Search

Increasing interest in spectroelectrochemisty due to its versatility and far-reaching applications has led to its rapid growth in recent years. Attenuated total reflectance (ATR), Fourier transform infrared spectroscopy (FTIR) is one technique that can provide valuable insights into surface interactions provided a viable working electrode can be fabricated. Though ATR working electrodes can be deposited using a range of techniques such as argon sputtering, vapor deposition and electrodeposition, published studies typically do not report electrode thicknesses which are a key in obtaining repeatable signal responses. Here we examine different approaches to quantify ATR electrode thicknesses and screen for optimal electrode configurations. The test system used to assess different approaches for validating electrode performance was the ferri-ferrocyanide redox reaction with an argon sputtered platinum surface. It was found that atomic force microscopy, ellipsometry and optical profilometry had practical limitations operating in the nanometer range and were not adequate for predicting spectroscopic outcomes. In contrast, electrochemical impedance spectroscopy (EIS) measurements accurately predicted high performing ATR working electrodes. For an 11 mm by 9 mm ATR wafer, favorable signal strengths corresponded to cell impedances greater than 500 Ω. Cell constant calculations determined that platinum thickness layers needed to be less than 4 nm to provide consistent spectroelectrochemical responses.

Greenhouse gas emissions produced during electricity generation can be reduced by increasing the energy efficiency of power generation systems. One approach to increase generation efficiency is by extracting additional electrical energy from low-grade waste heat (< 100 °C) using thermally regenerative redox flow batteries (TRABs). TRAB chemistries with the largest equilibrium potentials use two different active metals and are referred to as bimetallic TRABs (B-TRAB). In most designs, copper is the positive electrode and zinc is the negative electrode. Using thermal energy, ammonia is moved between battery electrolytes to favorably modify the cell potential of the two electrochemical steps within the cycle. A typical B-TRAB system has four steps to complete a full cycle: 1) high-voltage electrochemical discharge, 2) thermal ammonia separation, 3) low-voltage electrochemical charge, and 4) thermal ammonia separation. During step 1, copper is reduced at the positive electrode and zinc is oxidized at the negative electrode creating aqueous zinc-ammonia complexes. Next, ammonia is separated from the zinc complexes and introduced to the copper electrode electrolyte, thereby increasing the equilibrium potential of the negative electrode, and decreasing the potential of the positive electrode. This creates a smaller cell potential required for electrochemical charging during step 3. After charging, ammonia is thermally separated from the electrolyte and re-introduced to the zinc-containing electrolyte returning the system to step 1. Currently, B-TRAB designs use solid zinc and copper electrodes which lead to low coulombic efficiencies at the copper electrode. Here, we demonstrate how the B-TRAB system can be improved by the addition of ammonium bromide to the positive electrolyte stabilizing Cu(I) species and replacing copper deposition/dissolution reaction with the Cu(I, II) reaction on a carbon felt. As a result, the open circuit potential during charging can be as small as 0.8 V, and as large as 2 V for the discharge step. Polarization curves showed that peak power for the discharge cycle was 98 mW cm-2, a significant improvement over previous B-TRABs.

Efforts to reduce the cost of long-duration energy storage systems have drawn increased interest in all-iron redox flow battery systems due to their inexpensive electrolytes. Previous studies have shown that the kinetics of the positive electrode for the Fe(II, III) reaction in these systems are sluggish on carbon surfaces and require significant overpotentials. Here, we identify how changes to the electrode surface and electrolyte composition can influence the electrochemical rate constants to reduce required overpotentials using a rotating disc electrode (RDE) assembly. The impacts of solution composition in terms of pH, chloride content, and cation species were investigated using a range of electrolyte compositions reported for all-iron flow battery systems. The electrode materials studied were glassy carbon, pyrolytic graphite basal plane, and pyrolytic graphite edge plane. Untreated and electrochemically oxidized electrode materials were quantified using X-ray photoelectron spectroscopy (XPS), and the reaction rate constants were quantified using electron impedance spectroscopy (EIS). Rate constants of the Fe(II, III) reaction on untreated electrode materials were largest on platinum and lowest on glassy carbon with pyrolytic materials providing comparable results. For the materials examined, increases electrolyte chloride content generally decreased the obtained rate constant.

