University of Minnesota
University Relations
http://www.umn.edu/urelate
612-624-6868

Minnesota Supercomputing Institute


Log out of MyMSI

Research Abstracts Online
January 2008 - March 2009

University of Minnesota Twin Cities
College of Food, Agricultural, and Natural Resource Sciences and
Institute of Technology
Department of Bioproducts and Biosystems Engineering

PI: Shri Ramaswamy, Associate Fellow

Three-dimensional Structural Analysis of Porous Materials via Micro-tomography Image Processing

Porous materials such as paper present a very intricate structure which is difficult to understand and analyze, particularly when the goal is to characterize it to allow for accurate modeling and prediction of mechanical or optical or transport properties. The past two decades have given way to a non-intrusive and non-destructive method, x-ray micro-tomography, as an extremely powerful tool to obtain the internal rendering of porous media. The complete three dimensional understanding of the fiber matrix, combined with powerful image analysis tools, has shown great potential in providing accurate results, and a promising replacement technique for standardized bulk measurements that can only give crude averages for the overall media being studied.

Recently, access to high-resolution imaging facilities such as the European synchrotron has opened the door to very high resolution scans of porous materials, allowing for much higher accuracy in visualizing the structural features. The major drawback was the amount of noise in the data collected preventing further analysis.

A new set of fully automated image analysis tools was developed to de-noise the raw tomography data and accurately identify the various components of the porous media. Thus obtained higher quality images were then analyzed for structural parameters such as porosity, interfacial area, pore size distribution for samples of varying structure, fiber sources and process treatments and compared with experimental data from conventional techniques. The results indicate interesting differences between the high-resolution and low-resolution images as well as the different mechanical treatment of fibers.

An improved method for estimating transport properties of the porous media using random walk simulations in actual three-dimensional images of the samples is also presented.  In addition to single-phase transport, a method for simultaneous multi-phase transport through the porous media is also presented and results for varying sample structures are compared with experimental data. This provides a unique tool for directly predicting transport behavior of porous materials.

Group Members

Yves Defrenne, Graduate Student
Huanjiang Huang, Research Associate
Sho Takagaki, Research Associate
Vasili Zhdankin, Graduate Student