Laboratory evaluation of freezing-thawing resistance of aggregate for concrete pavement using Iowa Pore Index Tests

Ren, Jiaxi
Major Professor
Kejin Wang
Fatih Bektas
Committee Member
Journal Title
Journal ISSN
Volume Title
Research Projects
Organizational Units
Journal Issue
Civil, Construction, and Environmental Engineering

This study is aimed at evaluating Iowa pore index (IPI) test method for characterizing pores in aggregates that may have significant influence on freezing-thawing (F-T) durability of concrete. The study includes three phases: (a) evaluating the pore indexes of various concrete aggregates from different sources that mostly contain both carbonate and non-carbonate portions; (b) testing volumetric properties (such as specific gravity, absorption and desorption) of the aggregate; and (c) examining the correlations between these test results. Statistical analyses were also applied in the examination of these correlations.

The results of this study suggest that IPI test is a quick and simple test method for aggregate pore structure evaluation. Carbonate aggregates generally have higher absorption, lower specific gravity, higher primary pore index (PPI), and much higher secondary pore index (SPI) (for a given absorption) than non-carbonate aggregates. The PPI and PPI+SPI measurements are closely related to absorption of aggregates (for carbonate, non-carbonate, and bulk aggregates). The correlations of SPI with aggregate absorption are very weak, especially for carbonate aggregate. This may be attributed by other unidentified factors (such as aggregate pore tortuosity and the IPI test pressure and time). A pattern has been observed for the residual of the linear regression model for SPI-absorption correlation, which suggests that a higher order regression may be needed to better describe the correlation between SPI and absorption for carbonate aggregate.

More research should be conducted to further study the correlation between F-T durability and aggregate chemistry, mineralogy, and pore structure. Additional tests should also be done to develop statistical model better correlates the aforementioned testing results and aggregate property.