Thermodynamic Analysis of Metal-Ligand Cooperativity of PNP Ru Complexes: Implications for CO2 Hydrogenation to Methanol and Catalyst Inhibition

Article abstract of DOI:10.1021/jacs.9b06760

The hydrogenation of CO₂ in the presence of amines to formate, formamides, and methanol (MeOH) is a promising approach to streamlining carbon capture and recycling. To achieve this, understanding how catalyst design impacts selectivity and performance is critical. Herein we describe a thorough thermochemical analysis of the (de)hydrogenation catalyst, (PNP)Ru-Cl (PNP = 2,6-bis(di-tert-butylphosphinomethyl)pyridine; Ru = Ru(CO)(H)) and correlate our findings to catalyst performance. Although this catalyst is known to hydrogenate CO₂ to formate with a mild base, we show that MeOH is produced when using a strong base. Consistent with pK_a measurements, the requirement for a strong base suggests that the deprotonation of a six-coordinate Ru species is integral to the catalytic cycle that produces MeOH. Our studies also indicate that the concentration of MeOH produced is independent of catalyst concentration, consistent with a deactivation pathway that is dependent on methanol concentration, not equivalency. Our temperature-dependent equilibrium studies of the dearomatized congener, (*PNP)Ru, with various H-X species (to give (PNP)Ru-X; X = H, OH, OMe, OCHO, OC(O)NMe₂) reveal that formic acid equilibrium is approximately temperature-independent; relative to H₂, it is more favored at elevated temperatures. We also measure the hydricity of (PNP)Ru-H in THF and show how subsequent coordination of the substrate can impact the apparent hydricity. The implications of this work are broadly applicable to hydrogenation and dehydrogenation catalysis and, in particular, to those that can undergo metal-ligand cooperativity (MLC) at the catalyst. These results serve to benchmark future studies by allowing comparisons to be made among catalysts and will positively impact rational catalyst design.

Full text of DOI:10.1021/jacs.9b06760

Subscriber access provided by Nottingham Trent University
Article
Thermodynamic Analysis of Metal-Ligand Cooperativity
2
of PNP Ru Complexes: Implications for CO
Hydrogenation to Methanol and Catalyst Inhibition
Cheryl L. Mathis, Jackson Geary, Yotam Ardon, Maxwell S. Reese,
Mallory A. Philliber, Ryan T VanderLinden, and Caroline T. Saouma
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.9b06760 • Publication Date (Web): 07 Aug 2019
Downloaded from pubs.acs.org on August 8, 2019
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 14
Journal of the American Chemical Society
1
2
3
4
5
6
Thermodynamic Analysis of Metal-Ligand Cooperativity of
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
PNP Ru Complexes: Implications for CO Hydrogenation to
Methanol and Catalyst Inhibition
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
Cheryl L. Mathis, Jackson Geary, Yotam Ardon, Maxwell S. Reese , Mallory A. Philliber, Ryan T.
VanderLinden and Caroline T. Saouma*
Department of Chemistry, University of Utah, 315 S. 1400 E., Salt Lake City, Utah 84112, U.S.A.
Supporting Information Placeholder
2
ABSTRACT: Hydrogenation of CO in the presence of amines to formate, formamides, and methanol (MeOH) is a promising
approach to streamline carbon capture and recycling. To achieve this, understanding how catalyst design impacts selectivity
and performance is critical. Herein we describe a thorough thermochemical analysis of the (de)hydrogenation catalyst,
(
PNP)Ru-Cl (PNP = 2,6-bis-(di-tert-butylphosphinomethyl)pyridine; Ru = Ru(CO)(H)), and correlate our findings to catalyst
performance. While this catalyst is known to hydrogenate CO to formate with a mild base, we show that MeOH is produced
when using a strong base. Consistent with pK measurements, the requirement for a strong base suggests that deprotonation
2
a
of a 6-coordinate Ru species is integral to the catalytic cycle that produces MeOH. Our studies also indicate that the
concentration of MeOH produced is independent of catalyst concentration, consistent with a deactivation pathway that is
dependent on methanol concentration, not equivalency. Our temperature-dependent equilibrium studies of the dearomatized
congener, (*PNP)Ru, with various H-X species (to give (PNP)Ru-X; X = H, OH, OMe, OCHO, OC(O)NMe
2
) reveal that formic
acid equilibrium is ~ temperature independent; relative to H , it is more favored at elevated temperatures. We also measure
2
the hydricity of (PNP)Ru-H in THF and show how subsequent coordination of the substrate can impact the apparent hydricity.
The implications of this work are broadly applicable to hydrogenation and dehydrogenation catalysis and in particular, those
that can undergo metal-ligand cooperativity (MLC) at the catalyst. These results serve to benchmark future studies by
allowing comparisons to be made amongst catalysts and will positively impact rational catalyst design.
is heterolytically cleaved with ligand protonation
INTRODUCTION
concomitant to hydride installation on Ru. While this
2
a
Rising atmospheric carbon dioxide levels present a
significant incentive to develop new catalysts that are active
mechanism has been proposed by Sanford, the direct
pathway is suggested by Pidko in the analogous PNP-ligated
2c
for the transformation of CO
2
into renewable fuels. Catalysts
Ru system (Figure 1, right pathway). Their calculations
that engage in metal-ligand cooperativity (MLC)1 are
particularly attractive due to their proven ability to
suggest that deprotonation of the bound H in a cationic Ru-
2
H species occurs with the added base and an outer-sphere
2
2
,3,4
hydrogenate CO
2
to formate;
in the presence of amines,
formate. Their analysis moreover suggests that the species
with formate bound to the Ru is an off-cycle species,
contrary to the MLC pathway. These cycles illustrate how
5
,6
7,8
formamides and MeOH are also produced. Formic acid
and MeOH are both attractive to probe as hydrogen storage
9
mediums. However, existing studies often employ high
a
hydricity, pK , and propensity to react with H over other H-
2
pressures/temperatures, strong bases, and precious metals
to promote reactivity. A thermochemical understanding of
existing systems will help guide the rational design of next
X species must be balanced. Related computational studies
suggest that the direct pathway may be more favorable than
those that undergo MLC with both Ru and Ir catalysts,
but the calculations do no always take into account the
reaction conditions. These discrepancies reinforce the need
to experimentally establish the thermodynamic parameters
of such systems to determine their impact on catalysis.
2
c,10
generation catalysts; factors such as pK
a
of the ligand, Keq
with substrates, and ΔGH- can all modulate catalyst
performance.
The strategy of MLC is illustrated in Figure 1, which
1
1
considers the hydrogenation of CO
Milstein’s PNN-ligated Ru catalyst. In this mechanism, H
2
to formate with
For the successful hydrogenation of CO to MeOH, only
2
2a
2
catalysts that can or are speculated to undergo MLC
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 14
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. Mechanistic proposals for the hydrogenation of CO
2
to formate. All Ru species are L
4
X
2
and 18 e . Pertinent thermochemical
parameters are given outside the reaction arrows. (left): MLC mechanism, whereby H
2
addition occurs to the dearomatized
*
(
PNN)Ru. (right): The direct pathway does not undergo ligand deprotonation.
facilitate this transformation in the presence of amines.7-8
This is pertinent to carbon capture and recycling schemes,
Towards advancing catalysts that transfer net H
systems that undergo MLC have been developed.
Comparisons of catalyst performance often are based on
turnover numbers (TON), turnover frequencies (TOF),
and/or reaction times. With no further mechanistic
studies, these parameters do not reveal the underlying
reason(s) as to why the catalysts perform differently, nor do
they correlate these effects with thermochemical
parameters, limiting rational catalyst design.
2
, several
1
c
as current technologies rely on amines to capture CO
2
,
1
2
giving carbamates and carbonates.
However, these
1
c
systems are limited by the energetic demands required for
1
3
CO
amines would streamline carbon-capture and recycling.
Hydrogenation of carbamates (“captured CO ”) is thought
to occur via thermal release of CO , which is first
2 2
release. Thus, hydrogenation of CO in the presence of
2
2
7
b,14
hydrogenated to formate (Scheme 1, (i)).
Condensation
Recently, Prakash and co-workers investigated the role of
the ligand and amine additive for hydrogenation of CO to
2
MeOH. They found a correlation between how electron-
donating the pincer ligand is to the catalyst performance,
and through elegant mechanistic studies, established that
this is due to the relative stability of an off-cycle species.
with an amine then provides formamide, which is further
hydrogenated to MeOH (Scheme 1, (ii)). Not all catalysts
14
that hydrogenate CO
acid to MeOH (in the presence of amines), and generally
2
to formate can hydrogenate formic
7d
5,8b
harsher conditions are required for the latter.
Sometimes, the CO must first be removed and hence the
2
As the catalyst is implicated in each step of the
mechanism, it is important to know the thermodynamic
parameters associated with the catalyst. Hydricity, basicity,
and ease of cleaving an H-X bond (Figure 1) are likely
sensitive to modification of the catalyst. This approach has
been applied to electrocatalytic H
reduction. Expanding this knowledge to hydrogenations,
particularly those that undergo MLC, will help develop
catalysts whereby each step is energetically
thermoneutral, avoiding highly exothermic and subsequent
endothermic steps. It will also shed light on potential
inhibition pathways, which may clarify the correlation of
catalyst loading and TON. The results will be broadly
8
b
overall transformation is necessarily sequential. For Ru, it
is speculated that only hydrogenation of formamide to
MeOH occurs via MLC (Scheme 1, step (ii)), whilst with Mn
both steps are thought to proceed via MLC. Establishing
the mechanism for both hydrogenations as well as the
thermodynamic parameters of catalysts that can and cannot
do this transformation are warranted to advance catalyst
development.
1
4
8
b
15
2
evolution and CO
2
17
16
~
2
applicable to systems that transfer H .
Herein we provide a thorough thermochemical analysis
of the Ru PNP system. This catalyst is chosen for these
studies as it can perform several dehydrogenations and
1
8
19
2c,2d,20
hydrogenations including that of CO
2
to formate,
and
in the presence of amines, MeOH (vide infra). The analysis
presented provides the first hydricity measurements in THF
and the first hydricity measurements for complexes that
react via MLC. This work also considers how coordination
of a 6 ligand impacts the apparent hydricity and pK
Scheme 1. Hydrogenation of (DMA-H)(DMC) to MeOH is
Ph
H
3
mediated by aliphatic ( P NP)Ru-(HBH ) that can undergo
MLC. The hydrogenation is thought to proceed via the species
shown, with the hydrogenation steps labelled (i) and (ii).
