Appendix C
Stability and Materials Compatibility of Candidate Replacements for Halon
Storage Stability
Halon 1301 is normally stored in metal containers. It is known to be stable under these conditions for many years. Any candidate replacement must also be stable with respect to storage in metal containers for many years. The National Institute of Standards and Technology (NIST) has conducted two studies of the storage stability of candidate replacement agents. The first of these, by Peacock et al.,1 examined the agents listed in Table C.1. Each agent was used to fill a Teflon™-lined stainless steel cylinder containing the metal coupons listed in Table C.2.
Table C.1 Agents Examined in the National Institute of Standards and Technology Study
Agent |
Chemical Formula |
HCFC-22 |
CF2HCl |
HCFC-124 |
CF3-CHFCl |
HCFC-125 |
CF3-CHF2 |
HFC-32/HFC-125 |
CF2H2/CF3-CHF2 |
HFC-134a |
CF3-CHF2 |
FC-218 |
CF3-CF2-CF3 |
HFC-227ea |
CF3-CHF-CF3 |
FC-31-10 |
CF3-CF2-CF2-CF3 |
FC-116 |
CF3-CF3 |
FC-C-318 |
Cyclo-C4F8 |
HFC-236fa |
CF3-CH2-CF3 |
Halon 1301 |
CF3Br |
IFC-13I1 |
CF3I |
Table C.2 Metals Tested in the National Institute of Standards and Technology Agent Stability Study
Metal |
Major Constituents |
Typical Use |
Nitronic-40 (21-6-9 stainless) |
Cr, Mn, Ni, Fe |
Pressure vessels |
304 stainless |
Cr, Ni, Fe |
Pressure vessels and fittings |
AM 355 stainless |
Cr, Mo, Ni, Fe |
Outlet closures |
4130 alloy steel |
Cr, Mn, Si, Fe |
Pressure vessels |
13-8 Mo alloy steel |
A1, Cr, Mo, Ni, Fe |
|
Inconel 625 |
Cr, Fe, Mo, Ni |
Bourdon tubes |
606 l-T6 aluminum alloy |
A1, Si, Fe, Mg |
Compression gaskets and fittings |
CDA 172 Cu-Be alloy |
Cu, Be |
Bourdon tubes, outlet closures |
The storage cylinders were placed in 149°C ovens for 28 days. The coupons were weighed and checked for appearance before and after testing, and the agent was analyzed using Fourier transform infrared (FT-IR) spectroscopy before and after testing. Peacock et al. concluded the following:
- "No new compounds, observable above background, were evident after 28 day, 150°C exposure of any of the candidate agents."
- "A possible decrease in integrated areas for selected spectral peaks was observed only for CF3I. This observation may be within experimental error, and was without accompanying formation of new compounds. An observation of a dark solid on metal coupons post-exposure for this agent may have been I2 which could result from either degradation or impurity in the original agent. Longer term study is warranted for this agent."
- "For the chemicals studies, stability in long-term storage and agent residue should not be major deciding factors in determining selection of appropriate agents for further study. As noted above, CF3I could be an exception."
In a second study, Harris2 examined FC-218, HFC-125, HFC-227ea, and CF3I more closely, that is, over a longer time and at more than one temperature (see Table C.3).
All agents were tested in the presence of Nitronic-40, Ti-15-3-3-3, C4130, and Inconel 625. The storage vessels were stored at the prescribed temperatures for as many as 52 weeks. At specific times the cylinders were removed from storage, cooled, and the contents analyzed by FT-IR. After conducting the study, Harris concluded:
- "The fluorocarbon agents FC-218, HFC-125, and HFC-227ea were stable at temperatures as high as 150°C for as long as 48 weeks. No by-products were formed."
- "CF3I degraded at 100°C and was accelerated at 150°C."
- "CF3H, CO2, and CO were produced in low levels as degradation products of CF3I."
- "The presence of moisture accelerates the degradation of CF3I."
- "The presence of copper accelerates the degradation of CF3I."
- "The presence of copper and moisture accelerate the degradation of CF3I."
- "An abundance band at 950 cm-1 was generated in the CF3I samples that may be from a fluorinated alkene; the presence of copper at 150°C caused the double bond to break."
- "Storage at ambient conditions of any of the four agents is feasible, but storage at elevated temperatures for CF3I needs more study."
