increased by the deposition of other metals onto the surface of the
Pt cathode by CP and/or EP methods. The most effective modifier
was Cu, which increased the activity of the Pt cathode more than
+
25 times (EP). A very low yield of NH4 is another noticeable
merit of this modification, which is in contrast to the Ni-modified
cathodes. Neither Pd-modified Pt nor Pd alone were active for
nitrate reduction. The modification by Cu was also effective for
Pd; a bimetallic Pd–Cu cathode (9 wt% Cu), which was prepared
by the CP method using a mixed aqueous solution of CuSO4–
PdCl2 exhibited a high activity (k = 19.8 6 1023 min21). When Cu
alone was used as the cathode, however, the activity and stability
were too low to use. The low stability should be associated with the
ease of oxidation of copper.
In the catalytic hydrogenation by H2, it is reported that Pd or Pt
catalysts can reduce nitrite but are inactive for the reduction of
nitrate.10 We thus suspect that the high activity for the Cu-
modified Pt electrode is due to highly-dispersed metallic Cu that is
stabilized on Pt. This is compatible with the X-ray diffraction
pattern of the Cu-modified Pt electrode, which exhibited a single
phase of Pt but the deposition of Cu metal and oxides could not
detected. X-Ray photoelectron spectroscopy showed the presence
of Cu2+, Cu0 and Pt0 on the cathode surface.
Fig. 2 Concentration profiles of NO32, NO22, and NH4 during
The degradation of nitrate did not occur when the applied
voltage was not enough for water electrolysis (,ca. 1.2 V). Once
water electrolysis began, the rate constant, k, increased mono-
tonically with an increase in current and voltage, but the product
selectivity was almost constant. This is quite different from
conventional electrochemical nitrate reduction, where the product
is dependent on the electrolysis potential; reduction at low negative
potentials gives NO22, whereas reduction at high negative
+
application of a DC current (100 mA) (a) without a CO2 supply and (b)
with a CO2 supply to the Cu-modified Pt cathode (CP) at room
temperature.
which is not in accord with the experimental value, 11. Actually,
the cyclic voltammetry measurement exhibited no obvious signals
ascribable to these cathodic processes.
+ 2
These considerations suggest that the reaction is probably due to
2
catalytic hydrogenation of NO3 by H2, which is evolved
potentials gives mainly NH4 .
Fig. 2a exhibits the concentration profiles of NO32, NO22, and
+
electrochemically on the surface of the cathode. To confirm this
point, the catalytic nitrate hydrogenation was carried out by
supplying H2 by bubbling onto the cathode surface without
applying an external potential. The rate of H2 supply,
20 cm3 min21, corresponds to 26 times more than Faradic H2
evolution at the cathodic current of 100 mA. Nevertheless, the
degradation of nitrate was negligible (k , 0.1 6 1023 min21). The
result implies that the reduction of nitrate should basically be
catalytic but promoted significantly by applying a cathodic
potential.
NH4 for the MEA having a Cu-modified Pt cathode under
electrolytic conditions (100 mA). The concentration of nitrate
decreased monotonically, although nitrite and ammonium ions
were formed. About half of the nitrate reacted in the first 90 min
and was mainly converted into nitrite, which would be gradually
consumed to N2 after 180 min. The mass spectrometric analysis of
the gas evolved from a cathode chamber detected N2, but N2O and
NO were negligible. Thus, the reaction seems to take place in a
2
2
consecutive manner, NO3 A NO2 A (NO) A N2. The
formation of more toxic nitrite as an intermediate causes a serious
problem in the catalytic hydrogenation of nitrate. After 180 min of
the reaction, the pH of the aqueous solution in the cathode
chamber increased from 5.5 to 11.0, indicating that the nitrate
reduction yielded OH2 as in the conventional catalytic nitrate
reduction. The change of pH is nearly equivalent to the amount of
OH2 formed by the reaction shown in eqn (1).
To keep the pH constant during the catalytic nitrate
hydrogenation, bubbling CO2 gas into the reaction solution is
known as an useful buffer.11 Fig. 2b displays the result of
electrocatalytic nitrate reduction when CO2 was supplied into the
cathode chamber at a rate of 20 cm3 min21. The pH in this case
became almost constant at ca. 6. Clearly, the nitrate reduction
became about 5-times higher compared to the reaction without a
2
2NO3 + 5H2 A N2 + 2OH2 + 4H2O
2
(1)
CO2 feed (Fig. 2a). Moreover, the amount of NO2 released
intermediately during the reaction course was significantly
suppressed and totally disappeared after 90 min without increasing
One may consider that the following electrochemical reactions
of nitrate are also possible.
+
+
NH4 . The final selectivity to NH4 after 180 min of the reaction
was less than 5%, which is lower than those reported so far for
catalytic and electrocatalytic hydrogenation of nitrate.4–7 The
maximum Faradaic efficiency for the hydrogenation of nitrate to
N2 on the Cu–Pt cathode reached more than 25%, which is
extremely larger than the efficiency reported for conventional
catalytic reduction of nitrate by H2 (¡1%).4–7,12
2
2NO3 + 12H+ + 10e2 A N2 + 6H2O
(2)
(3)
2
2NO3 + 6H2O + 10e2 A N2 + 12OH2
However, eqn (2) cannot explain the pH increase. According to
eqn (3), the final pH of the cathode solution should exceed 13,
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 732–734 | 733
Machida, Masato
