Propellants, Explosives, PyrotechnicsVolume 43, Issue 5 p. 433-435 EditorialFree Access Hazard Assessments of Energetic Systems, A Field Still in Development First published: 14 May 2018 https://doi.org/10.1002/prep.201880531AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat I am delighted and honoured to be invited to write an editorial for this edition of Propellants, Explosives and Pyrotechnics, conscious of the immense contribution of this Journal to understanding of the behaviour of energetic materials. Throughout my career I have been seeking to contribute to improved understanding of the safety of energetic materials and munitions in abnormal and hostile environments. In recent years, I have also become interested in coaching and mentoring early-career scientists and engineers. This has led me to consider the lessons we have learnt from historical events and the critical role of high quality investigations into explosives mechanisms. When the consequences of unintended initiation of munitions are very high (for example, because they are part of systems containing significant quantities of energetic, toxic or radioactive material) accountable organisations may be required to perform a Quantified Risk Assessment (QRA). This enables these organisations to demonstrate that risks are acceptable or tolerable and have been reduced so far as is reasonably practicable and that the operations comply with regulatory requirements. This provides significant challenges. Energetic materials have been studied extensively for many decades but we do not yet know enough to predict the response of explosives in all operational and abnormal environments with high confidence and thus assessments are usually cautiously pessimistic. It is self-evident that process owners need to give high priority to reducing the potential for exposure to unintended stimuli. The response of energetic systems to the full range of credible environments is extremely complex involving: energy transfer functions; hot-spot formation at the microstructural level; endo- and exothermic decomposition; and combustion which is strongly dependent on the material composition, particle characteristics, material failure modes and confinement. Yet, most qualification tests only probe behaviour at the macroscopic level. Materials qualification usually involves comparative measurement in a small number of standard tests, loosely related to real environments; systems tests provide evidence for more representative scale, build standards and environments but limited opportunity to probe detailed mechanisms. Trials engineers are faced with a dilemma because they wish to test articles which are fully representative of the service build standard and are therefore reluctant to include intrusive diagnostics that might otherwise help gain insight into the underlying initiation mechanisms. These approaches provide insufficient numerical data to derive statistically significant estimates of likelihood of reaction and often only address a small subset of credible environments. High reliance is usually placed on benign response to a few ‘so-called' worst-case tests in standard maximum credible environments assuming a monotonic response to progressive increases in threat stimulus. The history of military and commercial manufacture, storage and deployment leads us to expect sporadic accidents, followed by investigation, targeted research and new test methods; this reinforces the fragility of our understanding. We have come to learn that maximum credible environment tests may not represent the worst-case tests after all. Over the past two decades, in response to its Health & Safety commitments, legal requirements and regulatory expectations, the UK's Atomic Weapons Establishment has evolved an approach to QRA for processes involving explosives assemblies. A multi-disciplinary team of: process operators; facility, process, and design engineers, and human factors, materials and explosives specialists work alongside risk analysts. Together they systematically consider potential faults during normal operations, foreseeable process deviations and abnormal events for each element of the munition's manufacturing to deployment and decommissioning lifecycle. Activities include: task analysis, engineering analysis to assess equipment failure modes and identification of threats and their characteristics. These are supported by the use of finite element analysis and, increasingly in recent years, virtual reality techniques combined with human reliability analysis. Bow-tie analysis, accident sequence and fault trajectory diagrams are then used to evaluate Lines of Defence against each identified threat and allow consideration of the likelihood of defeating or by-passing each protective feature and the potential consequences. This integrated activity takes into consideration the build standards and confinement of the assembly; the way in which they may be changed by the threat; the types and intensity of stimuli applied to the energetic materials