Immobilized Ru-Pincer Complexes for Continuous Gas-Phase Low-Temperature Methanol Reforming-Improving the Activity by a Second Ru-Complex and Variation of Hydroxide Additives

Article abstract of DOI:10.1002/ejic.202100042

Ru-pincer complexes were immobilized as supported liquid phase (SLP) materials to allow the methanol reforming reaction as continuous gas phase process. Under reaction conditions, the liquid phase forms from the hydroxide coating. Several hydroxides were screened and CsOH showed highest activity compared to the standard KOH coating. The well-known Ru-pincer complex carbonylchlorohydrido [bis(2-di-i-propylphosphinoethyl)amine]ruthenium(II) is limited in catalyzing the final step of the methanol reforming. Addition of a second complex, having a methylated backbone in the pincer-ligand, could overcome these limitations. Significant enhancement of the overall catalytic activity was observed.

Full text of DOI:10.1002/ejic.202100042

Full Papers
doi.org/10.1002/ejic.202100042
Immobilized Ru-Pincer Complexes for Continuous Gas-
Phase Low-Temperature Methanol Reforming-Improving
the Activity by a Second Ru-Complex and Variation of
Hydroxide Additives
[a]
[a]
[b, c]
[b]
Christian H. Schwarz, Dominik Kraus, Elisabetta Alberico,
Marco Haumann*
Henrik Junge, and
[a]
Ru-pincer complexes were immobilized as supported liquid
phase (SLP) materials to allow the methanol reforming reaction
as continuous gas phase process. Under reaction conditions,
the liquid phase forms from the hydroxide coating. Several
hydroxides were screened and CsOH showed highest activity
compared to the standard KOH coating. The well-known Ru-
pincer complex carbonylchlorohydrido [bis(2-di-i-propylphos-
phinoethyl)amine]ruthenium(II) is limited in catalyzing the final
step of the methanol reforming. Addition of a second complex,
having a methylated backbone in the pincer-ligand, could
overcome these limitations. Significant enhancement of the
overall catalytic activity was observed.
Introduction
NH3 to more complex structures like dibenzyltoluene. A
promising approach to hydrogen storage is the generation of
[14]
Renewable energies like wind and solar are subject to temporal
fluctuation and efficient storage of the excess energy is
methanol by hydrogenation of CO2.
Methanol has the
advantage of a relatively high hydrogen content of 12.5 wt% in
the molecule. In addition, methanol is liquid under ambient
conditions and can be stored in tanks without modifications of
the existing infrastructure. In a decentralized scenario, even for
mobile applications in transport, the hydrogen can be released
[
1]
mandatory for an efficient green energy scenario. Excess
energy can be stored in form of batteries, by converting
electrical energy into other forms of energy (e.g. by means of a
pumped storage power station) or by producing a chemical
energy carrier. Hydrogen, for example, can be used as a suitable
[15]
again by methanol steam reforming. The resulting reformate
has a hydrogen to carbon dioxide product ratio of 3:1
according to Scheme 1.
[2]
chemical energy carrier. Its generation by electrolysis and
conversion in fuel cells are established technologies along the
[3,4]
process chain.
However, storing pure hydrogen is challeng-
From a thermodynamic point of view, high equilibrium
conversions of reaction (R1) can be obtained at higher temper-
atures since the reaction is endothermic. However, in case the
reaction temperatures exceed approx. 200°C, the stronger
endothermic methanol decomposition (R2) will result in high
CO levels >1%, which would render the gas mixture not
[5]
ing, given the low volumetric storage capacity. In recent years,
storage of hydrogen in chemical form by catalytic hydro-
genation of organic molecules, so called liquid organic hydro-
gen carriers (LOHC), became an interesting alternative.
LOHC materials, hydrogen can be liberated by dehydrogenation
catalysis. Today, hydrogen storage materials range from simple
[6–13]
From
[16,17]
suitable for most fuel cell applications.
Since CO is
converted in the presence of water via reaction (R3), lower
reaction temperatures will lead to higher hydrogen purity.
Heterogeneous catalysts, however, will drastically lose activity
at such temperatures, making the process not feasible from a
[
a] C. H. Schwarz, D. Kraus, Dr. M. Haumann
Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Lehrstuhl für
Chemische Reaktionstechnik (CRT),
Egerlandstr. 3, 91058 Erlangen, Germany
E-mail: marco.haumann@fau.de
[17,18]
kinetic point of view.
https://www.crt.tf.fau.de/
In recent years, a wide range of homogeneous catalysts has
[
[
b] Dr. E. Alberico, Dr. H. Junge
successfully been employed in the field of low temperature
Leibniz-Institut für Katalyse, e. V.,
[19–23]
aqueous phase methanol reforming (AMR).
