Conformational twisting of a formate-bridged diiridium complex enables catalytic formic acid dehydrogenation

Article abstract of DOI:10.1039/c8dt03268h

We previously reported that iridium complex 1a enables the first homogeneous catalytic dehydrogenation of neat formic acid and enjoys unusual stability through millions of turnovers. Binuclear iridium hydride species 5a, which features a provocative C

Full text of DOI:10.1039/c8dt03268h

View Article Online
View Journal
Dalton
Transactions
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: P. J. Lauridsen, Z.
Lu, J. J. A. Celaje, E. A. Kedzie and T. Williams, Dalton Trans., 2018, DOI: 10.1039/C8DT03268H.
This is an Accepted Manuscript, which has been through the
Volume 45 Number
1 7 January 2016 Pages 1–398
Royal Society of Chemistry peer review process and has been
accepted for publication.
Dalton
Transactions
Accepted Manuscripts are published online shortly after
An international journal of inorganic chemistry
www.rsc.org/dalton
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1477-9226
PAPER
Joseph T. Hupp, Omar K. Farha et al.
Efficient extraction of sulfate from water using
framework
a Zr-metal–organic
rsc.li/dalton

Page 1 of 8
PleasDeadltoonnoTtraandsjaucsttiomnsargins
View Article Online
DOI: 10.1039/C8DT03268H
Journal Name
ARTICLE
Conformational Twisting of a Formate-Bridged Diiridium Complex
Enables Catalytic Formic Acid Dehydrogenation
Paul J. Lauridsen, Zhiyao Lu, Jeff J. A. Celaje, Elyse A. Kedzie, Travis J. Williams*
Received 00th January 20xx,
Accepted 00th January 20xx
We previously reported that iridium complex 1a enables the first homogeneous catalytic dehydrogenation of neat formic acid and
enjoys unusual stability through millions of turnovers. Binuclear iridium hydride species 5a, which features a provocative C2-symmetric
geometry, was isolated from the reaction as a catalyst resting state. By synthesizing and carefully examining the catalytic initiation of a
series of analogues to 1a, we establish here a strong correlation between the formation of C2-twisted iridium dimers analogous to 5a
and the reactivity of formic acid dehydrogenation: an efficient C2 twist appears unique to 1a and essential to catalytic reactivity.
DOI: 10.1039/x0xx00000x
www.rsc.org/
Synthetic studies of the initiation sequence for our catalyst
enabled us to identify several species present in the conversion
of 1a21 to our active catalyst (Scheme 2). The sequence features
Introduction
Formic acid (FA) is a promising hydrogen carrier, because it is a
solar fuel and it enables on-demand H2 release under mild
conditions with CO2 as the only by-product.1,2 Ongoing research
on FA dehydrogenation has identified many outstanding
catalytic systems, both heterogeneous3–8 and homogeneous.9–
13 Our group is interested in catalytic hydride manipulation,14–19
and we have recently reported the first homogeneous iridium
catalytic system for the dehydrogenation of neat FA,20 which
presents a unique challenge due to the corrosive nature of the
neat liquid. Catalytic systems based on iridium complex 1a have
shown a useful average turnover frequency (TOF ca. 0.83 s-1)
through greater than 2 million turnovers under mild conditions
(Scheme 1).20 When compared to other homogeneous systems,
we observe several unique advantages of ours: 1) it is highly
reactive with low loading of catalyst and base; 2) it works in
neat, technical-grade FA; 3) the selectivity against CO is nearly
absolute (< 10 ppm).
three intermediates that are observable under the reactive
conditions. The first is an (allyl)iridium hydride (2a).19,21,22 The
second is a Cs-symmetric hydride-bridged dimer (4a) that is
analogous to many related structures characterized by Pfaltz.23
In the presence of carboxylic acids, we find that 4a converts
easily and selectively to structurally-novel iridium complex 5a, a
binuclear, formate-bridged iridium species that is present at the
beginning and end of our catalytic reaction. The acetate
homologue of 5a (
5a’) was characterized crystallographically.21
With the aid of distinctive hydride peaks in NMR, we showed
that 5a and its formate addition adduct, 6a, are resting states of
the catalytic reaction. Whereas 4a is unstable to our FA
dehydrogenation conditions—it converts rapidly to 5a—we
could not test directly whether the twisting of 4a to 5a is
essential to our mechanism or simply an unavoidable
thermodynamically-driven outcome.
