Chemical equilibria in formic acid/amine-CO2 cycles under isochoric conditions using a ruthenium(ii) 1,2-bis(diphenylphosphino)ethane catalyst

Article abstract of DOI:10.1002/cctc.201300740

The equilibrium position in formic acid/amine-CO₂ systems has been examined as a function of pressure and temperature under isochoric conditions. The homogeneous ruthenium(II)-1,2-bis(diphenylphosphino)ethane catalyst was active in both reactions, that is, in formic acid cleavage producing pure hydrogen and CO₂, as well as in carbon dioxide hydrogenation under basic conditions. High yields of formic acid dehydrogenation into H₂ and CO₂ are favored by low gas pressures and/or high temperatures, and H₂ uptake is possible at elevated H ₂-CO₂ pressures. These results take us one step closer to the realization of a practical H₂ storage-discharge device. Copyright

Full text of DOI:10.1002/cctc.201300740

CHEMCATCHEM
COMMUNICATIONS
DOI: 10.1002/cctc.201300740
Chemical Equilibria in Formic Acid/Amine-CO2 Cycles
under Isochoric Conditions using a Ruthenium(II)
1,2-Bis(diphenylphosphino)ethane Catalyst
Katerina Sordakis,[a] Matthias Beller,*[b] and Gꢀbor Laurenczy*[a]
Abstract: The equilibrium position in formic acid/amine–CO2
systems has been examined as a function of pressure and tem-
perature under isochoric conditions. The homogeneous ruthe-
nium(II)-1,2-bis(diphenylphosphino)ethane catalyst was active
in both reactions, that is, in formic acid cleavage producing
pure hydrogen and CO2, as well as in carbon dioxide hydroge-
nation under basic conditions. High yields of formic acid dehy-
drogenation into H2 and CO2 are favored by low gas pressures
and/or high temperatures, and H2 uptake is possible at elevat-
ed H2–CO2 pressures. These results take us one step closer to
the realization of a practical H2 storage–discharge device.
Our groups have both reported the catalytic decomposition
of FA using Ru-phosphine-based catalysts. However, if we ad-
dress the formation of FA from gaseous H2 and CO2, it is neces-
sary to overcome the thermodynamic barriers that result from
the uphill nature of this reaction. In this sense, CO2 hydrogena-
tion can be shifted towards the direction of FA by the addition
of bases, which are often amines. These FA–amine adducts are
also known to act as hydrogen donors in transfer-hydrogena-
tion reactions.[32–34]
Beller and co-workers demonstrated the generation of H2
from mixtures of FA and amines at room temperature.[35–37] Un-
precedented high rates at ambient temperature were obtained
by using in situ generated Ru-phosphine catalytic systems; the
highest activity obtained to date for the dehydrogenation of
FA was observed by applying a [RuCl2(benzene)]2 pre-catalyst
and a 1,2-bis(diphenylphosphino)ethane (dppe) ligand. In addi-
tion, the hydrogenation of CO2 (more precisely bicarbonate)
was also possible with this catalyst, which makes concrete ap-
plications of this system possible.
Introduction
A hydrogen economy, the long-term goal of many nations, can
potentially confer energy security along with environmental
and economic benefits.[1] Hydrogen storage is an essential ele-
ment in the hydrogen cycle, from production and delivery to
energy conversion and applications and, so far, storage issues
present a significant obstacle to the large-scale utilisation of
hydrogen as an energy carrier. Improved storage technologies
are needed to satisfy end-user expectations and foster con-
sumer confidence in hydrogen-powered alternatives.[2]
Knowledge of the position of the chemical equilibria in
Equation (1) is important for the construction of a practical H2
storage–discharge device.