These catalysts,
Albert-Einstein Straße 29a, 18059 Rostock, Germany
c] Dr. E. Alberico
some of which are shown in Figure 1, facilitate the release of
hydrogen with carbon dioxide as a by-product from methanol
and water at temperatures below 100°C. Hence, the CO
contamination level can be as low as <1 ppm. One of the most
promising catalyst systems, which combines high productivity
Istituto di Chimica Biomolecolare, Consiglio Nazionale delle Ricerche,
tr. La Crucca 3, 07100 Sassari, Italy
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/ejic.202100042
Part of the Supported Catalysts Special Collection.
(turnover number >350,000), good long term stability (>
©
2021 The Authors. European Journal of Inorganic Chemistry published by
Wiley-VCH GmbH. This is an open access article under the terms of the
Creative Commons Attribution Non-Commercial NoDerivs License, which
permits use and distribution in any medium, provided the original work is
properly cited, the use is non-commercial and no modifications or adap-
tations are made.
3 weeks) and low concentrations of carbon monoxide as side
product (<10 ppm) is the ruthenium-based PNP-pincer catalyst
[24]
3
published by Beller and co-workers. Base metal homoge-
neous catalysts with similar pincer motifs (see e.g. 4) have also
Eur. J. Inorg. Chem. 2021, 1745–1751
1745
© 2021 The Authors. European Journal of Inorganic Chemistry
published by Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100042
Scheme 1. Reaction network of methanol reforming including the side reaction (R2) leading to CO formation and the water-gas shift reaction (R3).
Figure 1. Selected examples of transition metal complexes reported for the aqueous phase methanol reforming.
been investigated. However, such catalysts were not as active
as the ruthenium-based system and suffered from a high
oxygen sensitivity.
molar ratio of 3:5 to reach higher turn-over frequencies (TOF)
by up to 50% compared to the use of 3 alone. Furthermore, the
absolute amount of base could be reduced to 12.5% compared
to initial studies using 80 mmol KOH.
From a mechanistic point of view, the AMR reaction
proceeds in three steps via formaldehyde and formic acid to
finally yield a product composition of carbon dioxide and
hydrogen in a ratio of 1:3. When only complex 3 is applied, the
rate determining step of the reaction sequence was found to be
the final dehydrogenation of formate (step 3 in Scheme 2),
In our previous work complex 3 was successfully immobi-
[
27]
lized using the supported liquid phase (SLP) concept. Here,
the homogenous complex 3 is entrapped in a thin film of KOH,
which is liquid under reaction conditions due to dissolved
amounts of water and methanol. This aqueous KOH film is
dispersed over the large inner surface area of a porous support
material, in this case alumina. These SLP catalysts were then
[25]
resulting in the accumulation of formate during the reaction.
By screening for a second catalyst to lower the activation barrier
of this step, Beller and co-workers were able to improve the
employed in
a continuous vapor-phase methanol steam
[26]
dehydrogenation of formate.
reforming process (MSR). KOH is required for catalytic activity as
well as catalyst complex activation. Dissolving 2 wt% of
complex 3 in 10 wt.–% of KOH on porous alumina allowed
The best complex was based on a simple modification in
the backbone of the PNP-pincer ligand, indicated by structure 5
in Figure 1. Although the structures of complexes 3 and 5 are
very similar, their catalytic properties differ significantly. The
rate determining step of 5 is the dehydrogenation of methanol
to formaldehyde (step 1 in Scheme 2). The fact that both
complexes accelerate the rate determining step of their
counterpart is utilized in the bi-catalytic system with a 1:1
À 1 À 1
À 1
reasonable activity of 14 molH2 molRu
h
(shorth ) and a high
selectivity of 99.0% CO (corresponding to less than 2500 ppm
2
CO) at 150°C and 1 bar. More important for continuous
processes, the stability of the catalyst was maintained over 70 h
time on stream (TOS) with minor deactivation. Due to the
crucial role of the base additive for catalyst performance, we
became interested in investigating the influence of different
base additives for gas-phase methanol reforming. In addition,
we employed the combination of complexes 3 and 5, the so-
called bi-catalytic system, in order to investigate the impact of
complex 5 on overall performance.
Results and Discussion
In a first series of experiments we investigated different alkali
metal hydroxides as basic additives for the MSR catalyst. The
Scheme 2. Proposed reaction sequence for methanol reforming.
