catalyst 3.
Traditionally, formation of 3 has been viewed as inherent and necessary in the Shvo chemistry.
In consequence, attempts to improve the Shvo catalyst were focused on making the ligands on
ruthenium bulkier, to stabilize and protect the 16ꢀelectron ruthenium through increased sterics
against reverting to the preꢀcatalyst 1 with 2. The observations of Schemes 8 and 9 allow seeing
the fundamental problem of the Shvo catalyst in a completely different light: the Achilles' heel of
the system is the instability of the alkoxide Int 2. If reactions of 2 with carbonyls (or alcohols)
led to the corresponding Ru(II) alkoxides, the formation of 1 would have been avoided. A lesser
vulnerability of the Shvo catalytic system is the thermal instability of 2. This is not a real
problem under H or in the presence of an alcohol but can lead to 1 in the catalytic
2
disproportionation of aldehydes without added alcohol.
Conclusions
ACS Paragon Plus Environment
20

Page 21 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
Chemists generally cannot change the properties and reactivity of molecules at will. Also,
some changes even when possible may be prohibitively expensive for practical catalytic
applications. Therefore, it is an open question as to whether (and how) the Shvo catalyst can be
improved in future work. The objective, however, can be tentatively formulated as follows: it
may be desirable of the nextꢀgeneration Shvo 18ꢀelectron hydride catalyst, when reacted with an
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
alcohol such as ethanol, to give the corresponding 18ꢀelectron Ru(II) alkoxide and H2
accompanied by a reaction Gibbs energy near zero, e.g., within ±2 kcal/mol. This idea is based
on the expectation that an ideal “allꢀinꢀone” catalyst broadly applicable and efficient for
hydrogenations of aldehydes, ketones, and esters, as well as for the dehydrogenative coupling of
alcohols and the coupling reactions of aldehydes to esters and (with primary amines) to amides,
should maintain sufficient operating concentrations of the Ru(II) hydride and alkoxide
intermediates in the reaction solution. It will be appropriate for the search for the next generation
of the Shvo catalyst to be guided by DFT calculations.
Materials and Methods
Experimental details. The Shvo catalyst 1 was obtained from Strem Chemicals. All other
chemicals were purchased from SigmaꢀAldrich. Anhydrousꢀgrade acetaldehyde was reꢀdistilled
under argon and stored at ꢀ28 °C (this material contained 0.13 mol% of trimethyl trioxane
formed at the low temperature during storage). Ethyl acetate and toluene were filtered through
activated basic alumina under argon. Tolueneꢀd and ethanol were stored with 3 Å molecular
8
sieves. The NMR spectra were collected on a Agilent DD2 400 MHz spectrometer. For
1
13
quantitative integration, the H{ C} NMR spectra were obtained using 15° pulses and relaxation
delays of 30 – 60 seconds (depending on the nature of the samples). The choice of the relaxation
1
delay was guided by the H T relaxation times measured under argon in toluene, at room
1
ACS Paragon Plus Environment
21

