Transition-Metal-Free Acceptorless Decarbonylation of Formic Acid Enabled by a Liquid Chemical-Looping Strategy

Article abstract of DOI:10.1002/anie.201909039

The selective decarbonylation of formic acid was achieved under transition-metal-free conditions. Using a liquid chemical-looping strategy, the thermodynamically favored dehydrogenation of formic acid was shut down, yielding a pure stream of CO with no H₂ or CO₂ contamination. The transformation involves a two-step sequence where methanol is used as a recyclable looping agent to yield methylformate, which is subsequently decomposed to carbon monoxide using alkoxides as catalysts.

Full text of DOI:10.1002/anie.201909039

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201909039
German Edition: DOI: 10.1002/ange.201909039
Organocatalysis
Hot Paper
Transition-Metal-Free Acceptorless Decarbonylation of Formic Acid
Enabled by a Liquid Chemical-Looping Strategy
Arnaud Imberdis, Guillaume Lefꢀvre, and Thibault Cantat*
Abstract: The selective decarbonylation of formic acid was
achieved under transition-metal-free conditions. Using a liquid
chemical-looping strategy, the thermodynamically favored
dehydrogenation of formic acid was shut down, yielding
a pure stream of CO with no H2 or CO2 contamination. The
transformation involves a two-step sequence where methanol is
used as a recyclable looping agent to yield methylformate,
which is subsequently decomposed to carbon monoxide using
alkoxides as catalysts.
Scheme 1. Routes for the production and decomposition of formic
acid.
A
mong C1 chemicals, formic acid (HCOOH, FA) has been
the production of CO from FA would afford an attractive way
to produce a stream of pure CO, in a controlled way, from
a sustainable and storable precursor.[9]
the focus of renewed interests since it is an important product
in catalytic transformations related to the storage of sustain-
ably produced energy. Recent efforts have indeed showed
that HCOOH is a key intermediate in the hydrogenation of
CO2 to methanol and methane.[1] In addition, formic acid can
be produced from the (photo)electrolysis of CO2 or the
hydrogenation of CO2 and carbonates.[2a–c] Despite its simple
formulation, FA can undergo different decomposition reac-
tions depending on the reaction conditions and the presence
of catalysts. While homogeneous catalysts have been devel-
oped that can disproportionate FA to methanol, the main
decomposition pathway involves the dehydrogenation of
HCOOH to H2 and CO2.[3a–e] In fact, the reversible hydro-
genation of CO2 (and bicarbonates) to formates has led to the
concept of a hydrogen battery for the storage of H2 in liquid
form.[4]
In situ decomposition of formic acid to promote for
example formal carbonylation or hydroxycarbonylation, for
example, has been reported.[10] Only a few methods have been
developed to promote the acceptorless decarbonylation of
FA, which relied on the use of stoichiometric amounts of
sulfuric or phosphoric acids[11] or on thermolytic conditions.[12]
Catalytic strategies are scarce. These involve zeolite-based
catalysts able to decompose FA at high temperatures
(> 1508C), to remove water, and exhibit modest activity
with turnover frequencies up to 39 hÀ1.[13] Very recently, while
exploring the alkoxycarbonylation of alkenes with FA, Beller
et al. discovered that palladium complexes, supported by
chelating bis-phosphines decorated with pyridine bases, could
catalyze the acceptorless decarbonylation of FA.[14] Because
the dehydrogenation of HCOOH is facile, both thermody-
namically and kinetically, the authors noted the concomitant
release of at least 10% CO2 and H2. In the pursuit of
a practical system able to selectively decarbonylate FA, we
sought a transition-metal-free method. Under organocatalytic
conditions, the dehydrogenation of HCOOH is indeed
difficult and only a handful of catalysts have been shown to
decompose FA to CO2 and H2.[15] Herein, we report a system
that combines chemical looping and an organocatalytic
transformation for the decomposition of FA into CO and
H2O, without the formation of H2, at low temperature
(< 758C).