and components, and the damage which could result. It would be impossible to undertake physical testing for all of the possible stimuli that an article might see during its service life. The practical alternative is to estimate relative likelihoods of ignition and growth of reaction for each potential fault scenario. These are then rank-ordered and combined with estimated failure mode frequencies to inform the risk analysis and operational safety case, supported by comparison with data from observed process deviations, equipment failure and unplanned events. Where shortfalls in evidence are identified, specific process-related experiments and modelling are commissioned. Although it may not be possible to rigorously justify the numerical estimates, the rank-ordering process allows process owners to identify opportunities to reduce risk through process changes or introduction of additional controls. Confidence in these estimates is strongly dependent on the knowledge and understanding of our energetic materials specialists who have unfettered access to system specific evidence and results of world-wide research, characterisation and testing of explosives materials, components and munitions. It is important that as much detail as possible is included to describe the materials and test arrangements so that the results can be correctly interpreted and read-across to safety assessments. Despite best efforts, there are inevitably gaps in evidence and incomplete understanding for some of the environments considered, particularly where a combination of types of stimuli is involved. These gaps have to be filled by expert judgement but this leads to further uncertainty. A study of the role of expert judgement in risk analysis for radiation and nuclear safety has been reported by Reiman 1, 2. Reiman emphasises the importance of careful selection of experts and aggregation of opinions to reduce the likelihood of biases in judgement and cites evidence that in some fields experts are as likely to make errors as novices (although presumably the errors may be more complex and difficult to detect!). Estimation of detonation probabilities for high explosives has been reported by investigators at Los Alamos National Laboratory to inform hazard analysis of weapon dismantlement programmes 3, 4. In this study, a range of subject matter experts was consulted about the probability of violent reaction for drops of PBX-9404 from a range of heights. The investigators found that the experts took different approaches to deducing the probabilities and reached somewhat different conclusions which presumably depended on their differing experience and ‘appetite' for risk. It is therefore imperative that experts record the reasons for their judgement to inform subsequent peer review and allow uncertainties to be taken into account through sensitivity analysis as part of the overall risk assessment. Expert judgement plays an important role in munitions safety assessment to identify and aid interpretation of relevant evidence. However, we should apply caution to undue reliance on expert judgement and seek to fill evidence gaps through research into fundamental processes and development of well-instrumented process-related testing. The reliability of expert judgement in predicting the response of munitions to threats is only as good as the evidence available and the knowledge of their peer group. In the relatively small and shrinking community of warhead explosives safety specialists there is also a danger of ‘group think' because we all collaborate and have learnt our trade from the same expert pioneers in this field. Hindsight gained from accident investigations and novel research tell us that earlier preconceptions are sometimes flawed as indicated by the following examples. A collaborative study between explosives specialists from Los Alamos National Laboratory (Cary Skidmore), and Lawrence Livermore National Laboratory (Nicole Anderson) in the United States and AWE (Paul Deacon and Keith Fleming), has reviewed past explosives incidents, in particular, three separate fatal accidents in the US & UK in 1959 which led to the tragic loss of eight process operators in total 5. Seen through the lens of perfect hindsight, analysis of these accidents indicates how far we have progressed. However, it leaves us feeling that there is still more to learn and that explosives activities remain vulnerable to human error. Although these accidents involved different routine operations (drilling, explosives transport and waste disposal) we identified some common themes including: poorly controlled manual operations involving large charges; programme pressures associated with cold-war political threats; a directive culture with operators having little involvement in decisions or safety assessment; and ‘false comfort' drawn from substantial numbers of successful previous operations and absence of ignition following process failures. Perhaps the most striking observation from the historical records was the prevailing