• Login
• Admin

Chu, Jin Kiat (2023) Modelling of ultrasonic wave propagation for industrial sonoreactor design and optimisation. PhD thesis, University of Nottingham.

Preview

PDF (Thesis - as examined) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader Available under Licence Creative Commons Attribution. Download (8MB) | Preview

Abstract

The development of power ultrasound technology had raised interest in the design of sonoreactors. Numerical simulations were identified as a promising tool for tackling the challenges involved in scaling up and optimisation. The development of simulation strategies for sonoreactor acoustics had mainly focused on horn-type configurations, and little information was available for bath-type systems. This work evaluated the performance of three popular frequency-domain models for the prediction of acoustic pressure in bath-type configurations. The results of the pure-liquid, linear Commander and Prosperetti (CP), and nonlinear Helmholtz (NH) models were verified against hydrophone measurements and antinode characterisation results obtained from experiments. Emphasis was placed on the evaluation of the linear CP model, as it was hypothesised to best characterise the acoustic behaviour of the studied systems. In accordance with the objectives of this work, seven case studies were carried out.

Case Study 1 explored the suitability of frequency-domain solutions for the characterisation of acoustic pressure fields. Predictions of frequency-domain solutions compared well with time-domain models and experimental observations, notably in terms of antinode locations. The results justified the use of frequency-domain models to characterise sonoreactors. Case Study 2 explored the acoustic boundary conditions used in the modelling strategies of this work. The test cases showed that Dirichlet pressure boundaries can be used to represent the vibrating walls of studied sonoreactors. Numerical aspects of other boundary conditions were also identified to improve confidence in the simulation results.

Case Study 3 presented a pressure validation study using a 252 L ultrasonic bath to evaluate the performance of selected acoustic pressure models. The pure-liquid model was found to yield good agreements for lower-frequency cases (35 and 72 kHz, but failed to account for increased attenuation at higher-frequency cases (110 and 170 kHz). The linear CP model showed mixed results and performance was strongly affected by wall reflection. The results were found to depend on the bubble density magnitude. Empirically tuned bubble densities significantly improved the performance of the linear CP model for the 35 and 72 kHz cases. Preliminary studies for the nonlinear CP and NH models resulted in strong deviations from hydrophone measurements, which suggested an overtuned attenuation mechanism.

Case Study 4 presented a novel investigation on the effect of geometry on qualitative antinode validation methods. Heuristic and theoretical analysis supported the hypothesis which stated that the value of the effective-geometry-to-wavelength ratio, D/λ, affects the confidence in the antinode validation results. A simplified standing wave model was developed and proposed to show that wavelength-shortening effects can only be detected above a certain D/λ threshold. The findings of the case study were found to strongly influence current interpretations of antinode validation in sonoreactors and outlined the effect sonoreactor geometry on the validity of antinode verification results.

Antinode validation was performed for the pure-liquid, linear CP, and NH models in Case Study 5, using SCL images of a hexagonal sonoreactor. The performance of the selected models in terms of phase-speed prediction was evaluated. The pure-liquid model consistently predicted wavelengths longer than those of SCL observations. The outputs of the linear CP model once again showed strong sensitivity to the bubble density parameter. The results suggested that the best agreement with the SCL standing wave patterns could be obtained using a bubble density magnitude between 1E10 – 1E11 1/m3. The NH model also yielded mixed results. The phase speed prediction mechanism of the NH model was found to depend on many factors, in which a detailed analysis could not be completed within the scope of this work.

Case Study 6 served to demonstrate and discuss the proposed semi-empirical wavelength tuning approach for the characterisation of the bubble density in a sonoreactor. The novel method was developed based on the linear CP model to address the uncertainties regarding bubble field characterisation in sonoreactor simulations. The method allowed the bubble density parameter to be tuned against wavelength measurements which led to remarkable improvements for antinode characterisation results using the linear CP model.

Case Study 7 presents an exploratory study on the use of frequency-domain modelling strategies for multi-frequency sonoreactors. Based on Parseval’s theorem, the CP model was modified to account for multi-frequency harmonics. The antinode verification of the simulation results suggested that the use of linear wave superposition was a promising approach for simplifying the characterisation of multi-frequency acoustic fields. The remarkable agreement with SCL images also further supported the viability of the linear CP model.

Item Type:	Thesis (University of Nottingham only) (PhD)
Supervisors:	Tiong, Timm Joyce Chong, Faye Siew Hui
Keywords:	Sonoreactor; simulation; ultrasound; power ultrasound; computational modelling; fem, acoustics; sonoreactor acoustics; sonochemiluminesence
Subjects:	T Technology > TP Chemical technology
Faculties/Schools:	University of Nottingham, Malaysia > Faculty of Science and Engineering — Engineering > Department of Chemical and Environmental Engineering
Item ID:	73281
Depositing User:	CHU, Jin
Date Deposited:	22 Jul 2023 04:40
Last Modified:	22 Jul 2023 04:40
URI:	https://eprints.nottingham.ac.uk/id/eprint/73281

Actions (Archive Staff Only)

Edit View