Bioenergy systems for the future 10 formation of hydrogen rich gas via conversion of lignocellulosic biomass and its decomposition products

Formation of hydrogen-rich gas
via conversion of lignocellulosic
biomass and its decomposition
products

10

J. Grams, A.M. Ruppert
Lodz University of Technology, Lodz, Poland

10.1

Introduction

In recent years, the increased interest in the use of renewable energy sources has been
observed. It is related to the rapid depletion of fossil fuels and growing energy
demand. One of the most promising alternatives for traditional energy resources is
lignocellulosic biomass. The advantages of such a feedstock are usually associated
with global availability, relatively low price, and limited influence on the increase
in the greenhouse effect.
On the other hand, hydrogen is considered one of the most environmentally
friendly energy carriers. Moreover, it can be used in a number of chemical processes,
including the production and valorization of platform molecules originating from biomass. Unfortunately, currently, its main production methods are steam reforming of
natural gas or coal gasification. It is widely known that such processes require the use
of traditional carbon resources and due to that affect considerably the quality of the
environment.
Taking that into account researchers began the studies focused on the development
of methods that allow for direct production of hydrogen from lignocellulosic feedstock. The literature data shows that it can be formed by high temperature treatment
or in milder conditions, that is, formic acid decomposition where arising H2 can be
used as a reducing agent in different industrial processes related to the valorization
of intermediates formed in lignocellulosic processing. In the further part of this chapter, both mentioned directions will be discussed in more detail.

10.2

High-temperature conversion of lignocellulosic
biomass towards hydrogen-rich gas

The production of hydrogen by high-temperature methods can be attractive from both
economic and environmental points of view (Ni et al., 2006). However, the efficient
conversion of lignocellulosic biomass is not an easy task due to the problems with
obtaining high yield and selectivity of occurred reactions. The literature data
Bioenergy Systems for the Future. />© 2017 Elsevier Ltd. All rights reserved.


346

Bioenergy Systems for the Future

demonstrate that pyrolysis and gasification are the most popular methods of the hightemperature processing of lignocellulosic feedstock (Bridgwater, 2012; Saxena et al.,
2008; Balat et al., 2009). In this case, the feedstock is heated in the presence of inert
gas or gasifying agent (i.e., oxygen or steam), respectively. It leads to the production
of permanent gases, bio-oil, and carbonaceous residue as demonstrated in Eq. (10.1):
Lignocellulosic biomass ! H2 + CO + CO2 + CH4 + Cn Hm + bio
À oil + tars + char

(10.1)

However, it should be noted that thermal conversion of lignocellulosic biomass is a
very complex process that proceeds in several steps (Wang et al., 2014). The first one
consists of the decomposition of the feedstock in the temperature range of 300–500°C
that results in the formation of thermally produced oxygenates. In the further part of

the process, the oxygenates are submitted to dehydration and cracking reactions.
Moreover, decarboxylation, decarbonylation, and oligomerization processes can
proceed (Ruddy et al., 2014). They lead to the arising of liquid fraction containing
water, hydrocarbons, and their derivatives (i.e., carboxylic acids, ketones, aldehydes,
alcohols, esters, ethers, sugars, among others).
In the case of the formation of gaseous products, steam reforming, dry reforming,
water-gas shift, and methanation play a major role (Zhao et al., 2009). It results in the
production of gas mainly consisted of hydrogen, carbon oxide, carbon dioxide, methane, and lower amount of light hydrocarbons. The presence of hydrogen mixed
together with other gaseous compounds makes it necessary to separate and purify this
component before its use in industrial processes or as a fuel. It can be performed by
different methods including membrane separation, absorption of carbon dioxide, or
drying (Ni et al., 2006).
The composition of the gaseous products obtained in the thermal conversion of lignocellulosic feedstock depends on the final temperature of the process, heating rate,
residence time, and type of biomass, among others (Ni et al., 2006; Saxena et al., 2008;
Czernik and Bridgwater, 2004; Huber et al., 2006). Therefore, the change of the conditions of lignocellulosic biomass treatment makes it possible to control the concentration of particular chemical compounds in the formed mixture. However, due to the
fact that described process consists of a high number of chemical reactions, both
desired and undesired, this control is not satisfactory. It also affects the profitability
of the hydrogen production. That is why the scientists undertook research aimed at
developing new methods of high-temperature conversion of lignocellulosic biomass
that involved the use of heterogeneous catalysts.
The literature data demonstrates that an application of the catalyst leads to the
increase in the efficiency of hydrogen production (Fig. 10.1). It can be produced
directly from lignocellulosic feedstock or by the conversion of earlier formed biooil in steam reforming process. There are numerous examples of the studies concerning an application of the catalysts to the steam reforming of model bio-oil
compounds, such as acetic acid, acetone, ethanol, ethyl acetate, benzene, xylene, phenol, glycerol, or glucose (Chen and He, 2011; Braga et al., 2016; Nabgan et al., 2016;
Gao et al., 2016; Zou et al., 2015; Seung-hoon et al., 2014).


Formation of hydrogen-rich gas

347

Lignocellulosic
biomass

Intensification of gas
production, higher
selectivity to H2

Thermal
decomposition,
cracking, reforming,
water-gas shift

Liquid phase
(oxygenates)

Gaseous products
(H2, CO, CO2, CH4, C2Hx)

Tar and char

Intensification of cracking,
dehydration, decarboxylation,
decarbonylation

Fig. 10.1 Influence of the catalyst on the efficiency of lignocellulosic biomass conversion
process.

10.2.1 Effect of the type of catalyst
At first, the high-temperature treatment of lignocellulosic biomass was conducted

without the use of catalysts. However, it was observed that an addition of natural minerals such as dolomite or olivine increased the conversion of tar formed during the
decomposition of biomass (Yoon et al., 2010). In spite of the fact that those materials
are cheap and thermally stable, their performance in the mentioned process was moderate. In the further part of the studies, an influence of zeolites and various mesoporous
materials was investigated (Adam et al., 2006; Jeon et al., 2013; Iliopoulou et al.,
2012). Although it was demonstrated that the efficiency of thermal conversion of biomass can be increased when the supported catalysts containing the metallic phase are
used, there are several examples of an application of noble metals as an active phase of
the catalysts for fast pyrolysis of biomass (Kaewpengkrow et al., 2014; Lu et al.,
2010). However, the literature data exhibit that due to the lower price and high
performance in described process nickel is the most commonly used metal for
high-temperature conversion of lignocellulosic biomass (Melligan et al., 2012;
Swierczynski et al., 2007; Wu et al., 2013). Obviously, it should be noted that an application of nickel is associated with several drawbacks. A deactivation of the catalyst
by carbon deposit formation is the most important of them. This process cannot be
fully eliminated during the conversion of lignocellulosic feedstock in the presence
of metallic catalyst. However, it can be noticeably reduced by the choice of the suitable support for Ni. The influence of the support on the activity and stability of the


348

Bioenergy Systems for the Future

catalysts in high-temperature conversion of biomass will be discussed in the further
part of this work.
The mechanism of catalytic conversion of lignocellulosic material is not fully
understood yet. Although the performed studies suggest that intermediates formed
during initial decomposition of biomass are adsorbed on the surface of the catalyst
and undergo dehydrogenation reaction (Ruppert et al., 2014). Swierczynski et al.
(2007) demonstrated that the presence of nickel catalyst facilitates the cleavage of
CdO and CdC bonds in the molecules of primary products of the decomposition
of lignocelluloses. Subsequently, the smaller products formed in cracking reaction
are easier dehydrogenated, and this way, larger amount of hydrogen can be obtained.

Considering other components of gaseous phase formed in thermal treatment of lignocellulosic feedstock, it should be noted that the presence of catalyst promotes
reforming reaction that leads to a decrease in the methane content comparing with
the process performed without catalyst. Furthermore, water-gas shift reaction can proceed in the presence of residual water originated from the mixture of primary products
of biomass decomposition.
The investigations performed by Wang et al. (2014) suggested that intermediates
containing oxygen formed in dehydration step were rather subjected to
decarbonylation in the presence of H-ZSM5 material in comparison with the reaction
performed without catalyst. In the latter case, decarboxylation path was more favored.
Chen and He in their work (Chen and He, 2011) described the possible mechanism
of hydrogen formation in reforming of oxygenates originating from biomass decomposition in the presence of metallic catalyst. In the investigations, ethylene glycol was
chosen as a model compound. This molecule can be adsorbed on the surface of the
catalyst via two carbon atoms or one carbon atom and one oxygen atom, which results
in the formation of two bonds with an active phase. A desired pathway of dehydrogenation of oxygenates consists of the formation of hydrogen and carbon oxide via
the cleavage of CdC bond. In this case, the subsequent conversion of carbon oxide
to carbon dioxide and the production of additional amount of hydrogen via water-gas
shift reaction are also observed.
The second possibility is associated with the cleavage of CdO bond and the formation of alcohol. In the following step, the CdC or CdO breaking can proceed in
the alcohol molecule. It results in the formation of hydrogen and carbon dioxide or
light alkanes, respectively. The next path can lead to the production of acids that
can be transformed into alkanes, carbon oxides, hydrogen, and water. Taking that into
account, it seems that hydrogen production can be considerably enhanced by the initial
dehydrogenation of primary products of biomass decomposition and subsequent
cleavage of CdC bonds while their dehydration and successive breaking of CdO
bonds does not affect H2 formation so positively.
In spite of that the authors of the mentioned publication presented also the results of
theoretical calculation concerning the ability of different metals to the cleavage of
CdC and CdO bonds. It was noted that catalytic properties of particular elements
strictly depend on numerous factors such as the composition of the feedstock, type
of the precursor of the catalyst, its preparation method, type of the support, metal loading, and conditions of thermal treatment of biomass. Therefore, it is difficult to refer
these results to the real reaction systems.

