Study on the formation of micro-scale shaped charge jet based on diameter effect Authors: Dexu Li, Zhengxiang Huang, Qiangqiang Xiao, Xin Jia, Yiming Chen Published: 2026 ABSTRACT AbstractShaped charge jets (SCJ), characterized by high power, favorable directionality, and broad platform compatibility, are well suited for low-overload miniature weapon platforms. At the micro-scale, however, charge diameter introduces pronounced size effects that are not captured by classical shaped-charge theory, leading to a systematic overestimation of jet velocity. This study aims to address the limitations of conventional theory, particularly the inadequate prediction of jet velocity for micro-scale charges. Theoretical analyses and detonation experiments are first conducted to verify the existence of a size effect, with the experimental results confirming a marked reduction in detonation velocity at reduced charge diameters. Building on these findings and integrating classical jet-formation theory, the formulations for liner collapse velocity and jet velocity are refined. A pulsed X-ray test employing a 10 mm diameter shaped charge was performed to validate the proposed model. The measured jet tip velocity was [3077.0 m/s], while the conventional model predicted [3921.5 m/s] and the modified model predicted [3189.6 m/s]. The modified model exhibits strong agreement with the experimental data, reducing the prediction error relative to the conventional theory by approximately 20%. These results provide essential guidance for the design and optimization of micro-scale shaped charges.Keywords: Shaped charge, X-ray, Jet formation, Micro-scale chargeSubject terms: Engineering, Physics Introduction Recent years have witnessed substantial progress in microelectromechanical systems (MEMS) and artificial intelligence (AI), which has significantly advanced the development of micro aerial vehicles (MAVs). Contemporary MAVs achieve higher energy efficiency and lower operational costs, leveraging their compact size, agility, and stealth capabilities to undertake specialized missions in complex terrain and confined environments1. Regarding physical scale, MAV dimensions have undergone a marked reduction—from initial sizes of several meters, to tens of centimeters, and now to the centimeter scale—representing true miniaturization of aerial platforms. Concurrent weight reductions have yielded MAVs weighing only a few grams, as exemplified by NUS‑Robot2, the RF‑powered MAV3, KUBeetle4, and RoboBee5. Owing to their compact form, high maneuverability, and enhanced stealth, these vehicles are exceptionally well suited for specialized operations in complex terrains and confined spaces, underscoring their considerable potential for military applications. Employing these MAVs as weapon‑system platforms could substantially enhance their utility for specialized missions. However, the platforms’ compact size and limited payload capacity impose stringent constraints on ammunition integration, underscoring the need for advances in micro‑munition technology. Compared with most conventional munition types, shaped charges offer superior penetration capability, precise directional control, and high compatibility with compact platforms. Moreover, their ability to function without imposing excessive overload on the delivery system renders them especially well suited for low‑overload platforms such as MAVs. Nonetheless, existing shaped‑charge warheads—both domestically and internationally—typically range in diameter from 20 mm to 100 mm, and investigations into smaller calibers, particularly at the millimeter scale, remain scarce6. A shaped charge consists primarily of two main components: the liner and the explosive. Detonation of the explosive drives the liner to collapse, forming a focused jet characterized by high temperature, high velocity, and substantial kinetic energy, which is subsequently employed to damage or destroy the target7. The jet‑formation process involves a range of intricate, nonlinear dynamic mechanical phenomena, including the detonation behavior of high‑energy explosives and the high‑strain‑rate response of liner materials. These inherent complexities pose substantial challenges to the development of accurate theoretical models. Research on the formation characteristics of shaped charges has advanced considerably since Birkhoff et al. first proposed the jet‑formation theory in 19488. Early models, however, tended to overestimate jet velocities. To account for velocity gradients within the jet, Pugh, Eichelberger, and Rostoker9refined Birkhoff’s framework and developed a one‑dimensional unsteady theory (PER theory), which applies to conical and wedge‑shaped liners under planar detonation initiation. Subsequently, Carleone et al10. extended the PER jet‑formation model to accommodate liners of more general geometries—such as spherical or trumpet‑shaped configurations—while incorporating both spherical and annular detonation waves. These successive refinements culminated in the establishment of a generalized jet‑formation theory. Current research on shaped