H3PO3 electrochemical behaviour on a bulk Pt electrode: adsorption and oxidation kinetics

Article abstract of DOI:10.1016/j.electacta.2016.07.045

Polybenzimidazole-type polymer doped with H₃PO₄ is commonly used as the proton-conductive phase in high-temperature proton-exchange membrane fuel cells. However, H₃PO₄ is not stable during fuel cell operation and undergoes reduction by hydrogen on a Pt surface to phosphorus compounds in a lower oxidation state, such as H₃PO₃. In this work the kinetics of H₃PO₃ oxidation on Pt electrode was studied, including an investigation of H₄P₂O₆ as a possible oxidation intermediate. H₃PO₃ adsorption in hydrogen underpotential deposition region was described by a triple Langmuir isotherm corresponding to adsorption on specific Pt crystalline planes. Co-adsorption of hydrogen as well as SO₄^2?, HSO₄^? ions decreased the total amount of adsorbed H₃PO₃. The determined apparent charge transfer coefficients of H₃PO₃ anodic oxidation on a metallic Pt surface were found to be concentration and temperature-dependent, indicating that the nature of the anodic process is complex. From chronopotentiometric measurements of H₃PO₃ and H₄P₂O₆ oxidation on a preoxidised Pt surface it was concluded that, while H₃PO₃ is oxidised by means of a chemical reaction with PtO_x, H₄P₂O₆ undegoes anodic oxidation on the PtO_x surface. According to voltammetry and bulk electrolysis experiments H₄P₂O₆ is not formed as an intermediate product during electrochemical oxidation of H₃PO₃ on a metallic Pt surface.

Full text of DOI:10.1016/j.electacta.2016.07.045

Electrochimica Acta 212 (2016) 465–472
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
H3PO3 electrochemical behaviour on a bulk Pt electrode: adsorption
and oxidation kinetics
*
M. Prokop, T. Bystron , M. Paidar, K. Bouzek
Department of Inorganic Technology, University of Chemistry and Technology Prague, Technicka 5, Prague 6, 166 28, Czech Republic
A R T I C L E I N F O
A B S T R A C T
Article history:
Polybenzimidazole-type polymer doped with H3PO4 is commonly used as the proton-conductive phase
in high-temperature proton-exchange membrane fuel cells. However, H3PO4 is not stable during fuel cell
operation and undergoes reduction by hydrogen on a Pt surface to phosphorus compounds in a lower
oxidation state, such as H3PO3. In this work the kinetics of H3PO3 oxidation on Pt electrode was studied,
including an investigation of H4P2O6 as a possible oxidation intermediate. H3PO3 adsorption in hydrogen
underpotential deposition region was described by a triple Langmuir isotherm corresponding to
Received 7 March 2016
Received in revised form 10 June 2016
Accepted 9 July 2016
Available online 11 July 2016
Keywords:
ꢀ
adsorption on specific Pt crystalline planes. Co-adsorption of hydrogen as well as SO42ꢀ, HSO4 ions
Phosphorous acid
hypophosphoric acid
electrochemical oxidation
Pt electrode
decreased the total amount of adsorbed H3PO3. The determined apparent charge transfer coefficients of
H3PO3 anodic oxidation on a metallic Pt surface were found to be concentration and temperature-
dependent, indicating that the nature of the anodic process is complex. From chronopotentiometric
measurements of H3PO3 and H4P2O6 oxidation on a preoxidised Pt surface it was concluded that, while
H3PO3 is oxidised by means of a chemical reaction with PtOx, H4P2O6 undegoes anodic oxidation on the
PtOx surface. According to voltammetry and bulk electrolysis experiments H4P2O6 is not formed as an
intermediate product during electrochemical oxidation of H3PO3 on a metallic Pt surface.
ã 2016 Elsevier Ltd. All rights reserved.
adsorption
1. Introduction
The oxidation of phosphorus impurities, present at elevated
temperatures in concentrated H3PO4, on Pt electrode has been
Fuel cells represent one of the vital devices for energy
conversion in the hydrogen economy scheme [1]. Of all the
various types, high temperature fuel cells with a proton-exchange
membrane (HT PEM FC) are one of the most promising options due
to their operating temperature of 120–200 ꢁC and consequently
relatively high resistance to CO poisoning along with reasonable
material demands. On the other hand, such an operating
reported by several authors [7,9–11]. These impurities strongly
adsorb on a Pt surface. Their formation takes place at potentials
corresponding to hydrogen adsorption, evolution and their anodic
oxidation proceeds at potentials in the double layer region. The
exact nature of these impurities has been a matter of discussion for
a long time. In the work of Vogel and Baris these compounds were
assumed to be H3PO4 reduction products, namely Pt-P [7].
Sugishima et al. extended the state of knowledge by suggesting
that H3PO3 was one of the impurities [10]. In their following work
the negative influence of H3PO3 on a Pt electroactive surface was
described as well as the fact that H3PO3 oxidation is a competitive
reaction to O2 reduction [12]. The presence of Pt-P and H3PO3
impurities on a Pt surface during polarization of a H3PO4-doped
membrane at a temperature of 125 ꢁC was confirmed by Doh et al.
using scanning photoelectron microscopy [13]. The formation of
impurities was ascribed to the electrochemical reduction of H3PO4
as well as to its chemical reduction by H2, with Pt acting as the
catalyst of these processes. A brief study of the electrochemical
behaviour of H3PO3 on Pt was published by Bravacos et al. and
Trasatti et al. [14,15]. In the work of Prokop et al. both the
electrochemical behaviour of H3PO2 and H3PO3 and the surface
temperature requires the use of membranes based on
a
polybenzimidazole-type polymer doped with H3PO4 [2]. The
negative effects of H3PO4, including adsorption on Pt nanoparticles
and the impact on O2 reduction reaction kinetics, have already
been described in numerous studies [3–6]. Although not frequently
mentioned, the stability of H3PO4 under conditions corresponding
to HT PEM FC operation is also an issue [7,8]. Information on the
reduction of H3PO4 to phosphorus compounds in a lower oxidation
state and their impact on the Pt catalyst is, however, still not
consistent.
* Corresponding author. Tel.: +420 220 444 272.
E-mail addresses: prokopm@vscht.cz (M. Prokop), bystront@vscht.cz
(T. Bystron), paidarm@vscht.cz (M. Paidar), bouzekk@vscht.cz (K. Bouzek).
http://dx.doi.org/10.1016/j.electacta.2016.07.045
0013-4686/ã 2016 Elsevier Ltd. All rights reserved.

466
M. Prokop et al. / Electrochimica Acta 212 (2016) 465–472
coverage of Pt by these phosphorus oxoacids were examined.
According to this work, H3PO3 surface coverage of Pt increases with
rising temperature [11]. This unusual phenomenon was explained
by the tautomeric equilibria between a thermodynamically more
stable “inactive” tetrahedral and a strongly adsorbing “active”
pyramidal form of the acid [16–18].
