Skip to main content

Full text of "DTIC AD1044406: Construction of an Exploratory List of Chemicals to Initiate the Search for Halon Alternatives"

See other formats


ESL-TR-90-26 



CONSTRUCTION OF AN EXPLORATORY 
LIST OF CHEMICALS TO INITIATE THE 
SEARCH FOR HALON ALTERNATIVES 


A4W.NI. PITTS, M.R. NYDEN, R.G. GANN, W.G. MALLARD, 


TSANG 


NATIONAL INSTITUTE OF STANDARDS AND 

TECHNOLOGY 

GAITHERSBURG MD 20899 


JUNE 1991 
FINAL REPORT 

OCTOBER 1989-SEPTEMBER 1990 


APPROVED FOR PUBLIC RELEASE. DISTRIBUTION 
UNLIMITED. 



AIR FORCE ENGINEERING & SERVICES CENTER 
ENGINEERING & SERVICES LABORATORY 
TYNDALL AIR FORCE BASE, FLORIDA 32403 














































NOTICE 


Please do not request copies of this report from 
HQ AFESC/RD (Engineering and Services Laboratory), 
Additional copies may be purchased from: 

National Technical Information Service- 
5285 Port Royal Road 
Springfield, Virginia 22161 

Federal Government Agencies and their contractors 
registered with Defense Technical Information Center 
should direct requests for copies of this REPORT TO: 


Defense Technical Information Center 
Cameron Station 
Alexandria, Virginia 22314 



UNCLASSIFIED _ 

SECURITY CLASSIFICATION OF THIS PAGE 


la. REPORT SECURITY CLASSIFICATION 
UNCLASSIFIED 


2a SECURITY CLASSIFICATION AUTHORITY 


2b. DECLASSIFICATION/DOWNGRADING SCHEDULE 


4. PERFORMING ORGANIZATION REPORT NUMBER(S) 
NIST TN 1279 


REPORT DOCUMENTATION PAGE 


lb RESTRICTIVE MARKINGS 


Form Approved 
OMB No 0704 OIBfi 


3 DISTRIBUTION/AVAILABILITY OF REPORT 

Approved fo.r public release 
Distribution unlimited 


5. MONITORING ORGANIZATION REPORT NUMBER(S) 
ESL-TR-90-26 


6a. NAME OF PERFORMING ORGANIZATION 6b OFFICE SYMBOL 7a. NAME OF MONITORING ORGANIZATION 
National Institute of Standards (If applicable) 
and Technology Engineering and Services Laboratory 


6c. ADDRESS (City, State, and ZIP Code) 


7b ADDRESS (City, State, and ZIP Code) 

AIR FORCE Engineering and Services Center 
Tyndall AFB FL 32403 


8b OFFICE SYMBOL 9 PROCUREMENT INSTRUMENT IDENTIFICATION NUMBER 
(If applicable) 

Contract No. 89CS8205 


10 SOURCE OF FUNDING NUMBERS 


PROGRAM 


Gaithersburg MD 20899 


8a. NAME OF FUNDING / SPONSORING 
ORGANIZATION 


8c. ADDRESS (City, State, and ZIP Code) 


11. TITLE (Include Security Classification) 

Construction of an Exploratory List of Chemicals to Initiate the Search for Halon Alternatives 


12. PERSONAL AUTHOR(S) 

W.M. Pitts, M.R. Nyden, R.G. Gann, W.G. Mallard, and W. Tsang 


13a. TYPE OF REPORT 13b. TIME COVERED 14. DATE OF REPORT (Year, Month, Day) 15. PAGE COUNT 

Final Report from 10/89 TO 9/90 June 1991 _ 


16. SUPPLEMENTARY NOTATION 


PROJECT 

TASK 

WORK UNIT 

NO. 

NO 

ACCESSION NO 


Also published as NIST Technical Note 1278. 


COS ATI CODES 


SUB-GROUP 



18 SUBJECT TERMS ( Continue on reverse if necessary and identify by block number) 
Halons, Fire Suppression, Fire Safety, Halon Alternatives 


19 ABSTRACT ( Continue on reverse if necessary and identify by block number) 

Production of the currently-used halogenated fire suppressants (halons) will be curtailed 
because of their contribution to stratospheric ozone depletion. This report, one of the 
first efforts toward identifying alternatives, documents the rationale for and selection 
of a set of approximately one hundred gases and/or liquids, covering a range of chemical 
and physical principles thought to affect flame suppression capability and stratospheric 
ozone depletion. An Appendix provides extensive information on each of the selected 
chemicals. Also included in the report is an introduction to combustion concepts, fire 
suppression mechanisms, test approaches for flame suppression effectiveness, and the 
mechanisms by which the current commerical halons can decrease stratospheric ozone. 


20 DISTRIBUTION / AVAILABILITY OF ABSTRACT 21. ABSTRACT SECURITY CLASSIFICATION 

□ UNCLASSIFIED/UNLIMITED n SAME AS RPT □ DTIC USERS UNCLASSIFIED 


22a NAME OF RESPONSIBLE INDIVIDUAL 22b TELEPHONE (Include Area Code) 22c OFFICE SYMBOL 

John R. Floden, Capt, USAF 


DD Form 1473, JUN 86 Previous editions are obsolete SECURiTY CLASSIFICAT ON OF THIS PAGE 

i UNCLASSIFIED 

(The reverse of this page is blank.) 







































EXECUTIVE SUMMARY 


Production of the currently-used halogenated fire suppressants (halons) will be curtailed because of 
their contribution to stratospheric ozone depletion. This report, one of the first efforts toward identifying 
alternatives, documents the rationale for and selection of a set of approximately one hundred gases and/or 
liquids, covering a range of chemical and physical properties thought to affect flame suppression capability and 
stratospheric ozone depletion. An Appendix provides extensive information on each of the selected chemicals. 
Also included in the report are an introduction to combustion concepts, fire suppression mechanisms, test 
approaches for flame suppression effectiveness, and the mechanisms by which the current commercial halons 
can decrease stratospheric ozone. 


iii 

(The reverse of this page is blank.) 




PROJECT SUMMARY 


The objective of this project is to initiate a systematic search for optimal halon replacements by 
identifying approximately 100 gases and/or liquids, covering a range of chemical and physical principles thought 
to affect flame suppression capability. This comprehensive report is designed to provide a basis for the search 
for alternatives to the current commercial halons. As much detail as possible is provided for each of the 103 
chemicals to aid researchers who might test or evaluate these chemicals in the future. 

The current commercial halons have been identified as contributing to the depletion of the earth’s 
stratospheric ozone. As a result, their manufacture has been limited to 1986 levels by the Montreal protocol, 
and more severe limitations are anticipated in the near-future. This project is an early step in a governme¬ 
nt/industry program to identify and qualify candidate replacements for halons 1301 and 1211 that will satisfy 
the needs of the major users for existing applications. No such search has been conducted since the late 1940s, 
when the U.S. Army conducted the study that led to today’s predominant halogenated fire suppressants: halons 
1301 (CF 3 Br) and 1211 (CF 2 ClBr). Halon 2402 (GjT^B^) is in use to a lesser degree, as are halons 1001 and 
1011. 


Replacements for the current halons must have a number of critical properties: fire suppression 
efficiency, low residue level, low electrical conductivity, low metals corrosion, high materials compatibility, 
stability under long-term storage, low toxicity (inhalation and contact) of the chemical and its combustion 
products, and low (or zero) contributions to stratospheric ozone depletion and global warming. These 
constraints are expected to complicate the search. 

The search strategy selected by the National Institute of Standards and Technology (NIST) team 
focusses on principles for efficient fire suppression and low contribution to stratospheric ozone depletion. The 
widest range of chemical families were researched for their potential to test these principles. Some of these 
families have not previously been considered for fire suppression. Compounds are discussed which are clearly 
not candidates as alternatives, but which are included to test principles of fire suppression or ozone depletion. 


As a basis for deriving these principles, this report begins with a compilation of state-of-the-art 
knowledge in flame suppression and ozone depletion chemistry. The means by which a chemical can quench 
a flame fall under the following headings: 

■ Physical mechanisms 

■ smothering or blanketing 

■ cooling and dilution 

■ mechanical means (e.g., blowing out a match) 

■ flame radiation blockage 

■ Chemical action (interference with the chain reactions that propagate flames) 

All agents have physical effects on a fire. Agents which also have a chemical mechanism, such as the 
currently-used halons, are more effective due to the additional pathway for flame suppression. There have 
been several studies of the effectiveness of specific chemicals. Very few of those separate the contribution of 
physical and chemical mechanisms. 

The means for altering a chemical to decrease its contribution to ozone depletion also fall under a 
few headings: 


v 


■ Elimination of all bromine, chlorine, and iodine atoms; 

■ Increased reactivity in the lower atmosphere by 

■ enhanced reaction with OH radicals 

■ dissociation due to absorption of solar radiation 

■ enhanced rain out due to increased polarity 

Assessment of ozone-depletion effectiveness is based on atmospheric modeling. The only experimental 
work is the determination of possible reaction paths and their rates. 

An extensive search was performed of the literature on combustion suppression, flame inhibition, and 
fire retardancy. This led to the identification of nine families of chemicals with potential for workbench testing 
of the above principles. 


compounds, and analogs to the HCFCs. 


■ Halogenated ketones, anhydrides, and esters. These contain C=0 bonding and may be more 
prone to solar dissociation in the troposphere. 


■ Unsaturated halocarbons. These contain C=C bonding and may be more reactive with OH 
in the troposphere. 


■ Halogenated ethers. These contain an O-C-O linkage, which is thought to promote solar 
dissociation in the troposphere. 

■ Halons containing iodine. These are likely to be more reactive in the troposphere. 

■ Sulfur halides. These are analogs to the halons. 


■ Compounds containing phosphorus. These are non-ozone-depleting, possibly highly efficient 
fire suppressants. 


■ Silicon and germanium compounds. These are tropospherically-reactive analogs to the 
halons. 


■ Metallic compounds. These are extremely efficient fire inhibitors that are non-ozone-deplet- 
ing. 

■ Inert gases. These are baseline chemicals: less-efficient, only physically active agents with zero 
ozone-depletion potential. 


vi 



A list of 103 chemicals which were selected from these families follows: 


perfluoromethane 

perfluoroethane 

perfluoropropane 

perfluoro-n-butane 

perfluorocyclobutane 

trifluoromethane 

pentafluoroethane 

1.1.1.2- tetrafluoroethane 
dibromodifluoromethane 

2.2- dibromo-l ,1,1,2-tetrafluoroethanc 
chlorodifluoromethane 

1.1.1 -trichlorethane 

2.2- dichloro-l ,1,1 -trifluoroethane 
2-chloro-l,l,l,2-tetrafluoroethane 

1.1 -dichloro-1 -fluoroethane 

1- chloro-1,1-difluoroethane 
bromodifluorometliane 
bromoch lorofluoromet ha ne 

2- bromo-2-chIorol,1,1-trifluoroethane 
2-bromo-l -chloro-1,2,2-trifluoroethane 

1- bromo-l,l,2,2-tetrafluoroethane 

2- bromo-1,1,1 -trifluoroetha ne 

1.2- dibromo-l,l ,2-trifluoroethane 

1.2- dibromo-1,1-difluoroethane 

1.2- dibromo-l ,2-difluoroethane 

1 -bromo-1,1,2,3,3,3-hexafluoropropane 

1.3- dibromo-l,l,3,3-tetrafluoro propane 

2.2- dibromo-l ,1,3,3-tetrafluoropropane 
l-bromo-l,l,3,3,3-pentafluoropropane 
hexafluoracetone 

trifluoroacetic anhydride 
bis(perfluoroisopropyl) ketone 
methyltrifluoroacetate 

3- bromo-l,l,l-trifluoropropanone 
bromopen ta fl uoroace tone 
bromomethyltrifluoroacetate 
perfluoropropene 
perfluorobutene-2 
perfluorotoluene 

1.1.3.3.3- pentafluoropropene-l 
3,3,3 trifluoropropene 

1.2- bis( p>e rfluoro-n -bu tyl)ethy lene 
3 -bromop>erfluoropropene 

1 -bromoperfluoropropene 

1.2- bis(perfluoromethyl)ethylene 
l-bromoperfluoromethyl-2-perfluoromethylethylene 

1- bromo-bis(perfluoromethyl)ethylene 
tetris(perfluoromethyl)ethylene 
tetrafluorodimethyl ether 
pentafluorodimethyl ether 

2- chloro-l -(difluoromethoxy)-l ,1,2-trifluoroethane 
isoflurane 


perfluoro-2-butyltetrahydrofuran 

bis(bromodifluoroethyl) ether 

l-bromo-l,l,3,3,3-pentafluorodimethyl ether 

bromoenflurane 

octafluorofuran 

3-bromoperfluorofuran 

bis(perfluoromethyi) thioether 

tris(perfluoromethyl) amine 

iodotrifluoromethane 

chlorodifluoroiodomethanc 

1 -bromo-1,1,2,2-tetrafluoro-2-iodoethane 

1,1,2,2-tetrafl uoro-1,2-diiodoe thane 

iodomethane 

iodoethane 

1-iodopropane 

1,1,1,2,2,3,3-hep ta fluoro-3-iodopropane 

sulfur fluoride 

sulfur chloride fluoride 

sulfur bromide fluoride 

phosphorous trifluoride 

phosphorous trichloride 

phosphorous bromide difluoride 

phosphoryl fluoride 

phosphoryl chloride 

phosphoryl bromide fluoride 

tetrachlorosilane 

trichlorofluorosilane 

tetrafluorosilane 

bromotrifluorosilane 

tribromofluorosilane 

tetramethylsilane 

chlorotrimethylsilane 

trichloromethylsilane 

chloromethyltrimethylsilane 

tetrachlorogermane 

tetramethylgermane 

sodium hydrogen carbonate 

sodium acetate 

potassium hydrogen carbonate 

potassium oxalate 

potassium acetate 

potassium acetylacetonate 

chromium acetylacetonate 

chromyl chloride 

tin (IV) chloride 

titanium (IV) chloride 

tetraethyl lead 

iron pentacarbonyl 

nitrogen 

carbon dioxide 

argon 


vii 



While it is true that some of the chemicals which appear on this list can be considered candidates 
themselves, others were chosen to test principles of fire suppression. The appendix contains data sheets for 
each of the selected compounds that include formulae, common names, classification numbers, physical proper¬ 
ties, commercial sources and prices (if available), toxicity information, references to fire suppression results, 
and additional relevant comments. 

These compounds should be tested in a selective series of experiments based on the insights used in 
the development of the list. The study should determine the fire suppression effectiveness of these agents, and, 
if possible, characterize their mechanisms of chemical flame inhibition. Such insights will provide the 
knowledge base required for the intelligent design of alternative chemical fire suppressants with a full range 
of desirable properties. 


viii 



PREFACE 


This report was prepared by the Center for Fire Research, National Institute of Standards and 
Technology, Gaithersburg, Maryland 20899, under Contract Number 89CS8205, for the Air Force Engineering 
and Services Center, Engineering and Services Laboratory, Tyndall Air Force Base, Florida 32403-6001. This 
summarizes work done between October 1989 and May 1990. Capt. John M. Floden was the contract monitor. 

This report has been reviewed by the Public Affairs Office and is releasable to the National Technical 
Information Service (NTIS). At NTIS, it will be available to the general public, including foreign nations. 

The authors appreciate the review provided by the Halon Alternatives Research Corporation, chaired 
by Ms. Denise Mauzerall, and preparation of the final copy of this report by Ms. Paula Garrett 

This technical report has been reviewed and is approved for publication. 



Chief, Air Base Fire Protection Director,/Engineering and Services 

and Crash Rescue Systems Branch Laboratory 


ix 

(The reverse of this page is blank.) 





TABLE OF CONTENTS 


Section Title Page 

I INTRODUCTION . 1 

A. OBJECTIVE . 1 

B. BACKGROUND . 1 

1. Halogenated Hydrocarbon Fire Suppressants (halons). 1 

a. Definition and Designation . 1 

b. A Brief History. 1 

c. Current Use of Halogenated Fire Suppression Agents . 4 

2. The Search for Halon Alternatives. 5 

C. SCOPE/APPROACH . 5 

1. Nature of the Problem. 5 

2. Structure of the Research Program. 6 

3. Search Strategy. 6 

4. Report Philosophy. 8 

a. Breadth of Coverage . 8 

b. Report Structure. 8 

5. Relation of This Work to Other Efforts . 9 

II CURRENT UNDERSTANDING. 10 

A. COMBUSTION AND FIRE CONCEPTS. 10 

1. Definitions of Combustion and Fire. 10 

2. The Fire Tfiangle. 10 

3. Combustion and Fire Classifications. 10 

a. Flaming Combustion and Smoldering. 10 

b. Fuel Types. 11 

c. Premixed and Diffusion Flames . 11 

d. Laminar and Thrbulent Combustion. 12 

e. Lean, Stoichiometric, and Rich Combustion. 12 

4. Basic Combustion Concepts. 12 

a. Thermodynamics . 12 

xi 

































TABLE OF CONTENTS 
(CONTINUED) 


Section Title Page 

b. Chemical Kinetics . 13 

c. Combustion Chemistry. 14 

d. Thermal balance . 17 

e. Propagation Rates of Premixed Flames. 17 

f. Flame Structure. 18 

B. INTRODUCTION TO FIRE SUPPRESSION. 18 

1. Relevant Reviews. 19 

2. Fire Suppression Mechanisms. 19 

a. Physical Fire Suppression Mechanisms. 19 

b. Chemical Fire Suppression Mechanism . 20 

c. Chemical Fire Suppression and Physical Cooling . 20 

3. Characterizations of An Agent’s Flame Suppression Capabilities. 21 

a. Flame Chemistry and Flame Structure of Premixed Flames. 21 

b. Agent Effects on Flame Speed of Premixed Flames . 21 

c. Agent Effects on Premixed Flame Flammability Limits and Peak 

Concentrations. 22 

d. Full Kinetic Modeling of Premixed Flames. 23 

e. Types of Diffusion Flames Used for Inhibition Investigations. 24 

f. Flame Chemistry and Structure of Inhibited Diffusion Flames . 25 

g. Extinction of Laminar Diffusion Flames. 26 

h. Extinction of TUrbulcnt Diffusion Flames. 27 

4. Agent Effectiveness, Tbsting Methods, and the Role of Different 

Suppression Mechanisms . 28 

a. Agent Effectiveness and Tbsting Methods. 28 

b. Contributions of Physical and Chemical Mechanisms to Fire 

Suppression Effectiveness. 28 

c. Conclusions Regarding a Chemical Mechanism for Fire 

Suppression. 32 

C. HALON STABILITY .AND DEPLETION OF STRATOSPHERIC OZONE. 33 

1. Stability and Properties of Halons. 33 





























TABLE OF CONTENTS 
(CONTINUED) 

Section Title Page 

2. Brief Introduction to Atmospheric Structure and Chemistry. 33 

3. Current Commercial Halons and Depletion of Stratospheric Ozone . 34 

4. Means for Reducing ODP Values of Effective Firefighting Agents. 35 

III FAMILIES AND COMPOUNDS INCLUDED ON THE LIST. 37 

A. INTRODUCTION. 37 

B. SATURATED HALOCARBONS. 37 

1. Justification for Consideration . 37 

2. Past Flame-Suppression Measurements. 39 

3. Recommendations . 40 

a. Perfluoro Compounds . 40 

b. Photosensitive Compounds. 41 

c. Hydrogen-Containing Compounds . 41 

C. HALOGENATED KETONES, ANHYDRIDES AND ESTERS. 45 

1. Justification for Consideration . 45 

2. Past Flame Suppression Measurements. 48 

3. Recommendations . 48 

D. UNSATURATED HALOCARBONS . 50 

1. Justification for Consideration . 50 

2. Past Flame Suppression Measurements. 50 

3. Recommendations . 51 

E. HALOGENATED ETHERS AND RELATED COMPOUNDS . 51 

1. Justification for Consideration . 51 

2. Past Flame Suppression Measurements. 53 

3. Recommendations . 53 

F. HALONS CONTAINING IODINE. 55 

1. Justification for Consideration . 55 

2. Past Flame Suppression Measurements. 56 

3. Recommendations . 57 


xiii 


































TABLE OF CONTENTS 
(CONTINUED) 


Section 


Title 


Page 


G. SULFUR HALIDES . 57 

1. Justification for Consideration . 57 

a. Parent Properties. 57 

b. Chemical Analogs . 57 

2. Past Flame Suppression Measurements. 59 

3. Recommendations . 59 

H. COMPOUNDS CONTAINING PHOSPHORUS . 59 

1. Justification for Consideration . 59 

2. Past Flame Suppression Measurements. 61 

3. Recommendations . 61 

I. SILICON AND GERMANIUM FLAME INHIBITORS. 64 

1. Justification for Consideration . 64 

a. Family of Chemicals. 64 

b. Properties . 64 

2. Past Flame Suppression Measurements. 64 

3. Recommendations . 66 

J. METALLIC FLAME INHIBITORS. 66 

1. Justification for Consideration . 66 

2. Past Flame Suppression Measurements. 67 

a. Alkali Metal Inhibitors . 67 

b. Heavy Metal Inhibitors . 69 

3. Recommendations . 70 

K. INERT GASES. 71 

1. Justification for Consideration . 71 

2. Past Flame Suppression Measurements. 71 

3. Recommendations . 71 


xiv 





























TABLE OF CONTENTS 
(CONCLUDED) 


Section Title Page 

IV CONCLUSIONS AND RECOMMENDATIONS. 74 

REFERENCES. 76 

APPENDIX 

A DATA SHEETS FOR SELECTED COMPOUNDS. 91 


xv 







LIST OF TABLES 


Thble Title Page 

1 Compounds with no Bromine or Chlorine. 42 

2 Photosensitive Compounds . 43 

3 Compounds with no Bromine . 44 

4 Compounds Containing Bromine. 46 

5 Current Commercial Halons . 47 

6 Halogenated Ketones, Anhydrides and Esters . 49 

7 Unsaturated Halocarbons . 52 

8 Halogenated Ethers and Related Compounds . 54 

9 Recommended Halons Containing Iodine . 58 

10 Recommended Sulfur Halides . 60 

11 Phosphorous-Containing Gases and Liquids. 62 

12 Recommended Phosphorous-Containing Compounds. 63 

13 Silicon and Germanium Flame Inhibitors. 68 

14 Metallic Inhibitors. 72 

15 Inert Gases. 73 

16 Complete List of Recommended Compounds. 75 


xv i 



















SECTION I 


INTRODUCTION 


A. OBJECTIVE 

The objective of this project is to identify approximately 100 gases and/or liquids, covering a range 
of chemical and physical principles thought to affect flame suppression capability. This project is part of an 
overall effort to identify safe and effective fire suppression agents which are less detrimental to stratospheric 
ozone than the current commercial halons. A major element in the choice of compounds is also understanding 
the processes by which halons deplete stratospheric ozone. The list of compounds includes as much relevant 
information as possible to aid researchers who will use the findings of this work in the search for new flame 
suppressants. 


B. BACKGROUND 

1. Halogenated Hydrocarbon Fire Suppressants (halons) 

a. Definition and Designation 

Halogenated hydrocarbon fire suppressants refer to a group of chemical compounds 
widely used for firefighting and inerting (creation of an atmosphere which does not support combustion). Die 
compounds denoted as halons contain one or more halogen atoms 1 attached to a backbone of one or more 
carbon atoms. Only straight chain, fully saturated (each carbon atom is bonded to four other atoms) 
compounds are included. 


A simple nomenclature developed by the United States Army Corps of Engineers is 
widely employed to designate halons [1]. A particular halon is assigned a "halon number” where, reading from 
left to right, the first number is the number of carbon atoms and the remaining four numbers are the number 
of each halogen atom in order of increasing atomic weight, i.e., F, Cl, Br, and I. Any terminal zeros are 
dropped. As an example, trifluorobromomethane, CF a Br, is denoted as halon 1301. Note that some 
inconsistencies are possible when two or more carbon atoms are present. Consider CB^HCB^H and 
CBr 3 CBrH 2 . Both of these molecules would be designated as halon 2004. 

b. A Brief History 

Ford [2] has provided a brief perspective on the historical development of halons for 
firefighting purposes. Carbon tetrachloride (halon 104) was the first halon to be used. Starting in the early 
1900s and continuing until the mid-1960s, fire extinguishers containing this liquid were commercially available. 
In the late 1920s, methyl bromide (halon 1001) was found to be much more effective for fire-extinguishment. 
Because of its high toxicity, this compound was never widely utilized in the United States. It was employed 


x The five halogen atoms are: fluorine (F), chlorine (Cl), bromine (Br), iodine (I), and astatine (At). 
Astatine is very scarce and will not be considered further in this work. 


1 



militarily during World War II by the British (aircraft extinguishing systems) and Germans (aircraft 
extinguishing systems and marine systems) [1]. 

During World War II the Germans developed a fire-extinguishing agent known as 
"CB," chlorobromomethane (halon 1011), which was as effective as halon 1001 and less toxic. This agent was 
ordered to be installed on all German military aircraft, but apparently did not come into widespread use before 
the end of the war [1]. 


Following the war, the German research efforts were reviewed by the United States 
military; and, CB was extensively tested in this country [3]. CB was eventually adopted for use by the United 
States Air Force [4]. Despite the improvements represented by CB, it was recognized that significant 
toxicological problems remained and that fire suppression capabilities might be further refined. As a result, 
the Air Force [1] and the U.S. Army Corps of Engineers [5,6] began extensive testing of a wide range of 
possible compounds. These projects considered a number of possible firefighting agents, but the principal 
focus was the halons. 


The requirements for a new agent specified by the agencies are instructive because 
they generally apply to so-called "clean" firefighting agents. Klein [1] lists the Air Force requirements as: 

1. better extinguishing properties than CB over the temperature range of - 
54° C to 71° C, 

2. less toxic than CB, 

3. less corrosive than CB, 

4. suitable for class A, B, and C fires 2 , and 

5. higher specific gravity than CB. 

The Corps of Engineers’ specifications [6] were similar. The agent was to be: 

1. less toxic than carbon tetrachloride, 

2. less corrosive than carbon tetrachloride, 

3. a nonconductor of electricity, 

4. stable when stored for five years in any climate, and 

5. suitable for use over a temperature range of -54° C to 71° C. 

These investigations identified four halons as being particularly promising [2]: 
trifluorobromomethane (halon 1301), bromochlorodifluoromethane (halon 1211), dibromodifluoromethane 
(halon 1202), and dibromotetrafluoroethane (halon 2402). Halons 1301 and 1211 have since been developed 
and are utilized for a large number of applications. Halons 1202 and 2402 are also used, but less extensively. 


2 Fires are classified by the type of fuels and conditions present. Class A fires involve cellulosic materials 
such as wood or paper. Class B fires refer to liquid fuels such as hydrocarbons and alcohols. Fires where 
energized electronic components are present are denoted as Class C fires. If metal fuels such as sodium or 
aluminum are present, the fire is a Class D fire. 


2 


In this text these four halons will be denoted as "the current commercial halons" to differentiate them from 
the broader halon and halogenated organic molecular families. In the literature, the current commercial 
halons are often denoted simply as "halon." 

Since the 1950s, the current commercial halons have come into widespread use for 
firefighting and fire protection systems. The military uses these compounds extensively in a number of 
mission-critical areas. Similarly, current commercial halon systems have been incorporated into a number of 
civilian uses where costly or irreplaceable facilities, equipment, or items must be protected from fire. By 1985, 
the production and use of these halons in the United States had reached 6.4 x 10 6 kg [7], Halon systems are 
generally regarded as reliable, safe, and effective. As such, they have become an integral part of fire protection 
practice. 


Unfortunately, a major issue has arisen which threatens to curtail, if not eliminate 
use of the current commercial halons. This issue has been discussed extensively in the literature. A 
chronological summary of major events in this area has been compiled by S. Daws [8], Anderson [7] and 
Grant [9] provide excellent histories of the problem and prospects for future halon usage. A brief history is 
provided here to illustrate the nature of the problem and the suddenness with which it arose. 

In 1974, Molina and Rowland [10] hypothesized that the release of chlorine by 
photolysis of manmade chlorofluorocarbons (CFCs) in the stratosphere would result in a reduction of the 
ozone concentration which shields the earth’s surface from the sun’s harmful ultraviolet radiation. By 1978, 
the use of CFCs as propellants for aerosols was banned in the United States for all but essential uses. 

From 1978 to 1985, the movement of the global community toward the control of 
CFCs was very deliberate. This changed with the publication of results of ground-based observations indicating 
a 40 percent depletion of stratospheric ozone above the Antarctic at certain times of the year--the "ozone hole" 
[11], Satellite observations confirmed this to be a regional effect [12]. World-wide decreases in stratospheric 
ozone have since been detected [13]. By 1988, there was sufficient evidence for most scientists to conclude 
that stratospheric ozone depletion has occurred as a result of ozone reactions with Cl and Br atoms introduced 
into the stratosphere by manmade CFCs and current commercial halons [13]. 

In September of 1987, the Montreal Protocol was signed by the United States, 
Canada, members of the European Economic Community, and 23 other countries. This accord set limits on 
the production of CFCs and halons 1211 and 1301 in an effort to limit further destruction of stratospheric 
ozone. Initial discussions had focused primarily on the CFCs, but in the final document, production of the 
two halons is limited to their 1986 levels. Note that the limits require a cut in manufacture because 
production has increased significantly since 1986. The protocol was effective as of January, 1989 and limits 
go into force in January, 1992. 

Since the signing of the Montreal Protocol, evidence has continued to mount that 
halons contribute significantly to stratospheric ozone depletion. The current commercial halons are now 
recognized as being 3 to 11 times more destructive of stratospheric ozone on a molecule-per-molecule basis 
than the common CFCs due to the presence of bromine atoms in these fire suppressants [14]. It has been 
hypothesized that Br atoms can have even larger effects on ozone levels through a synergistic interaction with 
more abundant Cl atoms [15,16]. A recent study also suggests that atmospheric concentrations of halons 1211 
and 1301 are increasing at annual rates of 12 percent and 5 percent, respectively [17]. These rates are much 
higher than previously believed. 

These recent findings suggest that future accords may demand larger cuts in halon 
production than those mandated by the Montreal Protocol. During 1989, two bills were introduced in the 
United States Senate designed to limit the production of current commercial halons to levels lower than called 
for by the Montreal Protocol. The first mandates limiting production at 1986 levels as of July, 1989, with a 


3 




20 percent cut in 1992, a 50 percent cut in 1995, and a total ban in 1997. An exemption is provided for halon 
uses essential to national security needs. The second bill would place a surcharge on halons in 1992 of $1.50 
per pound multiplied by the chemical’s ozone depletion potential. For halon 1301 this would amount to 
approximately $15 per pound [9]. 

Future fire protection will, at a minimum, be considerably more expensive and more 
carefully controlled. It is also possible that all production of the current commercial halons will be halted and 
that the world will be dependent on supplies which have been stored or "banked" for fire protection 
requirements. Grant [9] refers to the current uncertainty concerning the future of halons as "one of the most 
challenging problems ever to confront the fire protection community.” 

c. Current Use of Halogenated Fire Suppression Agents 

The current commercial halons are rapid and effective fire inerting agents and 
suppressants. They have low toxicity at ambient temperatures (note that the toxicity of combustion products 
resulting from the use of halons is treated as a separate issue to be discussed in Section I-C-3) so that they 
can be effectively used in the presence of people. In addition, these compounds: are available as both liquids 
and gases and can therefore be used either as streaming or flooding agents; they are clean, leaving little or no 
residue (with the possible exception of combustion products); they are nonconductive, so that they can be 
applied safely to energized electronic equipment; and they are compatible with a wide variety of plastics and 
metals. 


The three most commonly used halons, 1301,1211, and 2402, have physical properties 
which are compatible with two distinct modes of deployment. Halon 1301 is a gas at ambient temperatures 
and is most often used in "total flooding" applications. The gas is introduced into an enclosure or space at 
levels sufficient to suppress fires and prevent further ignition. Halon 2402 is a low-boiling-point liquid at 
normal room temperatures. As such, it is an ideal streaming agent for local application, which is its most 
common use. Halon 1211 has physical properties which are intermediate between 1301 and 2402 and has been 
used much the same as halon 2402. 


The proceedings of a meeting held by the National Academy of Sciences cites many 
examples of where halons have been used [18]. The most important of these are facilities where "clean" agents 
are required and situations where personnel must remain during fire emergencies. 

"Clean" fire extinguishing agents are often used when costly and/or irreplaceable 
objects are present. Libraries, museums, and archives of important national documents are examples of 
facilities which are currently protected by halon total-flooding extinguishment systems. Extinguishing agents 
such as water or dry powders would be very damaging in these environments. Clean agents are also required 
where other alternatives would interfere with important functions of a facility or device. Many computer 
facilities, electronic record depositories, and telephone exchanges are protected by current commercial halon 
total flooding systems. In this context, the lack of electrical conductivity is also crucial. Additional examples 
include the mission-critical use of these halons on Air Force flight lines to combat aircraft and hangar fires 
and their use to fight nacelle fires in both civilian and military jet engines. 

There are also circumstances when the evacuation of personnel during a fire 
emergency is not feasible. Control rooms of nuclear reactors and aircraft carriers are examples of facilities 
that must be staffed even after deployment of an extinguishing agent. Numerous mission-critical military 
situations have the same requirement. Current commercial halon systems are employed to protect ground- 
based radar and military command and control centers, such as those on Navy ships, as well. Similar systems, 
which are designed to suppress fires and explosions in armored vehicles during combat, are also in place. 


4 



The use of the current commercial halons for the requirements mentioned above, as 
well as numerous others, has become very widespread during the past two decades. Fixed suppression systems 
number in the thousands. There are even more portable extinguishers in use, making halons an integral part 
of fire protection engineering. As such, they currently play a crucial role in the protection of national 
resources and national security of the United States. The same is true for other industrialized countries. 

2. The Search for Halon Alternatives 

The discussion in the last section suggests that the manufacture of the current commercial 
halons is likely to be phased out by the year 2000. On June 21,1988, E.I. du Pont de Nemours & Co., a major 
manufacturer of halon 1301, announced plans to initiate just such a phaseout [9]. Other companies have 
announced that production will cease as soon as safe alternatives are available [19]. 

The fire protection community has recognized the seriousness of the situation. In 1987, an 
International Halon Research Project (IHRP) was initiated under the auspices of the National Fire Protection 
Research Foundation to conduct collaborative research [9], The IHRP has released two reports [20,21] 
designed to be first steps toward reducing releases of halon 1301 into the atmosphere. A project designated 
as the "Best/Essential Use Project" was initiated by IHRP at the beginning of 1989. This project is designed 
to identify applications where commercial halons are currently employed which can use alternative fire 
protection methods [9]. 

Some essential fire protection needs cannot be met by existing alternatives to the current 
commercial halons. Users of such systems have recognized the seriousness of the situation and have begun 
to take action. Organizations are being formed to guide the search for alternatives which are effective fire 
suppression agents and have other desirable properties of the current commercial halons, but which do not 
pose a significant threat to stratospheric ozone. 

One of the first steps in this process was a meeting organized by the United States Air Force 
and the Environmental Protection Agency, which was held at Tyndall Air Force Base in late 1988 [22,23]. 
Numerous presentations were made delineating the problems faced, establishing the current understanding, 
and identifying available resources. The participants recognized the need for the development of new agents 
and a cooperative effort between private and public research interests was suggested as a mechanism for 
accelerating the development of new chemicals. Recommendations were provided for research areas to serve 
as the basis of a comprehensive plan. 

During the past year, the Halon Alternatives Research Corporation (HARC) was established 
by the private sector to foster efforts to develop alternatives. A government working group has been formed 
to serve as liaison between HARC and Federal agencies. Efforts are moving forward to implement an effective 
research plan. Reference [19] discusses these efforts. 

In the meantime, a comprehensive research plan for identifying suitable replacements for the 
current commercial halons has been formulated by an ad hoc technical committee as an outgrowth of the 
Tyndall meeting [24]. The project described in this report is an element of this plan. 


C. SCOPE/APPROACH 

1. Nature of the Problem 

The current commercial halons are likely to become prohibitively expensive or totally 
unavailable for fire suppression and inerting purposes by the year 2000. There is a clear requirement for 


5 




replacement chemicals which have the following properties: fire suppression effectiveness, low toxicity, low 
residue, favorable physical properties, low electrical conductivity, and materials compatibility. At the same 
time, these chemicals must be far less destructive to stratospheric ozone than the halons which they are 
expected to replace. Unfortunately, since the current commercial halons were thought to possess ideal 
properties, there have been very few research efforts since the late 1940s designed to identify alternatives. The 
last wide-ranging study was that carried out by the Purdue Research Foundation [25] during the late 1940s 
with support from the U.S. Army Engineers Research and Development Laboratories. (Even this study 
focused primarily on fully halogenated compounds.) As a result, very little research infrastructure and 
scientific expertise exists for the search. 

As discussed in Section II.C.1, the unusual chemical stability of the halons contributes to their 
effectiveness as fire extinguishing agents. Unfortunately, it is this same chemical stability which allows these 
chemicals to pass unreacted through the earth’s troposphere and penetrate the stratosphere. The culmination 
of this process is the catalytic destruction of stratospheric ozone by bromine atoms which are produced in the 
photolysis of halons (see discussion in Section II.C.3). The development of new fire suppression agents which 
have the desirable properties of the current commercial halons but which are less harmful to stratospheric 
ozone will probably require careful balancing of chemical reactivity behavior. Principles need to be developed 
to identify the best possible alternatives. 

TWo general search strategies for alternatives are immediately obvious. One approach is to 
consider chemical modifications of the commercial halons such that their fire suppression capability is 
maintained, but their atmospheric reactivity is increased to the point where transport to the stratosphere is 
minimized. The second approach is to identify completely different chemical families which offer the required 
characteristics. A wide-ranging search must incorporate both approaches. 

2. Structure of the Research Program 

Under U.S. Air Force sponsorship, the National Institute of Standards and Technology (NIST) 
has undertaken to identify approximately 100 compounds covering a range of chemical and physical properties 
believed to affect flame suppression capability. Following a thorough literature search, NIST staff members 
have devised a strategy for identifying these compounds. This report serves to document the existing 
knowledge of fire suppression, the search strategy employed, and the physical and chemical properties of the 
compounds which have been chosen for consideration. 

3. Search Strategy 

The first priority of this project was to define the boundaries for the search. No systematic 
search for new flame suppressants has been done since the late 1940s. We did a thorough literature search 
to define the current understanding of flame suppression. This understanding serves as the foundation for the 
remainder of the project. 

The next task was the identification of chemical families to be considered. The decision was 
made to consider as wide a range of chemical families as possible. As a result, chemicals which have been 
investigated in the past for fire suppression effectiveness were identified and assessed. The search was then 
expanded to include classes of chemicals which have not been considered for fire suppression purposes 
previously. 


Several different approaches were possible for the search. One was to identify the compounds 
offering the best possibilities as short-term alternatives for the current commercial halons. The project team 
quickly realized that the existing knowledge of fire suppression, ozone depletion, toxicity, and other physical 
properties was insufficient. An alternate strategy was to choose compounds in such a manner that the 


6 



principles necessary for the selection of alternatives for the current commercial halons would be tested. These 
principles would then be available to guide future work. An intermediate path was ultimately chosen. As a 
result, the final list of compounds contains members which are likely to be among those agents chosen in the 
near future for further testing as alternatives for the current commercial halons. At the same time, other 
molecules which are definitely not suitable for replacing these halons are included to allow the development 
of the principles required to guide the search for the best possible alternatives. This approach facilitates both 
short- and long-term efforts to find the best possible alternatives for the current commercial halons. The 
intention was to develop a broad base of knowledge to aide researchers and manufacturers in responding to 
future changes in the regulatory environment. 

Next, we decided upon a method to weight the various properties which make the current 
commercial halons so effective for fire suppression purposes. These properties are listed in the first paragraph 
of this section. We decided to focus on two: 

■ fire suppression and inerting, the principal functions of these agents; and 

■ stratospheric ozone impact, the principal reason the future of these chemicals has 
come into question. 

The project team realized that the decision to focus on fire suppression and ozone depletion 
required that several other important characteristics be deemphasized. Perhaps the most crucial of these is 
toxicity. Clearly, one of the most important characteristics of existing agents is that they can be deployed 
effectively in the presence of people. This is a result of their relatively low toxicity. Unfortunately, toxicity 
cannot be predicted confidently from chemical structure. Generally, the toxicity of a particular chemical is 
determined by a battery of expensive and time-consuming tests. The approach taken for this study is to report 
existing toxicological data for compounds included on the final list, but not to attempt to predict the 
toxicological behavior of compounds for which no previous studies are available. 

Some general guiding principles for toxicological behavior are known (e.g., compounds 
containing fluorine tend to be less toxic than their iodine analogs) and have been incorporated into discussions 
of the chemical families which appear later in the report. Compounds which are known to be toxic, and in 
some cases highly toxic, are included on the final list. These compounds are not considered to be viable 
alternatives for the current commercial halons (even though certain limited uses as fire suppressants are 
possible for some) but are included to test principles of fire suppression or ozone depletion. 

The toxicity referred to above is that for the chemicals themselves. Another important 
consideration is the toxicity of decomposition products which results from the deployment of an agent on fires 
or in other heated environments. Numerous discussions of this topic (e.g., references [6,26,27,28,29,30,31,32]) 
with regard to the current commercial halons indicate that hydrogen halides, halogens, and other halogenated 
compounds are formed. All of these compounds are noxious and toxic at relatively low levels. These species 
can also damage a wide range of materials and cripple electronic equipment. 

These dangerous combustion products do not usually pose serious toxicological and materials 
problems when the current commercial halons are deployed for fire suppression. This observation is traced 
to the effectiveness of these fire extinguishment agents. If properly applied, the fire is extinguished so rapidly 
that only small concentrations of these products are created [27,29,31]. Observed levels of fire products 
attributable to the use of these agents are usually not dangerous. Certain situations involve exposure of the 
current commercial halons to deep-seated fires [28] or hot appliances [31] where high levels of noxious gases 
can be formed by decomposition of the agents. Conditions where the combustion products resulting from 
usage of the current commercial halons pose a realistic danger have not been investigated extensively and no 
quantitative results were identified during the literature search. 


7 



The combustion products generated by the application of alternative compounds are not likely 
to be more toxic than those generated by the use of the current commercial halons. It is further assumed that 
any replacement agents will have roughly similar fire suppression effectiveness. On this basis, it can be 
concluded that toxic products generated by the use of alternatives will not represent a more significant threat 
than those resulting from the use of the current commercial halons. While recognizing that a potential 
problem exists, the likelihood of producing toxic products during fire suppression has not been considered in 
choosing the chemicals included on the list. 

Ideal alternatives for the current commercial halons would be "drop-in" agents. In other 
words, the physical and materials compatibility of the replacements would be such that they could be used in 
existing halon suppression systems. This criterion is much too confining for a wide-ranging search for 
alternatives. Some general selection of physical properties is enforced by requiring that compounds on the 
list be gases or easily vaporizable liquids, but compatibility with current commercial halons is not included as 
a criterion. Even the simple requirement of being a liquid or gas is waived for a few compounds. 

Generally, compounds which are gases or vaporizable liquids will leave no residue if pure. 
The electrical conductivity of most gases and liquids is low, but no attempt has been made to consider this 
property in the search strategy. Materials compatibility is very important for the design of practical fire 
suppression systems. Nonetheless, it was decided that consideration of this property should also be deferred 
and it was not used as a criterion for including compounds on the list. 

4. Report Philosophy 

a. Breadth of Coverage 

This project is intended to be comprehensive. The current understanding of fire 
suppression and ozone depletion is utilized to choose approximately 100 compounds for testing purposes. A 
summary of this understanding is provided to give the reader the needed background to evaluate the basis for 
recommendations which are made. Large numbers of references to original literature are provided to allow 
readers to obtain additional details. 

We have generated a detailed report which we hope will capture the current 
understanding of the problem, give clear explanations of approaches utilized in the study, and provide a 
comprehensive document which can serve as the basis of the search for alternatives to the current commercial 
halons. As much detail as possible is provided for each chemical species included on the list to aid researchers 
who might test or evaluate these chemicals in the future. 

b. Report Structure 

The report is divided into 4 major sections. This section began with a statement of 
the objective of this study. Next, the properties of the current commercial halon fire extinguishing agents were 
enumerated with regard to their composition, historical development and uses. In the following subsection, 
the relationship of this project to other ongoing efforts is discussed. 

Section II is devoted to providing an overview of relevant areas of combustion science. 
Included are descriptions of the current understanding of fire suppression mechanisms, various approaches 
which are utilized for testing flame suppression effectiveness and a brief introduction to the mechanisms by 
which the current commercial halons are thought to contribute to the depletion of stratospheric ozone. 

Section III is divided into subsections corresponding to the chemical families which 
were considered in this investigation. A justification for including each chemical family, as well as a discussion 


8 



of previous tests of family members for fire suppression effectiveness, is presented. Recommended compounds 
from each chemical family are tabulated, along with relevant properties, following each subsection. 

The conclusions of this investigation are presented in Section IV along with the 
complete exploratory list of chemical alternatives. Separate data sheets for each of the compounds on this 
list are included in an appendix. 

5. Relation of This Work to Other Efforts 

The Engineering & Services Laboratory of the U.S. Air Force is funding a number of research 
efforts related to usage of current commercial halons and alternatives. The goals are to provide a short-term 
reduction in halon releases and a long-term solution to the problem by identifying appropriate alternatives 
for training and use. An effort has been underway at the New Mexico Engineering Research Institute 
(NMERI) for several years [33]. This program has primarily considered means for reducing releases of the 
current commercial halons to the atmosphere and the replacement of these halons with close chemical 
relatives and mixtures. 

The National Institute of Standards and "technology (NIST) is performing three projects- 
"Construction of an Exploratory List of Potential Replacements for halons 1211 and 1301," "Preliminary 
Screening Procedures and Criteria for Replacements for halons 1211 and 1301," and "Halon Candidate Agent 
Database." The team members of these three projects have interacted closely so there is a great deal of 
coordination between these efforts. 

It is anticipated that the compounds included on the master list will be passed through the 
preliminary screening procedures under development at NIST. This initial screening will serve as a highly 
effective test of the screening process and will generate initial information with regard to the relative ordering 
of the compounds in terms of screen properties. Careful experimentation will be required to more fully 
develop the fire suppression and ozone-depletion principles which the list of compounds has been explicitly 
chosen to test and verify. 

