An alternative to hydrogenation processes. Electrocatalytic hydrogenation of benzophenone

Article abstract of DOI:10.3762/bjoc.14.40

The electrocatalytic hydrogenation of benzophenone was performed at room temperature and atmospheric pressure using a polymer electrolyte membrane electrochemical reactor (PEMER). Palladium (Pd) nanoparticles were synthesised and supported on a carbonaceous matrix (Pd/C) with a 28 wt % of Pd with respect to carbon material. Pd/C was characterised by transmission electron microscopy (TEM), and thermogravimetric analysis (TGA). Cathodes were prepared using Pd electrocatalytic loadings (L_Pd) of 0.2 and 0.02 mg cm^-2. The anode consisted of hydrogen gas diffusion for the electrooxidation of hydrogen gas, and a 117 Nafion exchange membrane acted as a cationic polymer electrolyte membrane. Benzophenone solution was electrochemically hydrogenated in EtOH/water (90/10 v/v) plus 0.1 M H₂SO₄. Current densities of 10, 15 and 20 mA cm^-2 were analysed for the preparative electrochemical hydrogenation of benzophenone and such results led to the highest fractional conversion (XR) of around 30% and a selectivity over 90% for the synthesis of diphenylmethanol upon the lowest current density. With regards to an increase by ten times the Pd electrocatalytic loading the electrocatalytic hydrogenation led neither to an increase in fractional conversion nor to a change in selectivity.

Full text of DOI:10.3762/bjoc.14.40

An alternative to hydrogenation processes. Electrocatalytic
hydrogenation of benzophenone
Cristina Mozo Mulero, Alfonso Sáez*, Jesús Iniesta and Vicente Montiel
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2018, 14, 537–546.
Instituto de Electroquímica, Universidad de Alicante, Apartado 99,
03080 Alicante, Spain
doi:10.3762/bjoc.14.40
Received: 22 December 2017
Accepted: 15 February 2018
Published: 01 March 2018
Email:
Alfonso Sáez* - alfonso.saez@ua.es
* Corresponding author
This article is part of the Thematic Series "Electrosynthesis II".
Guest Editor: S. R. Waldvogel
Keywords:
benzophenone; diphenylmethanol; electrocatalytic hydrogenation;
palladium nanoparticles; polymer electrolyte membrane
© 2018 Mozo Mulero et al.; licensee Beilstein-Institut.
License and terms: see end of document.
Abstract
The electrocatalytic hydrogenation of benzophenone was performed at room temperature and atmospheric pressure using a polymer
electrolyte membrane electrochemical reactor (PEMER). Palladium (Pd) nanoparticles were synthesised and supported on a carbo-
naceous matrix (Pd/C) with a 28 wt % of Pd with respect to carbon material. Pd/C was characterised by transmission electron
microscopy (TEM), and thermogravimetric analysis (TGA). Cathodes were prepared using Pd electrocatalytic loadings (LPd) of 0.2
and 0.02 mg cm−2. The anode consisted of hydrogen gas diffusion for the electrooxidation of hydrogen gas, and a 117 Nafion
exchange membrane acted as a cationic polymer electrolyte membrane. Benzophenone solution was electrochemically hydro-
genated in EtOH/water (90/10 v/v) plus 0.1 M H2SO4. Current densities of 10, 15 and 20 mA cm−2 were analysed for the prepara-
tive electrochemical hydrogenation of benzophenone and such results led to the highest fractional conversion (XR) of around 30%
and a selectivity over 90% for the synthesis of diphenylmethanol upon the lowest current density. With regards to an increase by ten
times the Pd electrocatalytic loading the electrocatalytic hydrogenation led neither to an increase in fractional conversion nor to a
change in selectivity.
Introduction
Hydrogenation is a common procedure applied in organic hance chemical kinetics, a number of side reactions reduces the
chemistry industry based on the use of an external hydrogen selectivity of the chemical reaction providing also complex and
source, generally carried out under moderate experimental cost-ineffective work-up procedures [1-3]. Alternatively, elec-
conditions of high temperature (until 673 K) and high-pressure trocatalytic hydrogenation has emerged as a technique driven by
(even 350 atm). Even though the hydrogenation is performed its operational mild conditions, i.e., in situ “active hydrogen”
using either homogeneous or heterogeneous catalysts to en- generation, room temperature and atmospheric pressure, and
537

Beilstein J. Org. Chem. 2018, 14, 537–546.
higher selectivity [4-10]. This electrochemical “active hydro- added value product in chemical industry. Even though several
gen” generation is performed under a cathodic polarization and research groups have reported in the literature fundamental
such mechanism of an unsaturated organic compound (e.g., aspects about the electrocatalytic hydrogenation of benzo-
Y=X) still remains undertrained. Nonetheless, the steps likely phenone as well as its electrolytic performance at laboratory
involved in this electrochemical process by the literature [11- scale using Pd, Ni/Pd or Ni/Pt based either on massive or
15] are as follows: (i) formation of adsorbed hydrogen MHads nanoparticulate cathodes [5,24-26]. The electrocatalytic hydro-
(defining “M” as a metal adsorption site and Hads as atomic genation of benzophenone has not been accomplished using a
adsorbed hydrogen), (ii) adsorption of the organic molecule on PEMER yet.
