Topics in Catalysis

https://doi.org/10.1007/s11244-022-01563-z

ORIGINAL PAPER

Catalytic Production of Renewable Hydrogen for Use in Fuel Cells:

A Review Study
Ilenia Rossetti1 · Antonio Tripodi1

Accepted: 31 December 2021

© The Author(s) 2022

Abstract
Hydrogen production from renewable sources is gaining increasing importance for application as fuel, in particular with
high efficiency and low impact devices such as fuel cells. In addition, the possibility to produce more sustainable hydrogen
for industrial application is also of interest for fundamental industrial processes, such as ammonia and methanol synthesis.
Catalytic processes are used in most options for the production of hydrogen from renewable sources. Catalysts are directly
involved in the main transformation, as in the case of reforming and of electro-/photo-catalytic water splitting, or in the
upgrade and refining of the main reaction products, as in the case of tar reforming. In every case, for the main processes that
reached a sufficiently mature development stage, attempts of process design, economic and environmental impact assessment
are presented, on one hand to finalise the demonstration of the technology, on the other hand to highlight the challenges and
bottlenecks. Selected examples are described, highlighting whenever possible the role of catalysis and the open issues, e.g.
for the ­H2 production from reforming, aqueous phase reforming, biomass pyrolysis and gasification, photo- and electro-
catalytic processes, enzymatic catalysis. The case history of hydrogen production from bioethanol for use in fuel cells is
detailed from the point of view of process design and techno-economic validation. Examples of steady state or dynamic
simulation of a centralised or distributed ­H2 production unit are presented to demonstrate the feasibility of this technology,
that appears as one of the nearest to market. The economic feasibility seems demonstrated when producing hydrogen start-
ing from diluted bioethanol.

Keywords Hydrogen production · Renewable sources · Fuel cells · Bioethanol · Reforming · Biomass gasification ·
Biomass pyrolysis

1 Introduction electrochemical devices that are characterised by higher

efficiency than internal combustion engines [2]. Indeed, the
Hydrogen is the most abundant element in the universe chemical to electric energy conversion may reach 60% by
and raised particular interest in recent times due to its high itself, while increasing efficiency is achieved when recover-
energy content per unit mass: its enthalpy of combustion ing also heat in Combined Heat & Power (CHP) co-genera-
is − 286 kJ/mol. Its direct combustion, though energy effi- tion plants [3]. These features are accompanied by the pro-
cient, is difficult due to hard thermal management for the duction of water as only reaction product, which adds further
very high adiabatic temperature raise, which imposes the advantages from the sustainability point of view [4, 5].
selection of appropriate materials for burners and the con- In spite of this, a key point is the need to obtain molecu-
trol of ­NOx emission, when combustion is carried out with lar hydrogen by transformation of hydrogen-rich substrates
air [1]. An alternative use is as fuel for fuel cells (FCs), available in nature. This step is not at all trivial, since it
affects the economic feasibility of the process and most
of all its real environmental impact. If it is true that H ­ 2
* Ilenia Rossetti conversion into FCs delivers only water, the synthesis of
ilenia.rossetti@unimi.it
hydrogen involves the use of raw materials that should be
1
Chemical Plants and Industrial Chemistry Group, Dip. extracted and transported (e.g. natural gas), harvested and
Chimica, CNR-SCITEC and INSTM Unit Milano- transformed (e.g. ethanol from the fermentation of biomass;
Università, Università degli Studi di Milano, via C. Golgi 19, biomass transportation to the point of its transformation),
20133 Milan, Italy

13
Vol.:(0123456789)
 Topics in Catalysis

etc. Irrespectively on the fossil or renewable source selected, temperatures in civilian buildings, allowing to design a CHP
every step needed for ­H2 production, purification, storage system suited for micro-scale distributed use.
and distribution, implies direct and indirect environmental,
social and economic impacts that must be carefully assessed
to select the most appropriate choices [6–8]. 2 Catalytic Hydrogen Production
More in general, the production of sustainable hydrogen from Renewable Raw Materials
is considered a promising solution to reduce the ­CO2 emis-
sions in the automotive and industrial sectors, especially 2.1 Current Processes for ­H2 Production
when considering efficient energy conversion systems as and Purification Suitable for Use in FCs
in the case of FCs. Furthermore, it opens new perspectives
for the use of many different renewable based materials. Currently, ca. 95% of the H ­ 2 produced is based on fossil
Research is focused mainly on two fields of application: the sources, accounting for ca. 55 million tons per year, increas-
production of hydrogen as energy vector and as chemical ing by ca. 6% annually [11, 12]. The steam reforming of
for industrial use. It is indeed the reactant for ammonia syn- natural gas accounts for ca. 50% of H
­ 2 production, reforming
thesis, a huge industrial process that uses most of the H ­ 2 of oil and naptha for ca. 30%, 18% is based on coal gasifica-
produced worldwide and accounts for ca. 3% of the global tion and ca. 4% on electrolysis of water [12].
energy consumption [9], stimulating the development of less Steam reforming (SR) occurs through the following
impacting solutions for ­H2 production. general reactions (R1) and (R2), exemplified for a general
In this work we consider different options for the produc- hydrocarbon [13]:
tion of ­H2 from renewable sources, exemplifying some cata-
lytic processes that are sufficiently mature and demonstrated Cm Hn + mH2 O(g) → mCO + (m + 0.5n)H2 (R1)
to be proposed in the near future for exploitation. At this
stage, the robust knowledge of kinetics, reaction mechanism Cm Hn + 2mH2 O(g) → mCO2 + (2m + 0.5n)H2 (R2)
and process design play a key role, and such topics should be
early developed whenever possible to assess the real feasibil- CO + H2 O ⇄ CO2 + H2 (R3)
ity of the solution proposed.
Case histories will be also discussed of a centralised which are connected through the water gas shift (WGS) reac-
­H2 production facility and of distributed CHP systems. A tion (R3) [14].
medium to large scale reforming plant can therefore be a Depending on the substrate, the reaction temperature is
solution to exploit bioethanol as a hydrogen source for cen- widely different, e.g. 180 °C for methanol, 400–600 °C for
tralised distribution facilities, e.g. a refuelling station. In different alcohols or oxygenate substrates, 700–900 °C for
addition, the production of electric energy via a fuel cell on hydrocarbons such as methane [15–17]. The catalysts are
board or for distributed CHP options, looks more suited to either based on noble metals (e.g. Pt or Rh) or on transition
small- and micro-scale applications. In both cases the use non-noble metals, the most studied being Ni. The latter is
of diluted bioethanol has been explored as a less expensive particularly successful since at very high temperature kinet-
feedstock than fuel-grade ethanol. Indeed, for application ics is not the limiting factor, especially when the effective-
in internal combustion engines ethanol must be used pure ness factor of pellettised catalysts is as low as 5% [18].
(when using dedicated fleets) or blended with gasoline (usu- The process is sufficiently energy efficient from methane
ally up to 10 vol% ethanol) [10]. Viceversa, in steam reform- (70–85%) and offers the highest ­H2/CO2 ratio. The inter-
ing a minimum 3:1 water:ethanol stoichiometric ratio must est is now turning towards more complex, and at the same
be used. In this view, the dehydration of ethanol from the time more sustainable substrates, but the limiting factor is
fermentation broth can be less intensive, with considerable a higher C­ O2 emission rate (up to 7–15 kg C ­ O2/kg ­H2), that
energy and cost savings and improvement of the environ- should be compensated in some way to become ­CO2 neutral
mental and economic sustainability of its use. [19].
Moving a step further on this route, it is observed that In order to limit the energy input to this endothermal
the water condensation (to purify the reformate) and the reaction, a variant is the autothermal reforming (ATR),
hydrogen reaction with air (in the fuel cell) release heat at which includes some feed of oxygen to partially oxidise the
temperatures below the foreseen pinch-points of the reform- substrate, supplying the heat needed. More compact ATR
ing process: one approximately at the boiling point of the reactors can be designed than for SR, though the gas purity
hydroalcoholic feed (cold feed—hot products exchange), and ­H2/CO2 ratio are more favourable for SR [20].
and the other at the catalyst activation temperature (react- All these processes lead to a considerable amount of CO
ing mixture—combustion gases exchange). Yet these sub- and ­CO2, that should be possibly removed before feeding
pinch heat loads are aligned with the typical sanitary water a FC. If a Polymer Electrolyte Membrane FC is used, the

