n 'States C'tin-fc- o' ; ::!:,.-;.. :^ >Jieven:ior E;pA ."44-R-94-001
onrnentai '"'roiftCT.on nnr; Toxic 'Sub^'.Tnce rfDr,: ~. ",' ';'.-
Agency (7406)
«>EPA Commercial Explosives
And Their Hazards
_ Hecvc!ed'Rec-/clable
' ", ' f'nnt&d on parje'Thai comai
i" ;/"" ;it Ittasi 5C% rojvcierl tibur
-------
REPORT DOCUMENTATION
PAGE
If REPORT NO.
EPA 744-R-94-001
0, P-*clpienf s Accession No.
4. Title and Subtitle
Commercial Explosives and Their Hazards
5. Report Date
Feb 1994
7. Authors)
8. Performing Organization Rept. No.
9. Performing Organization Name and Address
ICF Inc.
9300 Lee Hwy
Fairfax, VA 22031
10. Project/Task/Work Unit No.
Work Assignment #26
II. Contract(C) or Grant(G) No.
(O 68-C2-0107
(G)
12. Sponsoring Organization Name and Address
U. S. Environmental Protection Agency
Office of Pollution Prevention and Toxics (7406)
401 M. St. S.W.
Washington, D.C. 20460
13. Type of Report & Period Covered
Final
14.
IS. Supplementary Notes
16. Abstract (Limit: 200 words)
This report assesses the potential consequences of accidents involving commercial
explosives. The analysis includes the identification and evaluation of existing
listing and classification systems, along with any applicable criteria; review of
existing regulations and codes dealing with commercial explosive materials; analysis
of histories of accidents involving commercial explosives; and modeling potential
consequences of overpressures and heat generation of commercial explosives. The
results of this report .may be used to support regulatory and non-regulatory .objectives
of EPA under the Superfund Amendments and Reauthorization Act:' (SARA)' title III 1986
and theClean Air Act Amendments (CAAA) 1990.
17. Document Analysis a. Descriptors
Explosive
Hazard
b. Identifiers/Open-Ended Terms
c. COSATI Field/Group
18. Availability Statement
Release unlimited
19. Security Class (This Report)
2O. Security Class (This Page)
21. No. of Pages
81
22. Price
[See ANSI-Z39.18)
See Instruction! on Reverse
OPTIONAL FORM 272 (4-77)
(Formerly NTIS-35)
Department of Commerce
-------
COMMERCIAL EXPLOSIVES AND THEIR HAZARDS
Office of Pollution Prevention and Toxics
U.S. Environmental Protection Agency
Washington, D.C. 20460
February 1994
-------
ACKNOWLEDGEMENTS
The preparation of this report has been made possible through the efforts of the Office of Pollution
Prevention and Toxic Substances "Physical and Chemical Property Hazards" Workgroup over the many years
of its existence. The many contributors addressed the challenging task of reviewing, analyzing and summarizing
the myriad existing models and empirical data on commercial explosives and their hazards. A major
consideration in this work was the need to express the hazards of commercial explosives as correctly and simply
as possible for use by risk managers. Finally, the Environmental Protection Agency Science Advisory Board
provided comments that were helpful in preparation of the final draft of this document.
Environmental Protection Agency Report Workgroup
and Contributors
*Roger Garrett, Workgroup Co-Chair
*Paul Tobin, Workgroup Co-Chair
*Russell Farris, Workgroup Co-Chair
*Paul Anastas, Workgroup Co-Chair
*Robert Andrei
Paul Bickart
Kathy Bishop
*Gail Froiman
*Robert Lenahan
*Craig Matthiessen
*Fred Metz
Jerry Newsome
*Mark Pederson
*Ward Penberthy
*Vanessa Rodriguez
Research Contributors
Pamela Bridgen, 1CF Incorporated
Maravene Edelstein, 1CF Incorporated
David Goldbloom-Helzner, ICF Incorporated
Deborah Shaver, ICF Incorporated
-------
1
TABLE OF CONTENTS
Page
1.0 Introduction 1
1.1 SARA Section 302 1
1.2 Regulatory History 1
1.3 Approach J. 2
2.0 Characteristics of Explosive Materials 3
3.0 Manufacture, Handling, and Use of Commercial Explosives 4
3.1 Manufacture, Handling, and Use of High Explosives 4
3.2 Manufacture, Handling, and Use of Low Explosives 9
3.3 Manufacture, Handling, and Use of Blasting Agents 10
3.4 Storage of Explosives 11
4.0 Past Incidents Involving Commercial Explosives 12
4.1 Explosives Accidents Reported in EPA's AHE Data Base, and
Other Hazardous Materials Databases 12
4.2 Transportation Accidents 13
4.3 Blasting Accidents in Mines 18
4.4 Black Powder Accidents Reported by Department of Defense 21
5.0 Existing Regulations, Standards, and Classification Systems 21
5.1 DOT Standards 21
5.2 ATF Standards 24
5.3 NFPA Codes 26
5.4 State and Local Regulations : 26
5.5 Clean Air Act Regulations 26
5.5.1 OSHA Process Safety Management Standards 26
5.5.2 EPA Chemical Accident Prevention Regulations 27
5.6 Other Standards 27
6.0 Methodology for Determining Affected Distances
for Commercial Explosives , 27
6.1 Approach 27
6.1.1 Damage Criteria v 27
6.1.2 Assumptions 28
6.1.3 Quantities Modeled 29
-------
TABLE OF CONTENTS
(Continued)
Page
6.2 Methodology for Determining Explosive Distances 29
6.2.1 Scaling Law for Explosives 29
6.2.2 ATF Table of Distances for High Explosives 30
6.2.3 ATF Table of Distances for Low Explosives 32
6.3 Results 35
6.3.1 High Explosives and Blasting Agents 35
6.3.2 Low Explosives 35
6.4 Findings 35
References 39
APPENDICES
Appendix A: Damage Criteria/Overpressure Data 43
A.1 Damage Criteria 43
A.2 Effects of Varying Overpressures 51
References 53
Appendix B: Other Methods of Consequence Analysis Based on the Scaling Law 55
B.I ARCHE i 55
B.2 "K-Factor" Methodology 56
B.3 Results 58
Appendix C: Requirements For Explosives Magazines 65
C.1 General Requirements 65
C.2 Requirements by Type of Magazine 65
References 70
Appendix D: Tests For Explosives in Transportation 71
D.I Tests for Explosive Properties 71
D.2 Tests for Assignment of Hazard Division 72
References 75
-------
Ill
LIST OF EXHIBITS
1 Explosive Materials and their Applications 5
2 No. 8 Detonator 7
3 Incidents Involving Explosives 15
4 Incidents Involving Fireworks and Pyrotechnics 16
5 Transportation Accidents Involving Explosives 17
6 Fatal and Non-Fatal Injuries Due To Blasting Accidents in Coal Mining 19
7 Fatal and Non-Fatal Injuries Due To Blasting Accidents in Metal Mining 20
8 K-Factors as a Function of Peak Overpressures 31
9 American Table of Distances - High Explosives & Blasting Agents 33
10 Table of Distances - Low Explosives 34
11 Affected Distances for Commercial High Explosives, Determined
Using the ATF Method (Peak Overpressure = 0.5 psi - 3.0 psi) 36
12 Affected Distances for Low Explosives, Determined Using the
ATF Table of Distances for Low Explosives 37
A-l Human Injury Criteria 44
A-2 Property Damage Criteria 46
A-3 Lethality Curves 49
A-4 Lethality and Injury Curves 49
A-5 Air-Blast Criteria for Personnel Standing in the Open 50
A-6 Overpressure Versus Distance for 100 Pounds of High Explosives 52
B-l Sample Printout of ARCHIE Modeling Results for 1,000 Pounds of TNT 57
B-2 Affected Distances for Commercial High Explosives and Blasting Agents, Determined
Using Four Different Methods (Peak Overpressure = 0.5 psi) 59
-------
IV
LIST OF EXHIBITS
(Continued)
Page
B-3 Affected Distances for Commercial High Explosives and Blasting Agents, Determined
Using Four Different Methods (Peak Overpressure = 1.0 psi) 60
B-4 Affected Distances for Commercial High Explosives and Blasting Agents, Determined
Using Four Different Methods (Peak Overpressure = 2.0 psi) 61
B-5 Affected Distances for Commercial High Explosives and Blasting Agents, Determined
Using Four Different Methods (Peak Overpressure = 3.0 psi) 62
-------
COMMERCIAL EXPLOSIVES AND THEIR HAZARDS
1.0 Introduction
1.1 SARA Section 302
When the Superfund Amendments and Reauthorization Act (SARA) was enacted in 1986, a list of
extremely hazardous substances was published under section 302 of Title III. Section 302 requires a facility
that has an extremely hazardous substance (EHS) above its threshold planning quantity (TPQ) to notify the
State Emergency Response Commission (SERC) that it is subject to the requirements for emergency planning.
The original list of EHSs was created using only acute toxicity criteria. The purpose of the list was to
identify those substances that, if accidentally released, could cause death or serious irreversible health effects
off-site. However, toxicity is not the only hazard posed by chemicals. SARA states that the Administrator
may revise the list, and that such revisions "shall take into account the toxicity, reactivity, volatility,
dispersibility, combustibility, or flammability of a substance."
Commercial explosives are one of the many classes of substances that may be added to the Section 302
list of EHSs under the additional criteria. In order to regulate them, EPA must distinguish extremely
hazardous explosives from all other explosives. This report is intended to be used as technical support for the
rulemaking by describing various ways of classifying explosives and their hazards. The report describes
different types of explosives and how they are used, and provides analysis of past accidents involving explosives.
It also gives a summary of existing regulations, and analysis of the quantity-distance relationship methodology
on which those regulations are based.
n
EPA is aware that the Bureau of Alcohol, Tobacco, and Firearms (ATP) already has a rule on'
commercial explosives (27 CFR Pan 55 - Commerce in Explosives). However, the ATF regulation does not
require reporting of the presence of commercial explosives to local emergency planning authorities. Therefore,
EPA is considering a rule that would add explosives to the list of EHSs under section 302 of SARA Title III,
thereby requiring that the SERC be notified of the presence of explosives above a TPQ which EPA would set.
Local Emergency Planning Committees (LEPCs) would then receive information that would enable them to
carry out proper planning for the hazards associated with explosives at facilities in their communities. The
regulatory history of local emergency planning is presented below. .
1.2 Regulatory History
EPA originally developed the EHS list as pan of the voluntary Chemical Emergency Preparedness
Program (CEPP) which was designed to increase public awareness of chemical hazards in communities and
focus effort on emergency planning. EPA believed that communities needed a staning point and intended that
the list draw attention to the chemical substances and facilities that pose the most immediate concern from
an emergency planning and response perspective. EPA recognized and emphasized that there are tens of
thousands of compounds and mixtures in commerce that may pose a hazard under specific circumstances and
that this list only deals with lethality or serious irreversible health effects associated with acute toxicity. The
Agency chose lethality because it represents the most immediate concern in an emergency situation and is also
an approximate measure of the overall toxicity of chemical substances.
-------
-2-
In its analysis of toxic substances under SARA section 302, EPA used a 100-meter fenceline distance
to provide a guideline for community emergency planners to use in setting priorities for planning for hazards
in the community. The 100-meter fenceline distance has also been used by the Occupational Safety and Health
Administration (OSHA) in its analysis for thresholds for chemicals listed in its Process Safety Management
Standard and by the state of Delaware for thresholds under its prevention regulations. EPA recognizes that
this distance may not be appropriate for protection of emergency responders. There is no absolute distance
that would guarantee the safety of first responders in every situation.
Several commenters on the rulemaking for the EHS list noted that physical/chemical hazards such as
flammability or explosivity may also lead to serious consequences, including death and irreversible injury. In
an official inquiry to the Agency in 1987, Senator Frank Lautenburg also noted concern for hazards other than
toxicity and asked that EPA focus on other hazards. The Agency agreed and noted its intent to evaluate
hazards other than toxicity in the future.
A Physical/Chemical Criteria Workgroup was formed at EPA to address the concerns raised by Senator
Lautenburg and others and to focus on the problem of identifying materials that are hazardous based upon
their physical/chemical properties and could thereby be candidate EHSs. The Workgroup initially considered
the hazards referred to in the Act and determined that they fell into two general categories: hazards related
to toxic properties and hazards related to physical/chemical properties. The characteristics of reactivity,
volatility, dispersibility, combustibility, and flammability, mentioned in the Act, are related to a variety of
physical/chemical properties; the Workgroup chose to focus on the physical/chemical properties that lead to
the greatest potential hazards to people.
L3 Approach
Hazards associated with physical/chemical properties are primarily based on two phenomena that can
cause serious personal injury or death. These phenomena are the result of the release of energy from highly
reactive or flammable chemicals and are identified as overpressures from blast waves and thermal radiation
from fires. Overpressures result from nearly instantaneous energy release, or detonation, while thermal energy
is released during combustion, which occurs more slowly. Since explosive chemicals are highly reactive
substances that can detonate and create overpressures, and flammable chemicals can burn and produce thermal
radiation, the Workgroup focused on explosives and flammable chemicals as chemicals of concern.
The Workgroup elected to segregate the chemicals for review into separate categories for flammables,
explosives, and reactives for purposes of analysis and possible regulation even though it was recognized that
there may be overlap of the consequences; e.g., the detonation of a flammable material may yield the same
results as the detonation of a commercial explosive. However, the Workgroup wanted to determine which
parameters distinguish extremely hazardous explosives from all other explosives, extremely hazardous
flammable substances from all other flammables, and so on. This report focuses on commercial explosives.
Separate documents will address the consequences of accidents involving flammable materials, non-commercial
explosives, and other reactive chemicals.
The Physical/Chemical Criteria Workgroup investigated several possible approaches to characterizing
explosive materials, none of which provided any clear-cut means of placing certain commercial explosives into
extremely hazardous categories. The data analyzed are presented below. Section 2 describes physical and
chemical characteristics. Section 3 contains information on manufacture, handling, use, and storage of
explosives. Section 4 presents data on past accidents, Section 5 describes existing regulations, and Section 6
-------
-3-
presents methodology for determining affected distances and results of consequence analysis based on the
methodology.
2.0 Characteristics of Explosive Materials
An explosion is defined as a sudden and violent release of energy (Lees 1980) accompanied by high
pressure gases. The more rapid the release and the greater the available energy, the more violent the
explosion (U.S. DOT 1988). Most of the available energy from an explosion is convened into some form of
mechanical energy, manifesting itself in the form of blast waves, fragmentation, and ground shock. Of these
damage effects, blast overpressure contributes the greatest damage in chemical explosions. The remaining
ponion of the explosive energy remains in thermal form and is of concern because of its fire hazard. Typically,
the damage associated with the pressure wave is far greater such that the thermal wave can be ignored (Brasie
and Simpson 1968). Blast waves from accidental explosions can cause damage to people and property by
subjecting them to transient crushing pressures and winds. Relatively simple concepts have-been used
effectively to correlate blast wave properties with damage. These concepts and the results of consequence
analysis based on them are discussed in Section 6 and Appendices A and B.
Explosive materials typically are classified as either "high" or 'low' explosives or blasting agents. "High"
explosives are defined by ATF as those materials that detonate if ignited, shocked, or subjected to heat or
friction. Detonations result from exothermic chemical reactions that proceed at supersonic velocities,
generating large volumes of high pressure gas and heat instantaneously, even though no confining vessel or
structure exists. In an explosion caused by a high explosive, the rate of energy release is particularly rapid and
the shock wave has a very shon duration. Blasting agents also detonate, but are much less sensitive than high
explosives. In contrast, "low" explosives deflagrate, or burn at subsonic velocities. Some examples of high and
low explosives and blasting agents, and their manufacture, handling, and use, are discussed in Section 3 below.
A deflagration generates lower pressures and is less destructive than a detonation. A deflagration
reaction propagates through the unreacted material via a diffusion mechanism rather than a shock wave.
However, under the right conditions (e.g., confinement) some low explosives (or substances not normally
considered explosives) may detonate; conversely, some high explosives may burn if ignited. According to
several general references reviewed, factors that affect whether a deflagration or detonation may occur, or that
affect the transition from deflagration to detonation (or detonation to deflagration) include:
Type and intensity of initiation. Explosives that usually detonate may burn under carefully
controlled conditions involving gentle ignition to avoid shock-wave formation, and propellants
that normally burn quietly and controllably may detonate if initiated by a high-pressure, high-
intensity shock wave (Kirk-Othmer 1980).
Physical conditions, including:
Panicle size. Finely divided material is more likely to detonate (Kirk-Othmer 1980).
Thickness of layers of explosive. A large mound of material is more likely to detonate
than a thin layer (Kirk-Othmer 1980); there is a "critical thickness," which varies by
substance, below which a detonation cannot propagate (Medard 1989).
-------
Diameter of cylinder of explosive. There is a critical diameter which is the minimum
diameter of an explosive charge at which detonation can occur, it is texture dependent,
and may be very large for some materials (e.g., ammonium nitrate) (Meyer 1987).
Confinement. Materials under heavy confinement are more likely to detonate (Kirk-Othmer
1980). Note that some high explosives are designed so that they will only detonate under
confinement (Meyer 1987)..
Mass of explosive material. For some unconfined explosives, there is a critical mass below which
the material will be consumed by deflagration before detonation can occur, and above which a
detonation may be initiated by the heat of the deflagration (Medard 1989).
In general, conditions that minimize energy losses and promote the build-up of a shock wave increase
the likelihood of a deflagration becoming a detonation (Kirk-Othmer 1980). According to Medard (1989),
the transition from a deflagration to a detonation is not continuous, with the deflagration accelerating until
it reaches the velocity of a detonation, but is always a discontinuous leap.
3.0 Manufacture, Handling, Use, and Storage of Commercial Explosives
The manufacture, handling, use, and storage of high and low explosives as well as blasting agents are
briefly described in this section. Representative explosives of each type are used as examples. This section
is intended to provide an overview of the explosives industry and current industrial practices.
Explosives have been in use for more than ten centuries; the ancient Chinese used a crude form of
black powder in early missiles and firecrackers. Explosives were used during the Renaissance to increase
quarrying yields and black powder served as the chief propellent for early muskets and cannons. Current uses
of explosives include mining, quarrying, construction, fireworks, and military uses. According to the Bureau
of Mines (1991), total consumption of industrial explosives and blasting agents in the U.S. in 1990 was 4.7
billion pounds. The coal mining industry used 3.2 billion pounds, or 68 percent of the total (Bureau of Mines
1991). These explosives were produced by 24 companies. Blasting agents (discussed in Section 3.3 below)
account for by far the greatest volume of industrial explosives; most blasting agents are based on ammonium
nitrate. Exhibit 1 presents a breakdown of explosives by type and major applications.
3.1 Manufacture, Handling, and Use of High Explosives
The Bureau of Mines (1991) reports use of high explosives in coal mining, quarrying and nonmetal
mining, metal mining, construction work, and other (unspecified) uses. Sales of industrial high explosives
totalled 152 million pounds in 1990. Quarrying and nonmetal mining was reported to use the largest quantity
of high explosives, 61 million pounds, or 40 percent of the total, followed by construction, with 40 million
pounds, or 26 percent of the total (Bureau of Mines 1991).
High explosives are classified as primary or secondary based upon the explosive's sensitivity to energy
inputs. Primary explosives, such as lead azide or mercury fulminate, are sensitive to small energy inputs caused
by friction, shock, or static electricity (i.e., it takes a relatively small amount of energy to cause these explosives
to detonate). A small energy input generates a shock wave which moves through the unreacted explosive.