From a thermodynamic standpoint, FA could also decom-
pose to CO and H2O, with a Gibbs free energy of
À12.4 kJmolÀ1 at 2988C (Scheme 1).[5] This decarbonylation
reaction is less favoured than the classical dehydrogenation
(DG8 = À32.9 kJmolÀ1 at 2988C);[5] but it would provide
a convenient flow of CO from a renewable feedstock. Utilized
in large scale in the Fischer–Tropsch and Cativa processes,[6]
as well as in hydroformylation reactions, carbon monoxide is
currently produced from fossil sources, primarily through
methane steam reforming (SMR) or autothermal reformer
(ATR).[7] Alternatively, the reverse water–gas shift (RWGS)
reaction can convert CO2 and H2 into a mixture of CO, H2O,
CO2, and H2 at equilibrium.[8] Overall, these methods suffer
from severe disadvantages, such as the need for further
purification of the gas stream, to separate CO. In this context,
From a mechanistic standpoint, the decarbonylation of FA
À
must involve a C H activation step that results in formal
À
deprotonation of the C H group to reduce the carbon atom.
Computationally, the corresponding proton has a pKa of ca.
31.[16] The direct decarbonylation of HCOOH with an organic
Brønsted base is hence illusory in the presence of the more
acidic O-H functionalities of FA (pKa = 3.7) or water. To
circumvent this limitation, we envisioned a chemical-looping
strategy. Chemical looping is a powerful and practical
approach where a transformation is divided into several
sub-reactions in order to separate gases, prevent deleterious
[*] A. Imberdis, Dr. G. Lefꢀvre, Dr. T. Cantat
NIMBE, CEA, CNRS, Universitꢁ Paris-Saclay, CEA Saclay
91191 Gif-sur-Yvette cedex (France)
E-mail: thibault.cantat@cea.fr
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
https://doi.org/10.1002/anie.201909039.
Angew. Chem. Int. Ed. 2019, 58, 1 – 6
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Table 1: Scope of the reaction for the decarbonylation of alkyl formate.[a]
equilibria, and/or avoid incompatible substrates. This has
been applied using solid looping agents in a variety of
processes, including the RWGS reaction, the (super-)dry
reforming of methane, and CO2 capture from combustion.[17]
An esterification reaction is well-suited to separate the acidic
À
O-H functionalities from the C H bond in Equation (3)
because it would yield an alkylformate derivative with
a Gibbs free energy intermediate between FA and CO +
H2O (Scheme 2). An alcohol was hence chosen to set up the
Entry
R
R’XM
T [8C]
t [h]
Conv [%][c]
1
2
None
MeOLi
30
30
20
8
<5
54
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
MeONa
MeONa/18C6
MeONa
MeOK
MeOK/18C6
MeOK/222
MeORb
EtOK
19
19
30
19 (75)
19
19
19
19
30
19
30
19
19 (75)
19
3
3
3
3
3
3
3
3
3
20
3
4
2
1
3
3
<5
56
46
56 (85)
47
49
44
45
59
42
56
50
24 (77)
51
60 (81)
69 (94)
Me
tBuOK
Me2NLi
TBDNa
TBDK
MeOK
EtOK
Et
nBu
Bz
tBuOK
MeOK
30 (75)
19 (75)
Scheme 2. Principle of liquid chemical looping for the transition-metal-
free decarbonylation of formic acid, with the associated Gibbs free
energies using methanol and ethanol as looping reagents.