13
Topics in Catalysis

Pt-based electrodes impose very careful removal of CO, with calcined dolomite can be used for gas cleaning and
down to few ppm. By contrast a Solid Oxide FC can be upgrading [29]. The economy of scale suggests very large
tolerant to high CO levels. ­CO2 is inert to FCs, but it dilutes installations, which however do not fit the supply rate of
the reformate (ca. 20–25 vol%) and can induce challenges biomass in most applications. The challenge is therefore
for hydraulics in the design and rating of the cell. Very lim- the development of small decentralised plants that are eco-
ited demonstration data are available for FCs operating with nomically convenient [30]. Furthermore, the conversion
reformate than with pure ­H2 [21]. of tar is a further issue, that suggests the subsequent cata-
Very well assessed processes exist for CO conversion and lytic upgrade by reforming [31–35]. Catalytic cracking of
­CO2 separation [14, 22]. The WGS reaction (R3) is conven- tar can be performed at temperature higher than 1000 °C
tionally operated in two stages, the former at high tempera- and dolomite or olivine as catalyst. Tar formation can be
ture (HT, i.e. 350–450 °C) over Fe–Cr catalysts to improve prevented by using catalysts such as Ni, Pt, Pd, Rh or alka-
kinetics and convert most CO. The second stage is carried line metal oxides supported over ­CeO2/SiO2 [36]. Another
out at lower temperature (180–250 °C) over Cu–ZnO cata- strategy is to favour gasification and WGS by selective
lysts, more active at lower temperature but more expensive removal of ­CO2 by sorption with Ca(OH) 2, obtained by
[23]. Current trends are oriented towards a single stage of high pressure hydration of CaO. C ­ aCO3 forms, which is
WGS at intermediate temperature, taking advantage of the then decomposed by calcination to accomplish C ­ O2 recov-
development of high temperature PEM FCs that can operate ery and regeneration of the sorbent [36].
at ca. 160 °C and thus can allow slightly higher CO content Pyrolysis is an alternative thermochemical process for
[24, 25]. biomass conversion [37], carried out in absence of oxygen.
To achieve few ppm CO concentration a further refining Depending on the residence time and reaction temperature,
is needed, either through preferential oxidation or by metha- biomass can be converted into different proportions of liq-
nation. The latter consumes some more H ­ 2, but avoids addi- uid (bio-oil), gaseous ­(H2-containing) and solid products.
tion of air. Also in this case noble metal catalysts (Pt, Ru, Rh This process can be co-adjuvated by catalysts, but the most
over alumina) or Ni-based ones are the most employed [26]. foreseen application is the further conversion of bio-oil into
As for the C ­ O2 removal technologies, wide experience ­H2 by steam reforming. Bio-oil is a mixture of organic com-
has been derived on the post treatment of combustion flue ponents and can be easily separated into an aqueous frac-
gas and commercial technologies, fitting different flow rates tion, containing water and water-soluble organics and a non
and ­CO2 concentration, include pressure swing adsorption water-soluble fraction. The aqueous fraction is particularly
(PSA) or scrubbing with high-pressure or basic aqueous suitable for steam reforming, whereas the latter can be val-
solutions [27]. orised by extraction of specialties and fine chemicals [28].
Steam reforming of the aqueous fraction of poplar-derived
2.2 Emerging Processes for ­H2 Production bio-oil led to 75% ­H2 yield at 750 °C, but fast deactivation of
from Biomass or Biomass Derived Substrates the catalyst was observed due to coking. Regeneration was
accomplished by steaming at 800 °C and high ­H2O/oil ratio
Alternatives are currently based on the valorisation of (ca. 7 mol/mol C), while low space velocity was suggested
renewable feedstocks, but each option carries different chal- to prevent further deactivation [38]. Other reports indicate
lenges and degree of maturity. As a general classification H ­ 2 ­H2 yield between 71 and 78% [39, 40] using platinum group
is defined as grey if derived from fossil fuels, blue if carbon metals supported over ­ZrO2, ­Al2O3 or ­CeO2–ZrO2.
capture and storage is applied, green if derived from renewa- As for gasification, also pyrolysis fits large scale produc-
bles, bio if derived by bio-production [11]. tion facilities and needs stronger efforts for validation, since
One option for green ­H2 is biomass gasification to pro- its use is less defined in refinery and its development much
duce syngas and then hydrogen. Biomass is heated and more recent. In addition, the need to cope with the logistics
loses the volatile components, leaving a C-rich char, which of large scale biomass transportation is a key limit [41]. An
is then gasified at 800–850 °C. The products are ­H2, CO interesting review on the opportunities of biomass pyroly-
and ­CO2 and the latter can be further interconverted by sis for hydrogen production is reported in [42]. Biomass
the WGS reaction (R3). Furthermore, a set of organic is rapidly heated without addition of oxygen and water at
molecules, condensate after the high temperature reactor intermediate temperature (375–525 °C) forming a hydrogen-
zone, are collected as tar, which represents one of the main rich gas and a liquid fraction, called bio-oil. The latter can
technological issues. The reaction is carried out cofeed- be used as substrate for steam reforming, but challenges are
ing ­O2, air or steam, possibly using a catalyst in a fixed or related to the optimisation of the catalytic materials for this
fluidised bed reactor [12, 28]. The thermal efficiency is not application, which are currently based on Ni as active phase.
very high, 35–50%, especially when moisture in biomass One of the major issues is the improvement of the effective-
is significant. A second step with a Ni-based catalyst or ness and attrition resistance for application in large scale

13
 Topics in Catalysis

fluidised beds. Other interesting applications are envisaged [46]. As for non-noble metal formulations, transition metal
for the valorisation of the water-soluble fraction of bio-oil. carbides, chalcogenides or phosphides are interesting for
Aqueous phase reforming (APR) is a new option spe- HER, e.g. M ­ o2C or WC due to high electrical conductivity,
cifically developed for biomass valorization into H ­ 2 when H-adsorption and d-band density state [47]. As for OER,
aqueous solutions containing oxygenated hydrocarbons or two different mechanisms are discussed in the literature,
carbohydrates are available (better if waste substrates) [43]. conventional adsorbate evolution or oxygen-mediated. For
Low temperature is applied (up to 270 °C) under autogenous the former, four concerted transfers of protons and electrons
pressure to keep a liquid phase (up to 50 bar). The low tem- occur at the metal active site, producing an oxygen molecule
perature limits CO amount in the outflowing gas and water either in acidic or basic conditions. The latter mechanism is
evaporation does not occur, with considerable energy saving instead mediated by oxygen transfer into the lattice of the
for the heat input to the reactor. Also in this case Ni-based catalyst. This mechanism is particularly suitable for materi-
or noble metal-based (mainly Pt) catalysts are used, but the als rich of oxygen vacancies and with high ionic conductiv-
productivity and selectivity are still not optimal [12]. Cleav- ity, such as ­La1−xSrxCoO3−δ perovskites [46]. Suitable OER
age of C–C bonds can occur over Pt, Pd, Rh, Sn, Ni, Co, Cu, catalysts are Ru- and Ir-based oxides (also in multi-metal
Zn and various bi-metallic formulations [44], with activity configuration) in acidic electrolyte or Ni–Fe or Fe–Co–W
in the order Pt ≅ Ni > Ru > Rh ≅ Pd > Ir. Ni/SiO2 showed sig- in alkaline environment.
nificant deactivation, while Pt and Pd based catalysts exhibit Solar energy can be directly stored in form of hydrogen
low selectivity for C–O bond cleavage. As a compromise, by using semiconductors as photocatalysts. The water split-
Pt was the best choice, but the cost of the material makes it ting reaction is at the basis of the process, but given the slow
unsuitable for large scale applications [36]. kinetics of oxygen oxidation, faster alternative half-reactions
The substrate solutions for the APR can be obtained can be used, such as the oxidation of sacrificial agents. If
e.g. by hydrolysis of cellulosic biomass, that can be per- waste organic molecules already present in water are used
formed under acidic conditions to produce sugars. The for this purpose, the process is called photoreforming and
effectiveness is related to the pKa of the acid employed, e.g. becomes particularly interesting to improve the environmen-
HCl > ­H2SO4 > ­HNO3 > ­H3PO4. Alternatively, alkaline con- tal sustainability while removing a waste from water and
ditions can be used with high reaction rate, but low sugar producing an energy vector simultaneously [48–50]. The
yield due to degradation [36]. challenges in this case are mainly related to the development
H2 as energy vector has been also proposed as a mean to of the photocatalyst, that in principle should be activated by
store excess electricity from renewables. For instance, wind solar light, being visible-light responsive and above all suf-
turbine installations currently afford convenient electric ficiently active. Very scarce productivities are reported until
energy price (0.03–0.07 $/kWh), that suggests the opportu- now with respect to rival processes that have much more
nity to associate to wind parks electrolysers for H ­ 2 produc- demonstrated maturity and shorter time-to-market. On the
tion from water. Different devices are commercially avail- other hand, the investigation of appropriate catalysts is the
able, such as polymer electrolyte membrane electrolysers key issue, making the research on the catalytic material one
(PEM), alkaline electrolysers (AE) or solid oxide (SO) ones. of the most popular topics nowadays.
Alkaline systems are available in large scale applications Bio-enzymatic catalysis is at the basis of most fermenta-
with relatively high efficiency (62%) [28]. The main chal- tive processes. Some of them, based on hydrogenase from
lenges for this application are related to the need of more anaerobic bacteria, directly aim at producing H ­ 2 starting
active catalysts for both the hydrogen and oxygen evolution from simple organic substrates, e.g. glucose. However, H ­ 2
reactions (HER and OER, respectively), since large over- yield is insufficient to justify the current costs. Alternatively,
potentials are required in both cases due to slow kinetics fermentation can be exploited through much more consoli-
[45]. Other issues are the high temperature of operation in dated processes to produce simpler substrates from biomass.
the case of SO electrolysers and the acidic operating condi- For instance, carbohydrate-containing raw materials can be
tions of PEM that impose the use of noble metals as OER fermented to obtain bioethanol in high yield, which can be
catalysts (mainly Ru and Ir), thus high cost. Ca. 12% solar- further used as substrate for the production of H ­ 2 by steam
to-hydrogen efficiency is reported [46]. reforming [51–55]. Through not yet commercialised, this is
Volcano plots are usually reported, with maximum activ- one of the most studied routes for the production of H ­ 2 from
ity (exchange current density) as a function of the metal- renewable substrates and the production of bioethanol from
hydrogen binding energy for different noble metals. AE are non-edible second generation biomass is even more sustain-
preferable for large scale applications, but also in this case able on a long term perspective [56].
the highest activity is achieved with noble metal catalysts SR can be also applied to other renewable derived sub-
(Pt, Pd, Ru, Ir and Rh). Bi-metallic Ni-Pt formulations dec- strates, such as glycerol obtained as by-product of biodiesel
orated on carbon cloths also showed very active for HER production by transesterification of vegetable oils. For

13
Topics in Catalysis

instance, the steam reforming of glycerol at 400 °C and section we newly elaborated some aspects of both plants to
4.5 bar was performed over Rh/Al2O3 catalyst, with maxi- highlight their main features and application potential.
mum hydrogen yield of 2.6 mol/mol glycerol fed [57] or over A 5 k­ Welectric + 5 ­kWthermal CHP system has been vali-
transition metal catalysts at relatively higher temperature dated starting from 96 vol% bioethanol and using either a
[58–60]. standard PEM FC operating at 80 °C and a high tempera-
Comparing the costs of the different technologies, when- ture PEM FC working at 160 °C [21, 84]. The system was
ever available, state-of-the-art technology offers 7 US$/GJ based on a pre-reformer and a main reformer, high (HT-) and
for ­H2 produced from SR of natural gas or partial oxidation. low (LT-)temperature WGS reactors and two methanators
If ­CO2 capture and storage is added, the ­H2 price increases to achieve < 20 ppm CO to feed the low temperature PEM
by ca. 25–30%. Higher price is currently estimated for bio- FC. A modification of the layout with a PEM FC working at
mass gasification (10–14 US$/GJ) or pyrolysis (8.9–15.5 160 °C allowed the removal of both the methanation reac-
US$/GJ) [12, 61]. tors, but the FC operation was much less reliable.
Considering a recent economic assessment of a central- A new hybrid turbofan with a molten carbonate fuel cell
ised facility for hydrogen production from steam reform- (MCFC) has been proposed, including a steam reformer,
ing of bioethanol, competitive price was forecasted if using a water gas shift reactor and a catalytic burner [85]. Five
diluted bioethanol streams, which would avoid the energy feedstocks have been compared, ethanol, methane, hydro-
intensive dehydration of bioethanol. The calculated mini- gen, dimethyl ether and methanol, including fuel blends. The
mum hydrogen selling price would be 1.91 USD/kg, to be results were also compared with a kerosene-based turbofan
compared with a present standard value of the one obtained that can produce 42 MW with 59% energetic efficiency and
from methane steam reforming (1.80 USD/kg) [62]. This 71% exergetic efficiency, whereas the hybrid MCFC turbofan
calculation was made for a system capable of producing can produce 40 MW by mixing all the five alternative fuels,
7793 ton/year of H ­ m3 ­h−1) starting from 40,000
­ 2 (9886 N but with higher performance of 65% and 80% energetic and
ton/year of bioethanol [63, 64]. Due to its high technical exergetic efficiencies, respectively. The ­CO2 emissions were
maturity and competitive cost over rival technologies, the reduced by 75%.
steam reforming of bioethanol is further deepened in the The steam reforming of methane presented the lowest
following paragraphs, deepening and newly elaborating the water footprint per kg of H ­ 2 produced (0.257 m3 /kgH2 ), fol-
results from selected papers. lowed by glycerol (0.768 m3 /kgH2 ) and bioethanol reform-
ing (9.651 m3 /kgH2 ). The exergo-environmental analysis
shows that the main bottlenecks identified are related to
exergy destruction in the burners (52.46–57.32%), reform-
3 Case History: Ethanol Steam Reforming ers (2.48–21.72%) and heat exchangers (19.45–32.61%),
which are all aspects to be considered during optimisation
3.1 Design and Validation of Integrated Plants of this process [86]. Thus, preventing external water addition
for Hydrogen Production by Ethanol SR through the use of diluted bioethanol improves the water
footprint of the process, together with the heat integration
Ethanol SR has been selected as case history to present the and energy efficiency of the system.
main steps towards process deployment. This example is To this end, multi-scale energy integration is needed, con-
indeed at a sufficiently advanced research stage to allow the sidering the coupling of different high-and low temperature
presentation of demonstration projects, integrated process units of the plant, as exemplified in detail in paragraph 3.4
design and economic assessment. Furthermore, it is based and optimising the thermal coupling at the meso-scale, e.g.
on a widely exploitable liquid raw material, obtainable eve- between the SR reactor and the burner. For instance, the
rywhere form different biomass, adaptable to different plant integration of ethanol steam reforming and ethanol combus-
size and easy to transport [65–67]. Very recent reviews detail tion in a parallel-plates reactor (size 1 k­ Wthermal) is proposed
the production of hydrogen from bioethanol [13, 68, 69]. through a mathematical model of a non-adiabatic reactor and
Examples of this application are described in the following. the associated heat exchangers. The influence of the insula-
Ni, Co and Cu-based catalysts were developed for this appli- tion of the reactor and heat exchangers, the fuel concentra-
cation [70–75] and based on detailed kinetic modelling [76, tion, fuel distribution policy to the reactor and flowrates are
77] two different plants were simulated: (1) a small scale considered, pointing out the importance of heat losses that
fuel processor coupled with a fuel cell for residential size represent ca. 35–50% of the heat released by the combus-
CHP [78–83] and (2) an industrial scale plant for central- tion [87].
ised hydrogen production, with economic assessment [63, Originally developed kinetics was used to design an
64]. The details for each system, the equipment sizing and integrated process of ­H2 production from ethanol, which
models validation are reported in the cited references. In this includes a parallel-plate reactor, a shell-and-tube membrane

