Base-free hydrogen generation from methanol using a bi-catalytic system

Article abstract of DOI:10.1039/c3cc47306f

A bi-catalytic system, in which Ru-MACHO-BH and Ru(H)₂(dppe) ₂ interact in a synergistic manner, was developed for the base-free dehydrogenation of methanol. A total TON > 4200 was obtained with only trace amounts of CO contamination

Full text of DOI:10.1039/c3cc47306f

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Base-free hydrogen generation from methanol
using a bi-catalytic system†
Cite this: Chem. Commun., 2014,
50, 707
Angele Monney, Enrico Barsch, Peter Sponholz, Henrik Junge, Ralf Ludwig and
`
Matthias Beller*
Received 24th September 2013,
Accepted 6th November 2013
DOI: 10.1039/c3cc47306f
www.rsc.org/chemcomm
A bi-catalytic system, in which Ru-MACHO-BH and Ru(H)2(dppe)2 (4700 hÀ1) using ruthenium complexes bearing a cooperative
interact in a synergistic manner, was developed for the base-free PNP pincer ligand10–16 in the presence of base. From an
dehydrogenation of methanol. A total TON > 4200 was obtained environmental point of view and for practical applications,
with only trace amounts of CO contamination (o8 ppm) in the it would be advantageous to produce H2 from methanol under
produced gas.
neutral conditions. Parallel to our work, Gru¨tzmacher et al.
described another ruthenium-based catalyst capable of dehydro-
Hydrogen is attracting increasing attention as an alternative genating a methanol–water mixture without base with a
energy carrier and is considered as a potential fuel of the TOF = 50 hÀ1 17
future.1 It is therefore conceivable that a ‘‘hydrogen economy’’2–6
Here, we report an improved base-free methanol dehydro-
.
may solve the problems related to the use of fossil fuels whose genation using the ruthenium-based PNP pincer complex
reserves are depleting. Hydrogen can be used as an energy source Ru-MACHO-BH18,19 (A1, see Table 1 for structure). This catalyst
for fuel cells or as a direct fuel for combustion with the release of is comparable to Ru-MACHO (A0),20–22 used for the base-
water as the only side product. An important challenge towards the mediated methanol reforming process, but lacks the chloride
implementation of a hydrogen economy is the safe, practical and ligand and therefore does not require a base for its activation.
economical storage of the produced hydrogen.7 Unfortunately, the Complex A1 showed comparable catalytic activity to A0 in the
compression or liquefaction of H2 is costly and requires substan- presence of a base (Table 1, entries 1 and 2). The addition of
tial energy. Alternatively, the chemical storage of hydrogen in triglyme23 as solvent was found to increase the solubility of the
liquid organic compounds and its release on demand hold great catalyst (cf. ESI†) and the gas generation was improved, reach-
promise. In this respect, methanol is an ideal candidate due to its ing a gas evolution rate of 149 mL hÀ1, corresponding to a
high hydrogen content (12.6% by weight) and its liquid state at
Table 1 Base-free hydrogen generation from aqueous methanola
room temperature.
The so-called methanol-reforming process allows for the
production of H2 from a mixture of methanol and water.8
However, the high temperatures needed (>200 1C) and the
co-production of carbon monoxide render current methanol
steam reforming processes unsuitable for applications in fuel
cells which only tolerate very low CO concentrations.
Recently, we described an aqueous phase methanol reforming
process at temperatures below 100 1C using a homogeneous
ruthenium catalyst with only traces of released carbon monoxide
(o10 ppm).9 Hydrogen generation was observed with excellent
catalyst turnover numbers (>350 000) and turnover frequencies
Catalyst loadingb
Entry Cat. [mmol]/[ppm]
Gas evolution
Base Solvent ratec [mL hÀ1
]
TON3h
1
2
3
4
5
6
7
A0 5/23
A1 5/23
A1 5/23
A1 5/23
A1 20/90
A1 45/203
A1 95/428
KOH None
KOH None
113
133
2064
2392
2743
32
70
74
KOH Triglyme 150
None Triglyme
1.7
None Triglyme 17
None Triglyme 37
None Triglyme 61
59
¨
¨
Leibniz-Institut fur Katalyse e.V. and der Universitat Rostock,
Albert-Einstein-Strasse 29a, 18059 Rostock, Germany.