Eur. J. Inorg. Chem. 2021, 1745–1751
www.eurjic.org
1746
© 2021 The Authors. European Journal of Inorganic Chemistry published
by Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100042
2
Figure 2. Activity (filled symbols) and CO selectivity (open symbols) over
time on stream for methanol reforming using Ru-3-SLP catalysts with
different hydroxide coatings of KOH (black squares) and CsOH (red circles).
Figure 3. Maximum activity (TOFmax) of immobilized Ru-3-SLP catalysts in
continuous methanol reforming with different alkali hydroxides and different
Reaction conditions: pabs: 1 bar; pMeOHjH2O: 0.5 bar; T: 150°C; mSLP: 5.0 g; Qtot
:
À 1
À 1
5
0 mlmin . Catalyst composition: 50 μmolRu
g
Support ; 10 wt% MOH-loading;
loadings. Reaction conditions: T=150°C, pabs: 1 bar; pMeOH j H2O: 0.5 bar; mSLP
:
Aluminum oxide support material 1.0 mm Ø.
À 1
À 1
Support
;
5
4
.0 g; Qtot: 50 mlmin . Catalyst composition: 50 μmolRu
g
À 1
0 molMOH molRu (white bars, uniform amount of hydroxide); 10 wt% MOH-
loading (filled bars, uniform mass of hydroxide); Aluminum oxide support
material 1.0 mm Ø.
catalytic activity and CO2 selectivity was monitored over
prolonged time on stream as depicted for KOH and CsOH
coated Ru-3-SLP catalysts in Figure 2. Compared to the
previously reported KOH-based SLP catalyst, the activity was
discernable difference between the investigated bases (see
Figure S3 in the Supporting Information). Multiple studies have
shown the vital role of the base in the catalytic cycle of the Ru-
pincer complex, therefore a higher concentration of hydroxide
ions could lead to a more active catalyst. Indeed, the activity
À 1
À 1
now almost twice as high (29 h vs. initial 14 h ) due to better
handling of the catalyst and removal of dissolved oxygen from
both methanol and water. The CO2 selectivity of >99.9%
corresponded to 250 ppm CO in the product gas mixture. The
value would not allow direct use in low temperature PEM fuel
cells, however, established HT-PEMFC could be operated with
values correlated with the pK values of the various hydroxides
B
as shown in Table 1.
[16]
this gas stream. As can be seen from Figure 2, the activity of
the KOH coated Ru-3-SLP catalyst declined over time on stream.
As deactivation rate we defined the loss of activity in % per h,
On the other hand, the solubility of water and methanol in
the basic coating is important for the catalyst performance. The
hygroscopic nature of the salt leads to an absorption of vapors
into the coating and reduces the melting point of the salt to
appropriate levels. With increasing solubility and subsequently
higher hygroscopy a higher concentration of substrates is
available for the catalyst. Again, the activity values increased
with increasing hygroscopy. Most likely a combination of both
À 1
resulting in a value of 0.4%h for KOH. The second data set
included in Figure 3 represents the same Ru-SLP catalyst with
CsOH as hydroxide coating. As can be seen, this catalyst
À 1
showed a marginally lower activity at 23 h , while maintaining
a constant activity. The selectivity towards CO is comparable to
2
the SLP with KOH coating at 99.8%.
factors, pK and hygroscopic nature, is crucial for achieving the
B
To investigate the effect of the nature of hydroxide on the
catalytic performance further, Ru-3-SLP catalysts were prepared
with different hydroxide coatings, including LiOH, NaOH, CsOH
and RbOH. Since the density of these hydroxides varied
significantly, we prepared one set of SLP materials with a
constant weight loading of 10 wt% hydroxide, while the other
set was prepared with a constant molar ratio of hydroxide to
Ru-complex 3 of 40.
best performance.
Because of the different densities and molar masses of the
metal hydroxides the dry pore filling degrees of the SLP
Table 1. Selected physical parameters of used basic additives for Ru-3-SLP
catalysts and corresponding TOF of SLP catalyst. pK data taken from.
B
[28]
À 1 [a]
À 1 [b]
À 1 [c]
Hydroxide
TOF [h
]
pK
B
S
water [gL
]
Density [g L
]
Of all hydroxides tested, CsOH yielded far more active Ru-3-
À 1
LiOH
NaOH
KOH
RbOH
CsOH
4.5
8.4
29.4
29.7
50.2
0.18
128
1.5
2.1
2.0
3.0
3.7
SLP catalysts with peak performances of 50 h , while LiOH and
À 0.56
À 1.10
À 1.40
À 1.76
1090
1130
1800
3000
NaOH led to rather low activities. RbOH mshowed comparable
activity to the KOH system. This increase in activity in order of
Li<Na<K<Rb<Cs can on the one hand be attributed to the
increased base strength. The selectivity towards CO2 ranged
from 99.5 to 99.8% for the different hydroxides, with no
[a] TOF=turnover frequency. [b] Solubility of hydroxide in water at room
temperature. [c] Dry density of hydroxide.
