Flammability of gas mixtures Part 2: Influence of inert gases

Bundesanstalt für Materialforschung und -prüfung (BAM), Abt. Chemische Sicherheitstechnik, Unter den Eichen 87, D-12205 Berlin, Germany.
Journal of Hazardous Materials (Impact Factor: 4.33). 06/2005; 121(1-3):45-9. DOI: 10.1016/j.jhazmat.2005.01.033
Source: PubMed

ABSTRACT Ternary systems, which contain flammable gas, inert gas and air, were studied in order to give the user an evaluation of the ISO 10156 calculation method for the flammability of gas mixtures. While in Part 1 of this article the fire potential of flammable gases was the focal point, the influence of inert gases on the flammability of gas mixtures was studied in Part 2. The inerting capacity of an inert gas is expressed by the dimensionless K value, the so-called "coefficient of nitrogen equivalency". The experimental determination of K values is demonstrated by using explosion diagrams. The objective of this study was to compare the estimated results, given by ISO 10156, with measurements of explosion ranges based on the German standard DIN 51649-1, given by CERN and CHEMSAFE. The comparison shows that ISO 10156, Table 1, supplies conservative K values, which can be regarded as safe in all cases. Nevertheless, in a number of cases ISO underestimates the inerting capacity, so that non-flammable gas mixtures are considered flammable.

  • [Show abstract] [Hide abstract]
    ABSTRACT: Suppressions of the gasoline-air mixture explosion by non-premixed nitrogen were studied experimentally in a closed tunnel. During the process of experiments, the test tunnel was divided into three parts by two plastic films: igniting section, N2 suppression section and gasoline-air mixture section. Meanwhile, two flame intensity sensors were respectively deployed in front of and behind the N2 suppression section. Based on the analysis of the flame intensities, overpressures and concentrations of the gas components after ignition, the gasoline vapor concentration range in which the explosions can be effectively suppressed, critical length of the ignition section and critical O2 concentration in the suppression section were discussed in detail. It was indicated that values of maximum overpressure and overpressure rise rate of the explosions with non-premixed suppression were lower than that without such suppression. When the initial gasoline vapor concentration is constant, the critical length of the ignition section increases with the growth of the length of the N2 suppression section. The relationship between the critical length of the ignition section and the initial gasoline vapor concentration can be described as a negative exponential expression of y = ae−bx. There are three modes for the explosion suppression experiments: complete suppression mode, partial suppression mode and suppression failure mode. The critical O2 concentration in the N2 suppression section decreases with increase of the λ (a dimensionless parameter of ignition section length/N2 suppression section length) when the initial gasoline vapor concentration is 2.0%.
    Journal of Loss Prevention in the Process Industries 09/2014; 31:113–120. DOI:10.1016/j.jlp.2014.07.012 · 1.35 Impact Factor
  • Source
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The Domino Hazard Index used in this work was previously introduced by 19 and 20 to assess domino effect hazards caused by a unit in a given layout. This index has been used to compare several layouts in order to select the one with lower risk. A set of piecewise differentiable equations have been developed to describe all graphical descriptions given in the original proposal. Then, this work introduces a complete formulation aimed to achieve optimal facility layouts taking into account the main variables affecting this index. The result of this formulation is a mixed-integer non-linear program (MINLP), which can be solved using a GAMS code. The result can provide substantial support for decision-makers during the design stage.
    Journal of Loss Prevention in the Process Industries 07/2014; 30:219–227. DOI:10.1016/j.jlp.2013.07.007 · 1.35 Impact Factor