The 5a-catalyzed dehydrogenation of FA stands out in the
FA literature, because many recent mechanistic studies show
that homogeneous, iridium-catalyzed FA dehydrogenation
systems rely on mononuclear catalysts,24–26 while ours is
dimeric. This presages a central question of what, if any, role the
second iridium plays in the reactivity. Curiously, no other C2-
twisted complexes with the iridium hydride topology of 5a are
known, although several binuclear iridium hydride complexes
that structurally resemble 4a have been synthesized,23,27,28 and
binuclear iridium complexes such as these are not well known
for catalytic dehydrogenation.23,29–32 Herein, we show that the
C2 twisting behavior seen in the conversion of 4a to 5a is highly
OTf
N
Ir
P
tBu
1a
tBu
HCOOH
CO2
+
H2
neat, 90 oC, 0.83 s-1, > 2 million TON
Scheme 1. Iridium precatalyst 1a shows FA dehydrogenation
reactivity.
specific to the parent 4a/5a system and that closely-related
structural analogs do not participate in this transformation
efficiently (or at all). We further illustrate that this twisted
bimetallic structure is essential to the system’s remarkable
reactivity.
Loker Hydrocarbon Research Institute and Department of Chemistry, University of
Southern California, 837 Bloom Walk, Los Angeles, CA 90089-1661, USA. E-mail:
travisw@usc.edu;
Fax: +1 213 740 5961; Tel: +1 213 740 5961
Electronic Supplementary Information (ESI) available: preparative details, graphical
spectral data, and raw kinetics data. See DOI: 10.1039/x0xx00000x
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

PleasDeadltoonnoTtraandsjaucsttiomnas rgins
Page 2 of 8
ARTICLE
Journal Name
1 to see if these convert to 4 or 5 under our FA
View Article Online
of complexes
Results and Discussion
1. A proposed mechanism
DOI: 10.1039/C8DT03268H
dehydrogenation conditions and compare the results of these
synthetic studies with reaction kinetics data for the
dehydrogenation process.
Scheme 2 illustrates our working mechanistic model for
initiation of our FA dehydrogenation catalyst, 1a
Precatalyst 1a21 is first transformed to 4a through the
intermediacy of several species, including (allyl)iridium hydride
complex 2a,
21 that are not observed under catalytic conditions.
Iridium dimer 4a is one of a family of iridium species of the same
iridium hydride topology that are known to occur in iridium-
catalyzed hydrogenations.23 In our hands, 4a is stable at
ambient temperature, and we have observed it directly. While
we see that 4a converts completely and rapidly to 5a under
catalytic conditions, we have not found conditions for the
reverse reaction. In a formic acid/formate buffer, 5a and 6a are
in equilibrium, and both are observed as resting states under
catalytic conditions for FA dehydrogenation. No further iridium
19,20,22
.
2. Synthesis and characterization of Ir(COD) precatalysts
A general synthetic method was developed for iridium
complexes 1a-1i. They can be prepared by adding slowly the
appropriate bidentate ligand to a dichloromethane solution of
[Ir(COD)Cl]2 (Scheme 3). After stirring at room temperature for
2 hours, insoluble salts were filtered off and the reaction
mixture was treated with hexanes to yield the corresponding
products as yellow to pale red powders. The structures of the
products were confirmed by spectral and crystallographic data.
As expected, crystal structures of 1a, 1h, and 1i show that these
complexes adopt the structure with a square planar iridium
center.16,20,21
species are observed in the catalytic reaction.
