A Bimetallic Nickel-Gallium Complex Catalyzes CO2 Hydrogenation via the Intermediacy of an Anionic d10 Nickel Hydride

Article abstract of DOI:10.1021/jacs.7b07911

Large-scale CO₂ hydrogenation could offer a renewable stream of industrially important C₁ chemicals while reducing CO₂ emissions. Critical to this opportunity is the requirement for inexpensive catalysts based on earth-abundant metals instead of precious metals. We report a nickel-gallium complex featuring a Ni(0)→ Ga(III) bond that shows remarkable catalytic activity for hydrogenating CO₂ to formate at ambient temperature (3150 turnovers, turnover frequency = 9700 h^-1), compared with prior homogeneous Ni-centered catalysts. The Lewis acidic Ga(III) ion plays a pivotal role in stabilizing catalytic intermediates, including a rare anionic d¹⁰ Ni hydride. Structural and in situ characterization of this reactive intermediate support a terminal Ni-H moiety, for which the thermodynamic hydride donor strength rivals those of precious metal hydrides. Collectively, our experimental and computational results demonstrate that modulating a transition metal center via a direct interaction with a Lewis acidic support can be a powerful strategy for promoting new reactivity paradigms in base-metal catalysis.

Full text of DOI:10.1021/jacs.7b07911

Subscriber access provided by UNIVERSITY OF ADELAIDE LIBRARIES
Article
2
A bimetallic nickel-gallium complex catalyzes CO hydrogenation
10
via the intermediacy of an anionic d nickel hydride
Ryan C. Cammarota, Matthew V. Vollmer, Jing Xie, Jingyun Ye, John C. Linehan,
Samantha A. Burgess, Aaron M. Appel, Laura Gagliardi, and Connie C Lu
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.7b07911 • Publication Date (Web): 12 Sep 2017
Downloaded from http://pubs.acs.org on September 12, 2017
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 free 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 accessible to all
readers and 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.
Journal of the American Chemical Society 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 30
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
2
A bimetallic nickelꢀgallium complex catalyzes CO hydrogenation via the intermediacy
10
of an anionic d nickel hydride
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
1
1,2
1,2
3
Ryan C. Cammarota, Matthew V. Vollmer, Jing Xie, Jingyun Ye, John C. Linehan,
3
3
1,2
1,
Samantha A. Burgess, Aaron M. Appel, Laura Gagliardi, and Connie C. Lu *
1
Department of Chemistry, University of Minnesota, 207 Pleasant Street SE, Minneapolis, Minnesota
2
55455, USA, Supercomputing Institute, and Chemical Theory Center, University of Minnesota
3
Minneapolis, Minnesota 55455, USA, & Pacific Northwest National Laboratory, P.O. Box 999, MS
K2ꢀ57, Richland, Washington 99352, USA. Corresponding author: C.C.L. (eꢀmail: clu@umn.edu)
Abstract
Largeꢀscale CO hydrogenation could offer a renewable stream of industrially important C
2
1
2
chemicals while reducing CO emissions. Critical to this opportunity is the requirement for
inexpensive catalysts based on earthꢀabundant metals instead of precious metals. We report a
nickelꢀgallium complex featuring a Ni(0)→Ga(III) bond that shows remarkable catalytic
2
activity for hydrogenating CO to formate at ambient temperature (3150 turnovers, turnover
−
1
frequency = 9700 h ), compared with prior homogeneous Niꢀcentered catalysts. The Lewis
acidic Ga(III) ion plays a pivotal role in stabilizing catalytic intermediates, including a rare
10
anionic d Ni hydride. Structural and inꢀsitu characterization of this reactive intermediate
supports a terminal Ni−H moiety, for which the thermodynamic hydride donor strength rivals
those of precious metalꢀhydrides. Collectively, our experimental and computational results
demonstrate that modulating a transition metal center via a direct interaction with a Lewis
acidic support can be a powerful strategy for promoting new reactivity paradigms in baseꢀ
metal catalysis.
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 30
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
Introduction
Efficient recycling of CO
2
to industrial chemicals and liquid fuels, such as formic acid
or methanol, could generate valueꢀadded products from an abundant C1 feedstock while
alleviating adverse effects associated with rising CO levels. Although a CO ꢀtoꢀfuel scheme
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
2
1
would require efficient CO capture and sustainable H production, we focus on the
2
2
singularly formidable challenge of developing baseꢀmetal catalysts for selective CO
2
2
ꢀ5
hydrogenation under mild conditions. CO
2
hydrogenation to methanol remains a daunting
6
ꢀ
task; the few known molecular catalysts rely almost exclusively on precious metals (Ru, Ir),
1
1
6ꢀ10,12
with only a single instance of a baseꢀmetal (Co) catalyst.
Furthermore, baseꢀmetal
13ꢀ17
catalysts remain uncommon for the hydrogenation of CO to formate.
In the past decade,
2
impressive catalytic performance has been achieved using phosphineꢀligated Fe and Co
catalysts, which generate formate with turnover numbers (TONs) from 9000 to 59000 with
18ꢀ20
high turnover frequencies (TOFs).
Despite the aforementioned recent advances of Fe and Co catalysts, the progress of
21ꢀ23
Niꢀbased systems for CO
catalyst was reported over 40 years ago: Ni(dppe)
bis(diphenylphosphino)ethane, produced formate from H
2
hydrogenation has been limited.
, where dppe =
2
and CO , albeit with poor activity
The first homogeneous Ni
2
2
−1 21
(TON = 7, TOF = 0.35 h ). Recently, a waterꢀsoluble Ni bis(diphosphine) catalyst
mediated the H /CO ꢀtoꢀformate reaction in aqueous solution using NaHCO as the base, but
2
2
3
−1
23
the activity remained low (TOF of 0.40(5) h at 80 ºC and 34 atm of H
work, a Ni PCPꢀpincer catalyst generated formate from H and NaHCO
comparatively impressive TON of 3000 at 150 ºC and 54 atm; however, the catalyst was
2
/CO
2
). In related
2
3
with a
24
3 2
inactive when NaHCO was replaced with CO .
2
ACS Paragon Plus Environment

