Margie

The diagram that Prof Dan Crowl indicates is much more precise and is typically used in process design, operations and research. For flammability classification of gas mixtures under the Transportation Regulations, GHS or the Fire Codes, ISO 10156 as mentioned earlier is used. People like yourself are worried that if the gas mixture cylinder should leak would it cause a flammable environment at 1 atm. It is much simpler test but not as precise. This test method has superceded the old Bureau of Mines testing that was done in the 1950's and is referenced in CGA P-23 that was used by the DOT.

The testing is conducted in an open tube at 1 atm with air flowing through it. The mixture is then flowed into the tube and ignited with an electric spark. If a flame develops and propagates the specified distance it is considered to be flammable. Concentrations above and below this are also tested to determine the lowest concentration in this diluent gas that would be flammable. ISO 10156 has tested flammable mixtures in Nitrogen by BAM a German Agency. There is a simple correction factor for other inert gases that can be used. Pure hydrogen for example has a LFL of 4% in air, when diluted with nitrogen it is not flammable until it is 5.5% in N2.

Eugene Ngai

Chemically Speaking LLC

From: DCHAS-L Discussion List [mailto:dchas-l**At_Symbol_Here**med.cornell.edu] On Behalf Of margie.brazelton**At_Symbol_Here**AM.DYNONOBEL.COM
Sent: Thursday, May 5, 2016 6:25 PM
To: DCHAS-L**At_Symbol_Here**MED.CORNELL.EDU
Subject: Re: [DCHAS-L] Hydrogen

Maybe I am misunderstanding somewhere, but wouldn't a 5% H2/95% N2 mixture be non-flammable by itself perhaps, but once it leaks into the lab and mixes with the O2 in ambient air - wouldn't this be a flammable mixture then?  I have mixes of Ar/N2/H2/CH4 that concern me, should they leak into a 'normal' atmosphere.  Also have a mix of hydrocarbons in gas form (90% CH4 +C5's, butanes, propane, ethane, etc) - am VERY concerned about leakage from that cylinder causing a flammable atmosphere if leaked into the cylinder storage area...  Currently, we have at least 30 cylinders stored in a concrete block room attached to the lab bldg with NO hazardous atmosphere monitoring in there!

I'm worried...and think I should be!

Best regards,
Margie Brazelton
Senior Process Chemist

Dyno Nobel Inc.
A business of Incitec Pivot Limited
Cheyenne Plant, P.O. Box 1287 / Cheyenne, WY 82003, 8305 Otto Road / Cheyenne, WY 82001, USA
Office: +1 307 637 2766 | Fax: +1 307 771 5637  | Mobile: +1 307 631 8368
mailto:margie.brazelton**At_Symbol_Here**am.dynonobel.com
http://www.dynonobel.com

Groundbreaking Performance Through Practical Innovation



From:        Daniel Crowl <crowl**At_Symbol_Here**MTU.EDU>
To:        DCHAS-L**At_Symbol_Here**MED.CORNELL.EDU
Date:        05/05/2016 11:45 AM
Subject:        Re: [DCHAS-L] Hydrogen
Sent by:        DCHAS-L Discussion List <dchas-l**At_Symbol_Here**med.cornell.edu>

If you have non-air mixtures then the flammable limits don't work anymore.  The flammable limits only apply to air at 1 atm and usually 25 C.  You will need a flammability triangle diagram with the 3 axis composed of hydrogen, nitrogen and oxygen.  There is a flammability zone on the diagram, that includes the upper and lower flammable limits in air.  There are rules on how to use the diagram.  How to use this diagram is provided in my textbook:  Chemical Process Safety, Fundamentals with Applications, Prentice Hall,  2011.

The textbook contains a flammability diagram for hydrogen on page 268 that I did in my testing laboratory using a 20 liter sphere.  Several hundred different mixtures.

Clearly, a mixture without oxygen is not flammable.  But, if it escapes and mixes with air it can become flammable.  The triangle diagram can be used to make this determination.

