Electrochemical reactor for sustainable transformation of bio-mass derived allyl alcohol into acrylate and pure hydrogen

Article abstract of DOI:10.1016/j.ica.2021.120488

Acrylic acid is widely used in the chemical, polymer, cosmetic and food industries. Typically, it is produced through processes with a high environmental impact. In this paper, we demonstrate the co-production of the potassium acrylate salt and hydrogen gas from allyl alcohol in a liquid flow fed anion exchange membrane electrolysis cell operating at 60 °C and ambient pressure. We compare in electrolysis cell tests, two electrocatalysts Pd/C and Pd-CeO₂/C evaluating the activity and selectivity for acrylate production. Electrolysis cell parameters are tuned obtaining a maximum conversion of allyl alcohol of 96% and a selectivity to acrylate of 50% at an operating cell voltage of 1 V. Operating at a lower cell potential (0.7 V) the selectivity for acrylate increases to 74%. Hydrogen gas is produced in the separated cathode compartment at an energy cost of 26 KWh kg_H2^-1, which is around half when compared to state-of-the-art water electrolyzers. The electrochemical oxidation mechanism of allyl alcohol is also studied and discussed, providing for the first time an insight into the pathways for formation of acrylate with respect to the other principle oxidation products (propionate and 3-hydroxypropionate).

Full text of DOI:10.1016/j.ica.2021.120488

Inorganica Chimica Acta 525 (2021) 120488
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Research paper
Electrochemical reactor for sustainable transformation of bio-mass derived
allyl alcohol into acrylate and pure hydrogen
,
Maria Vincenza Pagliaroa, Hamish Andrew Millera *, Marco Bellinia, Benedetto Di Vicoa,
,
*
Werner Oberhausera, Giovanni Zangarib, Massimo Innocentic, Francesco Vizzaa
a Institute of Chemistry of Organometallic Compounds – National Research Council of Italy (ICCOM-CNR), Via Madonna del Piano 10, 50019 Sesto Fiorentino
(Florence), Italy
b University of Virginia, Dept Mat Sci & Engn, Charlottesville, VA 22904 USA
c University of Florence, Dept Chem Ugo Schiff, Via Lastruccia 3, I-50019 Sesto Fiorentino, FI, Italy
A R T I C L E I N F O
A B S T R A C T
Keywords:
Acrylic acid is widely used in the chemical, polymer, cosmetic and food industries. Typically, it is produced
through processes with a high environmental impact. In this paper, we demonstrate the co-production of the
potassium acrylate salt and hydrogen gas from allyl alcohol in a liquid flow fed anion exchange membrane
electrolysis cell operating at 60 ◦C and ambient pressure. We compare in electrolysis cell tests, two electro-
catalysts Pd/C and Pd-CeO2/C evaluating the activity and selectivity for acrylate production. Electrolysis cell
parameters are tuned obtaining a maximum conversion of allyl alcohol of 96% and a selectivity to acrylate of
50% at an operating cell voltage of 1 V. Operating at a lower cell potential (0.7 V) the selectivity for acrylate
increases to 74%. Hydrogen gas is produced in the separated cathode compartment at an energy cost of 26 KWh
kg-H12, which is around half when compared to state-of-the-art water electrolyzers. The electrochemical oxidation
mechanism of allyl alcohol is also studied and discussed, providing for the first time an insight into the pathways
for formation of acrylate with respect to the other principle oxidation products (propionate and 3-
hydroxypropionate).
Electrochemical reforming
Electrolysis
Hydrogen
Pd-CeO2/C
Acrylic acid
Allyl alcohol
1. Introduction
AcA in the gas phase over Mo ꢀ V ꢀ W ꢀ O catalysts. An overall AcA
yield of 80% was achieved. The current process of oxidizing AA to AcA
Acrylic acid (AcA) is a precursor to acrylic based polymers, useful for
their absorbent properties and used in a wide range of applications such
as surface coatings, paint formulations, textiles, adhesives, paper treat-
ment, leather, fibers and detergents [1–3]. Worldwide production of
acrylic acid was 4.5 million tons in 2012 with a growing demand of 4%
per year [3]. AcA is currently produced from crude oil through the two-
step gas-phase oxidation of propylene, producing acrolein as an inter-
mediate [3]. The growing industrial demand for AcA, combined with the
need for a progressive reduction of dependence on fossil fuels, means we
need new production processes based on renewable biomass resources.
Glycerol (Gly), is a major byproduct of the biodiesel industry, has
attracted attention as a precursor for the production of various chem-
icals (propanediol, lactic acid, acrolein, allyl alcohol) and their conse-
quent conversion to AcA [3,4]. For example, Li and Zhang [5] report a
novel two-step method through glycerol deoxydehydration (DODH)
using formic acid to form allyl alcohol (AA), followed by oxidation to
involves high temperatures (up to 340 ◦C in the second step), the use of
pure oxygen, and the co-production of a CO and CO2 waste mixture.