charges seldom accounts for diameter effects on detonation behavior. When a detonation wave propagates through a charge of uniformly small diameter, both the detonation velocity and the wave-front shape attain a quasi-steady state that is independent of the initial conditions11. In small-diameter charges, lateral expansion during detonation induces energy losses within the reaction zone. When the charge diameter is sufficiently large, these losses become negligible and the detonation wave propagates at near-ideal velocities. However, as the diameter decreases toward a threshold value (the limit diameter), the detonation velocity declines correspondingly. At the critical diameter, energy losses in the reaction zone become substantial enough to disrupt the balance between energy release and dissipation, thereby causing the detonation wave to fail to sustain itself—resulting in detonation failure12. Research on the diameter effect in explosives has been extensive. Eyring et al13. were among the first to propose empirical formulas and theoretical models describing the relationship between detonation velocity and charge diameter. Subsequently, Wood and Kirkwood14 established the conservation laws for mass, momentum, and energy behind curved detonation wave fronts. The majority of later investigations have built upon the Wood–Kirkwood framework to further elucidate the diameter effect in explosives. Lambourn, Bdzil, Stewart, and co-workers15–17successively proposed a theoretical model for detonation shock dynamics (DSD). This approach effectively characterizes the relationship between reaction rates and the stability of the detonation reaction zone, thereby establishing a quantitative link between the ideal detonation velocity and the curvature of the detonation wave front. Aslam et al18. pioneered a numerical method grounded in DSD theory, employing the Level Set technique to simulate detonation wave propagation under complex two-dimensional boundary conditions. Chiquete et al19. decomposed detonation performance calculations into separate timing and energy-release components. The timing analysis employed DSD theory to describe wave propagation, whereas energy release was modeled using the Pseudo-Reaction Zone (PRZ) approach. Analogous to a combustion reaction framework, the PRZ model simulates the finite-rate conversion of reactants to products, with the reaction rate governed by the normal surface velocity obtained from DSD computations. Yoo20investigated detonation mechanisms in non‑uniform explosives, deriving a correlation between normal detonation velocity and wave‑front curvature that aligns with DSD theory. These findings underscore the capability of DSD theory to accurately represent curved detonation waves under non‑ideal conditions. Short et al21. applied DSD theory to examine the stable propagation of detonation waves in curved geometries, establishing relationships among angular detonation velocity, arc radius, thickness, and confinement parameters. Jiao et al22. developed a theoretical model for curved detonation wave propagation in strongly confined cylindrical small‑diameter charges, combining DSD theory with a linear streamline deflection‑angle assumption. Jiang et al23. utilized DSD theory to investigate the Level Set method for simulating detonation‑front propagation in two‑dimensional body‑fitted coordinate systems. Moreover, Jiang et al24. employed the DSD framework to analyze non‑ideal detonation wave propagation under inert‑media confinement, demonstrating that the shock polar method effectively captures interfacial interactions. Zhong et al25., also relying on DSD theory, studied the Level Set method for three‑dimensional non‑ideal detonation‑front propagation, introducing a finite‑difference scheme for the governing equations in non‑orthogonal body‑fitted coordinates and presenting the corresponding numerical solutions. At present, DSD theory has emerged as a mainstream approach for describing non‑ideal detonation in explosives. Through a combination of numerical and analytical techniques, it enables the determination of key parameters—including detonation‑front propagation, detonation pressure, and detonation velocity—as functions of wave evolution. Consequently, DSD theory constitutes an exceptionally effective tool for investigating the non‑ideal detonation behavior of small‑scale shaped charges. This study focuses on micro‑scale shaped charges, specifically those with charge diameters below the explosive’s limit diameter. Detonation velocity measurements were conducted on multi‑scale charges to establish the quantitative relationship between detonation velocity and charge diameter. Starting from classical DSD theory, an equation analogous to geometric optics was employed to describe the evolution of the detonation wave‑front shape. This approach yielded a detonation‑wave propagation model for JH‑2 explosive at small scales, which was subsequently validated against experimental data. Building upon the derived non‑ideal detonation model for JH‑2 explosive, we integrated shaped‑charge formation theory to analyze