+
+
L
2 FA OR H2
Cl
+
NaOTf
L
H
Ir
+
Ir
N
P
tBu
N
L
CH2Cl2
rt
Ir
L
Ir
P
tBu
7
tBu
1a-i
tBu
+
-cod, -2 CO2
1a
(X-ray)
2a
(NMR)
+
N
R''
N
Ir
Ir
P
R'
N
R'
+
2+
1a R'=tBu R"=H
1b R'=tBu R"=Me
1c R'=iPr R"=H
1d R'=Ph R"=H
1g
solv
tBu
solv
Ir
N
H
solv
N
Ir
tBu
H
H
P
P
tBu
Ir
solv
H
P
tBu
H
N
+
tBu
H
+
tBu
-2 solv
R
R
4a
3a
N
P
P
Ir
(NMR, MS)
N
Ir
-
N
FA + HCO2
R
R
R
1e R=tBu
1f R=Ph
H2
1h R=Me
1i R=Mes
+
O
H
H
O
O
O
Scheme 3. Synthesis of iridium complexes 1a-1i. Reaction time
is 2 hours.
O
-
tBu
H
HCO2
O
Ir
tBu
H
N
N
Ir
H
Ir
tBu
P
Ir
H
O
tBu
P
N
O
O
N
bridge
opening
P
O
tBu
tBu
3. Conversion of mononuclear precursors to binuclear complexes
P
tBu
tBu
H
H
Knowing that C2-symmetrical iridium dimer 5a is a catalytic
resting state in FA dehydrogenation, we determined to evaluate
whether structures analogous to 5a are available to other
iridium complexes shown in scheme 3. To do so we developed
mild, stepwise conditions that simulate catalytic initiation, in
which we observe clean conversion of 1a to 4a’ and 5a’, as
monitored by proton NMR (Scheme 4). For this reaction, we
chose acetic acid over formic acid, because the corresponding
acetate complexes are resistant to dehydrogenation, thus
promoting the stability of our target iridium intermediates.
5a
(X-ray)
6a
(NMR)
Scheme 2. A proposed catalyst initiation mechanism of iridium
complex 1a
.
From kinetic evidence,33 we concluded that the mechanism
of catalytic formic acid dehydrogenation from 5a 6a proceeds
with rate-determining H-H bond formation followed by rapid
decarboxylation to return 5a This work is explained
/
.
elsewhere.19,20 Further, we are stunned that, despite extensive
investigation of highly analogous complexes over many years,34
precatalyst 1a was not identified for its reactivity with FA prior
to our work. Thus, we set about to track the initiation of a family
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 8
Dalton Transactions
Please do not adjust margins
Journal Name
ARTICLE
Complex 1e, with two tert-butyl supported phosphine ligands
converts to dimer 4e’ in appreciable amounts but does not
View Article Online
DOI: 10.1039/C8DT03268H
O
+
O
Ir
2+
H
solv
Ir
L
H
participate in a C2 twist to the 5e’ dimer. Complex 1g, with the
bipyridine ligands, behaves similarly: it converts to a 4g’
structure but fails to twist. Furthermore, according to MALDI
data, complex 4g’ converts back to the monomeric precatalyst
1g as the major species in solution. With these studies in mind,
the effects of both the specific pyridine and phosphine groups
Ir
CH3COOH
L
H2
H
L
H
L
O
1a-i
Ir
L
L
O
L
DCM, rt
DCM, rt
H
L
H
solv
H
5-type dimer
4-type dimer
Scheme 4. Attempted conversion of iridium complex 1a-i to 4-type
and 5-type dimers. Conditions: 10 mg iridium complex 1, 10 eq.
acetic acid, ambient temperature, 1 h.
present in 1a are needed cleanly to effect the C2 twist to a 5-
type structure.
