This paper presents a comparative investigation of the high-velocity perforation behavior of a composite projectile and a conventional ogive-nosed projectile. Based on oblique perforation experiments on multi-layered steel plates at high velocities, the asymmetric evolution of the projectile nose and the characteristic failure morphologies of the targets are obtained. Combined with numerical simulations, the projectile-target interaction mechanisms are elucidated. Observations of projectile motion during multi-layered perforation reveal that the variations in pitch angle and angle of attack of the composite projectile are significantly smaller than those of the single projectile. Further comparative analysis indicates that the enhanced attitude stability originates from the protective mechanism of the cap. Under high-velocity oblique perforation, the cap dissipates a substantial amount of energy through its own plastic deformation, effectively redistributing the energy across different regions of the nose. This significantly attenuates the compressive and shear loads transmitted to the main body, thereby suppressing stress concentration and asymmetric deformation on the distal side of the nose and ensuring high structural integrity and minimal deformation of the main nose. The mechanism underlying the smaller attitude deflection is that the plastic deformation of the cap attenuates the compressive and shear forces acting on the main body, thereby reducing the deflection moment and angular acceleration experienced by the projectile and enhancing attitude stability. This stabilization advantage becomes increasingly pronounced with increasing obliquity. This study on the perforation behavior of composite projectiles provides mechanistic insights for the design and analysis of similar projectile configurations.

This study investigated the influence of steel fiber content on the post-cracking mechanical behavior of ultra-high-performance fiber-reinforced concrete (UHPFRC) and quantified the evolution of the macroscopic fiber efficiency factor ηe at different fiber contents. A systematic experimental program was conducted, including fiber pullout tests at different inclination angles and direct tensile and flexural tests on notched specimens with fiber contents ranging from 0% to 3%. The results showed that both the peak bridging stress and flexural strength increased with increasing fiber content, whereas ηe exhibited a non-monotonic trend. Specifically, ηe increased at low fiber contents but decreased at higher fiber contents, with a reduction of approximately 30.6% as the fiber content increased from 1% to 3%. Based on these results, a post-cracking constitutive model incorporating fiber bridging stress and matrix softening stress was developed. In this model, ηe was determined by inverse analysis to quantify the variation in fiber efficiency associated with coupled mechanisms, including the group effect, matrix spalling, and fiber-induced matrix confinement. The proposed model was implemented in LS-DYNA using a user-defined material subroutine (UMAT) within the framework of the fixed smeared crack model. The numerical simulations accurately reproduced the crack propagation process during fracture and showed good agreement with the experimental results. These findings demonstrate that the proposed model can effectively characterize the tensile softening and fracture behavior of UHPFRC under monotonic static loading while accounting for the evolution of fiber efficiency.

Explosion mechanics originated from China’s “Two Bombs and One Satellite” project and was named by Hsue-Shen Tsien in 1963. It primarily studies the occurrence patterns of explosions, as well as the utilization, control, and protection of their mechanical effects. As an interdisciplinary subject, it involves fluid mechanics, solid mechanics, physics, and chemistry. The core of this discipline lies in understanding how high-power-density energy is transmitted through shock waves and other means within an extremely short time, leading to high-speed flow, large deformation, and even the failure of media. Since its inception, explosion mechanics has made a series of significant contributions to China’s national security and defense, aerospace, advanced manufacturing, and major engineering projects. Four distinguished scientists—Che-Min Cheng (2012), Kai-Jia Cheng (2013), Ze-Shan Wang (2017), and Qi-Hu Qian (2018)—have successively won the National Highest Science and Technology Award. In recent years, this field has continued to make breakthroughs following the “Four-Extreme” strategy, focusing on energetic materials, detonation technology, impact physics, and damage and protection. Moreover, the number of researchers engaged in explosion mechanics has grown rapidly. Nevertheless, the development of explosive mechanics still faces challenges, including the need for full-chain innovation from fundamental research to engineering applications, and the paradigm shift from Galileo’s “experiments-mathematics” approach to artificial-intelligence-empowered scientific research. In this Roadmap, we aim to provide a broad overview of recent and potential future activities in explosive mechanics and to present a development roadmap by gathering contributions from scientists with diverse backgrounds. We believe that these contributions will significantly stimulate the generation of original achievements and the advancement of state-of-the-art experimental, theoretical, and numerical tools. We also hope, this Roadmap will further establish the strategic importance of explosion mechanics as both a “shield” and a “spear” for safeguarding national security, and as a “masterful hand” in solving major engineering problems.

