A study of the electrochemical hydrogenation of o-xylene in a PEM hydrogenation reactor

Article abstract of DOI:10.1016/j.electacta.2012.04.116

In this study, we investigate the electrochemical hydrogenation of o-xylene in a proton exchange membrane hydrogenation reactor (PEMHR). The reactor was operated isothermally over the temperature range 20-68 °C and at a pressure of 1 atm in a semi-batch mode. Hydrogen was fed into the anode compartment and o-xylene into the cathode. The hydrogenation efficiency was investigated at different current densities and temperatures. Results obtained show that the hydrogenation efficiency increases with temperature but decreases with current density. At low current densities the hydrogenation efficiency approaches 100%. A zero dimensional model was used to fit the data and extract a rate constant for the hydrogenation reaction. The activation energy for this reaction was found to be 28 kJ/mole.

Full text of DOI:10.1016/j.electacta.2012.04.116

Electrochimica Acta 75 (2012) 171–178
Contents lists available at SciVerse ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
A study of the electrochemical hydrogenation of o-xylene in a PEM
hydrogenation reactor
R. Fonocho, C.L. Gardner∗, M. Ternan
Department of Chemical Engineering and Centre for Catalysis Research and Innovation, University of Ottawa, Ottawa, Ontario K1N 6N5, Canada
a r t i c l e i n f o
a b s t r a c t
Article history:
In this study, we investigate the electrochemical hydrogenation of o-xylene in a proton exchange mem-
brane hydrogenation reactor (PEMHR). The reactor was operated isothermally over the temperature range
20–68 ◦C and at a pressure of 1 atm in a semi-batch mode. Hydrogen was fed into the anode compart-
ment and o-xylene into the cathode. The hydrogenation efficiency was investigated at different current
densities and temperatures. Results obtained show that the hydrogenation efficiency increases with
temperature but decreases with current density. At low current densities the hydrogenation efficiency
approaches 100%. A zero dimensional model was used to fit the data and extract a rate constant for the
hydrogenation reaction. The activation energy for this reaction was found to be 28 kJ/mole.
© 2012 Elsevier Ltd. All rights reserved.
Received 17 November 2011
Received in revised form 23 February 2012
Accepted 25 April 2012
Available online 14 May 2012
Keywords:
O-xylene
Electrochemical hydrogenation
PEM fuel cell
Hydrogenation kinetics
1. Introduction
Hydrogenation is also an important process in the food indus-
try. In the hydrogenation of unsaturated fats to saturated fats
in the vegetable oil industry, the high temperature and pressure
used often results in the formation of trans-fats that have been
1.1. General introduction
Hydroprocessing is commonly used in the petroleum refining
industry. Current industrial hydro-processes operate over a range
of temperatures (∼300–425 ◦C) and pressures (20–130 atm). Much
of the hydrogen that is used in the refining industry comes from
the steam reformation of natural gas. Steam reforming has a high
energy demand to provide the endothermic heat of reaction plus
the sensible heat needed to attain its high operating temperature.
It also generates a large amount of carbon dioxide that contributes
to climate change.
The current study examines a technique that might decrease
both the energy demand for hydroprocessing and the accompany-
ing quantity of greenhouse gases. Electrochemical hydrogenation
can be performed at a much lower temperature and pressure
than hydroprocessing, but requires an input of electrical power.
The overall intent is to reduce the amount of energy required.
Both electrochemical hydrogenation and hydroprocessing produce
hydrogenated hydrocarbons, H2S, and NH3 as products. Capture
and storage of the carbon dioxide from the steam reforming reac-
tion is a financial burden to the petroleum industry which can
be eliminated by replacing steam reforming with electrochemical
hydrogenation.
associated with coronary heart disease. High temperatures and
pressures also cause a break down in good but unstable saturated
fats.
