DOI: 10.1002/cssc.

201600435 Concepts

Carbon Dioxide-Free Hydrogen Production with Integrated

Hydrogen Separation and Storage
Stefan Drr,[a] Michael Mller,[a] Holger Jorschick,[b] Marta Helmin,[c] Andreas Bçsmann,[a]
Regina Palkovits,*[c] and Peter Wasserscheid*[a, b]

An integration of CO2-free hydrogen generation through meth- solid carbon and a hydrogen/methane mixture that can be di-
ane decomposition coupled with hydrogen/methane separa- rectly fed to a hydrogenation unit to load a LOHC with hydro-
tion and chemical hydrogen storage through liquid organic hy- gen. This allows for a straight-forward separation of hydrogen
drogen carrier (LOHC) systems is demonstrated. A potential, from CH4 and conversion of hydrogen to a hydrogen-rich
very interesting application is the upgrading of stranded gas, LOHC material. Both, the hydrogen-rich LOHC material and the
for example, gas from a remote gas field or associated gas generated carbon on metal can easily be transported to desti-
from off-shore oil drilling. Stranded gas can be effectively con- nations of further industrial use by established transport sys-
verted in a catalytic process by methane decomposition into tems, like ships or trucks.

Introduction

Today, more than 60 million tons of hydrogen are produced to the stoichiometry of the methane reforming and water-gas
and utilized industrially.[1] The largest part of this quantity is ap- shift reactions, the energy consumption of the process, and
plied in the petrochemical and chemical industry, mainly for the molecular masses of the molecules involved.[5]
desulphurization of fuels and hydrogenation reactions, for ex- Aiming for the sustainable (zero emission) production of hy-
ample, the production of ammonia and methanol.[2] In the drogen, electrolysis of water using renewables is one possible
near future, hydrogen is expected to gain additional impor- option. However, electrolysis is still relatively expensive (3 E
tance as a green energy vector for the mobility sector, because per kg of hydrogen).[6] As an alternative, hydrogen production
its combustion in fuel cells or combustion engines produces from hydrocarbons avoiding CO2 emissions can be considered.
water as the only combustion product.[3, 4] Whereas CO2 capture and storage (CCS technology) is energy
However, to consider hydrogen as a sustainable fuel, it is im- and cost intensive,[7] and associated with ecological uncertain-
portant to assure that its production is free of CO2 emissions. ties, hydrogen production through catalytic decomposition of
This is not a trivial task given that 95 % of today’s industrial hy- hydrocarbons linked to elemental carbon sequestration is
drogen production is generated from fossil fuels by steam re- a promising option. In particular, the moderately endothermic
forming, partial oxidation, or gasification processes. All of catalytic methane decomposition to carbon and hydrogen
these produce significant amounts of CO2. For example, 1 ton holds great promise for the sustainable production of hydro-
of hydrogen produced by using methane as raw material was gen.[8]
reported to produce more than 10 tons of CO2 simply owing
CH4 ! C þ 2 H2 DH25  C ¼ 74:5 kJ mol1 ð1Þ
[a] S. Drr, M. Mller, A. Bçsmann, Prof. P. Wasserscheid
Lehrstuhl fr Chemische Reaktionstechnik Whereas purely thermal decomposition of methane occurs
Friedrich-Alexander-Universitt Erlangen-Nrnberg at process temperatures above 1200 8C,[9] the use of an appro-
Egerlandstr. 3, 91058 Erlangen (Germany)
priate catalyst has been reported to bring the operating tem-
[b] H. Jorschick, Prof. P. Wasserscheid
Helmholtz-Institut Erlangen-Nrnberg fr Erneuerbare Energien (IEK-11) perature down to 550 8C.[10] Regarding suitable catalysts for
Forschungszentrum Jlich GmbH catalytic methane decomposition (CMD), the literature pays
Egerlandstraße 3, 91058 Erlangen (Germany) special attention to supported group 8–10 base metal cata-
E-mail: [email protected]
lysts, in particular nickel, iron, and cobalt.[11] While nickel cata-
[c] M. Helmin, Prof. R. Palkovits
lysts are limited in CMD to operating temperatures below
Lehrstuhl fr Heterogene Katalyse und Technische Chemie
RWTH Aachen and JARA Energy 700 8C, iron systems have been reported to operate even
Worringerweg 2, 52074 Aachen (Germany) above this temperature.[11] Despite these differences, any cata-
E-mail: [email protected] lyst used for CMD eventually deactivates as result of deposition
Supporting Information and the ORCID identification number(s) for the of carbon-rich polyaromatic structures, amorphous, filamen-
author(s) of this article can be found under http://dx.doi.org/10.1002/
tous, or graphitic carbon on its surface.
cssc.201600435.
CMD is a complex process that can be divided into three
This publication is part of a Special Issue celebrating “10 Years of Chem-
SusChem”. A link to the issue’s Table of Contents will appear here once key steps: (i) hydrogen generation by a sequence of methane
it is complete. adsorption and dissociation at the metal surface and simulta-