Hydrocarbon fuels are a major focus of fuel application and research due to the high volumetric calorific value. The addition of ClF3O with strong oxidizability as an initiator in fuels is considered to be an effective method to improve the overall thermal decomposition rate. This study employs the quantum mechanics molecular dynamics (QM-MD) and density functional theory (DFT) to investigate the thermal decomposition reaction of C7H16 initiated by ClF3O, focusing on reaction mechanism, decomposition temperature, and product distribution. The decomposition reactions channels of ClF3O/C7H16 mainly occur through two types of reactions: initiating decomposition by ClF3O and decomposition of fuel itself. From the QM-MD simulations, we find ClF3O can accelerate C7H16 decompositions, since the H-abstraction on C1 site of C7H16 can be initiated by ClF3O and other small molecular radical at relatively lower temperatures, which is more likely to occur due to its low activation energy compared to the C-C bond cleavage. The H· and CH3· radicals generated by dehydrogenation and C-C bond cleavage in the self-decomposition stage further promoted the decomposition of C7H16 and produced a variety of alkane radicals. Meanwhile, the product HF could promote the self-decomposition of C7H16 because of its acid properties. This work might provide important theoretical insights for fluorine-containing compounds as initiators to induce thermal decomposition reactions of hydrocarbon fuels.

Deeply buried and ultrahigh-strength protective structures often require multiple sequential penetration-explosion cycles to be effectively neutralized. This study focuses on the secondary penetration behavior of ultrahigh-performance concrete (UHPC) targets after an initial penetration and explosion sequence, a subject that has received limited systematic attention. First, a series of penetration-explosion-penetration tests was performed on UHPC targets, with systematically varying secondary impact locations to examine their effect on penetration depth and local failure characteristics. Experimental results reveal that secondary penetration performance varied significantly with impact position, showing distinct differences in both the increase in penetration depth and the degree of projectile redirection across tested locations. In addition, a computational model incorporating the restart method was developed and rigorously validated through comparisons with experimental data. Furthermore, a systematic parametric study was conducted to examine the influence of impact location, velocity, and accumulated material damage on secondary penetration behavior, accompanied by a discussion of the underlying physical mechanisms.

Accurate prediction of sympathetic reaction in cylindrical charges is essential for ammunition safety design. However, the underlying mechanisms involving near-field coupled loads remain inadequately understood. This study employs a combined experimental and numerical approach to investigate the sympathetic responses of cylindrical charges subjected to the impact of detonation products and fragments. The results indicate that the near-field reflected loads of bare RX-1 cylindrical charges are primarily governed by the dynamic pressure of the detonation products. The inertial confinement provided by the casing significantly reduces the peak dynamic pressure compared with that of bare charges at an equivalent expansion radius. As the expansion radius increases, the number of fragments exceeding various kinetic energy thresholds initially increases and then decreases, whereas the total kinetic energy within the effective coverage area decays monotonically. For bare charges, sympathetic reaction is primarily governed by the dynamic pressure of the detonation products. In contrast, for cased charges, the dominant mechanism shifts from the integral impact of a fragment cluster to the coordinated impact of discrete fragments as the expansion radius grows. Compared with fragment clusters, the contribution of detonation products to sympathetic reaction shows a stronger distance dependence, leading cased charges to generate a wider region of graded reaction intensities than bare charges. This study clarifies the contribution of near-field loads to sympathetic reaction, thereby providing a critical basis for developing predictive assessment models.