The database project has been the source for many of the physical properties included in this 
report. At the same time, many of the compounds identified in the current research effort will be included 
in the database. Some of the findings of the literature search for this work will also be added to the database. 

The United States Environmental Protection Agency is funding a project entitled 
"Examination of Fire Types" which is being performed by Factory Mutual Research Corporation of Norwood, 
MA This project is designed to identify the classes of fires for which the current commercial halons are used 
and recommend generic fires for testing purposes. The results of this project will be used for practical fire 
testing of compounds identified as the most likely candidates as alternatives for the current commercial halons. 


9 



SECTION II 


CURRENT UNDERSTANDING 


A. COMBUSTION AND FIRE CONCEPTS 

This Section reviews the fundamental concepts of combustion and fire required for understanding the 
discussion on fire suppression and inerting as well as the literature discussed during the remainder of the 
report. The approach taken is intentionally introductory and many details are omitted. Any of several 
excellent texts on these subjects (e.g., [32,34,35,36,37]) can be consulted for more in-depth treatments. 

1. Definitions of Combustion and Fire 

"Combustion" and "fire" have general meanings which are recognized by everyone. At the 
same time it is difficult to provide specific definitions. For the purposes of this report, combustion will be 
defined as "a rapid and persistent chemical reaction which combines fuel and oxygen to produce heat and 
light." The quotation is adapted from an introductory book by Cottrell [38]. 

The above quotation was provided by the author to define fire and not combustion. 
Generally, a more restrictive definition, such as that provided by Berl [39], is used for fire. Fires are "special 
cases of combustion where the addition of fuel and oxidizer to a flame is not under human control" [39]. The 
class of fires of interest for this report are uncontrolled and unwanted fires. These fires are generally harmful, 
destructive, and/or dangerous. 

2. The Fire Wangle 

The fire triangle is an introductory concept often employed to remind readers of the three 
conditions which must be present simultaneously for combustion to take place and sustain itself. These 
requirements are fuel, oxygen, and heat. These are often drawn along the sides of a triangle to emphasize that 
all three are required. 

The fire triangle also provides insights into how fires can be suppressed. If any of the sides 
of the triangle is removed, the fire will be extinguished. This will become clearer in the section on 
extinguishment mechanisms. 

3. Combustion and Fire Classifications 

Numerous methods are used to classify types of combustion and fire. A few of these are 
mentioned to provide background and nomenclature for the discussions which follow. 

a. Flaming Combustion and Smoldering 

TWo types of combustion are common in fires. The first is known as flaming 
combustion and involves chemical reactions occurring primarily in the gas phase. The name arises from the 
fact that the combustion is characterized by the emission of visible and infrared (heat) electromagnetic 
radiation by the fire. These emissions are the means by which a person often detects the presence of a fire. 
The visible radiation is either (a) emission from electronically excited chemical species formed as the result 


10 



of the chemical reactions or (b) thermal radiation from small particles (soot) located in regions of high 
temperature which obey Planck’s black body radiation law. 

The second type of combustion is only important for burning solids such as wood or 
plastics and involves chemical reactions occurring directly at, or near, the surface within a solid. These are 
heterogenous reactions since the solid surface reacts directly with oxygen from the surrounding gases. This 
process is known as smoldering. 

Flaming combustion is usually more vigorous and generates higher temperatures than 
smoldering. As a result, flaming combustion is often more dangerous and poses a more immediate threat. 
On the other hand, smoldering can generate significantly higher concentrations of toxic gases than flaming 
combustion (this depends on oxygen availability) and can be more difficult to extinguish. Smoldering often 
evolves into flaming combustion. 

b. Fuel Types 

One example of fire classification based on fuel type is given in the footnote at the 
bottom of page 2. This scheme is employed to allow people who must control these types of fires to rapidly 
identify a fire class and to determine the best available means for firefighting. 

Combustion processes can also be classified based on the normal state-gas, liquid, 
or solid-of the fuel. This classification is important because it determines the mechanisms by which the fuel 
is made available for flaming combustion. Note that liquids and gases do not support smoldering combustion 
under normal conditions. 

For gases, the fuel is provided directly to the combustion region. An example is the 
flow of natural gas to a laboratory Bunsen burner. The supply of gaseous fuel is determined simply by the gas 
flow rate. 


Liquid fuels are considerably more complex. Burning conditions for liquid fuels can 
vary. As an example, consider the common configuration of flaming combustion occurring above a pool of 
the liquid fuel. The gaseous fuel results from vaporization from the liquid surface, and the fuel supply rate 
is determined by the rate of fuel vaporization and the surface area of the pool. The vaporization rate per unit 
area is highly sensitive to the liquid temperature and the radiation flux from the fire. Since flaming 
combustion can heat the liquid by thermal conduction, convection, and radiation, it is possible for the flames 
to affect the vaporization rate, and the fire can therefore influence its own behavior. 

Combustion involving solid fuels is by far the most complicated. Here, the heat 
feedback from the overhead flames is crucial for generating the gaseous fuels required for flaming combustion. 
In some cases a solid fuel can simply melt and produce a vaporizable liquid which burns as a pool fire. More 
generally, however, complex pyrolysis and chemical reactions occur near or at the solid surface resulting in the 
release of gaseous products which serve as the fuel. These processes can result in even more complicated 
behaviors in the solid such as charring. Wood burning is a common example of this type of combustion. 

c. Premixed and Diffusion Flames 

These terms refer to the means by which the fuel and oxidizer are mixed. When the 
fuel and oxidizer are mixed before combustion the flames are known as premixed. A flame propagating down 
a tube containing a mixture of a hydrocarbon fuel and air is an example of a premixed flame. A diffusion 
flame is one where the fuel and oxidizer are not premixed, and the combustion and mixing take place at the 
same positions). Common means for mixing are molecular or turbulent diffusion, hence the name. A candle 


11 



is an example of a diffusion flame, where the combustion is the result of molecular diffusion of fuel and 
oxidizer species. 


It is possible to have intermediate cases between premixed and diffusion flames. A 
Bunsen burner allows partial premixing of fuel and air, but the amount of air provided by the burner may be 
insufficient for complete combustion. In this case, additional reaction between excess fuel and ambient oxygen 
occurs. 


d. Laminar and Tbrbulent Combustion 

For gas-phase combustion, the behavior of the flow field is central to the flame 
characteristics for both premixed or diffusion flames. Gas flows are typically characterized as laminar or 
turbulent, and, while a complete discussion of these effects is beyond the scope of this report, a few general 
comments are necessary. Here we follow the discussion of Fox and McDonald [40]. 

These classifications of flow are based on macroscopic behavior. In laminar flow, 
there is no mechanical muting between adjacent layers of the fluid. The flow is "smooth." The only dispersive 
mechanisms are the result of molecular diffusion. By contrast, turbulent flow is characterized by the 
generation of numerous and apparently random velocity fluctuations. These velocity fluctuations are the result 
of the formation of vortices extending over a wide range of size scales. These motions provide a mixing 
mechanism for adjacent layers of fluid. As a result, transfer and mixing of momentum, heat, and mass is much 
greater for turbulent flows than observed in laminar cases. 

The type of flow which occurs is sensitive to a large number of system parameters. 
In general, small area flows at low velocities will be laminar. An example of a laminar flame is a small candle. 
As the area of the flow or its velocity increases a transition to turbulent behavior occurs. Most fires covering 
areas greater than a few centimeters involve turbulent combustion. 

As might be expected, the behaviors of combustion systems involving laminar and 
turbulent flows are different because of large variations in the rates of temperature, chemical species, and 
momentum mixing. 


e. Lean, Stoichiometric, and Rich Combustion 

These terms refer to the relative amounts of fuel and oxidizer available for 
combustion. When the amount of oxidizer is exactly that needed to convert all fuel to final products (CO z 
and H z O for most fuels) the combustion is said to be stoichiometric. If excess air is available, the combustion 
is lean, and in the opposite case, (fuel in excess), the combustion is rich. 

4. Basic Combustion Concepts 

The attempt to understand combustion is a part of the earliest of modern scientific research 
[41]. The complex nature of the interaction between the flow fields and the chemistry made progress difficult. 
However, today the major elements of the problem are understood and modeling calculations provide 
reasonable results for simple combustion systems. 

a. Thermodynamics 

Combustion occurs when a fuel and an oxidizer (typically air) mix and are ignited. 
The resulting reaction generates heat and light. Thermodynamics provides the tools to calculate the amount 
of heat which can be released for any given fuel. It also allows the calculation of the maximum possible 


12 



temperature which can be generated by a flame and the equilibrium concentrations of species which are 
produced. As an example, consider the combustion of the simplest hydrocarbon-methane. The overall 
reaction of methane with oxygen can be written as 

CH 4 + 20 2 - 2H 2 0 + C0 2 . 

The heat of reaction at room temperature is calculated as the sum of the standard heats of formation for the 
products minus that for the reactants. The net heat release for the above reaction is 802 kJ/mole of CH 4 
consumed. This quantity, the heat of reaction of fuel with oxygen to produce water and carbon dioxide, is 
called the heat of combustion. Similar calculations can be made for any fuel, whether it is a simple gas (as 
above) or a complex fuel such as wood. 

The heat of combustion can be used to estimate the maximum temperature a flame 
can achieve. This is called the adiabatic flame temperature. Tb calculate this quantity, it is necessary to know 
both the heat of combustion of the fuel and the composition of the reaction mixture. When methane is 
burning in air, this means that we must account for the nitrogen, which does not take part in the reaction. 
In addition, it must be assumed that all of the heat released by the combustion reaction goes into heating the 
product gases and the nitrogen. Calculation of the adiabatic flame temperature also requires input of the heat 
capacities of all product and inert gases. The heat capacity of a gas is simply the number of joules that are 
required to increase the temperature of one mole of the gas one Kelvin. In general, the heat capacity of a 
molecule increases with increasing molecular complexity-the more atoms in the molecule, the more energy 
required to heat the molecule. 

The calculation of equilibrium composition of the reaction mixture depends on the 
estimated flame temperature and requires knowledge of the heats of formation of stable species (such as CO), 
as well as, the heats of formation of the radicals having the highest concentrations in the flame (for example, 
hydrogen atoms, H; hydroxyl radicals, OH; oxygen atoms, O; hydroperoxide, H0 2 ; and methyl radical, CH 3 ). 
Finally, data on the entropy of formation for all of these species is needed. This process is iterated until 
convergence. These calculations can be extended to include other additives, such as inhibitors, provided that 
adequate thermodynamic data are available. The approximation of adiabaticity is generally good, but always 
produces an upper limit for the temperature. Equilibrium composition calculations are generally adequate 
if soot concentrations are not high. The data necessary for these calculations are, for the most part, well- 
known [42] and the calculations are readily performed. 

Thermodynamics can be used to predict the amount of heat and the composition of 
the gases produced, but it does not address the question of how fast the reaction occurs. Tb do this it is 
necessary to examine the kinetics of combustion processes. 

b. Chemical Kinetics 

Combustion can be described as rapid oxidation. The yellowing of a piece of paper 
over time is an oxidation process as well, but it is slow. When a flame is touched to the same piece of paper, 
the resulting oxidation is much more rapid. The difference is the result of variations in the overall rates of 
reaction for the processes or, in other words, the kinetics. In the methane/air example given above, the fuel 
and oxidizer could have been mixed in a cylinder at room temperature and there would have been no reaction. 
On the other hand, even a small spark passing through the mixture would have initiated a rapid, indeed almost 
explosive, reaction. 


Tb understand these differences in reaction behavior it is necessary to examine 
chemical kinetics in general, and, more specifically, the kinetics of ignition and propagation of flames. Overall 
reactions, such as the combustion of methane, generally occur by a number of individual reaction steps 
involving molecular species. These reaction steps are characterized by the reactants, the products, and the rate 


13 






at which they occur. The reaction of H atoms and 0 2 molecules to produce OH radicals and O atoms, an 
extremely important combustion reaction, is written as 

H + 0 2 - OH + O. 

The rate at which the reaction occurs can be defined in terms of the rate of loss of 
a reactant, -d[H]/dt or -d[0 2 ]/dt, or the rate of production of a product, d[OH]/dt or d[0]/dt. (The notation 
[X] denotes the concentration of X in moles/volume.) The reaction rate is equal to the product of the 
reactant concentrations times a rate constant, k, 

Rate = k[H][0 2 ], 

where the rate constant is usually expressed as 

k = AT n exp(-E a /RT). 

The energy of activation,E a , is a measure of the energy barrier that the reactants must overcome to produce 
products. The preexponential factor, AT", incorporates the rate of collision between the reactive species in 
the gas phase and the probability that if the reactants have sufficient energy, reaction will occur. In general, 
most of the temperature dependence for a reaction step is the result of the energy of activation since it 
appears in an exponent. The form of the equation is such that for E a > 0, the reaction rate increases with 
temperature. Tb understand why the methane/air mixture does not undergo combustion when mixed at room 
temperature, it is necessaty to examine the details of the chemistry of combustion. 

c. Combustion Chemistry 

Combustion chemistry is primarily the chemistry of free radicals-their production, 
reaction, and destruction. A free radical is a molecular species that is very reactive because it contains 
unpaired electrons. These species are not generally thermodynamically stable at room temperatures -- they 
recombine to form stable molecules. At flame temperatures, typically > 800 K, their recombination (loss) 
processes are not as fast as their generation rate. Thus, the radicals’ concentrations are relatively large, and 
their high reactivity dominates flame chemistiy. 

The most important free radicals in most practical flame systems are H and OH. In 
addition, CH 3 , H0 2 , and O are important in some situations. Beyond these species, dozens of other radicals 
play a role in the combustion of complex fuels. The chemistry of combustion is complex because of the very 
large number of species and because, at different temperatures and different ratios of fuel to oxidizer, different 
reactions among these species will dominate. We can characterize the major types of reactions in terms of 
their role in the overall chemistry. The usual classifications are: 

■ Initiation Reactions. These are reactions that provide the first radicals to start the 
combustion process for the fuel and oxidizer. 

■ Branching Reactions. These are reactions that increase the number of radicals and therefore 
the overall reaction rate. The most important of these reactions consume the oxidizer. 

■ Propagation Reactions. These reactions usually consume fuel or oxidizer with no net change 
in the number of radicals. 

■ Termination Reactions. These reactions reduce the net number of radicals or produce 
unreactive radicals from reactive radicals. 


14 



The branching reactions are common to all practical combustion systems. This is 
especially true for the branching reaction H + 0 2 -» OH + O discussed above. The individual initiation and 
propagation reactions often depend on the fuel. The important termination reactions are common to many 
systems. 


Initiation reactions are typically like these: 

H 2 + O-j —* HO 2 + H 

ch 4 + o 2 - ch 3 + ho 2 

C2H 6 - 2CH 3 . 

In these reactions, stable molecules react to generate radicals such as H and CH 3 . These reactions typically 
have activation energies greater than 200 kJ/mol. These are the only reactions that produce radicals directly 
from the stable fuel and oxidizer. The high activation energies of these reactions mean that the rate constants, 
k, vary by over 30 orders of magnitude between room temperature and combustion temperatures. Even at 
combustion temperatures, these are typically slow reactions with rate constants below all of the other reactions 
we will discuss. Tb illustrate how slow these processes are, the reaction between H 2 and 0 2 will produce less 
than one H atom per century at room temperature. Combustion does not occur unless the temperature is very 
high or there is some other source of radicals. Once the radicals are produced, the initiation reactions are 
completely unimportant; and the rate of reaction is set by the rates of the branching and propagation 
reactions. 


The most important branching reaction for practical combustion systems is: 

H + 0 2 -* OH + O. 

This reaction has an activation energy of 69 kJ/mole and an average pre-exponential factor. At combustion 
temperatures this reaction occurs about one in every hundred collisions. Most importantly, this reaction 
produces two reactive radicals, OH and O, for each radical (H atom) which is consumed. Reactions of O 
atoms with the fuel such as: 


O + H 2 - OH + H 
O + CH 4 - OH + CH 3 

are also important. The OH free radicals produced by these reaction steps react primarily with fuel molecules 
in propagation reactions. 

The propagation reactions are highly variable since they depend on the fuel type. 

Some examples are: 


H 2 + OH -» H 2 0 + H 
CH 4 + OH -* H z O + CH 3 
CO + OH — C0 2 + H. 

In each of these cases, the OH radical reacts with a fuel molecule (H 2 ,CH 4 ,CO) to produce a stable end 
product (H 2 0,C0 2 ) and another radical (H,CH 3 ). All of these reactions have low activation energies (5-20 
kJ/mol), although the third is an unusual reaction and is slower than the first two at flame temperatures. The 


15 




propagation and branching reactions are typically fast reactions that, once started, will rapidly consume all of 
the available fuel or oxidizer. 

The termination reactions either produce a radical with greatly reduced reactivity or 
produce stable products from two radicals. Examples are: 

H + 0 2 + M - H0 2 + M 

H + OH + M -* H 2 0 + M 

CH 3 + CH 3 + M -* C2H 6 + M 

where in each case M is some third body that does not participate in the reaction, e.g, N 2 . These reactions 
are typically slower than branching or propagation reactions at flame temperatures, but have negative 
activation energies, so that they become faster at lower temperatures. 

It is possible to understand very simple combustion systems, such as H 2 /0 2 within 
the context of a manageable set of reactions; and, for special cases, algebraic solutions of the rates of growth 
for the species involved can be found. In general, the reaction systems are so complex that computer modeling 
is essential just to understand the kinetics. When the effects of flow fields, mixing, and diffusion are included, 
even the simplest system must be modeled. 

It is nevertheless possible to use the simple ideas discussed here to understand the 
general principles of combustion. For example, the stability of combustible materials at room temperature 
is a result of the high activation energies for their initiation reactions. The discussion thus far is only strictly 
applicable to simple gaseous fuels, but the same general features pertain to more complex fuels; that is, the 
initiation reactions which involve the creation of radicals from stable species will be slow at room temperature. 
In addition, the rate of combustion, once initiated, can be understood in terms of the branching and 
propagation reaction rates. The rates for branching reactions increase with temperature and are reasonably 
fast, and the radical attack on the fuel molecules is generally fast and has low activation energy. As a result, 
the overall reaction rate is sufficient to maintain combustion for high temperatures. 

The same kind of analysis can be applied to the role of suppressants in the flame. 
Tb stop a combustion process one must either remove the radicals faster than the chain branching process can 
produce them (note that the propagation steps only maintain the number of radicals) or reduce the 
temperature so that the chain termination steps can dominate. These are really the same idea: in both cases 
the radicals are removed faster than the branching process can create them. 

Consider, for example, the primary chain branching reaction 

H + Oj —*• OH + O. 

This reaction has an alternate channel, 

H + O z + M -» H0 2 + M. 

Both of these channels are always present. The second channel is pressure-dependent, that is the second order 
rate constant increases with pressure. The rate expressions of these two processes are such that at atmospheric 
pressure, the rates are roughly equal at 1000 K. Below this temperature the recombination channel (producing 
H0 2 ) dominates, while above this temperature the branching channel dominates. The H + O, reaction is also 
important since it is the predominant path for 0 2 consumption in combustion. If there is competition for the 


16 



H atoms by other reactions, these will correspondingly reduce the net rate of the chain branching reaction by 
reducing the concentration of H atoms. 

Using the ideas expressed here, it is also possible to see what role an ignition source 
plays in the combustion process. The rate of radical production by the initiation reactions is so slow that, at 
room temperature, the termination reactions will soak up any radicals faster than the branching reactions. 
If a sudden, concentrated local source of radicals is introduced, the branching and propagation reactions will 
produce enough excess energy to heat the gas locally and start the combustion. This local radical source could 
be a spark or a small flame. Once the combustion has been started, the problem of flame inhibition goes back 
to the same ideas discussed above. For simple combustion systems, H 2 /0 2 or 0H 4 /O 2 , the kinetic data is well 
enough established that modeling is a useful tool for analyzing the combustion process. Although there is 
some kinetic data on halocarbons and other halogens, it is inadequate for computer modeling of the flame 
suppression process. 


d. Thermal balance 

All of the concepts discussed above can be incorporated into the idea of thermal 
balance. The discussion above can be simply summarized as follows: the energy released in the combustion 
process must be sufficient to keep the temperature high enough so that the branching reactions can stay ahead 
of the termination reactions. The thermal losses, heating up new fuel and oxidizer and breaking the initial 
bonds, are made up by the thermal gains, exothermic production of combustion products. In a practical system 
there are other loss mechanisms such as radiation, conduction and convection that the exothermic processes 
must also make up. The heating of new fuel will, in a diffusion flame, also include the loss of H atoms to the 
fuel. This loss of H atoms can be viewed as a mechanism for self-extinction and is responsible for rich 
combustion limits. 


These ideas can be translated into some simple rules of thumb for the limits of 
combustibility. For most organic fuels the heat of combustion is found to be approximately 425 kJ/(molc-0 2 ) 
[43,44,45,46,47], Note that the mole-0 2 means per mole of 0 2 consumed. Flammability limits, both rich and 
lean, and the limits in the presence of inerts have empirically been found to be expressed in terms of a limiting 
heat capacity for the reacting mixture (fuel/oxidizer/inert) of 210 J/(mole-0 2 K). That is, the mixture will not 
sustain combustion if the heat capacity of the mixture is greater than 210 J/K for each mole of oxygen present 
in the mixture [44,48]. Note that excess oxygen or fuel acts simply as an inert gas that adds to the heat 
capacity of the system. Thus, the addition to the heat capacity of the mixture can be accomplished by the 
addition of an inert, excess fuel, or excess oxidizer. The heat production is constant per mole of 0 2 consumed, 
so the increase in heat capacity must correspond to a decrease in the final temperature of the flame. The 
result is again the dominance of the recombination processes over the branching process. 

The use of the ideas of chemical kinetics can provide a qualitative understanding of 
the rationale behind thermal balance, but the results depend on the fluid mechanical details of the combustion 
and must be determined either from experiments or from models that mimic the system under consideration. 
Tb the extent that there are other loss mechanisms for the radicals needed for the branching reaction, the 
simple ideas developed from the thermal balance will not hold. Thermal balance concepts have been extended 
to include the definition of a limit temperature. That is mixtures having an adiabatic flame temperature below 
some lower limit will not support combustion. These temperatures have been found to be greater than that 
for which the rates of the branching and recombination channels of the H + 0 2 reaction intersect for a wide 
variety of combustion systems. This suggests that additional radical loss mechanisms are operant. 

e. Propagation Rates of Premixed Flames 

An important property for a burning fuel/oxidizer system is its flame speed. This 
property is generally only meaningful for gaseous and liquid fuels for which flammable gaseous mixtures can 


17 




be formed. There are a number of experimental methods available for flame spread measurements [49], but 
all attempt to report the one-dimensional flame propagation rate of the combustion front into the fuel/oxidizer 
mixture. 


Flame speed measurements vary with a number of system parameters. As expected, 
flame speeds depend strongly on the concentrations of the fuel and oxidizer as well as the total pressure. 
Flame speeds are generally found to increase with the temperature of the unburned fuel/oxidizer mixture. In 
turbulent mixtures, flame speeds are proportional to both the turbulence intensity and the laminar flame- 
spread rate [50]. The majority of all flame speed measurements are reported for fuel/air mixtures at ambient 
atmospheric conditions. 


f. Flame Structure 

The term "flame structure" refers to the details of the velocity, temperature, and 
concentration fields in the vicinity of the combustion reaction zone. The structures for both premixed and 
diffusion flames are characterized on the basis of experimental measurements. Usually, well-defined systems 
are investigated in which the combustion region can be assumed to be one-dimensional. Thus far, most 
interest has focused on laminar flames. Interactions of the heat release of combustion with the turbulent 
motion of fluids, along with temporal fluctuations of properties, makes flame structure determinations in 
turbulent flames very difficult. It is significant that the most widely accepted theoretical treatment of turbulent 
combustion-the laminar flamelet approach [51,52]~treats the local flame structure in a turbulent flame as 
being a laminar flame. 


Modeling of combustion and diffusion flames involving full-kinetic treatments of 
major chemical reactions [53] and realistic treatments of molecular diffusion transport [54] allows flame 
structures to be calculated for simple laminar fuel/oxidizer systems. In general, excellent agreement is obtained 
between experimental and calculated results. 


B. INTRODUCTION TO FIRE SUPPRESSION 

This Section is intended to serve three purposes. 

■ First, the current understanding of fire suppression is summarized. The goal is to help the 
reader to understand why and how choices were made concerning the compounds included 
on the final listing. 

■ Second, the types of experiments used to characterize fire suppression effectiveness are 
summarized. TTiis discussion provides the understanding necessary to evaluate past measure¬ 
ments of the fire suppression capabilities for compounds included on the list. 

■ Third, the results of past measurements are used to determine whether (or not) chemical fire 
suppression is a general phenomenon. This is important since many of the chemicals 
included on the final list are intended to test various principles concerning chemical 
mechanisms for flame suppression. Experimental findings strongly suggest that combustion 
suppression can occur by chemical pathways. 


18 



1 . 


Relevant Reviews 


There are numerous literature reviews and critical analyses of fire suppression. These should 
be consulted by any reader interested in developing a basic understanding of the field. Some of these reviews 
are mentioned here as a guide for the reader. 

An excellent review of early work was provided by Fiyburg [55]. This work considers fire 
suppression mechanisms in general, but focuses on the possibility of chemical suppression. A contemporary 
report prepared by Malcolm [56] is not a review article per se, but it does provide an invaluable summary of 
a number of investigations dealing with fire suppression and halogenated suppressants. In 1957, a thorough 
review of flame suppression mechanisms by Friedman and Levy [57] appeared. This review was followed by 
two shorter efforts designed to summarize the literature appearing during interim periods [58,59] and two 
additional short reviews [60,61]. Other authors have contributed to the interim reviews and a continuous 
record can be traced up through 1969 [62,63,64,65,66]. The monograph edited by Gann is an invaluable source 
of information on various aspects of flame suppression and the halogenated fire suppressants in particular [67]. 
A bibliography of halon literature up through 1976 has been prepared by Miller and Kenney [68]. 

More specialized reviews are also relevant. Workers at the U.S. Bureau of Mines have 
provided extensive reviews and critical analysis of flammability data for gases and vapors [69,70], A later 
review dealing with the same subject includes a section on flame inhibition [71]. TVvo authors have provided 
extensive reviews with a focus on the reactions important for flame suppression chemistry [72,73]. 

No reviews of this field have appeared since 1976 because very little new data have been 
reported after this date. Most of these additional studies will be discussed during the following sections. The 
dearth of recent studies can be traced directly to the success of the current commercial halons as fire 
suppressants. Their properties have been so well-suited to their use that there has been little incentive for 
improvement; as a result, research funding for this area had dried up. 

The lack of recent research in fire suppression suggests that it may be possible to make rapid 
progress. During the past 15 years, significant gains have been made in the understanding and modeling of 
combustion systems. These advances are available to researchers and should be directly applicable to the 
problem of flame suppression. 

2. Fire Suppression Mechanisms 

The possible mechanisms of flame suppression can be categorized in various ways. Here we 
follow the discussion of Friedman and Levy [57], but their discussion is typical of other authors. Tvo general 
classes-physical and chemical-for flame extinguishment mechanisms are recognized. 

a. Physical Fire Suppression Mechanisms 

There are 4 types of physical fire suppression mechanisms. Brief introductions to 
each are provided. Reference to the fire triangle discussed earlier in section II. A3 provides immediate insights 
into the physical basis for these processes. 

■ Smothering or Blanketing. This physical mechanism operates by separating 
the fuel and air. Covering a small fire with a blanket to extinguish it is a 
common example. Application of appropriate gases and liquids can also 
smother a fire. Carbon dioxide fire extinguishers are recognized as operating 
primarily by blanketing. 


19 





■ Cooling and Dilution. These processes absorb heat from the combustion 
zone and lower the temperature. As discussed in section II.A4.d, it is 
possible to extinguish a flame when sufficient heat is extracted. There are 
a number of physical means by which heat can be absorbed by agents added 
to a fire. These include simple heating of the agent (proportional to the 
heat capacity), phase changes such as vaporization or sublimation, and 
endothermic decomposition. One might also expect that the rate of heat 
transfer from the combustion zone, as well as the heat capacity, might be an 
important parameter for cooling mechanisms. The physical property which 
characterizes heat transfer behavior is thermal conductivity. The results of 
experiments suggest that the dominant physical property is heat capacity and 
that thermal conductivity plays only a minor role in flame inhibition. 

■ Mechanical Means. This term refers to mechanisms by which the fuel, heat, 
and/or air are forcibly separated. For example, blowing on a match removes 
the heated gases and soot necessary for sustaining combustion from the 
flame zone and extinguishes the flames. 

■ Flame Radiation Blockage. Friedman and Levy [57] also discuss a mecha¬ 
nism by which flame suppression occurs when an agent is interposed 
between the surface of a liquid or solid fuel and the flames which absorbs 
the thermal radiation which is necessary for the generation of gaseous fuel. 
Very little discussion of this process is found in the literature. 

b. Chemical Fire Suppression Mechanism 

This classification refers to an agent, which either by itself or through its 
decomposition products, interferes with the combustion chemistry to a degree sufficient to inhibit (or suppress) 
combustion. Based on the current understanding of combustion as consisting of chain reactions, the most 
effective chemical inhibitors would be expected to be those which remove chain carriers in, or near, the 
reaction zone. 


c. Chemical Fire Suppression and Physical Cooling 

When a chemically active agent is added to a combustion system the agent will have 
both physical and chemical effects. The concept of chemical fire suppression was suggested when agents were 
identified which were more effective fire suppressants than expected based simply on their physical properties. 
Only very recently have efforts been made to quantify the contributions of physical and chemical modes of fire 
suppression for an agent. Such studies are discussed in Section Il.B.4.b. 

The relative effects of cooling and chemical effects on the suppression of a fire are 
difficult to ascertain, in part, because the details of the heat removal processes are not well defined. Consider 
a diffusion flame. Based on existing combustion models it is expected that the location where temperatures 
must be lowered in order to cause flame suppression is in the vicinity of the high-temperature flame zone. 
This zone transfers heat to its immediate surroundings, and the process continues until the temperature 
reaches that for the surroundings. No definitive prescriptions are available for handling the different heat 
transfer processes implied by this discussion. For instance, room temperature heat capacities are sometimes 
used for correlating flame suppression efficiencies, while others use an integrated heat capacity from room 
temperature to some assumed flame temperature. For evaporation or endothermic decomposition, the 
location of the process relative to the combustion region might be expected to be very important. No analyses 
are available which treat this problem. Similar considerations will apply to premixed flames. 


20 



A concept which has been used implicitly in the discussions above requires further 
explanation. Simple endothermic thermal decomposition has been treated as a heat extraction mechanism 
capable of cooling a flame. It is therefore classed as a physical mechanism for flame suppression, despite the 
fact that thermal decomposition is recognized as a chemical process. Thermal decomposition is defined to 
contribute to chemical flame suppression only if it generates a chemical species which interferes with the 
combustion chemistry. Ttoo hypothetical examples serve to illustrate this point. Consider an agent capable 
of undergoing endothermic decomposition to produce two inert molecular species. The flame would be cooled 
by the heat capacity of the agent as it is heated, the heat absorbed during the thermal decomposition of the 
agent, and any additional heat absorbed by the inert species which are generated. Since the species which are 
generated by the thermal decomposition are inert, only a physical mechanism is considered to be active. The 
alternative case occurs when thermal decomposition produces one or more molecular species which are 
reactive and capable of interfering with combustion intermediates and inhibiting the flame chemistry. Both 
physical and chemical mechanisms are active in this case. Heat absorbed from the flame system by the agent, 
the agent’s thermal decomposition, and the decomposition products would be classed as a physical mechanism. 
Only the inhibition of the flame chemistry as the result of the active species generated by the decomposition 
is classed as a chemical mechanism of fire suppression. Fire suppression in the latter case is the result of both 
physical and chemical mechanisms. 

3. Characterizations of An Agent’s Flame Suppression Capabilities 

There are fundamental justifications for investigating flame suppression, but for the purposes 
of this report, the principal objective is to identify agents which are effective as fire suppressants for unwanted 
fires. As a result, the ultimate test of an agent is its ability to extinguish fires which are characteristic of real 
unwanted fires. Note that these will usually be relatively large, turbulent diffusion flames. 

Cost, material, and time limitations prevent testing every possible agent on a full-scale fire. 
As a result, numerous procedures have been developed which are designed to provide guidance to researchers 
in characterizing fire suppression behavior. These tests range from observations of flame structure changes 
in premixed flames to full-scale fire tests. The descriptions which follow provide the necessary background to 
understand the significance of previous tests of fire suppression agents. The methods are listed in a rough 
order such that their relevance to actual fire situations is increasing. Note that measurements in premixed 
flames are directly applicable to situations where the agent is employed to inert fuel-air mixtures. Nonetheless, 
diffusion flame investigations are deemed more relevant for extinguishment of unwanted fires. 

a. Flame Chemistry and Flame Structure of Premixed Flames 

In these experiments, the fuel and oxidizer are premixed and burned in the presence 
and absence of an added agent. It is possible to monitor the combustion products of the reaction and see if 
the agent undergoes a chemical transformation. An example of such an investigation is the work of Burdon 
et al. [74], who found that bromine and hydrogen bromide were produced when hydrogen and carbon 
monoxide were burned in the presence of methyl bromide. 

In a series of benchmark papers, Biordi and coworkers [75,76,77] have reported 
detailed concentration profiles for low-pressure premixed flames of methane, oxygen, and argon in the absence 
and presence of low concentrations of CF 3 Br. These investigations have provided important insights into the 
chemical mechanisms of flame suppression. 

b. Agent Effects on Flame Speed of Premixed Flames 

Flame speed, or burning velocity, is the rate at which the combustion front moves 
into a premixed fuel and oxidizer mixture. Burning velocities are experimentally determined in a variety of 


21 



ways [49]. The velocities observed are sensitive to parameters such as the fuel/air ratio, temperature and 
pressure. The addition of fuel additives also modifies the burning velocity. A decrease in burning velocity 
with the addition of varying amounts of flame suppressants is sometimes employed as a measure of flame 
suppression capability. 


An early example of such an investigation is the work of Rosser et al. [78]. These 
researchers reported the effects of a variety of agents on the flame speed of methane-air flames. Several of 
these compounds contained bromine atoms. A very good correlation was found of the effect of an agent on 
the flame speed and the number of bromine atoms contained in the agent. Ratios of inhibition effectiveness 
for Cl:Br:l were reported as 1:7:7. 

Lask and Wagner [79] reported the effects of a number of additives on hydrogen and 
hydrocarbon flames. Data are reported graphically for flame velocity versus percentage of added agent as well 
as in tables with the percentage of additive required to reduce measured flame speeds by 30 percent. One of 
the most interesting observations was that bromine-containing molecules reduced flame velocities much more 
efficiently than those containing chlorine. Limited measurements on species with iodine suggested that this 
atom was still more effective. This observation was confirmed by Haipern [80]. A comparison of the effects 
of several halogenated hydrocarbons on the flame speed of methane/air mixtures showed that compounds 
containing bromine were considerably more effective than those containing chlorine. 

Homann and Poss [81] measured the flame speed of ethylene/air flames inhibited with 
a series of compounds as a function of pressure. Pressure was varied from 18 kPa to 101 kPa (101 kPa is one 
atmosphere). As expected, flame speeds varied substantially with pressure. However, when data for inhibited 
flames were normalized by the flame speeds for uninhibited flames, the results for different pressures collapsed 
into a single curve. An interesting observation of this work was that CH 3 I was slightly more effective as a 
suppressant than CH 3 Br at both high and low pressures. 

c. Agent Effects on Premixed Flame Flammability Limits and Peak Concentrations 

Flammability limits refer to the relative amounts of fuel and oxidizer in the mixture 
necessary to sustain combustion. These measurements are often made in a device called an explosion burette. 
The mixture to be tested is placed in a long tube, open to the atmosphere with sufficient diameter to ensure 
that the walls have a minimal effect. A spark or small flame is used to ignite the gases. The mixture is 
considered to be flammable when the combustion propagates to the end of the tube and nonflammable when 
propagation does not occur. By varying the concentrations of the gases it is possible to map out the 
flammability curve. Other devices utilized for determining flammability diagrams are discussed by Ford [2]. 

The effect of an agent on combustion behavior is often characterized in terms of the 
lean and rich flammability limits as the concentration of the additive is increased. Usually the concentration 
range of fuel and air capable of supporting combustion decreases as the agent is added. Extensive 
compilations of flammability results are available [69,70]. The work of Moran and Bertschy is also of interest 
since they investigated the flammability limits of pentane/air mixtures for a number of halogenated 
hydrocarbons [82]. 


As the agent concentration is increased, a point is reached where the lean and rich 
limits coincide. The agent concentration at this point is called the flammability peak concentration or, more 
simply, the peak concentration. Farther increases in agent concentration lead to a situation where the fuel 
and air will not support combustion for any fuel to air ratio. Peak concentrations are often used to rate agent 
effectiveness. The most extensive series of measurements have been provided by the Purdue Foundation [25]. 
Results are provided for 56 compounds. 


22 


d. Full Kinetic Modeling of Premixed Flames 

In recent years, it has become possible to model important aspects of laminar flames. 
This does not imply that complete modeling of combustion systems is feasible. Soot formation, for example, 
remains an intractable problem. However, equations which include realistic treatments of mass, momentum, 
and heat transfer, and which incorporate complex kinetic mechanisms to treat the chemical reactions, can be 
solved. Frequently, these calculations are simplified by considering only one- or two-dimensional flames. 

Numerous comparisons have shown that it is possible to accurately calculate laminar 
flame speeds and flame structures for a number of fuel/oxidizer systems. The significance of this advance 
cannot be overstated. It is now possible to make useful predictions concerning practical combustion systems 
on the basis of measured chemical properties (e.g., thermal conductivities and rate constants for individual 
reaction steps). Such understanding offers great possibilities for tailoring and modifying combustion systems 
to meet specific engineering goals. 

A few workers have attempted to utilize these techniques for investigating flame 
inhibition. An early example of such an approach is a paper by Dixon-Lewis [83] in which the structure and 
properties for uninhibited and inhibited hydrogen/air premixed flames were investigated. These authors 
concluded that the primary mechanism of flame inhibition was the catalyzed removal of H atoms to form less 
reactive Br atoms by the reactions 

H + HBr *£ H? + Br 

H + Br 2 HBr + Br. 

Interestingly, the first of these reactions was found to be very close to partial equilibrium (i.e., equal reaction 
rates in forward and reverse directions), but the second was shifted far to the right of equilibrium (as written). 
The second reaction is primarily responsible for removing H atoms which are necessary to sustain combustion. 
As recognized by Day et al. [84], it is necessary to include recombination reactions such as 

Br + Br + M «£ Br 2 + M 

to explain the observed flame suppression effectiveness of Br. Day et al. [84], suggests that the greater flame 
suppression efficiency of Br atoms as compared to Cl atoms results from the higher Br atom concentrations 
possible when H atoms and the hydrogen halide are in equilibrium. 

Westbrook reported similar calculations for organic fuels in a series of three papers 
[85,86,87]. In the first paper, he considered the inhibition of methane and methanol-air flames by HBr [85], 
Calculated flame behaviors were found to be in agreement with available experimental data. The inhibitor 
was predicted to be more effective for lean flames than stoichiometric or rich flames. Flame inhibition was 
primarily the result of the catalytic removal of H atoms via the mechanism discussed above. 

In the second paper [86], the calculations were extended to include inhibition effects 
of HC1, HBr, HI, as well as, halogenated hydrocarbons formed by single substitution of Cl, Br, and I in 
methane, ethane, and ethylene. Preliminary findings for halon 1301, CF 3 Br, were also discussed. The catalytic 
removal of H atoms was found to be the principal inhibition mechanism for each molecule. The only 
additional suppression reactions included beyond those for the hydrogen/air system discussed above were 


23 





H + RX £ HX + R 


X + RX ^ X 2 + R, 

where RX is the halogenated hydrocarbon and X represents the halogen atom. 

The calculations agreed with experimental observations in that compounds containing 
iodine were predicted to be slightly more effective inhibitors than their bromine analogs, while both were 
considerably more effective than the corresponding chlorine compounds. In the case of the halogenated 
hydrocarbons, variations in the fuel content of the parent species led to variations in inhibition efficiency with 
equivalence ratio. Effects of pressure changes were also characterized. 

The final paper in the series by Westbrook [87] includes detailed numerical modeling 
results for CF 3 Br-inhibited flames of hydrogen, methane, methanol, and ethylene. Good agreement between 
computed flame structures and the experimental determinations of Biordi et al. [75,76,77] was found. 
Flammability limits and burning velocities for the laminar flames were predicted and compared with 
experimental findings. The effects of variations in pressure, ambient temperature, and equivalence ratio were 
characterized. 


The calculations described above must be considered as preliminary attempts to model 
fire suppression effectiveness. Our analysis has identified serious deficiencies with regard to some of the rate 
constants utilized in the calculations. On the other hand, these calculations are valuable because they indicate 
that flame modeling can provide important insights into flame suppression mechanisms. State-of-the-art 
calculations of this type are capable of modeling simple diffusion flames. Characterization of suppression 
behavior in flames of this type should be particularly illuminating. 

e. Types of Diffusion Flames Used for Inhibition Investigations 

Tvo types of configurations have been used to investigate the flame structure of 
inhibited diffusion flames. The first is the diffusion flame typical of a flowing fuel entering an ambient or 
coflowing oxidizing atmosphere. We will refer to this configuration as a "standard" diffusion flame. Normally 
these flames are stabilized at the mouth of the burner, and there is a region near the mouth where the flames 
are quenched. For certain fuels and burner configurations stable lifted flames can be created, while for other 
conditions blow out occurs. Note that it is possible to have some fuel and oxidizer mixing in the cool regions 
between the burner outlet and the flame base. 

The second type is known as a counterflow diffusion flame. As the name implies, 
these flames are formed al positions where head-on flows of fuel and oxidizer meet. Two configurations are 
common. One involves flows moving in opposite directions. The second utilizes a porous cylinder placed in 
a flowing oxidizer (or less often, fuel) stream. Fuel (or sometimes oxidizer) is passed through the walls of the 
cylinder and a diffusion flame is stabilized around the cylinder. A stagnation point for the flow exists on the 
upstream side of the cylinder. 

Since counterflow diffusion flames arc located at positions removed from solid 
surfaces they are less subject to heat loss effects and premixing of fuel and oxidizer. For these reasons, 
measurements in counterflow diffusion flames tend to be more reproducible between different facilities. When 
the velocities of the fuel and/or oxidizer are increased, these diffusion flames are eventually quenched. This 
quenching occurs first at the flow-stagnation point in the center of the flame sheet. The flow velocity required 
to "open up" a diffusion flame provides a reproducible measure of diffusion flame extinction known as the 
flame strength. 


24 



f. Flame Chemistry and Structure of Inhibited Diffusion Flames 

Most investigations of changes in flame structure due to the addition of an inhibitor 
have been made for standard diffusion (lames. Several examples are discussed to provide insight into the types 
of measurements and understanding which have been achieved. 

Simmons and Wolfhard [88] used spectroscopic techniques to investigate the flame 
structure of diffusion flames of ethane and carbon monoxide in air, both with and without, methyl bromide 
added to the fuel or oxidizer. A startling change in flame structure was found when the agent was added to 
the air. In the primary combustion zone emission intensities of continuum radiation (indicating increased soot 
concentration) and C^ emission were observed as the inhibitor concentration was increased. Most unusual 
was the formation of a second reaction zone on the oxidizer side of the flame which showed considerable 
emission from excited bromine atoms. It was emphasized that reaction at this secondary location was not 
expected in the absence of the agent. Further tests demonstrated that this reaction zone was the result of the 
addition of the agent, which acted like a fuel. These results provided strong evidence that chemical reactions 
are responsible for flame inhibition. 

Creitz [89] utilized gas chromatography to measure stable chemical species in the 
region outside of the flame zone for a propane diffusion flame. He observed that when CF 3 Br was added to 
the oxidizer, the agent disappeared at positions outside of the flame where temperatures were too low for 
thermal decomposition. This observation is consistent with the observations of Simmons and Wolfhard [88] 
and suggests that chemical species generated by the flame interact with the agent. 

Noda et al. [90] employed electron-spin resonance to measure the relative H atom 
concentrations within a methane Bunsen-type flame. Such a flame burns partly as a premixed flame and partly 
as a diffusion flame. The effects of adding a series of halogen-containing compounds on the H atom 
concentration were reported. For relatively low concentrations of the agents, H atom concentrations were 
decreased by amounts linearly related to the amount of agent added to the secondary air. The decreases were 
quite dramatic. For instance, the addition of 2 percent of methyl bromide to the air decreased the H atom 
concentration by 60 percent. The decreases in H atom concentrations were roughly correlated with inhibition 
efficiency with the amount of decrease increasing in the series F < Cl < Br. 

Relatively few flame structure characterizations of counterflow diffusion flames arc 
available. Ibiricu and Gaydon [91] reported spectroscopic results for ethylene-air and hydrogen-air flames to 
which inhibitors (methyl bromide, bromine, carbon tetrachloride, chlorine, or phosphorous oxychloride) were 
added. Their observations are similar to those of Simmons and Wolfhard [88]. They concluded that the 
inhibitors result in the removal of OH free radicals from the flames. 

A method for stabilizing counterflow diffusion flames for liquid fuels was used by 
Kent and Williams [92] and Seshadri and Williams [93] to investigate flame structures for several organic fuels. 
Gases added to the oxidizer stream included N 2 , C0 2 , He, and CF 3 Br. Some differences in the behavior of 
major species were found when these authors compared their findings to those of Creitz [89], However, the 
rapid disappearance of CF 3 Br as the flame zone is approached was observed in both types of diffusion flames. 

Temperature measurements have also been reported for counterflow diffusion flames 
[92,93]. It was concluded that flame temperatures for conditions close to extinction are much higher when 
CF 3 Br is added to the oxidizer than for nitrogen dilution. This is taken as support for a chemical mechanism 
for CF 3 Br inhibition since N 2 is believed to inhibit flames only as the result of heat capacity effects. 


25 



g. Extinction of laminar Diffusion Flames 

The relative effectiveness of agents can be quantified by comparing the amounts which 
must be added to either the fuel or oxidizer in order to extinguish a diffusion flame. Both standard and 
counterflow diffusion flames have been widely used for such studies. 