Formation of hydrogen-rich gas

349

Considering that an issue of hydrogen production from lignocellulosic biomass has
been already raised in several reviews mentioned earlier, the following part of this
chapter will be focused on the presentation of the examples of the latest developments
related to the use of heterogeneous catalysts in this process (Fig. 10.2).

10.2.1.1 Bimetallic containing nonnoble metals and
perovskie-type catalyst
Li et al. (2014) investigated an influence of the addition of copper to the catalyst prepared from hydrotalcite-like compounds containing nickel, magnesium, and aluminum on its performance in steam reforming of lignocellulosic biomass tar derived
from pyrolysis of cedar wood. The obtained results revealed that Ni-Cu/Mg/Al catalyst possessed the higher activity than monometallic systems. The best catalytic performance was observed for the sample with copper-to-nickel ratio ¼ 0.25. The authors
concluded that the most important factor for the enhancement of the catalytic activity
was the formation of Ni-Cu alloy that led to higher dispersion of nickel, increase in the
amount of the active sites on the catalyst surface, and its higher affinity to the oxygen.

Bimetallic catalysts based
on nonnoble metals

Catalysts containing
noble metals

Efficiency
improvement
of high temperature
conversion of biomass

Modification of support
of nickel catalyst

Development of new multi-step
processes increasing H2 yield

H2

Fig. 10.2 Routes of the application of heterogeneous catalysts for the improvement of the
efficiency of hydrogen production in high-temperature conversion of lignocellulosic biomass.


350

Bioenergy Systems for the Future

During the steam reforming process, the large molecules existing in tar undergo
dissociation and fragmentation on the surface of an active phase. Simultaneously,
steam dissociates on the metal particles giving hydrogen and oxygen atoms that
can interact with hydrocarbon fragments and form the next portion of hydrogen
and carbon monoxide molecules. Due to the presence of Ni-Cu alloy, the smaller
nickel species are formed on the catalyst surface. It increases an ability of the catalyst
to the creation of nickel active sites (better metal dispersion) and adsorption of hydrogen. Thus, a dissociation of tar takes place more efficiently. Furthermore, the smaller
bimetallic particles demonstrate higher affinity to oxygen in comparison with bigger
nickel species. It promotes the decomposition of tar fragments adsorbed on the catalyst surface and limits the formation of carbon deposit. The coke resistance of bimetallic catalyst can be also enhanced due to the presence of smaller nickel crystallites
because coke formation is stimulated by larger Ni particles than steam reforming reaction. The stability of Ni-Cu/Mg/Al catalyst was also linked with no aggregation of
bimetallic species during the reaction.
The similar investigations of the steam and dry reforming of model pyrolysis gas
with the use of Ni/Fe/Ce/Al2O3 catalyst were described by Xu et al. (2015). In this
case, the role of the second metal (iron) consisted of the promoting of cracking of

pyrolysis intermediates and deposited carbon, which results in the increase in hydrogen production. It was suggested that iron oxide can react with water and hydrocarbon
molecules according to the reactions (10.2), (10.3):
Fex OyÀ1 + H2 O ! Fex Oy + H2

(10.2)

ð2n + mÞFex Oy + Cn H2m ! ð2n + mÞFex OyÀ1 + nCO2 + mH2 O

(10.3)

Moreover, it was demonstrated that the presence of iron can enhance the conversion of
methane and selectivity to hydrogen and carbon dioxide in both steam reforming and
water-gas shift reactions.
Lang et al. (2015) developed iron catalyst for water-gas shift reaction for hydrogen
enrichment in a gas from steam gasification of biomass. They used ceramic foams
impregnated by ceria. The presence of ceria allowed for the increase in the surface
area of the support and formation of relatively well-dispersed crystallites of iron.
However, the submission of the synthesized catalyst to the activity test resulted in
the increase in iron oxide species from about 30 to 45 nm.
Hydrogen-rich gas was also produced by catalytic steam reforming of bio-oil or
bioslurry (containing both bio-oil and biochar with the ratio of 9:1) with the use of
perovskite-type materials (La1ÀxKxMnO3 or LaCo1ÀxCuxO3) (Chen et al., 2016a; Yao
et al., 2016). The advantage of perovskites is their stability at high temperatures that prevents agglomeration of metal atoms in the structure of the catalyst submitted to thermal
treatment. The smaller metal species suppress the formation of carbon deposit that
enhances the activity of the catalysts. Moreover, the substitution of lanthanum present
in the structure of the perovskite by potassium can lead to the enhancement of the oxygen
mobility and surface area of the prepared material. Both tested systems allowed for the
production of hydrogen with the yield up to 70%–75% of the stoichiometric yield.

Formation of hydrogen-rich gas

351

The mixture of KAl and NiAl catalysts was applied in the steam gasification of wheat
straw (Lv et al., 2014) that was conducted in double-bed reactor. In the first step, the
mixture of catalysts and biomass (gasification bed) was fluidized. Subsequently, intermediates were directed to the reforming bed for further conversion. The obtained results
showed that owing to the use of such reaction system it is possible to obtain about 97%
efficiency of carbon conversion. Moreover, it was suggested that KAl promotes decomposition and cracking of biomass and primary products, NiAl favors reforming of tar
and light hydrocarbons, while both catalysts enhance water-gas shift reaction.
The examples of the application of bimetallic containing nonnoble metals and
perovskite-type catalyst to the high-temperature conversion of biomass are also presented in Table 10.1.

10.2.1.2 Modification of support of Ni catalyst
As mentioned earlier, Ni-based systems are the most popular groups of catalysts used
in the high-temperature conversion of lignocellulosic biomass. In recent years, the
main attention of the researchers focused on the investigation of the influence of

Application of bimetallic containing nonnoble metals
and perovskite-type catalyst to the high-temperature conversion
of biomass

Table 10.1

No.

Catalyst

Feedstock

1

Ni-Cu/Mg/Al

Cedar wood

2

Ni/Fe/Ce/Al2O3

Model pyrolysis
gas

3

Model gas

4

Fe/CeO2
supported on
ceramic foam
La1ÀxKxMnO3

5

LaCo1ÀxCuxO3

6

KAl, NiAl

Bio-oil from
fast pyrolysis of
pinewood
sawdust
Bioslurry
(90 wt%
bio-oil, 10 wt%
biochar)
Wheat straw

Process, products,
and remarks
regarding the
influence of
catalyst

Reference

Steam reforming
of biomass tar
Steam and dry
reforming of model
pyrolysis gas
Water-gas shift for
hydrogen
enrichment
Steam reforming
of bio-oil

Li et al. (2014)

Steam gasification
of bioslurry

Yao et al.
(2016)

Steam gasification
in double-bed
reactor

Lv et al. (2014)

Xu et al.
(2015)
Lang et al.
(2015)
Chen et al.
(2016a)


352

Bioenergy Systems for the Future

support on the catalytic performance of such materials including catalyst activity, stability in high-temperature range, and susceptibility to deactivation (Table 10.2 and
Fig. 10.3).
The studies performed in our group (Matras et al., 2012) revealed that ZrO2 was the

most promising support of the nickel catalyst applied to the cellulose pyrolysis process
conducted in a stirred batch reactor at the temperature range up to 700°C. The Ni/ZrO2
sample appeared the most active among the investigated materials (Ni/Al2O3,
Ni/SiO2, Ni/CeO2, Ni/TiO2, and Ni/MgO). However, in this case, nickel was introduced on the surface of commercial oxides. The commercial ZrO2 possessed low

Modification of support of Ni catalyst used in
high-temperature conversion of biomass

Table 10.2

No.