More recently, Ueda et al. studied the oxidation of AA using four kinds of
crystalline (orthorhombic, trigonal, tetragonal and amorphous) MoVO
catalysts, and investigated the active sites for the reaction. The
maximum yields of AcA were 73% for the orthorhombic Mo3VOx cata-
lyst and 72% for the trigonal Mo3VOx catalyst at 350 ◦C [3,6]. Kim and
Lee have developed the liquid-phase oxidation of AA to AcA using an Au-
CeO2 catalyst; a 92% yield of AcA, at 25 ◦C and 10 bar O2 pressure in the
presence of NaOH, was obtained [7]. In this work, we demonstrate a
process for the preparation of the potassium salt of acrylic acid (potas-
sium acrylate) from AA through electrochemical oxidation in a mem-
brane electrolysis flow cell at 60 ◦C without the need for high pressure
O2. As shown in our previous works, this technology presents numerous
advantages combining the production of valuable chemicals from
biomass-derived starting materials with the production of clean
* Corresponding authors.
E-mail addresses: hamish.miller@iccom.cnr.it (H.A. Miller), francesco.vizza@iccom.cnr.it (F. Vizza).
https://doi.org/10.1016/j.ica.2021.120488
Received 30 March 2021; Received in revised form 27 May 2021; Accepted 31 May 2021
Available online 2 June 2021
0020-1693/© 2021 Elsevier B.V. All rights reserved.

M.V. Pagliaro et al.
Inorganica Chimica Acta 525 (2021) 120488
was used as substrate for the deposition of the catalyst ink. Before the
deposition, the glassy-carbon surface was polished with different CT
diamond suspensions featuring progressively smaller particle size (1 µm,
0.25 µm, 0.1 µm) and finally washed with distilled water. The catalyst
ink is composed of Pd/C or Pd/C-CeO2 (7 mg) in a 50:50 water/ethanol
solution (600 mg water and 600 mg ethanol) with about 12 mg of a 5%
Nafion® perfluorinated resin solution. The resulting suspension was
sonicated for 1 h with a FALC sonic bath to reach a uniform suspension.
The catalyst film was prepared by casting 5 µL of the catalyst ink onto
the glassy-carbon electrode. The exact amount of ink deposited was
determined using an analytical balance. The final metal loading on the
electrode is between 4 and 6 µg. Each electrode was dried for 30 min
before and finally it was mounted on the rotating disk electrode shaft
and immersed into the electrolyte solution. The reference electrode was
a commercial Ag/AgCl/KCl (1 M). The counter electrode was a platinum
gauze enclosed in a glass tube with porous bottom. All the potentials
reported are versus the reversible hydrogen electrode (RHE). The CV
experiments were undertaken in static N2 saturated aqueous solutions of
both 0.5 M KOH and 0.5 M KOH + 0.5 M allyl alcohol. All the solutions
were prepared with Millipore water (18 MΩ*cm) provided by a milli-Q
labo apparatus (Nihon Millipore Ltd.). Cyclic voltammetry was under-
Fig. 1. Schematic representation of the electrolysis cell.
taken from 0 to 1.4 V (vs RHE) at a scan rate of 10 mV sꢀ 1
.
hydrogen at low temperature and atmospheric pressure (see Fig. 1)
[8–11].
2.3. Electrolysis cell tests
The selectivity of the oxidation reaction for the desired product may
be tuned by varying the cell operating parameters (e.g. temperature and
anode potential). The nature of the anode catalyst may also influence the
oxidation selectivity [12]. To this regard, we compare two electro-
catalysts (Pd/C and Pd/C-CeO2) at the anode, evaluating the activity and
product selectivity. The Pd-CeO2/C catalyst has been shown to exhibit
enhanced alcohol oxidation activity [9,11]. The CeO2 promotes the
formation at low potentials of species adsorbed on Pd, (e.g. PdI-OHad),
that are involved in alcohol oxidation [13]. Electrolysis cell parameters
are also tuned to favor the formation of acrylate with respect to the main
The MEAs (membrane electrode assemblies) were composed of a
nickel foam support coated with either the Pd/C or Pd-CeO2/C catalyst
as anode, a commercial Tokuyama A-201 anion-exchange membrane,
and a commercial 40 wt% Pt/C (Aldrich) catalyst supported on carbon
cloth as cathode. The anode ink was prepared by mixing the catalyst
with a 5% aqueous suspension of PTFE to form a dense catalyst paste
which is spread onto a 5 cm2 Ni-foam support (Heze Tianyu Technology
Development Co., China) in order to obtain a catalyst coated electrode
with a Pd metal loading of 1 mg cmꢀ 2. The cathodic ink was prepared in
a 5 mL high density polyethylene vial, mixing 200 mg of the commercial
Pt (40 wt%)/C in 450 mg of distilled water, 790 mg of 1-propanol and
1.56 g of the ionomer Nafion® (5 wt% in 2-propanol). The mixture was
suspended with three pulses of ultrasound, 20 W power at the frequency
of 20 kHz (Bandelin Sonor pulse UW 2200 SERIES). Finally, this paste
was spread onto a carbon cloth W1S1005 (CeTech Co. Ltd.) gas diffusion
layer, with a Meyer rod (n◦150) obtaining a 0.4 mg cmꢀ 2Pt loading.