Early investigations involving standard diffusion flames are those of Simmons and 
Wolfhard [88] and Creitz [94]. Simmons and Wolfhard reported the percentages of methyl bromide which 
must be added to the fuel or air to extinguish flames of four organic fuels and hydrogen. In all cases the 
percentage of agent which had to be added to the fuel was much greater than required for the oxidizer. 
Extinction percentages were compared with measurements for premixed flames utilizing both the peak agent 
concentration and the percentage required for a stoichiometric mixture. The percentage of agent required for 
the oxidizer varied in similar ways as a function of fuel and, while larger for the diffusion flames, were of 
comparable magnitude to those necessary to prevent flame propagation in stoichiometric premixed flames. 
This is consistent with the view that the flame sheet in diffusion flames lies near positions of stoichiometric 
mixing. 

Creitz reported similar measurements for a variety of fuels utilizing nitrogen, methyl 
bromide, and halon 1301 as inhibitors [94]. In this investigation, the concentration of oxygen in the 
surrounding atmosphere was also varied. As found by Simmons and Wolfhard [88], a considerably higher 
percentage of agent was required for extinction when added to the fuel than when added to the oxidizer. 
Creitz concluded that the mechanism of inhibition is different for the two cases. It is not clear whether he 
treated his data in terms of concentrations for stoichiometric combustion. One interesting conclusion was that 
for nitrogen/oxygen mixtures containing more than 25 percent oxygen, methyl bromide became a fuel and was 
entirely ineffective as a suppression agent. This compound burned as a diffusion flame in 32 percent oxygen. 

Early measurements of percentages of inhibitor required to extinguish laminar 
diffusion flames were for gaseous fuels only. The results had a great deal of scatter for given fuel and inhibitor 
combinations [95,96]. This scatter was most likely due to variations between experimental apparatus and 
techniques. In response to this situation, Bajpai [96] developed an apparatus for determining the 
concentration of agents required to extinguish small pool fires of liquids. The device has come to be known 
as a "cup burner" after the shape of the vessel holding the liquid fuel. A coflow of oxidizer flows by the cup. 
In Bajpai’s device the fuel surface had a diameter or 25.4 mm. Hirst and Booth have described a very similar 
apparatus having a cup diameter of 28 mm [97]. It is unclear if the flame will be strictly laminar for these 
conditions, but it is also true that the flame is unlikely to be fully turbulent. These devices will be discussed 
in greater detail because of their widespread use for testing suppression agents. 

The percentage of agent required to extinguish a flame has been found to be a 
function of the oxidizer flow velocity, first increasing and then falling with increasing velocity [96,97]. Bajpai 
reported values of the peak percentages of halons 1301 and 1211 in air required to extinguish a wide variety 
of liquid fuels [96]. He found that his results agreed with the then-unpublished measurements made with a 
similar apparatus at Imperial Chemical Industries [97], Hirst and Booth reported results for halons 2402, 
1202, 1011, and 113 as well as halons 1301 and 1211. Bajpai [96] also compared his cup burner results with 
the findings of other researchers who measured percentages of agents required for extinguishment of much 
larger pool fires in enclosures [98,99,100]. Generally, peak percentages of agents for the two types of 
experiments were comparable, but values for the large-scale fires were smaller. Hirst and Booth made similar 
observations [97]. v 

Sheinson and coworkers have reported peak percentages for a number of agents 
utilizing heptane and 2-propanol as fuels [45,46], The results of these measurements are discussed in more 
detail in Section II-B-4-b. 


26 



Petrella and Sellers [101] had previously used a similar apparatus to investigate the 
extinction of propane diffusion flames. The gaseous fuel flowed from a round tube into the oxidizer coflow. 
Experimental results were reported in terms of the lowest oxygen index (LOI). This is simply the lowest 
oxygen concentration in air diluted with the agent which allows combustion to occur. Values of LOI were 
shown to be nearly independent of propane flow rate and to have only a mild dependence on the oxidizer flow 
rate. LOI values are reported for halons 1001, 1011, 1211, 1202, 1301, and 2402 as well as nitrogen. All of 
the brominated species gave LOI values of approximately 0.205 as compared to that for nitrogen of roughly 
0.16. 


Measurements have also been reported for percentage of agents in fuel or air required 
to extinguish opposed flow diffusion flames. One of the first investigations of this type is the study of 
Friedman and Levy [102]. These authors report the concentrations of various gaseous agents which must be 
added to the methane to open a hole in the flame sheet formed between counterflow jets of fuel and air. The 
effectiveness of various agents was ranked in the order CH 3 C1 < CC1 4 < CH^Br < CFjBr, which is the same 
ordering reported by Rosser et al. [78] for the inhibition effects on flame speeds of premixed methane/oxidizer 
mixtures. 


Milne et al. [103] used the counterflow geometry of Tkuji and Yamaoka [104] to 
investigate inhibition. In this configuration, fuel is supplied by a porous cylinder and the oxidizer flows by the 
cylinder. Extinction occurs at the stagnation point in front of the cylinder. Measurements are reported for 
agents added to both fuel and air for methane, propane and butane. Agents tested were CF 3 Br, CF 2 ClBr, 
CH 3 Br, C0 2 , Ar, and N 2 . The ordering of agent effectiveness was consistent with measurements by other 
techniques. 


Kent and Williams [92] and Sheshadri and Williams [93] report the amounts of CF 3 Br 
and nitrogen in the oxidizer which arc required to extinguish counterflow flames for liquid fuels. 

h. Extinction of Turbulent Diffusion Flames 

Ifests of the ability of agents to extinguish turbulent diffusion flames are expected to 
most closely resemble the agents’ use in the field to fight fires. At this point it is necessary to distinguish two 
common uses of halons: total flooding and local application. Tbtal flooding is employed for firefighting in 
enclosures. The aim here is to create an environment in which the fire cannot exist by adding the agent to 
the entire atmosphere in the room at a concentration sufficient to inhibit combustion. Usually the enclosure 
is designed to be relatively airtight so that the agent concentration can be maintained above a specified level 
for several tens of minutes. Gaseous agents are employed for total flooding. Local application involves 
spraying the agent directly on, or near, the fire in order to induce extinguishment. Since liquids are more 
easily applied, agents used for this purpose are usually liquids which evaporate near room temperature or gases 
which condense just below room temperature. 

Ifests have been developed to assess agent effectiveness in both types of applications. 
These tests are usually less standardized than the tests previously described and utilize fire systems which are 
subject to a great deal of variability. As a consequence, these results show more variation so that 
reproducibility between different laboratories tends to be poor. Since they are the final step in the 
development of firefighting agents, many large-scale tests have been performed. Only a few are mentioned 
to provide a flavor of the types of studies which have been carried out. 

Ford [2] discusses a number of large-scale fire tests which have been utilized to test 
the effectiveness of halon 1301. Results for both class A and class B fires are described. Three published 
examples of such tests were mentioned earlier [98,99,100], The results of McKee et al. [105] and Alvares [106] 
for the extinguishment of pool and spray fire by halon 1301 and halon 1211 are also of interest. Investigations 


27 




have shown that deep-seated smoldering fires in solid fuels often require considerably higher agent 
concentrations for extinguishment [28,107,108], 

Fire tests involving direct application to fires in the open are less numerous. Ford 
[2] again provides a short review of previous tests. Malcolm describes some very early tests in which various 
halons were tested for their ability to extinguish Class B fires and cotton waste saturated with gasoline [6]. 
An editor’s note in this paper (see Ihble XV, p. 128) points out many of the difficulties associated with this 
type of testing. Breen [109] has investigated the effectiveness of various mixtures of halons 1211 and 1301 for 
extinguishing heptane fires. A slight dependence of fire suppression efficiency on mixture fraction beyond a 
simple summation was identified. The author attributed this observation to differences in thermophysical 
properties between the two agents. Problems of reproducibility for this type of measurement are discussed. 

The results of practical studies are usually published as internal reports and are 
sometimes difficult to access. An example of such work is a report which appeared as a Wright Air 
Development Division technical report [110]. Results for tests on a number of different size pool fires are 
discussed. Halons 1011,1211,2402,1202, and 1301 were tested. Difficulties with reproducibility are discussed. 

4. Agent Effectiveness, Testing Methods, and the Role of Different Suppression 
Mechanisms 

a. Agent Effectiveness and Testing Methods 

Many of the test methods which have been used to characterize the flame suppression 
capability of agents were discussed in the previous section. It is important to consider whether these tests are 
true predictors for fire suppression efficiency when the agents are utilized in real fire situations. 

The effectiveness of an agent utilized as a total flooding agent is usually well 
correlated with its behavior in any of the several small-scale tests (e.g., explosion burette or cup burner). This 
is not surprising since cooling, dilution, and chemical flame suppression mechanisms (see the discussion in 
Sections II.B.2.a and .b) are expected to be most important in total flooding applications. In contrast, the 
performance of local application-agents does not correlate nearly as well with small-scale tests. This is due 
to the fact that these tests are insensitive to certain physical effects, such as evaporative cooling, smothering 
and mechanical blow off, which contribute to the effectiveness of streaming agents. Only test methods 
involving the direct application of these agents to fires (as liquids) will be able to account for these additional 
physical mechanisms. 


b. Contributions of Physical and Chemical Mechanisms to Fire Suppression 

Effectiveness 

Thus far, little mention has been made of the relative contributions of the various 
physical and chemical mechanisms to the flame suppression capability of an agent. This point is discussed here 
with reference to the published literature and the previous discussion. 

As noted in the last section, very few of the tests used to characterize agents for fire 
suppression effectiveness screen for smothering or mechanical suppression. Perhaps as a result of this, the 
understanding of these processes remains very poor. No literature has been identified which attempts to 
quantify the contributions of these processes to flame suppression. The possibility of flame radiation blockage 
is even less well understood. These physical mechanisms will not be discussed further. 

The majority of tests utilized for agent suppression efficiency characterization are 
expected to be sensitive to the physical processes of cooling and dilution and to any chemical mechanisms for 


28 



flame suppression. Even for these mechanisms, the relative contributions of each to total agent suppression 
effectiveness are not well characterized. 

Early studies focused on the role of inert agents in the extinction of flames. These 
studies are important since these gases are believed to play no chemical role in inhibiting combustion. As a 
result, their effectiveness can only be due to cooling and dilution. Therefore, these gases provide examples 
of flame suppression behavior where the mechanism can be characterized as well known. 

Coward and Hartwell [111] reported on the extinction of methane flames by carbon 
dioxide, nitrogen, helium, and argon. These workers showed that the "extinctive action" of CO z , Ar, and N 2 
were nearly in the inverse ratio of their heat capacities and concluded that heat removal from the reaction 
zone was the principal mechanism of flame inhibition. Helium was found to be more effective than predicted, 
based on heat capacity effects alone. This was attributed to the high thermal conductivity of this low 
molecular weight gas and suggests that thermal heat transfer plays a minor, but detectable, role in an agent’s 
effectiveness. 


In 1925 White [112] suggested that flame temperatures for a given fuel-air mixture 
were nearly constant at the lower flammability limit and were independent of the initial temperature of the 
mixture. The limiting temperature did vary from fuel to fuel. An experimental investigation by Simmons and 
Wolfhard [113] for a number of fuel-air mixtures diluted by nitrogen supports this conclusion. Many authors 
have applied the idea of a well-defined limit temperature to the understanding of the effects of inert gases on 
the flammability limits of fuel-oxidizer mixtures. 

Egerton and Powling [114] calculated adiabatic flame temperatures for hydrogen and 
methane flames at their lower flammability limits when the nitrogen in air was replaced with C0 2 , Ar, and 
He. With the exception of added helium, the calculated flame temperatures were roughly the same. The 
limiting temperature with helium was higher which is consistent with the findings of Coward and Hartwell 
[111] concerning the importance of the high thermal diffusivity of this light gas. 

Flame temperatures for methane flames at their fuel and oxidizer extinction limits 
were measured by Ishizuka and 'Buji [115] in a counterflow diffusion flame. The oxidizer used for the 
experiments was oxygen mixed with N 2 , Ar, or He. For N 2 and Ar, the flame temperatures were very close 
(± 30 K) to 1500 K. In the cases where the oxidizer contained helium, flame temperatures were roughly 1640 
K. The slightly higher temperatures which were observed for helium mixtures are consistent with previous 
conclusions concerning the limited role thermal conductivity plays in the determination of extinction limits. 

Ewing et al. [116] have also considered the effects of inert diluents on temperatures 
at flame limits. They calculated adiabatic flame temperatures and concluded that the limit temperatures were 
functions of the molecular weight of the inert extinguishants. Both diffusion flames of heptane in air and 
premixed flames of methane/air were considered. An explanation was not provided as to why the limit flame 
temperature behavior they find differs from that reported by other authors. 

There is widespread agreement that the fire suppression effectiveness of certain agents 
is beyond what would be expected on the basis of their ability to extract heat. Much of the literature discussed 
in Section II.B is based on the hypothesis that chemical mechanisms are active. Fryburg [55J provides a good 
review of the justifications for a chemical mechanism based on work published prior to 1950. The following 
paragraph quoted from this work is typical of the literature. 

"The importance of the chemical actions are indicated by the greatly enhanced 
extinguishing effectiveness of the gaseous and liquid halogen-containing compounds, as determined in 
inflammable-limits studies. These compounds have larger specific heats than the inert gases and the cooling 
action would therefore be increased, but the increase is not large enough to account for the much greater 


29 




effectiveness of many of the compounds investigated by Jorissen." (See references [117,118,119].) "In addition, 
there is no relationship between extinguishing effectiveness and specific heat among the halogen compounds. 
This effect therefore must be derived from strong chemical actions." 

Additional evidence for a chemical mechanism of flame suppression is available in 
results for the effect of suppressants on the flame temperature at limit concentrations. It has been shown 
earlier that physical suppression does not change limit temperatures significantly. Burdon et al. [74] 
investigated the effect of methyl bromide on H 2 -air and CO-air flames. As part of the investigation adiabatic 
flame temperatures were calculated for limit mixtures. These limit temperatures were found to increase 
substantially as the methyl bromide concentration was raised and the authors concluded that the temperature 
rise "may be attributed to the chemical intervention of methyl bromide in the flame reactions" [74]. 

Simmons and Wolfhard [88] reached an identical conclusion based on measurements 
on methane-air mixtures inhibited with bromine and methyl bromide. Simmons and Wright [120] considered 
the system of propane-air-hydrogen bromide. Adiabatic flame temperatures increased from 1590 K at the lean 
limit with no hydrogen bromide to 2100 K for a mixture containing a concentration of HBr which was just 
short of the peak percentage. 

As part of a study which argues that extinction of hydrocarbon flames can be 
understood based solely on heat-absorption processes (even for the "chemical" suppressants), Ewing et al. [116] 
found that for certain classes of suppressants the calculated limit flame temperature was raised considerably 
above that observed for inert suppressants. Even though it is not the aim of the authors, this observation 
provides additional support for a chemical mechanism. 

Section II.A.4.d noted that, for flames in which only heat loss mechanisms are 
important, roughly 210 J/K per mole of 0 2 are required to suppress a flame. Tbcker et al. [47] used this idea 
to analyze the effects of inert diluents and halon 1301 on various fuel-oxidizer combinations. Based on the 
nitrogen results, predictions were made for the heat capacity contributions required from CO,, He, and CF 3 Br. 
The experimental amount of C0 2 was predicted within 10 percent, but the amounts of He and CF 3 Br required 
were greatly overpredicted. We have seen previously that helium gas results in higher heat losses from the 
combustion zone due to its unusually high thermal conductivity and therefore more easily extinguishes flames 
than heavier molecules. Tbcker et al. [47] concluded that the overprediction of the amount of CF 3 Br required 
is due to a chemical effect of this halon on the flame chemistry. 

Using a cup burner test with n-heptane as the fuel Sheinson et al. [45,46] have 
determined the amounts of various agents required for extinguishment. They have analyzed their results in 
terms of the heat capacity of the flame gases per mole of oxygen. Inert agents arc found to require roughly 
210 J/K for extinguishment. Certain agents such as CF 3 Br and CF 3 I require considerably less heat capacity 
for the gas mixture. This decrease in the required heat capacity is attributed to a chemical mechanism for 
flame suppression. 


As we have seen, chemical mechanisms of flame suppression have been hypothesized 
and tested for many decades. Despite this, until recently there have been no attempts to quantify the relative 
contributions of physical and chemical mechanisms to the flame extinguishment process. The first attempts, 
of which we are aware, are those of Larsen [121,122,123] and Ewing et al. [116]. These authors reached the 
somewhat startling conclusion that flame suppression is the result solely of a physical mechanism. This was 
argued to be true even for the so-called chemical suppressants. 

In the earliest of these papers [121] Larsen found that the effectiveness of halogen 
atoms for fire suppression was directly proportional to their atomic weights. It was argued that this supported 
a physical mechanism for their extinguishment capabilities, but no hypothesis was made as to why fire 
suppression effectiveness should scale with atomic weight. It is very likely that some other physical property 


30 



which scales with atomic weight (e.g., hydrogen halide bond energies decrease with increasing halogen 
molecular weight) is the source of the observed correlation. 

In another paper [122], it was demonstrated that the flame suppression capabilities 
of inert agents could be understood in terms of heat capacity effects. Larsen argued that a similar analysis of 
peak concentration data for a number of halons suggested that the effectiveness of these agents was also due 
to exclusively heat capacity [123]. Instead of calculating the total heat capacity of the gases at the flame limit, 
Larsen reported values for three ratios (K 10 , K 1F , KC p ) of heat capacities. K I0 is the ratio of the heat 
capacity of the agent to the total heat capacity of a mixture of oxygen and inert which is chosen such that, 
when the fuel is added, the mixture passes' through the peak concentration. K 1F is a similar ratio for a fuel- 
agent mixture. KC p is the ratio of the total heat capacities of the inerts and the total heat capacity for all 
gases. All heat capacities are for a temperature of 298 K. Values of K JO and K 1F were found to be roughly 
constant. Larsen argued that this observation supports his contention that only heat capacity effects are 
important for flame suppression effectiveness. 

The monograph [123] where Larsen’s paper appeared includes comments from the 
participants of the conference [124]. Many of them questioned the analysis and conclusions which Larsen 
reached. Perhaps the most important critique was offered by J. W. Dehn. In his comment, Dehn noted that 
the ratios of heat capacities employed were determined primarily by the heat capacities of nitrogen and air. 
Variations in other additives would have minor effects on their values. Gmsequently, these ratios would not 
be expected to be very sensitive indicators of fire suppression efficiency. 

Sheinson et al. [45,46] and Ewing et al. [116] have performed analyses based on the 
total heat extraction capabilities of combustion product gases which seem to be more realistic than those 
employed by Larsen [122,123]. Sheinson et al. calculated the heat capacities available per mole of 0 2 for n- 
heplane diffusion flames burning at their limit in air/agent mixtures. Seven agents which are expected to be 
inert in the flame gave values covering a range of 180-205 J/K(mole-0 2 ). The lowest value was for helium 
which is expected to have a lower value due to its high thermal conductivity. Values for agents which are 
expected to take part in the chemistry of the flame (CF 3 Br and SF s Br) required only on the order of 155 
J/K(mole-0 2 ). This is taken as strong evidence for a chemical mechanism of flame suppression [45,46]. 

A different type of analysis was provided by Ewing et al. [116]. These authors 
calculated the heat extraction behaviors for a number of fuel/air combustion systems in the presence of various 
agents (including solids) and made the assumption that extinction occurs for well-defined flame temperatures. 
In this way they were able to correlate the findings for a wide variety of suppressants added to flames. As 
discussed above, inert species gave limit temperatures which varied little and are relatively low. This behavior 
was not observed for those agents which are traditionally considered to be chemically active [116]. The total 
heat extraction capability of these agents was too low to be the sole source of extinction. Note that this 
conclusion is at variance with Larsen’s [123] contention that only heat capacity effects need be considered. 
Ewing et al. noted that agents such as CF 3 Br can absorb additional heat by decomposing. Even though 
decomposition is a chemical process, this is classified as physical extinction since there is no direct interaction 
with the combustion chemistry (this viewpoint is consistent with that discussed in Section II.B.2.C). Better 
agreement was obtained when these additional heat extraction mechanisms were taken into account. The 
major conclusion of this work was that "for most substances the extinguishing capacity is related to heat- 
extraction and that many of the effects previously attributed to chemical mechanisms may be thermodynamic 
in nature rather than kinetic." 

Interestingly, these authors had to assume that the limit temperature varied for 
different types of agents [116]. For agents which in the past have been classed as chemically active, including 
some solids, the limit temperatures were considerably higher than for inert gaseous species. As discussed 
above, such changes in the limit temperature can be taken as evidence for a chemical mechanism. This work 
is important because it emphasizes that all mechanisms for heat removal must be considered in assessing the 


31 




physical effects of an agent. Furthermore, it provides a prescription for accomplishing this. On the other hand, 
the conclusion that only physical processes are responsible for extinguishment is not consistent with the 
increases in limit temperatures which are required in order to obtain correlations of flame suppression 
effectiveness for different chemical families. 

Sheinson et al. [46,125] incorporated some of the ideas of Ewing et al. [116] into a 
model which attempts to quantify the roles of physical and chemical mechanisms on agent effectiveness. Their 
model assumes that physical and chemical effects are additive. Physical contributions are determined by 
calculating the heat which can be extracted from the flame by the heat capacity and chemical decomposition 
of the agent. As discussed above, a total of roughly 210 J/K(mole-0 2 ) are required to inert a mixture. 

Cup burner measurements of extinguishing agent concentrations for an n-heptane 
pool fire were used for the analysis by Sheinson et al. If the mixtures required less than 210 J/K(moles-0 2 ) 
the difference was attributed to chemical suppression effects. The assumption of a linear relationship allowed 
the relative fractions of physical and chemical suppression to be reported. Interestingly, two agents tested, 
SF S C1 and S 2 F 10 , seemed to enhance the fire. In agreement with past investigations, it was concluded that 
CF 3 C1, CFjBr, CF 3 I, and SF 5 Br act as chemical flame suppressants. As an example, these authors conclude 
that the suppressant effects of CF 2 Br are 20 percent physical and 80 percent chemical [46,125]. 

The analysis of Sheinson et al. [46,125] is recent, and considerably more development 
and testing are required. However, the work does provide the first steps toward quantifying the contributions 
of physical and chemical mechanisms to flame suppression. If ideas of this type can be further developed, it 
may become possible to tailor molecules to maximize physical and chemical contributions to flame suppression 
behavior. 


c. Conclusions Regarding a Chemical Mechanism for Fire Suppression 

There is widespread agreement that combustion reactions can be interrupted by the 
application of certain fire suppression agents. Most researchers conclude that interference with the flame 
chemistry increases the effectiveness of such agents significantly as compared to that expected based on heat 
extraction effects alone. The modern understanding of combustion chemistry and kinetics identifies the most 
likely chemical mechanism for fire suppression as an interruption of the important free radical chain reactions 
responsible for maintaining the high heat release rate. The most likely species for attack by a chemical agent 
or its products is the H atom. 

The fact that some uncertainty remains concerning the chemical mechanism of flame 
suppression can be traced to the relatively poor characterization of the process. Recent attempts have been 
made to quantify the relative contributions of physical and chemical mechanisms to the effectiveness of fire 
suppression agents. The tools available to modern combustion scientists should allow these uncertainties to 
be resolved in the near future. 

Based on the extensive literature review discussed above, we have concluded that 
chemical fire suppression does occur and that chemical mechanisms contribute significantly to the fire 
suppression capabilities of many agents. For this reason, the ability to develop principles of chemical fire 
suppression has played a major role in selecting compounds for the exploratory list. 


32 



c. 


HALON STABILITY AND DEPLETION OF STRATOSPHERIC OZONE 


1. Stability and Properties of Halons 

As discussed above, the current commercial halons used for fire suppression have many 
desirable properties. In almost every case, these favorable properties can be traced to the unusual chemical 
stability of these molecules. In many systems, including the atmosphere and living organisms, the species most 
likely to be responsible for chemical attack on a given molecule is the highly reactive hydroxyl free radical, 
OH. Due to the relative strength of carbon-halogen bonds compared to the relatively low bond energies of 
possible products, OH reactions with fully halogenated species are slow. In contrast, for molecules containing 
carbon-hydrogen bonds, OH free radicals can abstract the hydrogen atom rapidly to create a new carbon-based 
free radical and a water molecule. The resulting free radical reacts rapidly with 0 2 and is ultimately destroyed. 


The lack of significant reactions and the low polarity of the current commercial halons mean 
these molecules are nonconducting and thus safe for use on electrical equipment. The nonreactivity of the 
molecules is the principal reason why the species have low toxicity. The normal processes which cause toxic 
responses in organisms do not take place. The low reactivity of totally halogenated species also means that 
they do not react easily with other materials and, therefore, have good compatibility properties. 

In combustion, molecules can be destroyed by unimolecular decomposition or attack by H and 
OH radicals. Commercial halons such as CF^Br (halon 1301) are unreactive towards OH. They tend to be 
about as reactive towards H atoms as are more typical fuel molecules (via H + CF 3 Br -+ CF 3 + HBr), but 
are more likely to undergo unimolecular decomposition (to CF 3 + Br). These reactions release Br and HBr 
which are thought to play important roles in the mechanism of chemical fire suppression. The subsidiary 
question, which will be addressed in the course of testing the candidate compounds, is whether the details of 
the mechanism and the rate of halogen release affect fire suppression activity. This can be addressed through 
experimental work coupled with chemical kinetic modeling. It may well be that there is only a narrow 
"window" for effective chemical suppressant activity. If this can be established, it will open the way to 
designing molecules for specific fire suppression situations. 

2. Brief Introduction to Atmospheric Structure and Chemistry 

The earth’s atmosphere is a dynamic and complicated physical system. In this section, we 
introduce some very simple descriptions of the structure and chemistry of two regions of the atmosphere, the 
troposphere and the stratosphere, to provide a background for the following discussion on depletion of 
stratospheric ozone by the current commercial halons. Details can be found in a number of textbooks (e.g., 
[126,127]). A report describing the outcome of the Alternative Fluorocarbon Environmental Acceptability 
Study (AFEAS) [128] is also relevant. It contains an excellent discussion of these topics including an 
examination of compounds designed to replace the widely-employed chlorofluorocarbons (a subclass of halons). 
Dr. John Herron discusses many of the concepts of atmospheric chemistry related to the current commercial 
halons in the companion report "Preliminary Screening Procedures and Criteria for Replacements for halons 
1211 and 1301 [129]." 

The troposphere is the region of the atmosphere extending from the earth’s surface to a 
height denoted as the tropopause. The tropopause separates the troposphere and the stratosphere. The 
height of the tropopause varies with location and season, but is on the order of 11 km [126]. The stratosphere 
lies above the tropopause and extends to a height of roughly 50 km where the stratopause is located. The next 
region of the atmosphere is known as the mesosphere. 


33 


The troposphere is a region of rapid air movements and intense mixing. As a result, it is 
reasonably uniform; and chemical species released here are dispersed and mixed within the layer quite rapidly 
(on the order of a few weeks). Air pollution in urban areas is the result of chemical processes occurring in 
this region of the atmosphere. The chemistry is very complex [127] and no attempt is made here to discuss 
it. It is worthwhile to reiterate that hydroxyl free radicals are by far the most important species responsible 
for chemical attack in the troposphere. 

There are a number of processes by which a molecule can be removed from the troposphere: 

■ Reactive molecules undergo chemical transformation in the troposphere and in most 
cases eventual removal from the atmosphere. 

■ The troposphere is subject to sunlight having wavelengths greater than 290 nm. If 
a molecule absorbs at these wavelengths, it is possible that photolysis may occur. 
Photolysis creates reactive species and eventually leads to the removal of the 
molecule from the troposphere. 

■ If the molecule is water-soluble or capable of hydrolysis, it can interact with moisture 
in the troposphere and eventually "rain out." 

■ If the molecule survives long enough it can eventually diffuse into the stratosphere. 

Temperatures generally fall with height in the troposphere. This trend is reversed in the 
stratosphere where temperatures begin to increase with altitude. This temperature rise occurs as the result 
of absorption of the sun’s radiant energy at wavelengths shorter than 300 nm by ozone, 0 3 [126]. As a result 
of the change in temperature behavior with height, mixing across the tropopause is very slow. Chemical 
species released in the troposphere require several years to diffuse into the stratosphere. 

Ozone is formed in the stratosphere as the result of photolysis of oxygen by sunlight to create 
oxygen atoms followed by recombination of the atoms with oxygen molecules to form ozone. These reaction 
processes can be written as: 

0 2 + hv -» 20 

O + 0 2 + M —* 0 3 + M, 

where hv represents a photon having a wavelength shorter than 220 nm and M is a third body required to 
remove excess thermal energy. 

There are also chemical processes taking place in the stratosphere which remove ozone 
molecules. The equilibrium concentration of ozone in the stratosphere is the result of a balance between these 
destruction and formation processes. 0 3 is found throughout the stratosphere with a maximum concentration 
at a height of approximately 25 km. 

3. Current Commercial Halons and Depletion of Stratospheric Ozone 

In section I.B.l.b it was pointed out that the current commercial halons are generally 
recognized as contributing to the depletion of ozone in the stratosphere. It is now possible to discuss the 
reasons for this conclusion in more detail. 


34 



Current commercial halons have been, and continue to be, released into the atmosphere by 
a variety of mechanisms. These include discharges during firefighting, training, testing of total flooding systems 
(efforts are underway to limit this activity), and leaks [7], In the vast majority of cases these releases take 
place in the troposphere. The current commercial halons have very long tropospheric lifetimes due to their 
unusual chemical stability. As a result, they eventually diffuse to the stratosphere. As the molecules rise in 
the stratosphere they are exposed to shorter wavelength light. Eventually, a height is reached where the 
current commercial halons begin to absorb this higher energy light and undergo photolysis which releases 
bromine atoms. The bromine atoms are highly reactive and initiate a catalytic chain reaction which converts 
ozone to oxygen molecules in the following manner: 

Br + O 3 —* BrO +0 2 
BrO + O — Br + 

Net: 0 3 + O -* 20,. 

As a consequence of this mechanism, the rate of destruction of ozone is increased. This leads to a reduction 
of the equilibrium concentration of ozone in the stratosphere. Once introduced into the stratosphere, bromine 
atoms have very long lifetimes; and the destruction of ozone can continue for decades. Chlorine and iodine 
atoms also catalytically destroy ozone in the stratosphere. Fluorine atoms are recognized as being relatively 
unimportant, since effective mechanisms are available for chain termination. 

A widely used concept to characterize the ability of a specific type of molecule to catalytically 
destroy ozone is the ozone depletion potential (ODP). Several parameters must be considered in deriving this 
number. These include: tropospheric lifetime, diffusion time from the troposphere to the stratosphere, 
effectiveness in destroying ozone, and possible sinks (removal processes) for the reactive species created from 
the molecule in the stratosphere. Such calculations are subject to a great degree of uncertainty and are thus 
difficult to quantify. As a consequence, values are usually reported relative to the results for CFC-11, CC1 3 F, 
as first suggested by Wuebbles [130], This ratio has come to be known as the ODP value. As noted above, 
ODP values for two of the current commercial halons arc estimated to be 2.7 (halon 1211) and 11.4 (halon 
1301). 


As discussed in Section I.B.l.b reductions of stratospheric ozone have been observed which 
are attributed to the release of manmade halons (both current commercial halons and CFCs) into the 
stratosphere. The production of these chemicals has been limited by the Montreal Protocol. It is ironic that 
the chemical stability of the current commercial halons, which makes them so useful for fire suppression, is 
responsible for their ability to destroy stratospheric ozone, and will ultimately limit their use. 

4. Means for Reducing ODP Values of Effective Firefighting Agents 

The use of current commercial halons will be strictly limited or curtailed entirely in the future. 
The challenge is to find replacements which are effective and safe as fire suppression agents and which are less 
harmful to stratospheric ozone than the current commercial halons. In this section possible means for 
identifying fire suppression agents which are safe for stratospheric ozone are discussed. 

The present understanding suggests that there are two general mechanisms for limiting the 
destruction of stratospheric ozone by molecules which are effective as fire suppressants. The first is to employ 
agents which contain no chlorine, bromine, or iodine. This requires that agents which are recognized as being 
the most effective fire suppression agents, including the current commercial halons, cannot be employed. The 
second approach is to use agents which are destroyed over short time scales within the troposphere and thus 
do not have time to diffuse to the stratosphere. 


35 


For the first approach it is necessary to utilize only perfluorinaled organics or other classes 
of molecules which do not contain Cl, Br, or I. This approach severely limits the number of possible agents 
and removes large classes of molecules which have been shown to be effective fire suppression agents. 

The second approach, employing molecules which are effectively destroyed in the troposphere, 
might allow some molecules containing the heavier halogens to be used. The discussion in Section II.C.2 
provides insights as to how halons might be destroyed in the troposphere. These destruction mechanisms will 
be discussed by considering the possibility of modifying one of the current commercial halons to dramatically 
reduce its lifetime in the troposphere. Keep in mind that the discussion is general, and similar considerations 
will apply to the entire halon family. 

Possible modifications to current commercial halons for reducing tropospheric lifetimes 

include: 


■ providing a site for OH free radical attack, 

■ extending the absorption region of the molecule toward the red so that solar 
wavelengths reaching the troposphere can photolyze the molecule, and 

■ increasing the solubility or susceptibility to hydrolysis so that the molecule will rain 
out of the atmosphere. 

As examples of these concepts, consider some possible molecular modifications for halon 
1301, Keep in mind that these modifications are provided only to demonstrate principles. No claim is made 
that these particular molecules will be appropriate alternatives for halon 1301. The susceptibility of halon 
1301 to chemical attack by OH can be increased by substituting an H atom for one of the fluorines to give 
CF 2 HBr (halon 1201). The lifetime of this molecule in the troposphere will be reduced compared to that for 
CF 3 Br. A second approach is to red-shift the absorption spectrum of the CF 3 Br. One means of doing this 
is to replace the bromine atom with iodine-yielding CF 3 I (halon 1301). CF 3 I should be an effective fire 
suppression agent. Its absorption spectrum is red-shifted from that for CF 3 Br which should increase the 
probability that it will photolyze in the troposphere. Since the solubility of a molecule in water and its 
susceptibility to hydrolysis both increase with molecular polarity, it is possible that tropospheric lifetimes might 
be decreased by increasing the polarity of CF 3 Br. For example, halon 1201 is more polar than halon 1301 and 
would, therefore, be expected to have a shorter tropospheric lifetime. 

Modifications discussed here are designed to increase the reactivity of the molecular species. 
Such changes are also likely to result in variations in physical properties and characteristics (e.g., fire 
suppression effectiveness, toxicity, and materials compatibility) which make the molecule less likely to be useful 
as a fire suppressant. It is certain that a careful balancing of properties will be required to provide an effective 
and safe fire suppression agent which is also less harmful to stratospheric-ozone than current commercial 
halons. This is the primary challenge to the fire protection community. 


36 



SECTION III 


FAMILIES AND COMPOUNDS INCLUDED ON THE LIST 


A. INTRODUCTION 

This section describes the development of the list of compounds. The discussion is broken down into 
a series of chemical families. A discussion is provided as to why these families were chosen, as well as, a 
summary of previous measurements of fire suppression effectiveness for representative compounds from each 
family. The subsections conclude with a tabulation of the names and relevant properties for the family 
members which are included in the final list. 


B. SATURATED HALOCARBONS 

1. Justification for Consideration 

The saturated halocarbons or halons (C n F 0 Cl p Br q I r where o+p+q + r=2n+2) are well 
established as flame suppressants [66], Halons containing iodine are treated as a special case and are discussed 
later. Their mechanism for suppression is not fully resolved between two proposals: 

■ first, chemical action via catalytic recombination of chain carriers [55]; and 

■ second, physical action via increasing the capacity of the system to absorb heat 
[121,122,123]. 

In either case, a large number of halocarbons exist that would provide the necessary recombination agent (Cl 
or Br) or adequate stability and heat capacity. The appeal of the current commercial halons was primarily 
their relative safety and lack of residue. The safety allowed compounds such as halons 1301, 2402, and 1211 
to be used without the need for alternative life support systems in a large number of environments. The lack 
of residue and the nonionic character of the compounds also made them ideal for dealing with fires in sensitive 
electrical equipment. All of these compounds are very stable to reaction, and this very stability makes them 
such a threat to the ozone layer. The halons used in fire suppression are not destroyed in the troposphere, 
but are reactive in the stratosphere. In addition, all of the species listed above contain bromine, which has 
proven to be much more destructive to the ozone on a per mole basis than chlorine. 

The strategy for dealing with the relatively high ozone depletion potentials of the current 
commercial halons is to provide alternative compounds that either do not reach the stratosphere or, if reaching 
the stratosphere, do not participate in catalytic destruction of ozone. In the case of the CFCs (a subclass of 
halons) used in refrigeration and for solvent applications, this has meant a fairly simple modification - adding 
a hydrogen to the molecule. The resulting molecule will be destroyed in the troposphere in a time that is 
short relative to its tropospheric residence time by reactions with atmospheric hydroxyl radicals. It is also 
possible to design molecules that will react in sunlight and be destroyed in the troposphere. Finally, within 
the halocarbons we may seek molecules that do not destroy stratospheric ozone. 

The difficulty in designing such molecules does not come from the requirement that it put 
out a fire, or even from the requirement that it not destroy ozone, but rather from the requirement that these 
compounds have low toxicity. Tb see this, one need simply to look at the effects of addition of H atoms to 
any brominated compound. As an example we can use halon 2402, CF 2 BrCF 2 Br, which can be hydrogenated 


37 


in a number of ways. The simplest for this discussion would produce CF 2 HCF 2 Br. This molecule would be 
expected to react reasonably rapidly in the atmosphere, since the H atom is available for reaction with OH. 
At the same time, it may be expected to undergo hydrolysis in an ionic medium to produce HBr and 
more rapidly, and, since it would be expected to have a larger dipole moment, to be more soluble in an ionic 
medium. Hydrolysis in rain droplets would be faster and thus aid in the destruction of the compound; but 
solubility in blood and subsequent reaction could make the compound more toxic. 

Even in cases where the halon is not acutely toxic, it may have side effects that must be 
considered. For example, one of the compounds listed below as a candidate, halothane (BrCIHCCF 3 ), is an 
anesthetic. Halothane has been in clinical use for over 30 years so there is a substantial body of data on its 
side effects [131). It is not especially toxic, but there is evidence, as there is for other halocarbons, that cardiac 
sensitization does occur. In addition, the very fact that the compound is an anesthetic makes it necessary to 
exercise caution in using it as a general replacement for the current commercial halons. While these 
precautionary comments must be noted, it is also true that, with regard to toxicity, very small changes in the 
structure of the compound can greatly change the physiological effect [132]. The compounds suggested should 
be viewed as a starting point for further work and not as a final list. As studies on the toxicity and the fire 
suppression effectiveness of these compounds are extended, new, closely related compounds may well suggest 
themselves for study. 

In producing a list of compounds, two basic criteria were applied: 

■ the compound should have some feature that suggests it would not be as destructive 
to the ozone layer as the current commercial halons, 

■ it should test the basic ideas as to how halocarbons affect flames. 

In addition, wherever possible, compounds with low toxicity were sought. In many cases, however, the toxicity 
data was not available. 

The result of applying the first criterion was that only compounds containing a hydrogen atom, 
compounds containing no chlorine or bromine, or compounds that might be photoactive were included. The 
first alterative has been discussed briefly above. It simply states that if there is an H atom in the molecule, 
the rate of reaction with atmospheric OH might be fast enough to destroy the molecule in the troposphere 
before it reaches the stratosphere. The second alternative requires that the molecule be fully fluorinated. 
(Iodine-containing halons are discussed separately in Section III.F.) The final alternative can be stated as 
requiring either that the molecule have more than one Br/Cl on a single carbon, or that there be some other 
feature in the molecule that shifts the absorption cross section toward the red enough that the solar radiation 
flux in the troposphere would be sufficient to photolyze the molecule in a time short compared to the time 
expected for the molecule to reach the stratosphere. 

The second alternative-the fully fluorinated compounds-is the simplest to discuss and so we 
shall begin with it. There has been extensive work on the use of perfluoroalkanes as fire suppressants, partly 
as a result of the desire to create atmospheres that would sustain life and not combustion [44,48,133]. In 
addition, the fluorinated analog of cyclobutane has been suggested as a possible fire suppressant. The 
inclusion of these compounds has the further benefit of testing one of the two proposed mechanisms of fire 
suppression and thus satisfies the second criterion for selection. The perfluorinated compounds arc less 
reactive than other halocarbons. It is generally assumed that CF 4 is inert and thus differences between it and 
Ci (single carbon) halocarbons containing Br or Cl may be analyzed in terms of the difference between the 
chemical and physical effects. The larger perfluorinated compounds are less reactive than the corresponding 
bromine substituted compounds, but since the C-C bonds are about the same strength as those of 
hydrocarbons they are not totally inert in a flame. 


38 



The third alternative-photoactive compounds-requires data that are difficult to measure. 
The solar flux in the troposphere starts to increase dramatically at exactly the point that most of the current 
fire suppressants have steep decreases in their photoabsorption cross sections [134,135]. However, by putting 
two bromines or a bromine and a chlorine on the same atom, the absorption cross section shifts toward the 
red and the resulting tropospheric lifetime is decreased. In addition, the cross section will shift toward the 
red if the molecule is made larger. Going from C1-C2-C3 provides a steady, but small shift toward the red 
[136], Data for most molecules is difficult to find in the literature, and because the absorption cross sections 
are small, the measurements are subject to impurity effects. 

The first alternative - adding H to the halocarbon - is essentially that being followed by the 
research leading to alterative refrigerants and solvents. For these compounds, all hydrogenated chlorofluoro- 
carbons, the problem of hydrolysis, discussed above, is much less acute because of the increased C-X bond 
strength of Cl/F compounds relative to Br compounds. Thus one mode of toxicity is reduced. Beyond the 
toxicity effects there are also the effects on the physical properties of the molecule. In general, if one of the 
fluorines is replaced with a hydrogen, the boiling point of the compound will increase due to increased 
polarity. This is slightly offset by the decrease in boiling point due to decreased molecular weight, but, in 
general, the boiling point will increase upon H substitution. Replacing chlorine by hydrogen generally results 
in a decrease in boiling point. Most other physical properties show similar predictable variations with 
substitution. The hydrogen-substituted compounds also offer the greatest scope for examining some of the 
proposals for understanding fire suppression. 

The idea of endolhermicity affecting fire suppression is at the heart of the argument that fire 
suppressants work by adding to the heat loss mechanisms in the fire. While the suggestion that endothermic 
processes are important is not new [116], it has not been systematically explored. In a theoretical calculation 
one could increase the heat capacity of an additive species indefinitely and at some point the fire would no 
longer propagate. Curiously, this has not been done within the context of modern flame modeling calculations. 
In practice, the larger the molecule, the greater its heat capacity. There are, however, other ways of increasing 
the heat-absorbing potential of a molecule. One of these is to find molecules that undergo endothermic 
reactions to produce stable products. A simple example of this sort of process would use a hydrogen- 
substituted analog of halon 2402. In this case, like the example earlier, one of the bromines would be replaced 
with an H atom, resulting in CF 2 HCF 2 Br. This molecule could eliminate HBr to form CF 2 =CF 2 [137] in a 
reaction that was estimated by functional group contributions to be about 138 kJ/mole endothermic. The 
reaction rate has not been measured, but can be estimated to occur readily in a temperature range of 300- 
800°C. The 138 kJ/mole exothermicity is converted to a heat absorbed per gram, it corresponds to about 760 
J/g. This should be compared to a total heat absorbed by halon 1301 in going from room temperature to 
500° C of about 230 J/g. The reaction endolhermicity is in addition to the heat capacity of the starting 
material, which would be expected to be larger than that of halon 1301. Even if only 20 percent of the halon 
underwent this reaction, it would still represent an addition of 150 J/g to the heat extraction capability of the 
system. 


There are other important endothermic pathways; one is simple bond breaking. The C-Br 
bond strength in CFyBr is 290 kJ/mole, but the rate constant for bond dissociation is too small for this 
process to have an effect at temperatures below about 700°C [138], If the bond energy were lower by about 
40 kJ/mol, there might be substantial effects from bond breaking processes of this type. There has been very 
little exploration of this class of reactions as an explanation of fire suppression effectiveness. 

2. Past Flame-Suppression Measurements 

The literature on halogenated hydrocarbons is extensive (see Section II.B). A note of caution 
must be included in any survey of fire test results. First, the results arc often indicative, not predictive: small 
changes in the test configuration will change the results dramatically. Second, there are two approaches that 


39 




are often in conflict: the "full-scale test" and the "research project." An experiment using a premixed flame 
and suppressant added to the fuel gas before combustion is likely to yield different results than an experiment 
with a diffusion flame with the suppressant added in the oxidizer stream. Both of these would probably be 
classed as "research projects" and might correspond to the results of spraying a suppressant agent at a fully 
turbulent pool fire. Each of these experiments has value, and all need to be done. However, the 
interpretation of the results for each kind of experiment relative to another must be made with extraordinary 
care. The role of chemical and heat-capacity effects is best determined with the more controlled research scale 
tests, but the final choice as to which of otherwise equally effective agents should be used must be made with 
a clear understanding of how the material will be used in practical applications. In general, when trying to 
assess data, it is best to find a single experiment done on a large number of compounds-even if the experiment 
is not ideal. 


The earliest, and, in many ways, the most complete, of the tests of fire suppression agents is 
the Purdue Foundation Study [25]. In this study over 50 compounds were tested. In the most extensive set 
of tests, the compounds were added to n-heptane/air mixtures and the amount of the test compound required 
to suppress combustion for all heptane/air mixture ratios was reported. The data were taken under reduced 
pressure conditions - typically 0.4 - 0.8 atm, and generally over a range of stoichiometric conditions (1-6 
percent heptane, corresponding to stoichiometric ratios of 0.5 to 6 in the absence of inhibitor). Not every 
inhibitor had the full range of coverage (2-5 percent was the most common), and in most cases only about lb- 
20 separate experiments were preformed. 

There is no comparable body of data on diffusion flames, or oxygen index type tests. For 
these tests, one would use a diffusion flame and add suppressant to the oxidizer stream. The results from such 
a test more nearly simulate the effects of a total flooding system. In broad outline, the results would be 
expected to be similar, but not identical, to the results of the flame propagation experiments from the Purdue 
Study. Very few experiments have been done along these lines, although the work of Creitz [89,94] examined 
not only the effect of agent addition to the fuel and oxidizer, but also the effect of different fuels. 
Unfortunately, only two inhibitors were investigated. Simmons and Wolfhard [88] did similar experiments in 
a different burner configuration. (The cup burner test series discussed above studied a diversity of fuels, but 
focussed on halons 1211 and 1301.) 

