Abstract
Bilayer coatings consisting of a 
inner layer and a 
outer layer were made on copper substrates 
and the photoelectrochemical properties of the derived bilayered photoelectrodes for corrosion protection application was investigated and compared with the performance of our previously studied 
bilayered photoelectrodes coating for similar applications. 
and 
sols were used for preparing the coatings. Investigations on the charge storage capacity of the two different bilayered coatings revealed that the 
system is capable of storing higher charge at less negative applied potentials, i.e., 
vs SCE, whereas, the 
system stores higher charge at more negative applied potentials, i.e., 
vs SCE.
Export citation and abstract BibTeX RIS
Copper has been proposed as the ideal material for construction of canisters to be used for the disposal of spent nuclear fuels because of its outstanding corrosion resistance, ease of fabrication, abundant availability, and overall cost-effectiveness. For these reasons, extensive research has been pursued for assessment of suitability of copper for these same purposes and, therefore, corrosion protection of copper in the conditions prevailing underground becomes a matter of concern because the passive and protective copper oxide layer does not form when the oxygen potentials are very low. In this context, cathodic protection offered by 
in the presence of ultraviolet (UV) light could be envisaged as appropriate. There have been substantial efforts to exploit the photocatalytic applications of nano 
in all fields that it is predicted to become a part of every individual's life in a few years. One of the most promising applications that has stimulated considerable interest in the recent past is its use as a photoanode for corrosion protection of metal and steel substrates. The working principle of photoaided cathodic protection is that when a metal coated with a thin layer of 
is exposed to UV irradiation, 
-hole pairs are created in the 
layer. The photogenerated electrons can be transferred to the metal substrate thereby making its electrode potential more negative than its corrosion potential. The main advantage of this method is that the photoanode, i.e., 
in this case, does not get consumed during the process of corrosion protection, unlike a sacrificial-type cathodic protection. However, in the case of the applicability of 
coatings for corrosion protection of copper under subterranean conditions, there would be no light available for such a purpose. But this problem is not a serious one as the containers would hold high-level nuclear waste and the gamma radiation being emitted from them could itself be visualized as an in situ source of UV illumination, provided suitable radiation converters are incorporated in the protective 
-based coating. The exact mechanism of corrosion protection offered by a 
coating on metal/steel substrate in presence of UV illumination has been discussed in detail in Refs. 1–4.
One of the inherent disadvantages of using a plain 
coating for anticorrosion applications is that the photoeffect ceases to be operative once the UV illumination stops shining on the 
-coated substrate. A possible method for solving this problem would be to couple another semiconductor like 
whose conduction band (CB) level is lower than that of 
5 so that the photogenerated electrons could be transferred from 
to 
and the coupled semiconductor, i.e., 
, should be capable of storing and releasing the photogenerated electrons. Another candidate semiconductor that could be coupled with 
would be nano 
. There are several reports on the use of 
electrodes for electrochromic windows, where the reversibility for lithium insertion in 
was used for transmissive electrochromic devices.6, 7 In addition, a recent report describes the application of hierarchically mesostructured 
for use in solar cells,8 where the active component is nanometric ceria using a constructed organic-dye-free solar cell. The reason for this is that the bandgap of nanometric ceria was found to have shifted by 
when compared to that of 
. In view of the potential applications of nanoceria, the present investigation was carried out with an objective to measure the photoelectrochemical properties of 
bilayer electrodes on copper substrates and to find out if such electrodes possess the desirable properties of an efficient photoanode for corrosion protection. The results obtained are compared with the performance of the 
electrode for the same application.
Experimental
Synthesis
A commercial 
sol solution (pH 7.0) supplied by Nihon Parkerizing Co., Ltd., Japan, was used as the source of 
. The average crystallite size of 
in the solution was 
. Ceria nanoparticles were synthesized using a low energy wet chemical synthesis technique known as microemulsion, the details of which are given as follows. The microemulsion system consists of surfactant sodium bis(2-ethylhexyl)-sulfosuccinate (AOT), toluene, and Millipore DI water. All the chemicals were purchased from (Aldrich, Inc.). AOT was dissolved in toluene and 
aqueous cerium nitrate solution was added. The mixture was stirred, and hydrogen peroxide was then added dropwise. The detailed synthesis procedure is explained in Ref. 9. The reaction was carried out for 
and then the reaction mixture was allowed to separate into two layers. The upper, pale yellow layer was toluene-containing nonagglomerated ceria nanoparticles and the lower layer was the aqueous phase. The morphology of 
nanoparticles was studied by high resolution transmission electron microscopy (HRTEM) using a Philips HRTEM operated at 
accelerating voltage. Figure 1 shows the HRTEM image of the synthesized cerium oxide nanoparticles.
