Modified mixed-mode caustics interpretation to study a running crack subjected to obliquely incident blast stress waves

https://doi.org/10.1016/j.ijimpeng.2021.103821Get rights and content

Highlights

  • Incident P and S waves, as well as reflected waves, were visualized.

  • A modified mixed-mode caustics interpretation was proposed and verified.

  • More precise mixed-mode SIFs and crack-tip positions were determined.

  • Transient stress superposition and delayed redistribution occur in the crack tip.

Abstract

Crack-wave interaction is common in blast, producing mixed-mode cracks when stress waves are oblique to cracks. How to measure mixed-mode crack-tip stress under blast stress waves is difficult. To this end, optical caustics method was employed to study a running crack in a PMMA plate subjected to obliquely incident blast stress waves, by which stress problem was transformed into optical problem. Incident P and S waves, as well as reflected waves, were visualized as fringe patterns, and mixed-mode crack-tip stress was represented by a caustics pattern (shadow spot). However, mixed-mode caustics patterns under P waves were not consistent with the classical interpretation, hence a modified mixed-mode caustics interpretation was proposed and then verified by P-wave compression and tension loading cases. In modified interpretation, an iteration procedure was proposed to obtain modified stress intensity factors, i.e. KI and KII, and crack-tip positions which were more precise than those by the classical interpretation. P-wave compression phase decreased KI and crack velocity sharply but produced the largest KII, while the following P-wave tension phase had an opposite effect. S waves and reflected waves produced higher KI and crack velocity with lower KII. KII and crack velocity histories explained curved crack path and microscopic fracture surface. Finally, the reason for the success of the modified interpretation was discussed, and it is indicated that transient stress superposition and delayed redistribution in the crack tip are the mechanism underlying the modified interpretation.

Introduction

The interaction between dynamic cracks and stress waves is common in blast applications, such as mining and tunnelling engineering [1,2]. Due to arbitrary crack propagation in multi-hole blasting, blast-induced cracks are always subjected to obliquely incident blast stress waves, thus mixed-mode cracks appear which are more general than mode I cracks in practice. For a mixed-mode crack, the local crack-tip stress is dominated by mixed mode I and mode II stress intensity factors (SIFs) which determine crack propagation. For a better understanding of mixed-mode crack propagation in blast, it is of fundamental importance to evaluate mixed-mode crack-tip stress under blast stress waves.

Mixed-mode crack-tip stress caused by stress waves was studied by dynamic fracture theory [3]. Based on stress superposition, Freund [4], Achenbach [5] and Ma [6] obtained analytical solutions for mode I and mode II SIFs of a crack subjected to obliquely incident stress waves. Specifically, Freund [4] restricted crack remaining straight, while Achenbach [5] and Ma [6] considered a kinked crack deflecting off the straight line. Due to complex blast stress waves, these analytical solutions were not applied to real blast cases, but they provided a critical information, i.e. stress superposition in the crack tip upon arrival of stress waves.

Due to complexity in analytical approaches, laboratory blast experiments are indispensable which were used to measure shock wave pressure [7], stress wave form [8] and dynamic fracturing behavior [9], [10], [11]. In blast experiments, visual observations of blast stress waves and cracks provide useful information for understanding the interaction between stress waves and cracks. Hence, optical methods combining with transparent testing materials are good choices, such as photoelasticity and caustics methods [12]. Despite a wide range of applications, both photoelasticity and caustics methods face challenges in obtaining SIFs from crack-tip optical patterns in blast. Historically, photoelasticity was introduced into blast experiments much earlier than caustics method, due to its advantages in rich full-field data for stress waves and crack-tip stress [13,14]. However, it is difficult to obtain SIFs from highly dense and distorted isochromatic fringes in the crack tip upon arrival of blast stress waves [15]. This challenge is also presented in Ref. [16]. About 30 years later than photoelasticity, caustics method was introduced into blast experiments by Yang [17], to obtain crack-tip SIFs by measuring the diameter of the caustics pattern (shadow spot). Compared with photoelasticity, caustics method provides simple caustics patterns, avoiding dense fringes in the crack tip [18]. The shape of the caustics pattern reflects stress distribution in the crack tip. Caustics patterns are always in an approximately circle shape under general dynamic loading, such as tensile machine [19,20], gas gun projectile [21], drop weight [22] and out-of-plane impact [23], indicating K dominated stress field in the crack tip. However, recent blast experiments [24,25] show that caustics patterns are distorted in an ellipse shape under blast-induced P wave loading, indicating that crack-tip stress is non-K dominated and SIFs obtained by the classical caustics interpretation have large errors. In order to better interpret distorted caustics patterns, a modified caustics interpretation for a mode I crack subjected to blast-induced P waves was built, by which more precise crack-tip SIFs were obtained [24,25].