Table C.3 Test Matrix for Harris Study
Agent |
Temperature |
FC-218 |
150°C |
HCFC-125 |
23°C and 150°C |
HFC-227ea |
23°C, 100°C, and 150°C |
CF3I (dry) |
23°C, 100°C, and 150°C |
CF3I with Cu (dry) |
23°C and 150°C |
CF3I (moist) |
150°C |
CF3I with Cu (moist) |
150°C |
Effects of Halon-Like Alternative Agents on Organic Materials and Metals
Storage Considerations
In considering of the storage of halon-like agents, the long-term effects of the agent on elastomeric sealing gaskets and lubricants in the storage system must be considered. In this connection, a specific evaluation of various fluorocarbons (FCs), hydrofluorocarbons (HFCs), and hydrochlorofluorocarbons (HCFCs), acting on relevant elastomeric gasket materials and lubricants, has been carried out by McKenna et al.3 at NIST. The elastomers investigated included silicones, fluorosilicones, fluorocarbons, neoprene, and nitriles. The lubricants included a fluorinated grease, a perfluoropolyether grease, and a commercial aircraft grease.
Two methods of evaluation were employed. The first method involved measurement of the degree of swelling of the elastomer or lubricant on exposure to agent vapor. If the amount of solvent absorbed was small, the agent was reckoned to have good compatibility; i.e., it did not damage the elastomer or the lubricant. Bad compatibility was implied by excessive swelling, and fair compatibility was intermediate between the two. The measurements were carried out at 35°C. This method is indirect in the sense that it does not measure the effect on the mechanical properties of the elastomer or lubricant. Good compatibility implies equilibrium sorption of a weight fraction less than 0.22, while bad compatibility implies sorption greater than 0.38.
The second method involved measurement of mechanical properties of the elastomers and lubricants after extended exposure (weeks) to agent vapor at elevated temperature and pressure (150°C, 5.86 Mpa). The authors concluded that testing at 150°C was too severe, but the data are indicative if not interpreted absolutely. For elastomers the mechanical properties are compression set resistance, i.e., the tendency to spring back after compression, and elongation reduction in ultimate tensile testing, i.e., the loss of stretchability. These mechanical factors are directly relevant to gasket performance. Bad compatibility is indicated when the ultimate elongation decreases by 65% after a 2-week exposure to agent. Fair compatibility, i.e., marginally acceptable, is indicated by a 65% loss in 4 weeks.
Lubricants responded to exposure to an agent in a different manner. The mobile fraction of the lubricant is extracted by the agent over time, leaving a powdery residue that is unsuitable as a lubricant. Bad compatibility signifies an agent that leaves the lubricant powdery after 4 weeks of exposure. Good compatibility implies that the lubricant does not become powdery after 6 weeks of exposure.
Results are given in Tables C.4 and C.5. The first letter (B, F, or G) is the poorer of the two mechanical ratings, i.e., compression set resistance or reduction of elongation. The second letter is based on swelling measurements. In drawing conclusions from the ratings one must be aware of the authors' admonition that the exposure conditions prior to mechanical testing (150°C) were too severe. Even so, until further test results are available, the information may be used for tentative screening. Examination of Table C.4 suggests that neither of the nitriles is promising as a gasket material with halon-like agents. Fluorocarbon elastomers are similarly unsuitable. Silicone, fluorosilicone, and neoprene elastomers emerge as superior candidates, but there are sufficient negative entries to make further testing necessary before selecting a material. In assessing the compatibility of lubricants, the mechanical property measure is probably most relevant. That is, the tendency to become powdery appears to be the failure mechanism, and swelling may not measure this tendency. Even so, the mechanism may be more complex, and further testing under milder conditions of exposure is indicated. The data in Table C.5 do not offer unambiguous clues for distinguishing among the three lubricants.
The study of McKenna et al. is an excellent beginning but, as the authors point out, further testing is required. Their study contains a wealth of interpretation not included herein. The results do lead to optimism that effective, long-lived gasket and lubricating materials can be found for a given halon-like fire fighting agent. It should be noted that other classes of elastomers could be considered (e.g., polyacrylates and polyphosphazenes). It is recommended that specific tests be carried out before designing a system.