Redox flow batteries are emerging as a promising method to provide grid-scale power and long-duration energy storage safely and economically. The thermally regenerative ammonia battery (TRAB) is a new redox flow battery category that can be recharged using low-grade waste heat rather than electric energy, adding further flexibility to the applicability of flow battery systems. Recently, a new TRAB with all-aqueous electroactive species (referred to as the Cuaq-TRAB), as opposed to deposition-dissolution reactions, was found to have superior energy and power densities relative to competing TRABs. The use of bromide and ammonia stabilizes for both Cu(I) and Cu(II) oxidation states, while also creating a cell potential of up to 1.0 V. The potential can be recovered by thermally separating ammonia from the electrolyte and adding it back to alternate electrolyte chambers in successive cycles. Potential improvements in overall cell performance are possible by reducing ohmic losses associated with membrane selection. However, the ideal membrane for the Cuaq-TRAB is not obvious and raises interesting transport questions relative to the dominant redox-active copper species being negatively charged in the catholyte but positively charged in the anolyte. Furthermore, membrane crossover of ammonia, a small, uncharged molecule, was previously shown to be a primary source of parasitic losses for TRABs. Therefore, we investigated how different membrane types (cation, anion, and non-selective) affected ion transport and TRAB performance. A batch symmetry cell was used to determine membrane conductivity in the Cuaq-TRAB environment and membrane diffusion coefficients of each species in the electrolytes. Flow cell experiments were also conducted to find peak power, energy density, average power during discharge, and capacity fade over successive electric charge/discharge cycles. Tradeoffs between membrane conductivity and permeability observed in the symmetry cell are manifested in flow cell results as either high peak and average power or high energy density and low capacity fade. These results expand on potential methods for controlling transport in redox flow batteries and demonstrate the potential for non-selective membranes for electrochemical energy technologies.

Reference electrodes (RE) are essential to many electrochemical measurements and monitoring technologies. True reference electrodes allow measurement of pH, oxidation-reduction potentials, corrosion potentials, and more. The primary role of these REs is to maintain a stable and predictable potential while withstanding an application environment. When these goals cannot be maintained for a suitable amount of time due to complications like contamination or degradation, a pseudo-reference electrode is often used (e.g., a platinum wire) but diminishes the value of the collected electrochemical data (i.e., potential measurements are no longer traceable to the standard hydrogen scale.) To limit the use of pseudo-reference electrodes, new RE designs are being produced to be more compatible with specific environments, to be more cost-effective, or to exhibit better potential stability for a longer time. Additionally, the operational lifetime/stability of the RE should reflect its application: a small, disposable reference electrode may only need to be stable for a few hours; a reference electrode for daily lab use may need maintenance once every other week; a reference electrode used in a remote location may need to be stable for months or many years without servicing. RE lifetimes are typically determined a posteriori (from observation) and not predicted at the design phase, which can make the process of designing a new RE very inefficient. In this paper, we will discuss key design considerations that impact the stability and lifetime for true reference electrodes.

The vanadium redox flow battery (VRFB) is a promising candidate for use in grid-scale battery applications to support a deeper reliance on intermittent renewable energy. However, it suffers from a sluggish redox reaction at the positive electrode. Previous studies indicate that its performance can be improved by oxidizing a graphene-based carbon electrode and thus increasing the concentration of its oxygen-containing surface sites. While such performance enhancement does not appear to be controversial, the underlying mechanisms is far from being clear. Some authors suggest that this procedure increases the number of electrocatalytically active sites; others simply attribute it to improved wettability. In this ongoing study, we use density functional theory (DFT) to examine the role of hydroxyl and carboxyl groups, as well as free edge carbon atoms, in the oxygen transfer process at the positive electrode. Preliminary results using carefully selected model compounds (with appropriate charge C and spin multiplicity M) indicate that oxygen functional groups are unlikely to be active sites in the electrochemical redox process, which in many respects is analogous to the oxidation reduction reaction (ORR). Both the thermodynamics (less favorable adsorption energy of vanadium oxide) and the kinetics (problematic transition states) are less plausible than in analogous mechanistic steps involving carbene-type active sites. The carbene site was probed through two possible reaction mechanisms: V-down and O-down (see graph below). The former adsorption step is essentially a ‘dead-end’ process. The latter is not only thermodynamically and kinetically more feasible; it is consistent with other oxygen-mediated electron transfer processes occurring on the surface of graphene-based materials, be they catalysts or reactants. Furthermore, it has been shown in such processes that the presence of oxygen functional groups can stabilize and enhance the concentration of free carbene-type sites at graphene edges. Figure 1