13
 Topics in Catalysis

unit, and auxiliary units. Satisfactory thermal integration various promoters) has often been preferred over noble metal
with efficiencies from 43 to 47% based on lower heating catalysts, because is as active but less expensive [70, 71,
values, for a stream of the order of 10 molH2 /h [88]. 74, 75, 95, 98–106]. Depending on the catalyst support and
The integration of thin Pd-based membranes for H ­ 2 sepa- active material, the adsorbed ethanol can be activated by a first
ration in a fluidized bed catalytic reactor for ethanol auto- abstraction of the oxygen-bound hydrogen. Some ethanol can
thermal reforming has been demonstrated for over 50 h. 5 then be lost by dehydration to ethylene, while the reforming
thin Pd-based alumina-supported membranes and a 3 wt% path is generally believed to involve a first oxidative dehy-
Pt–10 wt% Ni catalyst deposited on a mixed ­CeO2/SiO2 sup- drogenation to acetaldehyde followed by the C–C bond break
port were used, with H ­ 2 recovery factor up to 70. The most and steam-reforming of the resulting methane [76, 107, 108].
critical part of the system was the sealing of the membranes, The importance of the methane reforming reaction over the
where most of the gas leakage was detected. A fluidized ­CH4 conversion to ­CO2 and water–gas shift equilibria depends
bed membrane reactor model for ethanol reforming was also on the peculiar catalyst properties and reaction temperature
validated with the experiments and used to design a small (usually in the range 400–600 °C) [95, 109].
reactor unit for domestic use: 0.45 m ­ 2 membrane area was In general, the theoretical output of 6 mol of ­H2 per mole
needed to feed a 5 ­kWelectric PEM FC [89]. of ethanol (at the balanced ratio of 1 mol of ethanol and 3
A high temperature PEM FC integrated with a reformer of water) is not reached due a residual presence of methane
has been studied considering the efficiency of different [110]. Though water molecules may hamper the ethanol
layouts and fuels. Ca. 60% heat is needed to preheat the absorption on acidic supports (mainly alumina) [111], water/
reformer fuel when increasing its dilution with steam form alcohol ratios higher than 3 mol/mol have been successfully
S/C = 4 to 6. Also in this case the key to improve the effi- tested [79, 97, 111], either for the fact that water undergoes a
ciency of the system is the thermal integration [89]. dissociative absorption turning into the active moieties –OH·
The techno-economic analysis of an ethanol-to-H2 plant and –H, and for its role in reforming and removing the coke
to feed on board PEM FCs has been considered using Rh/ formed by ethylene polymerization or by –CH· aggregation
CeSiO2 as reforming catalyst and Pt/ZrO2 for the WGS step. into soot.
Monolith structured catalysts were developed, sufficiently A reaction network compliant with the above sketched
stable towards coking [90]. As also pointed out in [63, 64], mechanism (Fig. 1) can be built to different complexity levels
the ­H2 production cost was significantly OPEX sensitive and [76, 94, 112–117], with a heuristic or microkinetic approach
mainly affected by the ethanol price [90]. The ­H2 production
cost obtained using Brazilian ethanol (0.81 US$/Lethanol) was
8.87 US$∕kgH2 , lower than the current market prices (13.44
US$∕kgH2 ) practiced at ­H2 re-fuelling stations in California
[90].
Finally, examples of catalytic membrane microreactors
have been recently reviewed, highlighting examples of fuel
processing (e.g. reforming of alcohols) [91]. The inherent
advantage of micro-reactors and membrane separation are
the enhanced heat and mass transfer and better catalytic effi-
ciency. Compact configurations are made available through
lab-on-a-chip technologies, suited for portable or mobile
devices.

3.2 Mechanism, Kinetics and Thermodynamics

Whenever a catalytic process has been demonstrated valid

at lab scale, the scale up should rely on robust knowledge
of kinetics and thermodynamics, at the basis of reactor and
process design. Heterogeneous catalytic kinetics is for sure
more complex to assess than for homogeneous reactions and
implies also hypotheses on the future size and shape of the
catalyst to estimate its effectiveness factor.
The mechanism of the ethanol steam reforming has been
extensively studied in the past years, both theoretically [92,
93] and experimentally [70, 76, 77, 83, 93–97]. Nickel (with Fig. 1  Mechanism of ethanol reforming

13
Topics in Catalysis

[118]. The latter solution has been worked out successfully

( B
)
yCO yH2 O 1 − eA+ T Q
and furtherly adapted to a set of proprietary data [76, 77], and Ea3
r3 = k3 e RT (E4)
can be used to calculate in detail the thermal profile and the D2
material balances rising in a reformer:
yH2 O
CH3 CHO → CH4 + CO (R4) D = 1 + KA2 yEtOH + KA3 yH2 O + KA4 yH2 + KA5
y0.5
H
(E5)
2
yCH4
CH3 CH2 OH + H2 O → CO2 + 2H2 + CH4 (R5) + KA6 + KA7 yCO + KA8 y0.5
H
y0.5
H
2
2

CH4 + 2H2 O ⇄ CO2 + 4H2 (R6)

where Q are the reaction quotients and the parameters are
as reported in Table 1. ­Yi represent the molar fractions of
CO2 + H2 ⇄ CO + H2 O (reverse R3) the respective species, ­ri are the reaction rates of the speci-
fied reaction, k­ i the preexponential factors, E
­ ai the activation
CH3 CH2 OH → CH3 CHO + H2 (R7) energies and ­KAi are adsorption constants.
Notice that some equilibrium constants in the above
CH3 CH2 OH → CH2 CH2 + H2 O (R8) equations are modified, with respect to the original values
reported by the authors, to correct the fugacities at 10 bar
(i.e. the supposed pressure for the process described in the
CH2 CH2 → 2C + 2H2 (R9)
following) with respect to the atmospheric pressure of the
lab tests from which they were retrieved.
The kinetic equations and the regressed parameters are The first, complete formulation is used to simulate a
listed in the original work [76], depending on the dataset on reformer for distributed small scale CHP, while the abridged
which the retrofit has been operated (i.e. on the catalyst type). reaction set is used for the more extended diagram of a cen-
Nevertheless, it has to be pointed out that this set of rate tralised large scale ­H2 production. In addition, in the former
equations may present some calculation drawbacks Especially case, the water–gas shift and methanation steps after the
due to the presence of many enthalpy-related parameters, the reformer are assumed in equilibrium, whereas for centralised
system is very sensitive to temperature variations. Therefore, hydrogen production, the high temperature water gas shift
the adoption of a simpler system of heuristic equations can stage was still considered in equilibrium, while the lower
be considered to achieve a preliminary model of a reformer temperature reactor downstream (that can actually modify
(especially if included in a relatively complex flowsheet), after the process balances) was modelled after the reviewed lit-
the review of available literature models [119]: erature [120].
When designing a process using simulators, the appro-
CH3 CH2 OH ⇄ CH3 CHO + H2 (R7) priate thermodynamic methods must be selected in order to
avoid wrong estimation of phase and chemical equilibria.
Often, different sections of a complex system need definition
( B
)
yEtOH 1 − eA+ T Q
r10 = k10 e
Ea10
RT
(E1)
D
Table 1  The parameters K
­ A2–8 have the same form as the reaction
CH3 CHO ⇄ CO + CH4 (R4) constants ­k1−n

( ) k0 (kmol/s kg) Ea/R (K) A B

B
yAcH 1 − eA+ T Q
Ea11
(E2) r1 11.4 − 3825 − 14.7 8116
r11 = k11 e RT
D r2 8.42 − 2983 − 14.7 − 1609
r3 9.87e6 − 14,313 − 53.0 26,154
CH4 + H2 O ⇄ CO + 3H2 (R12) r4 1.06e6 − 7264 − 4.39 − 4657
KA2 4.1 596.5
( B
) KA3 5.0e−6 6845
yCH4 yH2 O 1 − eA+ T Q KA4 7.4e−4 8030
r12 = k12 e
Ea12
(E3)
KA5 1.2e−6 6440
RT
D2
KA6 6.8e−9 8025
KA7 5.6e−9 3260
CO + H2 O ⇄ CO2 + H2 (R3)
KA8 1.0e−8 9565