The electrochemical flow cell was purchased from Scribner-
Associates (USA). The MEAs were assembled by mechanically pressing
together the anode, cathode and membrane within the cell hardware.
The cell temperature was regulated at 60 ◦C using a Scribner 805e fuel
cell station. The aqueous fuel solution (0.5 M KOH + 0.5 M allyl alcohol)
was delivered to the anode at 1 mL minꢀ 1. Voltage scans and galvano-
static curves were determined using an ARBIN BT-2000 5A-4 channels
instrument. Polarization experiments were recorded by applying a linear
voltage ramp with a 10 mV sꢀ 1 scan rate between 0.2 and 1.2 V.
Chronopotentiometry experiments were performed applying a constant
electrolysis current of 125 mA until the cell voltage reached the value of
1.2 V. Amperometry experiments were also performed tuning cell po-
tential values from 0.4 V to 1 V. The hydrogen generation flow was
recorded by a Bronkhorst EL-FLOW mass flow meter model F-101C-002-
by-products (propionate and 3-hydroxypropionate) obtaining
a
maximum conversion of AA of 96% and a selectivity to Ac of 50% at a
constant operating voltage of 1 V. At the same time pure hydrogen is
produced at cathode side with an energy consumption of only 26 KWh
kgH2 -1, about the half when compared to state of the art water elec-
trolyzers [14]. To the best of our knowledge this is the first time AA has
been used as substrate in a complete electrochemical reforming cell with
the objective of producing acrylate. The reaction mechanism was
investigated, highlighting the existence of two separated oxidation
processes: a 4 e- electrochemical path for the oxidation of AA to AcA and
the simultaneous isomerization of allyl alcohol to propionaldehyde,
which is oxidized to propionate through a 2 e- path.
2. Experimental
2.1. Materials
Carbon black (Vulcan XC-72) was purchased from Cabot Corp., USA.
All metal salts and reagents were purchased from Aldrich and used
without further purification. A commercial 40 wt% Pt/C catalyst was
used as (Hydrogen Evolution) cathode catalyst, while the Pd/C and Pd/
C-CeO2 anode catalysts were prepared as already reported in our pre-
vious works [15]. The anion exchange membrane used was A-201, ob-
tained from Tokuyama Corp. (Japan).
AGD-11-V with a maximum H2 flow rate of 3 mL minꢀ 1
.
2.4. Model reaction investigations and product analysis
Model catalytic reactions were performed in a PTFE-coated, stainless
steel autoclave (volume 50 cm3), equipped with a magnetic stirrer and
temperature and pressure controllers. The fuel solutions after each
experiment were quantitatively and qualitatively analyzed by 13C{1H}
NMR spectroscopy and HPLC. A UFLC Shimadzu Chromatograph
equipped with refraction index detector (RID) was used; the column is a
GRACE- Alltech OA-1000 Organic Acids (300 mm × 6.5 mm),
2.2. Electrochemical tests
The electrochemical measurements were carried out using a Parstat
2273 potentiostat–galvanostat (Princeton Applied Research) equipped
with a Model 616 Rotating Disk electrode (PAR-Ametek). A 5 mm (A =
0.1963 cm2) Teflon potted glassy-carbon disk electrode tip (PINETM)
2

M.V. Pagliaro et al.
Inorganica Chimica Acta 525 (2021) 120488
Fig. 2. TEM images of representative portions of (a) Pd/C-CeO2 and (b) Pd/C.
Fig. 3. CVs in N2 saturated aqueous solutions of (a) 0.5 M KOH (b) 0.5 M allyl alcohol + 0.5 M KOH at 10 mV sꢀ 1
.
thermostated at 65 ◦C. The eluent is 0.01 N H2SO4; and the eluent flow is
0.4 mL minꢀ 1. NMR spectra were acquired with a Bruker Avance DRX
400 spectrometer. Chemical shifts (δ) are reported in ppm relative to
TMS. Deuterated solvents (Sigma-Aldrich) used for NMR measurements
were dried with activated molecular sieves; 1,4-dioxane was used as
internal standard for product quantification.
reforming cell.[9,11] In this study Pd-CeO2/C is compared to Pd/C with
AA as substrate.