Eur. J. Inorg. Chem. 2021, 1745–1751
www.eurjic.org
1747
© 2021 The Authors. European Journal of Inorganic Chemistry published
by Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100042
In line with the recently reported bi-catalytic system in
liquid phase methanol reforming we adapted our Ru-SLP
catalyst accordingly. By employing a SLP system consisting of
the two different Ru-pincer complexes 3 and 5 in a 1:1 ratio,
the TOF in gas-phase methanol reforming could be increased
À 1
À 1
from 35 h (pure Ru-3-SLP) to 85 h , resembling an increase
by 115%. (see Figure 5) The CO selectivity slightly declined
2
with increasing activity of the bi-metallic Ru-3-5-SLP system.
This is in agreement with previous studies from Beller and co-
workers of such bi-metallic systems in AMR. When using 5 in
excess compared to 3, both the activity and selectivity declined
drastically. While 3 prefers a high pH of 10 and beyond, 5
prefers neutral or slightly acidic conditions. This might also be
the reason for the poor performance of pure Ru-5-SLP with
basic coating. Interestingly, the lowest deactivation of
À 1
À 0.16%h is observed for the 1:1 ratio showing highest
activity, while pure Ru-3-SLP and Ru-5-SLP deactivated faster
À 1
with À 0.3 and À 0.62%h , respectively.
To investigate if the observed beneficial effect of the CsOH
coating can be expanded to the bi-catalytic system, a series of
Ru-3-5-SLP catalysts with different ratios of 3:5 with KOH and
CsOH as basic additives was studied. As seen in Figure 5, for a
Figure 4. Methanol reforming using Ru-3-SLP catalysts with different load-
ings of KOH (black) and CsOH (red) hydroxides. Activity (squares, left) and
averaged change in activity (triangles, right) of immobilized Ru-Pincer-based
SLP catalysts in continuous methanol reforming. Average values of a) 8, b) 3
1
:1 ratio there was virtually no difference between the KOH
and c) 4 measurements are shown for CsOH. Reaction conditions: T=150°C,
À 1
À 1
p
abs: 1 bar; pMeOHjH2O: 0.5 bar; mSLP: 5.0–6.5 g; Qtot: 50 mlmin . Catalyst
and CsOH coating, both having an initial TOF of around 70 h .
While the activity of the KOH-based Ru-SLP catalyst declined
linearly with reduced amount of 5, the catalyst with CsOH
showed a vastly different trend. A distinct activity maximum
À 1
composition: 50 μmolRu
g
Support ; Aluminum oxide support material 1.0 mm
Ø.
À 1
was found at a 5:1 ratio of 3:5 with a TOF of 91 h . Over the
catalysts vary vastly. While for LiOH, NaOH, KOH and RbOH the
diverging pore filling degrees showed only minor impact on the
activity of the SLP catalyst, the difference in CsOH loading more
whole range the more basic additive CsOH resulted in more
active Ru-3-5-SLP catalysts.
The CO selectivity was neither influenced by the ratio or
2
À 1
than doubled the TOF of the catalyst from 23 to 50 h (see
the hydroxide coating to a large extend, showing a selectivity
Figure 3). Therefore, the pore filling degree was investigated in
more detail. Figure 4 compares the activity and stability (shown
as average change in activity) over different loadings of either
KOH or CsOH.
of >99.8% (equal to 1250 ppm CO) throughout all experiments.
The values of the pore filling can only be determined based
on the synthesis protocol. The real pore filling degree at
reaction conditions was most likely far higher due to film
swelling from dissolved water and methanol. The Ru-3-SLP
catalyst with KOH coating showed only a minor dependency of
its activity upon changing the base loading. Merely a 30%
improvement was achieved when increasing the pore filling
from 12 to 26 vol%. This is also consistent to previously
[27]
published investigations using the same Ru-3-SLP system.
