Available online at www.sciencedirect.

com

ScienceDirect
Energy Procedia 114 (2017) 6893 – 6904

13th International Conference on Greenhouse Gas Control Technologies, GHGT-13, 14-18

November 2016, Lausanne, Switzerland

Carbon dioxide conversion into liquid fuels by hydrogenation and

photoelectrochemical reduction: Project description and preliminary
experimental results
Alberto Pettinau*, Mauro Mureddu, Francesca Ferrara
Sotacarbo S.p.A., Grande Miniera di Serbariu, 09013 Carbonia, Italy

Abstract

The increasing attention toward climate changes and the recent strategic policies to stabilize and reduce CO2 emission
is promoting research in the field of carbon capture, storage and more recently utilization technologies, whereby CO2
is captured from industrial flue gas, reused if possible or permanently stored in geological formations, such as depleted
oil or gas fields and saline aquifers.
In this scenario, Sotacarbo is leading the “Centre of Excellence on Clean Energy” research project, funded by the
Regional Government of Sardinia, which aims to develop technologies for low emission power generation and for
CO2 capture, utilization and storage. Among these technologies, CO2 utilization by hydrogenation and
photoelectrochemical reduction are experimentally studied in the Sotacarbo laboratories.
This work presents a general overview of the researches (carried out in close cooperation with the University of
Cagliari) on CO2 utilization by conversion into liquid fuels, with a description of the research facilities and a hint on
preliminary experimental results.
© 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
© 2017 The Authors. Published by Elsevier Ltd.
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of GHGT-13.
Peer-review under responsibility of the organizing committee of GHGT-13.

Keywords: CO2 utilization; hydrogenation; photoelectrochemical reduction; carbon-to-liquids

* Corresponding author. Tel.: +39-0781-1863086; fax: +39-0781-670552.

E-mail address: [email protected]

1876-6102 © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of GHGT-13.
doi:10.1016/j.egypro.2017.03.1832
6894 Alberto Pettinau et al. / Energy Procedia 114 (2017) 6893 – 6904

1.
Introduction

The utilization of carbon-rich fuels (coal, oil and natural gas) has allowed an outstanding era of prosperity and
advancement for human development. Nevertheless, carbon dioxide concentration in the atmosphere has consequently
risen from about 280 ppm before the industrial revolution to about 390 ppm in 2010. The increase in CO2 emission
can contribute to global warming and climate changes due to the enhanced greenhouse effect. The increasing attention
toward climate changes and the recent strategic policies to stabilize and reduce the amount of CO2 emission [1] is
promoting research in the field of carbon capture, storage and more recently utilization technologies [2], whereby CO2
is captured from industrial flue gas, reused if possible or permanently stored in geological formations, such as depleted
oil or gas fields and saline aquifers.
During the past decades, the concept of CO2 reused was mainly applied to promote the extraction of hydrocarbons
in depleted seams, through enhanced oil recovery (EOR) and enhanced gas recovery (EGR) techniques [3]. But in the
recent years the interest on CO2 conversion into liquid fuels and chemicals has significantly increased, with the
development of several research projects and the construction of the first commercial scale-plant, the George Olah
Renewable Methanol Plant installed near Reykjavik, in Iceland [4].
Several researches are currently focused on catalysis development for CO2 hydrogenation process, which could be
particularly interesting and environmental friendly when CO2 captured from power generation or industrial plants and
hydrogen from renewable sources are available [5,6]. Moreover, among CO2 utilization processes, electrochemical
and photoelectrochemical reduction technologies are very promising [7-9]. The electroreduction of CO2 has been
widely studied and the literature reports excellent overview of possible products obtained by electrochemical CO2
reduction, at different electrode materials and experimental conditions [10-12]. However, research on this topic is still
opened, being high energy efficiency achievable through a combination of high faradaic efficiency for the desired
product, and low cell potential [10]. In this contest, using a photoelectrochemical cell, in which at least one electrode
is a photosensitive material, could represent a valid solution to reduce the cell voltage.
In this scenario, Sotacarbo is leading the “Centre of Excellence on Clean Energy” research project, funded by the
Regional Government of Sardinia, which aims to develop technologies for low emission power generation and for
CO2 capture, utilization and storage. Among these technologies, CO2 utilization by hydrogenation and
photoelectrochemical reduction are experimentally studied at the Sotacarbo laboratories.
This work presents a general overview of the researches (carried out in close cooperation with the University of
Cagliari) on CO2 utilization by conversion into liquid fuels, highlighting the main experimental results and the
potential development of the technologies.