As an alternative to hydroprocessing, a number of researchers
have recently investigated electrochemical hydrogenation using a
proton exchange membrane (PEM) reactor. In a PEM hydrogena-
tion reactor (PEMHR), hydrogen, introduced directly or produced
at the anode by the electrolysis of water, is oxidized at the anode
to produce protons that migrate through the proton conducting
membrane in response to the applied potential. At the cathode, the
protons in the electrolyte are reduced to hydrogen atoms adsorbed
on the electrode metal, which can react to hydrogenate unsaturated
compounds flowing through the cathode compartment. PEMHRs
offer several advantages over conventional catalytic hydrogena-
tion reactors. High temperatures and elevated pressures are not
required. The lower temperatures reduce energy consumption and
allow thermally unstable molecules to be hydrogenated. The rate of
supply of hydrogen to the cathode is easily controlled, by adjusting
the current.
In this study we report the results of the electrochemical hydro-
genation of o-xylene. The o-xylene molecule has an aromatic ring
and two methyl groups. It was chosen to be a model of the larger
complex aromatic compounds that have both condensed aromatic
rings and side chains, and form a major part of the heavy oils in the
Canadian oil sand deposits.
∗
Corresponding author. Tel.: +1 613 832 3689.
E-mail address: cgardner88@sympatico.ca (C.L. Gardner).
0013-4686/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.electacta.2012.04.116

172
R. Fonocho et al. / Electrochimica Acta 75 (2012) 171–178
Fig. 1. Schematic illustrating the experimental set-up.
1.2. Hydrogenation in PEMHR
The electrogenerated hydrogen atoms at the cathode then either
react with the aromatic bonds in the o-xylene (Eq. (2)) or else com-
bine in an unwanted side reaction to form molecular hydrogen (Eq.
(3)).
Research on PEMHRs has been relatively limited. In pioneering
work, Langer et al. [1–4] studied the electrogenerative hydrogena-
tion of a number of alkenes. Ogumi et al. [5–7] have studied the
electrohydrogenation of olefins as well as a number of other organic
reduction and oxidation reactions. More recently, Pintauro et al.
[8–11] have studied both the hydrogenation of edible oils in a
PEMHR as well as the electrochemical hydrogenation of glucose
[12,13]. A series of studies have been carried out by Ma and co-
workers [14–16] using a PEMHR to hydrogenate a variety of organic
compounds. Recently, Sedighi and Gardner [17] have studied the
electrochemical hydrogenation of ethylene in a PEMHR.
–CH CH– + 2H − S → –CH2–CH2– + 2S; khydrog.
(2)
(3)
−→
2H − S ⇔ H2 + 2S; k H,des, KH
where S represents a surface site on the platinum catalyst.
In our experimental set-up, any hydrogen gas that is evolved is
carried with the o-xylene back to the o-xylene storage tank where
it is separated and its rate of production measured (Fig. 1). It is then
exhausted into the fume hood. The rate of hydrogen production was
measured as a function of current to determine the efficiency of the
hydrogenation process as per Eqs. (4) and (5).
2. Experimental
The reaction was carried out at a constant flow rate of o-xylene
and at temperatures ranging from 25 ◦C to 68 ◦C. The input pres-
sure of the hydrogen to the flow meter was maintained at 377 kPa.
The total pressure in the anode and cathode compartments of the
PEMHR was maintained at 1 atm. The hydrogen gas inflow rate was
controlled using a rotameter. The hydrogen flow rate was kept at
about 100 ml/min, significantly higher than the amount required to
support the maximum current (approximately 77 ml/min at 10 A).
The o-xylene recycle flow rate in the cathode was maintained
at 300 ml/min to ensure the system was isothermal. The recycle
flow rates that are needed to maintain thermal equilibrium and to
remove the evolved hydrogen are many orders of magnitude higher
than the rate of removal of o-xylene through hydrogenation. At a
10 A current, a maximum of 1.25 × 10−2 ml/min is required to sup-
port the hydrogenation reaction. The results that we report for the
rates of hydrogen production and utilization are thus expected to
be independent of o-xylene flow rates at practical flow rate values.
Since hydrogen is the only gaseous entity in the exit flow, the
amount of hydrogen that was consumed or utilized in the hydro-
genation reaction per second, Vreacted, is deduced by subtracting
the measured amount of hydrogen in the cathode exit, Vout, from
2.1. Cell and membrane electrode assembly
Experiments were made using a single PEM electrochemical cell
having a 25 cm2 surface area manufactured by ElectroChem Inc.
