In the semiconductor industry, the ongoing down-scaling of nanoelectronic elements has lead to an increasing complexity of their fabrication. Hence, the individual fabrication processes become increasingly difficult to handle. To minimize cross-contamination, intermediate surface cleaning and preparation steps are inevitable parts of the semiconductor process chain. Here, one major challenge is the removal of residual nano-particulate contamination resulting from abrasive processes such as polishing and etching. In the past, physical cleaning techniques such as megasonic cleaning have been proposed as suitable solutions. However, the soaring fragility of the smallest structures is constraining the forces of the involved physical removal mechanisms. In the case of "megasonic" cleaning -cleaning with ultrasound in the MHz-domain-the main cleaning action arises from strongly oscillating microbubbles which emerge from the periodically changing tensile strain in the cleaning liquid during sonication. These bubbles grow, oscillate and collapse due to a complex interplay of rectified diffusion, bubble coalescence, non-linear pulsation and the onset of shape instabilities. Hence, the resulting bubble size distribution does not remain static but alternates continuously. Only microbubbles in this distribution that show a high oscillatory response are responsible for the cleaning action. Therefore, the cleaning process efficiency can be improved by keeping the majority of bubbles around their resonance size. In this paper, we propose a method to control and characterize the bubble size distribution by means of "pulsed" sonication and measurements of acoustic cavitation spectra, respectively. We show that the so-obtained bubble size distributions can be related to theoretical predictions of the oscillatory responses of and the onset of shape instabilities for the respective bubbles. We also propose a mechanism to explain the enhancement of both acoustic and cleaning activity observed when turning the ultrasound periodically on and off in comparison to the case of continuous sonication.
CITATION STYLE
Hauptmann, M., Struyf, H., Mertens, P., Heyns, M., De Gendt, S., Glorieux, C., & Brems, S. (2012). Bubble size distribution analysis and control in high frequency ultrasonic cleaning processes. In IOP Conference Series: Materials Science and Engineering (Vol. 42). https://doi.org/10.1088/1757-899X/42/1/012016
Mendeley helps you to discover research relevant for your work.