[a] Reaction conditions: 1 mmol MF, 50 mmol catalyst (5 mol%), 500 mL
DMF. [b] Screening of the solvent (see details in the Supporting
Information). [c] Determined by 1H NMR of the crude reaction mixture.
chemical looping depicted in Scheme 2. This relies on the
esterification of formic acid with an alcohol and isolation of
the corresponding alkylformate by distillation (steps 1–2 in
Scheme 2). The subsequent catalytic decarbonylation of the
alkylformate would afford a stream of CO, while regenerating
the alcohol. The choice of the best-suited looping agent is
discussed at the end on this communication since it relies both
on the physicochemical properties of the alcohol/alkylfor-
mate couple and on the reactivity of the alkylformate in the
decarbonylation step.
catalyst, the solvent, and the alkylformate were investigated
(Table 1). Given the low exergonicity of the decarbonylation
of MF (DG8298K = À2.9 kJmolÀ1), the reaction operates under
an equilibrium and, in a sealed NMR tube, a maximum
conversion of 60% was evaluated for the decomposition of
MF vs. 90% at 758C, which is in agreement with an
endothermic reaction (DH8298K =+ 36.9 kJmolÀ1). A screen-
ing of solvents revealed that the decarbonylation of MF was
efficient in polar aprotic solvents with large dissociation
constants, such as DMF (e = 36.7) and NMP (e = 32.2), with
a conversion to CO and methanol of 56 and 58%, respec-
tively, after 3 h at 198C (see Table S1 in the Supporting
Information). Importantly, the decarbonylation was com-
pletely shut down in methanol and in neat conditions, thereby
indicating potential poisoning or deactivation of the catalyst
by both the product and the substrate.
While MeOLi and MeONa were found to be inactive at
198C, these catalysts decompose MF to CO and methanol at
308C, and a longer reaction time was required in the presence
of the lithium derivative (8 vs. 3 h; Entries 2, 3 and 5 in
Table 1). The rubidium salt MeORb exhibited catalytic
activity close to that of MeOK (Entry 9 in Table 1). Since
alkali metal cations have been shown to influence the
performances of catalytic systems through coordination to
oxygen-rich substrates, the coordination sphere of the potas-
sium cation in MeOK was modulated by the addition of
exogenous chelating ligands. Addition of 5.0 mol% of the 18-
C-6 crown ether or the 2,2,2-cryptand had no impact on the
catalytic performances of MeOK, thus suggesting an innocent
Interestingly, several metal catalysts have been reported
for the acceptorless decarbonylation of alkylformates based
on copper,[18a] ruthenium,[18b] rhodium,[18c] osmium,[18d] and
palladium[18e] complexes that are able to oxidatively add to
À
the formate C H bond. Organic alternatives exist, which
involve guanidines,[19] amines, or phosphines.[20] All these
catalysts nevertheless operate at elevated temperatures,
above 1408C. Reasoning that an alkoxide base (pKa = 29–32
in DMSO)[21] would be basic enough to deprotonate a C H
À
bond of a formate group, a DMF solution of methylformate
(MF) was investigated, in the presence of 5.0 mol% potas-
sium methoxide (MeOK). In a sealed NMR tube, rapid
decomposition of 56% MF was noted after 3 h at 198C in
DMF (Entry 6, Table 1). The concomitant formation of
1
methanol was observed by H and 13C NMR spectroscopy,
while the production of CO was identified in the gas phase
using GC. Importantly, neither H2 nor CO2 was detected by
GC in the gas phase. A conversion of 85% was reached at
758C, while no reaction occurred in the absence of the
catalyst. Encouraged by this success, the influence of the
2
www.angewandte.org
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2019, 58, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
role of K+. The difference in reactivity between MeOK and its
lithium and sodium congeners hence likely stems from the
tighter ion pairs formed between the MeOÀ anion and the
hard Li+ and Na+ cations. This trend is reflected in the
À
shortening of O M bond length in the alkali oxides M2O from
2.9, 2.8, 2.4, to 2.0 ꢀ across the series Rb, K, Na, Li.[22] This
interpretation is supported by the comparable catalytic
activities of MeOK and MeONa/18-C-6 at 198C (Entries 4
and 6 in Table 1). It is notable that bases stronger than MeOK,
such as EtOK, t-BuOK, and nitrogen-containing bases
(TBDK), exhibited catalytic performances similar to
MeOK; similarly, Me2NLi behaved like MeOLi.
