Thermal Conductivity of Explosives and Energetics in Support of Self-Accelerating Decomposition Temperature (SADT)

Written by Jacob Pierce, Market Analyst

Self-Accelerating Decomposition Temperature is defined as the lowest possible temperature that a packaged material may reach, at which it begins to undergo the process of self-accelerated decomposition within a week (typically when the center of the package heats to 6 °C/K above the environment temperature) [1]. Self-accelerated decomposition is the process in which the rate of chemical decomposition is generating heat quicker than it may be dissipated by the material into the environment.

Figure 1: Explosion in Beirut, Lebanon. In 2020 2750 tons of ammonium nitrate exploded, killing over 200 people [2]

The SADT is a crucial value to consider when it comes to dealing with certain hazardous chemicals such as energetic powders, general self-reactive materials, and organic peroxides, due mainly to the increased risk of mishaps including fire, explosions, and toxic fumes.

A report was presented by Whitmore, Gladwell, and Rutledge, in which a case of a chemical explosion is explored and summarized [3]. The explosion occurred during the manufacturing of an arzodicarbonate formulation (AC), specifically within the drying period of the formulation. The vessel housing the chemical was torn open completely and struck the floor above. There was extensive damage that occured, with windows being shattered from up to 90 m away from the vessel, and a TNT blast equivalence estimated at 3.3 kg. It was ultimately determined that the explosion was a result of a propagating decomposition, likely provoked by the deflagration (locally initiated decomposition). This case among others like it clearly illustrate the importance of thermal properties when dealing with dangerous chemical compounds, such as SADT tests.

An additional case demonstrating the importance of thermal property assessment of reactive materials is the 2020 Beirut port explosion. A cargo ship containing a large quantity of ammonium nitrate (approximately 2750 tons) was stored at the port for several years, among other known/unknown substances such as 23 tons of fireworks [4]. The exact cause of ignition is still unclear as of 2022; however, an explosion of such magnitude (seismic event of magnitude 3.3) [5], was sufficient to cause significant casualties in the hundreds, mass injuries, and property damages ranging in the billions, as well as an estimated 300 000 individuals becoming homeless.

UN/OECD seminar following the Beirut Explosion [6]

A seminar was presented by the UN/OECD following the explosion that aimed to manage future risks involving ammonium nitrate and can be seen above [6]. The seminar provided details on the classification of ammonium nitrate/ammonium nitrate-based fertilizers according to their explosive, oxidizing, and self-sustaining exothermic decomposition properties. 

SADT – Classification Procedures, Test Methods and Criteria relating to Self-Reactive Substances

__________________________________________________________

The values obtained for SADT are used to determine whether a certain packaged product requires special cooling/heating considerations during the storage or transport of the material. These procedures for obtaining SADT are defined by the United Nations in Part II of Classification Procedures, Test Methods and Criteria relating to Self-Reactive Substances of Division 4.1 and Organic Peroxides of Division 5.2 [1]. The results from this test are typically used to assess the hazard probability of the packaged material. The primary method presented in this document is test code H.1, United States SADT test (section 28.4.1). This method amongst others presented in the UN document is plagued by disadvantages, requiring 7 days to complete a test, needing large sample sizes ranging from 400 g to 200 kg [7], and presenting a considerable amount of danger with each test iteration. This presented a need for a better option, free of the above issues, which inevitably arrived in the form of simulation.

Case Study: Akoustis Technologies, armasuisse, University of Applied Sciences of Western Switzerland

__________________________________________________________

Figure 2: Experimental setup of a cook-off experiment as presented by Akoustis Technologies, armasuisse, and the University of Applied Sciences of Western Switzerland [8]

A paper was presented by Akoustis Technologies, armasuisse, and the University of Applied Sciences of Western Switzerland [8], which evaluates a kinetic-based simulation approach to determine the value of SADT, alongside the Frank-Kamenetskii approach for describing the rate of heat generation. This process of determining SADT is far more suitable for performing evaluations with less risk involved, and for laboratory environments dealing with small sample sizes. With this simulation, there are several factors that must be considered to properly characterize the decomposition process. These factors may be split into categories such as the following:

(1) Extrinsic Properties

1. The sample masses
2. The geometry of
   1. Sample holder
   2. Container
3. The applied heating mode
   1. Slow or fast cook-off
   2. Heat-wait-search mode
   3. Isothermal or adiabatic run

(2) Intrinsic Properties

1. Kinetic parameters of decomposition
   1. Activation energy
   2. Pre-exponential factor in Arrhenius equation
2. Physical-chemical properties
   1. Heat capacity
   2. Material density
   3. Thermal Conductivity

It was reported [8] that thermal conductivity has a large influence on the time-to-ignition value under quasi-thermal conditions. It was also noted that precise values of thermal conductivity are required to accurately simulate time/temperature dependence during the cook-off experiment. Thermal conductivity becomes increasingly important when applying the kinematic parameters to the material volume scale-up estimation [9], done to predict the thermal properties from the small-volume measured material to a theoretical larger quantity. This is illustrated through the use of the sample heating rate expression based on the Frank-Kamentetskii theory considering heat balance in solid materials, expressed below in equation 1:

Any packaging surface that is exposed to its surroundings is subject to radiant and convectional heat transfer. These heat exchange-with-surroundings properties causes the thermal characteristics of the sample to change significantly with varying volume size. The following boundary condition equations can be used to accurately predict the temperature at any location within the package, expressed below as equations 2 through 4 [9].

