Access to this full-text is provided by Wiley.
Content available from ChemSusChem
This content is subject to copyright. Terms and conditions apply.
Biomass Source Influence on Hydrogen Production through
Pyrolysis and in Line Oxidative Steam Reforming
Irati Garcia,[a] Gartzen Lopez,*[a, b] Laura Santamaria,[a] Enara Fernandez,[a] Javier Bilbao,[a]
Martin Olazar,[a] Maite Artetxe,[a] and Maider Amutio[a]
This study evaluates the potential of several biomasses differing
in nature and composition for their valorization by pyrolysis
and in line oxidative steam reforming. The first task involved
the fast pyrolysis of the biomasses in a conical spouted bed
reactor (CSBR) at 500°C, in which product yields were analyzed
in detail. Then, the oxidative steam reforming (OSR) of pyrolysis
volatiles (gases and bio-oil) was approached in a fluidized bed
reactor (FBR). The reforming experiments were performed at
600°C, with a steam/biomass (S/B) ratio of 3 and catalyst (Ni/
Al2O3) space times of 7.5 and 20 gcat mingvol1. Concerning
equivalence ratio (ER), a value of 0.12 was selected to ensure
autothermal operation. Remarkable differences were observed
in H2production depending on the type of biomass. Thus, pine
wood led to a H2production of 9.3 wt %. The lower productions
obtained with rice husk (7.7 wt %) and orange peel (5.5 wt %)
are associated with their higher ash and fixed carbon content,
respectively, which limit the efficiency of biomass conversion to
bio-oil. However, in the case of the microalgae, the poor
performance observed is because of the lower conversion in
the reforming step toward gases due to the composition of its
pyrolysis volatile stream.
1. Introduction
The current scenario of global H2production is marked by the
extensive use of fossil fuels, and therefore a pressing need for
changing towards sustainable sources. In fact, 96 % of the
global H2production is currently obtained from non-renewable
sources, such as natural gas, naphtha and coal.[1] Accordingly,
the use of biomass as an alternative and renewable source for
H2production is increasingly interesting within the current
scenario of energy transition and reduction of CO2emissions.[2]
H2can be obtained from biomass by either thermochemical
or biological methods.[3] It is to note that biological processes
use enzymes and microorganisms, whereas thermochemical
ones are based on chemical reactions at higher temperature.
Within the current technical scenario, thermochemical conver-
sion routes have considerable advantages over biological ones,
such as higher reaction rates, higher yields and greater
flexibility to valorize different types of biomasses.[4]
Thus, the most common biomass thermochemical conver-
sion processes for H2production are gasification and steam
reforming. Gasification is carried out at high temperatures (>
750°C) and using different gasifying agents (O2, steam, CO2, air
or their mixtures).[5] More specifically, steam gasification is the
best option for the production of a H2rich syngas, with its yield
being in the 5 to 8 wt% range under optimum process
conditions and using suitable catalysts.[6–9] However, the full
scale development of biomass gasification is mainly condi-
tioned by the process conversion efficiency and formation of
undesired by-products, especially tar.[10–12] Reforming on highly
active catalysts allows attaining full conversion of biomass
derived products into a H2rich stream free of tar. Moreover, this
process can be performed under milder conditions compared
with those used in biomass gasification.[13] Another interesting
option lies in the indirect approach based on the steam
reforming of the bio-oil obtained by biomass fast pyrolysis on
different metallic catalysts.[14–16] This strategy involves the onsite
production of bio-oil and its transportation as densified biomass
to centralized reforming units. This process has proven to lead
to high conversion efficiency and H2production values above
10 wt%.[17–20] More recently, the direct conversion of biomass
into H2has been proposed using a two-step process of pyrolysis
and in line catalytic steam reforming.[13] Thus, biomass is
converted into a volatile stream (gases and bio-oil) and biochar
in the pyrolysis step, and the volatiles are then reformed on a
suitable catalyst in a second reforming reactor. It is to note that
the H2production capacity of biomass by the strategy of
pyrolysis-reforming is of the same order as that reported for
bio-oil steam reforming, i.e., of around 10–11 wt% under
optimum process conditions.[21–26] However, the technical devel-
opment of the process is limited, as the vast majority of
previous studies have been performed in batch mode in lab
scale units.[13]
The biomass derived product stream obtained by steam
reforming is conditioned by two major limitations, as are
process endothermicity and fast catalyst deactivation.[13,16,27,28]
The incorporation of oxygen into the steam reforming step, i. e.,
[a] I. Garcia, G. Lopez, L. Santamaria, E. Fernandez, J. Bilbao, M. Olazar,
M. Artetxe, M. Amutio
Department of Chemical Engineering, University of the Basque Country
UPV/EHU, P.O. Box 644 - E48080 Bilbao (Spain)
E-mail: gartzen.lopez@ehu.eus
[b] G. Lopez
IKERBASQUE, Basque Foundation for Science, Bilbao, Spain
© 2024 The Authors. ChemSusChem published by Wiley-VCH GmbH. This is
an open access article under the terms of the Creative Commons Attribution
Non-Commercial License, which permits use, distribution and reproduction
in any medium, provided the original work is properly cited and is not used
for commercial purposes.
Wiley VCH Dienstag, 15.10.2024
2420 / 357252 [S. 303/314] 1
ChemSusChem 2024,17, e202400325 (1 of 12) © 2024 The Authors. ChemSusChem published by Wiley-VCH GmbH
ChemSusChem
www.chemsuschem.org
Research Article
doi.org/10.1002/cssc.202400325
the oxidative steam reforming (OSR), allows operating under
autothermal conditions, and therefore overcome the problems
associated with the heat supply to the reforming reactor, with
the reduction in H2production being of low significance.[15,29–31]
A previous study showed that the equivalence ratio (ER)
required to attain neutral energy balance in the pyrolysis-OSR
of several biomasses is between 0.07 and 0.13.[32] In this study,
an original continuous unit made up of a conical spouted bed
for the pyrolysis step and a fluidized bed for the steam
reforming step was used. It is to note that a continuous steady
process is required to operate under OSR conditions, as
otherwise the ER cannot be modified throughout the reforming
of pyrolysis volatiles.
Interestingly, the operation under OSR conditions has
proven to have a positive effect on catalysts deactivation, as it
leads to the attenuation of the deactivation rate due to the
partial in situ combustion of the deposited coke.[33–35]
The other main challenge in the biomass pyrolysis and in
line steam reforming process, i.e., the fast catalyst deactivation
by coke deposition, is greatly conditioned by the composition
of the volatile stream entering the reforming reactor.[36] Thus,
the identification of the main coke precursor compounds has
been of great interest in the literature.[37] However, due to the
high heterogeneity of bio-oil derived compounds, made up of
water and a wide range of oxygenated compounds, major
research efforts have been focused on the evaluation of
deactivation mechanisms using different model compounds
such as phenol, acetic acid, glycerol, furfural, among others.[35,38]
It is to note that the reactivity of these compounds varies
considerably when they are separately or jointly reformed in a
mixture,[39] being of special relevance the studies conducted
using real bio-oil.
The aim of this paper is the evaluation of the role played by
biomass composition in the OSR process of pyrolysis volatiles.
In fact, the composition of the pyrolysis stream greatly depends
on the original biomass nature, i.e., the content of main
components, such as cellulose, hemicellulose, lignin or
extractives.[40,41] In this scenario, the pyrolysis of diverse kind of
biomass wastes for bio-oil production has been analyzed in the
literature, with pinewood being the most used one due to its
high representativeness compared to other softwood forest
materials.[42,43] Moreover, the amount of agroforestry wastes
derived from food production, namely orange peel or rice husk,
has exponentially increased, and therefore, its valorization by
pyrolysis route has gained increasing attention.[44,45] However,
the bio-oil produced from these wastes has low energy density,
which hinders its direct used as fuel due to their high content
of water, oxygen and ash. A promising alternative resource for
bio-oil production is microalgae biomass, due to its high growth
rate, availability and high carbon fixing efficiency. Besides, bio-
oils derived from microalgae are mainly composed of hydro-
carbons, nitrogen containing compounds, ketones, alcohols,
acids, lactones, phenols and aldehydes.[46] It is to note that the
product yields and composition obtained in all pyrolysis studies
greatly depend on the operating conditions or reactor config-
uration used.
