NFPA 69-2014 Edition Standard on Explosion Prevention Systems TIA Log No.: 1211 Reference: 7.2.3.1.2, 7.2.3.1.1(new), A.3.3.25, and Table C.1(a) Comment Closing Date: February 19, 2016 Submitters: Martin Clouthier, Clouthier Risk Engineering, and Laurence Britton, AIChE Fellow and Process Safety Consultant 1. Revise Subparagraph 7.2.3.1.2 to read as follows: 7.2.3.1.2 For gases and vapors, if the LOC values according to ASTM E 2079 are not available, then the LOC values obtained in flammability tubes shall be used after adjustment by subtracting 1.5 2 percent by volume oxidant for LOC values of 10 percent or greater and by multiplying by a factor of 0.85 for LOC values less than 10 percent, as indicated in the adjusted columns in Table C.1(a). 2. Add new Subparagraph 7.2.3.1.3 to read as follows: 7.2.3.1.3 In no case shall the adjusted LOC value for carbon dioxide inerting result in a value lower than that required for nitrogen inerting. 3. Revise Annex A.3.3.25 to read as follows: A.3.3.25 Limiting Oxidant Concentration (LOC). Materials other than oxygen can act as oxidants. The LOC depends upon the temperature, pressure, and fuel concentration as well as the type of diluent. Preliminary results of the ASTM E 2079, Standard Test Methods for Limiting Oxygen (Oxidant) Concentration in Gases and Vapors, round robin tests for gases and vapors revealed that the LOC data that were obtained using different test methods and that are listed in a majority of reference publications are nonconservative. The old Bureau of Mines data were obtained mostly in a 50 mm diameter flammability tube. This diameter might be too small to mitigate the flame-quenching influence, thereby impeding accurate determination of the LOC of most fuels. The 4 L minimum volume specified in ASTM E 2079 would correspond to a diameter of at least 200 mm (7.9 in.). As a result, some LOC values determined using this standard are approximately 1 percent by volume oxygen lower than the previous values measured in the flammability tube, and a few are even up to 1.5 2 percent by volume lower. The lower LOC values obtained in larger chambers are more appropriate for use in fire and explosion hazard assessment studies. A data comparison can be found in Table A.3.3.25. 4. Replace Table C.1(a) and associated notes with the following:
Proposed Table C.1(a) for NFPA 69 Temporary Interim Amendment Table C.1(a): Limiting Oxidant Concentrations for Flammable Gases When Nitrogen or Carbon Dioxide Are Used as Diluents Updated or (Adjusted) Data
Original Data
N2 -Air Mixture
CO2 -Air Mixture
N2 -Air Mixture
CO2 -Air Mixture
Gas/Vapor
LOC
Note
LOC
Note
LOC
Note
LOC
Note
Paraffins (alkanes) Methane Ethane Propane n-Butane Isobutane (methylpropane) n-Pentane Isopentane (2-methylbutane) n-Hexane n-Heptane
11.1 (9.5) 10.7 (10.6) (10.5) (10.6) (10.5) (10.4) (10.0)
a b a b b b c b c
(13.1) (11.9) (12.8) (13.0) (13.3) (12.9) (13.0) (13.0) (13.0)
b b b b b b c b c
12.1 11.0 11.4 12.1 12.0 12.1 12.0 11.9 11.5
b b b b b b c b c
14.6 13.4 14.3 14.5 14.8 14.4 14.5 14.5 14.5
b b b b b b c b c
Cycloparaffins (cycloalkanes, naphthenes) Cyclopropane
(10.2)
b
(12.4)
b
11.7
b
13.9
b
Olefins (alkenes) Ethylene (ethene) Propylene (propene) α-butylene (1-butene) Isobutylene (2-methylpropene) Isopentene (3-methyl-1-butene)
8.5 (10.0) (10.1) (10.5) (10.0)
a b b c c
(10.2) (12.6) (12.5) (13.5) (12.5)
b b b c c
10.0 11.5 11.6 12.0 11.5
b b b c c
11.7 14.1 14.0 15.0 14.0
b b b c c
Diolefins (dienes) 1,3-Butadiene
(8.9)
b
(11.6)
b
10.4
b
13.1
b
