ACCELERATED STRENGTH TESTING OF SOIL-CEMENT.

Abstract

This research was conducted to investigate the possibility of devising and utilizing an accelerated curing procedure in order to estimate the 7-day and 28-day normal curing (72(DEGREES)F and 100% humidity) compressive strengths of soil-cement mixes. Based on information reported in the literature on accelerated curing in lime stabilization and concrete, the accelerated curing procedure adopted in this research was a modified version of the boiling water method given by ASTM 684-81. By this procedure compacted soil-cement specimens were boiled in distilled water for certain periods of time. After a 30-minute cooling period, the specimens were soaked in water for 24 hours and then tested for strength. Mechanical as well as physico-chemical tests were conducted on accelerated- and normally cured specimens. Mixtures of clay (kaolinite Hydrite 10), sand and portland cement (Type I/II) were utilized in preparing the specimens. The unconfined compression test results indicated that accelerated strength values increased with increasing boiling time and with increasing cement content. The strength values also increased with decreasing clay content down to 30%. Below 30% clay content the strength decreased. As for predicting normal curing strengths from accelerated curing strengths, two procedures were adopted. Procedure A predicted the 7-day and 28-day strengths by boiling specimens for 3 hours and 40 minutes and 4 hours and 20 minutes, respectively. Procedure B predicted the normal-cure strengths using linear regression equations. The predictions by both procedures were found to be satisfactory, within (+OR-) 15% of the normal-cure strengths. X-ray diffraction data suggested less hydration had occurred in the accelerated-cure specimens than in the normal-cure specimens. This probably was due to the shielding of cement grains by clay and hydration gels. pH measurements showed that specimens cured by both procedures exhibited highly alkaline environment. Electron micrographs of selected samples appeared to confirm the existence of more unhydrated and/or partially hydrated cement in the accelerated-cure samples.