Fuel Processing Technology 211 (2021) 106572

Contents lists available at ScienceDirect

Fuel Processing Technology

journal homepage: www.elsevier.com/locate/fuproc

Research article

Combined ZnO reduction and methane reforming for co-production of pure T

Zn and syngas in a prototype solar thermochemical reactor
Srirat Chuayboona,b, Stéphane Abanadesa,
⁎

a
Processes, Materials and Solar Energy Laboratory, PROMES-CNRS, 7 Rue du Four Solaire, 66120 Font-Romeu, France
b
Department of Mechanical Engineering, King Mongkut's Institute of Technology Ladkrabang, Prince of Chumphon Campus, Chumphon 86160, Thailand

ARTICLE INFO ABSTRACT

Keywords: Solar thermochemical conversion processes offer a promising avenue for storing intermittent solar energy into
Solar thermochemical high value chemical products. A directly-irradiated packed-bed reactor was developed to experimentally in­
Concentrated sunlight vestigate the solar thermochemical methane reforming combined with ZnO reduction for co-production of Zn
Solar reactor and syngas in a single process. On-sun experiments were conducted to demonstrate reliable process operation
Methane reforming
and to identify optimal operating conditions, considering the impact of reduced and atmospheric pressures,
Zinc
temperatures, and inlet methane flow-rates. A pressure decrease enhanced net ZnO conversion at the expense of
Syngas
lowered syngas selectivity attributed to the CO2 increase because of insufficient gas residence time. Increasing
temperature promoted syngas production rate, yield, ZnO and methane conversion at the expense of favored
methane cracking which can be alleviated by lowering pressure. The maximum H2 and CO yields of 28.4 and
4.6 mmol/gZnO, and the maximum net ZnO conversion and methane conversion (67.3% and 57.6%) were
achieved from non-isothermal tests. The calorific value of chemical products was upgraded by solar energy (U up
to 1.31), and ηsolar-to-chemical = 3.9% was achieved. Pure and highly reactive metallic Zn was produced with well
crystallized structure in micrometric size, demonstrating the feasibility of combined solar methane reforming
and ZnO reduction for synthetic fuel production and sustainable Zn metallurgy.

1. Introduction around 3.6 kgCO2-eq/kgZn, with values ranging from 2.8 to 4.9 kgCO2-eq/
kgZn [22].
Solar energy is a plentiful, clean, and sustainable energy source, To alleviate the issues associated with CO2 emissions, solar ther­
with important potential contribution to protect the environment and mochemical dissociation of zinc oxide (ZnO) towards Zn production has
climate. When concentrated, it can provide process heat to drive highly been extensively proposed [23–29]. ZnO can be thermally reduced to
endothermic thermochemical conversion processes such as two-step Zn vapor with the release of oxygen using concentrated solar energy at
H2O/CO2 splitting [1–5], gasification [6–11], carbothermal metallurgy temperature above 2000 K [25], according to Eq. (1). In addition,
[12], methane cracking [13–16], and methane reforming [17–20]. produced Zn can be exothermically oxidized with H2O and/or CO2 at
These processes offer an effective and promising pathway in capturing lower temperatures to produce hydrogen (H2) and/or carbon monoxide
and converting intermittent solar energy into long term storable che­ (CO) and ZnO, which can be recycled back to the reduction step (Eq.
mical energy carriers such as syngas and metal commodities [21]. (1)) [30–35], according to Eq. (2).
Zinc (Zn) has been regarded as one of the most widely used metal ZnO(s) + solar heat Zn(g) + ½O2 (1)
commodities in various industrial sectors such as automotive, aero­
space, food, pharmaceutical, galvanizing, electrical batteries or fuel Zn(s) + H2 O(g) (CO2) ZnO(s) + H2 (CO) (2)
cells, electronics, and mechanical engineering [12]. Conventionally, Zn
Abanades et al. [24] designed and simulated a solar reactor for the
is extracted from ores via smelting and refining processes, which dis­
direct ZnO decomposition using real concentrated solar energy. They
charge large amount of greenhouse gas emissions, especially CO2,
found that the reaction extent strongly depends on temperature of the
thereby contributing to climate changes. Global carbon footprint of Zn
absorbing particles, and complete reaction was achieved at temperature
caused by conventional metallurgical processes regarding pyr­
above 2200 K for a 1 μm initial particle diameter. The thermal
ometallurgy, hydrometallurgy, and electrometallurgy was averaged

⁎
Corresponding author.
E-mail address: [email protected] (S. Abanades).

https://doi.org/10.1016/j.fuproc.2020.106572
Received 30 April 2020; Received in revised form 12 August 2020; Accepted 15 August 2020
0378-3820/ © 2020 Elsevier B.V. All rights reserved.
S. Chuayboon and S. Abanades Fuel Processing Technology 211 (2021) 106572

dissociation of ZnO was also tested in solar reactors including rotating- reactant, while the others are secondary because they involve sec­
cavity receivers [23,25,36]. Furthermore, various concepts such as ondary species (reaction 3 is also secondary as solid carbon is a by-
moving-front [27], beam-down [28], and 100 kWth scale [37] solar product not linked to the presence of ZnO).
reactors were designed, fabricated, and assembled to facilitate the ZnO The advantages of the combined ZnO reduction and CH4 reforming
dissociation, and reactor performances were found to depend on reactor are: (i) both syngas and metallic Zn are produced in a single process, (ii)
concepts, designs and technologies. Regarding two-step H2O/CO2 syngas can be generated with a H2/CO ratio of two, suitable for me­
splitting thermochemical cycle based on ZnO/Zn for H2/CO production, thanol synthesis and compatible with the downstream conversion by
Abanades [34] discovered that Zn nano-powder synthesized from solar Fischer-Tropsch for synthetic liquid fuels, (iii) the use of catalysts is not
thermal ZnO dissociation (obtained after condensation of Zn vapor) required, bypassing catalysts deactivation issue, (iv) methane is re­
reacted effectively with both CO2 and H2O with rapid kinetics. Zn re­ formed without using an excess in oxidizer, thus lowering energy re­
combination with oxygen during the solar step usually required effi­ quirements as compared to conventional methane reforming
cient gas quenching to minimize this undesired reverse reaction and (H2O:CH4 ≥ 3) [18]. Likewise, syngas production via solar chemical-
remained a serious problem of the direct ZnO dissociation cycle [37]. looping methane reforming using non-volatile oxides such as ceria
Even though ZnO can be thermally dissociated via solar heat, high [41,42], iron oxides [43,44], and perovskites [45,46] can be considered
operating temperatures exceeding 2000 K have hindered commercial as an alternative route since non-volatile oxides remain in the solid
realization. Alternatively, ZnO reduction can be conducted at con­ state over the whole process while releasing only oxygen from their
siderably lower temperatures (above 1223 K) by employing carbonac­ structure, thereby circumventing the recombination issue. Never­
eous materials as reducing agents to reduce metal oxides and to pro­ theless, they exhibit drawbacks of non-stoichiometric reactions, which
duce metals [38]. ZnO can react with a carbonaceous feedstock to results in lower oxygen storage/release capacity leading to low fuel
produce Zn and CO (Eq. (3)). Accordingly, using carbon as reducing productivity (ceria) and weak thermal stability (iron oxides).
agent alleviates recombination of the Zn vapor with oxygen products Steinfeld et al. [47] examined the thermal ZnO reduction and CH4
[39]. reforming, and they reported that CH4 conversion and H2 yield in­
creased with increasing temperature. The extent of reduction rose
ZnO(s) + C Zn(g) + CO ( H 0 = 370. 4 kJ/mol) (3)
consistently with CH4 concentration, and the activation energy of
Osinga et al. [40] employed beech charcoal, activated charcoal, and 146 kJ/mol was obtained. Combined ZnO reduction and CH4 reforming
petroleum coke as reducing agents to reduce ZnO using a 5 kWth two- was performed in a solar reactor using a solar furnace between 1000
cavity solar reactor and reported that thermal efficiency of 20% and Zn and 1600 K under atmospheric pressure [48], yielding ZnO conversion
production rate of 0.16 mol/min were achieved at 1150 K. Then, the up to 90%. Thermodynamic analysis of ZnO/Zn based CH4 reforming
same group experimentally investigated ZnO carbothermal reduction and H2O splitting was investigated [49]. The CH4/ZnO ratio and water
with beech charcoal using a 300 kW two-cavity solar chemical pilot splitting temperature were varied. The ZnO reduction was possible at
plant operated in batch mode at 1300-1500 K. Zn production of 50 kg/h 1200 K thanks to CH4 reducing agent, and increasing the CH4/ZnO ratio
with 95% purity (2.5–5 μm mean particle size) was obtained, and the from 0.1 to 1.0 influenced the Zn and syngas production.
thermal efficiency was 30%. In addition, operating ZnO reduction under vacuum pressure and
Another attractive solar-driven approach, which proceeds at mod­ introducing inert gas for dilution are thermodynamically favorable
erate temperatures (above 1123 K), is ZnO reduction with methane. The routes, which in turn results in lowered operating temperature [50–52]
endothermic reactions involve both partial oxidation of methane to and enhanced reaction kinetics [53]. Brkic et al. [39] tested vacuum
form syngas and simultaneous ZnO reduction to yield Zn, thereby carbothermal reduction of aerosolized ZnO particles in a drop-tube
producing both Zn and syngas in a single process, according to Eq. (4). reactor at 1–960 mbar, and demonstrated that Zn production was
greater at 100 mbar compared with ambient pressure. Reaction at
ZnO(s) + CH 4 Zn(g) + CO + 2H2 ( H 0 = 444. 9 kJ/mol) (4)
1 mbar was hindered because of lowered particles residence time and
However, possible secondary side reactions may occur during ZnO inefficient heat transfer. Most of previous studies dealt with ZnO re­
reduction with methane according to (Eqs. (5)–(10)): duction with solid carbon. The combination of methane reforming with
CH 4 C + 2H2 (5) ZnO reduction has been scarcely addressed and the effect of operating
pressure has not been considered so far. Moreover, the influence of the
CH 4 + CO2 2CO + 2H2 (6) operating conditions (temperature, pressure, and CH4 flow-rate) on
products yields has not been studied. Prior studies on the ZnO reduction
CH 4 + H2 O CO + 3H2 (7)
with CH4 were focused on thermodynamics [49,54] and experiments
C + CO2 2CO (8) [48,55–57] at atmospheric pressure only. Therefore, the present work
aims to further explore the combined CH4 reforming and ZnO reduction
C + H2 O CO + H2 (9) in a 1.5 kWth prototype solar reactor operated under both vacuum and
CO2 + H2 CO + H2 O (10) atmospheric pressures using real concentrated sunlight. This reactor
was specifically designed for testing the process under vacuum pressure
Other possible reactions are ZnO reduction with carbon (Eq. (3)), condition [58]. Likewise, isothermal and non-isothermal on-sun ex­
CO (Eq. (11)), or H2 (Eq. (12)), yielding Zn with CO, CO2, or H2O. periments were realized to address a challenge in controlling the pro­
ZnO + CO Zn + CO2 (11) cess under transient solar radiation conditions. Moreover, a parametric
study was carried out in order to experimentally investigate the impact
ZnO + H2 Zn + H2 O (12) of process variables regarding operating pressures (0.15–0.90 bar),
Therefore, the overall side reaction of ZnO with methane yielding temperatures (isothermal tests: 800–1000 °C, non-isothermal tests:
Zn, CO2, and H2O can be represented as: 610–1171 °C), and methane flow-rates (0.1–0.3 NL/min affecting the
solid-gas contact time) on the performance of methane-driven ZnO re­
4ZnO + CH 4 4Zn + CO2 + 2H2 O (13)
duction in a prototype packed-bed solar reactor. The results obtained
The contact between ZnO and CH4 yields a gas/solid distribution of under realistic conditions provided insights into the process feasibility,
products, including mainly Zn, CO, H2, but also C, CO2, H2O, besides reliability and scalability of solar methane-driven reduction of ZnO for
the unconverted reactants, and the potential reactions (Eqs. (4)–(13)) co-production of Zn and syngas.
explain such a distribution. Among these reactions, the main reactions
(Eqs. (4), (11), (12), (13)) are dominant because they involve ZnO

