Date: Fri, 23 Sep 2011 09:55:18 -0400
Reply-To: DCHAS-L Discussion List <DCHAS-L**At_Symbol_Here**LIST.UVM.EDU>
Sender: DCHAS-L Discussion List <DCHAS-L**At_Symbol_Here**LIST.UVM.EDU>
From: Eugene Y Ngai <Eugene_Ngai**At_Symbol_Here**COMCAST.NET>
Subject: Re: 4% hydrogen mixture
Comments: To: Ken Kretchman
In-Reply-To: <4E7C4AA70200005E00095771**At_Symbol_Here**>
A similar question was asked on the SEMI Safety Grapevine. My answer with
some edits for this situation is as follows

The worldwide guidance document on gas mixture flammability is ISO 10156
"Gases and gas mixtures - Determination of fire potential and oxidizing
ability for the selection of cylinder valve outlets" 3rd edition, 2010. The
recent update to CGA P-23 was based on the more current data from ISO 10156

This is referenced in NFPA 55 Compressed Gases Flammability of Gas Mixtures. For gas mixtures other
than those containing ammonia and nonflammable gases,
flammability of gas mixtures shall be classified in accordance
with CGAP-23, Standard for Categorizing Gas Mixtures Containing
Flammable and Nonflammable Components; or by physical testing
in accordance with the requirements of ASTM E 681, Standard
Test Method for Concentration Limits of Flammability of Chemicals
(Vapors and Gases), or ISO 10156, Gases and gas mixtures -
Determination of fire potential and oxidizing ability for the selection of
cylinder valve outlets.

Under the global harmonization effort DOT, OSHA and EPA will adopt ISO10156
to classify flammable gas mixtures

The heat capacity of the gas that it is mixed with will influence the LEL,
in most cases it will increase the concentration. In 10156, the flammable
gas is tested mixed with Nitrogen, the lowest concentration that will burn
is called the Tci value while the lowest concentration of the pure gas in
air that will burn is the Li. In the case of hydrogen the Tci is 5.5% while
the Li is 4%. The Tci changes as the balance gas changes. If H2 is mixed
with N2, it will not be flammable until it reaches 5.5%.

As indicated, P-23 data is primarily based on the Bureau of Mines testing
and method which dates to the 1950's. This apparatus used a small diameter
open tube which conducts heat from a flame quicker therefore the data will
be higher than the more current ISO 10156 test which uses a larger diameter
open tube. The ASTM test is a closed sphere. There are also differences in
the definition of LFL based on the test. H2 mixed with N2 using 4 different
test methods yielded results of 5.4 - 5.7%

Hope this helps

Eugene Ngai
Chemically Speaking LLC

-----Original Message-----
From: Ken Kretchman [mailto:ken_kretchman**At_Symbol_Here**] 
Sent: Friday, September 23, 2011 9:00 AM
To: DCHAS-L Discussion List
Cc: Eugene_Ngai**At_Symbol_Here**
Subject: Re: [DCHAS-L] 4% hydrogen mixture

While I have been long operating with the understanding that this mixture is
well below a flammable concentration, based on company testing and other
data also from long ago, I am copying Eugene Ngai of Chemically Speaking who
has a great deal of experience with gas suppliers and CGA with extends over
many years...


>>> "Jeskie, Kimberly B."  9/23/2011 8:14 AM >>>
Long story short, the reason I ask this question is that we have a long
standing debate with our research community who feel strongly that they have
chosen the 4% mixture because it gives them the reducing environment they
need while limiting the hazard, that's why they use it.  The DOT
classification is based on P-32 and the building codes (which is where the
debate starts) generally site DOT as the go to place for deciding how to
class materials for the quantities allowed inside buildings. That's where we
keep coming up against a logistics nightmare. No one can agree on how to
account for this mixture and most everyone agrees that using the full volume
of a 4 % mixture just doesn't make common sense. 

One of our Fire Protection Engineers has suggested that the sources of
information we use to make these decisions may be dated and an analysis may
be the ticket to put this issue to bed. The P-23 pamphlet references Bureau
of Mines Bulletin 503-1952 and much of the 503 date is based on simple, ad
hoc tests on an apparatus dating back to the 1920s.  Some of the data is
based on earlier work dating back to the 1870s and 1880s.  DOT references
and accepts ASTM E-681 to classify flammability of gases. So...we're
debating on whether or not to test the mixture under this newer standard,
and it sure would be great if it has already been done.


Kimberly Begley Jeskie, MPH-OSHM
Operations Manager
Physical Sciences Directorate
Oak Ridge National Laboratory
(865) 574-4945

-----Original Message-----
From: DCHAS-L Discussion List [mailto:DCHAS-L**At_Symbol_Here**] On Behalf Of
Todd Perkins
Sent: Thursday, September 22, 2011 7:05 PM
To: DCHAS-L**At_Symbol_Here**LIST.UVM.EDU 
Subject: Re: [DCHAS-L] 4% hydrogen mixture

Do you mean test the mixture for flammability per CGA publication P-23?
 My understanding is that the data is based on experimental observation
as well as calculation. I've never had reason to question the data. Have you
observed something different?

Todd Perkins
Regional Safet Director
Airgas Mid America


- Sent from my mobile phone
On Sep 22, 2011 11:41 AM, Jeskie, Kimberly B. <jeskiekb**At_Symbol_Here**ORNL.GOV>

Has anyone actually tested a 4% hydrogen/ 96% argon mixture using ASTM
E-681, as opposed to just taking the P-23 data or the Bureau of Mines
Bulletin 503-1952 at face value? Kim Kimberly Begley Jeskie,
MPH-OSHMOperations ManagerPhysical Sciences DirectorateOak Ridge National
Laboratory(865) 574-4945

Previous post   |  Top of Page   |   Next post

The content of this page reflects the personal opinion(s) of the author(s) only, not the American Chemical Society, ILPI, Safety Emporium, or any other party. Use of any information on this page is at the reader's own risk. Unauthorized reproduction of these materials is prohibited. Send questions/comments about the archive to
The maintenance and hosting of the DCHAS-L archive is provided through the generous support of Safety Emporium.