As aforementioned, bio-oil composition has a remarkable
impact on the reforming step performance, affecting H2yield,
reforming reactivity and catalysts stability.[16,39] In fact, the full
scale development of any pyrolysis-OSR process is conditioned
by its capability to successfully valorize biomasses of different
nature. Indeed, the supply of waste biomass is of marked
seasonal nature, with great fluctuations in the features of the
raw materials and their price over the year.[47] Accordingly,
process flexibility to handle very different biomasses must be
ensured. In this study, four biomasses accounting for agro-
forestry wastes (pine wood and rice husk), food industry
(orange peel) and microalgae were selected as feedstock for the
pyrolysis-OSR process. Product yields and detailed compositions
of the gas and bio-oil obtained in the fast pyrolysis step were
first determined and discussed. Then, the role played by the
pyrolysis volatile stream composition in the OSR was evaluated.
Experimental
Biomass Characterization
This study approached the conversion of different representative
biomasses, namely, pinewood sawdust, rice husk, orange peel and
a microalgae (Nannochloropsis), by pyrolysis and in line OSR. These
four biomasses were dried, ground and sieved in order to obtain a
material with suitable physical features for ensuring correct
operation of the device designed to feed them into the pyrolysis
unit.
The proximate analysis of the samples was carried out in a TGA
Q5000IR thermogravimetric analyzer. Moreover, a LECO CHNS-932
was used for the determination of ultimate analysis. The determi-
nation of the biomasses macromolecular composition was based
on the deconvolution of the peaks obtained by differential
thermogravimetry (DTG). Table 1 summarizes the characterization
of the biomasses used in this study.
As observed in Table 1, there are remarkable differences in the
composition of the studied biomasses. Thus, pine wood is a typical
lignocellulosic biomass and has low ash content but high of carbon
and volatile matter. Rice husk is also made up of lignin, cellulose
and hemicellulose, but has a much higher ash content (12.9 wt %).
In the case of orange waste, it has a rather high pectin content,
which leads to a remarkable fixed carbon content. Finally, the
chemical composition of the microalgae is completely different to
the other biomasses. Thus, it contains carbohydrates, proteins and
lipids in its composition. In addition, it is the one with the highest
ash content (20.04 wt%) of the studied biomasses. According to a
preliminary evaluation of these compositions, the most interesting
biomass for the pyrolysis-reforming process is pine wood sawdust,
as it has the highest content of volatile matter, as well as those of
carbon and hydrogen. It is to note that only volatile products are
fed into the reforming step in order to produce H2.
Table 2 shows the detailed composition of the ashes contained in
the studied biomasses determined by X-ray Fluorescence (XRF),
(model AXIOS, PANalytical). It is noteworthy that the ash in the
biomass catalyzes the cracking reactions involving pyrolysis vola-
tiles, which usually lead to an increase in the gas yield at the
expenses of a decrease in that of bio-oil.[48] As observed in Tables 1
and 2, the ash content and composition greatly vary depending on
the type of biomass. Rice husk ashes are mainly made up of SiO2. In
fact, the recovery of amorphous SiO2from pyrolysis char has been
Wiley VCH Dienstag, 15.10.2024
2420 / 357252 [S. 304/314] 1
ChemSusChem 2024,17, e202400325 (2 of 12) © 2024 The Authors. ChemSusChem published by Wiley-VCH GmbH
ChemSusChem
Research Article
doi.org/10.1002/cssc.202400325
regarded as a promising valorization route.[49] However, composi-
tion is more complex in the case of the other biomasses, with high
contents of alkaline and alkaline earth metals (AAEM).
Catalyst Characterization
This study was carried out using a Ni based commercial catalyst
(ReforMax®330 or G90-LDP) supplied by Süd Chemie, which was
originally developed for methane steam reforming process. It is to
note that this catalyst showed a suitable performance for the steam
reforming of biomass pyrolysis volatiles.[36,50] It has a NiO content of
14 wt%, is supported on Al2O3, and is also doped with Ca. The
catalyst was supplied in the form of perforated rings, which were
ground and sieved to obtain a suitable particle size for the fluidized
bed reactor, i. e., 0.4–0.8 mm.[51]
The fact that it was a commercial catalyst hindered the perform-
ance of its detailed characterization, and therefore only physical
properties and reducibility were analyzed. The textural properties
were evaluated by N2adsorption-desorption in a Micromeritics
ASAP 2010, and temperature programmed reduction (TPR) was
conducted in an AutoChem II 2920 Micromeritics.
According to the N2adsorption-desorption results, the catalyst is a
mesoporous material with a total BET surface area of 19 m2g1and
an average pore size of 122 Å. The TPR profiles showed a main
peak located at 550°C ascribed to NiO reduction interacting with α-
Al2O3, and another one at higher temperature (at around 700°C)
attributed to NiAl2O4. Therefore, the catalyst was subjected to a
reduction process prior to use in order to ensure the conversion of
NiO to active Ni0phase. The reduction was performed at 710 °C for
4 h using a 10 vol % H2stream. Further details about catalyst
characterization can be found elsewhere.[52]
Experimental Equipment and Process Conditions
An original continuous bench scale unit was used for biomass fast
pyrolysis and in line oxidative steam reforming (see Figure 1). Thus,
biomass fast pyrolysis was conducted in a conical spouted bed
reactor (CSBR), whereas the reforming of the pyrolysis product
stream (permanent gases and bio-oil) was carried out in a fluidized
bed reactor (FBR).
The choice of a CSBR is based on its excellent performance for
biomass and waste fast pyrolysis proven in previous studies.[53–55] In
addition, an improved version of this reactor was used in this study.
Thus, a nonporous draft tube was incorporated in order to improve
the reactor’s hydrodynamic performance. This internal device
contributes to improving bed stability, controlling solid circulation
pattern and reducing the spouting flow rate.[56–59] Details about
reactor and draft tube dimensions were reported in previous
papers.[29]
Below the reactor there is a section for generating steam and
preheating the inlet stream. A coil evaporator is located within an
electric cartridge, which provides the heat to ensure quick and
uniform steam generation. In addition, the pyrolysis reactor is
located in a radiant oven (1250 W) that provides heat to both the
lower and the upper reactor sections, i. e., the preheater and
spouted bed itself. The CSBR is equipped with a lateral outlet for
the continuous removal of biochar, avoiding its accumulation in the
bed throughout continuous operation.
As mentioned above, the OSR of biomass fast pyrolysis volatiles is
conducted in a fluidized bed reactor, as it has clear advantages
over the fixed bed one, such as higher heat transfer rate, easier
control of temperature, lower coke deposition rate and easier
scaling up.[60] Moreover, the use of a FBR eases oxygen distribution
in the reactor and avoids the formation of hot spots. Thus, an
oxygen distributor was located at the axis of the bed. The design of
this device, as well as further details about the FBR were reported
elsewhere.[29] The mentioned device allows feeding independently
the pyrolysis volatile stream and oxygen. Hence, the pyrolysis
products enter the fluidized bed through the bottom distributor
plate, whereas oxygen is distributed along the bed using a
multipoint injector. Accordingly, oxidation reactions occur in the
catalytic bed, which supply the heat required by the highly
endothermic steam reforming reactions.
Table 1. Characterization of the biomasses used in this study.
Feedstock Pine
wood
Orange
waste
Rice
husk
Microalgae
Ultimate analysis (wt %)a
Carbon 49.08 42.70 42.00 35.54
Hydrogen 6.03 6.40 5.40 5.20
Nitrogen 0.04 1.00 0.40 6.20
Oxygenb44.35 47.60 39.30 33.02
Proximate analysis (wt %)c
Volatile mat-
ter
81.06 74.10 70.50 58.68
Fixed carbon 18.44 23.60 16.60 21.28
Ash 0.50 2.30 12.90 20.04
Moisture 9.4 1.5 1.1 2.0
Macromolecular composition (wt %)d
Hemicellulose 23.0 17.0 21.7 –
Cellulose 51.0 17.5 46.4 –
Lignin 26.0 29.4 31.9 –
Pectin – 36.1 – –
Carbohydrates – – – 28.9
Proteins – – – 45.5
Lipids – – – 25.6
HHV (MJ
kg1)
19.8 19.4 16.8 14.3
aas received, bby difference, con an air-dried basis, don an ash free basis.
Table 2. Chemical composition of the ashes.