13
 Topics in Catalysis

through different models, to be validated against experimen- (the number of moles in the gas phase increases during the
tal data. According to our experience the following choices reaction). The overall layouts of methane reforming plants
are appropriate for the different units. [128, 129] have been reviewed as starting points for the
Process simulation has been carried out with Aspen PLUS design of this new plant for ESR.
V8.8 or following, retrieving the thermodynamic data from Figure 3 reports the layout of the reforming reaction
the PURE32 databank. The thermodynamic model used for section. The feed is pumped through a heat exchanger that
the description of the gas phase is the Peng–Robinson equa- recovers part of the products heat and is fed to the reform-
tion of state, except for several heat exchangers dealing with ing reactor, which includes multiple catalytic beds and
a substantial liquid fraction of ethanol–water mixture, which is thermally sustained by a burner. The latter is fed with
is strongly non-ideal. The latter is better described by the ambient air and the purge flow from the Pressure Swing
NRTL method, implementing the Wilson mixing rules. Adsorption (PSA) unit. Both the radiative direct heat-
The ­CO2 separation is achieved by coupling an absorption ing power and the convective residual heat content of the
and a stripping column, or through an expanded pressure- burner off gases are fully exploited for heating. The prod-
swing section. In the latter case, the gas phase is treated with ucts of reaction are sent to a HT-WGS reactor and then to
the RKS equation, while the liquid phase with the NRTL a LT-WGS one. Various heat exchangers are shown in the
model, whereas for absorption using an aqueous solution of flowsheet to highlight the importance of thermal recovery
amines (MDEA), the electrolytic acid–base equilibria must to improve the overall process efficiency.
be computed. The fate of the reformate may be different depending
Methane, hydrogen and carbon monoxide solubilities are on the application. For instance, in Fig. 4 the production
calculated with the Henry constants. This model is useful to of pure H­ 2 is illustrated. Separation of excess water is
correlate the solubility of gases in liquids as a function of achieved by flash separation, while ­CO 2 is removed by
their absolute or partial pressure. The relationship between absorption with basic solutions of alcanolamines. A dou-
liquid phase concentration and pressure in the gas phase ble column is represented in Fig. 4, the former for the
is usually linear for dilute systems. Calculations of phase absorption of ­CO2 and the second for amine regeneration.
partitions with the Aspen ­Plus© database to reproduce litera- Final ­H2 purification is accomplished by PSA. The purge
ture data [121–126] are reviewed throughout Fig. 2, which gas of the PSA is used as fuel for the burner of Fig. 3.
reports the mole fractions of each compound in the liquid Figure 5 details the performance of the reforming
phase as a function of temperature. The same Fig. 2 also reactor, describing the profile of conversion and hydro-
compares the experimental data retrieved from the literature gen concentration vs. temperature along the three cata-
vs. the ones calculated in liquid phase through the Henry lytic beds. The respective temperature profiles, divided
model. between the zones where heating is done by irradiation or
Accurate predictions of solubility are available for by convection, are also reported. The detail of gas com-
­CO2(liq), ­H2O(vap) and C
­ H4 partition between the liquid and position outflowing from the different units are reported
the gas. The Henry constant for Hydrogen may need a revi- in Fig. 6. Hydrogen concentration progressively increases
sion in further developments, but in this specific case the through the three reforming beds and thanks to the follow-
operating conditions used to simulate the plant are in a low- ing WGS units. ­CO2 and the remaining impurities are fully
temperature region (below 50 °C) or in a high-pressure one, eliminated with the two units of ­CO2 absorption and PSA.
where the default parameters provide a satisfactory descrip- Possibly, a single PSA may be designed also to remove
tion of hydrogen partition between the gas and liquid phases. ­CO2, but it is a major component and this would require
oversizing of this unit.
3.3 Design of a Medium‑Scale Centralised The process depicted in Figs. 3 and 4 simulates a reformer
Hydrogen Production conceptually divided into a radiant and convective section
via the following criteria:
The size selected for a bioethanol to hydrogen produc-
tion facility is of 4.6 t/h of ethanol diluted with 3 mol of – part of the heat released by the burned fuel just before
water per mole of alcohol, i.e. at the stoichiometric ratio. the radiant section is actually used to heat the feed, so the
This corresponds to the ethanol produced by a medium- combustion is not adiabatic and the higher temperature is
size bio-refinery [127]. Starting from an already diluted fixed to 1000–1100 °C: the resulting heat surplus at the
mixture, it is implicitly considered that the prior distilla- burner is used as an input to the last feed heater;
tion up to azeotropic ethanol is not needed. The reaction – the hot flues are split in three, to simulate a parallel cross
pressure was set to 10 bar, values even higher than this are flow between the hot gases and the process stream in the
common in reforming plants in order to limit the piping radiant section: the first part heats up the ethanol before
volume, though unfavorable for thermodynamic reasons

13
Topics in Catalysis

Fig. 2  Capability of the library models to reproduce the solubility of (from top): ­H2 (atmospheric-high pressure), ­CO2 and ­CH4 in water

the catalytic bed, the second part constitutes the heating added to intercept the third hot flues crossflow still at
stream of a plug-flow reactor; 1000 °C. It is deemed that this twofold bed arrange-
– the first catalytic bed is supposed to 'end' where the ment and threefold flues partition is a reasonable rep-
outlet process and thermal fluid temperatures become resentation of the radiative section: the heat exchange
closer than 300 °C, so another virtual catalytic bed is

13
 Topics in Catalysis

Fig. 3  PFD of the reforming section. PSA pressure swing adsorption

coefficient for this part is conservatively set to 30 W/ can actually capture also this gas. In both cases, anyway,
m2 °C; the equilibrium condition at the condenser of the stripper
– the convective section is modelled via a single plug- makes a non-negligible amount of water to be lost with the
flow-reactor heated by a single flue stream, with a heat carbon dioxide vent (3–4 wt%) and the circuit between the
exchange coefficient of 15 W/m2 °C. columns is actually calculated with a continuous makeup
in the low pressure side, just upstream the charging pump.
The final (and most important) water–gas shift section The pressure-swing purification part is accounted for
is designed around 250 °C, resorting to a kinetic expres- after the results reported on multi-layer adsorption beds
sion available in the literature. After this step, the water still capable of trapping different gases [131, 132]. The purge
in excess can be condensed in one or two steps. The sec- stream is not recycled to recover the hydrogen [128], but
ond strategy is convenient if the second separator works at routed to the burner. This approach is based on the follow-
higher pressures, and it is adopted here because pressures ing considerations:
of 15–20 bar are more effective when C ­ O2 is to be absorbed
in basic aqueous solutions. – though hydrogen is a valuable building block, its use
The carbon dioxide treatment module is based on the as a fuel is nearly as important;
use of aqueous methyl diethanolamine (MDEA), 1.7 mol – being the reforming reaction globally endothermic,
per mole of C­ O2, and a 3% volumetric fraction of carbon recycling about the 30% of the produced hydrogen as
dioxide is allowed in the output stream [130], because a fuel keeps it within the process balances and also the
most adsorption beds as the one foreseen downstream

13
Topics in Catalysis

Fig. 4  PFD of the reformate purification

appreciable residual quantity of methane is turned into using a PSA system and an amine scrubbing module for
a fuel rather than a byproduct; the recovery of pure ­CO2, as detailed in Fig. 4.
– this option leads to the design of a closed process, that The conversion and thermal profile in the reforming reac-
does not rely neither on external fuel gas or azeotropic tor are detailed in Fig. 5, as well as the composition of the
ethanol while it can exploit as desired a diluted inex- gas outflowing from each stage is detailed in Fig. 6a. Finally,
pensive bioethanol, depending on the context mixture, Fig. 6b details the duty needed for each stage. Besides the
it is worthwhile to study also a stand-alone version of heat duty to the reactor, significant contribution is due to
the process. the condensation step, which needs compression of a large
flowrate of gases (prodromal also to ­CO2 removal and PSA)
Notice that, according to the reviewed references, also and for the C
­ O2 absorption unit, in particular for the regen-
large quantities of C
­ O2 can be treated via a pressure-swing eration of the amine.
apparatus relying on the same (or very similar) solids that
entrap methane and carbon monoxide. The preliminary
design has been devised in this way. A dedicated and 3.4 Energy Recovery
selective purification section makes ­CO 2 available for
re-utilization or capture purposes. The delay in hydrogen The reformate that cools down to the condenser temperature
production with respect to methane formation (due to the (chosen slightly above 80 °C) and the burner flues exiting
used reaction network) appears clearly in Fig. 5a, leav- the reformer still contain some thermal energy that is partly
ing to the last reformer stage the task to switch the ratio used to pre-heat the feed and partly to produce low pressure
between the two species, with almost full conversion of steam. In principle, the feed heating stages depends on how
the formed methane. In particular, Fig. 3, from left to right the diluted ethanol is actually provided:
reports three catalytic packed bed reactors, the former for
reforming (thermally sustained by a burner), the two high (1) as base case, it is supposed to start from a mildly pre-
and low temperature water gas shift steps. In our previous heated (50 °C) hydro-alcoholic mixture at the stoichio-
investigations [79, 132, 133], the further purification was metric ratio, charged at the process pressure; this work-
carried out by selective methanation of CO, to achieve a ing point can be taken as reference to consider at least
residual 20 ppm amount. Here a new route was explored other two options [134–136];

13
 Topics in Catalysis

Fig. 6  a Composition of the streams and b plant duties. PFR1-3 rep-
resent the three steam reforming beds, HTWGS and LTWGS the
high and low temperature water gas shift reactors, condensation the
unit for water discharge and C
­ O2 wash the C
­ O2 absorption unit with
Fig. 5  a profile of the energy-carrying species and conversion in the amines (and relative regeneration)
reformer; b temperature difference map between the reacting mix-
tures and the flues
(b) in this case the additional water could come from the
bottoms of the wine boiler, but another pre-treatment
would be needed;
(2) reboil a fermentation broth up to the desired ethanol (c) in this case the water could enter the system first in the
concentration and route it to the second feed heater; auxiliary steam section, then be used as process feed to
(3) distillate azeotropic ethanol and mix it with the steam dilute the azeotrope.
produced by the plant.
The heat-recovering utilities have been designed after
The water used to raise steam could belong to a closed reviewing and adapting the already cited process layouts.
cycle. If, instead, an open cycle should be designed, then Except for the ethanol/water feeding sections, subject to fur-
there would be at least three different entry-levels for water ther optimization according to the above sketched criteria,
into the system, according to what described above: the main thermal and power balances for the base case are
summarized in the Fig. 6b.
(a) as base case, fresh water enters the process and is mixed The basic Pinch Analysis, restricted to the process
to the process condensate stripped from the dissolved streams only, is reported in Fig. 7. It is possible to see that,
gases; even at the process pressure, the dew point of the refor-
mate is below the temperature level requested in the MDEA

13
Topics in Catalysis

stripper reboiler, leading to the poor overlap of the hot and

cold composite curves (HCC and CCC). The choice of burn-
ing part of the hydrogen to supply the reaction energy leads
naturally to the coupling of a Hot Utility profile (as indicated
in the same graph), following the criterion of setting a utility
pinch at the top kink of the grand composite curve (GCC,
Fig. 8). The excess heat carried by the hot flues is spent
rising steam at 15 bar (195–198 °C), which brings to the
curved shape of the hot curve below 430 °C. The analysis
lets foresee, anyway, at least two options for further energy
recovery: lowering the highest flue temperature (e.g. by add-
ing over-stoichiometric air) and/or rising more steam.
Table 2 highlights the main thermal parameters of the
process and the process plus utilities. Notice that the final
heat sink foreseen is atmospheric air. It is visible that a sig-
nificant reduction of the foreseen use of hot and cold utilities
is achieved after matching hot and cold streams according to
the results of the pinch analysis.