3. Recommendations 

The saturated halocarbons chosen for inclusion on the list will be subject to revision. Small 
changes in the structure - moving a hydrogen from the end to an interior carbon, replacing a bromine with 
a chlorine, switching the position of a fluorine and a bromine - can have very large effects on toxicity, with 
possibly only minor effects on the fire suppression effectiveness. The recommendations are broken up into 
three groups based on the rationale for believing the molecule will not deplete the ozone layer. The first 
group is selected because they have no chlorine or bromine in the molecule. The chemicals in the second are 
probably photoactive and are more likely to be destroyed by sunlight in the troposphere. In the third group, 
all the chemicals contain hydrogen. 

a. Perfluoro Compounds 

The fluorinated analogues of methane, ethane, propane, n-butane and cyclobutane 
are initially considered. In addition, it may prove useful to examine the fluorinated analog of neo-pentane. 
The choice among these compounds is based largely on physical properties. Some work on these compounds 
indicates a relative effectiveness in fire suppression of C^F 8 > cyp^ > CF 4 [48], For the straight-chain 
perfluoro compounds the boiling points are -128,-78,-37,-2 for Cl - C4 respectively (temperatures in °C), so 
that all of them might be considered for total flooding applications. The corresponding heat capacities are 
63, 105, 146, 188 J/(K-mole). Data are not available on neo- C 5 F 12> but the heat capacity should be in the 


40 



region of 230 J/(K-mole) and the boiling point in the range of 0-10° C. Cyclooctafluorobutane (C 4 F 8 ) has a 
boiling point of -6° C and a heat capacity of 156 J/(K-mole). (The heat capacity per mole is the proper value 
to be using for a total flood agent since these compounds will be designed to replace a given volume of room 
air, and thus a defined number of moles of gas will be delivered. For streaming agents, comparing heat 
capacity per unit weight, or possibly per unit volume may be more useful.) 

In general, the toxicity of the perfluorocarbons is not a serious issue - the primary 
toxic effect is that of asphyxia; that is, it simply replaces the air. For at least some of the fluorinated 
compounds, there are other possible problems, cyclooctafluorobutane has been investigated for mutagenic 
effects, and one of larger fluorinated hydrocarbons, n-C 7 F 16 , has been reported to be mildly toxic via 
inhalation [139]. Thble 1 summarizes the relevant physical property data on the recommended compounds. 

b. Photosensitive Compounds 

The photon flux in the troposphere falls off dramatically at wavelengths below 300 
nm [127] due to the absorption of ozone in the upper atmosphere. For the current commercial halons, this 
is about the point at which they begin to absorb. The net effect is that photolysis is not an important factor 
in the destruction of these halons. However, small shifts of the absorption cross section toward the red 
(longer wavelength) would provide substantial increases in total rates of solar photolysis. Research for 
refrigerant substitutes has resulted in data indicating that solar photolysis is significant when there are two or 
more chlorine atoms on the same carbon atom [140], This can be generalized to say that solar photolysis is 
important as a destruction pathway whenever there are any two halogens larger than fluorine on the same 
carbon atom. Only for some methane derivatives - Cl compounds - would the absorption spectrum be shifted 
far enough toward the red for photolysis to be effective. The suggested compounds are CBrClF 2 , CBr 2 F 2 , 
CF 3 CFBr 2 . Note that the last is an isomer of halon 2402 which is CF 2 BrCF 2 Br Halon 2402 should be 
carefully re-investigated since it is on the edge of having a tropospheric lifetime short enough to be acceptable. 

CBrClF 2 is classed as an asphyxiant and is thus not toxic in the normal sense of the 
word. It has a boiling point of -4°C. The other two compounds suggested are higher in boiling point and 
thus likely substitutes for the liquid halons. CBr 2 F 2 has a boiling point of 22° C and a room temperature 
vapor pressure of 1.15 atm. These physical properties are almost ideal for a liquid agent, since they insure 
that there will be a minimum of time for evaporation of the extinguishant after its use. The lifetime for solar 
pyrolysis in the troposphere for these two compounds has been estimated: for the chloro compound the 
lifetime is 14-21 years; for the bromo compound 0.9-1.4 years. The ethane derivative CF^CFBr, will probably 
have a lifetime in the 1 year range also, and its physical properties will be similar to those of halon 2402. A 
summary of the relevant physical data for these compounds is given in Table 2. 

c. Hydrogen-Containing Compounds 

This group logically divides into two subgroups. The first contains those compounds 
which are under active consideration as refrigerant substitutes; the second, bromine-containing compounds. 
Brominated halocarbons were never used as refrigerants or solvents since the solvation or refrigerant 
properties that were desired could be achieved without bromination, which adds to the cost and in general 
reduces the chemical stability. There is, however, a broad body of evidence that bromine adds to the fire 
suppression capability, so brominated compounds were used for this application. In addition, there are special 
uses for brominated halocarbons, such as halothane, which is used as an anesthetic. The majority of uses for 
the halocarbons did not require bromine; and so when non-ozone depicting compounds were sought, simple 
hydrogen substitution was used. The current set of possible replacement compounds for refrigerants should 
all be tested for fire suppression potential. These compounds are listed in Thble 3 along with relevant physical 
property data. 


41 


TABLE 1. COMPOUNDS WITH NO BROMINE OR CHLORINE 


Name 

Formula 

CASN 

Phase 

Normal 
boiling point 
(nbp °C) 

Vapor 

pressure 

(atm) 

(298 K) 

Heat 

capacity 

(J/K-mol) 

(298 K) 

Heat 

of 

vaporization 
(kJ at nbp) 

perfluoromethane 

cf 4 

75-73-0 

gas 

-128 

na 

61 

11.6 

perfluoroethane 

C2F6 

76-16-4 

gas 

-78 

na 

106 

16.1 

perfluoropropane 

C 3 F 8 

76-19-7 

gas 

-36 

8.69 

148 

19.7 

perfluoro-n-butane 

C 3 F 10 

355-25-9 

gas 

-2 

2.63 

189 

23.2 

peril uorocyclobu la ne 

C 4 F 8 

115-25-3 

gas 

-6 

3.07 

156 

23.0 

trifluoromethane 

CHFj 

75-46-7 

gas 

-82 

46.7 

51 

16.7 

pentafluoroethane 

CzHF 5 

354-33-6 

gas 

-48.5 

10 + 

94 

20 + 

l,LL2-tetrafluoroethane 

c 2 h 2 f 4 

881-97-2 

gas 

-26.5 

7+ 

87 + 

22 + 

^Estimated for this report by authors 
na = not available 
















TABLE 2. PHOTOSENSITIVE COMPOUNDS 


Name 

Formula 

CASN 

Phase 

Normal 
boiling point 
(nbp °C) 

Vapor 

pressure 

(atm) 

(298 K) 

Heat 

capacity 

(J/K-mol) 

(298 K) 

Heat 

of 

vaporization 
(kJ at nbp) 

dibromodifluoromethane 

CB r 2F 2 

75-63-8 

gas 

-58 

15.8 

69 

17.5 

2 ,2-dibromo-l,l,l>2- 

tetrafluoroethane 

c^b^f 4 

27336-28-8 

liquid 

50+ 

0.4 + 

116 

29 f 

^Estimated for this report by authors 
























TABLE 3. COMPOUNDS WITH NO BROMINE 


Name 

Formula 

CASN 

Phase 

Normal 
boiling point 
(nbp °C) 

Vapor 

pressure 

(atm) 

(298 K) 

Heat 

capacity 

(J/K-mol) 

(298 K) 

Heat 

of 

vaporization 
(kJ at nbp) 

chlorodifluoromethane 

CHC1F 2 

75-45-6 

gas 

-41 

10.1 

57 

20.2 

1 ,1,1-trichloroethane 

C 2 H 3 C1 3 

71-55-6 

liquid 

74 

0.16 

93 

29.8 

2 ,2-dichloro- 

1 ,1,1-trifluorocthane 

C 2 HC! 2 F 3 

306-83-2 

liquid 

24 

1 + 

102 

26 + 

2 -chloro- 

1 ,1,1,2-tetrafluoroethane 

c 2 hc.if 4 

2837-89-0 

gas 

-12 

4+ 

101 + 

23 f 

1 ,1-dichloro-l- 

fluorocthane 

c 2 h 3 ci 2 f 

1717-00-6 

liquid 

32 

0.5 + 

89 f 

27 + 

l-chloro-1,1- 

difluoroethane 

c 2 h 3 cif 2 

75-68-3 

gas 

-10 

3.33 

82 

22.4 

^Estimated for this report by authors 
















The second group, containing bromine, are chosen to try to examine the role of HBr 
elimination, to provide H atoms for OH attack, and to examine a range of compounds for toxicological effects. 
The range of compounds includes some that are simple substitutions of H for F or Br in the current 
commercial compounds, and some that are extensions in that they are derivatives of propane. These have been 
chosen so that the properties could later be better adjusted. The compounds are: 

Cl: CHF 2 Br and CHFBrCl 

C2: CF 3 CHBrCl,CF 2 BrCFHCl,CF 2 BrCF 2 H,CH 2 BrCF 3 ,CF 2 BrCFHBr,CF 2 BrCH 2 Br,CHBrFCH- 

BrF 

C3: CF 2 BrCHFCF 3 ,CF 2 BrCH 2 CF 2 Br,CF 2 HCBr 2 CF 2 H,CBrF 2 CH 2 CF 3 . 

This list contains a bias, in that most of these compounds are commercially available. 
Other isomers may turn out to be important when toxicity is considered, but those listed above provide a 
framework for examining the fire suppression and physical properties of the molecules. Physical property data 
are given in "Bible 4. 


For comparison purposes, a list of properties for the current commercial halons is 
included in Table 5. While these compounds are not included explicitly on the final list of potential 
replacements, it is anticipated that they will be tested concurrently with the compounds appearing on the final 
list. 


C. HALOGENATED KETONES, ANHYDRIDES AND ESTERS 
1. Justification for Consideration 

Fluorinated carbonyl compounds (with appropriate bromine or chlorine substitution) are 
considered for the present application because the insertion of a carbonyl group into an organic framework 
shifts the absorption spectrum of the molecule towards the red. The consequence is that the molecule is more 
likely to be photolyzed in the troposphere (see Section II.C). A typical case is the contrast between the 
absorption spectra of and CH 3 COCH 3 [141]. For the alkane, absorption does not occur for wavelengths 
above 155 nm. Acetone, however, has an absorption maximum near 280 nm. Thus, it is not surprising that 
the absorption spectrum of hexafiuoroacetone peaks at 313.0 nm [141]. The mechanism and quantum yields 
for the photolysis of hexafiuoroacetone have been studied by many workers [142,143,144,145], In order to 
mimic as closely as possible the fire suppressant properties of compounds such as halon 1301 or halon 2402, 
a bromine group must be substituted for a fluorine in the structure. This will have the effect of moving the 
absorption spectrum further into the red. 

Spectroscopic information on fluorinated esters and anhydrides does not appear to be 
available. However the similarities between the absorption and photochemical data for 1,1,1- trifluoroacetone 
and acetone [145] suggest that the photochemical processes are dominated by the presence of the carbonyl 
group. Thus the data on the hydrogenated esters and anhydrides indicate that their absorption spectra will 
not be shifted as much towards the red as the ketones. It will be necessary to verify that photolysis will in fact 
destroy these compounds in the troposphere. All of these compounds are quite soluble in water. This 
provides another possible mechanism for destruction or removal of these compounds in the troposphere. 

In comparison to compounds without the carbonyl group (ethers and alkanes), the biggest 
changes in physical properties are lower vapor pressures and correspondingly higher boiling points. There will 
also be a small increase in the heat capacity. A thorough discussion of the chemical consequences under 


45 


TABLE 4. COMPOUNDS CONTAINING BROMINE 


Name 

Formula 

CASN 

Phase 

Normal 
boiling 
point 
(nbp ’C) 

Vapor 

pressure 

(atm) 

(298 K) 

Heat 
capacity 
(J/K-mol) 
(298'C) 

Heat 

of 

vaporization 
(Id at nbp) 

bromodifluoromethane 

CHBrF 2 

1151-62-2 

gas 

-15 

4.4 

59 

23* 

bromochlorofluorome thane 

CHBrCIF 

593-98-6 

liquid 

36 

0.5 + 

63 

28* 

2-bromo-2-chloro-l,l,l-trifliJoroethane 

CjHBrCIFj 

151-67-7 

liquid 

50 

0.4 

104 

27.3 

2 -bromo-l -chloro-l ,2,2-trifluoroet hane 

C 2 HBrCIF 3 

354-06-3 

liquid 

53 

0.4* 

103 

29* 

l-bromo-l,l,2,2-tetrafluoroethane 

C 2 HBrF 4 

354-07-4 

gas 

-5 

3.1 

86 * 

23.8 

2 -bromo-l ,1,1 -trifluoroethane 

C 2 H2BrF3 

421-06-7 

liquid 

26 

1 * 

91* 

27* 

1 ,2-dibromo-l,l ,2-trifluoroethane 

C 2 HBr 2 F 3 

354-04-1 

liquid 

76 

0 .2* 

106* 

31* 

l,2-dibromo-l,l-difluoroethane 

C 2 FI 2 Br 2 F 2 

75-82-1 

liquid 

93 

0 .1* 

95* 

33* 

l,2-dibromo-l,2-difluoroethane 

C 2 H 2 Br 2 F 2 

20705-29-7 

liquid 

102 * 

0.06* 

97* 

33* 

1 - bromo-1,1,2,3,3,3- hexa fl uoropropa ne 

C 3 HBrF 6 

na 

liquid 

50* 

0.4* 

145* 

29* 

1,3-dibromo-l ,1,3,3-tetrafluoropropane 

C3H2Br 2 F 4 

460-86-6 

liquid 

118 t 

0.04* 

138* 

35* 


C 3 H 2 Br 2 F 4 

na 

liquid 

120 * 

0.03* 

137* 

35* 

1 -bromo-1,1,3,3,3-penta fluoropropane 

C 3 H 2 BrF 5 

460-88-8 

liquid 

51* 

0.4* 

133* 

29* 

^Estimated for this report by authors 
na = not available 
















•U 


TABLE 5. CURRENT 


Name 

Formula 

CASN 

Phase 

halon 1301 

CBrFj 

75-63-8 

gas 

halon 1211 

CBrClF 2 

353-59-3 

gas 

halon 2402 

^2® r 2^4 

124-73-2 

liquid 
























combustion conditions will be presented later. At the simplest level the introduction of small amounts of CO 
from the carbonyl group into a flame is expected to have minimal effects. 

2. Past Flame Suppression Measurements 

Compounds with a carbonyl group have not been used as fire suppressants [146]. In the 
Purdue report [25], mention was made of the synthesis of ethyltrifluoroacetate. However, no results on flame 
suppression properties were given. As can be inferred from the earlier discussion, the exact nature of the fire 
suppression process is not well characterized. In the following some of the issues will be discussed. The 
mechanism for the decomposition of the suppressant in a fire situation may be an important property when 
there is a chemical basis for such activity. Although there are no direct measurements on the mechanisms and 
rates for the decomposition of these compounds, some estimates can be made [147,148,149]. If the suppressant 
molecule contains no hydrogen atoms, the general decomposition process will involve the cleavage of the 
weakest bond [147] and/or hydrogen atom attack on the bromine or chlorine substituent [150,151], The 
gradation of bond strengths for the present discussion is C-F > C-Cl * C-C > C-Br [149], Thus carbon- 
bromine bonds are the most readily broken in fire situations. Alternatively, hydrogen atoms can abstract 
chlorine and bromine, leading directly to the formation of the hydrogen halides. This is a much less selective 
process. At lower temperatures, this is undoubtedly the prime mechanism for removing halides from the 
organic moieties. The uncertainty regarding the temperature at which chemical suppressant action is effective 
prevents a definitive choice between these alternatives. Interestingly, it is usually assumed that hydrogen atoms 
will not abstract fluorine atoms. However, there are almost no experimental tests of this assumption. 

The consequence of these considerations is that, if one starts with halon 2402 and inserts a 
carbonyl group between the carbon-carbon bond, the chemical kinetic stability properties (in terms of 
unimolecular decomposition to form bromine or hydrogen atom attack leading to hydrogen bromide 
formation) will not be greatly changed. There are, however, differences from halon 2402 in the nature of the 
perfluorinated fragments. For halon 2402, perfluoroethylene is formed almost immediately upon removal of 
the first bromine, whether by radical attack or thermal decomposition. For the carbonyl compounds listed 
here, their chemical structures suggest the rapid formation of a large variety of perfluorinated compounds and 
radicals. It is usually assumed that these properties will have no effect on fire suppression capabilities. On 
the other hand, as noted earlier, Sheinson and coworkers [46,125] claim that the CF 3 radical acts as a chemical 
flame suppressant. Because some of these compounds are toxic, probably the most important reason for 
investigating the fire suppressant power of this type of compound is to elucidate the role of the organic 
fluorine species generated during fire situations. 

3, Recommendations 

In the absence of any data on the fire suppression properties of compounds containing 
carbonyl groups, it is worthwhile to examine these compounds in some detail. Thble 6 includes the compounds 
which are recommended for testing along with relevant physical properties. The first five compounds in Table 
6 are commercially available. Only one of these compounds contains a bromine group; thus, a true 
comparison to halons 1301 and 2402 will require custom synthesis of the last two compounds in the table, 
which contain bromine atoms. Synthesis of these compounds from common materials should not be 
particularly difficult [152]. Direct comparison of the fire suppression properties of these compounds when 
bromine is substituted for hydrogen and with the suppression effectiveness of alkanes and ethers should be 
especially interesting. 

All of these compounds have a considerable degree of toxicity. This is probably the chief 
constraint to their use [139]. It is for this reason that fluorinated acids, aldehydes and acyl compounds have 
not been included in the list. It may be possible to use these compounds in dilute mixtures in the presence 
of nonchemical suppressants, such as the perfluorinated compounds. It is suspected that the situation with 


48 



TABLE 6. HALOGENATED KETONES, ANHYDRIDES AND ESTERS 


Name 

Formula 

CASN 

Phase 

Normal 
boiling 
point 
(nbp °C) 

Vapor 

pressure 

(atm) 

(298 K) 

Heat 
capacity 
(J/K-mol) 
(298 K) 

Heat 

of 

vaporization 
(kl at nbp) 

hexafluoracetone 

CF 3 COCF 3 

684-16-2 

gas 

-26 

6.65 

118 

20.9 

trifluoroacetic anhydride 

cf 3 coococf 3 

407-25-0 

liquid 

40 

.4+ 

155+ 

28.0+ 

bis(perfluoroisopropyl) 

ketone 

(iC3F 7 ) 2 CO 

813-44-5 

liquid 

73 

.15+ 

280+ 

29.7+ 

methyltrifluoroacetate 

cf 3 cooch 3 

431-47-0 

liquid 

44 

.4+ 

113+ 

28.2+ 

3-bromo-1,1,1- 
trifluoropropanone 

CF 3 COCH 2 Br 

431-25-6 

liquid 

86 

.08+ 

113+ 

32.0+ 

bromopentafluoroacetone 

BrCF 2 COOCF 3 

815-23-6 

liquid 

31 

.9+ 

126+ 

26.4+ 

bromomethyltrifluoro- 

acetate 

cf 3 cooch 2 br 

116587-92-4 

liquid 

105+ 

.03+ 

126+ 

33.7+ 

+ Estimated for this report by authors 





regard to toxicity will not improve appreciably with bromine substitution. Furthermore, bromine substitution 
will increase the boiling point by 50-60° C. 


D. UNSATURATED HALOCARBONS 
1. Justification for Consideration 

Hydroxyl radicals add readily to double bonds [153] near ambient temperatures to form a 
hydroxyl-alkyl radical. The newly-formed radical readily reacts with the oxygen in the troposphere and can 
therefore be destroyed before it is transported into the stratosphere where formation of chlorine and bromine 
atoms through photolysis will destroy ozone. Rate constants for OH attack on totally halogenated ethylenes 
are, however, much smaller than for ethylene itself [153]. An important issue will be whether this decrease 
will move it out of the range necessary for destruction of these molecules in the troposphere. The presence 
of the double bond and a bromine atom may shift the absorption spectrum of a particular species further 
toward the red and thus give photolysis [141] a chance to destroy the gas before it migrates into the 
stratosphere. 


The introduction of a double bond leaves physical properties relatively unchanged. There are 
small decreases in the boiling point and heat capacity; however, profound changes in the possible chemistry 
can be expected. These are discussed in a subsequent section. 

2. Past Flame Suppression Measurements 

The Purdue report [25] mentioned that perfluorinated ethylene is combustible and that, in 
general, unsaturated compounds are ineffective in decreasing the flammability of air and n-heptane mixtures. 
However, the exact nature of the unsaturated compounds investigated was not specified; and it was found that 
2,2-difluorovinyl bromide had suppression properties similar to methyl bromide. Aside from this cursory 
mention, Larsen [154] has published a table containing data on lower explosion limits for a variety of 
halogenated organics with possible application as anesthetics. These include a number of olefins. Although 
exact comparisons are difficult, inspection of the data suggests that the presence of a double bond lowers the 
flammability limit. 

A large number of fluorinated olefins are available. Assuming a special role for bromine in 
fire suppression, custom synthesis will be necessary' to place bromine in a variety of positions and exploit the 
special effects introduced with the presence of a double bond. A very interesting aspect of these compounds 
is that, depending on the site of bromination, one can have molecules where the strength of the C-Br bonds 
will cover a range of nearly a 100 kJ/mol [138,148,149]. This can be contrasted with the situation for halon 
1301 and halon 2402 where the difference in bond energies is only 18 kJ/mol. For example, a bromine in the 
allylic position in a compound such as 3-bromoperfluoropropene will have a carbon-bromine bond strength 
about 45 kJ/mol lower than that for halon 2402. The bond strength will in fact approach that of the CF 3 -I 
bond. Such a compound can therefore be expected to release bromine atoms through a thermal mechanism 
with great rapidity. 

Even more interesting is the introduction of possible new reaction pathways. Although 
carbon-bromine bond energies adjacent to double bonds are greatly increased, hydrogen atoms can add to 
double bonds; and thermodynamics dictates that it will displace all halogens except fluorine atoms [155]. 
Thermochemistry suggests that OH can displace bromine atoms [156,157], Alternatively, after addition, the 
hydrogen halide may be eliminated directly. It is well known that in the pyrolysis of hydrocarbons the 
formation of olefins is responsible for the observed self-inhibiting nature of most of these systems [158]. The 
Fire suppression capability of difluorovinylbromide can be rationalized in terms of the displacement of the 


50 



bromine atom by a hydrogen atom. This provides a rapid means of releasing bromine into the combustion 
region. 


3. Recommendations 

Table 7 lists the names, formulas, and relevant properties for those compounds which are 
recommended for study. The first six compounds are commercially available. None contains a bromine group. 
Therefore, custom synthesis will be necessary to widen the range of possibilities. Based on the earlier 
comments, the molecules which have been added to Table 7 will be particularly interesting. 

Bromine substitution will increase the boiling point. Some of the compounds contain 
hydrogen. This will render the double bond redundant. The possibility that under certain conditions they may 
be inflammable must also be considered [25,154]. 

Many of these compounds are mildly toxic [139] and therefore compare unfavorably with 
halon 1301. Perfluorinated isobutene is an extremely dangerous poison. It may well be necessary to consider 
the possibility of its formation during the use of some of the other perfluorinated olefins. Unsaturated olefins 
such as l,2-bis(perfluoro-n-butyl)ethylene or bis(di-n-butyl)ethylene have been used as blood substitutes 
[159,160]. This would seem to assure that these compounds are not toxic. As blood substitutes, these 
molecules must have low vapor pressures or, equivalently, high boiling points. Thus the boiling point for 1,2- 
bis(perfluoro-n-butyl)ethylene is 132* C and would seem to disqualify this compound on purely physical 
grounds. Compounds such as 1,2- bis(perfluoromethyl)ethylene, tetris(perfluoromethyl)ethylene, and those 
where there is bromine for fluorine or hydrogen substitution will probably have more reasonable boiling 
points. The important issue is whether such molecules will retain the favorable toxicity profiles. Bagnall and 
coworkers [161] have synthesized a number of partially-fluorinated butenes with chlorines in the allylic and 
vinyl positions in order to investigate possible anaesthetic properties. It appears that a vinyl CF 2 C1 group 
leads to unfavorable toxic effects. 

The main reason for studying these compounds is to define the role of the double bond. It 
should be noted that the experimental observations regarding self-inhibition and the lowered explosion limits 
appear to be contradictory. There are clearly fundamental issues that are not understood. Particularly 
important experiments are those that involve studies with bromine in a variety of positions. For example, for 
perfluorinated propene, studies with bromine substitution in the 1 and 3 positions may well lead to the 
determination of whether the manner or rate by which bromine atoms are released into a fire situation is of 
any consequence. 


E. HALOGENATED ETHERS AND RELATED COMPOUNDS 
1. Justification for Consideration 

The addition of an oxygen between two carbon groups in an alkane moves the absorption 
spectra about 30 nm toward the red [141]. This suggests that for a compound such as CF 2 BrCF 2 Br or 
CF 3 CFClBr, oxygen insertion may be sufficient to make it photochemically active [136]. The chances of ethers 
with only one bromine or chlorine atoms being photochemically active in the appropriate spectral region is 
much less likely. The addition of an extra CF 2 group is not expected to improve the situation. The effect of 
hydrogen substitution will probably be similar to that in the current commercial halons with respect to 
reactivity towards OH attack. 

Some of the compounds in this class are well known anesthetics [162,163,164,165]. These 
include methoxyflurane (CHCl 2 CF 2 OCH 3 ), enflurane (CHF 2 OCF 2 CHFCl), and isoflu rane (CF3CHC10CHF2). 


51 




Lh 

to 


TABLE 7. UNSAT 


Name 

Formula 

CASN 

perfluoropropene 


116-15-4 

perfluorobutene-2 

C 4 F 8 -2 

1 1 

perfluorotoluene 

cf 3 c 6 f 5 

434-64-0 

1,1,3,3,3-pentafluoropropene-l 

cf 3 ch=cf 2 

690-27-7 

3,3,3-trifluoropropene 

cf 3 ch=ch 2 

690-27-7 

l,2-bis(perfluoro-n-butyl)ethylene 

(n-C 4 F 9) CH= 

CH(n-C 4 F 9 ) 

8455143-9 

3-bromoperfluoroprop>ene 

BrCF 2 CF=CF 2 

431-56-1 

1-bromoperfluoropropene 

CF 3 CF=CFBr 

14003-53-33 

14003-61-3 

l,2-bis(perfluoromethyl)ethy!ene 

CF 3 CH=CHCF 3 

66711-86-2 

1 -bromoperfluoromethyl-2- 
perfluoromethylethylene 

CF 2 BrHC-CHCF 3 


1 -bromo-bis( perfluoromethy 1)- 
ethylene 

CF 3 BrC=CHCF 3 

400-41-9 

tetris(perfluoromethyl)ethylene 

(CF 3 ) 2 C=C(CF 3 ) 2 

360-57-6 


^Estimated for this report by authors 


HALOCARBONS 


Phase 

Normal 

boiling 

point 

(nbp'C) 

■n 

■9 

Heat 

capacity 

(J/K-mol) 

(298 K) 

Heat 

of 

vaporization 
(kJ at nbp) 

gas 

-29 

6.42 

116 

21 

gas 


2.62 

+- 

00 

t-H 

21 + 

liquid 


.03+ 

207 

34+ 

gas 

-21 

5+ 

109+ 

23+ 

liquid 

-17 

4+ 

92+ 

24+ 

liquid 

132 

,006 + 

377 f 

36+ 

liquid 

28+ 

1+ 

121+ 

27+ 

liquid 

30+ 

1+ 

121'*’ 

27+ 

liquid 

6+ 

1 .6+ 

130+ 

26 f 

liquid 

57 t 

0.1 + 

138+ 

30+ 

liquid 

65 + 

,06 t 

143 + 

33+ 


liquid 


55 


180 + 


29 + 






















It is claimed that they are non-flammable. This is however only with reference to their specific application. 
Larsen [154] has shown that their lower flammability limits for oxygen are only at the 5 to 7 percent level. 
This can be compared with a value of 10 percent for methyl bromide. The large number of studies on the use 
of fluorinated compounds as anesthetics provides a very important initial source of toxicity data 
[154,161,164,165,166,167]; however, these data should be used with care, since much of the focus is on short¬ 
term effects. A disturbing property of the ethers is their wide range of (generally unpredictable) biological 
responses. Hexafluorodiethyl ether, for example, is an extremely potent convulsant. Perfluorodiethyl ether 
is, however, only mildly toxic. Of course, a good anesthetic is an undesirable fire suppressant. Indeed, the 
ratio of concentrations necessary for satisfactory anesthesia and lethal dosage is rarely an order of magnitude 
[154]. Perfluorinated n-butyltetrahydrofuran has been used as a blood substitute. Certainly this application 
should provide a very satisfactory guarantee of safety. A high priority should be assigned to the bromination 
of the perfluorotetrahydrofuran. 

Physical properties of the ethers are quite favorable. The boiling points appear to be little 
changed with the substitution of the oxygen. 

The above discussion has focused on C, F, H, and O compounds. Some consideration has 
also been given to halogenated organic systems involving sulfur and nitrogen atoms. The specific compounds 
are the perfluorinated thioethers and amines. Although the absorption spectra of these compounds probably 
do not extend sufficiently far toward the red for photolysis to be an effective destruction mechanism in the 
troposphere [141], a particularly interesting aspect of their behavior is the possibility of addition of OH to the 
sulfur or nitrogen moiety followed by reaction with oxygen and subsequent removal from the troposphere. 
The evidence for such a process is quite strong for the organic sulfides [168] and less compelling for the 
amines where there are insufficient data. 

The sulfur analog of the well known anaesthetic methoxyflurane has been synthesized and it 
appears to be a much more powerful anaesthetic. This suggests that sulfur substitution may well have 
deleterious toxic effects in comparison to the ether. However it is known that bis(trifluoromethyl) sulphide 
resembles a perfluoro ether in its stability to heat and attack by strong acids and bases [169]. 1>is(per- 
fluorinated propyl) amine is used as a blood substitute. It can therefore be assumed that there will be no 
toxicity problems. It also suggests that smaller members of this group will have similar properties. This is 
borne out by the inertness of tris(perfluoromethyl) amine [170]. 

2. Past Flame Suppression Measurements 

Malcolm [ 6 ] has noted rough flame suppression studies with octadecafluoroperdibutyl ether. 
On a relative basis it is only half as effective as carbon dioxide. This is difficult to understand since the heat 
capacity of the perfluoro compound is much larger than that for C0 2 - There does not appear to have been 
any other prior work using halogenated species containing ether linkages as fire suppressants [25], The 
strengths of the carbon-bromine bonds are expected to be in the same range as halon 1301 and halon 2402. 
Perfluorinated phosgene as well as various fluorinated carbenes and methyl radicals are expected to be formed 
quite readily. Experimental studies will therefore bear on the question of the importance of such substitutions 
for the CF 3 and C 2 F 4 from halon 2402 and halon 1301 decomposition. Although it has been noted earlier 
that the three anesthetic ethers are not really non-flammable in the general sense, it may well be that 
bromination will improve this property. 

3. Recommendations 

The compounds listed in Thble 8 are recommended for study. The first five compounds are 
commercially available and may also be useful as possible reactants for subsequent synthetic work. With EPA 
support, bis(bromodifluoromethyl) ether and l-bromo-1,1,3,3,3- pentafluorodimethyl ether are now being 


53 



TABLE 8. HALOGENATED ETHERS AND RELATED COMPOUNDS 


Name 

Formula 

CASN 

Phase 

Normal 
boiling 
point 
(nbp °C) 

Vapor 

pressure 

(atm) 

(298 K) 

Heat 
capacity 
(J/K-mol) 
(298 K) 

Heat 

of 

vaporization 
(kJ at nbp) 

tetrafluorodimethyl ether 

CF 2 HOHCF 2 

1691-17-4 

gas 

2 

2 + 

92+ 

24.5+ 

pentafluorodimethyl ether 

cf 2 hocf 3 

3822-68-2 

gas 

-35 

7+ 

109+ 

21 .2+ 

2 -chloro-l-(difluoro- 

methoxy)-l,l,2-trifluoro- 

ethane 

chf 2 ocf 2 chfci 

13838-16-9 

liquid 

56.5 

.29 

162+ 

29.4+ 

isoflurane 

cf 3 chciochf 2 

26775-46-7 

liquid 

48.5 

.32 

167+ 

28.7+ 

perfluoro-2- 

butylteirahydrofuran 

c 4 f 9 c 4 f 7 o 

335-36-4 

liquid 

103 

.033+ 


33.5+ 

bis(bromodifluoroethyl) 

ether 

CF 2 BrOCF 2 Br 

na 

gas 

25+ 

1 + 

117+ 

26.7+ 

1-bromo-l,l,3,3,3-penta- 

fluorodimethyl ether 

CF 2 BrOCF 3 

na 

gas 

-30+ 

6 + 

104+ 

24.7+ 

bromoenflurane 

CBrF 2 OCF 2 CHFCl 

na 

liquid 

67+ 

2 + 

175+ 

. 30.3+ 

octafluorofuran 

c 4 f 8 o 

773-14-8 

liquid 

50+ 

.33+ 

151+ 

28.8+ 

3-bromoperfluorofuran 

BrC 4 F 7 0 

na 

liquid 

105+ 

.03+ 

159+ 

33.7+ 

bis(perfluoromethyl) 

thioether 

cf 3 scf 3 

371-78-87 

gas 

-22.2 

6.1 

126+ 

25.1+ 

tris(perfluoromethyl) 

amine 

(CF 3 )N 

432-03-1 

gas 

-10.1 

3.9 

167+ 

25.1+ 

"^Estimated for this report by authors 
na = not available 




synthesized. The same procedures should lead to the bromination of the enflurane and isoflurane. Perfluoro- 

2- butyltetrahydrofuran is included because of its use as a blood substitute. This is a large compound and its 
boiling point and vapor pressure are unfavorable. Thus it will be important to test the tetrafluorofuran and 

3- bromotetrafluorofuran. 

The first prerequisite will be to establish that the simplest brominated ethers can in fact be 
photolytically decomposed. Comparison of these compounds with the purely fluorinated analogs will be 
extremely interesting not only for the immediate problem, but to develop a better understanding of fire 
suppression mechanisms. For example, comparisons of the behavior of bis(bromodifluoromethyl) ether with 
halon 2402 and l,3-dibromo-l,l,2,2,3,3-hexafluoropropane should yield a great deal of information on 
mechanistic issues. It is interesting that enflurane (CHF 2 OCF 2 CHFCl) has apparently lower toxic effects in 
comparison to halothane (CF 3 CHClBr) and has overtaken the use of the latter as the most used volatile 
anesthetic agent in many institutions, despite the higher cost [162]. It is possible that the greater toxicity of 
the halothane is due to the presence of the bromine group. This is an issue that should be investigated. 

For the nitrogen and sulfur compounds, it will be extremely interesting to investigate 
bis(perfluoromethy) sulfide and tris(perfluoromethyl) amine, and, should the initial results be favorable, to 
go on to compounds where bromine is substituted for fluorine. These compounds are not available from 
commercial sources. However, the standard electrochemical method (Simons process) [171] is probably 
suitable for their preparation. This process has, in fact, been used to prepare some of the amines [170,172]. 
It is also very convenient that fluorination is not complete. There will, therefore, be a few hydrogens 
remaining for bromination. The sulfide can also be prepared from the pyrolysis of the disulfide [173]. 

F. HALONS CONTAINING IODINE 

1. Justification for Consideration 

Current commercial flame suppressants such as halons 1211 and 1301 include bromine atoms. 
Many experts in the field (e.g., [57,72,146] have reported the efficiency of halogen atoms for fire suppression 
as follows: 


I > Br > Cl > F. 

Some compounds containing iodine were investigated in the Purdue study [25], Malcolm [6] 
used these findings to assign "apparent atomic values of firefighting effectiveness" for the halogens. His results 
were: fluorine, 1; chlorine, 2\ bromine, 10; and iodine, 16. Later work showed that extinguishing 
concentrations of halons containing F, Cl, and Br were correlated accurately in terms of scores obtained simply 
by summing the products of the number of each halogen atom in the molecule by its firefighting effectiveness 
[97]. Unfortunately, no molecules containing iodine were tested during the development of this correlation. 

Despite the apparent effectiveness of iodine atoms for flame inhibition, no compounds 
containing iodine have been widely used or tested for flame extinguishment purposes. References in the early 
literature of the field [1,5] suggest that halons containing iodine were not considered primarily due to the 
scarcity and cost of iodine; it is also clear that the recognition that iodine compounds are generally more toxic 
played a role [1]. 

In general, iodine-containing halons are more reactive and have longer wavelength absorptions 
than analogs containing lower molecular weight halogens. These characteristics suggest that molecules 
containing iodine will be more easily destroyed in the troposphere than the corresponding bromine 
compounds. As a result, they are less likely to reach the stratosphere where they would contribute to the 
destruction of ozone. 


55 




Based on the above discussion, it is considered likely that halons containing iodine will have 
good fire suppression capabilities while being less damaging to stratospheric ozone than the current 
commercial halons. In addition to this practical reason for including a number of compounds containing 
iodine on the final list, it is also likely that new principles for the understanding of chemical fire suppression 
will emerge following the testing of a number of these species. 

2. Past Flame Suppression Measurements 

One very short, but intriguing reference, suggests that iodine-containing compounds may be 
very effective extinguishment agents for real fires [174]. de C. Ellis refers to early work of R.V. Wheeler and 
H.B. Dixon who, it is claimed, discovered "the combustion-inhibiting influence of the iodides" [174]. 
Measurements were made to test the effectiveness of methyl iodide as compared to carbon tetrachloride, the 
commonly-used chemical fire extinguisher of the day. Results indicated that only one fifth of the quantity of 
CH 3 I was required to extinguish a fire. Tfcsts are described for petrol burning on water with an indication that 
burning areas greater than 1.3 m 3 were used. Carbon tetrachloride sprayed on the fires failed to extinguish; 
but when methyl iodide was used, the flames were quickly quenched. 

Some common measurements of suppression effectiveness have been reported for iodine- 
containing compounds. In 1932 Jorissen et al. [118] reported the flammability limits for methane-air mixtures 
inhibited by ethyl iodide. The peak concentration was slightly greater than 5 percent. Flammability peak 
concentrations for CH 3 I (6.2 percent) and C 2 H 5 I (6.2 percent) in n-heptane/air mixtures were determined 
during the Purdue investigation [25]. Both agents required considerably lower concentrations to inert n- 
heptane/air mixtures than the corresponding species with bromine. The peak concentrations were somewhat 
higher than for fully-halogenated alkanes. Lask and Wagner reported the effects of CF 3 I on the flammability 
limits of n-hexane/air mixtures [79]. The peak concentration for this compound was roughly 3.5 percent. 

Rosser et al. [78] have reported measurements of reductions in flame speeds for methane/air 
mixtures due to the addition of a number of compounds including CH 3 I and C^Hyl. Their results indicated 
that the compounds containing iodine are roughly equivalent to their analogs containing bromine in fire 
suppression effectiveness. Homann and Poss measured reductions in flame speed as various additives were 
added to ethylene/air mixtures [81]. Measurements were recorded as a function of pressure. Direct 
comparisons were made for CH 3 I and CH 3 Br acting as inhibitors. For a range of pressure, the methyl iodide 
was found to inhibit the flame slightly more efficiently. 

Flame speeds for a one-dimensional ethylene/air premixed flame with 1 percent concentrations 
of various agents added have been calculated by Westbrook utilizing a detailed chemical kinetics model [86]. 
Agents included were HC1, HBr, HI, and methyl, ethyl, and vinyl radicals combined with Cl, Br, or I. The 
addition of the agents to the fuel/air mixture resulted in significant reductions in calculated flame speeds, with 
the relative reductions increasing in the order I « Br > Cl. These results suggest that species containing 
iodine should be effective fire suppressants. 

Sheirison et al. [46,125] have reported cup burner results for heptane fuel in which the 
concentrations of agent in air required to extinguish the flame were measured for a number of compounds 
including CF 3 Br (3.1 percent) and CF 3 I (3.2 percent). These two compounds yielded the lowest percentages 
of the compounds tested in this study. As is clear from the values reported, the effectiveness of these two 
similar compounds were nearly identical with perhaps a slight edge going to the agent containing bromine. 

A few workers have investigated the effects of agents containing iodine on fires. The work 
of de C. Ellis [174] was described above. Malcolm [56] quotes the results of two relevant studies. One 
investigation was performed by the Civil Aeronautics Authority in Indianapolis, IN (See Appendix E, Exhibit 1 
in Malcolm [56]). Fires in a model aircraft engine were extinguished by a variety of agents and the amount 


56 



required for extinguishment reported as a weight. On a per mole basis, the amounts of methyl iodide and 
methyl bromide required were identical. The second study quoted was work performed at the Minnesota 
Mining and Manufacturing Company and is summarized in Malcolm’s Appendix E, Exhibit 2 [56]. In this 
study flaming cotton was inserted into a small enclosure in which the inhibitor had been mixed with air and 
the amount of agent required for extinguishment measured. One of the gases tested in this way was methyl 
iodide. On a per mole basis the most effective agent tested was the halon with iodine. The agent CB (halon 
1011) was one of the molecules included in this investigation. The methyl iodide was roughly 10 percent more 
effective than this agent. 

3. Recommendations 

The fire suppression properties, as well as other relevant properties of halons with iodine 
atoms, have not been well-characterized despite indications that such compounds are very effective flame 
suppressants. A wide range of such compounds is included on the list in order to provide insight into fire 
suppression mechanisms and to investigate whether such compounds might be used as replacements for 
existing clean firefighting agents. Appropriate testing of these compounds is likely to provide significant 
insights into the effects of replacing bromine with iodine on the ODP and toxicity behaviors of halons. 

As a direct test of the effectiveness of compounds containing iodine compared to the three 
widely employed halons (1211, 1301, and 2402), four analogs of these compounds with iodine substituted for 
one or more of the bromine atoms are included on the list. In order to characterize the behavior of 
hydrocarbons with a single substituted iodine atom, the homologous series with 1-3 carbons has been added. 
Finally, a perfluorinated propyl iodide has been included to characterize the effect of substituting fluorines 
for hydrogens on the alkyl iodides fire suppression characteristics. The resulting list of compounds is given 
in Thble 9, along with appropriate physical properties. 


G. SULFUR HALIDES 

1. Justification for Consideration 
a. Parent Properties 

Sulfur halides are identified as possible fire suppressants, based principally on the 
properties of one member of this class of compounds — SF 6 . This gas has many desirable properties for a 
chemical fire suppressant. It is colorless, odorless, tasteless, nontoxic, and nonflammable [175]. Since SF 6 
contains numerous fluorine atoms, the molecule might be expected to have significant fire suppression 
capability. 


b. Chemical Analogs 

SF 6 is a unique molecule due to its high symmetry and strong bonds. The other 
sulfur halides are toxic, reactive, and thermally unstable. As noted above, this is not the case for SF 6 . Normal 
nucleophilic reagents cannot attack the highly symmetric octahedral structure [175], 

One might hope that it would be possible to exchange one or more of the fluorine 
atoms on SF 6 for chlorine or bromine to create a gas with properties similar to SF 6 , but with improved fire 
suppression capabilities. Unfortunately, this does not turn out to the case. Only two analogs, SF 5 C1 and 
SF 5 Br, are stable. The molecular properties of these species differ dramatically from those of the parent 
molecule. SF 5 C1 is stable up to 400° C while SF^Br decomposes at 150° C [176]. Both are easily hydrolyzed. 
SF 5 C1 has been reported in rat studies as a strong lung irritant and as being as toxic as phosgene [177], 


57 



TABLE 9. RECOMMENDED HALONS CONTAINING IODINE 


Name 

Formula 

CASN 

Phase 

Normal 
boiling 
point 
(nbp °C) 

Vapor 

pressure 

(atm) 

(298 K) 

Heat 
capacity 
(J/K-mol) 
(298 K) 

Heat 

of 

vaporization 
(kJ at nbp) 

iodotrifluoromethane 

cf 3 i 

2314-97-8 

gas 

-22.5 

na 

70.9 

22 + 

chlorodifluoroiodomethane 

cf 2 cii 

420-49-5 

liquid 

33 

na 

76 + 

27 f 

l-bromo-l,l,2,2-tetrafluoro-2- 

iodoethane 

CF 2 BrCF 2 I 

421-70-5 

liquid 

78 

na 

124 + 

31 f 

1,1,2,2-let ra fluoro-1,2- 
diiodoethane 

cf 2 icf 2 i 

354-65-4 

liquid 

112 

na 

126 + 

34 f 

iodomethane 

ch 3 i 

74-88-4 

liquid 

42.4 

0.53 

44.1 

27.4 

iodoethane 

ch 3 ch 2 i 

75-03-6 

liquid 

72.3 

0.18 

67.8 

29.4 

1-iodopropane 

ch 3 ch 2 ch 2 i 

107-08-4 

liquid 

102.4 

0.06 

88.3 

32.1 

1,1,1,2,2,3,3-heptafluoro-3- 
iodopropane 

cf 3 cf 2 cf 2 i 

754-34-7 

liquid 

41 

na 

197+ 

28+ 

^Estimated for this report by authors 
na = not available 




2. 


Past Flame Suppression Measurements 


Three references were identified which report the mole fraction of SF 6 at the flammability 
peak (concentration just sufficient to limit flame propagation for all concentrations) for hydrocarbon/air 
flames. The Purdue study [25] found that a mole fraction of 0.205 was required to inhibit n-heptane/air flames. 
This was one of the higher limit concentrations reported of the 56 compounds investigated. A similar 
investigation for pentane/air mixtures yielded a mole fraction of 0.158 [82]. Compared to the levels of C0 2 
necessary for these cases, SF 6 appears to be slightly more effective for pentane combustion than for n-heptane, 
but relatively high mole fractions of SF 6 are required for extinguishment with each fuel. 

A low-pressure premixed CH 4 /air flame only required a mole fraction of 0.057 SF 6 to cause 
extinguishment [178]. This mole fraction is somewhat lower than found in the atmospheric-pressure flame 
studies, but it is difficult to reach conclusions since the amounts of other gases required to extinguish the low- 
pressure flame were also smaller than found for the corresponding atmospheric-pressure flames. 

Sheinson et al. [45,46] have investigated the effectiveness of SF 6 as a flame extinguishment 
agent in a cup burner apparatus with n-heptane fuel. Their results and analysis indicate that the only 
mechanism responsible for flame suppression with SF 6 is the physical effect based on the molecular heat 
capacity. 


