Page 1 of 8

European Journal of Applied Sciences – Vol. 11, No. 4

Publication Date: August 25, 2023

DOI:10.14738/aivp.114.15364

Partom, Y. (2023). What May Happen After a Thermal Explosion. European Journal of Applied Sciences, Vol - 11(4). 268-275.

Services for Science and Education – United Kingdom

What May Happen After a Thermal Explosion

Y. Partom

18 HaBanim, Zikhron Ya'akov 3094017, Israel

ABSTRACT

Thermal explosion may happen when an explosive is in a temperature field, and

when its boundary temperature is between the critical and ignition temperatures.

Semenov and Frank-Kamenetzky solved the thermal explosion problem

analytically in 1D and for a constant boundary temperature [1,2]. Since the 1960s

thermal explosion problems have been simulated numerically in 3D and for

realistic boundary conditions. In thermal explosion tests it is easier to apply the

boundary temperature gradually, and in this way, they become cookoff tests. When

thermal explosion happens in a small region of an explosive body, the events that

follow, and their overall resultant violence, may be quite diverse, like: 1) a slow

decaying pressure wave; 2) a fast non-decaying wave; 3) a strengthening pressure

wave that builds up to shock initiation and detonation. The outcome of a thermal

explosion depends on: 1) the sensitivity of the explosive; 2) the temperature field

throughout the explosive body at the time of thermal explosion; 3) the geometry of

the explosive body; and 4) the degree of confinement of the explosive body. To

model what may happen after a thermal explosion event, we use our PDSR (=

Pressure Dependent Shear Reaction) together with our TDRR (= Temperature

Dependent Reaction Rate) reactive flow models. For each computational cell these

two models work in sequence. Initially there is a shear reaction handled by PDSR. If

as a result, pressure and temperature there go beyond the threshold for reaction

out of hot spots, TDRR takes over to compute shock initiation and detonation. We

present here computed examples of different outcomes of thermal explosion

events.

INTRODUCTION

Thermal explosion is a dangerous safety event when considering the response of explosives to

heat. A thermal explosion may happen in an explosive device when it is surrounded by a

temperature field that is between what is known as a critical temperature and the explosive

ignition temperature. Ignition temperature is the explosive boundary temperature above

which a self-sustaining combustion wave propagates into the explosive, given that the

combustion products flow away very quickly. Ignition temperature of common explosives is

around 500-600K. When the boundary temperature is above the ignition temperature, a steady

combustion wave propagates into the explosive, as long as the combustion products are

unconfined. Such a combustion wave is usually rather slow, around 1mm/s. When the explosive

boundary temperature is below the ignition temperature, a combustion wave is not formed,

and heat flows into the explosive at a rate determined by heat transfer. The front temperature

of this heat wave may slowly increase because of two reasons:1) slow bulk reaction with an

Arrhenius reaction rate; and 2) slow heat transfer towards the boundary, which decreases as

the heat front moves away from the boundary. Because of the highly nonlinear nature of the

Arrhenius rate equation, whenever the heat front temperature goes above a certain threshold,

Page 2 of 8

269

Partom, Y. (2023). What May Happen After a Thermal Explosion. European Journal of Applied Sciences, Vol - 11(4). 268-275.

URL: http://dx.doi.org/10.14738/aivp.114.15364

the reaction process accelerates exponentially, and the temperature at the front increases to

high values in less than a μs. As a result, the explosive there undergoes a fast bulk reaction,

which is known as thermal explosion. For decreasing values of the boundary temperature

below the ignition temperature, the thermal explosion location in the explosive body moves

further away from the boundary. There is therefore a low boundary temperature for which the

thermal explosion distance from the boundary is the furthest. For a spherical explosive body

this would be the center of the sphere. The boundary temperature for which the thermal

explosion location is the furthest from the boundary is known as the critical temperature Tc.

When the boundary temperature is below Tc, a thermal explosion event would not happen. The

phenomenon of thermal explosion in energetic materials was discovered and analyzed in

Russia some 80 years ago by Semenov and Frank-Kamenetzky [1, 2]. As there were no

computers at that time, analysis of this highly nonlinear problem was quite difficult. They were

able to overcome the difficulties by treating the 1D configurations of the problem with

asymptotic approximations. Since the 1960s thermal explosion problems are treated with

computer simulations [3], which make it possible to predict the time and location of thermal

explosions for explosive bodies of various geometries and boundary conditions.