Baudler and Schellenberg studied the electrochemical oxida-
tion of various phosphorus oxoacids as well as the H3PO4 reduction
mechanism on Pt and other metal electrodes. According to their
findings the oxidation of H3PO3 could proceed with a P(IV)
compound as an intermediate [19]. H4P2O6 is a white crystalline
compound that is stable at ambient temperature in the form of
hypophosphoric acid (H4P2O6) as a possible H3PO3 oxidation
reaction intermediate will be investigated.
2. Experimental
Extra pure chemicals, i.e. 98% phosphorous acid (Acros
Organics, extra pure), 96% sulphuric acid (Fluka, for trace analysis),
85% phosphoric acid (Acros Organics, extra pure, SLR) and 70%
perchloric acid (Acros Organics, for analysis) were used in the
experiments. Crystalline Na2H2P2O6
a method previously described by Remy and Falius [24]. The
presence of crystalline Na2H2P2O6 10H2O phase in the prepared
ꢂ10H2O was synthesised using
ꢂ
H4P2O6
ꢂ
2H2O. Hydrated hypophosphates, e.g. Na2H2P2O6
ꢂxH2O
sample was confirmed by means of X-ray diffraction using an
X’pert PRO (PANanalytical, Netherlands) diffractometer. The purity
(x = 6, 10), are also known. In aqueous solutions H4P2O6 behaves as
a weak tetrabasic acid (pKa1 = 2.2, pKa2 = 2.8, pKa3 = 7.3, pKa4 =
10.0) [20]. Therefore, hypophosphates undergo protonation to
H4P2O6 in strongly acidic aqueous solutions. The prolonged
presence of H4P2O6 in a strongly acidic environment and at
elevated temperatures leads to its disproportionation to H3PO3 and
H3PO4 according to Eq. (1). The standard redox potentials of
H4P2O6/H3PO3 and H3PO4/H4P2O6 couples are indicated in Eqs. (2)
and (3). The standard redox potential of H3PO4/H3PO3 is included
in Eq. (4) for comparison. Even at ambient temperature and when
anhydrous, H4P2O6 with a P(IV)-P(IV) bond undergoes a slow
rearrangement to form isohypophosphoric acid, containing P(III)-
O-P(V) bond, i.e. the two phosphorus atoms in different oxidation
states are connected via an oxygen atom [20]. Isohypophosphoric
acid is not stable and disproportionates to H4P2O7 and H4P2O5.
of Na2H2P2O6 10H2O was higher than 94% according to analyses
ꢂ
performed by means of thermogravimetry on STA PT 700 LT
(Linseis, Germany) with temperature scan rate of 10 ꢁC minꢀ1 from
room temperature up to 170 ꢁC, X-ray diffraction on X’pert PRO
(PANanalytical, Netherlands) and ICP-OES on Optima8000 (Perki-
nElmer, USA), while the remaining 6% probably correspond to
excess water in the salt. Before the experiment the appropriate
amount of Na2H2P2O6
10H2O was dissolved in 0.5 mol Lꢀ1 H2SO4 or
ꢂ
H3PO4 solution where Na2H2P2O6 undergoes protonation to
H4P2O6. All solutions were prepared from fresh deionised water
(conductivity 0.2 m
S cmꢀ1) made by the DIWA deionisation system
(WATEK, Czech Republic). Before each measurement the equip-
ment that would be in contact with the electrolyte and electrodes
was rinsed with distilled water and then treated with 96% H2SO4:
30% H2O2 (1:3) “piranha” solution for at least 15 hours. After
purification by “piranha” solution the equipment was thoroughly
rinsed with deionised water.
H4P2O6 + H2O ! H3PO3 + H3PO4
(1)
A pyrex glass electrochemical cell (50 cm3) tempered by a F12
cryostat (Julabo, Germany) was used for the experiments. The
measurements were performed in a three-electrode arrangement.
Pt foil (7 cm2) and Hg/Hg2SO4 in K2SO4(sat.) solution (MSE) were
used as the counter-electrode and reference electrode, respective-
ly. All potentials in this paper refer to this reference electrode. The
reference electrode was separated from the electrolyte by a double
liquid junction, the first was filled with supporting electrolyte used
in the electrochemical cell, the second with a saturated K2SO4
solution. The working electrode was planar Pt foil with a geometric
area of 1.11 cm2 and a roughness factor of about 1.25. Potentiostatic
batch electrolysis was carried out in a three-electrode arrangement
in a glass electrolytic cell with a Pt mesh (approx. 50 cm2) working
electrode and Pt wire counter-electrode separated from the
electrolyte by a ceramic frit. The same reference electrode was
used as in the voltammetry experiments. The electrolyte solution
was agitated by a PTFE stirrer (cross-like shape, 2 cm wide, rotation
speed 500 RPM). The applied potential was 0.1 V vs. MSE. Prior to
each experiment the working electrode was rinsed thoroughly
with deionised water and cycled for at least one hour between ꢀ0.7
and 0.9 V in corresponding electrolyte at a potential scan rate of
100 mV sꢀ1 After cycling, the used electrolyte was replaced by a
fresh one. All voltammetry measurements were performed with a
HEKA PG310 potentiostat. All experimental voltammograms
presented were corrected for the presence of background currents
and iR drop in the electrolyte. The electrolyte pH was measured by
an Easy Pro automatic titrator (Mettler Toledo, Switzerland) with
an EM45-BNC sensor. The concentration of H3PO3 and H3PO4 in the
electrolyte was analysed by a DIONEX ICS 1000 ion-exchange
liquid chromatograph (Thermo Scientific, USA). An IonPac AS4A-SC
4 mm analytical column with an IonPac AG4A-SC 4 mm guard
column and ASRS-ULTRA II (4 mm) anion self-regenerating
supressor were used. The mobile phase was 1.8 mmol Lꢀ1
Na2CO3 + 1.7 mmol Lꢀ1 NaHCO3 aqueous solution.
H4P2O6 + 2H+ + 2eꢀ Ð 2H3PO3 EH0 P O =H PO = ꢀ0.274 V vs. MSE (2)
4
2
6
3
3
2H3PO4 + 2H+ + 2eꢀ Ð H4P2O6 + 2H2O EH0 PO =H P O = ꢀ1.587 V vs.
3
4
4
2
6
MSE
(3)
2H3PO4 + 2H+ + 2eꢀ Ð 2H3PO3 EH0 PO =H PO = ꢀ0.930 V vs. MSE (4)
3
4
3
3
So far almost no information is available on H3PO3 oxidation
kinetics, except for the work of Trasatti and Alberti who studied the
kinetics of H3PO3 oxidation in 0.5 mol Lꢀ1 H2SO4 aqueous solution
at 25 ꢁC mainly on a bulk Pd electrode [21]. They proposed H3PO3
oxidation to a P(IV) product via catalytic dehydrogenation of H3PO3
on a Pd surface, followed by electrochemical oxidation of the
adsorbed hydrogen. The catalytic dehydrogenation to the P(IV)
compound was suggested as the rate-determining step. The P(IV)
product subsequently disproportionates to H3PO3 and H3PO4 [21].