To address the longstanding challenge of accurately evaluating the dynamic fracture toughness of ceramic materials, a new mode I dynamic fracture testing method was developed based on the conventional split-Hopkinson pressure bar (SHPB) technique. This approach introduced a miniature fracture specimen specifically designed to ensure pure mode I loading, along with a custom fixture system that enabled stable and repeatable dynamic fracture experiments on alumina ceramics with varying loading rates. The combined experimental-numerical method was used to obtain the variation of the mode I dynamic stress intensity factor at the crack tip under different loading rates. Fracture initiation time was obtained with high precision using the strain gauge method, allowing for the determination of mode I dynamic fracture toughness. To further validate the accuracy of the measured fracture initiation time, high-speed photography was employed to capture the entire failure process in real time and corroborate the onset of fracture of the tested specimens. The results show that as the applied loading rate increases from 0.45 TPa·m1/2·s−1 to 1.83 TPa·m1/2·s−1, the dynamic fracture toughness of alumina ceramics rises significantly from 8.39 MPa·m1/2 to 15.76 MPa·m1/2, indicating a pronounced strengthening effect induced by higher loading rates. Meanwhile, the crack initiation time decreases notably with increasing loading rate. Fractographic analysis using scanning electron microscopy reveals a clear fracture mode transition behavior. Under lower loading rates, the fracture of alumina ceramics predominantly exhibits intergranular fracture features. Under higher loading rates, the fracture shows a mixed-mode fracture involving both intergranular and transgranular features. This transition is attributed to the activation and propagation of more micro-defects under higher rates, resulting in increased microcracking. The emergence of this mixed fracture mode is associated with greater energy dissipation, which fundamentally contributes to the increase in mode I dynamic fracture toughness. The proposed method offers a robust framework for accurately assessing the mode I dynamic fracture properties of ceramic materials.

In response to the thermal protection requirements of ammunition under extreme thermal shock conditions, this study systematically investigated the effects of aramid pulp (AF) on the specific thermal protection properties of intumescent flame-retardant polyurea (PUA) by modifying PUA with AF In combination with an intumescent flame retardant. The results indicated that, compared with conventional intumescent flame-retardant PUA composites, the composites incorporating AF exhibited a lower limiting oxygen index (LOI) and a higher heat release rate in the microscale combustion calorimeter (MCC) test. However, in the cd ne calorimeter and oxyacetylene ablation tests, the addition of low levels of AF (0.5 wt% and 1.0 wt%) further reduced the heat release and significantly enhanced the ablation resistance of the materials. The rational explanation proposed in this paper for this specific thermal protection phenomenon will contribute to the further development of the flame-retardant mechanisms of one-dimensional materials and their applications in the field of thermal protection coatings for safe ammunition. [GRAPHICS]

With the rapid adoption of lithium-ion batteries (LIBs) in energy-storage and transportation systems, increasingly complex operating conditions heighten the fire and explosion hazards of thermal runaway (TR). To suppress its early flame spread and provide effective early rescue time, in this study, a multilayer perforated-plate flame arrester was designed. GC–MS was employed to analyze battery-generated gases and Fluent simulations of jet-flame velocity are used to determine arrestor parameters, and the structure is optimized via the throttling effect. Experimental verification is conducted thereafter. The results show that the gaseous products primarily consist of CO₂, CO, H₂, and C₂H₄, with total yields of 3.23 mol and 4.8 mol for cells at 75% and 100% state of charge (SOC), respectively. The simulated peak flame velocity of the ejected gases reaches 21.53 m/s. Considering manufacturing constraints and cost, a single-layer quenching plate with an aperture of 0.4 mm, a pitch of 0.3 mm, and a thickness of 0.8 mm was selected, giving a maximum quenching velocity of 4.86 m/s for one layer. Based on the throttling effect, a multilayer quenching-plate structure was designed; simulations indicate that three layers reduce the flame velocity to 4.53 m/s, while six layers reduce it further to 3.57 m/s. Experimental results confirm that the flame arrestor effectively suppresses both flames and sparks. Under the front-facing configuration, 75% SOC and 100% SOC conditions require five and seven layers, respectively; under the side-facing configuration, only three layers are required, and spark emission decreases with increasing numbers of quenching layers.

