Hazards from Salute/Flash/Star Compositions A brief literature survey. By donald j haarmann aka The WiZ Scanned in from: The PGII Bulletin No. 65. May 1989 www.pgi.org Parts that gagged the scanner and a few others have been deleted. Donald James Haarmann -------- A Compilation Of Hazard and Test Data For Pyrotechnic Compositions. F. L. Mclntyre, Report ARLCD-TR-80047, October 1980, NTIS AD 096248. 390 pages. "This report is a compilation of parametric, stability, sensitivity and output data on selected pyrotechnic compositions derived from hazards evaluation and classification testing. This report provides a readily accessible source of data for some 180 pyrotechnic compositions." "An accident survey was conducted to identify primary hazards and cause/effect relationships associated with pyrotechnic operations during development, manufacturing, transportation and thermally ultimate use. "There were 18% [103] explosions and 5% [27] accidents that transition from either a fire to an explosion or multiple explosions. As expected, the majority of the incidents were fires. The Significant factor here is that 23% of the incidents RESULTED IN SOME FORM OF AN EXPLOSION, since pyrotechnic compositions are not normally considered to be explosive in nature." [Emphasis added.] Of interest were the TNT equivalence (Hi Explosive equivalence) tests. Of the six compositions used for producing sound two were tested for TNT Equivalence with the following results: Air Blast Simulator Mixture, as used in the M74A1 and M74 Simulator. [Aluminum flake 9%, Black Powder 91%] TNT Equivalence was found to be 45%. Detonation Simulator Mixture, use: the infamous M80. [************] TNT equivalence 80%. It should be noted that; "The M80 fire cracker mixture is no longer manufactured but is reported along with the test data BECAUSE OF SEVERAL CATASTROPHIC ACCIDENTS THAT HAVE OCCURRED." [Emphasis added.] "Critical Height" and "Critical Diameter" were also measured. "In the critical height test the "Critical height to explosion data are reported as the greatest material height in a given container diameter which did not result in transition from burning to an" explosion. Critical diameter tests the sample material using C4 as an explosive donor, to determine the minimum diameter required to induce a explosive reaction. The critical Diameter for the M80 composition was found to be 0.01 meters [4 inches!!!]. and the Critical Height was measured as being 3.96cm. [1.5 inches!!!] [Fools rush in where angles fear to tread.] Of the photoflash mixtures tested, TNT Equivalence of: 30-36-50% were measured. And even closer to homer a Yellow Star Mixture [Magnesium 18%, Barium Nitrate 17%, Strontium Nitrate 16%, Potassium Perchlorate 17%, Sodium Oxalate 17%, and HCB 12%] when tested: "indicated that this mix would detonate and an explosive equivalence (as compared to TNT) was greater then 50% in a contained vessel ***" This mixture was also found to be sensitive to friction and impact. ------- Propagation Rates In Thermally Ignited Pyrotechnic Compositions. Richard W. Collett, Tech Report ARLCD-TR-77049, August 1978, NTIS ADA060809. "Work was performed to determine the propagation rates in loose, granular confined pyrotechnic compositions when initiated thermally. Representative materials included flash, igniter and flare compositions." All compositions were tested confined in steel pipe 48" long by 2"id. both ends of which were sealed with heavy end caps. An igniter pack placed in the bottom of the column was used for thermal ignition. Conclusions: "All of the compositions tested developed fast reactions which could cause explosions and be extremely hazardous ***. The reactions are therefore all classed as detonative." [Emphasis added.] Of the four basic compositions tested, PFP-555 [Aluminum 15u 40.0%, Barium nitrate 140u 30.0%, and Potassium perchlorate 20u 30.0%] "can develop either a low-velocity or high-velocity detonation when thermally ignited. Test 1- 920 meters/sec. Test 2- 546 meters/sec." Explosive Power of Pyrotechnic Compositions. 1.M. Jenkins, Et. All, 19th Explosives Safety Seminar, Calif. 1980 Page 77 &ff. "Various pyrotechnic compositions were assessed in three experiments: 1-To measure and assess the explosive power from various initiating stimuli. 