a support site (two distinct adsorption sites are considered) and,
(iii) electrocatalytic hydrogenation of the organic molecule This work aims at exploring the electrocatalytic hydrogenation
through the adsorbed atomic hydrogen. It is important to note of benzophenone over palladium nanoparticles supported on
that, formation of “active hydrogen” is the main step in this carbon black based electrode using a PEMER. A nanoparticu-
process and hydrogen-active powder electrocatalysts such as late platinum gas diffusion electrode is used for the hydrogen
Pd/C, Pt/C or Raney-nickel have been demonstrated as the oxidation reaction and a cationic exchange membrane is used as
optimal choice [16,17]. Moreover, the organic molecule adsorp- solid polymeric electrolyte. Fractional conversion, product yield
tion rate must be faster than that one associated with the recom- and selectivity are presented upon the Pd electrocatalytic
bination of MHads to form diatomic hydrogen gas; otherwise loading and current density. Moreover, the effect of the absence
electrocatalytic hydrogenation does not occur or displays a low and presence of the supporting electrolyte upon the conversion
Faraday efficiency because of the fact that diatomic hydrogen and product yield of the electrochemical hydrogenation of
formation is not the main reaction, but a competitive one. benzophenone is explored.
Clearly, from the above mechanistic steps, the lower the
“atomic active hydrogen” generation rate, the more efficient the Results and Discussion
electrocatalytic hydrogenation process; thus, the higher the used Pd nanoparticles supported on Vulcan XC72R as electrocata-
current density, the lower the obtained efficiency. Moreover, lyst in cathodic reaction (electrocatalytic hydrogenation) were
selectivity of electrocatalytic hydrogenation of ketones to alco- synthesised by reducing K2PdCl4 using NaBH4 in a water-in-oil
hols is performed in acidic medium as an optimal reaction me- (w/o) microemulsion (water/Brij@30/n-heptane). This method-
dium [5].
ology has been previously used in our laboratory [27,28]. We
first explored the morphology, size and dispersion of Pd nano-
Currently, developments in nanostructured materials and poly- particles supported on Vulcan XC72R carbonaceous material
meric solid exchange membranes have led the field of polymer (Pd/C electrocatalysts) using TEM micrographs. As shown in
electrolyte membrane fuel cells (PEMFC) to become a mature Figure 1a, a good dispersion of spherical Pd nanoparticles of
technology [18-20]. In this regard, a polymer electrolyte mem- about 4–5 nm size is obtained. As expected, EDX analysis of Pd
brane electrochemical reactor (PEMER) has already been nanoparticles displays palladium as the sole metal, as shown in
defined and manufactured to implementing all PEMFC technol- Figure 1b. With respect to the Pd loading in the Pd/C electrocat-
ogy advantages for both inorganic and organic electrosynthetic alyst, TGA results shown in Figure 1c lead to ca. 30 wt %
processes [21]. It is worth noting that the use of a polymer elec- content. A high drop of weight is observed between 600 K and
trolyte membrane electrochemical reactor (PEMER) allows ob- 700 K since carbonaceous support is volatilized until reaching a
taining several advantages compared to the use of conventional residual PdO content (Pd in a N2/O2 atmosphere becomes a
electrochemical reactors, as follows: i) nanostructurated electro- PdO species as stable residual weight when temperature is in-
catalysts can be utilized for both cathodic and anodic reactions, creased from 700 K to 1073 K). From the stable residual weight
ii) a solid polymer electrolyte is used instead of a conventional in Figure 1c at higher temperature it can be estimated that there
electrolyte and iii) a decrease of the anode–cathode gap reduces is a content of 32 wt % of PdO in the whole Pd/C electrocata-
ohmic drop at the whole process. In the case of organic elec- lyst, thereby confirming that the electrocatalyst can be consid-
trosynthesis, electrocatalytic hydrogenation of aromatic ke- ered as Pd/C with 28 wt %, by correcting the PdO value by
tones, specifically acetophenone, has been recently carried out subtracting the oxygen atomic percentage content. Thus, the
[22,23] achieving a high selectivity using the above-mentioned above experimental wt % value of Pd is in agreement with the
technology. In this work, we have chosen benzophenone, as a theoretical one of 30 wt % as described in the experimental
more complex aromatic ketone dictated by the presence of two section.
phenyl rings, in order to analyse fractional conversion and
selectivity of formed products. Here, the main obtained product Once the Pd/C electrocatalyst characterisation was performed,
is diphenylmethanol (ketone conversion to alcohol) as a high Pd/C based cathodes either with 0.2 or 0.02 mg Pd per cm2
538

Beilstein J. Org. Chem. 2018, 14, 537–546.
