Collaborative Research: International Development of Non-Destructive Tools to Quantify Sample Disturbance and Its Correlation to Measured Behavior

Northeastern University, Boston MA

Investigators

Abstract

CMS- 0218314 PI: Thomas Sheahan Institution: Northeastern University Title: "Collaborative Research: International Development of Non-Destructive Tools to Quantify Sample Disturbance and its Correlation to Measured Behavior" Abstract: While important progress has been made in the use of in situ testing and geophysical methods to determine soil stress-strain-strength parameters for design, the state of practice for the foreseeable future will often rely on physical soil sample collection for laboratory testing. Samples collected for laboratory testing are often highly disturbed, but the geotechnical profession does not have adequate non-destructive tools for evaluating that disturbance or correlating it to the measured behavior either in situ or in laboratory tests. Existing methods of assessing sample disturbance, such as volumetric strain upon laboratory reconsolidation, are not a priori methods, i.e., one does not know a sample's quality until a laboratory specimen has been trimmed and set up in a testing device. At the same time, we face increasingly demanding geotechnical design and analysis problems that require more accurate estimates of soil behavior based on both in situ tests and laboratory tests on physical samples. The demands of infrastructure design as well as larger scale problems such as submarine landslide analyses require geotechnical engineers to produce not just a conservative answer, but the right answer, i.e., the closest to the "true" in situ mechanical properties as possible. This three year project involves an international effort to improve our existing state of practice in soft ground sampling technology, develop better techniques for assessing sample disturbance, and develop improved relationships between field-measured mechanical properties, the degree of sample disturbance, and the laboratory-measured mechanical properties from those samples. The Principal Investigators from the University of Massachusetts Amherst and Northeastern University are building on over 15 years of individual and joint experience in sampling, laboratory testing and equipment innovation to help solve this longstanding problem in the field of soft ground engineering. In addition, the project extends an existing partnership with the Norwegian Geotechnical Institute (NGI) to jointly develop field-deployable non-destructive sample disturbance evaluation techniques. The project involves: 1) developing down-hole block sampling equipment for use with conventional U.S. drill rigs; 2) sampling a variety of soil types with different degrees of structure from 4 locations in North America, using both conventional tube and block sampling methods; 3) performing seismic piezocone (CPTU) tests at the four sites to obtain baseline CPTU measurements and in situ shear modulus (Gmax); 4) developing a field suction probe and bender element device to evaluate sample effective stress and Gmax, respectively, in block and conventional tube samples; 5) implementing bender elements in laboratory test devices to measure Gmax after specimen set-up and at small strain levels; 6) conducting a laboratory testing program with the modified equipment to comprehensively measure the stress-strain-strength behavior of the four soils, including that at small and medium strains; 7) developing accurate correlations between the field- and lab-measured behavior in the context of the degree of sample disturbance and degree of soil structure. It is anticipated that the results from this work will have an impact on the state of the art and state of practice in sampling soft soils and correlating disturbance effects to measured behavior. A comprehensive framework can be established to link sample disturbance with soil behavior fundamentals such as structure and anisotropy. The block sampling method will be demonstrated as a viable tool for U.S. practice, and a standard of practice is being developed for quantitative assessment of sample disturbance. The correlations between sample disturbance and measured behavior will ultimately lead to improved accuracy in laboratory results versus the true in situ behavior, and thus more reliable geotechnical design without excessive conservatism.