introduction to explosives - wordpress.com

CHAPTER – 1

1

INTRODUCTION TO EXPLOSIVES

1. DEVELOPMENT OF BLACKPOWDER

Blackpowder, also known as gunpowder, was most likely the first explosive

composition. In 220 BC an accident was reported involving blackpowder

when some Chinese alchemists accidentally made black-powder while

separating gold from silver during a low-temperature reaction. According to

Dr Heizo Mambo the alchemists added potassium nitrate [also known as

saltpeter (KNO3)] and sulfur to the gold ore in the alchemists’ furnace but

forgot to add charcoal in the first step of the reaction. Trying to rectify their

error they added charcoal in the last step. Unknown to them they had just

made blackpowder which resulted in a tremendous explosion.

Blackpowder was not introduced into Europe until the 13th century when an

English monk called Roger Bacon in 1249 experimented with potassium

nitrate and produced blackpowder, and in 1320 a German monk called

Berthold Schwartz (although many dispute his existence) studied the

writings of Bacon and began to make blackpowder and study its properties.

The results of Schwartz’s research probably speeded up the adoption of

blackpowder in central Europe. By the end of the 13th century many

countries were using blackpowder as military aid to breach the walls of

castles and cities.

Blackpowder contains a fuel and an oxidizer. The fuel is a powdered mixture

of charcoal and sulfur which is mixed with potassium nitrate (oxidizer). The

mixing process was improved tremendously in 1425 when the Corning, or

granulating, process was developed. Heavy wheels were used to grind and

press the fuels and oxidizer into a solid mass, which was subsequently

Chapter 1 Introduction to Explosives

2

broken down into smaller grains. These grains contained and intimate

mixture of the fuels and oxidizer, resulting in a blackpowder which was

physically superior. Corned blackpowder gradually came into use for small

guns and hand grenades during the 15th century and for big guns in the

16th century. Blackpowder mills (using the Corning process) were erected at

Rotherhithe and Waltham Abbey in England between 1554 and 1603.

The first recording of blackpowder being used in civil engineering was during

1548-1627 blackpowder was used as a blasting aid for recovering ore in

Hungary. Soon, blackpowder was being used for blasting in Germany,

Sweden and other countries. In England, the first use of blackpowder for

blasting was in the Corninsh copper mines in 1670. Bofors Industries of

Sweden was established in 1646 and became the main manufacturer of

commercial blackpowder in Europe.

2. DEVELOPMENT OF NYTROGLYCERINE

By the middle of the 19th century the limitations of blackpowder as a

blasting explosive were becoming apparent. Difficult mining and turnneling

operations required a ‘better’ explosive. In 1846 the Italian, Professor

Ascanio Sobrero discovered liquid nitroglycerine [C3H5O3(NO2)3]. He soon

became aware of the explosive nature of nitroglycerine and discontinued his

investigations. A few years later the Swedish inventor, Immanuel Nobel

developed a process for manufacturing nitroglycerine, and in 1863 he

erected a small manufacturing plant in Helenborg near Stockholm with his

son, Alfred. Their initial manufacturing method was to mix glycerol with a

cooled mixture of nitric and sulfuric acids in stone jugs. The mixture was

strried by hand and kept cool by iced water, after the reaction had gone to

completion the mixture was poured into excess cold water. The second

manufacturing process was to pour glycerol and cooled mixed acids into a

conical lead vessel which had perforations in the constriction. The product


3

nitroglycerine flowed through the restrictions into a cold water bath. Both

methods involved the washing of nitroglycerine with warm water and a warm

alkaline solution to remove the acids. Nobel began to license the

construction of nitroglycerine plants which were generally built very close to

the site of intended use, as transportation of liquid nitroglycerine tended to

generate loss of life and property.

The Nobel family suffered many setbacks in marketing nitroglycerine

because it was prone to accidental initiation, and its initiation in bore holes

by blackpowder was unreliable. There were many accidental explosions, one

of which destroyed the Nobel factory in 1864 and killed Alfred’s brother,

Emil. Alfred Nobel in 1864 invented the metal ‘blasting cap’ detonator which

greatly improved the initiation of blackpowder, The detonator contained

mercury fulminate [Hg(CNO)2] and was able to replace black powder for the

initiation of nitroglycerine in bore holes. The mercury fulminate blasting cap

produced an initial shock which was transferred to a separate container of

nitroglycerine via a fuse, initiating the nitroglycerine.

After another major explosion in 1866 which completely demolished the

nitroglycerine factory. Alfred turned his attentions into the safety problems

of transporting nitroglycerine. To reduce the sensitivity of nitroglycerine

Alfred mixed it with an absorbent clay, ‘Kieselguhr’. This mixture became

known as ghur dynamite and was patented in 1867.

Nitroglycerine (1.1) has a great advantage over blackpowder since it contains

both fuel and oxidizer elements in the same molecule. This gives the most

intimate contact for both components.

3. DEVELOPMENT OF MERCURY FULMINATE

Mercury fulminate was first prepared in the 17th century by the Swedish-


4

German alchemist, Baron Johann Kunkei von Lőwnstern. He obtained this

dangerous explosive by treating mercury with nitric acid and alcohol. At that

time, Kunkel and other alchemists could not find a use for the explosive and

the compound became forgotten unti Edward Howard of England

rediscovered it between 1799 and 1980. Howard examined the properties of

mercury fulminate and proposed its use as percussion initiator for

blackpowder and in1807 a Scottish Clergyman, Alexander Forsyth patented

the device.