Catalyst

Feedstock

Process, products,
and remarks
regarding the
influence of catalyst

1

Cellulose

Pyrolysis

Matras et al.
(2012)

2

Ni supported on
ZrO2, Al2O3, ZrO2Al2O3, SiO2, CeO2
Ni/ZrO2

Cellulose

Ruppert
et al. (2014)

3

Ni/CeO2-ZrO2

Cellulose

4

Ni/MexO-ZrO2
(Me ¼ Ca, Mg, Na,
and K)

Cellulose

5

Ni/Mesoporous
material (SBA-15,
SBA-16, KIT-6, and

MCM-41)
Ni/CaAlOx

Cellulose

Pyrolysis
Catalysts were
prepared by different
methods (precipitation
with organic template,
precipitation with
NaOH, calcination of
zirconium salt)
Pyrolysis
CeO2 was introduced
into ZrO2 structure by
impregnation,
precipitation, and
sol-gel method
Pyrolysis
Dopants were
introduced on the
catalyst surface by
impregnation method
Pyrolysis

6

Wood
sawdust

Pyrolysis-steam
reforming in fixed-bed
two-stage reactor

Reference

Grams et al.
(2016a)

Ryczkowski
et al. (2016)

Grams et al.
(2016b)

Chen et al.
(2016b)


Formation of hydrogen-rich gas

353

Nickel catalyst

Type of support

Different metal
oxide supports

Application of
mesoporous silicas

Modification of ZrO2 by
cerium and various alkali
and alkaline earth metals

Development of optimal method
of ZrO2 synthesis (precipitation,
calcination of Zr salts, use of organic
template)

Fig. 10.3 Routes of the modification of the support of Ni catalysts for high-temperature
conversion of lignocellulosic biomass.

surface area equal about 5 m2/g (considerably lower than SiO2 or Al2O3) and monoclinic phase. It indicated the high potential of the use of zirconium oxide as a catalyst
support in the described process. Therefore, in the next step of our studies, the investigations were focused on the choice of the optimal method of ZrO2 synthesis ensuring
the formation of the material having the best physicochemical properties, which
would allow for the achievement of the highest hydrogen yield in the hightemperature treatment of lignocellulosic biomass.
The zirconium oxides were prepared by precipitation with organic template, precipitation with sodium oxide, and calcination of zirconium salt (Ruppert et al., 2014).
The obtained results exhibited that the highest hydrogen yield was achieved in the
presence of the catalyst where nickel was introduced on the support synthesized from
ZrOCl2 by precipitation with NaOH, which was followed by calcination at 700°C in
air. Such system contained tetragonal zirconia phase, relatively small crystallites of
nickel oxide, and retained surface area in the reaction conditions. In contrast, the catalyst containing ZrO2 prepared with the use of organic template showed the highest
surface area, but it was not stable during the reaction and decreased of about 40% after
the reaction. Moreover, in the last case, zirconium oxide was amorphous. It is believed
that these differences were the reasons of low activity of that material.
Another crucial aspect was related to the ability of migration of zirconium ions on
the surface of the active phase. The XPS results revealed that the highest migration

tendency was characteristic for tetragonal phase of zirconia (followed by monoclinic
and amorphous). It was responsible for the closer contact between support and metallic phase during the reaction that can be associated with higher stability of the catalyst
and the enhancement of the catalytic activity.


354

Bioenergy Systems for the Future

The next step of the studies was devoted to the investigation of the effect of the
addition of various dopants to the structure of zirconium oxide. In the first case,
ZrO2 was doped by CeO2 (Grams et al., 2016a). The supports containing 15 and
50 wt% of cerium oxide were prepared by three different methods such as impregnation, precipitation, and sol-gel. As in the previous cases, nickel (20 wt%) was introduced onto the support surface by impregnation method, and the activity of the
catalysts was tested in high-temperature conversion of cellulose in the atmosphere
of inert gas. It was demonstrated that an addition of cerium oxide to the zirconium
oxide support considerably increased the amount of the formed hydrogen comparing
with the Ni/ZrO2 catalyst. The production of the highest amount of H2 was observed in
the case of the materials containing supports prepared by sol-gel and impregnation
methods.
It was suggested that zirconium oxide promotes the activity of nickel in the
reforming reaction. Support containing ZrO2 can accumulate H2O molecules (which
are present in the reaction mixture) and produce hydroxyl groups participating in the
hydrogen formation process. On the other hand, literature data show that cerium
oxide, due to its high oxygen storage/release capacity and thermal stability, can limit
the coke formation and increase an efficiency of carbon deposit removal (Ebiad et al.,
2012; Shao et al., 2014). It was also observed in the described measurements where
Ni/ZrO2 catalyst loses its activity in high-temperature conversion of cellulose noticeably faster than the material consisted of Ni supported on CeO2-ZrO2.
The performance of Ni/ZrO2 catalyst in the hydrogen production from biomass can
be also enhanced by the modification of zirconia support by alkali and alkaline earth
metals (Ryczkowski et al., 2016). As in the previous case, it is related to the inhibition

of carbon formation process (Nichele et al., 2014). An addition of alkali metals leads
to the formation of oxygen vacancies that can be responsible for the arising of OH and
O radicals able to stop the accumulation of carbon species on the catalyst surface.
Moreover, dopants take part in the formation of basic sites responsible for carbon
dioxide chemisorption and facilitation of the gasification of accumulated coke (Liu
et al., 2008). The adsorbed CO2 can also shift an equilibrium of the reactions that take
place in the high-temperature biomass conversion and additionally increase the hydrogen production. The obtained results revealed that in spite of the substantial decrease
in the surface area the presence of calcium resulted in the production of higher amount
of hydrogen in comparison with the catalysts containing sodium or potassium. Further
studies, performed by Chen et al. (2016b), showed also that an addition of Ca can be
responsible for the increase in CO selectivity and simultaneous drop in CO2 production in pyrolysis-steam reforming of wood sawdust.
The other group of the investigations was devoted to the application of mesoporous
silicas as supports for nickel catalyst used in the cellulose conversion to hydrogen-rich
gas (Grams et al., 2016b). The samples containing SBA-15, SBA-16, KIT-6, and
MCM-41 were tested. The obtained results revealed that an introduction of Ni on
the surface of mesoporous support can increase H2 yield in comparison with the catalyst supported on commercial SiO2. The highest amount of hydrogen was produced
in the presence of 20% Ni/SBA-15 and 20% Ni/KIT-6 samples. It was demonstrated
that catalytic performance of the studied materials depended not only on surface


Formation of hydrogen-rich gas

355

acidity but also on pore size and pore volume, stability of their structure, and accessibility of Ni on the surface and inside the pores of the catalyst. The high surface area
and bigger pore diameter favored the penetration of the structure of the catalysts by
larger intermediates formed in the initial step of biomass pyrolysis process, which
favors the hydrogen production. However, the comparison of the activity of the catalysts supported on mesoporous materials with the catalytic performance of the systems based on zirconium oxide exhibited that the latter are more efficient in the
formation of H2 in high-temperature treatment of lignocellulosic biomass.

10.2.1.3 Application of catalyst containing noble metals
In spite of that nickel catalysts are the most widely used in high-temperature biomass
conversion processes the researchers continue also the studies on the systems containing noble metals (Table 10.3). Czernik and French (2014) investigated the
hydrogen production by autothermal reforming of fast pyrolysis bio-oil originated
from oak, poplar, and pine. They applied commercial Pt/Al2O3 catalyst containing
0.5 wt% of platinum that favors the reforming of bio-oil vapors. The reaction system
consisted of reformer working at 850°C and WGS reactor operating at 350°C (in the
water-gas shift reaction step iron/chromium system was used). The reaction mixture
formed in the first step of the process besides hydrogen also contained nitrogen,

Application of catalysts containing platinum or nickel
to the novel processes of high-temperature conversion of biomass

Table 10.3

No.

Catalyst

Feedstock

1

Pt/Al2O3
(reforming)
Fe/Cr (WGS)

2

Pt/C,

Pt/TiO2, and
Pt/Al2O3

3

Ni/MgO

Bio-oil from
fast pyrolysis
of oak, poplar,
and pine
Wheat straw
hydrolysates
and glucose
solution
Cotton stalks

4

Ni
nanoparticles

Pinewood,
wheat straw

5

Ni/Al2O3
doped with
Ca

Pinewood
sawdust

Process, products,
and remarks regarding
the influence of catalyst

Reference

Autothermal reforming
of fast pyrolysis bio-oil

Czernik and
French (2014)

Aqueous-phase
reforming

Irmak et al.
(2015)

Simultaneous
gasification of pyrolysis
gases and char in
entrained flow-bed
reactor
Subcritical and
supercritical water
gasification of biomass

Continuous fast
pyrolysis and in-line
steam reforming

Chen et al.
(2015)

Nanda et al.
(2016)
Arregi et al.
(2016)


356

Bioenergy Systems for the Future

carbon oxides, and traces of methane (CH4 content was much lower than that obtained
in the process with the use of nickel catalyst (Czernik et al., 2002)). Additional amount
of H2 was produced by the conversion of steam and carbon oxide in the second reactor
for water-gas shift reaction. Of course, the steam reforming process contributes to the
production of the highest hydrogen yield, but it is endothermic and requires the delivery of the external heat. This can be provided by exothermic partial oxidation reaction
that also occurred in autothermal reforming.
One of the drawbacks of the described method is the presence of heavier molecules
(i.e., oligomeric lignin or carbohydrates) resulted in incomplete volatility of the
processed bio-oil. It leads to the decrease in the efficiency of the whole process. Moreover, the material that cannot be evaporated contributes in the formation of carbon
deposit on the walls of the reactors or on the surface of the catalyst, which results
in the further drop in the H2 production. The catalyst may also undergo deactivation
due to the presence of various inorganic impurities in the feedstock. Usually, the conversion of carbon remains at the level of 70%–90%, and nonvolatile residues are
trapped in the filter before their introduction to the reforming unit.