21
th
a
of the
2
ACS Paragon Plus Environment

Page 3 of 14
Journal of the American Chemical Society
catalyst, and participation in likely thermodynamic
catalysis via MLC (Figure 1). Equilibrium constants are shown
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
bottlenecks during catalysis. Finally, the thermodynamic
values are correlated to catalytic results. Overall, this study
benchmarks the thermodynamic parameters for
(de)hydrogenation catalysts that operate via MLC. This
allows for future comparisons to be made and
understanding of catalyst limitations.
above the arrow that corresponds to the direction of K. For
clarity, only balanced equations are shown for the reverse
equilibria. As all the congeners feature a hydride and a CO
ligand that remain unchanged, we do not include these ligands
in the shorthand nomenclature (boxed), and simply give the
th
identity of the X-type ligand that occupies the 6 coordination-
site. Deprotonation of the PNP ligand at one of the methylene
positions gives the dearomatized species (indicated by *),
whereby the pyridine L-type donor becomes an anionic LX-
type donor. Formal ligand type and electron count are provided
in the boxes for clarity.
RESULTS
Overview of system considered. The (PNP)Ru species
pertinent to this study are shown in Figure 2 (PNP = 2,6-bis-
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
(
di-tert-butylphosphinomethyl)pyridine). Central to all the
equilibria is the 5-coordinate dearomatized species,
*PNP)Ru (Figure 2, middle top).18a This species reacts with
Synthesis and characterization of complexes.
Towards measuring the equilibria shown in Figure 2,
several pertinent congeners needed to be prepared.
(
Brønsted acidic H-X species to give (PNP)Ru-X (K1,X); in this
2 2
Addition of H , Brønsted acidic H-X or Lewis acidic CO to
reaction proton transfer to the ligand is coupled to
-
1c
(*PNP)Ru readily gives (PNP)Ru-X (Figure 2), as described
nucleophilic attack of X to Ru (Figure 2, bottom left).
These equilibria are pertinent as they are competitive with
addition of H or other species to (*PNP)Ru in the catalytic
2
d,5,24
in the literature (X = H, OMe, OCHO, OH).
noted that in the presence of Lewis or Brønsted acids,
*PNP)Ru is in equilibrium with (PNP)Ru-X, where the
It should be
2
(
cycle (Figure 1).
equilibrium constant depends on the identity of H-X and in
some instances, the concentration (vide infra). Thus, the
various proton sources that may be produced during
16-electron species, [(PNP)Ru]+,22 is
3 2
X
The cationic L
the protonated congener of (*PNP)Ru and is related to
(PNP)Ru-X by loss of the anionic X-type ligand (Figure 2,
bottom). Thus, it is the product of hydride transfer to a
substrate. These three congeners are readily accessible
synthetically, and when X = H, together they are used to
establish the hydricity of (PNP)Ru-H (Figure 2, bottom
left).
catalysis are in competition with H
2
to coordinate the metal
(
i.e., competing K1,X) during catalysis.
-
Deprotonation of (PNP)Ru-X gives [(*PNP)Ru-X] , a
2
a,22-
5 2
formally L X 20-electron species. With few exceptions,
2
3
-
species of the type [(*PNP)Ru-X] are not stable and lose
X-,1c yielding (*PNP)Ru. For example, treatment of
t
(
PNP)Ru-Cl with 1 equiv of KO Bu gives dearomatized
*PNP)Ru.
(
2
3
Dearomatized (*PNP)Ru also reacts with Lewis acids
Figure 3. Solid-state structures (50% displacement ellipsoids)
2
4
such as CO
2
to give (PNP)Ru-CO
2
,
whereby the basic site
(Figure
, bottom right). This species is speculated to be
of a) (PNP)Ru-DMC and b) [K][(*PNP)Ru-CO ]. All calculated
2
on the ligand attacks the Lewis-acidic carbon of CO
2
hydrogen atoms and minor components of disorder are
omitted for clarity. Only one half of the dimeric unit of
2
2a,2d
[K][(*PNP)Ru-CO ] is shown. Select bond distances (Å) for
detrimental to hydrogenation of CO
2
.
The adduct can
2
(PNP)Ru-DMC: Ru1-N1 2.145(1), Ru1-O1 2.205(1), C1-C2
1.503(2), C6-C7 1.495(2). Select bond distances (Å) for
[K][(*PNP)Ru-CO ]: Ru1-N1 2.119(3), Ru1-O1 2.331(3), C1-
2
O1 1.269(5), C1-O2 1.244(5), C1-C2 1.552(6), C2-C3 1.510(6),
C7-C8 1.374(6). Analogous bonds in the other half of the dimer
have similar bond distances.
also be deprotonated to give [K][(*PNP)Ru-CO
2
] (vide
infra).
To probe the initial Ru species formed in catalytic
5
,7,14,25
hydrogenation of amine “captured" CO
2
,
we sought to
prepare a carbamate species. Addition of excess of
dimethylammonium dimethylcarbamate, (DMA-H)(DMC),
to dearomatized (*PNP)Ru cleanly generates (PNP)Ru-
DMC. This carbamate was chosen because (DMA-H)(DMC)
has been shown to be hydrogenated to MeOH using
Ph
H
7b
(
3
P NP)Ru-(HBH ) as a catalyst (Scheme 1). The identity
of (PNP)Ru-DMC as a 6-coordinate aromatized species was
confirmed by NMR spectroscopy and XRD (Figure 3).
Similar to other carboxylate bound species such as
2d
24
(PNP)Ru-OCHO
and (PNP)Ru-CO
2
,
the hydride
Figure 2. Thermodynamic parameters considered in this
study. Species in gold boxes are proposed to be pertinent to
resonance appears as a triplet centered at  = -16.74 ppm
1
(JPH = 19 Hz) in the H NMR spectrum (d
8
-THF). Though the
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
carbamate species (PNP)Ru-DMC is in equilibrium with
Page 4 of 14
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
(*PNP)Ru (vide infra), it is stable to vacuum. This is also
true for the other carboxylate species, (PNP)Ru-OCHO and
(PNP)Ru-CO , but contrasts the instability of (PNP)Ru-H,
2
(PNP)Ru-OH, and (PNP)Ru-OMe, all of which revert to
dearomatized (*PNP)Ru upon concentration in vacuo at
2
6
ambient temperature; this instability limits our ability to
readily isolate these species in the absence of H-X in
solution.
To gain insight into the true pK
PNP)Ru-X, we sought to deprotonate (PNP)Ru-CO
a
of 6-coordinate
. The
(
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
rationale being that the Ru-O bond will be weakened or if
cleaved, will remain formally associated with the Ru (Figure
2
a
2
). This is in contrast to the other (PNP)Ru-X species
-
which lose X and therefore have a ΔG*X- contribution
Figure 2, top left to top middle). Indeed, reaction of a d
THF solution of (PNP)Ru-CO with 1 equiv of KHMDS
KHMDS = potassium hexamethyldisilazide) cleanly gives
K][(*PNP)Ru-CO ]. The NMR spectral parameters for this
(
8
-
2
(
[
2
-
Figure 4. Stacked UV-vis spectra for the equilibria of (*PNP)Ru
(0.60 mM) and formic acid in THF. Concentrations of formic
2
species are similar to that of [(*PNN)Ru-CO ] which was
characterized by Sanford and coworkers (see Figure 1 for
PNN structure and SI for NMR characterization).
2
a
acid given in the legend are those at equilibrium. (inset): Plot
[
(퐏퐍퐏)퐑퐮 ― 퐎퐂퐇퐎]
[( ∗ 퐏퐍퐏)퐑퐮]
of
vs. [H-OCHO]. The slope gives K1,OCHO (eq 1).
To determine if the CO
or not, crystals of [K][(*PNP)Ru-CO
2
oxygen remained bound to the Ru
] were grown by vapor
2
Equilibria with H-X species. Dearomatized (*PNP)Ru
readily reacts with Brønsted and Lewis acids to give
PNP)Ru-X (Figure 2, K1,X and K , respectively). However,
given the importance of its reaction with H for
diffusion of diethyl ether into THF, and the solid-state
structure is shown in Figure 3. In the solid-state,
(
2
[
2 2 2
K][(*PNP)Ru-CO ] is a dimeric species bridged by a K O
2
diamond core (see SI). The loss of aromaticity is evident
from the C7-C8 bond distance of 1.374(6) Å. By contrast, C2-
C3 is 1.510(6) Å, which is similar to that found in the neutral
hydrogenation reactions, and the potential inhibitory effect
of this reaction with other species that may be present
during catalysis, we sought to establish the equilibrium
2
4
analogue (1.507(2) Å). Upon deprotonation, the N1-Ru1
bond distance remains unchanged (Figure 3 caption). This
contrasts with deprotonation of the related aliphatic PNP
ligands (Scheme 1), whereby the amide-Ru bond distance is
noticeably shorter (~0.18 Å) than the amine-Ru bond
constants with H
hydrogenation of CO
X species that may be present during catalysis: MeOH, H
2
, H-X (K1,X), and CO
prompted analysis of the following H-
O,
2 2
(K ). Our interest in
2
2
HC(O)OH, DMF, diethylamine, and [DMA-H][DMC]. Neither
DMF nor diethylamine reacts with (*PNP)Ru to an
observable extent (see Figures S39, S35, respectively); all
other species react with (*PNP)Ru to give (PNP)Ru-X. The
equilibrium is readily monitored by UV-vis spectroscopy by
observing the absorbance at 595 nm, which corresponds to
dearomatized (*PNP)Ru (Figure 4). Briefly, addition of
various concentrations of H-X to a THF solution of
(*PNP)Ru allows for the equilibrium constant (eq 1) to be
established.
27
distance. Notably, CO
2
still coordinates to Ru with a Ru1-
O1 bond distance of 2.331(3) Å. This represents a ~ 0.05 Å
2
4
elongation from neutral (PNP)Ru-CO
2
and a ~ 0.07 – 0.1
Å
elongation from other structurally characterized
2
d
carboxylate and carbonate derivatives.