Page 3 of 30
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
The dearth of Ni catalysts for CO
2
hydrogenation stems from several inherent
limitations. Ni, being more electronegative than Fe and Co, has a lower propensity for
25
binding and activating H
2
to generate Ni−H, and once generated, the resulting Ni−H species
are typically weak hydride donors. Furthermore, NiꢀH complexes which are potent enough
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
hydride donors to react with CO tend to form strong Ni−O bonds which prevent formate
2
26
liberation. This lack of strong hydride donors is illustrated by the fact that even for the most
26
+
hydridic Ni−H reported, [HNi(dmpe)
2
] (dmpe = bis(dimethylphosphino)ethane) with a
thermodynamic hydricity (ꢁG°H‾) of 50.7 kcal/mol in CH
3
CN, outerꢀsphere hydride transfer
to CO to generate free formate (c.f. ꢁG° = 44 kcal/mol) is thermodynamically unfavorable
2
H‾
27,28
by ~7 kcal/mol in CH CN.
However, it should be noted that there are several examples of
3
1
NiꢀH complexes which are sufficiently hydridic to react with CO
2
to form η ꢀO formate
26
adducts of Ni. In these cases, the more salient issue which precludes catalytic CO
2
hydrogenation is the formation of strong Ni−O bonds which impede formate release and thus
prevent catalytic turnover. Catalytic liberation of CO reduction products in these systems
2
can be accomplished in some cases using stoichiometric boranes or silanes to form strong
B−O or Si−O bonds which permit Ni−O bond cleavage and drive catalytic turnover. For
example, Ni POCOPꢀpincer complexes can rapidly insert CO
2
into Ni−H bonds and further
to methoxides using stoichiometric borane. The requirement of B−H or Si−H
reductants to drive catalysis, however, is a drawback in that the overall transformations lack
2
9
reduce CO
2
29 ,30
atom economy.
Circumventing the inherent limitations of Ni in CO
2
hydrogenation would involve
31,32
stabilizing Ni−H in more reduced charge states and/or lower Ni oxidation states,
while
33,34
still allowing the regeneration of Ni−H from H
2
and base.
A highly reduced Ni center
could potentially allow for both facile hydride transfer from a Ni−H species to CO , along
2
with a decreased propensity for undesirably strong formate binding in the subsequent step.
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 30
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
Specifically, our strategy involves the incorporation of a Lewis acidic Ga(III) metalloligand
3
5ꢀ38
which acts as a σꢀacceptor toward Ni in NiGaL (1).
We have previously shown that
positioning the supporting Ga(III) ion directly trans to the substrate binding site at Ni allows
2
39,40
for H binding to form a nonꢀclassical H adduct, (η ꢀH )NiGaL, under 1 atm H (Fig. 1).
2
2
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
In this combined experimental and theoretical study, we show that the Ga(III) support plays a
1
0
vital role in stabilizing a d terminal Ni−H species which is strongly hydridic, with an
estimated ꢁG°H‾ value of ~31 kcal/mol in CH CN that rivals those of precious metalꢀhydrides
and allows for spontaneous reaction with CO
3
41
2
to generate and release formate. Taken
together, the supporting Ga(III) ion modulates the properties of Ni so as to promote H2
heterolysis and stabilize a highly reactive Ni−H species, affording a Ni catalyst capable of
2
mediating efficient CO hydrogenation to formate at room temperature.
Results and discussion
Catalysis
Under ambient conditions (1 atm 1:1 H /CO , 293 K), complex 1 catalyzes CO
2
2
2
hydrogenation to formate in 91% yield (0.36 mol% catalyst loading; Table 1, entry 1). A
stoichiometric base is necessary for formate production, and high yields were obtained using
Verkade’s proazaphosphatrane, 2,8,9ꢀtriisopropylꢀ2,5,8,9ꢀtetraazaꢀ1ꢀ
4
2,43
phosphabicyclo[3,3,3]undecane (abbrev. as Vkd).
Catalyst performance was further
optimized by increasing the H /CO pressure to 34 atm and decreasing the catalyst loading by
2
2
tenꢀfold to 0.03 mol%, which gave near quantitative generation of formate (Table 1, entries
ꢀ3). The corresponding kinetic plot in Figure 2 shows that catalyst 1 attained 3150 formate
2
−1
turnovers in ~40 minutes with an initial TOF of 9700 h . The high activity of 1 sharply
21,23
contrasts that of prior Ni homogeneous catalysts (Table S3).
4
ACS Paragon Plus Environment

Page 5 of 30
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
A strong base is necessary for catalysis mediated by 1 based on trials with bases of
42
varying strengths (pK
a
of conjugate acid in CH
3
CN): Vkd (33.6), tꢀbutyl tetramethyl
44
45
guanidine (abbrev. tBuTMG, 26.5), and triethylamine (18.8). Under identical conditions
(
1 mM 1, 34 atm 1:1 H /CO , 293 K), reactions with tBuTMG resulted in a lower yield of
2 2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
formate (80%) and a 30ꢀfold decrease in the maximum rate (Table 1, c.f. entries 2 and 4).
Moreover, an induction period of ca. 3 h was observed for tBuTMG (Fig. 2 inset, SI Fig. S9).
2
Of relevance, an initial induction period was also reported for CO hydrogenation catalyzed
18,46
by HCo(dmpe)
2
using a base of comparable strength to tBuTMG.
3
With NEt , an even
weaker base, no formate was generated (entry 5). Presumably, NEt is not sufficiently basic
3
2
to deprotonate the H adduct, (η ꢀH )NiGaL, precluding formation of the catalytically active
2
2
Ni−H species (Fig. 1).
The striking effect of the supporting Ga(III) ion is appreciated by comparing 1 with
39,47
the isostructural Niꢀonly congener, NiLH
3
(2).
Basically, no formate was generated using
2
as the catalyst (entries 6ꢀ7). Adding GaCl as a coꢀcatalyst with 2 also did not yield any
3
formate (entry 8). Altogether, the catalytic results suggest that an intact Ni→Ga interaction
(vide infra) within our ligand framework provides the requisite tuning effect at Ni to enable
catalysis.
Isolating catalytic intermediates
Because of the remarkable CO hydrogenation activity by 1 relative to known Ni
2
homogeneous catalysts, we sought to elucidate the role of the supporting Ga(III) ion in
2
promoting catalysis. Previously, the H
2
adduct, (η ꢀH
2
)NiGaL, was characterized in situ using
various NMR techniques. In this work, we targeted two later stage intermediates: the Ni(0)
anionic hydride and the Ni(0) anionic formate adduct, which is formed after hydride transfer
to CO (Fig. 1).
2
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 30
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 anionic Ni(0) hydride, [VkdH][HNiGaL] ([VkdH][3]), was generated in situ by
31
subjecting a THFꢀd
8
solution of 1 to 3ꢀ5 equiv. of Vkd and 1 atm of H2. The P NMR
spectrum displayed two new resonances at 76 and −12 ppm in a 3:1 ratio for the ionꢀpaired
−
+
31
[
HNiGaL] and [VkdH] fragments, respectively (Fig. S10). The single P signal of 3 is
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
consistent with retention of C symmetry, while the hydride resonance in the H NMR
3
spectrum appears as a broad peak at −6.4 ppm with no discernible coupling even at low
31
temperatures (Fig. S11ꢀ12). Of note, the P NMR spectrum of a close analogue,
2
[
K(THF)
x
][3], resolves nicely into a doublet with JP−H of 31.6 Hz (Fig. S16). Based on the
spectroscopic data, we propose that the hydride ligand is terminally bound to the Ni center
and resides in the apical pocket trans to the Ga(III) ion. In support, density functional theory
(
DFT) calculations predicted the terminal hydride to be more stable than the bridged
Ni(ꢁ−H)Ga isomer (Fig. S25, Table S6).
To validate the proposed structure of 3, the bis(triphenylphosphine)iminium (PPN)
analogue was independently prepared and investigated by singleꢀcrystal Xꢀray diffraction.
The hydride species was synthesized by adding 1.1 equiv. nBuLi to 1, presumably via an in
26
situ Ni alkyl species that undergoes βꢀhydride elimination. Subsequent salt exchange with
−
[
PPN][tetrakis[3,5ꢀbis(trifluoromethyl)phenyl]borate] ([BArF] ) afforded [PPN][3]. Of note,
the chemical shifts of the hydride and the phosphines in [PPN][3] are nearly identical to those
of [VkdH][3], implying that [PPN][3] is a reasonable model of the catalytically relevant
species (Fig. S15). From a THF/pentane solution of [PPN][3], brightꢀyellow single crystals
were obtained that were suitable for Xꢀray diffraction. The structural characterization of
10
48
[
PPN][3] (Fig. 3) is noteworthy because anionic d hydride complexes are very rare. The
only other mononuclear anionic Ni(0) hydride is [Na(THF) ][HNi(CO) ], which is unstable at
4
3
49
room temperature. Moreover, 3 has catalytic utility in that it can be generated from H and
2
6
ACS Paragon Plus Environment