presumption, appearing to be based on ‘conventional wisdom' rather than objective evidence, that explosives powders were more hazardous than consolidated charges; it was thus inferred that acceptable powder safety properties provided assurance of the safety of charges. Following these accidents there was an intensive programme of research to understand the mechanisms associated with ignition and growth of reaction in explosives charges and this led to the development of new test methods (notably a range of charge safety tests) and new understanding that continues to evolve today. In a second example, for many years the ‘conventional wisdom' was that Deflagration-to-Detonation Transition (DDT) does not occur in high density systems despite sporadic contra-indications; thus most research focused on combustion in low-density porous beds. Recently however, researchers have demonstrated that DDT can occur in heated, high density, high energy confined systems through a complex sequence of events in which heating causes localised damage (phase transitions and void creation) which promotes rapid combustion with the potential to cause High Explosive Violent Reaction and eventual transition to detonation 6. This has provided new insights into the underlying mechanisms and changed our perception of risk from abnormal thermal environments. In the past, the majority of explosives safety model development was focused on shock initiation environments because of code and mesh resolution limitations, rather than the perhaps more likely mechanical and thermo-mechanical environments which could be induced by typical low energy accidents. These expose explosives assemblies to highly complex, multi-dimensional and multi-time-scale stimuli which require advanced constitutive modelling, adaptive meshing and faster coupled codes. Assessment of such real-world hazards demands high resolution diagnostics which can operate over a range of time-domains to enable refinement and validation of the models. The development by John (Jack) Reaugh from LLNL of the High Explosives Response to Mechanical Stimulation (HERMES) model, in collaboration with investigators at AWE 7 and LANL 8, is just one of many examples of advances in tools and techniques which bring the exciting prospect of significant gains in insight which will be of great value to munitions safety assessment specialists. I look forward with great interest to further developments in this important field of study. As indicated above we should be cautious of the continued reliance on expert judgement to bridge gaps in direct relevant evidence. Experts base their judgements on their own experience and the knowledge of their technical community. The explosives community (at least in the UK) has shrunk considerably over the past thirty years and there are fewer experts on whom to rely. It is all the more important that prestigious peer-reviewed journals like Propellants, Pyrotechnics and Explosives continue to share reliable, high quality sources of data and understanding of the behaviour of energetic materials. Keith A. Fleming Atomic Weapons Establishment Aldermaston, United Kingdom References 1L. Reiman, Expert judgment in analysis of human and organizational behaviour at nuclear power plants, Finnish Centre for Radiation and Nuclear Safety, Department of Nuclear Safety, STUK−A118, December 1994. 2L. Reiman, Expert Judgment in Human Reliability Analysis. In: Cacciabue P. C., Papazoglou I. A. (eds) Probabilistic Safety Assessment and Management'96. Springer, London, 1996. 3S. W Eisenhawer, T. F. Bott and T. R. Bement, Detonation Probabilities of High Explosives, LA-UR-95-1754, TSA-111-95-R124, Los Alamos National Laboratory, Los Alamos, New Mexico 87545, June 1995. 4W. E. Gilmore and J. C. Casias, Expert Judgement of Likelihood Estimates for High Explosives Reactions: Findings and Lessons Learned from an Analysis of Drop Scenarios, Los Alamos National Laboratory, LA-UR-07-2556, April 2007 5C. B. Skidmore and K. A. Fleming, The Essential Role of Science in Explosives Safety, 26th International Colloquium on the Dynamics of Explosions and Reactive Systems, Boston MA, August 2017. 6G. Parker, P. Dickson, B. W. Asay, L. B. Smilowitz, B. Henson, J. McAfee, Los Alamos National Laboratory, Violent cookoff reactions in HMX-based explosives in DDT tubes: Tracking luminous waves with streak imaging, AIP Conf. Proc. 2012, 1426, 701. 7J. E. Reaugh, J. Curtis and M.-A. Maheswaran, Modelling of Deflagration to Detonation Transition in Porous PETN of Density 1.4 g/cc with HERMES, 20th Biennial Conference of the APS Topical Group on Shock Compression of Condensed Matter, Volume 62, Number 9, St Louis, Missouri, July 2017. http://meetings.aps.org/link/BAPS.2017.SHOCK.H3.1 8J. W. Tringe, K. S. Vandersall, J. E. Reaugh, H. W. Levie, B. F. Henson, L. B. Smilowitz, G. R. Parker, Observation and modeling of deflagration-to-detonation transition (DDT) in low-density HMX, AIP Conf. Proc. 2017, 1793, 060024 Volume43, Issue5May 2018Pages 433-435 ReferencesRelatedInformation