3.5 Distributed Heat and Power Cogeneration

The main layout difference between the large-scale reform-

ing and the micro-scale hydrogen production are [79, 133]:

– only CO must be removed but not ­CO2, then just a metha-

nation reactor is foreseen upstream the water condenser
to allow CO concentration down to 20 ppm. This unit can
be discarded if high temperature PEMFC are used. Water
condensation is anyway compulsory because the fuel cell
would be flooded by excess water [137];
Fig. 7  Composite curves (CC) and Grand-Composite curve (GCC) – the fuel to heat the reformer is extracted before the FC,
for the process. The y-scale reports the shifted temperature
then the system is intrinsically stable (in terms of global
power output), because the electric power (hydrogen to
the cell) and the thermal one (hydrogen to the burner) are
dependent [78];
– the residual heat recovery is carried out in a series of
blocks that simulate a common household heating sys-
tem: since the quantity of ethanol needed to provide
the standard electrical supply (i.e. 70 mol/h of alcohol
diluted with 350 mol/h of water, that exceed the nominal
5 ­kWelectric target and leave a margin for fuel cell effi-
ciencies possibly lower than 50%) cannot cover, at the
same time, also the installed thermal need (usually up
to 20–25 k­ Wthermal for burners with hot convective flue
flow), the calculation is solved dynamically exploiting
the fact that most of the time the two kinds of appliances
do not deliver their target power simultaneously.

Fig. 8  Grand-Composite curve (GCC) for the process plus utilities Referring to the scheme in Fig. 9, the reforming system
after matching according to the pinch analysis. HU hot utility
is represented in the lower section, while the house-heating
system is drawn in the upper part, but only for its steady-
state functioning part (i.e. the production of hot water to
feed the radiators).

13
 Topics in Catalysis

Table 2  Main energetic Case Overall T pinch (°C) HU (kW) CU (kW) Heat input (kW) Heat Output
parameters of the simulation ∆T (°C) (kW)
cases
No matches 10 121 16,100 5740 21,800 12,200
Matches & utilities 10 111 0 1940 0 –

HU hot utility, CU cold utility

Fig. 9  Process flow diagram of a small-scale Heat & Power Cogeneration unit. SR steam reformer, FC fuel cell

The key point of the material and energy recycle is the 3.6 Dynamic Energy Integration
reformer behaviour, as is modelled rigorously via the reac-
tion network presented above. In Fig. 10 it can be seen that As already mentioned, the supply of sanitary water (up to 12
right flowrate and temperature of the burned gas is crucial L/min) cannot be covered with the above mentioned feed,
to avoid an anomalous temperature profile (with the given supposing to start from grid water at 5 °C to a high setpoint
activation energy), that means in turn loss of the hydrogen of 50 °C (see [82] and references therein), so a static heat
output. Figure 10 exemplifies the thermal behaviour of the reserve is put in order to store the power continuously dis-
process and burner flue gas streams (used to heat up the charged and not consumed. Also the steady-state operation,
reformer) along the catalyst bed. Good match is compulsory however, can shift from an enhanced electricity production
between the amount of reactants in the reforming process to the delivery of more thermal power.
line and the one of reformate burned to thermally sustain The working philosophy is as follows:
the reformer, in order to avoid instability and consequent
decreasing yield. – the hydrogen usage at the FC determines the electric
power and leaves more or less enthalpy after the burner;

13
Topics in Catalysis

Fig. 10  Thermal behaviour of the reformer at 400 (left) and same process flow inlet temperature and variable heating flue gas
600 mol/h (right) of hot flues (feed: 420 mol/h). Solid lines: refor- temperatures (readable in the temperature scale)
mate—dashed lines: flues. Differently coloured curves refer to the

– according to the feed water content, some hot gas has to in research and development are still needed in every case,
bypass the burner and/or the feed boiler to keep the reac- depending on the level of maturity reached.
tor at the rated condition (around 500 °C); In many cases the open issues still rely on the develop-
– a reduced electricity production leaves then enough heat ment or optimisation of the catalytic material. For instance,
available, at the cell and flue gas heat exchangers, to meet in the case of APR, and more in general of the reforming
the 6.5 kW to 70 °C wintertime utility requirement usu- of renewable substrates, transition metal catalysts have the
ally considered in most Northern Italy houses; advantage of low cost, good activity (even if under harsher
– when sanitary water is needed, the line passing though conditions than noble metals), but are characterised by
the dissipator is bypassed and all the heat available significant deactivation. This point is the main issue to be
remains in the system: this set-up, coupled with the res- addressed in future investigations.
ervoir thermal inertia, keeps a milder sanitary set-point On the contrary, when attempting to develop photocata-
of 47–48 °C for more than 30 min. lytic routes for the valorisation of biomass-derived feed-
stocks, the targets of hydrogen productivity are not yet met,
The main results are reported in Figs. 11 and 12. Fig- so catalyst development should rely mostly on optimising
ure 11 reports for two cases of different ­H2 utilisation in the activity.
fuel cell, 50 or 80%, the amount of heat available through For the direct exploitation of biomass, the economy of
different streams, i.e. from the FC itself, from the excess scale suggests very large installations, which however do not
water discharged in the condenser and from waste flue gases. fit the supply rate of biomass in most applications. The chal-
Different scenarios are compared (different colours) depend- lenge is therefore the development of small decentralised
ing on the flow rate of water that can be recycled internally. plants that must be economically convenient. Some of the
The same Fig. 11 also reports the profile of the sanitary fractions produced, e.g. biooil or its aqueous fraction, may
water temperature available for the time lapse evidenced offer interesting options for catalytic valorisation through
in abscissa, depending on the regulation of water supply. reforming. Yet, specific catalysts must developed to stably
Details on this dynamic control scheme are available in the convert a widely variable substrate.
original reference. At this stage, for all the considered processes the robust
Figure 12 summarises some data on the efficiency, whose knowledge of kinetics, reaction mechanism and process
total value may sum up to 60% with respect to the inlet etha- design plays a key role, and such topics should be early
nol feed. developed whenever possible to assess the real feasibility
of the solution proposed.
Examining the SR of bioethanol as case history at suf-
4 Conclusions ficiently high TRL (demonstrative phase), the exergo-
environmental analysis shows that the main bottlenecks
Some of the most investigated processes for catalytic hydro- are related to exergy destruction in the burners, reformers
gen production and for its recovery from biomass and bio- and heat exchangers. Therefore, optimisation mainly refers
mass-derived substrates have been reviewed. Many efforts to the integrated process, to better thermally integrate the

13
 Topics in Catalysis

Fig. 11  From top: heat released at different ethanol:water ratios (40% of burned gas to the reformer); adjustment of sanitary set point via time-
dependent water-mixing strategies. FC fuel cell

different plant sections. In this light, efforts in catalyst

development should be related in the decrease of operat-
ing temperature. Furthermore, the use of diluted bioethanol
improves the water footprint of the process, together with
the heat integration and energy efficiency of the system, but
poses key questions on the reliability of the catalyst with
possible impurities of the feed. Overall, the possibility to
produce hydrogen through diluted hydroalcoholic solutions
is demonstrated. This technology has been applied to the
centralized production of hydrogen (big scale plant) and to
the distributed co-generation of heat and power by using
fuel cells. In this latter case, the time-to-time availability of
sanitary water during winters has been considered dynami-
cally, when demanding working conditions are yet present
for heating, and demonstrated feasible.

Fig. 12  Efficiency (%) of all the tested working cases. The enthalpy Open Access This article is licensed under a Creative Commons
content of ethanol has been set to 1370 kJ/mol: the heat recovery is Attribution 4.0 International License, which permits use, sharing,
such as to maintain a nearly constant thermal efficiency

