Accepted Manuscript

Title: Computer code to predict the Heat of Explosion of High

Energy Materials

Authors: H. Muthurajan, R Sivabalan, N. Pon Saravanan,

M.B. Talawar

PII: S0304-3894(08)00536-0
DOI: doi:10.1016/j.jhazmat.2008.04.032
Reference: HAZMAT 8145

To appear in: Journal of Hazardous Materials

Received date: 6-10-2007

Revised date: 5-4-2008
Accepted date: 7-4-2008

Please cite this article as: H. Muthurajan, R. Sivabalan, N.P. Saravanan, M.B. Talawar,
Computer code to predict the Heat of Explosion of High Energy Materials, Journal of
Hazardous Materials (2007), doi:10.1016/j.jhazmat.2008.04.032

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
* 4. Revised Manuscript

Computer code to predict the Heat of Explosion of High Energy Materials

1* 2 2 2
H. Muthurajan , R Sivabalan , N Pon Saravanan , M.B. Talawar
1
Armament Research and Development Establishment, Pashan, Pune – 411 021, India
2
High Energy Materials Research Laboratory, Sutarwadi, Pune – 411 021, India

ABSTRACT

t
The computational approach to the thermochemical changes involved in the process of explosion

ip
of a High Energy Materials (HEMs) vis-à-vis its molecular structure aids a HEMs chemist / engineers to

cr
predict the important thermodynamic parameters such as heat of explosion of the HEMs. Such a

computer-aided design will be of useful in predicting the performance of a given HEM as well as in

us
conceiving futuristic high energy molecules that have significant potential in the field of explosives and

propellants. The software code viz., LOTUSES developed by authors predicts various characteristics of

an
HEMs such as explosion products including balanced explosion reactions, density of HEMs, velocity of

detonation, CJ pressure, etc. The new computational approach described in this paper allows the
M
prediction of heat of explosion (He) without any experimental data for different HEMs, which are

comparable with experimental results reported in literature. The new algorithm which doesn’t require any

complex input parameter is incorporated in LOTUSES (version 1.5) and the results are presented in this
d

2
paper. The linear regression analysis of all data point yields the correlation coefficient R = 0.9721 with a
te

linear equation y = 0.9262x + 101.45. The correlation coefficient value 0.9721 reveals that the computed
p

values are in good agreement with experimental values and useful for rapid hazard assessment of
ce

energetic materials.

Keywords: Heat of explosion, Heat of formation, High energy materials, Computer-aided design
*
corresponding author email: [email protected] or [email protected]
Ac

INTRODUCTION

One of the most important thermodynamic parameters that determine the performance of

explosives and propellants[1], which are broadly classified as high energy materials (HEMs) is their heat

of explosion (He). Heat of explosion (He) is the quantity of heat released (illustrated in figure 1) when a

HEM undergoes detonation as an explosive (eg. RDX, HMX, TNT etc.,) or deflagration (burning) as a

propellant (eg. a rocket propellant). Both these processes of detonation and deflagration occur even in

the absence of external oxygen or air, because of the fact that the molecule of a HEM carries its own

1
Page 1 of 11
oxygen. He is a quick and reliable criterion to evaluate the performance potential of a HEM[2]. For

example the higher the He (usually expressed in cal/g or kJ/kg) of a rocket propellant the higher will be

its “specific impulse”, an index of propelling power of a rocket using this propellant. In recent past,

theoretical calculations to predict detonation behaviour of explosives have evinced great interest[3–7].

t
Theoretical computation of notional materials allows for identification of promising candidates for

ip
additional study and elimination of poor candidates from further consideration, and thus, reducing costs

associated with synthesis, and evaluation of the materials. This capability of computer-aided design leads

cr
to better designs and shorter design cycles.