ChemSusChem 2016, 9, 1 – 7 1  2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim &

These are not the final page numbers! ÞÞ

 Concepts
neous desorption of hydrogen; (ii) formation of carbon nano- by an extremely low volumetric energy density (only 3 Wh L1
materials by dissolution of chemisorbed carbon atoms into the at ambient pressure), and its transport over long distances is
metal followed by diffusion of carbon through the bulk of the unsuitable in elemental form. Even the transport of com-
metal where carbon atoms precipitate onto the support form- pressed gaseous hydrogen (CGH2, typically up to 700 bar
ing the nanostructure; and (iii) catalyst deactivation caused by  79 MPa pressure) or liquefied cryogenic hydrogen (LH2, typi-
slower and slower carbon diffusion through the metal particle cally at 253 8C), would not solve the problem of an efficient
and formation of graphitic carbon at the particle surface thus transport of elemental hydrogen, as the volumetric storage
reducing the active surface area.[12] Figure 1 illustrates these density is still modest (see Table 1). Moreover, hydrogen com-
three steps. pression and cooling come along with additional energy con-
sumption and require special infrastructure, such as compres-
sor stations, compressed gas tankers, or transport vehicles with
cryogenic tank systems.

Table 1. Volumetric and gravimetric energy density of elemental hydro-

gen compared to fossil fuels.[16]

Energy source Energy carrier Energy density Energy density

[kWh L1] [kWh kg1]
Fossil fuels Gasoline 8.6 11.4
Diesel 9.8 11.8
Heavy oil 10.7 11.0
Natural gas 2.4 11.1
Hydrogen H2 (atm) 0.003 33.0
GCH2 (700 bar) 1.3 33.0
LH2 (liquid) 2.4 33.0
Figure 1. Representation of hydrogen production, carbon nanomaterial LOHC (H18-DBT) 1.87 2.05
(CNM) formation, and growth in the CMD process.

A significant part of the world’s methane reserves is located In this contribution, we propose the combination of CMD
far from industrial and population centers, and transportation with hydrogen storage in liquid organic hydrogen carrier
cost hinders its efficient utilization. Moreover, many of these (LOHC) systems as a very attractive option to overcome the
gas fields are far too small to justify chemical conversion of named limitations and to establish a technology to upgrade
methane by current gas-to-liquid technologies that gain eco- stranded methane in the form of storable and transportable
nomic attractiveness only in the form of world-scale units.[13] hydrogen equivalents ready for industrial use at any far distant
Thus, the term “stranded gas” refers to methane resources location. LOHC systems are composed of pairs of hydrogen-
at places and in quantities that do not justify the installation of lean, mostly aromatic compounds and hydrogen-rich, mostly
pipeline infrastructures, for example, associated gas from an alicyclic compounds. By reversible hydrogenation and subse-
off-shore oil drilling rig or from oil drilling at remote locations quent dehydrogenation, LOHCs can be used to safely bind,
in the desert. Currently, such stranded gas is typically flared, store, transport, and release hydrogen on demand.[17–20] De-
vented or re-injected.[14] Consequently, new technologies able pending on the LOHC system, hydrogen storage is possible at
to convert this gas in small facilities of low investment cost di- a material-based energy density of up to 72 gH2 LLOHC or
rectly at the gas source would benefit from a very low or even 2.4 kWh L1. Thus, the concept allows hydrogen storage over
negative feedstock price. It is common understanding among long times at high volumetric energy densities as well as hy-
experts that the amount of stranded gas is on the order 40 to drogen transport over long distances without losses using ex-
60 % of the proven global natural gas resources.[15] isting infrastructure for liquid fuels (e.g., ships, trucks, rail
CMD is, in principle, ideally suited for a treatment of strand- trucks, and tank farms). Figure 2 shows the proposed combina-
ed methane. Compared to common gas-to-liquid technology, tion of CMD and hydrogen storage through LOHC systems.
CMD represents a relatively simple one-step process that con- Herein, we demonstrate new insight into the combination of
verts methane directly into two valuable products, hydrogen CMD and LOHC hydrogenation by using representative hydro-
and carbon-loaded metal. However, two aspects prevent the gen-methane mixtures from CMD as substrate for charging
use of CMD as economic option in stranded gas upgrading: a selected, suitable LOHC system.
(i) for thermodynamic reasons, the CMD process provides no
full methane conversion in the desired temperature range of
Hydrogen production through CMD
its catalytic operation, and hydrogen–methane separation is re-
quired to recycle unconverted methane back into the CMD The influence of the reaction temperature on hydrogen yield
process and to obtain pure hydrogen; and (ii) whereas carbon- in CMD was studied using a 20 wt % Ni/Al2O3 catalyst in the
loaded metal can be easily shipped from the remote place to temperature range of 550 to 700 8C. Figure 3 shows the outlet
a potential consumer, the produced hydrogen is characterized hydrogen stream as well as methane conversion as a function

