Prediction of Maximum Dry Density and Angle of Internal Friction Based on Sieve Analysis

Document Type : Original Article

Authors

1 National Water Research Center - Construction Research Institute - Egypt

2 Fayoum University - Faculty of Engineering - Department of Civil Engineering - Egypt

3 Fayoum University- Faculty of Engineering - Department of Mathematics and Physics - Egypt

4 National Water Research Center - Construction Research Institute - Egypt Al-Baha University - Faculty of Engineering - Civil Engineering Department - Kingdom of Saudi Arabia

Abstract

Grain size distribution (GSD) is widely recognized as a critical index property of soils. This research seeks to derive empirical equations linking maximum dry density and internal friction angle to GSD. It also evaluates the consistency of maximum dry density results from standard and modified proctor compaction tests. The derived relationships provide a quick and pragmatic means to estimate maximum dry density and internal friction angle without extensive laboratory work. The study concentrates on clean sandy soils, some of which contain gravel fractions that vary from 0% to 30%. Equations for maximum dry density were formulated from a dataset of eighty-five compaction tests, while the correlations for internal friction angle were based on the results of ten tests. The findings demonstrate robust relationships between maximum dry density—regardless of whether it comes from the modified or standard proctor method—and sieve data, yielding coefficients of determination (R²) between 0.58 and 0.97. Higher R² values signal a more reliable prediction, confirming that the GSD parameters effectively characterize the compaction behavior of these sandy gravel soils. The best prediction for the internal angle of friction came from combining the particle sizes D10, D30, D50, D60, the sand ratio, and the maximum dry density determined by the modified proctor test, yielding an R² of 0.79. The least effective correlation arose when the internal angle of friction was linked solely to the coefficients of uniformity and curvature, along with the maximum dry density from the modified proctor test, resulting in the lowest R² value.

Keywords