Higher loading beyond 26 vol% resulted in a declined activity,
most likely due to pore blocking by the now strongly swollen
hydroxide film. Simultaneously, the stability dropped with
increasing hydroxide loading, which might hint an elevated
leaching of active material from fully liquid-filled pores. On the
contrary when employing CsOH as a coating material an
increase from 7 to 42 vol% pore filling consecutively led to
more active catalysts. The TOF seemed to follow an asymptotic
trend without a distinct maximum. At the same time the most
stable catalyst was achieved with a CsOH pore filling degree of
around 20 vol%. Hence, a compromise between activity and
long-term stability has to be accepted.
Figure 5. Continuous gas-phase methanol reforming using bi-metallic Ru-3-
5
-SLP catalysts with different ratios of 3:5 and varying loadings of KOH
(black) and CsOH (red). Activity (squares, TOF) and selectivity towards CO2
(
p
5
circles) over time on stream. Reaction conditions: T=150°C, pabs: 1 bar;
MeOH j H2O: 0.5 bar; mSLP: 5.7 g (CsOH), 5.0 g (KOH), 1.5 g (KOH 1:1); Qtot
:
À 1
À 1
À 1
0 mlmin . Catalyst composition: 50 μmolRu
g
Support ; 40 molKOH molRu
;
À 1
10 wt% KOH-loading, 30 molCsOH molRu ; 20 wt% CsOH-loading; Aluminum
oxide support material 1.0 mm Ø.
Eur. J. Inorg. Chem. 2021, 1745–1751
www.eurjic.org
1748
© 2021 The Authors. European Journal of Inorganic Chemistry published
by Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100042
The influence of both hydroxide and Ru complex on activity
Table 2. Activity (TOF) and effective activation energy (EA,eff) of immobi-
lized, mono- and bi-catalytic Ru-pincer-based SLP catalysts in continuous
methanol steam reforming with KOH or CsOH basic additive.
is shown in Figure 6 as a function of temperature. The activity
[a]
at 120°C with KOH as hydroxide coating was used to normalize
as
TOF
À 1
À 1
Catalyst
Hydroxide Temperature [°C] TOF [h
]
E
A,eff [kJ mol
]
the data sets. In doing so, an activity enhancement E
function of hydroxide or catalyst type could be calculated, the
results are summarized in Figure 6. The effect of Ru-complex
combination in methanol reforming is highlighted by the
strong enhancement of up to 1.33 when going from Ru-3-SLP
to Ru-3-5-SLP while keeping the KOH hydroxide coating. This
Ru-3-SLP
KOH
120
9
59
68
36
59
1
1
1
1
30
40
50
60
14
23
34
48
11
22
38
57
75
21
28
35
46
58
23
37
57
80
97
Ru-3-SLP
CsOH
120
1
1
1
1
30
40
50
60
1
33% increase in activity is observed at lowest temperature of
1
20°C. A steep decline in the enhancement is observed when
going to higher temperatures, limiting the enhancement to a
mere 20% at 160°C. This is in line with the different activation
energies obtained from liquid-phase reforming. The higher
Ru-3-5-SLP KOH
Ru-3-5-SLP CsOH
120
1
1
1
1
30
40
50
60
À 1
activation energy of complex 3 (82 kJmol ) leads to higher rate
acceleration with increasing temperature. At the same time, the
120
temperature dependency of complex 5 is lower, with only
1
1
1
1
30
40
50
60
À 1
67 kJmol . Hence, at higher temperatures the catalyst complex
3
is active enough to catalyze all three steps in the MSR cycle
effectively and the presence of complex 5 is not really required.
This picture is obviously different at lower temperatures, where
complex 5 helps to accelerate the final step in the cycle.
The effect of hydroxide is not as strong for activity
enhancement, approx. 80% can be obtained at 150°C for both
the Ru-3-SLP and the Ru-3-5-SLP system when changing from
KOH to CsOH.
[
a] Reaction conditions: pabs: 1 bar; pMeOHjH2O: 0.5 bar; mSLP: 5.7 g (CsOH); Qtot
:
À 1
À 1
5
0 mlmin ; T: 120–160°C. Catalyst composition: 50 μmolRu
g
Support ; 1:1
À 1
À 1
Ru-3-5-SLP; 91 molKOH molRu ; 20 wt% KOH-loading, 31 molCsOH molRu
;
2
0 wt% CsOH-loading; Aluminum oxide support material 1.0 mm Ø.
The long-term stability of the different catalysts followed no
distinct trend (see Supporting Information for details).