& ChemSusChem 2016, 9, 1 – 7 www.chemsuschem.org 2  2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ÝÝ These are not the final page numbers!

 Concepts

Figure 2. The proposed two-step upgrading process for stranded gas: meth-
ane is converted in a catalytic decomposition process to carbon and hydro-
gen, which is subsequent purified from remaining methane and transferred
into a transportable form by LOHC hydrogenation.

Figure 4. TEM micrographs of the spent Ni/Al2O3 catalyst at (a, b) 600 8C and
at (c, d) 700 8C.

ical differences of the deposited carbon as a function of differ-

ent reaction temperatures. Figure 4 a–d show some representa-
tive TEM micrographs of the spent catalyst at 600 (Figure 4 a, b)
and 700 8C (Figure 4 c, d). The TEM measurements reveal the
presence of carbon material on the catalyst surface, which is
mainly deposited in the form of nanostructures with varying
morphology depending on the reaction temperature. Compar-
ing the nanostructures formed at 600 8C with those formed at
700 8C, the material obtained at 600 8C consists mainly of
hollow cylindrical filaments emerging from the nickel particles
(Figure 4 a) coexisting with filaments containing a metal parti-
Figure 3. Effect of reaction temperature on the evolution of hydrogen from cle on their tip, which is uniformly coated with parallel planes
methane by CMD at a gas hourly space velocity (GHSV) of 12 500 h1; left of graphene (Figure 4 b). An examination of the TEM measure-
axis (black symbols) shows the generated volumetric hydrogen flow; right ments evidences that long filaments with no particle on their
axis (red symbols) shows the degree of methane conversion.
tip are more abundant than filaments with uniform coatings.
In contrast, carbon material formed at a reaction temperature
of reaction time. The nickel catalyst maintains its initial activity of 700 8C shows a less homogenous morphology distribution.
at a reaction temperature of 550 8C for the reaction time of 5 h Nanotubes with coated metal particles on their tip (Figure 4 c),
producing a constant hydrogen outlet stream of uniformly coated nickel particles referred to as nano-onions, as
54 mLN min1 gcat1 (where LN refers to the volume at normal well as bamboo shaped filaments (Figure 4 d) were obtained
conditions: 1 bar, room temperature) corresponding to a meth- during this reaction. The results suggest, that high tempera-
ane conversion of 14 %. A slight increase in reaction tempera- tures cause fast catalyst deactivation owing to the immediate
ture of 50 8C resulted in a hydrogen outlet stream of formation of a graphitic layer encapsulating the catalytically
92 mLN min1 gcat1. However, the catalyst experienced slight active metal particle. In contrast, the catalyst maintains its ac-
deactivation at 600 8C during the reaction time of 5 h; there- tivity owing to the generation of nanotubes at lower reaction
fore, a loss in methane conversion of around 5 % was observed temperatures.
during this time. Further increase in reaction temperature re-
sulted in an increase of the initial hydrogen yield to
Hydrogen separation and storage through LOHC hydroge-
122 mLN min1 gcat1 at 700 8C. However, at this temperature
nation
the evolving hydrogen content dropped quickly until complete
catalyst deactivation after approximately 1.5 h on stream. The LOHC hydrogen storage technology is based on reversible
The observed fast deactivation behavior of Ni/Al2O3 at hydrogenation of double bonds. In the past, various com-
700 8C can be attributed to the formation of different carbon pounds have been investigated for their suitability as LOHC
species during methane decomposition. TEM characterization systems ranging from small aromatics, for example, benzene/
of the spent catalyst gave us more insight into the morpholog- cyclohexane[21] and toluene/methylcyclohexane,[22–26] to con-