Bulkier ethyl and n-butyl formates were found to be less
reactive than MF and required stronger bases as catalysts
(entries 15–17). In contrast, 94% of the more activated benzyl
formate was decomposed at 758C. It is noteworthy that all the
results displayed in Table 1 involved very mild temperatures
(198C or 308C).
Figure 1. Plot of the volume of CO produced upon decarbonylation of
To suppress the thermodynamic constraint associated with
pressure buildup in a closed vessel, the decomposition of MF
was carried out in an open system and monitored using
a eudiometer (Figure 1). Using the reaction conditions of
Entry 6, Table 1, CO was generated in 82% yield after only
50 min at 198C (red plot in Figure 1). The setup enabled
a kinetic study, thereby establishing that the rate law is first
order in catalyst and has a À1 order with respect to MeOH
while the order for MF is non-integer (see the Supporting
Information).
methyl formate in an open system using different initial loadings of
MeOH (red: 0 mol% MeOH, green: 5 mol% MeOH, blue: 10 mol%
MeOH) and evolution of the rate constant with respect to the initial
concentration of MeOK and MeOH.
chemical looping depicted in Scheme 2. MeOH/MF was
selected as the best looping system, because both MeOH
and MF are liquids under ambient conditions and MF
presents a high CO content of 47 wt%. Using a two-chamber
system depicted in Table 2, a solution of MeOK in DMF
(0.07 molLÀ1) was connected to a vessel containing pure MF.
The device was equipped with a condenser and immersed in
a 758C oil bath. Thanks to the low boiling points of MF and
MeOH of 32 and 618C, respectively, accumulation of the
reagents and products in the solution containing the catalyst
(chamber B in Table 2) is avoided and methanol is collected in
These data confirm that both MF and MeOH have
a detrimental effect on the reactivity of the catalyst and, to
better comprehend this behavior, DFT calculations were
carried out on the mechanism of the decarbonylation
reaction.
The MeOÀ catalyst is a base strong enough to interact with
À
MF, and although the deprotonation of the C H bond is
slightly uphill (DG =+ 5.1 kcalmolÀ1), CO release occurs
readily via a low lying transition state (TS2, DG =+ 8.1 kcal
molÀ1; Scheme 3, green surface). Interestingly, the methanol
byproduct forms a strong H-bond with MeOÀ and the
separation of the free MeOÀ base from methanol requires
+7.3 kcalmolÀ1. As a result, the energy span governing the
activity of the MeOÀ catalyst reaches 15.4 kcalmolÀ1.
As the reaction proceeds, the accumulation of methanol
can be accounted for, mechanistically, by exploring the
catalytic behavior of the H-bonded [MeOÀ···HOMe] pair
(blue surface in Scheme 3). The decreased basicity of the
catalyst results in destabilization of the two transitions states
À
responsible for the deprotonation of the C H bond in MFand
the release of CO from the CH3OC(O)À anion (e.g., TS2,
DG =+ 14.5 kcalmolÀ1). As a consequence, in the presence of
methanol, the reaction span increases to 19.8 kcalmolÀ1
,
thereby explaining the deleterious influence of the reaction
product on the catalyst activity. In addition, the presence of
a large excess MF results in the trapping of the MeOÀ catalyst
to form a HC(OMe)2OÀ anion (with a Gibbs free energy of
À3.7 kcalmolÀ1), thereby slowing down the rate of the
decarbonylation.
Scheme 3. Decarbonylation of methyl formate catalyzed by MeOÀ.
Gibbs free energies in kcalmolÀ1 computed at the M062x/6-
311+ +G**/PCM level of theory.