A significant amount of the potential energy that is generated during energy harvesting worldwide is discarded as waste heat because of inefficient power generation cycles. Much of this wasted power source goes unused because it is trapped as low-grade thermal energy (< 100 °C), which traditional power cycles cannot viably harness. With the advent of electrochemical power systems such as redox flow batteries and fuel cells, researchers are investigating new methods of providing usable electric power from these unused low-grade thermal energy sources. The thermally regenerative ammonia battery (TRAB) is one promising technology in this space, as it operates with the same principles as a flow battery except that TRABs can be recharged using low-grade waste heat instead of electric power. Of the TRAB chemistries proposed, the all-aqueous copper TRAB chemistry is capable of both large power densities and energy densities, and high coulombic efficiencies. In this presentation, we discuss how key performance metrics relevant to power generation from low grade thermal energy sources are strongly influenced by electrolyte composition and battery operating parameters. Manipulating the ammonia to copper ratio demonstrated clear tradeoffs between achievable energy and power capacities. Increasing ligand concentration had a large impact on electroactive species solubility and cell potential differences, which increased the theoretical energy density limit of the battery. Increasing the amount of ammonia relative to dissolved copper raised peak power density, but adversely affected energy density. Moderate increases in discharge current density did not decrease the energy density due to reduced impact of ammonia crossover which appears to be a dominant source of energy loss in TRABs.

As the need for a sharp reduction in carbon emissions continues to grow, many researchers are looking to electrochemical technologies to provide sustainable power to the grid. One promising technology is the silver thermally regenerative ammonia battery (Ag-TRAB), which is a new approach for converting low-grade waste heat ( A 2D numerical model was created using COMSOL Multiphysics to study the relationship between fluid flow, electrode structure, and electrodeposition in the porous electrodes using cylinders in cross-flow to represent the internal structure of a porous carbon fiber electrode. Lower void fractions resulted in 2.5% higher power density, but less uniform deposition and the pores clogging 3-5 times faster due to nonuniform deposition normal to the membrane boundary. It was found that the electric field resultant from the void fraction and fiber arrangement was critical to the performance of the electrode. The electrode microstructure can be modified to better distribute the deposition rate relative to the pore size to allow for higher power output for a longer discharge time before the pores begin to clog. An electrode with variable void fraction increased the peak power by 7.5% but the pores clogged 23% faster compared to homogenous void fraction.

Deep cuts to global carbon emissions are only possible with a dramatic increase in energy storage capacity. However, the cost of the storage options available hinders the timely deployment of the necessary energy storage. Promising supplements to conventional storage are thermally regenerative ammonia batteries (TRABs), which have an additional advantage of being able to capture low-grade waste heat into dispatchable electricity for the grid. This new battery technology can potentially provide power on demand using low-cost chemistries and waste energy sources relative to their purely electric battery counterparts. Most of the research to date for TRABs has been proof-of-concept in nature. As such, little is known about how the electrolytes used in these devices impact their theoretical energy storage capacity. Here we present the general principles of TRABs from an electrochemical and aqueous chemistry perspective. Energy storage density values and Eh-pH diagrams are presented to highlight key factors that govern these complex but versatile energy storage devices. Ag-based and Cu-based TRABs are discussed, and some experimental data confirming the conclusions of the Eh-pH diagrams are presented.

Internal corrosion monitoring programs are essential for the safe management of natural gas transmission pipelines. Electrochemical technologies can provide reliable, real-time monitoring of key risk factors, such as solution conductivity, corrosion rates, and localized corrosion risk. A novel, membrane-based sensor is being developed to perform electrochemical tests in both aqueous and gaseous environments, allowing for continuous operation and data collection within a pipeline environment. This sensor can be used to determine water content in the presence of a number of gases (CH4, H2S, CO2), to monitor the pipeline’s corrosion rate with and without a condensed phase, and to provide localized corrosion data in the presence of a pitting agent. Here, we explore the sensor’s design, validation through finite-element analysis, and performance data for general and localized forms of corrosion.