Page 7 of 30
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
base, which is in sharp contrast to the synthesis of [Na(THF)
4 3
][HNi(CO) ] where
stoichiometric NaHAl(iꢀBu) is employed.
3
In comparison with NiGaL (1), [PPN][3] has subtle structural differences that are
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
instructive to consider. One effect of the hydride donor is that the Ni−Ga bond contracts
39
slightly from 2.3789(8) in 1 to 2.3549(7) Å in [PPN][3] (Fig. 3, Table S5). Hence, the
dative Ni→Ga interaction remains intact upon the introduction of a trans hydride. Perhaps
the most informative parameter is the position of the Ni and Ga centers relative to their
3 3
respective P and N donor sets. In [PPN][3], both Ni and Ga are displaced further above the
P
3
ꢀ and N ꢀplanes by 0.07 Å and 0.17 Å, respectively, than in 1. Intriguingly, the
3
displacement of Ga is more than double that of Ni upon introduction of the hydride ligand
despite the fact that the hydride only interacts directly with Ni. This striking structural feature
underscores the cooperativity of the NiꢀGa unit, as both metals reposition to accommodate
the incoming hydride ligand. A complementary interpretation is that a strong Ni→Ga dative
interaction assists in stabilizing the electronꢀrich, anionic Ni−H moiety.
−
Next, we sought to characterize the anionic Ni formate intermediate, [(HCO
2
)NiGaL]
(
4), which appears in the catalytic cycle following hydride transfer to CO (Fig. 1). The
2
2
formate species, [VkdH][4], was generated in situ by exposure of [VkdH][3] to CO (1 atm)
1
or by addition of excess (~4 equiv.) [VkdH][HCO
2
] to 1. A diagnostic H NMR resonance at
8.68 ppm is assigned to the proton of the coordinated formate (c.f. 8.79 ppm for free
[
VkdH][formate], Fig. S20). This intermediate was also isolated as the PPN ionꢀpair by
exposing [PPN][3] to CO
[PPN][(HCO
2
. The solidꢀstate structure of the resulting complex,
1
)NiGaL] ([PPN][4]), shows an η ꢀO formate ligand and an intact Ni−Ga bond
2
of 2.3789(5) Å, which is essentially identical to that of 1 (Fig. 3, Table S5).
Understanding the Ni(0) hydride intermediate (3)
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 30
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
To gain additional insight into the nature of the Ni(0)−H bond, we undertook a
−
complementary experimental and theoretical investigation of [HNiGaL] . By IR
spectroscopy, a KBr pellet of [PPN][3] displayed a broad band of medium intensity at 1696
−
1
−1
cm , which shifted to 1226 cm upon deuteration (Fig. S26). The Ni−H vibrational
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
frequency was further confirmed by a closely matched value of 1697 cm from DFT
calculations (Fig. S27). Of note, this frequency value is extremely low for terminal hydrides
−1
of Ni, and nearly ties with CpNi(IMes)H (1695 cm ) for the lowest Ni−H stretching
26,50
frequency.
Another useful parameter for comparing the reactivity of metal hydride complexes is
the thermodynamic hydricity (ꢁG° ), or the free energy needed to cleave a M−H bond to
H‾
41,51
−
generate a hydride ion.
The ꢁG°H‾ of [HNiGaL] was experimentally determined by
measuring the H
2
heterolysis equilibrium with NiGaL and Vkd base [equation (1)], in
+
conjunction with the known pK
a
of VkdH [equation (2)] and the H
2
heterolysis constant
2
7,41
[known in CH CN, equation (3)], as shown below.
3
−
+
(
1)
(2)
3)
[HNiGaL] + VkdH ⇌ NiGaL + H + Vkd
1
ꢁG° = −RTln(Keq)
2
+
+
Vkd + H ⇌ VkdH
ꢁG°2 = −1.364(pKa)
ꢁG°3 = +76 kcal/mol
+
−
(
H ⇌ H + H
2
−
−
[
HNiGaL] ⇌ NiGaL + H
1 2 3
ꢁG°H‾ = ꢁG° + ꢁG° + ꢁG°
As hydricity values are typically measured in CH
3 3
CN since ꢁG° is known [equation
−
(3)], one caveat to estimating the absolute hydricity of [HNiGaL] is that K of equation (1)
eq
was measured in THF due to complications arising from poor solubility of NiGaL in CH
and competitive binding between H and CH CN (see SI). Keq of equation (1) was measured
to be 0.16 in THF based on two independent trials employing different base concentrations. If
3
CN
2
3
K
3
eq of equation (1) is comparable in THF and CH CN solvents, one can estimate ꢁG°H‾ ≈
8
ACS Paragon Plus Environment