ChemSusChem 2016, 9, 1 – 7 www.chemsuschem.org 3  2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim &

These are not the final page numbers! ÞÞ

 Concepts
Rt Rt
densated aromatics (e.g., naphtaline/decalin[27, 28]) and t0 nH2 ;reaction ðt Þ dt t0 nH2 ;reaction ðt Þ dt
N-containing aromatics [e.g., N-ethyl-carbazole(NEC)/dedecahy- DoH ðtÞ ¼ ¼ ð2Þ
nH2 ;max 9  nH0DBT
dro-N-ethyl-carbazole[29–33]]. All the named systems have their
advantages and disadvantages regarding hydrogen capacity, In Equation (2), nH2 ,reaction (t) represents the amount of hydro-
hydrogenation enthalpy, thermal stability liquid range, toxicity, gen consumed at time t as measured by mass-flow-meter.
and technical availability. nH0DBT was calculated from mH0DBT, the mass of H0-DBT pro-
An excellent tradeoff of all these highly relevant properties vided in the reactor.
for suitable LOHC materials is found for the relatively recently In the here reported batch mode experiments, hydrogena-
reported system dibenzyltoluene (H0-DBT)/perhydrodibenzyl- tion is characterized by a constant reaction rate at the begin-
toluene (H18-DBT).[34–36] Dibenzyltoluene is a commercial heat ning of the reaction, that is, for low to medium degrees of hy-
transfer oil (marketed under the tradenames Marlotherm SH drogenation. The gradient of DoH versus time is almost con-
or Farolin WF0801; price is below 4 E kg1 on ton-scale) with stant until a DoH of 0.6. This observation is further confirmed
known and very favorable toxicology (non-toxic, non-mutagen- by the constant hydrogen consumption, as shown in Figure 6.
ic, non-carcinogenic, not labelled as hazardous good in trans- This zero-order-like behavior hints for the fact that transport of
portation), high thermal stability, and excellent physico-chemi- hydrogen gas into the liquid phase limits the observed macro-
cal properties for many LOHC applications. The isomeric mix- kinetics during the period of maximum hydrogen consumption
ture of dibenzyltoluenes has a melting point of 34 8C and at the beginning of the reaction. Further hydrogenation
a boiling point of 390 8C.[36] During hydrogenation, H0-DBT beyond the DoH of 0.6 is significantly slower owing to the fall-
binds 6.2 wt % hydrogen corresponding to an energy content ing concentration of hydrogenable LOHC left, but full conver-
of 2.05 kWh kg1. Figure 5 shows a depiction of the H0-DBT/ sion is always reached. The time to full conversion is a function
H18-DBT hydrogen storage system. of hydrogen pressure, reaction temperature, stirrer speed, and
the molar catalyst to LOHC ratio.
The effect of hydrogen pressure on the rate of H0-DBT hy-
drogenation is shown in Figure 6 for experiments conducted
at 180 8C and a molar catalyst to LOHC ratio of 1:400, corre-
sponding to 0.25 mol % of the applied, commercial Ru/Al2O3
(5 wt %ruthenium) catalyst. It is apparent that a higher hydro-
gen pressure leads to faster hydrogenation and complete for-
mation of H18-DBT is achieved in a shorter time. Full conver-
sion is reached after 100 min (50 bar  5 MPa), 140 min (40 bar
 4 MPa) or about 3 h at 30 bar hydrogen pressure. A reduc-
tion of hydrogen pressure below 30 bar leads to a massive in-
Figure 5. Storage and transport of hydrogen by reversible catalytic hydroge- crease in reaction time for full DoH under otherwise identical
nation/dehydrogenation of dibenzyltoluene/perhydro-dibenzyltoluene.
reaction conditions.
As can be seen in Figure 7, a linear relation between pres-
As shown by the recent studies of Brckner et al.,[34] com- sure and reaction rate is found for hydrogen pressures varying
plete hydrogen release from H18-DBT is possible by a catalytic, from 5 to 50 bar (0.5–5 MPa). Further investigations of hydro-
endothermic dehydrogenation reaction, for example, in the genation reactions applying hydrogen pressures of less than
presence of Pt on alumina catalysts. These studies could also 5 bar  0.5 MPa not only show poor reaction rates but also
show that dehydrogenation of H18-DBT proceeds under ap- seem to be affected by equilibrium limitations.
propriate conditions (i.e., 310 8C, 1 bar  0.1 MPa) with negligi-
ble fragmentation of the hydrogen carrier molecule. This ena-
bles the utilization of the H0-DBT/H18-DBT LOHC system in re-
peated hydrogenation-dehydrogenation cycles. Due to these
very favorable properties we will restrict ourselves in this publi-
cation to the catalytic hydrogenation of H0-DBT as most inter-
esting way to purify and store hydrogen from the CMD pro-
cess.
The hydrogenation of H0-DBT comprises the consecutive hy-
drogenation of three isolated phenyl rings. For a quantitative
comparison, the degree of hydrogenation (DoH) is defined as
the ratio of the bound hydrogen and the maximum hydrogen
up-take capacity. For the experiments reported here dealing
with the hydrogenation of H0-DBT with hydrogen/methane
mixtures using a batch reactor, the DoH has been determined
by mass flow measurements according to Equation (2) and in- Figure 6. Hydrogenation of H0-DBT at different hydrogen pressures;
dependently by 1H NMR spectroscopy.[37] T = 180 8C, 5.9 mol DBT, 0.25 mol % Ru applied as 5 wt % Ru on Al2O3.