Figure 1: Characterisation of Pd/C electrocatalyst. a) TEM micrograph. b) Energy dispersive X-ray analysis (EDX). c) Thermogravimety analysis in
N2/O2 (4/1) atmosphere at a heating slope of 10 K min−1 from 298 K to 1023 K.
were prepared and characterised by SEM and cyclic voltammet-
ry. A Pd/C to Nafion ratio was established at 60:40 for each
electrocatalytic layer irrespectively of the Pd loading in the
electrocatalytic layer. Figure 2 depicts the SEM micrographics
using back scattering electrons for both Pd0.02/C/T and
Pd0.20/C/T electrodes (nomenclature described in the experi-
mental section) at different magnifications. The SEM image for
the highest electrocatalytic loading discloses a brighter appear-
ance compared to the lowest one since backscattering electrons
from high atomic weight (e.g., palladium) are brighter than
lower atomic weight (e.g., carbon).
Apart from the SEM characterisation, the determination of the
electrochemically active surface area (ESA) of the cathodic
Figure 2: SEM images of (a) Pd0.02/C/T and (b) Pd0.20/C/T electrodes,
with different magnifications.
electrocatalytic layer is also crucial for the characterisation of
both electrodes. In this regard, Figure 3 depicts the electro-
chemical behaviour of a Pd0.20/C/T electrode in 0.5 M H2SO4
using cyclic voltammetry (CV) at a scan rate of 50 mV s−1 anodic peak is associated to the oxidation of the Pd surface to
(potential interval of 0.1 and 1.2 V vs Ag/AgCl). The CV PdO and then followed by a well-defined cathodic peak on the
diagram shown in Figure 3 represents a typical pattern for the negative scan ascribed to the reduction of PdO to Pd [29]. From
electrochemical behaviour of Pd bar electrode where a broad the coulombic charge integration of the cathodic peak on the
539

Beilstein J. Org. Chem. 2018, 14, 537–546.
CV in Figure 3 (shadowed region) can be accurately calculated particles, mostly due to the appearance of bottlenecks or
the ESA value and roughness factor, as defined in the experi- blockage of network across the electrocatalytic layer, or likely
mental section. The ESA value is calculated according to Equa- some of these Pd nanoparticles could be covered up each other
tion 1,
so those ones would be hidden.
Once Pd nanoparticle-based cathodes were manufactured and
characterised, electrolytic hydrogenation of benzophenone in a
cathodic process and hydrogen oxidation reaction as the anodic
(1)
where QPd (mC) is calculated from the coulombic charge inte- process was performed using a PEMER. Figure 4 describes
gration from the cathodic peak after subtracting the double each one of reactions involved in both cathodic and anodic pro-
layer, LPd denotes the electrocatalytic loading expressed in mg cesses and, also, the particulate elements of the electrochemical
of Pd per square centimetre of electrocatalytic layer, reactor.
Ageom (cm2) is the electrode geometric area and 0.424 refers to
charge density expressed in mC per cm2 of Pd [29]. On the Thus, preparative electrocatalytic hydrogenation of benzo-
other hand, the roughness factor rf is also calculated according phenone was performed using a PEMER with Pd0.02/C/T and
to Equation 2.
Pd0.20/C/T electrodes. We first investigated the influence of
current density, i.e., 10, 15 and 20 mA cm−2, on the electro-
chemical hydrogenation using a Pd0.02/C/T electrode with a
coulombic charge passed of 2 F (theoretical coulombic charge
established by Faraday’ s Law by considering reaction of one
mol of reagent). Figure 5 depicts the fractional conversion of
benzophenone (expressed in percentage) versus the coulombic
charge passed for all three current densities examined.
(2)
It should be noted that a higher fractional conversion of benzo-
phenone of 25.8% is obtained for the lowest current density
compared with values of 10.1% and 3.9% for current densities
of 15 and 20 mA cm−2, respectively (see Table 1). Our findings
are highly expected accordingly to the steps involved in the
electrochemical hydrogenation mechanisms (vide supra). In this
regard, after formation of PdHad, the recombination rate of this
“atomic active hydrogen” into diatomic hydrogen gas should be
slower than the adsorption rate of the organic compound over
the cathode in order to prompt electrocatalytic hydrogenation;
consequently, a low current density should favour the hydroge-
nation process instead of diatomic hydrogen formation as
competitive and side reaction by considering a slow PdHad for-
mation rate of this.
Figure 3: Cyclic voltammetric behaviour of Pd0.20/C/T electrode in
0.5 M H2SO4. Scan rate: 50 mV s−1. Starting potential: 0.2 V vs
Ag/AgCl. Positive scan.
Calculations of ESA and rf values were addressed for both
Pd0.02/C/T and Pd0.20/C/T electrodes. It is worth noting that
rf values denote real electrocatalytic surface area as a multiple
factor of the theoretical geometric area. In this regard, rf values
were 85 and 14 times the geometric area, whereas ESA values
were 707 and 424 for Pd0.02/C/T and Pd0.20/C/T electrodes, re-
spectively. The later ESA values indicate an increase of 67% of
active surface area per weight unit, so the higher the electrocata-
lytic loading shown in a cathode electrocatalytic layer, the
lower the electrochemically accessible surface per milligram.