To investigate the effects of gas transport on ignition response of polymer-bonded explosives (PBXs) during slow cookoff, this study develops a thermo-mechanically coupled pyrolytic gas transport porous model. The model resolves dynamically coupled physical fields across thermal, mechanical, gas transport, and chemical reaction throughout the slow cook-off. Results are consistent with Sandia Instrumented Thermal Ignition (SITI) experiments for PBX 9501 in sealed and vented systems. It is revealed that porosity evolves in two stages: the first stage is dominated by thermal expansion, while the second is governed by decomposition. Gas transport pathways are controlled by the coupled evolution of porosity and temperature. Convective heat transfer from gas products contributes less of the total thermal transport, while the concentration effect of gas product within pores is identified as the dominant factor in ignition delay. This study provides a foundational framework for understanding gas transport in the slow cookoff of polymer-bonded explosives.

Rapid assessment of damage under combined penetration-explosion loading remains challenging due to the high cost of conventional simulations and experiments. Data-driven models improve efficiency but often lack physical consistency and struggle to capture the damage evolution process of concrete under extreme loading. This study introduces GAT-ImpactNet-an attention-enhanced graph neural network designed to advance engineering informatics by enabling real-time, physics-guided damage assessment for defense-related structural analysis. The model leverages high-fidelity numerical simulations, validated through projectile penetration and explosion tests on ultra-high-performance concrete targets, to establish a reliable dataset. GAT-ImpactNet incorporates physical priors via the loss function, allowing the graph attention mechanism to effectively capture complex structural interactions while enhancing prediction accuracy and computational efficiency. Validation against LS-DYNA simulations shows relative errors of 4.84% for penetration depth and 3.92-4.19% for blast cavity dimensions. Attention visualizations further demonstrate the model's interpretability by highlighting critical structural interactions. This work contributes a novel, physics-guided deep learning framework that integrates fundamental engineering principles with neural networks, offering a robust informatics tool for real-time safety evaluation under extreme dynamic loads.

ABSTRACT Due to the increased oxidizing gas concentration in the detonation products, the addition of oxidizing agents to explosives could enhance the burning rate and the reaction degree of the aluminum powder, which would further improve the metal acceleration capabilities of the explosives. The metal acceleration capabilities of CL‐20‐based aluminized explosives that contained oxidizers were investigated by means of standard cylinder tests. The effects of the oxidizers on the crossover time and the reaction degree of the aluminum powder were analyzed by replacing the aluminum powder with lithium fluoride for comparison. Based on the sources of the cylinder‐wall velocity, an empirical expression that relates the explosive composition to the cylinder‐wall kinetic energy was fitted. The results show that the detonation velocity and aluminum concentration of the CL‐20‐based aluminized explosives were linearly correlated, and that the detonation velocities of the oxidizer‐containing explosives could be set equivalent to those of oxidizer‐free explosives within a certain margin of error. However, with the addition of the oxidizer, the relationship between the detonation pressure and the concentration of the base explosive changed. The experimental results show that the addition of the oxidizer advanced the crossover time by 4.81 µs and 1.66 µs for aluminum concentrations of 5% and 30%, respectively. According to the fitted relationship, it was predicted that the maximum value of the cylinder‐wall kinetic energy could be obtained when both the aluminum powder and the oxidizer had concentrations of 10%.

The blast simulator based on high-speed impact has the advantages of low cost, strong repeatability and easy data acquisition, and is expected to become a promising supplement to traditional blast tests. However, there is a lack of theoretical model describing the dynamic response of impact system to guide the setting of loading conditions and the evaluation of loading effects. This study focuses on the impact system of a metal/PWG composite impact module (MPCIM) impacting reinforced concrete (RC) beam, aims to develop an accurate dynamic model based on equivalent degree of freedom model for rapid prediction of the impact load and structural response of RC beam. Firstly, regardless of the influence of the impacted structure, the quasi-static and dynamic compression numerical simulations for PWG with different dimensions were carried out. The nonlinear dynamic response model of PWG considering the dimension and impact velocity of MPCIM was developed by combining machine learning methods. Then, based on Euler-Bernoulli beam theory and the impact process of MPCIM and RC beam, a resistance function of RC beam incorporating the dimension of MPCIM was derived. Finally, a novel dynamic model of the impact system that considers the characteristics of the RC beam, the MPCIM dimension, and the dynamic response of PWG was established and verified by numerical simulation results. The results show that the model can accurately predict the impact load and structural response of RC beam impacted by MPCIM.

