Spalling Tests Research Articles

A study of the mechanical response of polycrystalline ice subjected to dynamic tension loading using the spalling test technique

Polycrystalline ice has been extensively investigated during the last decades regarding its mechanical behaviour for quasi-static loadings. Conversely, only few studies can be found on its dynamic behaviour and scientists suffer from a lack of experimental observation to develop relevant modelling at high strain-rate ranges. Dynamic experiments have already been conducted in compression mode using Hopkinson bar set-up. Regarding tension, experimental observations and measurements are scarce. The literature gives only approximated strength values. The knowledge of the latter is essential to design structures that may experience ice impact. The present study aims at providing the first reproducible experimental data of the tensile strength of polycrystalline ice subjected to dynamic tensile loading. To do so, a spalling test technique has been used for the first time on ice to apply tensile loading at strain-rates from 41 s−1 to 271 s−1. The experimental results show that the tensile strength is sensitive to the applied strain-rate, evolving from 1.9 MPa to 16.3 MPa for the highest applied loading rate.

Effects of Recycled Steel and Polymer Fibres on Explosive Fire Spalling of Concrete

Modern high-performance concrete, increasingly used in tunnels and other important infrastructure, is susceptible to explosive fire-induced spalling. To prevent fire spalling, modern codes recommend the use of small quantities (e.g. 2 kg/m3 as recommended by the Eurocodes) of polypropylene fibres in the concrete mix. This paper presents an experimental study investigating, the effect of cleaned recycled fibres extracted from end-of-life tyres on the explosive fire-induced spalling of concrete. This paper presents 24 spalling tests, indicating that recycled tyre polymer fibres, at dosages equal to or larger than 2 kg/m3, might help prevent fire spalling. Recycled tyre steel fibres also show the potential of preventing fire spalling damage by keeping spalled concrete attached to the heated surface, thus protecting the main steel reinforcement. The use of these fibres might lead to safe and sustainable fire spalling mitigation solutions.

Three-dimensional mesoscopic simulation of the dynamic tensile fracture of concrete

Despite its ubiquitous presence in the built environment, the dynamic tensile fracture behaviour of concrete under a high strain rate is still not completely understood. In this study, a three-dimensional meso-scale finite element model is developed to investigate the dynamic tensile fracture behaviour of concrete. On the meso-scale, concrete is regarded as a composite of aggregate, mortar and the interface transition zone (ITZ). The nucleation, coalescence, and propagation of cracks are modelled by pre-insert rate-dependent cohesive elements. Predicted results of the proposed numerical model fit well with experimental data in spall tests. The evolution of cracks in concrete under dynamic loading is explicitly presented. The micro-cracks of concrete are widely distributed before the initiation of macro-cracks. The results suggest that the inertia effect plays an important role in the dynamic fracture behaviour of concrete. Moreover, the influence of the mesostructure on the dynamic tensile strength is investigated. The dissipated fracture energy under dynamic loading is also indicated to be affected by the mesostructure of concrete.

Identification of strength and toughness of quasi-brittle materials from spall tests: a Sigma-point Kalman filter approach

ABSTRACTUnder impacts, monolithic quasi-brittle specimens fail by cracking if the magnitude of the induced stress waves exceeds their tensile strength. In spall tests, depending on the velocity of the striker hitting the specimen, either a partial damage or a whole crack formation can be induced. Since cracking occurs almost instantaneously in this dynamic process, experimental data need to be accurately and reliably filtered to identify the material strength and toughness ruling a cohesive law that describes the damaging/cracking event. To this aim, a Sigma-point Kalman filter approach is here discussed. The performance of the proposed methodology, in terms of constitutive parameter estimation and tracked specimen state, especially crack opening, is first quantitatively assessed through pseudo-experimental testing. It is shown that, if the stress pulse leads to a complete failure of the specimen and if the out-of-plane velocity is measured at the back surface of the impacted specimen, a noteworthy accurate and (almost) initialization-independent calibration of the cohesive model is obtained. Next, some results are provided by processing real experimental data relevant to two ceramic materials: SiC and BC.