2
S. Chuayboon and S. Abanades Fuel Processing Technology 211 (2021) 106572

Fig. 1. Schematic diagram of solar furnace and complete experimental set up.

2. Experimental setup and methods bed at the middle of the ZnO layer (T1) is measured by a type-B ther­
mocouple inserted inside the cavity receiver. It is compared with the
Fig. 1 shows schematically the directly-irradiated solar chemical temperature (Tpyrometer) measured by a solar-blind pyrometer (operating
reactor associated with a real solar concentrating system. The reactor at 4.8–5.2 μm in a H2O absorption band) through a fluorine window
consists of a cylindrical cavity receiver made of alumina (Tmax: 1800 °C) (CaF2 with transmission above 95% in the IR wavelength measurement
with a 121.4 cm3 inner cavity volume (48 mm I.D., 58 mm O.D., and range), which points directly at the upper surface of ZnO powder bed
70 mm height), and it is designed particularly for methane-driven ZnO directly exposed to radiation. The reactor cavity pressure (P) is mea­
reduction under vacuum pressure condition. It features rapid solar sured via a pressure transducer.
heating thanks to a suitable cavity volume and accommodates fluc­ Experiments were conducted at the 1.5 kWth vertical-axis solar
tuations in the transient DNI. The cavity receiver is enveloped with a furnace of PROMES-CNRS laboratory, Odeillo, France. It consists of a
cylindrical porous ceramic alumino-silicate insulation layer (40 mm- primary 4-m2 automatic sun-tracking heliostat delivering sunlight ver­
thick). Its bottom center is drilled for the passage of the inlet tube with tically towards the secondary parabolic concentrator (0.85 m focal
a 4 mm internal diameter, where a gas mixture of reactive methane and distance, 10.5 MW/m2 peak flux density for a DNI value of 1 kW/m2).
inert carrier N2 is fed towards the reaction zone. The concentrated solar power is delivered to the reactor, and its flux
To support ZnO powder, a small amount of inert alumina particles intensity is manually controlled by the shutter positioned between the
(2 mm diameter, 10 mm bed height) is positioned at the cavity bottom heliostat and the concentrator (Fig. 1). Prior to on-sun experiments,
above an alumina wool layer (10 mm thickness). These bottom layers well grinded ZnO powders (99.8% chemical purity, 1–5 μm particle
also broaden the inlet gas stream, which promotes the gaseous reactant size, purchased from PROLABO) are placed in the reactor cavity re­
flow distribution along the cross section of the ZnO packed-bed. The ceiver, which is directly exposed to the solar radiation. The ZnO powder
insulated alumina cavity receiver is placed inside the cylindrical bed is uniformly spread at the cavity bottom (48 mm in diameter cy­
stainless-steel reactor shell (170 mm O.D., 150 mm I.D., and 151 mm linder) above the supporting alumina particles. Given the ZnO bed loose
height, volume: 2.67 L) cooled by a water circulation system. Its top is bulk density (~0.5–0.6 g/cm3), the bed height is in the range
covered with an alumina cap with a small aperture (17 mm diameter) to 1.8–2.2 mm (for 2 g ZnO). The low amount of ZnO allows avoiding
capture the incident concentrated solar energy inside the cavity and let steep thermal gradient across the bed height. After completely loading
incident radiation entering directly towards the reaction site while ZnO, the reactor is flushed with N2 flow for 2 min to purge the residual
minimizing radiation heat losses. The front reactor is enclosed by a air in the system, reflected by negligible amount of oxygen (< 10 ppm)
hemispherical transparent Pyrex window. The temperature inside the from outlet gas analysis. Subsequently, it is heated progressively by

3
S. Chuayboon and S. Abanades Fuel Processing Technology 211 (2021) 106572

Fig. 2. Photograph of the 1.5 kWth prototype solar reactor during on-sun testing conducted at the PROMES-CNRS solar furnace: (1) parabolic concentrator, (2)
pyrometer, (3) transparent window, (4) water-cooled reactor outlet port (zone A), (5) ceramic filter (zone B), (6) thermocouple type B, (T1), (7) water-cooled reactor
shell, (8) inlet tube where a mixture of carrier N2 and CH4 is injected, (9) tubes where protective N2 is injected, (10) filter outlet connected to pumping system.

highly concentrated sunlight. gas composition was quantitatively analyzed by an on-line gas analyzer
During heating, the reactor window is actively cooled and kept clear (using infrared detectors for CH4, CO2, and CO and thermal con­
of particles deposition by utilizing the protective N2 flow (2 NL/min), ductivity detector for H2, precision: 0.5% of full scale). The molar flow-
which is injected to the window area via two stainless steel tubes (6 mm rate of gas product species (Fi) was calculated using their outlet mole
O.D., 4 mm I.D.) vertically inserted through the insulation. Protective fraction (yi) and the inlet N2 molar flow-rate (FN2) as reference:
N2, carrier N2, and reactive methane gas flow-rates are electronically Fi = FN2 ∙ yi/yN2. X-ray powder diffraction analysis (XRD) was per­
controlled by mass flow controllers (Brooks, model SLA5850S, range formed for phase identification in the Bragg Brentano geometry using
0–5 NL/min ± 0.2% of full scale). Once approaching the desired tem­ Cu Kα radiation (αCu = 1.5418 Å, angular range = 20–100° in 2-
peratures (800, 900, and 1000 °C), the reactor is evacuated by a va­ Theta, step size of 0.02°, recording time = 2 s, Philips PW 1820 dif­
cuum pump to reduce its total pressure (either 0.15 or 0.45 bar), while fractometer) to characterize the solid products collected from zone A
at atmospheric pressure (0.90 bar) the vacuum pump is not used. When and zone B for all collected samples. Field emission scanning electron
pressure and temperature are in a steady state, methane (99.999% microscopy (FESEM, Hitachi S4800) was utilized to characterize the
chemical purity) is subsequently fed (0.1–0.3 NL/min) along with particles morphology. In addition, the reactivity of Zn product was
carrier N2 (0.2 NL/min) into the packed-bed of ZnO. Fig. 2 shows a examined using thermogravimetry analysis (TGA, Netzsch STA449).
photograph of the 1.5 kWth prototype solar reactor and its external To assess the quality of the syngas products, the H2 and CO se­
components during on-sun testing. Solar thermal power input was ad­ lectivity is quantified according to Eqs. (14) and (15):
justed in the range 0.5–0.9 kW by the degree of shutter opening, de­
n H2
pending on the required operating temperature and pressure, and DNI SH2 =
for each experimental day was stable in the range 950–1050 W/m2. n H2 + n H2 O (14)
At the high reactor temperature, solid ZnO was reduced to Zn vapor
that was continuously transported to the outlet by the carrier gas flow, nCO
SCO =
thereby enabling complete ZnO consumption and Zn extraction from nCO + nCO2 (15)
the reactor cavity (continuous products removal promotes the reaction
completion). Metallic Zn vapor and syngas exited the reactor via the where ni is the molar amount of species i, calculated from time in­
outlet port and then flowed through a water-cooled alumina tube (zone tegration of the gas production rates (Fi) over the experiment duration.
A) with a 20 mm in-diameter where Zn vapor primarily condensed. In The net ZnO conversion (XZnO), equivalent to Zn production yield, is
downstream, the product particles regarding condensed Zn particles the net ZnO fraction converted to Zn after Zn recombination. Oxygen is
and low amount of soot were trapped by a ceramic filter (zone B). Then, extracted from the ZnO and then recovered in the forms of CO, CO2, and
the syngas after exiting the ceramic filter was cleaned again with a H2O, as shown in Eqs. (4) and (13). Therefore, XZnO is determined based
micro filter (pore diameter: 0.1 μm) prior to gas analysis. The produced on outlet gas analysis using an oxygen mass balance (Eq. (16)):