Page 9 of 30
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
−
31.3(5) kcal/mol for [HNiGaL] in CH
3
CN. Alternatively, one can rigorously measure the
−
difference in ꢁG°H‾ values between [HNiGaL] and H
difference, or ꢁG°H‾(3)−ꢁG°H‾(H ), is −37.1(6) kcal/mol in THF, and of interest, is
comparable to ꢁG° (3)−ꢁG° (H ) in CH CN of −45 kcal/mol, using the estimated ꢁG° (3)
2
in THF [equations (1)ꢀ(2)]. This
2
H‾
H‾
2
3
H‾
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
of 31 kcal/mol.
To further test the assumption that measuring Keq of equation (1) in THF can lead to
meaningful comparisons with ꢁG°H‾ values obtained wholly in CH CN, we probed the
hydrideꢀtransfer equilibrium between 1 and HCo(dmpe) (known ꢁG°H‾ = 36 kcal/mol in
CH CN). No hydride transfer to NiGaL was detected in THF over 10 days, suggesting that
the ꢁG°H‾ value of [HNiGaL] is <33 kcal/mol (Fig. S33). In the reverse direction, essentially
3
2
3
−
−
+
full consumption of [HNiGaL] and [Co(dmpe)
2 3 2
] to give (CH CN)NiGaL and HCo(dmpe)
was observed within 3 days in 3:1 CH CN:THF, confirming that the lack of hydride transfer
3
between HCo(dmpe)
2
and NiGaL was due to the greater thermodynamic stability of
−
HCo(dmpe) relative to [HNiGaL] rather than sluggish hydride transfer kinetics.
2
−
Additionally, DFT studies of isodesmic hydride transfer between [HNiGaL] and
2
+
−
2 3
[Ni(dmpe) ] in CH CN predicted a hydricity of 28ꢀ31 kcal/mol for [HNiGaL] that matches
34
the experimental estimate (Table S11). The importance of Ga(III) for stabilizing the anionic
−
3
Ni−H is, perhaps, underscored by our inability to synthetically generate the [H−NiLH ]
analogue. We also note that we have not observed H binding to NiLH even at high pressure
2
3
and low temperature (34 atm H
2
, 193K) and have not observed H
2
deprotonation to give a
−
[
H−NiLH
3
] analogue using excess strong base (Vkd base or potassium tertꢀbutoxide).
+
For comparison, [HNi(diphosphine)
2
] complexes have hydricity values that range
4
1
−
from 50 to 68 kcal/mol. With a considerably lower ꢁG°H‾ value (31 kcal/mol), [HNiGaL]
is, by a wide margin, the strongest hydride donor of any Ni−H reported to date. The
−
exceptional hydricity of [HNiGaL] can be attributed to its anionic charge and the zeroꢀvalent
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 30
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
6
oxidation state of Ni, which is distinct from the cationic Ni(II) hydrides in the literature.
−
Notably, [HNiGaL] is among the strongest hydride donors of any firstꢀrow metal (c.f.
52
HCo(P
4
N
2
), ꢁG°H‾ = 31.8 kcal/mol), and is even on par with some of the more hydridic
4
1
precious metal complexes, (e.g. HRh(diphosphine) , ꢁG° = 26ꢀ34 kcal/mol). Of relevance
2
H‾
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
−
to catalysis, the low ꢁG° of [HNiGaL] would allow for direct outerꢀsphere hydride transfer
H‾
2
8
to CO
2
with a driving force of ~13 kcal/mol. It should be noted, however, that the driving
force in catalysis is likely greater for the conversion of [VkdH][3] and CO
2
to [VkdH][4] due
to enthalpic contributions from subsequent Ni−O bond formation in the formate adduct after
initial hydride transfer (vide infra).
10
48
Perhaps, the most unexpected feature of 3 is its stability as an anionic d hydride. A
10
−
simple bonding analysis between a d metal and H would require the M−H σꢀantibonding
orbital to be fully populated. Natural orbitals obtained from complete active space selfꢀ
53
consistent field (CASSCF) calculations, however, revealed a distinctly different bonding
scheme. While the five Ni 3d orbitals are indeed doubly occupied, consistent with Ni(0), they
show essentially no bonding to the hydride. Rather, the natural orbital involved in Ni−H σꢀ
bonding has both metal contributions [29% Ni, primarily 4p
hydridic character [51% H(1s)] (Fig. 4, Table S13).
z
; 10% Ga] and substantial
We further propose that the Ni→Ga interaction is critical for stabilizing the Ni(0)−H
bond because of the symbiotic nature of two σꢀbonding interactions: (1) donation of hydride
to Ni, via H(1s)→ Ni(4p
z
/4s), and (2) donation from Ni to Ga, via Ni(3dz2)→Ga(4p /4s). This
z
stabilization of the Ni(0)−H by Ga(III) can be interpreted as an inverse trans influence
54
55,56
exerted by the Ga metalloligand, which acts as a σꢀacceptor
toward Ni and strengthens
the Ni−H bond directly trans to it (“pullꢀpull”). In support, a natural bond orbital analysis
shows that the Ni→Ga stabilization energy increases by 10 kcal/mol from NiGaL (1) to
−
57
[
HNiGaL] (3) (Fig. S35, Table S14).
10
ACS Paragon Plus Environment