3.1. Electrochemical study of allyl alcohol oxidation
The electrochemical characteristics of Pd/C and Pd/C-CeO2 were
investigated by cyclic voltammetry (CV) in both 0.5 M KOH (Fig. 3a) and
0.5 M allyl alcohol + 0.5 M KOH (Fig. 3b).
3. Results and discussions
The CVs of Pd/C and Pd/C-CeO2 were firstly obtained in N2 saturated
0.5 M KOH (Fig. 3a) and show features characteristic of Pd-centered
transitions that have been assigned by comparison with literature data
[11,15,16]. On the forward anodic scan (0 to 1.4 V), there are three
transitions (A1-A2-A3). The broad peak A1, centered at 0.4 V, is due to
the oxidative desorption of hydrogen. The shallow approach to the
limiting current at about 0.6 V (A2) is ascribed to the formation of
palladium hydroxides. This is followed by the formation of an oxide
layer (A3) on the palladium surface that is associated with a gentle in-
crease in the anodic current density. On the reverse cathodic scan (1.4 to
0 V), the peak at 0.7 V (C1) is assigned to the reduction of Pd oxides,
whereas the broad cathodic feature at potentials between 0.5 and 0.2 V
The Pd/C and Pd/C-CeO2 anode electrocatalysts with 10 wt% Pd
have been fully characterized in our previous reports through XRD,
chemisorption, XAS and HR-TEM.[16] Representative TEM images of
both catalysts are shown in Fig. 2. As demonstrated in our previous
studies, the ceria in the support plays a key role in promoting the for-
mation, at low potentials, of hydroxide species adsorbed on Pd (PdI-
–OHad) that leads to enhanced electrocatalytic activity. Under strongly
alkaline conditions (pH ≥ 13), on palladium (Pd) electrocatalysts, al-
cohols are generally partially oxidized to carboxylate compounds
without formation of CO2.[12] The Pd/C-CeO2 catalyst readily trans-
forms a ranged of alcohols into carboxylates in the electrochemical
Fig. 4. MEA (1): (a) potentiodynamic experiments (b) galvanostatic experiments at 50 mA cmꢀ 2. Cell temperature 60 ◦C.
3

M.V. Pagliaro et al.
Inorganica Chimica Acta 525 (2021) 120488
Table 1
Electrolysis data with allyl alcohol.
TEST
Type of experiment
@Electrochemical Conversion/%
$Selectivity/%
H2/L minꢀ 1 mꢀ 2
Energy Consumption /KWh kgH2
*Ac *Prp *3-Hp
Pd/C-CeO2
(1)
(2)
(3)
(4)
JK = 50 mA cmꢀ 2
Vk = 0.4 V
64
14
63
96
41.3
44.3
43.0
42.7
50.4
8.3
1.6
4.4
3.1
1.17
0.25
1.15
1.95
20.2
10.8
18.7
26.8
54.1
52.6
53.2
Vk = 0.7 V
Vk = 1 V
Pd/C
(5)
JK = 50 mA cmꢀ 2
Vk = 0.4 V
19
26
61
96
34.3
34.2
73.9
50.1
63.3
22.7
16.7
49.8
2.4
43.1
9.4
3.12
0.01
0.98
1.54
24.7
10.7
18.7
26.8
(6)
(7)
Vk = 0.7 V
(8)
Vk = 1 V
1.1
◦Galvanostatic (constant current density jk), Potentiostatic (constant voltage Vk);
*Acrilate (Ac), Propionate (Prp), 3-hydroxypropionate (3-Hp);
@
Considering the double oxidation pathway at 4 and 2 e-, and a propionate/acrylate ratio of 1.2;
Data from 13C{1H} NMR spectroscopy and HPLC analysis.
$
(C2) is attributed to hydrogen adsorption. The CV of Pd-CeO2/C has also
the characteristic underlying redox features of the CeO2 part of the
support that are visible in the region 0 to 0.6 V (RHE).
Table 2
Comparison of electrochemical reforming data with Pd-CeO2/C at the anode at
Tcell of 60 ◦C. 1,2-P = 1,2-propandiol, 1,3-P = 1,3-propandiol, EG = ethylene
glycol, 1,4-B = 1,4-butanediol. Data from refs. 9 and 11 were obtained at a
The electrochemical activity of Pd/C and Pd/C-CeO2 towards the
electrooxidation of allyl alcohol was also investigated by cyclic vol-
tammetry in a solution of 0.5 M AA + 0.5 M KOH. The forward anodic
scans reported in Fig. 3b have the onset potential for AA oxidation at
approximately 0.5 V (vs RHE) and the recorded oxidative current den-
sity increases continuously with increasing potential. The subsequent
current decrease after the peak (0.8 to 1.0 V) corresponds to the pro-
gressive formation of an inactive palladium oxide layer [17]. In the
reverse scan there is e a small peak in current density (A2) at around 0.7
V, due to the reactivation of the palladium surface (removal of surface
oxides) that are active for allyl alcohol oxidation. The electrochemical
characteristics of allyl alcohol oxidation by Pd/C and Pd/C-CeO2
determined here are consistent with the activity profile with a range of
other alcohols previously reported (e.g. ethanol, methanol, ethylene
glycol, glycerol, 1,2-propandiol, 1,3-propandiol and 1,4-butandiol)
[9,11].
current density of 30 mA cmꢀ 2
Fuel
.
Jmax @ 1
V
H2
Energy
Conversion
(%)
Ref.
(mmol)
Consumption
(mA
(kWh kg-1
)
H2
cmꢀ 2
)
HCOOK
Methanol
Ethanol
1,2-P
355
425
444
420
336
372
395
400
150
46
6.0
13.4
14.5
18.9
16.9
15.6
17.5
16.3
20.2
77
81
96
77
71
52
67
68
64
9
85
9
68
9
107
96
9
EG
9
Glycerol
1,4-B
92
9
102
93
9
1,3-P
11
This
work
AA
40
present compared to 19% for the Pd/C catalyst.