Secondary explosives have faster detonation rates and produce a larger energy output than do primary
explosives. However, they require a larger energy input in order to initiate detonation (Kirk-Othmer 1980):
-------
Exhibit 1. Explosive Materials and Their Application
explosive matter
1
1
explosives
1 1
high explosives propellants pyrotechnics
1
1
primary (initiating) secondary gun propellants
explosives explosives
, . . . single base
^Bld? double base
lead stvphnate ...
mercury fulminate ' i ' \
diazodinitrophenol i. .
tetrazene uf^ Ti
others black powder
mixtures »
flashes
flares
fume
generators
optical and
acoustic
signals
fireworks
1
military explosives industrial explosives rocket propellants
explosive compounds, e.g.: gelatins double base
TNT powders composites
RDX (Hexogen) permitted explosives liquid fuels
PFTN fNiirnnfinia) ANFO and oxidizers
1
industrial chemical products
for non-explosive purpose
fertilizer grade ammonium nitrate
chlorates as weed killers
gas generating ingredients for
foam plastics
organic peroxides as polymerization
catalysts
nitroglycerin and PETN-solutions
for pharmaceutical purposes
salts of nitrated organics acids for
pest control chemicals
others
Tetryl. and others
mixtures, e.g.
Composition B
Torpex
RDX-based plastics, and others
slurries: emulsion slumes
Source: Meyer, R.. Explosives, 3rd ed., Essen, Germany: WASAGCHEMIE, 1987.
-------
-6-
Because of their sensitivity, primary explosives are used as initiator explosives to detonate secondary
high explosives, which require a much larger energy stimulus. Primary explosives are used in detonators,
blasting caps and percussion primers. A detonator usually contains a primary explosive, typically lead azide
in quantities less than a gram, and several secondary explosives, as needed. The function of a detonator is to
magnify a tiny energy input in a chain reaction which leads to the quick release of energy stored in secondary
explosives (Kirk-Othmer 1980). A typical detonator is shown in Exhibit 2.
According to the United States Bureau of Mines (Cantrell 1991), the majority of high explosives
manufactured in this country are nitro explosives. Nitroglycerine, a nitro explosive in common use in the
explosives and medical industries, is discussed here as an example of a secondary high explosive. Information
on its manufacture, handling, and use is presented below. Early references to nitroglycerine date back some
170 years. Alfred Nobel used this compound in his original formulation for guar dynamite, a mixture of
nitroglycerine and kieselguhr, an inert base. Today, nitroglycerine is used as a prescription drug, as well as
an explosive; it is used as a vasodilator and to relieve angina pectoris.
Nitroglycerine is very stable, and it does not decompose at room temperature. At higher temperatures,
above 60°C, it begins to aulocatalyze into its decomposition products: carbon monoxide, carbon dioxide, nitric
oxide and water. Nitroglycerine burns rather than explodes if present in small quantities or if unconfined.
Microbubbles of air present in liquid nitroglycerine increase its sensitivity to shock, since the shifting of
bubbles increases the kinetic energy in the system. Nitroglycerine's sensitivity to shock is unusual in a
secondary explosive. It is only employed when desensitized by other compounds, often triacetin, dibutyl
phthalate, or nitrocellulose (Kirk-Othmer 1980).
The earliest methods for manufacturing explosives were batch methods, used to prepare a large quantity
of an explosive at one time. The dangers of handling large amounts of explosives led to the development of
continuous production methods and the eventual phase-out of batch processes. Continuous methods result
in production of quantities of explosives that allow for safe handling. Production is continuous with the
explosives being removed from the reaction immediately after synthesis. Two methods are used to manufacture
nitroglycerine: the Biazzi Continuous Process and the Nitro Nobel Injector Process. Variations of these two
methods may be used to synthesize other nitro explosives as well (Kirk-Othmer 1980).
In the Biazzi Continuous Process, oleum, nitric acid, and glycerol are metered into a nitrator. An
emulsion quickly forms and a portion of the emulsion passes over cooling coils at 10-20°C. Nitroglycerine
is separated from the emulsified working solution and spent acid is sent through a mechanical centrifuge and
additional nitroglycerine is separated. Both the nitroglycerine and acidic working solution are washed with
water until stability is assured, and a non-explosive emulsion of water and nitroglycerine is formed for
temporary storage (Kirk-Othmer 1980).
There are many safety features built into the Biazzi Process. The emulsions formed ensure that the
nitroglycerine is insensitive; a 3:1 water to nitroglycerine emulsion is stable and unlikely to explode.
Production reactors are of highly polished, stainless steel, and operations are fully automated. Fail-safe
features include water spray jets, signaling devices, and automatic shut-down. Control features include an
overflow pipe and the ability to drown a runaway reaction with water to bring it under control (Kirk-Othmer
1980).
-------
-7-
EXHIBIT2. NO. 8 DETONATOR
shell 5052 vLuatlnua alloy
7.06 sa o.d. 0.035 an thick
ignition chug*
priaarchMB*
0.195g d«trlai.t«d
b*aa
O.A47 * 0.019f
pressed At 28Mp*
Pests and Criteria.
-------
In the Nobel Process, warm glycerol is slowly injected into a cooled stream of nitric acid. The flow of
glycerol is regulated by a vacuum generated by the flow rate of tbe acid. The glycerol quickly reacts, an
emulsion forms, and the solution is immediately cooled. After cooling, the nitroglycerine is separated, washed,
and temporarily stored as an emulsion. Safety features for this method are the same as those built into the
Biazzi Process. In both processes detonation traps are used to prevent propagation of an accidental explosion
throughout the entire reactor.
Pharmaceutical grade nitroglycerine can be manufactured using either the Biazzi or Nobel process.
After synthesis, it is stabilized and mixed so that its concentration ranges from 1 to 20 percent pure
nitroglycerine, in combination with inert chemicals, forming a non-explosive mixture (Wilcox 1991).
High explosives have many uses ranging from mining and quarrying to construction, seismology and
petrogeology. Primary explosives are very sensitive, and they are generally not handled until placed in blasting
caps and detonators. Secondary explosives have a wide range of uses; they are used in blasting caps and to
sensitize blasting agents. Secondary explosives (e.g., nitroglycerine) may also be used in propellants (see
Section 3.2 below). Modem dynamites are made from high explosives, usually nitroglycerine and an active
base nitrocellulose, which contribute to the force of the explosion. Most dynamites are variations of straight
dynamite or gelatin dynamite. Straight dynamite is manufactured by absorbing an active base in nitroglycerine.
Gelatin dynamites are made from nitroglycerine and nitrocbtton. Ammonium nitrate (which is widely used
in blasting agents) is also often used in dynamites, in combination with nitroglycerine and various fuels.
Dynamite is classified by strength according to its equivalence to or percentage of nitroglycerine (Wallace
1991). Dynamite is supplied in various cartridge sizes as needed to perform particular demolition and mining
tasks. Cartridges are stored and shipped in fiberboard crates usually weighing fifty pounds. Large cartridges
can weigh in excess of 50 pounds and are safely shipped uncased.
Storage facilities for explosives must meet the requirements for safe distances as specified in the
American Table of Distances for Storage of Explosives established by the Institute of Makers of Explosives
and adopted by ATF (see Section 6.22). These storage sites are required to be clearly marked with placards
indicating the presence of explosive material.
Transportation of explosives is strictly regulated by DOT. Operators of vehicles transporting explosives
are required to meet the qualifications specified in 49 CFR 391, Subpart B. High explosives must be
transported in a magazine or a day box, which contains only enough explosive for one day, as regulated by 49
CFR 174. High explosives are currently classified by DOT as Division 1.1 or 1.2 under the international
classification system, or Class A under the old system (see Section 5.1) and transportation of these explosives
is regulated under 49 CFR 173.61-173.91. Explosives transported on land may not contain more than 60
percent liquid explosive ingredient; explosives transported by water may contain no more than 75 percent.
DOT lists general packing requirements for high explosives and specific regulations for certain high explosives
(e.g., nitroglycerine, nitroguanidine and lead azide). All packages and transporting vehicles must be clearly
identified as containing explosives and possess the appropriate markings (i.e., labels and placards).
High grade nitroglycerine, a liquid high explosive, is regulated by DOT as Division 1.1 under the
international classification system, or Class A under the old system (see Section 5.1). Pure (undesensitized)
nitroglycerine is forbidden for transportation. Pharmaceutical grade nitroglycerine, however, passes the DOT
required No. 8 Blasting Cap and Impact Fire Tests for propagation; it is not classified as an explosive.
-------
-9-
3.2 Manufacture, Handling, and Use of Low Explosives
Low explosives deflagrate rather than detonate, that is, they burn at a rate that is subsonic, or less than
the speed of sound. Propellants, used in guns and rockets, and pyrotechnic compositions, used, for example.
in fireworks, flares, and theatrical devices, are examples of materials that may be classified as low explosives.
Propellants are mixtures of chemicals, including fuels and oxidizers, that function by burning to produce
large volumes of gas at controlled rates. Propellants for guns are usually based on nitrocellulose; liquid
explosives such as nitroglycerine and crystalline explosives such as nitroguanidine may also be included in the
composition. Rocket propellants may be based on nitrocellulose and a liquid explosive or may be polymer-
based (Kirk-Othmer 1980).
Pyrotechnic compositions are made up of a fuel and an oxidizer, reactions generally take place
independently of any external oxidizer. The oxidation-reduction reactions involving pyrotechnic compositions
produce little or no gas, in contrast to propellants. Magnesium and aluminum powders are the most
commonly used fuels; alkali metal salts may be used as oxidizers. Fireworks, which are made from pyrotechnic
compositions, are produced at approximately 100 plants nationwide that employ no more than 1,000 workers
(Conkling 1986). Common or consumer fireworks (e.g., firecrackers) are regulated during their manufacture
by ATF; however, they are not regulated as explosives by ATF in finished form. They are regulated in
transponation by DOT as explosives in Division 1.4. Special or display fireworks are regulated by ATF both
in manufacturing and in finished form. Pyrotechnic compositions also may be used as prime igniters for more
powerful high explosives, as in a safety fuse (Kirk-Othmer 1980).
Black powder, which may serve as a propellant and fuse powder in fireworks, is classified by the ATF
as a low explosive; however, it is listed as a high explosive under DOT regulations. Black powder is a mixture
with a composition that has changed little in over 1,000 years of use. According to the Bureau of Mines
(1991) report on consumption of industrial explosives and blasting agents, sales of black powder are "minor";
black powder sales have not been reported to the Bureau of Mines since 1971.
Manufacturing procedures for low explosives vary widely. Propellants of various types include a variety
of compositions; manufacturing methods include both batch and continuous processes, and may be very
different, depending on the propellant and its use.
Black powder, discussed above, is manufactured by only one company in the United States, Goex, Inc.
of Moosic, Pennsylvania. Black powder is essentially a mixture; therefore, its properties rely on the blending
of the ingredients. Either potassium nitrate or sodium nitrate is ground with charcoal and sulfur. The
resulting mixture is dried, pressed and packaged. The process is highly automated. Water deluge systems are
in place as a safety measure. If ultraviolet light sensors detect a flash, deluge systems are activated.
ATF regulations limit the quantities of explosive materials allowed in the process area of a special
fireworks manufacturing plant to quantities reasonably necessary for a day's manufacturing or assembly
operations. No more than ten pounds of flash powder and 500 pounds of other explosive materials may be
kept in any process building or area. All explosive powders and mixtures, unfinished special fireworks, and
finished fireworks must be stored in approved magazines at the end of the day (magazines are any buildings
or structures used for storage of explosives) (Conkling 1986).
-------
-10-
Burn tests on common fireworks conducted in 1983, in which 15,000 pounds of fireworks in steel
shipping containers were burned in a wood and kerosene fire, indicated that although the fireworks burned,
there was no detonation or pressure build-up strong enough to cause fragmentation of the container. Bum
tests in 1985 on special fireworks, however, showed that certain types would detonate under confinement or
in bulk quantities (quantities that might be found in manufacturing operations). The materials that detonated
were flash powder, which is usually a mixture of an oxidize: (potassium perchlorate) and two fuels (sulfur and
aluminum powder); finished "salutes", which consist of a cardboard casing containing about 57 grams of flash
powder, with a 28-gram propelling charge of black powder at the bottom; and "stars", which usually consist
of a mixture of an oxidizer (typically potassium perchlorate, potassium chlorate, or potassium nitrate), one or
more fuels (e.g., charcoal, aluminum, or a carbon-containing gum or resin), and a color-producing agent. Tests
on bulk quantities of finished "color shells" showed rapid burning, but no detonation or mass explosion
(Conkling 1986).
Low explosives are categorized and regulated by DOT as Division 13 under the international
classification system, or Class B under the old system (see Section 5.1 below). Individual pyrotechnic
compositions (e.g., some flash powder compositions) may be classed as high explosives (1.1) depending on test
results. Shipping containers and methods are listed in 49 CFR 173.88-114.
3.3 Manufacture, Handling, and Use of Blasting Agents
Blasting agents are mixtures of chemicals which, when used for blasting, do not detonate when initiated
by a.No. 8 Blasting Cap, according to 49 CFR 173.114a; blasting agents must also pass five other tests as
required by DOT. All blasting agents must be approved as such by the U.S. Bureau of Mines Associate
Director for Hazardous Materials Regulation. As noted earlier, blasting agents are in use in much larger
quantities than other types of explosives. Ammonium nitrate and fuel oil mixture (ANFO) is by far the most
common blasting agent. Roughly 85 percent of all blasting agents are variations of this mixture. The ease
of use and high stability of ANFO has led to its replacement of dynamite as the most widely used blasting
agent. Approximately 16 billion pounds of ammonium nitrate for all purposes was manufactured in the U.S.
in 1989 (C&E News 1990). According to the Bureau of Mines (1991), sales of ammonium nitrate blasting
agents included 3.2 billion pounds of unprocessed ammonium nitrate and 0.68 billion pounds of ANFO.
To manufacture ammonium nitrate, nitric acid is neutralized with ammonia, to yield the ammonium
nitrate salt either in the form of prills or granules. The prills are obtained by spraying concentrated, liquid
ammonium nitrate downward in a large tower. As the droplets fall and are cooled, they form spherical solids
known as prills. Granulation is achieved in a process similar to the prilling process; however, a smaller volume
of air is used and one-eighth inch granules are produced. Ammonium nitrate fuel oil is a simple mixture of
94 percent ammonium nitrate in unmilled porous prills, 6 percent absorbed fuel oil, and an anticaking agent.
Blasting agents have replaced dynamite as the explosive of choice for most uses, including mining,
construction, and demolition. Greater safety can be maintained with blasting agents than dynamites because
they may be transported as components and prepared on site. Blasting agents are frequently mixed on site
and pumped as needed by pneumatic devices. Ammonium nitrate, the most widely used blasting agent, is easy
to handle and inexpensive. Emulsions, slurries, and gels can also be classified as blasting agents, if they pass
the testing criteria as specified by DOT in 49 CFR.
In 1947, an explosion of ammonium nitrate in a cargo ship, believed to be caused by the presence of
flammable organics, led the Fire Prevention Bureau of Texas to issue the first storage guidelines for
-------
-11-
aramonium nitrate. The U.S. Bureau of Mines concurred with these findings and national guidelines were
adopted. If stored by itself, ammonium nitrate is quite stable. Ammonium nitrate must be considered a high
explosive when in the presence of a priming explosive, confined at elevated temperatures, or found in the
presence of readily oxidizable compounds. ATF specifies separation distances for ammonium nitrate (by itself)
and blasting agents from explosives and other blasting agents (ATF 1990).
Transportation of blasting agents, Division 1.5 under the international classification system, is regulated
by DOT (49 CFR 114a). Blasting agents may be stored in rigid containers which can withstand a four foot
drop onto a concrete surface or a package (e.g., bag or tube) which can withstand three consecutive four foot
drops. Blasting agents may not be transported in tanks, tankers or tank cars unless exempted by the Office
of Hazardous Materials Regulation.
3.4 Storage of Explosives
ATF regulations include specifications for storage magazines, as well as specifications for storage
distances. Type 1 (permanent) and type 2 (mobile and portable) magazines are designed for high explosives;
other types of explosives may also be stored in them. These magazines must be bullet-resistant, weather-
resistant, theft-resistant, and fire-resistant, except in the case of indoor type 2 magazines where the magazine
itself need only be fire-resistant and theft-resistant if the building in which it is housed adequately provides
bullet- and weather-resistance. Both of these magazine types must be adequately ventilated. Type 1 magazines
are typically igloo, Army-type, tunnel or dugout designs and must have a bullet-resistant roof (or 24-inch
earthen mound covering the top, sides, and rear) and no other openings besides entranceways and those for
ventilation. Type 2 magazines are mobile or portable indoor and outdoor facilities for explosive storage.
Common examples include boxes, trailers, and semi-trailers. Inside storage of high or other explosives in type
2 magazines may not exceed fifty pounds and magazine location may not be within a residence or dwelling.
Type 3 magazines are portable outdoor magazines for the temporary storage of high explosives, and also may
be used for low explosives and blasting agents. Often referred to as "day-boxes," these magazines must be fire-,
theft-, and weather-resistant and, when containing explosives on-site, be within the line-of-vision of the blaster
(ATF 1990).
Type 4 magazines are indoor or outdoor magazines for the storage of low explosives, and also may be
used for blasting agents, and electric blasting caps not subject to mass detonation. Typical structures meeting
type 4 requirements include army-type structures, tunnels, dugouts, trailers, semi-trailers, or buildings. These
magazines must be constructed so as to be fire-, theft-, and weather-resistant (if outdoors or if the building
itself fails to provide adequate weather protection).
Type 5 magazines are permanent or portable, indoor or outdoor, theft- and weather-resistant structures
for the storage of blasting agents. Typical structures that serve as type 5 magazines are igloo or army-type
structures, dugouts, bins, trailers, semi-trailers, bulk trailers, tank trailers, or other mobile containers. Indoor
type 5 magazines may not be located in a residence or dwelling and can not contain more than fifty pounds
of blasting agent (ATF 1990). Magazines of Types 1 through 4, for high and low explosives, are normally
smaller and of heavier construction than Type 5 magazines for blasting agents and are brightly colored (U.S.
Fire Administration).
Additional information on explosives magazines may be found in Appendix C. Storage distances for
explosives are discussed in Sections 5.2, 6.2.2, and 6.2.3.
-------
-12-
4.0 Past Incidents Involving Commercial Explosives
This section discusses accidents involving commercial explosives that have been reported in EPA, DOT,
and other hazardous materials databases and in Bureau of Mines reports. This analysis of past accidents was
undertaken because the lesson of Bhopal, India, cautions that knowledge of potentially catastrophic situations
may be of value even though the occurrence of a large accident is rare. Of special consideration would be any
emergency planning activities that require only minimal effort, but could provide maximum benefit in case of
a serious accident
4.1 Explosives Accidents Reported in EPA's AHE Data Base and Other Hazardous Materials
Databases
The Acute Hazardous Events Data Base (AHE/DB) was developed by EPA to provide a historical
perspective on accidents involving the release of hazardous substances in the United States (EPA 1988). The
1988 version of the database includes data on 5,827 events that occurred between the 1960s and 1987 (the
majority of the incidents appear to have occurred in the 1980s). The AHE database lists approximately 36
incidents resulting in fires or explosions that may have involved commercial explosives.1 This database does
not appear to provide clear evidence that commercial explosives should be considered extremely hazardous
based solely on their accident history. It does show that commercial explosives have been involved in accidents
that resulted in death and injury in the past
A recent accident not included in the AHE/DB occurred in Kansas City, MO, on November 29,1988.