Fig: Nitroglycerine (1.1)

4. DEVELOPMENT OF NITROCELLULOSE

At the same time as nitroglycerine was being prepared, the nitration of

cellulose to produce nitrocellulose (also known as guncotton) was also being

undertaken by different workers, notably Schőbein at Basel and Bőttger at

Frankfurt-am-Main during 1845-47. Earlier in 1833, Braconnot had

nitrated starch, and in 1838, Pelouze, continuing the experiments of

Braconnot, also nitrated paper, cotton and various other materials but did

not realize that he had prepared nitrocellulose. With the announcement by

Schőbein in 1846, and in the same year by Bőttger that nitrocellulose had

been prepared, the names of these two men soon became associated with

the discovery and utilization of nitrocellulose. However, the published

literature at that time contains papers by several investigators on the

nitration of cellulose before the process of Schőnbein was known.


5

Many accidents occurred during the preparation of nitrocellulose, and

manufacturing plants were destroyed in France, England and Austria.

During these years, Sir Frederick Abel was working on the instability

problem of nitrocellulose for the British Government at Woolwich and

Waltham Abbey, and in 1865 he published his solution to this problem by

converting nitrocellulose into a pulp. Abel showed through his process of

pulpling, boiling and washing that the stability of nitrocellulose could be

greatly improved. Nitrocellulose was not used in military and commercial

explosives until 1868 when Abel’s assistant, E.A. Brown discovered that dry,

compressed, highly-nitrated nitrocellulose could be detonated using a

mercury fulminate detonator, and wet, compressed nitrocellulose could be

exploded by a small quantity of dry nitrocellulose (the principle of a Booster).

Thus, large blocks of wet nitrocellulose could be used with comparative

safety.

5. DEVELOPMENT OF DYNAMITE

In 1875 Alfred Nobel discovered that on mixing nitrocellulose with

nitroglycerine a gel was formed. This gel was developed to produce blasting

gelatine, gelatine dynamite and later in 1888, ballistite, the first smokeless

powder. Ballistite was a mixture of nitrocellulose, nitroglycerine, benzene

and camphor. In 1889 a rival product of similar composition to ballistite was

patented by the British Government in the names of Abel and Dewar call

‘Cordite’ In its various forms Cordite remained the main propellant of the

British Forces until the 1930s.

In 1867, the Swedish chemists Ohlsson and Norrbin found that the

explosive properties of dynamites were enhanced by the addition of

ammonium nitrate (NH4NO3). Alfred Nobel subsequently acquired the patent

of Ohlsson and Norbbin for ammonium nitrate and used this in his

explosive compositions.


6

6. DEVELOPMENT OF AMMONIUM NITRATE

Ammonium nitrate was first prepared in 1654 by Glauber but it was not

until the beginning of the 19th century when it was considered for use in

explosives by Grindel and Robin as a replacement for potassium nitrate in

blackpowder. Its explosive properties were also reported in 1849 by Reise

and Millon when a mixture of powdered ammonium nitrate and charcoal

exploded on heating.

Ammonium nitrate was not considered to be an explosive although small

fires and explosions involving ammonium nitrate occurred throughout the

world.

After the end of World War II, the USA Government began shipments to

Europe of so-called Fertilizer Grade Ammonium Nitrate (FGAN), which

consisted of grained ammonium nitrate coated with about 0.75% wax and

conditioned with about 3.5% clay. Since this material was not considered to

be an explosive, no special precaution were taken during its handling and

shipment – workmen even smoked during the loading of the material.

Numerous shipments were made without trouble prior to 16 and 17 April

1947, when a terrible explosion occurred. The SS Grandchamp and the SS

Highflyer, both moored in the harbor of Texas City and loaded with FGAN,

blew up. As a consequence of these disasters, a series of investigation was

started in the USA in an attempt to determine the possible cause of the

explosions. At the same time a more thorough study of the explosive

properties ammonium nitrate and its mixtures with organic and inorganic

materials was also conducted. The explosion at Texas City had barely taken

place when a similar one aboard the SS Ocean Liberty shook the harbor of

Brest in France on 28 July 1947.


7

The investigations showed that ammonium nitrate is much more dangerous

than previously thought and more rigid regulations governing its storage,

loading and transporting in the USA promptly put into effect.

7. DEVELOPMENT OF COMMERCIAL EXPLOSIVES

7.1 DEVELOPMENT OF PERMITTED EXPLOSIVES

Until 1870, blackpowder was the only explosive used in coal mining, and

several disastrous explosions occurred. Many attempts were made to modify

blackpowder, these included mixing blackpowder with ‘cooling agents’ such

as ammonium sulfate, starch, paraffin, etc., and placing a cylinder filled

with water into the bore hole containing the blacpowder. None of these

methods proved to be successful.

When nitrocellulose and nitroglycerine were invented, attempts were made

to use these as ingredients for coal mining explosives instead of blackpowder

but they were found not to be suitable for use in gaseous coal mines. It was

not until the development of dynamite and blasting Gelatine by Nobel that

nitroglycerine-based explosives began to dominate the commercial blasting

and mining industries. The growing use of explosives in coal mining brought

a corresponding increase in the number of gas and dust explosions, with

appalling casualty totals. Some European governments were considering

prohibiting the use of explosives in coal mines and resorting to the use of

hydraulic devices or compressed air. Before resorting to such drastic

measures, some governments decided to appoint scientists, or commissions

headed by them, to investigate this problem. Between 1877 and 1880,

commissions were created in France, Great Britain, Belgium and Germany.