and characterize the formation of micro‑scale shaped‑charge jets (MSCJ). The theoretical model was further verified through pulsed X‑ray experiments, revealing the influence of diameter effects on MSCJ formation. The findings of this study offer valuable insights for the design and application of micro‑scale shaped charges. Theoretical model Over recent decades, research efforts have predominantly focused on enhancing jet velocity to improve penetration depth. Accordingly, studies of shaped charges have prioritized liner material optimization (e.g., copper versus tantalum alloys) and geometric innovations (e.g., conical, trumpet, and multi‑layered liners), while micro‑scale charge dynamics have received comparatively limited attention. The widely adopted Pugh–Eichelberger–Rostoker (PER) theory9accurately predicts jet‑formation parameters for planar‑initiated conical and wedge‑shaped liners. Carleone et al10. subsequently extended this framework by developing a generalized semi‑analytical model capable of simulating jet formation under diverse initiation conditions and liner configurations. In micro‑scale shaped charges, diameter effects induce lateral rarefaction waves during the detonation process, giving rise to two critical non‑ideal phenomena: Detonation velocity reduction below the ideal Chapman-Jouguet value. Detonation velocity reduction below the ideal Chapman-Jouguet value. Curved detonation fronts with convex morphology. Curved detonation fronts with convex morphology. These effects fundamentally alter jet‑formation dynamics relative to those of conventional charges, imposing constraints on key parameters such as the jet velocity gradient and mass distribution. While current theoretical frameworks are effective for standard‑scale charges (diameter > 20 mm), they exhibit significant limitations in predicting micro‑scale jet behavior, owing to their inherent assumptions of steady‑state detonation fronts and steady‑state detonation. Consequently, it is imperative to refine the jet‑formation theory from new perspectives in order to enhance its engineering applicability. Charge diameter effect The DSD theory plays a critical role in studying the diameter effect of small-scale charges. Current research primarily focuses on charge types (e.g., PBX-950226, LX-1627, and JO-11C28 and confinement conditions, with limited application of diameter effect studies to shaped charges. Additionally, for shaped charges, JH-2 is a commonly used explosive, yet few studies have investigated its diameter effect. Therefore, this paper will study the diameter effect of JH-2, develop a detonation model tailored for JH-2, and provide insights into its detonation propagation characteristics. Based on the DSD framework, an equation analogous to geometric optics is established by coupling the relationship between the normal detonation velocity and local mean curvature, along with the kinematic description of wave front evolution. This formulation enables quantitative modeling of detonation front morphology over time15–17. We propose that the detonation wave behaves as a geometric surface of infinitesimal thickness, where each point acts as a discrete initiation source. When the normal velocity is defined at these points, the collective envelope of their propagated micro-detonation waves reconstructs the macroscopic wave front at subsequent time steps. A schematic representation of this mechanism is illustrated in Fig. 1. Schematic diagram of detonation wave propagation. Figure 2 illustrates the wavefronts at two distinct time instants, as described by the function . Based on the geometric relationship between these two wavefronts, the following equation can be derived: Schematic diagram of detonation incorporating boundary conditions. Once the detonation wavefront has stabilized, its spatial profile becomes independent of time and can be expressed as zThis yields the following relationship: In conventional DSD theory, D can typically be expressed as15–17: where denotes the ideal detonation velocity (Chapman-Jouguet velocity), is a parameter related to the explosive properties, is the mean curvature, which can be expressed in terms of the spatial derivatives of the wave front curve z as: By combining the above Eq. (4) and neglecting higher-order infinitesimal terms, a typical viscous Burgers equation can be derived. To this end, a new function is introduced, which can be expressed as: In the above equation, C is a constant. we obtain: By applying the boundary conditions z=0 at r=0, z=0 at r=2R and dz/dr=0 at x=R, the solution can be determined as: At this stage, the relationship between detonation velocity and steady-state wavefront geometry has been established. However, to further elucidate diameter-dependent effects, it is necessary to analyze the charge boundary conditions, which requires introducing mechanical constraints rather than relying exclusively on geometric relationships. Figure 2 depicts the boundary conditions during charge detonation. Ddenotes the detonation velocity, is the angle between the detonation wavefront and the z-axis, represents