The above behaviour is likely attributed to a much thicker elec-
trocatalytic layer for the Pd0.20/C/T electrode compared to
Pd0.02/C/T electrode. Nonetheless, a thicker electrocatalytic
layer is also detrimental for the accessibility of the electroac-
tive species to the electrocatalytic metal surface, i.e., Pd nano-
HPLC analysis was performed for the identification and quan-
tification of the benzophenone depletion and final product for-
mation from the electrocatalytic hydrogenation at Pd0.02/C/T
electrode. An inspection of the chromatograms confirms the
formation of two final products corresponding to diphenyl-
methanol (Ph2CHOH) and diphenylmethane (Ph2CH2). Table 1
compiles fractional conversion of Ph2CO (XR), product yield of
Ph2CHOH and Ph2CH2 (η), selectivity of Ph2CHOH (ξ) and
cell voltage (∆Ucell) for each current density over coulombic
charge passed. It is worth noting that product yield for the
Ph2CHOH is higher at low current density with an almost linear
increase as a function of coulombic charge passed. Moreover, it
540

Beilstein J. Org. Chem. 2018, 14, 537–546.
Figure 4: Scheme of all components of the electrochemical reactor including reactions involved in both anode and cathode.
Table 1 also depicts the average cell voltage from the electro-
catalytic hydrogenation of benzophenone, with values of 0.2,
0.3 and 0.5 V for 10, 15 and 20 mA cm−2, respectively. More
precisely, Figure 6 plots cell voltage versus time at
10 mA cm−2. In this case, a high cell voltage increase is ob-
served from the open circuit voltage of around 0 V until the first
stage of the electrocatalytic hydrogenation, followed by a cell
voltage drop attributed to a decrease in IR drop of the mem-
brane provoked by humidity lack of the membrane at either the
anodic or cathodic sides when starting the electrocatalytic
hydrogenation reaction. A similar variation of cell voltage
versus time was also observed when using the rest of current
densities.
Figure 5: Fractional conversion of benzophenone as a function of
coulombic passed charge using the PEMER; 0.5 M benzophenone as
starting solution in ethanol/H2O (90:10 v/v) + 0.1 M H2SO4. Pd0.02/C/T
electrode.
Next we turn out to explore the influence of the Pd electrocata-
lytic loading on benzophenone conversion. Once again, electro-
catalytic hydrogenation experiments were performed at con-
stant current densities of 10, 15 and 20 mA cm−2 using a manu-
is noticed that the selectivity value generally over 90% is irre- factured cathode with a Pd electrocatalytic loading of
spective of the current density. Electrocatalytic hydrogenation 0.2 mg cm−2 (Pd0.20/C/T electrode). Results depicted in
of benzophenone was also performed using Pd0.00/C/T Figure 7 show again a higher benzophenone fractional conver-
electrodes, i.e., in the absence of the Pd electocatalyst, at sion at lower current densities, as happened for the electrocata-
10 mA cm−2 demonstrated neither the electrochemical reduc- lytic hydrogenation using an electrode (Pd0.02/C/T) with a
tion of benzophenone nor the formation of alcohol or alkane de- 10 times lower Pd loading. Interestingly, benzophenone conver-
rivatives.
sion performed at 15 and 20 mA cm−2 turned out to be slightly
541

Beilstein J. Org. Chem. 2018, 14, 537–546.
Table 1: Fractional conversion (XR), product yield for diphenylmethanol (ηPh2CHOH) and diphenylmethane (ηPh2CH2), selectivity for diphenylmethanol
(ξPh2CH2) versus coulombic charge passed at constant current densities of 10, 15 and 20 mA cm−2. Pd0.02/C/T electrode. Benzophenone (0.5 M) was
used as starting solution in ethanol/water (90:10 v/v) + 0.1M H2SO4.
j
Q passed
[F]
∆Ucell
[mA cm−2]
0.5
1.0
1.5
2.0
[V]
XR
6.5
4.1
0.9
83
11.5
9.2
1.1
89
18.9
16.9
1.6
25.8
22.4
1.6
10
15
20
η Ph2CHOH
η Ph2CH2
ξ Ph2CHOH
0.2
0.3
0.5
91
93
XR
3.1
3.7
0
5.7
5.4
0
7.9
7.4
0.3
96
10.1
9.9
0.4
96
η Ph2CHOH
η Ph2CH2
ξ Ph2CHOH
100
100
XR
1.8
0.9
0
1.3
1.7
0
2.9
2.5
0.3
88
3.9
3.0
0.4
88
η Ph2CHOH
η Ph2CH2
ξ Ph2CHOH
100
100
Figure 7: Fractional conversion of benzophenone as a function of
coulombic charge passed; 0.5 M benzophenone as starting solution in
ethanol/H2O (90:10 v/v) plus 0.1 M H2SO4. Pd0.20/C/T electrode.