As a result of the work of the French Commission, maximum temperatures

were set for explosions in rock blasting and gaseous coal mines. In Germany

and England it was recongnized that regulating the temperature of the

explosion was only one of the factors in making an explosive safe and that


8

other factors should be considered. Consequently, a Testing gallery was

constructed in 1880 Gelsenkirchen in Germany in order to test the newly-

developed explosives. The testing gallery was intended to imitate as closely

as possible the conditions in the mines. A Committee was appointed in

England in 1888 and a trial testing gallery at Hebburn Colliery was

completed around 1890. After experimenting with various explosives the use

of several explosive materials was recommended, mostly based on

ammonium nitrate. Explosives which passed the tests were called ‘permitted

explosives’. Dynamite and blackpowder both failed the tests and were

replaced by explosives based on ammonium nitrate. The results obtained by

this Committee led to the Coal Mines Regulation Act of 1906. Following this

Act, testing galleries were constructed at Woolwich Arsenal and Rotherham

in England.

7.2 DEVELOPMENT OF ANFO AND SLURRY EXPLOSIVES

By 1913, British coal production reached an all-time peak of 287 million

tons, consuming more than 5000 tons of explosives annually and by 1917,

92% of these explosives were based on ammonium nitrate. In order to

reduce the cost of explosive compositions the explosives industry added

more of the cheaper compound ammonium nitrate to the formulations, but

this had an unfortunate side effect of reducing the explosives regularly filled

with water. Chemists overcame this problem by coating the ammonium

nitrate with a various inorganic powders before mixing it with dynamite, and

by improving the packaging of the explosives to prevent water ingress.

Accidental explosions still occurred involving minimum explosives, and in

1950 manufacturers started to develop explosives which were waterproof

and solely contained the less hazardous ammonium nitrate. The most

notable compositions was ANFO (Ammonium Nitrate Fuel Oil). In the 1970s,

the USA companies Ireco and DuPont began adding paint-grade aluminium

and mono-methylamine nitrate (MAN) to their formulations to produce gelled


9

explosives which could detonate more easily. More recent developments

concern the production of emulsion explosives which contain droplets of a

solution of ammonium nitrate in oil. These emulsions are waterproof

because the continuous phase is a layer of oil, and they can readily detonate

since the ammonium nitrate and oil are in close contact. Emulsion

explosives are safer than dynamite, and are simple and cheap to

manufacture.

8. DEVELOPMENT OF MILITARY EXPLOSIVES

8.1 DEVELOPMENT OF PICRIC ACID

Picric acid [(trinitrophenol) (C6H3N3O7)] was found to be a suitable

replacement for blackpowder in 1885 by Turpin and in 1888 blackpowder

was replaced by picric acid in British munitions under the name Liddite.

Picric acid is probably the earliest known nitrophenol: it is mentioned in the

alchemical writings of Glauber as early as 1742. In the second half of the

19th century, picric acid was widely used as fast dye for silk and wool. It was

not until 1830 that the possibility of using picric acid as and explosive was

explored by Welter.

(1.2)

Designolle and Brugere suggested that picrate salts could be used as a

propellant, while in 1871, Abel proposed the use of ammonium picrate as an

explosive, In 1873, Sprengel showed that picric acid could be detonated to

an explosion and Turpin, utilizing these results, replaced blackpowder with


10

picric acid for the filling of munition shells. In Russia, Panpushko prepared

picric acid in 1894 and soon realized its potential as an explosive.

Eventually, picric acid (1.2) was accepted all over the world as the basic

explosive for military uses.

Picric acid did have its problems: in the presence of water it caused

corrosion of the shells, its salts were quite sensitive and prone to accidental

initiation, and picric acid required prolonged heating at high temperatures

in order for it to melt.

8.2 DEVELOPMENT OF TETRYL

An explosive called tetryl was also being developed at athe same time as

picric acid. Tetryl was first prepared in 1877 by Mertens and its structure

established by Romburgh in 1883. Tetryl (1.3) was used as an explosive in

1906, and in the early part of this century it was frequently used as the base

charge of blasting caps.

(1.3)

8.3 DEVELOPMENT OF TNT

Around 1902 the Germans and British had experimented with

trinitrotoluene [(TNT) (C7H5N3O6)], first prepared by Wilbrand in 1863. The

first detailed study of the preparation of 2,4,6- trinitrotoluene was by

Beilstein and Kuhlberh in 1870 when they discovered the isomer 2,4,5-


11

trinitrotoluene. Pure 2,4,6- trinitrotoluene was prepared in 1880 by Hepp

and its structure established in 1883 by Claus and Beeker. The

manufacture of TNT began in Germany in 1891 and in 1899 aluminium was

mixed with TNT to produce and explosive composition. In 1902, TNT was

adopted for use by the German Army replacing picric acid, and in 1912 the

US Army also started to use TNT. By 1914, TNT (1.4) became the standard

explosive for all armies during World War I.

(1.4)

Production of TNT was limited by the availability of toluene from coal tar and

it failed to meet demand for the filling of munitions. Use of a mixture of TNT

and ammonium nitrate, called amatol, became wide-spread to relieve the

shortage of TNT. Underwater explosive used the same formulation with the

addition of aluminium and was called animal.

8.4 DEVELOPMENT OF NITROGUANIDINE

The explosive nitroguanidine was also used in World War I by the Germans

as an ingredient for bursting charges. It was mixed with ammonium nitrate

and paraffin for filling trench mortar shells. Nitroguanidine was also used

during World War II and later in triple-base propellants.

Nitroguanidine (CH4N4O2) was first prepared by Jousselin in 1877 and its

properties investigated by Vieille in 1901. In World War I nitroguanidine was

mixed with nitrocellulose and used as a flashless propellant. However, there

were problems associated with this composition; nitroguanidine attached


12

nitrocellulose during its storage. This problem was overcome in 1937 by the

company Dynamite AG who developed a propellant composition containing

nitroguanidine called ‘Gudol Pulver’ Gudol Pulver produced very little

smoke, had no evidence of a muzzle flash on firing , and was also found to

increase the life of the gun barrel.