The membrane electrode assembly (MEA) used in these experi-
ments had a carbon supported platinum catalyst at both the anode
and cathode (loading: 1 mg Pt cm−2) and used NafionTM 115 as
membrane material. Pure hydrogen, which had been humidified
by bubbling it through water (Fig. 1), was fed to the anode contin-
uously on a once-through basis and this electrode was used as a
reference electrode for the measurement of the cathode potential.
In this study, a batch of o-xylene was stored in an external tank
and then pumped through the cathode compartment of the PEMHR
and then recycled back to the storage tank. Molecular hydrogen
is oxidized at the anode to form protons that migrate across the
Nafion membrane in response to the applied potential. At the cath-
ode, atomic hydrogen is generated on the platinum catalyst surface
by the reduction of these protons as shown in Eq. (1).
−→
H+ + e− + S ⇔ H − S; k H,red
(1)

R. Fonocho et al. / Electrochimica Acta 75 (2012) 171–178
173
Fig. 2. Comparison of the theoretical and measured rate of hydrogen production
at the cathode with no o-xylene present at the cathode: electrode area = 25 cm2
corresponding to a maximum current of 10 A.
Fig. 3. Rate of hydrogen production at the cathode with o-xylene present at the
cathode; electrode area = 25 cm2 corresponding to a maximum current of 10 A.
membrane based on the current that is being passed through the
cell (Eq. (4)). Results for the rate of molecular hydrogen gas evo-
lution from the cathode as a function of current and temperature
are shown in Fig. 3. The results show clearly that the amount of
hydrogen that is evolved generally decreases with increasing tem-
perature indicating that more hydrogen is being consumed in the
hydrogenation reaction as the temperature increases.
Fig. 4 shows the rate of hydrogen utilization, Vreacted, as a
function of current and temperature. The results show that the
hydrogen utilization rate increases up to some value and then
remains essentially constant. There are two possible explanations
for this constant value that is reached: (a) the rate is limited by the
hydrogenation rate; or else (b) it is limited by the rate of diffusion of
o-xylene through the gas diffusion layer (GDL) to the cathode. Cal-
culation shows that it is unlikely that the rate is being limited by
the rate of diffusion of o-xylene to the cathode. Using typical values
for the GDL thickness and porosity [18], we estimate that only 2.5%
of the o-xylene contained in the GDL would be consumed to sup-
port the 10 A current for 1 min. This conclusion is also supported
the theoretical amount of hydrogen that migrates across the mem-
brane, Vtheor, in response to the current flow as shown in Eq. (4).
IRT
Vreacted = Vtheor − Vout
=
− Vout
(4)
2PF
The hydrogenation efficiency can then be defined as
ꢀ
ꢁ
Vreacted
2PFVout
IRT
ꢀhydrog.
=
× 100 = 100 × 1 −
(5)
Vtheor
A hydrogen production experiment (Fig. 2) carried out without
o-xylene at the cathode confirmed that the amount of hydrogen
produced agreed with the amount expected theoretically and that
there was no significant permeation of hydrogen gas through the
membrane.
2.2. Electrochemical measurements and data acquisition
A Kikusui Model PBX 20-10 Bi-Polar Power Supply, connected
in parallel with the electrochemical cell, was used to control the
cell current. The power supply was programmable so that the cell
could be cycled through a predetermined current profile. A Fluke
Hydra data logger was used to monitor and capture cell potential
and current for final analyses using a desktop computer. One chan-
nel of the data logger was directly connected to the electrochemical
cell to measure the cell potential while a second channel monitored
cell current from a connection on the bipolar power supply. Con-
stant current measurements were made at specific currents over
a current range from 0 to 10 A to determine the efficiency of the
hydrogenation process. Polarization curve experiments were per-
formed, by ramping the current from 0 to 10 A at a rate of 20 mA/s.
3. Results and discussion
3.1. Effects of current density and temperature on the
hydrogenation efficiency
The hydrogenation efficiency (Eq. (5)) was measured as a func-
tion of current density and temperature. As explained in the
experimental section, to do this the rate of molecular hydrogen
gas production was measured as a function of current and then the
o-xylene hydrogenation reaction or utilization rate is deduced by
subtracting the measured amount of hydrogen in the cathode exit
from the theoretical amount of hydrogen that migrates across the
Fig. 4. Rate of hydrogen utilization at the cathode with o-xylene present at the cath-
ode at various temperatures: electrode area = 25 cm2 corresponding to a maximum
current of 10 A.