Page 11 of 30
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Mechanistic insights
A simple catalytic cycle is proposed (Fig. 1): (1) H
2
2
binding to NiGaL forms (η ꢀ
−
2 2
H )NiGaL, (2) deprotonation by base generates [HNiGaL] , (3) hydride transfer to CO
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
−
forms [(η ꢀHCO
2
)NiGaL] , and (4) liberation of formate regenerates NiGaL. This catalytic
+
mechanism is commonly proposed, albeit typically with M(H) and a neutral M−H as the
2
2
3
catalytic intermediates instead of M(η −H
The feasibility of the proposed mechanism is further demonstrated by complementary
reactivity studies. Initial binding of H is favored as 1 showed no propensity to bind CO
2
) and the anionic M−H proposed here.
2
2
2
−
even under 34 atm. Deprotonation of (η ꢀH )NiGaL to generate [HNiGaL] occurred readily
2
THF
2
in the presence of H
2
(1 atm) and excess Vkd (Fig. S10ꢀ11). Of interest, the pK
a
of (η ꢀ
2
H
H
2
2
)NiGaL was measured to be 27.5(2) via the protonꢀtransfer equilibrium between (η ꢀ
2
)NiGaL and Vkd in the overall H heterolysis reaction [equation (1), Table S9]. This
−
supports the hypothesis that a strong base is necessary to form the active species, [HNiGaL] .
THF
58
For comparison, the pKa
increases by 21 pK
upon binding to Ni, which allows for deprotonation to give the catalytically relevant anionic
hydride, is not observed for NiLH , rendering it essentially catalytically inactive (vide supra).
Finally, the reaction between [VkdH][HNiGaL] and CO (1 atm) was observed to form
VkdH][(HCO )NiGaL] (Fig. S22). Additionally, isotopic labelling studies confirmed that
of H is estimated to be 49, which means that the acidity of H
2
2
a
units upon binding to NiGaL. This drastic increase in the acidity of H
2
3
2
[
2
13
2 2
formate was derived from D and CO (Fig. S38).
Monitoring the Ni speciation during catalysis also provided unique insights.
Throughout catalysis, the observed catalyst resting state is the formate species, [VkdH][4]
(Fig. S36). From [VkdH][4], regeneration of the catalytically active [VkdH][3] was observed
to occur upon the addition of H (1 atm) in the presence of Vkd (Fig. S37). The buildup of the
2
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 30
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
formate complex during catalysis suggests the rateꢀlimiting step is the liberation of
[
VkdH][HCO
2
] to regenerate NiGaL. In highly active Fe and Co pincer systems with TONs
4
+
of ~10 , formate dissociation is also sluggish, and substoichiometric Li additives are needed
19,20
to promote formate loss and drive catalytic turnover.
In the present system, no additives
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
were used. Presumably, formate extrusion is enhanced by the overall anionic charge of the Ni
formate intermediate and the weaker Ni(0)−OCHO bond.
Conclusion
In summary, we have shown that a Lewis acidic Ga(III) supporting ion can interact
with Ni to enable direct CO hydrogenation to formate at ambient temperature with excellent
2
ꢀ1
yields. The high formate turnovers (3150) and TOF (9700 h ) are unprecedented for Ni and
respectable compared to stateꢀofꢀtheꢀart homogeneous firstꢀrow metal catalysts (i.e. Fe, Co).
−
The Ga(III) support plays a pivotal role in stabilizing the unusual anionic [HNiGaL] species,
which is by far the strongest Ni−H donor reported to date (ꢁG°H‾ ≈ 31 kcal/mol) and on par
with the most hydridic firstꢀrow and many precious metal hydrides. The significance of the
Lewis acidic Ga(III) support is further emphasized via comparison to NiLH
3
, which is
toward
essentially inactive for CO hydrogenation due to its inability to bind and activate H
2
2
deprotonation to stabilize an analogous anionic hydride. Future studies will further elucidate
the catalytic mechanism and detail the roles of the Ga(III) supporting ion therein, as well as
investigate the influence of other group 13 Lewis acidic supports on Niꢀcatalyzed CO2
hydrogenation.
Experimental Section
Highꢀpressure catalytic reactions
2 2 2
CO hydrogenation reactions at 34 atm of ~1:1 H /CO were performed in PEEK
highꢀpressure NMR spectroscopy tubes designed and built at Pacific Northwest National
12
ACS Paragon Plus Environment

Page 13 of 30
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
5
9
Laboratory, as reported previously. In a typical catalytic experiment (0.25 mM NiGaL (1),
00 mM Vkd base), a stock solution of 1 (11.8 mg, 14.6 ꢂmol) in 1 mL THFꢀd was
prepared. A 0.875 M solution of Vkd (168.3 mg, 560 ꢂmol) in 640 ꢂL THFꢀd was also
8
8
8
prepared. From these stock solutions, a catalyst mixture stock solution was prepared (700 ꢂL
of 800 mM Vkd, 0.25 mM NiGaL). Note that preparations were adjusted accordingly to
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
afford other concentrations of NiGaL (or NiLH
NEt . Lastly, 300 ꢂL of the reaction solution was added to two different PEEK NMR
spectroscopy cells.
3
) and/or other bases, including tBuTMG and
3
The PEEK cell was sealed and connected to a purged highꢀpressure line equipped
with a vacuum pump and an ISCO syringe pump. The headspace was degassed by opening
the PEEK cell to static vacuum (3 × 10 s). Gas was delivered to the cell from an ISCO
syringe pump running constantly at 34 atm (i.e. continuous gas feed). The contents of the
PEEK cells were mixed using a vortex mixer for approximately 3 minutes until the pressure
stabilized. After stabilization, the cell was inserted into the NMR spectrometer periodically to
acquire data over time. The contents of the PEEK cell were mixed using a vortex mixer when
NMR data was not being collected to promote optimal gasꢀliquid mixing throughout
catalysis. The concentration of formate was determined by integrating the formate resonance
2 2
relative to the residual HDO resonance in the CoCl in D O capillary standard.
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 30
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
i
Synthesis of [N(P(C
6
H
5
)
3
)
2
][HNiGa(N(oꢀ(NCH
2
P Pr
2
)C
6
H
4
)
3
)] (abbrev. as [PPN][3])
A stirring solution of NiGaL (114.1 mg, 138 µmol) in 8 mL THF was cooled to –78˚C over
15 minutes. To this solution, 62.4 µL (156 µmol) of nꢀBuLi solution (2.5 M in hexanes) was
added, and the color rapidly changed from deepꢀred to redꢀyellow. This solution was allowed
to warm to room temperature over the course of 12 h. Subsequent salt exchange was
performed by adding solid bis(triphenylphosphine)iminium tetrakis[3,5ꢀ
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
bis(trifluoromethyl)phenyl]borate ([PPN][BArF24], 415 mg, 283 µmol), and the solution was
concentrated in vacuo, resulting in the precipitation of solids. Et
volume was ~20 mL and the mixture was stirred vigorously for 3 h, affording a bright yellow
precipitate. This precipitate was washed with Et O (2 x10 mL) to give [PPN][3] as a bright
2
O was added until the total
2
yellow powder (101 mg, 731 ꢂmol, 40% yield). Single crystals of [PPN][3] suitable for Xꢀray
diffraction were grown by layering a THF solution with pentane. Additionally, the analogous
deuteride complex, [PPN][3(D)], was synthesized in situ by overnight exposure of [PPN][3]
(
10 mg, 7.4 µmol) to D (4 atm) in THFꢀd in a J. Young NMR tube.
2
8
1
31
H{ P} NMR (ppm, THFꢀd
8
, 400 MHz): 7.62 (t, J=7.4 Hz, PPN, 6H), 7.56 (d, J=7.7 Hz,
PPN, 12H), 7.44 (t, J=7.5 Hz, PPN, 12H), 7.12 (d, J=7.4 Hz, aryl, 3H), 6.56 (t, J=7.5 Hz,
i
aryl, 3H), 6.14 (d, J=7.9 Hz, aryl, 3H), 5.94 (t, J=7.3 Hz, aryl, 3H), 2.60 (br, CH
2
P( Pr)
2
, 6H),
1.77 (br, CH(CH
3
)2, 3H), 1.56 (br, C′H(CH
3
)
3 2
2, 3H), 1.05−0.60 (br, CH(CH ) , 36H), −6.45
31
1
(
br, NiH, 1H, T (min)≤0.84(4) s at 233K). P{ H} NMR (ppm, THFꢀd , 162 MHz): 75.3 (s,
1
8
−
−1
3
P, HNiGaL ), 21.0 (s, 2P, PPN). IR (KBr pellet, cm ): ν(Ni−H) 1696 [1226 for 3(D)]. Anal.
NiGa] (%): C, 66.93; H, 6.82; N, 5.20. Anal.
] (%): C, 65.38; H, 6.66; N, 5.08. Found: C, 65.24; H, 6.90; N, 4.86.
calcd for [PPN][3], [C36
2 61 4 3
H30NP ][C39H N P
calcd for [PPN][3·O
2
Elemental analysis consistently showed a low carbon percentage that could be consistent with
oxidation by one equivalent of O2.
14
ACS Paragon Plus Environment