us
For the existing computer codes such as BKW code, TIGER code, VLW code, etc., the user has

to give molecular weight, density, heat of formation and many complex thermodynamic parameters as

an
input parameter. Also these software’s requires huge constants such as coefficients of EOS -

coefficients for Entropy fit - A, B, C, D, F, Enthalpy integration constant - IC, Heat of formation
M
(Elements at 0K to Species at 0K) co-volume, etc.. Also the number of coefficients different when different

type of EOS is used. It is also observed that for the same EOS like BKW EOS, coefficients of EOS -
d

has been adjusted to different extent by different researchers. Apartment from the huge input
te

constants, the user has to input some guess values for P, V, T to start the iteration as well as initial guess
p

of the concentrations of the products. Some time the iteration won’t converge due to improper / wrong
ce

initial guess values.

HEMs researchers all over the world are in the constant pursuit of conceiving and synthesizing a
Ac

large number of compounds, the molecules of which are expected to have high He values. Such a

synthesis involves a lot of procedural challenges apart from high cost and hazards[8]. Thermochemical /

hydrodynamic computer codes such as BKW (Charles L. Madder, first in 1956 for IBM -704, STRETCH

BKW in 1961 for IBM 7030, using Fortran IV BKW in 1967)[9–12], RUBY (Levine and Sharples in

1962)[13], PANDA (Gerald I. Kerley in 1981) [14] and JCZS [15]have been reported in literature for the

prediction of various thermodynamic as well as detonation parameters. But these codes involves too

many complex input parameters. It is therefore useful to develop a computational approach to predict the

He without any complex input parameters and without any experimental data as input parameters. In the

2
Page 2 of 11
present work, the computational approach combines Stines & Kramer method[16] to compute heats of

formation of HEMs and Kistiakowsky Wilson (K-W) rule for predicting possible explosion products[17].

The results obtained by this computational approach are comparable with experimental results reported in

literature.

COMPUTATIONAL APPROACH

t
ip
Computational approach described in this paper involves a systematic procedure in which

thermodynamic parameters are computed from its structural information. The advent of advanced

cr
computer system has dispelled one of the principal drawbacks of laborious time consuming mathematical

us
calculations. Sufficient experimental data on estimating Hf for HEMs in different phase states has been

collected[18–20]. This set of data forms the basis for computing He for various HEMs. Computation of

an
Hf and He from the model described below is versatile, convenient and require structural information

only.
M
In the present computational approach, the structural information of HEMs is given as input for

the computation of the heat of formation of energetic reactants using Stine and Kramer method[16]

extended to terminal groups such as –OH, -NH, -NH2, C=O and hydrogen attached to aromatic and
d

nonaromatic carbon[21]. To calculate the He and other explosive performance parameters, a knowledge
te

of the composition of the gaseous products of explosion is essential[19]. The software developed by the
p

authors ensures that once the molecular structure of a HEM is known, it is converted in to coded
ce

structural information giving various types of bonding involved in the molecule thereby, predicting the

following: (i) most probable products of explosion based on Kistiakowsky–Wilson hierarchy rules (ii) Hf of

the products (iii) Hf of the energetic reactant and (iv) the resultant heat of explosion. The logic of above
Ac

discussed is illustrated in Figure 2. The heat of formation of most common explosion products are given

in Table – I. Since the data shown in table – I is stored inbuilt in LOTUSES (version 1.5), it will compute

the sum of heat of formation of explosion products (Hf of the products). Once the Hf of the energetic

reactant and Hf of the products are computed by LOTUSES (version 1.5) , it will automatically compute

the heat of explosion (He) using the following well known relation.