174
R. Fonocho et al. / Electrochimica Acta 75 (2012) 171–178
Fig. 5. Change in cell potential as a function of time in response to a 9 A to 1 A current
step.
Fig. 7. Hydrogenation efficiency as a function of temperature at various current
densities.
by potential step measurements. Fig. 5 shows how the cell poten-
tial changes in response to a step from 9 A to 1 A. If the rate was
being controlled by the rate of diffusion of o-xylene to the cathode,
it would be expected that the cell potential would slowly decrease
as the concentration of o-xylene increased when the current was
stepped from 9 A to 1 A. This behaviour is not observed. The small
increase in cell potential is consistent with a reduction in hydrogen
atom concentration at the lower current (potential). We thus con-
clude that the rate of hydrogen utilization is being limited by the
rate of hydrogenation and not by the rate of diffusion. The limiting
hydrogen utilization rate is strongly temperature dependent indi-
cating that the hydrogenation rate is increasing with temperature
as might have been expected.
Fig. 6 shows the effects of current density on the hydrogena-
tion efficiency of the PEM reactor at various temperatures. At low
current densities, the hydrogenation efficiency can approach 100%.
Similar results have also been reported [11] for the hydrogena-
tion of vegetable oils. Increasing the temperature (Fig. 7) generally
results in an increase in the hydrogenation efficiency presumably
because of an increase in the hydrogenation rate constant. The
reason for the decrease observed in the 10 A curve at high temper-
Fig. 8. Polarization curves: (a) hydrogen at the cathode; (b) o-xylene at the cathode.
atures is unclear but probably results from increased experimental
uncertainty at the higher temperatures.
3.2. Cathodic polarization curves
Fig. 8 shows the polarization curves at 46 ◦C for the cases where
o-xylene is flowing through the cathode compartment (hydrogena-
tion and hydrogen evolution) and where the cathode compartment
contains only hydrogen (hydrogen evolution). Noticeable in this fig-
ure is the positive shift of the open circuit cathode potential at low
current densities when o-xylene flows through the cathode com-
partment compared to hydrogen gas being at both electrodes. This
positive shift results from the reduction in hydrogen partial pres-
sure at the cathode as a result of the hydrogenation reaction. While
the hydrogen pressure at the anode remains constant at 1 atm., the
hydrogen pressure at the cathode is reduced as a result of the hydro-
genation reaction. The shift in cathode potential can be related to
the hydrogen pressure at the cathode by the Nernst equation:
ꢂ
ꢃ
H
2
Panode
RT
ꢁꢀC
=
ln
(6)
H
2
2F
Fig. 6. Hydrogenation efficiency as a function of current density at various temper-
atures.
Pcathode

R. Fonocho et al. / Electrochimica Acta 75 (2012) 171–178
175
The observed shift in cathode potential (∼0.03 V) corresponds to
an equilibrium hydrogen pressure of only about 0.1 atm at low
current density indicating that the hydrogenation reaction con-
sumes a substantial portion of the hydrogen that migrates across
the membrane into the cathode compartment under these con-
ditions. At very low currents, the hydrogenation reaction occurs
spontaneously and does not involve the consumption of external
power. The main loss mechanism in the reactor is, in fact, due to IR
loss in the membrane. Selection of the optimum operating current
for the reactor will depend on a trade-off between operating and
capital costs.
result, adsorbed hydrogen radicals are expected to become readily
available on the cathode catalyst. We suspect that any adsorbed o-
xylene will be quickly removed and replaced by adsorbed hydrogen
radicals.