ACS Catalysis
Page 22 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
temperature as follows: toluene (CH 3.3 s, CH ca. 20 s), ethyl acetate (CH 13.0 s, COCH 5.2 s,
3
2
3
CH 9.8 s), acetaldehyde (CH ca. 50 s, CH ca. 19 s), complex 1 ( RuH 2.2 s, CH 1.6 – 1.7 s),
3
3
complex 2 (RuH 3.9 s, CH 2.8 s).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Preparation of complex 2. 1.8 g of 1 (1.66 mmol) in 29 ml of anhydrous THF was pressurized
to 50 bar H in a 300 mL reactor and heated at 100 °C for 12 h. The reactor was opened in an Ar
2
glovebox to display a clear lightlyꢀcolored solution. THF was evaporated and the residue was
dried under vacuum of an oil pump for 1 h. Addition of 20 mL of hexane with trituration
afforded a yellow solid that was filtered, carefully washed with 3 × 10 mL of hexane while
stirring and pressing the solid against the walls of the fritted funnel with a spatula. The solid was
dried under vacuum for 20 min and stored at ꢀ28 °C. The product was characterized in CD Cl
2
2
(
0.1 M) and displayed ca. 7.5 mol% of THF and 4.1 mol% of hexane (effective M = 552.52 g).
w
1
Yield: 1.42 g (77%). H NMR (400 MHz, CD Cl ): δ 7.49 (m, 4H, CH), 7.32 (m, 6H, CH), 7.06
2
2
– 7.13 (m, 10H, CH), 4.86 (s, 1H, OH), −9.79 (s, RuH). A lowꢀintensity peak (ca. 1.7% of the
1
3
1
hydride of 2) was present at ꢀ10.06 ppm. The nature of this minor species is unknown. C{ H}
NMR (100.46 MHz, CD Cl ): δ 201.8 (CO), 135.7 (C), 133.3 (CH), 132.9 (CH), 131.8 (C),
2
2
1
30.8 (C), 129.0 (CH), 128.5 (CH), 128.1 (CH), 128.0 (CH), 104.4 (C), 91.2 (C).
1
Equilibrium between 1 and 2 was followed by H NMR using solutions containing the total
0
1
.01 M [Ru] in 0.9 mL of tolueneꢀd sealed under H under the normal atmospheric pressure (ca.
8
2
atm H ) in J. Young NMR tubes. The tubes were heated with the solution part fully immersed
2
in an oil bath until the observed 2/1 ratio stopped changing (i.e., stopped increasing from 1 or
decreasing from 2). For the NMR measurements, the samples were taken from the bath for 15 –
2
0 min without disturbing the solutions. It is presumed that liquidꢀtoꢀgas phase H diffusion must
2
be slow at room temperature, and the recorded H concentrations reflect those at the bath
2
ACS Paragon Plus Environment
22

Page 23 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
22
temperatures. Following the approach of Casey and coꢀworkers, the concentration of H in
2
solution was corrected for underꢀintegration of the H resonance due to paraꢀhydrogen by
2
multiplying the integral by 4/3 to give 3.8 (43.5°), 4.1 (60°), 4.3 (74.5°), and 4.6 mM (85 °C)
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
concentration values. For comparison, H concentration of 3.2 mM in toluene was reported at 35
2
23
°
C , lower than 3.6 mM expected based on the trend in this work. Also, the solubility in C H is
6
6
3
.45 mM at 45 °C; thus, it appears that the values determined in this work might be slightly
overestimated, however the origin of the difference is not clear.
Acetaldehyde disproportionation. All reactions were performed under argon, in 20 mL vials
equipped with micro stir bars and tight polypropylene screwꢀon caps, according to the following
procedure. Acetaldehyde was taken from the freezer (at −28 °C) immediately prior to the
preparation of the reaction solutions. First, 5.4 mg of the catalyst was taken in a vial on a
calibrated analytical balance accurate to 0.1 mg. Next, toluene (0.1 moles) was added (followed
by 5 mol% ethanol, when ethanol was used) and the vial was placed on a magnetic stirrer.
Acetaldehyde (0.1 moles) was added next, and the vial was securely sealed. Ca. 0.8 mL samples
of the reaction solution were retrieved from the vial for regular NMR analyses. The TON values
1
13
were determined by integration of the H{ C} NMR spectra, using the toluene ꢀCH resonance
3
as an internal standard.
Hydrogenation. The hydrogenation of ethyl acetate was performed under initial p(H ) = 50 bar
2
using 0.1 moles of the neat substrate (ca. 9.8 mL) in a 300 mL stainlessꢀsteel Parr reactor. The
reactor was loaded and assembled inside of an argon glovebox. The pressurized reactor was
placed into an oil bath preheated to 100 °C on a hotplate stirrer. This temperature was maintained
for 10 h while stirring at 500 rpm.
ACS Paragon Plus Environment
23