As the need for a sharp reduction in carbon emissions continues to grow, many researchers are looking to electrochemical technologies to provide sustainable power to the grid. One promising technology is the silver thermally regenerative ammonia battery (Ag-TRAB), which takes advantage of low-grade waste heat (

One aspect of effective steam/water cycle chemical treatment programs is their ability minimize corrosion and corrosion related failures. In this study, we examine how corrosion product layers on the surface of tubing samples impact corrosion rate for a series of test specimens prepared from boiler/steam piping provided from operating power plants. Corrosion rates were estimated using the following electrochemical techniques: linear sweep voltammetry (LSV), linear polarization resistance (LPR) and electrochemical impedance spectroscopy (EIS). Two sources of Tafel slopes were the standard assumptions used if the corrosion mechanism is not known and those obtained from LSV curves. In total, five variations of corrosion rate measurements were compared. For these corrosion tests, water chemistries representative of conditions that induced online hydrogen damage and offline pitting were used. Field emission scanning electron microscopy (FESEM) and energy dispersive X-ray spectroscopy (EDS/EDX) were the electron microscopy techniques applied to examine the corrosion product layers. Using these techniques, the microstructure and elemental composition of the tubing sample corrosion product layers were examined. These techniques could be utilized to evaluate the effectiveness or reduction in corrosion rates with the application of film forming products.

For meeting ambitious carbon-free electricity goals, low-cost electrical energy storage solutions are needed. Although redox flow batteries are a promising option to satisfy this need, the cost of these devices must decrease before widespread adoption can take place. As the electrolytes are the most significant expense for MW-scale flow batteries, new chemistries derived from inexpensive metals is one approach that may alleviate this bottleneck towards a carbon-free power grid. Here we use an electrochemical approach by applying a rotating disc electrode (RDE) to quantify how ligand chemistries can favorably manipulate the electrochemical parameters of the Cu(I/II) redox reaction, progressing these chemistries forward as potential low-cost electrolytes for redox flow batteries. In this research, multiple ligands were used to analyze such impacts on aqueous copper metal complexes by using both a platinum and carbon tipped electrode installed into the RDE. Electrochemical Impedance Spectroscopy (EIS), Linear Sweep Voltammetry (LSV) and Open Circuit Potential (OCP) were the three techniques utilized to analyze the effects of these metal complexes. The experimental data was analyzed using the Butler-Volmer and Nernst equations. As step reactions are mostly unknown for these reactions, LSV and EIS data gave insights into what may be occurring at the electrode interfaces. OCP data quantified how different ligands modified the standard electrode potentials of the Cu(I/II) redox reaction.

Monitoring the internal corrosion of natural gas transmission lines is required for safe management of natural gas infrastructures. Internal corrosion programs can be strengthened by real-time monitoring of key risk factors such as relative humidity and corrosion rates of the pipeline material. We have developed a membrane-based electrochemical sensor, which functions in gaseous environments, to provide this real-time monitoring capability. The design is based on the sensitivity of the membrane’s conductivity to gaseous water content. We match relative humidity curves to 4-electrode, electrochemical impedance spectroscopy results, and use 3-electrode, linear polarization resistance tests to calculate corrosion rates. Due to the membrane’s atypical current pathways, precise estimation of the sensor’s cell constant was required to interpret the impedance results into relative humidity. We utilize finite-element analysis to obtain our cell constant (14.84 cm-1), and the resulting membrane conductivity values were in good agreement with published data. The sensor was tested in H2S environments to gauge how these results may change with acidic gases that may be present in the natural gas stream.

Water treatment programs are a key factor in steam cycle power station safety. They are one of the first lines of defense against a wide range of corrosion-induced boiler system failures. To further facilitate the development of effective boiler water treatment programs, it is important to fully understand the nuances of the underlying corrosion process. Though waterside corrosion of tubing in steam cycle power stations is an old problem, many aspects of the underlying processes are remain poorly defined at elevated temperatures and pressures. These gaps in our current understanding complicate the development and validation of new corrosion mitigation treatments. Here we demonstrate how a series of electrochemical techniques can be used to fill in some of these gaps when coupled with electrochemical theory and a high temperature & pressure condensed phase corrosion sensor. In this study, we investigated the kinetics of carbon steel corrosion in simulated boiler waters at elevated temperatures and pressures. Electrochemical properties (Tafel slopes and corrosion currents) of the corrosion processes were obtained and compared between techniques such as linear polarization resistance, electrochemical impedance spectroscopy and potentiodynamic scans. These comparisons demonstrated the power of using multiple techniques to better understand the behavior of elevated temperature corrosion. Furthermore they can serve as the benchmarks needed to verify the effectiveness of new water treatment programs.