Figure 1. HRTEM image of nanoceria particles (size 
). (Inset shows the SAED pattern indicating the fluorite structure of nanoceria sample.)
Copper specimens of dimensions 
were used as the substrates. The copper plates were polished to a mirror finish with 
prior to coating and thoroughly degreased in acetone for 
in an ultrasonic cleaner, which were subsequently used as substrates for coating the photoanode material. The photoanode coating consisted of two layers, the first one being 
and the outer layer 
. 
coatings on copper coupons were made using a dip coater, and the coating thickness was calculated using the weight difference of coupons before and after coating. The coating thickness was approximately 
on each sample. 
coatings on the 
layer were made by the spin-coating technique in order to achieve a thickness of 
for the 
layer. The total thickness of the photoanode was measured to be approximately 
. The bilayered photoelectrode coating on the copper substrate was co-fired at 
for 
in a 
atmosphere to obviate oxidation of copper during heating. The heat treated coupns were subsequently subjected to photoelectrochemical characterization.
Photoelectrochemical characterization
Electrochemical characterization was carried out by measuring the open circuit potential (OCP) of the samples under dark and UV illuminated conditions with reference to a saturated calomel electrode (SCE). A deaerated 0.3% NaCl solution was used as the electrolyte unless otherwise mentioned, and a Pt strip was used as the counter electrode for polarization measurements. In order to simulate the conditions prevailing underground, a deaerated atmosphere was created during the electrochemical studies by purging the electrolyte solution with 
(purity, 99.9%). A 
Hg lamp was used as the source of UV illumination. The light first passed through an optical filter (UV-34, Kenko), which mostly 
allowed only light of wavelengths greater than 
. After this light passed through a quartz window, it was then made to fall on the coated side of the copper plate. A transparent electrochemical cell made of acryl fitted with a quartz window on one side was used. The sample was fixed on the opposite side with the aid of an O ring. A more detailed description of the experimental details is given in Ref. 10.
For impedance measurements, the same conditions with respect to atmosphere and electrolyte as used for the OCP measurements were maintained. Impedance was measured as the current response to a superimposed ac signal of amplitude 
with a frequency ranging from 
up to 
at various applied potentials provided by a potentiostat, model 2090 HS supplied by Toho Techniques Research, Japan. After a preliminary analysis of the impedance data obtained for various applied potentials as a function of frequency, it was found that at lower frequencies 
, the system showed a capacitive behavior, and this was ascertained to be arising from the bilayer electrode coating. Hence, for measurement of the capacitance of the coating, the impedance was measured at potentials ranging from 
down to 
in steps of 
at a low frequency of 
. The system was kept for 
at any applied potential before measurement, as it was verified that a constant current state was reached at 
after application of potential.
Results and Discussion
As a first step, the response of the OCP of different samples to UV illumination was studied and their time evolution observed after exposure to UV source and after cutting it off.
Figure 2 shows the time evolution of the OCP on exposure to UV illumination for 
. It can be seen that there is an initial p-type effect (OCPs instantaneously becoming more positive) on exposure to UV light. But after sometime, the electrode potential slowly reverts back to negative values. Nevertheless, the values do not become more negative than 
whereas, in the case of a plain 
coating (also included in Fig. 2), though the same initial p-type effect is observed, after some time, the electrode potentials slowly become more negative and finally reach values of 
, which is the photopotential of 
. The initial p-type effect is due to the presence of a thin layer of copper oxide 
present between the copper and the semiconductor oxide coating, that could have formed due to the interaction of the copper surface with the water-based sol solution during heat-treatment. 
and CuO are both p-type semiconductors with bandgap equal to 2 and 
, respectively, and their presence adversely affects the normal n-type photoeffect in 
as shown in Fig. 2. The copper oxide layer can be reduced by photogenerated electrons from 
,10 but the photoeffect in 
, though capable of reducing the copper oxide layer, cannot make the OCP reach large negative values like 
(cf. Fig. 2). Hence, there is a necessity to include 
in the photoanode containing 
. But since the electrical conductivity of 
is greater than that of 
, the design of a bilayer coating was thought of, where the photoeffect could be generated by an outer 
layer, from which the photogenerated electrons could be passed on to the substrate via an intermediate 
layer, whose electrical conductivity was substantial to facilitate charge transfer. In addition, since 
is also capable of exhibiting multiple valence, it is expected to store charge in the form of photogenerated electrons according to Eq. 1, which will sustain the photoeffect even after cessation of UV illumination, by releasing the stored electrons to the substrate
where 
or Li and 
.
