Zp. Bazant et M. Jirasek, R-CURVE MODELING OF RATE AND SIZE EFFECTS IN QUASIBRITTLE FRACTURE, International journal of fracture, 62(4), 1993, pp. 355-373
The equivalent linear elastic fracture model based on an R-curve (a cu
rve characterizing the variation of the critical energy release rate w
ith the crack propagation length) is generalized to describe both the
rate effect and size effect observed in concrete, rock or other quasib
rittle materials. It is assumed that the crack propagation velocity de
pends on the ratio of the stress intensity factor to its critical valu
e based on the R-curve and that this dependence has the form of a powe
r function with an exponent much larger than 1. The shape of the R-cur
ve is determined as the envelope of the fracture equilibrium curves co
rresponding to the maximum load values for geometrically similar speci
mens of different sizes. The creep in the bulk of a concrete specimen
must be taken into account, which is done by replacing the elastic con
stants in the linear elastic fracture mechanics (LEFM) formulas with a
linear viscoelastic operator in time (for rocks, which do not creep,
this is omitted). The experimental observation that the brittleness of
concrete increases as the loading rate decreases (i.e. the response s
hifts in the size effect plot closer to LEFM) can be approximately des
cribed by assuming that stress relaxation causes the effective process
zone length in the R-curve expression to decrease with a decreasing l
oading rate. Another power function is used to describe this. Good fit
s of test data for which the times to peak range from 1 sec to 250000
sec are demonstrated. Furthermore, the theory also describes the recen
tly conducted relaxation tests, as well as the recently observed respo
nse to a sudden change of loading rate (both increase and decrease), a
nd particularly the fact that a sufficient rate increase in the post-p
eak range can produce a load-displacement response of positive slope l
eading to a second peak.