Print ISSN: 1681-6900

Online ISSN: 2412-0758

Keywords : destructive Tests


Predicting Mechanical Properties of High Performance Concrete by Using Non-destructive Tests

Sura F. Al-Khafaji; Waleed A. Al-Qaisi; Shakir A. Al-Mishhadani

Engineering and Technology Journal, 2009, Volume 27, Issue 3, Pages 425-444

In this study, high performance concrete mixes were produced by using high
range water reducing agent and also by using 10% silica fume or 10% high
reactivity metakaolin as a partial replacement by weight of cement. Three cement
contents (350, 450, and 550) kg/m3 were used through this study. A total of 330
(100 mm) cubes, 132 (100×200 mm) cylinders, 132 (100×100×400 mm) prisms,
and 66 (150×300 mm) cylinders were casted and cured to the required age of test .
All specimens were cured in tap water except 165 cubes, which were submerged in
Cl ˉ + SO4ˉ ˉ solution at concentration identical to those present in severe
aggressive environment to study the effect of this solution on the compressive
strength of high performance concrete mixes. Compressive strength, splitting
tensile strength, modulus of rupture, static modulus, rebound number, ultrasonic
pulse velocity, dynamic modulus, initial surface absorption, density ,and total
absorption tests were investigated for all mixes at 7, 28, 90, and 120 days age.
Results of the destructive tests (compressive tensile strength, strength, splitting
modulus of rupture, and static modulus) and non–destructive tests (hammer,
ultrasonic pulse velocity, and dynamic modulus) are statistically analyzed by using
SPSS Ver.15 software to study the possibility of predicting the mechanical
properties of high performance concrete by using non–destructive tests. Simple and
multiple linear regression analysis of the obtained results leads to the proposed
statistical models for evaluating the compressive strength, splitting tensile
strength, modulus of rupture, and static modulus by using one or two or three of
the above mentioned non–destructive tests. Analysis of variance (ANOVA)
and t–test was also used to investigate the adequacy of the statistical models.