The Plasma Water Reactor: A Geometric Approach to Scaling Electric Discharges for Water Treatment

Abstract

Advanced oxidation processes (AOPs) are established disinfection methods that can remove contaminants of emerging concern (CECs) by producing hydroxyl radicals and other reactive oxygen species (ROS) in situ. Production of hydrogen peroxide (H2O2), ozone (O3), and ultraviolet (UV) light in traditional AOPs can significantly contribute to the cost of water due to required consumables and associated infrastructure.

Plasma interacting with liquid water can generate additional transient ROS in solution while eliminating consumables and conversion efficiencies. Subsequently, in principle, plasma-based AOPs should be considerably cheaper and more effective than conventional AOPs. Although the plasma-water interface can facilitate vital kinetics through various pathways, approaches to date fail to scale-up to practical flow rates due to limited oxidant transport. In this work, the Plasma Water Reactor (PWR) is proposed as a scalable high-throughput system that advantageously uses flowing water to enhance plasma formation and propagation. The PWR utilizes a close-packed lattice of water jets to mimic packed bed dielectric barrier discharges where water streams serve as the dielectric media.

To sustain and maximize the plasma-water interface, the PWR design criteria incorporate jet stability and structure by considering fluid and electrohydrodynamic effects. Assuming steady jets, simulations for a hexagonal lattice with cylindrical electrodes indicated excellent plasma-water contact area with >82% of the PWR achieving the breakdown electric field in atmospheric air (|E| > 30kV/cm). Computations also suggested that changes in power density result in varied oxidant production, though this needs to be further verified.

The PWR was assessed in pure or single-CEC-spiked distilled water matrices. Parametric kinetic studies using pulsed power were performed by measuring power and species concentrations while varying the pulse voltage, width, and repetition frequency. In exclusively distilled water, H2O2 and O3 were measured for various combinations of pulser parameters. Though the pulser was power-limited, the PWR produced relevant oxidant concentrations and variations in H2O2/O3 based on chemical probe measurements. For a given set of pulser parameters that corresponded to high oxidant dose, methylene blue, methyl tert-butyl ether, and 1,4-dioxane were decomposed. These indicator compounds demonstrated effective flow rates on the order of 0.1—0.75gal/min for 90% removal. Since 1,4-dioxane exhibited the slowest destruction, the PWR was optimized using this compound. In addition to 1,4-dioxane, two transformation products, formate and acetate, and a plasma byproduct, nitrate, were measured. For three different pulser configurations, 1,4-dioxane kinetics were analyzed and the PWR achieved at least 0.5-log reduction, which implied sufficient oxidation. 1,4-dioxane decay displayed different order of reaction rates and real-time oxidant measurements confirmed H2O2/O3 ratios spanned several orders of magnitude. Thus, the PWR demonstrated the ability to vary kinetics. In the future, pilot studies must be performed to assess the PWR's efficacy for custom water matrices.