Page 15 of 30
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Computational methods
Density functional theory (DFT) and ab initio wave function calculations were performed for
anionic complexes 3 and 4. Geometries were fully optimized in the gas phase using Gaussian
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
60
61
62
0
9 software and the M06ꢀL local functional with def2ꢀseries basis sets, denoted as M06ꢀ
L/DEF2 (see SI for details). Vibrational frequency calculations were performed to confirm
the stationary point nature of the structures. Solvation effects were considered by performing
6
3
singleꢀpoint calculations for all stationary points using the SMD solvation model. Isodesmic
hydride transfer reactions were modeled by DFT using several different functionals to
benchmark the hydricity of 3 relative to those of other previously reported MꢀH (see Tables
3
3
S10ꢀ11). Complex 3 was further investigated by ab initio calculations using the complete
5
3
active space selfꢀconsistent field (CASSCF) method. CASSCF calculations were performed
6
4
on the M06ꢀL/DEF2 optimized structures using the MOLCASꢀ8.1 package with relativistic
basis sets of atomic natural orbitals types, i.e. ANOꢀRCCꢀVDZ were used for N, P, C, H
65
atoms, and ANOꢀRCCꢀVTZ were used for Ni and Ga atoms. The Ni→Ga dative bond was
also interrogated by calculating the donorꢀacceptor stabilization energies for 3 based on
57
natural bond orbital (NBO) analysis.
Supporting Information
The Supporting Information is available free of charge on the ACS Publications website at
DOI:
Additional data for catalysis studies, synthesis of 4, spectroscopic characterization,
computational studies, and hydricity measurements. XYZ files for DFTꢀoptimized
geometries. (PDF)
Crystallographic data for [PPN][3] and [PPN][4] (CIF)
(These cif files have been deposited to the Cambridge CCDC 1553935ꢀ1553936).
15
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 30
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
Corresponding Author
clu@umn.edu
Notes
*
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 authors declare no competing financial interests.
Acknowledgements
R.C.C. and M.V.V. were supported by DOE Office of Science Graduate Student Research
and National Science Foundation (NSF) Graduate Research Fellowship programs,
respectively. J.X., J.Y., and L.G. were supported as part of the Inorganometallic Catalyst
Design Center, an Energy Frontier Research Center funded by the U.S. Department of Energy
(DOE), Office of Science, Basic Energy Sciences under Award DEꢀSC0012702. J.C.L.,
S.A.B., and A.M.A. were supported by the U.S. Department of Energy, Office of Science,
Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences &
Biosciences. C.C.L. acknowledges NSF (CHEꢀ1665010) for support.
16
ACS Paragon Plus Environment

Page 17 of 30
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2 2
Figure 1. General catalytic scheme for H /CO to formate by NiGaL (1). A key step is the
2
baseꢀassisted H
2
heterolysis of (η ꢀH
2
)NiGaL to form the anionic Ni hydride intermediate,
−
[
HNiGaL] .
Figure 2. Formate turnovers versus time plots for CO hydrogenation reactions catalyzed by
2
1
and 2 (0.25 mM catalyst, 800 mM Vkd, 1:1 H
each; see Table 1, entries 3 and 7). Inset shows the kinetic plots for 1 (1 or 4 mM) with
various bases (Vkd, tBuTMG, and NEt ; see Table 1, entries 2, 4, and 5). Full kinetic profiles
of all trials are shown in Figures S5ꢀS9.
2 2
/CO ) of 34 atm, 293 K, avg. of two trials for
3
17
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 30
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 3. Solidꢀstate structures of [PPN][HNiGaL] ([PPN][3], left) and
[
2
PPN][(HCO )NiGaL] ([PPN][4], right) displayed with 50% thermal ellipsoids. The PPN
cation and H atoms, except for those on or interacting with the apical ligands, were omitted
for clarity. A nonꢀclassical hydrogenꢀbonding interaction, O(formate)ꢀꢀꢀHC(methine), is
shown in 4 (light gray line). Relevant structural parameters of [PPN][3] and [PPN][4] are
shown in Tables S4ꢀS5.
−
Figure 4. Selected natural orbitals obtained from a CASSCF calculation of [HNiGaL] (3).
Occupation numbers (shown in parentheses) indicate that these six orbitals are doubly
occupied. Refer to Figure S34 and Table S13 for additional details.
18
ACS Paragon Plus Environment

Page 19 of 30
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Key
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
19
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 30
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
Table 1. Catalytic CO
2
hydrogenation to formate using NiGaL (1) or NiLH
3
(2) with various
bases. Reaction conditions: catalyst (0.25 to 4 mM), 800 mM base, 1:1 H
2
/CO (1 or 34 atm),
2
a
8
0.30 mL THFꢀd solution, 293 K.
−1
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
[catalyst]
P(H
2
/CO
2
)
theor.
TON
275
actual
%
TOF (h )
entry
catalyst
base
b
c
d
e
(
mM)
2.9
1
(atm)
1
TON
yield
91
initial
overall
41
f
1
1
1
1
1
1
2
2
Vkd
Vkd
Vkd
250
800
3150
640
0
67
2
3
4
34
34
34
34
1
800
quant.
99
3680
9700
2130
6900
74
0.25
1
3200
800
g
tBuTMG
NEt
80
120
h
5
4
3
200
0
N.A.
N.A.
0.04
N.A.
N.A.
f
i
6
2.9
0.25
0.25
Vkd
Vkd
Vkd
275
0.8
0
0.3
0
0.14
7
34
34
3200
3200
N.A.
N.A.
8
2 + GaCl
3
0
0
a
Conditions apply to all entries unless otherwise noted. TON, % yield, and TOF are reported as averages of two
b
1
−
trials. See SI Table S2 for details. TON based on H NMR integration of HCO
2
relative to an internal capillary
c
1
−
standard. % yield based on TON/maximum TON, which closely matched H NMR integration of HCO
2
+
d
relative to H(base) . Initial TOF is the initial linear slope of the formate turnovers vs. time plot (initial ~40 min
e
f
at 1 atm, or 6ꢀ10 min at 34 atm). Overall TOF = turnovers/time for >90% of final yield achieved. 0.40 mL
g
volume for 1 atm runs. An induction period was observed. Initial TOF defined from ~3.5 h to 7.5 h over which
h
turnovers/time is linear (after induction period). Single run in 0.75 mL THF in a highꢀpressure vessel. No
−
i
−
HCO
2 2
was detected after 150 h at 323 K. Initial TOF determined for the first time point that HCO was
observed (~3.5 h).
20
ACS Paragon Plus Environment