Figure 6: Plot of cell voltage versus time obtained from a preparative
electrosynthesis performed at 10 mA cm−2, Pd0.02/C/T electrode;
0.5 M benzophenone as starting solution in ethanol/water (90:10 v/v)
plus 0.1 M H2SO4.
lower for the Pd0.20/C/T electrode (vide infra). In this regard,
Figure 8 shows the comparative results in terms of fractional
conversion and product yield for both electrocatalytic hydroge-
nation reaction performed at Pd0.02/C/T and Pd0.20/C/T elec-
trodes and no significantly electrocatalytic hydrogenation reac-
tion at Pd0.00/C/T electrode appears. In addition, the selectivity
of diphenylmethanol is still over 90% irrespectively of the cur-
rent density examined; the alcohol and alkane derivatives are
also obtained, though prevailing the diphenylmethanol with
high product yield (see Figure 8), being similar behaviour as
that shown for the electrochemical hydrogenation using the
Pd0.02/C/T electrode.
Figure 8: Comparison of fractional conversion (XR) and product yield
(η) between Pd0.02/C/T, Pd0.20/C/T and Pd0.00/C/T electrodes.
542

Beilstein J. Org. Chem. 2018, 14, 537–546.
The comparison of fractional conversion and product yield benzophenone was performed at 10 mA cm−2 using both Pd0.02/
displayed in Figure 8 shows that the electrocatalytic hydrogena- C/T and Pd0.20/C/T electrodes in the absence of sulfuric acid.
tion of benzophenone is nearly irrespective of the Pd electro- Figure 9 depicts the fractional conversions and product yields as
catalytic loading. Indeed, one would respect a much higher frac- a function of the coulombic charge passed for both electrodes in
tional conversion of benzophenone when using the Pd0.20/C/T the absence and presence of 0.1 M H2SO4 for comparative
electrode. In other words, the higher Pd electrocatalytic loading, purposes. Even though no changes were observed in terms of
the higher adsorption of the organic molecule, and likely the selectivity in the absence of the electrolyte, a remarkable frac-
higher reactivity of the adsorbed organic molecule with tional conversion decrease of around 10% can be calculated. As
adsorbed hydrogen. At this point, we have defined a parameter, summary, the sole contribution of hydronium ions originating
XN, as fractional conversion normalised towards the electro- from the hydrogen oxidation reaction is not enough yet for an
chemically Pd accessible surface area, i.e., Pd accessible sites, optimal performance of the electrocatalytic hydrogenation of
according to Equation 3 shown below in order to shed light on benzophenone.
the real effect of the Pd loading on the electrochemical hydroge-
nation of benzophenone.
(3)
This parameter denotes the usefulness of electrocatalytic
loading and is expressed as % (cm2 Pd)−1 where, XR and rf
have the usual meanings and Ageom corresponds to geometric
area.
We have calculated the XN value accordingly to the optimum
current density of 10 mA cm−2 for both electrocatalytic
loadings. These values are 0.073 and 0.014 for 0.02 and
0.2 (mg Pd) cm−2 respectively, confirming higher electrocata-
lytic usefulness for lower catalytic loading. By comparing Pd
electrocatalytic loadings between 0.02 and 0.2, only 16% of the
total accessible surface area in Pd electrocatalytic loading of
0.2 mg cm−2 resulted to be active to electrocatalytic hydrogena-
tion of benzophenone. Accessibility problems of benzophenone
molecules reaching electroactive sites within a thicker electro-
catalytic layer are the most plausible explanation for getting a
lower XN value. Consequently, an increment of the electrocata-
lytic layer does not guarantee an enhancement of fractional
conversion. For this reason, lower electrocatalytic loadings are
the best approach to take into consideration.
Figure 9: Fractional conversions of benzophenone and product yield
of diphenylmethanol at both electrodes. (a) Pd0.02/C/T electrode;
(b) Pd0.20/C/T electrode; 0.5 M benzophenone as starting solution in
ethanol/water (90:10 v/v) in the presence of 0.1 M H2SO4 (black
symbols) and absence of H2SO4 (red symbols).
Finally, the influence of acidic medium in catholyte solutions
was explored to simplify the work-up of the crude mixture from
the electrocatalytic hydrogenation of benzophenone. It is worth
noticing that acidic medium is a parameter that must be taken
into consideration in terms of selectivity for electrocatalytic Conclusion
hydrogenations of several organic compounds. In this regard, The electrocatalytic hydrogenation of benzophenone has been
we performed the electrochemical hydrogenation of benzo- performed at pre-pilot scale using a polymer electrolyte mem-
phenone in the absence of 0.1 M H2SO4 in order to investigate brane electrochemical reactor (PEMER). The benzophenone
the contribution of hydronium ion formation from hydrogen ox- conversion and diphenylmethanol and diphenylmethane yields
idation reaction passing through the polymer electrolyte mem- are driven mostly by the current density used. Even though the
brane as the sole hydronium ions source reaching the catholyte lowest current density favours the benzophenone conversion
compartment. In doing so, the electrocatalytic hydrogenation of and the alcohol yield, higher current density leads to a high
543

Beilstein J. Org. Chem. 2018, 14, 537–546.
selectivity. In terms of Pd loading within the electrocatalytic Morphology, size and dispersion of Pd nanoparticles in Pd/C
layer, no significant difference is achieved with respect to frac- were analysed by a transmission electron microscopy (TEM)
tional conversion and product yields, however, Pd accessible using a JEOL JEM-2012 instrument with an accelerating
surface area per weight unit turns out to be enhanced for the voltage of 300 kV. The TEM was connected with an energy
lowest electrocatalytic loading, indicating a benzophenone dispersion X-ray (EDX) for the analysis of the Pd nanoparticles.
accessibility to the Pd nanoparticulate electrocatalysts.