The observed
elongation is consistent with the Ru center now being
formally 20-electron.
[(퐏퐍퐏)퐑퐮–퐗]
(
퐾1,푋
)
=
( ∗ 퐏퐍퐏)퐑퐮
[
(
1)
]
∙ [H–X]
Equation 1 represents the binding of H-X to the Ru and the
stoichiometry suggests that there should be
a
concentration-dependence to the equilibrium constant.
This is confirmed by diluting equilibrium samples and
noting the shift in the ratio of the two Ru species (see SI).
Apart from (DMA-H)(DMC) (eq 2-3), this concentration-
dependence was observed for all H-X species.
NMe2
t
O
O
Ru
H
P Bu2
O
K1,DMC
PtBu2
N
Ru
H
C
+ (DMA-H)(DMC)
N
C
O
+
2
Me NH
P
H
P
t
t
Bu2
Bu2
(2)
H
[
(퐏퐍퐏)퐑퐮–퐃퐌퐂] ∙ [DMA]
(
3) 퐾1,퐷푀퐶 = [( ∗
퐏퐍퐏)퐑퐮
]
∙ [(DMA–H)(DMC)]
4
ACS Paragon Plus Environment

Page 5 of 14
Journal of the American Chemical Society
The lack of concentration-dependence for the
accurately measure, no (*PNP)Ru is observed at room
temperature for the equilibria with H or CO ; the
equilibrium constants are therefore extrapolated from
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
equilibrium with (DMA-H)(DMC) is rationalized on the
basis that one equivalent of DMA is formed and the (DMA-
H)(DMC) substrate is strongly ion-paired in solution.
2
2
higher temperature data (vide infra).
[
(퐏퐍퐏)퐑퐮–퐇]
퐾1,퐻
)
=
( ∗ PNP)퐑퐮
[
Similarly, the equilibrium constant for the binding of H
and CO is measured analogously (eqs and 5,
2
(
(
(
(
4)
5)
]
∙ pH2
2
4
[
(퐏퐍퐏)퐑퐮–퐂퐎ퟐ]
퐾2
)
=
( ∗ 퐏퐍퐏)퐑퐮
[
respectively), with the gas pressures given in atm. These
measurements necessitate knowledge of the equilibrium
partial pressure due to the gas, and hence require a large
head-space such that minimal changes occur during
equilibration (see SI for details). At pressures that we can
]
∙ pCO2
From the equilibrium constants, the free energy changes
can be obtained. To note, the equilibria that are
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
Table 1. Equilibrium constants and free energy change associated with coordination of H-X and CO to (*PNP)Ru.
퐨
퐨
퐨
H-X
횫퐆
횫퐇
횫퐒
ퟏ,퐗
K
1,X
ퟏ,퐗
ퟏ,퐗
kcal∙mol-1)
(kcal∙mol-1)
(cal∙mol ∙K )
-1 -1
(
293 K
428 K
428 K
(1 atm, 1 M)
(1 atm, 1 M)
(50 bar, 1 mM)
H-H
(DMA-H)(DMC)
H-OCHO
-4.1 ± 0.2h
-0.03 ± 2f,h
-3.5 ± 0.2h
-2.0 ± 1.9h
-2.0 ± 0.6h
-3.8 ± 0.8c,h
-17.4 ± 0.2
-12.6 ± 1.5f
0.06 ± 0.15
-3.5 ± 1.3
-45 ± 0.5
-42 ± 5f
1100 ± 1.4a,d,g
0.13a,d
6.6a,d
1.3 ± 0.2
N.A.e
N.A.e
355 ± 38b
370b,d
0.37b,d
12 ± 0.5
-5.2 ± 4.6
8.7 ± 1.4
-33 ± 2c
25 ± 5b
4.6b,d
4.6 × 10-3 b,d
H-OMe
H-OH
-0.6 ± 0.4
-13.5 ± 0.6c
5.4 ± 0.5b
41b,d
4.1 × 10-2 b,d
c
633 ± 4a,c,d,g
0.55 a,c,d
27.6 a,c,d
CO
2
a
-1
b
-1
c
d
e
Units of atm . Units of M . Equilibrium with CO
2
corresponds to K
2
. Extrapolated from temperature-dependent data. Not
f
applicable, due to limited stability of DMC-H and (DMA-H)(DMC) at elevated temperatures. See text for details. Energies obtained
표
from a limited temperature-range (20-50 C). Error from propagation of the error of Δ퐺 , 휎퐾 = exp (휎Δ퐺 푅푇
o
g
hError from
)
.
표
표
2
2
propagation of the error of Δ퐻 and Δ푆 , 휎Δ퐺 =
Δ푆 . Error on the entropy and enthalpy are from the error in
(휎Δ퐻) + (|푇|휎 )
intercept and slope, respectively.
concentration dependent will also have a concentration-
dependent free energy (eq 6; standard-conditions are 1 M
and 1 atm).
a dearomatized Ru species.28 The study considered the
reaction of various Lewis acids in the related PNN system,
all of which bind akin to CO . While different equilibrium
2
constants were obtained, concentration-dependence is not
provided.
[
(PNP)Ru–X]
(
6a)
Δ퐺1,푋 = ―RT ∙ ln([
( ∗ 퐏퐍퐏)퐑퐮 ∙ [H–X]o
∙ [H–X]
)
]
표
=
Δ퐺 ― RT ∙ ln ([H–X])
The free energy associated with addition of H is similar
1
,푋
2
[
(PNP)Ru–H]
to that observed by others whereby H adds to give a metal
2
(
6b)
Δ퐺1,퐻 = ―RT ∙ ln([
o
∙ pH2
( ∗ 퐏퐍퐏)퐑퐮
∙ p H
]
2
)
29
dihydride. It is smaller by ~ 2 kcal/mol than that found for
a series of (P N )Ni species, whereby H is heterolytically
2 2 2
표
30
=
Δ퐺 ― RT ∙ ln
(
pH2
)
1
,퐻
cleaved, protonating the ligand with hydride attack at the
Ni.
The free energy for the equilibrium with CO
2
is analogous
2
to that for H given in eq 6b. Thus, under non-standard
Given that catalysis ensues at elevated temperatures, the
temperature-dependence of the equilibria was also
determined. An overlay of the van’t Hoff plots is shown in
Figure 5 and the enthalpy and entropy are given in Table 1.
During these studies, we found that (PNP)Ru-OMe and
states, the equilibrium constant is given by eq 7.
표
―
∆퐺1,푋
(7) 퐾1,푋 = [H–X]exp
(
)
RT
From equations 6 and 7, increasing the concentration of
H-X at equilibrium will favor formation of (PNP)Ru-X.
Equilibrium constants and equilibrium energies are given in
Table 1.
(
PNP)Ru-OH are not stable at elevated temperatures, and
hence a shifted temperature range was employed (see
Figures S30-31).
The data in Table 1 indicate that under standard
conditions, the reaction of (*PNP)Ru with H
favorable, followed by that with CO and formic acid; that
with (DMA-H)(DMC) and H O are least favored. These
findings are consistent with the notion that if (PNP)Ru-CO
2
is most
2
2
2
2c
is an off-cycle species, it can form competitively. Upon
dilution, the equilibrium constants decrease accordingly for
all H-X, with the exception of (DMA-H)(DMC). To our
knowledge, only one other report considers equilibria with
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 14
for all others, they are modest (> -4 kcal/mol). In all
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
instances, proton-transfer to the ligand ensues, so the
difference in enthalpies must be attributed to the formation
of strong Ru-X bonds and breaking of (relatively) weak
bonds (vide infra). The three species that have the least
negative enthalpies, (PNP)Ru-OH, (PNP)Ru-OCHO, and
(
PNP)Ru-OMe, are all produced from reaction with
Brønsted acids. Ion-pairing and hydrogen-bonding may
lead to a dissolution enthalpy and hence the larger (less
negative) enthalpies. While (PNP)Ru-DMC should fall in
this category as well, the discrepancy may be attributed to
concomitant formation of DMA.
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 large and negative entropy for (PNP)Ru-CO
PNP)Ru-H is consistent with adding rigidity to the linear
2
and
(
gas molecules. The entropy associated with (PNP)Ru-DMC
-1 -1
of -42 cal∙mol ∙K suggests significant ordering upon its
formation. Because these equilibria take two molecules to
one, we anticipate large and negative entropies. However,
2
this is not the case for formic acid, MeOH, and H O. This may
indicate significant solvent ordering in the free H-X, or
3
2
Figure 5. Overlay of van’t Hoff plots for (*PNP)Ru + H-X or CO
in THF. Extrapolation shows the relative K1,X and K at catalytic
temperatures. Keq has units of: M for formic acid, water,
MeOH; atm for H , CO ; unitless for (DMA-H)(DMC). For the
2 2
equilibrium with (DMA-H)(DMC), the two points that are not
filled are not included in the fit.
2
homoconjugation. Formic acid, MeOH, and H O can all
2
2
hydrogen-bond, and these interactions may in part increase
the entropic contributions in THF.
-1
-1
pK
or from a substrate, the H
proton. This prompted us to investigate the pK
a
Measurements. In reactions that transfer net H
is delivered as a hydride and a
of
2
to
2
a
+
We also note that upon heating (PNP)Ru-DMC in the
(PNP)Ru , (Figure 2, K ) which is produced upon hydride
3
absence of DMA or (DMA-H)(DMC), some (PNP)Ru-CO
2
transfer from (PNP)Ru-H to a substrate. Subsequent proton
transfer then closes the catalytic cycle to give (*PNP)Ru.
forms that remains present upon cooling (see Figures S44-
S45). This indicates that the free (DMC-H) is prone to
Moreover, knowledge of this pK , combined with the
a
2
release of CO , complicating equilibrium measurements at
equilibrium with H , allows the hydricity of (PNP)Ru-H to
2
2
1
elevated temperatures (see SI). Indeed, the van’t Hoff plot
shown in Figure 5 shows a digression from linearity at 60
be determined.
