Methanol flammability in depleted air 4

[c2sco]

I'm looking at explosion hazards in a formalin plant which burns methanol in oxygen-depleted air. I can't find a flammability diagram between oxygen, nitrogen and methanol. Can anyone point me to one please?
Thanks,
c2sco

 

[25362]

May be the following thread is of help:

thread798-79722.

 

[c2sco]

Thanks - it led to some interesting findings, but still not the full diagram. I'd also be interested in the same but at elevated pressures (up to say 5 barg)as well, but the atmospheric case is the most pressing.

 

[Guidoo]

This document will give you an idea of what the diagram looks like at atmospheric pressure: web.umr.edu/~dludlow/classes/che258/INERTING.doc

You know two points of the diagram: UFL and LFL in air. You can estimate a third point: the MOC point. This point is where the MOC line crosses the stochiometric line. By drawing straight lines between LFL in air and the MOC point and between UFL in air air and the MOC point you get a good idea of the complete diagram.

 

[Guidoo]

The link should be as follows:

http://web.umr.edu/~dludlow/classes/che258/INERTING.doc

 

[Montemayor]

Guido:

Please receive my profound thanks for your useful, practical, and important contribution to us all. This kind of information, as well as your succinct explanation, is of major importance to all engineers who labor in the process area with hazards at every turn that must be confronted and neutralized. Once again, as on other occasions, you've come up with practical and important information for all engineers to share in. Keep up the great work.

 

[Guidoo]

Art:

My pleasure.

 

[c2sco]

Yes, well done for your help. It's still not the full story, but we are nibbling away at it. The tables in the paper don't give LFL/O2 figures for methanol / air + inerts, but as you say the diagram doesn't vary in overal shape that much between substances. "Understanding explosions" by Daniel Crowl also has some good advice on constructing triangular diagrams (on Knovel, if you subscribe).
I've just found there's a part diagram on Aire Liquide's web site - that solves it.
Thanks to you both.
Stuart

 

[Guidoo]

c2sco,

What do you mean, the paper doesn't give LFL figures for methanol? For methyl alcohol, it shows an LFL in air of 7.3 mole% and an UFL in air of 36.0 mole%. LFL and UFL in pure oxygen are not given.

About the diagrams on Air Liquides web site (

http://www.airliquide.com/en/medias/pdf/business/industry/chemicals/flammibility.pdf),

they are given in this brochure just to show Air Liquide's abilities. They don't indicate for what pressure and temperature these diagrams were made.

When I look at the yellow envelope, it appears that LFL in air is about 25% and UFL in air is about 55%. This is not in line with above mentioned data in literature at room temperature and atmpospheric pressure. In my opinion, you cannot blindly use this diagram. Of course you can contact Air Liquide to discuss your problem. They can also do tests at the elevated pressures you are talking about.

 

[c2sco]

LFL/O2 was meant to indicate LFL in oxygen - sorry, I thought you'd realise.
I think you're mis-reading Aire Liquide's diagram. On pg 2, the only yellow envelope is showing LFL of about 4%, UFL in air about 30%, but it's an example and doesn't specify the fuel. However it does specify the temperature - their point is the composition range giving an AIT of 300°C
On their pg 3, they do specify methanol on the lower diagrams, and seem to show LFL of about 6% and UFL of about 47% in air, but you're right, they don't specify the temperature. I agree, it is an advertising leaflet - I can't see anything wrong with that. They are a competent company, and I'd trust the data insofar as it is specified. Caveat emptor, we didn't pay for it!

 

[Guidoo]

The diagram on pg 2 is not for methanol since AIT, LFL and UFL don't match literature values (e.g. AIT for methanol is about 385 °C).

The diagram on the lower left-hand corner on page 3 is for methanol, and LFL in air is indeed about 6% and UFL in air is about 45% (I must have been confused by the partial diagram...). Since the UFL is much higher than the reported value of 30%, I doubt whether this diagram is for atm pressure and room temperature.

I still recommend to construct your own diagram based on literature values for LFL in air, UFL in air and crossing of MOC line(9.9% Oxygen)with the stochiometric line. Alternative is have some tests done at the pressure/temperature that you are interested in (see for example

http://www.tno.nl/defensie_en_veili...lised_areas_of_expe/gas_explosions/index.xml)

 

[Guidoo]

U.S. Bureau of Mines Bulletin 627 (Flammability Characteristics of combustible gases and vapors, by M.G. Zabetakis) contains a curve (figure 75 on page 66) showing limits of flammability of methanol/nitrogen/air mixtures at 25°C and atmospheric pressure. So it shows what happens to LFL in air and UFL in air when nitrogen is added. This you can convert to the part of flammability triangle on the right hand side of the air line.

 

[owg]

c2sco - You asked about flammability limits at elevated pressure. Bodurtha in "Industrial Explosion Prevention and Protection" says "The effect of higher pressures on L (lower flammability limit) and correspondingly on MOC is slight. A table illustrates the point for ethane in air. Pressure is in MPa gauge.
Pressure 0, L=2.85 % v/v, MOC=11
0.69, 2.8,-
3.45, 2.55, 8.9
6.2, -, 8.8
6.9, 2.2,-
The original reference is Lewis and von Elbe, Combustion, Flames, and Explosions of Gases, 2d ed. Academic, New York, 1961.

HAZOP at

http://www.curryhydrocarbons.ca

 

[c2sco]

Many thanks!
Stuart

 

[Guidoo]

Effect of elevated initial pressure on LFL is small, effect of elevated pressure on UFL is not so small. For example, for natural gas (85-95% methane and 15-5% ethane), following relations were found:

LFL(vol%) = 4.9-0.71Log P (atm)
UFL(vol%)=14.1+20.4Log P (atm)

(Source: US Bureau of Mines bulletin 627, equations 41 and 42 on page 26)

So LFL changes from 4.9 to 4.4 % and UFL changes from 14,1 to 28.4 % when pressure is increased from 1 to 5 atm.

I don't have such a relationship for methanol, but expect that qualitatively effect will be similar as for natural gas.

 

[c2sco]

Thanks, Guidoo.

Bearing in mind that the qualitative explanation for the existance of an LFL is that below this, there is insufficient heat of reaction to sustain the flame, it seems logical that the LFL doesn't change much with pressure.

It's a bit harder though to rationalise this for the UFL trend. I've heard of this relationship in principle before, but hadn't thought much about it. Usually the converse argument is applied - above the UFL there's too little oxygen and hence insufficient heat of reaction to sustain the flame again - but if so, why should less oxygen be tolerable at higher pressure?

I managed to find the info on methanol from the US Bureau of Mines - from another customer who operates a methyl chloride process (Methanol + HCl). Small world!

 

[Guidoo]

c2sco,

See

http://www.airproducts.com/metals_newsletter/TempInfluenceFlammability.pdf

According to this article, influence of pressure on flammability limits cannot be described by CAFT (Calculated Adiabatic Flame Temperature) theory.