An Improved Simple Method for the Calculation of the Detonation Performance of CHNOFCl, Aluminized and Ammonium Nitrate Explosives

Abstract

An improved simple method is presented for calculation of the detonation velocity of CHNO and CHNOFCl explosives, as well as non-ideal explosives containing aluminum (Al) and ammonium nitrate (AN) additives. In contrast to the available complex computer codes, where the estimated detonation velocities of non-ideal explosives for equilibrium and steady state calculations show significant differences from the measured data, this simple method gives more reliable results. Suitable decomposition paths are suggested in which the partial interaction of Al with the gaseous products and the decomposition of AN are assumed for composite explosives containing Al/AN additives. The predicted detonation velocities using the new method are good compared to those from one of the well-known empirical methods and from computer codes using full and partial equilibrium of Al/AN.

Prediction of Detonation Velocity and N−O Composition of High Energy C−H−N−O Explosives by Means of Artificial Neural Networks

“…Ammonium nitrate (AN) and AN based explosives like amatol 80/20 (80/20 AN/TNT) appears to show large deviation of over 20 % as shown in Figure 7. Ammonium nitrate explosive and explosive mixtures containing ammonium nitrate is said to exhibit non-ideal behavior [11] unlike all the other explosives considered in the dataset. The empirical method of Keshavarz et al [10] could not treat this behavior of ammonium nitrate and explosive mixtures of ammonium nitrate and they had to come up with an improved empirical model to account for this [11].…”

“…Ammonium nitrate explosive and explosive mixtures containing ammonium nitrate is said to exhibit non-ideal behavior [11] unlike all the other explosives considered in the dataset. The empirical method of Keshavarz et al [10] could not treat this behavior of ammonium nitrate and explosive mixtures of ammonium nitrate and they had to come up with an improved empirical model to account for this [11]. This non-ideal behavior of the explosive along with the lack of more data for the ANN model on AN and AN mixtures could be a reason for the ANN to not predict detonation velocity of these explosives accurately (since the accuracy of ANN model depends on the quality and quantity of data available to train).…”

“…Jain [8] showed that properties of explosives could be correlated using their oxidation and reduction valences. More recent studies on prediction of detonation velocity has been done for both CHNO explosives, CHNFOCl and aluminized explosives [9][10][11]. Keshavarz [9] gives an empirical relation for predicting detonation velocity for CHNOF and aluminized explosives at maximum nominal density.…”

Effect of Nitrogen-Doping on Detonation and Stability Properties of CL-20 Derivatives from a Theoretical Viewpoint

“…where D K-J is detonation velocity (km s À1 ); P K-J is detonation pressure (GPa); N is the number of moles of gas products for per gram explosive bombed (mol g À1 ); M is the molar mass of these gases, Q K-J is the heat of detonation (J g À1 ) and 1 is the loaded density of explosives (g cm À3 ), which was calculated from a suitable model for nitramine energetic materials with general formula C a H b N c O d [14], where M W is the molecule mass (g mol À1 ). Also, another faithful correlations were also used for prediction of D and P, which raised by Mohammad Hossein Keshavarz in 2004 [15] and 2016 [16].…”

A New Method for Assessment of Performing Mechanical Works of Energetic Compounds by the Cylinder Test

“…Kamlet and Jacobs 16 used these detonation products for estimation of detonation pressure and velocity of C a H b N c O d explosives above 1.0 g · cm –3 . It was shown that the predicted of detonation pressure and velocity by Kamlet and Jacobs 16 method give large deviations for loading density less than 1.0 g · cm –3 3a,4a . Short et al14a used these detonation products to derive wall velocity in the cylinder test of C a H b N c O d explosives by least‐squares fitting of the experimental data as follows:…”