Feedstock Pine wood Orange waste Rice husk Microalgae
SiO2
Al2O3
Fe2O3
MnO
MgO
CaO
Na2O
K2O
TiO2
P2O5
SO3
CuO
SrO
8.84
2.38
2.30
2.46
10.44
32.34
1.93
11.30
0.11
2.55
3.59
–
–
0.29
0.33
0.09
–
4.78
29.47
1.98
30.9
0.02
8.34
3.46
–
–
98.02
0.52
0.11
0.01
0.11
0.23
0.10
0.38
0.02
0.08
–
–
–
2.31
0.60
2.62
0.49
7.04
46.92
1.97
0.98
–
32.99
3.89
0.14
0.08
Wiley VCH Dienstag, 15.10.2024
2420 / 357252 [S. 305/314] 1
ChemSusChem 2024,17, e202400325 (3 of 12) © 2024 The Authors. ChemSusChem published by Wiley-VCH GmbH
ChemSusChem
Research Article
doi.org/10.1002/cssc.202400325
As observed in Figure 1, the pyrolysis and reforming reactors are
located in a forced convection oven, where temperature is of
around 300°C to avoid the condensation of bio-oil prior analysis.
In order to operate in continuous regime, the plant is equipped
with systems for continuous and uniform feed of steam and
biomass. Thus, water was fed using a high precision Gilson 307
pump, which was vaporized in the lower section of the pyrolysis
reactor in the coil evaporator heated by an electric cartridge. The
different biomasses were fed using a vibrated piston feeder, whose
detailed design was reported in previous studies.[61] It is to note
that, given the different density and granulometry of the biomasses
used, the upward velocity of the piston had to be calibrated for
each one.
Moreover, the plant is equipped with several mass flow meters for
the control of different gases: i) N2, used for reactor heating and
cooling processes, ii) air, used for burning the coke and so cleaning
the reactors, iii) H2, required for the catalyst reduction prior to the
reforming reaction and iv) O2, directly injected into the reforming
reactor to attain OSR conditions.
Downstream each reactor, there is a device to remove any solid
particle entrained from the corresponding bed. In the case of
pyrolysis in the CSBR, a cyclone was used to retain any char or sand
fine particle from the pyrolysis volatile stream. In the case of the
FBR, a sintered steel retained any catalyst fine particle entrained.
Both elements were located within the forced convection oven to
avoid bio-oil condensation.
Experimental Conditions
The selection of the conditions for the pyrolysis and in line OSR of
different biomasses is based on previously conducted studies. On
the one hand, biomass fast pyrolysis step was carried out at 500 °C,
as this temperature ensures an efficient biomass conversion and
maximum bio-oil yield.[45,54,62,63] On the other hand, the OSR step
conditions are based on a parametric study performed for the
reforming of pine wood sawdust pyrolysis volatiles.[29] Thus, the
following conditions were used: i) temperature of 600 °C, ii)
equivalence ratio (ER) of 0.12, iii) steam/biomass (S/B) ratio of 3 and
iv) space times of 7.5 and 20 gcat mingvol1. These conditions were
selected to ensure suitable catalysts activity, stability and operation
under autothermal conditions. Moreover, the effect of space time
was assessed by modifying the amount of reforming catalyst in the
bed, taking into account the different amount of the pyrolysis
volatiles obtained with each biomass sample.
In all the cases biomass feed rate was of 1 gmin1and that of water
3 mLmin1. It should be noted that only steam was in the feed into
the pyrolysis reactor, as it is an inert fluidizing agent at the
moderate temperatures used, i.e., there is no reforming activity in
the pyrolysis reactor. This steam flow rate allows operating under
stable spouting conditions using a bed made up of 75 g of sand
with a particle size in the 0.2 to 0.3 mm range. In the case of the
FBR, the bed consists of 25 g of a mixture of catalyst and sand, with
their particle sizes being 0.4–0.8 mm and 0.3–0.35 mm, respectively.
It should be noted that the oxygen flow rate fed into the reforming
reactor was calculated for each biomass considering the yields and
composition of pyrolysis volatiles in order to attain an ER value of
0.12 in the operation.
Product Analysis and Reaction Indexes
The pyrolysis products obtained with the different biomasses were
analyzed using on-line chromatographic techniques, i.e., a chroma-
tography (GC) and micro-chromatography (microGC). FID response
factors of different bio-oil families (phenols, ketones, aldehydes and
so on) were determined in a previous study[63] and used in the GC
for the quantification of the bio-oil compounds. Moreover, the
identification of bio-oil compounds was carried out by means of a
gas chromatograph-mass spectrometer (GC-MS Shimadzu QP-
Figure 1. Scheme of the continuous bench scale unit for pyrolysis-oxidative steam reforming.
Wiley VCH Dienstag, 15.10.2024
2420 / 357252 [S. 306/314] 1
ChemSusChem 2024,17, e202400325 (4 of 12) © 2024 The Authors. ChemSusChem published by Wiley-VCH GmbH
ChemSusChem
Research Article
doi.org/10.1002/cssc.202400325
2010S) by analyzing the collected liquid using a similar column as
in the GC.
The analysis of biomass pyrolysis-reforming was carried out
combining gas chromatography (GC) and micro-chromatography
(microGC). Thus, the GC was used for the analysis of oxygenates
and hydrocarbons, which ranged from methane to heavy bio-oil
compounds. A Varian 3900 GC equipped with a HP-Pona column
and a flame ionization detector (FID) was used for that purpose. As
observed in Figure 1, the sample was taken from the product
stream downstream the reforming reactor, and was injected into
the GC using a thermostated line maintained at 280 °C.
Moreover, the gaseous stream including H2, CH4, CO, CO2and light
hydrocarbons (C2
C4) was analyzed in the microGC after condensa-
tion and filtering (see Figure 1). A Varian 4900 microGC with three
analytical modules (molecular sieve, Porapak (PPQ) and plot
alumina) and thermal conductivity detectors (TCD) was used for
gaseous products analysis. The gas samples for the microGC were
taken in Tedlar gas sampling bags throughout continuous runs.
In order to determine the product yields, information from the GC
and microGC was gathered. Thus, global and elemental mass
balances (carbon, hydrogen and oxygen) in the reforming step
were solved. It should be noted that the mass balance closure was
above 95% in all the cases.
Once the mass balances were solved, different reaction indices
were calculated in order to assess the reforming step performance.
Therefore, the conversion of biomass fast pyrolysis volatiles to
gases was calculated as the ratio between the moles of C contained
in the outlet stream of the reforming step and those in the pyrolysis
volatile stream:
X¼Cgas
Cvolatiles �100 (1)
The yield of each carbon containing product (such as CH4, CO or
CO2) was calculated as follows:
Yi¼Fi
Fvolatiles �100 (2)
where Fiand Fvolatiles are the molar flow rates of compound i and the
pyrolysis product stream entering the OSR reactor, respectively,
expressed in C equivalent moles.
The H2yield was calculated as the ratio between the molar flow of
H2(FH2) and the maximum allowable by stoichiometry (F0
H2):
YH2¼FH2
F0
H2�100 (3)
The stoichiometry considered for the reforming of biomass fast
pyrolysis volatiles is as follows:
CnHmOkþaO2þ2n2akð ÞH2O
!nCO2þ ð2n2aþm=2kÞH2
(4)
Moreover, the H2production per biomass mass unit fed into the
pyrolysis-reforming process was also determined to assess the
overall process efficiency:
Prod:H2¼mH2
m0
Biomass �100 (5)
where mH2 and m0
biomass are the mass flow rates of H2obtained in
the product stream and biomass fed into the pyrolysis-reforming
process, respectively.
2. Results
2.1. Pyrolysis Step Results
In order to understand the influence of biomass composition
on the pyrolysis and in line OSR processes, the results obtained
in the pyrolysis step must be analyzed in detail. Thus, product
yields and the composition of the pyrolysis streams were first
studied, with especial emphasis being placed on the volatile
fraction (gases and bio-oil), as this is the stream to be upgraded
in the OSR step.
Figure 2 shows the product distribution obtained in the
pyrolysis of the different biomasses studied at 500 °C. As
observed, there are remarkable differences in the product
distributions obtained in the pyrolysis of the biomasses
considered. Although the fraction to be reformed is that made
up of biomass pyrolysis volatiles, the interest is to obtain the
highest bio-oil yield; that is, the gas fraction also contains non-
condensable components, such as CO and CO2, whose H2
production potential is rather limited. In this respect, pine wood
sawdust showed the best potential with a bio-oil yield of
75.3%. In spite of the similar composition of rice husk
(lignocellulosic nature), the bio-oil yield obtained is slightly
lower, 69.0%. This result is mainly associated with the
remarkable ash content, mainly made up of SiO2, which leads to
an increase in char formation. However, the notable presence of
Figure 2. Product distributions obtained for the biomasses studied in the
fast pyrolysis at 500°C.