4
S. Chuayboon and S. Abanades Fuel Processing Technology 211 (2021) 106572

nCO + 2nCO2 + n H2 O
XZnO =
nZnO (16)
To determine how much inlet methane is converted to syngas, solid
carbon, or soot, the CH4 conversion (XCH4) is calculated according to
Eq. (17):
munreacted CH4
XCH4 = 1
mCH4 (17)
where XCH4 is the methane conversion, is the mass flow-rate of un­
reacted methane in the off-gas, and mCH4 is the mass flow-rate of in­
jected methane.
To evaluate the amount of initial carbon in injected methane com­
pared with the total carbon in the outlet gas flow, the carbon mass
balance (Xc) is calculated (Eq. (18)):
t t t
FCO (t ) dt + FCO2 (t ) dt + Funreacted CH4 (t ) dt
0 0 0
XC = t
Finjected CH4 (t ) dt Fig. 3. ∆G° of ZnO reduction reactions as a function of temperature at 1 bar.
0 (18)
Fi represents the molar flow-rate of species i (mol/s).
The solar reactor performance indicators are characterized by the
energy upgrade factor and solar-to-chemical energy conversion effi­
ciency.
The energy upgrade factor (U) represents the ratio of the total
chemical energy content in the outlet flow to the energy content in the
inlet flow (Eq. (19)).

(m H2 LHVH2 + mCO LHVCO ) + ((1 XCH4 ) mCH4 LHVCH4 )

+ (mZn HZnO Zn + 0.5O2 )
U=
(mCH4 LHVCH4 ) (19)
where m , LHV, and ∆H are the mass flow-rate (kg/s), Lower Heating
Value (J/kg), and standard enthalpy change equivalent to the LHV of
Zn (J/kg), respectively.
The solar-to-chemical energy conversion efficiency
(ηsolar−to−chemical) is defined as the chemical energy content of products
divided by the total energy input (Eq. (20)).
(m H2 LHVH2 + mCO LHVCO ) + (mZn HZnO Fig. 4. Thermodynamic equilibrium products distribution of the ZnO + CH4
Zn + 0.5O2 )
solar to chemical = system at 1 bar.
Qsolar + (XCH4 mCH4 LHVCH4 )
(20)
675 °C, respectively.
Qsolar is the total solar power input (W). Fig. 4 presents the thermodynamic equilibrium product composi­
The energy content of carbon was not considered because it is a side tions of an equimolar mixture of ZnO + CH4 at 1 bar, calculated by
product that is not targeted. Because both the solar power input and Gibbs free energy minimization (HSC Chemistry). Higher hydrocarbons
products flow-rate change with time, the efficiency was calculated (CnHm) were included in the calculation but their amount is negligible
using integrated data over the experiment duration, for determining at equilibrium. At ambient temperature, both ZnO and CH4 are ther­
both the total solar energy input and the energy content of products modynamically stable. Then, at above 400 °C, CH4 is decomposed to C
given that the total amounts of H2, CO, Zn products were quantified. and H2(g). Between 600 and 800 °C, ZnO(s) is reduced to Zn (in con­
densed and gaseous states), while C is gasified with CO2 and H2O to
3. Results and discussion form H2 and CO. The reduction approaches completion above 800 °C
and then reaches it above 1000 °C, consisting of a single gas phase
3.1. Thermodynamic equilibrium analysis containing Zn vapor and a mixture of H2 and CO with a H2/CO ratio of
two. The low quantity of CO2 and H2O could actually be the result of
Fig. 3 shows the Gibbs free energy (∆G°) of the possible reduction their consumption by the reaction of solid C with CO2 and H2O yielding
reactions of ZnO with or without the aid of carbonaceous materials as CO and H2. The other anticipated mechanism consisting of CH4 de­
reducing agents at 1 bar. Overall, ∆G° declines with increasing tem­ composition followed by reaction of C with ZnO is also realistic.
perature, and all the considered reactions become thermodynamically Thermodynamic equilibrium products distribution of the
favorable. On the one hand, the ∆G° of the direct ZnO dissociation ZnO + CH4 system at temperatures of 700 and 750 °C as a function of
approaches zero at above 2060 °C. On the other hand, when using pressure (0–1 bar) is shown in Fig. 5. The temperatures of 700 and
carbonaceous materials as reducing agents, the required temperature 750 °C were selected because they illustrate well the transition between
drops considerably. For example, ZnO + C and ZnO + CH4 tempera­ reactants and products (Zn + CO + 2H2) when varying the pressure.
tures at ∆G° equal to zero are 948 and 840 °C, respectively. The for­ Indeed above 800 °C, the reaction is complete and the gas products are
mation of H2O can theoretically be achieved by ZnO + H2 at above the only predicted species at equilibrium. Overall, decreasing pressure
1215 °C. ZnO + CO, 2ZnO + C, C + CO2 (Boudouard), and C + H2O enables the reduction of ZnO with CH4 to reach completion. Regardless
reactions can take place at temperatures above 1317, 1053, 700, and

5
S. Chuayboon and S. Abanades Fuel Processing Technology 211 (2021) 106572

Fig. 5. Thermodynamic equilibrium products distribution of the ZnO + CH4 system as a function of pressure at 700 and 750 °C. H2O and CO2 were omitted because
in negligible amounts.

reactor pressure (P) = 0.15–0.90 bar, total N2 flow-rate = 2.2 NL/min,

reactor temperature (T1) = 800–1000 °C (isothermal tests) and
610–1171 °C (non-isothermal tests), and solar power input
(Q solar) = 0.48–0.93 kW(thermal).
Overall, the H2 and CO selectivity was found in the range of
0.50–0.94 and 0.35–0.83, respectively, indicating higher H2 produc­
tion. XZnO, XCH4, and Xc in the ranges of 18.3–67.3%, 7.4–57.6%, and
61.0–96.1%, respectively, were accomplished. ηsolar-to-chemical and U va­
lues were in the ranges of 1.9–3.9% and 0.90–1.31, respectively. ηsolar-to-
chemical was not high due to sensible heat losses resulting from long
experimental duration and vacuum pressure condition. Almost all U
values exceeded one and some of them approached the theoretical
maximum U of 1.39 (calculated from Eq. (4)), thereby confirming ef­
ficient solar energy storage into chemical fuels.

3.3. Temperature and pressure influence

Fig. 6. Stability (P, T) domain of ZnO showing the boundary at which ZnO is On-sun combined ZnO reduction and CH4 reforming was examined
fully converted at equilibrium.
at different isothermal temperatures (800, 900, and 1000 °C) and
pressures (0.15, 0.45, and 0.90 bar). The introduction of the mixed
of the pressure, CH4 is decomposed to 2H2 and C, while the pressure flow-rate of reactive CH4 (0.2 NL/min) and inert carrier N2 (0.2 NL/
decrease results in the shift of the main species from (ZnO + C) to (Zn min), which results in 50% inlet CH4 concentration, was set. The solid-
(g) + CO) according to the carbothermal reduction process (Eq. (3)). In gas contact time was calculated as ~0.6 s (also referred as space time),
addition, when decreasing the operating temperature, the required taking into account the initial ZnO bed volume and the total inlet gas
pressure to reach completion is lowered. For example, the reaction flow-rate (CH4 + N2 at entrance conditions). This space time in the
reaches completion at pressure below 0.15 bar at 700 °C compared to reacting layer refers to the initial state since the solid-gas contact time
below 0.4 bar at 750 °C. Fig. 6 plots the ZnO boundary point at which then decreases as the ZnO bed is consumed continually. The dynamic
ZnO totally disappears (is fully converted) at any temperature, which species production rates from ZnO reduction and CH4 reforming (2 g
highlights the benefit of decreasing pressure for lowering the tem­ ZnO particles) are plotted for temperatures of 800, 900, and 1000 °C at
perature of complete ZnO conversion. both low pressure (0.15 bar, Fig. 7a, c, e) and atmospheric pressure
(0.90 bar, Fig. 7b, d, f).
3.2. Reactor performance assessment As shown in Fig. 7, differences due to the effect of both tempera­
tures and pressures are obvious. When the flow of methane was in­
Table 1 lists the operational conditions and experimental results for itiated, CO2, CO, and H2 were produced and reached a peak at initial
combined ZnO reduction and methane reforming for 12 on-sun tests. state, and they then declined progressively until reaching reaction
Experiments were carried out with the following parameter ranges: completion, reflected by CO and CO2 approaching zero (methane
initial ZnO mass = 2–5 g, inlet methane flow-rate = 0.1–0.3 NL/min, cracking can still occur even after complete ZnO reduction: H2 thus