We find that carbene-supported complex 1h does not
dimerize at all according to NMR and MS evidence. This is
consistent with our observation of catalytic reactions involving
1h in other catalytic reactions. For example, our processes for
glycerol oxidation16 and triglyceride upgrading18 proceed
through monomeric iridium complexes derived from 1h. Our
method for primary alcohols oxidation,22 proceeds through a
We used NMR and MS to monitor the conversion of iridium
chelates 1bcdegh under the conditions of scheme 4. Much to
our surprise, despite the close structural homology of 1b-i with
1a, products found upon treatment with the conditions
described in scheme 4 do not conform to the structure of
fact, only 1c 1d 1e, and 1g seem to have access to a -type
dimer, and even these reactions are not as selective as the
reaction of 1a
Results for the conversion of iridium complexes
5. In
,
,
4
dimeric derivative of 1a that is not related to
4 or 5, and we
.
suspect that 1h behaves like 1a in this reaction. These
observations could help explain the complexity of the data
shown for 1h in figure 1, because we’ve shown that there are
many productive pathways for initiation of 1h and that none of
1
are varied
and complex. The data are best summarized through the
1
analysis of iridium hydride signals visible in their respective H
NMR spectra (Figure 1). The first two lines of this figure show
the iridium hydride signals that result from the conversion of 1a
to 4a’ and 5a’, respectively. While a small fraction of 6a’ can be
seen in the spectrum of 5a’, the assignments of these spectra
them involve dimeric structures analogous to 4 or 5.
can be corroborated by comparison to their mass spectra and
20
NMR data known for 4a and 5a
.
By remarkable contrast, complex 1b, which differs from 1a
only by the addition of a methyl group in the complex’s pyridine
backbone, forms a complex mixture of products, none of which
can be comfortably assigned as analogs of
4 or 5 according to
NMR or MS. While it is possible that the aryl methyl group is
perturbing the geometry of the catalyst’s phosphinomethyl
arm, we propose that this result shows that even a subtle
perturbation of the electronics of the (pyridyl)phospine ligand
of 1a can disrupt its conversion to 4/5. Complex 1c, differing
from the parent by only the deletion of a methyl from the
phosphine tert-butyl group, shows high conversion to species
4c’, but no evidence of any C2 twisting to a 5c’ structure. It
appears therefore that steric bulk in the phosphine groups of 1a
is critical to enabling the C2 twist. Complex 1d, which differs
from the parent by the substitution of tert-butyl groups with
phenyl groups, also shows high conversion to a 4d’ dimer, but
no C2 twisting. This further supports the finding that the steric
bulk in the phosphine groups of 1a is needed for a clean
conversion to dimer 5a’
.
We investigated complexes 1e and 1g that contain only the
phosphine or pyridine moieties present in the parent 1a
.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

PleasDeadltoonnoTtraandsjaucsttiomnas rgins
Page 4 of 8
ARTICLE
Journal Name
View Article Online
DOI: 10.1039/C8DT03268H
Figure 1. Results from treatment of various precursors with catalyst initiation conditions for formic acid dehydrogenation.
Structures illustrated are predicted target structures, which were confirmed in some cases by identification of characteristic
1
hydride signals. See supporting information for full H spectra and, where available 31P spectra. Also included are spectra for the
conversion of 1a to 4a’ upon exposure to hydrogen in acetonitrile.
4. Catalytic FA dehydrogenation by iridium complexes
Even though 1b-1i do not form a -type dimer efficiently, some
of them are catalysts for FA dehydrogenation under the same
conditions that are optimized for 1a. Kinetic data collected for
FA dehydrogenation by iridium complexes 1 are striking (Figure
2). The reaction catalyzed by 1a is much faster than any other,
reaching completion in only 3 hours. By contrast, iridium
complex 1c, the most reactive of all of the derivatives studied,
delivers less than 80% conversion after 24 hours. While 1b and
1c deliver some turnover of formic acid, all other complexes are
unreactive compared to 1a. The observation of an upward
curvature of the kinetic profile of 1a can be explained in that the
decomposition of FA to gaseous H2 and CO2 reduces the volume
of the solution and therefore increases the concentrations of
both formate and the iridium catalyst.
