Skip to Main content Skip to Navigation
Theses

Emballements thermiques de réactions : Etude des méthodes de dimensionnement des évents de sécurité applicables aux systèmes hybrides non tempérés.

Abstract : DIERS developed simplified emergency vent sizing methods to protect vessels from overpressures. When applied to untempered systems, DIERS methodology can be overly conservative. Some similarity tools (like UN 10 L reactor) lead to more realistic vent sizes. They are however very constraining. A former study led to building a new similarity vent sizing tool at laboratory scale: the 0.1 L scale model. It was partially validated by a comparative study between the new tool and UN 10 L reactor, of the thermal decomposition of cumene hydro peroxide (CHP) 30 % w/w in 2,2,4-triméthyl-1,3-pentanediol diisobutyrate (butyrate). The 0.1 L scale model then allowed a better understanding of the blowdown course and assessment of vent sizes from DIERS methodology for untempered systems.

This work was aimed at widening that understanding and at a better identification of the origin of DIERS method being so much oversizing. The method was improving the scale model, testing new chemical systems and especially changing the vapour contents of these chemical systems.

We added a real time measurement of vented gas volume to the scale model. A study of thermal leaks allowed verifying that the scale model can be used for simulating not only fire scenarios but also adiabatic ones.

We then looked for solutions as near as possible from the pure gassy case (vapour influence as low as possible): dycumyl peroxide (DCP) and tert-butylperoxy-2-ethylhexanoate (tBPEH) in butyrate. Study of the decomposition of the same peroxides in a more volatile solvent (xylene) then allowed measuring the sensitivity of the blowdown and the DIERS method to vaporisation. Studying these systems in both closed and open test cells (adiabatic calorimetry) incidentally showed that these two methods lead to very different assessments for gas flow rate.

Studying blowdown course allowed confirmation of forecast qualitative trends: the more vaporisation exits, the more kinetics are sensitive to vent size. A more surprising observation is that a temperature stabilisation due to ebullition is always observed after the second pressure peak, even for the most gassy system (DCP in butyrate).

For nearly pure gassy system, we concluded that the main origin of DIERS method being oversizing is the assumption of a homogenous flow regime inside the reactor (level-swell) and thus two-phase flow through the safety vent whereas real flow is gaseous. A less important source is the type of calorimetric test used for sizing (closed or open test). For untempered systems sensitive to vaporisation, oversizing is moreover due to the vaporisation effect, which is not taken into account in DIERS methods.
Document type :
Theses
Complete list of metadatas

https://tel.archives-ouvertes.fr/tel-00372536
Contributor : Andrée-Aimée Toucas <>
Submitted on : Wednesday, April 1, 2009 - 2:43:27 PM
Last modification on : Wednesday, June 24, 2020 - 4:18:20 PM
Long-term archiving on: : Thursday, June 10, 2010 - 7:30:19 PM

Identifiers

  • HAL Id : tel-00372536, version 1

Citation

Wilfried Minko. Emballements thermiques de réactions : Etude des méthodes de dimensionnement des évents de sécurité applicables aux systèmes hybrides non tempérés.. Sciences de l'ingénieur [physics]. Ecole Nationale Supérieure des Mines de Saint-Etienne, 2008. Français. ⟨tel-00372536⟩

Share

Metrics

Record views

594

Files downloads

2751