Wiley VCH Dienstag, 15.10.2024
2420 / 357252 [S. 307/314] 1
ChemSusChem 2024,17, e202400325 (5 of 12) © 2024 The Authors. ChemSusChem published by Wiley-VCH GmbH
ChemSusChem
Research Article
doi.org/10.1002/cssc.202400325
pectin in the composition of orange waste promotes char
formation in the pyrolysis process,[63–65] with the subsequent
reduction in the bio-oil yield (54.9 %). Likewise, the contents of
cellulose, hemicellulose and lignin also have a great influence
on biomass pyrolysis product distribution. On the one hand, the
presence of cellulose and hemicellulose enhances bio-oil
production.[53,66] On the other hand, high lignin contents
promote char formation.[67] The char and bio-oil yields obtained
in the fast pyrolysis of pine wood, rice husk and orange peel are
consistent with the aforementioned influence of their macro-
molecular composition. Moreover, the role played by biomass
ashes should also be taken into account. Thus, alkaline and
alkaline earth metals catalyze the cracking of pyrolysis volatiles,
thus promoting gas formation at the expenses of decreasing
that of bio-oil.[68]
Finally, the microalgae showed a limited potential for
pyrolysis-reforming, with the bio-oil yield obtained being the
lowest, 46.6%. In spite of the fact that this bio-oil yield is higher
than those reported in algae pyrolysis in other technologies,[69]
the high ash content of the microalgae hindered bio-oil
production.
Figure 3 shows the yields of gaseous compounds obtained
in the pyrolysis of different biomasses at 500 °C. The pyrolysis of
lignocellulosic biomasses (pine wood and rice husk) led to
similar compositions of the non-condensable gaseous products.
Thus, the gaseous fraction was made up of a mixture of CO and
CO2, with a slightly higher amount of the former than the latter,
and low concentrations of hydrocarbons and H2, which is
consistent with previous literature results.[70] However, in the
case of the microalgae and, especially, the orange peel, the CO2
contents are much higher than those of CO. The remarkable
CO2formation in the orange waste pyrolysis has been
associated with pectin thermal degradation.[65,71] Moreover, in
the case of the microalgae, CO2release takes place by the
degradation of carbonyl and carboxyl groups of proteins and
carbohydrates.[72]
There are significant differences in the bio-oil nature derived
from different biomasses, as observed in Figure 4. Thus, bio-oil
composition has a remarkable effect on the reforming step
performance due to the different reforming reactivity and
deactivation rate associated with different bio-oil chemical
families.[39] The composition of bio-oil is directly related to the
biomass components, i.e., lignin, cellulose, proteins etc. Thus, in
the case of pine wood sawdust and rice husk, phenols are the
prevailing compounds (16.49 and 9.00 wt%, respectively),
which is associated with the high lignin content of these
biomasses.[62] In steam reforming reactions, phenolic com-
pounds have been reported to have lower reactivity compared
to other oxygenated compounds.[35,73,74] The degradation of
hemicellulose and cellulose contributes to the formation of
furans, ketones, and aldehydes (the former), and saccharides
and aldehydes (the latter).[53,67] Besides, the presence of nitrogen
compounds in the rice husk derived bio-oil is noteworthy, with
the yield of CxHyNzO compounds being 1.45 wt%. In the case of
orange peel, furans (particularly furfural) are the most important
fraction in the bio-oil (11.79 wt %), whose formation is related
to pectin decomposition.[63] Despite furfural has high reactivity
in reforming reactions,[75] this compound is prone to coke
formation, leading to fast catalyst deactivation.[35,39] The pres-
ence of ketones in the bio-oil derived from orange peel is also
remarkable (7.70 wt %), and their origin is diverse, as they are
formed via decomposition reactions involving several types of
oxygenates. Similarly to rice husk, a small fraction of N-
containing compounds is observed in the orange peel pyrolysis
Figure 3. Yields of the gaseous compounds obtained for the biomasses
studied in the fast pyrolysis at 500 °C.
Figure 4. Yields of the identified bio-oil compounds grouped into chemical
families for the biomasses studied in the fast pyrolysis at 500 °C.
Wiley VCH Dienstag, 15.10.2024
2420 / 357252 [S. 308/314] 1
ChemSusChem 2024,17, e202400325 (6 of 12) © 2024 The Authors. ChemSusChem published by Wiley-VCH GmbH
ChemSusChem
Research Article
doi.org/10.1002/cssc.202400325
oil (0.28 wt %), with pyridine (C5H5N) being the only compound
contained in this sample.
The difference in the nature of the bio-oil obtained in the
microalgae pyrolysis is due to the composition of this biomass.
Thus, the lack of lignin and polysaccharides like cellulose and
hemicellulose in the microalgae avoids the formation of
phenols, saccharides and furans. However, the decomposition
of proteins and lipids leads to significant yields of hydrocarbons
(2.78 wt%) and N-containing compounds (2.79 wt %),
respectively.[46] The latter are mainly nitrile, amine, amide, azole,
indole and nitrogen heterocyclic compounds, with CxHyNzand
CxHyNzO yields being 1.59 and 1.19 wt %, respectively.
In all the cases, the yield of water was in the 20 to 25 wt%
range, which comes from, on the one hand, the biomass
original moisture content and, on the other hand, the pyrolysis
reactions via dehydration of cellulose and hemicelluloses and
other intermediate components.[66,76]
The biochar obtained in the biomass pyrolysis-reforming
process is a valuable final product, which is removed from the
pyrolysis reactor, and therefore not fed into the reforming step.
In fact, the biochar has been successfully applied for soil
remediation, carbon sequestration, composting, wastewater
treatment, catalyst or catalyst support and electrodes.[77] Biochar
valorization and its applications depend on its specific features.
Table 3 shows proximate and ultimate analyses of the biochars
obtained from different biomasses. The properties of the
biochars are mainly conditioned by the original biomass ash
content. Thus, the high ash content of the microalgae led to a
biochar with an ash content close to 46 %, and therefore very
low carbon content and heating value (8.3 MJ kg1). The high
SiO2content of the rice husk is responsible for the poor heating
value of the biochar produced from this biomass. However,
those obtained by pine wood and orange peel pyrolysis have
much higher carbon contents and heating values, and therefore
better perspectives for energy production and other alternative
upgrading routes.
2.2. Pyrolysis Oxidative Reforming of Biomasses of Different
Composition
This section deals with the influence of the composition of the
biomass pyrolysis volatile stream on the oxidative steam
reforming step. The experiments were carried out at 600 °C,
using an ER of 0.12 and catalyst space times of 7.5 and
20 gcat mingvol1, which have been established based on a
previous study.[29] The mass of the reforming catalyst used in
the bed was different depending on the type of biomass, since
space time was defined by mass unit of the pyrolysis volatile
stream (gases+bio-oil) in order to make easier comparison of
the obtained results.
Figure 5 shows the conversion obtained in the oxidative
steam reforming of the pyrolysis volatiles from different
biomasses with the two space times studied. As observed, an
increase in space time enhanced reforming reactions and
improved the conversion of bio-oil into gaseous products. The
same qualitative effect has been reported in the reforming of
other biomasses derived volatile streams.[29,78] Considerable
differences are observed in the results obtained for the
biomasses studied. Thus, the conversions attained with pine
wood, rice husk and orange peel are similar, whereas the
reforming efficiency was very poor with the microalgae. Thus,
almost full conversion was attained (>99.7 %) with a space time
of 20 gcat min gvol1in the reforming of pine wood, rice husk and
orange peel pyrolysis volatiles. However, the conversion
attained in the OSR of the microalgae pyrolysis stream was
remarkably lower (66.8%), evidencing the high influence
volatile stream composition has on the reforming step.
Table 3. Proximate and ultimate analyses of the pyrolysis biochars
obtained at 500°C.