Nonlocal formulation of the modified K&C model to resolve mesh-size dependency of concrete structures subjected to intense dynamic loadings

With the aid of commercial concrete material models, numerical simulation plays an increasingly important role in the analyses of dynamic responses of concrete structures subjected to intense dynamic loadings. Mesh-size dependency is a known problem of local material models used to simulate concrete due to strain softening, which is observed in numerous simulations. In this paper, the widely used regularization technique for quasi-static problems, smeared crack approach, is evaluated and proven to be ineffective for dynamic situations. To resolve the mesh-size dependency of concrete under dynamic loadings, the nonlocal formulation of the modified K&C (MKC) model proposed in our recent work is presented. The nonlocal MKC model is validated against four experiments, covering a wide range of stress/loading conditions, namely: Modified Split-Hopkinson Bar (spalling) test, Split-Hopkinson Tension test, Split-Hopkinson Pressure Bar test, and reinforce concrete beam subjected to blast loading. The numerical predictions are mesh objective and in good agreement with the corresponding experimental data.

Shockless spalling characterization of ceramics in a 1D-stress state

Ceramics are particularly interesting as protective materials due to their high compressive strength. However the maximum tensile stress withstood by these materials is usually lower by one order of magnitude. To study the dynamic strength in tension, spalling tests are commonly performed. In this work, a new spalling configuration is presented to characterize brittle materials such as ceramics at ultra-high strain-rates of about 103–104 s−1 in a one-dimensional stress state. To do so, a specific test is designed in which the ceramic specimen is a bar with a 3 mm radius. This geometry is chosen to ensure a one-dimensional stress state in the cylindrical specimen avoiding wave dispersion during the propagation of the loading pulse. Finally, the first experimental validation tests conducted at the CEA-Gramat with the pulsed power generator called GEPI are reported.

Comparative study of tensile tests based on Hopkinson bar for recycled aggregate concrete

To comparatively study the tensile properties and fracture patterns of recycled aggregate concrete with various replacement percentages (i.e. 0%, 25%, 50%, 75%, and 100%) of recycled coarse aggregate, the dynamic direct tensile tests, splitting tests, and spalling tests of recycled aggregate concrete in the strain-rate range of 100–102 s−1 were carried out using large diameter (75 mm) split Hopkinson tensile bar and pressure bar. Test results show that for recycled aggregate concrete, the quasi-static direct tensile strength is more marvelous than its quasi-static splitting strength. When recycled coarse aggregate replacement percentage is 0%–75%, the replacement percentage impact minimally on the quasi-static tensile strength of recycled aggregate concrete. In dynamic tensile tests, there exists apparent difference between the dynamic direct tensile strength and dynamic splitting. The dynamic tensile strength of recycled aggregate concrete increases with the increase of average strain-rate in all three kinds of tests. The average strain-rate affects the damage form of recycled aggregate concrete, which indicates that the recycled aggregate concrete has obvious rate sensitivity. There shows no obvious regularity between the dynamic tensile strength and the recycled coarse aggregate replacement percentage. And the indirect tensile strength calculation method used in this article offers the theoretical basis for the engineering application of recycled aggregate concrete.

Effects of steel fiber content and type on dynamic tensile mechanical properties of UHPCC

Ultra-high performance cementitious composites (UHPCC) has become the most prospective construction materials for both civil and military protective structures. In order to facilitate the structural design and calibration/validation of the constitutive model, the dynamic tensile (spalling) behavior of UHPCC is experimentally studied. By using a 50 mm-diameter conic variable cross-sectional Split Hopkinson Pressure Bar (SHPB), the dynamic spalling test on Φ48mm × 400 mm cylindrical UHPCC specimen is conducted. The tensile strain rate ranges from 14.3 s−1 to 214.8 s−1, and two typical steel fibers (micro-straight and hooked) with three volume fractions of 0%, 1.0% and 2.0% are investigated. The effects of steel fiber content and type as well as the strain rate on the dynamic spalling strength of UHPCC are experimentally assessed. It indicates that the dynamic spalling strength of UHPCC increases obviously with the strain rate rising up to ∼110 s−1, and also improves with increasing the content of both two typical steel fibers. When the strain rate increases higher than 60 s−1, the micro-straight steel fiber has relatively better effect on improving the dynamic spalling strength than the hooked one with the volume fraction of 1.0%. Furthermore, corresponding to each steel fiber content and type, the empirical dynamic increase factors (DIFs) for the dynamic tensile strength of UHPCC are formulated.