In this incident, six fire fighters were killed in an explosion of a blend of ammonium nitrate and fuel oil
(ANFO). Two other firefighters about a quarter of a mile away received minor injuries; the windshield of
their car was blown in. The explosion broke windows tax from the site. The explosion involved about 50,000
pounds of ANFO, a blasting agent, stored in two trailers at a construction site. Most of the stored ANFO was
mixed with aluminum pellets, which increase the explosive power of ANFO. The trailers were Type 5
magazines specifically designed to transport and store blasting agents. The trailers were less than 100 feet
apart; according to ATF regulations, they should have been 224 feet apart The fire that led to the explosion
is believed to have been caused by arson. The firefighters may not have been aware that the trailers were
explosives magazines containing blasting agents (U.S. Fire Administration).
There have been several accidents in the past involving fireworks manufacturing plants. An explosion
in a fireworks plant in Oklahoma in 1985 killed 21 workers and injured five. Although the exact cause of this
explosion could not be determined, it is probable that it was caused by a quantity of spilled flash-and-sound
powder (a combination of potassium perchlorate blended with aluminum or sulfur) that was ignited by friction.
The initial explosion caused a series of explosions that leveled most of the facility and were felt 13 miles away
(Kyte 1985).
The Fatal Hazardous Materials Accidents Database, developed by Resources for the Future, contains
information on fatal accidents which involved the release of hazardous materials. Data were obtained from
1 The identities of the substances involved are not always known. In some cases the substances reported
may be considered commercial explosives under certain conditions but not under others. For these reasons,
the total number of incidents involving commercial explosives is an estimate.
-------
-13-
encyclopedias, almanacs, books, reports, articles, newspapers and computer files (including AHE/DB). The
database contains reported accidents which occurred during industrial production, storage, handling and
transportation of hazardous materials. Only those accidents involving at least one reported death for U.S.
accidents and five reported deaths for accidents outside the U.S. are included. Every such event between 1945
and 1991 is in the database, with the following exceptions: mining or mineral extraction accidents; handling,
transportation, and storage of munitions, fireworks, and manufactured explosives (manufacture of such
products was counted); and transmission and distribution of natural gas (i.e., pipelines).
There are 1,068 accidents in the database, of which 758 occurred in the U.S. Explosions of materials
described as explosives (not including explosions of flammable substances) at fixed facilities accounted for 38
of these reports, with 446 to 500 associated deaths and 769 to 785 injuries. The database does not distinguish
between on-site and off-site fatalities. Fireworks, pyrotechnics, and firecracker explosions comprised 11 of the
explosion accidents. Additionally, 17 of the 38 accidents took place in the U.S. (6 of these were also reported
in the AHE/DB).
The Major Hazard Incident Data Service (MHIDAS) is a world-wide databank maintained by the Safety
and Reliability Business (SRD) of AEA Technology and the UK Health and Safety Executive. It is continually
updated, and records and analyzes incidents involving hazardous materials that resulted in or had the potential
to produce an off-site impact Off-site impact includes incidents which involved evacuation and those which
could, but for mitigating circumstances, have led to evacuation, casualties, or damage to property or the
environment.
MHIDAS currently contains information on over 4,000 incidents. There were 193 accidents involving
explosives, which includes munitions, explosives used in construction, fireworks, and pyrotechnics. Exhibits
3 and 4 describe some of the most serious of these 193 incidents. Of the 15 explosives incidents in Exhibit
3, four occurred in the U.S.; all U.S. incidents were transportation related. Exhibit 4 contains 16 fireworks
and pyrotechnics incidents, all related to factories or transportation, including five in the U.S.
4.2 Transportation Accidents
Another database containing information on accidents involving hazardous substances is the
Department of Transportation's (DOT) Hazardous Materials Information System (HMIS). The HMIS
database contains transportation accident reports from 1971 through 1991 involving release of DOT regulated
hazardous materials, including high explosives, low explosives, minimum hazard explosives (e.g., common
fireworks), and blasting agents (reports for blasting agents are primarily from the 1980s and 1990s; the blasting
agent hazard class did not exist earlier). A summary of information on the explosives incidents in the database
is presented in Exhibit 5. Some general findings from the database include:
The database contains reports on 351 incidents involving explosives for the twenty year
period, an average of more than 17 reports per year. An explosive "incident" may involve
a spill, a fire, an explosion or a combination of these.
In these incidents, there were 10 deaths reported, no major injuries, 130 minor injuries,
two evacuations, and over five million dollars in damages.
-------
-14-
Most high explosives incidents (approximately 90 percent) were only spills and did not
lead to Ores or explosions; a somewhat higher percentage of low explosives incidents
involved fires, but the difference is probably not significant
The deaths and injuries reported resulted from only a few incidents, primarily involving
high explosives (eight deaths and 121 injuries in six incidents).
Blasting agents were not involved in any deaths, injuries, fires, or explosions.
No conclusion could be drawn regarding the causes of incidents or ways to avert them because no
particular mode of release of materials appeared more often than any other or was linked to fires or
explosions.
A report from the National Transportation Safety Board (NTSB) suggests that a single incident, rather
than a historical pattern, can contribute to the regulation of explosives. This report describes an explosion
of 2,600 bombs being carried by train through a sparsely populated area of Arizona. The explosion created
a huge crater, scorched the desert for a quarter mile in all directions, and blew parts of railcars up to three
quarters of a mile away. NTSB analysis indicated that the accident was probably initiated by sparks from the
brakeshoes which started a fire in floorboards of one of the boxcars containing bombs. As a result of this
accident, the Federal Railroad Administration issued a rule including requirements for special types of
brakeshoes and steel subfloors or spark shields in railcars carrying Class A explosives, as well as additional
inspection requirements for such cars
4.3 Blasting Accidents in Mines
Death and injuries due to blasting accidents are not uncommon in the coal and metal mining industries.
Data on blasting accidents in coal mining indicate that deaths and injuries usually occur during planned rather
than accidental explosions and do not affect the surrounding communities. Use of explosives and breaking
agents in underground and surface coal mining and coal preparation plants caused a total of 64 deaths and
1,109 nonfatal injuries from 1971 through 1992. These data are presented in Exhibit 6. This is an average
of 3 deaths and SO nonfatal injuries per year (MSHA 1992). As Exhibit 6 indicates, the death and injury rate
has declined since the "70's.
Blasting accidents are responsible for a relatively small percentage of the total number of deaths and
injuries in the coal mining industry. Blasting accidents accounted for 2.3 percent of the annual average of 134
fatalities and 03 percent of the annual average of 20,014 nonfatal injuries in coal mining from 1971 through
1984 (Peltier, Fletcher, and D'Andrea 1986).
The data indicate that the great majority of the blasting accidents that occur were the result of human
error and were not due to faulty or unsafe products (Peltier, Fletcher, and D'Andrea 1986). Abstracts from
297 accidents that occurred from 1978 through 1984 indicate that most accidents were due to failure of the
blast area security system, i.e., failure to clear the blast zone, failure of personnel to follow instructions,
inadequate guarding, and failure to retreat to a safe location and/or take adequate cover. This was especially
true of underground coal mining, where 166 of 208 incidents were caused by some failure in the blast area
security system (Peltier, Fletcher, and D'Andrea 1986).
-------
-15-
EXmBIT3
INCIDENTS INVOLVING EXPLOSIVES
Incident LocaUon,Dale
Thung Maphrao,
Thailand 02/15/91
Oaram Chashma,
Pakistan 1 in 6/89
Gwynedd. UK 06/14^8
Arzamas, USSR
06/04/88
Chaguaramas, Trinidad
04/26/88
Islamabad. Pakistan
04/10/88
Grantsville. UT
07/30/86 -
Checotah, OK
08/04/85
Deh-Boz Org; Iran
08/19/80
Shelbyvillc. KY
01/29/81
Bangkok, Thailand
11/16/80
Tauriano. Italy
10/12/79
Roseville, CA
04/28/78
In. South Korea
11A2/77
Seoul. South Korea
08/14/70
Cause/Description
Truck carrying dynamite tipped at road. Lighted cigarette from looting villagers suspected
explosion cause. 50 houses destroyed.
Ammunition depot explosion threw rockets and shrapnel onto nearby homes.
Nitroglycerine blast destroyed nitroglycerine mixing building and effluent dispersal
building. Shops damaged up to 0.8 kilometers away, blast heard 3.2 km away.
Explosion of 3 box cars (120 metric tons) of industrial-type explosive produced a crater 26
meters deep and 53 m wide. 150 homes destroyed. Improper segregation and high
ambient temperatures suspected causes.
Mountain brush fire engulfed ammunition bunker. Explosion levelled everything within
305 meters; nearby jeep and fire truck thrown 200 meters.
Series of explosions in a ammunition dump threw munitions, including rockets, bombs,
shells, and shrapnel (up to 10 kilometers) into town.
Three detonations initiated during unloading of four metric tons of explosives produced a
hole 14 meters deep and threw debris 1.6 kilometers.
Collision between munitions vehicle and car and subsequent rupture of fuel tank resulted
in three bomb explosions. Produced a crater 8 meters x 10 meters.
Welding on door of construction firm explosives store resulted in fire and subsequent
explosion of 0.5 metric ton of contents.
Overturned truck carrying explosives caught Ore and exploded. Explosion felt 40
kilometers away.
Munitions factory accidents, including poor rocket fuse connection, led to series of
explosions at ordnance depot. Most victims were local families. Area 500 m* levelled.
Flying debris from explosion in a factory making dynamite and emptying shells killed five,
including one 11 -year old child offsite. Buildings hundreds of meters away were damaged;
electric lines down, resulting in Tauriano power loss.
Explosion on ammunition train caused fire and subsequent explosions over a period of
seven hours. Railyard, 6,000 unfused bombs, and 24 buildings destroyed. Area of
4.8-kilometer radius evacuated.
Watchperson falling asleep knocked over candle which caught fire and spread to cargo
load of dynamite. Explosion of load occurred at a crowded station. Caused damage to
buildings and produced 15 meter crater.
Explosion of 50 40kg bags of gunpowder exploded during attempted disposal. Explosion
leveled 2 houses and damaged as many as 200 houses up to 2 km away. Fire consumed
over 3.000 drums of chemicals.
Death, Injuries,
Damage
171 killed
>100 injured
40 killed
>20 injured
2 killed, 8 injured
73 killed
230 injured
600 evacuated
6 killed
. 15 injured
> 100 killed >1000
injured
5 killed
49 injured
6.000 evacuated
80 killed
45 injured
3 fire fighters 200 yards
away hurt
54 killed
353 injured .
5 killed, 19 injured
52 injured, 2000
evacuated, $10
mil.damage
57 killed
1300 injured
10.000 homeless
9 killed (5 off-site), 59
injured, 1,500 evacuated
Source: MHIDAS Database 1993.
-------
-16-
EXHIBIT4
INCIDENTS INVOLVING FIREWORKS AND PYROTECHNICS
Incident Location, Dale
Candelaria, Philippines,
08/08/90
Columbus, Mississippi,
06A3/90
Henan Province, China,
01/19/89
Surabaya, Indonesia, 03/18/88
Enid, Oklahoma
06/16/87
Bocaue, Philippines
06/04/86
Salisbury, UK 02/21/86
Hallett, Oklahoma
06/25/85
Monte Carlo, Monaco
05/03/85
Bellport, New York
11/26/83
Ipswich: Suffolk; UK. 10/14/82
Ponte Da Barca, Ponugal.
08/08/80
Raipun West Bengal; India,
05/04/80
Hery, France
12/11/79
Jakarta, Indonesia
08/29/73
Chicago. Illinois
03/06/72
Cause/Description
Four explosions and a fire occurred at a fireworks factory.
Explosions that leveled fireworks factory were felt 48 kilometers away.
Illegal fireworks factory explosion destroyed 45 houses and damaged 176
others.
Heat of sun thought to have caused ignition of truckload of fireworks
Small explosion in railcar carrying fireworks led to a subsequent explosion that
spread fire to several fireworks warehouses.
A series of three explosions in a fire-cracker factory obliterated the factory
and shattered windows more than 1.6 kilometers away.
Explosion at Dare factory resulted in extensive damage to the factory.
Explosion at a fireworks company possibly caused by fire in a pick-up truck.
If not extinguished, grass fires could have spread to other magazines.
Container of fireworks exploded while being loaded onto a truck. Those
killed were standing next to truck during loading, while the injured were in a
warehouse that caught fire.
Three blasts at a fireworks company cut power to 7000 homes, severely
damaged residential property up to 334 meters away, and threw empty
container 4 miles. Shock wave felt 16 kilometers away. Caused $1 million in
damage.
Welding started a fire in a warehouse containing ammonium/potassium
nitrate/charcoal which led to a series of explosions and evacuation of 750 local
residents.
Passers-by among killed and wounded after fireworks factory explosion.
Children offsite among those killed by explosion at fireworks factory.
A blast from a machine coiling detonator fuzes completely destroyed the
pyrotechnics factory in which it was located. Blast felt 10 kilometers away.
There were 170 workers in fireworks factory when gunpowder exploded.
Those killed were trapped in locked section of the factory. Ten nearby
residents were evacuated.
Nine explosions at a fireworks factory destroyed 16 out of 18 aerial buildings
of the 3.2-square kilometer factory, including the manufacturing facility.
Hundreds of windows in nearby bouses and buildings broken.
Death, Injuries, Damage
>10 lulled
19 injured
2 killed
2 injured
27 killed
22 injured
11 killed
26 injured
1 injured
11 killed
8 injured (2 seriously)
2 killed. 1 injured
22 killed
>4 injuries
3 dead
14 injured
2 killed
24 injured
300 evacuated.
$2 million (U.S.)
7 lulled
14 injured
40 killed
11 injured
20 injured
52 killed, 24 injured
3 killed
17 injured
1 missing
Source: MHIDAS Database 1993.
-------
EXHIBITS. TRANSPORTATION ACCIDENTS INVOLVING EXPLOSIVES
Numbe
Explosive Inciden
Type Total
High 119
Explosives
(A or 1.1)
Low 52
Explosives
(B)
Low Hazard 128
Explosives
(C)
Blasting 52
Agents
(1.5)
Total 351
r or Number of
ts Incidents
With Deaths
2
(2% of total)
(8 deaths)
1
(2% of total)
(1 death)
1
(1% of total)
(1 death)
0
4
(1% of total)
(10 deaths)
Number of
Incidents
With Injuries
5
(4% of total)
(121
injuries)
1
(2% of total)
(3 injuries)
5
(4% of total)
(6 injuries)
0
11
(3% of total)
(130
injuries)
Number of Numbei
Incidents Inciden
With Fires With
Explosioi
9 5
r of Number of
ts Incidents
With Both
is Fire & Exp.
4
(8% of total) (4% of total) (3% of total)
9 3
3
(17% of (6% of total) (6% of total)
total)
13 8
2
(10% of (6% of total) (2% of total)
total)
0 0
31 16
0
9
(9% of total) (5% of total) (3% of total)
Source: U.S. DOT, Hazardous Materials Information System (HMIS), 1971-1991.
-------
-18-
Injuries due to blasting accidents are less frequent in metal mining than in coal mining. Use of
explosives and breaking agents in underground and surface metal mines and metal mills caused a total of three
deaths and 63 nonfatal injuries from 1985 to 1992 (see Exhibit 7). This is an annual average of 0.4 deaths and
8 nonfatal injuries per year. No data on the causes of blasting accidents in metal mines were found.
4.4 Black Powder Accidents Reported by Department of Defense
According to a list of accidents involving black powder provided by the Department of Defense
Explosives Safety Board (DODESB 1992), 91 world-wide accidents over a 128-year period from 1844-1972
accounted for 337 deaths and 336 injuries. These accidents included 46 in the U.S. with 97 deaths and 106
injuries. Worldwide in the SO years prior to and including 1972,75 black powder accidents accounted for 256
deaths and 284 injuries. During this period, there were 40 accidents in the U.S. accounting for 54 injuries and
53 deaths. Accidents in the U.S. with black powder involved mostly manufacturing.
In addition, somewhat more detailed information on some black powder incidents was provided by
DODESB. This information indicates that in at least some cases, there was detonation of the black powder;
little information was provided in these cases about the quantities involved or the conditions under which the
detonations occurred. For most of the reported incidents, information was not available as to whether a
detonation or deflagration occurred (DODESB 1992).
5.0 Existing Regulations, Standards, and Classification Systems
The Bureau of Alcohol, Tobacco and Firearms (ATF) and the Department of Transportation (DOT)
are the two principal regulatory groups that oversee explosives. Explosives and explosive mixtures generally
are not distinguished in any classification schemes. Most organizations recognize that a "mixture" containing
an explosive substance can be just as hazardous as a "pure" explosive substance. Therefore, the classification
of explosives by ATF, DOT, and others is not based directly on chemical propenies, rather it is based on
empirical data obtained from tests performed on the substance or mixture. These standardized tests typically
are performed by the Bureau of Mines and other agencies. Because classification is based on empirical tests
rather than chemical formulation, and there are many different chemical formulations for explosives, the
number of specific chemicals and chemical mixtures classified as explosives is unknown (White 1989, Dowling
1989, McCune 1989, Schultz 1989).
5.1 DOT Standards
The DOT regulates the transportation of explosives. DOT published a final rule on December 21,1990
(55 FR 52402) including changes to the Hazardous Materials Regulations with regard to explosives, based on
the United Nations Recommendations on the Transport of Dangerous Goods. The rule harmonizes domestic
regulations for explosives with those used internationally. Under this rule, an explosive is defined as, "any
substance or article, including a device, which is designed to function by explosion (i.e., an extremely rapid
release of gas and heat) or which, by chemical reaction within itself, is able to function in a similar manner
even if not designed to function by explosion...." Explosives are listed in Class 1, which is divided into the
following six divisions:
-------
-19-
EXHIBIT 6. FATAL AND NONFATAL INJURIES DUE TO BLASTING ACCIDENTS IN COAL MINING
Year
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984*
1985
1986
1987
1988
1989
1990
1991
1992*
TOTAL
Annual
Average
Underground
F
2
2
1
1
0
0
3
3
0
7
3
3
0
1
6
0
0
2
0
3
2
0
39
1.8
NF
74
94
66
87
64
58
57
51
49
48
28
19
18
16
32
18
23
20
15
15
10
2
864
39J
Surface
F
2
1
1
2
1
0
3
0
1
1
2
1
0
1
0
3
1
0
2
1
1
0
24
1.1
NF
6
17
16
4
18
23
26
19
12
13
9
5
7
13
3
8
2
11
12
6
5
0
235
10.7
Preparation Plants
F
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
NF
0
0
0
0
1
0
2
0
2
1
1
0
2
0
0
0
0
1
0
0
0
0
10
0.5
TOTAL
F
4
3
2
3
1
0
6
3
2
8
5
4
0
2
6
3
1
2
2
4
3
0
64
2.9
NF
80
111
82
91
83
81
85
70
63
62
38
24
27
29
35
26
25
32
27
21
15
2
1,109
50.4
F= Fatal Injuries
NF=Nonfaial Injuries
* Preliminary data, January-December 1984, January-March 1992.
Sources: Peltier, M.A., L.R. Fletcher, and D.V. D'Andrea, "Coal Mining Blasting Accidents," Engineering Health and Safety
in Coal Mining, Proceedings of the Society of Mining Engineers Symposium, March 2-6, 1986; Department of Labor, Mine
Safety and Health Administration.