Based on these key findings, an optimized system was
found for the decarbonylation of formic acid, using the
Angew. Chem. Int. Ed. 2019, 58, 1 – 6
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
Table 2: Catalytic decarbonylation of MF using a two-chamber setup.[a]
Scheme 5. Tandem carbonylation reactions enabling the preparation of
amides with CO generated from MF.
this palladium-catalyzed reaction. Interestingly, the possibil-
ity of generating high pressures of CO (up to 26 bars, allowed
[b]
Entry Reagent (mL/mol)
t
TON[b] TOF (hÀ1
)
Yield [%][b]
À
iron-catalyzed carbonylation of the N CH3 bond in N,N-
1
2
3
MeOCOH 10/0.16 20 h 1618
MeOCOH 10/0.16 40 h 2171
81
54
52
70
92
22
dimethylaniline in 73% yield under 9 bars of CO, using
a double autoclave system (see Scheme 5 and the Supporting
Information).[25] While chemical looping is common strategy
in the realm of heterogenous catalysis, the present work hence
exemplifies how liquid chemical looping can unlock thermo-
dynamically unfavored transformations, at low temperatures,
without the need for sophisticated metal catalysts.
MeOCOH 100/1.6 4 d
5000
[a] Reaction conditions: In a two-chamber system surmounted with
a cooler, pure MF is introduced in the reserve chamber (A) and the
catalytic chamber (B) is charged with 7 mmol catalyst (5 mg) and 1 mL
DMF. The setup is immersed in an oil bath at 758C. [b] Determined by
1H NMR of an aliquot of compartment A mixture.
the reservoir vessel (chamber A). Using this setup, an
excellent TON of 5000 was measured, with a TOF up to
81 hÀ1 at 758C. Overall, 10 mL MF were successfully decar-
bonylated in 92% yield to afford 3.7 L CO, after 40 h
(Table 2). 6.1 mL of methanol were recovered, with a purity
of 96% (along with 4% unreacted MF). To demonstrate the
liquid chemical looping depicted in Scheme 2, the regener-
ation of MF was carried out by esterification of the produced
methanol with HCOOH at 328C. Continuous distillation of
MF enabled the isolation of the alkylformate in 86% yield
after 5 h, thanks to the absence of azeotrope between MF and
formic acid, water, or methanol.
Acknowledgements
The authors acknowledge for financial support of this work:
CEA, CNRS, CINES (project sis6494), the CHARMMMAT
Laboratory of Excellence, and the European Research
Council (ERC Consolidator Grant Agreement no. 818260).
Conflict of interest
The authors declare no conflict of interest.
All together, the decarbonylation of formic acid to CO
and water was achieved in 79% yield at low temperature
(758C), using transition-metal-free catalysts, for the first time
(Scheme 4). No H2 nor CO2 contamination of the gas stream
was detected by GC, although the dehydrogenation of formic
acid is thermodynamically preferred.
Keywords: carbon monoxide · chemical looping ·
decarbonylation · formic acid · organocatalysis
[1] a) S. Wesselbaum, V. Moha, M. Meuresch, S. Brosinski, K. M.
Thenert, J. Kothe, T. vom Stein, U. Englert, M. Hçlscher, J.
Klankermayer, et al., Chem. Sci. 2015, 6, 693 – 704; b) M.
Behrens, F. Studt, I. Kasatkin, S. Kuhl, M. Havecker, F. Abild-
Pedersen, S. Zander, F. Girgsdies, P. Kurr, B.-L. Kniep, et al.,
Science 2012, 336, 893 – 897.
[2] a) A. S. Agarwal, Y. Zhai, D. Hill, N. Sridhar, ChemSusChem
2011, 4, 1301 – 1310; b) C. Federsel, R. Jackstell, M. Beller,
Angew. Chem. Int. Ed. 2010, 49, 6254 – 6257; Angew. Chem. 2010,
122, 6392 – 6395; c) C. Federsel, C. Ziebart, R. Jackstell, W.
Baumann, M. Beller, Chem. Eur. J. 2012, 18, 72 – 75.
Recently, several surrogates of CO have been designed to
generate a low pressures of CO for synthesis at the laboratory
scale, for instance using two-chamber systems (e.g.,
COgen).[23] Similarly, we were able to perform the gram-
scale aminocarbonylation of 4-bromoanisole with morpho-
line, using CO produced from MF (Scheme 5).[24] The absence
of H2 circumvented the dehalogenation of the arylbromide in
[3] a) A. J. M. Miller, D. M. Heinekey, J. M. Mayer, K. I. Goldberg,
Angew. Chem. Int. Ed. 2013, 52, 3981 – 3984; Angew. Chem.