Nomenclature

BDD boron doped diamond

EGR enhanced gas recovery
EOR enhanced oil recovery
GC gas chromatograph
GHSV gas hourly space velocity
PEC photoelectrochemical
PID proportional integral derivative control
PLC programmable logic controller
RWGS reverse water-gas shift reaction
SEM scanning electron microscope
XtL X(coal/biomass/CO2)-to-liquids
 Alberto Pettinau et al. / Energy Procedia 114 (2017) 6893 – 6904 6895

2.
Equipment and procedure

With the aim of developing advanced technologies for CO2 (but also coal and biomass syngas) conversion to liquid
fuels (mainly methanol and dimethilether), Sotacarbo has built up a very flexible bench-scale unit to characterized
catalysts and analyze the performance of the CO2 hydrogenation process and a laboratory for the experimental
development of the photoelectrochemical reduction process.

2.1.
X-to-liquids unit

The X-to-liquids (XtL) unit, which is schematically shown in figure 1, is based on the following sections: (a) gas
distribution and mixing, (b) synthesis reactor, (c) wax separation section, (d) online gas analysis system and (e) liquid
products characterization.

Fig. 1. Simplified scheme of the Sotacarbo XtL bench-scale unit.

The feeding section includes six different lines for gas injection. In particular, the unit can be fed with H2, CO, N2
and CO2 provided in high pressure bottles (equipped with pneumatic electrovalves to close the circuit in case of
emergency); pure gases can be mixed in order to simulate a gas mixture on the basis if the specific aim of the
experimental test. A fifth line can be used to feed the unit with other gases (for example H2S or SO2 to analyze the
effects of a potential contamination of the CO2-rich gas on catalyst activity, thus simulating the potential application
of the technology in integration with pre- and post-combustion capture, respectively). Finally, a sixth line is used to
feed the unit with syngas from a bench-scale fluidized bed coal and/or biomass gasification unit (which is currently
under commissioning and will allow methanol, dymethylether and Fisher-Tropsch liquids production by coal and
biomass indirect liquefaction). The injection of pure gases is managed by four mass flow controller Bronkhorst F-
211CV-1K0-MG-11-V), whereas a Bronkhorst mini Coriolis mass flow controller allows to regulate the injection of
gas mixtures, including syngas.
A mixing chamber allows to mix each gases in order to feed the reactor with a uniform pre-heated flow. With the
aim to protect catalysts, mainly during the experimental tests with syngas from the gasification unit, two filters are
disposed in series downwards of the mixing chamber.
The synthesis section, including the mixing chamber, the reactor and the discharging system is located into a hot-
box, sealed and electrically heated, in order to allow the pre-heating of the reactants, the condensation of condensable
products outside the reactor and the solidification of waxes eventually produced into the reactor. A specific module
allows to trap waxes and to separate them during the shut-down phase.
The fixed bed reactor, which represents the core of the plant, includes a porous plate designed to ensure that the
catalyst bed remains in the same position and inside the furnace's isothermal temperature zone. The porous plate is
6896 Alberto Pettinau et al. / Energy Procedia 114 (2017) 6893 – 6904