-------
-20-
EXHIBIT 7. FATAL AND NONFATAL INJURIES DUE TO BLASTING ACCIDENTS
IN METAL MINING
Year
1985
1986
1987
1988
1989
1990
1991
1992*
TOTAL
Annual
Average
Underground
F
0
0
0
0
1
0
2
0
3
0.4
NF
5
4
5
9
11
7
4
5
SO
6.2
Surface
F
0
0
0
0
0
0
0
0
0
0
NF
0
2
3
0
3
2
0
0
10
1.2
Mills
F
0
0
0
0
0
0
0
0
0
0
NF
0
0
1
0
1
0
1
0
3
0.4
TOTAL
F
0
0
0
0
1
0
2
0
3
0.4
NF
5
6
9
9
15
9
5
5
63
7.9
F= Fatal Injuries
NF=Nonfatal Injuries
* Preliminary data, January-March 1992.
Source: Department of Labor, Mine Safety and Health Administration.
-------
-21-
Division 1.1: Explosives that have a mass explosion hazard, where a mass explosion hazard is
one that affects almost the entire load instantaneously.
Division 1.2: Explosives that have a projection hazard but not a mass explosion hazard.
Division 1.3: Explosives that have a fire hazard and either a minor blast hazard or a minor
projection hazard or both, but not a mass explosion hazard.
Division 1.4: Explosive devices that present a minor explosion hazard and do not contain more
than 25 grams (0.9 ounce) of a detonating material.
Division 1.5: Explosives that have a mass explosion hazard but are so insensitive that there is
very little probability of initiation or of transition from burning to detonation under normal
conditions of transport.
Division 1.6: Articles that contain only extremely insensitive detonating substances and which
demonstrate a negligible probability of accidental initiations or propagation.
Test methods and procedures for classifying explosives according to these divisions are described in the
U.N. document, Recommendations on the Transport of Dangerous Goods. Tests and Criteria. Classification
procedures are as follows:
Any material that is designed to have an explosive effect is first tested to determine if it is too
sensitive for transport, using tests for shock sensitivity and thermal sensitivity. Materials found
to be thermally unstable are not permitted to be transported. Materials that do not pass the
tests for shock sensitivity must be de-sensitized or encapsulated and submitted for another series
of tests before they can be transported.
Materials that are designed to have an explosive effect and that are not too sensitive for
transport are accepted for Class 1 (explosives). Further tests (described in Appendix D) are
carried out to determine the hazard division (1.1, 1.2, 1.3, or 1.4).
Materials that are not designed to have an explosive effect are also tested for shock sensitivity
and thermal sensitivity. If they are not too sensitive for transport, they are further tested to
determine whether they have explosive properties, using tests that determine whether the
materials detonate under confinement (see Appendix D for a description of the tests). If they
are found to have explosive properties, they are tested to determine the hazard division, as
described for materials designed to have an explosive effect.
Very insensitive explosive materials with a mass explosion hazard are subjected to a series of
tests to determine cap sensitivity, tendency to undergo a transition from deflagration to
detonation, and behavior under Ore conditions (explosion in a fire or tendency to ignite readily).
Assignment of a substance to specific hazard classes and divisions depends on the composition of the
substance and on the tests the Department of Transportation performs on the substance. Theoretically, the
same substance could receive different classifications depending on test results. Nitrocellulose, which is used
in furniture finishes, nail polish, and ink (approximately two-thirds of total production), as well as in
-------
-22-
gunpowder and propulsion systems (Aqualon 1992) has eight individual listings in the DOT Hazardous
Materials Table. These listings are as follows:
Description
Nitrocellulose, dry or wetted with less than 25 per cent water (or
alcohol), by mass
Nitrocellulose, plasticized with not less than 18 per cent
plasticizing substance, by mass
Nitrocellulose, solution, flammable with not more than 12.6 per
cent nitrogen, by mass, and not more than 55 per cent
nitrocellulose
Nitrocellulose, unmodified or plasticized with less than 18 per
cent plasticizing substance by mass
Nitrocellulose, wetted with not less than 25 per cent alcohol, by
mass
Nitrocellulose with alcohol not less than 25 per cent alcohol by
mass, and not more than 12.6 per cent nitrogen, by dry mass
Nitrocellulose with plasticizing not less than 18 per cent
plasticizing substance, by mass, and not more than 12.6 per cent
nitrogen, by dry mass
Nitrocellulose with water not less than 25 per cent water, by
mass
Hazard Class and Division
1.1 (high explosive)
1.3 (low explosive)
3 (flammable liquid)
1.1 (high explosive)
1.3 (low explosive)
4.1 (flammable solid)
4.1
4.1
Black powder is listed on the DOT Hazardous Materials Table as a high explosive (1.1). Fireworks are
listed with five classifications: 1.1,1.2,1.3, and 1.4, depending on test results. Common or consumer fireworks
(e.g., firecrackers) would fall in Division 1.4, while special fireworks used for display purposes usually would
be classed as 1.3.
DOTs rule, containing the classifications presented above, provides a five-year transition period for the
changeover to the new classifications and requirements. Under the previous DOT regulations, an explosive
is defined as "any chemical compound, mixture, or device, the primary or common purpose of which is to
function by explosion, i.e., with substantially instantaneous release of gas and heat..." (49 CFR 173.50).
Explosives are further defined by the DOT according to the following classifications:
Class A Detonating or otherwise of maximum hazard:
Type 1: Solid explosives which can be caused to deflagrate by contact with sparks
or flame but cannot be detonated by means of No. 8 test blasting cap.2 Example:
black powder.
2. A No. 8 blasting cap is one containing two grams of a mixture of 80% mercury fulminate and 20%
potassium chlorate, or a cap of equivalent strength (49 CFR 173.53).
-------
-23-
Tvpe 2: Solid explosives which contain a liquid ingredient and which when
unconfined can be detonated by means of a No. 8 test blasting cap; or which can
be exploded in at least 50% of the trials in the Bureau of Explosives' Impact
Apparatus under a drop of 4 inches or more, but cannot be exploded in more than
50% of the trials under a drop of less than 4 inches. Example: commercial
dynamite containing a liquid explosive ingredient.
Type 3: Solid explosives which contain no liquid ingredient and which can be
detonated by means of a No. 8 test blasting cap; or which can be exploded in at
least 50% of the trials in the Bureau of Explosives' Impact Apparatus under a
drop of 4 inches or more, but cannot be exploded in more than 50% of the trials
under a drop of less than 4 inches. Examples: commercial dynamite containing
no liquid ingredient, TNT, amatol, picric acid, urea nitrate, pentolite, and
commercial boosters.
Type 4: Solid explosives which can be caused to detonate when unconfined, by
means of contact with sparks or flame; or which can be exploded in the Bureau
of Explosives' Impact Apparatus, in more than 50% of the trials under a drop of
less than 4 inches. Examples: lead azide, fulminate of mercury.
Type 5: Desensitized liquid explosives are explosives which may be detonated
separately or when absorbed in sterile absorbent cotton, by a No. 8 test blasting
cap; but which cannot be exploded in the Bureau of Explosives' Impact Apparatus
under a drop of less than 10 inches. Example: desensitized nitroglycerine.
Type 6: Liquid explosives that can be exploded in the Bureau of Explosives'
Impact Apparatus under a drop of less than 10 inches. Example: nitroglycerine.
Type 7: Detonators and initiating devices.
Type 8: Any solid or liquid compound, mixture, or device which is not specifically
included in any of the above types, and which under special conditions may be so
designated and examined by the Bureau of Explosives. Example: shaped charges.
Type 9: Propellant explosives, class A (solid chemicals or solid chemical mixtures
which are designed to function by rapid combustion of successive layers) which
detonate in their package in one out of five test trials, when ignited.
Class B Flammable Hazard:
Those explosives that in general function by rapid combustion rather than
detonation and include some explosive devices such as special fireworks, flash
powders, some pyrotechnic signal devices and liquid or solid propellant explosives
which include some smokeless powders.
-------
-24-
Class C Minimum Hazard:
Certain types of manufactured articles which contain Class A, or Class B
explosives, or both, as components but in restricted quantities, and certain types
of fireworks.
DOT also regulates transportation of blasting agents, defined under the old rule as materials designed
for blasting that have been tested in accordance with a series of specified tests and found to be so insensitive
that there is very little probability of accidental explosion.
DOT provides the following comparison of the old and new hazard classifications:
Old Classification New Classification
Class A explosives Division 1.1
Class A or Class B explosives Division 1.2
Class B explosives Division 1.3
Class C explosives Division 1.4
Blasting agents Division 1.5
No applicable class Division 1.6
5.2 ATF Standards
The manufacture, processing, use, distribution, and storage of explosive materials is regulated also by
the Bureau of Alcohol, Tobacco, and Firearms (ATF) as set forth in 27 CFR Pan 55. The ATF defines an
explosive material as follows:
"Any chemical compound, mixture, or device, the primary or common purpose
of which is to /unction by explosion. The term includes, but is not limited to,
dynamite and other high explosives, black powder, pellet powder, initiating
explosives, detonators, safety fuses, squibs, detonating cord, igniter cord, and
igniters."
Explosive materials have been divided into three classes by the ATF for purposes of regulating storage:
High Explosives: those explosives which can be detonated by means of a
blasting cap when unconfined (e.g., TNT, dynamite).
Low Explosives: those explosive materials which can be caused to deflagrate
when confined (e.g., black powder, safety fuses, and DOT Class B explosives).
Blasting Agents: any material or mixture consisting of a fuel and oxidizer
that is intended for blasting and not otherwise defined as an explosive,
provided that the finished product cannot be detonated by a No. 8 blasting
cap when unconfined (e.g., ammonium nitrate/fuel oil (ANFO) and certain
water gels).
-------
-25-
Note that blasting agents are very insensitive and unlikely to explode: however, in the event that they do
explode, the consequence may be a detonation, with accompanying overpressure, as for high explosives, rather
than a deflagration (intense fire), as for low explosives. For example, an explosion of the blasting agent ANFO
killed six fire fighters in Kansas City, MO (see Section 4.1).
The ATF regulations contain a table, called the "Table of distances for storage of explosive materials"
(ATF 1990), which must be used by explosive materials handlers to determine acceptable minimum separation
distances for storing and handling high explosives and blasting agents. The table also states the maximum
quantities of explosives allowed in any one location. The table of distances is further discussed in Section 6.2
and is presented in Exhibit 9 in that section. Locations that contain explosives must be separated from:
Inhabited buildings, including structures or other places not directly related
to explosives operations where people usually assemble or work;
Public highways;
Passenger railways; and
Other locations which contain explosives.
For example, if a site has from 40 to 50 pounds of high explosives in one magazine, there can be no
unbarricaded inhabited building (e.g., a home) within 300 feet (91 meters) of thai magazine. A separate table
is provided for determining separation distances for the storage of low explosives (see discussion in Section
6.2 and Exhibit 10). These distance requirements differ from those for high explosives because the
consequences of an accident involving low explosives are likely to result from thermal radiation rather than
overpressure. In all instances, the separation distances increase as the quantity of explosives stored increases.
The ATF table of distances is based on the American Table of Distances for storage of explosives as
revised by the Institute of Makers of Explosives (IME). The original American Table of Distances was
published in 1910 based on the work of a special committee of the Association of Manufacturers of Powder
and High Explosives. This committee gathered and evaluated information on a number of explosions ranging
from small amounts of explosive materials to one million pounds. Since then, the table has been revised
periodically to incorporate information obtained from past accidents. The explosions studied have covered
a period of almost fifty years and have occurred in manufacturing, transportation, and storage, both in the
United States and abroad. The last revision to the table was in 1985. The table is designed to cover
manufacture and permanent storage of explosive materials and is not applicable for the incidental handling
or temporary storage of explosive materials being transported.
ATF regulations also include requirements for special fireworks, pyrotechnic compositions, and
explosive material used in assembling fireworks. Tables are provided for distances between fireworks process
and fireworks nonprocess buildings, between fireworks process buildings and other areas (e.g., passenger
railways, public highways, inhabited buildings), and distances for the storage of special fireworks. The ATF
does not regulate common fireworks (e.g., firecrackers), except during manufacture.
-------
-26-
5.3 NFPA Codes
The National Fire Protection Association (NFPA) publishes the National Fire code, intended to be
referenced by public authorities in laws, ordinances, regulations, and administrative orders. NFPA 495,
Explosive Materials Code (NFPA 1990) applies to manufacture, transportation, storage, sale, and use of
explosive materials. The code includes detailed requirements related to explosives safety, including permitting,
record keeping, and reporting; handling of blasting agents; transportation of explosives; storage requirements,
including use of the American Table of Distances; and use and disposal of explosives. The code is intended
to conform to federal regulations, including those promulgated by ATF, DOT, and the Mine Safety and Health
Administration. NFPA has also published two codes for fireworks: NFPA 1124, Manufacture, Transportation
and Storage of Fireworks, and NFPA 1123, Code for the Outdoor Display of Fireworks.
5.4 State and Local Regulations
Many state and local governments have regulations regarding licensing and safety requirements for
explosives. Some of them may adopt the NFPA code for explosives, NFPA 495, described above, or the
explosives provisions of the Uniform Fire Code, which are similar to the provisions of NFPA 495. In some
localities there may be three or four sets of requirements (Federal, state, and one or more local, e.g., county
and city). Some states have "home rule" provisions, leaving regulation of explosives to local authorities. There
is no uniformity in the state and local regulations, however. For example, the permitting authority for
explosives may be the fire marshal or the fire department, or it may be the police department, labor
department, or other agency. The state of Massachusetts, for example, gives the fire department authority over
explosives (Henry 1989). In Kansas City, MO, where six firefighters were killed in an explosion, the permitting
authority was the City Engineer's office; at the time of the accident, there was no requirement that the fire
department be notified (the regulation was subsequently modified to require that blasting permits be cleared
through the Fire Marshal's office) (U.S. Fire Administration).
5.5 Clean Air Act Regulations
Under the Clean Air Act Amendments of 1990, OSHA was required to promulgate regulations
intended to prevent accidental releases of chemicals which could pose a threat to employees. OSHA has
published a standard, described below in 5.5.1. EPA is required to develop regulations for protection of the
public, as discussed in 5.5.2.
5.5.1 OSHA Process Safety Management Standard
OSHA has developed a chemical process safety management standard (57 FR 6356, February 24,1992)
that applies to any process involving a highly hazardous chemical at or above a specified threshold quantity.
The requirements of this standard also apply to the manufacture of explosives and pyrotechnics (no threshold
quantity is specified in this case). The OSHA Process Safety Management Standard is intended to protect
employees by preventing or minimizing the consequences of catastrophic releases of toxic, reactive, flammable,
or explosive chemicals. The requirements include development of a compilation of written process safety
information; process hazards analysis, carried out by a team with expertise in engineering and process
operations; development of written operating procedures for processes involving highly hazardous chemicals;
employee training; pre-startup safety reviews for new and significantly-modified facilities; maintenance of
mechanical integrity of critical equipment; and establishment of procedures for management of changes to
-------
-27-
process chemicals, technology, equipment, and procedures; investigation of incidents; emergency action plans;
and compliance and safety audits to ensure programs are in-place and operating properly.
5.5.2 EPA Chemical Accident Prevention Regulations
Under section 112(r) of the Clean Air Act (CAA), EPA is required to develop a list of at least 1(X)
regulated substances that, when released, can cause death, injury, or serious adverse effects to human health
or the environment. This list could include explosives, although Congress did not specifically mandate
inclusion of explosives. Facilities that use these chemicals above threshold quantities developed by EPA need
to comply with new CAA regulations on release prevention, detection, and emergency response. One accident
prevention provision of the CAA mandates the development of regulations requiring facilities to prepare and
implement risk management plans. EPA is currently developing a list of regulated substances and threshold
quantities, as well as risk management plan regulations. EPA's proposed rule on the list of regulated
substances and threshold quantities for accidental release prevention (58 FR 5102, January 19, 1993) includes
commercial explosives as defined by DOT in Division 1.1; the proposed threshold quantity is 5,000 pounds.
5.6 Other Standards
Other agencies such as the Mine Safety and Health Administration (MSHA), Department of Defense
(DOD), the International Maritime Organization (IMO), and others regulate certain aspects of the explosives
industry. For example, DOD regulates the storage of munitions and military explosives, OSHA regulates
worker safety at explosives manufacturing sites, the IMO is concerned with international shipments of
hazardous materials, and MSHA regulates worker safety and includes provisions for explosives. The standards
of these organizations are not related to hazards to the community from explosives; therefore, they were not
considered further in this analysis.
6.0 Methodology for Determining Affected Distances for Commercial Explosives
6.1 Approach
The Agency's Physical/Chemical Criteria Workgroup used the ATFs American Table of Distances,
and the equations on which the table is based, to estimate the consequences of explosions and evaluate the
relationship of quantity exploded to the distance affected at various overpressure levels. Several other widely
recognized explosives consequence methodologies, including the K-factor and the Automated Resource for
. Chemical Hazard Incident Evaluation (ARCHIE) computer program, are discussed in Appendix B. For this
analysis, commercial explosives have been grouped into three categories: high, low, and blasting agents, in the
same way ATF has classified explosives for storage purposes. These categories can also be related to the new
DOT/international hazard classes and divisions, described in Section 5.1 above; i.e., high explosives correspond
to materials in Division 1.1, low explosives correspond to materials in Division 1.3, and blasting agents
correspond to materials in Division 1.5 (as noted in Section 3, blasting agents make up the bulk of explosives
in commercial use). Materials in Division 1.2 have a projection hazard rather than a mass explosion hazard;
the method of consequence analysis described here is not applicable to such materials, nor is it applicable to
materials in Divisions 1.4 or 1.6, which have only minor blast hazards.
6.1.1 Damage Criteria
The next step in the consequence analysis is the identification of credible accident scenario(s). For
high explosives, only one scenario is likely: a quantity of explosive is somehow detonated. Damage from
accidents involving explosives is a function of the peak overpressure resulting from the blast wave. This is
discussed further in Section 6.2.1 below and in Appendix A In the analysis which was used to determine the
threshold planning quantities for acutely toxic materials, lethality was chosen as the endpoint. In the case of
-------
-28-
explosions, human deaths and injuries may be caused directly by blast waves or indirectly by flying glass or
other objects. Very high overpressures are needed for fatality as a direct result of an explosion: various
sources cite threshold levels for fatality of 14.5 pounds per square inch (psi) to 120 psi (see Appendix A).
However, much lower overpressures may cause property damage that might lead to human injury or death.
For this analysis, EPA considered blast overpressures of 0.5 psi, 1.0 psi, 2.0 psi, and 3.0 psi. An overpressure
of 0.5 psi can cause shattering of glass windows (Brasie and Simpson 1968, US Air Force 1983, US DOT 1988,
Lees 1980). An overpressure of 1.0 psi can cause partial demolition of houses, and is the threshold for skin
laceration from flying glass (Lees 1980). An overpressure level of 2.0 psi can lead to the partial collapse of
walls and roofs of houses and shattering of non-reinforced concrete or cinder block wall panels (Brasie and
Simpson 1968, Air Force 1983, U.S. DOT 1988, Lees 1980). Overpressures between 2.0 and 3.0 psi can cause
eardrum rupture (Lees 1980). Overpressure of 3.0 psi can cause a steel frame building to become distorted
and pull away from its foundations (Lees 1980); it can also hurl a person to the ground (Brasie and Simpson
1968).
Appendix A contains more detailed background information on blast overpressures and their effects.
The results of consequence analysis at different overpressures are presented in Section 6.3.1. The results show
that, as would be expected, if a higher overpressure is chosen, the blast effect extends over a shorter distance.
