Effect of bubble generation on hot zone formation in tank fires

https://doi.org/10.1016/j.jlp.2020.104314Get rights and content

Highlights

  • Heterogeneous bubble nucleation occurs at the interface of the metal mesh and fuel.

  • Bubble generation enhances the formation of hot zone.

  • The time to hot zone formation is decreased when the metal mesh is close to the fuel surface.

Abstract

Boilover is a phenomenon that both stakeholders and fire-fighters in the petrochemical industry try to avoid. This phenomenon results in an explosion of liquid hydrocarbon materials (e.g., crude oil) due to prolonged oil tank fires. The elevated temperature provides energy to vaporize the water sub-layer, which commonly resides at bottom of storage tanks, leading to tremendous fire enlargement as fuel tends to spill over around the tank. Boilover follows the formation of an isothermal layer called ‘hot zone’, and is typically accompanied by continuing bubble appearance in the hot zone. Previous studies have suggested that bubble generation could be a driving force for boilover, as it accelerates heat mixing. However, the effect of bubble generation on the mechanism of hot zone formation has not yet been systematically studied. This work investigates the effect of bubble formation on hot zone formation by installing a metal mesh in a burning fuel container. The size of the mesh grid and the position of the mesh are varied to generate bubbles with different volumes from different depths from the fuel surface. Experimental results demonstrate that the metal mesh definitely increases the volume of bubbles, and significantly reduces the time to form a hot zone. The mesh with small grids generates more bubbles than that with large grids. Additionally, bubbles start to generate earlier when the mesh is fixed nearer the fuel surface. Experimental results provide direct evidence of the bubble effect on hot zone formation.

Introduction

Boilover may occur with fuel containing water, and heated by either a fire above or other means. Once the water temperature is increased beyond the superheated limits (Garo et al., 1994), the water would rapidly expand and push the fuel out of the fuel container, enlarging the fire. This phenomenon, called ‘boilover’, inevitably results in catastrophic disasters. The most common boilover type in the petrochemical industry is ‘hot zone’ boilover (Shaluf and Abdullah, 2011), while ‘thin layer’ boilover may occur from refined fuel such as petrol leaked onto the sea.

Hot zone boilover is more likely to occur if the hot zone descends more quickly to the water layer. The energy of the hot zone vaporizes the water to generate a phase transformation, from liquid to gas phase, and this superheating limit is approximately 120 °C for crude oil boilover (Koseki et al., 2006a).

Previous researchers focused on the observation of hot zone formation before boilover occurrence (Koseki et al., 2006a, 2006b; Kamarudin and Buang, 2016; Broeckmann and Schecker, 1995). However, the clear mechanism of hot zone formation has not been established (Drysdale, 2011) although previous research indicates that the formation of hot zones depends on the type of fuels and, in certain circumstances, both the material and diameters of the tanks (Hasegawa, 1989; Blinov and Khudyakov, 1961). Hasegawa (1989) observed that bubble generation appeared throughout the hot zone and led to vigorous convection. Kamarudin et al. (Kamarudin and Buang, 2016) also observed floating bubbles near the surface of the fuel during tests, and concluded that the evaporation of the light component enhances the convective heat transfer process. Clearly, a fuel may contain components with a wide range of boiling points, and the light component tends to boil first and the heavy ones boil subsequently. Scattered vapor bubbles are typically observed in most hot zones (Hasegawa, 1989) as the fuel is heated. Moreover, the bubble generation would perform a convection motion (Xu et al., 2015) and unify the temperature inside fuel layer. Thus, the heat convection, which is induced by the bubble formation, in the liquid fuel definitely plays an important role. Nevertheless, the previous studies (Kamarudin and Buang, 2016; Hasegawa, 1989) implied that the bubble generation is one of the critical reasons for hot zone formation. No obvious evidence has been present.

Section snippets

Bubble generation-energy required for homogeneous and heterogeneous nucleation

Previous studies have suggested that the formation of hot zones may be relevant to intense bubble nucleation process (Kamarudin and Buang, 2016; Hasegawa, 1989). Researchers need to discuss the nucleation mechanism. Typically, bubble nucleation may occur at the interface between liquids/fuels with different saturation temperatures during the time of burning (Kong et al., 2017; Labouerur et al., 2013) or between solids and liquids (Jo et al., 2014; Sarangi et al., 2015). The nucleation includes

Experimental setups and procedure

This study focused on the formation of hot zone on a basis of pool fire experiments. Fig. 1 illustrates the sketch of the small-scaled experimental setups. A round stainless-steel pan (300 mm in diameter and 100 mm in height) was used as the container to mimic the hot zone formation. In parallel, the formation of bubbles and hot zone was observed in a round transparent quartz glass pan of the same size as the stainless-steel pan as shown in Firuge 2. The quartz glass series helped observe the

Results and discussion

After the fuel was ignited, some radiation from the flame transferred to the fuel, and gradually reached the deep fuel. When a metal mesh was present, the mesh absorbed some heat, increasing the temperature of the mesh and surrounding fuel. The enhanced temperature helped generate bubbles (Shi et al., 2017). This experimental method to generate bubbles was not similar to a real tank fire, in which the bubbles were generated from water, fuel itself, fuel-water interface or fuel-tank interface.

Conclusion

This study investigated the effects of bubble generation on the formation of hot zone. The extra bubbles were generated by the installation of a metal mesh with varying sizes of the grids and locations, i.e. heights. Two series of experiments were conducted, using a quartz glass pan and a stainless-steel pan. The first series helped observe the bubble generation, including the amount of bubbles and the place to achieve heterogeneous nucleation; while the second provided information of time to

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors would like to thank the Ministry of Science and Technology of the Republic of China, Taiwan, for financially supporting this research under Contract No. MOST 108-2221-E-992-056-. Additionally, the assistance of Mr. Po-Shun Chang is appreciated for conducting experiments.

References (23)

  • W.N.I. Wan Kamarudin et al.

    Small scale boilover and visualization of hot zone

    J. Loss Prev. Process. Ind.

    (2016)
  • Cited by (0)

    View full text