Not only Czernik and French but also Irmak et al. (2015) studied the performance
of platinum catalysts in the conversion of biomass to hydrogen-rich gas. In this case,
the investigations were focused on the influence of the pretreatment method on the
activity of Pt supported on active carbon, titanium oxide, and aluminum oxide in
aqueous-phase reforming of wheat straw hydrolysates and glucose solution conducted
at 250°C in the stainless steel microbench reactor. This time, the prepared catalysts
contained higher amount of noble metal equal 8 wt%. The obtained results confirmed
that the best distribution of metal particles can be obtained on the surface of active
carbon (about 10 nm), while in the case of other supports the control of the size of
Pt crystallites was hindered due to the agglomeration of the metal species on their
surface. It was indicated that the size of metal particles is strictly connected with catalytic activity of the tested materials. The catalyst with smaller size of Pt crystallites
enhanced the efficiency of aqueous-phase reforming process in a higher degree than
those with larger metal particles. The catalyst activity was additionally increased
by the choice of the optimal reduction of the investigated samples. It was demonstrated that the best effect was achieved when the catalyst was initially submitted
to the chemical reduction with the use of NaBH4 and subsequent thermal treatment
in the nitrogen atmosphere. The chemical reduction influenced not only the size of
Pt particles but also the structure of active carbon. The IR spectra revealed a significant decrease in the intensity of the band corresponding to the presence of CdO
groups, among others.
The independent studies have been devoted to the application of bimetallic Ni-Pt
and Ni-Rh catalyst in the low-temperature gasification process (Dı´az-Rey et al.,
2015). In spite of the fact that algae biomass was used as a feedstock, it is worth to
look at the obtained results that showed that the presence of noble metals inhibited
the migration of nickel atoms on the surface of the catalyst, stabilized the Ni species,
and promoted their reducibility. Owing to that, the bimetallic catalysts were more
resistant to deactivation related to not only carbon deposition but also poisoning by


Formation of hydrogen-rich gas

357

sulfur present in the feedstock. Moreover, the formation of Ni-Pt species can be
responsible for the further improvement of the efficiency of secondary reactions
(i.e., cracking or reforming of biomass volatiles and tars produced in the initial step
of the process), which allows the enhancement of hydrogen production. It is also associated with the increase in the selectivity of the process toward hydrogen and carbon
oxide rather than to the formation of coke.

10.2.1.4 Development of new methods of lignocellulosic
biomass conversion
Literature data demonstrate that not only new but also commercially available catalysts can be used for the production of hydrogen-rich gas via high-temperature treatment of lignocellulosic feedstock (Table 10.3). However, in this case, the studies are
rather focused on the development of new methods of lignocelluloses conversion.
Chen et al. (2015) investigated simultaneous gasification of gas and char formed in
the pyrolysis of cotton stalks in entrained flow-bed reactor. Apart from the application
of commercial Ni/MgO catalyst that was responsible for the increase in the efficiency
of steam reforming, the authors mentioned about the catalytic activity of char. It was
demonstrated that simultaneous gasification of char and pyrolysis gas allowed for the
increase in the total conversion of carbon from 79% (observed in the case of two-stage
pyrolysis and steam reforming process) to about 92%. The presence of char influenced
the composition of tar by the promotion of transformation of polycyclic aromatic compounds to low-ring or even low-chain compounds. The described changes, resulting
from the interactions between pyrolysis gas and char, led also to considerable decrease
in the carbon deposit formation on the surface of nickel catalyst and almost twofold
growth in the yield of the produced hydrogen.
Nanda et al. (2016) performed studies on subcritical and supercritical water gasification of pinewood and wheat straw. That process has already been conducted previously, but in the present case, biomass submitted to gasification was additionally
impregnated by nickel salt. It was showed that the efficiency of the incorporation
of Ni into the structure of lignocelluloses depended on the amount of lignin that hinders the penetration of the material by dopants. This is why pinewood containing more
lignin accumulated lower amount of nickel in comparison with wheat straw. It is worth
noting that wheat straw possessed also higher content of alkali metals such as sodium,
potassium, calcium, and magnesium. The presence of the catalyst resulted in substantial increase in the total gas yield (about 60%) and the amount of hydrogen (even
100%) with simultaneous growth in the gasification efficiency of carbon (about
65%). The highest hydrogen yield was achieved at 500°C with biomass-to-water ratio

1:10 and longer residence time that enhanced cracking reaction. The higher temperature of supercritical water promoted the production of hydrogen, carbon dioxide, and
methane due to the increase in the efficiency of water-gas shift and methanation
processes.
The efficiency of the hydrogen production can be also increased by the use of the
integrated process consisted of continuous fast pyrolysis of pinewood sawdust


358

Bioenergy Systems for the Future

followed by in-line steam reforming of the pyrolysis vapors with the use of commercial Ni/Al2O3 catalyst doped by calcium (Arregi et al., 2016). In this case, the first step
of biomass conversion is conducted in a conical spouted-bed reactor, while the second
one in a fluidized-bed reactor. The char formed during pyrolysis is continuously
removed from the reaction system. The main product of this step is bio-oil (about
75 wt%) that is passed together with gases (about 7 wt%) to steam reforming unit.
However, some part of that fraction can be lost due to not complete vaporization that
influences the efficiency of the whole process.
According to the authors of that work, the advantage of the proposed system in
comparison with gasification process is the production of gases that are free of
tars. It was showed that the use of steam in the pyrolysis process does not significantly affect the distribution of produced substances that is similar to that
obtained with the use of nitrogen as a fluidizing agent. An application of nickel
catalyst increases the conversion of the volatiles present in the reaction mixture
and hydrogen yield. The optimization of the reaction conditions (temperature,
steam-to-biomass ratio, and space time) allows for the formation of about 96%
of theoretically possible amount of H2 taking into account stoichiometry of the
occurred reactions.

10.3

Hydrogen not only as a source of energy

When hydrogen is produced for energy purposes, very often, the gasification of the
bio-based feedstock is the process of choice. There are however plenty other than
energy-based applications for hydrogen-rich gas, and among them are all reactions
that require external use of hydrogen, fuel cell applications, or many others. It can
be even considered that hydrogen can play pivotal role in future biorefinery schemes
and its demand can increase. Many processes based on cellulose valorization concerning platform molecule production (e.g., sugar alcohols) or biofuel additives
(e.g., γ-valerolactone) require hydrogen. Just mentioning the process of biofuel
manufacture, where for the reduction of oxygen-containing compounds, a great
amount of reductant typically represented by molecular fossils delivered hydrogen
is required.
Therefore, it is essential that hydrogen for those purposes will be also bio-based.
Hydrogen can be obtained from carbohydrates via reforming reaction; however, this
process requires harsh conditions and noble metal catalysts (Nguyen-Phan et al.,
2016). On the other hand, formic acid (FA) can be used for hydrogen storage, hydrogen donor, or molecular hydrogen for many reactions. It can be easily obtain from
lignocellulosic biomass. One way of its synthesis is acid-catalyzed hydrolysis of cellulose where sugars formed in the first step undergo subsequent processes to form
equimolar amount of formic and levulinic acids (Weingarten et al., 2012). Much
higher yield of FA (up to 60%–70%) can be obtained when hydrolysis is combined
with high oxygen pressure. Glucose of hydroxymethylfurfural formed in the first step
of cellulose hydrolysis is subsequently oxidized to formic acid ( Jin et al., 2008)
(Scheme 10.1).


Formation of hydrogen-rich gas

359

O
C6 sugar

H2O

O

O

OH

–HCOOH
H2O

5-Hydroxymethylfurfural
OH
O
HO

O

HO
O

OH

OH
O
OH

H2O2

O
OH

+

O
Levulinic acid

OH
Formic acid

O
+ H 2O
OH
Formic acid

Cellulose

Scheme 10.1 Two different ways of the synthesis of formic acid from biomass.

10.3.1 Factors which influence the decomposition of FA
The decomposition of formic acid has been mostly characterized by two reaction
pathways: dehydrogenation (reaction 10.4) to form H2 and CO2 and dehydration
(reaction 10.5) to form H2O and CO:
HCOOH ! H2 + CO2

(10.4)

HCOOH ! H2 O + CO

(10.5)

CO + H2 O Ð CO2 + H2

(10.6)

CO + 3H2 ! CH4 + H2 O

(10.7)

CO2 + 4H2 ! CH4 + 2H2 O

(10.8)

2HCOOH ! HCHO + CO2 + H2 O

(10.9)

In real reaction conditions, however, the situation is much more complex. Subsequent
water-gas-shift reaction (WGS, reaction 10.6) can occur. In the presence of some catalysts like Ru, it has been also identified that Fischer-Tropsch reaction can take place
(reactions 10.7, 10.8). Although less often, it is also mentioned in the literature that the
formation of formaldehyde is possible due to the reaction of formate ions (HCOOÀ)
(reaction 10.9) (Redondo et al., 2014). However, those important issues of side reactions are mostly omitted by researchers, and only very few such examples have been
described (Zhang et al., 2013; Yi et al., 2013; Bulushev et al., 2013; Ciftci et al., 2013).
While the side reactions decrease the hydrogen selectivity, CO is also a poison of
the catalyst active centers. Therefore, the design of selective catalysts for this process
still remains a great challenge. FA decomposition has been performed in a broad
range of temperatures from room temperature to 300°C, mostly in the gas phase
and flow system ( Jia et al., 2013, 2014). It was demonstrated that FA decomposition
in the gas phase starts at 80°C and is finished in the temperature range of 200–260°C.