Trassati and Alberti also briefly studied H3PO3 oxidation on a Pt
electrode, however no kinetic information was provided in their
publication [15]. They suggested that H3PO3 oxidation on a Pt
electrode proceeds by the same mechanism as on a Pd electrode.
However, this conclusion was not supported by sufficient
experimental data, and when large differences in hydrogen
adsorption energies on Pt and Pd are taken into account
(approximately ꢀ45 and ꢀ36 kJ molꢀ1 for Pd-H and Pt-H bond
respectively) their theory still remains open [22,23].
The main goal of the present study is to extend the findings
presented in our previous work [11], to include information on
H3PO3 adsorption in the presence of SO42ꢀ, HSO4ꢀ and ClO4ꢀ in the
electrolyte solution. The kinetics of H3PO3 electrochemical
oxidation will be determined for a polycrystalline Pt electrode
and for 0.5 mol Lꢀ1 H2SO4 aqueous solution at various temper-
atures. Furthermore, the electrochemical behaviour of

M. Prokop et al. / Electrochimica Acta 212 (2016) 465–472
467
[F(ig._1)TD$FIG]
3. Results and discussion
3.1. Adsorption of H3PO3 on a Pt surface
Adsorption isotherms of H3PO3 determined in 0.5 mol Lꢀ1
H2SO4 aqueous solution in the electrode potential range of
hydrogen underpotential deposition on
a polycrystalline Pt
electrode presented previously in [11] were fitted using Athena
Visual Studio with a Langmuir triple-adsorption model according
to Eq. (5). The following assumptions were made: three indepen-
dent types of adsorption sites are present on the electrode surface;
H3PO3 adsorption on each type of adsorption site at constant
electrode potential follows the Langmuir isotherm. The fitted
values of equilibrium adsorption constants and maximum cover-
ages are included in Table 1. Triple adsorption can be interpreted as
independent adsorption of H3PO3 on three different low-index
crystalline planes present on a bulk polycrystalline Pt electrode.
Fig. 1. H3PO3 adsorption isotherms on a polycrystalline Pt electrode in 0.5 mol Lꢀ1
H2SO4 aqueous solution. The logarithm was calculated using concentrations in
mol mꢀ3. Temperatures are included in plot inset.
The fitted isotherms, showing
a good agreement with the
experimental data, are presented in Fig. 1.
Kads;1
ads;1c þ 1
c
Kads;2
ads;2c þ 1
c
Kads;3
ads;3c þ 1
c
u
u
¼
umax;1
þ
umax;2
þ
umax;3
ð5Þ
K
K
K
[(Fig._2)TD$FIG]
—partial surface occupation by H3PO3, c—H3PO3 concentration in
mol mꢀ3, Kads,x—equilibrium adsorption constant on a crystalline
plane numbered x,
crystalline plane x. It is assumed that umax,1
Here, the partial surface occupation is defined as a fraction of the
Pt electrode surface covered by the studied adsorbent (the
remaining part of the surface, 1 – , is not covered by the studied
u
max,x—maximum partial surface occupation on
+
umax,2 +
umax,3 = 1.
u
u
adsorbent). Such a quantity does not allow one to distinguish
between mono- and multilayer adsorption. It is thus not to be
confused with surface coverage.
The effect of co-adsorbing H, SO42ꢀ, HSO4ꢀ and ClO4ꢀ on H3PO3
adsorption on a Pt surface was examined by cyclic voltammetry
using varying lower reversal potential in 0.5 mol Lꢀ1 H2SO4 and
HClO4 aqueous solutions with an addition of H3PO3. Aqueous
H2SO4 solution was chosen as
a supporting electrolyte for
continuity with previous experiments, while HClO4 represents a
non-adsorbing electrolyte for comparison purposes. During the
experiment, an electrode potential was scanned from a fixed
starting potential (ꢀ0.25 V) to a more negative lower-reversal
potential where H3PO3 adsorbs and then a potential was scanned
to positive electrode potentials in order to oxidise H3PO3 present in
the vicinity of the electrode. This oxidation process is on the
Fig. 2. Cyclic voltammetry of a polycrystalline Pt electrode in 0.5molLꢀ1 H2SO4 +
1 mmol Lꢀ1 H3PO3 at 70ꢁC. Electrolyte saturated with N2. Scan description:
ꢀ0.25 V ! lower reversal potential ! 0.84 V ! ꢀ0.25 V, potential sweep rate
50mV sꢀ1. The value of the lower reversal potential is stated in the plot inset.
Inserted plot shows evolution of electrode potential in time.
experimental series was selected from QOx;i values. The change in
H3PO3 adsorbed amount expressed as number of monolayers
desorbed DuM (dimensionless) was determined using Eq. (6),
z (= 2) is number of electrons removed per molecule of H3PO3,
voltammograms visible as the peak pa [11], see Fig. 2. Peak pa’
resulting from H3PO3 oxidation on Pt surface freed from oxides
during negative going potential scan was not used for study due to
non-defined conditions of H3PO3 adsorption. The objective of these
experiments was to examine the influence of potential on the
amount of adsorbed H3PO3. The evaluation was performed using
the following treatment. First, the anodic charge QOx;i in C was
obtained by the current (measured during the positive going
sweep) integration in the potential region between ꢀ0.2 and 0.8 V
for each voltammogram. As one can see from the voltammograms
presented in Fig. 2, the only differences in these parts of
F (= 96 485 C molꢀ1) is Faraday constant, GPt (= 2.2 10ꢀ5 mol mꢀ2
)
ꢂ
is surface concentration of Pt atoms and EASA in m2 is
electrochemically active surface area of the electrode determined
by means of hydrogen underpotential deposition. The calculation
of DuM is based on Faraday law and, in principle, only allows one to
evaluate changes in the extent of H3PO3 adsorption and not the
total amount of H3PO3 adsorbed on the electrode surface.
voltammetry curves are due to variation of oxidation peaks pa.
Subsequently, the maximum charge QOx;max obtained within the
QOx;i ꢀ QOx;max
DuM
¼
ð6Þ
z
ꢂFꢂGPtꢂEASA
In 0.5 mol Lꢀ1 H2SO4 solution at 25 ꢁC the H3PO3 oxidation peak
(and the amount of originally adsorbed H3PO3) was largest when
the lower reversal potential had a value of approximately ꢀ0.40 V
and it decreased when this potential was either more positive or
negative, see Fig. 3. H3PO3 desorption was most pronounced in the
potential region of hydrogen adsorption/evolution, resulting in a
lower charge of the H3PO3 oxidation peak in the following positive-
Table 1
Equilibrium adsorption constants and maximal partial surface occupation by H3PO3
of polycrystalline Pt electrode in 0.5 mol Lꢀ1 H2SO4 aqueous solutions saturated
with N2.