made of Hastelloy C and is 20 μm in size. A K-type thermocouple with a high-temperature connector, single reading
point and Inconel sheath is housed inside the reactor (in a fixed position 5 mm above the porous plate) in order to
monitor the process temperature.
The reactor is located into a furnace (which could operate up to 1120 °C), in order to compensate for the significant
heat losses of the catalyst (characterized by a very high surface/volume ratio). The furnace is made of two ceramic
fibers which house the heating resistors. The design of the furnace distributes the capacity along the entire length,
creating an isothermal zone inside which the catalyst bed is located. The furnace therefore has very little mass, thus
providing more precise temperature control due to the system's limited inertia. The heat provided to the catalyst bed
is regulated by a multiloop controller on the basis of the bed temperature and the furnace temperature. This system
can be managed by the user through a PLC (programmable logic controller, which also allows to manage eventual
alarms), based on a PID (proportional integral derivative) controller to heat the furnace.
The XtL plant is equipped with three different reactors (figure 2): (a) a 9.1 mm inner diameter and 304.8 mm long
reactor made of SS316 stainless steel, (b) an identical reactor made of Hastelloy C and (c) a 23.8 mm inner diameter
and 486.9 mm long reactor made of Hastelloy C.
These reactors allow to operate with both commercial catalysts (typically characterized by large particle size, thus
requiring the 23.8 mm reactor) and properly developed catalysts (in powders and typically prepared in small amounts,
thus requiring the smaller reactors). They also allow to work within a wide range of temperatures (up to 800 °C, with
typical values between 220 and 250 °C), pressures (typically 20 to 80 bar, but in general between atmospheric pressure
and 100 bar) and gas hourly space velocity (3000 to 10,000 h-1).

Fig. 2. Overview of the three reactors and the furnace.

At the outlet of the hot box, condensable and incondensable products of the process are collected and separated. To
this aim, the unit is equipped with a specific system (patented by PID Eng&Tech, which provided the whole
equipment) for high pressure separation of unreacted gas and produced liquids. In particular, the separation of liquid
products is carried out through a Peltier cell that allows to reduce the temperature to 5 °C. Liquid products are then
separated into hydrophilic liquids (such as methanol, dymethilether, etc.) and hydrophobic liquids (i.e. hydrocarbons).
Outlet gas is depressurized and sent to a mass flow meter Bronkhorst, mod. F-111B-500-MGD-11-K and to a gas
chromatograph (GC) SRA R-3000 to analyze gas composition. GC is based on (a) a 10 m MolSieve column for H2,
O2, N2, CO and CH4, using Ar as gas carrier; (b) a PoraPlot U column for CO2, C2H6, H2S, COS, C3H8, using He as
gas carrier; (c) a OV1 column for C3-C8 hydrocarbons, using He as gas carrier. After the analysis, purge gas is vented
in atmosphere. In parallel, liquid products are analysed off-line by using a gas chromatograph (Agilent 7890A) coupled
with a mass spectrometer. In particular, since the hydrogenation reaction involves the production of water, the gas
chromatograph includes a 25 m PoraPlot Q cross linked column, with an inner diameter of 0.32 mm. Such a column
is located into the furnace of the GC and subjected to a temperature ramp predetermined to optimize the performance
 Alberto Pettinau et al. / Energy Procedia 114 (2017) 6893 – 6904 6897

of the analyzer. Since the GC does not measure water amount in the liquid product, it is subsequently analysed by the
mass spectrometer.
The design, assembly and operation of the plant are intended to provide the user and work area with proper
protection against shock or burns caused by electricity, mechanical hazards, high temperatures, fire, the effect of fluids
and fluid pressure, the effects of radiation, gas leaks, explosion and implosion.

