From: Eugene Ngai <eugene_ngai**At_Symbol_Here**COMCAST.NET>
Subject: Re: [DCHAS-L] Pyrophoric gases
Date: Wed, 19 May 2021 19:13:08 -0400
Reply-To: ACS Division of Chemical Health and Safety <DCHAS-L**At_Symbol_Here**Princeton.EDU>
Message-ID: 047f01d74d04$89a88690$9cf993b0$**At_Symbol_Here**

The key safety rule worldwide for all flammable gases is never to put out the fire. If safe to do so shutoff the flow. Putting out the fire creates an even more dangerous situation. The flammable gas is continuing to leak and mix with air. It could suddenly find an ignition source and an explosion occurs. With a pyrophoric gas it is even more dangerous since it can suddenly ignite by itself. I investigated a diborane leak 20 years ago. Since it’s lighter than air it went to the ceiling. With the moisture in the air it hydrolyzed to form hydrogen. It suddenly ignited and blew out the roof. Windows were shattered 100 ft away.

I have made and tested silane for almost 50 years. I teach silane Safety and ER. Silane can create a different problem. As it burns it creates SiO2 that will choke off the gas flow. Suddenly the plug will fall off and the flow will start up again only to be choked off again. This process repeats it self many times. A normal gas cylinder would empty itself in hours while silane could take days. With mixtures silane has a blowout velocity limit. The mixture will ignite at the beginning as the flow increases the flame suddenly goes out. Now it becomes a dangerous situation like the diborane incident above. When designing systems for pyrophoric or highly toxic gases, a shutoff valve must be as close to the cylinder as possible under the Fire Code.

Diborane, Disilane and Silane are extremely strong reducing gases. They will react with CO2 or any of the Halocarbon extinguishers


I did extensive testing with trimethylaluminum in 2004 and even tried to burn down a gas cabinet with it. Metal Organics like TMA, TMI, TMG are also violently water reactive (explosive). Extinguishing with CO2 or Halocarbon they would reignite once there is no more gas. The only material that would work are dry powder, vermiculite or ane. All have to be extremely dry otherwise there is an explosive reaction. The rule of thumb is to use 4-5 times the weight of the pyrophoric to smother it. All will reignite when they are disturbed.


Eugene Ngai

Chemically Speaking LLC




From: ACS Division of Chemical Health and Safety <DCHAS-L**At_Symbol_Here**Princeton.EDU> On Behalf Of David C. Finster
Sent: Wednesday, May 19, 2021 4:16 PM
To: DCHAS-L**At_Symbol_Here**Princeton.EDU
Subject: [DCHAS-L] Pyrophoric gases


I am working on a document about pyrophorics and how to extinguish fires involving them.


I have read that:  for pyrophoric solids (and for active metal fires involving water – which “behave” like pyrophorics) one can use a Type D extinguisher.  This makes sense to me since the Type D material will largely exclude air and probably cool the solid (a bit).  For liquids, a Type ABC extinguisher is recommended, and this makes sense to me since the ABC agent will cover the liquid (excluding air) and reduce vaporization.


For pyrophoric gases, the situation is less clear.  The 8th edition of the GHS Purple Book does not have pyrophoric gases as a hazard category.  (There are pyrophoric solids and pyrophoric liquids/solutions.)   The hazard statements for a “pyrophoric gas” will include H220 (extremely flammable gas) and H250 (catches fire spontaneously if exposed to air).


I have poked around a bit using phosphine as the test case.


The Linde SDS for phosphine (phosphine-ph3-safety-data-sheet-sds-p4643.pdf ( revised in 8/31/18 says:  “Suitable extinguishing media : Carbon dioxide, Dry chemical, Water spray or fog”.  This is shocking to me since none of these materials would adequately eliminate a leg of the fire triangle (or tetrahedron).  On a lucky day, I can imagine that CO2 could momentarily displace enough air to extinguish a small volume of the burning gas but assuming that the gas is leaking from some system, it would reappear and catch fire. The SDS continues, (colored fonts added by me):


DANGER! Toxic, flammable, corrosive, liquid and gas under pressure.   If venting or leaking gas catches fire, do not extinguish flames. Flammable vapors may spread from leak, creating an explosive reignition hazard. Vapors can be ignited by pilot lights, other flames, smoking, sparks, heaters, electrical equipment, static discharge, or other ignition sources at locations distant from product handling point. Explosive atmospheres may linger. Before entering an area, especially a confined area, check the atmosphere with an appropriate device.


The phrase in red seems appropriate to me.  Important, too, since the natural intuition for chemists and emergency responders is to put out the fire


The section in green seems entirely inappropriate since no “ignition source” is needed for a pyrophoric substance.


As a former firefighter the phrase in blue is not helpful since only 0% oxygen would be safe, and this is not realistic at any fire scene. 


FYI, the 2021 DOT Emergency Response Guide lists phosphine with a Guide #119.  It says:


Small Fire

• Dry chemical, CO2, water spray or alcohol-resistant foam.

Large Fire

• Water spray, fog or alcohol-resistant foam.

• FOR CHLOROSILANES, DO NOT USE WATER; use AFFF alcohol-resistant medium-expansion foam.

• If it can be done safely, move undamaged containers away from the area around the fire.

• Damaged cylinders should be handled only by specialists




I found two useful documents from the University of Nebraska-Lincoln:

Safe Operating Procedure (

For (pyrophoric) gases, close the valve at the cylinder if it is safe to do so. If not, evacuate the area. Most burning pyrophoric gases are difficult to extinguish. Silane is reported to be impossible to extinguish.



DO NOT use carbon dioxide or water fire extinguishers as these types of extinguishers can actually enhance the combustion of some pyrophoric compounds.

(Phosphine might not be applicable for this statement.)


It seems that the best remedy for a burning gas (presumably arising from a leak in a system) is to isolate the leak or close the valve – if these steps are even possible.


Perhaps in some situations there is no way to stop the fire except to let it burn itself when the pyrophoric gas concentration is exhausted.  Best to be working in a hood, I imagine.  (Preferably an uncluttered hood not storing bottles of flammable liquids!)


Additional thoughts are welcome.





David C. Finster
Professor Emeritus, Department of Chemistry
Wittenberg University


--- For more information about the DCHAS-L e-mail list, contact the Divisional membership chair at membership**At_Symbol_Here** Follow us on Twitter **At_Symbol_Here**acsdchas

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.