ACS Catalysis
Page 24 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Acceptorless dehydrogenative coupling. In an argon glovebox, 11.5 mg of 2 was added into a
5
0 mL Schlenk tube (also containing a micro stirbar) on a calibrated analytical balance accurate
to 0.1 mg. Next, 9.59 g of ethanol was added. After taking the tube out of the box, it was
connected to a vacuum/Ar manifold. Under argon, the stopper was replaced by a finger
condenser connected to a circulating refrigerated bath. When the temperature in the bath reached
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
−
5 °C, the flask was placed in a bath, the argon tank was closed, and the gas released as the bath
temperature increased to 90 °C was allowed to pass freely through a mineral oil bubbler.
Throughout the reaction, the temperature in the cold finger was maintained between −10 and −15
°
C and in the bath at 90 °C while stirring at 1100 rpm.
Crystal structure determination. The singleꢀcrystal Xꢀray diffraction (SCꢀXRD) data for
compounds 1 and 2 were collected using the Canadian Macromolecular Crystallography Facility
5
3
CMCFꢀBM beamline at the Canadian Light Source (CLS). CMCFꢀBM is a bending magnet
beamline equipped with a Si(111) doubleꢀcrystal monochromator, Rayonix MX300HE CCD
detector and MD2 microꢀdiffractometer equipped with Mini Kappa Goniometer Head. The data
for compounds 1 and 2 were collected at 18.000 keV (0.689 Å) and 19.000 keV (0.653 Å)
5
4
respectively, at 100 K. Cell refinement and data reduction were performed using SAINT. An
empirical absorption correction, based on the multiple measurements of equivalent reflections,
55
and merging of data was performed using SADABS. The space groups were confirmed by the
56
57
XPREP routine. The structures of 1 and 2 were solved with SHELXLꢀ2016 using conjugateꢀ
gradient refinement (CGLS) in the initial stages and then by fullꢀmatrix leastꢀsquares and
difference Fourier techniques in the final stages. All nonꢀhydrogen atoms were refined with
anisotropic displacement parameters. The hydrides were extracted from the Fourier difference
maps and refined using a fullꢀmatrix leastꢀsquare method. All other hydrogen atoms were set in
ACS Paragon Plus Environment
24

Page 25 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
the calculated positions and refined as riding atoms with a common thermal parameter. All
5
8
59
publication materials were prepared using LinXTL and Mercury programs. Routine checkCIF
60
and structure factor analyses were performed with Platon.
6
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Computational details. The DFT calculations were carried out with Gaussian 16 (with
62
G09Defaults) using the MN15ꢀL functional. The basis sets used for the geometry optimizations
included Def2QZVP (with def2 ECP) for Ru and Def2SVP for all other atoms (together with the
6
3,64
W06 density fitting basis set).
The polarizable continuum model (IEFPCM) was used in all
without exception calculations (solvent=toluene or ethanol), with the radii and nonꢀelectrostatic
6
5
terms of Truhlar and coꢀworkers’ SMD solvation model (scrf=smd). Every geometry
optimization was accompanied by a frequency calculation, followed by a singleꢀpoint energy
calculation at the MN15ꢀL/Def2QZVP level. The reaction Gibbs energies were calculated by
combining the singleꢀpoint energies with the thermal corrections to enthalpy from the frequency
ꢀ1
ꢀ1
calculations and the computed molar entropies corrected by ꢀ22 calꢁmol ꢁK following the
5
2
approach developed in our recent work. The molar entropy of H in ethanol was assigned the
2
ꢀ1 ꢀ1
ꢀ1 ꢀ1
23
value of S = 17.3 calꢁmol ꢁK at 298.15 K, i.e. 13.9 calꢁmol ꢁK smaller than the calculated
ꢀ1
ꢀ1
gasꢀphase S = 31.2 calꢁmol ꢁK .
AUTHOR INFORMATION
Corresponding Author
*
Eꢀmail: dgoussev@wlu.ca.
Author Contributions
The manuscript was written by D. G. Gusev. Both authors have given approval to the final
version of the manuscript.
ACS Paragon Plus Environment
25

