Public defence: Petar Bosnic

Petar Bosnic is defending his dissertation for the degree philosophiae doctor (PhD) at the University of South-Eastern Norway. 

The doctoral work has been carried out at the Faculty of Technology, Natural Sciences and Maritime Sciences at campus Vestfold. 

You are welcome to follow the trial lecture and the public defence.

• Link to the dissertation (to be updated) 

Summary

Achieving climate targets requires the development and deployment of scalable, flexible, and sustainable energy carriers capable of reducing emissions in industries that are difficult to decarbonize. Among these, hydrogen has emerged as a key enabler of deep decarbonization across industry, transport, and future energy systems. However, despite its potential as a clean energy carrier, hydrogen also introduces important safety challenges. If accidentally released and mixed with air, hydrogen can ignite and, under certain conditions, lead to severe explosions. Several serious accidents in Norway, including the ammonia plant explosions in Porsgrunn in 1985 and 1997 and the hydrogen refuelling station explosion in Sandvika in 2019, demonstrate why these risks must be better understood and managed.

The PhD project investigates hydrogen explosions and a specific worst-case scenario in which an initially slow explosion transitions into a supersonic explosion in a phenomenon known as deflagration-to-detonation transition (DDT). The research began by investigating this phenomenon in the explosion channel, where high-speed imaging and pressure measurements showed that explosions pass through several stages of flame acceleration before detonation can occur. A key finding was the formation of localised “hot spots”, strong local explosion events formed inside an already shock-inducing explosion. The results indicate that, for a stable detonation to develop, multiple such events must occur and interact with each other and with the channel walls. The experiments also showed the unpredictability and sensitivity of explosions transitioning into detonation, even under nearly identical conditions. Hydrogen concentration was found to have a strong influence on the outcome. 

The second part of the research focused on developing a simulation tool capable of predicting the full sequence of events that can occur following an accidental gas release and ignition. A key methodological finding was the importance of chemical kinetics in defining the critical hot-spot events identified in the experiments and the subsequent rapid energy release that ultimately leads to detonation. The development was carried out in OpenFOAM, an open-source software framework for modelling reactive fluid flow. This resulted in the development of a new explosion simulation framework called Multi-Mode-eXplosion-Foam (MMXFoam). The solver was validated against different experiments and showed that it is capable of reliably predicting explosion outcomes for different gas compositions and geometries, including fully three-dimensional explosion scenarios.

The code was designed for engineering-scale safety analysis, balancing physical accuracy with computational efficiency. Instead of attempting to resolve every small-scale detail directly, the solver focuses computational power on the key physical processes that govern how explosions develop and transition into detonations.