A modified slant shear test designed to enforce adhesive failure
Highlights
► The Slant Shear Test (SST) is discussed. ► Major drawbacks are identified: possibility of cohesive failures. ► A Modified-SST is proposed to overcome these disadvantages. ► It is shown that the Modified-SST enforces adhesive failure.
Introduction
Structural concrete-to-concrete interfaces are present in reinforced concrete members when certain strengthening operations of existing structures are performed or when precast concrete members with cast-in-place parts are used. Presently, it is widely accepted that three load transfer mechanisms contribute for the shear strength of the interface between concrete layers cast at different times: (i) adhesion, understood as a chemical connection between old and new concrete and as aggregate interlock after debonding; (ii) friction, which appears when relative slippage between concrete layers takes place in the presence of normal stresses to the interface; and (iii) dowel action, which results from the deformation of the reinforcement crossing the interface due to the relative slippage between concrete layers.
There are several tests available to evaluate the behaviour and/or the strength of concrete-to-concrete interfaces. These tests can be classified according to the stress resultant at the interface in: (i) axial; (ii) bending; or (iii) shear (Table 1) tests.
From all the tests referred to, the Slant Shear Test (SST) (Fig. 1) is the most commonly used to assess the adhesion between concrete layers. In the SST, the interface is subjected to both shear and compressive stress states, which allows a relatively uniform stress distribution at the interface. The wide spread use of the SST is mainly due to the simplicity of the experimental setup and to the high sensitivity of the test to variations in influencing parameters [26], [27], [28], [29], [30], namely: (i) the specimens’ geometry (angle of the shear plane and ratio height/width of the cross section); (ii) the casting procedure; (iii) the preparation of the interface (surface roughness); (iv) the differential stiffness between concrete layers; (v) the differential shrinkage between concrete layers; and (vi) the use of a bonding agent at the interface.
The SST was proposed in 1976 by Kriegh [31] to evaluate the bond strength of epoxy-based resins using cylinder-shaped specimens of 150 mm diameter by 300 mm height and with the interface at 30° to the vertical. In 1978, this test was adopted by Tabor [32] for studying the bond between concrete-to-concrete interfaces using prismatic specimens. Currently, there are several SST standards (see Table 2), using different specimen geometries.
In spite of its popularity, this test presents as major drawback the fact of leading to two possible failure mechanisms: (i) adhesive, interface debonding (Fig. 2); and (ii) cohesive, crushing of the weakest concrete (Fig. 3). In the latter case, only a lower estimate of the interface strength can be obtained, since failure depends on the compressive strength of the weakest concrete.
According to Austin et al. [35], who investigated the SST failure modes, the normal/shear stress ratio is controlled by the interface angle, having for this reason a crucial influence on the SST failure mode and, thus, on the ultimate load. Therefore, it is suggested to test different interface angles, for each surface treatment, in order to obtain a bond failure envelope.
According to Naderi [36], adhesion under zero normal stress can only be estimated with the SST if, at least, three interface angles are considered. Furthermore, this method does not agree with the results of other methods, namely, the friction transfer method. Additionally, it was also stated that the coefficient of variation presented by this method is relatively high.
Although the SST might not be adequate to assess the pure shear strength, it is possible to define bond failure envelope. This envelope can be obtained using SST results in shear (σ, τ) combined with: (i) results of a tensile test (ft,i); and (ii) the pure shear strength (τ0) estimated using the Mohr–Coulomb criterion (Fig. 4). If cohesive failure occurs, the envelope can also be determined, nevertheless the obtained value is a lower bond of the interface strength, as represented by the dashed line in Fig. 4. More details are given in [37].
Clímaco and Regan [26] adopted a Mohr Coulomb failure criterion to select an angle to ensure adhesive failure. 223 specimens have been tested adopting three different angles (0°, 20° and 26.7°). The obtained failure types were divided in three categories: (i) concrete failure, adjacent to the interface; (ii) shear failure, along the interface with simultaneous concrete failure neighbouring the interface; and (iii) cohesive failure with crushing of the weakest concrete. Even though a 20° angle was defined to always obtain adhesive failures, cohesive failures were still reported. Therefore, it can be stated that although the interface angle plays an important role in the failure mode, it is not always possible to obtain adhesive failure only by adequately selecting this parameter.
In Júlio et al. [38], SST specimens were produced in order to verify the influence of differential stiffness in the shear strength of concrete-to-concrete interfaces. Keeping the substrate concrete unchanged with a compressive strength of 30 MPa, three different concrete mixtures were used for the added concrete layer, with the following compressive strengths: 30, 50 and 100 MPa. The surface roughness was increased in all specimens by sand-blasting. All specimens with both halves of the same compressive strength (30/30) presented adhesive failures, whereas all the others (30/50 and 30/100) presented cohesive failures, showing the importance of differential stiffness on the failure mechanism.