The only fire suppression investigation of halogen-substituted SF 6 compounds identified was 
the work of Sheinson et al. [45,46] These workers measured the amount of SFgCl and SF 6 Br required to 
extinguish n-heptane flames. Their analysis suggests that SF5CI has no chemical suppression capability and 
may actually enhance combustion. On the other hand, SF 5 Br appears to have a definite flame suppression 
capability. 


3. Recommendations 

SF 6 is readily available commercially. SF 5 C1 is also produced by specialty manufacturers. No 
commercial source of SF s Br has been identified, but an effective procedure for its preparation has been 
published [179] and it should be possible to synthesize this molecule at a reasonable cost for testing purposes. 

Since a consensus does not exist on the fire suppression capabilities of SF 6 , this molecule 
should definitely be included on the list for testing. Clearly, SF 5 C1 and SF s Br are not suitable as replacements 
for the current commercial halons. However, it seems likely that the understanding of fire suppression 
mechanisms can be significantly enhanced if these compounds are included in the testing protocol. It is 
recommended that SF^Cl, SF 5 Br and SF 6 be included on the list of test compounds. Relevant properties are 
included in Thble 10. 


H. COMPOUNDS CONTAINING PHOSPHORUS 
1. Justification for Consideration 

Phosphorus-containing compounds are widely used as fire retardants for natural and man¬ 
made polymers [146], This suggests that gas-phase phosphorus compounds could prove to be effective flame 
extinguishment agents. It is generally believed that antimony halides are effective radical traps [146] and thus 
act as flame suppressants. Since phosphorus and antimony are both group V elements, a similar behavior 
might be expected for phosphorus trihalides. 


59 





TABLE 10. RECOMMENDED SULFUR HALIDES 


Name 

Formula 

CASN 

Phase 

Normal 
boiling point 
(nbp °C) 

Vapor 

pressure 

(atm) 

(298 K) 

Heat 

capacity 

(J/K-mol) 

(298 K) 

Heat 

of 

vaporization 
(kJ at nbp) 

sulfur fluoride 

sf 6 

2551-62-4 

gas 

-63.8 (sub) 

21.8 

97.0 

23.6 

sulfur chloride fluoride 

sf 5 ci 

13780-57-9 

gas 

-21 

na 

104 

21.7 

sulfur bromide fluoride 

SF 5 Br 

15607-89-3 

gas 

3.1 

na 

107 

24+ 

^Estimated for this report by authors 
na = not available 




2. 


Past Flame Suppression Measurements 


An immense literature exists which deals with the use of phosphorus-containing compounds 
as flame retardants in solids [146]. However, only a few references have been identified in which the effective¬ 
ness of a phosphorus-containing liquid or gas as a flame suppressant has been tested. Phosphorus trichloride 
was included as one of a large number of compounds tested for fire-suppression capabilities during the study 
carried out at Purdue University [25]. Results for this study are reported in terms of the volume percent of 
the tested molecule required to prevent flame propagation in heptane/air mixtures. PC1 3 required one of the 
highest volume fractions, 22.5 percent, of any species studied during this investigation. Of the 56 compounds 
considered in the Purdue study, only C0 2 , CF 4 , and HC1 required higher volume fractions to prevent flame 
propagation. 


Lask and 3\&gner [79] measured the concentrations of a variety of species required to reduce 
the flame velocity of a n-hexane/air flame by 30 percent. Included in the tests were PC1 3 , PBr 3 , PSC1 3 , and 
PSBr 3 . These compounds were all found to be considerably more effective inhibitors than Cl 2 , Br 2 , or CC1 4 . 
These authors concluded that the phosphorus compounds inhibit flames by a second chemical mechanism 
which occurs in conjunction with the mechanism responsible for flame inhibition due to halogen molecules. 

POCl 3 was one of the molecules Miller tested for extinguishment of low-pressure premixed 
flames of methane [178]. A considerably lower mole fraction of POCl 3 than Br 2 , which is recognized as an 
effective suppressant, was required for extinguishment. Jorissen et al. [118] also noted that POCl 3 is a very 
powerful flame inhibitor. They claimed that less than 1 percent of this compound is sufficient to prevent flame 
propagation in methane/air mixtures. 

The papers summarized above suggest that some phosphorus-containing molecules may be 
very efficient fire suppression agents. At the present time, the fire suppression behavior of this family of 
molecules must be classed as poor. Investigation of the fire suppression behavior of phosphorus-containing 
molecules is certain to result in an improved understanding of chemical fire suppression mechanisms and 
might point the way to possible replacements for the current commercial halons. 

3. Recommendations 

Only a relatively few liquids and gases containing phosphorus are known [180,181]. These 
compounds are the hydrides, trihalides, pentahalides, and oxo-halides. In addition, the phosphonium halides 
have been observed. Most of the compounds are highly toxic and hygroscopic. For this reason they are poor 
candidates for widespread replacement of existing halons. However, it is possible that one or more of these 
compounds may provide important tests for the screens being developed or new insights into flame 
extinguishment mechanisms. A choice of appropriate compounds for screen testing is aided by the information 
in Thble 11 [181,182,183,184]. Only a few of these compounds are available commercially. These include 
PC1 3 , PBr 3 , POCl 3 , PSC1 3 , PH 3 , and PC1 5 . 

Based on previous results of fire suppression investigations and the information shown in 
Thble 11, the compounds included in Thble 12 have been chosen for inclusion on the final list of compounds. 
These compounds should allow conclusions concerning the role of phosphorus in fire suppression effectiveness 
to be formulated. Comparison between the results for the phosphorus halides and the phosphoryl halides will 
allow the fire suppression effectiveness of these two families of phosphorus compounds to be characterized. 
Finally, by comparing findings for cases where different halogen atoms are substituted on the phosphorus, the 
importance of chemical flame suppression due to halogen molecules to the overall suppression efficiency of 
this class of molecules can be assessed. 


61 




TABLE 11. PHOSPHOROUS-CONTAINING GASES AND LIQUIDS 


Compound 

m. P . (°C) 

b.p. (»C) 

Comments 

PH, 

-209 

-126 

Highly flammable, can form explosive mixtures 

ph 4 ci 

28 

subl 

This compound is very unstable 

PH 4 Br 

*30 

subl 

This compound is very unstable 

ph 4 i 

*19 

39 

This compound is very unstable 

pf 3 

-152 

-102 

Complexes with hemoglobin, hydrolyses slowly 

pci. 

-112 

76 

Violently hydrolyzed by water 

PBr, 

-40 

173 

Violently hydrolyzed by water 

pf 2 ci 

-165 

-47 


pfci 2 

-144 

14 


PF 2 Br 

-134 

-16 


PFBr 2 

-115 

78.4 


pf 5 

-136 

-121 

Very hazardous due to hydrolysis by water 

pof 3 

-39 

-40 


pof 2 ci 

-96 

3 


pofci 2 

-80 

53 


POF 2 Br 

-85 

32 


PSF, 

-149 

-52 


PSC1 3 

-35 

125 


PSF 2 Br 

-137 

36 



62 











TABLE 12. RECOMMENDED PHOSPHOROUS-CONTAINING COMPOUNDS 


Name 

Formula 

CASN 

Phase 

Normal 
boiling point 
(nbp °C) 

Vapor 

pressure 

(atm) 

(298 K) 

Heat 

capacity 

(J/K-mol) 

(298 K) 

Heat 

of 

vaporization 
(kJ at nbp) 

phosphorous trifluoride 


7783-55-3 

gas 

-101 

66 

58.7 

14.6 

phosphorous trichloride 

PC1 3 

7719-12-2 

liquid 

+76 

.16 

71.6 

30.5 

phosphorous bromide 
difluoride 

PF 2 Br 

15597-40-7 

gas 

-16 

4.5 

64+ 

23.9 

phosphoryl fluoride 

POF 3 

13478-20-1 

gas 

-40 

13.2 

68.8 

22.1 

phosphoryl chloride 

POCI 3 

10025-87-3 

liquid 

106 

.05 

84.9 

33.7 

phosphoryl bromide fluo¬ 
ride 

POF,Br 

14014-18-7 

liquid 

32 

.77 

75 f 

29.7 

'’’Estimated for this report by authors 






I. SILICON AND GERMANIUM FLAME INHIBITORS 

1. Justification for Consideration 

a. Family of Chemicals 

The focus of this section is on the compounds of silicon (Si) and, to a lesser extent, 
on some germanium (Ge) analogs. 1\vo of the other Group IVA elements, tin (Sn) and lead (Pb), are more 
appropriately classified as metals. Accordingly, compounds involving these elements will be considered in the 
section on metallic flame inhibitors (Section J). 

b. Properties 

The position of silicon directly below carbon in the periodic table suggests that the 
chemistry of these elements will be similar. Like carbon, silicon has a valency of four. There do not appear 
to be any compounds derived from the divalent state ^s^p 2 ) which are stable at room temperature [188]. 
However, the tendency to form divalent compounds increases with atomic weight so that Sn(II) and Pb(II) are 
relatively common. The Group FVA elements, other than carbon, do not form strong bonds with like atoms. 
The Si-Si bond, for example, is notably weaker (213 kJ/mole) than the Si-C bond (301 kJ/mole). 

All of the Group IVA elements form stable halides. The silicon and germanium 
analogs of the halons, in particular, would be expected to be effective flame inhibitors. This hypothesis was 
confirmed very early, at least with respect to silicon tetrachloride (SiCl 4 ), which was one of the chemicals 
tested in the Purdue Research Foundation study [25]. 

Obviously, if both families of chemicals behaved similarly with respect to all 
properties, including ozone depletion and global warming potentials, there would be no reason to continue 
this discussion. There are, however, significant differences in the behavior of these compounds in the 
atmosphere. Unlike the halons, all of the halosilanes readily hydrolyze in moist air [185]. An important 
consequence is that these compounds will undergo rapid decomposition in the troposphere and would 
therefore be expected to have correspondingly low potentials for ozone depletion and global warming. 
Unfortunately, this beneficial property is offset by the fact that hydrogen halides are produced in the hydrolysis 
of halosilanes. This effect is so pronounced that the presence of a single silicon-halogen bond in a molecule 
is sufficient to make its vapors corrosive and dangerous to breathe [185]. 

On the basis of this consideration, it is clear that along with fire suppression 
effectiveness, the propensity for hydrolysis will be the critical factor in determining whether any of the 
halosilanes will be viable candidates for the replacement of halons. The strategy, therefore, is to examine the 
chemical properties of these molecules in an effort to identify structural permutations which result in a balance 
between toxicity and tropospheric lifetime, while at the same time retaining their ability to inhibit flames. 

2. Past Flame Suppression Measurements 

A list of the results obtained by the Purdue Research Foundation [25] is duplicated in Lyon’s 
book on fire retardants [146] and in a review article by McHale [66]. The tabulated values are the volume 
percent of inhibitor corresponding to the peak in the flammability curve for a premixed n-heptane flame. The 
value reported for SiCl 4 was 9.9 percent. On the basis of this criterion, the flame suppression efficiency of 
SiCl 4 is between those for halon 1301 (CF 3 Br) and carbon tetrachloride (CC1 4 ), which were found to have 
peak values of 6.1 percent and 11.5 percent, respectively. 


64 



In an independent study, Lask and Wagner conducted flame velocity measurements for a series 
of additives including some halosilanes and related compounds [79,186,187]. The figure of merit was the 
volume percent of inhibitor required to reduce the burning velocity of a premixed (stoichiometric) n-hexane 
flame by 30 percent. On this basis, it was determined that the flame inhibition activity of SiCl 4 (0.56 percent) 
was comparable to Br 2 (0.7 percent) but considerably higher than CC1 4 (1.38 percent). 

It is difficult to make quantitative comparisons between data obtained from flammability curve 
and flame velocity measurements; however, the relative rankings of the inhibitor efficiencies would appear to 
be consistent. Thus, in addition to the compounds cited above, both sets of measurements have been made 
on chloroform (CHC1 3 ). The Purdue Research Foundation reported that the peak in the CHC1 3 flammability 
curve occurred at 17.5 percent. This implies that it is a less efficient inhibitor than either SiCl 4 or CC1 4 . The 
data on flame speeds justify a similar ranking. Lask and Wagner found that it took 1.97 percent of CHC1 3 
(in the mixture of fuel and oxidizer) to reduce the velocity by 30 percent of the value obtained for the clean 
flame. This was considerably more than the amounts of SiCl 4 and CC1 4 which were required to achieve the 
same velocity reduction. 

A smaller effect was observed for the silicon analog of chloroform - trichlorosilane (SiHCl 3 ), 
It required 2.9 percent of this halosilane to reduce the flame speed by 30 percent. Although this is 
considerably larger than the values obtained for some of the other halogenated compounds, it still indicates 
a significant degree of flame inhibition. Mixtures of 6.8 percent C0 2 , which is widely used as fire extinguishing 
agent, and 8 percent of N 2 , were needed to produce the same degree of inhibition. 

Flame velocity measurements were also reported by Lask and Wagner on two additional 
tetrachlorides of the Group IVA elements. The values of 0.50 percent and 0.19 percent were reported for 
GeCl 4 and SnCl 4 , respectively. The hierarchy for inhibition: SnCl 4 > GeCl 4 > SiCl 4 > CC1 4 was also found 
to apply to increases in the ignition temperatures of hydrocarbon/(0 2 + N 2 ) mixtures [187], No explanation 
for this trend has been found; however, it may have some relevance to the problem of interest. It is known 
that the susceptibility to hydrolysis of compounds involving Group IVA elements decreases with increasing 
atomic weight [188]. Thus, the most potent inhibitors may, in fact, be the least corrosive. Unfortunately, the 
drop-off may not be fast enough to yield practical benefits. Thus, SnCl 4 readily hydrolyzes and, as a 
consequence, it is highly corrosive [139]. On the other hand, lead chloride (PbCl 2 ), which does not hydrolyze, 
is a solid. The toxicity of this compound is probably due more to the presence of a heavy metal than to HC1. 
The possibility that some of the mixed halogermanes may be more suitable, however, should not be dismissed 
out of hand. 


The mechanism by which the halosilanes affect flame inhibition is probably similar, if not 
identical, to the halons (vide-supra). The same mechanism, however, cannot explain the flame inhibition which 
has been observed for tetramethylsilane (Si(CH 3 ) 4 ) [79,186,187]. This compound, which is commonly referred 
to as TMS, is routinely used as a standard in nuclear magnetic resonance (NMR) spectroscopy. Although 
TMS does have a large heat capacity, the effect is considered to be insufficient to explain the degree of flame 
inhibition exhibited by this compound [189], The magnitude of flame inhibition suggests, instead, that a 
chemical mechanism is operant. One possibility is that the alkyl groups play a role in promoting radical 
recombination of H and OH radicals in the flame. Lask and Wagner reported that a mixture of 1.5 percent 
of TMS was required to reduce the velocity of their stoichiometric n-hexane flame by 30 percent. This 
indicates that the inhibiting effect of TMS is comparable to that of CC1 4 . The possibility that halogenation 
of alkyl groups may increase the ability of alkysilanes to inhibit flames should be examined further. 


65 



3. 


Recommendations 


Many compounds of the Group IVA elements, including the halosilanes, halogermanes and 
alkylsilanes are effective flame inhibitors. The possibility that some of these compounds may be viable 
candidates for the replacement of halon fire extinguishing agents warrants further consideration. 

A variety of halogenated and alkylated compounds of silicon and germanium are listed, along 
with some of their relevant properties in Thble 13 [190,191], These particular compounds were selected 
because they were thought to represent structural permutations which are likely to affect susceptibility to 
hydrolysis and flame-inhibition activity. We recommend conducting corrosion and fire suppression efficiency 
testing of these compounds. This information will be useful in identifying silicon and germanium compounds 
which are viable alternatives for the existing halon extinguishing agents. 


J. METALLIC FLAME INHIBITORS 
1. Justification for Consideration 

The ability of alkali metal compounds to extinguish fires has been recognized for over 100 
years [32]. Sodium hydrogen carbonate (NaHC0 3 ) was probably the first dry chemical extinguishing agent. 
In addition to NaHC0 3 , formulations based on potassium hydrogen carbonate (KHC0 3 ), potassium chloride 
(KC1) and potassium oxalate (K^C^C^.fLO) are now in widespread use as extinguishing agents. More 
recently, it was discovered that certain heavy metal compounds, including tin chloride (SnCl 4 ), titanium 
chloride (TiCl 4 ), tetraethyl lead (Pb(C,H 5 ) 4 ), chromyl chloride (Cr0 2 Cl 2 ) and iron pentacarbonyl (Fe(CO) 5 ) 
are extremely effective flame inhibitors [189]. 

Metallics have physical and chemical properties which would seem to disqualify them from 
consideration as candidates for the replacement of halons. Since these compounds are usually not volatile, 
they would be expected to leave a residue, which in many cases, would be highly corrosive. Commercial 
extinguishing agents such as ”Purple-K" and "Super-K," which are based on KHC0 3 and KC1, respectively, 
leave a sticky residue which is destructive to electronics and other delicate equipment. Tbxicity is also a 
concern, particularly with heavy metals. The vapor pressure of Fe(CO) s is about 30 Tbrr, while SnCl 4 has a 
vapor pressure of about 20 Tbrr at ambient temperature. Thus, although relatively volatile, both compounds 
are highly toxic [139] and could not be deployed in the presence of people. 

The high degree of activity exhibited by metallic flame inhibitors is the basis for their further 
consideration. In one study, Hayes and Kaskan determined that NaHC0 3 was 20 times more effective than 
CH 3 Br, on a weight percent basis, in inhibiting premixed CH 4 /air flames [192]. Milne, Green and Benson 
arrived at a similar conclusion regarding the relative effectiveness of alkali metals and the halons after studying 
inhibition of CH 4 /air counterflow diffusion flames with CF 3 Br and "Purple K" [103], Three of the heavy metal 
compounds, Fe(CO) 5 , Cr0 2 CI 2 and Pb(C 2 H 5 ) 4 , are measurably more effective than the alkali metals and may 
be as much as two orders of magnitude more efficient than the halons in suppressing hydrocarbon flames 
[79,189]. Thus, it is conceivable that a solution containing a small amount of a metallic inhibitor dissolved 
in a volatile low toxicity solvent may result in a viable alternative to the halon extinguishing agents. Further 
elucidation of the operant principles of metallic flame inhibition will also serve to provide additional guidance 
in the search for halon replacements. 


66 



2. 


Past Flame Suppression Measurements 


a. Alkali Metal Inhibitors 

The fire suppression properties of alkali metal salts are well established. They are 
used as extinguishing agents in a wide variety of situations ranging from small household fires to large aircraft 
fuel fires. A large number of flame velocity and extinction measurements on these compounds have been 
reported in the literature [102,193,194,195,196,197,198,199,200,201], 

The consensus which has developed on the basis of these experiments is that flame 
inhibition by the alkali metals is the result of a chemical, rather than a thermal process. The distinction 
between chejnical and thermal suppression is clarified in Section II.B.4.b. Support for this hypothesis was 
obtained in early investigations of heptane and gasoline fires conducted by McChmy, Shroud and Lee [193] 
and independently by Lee and Robertson [194], A major conclusion of both studies was that there was no 
correlation between heat capacity and fire suppression effectiveness of the alkali metal salts. 

The effects of decomposition of the metallic extinguishing agents were not considered 
in these early studies and have yet to be fully resolved. Decomposition of alkali metal salts is a highly 
endothermic process and might be expected to facilitate suppression by reducing flame temperatures. This 
possibility was considered in a paper by Dodding, Simons and Stephens [195]. Their calculations indicated 
that the reduction in flame temperatures (at extinction) due to the presence of NaHC0 3 , varied from about 
35 to 75°C, depending on the average particle size of the powder. They went on to conclude that this was 
insufficient to account for the observed inhibition. Unfortunately, these results are open to interpretation, 
Ewing, Hughes and Carhart performed a similar analysis of the data for a wide variety of extinguishing agents, 
including some compounds of the alkali metals [116], They arrived at just the opposite conclusion - that the 
data is consistent with thermal suppression, provided that the energy requirements for decomposition of the 
extinguishing agents are taken into account. At the present time, the chemical hypothesis seems to be more 
widely accepted. However, the alternative point of view, that the dominant mode of suppression is thermal 
in nature, cannot be dismissed out of hand. (See the discussion in Section Il.B.4,b.) 

The dependence of flame suppression effectiveness on particle size is an important 
concern because it has a critical impact on the design of efficient delivery systems. Dolan observed that the 
surface area of powders was a major factor in determining the ability of alkali metal salts to inhibit explosions 
in methane/air mixtures [196]. More recently, Birchall demonstrated that in some alkali metal salts, there is 
an optimal particle size for reduction in the velocities of town gas/air diffusion flames [197], Birchall’s 
explanation for this behavior was based on the argument that small particles will decompose before penetrating 
the flame front, whereas large particles will pass through before they can volatilize. In the same study, Birchall 
confirmed earlier observations reported by Friedrich [198] and by Lee and Robertson [194] that the oxalates 
of the alkali metals are particularly effective. It is now generally recognized that the order of effectiveness for 
a specified alkali metal is: oxalate > cyanate > carbonate > iodide > bromide > chloride > sulfate > 
phosphate, whereas the hierarchy for the metals is Rb > K > Na > Li [189], 

The elucidation of the chemical mechanisms of alkali metal flame inhibition is a focal 
point of the research effort in this area. DeWitte, Vrebosch and van Tiggclen observed that alkali metal salts 
were incompletely volatilized in their downward burning CH 4 /(N 2 + 0 2 ) premixed flames [199]. On this basis, 
they reasoned that the observed reduction in flame velocities was due to the recombination of combustion 
radicals on surfaces of the alkali metal particulates. Although a mechanism based on catalytic radical 
recombination seemed very likely, there was considerable debate over whether this occurs on particles 
(heterogeneous) or in the gas phase (homogeneous). Evidence supporting a homogeneous mechanism was 
obtained from measurements of temperature changes in Na-inhibited premixed CH 4 /air flames by Iya, 
Wollowitz and Kaskan [200]. On the basis of spatially resolved concentration measurements, these researchers 


67 




TABLE 13. SILICON AND GERMANIUM FLAME INHIBITORS 


Name 

Formula 

CASN 

PHASE 

Normal 

Boiling 

Point 
(nbp °C) 

Vapor 

Pressure 

(atm) 

(298 K) 

Heat 

Capacity 

(J/K-mole) 

(298 K) 

Heat 

of 

Vaporization 
(kJ at nbp) 

tetrachlorosilane 

SiCI 4 

10026-04-7 

liquid 

58 

0.31 

145 

32 

trichlorofluorosilane 

SiCIjF 

14965-52-7 

gas 

15 

1.46 

92 

26 

tetrafluorosilane 

SiF 4 

7783-61-1 

gas 

-65 

1510 

74 

22 

bromotrifluorosilane 

SiBrFj 

14049-39-9 

gas 

-42 

10.5 

80 f 

18 

tribromofluorosilane 

SiBrjF 

18356-67-7 

liquid 

84 

0.11 

150 f 

34 

tetramethylsilane 

Si(CH 3 ) 4 

75-76-3 

liquid 

27 

0.95 

204 

27 

chlorotrimethylsilane 

Si(CH 3 ) 3 Cl 

75-77-4 

liquid 

57 

0.31 

175 T 

32 

trichloromethylsilane 

Si(CH 3 )Cl 3 

75-79-6 

liquid 

66 

0.23 

100 

31 

chloromethyltrimethylsilane 

Si(CH 3 ) 3 CH 2 Cl 

2344-80-1 

liquid 

98 

na 

200 1 

na 

tetrachlorogermane 

GeCl 4 

10038-98-9 

liquid 

83 

0.11 

150* 

35 

tetramethylgermane 

Ge(CH 3 ) 4 

865-52-1 

liquid 

44 

0.52 

200* 

30 

'Estimated for this report by authors 
na = not available 




were able to demonstrate that the degree of inhibition depended only on the concentration of Na atoms that 
had penetrated the reaction zone and was independent of initial particle size distribution and the nature of 
the ligand. This is compelling evidence in favor of a homogeneous mechanism as the efficiency of a 
heterogeneous catalyst would be expected to increase with the average size of the particles. 

Further progress in determining the mechanism of alkali metal flame inhibition was 
made by Friedman and Levy [102]. They observed that the introduction of elemental potassium in the fuel 
stream of their counterflow diffusion burner had no effect on the strength of CH 4 /air flames. (See the 
discussion of flame strength in Section ll.B.3.e.) This observation could be cited in support of the thermal 
suppression mechanism discussed above. Thus, it may be argued that elemental potassium has no effect on 
flame speeds because it does not decompose (as do potassium salts) and, therefore, does not withdraw heat 
from the flame. Friedman and Levy, however, reasoned that this would also be expected if there was a critical 
intermediate that could not be produced from elemental potassium in sufficient concentrations to inhibit the 
flame. The results of thermochemical calculations suggested that KOH was a good possibility. Potassium 
salts, for example carbonates and oxalates, decompose by forming K 2 0 which readily reacts with H 2 0 to give 
KOH. Friedman and Levy attributed the failure of elemental potassium to inhibit the flame to the slow rate 
of conversion of K to KOH via the termolecular reaction K + OH + X -* KOH + X. The third body, X, 
is needed to dissipate the kinetic energy generated in the collision [202], Independent results obtained by lya, 
Wollowitz and Kaskan support the contention that this termolecular reaction is too slow to have a significant 
effect on the chemistry of alkali metal-inhibited flames [200]. Calculations made by these researchers using 
a kinetic model indicated that the reaction Na + OH + X -* NaOH + X could not account for the observed 
OH concentration profiles. They went on to propose the alternative: Na + H 2 0 -* Na-H 2 0; Na-H 2 0 + 
OH -* H 2 0 + NaOH. Although this mechanism removes the discrepancy between predicted and observed 
OH concentrations, it is inconsistent with Friedman and Levy’s original observation that elemental K, and by 
implication the other alkali metals as well, is not an effective flame inhibitor. 

The details of the mechanism of alkali metal flame inhibition, and in particular 
whether the catalysis is homogeneous or heterogenous, have not been fully resolved. Advocates of a 
homogeneous mechanism tend to agree that the critical intermediate is the alkali metal hydroxide (MOH). 
Flame inhibition is thought to result from removal of hydrogen radicals via the reaction MOH + H -* M + 
H 2 0 [102,197,200,201]. Note that the inhibitor does not regenerate itself in this mechanism. It has been 
postulated, however, that a single functioning of each inhibitor molecule may be sufficient because of the low 
concentration of hydrogen radicals in hydrocarbon flames [102]. This mechanism also accounts for the 
surprising observation that halide salts are amongst the least effective alkali metal inhibitors [79,189], The 
poisoning effect is due to the tendency of the halogen to tie-up the metal, thereby reducing the concentration 
of MOH in the flame. 


b. Heavy Metal Inhibitors 

Bulewicz and Padley demonstrated that low levels (ppm) of a wide range of metallic 
elements including Mg, Cr, Mn, Sn, U and Ba have a measurable effect on the recombination of hydrogen 
radicals in premixed H 2 /(0 2 + N 2 ) flames [203]. Despite this observation, which would appear to be good 
evidence that the ability to inhibit flames is widespread in metals, the data on metallic compounds which do 
not contain either Na or K is very limited. 

A paper by Lask and Wagner appears to be the first serious attempt to investigate 
the influence of heavy metals on flames [79]. The effects of a wide range of additives on the flame velocities 
of premixed (stoichiometric) n-hexane flames were measured. These additives included TiCl 4 , SnCl 4 , Fe(CO) s , 
Pb(C 2 H s ) 4 , Cr0 2 Cl 2 , as well as a series of phosphorous and silicon compounds. The halogens, Br 2 and Cl 2 , 
and halogenated hydrocarbons like CH 3 C1 and CC1 4 were tested as well. An extended list which was compiled 
for a report [186] appears in a later paper by Morrison and Scheller [187]. Lask and Wagner reported that 
the flame inhibition exhibited by Fe(CO) 5 , Pb(C 2 H 5 ) 4 and Cr0 2 Cl 2 was at least an order of magnitude greater 


69 




than what was observed with any of the other compounds that were tested. It required less than 0.02 percent 
(by volume) of these compounds in the fuel mixture to reduce the flame speed by 30 percent. The other two 
metallics, TiCl 4 and SnCl 4 , were also effective requiring about 0.2 percent to achieve a 30 percent velocity 
reduction. For comparison, it took a mixture of 0.7 percent of Br 2 and almost 1.4 percent of CC1 4 to produce 
the same results. 


The high degree of flame suppression activity exhibited by Fe(CO) 5 had been 
previously reported by Jost, Bonne and Wagner [204], At the same time, it was noted that the catalytic 
efficiency depends on the concentration Fe(CO) 5 in the flame. This suggests that there may be two distinct 
mechanisms. A heterogeneous process may be operant at concentrations which exceed the saturation pressures 
of iron containing species (e. g. FeO) in the flame, whereas at lower concentrations of Fe(CO) 5 a homoge¬ 
neous reaction predominates [205], Hastie cites the observed fall-off in activity at higher concentrations as 
an indication that heterogeneous recombination may be less efficient [189], 

There is, however, no direct evidence that any of these metallics exhibit homogeneous 
flame inhibition. Arguments for a heterogeneous mode have cited the streaky white appearance (indicative 
of the presence of nonvolatile oxides) and luminosity which are evident in transition metal inhibited flames 
[203]. It is known that Fe(CO) 5 , in particular, produces finely divided oxides in flames [73,206]. In addition, 
particle temperatures in considerable excess of the adiabatic flame temperature (by as much as 400 k) have 
been observed [189,203]. This effect has been attributed to the heat released during surface recombination 
reactions [203,207,208]. Further evidence of heterogeneous inhibition was presented by Miller [209] and by 
Miller and Vree [210]. These investigators assigned continuum emission bands which were observed in low 
pressure methane flames (both diffusion and premixed) inhibited by Fe(CO) 5 and Cr0 2 Cl 2 to hot particles. 

In a more recent study, Vanpee and Shirodkar examined the inhibition exhibited by 
a wide variety of metal chlorides, acetates and acetylacetonates using a counterflow diffusion burner with an 
ethanol/(air + N 2 ) flame [211]. After dissolving the compounds in ethanol, the solutions were atomized in 
a sonic nozzle and the resulting mist was mixed with the oxidizer. In all cases, with the exception of aluminum 
acetylacetonate, the solutions were effective inhibitors. The figure of merit was the increase in oxygen 
concentration in the oxidizer stream per unit concentration of the additive at extinction. The transition metal 
compounds were most effective when compared on a molar basis. However, when compared on a mass basis 
the alkali metal compounds were superior. Surprisingly, Fe(CO) 5 was rated comparatively low using either 
criterion. 


3. Recommendations 

Compounds of the alkali metals, as well as some heavy metals including Sn, Ti, Pb, Cr and 
Fe are flame inhibitors. Some of these compounds, for example Fe(CO) s , Pb(C 2 H 5 ) 4 , CrO 2 01 2 , may be as 
much as two orders of magnitude more effective than the halons in inhibiting hydrocarbon flames. As is also 
the case with the halons, the dominant mechanism of metallic flame inhibition is thought to be chemical - 
involving the catalytic recombination of hydrogen and hydroxyl radicals, rather than thermal in nature. 
Although inconclusive, the evidence favors the hypothesis that the catalytic action of alkali metal inhibitors 
is homogeneous, whereas a heterogeneous mechanism appears to predominate in heavy metal inhibited flames. 
The efficiency of metallic inhibitors depends on the physical states and relative stabilities of the intermediates 
(for example metal oxides and hydroxides) which perform the catalytic function. Properties, such as volatility 
and average particle size, are also important because they determine the ability of the agent to penetrate the 
flame front. 


An intensive search has not been successful in identifying low toxicity, high volatility, metallic 
flame inhibitors. The unfortunate conclusion is that pure metallic compounds are not acceptable halon 
replacements. This does not, however, preclude the possibility that a solution consisting of a small amount 


70 



of a metallic inhibitor dissolved in a volatile solvent may make an effective fire extinguishing agent. The 
prospect of using Fe(CO) 5 as a fire suppression additive was first considered by Jost, Bonne and Whgner [204]. 
Mixtures consisting of halogenated fire inhibitors adsorbed on metal powders have also been examined [212]. 

A list of the metallic flame inhibitors recommended for further study is presented in Thble 
14 along with some of their relevant properties [190,191,213,214]. All of these chemicals, with the exception 
ofPb(C 2 H 5 ) 4 , are listed in the 1990 - 1991 Aldrich Catalogue. Many of these compounds, including the alkali 
metal acetates, Fe(CO) 5 , and both potassium and chromium acetylacetonates, are soluble in ethanol. Flame 
inhibition measurements conducted by Vanpee and Shirodkar [211] suggest that these solutions might be 
effective as fire extinguishing agents. Additional experiments, involving solvents which are themselves flame 
inhibitors, are warranted. 


K. INERT GASES 

1. Justification for Consideration 

In the present context, gases are considered to be inert if they do not undergo chemical 
transformation when they pass through a flame. As a result, these compounds can have no chemical effect 
on the flame chemistty and therefore must effect suppression by purely physical mechanisms. This point is 
discussed extensively in Section Il-B. Comparitive studies involving inert and chemical agents should help to 
further elucidate the differences between physical and chemical suppression. 

2. Past Flame Suppression Measurements 

There are numerous flame inhibition and suppression investigations utilizing inert species. 
These will not be discussed in detail. Several were discussed in Section II.B. It should suffice to point out 
that C0 2 fire extinguishers are widely employed for local firefighting and that their effectiveness has been the 
subject of a great deal of investigation. Nitrogen has been proposed as an inerting agent [215,216,217] but 
has not been widely used. The third agent to be recommended here, argon, has not been tested as a practical 
firefighting agent. 

3. Recommendations 

The primary reason for including inert compounds on the list is to establish a baseline for 
purely physical fire suppression and to provide a basis for the comparison of chemical agents. It is recognized 
that these agents will not be suitable as replacements for all situations where the current commercial halons 
are used. 


Three inert agents, N 2 , C0 2 , and Ar, have been chosen for inclusion on the list. These 
compounds are relatively inexpensive and have a range of heat capacities and densities. Table 15 summarizes 
the molecular properties of these three species. 

As discussed in Section III.B.l carbon tetrafluoride is also included on the list. This species 
is believed to be inert at flame temperatures and will thus provide an example of a still more complicated inert 
species for comparison purposes. 


71 




TABLE 14. METALLIC INHIBITORS 


Name 

Formula 

CASN 

PHASE 

Normal 

Boiling 

Point 
(nbp °C) 

Vapor 

Pressure 

(atm) 

(298 K) 

Heat 

capacity 

(J/K-mol) 

(298 K) 

Heat 

of 

vaporization 
(kJ at nbp) 

sodium hydrogen carbonate 

NaHC0 3 

144-55-8 

solid 

decomposes 

0 f 

88 

na 

sodium acetate 

NaC 2 H 3 0 2 

127-09-3 

solid 

decomposes 

0 f 

80 

na 

potassium hydrogen carbonate 

KHCOj 

298-14-6 

solid 

decomposes 

0 f 

na 

na 

potassium oxalate 

K 2 C 2 O 4 • H 2 O 

6487-48-5 

solid 

decomposes 

0 f 

na 

na 

potassium acetate 

k 2 c 2 h 3 o 2 

127-08-2 

solid 

decomposes 

o f 

na 

na 

potassium acctylacetonate 

kc 5 h 7 o 2 *=/ 2 h 2 o 

57402-46-7 

solid 

decomposes 

na 

na 

na 

chromium acctylacetonate 

Cr ( C 5 H 7°2)3 

21679-31-2 

solid 

340 

na 

na 

na 

chromyl chloride 

Cr0 2 Cl 2 

14977-61-8 

liquid 

116 

0.024 

na 

41 

tin (IV) chloride 

SnCl 4 

7646-78-8 

liquid 

114 

0.030 

165 

37 

titanium (IV) chloride 

t.ci 4 

7550-45-0 

liquid 

136 

0.015 

145 

39 

tetraethyl lead 

Pb(C 2 H 5 ) 4 

78-00-2 

liquid 

~ 200 

0.00055 

na 

54 

iron pentacarbonyl 

Fe(CO ) 5 

13463-40-6 

liquid 

103 

0.040 

241 

38 

^Estimated for this report by authors 
na = not available 





GASES 


Normal 
boiling point 
(nbp °C) 

Vapor 

pressure 

(atm) 

(298 K) 

Heat 

capacity 

(J/K-mol) 

(298 K) 

Heat 

of 

vaporization 
(Id at nbp) 

-195.8 

na 

29.3 

5.6 

-78.5(sub) 

29.9 

37.7 

15.3 

-185.4 

na 

20.9 

6.5 


















SECTION IV 


CONCLUSIONS AND RECOMMENDATIONS 


A total of 103 compounds are included on the list which has been generated during the course of this 
study. Table 16 list the names of these compounds. 

A quick review of these compounds reveals that a wide range of chemical families has been 
recommended for study. A more careful check reveals that the list meets the goals of the project by including 
species which are likely to be considered as immediate replacements for the current commercial halons, while 
at the same time, allowing the development of principles for fire suppression and ozone depletion which can 
be applied to a long-term search for replacements. 

It is recommended that these compounds be tested in a very selective series of experimental 
investigations chosen to maximize the knowledge to be gained using the insights which have been used in the 
development of the list. For instance, certain halons are included on the list to characterize their behavior 
in the troposphere. Particular emphasis should be focused on whether or not these molecules will be attacked 
by OH radicals or undergo photolysis at wavelengths greater than 300 nm. The effects of changes in chemical 
structure on fire suppression and toxicity behavior must also be characterized. Similarly, compounds 
containing phosphorous and the metal compounds have been added to the list because literature reports 
indicate that these compounds arc unusually effective fire suppressants. Efforts should be made to characterize 
the fire suppression effectiveness of these agents, and, if possible, characterize their mechanisms of chemical 
flame inhibition. Such insights will provide the knowledge base required for the intelligent design of alterna¬ 
tive chemical fire suppressants. 

The literature review performed as part of this investigation suggests that there has been no systematic 
search for new fire suppressants since the early 1950s. For this reason the review of the literature and 
recommendation of initial chemicals to investigate was a necessary and appropriate early step in the search 
for suitable alternatives. It is unlikely that this search will be easy, and the effort may not even lead to 
chemicals which are as effective and safe as the current commercial halons. On the other hand, the need for 
effective agents is so great that alternatives will be necessary for certain critical applications. The authors 
believe that this document will serve as the guidepost in the important effort to develop alternative chemicals 
for use in place of the current commercial halons. 


74 



TABLE 16. COMPLETE LIST OF RECOMMENDED COMPOUNDS 


perfluoromethane 

perfluoroethane 

perfluoropropane 

perfluoro-n-bulane 

perfluorocyclobutane 

trifluoromethane 

pentafluoroethane 

1 . 1 . 1 . 2 - tetrafluoroethane 
dibromodifluoromethane 

2 . 2 - dibromo-l,l,l, 2 -tetrafluoroethane 
chlorodifluoromethane 

1 . 1 . 1 - trichlorethane 

2 . 2 - dichloro-l , 1,1 -trifluoroethane 
2 -chloro-l,l,l, 2 -tetrafluoroethane 

1 . 1 - dichloro-l-fluoroethane 

1- chloro-l,l-difluoroethane 
bromodifluoromethane 
bromochlorofluoromethane 

2 - bromo- 2 -chloro-l, 1 , 1 -trifluoroethane 
2 -bromo-l -chloro- 1 , 2 , 2 -trifluoroethane 

1- bromo-l,l, 2 , 2 -tetrafluoroethane 

2 - bromo-l , 1,1 -trifluoroethane 

1 . 2 - dibromo-l,l , 2 -trifluoroethane 

1. 2 - dibromo-l,l-difluoroethane 

1 . 2 - dibromo-l , 2 -difluoroethane 
l-bromo-l,l,2,3,3,3-hexafluoropropane 

1.3- dibromo-l,l,3,3-tetrafluoropropane 

2.2- dibromo-l,l,3,3-tetrafluoropropane 
l-bromo-l,l,3,3,3-pentafluoropropane 
hexafluoracetone 

trifluoroacetic anhydride 
bis(perfluoroisopropyl) ketone 
methyltrifluoroacetate 

3- bromo-l ,1,1 -trifluoropropanone 
bromopentafluoroacetone 
bromomethyltrifluoroacetate 
perfluoropropene 
perfluorobutene -2 
perfluorotoluene 

1.1.3.3.3- pentafluoropropene-l 
3,3,3 trifluoropropene 

1 . 2 - bis(perfluoro-n-butyl )ethylene 

3-bromoperfluoropropene 

1 -bromoperfl uoropropene 

1 . 2 - bis( perfluoromethyl)ethy lene 
l-bromoperfluoromethyi- 2 -perfluoromethylethylene 

1- bromo-bis(perfluoromethy!)ethylene 
tetris(perfluoromethyl)ethyiene 
tetrafluorodimethyl ether 
pentafluorodimethyl ether 

2 - chloro-l-(difluoromethoxy)-l,l, 2 -trifluoroethane 
isoflurane 


perfluoro-2-butyltetrahydrofuran 

bis(bromodifluoroethyl) ether 

1 -bronio-1,1,3,3,3-pentafl uorodimethy 1 ether 

bromoenflurane 

octafluorofuran 

3-bromoperfluorofuran 

bis(perfluoromethyl) thioether 

tris(perfluoromethyl) amine 

iodotrifluoromethane 

chlorodifluoroiodomethane 

l-bromo-l,l,2,2-tetrafluoro-2-iodoethane 

l,l,2,2-tetrafluoro-l,2-diiodoethane 

iodomethane 

iodoethane 

1 -iodopropane 

l,l,l,2,2,3,3-heptafluoro-3-iodopropane 

sulfur fluoride 

sulfur chloride fluoride 

sulfur bromide fluoride 

phosphorous trifluoride 

phosphorous trichloride 

phosphorous bromide difluoride 

phosphoryl fluoride 

phosphoryl chloride 

phosphoryl bromide fluoride 

tetrachlorosilane 

trichlorofluorosilane 

tetrafluorosilane 

bromotrifluorosilane 

tribromofluorosilane 

tetramethylsilane 

chlorotrimethylsilane 

trichloromethylsilane 

chloromethyltrimethylsilane 

tetrachlorogcrmane 

tctramethylgermane 

sodium hydrogen carbonate 

sodium acetate 

potassium hydrogen carbonate 

potassium oxalate 

potassium acetate 

potassium acetylacetonate 

chromium acetylacetonate 

chromyl chloride 

tin (IV) chloride 

titanium (IV) chloride 

tetraethyl lead 

iron pentncarbonyl 

nitrogen 

carbon dioxide 

argon 


75 




REFERENCES 


[1] Klein, H.A, "New Fire Extinguishing Agents for Aircraft," NFPA Quarterly , vol. 43, pp, 235-240, April 
1950. 

[2] Ford, C.L., "An Overview of Halon 1301 Systems," in Halogenated Fire Suppressants (Gann, R.G., 
Editor), pp. 1-63, American Chemical Society, Washington, D. C., 1975. 

[3] "Chlorobromomethane," NFPA Quarterly , vol. 42, pp. 46-50, July 1948. 

[4] Strasiak, R.R., The Development History of Bromochloromethane (CBL WADC Technical Report 
53-279, Wright Air Development Center, Wright-Patterson AFB, OH, January 1954. 

[5] Downing, R.C., and Eiseman, Jr., B.J., "Halogenated Extinguishing Agents-Part I, Research at the 
Jackson Laboratory," NFPA Quarterly , vol. 45, pp. 110-119, October, 1951. 

[6] Malcolm, J.E., "Halogenated Extinguishing Agents-Part II, Research at the Corps of Engineers’ 
Laboratories," NFPA Quarterly, vol. 45, pp. 119-131, October, 1951. 

U\ Anderson, S.O., "Halons and the Stratospheric Ozone Issue," Fire Journal , vol. 81, pp. 56-62,118-125, 
May/June 1987. 

[8] "The Halon/Ozone Issue," Fire Surveyor, vol. 18, pp. 13-15, October 1989. 

[9] Grant, C.C., "Halons and the Ozone Layer: An Overview," Fire Journal, vol. 83, pp. 58-62 and 78-81, 
September, 1989. 

[10] Molina, M.J. and Rowland, F.S., "Stratospheric Sink for Chlorfluoromethanes: Chlorine Atom- 
Catalyzed Destruction of Ozone," Nature , vol. 249, pp. 810-812, 28 June 1974. 

[11] Farman, J.C., Gardiner, B.G., and Shanklin, J.D., "Large Losses of Tbtal Ozone in Antarctica Reveal 
Seasonal C10 X /N0 X Interaction," Nature , vol. 315, pp. 207-210, 16 May 1985. 

[12] Stolarski, R.S., Krueger, AJ., Schoeberl, M.R., McPeters, R.D., Newman, P.A, and Alperl, J.C., 
"Nimbus 7 Satellite Measurements of the Springtime Antarctic Ozone Decrease," Nature, vol 322, pp. 
808-811, 28 August 1986. 

[13] "Stratospheric Ozone is Decreasing," Science , vol. 239, pp. 1489-1491, 25 March 1988. 

[14] Hammitt, J.K., Camm, F., Connell, P.S., Mooz, W.E., Wolf, KA, Wuebbles, D.J., Bamezai, A, 
"Future Emission Scenarios for Chemicals That May Deplete Stratospheric Ozone," Nature, vol. 330, 
pp. 711-716, 24/31 December 1987. 

[15] Yung, Y.L., Pinto, J.P., Watson, R.T., and Sander, S.P., "Atmospheric Bromine and Ozone 
Perturbations in the Lower Stratosphere," Journal of the Atmospheric Sciences , vol. 37, pp. 339-353, 
February 1980. 

[16] McElroy, M.B., Salawitch, R.J., Wofsy, S.C., and Logan, J.A., "Reduction of Antarctic Ozone Due to 
Synergistic Interactions of Chlorine and Bromine," Nature , vol. 321, pp. 759-762, 19 June 1986. 


76 



117] 


Singh, O.N., Borchers, R., Fabian, P., Lai, S., and Subbaraya, B.H., "Measurements of Atmospheric 
BrO x Radicals in the Tropical and Mid-Latitude Atmosphere," Nature , vol. 334, pp. 593-595, 18 
August 1988. 

[18] Published Proceedings of the Symposium An Appraisal of Haloeenated Fire Extinguishing Agents. 
Washington, DC, 11-12 April 1972, National Academy of Sciences, 1972. 

