Date of Award

Spring 5-3-2024

Level of Access Assigned by Author

Open-Access Thesis

Degree Name

Master of Science (MS)

Department

Civil Engineering

Advisor

Eric N. Landis

Second Committee Member

Edwin Nagy

Third Committee Member

Stephanie Wood

Abstract

With the goal of detecting and quantitively characterizing pore distribution using ultrasonic analysis in concrete specimens with fine and coarse aggregate included, specimens were created with varying air contents, two different w/c ratios, and with and without coarse aggregate, which consisted of glass beads. The glass beads were used because of the close match of acoustic impedance with cement paste. The presumption being that any scattering at the interface between the paste and the aggregate would be due to the presence of an interfacial zone rather than the impedance mismatch. The specimens were subjected to through-transmission ultrasonic interrogation using Yushi 4 MHz transducers along with an Olympus pulser and a Moku:Go data acquisition device (DAQ) to record the ultrasonic pulses. For attenuation measurements, the recorded ultrasonic signals were processed first by obtaining a baseline signal with the transducers face to face with only the coupling grease between them. This is compared to a through transmission pulse recorded through the specimen to determine the amount of signal lost through the specimen. The frequency-based attenuation for each specimen was determined through division in the frequency domain. Some dependence on air content was observed, but it did not provide a clear picture of pore sizes. A new method was developed to obtain the pore size distribution. This is done by first converting the time-based data to a frequency-based data set using an FFT using the same code as the attenuation calculations. The frequency spectrum is integrated, resulting in the magnitude of energy received at each frequency as well as the total magnitude of energy received within the full frequency band for the pulse. For this case, a frequency band from 1-10 MHz was used. These frequencies are then converted to wavelengths using the ultrasonic pulse velocity data for each specimen. For each wavelength, a corresponding percentage is calculated representing how much of the total pulse energy received in this frequency range was at the current wavelength being looked at. This was done for all the wavelengths within the frequency range. This results in a curve of the cumulative percentage of non-scattered wavelengths. By taking the compliments of these percentages, the curve now shows the cumulative wavelength percentage scattered within the specimen which is interpreted as a pore size cumulative percentage curve. For validation, a select set of specimens were then CT-scanned, and the resulting images processed to get an independent measurement of pore size distribution. This resulted in pore size cumulative percentage curves for both the ultrasonic and CT scans that were comparable for different air contents and aggregate inclusions. While there is not a clear fundamental basis for the proposed technique, the results are important and provide the next step to a field applicable ultrasonic testing approach to characterizing microstructure in concrete structures without using destructive methods.

Download

Share