T / ꢁC
Kads,1
umax,1
Kads,2
umax,2
Kads,3
25
70
3.5
4.5
ꢂ
ꢂ
107
104
0.022
0.121
3.07
5.51
ꢂ
ꢂ
101
101
0.541
0.525
5.39
1.25
ꢂ
ꢂ
10ꢀ4
10ꢀ3

468
M. Prokop et al. / Electrochimica Acta 212 (2016) 465–472
[(Fig._3)TD$FIG]
(mainly as H4P2O5 and H5P3O7 produced by condensation of
“active” P(OH)3 molecules). Their concentration in the electrolyte
solution at a higher temperature increases, as already discussed in
our previous work [11]. The presence of adsorbed polyacid also
explains the more pronounced desorption with decreasing
electrode potential at elevated temperatures. It is obvious from
Fig. 3 that, if the temperature is raised from 25 ꢁC to 55 and 70 ꢁC,
the maximum adsorption appears at less negative potentials
(though the minimum is less pronounced for the data measured at
55 and 70 ꢁC). A likely explanation for this observation is a shift in
the potential of zero charge with temperature in the studied
system. However, no information on this effect could be found in
the available literature. A further phenomenon is also apparent
from the data, which confirms the presence of adsorption, namely
a shift of the H3PO3 oxidation peak pa potential towards negative
potential values with its increasing overall charge, see Fig. 2. If the
peak were governed purely by the oxidation kinetics and the
diffusion of electroactive species towards the electrode, the
opposite shift would be expected.
Fig. 3. Dependence of desorbed H3PO3 monolayers on lower reversal potential
during cyclic voltammetry measurements in 0.5 mol Lꢀ1 H2SO4 aqueous solution
with an addition of 1 mmol Lꢀ1 H3PO3 (ꢀ0.25 V ! lower reversal potential ! 0.84
The behaviour of H3PO3 in 0.5 mol Lꢀ1 HClO4 electrolyte was
comparable to that in H2SO4 solution, except for the intensities of
the H3PO3 oxidation peaks. The peak charges in HClO4 electrolyte
at 25 ꢁC were found to be approximately 40% higher than those
observed in H2SO4 solution. This difference can be ascribed to
stronger adsorption of SO42ꢀ and especially HSO4ꢀ anions on the Pt
electrode, in comparison to relatively inert ClO4ꢀ, which leaves
lower number of surface sites available for H3PO3 adsorption. Fig. 4
clearly shows that the amount of H3PO3 adsorbed during polar-
isation in the H adsorption/evolution potential region at 25 ꢁC was
highest at a lower reversal potential of ꢀ0.35 V (Fig. 4). This value
again corresponds to the reported zero charge potential of
polycrystalline Pt in 0.1 mol Lꢀ1 HClO4 at 25 ꢁC [25]. However,
unlike in the H2SO4 solution, the desorbed amount of H3PO3 at
25 ꢁC did not change dramatically until the lower reversal potential
reached a value of ꢀ0.6 V, corresponding to the onset of H2
evolution. This is due to the lack of competitive adsorption of the
supporting electrolyte anion. Hence, in the HClO4 electrolyte the
adsorption of H3PO3 is negatively affected mainly below ꢀ0.6 V, i.e.
by hydrogen adsorption in the H2 evolution potential region. In the
case of temperatures of 55 and 70 ꢁC the dependence of the
amount of desorbed H3PO3 on lower reversal potential is almost
identical, which is slightly surprising. The desorbed amount of
H3PO3 rises substantially when the lower reversal potential is
V ! ꢀ0.25 V), potential sweep rate 50 mV sꢀ1
. Electrolyte saturated with N2.
Electrolyte temperature is stated in the figure inset. Inserted plot shows detail of
data recorded at temperature of 25 ꢁC.
going potential scan. The value of ꢀ0.40 V coincides with the
reported potential of zero charge (ꢀ0.384 V) of a Pt electrode in
0.5 mol Lꢀ1 H2SO4 at 25 ꢁC [25]. The adsorption of H3PO3 in the
form of an anion would lead to dramatic changes in the potential of
zero charge value. It can thus be assumed that, in the system
studied, H3PO3 was adsorbed on the Pt surface in the protonated
form. This is also in agreement with our previous conclusions that
it is the “active” P(OH)3 form of H3PO3 that adsorbs on the Pt
surface. Finally, it agrees with P(OH)3 reported acido-basic
properties (pKa1 = 7.4). As the pH value of the studied solution
was about 0.35, practically all P(OH)3 molecules in the electrolyte
were protonated [17].
At 25 ꢁC the maximum desorption of 0.78 of a H3PO3 monolayer
was observed when the lower reversal potential was shifted from
ꢀ0.4 to ꢀ0.7 V. From the adsorption isotherm for 25 ꢁC, presented
in Fig. 1 (determined from cyclic voltammograms with a lower
reversal potential of ꢀ0.65 V, H3PO3 concentration of 1 mmol Lꢀ1
)
it is clear that, at a potential of ꢀ0.65 V, approximately 50% of the
electrode surface is still covered by H3PO3. At the same time,
desorption of about 0.5 monolayer at this potential, compared to
ꢀ0.4 V, is visible in Fig. 3. Thus, the maximum amount of adsorbed
H3PO3 observed at potentials around ꢀ0.4 V attains at least 1
monolayer. According to the voltammograms similar to these
presented in Fig. 2, the co-adsorption of H on the Pt surface at
temperatures of 55 and 70 ꢁC has a pronouncedly negative effect on
the amount of H3PO3 adsorbed. At these temperatures the
maximum amount of desorbed H3PO3 is equivalent to at least
1.7 and 2.3 monolayers, respectively. This corresponds to the
situation when the lower reversal potential is lowered from about
ꢀ0.25 to ꢀ0.68 V (Fig. 3). If the information about the adsorption
[F(ig._4)TD$FIG]
ꢁ
from Fig. 1 (isotherm at 70 C determined from cyclic voltammo-
grams with lower reversal potential of ꢀ0.62 V, H3PO3 concentra-
tion of 1 mmol Lꢀ1) and Fig. 3 (for ꢀ0.62 V) at a temperature of 70
ꢁC is combined, it yields a value of about 2.3 monolayers. This is in
agreement with the maximum amount of desorbed H3PO3
observed at ꢀ0.68 V in Fig. 3 and it documents the agreement
between the previous and the new published data. Surprisingly,
the maximum amount of adsorbed H3PO3 largely exceeds the
amount corresponding to the monolayer coverage. This is
interesting because direct contact between the Pt surface and
the “second” and “third” H3PO3 layers, required for chemisorption,
is not possible. Therefore, it seems that at least part of the H3PO3
molecules is present on the Pt surface in the form of a polyacid
Fig. 4. Dependence of desorbed H3PO3 monolayers on lower reversal potential
during cyclic voltammetry measurements in 0.5 mol Lꢀ1 HClO4 aqueous solution
with an addition of 1 mmol Lꢀ1 H3PO3 (ꢀ0.25 V ! lower reversal potential ! 0.84
V ! ꢀ0.25 V), potential sweep rate 50 mV sꢀ1
. Electrolyte saturated with N2.
