An accurate method for structural analysis of 2D foam

Dimarco, Steven Lance

Permalink

https://hdl.handle.net/2142/116284 Copy

Description

Title

An accurate method for structural analysis of 2D foam

Author(s)

Dimarco, Steven Lance

Issue Date

2022-07-26

Director of Research (if dissertation) or Advisor (if thesis)

Hilgenfeldt, Sascha

Department of Study

Mechanical Sci & Engineering

Discipline

Mechanical Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• foam
• foams
• soap films
• bubbles
• Plateau
• Matlab
• ImageJ
• quasi-2D
• Fermat

Abstract

Foams are prototypical systems of Soft Matter, filling space with bubbles whose shape, size, and relative positioning co-determine many properties of the material, in particular its mechanical response. Recent theoretical and simulational work suggests that understanding foam structure in detail allows for quantitative statements about the mechanics of a whole class of cellular soft materials sharing aspects of foam structure, such as biological tissues. These theories can be tested most easily in an experimental aqueous foam system, where a single layer of bubbles generates a quasi-2D structure. For a stringent test, it is crucial to obtain a highly accurate and completely consistent quantification of certain aspects of the foam geometry, in particular bubble size, bubble topology, and the total perimeter of the quasi-2D foam’s cross section. This perimeter value (a measure of empirical mechanical foam energy) is traditionally evaluated from image analysis identifying all edges between polygonal bubbles. A consistent image analysis and subsequent reconstruction of the foam is cumbersome for large samples. In this thesis we suggest a much faster, less CPU-intensive, and less storage space-intensive method for foam reconstruction called Vertex Reconstruction Method (VRM). Here, only the vertex regions of the foam (vertical Plateau borders of the quasi-2D foam) need to be extracted through image analysis, a much simpler task. Because of the known local geometry of foam vertices, the location and orientation of the triangular vertex regions is sufficient to fully reconstruct the entire foam. Using our own experimental samples imaged at high resolution under a scanner, we show that the VRM is at least as accurate as traditional methods while using much less computational resources. Using the extracted foam energy and statistical measures, we also demonstrate that foams indeed conform to previously developed theories relating these quantities to the value of the system’s mechanical energy functional.

Graduation Semester

2022-08

Type of Resource

Thesis

Copyright and License Information

Copyright 2022 Steven Dimarco

Owning Collections