The different trends and correlations between the mono- Conclusion and outlook
and bi-catalytic systems, as well as different basic additives can
have a plethora of physical or chemical reasons. While the
catalytic cycle of the Ru-Pincer complex 3 is strongly dependent
on the concentration of hydroxide ions in solution, 5 favors a
In the present work we have expanded the investigations on
the novel SLP catalyst for methanol steam reforming. Variation
of the base showed a strong impact on the overall performance
different pH regime. Additionally the different pK values and
B
hygroscopic nature of the metal hydroxides KOH and CsOH
lead to different pore filling degrees at reaction conditions and
concentrations of reactants in the active film. In Table 2 the four
distinct catalytic systems have been investigated over a temper-
ature range of 120–160°C.
From the set of temperature dependent studies, the
effective activation energies could be calculated for the mono-
catalytic Ru-3-SLP in both hydroxides as well as the bi-catalytic
Ru-3-5-SLP in both hydroxides. Compared to Ru-3-SLP, the bi-
catalytic system showed a lower E
with a stronger decline by 23 kJmol from 59 to 36 kJmol in
KOH. The effect is smaller in CsOH, where the effective
in both KOH and CsOH,
A,eff
À 1
À 1
À 1
activation energy is lowered by 9 kJmol only. Since several
effects-e.g. pore diffusion, liquid film viscosity, solubility in the
liquid mixture – contribute to the effective activation energy in
SLP catalysis, no clear insight on molecular level can be made
here. In liquid-phase methanol reforming, activation energies of
Figure 6. Activity enhancement for Ru-SLP catalysts in gas-phase methanol
reforming. Enhancement as a function of catalyst (Ru-3-5-SLP compared to
Ru-3-SLP) is shown at the top for both hydroxides KOH (black) and CsOH
(red). Enhancement as a function of hydroxide (CsOH compared to KOH) is
shown on the right y-axis for both Ru-3-SLP (blue) and Ru-5-SLP (green).
À 1
À 1
8
2 kJmol for complex 3 and 67 kJmol for complex 5 have
[29]
been determined. While these data do not allow a direct
comparison, the range is similar for all systems studied. More
detailed kinetic investigations would be mandatory to further
elucidate the SLP mechanism at hand.
Reaction conditions: pabs: 1 bar; pMeOH j H
2
O: 0.5 bar; mSLP: 5.7 g (CsOH); Qtot
:
À 1
À 1
5
0 mlmin ; T: 120–160°C. Catalyst composition: 50 μmolRu
g
Support
;
À 1
À 1
91 molKOH mol ; 20 wt% KOH-loading, 31 mol
Ru
CsOH
Ru
mol ; 20 wt% CsOH-
loading; Aluminum oxide support material 1.0 mm Ø.
Eur. J. Inorg. Chem. 2021, 1745–1751
www.eurjic.org
1749
© 2021 The Authors. European Journal of Inorganic Chemistry published
by Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100042
of the catalyst. KOH and CsOH were identified as most
promising basic additives and coating materials. Peak perform-
Instruments GmbH). The solid samples were digested with concen-
trated HCl:HNO :HF in a 3:1:1 ratio in volume, using microwave
3
À 1
À 1
heating up to 220°C for 40 min. (CAUTION: HF is extremely harmful,
relevant safety precautions must be taken). The instrument was
calibrated with standard solutions of the relevant elements prior to
the measurements.
ances of 41 h (SLP with KOH) and 74 h (SLP with CsOH)
could be achieved at optimized pore filling degrees. Addition-
ally, a bi-catalytic system, consisting of two distinct Ru-pincer
complexes was employed to further enhance the activity and
À 1
long-term stability of the SLP catalysts. A TOF of 91 h was
Continuous gas-phase reactor setup
achieved with this optimized bi-catalytic SLP catalyst with CsOH
coating, which is a threefold improvement of TOF compared to
our recent results at comparable reaction conditions.
The continuous vapor phase methanol steam reforming experi-
ments were conducted using a fixed bed reactor setup. A simplified
flow scheme is shown in Figure 7, for a detailed flow scheme see
Figures S1 and S2 in the Supporting Information. The liquid
substrates methanol and bi-distilled water were delivered via two
independent mass flow controllers (Liqui-Flow; Bronkhorst) into a
heating coil followed by an evaporating column in order to
minimize pulsations. No inert gas as diluting or balancing agent
was applied in the reactor.
Experimental Section
All syntheses of Ru-SLP catalysts were carried out under an inert
atmosphere of Ar (99.999 vol%, Air Liquide) using standard Schlenk
techniques. The carbonylchlorohydrido [bis(2-di-i- propylphosphi-
noethyl)amine]ruthenium(II) complex 3 was purchased from Strem
Nitrogen (99.9990 vol%, Linde Gas) was used for starting up and
shutting down the reactor in order to avoid condensation and was
Chemicals and stored in
a glovebox (Vacuum Atmosphere
Company, OMNI-LAB) prior to use. The complex 5 was prepared at
supplied and regulated by
a mass flow controller (El-Flow;
[
30]
LIKAT according to a literature procedure. It was shipped in a
sealed ampule and stored at FAU in a glovebox prior to use.
