TY - JOUR
T1 - Size effect in Brazilian split-cylinder tests. Measurements and fracture analysis
AU - Bazant, Zdenek P.
AU - Kazemi, Mohammad Taghi
AU - Hasegawa, Toshiaki
AU - Mazars, Jacky
N1 - Copyright:
Copyright 2004 Elsevier B.V., All rights reserved.
PY - 1991/5
Y1 - 1991/5
N2 - Size effect in the split-cylinder (Brazilian) tensile test is studied experimentally and analyzed theoretically. Tests of a very broad size range, 1.26, were conducted on cylindrical discs of constant thickness made from concrete with aggregate of a maximum size of 5 mm (2 in.). The results confirm the existence of size effect and show that up to a certain critical diameter dt, the curve of nominal strength versus diameter approximately agrees with the law proposed by Bazant for the size effect caused by energy release due to fracture growth. For larger sizes, there appears to be a deviation from the size-effect law, although large scatter of the test results does not permit a strong conclusion in this regard. The trend of the size-effect curve is probably an approach to a horizontal asymptote with or without a reversal of slope of the size-effect curve. The reversal can be explained by a modification of the size-effect law in which the crack length at failure ceases to increase in proportion to the diameter and may remain constant for sizes larger than a certain characteristic size. The approach to a horizontal asymptote can be explained by plastic slip-on triaxially confined wedge-shaped regions below the platens.
AB - Size effect in the split-cylinder (Brazilian) tensile test is studied experimentally and analyzed theoretically. Tests of a very broad size range, 1.26, were conducted on cylindrical discs of constant thickness made from concrete with aggregate of a maximum size of 5 mm (2 in.). The results confirm the existence of size effect and show that up to a certain critical diameter dt, the curve of nominal strength versus diameter approximately agrees with the law proposed by Bazant for the size effect caused by energy release due to fracture growth. For larger sizes, there appears to be a deviation from the size-effect law, although large scatter of the test results does not permit a strong conclusion in this regard. The trend of the size-effect curve is probably an approach to a horizontal asymptote with or without a reversal of slope of the size-effect curve. The reversal can be explained by a modification of the size-effect law in which the crack length at failure ceases to increase in proportion to the diameter and may remain constant for sizes larger than a certain characteristic size. The approach to a horizontal asymptote can be explained by plastic slip-on triaxially confined wedge-shaped regions below the platens.
UR - http://www.scopus.com/inward/record.url?scp=0026157526&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0026157526&partnerID=8YFLogxK
M3 - Article
AN - SCOPUS:0026157526
SN - 0889-325X
VL - 88
SP - 325
EP - 332
JO - ACI Materials Journal
JF - ACI Materials Journal
IS - 3
ER -