Thermogravimetric analysis (TGA) was performed using a
Mettler Toledo model TGA/SDTA851 and /SF/1100 using a
Hydrogen gas evolution reaction in the cathode compartment heating slope of 10 K min−1 from 298 K to 1023 K under a
can be reused to feed a gas diffusion electrode for the hydrogen N2:O2 (4:1) atmosphere. For the characterisation of the Pdx/C/T
oxidation reaction. Moreover, in the absence of an acid medi- scanning electron microscopy (SEM) micrographs were ob-
um as supporting electrolyte commonly used at the cathode tained using a Hitachi S-3000N microscope with backscattered
compartment, the electrocatalytic hydrogenation is feasible by electron signal. Electrochemical surface characterisation of the
the sole supply of hydronium ions generated at the anode com- distinct Pdx/C/T electrodes was explored by cyclic voltammet-
partment coming from the hydrogen oxidation reaction.
ry (CV). CV measurements were performed using a PGSTAT30
Autolab system. A 0.5 cm × 1.0 cm Pdx/C/T electrode acted as
Finally, it is worth noting that the hydrogen oxidation reaction a working electrode (WE), the counter electrode was a plati-
was used as anodic process decreasing cell voltage of the num wire and an Ag/AgCl (3.0 M KCl) acted as reference elec-
process; besides, the use of hydrogen evolved at the cathode as trode through a Luggin capillary. CV measurements were per-
competitive reaction must be considered as hydrogen source to formed at room temperature and under argon atmosphere. CV
partially feed the hydrogen anode.
curves were recorded between 0.1 and 1.2 V vs Ag/AgCl
(3.0 M KCl) with a scan rate of 50 mV s−1 using a 0.5 M
H2SO4 aqueous solution. The electrochemical accessible sur-
Experimental
Benzophenone, diphenylmethanol, diphenylmethane, ethanol face area (ESA) was defined as the Pd electrochemically active
96% (Sigma-Aldrich, Spain) were used without further purifica- surface divided into the total Pd weight according to the total Pd
tion. K2PdCl4 salt was purchased from Sigma-Aldrich (purity loading, expressed in (cm2 Pd) (mg Pd)−1, while the roughness
higher than 99%). Vulcan XC72R carbon material was pur- factor was defined as the Pd electrochemically active surface
chased from Cabot Corporation. The other chemicals were pur- divided by the geometric total area, expressed in (cm2 Pd) cm−2
chased in the highest purity available and used as received. The (vide supra).
solutions were prepared with doubly distilled water.
Electrosynthesis reactions were performed using a 25 cm2 PEM
The synthesis of Pd nanoparticles was performed following a single cell fuel cell (EFC-25-01 model from ElectroChem Inc.)
procedure previously used in our laboratory and described in acting as polymer electrolyte membrane electrochemical reactor
the literature [27,28]. Briefly, a K2PdCl4 salt solution was (PEMER), as depicted in Figure 10. Such an electrochemical
reduced using NaBH4 as reducing agent in a water-in-oil (w/o) reactor consisted of (i) a Pdx/C/T cathode, (ii) an ETEK gas
microemulsion in the presence of polyethylene glycol hexa- diffusion-type EFGC (Pt/C/T) with a 40 wt % of Pt with respect
decyl ether, Brij@30 as capping agent and n-heptane as organic to the amount of carbon, and a Pt electrocatalytic loading of
solvent (water/Brij@30/n-heptane). After precipitation and 2.0 mg cm−2 as anode to carry out the hydrogen oxidation reac-
copiously washing with acetone and water, Pd nanoparticles tion, and finally, (iii) a Nafion 117 cation exchange membrane
were supported on Vulcan XC72R to get a nominal 30 wt % Pd in contact between both electrodes acting as solid polymer elec-
loading (Pd/C electrocatalyst).
trolyte.
The cathode layer was prepared by air-brushing an electrocata- The catholyte solutions for the electrochemical hydrogenation
lytic ink onto a Toray Paper TGPH-90 (carbonaceous compos- of benzophenone consisted of either (i) 0.5 M benzophenone in
ite paper supplied by Toray Industries Inc., thickness 280 µm). ethanol 96% /water (90:10 v/v) and 0.1 M H2SO4, or (ii) 0.5 M
The electrocatalytic ink consisted of Pd nanoparticles sup- benzophenone in ethanol 96% /water (90:10 v/v) in the absence
ported in Vulcan XC72R 30 wt % dispersion in isopropanol of the sulfuric acid electrolyte in order to explore the influence
containing a Nafion dispersion of 5 wt %. All electrodes have a of the medium acidity upon the electrochemical hydrogenation.