After World War I, major research programmes were inaugurated to find new

and more powerful explosive materials. From these programmes came

cyclotrimethylenetrinitramine [(RDX) (C3H6N6O6)] also called Cyclonite or

Hexogen, and pentaerythritol tetranitrate [(PETN) (C5H8N4O12)].

8.5 DEVELOPMENT OF PETN

PETN was first prepared in 1894 by nitration of pentaerythritol. Commercial

production of PETN could not be achieved until formaldehyde and

acetaldehyde required in the synthesis of pentaerythritol became readily

available about a decade before World War II. During World War II, RDX was

utilized more than PETN because PETN was more sensitive to impact and its

chemical stability was poor. Explosive compositions containing 50% PETN

and 50% TNT were developed and called ‘Pentrolit’ or ‘Pentolite; this

composition was used for filling hand and anti-tank grenades and detonator.

8.9 DEVELOPMENT OF RDX AND HMX

RDX was first prepared in 1899 by German, Henning for medicinal use. Its

value as an explosive was not recognized until 1920 by Herz.

Herz succeeded in preparing RDX by direct nitration of hexamine, but the

yields were low and the process was expensive and unattractive for large

scale production. Hale, at Picatinny Arsenal in 1925, developed a process for

manufacturing RDX which produced yields of 68%. However, no further


13

substantial improvements were made in the manufacture of RDX until 1940

when Meissner developed a continuous method for manufacturing process

for RDX (1.5) from hexamine which gave the greatest yield.

(1.5)

Bachmann’s products were known as Type B RDX and contained a

constant impurity level of 8-12%. The explosive properties of this impurity

were later utilized and the explosive HMX, also known as Octogen, was

developed. The Bachmann process was adopted in Canada during World

War II, and later in the USA by the Tennessee-Eastman Company. This

manufacturing process was more economical and also led to the discovery of

several new explosives. A manufacturing route for the synthesis of pure RDX

(no impurities) was developed by Brockman, and this became know as Type

A RDX.

In Great Britain the Armament Research Department at Woolwich began

developing a manufacturing route for RDX after the publication of Herz’s

patent in 1920. A small-scale pilot plant producing 75 lbs of RDX per day

was installed in 1933 and operated until 1939. Another plant was installed

in 1939 at Waltham Abbey and a full-scale plant was erected in 1941 near

bridgewater. RDX was not used as the main filling in British shells and

bombs during World War II but was added to TNT to increase the power of

the explosive compositions. RDX was used in explosive compositions in

Germany, France, Italy, Japan, Russia, USA, Spain and Sweden.

Research and development continued throughout World War II to develop


14

new and more powerful explosives and explosive compositions. Torpex

(TNT/RDX/aluminium) cyclotetramethylenetetranitramine, known as

Octogen [(HMX) (C4H8N8O8)], became available at the end of the World War

II. In 1952 and explosive composition called ‘Octol’ was developed; this

contained 75% HMX and 25% TNT. Mouldable plastic explosives were also

developed during World War II; these often contained Vaseline or gelatinized

liquied nitro compounds to give a plastic –like consistency. A summary of

explosive compositions used in World War II is presented in Table 1.1.

Table 1.1 Example of explosive compositions used in World War II

Name Composition

Baronal Barium nitrate, TNT and alunimium

Composition A 88.3% RDX and 11.7% non-explosive plasticizer

Composition B RDX, TNT and wax (cyclotol)

H-6 45% RDX, 30% TNT, 20% aluminium and 5% wax

Minol-2 40% TNT, 40% ammonium nitrate and 20% aluminium

Pentolites 50% PETN and 50% TNT

Picratol 52% Picric acid and 48% TNT

PIPE 81% of PETN and 19 % Gulf Crown E Oil

PTX-1 30 % of RDX, 50% tetryl and 20% TNT

PTX-2 41-44% RDX, 26-28% PETN and 28-33% TNT

PVA-4 90% RDX, 8% PVA and 2% dibutyl phthalate

RIPE 85% RDX, and 15% Gulf Crown E Oil

Tetrytols 70% Tetryl and 30% TNT

Torpex 42% RDX, 40% TNT and 18% aluminium

9. POLYMER BONDED EXPLOSIVES

Polymer bonded explosives (PBXs) were developed to reduce the sensitivity of

the newly-synthesized explosive crystals by embedding the explosive crystals

in a rubber-like polymeric matrix. The first PBX composition was developed


15

at Los Alamos Scientific Laboratories in USA in 1952. The composition

consisted of RDX crystals embedded in plasticized polystyrene, since 1952,

Lawrence Livermore Laboratories, the US Navy and many other

organizations have developed a series of PBX formulations, some of which

are listed in Table 1.2.

HMX-based PBXs were developed for projectiles and lunar scismic

experiments during the 1960s and early 1970s using Teflon (polytetra-

fluoroethylene) as the binder. PBXs based on RDX and RDX/PETN have also

been developed and are known as Semtex. Development is continuing in this

area to produce PBXs which contain polymers that are energetic and will

contribute to the explosive performance of the PBX. Inert prepolymer have

been substituted by energetic prepolymers [(mainly hydroxyl terminated

polybutadiene (HTPB)] in explosive composition, in order to increase the

explosive performance, without compromising its vulnerability to accidental

initiation. In the last ten years it has become apparent that PBXs containing

inert of energetic binders are more sensitive to impact compared to

traditional explosive compositions. The addition of a plasticizer has reduced

the sensitivity of PBXs whilst improving its processability and mechanical

properties. Energetic plasticizers have also been developed for PBXs.

Examples of energetic polymers and energetic plasticizers under

investigation are presented in Table 1.3 and 1.4, respectively.