Electrolyte temperature is stated in the figure inset. Inserted plot shows detail of
data recorded at temperature of 25 ꢁC.

M. Prokop et al. / Electrochimica Acta 212 (2016) 465–472
469
Table 2
decreased from ꢀ0.3 V to ꢀ0.6 V, where it attains a value of about
1.5 monolayers. Below ꢀ0.6 V the intensity of desorption again
increases even more due to the hydrogen adsorption in the region
of H2 evolution and reaches a value of 2.2 monolayers at ꢀ0.66 V.
This value of the total amount of H3PO3 desorbed at this potential is
very similar for both supporting electrolytes studied, i.e. it does not
seem to be influenced by the character of the anion.
Temperature and concentration dependence of H3PO3 electrochemical oxidation
charge transfer coefficient
a and temperature dependence of apparent reaction
order on a polycrystalline Pt electrode in 0.5 mol Lꢀ1 H2SO4 aqueous solution
saturated with N2. The Tafel equation is in the form E = Elogj=0 + logj(273.15 + T)R/
(Fa).
a
Elogj=0/V
reaction order
c / mmol Lꢀ1
c / mmol Lꢀ1
T / ꢁC
0.2
1
5
0.2
1
5
3.2. Kinetics of H3PO3 oxidation
25
55
70
0.77
0.79
0.95
0.57
0.74
0.75
0.53
0.67
0.68
0.09
0.01
ꢀ0.03
0.10
0.03
0.01
0.09
0.03
0.01
0.18
ꢀ0.11
ꢀ0.04
H3PO3 polarisation curves were obtained by potentiostatic
sampling voltammetry at 25, 55 and 70 ꢁC for H3PO3 concen-
trations of 0.2, 1 and 5 mmol Lꢀ1. The voltammograms recorded at
25 ꢁC are presented in Fig. 5. Oxidation of H3PO3 starts at around
ꢀ0.05 V and the limiting current plateau can be found on the
The apparent reaction orders of H3PO3 electrochemical oxida-
tion in the potential range of interest were determined from the
dependence of current density at constant potentials on the
logarithm of H3PO3 concentration. The slope of this dependence is
the apparent order of reaction with respect to H3PO3. The current
densities were read for each concentration and potential in the
Tafel region of the polarisation curve. The calculated values of the
reaction orders at 25, 55 and 70 ꢁC are included in Table 2. With
increasing temperature the reaction order of H3PO3 oxidation
decreased from a value of 0.18 at 25 ꢁC to ꢀ0.04 at 70 ꢁC. This is in
agreement with the above discussed dependence of H3PO3
adsorption on temperature because the rate of H3PO3 oxidation
is not directly influenced by its bulk concentration. Instead, it is
largely affected by the surface excess of H3PO3.
In the framework of this study an attempt was also made to
determine the H3PO3 apparent diffusion coefficients, Dapp, by
integrating the current response obtained during polarisation
measurements by the potential step measurements. In the method
described earlier by Bard and Faulkner [28] the Dapp is calculated
from a linearised coulometric plot using Eq. (8). The data selected
from the potential step measurements were located in the
diffusion-limited area, thus permitting the effects of adsorption
and double layer charging to be neglected.
voltammograms. In the case of H3PO3 concentration of 5 mmol Lꢀ1
,
the limiting plateau is not visible due to the formation of a PtO
layer by which the mechanism of H3PO3 oxidation changes, as will
be discussed later in section 3.4. An increase in temperature to 55
and 70 ꢁC shifted the onset of H3PO3 oxidation to a more negative
potential of around ꢀ0.1 V. At the same time the limiting plateau
was significantly less pronounced than at 25 ꢁC. This is especially
apparent in the case of the solution containing 5 mmol Lꢀ1 H3PO3.
To widen the kinetics-governed potential region of the polarisation
curves mass transfer corrected currents were calculated according
to Eq. (7) [9,26].
I
ꢂ
Ilim
Icorr
¼
ð7Þ
Ilim ꢀ I
Icorr—mass transport corrected current in A, I—measured current in
A, Ilim—average current in the limiting plateau of the polarisation
curve in A.
The apparent charge transfer coefficient values determined,
presented in Table 2, vary between 0.52 and 0.95. The values
increase with both the rising temperature and the decreasing
H3PO3 concentration in the electrolyte. It has been shown, e.g. by
Holewinski et al., that in the presence of adsorption of species
involved in the electrochemical reaction, the charge transfer
coefficient value is not only a function of the position of the rate-
determining step in the reaction mechanism, but also of the extent
of adsorption of all species involved [27]. Since the extent of H3PO3
(and its oxidation intermediates) adsorption in the potential range
of its oxidation is unknown, the charge transfer coefficient values
are presented without a mechanistic interpretation. This will be
the subject of future work.
ꢀ
ꢁ
2t1=2
1
H1=2
QðtÞ ¼ nFAkf cr
ꢀ
p1=2
ð8Þ
H
where
p1=2
2t1i =2
kf
H ¼
¼
D
ð9Þ
1=2
app
Q(t)—charge in time in C, n—molar amount in mol, kf—heteroge-
nous rate constant of oxidation in m sꢀ1, cr—concentration of H3PO3
in mol mꢀ3, t—time in s, ti—time in s (obtained from the intercept of
dependence Q vs. t1/2 with x-axis), Dapp—apparent diffusion
[(Fig._5)TD$FIG]
coefficient of H3PO3 in m2 sꢀ1
.
The calculated Dapp values are summarised in Table 3. The Dapp
values were found to be dependent on both the temperature and
the concentration of H3PO3 in the electrolyte. As expected, Dapp
increases with rising temperature, namely at each concentration
Dapp is an approximately linear function of the temperature.
However, Dapp was found to change excessively with H3PO3
concentration. In particular, an increase in H3PO3 concentration
Table 3
Temperature and concentration dependence of H3PO3 apparent diffusion coeffi-
cient in 0.5 mol Lꢀ1 H2SO4 aqueous solution saturated with N2.
T / ꢁC
c / mmol Lꢀ1
0.2
1
5
25
55
70
0.63
2.62
3.48
0.26
1.52
2.15
0.05
0.43
0.56
Dapp / 10ꢀ9 m2 sꢀ1
Fig. 5. Potentiostatic voltammetric curves of H3PO3 oxidation on a polycrystalline
Pt electrode recorded at 25 ꢁC in 0.5 mol Lꢀ1 H2SO4 with an addition of H3PO3.
Electrolyte saturated with N2. H3PO3 concentrations are stated in the plot inset.