2-To measure the explosive power expressed in terms of the equivalent mass of TNT per unit mass. 3-The likelihood and effect of sympathetic initiation in a practical storage situation." Three initiating stimuli were used: 1/fuzehead 2/electric detonator, and 3/a detonator boosted with a tetryl pellet. The composition being placed in a paper mache pot, with the initiator being placed at the geometric center of the charge mass. Composition #11: Photoflash [40/60 Aluminum/Potassium Perchlorate] when ignited by source number three, resulted in an "equivalent mass approximation kg. TNT per unit mass" of 0.42. More rigorous testing using piezo-electric pressure transducers to measure air blast and other experiments using foil gauges raised the TNT equivalence to 50%. --------- TNT Equivalencies of Black Powder. Volume 1: Management Summary and Technical Discussion, H.S. Napadensk and J.J. Swatosh Jr., lTIRJ6265-3, Sept. 1972, NTIS ADA-044444. 69 pages + vii. "Black powder charges ranging in weight from 8 to 150 pounds were evaluated under different levels of confinement. The TNT equivalence for the final product were found to range between zero to 43% for impulse and zero to 24% for pressure, depending upon the level of confinement, the weight of explosive and booster, and the distance form the explosion." The generally quoted figure for the detonation velocity of BP is 400 meters/sec. However A.F. Belyaev and RKh. Kurbangalina; Russ. J. Phy- s.Chem. 38:309-310,1964, as quoted in the LLNL Explosives Handbook, URCL-52997, provide the following figures Density g/cm3 appx. 0.7, det velocity appx. 1.3 km/cm3, 1.35 km/sec. -------- Hazards Testing of Ammonium Perchlorate. F.L. McIntyre, et al, 58 pages. NTIS ADA-114966 A series of hazard classification tests were conducted on ammonium perchlorate, nominal 200 micron size, packed in 30 gallon, 20 Ga. steel drums with bolted ring closures, each container containing approximately 250 lbs. of material. Tests using a S94 squib and 2 oz. of FFF black powder resulted in NO explosion, NO over pressure detected, and NO rupture, splitting, or fragmenting of the drums. A second series of tests using a number 8 blasting cap produced the same results. Thermally decomposition, with NO evidence of an explosion. It would be well to remember however that Ammonium perchlorate in particle size below 15 micron is considered to be an explosive material under 18 U.S.C. Chapter 40. And that it can be sensitized with reducing agents. -------- PATR 2700 provides the following detonation velocities for Ammonium perchlorate: (Original reference: RH. Richardson, Hazards Evaluation of the Cast Double-Based Manufacturing Process, ABL/X-47 (1960) AD 250858. [Not seen by me.] Dry 400 m/s Wet-ethyl alcohol Wet-acetone 4200 m/s 4500 m/sec. ------- Studies on Fireworks compositions. 1: Combustion or explosion of crackers and bursting compositions. Noboru Ishikawa and Masao Kusakabe. "Kogyo Kayaku" 1976,37(6)310-15. [In Japanese] As almost all of this is in Japanese, only the English summery is available for inspection: "Crackers always detonated in spite of their small quantities or weak initiation with igniters. "Reaction of bursting compositions was always initiated as combustion and accelerated to detonation in the case of sufficiently large amounts of the charge. "It was shown that the busting composition with potassium erchlorate was safer than those with potassium chlorate." The following compositions were tested: >Deleted< -------- Studies on Fireworks Compositions. (11) Combustion Characteristics of Piled Fireworks Compositions: Gerbs, Star Grains and Star Composition as Powder; Noboru Ishikawa and Masao Kusakabe. "Kogyo Kayaku" 1979, 40(4), 277-82 (In Japanese) This paper reports on work performed by the Japanese government some six years ago. This report may have served as the model for the ATF test as the Bureau of Mines has this journal translated on a regular bases, although the three articles on fireworks that have been published do not appear in the translated edition. Apparently the Bureau of Mines feels that information on blowing up fish is more important then preventing accidents in the fireworks industry! The following is from the English summery: Fireworks compositions "were piled on the ground or on a concrete-floor in 5kg, 30kg, 50kg or 100kg and they were ignited with two squibs combined with powder pasted paper, Yakushi, or for some samples with two detonators. From 37 tests the reaction modes were classified into three: combustion, deflagration