Table C.4 Elastomer Compatibility
Agent |
Silicone |
F-silicone |
Neoprene |
Fluorocarbon |
Nitrile |
Nitrile (low T) |
FC-116, C2F6 |
G/G |
G/G |
G/G |
B/G |
F/G |
B/G |
FC-218, C3F8 |
G/G |
G/G |
G/G |
B/G |
F/G |
B/G |
FC-31-10, C4F10 |
G/F |
F/G |
F/G |
B/F |
B/F |
B/F |
FC-318, c-C4F8 |
G/G |
G/G |
G/G |
G/G |
F/G |
F/F |
HFC-236fa, C3H2F6 |
G/F |
G/B |
B/B |
B/B |
B/F |
B/B |
HFC-32-325* |
G/F |
B/G |
F/G |
B/G |
B/G |
B/G |
HFC-227ea, C3HF7 |
G/F |
G/F |
F/G |
B/F |
B/G |
B/B |
HCF-134a, C2H2F4 |
G/F |
G/F |
F/G |
B/G |
B/G |
B/G |
HFC-125, C2HF5 |
G/G |
B/G |
G/G |
B/F |
B/G |
B/G |
HCFC-22, CHF2Cl |
B/F |
G/F |
B/G |
B/F |
B/B |
F/B |
HCFC-124, C2HF4Cl |
B/B |
G/F |
B/F |
B/F |
B/F |
B/F |
NOTE: G, F, and B mean good, fair (i.e., between good and bad) and bad, respectively. In this context, a good solvent is one that does not diminish the desired behavior of an elastomer or lubricant. Two measures are given for each, representing the separate mechanical and sorption methods of characterization. See text. * HFC-32-325 is an azeotrope of CH2F2 and C2HF5. |
Table C.5 Lubricant Compatibility
Agent |
Krytox 240AC |
Braycote 600 |
Braycote 807 |
FC-116, C2F6 |
F/F |
G/F |
F/F |
FC-218, C3F8 |
B/G |
B/G |
B/G |
FC-31-10, C4F10 |
B/F |
B/B |
B/B |
FC318, c-C4F8 |
B/F |
B/F |
B/F |
HFC-236fa, C3H2F6 |
G/G |
G/G |
G/G |
HFC-33-125* |
G/G |
G/G |
G/G |
HFC-227ea, C3HF7 |
B/F |
G/F |
G/F |
HFC-134a, C2H2F4 |
B/G |
B/F |
B/G |
HFC-125, C2HF5 |
G/F |
G/G |
G/G |
HCFC-22, CHF2Cl |
B/F |
B/G |
F/F |
HCFC-124, C2HF4Cl |
G/G |
G/G |
G/G |
NOTE: G, F, and B mean good, fair (i.e., between good and bad) and bad, respectively. In this context, a good solvent is one that does not diminish the desired behavior of an elastomer or lubricant. Two measures are given for each, representing the separate mechanical and sorption methods of characterization. See text. * HFC-32-125 is an azeotrope of CH2F2 and C2HF5. |
The possibility of corrosion of the storage and distribution equipment induced by halon-like agents has been studied by Ricker et al.4 as reported in the NIST document cited above. These authors document studies that indicate that the halon-like agents do not pose serious corrosion problems for metals likely to be employed. Again, specific tests are recommended before implementation of materials choices.
Effects on Plastics and Other Organic Materials
The halon-like alternative agents tend to be chemically inert under most anticipated storage and discharge conditions. They would not be expected to exhibit deleterious effects on organic materials present, for example, in equipment or in protected spaces. Examples of organic materials that may be of concern include wire insulation, packaging for solid-state circuitry, circuit boards, floor coverings, paint, furniture coatings, and so on. Machinery spaces aboard naval ships, in general, have less exposed organic material than is typical of other occupied spaces.
Owing to the short exposure time associated with agent discharge in a fire fighting incident, no significant chemical deterioration of exposed organic materials is expected. A mode of failure called environmental stress cracking does exist, and it may be prudent to conduct specific tests on relevant plastics under stress in the presence of the agent in critical areas such as wire insulation. For materials used in typical shipboard machinery spaces, the likelihood of material failure resulting from agent discharge is very low. The reader is referred to the Modern Plastics Encyclopedia5 and a document from the American Society for Testing Materials6 for more detail.
Effects of Discharged Agents on Metals
During flame extinguishment, some acid is formed by the decomposition of halons and halon-like alternative agents. This effect is smaller when the time to extinguishment is shorter. Thus, the release of a larger quantity of agent can result in a smaller quantity of acidic decomposition by-products such as HF.
References
1. R.D. Peacock, T.G. Cleary, and R.H. Harris, Jr., pp. 643-668 in Evaluation of Alternative In-Flight Fire Suppressants for Full-Scale Testing in Simulated Aircraft Engine Nacelles and Dry Bays, NIST Special Pub. 861, U.S. Department of Commerce, Washington, D.C. (1994).
2. R.H. Harris, Jr., Fire Suppression System Performance of Alternative Agents in Aircraft Engine and Dry Bay Laboratory Simulations, Vol. 1, NIST Special Pub. 890, pp. 249-403, U.S. Department of Commerce, Washington, D.C. (1995).
3. Evaluation of Alternative In-Flight Fire Suppressants for Full-Scale Testing in Simulated Aircraft Engine Nacelles and Dry Bays, W.L. Grosshandler, R.G. Gann, and W.M. Pitts, Eds., pp. 729-763, U.S. Department of Commerce, Washington D.C. (1994).
4. Evaluation of Alternative In-Flight Fire Suppressants for Full-Scale Testing in Simulated Aircraft Engine Nacelles and Dry Bays, W.L. Grosshandler, R.G. Gann, and W.M. Pitts, Eds., pp. 669-728, U.S. Department of Commerce, Washington D.C. (1994).
5. Modern Plastics Encyclopedia, McGraw-Hill, New York (1996).
6. American Society for Testing Materials (ASTM), Resistance of Plastics to Chemical Reagents, D543, ASTM, West Conshohocken, Pa. (1995).