Bronkhorst). Furthermore, in order to achieve reasonable resolution
concerning measurement times, the product gas was diluted after
À 1
the reactor with a defined N stream (F , typically 200 mL min ).
2
N2
N
The gaseous compounds were dried over silica beds, quantified by
a mass flow meter (El-Flow; Bronkhorst) and analyzed by an on-line
IR- and TCD-based Gas Analyzer (X-Stream Enhanced; Emerson).
The hydrogen flow at the reactor outlet FH2 was calculated
according to Equation (1), taking into account the measured
volume flow FH2,out of the effluent gas, the correction factors for the
respective compounds KH2 and temperature Tcorr as well as the
Synthesis of supported Ru-pincer catalysts
For the impregnation of the support materials the homogeneous
Ru-complex, powdery LiOH (Acros Organics; CAS 1310-65-2), NaOH
.
(
Alfa Aesar; CAS 1310-73-2), RbOHH O (Alfa Aesar; CAS 12026-05-0),
2
.
CsOHH O (Alfa Aesar; CAS 35103-79-8) and KOH (Merck; CAS 1310-
2
5
8-3) were dissolved in methanol (LC-MS grade; VWR CAS: 67-56-1)
volume fraction xH2.
and stirred for at least an hour. Next, the aluminum oxide support
material (Sasol Germany, alumina spheres 1.0/160, 1.06 mm average
2
À 1
particle diameter, 163 m g specific surface area) was added and
dispersed in the solution for approx. 10 minutes in order to avoid
mechanical stress on the support. The solvent was then slowly
removed under reduced pressure in a rotary evaporator to facilitate
homogeneous coating of the support material. Afterwards, the
optically dry catalyst was again evacuated at <10 mbar for half an
hour and stored under argon atmosphere prior to use. Solid
substances have been dried, and liquids have been degassed prior
to use.
(1)
The maximum product flow FH2,max, based on the inlet composition,
was calculated according to Equation (2), taking into account the
stoichiometry. The mass flow of methanol m_ MeOHwas additionally
checked by taking samples from the condensate from time to time.
(2)
From these two flows, the conversion with respect to methanol was
calculated.
Characterization of the supported molecular catalysts
The content of active metals and other elements, and their ratio,
was determined by inductively coupled plasma-atom emission
spectroscopy (ICP-AES) using a Ciros CCD (Spectro Analytical
(3)
Figure 7. Simplified flow scheme of the continuous vapor phase methanol steam reforming setup.
Eur. J. Inorg. Chem. 2021, 1745–1751
www.eurjic.org
1750
© 2021 The Authors. European Journal of Inorganic Chemistry published
by Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100042
The selectivity towards CO was calculated from the gas analyzer
data directly according to Equation (4). No hints of other carbona-
ceous compounds were found in random samples.
[6] D. Teichmann, W. Arlt, P. Wasserscheid, Int. J. Hydrogen Energy 2012, 37,
8118–18132.
2
1
[
7] P. Preuster, C. Papp, P. Wasserscheid, Acc. Chem. Res. 2017, 50, 74–85.
8] P. Preuster, A. Alekseev, P. Wasserscheid, Annu. Rev. Chem. Biomol. Eng.
[
2
017, 8, 445–471.
(4)
[9] P. M. Modisha, C. N. M. Ouma, R. Garidzirai, P. Wasserscheid, D.
Bessarabov, Energ. Fuels 2019, 33, 2778–2796.
The activity of the Ru-SLP catalysts was assigned by means of turn
over frequency TOF, taking into account the known number of
moles of Ru nRu inside the SLP catalyst.
[10] Y. Okada, E. Sasaki, E. Watanabe, S. Hyodo, H. Nishijima, Int. J. Hydrogen
Energy 2006, 31, 1348–1355.
[
[
[
[
11] Y. Shen, Y. Zhan, S. Li, F. Ning, Y. Du, Y. Huang, T. He, X. Zhou, Chem. Sci.
017, 8, 7498–7504.
12] B. Loges, A. Boddien, H. Junge, M. Beller, Angew. Chem. Int. Ed. 2008, 47,
3962–3965; Angew. Chem. 2008, 120, 4026–4029.