In initial attempts at modelling the hydrogenation of o-xylene,
we attempted to fit the hydrogen utilization and polarization data
to a Jenkins–Rideal type model assuming that once the initial
aromatic bond had been hydrogenated by two hydrogen atoms,
the hydrogenation of the remaining bonds proceeded very rapidly
once again with the consumption of 6 hydrogen atoms. Using this
assumption, we found we were unable to fit both the polariza-
tion and hydrogen utilization data adequately. The high rate of
hydrogen consumption shifts the cell potential to higher positive
potentials than are observed experimentally at low currents. We
found, however, that we were able to get a good fit to the exper-
imental data using the Jenkins–Rideal model if we assumed that
the hydrogenation of each aromatic bond proceeded sequentially
and that after one aromatic bond had been hydrogenated, then
the next two pi bonds would begin to be rapidly hydrogenated in
series at the same rate. As mentioned earlier, in our experiments
we have not analysed the product stream to determine if signif-
icant amounts of mono- or di-enes are being formed but based
on previous studies with benzene [19] in our model we assume
that 1,2-dimethylcyclohexane and molecular hydrogen are the only
products.
4. Modelling o-xylene hydrogenation
4.1. Hydrogenation mechanism
In our studies, we have not analysed the product stream and
do not know if significant amounts of mono- or di-enes are being
formed. It is our assumption, that similar to the catalytic hydro-
genation of benzene [19], once the first bond in an o-xylene is
hydrogenated, hydrogenation of the other bonds will follow to give
a completely saturated product. The anodic reaction can be written
as
6H2 ⇔ 6H+ + 6e−
(7)
While the cathodic reaction is assumed to be
6H+ + 6e− + o-xylene ⇔ 1, 2-dimethylcyclohexane
4.2. Modelling results
(8)
The overall reaction in the cell is assumed to be the hydrogena-
tion of o-xylene to form 1,2-dimethylcyclohexane according to the
reaction:
A zero-dimensional model, similar to that used by Zhang [24],
has been used to model the reactions at the cathode. In this
model it is assumed that the cathode chamber is perfectly mixed
and that diffusion across the gas diffusion layer is rapid so that
bulk concentrations are present at the catalyst surface. The cath-
ode reactions are assumed to consist of the reversible hydrogen
adsorption/desorption reaction, the reversible hydrogen electro-
oxidation/reduction reaction as well as an irreversible xylene
hydrogenation reaction as shown in Eqs. (1)–(3).
6H2 + o-xylene ⇔ 1, 2-dimethylcyclohexane
(9)
The above equations describe the overall reactions in the cell but do
not describe the detailed mechanism by which the hydrogenation
occurs. Of the many mechanisms postulated for the hydrogena-
tion of unsaturated hydrocarbons, the following three, based on
experiments with ethylene, are often referenced:
Adopting a zero dimensional model leads to the following set of
equations for the rates of these reactions:
4.1.1. Jenkins–Rideal mechanism
ꢀ
ꢁ
˛FꢀC
RT
−→
In the Jenkins–Rideal mechanism [20], the hydrogen radical that
is absorbed on the platinum catalyst is assumed to react directly
with the unadsorbed unsaturated hydrocarbon.
rH,red = −rH,ox = −2 k H,oxꢂH sinh
(10)
Eq. (10) gives the rate at which protons are reduced at the cathode
to form adsorbed hydrogen atoms as shown in Eq. (1).
4.1.2. Horiuti–Polanyi mechanism
−→
rhydrog. = k hydrog.CoXꢂH2
(11)
The Horiuti–Polanyi mechanism [21] postulates that unsatu-
rated hydrocarbon is first adsorbed on the catalyst surface with
hydrogenation then proceeding via reaction with atomic hydrogen
radicals that were dissociatively adsorbed on the catalyst surface.
Polanyi and Greenhalgh [22] extended this concept to the hydro-
genation of benzene.
Eq. (11) gives the rate of the desirable o-xylene hydrogenation reac-
tion as shown in Eq. (2).
−→
−→
rH,des = −rH,ads = k H,adsKHꢂH2 − k H,adsPHb ꢂ02; ꢂ0 = 1 − ꢂH
(12)
s
Eq. (12) gives the rate of the undesirable reaction, the formation
of hydrogen gas. Since the hydrogen atom surface coverage, ꢂH,
is potential dependent, all of the above rates (Eqs. (10)–(12)) are
expected to be potential (current) dependent.
4.1.3. Twigg mechanism
The Twigg mechanism [23] combines features of both of the
above models in that it is assumed that molecular hydrogen from
the gas phase reacts with adsorbed hydrocarbon to add one of its
hydrogen atoms to the hydrocarbon with the other forming an
adsorbed hydrogen radical and that further hydrogenation occurs
with the addition of an atomic hydrogen radical.