Feedstock Pine
wood
Rice
husk
Orange
waste
Microalgae
Ultimate analysis (wt %)a
Carbon 82.7 45.2 71.8 39.1
Hydrogen 2.9 1.5 2.9 2.4
Nitrogen 0.1 0.4 1.5 6.2
Oxygenb11.4 1.7 13.9 6.3
Proximate analysis (wt %)c
Volatile
matter
23.5 12.8 18.9 7.7
Fixed carbon 73.6 36.0 71.2 46.3
Ash 2.9 51.2 9.9 46.0
HHV (MJ kg1)30.4 14.4 26.8 8.3
aWet basis, bby difference, con an air-dried basis.
Figure 5. Conversions obtained in the pyrolysis-OSR for the biomasses
studied at 600°C, with ER of 0.12 and catalyst space times of 7.5 and
20 gcat mingvol1.
Wiley VCH Dienstag, 15.10.2024
2420 / 357252 [S. 309/314] 1
ChemSusChem 2024,17, e202400325 (7 of 12) © 2024 The Authors. ChemSusChem published by Wiley-VCH GmbH
ChemSusChem
Research Article
doi.org/10.1002/cssc.202400325
The results obtained with a low space time,
7.5 gcat mingvol1, allow evaluating the reactivity of the biomass
pyrolysis volatile stream, as these experiments correspond to
kinetic control conditions. In this respect, the highest con-
version (and so reactivity) was observed in the reforming of the
volatile stream obtained in the pyrolysis of pine wood, which
was of 96.4%, followed by those obtained for rice husk and
orange peel, which were of 95.8 and 92.6 %, respectively.
Therefore, lignocellulosic biomasses like pinewood and rice
husk led to similar values of initial volatile conversion, whereas
the pyrolysis-OSR of orange waste resulted in lower conversion
values due to the different composition of the volatile stream
from its pyrolysis. Thus, the high yields of furans (11.79 wt %)
and N-containing compounds, consisting the latter exclusively
of CxHyNzcompounds (pyridine), are responsible for the lower
initial activity in the process. In the case of the microalgae,
conversion only reached 50.9%, which can be partly attributed
to the presence of highly refractory compounds in the volatile
stream, such as the hydrocarbon fraction (2.78 wt %), which led
to a lower initial catalyst activity.[36] However, the results
obtained in the OSR of the microalgae pyrolysis cannot be
related solely to the lower reactivity of the pyrolysis-derived
volatiles, but a very fast deactivation also occurs in the
reforming process. Thus, the presence of N-containing com-
pounds in the volatile stream led to a fast catalyst poisoning of
Ni0active sites and so fast catalyst activity decay. In fact, NOx
precursors, namely, NH3, HCN and HNCO, are formed from N-
pyrolysis compounds, with their distribution depending on
both biomass composition and operating conditions in the
process.[79,80] Within this scenario, Ni-based catalysts have been
reported to effectively remove HCN by hydrogenation (Eq. (6))
to form NH3and its subsequent decomposition into N2and H2
(Eq. (7)):[81,82]
HCN þ3H2!NH3þCH4(6)
2NH3!N2þ3H2(7)
However, the activity of the reforming catalyst is hampered
due to the competition of the pyrolysis volatiles involved in
reforming reactions and the NH3in the decomposition reaction
in the coverage of Ni active sites.[79] Thus, NH3adsorbs on the Ni
surface, leading to fast catalyst activity decay for reforming.[83]
The product yields obtained in the reforming of the studied
biomasses are shown in Figure 6. An increase in space time
from 7.5 to 20 gcat mingvol1promoted the reforming of bio-oil
oxygenates (Eq. (8)) and hydrocarbons (Eq. (9)), as well as the
WGS reaction (Eq. (10)). Accordingly, there is an increase in H2
and CO2yields and a decrease in those of bio-oil, CO and
hydrocarbons. Thus, in the experiment performed with a space
time of 7.5 gcat mingvol1, a partial conversion of bio-oil, CH4and
light hydrocarbons was observed, but these products were
almost completely converted when space time was increased to
20 gcat mingvol1, with the sole exception of the microalgae,
whose bio-oil conversion is far from being full.
Oxygenates reforming:
Oxygenates þH2O!CO þH2(8)
Methane steam reforming:
CH4þH2O$CO þ3H2(9)
Water gas shift (WGS):
CO þH2O$CO2þH2(10)
Figure 7 shows the gas product composition obtained for
the biomasses studied when space times of 7.5 and
20 gcat mingvol1were used. In spite of the great differences
reported in the product yields, the gas composition obtained
Figure 6. Product yields obtained in the pyrolysis-OSR of the biomasses studied at 600 °C, with ER of 0.12 and space times of 7.5 and 20 gcatmin gvol1.
Wiley VCH Dienstag, 15.10.2024
2420 / 357252 [S. 310/314] 1
ChemSusChem 2024,17, e202400325 (8 of 12) © 2024 The Authors. ChemSusChem published by Wiley-VCH GmbH
ChemSusChem
Research Article
doi.org/10.1002/cssc.202400325
with all the biomasses is similar when operating with the
highest space time value. This result is related to the fact that
the gas obtained is mainly formed in reforming reactions and
its composition is not conditioned by the low extension of bio-
oil conversion, as is the case in the OSR of the microalgae
derived volatile stream. In all cases, the H2content is of around
60 vol%, which is slightly lower than that reported for the
conventional process of biomass steam reforming due to the
oxidation of H2(Eq. 11) under OSR conditions. Thus, the H2
concentration values reported for biomass pyrolysis-steam
reforming under optimum conditions range from 60 to
65 vol%.[21,24,36,84,85]
Hydrogen combustion:
H2þ1
2O2!H2O(11)
The main goal of the pyrolysis-reforming process is the
conversion of biomass into H2rich gas. Table 4 compares the
main reaction indices for the different biomasses studied
operating with a space time of 20 gcat mingvol1. As observed,
there are significant differences in the H2production potential
of the biomasses. In the case of the microalgae, the lower H2
production is explained by the low bio-oil yield in the pyrolysis
step (46.6 wt %) and non-full conversion in the subsequent
reforming one (66.8%). However, for pine wood, rice husk and
orange waste, almost full conversion of pyrolysis volatiles was
reached, i.e., bio-oil content in the outlet stream was negligible
in all cases. Therefore, the variations in H2production are
associated with the differences in the volatile stream formed in
the pyrolysis step. In fact, the results observed in H2production
are as follows: pine wood (9.3 wt %) >rice husk (7.7 wt %) >
orange waste (5.5 wt %) >microalgae (3.3 wt %), with this trend
being closely related to that of bio-oil production in the
pyrolysis step (Figure 2). Thus, the bio-oil fed into the reforming
reactor is converted into H2by means of steam reforming
(Eqs. (8, 9)) and WGS (Eq. (10)) reactions. Accordingly, the
conversion of bio-oil in the reforming step is related to the
mass of steam reacted and the specific gas production, see
Table 4. More specifically, the mass of gas produced accounts
for the reforming of the carbon moles contained in the bio-oil,
the conversion of steam and incorporation of products into the
gaseous stream (H2, CO an CO2), which is especially evident in
the case of pine wood (130.8 g 100 g1biomass) and rice husk
(115.6 g 100 g1biomass). Similar trends were obtained in the
pyrolysis and in line conventional steam reforming of similar
biomasses.[81] Moreover, in order to avoid the influence of
moisture and ash content, the H2productions reported in
Table 4 correspond to dry and ash free (daf) biomasses. In this
case, the H2production of rice husk is closer to that of pine
wood (8.9 and 10.2 wt %, respectively). In addition, H2produc-
tions on a daf basis reveal that the poor results obtained by
Figure 7. Composition of the gaseous streams obtained in the pyrolysis-OSR of the biomasses studied at 600 °C, with ER of 0.12 and space times of 7.5 and
20 gcat mingvol1.
Table 4. Reaction indices obtained in the pyrolysis-OSR of the biomasses
studied at 600°C, with ER of 0.12 and space time of 20 gcatmin gvol1.
Pine
wood
Rice
husk
Orange
waste
Microalgae
H2concentration (vol%) 61.8 60.3 60.3 61.8
H2production (wt%) 9.3 7.7 5.5 3.3
H2production daf (wt %)a10.2 8.9 5.7 4.2
Gas production (m3gas
kg1biomass)
1.7 1.4 1.0 0.6
Gas yield (g 100g1biomass) 130.8 115.6 82.3 44.8
Bio-oil yield (g
100g1biomass)
0.0 0.2 0.2 19.1
Reacted water (g
100g1biomass)
42.3 35.4 8.6 2.7
aby mass unit of dry and ash free (daf) biomasses.
