Gilbart, Benjamin
(2022)
Adapting Numerical Models of Surface Barrier Discharges to Real-World Conditions.
PhD thesis, University of Liverpool.
|
Text
201311074_June2022.pdf - Unspecified Download (9MB) | Preview |
Abstract
The work described here aims to advance the numerical modelling used for Surface Barrier Discharges (SBDs) so that their design takes into account the non-idealised conditions encountered in real life. Starting from a two-dimensional fluid model of an SBD describing a single discharge gap, the model was used to analyse the effect of varying humidity on the species generated in the discharge. Furthermore, the model was upgraded to take into account heat transfer processes induced by the plasma, in addition to the mutual interaction between multiple discharge gaps when they are placed closely together. Wherever possible, the predictions of the model were compared to experimental data. For any application of SBDs where ambient air is the working gas, humidity varies at different times of the day and during different seasons. Therefore, it is vital to understand how the variation in humidity affects the generation of Reactive Oxygen and Nitrogen species (RONS). This motivated the first investigation in this work, in which the humidity level was used as an input to the model and the behaviour of the resultant species was analysed as a function of humidity. It is reported that the densities of HNOx species increase as humidity is increased, with the rate of increase slowing at higher humidity. Hydrogen-free species were only marginally influenced by humidity. These findings indicate that for applications where Hydrogen-free species are key, varying humidity from day to day is not a concern. For applications where HNOx species are important, the application should use synthetic air with a controllable added H2O fraction to maintain a steady dose of reactive species. A second observation on the operation of SBDs in ambient air is the rapid rise in their temperature. Considering that many reaction coefficients are functions of gas temperature, it is vital to understand how the increase in gas temperature impacts the performance of the SBD as a source of reactive species. This motivated a key development made in this work which was incorporating heat transfer treatment into the SBD model. This was achieved by coupling the heat equation to the model. Two sources of heat were computed: the heat flux to the dielectric surface due to ion bombardment and the volumetric heat source in the gas due to inelastic collisions between the background gas and energetic electrons. The work revealed that ion bombardment was the primary heating mechanism of the dielectric. The impact of accounting for the increase in temperature was also investigated, where it was shown that it can cause a difference of up to 40% in the densities of some species, particularly the Reactive Nitrogen species (RNS). The impact of this finding is that it paves the way for controlling the long-lived species chemistry of the discharge by controlling the temperature of the dielectric. Consequently, for a practical application where Reactive Oxygen Species (ROS) are of interest, active cooling of the dielectric is recommended, while for an application focused on RNS, active heating of the dielectric is advantageous. Another impact of this investigation was quantifying the errors in species density predictions from numerical models describing SBDs when the temperature effect is ignored, which were up to 40%. The third aspect of SBDs applications investigated was the use of an SBD array, consisting of closely spaced discharge gaps, instead of a single discharge gap configuration as is typically used for research studies. The proximity of discharge gaps may induce emergent phenomena which cannot be observed in a single discharge. To capture such phenomena, the model was upgraded to investigate an array of 6 discharge gaps with a controllable distance between them. Supporting evidence was provided by Particle Image Velocimetry (PIV) experimental data. It was shown in this work that increasing the electrode width resulted in the discharge power decreasing exponentially for a fixed applied voltage. It was also shown that decreasing the distance between the discharge gaps forced flow vortices to overlap, creating a ripple in the flow downstream of the discharge, where the velocity varies by 200% from maximum to minimum value. This ripple has a significant impact on the flux of species to a downstream sample when the flux is convection dominated. These findings show that while bringing the sample to be treated closer to the SBD array increases the flux to it as it is convection dominated, it comes at the expense of uniformity. Thus, a trade-off must be made between the magnitude of the arriving flux to a sample and its uniformity. Briefly, the work presented in this thesis provides a set of recommendations to be considered when designing an SBD for a particular application. The work described here aims to advance the numerical modelling used for Surface Barrier Discharges (SBDs) so that their design takes into account the non-idealised conditions encountered in real life. Starting from a two-dimensional fluid model of an SBD describing a single discharge gap, the model was used to analyse the effect of varying humidity on the species generated in the discharge. Furthermore, the model was upgraded to take into account heat transfer processes induced by the plasma, in addition to the mutual interaction between multiple discharge gaps when they are placed closely together. Wherever possible, the predictions of the model were compared to experimental data. For any application of SBDs where ambient air is the working gas, humidity varies at different times of the day and during different seasons. Therefore, it is vital to understand how the variation in humidity affects the generation of Reactive Oxygen and Nitrogen species (RONS). This motivated the first investigation in this work, in which the humidity level was used as an input to the model and the behaviour of the resultant species was analysed as a function of humidity. It is reported that the densities of HNOx species increase as humidity is increased, with the rate of increase slowing at higher humidity. Hydrogen-free species were only marginally influenced by humidity. These findings indicate that for applications where Hydrogen-free species are key, varying humidity from day to day is not a concern. For applications where HNOx species are important, the application should use synthetic air with a controllable added H2O fraction to maintain a steady dose of reactive species. A second observation on the operation of SBDs in ambient air is the rapid rise in their temperature. Considering that many reaction coefficients are functions of gas temperature, it is vital to understand how the increase in gas temperature impacts the performance of the SBD as a source of reactive species. This motivated a key development made in this work which was incorporating heat transfer treatment into the SBD model. This was achieved by coupling the heat equation to the model. Two sources of heat were computed: the heat flux to the dielectric surface due to ion bombardment and the volumetric heat source in the gas due to inelastic collisions between the background gas and energetic electrons. The work revealed that ion bombardment was the primary heating mechanism of the dielectric. The impact of accounting for the increase in temperature was also investigated, where it was shown that it can cause a difference of up to 40% in the densities of some species, particularly the Reactive Nitrogen species (RNS). The impact of this finding is that it paves the way for controlling the long-lived species chemistry of the discharge by controlling the temperature of the dielectric. Consequently, for a practical application where Reactive Oxygen Species (ROS) are of interest, active cooling of the dielectric is recommended, while for an application focused on RNS, active heating of the dielectric is advantageous. Another impact of this investigation was quantifying the errors in species density predictions from numerical models describing SBDs when the temperature effect is ignored, which were up to 40%. The third aspect of SBDs applications investigated was the use of an SBD array, consisting of closely spaced discharge gaps, instead of a single discharge gap configuration as is typically used for research studies. The proximity of discharge gaps may induce emergent phenomena which cannot be observed in a single discharge. To capture such phenomena, the model was upgraded to investigate an array of 6 discharge gaps with a controllable distance between them. Supporting evidence was provided by Particle Image Velocimetry (PIV) experimental data. It was shown in this work that increasing the electrode width resulted in the discharge power decreasing exponentially for a fixed applied voltage. It was also shown that decreasing the distance between the discharge gaps forced flow vortices to overlap, creating a ripple in the flow downstream of the discharge, where the velocity varies by 200% from maximum to minimum value. This ripple has a significant impact on the flux of species to a downstream sample when the flux is convection dominated. These findings show that while bringing the sample to be treated closer to the SBD array increases the flux to it as it is convection dominated, it comes at the expense of uniformity. Thus, a trade-off must be made between the magnitude of the arriving flux to a sample and its uniformity. Briefly, the work presented in this thesis provides a set of recommendations to be considered when designing an SBD for a particular application.
| Item Type: | Thesis (PhD) |
|---|---|
| Divisions: | Faculty of Science and Engineering > School of Electrical Engineering, Electronics and Computer Science |
| Depositing User: | Symplectic Admin |
| Date Deposited: | 05 Sep 2022 13:09 |
| Last Modified: | 18 Jan 2023 20:56 |
| DOI: | 10.17638/03157317 |
| Supervisors: |
|
| URI: | https://livrepository.liverpool.ac.uk/id/eprint/3157317 |
Dimensions
Dimensions