470
M. Prokop et al. / Electrochimica Acta 212 (2016) 465–472
[F(ig._7)TD$FIG]
from 0.2 to 5 mmol Lꢀ1 leads to a decrease in Dapp by a factor of 10
at constant temperature. As a result, Dapp values in the range of
10ꢀ9–10ꢀ11 m2 sꢀ1 were calculated. It is obvious that these Dapp
values (especially at higher concentrations) have no physical
meaning due to some process hindering electrode activity,
especially at higher H3PO3 concentrations. A likely explanation
is blockage of a substantial part of the Pt surface by slowly
desorbing product (or H3PO3 oxidation intermediate) that
effectively decreases the rate of H3PO3 anodic oxidation at the
ꢀ
electrode. Most probably the H2PO4 anions produced at the
electrode surface remain adsorbed on the Pt surface and are
responsible for blocking its active sites. This hypothesis is
ꢀ
supported by the fact that H2PO4 anions adsorb mainly at
potentials roughly corresponding to those of H3PO3 oxidation [6].
3.3. H4P2O6 electrochemical behaviour on a Pt electrode
The electrochemical behaviour of H4P2O6 as a potential H3PO3
oxidation intermediate product was examined in 0.5 mol Lꢀ1
H2SO4 solution. The corresponding cyclic voltammograms are
presented in Fig. 6. The addition of H4P2O6 to 0.5 mol Lꢀ1 H2SO4
electrolyte caused a slight delay in PtOx formation. A peak related
to H4P2O6 electrochemical oxidation to H3PO4 was observed in the
potential region of 0.4–0.7 V in all cases. This is well in the PtOx
region, which shows that H4P2O6 only oxidises on an already
formed PtOx surface. Similar to H3PO3, H4P2O6 oxidation requires a
substantial overpotential to proceed, since the standard redox
potential of H4P2O6/H3PO4 couple is ꢀ1.587 V, see Eq. (2). The
voltammograms in Fig. 6 allow one to conclude that H4P2O6 has no
impact on the electro-active surface area of Pt. Slight differences,
especially visible on the voltammogram recorded in the presence
of 10 mmol Lꢀ1 H4P2O6 in the electrolyte, can be attributed to the
effect of H3PO3 emerging in the electrolyte due to H4P2O6
disproportionation, according to Eq. (1). This conclusion is
supported by the appearance of a small oxidation peak around
0.1 V, corresponding to H3PO3 anodic oxidation. Such effects due to
H4P2O6 chemical decomposition are even more pronounced at
elevated temperatures, see Fig. 7.
Fig. 7. Cyclic voltammograms of 1 mmol Lꢀ1 H4P2O6 solution showing electro-
chemical behaviour on
polycrystalline Pt electrode in different 0.5 mol Lꢀ1
supporting electrolytes. Electrolyte saturated with N2. Electrolyte compositions and
temperatures used are included in plot inset.
a
ꢀ
ꢀ
H2PO4 compared to HSO4 and consequently a lower electro-
active area available for H4P2O6 oxidation. The same effect is
responsible for an approximately 50 mV shift of PtOx formation in
H3PO4 solution to more positive potentials than in H2SO4 solution.
On the other hand, the observed chemical decomposition of
H4P2O6 was much slower in H3PO4 than in H2SO4. This is
understandable because of the substantial difference in acidity
of the solutions (pH of 0.5 mol Lꢀ1 H2SO4 aqueous solution is 0.35
and that of 0.5 mol Lꢀ1 H3PO4 is 1.04). Thus, H4P2O6 decomposes
faster in H2SO4 solution due to its higher acidity. At the same time,
H3PO4 is the final product of H4P2O6 decomposition in an acidic
environment, therefore its presence decreases the driving force of
the disproportionation.
3.4. H3PO3 and H4P2O6 oxidation on an oxidised Pt surface
Both compounds studied, i.e. H3PO3 and H4P2O6, were found to
be oxidised in the region of PtOx formation, where the parallel Pt
surface oxidation complicates an analysis of the data. In order to
elucidate the mechanism of these oxidation processes a series of
chronopotentiometric experiments was performed. First, the Pt
working electrode was oxidised at 0.6 V for 5 s and then a decrease
in the open circuit potential (OCP) of the electrode was recorded.
Oxidation of the electrode in pure 0.5 mol Lꢀ1 H2SO4 solution led to
oxide coverage of approximately 60% (if PtO is assumed) of the
electrode surface. The results of these experiments are presented
in Fig. 8. OCP decay, observed after initial oxidation in 0.5 mol Lꢀ1
H2SO4, was slow and after 6000 s OCP attained a value of 0.19 V and
further decreased. In this case, the decay of OCP can be explained
by gradual dissolution of the PtOx layer in H2SO4 solution until a
state of equilibrium between Pt/PtOx and Pt2+ in the solution is
reached [29]. In order to ensure that this effect is not due to the
oxidation of impurities in the electrolyte solution, it was treated
with H2O2 prior to the experiements. H2O2 was consequently
decomposed by heating in a closed vessel.
The electrochemical behaviour of H4P2O6 was also examined in
ꢀ
H3PO4 aqueous solution in order to evaluate the effect of H2PO4
anion on H4P2O6 oxidation. The voltammograms measured in
H2SO4 and H3PO4 solutions at 25 and 70 ꢁC are compared in Fig. 7.
In H3PO4 solution the oxidation peak of H4P2O6 around 0.7 V was
much less pronounced and its maximum shifted to more positive
potentials than in H2SO4. The reason is stronger adsorption of
[(Fig._6)TD$FIG]
In the case of preoxidation in the presence of H3PO3 in the
solution the passed charge was higher because at 0.6 V the anodic
oxidation of H3PO3 takes place in parallel with PtOx formation. As a
first approximation it was assumed that coverage by PtOx after
preoxidation was comparable to the case of preoxidation without
H3PO3. The presence of H3PO3 in the electrolyte led to a dramatic
decrease in the OCP of the electrode after a few seconds of
chronopotentiometric measurement and an OCP of ꢀ0.24 V was
attained after 100 s of the experiment. This indicates fast chemical
Fig. 6. Cyclic voltammograms of H4P2O6 solution showing electrochemical
behaviour on a polycrystalline Pt electrode in 0.5 mol Lꢀ1 H2SO4 with an addition
of H4P2O6 at 25 ꢁC. Electrolyte saturated with N2. Electrolyte compositions are
included in plot inset.