the wavefront-z-axis angle at the boundary, indicates the streamline deflection angle, corresponds to the boundary streamline deflection angle, refers to the particle velocity of the confining material, is the propagation velocity of the inwardly traveling rarefaction wave. Given the small charge diameter in our study, we postulate that the streamline deflection angle exhibits a linear radial variation, attaining its maximum value at the boundary. For the streamline deflection angle, Brain29 derived an analytical formula expressed as: where represents the isentropic exponent of the energetic charge material. By combining the linear variation relationship of the streamline deflection angle with Eq. (10), and differentiating both sides with respect to r, the following expression can be derived: By combining Eq. (6), Eq. (11), and the wavefront relationship depicted in Fig. 3, the curvature κ can be expressed as: Schematic diagram of the liner collapse process. Considering the boundary conditions at and combining Eqs. (5) and (12), the detonation velocity D can be expressed as: Based on Fig. 3, the boundary streamline deflection angle can be expressed by the following equation: The particle velocity relative to the material can be expressed by the following empirical formulation30: where represents its shock wave velocity, c represents the bulk sound speed under ambient conditions, and is a dimensionless material constant dependent on the equation of state. Thus, by applying the conservation of momentum and the continuity conditions at the detonation interface, we can derive the expression for : where represents the detonation product pressure at the interface, is the density of the detonation products. Parameters with subscript 0 (e.g., , ) correspond to their initial states. For the confining material, both the particle velocity and the pressure vanish under initial conditions. Additionally, the values of , , and in this study are based on data from reference. With the established expression for , the detonation velocity D can be determined through the iterative computational scheme enabled by Eqs. (13) and (14). Furthermore, parametric variation of the charge radius R facilitates characterization of steady-state detonation velocities across different charge diameters. This methodology quantitatively elucidates the underlying mechanism of the charge diameter effect. Micro-scale shaped charge jet formation The PER theory is one of the most fundamental and widely used frameworks in shaped charge jet formation theory. However, the PER theory is based on the Chapman-Jouguet (CJ) detonation model for jet formation research, which is suitable for large-scale shaped charge calculations. In contrast, for micro-scale shaped charges, the previous section’s findings indicate that the diameter effect of explosive detonation cannot be neglected, as it is a critical factor influencing jet formation. Moreover, the morphology of the detonation wave front varies with the charge diameter. Under these conditions, both the magnitude and direction of the detonation velocity on the liner surface change along the generatrix. Therefore, it is imperative to accord greater consideration to the explosive diameter effect in computations of the liner collapse process. The formation process of SCJ can be divided into three main stages31: first, the collapse phase of the liner; second, the collision of the liner near the axis to form the jet; and finally, the elongation and eventual breakup of the shaped charge jet al.ong the axis. The most pronounced distinction between MSCJ and SCJ manifests during the liner collapse phase. Variations in detonation parameters induced by charge diameter effects result in non-uniform liner collapse velocities across the charge cross-section. Computational analyses based on conventional shaped charge jet theory exhibit deviations in predicted jet velocity. The primary objective of this section is to determine the liner collapse velocity. Upon initiation of the shaped charge, the detonation wave and subsequent detonation products propagate radially outward, inducing inward collapse of the liner toward the central axis. During this dynamic process, each liner element attains an initial collapse velocity , conventionally calculated using the Gurney velocity formulation. The classical Gurney formula is generally expressed as Equation32: where represents the ratio of the liner mass to the explosive mass, and E is the Gurney energy. However, the classical Gurney formula is more suitable for sandwich structures (two metal plates of equal mass with an explosive layer in between). For shaped charge applications, Chou32, Hennequin33, and Chanteret34 have developed distinct empirical formulations specifically tailored to implosive curved geometries (e.g., conical liners). These models demonstrate good agreement with experimental measurements of collapse velocities in large-diameter charge configurations. First, for the Gurney energy E, it is typically obtained through experiments and empirical formulas. However, it is often treated