2013, 125, 4073 – 4076; b) S. Savourey, G. Lefꢁvre, J.-C. Berthet,
P. Thuꢂry, C. Genre, T. Cantat, Angew. Chem. Int. Ed. 2014, 53,
10466 – 10470; Angew. Chem. 2014, 126, 10634 – 10638; c) M. C.
Scheme 4. Global process (YG =global yield of the process).
4
www.angewandte.org
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2019, 58, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
Neary, G. Parkin, Chem. Sci. 2015, 6, 1859 – 1865; d) K. Sordakis,
and T in K); K. Shen, Y. Fu, J.-N. Li, L. Liu, Q.-X. Guo,
A. Tsurusaki, M. Iguchi, H. Kawanami, Y. Himeda, G. Lau-
renczy, Green Chem. 2017, 19, 2371 – 2378; e) A. Tsurusaki, K.
Murata, N. Onishi, K. Sordakis, G. Laurenczy, Y. Himeda, ACS
Catal. 2017, 7, 1123 – 1131.
Tetrahedron Lett. 2007, 63, 1568 – 1576.
[17] a) M. Keller, J. Fung, H. Leion, T. Mattisson, Fuel 2016, 180,
448 – 456; b) M. Wenzel, L. Rihko-Struckmann, K. Sundmacher,
Chem. Eng. J. 2018, 336, 278 – 296; c) T. Prçll, P. Kolbitsch, J.
Bolhꢄr-Nordenkampf, H. Hofbauer, AIChE J. 2009, 55, 3255 –
3266; d) M. Ryden, A. Lyngfelt, Int. J. Hydrogen Energy 2006,
31, 1271 – 1283; e) M. M. Hossain, H. I. de Lasa, Chem. Eng. Sci.
2008, 63, 4433 – 4451.
[18] a) G. Doyle, US Pat. 4.319.050, 1981; b) G. Jenner, E. M.
Nahmed, H. Leismann, Tetrahedron Lett. 1989, 30, 6501 – 6502;
c) H. A. Zahalka, H. Alper, Tetrahedron Lett. 1987, 28, 2215 –
2216; d) C. Legrand, Y. Castanet, A. Mortreux, F. Petit,
Tetrahedron Lett. 1992, 33, 3753 – 3756; e) J. S. Lee, J. C. Kim,
Y. G. Kim, Appl. Catal. 1990, 57, 1 – 30.
[4] a) A. Boddien, F. Gꢃrtner, C. Federsel, P. Sponholz, D.
Mellmann, R. Jackstell, H. Junge, M. Beller, Angew. Chem.
Int. Ed. 2011, 50, 6411 – 6414; Angew. Chem. 2011, 123, 6535 –
6538; b) K. Sordakis, C. Tang, L. K. Vogt, H. Junge, P. J. Dyson,
M. Beller, G. Laurenczy, Chem. Rev. 2018, 118, 372 – 433.
[5] T. Zell, R. Langer, Phys. Sci. Rev. 2018, 3, 57 – 93.
[6] J. H. Jones, Platinum Met. Rev. 2000, 44, 94 – 105.
[7] I. Dybkjaer, Fuel Process. Technol. 1995, 42, 85 – 107.
[8] F. D. Doty, G. N. Doty, J. P. Staab, L. L. Holte, ES2010-90362
ASME conference, 2010, 1 – 10.
[19] M. J. Green, Eur Pat. EP0115387, 1984.
[20] F. R. Vega, J. C. Clꢂment, H. des Abbayes, Tetrahedron Lett.
1993, 34, 8117 – 8118.
[21] W. N. Olmstead, Z. Margolin, F. G. Bordwell, J. Org. Chem.
1980, 45, 3295 – 3299.