[19] Standardization and Data Management Newsletter. No. 32, Defense Quality and Standardization 
Office, January, 1990. 

[20] DiNenno, P. and Budnick, E., Halon 1301 Discharge Testing: A Technical Analysis . U.S. 
Environmental Protection Agency and National Fire Protection Research Foundation, October 1988. 

[21] Grant, C.C., Enclosure Integrity Procedure . National Fire Protection Research Foundation, Quincy, 
MA, January 1989. 

[22] Presentation Material from The Tyndall Conference on "Halon, the Ozone Layer, and Research on 
Alternative Chemicals," U.S. Environmental Protection Agency and United State Air Force, Tyndall 
AFB, Panama City, FL, 15-17 November 1988. 

[23] Summary of The Tyndall Conference on "Halon, the Ozone Layer, and Research on Alternative 
Chemicals," U.S. Environmental Protection Agency and United State Air Force, Tyndall AFB, Panama 
City, FL, 15-17 November 1988. 

[24] Replacement Fire Suppressants: A Research Strategy and Plan . Technical Committee of the Halon 
Alternatives Research Consortium, 1 August 1989. 

[25] Final Report on Fire Extinguishing Agents for the Period September 1.1947 to June 30.1950. Purdue 
Research Foundation and Department of Chemistry, Purdue University, West Lafayette, IN, July 1950. 

[26] MacEwen, J.D., "Tbxicology of Pyrolysis Products of Halogenated Agents," An Appraisal of 
Haloeenated Fire Extinguishing Agents . Washington, DC, 11-12 April 1972, pp. 53-59, National 
Academy of Sciences, 1972. 

[27] Rainaldi, N., "Appraisal of Halogenated Fire Extinguishing Agents," An Appraisal of Halogenated Fire 
Extinguishing Agents, Washington, DC, 11-12 April 1972, pp. 79-90, National Academy of Sciences, 
1972. 

[28] Fielding, G.H., Woods, F.J., and Johnson, J.E., "Halon 1301: Mechanism of Failure to Extinguish 
Deep-Seated Fires," Journal of Fire and Flammability , vol. 6, pp. 37-43, January 1975. 

[29] "General Discussion," in Halogenated Fire Suppressants (Gann, R.G., Editor), pp. 436-438, American 
Chemical Society, Washington, D. G, 1975. 

[30] Johns, L.C., “World "Rends in the Use of Halons,” Fire Engineers Journal , vol. 38, pp. 33-38, 
September, 1978. 

[31] Ashmore, F., "How Hazardous Is Halon 1301?," Fire Prevention , vol. 209, pp. 34-36, May 1988. 

[32] Cote, A. and Bugbee, P., Principles of Fire Protection, p. 208, National Fire Protection Association, 
Quincy, MA, 1988. 


77 







[33] Thpscott, R.E., Morehouse, E.T Jr., Next-Generation Fire Extinguishing Agent, Phase I-Suppression 
Concepts. ESL-TR-87-03, Engineering and Services Laboratory, Air Force Engineering and Services 
Center, Tyndall AFB, FL, July 1987. 

[34] Williams, F.A., Combustion Theory . 2nd ed., Benjamin/Cummings, Menlo Park, CA, 1985. 

[35] Glassman, I, Combustion. 2nd Ed., Academic Press, New York, 1987. 

[36] Flaessler, W.M., Fire: Fundamentals and Control . Marcel Dekker, New York, 1989. 

[37] Gardiner, Jr., W.C., Combustion Chemistry. Springer-Verlag, New York, 1984. 

[38] Cottrell, Jr., W.H., The Book of Fire . Mountain Press, Missoula, MT, 1989. 

[39] Berl, W.G., "Foreword," Fire Research Abstracts and Reviews, vol. 1, p. iii, September 1958. 

[40] Fox, R.W. and McDonald, A.T., Introduction to Fluid Mechanics , pp. 43-44, John Wiley & Sons, New 
York, 1973. 

[41] Faraday, M.; The Chemical History of a Candle Flame , Thomas Y. Crowell Co., New York, 1957. 

[42] Stull, D. R. and Prophet, H., JANAF Thermochemical Thbles . Second Edition, National Standard 
Reference Data System, 1971. 

[43] Huggett, C., "Combustion Processes in the Aerospace Environment," Aerospace Medicine, vol. 40, pp. 
1176-1180, November 1969. 

[44] Huggett, C., "Habitable Atmospheres Which Do Not Support Combustion," Combustion and Flame. 
vol. 20, pp. 140-142, 1973. 

[45] Sheinson, R. S., Hahn, J. E., Geary, K., and Williams, F. W., "Quantification of Physical Fire 
Suppression of Heptane Pool Fires," Abstract for paper presented at the 1977 Fall Technical Meeting 
of the Eastern Section of the Combustion Institute, East Hartford, CT, 10-11 November 1977. 

[46] Sheinson, R.S., Penner-Hahn, J.E., and Indritz, D., "The Physical and Chemical Action of Fire 
Suppressants," Fire Safety Journal , accepted for publication. 

[47] Tlicker, D.M., Drysdale, D.D., and Rasbash, D.J., "The Extinction of Diffusion Flames Burning in 
Various Oxygen Concentrations by Inert Gases and Bromolrifluormethane," Combustion and Flame , 
vol. 41, pp. 293-300, June 1981. 

[48] McHale, E.T., "Life Support Without Combustion Hazards," Fire Technology , vol. 10, pp. 15-24, 
February 1974. 

[49] Williams, F.A., Combustion Theory , 2nd ed., pp. 131-135, Benjamin/Cummings, Menlo Park, CA, 
1985. 

[50] Andrews, G.E., Bradley, D., and Lwakabamba, "Tbrbulence and Thrbulent Flame Propagalion-A 
Critical Appraisal," Combustion and Flame , vol. 24, pp. 285-304, 1975. 

[51] Bilger, R.W., "Turbulent Jet Diffusion Flames," Progress in Energy and Combustion Science, vol. 1, 
pp. 87-109, 1976. 


78 



Peters, N., "Laminar Flamelet Concepts in Tbrbulent Combustion," Twenty-First Symposium 
(International) on Combustion, pp. 1231-1250, The Combustion Institute, Pittsburgh, 1988. 


Gardiner, Jr., W.C., "Introduction to Combustion Modeling," in Combustion Chemistry (ed. W.C. 
Gardiner, Jr.), pp. 1-19, Springer-Verlag, New York, 1984. 

Dixon-Lewis, G., "Computer Modeling of Combustion Reactions in Flowing Systems with Transport," 
in Combustion Chemistry (ed. W.C. Gardiner, Jr.), pp. 21-125, Springer-Verlag, New York, 1984. 

Fryburg, G. Review of Literature Pertinent to Fire-Extinguishing Agents and Basic Mechanisms 
Involved in Their Action . NACA Tfechnical Note 2102, National Advisory Committee For 
Aeronautics, Washington, DC, May 1950. 

Malcolm, J.E., Interim Report-Vaporizing Fire Extinguishing Agents. Report 1177, Engineer 
Research and Development Laboratories, U.S. Army Corps of Engineers, Fort Belvoir, VA, 18 August 
1950. 

Friedman, R. and Levy, J.B., Survey of Fundamental Knowledge of Mechanisms of Action of Flame- 
Extinguishing Agents. WADC Tfechnical Report 56-568, Wright Air Development Center, Air 
Research and Development Command, Wright-Patterson AFB, OH, January 1957. 

Friedman, R. and Levy, J.B., Survey of Fundamental Knowledge of Mechanisms of Action of Flame- 
Extinguishing Agents, WADC Tfechnical Report 56-568, Supplement I, Wright Air Development 
Center, Air Research and Development Command, Wright-Patterson AFB, OH, September 1958. 

Friedman, R. and Levy, J.B., Survey of Fundamental Knowledge of Mechanisms of Action of Flame- 
Extinguishing Agents. WADC Tfechnical Report 56-568, Supplement II, Wright Air Development 
Center, Air Research and Development Command, Wright-Patterson AFB, OH, April 1959. 

Friedman, R. and Levy, J.B., "Mechanisms of Action of Chemical Agents for Flame Extinguishment," 
Fire Research Abstracts and Reviews, vol. 1, pp. 81-88, May 1959. 

Friedman, R., "Survey of Chemical Inhibition in Flames," Fire Research Abstracts and Reviews, vol. 
3, pp. 128-132, September, 1961. 

Skinner, G.B., Survey of Chemical Aspects of Flame Extinguishment. ASD Tfechnical Report 61-408, 
Aeronautical Systems Division, Wright-Patterson AFB, OH, December 1961. 

Skinner, G.B., Survey of Recent Research on Flame Extinguishment , ASD Tfechnical Report 61-408, 
Supplement I, Aeronautical Systems Division, Wright-Patterson AFB, OH, December 1962. 

Skinner, G.B., Survey of Recent Research on Flame Extinguishment . ASD Tfechnical Report 61-408, 
Supplement II, Aeronautical Systems Division, Wright-Patterson AFB, OH, February 1964. 

Fristrom, R.M. "Combustion Suppression (A Literature Survey with Commentary)," Fire Research 
Abstracts and Reviews, vol. 9, pp. 125-152, 1967. 

McHale, E.T, "Survey of Vapor Phase Chemical Agents for Combustion Suppression," Fire Research 
Abstracts and Reviews, vol. 11, pp. 90-104, 1969. 


Halogenated Fire Suppressants (Gann, R.G., Editor), American Chemical Society, Washington, D. 

C 1975 




[ 68 ] 


Miller, M.J. and Kenney, J.M., Bibliography of Halon Literature . Factory Mutual Research 
Corporation, May 1978. 


[69] Coward, H.F. and Jones, G.W., Limits of Flammability of Gases and Vapors. Bulletin 503, Bureau of 
Mines, U.S. Department of Interior, Washington, DC, 1952. 

[70] Zabetakis, M.G., Flammability Characteristics of Combustible Gases and Vapors. Bulletin 627, Bureau 
of Mines, U.S. Department of Interior, Washington, DC, 1965. 

[71] Lovachev, L.A., "Flammability Limits-A Review," Combustion Science and Tfechnology, vol. 20, pp. 
209-224, 1979. 

[72] Belles, F.E., Chemical Action of Halogenated Agents in Fire Extinguishment , NACA Technical Note 
3565, National Advisory Committee For Aeronautics, Washington, DC, September 1955. 

[73] Creitz, E.C., "A Literature Survey of the Chemistry of Flame Inhibition," Journal of Research of the 
National Bureau of Standards , vol. 74A, pp. 521-530, July-August 1970. 

[74] Burdon, M.C., Burgoyne, J.H., and Weinberg, F.J., "The Effect of Methyl Bromide on the Combustion 
of Some Fuel-Air Mixtures," Fifth Symposium (International) on Combustion , pp. 647-651, Reinhold, 
New York, 1955. 

[75] Biordi, J.C., Lazzara, C.P., and Papp, J.F., "Flame Structure Studies of CF 3 Br-Inhibited Methane 
Flames. II. Kinetics and Mechanisms," Fifteenth Symposium (International! on Combustion, pp. 917- 
932, The Combustion Institute, Pittsburgh, 1974. 

[76] Biordi, J.C., Lazzara, C.P., and Papp, J.F., "Flame Structure Studies of CF 3 Br-Inhibited Methane 
Flames. 3. The Effect of 1 percent CF 3 Br on Composition, Rate Constants, and Net Reaction 
Rates," The Journal of Physical Chemistry , vol. 81, pp. 1139-1145, 16 June 1977. 

[77] Biordi, J.C., Lazzara, C.P., and Papp, J.F., "Flame Structure Studies of CF 3 Br-lnhibited Methane 
Flames. 4. Reactions of Inhibitor-Related Species in Flames Containing Initially 1.1% 
Bromotrifluoromethane," The Journal of Physical Chemistry , vol. 82, pp. 125-132, 26 January 1978. 

[78] Rosser, W.A., Wise, H., and Miller, J., "Mechanism of Combustion Inhibition by Compounds 
Containing Halogen," Seventh Symposium (International! on Combustion , pp. 175-182, Butterworths, 
London, 1958. 

[79] Lask, G. and Wagner, H.Gg., "Influence of Additives on the Velocity of Laminar Flames," Eight 
Symposium (International! on Combustion, pp. 432-438, Williams and Wilkins, Baltimore, 1962. 

[80] Halpern, C. "Effect of Some Halogenated Hydrocarbons on the Flame Speed of Methane," Journal 
of Research of the National Bureau of Standards , Vol. 70A, pp. 133-141, March-April 1966. 

[81] Homann, KH. and Poss, R., "The Effect of Pressure on the Inhibition of Ethylene Flames," 
Combustion and Flame, vol. 18, pp. 300-302, April 1972. 

[82] Moran, Jr., H.E. and Bertschy, A.W., Flammability Limits for Mixtures of Hydrocarbon Fuels. Air, 
and Halogen Compounds, NRL Report 4121, Engineering Research Branch, Chemistry Division, 
Naval Research Laboratory, Washington, DC, 25 February 1953. 


80 


[83] 


Dixon-Lewis, G., "Mechanism of Inhibition of Hydrogen-Air Flames by Hydrogen Bromide and Its 
Relevance to the General Problem of Flame Inhibition,” Combustion and Flame, vol. 36, pp. 1-14, 
1979. 

[84] Day, M.J., Stamp, D.V, Thompson, K., and Dixon-Lewis, G., "Inhibition of Hydrogen-Air and 
Hydrogen-Nitrous Oxide Flames by Halogen Compounds," Thirteenth Symposium (International! on 
Combustion , pp. 705-712, The Combustion Institute, Pittsburgh, 1971. 

[85] Westbrook, C.K., "Inhibition of Laminar Methane-Air and Methanol-Air Flames by Hydrogen 
Bromide," Combustion Science and Technology , vol. 23, pp. 191-202, 1980. 

[86] Westbrook, C.K., "Inhibition of Hydrocarbon Oxidation in Laminar Flames and Detonations by 
Halogenated Compounds," Nineteenth Symposium ('International') on Combustion, pp. 127-141, The 
Combustion Institute, Pittsburgh, 1982. 

[87] Westbrook, C.K., "Numerical Modeling of Flame Inhibition by CF 3 Br," Combustion Science and 
Technology, vol. 34, pp. 201-225, 1983. 

[88] Simmons, R.F. and Wolfhard, H.G., "The Influence of Methyl Bromide on Flames: Part 2.-Diffusion 
Flames," Transactions of the Faraday Society, vol. 52, pp. 53-59, 1956. 

[89] Creitz, E.C., "Gas Chromatographic Determination of Composition Profiles of Stable Species Around 
a Propane Diffusion Flame," Journal of Chromatographic Science, vol. 10, pp. 168-173, March 1972. 

[90] Noda, S., Fujimoto, S., Claesson, O., and Yoshida, H., "ESR Studies of Bunsen-type Methane-Air 
Flames. II. The Effects of the Addition of Halogenated Compounds to the Secondary Air on the 
Hydrogen Atoms in the Flame," Bulletin of the Chemical Society of Jap an, vol. 56, pp. 2562-2564, 
1983. 

[91] Ibiricu, M.M. and Gaydon, A.G., "Spectroscopic Studies of the Effect of Inhibitors on Counterflow 
Diffusion Flames," Combustion and Flame , vol. 8, pp. 51-62, March 1964. 

[92] Kent, J.H. and Williams, F.A., "Extinction of Laminar Diffusion Flames," Fifteenth Symposium 
(International! on Combustion , pp. 315-325, The Combustion Institute, Pittsburgh, 1975. 

[93] Seshadri, K. and Williams, F.A., "Effect of CF 3 Br on Counterflow Combustion of Liquid Fuel with 
Diluted Oxygen," in Halogenated Fire Suppressants (Gann, R. G., Editor), pp. 149-180, American 
Chemical Society, Washington, D. C„ 1975. 

[94] Creitz, E.C., "Inhibition of Diffusion Flames by Methyl Bromide and Th'fiuoromethyl Bromide 
Applied to the Fuel and Oxygen Sides of the Reaction 2tone," Journal of Research of the National 
Bureau of Standards, vol. 65A, pp. 389-396, July-August 1961. 

[95] Miller, M.J., "The Relevance of Fundamental Studies of Flame Inhibition to the Development of 
Standards for the Halogenated Extinguishing Agent Systems," in Halogenated Fire Suppressants 
(Gann, R. G., Editor), pp. 64-91, American Chemical Society, Washington, D. C., 1975. 

[96] Bajpai, S.N., "An Investigation of the Extinction of Diffusion Flames by Halon," Journal of Fire and 
Flammability , vol. 5, pp. 255-267, October 1974. 

[97] Hirst, R. and Booth, K., Fire Technology, vol. 13, pp. 296-315, November 1977. 


81 




[98] 


Poeschl, P.M., "Large-Scale Halon 1301 Fire "test Program," Fire Journal , vol. 67, pp. 35-38, 
November 1973. 

[99] Yamashika, S., "Studies on the Required Quantity of Various Fire Extinguishing Agents," Report of 
the Fire Research Institute of Japan. Serial No. 36, pp. 7-12, March 1973. 

[100] Miller, M.J., "Evaluation of Fire Extinguishing Characteristics of Freon 1301 on Flammable Liquid 
Fires," Serial No. 16234.1, Factory Mutual Research Corporation, 1967. 

[101] Petrella, R.V. and Sellers, G.D., "Flame Inhibition by Bromine Compounds,” Fire "technology, vol. 
6 , pp. 93-101, May 1970. 

[102] Friedman, R. and Levy, J.B., "Inhibition of Opposed-Jet Methane-Air Diffusion Flames. The Effects 
of Alkali Metal Vapors and Organic Halides," Combustion and Flame, vol. 7, pp. 195-201, June 1963. 

[103] Milne, T.A., Green, C.L., and Benson, D.K., "The Use of the Counterflow Diffusion Flame in Studies 
of Inhibition Effectiveness of Gaseous and Powered Agents," Combustion and Flame, vol. 15, pp. 255- 
264, 1970. 

[104] Tuji, H. and Yamaoka, I., "The Counterflow Diffusion Flame in the Forward Stagnation Region of 
a Porous Cylinder," Eleventh Symposium (International! on Combustion , pp. 979-984, The 
Combustion Institute, Pittsburgh, 1967. 

[105] McKee, R.G., Wiersma, S.J., and Alvares, N.J., "Extinguishment of Turbulent Pool and Spray Fires 
by CF 3 Br and CF 2 ClBr," Fireline, pp. 12-16, April 1976. 

[106] Alvares, N.J., "CF 3 Br Suppression of Tarbulent, Class-B Fuel Fires," in Halogenated Fire Suppressants 
(Gann, R. G., Editor), pp. 94-117, American Chemical Society, Washington, D. C., 1975. 

[ 107] Cholin, R. "How Deep is Deep? Use of Halon 1301 on Deep-Seated Fires," Fire Journal, vol. 66, pp. 
19-23, March 1972. 

[108] Williamson, H.V., "Halon 1301-Minimum Concentration for Extinguishing Deep-Seated Fires," Fire 
"technology, vol. 8, pp. 269-277, November 1972. 

[109] Breen, D.E., "Interactions in Binary Halon Mixtures Used as Fire Suppressants," Fire "technology, vol. 
13, pp. 261-265, 281, November 1977. 

[110] Engibous, D.L. and Tbrkelson, T.R., A Study of Vaporizable Extinguishants . WADC Technical Report 
59-463, Wright Air Development Division, Air Research and Development Command, Wright- 
Patterson AFB, OH, January 1960. 

[111] Coward, H.F. and Hartwell, F.J., "Extinction of Methane Flames by Diluent Gases," Journal of the 
Chemical Society, pp. 1522-1532, 1926. 

[112] White, A.G., "Limits for the Propagation of Flame in Inflammable Gas-Air Mixtures. Part III. The 
Effect of "temperature on the Limits," Journal of the Chemical Society , vol. 127, pp. 672-684, 1925. 

[113] Simmons, R.F. and Wolfhard, H.G., "Some Limiting Oxygen Concentrations for Diffusion Flames in 
Air Diluted with Nitrogen," Combustion and Flame , vol. 1, pp. 155-161, 1957. 


82 



[114] Egerton, A. and Powling, J., "The Limits of Flame Propagation at Atmospheric Pressure. II. The 
Influence of Changes in the Physical Properties," Proceedings of the Royal Society of London, vol. 
A197, pp. 190-209, 27 May 1948. 

[115] Ishizuka, S. and Tkuji, H., "An Experimental Study of Effect of Inert Gases on Extinction of Laminar 
Diffusion Flames," Eighteenth Symposium ('International! on Combustion , pp. 695-703, The 
Combustion Institute, Pittsburgh, 1981. 

[116] Ewing, C.T., Hughes, J.T., and Carhart, H.W, "The Extinction of Hydrocarbon Flames Based on the 
Heat-Absorption Processes Which Occur in Them." Fire and Materials, vol. 8, pp. 148-156, September 
1984. 

[117] Jorissen, W.P., "Explosive Reactions and Negative Catalysis," Collection of Czechoslovak Chemical 
Communications, vol 2., pp. 288-291, May-June 1930. 

[118] Jorissen, W.P., Booy, J., and van Heiningen, J., "Reaction-Regions XXII. On the Prevention of 
Explosive Reactions in Gas and Vapour Mixtures by Small Amounts of Various Substances," Recueil 
des Travaux Chimiques des Pav-Bas. vol. 51, pp. 868-877, 15 July 1932. 

[119] Jorissen, W.P. and Hermans, J.J., "Reaction-Regions XXIII. On the Prevention of Explosive 
Reactions in Gas and Vapour Mixtures by Small Amounts of Various Substances. II.," Recueil des 
Travaux Chimiques des Pav-Bas . vol. 52, pp. 271-274, 15 March 1933. 

[120] Simmons, R.F. and Wright, N., "The Burning Velocities of Near Limit Mixtures of Propane, Air, and 
Hydrogen Bromide," Combustion and Flame, vol. 18, pp. 203-206, 1972. 

[121] Larsen, E.R., "Mechanism of Flame Inhibition I: The Role of Halogen," Journal of Fire & 
Flammabilitv/Fire Retardant Chemistry, vol. 1, pp. 4-12, February 1974. 

[122] Larsen, E.R., "Mechanism of Flame Inhibition II: A New Principle of Flame Suppression," Journal 
of Fire & Flammabilitv/Fire Retardant Chemistry, vol. 2, pp. 5-20, February 1975. 

[123] Larsen, E.R., "Halogenated Fire Extinguishants: Flame Suppression by a Physical Mechanism?," in 
Halogenated Fire Suppressants (Gann, R. G., Editor), pp. 376-388, American Chemical Society, 
Washington, D. C., 1975. 

[124] "Discussion of Halogenated Fire Extinguishants: Flame Suppression by a Physical Mechanism?," in 
Halogenated Fire Suppressants (Gann, R. G., Editor), pp. 389-402, American Chemical Society, 
Washington, D. C., 1975. 

[125] Sheinson, R.S., Gellene, G.I., Williams, F.W., and Hahn, J.E., "Quantification of Fire Suppressant 
Action on Liquid Pool Fires," Abstract for paper presented at the 1978 Fall Technical Meeting of the 
Eastern Section of the Combustion Institute, Miami Beach, FL, 29 November-1 December 1978. 

[126] Campbell, I.M., Energy and the Atmosphere , 2nd Edition, Wiley, New York, 1986. 

[127] Finlayson-Pitts, B.J. and Pitts, J.N., Atmospheric Chemistry': Fundamentals and Experimental 
Techniques , Wiley, New York, 1986. 

[128] Scientific Assessment of Stratospheric Ozone: 1989 . Vol. 2, Appendix: AFEAS Report, Report No. 
20, World Meteorological Organization, Global Ozone Research and Monitoring Project. 


83 




[129] Herron, J., "Ozone Depletion Potential," Chapter 4 in Preliminary Screening Procedures and Criteria 
for Replacements for Halons 1211 and 1301. NIST TN-1278, National Institute of Standards and 
Tfechnology, Gaithersburg, MD, 1990. 

[130] Wuebbles, D.J., The Relative Efficiency of a Number of Halocarbons for Destroying Stratospheric 
Ozone. Report UCID-18924, Lawrence Livermore National Laboratories, Livermore, CA, 1983. 

[131] Zavisca, F.G., "General Anesthetics: Gases and Volatile Liquids," in Modern Pharmacology. 2nd Ed., 
Eds.: Craig, C.R. and Stitzel, R.E., Little Brown, Boston, 1986. 

[132] Clayton, Jr., J.W., "Fluorocarbon Tbxicity and Biological Action," Fluorine Chemistry Reviews, vol. 
1, pp. 197-253, 1967. 

[133] McHale, E.T., Geary, R.W., von Elbe, G., and Huggett, C., "Flammability Limits of H 2 -0 2 -Fluorocar- 
bon Mixtures," Combustion and Flame, vol. 16, pp. 167-175, April 1971. 

[134] Gillotay, D. and Simon, P.C., "Ultraviolet Absorption Spectrum of THfluorobromomethane, 
Difluorodibromomethane and Difluorobromochloromethane in the Vapor Phase," Journal of 
Atmospheric Chemistry, vol. 8, pp. 41-62, 1989. 

[135] Simon, P.C., Gillotay, D., Vanlaethem-Meuree, N., Wisemberg, J., "Ultraviolet Absorption Spectrum 
of Chloro and Chlorofluoromethanes at Stratospheric Tbmperalures," Journal of Atmospheric 
Chemistry , vol. 7, pp. 107-135, 1988. 

[136] Molina, L.T., Molina, M.J., and Rowland, F.S., "Ultraviolet Absorption Cross Sections of Several 
Brominated Methanes and Ethanes of Atmospheric Interest," Journal of Physical Chemistry, vol. 86, 
pp. 2672-2680, 1982. 

[137] March, J., Advanced Organic Chemistry: Reactions. Mechanisms, and Structure , p 765, McGraw-Hill, 
New York, 1968. 

[138] TSang, W., "Single-Pulse Shock Tlibe Study on the Stability of Perfluorobromomethane," Journal of 
Physical Chemistry, vol. 90, pp. 414-418, 1986. 

[139] Sax, N.I. and Lewis, R.J., Dangerous Properties of Industrial Materials. 7th Edition, Van Nostrand 

Reinhold, New York, 1989. ..~ . 

[140] Molina, M.J., "Review of Ultraviolet Absorption Cross Sections of a Series of Alternative 
Fluorocarbons," in Scientific Assessment of Stratospheric Ozone: 1989 . Vol. 2, Appendix: AFEAS 
Report, Report No. 20, World Meteorological Organization, Global Ozone Research and Monitoring 
Project. 

[141] Calvert, J.G. and Pitts, J. N., Photochemistry , Wiley and Sons, New York, 1966. 

[142] Ayscough, P. B. and Steacie, E. W. R., "The Photolysis of Hexafluoroacetone," Proceedings of the 
Royal Society. Part A. vol. 234, pp. 476-488, 1956. 

[143] Giacometti, G., Okabe, H., and Steacie, E. W. R., "The Wave Length and Tbmperature-Dependence 
of the Fluorescence Efficiency and of the Primary Photochemical Yield in Hexafluoroacetone Vapor," 
Proceedings of the Royal Society. Part A . vol. 250, pp. 287-300, 1959. 


84 


[144] Okabe, H. and Steacie, E. W. R., "The Fluorescence and its Relationship to Photolysis in 
Hexafluoroacetone Vapor," Canadian Journal of Chemistry, vol. 36, pp. 137-146, 1958. 

[145] Ausloos, P., and Murad, E., Journal of Physical Chemistry , vol. 65, pp. 1519-1523, 1961. 

[146] Lyons, J.W., The Chemistry and Uses of Fire Retardants . John Wiley & Sons, New York, 1970. 

[147] TSang, W., "Single Pulse Shock Tlibe Study on the Thermal Stability of Ketones," International 
Journal of Chemical Kinetics, vol. 16, pp. 1543-1556, 1984. 

[148] Iking, W., "Single Pulse Shock-Tibe Studies on the Decomposition of 1,2-Dibromoperfluoroethane 
and Allyl Bromide," Journal of Physical Chemistry , vol. 88, pp. 2812-2817, 1984. 

[149] McMillen, D. F. and Golden, D. M., "Hydrocarbon Bond Dissociation Energies," Annual Review of 
Physical Chemistry, vol. 33, pp. 493-532, 1982. 

[150] Baulch, D. L., Duxbury, J., Grant, S. J. and Montague, D. C., "Evaluated Kinetic Data for High 
Temperature Reactions, Vol 4. Homogeneous Gas Phase Reaction of Halogen and Cyanide 
Containing Species," Journal of Physical and Chemical Reference Data, vol. 10, Supplement 1, 1981. 

[151] "Rang, W. and Walker, J. A., "Hydrogen Atom Attack on Perchloroethylene" TWentv-Third Symposium 
flnternationall on Combustion . 1990, in press. 

[152] Belenki, G. G., Fokin, A V., Rondarev, D. S., Ryazanovam, R. M., Sokolov, S. V., Sterlin, S. R., 
Voronkov, Zeifman, Y. V. "Fluoroaliphatic Compounds," in Synthesis of Fluorooreanic Compounds . 
(Knunyants, I. L. and Yakobson, I. L, Editors) Springer-Verlag, New York, pp. 3-108, 1985. 

[153] Atkinson, R., Baulch, D. L., Cox, R. A, Hampson, R. F., Kerr, J. A. and Hoe, J., "Evaluated Kinetic 
and Photochemical Data for Atmospheric Chemistry," Journal of Physical and Chemical Reference 
Data , vol. 18, pp. 881-1097, 1989. 

[154] Larsen, E. R., "Fluorine Compounds in Anesthesiology" in Fluorine Chemistry Reviews, (Thrrant, 
P„ Editor.) Marcel Dekker, New York, 1969. 

[155] Hang, W., J. P. Cui and J. A Walker, "Single Pulse Shock TUbe Study of the Reactions of Hydrogen 
Atoms with Complex Aromatics," Proceedings of the 17th International Symposium on Shock Tlibes 
and Waves, in press. 

[156] Benson, S. W. and O’Neal, H. E., Unimolecular Reactions. NSRDS-NBS-21, U.S. Government 
Printing Office, 1970. 

[157] Stull, D. R., Westrum, E. F. and Sinke, G. C., The Chemical Thermodynamics of Organic Compounds. 
John Wiley and Sons, New York, 1969. 

[158] Leathard, D. A and Purnell, J. H., "Paraffin Pyrolysis,” Annual Reviews of Physical Chemistry , vol. 
21, pp. 197-233, 1970. 

[159] Le Blanc, M. and Riess, J. G., "Artificial Blood Substitutes Based on Perfluorochemicals" in 
Preparation. Properties and Industrial Applications of Organofluorine Compounds (Banks, R. E., 
Editor), pp. 83-138, Halsted Press, New York, 1982. 


85 





[160] Geyer, R. P., "Perfluorochemicals as Oxygen Transport Vehicles" in Blood Substitutes (Chang, T M. 
S. and Geyer, R. P., Editors), pp. 31-49, Marcel Dekker, New York, 1989. 

[161] Bagnall, R. D., Bell, W. and Pearson K., "Inhalation Anesthetics: V. Fluorinated Butanes (and 
Butenes)" Journal of Fluorine Chemistry, vol. 13, pp. 325-335, 1979. 

[162] Jones, W. G. M. "Inhalation Anesthetics of the Organofluorine Class" in Preparation, Properties and 
Industrial Applications of Organofluorine Compounds. (Banks, R. E., Editor.), pp. 157-192, Halsted 
Press, New York, 1982. 

[163] Craig, C. R. and Stitzel, R. E. Modern Pharmacology, pp. 462-465, Little, Brown and Company, 1982. 

[164] Bagnall, R. D., Bell, W., Pearson, K. and Jeater, A,, "Inhalation Anesthetics, III. Fluorinated Aliphatic 
Ethers," Journal of Fluorine Chemistry, vol. 13, pp. 123-140, 1979. 

[165] Bagnall, R. D., Bell, W., and Pearson, K., "Inhalation Anesthetics, II. Fluorinated Methyl Propyl 
Ethers," Journal of Fluorine Chemistry, vol. 13, pp. 93-107, 1979. 

[166] Bagnall, R. D., Bell W., and Pearson K., "Inhalation Anesthetics, I. Fluorinated Dioxalane 
Derivatives," Journal of Fluorine Chemistry, vol. 9, pp. 359-375, 1977. 

[167] Bagnall, R. D., Bell, W., and Pearson K., "Inhalation Anesthetics, IV. Fluorinated Propanes," Journal 
of Fluorine Chemistry, vol. 13, pp. 209-223, 1979. 

[168] Atkinson, R. "Kinetics and Mechanisms of the Gas Phase Reactions of the Hydroxyl Radical with 
Organic Compounds," Journal of Physical and Chemical Reference Data. Monograph 1, American 
Institute of Physics, New York, 1989. 

[169] Banks, R. E. Fluorocarbons and Their Derivatives. Second Edition, American Elsevier, New York, 
1970. 

[170] Burger, H., Niepel, H., Pawelke, G., Frohn, H. J. and Sartori, "Hexa-, Hepta- und Octafluorotrimethyl- 
amin" Journal of Fluorine Chemistry, vol. 15, pp. 231-237, 1980. 

[171] Abe, T. and Nagase, S. "Electrochemical Fluorination (Simons Process) as a Route to Perfluorinated 
Organic Compounds of Industrial Interest," in Pre paration, Properties and Industrial Applications of 
Organofluorine Compounds (Banks, R. E., Editor), pp. 19-43, Halsted Press, New York, 1982. 

[172] Pawelke, G., Heyder, F., and Burger, H., "Halogen Exchange of Fluoroalkyl Amines-- Synthesis of 
Polychloro- and Bromo-'IHalkylamines," Journal of Fluorine Chemistry, vol. 20, pp. 53-63, 1982. 

[173] Brandt, G. A. R., Emeleus, H. J., and Haszeldine, R. N., "Organometallic and Organometalloidal 
Fluorine Compounds. Part III. THfluoromethyl Derivatives of Sulfur," Journal of the Chemical 
Society , pp. 2198-2205, 1952. 

[174] de C. Ellis, O.C., "Extinction of Petrol Fires by Methyl Iodide," Nature , vol. 161, pp. 402-403, 13 
March 1948. 

[175] von Halasz, S.P. and Glemser, O., "The Sulfur-Fluorine Bond," Chapter 7 in Sulfur in Organic and 
Inorganic Chemistry . Vol. 1 (Senning, A, Editor), Marcel Dekker, New York, 1971. 


86 



[176] Schmidt, M. and Siebert, W., "Sulfur," Chapter 23 in Comprehensive Inorganic Chemistry. Vol. 2 
(Bailer, J.C., Emelesus, H.J., Nyholm, R., and Ttotman-Dickenson, A.F., Editors), Pergamon Press, 
Oxford, 1973. 

[177] Gage, J.C., "Subacute Inhalation Tbxicity of 109 Industrial Chemicals," British Journal of Industrial 
Medicine, vol. 27, pp. 1-18, 1970. 

[178] Miller, W.J., "Inhibition of Low Pressure Flames," Combustion and Flame, vol. 13, pp. 210-212, April 
1969. 

[179] Kovacina, T.A., Berry, AD., and Fox, W.B., "Improved Preparation and Purification of Pentafluoro- 
sulfur Bromide," Journal of Fluorine Chemistry, vol. 7, pp 430-432, 1976. 

[180] Cotton, F.A and G. Wilkinson, G., Advanced Inorganic Chemistry. John Wiley & Sons, New York, 
1972. 

[181] Tby, AD.F. "Phosphorous," Chapter 20 in Comprehensive Inorganic Chemistry, vol. 2 (Bailer, J.C., 
Emelesus, H.J., Nyholm, R., and Ttotman-Dickenson, A.F., Editors), Pergamon Press, Oxford, 1973. 

[182] CRC Handbook of Chemistry and Physics . 70th Edition, CRC Press, Boca Raton, FL, 1989. 

[183] Braker, W. and Mossman, AL., Matheson Gas Data Book . 5th Edition, Matheson Gas Products, East 
Rutherford, NJ, 1971. 

[184] Schmutzler, R., "Fluorides of Phosphorous," in Advances in Fluorine Chemistry (Stacey, M., Thtlow, 
J.C., and Sharpe, AG., Editors), vol. 5, pp. 34-279, 1965. 

[185] Rochow, E. G. in Comprehensive Inorganic Chemistry, vol. 11, Pergamon Press, New York, 1973. 

[186] W&gner, H. G., "Studies of Inhibitors as Anticatalytic Extinction Agents (Preliminary Report)," 
Research Contract No. 3/55, 1955. 

[187] Morrison, M. E. and Scheller, K., "The Effect of Burning Velocity Inhibitors on the Ignition of Hydro¬ 
carbon - Oxygen - Nitrogen Mixtures," Combustion and Flame , vol. 18, pp. 3-12, 1972. 

[188] Ebsworth. E.AV, Volatile Silicon Compounds. Academic Press, New York, 1975. 

[189] Hastie, J. W., High Temperature Vapors, pp. 335-357, Academic Press, New York, 1975. 

[190] Handbook of Chemistry and Physics. Fifty Second Edition, Weast, R. C. (Ed.), The Chemical Rubber 
Co., Cleveland, OH, 1971. 

[191] Shuzo, O., Computer Aided Data Book of Vapor Pressure. Data Book Publishing Co., Tbkyo, Japan, 
1976. 

[192] Hayes, K. and Kaskan, W. E., "Inhibition by CH 3 Br of CH 4 /A'r Flames Stabilized on a Porous 
Burner," Combustion and Flame, vol. 24, pp. 405-407, 1975. 

[193] McCamy, C. S., Shoub, H. and Lee, T G., "Fire Extinguishment by Means of Dry Powder," Sixth 
Symposium (International! on Combustion , pp. 795-801, Reinhold, New York, 1957. 


87 


[194] Lee, T. G. and Robertson, A F., "Extinguishing Effectiveness of Some Powdered Materials on Hydro¬ 
carbon Fires," Fire Research Abstracts and Reviews, vol. 2, pp. 13-17, 1960. 


[195] Dodding, R. A, Simmons, R. F., Stephens, A, "The Extinction of Methane - Air Diffusion Flames 
by Sodium Bicarbonate Powders," Combustion and Flame , vol. 15, pp. 313-315, 1970. 

[196] Dolan, J. E., "The Suppression of Methane/Air Ignitions By Fine Powders," Sixth Symposium 
flnternationan on Combustion, pp. 787-794, Reinhold, New York, 1957. 

[197] Birchall, J. D., "On the Mechanism of Flame Inhibition by Alkali Metal Salts," Combustion and 
Flame, vol. 14, pp. 85-96, 1970. 

[198] Friedrich, M., "Extinguishment Action of Powders," Fire Research Abstracts and Reviews, vol. 2, pp. 
132-135, 1960. 

[199] Dewitte, M., Vrebosch, J. and van Tiggelen, A, "Inhibition and Extinction of Premixed Flames by 
Dust Particles," Combustion and Flame. Vol. 8, pp. 257-266, 1964 

[200] Iya, K. S., Wollowitz, S. and Kaskan, W. E., "The Mechanism of Flame Inhibition by Sodium Salts," 
Fifteenth Symposium (International! on Combustion, pp. 329-336, The Combustion Institute, 
Pittsburgh, PA, 1975. 

[201] Rosser, W. A, Inami, S. H., and Wise, H,, "The Effect of Metal Salts on Premixed Hydrocarbon -Air 
Flames," Combustion and Flame, vol. 7, pp. 107-119,1961. 

[202] Gardiner, W. C. Jr., Rates and Mechanisms of Chemical Reactions , pp. 136-149, Benjamin- 
Cummings, Menlo Park, 1972. 

[203] Bulewicz, E. M. and Padley, P. J., "Catalytic Effect of Metal Additives on Free Radical Recombination 
Rates in H 2 + 0 2 + N 2 Flames," Thirteenth Symposium (International! on Combustion, pp. 73-80, 
The Combustion Institute, Pittsburgh, PA 1971. 

[204] Jost, W., Bonne, U. and Wagner, H. G., Chemical and Engineering News, vol. 39, p. 76,11 September, 
1961. 

[205] Bonne, U., Jost, W. and Wagner, H. G., Fire Research Abstracts and Reviews , vol. 4, pp. 6-18,1962. 

[206] Kaufman, F., "The Air Afterglow and its Use in the Study of Some Reactions of Atomic Oxygen," 
Proceedings of the Royal Society ('London!, vol A247, pp. 123-139, 1958. 

[207] Tischer, R. L. and Scheller, PL, "The Behavior of Uranium Oxide Particles in Reducing Flames," Com¬ 
bustion and Flame, vol. 15, pp 199-202, 1970. 

[208] Bulewicz, E. M. and Padley, P. J., "Metal Oxide Particle 'Temperatures in Flames," Combustion and 
Flame, vol, 15, pp. 203-205, 1970. 

[209] Miller, W. J., "Flame Ionization and Combustion Inhibition," Fire Research Abstracts and Reviews. 
vol. 10, p. 191, 1968. 

[210] Miller, W. J. and Vree, P. H., "Flame Ionization and Combustion Reactions," Fire Research Abstracts 
and Reviews , vol. 10, pp. 190-191, 1968. 


88 



[211] Vanpee, M. and Shirodkar, P. P., "A Study of Flame Inhibition by Metal Compounds," Seventeenth 
Symposium (International! on Combustion, pp. 787-793, The Combustion Institute, Pittsburgh, 1978. 


[212] Fristrom, R. M., "Combustion Suppression," Fire Research Abstracts and Reviews, vol. 9, pp. 125-152, 
1967. 

[213] The Sigma-Aldrich Library of Chemical Safety Data . First Edition (Lenga, R.E., Editor), Sigma- 
Aldrich, 1985. 

[214] legman, D. D., Evans, W. H., Parker, V. B., Schamn, R. H., Halow, F., Bailey, S. M., Churney, K. 
L. and Nuttall, R. L., Journal of Physical and Chemical Reference Data, vol. 11, supplement No. 2. 

[215] Thtem, P.A., Gann, R.G., and Carhart, H.W., "Pressurization with Nitrogen as an Extinguishant for 
Fires in Confined Spaces," Combustion Science and Technology. vol. 7, pp. 213-218, 1973. 

[216] Thtem, P.A., Gann, R.G., and Carhart, H.W., "Pressurization with Nitrogen as an Extinguishant for 
Fire in Confined Spaces. II. Cellulosic and Fabric Fuels," Combustion Science and Technology, vol. 
9, pp. 255-259, 1974. 

[217] Gann, R.G., Stone, J.P., Thtem, F.W., and Carhart, H.W., "Suppression of Fires in Confined Spaces 
by Nitrogen Pressurization: III. Extinction Limits for Liquid Pool Fires," Combustion Science and 
Technology , vol. 18, pp. 155-163, 1978. 


89 



90 



APPENDIX A 

DATA SHEETS FOR SELECTED COMPOUNDS 


This Appendix contains data sheets for each of the selected compounds that include formulae, 
common names, classification numbers, physical properties, commercial sources and prices (if available), 
toxicity information, references to fire suppression results, and additional relevant comments. 


91 




Name: perfluoromethane 


Compound group: 
Formula: 
Alternate Names: 


Halogenated hydrocarbons 

CF 4 

halon 14, cfc 14 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


75-73-0 

-128 

Pressure greater than critical pressure. 

61 

11.6 

PCR, Inc., Gainesville, FL 32602, (800)-331-6313 
$65/50g 


Tbxicity: 


Essentially non-toxic. CF 4 can be view as a simple asphyxiant, that is its action is only to 
replace air. 


Comments: There is a substantial body of evidence to indicate the effectiveness of CF 4 ; however, it will 

require a much larger concentration than halon 1301 and it may be a significant greenhouse 
gas. 

Fire suppression studies: 

Ewing, C.T, Hughes, J.T and Carhart, H.W., "The Extinction of Hydrocarbon Barnes based 
on the Heat-absorption Processes which Occur in them," Fire and Materials, vol. 8, pp 148- 
156, 1984. 

McHale, E.T, "Survey of Vapor Phase Chemical Agents for Combustion Suppression," Fire 
Research Abstracts and Reviews, vol. 11, pp 90-104, 1969. 


92 



Name: perfluoroethane 


Compound group: 
Formula: 
Alternate Names: 


Halogenated hydrocarbons 

<^6 

halon 26, cfc 116 


CASN: 76-16-4 

Normal boiling point (nbp °C): -78 

Vapor pressure (atm at 298 K): Pressure greater than critical pressure. 

Heat capacity (J/K-mol at 298 K): 106 

Heat of Vaporization (kJ/mol at nbp): 16.1 

Source: PCRJnc, Gainesville,FL 32602, (800)-331-6313 

Price: $40/100g 


Tbxicity: Very low, also primarily an asphyxiant. 


Comments: A more effective fire suppressant than CF 4 , less effective than CjFg, also possible problems 

with greenhouse effect. 

Fire suppression studies: 

McHale,E.T., "Survey of Vapor Phase Chemical Agents for Combustion Suppression," Fire 
Research Abstracts and Reviews, vol. 11, pp 90-104, 1969. 


93 





Name: perfluoropropane 


Compound group: Halogenated hydrocarbons 

Formula: CjFg 

Alternate Names: halon 38, cfc 218 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


76-19-7 

-36 

8.69 

148 


19.7 

PCR, Inc., Gainesville, FL 32602, (800)-331-6313 
$75/50g 


Tbxicity: Sax and Lewis give inhalation data for rats and mice. Rated as mildly toxic by inhalation. 

Sax, N.I.; Lewis, R.J.; "Dangerous Properties of Industrial Materials," Van Norstrand 
Reinhold, New York (1987). 


Comments: This compound has the right range for physical properties for a replacement for halon 1301. 

As with the other perfluorinated compounds it is likely to be a problem as greenhouse gas. 


94 





Name: perfluoro-n-butane 


Compound group: 
Formula: 
Alternate Names: 


Halogenated hydrocarbons 
C 4 Fio 

Outside of halon naming conventions. 


CASN: 355-25-9 

Normal boiling point (nbp °C): -2 

Vapor pressure (atm at 298 K): 2.63 

Heat capacity (J/K-mol at 298 K): 189 

Heat of Vaporization (kJ/mol at nbp): 23.2 

Source: PCR, Inc., Gainesville, FL 32602, (800)-331-6313 

Price: $50/25g 


"toxicity: As the perfluorinated saturated hydrocarbons become larger, there are increasing toxicity 

concerns. There has been very limited long term study of these compounds. 


Comments: 


Previous fire suppression studies: 

McHale, E.T., "Survey of Vapor Phase Chemical Agents for Combustion Suppression," Fire 
Research Abstracts and Reviews, vol. 11, pp 90-104, 1969. 