2.2.
Experimental procedure

The experimental procedure is composed by the following main steps: (a) reactor loading with catalyst, (b) leak
tests, (c) plant start-up, (d) operation phase and on-line analyses and (e) plant shut-down.
During the reactor loading phase, a pre-determined amount of catalyst (between 0.5 and 3 cm3, with reference to
the smaller reactor) is introduced into the reactor through its head; two layers of quartz wool are used to maintain the
catalyst bed in its position. Then the head of the reactor is properly locked by using a dynamometric wrench. The
reactor is introduced into the furnace, located in the hot box, which is closed through its pneumatic door.
Before the starting of the experimental test, a leak test is essential to avoid the leakage of gases and to assure the
safety of the operators. The leak test is carried out by locking each plant outlet and pressurizing the whole plant with
nitrogen, at a pressure 5-10 bar higher than the experimental one. If, after a proper time, the internal pressure remains
constant, it means that no nitrogen leakage occurred during the test and the experimental run can be performed safely.
The start-up phase is the more complex one. The first step consists in the gradual heating of the whole unit, under
a nitrogen flow, at a heating rate of 10 °C/min and up to the operating temperature (typically 250 °C). Then the catalyst
must be activated according – for commercial products – to the specific procedure indicated by the producer: in this
specific case, the procedure consists in the reduction of copper oxide to metallic copper by fluxing the catalyst with
270 ml/min of a mixture of hydrogen (5% by volume) and nitrogen. A third step involves the gradual increasing of
reactor pressure through the injection of the same mixture used for catalyst activation. Finally, the gas mixture selected
for the experimental test is injected into the reactor instead of the activation mixture.
Conventionally, the operating phase begins as soon as the reactor reaches the operating conditions (in terms of
temperature, pressure and gas composition). During this phase, outlet gas is continuously analyzed by the online gas
chromatograph and the liquid products are sampled through proper vials for an off-line analysis by gas chromatography
and mass spectrometry.

2.3.
Photoelectrochemical laboratory

In order to test the photoelectrochemical process with different materials under different operating conditions, a
bench-scale unit is currently under construction at the Sotacarbo laboratories. The unit is composed by a gas feeding
system and the photoelectrochemical (PEC) unit, which includes the cell and a sun simulator.
The gas feeding is currently composed by a bi-component mixture of CO2 and N2. Both these gases are provided
in bottles and their mixture is regulated by using two EL-Bronkhorst FLOW flow meters, both calibrated for a volume
flow between 0.1 and 10.0 Ndm3/h. A photoanode-driven PEC cell, currently under construction, has been designed
for a very flexible operation, whereas the sunlight is simulated by a LOT-QuantumDesign solar simulator. The system
is equipped with a potentiostat/galvanostat PGSTAT204, provided by Metrohm Italiana SpA.

3.
Catalysts development for CO2 hydrogenation

Most of the Sotacarbo research activities in the field of CO2 hydrogenation have been focused on the optimization
of catalysts. In general, many techniques are used for complex oxide catalyst preparations, including, co-precipitation,
sol-gel, solidstate, and mechano-synthesis methods [13-16]. An attractive low-cost new method to prepare complex
oxides and advanced solid state materials is by combustion synthesis [17,18]. Today sol-gel auto-combustion method
has become a very popular approach for preparation of nanomaterials and a number of important breakthroughs in this
6898 Alberto Pettinau et al. / Energy Procedia 114 (2017) 6893 – 6904

field have been made, notably for development of new catalysts and nanocarriers with properties better than those for
similar traditional materials. The specific feature of this technique is that, after local ignition, a self-sustaining reaction
propagates throughout the mixture of precursors, leaving behind the desired product oxides. Precursor gels are often
prepared from an aqueous solution of the metal nitrates and an organic complexing agent such as citric acid,
carboxylate azides, urea or glycine [19-22]. The glycine and the citric acid play two important roles: on one hand, they
are the fuel for the combustion reaction; on the other hand, they form complexes with metal ions preventing the
precipitation of hydroxilated compounds; the nitrate ion is the burning oxidizer.
The above described procedure has the advantages of inexpensive precursors, simplicity, short preparation time,
good quality of the powders, moderate heating; furthermore, it does not require specialized equipment, so large
quantities can be easily prepared.

3.1.
Experimental

Nanocrystalline Cu/ZnO/ZrO2 powders have been prepared following the procedure schematized in form of a flow
diagram in figure 3.

*
(%.-


("
.  
$
'   *
(%.-


("
$
+    $-+$
$
(+
%1$'!

)('-'!(.,

)
(+
 $-$('
("

(&)%!0-$('  

/)(+-$('
-






!%
"(+&-$('

#'$-$('
$'
'
(/!'
 


.-((&.,-$('

 


.'+

Fig. 3. Schematic representation of catalyst preparation procedure.