E-mail: matthias.beller@catalysis.de; Fax: +49 381 1281 51113;
Tel: +49 381 1281 113
a
Reaction conditions: MeOH (9.0 mL), H2O (1.0 mL), solvent (4.0 mL)
when indicated, base (80 mmol) when indicated, catalyst (5–95 mmol),
Tset = 93.5 1C. The gas evolution was measured manually using burettes.
b
c
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c3cc47306f
ppm relative to MeOH. Average calculated over the first 3 hours.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 707--709 | 707

View Article Online
Communication
ChemComm
TOF > 900 hÀ1 after 3 hours (Table 1, entry 3). Although no
activity was observed in the absence of base with 5 mmol of A1
(Table 1, entry 4), an increase in catalyst loading allowed the
reaction to proceed. With 20 mmol of A1, the gas evolution was
measured at an average rate of 17 mL hÀ1 (Table 1, entry 5).
When the catalyst loading was increased to 45 mmol, a propor-
tional increase in the gas evolution rate was observed reaching
an average value of 37 mL hÀ1 (Table 1, entry 6). Finally, 95 mmol
of A1 induced the highest gas evolution rate with 61 mL hÀ1
(Table 1, entry 7). The increase in catalyst loading is believed
to significantly accelerate the decomposition of formic acid,
an intermediate suspected of deactivating the catalyst. Even
though the evolution of gas is increased by the use of higher
catalyst loadings, the overall productivity of the catalytic system
stayed low.
Table 2 Catalyst combinations for the base-free dehydrogenation of
methanola
c
Gas evolution rateb TOF1h TOF3h TOF7h
Entry Cat. A Cat. B [mL hÀ1
]
[hÀ1
]
[hÀ1
]
[hÀ1
]
In order to improve catalyst turnover numbers, the catalytic
step that involves the dehydrogenation of formic acid needs to
be accelerated in a different way. Therefore, we envisioned that
the addition of a second catalyst capable of efficiently dehydro-
genating formic acid would generate a bi-catalytic system that
allows the reaction to proceed under neutral conditions at low
catalyst loading. Although a similar approach was proposed by
Cole-Hamilton in 1989 for the dehydrogenation of ethanol
using a series of rhodium and ruthenium catalysts,24 only modest
improvements in H2 production were achieved in certain cases.
To our delight, the combination of Ru-MACHO-BH (A1) with
Ru(H)2(dppm)2 (B1, Table 2) at a catalyst loading as low as
5 mmol per catalyst (i.e. 45 ppm total [Ru] relative to MeOH) was
able to effectively dehydrogenate methanol under neutral con-
ditions (Table 2, entry 1). Despite a rapid gas evolution at the
beginning of the reaction with a turnover frequency of 239 hÀ1
after 1 hour, this catalytic system however slowly deactivated.
Gratifyingly, when B1 was replaced with the less constrained
Ru(H)2(dppe)2 (B2), a constant gas evolution was measured over
7 hours with an average TOF of 93 hÀ1 (Table 2, entry 2).25 Other
variations of the bisphosphine ligand on the ruthenium
catalyst B gave only a slower gas evolution rate (Table 2, entries
3–5). The combination of A1 with an iron catalyst decorated
with the tetraphos26 ligand showed no activity in the dehydro-
genation of methanol either as a defined (B6) or as an in situ
prepared (B7) catalyst (Table 2, entries 6 and 7, respectively)
even though these iron complexes were previously found to be
highly active in the base-free dehydrogenation of formic acid.27
When catalyst A1 was replaced by A2 (Table 2, entry 8) or A3
(Table 2, entry 9) in combination with B2 similar activities were
observed. Unlike previous observations,9 the substituents on
the PNP ligand (phenyl in A1 and isopropyl in A2 and A3) did
not influence considerably the catalytic activity of the system.
The use of the iridium PNP complex A4 lowered the gas
evolution rate (Table 2, entry 10) as previously observed in the
catalytic dehydrogenation of alcohols.22,28 A Ru–arene complex
bearing a triazolylidene ligand (A5), reported for its activity in
base-free alcohol oxidation,29 was also tested here in combi-
nation with B2 but showed no activity (Table 2, entry 11).
Interestingly, when catalyst A1 and catalyst B2 were used
1
2
3
4
5
6
7
8
A1
A1
A1
A1
A1
A1
A1
A2
A3
A4
A5
—
B1
B2
B3
B4
B5
B6
B7
B2
B2
B2
B2
B2
60
30
3.6
6.6
5.3
1.2
0.6
26
25
13
2.0
9.1
239
87
13
19
20
4
138
94
10
22
13
—
2
79
76
36
—
83
93
—
—
—
—
—
77
—
—
—
2
80
76
44
8
9
10
11
12
a
Reaction conditions: MeOH (9.0 mL), H2O (1.0 mL), triglyme (4.0 mL),
cat. A (5 mmol), cat. B (5 mmol), Tset = 93.5 1C. See ESI for graphical
b
c
representation. Average calculated over the first 3 hours. Calculated
using the theoretical value of 75% of H2 content of the gas phase. See
ESI for details.
