Multiphysics modeling of the initiating capability of detonators. II. Booster initiation

Abstract

Detonators are explosive devices used for the initiation of secondary explosives in commercial and military applications. They are characterized by their initiating capability which is a critical factor for their safe and effective use but challenging to assess accurately. In this two-part study, we employ numerical simulations to investigate the blast wave generated by detonators and examine their initiating capability. The first part follows the European underwater test which evaluates detonators in isolation (direct method). The second part, presented here, investigates detonators placed within a receiving explosive charge (indirect method). Specifically, the detonator is placed inside a booster device which contains secondary explosive and together form an initiating system used to ignite mining blastholes.The physical system is modeled using a multiphysics methodology to accurately capture the response of the materials present in the configuration (explosives, metals and fluids). The reactive model is extended to account for shock desensitization where explosives become more difficult to initiate after the passage of weak shock waves. The variability of the blast wave generated by detonators, observed in the first part of the study, can lead to partial desensitization resulting in pockets of unreacted explosive which inhibit booster initiation and performance. The computational implementation is extensively validated and calibrated against experiment before employed for the study of booster initiation by a range of detonators. Results show that the booster is susceptible to shock desensitization which occurs in varying degrees for different types of detonators and can significantly impact the performance of the initiating system.

Reacting condensed phase explosives in direct contact

“…As the detonation travels faster than the inert shock, it eventually reaches the vertical wall of the rigid well, producing a region within the explosive sample where no reactions have occurred; this results in a dead zone that is observed to persist for the rest of the simulation. The results are in good agreement with the study by DeOliveira et al [12] and the numerical solutions reported in Ioannou and Nikiforakis [16], where high pressure LX-17 products are used to initiate the main sample of explosive. Indeed, the use of UF-TATB products does not qualitatively affect the overall results, but illustrates the applicability of the proposed model to scenarios involving two distinct explosive materials.…”

“…In order to capture desensitisation effects and the formation of dead zones, the desensitisation model proposed by DeOliveira et al [12] is included in the formulation with the same parameters used therein. The computational domain is discretised with a coarse mesh of size 160 mm and two levels of refinement, with refinement factors ×4 and ×2 are used to adaptively increase the resolution around the detonation front, material interfaces and rigid boundaries; this results in an effective resolution of ∆x = ∆y = 20 µm which is consistent with that used by Ioannou and Nikiforakis [16]. Numerical results are shown in Fig.…”

“…Level-set based algorithms have been employed in the past to address the challenges of more than one reacting explosives in direct contact by Ioannou and Nikiforakis [15,16], in the context of resolved simulations of detonators, and by Kim et al [19], who investigated the enhanced blast effect generated by the combustion of metal particles dispersed in a high explosive. Diffuse interface approaches have been extensively used for the numerical simulation of explosives and their interaction with confiners.…”