The chosen molar ratio between copper, zinc oxide, and zirconia has been 0.5/0.2/0.3 respectively. Cu(NO3)3.3H2O
(Aldrich, 99%), Zn(NO3)2.6H2O (Aldrich, 98%), Zr(NO3)4.5H2O (Aldrich, 99%), and NH2CH2COOH, (Aldrich, 99%)
have been used as sources of metal ions and chelating-fuel agent, respectively. Assuming a complete combustion, the
reaction that takes place between nitrates and glycine can be expressed in the following way:
    
  
   
    
        
        
 
 
    
         (1)
   

The starting aqueous solutions have been mixed together into a pyrex beaker with a equivalence ratio of 1.44
between oxidising and reducing agents. The clear light blue solution thus obtained presented a pH of 1.50±0.01.
 Alberto Pettinau et al. / Energy Procedia 114 (2017) 6893 – 6904 6899

Aqueous ammonia (NH4OH at 30%) solution, in some cases, has been then added, drop by drop, under constant stirring
in order to adjust the pH value to about 9.5 and with the aim of study the influence of the pH solution on the
morphological and structural features of the final powder. The dark blue solution has been allowed to evaporate on a
hot plate maintaining the solution temperature at 80-90 °C. As a result of the increasing concentration, the viscosity
has risen due to the crosslinking of carboxylato-metal complexes into a three dimensional structure and, on further
dehydratation, a gel has started to form. The viscous dark blue gel collected after evaporation, has been submitted to
a sudden raising of the temperature up to 250-300 °C, activating the combustion mechanism and generating a dark
brown dry powder (figure 4).

Fig. 4. Sample of catalyst.

4.
CO2 hydrogenation process: preliminary results

This section presents a brief summary of the preliminary results obtained on the CO2 hydrogenation process for
methanol synthesis, represented by the following reaction

         ΔH298K=-49.5 kJ/mol (2)

In terms of thermodynamic equilibrium, the reaction is promoted by low temperatures and high pressures.
The system has been fed with a mixture of H2 (75% by volume) and CO2 (25%), which means a H2/CO2 molar ratio
of 3:1. The reactor (the 9.1 mm one, made of Hastelloy C) has been filled with 0.45 g of a pulverized commercial
catalyst composed by CuO/ZnO/Al2O3/MgO, with the addition of 3 g of pulverized α-Al2O3 (with the aim of improving
the control of temperature profile by diluting the catalyst bed).
After the sealing of the reactor and the leak test, the temperature of the system has been increased up to the operating
value of 250 °C with a heating rate of 10 °C/min. The catalyst activation phase has been carried out by injecting 270
ml/min of a gaseous mixture of hydrogen (5% by volume) and nitrogen, as resulted by a series of preliminary tests.
Then pressure is gradually increased up to 40 bar.
As temperature and pressure have reached the operating condition for the specific test, the unit has been fed with
the selected gas mixture for the operating phase. In particular, a 300 Nml/min flow of H2/CO2 mixture with a 3:1 molar
ratio has been sent to the reactor. A small nitrogen flow (30 Nml/min) has been added to simplify the operation of the
gas chromatograph. These conditions correspond to a gas hourly space velocity (GHSV) of about 6000 h-1, referred to
the whole catalytic bed (3 cm3).
6900 Alberto Pettinau et al. / Energy Procedia 114 (2017) 6893 – 6904

During the preliminary test, which operation phase has been characterized by a duration of about 3 hours, the liquid
production (at the exit of the liquid/liquid/gas separator, with a condensation temperature of the Peltier cell set at 15
°C) was 1.7 ml; one more milliliter of liquid products remains into the system to assure the pressurization of the
separator. As expected, all the liquid product has been collected at the exit of the hydrophilic liquids, whereas no
production has been collected at the exit of the hydrophobic liquids: this is a first indication of the correct operation
of the catalysis, that promotes the production of polar liquids and not of hydrocarbons.
As mentioned, purge gas is analysed by a GC. Figure 5 shows two chromatograms of gaseous effluent during the
preliminary tests, as measured by both the MolSieve and the PoraPlot U columns.

Fig. 5. Purge gas chromatogram with (a) MolSieve column and (b) PoraPlot U column.