In Santos and Júlio [30], two failures mechanisms (cohesive and adhesive) are referred to. It was observed that the number of cohesive failures increases with the surface roughness. It is also stated that the number of cohesive failures increased with the differential stiffness. Differential shrinkage was also analysed by means of different curing conditions and different ages between substrate and added layer. It was also concluded that this parameter influences the failure type.
In summary, according to previous studies [26], [28], [30], [35], [38], it can be stated that the failure mechanism of the SST is mainly influenced by: (i) the interface angle; (ii) the interface roughness; (iii) the differential stiffness between concrete layers; and (iv) the differential shrinkage between concrete layers. Moreover, it can be stated that adhesive failure cannot be prevented only by adequately defining the interface angle.
As it was previously discussed, it is not possible to enforce adhesive failures by exclusively optimising the geometry of slant shear specimens. A different approach is herein presented which consists on a Modified-SST (M-SST), specifically designed to enforce adhesive failure and to ensure a uniform stress distribution at the interface.
Section snippets
Concept
In a structural connection the weakest component has an important role in both the strength and the failure type. In the case of slant shear specimens, made of two concrete layers, three components can be identified: (i) interface; (ii) substrate concrete; and (iii) concrete overlay. Cohesive failure occurs whenever the weakest component is the lowest compressive strength concrete. Therefore, adhesive failures can be only enforced if both parts of the slant shear specimen are strengthened. This
Numerical analysis
In order to check if the stress distribution at the interface is influenced (or not) by the added reinforcement, a simplified numerical study was performed [40]. Two linear elastic plane stress models have been considered (with and without reinforcement) using 492 bilinear finite elements for concrete (Fig. 6a). At the interface, 15 two-node zero thickness finite elements, connecting each half of the specimen, have been introduced. Finally, the reinforcement was simulated with 197 linear truss
Experimental validation
To experimentally validate the efficiency of the M-SST, pointed out by the numerical analysis, the most unfavourable situation has been selected: monolithic specimens.
First, the M-SST specimens were casted considering two situations, plain and reinforced concrete (see Fig. 8). Two specimens were adopted for each situation. The adopted concrete mixture was, by cubic meter: (i) 320 kg of ordinary Portland cement (Type II 32.5R); (ii) 421 kg of fine sand (0–2 mm); (iii) 181 kg of medium sand (0–4 mm);
Summary and conclusions
The Slant Shear Test (SST) has been proposed in 1976. After that, several modifications were introduced by different researchers. Nevertheless, none of these prevent cohesive failures. Even when the geometry of the SST specimen is optimised, namely by adequately defining the interface angle with the vertical, cohesive failures can still be obtained. Aiming to overcome this disadvantage of the SST, i.e. to always enforce adhesive failures, the authors designed an innovative test, called Modified
Acknowledgement
This research project has been funded by the Portuguese Science and Technology Foundation (FCT) with reference PTDC/ECM/098497/2008.
References (42)
- et al.
Application of tension softening diagrams to evaluation of bond properties at concrete interfaces
Eng Fract Mech
(2000) - et al.
The effects of surface preparation on the fracture behaviour of ECC/concrete repair system
Cem Concr Compos
(2000) - et al.
Improving the bonding between old and new concrete by adding carbon fibres to the new concrete
Cem Concr Res
(1995) - et al.
Comparison of methods for evaluating bond strength between concrete substrate and repair materials
Cem Concr Res
(2005) - et al.
Concrete-to-concrete bond strength. Influence of the roughness of the substrate surface
Constr Build Mater
(2004) - et al.
Shear bond testing of concrete repairs
Cem Concr Res
(1999) - et al.
Influence of added concrete compressive strength on adhesion to an existing concrete substrate
Build Environ
(2006) - BS EN 1542. Products and systems for the protection and repair of concrete structures – test methods – measurement of...
- ASTM C1583. Standard test method for tensile strength of concrete surface and the bond strength or tensile strength of...
- BS 1881: Part 207. Testing the interface bond strength between existing and new concrete. British Standards...
The pull-off partially destructive test for concrete
Laboratory and field testing of concrete bond development for expedited bonded concrete overlays
ACI Mater J
Cited by (53)
Bond performance between cementitious grout and normal concrete subjected to freeze–thaw damage
2023, Construction and Building MaterialsCharacterization of Substrate-to-Overlay Interface Bond in Concrete Repairs: A Review
2023, Construction and Building MaterialsAutomatic strength assessment of the virtually modelled concrete interfaces based on shadow-light images
2022, Construction and Building MaterialsInterfacial bond in concrete-to-concrete composites: A review
2022, Construction and Building MaterialsExperimental study on factors influencing ERC-NSC interface bonding performance
2022, Engineering Failure AnalysisExperimental testing correlation of the different aged concrete bond strength involving the new tilted cube test (TCT)
2022, Journal of Building Engineering