SPINEL-TYPE COBALT OXIDE THIN FILM ELECTRODES IN ALKALINE MEDIUM

For preparing spinel-type cobalt oxide thin film electrodes in alkaline medium, an important problem is the development of highly active, stable, and inexpensive electrocatalysts. Spinel-type cobalt oxide films have attracted much attention due to their excellent electrocatalytic properties with respect to the OER and good chemical stability in an alkaline solution. Due to their low cost, good conductivity, and electrocatalytic activity for OER, cobalt spinel oxides (e.g., C03O4) have been the target of several studies (Hamdani et al., 2004). The electrochemical behavior of thin C03O4 spinel films for the OER in 1 M KOH on stainless steel substrates using the thermal decomposition method at 400°C was studied by cyclic voltammetry and impedance methods (Laouini et al., 2008). The impedance measurements were carried out at various positive potentials, from the open circuit potential to the potential in the OER region. The value of the roughness coefficient, determined in the potential region where the charge transfer reaction is negligible, is similar to the value obtained by cyclic voltammetry. The importance of the oxygen evolution reaction stems from the fact that it is an anodic reaction in key processes such as aqueous electrolysis, electrochemical metal extraction, and organic electrosynthesis. However, due to the high overvoltage, electrolysis of alkaline water usually shows low energy efficiency. Efforts are being made to improve reaction kinetics and reduce overvoltage.

In recent years, the use of EIS has been expanded to characterize the electrochemical and electrocatalytic surface properties of oxide electrodes toward OER in order to understand solid state redox transitions. Palmas et al. (2007) have investigated the sol-gel behavior of prepared C03O4 powder electrodes using cyclic voltammetry and EIS to gain access to solid-state surface redox transitions, which lead to the formation of active sites on the surface electrode for OER in alkaline solutions. These authors confirmed that the kinetic parameters - the exchange current density, and the electron transfer coefficient obtained from the EIS data are in good agreement with the data obtained in polarization n experiments. Castro, Real, and Pinheiro Dick (2004) studied porous nickel and cobalt oxides obtained by cathodic electrodeposition using approximation in terms of a finite conical pore transmission model connected in parallel.

GRAPHENE OXIDE SOLUTIONS

Graphene has been extensively studied for various applications due to its excellent electrical, mechanical, thermal, and optical properties (Neto, Guinea, Peres, Novoselov, & Geim, 2009). Alternatively, graphene oxide (GO) dispersed in a solvent could be a promising method for mass production of wafers in any size at low cost (Zhu et al., 2010). GO can be easily applied to the target substrate using centrifugation and/or spraying methods as a preliminary cursor, and then converted to graphene by thermal, chemical, or photocatalytic reduction (Stanko- vich et al., 2007). Meanwhile, GO performs its own functions with a hydroxyl/epoxy group on the basal plane and a carbonyl/carboxyl group on the edge, which is not like the final graphene (Geim, & Novoselov, 2007).

Thanks to such oxygen-containing functional groups, GO exhibits excellent chemical sensory properties and can be uniformly and stably dispersed in deionized (Dl) water. In Yoon et al. (2017), the optical and electrical characteristics of a solution of GO (GS) with different concentrations of GO in Dl water (see Table 3.4) were studied by EIS. GO has become very conductive in Dl water. By oxidizing graphite and dispersing it in Dl water, the authors formed GS for GO concentration ranges (from 0.25 to 7.00 g/L). The transmittance GS becomes completely opaque in the visible range from 300 to 700 nm at GO concentrations exceeding 2 g/L. The measurement results obtained by EIS for GS are presented by the Bode and Nyquist plots in the frequency range from 1 kHz to 10 MHz. In detail, EIS was performed, which was known to be a suitable method for studying the electrical properties of liquid materials. The observed experimental results were correlated with the simulation of an equivalent circuit for different GSs, which made it possible to extract physical parameters for explaining and interpreting the current flow mechanism of various GO concentrations in GS samples.