95 




Name: perfluorocyclobutane 


Compound group: 
Formula: 
Alternate Names: 


Halogenated hydrocarbons 
C 4 F 8 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


115-25-3 

-6 

3.07 

156 

23.0 

PC.R, Inc., Gainesville, FL 32602, (800)-331-6313 
$75/25g 


Ibxicity: Tbxicity suspect. Decomposition products may be exceptionally toxic. There is some 

anecdotal evidence that this compound will decompose in a flame or on hot surfaces to 
produce perfluoroisobutene which is very toxic. This report does not attempt to cover 
combustion product toxicity, but the toxicity of the perfluoroisobutene has been reported to 
be so high that some caution needs to be advised with regard to this compound. Sax and 
Lewis give only limited data to indicated low direct toxicity. 

Sax, N.I.; Lewis, R.J.; "Dangerous Properties of Industrial Materials," Van Norstrand 
Reinhold, New York (1987). 


Comments: Except for its toxic effects, this compound would be an interesting candidate for the 

investigation of chemical energy absorption since it breaks down into two molecules in 
the gas phase. 

Fire suppression studies: 

McHale, E.T., "Survey of Vapor Phase Chemical Agents for Combustion Suppression," Fire 
Research Abstracts and Reviews, vol. 11, pp 90-104, 1969. 


96 






Name: trifluoromethane 


Compound group: Halogenated hydrocarbons 

Formula: CHFj 

Alternate Names: halon 13, cfc 23 


CASN: 

Normal boiling point (nbp ”C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


75-46-7 

-82 

46.7 
51 

16.7 


Aldrich Chemical, Milwaukee, WI 53201, (800)-558-9160 

$120/200g 


Tbxicity: Tbxicity is expected to be very low. Clayton has reported data showing no fatalities in a 2 

hour exposure of guinea pigs at 20% trifluoromethane. 

Clayton, J.W., "Fluorocarbon Tbxicity and Biological Action," Fluorine Chemistry Reviews 
(ed. Thrrant, P.), vol. 1, 197-252, 1967. 


Comments: This compound is undergoing tests as an alternative cfc for uses other than fire suppression. 

Fire suppression studies: 

McHale, E.T., "Survey of Vapor Phase Chemical Agents for Combustion Suppression," Fire 
Research Abstracts and Reviews, vol. 11, pp 90-104, 1969. 


97 




Name: pentafluoroethane 


Compound group: 
Formula: 
Alternate Names: 


Halogenated hydrocarbons 

C2HF5 

halon 25, cfc 125 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


354-33-6 

-48.5 

10 f 

94 

20 t 

PCR, Inc., Gainesville, FL 32602, (800)-331-6313 
S120/50g 


Tbxicity: DuPont has given a value for toxicity of >>100,000 ppm for rats exposed for 4 hrs. In the 

same note the stated toxicity of CF 3 Br was 400,000 - 800,000 ppm. This data comes from 
Clayton who reports essentially the same value (Clayton was at Haskell Labs - a branch of 
DuPont). 

Clayton, J.W., "Fluorocarbon Tbxicity and Biological Action," Fluorine Chemistry Reviews 
(ed. Thrrant, P.), vol. 1, 197-252, 1967. 


Comments: This compound is undergoing tests as an alternative cfc for uses other than fire suppression. 

In addition DuPont has recently announced that it will be making Cy-IFj available for testing 
as a fire suppressant. 


+ Estimated for this report by authors. 




Name: l,l,l>2-tetrafluoroethane 


Compound group: Halogenated hydrocarbons 

Fbrmula: C 2 H 2 F 4 

Alternate names: halon 24, cfc 134 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


881-97-72 

-26.5 

7+ 

87 + 

22 + 

PCR, Inc., Gainesville, FL 32602, (800)-331-6313 
$195/100g 


Ibxicity: Data was not found for this compound, but the 1,1,1 trifluoroethane was reported to have 

anesthetic dose for 50% of the mice tested of 50-60% and no fatal dose was reported. In 
general the more highly fluorine substituted the molecule, the lower its toxicity, so that this 
compound would be expected to be at least as safe as the 1,1,1 trifluoro compound. 

Robbins, B.H., "Preliminary studies of the anesthetic activity of fluorinated hydrocarbons," J. 
Pharmacology & Experimental Therapeutics, vol. 86, pp 197-204, 1946. 


Comments: This compound is undergoing tests as an alternative cfc for uses other than fire suppression. 

Note that only compounds with small numbers of H atoms were considered among the 
alternative cfc’s. As the number of H atoms increases and the corresponding halogen content 
of the molecule decreases, the fire suppression effectiveness of the molecule will decline. 

* Estimated for this report by authors. 


99 





Name: dibromodifluoromethane 


Compound group: 
Formula: 
Alternate Names: 


Halogenated hydrocarbons 
CBr 2 F 2 

halon 1202, cfc 12B2 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


75-61-6 

-58 


15.8 

69 

17.5 


PCR, Inc., Gainesville, FL 32602, (800)-331-6313 
$35/1 OOg 


Tbxicity: Clayton reports LC50 in rats of 5.5% for a 15 min exposure. 

Clayton, J.W., "Fluorocarbon Tbxicity and Biological Action," Fluorine Chemistry Reviews 
(ed. Thrrant, P.), vol. 1, 197-252, 1967. 


Comments: This compound will probably be destroyed by solar photolysis in the troposphere. 

Fire suppression studies: 

Ewing, C.T., Hughes, J.T and Carhart, H.W., The Extinction of Hydrocarbon Flames based 
on the Heat-absorption Processes which Occur in them," Fire and Materials, vol. 8, pp 148- 
156, 1984. 

McHale, E.T, "Survey of Vapor Phase Chemical Agents for Combustion Suppression," Fire 
Research Abstracts and Reviews, vol. 11, pp 90-104, 1969. 


100 





NAME: 2,2-dibromo-l,l,l>2-tetrafluoroethane 


Compound group: Halogenated hydrocarbons 

Formula: C^BrjF,* 

Alternate Names: 1,1 dibromotetrafluoroethane, FC-114aB2 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


27336-28-8 

50+ 

0.4 + 

116 + 

29 + 

PCR,Inc., Gainesville,FL 32602, (800)-331-6313 
Sl60/100g 


Tbxicity: No data found. This is probably more toxic than the symmetric analog since it is less stable. 


Comments: This compound is the asymmetric form of halon 2402, and as such could also be called halon 

2402, to avoid confusion this name is not used. Because of the two bromine atoms on a 
single carbon, this compound should be very sensitive to solar photolysis in the troposphere. 

* Estimated for this report by authors. 





Name: chlorodifluoromethane 


Compound group: 
Formula: 
Alternate Names: 


Halogenated hydrocarbons 
CHC1F 2 

halon 121, cfc 22 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


75-45-6 

-41 

10.1 

57 

20.2 


Aldrich Chemical, Milwaukee, WI 53201, (800)-558-9160 
$30/100g 


Tbxicity: This compound has been tested as an anesthetic and the reported AD50 (the dose at which 

50% of the test animals are anesthetized) is 20%. No data on fatality was found. 

Davies, R.H., Bagnall, R.D., Jones, W.G.M., "A Quantitative Interpretation of Phase Effects 
in Anaesthesia," Int. J. Quantum Chem: Quantum Biology Symp. No. 1, pp 201-212, 1974. 


Comments: 


This compound is undergoing tests as an alternative cfc for uses other than fire suppression. 





Name; 1,1,1-trichIoroethane 


Compound group: 
Formula: 
Alternate Names: 


Halogenated hydrocarbons 

GjHjCIj 

halon 203, cfc 14 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


71-55-6 

74 

0.16 

93 

29.8 


Aldrich Chemical, Milwaukee, WI 53201, (800)-558-9160 
$8/100ml 


Tbxicity: Sax and Lewis give extensive data and conclude that the compound is moderately toxic by 

inhalation. In addition it is an experimental teratogen and sensitizes the heart. Lowest 
effective concentration in humans for an effect on central nervous system is reported at 
200ppm over a 4H exposure. 

Sax, N.I.; Lewis, R.J.; "Dangerous Properties of Industrial Materials," Van Norstrand 
Reinhold, New York (1987). 


Comments: 


This compound is undergoing tests as an alternative cfc for uses other than fire suppression. 






Name: 2,2-dichIoro-l,l,l-trifl u °n)ethane 


Compound group: 
Formula: 
Alternate Names: 


Halogenated hydrocarbons 

C2HC1 2 F 3 

halon 2302, cfc 123 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source; 

Price: 


306-83-2 

24 

1 + 

102 

26 + 

PCR, Inc., Gainesville, FL 32602, (800)-331-6313 
S75/lkg 


Tbxicity: Sax and Lewis report 14% in 4 min for mouse toxicity. Robbins reports an LD50 of 7.7%. 

Both Robbins and Davies et al. report AD50 of about 2.5%. 

Sax, N.I.; Lewis, R.J.; "Dangerous Properties of Industrial Materials," Van Norstrand 
Reinhold, New York (1987). 

Robbins, B.H., "Preliminary studies of the anesthetic activity of fluorinated hydrocarbons," J. 
Pharmacology & Experimental Therapeutics, vol. 86, pp 197-204, 1946. 

Davies, R.H., Bagnall, R.D., Jones, W.G.M., "A Quantitative Interpretation of Phase Effects 
in Anaesthesia," Int. J. Quantum Chem: Quantum Biology Symp. No. 1, pp 201-212, 1974. 


Comments: This compound is undergoing tests as an alternative cfc for uses other than fire suppression. 


+ Estimated for this report by authors. 





Name: 2-chloro-l,l,l>2-tetrafluoroethane 


Compound group: 
Formula: 
Alternate Names: 


Halogenated hydrocarbons 

C2HC1F 4 

halon 241 


CASN: 2837-89-0 

Normal boiling point (nbp °C): -12 

Vapor pressure (atm at 298 K): 4* 

Heat capacity (J/K-mol at 298 K): 101* 

Heat of Vaporization (kJ/mol at nbp): 23* 

Source: PCR, Inc., Gainesville, FL 32602, (800)-331-6313 

Price: $48/100g 


Tbxicity: Davies et al. report an AD50 of 15% in mice. 

Davies, R.H., Bagnall, R.D., Jones, W.G.M., "A Quantitative Interpretation of Phase Effects 
in Anaesthesia," Int. J. Quantum Chem: Quantum Biology Symp. No. 1, pp 201-212, 1974. 


Comments: This compound is undergoing tests as an alternative cfc for uses other than fire suppression. 


* Estimated for this report by authors. 





Name: 1,1-dichIoro-l-fluoroethane 


Compound group: Halogenated hydrocarbons 

Formula: Cy-J^C^F 

Alternate Names: halon 212 


Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


1717-00-6 

32 

0.5+ 

89 + 

27 t 


PCR, Inc., Gainesville, FL 32602, (800)-331-6313 
$65/250g 


Tbxicity: Robbins reports an AD50 of 2 . 5 % and an LD50 of 5% in mice. Davies et al. report an AD50 

of 6% in mice. 


Robbins, B.H., "Preliminary studies of the anesthetic activity of fluorinated hydrocarbons," J. 
Pharmacology & Experimental Therapeutics, vol. 86, pp 197-204, 1946. 

Davies, R.H., Bagnall, R.D., Jones, W.G.M., "A Quantitative Interpretation of Phase Effects 
in Anaesthesia," Int. J. Quantum Chem: Quantum Biology Symp. No. 1, pp 201-212, 1974. 


Comments: This compound is undergoing tests as an alternative cfc for uses other than fire suppression. 

* Estimated for this report by authors. 


106 





Name: l-ch!oro-l,l-clifluoroethane 


Compound group: 
Formula: 
Alternate Names: 


Halogenated hydrocarbons 

C2H 3 CtF 2 

halon 221 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


75-68-3 

-10 

3.33 

82 

22.4 

Aldrich Chemical, Milwaukee, WI 53201, (800)-558-9160 
$40/1 OOg 


Toxicity: Davies et al. report AD50 of 23% for mice, Robbins 2 reported AD50 of 25% and no 

fatalities. 


Davies, R.H., Bagnall, R.D., Jones, W.G.M., "A Quantitative Interpretation of Phase Effects 
in Anaesthesia," Int. J. Quantum Chem: Quantum Biology Symp. No. 1, pp 201-212, 1974. 

Robbins, B.H., "Preliminary studies of the anesthetic activity of fluorinated hydrocarbons," J. 
Pharmacology & Experimental Therapeutics, vol. 86, pp 197-204, 1946. 


Comments: 


This compound is undergoing tests as an alternative cfc for uses other than fire suppression. 



Name: bromodifluoromethane 


Compound group: Halogenated hydrocarbons 

Formula: CHBrF 2 

Alternate Names: halon 1201 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


1151-62-2 

-15 

4.4 

59 

23 + 

Great Lakes Chemicals will be selling this in the near future, 
na 


Tbxicity: Toxicity, as reported by Great Lakes Chemicals, is greater than halon 1211 and much greater 

that halon 1301. Cardiac sensitization not reported. 


Comments: Great Lakes Chemicals has recently announced that they will be making this compound 

available as a commercial fire suppressant. 

Fire suppression studies: 

McHale, E.T., "Survey of Vapor Phase Chemical Agents for Combustion Suppression," Fire 
Research Abstracts and Reviews, vol. 11, pp 90-104, 1969. 


t Estimated for this report by authors, na = not available. 


108 







Name: bromochlorofluoromethane 


Compound group: 
Formula: 
Alternate Names: 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


Halogenated hydrocarbons 

CHBrCIF 

halon 1111 


593-98-6 

36 

0.5 + 

63 

28 + 

No commercial source was identified 
na 


Tbxicity: Davies et al. report an AD50 of 1% for mice. This compound may be expected to be 

substantially more toxic than many of the other halons. 

Davies, R.H., Bagnall, R.D., Jones, W.G.M., "A Quantitative Interpretation of Phase Effects 
in Anaesthesia," Int. J. Quantum Chem: Quantum Biology Symp. No. 1, pp 201-212, 1974. 


Comments: This compound will have a relatively short lifetime due to solar photolysis as well as reacting 

with OH. 


+ Estimated for this report by authors, na = not available. 


109 




Name: 2-bromo-2-chIoro-l,l>l-trifluoroethane 


Compound group: Halogenated hydrocarbons 

Formula: C^HBrClFg 

Alternate Names: halothane, narcotane 


CASN: 151-67-7 

Normal boiling point (nbp °C): 50 

Vipor pressure (atm at 298 K): 0.4 

Heat capacity (J/K-mol at 298 K): 104 

Heat of Vaporization (kJ/mol at nbp): 27.3 

Source: PCR, Inc., Gainesville, FL 32602, (800)-331-6313 

Price: $25/100g 


Tbxicity: Sax and Lewis give extensive data on this molecule since it is widely used as an anaesthetic. 

Human toxicity over 3-hour periods starts to occur at about 7000 ppm. In addition it is a 
sever eye irritant which may make it undesirable as a suppressant where continued occupation 
of the site is necessary. Davies et al. report AD50 in mice of less than 1%, Robbins does not 
report data on this compound but has data on the 2-bromo-l-chloro-l,l-difluoro ethane and 
gives AD50 of 0.8% and LD50 of 3.7%. 

Sax, N.I.; Lewis, R.J.; "Dangerous Properties of Industrial Materials," Van Norstrand 
Reinhold, New York (1987). 

Davies, R.H., Bagnall, R.D., Jones, W.G.M., "A Quantitative Interpretation of Phase Effects 
in Anaesthesia," Int. J. Quantum Chem: Quantum Biology Symp. No. 1, pp 201-212, 1974. 

Robbins, B.H., "Preliminary studies of the anesthetic activity of fluorinated hydrocarbons," J. 
Pharmacology & Experimental Therapeutics, vol. 86, pp 197-204, 1946. 


Comments: This compound is included as a research topic. As an anaesthetic it is not appropriate as a 

general fire suppressant but it represents an important class of compounds that can undergo 
HBr elimination. 


110 




Name: 2-bromo-l-chIoro-l,2,2-trifluoroethane 


Compound group: Halogenated hydrocarbons 

Formula: CjHBrCIFg 

Alternate Names: 


CASN: 

Normal boiling point (nbp 6 C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


354-06-3 

53 

0.4* 

103 

29* 

PCR, Inc., Gainesville, FL 32602, (800)-331-6313 
S25/100g 


Ibxicity: Sax and Lewis give data for mice. Lowest toxic concentration reported is 35 ppt over 17 

minutes. Davies et al. give an AD50 of 1.1% for mice. 

Sax, N.I.; Lewis, R.J.; "Dangerous Properties of Industrial Materials," Van Norstrand 
Reinhold, New York (1987). 

Davies, R.H., Bagnall, R.D., Jones, W.G.M., "A Quantitative Interpretation of Phase Effects 
in Anaesthesia," Int. J. Quantum Chem: Quantum Biology Symp. No. 1, pp 201-212, 1974. 


Comments: This compound is more interesting from the point of view of undergoing HBr elimination 

since the H and Br are on adjacent carbon atoms rather than on the same carbon atom. 

* Estimated for this report by authors. 





Name: l-bromo-l,l,2,2-tetrafluoroethane 


Compound group: Halogenated hydrocarbons 

Formula: CjHB^ 

Alternate Names: halon 2401 


CASN: 354-07-4 

Normal boiling point (nbp °C): -5 

Vapor pressure (atm at 298 K): 3.1 

Heat capacity (J/K-mol at 298 K): 86 + 

Heat of Vaporization (kJ/mol at nbp): 23.8 

Source: No commercial sources identified. 

Price: R. Du Boisson of PRC states that this should be easy to make but 

start up costs alone will make the cost in the range of $2000+ 
(amount does not matter greatly) 


Tbxicity: Davies et al. give an AD50 of 7%. Based on the average value of the AD50/LD50 observed 

for bromo-fluoro compounds the LD50 might be estimated to be in the range of 20%. 

Davies, R.H., Bagnall, R.D., Jones, W.G.M., "A Quantitative Interpretation of Phase Effects 
in Anaesthesia," Int. J. Quantum Chem: Quantum Biology Symp. No. 1, pp 201-212, 1974. 


Comments: This is the most ideal candidate for undergoing HBr elimination. The resulting perfluoro- 

ethene is not stable and would probably take up radicals in the flame adding to its 
effectiveness. This molecule should be given high priority in testing and is a strong candidate 
for fundamental research since data on the elimination reaction does not exist and estimation 
techniques for this class of reactions are not very accurate. The relatively low toxicity also 
makes this a strong candidate. 

+ Estimated for this report by authors. 


112 





Name: 2-bromo-l,l,l-trifluoroethane 


Compound group: Halogenaied hydrocarbons 

Formula: Cjh^BrFj 

Alternate Names: 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


421-06-7 

26 

1 + 

91+ 

27+ 

Aldrich Chemical, Milwaukee, WI 53201, (800)-558-9160 
$28/25g 


Toxicity: Robbins reports AD50 of 2.8% and LD50 of 11.7% 

Robbins, B.H., "Preliminary studies of the anesthetic activity of fluorinated hydrocarbons," J. 
Pharmacology & Experimental Therapeutics, vol. 86, pp 197-204, 1946. 


Comments: See comments on previous molecule. Increasing the fluorine content raised the AD50 from 

2.8% in this molecule to 7% in the previous molecule. These small differences in the 
molecules may change the toxicity a great deal and possibly only marginally affect the fire 
suppression. Note that in this case the increase in F atom content would be expected to 
increase the fire suppression. 

+ Estimated for this report by authors. 


113 





Name: l,2-dibromo-l,l,2-trifluoroethane 


Compound group: Halogenated hydrocarbons 

Formula: C^HBr^Fj 

Alternate Names: 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


354-04-1 

76 

0 . 2 + 

106 + 

31 f 

PCR, Inc., Gainesville, FL 32602, (800)-331-6313 
$145/250g 


Tbxicity: Davies et al. report AD50 of 0.6%, so that this molecule must be considered more toxic than 

most in the halogenated hydrocarbon group. 

Davies, R.H., Bagnall, R.D., Jones, W.G.M., "A Quantitative Interpretation of Phase Effects 
in Anaesthesia," Int. J. Quantum Chem: Quantum Biology Symp. No. 1, pp 201-212, 1974. 


Comments: Also a candidate for HBr elimination. Vapor pressure may be too low for many applications, 

but in adverse environments where wind shear of the suppressant stream is a factor, 
compounds with higher boiling points and therefore greater resistance to dispersion of the 
stream may be advantageous. 

* Estimated for this report by authors. 


114 





Name; l,2-dibromo-l,l-difluoroethane 


Compound group: Halogenated hydrocarbons 

Formula: Cyr^BrjFj 

Alternate Names: 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


75-82-1 

93 

0 . 1 + 

95 + 

33 + 

PCR, Inc., Gainesville, FL 32602, (800)-331-6313 
$25/100g 


Tbxicity: Sax and Lewis rate this as mildly toxic by inhalation. Data given indicate toxicity in rats at 

the 5 ppl level for long exposure. 

Sax, N.I.; Lewis, R.J.; "Dangerous Properties of Industrial Materials," Van Norstrand 
Reinhold, New York (1987). 


Comments: May also eliminate HBr. Boiling point is very high. 

+ Estimated for this report by authors. 


115 







Name: l,2-dibromo-l,2-difluoroethane 


Compound group: 
Formula: 
Alternate Names: 


Halogenated hydrocarbons 

^ 2 ^ 2 ® r 2^2 
halon 2202 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


20705-29-7 

102 * 

0.06+ 

97+ 

33* 

R. Du Boisson of PRC indicated that this compound is very difficult 
to make, 
na 


Tbxicity: 

Comments: 

* Estimated for this report by authors, na = not available. 


116 





Name: l-bromo-l,l,2,3,3,3-hexafluoropropane 


Compound group: 
Formula: 
Alternate Names: 


Halogenated hydrocarbons 
qHBrF 6 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


2252-78-0 

35.5 

0.5+ 

145+ 

29+ 

PCR, Inc., Gainesville, FL 32602, (800)-331-6313 
$85/1 OOg 


Ibxicity: Sax and Lewis have data on a related compound 3-bromo-l,l,2,2-tetrafluoropropane (CASN 

679-84-5) : human lowest toxic concentration was given as 40 ppt. Generally increasing the 
fluorine content at the expense of the hydrogen reduces toxicity so this compound may be 
expected to be less toxic. Davies et al. report an AD50 of <2% for the l-bromo-1,1,3,3,3- 
pentafluoro compound. Again, increasing the fluorine content should increase the safety of 
the compound. The data does seem to indicate that for the larger halocarbons the ratio of 
AD50 to LD50 is lower, but the data are not conclusive. 

Sax, N.I.; Lewis, R.J.; "Dangerous Properties of Industrial Materials," Van Norstrand 
Reinhold, New York (1987). 

Davies, R.H., Bagnall, R.D., Jones, W.G.M., "A Quantitative Interpretation of Phase Effects 
in Anaesthesia," Int. J. Quantum Chem: Quantum Biology Symp. No. 1, pp 201-212, 1974. 


Comments: This compound is one of the test cases for the HBr elimination. 


+ Estimated for this report by authors. 


117 





Name: l,3-dibromo-l,l,3,3-tetrafluoropropane 


Compound group: 
Formula: 
Alternate Names: 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


Halogenated hydrocarbons 
C3H 2 Br 2 F 4 


460-86-6 

118* 

0.04 + 

138+ 

35+ 

No commercial source was identified 
na 


Tbxicity: Tbxicity probably much greater that the 1-bromo-hexafluoropropane discussed above. 


Comments: Vapor pressure probably eliminates this compound as a practical suppressant, but it is 

included as an example of the propanes which may be capable of delivering much higher heat 
capacities. 

+ Estimated for this report by authors, na = not available. 


118 





Name: 2,2-dibromo-l,l,3,3-tetrafluoropropane 


Compound group: 
Formula: 
Alternate Names: 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


Halogenated hydrocarbons 
C3H 2 Br 2 F 4 


na 

120 + 

0.03+ 

137 + 

35 + 

No commercial source was identified 
na 


Tbxicity: Probably greater than 1-bromo-hexafluoro propane discussed above. 


Comments: 

* Estimated for this report by authors, na = not available. 


119 





Name: l-bromo-l,l,3,3,3-pentafluoropropane 


Compound group: Halogenated hydrocarbons 

Formula: C 3 H 2 BrF 5 

Alternate Names: 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


460-88-8 

51 + 

0.4 + 

133 + 

29 f 

No commercial source was identified 
na 


Toxicity: Probably greater than the 1-bromo-hexafluoro propane discussed above. 

Comments: 

t Estimated for this report by authors, na = not available. 


120 





Name: hexafluoracetone 


Compound group: 
Formula: 
Alternate Names: 


Halogenated Ketones, Anhydrides and Esters 
CF 3 COCF 3 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


684-16-2 

-26 

6.65 

118 

20.9 

PCR Incorporated, Gainesville, FL 32602 800-331-6313 
$175.00/100g 


Tbxicity: Hexafluoroacetone is given a hazard rating of 3 in Sax and Lewis. 

Sax, N. I., and Lewis, R. J. Dangerous Properties of Industrial Materials 7th Edition, Van 
Nostrand Reinhold, 1989 


Comments: A detailed description of the physics and chemistry of hexafluoroacetone can be found in a 

review by Krespan and Middleton 

Krespan, C. G. and Middleton, W. J., "Hexafluoroacetone" in Fluorine Chemistry Reviews. 
1, 145-196, 1967 


121 





Name: trifluoroacetic anhydride 


Compound group: Halogenated Ketones, Anhydrides and Esters 

Formula: CF 3 COOCOCF 3 

Alternate Names: 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


407-25-0 

40 

.4+ 

155 + 

28 f 

PCR Incorporated, Gainesville, FL 32602 800-331-6313 
$125.00/1 kg 


Toxicity: Thifluoroacetic anhydride has a hazard rating of 2 in Sax and Lewis. It is a severe skin and 

eye irritant. 

Sax, N. I., and Lewis, R. J. Dangerous Properties of Industrial Materials 7th Edition, Van 
Nostrand Reinhold, 1989 


Comments: 

"^Estimated for this report by authors. 


122 





Name: bis(perfluoroisopropyl)ketone 


Compound group: 
Formula: 
Alternate Names: 


Halogenated Ketones, Anhydrides and Esters 
(iC^feCO 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


813-44-5 

73 

,15 + 

280+ 

29.7 + 

PCR Incorporated, Gainesville, FL 32602 800-331-6313 
$160.00/25g 


Ibxicity: There are no toxicity data on this compound. Note however the toxic properties of 

hexafluoroacetone. 


Comments: Comparison of the fire suppressant capacity of this compound with hexafluoroacetone will 

highlight heat capacity effects. 


’•‘Estimated for this report by authors. 






Name: methyltrifluoroacetate 


Compound group: 
Formula: 
Alternate Names: 


Halogenated Ketones, Anhydrides and Esters 
CF 3 COOCH 3 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


431-47-0 

44 

.4 + 

113 + 

32 + 

PCR Incorporated, Gainesville, FL 32602 800-331-6313 
$130.00/1OOg 


Tbxicity: There are no toxicity data on this compound. From Sax and Lewis one notes that fluorinated 

esters have a general lower toxicity rating, 2 , than the ketones 

Sax, N. I., and Lewis, R. J. Dangerous Properties of Industrial Materials 7th Edition, Van 
Nostrand Reinhold, 1989. 


Comments: This compound is selected with the view that it will be studied in tandem with the brominated 

compound. 

^Estimated for this report by authors. 


124 





Name: 3-bromo-l,l,l-trifluoropropanone 


Compound group: Halogenated Ketones, Anhydrides and Esters 

Formula: CF 3 COCH 2 Br 

Alternate Names: 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


431-35-6 

86 

.08* 

113+ 

32 * 

PCR Incorporated, Gainesville, FL 32602 800-331-6313 
$140.00/100g 


Tbxicity: There are no toxicity data. Tbxicity properties should not be any better than the 

hexafluoroacetone. 


Comments: 1,1,1 trifluoroacetone can be purchased and it should be possible to brominate this 

compound. Comparison with the non-brominated compounds should be extremely interesting. 


^Estimated for this report by authors. 







Name: bromopentafluoroacetone 


Compound group: 
Formula: 
Alternate Names: 


Halogenated Ketones, Anhydrides and Esters 
BrCF 2 COCF 3 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


815-23-6 

31 

,9 + 

126 + 

26.4 

No commercial source was identified 
na 


Tbxicity: There are no toxicity data on bromopentafluoroacetone. However there is no reason to think 

that the reactivity of hexafluoroacetone will be decreased by bromination. 


Comments: Bromopentafluoroacetone can be prepared from the reaction of bromine with pentafluoropro- 

penol-2 following by beta-dehydrobromination of the pentafluoro-l,2ibromopropanol-2 by 
heating in N-methylpyrollidone. It is a liquid with an unpleasant odor, soluble in common 
solvents. In water it forms a hydrate. It is a Iachrymator. 

This compound is selected for comparison with hexafluoroacetone and will give direct 
information on the effect of bromine. 

Belenki, G. G., Fokin, A. V., Rondarev, D. S., Ryazanovam R. M., Sokolov, S. V., Sterlin, S. 
R., Voronkov, Zeifman, Y. V "Fluoroaliphatic Compounds" in Synthesis of Fluoroorganic 
Compounds. (Knunyants, I. L. and Yakobson, I. L, Editors) Springer-Verlag, New York, pp 
3- 108, 1985 


"^Estimated for this report by authors, na = not available 


126 





Name: bromomethyltrifluoroacetate 


Compound group: Halogenated Ketones, Anhydrides and Esters 

Formula: CF 3 COOCH 2 Br 

Alternate Names: 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


116478-92-4 

105 + 

.03 + 

126 + 

33.7 + 

No commercial source was identified 
na 


Tbxicity: There are no toxicity data. Tbxic properties may be similar to other esters. 


Comments: It should be possible to brominate the methyltrifluoroacetate. Studying the brominated and 

non-brominated compounds will lead to important information on the effect of bromine. 


^Estimated for this report by authors, na = not available 




Name: perfluoropropene 


Compound group: 
Formula: 
Alternate Names: 


Unsaturated Halocarbons 

C3F6 

Hexafluoropropene 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


116-15-4 

-29 

6.42 

116 

21 

PCR Incorporated, Gainesville, FL 32602 800-331-6313 
S 105.00/1 kg 


Tbxicity: Sax and Lewis assign a hazard rating of 3. It is rated as mildly toxic by inhalation, ihl-rat 

LC50:11200mg/m 3 /4H, ihl-mus LC50: 750 ppm/4H 

Larsen has summarized findings on the possible uses of this compound as an anaesthetic. 
With mixtures in the 50-75% range anaesthetic effects were mild. However convulsions and 
delayed death also occurred. 

Sax, N. I., and Lewis, R. J. Dangerous Properties of Industrial Materials 7th Edition, Van 
Nostrand Reinhold, 1989 

Larsen, E. R. "Fluorine Compounds in Anaesthesiology" in Fluorine Chemistry Reviews. 
(Thrrant, P., Editor.) Marcel Dekker, New York, 1969 


Comments: This compound is selected to determine the effect of double bonds on fire suppressant 

activity. It is a base for comparisons with various brominated compounds. 


128 





Name: perfluorobutene-2 


Compound group: 
Formula: 
Alternate Names: 


Unsaturated Halocarbons 
C 4 F 8 -2 

1,1,1,2,3,4,4,4,octafluorobutene 


CASN: 

Normal boiling (nbp°C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


360-89-4 

0 

2.62 

138 + 

21 + 

PCR Incorporated, Gainesville, FL 32602 800-331-6313 
$200.00/1OOg 


Tbxicity: Sax and Lewis assign a hazard rating of 1. It is mildly toxic by inhalation; ihl-rat LCLo:6100 

ppm/4H 

Sax, N. I., and Lewis, R. J. Dangerous Properties of Industrial Materials 7th Edition, Van 
Nostrand Reinhold, 1989 


Comments: The isomer perfluoroisobutene is a deadly poison by inhalation. In studies with this 

compound some thought should be given to the possibility of conversion to this toxic Lsomer. 

^Estimated for this report by authors. 


129 





Name: perfluorotoluene 


Compound group: 
Formula: 
Alternate Names: 


Unsaturated Halocarbons 

cf 3 c 6 f 5 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


434-64-0 

104 

.03 + 

207+ 

34 + 

PCR Incorporated, Gainesville, FL 32602 800-331-6313 
$156.00/50g 


Ibxicity: Perfluorotoluene is given a hazard rating of 1 by Sax and Lewis and is stated to be mildly 

toxic by inhalation. Larsen has summarized studies on the possible use of this compound as 
an anaesthetic and reported delayed death at levels as low as 0.5-0.9% 

Sax, N. I., and Lewis, R. J. Dangerous Properties of Industrial Materials 7th Edition, Van 
Nostrand Reinhold, 1989 

Larsen, E. R. "Fluorine Compounds in Anaesthesiology" in Fluorine Chemistry Reviews . 
(Tarrant, P., Editor.) Marcel Dekker, New York, 1969 


Comments: This compound has been selected in order to test the effects arising from a fully iluorinated 

aromatic structure. 


'''Estimated for this report by authors. 


130 





Name: 1,1,3,3,3-pentafluoropropene-l 


Compound group: Unsaturated Halocarbons 

Formula: CF 3 CH=CF 2 

Alternate Names: 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization at nbp (kJ/mol): 
Source: 

Price: 


690-27-7 

-21 

5 * 

109 f 

23* 

PCR Incorporated, Gainesville, FL 32602 800-331-6313 
$500.00/1OOg 


Tbxicity: 


There are no toxicity data. 


Comments: This compound has been selected for comparison with the fully fluorinated propene. 

Important information will also be obtained through comparison with the compound where 
the hydrogen is replaced by bromine. 


^Estimated for this report by authors. 





Name; 3,3,3-trifluoropropene 


Compound group: 
Formula: 
Alternate Names: 


Unsaturated Halocarbons 

cf 3 ch=ch 2 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


677-21-4 

-17 

4 + 

92 f 

24+ 

PCR Incorporated, Gainesville, FL 32602 800-331-6313 
$220.00/500g 


Tbxicity: There is no toxicity data. 


Comments: This compound has been selected for comparison with the fully fluorinated propene. 

Important information will also be obtained through comparison with the compound where 
the hydrogen is replaced by bromine. 

^Estimated for this report by authors. 


132 




Name: 1,2-bis (perfluoro-n-butyl) ethylene 


Compound group: Unsaturated Halocarbons 

Formula: (nC 4 F 9 )CH=CH(nC 4 F 9 ) 

Alternate Names: 


CASN: 84551-43-9 

Normal boiling (nbp 6 C): 132 

Vapor pressure (atm at 298 K): .006 t 

Heat capacity at (J/K-mol at 298 K): 377 f 

Heat of Vaporization (kJ/mol at nbp): 36 + 

Source: Produits Chimiques Ugine Kuhlmann, France 

Price: na 


Tbxicity: This compound has been used as a blood substitute. Short-term toxic properties should 

therefore be satisfactory 


Comments: The boiling point is probably too high for use as a fire suppressant. However the ready 

availability means that tests can be made immediately. The data summarized by Larsen 
indicate that flammability limits are lowered by the presence of the double bond. It will be 
very interesting to see how this is carried over to this compound which is extremely inert. 
A particularly interesting property is that it is readily emulsified. This may prove to be an 
interesting way of delivering the suppressant. 

Riess, J. G. and LeBlanc, M. "Pcrfluoro Compounds as Blood Substitutes" Angewandte 
Chemie . International Edition, 9, 621-634, 1978 


Larsen, E. R. "Fluorine Compounds in Anaesthesiology" in 
(Thrrant, P., Editor.) Marcel Dekker, New York, 1969 



Estimated for this report by authors, na = not available 





Name: 3-bromoperfluoropropene 


Compound group: Unsaturated Halocarbons 

Formula: BrCF 2 CF=CF 2 

Alternate Names: 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


431-56-1 

28 

1 + 

12lt 

27 f 

No commercial source was identified 
na 


Tbxicily: 


Comments: This compound is selected for comparison with the fully and partially fluorinated propenes. 

3-Bromoperfluoropropene can be synthesized through the reaction of 
dibromodifluoromethane in the presence of benzoyl peroxide at 100° C. The major product, 
3-dibromo-l,l,2,3,3-pentafluoropropane can be dehydro-brominated in the presence of KOH 
to form the desired compound. 

Thrrant, P. Lovelace, A. M. and Lilyquist, M. R. "Free Radical Additions Involving Fluorine 
Compounds," Journal of the American Chemical Society. 77, 2783-2787 


Coda, A, T., "Free Radical Addition to Tlifiuoroethylene," 
2995-2996, 26, 1961 



tEstimated for this report by authors, na = not available 


134 





Name: 1-bromoperfluoropropene 


Compound group: 
Formula: 
Alternate Names: 


Unsaturated Halocarbons 
CF 3 CF=CFBr 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


14003-53-3 

14003-61-3 

30 + 

1 + 

121 + 

27 + 

No commercial source was identified 


Tbxicity: 


There are no toxicity data. 


Comments: This compound has been selected for comparison with the fully and partially fluorinated 

substances that are also recommended for study. 


’’Estimated for this report by authors, na = not available 





Name: 1,2-bis (perfluoromethyl) ethylene 


Compound group: 
Formula: 
Alternate Names: 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


Unsaturated Halocarbons 
CF 3 CH=CHCF 3 


66711-86-2 

6 

1 . 6 + 

130+ 

26+ 

No commercial source was identified 
na 


"toxicity: There are no toxicity data for this compound. The perfluorobutyl analogy is a blood 

substitute and it may well be that the properties of this compound will also be favorable. 
However, it is known that the reactivity of these compound decreases as the size of the 
perfluorinated alkyls are increased. It may well be that there is an optimum size for fire 
suppression purposes. 

Riess, J. G. and LeBlanc, M. "Perfluoro Compounds as Blood Substitutes" Angewandte 
Chemie . International Edition, 9, 621-634, 1978 


Comments: This compound is proposed in the expectation that the brominated analog will also be 

studied. Larsen has shown that the presence of a double bond generally lowers the 
flammability limit. It is expected that bromination will raise this value. Comparisons with 
perfluorinated butene-2, as well as, 1,1,1,2,3,4,4,4-octafiuorobutane will provide an extremely 
important set of results regarding the role of double bond in flame suppression. 

This compound has been synthesized via the fluorination of fumaric acid by SF 4 at 130° C. 
An alternate synthetic path is telomerization of perfluoroethylene with methyl iodide. 

Larsen, E. R. "Fluorine Compounds in Anaesthesiology" in Fluorine Chemistry Reviews . 
(Thrrant, P., Editor.) Marcel Dekker, New York, 1969 

Hasek, W. R., Smith, W. C. and Engelhardt, V. A., "The Chemistry of Sulfur Tfetrafluoride. 
II. The Fluorination of Organic Carbonyl Compounds" Journal of the American Chemical 
Society. 82, 543-551, 1960 

+Estimated for this report by authors, na = not available 


136 






Name: l-bromoperfluoromethyI-2-perfluoromethylethylene 


Compound group: Unsaturated Halocarbons 

Formula: CF 2 BrHC=CHCF 3 

Alternate Names: 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


na 

57* 

. 1 + 

138 + 

30* 

No commercial source was identified 
na 


Ibxicity: There are no toxicity data. 

Comments: This compound has been selected for comparison with 1,2 bis(perfluoromethyl)ethylene. 

^Estimated for this report by authors, na = not available 


137 






Name: 1-bromo-bis (perfluoromethyl) ethylene 


Compound group: 
Formula: 
Alternate Names: 


Unsaturated Halocarbons 
CF 3 BrC=CHCF 3 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


400-41-9 

65 + 

.06+ 

143 + 

33 f 

No commercial source was available 
na 


Tbxicity: There are no toxicity data. 


Comments: This compound is selected with the expectation that it will be studied with the non- 

brominated analog. The comparison of the results from the two studies will yield information 
on the effect of bromine on fire suppression properties. 

It should be possible to synthesize this compound through the hydrobromination of 
perfluorobutyne-2. 

^Estimated for this report by authors, na = not available 


13 ! 





Name: tetris (perfluoromethyl) ethylene 


Compound group: Unsaturated Halocarbons 

Formula: (CF 3 ) 2 C=C(CF 3 ) 2 

Alternate Names: 


CASN: 

Normal boiling (nbp a C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


360-57-6 

55 

. 28 * 

180 

29 

No commercial source was available 
na 


Tbxicity: The tetris(perfluoro-n-butyl)ethylene version of this molecule is a blood substitute. This 

molecule is proposed for study in the expectation that it will have similar favorable 
properties. 

Geyer, R. P., "Perfluorochemicals as Oxygen Transport Vehicles" in Blood Substitutes 
(Chang, T. M. S. and Geyer, R. P., Editors) Marcel Dekker, New York, pp31-49, 1989 


Comments: Possible synthesis routes are outlined in the paper by Bell et al. 

Bell, A. N., Fields, R. and Haszeldine, R. N., "Fluoro-olefin Chemistry. Part 13. A Further 
Route to Perfluoro-2,3-dimethylbut-2-ene and the Photochemical Rearrangement of Some 
Perfluoro-Alkyl Olefins," J. Chem. Soc. 487, 1980 


* Estimated for this report by authors, na = not available 


139 





Name: tetrafluorodimethyl ether 


Compound group: Halogenated Ethers 

Formula: CF,HOHCF 2 

Alternate Names: 


CASN: 

Normal boiling (nbp “C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


1691-17-4 

2 

2 f 

92 f 

24.5'*' 

PCR Incorporated, Gainesville, FL 32602 800-331-6313 
$150.00/100g 


Tbxicity: There are no toxicity data. 


Comments: This is the starting material for bromination. Comparison with the brominated compound 

can lead to important mechanistic information on chemical fire suppressant mechanisms. 

^Estimated for this report by authors. 


140 





Name: pentafluordimethyl ether 


Compound group: 
Formula: 
Alternate Names: 


Halogenated Ethers 
CF 2 HOCF 3 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


3822-68-2 

-35 

7 + 

109 + 

21 . 2 + 

PCR Incorporated, Gainesville, FL 32602 800-331-6313 
$150.00/1OOg 


Tbxicity: There are no toxicity data. 


Comments: This is the starting material for bromination. Comparison with the brominated compound 

can lead to important mechanistic information on chemical fire suppressant mechanisms. 


^Estimated for this report by authors, na = not available 





Name: 2-chIoro-l-(difluoromethoxy)-l,l,2-trifluoroethane 


Compound group: 
Formula: 
Alternate Names: 


Halogenated Ethers 

CHF 2 OCF 2 CHFCl 

enflurane, enthrane, methylfluorether 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


13838-16-9 

56.5 

.29 

162 + 

29.4 + 


Anaquest, 2005 West Beltline Highway, Madison Wisconsin 53713- 
2318 

$400.00/750 cm 3 


Tbxicity: This compound is a widely used anaesthetic. Health effects are well documented. It is 

considered to be mildly toxic by inhalation, ingestion and subcutaneous routes and is given 
a hazard rating of 2 by Sax and Lewis 


Comments: As a successful anaesthetic it is disqualified as a fire suppressant. We have selected this 

compound in order to study it in tandem with the brominated version. 

"^Estimated for this report by authors. 


142 





Name: isoflurane 


Compound group: Halogenated Ethers 

Formula: CF 3 CHC!OCHF 2 

Alternate Names: 


CASN; 26775-46-7 

Normal boiling (nbp °C): 48.5 

Vapor pressure (atm at 298 K): .32 

Heat capacity at (J/K-mol at 298 K): 167 + 

Heat of Vaporization (kJ/mol at nbp): 28.7* 

Source: Anaquest, 2005 West Beltline Highway, Madison Wisconsin 53713- 

2318 

Price: $400.00/750 cm 3 


Tbxicity: This is a well known anaesthetic. A great deal of toxicity data exists. 


Comments: As an anaesthetic this is not an acceptable fire suppressant. However, it is known to be non¬ 

flammable in anaesthetic contexts and since it is readily available it will be interesting to 
compare this compound with its structural isomer, 2-chloro-l-(difluoromethoxy)-l,l,2- 
trifluoroethane or enflurane (see other data sheet). 


^Estimated for this report by authors. 





Name: perfluoro-2-butyItetrahydrofuran 


Compound group: Halogenated Ethers 

Formula: C 4 F 9 C 4 F 7 0 

Alternate Names: 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


335-36-4 

103 

.033 f 

na 

33.5 + 

PCR Incorporated, Gainesville, FL 32602 800-331-6313 
$85.00/500g 


Tbxicity: This compound has been used as a blood substitute. It should not have any short term health 

effects. 


Comments: The vapor pressure of this compound is probably too low for fire suppressant use. However, 

its availability and favorable toxic properties suggest that studies be carried out as a 
preliminary to further test with the side chain removed and with bromine substitution. 

‘''Estimated for this report by authors, na = not available 


144 





Name: bis(bromodifluoroethyl) ether 


Compound group: Halogenated Ethers 

Formula: BK^OCFjBr 

Alternate Names: 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


na 
25 + 

1 + 

117 + 

26.7 + 

No commercial source was identified 
na 


Tbxicity: 

Comments: This is one of the compounds that is being synthesized for EPA. First priority should be 

assigned to determining whether this compound can be photolyzed in the troposphere. 

^Estimated for this report by authors, na = not available 


145 




Name: l-bromo,l>l>3,3,3-pentafluorodimethyl ether 


Compound group: Halogenated Ethers 

Formula: CF 2 BrOCF 3 

Alternate Names: 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


na 

-30+ 

6 + 

104 + 

24.7 + 

No commercial source was identified 
na 


Tbxicity: There are no toxicity data. 


Comments: The compound is being synthesized through an EPA-supported program. An important issue 

is whether the addition of the bromine will change the photolytic property sufficiently. 

^Estimated for this report by authors, na = not available 


146 





Name: bromoenflurane 


Compound group: 
Formula: 
Alternate Names: 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


Halogenated Ethers 
CBrF 2 OCF 2 CHFCl 


na 

67 f 

. 2 + 

175 + 

30.3 + 

No commercial source was identified 
na 


Tbxicity: There are no toxicity data. The nonbrominated version of the compound is the well known 

anaesthetic enflurane. One does not expect that toxicity will decrease with behavior. 


Comments: The nonbrominated compound is easily obtainable. Bromination should be straightforward. 

Studies on the fire suppressant capabilities of the pair of compounds should be extremely 
informative. 