5
The extraordinary reactivity of 1a cannot simply be ascribed
to electronic and/or steric effects. For example, complex 1a has
only two methyl groups more than 1c and one fewer than 1b
.
Poor reactivity of 1b teaches us that there is most likely an
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 5 of 8
Dalton Transactions
Please do not adjust margins
Journal Name
ARTICLE
electronic effect in the C2 twist, and the performance of 1c
shows that sterics also are important.
View Article Online
DOI: 10.1039/C8DT03268H
Experimental
I. General Procedures
We observe dramatic contrast between P—N supported 1a
and C—N supported 1h and 1i. N-Heterocyclic carbenes (NHCs)
are often compared to phosphine ligands in coordination
chemistry, as both are spectator ligands and strong sigma
donors.35,36 The “on” and “off” reactivity difference observed on
the P—N and C—N motifs strongly suggests a reaction pathway
available only to the P—N catalysts. We have previously
proposed that this is because strongly-donating carbene ligands
cause iridium hydride complexes to be high in energy and short-
lived.19
All air- and water-sensitive procedures were carried out either
in a Vacuum Atmospheres glovebox under nitrogen (0.5−10
ppm of O2 for all manipulations) or using standard Schlenk
techniques under nitrogen. Deuterated NMR solvents were
purchased from Cambridge Isotopes Laboratories. Benzene and
dichloromethane-d2 were dried over sodium benzophenone
ketyl and distilled prior to use. Formic acid (HCOOH) was
purchased from Sigma-Aldrich and used under a N2 atmosphere
without further purifications. Acetic acid was dried over CaH2
and distilled prior to use. The integrity of this material was
checked regularly by NMR. Cyclooctadiene iridium chloride
dimer (7) was purchased from Sigma-Aldrich and used as
received. Sonication procedures were done in a VWR desktop
sonic cleaner bath. Silver trifluoromethylsulfonate was
1a
1b
1c
1d
1f
purchased from Alfa Aesar and used as received. 1H, 13C, and 31
P
NMR spectra were obtained on Varian 600 or 500 MHz
spectrometers with chemical shifts reported in units of ppm. All
1
1H chemical shifts were referenced to the residual H solvent
1g
1h
1i
(relative to TMS). NMR spectra were taken in 8-inch J-Young
tubes (Wilmad) with Teflon valve plugs. All spectra were
processed using MesRe Nova (v.11.0.4-11998). MALDI mass
spectrometry was conducted on a Bruker Autoflex system.
0
3
6
9
12
15
18
21
24
TIME (HOUR)
II. Preparative Details of Precatalysts 1
Iridium complex 1a: complex 1a was prepared in the drybox
under nitrogen. 2-((Di-t-butylphosphino)methyl)pyridine37
(105ꢀmg, 0.440ꢀmmol) was dissolved in a dry vial in 5ꢀmL of dry
dichloromethane. In another vial containing a Teflon stir bar,
Figure 2. Catalytic dehydrogenation of neat FA monitored by
eudiometer. Conditions: 90 oC, 1.0 mg NaOOCH, [Ir] = 15 µmol,
2.0 mL neat formic acid.
chloro(1,5-cyclooctadiene)iridium(I)
dimer
(149ꢀmg,
Figure 2 shows that iridium complex 1a has a unique
reactivity advantage over its analogues 1b-1i. Connecting this
observation to the structural observation that only 1a converts
cleanly to a C2-symmetric binuclear complex (5a) under
catalytic conditions while 1b-1i either stop the initiation
0.220ꢀmmol) and sodium triflate (130ꢀmg, 0.750ꢀmmol) were
suspended in 10ꢀmL of dry dichloromethane. The suspension
was stirred vigorously and then the phosphinopyridine solution
was added slowly dropwise. The phosphinopyridine vial was
rinsed with 5ꢀmL of dichloromethane and added to the stirred
suspension. After stirring for 1ꢀh, the solution was filtered to
remove insoluble materials. The solvent was evaporated under
reduced pressure to yield an orange, glassy solid. A 5:1 mixture
of dry hexanes/ethyl ether (10ꢀmL) was added to the residue,
which was then triturated with the aid of sonication. The hexane
was decanted and the residue washed with an additional 10ꢀmL
of hexanes/ethyl ether. The pure iridium complex 1a was dried
under reduced pressure to give an orange solid (235ꢀmg,
77.3%).20
sequence at the 4-like dimer or turn into intractable mixtures of
multiple species, we find that failure of 1b-1i to enable efficient
catalysis is related to a failure of the catalyst precursor to
engage in the C2 twisting behavior observed in the conversion
of 4a to 5a. We believe that the relationship is causal, i.e., that
a successful C2 twist causes good catalytic performance.
Conclusions
We show here that the formation of C2-symmetrical binuclear
species 5a is essential for the formation of the catalytically
active species in neat formic acid dehydrogenation by 1a. We
propose that only iridium complexes in this series that can
1b-d: Iridium complexes 1b-d were prepared following the
same method as 1a. The pyridylphospine ligands for 1b and 1c
were prepared similarly to literature methods as in 1a. The
pyridylphosphine ligand for 1d was modified from known
literature (see Supplementary Information).38
access
5-type geometry have useful reactivity in
dehydrogenation of neat FA. Kinetic data and structural analysis
of an array of eight iridium complexes support this proposal.
Future experiments on elucidation of the mechanism of FA
dehydrogenation are under way.
1e-i: Iridium complexes 1e-i were prepared according to
literature procedure.16,39–41
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

PleasDeadltoonnoTtraandsjaucsttiomnas rgins
Page 6 of 8
ARTICLE
Journal Name
III. Conversion of Precatalysts 1 to 4- and 5-type dimers
-Type species: In a typical reaction, iridium complex (5.0 mg)
View Article Online
DOI: 10.1039/C8DT03268H
M. Grasemann and G. Laurenczy, Energy Environ. Sci.,
2012, 5, 8171–8181.
Notes and references
1
4
was placed in a J-Young NMR tube in a glovebox. Then, 0.60 mL
of dried deuterated dichloromethane was placed in the tube
and the tube was sealed. The complexes dissolved readily to
form a red solution. The tube was then taken out of the
glovebox and affixed to a Schlenk line. The tube was evacuated
under reduced pressure, but only very briefly so as to prevent
volatiles from escaping. The tube was then, while still on the
Schlenk line, filled with H2 gas. The tube was sealed and shaken
until the red solution turned yellow, which happens rapidly
upon mixing.
2
3
F. Joó, ChemSusChem, 2008, 1, 805–808.
A. Bulut, M. Yurderi, Y. Karatas, Z. Say, H. Kivrak, M. Kaya,
M. Gulcan, E. Ozensoy and M. Zahmakiran, ACS Catal.,
2015, 5, 6099–6110.
A. Beloqui Redondo, F. L. Morel, M. Ranocchiari and J. A.
Van Bokhoven, ACS Catal., 2015, 5, 7099–7103.
S. Zhang, Ö. Metin, D. Su and S. Sun, Angew. Chem. Int. Ed.,
2013, 52, 3681–3684.
M. Ojeda and E. Iglesia, Angew. Chem. Int. Ed., 2009, 48,
4800–4803.
J. Li, Q.-L. Zhu and Q. Xu, Chimia (Aarau)., 2015, 69, 348–
352.
M. D. Marcinkowski, J. Liu, C. J. Murphy, M. L. Liriano, N. A.
Wasio, F. R. Lucci, M. Flytzani-Stephanopoulos and E. C. H.
Sykes, ACS Catal., 2017, 7, 413–420.
T. C. Johnson, D. J. Morris, M. Wills, G. Laurenczy, J. R.
Noyes, S. Gladiali, M. Beller and A. L. Spek, Chem. Soc. Rev.,
2010, 39, 81–88.