Wiley VCH Dienstag, 15.10.2024
2420 / 357252 [S. 311/314] 1
ChemSusChem 2024,17, e202400325 (9 of 12) © 2024 The Authors. ChemSusChem published by Wiley-VCH GmbH
ChemSusChem
Research Article
doi.org/10.1002/cssc.202400325
orange waste P-OSR are related to the low bio-oil yield
obtained in the pyrolysis step and not to the moisture and ash
content of this biomass. Therefore, these results are clear
evidence that key point in the P-OSR process efficiency is the
biomass macromolecular composition. Thus, the concentration
of cellulose, hemicellulose, lignin and other components
determines biomass conversion to bio-oil in the pyrolysis step,
as well as in the subsequent H2production in the OSR reactor.
The H2productions obtained in the biomass pyrolysis-OSR
are below than those reported under conventional reforming
conditions due to the partial H2combustion (Eq. (11)). Thus,
Arregi et al.[50] obtained a H2production of 11.2 wt % with the
same pine wood in a similar pyrolysis-reforming unit and
working under steam reforming conditions (ER =0). However,
H2production decreased to 9.3 % in this study operating with
an ER of 0.12. Likewise, other pyrolysis and in line steam
reforming studies in different reaction technologies and
performed with low ash content biomasses also reported H2
productions above 10 wt%.[21,22,24,86,87] It should be noted that
there is no previous reference dealing with biomass P-OSR
process, as it requires operation to be conducted in continuous
regime and the previous studies were carried out mainly in
batch conditions.
In spite of the reduction in H2production compared to the
conventional SR, operation under OSR is an interesting
alternative for the large scale implementation of the two step
pyrolysis-reforming process. Thus, the P-OSR process has great
operational advantages, such as the feasibility to reach
autothermal regime in the reforming reactor or the attenuation
of catalyst deactivation. Moreover, this process has proven to
have great flexibility to treat diverse biomasses, even though
the poor performance obtained for microalgae should be
further investigated.
3. Conclusions
A continuous process is proposed for the direct conversion of
several biomass wastes into H2. The process combines biomass
fast pyrolysis in a CSBR with in line oxidative steam reforming
of the pyrolysis volatiles in a FBR. The pyrolysis step was carried
out at 500°C, as it is the optimum temperature for bio-oil
maximization. The reforming process conditions were based on
previous studies; temperature of 600°C, S/B ratio of 3 and
catalyst space times of 7.5 and 20 gcat min gvol1. The flexibility of
the process was evaluated by feeding biomasses of diverse
composition, and significant differences in H2production were
observed depending on the biomass in the feed. The behaviour
of each biomass in the pyrolysis process was analyzed in detail,
which allowed proving that biomass macromolecular composi-
tion and ash content greatly influenced the efficiency of
biomass conversion into bio-oil. In fact, pine wood led to high
bio-oil yields (75.3 wt%) due to the high content of cellulose
and hemicellulose and low ash content. The bio-oil yield
obtained in the case of rice husk decreased to 69.0 wt% due to
the high ash content of this biomass. The lower bio-oil
productions obtained in the cases of orange waste (54.9 wt%)
and microalgae (46.6 wt%) are due to their high pectin and ash
content, respectively. Furthermore, the reactivity of the volatile
stream also affected the performance of the oxidative reforming
step. Thus, the pyrolysis volatile streams obtained with pine
wood, rice husk and orange waste were efficiently reformed,
with conversion values above 92%. However, the one obtained
in the microalgae pyrolysis showed low reactivity. The high
content of N-compounds led to the formation of NOx
precursors, which compete with the reforming reaction by their
adsorption on Ni0sites. Thus, a severe catalyst deactivation by
poisoning of Ni0active sites was observed in the microalgae
pyrolysis-oxidative reforming. Accordingly, significant differ-
ences in H2production were observed depending on the
biomass, when they were processed with a space time of
20 gcat mingvol1. Pine wood and rice husk led to the higher H2
yields, 9.3 and 7.7 wt %, respectively. The low bio-oil yield
obtained with orange waste limited the H2production to
5.5 wt%. In the case of the microalgae, the low bio-oil yield
combined with its non-full conversion led to the lowest H2
production of 3.3 wt %.
The results obtained are clear evidence of the interest and
potential of the two-step process of pyrolysis-oxidative steam
reforming for H2production from biomass. Furthermore,
biomass source and features are aspects to be also considered
as they have great influence on the process performance.
Acknowledgements
This work was carried out with the financial support of the
grants PID2022-140704OB-I0 and PID2022-139454OB-I00
funded by MICIU/AEI/10.13039/501100011033 and by “ERDF/
EU”, the grants TED2021-132056B-I00, PLEC2021-008062 and
CNS2023-144031 funded by MICIU/AEI/10.13039/501100011033
and by the “European Union NextGenerationEU/PRTR”, and the
grants IT1645-22 and KK-2023/00060 funded by the Basque
Government.
Conflict of Interests
The authors declare no conflict of interest.
Data Availability Statement
The data that support the findings of this study are available
from the corresponding author upon reasonable request.
Keywords: Pyrolysis ·Hydrogen ·Steam reforming ·Catalyst ·
Biomass
[1] M. Shahbaz, T. Al-Ansari, M. Aslam, Z. Khan, A. Inayat, M. Athar, S. R.
Naqvi, M. A. Ahmed, G. McKay, Int. J. Hydrogen Energy 2020,45, 15166–
15195, 10.1016/j.ijhydene.2020.04.009.
[2] D. B. Pal, A. Singh, A. Bhatnagar, Int. J. Hydrogen Energy 2022,47, 1461–
1480, 10.1016/j.ijhydene.2021.10.124.
Wiley VCH Dienstag, 15.10.2024
2420 / 357252 [S. 312/314] 1
ChemSusChem 2024,17, e202400325 (10 of 12) © 2024 The Authors. ChemSusChem published by Wiley-VCH GmbH
ChemSusChem
Research Article
doi.org/10.1002/cssc.202400325
[3] B. Pandey, Y. K. Prajapati, P. N. Sheth, Int. J. Hydrogen Energy 2019,44,
25384–25415, 10.1016/j.ijhydene.2019.08.031.
[4] N. J. Rubinsin, N. A. Karim, S. N. Timmiati, K. L. Lim, Isahak, Wan
Nor Roslam Wan, M. Pudukudy, Int. J. Hydrogen Energy 2024,49, 1139–
1164, 10.1016/j.ijhydene.2023.09.043.
[5] A. Molino, S. Chianese, D. Musmarra, J. Energy Chem. 2016,25, 10–25,
10.1016/j.jechem.2015.11.005.
[6] S. Rapagnà, K. Gallucci, P. U. Foscolo, Waste Manage. 2018,71, 792–800,
10.1016/j.wasman.2017.07.038.
[7] L. Wei, S. Xu, L. Zhang, C. Liu, H. Zhu, S. Liu, Int. J. Hydrogen Energy
2007,32, 24–31, 10.1016/j.ijhydene.2006.06.002.
[8] S. Fremaux, S. Beheshti, H. Ghassemi, R. Shahsavan-Markadeh, Energy
Convers. Manage. 2015,91, 427–432, 10.1016/j.enconman.2014.12.048.
[9] M. Cortazar, G. Lopez, J. Alvarez, M. Amutio, J. Bilbao, M. Olazar, Energy
2018,153, 455–463, 10.1016/j.energy.2018.04.067.
[10] M. Cortazar, L. Santamaria, G. Lopez, J. Alvarez, L. Zhang, R. Wang, X. Bi,
M. Olazar, Energy Convers. Manage. 2023,276, 116496, 10.1016/j.encon-
man.2022.116496.
[11] M. L. Valderrama Rios, A. M. González, E. E. S. Lora, Almazán del Olmo,
Oscar Agustin, Biomass Bioenergy 2018,108, 345–370, 10.1016/j.
biombioe.2017.12.002.
[12] V. Claude, C. Courson, M. Köhler, S. D. Lambert, Energy Fuels 2016,30,
8791–8814, 10.1021/acs.energyfuels.6b01642.
[13] G. Lopez, L. Santamaria, A. Lemonidou, S. Zhang, C. Wu, A. T. Sipra, N.