+
Initial attempts to probe the pK
a
of (PNP)Ru were
o
C, which coincides with the temperature at which (DMA-
hampered by coordination of the base to the Ru. The
cationic 16-electron species has a vacant coordination site
and allows for this. For instance, addition of TBMP (TBMP =
31
2
H)(DMC) decomposes to DMA and CO . We thus limit our
analysis of the plot to temperatures less than 60 C to
extrapolate the enthalpy and entropy.
o
2
,6-di-tert-butyl-4-methylpyridine)
instead
gave
Figure 5 indicates that the relative equilibrium constants
coordination of the base. This is evident by NMR
spectroscopy, whereby the H and P resonances for
(PNP)Ru shift, and no resonances ascribed to (*PNP)Ru
1
31
vary with temperature. At the temperature of catalysis, 155
o
+
C, the equilibria with H
. This is notable as these are the most pertinent to
catalysis. We note that through extrapolation, formation of
PNP)Ru-DMC is also expected to be unfavored at catalytic
2
is least favored, followed by that
with CO
2
are observed (see Figure S28). In some instances, treatment
with the base gave (*PNP)Ru as the kinetic product, but
over the course of hours, partially converts to the base-
coordinated thermodynamic product. This is the case with
(
temperatures, though this equilibrium may not be pertinent
at elevated temperatures (vide supra). Rather, formation of
(Li)(NMe ), which converts to (*PNP)Ru prior to further
2
(
PNP)Ru-OCHO, (PNP)Ru-OH, and (PNP)Ru-OMe are all
favored. These represent equilibria with product species.
To favor binding of H to give (PNP)Ru-H at elevated
temperatures, higher pressures of H can be used.
reacting to give a species that is consistent with (PNP)Ru-
NMe (see Figure S30).
2
2
To circumvent this, we turned to the non-nucleophilic
phosphazene bases and their conjugate acids. Upon
2
3
3
Increasing the pressure to 50 bar (as in catalysis, vide infra)
results in a factor of ~ 50 change in the equilibrium constant
4 a
titration of (BTPPH)(BF ) (pK = 20.2 in THF; BTPP = tert-
butylimino-tri(pyrrolidino)phosphorane) to a THF solution
+
(
Table 1). Non-standard conditions, whereby the
of (*PNP)Ru, (PNP)Ru is cleanly generated, as confirmed
concentration of H-X is less than 1 M, give lower equilibrium
constants. This suggests that catalyst inhibition will indeed
increase as product concentrations rise, regardless of the
catalyst loading.
by NMR and UV-vis spectroscopies. Monitoring the
titrations by UV-vis spectroscopy allows for the pK of 20.7
a
± 0.2 to be determined (Figure 2, K ). By contrast, the pK of
3
a
the ligand, PNP, was found to be 28.6 ± 0.1 (see SI), similar
34
to that estimated by others. Thus, the ability for the
The equilibrium enthalpies appear to fall into two
anionic nitrogen to interact strongly with a metal renders
the ligand more acidic by about 8 pK units.
a
categories. For (PNP)Ru-CO
2
, (PNP)Ru-H, and (PNP)Ru-
DMC, the enthalpies are significant (< -12 kcal/mol), whilst
6
ACS Paragon Plus Environment

Page 7 of 14
Journal of the American Chemical Society
pKaa
K6X (M)a
횫퐆
ퟔ,퐗
표
We are also interested in establishing the effect that
Species
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
coordinating a 6 ligand has on the pK
th
. This would allow
a
kcal∙mol-1)
(
for a suitable base to be used to bias the system and more
-
(PNP)
(PNP)Ru+
28.6 ± 0.1
20.7 ± 0.2
24.6 ± 0.4
> 25
--
--
--
favorably release X during catalysis. The ability to isolate
both [K][(*PNP)Ru-CO
a comparison to be made (Figure 2, K
K][(*PNP)Ru-CO ] with (P EtH)(BF
.4 (P Et
2
] and (PNP)Ru-CO
2
allows for such
). Titration of
--
--
4
(
PNP)Ru-CO
2
--
[
0
2
2
4
) gives a pK
a
of 24.6 ±
=
2 × 10-3
(
PNP)Ru-
OCHO
3.7 ± 0.7
2
tetramethyl(tris(dimethylamino)phosphoranylidene)phos
9.6 × 10-11
6.3 × 10-12
5.0 × 10-32
13.7 ± 0.7
15.3 ± 0.5
42.7 ± 0.6
3
3
(PNP)Ru-DMC
PNP)Ru-Cl
PNP)Ru-H
> 32
> 34
--
phorictriamide-Et-imin; pK
a
= 25.3 in THF ). Coordination
of a 6 ligand increases the ligand basicity by ~ 4 pK units.
Together, these pK measurements indicate that the Ru
th
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
a
(
center can greatly modulate the acidity of the ligand.
a
o
X
Measurements done at 20 C. We assume that the
t
t
P Bu2
K8x
P Bu
2
o
o
+ BH+ + X-
equilibrium at 25 C is ~ that at 20 C.
N
Ru C O
+ B
N
Ru
H
C O
H
P
Bu2
P
t
H
t
H
Bu2
Hydricity and binding constants. The hydricity of
PNP)Ru-H can be determined from the equilibria shown in
(
8)
Titration of other (PNP)Ru-X (X = Cl, DMC, OCHO) species
(
Scheme 3.
with a suitable base cleanly generates (*PNP)Ru (eq 8, K8x).
This equilibrium is formally a proton transfer followed by
Scheme 3. Equilibria employed to determine the hydricity
of (PNP)Ru-H.
ligand loss and thus cannot give a true pK
However, knowledge of the pK for the added base (KBH+
and estimation of K7x allows for a lower-limit to the pK
PNP)Ru-X to be obtained.
Scheme 2. Relationship between the equilibrium of eq 8
a
. (Scheme 2).
a
)
∗
(퐾1,퐻) ―1
퐾3
(퐏퐍퐏)퐑퐮–퐇 ⇌ ( 퐏퐍퐏)퐑퐮 + H
( ∗ 퐏퐍퐏)퐑퐮 + H + ⇌(퐏퐍퐏)퐑퐮 +
H ⇌ H + + H ―
2
a
of
(
퐾H2 ―
2
and the pK
a
of (PNP)Ru-X.
(퐏퐍퐏)퐑퐮–퐇 ⇌ (퐏퐍퐏)퐑퐮 + + H ―
퐾6,H
(퐏퐍퐏)퐑퐮–퐗⇌ [( 퐏퐍퐏)퐑퐮–퐗] + H +
∗
―
K
K
5x
7x
a
The hydricity is obtained by combining the pK of
( ∗ 퐏퐍퐏)퐑퐮–퐗] ― ⇌( 퐏퐍퐏)퐑퐮 + X
B + H + ⇌ BH +
퐏퐍퐏)퐑퐮–퐗 + 퐁⇌ ( 퐏퐍퐏)퐑퐮 + BH + + 푋 ―
∗
―
(PNP)Ru
and the free energy of H
+
, the free energy of hydride transfer from H2,36,37
loss from (PNP)Ru-H. The
[
(KBH+)-1
2
thermochemical cycle gives a hydricity of 42.7 ± 0.6
kcal/mol (44.6 ± 0.6 kcal/mol using the value of KH2- given
in ref. 37) It is difficult to draw comparisons to the hydricity
of other metal complexes, as this is the first reported
hydricity in THF. As noted by others, hydricity values for
∗
(
K
8x
–
Ligand loss from [(*PNP)Ru-X] to (*PNP)Ru (K7x) is
expected to be favorable as the electron count drops from
3
8
metal complexes are solvent dependent,
with no
20 to 18 (Figure 2). To probe this, 1 – 10 equivalents of
qualitative trends that convert hydricity amongst solvents.
This is also the first hydricity value reported for a metal
n
( Bu
4
N)(X) (X = Cl or OCHO) were added to a 8.4 mM d
8
-
THF solution of (*PNP)Ru. No immediate change was
observed by NMR spectroscopy (Figure S40), but for the
chloride sample, (PNP)Ru-Cl is produced over the course
of hours. This transformation necessitates a proton and
likely occurs from Hofmann degradation35 of the
tetrabutylammonium cation, facilitating binding of the
chloride. Addition of 10 equiv (Li)(DMC) to (*PNP)Ru gave
no reaction (see Figure S43). As dimethylcarbamate is more
basic than formate or chloride and hence should bind most
favorably to the Ru, we assume that there is no appreciable
reaction of the X species binding to (*PNP)Ru. Given the
limits of what we can observe, we assign a lower limit of 100
to K7x. The equilibrium constant is likely to be much larger,
the extent of which will vary with the identity of X (vide
complex that undergoes MLC; others have measured
30a,39
hydricities for P
2 2
N -ligated metal hydrides,
whereby
protonation also occurs at the ligand, but subsequent
proton transfer does not impact the nature of the metal-
ligand bonds.
The hydricity allows for evaluation of whether hydride
transfer to a substrate, for example, CO , is favorable. For
2
that to be the case, the hydricity of (PNP)Ru-H must be less
than that of formate. While it may be tempting to
approximate that the hydricity of formate in THF must be
greater than 42.7 kcal/mol, this is not valid in this system
because the resulting formate binds to the Ru and this
energy needs to be accounted for (Scheme 4). This binding
may give an apparent increase in hydricity for (PNP)Ru-H,
which can differ based on subsequent binding of the
hydride-transferred product (Scheme 4, left).
-
-
infra).
This analysis provides lower-limits of 25 and 32 for the
respective pK of (PNP)Ru-OCHO and (PNP)Ru-DMC.
a
Shifting from a carboxylate to a chloride, (PNP)Ru-Cl
further increases the basicity of the ligand, with a lower
Scheme 4. Relationship between hydricity and binding of
+
formate to [(PNP)Ru] (two vertical equilibria).
limit of the pK
a
now being 34 (Table 2).
s and thermodynamic parameters
pertinent to X release from (PNP)Ru-X.
Table 2. Summary of pK
a
-
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 14
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
The free energy associated with release of formate from
the Ru (Scheme 4, left) can be determined from one of the
two thermochemical cycles, which are described in Scheme
5.
From this analysis, the free energy for formate release is
3.7 ± 0.7 kcal/mol (Table 2). This is appreciably less than
the energy associated with DMC or chloride release, 13.7
and 15.3 kcal/mol respectively.
Binding of these species modifies the apparent hydricity.