as a quantity related to the detonation velocity D. For MSJC, the detonation velocity D varies along the generatrix of the liner. Consequently, in the calculations, the Gurney energy E must also be treated as a variable, which can be determined through the relationship between detonation velocity D and charge radius R established in the preceding subsection. Figure 3 illustrates the schematic diagram of the liner collapse process. The collapse velocity expression proposed by Chou32 incorporates an impulse term derived from an empirical formula obtained through two-dimensional numerical simulations. However, this impulse calculation is only valid within , resulting in significant deviations when predicting the collapse velocity of annular elements near the liner apex. To address this limitation, we adopt Chanteret’s collapse velocity formulation34, which postulates a stationary boundary (zero radial expansion velocity). Within the region bounded by and the liner surface, the explosive energy is fully converted into the kinetic energy of both the liner and detonation products. According to the momentum conservation principle, this boundary can be determined by solving the following governing equation: where represents the density of detonation products, and represent the outer and inner radius of the charge. Based on the energy conservation equation, the collapse velocity can be expressed as: Since many parameters are related to the generatrix position of the liner, we use the subscript z to denote these varying parameters. When the liner element reaches the axis and collides, it splits into two parts moving in opposite directions: one forms the jet, and the other forms the slug. For the theoretical model of shaped charge formation, it is essential to obtain the velocities of the jet and the slug. According to the steady-state incompressible fluid theory, the velocities of the jet and slug can be expressed as: where, represents the collapse angle, represents the half-cone angle, and represents the projection angle. Therefore, to calculate the velocities of the jet and the slug, it is necessary to determine the projection angle and the collapse angle. As indicated by Eq. (20), the collapse velocity varies along the generatrix of the liner. Consequently, the projection angle calculated using the conventional Taylor relation tends to be underestimated. Therefore, based on the exponentially accelerated collapse model of the liner, a relatively accurate formula for calculating the projection angle is derived as follows: In the equation, the prime symbol (′) denotes differentiation with respect to the position coordinate along the liner generatrix, and are empirical constants, M is the mass of the liner element, and represents the detonation pressure at different position coordinates. In practical scenarios, the generatrix of the collapsing liner is often curved rather than straight. Therefore, the tangent of the collapse angle is represented by the slope of this curved generatrix at the axis of the liner. The displacement lof an annular element along the direction of the collapse velocity can be obtained by integrating the collapse velocity. Once the collapse velocity, projection angle , and collapse angle are calculated using Eqs. (20), (22), and (23), the jet velocity and slug velocity can be determined through Eq. (21). This model accounts for the variations in detonation behavior caused by diameter effects in miniaturized charges, offering valuable insights for the design of microscale shaped charges. Experimental Research This subsection outlines two principal experimental components: (1) detonation velocity tests investigating diameter-dependent effects, and (2) pulsed X-ray radiography experiments characterizing jet formation dynamics. A systematic approach is presented, providing comprehensive descriptions of specimen configurations, experimental setups, and diagnostic arrangements for each test series. Charge diameter effect experiment As shown in Fig. 4, the physical diagram of the JH-2 (a high explosive commonly used in shaped charges) charge used in the detonation velocity measurement experiment is displayed. The charges employed in this experiment are all compression charges, with five different diameters: 30 mm, 20 mm, 10 mm, 6 mm, and 4 mm. All charges have a length-to-diameter ratio of 1:1 and a density of 1.68 g/cm³. The parameters of the charges, including size and density, are listed in Table 1. Image of JH-2 Charges. Table 1Characteristics of Charges.ExplorsiveDensity(g/cm³)Charge diameter(mm)Charge height(mm)JH-21.683030202010106644 Characteristics of Charges. Figure 5 provides a schematic diagram of the experimental setup for the detonation velocity measurement. The test uses an electrical method to determine the detonation velocity of multi-scale charges. Three probe wires are attached to the charge. The core principle of this measurement technique relies on the detonation wave’s instantaneous energy severing the probe wires, altering their resistance and producing three pulse signals. This enables accurate measurement