The main component found in purge gas are non-reacted hydrogen and carbon dioxide, together with nitrogen and
small amounts of carbon monoxide, which formation comes from the so-called reverse water-gas shift reaction
(RWGS)

        ΔH298K=+37.6 kJ/mol (3)

The analysis of purge gas composition with respect to the inlet gas mixture shows an overall CO2 conversion rate
into methanol of about 20%, this confirming most of the results available in the scientific literature.
As mentioned, liquid products, an aqueous solution of oxygenated compounds, are sampled and analysed off-line
by a GC coupled with a mass spectrometry. Figure 6 shows the result of the GC analysis.

Fig. 6. Liquid effluent chromatogram.

It can be observed that the outlet of signals referred to different species takes place at different times and there is
not any peak overlapping. The identification of each component by using the mass spectrometer indicates that the
main components are, as expected, methanol and water (the two products of the hydrogenation reaction) with small
amounts of nitrogen and unreacted carbon dioxide, absorbed in the aqueous solution. A very small peak can be noticed
 Alberto Pettinau et al. / Energy Procedia 114 (2017) 6893 – 6904 6901

at a high retention time; it seems to refers to a compound (not identified yet) characterized by a boiling point slightly
higher than the methanol one.

5.
Photoelectrochemical reduction of CO2: preliminary analyses

A series of preliminary experimental tests have been carried out, in close cooperation with the University of
Cagliari, to optimise the design of the photoanode-driven PEC bench-scale unit.
This section provides a general overview of the preliminary results, whereas more information can be found in
previous publications by the authors [23].Several experimental tests have been carried out to select the material of the
electrodes. Semiconductor materials based on metal oxides are very promising as photoanode and several tests have
been performed on TiO2, which is characterized by low cost, low toxicity, high stability and high reactivity. In
particular, nanotubular structures of TiO2 have been produced (with a nominal surface up to 15 cm2) and characterized.
In order to verify the uniformity of the structure and its photoactivity, cyclic voltammetries and photo-currents were
measured at different wave lenghts. SEM (scanning electron microscope) analysis was performed to investigate on the
morphology of the structures. Figure 7 shows some examples of the results of the SEM analyses performed at different
section points in order to verify the uniformity of the structures. If the border line effects are excluded, a regular
structure has been observed with a mean tube diameter between 40 and 50 nm measured in different points of the
sample [23].

Fig. 7. SEM micrographs of photoanode at three investigation points [23].

Similar results have been obtained with the different samples prepared under the same conditions, confirming a
good uniformity and, overall, a good reproducibility of the synthesis. Moreover, in order to improve the final
photocatalytic performance in the near visible range, the possibility to couple TiO2 with another semiconductor (e.g
tungsten oxide) has been considered. Such a coupling allows to shift the sensibility of the photoanode toward the
highest wave lengths, thus increasing the efficiency of the process.
As shown in figure 8, which represents the effect of electrodeposition method on the photocurrent density response
at different wave lengths, a slight increase of the performance can be observed at samples where tungsten oxide was
deposited (NTWx samples) with respect to the TiO2 (NT) samples. Moreover, as expected, at the highest wave length
lower values of photocurrent are obtained (full bars) [23].
6902 Alberto Pettinau et al. / Energy Procedia 114 (2017) 6893 – 6904

Fig. 8. Effect of electrodeposition method on the photocurrents density response at different wave lenghts [23].

As for the cathode, copper (Cu) and copper oxide (CuO) deposition on different kinds of supports (i.e. boron doped
diamond - BDD, graphite and titanium nanotubes) has been tested by using different electrolytes, such as NaHCO3,
KHCO3, KCl and KNO3. The experimental tests shows that the deposition of CuO presents a strong instability, since
the negative potential involves the reduction of CuO to metallic Cu. The preparation and characterization of other
materials is still in progress.