Formate binding gives an apparent hydricity of 39.0
kcal/mol (40.9 kcal/mol using the value of KH2- given in ref.
Catalytic Hydrogenation of CO
DMA-H)(DMC) by (PNP)Ru-Cl was probed (Table 3). The
use of dimethylammonium dimethylcarbamate as a CO
surrogate for hydrogenation to MeOH was first shown by
2
. The hydrogenation of
(
-
2
37). Given that other X species bind more tightly, it is
conceivable that the apparent hydricity can further be
7
b
-1
Sanford by use of a related Ru catalyst. To draw
comparisons, similar conditions were employed in terms of
substrate and base equivalents, temperature, and solvent.
We first examined the effect of catalyst concentration on
conversion to MeOH, formate, and DMF, while maintaining
a constant loading of 1 mol% (entries 1-5). As the catalyst
concentration is increased from 0.05 mM to 5.3 mM, the
TON of MeOH decreases. When the amount of MeOH
produced is viewed instead as the concentration of MeOH,
we see that except for 0.05 mM catalyst, the final
concentration of MeOH is essentially constant (~2 mM).
This suggests to us that there might be a deactivation
mechanism that is dependent on the concentration of MeOH
reduced. From this, and a lower-limit of 520 atm for K
Scheme 4 and Figure S24), we determine that the lower-
limit for the hydricity of formate in THF is 42.7 kcal/mol
44.7 kcal/mol when using the values of ref. 37). For
comparison, GH-(OCHO) is estimated to be 44 kcal/mol in
9
(
(
2
1
MeCN, and 24.1 kcal/mol in water.
The thermochemical cycles of Scheme 5 allow for the
evaluation of the pK s of H-OCHO and H-DMC in THF. Values
a
of 20.8 ± 0.6 and 30.6 ± 0.6 are obtained, respectively. The
value obtained for formic acid is similar to that of acetic acid
40
in THF (22.5), further validating our equilibration studies.
Scheme 5. Thermochemical cycles that can be employed to
determine K6,X (highlighted in blue).
and/or H O. Indeed, during our equilibrium studies of
2
(
2
*PNP)Ru and MeOH/H O, we saw irreversible
decomposition at elevated temperatures to intractable Ru
species (See SI).
We then consider the effect of the base strength (entries
t
1
, 6-9). Going from K
PO
3 4
to BuOK gives an increase in
t
a 3 4
MeOH production. While the pK s of K PO and BuOK are
not known in THF, we note that the former is not capable of
converting (PNP)Ru-Cl to (*PNP)Ru at room temperature,
t
whilst the latter can, suggesting that BuOK is more basic.
When we use (Li)(DMC) as the base, we see an order of
magnitude increase in TON to 39.4. Now, we reach our
limiting concentration of MeOH produced with 0.05 mM
2
catalyst loading. However, with (Li)(Me N) the TON drops
to 1.6. We attribute this decrease to competitive amine
binding which may occur at elevated temperatures. With
Ph
H
(
P NP)Ru-(HBH
3
), which is known to hydrogenate
8
ACS Paragon Plus Environment

Page 9 of 14
Journal of the American Chemical Society
7
a,7b
carbamates to MeOH,
base ( BuOK vs. K PO
3 4
we see no effect on varying the
) on MeOH yield (see SI).
Given that the produced H O may quench one equiv of
To determine if there is a substrate which is limiting TON,
we compared the substrate (entries 1, 10-12). Going from
(DMA-H)(DMC) to (Li)(DMC), we see a doubling of the
MeOH TON. This may be due to lower concentrations of
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
t
2
base, we also investigated using equimolar substrate and
base. When we use equimolar BuOK and (DMA-H)(DMC),
we see a slight decrease in TON compared to when we use
1:2 base:substrate (Table S4). A decrease is also observed
when we use 1:4 base:substrate. We interpret these results
as suggesting an intricate balance between the requirement
of the base, and the impact of adding base to the equilibrium
+
+
t
amine (DMA-H versus Li ) or discrepancies in how readily
CO is released from DMC. Formic acid is poorly
2
hydrogenated, which may be due to the now acidic nature
of the reaction medium, as only 50 equivalents of base was
added. When DMF is employed as a substrate, we see MeOH
produced with a TON of 59.2, again giving a similar limiting
final concentration.
of (DMA-H)(DMC) and DMA + CO
2
(with base favoring
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
formation of DMC salts).
Table 3. Hydrogenation of (DMA-H)(DMC) by (PNP)Ru-Cl. All reactions are run in 10 mL THF.
0.05-5.3 mol (PNP)Ru-Cl
Cl
t
P Bu2
50 equiv Base
50 bar H2
N
Ru C O
substrate
(100 equiv)
MeOH, Formate, DMF
H
P
Bu2
H
THF, 155 °C, 18 h
t
H
Entry
1
[catalyst]
mM) (mol%)
Substrate
Base
TON MeOH (mM)a
3.9 ± 0.3
TON Formate (mM)b
TON DMF
(mM)
a
(
0.05 (1%)
0.5 (1%)
1.5 (1%)
3.3 (1%)
5.3 (1%)
0.05 (1%)
0.05 (1%)
0.05 (1%)
0.05 (1%)
(DMA-H)(DMC)g
(DMA-H)(DMC)g
tBuOK
3.3 ± 0.2
(0.17)
19.8 ± 5.4
(4.1)
(
0.20)
4.0 ± 0.2
2.1)
2
3
tBuOK
5.3 ± 0.2
(2.6)
57.2 ± 15.8
(140)
(
(DMA-H)(DMC)g
tBuOK
1.2 ± 0.2
(1.7)
0.6 ± 0.1
(0.90)
0.5 ± 0.3
(57)
(DMA-H)(DMC)g
tBuOK
n.d.e
4
0.5 ± 0.1
0.3 ± 0.1
(0.98)
(
1.7)
0.4 ± 0.1
2.0)
5
(DMA-H)(DMC)g
(DMA-H)(DMC)g
(DMA-H)(DMC)g
(DMA-H)(DMC)g
(DMA-H)(DMC)g
tBuOK
None
0.8 ± 0.1
(4.5)
0.3 ± 0.3
(0.016)
(
6c
7
2.0 ± 0.7
(0.065)
0.7 ± 0.4
(0.046)
n.d.e
K
3
PO
4
2.6 ± 0.5
1.6 ± 0.7
(0.23)
10.9 ± 7.7
(4.1)
(
0.14)
39.4 ± 0.4
2.0)
8
(Li)(DMC)d
6.3 ± 5.7
(0.32)
0.5 ± 0.4
(0.00095)
(
9
(Li)(Me
2
N)
1.6 ± 0.3
(0.078)
0.7 ± 0.4
(0.035)
1.0 ± 1.0
(0.001)
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 14
1
0
1
0.05 (1%)
0.05 (1%)
0.05 (1%)
(Li)(DMC)
HCOOH
DMF
None
tBuOK
tBuOK
7.5 ± 4.9
0.38)
1.2 ± 0.6
0.062)
59.2 ± 2.3
2.9)
1.3 ± 0.5
(0.065)
n.d.e
n.d.e
--f
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
--f
(
12
0.7 ± 0.3
(3.6 x 10-5)
(
a
b
c
Quantified by GC. Error from duplicate catalytic runs. Quantified by IC. Error from duplicate catalytic runs. (*PNP)Ru used as
the catalyst. 100 equiv of base employed. n.d. = not detected. not quantified. 92 equiv of substrate employed.
d
e
f
g
classes that differ by the magnitude of this effect. Our study
shows that the decrease is not uniform across all substrates.
Thus, the equilibrium with formic acid is more or less
temperature independent, whilst that with H shows a
strong decrease with increasing temperatures. If formation
of (PNP)Ru-OCHO does indeed represent a thermodynamic
bottleneck, or the most favorable equilibrium, then
increasing the temperature would not favor formation of
(PNP)Ru-H; only raising the pressure or lowering the
temperature would do so. Varying the temperature and
concentrations can likewise impact the thermodynamic
bottlenecks; the extent of which would not be known
without studies such as this.
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
DISCUSSION
Ramifications of H-X equilibria on catalysis. Given the
concentration (or pressure) dependence for the equilibria
of (*PNP)Ru and H-X or CO
2
2
, it is imperative to understand
how different catalyst loadings impact TON. Regardless of
the catalyst concentration, only the H-X concentration will
determine the equilibrium position. This is illustrated when
considering the TON of MeOH produced. While varying the
catalyst loading impacts the TON for MeOH, at the four
highest loadings considered it did not impact the overall
concentration of MeOH produced, ~ 2 mM. This same
concentration can be obtained at lower catalyst loading
when using (Li)(DMC) as the base. Strikingly, when DMF is
instead used as a substrate, the same concentration of
MeOH is produced. This suggests that competitive
equilibria may limit the overall MeOH production. Indeed,
at elevated temperatures, the equilibrium to give (PNP)Ru-
OMe is favored over that of (PNP)Ru-H. In this system,
further complication likely arises from the instability of
These findings are not unique to hydrogenation of
carbamates
to
( P NP)Ru(HBH
hydrogenation to DMF (in the presence of DMA) is found to
MeOH.
For
example,
using
Ph
H
3
), the optimal temperature for CO
2
o
7b
o
be 95 C, well below the 155 C used to hydrogenate the
produced DMF to MeOH. This is consistent with the idea of
the relative equilibria shifting with temperature. Also, a
recent study on dehydrogenative coupling of
(
*PNP)Ru to excess MeOH at elevated temperatures; the
ability for the catalyst to turnover suggests that this
deactivating equilibrium is less favorable than that which
gives productive turnover.
ethylenediamine and MeOH to give ethylene urea and H
showed that increasing the headspace volume, effectively
decreasing the [H ], enhances this Ru catalyzed reaction.
2
These studies on related Ru catalysts that undergo MLC
reinforce the need to understand the thermodynamic
parameters associated with the catalyst; this allows for
optimization of the catalytic conditions.