The reaction was mostly found to be zero-order. In the gas phase, the reaction conditions seem not to have a deciding influence on the catalyst performance. It was


360

Bioenergy Systems for the Future

observed that neither the temperature nor the FA concentration has much influence on
the process selectivity (Bulushev et al., 2010; Ojeda and Iglesia, 2009). In the aqueous
phase, the reaction has been examined to a lesser extent, but the conditions are often
milder (25–90°C). It is expected however that it proceeds in a different way, as solvent
effects may change the reaction rate by several orders of magnitude or completely
alter the mechanism for reactions in aqueous phase. In the gas phase on metal surface,
HCOOÀ was identified as a reactive intermediate, and the decomposition of HCOOÀ
into CO2 and H2 was the rate-limiting step in this reaction. By contrast, in the liquid
phase, HCOOÀ is rather considered only a spectator that adsorbs on the catalyst surface (Hu et al., 2012).
Additionally, the presence of water is reported to lower the activation barrier of
the decarboxylation reaction in comparison with dehydration, therefore shifting the
selectivity of FA decomposition toward hydrogen path (Akiya and Savage, 1998).
The presence of water—beside influencing the reaction pathways—can also change
the reactivity of the catalyst itself. It was observed by us that some metals in its presence can significantly lower their activation barrier for hydrogenation or dehydrogenation reactions, for example, Ru that consequently is much more active in aqueous
phase (Michel et al., 2014). It was also noted that, in the case of some metals, the selectivity of the FA decomposition can be significantly improved in the presence of water;
for example, Ir gave 98.3%–99% selectivity for H2, additionally giving CO-free H2
at the temperatures of 110–150°C (Solymosi et al., 2011).

10.4

Catalysts used for FA decomposition

Both hetero- and homogeneous catalysts, with mono- and bimetallic nanoparticles,

supported on various materials or just as bulk nanoparticle catalysts, were tested in
FA decomposition.

10.4.1 Homogeneous catalysts
Work concerning homogeneous catalysts was recently described in review paper
summarizing the recent discoveries on that topic. Homogeneous catalysts used for this
reaction are restricted mainly but not only to different Ru complexes formed in situ.
Among different examples, it is interesting to mention Ru catalysts formed from [Ru
(H2O)6]2 or commercial RuCl3xH2O with two equivalents of meta-trisulfonated
triphenylphosphine (mTPPTS), which especially with the addition of HCOONa,
resulted in excellent activity (Fellay et al., 2008, 2009).
Czaun et al. used homogeneous ruthenium trichloride and triphenylphosphines as
catalyst precursors in emulsion and in biphasic (aqueous/organic) systems. Activity
and reaction selectivity of in situ formed Ru(HCO2)2(CO)2(PPh3)2, Ru(CO)3(PPh3)2,
and Ru2(HCO2)2(CO)4(PPh3)2 catalysts were as well as in previous case additionally
improved by the addition of the surfactants, especially sodium dodecyl sulfate to the
biphasic system of toluene/water (Czaun et al., 2014). Also in another example, tetranuclear ruthenium complex [Ru4(CO)12H4], identified as active species formed in situ
from precursors and additional CO resulted from FA decomposition, showed superior


Formation of hydrogen-rich gas

361

activity in reaction that was boosted by the addition of NaCOOH. Although the catalysts
showed high activity and very often high selectivity in the reaction conditions, their
separation from reactants and reaction products and their reuse would be challenging.
Additionally, the necessity of using surfactants or other adducts like HCOONa pushed
scientist for more intensive development of heterogeneous catalysts.

10.4.2 Heterogeneous catalysts
Consequently, the large part of research concerns the use of different heterogeneous
catalysts based mostly on the following metals: Pt Ag, Pd, Rh, Au Cu, Ni, and Fe
(Zhang et al., 2013; Ojeda and Iglesia, 2009; Park et al., 2002; Gazsi et al., 2011;
Luo et al., 2012; Boddien et al., 2011). The earlier work on this topic was classified
in some reviews ( Johnson et al., 2010; Grasemann and Laurenczy, 2012; Enthaler
et al., 2010). The recent research can be categorized into the following areas:
(a) the search of selective catalysts producing CO-free gas, which are operating in
ambient-temperature conditions mainly for fuel cell applications; (b) basic research
focusing on understanding the principles of the catalytic active sites for this reaction
and related mechanisms of FA decomposition; and (c) formic acid decomposition and
its direct use for hydrogenation reactions—mainly one-pot reactions (Wa˛chała et al.,
2016) (Fig. 10.4).
There are many methodologies of preparing selective catalysts. One of the methods
is based on synthesis modification in a way that would allow to obtain ultrafine welldispersed nanoparticles of the metal. Zhu et al. (2014) synthesized highly dispersed Pd
nanoparticles deposited on nanoporous carbon MSC-30. They exchange the ligand of
the chlorine precursor of Pd to hydroxide ones by the addition of NaOH. The excellent
catalytic properties were assigned not only to enhanced electronic interaction of
chlorohydroxypalladium(II) complex and support that led to small nanoparticles
but also to the crucial role of the carbon support. Authors obtained high TOF value
of 750 hÀ1 at 25°C and evidenced also that formic acid was decomposed exclusively
to H2 and CO2.
Also in the work of Cao and coworkers, the high activity of the catalyst was attributed to small metal particles. Working on gold deposited on zirconia catalysts, they
applied slightly modified deposition-precipitation method for synthesizing ultradispersed gold catalysts consisting in gold subnanoclusters deposited on zirconia.
They were able to reach TOF values as high as 1590 hÀ1 for the reaction performed
in 40°C (Bi et al., 2012). Another approach was related with changing the properties of
the metal by its doping/alloying with a second one. Recent studies pointed out that
Pd-based catalysts, such as Pd-Au (Huang et al., 2010; Zhou et al., 2008; Yuan
and Liu, 2013) and Pd-Ag (Tedsree et al., 2011; Xu et al., 2014; Zhang et al.,
2013), showed great potential, especially by selectively producing almost CO-free

H2. Additionally, the activity of such bimetallic catalysts was much higher than monometallic counterparts. Especially a lot of research was devoted to the Ag-Pd system.
Many different modifications, synthesis approaches, and component ratios were
considered. From recent examples, the work of Zhang et al. is highly interesting that
investigated the AgPd nanoparticles prepared via coreduction of organic precursors of


362

Bioenergy Systems for the Future

Approaches for catalysts design for
formic acid decomposition

Small metal particles
Uniform particles of the very small average
size in average of 2–5 nm. Activity
improvement assigned mainly to small
nanoparticles and support influence

Alloy structure
(a) Synergistic effect between two metals in the
formed alloy responsible for increase of activity.
(b) Increased dispersion identified
(c) Search of optimum composition of two
metals for optimum performance

Further modifications
of synthesis

Further improvement

of bimetal composition

CO free catalyst
for low temperature
purposes

Highly dispersed Ag-Pd hollow spheres
anchored on graphene

Core shell structure
-Synthesis with PVP or PVA used as stabilizing
agents or via thermal modification

Average diameter 18 nm, and wall 5 nm.
Activity increase due to higher availibility
of active sites and specific interactions of
bimetal with the support

Activity increase mainly due to
(a) Charge transfer from the metal cores to Me shell
This can strength the adsorption and bridge the
formate via stronger back donation and increase
the amount of produced hydrogen.
(b) Change in particle size distribution
(c) Change of CO adsorption strength

Fig. 10.4 Schematic representation of the factors that influence the design of the catalysts
used in formic acid decomposition.

the metals, assisted by the presence of oleic acid 1-octadecene used as surfactants

where the latter one holds also as in situ reducing agent. By this way, they managed
to obtain uniform particles of the very small average size of 2.2 nm. Very high activity
(TOF 382 hÀ1 for reaction at 50°C) and selectivity observed with this material were
attributed to the very small NPs size, and synergistic effect between those two metals
in the formed alloy was also identified, with an optimum composition of Ag42Pd58
(Zhang et al., 2013). In another example, the group of Chen ( Jiang et al., 2016) managed to synthesize highly dispersed AgPd hollow spheres anchored on graphene by
using facile one-pot hydrothermal route. Those hollow nanoparticles (NPs) had an
average diameter of about 18 nm and a wall thickness of about 5 nm. The excellent
activity of NPs was attributed not only to the presence of this very thin shell that
increases the availability of active sites for regents but also to specificity of the interactions of bimetallic particles with the graphene layer and the very good dispersion of