of the time intervals between the detonation wave passing through the three probe wires. By carefully spacing the probe wires, the average detonation velocity can be computed. The test procedure involves initiating a electric detonator with a DC power source, which then detonates the charge. As the detonation progresses, it breaks each probe wire in sequence, generating signals that are converted by a multi-channel signal generator and recorded by a time interval measurement instrument. Schematic diagram of the experimental setup. In this section of the experiment, five test conditions with charge diameters of 30 mm, 20 mm, 10 mm, 6 mm, and 4 mm were set up. For each test, the sample consisted of multiple charges with a length-to-diameter ratio of 1:1 (to avoid inconsistencies in charge density due to excessive charge length). The samples were placed in a specially designed fixture (made of nylon), and the probe wires were fixed at specific positions. After securing the probe wires, the distances between them were measured and recorded. The test site was then arranged according to the schematic diagram of the experimental setup, and all test lines were connected. The test site is shown in Fig. 6. Before conducting the test, a multimeter was used to recheck the status of each circuit to ensure that the test could yield valid data. Field photo of the experiment. Micro-scale shaped charge jet formation experiments As shown in Fig. 7, the physical and structural diagrams of the micro-scale shaped charge designed in this study are presented. Based on the research conclusions from Sect. 2, the charge diameter was selected as 10 mm, with a charge height of 1.5 times the diameter. The charge material is JH-2 explosive, with a density of 1.68 g/cm³. We selected a conical liner, a standard choice for such applications, made of conventional copper with a cone angle of 60° and a uniform thickness of 0.5 mm. Micro-scale shaped charge. As shown in Fig. 8, the on-site arrangement of the X-ray test and a top-view schematic of the setup are presented. During the experiment, the X-ray film is placed inside the film holder, positioned perpendicular to the horizontal plane, while the test object is located at the intersection between X-ray triggers 1 and 2 and their corresponding films. In the test, both X-ray machines are activated simultaneously but triggered at different times to capture images at moments and . Due to the distance between the film and the test object, the captured images of the shaped charge jet are magnified. To calculate the magnification factor for the two images, the following proportional relationship can be used, with specific parameter definitions illustrated in Fig. 9 (C). Alternatively, the magnification can be determined by measuring and calculating based on reference objects of known dimensions in the photographs. X-ray test setup and schematic diagram. Detonation front at deffrent radius. Based on the distances , , , and between the head and tail of the shaped charge jet and the same reference point on the film at times and ., the head and tail velocities of the jet can be calculated as follows: Since the shaped charge used in this experiment is significantly smaller than conventional ones, the distances d and c were set to smaller values to capture clearer X-ray images. Additionally, marker lines were placed on the film holder to facilitate the calibration of axial and radial scales and distances. The parameters of the experimental setup in Fig. 9 are listed in Table 2, and the magnification factor was calculated accordingly. Table 2Parameters for X-ray Test Setup.Charge diameter(mm)a(m)b(m)c(m)d(m) (μs) (μs)101.4452.1251.4352.0201.9931.9505.65516.7471.4402.1251.4102.0181.9791.9505.77313.0511.4452.1231.4252.0151.9861.9495.71113.160 Parameters for X-ray Test Setup. Results and discussion Charge diameter effect For each test condition, three rounds of experiments were conducted to ensure the accuracy of the detonation velocity measurements. The results of the detonation velocity tests for different charge diameters are presented in Table 3. Table 3Experiment data of detonation velocity.Diameter(mm)Time Interval(µs)Distance interval(mm)Detonation velocity(m/s)303.7030.008108.11204.9040.008163.27106.1045.687488.526.03.1020.826716.134.03.6020.645733.33 Experiment data of detonation velocity. Analysis of experimental data reveals distinct diameter-dependent detonation characteristics in JH-2 explosive charges. Specifically: Charges exceeding 20 mm diameter exhibit negligible dependence of detonation velocity on dimensional variations, Furthermore, the ideal detonation velocity of JH-2 explosive at a density of 1.68 g/cm³ is determined to be approximately 8160 m/s.; Charges exceeding 20 mm diameter exhibit negligible dependence of detonation velocity on dimensional variations, Furthermore, the ideal detonation velocity of JH-2 explosive at a density of 1.68 g/cm³ is determined to be approximately 8160 m/s.; Substantial velocity attenuation occurs when diameters fall below 10 mm, At a diameter of 10 mm, the velocity deficit