Gao, Nat. Rev. Methods Primers 2022,2, 20, 10.1038/s43586-022-00097-
8.
[14] W. Nabgan, T. A. Tuan Abdullah, R. Mat, B. Nabgan, Y. Gambo, M.
Ibrahim, A. Ahmad, A. A. Jalil, S. Triwahyono, I. Saeh, Renewable
Sustainable Energy Rev. 2017,79, 347–357, 10.1016/j.rser.2017.05.069.
[15] R. Trane, S. Dahl, M. S. Skjoth-Rasmussen, A. D. Jensen, Int. J. Hydrogen
Energy 2012,37, 6447–6472, 10.1016/j.ijhydene.2012.01.023.
[16] J. Chen, J. Sun, Y. Wang, Ind. Eng. Chem. Res. 2017,56, 4627–4637, 10.
1021/acs.iecr.7b00600.
[17] F. Bimbela, J. Ábrego, R. Puerta, L. García, J. Arauzo, Appl. Catal. B 2017,
209, 346–357, 10.1016/j.apcatb.2017.03.009.
[18] A. Remiro, A. Arandia, L. Oar-Arteta, J. Bilbao, A. G. Gayubo, Appl. Catal.
B2018,237, 353–365, 10.1016/j.apcatb.2018.06.005.
[19] D. Yao, C. Wu, H. Yang, Q. Hu, M. A. Nahil, H. Chen, P. T. Williams, Int. J.
Hydrogen Energy 2014,39, 14642–14652, 10.1016/j.ijhydene.2014.07.
077.
[20] S. Czernik, R. French, C. Feik, E. Chornet, Ind. Eng. Chem. Res. 2002,41,
4209, 10.1021/ie020107q.
[21] X. Xiao, X. Meng, D. D. Le, T. Takarada, Bioresour. Technol. 2011,102,
1975–1981, 10.1016/j.biortech.2010.09.016.
[22] J. P. Cao, P. Shi, X. Y. Zhao, X. Y. Wei, T. Takarada, Fuel Process. Technol.
2014,123, 34–40, 10.1016/j.fuproc.2014.01.042.
[23] L. Santamaria, G. Lopez, A. Arregi, M. Artetxe, M. Amutio, J. Bilbao, M.
Olazar, J. Ind. Eng. Chem. 2020,91, 167–181, 10.1016/j.jiec.2020.07.050.
[24] C. Gai, N. Zhu, S. K. Hoekman, Z. Liu, W. Jiao, N. Peng, Energy Convers.
Manage. 2019,183, 474–484, 10.1016/j.enconman.2018.12.121.
[25] H. Yu, Y. Liu, J. Liu, D. Chen, Fuel 2019,254, 115622, 10.1016/
j.fuel.2019.115622.
[26] L. Santamaria, A. Arregi, J. Alvarez, M. Artetxe, M. Amutio, G. Lopez, J.
Bilbao, M. Olazar, J. Anal. Appl. Pyrolysis 2018,136, 222–231, 10.1016/j.
jaap.2018.09.025.
[27] G. Guan, M. Kaewpanha, X. Hao, A. Abudula, Renewable Sustainable
Energy Rev. 2016 58, 450–461, 10.1016/j.rser.2015.12.316.
[28] S. Ayalur Chattanathan, S. Adhikari, N. Abdoulmoumine, Renewable
Sustainable Energy Rev. 2012,16, 2366–2372, 10.1016/j.rser.2012.01.051.
[29] I. Garcia, L. Santamaria, G. Lopez, J. Bilbao, M. Olazar, M. Amutio, M.
Artetxe, Chem. Eng. J. 2023,475, 146223, 10.1016/j.cej.2023.146223.
[30] E. J. Leijenhorst, W. Wolters, L. Van de Beld, W. Prins, Fuel Process.
Technol. 2013,115, 164–173, 10.1016/j.fuproc.2013.05.015.
[31] A. Remiro, A. Arandia, L. Oar-Arteta, J. Bilbao, A. G. Gayubo, Energy Fuels
2018,32, 3588–3598, 10.1021/acs.energyfuels.7b04141.
[32] H. I. Villafán-Vidales, G. Lopez, L. Santamaria, M. Artetxe, M. Olazar, J.
Anal. Appl. Pyrolysis 2022,163, 105482, 10.1016/j.jaap.2022.105482.
[33] W. Xie, J. Yang, Q. Wang, L. Huang, N. Wang, Catal. Sci. Technol. 2018,8,
3015–3024, 10.1039/C8CY00116B.
[34] Q. Zhou, X. Zhong, X. Xie, X. Jia, B. Chen, N. Wang, L. Huang, Renewable
Energy 2020,145, 2316–2326, 10.1016/j.renene.2019.07.078.
[35] R. Trane-Restrup, A. D. Jensen, Appl. Catal. B 2015,165, 117–127, 10.
1016/j.apcatb.2014.09.026.
[36] E. Fernandez, L. Santamaria, M. Artetxe, M. Amutio, A. Arregi, G. Lopez,
J. Bilbao, M. Olazar, Fuel 2022,312, 122910, 10.1016/j.fuel.2021.122910.
[37] E. Fernandez, L. Santamaria, I. García, M. Amutio, M. Artetxe, G. Lopez, J.
Bilbao, M. Olazar, Chin. J. Catal. 2023,48, 101–116, 10.1016/S1872-
2067(23)64407-9.
[38] C. Rioche, S. Kulkarni, F. C. Meunier, J. P. Breen, R. Burch, Appl. Catal. B
2005,61, 130–139, 10.1016/j.apcatb.2005.04.015.
[39] J. Remón, F. Broust, G. Volle, L. García, J. Arauzo, Int. J. Hydrogen Energy
2015,40, 5593–5608, 10.1016/j.ijhydene.2015.02.117.
[40] W. Cai, Z. Luo, J. Zhou, Q. Wang, Fuel Process. Technol. 2021,221,
106919, 10.1016/j.fuproc.2021.106919.
[41] S. Singh, K. K. Pant, M. Krishania, J. Anal. Appl. Pyrolysis 2023,175,
106160, 10.1016/j.jaap.2023.106160.
[42] G. Ningbo, L. Baoling, L. Aimin, L. Juanjuan, J. Anal. Appl. Pyrolysis 2015,
114, 155–162, 10.1016/j.jaap.2015.05.011.
[43] R. E. Guedes, A. S. Luna, A. R. Torres, J. Anal. Appl. Pyrolysis 2018,129,
134–149, 10.1016/j.jaap.2017.11.019.
[44] B. Tian, L. Xu, M. Jing, N. Liu, Y. Tian, Fuel Process. Technol. 2021,214,
106715, 10.1016/j.fuproc.2020.106715.
[45] J. Alvarez, G. Lopez, M. Amutio, J. Bilbao, M. Olazar, Fuel 2014,128, 162–
169, 10.1016/j.fuel.2014.02.074.
[46] K. Azizi, M. Keshavarz Moraveji, H. Abedini Najafabadi, Renewable
Sustainable Energy Rev. 2018,82, 3046–3059, 10.1016/j.rser.2017.10.033.
[47] D. Aboelela, H. Saleh, A. M. Attia, Y. Elhenawy, T. Majozi, M. Bassyouni,
Sustainability 2023,15, 10.3390/su151411238.
[48] A. V. Bridgwater, Biomass Bioenergy 2012,38, 68–94, 10.1016/j.bio-
mbioe.2011.01.048.
[49] M. Kordi, N. Farrokhi, M. I. Pech-Canul, A. Ahmadikhah, Rice Sci. 2023,
10.1016/j.rsci.2023.08.005.
[50] A. Arregi, G. Lopez, M. Amutio, I. Barbarias, L. Santamaria, J. Bilbao, M.
Olazar, J. Ind. Eng. Chem. 2018,68, 69–78, 10.1016/j.jiec.2018.07.030.
[51] I. Barbarias, G. Lopez, J. Alvarez, M. Artetxe, A. Arregi, J. Bilbao, M.
Olazar, Chem. Eng. J. 2016,296, 191–198, 10.1016/j.cej.2016.03.091.
[52] G. Lopez, A. Erkiaga, M. Artetxe, M. Amutio, J. Bilbao, M. Olazar, Ind. Eng.
Chem. Res. 2015,54, 9536–9544, 10.1021/acs.iecr.5b02413.