In order to identify and quantify the ally alcohol electrooxidation
product distribution for each catalyst, the fuel solutions after constant
current density experiments (50 mA cmꢀ 2) were quantitatively and
qualitatively analyzed by 13C{1H} NMR spectroscopy and HPLC. The
formation of three main products was detected and quantified: acrylate,
propionate and 3-hydroxypropionate. Additionally, the electrical energy
consumption of the process was determined from the integration of the
instantaneous charging power over the duration of the galvanostatic
experiment. The H2 produced during each run was determined using a
flow meter connected to the outlet of the cathode compartment. Results
are summarized for all tests in Table 1.
3.2. Electrolysis cell testing
The anode, membrane, and cathode were pressed together within the
cell hardware to form the membrane electrode assembly (MEA) for
testing. The anode compartment of the cell was fed by recycling 50 mL of
aqueous solutions of 0.5 M allyl alcohol + 0.5 M KOH while the cathode
compartment was connected to a flow meter to measure the H2 pro-
duction. The cell performance was firstly measured using potentiody-
namic scans at a cell temperature of 60 ◦C. The potentiodynamic scans
(examples are shown in Fig. 4a) were performed from 0 to 1.2 V at 10
mV sꢀ 1. The curves reach approximately 150 mA cmꢀ 2 maximum cur-
rent density at 1 V for Pd/C and 1.2 V for Pd-CeO2/C respectively. Under
the same cell conditions higher current densities have been obtained
with other alcohols, demonstrating the slower kinetics of AA electro-
oxidation due to the strong adsorption of intermediates, as will be dis-
cussed in detail below (see Table 2 for comparison). For example, Pd-
CeO2/C achieved up to 400 mA cmꢀ 2 at 1 V for 1,3-propandiol and
Ethylene Glycol oxidation under the same conditions [11]. The current
density drops after 1 V for Pd/C while for Pd-CeO2/C the CD increases up
to 1.2 V. Demonstrating a poisoning of the Pd/C catalyst by progressive
oxidation that does not occur below 1 V on Pd-CeO2/C. Pd-CeO2/C has
been shown to be more stable under potential cycling in alkaline media
[16]. A constant current experiment (50 mA cmꢀ 2) using a single batch
of fuel (50 cm3) was run for each catalyst. From Fig. 4b it is clear that the
Pd-CeO2/C catalyst is more stable able to transform 64% of the AA
To study the AA conversion efficiency and selectivity to AcA for each
catalyst, a series of potentiostatic (constant cell voltage) experiments
were carried out applying three cell potentials: TEST (2) which operates
at 0.4 V, TEST (3) at 0.7 V and TEST (4) at 1 V. The corresponding
potentiostatic curves are shown in Fig. 5a and 5b and their extrapolated
electrochemical data are reported in Table 1. At the highest operating
voltage of 1 V, the catalysts show the highest activity for AA oxidation,
starting at 140 and 100 mA cmꢀ 2 current density for Pd-CeO2 and Pd/C
respectively. The AA total conversion at 1 V reaches 96% for both cat-
alysts (Table 1). This result is consistent with the cell voltage scan ex-
periments (Fig. 2a) where the cell exhibits the maximum
electrochemical activity for AA oxidation at approximately 1 V. At lower
cell potentials, significantly lower current densities and conversion ef-
ficiencies are observed. The current density decreases over time due to
4

M.V. Pagliaro et al.
Inorganica Chimica Acta 525 (2021) 120488
Fig. 5. Potentiostatic experiments carried out at cell voltages of 0.4 V, 0.7 V, and 1 V for the cell equipped with (a) Pd/C-CeO2 and (b) Pd/C anode.
Fig. 6. 1H NMR and 13C NMR spectra of the product solution after a typical potentiostatic experiment @1 V.
the progressive consumption of AA, which is oxidized to Acrylate and
other products. The higher energy introduced to the cell at 1 V is what
forces the reaction to a higher conversion (KWh Kgꢀ 1 26.8 vs 10.8 @ 0.4
V).
Table 1. Acrylate (Ac) and propionate (Prp) are the main products
detected together with a small amount of 3-hydroxypropionate (3-Hp).
Typical NMR spectra for the AA fuel solutions after the potentiostatic
experiments are presented in Fig. 6. The origin of each peak was
determined by comparison with standards. The concentration of each
compound was calculated by the HPLC peak area through the use of
predetermined calibration curves. From Table 1, the product
The fuel solutions at the end of each experiment were quantitatively
and qualitatively analyzed by 13C{1H} NMR spectroscopy and HPLC to
determine the AA oxidation products. The results are summarized in
5

M.V. Pagliaro et al.
Inorganica Chimica Acta 525 (2021) 120488
regardless of the cell voltage with the ratio Ac:Prp of approximately
45:55. By comparison the amount of acrylate produced by the Pd/C
catalyst is affected by the working cell voltage. At low cell voltage
acrylate formation is not favored while at 0.7 V selectively for acrylate
reaches 74%.