Page 21 of 30
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
References
(
(
1)
2)
Olah, G. A.; Prakash, G. K. S.; Goeppert, A. J. Am. Chem. Soc. 2011
Appel, A. M.; Bercaw, J. E.; Bocarsly, A. B.; Dobbek, H.; DuBois, D. L.; Dupuis, M.; Ferry,
, 133, 12881ꢀ12898.
J. G.; Fujita, E.; Hille, R.; Kenis, P. J. A.; Kerfeld, C. A.; Morris, R. H.; Peden, C. H. F.; Portis, A. R.;
Ragsdale, S. W.; Rauchfuss, T. B.; Reek, J. N. H.; Seefeldt, L. C.; Thauer, R. K.; Waldrop, G. L.
Chem. Rev. 2013
(3) Wang, W.ꢀH.; Himeda, Y.; Muckerman, J. T.; Manbeck, G. F.; Fujita, E. Chem. Rev. 2015
15, 12936ꢀ12973.
4) Klankermayer, J.; Wesselbaum, S.; Beydoun, K.; Leitner, W. Angew. Chem. Int. Ed. 2016
, 7296ꢀ7343.
, 113, 6621ꢀ6658.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
,
1
(
,
55
(
(
5)
6)
Aresta, M.; Dibenedetto, A. Dalton Trans. 2007, 2975ꢀ2992.
Wesselbaum, S.; vom Stein, T.; Klankermayer, J.; Leitner, W. Angew. Chem. Int. Ed. 2012
,
51, 7499ꢀ7502.
(7) Wesselbaum, S.; Moha, V.; Meuresch, M.; Brosinski, S.; Thenert, K. M.; Kothe, J.; Stein, T.
v.; Englert, U.; Holscher, M.; Klankermayer, J.; Leitner, W. Chem. Sci. 2015 , 693ꢀ704.
133, 18122ꢀ18125.
Kothandaraman, J.; Goeppert, A.; Czaun, M.; Olah, G. A.; Prakash, G. K. S. J. Am. Chem.
138, 778ꢀ781.
Rezayee, N. M.; Huff, C. A.; Sanford, M. S. J. Am. Chem. Soc. 2015
Sordakis, K.; Tsurusaki, A.; Iguchi, M.; Kawanami, H.; Himeda, Y.; Laurenczy, G. Chem.
Eur. J. 2016 22, 15605ꢀ15608.
12)
017
13)
,
6
(
(
8)
9)
Huff, C. A.; Sanford, M. S. J. Am. Chem. Soc. 2011,
Soc. 2016
10)
(11)
,
(
, 137, 1028ꢀ1031.
,
(
2
Schneidewind, J.; Adam, R.; Baumann, W.; Jackstell, R.; Beller, M. Angew. Chem. Int. Ed.
, 56, 1890ꢀ1893.
(
Federsel, C.; Boddien, A.; Jackstell, R.; Jennerjahn, R.; Dyson, P. J.; Scopelliti, R.;
Laurenczy, G.; Beller, M. Angew. Chem. Int. Ed. 2010
(14) Langer, R.; DiskinꢀPosner, Y.; Leitus, G.; Shimon, L. J. W.; BenꢀDavid, Y.; Milstein, D.
Angew. Chem. Int. Ed. 2011 50, 9948ꢀ9952.
15) Ziebart, C.; Federsel, C.; Anbarasan, P.; Jackstell, R.; Baumann, W.; Spannenberg, A.; Beller,
M. J. Am. Chem. Soc. 2012 134, 20701ꢀ20704.
16) Federsel, C.; Ziebart, C.; Jackstell, R.; Baumann, W.; Beller, M. Chem. Eur. J. 2012
5.
17)
(18)
11533ꢀ11536.
19) Zhang, Y.; MacIntosh, A. D.; Wong, J. L.; Bielinski, E. A.; Williard, P. G.; Mercado, B. Q.;
Hazari, N.; Bernskoetter, W. H. Chem. Sci. 2015 , 4291ꢀ4299.
Spentzos, A. Z.; Barnes, C. L.; Bernskoetter, W. H. Inorg. Chem. 2016
Inoue, Y.; Izumida, H.; Sasaki, Y.; Hashimoto, H. Chem. Lett. 1976 , 863ꢀ864.
Tai, C.ꢀC.; Chang, T.; Roller, B.; Jessop, P. G. Inorg. Chem. 2003 42, 7340ꢀ7341.
Burgess, S. A.; Kendall, A. J.; Tyler, D. R.; Linehan, J. C.; Appel, A. M. ACS Catal. 2017
089ꢀ3096.
24) Enthaler, S.; Brück, A.; Kammer, A.; Junge, H.; Irran, E.; Gülak, S. ChemCatChem 2015
, 49, 9777ꢀ9780.
,
(
,
(
7
, 18, 72ꢀ
(
Zell, T.; Milstein, D. Acc. Chem. Res. 2015
, 48, 1979ꢀ1994.
Jeletic, M. S.; Mock, M. T.; Appel, A. M.; Linehan, J. C. J. Am. Chem. Soc. 2013
,
135,
(
,
6
(
(
(
20)
21)
22)
, 55, 8225ꢀ8233.
,
5
,
(23)
3
(
, 7,
, 7,
65ꢀ69.
(
(
25)
26)
Harman, W. H.; Peters, J. C. J. Am. Chem. Soc. 2012
Eberhardt, N. A.; Guan, H. Chem. Rev. 2016 116, 8373ꢀ8426.
, 134, 5080ꢀ5082.
,
(27)
Curtis, C. J.; Miedaner, A.; Ellis, W. W.; DuBois, D. L. J. Am. Chem. Soc. 2002, 124, 1918ꢀ
1925.
21
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 30
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
(28)
(29)
DuBois, D. L.; Berning, D. E. Appl. Organomet. Chem. 2000
Chakraborty, S.; Zhang, J.; Krause, J. A.; Guan, H. J. Am. Chem. Soc. 2010
A Ni dicarbene catalyst: Lu, Z.; Williams, T. J. ACS Catal. 2016, 6670ꢀ6673.
Zimmermann, P.; Limberg, C. J. Am. Chem. Soc. 2017 139, 4233ꢀ4242.
Yoo, C.; Lee, Y. Angew. Chem. Int. Ed. 2017 56, 9502ꢀ9506.
Mondal, B.; Neese, F.; Ye, S. Inorg. Chem. 2016 55, 5438ꢀ5444.
Qi, X.ꢀJ.; Fu, Y.; Liu, L.; Guo, Q.ꢀX. Organometallics 2007 26, 4197ꢀ4203.
Rudd, P. A.; Liu, S.; Gagliardi, L.; Young, V. G.; Lu, C. C. J. Am. Chem. Soc. 2011
0724ꢀ20727.
36) Bouhadir, G.; Bourissou, D. Chem Soc Rev 2016
refs. 37ꢀ38 for recent examples of bifunctional catalysts with a sigma acceptor ligand at the metal.