6
S. Chuayboon and S. Abanades Fuel Processing Technology 211 (2021) 106572

Table 1
Summary of operational parameters and solar reactor performance assessment.
Run No. initial CH4 P T1 Qsolar H2 yield CO yield CO2 yield SH2 SCO XZnO XCH4 Xc U ( ± 2%) ηsolar-to-chemical
ZnO flow- (bar) (°C) (kW) (mmol/gZnO) (mmol/gZnO) (mmol/gZnO) ( ± 2%) ( ± 2%) ( ± 2%) ( ± 2%) ( ± 2%) ( ± 2%)
mass rate ( ± 2%) ( ± 2%) ( ± 2%)
(g) (NL/
min)

1 2 0.1 0.15 800 0.52 3.74 0.55 0.93 0.67 0.37 34.8 7.4 96.1 1.10 1.9
2 2 0.1 0.90 800 0.48 4.63 0.87 0.34 0.87 0.72 18.3 11.2 92.4 1.11 2.0
3 2 0.1 0.15 900 0.63 3.31 0.92 1.49 0.53 0.38 55.8 16.0 93.7 1.12 2.3
4 2 0.1 0.45 900 0.59 5.21 1.61 1.10 0.70 0.60 48.8 16.6 93.3 1.12 2.4
5 2 0.1 0.90 900 0.58 7.03 2.04 0.89 0.80 0.70 45.8 17.3 91.8 1.09 2.4
6 2 0.2 0.90 900 0.62 11.57 2.55 0.81 0.88 0.76 47.0 23.4 85.1 1.02 3.3
7 2 0.2 0.90 900 0.63 10.66 1.71 0.68 0.89 0.71 36.1 15.8 88.3 1.01 2.4
8 2 0.3 0.90 900 0.55 12.47 1.81 0.72 0.90 0.72 38.1 19.7 82.8 0.90 2.6
9 2 0.1 0.15 1000 0.77 3.26 0.90 1.64 0.50 0.35 60.6 26.6 89.7 1.31 2.6
10 2 0.1 0.90 1000 0.70 14.34 2.28 1.12 0.87 0.67 54.8 47.0 68.9 1.04 2.7
11a 5 0.1 0.90 755–1104 0.93 13.08 3.60 0.88 0.88 0.80 58.0 57.6 71.5 1.13 3.9
12a 4 0.1 0.90 610–1171 0.83 28.40 4.55 0.93 0.94 0.83 67.3 53.5 61.0 0.90 3.0

a
Non-isothermal test.

continues to evolve even after CO and CO2 drop to zero because CH4 by thermodynamic equilibrium calculations. In contrast, experiments
can decompose on the inner wall surfaces inside the reactor and at the show that decreasing pressure favors the CO2 formation at the expense
surface of the supporting alumina particles at the cavity bottom). In of lowered CO yield.
fact, H2O was also formed concomitantly with CO2 according to Eq. Importantly, at 0.90 bar and 1000 °C (Fig. 7f), the H2 production
(13); however, it cannot be detected by gas analysis, and H2O formation rate was extremely high and kept quite stable over the entire run al­
was not observable, because of the low amount formed and the possible though CO2 and CO reached zero after 9 min. This is attributed to the
condensation before reaching the filter. This H2O formation was evi­ favored side reaction associated with CH4 cracking. However, the me­
denced by thermodynamics (Fig. 4) and prior studies [47]. To quantify thane cracking reaction was not favored when decreasing the pressure
the evolved H2O, H-atom balance was performed as follows: to 0.15 bar (Fig. 7e), unveiling the beneficial effect of lowered pressure
on alleviating methane cracking.
(4nCH4 )inlet = (2nH2 + 4nCH4 + 2nH2O )outlet (21)
Fig. 8 shows the H2, CO, CO2 yields as a function of pressure at 800,
Thus, the amount of H2O can be calculated and compared to the 900, and 1000 °C, calculated by time integration of each gas production
theoretical value (given by Eq. (13)). As a result, H2O amounts from H- rate from Fig. 7 (the error bars show the uncertainty of the values due
atom balance (Eq. (21)) were consistent with the theoretical values (Eq. to the measurements linked to the gas analysis mainly). The H2 yield
(13)). For example at Run#9, H2O yield obtained from Eq. (21) was produced from the main reaction associated with CH4 + ZnO (Eq. (4),
3.14 mmol/gZnO compared to 3.28 mmol/gZnO from Eq. (13) (twice the Fig. 8a) and from the side reaction associated with methane cracking
CO2 yield, Table 1). (Eq. (5), Fig. 8b) were quantified and reported separately. The H2 yield
During reaction progress, temperatures (T1 and Tpyrometer) remained (from CH4 cracking) was estimated using the total H2 yield measured by
constant over entire runs, confirming isothermal operation. Methane gas analysis minus the H2 yield from the main reaction (Eq. (4)) that
evolution trends were always opposite compared to syngas production doubles the CO yield. In addition, to confirm this approach, carbon-
rates, demonstrating the conversion of CH4 to syngas. In comparison, atom balance was also performed (Eq. (22)) to directly determine the
CO2 and CO production rates at 0.15 bar were evolved slightly before solid carbon formation, and then quantify the H2 from CH4 cracking
H2, whereas they were produced at the same time at 0.90 bar, pointing (twice the amount of C, Eq. (5)).
out the influence of reduced pressure on reaction mechanism. (nCH4 )inlet = (nC + nCO + nCO2 + nCH4 )outlet (22)
Increasing temperature or decreasing pressure consistently im­
proved reaction kinetics, as reflected by a decrease in reaction duration, As a result, the H2 yields from CH4 cracking were consistent. For
in agreement with the thermodynamically favored reaction (Figs. 4 and example at Run#9, H2 yield from CH4 cracking was 1.33 mmol/gZnO
5) and with previously reported results [53]. For instance, at 0.15 bar (obtained from carbon-atom balance, Eq. (22)) compared to
the duration was 21 min at 800 °C compared to 9 min at 1000 °C, and at 1.46 mmol/gZnO (obtained from total H2 yield minus twice the CO
900 °C the duration was 17.5 min at 0.90 bar compared to 13.5 min at yield, Table 1).
0.15 bar (the difference is negligible at 800 °C, and reaction duration is On the one hand, H2 (ZnO + CH4) (Fig. 8a), H2 (CH4 cracking)
almost the same, probably because the effect of pressure at the lowest (Fig. 8b), and CO (Fig. 8c) yields dropped with decreasing pressure
temperature is not detectable due to low kinetics). In addition, a re­ regardless of temperature (e.g. at 900 °C: 4.1 (H2 (ZnO + CH4)), 2.9
markable increase in syngas production rates with temperature was (H2(CH4 cracking)) and 2.0 (CO) mmol/gZnO at 0.90 bar vs. 1.8 (H2
noticed for both pressures. For example, at 0.15 bar, the peak H2, CO2, (ZnO + CH4)), 1.5 (H2 (CH4 cracking)) and 0.9 (CO) mmol/gZnO at
and CO production rates were 7.39, 2.07, and 1.61 mL min−1 gZnO−1 at 0.15 bar). On the other hand, CO2 rose considerably with decreasing
800 °C compared to 23.10, 16.36, and 9.54 mL min−1 gZnO−1 at pressure whatever the temperature (Fig. 8d) (e.g. at 900 °C: from
1000 °C. 0.9 mmol/gZnO at 0.90 bar to 1.5 mmol/gZnO at 0.15 bar, leading to a
Regarding pressure influence, at 0.90 bar the syngas production decrease in syngas selectivity from 0.80 (H2) and 0.70 (CO) at 0.90 bar
rates seemed to be greater than those at 0.15 bar. In contrast to the to 0.53 (H2) and 0.38 (CO) at 0.15 bar). These variations can be ex­
results obtained at 0.90 bar, the CO production rate at 0.15 bar was plained by the fact that the reactions proceed through consecutive
always lower than CO2 regardless of temperature, indicating pressure steps. First, ZnO is chiefly reduced by CH4 to Zn vapor, CO, and H2 (Eq.
influence (at 900 °C peak CO and CO2 production rates were 9.06 and (4)), and further reduced by produced CO (Eq. (11)), H2 (Eq. (12)), or
2.65 mL min−1 gZnO−1 at 0.90 bar vs. 6.33 and 8.15 mL min−1 gZnO−1 carbon (Eq. (3)) to Zn vapor, CO2, H2O, or CO. However, the gas re­
at 0.15 bar, respectively). CO2 formation is not predicted by thermo­ sidence time is lowered consistently with decreasing pressure (by a
dynamics (Fig. 5) because kinetic limitations are not taken into account factor of ~6). Thus, the intermediate gas-phase secondary reactions

7
S. Chuayboon and S. Abanades Fuel Processing Technology 211 (2021) 106572

Fig. 7. Outlet gas product species flow-rates along with reactor temperatures (T1 and Tpyrometer) for combined ZnO reduction and CH4 reforming at temperatures of
800, 900, and 1000 °C and pressures of 0.15 and 0.90 bar.