H e   H f (products) -  H f (reactant)

3
Page 3 of 11
ABOUT LOTUSES (version 1.5)

The software LOTUSES (version 1.5) is developed to run on windows operating system using

modern modular and object-oriented techniques in Visual Basic 6.0. This installation package occupies

about 50 MB of hard disk space in the computer and can be installed in Windows 98, 2000 as well as in

Windows XP. In addition to thermodynamic parameters such as heat of explosion, heat of formation it

t
ip
also predicts performance parameters of HEMs such as velocity of detonation, CJ Pressure, density,

relative strength, power index etc.[21–25]. The output of the programme can be directly printed (attached

cr
to the standalone PC or LAN) as well as stored in various formats (.txt, .html, .doc, .xls, .pdf).

us
RESULTS AND DISCUSSIONS

an
Heat of explosion for various HEMs are computed by using above discussed logic (illustrated in

Figure 2) and the results are presented in Table – II. The Hf of reactants were computed using Stine &

Kramer method. The possible explosion products were predicted by K-W rule applicable to most of CHNO
M
type HEMs and corresponding standard heat of formation were computed. The computational approach

has been validated by comparing the predicted He values with the experimentally determined results
d

reported in the literature[20]. It does not include any heat generated by secondary reactions of the
te

explosive or its products with air.

The predicted heats of explosion and corresponding experimental values for twenty HEMs are
p

plotted in Figure-3. The linear regression analysis of all data points in Figure-3 yields the correlation
ce

2
coefficient R = 0.9721 with a linear equation y = 0.9262x + 101.45. The correlation coefficient value

0.9721 reveals that the computed values are in good agreement with experimental values. Hence the
Ac

algorithm included in the software code LOTUSES (version 1.5) for the computation of heat of explosion

will be useful to design futuristic high energy molecules without any experimental data.

CONCLUSION

The computational approach described in this paper predicts He values from the structural

information of HEMs. The methodology combines the prediction of heat of formation (Hf) of energetic

reactants from its structural information using Stine and Kramer method and explosion products required

for the estimation of heat of explosion, which is calculated by LOTUSES based on Kistiakowsky–Wilson

4
Page 4 of 11
hierarchy rules. Hence less input parameter (only structural information) is sufficient to compute the heat

of explosion using LOTUSES (version 1.5). The variation between the computed and experimentally

determined He values was calculated and the regression coefficient value 0.9721 shows that predicted

results are in good agreement with the experimental results reported in literature. The output of the

t
LOTUSES (version 1.5) can be directly printed (attached to the standalone PC or LAN) as well as stored

ip
in various formats (.txt, .html, .doc, .xls, .pdf). The computational approach for this computation is simple,

convenient and doesn’t require any complex input parameters. This computer-aided design will leads to

cr
better designs and shorter design cycles as well as hazard assessment of energetic materials.

us
ACKNOWLEDGEMENT

an
Authors are highly grateful to thank Shri Surendra Kumar, Director, Armament Research and

Development Establishment and Shri. A. Subhananda Rao, Director, High Energy Materials Research

Laboratory, Sutarwadi, Pune for providing infrastructure to carryout this work.

M
REFERENCE
d

1. Edward E. Baroody and Susan T. Peters, Heats of Explosion, Detonation and Reaction Products:
te

Their estimation and relation to the first law of Thermodynamics, 1990 JANNAF, Propulsion meeting

Vol – 1, CPIA Publications 550, AD-A240864, Columbia, USA, (1990)

p

2. Edward E. Baroody and Joan S. Ross, The use of Heats of Explosion and Heats of Formation to
ce

evaluate new energetic materials, Proceedings of the American Defence preparedness Associations,

Energetic Materials Technology (1992)

Ac

3. C. L. Mader, Numerical Modeling of Explosives and Propellants, 2nd Edition, CRC Press, New York,

1998.

4. P. Politzer, P. Lane and M. C. Concha, in: Peter Politzer and Jane S. Murray (Editors), Energetic

Materials – Part 1 - Theoretical and Computational Chemistry, Vol 12, Elsevier, Amsterdam, 2003,

ch. 9, p.247–277

5. Betsy M. Rice, Jennifer Hare, Predicting heats of detonation using quantum mechanical calculations,

Thermochimica Acta 384 (2002) 377-391

5
Page 5 of 11
6. Dmitrii V. Sukhachev and Tatyana S. Pivina, Estimation and Prediction of the Heats of Formation for

Non-Aromatic polynitro compounds on the basis of the QSPR approach, Propellants, Explosives,