Based on the impedance analysis of Figure 3.27, authors proposed an equivalent GS circuit model, as shown in Figure 3.28. At first, the circuit model was developed for a stacked three- dimensional (3D) GO with an inductor (Lc,o), resistor (Rco), and two CPEs: (£?goi and Qgoi)-> as shown in Figure 3.28. Generally, when circuits are not expressed as simple RC circuits, the CPE can be introduced with a frequency-independent (9-value, an imperfective resistive capacitance, and index a (0 < a < 1; a = 0 for a pure resistor and a = 1 for an ideal capacitor). The stacked 3D graphite showed an inductive conductor property, while the fully oxidized GO stack showed highly resistive properties consisting of a large resistor and a CPE pair with a long phase delay (shown in Figure 3.28a). Once the oxidation was sufficiently completed, Qooi became similar to an ideal capacitor with an a of ~1. Although GOs were scattered in Dl water in this work from the EIS measurements shown in Figure 3.28b, both circuit diagrams can be similar in principle. In circuit models of dielectric water, even a very small

TABLE 3.4

Various GO concentration in GS samples

Low-GO Samples

GS1

GS2

GS3

GS4

GSS

GS6

Concentration (g/L)

0.25

0.49

0.73

0.96

1.19

2.24

I ligh-GO Samples

GS12

GSI3

GS14

GS15

GS16

GS17

Concentration (g/L)

5.08

5.24

5.40

5.56

6.10

6.42

Reproduced from Yoon et al. (2017) under the Creative Commons Attribution License

Nyquist plot of the GS samples

FIGURE 3.27 Nyquist plot of the GS samples: (a) for GS1 to GS11, which have low GO concentrations and (b) for GS12 to GS20, which have high GO concentrations.

Reproduced from Yoon ct al. (2017) under the Creative Commons Attribution License.

Equivalent circuit models

FIGURE 3.28 Equivalent circuit models: (a) stacked 3D GOIO; (b) DI, and (c) GS inductor in series. Reproduced from Yoon ct al. (2017) under the Creative Commons Attribution License.

amount of water can contribute to a current conducting current.. Considering GO in GS as a component of a conducting network and Dl water as a matrix material, it can be assumed that the model of the equivalent GS circuit may be similar to the GO model. However, one of the CPE GS elements (Figure 3.28a) was closer to the ideal capacitor due to the condenser component in the Dl water, as shown in Figure 3.28c.

DENDRONIZED CaSi03-Si02-Si Nanoheterostructures

In recent years, calcium silicate (CaSiOj) has been given an increasing attention for its promising applications based on its good bioactivity, biocompatibility, and biodegradability (Ni, Chang, Chou, & Zhai, 2007; Pan. Thierry, & Leygraf, 1996). At high-enough pressures, it is believed to have been crystallized with a perovskite structure and is, therefore, referred to as Ca-Si-perovskite. At lower pressures, Ca-Si-perovskite is not stable and converts to wollastonite. The tremendous improvements in high-quality film-formation techniques and compositional engineering of perovskite materials over the past 5 years have led to rapid improvements in the power conversion efficiency of perovskite solar cells (Yang et al., 2015). Although solar-to-electric conversion efficiencies of up to 18% have been reported for perovskite solar cells (Jeon et al., 2015), developing technologies further to achieve the efficiencies near theoretical values (>30%) continues to be an important challenge in making the solar cell industry economically competitive. The largest light loss mechanism for the perovskite/silicon solar cell is reflection. The majority of this reflection can be attributed to the sets of layer interfaces (Grant, Catchpole, Weber, & White, 2016), as the large index contrasts between the adjacent layers result in high Fresnel reflection. To increase light absorption within the silicon layer, a scattering surface can be introduced at the interlayer interface. A number of textured surfaces, such as random pyramids and random spherical caps, can be formed by a number of processes (Baker-Finch. McIntosh, & Terry, 2012).