Formation of hydrogen-rich gas

363

NPs on this support. Bimetallic nanoparticles belong to a group of highly important
nanomaterials that are largely unexplored. Besides, numerous possible combinations
of metallic elements in nanoparticles and electronic interactions among metals, bimetallic NPs may produce totally new physical and chemical properties. Moreover,
change in the particle morphology constitutes another way of tailoring their physicochemical properties. For the above reasons, the choice of the synthesis method plays
an important role in obtaining the catalysts with desired properties. Therefore, a lot of
place in the literature was dedicated to the investigations of many different synthesis
methods, as it has been recognized that controlling the bimetallic structures plays a
crucial role in achieving the demanding properties. Especially, bimetallic core-shell
structure that contains an inner core of one metal element and an external shell of the
other metal element has shown some unique physical and chemical properties. The
design of the catalysts is therefore crucial, and that is why there were several interesting reports that claim that unique catalytic properties can be obtained when the coreshell structure is implemented. This synthesis was investigated for several different
alloys like Pd-Au or Pd-Ag (Huang et al., 2010). Authors claimed that such a structure
allows superior activity for hydrogen generation. The different methods were investigated with the use of stabilizing agent or just with the different reduction/calcination
steps combined with chemical or thermal reduction. In the later case, advantage was

taken from the ability of metals to migrate in certain conditions. For example, in the
case of Au-Pd systems, Pd has ability to migrate during calcination steps as it is more
easily oxidized than Au and forms the surface PddO bonds. By tuning, the conditions
of subsequent reduction are possible to influence the composition of the core versus
shell (Herzing et al., 2008; Hilaire et al., 1981). By implementing this approach,
PdAu@Au core-shell structure was synthesized by the use of chemical reduction without the presence of stabilizing agent. Such catalysts showed superior activity and high
H2 selectivity in the formic acid decomposition (Huang et al., 2010). Also, Ag@Pd
core-shell synthesized by the approach using PVP as a stabilizing agent was proved
to be much more active in FA decomposition than the corresponding alloy possessing
the same composition or Pd alone (Tedsree et al., 2011).
Despite that many researches have been done in this area, the question emerges
why alloy could shape selectivity. There are several explanations of the activity
improvement. One of the explanations is related with the existence of a synergistic
effect between two metals in the alloy. For example, in the case of Ag-Pd, the electrons can transfer from Ag to Pd, so that the bimetallic system was strongly modified
in the way that it becomes more active. In some reports, this close interaction of two
metals is claimed to be responsible for activity and selectivity increase ( Jiang et al.,
2016). On the other hand, it is claimed that the addition of, for example, Ag and Au,
reduces the size of nanoparticles in comparison with the corresponding monometallic
systems (Zhou et al., 2008). Although the same information can be found for Ag-Pd,
some reports claimed that the effect of Au addition is better than that of Ag judging the
particle size effect, which may be one of the reasons for the better performance of
Pd-Au/C compared with Pd-Ag/C (Zhou et al., 2008). Additionally, the adsorption
strength of CO is changed for such a bimetallic composition, being much weaker


364

Bioenergy Systems for the Future

so that Ag and Au do not form stable complexes with CO ( Judai et al., 2003) and that

is why such a catalyst is more resistant to poisoning.
Tedsree et al. explained the activity of metals and bimetallic systems by the use of
density functional theory (DFT). Based on the previous findings of the group of
Norskov (Ruban et al., 1997), they systematize the activity of metals in FA dehydrogenation as Pd > Rh > Pt % Ru > Au > Ag and relate this difference to the distance of
the d-band center to the Fermi level. The closer was the d-band center to the Fermi
level of a metal, the higher was the adsorption the optimum was reached for Pd, which
is in agreement with the literature as palladium is one of the most often used monometallic catalysts for FA decomposition (Larsen et al., 2006; Tedsree et al., 2011).
Based on the same principle, authors explained the activity increase of the coreshell structure of bimetallic particles particularly Ag-Pd. They explained it by the
charge transfer from the metal cores to Pd shell, which is the reason for the net difference in the work function. This can strengthen the adsorption and bridge the formate via stronger back donation, and the same can increase the amount of produced
hydrogen.

10.5

Decomposition of formic acid to hydrogen
and subsequent hydrogenation reaction

There are some interesting examples in the literature showing the use of formic acid as
direct hydrogen source for hydrogenation reactions. They are mostly related with biomass valorization approach where the increase of sustainability by the use of internal
hydrogen source is especially desirable. As formic acid is obtained in equimolar
amount with levulinic acid in cellulose hydrolysis process, it is therefore very often
directly used for LA hydrogenation. γ-Valerolactone, a product of this hydrogenation,
has many potential applications among them and is considered as a platform molecule
for conversion to many useful chemicals such as polymers of high thermal stability or
biofuels (Ruppert et al., 2015; Luo et al., 2015; Alonso et al., 2013). The challenge of
this process is often related with the use of the same catalysts for FA decomposition
and LA hydrogenation. This is however not the only requirement. When FA decomposition is done in batch reactor, a high conversion of formic acid is required before
the LA hydrogenation can take place. The reason of that is related with the nature of
FA adsorption. Formic acid can in the easy and strong way adsorb dissociatively on
metal surfaces in the form of formate at the temperature as low as À193°C (Sun and
Weinberg, 1991; Avery et al., 1982). For comparison, levulinic acid and hydrogen

have a much weaker energy of adsorption (Ruppert et al., 2016). As a result, the surface of the catalyst is completely covered by formate until a high enough (very often
full) conversion is reached in formic acid dehydrogenation, liberating catalytic sites
for H2 and the levulinic acid. It can be even said that formate acts as an inhibitor for the
LA hydrogenation when the same catalysts is used (Ruppert et al., 2016). Therefore,
intensive search for adequate catalyst for these subsequent reactions can be observed.
Both hetero- and homogeneous catalysts were investigated in this reaction.
Ruthenium-based homogeneous catalysts, such as [(η6-C6Me6)Ru(bpy)(H2O)][SO4]
or RuCl3/PPh3, were proved to be efficient for this reaction (Mehdi et al., 2008;


Formation of hydrogen-rich gas

365

Deng et al., 2009; Tang et al., 2014). However, drawbacks such as poor stability and a
weak resistance to water and to mineral acids that are often present in real biomass
feedstock after hydrolysis process consequently forced scientists to develop more
stable heterogeneous catalysts.
Considering the heterogeneous catalysts, Ru was shown to be the most active
among Pd-Pt ones when reaction was performed in batch reactor in aqueous phase.
It was also demonstrated that the further modification of Ru can influence the activity
in both dehydrogenation and subsequent hydrogenation reactions. Ru/C that was
reduced at higher temperature was much more active in formic acid dehydrogenation.
This could be related to the presence of larger nanoparticles of the metal than after a
lower-temperature treatment. Luo et al. (2013) suggested that the H2 production from
HCOOH could be then favored on larger particles, so on (0001) facets, which was
also observed in the literature before for other oxophilic metallic surfaces like Ni
for which the activity predicted by DFT calculations was higher on Ni(111) surfaces
versus Ni(211).
Also, Au catalysts were tested in mentioned process. Du et al. was evaluating different oxides and active carbon as support for metallic Au nanoparticles. Du et al.

(2011) evidenced that the role of the support is crucial and that zirconia was the most
promising one, with an excellent 99% GVL yield being achieved over Au/ZrO2 after
6 h of reaction at 150°C with equimolar amount of FA to LA.
Ag and Ag-Ni catalytic systems were also investigated, and high activity with
almost full GVL yield was shown over 10% Ag 20% Ni supported as well on zirconia
after 5 h of reaction performed at 220°C. The outstanding performance of Ag-Ni-ZrO2
catalyst was attributed to the surface synergy between Ag and Ni. Not only noble
metals were investigated. Upare et al. showed that nickel-promoted copper-silica
can be very active in described process. Authors tested their nanocomposites in high
temperature of 285°C in vapor phase in the flow system. The high activity and stability
were attributed to synergetic effect between the two metals, and they claimed additionally that the addition of Ni was preventing the sintering of the nanoparticles
(Upare et al., 2015).
Also, bimetallic nanoparticles of Ni-Pt and Ni-Ru supported on ZrO2 and gammaAl2O3 were investigated in the solvent-free hydrogenation of levulinic acid using
formic acid as a hydrogen source. A conversion of 35% with a selectivity to GVL over
99% was obtained on 0.6 Ni-1.9 Ru/gamma-Al2O3 in the flow reaction at 50 bar of
hydrogen at 90°C for 20 min.
Authors indicated that the catalytic performance was depending on the metal
dispersion on the surface and textural and surface properties of the support material
(Al-Najia et al., 2016).