[9] a) J. Klankermayer, S. Wesselbaum, K. Beydoun, W. Leitner,
Angew. Chem. Int. Ed. 2016, 55, 7296 – 7343; Angew. Chem.
2016, 128, 7416 – 7467; b) W. Supronowicz, I. A. Ignatyev, G.
Lolli, A. Wolf, L. Zhao, L. Mleczko, Green Chem. 2015, 17,
2904 – 2911.
[22] N. K. McGuire, M. OꢅKeeffe, J. Solid State Chem. 1984, 54, 49 –
53.
[10] a) X. Qi, C.-L. Li, X.-F. Wu, Chem. Eur. J. 2016, 22, 5835 – 5838;
b) F.-P. Wu, J.-B. Peng, X. Qi, X.-F. Wu, Catal. Sci. Technol. 2017,
7, 4924 – 4928; c) J.-P. Simonato, J. Mol. Catal. A 2003, 197, 61 –
64; d) T. G. Ostapowicz, M. Schmitz, M. Krystof, J. Klanker-
mayer, W. Leitner, Angew. Chem. Int. Ed. 2013, 52, 12119 –
12123; Angew. Chem. 2013, 125, 12341 – 12345.
[11] J. S. Morgan, J. Chem. Soc. Trans. 1916, 109, 274 – 283.
[12] W. L. Nelson, C. J. Engelder, J. Phys. Chem. 1925, 30, 470 – 475.
[13] P. Losch, A. S. Felten, P. Pale, Adv. Synth. Catal. 2015, 357, 2931 –
2938.
[23] a) C. Veryser, S. Van Mileghem, B. Egle, P. Gilles, W. M.
De Borggraeve, React. Chem. Eng. 2016, 1, 142 – 146; b) C.
Veryser, G. Steurs, L. Van Meervelt, W. M. De Borggraeve, Adv.
Synth. Catal. 2017, 359, 1271 – 1276; c) S. D. Friis, A. T. Lind-
hardt, T. Skrydstrup, Acc. Chem. Res. 2016, 49, 594 – 605.
[24] J. R. Martinelli, D. A. Watson, D. M. M. Freckmann, T. E.
Barder, S. L. Buchwald, J. Org. Chem. 2008, 73, 7102 – 7107.
[25] T. Nasr Allah, S. Savourey, J.-C. Berthet, E. Nicolas, T. Cantat,
Angew. Chem. Int. Ed. 2019, 58, 10884 – 10887; Angew. Chem.
2019, 131, 11000 – 11003.
[14] R. Sang, P. Kucmierczyk, K. Dong, R. Franke, H. Neumann, R.
Jackstell, M. Beller, J. Am. Chem. Soc. 2018, 140, 5217 – 5223.
[15] C. Chauvier, A. Tlili, C. Das Neves Gomes, P. Thuꢂry, T. Cantat,
Chem. Sci. 2015, 6, 2938 – 2942.
Manuscript received: July 19, 2019
[16] The pKa-values were calculated with respect to the furan anion
as reference base (experimental value of 35) according to the
formula: pKa(THF) = 35 + DG/(2.303RT) (DG in kcalmolÀ1
Revised manuscript received: September 10, 2019
Accepted manuscript online: September 17, 2019
Version of record online: && &&, &&&&
Angew. Chem. Int. Ed. 2019, 58, 1 – 6
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Organocatalysis
A. Imberdis, G. Lefꢁvre,
T. Cantat*
&&&& — &&&&
Transition-Metal-Free Acceptorless
Decarbonylation of Formic Acid Enabled
by a Liquid Chemical-Looping Strategy
In the loop: Selective and acceptorless
decarbonylation of formic acid was ach-
ieved for the first time under transition-
metal-free conditions. The approach,
inspired by chemical-looping strategies,
shuts down the thermodynamically
favored dehydrogenation of formic acid,
yielding a pure stream of CO without H2
or CO2 contamination.
6
www.angewandte.org
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2019, 58, 1 – 6
These are not the final page numbers!