Table 1.2 Examples of PBX compositions, where HMX is cyclo-

tetramethylene-tetranitramine (Octogen), HNS is hexanitrostilebene,

PETN is pentaerythritol tetranitrate, RDX is cyclotetramethylene-

tetranitramine (Hexogen) and TATB is 1, 3, 5-triamino-2, 4, 6-

trinitrobenzene.

Explosive Binder and plasticizer

HMX Acetyl-formyl-2, 2-dinitropropanol (DNPAF) and polyurethane


16

HMX Cariflex (thermoplastic elastomer)

HMX Hydroxy-terminated polybutadiene(polyurethane)

HMX Hydroxy-terminated polyester

HMX Kraton(block copolymer of styrene and ethylene-butylene)

HMX Nylon (polyamide)

HMX Polyester resin-styrene

HMX Polyethylene

HMX Polyurethane

HMX Poly (vinyl) alcohol

HMX Poly (vinyl) butyral resin

HMX Teflon (polytetrafluoroethylene)

HMX Viton (fluoroelastomer)

HNS Teflon (polytetrafluoroethylene)

NTO Cariflex (block copolymer of butadiene-styrene)

NTO/HMX Cariflex (block copolymer of butadiene-styrene)

NTO/HMX Estane (polyester polyurethane copolymer)

NTO/HMX Hytemp (thermoplastic elastomer)

PETN Butyl rubber with acetyl tributylcitrate

PETN Epoxy resin – diethylenetriamine

PETN Kraton (block copolymer of styrene and ethylene – butylene)

Explosive Binder and plasticizer

PETN Latex with bis-(2-ethylhexyl adipate)

PETN Nylon (polyamide)

PETN Polyester and styrene copolymer

PETN Poly (ethyl acrylate) with dibutyl phthalate

PETN Silicone rubber

PETN Viton (fluoroelastomer)

PETN Teflon (polytetrafluoroethylene)

RDX Epoxy ether

RDX Exon (polychlorotrifluoroethylene / vinylidine chloride)

RDX Hydroxy – terminated polybutadiene (polyurethane)


17

RDX Kel – F (polychlorotrifluoroethylene)

RDX Nylon (polymide)

RDX Nylon and aluminium

RDX Nitro-fluoroalkyl epoxides

RDX Polyacrylate and paraffin

RDX Polyamide resin

RDX Polyisobutylene / Teflon (polytetrafluoroethylene)

RDX Polyester

RDX Polystyrene

RDX Teflon (polytetrafluoroethylene)

TATB/HMX Kraton (block copolymer of styrene and ethylene – butylene)

Table 1.3 Example of energetic polymers

Common Name Chemical Name Structure

GLYN Clycidyl nitrate

polyGLYN Poly(glycidyl nitrate)


NIMMO (monomer) 3-Nitratomythyl-3methyl oxetane

poly NIMMO Poly(3-nitratomethyl-3- methyl oxetane)

GAP Glycidyl azide polymer


18

AMMO (monomer) 3-Azidomethyl-3-methyl oxetane

PolyAMMO Poly(3-azidomethyl-3-methyl Oxetane)

BAMMO (monomer) 3,3-Bis-azidomethyl oxetane

PolyBAMMO (monomer)

Poly(3,3-bis-azidomethyl oxatane)

Table 1.4 Examples of energetic plasticizers


NENAs Alkyl nitratoethyl nitramines


EGDEN Ethylene glycol dinitrate

MTN Metriol trinitrate

BTTN Butane-1, 2, 4-triol

Trinitrate


19

K10 Mixture of di- and tri-nitroethylbenzene

BDNPA/F Mixture of bis-

dinitropropylacetal and bis-dinitropropylformal

10. RECENT DEVELOPMENTS

Recent developments in explosives have seen the production of

hexanitrostilbene [(HNS) (C14H6N6O12)] in 1966 by Shipp, and

traminotrinitrobenzene {(TATB) [(NH2)3 C6 (NO2)3]} in 1978 by Adkins and

Norris. Both of these materials are able to withstand relatively high

temperatures compared with other explosives. TATB was first prepared in

1988 by Jackson and Wing, who also determined its solubility

characteristics. In the 1950s, the USA Naval Ordance Laboratories

recognized TATB as useful heat-resistant explosive, and successful small

scale preparations and synthetic routes for large-scale production were

achieved to give high yields.

Nitro-1,2,4-triazole-3one [(NTO) (C2H2N4O2)] is one of t he new explosives

with high energy and low sensitivity. It has a high heat of reaction and

shows autocatalytic behavior during thermal decomposition. NTO was first

reported in 1905 from the nitration of 1,2,4-triazole-3one. There was

renewed interest in NTO in the late 1960s, but it wasn’t until 1987 that Lee,

Chapman and Coburn reported the explosive properties of NTO. NTO is now

widely used in explosive formulations, PBXs and gas generation for

automobile inflatable airbag systems. The salt derivatives of NTO are also

insensitive and are potential energetic ballistic additives for solid rocket


20

propellants.

2,4,6,8,10,12-Hexanitrohexaazaisowurtzitane (C6H6N12O12) or HNIW, more

commonly called CL-20 was first eynthesized in 1987 by Anie Nielsen, and is

now being produced at SNPE in France in quantities of 50-100kg on an

industrial pilot scale plant.

Nitrocubanes are probably athe most powerful explosives with a predicted

detonation velocity of > 10,000 ms-1. Cubanes were first synthesized at the

University of Chicago, USA by Eaton and Cole in 1964. The US Army

Armament Research Development Centre (ARDEC) them funded

development into the formation of octanitrocubane [(ONC) (C8N8O16)] and

heptanitrocubane [(HpNC) (C8N7O14)]. ONC and HpNC were successfully

synthesized in 1997 and 2000 respectively by Eaton and co-workers. The

basic structure of ONC is a cubane molecule where all the hydrogens have

been replaced by nitro groups (1.6) HpNC is denser than ONC and predicted

to be a more powerful, shock – insensitive explosive.