Pyrotechnics 19, (1994) 159-164

7. S.Bourasseau, A systematic procedure for estimating the standard heats of Formation in the

condensed state of non aromatic polynitro compounds, Journal of Energetic Materials, 8(5), (1990)

t
ip
416

8. Tatyana S.Pivina, Vladimir V.Shcherbukhin, Marina S.Molchanova and Nikolai S.Zefirov, Computer-

cr
Assisted Prediction of Novel Target High-Energy Compounds, Propellants, Explosives, Pyrotechnics,

us
20 (1995)144-146

9. C.L. Mader, Stretch BKW - A Code for Computing the Detonation Properties of Explosives, Los

an
Alamos Scientific Laboratory, Report LADC-5691, 1962.

10. C.L. Mader, Detonation Performance Calculations Using the Kistiakosky and Wilson Equation of

State, Los Alamos Scientific Laboratory, Report U-2613, 1961.

M
11. C.L. Mader, Detonation properties of condensed explosives computed using the Becker-Kistiakosky-

Wilson equation of state, Los Alamos Scientific Laboratory Report LA-2900, 1963.
d

12. C.L Mader, Numerical modeling of detonations, University of California Press, 1979.
te

13. H.B. Levine, R.E. Sharples, Operator's Manual for RUBY, Lawrence Livermore Laboratory Report

UCRL-6815, Livermore, CA, 1962.

p

14. Gerald I. Kerley, User's Manual for PANDA: A Computer Code for calculating Equations of State, Los
ce

Alamos National Laboratory Report No. LA-8833-M, 1981.

15. Michael L. Hobbs, Melvin R. Baer, Bruce C. McGee, JCZS: An Intermolecular Potential Database for
Ac

Performing Accurate Detonation and Expansion Calculations , Propellants, Explosives, Pyrotechnics,

24(5), (1999) 269 - 279

16. J.R.Stine and J.F.Eramer, Estimation of Heats of formation and the development of chemical
th
databases useful for energetic materials, 26 JANNAF Combustion meeting, Jet Propulsion

Laboratory, Pasadena, California, CPIA Publication, Vol. II, (1989) p.53-56.

17. Jacqueline Akhavan, The chemistry of Explosives, The Royal Society of Chemistry, UK (1998)

18. ICT Database of Thermochemical Values, Version 3.0, Fraunhofer Institut Chemische Technologie,

(2001)

6
Page 6 of 11
19. A.Bailey and S G Muray, Explosives, Propellants and Pyrotechnics, Royal Military college of Science,

Brassey’s, UK (1989)
nd
20. Rudolf Meyer, Explosives, 2 revised and extended edition, Verlag Chemie, Germany (1981)

21. H. Muthurajan, R Sivabalan, M.B.Talawar and S.N.Asthana, Development of Novel Computer code

for prediction of heat of formation of High Energetic Materials, proceeding of 7th International

t
ip
Seminar on New Trends in research of Explosives, Czech Republic, (2004) p. 202-223

22. H. Muthurajan, R. Sivabalan, N. Venkatesan, M.B.Talawar and S.N.Asthana, Computational

cr
th
approaches for performance prediction of High energy Materials, Proceedings of 4 International

us
High Energy Materials Conference, India (2003) p.470-486

23. H Muthurajan, R Sivabalan, M.B.Talawar and S.N.Asthana, Artificial Intelligence methodology for

an
th
Thermodynamic Analysis of High Energetic Materials, Proceedings of 14 National Symposium on

Thermal Analysis, India, (2004) p.225-228

24. H. Muthurajan, R Sivabalan, M.B. Talawar and S. N. Asthana, Computer simulation for prediction of
M
performance and thermodynamic parameters of high energy materials, J. Hazard. Materials. 112