A homogenized localizing gradient damage model with micro inertia effect

The conventional gradient enhancement regularizes structural responses during material failure. However, it induces a spurious damage growth phenomenon, which is shown here to persist in dynamics. Similar issues were reported with the integral averaging approach. Consequently, the conventional nonlocal enhancement cannot adequately describe the dynamic fracture of quasi-brittle materials, particularly in the high strain rate regime, where a diffused damage profile precludes the development of closely spaced macrocracks. To this end, a homogenization theory is proposed to translate the micro processes onto the macro scale. Starting with simple elementary models at the micro scale to describe the fracture mechanisms, an additional kinematic field is introduced to capture the variations in deformation and velocity within a unit cell. An energetic equivalence between micro and macro is next imposed to ensure consistency at the two scales. The ensuing homogenized microforce balance resembles closely the conventional gradient expression, albeit with an interaction domain that decreases with damage, complemented by a micro inertia effect. Considering a direct single pressure bar example, the homogenized model is shown to resolve the non-physical responses obtained with conventional nonlocal enhancement. The predictive capability of the homogenized model is furthermore demonstrated by considering the spall tests of concrete, with good predictions on failure characteristics such as fragmentation profiles and dynamic tensile strengths, at three different loading rates.

Concrete propensity to fire spalling: testing and observations

This article presents results of fire spalling tests on small concrete slabs and studies of material parameters that may increase its occurrence. Experimental techniques enabling to study and determination of material features are presented and discussed. Experimental studies on spalling behaviour of elements were carried out on seven different concrete mixes with constant content of cement paste and mortar. Research aimed at determining influence of the following parameters: w/c ratio (0.30; 0.45; 0.60), cement type (CEM I, CEM III) and type of aggregates (riverbed gravel, granite, basalt) on fire concrete spalling. Paper discusses also the influence of cold rim that forms while testing slab-like element is subjected to one-side heating.

Investigation of Spalling Damage in Ultra-High Performance Concrete Through X-ray Computed Tomography

Ultra-high performance concretes (UHPC) are increasingly used to build protective structures such as headquarters, nuclear power plants or critical civil engineering structures. However, under impact or contact detonation, concrete is exposed to high-rate tensile loadings that can lead to intense damage modes. Such complex damage modes need to be correctly characterised. When a UHPC sample is subjected to a dynamic tensile loading by means of the spalling technique the post-mortem pattern shows a large number of fractures that cannot be seen with a classical observation of the external face (inner crack network). In the framework of the Brittle’s CODEX chair project, the fracturing process in spalled samples of UHPC is investigated with X-ray computed tomography. The tensile loading is applied thanks to a spalling technique that is based on the reflection of a compressive wave on a free boundary. The concrete samples are entirely scanned using X-ray tomography prior spalling test to identify the initial microstructure, and post spalling test to analyse the damage pattern. Image analysis tools are used in both steps. The main fracturing properties are related to the microstructure of the tested concrete.

On the Processing of Spalling Experiments. Part I: Identification of the Dynamic Tensile Strength of Concrete

The spalling technique based on the use of a single Hopkinson bar put in contact with the tested sample has been widely adopted as a reliable method for obtaining the tensile response of concrete and rock-like materials at strain rates up-to 200 s− 1. However, the traditional processing method, based on the use of Novikov acoustic approach and the rear face velocity measurement, remains quite questionable due to strong approximations of this data processing method. Recently a new technique for deriving cross-sectional stress fields of a spalling sample filmed with an ultra-high speed camera and based on using the full field measurements and the virtual fields method (VFM) was proposed. In the present work, this topic is perused by performing several spalling tests on ordinary concrete at high acquisition speed of 1Mfps to accurately measure the tensile strength, Young’s modulus, strain-rate at failure and stress–strain response of concrete at high strain-rate. The stress–strain curves contain more measurement points for a more reliable identification. The observed tensile stiffness is up-to 50% lower than the initial compressive stiffness and the obtained peak stress was about 20% lower than the one obtained by applying the Novikov method. In order to support this claim, numerical simulations were performed to show that the change of stiffness between compression and tension highly affects the rear-face velocity profile. This further suggests that the processing based only on the velocity “pullback” is quite sensitive and can produce an overestimate of the tensile strength in concrete and rock-like materials.

