A robust multiple linear regression model for accurate density prediction of metal-containing energetic complexes and energetic metal-organic frameworks to enhance detonation performance forecasting

High-energy-density materials such as metal-containing energetic complexes (MCECs) and energetic metal-organic frameworks (EMOFs) are critical in defense and industrial applications due to their superior energy output and tunable properties. Accurate prediction of their density is essential for reliable assessment of detonation parameters, including pressure, velocity, and heat release, which ultimately dictate explosive performance and safety. This study presents a novel multiple linear regression (MLR) model developed using a comprehensive dataset of 365 compounds. The model integrates three key molecular descriptors to predict densities with high precision ( R 2 = 0.957, RMSE = 0.104 g/cm 3 ). The model captures the complex influence of metal types, ligand architecture, and molecular packing effects—outperforming previous methods, especially for compounds with polycyclic and heterogeneous frameworks. Validation against experimental densities and thermochemical calculations demonstrates its reliability for model-driven design and optimization of next-generation energetic materials. This methodological advancement offers defense researchers and engineers a practical, computationally efficient tool for screening novel energetic compounds, supporting safer and more effective explosive formulations tailored for specific performance and stability requirements.