Electrochemical frequency modulation (EFM), a non-linear electrochemical technique was applied for the first time to solid oxide fuel cells (SOFC) to obtain effective exchange current density and transfer coefficient values. EFM has been used extensively in corrosion studies and this study investigates the feasibility of the technique in determining kinetic parameters in a SOFC. The study was carried out on a sub-commercial SOFC stack at 750 °C running on 3 % humidified H2. Values obtained from EFM data were compared to values obtained from electrochemical impedance spectroscopy (EIS) and linear sweep voltammetry (LSV) at similar operating conditions. A kinetic model based on generalized Butler-Volmer equation was used as the basis for the analysis. The results indicated that the values obtained via EFM were in agreement with EIS and Tafel analyses. The obtained values for a cell within a SOFC stack closely ranged from 11 - 16 mA cm-2 for effective exchange current density and 0.88 – 1.3 for transfer coefficient with good consistency. The values obtained from EFM were also further validated using theoretically developed equations for the specific operating condition. Causality factors, an internal check in EFM which relates the input potential to the output current, was also used to verify the validity of the data obtained by EFM.

A study has been carried out to investigate sweet and sour corrosion at the cement-casing interface for offshore oil and gas wells in cold climate. The test solution chemistry was determined by sampling pore water from cement exposed in an autoclave system to brine and CO2 at 100 °C and 10 MPa. The extracted pore water was used to define the composition of a cement simulated pore solution (CSPS), which was then used for the ensuing corrosion tests. Samples made of high strength casing steel, grade Q125, were exposed to the CSPS at 4 °C and 10 MPa in CO2-only and two mixed CO2-H2S environments. The corrosion rate measured in-situ with linear polarization resistance and electrochemical impedance spectroscopy was generally found to range between 10 and 20 μm y-1 after 60 hours of exposure, with the average corrosion rate decreasing slightly with increasing H2S content. The corrosion potential, measured versus a custom high pressure Ag/AgCl electrode, was also found to shift slightly in the negative direction as the H2S content increased. The corrosion products were determined using scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDS). The corrosion products had the chemical compositions close to iron carbonate and iron oxide. Sulfur was detected in the inner film region when H2S was present. These results were compared to the corrosion rates and Pourbaix diagrams modeled using commercial software. The reaction mechanisms are discussed in the paper based on the potentiodynamic polarization measurements and electrochemical impedance spectroscopy taking into account the modeled solution speciation of the CSPS and corrosion product film.

Under the high pressure high temperature (HPHT) conditions in boilers, iron oxides, like magnetite and hematite, are the most common corrosion products. Apart from just degrading the base metal in the feedwater system, corrosion products can also be transported downstream, where they can potentially deposit on surfaces in the boiler. Deposition of magnetite on orifice plates in lower headers of forced circulation boilers leads to improper flow patterns, poor heat transfer and in some cases lead to tube failures. Field data suggests that the boiler operating conditions, like temperature, pressure and boiler water chemistry could influence the deposition of magnetite from the aqueous stream on to the metal surface. It has been experimentally shown previously that boiler water conditioned to pH25 ⁰C 9.3 using NH4OH, favors the deposition of suspended magnetite nanoparticles on 304 stainless steel surfaces at 300 ⁰C and 100 bar. As an extension, the work presented here, explores the effects of additional chemistries on the deposition process and behavior of suspended magnetite particles. This study looks at the influence of all-volatile treatment (AVT) regime with the addition of solid alkali to the boiler water. The aim is to understand the influence of the phosphate treatment-addition of sodium triphosphate-and caustic treatment-addition of sodium hydroxide- on the deposition process with respect to the conventional AVT regime. In addition, the study also investigates the influence of solid alkali addition on the suspended magnetite and hence the nature of the deposits formed on the stainless steel substrate. The boiler environment was simulated in an autoclave pressure vessel. The test solutions were prepared in de-aerated conditions to closely match the boiler feedwater in terms of chemistry. The concentrations of sodium triphosphate and sodium hydroxide used were decided based on the standard operating guidelines followed in the industry. A standard three electrode assembly was used to monitor the deposition process using Electrochemical Impedance Spectroscopy (EIS). The nature of the deposits were investigated using surface characterization techniques. The results showed a clear influence of the zeta potential and double layer compression on the behavior of the suspended particles and their affinity to deposit at 300 ⁰C and 100 bar. The results will add to the existing work on understanding magnetite deposition and suspended particle behavior in high pressure high temperature environments.

Suspended magnetite [Iron (II,III) Oxide] particles in boiler water deposit on the walls and orifice plates in boilers, impeding flow and causing inefficiencies. Operating data suggests that reducing conditions and boiler water chemistries in the boiler enhances development of magnetite deposits. The inter particle interaction due to the zeta potential developed on the colloidal magnetite particles could also be a factor in deposition. In order to study the mechanism behind the deposition of suspended magnetite on stainless steel a high temperature high pressure electrophoretic deposition cell has been developed. An autoclave pressure vessel with graphite O-ring gaskets is serve as the electrophoretic deposition cell simulating the boiler conditions (300 °C and 9 MPa). Carefully prepared test solution imitating boiler water chemistries containing suspended magnetite particles are pumped in to the autoclave. The test solution is prepared in an anaerobic environment to prevent oxygen contamination. The tests simulate the boiler conditions in terms of temperature, pressure and boiler water chemistry. The experimental setup has been designed to allow the sampling of test solution to monitor the pH, dissolved oxygen level and solution potential. 304 Stainless Steel samples were placed in the autoclave to serve as the substrate for deposition. A three-electrode assembly, consisting of 304 SS substrate (working electrode), a counter electrode and a quasi-reference electrode, has been used to characterize the deposition process using in-situ Electrochemical Impedance Spectroscopy measurements. The goal is to understand the influence of boiler water chemistries on the surface charge properties of the suspension. The surface properties can be related to the stability of suspended matter, which in turn relates to the deposition of particles on the substrate. The growth of the deposits is characterized using in-situ Electrochemical Impedance Spectroscopy measurements. Post-test analysis of the steel sample was carried out using Scanning Electron Microscopy (SEM) techniques. Preliminary tests have been carried out with a test solution pH of about 9.4, controlled by adding NH4OH (All Volatile Treatment). Deposition was observed on all samples from all the tests. In-situ EIS measurements have been recorded at test conditions of 300 °C and 9 MPa over the duration of the test. The results will be significant in understanding the inter-particle interactions and electrophoretic deposition at high temperature-high pressure conditions.

Recent developments in membrane and electrocatalyst characterization techniques for materials used in the CuCl(aq)/HCl(aq) electrolytic cell have significantly increased the amount of experimental data available in literature. Rotating disc electrode (RDE) and membrane conductivity cells capable of operating in highly concentrated HCl(aq) were developed to characterize the material performance of key MEA materials for this electrolytic cell. Specifically, exchange current density and transfer coefficient values for platinum and carbon as the electrocatalyst material for the anode reaction of this electrolytic cell were quantified in highly concentrated HCl(aq)1. Additionally, conductivity and permeability parameters of hot-pressed Nafion 117 were characterized for applications in highly concentrated HCl(aq) and CuCl(aq) solutions2. Though these fundamental properties are now available, there remained a need for a simple model to connect these properties appropriately to overall cell performance. While equilibrium thermodynamics of this electrolytic cell was carefully studied3, the cell electrode kinetics and transport in the membrane have not been modeled yet. Principles of non-equilibrium thermodynamics have the potential to incorporate membrane transport properties and electrochemical kinetic data together into a simple but useful model of this electrolytic cell. The use of linear phenomenological equations with phenomenological coefficients can be used to model the effects of permeability and conductivity on the ohmic resistance of the electrolytic cell. In addition to linear non-equilibrium processes, non-linear nonequilibrium thermodynamic expressions of the electrochemical reactions can be used to model the effects of electrocatalyst properties on the overall cell performance. As a result, suitable electrocatalysts and membrane materials were identified and their effects on the electrolytic cell performance were quantified in the form of a non-equilibrium thermodynamic model. References: 1. D. M. Hall, E. G. LaRow, R. S. Schatz, J. R. Beck, and S. N. Lvov, J. Electrochem. Soc., 162, F108–F114 (2015). 2. R. Schatz, S. Kim, S. Khurana, M. Fedkin, and S. N. Lvov, ECS Trans., 50, 153–164 (2013). 3. D. M. Hall, N.N. Akinfiev, E. G. LaRow, R. S. Schatz, and S. N. Lvov, Electrochim. Acta, 143, 70-82 (2014).