^Estimated for this report by authors, na = not available 


147 





Name: octafluorofuran 


Compound group: 
Formula: 
Alternate Names: 


CASN: 

Normal boiling (nbp e C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


Halogenated Ethere 

c 4 f 8 o 


773-14-8 

50 f 

,33 + 

151 + 

28.8 + 

No commercial source was identified 
na 


Tbxicity: There are no toxicity data. The perfluorobutyl substituted compound is a blood substitute. 

This may mean that the toxic properties of this compound will be within acceptable limits. 


Comments: It may be possible synthesize this compound through the electrofluorination of tetrahydro- 


^Estimated for this report by authors, na = not available 


148 





Name: 3-bromoperfluorofuran 


Compound group: 
Formula: 
Alternate Names: 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


Tbxicity: There are no toxicity 


Halogenated Ethers 
BrC 4 F 7 0 


na 

105 + 

.03 + 

159 + 

33,7’*" 

No commercial source was identified 
na 


this compound. 


Comments: It may be possible to synthesize this compound through the electrofluorination of 

telrahydrofuran and then brominating the incompletely fluorinated compounds. 


^Estimated for this report by authors, na = not available 






Name: bis(perfluoromethyl) thioether 


Compound group: 
Formula: 
Alternate Names: 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


Miscellaneous 

CF 3 SCF 3 


371-78-87 

- 22.2 

6.1 

126* 

25.1* 

No commercial source was identified 
na 


Tbxicity: 


No toxicity data are available. However, note that the thio-analog of the well known 
anaesthetic methoxyflurane has been synthesized and tested and showed enhanced properties. 
This general trend may hold for all such substitutions. 


Le Blanc, M. and Riess, J. G., "Artificial Blood Substitutes Based on Perfluorochemicals” in 
Preparation. Properties and Industrial Applications of Organofluorine Compounds, (Banks, 
R. E., Editor.) Halsted Press, New York, pp 83-138 ,1982 


Comments: Bis(perfluoromethyl)thioether is not available from any commercial source. It should be 

possible to synthesize the compound via electrochemical fluorination (Simons Process) 
starting with dimethyl ether. The process is not complete. Thus it should also be possible to 
recover thioethers with one or more hydrogens. These should be convenient starting products 
for bromination, if these compounds are necessary for subsequent study. 

This compound has also been prepared via the photolysis of the disulfide. 

Brandt, G. A. R., Emeleus, H. J., and Haszeldine, R. N., "Organometallic and Organo- 
metalloidal Fluorine Compounds," Journal of the Chemical Society. 2198-2205, 1952 

*Estimated for this report by authors, na = not available 


150 





Name: tris(perfluoromethyl) amine 


Compound group: 
Formula: 
Alternate Names: 


Miscellaneous 

(CF 3 ) 3 N 


CASN: 

Normal boiling (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity at (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


432-03-1 

- 10.1 

3.9 

167 + 

25.1 

No commercial source was available 
na 


Tbxicity: No toxicity data are available. Higher members of this series are used as blood substitutes. 

This may be an indication that there will be no serious health effects at least in the short 
term. 


Comments: Tfis(perfluoromethyl) amine is not available from any commercial source. It is readily 

prepared from trimethyl amine via electrochemical fluorination (Simons Process). Should 
initial tests on these compounds prove to be satisfactory the next step will be to replace one 
or two of the fluorine by bromine. This can be accomplished by brominating the hydrogens 
from the incomplete fluorination of the trimethyl amine. 

Pawelke, G., Heyder, F. and Burger, H., "Halogen Exchange of Fluoroalkyl Amines , 
Synthesis of Polychloro- and Bromo-Th'alkylamines Journal of Fluorine Chemistry . 20, 53-63, 
1982. 

^Estimated for this report by authors, na = not available 


151 




Name: iodotrifluoromethane 


Compound group: 
Formula: 
Alternate Names: 


Halons Containing Iodine 

cf 3 i 

perfluoromethyl iodide, trifluoromethyl iodide, halon 13001 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


2314-97-8 

-22.5 

na 

70.9 

22 t 

Aldrich Chemical Company, Milwaukee, WI, 53233, 800-558-9160 
$255.90/100 g 


Tbxicity: No published data were identified for this compound. Researchers should use caution. 


Comments: Fire suppression references: 

Lask, G. and Wagner, H.Gg., "Influence of Additives on the Velocity of Laminar Flames," 
Eight Symposium (International^ on Combustion, pp. 432-438, Williams and Wilkins, 
Baltimore, 1962. 

Sheinson, R.S., Gellene, G.I., Williams, F.W., and Hahn, J.E., "Quantification of Fire 
Suppression Action on Liquid Pool Fires," Abstract for paper presented at the 1978 Fall 
Technical Meeting of the Eastern Section of the Combustion Institute, Miami, FL, 29 
November-1 December 1978. 

Sheinson, R.S., Penner-Hahn, J.E., and Indritz, D., "The Physical and Chemical Action of Fire 
Suppressants, Fire Safety Journal, accepted for publication. 

^Estimated for this report by authors, na = not available 


152 





Name: chlorodifluoroiodomethane 


Compound group: 
Formula: 
Alternate Names: 


Halons Containing Iodine 

CF 2 C1I 

halon 12101 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


420-49-5 

33 

na 

76 + 

27 + 

No commercial source was identified, 
na 


Tbxicity: No published data were identified for this compound. Researchers should use caution. 


Comments: A preparative procedure is available in the literature. 

Burton, D.J., Shin-Ya, S., and Kesling, H.S., "Preparation of Halo-F-Methanes via Potassium 
Fluoride-Halogen Cleavage of Halo-F-Methylphosphonium Salts," Journal of Fluorine 
Chemistry, vol. 20, pp. 89-97, 1982. 

No fire suppression measurements were identified for this compound. 

^Estimated for this report by authors, na = not available 






Name: l-bromo-l,l,2,2-tetrafluoro-2-iodoethane 


Compound group: 
Formula: 
Alternate Names: 


Halons Containing Iodine 
CF 2 BrCF 2 I 

l-bromo-2-iodotetrafluoromethane, halon 24011 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


471-70-5 

78 

na 

124 f 

31 f 

PCR Incorporated, Gainesville, FL 32602, 800-331-6313 
S195/100 g 


Tbxicity: No published data were identified for this compound. Researchers should use caution. 


Comments: No fire suppression measurements were identified for this compound. 


^Estimated for this report by authors, na = not available 





Name: l,l,2,2-tetrafluoro-l,2-diiodoethane 


Compound group: 
Formula: 
Alternate Names: 


Halons Containing Iodine 

cf 2 icf 2 i 

halon 24002 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


354-65-4 

112 

na 

126 + 

34 + 

PCR Incorporated, Gainesville, FL 32602, 800-331-6313 
$165/100 g 


Tbxicity: No published data were identified for this compound. Researchers should use caution. 


Comments: No fire suppression measurements were identified for this compound. 


^Estimated for this report by authors, na = not available 


155 





Name: iodomethane 


Compound group: 
Formula: 
Alternate Names: 


Halons Containing Iodine 
CH 3 I 

methyl iodide, halon 10001 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


74-88-4 

42.4 
0.53 
44.1 

27.4 

Aldrich Chemical Company, Milwaukee, WI, 53233, 800-0558-9160 
$74.70/500 g 


Tbxicity: Sax and Lewis: HR:3, High toxic, fire, or reactivity hazard Poison, suspected carcinogen, a 

human skin irritant, human mutagenic data. 

Sigma Aldrich: Carcinogen, highly toxic, vesicant, mutagen 

Sax, N.I. and Lewis, Sr., R.J., Dangerous Properties of Industrial Materials. 7th Ed., Van 
Nostrand Reinhold, New York, 1989. 

Lenga, R.E., The Sigma-Aldrich Library of Chemical Safety Data, Sigma-Aldrich Corporation, 
1985. 


Comments: Fire suppression references: 

de C. Ellis, O.C., "Extinction of Petrol Fires by Methyl Iodide," Nature , vol. 161, pp. 402-403, 
13 March 1948. 

"Final Report on Fire Extinguishing Agents for the Period September 1, 1947 to June 30, 
1950," Purdue Research Foundation and Department of Chemistry, Purdue University, West 
Lafayette, IN, July 1950. 

Rosser, W.A., Wise, H., and Miller, J., "Mechanism of Combustion Inhibition by Compounds 
Containing Halogen," Seventh Symposium ('International'! on Combustion , pp. 175-182, 
Butterworths, London, 1958. 

Homann, K.H. and Poss, R., "The Effect of Pressure on the Inhibition of Ethylene Flames," 
Combustion and Flame, vol. 18, pp. 300-302, April 1972. 

Westbrook, C.K., "Inhibition of Hydrocarbon Oxidation in Laminar Flames and Detonations 
by Halogenated Compounds," Nineteenth Symposium (TnternationaO on Combustion, pp. 
127-141, The Combustion Institute, Pittsburgh, 1982. 


156 





Name: iodoethane 


Compound group: 
Formula: 
Alternate Names: 


Halons Containing Iodine 

ch 3 ch 2 i 

ethyl iodide, halon 20001 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


75-03-6 

72.3 
0.18 
67.8 

29.4 

Aldrich Chemical Company, Milwaukee, WI, 53233, 800-558-9160 
$107.50/5 g 


Tbxicity: Sax and Lewis: HR:3, High toxic, fire, or reactivity hazard 

Sigma Aldrich: Tbxic, severe irritant, vesicant 

Sax, N.I. and Lewis, Sr., R.J., Dangerous Properties of Industrial Materials. 7th Ed., Van 
Nostrand Reinhold, New York, 1989. 

Lenga, R.E., The Sigma-Aldrich Library of Chemical Safety Data, Sigma-Aldrich Corporation, 
1985. 


Comments: Fire suppression references: 

Jorissen, W.P., Booy, J., and van Heiningen, J., "Reaction-Regions XXII. On the Prevention 
of Explosive Reactions in Gas and Vapour Mixtures by Small Amounts of Various 
Substances," Recueil des Travaux Chimiques des Pay-Bas . vol. 51, pp. 868-877, 15 July 1932. 

"Final Report on Fire Extinguishing Agents for the Period September 1, 1947 to June 30, 
1950," Purdue Research Foundation and Department of Chemistry, Purdue University, West 
Lafayette, IN, July 1950. 

Westbrook, C.K., "Inhibition of Hydrocarbon Oxidation in Laminar Flames and Detonations 
by Halogenated Compounds," Nineteenth Symposium (International - ) on Combustion, pp. 
127-141, The Combustion Institute, Pittsburgh, 1982. 


157 






Name: 1-iodopropane 


Compound group: 
Formula: 
Alternate Names: 


Halons Containing Iodine 

ch 3 ch 2 ch 2 i 

propyl iodide, halon 30001 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


107-08-4 

102.4 

0.6 

88.3 

32.1 

Aldrich Chemical Company, Milwaukee, WI, 53233, 800-558-9160 
$47.30/500 g 


Tbxicity: Sax and Lewis: HR:3, High toxic, fire, or reactivity hazard 

Poison by intraperitoneal route, experimental neoplastigen, very mildly toxic by inhalation 

Sigma Aldrich: Harmful if swallowed, inhaled, or absorbed through skin; may cause 
irritation; chronic effects: carcinogen. 

Sax, N.I. and Lewis, Sr., R.J., Dangerous Properties of Industrial Materials. 7th Ed., Van 
Nostrand Reinhold, New York, 1989. 

Lenga, R.E., The Sigma-Aldrich Library of Chemical Safety Data, Sigma-Aldrich Corporation, 
1985. 


Comments: Fire suppression references: 

Rosser, W.A., Wise, H., and Miller, J., "Mechanism of Combustion Inhibition by Compounds 
Containing Halogen," Seventh Symposium (Internationall on Combustion, pp. 175-182, 
Butterworths, London, 1958. 


158 



Name: l,l,l>2,2,3,3-heptafluoro-3-iodopropane 


Compound group: 
Formula: 
Alternate Names: 


Halons Containing Iodine 
CF 3 CF 2 CF 2 I 

perfluoropropyl iodide, halon 37001 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


754-34-7 

40 

na 

197+ 

28 + 

Aldrich Chemical Company, Milwaukee, WI, 53233, 800-558-9160 
$76.70/25 g 


Toxicity: No published data were identified for this compound. Researchers should use caution. 


Comments: No fire suppression measurements were identified for this compound. 


+ Estimated for this report by authors, na = not available 








Name: sulfur fluoride 


Compound group: 
Formula: 
Alternate Names: 


Sulfur halides 


SF 6 

sulfur hexafluoride 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


2551-62-4 
-63.8 (sublimes) 

21.8 

97.0 

23.6 

Aldrich Chemical Company, Milwaukee, WI, 53233, 800-558-9160 
$96.40/227 g 


Toxicity: Sax and Lewis: HR:1, Low toxic, fire, or reactivity hazard 

Chemically inert and considered to be physiologically inert as well. 

Sax, N.I. and Lewis, Sr., R.J., Dangerous Properties of Industrial Materials. 7th Ed., Van 
Nostrand Reinhold, New York, 1989. 


Comments: Fire suppression references: 

"Final Report on Fire Extinguishing Agents for the Period September 1, 1947 to June 30, 
1950," Purdue Research Fbundation and Department of Chemistry, Purdue University, West 
Lafayette, IN, July 1950. 

Moran, Jr., H.E. and Bertschy, A.W., Flammability Limits for Mixtures of Hydrocarbon Fuels. 
Air, and Halogen Compounds. NRL Report 4121, Engineering Research Branch, Chemistry 
Division, Naval Research Laboratory, Washington, DC, 25 February 1953. 

Miller, W.J., "Inhibition of Low Pressure Flames," Combustion and Flame , vol. 13, pp. 210- 
212, April 1969. 

Sheinson, R.S., Penner-Hahn, J.E., and Indritz, D., "The Physical and Chemical Action of Fire 
Suppressants, Fire Safety Journal, accepted for publication. 



Name: sulfur chloride fluoride 


Compound group: 
Formula: 
Alternate Names: 


Sulfur halides 
SF 5 C1 

sulfur chloropentafluoride 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


13780-57-9 

-21 

na 

104 

21.7 

PCR Incorporated, Gainesville, FL 32602, 800-331-6313 
$175/25 g ($110 cylinder charge) 


Toxicity: Sulfur chloride fluoride has been cited as being as toxic as phosgene by Gage. Phosgene is 

given a HR rating of 3 (high toxic, fire, or reactivity hazard) by Sax and Lewis and is 
described as a human poison. 

Gage, J.C., "Subacute inhalation toxicity of 109 Industrial Chemicals," British Journal of 
Industrial Medicine , vol. 27, pp. 1-18, 1970. 

Sax, N.I. and Lewis, Sr., R.J., Dangerous Properties of Industrial Materials. 7th Ed., Van 
Nostrand Reinhold, New York, 1989. 


Comments: Fire suppression references: 

Sheinson, R.S., Penner-Hahn, J.E., and Indritz, D., "The Physical and Chemical Action of Fire 
Suppressants, Fire Safety Journal , accepted for publication. 


na = not available 






Name: sulfur bromide fluoride 


Compound group: Sulfur halides 

Formula: SF 5 Br 

Alternate Names: sulfur bromopentafluoride, pentafluorosulfur bromide 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


15607-89-3 

3.1 

na 

107 

24+ 

No commercial source was identified, 
na 


Tbxicity: No toxicity data were identified in the literature. By analogy with sulfur chloride fluoride, 

this compound is likely to be at least as toxic as phosgene. 


Comments: A preparative procedure is available in the literature. 

Kovacina, T.A., Berry, A.D., and Fox, W.B., "Improved Preparation and Purification of 
Pentafluorosulfur Bromide," Journal of Fluorine Chemistry, vol. 7, pp. 430-432, 1976. 

Fire suppression references: 

Sheinson, R.S., Penner-Hahn, J.E., and Indritz, D., "The Physical and Chemical Action of Fire 
Suppressants, Fire Safety Journal , accepted for publication. 

+Estimated for this report by authors, na = not available 


162 






Name: phosphorus bromide difluoride 


Compound group: 
Formula: 
Alternate Names: 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


phosphorus-containing compounds 

PF 2 Br 

bromodifluorophosphine 


15597-40-7 

-16 

4.5 

64 + 

23.9 

No commercial source was identified, 
na 


Tbxicity: No toxicity information was identified for this compound. By analogy with other phospho¬ 

rous halides it is very likely that it would be assigned a HR rating of 3 using the Sax and 
Lewis (high toxic, fire, or reactivity hazard) system. 

Sax, N.I. and Lewis, Sr., R.J., Dangerous Properties of Industrial Materials. 7 Ed., Van 
Nostrand Reinhold, New York, 1989. 


Comments: This compound is included on the list to test mechanisms for fire suppression. It is a 

dangerous and unstable species. Particular care must be exercised in its synthesis and testing. 

A standard preparative procedure is available in the literature. 

Morse, J.G., Cohn, K., Rudolph, R.W., and Parry, R.W., ”3B Phosphorous. 22. Substituted 
Difluoro- and Dichlorophosphines," Inorganic Syntheses, vol. 10, pp. 147-156, 1967. 

No fire suppression measurements were identified for this compound. 

'''Estimated for this report by authors, na = not available 


165 




Name; phosphoryl fluoride 


Compound group: phosphorus-containing compounds 

Formula: POF 3 

Alternate Names: 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


13478-20-1 

-40 

13.2 

68.8 

22.1 

Atomergic Chemetals Corporation, Plainview, NY 

516-349-8800 

$400/200 g 


11803, 


Tbxicity: 


No toxicity information was identified for this compound. By analogy with other phosphoryl 
halides it is very likely that it would be assigned a HR rating of 3 using the Sax and Lewis 
(high toxic, fire, or reactivity hazard) system. 


Sax, N.I. and Lewis, Sr., R.J., Dance 
Nostrand Reinhold, New York, 1989. 


7 Ed., Van 


Comments: No fire suppression measurements were identified for this compound. 


166 




Name: phosphoryl chloride 


Compound group: 
Formula: 
Alternate Names: 


phosphorus-containing compounds 
POCl 3 

phosphorus oxychloride, phosphorous oxytrichloride 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


10025-87-3 

106 

0.05 

84.9 

33.7 

Aldrich Chemical Company, Milwaukee, WI, 53233, 800-558-9160 
$207.05/100 g 


Tbxicity: Sax and Lewis: HR:3, High toxic, fire, or reactivity hazard 

Poison by inhalation and ingestion; a corrosive eye, skin, and mucous irritant. 

Sax, N.l. and Lewis, Sr., R.J., Dangerous Properties of Industrial Materials. 7th Ed., Van 
Nostrand Reinhold, New York, 1989. 


Comments: Fire suppression references: 

Miller, W.J., "Inhibition of Low Pressure Flames," Combustion and Flame, vol. 13, pp. 210- 
212, April 1969. 

Jorissen, W.P., Booy, J., and van Heiningen, J., "Reaction-Regions XXII. On the Prevention 
of Explosive Reactions in Gas and Vapour Mixtures by Small Amounts of Various 
Substances," Recueil des Travaux Chimiques des Pav-Bas . vol. 51, pp. 868-877, 15 July 1932. 


167 




Name: phosphoryl bromide fluoride 


Compound group: 
Formula: 
Alternate Names: 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


phosphorus-containing compounds 
POF 2 Br 


14014-18-7 

32 

0.77 

75 + 

29.7 

No commercial source was identified, 
na 


Tbxicity: No toxicity information was identified for this compound. By analogy with other phosphoryl 

halides it is very likely that it would be assigned a HR rating of 3 (high toxic, fire, or 
reactivity hazard) using the Sax and Lewis system. 

Sax, N.I. and Lewis, Sr., R.J., Dangerous Properties of Industrial Materials. 7 Ed., Van 
Nostrand Reinhold, New York, 1989. 


Comments: This compound is included on the list to test mechanisms for fire suppression. It is a 

dangerous and unstable species. Particular care must be exercised in its synthesis and testing. 

A preparative procedure is available in the literature. 

Bernstein, P.A, Hohorst, F.A., Eisenberg, M., and DesMartcau, D.D., "Preparation of Pure 
Difluorophosphoric Acid and p-Oxo-bis(phosphoryl difluoride)," Inorganic Chemistry, vol. 
10, pp. 1549-1551, July 1971. 

No fire suppression measurements were identified for this compound. 

^Estimated for this report by authors, na - not available 


168 





Name: tetrachlorosilane 


Compound group: 
Formula: 
Alternate Names: 


Silicon and Germanium 


SiCl 4 

Silicon tetrachloride 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price (per kg): 


10026-04-07 

58 

0.31 

145 

32 


Liquid Carbonic, Corporation, Chicago, IL, 60603, 

312-855-2500 

Sll 


Tbxicity: Sax and Lewis assign a hazard rating of 3. (high toxic, fire, explosive, or reactivity hazard). 

Sax, N. I. and Lewis, R. J., Dangerous Properties of Industrial Materials. Seventh Edition, Van 
Nostrand Reinhold, New York, 1989. 


Comments: Fire suppression references: 

"Final Report on Fire Extinguishing Agents for the Period September 1, 1947 to June 30, 
1950," Purdue Research Fbundation and Department of Chemistry, Purdue University, West 
Lafayette, IN, July 1950. 

Lask, G. and Wagner, H. Gg., "Influence of Additives on the Velocity of Laminar Flames," 
Eighth Symposium (International! on Combustion , pp. 432-438, Williams and Wilkins, 
Baltimore, 1962. 

Morrison, M. E. and Scheller, K., "The Effect of Burning Velocity Inhibitors on the Ignition 
of Hydrocarbon-Oxygen-Nitrogen Mixtures," Combustion and Flame , vol. 18, pp. 3-12,1972. 


169 





Name: trichlorofluorosilane 


Compound group: 
Formula: 
Alternate Names: 


Silicon and Germanium 
SiCljF 


CASN: 

Normal boiling point (nbp a C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mole at 298 K): 
Heat of Vaporization (kJ/mole at nbp): 
Source: 

Price: 


14965-52-77 

15 

1.46 

92 

26 

No commercial source was identified 
na 


Tbxicity: High. Halosilanes hydrolyze in moist air - forming hydrogen halides in the process. 

Rochow, E. G. in Comprehensive Inorganic Chemistry, vol. 11, Pergamon Press, New York, 
1973. 


Comments: No fire suppression measurement data was uncovered. 

References pertaining to the synthesis of this compound can be found in: 

Ebsworth, E. A V., Volatile Silicon Compounds . Academic Press, New York, NY, 1975. 


na = not available. 




Name: tetrafluorosilane 


Compound group: 
Formula: 

Name: 


Silicon and Germanium 
SiF 4 

Silicon tetrafluoride 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price (per kg): 


7783-61-1 

-65 


1510 

74 

22 


Liquid Carbonic, Corporation, Chicago, IL, 60603, 

312-855-2500 

S250 


Tbxicity: Sax and Lewis assign a hazard rating of 3 (high toxic, fire, explosive, or reactivity hazard). 

Sax, N. I. and Lewis, R. J., Dangerous Properties of Industrial Materials , Seventh Edition, Van 
Nostrand Reinhold, New York, 1989. 


Comments: 


No fire suppression measurement data was uncovered. 





Name: bromotrifluorosilane 


Compound group: Silicon and Germanium 

Formula: SiBrF 3 

Alternate Names: 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price (per kg): 


14049-39-9 

-42 

10.5 

80 t 

18 

No commercial source was identified 
na 


Tbxicity: 


High. Halosilancs have a tendency to hydrolyze in moist air - forming hydrogen halides in the 
process. 


Rochow, E. G. in 
1973. 



, vol. 11, Pergamon Press, New York, 


Comments: No fire suppression measurement data was uncovered. 

References pertaining to the synthesis of this compound can be found in: 

Ebsworth, E. A. V, Volatile Silicon Compounds . Academic Press, New York, NY, 1975. 

fEstimated for this report by authors, na = not available 


172 





Name: tribromofluorosilane 


Compound group: 
Fbrmula: 
Alternate Names: 


Silicon and Germanium 
SiBr 3 F 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298): 

Heat of Vaporization (kJ/mole at nbp): 
Source: 

Price (per kg): 


18356-67-7 

84 

0.11 

150 + 

34 

No commercial source was identified 
na 


Tbxicity: High. Halosilanes have a tendency to hydrolyze in moist air - forming hydrogen halides in the 

process. 

Rochow, E. G. in Comprehensive Inorganic Chemistry, vol. 11, Pergamon Press, New York, 
1973. 


Comments: No fire suppression measurement data were uncovered. References pertaining to the synthesis 

of this compound can be found in: 

Ebsworth, E. A. V., Volatile Silicon Compounds, Academic Press, New York, NY, 1975. 


fEstimated for this report by authors, na = not available 





Name: tetramethylsilane 


Compound group: 
Formula: 
Alternate Names: 


Silicon and Germanium 
Si(CH 3 ) 4 


TMS 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 

Heal of Vaporization (kJ/mole at nbp): 
Source: 

Price (per kg): 


75-76-3 

27 

0.95 

204 

27 


Aldrich Chemical Company, Milwaukee, WI, 53233, 

800-558-9160 

S185 


Tbxicity: High. 

The Sigma-Aldrich Library of Chemical Safety Data , First Edition (Lenga, R. E., Editor), 
Sigma-Aldrich, 1985. 


Comments: Fire suppression references: 

Lask, G. and Wagner, H. Gg., "Influence of Additives on the Velocity of Laminar Flames," 
Eighth Symposium flnternationall on Combustion , pp. 432-438, Williams and Wilkins, 
Baltimore, 1962. 

Morrison, M. E. and Scheller, K., "The Effect of Burning Velocity Inhibitors on the Ignition 
of Hydrocarbon-Oxygen-Nitrogen Mixtures," Combustion and Flame , vol. 18, pp. 3-12, 1972. 


174 





Name: chlorotrimethylsilane 


Compound group: 
Formula: 
Alternate Names: 


Silicon and Germanium 
Si(CH 3 ) 3 Cl 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price (per kg): 


75-77-4 

57 

0.31 

175 + 

32 


Aldrich Chemical Company, Milwaukee, WI, 53233, 

800-558-9160 

$39 


Tbxicity: Sax and Lewis assign a hazard rating of 3 (high toxic, fire, explosive, or reactivity hazard). 

Sax, N. I. and Lewis, R. J,, Dangerous Properties of Industrial Materials, Seventh Edition, Van 
Nostrand Reinhold, New York, 1989. 


Comments: No fire suppression measurement data was uncovered. 


fEstimated for this report by authors 




Name: trichloromethylsilane 


Compound group: Silicon and Germanium 

Formula: Si(CH 3 )Cl 3 

Alternate Names: 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price (per kg): 


75-79-6 

66 

0.23 

100 f 

31 


Aldrich Chemical Company, Milwaukee, WI, 53233, 

800-558-9160 

$29 


Tbxicity: Sax and Lewis assign a hazard rating of 3 (high toxic, fire, explosive, or reactivity hazard). 

Sax, N. I. and Lewis, R. J., Dangerous Properties of Industrial Materials . Seventh Edition, Van 
Nostrand Reinhold, New York, 1989. 


Comments: No fire suppression measurement data was uncovered. 

fEstimated for this report by authors 


176 




Name; chloromethyltrimethylsilane 


Compound group: 
Formula: 
Alternate Names: 


Silicon and Germanium 
Si(CH 3 )CH 2 Cl 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price (per kg): 


2344-80-1 

98 

na 

200 t 

na 

Aldrich Chemical Company, Milwaukee, WI, 53233, 

800-558-9160 

$1246 


Tbxicity: High. 

The Sigma-Aldrich Library of Chemical Safety Data. First Edition (Lenga, R. E., Editor), 
Sigma-Aldrich, 1985. 


Comments: No fire suppression measurement data was uncovered. 


tEstimated for this report by authors, na = not available. 





Name: tetrachlorogermane 


Compound group: 
Formula: 
Alternate Names: 


Silicon and Germanium 


GeCl 4 

Germanium tetrachloride 


CASN: 10038-98-9 

Normal boiling point (nbp °C): 83 

Vapor pressure (atm at 298 K): 0.11 

Heat capacity (J/K-mol at 298 K): 150 f 

Heat of Vaporization (kJ/mol at nbp): 35 

Source: Aldrich Chemical Company, Milwaukee, WI, 53233, 

800-558-9160 

Price (per kg): $1652 


Tbxicity: Sax and Lewis assign a hazard rating of 3 (high toxic, fire, explosive, or reactivity hazard). 

Sax, N. I. and Lewis, R. J., Dangerous Properties of Industrial Materials. Seventh Edition, Van 
Nostrand Reinhold, New York, 1989. 


Comments: Fire suppression references: 

Lask, G. and Wagner, H. Gg., "Influence of Additives on the Velocity of Laminar Flames," 
Eighth Symposium ('International! on Combustion , pp. 432-438, Williams and Wilkins, 
Baltimore, 1962. 

Morrison, M. E. and Scheller, K., "The Effect of Burning Velocity Inhibitors on the Ignition 
of Hydrocarbon-Oxygen-Nitrogen Mixtures," Combustion and Flame, vol. 18, pp. 3-12, 1972. 


fEstimated for this report by authors. 


178 





Name: tetramethylgermane 


Compound group: 
Formula: 
Alternate Names: 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 

Heat of Vaporization (kJ/mole at nbp): 
Source: 

Price (per kg): 


Silicon and Germanium 
Ge(CH 3 ) 4 


865-52-1 

44 

0.52 

200 + 

30 

No commercial source was identified 
na 


Tbxicity: No information on the toxicity of this specific chemical was found. However, related 

compounds are not particularly toxic. 

Sax, N. I. and Lewis, R. J., Dangerous Properties of Industrial Materials . Seventh Edition, Van 
Nostrand Reinhold, New York, 1989. 


Comments: No fire suppression data have been reported for this compound. 

References pertaining to the synthesis of this compound can be found in: 
Glockling, F., The Chemistry of Germanium. Academic Press, New york, NY, 1969. 


fEstimated for this report by authors, na = not available. 




Name: sodium hydrogen carbonate 


Compound group: 
Fbrmula: 
Alternate Names: 


Metallic 

NaHCO a 

Sodium bicarbonate 


CASN: 

Normal boiling point (nbp °C): 
Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of vaporization (kJ/mol at nbp): 
Source: 

Price (per kg): 


144-55-8 

decomposes 

0 + 

88 

na 

Aldrich Chemical Company, Milwaukee, WI, 53233, 

800-558-9160 

S16 


Tbxicity: Low, Sax and Lewis assign a hazard rating of 1. 

Sax, N. I. and Lewis, R. J,, Dangerous Properties of Industrial Materials . Seventh Edition, Van 
Nostrand Reinhold, New York, 1989. 


Comments: Fire suppression references: 

Birchall, J. D., "On the Mechanism of Flame Inhibition by Alkali Metal Salts," Combustion 
and Flame, vol. 14, pp 85-96, 1970. 

Friedman, R. and Levy, J. B., "Inhibition of Opposed - Jet Methane - Air Diffusion Flames. 
The Effects of Alkali Metal Vapors and Organic Halides," Combustion and Flame, vol. 7, pp. 
195-201, 1963. 

fEstimated for this report by authors, na = not available. 


180 





Name: sodium acetate 


Compound group: 
Formula: 
Alternate Names: 


Metallic 

NaGjHgO-j 

Acetic acid, sodium salt 


CASN: 

Normal boiling point (nbp *C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of vaporization (kJ/mol at nbp): 
Source: 

Price: 


127-09-3 

decomposes 

0 + 

80 

na 

Aldrich Chemical Company, Milwaukee, WI, 53233, 

800-558-9160 

$36 (per kg) 


Tbxicity: Medium. Sax and Lewis assign a hazard rating of 2. 

Sax, N. I. and Lewis, R. J., Dangerous Properties of Industrial Materials . Seventh Edition, Van 
Nostrand Reinhold, New York, 1989. 


Comments: Fire suppression references: 

Dewitte, M., Vrebosch, J. and van Tiggelen, A., "Inhibition and Extinction of Premixed Flames 
by Dust Particles,* Combustion and Flame, vol. 8, pp. 257-266, 1964. 


fEstimated for this report by authors, na = not available. 






Name: potassium hydrogen carbonate 


Compound group: 
Formula: 
Alternate Names: 


Metallic 

KHC0 3 

Potassium bicarbonate 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of vaporization (kJ/mol at nbp): 
Source: 

Price (per kg): 


298-14-6 

decomposes 

0 + 


na 

na 

Aldrich Chemical Company, Milwaukee, WI, 53233, 


800-558-9160 

$24 


Tbxicity: Low. 


Comments: Fire suppression references: 

Birchall, J. D., "On the Mechanism of Flame Inhibition by Alkali Metal Salts," Combustion 
and Flame, vol. 14, pp 85-%, 1970. 

Friedman, R. and Levy, J. B., "Inhibition of Opposed - Jet Methane - Air Diffusion Flames. 
The Effects of Alkali Metal Vapors and Organic Halides," Combustion and Flame, vol. 7, pp. 
195-201, 1963. 


fEstimated for this report by author, na = not available. 


182 





Name: potassium oxalate 


Compound group: 
Formula: 
Alternate Names: 


Metallic 

k 2 c 2 o 4 *h 2 o 

Potassium oxalate monohydrate 


CASN: 

Normal boiling point (nbp °C): 
Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of vaporization (kJ/mol at nbp): 
Source: 

Price (per kg): 


6487-48-5 

decomposes 

0 + 


na 

na 

Aldrich Chemical Company, Milwaukee, WI, 53233, 


800-558-9160 

$46 


Tbxicity: Sax and Lewis assign a hazard rating of 3. (high toxic, fire, explosive, or reactivity hazard). 

Sax, N. I. and Lewis, R. J., Daneerous Properties of Industrial Materials. Seventh Edition, Van 
Nostrand Reinhold, New York, 1989. 


Comments: Fire suppression references: 

Friedrich, M. "Extinguishment Action of Powders," Fire Research Abstracts and Reviews, vol. 
2, pp. 132-135, 1960. 


Dewitte, M., Vrebosch, J. and van Tiggelen, A., "Inhibition and Extinction of Premixed Flames 
by Dust Particles," Combustion and Flame , vol. 8, pp. 257-266, 1964. 


fEstimated for this report by authors, na = not available. 





Name: potassium acetate 


Compound group: 
Formula: 
Alternate Names: 


Metallic 

K2C2H5O2 

Acetic acid, potassium salt 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of vaporization (kJ/mol at nbp): 
Source: 

Price (per kg): 


127-08-2 

decomposes 

ot 

na 

na 

Aldrich Chemical Company, Milwaukee, WI, 53233, 

800-558-9160 

S39 


Ibxicity: Medium. Sax and Lewis assign a hazard rating of 2. 

Sax, N. I. and Lewis, R. J., Dangerous Properties of Industrial Materials. Seventh Edition, Van 
Nostrand Reinhold, New York, 1989. 


Comments: Fire suppression references: 

Dewitte, M., Vrebosch, J. and van Tiggelen, A, "Inhibition and Extinction of Premixed Flames 
by Dust Particles," Combustion and Flame , vol. 8, pp. 257-266, 1964. 


fEstimated for this report by authors, na = not available. 


184 






Name: potassium acetylacetonate 


Compound group: 
Formula: 
Alternate Names: 


Metallic 

KC 5 H 7 0 2 # 1 /2H 2 0 

Potassium acetylacetonate hemihydrate 


CASN: 

Normal boiling point (nbp °C): 
Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of vaporization (kJ/mol at nbp): 
Source: 

Price (per kg): 


57402-46-7 

decomposes 

na 

na 

na 

Aldrich Chemical Company, Milwaukee, WI, 53233, 

800-558-9160 

S300 


Tbxicity: High. 


Comments: No fire suppression measurement data were uncovered. 


na = not available 




Name: chromium acetylacetonate 


Compound group: 
Formula: 
Alternate Names: 


Metallic 

Cr ( C 5 H 7°2)3 


CASN: 

Normal boiling point (nbp “C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of vaporization (kJ/mol at nbp): 
Source: 

Price (per kg): 


21679-31-2 

340 


Aldrich Chemical Company, Milwaukee, WI, 53233, 
800-558-9160 
$120 


Toxicity: High. 


Comments: Fire suppression references: 


Vanpee, M. and Shirodkar, P. P., "A Study of Flame Inhibition by Metal Compounds,' 
Seventeenth Symposium (International) in Combustion , pp 787-793, 1978. 


na = not available. 


186 




Name: chromyl chloride 


Compound group: Metallic 

Formula: Cr0 2 Cl 2 

Alternate Names: 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of vaporization (kJ/mol at nbp): 
Source: 

Price (per kg): 


14977-61-8 

116 

0.024 

na 

41 

Aldrich Chemical Company, Milwaukee, Wl, 53233, 

800-558-9160 

$2216 


Tbxicity: Sax and Lewis assign a hazard rating of 3. (high toxic, fire, explosive, or reactivity hazard). 

Sax, N. 1. and Lewis, R. J., Dangerous Properties of Industrial Materials , Seventh Edition, Van 
Nostrand Reinhold, New York, 1989. 


Comments: Fire suppression references: 

Lask, G. and Wagner, H. Gg., "Influence of Additives on the Velocity of Laminar Flames," 
Eighth Symposium ('International! on Combustion , pp. 432-438, Williams and Wilkins, 
Baltimore, 1962. 

Morrison, M. E. and Scheller, K., "The Effect of Burning Velocity Inhibitors on the Ignition 
of Hydrocarbon-Oxygen-Nitrogen Mixtures," Combustion and Flame, vol. 18, pp. 3-12, 1972. 


na = not available. 






Name: tin (IV) chloride 


Compound group: 
Formula: 
Alternate Names: 


Metallic 

SnCl 4 

Stannic chloride 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of vaporization (kJ/mol at nbp): 
Source: 

Price (per kg): 


7646-78-8 

114 

0.030 

165 

37 


Aldrich Chemical Company, Milwaukee, WI, 53233, 

800-558-9160 

S45 


Tbxicity: Sax and Lewis assign a hazard rating of 3. (high toxic, fire, explosive, or reactivity hazard). 

Sax, N. I. and Lewis, R. J., Dangerous Properties of Industrial Materials . Seventh Edition, Van 
Nostrand Reinhold, New York, 1989. 


Comments: Fire suppression references: 

Lask, G. and Wagner, H. Gg., "Influence of Additives on the Velocity of Laminar Flames," 
Eighth Symposium (International! on Combustion , pp. 432-438, Williams and Wilkins, 
Baltimore, 1962. 

Morrison, M. E. and Scheller, K., "The Effect of Burning Velocity Inhibitors on the Ignition 
of Hydrocarbon-Oxygen-Nitrogen Mixtures," Combustion and Flame, vol. 18, pp. 3-12, 1972. 





Name: titanium (IV) chloride 


Compound group: Metallic 

Formula: TiCl 4 

Alternate Names: 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of vaporization (kJ/mol at nbp): 
Source: 

Price (per kg): 


7550-45-0 

136 

0.015 

145 

39 


Aldrich Chemical Company, Milwaukee, WI, 53233, 

800-558-9160 

$40 


Ibxicity: Sax and Lewis assign a hazard rating of 3. (high toxic, fire, explosive, or reactivity hazard). 

Sax, N. I. and Lewis, R. J., Dangerous Properties of Industrial Materials . Seventh Edition, Van 
Nostrand Reinhold, New York, 1989. 


Comments: Fire suppression references: 

Lask, G. and Wagner, H. Gg., "Influence of Additives on the Velocity of Laminar Flames," 
Eighth Symposium ('International') on Combustion , pp. 432-438, Williams and Wilkins, 
Baltimore, 1962. 

Morrison, M. E. and Scheller, K., "The Effect of Burning Velocity Inhibitors on the Ignition 
of Hydrocarbon-Oxygen-Nitrogcn Mixtures," Combustion and Flame, vol. 18, pp. 3-12, 1972. 




Name: tetraethyl lead 


Compound group: 
Formula: 
Alternate Names: 


Metallic 

Pb(C>H 5 ) 4 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of vaporization (kJ/mol at nbp): 
Source: 

Price (per kg): 


78-00-2 

200 

0.00055 

na 

54 

Alfa Products, Danvers, MA, 01923, 800-342-0660 
$15500 


Tbxicity: Sax and Lewis assign a hazard rating of 3. (high toxic, fire, explosive, or reactivity hazard). 

Sax, N. L and Lewis, R. J., Dangerous Properties of Industrial Materials . Seventh Edition, Van 
Nostrand Reinhold, New York, 1989. 


Comments: Fire suppression references: 

Lask, G. and Wagner, H. Gg., "Influence of Additives on the Velocity of Laminar Flames," 
Eighth Symposium (International) on Combustion , pp. 432-438, Williams and Wilkins, 
Baltimore, 1962. 

Morrison, M. E. and Scheller, K., "The Effect of Burning Velocity Inhibitors on the Ignition 
of Hydrocarbon-Oxygen-Nitrogen Mixtures," Combustion and Flame , vol. 18, pp. 3-12, 1972. 


na = not available 





Name: iron pentacarbonyl 


Compound group: 
Formula: 
Alternate Names: 


Metallic 
Fe(CO) 5 
Iron carbonyl 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of vaporization (kJ/mol at nbp): 
Source: 

Price (per kg): 


13463-40-6 

103 

0.040 

241 

38 


Aldrich Chemical Company, Milwaukee, WI, 53233, 

800-558-9160 

$79 


Ibxicity: Sax and Lewis assign a hazard rating of 3. (high toxic, fire, explosive, or reactivity hazard). 

Sax, N. I. and Lewis, R. J,, Dangerous Properties of Industrial Materials . Seventh Edition, Van 
Nostrand Reinhold, New York, 1989. 


Comments: Fire suppression data references: 

Lask, G. and Wagner, H. Gg., "Influence of Additives on the Velocity of Laminar Flames,” 
Eighth Symposium (International) on Combustion, pp. 432-438, Williams and Wilkins, 
Baltimore, 1962. 

Morrison, M. E. and Scheller, K., "The Effect of Burning Velocity Inhibitors on the Ignition 
of Hydrocarbon-Oxygen-Nitrogen Mixtures," Combustion and Flame , vol. 18, pp. 3-12, 1972. 

Vanpee, M. and Shirodkar, P. P., "A Study of Flame Inhibition by Metal Compounds," 
Seventeenth Symposium (International! in Combustion , pp 787-793, 1978. 





Name: nitrogen 


Compound group: 
Formula: 
Alternate Names: 


inert gases 

n 2 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


7727-37-9 

-195.8 

na 

29.3 

5.6 

Matheson Gas Products, Baltimore, MD 21227,301-796-0517 
(99.998%) $42/8.3 kg 


Tbxicity: Sax and Lewis: HR:1, low toxic, fire, or reactivity hazard. 

Low toxicity; in high concentrations it is a simple asphyxiant. 


Comments: The effects of nitrogen on flames have been investigated extensively. A few representative 

references are provided. 

Coward, H.F. and Hartwell, F.J., "Extinction of Methane Flames by Diluent Gases," Journal 
of the Chemical Society , pp. 1522-1532, 1926. 

Simmons, R.F. and Wolfhard, H.G., "Some Limiting Oxygen Concentrations for Diffusion 
Flames in Air Diluted with Nitrogen," Combustion and Flame , vol. 1, pp. 155-161, 1957. 

Thtem, P.A., Gann, R.G., and Carhart, H.W., "Pressurization with Nitrogen as an Extinguish- 
ant for Fires in Confined Spaces," Combustion Science and Technology , vol. 7, pp. 213-218, 
1973. 

Ishizuka, S. and Tuji, H., "An Experimental Study of Effect of Inert Gases on Extinction of 
Laminar Diffusion Hames," Eighteenth Symposium (International! on Combustion , pp. 695- 
703, The Combustion Institute, Pittsburgh, 1981. 

Ticker, D.M., Drysdale, D.D., and Rasbash, D.J., "The Extinction of Diffusion Flames Burning 
in Various Oxygen Concentration by Inert Gases and Bromotrifluormethane." Combustion and 
Flame, vol. 41, pp. 292-300, 1981. 

na = not available. 


192 





Name: carbon dioxide 


Compound group: inert gases 

Formula: C0 2 

Alternate Names: 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


124-38-9 
-75.8 (sublimes) 

29.9 

37.7 

15.3 

Matheson Gas Products, Baltimore, MD 21227,301-796-0517 
(99.9 9 %) $95/27.2 kg 


Tbxicity: Sax and Lewis: HR:1, Low toxic, fire, or reactivity hazard 

An asphyxiant; an experimental teratogen. 


Comments: The effects of carbon dioxide on flames have been investigated extensively. A few 

representative references are provided. 

Coward, H.F. and Hartwell, F.J., "Extinction of Methane Flames by Diluent Gases," Journal 
of the Chemical Society , pp. 1522-1532, 1926. 

Egerton, A and Powling, J., "The Limits of Flame Propagation at Atmospheric Pressure. II. 
The Influence of Changes in the Physical Properties," Proceedings of the Royal Society of 
London , vol. A197, pp. 190-209, May 27, 1948. 

"flicker, D.M., Drysdale, D.D., and Rasbash, D.J., "The Extinction of Diffusion Flames Burning 
in Various Oxygen Concentration by Inert Gases and Bromotrifluormethane." Combustion and 
Flame, vol. 41, pp. 292-300, 1981. 


193 






Name: argon 


Compound group: 
Formula: 
Alternate Names: 


inert gases 
At 


CASN: 

Normal boiling point (nbp °C): 

Vapor pressure (atm at 298 K): 

Heat capacity (J/K-mol at 298 K): 
Heat of Vaporization (kJ/mol at nbp): 
Source: 

Price: 


7440-37-1 

-185.4 

na 

20.9 

6.5 

Matheson Gas Products, Baltimore, MD 21227,301-796-0517 
(99.998%) $77/13.5 kg 


Tbxicity: Sax and Lewis: HR:1, low toxic, fire, or reactivity hazard 

A simple asphyxiant gas. 

Comments: The effects of argon on flames have been investigated extensively. A few representative 

references are provided. 

Coward, H.F. and Hartwell, F.J., "Extinction of Methane Flames by Diluent Gases," Journal 
of the Chemical Society , pp. 1522-1532, 1926. 

Egerton, A and Powling, J., "The Limits of Flame Propagation at Atmospheric Pressure. II. 
The Influence of Changes in the Physical Properties," Proceedings of the Royal Society of 
London, vol. A197, pp. 190-209, May 27, 1948. ~ 

Ishizuka, S. and Tiuji, H., "An Experimental Study of Effect of Inert Gases on Extinction of 
Laminar Diffusion Flames," Eighteenth Symposium (International! on Combustion, pp. 695- 
703, The Combustion Institute, Pittsburgh, 1981. 


na = not available.