related to methane cracking (Eq. (5)), dry and steam methane re­ yield (9.79 mmol/gZnO). This issue could be alleviated by decreasing
forming (Eqs. (6) and (7)), Boudouard (Eq. (8)), carbon gasification (Eq. the pressure to lower the gas residence time (indeed, H2 (CH4 cracking)
(9)), or reverse water-gas shift (Eq. (10)) can be kinetically limited, yield at 0.15 bar and 1000 °C was lowered to 1.46 mmol/gZnO, Fig. 8b).
leading to a decrease in H2 and CO yield at the expense of an increase in Fig. 9a shows that net ZnO conversion (XZnO) increased with tem­
CO2 yield. The decrease of gas residence time more likely effects the perature, and it was also upgraded when decreasing pressure as clearly
gas-phase reactions than the reactions involving solid ZnO inside the seen at 0.15 bar (up to 60.6% at 1000 °C), in agreement with literature
packed-bed. [39]. Importantly, the lowest XZnO of 18.3% was obtained at atmo­
In agreement with thermodynamic analysis (Fig. 4) and previous spheric pressure (0.90 bar) and 800 °C with the presence of ZnO re­
studies [47], increasing the temperature enhanced syngas yield re­ maining in the reactor cavity receiver after the experiment according to
gardless of pressure, although the effect of temperature on syngas yield Fig. S1 (Supporting Information). This is because of low thermo­
is more pronounced at 0.90 bar than at 0.15 bar (e.g. at 0.90 bar: 4.63 dynamic driving force for the reaction at this temperature. Never­
(total H2), 0.87 (CO), and 0.34 (CO2) mmol/gZnO at 800 °C vs. 14.34 theless, it was upgraded to 34.8% when lowering pressure to 0.15 bar.
(total H2), 2.28 (CO), and 1.12 (CO2) mmol/gZnO at 1000 °C and syngas Note that an incomplete XZnO was assumed to also be due to Zn re­
selectivity: 0.87 (H2) and 0.67 (CO)). Nevertheless, it comes at the combination as well as losses caused by diffusion into insulation ma­
expense of favored methane cracking reaction (Fig. 8b), particularly at terials [40]. When decreasing pressure, methane conversion (XCH4)
1000 °C and 0.90 bar, as evidenced by the highest H2 (CH4 cracking) (Fig. 9b) tended to drop, particularly at 0.15 bar, due to lowered gas

8
S. Chuayboon and S. Abanades Fuel Processing Technology 211 (2021) 106572

Fig. 8. Yields of (a) H2 (ZnO + CH4), (b) H2 (CH4 cracking), (c) CO, and (d)
CO2 as a function of pressure for the temperatures of 800, 900, and 1000 °C.

residence time despite higher XZnO (Fig. 9a). However, it was enhanced
by increasing the temperature. Note that at 1000 °C and 0.90 bar, XCH4
was high (47%) because of strong CH4 cracking, as previously described
in Fig. 8b. Besides, incomplete carbon balance (Xc) was observed be­
cause of carbon particles formation (Table 1). However, it was close to
100% when lowering pressure and temperature, e.g. at 800 °C: Xc rose
from 92.4% at 0.90 bar to 96.1% at 0.15 bar, thanks to low solid carbon
formation.
Energy upgrade factor (U) rose minimally when decreasing pressure
(Fig. 9c) because of an increase in the unreacted methane, caused by
the reduced gas residence time. In addition, U values were above one
for all the runs (in the range of 1.04–1.31), confirming successful solar
energy storage into chemical products. Besides, U dropped with tem­
perature at 0.90 bar, but not at 0.15 bar. This confirmed that tem­
perature played stronger role on carbon deposition associated with
methane cracking at atmospheric pressure than at low pressure. In
addition, solar-to-chemical energy conversion efficiencies (ηsolar-to-che­
mical) (Fig. 9d) were obtained in the range 1.9–2.7%. They dropped
when reducing pressure due to the lowered H2 and CO yields (to the
benefit of CO2 and H2O), and the higher solar power input (Table 1)
needed to compensate the heat losses caused by vacuum pumping (hot
gases were rapidly sucked by vacuum pump, especially at 0.15 bar,
resulting in lower heat transfer rate). Nevertheless, ηsolar-to-chemical can be
upgraded by raising the temperature thanks to an improvement in
syngas yield and reaction rate. Indeed, in spite of higher solar power
input and CH4 conversion at high temperatures (both showing an ad­
verse impact on the efficiency, Eq. (20)), the shortened reaction dura­
tion (due to enhanced kinetics) inherently results in lowering the total
amounts of both injected methane and consumed solar energy (for in­
stance at 0.15 bar, the total solar energy input decreased from 0.61 MJ
at 800 °C to 0.42 MJ at 1000 °C), which in turn enhances ηsolar-to-chemical.
The higher efficiency when increasing the temperature is thus chiefly
the combined result of the enhanced syngas yields (H2 and CO) and
kinetics. Therefore, ηsolar-to-chemical increased when increasing the tem­
perature regardless of the pressure (Fig. 9d).
In summary, an improvement of the reaction kinetics was demon­
strated when increasing temperature or decreasing pressure, as evi­
denced by the enhanced syngas production rates and the shortened
reaction duration. However, H2 and CO yields declined with decreasing
pressure at the expense of increased CO2 (and H2O) yield caused by the
lowered gas residence time. Increasing the temperature promoted
syngas yield despite favored methane cracking reaction that can be
alleviated by decreasing the pressure. Net XZnO was upgraded by either
rising temperature or reducing pressure, while XCH4 was upgraded
when increasing temperature.

3.4. Inlet methane flow-rate influence

The influence of methane flow-rate (CH4 mole fraction) on syngas

production rates, yields, and performance metrics was experimentally
investigated at 900 °C and 0.90 bar (these conditions alleviate the CH4
cracking issue). The syngas production rates were plotted along with
temperatures for ZnO reduction with CH4 (2 g ZnO particles) at inlet
methane flow-rates of 0.1, 0.2, and 0.3 NL/min (33%, 50%, and 60%
CH4 mole fraction, respectively) in Fig. 10. Increasing the CH4 flow-rate
also decreases the contact time of the gas with the initial ZnO bed (from
0.80 s to 0.48 s).
When increasing methane flow-rate, the H2 production rate in­
creased considerably, especially between 0.1 and 0.2 NL/min, while CO
and CO2 production rates remained similar. For instance, the peak H2,
CO, and CO2 production rates were 21.26, 9.06, and 2.65 mL min−1
gZnO−1 at 0.1 NL/min compared to 34.54, 10.11, and 2.50 mL min−1

9
S. Chuayboon and S. Abanades Fuel Processing Technology 211 (2021) 106572

Fig. 9. (a) ZnO conversion (XZnO), (b) CH4 conversion (XCH4), (c) energy up­
grade factor (U), and (d) solar-to-chemical energy conversion efficiency (ηsolar-to-
chemical) as a function of pressure for the temperatures of 800, 900, and 1000 °C.

Fig. 10. Outlet gas product species flow-rates along with reactor temperatures
(T1 and Tpyrometer) for combined ZnO reduction and CH4 reforming at methane
flow-rates of 0.1, 0.2, and 0.3 NL/min and a temperature of 900 °C (constant
carrier N2 flow-rate of 0.2 NL/min was fed along with CH4 flow).

gZnO−1 at 0.3 NL/min. Differences in operating duration between the

three runs were negligible even though it tended to decrease slightly
(around 15 min duration) when rising the inlet methane flow-rate to
0.2 and 0.3 NL/min.
Fig. 11 confirms that an increase in the inlet methane flow-rate
improved H2 yield, especially H2 associated with CH4 cracking; how­
ever, CO and CO2 yields tended to decrease within the considered range
(4.09 (H2 (ZnO + CH4)), 2.94 (H2 (CH4 cracking)), 2.04 (CO), and 0.89
(CO2) mmol/gZnO at 0.1 NL/min vs. 3.61 (H2 (ZnO + CH4)), 8.86 (H2

10
S. Chuayboon and S. Abanades Fuel Processing Technology 211 (2021) 106572

Fig. 11. H2, CO, and CO2 yields as a function of methane flow-rate at 900 °C
and 0.90 bar.