10.6

Summary

The two showcases of hydrogen production were discussed in this chapter. One of
them is based on high-temperature processes. Here, we highlighted the new methods
of lignocellulosic treatment that prevent the catalysts deactivation and development of
new stable catalysts based mainly on nonnoble metals. Moreover, we described the

366

Bioenergy Systems for the Future

influence of the catalysts support on the efficiency of the performed reactions. In the
second approach, we discussed the application of formic acid as a hydrogen source.
We concentrated on its use for direct hydrogenation reactions mainly in liquid phase.
We focused on the development and use of novel catalysts for this process. The new
synthesis methods aiming at metal nanoparticle formation were highlighted. Also,
bimetallic catalysts were described taking into account experimental and theoretical
approach.
Presented studies showed a huge potential for the catalyst application for hydrogen
production processes. Their use can considerably limit the costs of the nowadays and
future chemical technologies and improve the efficiency of the H2 formation. Therefore, we believe that the increase of bio-based hydrogen production will take the major
place in the future biorefinery schemes.

References
Adam, J., Antonakou, E., Lappas, A.A., Stocker, M., Nilsen, M.H., Bouzga, A., Hustad, J.E.,
Rye, G., 2006. In situ catalytic upgrading of biomass derived fast pyrolysis vapours in a fixed
bed reactor using mesoporous materials. Microporous Mesoporous Mater. 96, 93–101.
Akiya, N., Savage, P.E., 1998. Role of water in formic acid decomposition. AICHE J. 44, 405–415.
Al-Najia, M., Yepezb, A., Balub, A.M., Romerob, A.A., Chena, Z., Wildea, N., Lic, H.,
Shihc, K., Gl€asera, R., Luquebb, R., 2016. Insights into the selective hydrogenation of
levulinic acid to γ-valerolactone using supported mono- and bimetallic catalysts.
J. Mol. Catal. A: Chem. 417, 145–152.
Alonso, D.M., Wettstein, S.G., Dumesic, J.A., 2013. Gamma-valerolactone, a sustainable platform molecule derived from lignocellulosic biomass. Green Chem. 15, 584–595.
Arregi, A., Lopez, G., Amutio, M., Barbarias, I., Bilbao, J., Olazar, M., 2016. Hydrogen
production from biomass by continuous fast pyrolysis and in-line steam reforming.
RCS Adv. 5, 25975–25985.
Avery, N.R., Toby, B.H., Anton, A.B., Weinberg, W.H., 1982. Decomposition of formic acid on

Ru(001): an EELS search for a formic anhydride intermediate. Surf. Sci. 122, 574–578.
Balat, M., Balat, M., Kirtay, E., Balat, H., 2009. Main routes for the thermo-conversion of biomass into fuels and chemicals. Part 1: pyrolysis systems. Energy Convers. Manage.
50, 3147–3157.
Bi, Q.-Y., Du, X.-L., Liu, L.-M., Cao, Y., He, H.-Y., Fan, K.-N., 2012. Efficient subnanometric
gold-catalyzed hydrogen generation via formic acid decomposition under ambient conditions. J. Am. Chem. Soc. 134, 8926–8933.
Boddien, A., Mellmann, D., G€artner, F., Jackstell, R., Junge, H., Dyson, P.J., Laurenczy, G.,
Ludwig, R., Beller, M., 2011. Efficient dehydrogenation of formic acid using an iron catalyst. Science 333 (6050), 1733–1736.
Braga, A.H., Sodre, E.R., Santos, J.B.O., Marques, C.M.P., Bueno, J.M.C., 2016. Steam
reforming of acetone over Ni- and Co-based catalysts: effect of the composition of reactants and catalysts on reaction pathways. Appl. Catal. B 195, 16–28.
Bridgwater, A.V., 2012. Review of fast pyrolysis of biomass and product upgrading. Biomass
Bioenergy 38, 68–94.
Bulushev, D.A., Beloshapkin, S., Ross, J.R.H., 2010. Hydrogen from formic acid decomposition over Pd and Au catalysts. Catal. Today 154, 7–12.


Formation of hydrogen-rich gas

367

Bulushev, D.A., Beloshapkin, S., Plyusnin, P.E., Shubin, Y.V., Bukhtiyarov, V.I.,
Korenev, S.V., Ross, J.R.H., 2013. Vapour phase formic acid decomposition over
PdAu/γ-Al2O3 catalysts: effect of composition of metallic particles. J. Catal. 299, 171–180.
Chen, D., He, L., 2011. Towards an efficient hydrogen production from biomass: a review of
processes and materials. ChemCatChem 3, 490–511.
Chen, Z., Zhang, S., Chen, Z., Ding, D., 2015. An integrated process for hydrogen-rich gas production from cotton stalks: the simultaneous gasification of pyrolysis gases and char in an
entrained flow bed reactor. Bioresour. Technol. 198, 586–592.
Chen, G., Yao, J., Liu, J., Ya, B., Shan, R., 2016a. Biomass to hydrogen-rich syngas via catalytic
steam reforming of bio-oil. Renew. Energy 91, 315–322.
Chen, F., Wu, Ch., Dong, L., Vassallo, A., Williams, P.T., Huang, J., 2016b. Characteristics and
catalytic properties of Ni/CaAlOx catalyst for hydrogen-enriched syngas production from
pyrolysis-steam reforming of biomass sawdust. Appl. Catal. B 183, 168–175.

Ciftci, A., Ligthart, D.A.J.M., Pastorino, P., Hensen, E.J.M., 2013. Nanostructured ceria
supported Pt and Au catalysts for the reactions of ethanol and formic acid. Appl. Catal.
B: Environ. 130–131, 325–335.
Czaun, M., Goeppert, A., Kothandaraman, J., May, R.B., Haiges, R., Surya Prakash, G.K.,
Olah, G.A., 2014. Formic acid as a hydrogen storage medium: ruthenium-catalyzed generation of hydrogen from formic acid in emulsions. ACS Catal. 4, 311–320.
Czernik, S., Bridgwater, A.V., 2004. Overview of applications of biomass fast pyrolysis oil.
Energy Fuels 18, 590–598.
Czernik, S., French, R., 2014. Distributed production of hydrogen by auto-thermal reforming of
fast pyrolysis bio-oil. Int. J. Hydrogen Energy 39, 744–750.
Czernik, S., French, R., Feik, C., Chornet, E., 2002. Hydrogen by catalytic steam reforming of
liquid byproducts from biomass thermoconversion processes. Ind. Eng. Chem. Res.
41, 4209–4215.
Deng, L., Li, J., Lai, D.M., Fu, Y., Guo, Q.X., 2009. Catalytic conversion of biomass-derived
carbohydrates into gamma-valerolactone without using an external H2 supply. Angew.
Chem. Int. Ed. Engl. 48 (35), 6529–6532.
Dı´az-Rey, M.R., Cortes-Reyes, M., Herrera, C., Larrubia, M.A., Amadeo, N., Laborde, M.,
Alemany, L.J., 2015. Hydrogen-rich gas production from algae-biomass by low temperature catalytic gasification. Catal. Today 257, 177–184.
Du, X.L., He, L., Zhao, S., Liu, Y.M., Cao, Y., He, H.Y., Fan, K.N., 2011. Hydrogenindependent reductive transformation of carbohydrate biomass into γ-valerolactone and
pyrrolidone derivatives with supported gold catalysts. Angew. Chem. Int. Ed. Engl.
50 (34), 7815–7819.
Ebiad, M.A., Abd El-Hafiz, D.R., Elsalamony, R.A., Mohamed, L.S., 2012. Ni supported high
surface area CeO2-ZrO2 catalysts for hydrogen production from ethanol steam reforming.
RSC Adv. 2, 8145–8156.
Enthaler, S., von Langermann, J., Schmidt, T., 2010. Carbon dioxide and formic acid—the couple for environmental-friendly hydrogen storage? Energy Environ. Sci. 3, 1207–1217.
Fellay, C., Dyson, P.J., Laurenczy, G., 2008. A viable hydrogen-storage system based on selective formic acid decomposition with a ruthenium catalyst. Angew. Chem. Int. Ed.
47, 3966–3968.
Fellay, C., Yan, N., Dyson, P.J., Laurenczy, G., 2009. Selective formic acid decomposition for
high-pressure hydrogen generation: a mechanistic study. Chem. Eur. J. 15, 3752–3760.
Gao, N., Wang, X., Li, A., Wu, Ch., Yin, Z., 2016. Hydrogen production from catalytic steam
reforming of benzene as tar model compound of biomass gasification. Fuel Process.

Technol. 148, 380–387.