Numerical modelling of dynamic spalling test on rock with an emphasis on the influence of pre-existing cracks

This article deals with numerical modeling of rock fracture under dynamic tensileloading and the related prediction of dynamic tensile strength. A special emphasis is laid on theinfluence of pre-existing natural microcrack populations as well as structural (articial) cracks.For this end, a previously developed 3D continuum viscodamage-embedded discontinuity modelis employed in the explicit dynamic nite element simulations of the spalling test. This modelis capable of modelling the eect of natural microcracks populations always present in rocks aswell as to capture the strain rate hardening eect of quasi-brittle materials. In the numericalsimulations of spalling test on Bohus granite, it is shown that the model can predict the pull-pack velocity of the free end of the intact rock sample and the eect of structural cracks witha good accuracy. According to the simulations, the effect of microcrack populations, modeledhere as pre-embedded discontinuity populations, is weaker than the corresponding eect underquasi-static loading

3D polycrystalline discrete element method (3PDEM) for simulation of crack initiation and propagation in granular rock

A three-dimensional Voronoi tessellation model based on the distinct element method (DEM) is proposed to model the representative part of the microstructures of granular brittle rocks. Regularization is employed to decrease the frequency of polyhedrons with large edge ratio and contributes to a higher efficiency for element meshing. Sensitivity analyses are performed for a series of micro contact parameters in accordance with the macro responses observed in laboratory experiments (e.g. the uniaxial compression test, the Brazilian disc test and the triaxial compression test). Verifications by simulating the spalling test and plate impact test indicate that the 3D polycrystalline discrete element method (3PDEM) can be employed for efficiently simulating nonlinear mechanical behaviors, large deformation, strain softening and rock dynamics.

Failure-delay effect in destruction of steel samples under spalling conditions

Dynamic spalling tests have been run on two batches of 30KhN4M steel samples. Experimental data have been processed with the classical technique based on solution of the elastic wave equation. Three samples have been revealed that demonstrated the failure-delay effect under testing. The incubation-time criterion has been used to show the conditions of emergence of failure delay with the example of triangular loading pulses. A rate strength curve has been constructed for the other samples. It has been shown that the limiting strengths under dynamic loads considerably differ for samples from different batches despite the same chemical composition and static strength.

Strain rate sensitivity of the tensile strength of two silicon carbides: experimental evidence and micromechanical modelling.

Ceramic materials are commonly used to design multi-layer armour systems thanks to their favourable physical and mechanical properties. However, during an impact event, fragmentation of the ceramic plate inevitably occurs due to its inherent brittleness under tensile loading. Consequently, an accurate model of the fragmentation process is necessary in order to achieve an optimum design for a desired armour configuration. In this work, shockless spalling tests have been performed on two silicon carbide grades at strain rates ranging from 103 to 104 s-1 using a high-pulsed power generator. These spalling tests characterize the tensile strength strain rate sensitivity of each ceramic grade. The microstructural properties of the ceramics appear to play an important role on the strain rate sensitivity and on the dynamic tensile strength. Moreover, this experimental configuration allows for recovering damaged, but unbroken specimens, giving unique insight on the fragmentation process initiated in the ceramics. All the collected data have been compared with corresponding results of numerical simulations performed using the Denoual-Forquin-Hild anisotropic damage model. Good agreement is observed between numerical simulations and experimental data in terms of free surface velocity, size and location of the damaged zones along with crack density in these damaged zones.This article is part of the themed issue 'Experimental testing and modelling of brittle materials at high strain rates'.

3D meso-mechanical modeling of concrete spall tests

Numerical simulation of concrete spall tests is carried out in this paper by means of a 3D meso-mechanical model. The model is developed by assuming that concrete can be considered as a three-phase material containing cement mortar, aggregates and interfacial transition zone (ITZ) and the difficulty of simulating the very thin ITZ layer is overcome by employing element reconstruction technique (ERT) which results in great improvement for the computational efficiency. It transpires that the present numerical simulation results are in good agreement with the data obtained for the spall tests of concrete in terms of failure pattern, strain and dynamic tensile strength. It also transpires that the present model can reproduce the experimentally observed phenomenon that the amplitude of reflected tensile wave increases with increasing propagation distance which is caused largely by damage growth in concrete materials.

Dynamic localization of damage and microstructural length influence

In the present contribution a localization analysis is performed for a two-scale dynamic damage model. The damage evolution law considered here results by homogenization from micro-crack propagation criteria and involve a characteristic length of the microstructure. For the one-dimensional dynamic system, we study the localization of strain and damage using analytical and numerical methods. An instability analysis based on harmonic perturbations of a homogeneous state for the linearized equations reveals wave dispersion properties and an intrinsic length of the model is deduced in the high frequency regime. The explicit dependence of this intrinsic length on the microstructural length and the damage level is obtained. Numerical simulations of dynamic tests illustrate the localization of damage, strain-rate sensitivity of the tensile strength and the influence of the size of the microstructure. The model predictions are compared with experimental results for spalling tests on concrete specimens. It is shown that the damage model is able to reproduce the observed influences of the strain rate and the size of the microstructure on the dynamic tensile strength.

Mesoscopic simulation of the dynamic tensile behaviour of concrete based on a rate-dependent cohesive model

We present a two-dimensional meso-scale finite element model for simulating the dynamic tensile behaviour of concrete. Aggregate, mortar and interface transition zone (ITZ) are explicitly modelled at the meso-scale. Rate-dependent cohesive elements, which take into account the effects of the viscosity occurring in the fracture process, are dispersedly pre-inserted to simulate the nucleation, coalescence, and propagation of cracks. The proposed model is validated by comparing with the experimental data of spall tests. The evolution of microcracks of concrete under dynamic loading is presented. The influences of the fracture parameters and the mesostructure on the dynamic tensile strength are investigated. The results suggest that the property of ITZ plays an important role on the dynamic tensile behaviour of concrete. The tensile strength and dissipated fracture energy under dynamic loading are affected by the mesostructure of concrete as well.

Spall Strength of Resistance Spot Weld for QP Steel**Supported by the National Natural Science Foundation of China under Grant Nos 11372149, 11572164 and 11502074, and the K. C. Wong Magna Fund in Ningbo University.

The spall tests under the plane tensile pulses for resistance spot weld (RSW) of QP980 steel are performed by using a gun system. The velocity histories of free surfaces of the RSW are measured with the laser velocity interferometer system for any reflector. The recovered specimens are investigated with an Olympus GX71 metallographic microscope and a scanning electron microscope (SEM). The measured velocity histories are explained and used to evaluate the tension stresses in the RSW applying the characteristic theory and the assumption of Gathers. The spall strength (1977–2784 MPa) of the RSW for QP980 steel is determined based on the measured and simulated velocity histories. The spall mechanism of the RSW is brittle fracture in view of the SEM investigation of the recovered specimen. The micrographs of the as-received QP980 steel, the initial and recovered RSW of this steel for the spall test are compared to reveal the microstructure evolution during the welding and spall process. It is indicated that during the welding thermal cycle, the local martensitic phase transformation is dependent on the location within the fusion zone and the heat affected zone. It is presented that the transformation at high strain rate may be cancelled by other phenomenon while the evolution of weld defects is obvious during the spall process. It may be the stress triaxiality and strain rate effect of the RSW strength or the dynamic load-carrying capacity of the RSW structure that the spall strength of the RSW for QP980 steel is much higher than the uniaxial compression yield strength (1200 MPa) of the martensite phase in QP980 steel. Due to the weld defects in the center of the RSW, the spall strength of the RSW should be less than the conventional spall strength or the dynamic load-carrying capacity of condensed structure.