2
42
A threshold MeOH concentration is not unique to our
system. Both Sanford and Prakash have used ( P NP)Ru-
Cl and ( P NP)Ru-(HBH ) to hydrogenate CO to MeOH in
3 2
the presence of a variety of amines. Under their conditions,
Sanford reports a maximum MeOH concentration of about 1
M. In three separate reports, Prakash reports maximum
MeOH concentrations of 1.3 M, and 2.1 M.
same catalyst, run under different conditions, in different
solvents, amines, and base additives, all give a similar
maximum MeOH concentration suggests a concentration-
dependent equilibrium between MeOH and the catalyst
contributes to catalyst performance. Related Ru catalysts
that undergo MLC via dearomatization give ~1 M limiting
Ph
H
Ph
H
Knowledge of the temperature and pressure dependence
7
b
for the H
to CO hydrogenation. With regards to hydrogenation to
formate, Pidko has suggested that (PNP)Ru-CO is an
inhibitive, off-cycle species and Sanford suggests that the
related (PNN)Ru-CO may be able to do catalysis, albeit at a
2 2
equilibrium versus that with CO is also pertinent
1
4
7a,41
That the
2
2
2c
2
2a
slower rate; this species should therefore be minimized
under catalytic conditions. Our studies indicate that at all
elevated temperatures, the equilibrium with CO
over that of H and thus pressure differences must be
employed. While this is usually achieved by changing the
partial pressures of the two gases, it can also be achieved by
2
is favored
7c,26b
MeOH concentrations in other types of hydrogenations.
2
The discrepancy between the catalysts likely has
contributions from differences in equilibrium constants,
and in the latter systems, different H-X species present and
hence different competing equilibria. Regardless, these
similarities suggest that a better parameter for evaluating
catalysis is product concentration, not TON.
using “captured CO
carbamic acid. These are in equilibria with free CO
2
” in the form of carbamates and
2
(eq 9).
푅 ′2 푁퐻
→
←
→
←
+
퐶푂푂 ―
퐶푂 + 푅 푁퐻 푅2푁 (퐻)
(
)
푅2푁 ― 퐶푂푂퐻
[푅2
( 9 )
2
2
←
Knowing both the concentration and temperature-
dependence of these equilibria (K1,X) allows for a better
understanding of how to optimize reaction conditions. In
general, the equilibrium constant towards (PNP)Ru-X
decreases with temperature, and the substrates fall into two
′
푁 ― 퐶푂푂][푅 푁퐻 ]
Moreover, it has been shown that by changing the amine
identity or using mixtures of amines, the pressure and
2
2
temperature requirements for releasing CO
2
can be greatly
43
modified. This relative equilibria is particularly pertinent
1
0
ACS Paragon Plus Environment

Page 11 of 14
Journal of the American Chemical Society
to carbon capture and recycling schemes, as CO
energy-intensive. Coupling release to catalysis will
streamline the process, and by judicious choice of amine,
will allow for tailoring the H :CO to optimize catalysis.
2
release is
Hydricity is modified by binding of substrate. Miller
and coworkers recently disclosed how hydricity of Ru and
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
2
Ir hydrides varies in H O with added buffers that can
4
4
2
2
coordinate to the metal. In their work, they found that
-
2
binding of H O versus Cl can alter the apparent hydricity by
Rationalization of the need for a strong base during
catalysis. It is established that using (PNP)Ru-Cl with DBU
about 3-5 kcal/mol. While their studies focused on buffers,
here we extend this idea to how coordination of the product
of hydride transfer impacts the apparent hydricity.
2
as a co-substrate base allows for CO hydrogenation to
2
d,20
40
formate.
a
While DBU, having a pK of 16.1 in THF, is not
sufficiently basic enough to deprotonate (PNP)Ru-OCHO at
room temperature, precipitation of (DBU-H)(OCHO) is not
accounted for, nor are subsequent equilibria with the
product (*PNP)Ru, both of which could drive the reaction.
It is thus conceivable that under reaction conditions, which
include elevated temperatures, this deprotonation readily
Our analysis gives a hydricity of 42.7 kcal/mol for
(PNP)Ru-H in THF. This is the value to consider if the
substrate does not coordinate to the Ru as part of the
catalytic cycle. However, our thermodynamic studies
indicate that substrate binding is favorable, giving
(PNP)Ru-X. This binding is favorable by ~4-15 kcal/mol,
and hence can greatly impact the apparent hydricity. If
hydrogenation does occur via coordination of the product
of hydride transfer, for example, formate, then catalyst
hydricity does not need to be great er than that of formate.
Rather, the catalyst hydricity and binding affinity for the
product must be larger than that of formate. Due to the
inability to isolate (PNP)Ru-OMe, we are unable to
determine the methoxide binding affinity. However, Table 2
suggests that it may differ from that of formate and require
different hydricity parameters from the catalyst.
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
occurs. Indeed, treatment of (PNP)Ru-Cl with K
3 4
PO at
room temperature gives no reaction, but upon warming to
o
1
20 C, formation of (*PNP)Ru ensues, as evident by the
characteristic color change from yellow to green. Cooling
the sample causes it to revert back to (PNP)Ru-Cl. There is
thus
a
a temperature dependence on the pK , but
deconvoluting the contribution of the inorganic base and
the Ru species is not trivial.
Our results suggest that using a stronger base allows for
and increases MeOH production when (PNP)Ru-Cl is used
as a catalyst for the hydrogenation of (DMA-H)(DMC). Given
the amount of base added, we cannot conclude whether it is
a co-reagent or a co-catalyst. Others have noted that in
related hydrogenations, addition of a base enhances
catalysis, though the base is added in sub-stoichiometric
CONCLUSIONS
We presented a thorough thermochemical analysis of the
Ru PNP hydrogenation catalyst. The temperature-
dependent equilibrium studies suggest what may be
thermodynamic bottlenecks and how they vary with
temperature. At elevated temperatures, (PNP)Ru-OCHO
formation is favored over (PNP)Ru-H by 4 orders of
magnitude in K1,X. This explains why formic acid is not
readily released, but rather a base is required to convert
(PNP)Ru-OCHO to (*PNP)Ru. A stronger base is required
in this system to drive hydrogenation of (DMA-H)(DMC) to
MeOH, while this is not a requirement in related systems.
7
b,14
amounts.
It is speculated that the base enhances the
MLC mechanism, and thus the base strength may alter the
mechanism.
As noted by others, K
production when ( P NP)Ru(HBH
3
PO
4
is sufficient for MeOH
) is used as the catalyst
Ph
H
3
7
a
for hydrogenation of (DMA-H)(DMC). Changing the base
t
to KO Bu does not improve MeOH production. By contrast,
with (PNP)Ru-X a stronger base is required. This suggests
that the base is used to deprotonate the catalyst. The
alternative, deprotonation of an organic intermediate,
would show the same base-dependence, regardless of the
catalyst.
a
We find that the pK of (PNP)Ru-X varies greatly with the
identity of the X-ligand, and hence suggest that the stronger
base is required to deprotonate a 6-coordinate Ru species
that is bound by an intermediate en route to MeOH
production. Our hydrogenation studies indicate that there
is a limiting methanol concentration that is reached with
this system. While there are likely many factors to this, the
equilibria studies are consistent with this observation and
likely contributes to performance. We also give a lower
estimate for the hydricity of formate in THF, and indicate
how substrate binding can impact the apparent hydricity of
a catalyst by several kcal/mol. This work mirrors that done
in the electrochemical field, whereby such findings have
This can be rationalized by considering our pK
measurements, which indicate that 6-coordinate (PNP)Ru-
X species are at least 5 pK units more basic than 5-
a
a
+
coordinate (PNP)Ru . Thus, addition of a strong base can
bias the system to regenerate (*PNP)Ru. It should be noted
a
that of the 6-coordinate (PNP)Ru-X species whose pK we
could estimate, (PNP)Ru-OCHO gave the lowest value,
consistent with a relatively weak base (DBU) facilitating
this hydrogenation. Additionally, our equilibrium studies
indicate that (PNP)Ru-OCHO does not readily dissociate
formic acid, even at elevated temperatures. This suggests
that the only way to turn over the catalyst is to use a base.
Our best results for MeOH production used (Li)(DMC) as the
2
allowed for better H evolution catalysts to be developed
and have helped ascertain catalyst requirements for
15-16
electrocatalytic CO
benchmarks several thermochemical properties pertinent
to CO hydrogenation catalysts, we envision that our
2
reduction.
Although this work
2
a
base. From our thermochemical cycles, the pK of DMC-H,
findings will be broadly applicable to a variety of other
catalytic reactions performed by this and related catalysts
that require cooperation between the metal and
bifunctional ligand.
dimethylcarbamic acid, is 30.6, significantly higher than the
DBU that is needed for formate. It therefore may serve to
deprotonate a (PNP)Ru-X species that is pertinent to DMF
hydrogenation.
1
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 14
Carbon dioxide hydrogenation catalysed by well-defined Mn( i )
PNP pincer hydride complexes Chem. Sci. 2017, 8, 5024.
(5) Examples with Ru: Zhang, L.; Han, Z.; Zhao, X.; Wang, Z.; Ding,
K. Highly Efficient Ruthenium-Catalyzed N-Formylation of Amines
ASSOCIATED CONTENT
Supporting Information
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
The Supporting Information is available free of charge on the
ACS Publications website.
with H
2 2
and CO Angew. Chem. Int. Ed. 2015, 54, 6186.
st
(
6) Examples with 1 row transition metals: (a) Jayarathne, U.;
Hazari, N.; Bernskoetter, W. H. Selective Iron-Catalyzed N-
Formylation of Amines using Dihydrogen and Carbon Dioxide ACS
Catal. 2018, 8, 1338; (b) Daw, P.; Chakraborty, S.; Leitus, G.; Diskin-
Posner, Y.; Ben-David, Y.; Milstein, D. Selective N-Formylation of
Experimental details, including syntheses, NMR and UV-vis
data, and equilibrium plots (PDF)
Crystallographic data (CIF)
Amines with H
Catal. 2017, 7, 2500.
(7) Examples with Ru: (a) Kothandaraman, J.; Goeppert, A.;
Czaun, M.; Olah, G. A.; Prakash, G. K. S. Conversion of CO from Air
2 2
and CO Catalyzed by Cobalt Pincer Complexes ACS
AUTHOR INFORMATION
Corresponding Author
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
into Methanol Using a Polyamine and a Homogeneous Ruthenium
Catalyst J. Am. Chem. Soc. 2016, 138, 778; (b) Rezayee, N. M.; Huff,
C. A.; Sanford, M. S. Tandem Amine and Ruthenium-Catalyzed
*caroline.saouma@utah.edu
Notes
The authors declare no competing financial interest.
