Modeling the Formation of Gas Bubbles inside the Pores of Reactive Electrochemical Membranes in the Process of the Anodic Oxidation of Organic Compounds

Abstract

The use of reactive electrochemical membranes (REM) in flow-through mode during the anodic oxidation of organic compounds makes it possible to overcome the limitations of plate anodes: in the case of REM, the area of the electrochemically active surface is several orders of magnitude larger, and the delivery of organic compounds to the reaction zone is controlled by convective flow rather than diffusion. The main problem with REM is the formation of fouling and gas bubbles in the pores, which leads to a decrease in the efficiency of the process because the hydraulic resistance increases and the electrochemically active surface is shielded. This work aims to study the processes underlying the reduction in the efficiency of anodic oxidation, and in particular the formation of gas bubbles and the recharge of the REM pore surface at a current density exceeding the limiting kinetic value. We propose a simple one-dimensional non-stationary model of the transport of diluted species during the anodic oxidation of paracetamol using REM to describe the above effects. The processing of the experimental data was carried out. It was found that the absolute value of the zeta potential of the pore surface decreases with time, which leads to a decrease in the permeate flux due to a reduction in the electroosmotic flow. It was shown that in the solution that does not contain organic components, gas bubbles form faster and occupy a larger pore fraction than in the case of the presence of paracetamol; with an increase in the paracetamol concentration, the gas fraction decreases. This behavior is due to a decrease in the generation of oxygen during the recombination reaction of the hydroxyl radicals, which are consumed in the oxidation reaction of the organic compounds. Because the presence of bubbles increases the hydraulic resistance, the residence time of paracetamol—and consequently its degradation degree—increases, but the productivity goes down. The model has predictive power and, after simple calibration, can be used to predict the performance of REM anodic oxidation systems.

A Simple 1D Convection-Diffusion Model of Oxalic Acid Oxidation Using Reactive Electrochemical Membrane

“…After permeate flux measurements, it was found that when an electric current is applied, the solution flux increases due to the electroosmotic flow. The calculation results showed that its contribution is about 12 percent at TMP = 40 mbar [31]. In the current work, the value of the parameter σ is assumed constant and equal to 1.7 × 10 −14 m 2 .…”

“…It should be noted that at higher current densities or lower OA fluxes, it is mandatory to take into account the evolution of bubbles since the oxygen concentration will overcome the solubility limit. This phenomenon is considered in detail in [31]. In the pores of REM, the concentration of dissolved molecular oxygen increases when the oxygen evolution reaction occurs.…”

“…Misal and coauthors used the one-dimensional model to estimate the potential and current density distribution and the evolution of the concentration of organic compounds within REM [29]. In our recent works [30,31], we used models concerning the transport and reaction of organic species with hydroxyl radicals generated at a TiOx REM operated in a flow-through mode to obtain the dependences of the paracetamol mineralization efficiency on the pore radius and porosity of REM [30] and the oxygen bubbles formation [31].…”

“…The oxygen concentration in the solution exceeds the solubility limit only at the lowest TOC fluxes and is insufficient at given current density [31]; thus, the bubble-formation caused by oxygen evolution is not taken into account;…”

“…In [30], the processes occurring at a depth of REM from 0 to 30 µm from its surface were considered, and the concentration of organic compound (paracetamol) in all calculations was extremely low (0.19 mM), which makes it possible to use the above assumptions. In our recent work [31], it was shown that during the simulation of anodic oxidation of organics, it is necessary to take into account the oxygen evolution reaction, since in the case of Magnelli phases of stoichiometric titanium oxide even at low current densities, the contribution of the reaction to electron transfer is very high. When considering the entire volume of REM, 2D simulation is difficult due to the need for large computing power, which extremely complicates the calculations.…”