Aromatics Benzene Ethylbenzene Diethylbenzene Divinylbenzene Toluene Vinyltoluene Styrene (phenylethene)
11.4 9.0 8.5 8.5 9.5 9.0 9.0
d d, k d, ## d, ## e, † d, # d, ††
b ··· ··· ··· ··· ··· ···
11.4 9.0 8.5 8.5 9.5 9.0 9.0
d d, k d, ## d, ## e, † d, # d, ††
Alcohols Methyl alcohol (methanol) Ethyl alcohol (ethanol) Ethyl alcohol (ethanol) n-Propyl alcohol (n-propanol) Isopropyl alcohol (2-propanol) t-Butyl alcohol (t-butanol) Isobutyl alcohol (2-methyl-1-propanol) Isohexyl alcohol (2-ethyl-1-butanol)
(12.4) ··· ··· ··· ··· ··· ···
13.9 ··· ··· ··· ··· ··· ···
b ··· ··· ··· ··· ··· ···
(8.5) (9.0) 8.7 8.6 9.5 ··· 9.1 (7.9)
c c e, † e, † f, † ··· e, † c, §
c c ··· ··· ··· c, § ··· ···
10.0 10.5 8.7 8.6 9.5 ··· 9.1 9.3
c c e, † e, † f, † ··· e, † c, §
Esters Methyl formate Methyl acetate n-Propyl acetate Isopropyl acetate n-Butyl acetate Isobutyl acetate Isobutyl formate
(10.5) (11.5) ··· ··· ··· (15.0) ··· ···
12.0 13.0 ··· ··· ··· 16.5 ··· ···
c c ··· ··· ··· c, § ··· ···
(8.5) (9.5) 10.1 8.8 9.0 9.1 (11.0)
c c f, † e, † e, † e, † c
(11.0) (12.0) ··· ··· ··· ··· (13.5)
c c ··· ··· ··· ··· c
10.0 11.0 10.1 8.8 9.0 9.1 12.5
c c f, † e, † e, † e, † c
12.5 13.5 ··· ··· ··· ··· 15.0
c c ··· ··· ··· ··· c
Ethers Methyl ether Ethyl ether Propylene oxide
(9.0) (9.0) (6.6)
c c g
c c ···
10.5 10.5 7.8
c c g
13.0 13.0 ···
c c
Ketones Acetone Methyl ethyl ketone
(11.5) (11.5) ···
(10.0) (9.5)
c c
(12.5) (12.0)
c c
11.5 11.0
c c
14.0 13.5
c c
(12.5) (10.5) (17.5) (15.5) (11.5) (10.0) (12.5)
c b, ‡ b, ∗∗ c, ‡ c b, ‡ c
··· ··· ··· ··· ··· (15.0) ···
··· ··· ··· ··· ··· b, ‡ ···
14.0 12.0 19.0 17.0 13.0 11.5 14.0
c b, ‡ b, ∗∗ c, ‡
··· ··· ··· ··· ··· 16.5 ···
··· ··· ··· ··· ··· b,‡ ···
Organo-chlorides n-Butyl chloride Methylene chloride Ethylene dichloride 1,1,1-Trichloroethane
8
b, ‡ c
(Continues)
Table C.1(a): Continued Updated or (Adjusted) Data
Original Data
N2 -Air Mixture
CO2 -Air Mixture
N2 -Air Mixture
CO2 -Air Mixture
LOC
Note
LOC
Note
LOC
Note
LOC
Note
(7.7) 13.4 15.0
c, ‡ d, † d
··· ··· ···
9.0 13.4 15.0
c, ‡ d, † d
Inorganic Compounds Carbon disulfide Carbon monoxide (in air) Hydrogen (in air) Hydrogen sulfide (in air)
··· ··· ···
··· ··· ···
··· ··· ···
(4.3) 5.1 4.6 (6.4)
c a a c
(6.4) (5.1) (4.6) (10.0)
c c c c
5.0 5.5 5.0 7.5
c c c c
7.5 5.5 5.2 11.5
c c c c
Miscellaneous nitrogen-containing compounds UDMH (1,1-dimethyl hydrazine)
(6.0)
c
···
···
7.0
c
Commercial Fuels Motor Gasolines (70/100) (100/130) (115/145)
···
···
(10.5) (10.5) (10.5)
c c c
(13.5) (13.5) (13.0)
c c c
12.0 12.0 12.0
c c c
15.0 15.0 14.5
c c c
Aviation Fuels Kerosene JP-1 fuel JP-3 fuel JP-4 fuel
(8.5) (9.0) (10.5) (10.0)
c, § c, § c c
(11.5) (12.5) (13.0) (13.0)
c, § c, § c c
10.0 10.5 12.0 11.5
c, § c, § c c
13.0 14.0 14.5 14.5
c, § c, § c c
Natural gas (Pittsburgh natural gas)
(10.5)
b
(12.9)
b
12.0
b
14.4
b
Gas/Vapor Trichloroethylene Vinyl chloride Vinylidene chloride
Notes: (a) 120-L apparatus – Zlochower and Green (2009). (b) Flammability Tube – Bureau of Mines Bulletin 503, Table 44, J. F. Coward and G. W. Jones (1952). (c) Flammability Tube – Bureau of Mines Bulletin 680, Table 11, J. M. Kuchta, A. L. Furno, A. Bartkowiak, and G. H. Martindill (1968). (d) ~5-L Vessel, ASTM E681 – The Dow Chemical Company (Unpublished) (e) ~5-L Vessel, ASTM E2079 – L.G. Britton “Using Heats of Oxidation to Evaluate Flammability Hazards”, Process Safety Progress, (2002). (f ) ~5-L Vessel, ASTM E2079 – L.G. Britton, The Dow Chemical Company, 1999 (Unpublished Report) (g) R.M. Jones, “Reducing the Inflammability of Fumigants with Carbon Dioxide,” Ind. Eng. Chem., Vol. 25, 1933, pp 394-396. ∗ ∗∗ † k †† ‡ # ## §