E. A. Bielinski, P. O. Lagaditis, Y. Zhang, B. Q. Mercado, C.
Würtele, W. H. Bernskoetter, N. Hazari and S. Schneider, J.
Am. Chem. Soc., 2014, 136, 10234–10237.
C. Chauvier, A. Tlili, C. Das Neves Gomes, P. Thuéry and T.
Cantat, Chem. Sci., 2015, 6, 2938–2942.
A. Boddien, D. Mellmann, F. Gartner, R. Jackstell, H. Junge,
P. J. Dyson, G. Laurenczy, R. Ludwig and M. Beller, Science,
2011, 333, 1733–1736.
J. F. Hull, Y. Himeda, W.-H. Wang, B. Hashiguchi, R. Periana,
D. J. Szalda, J. T. Muckerman and E. Fujita, Nat. Chem.,
2012, 4, 383–388.
Z. Lu, B. L. Conley and T. J. Williams, Organometallics,
2012, 31, 6705–6714.
Z. Lu and T. J. Williams, ACS Catal., 2016, 6, 6670–6673.
Z. Lu, I. Demianets, R. Hamze, N. J. Terrile and T. J.
Williams, ACS Catal., 2016, 6, 2014–2017.
X. Zhang, L. Kam, R. Trerise and T. J. Williams, Acc. Chem.
Res., 2017, 86–95.
Z. Lu, V. Cherepakhin, T. Kapenstein and T. J. Williams, ACS
Sustain. Chem. Eng., 2018, 6, 5749–5753.
Z. Lu, V. Cherepakhin, I. Demianets, P. J. Lauridsen and T. J.
Williams, Chem. Commun., 2018, 54, 7711–7724.
J. J. A. Celaje, Z. Lu, E. A. Kedzie, N. J. Terrile, J. N. Lo and T.
J. Williams, Nat. Commun., 2016, 7, 11308.
Crystallographic data for compounds 1a (CCDC 1415049),
2a (CCDC 1815570), 5a’ (CCDC 1415050), 1h (CCDC
1438247), and 1i (CCDC 1438246) can be obtained free of
charge from The Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/data_request/cif.
V. Cherepakhin and T. J. Williams, ACS Catal., 2018, 8,
3754–3763.
4
5
6
7
8
In the case of figure 1, the isolation of complex 4a was done by
following this same method substituting acetonitrile for
dichloromethane.
5-Type species: In a typical reaction, the J-Young NMR tube from
the previous conversion to a -type species was taken into a
4
9
glovebox. Acetic acid (glacial, dry, 10 equivalents) were added
to the tube solution via a microsyringe. The tube was then
sealed and shaken and allowed to sit for 1 hour prior to NMR
studies.
10
IV. Hydrogen Quantification
11
12
In a typical reaction, formic acid (2.0 mL, 53 mmol) and sodium
formate (1.0 mg, 0.022 mmol) were combined with the analyte
iridium complex
1 (15 µmol) in a 2 mL Schlenk bomb equipped
with a Teflon stir bar while in a glovebox under nitrogen. A
eudiometer was constructed as follows: the side arm of the
valve of the Schlenk flask was connected to a piece of Tygon
tubing, which was adapted to 20 gauge (0.03 inch) Teflon tubing
with a needle. The tubing was threaded through an inverted 1 L
graduated cylinder filled with water. The reactor’s valve was
opened to release gas from the reactor headspace while heating
in a regulated oil bath. The volume of liberated gas was
recorded periodically until gas evolution ceased. Liberated
hydrogen was quantified by recording its volume displacement
in the eudiometer.
13
14
15
16
17
18
19
20
21
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
This work is sponsored by the NSF (CHE-1566167), and the
Hydrocarbon Research Foundation. We thank the NSF (DBI-
0821671, CHE-0840366, CHE-1048807) and the NIH (S10
RR25432) for analytical instrumentation. Fellowship assistance
from the USC undergraduate research apprentice and provost
fellowship programs (E. A. K., P. J. L.) is gratefully acknowledged.