& ChemSusChem 2016, 9, 1 – 7 www.chemsuschem.org 4  2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ÝÝ These are not the final page numbers!

 Concepts
from 700 rpm (standard reaction conditions) to 900 or
1100 rpm, led to a linear increase of hydrogen consumption at
the beginning of the hydrogenation (see the Supporting Infor-
mation for details). This confirms the hypothesis that hydrogen
gas mass transport into the liquid phase limits the observed
reaction rate under the applied conditions. Moreover, it indi-
cates that the increased rate in presence of methane gas is re-
lated to a faster mass transfer of hydrogen into the liquid
phase very probably owing to a lower viscosity of the meth-
ane-saturated LOHC mixture compared to the methane-free
system.

Conclusions
Figure 7. Effect of hydrogen pressure on hydrogenation of H0-DBT; Our contribution highlights the fact that a combination of cat-
T = 180 8C, 5.9 mol DBT, 0.25 mol % Ru applied as 5 wt % Ru on Al2O3, 3 h re- alytic methane decomposition (CMD) and hydrogen purifica-
action time. tion and storage by hydrogenation of LOHC systems offers
a versatile, economically attractive, and new way for upgrading
To evaluate whether H0-DBT hydrogenation is a suitable stranded gas towards green, CO2-free hydrogen production.
method to purify hydrogen from hydrogen/methane mixtures Whereas the CMD step has been shown to operate in
out of the CMD process, our next set of experiments was dedi- steady-state for at least 5 h converting 14 % of the offered
cated to batch reactor hydrogenation studies using hydrogen/ methane into hydrogen and carbon, the LOHC charging step
methane mixtures. As shown in Figure 8, no negative impact operates beneficially with diluted hydrogen/methane streams
of methane on the reaction rate or H0-DBT conversion was ob- to hydrogenate dibenzyltoluene to perhydro-dibenzyltoluene.
served for comparative experiments in which 20 bar  2 MPa of In the scenario of a stranded gas resource, for example, at
methane were added to 40 bar  2 MPa of hydrogen. During an off-shore oil platform or at a remote drilling site in the
these runs (pH2 = 40 bar, pCH4 = 20 bar) even a remarkable in- desert, methane would be converted in the proposed technol-
crease of the reaction rate was observed at low degrees of H0- ogy sequence into carbon rich metal (for metallurgic process-
DBT conversion. es) and transportable hydrogen in the form of a hydrogen-rich
Figure 8 clearly illustrates increased reaction rates for both organic liquid.
experiments applying a mixed gas phase hydrogen substrate. It is fair to note, though, that nearly all studies on CMD to
Compared to the runs with pure hydrogen, higher degrees of date have been performed at ambient pressure as an increase
hydrogenation were achieved after given reaction times. Fur- in total pressure is expected to have a detrimental effect on
ther experiments featuring different methane partial pressures the reaction rate owing to thermodynamic effects.[38] Actually,
(see the Supporting Information) did not reveal any further in- CMD conducted under pressure is a largely uninvestigated
fluence on reaction rate or reaction time (until full hydrogena- field of research in the area of methane decomposition. There-
tion was reached). In addition, the influence of stirrer speed fore, it is expected that some compression of the methane/hy-
was examined using the same hydrogen/methane mixture drogen mixture from CMD is required to provide the pressure
(40 bar hydrogen/20 bar methane). Increase in stirrer speed for the hydrogenation step. However, the energy input into
this compression step will depend on the compression ratio
and with that the starting pressure delivered by the CMD step
is important. This turns studies of CMD under even slight pres-
sures into a highly relevant field for further investigations.

Experimental Section
Catalytic methane decomposition
The 20 wt % Ni/Al2O3 catalyst used in this work was prepared by in-
cipient wetness impregnation of g-Al2O3 (Sasol Germany) with an
aqueous solution of Ni(NO3)2·6H2O (Sigma–Aldrich). Prior to im-
pregnation the g-Al2O3 spheres were ground and sieved to a parti-
cle size of 500–630 mm. The impregnated sample was dried at
100 8C for 24 h and subsequently calcined at 500 8C for 4 h. The
Figure 8. Comparison of hydrogenation performance by using pure hydro- metal loading was verified by inductively coupled plasma atomic
gen and a hydrogen/methane mixture; T = 180 8C, 5.9 mol DBT, 0.25 mol % emission spectroscopy (ICP-AES). Prior to reaction the catalyst was
Ru applied as 5 wt % Ru on Al2O3. reduced in a 50 mL min1 hydrogen (Air products) stream for 2 h.

ChemSusChem 2016, 9, 1 – 7 www.chemsuschem.org 5  2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim &

These are not the final page numbers! ÞÞ

 Concepts
The methane decomposition experiments were carried out at at- [5] J. Dufour, D. Serrano, J. Galvez, J. Moreno, C. Garcia, Int. J. Hydrogen
mospheric pressure in a fixed-bed quartz reactor (internal diame- Energy 2009, 34, 1370 – 1376.
ter: 11 mm, length: 700 mm) at temperatures ranging from 550 to [6] A. Tremel, P. Wasserscheid, M. Baldauf, T. Hammer, Int. J. Hydrogen
Energy 2015, 40, 11457 – 11464.
700 8C. 0.44 g of catalyst, corresponding to a bed height of 5 mm
[7] B. Metz, O. Davidson, H. de Coninck, M. Loos, L. Meyer, Carbon Dioxide
was used for each run. The feed consisted of pure methane
Capture and Storage, International Panel On Climate Change, Cam-
(grade 3.5 corresponding to a purity of > 99.95%, Westfalen AG) bridge University Press, 2009.
adjusted to a gas hourly space velocity (GHSV) of 12 500 h1. The [8] J. I. Villacampa, C. Royo, E. Romeo, J. A. Montoya, P. Del Angel, A.
composition of the outlet stream was analyzed by an on-line gas Monzn, Appl. Catal. 2003, 252, 363 – 383.
chromatograph (Bruker 456-GC) equipped with two thermal con- [9] H. F. Abbas, W. M. A. W. Daud, Int. J. Hydrogen Energy 2010, 35, 1160 –
ductivity detectors. No methane decomposition products other 1190.
than hydrogen were detected in the effluent gas during the experi- [10] M. A. Ermakova, D. Y. Ermakov, G. G. Kuvshinov, Appl. Catal. 2000, 201,
ments. TEM studies were performed using a JEOL JEM-2200FS high 61 – 70.
[11] Y. Li, D. Li, G. Wang, Catal. Today 2011, 162, 1 – 48.
resolution, aberration corrected transmission electron microscope.
[12] G. B. Nuernberg, H. V. Fajardo, D. Z. Mezalira, T. J. Casarin, L. V. Carren,
L. F. D. Probst, Kinet. Catal. 2008, 87, 1698 – 1704.
[13] K. Ding, H. Metiu, G. D. Stucky, ACS Catal. 2013, 3, 474 – 477.
LOHC hydrogenation [14] R. Khalilpour, I. A. Karimi, Energy 2012, 40, 317 – 328.
[15] D. Coyle, C. Durr, P. Shah in SPE Annu. Tech. Conf. Exhib. 5 – 8 October,
The hydrogenation experiments were performed by using a 5.3 L Denver, Color., Society Of Petroleum Engineers, 2003.
stainless-steel batch autoclave applied with a gas inducing stirrer [16] D. Stolten, B. Emonts, Hydrogen Science and Engineering: Materials, Pro-
(Bchi Glas). All experiments were performed with a 5 wt % Ru/ cesses, Systems and Technology, Wiley-VCH, 2016.
Al2O3 hydrogenation catalyst supplied by Alfa Aesar. The molar cat- [17] D. Teichmann, W. Arlt, P. Wasserscheid, Int. J. Hydrogen Energy 2012, 37,
alyst to LOHC ratio was constant at 1:400 (0.25 mol %). The pres- 18118 – 18132.