All three of these mechanisms result in the same final products.
We have given preference, however, to the Jenkins–Rideal mech-
anism because, of the three, it appears to be most applicable to
electrochemical hydrogenation and we have used this mechanism
in our model. This mechanism is expected to be dominant because
hydrogen is being continuously transferred from the anodic com-
partment to the cathodic compartment by the current and, as a
Using the rate expressions given above, the following set of six
coupled ordinary differential equations can be derived considering
the time dependence of the surface species, ꢂH, the material balance
at the cathode (CobX, Cb , PHb and ꢃH2 ) and conservation of charge
dMC
at the cathode (ꢀc). All of the2se parameters depend on current and
time and are thus variable in the calculation.
Using the rate expressions given above, the time dependence of
the hydrogen surface coverage can then be expressed as follows.
dꢂH
dt
1
FꢄCt
=
(rH,ads − rH,ox − rhydrog.
)
(13)

176
R. Fonocho et al. / Electrochimica Acta 75 (2012) 171–178
The material balance for species i in the cathode chamber can be
written as follows.
dCib
riA
V
= ꢃ0C0 − ꢃCib − NiA = ꢃ0C0 − ꢃCib
−
(14)
i
i
dt
niF
If the reactor is treated as a well stirred batch reactor of volume
V, there is no net inflow or outflow of o-xylene (oX) or 1,2-
dimethylcyclohexane (dMC) from the reactor and, for these species,
Eq. (14) reduces to:
dCib
riA
V
= −
(15)
dt
niF
In Eq. (14), Ni represents the flux of species i to the surface which
involves one or more of the rate expressions. This then leads to
the following equation for the time dependence of the oX and dMC
concentrations in the reactor.
dcobX
dt
A
= −
rhydrog.
(16)
(17)
6VF
dcdbMC
dt
A
=
rhydrog.
Fig. 9. Comparison of the measured and calculated rate of hydrogen production at
various temperatures: electrode area = 25 cm2 corresponding to a maximum current
of 10 A.
6VF
To calculate the material balance equation for the hydrogen,
account must be taken of the hydrogen that flows out from the reac-
tor, ꢃCHb , and both the second and third terms of Eq. (14) must be
2
retained. Conversion of the hydrogen concentration term to partial
pressures using the ideal gas and Dalton’s laws, CH = PH2 /RT, gives
2
the following expression for the time variation of the gas partial
pressure in the cathode compartment.
dPHb
dt
ART
=
[PHb − P]rH,ads
(18)
2
2FPV
The rate of hydrogen production in the reactor is governed by the
difference in the rate of hydrogen reduction and the rate of hydro-
genation and is given by the equation
dꢃH
dt
ART
2
=
(−rH,ox − rhydrog.
)
(19)
2FP
From the charge conservation consideration, the following expres-
sion is obtained for the time dependence of the cathode potential.
dꢀC
dt
A
=
[j − rH,ox
]
(20)
Cdl
The double layer capacitance, Cdl, appears in Eq. (20) because, when
there is a change in potential, some of the current is used for double
layer charging. A typical literature value [24] has been used for Cdl
in our calculation.
Fig. 10. Comparison of the measured and calculated hydrogenation efficiencies at
various temperatures.
The resulting system of six coupled ordinary differential equa-
tions (ODE – Eqs. (13), (16)–(20)) were solved using Scilab©
software. In our simulations, literature values [24] for KH, kH,ads and
kH,ox have been used and are included in Table 1. The only kinetic
parameter that was adjusted to fit the modelled data to the experi-
mental data was the hydrogenation rate constant, khydrog.. A full list
of the parameters and symbols used in our calculations is given in
Table 1. The values of khydrog. at various temperatures were obtained
by fitting the calculated hydrogen production data to the experi-
mental data. Several of these fits are shown in Fig. 9. A comparison
of the experimental and calculated hydrogenation efficiencies of
the reactor is shown in Fig. 10. It is seen that the fitted data generally
agrees well with the experimental data.
From the temperature dependence of the hydrogenation rate
constant (Fig. 11), the activation energy for the electrochemical
o-xylene hydrogenation reaction is determined to be 28 kJ/mole.