3.3. Study of the allyl alcohol electrooxidation reaction mechanism
Both of the co-products identified during the electrolysis of AA (PrP
and 3-Hp) cannot be formed from solely electrochemical reactions. To
help us understand better the reaction mechanism and the origin of
formation of propionate and 3-hydroxypropionate during the electrol-
ysis, model experiments were performed in a sealed 50 cm3 stainless
steel pressure autoclave with a PTFE® liner. The first experiment (TEST
A) was carried out with 100 mg of Pd/C-CeO2 powder in 25 cm3 of an
aqueous solution of 0.5 M acrylate + 0.5 M KOH, at 60 ◦C for 10 h. This
model of reaction was performed in order to verify if acrylate could
–
undergo the addition of a water molecule to the C C double bond
catalyzed by Pd/C-CeO2 in alkaline conditions with the resultant for-
mation of 3-hydroxypropionate (step 3a of Scheme 1). NMR analysis
after the experiment confirms the formation of 3-hydroxypropionate,
through the “a” route for the formation of 3-hydroxypropionate from
allyl alcohol. The acrylate formed electrochemically (1a + 1b) is then
transformed to 3-hydroxypropionate through a non-electrochemical
catalytic process.
Scheme 1. Allyl alcohol oxidation mechanism.
Propionate which is also formed, cannot derive from a direct
oxidation electrochemical process from allyl alcohol. Using data from
the literature regarding palladium hydroxide catalyzed isomerization of
primary allyl alcohols to aldehydes (scheme 2) [18–22,18–22] we hy-
pothesized a second double step route: the first step (1b of scheme 1)
involves the isomerization of allyl alcohol to propionaldehyde, and a
second step (2b) involves a 2 e- electrochemical process of oxidation to
propionate. To demonstrate that the formation of propionate occurs
through the “b” pathway, we performed two tests (TEST B and TEST C)
in the autoclave. Both tests were carried out with 100 mg of Pd/C-CeO2
powder in 25 cm3 of an aqueous solution of 0.5 M allyl alcohol (TEST B)
and 0.5 M allyl alcohol + 0.5 M KOH (TEST C), at 60 ◦C for 5 h. At the
end of experiments, both the solutions were analyzed through 1H NMR
Scheme 2. Mechanism for the isomerization of allyl alcohol18
.
1
distribution of Pd-CeO2/C catalyst remains relatively constant
to verify the formation of propionaldehyde. The H NMR spectrum of
1
Fig. 7. H NMR spectra of the TEST B aqueous solution of 0.5 M allyl alcohol after a 5 h reaction at 60 ◦C, which evidences the formation of propionaldehyde at
9.6 ppm.
6

M.V. Pagliaro et al.
Inorganica Chimica Acta 525 (2021) 120488
the process is here obtained from the water reduction at the cathode that
produces hydroxyl anions, which cross the anion exchange membrane to
the anode where they are involved in the oxidation of the alcohol
(Fig. 1). Table 2 compares literature data for alcohol electrochemical
reforming with the Pd-CeO2/C anode catalyst. The conversion effi-
ciencies are related to constant current tests rather than fixed cell po-
tential tests. The data shows how the nature of the substrate determines
the energy cost and conversion efficiency. Generally, the simpler alco-
hols (e.g. methanol, ethanol) consume less energy and are more kinet-
ically favored. AA demonstrates slower kinetics and higher energy
requirement compared to the other alcohols reported.
At a cell potential of 1 V the conversion of allyl alcohol is close to
100%. The selectivity for acrylate is around 43%. The competing process
which suppresses the selectivity has been identified as the Pd catalyzed
isomerization of AA to the corresponding aldehyde which is then
oxidized electrochemically to proprionate. Strategies to prevent the
isomerization of AA will require the development of alternative catalysts
that do not catalyze this parasitic reaction. In the case of Pd/C improved
selectivity for acrylate can be obtained by setting the cell potential at
0.7 V although the conversion efficiency drops to 60%. We are currently
investigating nanostructured Pd-M alloy materials as alternative cata-
lysts to this end. Regarding the stability of the MEA we have performed 3
consecutive galvanostatic experiments with both Pd-CeO2/C and Pd/C
catalysts. After each run the used fuel solution was replaced by a fresh
batch. The cell voltage was monitored during each experiment (Fig. 9).
There does not appear to be significant degradation in performance over
the three experiments. To determine the future economic and environ-
mental feasibility of acrylate production by electroreforming will
require a study of long term stability of the catalyst materials in scaled
up processes along with dedicated LCA studies.
Fig. 8. HPLC trace of a 0.5 M solution of 0.5 M allyl alcohol and 0.5 M KOH,
fed to the cell at OCV and recycled at 60 ◦C for 18 h.
TEST B solution shows the typical triplet proton signal at 9.6 ppm that
confirms the presence of the aldehyde (Fig. 7); Analysis of the TEST C
solution did not show the presence of aldehyde. This is due to the
instability of propionaldehyde in alkaline media (pH 13.7) to fast re-
action with allyl alcohol through aldol condensation (Scheme 1 step 2c).
HPLC analysis of the TEST C solution, showed the appearance in the
chromatogram of a large peak at 53 min that we assigned to various
aldol condensation products. The formation of aldol condensation
products was also verified in the electrolysis cell, fueling a 0.5 M allyl
alcohol + 0.5 M KOH solution to the cell at open circuit voltage (no
current passed to the cell) for 18 h at 60 ◦C. The HPLC trace of the
recovered solution shows the presence of the same aldol condensation
products peak found for TEST C (Fig. 8). On the contrary, at operating
conditions of 1 V (high current density), the propionaldehyde is rapidly
electro-oxidized to propionate, limiting the collateral aldol
condensations.