ACS Catalysis
Page 26 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Funding Sources
Natural Sciences and Engineering Research Council (NSERC) of Canada, Discovery grant
program.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ASSOCIATED CONTENT
Supporting Information. The following files are available free of charge.
Crystallographic data for 1 and 2 (CIF file). Representative NMR spectra, computed energies,
IRC plots, and a summary of the crystal data collection and refinement parameters for 1 and 2 (a
PDF file). The supplemental file structures.xyz contains the Cartesian coordinates of the
ruthenium species computed in this study. This file may be opened as a text file to read the
coordinates or opened directly by a molecular modeling program such as Mercury (version 3.3 or
later, http://www.ccdc.cam.ac.uk/pages/Home.aspx) for visualization and analysis.
ACKNOWLEDGMENT
The author is thankful to the Natural Sciences and Engineering Research Council of Canadaꢀ
Discovery grant program, CFI LOF program, SHARCNET, and Wilfrid Laurier University for
support. Insightful comments and suggestions from Prof. Faraj Hasanayn are gratefully
appreciated.
REFERENCES
(1) Blum, Y.; Shvo, Y. Catalytically Reactive Ruthenium Intermediates in the Homogeneous
Oxidation of Alcohols to Esters. Isr. J. Chem. 1984, 24, 144ꢀ148.
ACS Paragon Plus Environment
26

Page 27 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
(
2) Blum, Y.; Czarkie, D.; Rahamim, Y.; Shvo, Y. (Cyclopentadienone)ruthenium Carbonyl
ComplexesꢀA New Class of Homogeneous Hydrogenation Catalysts. Organometallics 1985, 4,
1459ꢀ1461.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(3) Shvo, Y.; Czarkie, D.; Rahamim, Y. New Group of Ruthenium Complexes: Structure and
Catalysis. J. Am. Chem. Soc. 1986, 108, 7400ꢀ7402.
(4) Karvembu, R.; Prabhakaran, R.; Natarajan, K. Shvo’s diruthenium complex: a robust
catalyst. Coord. Chem. Rev. 2005, 249, 911ꢀ918.
(5) Samec, J. S. M.; Bäckvall, J.ꢀE.; Andersson, P. G.; Brandt, P. Mechanistic Aspects of
Transition MetalꢀCatalyzed Hydrogen Transfer Reactions. Chem. Soc. Rev. 2006, 35, 237ꢀ248.
(6) Conley, B. L.; PenningtonꢀBoggio, M. K.; Boz, E.; Williams, T. J. Discovery, Applications,
and Catalytic Mechanisms of Shvo’s Catalyst. Chem. Rev. 2010, 110, 2294ꢀ2312.
(7) Warner, M. C.; Casey, C. P.; Bäckvall, J.ꢀE. Shvo’s Catalyst in Hydrogen Transfer
Reactions. Top. Organomet. Chem. 2011, 37, 85 −125.
(8) Yamakawa, M.; Ito, H.; Noyori, R. The MetalꢀLigand Bifunctional Catalysis: A Theoretical
Study on the Ruthenium(II)ꢀCatalyzed Hydrogen Transfer between Alcohols and Carbonyl
Compounds. J. Am. Chem. Soc. 2000, 122, 1466ꢀ1478.
(9) Casey, C. P.; Singer, S. W.; Powell, D. R.; Hayashi, R. K.; Kavana, M. Hydrogen Transfer
to Carbonyls and Imines from a Hydroxycyclopentadienyl Ruthenium Hydride: Evidence for
Concerted Hydride and Proton Transfer. J. Am. Chem. Soc. 2001, 123, 1090ꢀ1100.
ACS Paragon Plus Environment
27