HCOOH ꢀ NR3 Ð H2þCO2þNR3
ð1Þ
Recently, successful attempts reported the use of CO2 as
a hydrogen carrier.[3–14] Formic acid (FA), or more precisely for-
mates, can be produced by the reduction of CO2 in basic
media (that is from bicarbonates, carbonates) with H2 in the
presence of appropriate catalysts[15–25] and can also be convert-
ed back to H2 when needed. In this manner, FA can be regard-
ed as a liquid carrier of H2 if H2 is combined with CO2. There-
fore, the combination of CO2 reduction and FA or formate
As a result of entropy changes during the reaction, the equi-
librium is shifted largely to the direction of H2 and CO2 produc-
tion, which facilitates the generation of H2 from FA. However,
the reverse reaction requires a base and high pressures of H2
and CO2.[22] An essential feature for practical applications, in
which the gas pressure can be used as the driving force in
a downstream H2 separation and purification step, is the CO2
and H2 pressure effect on the catalytic activity and on the equi-
librium.[38] Fellay et al.[3] reported no inhibition of catalytic ac-
tivity during FA decomposition (up to a pressure of 750 bar)
on application of [Ru(H2O)6]2+ or RuCl3 and a water-soluble
meta-trisulfonated triphenylphosphine (mTPPTS) ligand in
a 4m HCOOH/HCOONa (9:1) aqueous solution. Beller and co-
workers[39] presented a highly active iron catalytic system for
the liberation of H2 from FA, by applying Fe(BF4)2·6H2O and
a tris[(2-diphenylphosphino)ethyl]phosphine [P(CH2CH2PPh2)3,
PP3] ligand in propylene carbonate (PC) solvent. This system
showed good tolerance towards CO2 pressure (up to 30 bar),
but significant catalytic activity loss has been detected even at
moderate H2 pressures. Therefore, it was considered to be of
major importance to examine the stability and activity of the
[RuCl2(benzene)]2 dppe catalytic system under elevated CO2
and H2 pressures.
dehydrogenation under mild conditions is
system.[26–31]
a CO2-neutral
[a] K. Sordakis, Prof. G. Laurenczy
Institut des Sciences et Ingꢀnierie Chimiques
Ecole Polytechnique Fꢀdꢀrale de Lausanne (EPFL)
1015 Lausanne (Switzerland)
Fax: (+41)21-693-9780
E-mail: gabor.laurenczy@epfl.ch
[b] Prof. M. Beller
Leibniz-Institut fꢁr Katalyse e.V. an der Universitꢂt Rostock
Albert Einstein Str. 29a, 18059 Rostock (Germany)
Fax: (+49)381-1281-51113
E-mail: matthias.beller@catalysis.de
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201300740.
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 96 – 99 96

CHEMCATCHEM
COMMUNICATIONS
www.chemcatchem.org
To realise a multiple-use H2 storage device, only closed ves-
sels that contain the catalyst and the chemical(s) used for stor-
age are appropriate. Charging would then be possible at ele-
vated pressure and discharging (H2 release) at low pressure.[22]
The thermodynamic requirement of such a device is that H2
must be involved in an equilibrium sufficiently mobile in the
expected pressure (and temperature) range.[41]
Results and Discussion
For our investigations, 5HCOOH/2triethylamine (NEt3) was
chosen as the substrate. It was observed that with the [RuCl2-
(benzene)]2/dppe catalyst, FA alone does not produce H2 and
CO2; a base as additive is required, in agreement with Beller’s
findings. The decomposition of an FA/amine mixture under at-
mospheric pressure, catalysed by [RuCl2(benzene)]2 and six
equivalents of dppe, yielded 100% FA conversion into H2 and
CO2 after heating at 408C. Even at 308C, no traces of FA could
be detected by using 1H and 13C NMR spectroscopy on the
equilibrium position. However, on performing the reaction
under isochoric conditions, in a closed reactor, the FA dehydro-
genation decreases to 44%, which indicates a shift of the
chemical equilibrium position towards the direction of FA for-
mation with increasing gas pressures. The pressure inside the
vessel increased rapidly and levelled off at 40 bar (Table 1,
To determine the position of the equilibrium in Equation (1),
the reaction was allowed to proceed from both directions in
presence of the Ru-dppe catalyst. For the FA decomposition re-
action, the volume of the 5HCOOH/2NEt3 mixture was varied
between 0.05 and 3.3 mL, keeping the catalyst concentration
and the temperature constant (Figure 1), and the pressure in-
Table 1. Effect of initial gas pressure on FA/amine decomposition under
isochoric conditions[a] with [RuCl2(benzene)]2 and a dppe ligand.[b]
Entry
pH
[bar]
pCO
[bar]
Final pressure
[bar]
Conversion
[%]
2
2
1
2
3
4
5
6
7
0
20
40
0
0
10
20
0
0
0
20
40
10
20
40
60
75
50
64
58
77
44
40
36
32
25
38
36
Figure 1. Equilibrium position for FA decomposition/formation in an amine
solution at 408C with [RuCl2(benzene)]2 (4.775 mmol) and dppe (6 equiv.) in
&
DMF (0.5 mL). FA decomposition reaction ( ), CO2 hydrogenation
~
reaction ( ).