produced little gas evolution (Fig. 1). More specifically, complex
B2 generated only about 30 mL of gas in 3 hours (Table 2, entry 12)
while A1 was essentially inactive (Table 1, entry 4). However,
when these two complexes are combined together they interact
in a synergistic manner and are able to dehydrogenate methanol
with an increased rate. A stepwise addition of the catalysts showed
that the effective dehydrogenation of methanol is triggered by
the addition of B2 to A1 or vice versa by the addition of A1 to
B2 (Fig. 1). Remarkably, the volume of gas generated by the
bi-catalytic system is significantly greater than the sum of the gas
volumes produced by each catalyst separately. The possibility
Fig. 1 Gas evolution from methanol catalysed by A1 (blue line) and B2
(red line), individually and together (green line). Rapid gas evolution starts
individually for the dehydrogenation of methanol, both of them when the two catalysts are mixed together (green dotted lines).
708 | Chem. Commun., 2014, 50, 707--709
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
of a ligand exchange process between the catalysts was ruled out The synergistic interaction between Ru-MACHO-BH and Ru(H)2-
since neither the addition of dppe (Ph2PCH2CH2PPh2) to complex (dppe)2 allows the reaction to proceed at low catalyst loading.
A1 nor the addition of the PNP ligand (HN(CH2CH2PPh2)2) to B2 The unprecedented mild reaction conditions and the low CO
altered the activity of the individual catalysts. This positive contamination of the produced gas render this system interesting
interaction between the catalysts however decreases at higher for a combination with PEM fuel cells. Additional experiments
catalyst loading (cf. ESI†). This is tentatively explained by the low using in situ IR spectroscopy supported the proposed reaction
solubility of B2 at higher catalyst loading, preventing an optimal pathway through formaldehyde and formic acid.
interaction between the catalysts.
Notes and references
1 J. O. M. Bockris, Science, 1972, 176, 1323.
In order to assess the role of formic acid in this methanol
reforming process, HCOOH was added to the reaction mixture.
The bi-catalytic system A1/B2 (5 mmol of each catalyst) dehydro-
genated quickly and completely the extra formic acid (0.5 mL)
with a gas evolution rate of 464 mL hÀ1, corresponding to a
TOF = 940 hÀ1 (cf. ESI†).30–32 A positive synergistic effect
between the two catalysts was observed here as well, although
to a lower extent. Catalyst B2 (5 mmol) on its own was able
to decompose the added HCOOH with a gas evolution rate of
352 mL hÀ1. While formic acid decomposition by A1 at a
catalyst loading of 5 mmol was slow (31 mL hÀ1), an increased
loading (45 mmol) allowed for the efficient and complete HCOOH
decomposition with a gas evolution rate of 306 mL hÀ1. When
the added formic acid was fully converted, methanol dehydro-
2 J. O. M. Bockris, Int. J. Hydrogen Energy, 2013, 38, 2579–2588.
´
´
´
´
3 V. A. Blagojevic, D. M. Minic, D. G. Minic and J. G. Novakovic,
Hydrogen Economy: Modern Concepts, Challenges and Perspectives,
2012.
4 J. Andrews and B. Shabani, Procedia Eng., 2012, 49, 15–25.
5 N. Armaroli and V. Balzani, ChemSusChem, 2011, 4, 21–36.
6 G. W. Crabtree, M. S. Dresselhaus and M. V. Buchanan, Phys. Today,
2004, 57, 39–44.
7 P. Jena, J. Phys. Chem. Lett., 2011, 2, 206–211.
8 D. R. Palo, R. A. Dagle and J. D. Holladay, Chem. Rev., 2007, 107,
3992–4021.
9 M. Nielsen, E. Alberico, W. Baumann, H.-J. Drexler, H. Junge,
S. Gladiali and M. Beller, Nature, 2013, 495, 85–89.
10 R. Langer, I. Fuchs, M. Vogt, E. Balaraman, Y. Diskin-Posner, L. J. W.
Shimon, Y. Ben-David and D. Milstein, Chem.–Eur. J., 2013, 19,
3407–3414.
genation was carried out at the same rate as before the addition 11 S. Musa, S. Fronton, L. Vaccaro and D. Gelman, Organometallics,
2013, 32, 3069–3073.
12 K.-N. T. Tseng, J. W. Kampf and N. K. Szymczak, Organometallics,
of HCOOH, suggesting no significant deactivation of the cata-
lysts. These results support our initial hypothesis that either an
2013, 32, 2046–2049.
increase in catalyst loading or the addition of a second parti- 13 K. E. Allen, D. M. Heinekey, A. S. Goldman and K. I. Goldberg,
Organometallics, 2013, 32, 1579–1582.