(CH4 cracking)), 1.81 (CO), and 0.72 (CO2) mmol/gZnO at 0.3 NL/min).
This implies that raising the inlet methane flow-rate would rather
promote CH4 cracking reaction than the main reaction of ZnO reduction
with methane, as reflected by a substantial increase in H2 and a drop in
CO and CO2 yields with methane flow-rate. In addition, the associated
solid carbon formation could be estimated to be half the H2 (from CH4
cracking) yield, thus increasing in the range 1.47–4.43 mmol/gZnO at
0.1–0.3 NL/min, and inducing a drop in Xc with inlet methane flow-rate
(from 91.8% at 0.1 NL/min to 82.8% at 0.3 NL/min). Note that yellow
solid chunks were found in the reactor cavity after the experiments
(Run#6 and Run#8) when employing the methane flow-rates of 0.2
and 0.3 NL/min (Fig. S2). These formations looked like sulfur im­
Fig. 12. (a) ZnO conversion (XZnO) and CH4 conversion (XCH4), (b) energy
purities as a result of an excess in inlet methane.
upgrade factor (U) and solar-to-chemical energy conversion efficiency (ηsolar-to-
XZnO and XCH4 fluctuated slightly in the range of 36.1–45.8% and
chemical) as a function of methane flow-rate at 900 °C.
15.8–19.7%, respectively (Fig. 12a). The impact of the methane flow-
rate on XZnO and XCH4 was not obvious although their trends increased
slightly above 0.2 NL/min, particularly XCH4 due to methane cracking. while T1 is still above 1000 °C, thus high enough for the reaction to
U decreased from 1.09 to 0.90 (Fig. 12b), resulting from the increase of proceed; the amount of CO (and CO2) reaching zero means that there is
XCH4 caused mainly by methane cracking reaction at the expense of no more ZnO to react with CH4). This is because of the higher operating
solid carbon formation. Part of the formed carbon particles was carried temperature of Run#11, which resulted in greater reaction rates,
by outlet gases to filters, thereby losing their heating value and even­ leading to shorter duration. As shown in Fig. 13, syngas production
tually leading to a reduction in U. The heating value contained in rates varied strongly with a change in temperature, especially H2. The
carbon is lost if carbon is not recovered; thus, the calorific value of the H2 production rate increased with temperature and was dramatically
carbon was not included to calculate U, because solid carbon was higher than CO, indicating additional H2 formation from thermal CH4
considered as an undesired by-product. ηsolar-to-chemical increased slightly dissociation. In contrast to the results obtained from isothermal tests at
from 2.4% at 0.1 NL/min to 2.6% at 0.3 NL/min, ascribed to higher 800, 900, and 1000 °C (0.15 bar) (Figs. 7 and 10), CO production rate
syngas yield (Fig. 11). Therefore, in order to maximize XZnO and U while reached zero after that of CO2 (Fig. 13). As indicated in Table 1, syngas
minimizing CH4 cracking, the lowest inlet methane mole fraction is yields obtained from Run#12 (H2 and CO yields were 28.4 and
recommended. 4.6 mmol/gZnO, respectively, with H2/CO = 6.2) were greater than
those obtained from Run#11, especially H2 yield owing to stronger CH4
dissociation at the expense of high solid carbon formation, which ex­
3.5. Non-isothermal ZnO reduction with CH4 plained the lower Xc (61%). The highest values of XZnO (67.3%) and
XCH4 (57.6%) were obtained from non-isothermal tests, confirming the
Two non-isothermal on-sun tests of ZnO reduction with CH4 were strong temperature effect on ZnO and CH4 conversion.
performed with higher amounts of ZnO (5 and 4 g) at atmospheric
pressure to investigate dynamic chemical reaction behavior and asso­
ciated mechanism while increasing the temperature under CH4 flow. 3.6. Zn product characterization
Fig. 13 shows the evolution of syngas production rates together with
temperatures. To assess the impact of starting reaction temperature on To collect solid products separately, the removable outlet compo­
reaction rate, CH4 was injected at 755 °C (Table 1) for Run#11 (5 g nents where solid products deposited were separated into two zones: (i)
ZnO) vs. 610 °C for Run#12 (4 g ZnO). Note that the temperature of alumina tube and connector (zone A), and (ii) ceramic filter (zone B).
Run#11 was higher than Run#12 by above 100 °C (T1) over 14 min. It They were weighed before and after experiments in order to quantify
was found that Run#11 took shorter duration to reach reaction com­ the amount of solid products inside them. As a result, a large amount of
pletion despite a higher amount of ZnO (as evidenced by CO production solid products was collected from zone A in the range 0.5–2.2 g, fol­
rate reaching zero after 18 min for Run#11 vs. 27 min for Run#12 lowed by zone B in the range 0.2–1.3 g (total amount in the range

11
S. Chuayboon and S. Abanades Fuel Processing Technology 211 (2021) 106572

Fig. 13. Outlet gas product species flow-rates along with reactor temperatures (T1 and Tpyrometer) during non-isothermal combined ZnO reduction and CH4 reforming
for two runs at different heating rates and starting reaction temperatures.

0.82–3.53 g), according to Table S1 (Supporting Information). the water-cooled alumina tube (zone A, Fig. S4), condensation zone of
However, solid carbon particles contained in the products were also Zn vapor, thus implying the existence of Zn recombination, which ex­
observed after experiments, especially Run#12, in agreement with in­ plained the incomplete XZnO. Downstream (zone B), the solid products
complete Xc. A portion of the collected solid products was sampled for were carried by outlet gases, and in this zone the temperature was
XRD, FESEM, and TGA analysis. lower. As a result, only pure, fine Zn particles were recovered regardless
The XRD patterns of the solid products from zone A and zone B are of pressure and temperature. Therefore, the solar reactor was suitable
plotted for temperatures of 800, 900, and 1000 °C in Fig. 14a (0.15 bar) for producing both Zn and syngas in a single process.
and Fig. 14b (0.90 bar). Additional XRD analyses were performed for The morphology of solid product particles from zones A and B at the
powders obtained with different methane flow-rates at 0.90 bar and pressures of 0.15 and 0.90 bar, characterized by FESEM, is shown in
900 °C (Fig. S3). Overall, the products were well crystallized with clear Fig. 15 (1.2 μm scale) and Fig. S5 (3.0 μm scale). Overall, condensed Zn
patterns. Only the Zn pattern was detected regardless of pressure and products were formed in the shape of a clear crystal hexagonal structure
temperature, demonstrating pure Zn production. However, a small with most diameters below 1.2 μm. The products collected in zone A
amount of crystallized gray grains was found at the 1-cm outlet tip of (Fig. 15a) exhibited Zn in plane layers because Zn vapor condensed in

Fig. 14. XRD patterns of the collected solid products deposited in the aluminum tube and connector (zone A) and ceramic filter (zone B) for temperatures of 800, 900,
and 1000 °C at (a) 0.15 bar and (b) 0.90 bar.

12
S. Chuayboon and S. Abanades Fuel Processing Technology 211 (2021) 106572

Fig. 15. FESEM images of the solid products collected on (a,c) alumina tube and connector (zone A) and (b,d) ceramic filter at a temperature of 800 °C and pressures
of 0.15 and 0.90 bar.

contact with cold surfaces, uniformly growing along plane layers. Some and reactor performance during ZnO reduction with methane were
low amounts of the solid carbon deposition from CH4 dissociation in the assessed via a detailed parametric study. Results showed the benefits of
form of fine particles can be observed on the surface of Zn structure, e.g. the combined process involving both methane as gaseous reducer for Zn
Fig. 15b. However, such small amounts of fine carbon particles are extractive metallurgy and ZnO as solid oxidant for methane reforming.
mainly observed in the case of low pressure operation (favoring high Regarding isothermal tests, gas species production rates were found
gas velocities), denoting some carbon particles entrainment to the re­ to be dependent on temperature, pressure, and methane flow-rate.
actor outlet by the carrier gas. This carbon is not problematic as it can Increasing temperature and decreasing pressure improved reaction ki­
be avoided by decreasing the carrier gas. If the produced Zn powder is netics. In comparison, CO2 production rate was higher while operating
further used to split H2O or CO2 in a closed cycle, the carbon will be duration was shorter at 0.15 bar compared to atmospheric pressure.
consumed (gasified) by the reaction producing additional H2 or CO, and Decreasing pressure lowered syngas yield and selectivity (H2 and CO),
ZnO that can be recycled. favored CO2 and lowered methane conversion due to insufficient gas
In addition, the produced Zn powder reactivity was investigated residence time. Net ZnO conversion increased with decreasing pressure
using TGA. The purpose is to prove its recombination ability occurring up to 60.6% at 0.15 bar, in agreement with thermodynamic predictions.
at the condensation zone (zone A) and to further reveal its potential to Increasing temperature enhanced syngas production rate, syngas yield,
produce CO (H2) in a CO2 (H2O) splitting scheme. As a result, high conversion of ZnO and methane at the expense of favored methane
reactivity of Zn product with CO2 was highlighted (Fig. S6). cracking, which can be alleviated by lowering pressure. Increasing
Consequently, the pure Zn product obtained from this process is re­ methane flow-rate also favored methane cracking rather than main
active towards oxidation, and can be employed as an oxygen carrier in a reactions, as evidenced by the H2/CO ratio significantly higher than
two-step solar thermochemical H2O/CO2 splitting system for solar fuel two and an increase in XCH4 conversion.
production. Regarding non-isothermal tests, gas production rates were strongly
dependent on the operating temperature. The maximum XZnO and XCH4
up to 67.3% and 57.6%, respectively, were achieved from non-iso­
4. Conclusions
thermal tests, confirming the strong temperature impact. Therefore, to
enhance syngas selectivity, XZnO, and XCH4 while diminishing the CO2
Solar thermochemical methane-driven reduction of ZnO for co-
formation and methane cracking issues, the process should be operated
production of Zn and syngas was evaluated thermodynamically and
isothermally at an intermediate pressure of 0.45 bar, inlet methane
experimentally in a directly-irradiated packed-bed reactor prototype.
flow-rate of 0.1 NL/min, and temperature of 1000 °C. The highest U and
The process was successfully demonstrated under both isothermal and
ηsolar-to-chemical of 1.31 and 3.9%, respectively, were accomplished, and
non-isothermal conditions using a real solar concentrating system. Gas
pure metallic Zn product with well crystallized structure, micrometric
species production rates, yields, ZnO conversion, methane conversion,

13
S. Chuayboon and S. Abanades Fuel Processing Technology 211 (2021) 106572

size and high reactivity was highlighted. This solar-driven ZnO reduc­ solar energy, Energy. 31 (2006) 2805–2822, https://doi.org/10.1016/j.energy.
tion with methane was demonstrated to be promising for producing 2005.11.002.
[2] A. Haeussler, S. Abanades, A. Julbe, J. Jouannaux, M. Drobek, A. Ayral, B. Cartoixa,
both pure Zn and syngas in a single process, as a suitable means of Remarkable performance of microstructured ceria foams for thermochemical
storing intermittent solar energy into sustainable chemical products. splitting of H2O and CO2 in a novel high–temperature solar reactor, Chem. Eng. Res.
Des. 156 (2020) 311–323, https://doi.org/10.1016/j.cherd.2020.02.008.
[3] R.R. Bhosale, G. Takalkar, P. Sutar, A. Kumar, F. AlMomani, M. Khraisheh, A decade
CRediT authorship contribution statement of ceria based solar thermochemical H2O/CO2 splitting cycle, Int. J. Hydrog. Energy
44 (2019) 34–60, https://doi.org/10.1016/j.ijhydene.2018.04.080.
Srirat Chuayboon: Methodology, Validation, Formal analysis, [4] R.C. Pullar, R.M. Novais, A.P.F. Caetano, M.A. Barreiros, S. Abanades,
F.A.C. Oliveira, A review of solar thermochemical CO2 splitting using ceria-based
Investigation, Writing - original draft. Stéphane Abanades: ceramics with designed morphologies and microstructures, Front. Chem. 7 (2019)
Conceptualization, Methodology, Formal analysis, Investigation, 601, https://doi.org/10.3389/fchem.2019.00601.
Writing - review & editing, Supervision, Project administration, [5] A. Haeussler, S. Abanades, J. Jouannaux, A. Julbe, Non-stoichiometric redox active
perovskite materials for solar thermochemical fuel production, A Review, Catal. 8
Funding acquisition.
(2018) 611, https://doi.org/10.3390/catal8120611.
[6] F. Müller, H. Patel, D. Blumenthal, P. Poživil, P. Das, C. Wieckert, P. Maiti, S. Maiti,
Declaration of competing interest A. Steinfeld, Co-production of syngas and potassium-based fertilizer by solar-driven
thermochemical conversion of crop residues, Fuel Process. Technol. 171 (2018)
89–99, https://doi.org/10.1016/j.fuproc.2017.08.006.
The authors declare that they have no known competing financial [7] N. Piatkowski, C. Wieckert, A. Steinfeld, Experimental investigation of a packed-bed
interests or personal relationships that could have appeared to influ­ solar reactor for the steam-gasification of carbonaceous feedstocks, Fuel Process.
Technol. 90 (2009) 360–366, https://doi.org/10.1016/j.fuproc.2008.10.007.
ence the work reported in this paper. [8] P.G. Loutzenhiser, A.P. Muroyama, A review of the state-of-the-art in solar-driven
gasification processes with carbonaceous materials, Sol. Energy 156 (2017) 93–100,
Acknowledgements https://doi.org/10.1016/j.solener.2017.05.008.
[9] X. Zhao, H. Zhou, V.S. Sikarwar, M. Zhao, A.-H.A. Park, P.S. Fennell, L. Shen, L.-
S. Fan, Biomass-based chemical looping technologies: the good, the bad and the
King Mongkut's Institute of Technology Ladkrabang (KMITL), future, Energy Environ. Sci. 10 (2017) 1885–1910, https://doi.org/10.1039/
Thailand is acknowledged for fellowship granting under an academic C6EE03718F.
[10] N. Piatkowski, C. Wieckert, A.W. Weimer, A. Steinfeld, Solar-driven gasification of
joint project between KMITL and PROMES-CNRS. The authors also
carbonaceous feedstock-a review, Energy Environ. Sci. 4 (2011) 73–82, https://doi.
would like to gratefully thank R. Garcia for solar reactor technical org/10.1039/C0EE00312C.
support. [11] S. Chuayboon, S. Abanades, S. Rodat, Comprehensive performance assessment of a
continuous solar-driven biomass gasifier, Fuel Process. Technol. 182 (2018) 1–14,
Nomenclature
https://doi.org/10.1016/j.fuproc.2018.10.016.
[12] E. Koepf, I. Alxneit, C. Wieckert, A. Meier, A review of high temperature solar
ΔG° Gibbs free energy (J/mol) driven reactor technology: 25years of experience in research and development at
ΔH Enthalpy change of reaction (J/kg) the Paul Scherrer Institute, Appl. Energy 188 (2017) 620–651, https://doi.org/10.
1016/j.apenergy.2016.11.088.
T Temperature (°C) [13] S. Abanades, G. Flamant, High-temperature solar chemical reactors for hydrogen
LHV Lower heating value (J/kg) production from natural gas cracking, Chem. Eng. Commun. 195 (2008)
m mass flow-rate (kg/s) 1159–1175, https://doi.org/10.1080/00986440801943602.
[14] S. Rodat, S. Abanades, G. Flamant, High-temperature solar methane dissociation in
Fi molar flow-rate of species i (mol/s) a multitubular cavity-type reactor in the temperature range 1823−2073 K, Energy
Qsolar Solar power input (W) Fuel 23 (2009) 2666–2674, https://doi.org/10.1021/ef900037v.
U Energy upgrade factor (−) [15] S. Rodat, S. Abanades, J.-L. Sans, G. Flamant, Hydrogen production from solar
thermal dissociation of natural gas: development of a 10kW solar chemical reactor
SH2 Hydrogen selectivity (−) prototype, Sol. Energy 83 (2009) 1599–1610, https://doi.org/10.1016/j.solener.
SCO Carbon monoxide selectivity (−) 2009.05.010.
XZnO ZnO conversion (%) [16] S. Abanades, H. Kimura, H. Otsuka, Hydrogen production from CO2-free thermal
decomposition of methane: design and on-sun testing of a tube-type solar thermo­
XCH4 Methane conversion (%) chemical reactor, Fuel Process. Technol. 122 (2014) 153–162, https://doi.org/10.
Xc Carbon mass balance (%) 1016/j.fuproc.2014.02.002.
ηsolar-to-chemical Solar-to-chemical energy conversion efficiency (%) [17] P.T. Krenzke, J.R. Fosheim, J.H. Davidson, Solar fuels via chemical-looping re­
forming, Sol. Energy 156 (2017) 48–72, https://doi.org/10.1016/j.solener.2017.
05.095.
[18] D.S.A. Simakov, M.M. Wright, S. Ahmed, E.M.A. Mokheimer, Y. Román-Leshkov,
Solar thermal catalytic reforming of natural gas: a review on chemistry, catalysis
and system design, Catal. Sci. Technol. 5 (2015) 1991–2016, https://doi.org/10.
1039/C4CY01333F.
Abbreviations [19] M.M. Nair, S. Abanades, Tailoring hybrid nonstoichiometric ceria redox cycle for
combined solar methane reforming and thermochemical conversion of H2O/CO2,
DNI Direct Normal Irradiance Energy Fuel 30 (2016) 6050–6058, https://doi.org/10.1021/acs.energyfuels.
6b01063.
CSP Concentrated solar power [20] S. Chuayboon, S. Abanades, S. Rodat, Solar chemical looping reforming of methane
FESEM Field emission scanning electron microscopy combined with isothermal H2O/CO2 splitting using ceria oxygen carrier for syngas
XRD X-ray powder diffraction production, J. Energy Chem. 41 (2019) 60–72, https://doi.org/10.1016/j.jechem.
2019.05.004.
TGA Thermogravimetry analysis
[21] M. Romero, A. Steinfeld, Concentrating solar thermal power and thermochemical
O.D. outer diameter fuels, Energy Environ. Sci. 5 (2012) 9234–9245, https://doi.org/10.1039/
I.D. inner diameter C2EE21275G.
[22] A. Ekman Nilsson, M. Macias Aragonés, F. Arroyo Torralvo, V. Dunon, H. Angel,
NL Normal liter (at 0 °C, 1 atm)
K. Komnitsas, K. Willquist, A review of the carbon footprint of Cu and Zn production
from primary and secondary sources, Minerals 7 (2017) 168, https://doi.org/10.
Appendix A. Supplementary data 3390/min7090168.
[23] P. Haueter, S. Moeller, R. Palumbo, A. Steinfeld, The production of zinc by thermal
dissociation of zinc oxide—solar chemical reactor design, Sol. Energy 67 (1999)
Supplementary data to this article can be found online at https:// 161–167, https://doi.org/10.1016/S0038-092X(00)00037-2.
doi.org/10.1016/j.fuproc.2020.106572. [24] S. Abanades, P. Charvin, G. Flamant, Design and simulation of a solar chemical
reactor for the thermal reduction of metal oxides: case study of zinc oxide dis­
sociation, Chem. Eng. Sci. 62 (2007) 6323–6333, https://doi.org/10.1016/j.ces.
References 2007.07.042.
[25] L.O. Schunk, P. Haeberling, S. Wepf, D. Wuillemin, A. Meier, A. Steinfeld, A re­
[1] S. Abanades, P. Charvin, G. Flamant, P. Neveu, Screening of water-splitting ther­ ceiver-reactor for the solar thermal dissociation of zinc oxide, J. Sol. Energy Eng.
mochemical cycles potentially attractive for hydrogen production by concentrated 130 (2008) 21006–21009, https://doi.org/10.1115/1.2840576.