is approximately 700 m/s; at 6 mm, the deficit reaches 1400 m/s; and at 4 mm, the deficit even reaches 2400 m/s; Substantial velocity attenuation occurs when diameters fall below 10 mm, At a diameter of 10 mm, the velocity deficit is approximately 700 m/s; at 6 mm, the deficit reaches 1400 m/s; and at 4 mm, the deficit even reaches 2400 m/s; The limit diameter, where detonation velocity begins significant decline, is empirically determined to reside within the 10–20 mm range. A systematic parametric study with finer diameter increments (e.g., 1 mm steps) is recommended to precisely identify the critical transition threshold. The limit diameter, where detonation velocity begins significant decline, is empirically determined to reside within the 10–20 mm range. A systematic parametric study with finer diameter increments (e.g., 1 mm steps) is recommended to precisely identify the critical transition threshold. Based on the micro-scale detonation computational model proposed in Sect. 2.1, combined with the standard parameters of JH-2 explosive and the nylon confinement properties, key characteristic parameters of JH-2 detonation were calculated. These include detonation velocity, wave front, and curvature for micro-diameter charges. Utilizing this model, theoretical predictions of detonation velocity versus charge radius were generated and compared with experimental measurements, as shown in Fig. 10. Detonation velocity- charge radius relationship curve: comparative analysis of experimental measurements and theoretical simulations. As shown in Fig. 10, the computational results show close agreement with the experimental data across all tested charge radii, with consistent trends and a maximum deviation of 3.1%. The detonation velocity increases monotonically with charge radius and asymptotically approaches the ideal detonation velocity at larger radii. Notably, the rate of increase in detonation velocity accelerates markedly for smaller charge radii (e.g., < 10 mm). By the detonation wave model discussed in this section, the wave front profiles at charge radiu of 2 mm, 3 mm, and 5 mm were obtained, as shown in Fig. 9. The results demonstrate an inverse relationship between charge radius and wave front curvature: smaller radii correlate with higher curvature magnitudes. Notably, the curvature exhibits a pronounced increase near the charge boundary. Under the combined effects of the diameter dependency and steady-state non-ideal detonation processes, reductions in lateral expansion, energy density in the reaction zone, detonation wave intensity, and chemical reaction initiation rate lead to decreased detonation velocity and curvature of the shock front. The impact of this energy loss on shaped charge performance will be investigated in subsequent studies. Micro-scale shaped charge jet formation For this shaped charge, we conducted three rounds of tests to ensure the reproducibility of the experiments. One of the X-ray images obtained from the test results are shown in Fig. 11. Pulsed X-ray radiography of MSCJ morphology. Figure 11 presents pulsed X-ray test images of jet formation from a 10 mm diameter shaped charge. A steel ingot positioned directly beneath the charge serves as the spatial reference point for dimensional calibration. The annotated dimensions in the figure represent actual measurements, enabling calculation of jet tip velocity and tail velocity based on these markers and temporal intervals (summarized in Table 4). Experimental results indicate: At 13.16 µs, the jet is fully formed, with no significant head accumulation compared to conventional jets, and exhibits a smaller diameter; the jet maintains good continuity during free flight without fracture; Due to the larger apex size of the liner, the resulting slug mass is significantly increased. Table 4Comparative results of micro-scale jet formation: X-ray testing vs. theoretical modeling.NumberExperimental resultsTheoretical results(modifed model)Theoretical results(conventional model)(m/s)(m/s)(m/s)(m/s)(m/s)(m/s)13024.6579.53189.6518.43921.5796.123019.3543.233187.1471.0 Comparative results of micro-scale jet formation: X-ray testing vs. theoretical modeling. Table 4 compiles the calculated velocities of both the jet tip and tail during complete formation, derived from X-ray test image data, alongside corresponding theoretical modeling results of jetformation. As shown in Table 4, the experimentally measured jet velocity of the formed shaped charge jet is 3077.0 m/s, and the slug velocity is 531.2 m/s. The modified theoretical model incorporating charge diameter effects yields a calculated jet velocity of 3189.6 m/s (relative deviation: 3.7%) and a slug velocity of 518.4 m/s (relative deviation: 2.4%). In comparison, the conventional theoretical model that neglects charge diameter effects produces a jet velocity of 3921.5 m/s (relative deviation: 27.4%) and a slug velocity of 796.1 m/s (relative deviation: 49.8%). Based on the theoretical model established in Sect. 2, the variation of collapse velocity with respect to the liner coordinate position is illustrated in