13
Topics in Catalysis

adaptation, distribution and reproduction in any medium or format, 16. Chen Y, Mu X, Luo X et al (2020) Catalytic conversion of meth-
as long as you give appropriate credit to the original author(s) and the ane at low temperatures: a critical review. Energ Technol. https://​
source, provide a link to the Creative Commons licence, and indicate doi.​org/​10.​1002/​ENTE.​20190​0750
if changes were made. The images or other third party material in this 17. Zhang H, Sun Z, Hu YH (2021) Steam reforming of methane:
article are included in the article's Creative Commons licence, unless Current states of catalyst design and process upgrading. Renew
indicated otherwise in a credit line to the material. If material is not Sustain Energy Rev. https://​d oi.​o rg/​1 0.​1 016/J.​R SER.​2 021.​
included in the article's Creative Commons licence and your intended 111330
use is not permitted by statutory regulation or exceeds the permitted 18. Adris AM, Pruden BB, Lim CJ, Grace JR (1996) On the reported
use, you will need to obtain permission directly from the copyright attempts to radically improve the performance of the steam meth-
holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​ ane reforming reactor. Can J Chem Eng 74:177–186
org/​licen​ses/​by/4.​0/. 19. Mehmeti A, Angelis-Dimakis A, Arampatzis G et al (2018) Life
cycle assessment and water footprint of hydrogen production
methods: from conventional to emerging technologies. Environ-
ments. https://​doi.​org/​10.​3390/​envir​onmen​ts502​0000
References 20. Moreira R, Bimbela F, Gandía LM et al (2021) Oxidative steam
reforming of glycerol. A review. Renew Sustain Energy Rev.
1. Yip HL, Srna A, Yuen ACY et al (2019) A review of hydrogen https://​doi.​org/​10.​1016/J.​RSER.​2021.​111299
direct injection for internal combustion engines: towards carbon- 21. Rossetti I, Biffi C, Tantardini GF et al (2012) 5 kW e + 5 kW t
free combustion. Appl Sci 9:4842. https://d​ oi.o​ rg/1​ 0.3​ 390/a​ pp92​ reformer-PEMFC energy generator from bioethanol first data on
24842 the fuel processor from a demonstrative project. Int J Hydrogen
2. (2015) https://w ​ ww.e​ nergy.g​ ov/s​ ites/p​ rod/fi​ les/2​ 015/1​ 1/f​ 27/f​ cto_​ Energy 37:8499–8504. https://​doi.​org/​10.​1016/j.​ijhyd​ene.​2012.​
fuel_​cells_​fact_​sheet.​pdf. Accessed Oct 2021 02.​095
3. https://​www.​energy.​gov/​eere/​fuelc​ells/​fuel-​cells 22. Lee WJ, Li C, Prajitno H et al (2021) Recent trend in thermal
4. Ferraren-De Cagalitan DDT, Abundo MLS (2021) A review catalytic low temperature ­CO2 methanation: a critical review.
of biohydrogen production technology for application towards Catal Today 368:2–19. https://​doi.​org/​10.​1016/j.​cattod.​2020.​
hydrogen fuel cells. Renew Sustain Energy Rev. https://​doi.​org/​ 02.​017
10.​1016/J.​RSER.​2021.​111413 23. Rhodes C, Williams BP, King F, Hutchings GJ (2002) Promo-
5. Tzelepis S, Kavadias KA, Marnellos GE, Xydis G (2021) A tion of ­Fe3O4/Cr2O3 high temperature water gas shift catalyst.
review study on proton exchange membrane fuel cell electro- Catal Commun 3:381–384
chemical performance focusing on anode and cathode catalyst 24. Alijani A, Irankhah A (2013) Medium-temperature shift catalysts
layer modelling at macroscopic level. Renew Sustain Energy Rev. for hydrogen purification in a single-stage reactor. Chem Eng
https://​doi.​org/​10.​1016/J.​RSER.​2021.​111543 Technol 36:209–219. https://​doi.​org/​10.​1002/​ceat.​20120​0151
6. Melo SP, Barke A, Cerdas F et al (2020) Sustainability assess- 25. Pal DB, Chand R, Upadhyay SN, Mishra PK (2018) Performance
ment and engineering of emerging aircraft technologies-chal- of water gas shift reaction catalysts: a review. Renew Sustain
lenges, methods and tools. Sustainability (Switzerland). https://​ Energy Rev 93:549–565. https://​doi.​org/​10.​1016/J.​RSER.​2018.​
doi.​org/​10.​3390/​su121​45663 05.​003
7. Savla N, Suman PS et al (2021) Techno-economical evaluation 26. Song C (2002) Fuel processing for low-temperature and high-
and life cycle assessment of microbial electrochemical systems: a temperature fuel cells: challenges, andopportunities for sustain-
review. Curr Res Green Sustain Chem. https://​doi.​org/​10.​1016/J.​ able development in the 21st century. Catal Today 77:17–49
CRGSC.​2021.​100111 27. de Guido G, Compagnoni M, Pellegrini LA, Rossetti I (2018)
8. Mahian O, Mirzaie MR, Kasaeian A, Mousavi SH (2020) Exergy Mature versus emerging technologies for ­CO2 capture in power
analysis in combined heat and power systems: a review. Energy plants: Key open issues in post-combustion amine scrubbing and
Convers Manage 226:113467. https://d​ oi.o​ rg/1​ 0.1​ 016/J.E ​ NCON​ in chemical looping combustion. Front Chem Sci Eng. https://​
MAN.​2020.​113467 doi.​org/​10.​1007/​s11705-​017-​1698-z
9. Valera-Medina A, Amer-Hatem F, Azad AK et al (2021) Review 28. NREL www.​nrel.​gov/​hydro​gen/​New Horizons For Hydrogen
on ammonia as a potential fuel: from synthesis to economics. 29. Asadullah M, Ito SI, Kunimori K, Tomishige K (2002) Energy
Energy Fuels 35:6964–7029. https://​doi.​org/​10.​1021/​acs.​energ​ efficient production of hydrogen and syn- gas from biomass:
yfuels.​0c036​85 development of low-temperature catalytic process for cellulose
10. https://a​ fdc.e​ nergy.g​ ov/f​ uels/e​ thano​ l_f​ uel_b​ asics.h​ tml. Accessed gasification. Environ Sci Technol 36:4476–4481
Oct 2021 30. Yang K, Zhu N, Yuan T (2017) Analysis of optimum scale of bio-
11. Javaid R (2021) Catalytic hydrogen production, storage and mass gasification combined cooling heating and power (CCHP)
application. Catalysts 11:12–13. https://​doi.​org/​10.​3390/​catal​ system based on life cycle assessment(LCA). Procedia Eng
11070​836 205:145–152. https://​doi.​org/​10.​1016/j.​proeng.​2017.​09.​946
12. Kalamaras CM, Efstathiou AM (2013) Hydrogen production 31. Wang S, Shan R, Gu J et al (2020) Pyrolysis municipal sludge
technologies: current state and future developments. Conf Pap char supported Fe/Ni catalysts for catalytic reforming of tar
Energy 2013:1–9. https://​doi.​org/​10.​1155/​2013/​690627 model compound. Fuel. https://​doi.​org/​10.​1016/J.​FUEL.​2020.​
13. Panagiotopoulou P, Papadopoulou C, Matralis H, Verykios X 118494
(2014) Production of renewable hydrogen by reformation of bio- 32. He L, Hu S, Yin X et al (2020) Promoting effects of Fe–Ni alloy
fuels. Wiley Interdiscip Rev Energy Environ 3:231–253. https://​ on co-production of ­H2 and carbon nanotubes during steam
doi.​org/​10.​1002/​wene.​93 reforming of biomass tar over Ni-Fe/α-Al2O3. Fuel. https://​doi.​
14. Baraj E, Ciahotný K, Hlinčík T (2021) The water gas shift reac- org/​10.​1016/J.​FUEL.​2020.​118116
tion: catalysts and reaction mechanism. Fuel. https://​doi.​org/​10.​ 33. Yoon SJ, Kim YK, Lee JG (2011) Catalytic oxidation of biomass
1016/j.​fuel.​2020.​119817 tar over platinum and ruthenium catalysts. Ind Eng Chem Res
15. Nisbet EG, Dlugokencky EJ, Bousquet P (2014) Methane on 50:2445–2451. https://​doi.​org/​10.​1021/​IE102​0365
the rise—again. Science 343:493–495. https://​doi.​org/​10.​1126/​ 34. Gao N, Salisu J, Quan C, Williams P (2021) Modified nickel-
SCIEN​CE.​12478​28 based catalysts for improved steam reforming of biomass tar: a

13
 Topics in Catalysis

critical review. Renew Sustain Energy Rev. https://​doi.​org/​10.​ 55. Posada JA, Patel AD, Roes A et al (2013) Potential of bioethanol
1016/J.​RSER.​2021.​111023 as a chemical building block for biorefineries: preliminary sus-
35. Gao N, Chen K, Lai X, Quan C (2021) Catalytic steam reforming tainability assessment of 12 bioethanol-based products. Biore-
of real tar under high-efficiency Ni/USY catalyst for ­H2 produc- sour Technol 135:490–499. https://​doi.​org/​10.​1016/J.​BIORT​
tion. Fuel. https://​doi.​org/​10.​1016/J.​FUEL.​2021.​121676 ECH.​2012.​09.​058
36. Tanksale A, Beltramini JN, Lu GQM (2010) A review of catalytic 56. Malode SJ, Prabhu KK, Mascarenhas RJ et al (2021) Recent
hydrogen production processes from biomass. Renew Sustain advances and viability in biofuel production. Energy Convers
Energy Rev 14:166–182. https://​doi.​org/​10.​1016/j.​rser.​2009.​08.​ Manage: X 10:100070. https://​doi.​org/​10.​1016/j.​ecmx.​2020.​
010 100070
37. Nachenius RW, Ronsse F, Venderbosch RH, Prins W (2013) Bio- 57. Ayodele BV, Mustapa SI, Tuan Abdullah TAR, Bin SSF (2019)
mass pyrolysis. Adv Chem Eng 42:75–139. https://​doi.​org/​10.​ A mini-review on hydrogen-rich syngas production by thermo-
1016/​B978-0-​12-​386505-​2.​00002-X catalytic and bioconversion of biomass and its environmental
38. Wang D, Czernik S, Chornet E (1998) Production of hydrogen implications. Front Energy Res 7:1–6. https://​doi.​org/​10.​3389/​
from biomass by catalytic steam reforming of fast pyrolysis oils. fenrg.​2019.​00118
Energy Fuels 12:19–24 58. Rossetti I, Gallo A, DalSanto V et al (2013) Nickel catalysts
39. Czernik S, Evans R, French R (2007) Hydrogen from biomass— supported over ­TiO2, ­SiO2 and ­ZrO2 for the steam reforming
production by steam reforming of biomass pyrolysis oil. Catal of glycerol. ChemCatChem 5:294–306. https://​doi.​org/​10.​1002/​
Today 129:265–268 cctc.​20120​0481
40. Czernik S, French R, Feik C, Chornet E (2002) Hydrogen by 59. Conte F, Esposito S, Dal Santo V et al (2021) Flame pyrolysis
catalytic steam reforming of liquid byproducts from biomass synthesis of mixed oxides for glycerol steam reforming. Materi-
thermoconversion processes. Ind Eng Chem Res 41:4209–4215 als. https://​doi.​org/​10.​3390/​ma140​30652
41. Searcy E, Hess JR, Tumuluru J et al (2014) Optimization of bio- 60. Villa A, Gaiassi A, Rossetti I et al (2010) Au on M ­ gAl2O4 spi-
mass transport and logistics. In: Lecture Notes in Energy (LNEN, nels: the effect of support surface properties in glycerol oxida-
volume 17). pp 103–123 tion. J Catal 275:108–116. https://​doi.​org/​10.​1016/j.​jcat.​2010.​
42. Bakhtyari A, Makarem MA, Rahimpour MR (2019) Hydrogen 07.​022
production through pyrolysis. Fuel cells and hydrogen produc- 61. Balat M, Balat M (2009) Political, economic and environmen-
tion. Springer, New York, pp 947–973 tal impacts of biomass-based hydrogen. Int J Hydrogen Energy
43. He L, Yang J, Chen D (2013) Hydrogen from biomass: advances 34:3589–3603
in thermochemical processes. Renew Hydrogen Technol Prod 62. Bornapour M, Hooshmand RA, Khodabakhshian A, Parastegari
Purif Storage Appl Saf. https://​doi.​org/​10.​1016/​B978-0-​444-​ M (2017) Optimal stochastic coordinated scheduling of proton
56352-1.​00006-4 exchange membrane fuel cell-combined heat and power, wind
44. Davda R, Shabaker J, Huber G et al (2003) Aqueous-phase and photovoltaic units in micro grids considering hydrogen
reforming ofethylene glycol on silica-supported metal catalysts. storage. Appl Energy 202:308–322. https://​doi.​org/​10.​1016/j.​
Appl Catal B 43:13–26 energy.​2016.​10.​072
45. Suen N-T, Hung S-F, Quan Q et al (2017) Electrocatalysis for 63. Compagnoni M, Mostafavi E, Tripodi A et al (2017) Techno-
the oxygen evolution reaction: recent development and future economic analysis of a bioethanol to hydrogen centralized
perspectives. Chem Soc Rev 46:337–365 plant. Energy Fuels 31:12988–12996. https://​doi.​org/​10.​1021/​
46. Wang S, Lu A, Zhong CJ (2021) Hydrogen production from acs.​energ​yfuels.​7b024​34
water electrolysis: role of catalysts. Nano Convergence. https://​ 64. Compagnoni M, Tripodi A, Mostafavi E et al (2017) Hydrogen
doi.​org/​10.​1186/​s40580-​021-​00254-x production by steam reforming of bio-ethanol: Process design
47. Vrubel H, Hu X (2012) Molybdenum boride and carbide catalyze and economic assessment. In: DGMK Tagungsbericht
hydrogen evolution in both acidic and basic solutions. Angew 65. Dahman Y, Syed K, Begum S et al (2019) Biofuels: their char-
Chem Int Ed 51:12703–12706 acteristics and analysis. In: Verma D, Fortunati E, Zhang X
48. Bahadori E, Ramis G, Zanardo D et al (2020) Photoreforming (eds) Biomass, biopolymer-based materials, and bioenergy:
of glucose over CuO/TiO2. Catalysts 10:477. https://​doi.​org/​10.​ construction, biomedical, and other industrial applications.
3390/​catal​10050​477 Elsevier, New York, pp 277–325. https://​d oi.​o rg/​1 0.​1 016/​
49. Ramis G, Bahadori E, Rossetti I (2019) Photoreactors design for B978-0-​08-​102426-​3.​00014-X
hydrogen production. Chem Eng Trans 74:481–486. https://​doi.​ 66. Sindhu R, Binod P, Pandey A et al (2019) Biofuel production
org/​10.​3303/​CET19​74081 from biomass: toward sustainable development. In: Kumar S,
50. Ramis G, Bahadori E, Rossetti I (2021) Design of efficient photo- Kumar R, Pandey A (eds) Current developments in biotechnol-
catalytic processes for the production of hydrogen from biomass ogy and bioengineering: waste treatment processes for energy
derived substrates. Int J H2 Energy 46:12105–12116. https://d​ oi.​ generation. Elsevier, New York, pp 79–92. https://​doi.​org/​10.​
org/​10.​1016/j.​ijhyd​ene.​2020.​02.​192 1016/​B978-0-​444-​64083-3.​00005-1
51. Contreras JL, Salmones J, Colín-Luna JA et al (2014) Catalysts 67. Chakraborty R, Mukhopadhyay P (2020) Green Fuel Blending:
for H2 production using the ethanol steam reforming (a review). A Pollution Reduction Approach. In: Hashmi S, Choudhury
Int J Hydrogen Energy 39:18835–18853. https://​doi.​org/​10.​ IA (eds) Encyclopedia of Renewable and Sustainable Materi-
1016/J.​IJHYD​ENE.​2014.​08.​072 als. Elsevier, New York, pp 487–500. https://​doi.​org/​10.​1016/​
52. Mekhilef S, Saidur R, Safari A (2012) Comparative study of B978-0-​12-​803581-​8.​11019-7
different fuel cell technologies. Renew Sustain Energy Rev 68. Palma V, Ruocco C, Cortese M, Martino M (2020) Bioalcohol
16:981–989. https://​doi.​org/​10.​1016/J.​RSER.​2011.​09.​020 reforming: an overview of the recent advances for the enhance-
53. Song H (2012) Catalytic hydrogen production from bioethanol. ment of catalyst stability. Catalysts 10:665. https://​doi.​org/​10.​
Bioethanol. https://​doi.​org/​10.​5772/​20396 3390/​catal​10060​665
54. Sanchez N, Ruiz R, Hacker V, Cobo M (2020) Impact of bioetha- 69. Palanisamy A, Soundarrajan N, Ramasamy G (2021) Analysis
nol impurities on steam reforming for hydrogen production: a on production of bioethanol for hydrogen generation. Environ
review. Int J Hydrogen Energy 45:11923–11942. https://​doi.​org/​ Sci Pollut Res. https://​doi.​org/​10.​1007/​s11356-​021-​14554-6
10.​1016/J.​IJHYD​ENE.​2020.​02.​159