4. Conclusions
In this work, we demonstrate the first study of acrylate production by
the oxidation of allyl alcohol in an electrolysis cell. The process has a low
energy cost, operates at mild temperature and at ambient pressure, and
it also produces clean and pure hydrogen that can be used to produce
electrical energy in a fuel cell. A nanostructured Pd/C-CeO2 was used as
anode material, which presents enhanced activity and stability for
3.4. Discussion
This study has demonstrated that allyl alcohol can be readily
oxidized to acrylate in an electrolysis cell with the combined production
of H2. Hydrogen is produced in the membrane separated cathodic
compartment and can be fed directly to a fuel cell to be transformed into
alcohol electrooxidation under alkaline conditions, providing
a
electrical energy. Using the cell running at 1 V as an example, the energy
maximum conversion of AA of 96% and a selectivity to Ac of 43.0% at a
constant cell operating voltage of 1 V. An equivalent Pd/C catalyst
shows improved selectivity at 0.7 V (74%). To make the process a valid
alternative to current petrochemical-based technologies, the selectivity
for the formation of acrylate compared to the other main oxidation
products (propionate and 3-hydroxypropionate) must be optimized. A
detailed reaction mechanism investigation was performed in this study
that will help in finding strategies for improving the selectivity for
acrylate.
-1
cost of the hydrogen produced is 26.8 KWh kgH2
(Table 1). The
hydrogen produced has an energy content of 33 KWh kgH2 -1 (LHV).[23
If this hydrogen is fed to a state of the art PEM-FC stack with an energy
efficiency of approximately 50%, a 16.5 KWh kgH2 -1 electrical energy
output will be generated. This energy can be recycled back to the AA
electrolysis process. In this way more than half of the energy cost of the
electrolysis is recovered.
There are additional advantages of this electrolysis process,
including a low temperature and ambient pressure operation, as well as
the possibility to obtain H2 at high pressures (up to 10 bar). In addition,
no pure O2 is required for the oxidation reaction: the oxygen required by
Fig. 9. Three consecutive galvanostatic experiments (50 mA cmꢀ 2) carried out at 60 ◦C with three batches of fresh fuel solution (50 mL aqueous solution of 0.5 M
allyl alcohol + 0.5 M KOH) (a) Pd-CeO2/C and (b) Pd/C.
7

M.V. Pagliaro et al.
Inorganica Chimica Acta 525 (2021) 120488
[8] M.V. Pagliaro, M. Bellini, M. Bevilacqua, J. Filippi, M.G. Folliero, A. Marchionni, H.
A. Miller, W. Oberhauser, S. Caporali, M. Innocenti, F. Vizza, Carbon supported Rh
nanoparticles for the production of hydrogen and chemicals by the
electroreforming of biomass-derived alcohols, RSC Adv. 7 (23) (2017)
13971–13978.
CRediT authorship contribution statement
Maria Vincenza Pagliaro: Investigation, Visualization, Formal
analysis. Hamish Andrew Miller: Investigation, Formal analysis.
Marco Bellini: Investigation, Formal analysis. Benedetto Di Vico:
Investigation, Formal analysis. Werner Oberhauser: Investigation.
Giovanni Zangari: Supervision. Massimo Innocenti: Conceptualiza-
tion. Francesco Vizza: Conceptualization, Supervision, Project admin-
istration, Funding acquisition.
[9] J. Mahmoudian, M. Bellini, M.V. Pagliaro, W. Oberhauser, M. Innocenti, F. Vizza,
H.A. Miller, Electrochemical coproduction of acrylate and hydrogen from 1,3-
Propandiol, ACS Sustain. Chem. Eng. 5 (7) (2017) 6090–6098.
[10] M.V. Pagliaro, M. Bellini, J. Filippi, M.G. Folliero, A. Marchionni, H.A. Miller,
W. Oberhauser, F. Vizza, Hydrogen production from the electrooxidation of
methanol and potassium formate in alkaline media on carbon supported Rh and Pd
nanoparticles, Inorg. Chim. Acta 470 (2018) 263–269.
[11] M. Bellini, M.V. Pagliaro, A. Marchionni, J. Filippi, H.A. Miller, M. Bevilacqua,
A. Lavacchi, W. Oberhauser, J. Mahmoudian, M. Innocenti, P. Fornasiero, F. Vizza,
Hydrogen and chemicals from alcohols through electrochemical reforming by Pd-
CeO2/C electrocatalyst, Inorg. Chim. Acta 518 (2021), 120245.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
[12] L.Q. Wang, A. Lavacchi, M. Bevilacqua, M. Bellini, P. Fornasiero, J. Filippi,
M. Innocenti, A. Marchionni, H.A. Miller, F. Vizza, Energy efficiency of alkaline
direct ethanol fuel cells employing nanostructured palladium electrocatalysts,
ChemCatChem 7 (14) (2015) 2214–2221.