[53] M. Amutio, G. Lopez, J. Alvarez, R. Moreira, G. Duarte, J. Nunes, M.
Olazar, J. Bilbao, Bioresour. Technol. 2013,129, 512–518, 10.1016/j.
biortech.2012.11.114.
[54] K. Azizi, M. Keshavarz Moraveji, A. Arregi, M. Amutio, G. Lopez, M.
Olazar, Bioresour. Technol. 2020,311, 123561, 10.1016/j.bio-
rtech.2020.123561.
[55] S. Orozco, M. Artetxe, G. Lopez, M. Suarez, J. Bilbao, M. Olazar,
ChemSusChem 2021,14, 4291–4300, 10.1002/cssc.202100889.
[56] P. K. Mollick, A. B. Pandit, T. Mukherjee, P. K. Vijayan, Ind. Eng. Chem. Res.
2020,59, 3229–3237, 10.1021/acs.iecr.9b04035.
[57] P. K. Mollick, A. B. Pandit, P. K. Vijayan, Ind. Eng. Chem. Res. 2018,57,
8605–8611, 10.1021/acs.iecr.8b01691.
[58] J. Makibar, A. R. Fernandez-Akarregi, L. Diaz, G. Lopez, M. Olazar, Powder
Technol. 2012,219, 49–58, 10.1016/j.powtec.2011.12.008.
[59] H. Altzibar, G. Lopez, R. Aguado, S. Alvarez, M. J. San Jose, M. Olazar,
Chem. Eng. Technol. 2009,32, 463–469, 10.1002/ceat.200800605.
[60] E. Fernandez, M. Amutio, M. Artetxe, A. Arregi, L. Santamaria, G. Lopez,
J. Bilbao, M. Olazar, Process Saf. Environ. Prot. 2021,145, 52–62, 10.1016/
j.psep.2020.07.039.
[61] M. Cortazar, J. Alvarez, G. Lopez, M. Amutio, L. Santamaria, J. Bilbao, M.
Olazar, Energy Convers. Manage. 2018,171, 1589–1597, 10.1016/j.encon-
man.2018.06.071.
[62] M. Amutio, G. Lopez, M. Artetxe, G. Elordi, M. Olazar, J. Bilbao, Resour.
Conserv. Recycl. 2012,59, 23–31, 10.1016/j.resconrec.2011.04.002.
[63] J. Alvarez, B. Hooshdaran, M. Cortazar, M. Amutio, G. Lopez, F. B. Freire,
M. Haghshenasfard, S. H. Hosseini, M. Olazar, Fuel 2018,224, 111–120,
10.1016/j.fuel.2018.03.028.
[64] J. Aburto, M. Moran, A. Galano, E. Torres-García, J. Anal. Appl. Pyrolysis
2015,112, 94–104, 10.1016/j.jaap.2015.02.012.
[65] R. K. Sharma, J. B. Wooten, V. L. Baliga, M. R. Hajaligol, Fuel 2001,80,
1825–1836, 10.1016/S0016-2361(01)00066-7.
[66] F. Collard, J. Blin, Renewable Sustainable Energy Rev. 2014,38, 594–608,
10.1016/j.rser.2014.06.013.
[67] S. Wang, X. Guo, K. Wang, Z. Luo, J. Anal. Appl. Pyrolysis 2011,91, 183–
189, 10.1016/j.jaap.2011.02.006.
[68] A. Oasmaa, Y. Solantausta, V. Arpiainen, E. Kuoppala, K. Sipila, Energy
Fuels 2010,24, 1380–1388, 10.1021/ef901107f.
[69] C. Yang, R. Li, B. Zhang, Q. Qiu, B. Wang, H. Yang, Y. Ding, C. Wang, Fuel
Process. Technol. 2019,186, 53–72, 10.1016/j.fuproc.2018.12.012.
[70] T. Kan, V. Strezov, T. J. Evans, Renewable Sustainable Energy Rev. 2016,
57, 1126–1140, 10.1016/j.rser.2015.12.185.
Wiley VCH Dienstag, 15.10.2024
2420 / 357252 [S. 313/314] 1
ChemSusChem 2024,17, e202400325 (11 of 12) © 2024 The Authors. ChemSusChem published by Wiley-VCH GmbH
ChemSusChem
Research Article
doi.org/10.1002/cssc.202400325
[71] S. Zhou, Y. Xu, C. Wang, Z. Tian, J. Anal. Appl. Pyrolysis 2011,91, 232–
240, 10.1016/j.jaap.2011.02.015.
[72] W. Chen, H. Yang, Y. Chen, M. Xia, Z. Yang, X. Wang, H. Chen, Energy
2017,131, 1–12, 10.1016/j.energy.2017.05.019.
[73] S. Wang, Q. Cai, F. Zhang, X. Li, L. Zhang, Z. Luo, Int. J. Hydrogen Energy
2014,39, 18675–18687, 10.1016/j.ijhydene.2014.01.142.
[74] M. Artetxe, J. Alvarez, M. A. Nahil, M. Olazar, P. T. Williams, Energy
Convers. Manage. 2017,136, 119–126, 10.1016/j.enconman.2016.12.092.
[75] S. Wang, X. Li, F. Zhang, Q. Cai, Y. Wang, Z. Luo, Int. J. Hydrogen Energy
2013,38, 16038–16047, 10.1016/j.ijhydene.2013.10.032.
[76] A. Oasmaa, I. Fonts, M. R. Pelaez-Samaniego, M. E. Garcia-Perez, M.
Garcia-Perez, Energy Fuels 2016,30, 6179–6200, 10.1021/acs.energy-
fuels.6b01287.
[77] J. Wang, S. Wang, J. Cleaner Prod. 2019,227, 1002–1022, 10.1016/j.
jclepro.2019.04.282.
[78] A. Arandia, A. Remiro, L. Oar-Arteta, J. Bilbao, A. G. Gayubo, Int. J.
Hydrogen Energy 2017,42, 29175–29185, 10.1016/j.ijhydene.2017.10.
095.
[79] D. K. Binte Mohamed, A. Veksha, Q. L. M. Ha, W. P. Chan, T. Lim, G. Lisak,
Fuel 2022,318, 123602, 10.1016/j.fuel.2022.123602.
[80] H. Chen, R. Shan, F. Zhao, J. Gu, Y. Zhang, H. Yuan, Y. Chen, Chem. Eng.
J. 2023,451, 138979, 10.1016/j.cej.2022.138979.
[81] A. Arregi, L. Santamaria, G. Lopez, M. Olazar, J. Bilbao, M. Artetxe, M.
Amutio, J. Environ. Manage. 2023,347, 119071, 10.1016/j.jenv-
man.2023.119071.
[82] S. Kumagai, R. Yabuki, T. Kameda, Y. Saito, T. Yoshioka, Ind. Eng. Chem.
Res. 2020,59, 9023–9033, 10.1021/acs.iecr.0c00931.
[83] W. Yin, N. Guilhaume, Y. Schuurman, Chem. Eng. J. 2020,398, 125534,
10.1016/j.cej.2020.125534.
[84] D. Yao, Q. Hu, D. Wang, H. Yang, C. Wu, X. Wang, H. Chen, Bioresour.
Technol. 2016,216, 159–164, 10.1016/j.biortech.2016.05.011.
[85] Q. M. K. Waheed, C. Wu, P. T. Williams, J. Energy Inst. 2016,89, 657–667,
10.1016/j.joei.2015.05.006.
[86] B. Wang, J. Cao, X. Zhao, Y. Bian, C. Song, Y. Zhao, X. Fan, X. Wei, T.
Takarada, Fuel Process. Technol. 2015,136, 17–24, 10.1016/j.fuproc.2014.
07.024.
[87] J. Cao, T. Liu, J. Ren, X. Zhao, Y. Wu, J. Wang, X. Ren, X. Wei, J. Anal. Appl.
Pyrolysis 2017,127, 82–90, 10.1016/j.jaap.2017.08.020.
Manuscript received: February 19, 2024
Revised manuscript received: May 2, 2024
Accepted manuscript online: May 14, 2024
Version of record online: June 10, 2024
Wiley VCH Dienstag, 15.10.2024
2420 / 357252 [S. 314/314] 1
ChemSusChem 2024,17, e202400325 (12 of 12) © 2024 The Authors. ChemSusChem published by Wiley-VCH GmbH
ChemSusChem
Research Article
doi.org/10.1002/cssc.202400325
