Effect of Competing Oxidizing Reactions and Transport Limitation on the Faradaic Efficiency in Plasma Electrolysis

Abstract

Plasma electrolysis, where a solid electrode in an electrolytic cell is replaced by a plasma (or gas discharge), differs from conventional electrolysis by not being dictated by the surface characteristics of an electrode, but by the chemical species injected into the solution from the plasma. Reduction in a plasma cathode configuration occurs mostly by plasma-injected solvated electrons (e − aq ), which may engage in side reactions, such as the second order recombination of e − aq , that ultimately reduce the faradaic efficiency for the production of a desired product. In this work, we show that the depletion of reactants at the plasma-liquid interface due to insufficient transport can reduce the predicted faradaic efficiency for a plasma cathode at low concentrations. Measurements of the faradaic efficiency using the dissociative electron attachment to chloroacetate and the ferri/ferrocyanide redox couple confirm this behavior. The effect of other mechanisms on the faradaic efficiency, such as competing oxidation reactions with the hydroxyl radical, are also evaluated and found to be far less significant. Unlike conventional electrolysis, stirring the solution does not increase the faradaic efficiency, but increasing the species concentration does.

Chemical Analysis of Secondary Electron Emission from a Water Cathode at the Interface with a Nonthermal Plasma

“…To confirm that e aq − are present in the plasma anode configuration, we used the dissociative attachment reaction of e aq − with chloroacetate and measured the ensuing chloride concentration. 14,19 As shown in Figure S5, the results confirm the presence of e aq − , and an initial estimate is that approximately one e aq − is produced per incoming argon ion from the plasma; this is less than other estimates, 14 and more detailed follow-up studies are needed. However, as shown in Figure 4, these e aq − do not contribute to the electron emission process.…”

“…A possible explanation for the lack of effect of e aq – and e pre – scavengers on the plasma voltage is that these electrons simply are not present in the plasma–liquid interface. Contrary to ·OH and H·, which have been detected and measured in many experiments in the plasma phase, , the study of e aq – has focused mostly on plasma cathodes, where electrons are directly injected into the solution from the plasma. ,, In the plasma anode configuration, in contrast, e aq – are likely produced by bombarding plasma ions ionizing liquid water. To confirm that e aq – are present in the plasma anode configuration, we used the dissociative attachment reaction of e aq – with chloroacetate and measured the ensuing chloride concentration. , As shown in Figure S5, the results confirm the presence of e aq – , and an initial estimate is that approximately one e aq – is produced per incoming argon ion from the plasma; this is less than other estimates, and more detailed follow-up studies are needed.…”

“…Contrary to •OH and H•, which have been detected and measured in many experiments in the plasma phase, 41,42 the study of e aq − has focused mostly on plasma cathodes, where electrons are directly injected into the solution from the plasma. 6,19,43 In the plasma anode configuration, in contrast, e aq − are likely produced by bombarding plasma ions ionizing liquid water. To confirm that e aq − are present in the plasma anode configuration, we used the dissociative attachment reaction of e aq − with chloroacetate and measured the ensuing chloride concentration.…”

“…Electrochemical System. Experiments were conducted in an Hcell electrochemical reactor, such as the one in our previous work, 19 using an argon (UHP, Airgas) plasma. A schematic of the electrochemical cell and electrical circuit is shown in Figure 1.…”

“…Chloride was used as an indicator for the presence of e aq − based on the dissociative attachment of e aq − to chloroacetate, as reported in previous works. 14,19 The chloride concentration resulting from exposure of chloroacetate solutions to plasma was measured with a chloride-sensitive electrode (double junction, Cole-Parmer) connected to a 10 MΩ digital multimeter (Keysight 34401A). As in our previous work, 19 a calibration curve was prepared prior to each experiment in a given aqueous solution to account for any matrix effects.…”