However, the affected distance does not change linearly with overpressure; the rate of decrease in distance is
considerably greater between overpressures of 0.5 psi and 1.0 psi than between 1.0 psi and 2.0 psi or 2.0 psi
and 3.0 psi (see Section 6.3.1 and Exhibit A-6).
Effects of explosions can also include damage and injury from fragments of the casing of the explosive
(primary fragments) and debris from the blast. Such fragments and debris can have high velocities and travel
some distance from the site of the blast. Their effects can be very serious, including secondary explosions and
fires, structural damage, and human injury. However, these effects are highly variable, depending on site-
specific and incident-specific factors. Therefore, it would be very difficult to take these effects into account
in a generalized consequence analysis.
Fragments and debris from explosions can injure people by skin penetration and by the force-.of
impact. People can also be injured or killed by impact if the force of the blast wave throws them against a
stationary object, such as a wall. The severity of the injuries depend on the mass and velocity of the fragments
and debris, as well as a number of other factors, such as the shape of the fragments, the angle of impact, and
the point of impact. In the case of glass fragments from windows broken by an explosion, the initial mass and
velocity depend on the dimensions of the window pane and the blast overpressure compared to the minimum
overpressure at which the glass fails. Whether or not the glass causes injury, and the severity of the injury,
depend on how close a person is to the window and the position of the person (TNO 1992).
6.1.2 Assumptions
The following assumptions were made in the consequence analysis:
Explosions occur at ground level; and
The area surrounding the explosion is characterized by flat open terrain.
In actuality, potential reflections of the blast or shock wave from building walls or the sides of other obstacles
and surfaces may cause damage to be more erratic than that predicted using these assumptions; in fact, damage
may be increased, especially if there are obstacles parallel to the shock wave front. However, the above
assumptions are used widely in performing consequence analyses for explosive materials.
-------
-29-
6.1.3 Quantities Modeled
The following quantities of material were utilized in the analysis: 50, 100, 250, 500, 1,000 and 5,000
pounds. These amounts were chosen to provide a wide range of values in the analysis. In addition, quantities
of 10,000 and 100,000 pounds were considered for low explosives, to provide a wider range of distances.
6.2 Methodology for Determining Explosive Distances
6.2.1 Scaling Law for Explosives
Damage from explosions is primarily a function of the peak overpressure resulting from the blast wave
and is not dependent upon one specific physical or chemical property. It is a function of an empirically
derived function, known as the scaling law:
D = (K)(W1/3)
where: D = Required, or potentially affected, distance, ft
K = Protection factor depending upon the degree of risk assumed or permitted; the relationship
between the peak overpressure and the K factor is shown in Exhibit 8 (e.g., using this
exhibit, a K-factor of 11 (Kll) corresponds to a peak overpressure level of 8 psi)
W =TNT equivalent weight, Ib
The TNT equivalent weight is the weight of TNT that would yield the same peak pressure or impulse at a
given distance as the total weight of explosive material under consideration.
The scaling law relates peak overpressure, explosive weight, and distance for the purpose of evaluating
the effects of blast waves. Blast waves generated during an explosion of a given material are compared to
those which would have resulted from an explosion involving TNT (i.e., TNT is used as a reference material).
This approach is used because of the large amount of experimental data available on blast waves from TNT
explosions and on damage produced by TNT explosions. This relationship is applicable to any explosion
provided the equivalent weight of TNT is known. The scaling law is the basis for the Table of Distances used
for consequence analysis as discussed below. The scaling law is not applicable to assessment of consequences
of flying fragments from a blast
There are several limitations of the scaling law using TNT equivalent weights that need to be
recognized (U.S. DOT 1988). They are:
Not all of the accidentally-released material may be involved in the
explosion. Pan of the material may disperse without reacting and part may
react at a different time or location from the initial explosion. As such,
measured TNT yields of explosive materials depend upon accident
conditions, duration, and failure mode;
Of the portion of released material that reacts in the explosion, pan may
detonate and pan may deflagrate, with the latter contributing little energy
to the blast. Predicting whether a detonation or a deflagration (or a
combination) will occur is very complex. The outcome depends on the
materials' properties and on the conditions of the accident;
The blast characteristics of the released material may be different from those
of a TNT charge with an equivalent energy. For example, measured
-------
-30-
overpressure amplitudes are generally lower and durations are longer for low
explosives than for TNT because of slower reaction rates; and
TNT yields are only available for materials which have been tested. New
chemicals would require testing before consequence analyses could be
performed.
As the above limitations indicate, the TNT yield of a material is not an absolute property such as density or
molecular weight. Instead, it depends upon the test conditions under which it is measured. These limitations,
however, do not present a serious problem in the use of the scaling law equation since the dependence of blast
parameters on yield is low due to the cube-root exponent in the equation. The prediction of a hazard distance
is not very sensitive to the TNT equivalent weight Further, the limitations of the equation will generally lead
to a worst-case estimate and, hence, a conservative analysis. The TNT method of analysis has been used
effectively over many years despite the limitations mentioned above.
The scaling law is accepted widely and well established in the published technical literature. It has
been adopted by several government agencies, including the Bureau of Alcohol, Tobacco, and Firearms, the
Depanment of Defense, and the U.S. Air Force, and private companies for related uses, including hazards
assessment and emergency response and planning. It is used in some of the existing classification systems for
commercial explosives which are described above in Section S.O. No other methods of estimating blast wave
effects have been found. The results of several other methods of consequence analysis for explosives, all based
on the scaling law, are presented in Appendix B for comparison with the results presented in this section.
6.2.2 ATF Table of Distances for High Explosives
The primary purpose of the ATF regulations is to control the distribution of explosive materials to
authorized persons. As pan of this goal, the ATF regulations specify a set of standards to be used in the safe
storage of explosives. The standards are set forth in tabular form in two tables: (1) the Table of Distances"
for high explosives and blasting agents as shown in Exhibit 9, and (2) a second table for low explosives as
shown in Exhibit 10. The distance requirements in these tables are designed to ensure that explosives stored
in accordance with the regulations will not present an off-site hazard. The original Table of Distances is based
upon the scaling law and was developed by the Institute of Makers of Explosives (IME) (Dowling 1989). As
noted in Section 5.2, the table takes into account the results of explosions that have occurred over a 50-year
period in manufacturing, transportation, and storage, both in the United States and abroad.
Since the equation that the table is based upon is known, additions may be made to the table
depending upon the needs of the analysis. For example, the table for high explosives and blasting agents
(Exhibit 9) has several columns representing different overpressures and damage criteria. In calculating the
safe distance to an unbarricaded inhabited building (fourth column), an overpressure of 0.4 psi is used. In
contrast, in determining the safe distance to a barricaded inhabited building (third column), an overpressure
of 1.0 psi is used, and distance requirements are lower. Barricades, by ATF definitions, can be natural barriers,
such as hills or dense timber, or artificial barriers, such as mounds or walls of earth. Such barricades must
be high enough that a straight line from the top of the barricaded magazine (explosives storage building) to
the eave line of another magazine or building, or to a point 12 feet above a railway or highway, will pass
through the barricade. The table of distances provides data for six overpressure levels.3 However, distances
In determining the safe distance to a barricaded public highway, an overpressure of 4.5 psi is used;
for unbarricaded public highways and barricaded passenger railways and high volume public highways, an
overpressure of approximately 1.75 psi is used; for unbarricaded railways and high traffic public highways, an
overpressure of between 0.6 and 0.7 psi is used; and for unbarricaded inhabited buildings, an overpressure of
-------
-31-
EXHIBIT 8. K-FACTORS AS A FUNCTION OF PEAK OVERPRESSURES
I'
' '" :' T \f=- : : ::?;
i a.
i-
.0-
t-
I.. 1 III *
rx
.10-
Ul
.1 .2
4 .» .1 L»
« N
* * i in
t « » i to
K-fACTM
Source: U.S. Air Force. 1983. "Explosives Safety Standards," Air Force Regulation 127-100. 20 May 1983.
-------
-32-
based on various other overpressures can be calculated since the equation on which the data in the table are
based is known.
The differences between the scaling law and the table for high explosives are the following:
The ATF assumes that all high explosives and blasting agents have roughly
equivalent explosive energies and therefore does not calculate an explosive
weight based upon TNT equivalents (W = MAem rather than W =
[Mcbem x %TNT]) for each high explosive material,
where:
W = TNT equivalent weight
Mchem = Total mass of the explosive material
% TNT = TNT equivalent (fraction).
The Tables provide specific distances for ranges of weights, resulting in some
degree of error, especially at the endpoints. For example, the distance
determined using the table for a high explosive weighing 4,000 pounds is
the same as the distance for one weighing 5,000 pounds.
Although the ATF Table of Distances allows shorter distances in cases where barricades are present,
there is evidence that barricades may be ineffective for protection against the effects of blast waves; although
peak pressure may be reduced in a region adjacent to the site of the explosion, it may be increased at greater
distances (Filler et al. 1967). A study of accidental explosions has shown that barricades did not reduce
damage to adjacent buildings in any significant way (Filler et al. 1967). Barricades may protect against
fragments from a blast; however, if the barricade is destroyed by the blast, which often happens in an accident,
it may be a source of additional fragments (Filler et aL 1967).
623 ATF Table of Distances for Low Explosives
The Table of Distances for low explosives (see Exhibit 10) was originally developed by the Department
of Defense and adopted by ATF for application to the private sector. It is designed for use with materials that
deflagrate rather than detonate. The distances are based on levels of heat flux produced by a fireball rather
than blast overpressures. The heat flux level used for determining distances is 4 calories per square inch; no
exposure time is specified. When the table was developed, this heat flux was considered to be a level at which
people could avoid injury by moving away. To calculate distances, the following equation (empirically derived)
is used for inhabited buildings and public railroads and highways (Moran 1989):
D = 8 W 1/3
where:
D = Required, or affected, distance in feet
W = Weight of material in pounds.
For determination of distance to magazines, the equation is D = 5 W 1/3 (Moran 1989). Because the table
of distances for low explosives is based on thermal radiation, the scaling law for explosives and the
methodologies based on it do not give comparable results. None of the models reviewed that give thermal
radiation results (e.g., models for pool fires and BLEVEs) appear to be applicable to explosives that deflagrate
(i.e., low explosives). Therefore, only the equation presented above, used for the ATF Table of Distances for
low explosives, was used for analysis of low explosives.
-------
33-
EXHIBIT 9. AMERICAN TABLE OF DISTANCES - HIGH EXPLOSIVES & BLASTING AGENTS*
OU*NITTVOJ
CAOCO
1*0
180
220
290
280
300
340
380
400
430
470
S10
540
590
640
680
710
750
780
800
850
900
940
980
1.010
1.090
1.160
1.270
1.370
1.460
1.540
1.600
1.670
1.730
1.750
1.770
1.600
1.880
1.950
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.010
2.030
2.098
2.100
2.158
2.219
2.275
PUMJC
CLAW
MX*
CAOCO
30
35
45
SO
59
60
70
75
80
85
99
108
110
120
130
139
148
ISO
198
180
169
170
179
180
188
190
198
210
229
239
248
250
289
280
270
27S
280
288
290
318
340
380
380
400
420
440
488
470
488
900
910
920
930
940
948
980
589
580
988
970
580
590
600
608
610
620
638
690
670
690
oat
AtQHWffSt
At»D"
CAMO
60
70
90
100
110
120
140
ISO
160
170
190
210
220
240
260
270
290
300
310
320
330
340
390
380
370
380
390
420
490
470
490
900
510
520
540
590
580
570
560
630
680
720
780
600
640
680
910
940
970
1.000
1.020
1.040
1.060
1.060
1.090
1.100
1.110
1.120
1.130
1.140
1.160
1.160
1.200
1.210
1.220
1.240
1.270
1.300
1.340
1.380
kNCniMFOT
MSMN00
HtCUCMO
nflM«Wnm
1.000 MM
AM*
CAOtD
51
64
61
93
103
110
127
139
ISO
.159
175
180
201
221
238
283
268
278
280
300
318
338
391
388
378
408
432
474
913
548
973
600
624
648
687
723
798
788
813
878
933
981
1.028
1.068
1.104
1.140
1.173
1.208
1.238
1.283
1.293
1.317
1.344
1.388
1.392
1.437
1.478
1.521
1.557
1.593
1.62*
1.682
1.698
1.728
1.798
1.782
1.838
1.890
1.990
2.000
IAA*.M«ra«
WWMrfcWMI
«»»«»
UMtjAUM
CAOCO
102
128
162
186
208
220
254
278
300
318
350
378
402
442
478
508
532
598
578
600
638
672
702
732
788
618
884
948
1.028
1.092
1.148
1.200
1.248
1.290
1.374
.448
.512
.573
.628
.792
1.888
1.982
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
2.000
MMUUrtON
AM*
CAOCO
6
8
10
11
12
14
15
16
16
19
21
23
24
27
29
31
32
33
35
36
39
41
43
44
49
49
52
58
61
68
68
72
75
78
82
87
90 .
94
98
108
112
119
124
129
139
140
149
150
159
160
169
170
175
160
168
198
208
218
228
238
248
259
268
273
283
299
319
339
380
389
OTMAOAflNa
UNMJMh .
CAOCO
12
16
20
22
24
28
30
32
36
38
4}
44
4«
54
56
62
64
66
70
72
78
62
86
88
90
98
104
116
\zt
130
136
144
150
158
164
174
180
188
'»
210
224
238
246
2S8
270
280
290
300
310
320
330
340
350
360
370
390
410
430
450
470
490
510
530
550
570
590
630
670
720
770
A separate table i* provided by ATF for determining required distances between Mailing agent* and explosives or other blasting
agenu.
Highway! with an average traffic volume of 3.000 or fewer vehicle* per day.
-------
34-
EXHIBIT 10. TABLE OF DISTANCES - LOW EXPLOSIVES
Distances (ft)
Pounds
Over
0
1,000
5,000
10,000
20,000
30,000
40,000
50,000
60,000
70,000
80,000
90,000
100,000
200,000
Not Over
1,000
5,000
10,000
20,000
30,000
40,000
50,000
60,000
70,000
80,000
90,000
100,000
200,000
300,000
From
Inhabited
Building
75
115
150
190
215
235
250
260
270
280
295
300
375
450
From Public
Railroad &
Highway
75
115
150
190
215
455
250
260
270
280
295
300
375
450
From
Above Ground
Magazine
50
75
100
125
145
155
165
175
185
190
195
200
250
300
Source: ATF P 5400.7 (6/90), "ATF: Explosives Law and Regulations,* Bureau of Alcohol, Tobacco, and
Firearms, Department of the Treasury.
-------
-35-
6.3 Results
This section presents the results of application of the methodology discussed in the previous section.
Results for application of the scaling law of distances for high explosives and blasting agents are presented
first, followed by results for low explosives obtained from the equation for thermal radiation effects used by
ATF.
63.1 High Explosives and Blasting Agents
Exhibit 11 presents the results of analysis of high explosives (Division 1.1) and blasting agents
(Division 1.5) using the scaling law as applied by ATF for the table of distances, using unbarricaded distances,
for overpressure levels of 0.4 psi (the level used for development of the table), 1.0 psi, 2.0 psi, and 3.0 psi.
The analyses were carried out for 50, 100, 250, 500, 1,000, and 5,000 pounds of high explosive material. As
noted above, TNT equivalents are not taken into account in using the Table of Distances (i.e., all materials
are assumed to have the same explosive potential); therefore, distances are the same for all high .explosives.
At an overpressure level of 0.5 psi (a level that will shatter glass), distances range from 79 meters for 50
pounds of high explosive to 365 meters for 5,000 pounds of high explosive. At an overpressure level of 1.0
psi (which can cause partial demolition of houses), distances of 48 meters, for 50 pounds of explosive, to 224
meters, for 5,000 pounds of explosive, were calculated. At an overpressure level of 3.0 psi, the distance
calculated for a given quantity of explosive is approximately half the distance calculated for 1.0 psi.
6.3.2 Low Explosives
Exhibit 12 presents the distances calculated for the deflagration of a low explosive (Division 1.3),
based on the equation used in developing the ATF Table of Distances for low explosives. Distances are
reported in meters for comparison to the distances calculated for high explosives. The results for deflagration
were obtained from the following equation, used for the derivation of the ATF Table of Distances for low
explosives (see Exhibit 10):
D = 8 W 1/3
where:
D = required, or affected, distance, ft
W = weight of material, Ib
(The table itself was not used because of the wide ranges of quantities presented.)
Distances calculated for low explosives are much smaller than those calculated for high explosives, as
would be expected from the equation used and from the fact that the primary hazard of low explosives is heat
rather than blast waves.
6.4 Findings
Analysis of the consequences of detonations of high explosives (Division 1.1) and blasting agents
(Division 1.5), based on the ATF Table of Distances for high explosives and blasting agents, indicates that the
effects of the blast from the detonation of such substances can extend for considerable distances. The distance
affected is dependent on the quantity of explosive (to the one-third power) and on the overpressure or damage
level of interest. For example, the detonation of about 440 pounds of a high explosive would produce an
overpressure of about 1.0 psi at a distance of 100 meters from the site of the explosion; it would take 2,000
pounds of the same explosive to produce an overpressure of 2.0 psi at the same distance. The results of the
analysis do not consider the probability of an accidental explosion. The magnitude of the consequences of
-------
-36-
EXHIBIT 11. AFFECTED DISTANCES FOR COMMERCIAL HIGH EXPLOSIVES
AND BLASTING AGENTS, DETERMINED USING THE ATF METHOD
(Peak Overpressure = 0 J psi - 3.0 psi)
Affected Distance for Various Overpressures (meters)
Quantity
(Ibs)
50
100
250
500
1,000
5,000
0.4 psi
97
119
160
206
251
431
0.5 psi
79
99
134
168
213
365
1.0 psi
48
61
83
104
131
224
2.0 psi
29
37
50
63
79
136
3.0 psi
. 22
28
38
48
61
104
-------
-37-
EXHIBIT 12. AFFECTED DISTANCES FOR LOW EXPLOSIVES, DETERMINED USING THE ATF
TABLE OF DISTANCES FOR LOW EXPLOSIVES
(Assuming Deflagration; Results Based on Thermal Radiation)
Quantity Affected
(Ibs) Distance
(meters)
50 9
100 11
250 15
1,000 24
5,000 42
10,000 66
100,000 113
-------
-38-
accidents appears to be the same for blasting agents and high explosives; however, blasting agents are much
less sensitive to shock and much less likely to explode than high explosives, and individual high explosives
differ in sensitivity.
The consequences of explosions of materials whose blast hazards are primarily due to projectiles have
not been analyzed. These hazards may be significant; however, the scaling law is based on damage from
overpressure waves and is not applicable to projectiles.
ATFs table of distances for high explosives and blasting agents applies to all high explosives and
blasting agents for setting storage distances, regardless of whether the explosive has a mass explosion hazard
or a projection hazard. Under the ATF rules, high explosives and blasting agents in quantities greater than
SO pounds must be stored more than 100 meters from inhabited buildings if unbarricaded. If there are
barricades, the distances are based on an overpressure of 1.0; 444 pounds could be stored 100 meters from a
barricaded inhabited building.
The primary hazard of low explosives (Division 1.3) is thermal radiation from a fire, although
projectiles from a deflagration of low explosives may also represent a significant hazard. Consequence analysis
based on ATFs Table of Distances for low explosives indicates that in the event of a deflagration of low
explosives, the effects would not extend over a very great distance. To reach the heat radiation level used to
develop the Table of Distances for low explosives at a distance of 100 meters from the site, about 70,000
pounds of low explosives would have to be involved in the deflagration.
-------
-39-
REFERENCES
ATF 1990. "ATF: Explosives Law and Regulations," P 5400.7, Bureau of Alcohol, Tobacco, and Firearms.