368

Bioenergy Systems for the Future

Gazsi, A., Ba´nsa´gi, T., Solymosi, F., 2011. Decomposition and reforming of formic acid on
supported Au catalysts: production of CO-free H2. Phys. Chem. C 115 (31), 15459–15466.
Grams, J., Niewiadomski, M., Ryczkowski, R., Ruppert, A.M., Kwapi
nski, W., 2016a. Activity
and characterization of Ni catalyst supported on CeO2-ZrO2 for thermo-chemical conversion of cellulose. Int. J. Hydrogen Energy 41, 8679–8687.
Grams, J., Potrzebowska, N., Goscianska, J., Michalkiewicz, B., Ruppert, A.M., 2016b. Mesoporous silicas as supports for Ni catalyst used in cellulose conversion to hydrogen rich gas.
Int. J. Hydrogen Energy 41, 8656–8667.
Grasemann, M., Laurenczy, G., 2012. Formic acid as a hydrogen source—recent developments
and future trends. Energy Environ. Sci. 5 (8), 8171–8181.
Herzing, A.A., Carley, A.F., Edwards, J.K., Hutchings, G.J., Kiely, C.J., 2008. Microstructural
development and catalytic performance of AuÀPd nanoparticles on Al2O3 supports: the
effect of heat treatment temperature and atmosphere. Chem. Mater. 20 (4), 1492–1501.
Hilaire, L., Legare, P., Holl, Y., Maire, G., 1981. Interaction of oxygen and hydrogen with
Pd-Au alloys: an AES and XPS study. Surf. Sci. 103, 125–140.
Hu, Ch., Ting, S.-W., Chan, K.-Y., Huang, W., 2012. Reaction pathways derived from DFT for
understanding catalytic decomposition of formic acid into hydrogen on noble metals. Int.
J. Hydrogen Energy 37, 15956–15965.
Huang, Y., Zhou, X., Yin, M., Liu, C., Xing, W., 2010. Novel PdAu@Au/C core–shell catalyst:
superior activity and selectivity in formic acid decomposition for hydrogen generation.
Chem. Mater. 22, 5122–5128.
Huber, G.W., Iborra, S., Corma, A., 2006. Synthesis of transportation fuels from biomass:
chemistry, catalysts, and engineering. Chem. Rev. 106, 4044–4098.
Iliopoulou, E.F., Stefanidis, S.D., Kalogiannis, K.G., Delimitis, A., Lappas, A.A.,

Triantafyllidis, K.S., 2012. Catalytic upgrading of biomass pyrolysis vapors using transition metal-modified ZSM-5 zeolite. Appl. Catal. B 127, 281–290.
Irmak, S., Meryemoglu, B., Ozsel, B.K., Hasanoglu, A., Erbatur, O., 2015. Improving activity of
Pt supported metal catalysts by changing reduction method of Pt precursor for hydrogen
production from biomass. Int. J. Hydrogen Energy 40, 14826–14832.
Jeon, M.-J., Jeon, J.-K., Suh, D.J., Park, S.H., Sa, Y.J., Joo, S.H., Park, Y.-K., 2013. Catalytic
pyrolysis of biomass components over mesoporous catalysts using Py-GC/MS. Catal.
Today 204, 170–178.
Jia, L., Bulushev, D.A., Podyacheva, O.Yu., Boronin, A.I., Kibis, L.S., Gerasimov, E.Yu.,
Beloshapkin, S., Seryak, I.A., Ismagilov, Z.R., Ross, J.R.H., 2013. Pt nanoclusters stabilized by N-doped carbon nanofibers for hydrogen production from formic acid. J. Catal.
307, 94–102.
Jia, L., Bulushev, D.A., Beloshapkin, S., Ross, J.R.H., 2014. Hydrogen production from formic
acid vapour over a Pd/C catalyst promoted by potassium salts: evidence for participation of
buffer-like solution in the pores of the catalyst. Appl. Catal. B: Environ. 160–161, 35–43.
Jiang, Y., Fan, X., Xiao, X., Qin, T., Zhang, L., Jiang, F., Li, M., Li, S., Ge, H., Chen, L., 2016.
Novel AgPd hollow spheres anchored on graphene as an efficient catalyst for dehydrogenation of formic acid at room temperature. J. Mater. Chem. A 4, 657–666.
Jin, F., Yun, J., Li, G., Kishita, A., Tohji, K., Enomoto, H., 2008. Hydrothermal conversion of
carbohydrate biomass into formic acid at mild temperatures. Green Chem. 10, 612–615.
Johnson, T.C., Morris, D.J., Wills, M., 2010. Hydrogen generation from formic acid and alcohols using homogeneous catalysts. Chem. Soc. Rev. 39 (1), 81–88.
Judai, K., Abbet, S., W€orz, A.S., Heiz, U., Giordano, L., Pacchioni, G., 2003. Interaction of Ag,
Rh, and Pd atoms with MgO thin films studied by the CO probe molecule. J. Phys. Chem. B
107, 9377–9387.


Formation of hydrogen-rich gas

369

Kaewpengkrow, P., Atong, D., Sricharoenchaiku, V., 2014. Effect of Pd, Ru, Ni and ceramic
supports on selective deoxygenation and hydrogenation of fast pyrolysis Jatropha residue
vapors. Renew. Energy 65, 92–101.

Lang, Ch., Secordel, X., Zimmermann, Y., Kiennemann, A., Courson, C., 2015. Hightemperature water–gas shift catalysts for hydrogen enrichment of a gas produced by biomass steam gasification. C R Chim 18, 315–323.
Larsen, R., Ha, S., Zakzeski, J., Masel, R.I., 2006. Unusually active palladium-based catalysts
for the electrooxidation of formic acid. J. Power Sources 157, 78–84.
Li, D., Koik, M., Che, J., Nakagawa, Y., Tomishig, K., 2014. Preparation of Ni–Cu/Mg/Al catalysts from hydrotalcite-like compounds for hydrogen production by steam reforming of
biomass tar. Int. J. Hydrogen Energy 39, 10959–10970.
Liu, S., Guan, L., Li, J., Zhao, N., Wei, W., Sun, Y., 2008. CO2 reforming of CH4 over stabilized
mesoporous Ni–CaO–ZrO2 composites. Fuel 87 (12), 2477–2481.
Lu, Q., Tang, Z., Zhang, Y., Zhu, X., 2010. Catalytic upgrading of biomass fast pyrolysis vapors
with Pd/SBA-15 catalysts. Ind. Eng. Chem. Res. 49, 2573–2580.
Luo, Q., Feng, G., Beller, M., Jiao, H.J., 2012. Formic acid dehydrogenation on Ni(111) and
comparison with Pd(111) and Pt(111). Phys. Chem. 116 (6), 4149–4156.
Luo, Q., Wang, T., Beller, M., Jiao, H., 2013. Hydrogen generation from formic acid
decomposition on Ni(2 1 1), Pd(2 1 1) and Pt (2 1 1). J. Mol. Catal. A: Chem.
379, 169–177.
Luo, W., Sankar, M., Beale, A.M., He, Q., Kiely, C.J., Bruijnincx, P.C.A., Weckhuysen, B.M.,
2015. High performing and stable supported nano-alloys for the catalytic hydrogenation of
levulinic acid to γ-valerolactone. Nat. Commun. 6, 6540–6549.
Lv, X., Xiao, J., Du, Y., Shen, L., Zhou, Y., 2014. Experimental study on biomass steam gasification for hydrogen-rich gas in double-bed reactor. Int. J. Hydrogen Energy
39, 20968–20978.
Matras, J., Niewiadomski, M., Ruppert, A., Grams, J., 2012. Activity of Ni catalysts for hydrogen production via biomass pyrolysis. Kinet. Catal. 53, 565–569.
Mehdi, H., Fa´bos, V., Tuba, R., Bodor, A., Mika, L.T., Horva´th, I.T., 2008. Integration of
homogeneous and heterogeneous catalytic processes for a multi-step conversion of
biomass: from sucrose to levulinic acid, γ-valerolactone, 1,4-pentanediol, 2-methyltetrahydrofuran, and alkanes. Top. Catal. 48, 49–54.
Melligan, F., Hayes, M.H.B., Kwapinski, W., Leahy, J.J., 2012. Hydro-pyrolysis of biomass and
online catalytic vapor upgrading with Ni-ZSM-5 and Ni-MCM-41. Energy Fuels
26, 6080–6090.
Michel, C., Zaffran, J., Ruppert, A.M., Matras-Michalska, J., Jedrzejczyk, M., Grams, J.,
Sautet, P., 2014. Role of water in metal catalyst performance for ketone hydrogenation:
a joint experimental and theoretical study on levulinic acid conversion into gammavalerolactone. Chem. Commun. 50, 12450–12453.
Nabgan, W., Amran, T., Abdullah, T., Mat, R., Nabgan, B., Gambo, Y., Moghadamian, K.,

2016. Acetic acid-phenol steam reforming for hydrogen production: effect of different
composition of La2O3-Al2O3 support for bimetallic Ni-Co catalyst. J. Environ. Chem.
Eng. 4 (3), 2765–2773.
Nanda, S., Reddy, S.N., Dalai, A.K., Kozinski, J.A., 2016. Subcritical and supercritical water
gasification of lignocellulosic biomass impregnated with nickel nanocatalyst for hydrogen
production. Int. J. Hydrogen Energy 41, 4907–4921.
Nguyen-Phan, D.T., Baber, A.E., Rodriguez, J.A., Senanayake, S.D., 2016. Hydrogen from
oxygenated molecules Au and Pt nanoparticle supported catalysts tailored for H2 production: from models to powder catalysts. Appl. Catal. A Gen. 518, 18–47.