Non-destructive soil testing using x-ray computed tomography
Nielsen, Brent Daniel.
MetadataShow full item record
The mechanical behavior of soils is highly dependent on the particle microstructure. Traditional geotechnical engineering soil tests generally do not measure soil properties on a micro-scale; instead, macro scale properties are commonly used as estimates of microstructure properties in determining soil engineering behavior. Additionally, traditional geotechnical engineering soil tests are destructive in nature, and many test methods destroy the same soil properties they intend to measure. The goal of this research was to develop non-destructive soil test methods using x-ray computer-aided tomography (CT) scanning techniques to determine soil index properties. The CT scanning process provides a promising method for examining soil microstructure in a non-destructive manner. This research had two main objectives. The first was to configure the Montana State University Civil Engineering Department's computer-aided tomography scanner to perform CT scans on soil samples. The second objective was to use the CT scanner to develop nondestructive test procedures to determine geotechnical index properties of soils. Test methods were developed in this study to determine porosity, grain size distribution, and pore size distribution. The results from the first objective showed that the MSU CT scanning equipment is capable of producing high quality CT scans of soil materials. Resolution limitations of the scanner define the smallest soil grain size that is detectable in a CT scan, but the scan resolution may be improved by using smaller sample sizes for small particle soils. The results of the second portion of the study show that the non-destructive CT scanning test methods compare favorably with traditional geotechnical laboratory mechanical test methods. CT-measured porosity values and grain size distributions compared well with mechanical testing results, which were used to validate the new test methods. In addition, the CT-measured pore size distributions were in good agreement with an accepted pore size mathematical model. Since traditional pore size distribution tests are time-consuming, labor intensive, and destructive in nature, the non-destructive x-ray CT scanning test methods developed in this study show strong promise as a means for measuring an elusive soil property that cannot be accurately measured using traditional geotechnical testing procedures.