However, in blast experiments using caustics method, mode I and mode II SIFs of a mixed-mode crack are still obtained by the classical mixed-mode caustics interpretation, which may cause errors if mixed-mode caustics patterns are distorted under blast stress waves. Therefore, a modified mixed-mode caustics interpretation is required. The present work is the extension of the previous work by authors [26] where a mode I crack was subjected to normally incident blast stress waves.

In this paper, a running crack in a PMMA plate subjected to obliquely incident blast stress waves was investigated by caustics method and high-speed photography. Incident P and S waves, as well as reflected waves, were visualized as fringes and distinguished based on different fringe patterns. Upon arrival of blast-induced P waves, crack tip was surrounded by a mixed-mode caustics pattern which was not consistent with the classical caustics interpretation. To solve this problem, a modified mixed-mode caustics interpretation was proposed based on the superposition of geometrical optics and verified by P-wave compression and tension loading cases. An iteration procedure was proposed to obtain KI and KII in the modified caustics interpretation. SIFs and crack velocity histories were obtained in the whole crack propagation, which explained variations of crack path and microscopic fracture surface well. Finally, the mechanism underlying the modified mixed-mode caustics interpretation was discussed.

Section snippets

Specimen

PMMA (polymethyl methacrylate) specimen with 320 mm × 192 mm × 5 mm dimension shown in Fig. 1(a) was used in the present blast experiments due to its easy preparations and good transparency, which was widely used in various blast experiments [2,8,10,11,24,25]. A pre-crack with 60 mm in length and 0.5 mm in width was first produced by a laser cutter in the center of a long side of the specimen and then sharpened by a utility knife. Under the pre-crack, a field of view of 52.02 × 75.87 mm2 was

Post-blast crack path and surface

All 20 specimens were tested in the same experimental setup, yielding similar crack paths including straight and curved parts. For brevity, Fig. 3 shows a typical post-blast crack path and fracture surface in specimen 16. It is observed that a straight crack appears under drop weight and blade loading before blast stress waves arrive, while it changes to a curved crack when subjected to blast stress waves, shown in Fig. 3(a). Correspondingly, optical microscopic observations of fracture surface

Modified interpretation for mixed-mode caustics patterns

Because classical mixed-mode caustics interpretation is not consistent with experimental results, a modified interpretation is proposed here. The inspiration of the modified interpretation comes from the principle of caustics method, i.e. the superposition of light deflection in the crack tip based on geometrical optics. When a crack is subjected to blast-induced P waves, there are two kinds of light deflection in the crack tip: one caused by the crack and the other caused by P waves, which are

Verification of modified mixed-mode caustics interpretation

For verification, the modified mixed-mode caustics interpretation is employed in P-wave compression and tension loading cases in specimen 5 and 10. Firstly, parameters ω and z0/f are determined by the iteration procedure proposed in Fig. 13. Then, analytical mixed-mode caustics patterns are plotted and compared with experimental results, yielding a good agreement. Finally, KI and KII are obtained and compared with those from the classical mixed-mode caustics interpretation. In addition,

Mechanism underlying the modified interpretation

Through verifications in P-wave compression and tension loading cases, the modified mixed-mode caustics interpretation has been proved to be more consistent with experimental results than the classical interpretation. However, this conclusion is only based on the comparison between crack-tip optical patterns, but the mechanism underlying the modified mixed-mode caustics interpretation is not known, which will be discussed here.

The modified mixed-mode caustics interpretation is based on the

Conclusions

Mixed-mode crack propagation under stress wave loading is common in blast. In order to measure mixed-mode crack-tip stress, a running crack in a PMMA plate subjected to obliquely incident blast stress waves was studied experimentally by caustics method and high-speed photography. Through caustics method, stress problem was transformed into geometrical optics problem. However, classical mixed-mode caustics interpretation was not suitable for experimental results under P wave loading, hence

Data availability

The data required to support this paper is available from authors upon request.

CRediT authorship contribution statement

Peng Qiu: Conceptualization, Methodology, Investigation, Writing - original draft, Funding acquisition. Zhongwen Yue: Resources, Supervision, Funding acquisition. Renshu Yang: Supervision. Yang Ju: Supervision, Funding acquisition.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (no. 51974318, 51727807), China Postdoctoral Science Foundation (no. 2020M670524), and State Key Laboratory for Geomechanics and Deep Underground Engineering (no. SKLGDUEK2026).

References (39)

Cited by (0)

View full text