crease as a result of gas evolution was monitored. The mea-
sured pressures were then converted to percentage FA conver-
sion by normalisation to the calculated pressure yielded by
100% FA decomposition. An equimolar mixture of H2 and CO2
was assumed to be produced by the FA decomposition reac-
tion according to previous investigations by Beller et al. with
the same catalytic system.[37] As CO2 is soluble in amine solu-
tions and as the FA conversion was calculated based on the
measured pressure, the amount of dissolved CO2 was deter-
mined by quantitative 13C NMR spectroscopy by using appro-
priate relaxation delays.[40]
[a] 1.5 mL of 5HCOOH/2NEt3 mixture, experiments performed in sapphire
NMR tubes, data correspond to the average of several experiments.
[b] 4.775 mmol [RuCl2(benzene)]2, Ru/P=1:6, pre-treatment of the catalyst
2 h in 0.5 mL DMF. Reproducibility was ꢁ10% relative to the given
values.
entry 1). The release of the gas pressure (to 1 bar) triggered
further FA decomposition until the dehydrogenation reached
100%, which indicates that the catalytic activity was not lost
with increasing pressure. To examine the effect of CO2 and H2
pressure on the equilibrium position, several experiments with
different initial CO2 and H2 pressures were performed (Table 1,
entries 2–5). It was found that the equilibrium position is shift-
ed towards the direction of FA formation because of the pres-
ence of both H2 and CO2 pressures, although the effect of the
latter was more pronounced. The equilibrium position was de-
termined by recording multiple quantitative 13C NMR spectra
(using appropriate delay times for complete relaxation[40]
during the measurements) after the reaction was initiated. If
the concentrations of the various species in solution (and also
the pressure) were found to be stable over an adequate period
of time, chemical equilibrium was considered to be attained. In
all the cases, the experiments could be reproduced with a max-
imum relative error of ꢁ10% (however, in about 90% of the
experiments conducted, the deviation was in the range of rela-
tive error ꢁ5% for the average values given in Table 1).
The CO2 concentration in the liquid phase varies according
to the applied temperature and pressure. High pressures and
low temperatures promote the dissolution of CO2. The concen-
tration of dissolved CO2 was evaluated by using NEt3 as an in-
ternal reference and then used to correct the conversions cal-
culated based on the pressure increase. The decomposed FA
quantities were confirmed, and the conversions were in good
agreement (ꢁ5%) with those determined by the integration
of the FA peaks obtained by quantitative NMR spectroscopy,
relative to those of NEt3. For the reverse reaction, a solution
that contained the catalyst and the amine was pressurised
with 50 bar H2 and 50 bar CO2, then heated and left to equili-
brate.
Low pressures allow the almost complete decomposition of
FA into H2 and CO2 (Figure 1), the rate of which depends on
the applied temperature as described below. However, an in-
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 96 – 99 97

CHEMCATCHEM
COMMUNICATIONS
www.chemcatchem.org
creased pressure leads to a shift of the equilibrium position to-
wards the direction of FA formation. Hence, the higher the ap-
plied pressure, the lower the FA conversion into H2 and CO2. It
can also be seen that the reverse reaction, that is, the produc-
tion of FA through the hydrogenation of CO2, is subject to
thermodynamic barriers as discussed above.
sition reaction on application of a catalytic system that con-
tained [RuCl2(benzene)]2 and dppe.[37] The reason is probably
the slow formation of the catalytically active species during
the first run.
The composite equilibrium constant at a given temperature
is defined by Equation (2).