Support from the Jerome A. Sonosky fellowship of the USC
Wrigley Institute (Z. L. and J. J. A. C.) is also appreciated.
22
23
24
S. Gruber, M. Neuburger and A. Pfaltz, Organometallics,
2013, 32, 4702–4711.
S. Cohen, V. Borin, I. Schapiro, S. Musa, S. De-Botton, N. V.
Belkova and D. Gelman, ACS Catal., 2017, 8139–8146.
6 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 7 of 8
Dalton Transactions
Please do not adjust margins
Journal Name
ARTICLE
25
26
27
28
29
30
A. Matsunami, S. Kuwata and Y. Kayaki, ACS Catal., 2017, 7,
4479–4484.
View Article Online
DOI: 10.1039/C8DT03268H
M. Iguchi, H. Zhong, Y. Himeda and H. Kawanami, Chem. - A
Eur. J., 2017, 23, 17017–17021.
Z. Hou, T. A. Koizumi, A. Fujita, H. Yamazaki and Y.
Wakatsuki, J. Am. Chem. Soc., 2001, 123, 5812–5813.
T. M. Gilbert and R. G. Bergman, J. Am. Chem. Soc., 1985,
107, 3502–3507.
Y. E. Begantsova, A. E. Varvarin, V. A. Ilichev and L. N.
Bochkarev, Russ. J. Gen. Chem., 2017, 87, 1192–1197.
W. I. Dzik, S. E. Calvo, J. N. H. Reek, M. Lutz, M. A. Ciriano,
C. Tejel, D. G. H. Hetterscheid and B. De Bruin,
Organometallics, 2011, 30, 372–374.
31
32
K. Ogata, T. Nagaya and S. I. Fukuzawa, J. Organomet.
Chem., 2010, 695, 1675–1681.
E. S. Andreiadis, D. Imbert, J. Pécaut, A. Calborean, I.
Ciofini, C. Adamo, R. Demadrille and M. Mazzanti, Inorg.
Chem., 2011, 50, 8197–206.
33
Most striking in this study is the finding of half-order
dependence on formate, which we rationalize by noting
that in the conversion of 5a to 6a, one formate activates
two sites on the catalyst. A free radical chain mechanism
can also explain these kinetics, but we find no impact on
rate with the addition of radical inhibitors such as BHT.
W.-H. Wang, Y. Himeda, J. T. Muckerman, G. F. Manbeck
and E. Fujita, Chem. Rev., 2015, 115, 12936–12973.
S. P. Nolan, N-Heterocyclic Carbenes in Synthesis, 2006.
R. A. Moss and M. P. Doyle, Contemporary Carbene
Chemistry, 2013.
34
35
36
37
38
39
C. Beddie, P. Wei and S. Douglas, Can. J. Chem., 2006, 84,
755–761.
R. Sun, H. Wang, J. Hu, J. Zhao and H. Zhang, Org. Biomol.
Chem., 2011, 12, 5954–5963.
U. Hintermair, S. W. Sheehan, A. R. Parent, D. H. Ess, D. T.
Richens, P. H. Vaccaro, G. W. Brudvig and R. H. Crabtree, J.
Am. Chem. Soc., 2013, 135, 10837–10851.
M. Brill, D. Marrwitz, F. Rominger and P. Hofmann, J.
Organomet. Chem., 2015, 775, 137–151.
40
41
W. A. Fordyce and G. A. Crosby, Inorg. Chem., 1982, 21,
1455–1461.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

Dalton Transactions
Page 8 of 8
View Article Online
DOI: 10.1039/C8DT03268H
Conformational Twisting of a Formate-Bridged Diiridium Complex Enables
Catalytic Formic Acid Dehydrogenation
Paul J. Lauridsen, Zhiyao Lu, Jeff J. A. Celaje, Elyse A. Kedzie, Travis J. Williams*
Catalytic reactivity is switched on for formic acid dehydrogenation by a single precursor’s unique ability
to form a geometrically twisted dimer.