13
Topics in Catalysis

70. Compagnoni M, Tripodi A, Di Michele A et al (2017) Low reforming in Brazil. J Clean Prod 311:127577. https://d​ oi.o​ rg/1​ 0.​
temperature ethanol steam reforming for process intensifica- 1016/J.​JCLEP​RO.​2021.​127577
tion: new Ni/MxO-ZrO2 active and stable catalysts prepared by 87. Izurieta EM, Pedernera MN, López E (2019) Study of a thermally
flame spray pyrolysis. Int J Hydrogen Energy 42:28193–28213. integrated parallel plates reactor for hydrogen production. Chem
https://​doi.​org/​10.​1016/j.​ijhyd​ene.​2017.​09.​123 Eng Sci 196:344–353. https://​doi.​org/​10.​1016/J.​CES.​2018.​11.​
71. Rossetti I, Biffi C, Bianchi CL et al (2012) Ni/SiO2 and Ni/ 011
ZrO2 catalysts for the steam reforming of ethanol. Appl Catal 88. Izurieta EM, Adrover ME, Pedernera MN, Lópezlópez E (2018)
B 117–118:384–396. https://​doi.​org/​10.​1016/j.​apcatb.​2012.​02.​ Ethanol processor design for hydrogen production. Kinet Anal
006 Process Integr. https://​doi.​org/​10.​1021/​acs.​iecr.​8b023​24
72. Di Michele A, Dell’Angelo A, Tripodi A et al (2019) Steam 89. Spallina V, Matturro G, Ruocco C et al (2018) Direct route from
reforming of ethanol over Ni/MgAl2O4 catalysts. Int J Hydrogen ethanol to pure hydrogen through autothermal reforming in a
Energy. https://​doi.​org/​10.​1016/j.​ijhyd​ene.​2018.​11.​048 membrane reactor: experimental demonstration, reactor model-
73. Compagnoni M, Lasso J, Di Michele A, Rossetti I (2016) Flame- ling and design. Energy 143:666–681. https://​doi.​org/​10.​1016/J.​
pyrolysis-prepared catalysts for the steam reforming of ethanol. ENERGY.​2017.​11.​031
Catal Sci Technol 6:6257. https://​doi.​org/​10.​1039/​c5cy0​1958c 90. Moraes TS, Cozendey da Silva HN, Zotes LP et al (2019) A
74. Rossetti I, Lasso J, Finocchio E et al (2014) TiO2-supported cata- techno-economic evaluation of the hydrogen production for
lysts for the steam reforming of ethanol. Appl Catal A 477:42– energy generation using an ethanol fuel processor. Int J Hydrogen
53. https://​doi.​org/​10.​1016/j.​apcata.​2014.​03.​004 Energy 44:21205–21219. https://​doi.​org/​10.​1016/J.I​ JHYD​ENE.​
75. Rossetti I, Lasso J, Nichele V et al (2014) Silica and zirconia sup- 2019.​06.​182
ported catalysts for the low-temperature ethanol steam reforming. 91. Kiani MR, Meshksar M, Makarem MA, Rahimpour E (2021)
Appl Catal B 150–151:257–267. https://d​ oi.o​ rg/1​ 0.1​ 016/j.a​ pcatb.​ Catalytic membrane micro-reactors for fuel and biofuel pro-
2013.​12.​012 cessing: a mini review. Top Catal. https://​doi.​org/​10.​1007/​
76. Tripodi A, Compagnoni M, Rossetti I (2016) Kinetic model- s11244-​021-​01505-1
ling and reactor simulation for ethanol steam reforming. Chem- 92. Wang JH, Lee CS, Lin MC (2009) Mechanism of ethanol
CatChem 8:3804–3813 reforming: theoretical foundations. J Phys Chem C 113:6681–
77. Compagnoni M, Tripodi A, Rossetti I (2017) Parametric study 6688. https://​doi.​org/​10.​1021/​jp810​307h
and kinetic testing for ethanol steam refroming. Appl Catal B: 93. Sutton JE, Panagiotopoulou P, Verykios XE, Vlachos DG
Environ 203:899–909 (2013) Combined DFT, microkinetic, and experimental study
78. Rossetti I, Compagnoni M, Torli M (2015) Process simulation of ethanol steam reforming on Pt. J Phys Chem C 117:4691–
and optimization of H ­ 2 production from ethanol steam reforming 4706. https://​doi.​org/​10.​1021/​jp312​593u
and its use in fuel cells. 2. Process analysis and optimization. 94. Llera I, Mas V, Bergamini ML et al (2012) Bio-ethanol steam
Chem Eng J 281:1036–1044. https://​doi.​org/​10.​1016/j.​cej.​2015.​ reforming on Ni based catalyst. Kinetic study. Chem Eng Sci
08.​045 71:356–366. https://​doi.​org/​10.​1016/j.​ces.​2011.​12.​018
79. Tripodi A, Compagnoni M, Ramis G, Rossetti I (2017) Process 95. Graschinsky C, Laborde M, Amadeo N et al (2010) Ethanol
simulation of hydrogen production by steam reforming of diluted steam reforming over Rh (1%) M ­ gAl2O4/Al2O3: a kinetic study.
bioethanol solutions: Effect of operating parameters on electri- Ind Eng Chem Res 49:12383–12389
cal and thermal cogeneration by using fuel cells. Int J Hydrogen 96. Rossetti I, Compagnoni M, Finocchio E et al (2016) Syngas
Energy. https://​doi.​org/​10.​1016/j.​ijhyd​ene.​2017.​04.​056 production via steam reforming of bioethanol over Ni-BEA
80. Tripodi A, Compagnoni M, Bahadori E et al (2018) Process catalysts: a BTL strategy. Int J Hydrogen Energy 41:16878–
intensification by exploiting diluted 2nd generation bio-ethanol 16889. https://​doi.​org/​10.​1016/j.​ijhyd​ene.​2016.​07.​149
in the low-temperature steam reforming process. Top Catal. 97. Rossetti I, Lasso J, Compagnoni M et al (2015) ­H2 produc-
https://​doi.​org/​10.​1007/​s11244-​018-​1002-6 tion from bioethanol and its use in fuel-cells. Chem Eng Trans
81. Tripodi A, Pizzonia A, Bahadori E, Rossetti I (2018) Integrated 43:229–234. https://​doi.​org/​10.​3303/​CET15​43039
plant layout for heat and power cogeneration from diluted bioeth- 98. Ramis G, Rossetti I, Tripodi A, Compagnoni M (2017) Diluted
anol. ACS Sustain Chem Eng. https://​doi.​org/​10.​1021/​acssu​ bioethanol solutions for the production of hydrogen and ethyl-
schem​eng.​8b001​44 ene. Chem Eng Trans. https://​doi.​org/​10.​3303/​CET17​57278
82. Tripodi A, Bahadori E, Ramis G, Rossetti I (2019) Feasibility 99. Vizcaino A, Carrero A, Calles J (2007) Hydrogen production
assessment, process design and dynamic simulation for cogenera- by ethanol steam reforming over Cu–Ni supported catalysts. Int
tion of heat and power by steam reforming of diluted bioethanol. J Hydrogen Energy 32:1450–1461. https://​doi.​org/​10.​1016/j.​
Int J Hydrogen Energy 44:2–22. https://​doi.​org/​10.​1016/j.​ijhyd​ ijhyd​ene.​2006.​10.​024
ene.​2018.​02.​122 100. Wu Y-J, Santos JC, Li P et al (2014) Simplified kinetic model
83. Rossetti I, Tripodi A, Ramis G (2020) Hydrogen, ethylene and for steam reforming of ethanol on a Ni/Al2O3 catalyst. Canad
power production from bioethanol: ready for the renewable mar- J Chem Eng 92:116–130. https://​doi.​org/​10.​1002/​cjce.​21773
ket? Int J Hydrogen Energy. https://​doi.​org/​10.​1016/j.​ijhyd​ene.​ 101. Cunha AF, Wu YJ, Santos JC, Rodrigues AE (2012) Steam
2019.​07.​201 reforming of ethanol on copper catalysts derived from hydro-
84. Rossetti I, Biffi C, Forni L et al (2010) Integrated 5 kWe + 5 kWt talcite-like materials. Ind Eng Chem Res 51:13132–13143.
PEM-FC generator from bioethanol: a demonstrative project. In: https://​doi.​org/​10.​1021/​ie301​645f
ASME 2010 8th International Conference on Fuel Cell Science, 102. Zanchet D, Santos JBO, Damyanova S et al (2015) Toward
Engineering and Technology, FUELCELL 2010. pp 465–471 understanding metal-catalyzed ethanol reforming. ACS Catal
85. Seyam S, Dincer I, Agelin-chaab M (2021) Investigation of 5:3841–3863. https://​doi.​org/​10.​1021/​cs502​0755
potential fuels for hybrid molten carbonate fuel cell-based air- 103. Finocchio E, Rossetti I, Ramis G (2013) Redox properties of
craft propulsion systems. Energy Fuels 35:10156–10168. https://​ Co- and Cu-based catalysts for the steam reforming of etha-
doi.​org/​10.​1021/​acs.​energ​yfuels.​1c009​15 nol. Int J Hydrogen Energy 38:3213–3225. https://​doi.​org/​10.​
86. de Souza TAZ, Rocha DHD, Julio AAV et al (2021) Exergoen- 1016/j.​ijhyd​ene.​2012.​12.​137
vironmental assessment of hydrogen water footprint via steam 104. Nichele V, Signoretto M, Pinna F et al (2014) Ni/ZrO2 catalysts
in ethanol steam reforming: Inhibition of coke formation by