ACS Catalysis
Page 28 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
10) AbdurꢀRashid, K.; Clapham, S. E.; Hadzovic, A.; Harvey, J. N.; Lough, A. J.; Morris, R.
H. Mechanism of the Hydrogenation of Ketones Catalyzed by transꢀDihydrido(diamine)
Ruthenium(II) Complexes. J. Am. Chem. Soc. 2002, 124, 15104ꢀ15118.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(11) Casey, C. P.; Johnson, J. B. Kinetic Isotope Effect Evidence for a Concerted Hydrogen
Transfer Mechanism in Transfer Hydrogenations Catalyzed by [pꢀ(Me CH)C H Me]
2
6
4
Ru(NHCHPhCHPhNSO C H ꢀpꢀCH ). J. Org. Chem. 2003, 68, 1998ꢀ2001.
2
6
4
3
(12) Sandoval, C. A.; Ohkuma, T.; Muniz, K.; Noyori, R. Mechanism of Asymmetric
Hydrogenation of Ketones Catalyzed by BINAP/1,2ꢀDiamineꢀRuthenium(II) Complexes. J. Am.
Chem. Soc. 2003, 125, 13490ꢀ13503.
(13) Casey, C. P.; Johnson, J. B. Kinetic Isotope Effect Evidence for the Concerted Transfer of
Hydride and Proton from Hydroxycyclopentadienyl Ruthenium Hydride in Solvents of Different
Polarities and Hydrogen Bonding Ability. Can. J. Chem. 2005, 83, 1339ꢀ1346.
2
(14) Washington, J.; McDonald, R.; Takats, J.; Menashe, N.; Reshef, D.; Shvo, Y. Os(CO) (η ꢀ
4
4
C Me )ꢀPromoted Coupling of Alkynes and CO: Formation of (η ꢀC Me R CO)Os(CO) (R =
2
2
4
2
2
3
n
4
Me, Et, Pr) and Catalytic Activity of (η ꢀC R CO)Os(CO) (R = Me, Ph). Organometallics
4
4
3
1
996, 14, 3996ꢀ4003.
(
15) Mays, M. J.; Morris, M. J.; Raithby, P. R.; Shvo, Y.; Czarkie, D. Xꢀray Structure,
Reactivity, and Catalytic Properties of (Cyclopentadienone)ruthenium Dimer,
[(C Ph CO)(CO) Ru] . Organometallics 1989, 8, 1162ꢀ1167.
a
4
4
2
2
(16) Menashe, N.; Shvo, Y. Catalytic Disproportionation of Aldehydes with Ruthenium
Complexes. Organometallics 1991, 10, 3885ꢀ3891.
ACS Paragon Plus Environment
28

Page 29 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
(
17) Törmäkangas, O. P.; Koskinen, A. M. P. The Tishchenko Reaction and Its Modifications
in Organic Synthesis. Recent Res. Dev. Org. Chem. 2001, 5, 225ꢀ255.
4
(18) Blum, Y.; Shvo, Y. Catalytically Reactive (η ꢀtetracyclone)(CO) (H) Ru and Related
2
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Complexes in Dehydrogenation of Alcohols to Esters. J. Organomet. Chem. 1985, 282, C7ꢀC10.
(19) Pasini, T.; Solinas, G.; Zanotti, V.; Albonetti, S.; Cavani, F.; Vaccari, A.; Mazzanti, A.;
Ranieri, S.; Mazzoni, R. Substrate and Product Role in the Shvo's Catalyzed Selective
Hydrogenation of the Platform Bioꢀbased Chemical 5ꢀHydroxymethylfurfural. Dalton Trans.
2014, 43, 10224ꢀ10234.
(20) The coordination geometry around ruthenium in 2 closely resembles the structure of the
5
related chloride complex, RuCl(CO) [η ꢀC (OH)Ph ], see Jung, H. M.; Choi, J. H.; Lee, S. O.;
2
5
4
5
Kim, Y. H.; Park, J. H.; Park, J. Facile Synthesis of (η ꢀPh C COH)(CO) RuCl and Catalytic
4
4
2
Oxidation of Alcohols with Chloroform. Organometallics 2002, 21, 5674ꢀ5677.
(21) Casey, C. P.; Johnson, J. B.; Singer, S. W.; Cui, Q. Hydrogen Elimination from a
Hydroxycyclopentadienyl Ruthenium(II) Hydride: Study of Hydrogen Activation in a Ligandꢀ
Metal Bifunctional Hydrogenation Catalyst. J. Am. Chem. Soc. 2005, 127, 3100ꢀ3109.
(22) Casey, C. P.; Beetner, S. E.; Johnson, J. B. Spectroscopic Determination of Hydrogenation
Rates and Intermediates during Carbonyl Hydrogenation Catalyzed by Shvo’s
Hydroxycyclopentadienyl Diruthenium Hydride Agrees with Kinetic Modeling Based on
Independently Measured Rates of Elementary Reactions. J. Am. Chem. Soc. 2008, 130, 2285ꢀ
2
295.
ACS Paragon Plus Environment
29