, 14, 860ꢀ862.
,
132, 8872ꢀ8873.
(
(
(
(
(
30)
31)
32)
33)
34)
,
,
,
,
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
(35)
2
(
, 133,
, 45, 1065ꢀ1079, and references therein. See
(
37)
Takaya, J.; Iwasawa, N. J. Am. Chem. Soc. 2017
You, D.; Gabbaï, F. P. J. Am. Chem. Soc. 2017 139, 6843ꢀ6846.
Cammarota, R. C.; Lu, C. C. J. Am. Chem. Soc. 2015 137, 12486ꢀ12489.
Cammarota, R. C.; Clouston, L. J.; Lu, C. C. Coord. Chem. Rev. 2017 334, 100ꢀ111.
Wiedner, E. S.; Chambers, M. B.; Pitman, C. L.; Bullock, R. M.; Miller, A. J. M.; Appel, A.
M. Chem. Rev. 2016 116, 8655ꢀ8692.
Kisanga, P. B.; Verkade, J. G.; Schwesinger, R. J. Org. Chem. 2000
Wróblewski, A. E.; Pinkas, J.; Verkade, J. G. Main Group Chem. 1995
Zall, C. M.; Linehan, J. C.; Appel, A. M. ACS Catal. 2015 , 5301ꢀ5305.
Kaljurand, I.; Kütt, A.; Sooväli, L.; Rodima, T.; Mäemets, V.; Leito, I.; Koppel, I. A. J. Org.
Chem. 2005 70, 1019ꢀ1028.
46) Jeletic, M. S.; Helm, M. L.; Hulley, E. B.; Mock, M. T.; Appel, A. M.; Linehan, J. C. ACS
Catal. 2014 , 3755ꢀ3762.
47) Clouston, L. J.; Siedschlag, R. B.; Rudd, P. A.; Planas, N.; Hu, S.; Miller, A. D.; Gagliardi,
L.; Lu, C. C. J. Am. Chem. Soc. 2013 135, 13142ꢀ13148.
(48) Darensbourg, M. Y.; Ash, C. E. In Adv. Organomet. Chem.; Stone, F. G. A., West, R., Eds.;
Academic Press, Inc.: San Diego, CA, 1987
, 139, 6074ꢀ6077.
(
38)
,
(39)
(40)
,
,
(
41)
,
(
(
(
42)
43)
44)
,
65, 5431ꢀ5432.
, 69ꢀ79.
,
1
,
5
(45)
,
(
,
4
(
,
.
(
(
49)
50)
Kleimann, W.; Pörschke, K.ꢀR.; Wilke, G. Chem. Ber. 1985, 118, 323ꢀ331.
Abernethy, C. D.; Baker, R. J.; Cole, M. L.; Davies, A. J.; Jones, C. Transition Met. Chem.
28, 296ꢀ299.
2003
,
(
(
51)
52)
Berning, D. E.; Noll, B. C.; DuBois, D. L. J. Am. Chem. Soc. 1999
Wiedner, E. S.; Appel, A. M.; DuBois, D. L.; Bullock, R. M. Inorg. Chem. 2013
,
121, 11432ꢀ11447.
52, 14391ꢀ
,
14403.
(53)
Roos, B. O.; Taylor, P. R. Chem. Phys. 1980
Vollmer, M. V.; Xie, J.; Lu, C. C. J. Am. Chem. Soc. 2017
Parkin, G. Organometallics 2006 25, 4744ꢀ4747.
Amgoune, A.; Bourissou, D. Chem. Commun. 2011
Weinhold, F.; Landis, C. R. Chem. Edu. Res. Pract. 2001
AbdurꢀRashid, K.; Fong, T. P.; Greaves, B.; Gusev, D. G.; Hinman, J. G.; Landau, S. E.;
122, 9155ꢀ9171.
Yonker, C. R.; Linehan, J. C. J. Organomet. Chem. 2002 650, 249ꢀ257.
Frisch, M.; Trucks, G.; Schlegel, H.; Scuseria, G.; Robb, M.; Cheeseman, J.; Scalmani, G.;
, 48, 157ꢀ173.
(
(
(
(
(
54)
55)
56)
57)
58)
, 139, 6570ꢀ6573.
,
,
47, 859ꢀ871.
, 91ꢀ104.
,
2
Lough, A. J.; Morris, R. H. J. Am. Chem. Soc. 2000
,
(
(
59)
60)
,
Barone, V.; Mennucci, B.; Petersson, G., Gaussian 09, revision D. 01. Gaussian, Inc., Wallingford
CT: 2009.
(
61)
Zhao, Y.; Truhlar, D. G. J. Chem. Phys. 2006, 125, 194101.
(62)
(63)
Weigend, F.; Ahlrichs, R. PCCP 2005 , 3297ꢀ3305.
, 7
Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B 2009, 113, 6378ꢀ6396.
22
ACS Paragon Plus Environment

Page 23 of 30
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
(64)
Aquilante, F.; Autschbach, J.; Carlson, R. K.; Chibotaru, L. F.; Delcey, M. G.; De Vico, L.;
Ferré, N.; Frutos, L. M.; Gagliardi, L.; Garavelli, M. J. Comput. Chem. 2016, 37, 506ꢀ541.
(
65)
Roos, B. O.; Lindh, R.; Malmqvist, P.ꢀÅ.; Veryazov, V.; Widmark, P.ꢀO. J. Phys. Chem. A
, 108, 2851ꢀ2858.
2004
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
23
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 30
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
TOC Graphic
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
24
ACS Paragon Plus Environment

Page 25 of 30
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1
103x90mm (600 x 600 DPI)
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 26 of 30
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 2
8
4x65mm (300 x 300 DPI)
ACS Paragon Plus Environment

Page 27 of 30
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3
84x44mm (300 x 300 DPI)
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 28 of 30
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 4
7
1x61mm (300 x 300 DPI)
ACS Paragon Plus Environment

Page 29 of 30
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Key
4
0x19mm (300 x 300 DPI)
ACS Paragon Plus Environment