13
 Topics in Catalysis

CaO-doping. Appl Catal B 150–151:12–20. https://​doi.​org/​10.​ 121. Sander R (2015) Compilation of Henry’s law constants (ver-
1016/j.​apcatb.​2013.​11.​037 sion 4.0) for water as solvent. Atmos Chem Phys 15:4399–4981.
105. Nichele V, Signoretto M, Menegazzo F et al (2014) Hydrogen https://​doi.​org/​10.​5194/​acp-​15-​4399-​2015
production by ethanol steam reforming: effect of the synthesis 122. Young CL, Clever HL (1987) IUPAC Solubility Data Series -
parameters on the activity of Ni/TiO2 catalysts. Int J Hydrogen Methane, volume 27/28
Energy 39:4252–4258. https://​doi.​org/​10.​1016/j.​ijhyd​ene.​2013.​ 123. Wilhelm E, Battino R, Wilcock RJ (1977) Low-pressure solubil-
12.​178 ity of gases in liquid water. Chem Rev 77:219–262. https://​doi.​
106. Ramis G, Rossetti I, Finocchio E et al (2015) Metal disper- org/​10.​1021/​cr603​06a003
sion and interaction with the supports in the coke production 124. Valtz A, Chapoy A, Coquelet C et al (2004) Vapour-liquid equi-
over ethanol steam reforming catalysts. Progress Clean Energy libria in the carbon dioxide-water system, measurement and mod-
1:695–711 elling from 278.2 to 318.2 K. Fluid Phase Equilib 226:333–344.
107. Nichele V, Signoretto M, Pinna F et al (2015) Bimetallic Ni– https://​doi.​org/​10.​1016/j.​fluid.​2004.​10.​013
Cu catalysts for the low-temperature ethanol steam reforming: 125. Crovetto R, Fernández-Prini R, Japas ML (1982) Solubilities of
importance of metal-support interactions. Catal Lett 145:549– inert gases and methane in ­H2O and in ­D2O in the temperature
558. https://​doi.​org/​10.​1007/​s10562-​014-​1414-2 range of 300 to 600 K. J Chem Phys 76:1077–1086. https://​doi.​
108. Furtado AC, Alonso CG, Cantão MP, Fernandes-Machado org/​10.​1063/1.​443074
NRC (2009) Bimetallic catalysts performance during ethanol 126. Young CL (1981) IUPAC solubility data series: hydrogen and
steam reforming: influence of support materials. Int J Hydro- deuterium. Pergamon Press, Oxford
gen Energy 34:7189–7196. https://​doi.​org/​10.​1016/j.​ijhyd​ene.​ 127. https://​biorr​efine​r ia.​blogs​pot.​com/​2015/​08/​cellu​losic-​ethan​ol-​
2009.​06.​060 biore​finer​ies.​html
109. Xu J, Froment GF (1989) Methane steam reforming, methana- 128. Miltner A, Wukovits W, Pröll T, Friedl A (2010) Renewable
tion and water-gas shift: I. Intrinsic kinetics. AIChE J 35:88–96. hydrogen production: a technical evaluation based on process
https://​doi.​org/​10.​1002/​aic.​69035​0109 simulation. J Clean Prod 18:S51–S62. https://​doi.​org/​10.​1016/J.​
110. Mas V, Kipreos R, Amadeo N, Laborde M (2006) Thermody- JCLEP​RO.​2010.​05.​024
namic analysis of ethanol/water system with the stoichiometric 129. Carrara A, Perdichizzi A, Barigozzi G (2010) Simulation of an
method. Int J Hydrogen Energy 31:21–28. https://​doi.​org/​10.​ hydrogen production steam reforming industrial plant for ener-
1016/j.​ijhyd​ene.​2005.​04.​004 getic performance prediction. Int J Hydrogen Energy 35:3499–
111. Mas V, Bergamini ML, Baronetti G et al (2008) A kinetic study 3508. https://​doi.​org/​10.​1016/J.​IJHYD​ENE.​2009.​12.​156
of ethanol steam reforming using a nickel based catalyst. Top 130. Lang P, Denes F, Hegely L (2017) Comparison of different amine
Catal 51:39–48. https://​doi.​org/​10.​1007/​s11244-​008-​9123-y solvents for the absorption of C ­ O2. Chem Eng Trans 61:1105–
112. Rossetti I, Compagnoni M, Torli M (2015) Process simulation and 1110. https://​doi.​org/​10.​3303/​CET17​61182
optimisation of H ­ 2 production from ethanol steam reforming and 131. Yang S-I, Choi D-Y, Jang S-C et al (2008) Hydrogen separation
its use in fuel cells. 1. Thermodynamic and kinetic analysis. Chem by multi-bed pressure swing adsorption of synthesis gas. Adsorp-
Eng J 281:1024–1035. https://​doi.​org/​10.​1016/j.​cej.​2015.​08.​025 tion 14:583–590. https://​doi.​org/​10.​1007/​s10450-​008-​9133-x
113. Akpan E, Akande A, Aboudheir A et al (2007) Experimental, 132. Lopes FVS, Grande CA, Ribeiro AM et al (2009) Adsorption of
kinetic and 2-D reactor modeling for simulation of the produc- ­H2, ­CO2, ­CH4, CO, N ­ 2 and H­ 2O in activated carbon and zeolite
tion of hydrogen by the catalytic reforming of concentrated crude for hydrogen production. Sep Sci Technol 44:1045–1073. https://​
ethanol ( CRCCE ) over a Ni-based commercial catalyst in a doi.​org/​10.​1080/​01496​39090​27291​30
packed-bed tubular reactor. Chem Eng Sci 62:3112–3126. https://​ 133. Tripodi A, Pizzonia A, Bahadori E, Rossetti I (2018) Integrated
doi.​org/​10.​1016/j.​ces.​2007.​03.​006 plant layout for heat and power cogeneration from diluted bioeth-
114. Akande A, Aboudheir A, Idem R, Dalai A (2006) Kinetic mod- anol. ACS Sustain Chem Eng 6:5358–5369. https://​doi.​org/​10.​
eling of hydrogen production by the catalytic reforming of crude 1021/​acssu​schem​eng.​8b001​44
ethanol over a co-precipitated Ni–Al2O3 catalyst in a packed bed 134. Aden a, Ruth M, Ibsen K, et al (2002) Lignocellulosic biomass to
tubular reactor. Int J Hydrogen Energy 31:1707–1715. https://d​ oi.​ ethanol process design and economics utilizing co-current dilute
org/​10.​1016/j.​ijhyd​ene.​2006.​01.​001 acid prehydrolysis and enzymatic hydrolysis for corn stover.
115. Patel M, Jindal TK, Pant KK (2013) Kinetic study of steam National Renewable Energy Laboratory-NREL Medium: ED;
reforming of ethanol on Ni-based ceria-zirconia catalyst. Ind Eng Size, p 154. NREL/TP-510-32438
Chem Res 52:15763–15771. https://​doi.​org/​10.​1021/​ie401​570s 135. Dias MOS, Ensinas AV, Nebra SA et al (2009) Production
116. Sahoo DR, Vajpai S, Patel S, Pant KK (2007) Kinetic modeling of bioethanol and other bio-based materials from sugarcane
of steam reforming of ethanol for the production of hydrogen bagasse: integration to conventional bioethanol production
over Co/Al2O3 catalyst. Chem Eng J 125:139–147. https://​doi.​ process. Chem Eng Res Des 87:1206–1216. https://​doi.​org/​10.​
org/​10.​1016/j.​cej.​2006.​08.​011 1016/J.​CHERD.​2009.​06.​020
117. Patel MP, Shah N, Ashe R (2011) Robust optimisation method- 136. Pardo-Planas O, Atiyeh HK, Phillips JR et al (2017) Process
ology for the process synthesis of continuous technologies. In: simulation of ethanol production from biomass gasification and
Computer Aided Chemical Engineering. pp 351–355 syngas fermentation. Biores Technol 245:925–932. https://​doi.​
118. Tripodi A, Compagnoni M, Martinazzo R et al (2017) Process org/​10.​1016/j.​biort​ech.​2017.​08.​193
simulation for the design and scale up of heterogeneous catalytic 137. Xuan J, Leung MKH, Leung DYC, Ni M (2009) A review of
process: kinetic modelling issues. Catalysts. https://​doi.​org/​10.​ biomass-derived fuel processors for fuel cell systems. Renew
3390/​catal​70501​59 Sustain Energy Rev 13:1301–1313. https://​doi.​org/​10.​1016/J.​
119. Wu YJ, Li P, Yu JG et al (2013) Sorption-enhanced steam reform- RSER.​2008.​09.​027
ing of ethanol on NiMgAl multifunctional materials: experimen-
tal and numerical investigation. Chem Eng J 231:36–48. https://​ Publisher's Note Springer Nature remains neutral with regard to
doi.​org/​10.​1016/j.​cej.​2013.​06.​125 jurisdictional claims in published maps and institutional affiliations.
120. Choi Y, Stenger HG (2003) Water gas shift reaction kinetics and
reactor modeling for fuel cell grade hydrogen. J Power Sources
124:432–439. https://​doi.​org/​10.​1016/​S0378-​7753(03)​00614-1

13