13] K. Sordakis, C. Tang, L. K. Vogt, H. Junge, P. J. Dyson, M. Beller, G.
Laurenczy, Chem. Rev. 2018, 118, 372–433.
14] G. A. Olah, A. Goeppert, G. K. Surya Prakash, in Beyond Oil and Gas: The
Methanol Economy (Eds.: G. A. Olah, A. Goeppert, G. K. Surya Prakash),
Wiley-VCH, Weinheim, 2018.
2
(5)
The activity enhancement by either hydroxide coating or nature of
the catalyst was calculated by normalizing the TOF to the standard
one obtained for Ru-3-SLP with KOH coating at 120°C.
[
15] A. Iulianelli, P. Ribeirinha, A. Mendes, A. Basile, Renewable Sustainable
Energy Rev. 2014, 29, 355–368.
(6)
[
16] A. L. Dicks, D. A. J. Rand, Fuel Cell Systems Explained, Wiley-VCH,
Weinheim, 2018.
[
[
17] D. R. Palo, R. A. Dagle, J. D. Holladay, Chem. Rev. 2007, 107, 3992–4021.
18] S. Sá, H. Silva, L. Brandão, J. M. Sousa, A. Mendes, Appl. Catal. B 2010, 99,
Acknowledgements
43–57.
[
[
19] E. Alberico, M. Nielsen, Chem. Commun. 2015, 51, 6714–6725.
20] N. Onishi, G. Laurenczy, M. Beller, Y. Himeda, Coord. Chem. Rev. 2018,
The authors gratefully acknowledge funding from the German
Federal Ministry of Economic Affairs and Energy (BMWi) within the
project Metha-Cycle (grant number 03ET6071F). Open access
funding enabled and organized by Projekt DEAL.
373, 317–332.
[21] J. Kothandaraman, S. Kar, A. Goeppert, R. Sen, G. K. Surya Prakash, Top.
Catal. 2018, 61, 542–559.
[
22] P. Hu, Y. Diskin-Posner, Y. Ben-David, D. Milstein, ACS Catal. 2014, 4,
2
649–2652.
[
[
23] A. Mukherjee, D. Milstein, ACS Catal. 2018, 8, 11435–11469.
24] a) M. Nielsen, E. Alberico, W. Baumann, H.-J. Drexler, H. Junge, S.
Gladiali, M. Beller, Nature 2013, 495, 85–89; b) E. Alberico, A. J. J. Lennox,
L. K. Vogt, H. Jiao, W. Baumann, H.-J. Drexler, M. Nielsen, A. Spannen-
berg, M. P. Chercinski, H. Junge, M. Beller, J. Am. Chem. Soc. 2016, 138,
Conflict of Interest
The authors declare no conflict of interest.
14890–14904.
[
25] V. Strobel, J. J. Schuster, A. S. Braeuer, L. K. Vogt, H. Junge, M. Haumann,
React. Chem. Eng. 2017, 2, 390–396.
Keywords: Methanol reforming · Ru-pincer catalyst · Supported
liquid phase · Immobilization · Alkali hydroxide
[
[
26] A. Agapova, H. Junge, M. Beller Chem. Eur. J. 2019, 25, 9345–9349.
27] C. H. Schwarz, A. Agapova, H. Junge, M. Haumann, Catal. Today 2020,
3
42, 178–186.
[
[
28] http://www.periodensystem-online.de/ (accessed December 2020).
29] A. Agapova, E. Alberico, A. Kammer, H. Junge, M. Beller, ChemCatChem
[
1] G. A. Olah, G. K. Prakash, A. Goeppert, J. Am. Chem. Soc. 2011, 133,
2881–12898.
2] P. Makowski, A. Thomas, P. Kuhn, F. Goettmann, Energ. Environm. Sci.
009, 2, 480–490.
2
019, 11, 1910–1914.
1
[
30] Supporting Information of.[24b].
[
2
[
[
3] K. Zeng, D. Zhang, Progr. Energ. Comb. Sci. 2010, 36, 307–326.
4] M. Carmo, D. L. Fritz, J. Mergel, D. Stolten, Int. J. Hydrogen Energy 2013,
38, 4901–4934.
Manuscript received: January 18, 2021
[5] K. Müller, W. Arlt, Energ. Tech. 2013, 1, 501–511.
Revised manuscript received: March 8, 2021
Eur. J. Inorg. Chem. 2021, 1745–1751
www.eurjic.org
1751
© 2021 The Authors. European Journal of Inorganic Chemistry published
by Wiley-VCH GmbH