and detonation. Most gerbs showed combustion or deflagration except when a composition contained fine aluminium, the particle size of which was less then 300 mesh. The reaction of the composition with fine aluminium was promoted to detonation. [Emphasis added. WiZ] The star grains of 1ookg shifted to detonation from several ten millisecond combustion and in other cases they showed combustion or deflagration. The star composition powders showed combustion even with a quantity of 100kg." The above information was taken from those parts of the paper that were in English, i.e. the tables. Just what "star grains" are, is not reported in English. -------- Measurement of Pressure and Related Energy Output from Thermally Ignited Pyrotechnic Compositions Burning in a Partally Vented Vessel P.L. Farnell. 1981. NTIS ADA-100728. "The results of the tests described in the report indicate that pyrotechnic compositions are indeed hazardous and that new criteria are required to judge their hazardous nature, rather then attempting to apply nonapplicable ones used for explosives. For example, explosives reach a high pressure very quickly resulting in a large blast wave, but the extremely short duration yields a relatively small impulse imparted to contingent walls of a room. The blast wave also tends to be more directional. This is more likely to punch a hole in a wall or break it into small pieces, as form a hammer blow. In addition, a large pressure from the blast wave is relayed outside the room, if one wall is left open. Pyrotechnics, on the other hand, produce lower pressure but last longer, giving a large impulse to the whole wall which can push the wall down. Little pressure is relayed outside since the buildup is slow and there is little or no blast wave. Thus; adjacent buildings would be less endangered from the blast wave, at a closer distance, then from an explosive. However the extreme heat developed by some burning pyrotechnics can be of greater danger than the pressure; ** With proper venting, pressure from the combustion of pyrotechnics could be held to small values; but the heat and flame generated could harm people in the area, could set fires, or could even ignite other compositions located nearby. Finally, it is possible, with sufficient confinement producing a large pressure buildup, to cause the burning of some of the pyrotechnics to become a low velocity detonation at which point explosives criteria would apply. One should bear in mind, however extremely hazardous nature of pyrotechnics deflagrations, and the need to develop appropriate criteria for describing their outputs." --------- Deficiencies in the Testing and Classification of Dangerous Materials. J.E. Settles. 1968. Annals New York Academy of Sciences, Volume 152, Art.1. Pages 199-205. "A total of 103 persons suffered injuries in the 81 accidents. Seventy-eight fatalities resulted from these 81 accidents. "Of the 81 accidents included in this analysis, it was concluded that 23 of them involved only fire, and the principal hazard was radiant heat. It was further concluded that 44 of the accidents in- volved both fire and explosion. From information available, it seemed justified to assume that no more then 14 of the accidents were characterized by supersonic shock waves that would fall within the accepted definition of "detonating" reactions. "The 14 accidents in which detonating forces were present resulted in injuries to 35 persons and 34 fatalities. It appears from the information available that only one of these 34 deaths resulted from the blast overpressures that are associated with a detonating reaction. However, this one fatality was not the result of blast damage to human tissue. Rather, the blast pressure caused this individual to be propelled as a projectile. The other 33 persons who died in these 14 accidents were located at points where the density of flying fragments, and in some cases, the lethal searing of radiant heat were so great that their deaths were certain, even though there had been no blast effects. "A SERIOUS AND DISTURBING INCONSISTENCY IS RELATED TO THE PRACTICE OF ACCEPTING A "FIRE HAZARD ONLY" LABEL ON REACTIONS OF SUCH VIOLENCE AND DESTRUCTIVE ENERGY AS MEDIUM-VELOCITY DETONATION, LOW-VELOCITY DETONATIONS, HIGH-RATE EXPLOSIONS, MEDIUM-RATE EXPLOSIONS, LOW-RATE EXPLOSIONS, AND EVEN REACTIONS THAT DON'T EXPLODE AT ALL BUT KILL PEOPLE BY BURNING THEM TO DEATH."