sure vessel was filled with catalyst and approximately 5.9 mol fully [18] M. Ichikawa in Solid-State Hydrogen Storage, 2008.
dehydrogenated DBT (Sasol Germany). Inert atmosphere was en- [19] J. Graetz, D. Wolstenholme, G. Pez, L. Klebanoff, S. McGrady, A. Cooper
in Hydrog. Storage Technol., Taylor & Francis, 2012, pp. 239 – 328.
sured by repeatedly purging the reactor with nitrogen (grade 5.0
[20] D. Teichmann, W. Arlt, P. Wasserscheid, R. Freymann, Energy Environ. Sci.
corresponding to a purity of > 99.999 %, Linde AG). The reactor
2011, 4, 2767.
was heated to reaction temperature using a thermostat (Julabo [21] G. Cacciola, N. Giordano, G. Restuccia, Int. J. Hydrogen Energy 1984, 9,
HT60 m2) and then pressurized with hydrogen (grade 5.0 corre- 411 – 419.
sponding to a purity of > 99.999 %, Linde AG) in a range of 1 to [22] A. U. Pradhan, A. Shukla, J. V. Pande, S. Karmarkar, R. B. Biniwale, Int. J.
55 bar (0.1–5.5 MPa). In case of the hydrogenation experiments ap- Hydrogen Energy 2011, 36, 680 – 688.
plying a hydrogen/methane mixture, the pressure vessel was pres- [23] Y. Okada, E. Sasaki, E. Watanabe, S. Hyodo, H. Nishijima, Int. J. Hydrogen
surized with a defined amount of methane (grade 2.5 correspond- Energy 2006, 31, 1348 – 1356.
ing to a purity of > 99.5 %, Linde AG) resulting in a pressure in the [24] M. Taube, D. Rippin, D. Cresswell, W. Knecht, Int. J. Hydrogen Energy
1983, 8, 213 – 225.
range of 5 to 55 bar (0.5–5.5 MPa). Subsequently hydrogen was in-
[25] F. Alhumaidan, D. Cresswell, A. Garforth, Ind. Eng. Chem. Res. 2011, 50,
duced to achieve a total pressure of 60 bar (6 MPa). During the ex-
2509 – 2522.
periment, pressure was kept constant by continuous dosing of hy- [26] M. R. Usman, Energy Sources Part A 2011, 33, 2231 – 2238.
drogen while measuring the hydrogen flow by use of a Bronkhorst [27] S. Hodoshima, S. Takaiwa, A. Shono, K. Satoh, Y. Saito, Appl. Catal. A
Mass Flow Meter (F-111AC). 2005, 283, 235 – 242.
Samples of the liquid phase were withdrawn periodically during [28] D. Sebastin, E. G. Bordej, L. Calvillo, M. J. Lzaro, R. Moliner, Int. J. Hy-
the reaction and analyzed by using a JNM-ECX 400 NMR spectrom- drogen Energy 2008, 33, 1329 – 1334.
eter (Jeol Ltd.). [29] K. M. Eblagon, D. Rentsch, O. Friedrichs, A. Remhof, A. Zuettel, A. J.
Ramirez-Cuesta, S. C. Tsang, Int. J. Hydrogen Energy 2010, 35, 11609 –
11621.
[30] F. Sotoodeh, B. J. M. Huber, K. J. Smith, Int. J. Hydrogen Energy 2012, 37,
Acknowledgements 2715 – 2722.
[31] H. Cheng, A. C. Cooper, G. P. Pez, A. R. Scott, Wasserstoffspeicherung Mit-