The equilibrium composition is dependent on the tempera-
ture. Hence, the effect of temperature on the equilibrium posi-
tion in Equation (1) was examined (Figure 2). As the dehydro-
p p
H2 CO2
K0
¼
ð2Þ
eq
½HCOOHꢃL
Based on Equation (2) and by taking into account the gas–
liquid distribution of CO2, hence using the corrected CO2 pres-
sure, the enthalpy of reaction can be determined by combina-
tion of Equations (3) and (4) as usual:
DG ¼ ꢂRT lnðK0eqÞ
DG ¼ DHꢂTDS
ð3Þ
ð4Þ
The enthalpy of reaction was calculated from the slope of
the plot shown in Figure 4 and was found to be (+32.6ꢁ
0.5) kJmolꢂ1.
Figure 2. Effect of temperature on FA decomposition under isochoric condi-
tions with [RuCl2(benzene)]2 (4.775 mmol) and dppe (6 equiv.) in 5HCOOH/
2NEt3 solution (0.3 mL). Total solution volume=1.5 mL.
genation of FA is a slightly endothermic reaction (DH8=
+31.2 kJmolꢂ1), an increase in temperature is expected to pro-
mote the production of gases and, therefore, the decomposi-
tion of FA. Indeed, an increase in temperature increases the
rate of FA decomposition as well as the overall FA conversion
into H2 and CO2. The increase in FA conversion is more pro-
nounced at lower temperatures (Le Chatelier–Braun principle).
Despite the shift of equilibrium towards the direction of FA for-
mation with increasing pressure, high temperatures allow high
FA dehydrogenation yields.
Figure 4. Temperature dependence of the composite equilibrium constant
for FA/amine decomposition under isochoric conditions in the temperature
range of 10–1108C with [RuCl2(benzene)]2 (4.775 mmol) and dppe (6 equiv.)
in DMF (0.5 mL).
At a reaction temperature of 408C, a long induction period
is observed (Figure 3). This is in agreement with the findings of
Beller et al., who reported the same behaviour during the first
catalytic cycle of a 5FA/4N,N-dimethyl-n-hexylamine decompo-
From a practical point of view, it is essential that the desir-
able FA dehydrogenation yield can be obtained under isochor-
ic conditions. Therefore, the catalytic system needs to remain
active under elevated gas pressures that result from FA decom-
position. We have determined the equilibrium position in
Equation (1) as a function of pressure and shown that high
yields of FA dehydrogenation can be obtained even under ele-
vated pressures by increasing the reaction temperature.
Experimental Section
[RuCl2(benzene)]2 was prepared according to a literature method[42]
1
and checked by H NMR spectroscopy. The amine was distilled, and
DMF and the 5HCOOH/2NEt3 solution were degassed before use.
All other chemicals were commercial products and were used as
received. NMR spectra were recorded by using a Bruker DRX
400 MHz spectrometer.
Figure 3. Effect of temperature on the rate of FA decomposition under iso-
choric conditions with [RuCl2(benzene)]2 (4.775 mmol) and dppe (6 equiv.) in
All catalytic experiments were performed in 10 mm external diame-
ter, medium pressure sapphire NMR tubes.[43] In a typical experi-
~
&
^
^
*
5HCOOH/2NEt3 solution (0.3 mL), T=40 ( ), 50 ( ), 70 ( ), 80 ( ), 90 ( ),
*
1008C ( ).
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 96 – 99 98

CHEMCATCHEM
COMMUNICATIONS
www.chemcatchem.org
[8] J. F. Hull, Y. Himeda, W.-H. Wang, B. Hashiguchi, R. Periana, D. J. Szalda,
J. T. Muckerman, E. Fujita, Nat. Chem. 2012, 4, 383–388.
[9] W.-H. Wang, J. T. Muckerman, E. Fujita, Y. Himeda, ACS Catal. 2013, 3,
856–860.
[10] Y. Maenaka, T. Suenobu, S. Fukuzumi, Energy Environ. Sci. 2012, 5,
7360–7367.
ment, [RuCl2(benzene)]2 (4.775 mmol) and dppe (6 equiv.) were dis-
solved in DMF (0.5 mL) and stirred for 2 h at RT under a N2 atmo-
sphere. The pre-mixed solution of FA and amine was heated to
40.08C in a Schlenk tube and then transferred to an NMR tube.