(1.6)

The research into energetic molecules which produce a large amount of gas

per unit mass, led to molecular structures which have a high hydrogen to

carbon ratio. Examples of these structures are hydrazinium nitroformate

(HNF) and ammonium dinitramide (ADN). The majority of the development of

HNF has been carried out in The Netherlands whereas the development of


21

ADN has taken place in Russia, USA and Sweden. ADN is a dense non

chlorine containing powerful oxidizer and is an interesting candidate for

replacing ammonium perchlorate as an oxidizer for composite propellants.

ADN is less sensitive to impact than RDX and HMX, but more sensitive to

friction and electrostatic spark.

11. INTENSIVE MUNITIONS

Recent developments of novel explosive materials have concentrated on

reducing the sensitivity of the explosive materials to accidental initiation by

shock,, impact and thermal effects. The explosive materials, which have this

reduced sensitivity, are call Insensitive Munitions, (IM). Although these

explosive materials are insensitive to accidental initiation they still perform

very well when suitably initiated. Examples of some explosive molecules

under development are presented in Table 1.5 A summary of the significant

discoveries in the history of explosives throughout the world is presented in

Table 1.6.

12. POLLUTION PREVENTION

Historically waste explosive compositions (including propellants) have been

disposed of by dumping the waste composition in the sea, or by burning or

detonating the composition in an open bonfire. In 1994 the United Nations

banned the dumping of explosive waste into the sea, and due to an increase

in environmental awareness burning the explosive waste in an open bonfire

will soon be banned since it is environmentally unacceptable. Methods are

currently being developed to remove the waste explosive compositions safely

from the casing using a high pressure water jet. The recovered material then

has to be disposed; one method is to reformulate the material into a

commercial explosive. In the future, when formulating a new explosive

composition, scientists must not only consider its overall performance but

must make sure that it falls into the ‘insensitive munitions’ category and


22

that it can easily be disposed or recycled in an environmentally friendly

manner.

Table 1.5 Examples of explosive molecules under development


NTO 5-Nitro-1,2,4-triazol-3-one

ADN Ammonium dinitramide

TNAZ 1,3,3-Trinitroazetidine

CL-202 2,4,6,8,10,12-

Hexanitro-2,4,6,8,10,12-hexa-azatetracyclododecane

Table 1.6 Some significant discoveries in the history of incendiaries, fireworks, blackpowder and Explosive

Date Explosive

220 BC Chinese alchemists accidentally made blackpowder

222 - 235 AD Alexander VI of the Roman Empire called a ball of quicklime and asphalt ‘automatic fire’ which Spontaneously ignited on coming into contact with water.

690 Arabs used blackpowder at the siege of Mecca.

940 The Chinese invented athe ‘Fire Ball’ which is made of an explosive composition similar to blackpowder.


23

1040 The Chinese built a blackpowder plant in Pein King.

1169 – 1189 The Chinese started to make fireworks.