This value is close to the value of 25 kJ/mole obtained [25] for the
electrochemical hydrogenation of ethylene at a platinum electrode.
The activation energy for hydrogenation at a gas–solid (platinum)
interface [25] is almost double this value (45.8 kJ/mole). This higher
Fig. 11. Plot of log(hydrogenation rate constant) vs 1/T to determine the activation
energy.

R. Fonocho et al. / Electrochimica Acta 75 (2012) 171–178
177
Table 1
Symbols and model parameters.
Symbol
Definition
Value
Units
Symbol
Definition
Value
Units
atm
A cm−2
A cm−2
A cm−2
A cm−2
V
F
Faraday constant
96,485
C mol−1
cm2
PHb
Partial pressure of hydrogen
Variable
2
A
kH,ox
kH,ads
KH
khydrog
V
ꢃ
ꢃ0
c
Surface area
25
4
402
0.5
Variable
rhydrog.
rH,ads
rH,ox
j
ꢀc
Cdl
ꢄ
Ni
Ct
ꢂH
ꢂ0
˛
Hydrogenation reaction rate
Hydrogen adsorption reaction rate
Rate of hydrogen oxidation/reduction
Current density
Variable
Variable
Variable
Variable
Variable
0.45
Hydrogen oxidation constant
Hydrogen adsorption constant
Hydrogen adsorption equilibrium constant
Xylene hydrogenation constant
Volume
Volumetric outlet flow rate
Volumetric inlet flow rate
Concentration
Current
Gas constant
Temperature
Total pressure
A cm−2 atm−1
A cm−2 atm−1
A cm−2 atm−1
A cm−2 atm−1
cm3
Cathode overpotential
Double layer capacitance
Electrode roughness factor
Molecular flux of species i
Pt atom mole density per cm2
Fraction of surface covered by H atoms
Fraction of surface uncovered
Transfer coefficient
Variable
0
Variable
Variable
8.314
298–346
1
Variable
Variable
cm3 min−1
cm3 min−1
mol cm−3
A
100
Variable
2.2 × 10−9
Variable
Variable
0.5
mol cm−2
I
R
T
J mol−1 K−1
K
atm
mol/l
mol/l
P
ni
Number of electrons
Outlet flow rate of hydrogen
cobX
Bulk concentration of o-xylene
Bulk concentration of 1,2-dimethylcyclohexane
ꢃH
Variable
cm3 min−1
2
cdbMC
activation energy for thermochemical hydrogenation presumably
is associated with the dissociation of molecular hydrogen on the
platinum surface.
hydrogen production rate is also initially low and then increases
linearly at the higher current densities. Fig. 12c illustrates the
consumption of o-xylene during the current sweep. The starting
o-xylene concentration corresponds to pure o-xylene (8.3 mole/l).
Over the short time of the sweep (500 s), only a minor decrease in
o-xylene concentration occurs as a result of hydrogenation.
The results presented have been interpreted in terms of two
competing reactions: the hydrogenation of o-xylene to form 1,2-
dimethylcyclohexane and the undesirable reaction of two adsorbed
hydrogen radicals to form hydrogen gas. At very small currents all of
the protons entering the cathode are used to hydrogenate o-xylene.
An insight into the effect of current density on the hydrogena-
tion process can be obtained from the simulated results showing
how the model parameters change during a 0–10 A sweep at
20 mA/s (a 500 s sweep time). From Fig. 12, it is seen that, at low
current densities, both the hydrogen surface coverage (Fig. 12a)
and the hydrogen pressure (Fig. 12b) are held low as a result of the
hydrogenation reaction which removes the hydrogen atoms that
are formed at the electrocatalyst surface. Fig. 12d shows that the
a
b
c
d
Fig. 12. Effect of current density on the various parameters in the calculation: (a) hydrogen surface coverage; (b) hydrogen pressure in the cathode compartment; (c) o-xylene
concentration; (d) hydrogen production rate.

178
R. Fonocho et al. / Electrochimica Acta 75 (2012) 171–178
As the current increases the protons are used for both reactions.
When the current has been increased sufficiently, a limiting rate of
o-xylene hydrogenation is reached beyond which all of the incre-
mental protons entering the cathode are used to form hydrogen
gas. The limiting rate increases with increasing temperature.