		— Eq 2
where:		— Eq 3
and:		— Eq 4

How to measure thermal conductivity of energetic/explosive materials?

__________________________________________________________

It is well known that measuring the thermal conductivity of energetic materials is rather challenging to execute safely. A paper released by Sorenson and Harris [10] highlights the main issues faced when attempting to perform these measurements.

The Modified Transient Plane Source (MTPS) sensor, able to quickly measure the thermal conductivity of energetic materials with a low heat pulse. This sensor was employed in the work of Sorenson and Harris.

Energetic materials are typically sensitive to initiation due to impact, friction, and electrostatic discharge. Additionally, large sample volumes being dangerous to handle, and some sample geometries being impractical to work with, often result in an estimated value of thermal conductivity. This often results in using book values for thermal conductivity, which often reduces the accuracy of the SADT estimation. This presented a need for the measurement of λ, but with the potential for ignition when dealing with laser flash, and the large volumes required for steady-state methods, a non-destructive measuring method with lower volume capabilities will be the easiest and safest choice for measurements.

As for the paper mentioned previously, the researchers used the simulation method of predicting the SADT for AIBN (azobisisobutyronitrile). The critical thermal conductivity values used in the simulation were found using the MTPS method of measuring thermal conductivity from C-Therm Technologies, capable of measuring values of λ from 0 to 500 W/mK with accuracy better than 5%. The figure below outlines the results of the simulation done for AIBN with varying masses, dependent on the measured thermal conductivity λ, and heat transfer coefficient U. Actual values of SADT are noted above their respective curves.

Figure 2: AIBN thermal behavior simulation for masses of 50 kg (a), 20 kg (b), and 5 kg (c), respectively

From the figure above, and through the conclusions established by the researchers involved in the study, it was determined that software-based simulations of self-accelerating decomposition are successful in determining values of SADT, and in determining accurate values of SADT for scaled-up volumes of the sample material, if there was an initial accurate measurement of thermal conductivity. 

For more in depth information on the thermal conductivity measurement of explosive materials, a webinar was co-presented by Harris on the precautions that need to be taken when measuring explosive, energetic, or unstable materials. This webinar is available here, and provides an expansion on the work presented above.

This has been a part of our Explosives application.

______________________________________________________________________

References

[1] United Nations. (2019). Classification Procedures, Test Methods and Criteria Relating to Self-Reactive Substances of Division 4.1 and Organic Peroxides of Division 5.2. unece.org. Retrieved August 8, 2022, from https://unece.org/DAM/trans/danger/publi/manual/Rev5/English/02en_part2.pdf

[2] Hubbard, B, et al. (2021).  How a Massive Bomb Came Together in Beirut’s Port. New York Times. https://www.nytimes.com/interactive/2020/09/09/world/middleeast/beirut-explosion.html 

[3] Whitmore, M. W., Gladwell, J. P., & Rutledge, P. V. (1993). Report of an explosion during the manufacture of an azodicarbonamide formulation. Journal of Loss Prevention in the Process Industries, 6(3), 169–175. https://doi.org/10.1016/0950-4230(93)85006-7

[4] Moafi, S. (2020, November 17). The Beirut Port Explosion. Forensic architecture. Retrieved August 2022, from https://forensic-architecture.org/investigation/beirut-port-explosion

[5] National Institute of Standards and Technology (NIST). (2020, August 4). M 3.3 Explosion – 1 km ENE of Beirut, Lebanon. USGS earthquake hazards program. Retrieved August 2022, from https://earthquake.usgs.gov/earthquakes/eventpage/us6000b9bx/executive

[6] United Nations. (2021). Un/Oecd Seminar in follow-up to the 2020 Beirut Port Explosion. United Nations. Retrieved August 2022, from https://media.un.org/en/asset/k17/k17jc6msyv.

[7] Yu, Y., & Hasegawa, K. (1996). Derivation of the self-accelerating decomposition temperature for self-reactive substances using isothermal calorimetry. Journal of Hazardous Materials, 45(2-3), 193–205. https://doi.org/10.1016/0304-3894(95)00092-5

[8] Roduit, B., Folly, P., Sarbach, A., Berger, B., & Mathieu, J. (2019). Determination of SADT and Cook-off Ignition Temperature by Advanced Kinetic Elaboration of DSC Data. Journal of Thermal Analysis and Calorimetry, 117(3), 1017–1026.

[9] Roduit, B., Hartmann, M., Folly, P., Sarbach, A., Brodard, P., & Baltensperger, R. (2014). Determination of thermal hazard from DSC measurements. investigation of self-accelerating decomposition temperature (SADT) of AIBN. Journal of Thermal Analysis and Calorimetry, 117(3), 1017–1026. https://doi.org/10.1007/s10973-014-3903-3

[10] Harris, A., & Sorensen, D. N. (2007). Thermal Conductivity Testing of Minimal Volumes of Energetic Powders. Journal of Pyrotechnics, (25), 49–54.