6.
Conclusions

Due to the increasing interest in CO2 utilization technologies, Sotacarbo is managing different kinds of research
activities in the fields of carbon dioxide catalytic hydrogenation and in photoelectrochemical reduction of CO2.
Catalytic hydrogenation processes are almost mature from the industrial point of view, but further researches should
be done to improve the process and to reduce the costs. In parallel, photoelectrochemical reduction processes are still
far from the commercial phase, but they are very promising for a mid-term future.
The very flexible X-to-liquids (XtL) facility recently built up at the Sotacarbo Research Centre allows the operating
characterization of commercial catalysis and the development and optimization of advanced catalysts. Preliminary
tests have been carried out to synthesize methanol by using a commercial catalyst. The tests, carried out with 300
Nml/min of a mixture of H2 and CO2 (with a molar ratio of 3:1) at 250 °C and 40 bar and with a GHSV of 6000 h-1,
have allowed the production of about 1 ml/hour of liquid products, mainly proposed by an aqueous solution of
methanol. In parallel, several nanocrystalline Cu/ZnO/ZrO2 pulverized catalysts (with a molar ratio between copper,
zinc oxide, and zirconia of 0.5/0.2/0.3) have been properly prepared through the sol-gel auto-combustion method and
preliminarily characterized.
As for the photoelectrochemical reduction of CO2, even if the experimental facility is still not ready for operation,
several preliminary tests have been carried out to optimize the material for both the photoanode and the cathode. In
particular, a photoanode-driven PEC cell is considered; the lowering of the cell voltage is obtained from the
photopotential generated at the nanotubular structure of TiO2, irradiated by a suitable light wavelenght. This anode
could be coupled with metals or metal composite oxides. To this aim, nanotubular structures of TiO2 (without and
with the deposition of tungsten oxide) have been produced and characterized. The results confirm the quite regular
structure of the samples, in terms of morphology and photoactivity. Moreover, the results indicate the presence of
tungsten oxide on the nanotubular structure of the electrodes as well as its positive effect on the photoactivity. As for
the cathode, copper (Cu) and copper oxide (CuO) deposition on different kinds of supports have been tested by using
different electrolytes.
 Alberto Pettinau et al. / Energy Procedia 114 (2017) 6893 – 6904 6903

Even if the results here presented are preliminary, the very flexible experimental facilities allow a wide series of
experimental tests, which will be carried out soon. Most of the effort will be made to the development of advanced
catalysts for methanol synthesis by CO2 hydrogenation and to the optimization of the electrodes for the
photoelectrochemical reduction process.

Acknowledgements

The work here presented has been carried out within the “Centre of Excellence on Clean Energy” research project
(CUP D82I3000250001), funded by the Regional Government of Sardinia. The authors would like to thank prof.
Simonetta Palmas and prof. Elisabetta Rombi (University of Cagliari) for their support.