Pd/C to Nafion ratio of 60:40 in the electrocatalytic layer with
Pd loadings of 0.20 and 0.02 mg cm−2. Cathodes were named as A peristaltic pump Ismatel Reglo DIG MS/CA 2–8 provided a
Pdx/C/T where x stands for the Pd loading and T is the Toray catholyte flow of 12 mL min−1 through the cathodic compart-
paper support.
ment and the total catholyte volume was 40 mL. At the same
544

Beilstein J. Org. Chem. 2018, 14, 537–546.
Figure 10: (a) General scheme of a PEMER; (b) itemisation of the main parts of PEMER: 1) endplates, 2) gas diffusion electrode, 3) Pt electrocata-
lytic layer, 4) polymer electrolyte membrane (solid polymer electrolyte), 5) carbonaceous support (Toray paper TGPH-90), 6) Pd/C electrocatalytic
layer.
time, hydrogen (Praxair, purity: 99.999%) was fed to the anodic Hypersil ODS column 4 × 250 mm, 5 µm particle size (Agilent
compartment at a flow rate of 50 mL min−1 as reactant in Technologies). An acetonitrile/water (1:1 v/v, acetonitrile
hydrogen oxidation reaction, using a MTS-module A-150 from isocratic HPLC grade and water from Elix 3 Millipore system)
ElectroChem Inc. Electrocatalytic hydrogenation reactions were was used as mobile phase. The flow rate was 1 mL min−1; the
carried out at constant current densities of 10, 15 and injection volume was 100 µL and the working temperature was
20 mA cm−2 considering a projected geometric area of 25 cm2 303 K. Such experimental conditions provided linear calibra-
for both Pd0.02/C/T and Pd0.20/C/T electrodes using a potentio- tion curves between 20 and 100 ppm with all benzophenone,
stat OrigaFlex OGF05A. Total coulombic charge passed was set diphenylmethanol and diphenylmethane standard organic solu-
at 2 Faradays for each experiment. The cell voltage was re- tions. Preparative electrosyntheses were given as fractional
corded during each experiment. All preparative electrochemical conversion (XR) of benzophenone (Ph2CO) and product yield
hydrogenations were carried out at room temperature. In this (η) of diphenylmethanol (Ph2CHOH) and diphenylmethane
case, counter electrode plug was connected to reference elec- (Ph2CH2). Selectivity values (ξ) of diphenylmethanol and
trode one and chronopotentiometries were performed regis- diphenylmethane were also calculated. For comparative
tering cell voltage versus time.
purposes, a free-Pd electrocatalytic layer (Pd0.00/C/T) cathode
was also considered for the electrocatalytic hydrogenation of
The fractional conversion of benzopheneone and product yield Ph2CO.
and selectivity of final products from the electrocatalytic hydro-
genation of benzophenone was examined as a function of cur- Acknowledgements
rent density and Pd loading in the electrocatalytic layer. On Authors thank the Ministerio de Economia y Competitividad
doing so, the catholyte solution was analysed by using a high- MINECO, Spain for its financial support by the research proj-
resolution liquid chromatograph (HPLC Agilent 1200) with a ects CTQ2013-48280-C3-3-R and CTQ2016-76231-C2-2-R.
545

Beilstein J. Org. Chem. 2018, 14, 537–546.
22.Sáez, A.; García-García, V.; Solla-Gullón, J.; Aldaz, A.; Montiel, V.
Electrochem. Commun. 2013, 34, 316–319.
ORCID® iDs
Cristina Mozo Mulero - https://orcid.org/0000-0002-6490-9710
Alfonso Sáez - https://orcid.org/0000-0002-2949-6711
Jesús Iniesta - https://orcid.org/0000-0003-4350-2169
Vicente Montiel - https://orcid.org/0000-0003-2353-2993
doi:10.1016/j.elecom.2013.07.018
23.Sáez, A.; García-García, V.; Solla-Gullón, J.; Aldaz, A.; Montiel, V.
Electrochim. Acta 2013, 91, 69–74.
doi:10.1016/j.electacta.2012.12.097
24.Vilar, M.; Oliveira, J. L.; Navarro, M. Appl. Catal., A 2010, 372, 1–7.
doi:10.1016/j.apcata.2009.09.041
References
25.Villalba, M.; Bossi, M. L.; Calvo, E. J. Phys. Chem. Chem. Phys. 2015,
17, 10086–10092. doi:10.1039/C5CP00225G
1. Chen, B.; Dingerdissen, U.; Krauter, J. G. E.;
Lansink Rotgerink, H. G. J.; Möbus, K.; Ostgard, D. J.; Panster, P.;
Riermeier, T. H.; Seebald, S.; Tacke, T.; Trauthwein, H. Appl. Catal., A
2005, 280, 17–46. doi:10.1016/j.apcata.2004.08.025
2. Rilander, P. N. Hydrodenation Methods; Academic Press, 1985; p 193.
3. Magnoux, P.; Lavaud, N.; Guisnet, M. Top. Catal. 2000, 13, 291–299.
doi:10.1023/a:1009009727910
26.Villalba, M.; del Pozo, M.; Calvo, E. J. Electrochim. Acta 2015, 164,
125–131. doi:10.1016/j.electacta.2015.02.113
27.Solla-Gullón, J.; Montiel, V.; Aldaz, A.; Clavilier, J. J. Electrochem. Soc.
