Kier John C. Aboy Mrs.
Cherry Flores, RN MN
Group A7
Title no. 101-S33
Fracture of Rock-Concrete Interfaces:
Laboratory Tests
and Applications
by J. M. Chandra Kishen and Victor E. Saouma
The fundamental understanding of the fracture behavior at the rock-concrete interface requires
evaluation of the fracture energy, which is an interface material property. As in concrete or rock,
fracture at the interface is characterized by a steady degradation of its structure. With increasing
degradation of the material (opening of the crack), less stress is transferred across the interface until
it is completely separated. The interface undergoes tensile softening. Wedge-splitting tests are
performed on the limestone-concrete interface to evaluate the Mode I fracture energy. These
experiments are followed by linear and nonlinear fracture mechanics based on numerical analysis to
obtain the fracture toughness. The fracture parameters so obtained are used in the fracture
mechanics-based analysis of a gravity dam for the determination of crack length between the concrete
dam and rock foundation.
Keywords: concrete; fracture; toughness.
INTRODUCTION
Dam safety programs are of utmost importance to society and call for combined use of
multidisciplinary efforts. The concept of safety should apply not only to the preplanning
stages of design but also to the post-operational and maintenance stages. Due to this,
recent years have witnessed a major research interest from the academic community in
fracture mechanics of concrete and a high concern from the engineering community and
power utility companies owning dams, in dam safety. From recent studies,1 it is well
understood that the concepts of fracture mechanics could be usefully applied for the failure
analysis of concrete dams. In a concrete dam, the interface between the concrete
superstructure and the rock foundation is one of the potential sites of crack formation and
subsequent failure. Not only do they contribute in weakening the mechanical strength, but
they also constitute conduits for water to seep through and exert uplift pressure. Hence, it is
important that proper mechanical behavior of this interface is understood in light of realistic
loading conditions. The fracture mode at an interface of dissimilar materials is often mixed.
Differences between elastic properties across an interface will generally disrupt the
symmetry even when the geometry and loading are otherwise symmetric with respect to the
crack. Mixed mode crack propagation involves the presence of in-plane normal and shearing
tractions near the front of an existing crack (notch). The stress and displacement fields near
the tip of a crack present between dissimilar materials are given in the Appendix. The
Federal Energy Regulation Commission (FERC) guidelines2 for the evaluation of hydroelectric
projects include provisions addressing the use of finite element analysis instead of hand
calculation methods. Therefore, besides theoretical considerations, valid fracture mechanics
material properties should be determined for dam concrete-rock foundation interface for the
use in fracture mechanics-based finite element models. Hence, it is vitally important to
conduct experiments on interface specimens to extract valid material properties.
This paper presents the wedge-splitting test3 performed on rock-concrete interface
specimens. The specific fracture
energy GF, which is the energy required to fracture a unit area of the interface, is evaluated.
In addition, numerical
analyses are performed on these wedge-splitting specimens using the concepts of linear and
nonlinear fracture mechanics to obtain the fracture toughness. Further, the fracture
parameter obtained in the wedge-splitting tests and their numerical analysis is used in the
analysis of the Greyrock gravity dam for determining the crack length at the dam-foundation
interface.
RESEARCH SIGNIFICANCE In this paper, the specific fracture energy, which is the energy
required to fracture a unit area, is determined experimentally for cracks at the concrete-
limestone interface. Further, through a numerical compliance analysis, the Mode I and Mode
II fracture toughnesses are computed. The fracture toughness and the fracture energy serve
as principal fracture properties in most of the linear and nonlinear fracture mechanics-based
numerical models, respectively.
EXPERIMENTAL PROCEDURE
Test specimen
The geometry of the wedge-splitting specimen is shown in Fig. 1. One half of this specimen
is rock while the other half is concrete. The type of rock used is Indiana limestone. The
limestone portion was machine-cut to its corresponding size and was placed on its side
inside a mold. Freshly mixed concrete (cement:sand:coarse aggregate: water =
1:2.2:2.8:0.45) was placed over the limestone, thus providing a cold-jointed interface. Before
casting the concrete portion, the interface portion of limestone was cleaned with a wire
brush. The molds were stripped after 2 days following casting and the specimens were cured
in a fog room with 100% humidity for more than 21 days. A total of 12 specimens was
prepared. The experiments were conducted in a closed-loop servohydraulic testing machine
under crack mouth opening displacement (CMOD) control. The rate of CMOD was maintained
around 1.4 × 10–4 mm/s. The CMOD was monitored using a clip gage, fixed at a level, where
the resultant splitting force acted on the specimen. A representative load versus CMOD
response diagram4 is shown in Fig. 2. In this figure, it is seen that unloading-reloading was
performed at shows the mean value of fracture energy obtained from different specimens.
Figure 3 shows the vertical load plotted against the CMODs for different material contacts.
The maximum splitting load and fracture energy for solid concrete, solid limestone, 5 and
limestone-concrete contacts are shown in Table 2. It should be mentioned that the crack
started at the notch and propagated all along the interface and, hence, the fracture energy
measured is that of the interface.