14 S. Smith, D. Spasyuk and D. G. Gusev, Angew. Chem., Int. Ed., 2012,
cular catalyst can improve the rate of the methanol dehydro-
genation by accelerating the decomposition of in situ formed
51, 2772–2775.
formic acid.
15 S. Schneider, J. Meiners and B. Askevold, Eur. J. Inorg. Chem., 2012,
412–429.
Further insights into the mechanism of the methanol dehydro-
genation were provided by in situ infrared spectroscopy. During
16 J. I. van der Vlugt and J. N. H. Reek, Angew. Chem., Int. Ed., 2009, 48,
8832–8846.
the reaction, a vibration band was detected at 1730 cmÀ1 and was 17 R. E. Rodrıguez-Lugo, M. Trincado, M. Vogt, F. Tewes, G. Santiso-
´
Quinones and H. Gru¨tzmacher, Nat. Chem., 2013, 5, 342–347.
18 Takasago International Corp., PCT Int. Appl., WO2011048727A,
attributed to formic acid since its shape and position were very
similar to free formic acid under the same conditions (cf. ESI†).
2011.
This signal appeared as soon as the two catalysts were added and 19 For another complex with a [Ru(H)(H–BH3)] motif, see: T. Ohkuma,
˜
M. Koizumi, K. Muniz, G. Hilt, C. Kabuto and R. Noyori, J. Am. Chem.
its intensity continuously increased until it reached a steady-state
after about 20 hours. Our catalytic system therefore seems to
reach an equilibrium in which a certain amount of formic acid is
present. The appearance of this band was accompanied by a
negative contribution at 1650 cmÀ1 which we attributed to the
Soc., 2002, 124, 6508–6509.
20 W. Kuriyama, T. Matsumoto, O. Ogata, Y. Ino, K. Aoki, S. Tanaka,
K. Ishida, T. Kobayashi, N. Sayo and T. Saito, Org. Process Res. Dev.,
2012, 16, 166–171.
21 Z. Han, L. Rong, J. Wu, L. Zhang, Z. Wang and K. Ding, Angew.
Chem., Int. Ed., 2012, 51, 13041–13045.
consumption of water since this region of the spectrum hosts the 22 M. Nielsen, H. Junge, A. Kammer and M. Beller, Angew. Chem.,
Int. Ed., 2012, 51, 5711–5713.
23 Triethylene glycol dimethyl ether.
24 D. Morton, D. J. Cole-Hamilton, I. D. Utuk, M. Paneque-Sosa and
frequency of the deformation vibration of water. The stepwise
addition of A1 to B2 and of B2 to A1 was also followed by IR
spectroscopy and led to comparable conclusions to those
obtained from gas evolution measurements (cf. ESI†). Further
experiments involving the addition of formic acid and formalde-
M. Lopez-Poveda, J. Chem. Soc., Dalton Trans., 1989, 489–495.
25 These optimal conditions were reproduced several times with a
maximum error of 6%.
26 Tris-[(2-diphenylphosphino)ethyl]phosphine.
¨
hyde supported the proposed reaction pathway via these two 27 A. Boddien, D. Mellmann, F. Gartner, R. Jackstell, H. Junge,
intermediates9 and suggested the release of the first molecule of
P. J. Dyson, G. Laurenczy, R. Ludwig and M. Beller, Science, 2011,
333, 1733–1735.
hydrogen as a rate limiting step (cf. ESI†). Finally, a long term
28 M. Nielsen, A. Kammer, D. Cozzula, H. Junge, S. Gladiali and
experiment was performed resulting in the production of a total
gas volume of about 1400 mL. This corresponds to a hydrogen
yield of 26% (relative to H2O) and a TON > 4200 (cf. ESI†).
Importantly, less than 8 ppm of carbon monoxide was observed
by gas phase GC measurements throughout the reaction course.
In conclusion, we developed a base-free bi-catalytic system for
the production of H2 from aqueous methanol at low temperature.
M. Beller, Angew. Chem., Int. Ed., 2011, 50, 9593–9597.
29 D. Canseco-Gonzalez and M. Albrecht, Dalton Trans., 2013, 42,
7424–7432.
30 I. Mellone, M. Peruzzini, L. Rosi, D. Mellmann, H. Junge, M. Beller
and L. Gonsalvi, Dalton Trans., 2013, 42, 2495–2501.
31 M. Grasemann and G. Laurenczy, Energy Environ. Sci., 2012, 5,
8171–8181.
¨
32 B. Loges, A. Boddien, F. Gartner, H. Junge and M. Beller, Top. Catal.,
2010, 53, 902–914.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 707--709 | 709