Trace components and autoignition temperature

[Jobst]

Hi All,

Can anyone point me in the right direction for dealing with trace components in a gas mixture, and whether these need to be taken into account when determining temperature code for equipment?

E.g. I have a gas where the bulk components would warrant T2 (<300 °C). However, there's <100 ppm of a component which would warrant T6 (<85 °C) if it were a pure component. There is no way for the trace component concentration to increase beyond a few hundred ppm.

Somebody is asserting that because this component is in the gas, we must specify T6 for all equipment in the area. This seems a little over the top to me. If this was a flammability range determination, I'd do a calculation of the mixture flammability range, rather than using pure components, and use the mixture limits for design (with a healthy safety factor). Without any discussion of how to deal with mixtures in the standard, it seems that this person favours following the standard to the point of absurdity, and asserts that even 1 ppb of this substance would warrant a T6 rating, even if the remaining components were completely inert.

Any help appreciated!

 

[zdas04]

In similar [stupid] conversations, I've have done a mass wieghted average autoignition calc.

(0.999995 times MW₁ times Auto₁ + 0.000005 time MW₂ times Auto₂)/MW_mixture always equals Auto₂

The only exception that I've seen to this is with liquids that can stratify and the trace can accumulate. With gases this doesn't happen.

David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[georgeverghese]

Take a look at table 26-10 in Perry Chem Engg Handbook 7th edn for autoignition temps - mostly T >200degC, so what is this component with such a low autoignition temp? Also have a read of this section for aother guidance on how to use these values for setting T ratings.

 

[Jobst]

CS₂ is on that list at 90°C. Unfortunately Perry's doesn't have any discussion on treating mixtures of gases with a single combined autoignition temperature.

You're right though, in that most of the gases in that table are in a similar range of temperatures, and if I had a mixture of a few of them, I'd pick the lowest and call it a day. In my situation though, there is an outlier in terms of ignition temperature, that also only happens to be present at ppm levels. I'm arguing that we should neglect this component, or do a weighted average as per David's example (effectively neglecting it), and thereby not have to specify every piece of equipment in the plant at T6.

The situation reminds me of a similar one trying to explain distillation to some people:
-"the mixture boils at X temperature"
-"but the boiling point of Y component is below X temperature, it'll all be gone by then!"
-patiently explain once again that mixtures are different to pure components

 

[georgeverghese]

There is an interesting equation in Wikipedia (search for autoignition) that gives the time for ignition for a gas based on inputs for auto ignition temp, and some other factors. If you were inclined, you could work out what some of these time and heat flux values are for some pure component based on its known AIT. Then we could work out AIT for this gas mix with trace CS2 in it.

With a lot less effort, by examination of this equation, and knowing how mixing rules for these physical props are configured, we can see that thermal conductivity, density and specific heat for this NG mix would hardly register a change with 100ppmv CS2 in it. Hence we could safely say that AIT for this gas is >400degC, by looking at table 26-10.