Aim of this project was to get a basis for judging whether an extrapolation of auto ignition temperatures, determined in small volumina, to larger volumes is possible with respect to safety considerations, to get, based on the experimental data determined during this research project, reliable information on the amount of auto ignition temperature reduction with increasing volume as well as on the enlargement of the ignition delay time with increasing volume and to offer estimation methods for both auto ignition temperature reduction and ignition delay time enlargement.
Experimental determination of the ignition temperature of selected substances in vessels of 0,15 m³ and 0,5 m³. The determination method is as far as possible identical to the standardized procedure.
Experimental determination of the ignition temperature of homogenous vapour/air mixtures of the selected substances in vessels of 0,15 m³ and 0,5 m³ with a modified method.
Checking if the Beerbower-correlation is usable without a loss in safety.
The following cases can be distinguished:
Substances showing no cool flames, or cool flames only within a narrow range below the
auto ignition temperature according a standard: The auto ignition temperature of such substances follows a logarithmic decrease with increasing volume, comparable to the above mentioned correlation known from literature. The same is valid for mixtures of such substances
Substances showing cool flames well below the auto ignition temperature according a standard: Even a small increase in the volume decreases the auto ignition temperature remarkably (> 100 K). Only in volumes larger than 1 l - 5 l (depending on substance) it follows a logarithmic correlation. However, if the "cool flame temperature" is used as reference temperature, the volume dependence of the auto ignition temperature can be described by a correlation similar to that known from literature. The same is valid for mixtures of such substances
Mixtures of substances with "cool flame temperatures" well below the auto ignition temperature according a standard and substances showing no or only a slightly lower "cool flame temperature" behave inconsistent. Depending on the amount of the respective substance in the mixture the one or the other component will predominant the influence on the autoignition temperature. Following its "cool flame temperature" or its autoignition temperature has to be used for the estimation.
The results of the investigations carried out allow to estimate the auto ignition temperature in larger volumes from that determined in standard vessels. However, a safety margin is necessary in every case. The decreasing auto ignition temperature can be accompanied by a remarkable increase in ignition delay time.