The reaction was initiated by adding the catalyst to the mixture in
the sapphire NMR tube.
[11] D. J. Morris, G. J. Clarkson, M. Wills, Organometallics 2009, 28, 4133–
4140.
The tube was then thermostatted either with an electric heating
jacket or directly in the spectrometer, the temperature of which
was determined before and after measurement by using an exter-
nal temperature probe. The reaction was followed by monitoring
the pressure increase as a result of gas formation as a function of
time by using a pressure transducer connected to the tube via
a high pressure capillary, either manually or with an in-house Lab-
View 8.2 program with a NI USB 6008 interface, and/or in situ by
1H NMR spectroscopy. To verify that the kinetics of the reaction can
be followed by recording the change in pressure, FA decomposi-
[12] A. Majewski, D. J. Morris, K. Kendall, M. Wills, ChemSusChem 2010, 3,
431–434.
[13] S. Wesselbaum, U. Hintermair, W. Leitner, Angew. Chem. 2012, 124,
8713–8716; Angew. Chem. Int. Ed. 2012, 51, 8585–8588.
[14] J. H. Barnard, C. Wang, N. G. Berry, J. Xiao, Chem. Sci. 2013, 4, 1234–
1244.
[15] G. Laurenczy, F. Joꢂ, L. Nꢀdasdi, Inorg. Chem. 2000, 39, 5083–5088.
[16] F. Joꢂ, G. Laurenczy, P. Karꢀdy, J. Elek, L. Nꢀdasdi, R. Roulet, Appl. Orga-
nomet. Chem. 2000, 14, 857–859.
[17] J. Elek, L. Nꢀdasdi, G. Papp, G. Laurenczy, F. Joꢂ, Appl. Catal. A 2003,
255, 59–67.
[18] G. Kovꢀcs, G. Schubert, F. Joꢂ, I. Pꢀpai, Catal. Today 2006, 115, 53–60.
[19] R. Tanaka, M. Yamashita, K. Nozaki, J. Am. Chem. Soc. 2009, 131, 14168–
14169.
[20] C. Federsel, A. Boddien, R. Jackstell, R. Jennerjahn, P. J. Dyson, R. Scopel-
liti, G. Laurenczy, M. Beller, Angew. Chem. 2010, 122, 9971–9974;
Angew. Chem. Int. Ed. 2010, 49, 9777–9780.
[21] C. Federsel, R. Jackstell, A. Boddien, G. Laurenczy, M. Beller,
ChemSusChem 2010, 3, 1048–1050.
[22] G. Papp, J. Csorba, G. Laurenczy, F. Joꢂ, Angew. Chem. 2011, 123,
10617–10619; Angew. Chem. Int. Ed. 2011, 50, 10433–10435.
[23] C. Federsel, C. Ziebart, R. Jackstell, W. Baumann, M. Beller, Chem. Eur. J.
2012, 18, 72–75.
1
tion was followed by H NMR spectroscopy and pressure increase
1
simultaneously. The kinetic curve obtained from the H NMR spec-
tra that shows the decomposition of FA was compared to the data
obtained from pressure measurements. The measured pressures
were converted to percentage FA conversion by normalisation of
the final pressure to the final conversion determined by quantita-
tive NMR spectroscopy. As the data are in good agreement, it is
a valid method to follow the reaction only by recording the
change in pressure and recording NMR spectra before and after
the reaction to determine the conversion.
The integrals of the NMR spectroscopic signals that correspond to
FA and CO2 are proportional to their concentrations if appropriate-
ly long delay times (D1 >5T1) were chosen in the experiments. The
relaxation times (T1) of the CO2, HCOOH and NEt3 solutions were
determined by using the inversion recovery method.[41]
[24] M. S. Jeletic, M. T. Mock, A. M. Appel, J. C. Linehan, J. Am. Chem. Soc.
2013, 135, 11533–11536.
[25] C. A. Huff, M. S. Sanford, ACS Catal. 2013, 3, 2412–2416.