1249 Roger Bacon first made blackpowder in England

1320 The German, Schwartz studied blackpowder and helped it to be introduced into central Europe.

1425 Corning, or granulating, process was developed.

1627 The Hungarian, Kaspar Weindl used blackpowder in blasting.

1646 Swedish Bofors Industries began to manufacture blackpowder.

1654 Preparation of ammonium nitrate was undertaken by Glauber.

1690 Ther German, Kunkel prepared mercury fulminate.

1742 Glauber preapared picric acid.

1830 Welter explored the use of picric acid in explosive

1838 The Frenchman, Pelouze carried out nitration of paper and cotton.

1646 Schőbein and Bőttger nitrated cellulose to produce guncotton.

1846 The Italian, Sobrero discovered liquid nitroglycerine.

1849 Reise and Millon reported that a mixture of charcoal and ammonium nitrate exploded on heating.

1863 The Swedish inventor, Nobel manufactured nitroglycerine.

1863 The German, Wilbrand prepared TNT.

1864 Schultze prepared nitrocellulose propellants.

Date Explosive

1864 Nitrocellulose propellants were also prepared by Vieile.

1864 Nobel developed the mercury fulminate detonator.

1865 An increase in the stability of nitrocellulose was achieved by Abel.

1867 Nobel invented dynamite.

1867 The Swedish chemists, Ohlsson and Norrbin added ammonium nitrate to dynamites.

1868 Brown discovered that dry, compressed guncotton could be detonated.

1868 Brown also found that wet, compressed nitrocellulose could be exploded by a small quantity of dry nitrocellulose.


24

1871 Abel proposed that ammonium picrate could be used as an explosive.

1873 Sprengel showed that picric acid could be detonated.

1875 Nobel mixed nitroglycerine with nitrocellulose to form a gel.

1877 Mertens first prepared tetryl.

1879 Nobel manufactured Ammonium Nitrate Gelatine Dynamite.

1880 The German, Hepp prepared pure 2, 4, 6-trinitrotoluene (TNT).

1883 The structure of tetryl was established by Romburgh.

1883 The structure of TNT was established by Claus and Becker.

1855 Turpin replaced blackpowder with picric acid.

1888 Jackson and Wing first prepared TATB.

1888 Picric acid was used in British Munitions called Liddite.

1888 Nobel invented Ballistite.

1889 The British Scientist, Abel and Dewar Patented Cordite.

1891 Manufacture of TNT began in Germany.

1894 The Russian, Panpushko prepared picric acid.

1894 Preparation of PETN was carried out in Germany.

1899 Preparation of RDX for medicinal use was achieved by Hennings.

Date Explosive

1899 Aluminium was mixed with TNT in Germany.

1900 Preparation of nitroguanidine was developed by Jousselin.

1902 The German Army replaced picric with TNT.

1905 NTO was first reported from the nitration of,2,4-triazol–3 one.

1906 Tertryl was used as an explosive.

1912 The US Army started to use TNT in munitions.

1920 Preparation of RDX by the German, Herz.

1925 Preparation of a large quantity of RDX in the U.S.A

1940 Meissner developed the continuous method for the manufacture of RDX.

1940 Bachman developed the manufacturing process for RDX.

1943 Bachman prepared HMX.


25

1952 PBXs were first prepared containing RDX, polystyrene and dioctyl phthalate in the USA.

1952 Octols were formulated

1957 Slurry explosives were developed by the American, Cook.

1964 Cubanes were first synthesized at the University of Chicago, USA by Eaton and Cole.

1966 HNS was prepared by Shipp.

1970 The USA companies, Ireco and Dupont produced a gel explosive by adding paint-grade Aluminium and MAN to ANFO.

1978 Adkins and Norris prepared TATB

1983 TNAZ was first prepared at Flurochem Inc.

1987 Lee, Chapmand Coburn reported the explosive properties of NTO.

1987 CL20 was first synthesized by Arnie Nielsen.

1997 ONC was successfully synthesized by Eaton and coworkers.

2000 HpNC was successfully synthesized by Eaton and coworkers.

13. CHEMISTRY OF COMBUSTION AND EXPLOSION

For a simple understanding of explosive, it is helpful to compare an

explosive reaction with the more familiar combustion or burning reaction.

Three components are needed to have a common fire: fuel, oxygen from the

air, and a source of ignition. The process of combustion is basically an

oxidation- reduction (redox) reaction between the fuel and oxygen. Once

initiated, this reaction can become self-sustaining, producing large volumes

of gases and heat. The heat given off further expands the gases and provides

the stimulus for the reaction to continue by heating and igniting

surrounding fuels.

The basic burning reaction is a relatively slow, diffusion-controlled process

that occurs within the flames or near the surface of glowing embers. The size

of the fire depends upon how much fuel is involved and on the rate of the


26

combustion reaction. The rate of combustion reaction depends on how

finely divided the fuel is and how rapidly the oxygen reaches the flame, that

is the intimacy of contact between the fuel and the oxygen in the air.

Burning rate is greatly increased when convection of the air, natural (wind)

or man-made (fanning the flame), joins diffusion in supplying oxygen to the

flame.

Another result of an intimate mixture of the fuel and air is the completeness

or efficiency of the reaction. In a complete combustion all the fuel elements

are oxidized to their highest oxidation state. Thus, burning of wood, being

mainly cellulose and gasoline being generally a hydrocarbon (e.g., octane),

produces primarily carbon dioxide and water vapor upon complete

combustion. Once initiated, these burning reactions give off heat energy,

which sustain the reactions. Heat is released because the oxidized products

of the reactions are in a lower energy state (more stable) than the reactants.

The maximum potential energy release can be calculated from the respective

heats of formation of the products and reactants. Actual heats of

combustion can be measured experimentally by causing the reaction to

occur in a bomb calorimeter. The calculated energy values for the above

reactions are 3.875 cal/g for cellulose and 10,704 cal/g for octane.

In the case of an inefficient burn, some less stable ore higher-energy

products are formed so that the resultant heat energy given off is lower than

that for complete combustion. In the above examples, inefficient combustion

could result from lack or accessible oxygen, producing carbon monoxide or

even carbon particles instead of carbon dioxide. A smoky flame is evidence

of unburned carbon particles and results from inefficient combustion where

fuel particles are so large or so dense that oxygen cannot diffuse to the

burning surface fast enough. If this inefficiency is great enough, insufficient

heat is given off to keep the reaction going, and the fire will die out.


27

All chemical explosive reactions involve similar redox reactions; so the above

principles of combustion can help illustrate, in a very basic way, the

chemistry involved in explosions. As in a fire, there components (fuel,

oxidizer, ignition source) are needed for an explosion. Figure 2.1 shows the

explosion triangle, which is similar to a fire triangle. In general, the products

of an explosion are gases and heat, although some solid oxidation products

may be

Oxidation-Reduction gases + heat

(fast)

Fig. 2.1 An explosion triangle

Fig. 2.2 Chemical Structure of three molecular explosive

Produced, depending upon the chemical explosive composition. As in normal

combustion, the gases produced usually include carbon dioxide and water

vapor plus other gases such as nitrogen, again depending upon the


28

composition of the chemical explosive.

It should be noted that an explosion differs from ordinary combustion in two

very significant ways. First, oxygen from the air is not major reactant in the

redox reactions of most explosives. The source of oxygen (or other reducible

species) needed for reaction with the fuel-oxidizer-maybe part of the same

molecule as the fuel or a separate intermixed material. Thus, an explosive

may be thought of as merely an intimate mixture of oxidizer and fuel. This

high degree of intimacy contributes to the second significant difference

between an explosion and normal combustion – the speed with which the

reaction occurs.

Explosives in which the oxidizer and fuel portions are part of the same

molecule are called molecular are called molecular explosives. Classical

examples or molecular explosives are 2, 4, 6-trinitroluene (TNT),

penaerythhritol tetranitrate (PETN) and nitroglycerine (NG) or, more

precisely. Glycerol trinitrate. The chemical structures of these explosives are

shown in Fig. 2.2.

