Scale up design study on process vessel dimensions for ultrasonic processing of water and liquid aluminium

Abstract

Scaling up ultrasonic cavitation melt treatment (UST) requires effective flow management with minimised energy requirements. To this end, container dimensions leading to the resonance play a crucial role in amplifying pressure amplitude for cavitation. To quantify the importance of resonance length during the treatment of liquid aluminium, we used calibrated high-temperature cavitometers (in the range of 8–400 kHz), to measure and record the acoustic pressure profiles inside the cavitation-induced environment of liquid Al and deionized water (used as an analogue to Al) excited at 19.5 kHz. To achieve a comprehensive map of the acoustic pressure field, measurements were conducted at three different cavitometer positions relative to the vibrating sonotrode probe and for a number of resonant and non-resonant container lengths based on the speed of sound in the treated medium. The results showed that the resonance length affected the pressure magnitude in liquid Al in all cavitometer positions, while water showed no sensitivity to resonance length. An important practical application of UST in aluminium processing concerns grain refinement. For this reason, grain size analysis of UST-treated Al-Cu-Zr-Ti alloy was used as an indicator of the melt treatment efficiency. The result showed that the treatment in a resonance tank of (the wavelength of sound in Al) gave the best structure refinement as compared to other tested lengths. The data given here contribute to the optimisation of the ultrasonic process in continuous casting, by providing an optimum value for the critical compartment (e.g. in a launder of direct-chill casting) dimension.

In-situ observations and acoustic measurements upon fragmentation of free-floating intermetallics under ultrasonic cavitation in water

“…A suitable transparent candidate for such studies is water. Water shares similar cavitation behaviour with liquid aluminium, though the acoustic pressure can be up to 2 times larger in Al melts [17] , [18] . For this reason it has been systematically used to replicate the conditions of acoustic cavitation in liquid Al during UST [19] , [20] , [21] .…”

“…The raw voltage data were acquired with a Peripheral Component Interconnect (PCI) data acquisition device with sampling rate of 20 × 10 6 samples/s which allowed real-time signal tracking and recording of high-speed images and the FOH data. Fast Fourier transformation was applied to the voltage–time data to obtain the pressure values following the deconvolution process defined in our earlier work [17] , [18] , [44] and in [45] . For repeatability of the results, the experiments were conducted at least 4–5 times for each position and input power, including 3 extra runs without the presence of intermetallics.…”

Mechanisms of ultrasonic de-agglomeration of oxides through in-situ high-speed observations and acoustic measurements

“…Synchronised pressure measurement coupled with in-situ imaging of ultrasonic de-agglomeration were performed using a cavitometer sensor positioned close to the tip of the sonotrode as mentioned in Section 2.2 . It should be noted that the acoustic pressure values reported here were measured with reference to barometric pressure since the cavitometer was calibrated in water at ambient pressure conditions as previously reported in [30] . Fig.…”

“…This bespoke cavitometer comprised an in-built tungsten waveguide (Ø = 4 mm) connected to a piezoelectric sensor that converted the acquired mechanical vibrations into a voltage signal. The cavitometer had a spatial resolution of 40 mm and was calibrated in the National Physical Laboratory (NPL, UK) over a frequency range of 8–400 kHz with a sensitivity function as described in [30] . Spatial resolution of any sensor defines its ability to resolve the cavitation activity occurring within the ultrasonically treated volume.…”

“…The acoustic measurements were acquired from the onset of cavitation activity extending for up to tens of milliseconds to account for the de-agglomeration of particles confined within the imaging field of view. The analysis of the raw data (voltage–time) obtained from the cavitometer was performed using an in-house developed MATLAB code by applying a Fast Fourier Transformation (FFT) to the captured signal as described elsewhere [30] , [33] , [34] , to appropriately apply the sensitivity values of the sensor. The output of the acoustic emissions was displayed in the pressure–time domain after subtracting the background noise.…”