Understanding defect evolution and ignition in polymer-bonded explosives (PBXs) during slow cookoff is critical for thermal safety. This study develops a predictive model coupling gas transport with a micromechanics-based pressurization framework. A key innovation is the explicit distinction between pre-existing micropores and gas-induced opening-mode microcracks. We demonstrate that microcrack tip propagation dictates the onset of permeability, while the crack opening displacement regulates subsequent gas flow. The model successfully captures the localized pressure accumulation — a phenomenon driven by transport resistance — which significantly accelerates thermal decomposition. Validation against SITI and ODTX experimental data confirms the model's predictive accuracy and physical consistency. Furthermore, uncertainty quantification reveals high numerical robustness, showing that ignition time is minimally sensitive to microcrack parameters. Finally, statistical analysis of mesoscopic heterogeneity indicates that increased structural non-uniformity not only delays the median ignition time but also markedly amplifies its stochastic dispersion. This work provides a novel theoretical framework for gas-induced defect evolution, offering a robust tool for the thermal safety assessment of energetic materials.

The effects of strain rate, temperature and stress state on the plastic flow and failure behavior of Ti-5553 (Ti-5Al5Mo-5V-3Cr) alloy are studied separately. The results indicate that the alloy exhibits strong sensitivity to strain rate and temperature, with the yield and flow stresses increasing as strain rate rises, but its strength and workhardening capacity weakened at elevated temperatures. The material shows the highest flow stress in uniaxial tension but the lowest in simple shear, due to the stress state effect induced by the evolution of dislocation density and grain boundary migration. Fractographic analysis reveals that under different stress states, the failure behavior gradually transitions with increasing strain rates from ductile fracture to different fracture modes involving dimples, shear facets or adiabatic shear bands. To capture these behaviors for engineering design and calculation, two phenomenological and one physically-based constitutive models are revised to incorporate the effects of strain rate, temperature, and stress state simultaneously. The performance of the models in capturing work hardening, thermal sensitivity, and strain rate effects of the material is analyzed comprehensively.

2,4-Dinitroanisole (DNAN)-based melt-cast explosives, particularly the DNAN/Octahydro-1,3,5,7-tetranitro1,3,5,7-tetrazocine (HMX)/3-Nitro-1,2,4-triazol-5-one (NTO) composite, exhibit excellent high-overload performance. However, thermal cycling near the melting point of DNAN induces interfacial micro-damage, deteriorating mechanical properties. Dynamic Mechanical Analysis and Split Hopkinson Pressure Bar experiments revealed that after thermal cycling at 75 degrees C, the storage modulus decreased by 12.8% and dynamic compressive strength declined by 7.99%, indicating irreversible damage. Micro-Computed Tomography and Scanning Electron Microscope characterizations showed the micro-defect volume fraction doubled (from 0.81% to 1.63%) alongside the formation of numerous needle-like crystals, linking interfacial damage to sharp defects. In-situ Xray Diffraction and optical microscopy revealed that NTO undergoes significant dissolution within molten DNAN (22 wt%) and recrystallizes into needle-like structures upon cooling, whereas HMX dissolution is minor (8 wt%). Density Functional Theory-based molecular dynamics simulations elucidated the molecular origin: NTO's planar structure forms a stable bidirectional hydrogen bond network with DNAN. Compared to HMX, NTO exhibits a 34.8% lower dissolution barrier, higher interfacial binding energy (-2.09 eV), and significant electron transfer (356.68 e), explaining its preferential dissolution and needle growth. This study links macroscopic degradation to mesoscopic structural evolution and molecular mechanisms, providing a theoretical basis for designing highoverload-resistant explosives.