Figure 2. Comparison of the response of electrode potentials to UV illumination for 
and 
.
Figure 3 depicts a comparison of the response of the OCP response on exposure to UV illumination for the 
which is also compared with that of 
(taken from Ref. 10). It can be seen that, though both the profiles are comparable with respect to the absolute values of the photopotentials, the recovery of the electrode potentials after the source of UV illumination has been cut off, differs to a slight extent. The electrode potential after the stopping of illumination is more negative for the bilayer electrode containing 
than that obtained in the case of 
and remains at such values for a longer period of time. This clearly indicates that the charge-storage property of 
is different from that of 
, which result is expected. The charge-storage property could be quantified by measuring the capacitance as a function of applied voltage, which was carried out by measuring the impedance at low frequencies such as 
, with varying applied potentials.
Figure 3. Time evolution of electrode potential and its response to UV illumination compared for 
and 
.
Figure 4 shows the Bode plot for the sample polarized at 
vs SCE. An equivalent circuit corresponding to the configuration given in Fig. 5a was assumed for the analysis of the impedance data. The impedance information can be divided into three distinct regions. 
represents the solution resistance (appears at very high frequencies); 
are those from the high frequency response (due to electrode solution interface); 
are the resistance and capacitance components of the low frequency response (due to the semiconductor oxide coating). The value of each component was determined from the absolute value of the impedance determined at specific frequencies where only one of the components would be a major contributor to the total impedance, neglecting the effect due to others. The calculated data assuming the equivalent circuit fitted well with the measured data and is shown as a solid line in Fig. 4.
Figure 4. Impedance measured as a function of frequency for the 
HT Cu∣bilayered electrode at an applied potential of 
vs SCE; solid line is the curve fitting according to equivalent circuit given in Fig. 5a.
Figure 5. (a) Equivalent circuit proposed conforming to the measured impedance data and (b) reduced circuit at low frequencies.
From the impedance measurements at 
for various applied potentials, the capacitance of the bilayer coating was determined at each applied potential at this frequency assuming the equivalent circuit shown in Fig. 5b and plotted in Fig. 6. Figure 6 shows a comparison of the capacitance for 
, 
, and 
samples measured as a function of the applied electrode potential vs SCE. The 
electrode obviously stores less charge throughout the range of applied potentials of measurement when compared to the other two electrodes. This is due to the fast charge recombination (electron-hole) on the 
surface. However, when one compares the charge-storage property of 
and 
, it is seen that 
exhibits a constant charge, i.e., 
stored over potentials from 
down to 
and then there is a sudden increase in the amount of charge stored for still lower potentials. The large values of capacitance at 
corresponds to the one electron reduction from 
to 
followed by subsequent reduction to 
as indicated by the sudden increase in the capacitance beyond 
. The high capacitance values of 
at 
confirm that when the electrode potential reaches the value of 
during illumination (photopotential), there is a charging up of the electrode. When the source of illumination is cut off, this stored charge is slowly discharged to the substrate such that a highly negative electrode potential of the substrate is still maintained.
Figure 6. Capacitance as a function of applied potential for 
and 
.
In the case of 
there is low charge ranging from 
stored at potentials from 
down to 
and then there is a sudden increase in the stored charge, which is even higher than that stored by the 
electrode at even potentials. This indicates that the reduction of 
to 
takes place at a more negative electrode potential when compared to 
reduction.
It is desirable that the reduction of the multivalent species (
or 
) in the photoanode should take place at its photopotential, so that the photogenerated electrons can be transferred from 
and can be stored simultaneously. In view of this requirement, from the present study, it can be inferred that the 
electrode is preferable as a photoanode when compared to 
. However, the 
electrode could also be made more efficient, if the photopotentials can be made around 
, where an in situ reduction of Ce takes place and the charge can be reversibly stored and released to the substrate, thereby sustaining the photoeffect.
Conclusion
The present study was carried out to evaluate and compare the performance of 
bilayered photoelectrode with 
. The photoelectrochemical properties were characterized by measuring the electrode potential response to UV illumination and the charge-storage capacity by impedance measurements as a function of applied electrode potential. The investigations revealed that 
bilayered photoelectrode showed good charge-storage capacity at the photopotential of 
. Nevertheless, the results also showed that if the photopotential of the electrode could be made 
or lower, 
bilayered photoelectrode is also capable of very high charge storage.
Acknowledgment
R.S. acknowledges the National Institute for Materials Science, Tsukuba, Japan, for financial support in the form of a fellowship.
National Institute for Materials Science assisted in meeting the publication costs of this article.