[26] P. G. Jessop, T. Ikariya, R. Noyori, Chem. Rev. 1995, 95, 259–272.
[27] P. G. Jessop, F. Joꢂ, C.-C. Tai, Coord. Chem. Rev. 2004, 248, 2425–2442.
[28] S. Fukuzumi, Eur. J. Inorg. Chem. 2008, 1351–1362.
[29] F. Joꢂ, ChemSusChem 2008, 1, 805–808.
[30] S. Enthaler, ChemSusChem 2008, 1, 801–804.
[31] S. Enthaler, J. von Langermann, T. Schmidt, Energy Environ. Sci. 2010, 3,
1207–1217.
For reactions under various pressures, the sapphire tube that in-
cluded all reagents was pressurised at room temperature with CO2
and/or H2. High pressure techniques, which include pressurisation
and depressurisation with H2 and CO2 were performed in a fume
hood with appropriate precautions. All experiments were per-
formed at minimum in triplicate.
[32] S. Gladiali, E. Alberico, Chem. Soc. Rev. 2006, 35, 226–236.
[33] S. Gladiali, E. Alberico in Transit. Met. Org. Synth. (Eds.: M. Beller, C.
Bolm), Wiley-VCH, Weinheim, 2004, pp. 145–166.
[34] M. Kitamura, R. Noyori in Ruthenium Org. Synth. (Ed.: Murahashi), Wiley-
VCH, Weinheim, 2004, pp. 3–52.
[35] A. Boddien, B. Loges, H. Junge, M. Beller, ChemSusChem 2008, 1, 751–
758.
[36] H. Junge, A. Boddien, F. Capitta, B. Loges, J. R. Noyes, S. Gladiali, M.
Beller, Tetrahedron Lett. 2009, 50, 1603–1606.
Acknowledgements
The Swiss National Science Foundation and EPFL are thanked for
financial support.
[37] A. Boddien, B. Loges, H. Junge, F. Gꢃrtner, J. R. Noyes, M. Beller, Adv.
Synth. Catal. 2009, 351, 2517–2520.
Keywords: amines
ruthenium · thermodynamics
·
dehydrogenation
·
hydrogen
·
[38] M. Grasemann, G. Laurenczy, Energy Environ. Sci. 2012, 5, 8171.
[39] A. Boddien, D. Mellmann, F. Gꢃrtner, R. Jackstell, H. Junge, P. J. Dyson, G.
Laurenczy, R. Ludwig, M. Beller, Science 2011, 333, 1733–1736.
[40] S. Moret, P. J. Dyson, G. Laurenczy, Dalton Trans. 2013, 42, 4353–4356.
[41] A. Boddien, C. Federsel, P. Sponholz, D. Mellmann, R. Jackstell, H. Junge,
G. Laurenczy, M. Beller, Energy Environ. Sci. 2012, 5, 8907–8911.
[42] M. A. Bennett, A. K. Smith, J. Chem. Soc. Dalton Trans. 1974, 233–241.
[43] A. Cusanelli, U. Frey, D. T. Richens, A. E. Merbach, J. Am. Chem. Soc.
1996, 118, 5265–5271.
[1] P. Tseng, J. Lee, P. Friley, Energy 2005, 30, 2703–2720.
[2] National Hydrogen Energy Roadmap, “DOE”, 2002.
[3] C. Fellay, P. J. Dyson, G. Laurenczy, Angew. Chem. 2008, 120, 4030–4032;
Angew. Chem. Int. Ed. 2008, 47, 3966–3968.
[4] B. Loges, A. Boddien, H. Junge, M. Beller, Angew. Chem. 2008, 120,
4026–4029; Angew. Chem. Int. Ed. 2008, 47, 3962–3965.
[5] S. Fukuzumi, T. Kobayashi, T. Suenobu, ChemSusChem 2008, 1, 827–
834.
Received: September 2, 2013
Revised: October 11, 2013
[6] S. Fukuzumi, T. Kobayashi, T. Suenobu, J. Am. Chem. Soc. 2010, 132,
1496–1497.
[7] Y. Himeda, Green Chem. 2009, 11, 2018–2022.
Published online on November 7, 2013
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 96 – 99 99