The increase in complexity arising from topology design and morphology optimization hinders a deep understanding of the dynamic response of sandwich structures (SSs). Accordingly, a tailored framework of Sandwich Structure Performance Lifecycle Engine (SSPLE) is proposed by integrating eXtreme Gradient Boosting (XGBoost), SHapley Additive exPlanations (SHAP), and Nondominated Sorting Genetic Algorithm II (NSGA-II). To demonstrate its superiority, an SS with three-layer aluminum foam cores is selected as the case study. Firstly, specific energy absorption (SEA) and peak deflection (PD) are employed to characterize the SS's blast resistance performance, considering the influence of 11 features encompassing geometry, material, bonding, and loading parameters. Subsequently, 600 instances are collected from validated numerical simulations. Employing the trained XGBoost (R2 = 0.9443 on the test set) and SHAP, Ttop (thickness of the top sheet) is identified as the most influential feature with an effect range from -0.32 kJ center dot kg-1 to 0.91 kJ center dot kg-1, and is confirmed to exhibit a substantial interaction effect with SoD (standoff distance) on SEA. As for PD, the main effect of AM (adhesive material) is restricted to the range of -2.35 mm to 7.27 mm following the exclusion of interaction effects. In addition, two optimization strategies provide explainable schemes for performance enhancement, with an increase of 67.39 % in SEA and a decrease of 32.60 % in PD. These findings confirm that the SSPLE can provide a systematic solution for dynamic response analysis and design optimization of SSs. Lastly, a software tool is developed to facilitate the implementation of SSPLE functionalities in practical applications.

To address the breaching requirements of reinforced concrete walls, this study analyzes the damage characteristics of reinforced concrete slabs (RCSs) subjected to penetration and explosion. Experiments were conducted using explosion of prefabricated hole and static explosion following penetration. Validated numerical models were employed to investigate the damage mechanisms induced by explosion shock waves and detonation gas. A decoupled approach was used to examine the damage capacities of cylindrical bare charges and the influences of pre-penetration damage on explosive performance. Results indicate that front surface damage primarily resulted from pressure crushing caused by shock waves and detonation gas, while rear surface damage arose from tensile failure due to stress wave reflection, superposition, and the shear plugging of detonation gas. Pre-penetration damage exerted a weakening effect on targets, thereby making cracks easier to form, and rendering cracks to extension during internal explosions. An optimal length-to-diameter ratio of 3.37, with constant charge mass, maximized damage to the RCSs under the studied conditions. Similarly, the highest damage occurred when the charge position-to-target thickness ratio was 0.50. The target dimensionless critical thickness for realizing effective hole expansion was set to 2.73. Perforation diameters after the explosion exhibited linear correlations with the dimensionless impact coefficient. Compared to the penetration at the rebar grid center, that at the midpoint of a single rebar and the intersection of rebars reduced the perforation diameter by 5.7% and 8.9%, respectively. The study offers practical guidance for optimizing charge design and placement, and informing breaching strategies through a deeper understanding of damage mechanisms and rebar configuration effects.

Existing single-degree-of-freedom (SDOF) methods inadequately capture the influence of localized damage on the global response of reinforced concrete (RC) structures subjected to close‑in blast loading. This study aims to addresses this limitation. Close‑in explosion experiments were conducted on RC beams at five scaled height of burst, demonstrating that localized effects are non-negligible in the structural response. A validated numerical model was subsequently developed, and regression analysis of the simulation data was used to establish a predictive model for local damage characteristic dimensions. Furthermore, a modified SDOF framework was formulated to capture the coupled local–global response of RC beams under close-in explosions. A local response model for RC beams under close‑in explosions was first formulated based on the traveling hinge concept. A structural resistance-deflection relationship was then developed to account for cross-sectional loss induced by localized damage. The influence of local damage evolution on the mass transformation coefficient and stiffness transformation coefficient was subsequently analyzed, and an equivalent SDOF model was established to characterize the global structural response of RC beams under varying degrees of cross-sectional damage. By integrating the local response model with the equivalent SDOF approach, a global response model incorporating localized damage effects was established and validated against experimental data. The model improves the prediction accuracy, especially for scaled heights of burst below 0.45 m/kg1/3. For maximum displacement prediction, the traditional SDOF model yields a mean absolute percentage error of 23.28%, while the improved model reduces this error to 6.49%.