A systematic study of charge carrier relaxation processes was carried out in sonochemi- cally nanostructured silicon wafers, for which IS and transient photovoltage techniques are used (Savkina et al., 2019). Figure 3.29a shows the complex impedance plane plots of the samples (Savkina et al., 2019); they are semicircle shaped. The arrow shows the direction of the increase in frequency. Symbols are the experimental data and solid lines are the best-

(a) Nyquist plots showing the complex impedance associated to the sample #Si-26-06

FIGURE 3.29 (a) Nyquist plots showing the complex impedance associated to the sample #Si-26-06 (triangles) and #Si-24-12 (circles) surfaces; (c) equivalent electrical circuits representing the interface behavior between the CaO-SiCb species and Si wafer for samples #Si-24-12 and #Si-26-06, respectively; (d) and (e) changes in the frequency-dependent impedance  and the phase shift 0, respectively, in samples #Si-24-12 and #Si-26-06.

TABLE 3.5

Parameters of the equivalent circuits

Samples

SI, Ohm

S2, kOhm

S3, kOhm

Cl, F

CPE:

ZCpc = A '(iro) "

т i = R2C1

r2 = R3C1

A

n

Si-26-06

200

2.1

1.2

2.4 x Ю’9

5.7 x Ю'10

0.98

~5 ps

~3ps

Si-24-12

1700

14

_

КГ4

4 x 10~w

0.93

~ l .4 s

_

Reproduced with permission from Savkina et al. (2019). Copyright 2019, Springer.

fit curves to the measured spectra using the modified equivalent circuits of Figures 3.29c and 3.29d,e, which are obtained with the EIS spectrum analyzer (http://www.abc.chernis try.bsu.by/vi/analyser). The fitting was performed using the values of the circuit elements given in Table 3.5.

The conventional equivalent circuits used for samples investigated have series resistance R1 followed by the parallel circuit of CPE1. series R2-CI and the parallel circuit ЛЗ-С1. The resistance Л1 is usually associated with the contact resistance. However, since in this case /?1»1 Ohm, it was believed that R1 combines both the contact and the bulk material resistances.

CPE is used to accommodate the nonideal behavior of the capacitance which may have its origin in the presence of more than one relaxation process with similar relaxation times (Bis- quert & Fabregat-Santiago, 2010). The series R2-C1 and the parallel circuit /ТЗ-С1 (equivalent electrical circuit parameters see Figure 3.29c) corresponds to the charge transport in the space charge region of the sonochemically structured subsurface layer. Frequency dependence of the impedance |Z| and phase shift в is shown in Figure 3.29d and e, respectively. The frequency dispersion is not observed within the range of 10 Hz to 10 kHz (see Figure 3.29d). The experimental data give a cutoff frequency (the frequency that characterizes a boundary between a passband and a stopband) of about 8.5 kHz in #Si-26-06, which corresponds to the time constant of about 0.1 ms. The cutoff frequency of =18 kHz (with the time constant of ~56 ps) is obtained in #Si-24-12. Based on the impedance measurements, it was concluded that the set #2 and #3 samples exhibit a capacitance-type impedance, which can be associated with several charge carrier relaxation processes. At least two of them can be revealed by the two different cutoff frequencies shown in Figure 3.29d.

It is found that interface potential in Si wafers remarkably increases upon their exposure to sonochemical treatments in Ca-rich environments. In contrast, the density of fast interface electron states remains almost unchanged. It is found that the initial photovoltage decay, taken before ultrasonic treatments, exhibits the involvement of shorter- and longer-time recombination and trapping centers. The decay speeds up remarkably due to cavitation treatments, which is accompanied by a substantial quenching of the photovoltage magnitude. It is also found that, before processing, the photovoltage is noticeably inhomogeneous over the surface of the plate, which implies the presence of distributed areas that affect the distribution of photoexcited carriers. Treatments cause a general expansion of the distribution of photovoltaic voltage. In addition, it was found that samples of sonochemically structured silicon are characterized by a capacitive-type impedance and demonstrate more than one process of relaxation of charge carriers after treatment in a cryoreactor and annealing after treatment with ultrasound. Sonochemical nanostructuring of silicon wafers with dendronized CaSiCb can provide a new promising way to create inexpensive multilayer solar cell structures with efficient use of solar energy.

 
Source
< Prev   CONTENTS   Source   Next >