M. Prokop et al. / Electrochimica Acta 212 (2016) 465–472
471
[(Fig._8)TD$FIG]
the onset of electrode polarisation it reached a value of about 10%
of the initial current density. Therefore, in order to ensure a
reasonable average current density of the electrolysis, the potential
was changed to ꢀ0.25 V for 1 s after each 60 s of electrode
polarisation at a potential of 0.1 V. It is conceivable that, as the
H4P2O6 molecule contains a PꢀꢀP bond, it would most likely form
from H3PO3 oxidation intermediate products at a high concentra-
tion of H3PO3 at or near the electrode surface. The electrode
potential value of ꢀ0.25 V was chosen so as to allow the adsorption
of substantial amount of H3PO3 (as was discussed in section 3.1)
and to avoid unnecessarily large charging currents during the
potential step. This situation is similar to that observed during the
chronoamperometric determination of Dapp, discussed in section
3.2 and it corresponds to blockage of the electrode surface by
(intermediate) H3PO3 oxidation products. From the changes in
H3PO3 concentration, the amount of produced H3PO4, the value of
anodic charge passed during electrolysis and assuming the transfer
of 2 electrons per oxidised molecule of H3PO3, 100% selectivity for
H3PO3 to H3PO4 oxidation and current efficiency of 99.9% were
obtained for this process. As mentioned above, H4P2O6 could not be
oxidised at the electrode potential of 0.1 V applied during
electrolysis. The hydrolytic disproportionation described by
Eq. (1) is, under the conditions of electrolysis, a relatively slow
process. This shows that, if a reasonable amount of H4P2O6 were to
be generated in the course of anodic electrolysis, it would have to
be present in the electrolyte and both the current efficiency and the
selectivity values of H3PO3 oxidation to H3PO4 as calculated above
would be substantially lower. As a result, it can be summarised that
H4P2O6 is not the intermediate in H3PO3 oxidation to H3PO4.
Fig. 8. Chronopotentiometric (I = 0) curves recorded on
a polycrystalline Pt
electrode at temperature 25 ꢁC. 5 s electrode prepolarisation at E = 0.6 V. Supporting
electrolyte 0.5 mol Lꢀ1 H2SO4 saturated with N2. Electrolyte compositions are
included in plot inset.
oxidation of H3PO3 on an oxidised platinum surface. In order to
confirm that the observed rapid drop in OCP was not caused by an
insufficient degree of Pt surface oxidation, a second measurement
with H3PO3 was performed. This time the Pt was preoxidised in the
absence of H3PO3 in the electrolyte, so initial surface coverage by
PtOx was defined and not affected by H3PO3. It can be seen from
Fig. 8 that the decay in OCP follows that previously observed in
0.5 mol Lꢀ1 H2SO4 solution. H3PO3 was added to the electrolyte
solution after 30 s of chronopotentiometric measurement and OCP
immediately decreased. Within 100 s OCP value stabilised at
ꢀ0.1 V. This potential is slightly higher than in the case of
preoxidation in the presence of H3PO3, indicating a different
amount of PtOx on the electrode surface (most likely due to
interference of O2 absorbed into the electrolyte during the addition
of H3PO3).
Unlike H3PO3, the presence of H4P2O6 in the solution had almost
no impact on the decay in OCP observed after the preoxidation
period. The recorded results were comparable to those measured
in pure H2SO4 solution, showing that H4P2O6 is fairly stable in
contact with a PtOx surface under the experimental conditions
used. A minor difference is visible only within the first few seconds
of the experiment when a decrease in OCP was relatively rapid in
solution containing H4P2O6, presumably due to the partial
disproportionation of H4P2O6 in an acidic enviroment with
H3PO3 as one of the products. These results also suggest that
H4P2O6 oxidation previously discussed in the context of the
voltammograms in Figs. 6 and 7 is electrochemical in nature. In
other words, it proceeds by means of electron transfrer between
the H4P2O6 molecule and the electrode surface, not by a chemical
reaction between H4P2O6 and PtOx present on the electrode
surface.
4. Conclusion
It can be concluded that the electrochemical behaviour of
H3PO3 on a Pt electrode is quite complex. First, the adsorption of
H3PO3 on Pt electrode takes place. Its extent was found to depend
on several parameters: besides temperature, it is mainly the
electrode potential and nature of the supporting electrolyte anion.
In particular, the adsorbed amount is highest at the electrode
potential near the potential of zero charge of the surface and in the
non-adsorbing supporting electrolyte solutions such as HClO4. The
adsorbed amount of H3PO3 drops in H2SO4 electrolyte solutions
(due to co-adsorption of supporting electrolyte anions) and at
potentials of hydrogen underpotential deposition and in the
hydrogen evolution region (adsorption of hydrogen). At temper-
atures above room temperature the adsorbed amount exceeds that
corresponding to monolayer coverage and it is probable that,
besides H3PO3, species like H4P2O5 and H5P3O7 are present at the
electrode interface.
Secondly, H3PO3 can be oxidised on both a metallic Pt surface
and a PtOx surface. While in the former case the reaction proceeds
electrochemically via electron removal from a H3PO3 molecule, in
the latter case it involves chemical oxidation of H3PO3 by PtOx,surf
.
The anodic charge transfer coefficient of the process was found to
depend on the bulk concentration of H3PO3 and the temperature of
the electrolyte solution. It seems that the desorption of an
oxidation (intermediate) product (most likely H2PO4ꢀ) from the Pt
surface is a fairly slow process. As a result, it accumulates on the Pt
surface and substantially reduces the rate of the overall oxidation
process.
In this work the electrochemical behaviour of hypophosphoric
acid, H4P2O6, was also studied. In contrast to H3PO3, H4P2O6
behaves like a fairly inert compound. It neither adsorbs on the Pt
surface nor does it react with Pt oxides, and its oxidation proceeds
only on a preoxidised Pt surface by means of electron transfer.
Using batch electrolysis it was confirmed that H4P2O6 does not
3.5. H3PO3 batch electrolysis
To examine the possibility that H4P2O6 is formed as an
intermediate product during H3PO3 oxidation, batch electrolysis
experiments were performed. The electrolytes consisted of 0.5 and
5 mmol Lꢀ1 H3PO3 dissolved in 0.5 mol Lꢀ1 H2SO4 aqueous solution
at 25 ꢁC. Electrolysis was performed at a constant potential of 0.1 V.
At this potential H3PO3 is anodically oxidised and PtOx has not yet
formed on the electrode surface. Thus, the passed charge only
corresponds to the electrochemical oxidation of H3PO3. Even
though the electrolyte solution was stirred vigorously, the
measured current density decayed very rapidly and 60 s after

472
M. Prokop et al. / Electrochimica Acta 212 (2016) 465–472
form in detectable amounts during H3PO3 anodic oxidation at a Pt
electrode at ambient temperature in H2SO4 electrolyte solution.
Though the presented results of H3PO3 adsorption and
electrochemical oxidation were obtained at conditions far from
these present in the HT PEM FC environment, they can still shed
some light on the nature of the processes taking place in HTP EM FC
during its operation. It also becomes clear that better understand-
ing the phosphorus compounds behaviour is fundamental for
better understanding of the processes occuring at the HT PEM FC
during its operation.
[12] N. Sugishima, J.T. Hinatsu, F.R. Foulkes, Phosphorous acid impurities in
phosphoric acid fuel cell electrolytes: II. Effects on the oxygen reduction
reaction at platinum electrodes, Journal of The Electrochemical Society 141
(1994) 3332–3335.
