Analysis of acoustic emission characteristics of ice under three point bending

Highlights

• Investigation of acoustic emission (AE) response of ice beam under three point bending at varying strain rates.

• Analysis of various AE characteristics (hits, amplitude, wave duration, AE energy) with loading behaviour of ice.

• b-value estimation from AE amplitude distribution using Gutenberg-Richter (GR) relation.

• Fracture mode identification in terms of RA parameter and average frequency (AF).

Abstract

Acoustic emission (AE) is one of the promising techniques to understand the damage mechanism within a material and structural health monitoring non-invasively. The transient elastic energy released during the microscopic failure within a material is termed as acoustic emission. AE signals carry information about the mechanism and intensity of damage within a material. In this study, AE signatures were recorded during the mechanical loading of notched ice beams subjected to three point bending tests at three different strain rates i.e. 1 × 10⁻⁵ s⁻¹, 1 × 10⁻⁴ s⁻¹ and 1 × 10⁻³ s⁻¹. Various AE characteristics such as hits, amplitude, counts, rise time, energy release rate etc. were analysed in relation to the loading behaviour of ice beams. The AE amplitude distribution data was analysed to estimate b-value and its temporal variation throughout loading for all the three strain rate experiments. A decreasing trend in b-value with increasing bending stress was observed in all the three tests. Characterization of dominant failure mode was also attempted through the analysis of RA parameter and average frequency (AF) which indicates tensile mode as the dominant mode of fracture during three-point bending.

Introduction

Acoustic emission is a widely used non-destructive technique for assessment of progressive damage in various materials such as rocks, concrete, wood, composites etc. at both laboratory and larger scales (Lockner, 1993; Ohtsu, 1996; Vidya Sagar and Raghu Prasad, 2012; Bucur, 2006; Guarino et al., 1998; Amitrano et al., 2005). The stress waves within a material are the results of various non-reversible processes such as plastic deformation, crack opening, coalescence and growth of micro-cracks etc. Acoustic signal generated during the failure process propagates through the bulk and is detected at the surface by use of piezo transducers. The number of AE signals acquired by the detection system is related to the failure density or crack events occurred in the material. The nature of AE signal depends on the source and type of failure mode (Carpinteri et al., 2013). Various characteristics of AE waveform such as amplitude, rise time, energy and frequency carry information about the damage mode and fracture process (Carpinteri et al., 2013). Vidya Sagar and Raghu Prasad, 2012, recorded the energy released during fracture process in concrete and suggested that AE could be an efficient tool for fracture energy monitoring provided the attenuation of AE wave is not appreciable within the material. In fact, in most of the cases the AE wave undergoes attenuation during its propagation in material. Therefore, the measured AE energy can be in proportion and a small fraction of the released fracture energy during failure process. AE amplitude is one of the important parameter since it is related to the magnitude of failure event. However, the analysis of individual AE amplitude values can sometimes be misleading because the signal undergoes attenuation while propagating in the medium. Even a very high amplitude AE signal emitted at far could be recorded by the sensor as low amplitude (Shiotani et al., 2007). Therefore, relative distribution of AE amplitude is generally used to correlate the deformation process undergoing within the material. The b-value proposed by Gutenberg and Richter (1944) gives the relationship between frequency and magnitude of AE events emanated from a deformational process in a material. Investigations carried out by previous researchers suggest that there is a decrease in the b-value with increasing damage/micro-fractures within the material subjected to loading (Main et al., 1989; Lockner, 1993; Reiweger et al., 2015). Classification of cracking mode is another important area which is derived from the analysis of AE parameters i.e. RA parameter and average frequency (AF) (Ohno and Ohtsu, 2010; Shahidan et al., 2013; Li and Du, 2016). It was found that high RA and low AF values correspond to the shear crack whereas low RA and high AF correspond to the tensile crack. Researchers suggested that analysis of RA and AF along with the b-value can be used as a possible indicator of accelerating damage within a material which can lead to catastrophic failure.

Ice is one of the naturally occurring materials in the earth's cryosphere. Physical and mechanical properties of ice greatly affect the formation and dynamics of glaciers. Also, it is a cost effective construction material in cold regions for various applications such as embankment, highway and runways etc. (Masterson, 2009; Li and Du, 2016). Apart from the construction activities, ice in glaciated terrain poses a severe threat to the persons venturing in those areas. It provides a low friction sliding base to the newly fallen snow which is helpful in the formation of avalanches. Moreover, the Himalaya, having the highest glaciated peaks of the world witnesses hazards due to frequent ice cliff breaks (ice avalanches), crevasse openings etc. every year (Singh et al., 2013). In one such event 10 Indian soldiers got buried under deep snow and ice in Siachen glacier avalanche in Karakoram range of Indian Himalaya (Wikipedia, 2016). Prediction of such failure requires understanding of the damage mechanism of ice under different loading and temperature conditions. The plastic deformation of ice also contributes in the movement and break-up of glaciers over time (Andrews, 1985). Ice matrix is also the main component of snow structure and its mechanical properties determine the mechanical behaviour of snow (Chandel et al., 2014; Pinzer, 2009). The mechanical properties such as yield stress, fracture energy of ice are needed to implement the damage initiation and evolution law for micromechanical modelling of snow deformation (Chandel et al., 2014). In the past, various researchers have used acoustic emission to investigate the deformation of ice (Gold, 1960; Zaretsky et al., 1976; St Lawerence and Cole, 1982; Sinha, 1985, Sinha, 1996; Sinha et al., 2012). Most of them have used the ring down count as the main AE parameter which is equivalent to the AE ‘counts’ in current AE systems. Other parameters such as amplitude, acoustic energy, b-value etc. were not correlated with the deformation process of ice. Recently, Li and Du (2016) attempted to correlate the acoustic emission signature to the ice deformation during compressive and three point beam loading tests in a laboratory setup. Authors analysed the AE parameters such as amplitude, rise time, counts, wave duration and energy in relation to the loading of samples. AE amplitude distribution data was further analysed for the calculation of b-value and its variation during the course of loading. They further suggested that analysis of AE features can be used to assess the onset of damage within the material. However, the analysis presented by them was based on only one displacement rate which was 2 mm/min which corresponds to strain rate: 1.6 × 10⁻⁴ s⁻¹.

From the above discussions, we can say that previously not much research was undertaken to study acoustic emission characteristics from ice. Therefore there is a need to investigate the deformation behaviour of ice to improve its understanding in terms of various AE parameters. In this work, we have taken the studies further by subjecting ice samples to bending load at three different strain rates and corresponding acoustic signature were detected and analysed to correlate the damage behaviour of ice. An analysis of dissipated energy components is carried out and comparison was made with the recorded acoustic energy. The variation of b-value, RA and AF was also analysed in relation to the evolution of damage.