All experiments performed at 25 ◦C unless otherwise indicated Experiments performed at 30 ◦C Experiments performed at 60 ◦C Experiments performed at 70 ◦C Experiments performed at 73 ◦C Experiments performed at 100 ◦C Experiments performed at 105 ◦C Experiments performed at 114 ◦C Experiments performed at 150 ◦C
9
Substantiation: This TIA will result in changes to Subparagraph 7.2.3.1.2, addition of a new Subparagraph 7.2.3.1.3 and changes to Table C.1(a) of NFPA 69-2014. Subparagraph A.3.3.25 will also be revised to clarify statements related to LOC and to align with the discussion in this document. SUMMARY: Two Temporary Interim Amendments (TIAs) are proposed. This first TIA resolves issues with the table of limiting oxygen concentration (LOC) values for gases (Table C.1(a)), which had been modified and overcorrected in the 2008–2014 editions. BACKGROUND: In the current (2014) and many previous editions of the NFPA 69, “Standard on Explosion Prevention Systems,” reported LOC values for numerous fuels have been tabulated in the Annex material. Most of the tabulated LOCs were obtained using a 5-cm diameter, vertical flammability tube apparatus; since this test method is no longer in use, both the apparatus and the corresponding data are referred to as “classical.” During the 2008 revision cycle, the NFPA Technical Committee for Explosion Protection Systems was made aware of unpublished LOC data measured by Zlochower and Green at NIOSH. The new ASTM E2079 method was followed using a large (120-L) test vessel capable of generating reference-quality data. Of the five gases tested, ethylene showed by far the largest deviation from classical data with a measured LOC of only 8.5 mol% oxygen, or 15% less than the classical value of 10.0 mol%. New LOC data for hydrogen, carbon monoxide, methane and propane were on average 7.5±0.5% smaller than classical values. Based on the Technical Committee’s concern that tabulated classical LOC values were nonconservative, NFPA 69 was revised to state “LOC values obtained in flammability tubes shall be used after adjustment by subtracting two percent by volume oxidant. . . ” This was considered prudent, at least until more information became available to the Committee. In fact, this issue was initially raised for ethylene by Britton [1] in 1996. In 2009 Zlochower and Green [2] subsequently published the 120-L results for five fuels, concluding “. . . the traditional flammability tube values for the LOCs are too high—at least for N2 inerting. The differences from the 120-L results are typically less than 1% (e.g. 12 vs. 11.1% for CH4, 11.5 vs. 10.7 for C3H8), but the difference is 1.5% for ethylene (10 vs. 8.5%).” This issue is also recognized in ASTM E2079, “Standard Test Methods for Limiting Oxygen (Oxidant) Concentration in Gases and Vapors” [3], which warns the 5-cm vertical flammability tube “diameter may be too small to mitigate the flame quenching influence impeding accurate determination of the LOC of most fuels” and, “as a result, some LOC values determined using [newer methods] are approximately 1.5 mol% lower than the previous values measured in the flammability tube.” There’s no direct evidence that the 5-cm vertical tube causes more flame quenching at the LOC than it does at the lower and upper flammable limits, the likelihood of which was discounted by the U.S. Bureau of Mines following extensive testing in larger diameter tubes; see for example Lewis and von Elbe [4].