14
S. Chuayboon and S. Abanades Fuel Processing Technology 211 (2021) 106572

[26] C. Perkins, P.R. Lichty, A.W. Weimer, Thermal ZnO dissociation in a rapid aerosol analysis of isothermal and non-isothermal CeO2-based solar thermochemical cycle
reactor as part of a solar hydrogen production cycle, Int. J. Hydrog. Energy 33 with methane-driven reduction, Renew. Energy 143 (2019) 915–921, https://doi.
(2008) 499–510, https://doi.org/10.1016/j.ijhydene.2007.10.021. org/10.1016/j.renene.2019.05.047.
[27] M. Chambon, S. Abanades, G. Flamant, Thermal dissociation of compressed ZnO [43] L. Wang, T. Ma, Z. Chang, H. Li, M. Fu, X. Li, Solar fuels production via two-step
and SnO2 powders in a moving-front solar thermochemical reactor, AICHE J. 57 thermochemical cycle based on Fe3O4/Fe with methane reduction, Sol. Energy 177
(2011) 2264–2273, https://doi.org/10.1002/aic.12432. (2019) 772–781, https://doi.org/10.1016/j.solener.2018.12.009.
[28] E. Koepf, S.G. Advani, A. Steinfeld, A.K. Prasad, A novel beam-down, gravity-fed, [44] S. Chuayboon, S. Abanades, S. Rodat, Stepwise Solar methane reforming and water-
solar thermochemical receiver/reactor for direct solid particle decomposition: de­ splitting via lattice oxygen transfer in iron and cerium oxides, Energy Technol. 8 (8)
sign, modeling, and experimentation, Int. J. Hydrog. Energy 37 (2012) (2020) 1900415, https://doi.org/10.1002/ente.201900415.
16871–16887, https://doi.org/10.1016/j.ijhydene.2012.08.086. [45] Q. Jiang, H. Zhang, Y. Cao, H. Hong, H. Jin, Solar hydrogen production via per­
[29] C. Perkins, P. Lichty, A.W. Weimer, Determination of aerosol kinetics of thermal ovskite-based chemical-looping steam methane reforming, Energy Convers. Manag.
ZnO dissociation by thermogravimetry, Chem. Eng. Sci. 62 (2007) 5952–5962, 187 (2019) 523–536, https://doi.org/10.1016/j.enconman.2019.01.112.
https://doi.org/10.1016/j.ces.2007.06.039. [46] A. Bayon, A. de la Calle, K.K. Ghose, A. Page, R. McNaughton, Experimental,
[30] S. Chuayboon, S. Abanades, S. Rodat, Syngas production via solar-driven chemical computational and thermodynamic studies in perovskites metal oxides for ther­
looping methane reforming from redox cycling of ceria porous foam in a volumetric mochemical fuel production: a review, Int. J. Hydrog. Energy 45 (2020)
solar reactor, Chem. Eng. J. 356 (2019) 756–770, https://doi.org/10.1016/J.CEJ. 12653–12679, https://doi.org/10.1016/j.ijhydene.2020.02.126.
2018.09.072. [47] A. Steinfeld, A. Frei, P. Kuhn, D. Wuillemin, Solar thermal production of zinc and
[31] A. Stamatiou, P.G. Loutzenhiser, A. Steinfeld, Solar syngas production via H2O/ syngas via combined ZnO-reduction and CH4-reforming processes, Int. J. Hydrog.
CO2-splitting thermochemical cycles with Zn/ZnO and FeO/Fe3O4 Redox reactions, Energy 20 (1995) 793–804, https://doi.org/10.1016/0360-3199(95)00016-7.
Chem. Mater. 22 (2010) 851–859, https://doi.org/10.1021/cm9016529. [48] A. Steinfeld, M. Brack, A. Meier, A. Weidenkaff, D. Wuillemin, A solar chemical
[32] A. Stamatiou, P.G. Loutzenhiser, A. Steinfeld, Solar syngas production from H2O reactor for co-production of zinc and synthesis gas, Energy. 23 (1998) 803–814,
and CO2 via two-step thermochemical cycles based on Zn/ZnO and FeO/Fe3O4 https://doi.org/10.1016/S0360-5442(98)00026-7.
redox reactions: kinetic analysis, Energy Fuel 24 (2010) 2716–2722, https://doi. [49] R.R. Bhosale, Thermodynamic efficiency analysis of ZnO/Zn based solar thermo­
org/10.1021/ef901544v. chemical CH4 reforming and H2O splitting cycle, Int. J. Hydrog. Energy 45 (2020)
[33] A. Stamatiou, P.G. Loutzenhiser, A. Steinfeld, Syngas production from H2O and CO2 5760–5771, https://doi.org/10.1016/j.ijhydene.2019.05.191.
over Zn particles in a packed-bed reactor, AICHE J. 58 (2012) 625–631, https://doi. [50] T. Osinga, G. Olalde, A. Steinfeld, Solar carbothermal reduction of ZnO: shrinking
org/10.1002/aic.12580. packed-bed reactor modeling and experimental validation, Ind. Eng. Chem. Res. 43
[34] S. Abanades, Thermogravimetry analysis of CO2 and H2O reduction from solar (2004) 7981–7988, https://doi.org/10.1021/ie049619q.
nanosized Zn powder for thermochemical fuel production, Ind. Eng. Chem. Res. 51 [51] W. Villasmil, M. Brkic, D. Wuillemin, A. Meier, A. Steinfeld, Pilot scale demon­
(2012) 741–750, https://doi.org/10.1021/ie202518k. stration of a 100-kWth solar thermochemical plant for the thermal dissociation of
[35] D. Weibel, Z.R. Jovanovic, E. Gálvez, A. Steinfeld, Mechanism of Zn particle oxi­ ZnO, J. Sol. Energy Eng. 136 (2013) 11011–11016, https://doi.org/10.1115/1.
dation by H2O and CO2 in the presence of ZnO, Chem. Mater. 26 (2014) 6486–6495, 4025512.
https://doi.org/10.1021/cm503064f. [52] M. Halmann, A. Frei, A. Steinfeld, Vacuum carbothermic reduction of Al2O3, BeO,
[36] M. Chambon, S. Abanades, G. Flamant, Design of a lab-scale rotary cavity-type solar MgO-CaO, TiO2, ZrO2, HfO2 + ZrO2, SiO2, SiO2 + Fe2O3, and GeO2 to the metals. A
reactor for continuous thermal dissociation of volatile oxides under reduced pres­ thermodynamic study, Miner. Process. Extr. Metall. Rev. 32 (2011) 247–266.
sure, J. Sol. Energy Eng. 132 (2010) 021006, , https://doi.org/10.1115/1.4001147. [53] G. Levêque, S. Abanades, Investigation of thermal and carbothermal reduction of
[37] E. Koepf, W. Villasmil, A. Meier, Pilot-scale solar reactor operation and character­ volatile oxides (ZnO, SnO2, GeO2, and MgO) via solar-driven vacuum thermo­
ization for fuel production via the Zn/ZnO thermochemical cycle, Appl. Energy 165 gravimetry for thermochemical production of solar fuels, Thermochim. Acta 605
(2016) 1004–1023, https://doi.org/10.1016/j.apenergy.2015.12.106. (2015) 86–94, https://doi.org/10.1016/j.tca.2015.02.015.
[38] C. Wieckert, R. Palumbo, U. Frommherz, A two-cavity reactor for solar chemical [54] A. Steinfeld, C. Larson, R. Palumbo, M. Foley, Thermodynamic analysis of the co-
processes: heat transfer model and application to carbothermic reduction of ZnO, production of zinc and synthesis gas using solar process heat, Energy. 21 (1996)
Energy. 29 (2004) 771–787, https://doi.org/10.1016/S0360-5442(03)00183-X. 205–222, https://doi.org/10.1016/0360-5442(95)00125-5.
[39] M. Brkic, E. Koepf, A. Meier, Solar carbothermal reduction of aerosolized ZnO [55] S. Kräupl, A. Steinfeld, Pulsed gas feeding for stoichiometric operation of a gas-solid
particles under vacuum: modeling, experimentation, and characterization of a drop- vortex flow solar chemical reactor, J. Sol. Energy Eng. 123 (2000) 133–137,
tube reactor, Chem. Eng. J. 313 (2017) 435–449, https://doi.org/10.1016/j.cej. https://doi.org/10.1115/1.1351172.
2016.12.057. [56] S. Kräupl, A. Steinfeld, Experimental investigation of a vortex-flow solar chemical
[40] T. Osinga, U. Frommherz, A. Steinfeld, C. Wieckert, Experimental investigation of reactor for the combined ZnO-reduction and CH4-reforming, J. Sol. Energy Eng. 123
the solar carbothermic reduction of ZnO using a two-cavity solar reactor, J. Sol. (2001) 237–243, https://doi.org/10.1115/1.1384569.
Energy Eng. 126 (2004) 633–637, https://doi.org/10.1115/1.1639001. [57] C. Wieckert, A. Steinfeld, Solar thermal reduction of ZnO using CH4:ZnO and C:ZnO
[41] P. Holzemer-Zerhusen, S. Brendelberger, H. von Storch, M. Roeb, C. Sattler, R. Pitz- molar ratios less than 1, J. Sol. Energy Eng. 124 (2001) 55–62, https://doi.org/10.
Paal, Efficiency assessment of solar redox reforming in comparison to conventional 1115/1.1434980.
reforming, Int. J. Hydrog. Energy 45 (2020) 4137–4151, https://doi.org/10.1016/j. [58] S. Chuayboon, S. Abanades, Clean magnesium production using concentrated solar
ijhydene.2019.12.065. heat in a high-temperature cavity-type thermochemical reactor, J. Clean. Prod. 232
[42] T. Ma, L. Wang, C. Chang, J.S. Akhatov, M. Fu, X. Li, A comparative thermodynamic (2019) 784–795, https://doi.org/10.1016/j.jclepro.2019.05.371.

15