[2] S.H. Langer, H.P. Landi, Journal of the American Chemical Society 86 (1964)
4694.
[3] S.H. Langer, I. Feitz, C.P. Quinn, Journal of the American Chemical Society 93
(1975) 1092.
[4] S.H. Langer, G.P. Sakellaropoulos, Journal of the Electrochemical Society 122
(1975) 1619.
[5] Z. Ogumi, K. Nishio, S. Yoshizawa, Electrochimica Acta 26 (1981) 1779.
[6] Z. Ogumi, S. Ohashi, Z. Takehara, Electrochimica Acta 30 (1985) 121.
[7] Z. Ogumi, et al., Electrochimica Acta 28 (1983) 1687.
[8] W. An, J.K. Hong, P.N. Pitauro, K. Warner, W. Neff, Journal of the American Oil
Chemists Society 75 (1998) 917.
5. Conclusions
The results of this study have shown that o-xylene can be
hydrogenated in a PEM reactor at low temperatures. Increas-
ing the temperatures or decreasing the current density increases
the hydrogenation efficiency. The efficiency of hydrogenation can
approach 100% at low current densities. The experimental results
have been shown to be consistent with a zero dimensional model of
the hydrogenation process based on a Jenkins–Rideal type mecha-
nism whereby adsorbed hydrogen reacts with o-xylene in the liquid
phase. The activation energy for the hydrogenation reaction was
determined to be 28 kJ/mole.
[9] W. An, J.K. Hong, P.N. Pintauro, Journal of Applied Electrochemistry 28 (1998)
947.
[10] W. An, J.-K. Hong, P.N. Pintauro, K. Warner, W. Neff, Journal of the American Oil
Chemists’ Society 76 (1999) 215.
[11] P.N. Pintauro, M.P. Gil, K. Warner, G. List, W. Neff, Industrial & Engineering
Chemistry Research 44 (2005) 6188.
[12] V. Anantharaman, P.N. Pintauro, Journal of the Electrochemical Society 141
(1994) 2729.
[13] V. Anantharaman, P.N. Pintauro, Journal of the Electrochemical Society 141
(1994) 2742.
[14] X.-Z. Yuan, Z.-F. Ma, Q.-Z. Jiang, W.-S. Wu, Electrochemistry Communications
3 (2001) 599.
[15] Z.-F. Ma, X.-Z. Yuan, W.-S. Wu, J.-X. Zhou, Q.-Z. Jiang, 201st ECS Meeting
Abstracts, vol. 2002-1, Philadelphia, No. 1123 (2002).
[16] X.-Z. Yuan, Z.-F. Ma, Q.-G. He, J. Hagen, J. Drillet, V.M. Schmidt, Electrochemistry
Communications 5 (2003) 189.
Acknowledgements
[17] S. Sedighi, C.L. Gardner, Electrochimica Acta 55 (2010) 1701.
[18] M. Saeys, M.-F. Reyniers, M. Neurock, G.B. Marin, Journal of Physical Chemistry
B 109 (2005) 2064.
The authors gratefully acknowledge funding from the Canadian
Government’s Natural Sciences and Engineering Research Council.
We are grateful to the referees for helpful comments and sugges-
tions.
[19] C. Xu, T.S. Zhao, Q. Ye, Electrochimica Acta 51 (2006) 5524.
[20] G.I. Jenkins, E. Rideal, Journal of the Chemical Society (1955) 2490.
[21] J. Horiuti, M. Polanyi, Transactions of the Faraday Society 30 (1934) 1164.
[22] M. Polanyi, R.K. Greenhalgh, Journal of the Chemical Society 85 (1939) 520.
[23] G.H. Twigg, Discussions of the Faraday Society 8 (1950) 152.
[24] J. Zhang, Investigation of CO tolerance in proton exchange membrane fuel cells,
Ph.D. Dissertation, Worchester Polytechnic Institute, June 2004.
[25] A.J. Wieckowski, et al., Journal of the American Chemical Society 107 (1985)
5910.
References
[1] S.H. Langer, H.P. Landi, Journal of the American Chemical Society 85 (1963)
3043.