(2004) 17.
d

25. H. Muthurajan, R. Sivabalan, M.B.Talawar, S.Venugopalan and B.R.Gandhe, Development of

te

Computer Code for the Theoretical Performance Prediction of Explosive Formulations, Proceedings
th
of 5 International High Energy Materials Conference, India (2005) 29
p
ce
Ac

7
Page 7 of 11
 Input (Energy of activation)

t
ip
Reactant
(explosive)
Energy

cr
Output ( Heat of explosion )

us
Explosion Products

an
Reaction Coordinate

Fig. 1. Energy profile of an explosion reaction

M
d
p te
ce
Ac

8
Page 8 of 11
 INPUT
Structure of HEMs

Splitting of structure into

individual element Balance the explosion reaction
by K-W Rule

t
ip
cr
Scan the nature of bonds
attached to the each element Explosion products

us
Comparison
with database, Summation an Comparison
with database, Summation
M
of matched values of matched values
d
te

Predicted Heat of formation of Calculate heat of formation of

p

the reactant the explosion products

ce

Computation of
Ac

Heat of Explosion

OUTPUT
Heat of explosion

Fig. 2. Logic for computation of heat of explosion by LOTUSES

9
Page 9 of 11
 7500
y = 0.9262x + 101.45
6500 2
R = 0.9721
D He (Predictedl)

5500

t
ip
4500

3500

cr
2500

us
1500
1500 2500 3500 4500 5500 6500 7500 8500

DHe (Experimental)

an
Fig. 3. Comparison of Heat of Explosion
M
Table – I. Heat of formation of common explosion products
d

Molecular weight Heat of formation

Formula Name
te

(g/mol) (Hf ) (kJ/mol)

CO Carbon monoxide 28 -111.8
p

CO2 Carbon dioxide 44 -393.5

H2O Water 18 -240.6
ce

H2 Hydrogen molecule 2 0
O2 Oxygen molecule 12 0
N2 Nitrogen molecule 14 0
Ac

10
Page 10 of 11
 Table – II Comparison of experimental value and predicated by LOTUSES

Heat of explosion
Sl Empirical Error
HEM's Name (kJ/kg)
No formula %
Experimental LOTUSES

1 Bi-trinitroethyl nitramine (BTENA) C5H5N8O14 4857 5321 -9.55

2 Bitrinitroethylurea (BTNEU) C5H6N8O13 6542 6382 2.45

t
ip
3 Butene triol trinitrate C4H7N3O9 6153 5737 6.76

4 Cyclonite (RDX) C3H6N6O6 5723 5087 11.11

cr
5 Dinitronaphthalene C10H6N2O4 2635 2506 4.9

6 Dioxyethylnitramine dinitrate (DINA) C4H8N4O8 5384 5021 6.75

us
7 Dipentaerythritol hexanitrate (DIPEHN) C10H16N6O19 5208 4814 7.57

8 Erythritol trinitrate C6H11N3O9 4110 3725 9.36

an
9 Ethylene dinitramine (EDNA) C2H6N4O4 5343 5179 3.06

10 Mannitolhexa nitrate (MHN) C6H8N6O18 6380 5937 6.95

11 Methyl nitrate CH3NO3 6869 6562 4.46

M
12 Metriol trinitrate C5H9N3O4 4992 4532 9.21

13 Ethylene glycoldinitrate (EGDN) C2H4N2O6 7390 6881 6.89

d

14 Nitroisobutylglycerol trinitrate C4H6N4O11 7755 7244 6.59

te

15 Nitromethane CH3NO2 4763 4718 0.94

16 Nitrourea CH3N3O3 3865 3645 5.69

p

17 Pentaerythritol trinitrate (PETRIN) C5H9N3O10 5301 4704 11.27

ce

18 Pentaerythritol tetranitrate (PETN) C5H8N4O12 6404 6130 4.28

19 Tetranitromethane (TNM) CN4O8 2259 2206 2.33

Ac

20 Tetryl C7H5N5O8 4773 4527 5.16

11
Page 11 of 11