References

[1] United Nations Framework Convention on Climate Change. Paris Agreement. Available at
http://unfccc.int/files/essential_background/convention/application/pdf/english_paris_agreement.pdf [accessed: 26/07/2016].
[2] Mukherjee D, Park S-E, Reddy BM. CO2 as a soft oxidant for oxidative dehydrogenation reaction: An eco benign process for industry. J CO2
Utilization 2016;16:301-312.
[3] Jeon SH, Kim MS. Compressor selection methods for multi-stage re-liquefaction system of liquefied CO2 transport ship for CCS. Appl Thermal
Eng 2015;82:360-367.
[4] Carbon Recycling International website www.carbonrecycling.is [accessed 24/07/2016].
[5] Al-Kalbani H, Xuan J, García S, Wang H. Comparative energetic assessment of methanol production from CO2: Chemical versus
electrochemical process. Appl Energy 2016;165:1-13.
[6] Rivarolo M, Bellotti D, Magistri L, Massardo AF. Feasibility study of methanol production from different renewable sources and thermo-
economic analysis. Int J Hydrogen Energy 2016;41:2105-2116.
[7] Liu J, Shi H, Huang X, Shen Q, Zhao G. Efficient Photoelectrochemical Reduction of CO2 on Pyridyl Covalent Bonded Ruthenium(II) Based-
Photosensitizer. Electrochimica Acta 2016;216:228-238.
[8] Zhang Y, Luc W, Hutchings GS, Jiao F. Photoelectrochemical Carbon Dioxide Reduction Using a Nanoporous Ag Cathode. ACS Applied
Materials and Interfaces 2016;8:24652-24658.
[9] Ganesh I. Electrochemical conversion of carbon dioxide into renewable fuel chemicals - The role of nanomaterials and the commercialization.
Ren Sust En Rev 2016;59:1269-1297.
[10] Centi G, Quadrelli EA, Perathoner S. Catalysis for CO2 conversion: a key technology for rapid introduction of renewable energy in the value
chain of chemical industries. Energy Environ Sci 2013;6,1711-1731.
[11] Magesh G, Kim ES, Kang HJ, Banu M, Kim JY, Kimb JH, Lee JS. A versatile photoanode-driven photoelectrochemical system for conversion
of CO2 to fuels with high faradaic efficiencies at low bias potentials. Journal of Materials Chemistry A 2014;2:2044-2049.
[12] Jhong HR, Ma S, Kenis PJA. Electrochemical conversion of CO2 to useful chemicals: current status, remaining challenges, and future
opportunities. Current Opinion in Chemical Engineering 2013;2:191-199.
[13] Xi JY, Wang ZF, Lu GX. Improvement of Cu/Zn-based catalysts by nickel additive in methanol decomposition. Appl Catal A Gen
2002;225:77–86.
[14] Yuh J, Nino JC, Sigmund WA. Synthesis of barium titanate (BaTiO3) nanofibers via electrospinning. Mater Lett 2005;59:3645–3647.
[15] Shi M, Liu N, Xu YD, Wang C, Yuan YR, Majewski P, Aldinger F. Preparation of electrolyte foils La0.85Sr0.15Ga0.85Mg0.15O2.85 (LSGM)
by means of tape casting J Mater Process Technol 2005;169:179–183.
[16] Zyryanov VV, Sadykov VA, Uvarov NF, Alikina GM, Lukashevich AI, Neophytides S, Criado JM . Mechanosynthesis of complex oxides
with fluorite and perovskite-related structures and their sintering into nanocomposites with mixed ionic
electronic conductivity. Solid State
Ionics 2005;176:2813–2818.
[17] Patil KC, Aruna ST, Mimani T. Combustion synthesis: an update Curr Opin Solid State Mater Sci 2002;6:507–512.
[18] Mukasyan AS, Epstein P, Dinka P. Solution combustion synthesis of nanomaterials (2007) Proc Combust Inst 2007;31(2):1789–1795.
[19] Pederson LR, Maupin GD, Weber WJ, McReady DJ, Stephen RW.. Combustion synthesis of YBa2Cu3O7)x glycine/metal nitrate method.
Mater Lett 1991;10:437–443[J] Roy S, Sharma A,. Roy SN, Maito HS.Synthesis of YBa2Cu3O7)x powder by autoignition of citrate–nitrate
gel. J Mater Res 1993;8: 2761–2766.
[20] Chakraborty A, Sujatha Devi P, Roy S, Maiti HS,. Low-temperature synthesis of ultrafine La0.84Sr0.16MnO3 powder by an autoignition
process. J Mater Res 1994;9: 986–991.
[21] Devi PS, Maiti HS. A modified citrate gel route for the synthesis of phase pure Bi2Sr2CaCu2O8 superconductor. J Mater Res 1994; 9:1357–
1362.
6904 Alberto Pettinau et al. / Energy Procedia 114 (2017) 6893 – 6904

[22] Shafer S, Sigmund W, Roy S, Aldinger F. Low temperature synthesis of ultrafine Pb(Zr,Ti)O3 powder by sol–gel combustion. J Mater Res
1997; 12;2518–2521.
[23] Palmas S, Mascia M, Vacca A, Ampudia P, Mais L, Ferrara F, Pettinau A. On the behavior of modified TiO2 nanotubes for a photoanode-
driven photoelectrochemical reduction of CO2. Technical Proceedings of the 2015 TechConnect World Innovation Conference & Expo,
Washington, DC, USA, June 14-17, 2015:128-131.