Department of the Treasury, June, 1990.
Aqualon. 1992. Personal communication from David Vanderveer, Safety & Environmental Mgr. at Aqualon
Co. September 18, 1992.
Brasie, W., Simpson, D. 1968. "Guidelines for Estimating Damage Explosion," AIChE Symposium on Loss
Prevention in the Process Industries, February 18-23, 1968.
Bureau of Mines. 1991. Branch of Industrial Minerals, U.S. Bureau of Mines, Mineral Industry Surveys,
"Apparent Consumption of Industrial Explosives and Blasting Agents in the United States, 1990."
Cantrell, R. 1991. Telephone conversation between Ray Cantrell, Bureau of Mines, and ICF Incorporated,
August 15, 1991.
Conlding, J.A. 1986. American Fireworks Manufacturing: An Industry in Transition. Fire Journal, Vol. 80,
No.5 (September 1986).
DODESB. 1992. Department of Defense Explosives Safety Board. Information about black powder
accidents/incidents provided by DODESB to EPA. 1992.
Dowling, T. 1989. Telephone conversation between Tom Dowling, Institute of Makers of Explosives,
Washington D.C. and ICF Incorporated. June 12, 1989.
EPA 1988. "Acute Hazardous Events Data Base (1988)," Prepared for the Office of Policy Analysis, Office
of Policy, Planning, and Evaluation, U.S. Environmental Protection Agency, Prepared by Industrial Economics,
Incorporated, Cambridge, Massachusetts, 31 August 1988.
Filler, W.S., Gott, R.W., Davis, V., Perkins, R.G. 1967. Panel Discussion, The Barricade Question: What
We Know and What We Don't Know." Minutes of the Ninth Explosives Safety Seminar, Naval Training
Center, San Diego, CA, 15-17 August 1967. 1 November 1967. AD #824-044.
Henry, M. 1989. Telephone conversation between Martin Henry, NFPA, and ICF Incorporated. May 1989.
Kirk-Othmer 1980. "Encyclopedia of Chemical Technology," Third edition, volume 9, John Wiley & Sons, Inc.
1980.
Kyte, G. 1986. Oklahoma Fireworks Plant Explosion Kills 21, Injures 5. Fire Journal, July 1986, pp 58-65,
73.
Lees, F. 1986. "Loss Prevention in the Process Industries: Hazard Identification, Volume I," Butterworth &
Co. (Publishers) Ltd, 1980 (Reprinted 1986).
McCune, L. 1989. Telephone conversation between Larry McCune, Bureau of Alcohol, Tobacco and Firearms,
Explosives Technology Branch and ICF Incorporated. June 13, 1989.
-------
-40-
REFERENCES (Continued)
Medard, L. 1989. "Accidental Explosions," Volume 1: "Physical and Chemical Properties." Chichester,
England: Ellis Horwood Limited, 1989.
Meyer, R. 1987. "Explosives," third, revised and extended edition, Sponsored by WASAGCHEMIE.
Essen/Germany. 1987.
MHIDAS Database. 1993. Major Hazard Incidents Data Service. Developed by Safety and Reliability
Consultants of AEA Technology and UK Health and Safety Executive. Part of Silver Platter CD-ROM
database.
Moran, P. 1989. Telephone conversation between Paul Moran, Department of Defense Explosives Safety
Board, Washington, D.C. and ICF Incorporated, November 1, 1989.
MSHA. 1992. Mine Safety and Health Administration, U.S. Department of Labor. Statistics for 1985-1992.
NFPA. "Fire Protection Handbook," 16th edition, National Fire Protection Association.
NASA 1988. "Ignition and Thermal Hazards of Selected Aerospace Fluids," RD-WSTF-0001, National
Aeronautics and Space Administration, Lyndon B. Johnson Space Center, White Sands Test Facility, Post
Office Drawer MM, Las Cruces, New Mexico, October 14, 1988.
NTIS 1971. "Engineering Design Handbook Explosives Series Properties of Explosives of Military Interest,"
AD-764 340 (AMCP 706-177), Headquarters, U.S. Army Materiel Command, Washington D.C., January 1971.
Peltier, M.A., L.R. Fletcher, and D.V. D'Andrea 1986. "Coal Mine Blasting Accidents," Engineering Health
and Safety in Coal Mining, Proceedings of the Society of Mining Engineers Symposium, March 2-6,1986, New
Orleans, Louisiana, A.W. Khair, Editor; pages 174-181.
Resources for the Future. 1992. Fatal Hazardous Materials Database.
Sawyer, R. 1989. Telephone conversation between Ray Sawyer, Department of Defense Explosives Safety
Board, Washington D.C. and ICF Incorporated. June 13> 1989.
Schultz, C. 1989. Telephone conversation between Charlie Schultz, Department of Transportation and ICF
Incorporated. July 5, 1989.
TNO. 1992. The Netherlands Organisation of Applied Scientific Research. Methods for the determination
of possible damage to people and objects resulting from releases of hazardous materials. 1st ed. CPR 16E.
The Hague: Directorate-General of Labour of the Ministry of Social Affairs and Employment, 1992.
U.N. 1986. United Nations. Recommendations on the Transport of Dangerous Goods, Tests and Criteria.
New York: United Nations. Sales No. E.85.VIII.2. ISBN 92-1-139021-4.
U.S. Air Force. 1983. "Explosives Safety Standards," Air Force Regulation 127-100. 20 May 1983.
U.S. DOT. 1988. "Hazard Analysis of Commercial Space Transportation," U.S. Department of Transportation,
Office of Commercial Space Transportation Licensing Programs Division. May 1988.
-------
-41-
REFERENCES (Continued)
U.S. Fire Administration, n.d. "Six Firefighter Fatalities in Construction Site Explosion, Kansas City, Missouri.
November 29,1988." Report 024 of the Major Fires Investigation Project conducted by TriData Corporation
under Contract EMW-88-C-2649 to the U.S. Fire Administration, Federal Emergency Management Agency.
Wallace, B. 1991. Telephone conversation between Brian Wallace, Atlas Powder Company, and 1CF
Incorporated, August 19, 1991.
White, L. 1989. Telephone conversation between Larry White, Bureau of Alcohol, Tobacco and Firearms and
ICF Incorporated. June 13, 1989.
Wilcox, I. 1991. Telephone conversation between Ian Wilcox, ICI Americas, and ICF Incorporated. August
15, 1991.
-------
-43-
Appendix A
Damage Criteria/Overpressure Data
This appendix contains background on blast overpressure damage criteria and its effects. Also
contained within this appendix are graphs illustrating the effect of varying overpressures on consequences.
A.I Damage Criteria
Blast waves from accidental explosions can cause damage to people and property by subjecting them
to transient crushing pressures and winds. Relatively simple concepts have been used to quite effectively
correlate blast wave properties with damage. The concept states that damage is primarily a function of either
the peak overpressure, the impulse or some combination of these factors. Peak overpressure is most
commonly used.
. Criteria for peak overpressures causing personal injury are given in Exhibit A-l, including injury from
direct blast effects and from flying glass. Guidelines for peak overpressures required to produce property
damage are presented in Exhibit A-2. These data are largely based on empirical observations. There seems
to be general agreement between sources on the data presented in Exhibit A-2; however, the same is not true
for the data presented in Exhibit A-l. According to Exhibit A-l, lung damage may occur at overpressures of
10 psi. According to other sources, threshold lung damage may not occur until peak overpressures reach 30
to 40 psi (Grelecki, Richmond 1968). The values for corresponding fatalities also vary. According to the
Exhibit, the lowest overpressure cited for fatality from direct blast effects is 14.5 psi. According to another
source, a 1% mortality rate is probable at 27.0 psi. Others cite values in the range of 35 to 120 psi as the
threshold overpressure value for fatalities, 20.5 to 180 psi for a 50% fatality rate, and 29 to 250 psi for a 99%
fatality rate (Grelecki, Richmond 1968, Lees 1980).
Exhibits A-3 and A-4 show how lethality and injury vary with both overpressure and the duration of
the blast wave. Exhibit A-5 shows similar data as a function of both pounds of TNT and range in feet.
Exhibits A-3 through A-5 present potential injury levels resulting from direct exposure to blast
overpressures in an open area. They do not take into account injuries resulting from property damage
presented in Exhibit A-2. For example, substantial injury and possible deaths may result from the shattering
of concrete walls at 2.0 to 3.0 psi. In planning with regard to injury to personnel, the following guidelines have
been recommended (Brasie and Simpson 1968):
Personnel in areas subject to overpressures greater than about 2 psi are
likely casualties from fragmentation or self-impact against objects.
Personnel beyond the under 1 psi overpressure range should be reasonably
safe inside a reinforced structure away from windows or, if outdoors, lying
on the ground.
-------
-44-
EXHIBIT A-l. HUMAN INJURY CRITERIA
(Includes Injury from Flying Glass and Direct Overpressure Effects)
Overpressure
(psi)
Injury
Comments
Source
0.6
1.0 - 2.0
1.5
2.0 - 3.0
2.4
2.8
3.0
3.4
4.0 - 5.0
5.8
6.3
Threshold for injury
from flying glass*
Threshold for skin
laceration from flying glass
Threshold for multiple
skin penetrations from
flying glass (bare skin)*
Based on studies
using sheep and dogs
Based on Army data
Based on studies
using sheep and dogs
Threshold for serious wounds Based on Army data
from flying glass
Threshold for eardrum rupture Conflicting data on
eardrum rupture
10% probability of eardrum
rupture
Overpressure will hurl
a person to the ground
1% eardrum rupture
Serious wounds from flying
glass near 50% probability
Threshold for body-wall
penetration from
flying glass (bare skin)*
50% probability of eardrum
rupture
Conflicting data on
eardrum rupture
One source suggested an
overpressure of 1.0 psi
for this effect
Not a serious lesion
Based on Army data
Based on studies
using sheep and dogs
Conflicting data on
eardrum rupture
(Fletcher, Richmond, and
Yelverton 1980)
(Lees 1980)
(Fletcher, Richmond, and
Yelvenon 1980)
(Lees 1980)
(Lees 1980)
(Lees 1980)
(Brasie and Simpson
1968)
(U.S. DOT 1988)
(Lees 1980)
(Fletcher, Richmond, and
Yelverton 1980)
(Lees 1980)
Interpretation of data tables presented in reference.
-------
-45-
EXHIBIT A-l. HUMAN INJURY CRITERIA
(Includes Injury from Flying Glass and Direct Overpressure Effects)
(continued)
Overpressure
(psi)
Injury
Comments
Source
7.0 - 8.0
10.0
14.5
16.0
17.5
20.5
25.5
27.0
Serious wounds from flying
glass near 100% probability
Threshold lung
hemorrhage
Fatality threshold for
direct blast effects
50% eardrum
rupture
10% probability of fatality
from direct blast effects
50% probability of fatality
from direct blast effects
90% probability of fatality
from direct blast effects
1% Mortality
29.0
99% probability of fatality
from direct blast effects
Based on Army data
Not a serious lesion
(applies to a blast of
long duration (over 50
msec); 20-30 psi required
for 3 msec duration waves)
Fatality primarily from
lung hemorrhage
Some of the ear injuries
would be severe
Conflicting data on
mortality
Conflicting data on
mortality
Conflicting data on
mortality
A high incidence of severe
lung injuries (applies to a
blast of long duration (over
50 msec); 60-70 psi required
for 3 msec duration waves)
Conflicting data on
mortality
(Lees 1980)
(U.S. DOT 1988)
(Lees 1980)
(U.S. DOT 1988)
(Lees 1980)
(Lees 1980)
(Lees 1980)
(U.S. DOT 1988)
(Lees 1980)
-------
-46-
EXHIBIT A-2. PROPERTY DAMAGE CRITERIA
Overpressure (psi)
Damage
Source(s)
0.03
0.04
0.10
0.15
0.30
0.4
0.5 - 1.0
0.7
1.0
1.0 - 2.0
1.3
Occasional breaking of large glass
windows already under strain
Loud noise (143dB). Sonic boom
glass failure
Breakage of windows, small, under strain
Typical pressure for glass failure
"Safe distance" (probability 0.95 no
serious damage beyond this value). Missile
limit. Some damage to house ceilings.
10% window glass broken.
Minor structural damage
Shattering of glass windows, occasional
damage to window frames. One source
reported glass failure at 0.147 psi
Minor damage to house structures
Partial demolition of houses, made
uninhabitable
Shattering of corrugated asbestos siding
Failure of corrugated aluminum/steel paneling
Failure of wood siding panels (standard
housing construction)
Steel frame of clad building slightly
distorted
(Lees 1980)
(Lees 1980)
(Lees 1980)
(Lees 1980)
(Lees 1980)
(McRae 1984, Lees 1980)
(Brasie and Simpson 1968,
Air Force 1983,
U.S. DOT 1988,
Lees 1980)
(Lees 1980)
(Lees 1980)
(Brasie and Simpson 1968,
Air Force 1983,
U.S. DOT 1988,
Lees 1980)
(Lees 1980)
2.0
Partial collapse of walls and roofs of houses (Lees 1980)
-------
-47-
EXHIBIT A-2. PROPERTY DAMAGE CRITERIA
(continued)
Overpressure (psi)
Damage
Source(s)
2.0 - 3.0
2.3
2.5
3.0
3.0 - 4.0
4.0
4.8
5.0
5.0 - 7.0
7.0
7.0 - 8.0
Shattering of non-reinforced concrete or cinder
block wall panels (1.5 psi according to
another source)
Lower limit of serious structural damage
50% destruction of brickwork of house
Steel frame building distorted and
pulled away from foundations
Collapse of self-framing steel panel buildings
Rupture of oil storage tanks
Snapping failure - wooden utility tanks
Cladding of light industrial buildings
ruptured
Failure of reinforced concrete structures
Snapping failure - wooden utility poles
Nearly complete destruction of houses
Loaded train wagons overturned
Shearing/flexure failure of brick wall panels
(8-12 inches thick, not reinforced)
Sides blown in of steel frame buildings
Overturning of loaded rail cars
(Brasie and Simpson 1968,
Air Force 1983,
U.S. DOT 1988,
Lees 1980)
(Lees 1980)
(Lees 1980)
(Lees 1980)
(Brasie and Simpson 1968,
U.S. DOT 1988,
Lees 1980)
(Lees 1980)
(McRae 1984)
(Brasie and Simpson 1968,
Lees 1980)
(Lees 1980)
(Lees 1980)
(Brasie and Simpson 1968,
Air Force 1983,
U.S. DOT 1988,
Lees 1980)
(Air Force 1983)
(Brasie and Simpson 1968,
U.S. DOT 1988)
-------
-48-
EXHIBIT A-2. PROPERTY DAMAGE CRITERIA
(continued)
Overpressure (psi)
Damage
Source(s)
9.0
10.0
30.0
Loaded train box-cars completely demolished
Probable total destruction of buildings
Steel towers blown down
(Lees 1980)
(Lees 1980)
(Brasie and Simpson 1968,
Air Force 1983)
88.0
Crater damage
(McRae 1984)
-------
-49-
EXHIBITA-3. LETHALITY CURVES
(For a 154 Ib Person in Free Stream Situations)
kPa
Maiimum
incidtnt
overpres-
sure
10.000
5,000
2.000
1.000
500
200
99% Lethality
50H Lethality
1H lethality
10
0.2 0.4 1.0 2 4 710 20 40 100 200 1000
Duration of positive incident overpressure, msec
5000
kPa
10*
10*
Maiimum
incident
overprev
sure
10*
10
0.1
1.000
100
10
EXHIBIT A-4. LETHALITY AND IN JURY CURVES
(For a 154 Ib Person in Free Stream Situations)
1H Lethality
Threshold lung damage
50% Eardrum rupture
95H Eardrum protection with muffs
95H Eardrum protection w/o muffs
0.1
0.01
4-
0.2 10 10 100 1000
Duration of positive incident overpressure, msec
10.000
Notes: msec = milliseconds
kPa = kilopascals = 6.89S psi
psi = pounds per square inch = 0.1450 kPa
Source: U.S. DOT 1988.
-------
-50-
EXHIBIT A-S. AIR-BLAST CRITERIA FOR PERSONNEL STANDING IN THE OPEN
1000
II 1 I 11II 1II I I I III
1% SERIOUS
INJURIES FROM
DISPLACEMENT
50% OF PERSONNEL
BLOWN DOWN
0 PERSONNEL
BLOWN DOWN
IX MORTALITY FROM
DIRECT OVERPRESSURE-^,
EFFECTS /
THRESHOLD LUNG
DAMAGE
EARDRUMS RUPTURED
1% OF EARDRUMS RUPTURED
r i \/\/\ t i
i i i i i i i i i i i i i i i i i < i I i i i i 1-
IOO
RANGE IN FEET
IOOO
- O.OOI
I r
10,000
Source: U.S. DOT 1988.
-------
-51-
The values presented in the Exhibits may be used to relate damage or injury criteria to safe distances
for emergency planning, using the scaling law described in the text.
Note that although glass is reported to shatter at an overpressure level of 0.5 psi (see Exhibit A-2),
there is evidence that human injury from flying glass is unlikely at this level. Lees (1980) cites a report stating
that risk of injury from flying window glass is negligible for an explosion that gives a peak overpressure of 0.6
psi or less. Lees also cites a report giving the skin laceration threshold for flying glass as 1-2 psi and the
serious wound threshold as 2-3 psi (see Exhibit A-l). Fletcher, Richmond, and Yelverton (1980) carried out
experiments on the effects on sheep and dogs of glass fragments from windows shattered by various blast
overpressures. Their results (briefly cited in Exhibit A-l) appear to agree reasonably well with the data
reported by Lees. Their data indicate that the threshold for injury from flying glass is about 0.6 psi. Multiple
injuries from skin penetration (i.e., ten or more) may occur for bare skin at overpressures of 1.5 psi or higher.
Rying glass may penetrate the body-wall at 5.8 psi or higher if the skin is bare. The authors point out that
the skin and body-wall thickness of men and sheep are approximately the same (Fletcher, Richmond, and
Yelverton 1980); therefore, the data reported should be roughly applicable to humans.
A.2 Effects of Varying Overpressures
To determine the effect on distance of choosing different overpressure levels, analysis was carried out
at four different overpressures for high explosives. ATF methodology was used, and calculations were carried
out for a quantity of 100 pounds at overpressures of 0.4, 0.5, 1.0, 2.0, and 3.0 psi.
Exhibit A-6 presents a graph of overpressure versus distance for commercial high explosives. The
exhibits indicate that higher overpressures occur at lower distances (i.e., closer to the explosion site), and
overpressure falls off as distance increases. This is as expected from the scaling law. Distance and K are
directly proportional in the scaling law. Therefore, as K increases, so does the affected distance. However,
as K increases, overpressure decreases (as is indicated by Exhibit 5 in the main text of this study).
The decrease in distance affected with increase in overpressure appears to occur more rapidly at low.
overpressures than at higher overpressures; note that the distance decreases much more sharply between 0.4
and 0.5 psi than between 1.0 and 2.0 psi or 2.0 and 3.0 psi (see Exhibit A-6).
-------
-52-
EXHIBIT A-6
OVERPRESSURE VERSUS DISTANCE FOR 100 POUNDS OF HIGH EXPLOSIVES
i:tO
0.4
0.8
1.2 1.6 2.0
Overpressure (psi)
2.4
2.8
-------
-53-
REFERENCES
Brasie, W., Simpson. D. 1968. "Guidelines For Estimating Damage Explosion," AICHE Symposium on Loss
Prevention in the Process Industries, February 18-23, 1968.