=E2=80=8BDan Crowl
Adjunct Professor, University of Utah
Professor Emeritus, Michigan Tech

On Thu, May 5, 2016 at 9:56 AM, Jeff Tenney <Jeff.Tenney**At_Symbol_Here**sdmyers.com> wrote:
I would agree if the remaining gas is Nitrogen. Matheson Gas list the lower explosive limit of Hydrogen to be 4.0%. So a 5.5% Hydrogen with 64.5% Nitrogen and the remaining is 30% Oxygen then I would consider that flammable and explosive. Unless the 64.5% is a typo and is supposed to be 94.5 then I would agree you would have a nonflammable and nonexplosive mixture.  

Jeff

From: DCHAS-L Discussion List [mailto:dchas-l**At_Symbol_Here**med.cornell.edu] On Behalf Of Don Abramowitz
Sent: Thursday, May 05, 2016 11:02 AM
To: DCHAS-L**At_Symbol_Here**MED.CORNELL.EDU
Subject: Re: [DCHAS-L] Hydrogen

Martin,

Airgas sells a 5% Hydrogen/95% Nitrogen mixture and they characterize it as a non-flammable, compressed gas.  See http://www.airgas.com/msds/002119.pdf

          Don

Donald Abramowitz
Environmental Health & Safety Officer
Bryn Mawr College
101 N. Merion Avenue
Bryn Mawr, PA 19010-2899
(610) 526-5166

In light of the most recent incidents I would like to know if anyone considers a Hydrogen(0.1-5.5%)/Nitrogen(64.5-99%) mixture flammable? . 

Martin 

From: DCHAS-L Discussion List <dchas-l**At_Symbol_Here**med.cornell.edu> on behalf of Eugene Ngai <eugene_ngai**At_Symbol_Here**COMCAST.NET>
Reply-To: DCHAS-L <DCHAS-L**At_Symbol_Here**MED.CORNELL.EDU>
Date: Wednesday, May 4, 2016 at 3:13 PM
To: "DCHAS-L**At_Symbol_Here**MED.CORNELL.EDU" <DCHAS-L**At_Symbol_Here**MED.CORNELL.EDU>
Subject: [DCHAS-L] Hydrogen

Hi!

My name is Eugene Ngai, Chemically Speaking LLC. I have spent over 40 years in the compressed gas industry and currently consult for many universities, national labs and private companies. I have done projects for some of you.

I was retained to help the UC Center of Laboratory Safety investigate the U of Hawaii incident.

They have forwarded me a number of questions regarding hydrogen. I would like to comment as follows

In the US there has been considerable work done on H2 safety, currently NFPA 55 Compressed Gases and Cryogenic Fluids Code  chapter 10 addresses gaseous systems and Chapter 11 liquid systems. We have been working closely with NFPA-2 which will be the new standard for H2 used as a fuel. Sandia National Labs is the lead research group on this. We are developing a test protocol to measure dispersion from a liquid vent stack so that we can accurately model it. Numerous meetings have been held to conduct a fault tree on siting a H2 fueling stations.

As to the Tsinghua incident, it is very difficult to get accurate information on an incident like this especially since it involves one of China's primer universities. I am told  that the cylinder ruptured. Without seeing the cylinder it would be hard for me to speculate on what caused it. I don't know if he was filling the cylinder. As you will note in my article "Dangerous Gas Mixtures Avoiding Cylinder Accidents" which is hyperlinked in the webpage. People fill cylinders all the time with gases that could compromise the cylinder. There have been many incidents involving fuel cell research.

http://www.chemicallyspeakingllc.com/

I had sent out an alert about this problem in 2011

http://www.chemicallyspeakingllc.com/home/our-company/files

A second possibility is hydrogen embrittlement. This typically occurs with high strength steels used for some cylinders. The Europeans suffered from hydrogen cylinder ruptures for many years. There is an ISO standard 11114-4 on how to test metals for H2 embrittlement. The typical US DOT 3AA and 3 A cylinders have a high enough alloy content to not become embrittled. Since it was H2 it was most likely to be filled in China. China has some very strict regulations under the GB standards on what cylinders can be used, but again not knowing the specifications or history of the cylinder I can only speculate.

Eugene