Fig. 12, showing a comparison between the collapse velocities calculated by the modified model and the conventional model. It can be observed that after accounting for charge diameter effects, the collapse velocity is significantly reduced, with values approaching 1000 m/s. Moreover, as the coordinate position increases, the improved model exhibits a more pronounced decreasing trend in collapse velocity. Comparative analysis of calculation results: conventional model vs. modified model. Figure 13 illustrates the variation of jet velocity with respect to the liner coordinate position, comparing the calculated values from the modified model s with those from the conventional model. It can be seen that, compared to the conventional model, the modified model yields lower jet velocities. Furthermore, the modified model reaches its maximum velocity earlier than the conventional model, but also exhibits faster velocity decay. These changes lead to a micro-scale jet that is thinner and longer than its large-diameter counterpart upon formation, with a corresponding reduction in the mass of the jet tip. Comparative analysis of calculation results: conventional model vs. modified mode. In conclusion, the jet formation theory incorporating charge diameter effects proves particularly applicable to micro-scale shaped charges, demonstrating reliable predictive capability for both jet velocity and collapse processes in such configurations. Conclusions The purpose of this study is to overcome the limitations of conventional shaped charge theory, particularly its inadequacy in predicting the jet velocity of micro-scale shaped charges. These shortcomings primarily stem from the neglect of energy losses caused by diameter effects in miniaturized charges, which leads to reduced prediction accuracy. The main conclusions drawn are as follows: (1) This study conducted both experimental and theoretical investigations into the diameter effects of JH-2 explosive, analyzing the influence of confinement materials and lateral rarefaction waves on the propagation of curved detonation waves. The resulting detonation wave propagation model effectively captures the actual behavior of detonation wave propagation, with calculated results showing good agreement with experimental data. (2) For JH-2 explosive, the smaller the charge diameter, the more pronounced the diameter effects become in detonation wave propagation. Within the calculated range, when the charge diameter is 10 mm, the detonation velocity deficit is approximately 8.6%. As the charge diameter decreases, this velocity deficit gradually increases, reaching about 29.4% at a charge diameter of 4 mm. (3) By incorporating detonation wave parameters into the conventional shaped charge calculation model, this approach explicitly accounts for the coupled effects of micro-scale charge diameter effects and the liner collapse process. Compared to the conventional model, this method reduces the deviation between calculated jet velocity and experimental values from approximately 27.4% to below 5%. This advancement significantly improves calculation accuracy and provides a solid theoretical foundation for optimizing the structural design of micro-scale shaped charges. (4) Comparison between the conventional model and the modified model reveals that the conventional theory overestimates both the collapse velocity and the jet velocity. Additionally, deviations emerge in the velocity trends as they develop along the generatrix of the liner. This phenomenon suggests that the design of micro-scale shaped charges may deviate from conventional scaling laws, indicating that a direct scaling-down approach from large-diameter shaped charges is not feasible for micro-scale applications. The results obtained in this study provide new perspectives and conclusive findings regarding the formation mechanisms of micro-scale shaped charges, offering valuable references for advancing the miniaturization and micro-scale development of shaped charge technology. Author contributions Dexu Li: Conceptualization, Methodology, Software, Investigation, Formal analysis, Validation, Data curation, Writing – original draft. Zhengxiang Huang : Conceptualization, Resources, Supervision, Writing – review & editing. All authors have read and agreed to the published version of the manuscript. Qiangqiang Xiao: Investigation, Supervision, Visualization, Writing – review & editing.Xin Jia: Investigation, Supervision, Visualization, Writing – review & editing.Yiming Chen: Software, Validation. All authors have read and agreed to the published version of the manuscript. Funding This research was funded by the National Natural Science Foundation of China (12372360). Data availability All data generated or analyzed during this study are included in this published article. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Zhengxiang Huang, Email: huangyu@mail.njust.edu.cn. Xin Jia, Email: jiaxin@mail.njust.edu.cn. References Associated Data Data Availability Statement All data generated or analyzed during this study are included in this published article.