Fletcher, E.R., Richmond, D.R., Yelverton, J.T. 1980 (May 30). "Glass Fragment Hazard from Windows
Broken by Airblast." Washington, D.C.: Defense Nuclear Agency. Report Number DNA 5593T.
Lees, P.P. 1980. Loss Prevention in the Process Industries, Vol. 1. London: Butterworths.
U.S. Air Force. 1983. "Explosives Safety Standards," AF Regulation 127-100, 20 May 1983.
U.S. DOT. 1988. "Hazard Analysis of Commercial Space Transportation," Office of Commercial Space
Transportation Licensing Programs Division, U.S. Department of Transportation. May 1988.
NFPA. "Fire Protection Handbook," 16th edition. National Fire Protection Association (NFPA).
McRae, T. 1984. "The Effects of Large Scale LNG/Water RPT Explosions," McRae, T.G., et.al., Lawrence
Livermore National Laboratory. April 27, 1984.
Grelecki, C. "Fundamentals of Fire and Explosion Hazards Evaluation," C. Grelecki, Hazards Research Corp.,
AICHE Today Series.
Richmond, D., Damon, E.G., Fletcher, E.R., Bowen, I.G., White, C.S. 1968. "The Relationship Between
Selected Blast-Wave Parameters and the Response of Mammals Exposed to Air Blast." Ann. N.Y. Acad. Sci.,
1968.
Eichler, T., Napadensky, H. 1977. "Accidental Vapor Phase Explosions on Transportation Routes Near
Nuclear Plants," IIT Research Institute, Prepared for Argonne National Laboratory. April 1977.
Wiekema, B. J. 1984. "Vapor Cloud Explosions - An Analysis Based on Accidents," Journal of Hazardous
Materials, 8 (1984).
-------
-55-
Appendix B
Other Methods of Consequence Analysis Based on the Scaling Law
EPA selected two commonly used methodologies for estimating distances affected by explosions, both
based on the scaling law, for comparison with the ATF standards. These methods are:
ARCHIE: A computerized hazard assessment model developed for DOT, FEMA, and
the EPA; and
K-Factor. Use of the scaling law, as described in the text, with TNT equivalence.
The methods and the results obtained for examples of high explosives and blasting agents are described
below. These methods are not appropriate for substances that deflagrate rather than detonate; therefore, no
results are presented for low explosives.
B.I ARCHIE
The Automated Resource for Chemical Hazard Incident Evaluation (ARCHIE) computer program
was developed for FEMA, DOT, and EPA to provide emergency planning personnel with the tools necessary
to assess the vapor dispersion, fire, and explosion impacts associated with hazardous materials accidents.
Contained within the program are several methods for evaluating the consequences of an explosion, depending
upon the type of material involved. Model "K", the Condensed-Phase Explosion Model, was used for this
analysis. The model is based upon the scaling law discussed earlier and is appropriate for use with commercial
explosives. The Condensed-Phase Explosion Model differs from the scaling law in that the TNT-equivalent
weight is calculated as a function of heat of combustion and empirically derived yield factors, rather than as
a function of an empirically derived TNT equivalence. (Heat of combustion represents the total combustion
energy of a substance and is likely to be higher than the heat of explosion.) In addition, the curve relating .
K to overpressure has been approximated by a logarithmic expression. The ARCHIE manual presents the
following equation for determining distance:
X = (MTNT1/3)(exp (3.5031 - 0.72411o(Op) + 0.0398(lnOp)2))
where: X = Required distance, ft
= Adjusted equivalent mass of TNT, Ib
Mchem = Original mass of explosive material
AHC = Lower heat of combustion of explosive material, kcal/kg
Yf = Yield factor
On = Peak overpressure, psi
1 155 = Heat of explosion of TNT, kcal/kg
The ARCHIE manual recommends using a yield factor of one for condensed-phase explosions (i.e.,
it is assumed that all of the energy is available for the explosion).
-------
-56-
The following expression, from the above equation, gives an approximation of the K-factor at a given
overpressure:
exp (3.5031 - 0.7241 ln(Op) + 0.0398(lnOp)2)
The primary limitation of this method is that it attempts to estimate TNT equivalence, an empirically
derived parameter, with heats of combustion using yield factors, another empirically derived parameter. Note
also that the heat of explosion given for TNT, 1155 kcal/kg, is higher than the heat of explosion reported
elsewhere, 1011 to 1090 kcal/kg (Kirk-Othmer 1980, Meyer 1987). In addition, the equation approximating
the curve relating overpressure to K is imprecise at the endpoints.
While the ARCHIE manual presents the above equation for estimating the K-factor, it appears, based
on an analysis of the model results for several chemicals, that the model actually uses selected values of the
K-factor corresponding to specific consequences rather than the equation for the K-factor and, therefore, the
results produced by the computer model are somewhat different from the results obtained from the equation.
The model results are presented as distances (or ranges of distances) to 14 damage levels. Apparently the K-
factor for the overpressure related to each of these damage levels is included in the model.
Exhibit B-l presents a sample printout of ARCHIE model results for 1,000 pounds of TNT (heat of
combustion 6,512 BTU/lb). The expected damage levels presumably correspond to the overpressures and
expected damage presented in the ARCHIE manual, i.e., "Windows usually shattered; some frame damage"
corresponds to overpressures of 0.5 psi - 1.0 psi. An overpressure of 0.5 psi would be reached at 368 feet from
the explosion; an overpressure of 1.0 psi would be reached at 212 feet. (See Appendix A for more information
on the effects of various overpressure levels.)
B.2 "K-Factor" Methodology
The K-factor methodology is equivalent to the straight usage of the scaling law. The scaling law is
represented by the following equation:
D = (K)(W1/3)
where:
D = Required, or affected, distance, ft
K = Protection factor (see Exhibit 1)
W = TNT equivalent weight, Ib.
W in the above equation is equal to the TNT equivalent weight of the material of concern as follows:
W = [M^x %TNT]
where:
Mchem = Total mass of chemical
%TNT = TNT equivalent; the ratio of the quantity of TNT to the quantity of explosive
material that will have an equivalent effect.
The %TNT, or TNT equivalent, varies and is derived for each explosive material from empirical tests and
actual accidents involving TNT and other explosive materials.
-------
-57-
EXHIBIT B-l. SAMPLE PRINTOUT OF ARCHIE MODELING RESULTS FOR 1,000 POUNDS OF TNT
(HEAT OF COMBUSTION 6^12 BTU/LB.)
CONDENSED-PHASE EXPLOSION EFFECTS
DISTANCE FROM EXPLOSION EXPECTED DAMAGE
(feet)
11751 Occasional breakage of large windows under stress.
1656 Some damage to home ceilings; 10% window breakage.
618 - 1072 Windows usually shattered; some frame damage.
618 Partial demolition of homes; made uninhabitable.
159 - 618 Range serious/slight injuries from flying glass/object.
373 Partial collapse of home walls/roofs.
285 - 373 Non-reinforced concrete/cinder block walls shattered.
127 - 330 Range 90-1% eardrum rupture among exposed population.
321 50% destruction of home brickwork.
238 - 285 Frameless steel panel buildings ruined.
208 Wooden utility poles snapped.
171 - 208 Nearly complete destruction of houses.
141 Probable total building destruction.
85 - 117 Range for 99-1% fatalities among exposed populations
due to direct blast effects.
Note: The explosion is assumed to take place on or near the ground.
Note that the distances provided by the ARCHIE model are in feet. Since the Agency has used meters as the
unit of measure in its evaluations, the values here should be divided by 3.25 for purposes of comparison.
-------
-58-
The parameter K in the scaling law is referred to as the protection factor. It is a function of the peak
overpressure and is independent of the material of concern. The value of K at a given overpressure is
determined from the graph shown in Exhibit 5 of the main text.
B.3 Results
Several high explosives and ammonium nitrate fuel oil (ANFO), the most commonly used blasting
agent, were analyzed to estimate the distance at which various overpressures would be reached. The high
explosives included in this analysis are nitroglycerin, which is used in some dynamite formulations and other
explosive formulations; pentaerythritol tetranitrate, which may be used as the base charge in blasting caps, the
core explosive in commercial detonating cord, or in other formulations; and lead azide, which is a primary
explosive used only in small quantities as an initiator for other explosives because of its sensitivity. Exhibit
B-2 presents the factors used and the results of the analysis using the ARCHIE equation, the ARCHIE model,
and the K-factor method at an overpressure of 0.5 psi, and the ATF table of distances for overpressure levels
of 0.4 psi (the level used for development of the table) and 0.5 psi. Exhibits B-3, B-4, and B-5 show the results
at overpressures of 1.0 psi, 2.0 psi, and 3.0 psi, respectively. Distances for 0.5, 1.0, 2.0, and 3.0 psi are not
included in the ATF Table of Distances; the ATF distances at these overpressures were estimated from the
K-factor for purposes of comparison. Distances from the ATF Table of Distances (based on 0.4 psi) are also
included in these exhibits for comparison. The analyses were carried out for 50, 100, 250, 1,000, and 5,000
pounds of each high explosive material. The exhibits show the heat of combustion used for ARCHIE
modeling and the TNT equivalent and K value used in carrying out the distance calculations using the K-factor
method. A yield factor of one was used for ARCHIE modeling in all cases, as recommended for condensed-
phase explosions. Affected distance in meters based on each of the methods is shown.
TNT equivalents are not taken into account in using the Table of Distances (i.e., all materials are
assumed to have the same explosive potential); however, the TNT equivalent is taken into account in the K-
factor method. Therefore, the Table of Distances gives greater distances than the K-factor method for
materials with TNT equivalents less than one, while the K-factor method gives greater distances than the Table
of Distances for materials with TNT equivalents greater than one. (It should be noted that the TNT
equivalent is an experimentally-derived quantity, not a physical property, and values may vary, depending on
the test used.)
The equation presented in the ARCHIE manual for calculating distance (see Section C.1) does not
give results that are consistently higher or lower than the K-factor method or the Table of Distances. This
equation is based on a calculated K-factor, which may be significantly different at a given overpressure from
the K-factor value read from the graph, and a TNT equivalent weight calculated from the heat of combustion
of the material. For an overpressure of 0.5 psi, the K-factor calculated using the ARCHIE equation is 56,
compared to about 70 from the graph. At 1.0 psi, the equation gives 33, while the graph shows 43. At 2.0
psi, a K-factor of 20 is calculated from the equation, compared to 26 from the graph, and at 3.0 the calculated
value is 16 and graph value is 20. The TNT equivalent used in the K-factor method is based on empirical tests
comparing the substance to TNT. In all the cases considered, the TNT equivalent calculated from the heat
of combustion using the ARCHIE method is greater than TNT equivalents reported in the literature. Because
both the calculated K-factor and the calculated TNT equivalent may affect the results of the ARCHIE
equation, but in opposite ways (the lower K-factor would tend to lead to lower distances and the higher TNT
equivalents to higher distances), no clear-cut comparisons can be made between the ARCHIE equation and
the other methodologies.
The ARCHIE model yields distances quite comparable to those calculated by the K-factor
method. The ARCHIE model uses heat of combustion to calculate an approximate TNT equivalent, while
TNT equivalent values derived from test results are used for the K-factor method. The K-factors used by the
ARCHIE model (as estimated using the scaling law) are very close to the values read from the graph for use
-------
EXHIBIT B-2. AFFECTED DISTANCES FOR COMMERCIAL HIGH EXPLOSIVES
AND BLASTING AGENTS, DETERMINED USING FOUR DIFFERENT METHODS
(Peak Overpressure = 0.5 psi)
Explosive
Material
Nitroglycerin
Pentaerv-thHtol
Tetranitrate
(PEIN)
lead Azide
(Primry
explosive)
TNf
Aaanniuft
Nitrate -
Fuel Oil
(ANFO)
(Blasting
Agent)
Sources of TNT
Estimated for
« The value of K
Heat of
Amount Combustion TNT Overpressure
(Its) (kcal/kg) Equivalent (psi) K Value
50 162S 1.52 0.5 70
too
250
500
1000
5000
SO 1960 1.52 0.5 70
100
250
500
1000
5000
50 630 0.40 0.5 70
100
250
500
1000
5000
50 3620 1.00 0.5 70
100
250
500
1000
5000
50 626 *» 0.77 0.5 70
100
250
500
1000
5000
equivalents: NTIS 1971. NASA 1988. Kirk-Othmer 1980.
purposes of comparison.
corresponding to an overpressure of 0.4 psi is 80 - 90.
Affected Distance ()
ARCHIE
Model
88
111
151
190
239
409
94
118
160
202
254
435
67
85
115
138
183
312
120
152
206
248
327
559
67
85
115
138
182
312
ARCHIE
Equation
70
89
120
152
191
327
75
94
128
161
203
348
51
65
88
111
139
238
92
116
157
198
249
427
51
65
88
110
139
238
K-Factor
90
114
155
195
245
419
90
114
154
195
245
419
SB
73
99
125
157
269
79
99
134
169
213
365
72
91
123
155
196
334
ATF (0.5 psi)" ATF
79
99
134
168
213
365
79
99
134
169
213
365
79
99
134
169
213
365
79
99
134
169
213
365
79
99
134
169
213
365
(0.4 psi)*
97
119
160
206
251
431
97
119
160
206
251
431
97
119
160
206
251
431
97
119
160
206
251
431
97
119
160
206
251
431
»» Heat of combustion for pure ammonium nitrate (Kirk-Othner 1980).
-------
EXHIBIT B-3. AFFECTED DISTANCES FOR COMMERCIAL HIGH EXPLOSIVES
AND BLASTING AGENTS, DETERMINED USING FOUR DIFFERENT METHODS
(Peak Overpressure = 1.0 psi)
Explosive
Material
Nitroolycerin
Pentaerythritol
letranitrate
(PETN)
Lead Atide
(Primary
explosive)
INT
Anmoniin
Nitrate
Fuel Oil
(AMfO)
(lasting
Agent)
Heat of
Amount Combustion TNT Overpressure
(Ibs) (kcal/kg) Equivalent * (psi) 1C Value
SO 162S 1.52 1.0 43
100
2SO
500
1000
5000
SO I960 1.52 1.0 43
100
250
500
1000
5000
50 630 0.40 1.0 43
100
250
500
1000
5000
50 3620 1.00 1.0 43
100
250
500
1000
5000
50 626 «« 0.77 1.0 43
100
250
500
1000
5000
Affected Distance (en)
ARCHIE
Model
54
68
93
117
147
251
58
73
98
124
156
267
39
49
66
85
105
180
70
88
119
152
188
322
39
49
66
85
105
180
ARCHIE
Equation
42
53
71
90
113
194
44
56
76
96
121
207
30
38
52
66
83
141
55
69
93
118
148
253
30
38
52
66
83
141
K- Factor
56
70
95
120
151
258
55
70
95
120
151
257
36
45
61
77
97
165
48
61
83
104
131
224
44
56
76
95
120
205
AIF (1.0 psi)** AIF
48
61
83
168
131
224
48
61
83
104
131
224
48
61
83
104
131
224
48
61
83
104
131
224
48
61
83
104
131
224
(0.4 psi)*
97
119
160
206
251
431
97
119
160
206
251
431
97
119
160
206
251
431
97
119
160
206
251
431
97
119
160
206
251
431
Sources of TNT equivalents: NIIS 1971, NASA 1988, Kirk-Othawr 1980.
Estimated for purposes of comparison.
* The value of K corresponding to an overpressure of 0.4 psi is 80 - 90.
« Heat of combustion for pure aawoniu* nitrate (Kirk-Othmer 1980).
-------
EXHIBIT B-4. AFFECTED DISTANCES FOR COMMERCIAL HIGH EXPLOSIVES
AND BLASTING AGENTS, DETERMINED USING FOUR DIFFERENT METHODS
(Peak Overpressure = 2.0 psi)
Explosive
Material
Nitroglycerin
Pent aery thritol
1e t rani t rate
(PEIN)
lead Alide
(Priaary
explosive)
INI
Amoniua
Nitrate
fuel Oil
(ANFO)
(lading
Agent)
Heat of
Amount Coafcustion TNT Overpressuri
(lb») (kcal/kg) Equivalent * (psi)
50 1625 1.52 2.0
100
250
500
1000
5000
50 1960 1.52 2.0
100
250
500
1000
5000
SO 630 0.40 2.0
100
250
500
1000
5000
SO 3620 1.00 2.0
100
250
500
1000
5000
50 626 *» 0.77 2.0
100
250
500
1000
5000
K Value ARCHIE
Model
26 33
41
56
70
89
152
26 35
44
60
75
95
162
26 23
30
40
51
64
109
26 42
53
72
92
114
195
26 23
30
41
51
63
109
Affected Distance ()
ARCHIE
Equation
26
32
44
56
70
120
27
35
47
59
75
127
19
24
32
41
51
87
34
42
58
73
91
156
19
24
32
40
51
87
K-factor
34
42
57
72
91
156
34
42
57
72
91
156
22
27
37
46
58
100
29
37
SO
63
79
136
27
34
46
58
73
124
ATF (1.0 psi)" ATF
29
37
50
168
79
136
29
37
SO
63
79
136
29
37
SO
63
79
136
29
37
SO
63
79
136
29
37
50
63
79
136
(0.4 p«i)»
97
119
160
206
251
431
97
119
160
206
251
431
97
119
160
206
251
431
97
119
160
206
251
431
97
119
160
206
251
431
Sources of TNT equivalents; MIIS 1971, NASA 1988, Kirk Othroer 1980.
Estinated for purposes of comparison.
* The value of K corresponding to an overpressure of 0.4 psi is 80 - 90.
«» Heat of combustion for pure annoniua) nitrate (Kirk-Othner 1980).
Draft November 17,1992
-------
EXHIBIT B-5. AFFECTED DISTANCES FOR COMMERCIAL HIGH EXPLOSIVES
AND BLASTING AGENTS, DETERMINED USING FOUR DIFFERENT METHODS
(Peak Overpressure = 3.0 psi)
Explosive
Material
Nitroglycerin
Pent aery thri tot
Tetranitrata
(PETN)
lead Azide
(Primary
explosive)
INT
Aononiu*
Nitrate -
fuel Oil
(ANfO)
(lasting
Agent)
Heat of
Amount Combustion TNT Overpressure
(Its) (kcal/kg) Equivalent * (psi) K Value
50 1625 1.52 3.0 20
100
250
500
1000
5000
50 1960 1.52 3.0 20
100
250
500
1000
5000
50 630 0.40 3.0 20
100
250
500
1000
5000
50 3620 1.00 3.0 20
100
250
500
1000
5000
50 626 «» 0.77 3.0 20
100
250
500
1000
5000
Affected Distance ()
ARCHIE
Model
25
32
43
54
68
117
27
34
46
58
73
124
18
23
31
40
48
83 .
32
40
55
71
87
148
18
23
30
39
48
83
ARCHIE
Equation
20
25
34
43
54
92
21
27
36
45
57
98
14
IB
25
31
39
67
26
33
44
56
70
120
14
18
25
31
39
67
1C -factor
26
33
44
56
70
120
26
33
44
56
70
120
17
21
28
36
45
77
22
28
38
48
61
104
21
26
35
44
56
96
Alf (1.0 psi)** Alf
22
28
38
168
61
104
22
28
38
48
61
104
22
28
38
48
61
104
22
28
38
48
61
104
22
28
38
48
61
104
(0.4 psi)*
97
119
160
206
251
431
97
119
160
206
251
431
97
119
160
206
251
431
97
119
160
206
251
431
97
119
160
206
251
431
Sources of TNT equivalents: NIIS 1971, NASA 1988, Kirk-Othner 1980.