We propose that much of the LOC discrepancy can be attributed to a difference in LOC definition, since the classical 5-cm vertical tube test method required a flame to propagate the full length of a tube at least 1.5 m tall (which in a closed tube would typically produce a pressure rise in excess of 100%) while the ASTM E2079 closed vessel test method requires only a 7% pressure rise. Indeed, in cases where Zlochower and Green observed a pressure rise of about 30% the corresponding LOC values (apart from ethylene) were much the same as the classical ones. Moreover it’s proposed that the classical LOC data were subject to a number of experimental errors that were deliberately avoided by Zlochower and Green. In all cases where experimental errors resulted in overestimated classical LOC values, the experimental error and the “definition” error become additive. This is because the new definition always results in smaller LOC values. PROBLEM STATEMENT There are two key issues for the current 2% absolute adjustment from classical LOC values. 1. The 2% percent adjustment is overly conservative, which means inerting systems will require greater quantities of purge gas than necessary, resulting in higher costs and higher release to the environment. 2. An absolute adjustment factor is impracticable for gases with small LOCs as illustrated by the example of silane, which would be assigned a negative “adjusted LOC” value. Emergency Nature: Justification for the emergency nature of the proposed TIAs is based on Section 5.3, Item (f) of the “Regulations Governing the Development of NFPA Standards:” (f) The proposed TIA intends to correct a circumstance in which the revised document has resulted in an adverse impact on a product or method that was inadvertently overlooked in the total revision process, or was without adequate technical (safety) justification for the action. The Technical Committee made over-corrections to classical LOC data. Although this was the prudent approach based on available information at the time the standard was published, a more reasonable adjustment is warranted which diminishes the adverse impact on the method of deflagration prevention by oxidant concentration reduction. We therefore propose two Tentative Interim Amendments (TIAs). This TIA changes the LOC adjustment process and the associated table of LOC values; this effectively increases all of the classical LOC values that were previously overcorrected. The detailed TIA is presented here. For this TIA: First, instead of decreasing LOCs measured in the 5-cm vertical flammability tube by 2 mol% oxygen, multiply classical LOC values of 10 mol% or less by the factor 0.85. Since the 0.85 adjustment factor would otherwise lead to smaller adjusted LOC values for CO2 inerting than for N2 inerting, the adjustment is changed to a subtraction of 1.5 mol% oxygen where the classical LOC value is above 10 mol%. Second, reorganize the tabulated data according to chemical type rather than an alphabetical listing using different conventional names. Third, accept the reference quality LOC data of Zlochower and Green [1] with no adjustment. Fourth, where existing LOC values were measured using ASTM E2079 or a method closely approximating to it, such as ASTM E681, retain the existing LOC values with no adjustment.
The 0.85 adjustment factor was initially proposed to account for the apparent 15% decrease in the LOC of ethylene using ASTM E 2079 versus the classical method. After concluding that the classical ethylene data were probably flawed, it was decided to retain the basic 0.85 adjustment factor in view of the small number of gases that have so far been retested and the fact that the 0.85 factor represents only twice the average LOC decrease observed for the other four gases. REFERENCES [1] I. A. Zlochower and G. M. Green, The Limiting Oxygen Concentration and Flammability Limits of Gases And Gas Mixtures, J. Loss Prev. Process Ind. 22 (2009), no. 4, 499–505. Anyone may submit a comment by the closing date indicated above. To submit a comment, please identify the number of the TIA and forward to the Secretary, Standards Council, 1 Batterymarch Park, Quincy, MA 02169-7471.