[13] V. Bambagioni, C. Bianchini, Y.X. Chen, J. Filippi, P. Fornasiero, M. Innocenti,
A. Lavacchi, A. Marchionni, W. Oberhauser, F. Vizza, Energy efficiency
enhancement of ethanol electrooxidation on Pd-CeO2/C in passive and active
polymer electrolyte-membrane fuel cells, ChemSusChem 5 (7) (2012) 1266–1273.
[14] Y.X. Chen, A. Lavacchi, H.A. Miller, M. Bevilacqua, J. Filippi, M. Innocenti,
A. Marchionni, W. Oberhauser, L. Wang, F. Vizza, Nanotechnology makes biomass
electrolysis more energy efficient than water electrolysis, Nat. Commun. 5 (2014).
[15] H.A. Miller, A. Lavacchi, F. Vizza, M. Marelli, F. Di Benedetto, F.D.I. Acapito,
Y. Paska, M. Page, D.R. Dekel, A Pd/C-CeO2 anode catalyst for high-performance
platinum-free anion exchange membrane fuel cells, Angew. Chem. Int. Ed. 55 (20)
(2016) 6004–6007.
Acknowledgements
We acknowledge funding of a PRIN 2017 Project funded by the
Italian Ministry MUR Italy (Grant No. 2017YH9MRK). We also
acknowledge the Italian Ministry MISE for the FISR 2019 Project
AMPERE (FISR2019_01294).
References
[16] H.A. Miller, F. Vizza, M. Marelli, A. Zadick, L. Dubau, M. Chatenet, S. Geiger,
S. Cherevko, H. Doan, R.K. Pavlicek, S. Mukerjee, D.R. Dekel, Highly active
nanostructured palladium-ceria electrocatalysts for the hydrogen oxidation
reaction in alkaline medium, NANO Energy 33 (2017) 293–305.
[17] L.Q. Wang, A. Lavacchi, M. Bellini, F. D’Acapito, F. Di Benedetto, M. Innocenti, H.
A. Miller, G. Montegrossi, C. Zafferoni, F. Vizza, Deactivation of palladium
electrocatalysts for alcohols oxidation in basic electrolytes, Electrochim. Acta 177
(2015) 100–106.
[1] A. Corma, S. Iborra, A. Velty, Chemical routes for the transformation of biomass
into chemicals, Chem. Rev. 107 (6) (2007) 2411–2502.
[2] R. Beerthuis, G. Rothenberg, N.R. Shiju, Catalytic routes towards acrylic acid,
adipic acid and epsilon-caprolactam starting from biorenewables, Green Chem. 17
(3) (2015) 1341–1361.
[3] D.L. Sun, Y. Yamada, S. Sato, W. Ueda, Glycerol as a potential renewable raw
material for acrylic acid production, Green Chem. 19 (14) (2017) 3186–3213.
[4] E.V. Makshina, J. Canadell, J. van Krieken, E. Peeters, M. Dusselier, B.F. Sels, Bio-
acrylates production: recent catalytic advances and perspectives of the use of lactic
acid and their derivates, ChemCatChem 11 (1) (2019) 180–201.
[5] X.K. Li, Y.G. Zhang, Highly Efficient Process for the Conversion of Glycerol to
Acrylic Acid via Gas Phase Catalytic Oxidation of an Allyl Alcohol Intermediate,
ACS Catal. 6 (1) (2016) 143–150.
[18] G. Sabitha, S. Nayak, M. Bhikshapathi, J.S. Yadav, Palladium hydroxide catalyzed
isomerization of primary allylic alcohols to aldehydes: application to the formal
synthesis of (-)-Brevisamide, Org. Lett. 13 (3) (2011) 382–385.
[19] C. Bianchini, A. Meli, W. Oberhauser, Isomerization of allylic alcohols to carbonyl
compounds by aqueous-biphase rhodium catalysis, New J. Chem. 25 (1) (2001)
11–12.
[20] J. Muzart, Palladium-catalysed oxidation of primary and secondary alcohols,
Tetrahedron 59 (31) (2003) 5789–5816.
[6] T. Murayama, B. Katryniok, S. Heyte, M. Araque, S. Ishikawa, F. Dumeignil,
S. Paul, W. Ueda, Role of crystalline structure in allyl alcohol selective oxidation
over Mo3VOx complex metal oxide catalysts, ChemCatChem 8 (14) (2016)
2415–2420.
[21] P. Lorenzo-Luis, A. Romerosa, M. Serrano-Ruiz, Catalytic Isomerization of allylic
alcohols in water, ACS Catal. 2 (6) (2012) 1079–1086.
[22] R.C. van der Drift, E. Bouwman, E. Drent, Homogeneously catalysed isomerisation
of allylic alcohols to carbonyl compounds, J Organomet Chem 650 (1–2) (2002)
1–24.
[7] S. Yang, M. Kim, S. Yang, D.S. Kim, W.J. Lee, H. Lee, Production of acrylic acid
from biomass-derived allyl alcohol by selective oxidation using Au/ceria catalysts,
Catal. Sci. Technol. 6 (10) (2016) 3616–3622.
8