Hydrogenation of CO to Methanol J. Am. Chem. Soc. 2015, 137,
2
1028; (c) Khusnutdinova, J. R.; Garg, J. A.; Milstein, D. Combining
ACKNOWLEDGMENT
Low-Pressure CO
ACS Catal. 2015, 5, 2416; (d) Everett, M.; Wass, D. F. Highly
productive CO hydrogenation to methanol – a tandem catalytic
approach via amide intermediates Chem. Commun. 2017, 53, 9502.
2
Capture and Hydrogenation To Form Methanol
The authors gratefully acknowledge start-up funding from the
University of Utah and USTAR, ACS PRF (57058-DNI3), and NSF
REU (1659579).
2
st
(
8) Examples with 1 row transition metals: (a) Ribeiro, A. P. C.;
Martins, L. M. D. R. S.; Pombeiro, A. J. L. Carbon dioxide-to-methanol
single-pot conversion using a C-scorpionate iron(ii) catalyst Green
Chemistry 2017, 19, 4811; (b) Kar, S.; Goeppert, A.;
Kothandaraman, J.; Prakash, S. G. K. Manganese-Catalyzed
REFERENCES
(
1) (a) Khusnutdinova, J. R.; Milstein, D. Metal–Ligand
Cooperation Angew. Chem. Int. Ed. 2015, 54, 12236; (b) Alig, L.;
Fritz, M.; Schneider, S. First-Row Transition Metal
Sequential Hydrogenation of CO to Methanol via Formamide ACS
2
(
De)Hydrogenation Catalysis Based On Functional Pincer Ligands
Catal. 2017, 6347.
(9) Wang, W.-H.; Himeda, Y.; Muckerman, J. T.; Manbeck, G. F.;
Fujita, E. CO2 Hydrogenation to Formate and Methanol as an
Chem. Rev. 2019, 119, 2681; (c) Gunanathan, C.; Milstein, D. Bond
activation and catalysis by ruthenium pincer complexes Chem. Rev.
2
014, 114, 12024.
Alternative to Photo- and Electrochemical CO Reduction Chem.
2
(
2) Examples with Ru: (a) Huff, C. A.; Sanford, M. S. Catalytic CO
2
Rev. 2015, 115, 12936.
Hydrogenation to Formate by a Ruthenium Pincer Complex ACS
Catal. 2013, 3, 2412; (b) Filonenko, G. A.; Smykowski, D.; Szyja, B.
M.; Li, G.; Szczygieł, J.; Hensen, E. J. M.; Pidko, E. A. Catalytic
(10) Yang, X. Hydrogenation of Carbon Dioxide Catalyzed by PNP
Pincer Iridium, Iron, and Cobalt Complexes: A Computational
Design of Base Metal Catalysts ACS Catal. 2011, 1, 849.
Hydrogenation of CO
2
to Formates by a Lutidine-Derived Ru–CNC
(11) (a) Schneidewind, J.; Adam, R.; Baumann, W.; Jackstell, R.;
Beller, M. Low-Temperature Hydrogenation of Carbon Dioxide to
Methanol with a Homogeneous Cobalt Catalyst Angew. Chem. Int.
Ed. 2017, 56, 1890; (b) Wesselbaum, S.; Moha, V.; Meuresch, M.;
Brosinski, S.; Thenert, K. M.; Kothe, J.; Stein, T. v.; Englert, U.;
Holscher, M.; Klankermayer, J.; Leitner, W. Hydrogenation of
carbon dioxide to methanol using a homogeneous ruthenium-
Triphos catalyst: from mechanistic investigations to multiphase
catalysis Chem. Sci. 2015, 6, 693; (c) Wesselbaum, S.; vomꢀStein, T.;
Klankermayer, J.; Leitner, W. Hydrogenation of Carbon Dioxide to
Pincer Complex: Theoretical Insight into the Unrealized Potential
ACS Catal. 2015, 5, 1145; (c) Filonenko, G. A.; Hensen, E. J. M.; Pidko,
E. A. Mechanism of CO
homogeneous Ru-PNP pincer catalyst: from
description to performance optimization Catal. Sci. Technol. 2014,
, 3474; (d) Filonenko, G. A.; Conley, M. P.; Copéret, C.; Lutz, M.;
2
hydrogenation to formates by
a
theoretical
4
Hensen, E. J. M.; Pidko, E. A. The impact of Metal–Ligand
Cooperation in Hydrogenation of Carbon Dioxide Catalyzed by
Ruthenium PNP Pincer ACS Catal. 2013, 3, 2522.
(3) Examples with Ir: (a) Tanaka, R.; Yamashita, M.; Nozaki, K.
Catalytic Hydrogenation of Carbon Dioxide Using Ir(III)−Pincer
Complexes J. Am. Chem. Soc. 2009, 131, 14168; (b) Schmeier, T. J.;
Dobereiner, G. E.; Crabtree, R. H.; Hazari, N. Secondary
Coordination Sphere Interactions Facilitate the Insertion Step in an
Methanol by Using a Homogeneous Ruthenium–Phosphine
Catalyst Angew. Chem. Int. Ed. 2012, 51, 7499; (d) Huff, C. A.;
Sanford, M. S. Cascade Catalysis for the Homogeneous
Hydrogenation of CO to Methanol J. Am. Chem. Soc. 2011, 133,
2
18122.
Iridium(III) CO
9274.
2
Reduction Catalyst J. Am. Chem. Soc. 2011, 133,
(12) Vitillo, J. G.; Smit, B.; Gagliardi, L. Introduction: Carbon
Capture and Separation Chem. Rev. 2017, 117, 9521.
(4) Examples with 1st row transition metals: (a) Langer, R.;
Diskin-Posner, Y.; Leitus, G.; Shimon, L. J. W.; Ben-David, Y.;
Milstein, D. Low-Pressure Hydrogenation of Carbon Dioxide
Catalyzed by an Iron Pincer Complex Exhibiting Noble Metal
Activity Angew. Chem. Int. Ed. 2011, 50, 9948; (b) Bernskoetter, W.
H.; Hazari, N. Reversible Hydrogenation of Carbon Dioxide to
Formic Acid and Methanol: Lewis Acid Enhancement of Base Metal
Catalysts Acc. Chem. Res. 2017, 50, 1049; (c) Bielinski, E. A.;
Lagaditis, P. O.; Zhang, Y.; Mercado, B. Q.; Würtele, C.; Bernskoetter,
W. H.; Hazari, N.; Schneider, S. Lewis Acid-Assisted Formic Acid
Dehydrogenation Using a Pincer-Supported Iron Catalyst J. Am.
Chem. Soc. 2014, 136, 10234; (d) Bertini, F.; Glatz, M.; Gorgas, N.;
Stöger, B.; Peruzzini, M.; Veiros, L. F.; Kirchner, K.; Gonsalvi, L.
(13) Heldebrant, D. J.; Koech, P. K.; Glezakou, V.-A.; Rousseau, R.;
Malhotra, D.; Cantu, D. C. Water-Lean Solvents for Post-Combustion
CO2 Capture: Fundamentals, Uncertainties, Opportunities, and
Outlook Chem. Rev. 2017, 117, 9594.
(14) Kar, S.; Sen, R.; Kothandaraman, J.; Goeppert, A.;
Chowdhury, R.; Munoz, S. B.; Haiges, R.; Prakash, G. K. S.
Mechanistic Insights into Ruthenium-Pincer-Catalyzed Amine-
Assisted Homogeneous Hydrogenation of CO to Methanol J. Am.
2
Chem. Soc. 2019, 141, 3160.
(15) (a) DuBois, D. L. Development of Molecular Electrocatalysts
for Energy Storage Inorg. Chem. 2014, 53, 3935; (b) Helm, M. L.;
Stewart, M. P.; Bullock, R. M.; DuBois, M. R.; DuBois, D. L. A Synthetic
Nickel Electrocatalyst with a Turnover Frequency Above 100,000
1
2
ACS Paragon Plus Environment

Page 13 of 14
Journal of the American Chemical Society
s−1 for H
Fettinger, J. C.; Berben, L. A. Control of Ligand pK
Production Science 2011, 333, 863; (c) Sherbow, T. J.;
Values Tunes the
(27) Zhang, L.; Nguyen, D. H.; Raffa, G.; Trivelli, X.; Capet, F.;
2
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
Desset, S.; Paul, S.; Dumeignil, F.; Gauvin, R. M. Catalytic Conversion
of Alcohols into Carboxylic Acid Salts in Water: Scope, Recycling,
and Mechanistic Insights ChemSusChem 2016, 9, 1413.
(28) Huff, C. A.; Kampf, J. W.; Sanford, M. S. Reversible carbon–
carbon bond formation between carbonyl compounds and a
ruthenium pincer complex Chem. Commun. 2013, 49, 7147.
(29) Lilio, A. M.; Reineke, M. H.; Moore, C. E.; Rheingold, A. L.;
Takase, M. K.; Kubiak, C. P. Incorporation of Pendant Bases into
Electrocatalytic Dihydrogen Evolution Mechanism in a Redox-
Active Aluminum(III) Complex Inorg. Chem. 2017, 56, 8651; (d)
Rountree, E. S.; Martin, D. J.; McCarthy, B. D.; Dempsey, J. L. Linear
Free Energy Relationships in the Hydrogen Evolution Reaction:
Kinetic Analysis of a Cobaloxime Catalyst ACS Catal. 2016, 6, 3326.
(16) (a) Taheri, A.; Thompson, E. J.; Fettinger, J. C.; Berben, L. A.
An Iron Electrocatalyst for Selective Reduction of CO
in Water: Including Thermochemical Insights ACS Catal. 2015, 5,
140; (b) Waldie, K. M.; Ostericher, A. L.; Reineke, M. H.; Sasayama,
A. F.; Kubiak, C. P. Hydricity of Transition-Metal Hydrides:
2
to Formate
Rh(diphosphine)
And Catalytic CO
Complexes J. Am. Chem. Soc. 2015, 137, 8251.