2003, 150, E104–E109. doi:10.1149/1.1534600
28.Solla-Gullón, J.; Rodes, A.; Montiel, V.; Aldaz, A.; Clavilier, J.
J. Electroanal. Chem. 2003, 554–555, 273–284.
doi:10.1016/S0022-0728(03)00214-6
4. Yoshida, J.-i.; Kataoka, K.; Horcajada, R.; Nagaki, A. Chem. Rev.
2008, 108, 2265–2299. doi:10.1021/cr0680843
29.Fang, L.-l.; Tao, Q.; Li, M.-f.; Liao, L.-w.; Chen, D.; Chen, Y.-x.
Chin. J. Chem. Phys. 2010, 23, 543–548.
5. Cirtiu, C. M.; Brisach-Wittmeyer, A.; Ménard, H. Catal. Commun. 2007,
8, 751–754. doi:10.1016/j.catcom.2006.09.014
doi:10.1088/1674-0068/23/05/543-548
6. Coche, L.; Moutet, J. C. J. Am. Chem. Soc. 1987, 109, 6887–6889.
doi:10.1021/ja00256a072
7. Moutet, J.-C. Org. Prep. Proced. Int. 1992, 24, 309–325.
doi:10.1080/00304949209355892
License and Terms
8. Pintauro, P. N.; Bontha, J. R. J. Appl. Electrochem. 1991, 21, 799–804.
doi:10.1007/bf01402817
This is an Open Access article under the terms of the
Creative Commons Attribution License
9. Polcaro, A. M.; Dernini, M. S.; Palmas, S. Electrochim. Acta 1992, 37,
365–367. doi:10.1016/0013-4686(92)85025-G
(http://creativecommons.org/licenses/by/4.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
10.Polcaro, A. M.; Palmas, S.; Dernini, S. Electrochim. Acta 1993, 38,
199–203. doi:10.1016/0013-4686(93)85129-M
11.Dabo, P.; Cyr, A.; Lessard, J.; Brossard, L.; Ménard, H. Can. J. Chem.
1999, 77, 1225–1229. doi:10.1139/v99-120
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
12.Laplante, F.; Brossard, L.; Ménard, H. Can. J. Chem. 2003, 81,
258–264. doi:10.1139/v03-027
13.St-Pierre, G.; Chagnes, A.; Bouchard, N.-A.; Harvey, P. D.;
Brossard, L.; Ménard, H. Langmuir 2004, 20, 6365–6373.
doi:10.1021/la048977v
(https://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
14.Cirtiu, C. M.; Hassani, H. O.; Bouchard, N.-A.; Rowntree, P. A.;
Ménard, H. Langmuir 2006, 22, 6414–6421. doi:10.1021/la0519002
15.Cirtiu, C. M.; Brisach-Wittmeyer, A.; Ménard, H. J. Catal. 2007, 245,
191–197. doi:10.1016/j.jcat.2006.10.010
doi:10.3762/bjoc.14.40
16.Osa, T.; Matsue, T.; Yokozawa, A.; Yamada, T. Denki Kagaku 1986,
54, 484.
17.Osa, T.; Matsue, T.; Yokozawa, A.; Yamada, T. Denki Kagaku 1984,
52, 629.
18.Borup, R.; Meyers, J.; Pivovar, B.; Kim, Y. S.; Mukundan, R.;
Garland, N.; Myers, D.; Wilson, M.; Garzon, F.; Wood, D.; Zelenay, P.;
More, K.; Stroh, K.; Zawodzinski, T.; Boncella, J.; McGrath, J. E.;
Inaba, M.; Miyatake, K.; Hori, M.; Ota, K.; Ogumi, Z.; Miyata, S.;
Nishikata, A.; Siroma, Z.; Uchimoto, Y.; Yasuda, K.; Kimijima, K.-i.;
Iwashita, N. Chem. Rev. 2007, 107, 3904–3951. doi:10.1021/cr050182l
19.Litster, S.; McLean, G. J. Power Sources 2004, 130, 61–76.
doi:10.1016/j.jpowsour.2003.12.055
20.Smitha, B.; Sridhar, S.; Khan, A. A. J. Membr. Sci. 2005, 259, 10–26.
doi:10.1016/j.memsci.2005.01.035
21.Montiel, V.; Sáez, A.; Expósito, E.; García-García, V.; Aldaz, A.
Electrochem. Commun. 2010, 12, 118–121.
doi:10.1016/j.elecom.2009.11.002
546