Contact-free surface chemistry with acoustic levitation
Licentiate thesis, 2022
In this licentiate thesis, three different designs of multiple-transducers acoustic levitators are presented and evaluated in terms of both simulated acoustic pressure and experimental performance. It was found that by tuning the cavity length of the device, it was possible to reduce the number of transducers while generating equally high acoustic pressure fields. Furthermore, the levitator with the densest packing of transducers exhibited the best performance, in terms of stability and levitation capacity. This framework can be applied in the customization of the design for specific applications.
A highly stable acoustic levitator was utilized for determining the surface tension of aqueous surfactant solutions through a data-driven approach. Approximately 50,000 photographs of acoustically levitated droplets were used for the training of a deep neural network while the predicting evaluation was based on ~10,000 photographs. The mean absolute error of the neural network surface tension predictions was below 0.9 mN/m. The methodology presented here surpassed previous limitations related to droplet size and deformation, while generating equally high, and in specific cases higher accuracy.
Acoustic levitation
machine learning
surface chemistry
Author
Smaragda Maria Argyri
Chalmers, Chemistry and Chemical Engineering, Applied Chemistry
NMR-Lev: Nuclear Magnetic Resonance spectroscopy applied to Levitating material
Swedish Research Council (VR) (2018-04196), 2019-01-01 -- 2021-12-31.
Swedish Foundation for Strategic Research (SSF) (ITM17-0436), 2019-01-01 -- 2021-12-31.
Subject Categories
Analytical Chemistry
Physical Sciences
Building Technologies
Chemical Sciences
Licentiatuppsatser vid Institutionen för kemi och kemiteknik, Chalmers tekniska högskola: Nr 2022:15
Publisher
Chalmers