[13] W.H. Doh, L. Gregoratti, M. Amati, S. Zafeiratos, Y.T. Law, S.G. Neophytides, A.
Orfanidi, M. Kiskinova, E.R. Savinova, Scanning photoelectron microscopy
study of the Pt/phosphoric-acid-imbibed membrane interface under
polarization, ChemElectroChem 1 (2014) 180–186.
[14] J. Bravacos, M. Bonnemay, E. Levart, A.A. Pilla, Contribution á l'étude du
systéme électrochimique platine-acide phosphorique, C.R. Acad. Sc. Paris 265
(1967) 336–340.
[15] S. Trasatti, A. Alberti, Ossidazione anodica degli acidi ipofosforoso e fosforoso
su palladio e platino, La Ricerca Scientifica 8 (1965) 1012–1025.
[16] Y.V. Babin, A.V. Prisyazhnyuk, Y.A. Ustynyuk, The dynamic structure of cis and
trans conformers of substituted phosphinous acids, Russian Journal of Physical
Chemistry B 2 (2009) 684–689.
Acknowledgement
[17] J.P. Guthrie, Tautomerization equilibria for phosphorous acid and its esters,
free energies of formation of phosphorus and phosphonic acids and their ethyl
esters, and pKa values for ionization of the P-H bond in phosphonic acid and
phosphonic esters, Can. J. Chem. 57 (1979) 236–239.
[18] G. Manca, M. Caporali, A. Ienco, M. Peruzzini, C. Mealli, Electronic aspects of
the phosphine-oxide ! phosphinous acid tautomerism and the assisting role
of transition metal centers, Journal of Organometallic Chemistry 760 (2014)
177–185.
[19] V.M. Baudler, D. Schellenberg, Elektrolytische Untersuchungen von
Phosphorsauren in wasriger Losung, Zeitschrift fur anorganisehe und
allgemeine Chemie 340 (1965) 113–224.
[20] 12—Phosphorus, in: N.N. Greenwood, A. Earnshaw (Eds.) Chemistry of the
Elements (Second Edition), Butterworth-Heinemann Oxford, 1997, pp. 473–
546.
Financial support by the FCH JU within the framework of the
DEMSTACK project, No. 325368, MSMT CR project No. 7HX13002,
and specific university research (MSMT No. 20/2015) is gratefully
acknowledged.
References
[1] J. Andrews, B. Shabani, Re-envisioning the role of hydrogen in a sustainable
energy economy, International Journal of Hydrogen Energy 37 (2012) 1184–
1203.
[2] J. Zhang, Z. Xie, J. Zhang, Y. Tang, C. Song, T. Navessin, Z. Shi, D. Song, H. Wang, D.
P. Wilkinson, Z.-S. Liu, S. Holdcroft, High temperature PEM fuel cells, Journal of
Power Sources 160 (2006) 872–891.
[3] Q. He, X. Yang, W. Chen, S. Mukerjee, B. Koel, S. Chen, Influence of phosphate
anion adsorption on the kinetics of oxygen electroreduction on low index Pt
(hkl) single crystals, Physical chemistry chemical physics: PCCP 12 (2010)
12544–12555.
[4] K.-L. Hsueh, E.R. Gonzales, S. Srinivasan, D.T. Chin, Effects of phosphoric acid
concentration on oxygen reduction kinetics at platinum, J. Electrochem. Soc.
131 (1984) 823–828.
[21] S. Trasatti, A. Alberti, Anodic oxidation mechanism of hypophosphorous and
phosphorous acid on palladium, Journal of Electroanalytical Chemistry 12
(1966) 236–249.
ꢀ
[22] S. Cerný, M. Smutek, F. Buzek, The Calorimetric Heats of Adsorption on
Platinum Films, Journal of Catalysis 38 (1975) 245–256.
[23] W. Dong, V. Ledentu, P. Sautet, A. Eichler, J. Hafner, Hydrogen adsorption on
palladium: a comparative theoretical study of different surfaces, Surface
Science 411 (1998) 123–136.
[24] H. Remy, H. Falius, Uber das Verhalten der Unterphosphorsaure, Naturwiss 43
[5] S. Kaserer, K.M. Caldwell, D.E. Ramaker, C. Roth, Analyzing the influence of
H3PO4 as catalyst poison in high temperature PEM fuel cells using in-
operando X-ray absorption spectroscopy, The Journal of Physical Chemistry C
117 (2013) 6210–6217.
[6] F.C. Nart, T. Iwasita, On the adsorption of H2PO4- and H3PO4 on platinum: An
is situ FT-ir study, Electrochimica acta 37 (1992) 385–391.
[7] W.M. Vogel, J.M. Baris, Changes in the surface of platinum in hot concentrated
phosphoric acid at low potentials, Electrochimica Acta 23 (1977) 463–466.
[8] R. Jasinski, J. Huff, S. Tomster, L. Swette, Electrochemical oxidation of
hydrocarbons, Berichte der Bunsengesellschaft 68 (1964) 400–404.
[9] S.J. Clouser, J.C. Huang, E. Yeager, Temperature dependence of the Tafel slope
for oxygen reduction on platinum in concentrated phosphoric acid, Journal of
applied electrochemistry (1993) 597–605.
[10] N. Sugishima, J.T. Hinatsu, F.R. Foulkes, Phosphorous acid impurities in
phosphoric acid fuel cell electrolytes: I. Voltammetric study of impurity
formation, Journal of The Electrochemical Society 141 (1994) 3325–3331.
[11] M. Prokop, T. Bystron, K. Bouzek, Electrochemistry of Phosphorous and
Hypophosphorous Acid on a Pt electrode, Electrochimica Acta 160 (2015) 214–
218.
(1956) 1.
[25] Q.-S. Chen, J. Solla-Gullón, S.-G. Sun, J.M. Feliu, The potential of zero total
charge of Pt nanoparticles and polycrystalline electrodes with different
surface structure: The role of anion adsorption in fundamental
electrocatalysis, Electrochimica Acta 55 (2010) 7982–7994.
[26] M.M. Ghoneim, S.J. Clouser, E. Yeager, Oxygen Reduction Kinetics in Deuterated
Phosphoric Acid, J. Electrochem. Soc. 132 (1985) 1160–1162.
[27] A. Holewinski, S. Linic, Elementary Mechanisms in Electrocatalysis: Revisiting
the ORR Tafel Slope, Journal of the Electrochemical Society 159 (2012) H864–
H870.
[28] A.J. Bard, L.R. Faulkner, Electrochemical methods: Fundamentals and
applications, 2nd ed., John Wiley & Sons, USA, 2001.
[29] Y. Furuya, T. Mashio, A. Ohma, M. Tian, F. Kaveh, D. Beauchemin, G. Jerkiewicz,
Influence of Electrolyte Composition and pH on Platinum Electrochemical and/
or Chemical Dissolution in Aqueous Acidic Media, ACS Catalysis 5 (2015) 2605–
2614.