Sécurité des procédés. Emballement de réaction. Dimensionnement des évents de sécurité pour systèmes gassy ou hybrides non tempérés : outil, expériences et modèle

Abstract : Safety vents allow avoiding explosion of chemical reactors in case of thermal runaway reaction. Vent sizing methods designed by the DIERS often lead to large oversizing for non tempered systems. An alternative method, based on the principle of similarity, was developed within UN for peroxide transportation. It provides more realistic vent areas but is very constraining. This PhD work allowed us to build a new similarity vent sizing tool: the 0,1 litre scale model. This new tool is based on the VSP2 adiabatic calorimeter. It allows for direct determination of the necessary A/V ratio at laboratory scale. It also allows real time measurement of vented mass during the relief.
We validated the 0,1 litre scale model (1 x 10-3 m-1 < A/V < 3,5 x 10-3 m-1) by comparing with similar tests carried out by INERIS with UN 10 litre reactor. These blowdowns were conducted with cumene hydroperoxide (30% w/w) in 2,2,4-triméthyl-1,3-pentanediol diiso-butyrate. They showed that the 0,1 litre scale model leads to vents a bit larger (0 to 50%) than the UN 10 l reactor. It is thus conservative, but much less oversizing than DIERS method for gassy systems, while at the lab scale. The main limit is related to thermal leaks. It has to be verified that their influence is negligible for all studied systems.
Decomposition of the studied systems looks like that for a non tempered system (2 pressure peaks). However some products of the decompositions are vapour. These latter have a large influence on 2nd pressure peak. They slow the reaction and decrease the maximum reached temperature. One can even observe a correlation Pmax = f(Tmax). This behaviour could concern most decompositions (all ones?).
Vented mass measurements allowed us to distinguish three types of behaviour which illustrate the influence of pressure in the vessel on level swell. Comparison with a purely gassy dynamic model showed that the vented mass can be either two-phase venting or an alternation of gas and two-phase venting. It also showed that purely gas venting occurs at turnaround for high pressures and that two phase venting during pressure decrease after the second pressure peak can only be explained by presence of vapour (boiling phenomenon).
Finally, we identified the oversizing assumptions of the DIERS method for gassy systems. It appeared that, the main cause of oversizing is assuming that the turnaround is controlled by volume flow equality (volume vented = generated gas volume).
Complete list of metadatas

https://tel.archives-ouvertes.fr/tel-00136503
Contributor : Andrée-Aimée Toucas <>
Submitted on : Wednesday, March 14, 2007 - 11:28:04 AM
Last modification on : Tuesday, October 23, 2018 - 2:36:05 PM
Long-term archiving on : Wednesday, April 7, 2010 - 1:34:21 AM

Identifiers

  • HAL Id : tel-00136503, version 1

Citation

Luc Véchot. Sécurité des procédés. Emballement de réaction. Dimensionnement des évents de sécurité pour systèmes gassy ou hybrides non tempérés : outil, expériences et modèle. Génie des procédés. Ecole Nationale Supérieure des Mines de Saint-Etienne, 2006. Français. ⟨tel-00136503⟩

Share

Metrics

Record views

574

Files downloads

7863