The authors thank JARA Energy for supporting the presented tels Reversibler Hydrogenierung Pi-Konjugierter Substrate, 2004,
joint study. M.M. and P.W. acknowledge additional support from EP1475349A2.
[32] M. Yang, Y. Dong, S. Fei, H. Ke, H. Cheng, Int. J. Hydrogen Energy 2014,
the Bavarian Hydrogen Center and the Energie Campus Nrn- 39, 18976 – 18983.
berg. R.P. and M.H. acknowledge support by the project house [33] W. Peters, M. Eypasch, T. Frank, J. Schwerdtfeger, C. Korner, A. Bosmann,
Power2Fuel of RWTH Aachen University funded by the Excellence P. Wasserscheid, Energy Environ. Sci. 2015, 8, 641 – 649.
Initiative of the German federal and state government to pro- [34] N. Brckner, K. Obesser, A. Bçsmann, D. Teichmann, W. Arlt, J. Dungs, P.
Wasserscheid, ChemSusChem 2014, 7, 229 – 235.
mote science and research at German universities. [35] K. Mller, K. Stark, V. N. Emel’yanenko, M. a. Varfolomeev, D. H. Zaitsau,
E. Shoifet, C. Schick, S. P. Verevkin, W. Arlt, Ind. Eng. Chem. Res. 2015, 54,
7967 – 7976.
Keywords: hydrogen storage · hydrogenation · liquid organic [36] Sasol Germany, EC-Safety Data Sheet: Marlotherm SH, 2013.
hydrogen carriers · methane decomposition · upgrading [37] G. Do, P. Preuster, R. Aslam, A. Bosmann, K. Muller, W. Arlt, P. Wassersc-
heid, React. Chem. Eng. 2016, 1, 313 – 320.
[1] J. D. Holladay, J. Hu, D. L. King, Y. Wang, Catal. Today 2009, 139, 244 – [38] N. Muradov, Z. Chen, F. Smith, Int. J. Hydrogen Energy 2005, 30, 1149 –
260. 1158.
[2] N. Armaroli, V. Balzani, ChemSusChem 2011, 4, 21 – 36.
[3] B. G. Pollet, I. Staffell, J. L. Shang, Electrochim. Acta 2012, 84, 235 – 249. Received: April 2, 2016
[4] S. Singh, S. Jain, V. PS, A. K. Tiwari, M. R. Nouni, J. K. Pandey, S. Goel, Re- Revised: May 10, 2016
newable Sustainable Energy Rev. 2015, 51, 623 – 633. Published online on && &&, 0000

& ChemSusChem 2016, 9, 1 – 7 www.chemsuschem.org 6  2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ÝÝ These are not the final page numbers!

CONCEPTS
S. Drr, M. Mller, H. Jorschick,
M. Helmin, A. Bçsmann, R. Palkovits,*
P. Wasserscheid*
&& – &&

Carbon Dioxide-Free Hydrogen

Production with Integrated Hydrogen
LOHC it in: Combining catalytic meth- economically attractive way for upgrad- Separation and Storage
ane decomposition with hydrogen pu- ing stranded gas towards green, CO2-
rification and storage by hydrogenation free hydrogen production.
of LOHC systems offers a versatile and

ChemSusChem 2016, 9, 1 – 7 www.chemsuschem.org 7  2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim &

These are not the final page numbers! ÞÞ