Evaluation of Ratio between Splitting Tensile Strength and Compressive Strength for Concretes up...

Discussion by Kenneth W. Day

FACI, Consulting Concrete Technologist, Melbourne, Australia.

The discusser would anticipate that the ratio of splitting to compressive strength would be substantially dependent on coarse aggregate particle shape. The paper, however, gives no information as

Could the authors please comment?

AUTHORS' CLOSURE

The authors would like to thank the discussers: Y?zer for his stimulating remarks and Day for his keen interest in the paper.

Reply to Nabi Y?zer

The purpose of the research was to derive a reliable relationship between the ratio of splitting tensile strength to cylinder compressive strength and the compressive strength for concretes up to 120 MPa. Moreover, the relationship was used to verify Johnston's strength criterion in intact rocks for the ultimate strength of very high-strength concrete under triaxial compression. The authors entirely agree with the discusser that, in the case of concretes subjected to effect of chloride ion, use of any of the relationships in which the splitting strength is estimated on the basis of the compressive strength will considerably overestimate the splitting strength of concrete. Recalling the study of Rezansoff and Corbett,41 a chloride-based setting and accelerating admixture reduces the splitting tensile strength as compared with normal-no accelerated-concrete. In this discussion, our equation, Eq. (7),36 introduced a modification factor a^sub c^ to account for the effect of chloride on the tensile strength of concrete as displayed in Fig. B. The value of a was obtained by means of least-squares regression analysis. For this analysis, the data were gathered from Rezansoff and Corbett's41 and Issa et al.'s42 studies. According to the analysis, the statistical values of a^sub c^ were determined as the average a^sub c^= 0.84, the standard deviation S = 0.05, and the coefficient of variation V = S/a^sub c^ x 100 = 5.9%. From the value of V, it can be concluded that there is a very limited variation in the value of a^sub c^. In other words, the mean a^sub c^ of 0.84 can be taken as "constant of proportionality" for Eq. (7) correlating the splitting tensile strength with the compressive strength of concrete subjected to chloride effect.

In passing, it should be mentioned that the influence of strength range, curing condition, time, and concentration of chloride ion on the value of a^sub c^ are still unknown. In brief, for the assessment of the influences in question, the detailed experimental test program and statistical analysis are required.

To evaluate the accuracy of the modified Eq. (7) against experimental results, the data (Table A) reported by Y?zer and Ak?z38 were used as control data. The splitting tensile strengths predicted from the equation under review are compared with the experimental data in Fig. C by making use of the 1:1 technique. The average derivations (? = -10.1%; ? = +7.7%) obtained can be considered to be acceptable.

Reply to Kenneth W. Day

It is well known from the literature that the aggregate characteristics (type, particle shape, surface texture) have an important effect on the mechanical properties.45,46 In particular, the bond between the aggregate and the paste is a key factor affecting the tensile strength of concrete. This mechanical interlocking, or bond strength, can be enhanced by using clean aggregates with angular shape and rough texture.47 Hence, the tensile strength of concrete can be increased at a given water-cement ratio. As far as the compressive strength is concerned, this mechanical property is less affected by the variation in aggregate characteristics45 compared with the splitting tensile strength.

The ratio between splitting tensile strength and compressive strength f^sub tsp^/f^sub c^ is likely to be affected by aggregate properties in question. The majority of data used in our regression analysis belongs to the concretes made with crushed aggregates and the resulting equation, Eq. (7),36 is valid for the concretes made from these aggregates.

To quantitatively determine the effect of aggregate with round shape and smooth texture on the ratio of f^sub tsp^/f^sub c^, an additional statistical analysis (Fig. D) was performed through Franklin and King's experimental data45 (the concretes made with a constant water-cement ratio of 0.50, aggregate of rounded/irregular shape and smooth texture- uncrushed gravel-maximum aggregate size D^sub max^ = 19 mm, various curing ages, the compressive strengths of 150 mm cube were converted to the strength of f 150 x 300 mm cylinder via the conversion factor of 0.848). From Fig. D it is evident that, for a given compressive strength (10 MPa = f^sub c^ = 50 MPa), the ratio of f^sub tsp^/f^sub c^ for concretes with uncrushed gravels results in a reduction of almost 16% compared to the concretes containing crushed rock aggregates. Briefly, in the concretes with uncrushed gravel for a given compressive strength, there is a decrease in the splitting tensile strength. Its main reason is the weaker bond developed between the aggregate and the surrounding hydrated cement paste.

REFERENCES

41. Rezansoff, T., and Corbett, J. R., "Influence of Accelerating Admixtures on Strength Development of Concrete under Wet and Dry Curing," ACI Materials Journal, V. 85, No. 6, Nov.-Dec. 1988, pp. 519-528.

42. Issa, M. A.; Issa, M. A.; Faraj, M.; and Reddy, K., "Mechanical Properties and Durability of High-Performance Protecrete Concrete," Progress Report by University of Illinois submitted to Applied Concrete Technology, Feb. 1999, 39 pp.

43. Torrent, R. J., and Brooks, J. J., "Application of the Highly Stressed Volume Approach to Correlated Results From Different Tensile Tests of Concrete," Magazine of Concrete Research, V. 37, No. 132, Sept. 1985, pp. 175-184.

44. Kadlecek, V.; Modry, S. S.; and Kadlecek, J., "Size Effect of Test Specimens on Tensile Splitting Strength of Concrete: General Relation," Materials and Structures, V. 35, Jan.-Feb. 2002, pp. 28-34.

45. Franklin, R. E., and King, T. M., "Relations Between Compressive and Indirect Tensile Strengths of Concrete," RRL Report LR 412, Road Research Laboratory, UK, 1971.

46. Ar?oglu, E.; Girgin, Z. C.; and Ar?oglu, N., "Evaluation of the Ratio of Splitting Tensile Strength to Compressive Strength in Concrete," Ready Mixed Concrete Magazine, Jan.-Feb. 2002, pp. 58-63. (in Turkish)

47. Hannant, D. J., "The Tensile Strength of Concrete: A Review Paper," The Structural Engineer, V. 50, No.7, July 1972, pp. 253-258.

48. Ar?oglu, E.; Ar?oglu, N.; and Girgin, Z. C., "Shape-Size Effect in Normal and High-Strength Concrete Samples," Ready Mixed Concrete Magazine, Jan.-Feb. 1999, pp. 40-51. (in Turkish)