The problem of predicting the time and location of a thermal explosion (known as the Frank- Kamenetzky problem) is quite important for explosive safety considerations. Sometimes it is

enough to know if a thermal explosion is to be expected, and if so, how to avoid it by some

design change. But sometimes it is also important to predict what may happen after a specific

thermal explosion. This part is known as level of violence of a thermal explosion event. In

reality these are two stages of the same process. The first stage (heating stage) includes heat

transfer and slow reaction, and pressure rise is minimal. It ends with a very fast pressure rise

over a small volume of the explosive. In the second stage a weak pressure wave spreads out of

this high-pressure volume, which may subsequently ignite the explosive around it by means of

plastic flow localization and shear band formation [4, 5]. Deflagration fronts then spread out of

the ignited shear bands, similar to deflagration waves out of hot spots. But reaction out of shear

bands (called shear reaction) is much slower and less violent than reaction out of hot spots.

This is mainly because the distance between shear bands is an order of magnitude larger than

the distance between hot spots. On the other hand, ignition threshold of shear bands is much

lower than that of hot spots. Shear bands may ignite from a pressure wave of 0.1-0.2GPa, while

hot spots ignite at 1-2GPa.

The second stage of a thermal explosion process may become rather involved, because the

shear reaction wave may strengthen to become a detonation wave. Such strengthening may

happen in various ways: 1) self-strengthening, when the initial pressurized volume is relatively

high; 2) geometric convergence; 3) reflection from a rigid boundary; and 4) interaction of two

shear waves. This is why transition from shear reaction to shock initiation and detonation

depends strongly on: 1) thermal explosion location; 2) device geometry and 3) its level of

confinement.

From the view point of computer simulations, the two stages of thermal explosion are two

entirely separate problems, to be handled with different computational tools. The first stage

(heat transfer + slow reaction) is slow, and its computer simulation may ignore pressure

variations and use ms time steps. The second stage may be a fast hydrodynamic process, with

Page 3 of 8

Services for Science and Education – United Kingdom 270

European Journal of Applied Sciences (EJAS) Vol. 11, Issue 4, August-2023

ns time steps and high-pressure waves. This is why the two stages should be treated with

different types of codes, the first stage with a heat transfer code augmented by an Arrhenius

reaction rate, and the second stage with a hydrocode. What needs to be transferred from one

code to the other are the temperature field, the reaction parameter field, and the thermal

explosion location.

In what follows we show results from computed examples of the second stage of thermal

explosion events.

COMPUTED EXAMPLES

The hydrocode we’re using includes two of our previously developed reactive flow models that

work consecutively for each computational cell. The first is a shear reaction model that we call

PDSR (= Pressure Dependent Shear Reaction). We calibrated PDSR for an LX07 like explosive

from our Steven test data. The second is our shock initiation and detonation (or surface burn)

model which we call TDRR (= Temperature Dependent Reaction Rate) [6], that we calibrated

in the past for the same explosive from pop plot data.

The explosive body in our computed examples is a cylinder of 180mm diameter and 380mm

length. We compute only one half of this cylinder, z>0, as it has a symmetry plane at z=0. Around

the explosive we have an aluminum envelop 10mm thick. We assume axial symmetry, and the

computational mesh is 1 cell per mm in both directions. To start stage 2 of the thermal

explosion event we: 1) put an initial temperature T0 over all the explosive cells. T0 represents

an average of the boundary temperature and the somewhat increased temperature near the

thermal explosion location just before the thermal explosion. In the examples that follow we

use T0=350K, and the initial temperature of the explosive at the beginning of the second stage

has been specified accordingly; 2) choose the thermal explosion location cell. In the examples

that follow there are two such locations; 3) let part of the explosive around the cell center

detonate out of the cell center with a programmed burn scheme, and we control the mass of the

explosive that undergoes thermal explosion by changing the radius of the detonating sphere;

and 4) run our combined PDSR-TDRR reactive flow model.

As explained above, the pressure wave spreading out of the thermal explosion is too weak to

cause shock initiation, but it may cause shear initiation, which is handled by PDSR. Reaction

ignition in PDSR depends on two consecutive criterions: 1) shear localization criterion for

starting shear band formation given by:

( )L

PD 0.01GPa / s =  (1)

where P=pressure, D=effective plastic strain rate, and L stands for Localization. In the past we

calibrated this localization criterion from simulations on the mesoscale [7], and it may be

generally stated that around a thermal explosion this location criterion is usually met; and 2)

ignition delay (from localization to ignition). Ignition delay (h) depends on the product PD as

well, and we calibrated it in the past from Steven test data [7]. Ignition delay is low for high PD

and high for low PD. Ignition delay in a cell is determined by the coefficient Ah, calibrated from

a preliminary short simulation as explained next. Ah is given by: