Degradation of graphite electrodes in acidic bromine electrolytes

• Main Author: Bistrika, Alexander A.
• Other Authors: Yokochi, Alexandre
• Language: en_US
• Published: 2013
• Subjects: Redox Flow Batteries; Engineering; Degradation; Cost; Storage batteries > Evaluation; Storage batteries > Testing; Storage batteries > Deterioration; Storage batteries > Economic aspects; Electric batteries > Electrodes > Deterioration
• Online Access: http://hdl.handle.net/1957/37935

As the world's power needs grow, the demand for power from renewable resources, such as wind or solar is increasing. One major drawback associated with these renewable resources is that the power output is dependent on environmental factors, such as cloud cover and wind speeds. This allows the possibility of either power output exceeding or falling short of forecast levels that may lead to grid instabilities. Therefore, Large Scale Energy Storage (LSES) systems are critical to store excess power when the output exceeds demand in order to supplement output power when it falls short of demand.¹ The Zinc/Bromine Redox Flow Battery (RFB) is a promising technology because of previously reported long cycle-life (CL) capability, high efficiencies, low cost materials, and scalable operating conditions.² The excellent energy storage performance of the Zinc/Bromine system was confirmed by measuring both Faradaic and Coulombic electrochemical cell efficiency dependence on temperature of a bench scale Zinc/Bromine flow cell. At room temperature, near 75% Faradaic efficiency was measured when cycling the system between 20% and 100% State of Charge (SOC), which is in good agreement with published values,³ and was measured to be over 80% efficient when operating at an elevated temperature of 50°C. To elucidate capital and operational costs, key system operation parameters especially focused on degradation mechanisms were investigated. Since deep discharge cycling is perceived as highly damaging to electrochemical systems, a system was cycled between 0% and 5% (SOC) 10,000 times. Performance was quantified by measuring the frequency factor (i[subscript 0]) and relative activation energy (α) for the reactions using Tafel scans. No statistically significant degradation or change to the electrodes was observed during the zero point cycling experiment. However, it was found that under conventional operation damage to the electrodes does accumulate, presumably due to the highly oxidative environment caused by the presence of high concentrations of dissolved bromine or tri-bromide. While the performance of both electrodes shows decreases in frequency factor attributed to the damage process, the bromide oxidation process seems to be more damaging (i.e., at the positive electrode during the charging process). Long term measurements show a degradation of the electrocatalytic parameters at an applied overpotential of 100 mV from ca. 40 mA/cm² to ca. 5 mA/cm² at the positive electrode and from ca. 20 mA/cm² to ca. 10 mA/cm² for the negative electrode. A degradation rate model was proposed to predict the service life expectancy of graphite electrodes in a bromine system based on processes showing a combined second order reaction rate coupled with a negative first order reaction rate. The model can be used to predict the cost of energy when operating any device using graphite electrodes, based on the operating power ratio, defined here as the quotient between operating power and system rated power. This damage could be partially reversed by exposing the electrode surfaces to concentrated potassium hydroxide dissolved in isopropanol, presumably due to exfoliation of the electrocatalytic surface leading to the exposure of a clean surface with electrocatalytic performance close to the original. Further, a chemical pretreatment for the graphite surface imparting enhanced stability in aqueous bromine systems was developed that shows negligible damage when similar amounts of current have passed through the electrode surface. After bromide oxidation equivalent to passing ca. 10 Ah/cm² the treated surface showed a change in steady state current density at an applied overpotential of 100 mV from ca. 50 mA/cm² to ca. 48 mA/cm². === Graduation date: 2013 === Access restricted to the OSU Community at author's request from April 1, 2013 - April 1, 2015