ESTIMATION OF CONCRETE STRENGTH BASED ON PORE STRUCTURE CONSIDERATION : Part 1 Correlation between Strength and Pore Structure

Abstract

The purpose of this study is to establish a new method for the estimation of concrete strength by means of pore structure consideration. In this paper, the correlation between concrete strength and pore structure was investigated, and the calculated value, by using the empirical equations derived in this study, and the value obtained by using already known equations which express the relation between strength and pore structure were compared. Three tests, as shown in Table 1, were carried out; one preliminary test (Series I) and two fundamental tests (Series II, III). In the preliminary test, several mix proportions of concrete with or without air entraining agent were chosen under combinations of various water cement ratio, slump and aggregates of different sources. The preliminary test was carried out to judge the factors which diffuse the correlation between compressive strength and porosity of concrete. The object of the fundamental tests was to derive the empirical equations for estimation of concrete strength. Test of Series II was carried out to investigate the influence of age and conditions of curing on the concrete strength, and test of Series III was carried out to study the effect of the kind of cements, respectively. Strength tests were carried out by using 10φ×20cm cylinderical test specimens. The investigation of pore structure of concrete was carried out by using mercury porosimeter and small crushed pieces of concrete collected from specimens after strength test. Weight loss of concrete at a temperature of 600℃ was also measured, and then the dissolving test of concrete using 10% HCI solution was carried out after weight loss test. These test results of crushed sample were applied to determine several indicators concerning pore structure and degree of hydration of cement. As the results of preliminary test, the porosity was expressed as the value of "TPV" which was the amount of total pore volume divided by the total weight of sample, and it was found that the correlation between concrete strength and TPV differed by the kind of aggregates, and the slump of concrete (Fig. 3). Therefore, it was found that the modification of the difference between concrete strength and TPV was very difficult, even if some other factors, which were able to be obtained from pieces of concrete, would be applied for the modification. On the other hand, in the value of the case where "ETPV", which was expressed as the value of total pore volume divided by the weight of hardened cement paste in the sample, a simple proportional relation was obtained under the same age or same curing conditions (Fig. 4). On this account, it seemed that the latter relation could be adopted to determine the empirical equation for the estimation of concrete strength. The empirical equation was presented by means of multiple regression analysis. In this analysis, four predictor variables, such as three indicators concerning the fineness and shape of pore structure ("Me" and "Re"), the degree of hydration ("Wc" or "Age") and ETPV were used to estimate concrete strength simultaneously. For the regression analysis, 140 pieces of data (shown in Table 1) were used. The analytical result was given by equation (4), which led to coefficient of correlation of 0.984. And the application of Age instead of Wc yielded equation (5) where coefficient of correlation was 0.981. The relation between observed value and calculated value of compressive strength by equation (4) is shown in Fig. 5. Fig. 6 shows the relations obtained by equations (6)′, (8)′, (9)′ and (10)′ respectively, and the value of coefficient of correlation are shown in Table 5. Thus, the calculated value of compressive strength by equation (4) gave a better correspondence to the observed value than that by these existing equations. These empirical equations derived in this paper were

(View PDF for the rest of the abstract.)