Title:
Models for Property Prediction of Pervious Concretes
Author(s):
Omkar Deo, Milani S. Sumanasooriya, and Narayanan Neithalath
Publication:
Symposium Paper
Volume:
282
Issue:
Appears on pages(s):
1-20
Keywords:
Clogging, Compressive strength, Particle capture model, Permeability, Pervious Concrete, Pore structure, Porosity, Statistical analysis.
DOI:
10.14359/51683642
Date:
12/27/2011
Abstract:
Properties of a random porous material such as pervious concrete are strongly dependent on its pore structure features. This study describes the development of different models to understand the material structure – property relationships in pervious concretes. Several pervious concrete mixtures with different pore structure features are proportioned. The pore structure features such as pore area fractions, pore sizes, mean free spacing of the pores, specific surface area, and the three-dimensional pore distribution density are extracted using image analysis methods. The performance features modeled as a function of the pore structure features are: (1) the unconfined compressive strength, (2) permeability, and (3) permeability reduction due to particle trapping in the pores (clogging). A statistical model is used to relate the compressive strength to the relevant pore structure features, which is then used as a base model in a Monte-Carlo simulation for feature sensitivity evaluation. Permeability prediction is accomplished using the well-known Katz-Thompson equation that employs the pore structure features. An idealized 3-D geometry obtained from 2-D planar images of pervious concrete sections is used along with a probablistic particle capture model to predict the particle retention associated with clogging material addition and simulated runoff. These models are anticipated to be useful in designing pervious concrete systems of desired pore structure for requisite performance.