Estimated for purposes of comparison.
* The value of 1C corresponding to an overpressure of 0.4 psi is 80-90.
»» Heat of combustion for pure ammoniua nitrate (Kirk Otheer 1980).
a\
N>
-------
-63-
in the K-factor method. The ARCHIE model gives greater distances than the ATF Table of Distances in cases
where the calculated TNT equivalent is greater than one (i.e., where the heat of combustion of the explosive
is greater than 1155 kcal/kg, the heat of explosion of TNT cited by ARCHIE), and smaller distances where
the TNT equivalent is less than one (i.e., where the heat of combustion is less than 1155 kcal/kg).
In general, the ATF Table of Distances, the ARCHIE model, the ARCHIE equation, and the K-factor
method give results for high explosives and blasting agents that are of the same order of magnitude. As all
three methods are based on the scaling law, similar results would be expected. In some cases, however, the
results differ significantly, apparently because of the approximations made.
It is not clear that any of the three methods can be considered to have more validity or to give more
conservative results than the others. However, it should be noted that the ATF Table of Distances is widely
used and accepted, and that the Institute of Makers of Explosives and the Department of Defense favor the
ATF method (Sawyer 1989, Dowling 1989).
-------
-65-
Appendix C
Requirements for Explosives Magazines
This appendix provides additional details on ATF requirements for explosives magazines (discussed
in Section 3.4 of the text).
CM General Requirements
Outdoor magazines must be located according to the distances specified in the table of distances for
storage of explosive materials. The ground around an outdoor magazine must be sloped to provide for
drainage away from the magazine or sufficient drainage must otherwise be provided.
Where lighting is necessary, it should conform to the following guidelines as put forth in ATF-
Explosives Law and Regulations. 1990:
Electric lighting must comply with National Electrical Code standards for conditions present.
Electric switches must be placed outside the magazine and comply with the National Electric
Code.
Battery-activated safety lighting is permitted in explosives magazines of all types.
C.2 Requirements by Type of Magazine
Type 1 Magazines are permanent magazines for storage of high explosives; other classes of explosives
also may be stored in type 1 magazines. A type 1 magazine may be a building, an igloo or "Army-type
structure," a tunnel, or a dugout. The following requirements apply to type 1 magazines:
1) Must be bullet-resistant, fire-resistant, weather-resistant, theft-resistant, and ventilated.
2) Masonry wall construction can be no less than six inches in thickness and must have hollows packed
with weak concrete (one part cement, eight parts sand, and sufficient water for packing) or coarse,
dry sand. Interior walls must be nonsparking or covered with nonsparking material.
3) Metal wall construction must consist of at least 14-gauge steel or aluminum sheets fastened to a metal
framework. Either brick, solid cement blocks, or hardwood not less than four inches thick or a six-
inch sand fill between inner and outer walls must be used in conjunction with the metal wall
construction. Interior walls must be made of or covered with nonsparking material.
4) Wooden frame wall construction must be covered with at least 26-gauge steel or aluminum. An inner
wall (nonsparking) must be separated by at least six inches from the outer wall; this six-inch space
must be filled with weak concrete or coarse, dry sand.
5) Floors must be constructed of or covered by a nonsparking material. Floors must be sufficiently
strong to support the weight of the maximum amount of explosives that could be stored in that
magazine. Iron nails in floor and walls which would come in contact with explosives must be covered
by nonsparking material or otherwise countersunk or blind nailed.
6) Foundations must be constructed of brick, concrete, cement, block, stone, or wood posts. When posts
are used, the space under the building must be enclosed with metal.
7) External roofing for type 1 magazines must be 26-gauge iron or aluminum affixed to 7/8-inch
sheathing, except for buildings with fabricated metal roofs. If it is possible for a bullet to be fired
through the roof, striking the explosives, protection must be provided by one of the following:
-------
-66-
A sand tray lined with paper, plastic, or some other nonporous material and
filled with at least four inches of coarse, dry sand covering the entire ceiling
area; or
A metal roof constructed of 3/16 inch plate steel lined with four inches of
hardwood, where for each additional 1/16 inch plate steel, hardwood may be
decreased by one inch.
8) Doors are to be made of not less than 1/4 inch plate steel and lined with at least two inches of
hardwood. Hinges and hasps must be attached by welding, riveting, or bolting, and must be not be
removable when the door is closed and locked.
9) Each door must be equipped with two mortise locks, or two padlocks fastened in separate hasps and
staples, or a combination of mortise lock and padlock, or a mortise lock requiring two keys to open,
or a three-point lock. Padlocks themselves must have a 3/8 inch case-hardened shackle and have a
1/4 inch steel hood to prevent sawing or lever action on the lock, hasps, or staple. Lock requirements
need not apply to doors adequately secured by a bolt, lock, or bar on the inside that can not be
actuated from the outside.
10) Ventilation is to be provided to prevent dampness and heating of stored explosives. Clearance of roof
and foundation ventilators from magazine contents must be maintained by a lattice lining or some
sufficient equivalent: air circulation must be maintained. Ventilation openings must be screened to
prevent entrance of sparks; side and foundation ventilation openings must be shielded or offset for
bullet-resistance.
Type 2 Magazines are mobile and portable indoor and outdoor magazines for storage of high
explosives; other classes of explosives also may be stored in type 2 magazines. A type 2 magazine may be a
box, trailer, semitrailer, or other mobile facility. The following are requirements for type 2 magazines:
1) Outdoor Magazines.
a) Must be bullet-resistant, fire-resistant, weather-resistant, theft-resistant, and ventilated.
b) Must be supported to prevent direct contact with the ground.
c) Exteriors and doors must be constructed of at least 1/4 inch steel and at least two inches of
hardwood. Top-opening magazines must have lids that overlap the sides by one inch or have
water-resistant seals.
d) Hinges and hasps are to be attached to doors by welding, riveting, or bolting such that they
can not be removed when the door is locked and closed.
e) Each door is to be equipped with two mortise locks, or two padlocks fastened in separate
hasps and staples, or a combination of a mortise lock and padlock, or a mortise lock that
requires two keys to open, or a three-point lock.
2) Indoor Magazines.
a) Must be fire-resistant and theft-resistant; need not be bullet-resistant and weather-resistant
if located in a bullet-resistant and weather-resistant building.
b) May not be located in a residence or dwelling.
-------
-67-
c) Indoor, storage of high explosives must not exceed a total of 50 pounds in one or more
magazines. Detonators must be stored in a separate magazine; the total quantity of
detonators must not exceed 5,000.
d) Sides, bottoms, and covers (or doors) shall be either two inch hardwood (with braced corners)
and covered with at least 26-gauge sheet metal or at least 12 gauge metal lined inside with
nonsparking material.
e) Hinges and hasps must be attached by welding, riveting, or bolting such that they can not be
removed when the door is closed and locked.
f) Each door is to be equipped with two mortise locks, or two padlocks fastened in separate
hasps and staples, or a combination of a mortise lock and padlock, or a mortise lock that
requires two keys to open, or a three-point lock.
Type 3 Magazines are "day-boxes" or other portable outdoor magazines for the temporary storage of
high explosives while attended; other classes of explosives also may be stored in type 3 magazines. The
following requirements apply:
1) Must be fire-resistant, weather-resistant, and theft-resistant.
2) Must be constructed of at least 12-gauge steel and lined with 1/2 inch plywood or hardboard.
3) Doors must overlap openings by at least one inch.
4) At least one steel padlock with a 3/8 inch case-hardened shackle must be used. Explosives may not
be left unattended and if they are to be unattended, must be moved to a type 1 or type 2 magazine.
Type 4 Magazines are magazines for the storage of low explosives; blasting agents also may be stored
in type 4 magazines. A type 4 magazine may be a building, igloo or "Army-type structure," tunnel, dugout,
box, trailer, semitrailer, or other mobile facility. Type 4 magazines must meet the following specifications:
1) Outdoor Magazines.
a) Must be fire-resistant, weather-resistant, and theft-resistant.
b) When unattended, vehicular magazines must be immobilized by removal of the wheels or use
of a locking device.
c) Must be constructed of masonry, metal-covered wood, fabricated metal, or some combination
of these and built on a brick, concrete, cement block, stone, or metal/wood post foundation.
When built on piers or posts, the space under the building must be enclosed with fire-
resistant material. The interior must be nonsparking.
d) Doors must be metal or solid wood covered with metal.
e) Hinges and hasps must be attached to doors by welding, riveting, or bolting such that they
can not be removed when doors are closed and locked.
f) Each door is to be equipped with two mortise locks, or two padlocks fastened in separate
hasps and staples, or a combination of a mortise lock and padlock, or a mortise lock that
requires two keys to open, or a three-point lock.
-------
-68-
2) Indoor Magazines.
a) Must be fire-resistant and theft-resistant; need not be weather-resistant if located in a
weather-resistant building.
b) May not be located in a residence or dwelling.
c) More than one magazine may be located in the same building if the total quantity of
explosives does not exceed 50 pounds. Detonators that will not mass detonate must be stored
in a separate magazine; the total number of electric detonators must not exceed 5,000.
d) Must be constructed of masonry, metal-covered wood, fabricated metal, or a combination of
these materials; interiors must be non-sparking.
e) Hinges and hasps must be attached to doors by welding, riveting, or bolting such that they
can not be removed when door is closed and locked.
f) Each door is to be equipped with two mortise locks, or two padlocks fastened in separate
hasps and staples, or a combination of a mortise lock and padlock, or a mortise lock that
requires two keys to open, or a three-point lock.
Type 5 Magazines are magazines for the storage of blasting agents. A type 5 magazine may be a
building, igloo or "Army-type structure," tunnel, dugout, bin, box, trailer, semitrailer, or other mobile facility.
The following requirements apply to type 5 magazines:
1) Outdoor Magazines.
a) Must be weather-resistant and theft-resistant.
b) When unattended, vehicular magazines must be immobilized by removal of the wheels or use
of a locking device.
c) Doors and covers shall be constructed of either solid wood or metal.
d) Hinges and hasps must be attached to doors by welding, riveting, or bolting such that they
can not be removed when doors are closed and locked.
e) Each door is to be equipped with two mortise locks, or two padlocks fastened in separate
hasps and staples, or a combination of a mortise lock and padlock, or a mortise lock that
requires two keys to open, or a three-point lock.
2) Indoor Magazines.
a) Must be theft-resistant.
b) Doors shall be constructed of wood or metal.
c) Hinges and hasps must be attached to doors by welding, riveting, or bolting such that they
can not be removed when doors are closed and locked.
-------
-69-
d) Each door is to be equipped with two mortise locks, or two padlocks fastened in separate
hasps and staples, or a combination of a mortise lock and padlock, or a mortise lock that
requires two keys to open, or a three-point lock.
-------
-70-
REFERENCE
ATF 1990. "ATF: Explosives Law and Regulations," P 5400.7, Bureau of Alcohol, Tobacco, and Firearms,
Department of the Treasury, June, 1990.
-------
-71-
Appendix D
Tests for Explosives in Transportation
This appendix briefly describes the tests used for materials in transportation to determine whether the
explosives hazard class is applicable, and tests for assignment of hazard divisions for explosive materials.
D.I Tests for Explosive Properties
Materials that are not designed to have an explosive effect, and that are tested and found to be not
too sensitive for transport, are further tested to determine whether they have explosive properties. If positive
results are obtained in one of the shock tests or thermal tests described below, the substance is considered to
have explosive properties. The same tests are used, with less stringent conditions, to determine whether a
substance is too insensitive to be eligible for Class 1.
Shock Tests
BAM 50/60 steel tube test. This test measures sensitivity to shock under confinement by
determining whether the test material propagates a detonation. The material to be tested is
packed into a steel tube. A booster (50 grams of RDX/wax) is placed in the upper end of
the tube and a detonator is inserted into the booster. The detonator is fired and propagation
of detonation is noted if:
the tube is completely fragmented, or
the tube is fragmented at both its upper and lower ends, or
a length of tube less than 21 centimeters from the upper end is fragmented and the
substance has reacted completely with a velocity higher than the velocity of sound in
the substance (method of determining the velocity is not described).
TNO 50/70 steel tube test. This test also uses a steel tube and a booster to determine
detonability of a substance. A velocity probe also is inserted in the tube. A determination
is made as to whether or not the material has detonated based on fragmentation of the tube.
If the tube is fragmented over its entire length, the substance is considered to have detonated.
If fragmentation is not complete, the substance is still considered to have detonated if velocity
measurements meet the following conditions:
the total decrease in resistance measured by the velocity probe should correspond
with the total original length of resistance wire in the tube,
the propagation velocity should be constant in the second part of the tube,
the constant velocity should be higher than the sound velocity in the substance.
Gap test for solids and liquids. This test is used to measure shock sensitivity and detonation
propagation. The substance is packed in a steel tube with a velocity of detonation probe.
A booster is placed at the end of the tube and initiated with a detonator. The substance is
considered to propagate detonation if two of the following criteria are met:
stable propagation velocity greater than the velocity of sound in the substance is
observed,
-------
-72-
a hole is punched through the witness plate,
the sample tube is fragmented along its entire length.
Second gap test for solids and liquids. This test subjects the substance to an intense shock
to determine the possibility of propagation of detonation in a solid and occurrence of low
velocity detonation in a liquid. The substance is packed in a steel tube with a booster at the
end (a second booster is placed at the other end of the tube for solids as a reference). A
detonator is used to initiate the booster. For solids, the substance is considered to propagate
detonation if the witness plate is punctured in one of three tests. For liquids, fragmentation
of the tube, a bulge in the witness plate, and measured velocity are considered in
combination.
Combustion or Thermal Tests
Koenen test. This test subjects solids and liquids to intense heat under partial confinement
to determine whether an explosion will occur and to measure the limiting diameter for the
explosion. The substance is put in a steel tube which is heated using a propane heater.
Trials are carried out with orifice plates of various sizes, from 1 to 20 millimeters, attached
to the tube, starting with 20 millimeters. Orifice sizes are varied until an explosion is
observed. The limiting diameter of the substance is the largest diameter at which an
explosion occurs in at least one of three trials. Occurrence of an explosion is determined by
the condition of the tube; if the tube fragments into three or more pieces, an explosion has
occurred.
Internal ignition test. This test determine the response of substances to rapidly rising
temperatures and pressures. The substance is put in a steel pipe with an igniter in the center
and capped at both ends. The test is considered positive if firing the igniter leads to
fragmentation of either the pipe or one of the end caps into at least two pieces.
Small-Scale Cook-Off Bomb (SCB) test. This test subjects the substance to slow external
heating. The substance is heated electrically in a steel vessel; the temperature is raised at a
rate of 3° C per minute until a reaction occurs or up to 400° C. The test is considered
positive if the vessel is ruptured or fragmented or the witness plate is deformed or punctured.
0.2 Tests for Assignment of Hazard Division
Materials designed to have an explosive effect or found by testing to have an explosive effect are
assigned to hazard divisions based on tests carried out on the materials packaged as they are to be transported
(or unpackaged if they are to be carried without packaging). The following tests are used to determine the
hazard division:
Single package test. Substances intended to function by detonation are tested with a standard
detonator; substances intended to function by deflagration are tested with an igniter sufficient
to ensure ignition of the substance; substances not intended for use as explosives are tested
first with a standard detonator and then, if no explosion occurs, with an igniter. A single
package or article is tested in the form in which it is to be shipped. If the total contents
explodes, the material is considered a candidate for Division 1.1. Evidence of explosion of
the total contents includes:
a crater at the test site.
-------
-73-
damage to the "witness plate" under the package,
measurement of a blast,
disruption and scattering of most of the confining material.
Stack test. Substances that are candidates for Division 1.1, based on results described above,
are tested further. A similar test (using a detonator or igniter as described above) is carried
out on a stack of packages or articles with a total volume of 0.15 cubic meters. If explosion
of the total contents occurs practically instantaneously, the material is assigned to Division
1.1. Evidence of explosion of the total contents includes:
a crater at the test site appreciably larger than the crater resulting from explosion of
a single package,
damage to the "witness plate" under the package that is appreciably greater than that
from a single package,
measurement of a blast that significantly exceeds that from a single package,
violent disruption and scattering of most of the confining material.
External fire (bonfire) test. Substances that are not identified as candidates for Division 1.1
in the single package test, or that are not assigned to Division 1.1 in the stack test, are tested
to determine their behavior in a fire and the effect on the surroundings from blast waves,
thermal effects, and projectiles. A stack of packages or articles with a total volume of 0.15
cubic meters or a minimum of three packages, whichever is greater, is exposed to external
heating (e.g., from a wood fire or propane burner). The results are assessed as follows:
For assignment to Division 1.1 (explosives with a mass explosion hazard:
explosion of the total contents occurs practically instantaneously.
For assignment to Division 1.2 (explosives with a projection hazard but not a mass explosion
hazard):
explosion of the total contents does not occur practically instantaneously, and
there is perforation of the "witness screens", or
more than ten metallic projectiles, each weighing more than 25 grams, are thrown
more than 50 meters from the stack, or
any metallic projectile weighing more than 150 grams is thrown more than 15 meters
from the stack.
For assignment to Division 1.3 (explosives with a fire hazard):
criteria for assignment to Division 1.1 or 1.2 are not met, and
a fireball is produced that extends beyond the witness screens, or
-------
tests:
-74-
. a jet flame is produced that extends more than three meters from the flames of the
fire, or
thermal radiation from the burning product exceeds that of the fire by more than
4kW/m3 at a distance of 15 meters from the stack (thermal radiation is measured for
five seconds and corrected to correspond to 100 kilograms of explosive material).
For assignment to Division 1.4 (explosive devices with a minor explosion hazard):
criteria for assignment to Division 1.1, 1.2, or 1.3 are not met, and
there is indentation of the witness screens, or
there are projectiles, thermal effects, or blast effects that could significantly hinder
emergency response efforts in the vicinity (five meters), or
the product tested is an article intended to produce a practical explosive or
pyrotechnic effect and there is some effect (projection, fire, smoke, heat, or loud
noise) external to the device.
Tests for Very Insensitive Explosives
For assignment to Division 1.5 (very insensitive explosives), an explosive must pass all of the following
Cap sensitivity test. A standard detonator or blasting cap is used in three attempts to
detonate the substance. A substance that detonates is too sensitive for Division 1.5.
DDT test. The explosive is packed into a steel tube and ignited by a hot wire. The test is
performed twice. A probe is used to monitor the shock wave. A substance that detonates
in this test is too sensitive for Division 1.5.
Second DDT test. This test is similar to the first DDT test, but a five gram igniter is used
and a witness plate is used to determine whether a detonation occurs. The test is performed
three times. If a hole is blown in the witness plate, the substance is too sensitive for Division
1.5.
External fire test. A package or packages with a total volume of not less than 0.15 cubic
meters, where the weight of explosive does not have to exceed 200 kilograms, is exposed to
an external fire large enough to engulf the bottom of the package. The flames must reach
at least halfway up the sides of the package. A substance that explodes in this test is too
sensitive for Division 1.5. Evidence of explosion in the fire may be a loud noise or the
projection of fragments from the fire area.
Princess incendiary spark test. The explosive is tested in powder form (it is crushed and
passed through a sieve if necessary). Three grams of the powdered explosive are put in a test
tube and a fuse is placed on the surface of the sample. The fuse is lit; if the explosive is
ignited, it is considered too sensitive for Division 1:5.
-------
-75-
REFERENCE
U.N. 1986. United Nations. Recommendations on the Transport of Dangerous Goods, Tests and Criteria.
New York: United Nations. Sales No. E.85.VIII.2. ISBN 92-1-139021-4.
------- |