2
Complexes: Synthesis, Thermodynamic Studies,
+
7
2
2 2 2
Hydrogenation Activity of [Rh(P N ) ]
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
Thermodynamic Considerations for CO
Reduction ACS Catal.
(30) (a) Lense, S.; Ho, M.-H.; Chen, S.; Jain, A.; Raugei, S.; Linehan,
J. C.; Roberts, J. A. S.; Appel, A. M.; Shaw, W. Incorporating Amino
Acid Esters into Catalysts for Hydrogen Oxidation: Steric and
Electronic Effects and the Role of Water as a Base Organometallics
2012, 31, 6719; (b) Wilson, A. D.; Newell, R. H.; McNevin, M. J.;
Muckerman, J. T.; DuBois, M. R.; DuBois, D. L. Hydrogen oxidation
and production using nickel-based molecular catalysts with
positioned proton relays J. Am. Chem. Soc. 2006, 128, 358.
2
2018, 8, 1313.
(
17) Jeletic, M. S.; Hulley, E. B.; Helm, M. L.; Mock, M. T.; Appel, A.
M.; Wiedner, E. S.; Linehan, J. C. Understanding the Relationship
Between Kinetics and Thermodynamics in CO Hydrogenation
2
Catalysis ACS Catal. 2017, 7, 6008.
(18) (a) Gnanaprakasam, B.; Zhang, J.; Milstein, D. Direct
Synthesis of Imines from Alcohols and Amines with Liberation of
H2 Angew. Chem. Int. Ed. 2010, 49, 1468; (b) Gnanaprakasam, B.;
Balaraman, E.; Ben-David, Y.; Milstein, D. Synthesis of Peptides and
Pyrazines from β-Amino Alcohols through Extrusion of H2
Catalyzed by Ruthenium Pincer Complexes: Ligand-Controlled
Selectivity Angew. Chem. Int. Ed. 2011, 50, 12240; (c) Xie, Y.; Ben-
David, Y.; Shimon, L. J. W.; Milstein, D. Highly Efficient Process for
Production of Biofuel from Ethanol Catalyzed by Ruthenium Pincer
Complexes J. Am. Chem. Soc. 2016, 138, 9077.
(31) Chen, L.; Guo, S.-X.; Li, F.; Bentley, C.; Horne, M.; Bond, A. M.;
Zhang, J. Electrochemical Reduction of CO at Metal Electrodes in a
2
Distillable Ionic Liquid ChemSusChem 2016, 9, 1271.
(32) McCarthy, B. D.; Martin, D. J.; Rountree, E. S.; Ullman, A. C.;
Dempsey, J. L. Electrochemical Reduction of Brønsted Acids by
Glassy Carbon in Acetonitrile—Implications for Electrocatalytic
Hydrogen Evolution Inorg. Chem. 2014, 53, 8350.
(33) Kaljurand, I.; Saame, J.; Rodima, T.; Koppel, I.; Koppel, I. A.;
Kögel, J. F.; Sundermeyer, J.; Köhn, U.; Coles, M. P.; Leito, I.
Experimental Basicities of Phosphazene, Guanidinophosphazene,
and Proton Sponge Superbases in the Gas Phase and Solution J.
Phys. Chem. A 2016, 120, 2591.
(34) Simler, T.; Karmazin, L.; Bailly, C.; Braunstein, P.;
Danopoulos, A. A. Potassium and Lithium Complexes with
Monodeprotonated, Dearomatized PNP and PNCNHC Pincer-Type
Ligands Organometallics 2016, 35, 903.
(35) Francke, R.; Schille, B.; Roemelt, M. Homogeneously
Catalyzed Electroreduction of Carbon Dioxide—Methods,
Mechanisms, and Catalysts Chem. Rev. 2018, 118, 4631.
(36) (a) Buncel, E.; Menon, B. Carbanion mechanisms. 6.
Metalation of arylmethanes by potassium hydride/18-crown-6
ether in tetrahydrofuran and the acidity of hydrogen J. Am. Chem.
Soc. 1977, 99, 4457; (b) Morris, R. H. Brønsted–Lowry Acid
Strength of Metal Hydride and Dihydrogen Complexes Chem. Rev.
2016, 116, 8588.
(
19) (a) Zeng, R.; Feller, M.; Ben-David, Y.; Milstein, D.
Hydrogenation and Hydrosilylation of Nitrous Oxide
Homogeneously Catalyzed by a Metal Complex J. Am. Chem. Soc.
2017, 139, 5720; (b) Mukherjee, A.; Srimani, D.; Ben-David, Y.;
Milstein, D. Low-Pressure Hydrogenation of Nitriles to Primary
Amines Catalyzed by Ruthenium Pincer Complexes. Scope and
mechanism ChemCatChem 2017, 9, 559.
(
20) Li, Z.; Rayder, T. M.; Luo, L.; Byers, J. A.; Tsung, C.-K.
Aperture-Opening Encapsulation of a Transition Metal Catalyst in
a Metal–Organic Framework for CO Hydrogenation J. Am. Chem.
2
Soc. 2018, 140, 8082.
(
21) Wiedner, E. S.; Chambers, M. B.; Pitman, C. L.; Bullock, R. M.;
Miller, A. J. M.; Appel, A. M. Thermodynamic Hydricity of Transition
Metal Hydrides Chem. Rev. 2016, 116, 8655.
(22) Bauer, J. O.; Leitus, G.; Ben-David, Y.; Milstein, D. Direct
Synthesis of Symmetrical Azines from Alcohols and Hydrazine
Catalyzed by a Ruthenium Pincer Complex: Effect of Hydrogen
Bonding ACS Catalysis 2016, 6, 8415.
a 2
(37) The pK of H in THF was first measured in ref 36a and a
(
23) Montag, M.; Zhang, J.; Milstein, D. Aldehyde Binding through
value of 35.5 was obtained. This was later refined to be ~ 49 (ref
36b). This gives a value of 66.8 +/- 1.4 kcal/mol for the free energy
change associated with KH2- which is used throughout this
manuscript. A value of 68.7 kcal/mol was recently calculated and
we report hydricities using the new value reported in the following
reference in parentheses. Brereton, K; Jadrich, C.; Stratakes, B.,
Miller, A. J. M. Organometallics, 2019, accepted.
(38) Tsay, C.; Livesay, B. N.; Ruelas, S.; Yang, J. Y. Solvation Effects
on Transition Metal Hydricity J. Am. Chem. Soc. 2015, 137, 14114.
(39) Lilio, A. M.; Reineke, M. H.; Moore, C. E.; Rheingold, A. L.;
Takase, M. K.; Kubiak, C. P. Incorporation of Pendant Bases into
Reversible C–C Coupling with the Pincer Ligand upon Alcohol
Dehydrogenation by a PNP–Ruthenium Catalyst J. Am. Chem. Soc.
2
012, 134, 10325.
24) Vogt, M.; Gargir, M.; Iron, M. A.; Diskin-Posner, Y.;
Ben-David, Y.; Milstein, D. A New Mode of Activation of CO by
(
2
Metal–Ligand Cooperation with Reversible C-C and M-O Bond
Formation at Ambient Temperature Chem. Eur. J. 2012, 18, 9194.
(25) Kothandaraman, J.; Goeppert, A.; Czaun, M.; Olah, G. A.;
Surya Prakash, G. K. CO capture by amines in aqueous media and
2
its subsequent conversion to formate with reusable ruthenium and
iron catalysts Green Chemistry 2016, 18, 5831.
Rh(diphosphine)
And Catalytic CO
Complexes J. Am. Chem. Soc. 2015, 137, 8251.
(40) Kütt, A.; Selberg, S.; Kaljurand, I.; Tshepelevitsh, S.; Heering,
A.; Darnell, A.; Kaupmees, K.; Piirsalu, M.; Leito, I. pK values in
2
Complexes: Synthesis, Thermodynamic Studies,
+
(
26) For instability of related species with the PNN ligand, see:
2
2 2 2
Hydrogenation Activity of [Rh(P N ) ]
(a) Kohl, S. W.; Weiner, L.; Schwartsburd, L.; Konstantinovski, L.;
Shimon, L. J. W.; Ben-David, Y.; Iron, M. A.; Milstein, D. Consecutive
Thermal H
2
and Light-Induced O
2
Evolution from Water Promoted
a
by a Metal Complex Science 2009, 324, 74; (b) Balaraman, E.;
Gunanathan, C.; Zhang, J.; Shimon, L. J. W.; Milstein, D. Efficient
hydrogenation of organic carbonates, carbamates and formates
indicates alternative routes to methanol based on CO2 and CO Nat
Chem 2011, 3, 609.
organic chemistry – Making maximum use of the available data
Tetrahedron Lett. 2018, 59, 3738.
(41) Kar, S.; Sen, R.; Goeppert, A.; Prakash, G. K. S. Integrative
CO2 Capture and Hydrogenation to Methanol with Reusable
1
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 14
Catalyst and Amine: Toward a Carbon Neutral Methanol Economy
J. Am. Chem. Soc. 2018, 140, 1580.
42) Xie, Y.; Hu, P.; Ben-David, Y.; Milstein, D. A Reversible Liquid
Organic Hydrogen Carrier System Based on
Methanol-Ethylenediamine and Ethylene Urea Angew. Chem. 2019,
5967; (b) Kortunov, P. V.; Siskin, M.; Baugh, L. S.; Calabro, D. C. In
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
Situ Nuclear Magnetic Resonance Mechanistic Studies of Carbon
Dioxide Reactions with Liquid Amines in Non-aqueous Systems:
Evidence for the Formation of Carbamic Acids and Zwitterionic
Species Energy & Fuels 2015, 29, 5940.
(44) Pitman, C. L.; Brereton, K. R.; Miller, A. J. M. Aqueous
Hydricity of Late Metal Catalysts as a Continuum Tuned by Ligands
and the Medium J. Am. Chem. Soc. 2016, 138, 2252.
(
1
31, 5159.
(43) (a) Kortunov, P. V.; Baugh, L. S.; Siskin, M.; Calabro, D. C. In
Situ Nuclear Magnetic Resonance Mechanistic Studies of Carbon
Dioxide Reactions with Liquid Amines in Mixed Base Systems: The
Interplay of Lewis and Brønsted Basicities Energy & Fuels 2015, 29,
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
4
ACS Paragon Plus Environment