As can be seen in the structures, the oxidizer portions of the explosives are

the nitro (-NO2) groups in PETN and NG. The fuel portions of all three

explosives are the carbon and hydrogen (C and H) atoms. Comparison of the

ratios of carbon to oxygen in these explosives (i.e. approximately 1:1 for TNT,

approximately 1:2 for PETN, and 1:3 for NG) shows that TNT and PETN are

deficient in oxygen; that is, there is insufficient oxygen present in the

molecule to fully oxidize the carbon hydrogen. Consequently, products such

as carbon monoxide, solid carbon (soot), and hydrogen are produced, as well

as carbon dioxide and water vapor. Prediction, as well as carbon dioxide and

water vapor. Prediction of exact products of explosives, because the amounts

of CO2, CO, H2O and H2 will vary as will a host of trace products such as

residual hydrocarbons, depending upon reaction conditions (explosive


29

density, degree of confinement of the explosive, etc.) [4, 5]. The following

equations show typical ideal reaction products along with calculated heats

of reaction for these molecular explosives:

TNT: C7H5 N3O6 1.5CO2 + 0.5CO + 2.5H2O + 1.5N2 + 5C + 1,290cal/g PETN: C5HgN4O12 4CO2 + 4H2O + 2N2 + C + 1,510cal/g NG: C3H5N3O9 3CO2 + 2.5H2O + 1.5N2 + 0.25O2 + 1,480cal/g

Explosive in which the oxidizer and fuel portions come from different

molecules are called composite explosives because they are a mixture of two

of more chemicals, A classic industrial example is a mixture of solid

ammonium nitrate (AN) and liquid fuel oil (FO), The common designation for

this explosive is the acronym, ANFO. The oil used (typically #2 diesel fuel) is

added in sufficient quantity to react with the available oxygen from the

nitrate portion of AN. The redox reaction of ANFO is as follows.

3NH4NO3 + CH2 CO2 + 7H2O + 3N + 880cal/g AN FO

“Oxygen balance” (O.B) is the term applied to quantify either the excess

oxygen in an explosive compound or mixture (beyond what is needed for

complete combustion of the fuel elements) or oxygen deficiency (compared to

the amount required for complete combustion). It is expressed as either a

percentage or a decimal fraction of the molecular weight of the oxygen in

excess (+) or deficiency (-) divided by the molecular weight of the explosive or

the ingredient being considered. Individual components of an explosive

mixture have O.B. values that may be summed for the mixture. Shown

below are the O.B. calculations for AN and FO.

NH4NO3 2H2O + N2 + 2

1O2

AN O2


30

Mol. Wt. =80 Mol. Wt. =32

O.B =

80

322

1

= 0.20 (a)

(CH2) n + 2

n3O2 nCO2 + nH2O

Mol. Wt. = ~14n

O.B =

n14

322

n3

= 3.43 (b)

From the O.B values, one can readily determine the ratio of ingredients to

give a zero O.B. mixture for optimum efficiency and energy.

Thus, the weight ratio for ANFO is 94.5 parts of AN and 5.5 parts of FO (94.5 × 0.20 = 5.5 × 3.43).

For the molecular explosives shown previously, the respective oxygen

balances are: TNT, -0.74; PETN,-0.10; and NG ± 0.04. Thus, NG is nearly

perfectly oxygen-balanced; PETN is only slightly negative; but TNT is very

negative, meaning of TNT and AN have been employed to provide additional

oxygen for the excess fuel, as, for example, in the Amatols developed by the

British in World War I.

Modern commercial explosives react in a very rapid and characteristic

manner referred to as a detonation. Detonation has been defined as a

process in which combustion –induced, supersonic shock wave propagates

through a reactive mixture or compound. This high pressure shock wave

compresses and interacts with the reactive material it contacts, resulting in

very rapid heating of the material, initiation of chemical reaction, and

liberation of energy. This energy, in turn, continues to drive the shock wave,

thereby sustaining the detonation. Pressure in a detonation shock wave may


31

reach millions of pounds per square inch. Once initiated, molecular

explosives also have steady-state detonation velocities, but these velocities

are more variable than those of molecular explosives and are influenced by

such factors as diameter of the charge, temperature, density, and

confinement.

14. MODERN DEVELOPMENT

Soon after the advent of porous AN prills, introduced in the early 1950s,

investigators realized that these prills could readily absorb just the right

amount of FO to produce an oxygen-balanced mixture that was both an

inexpensive and effective blasting agent, in addition to being safe and simple

to manufacture. This technology was widely adopted and soon constituted

85% of the industrial explosives produced in the United States [17]. With

ANFO’S cost and safety characteristics, it became practical for surface

miners to drill larger boreholes and to utilize bulk ANFO delivery systems.

Nevertheless, ANFO had two significant limitations: AN is very water soluble,

so wet boreholes readily deactivated the explosives; and ANFO’s low density

of 0.85% g/cc limited its bulk explosive strength. Cook [18] hit upon the

idea of dissolving the AN in a small amount of hot water, mixing in fuels

such as aluminum powder, sulfur, or charcoal, and adding a thickening

agent to gel the mixture and hold the slurried ingredients in place. As this

mixture cooled down, the AN salt crystals would precipitate, but the gel

would preserve the close contact between the oxidizer and the fuels,

resulting in a detonable explosive. Other oxidizers also could be added, and

the density could be adjusted with chemical foaming agents to vary the bulk

explosive strength of the product. With the addition of cross linking agent,

the slurry could be converted to a semisolid material, generally called a

water gel, having some water resistance. The latest significant development

in industrial explosives actually was invented only a few years after slurries.

Water-in-oil emulsion explosives involve essentially the same ingredients


32

that slurry composite explosives do, but in a different physical form.

Emulsion explosives are discussed fully under the section titled “Explosives

Manufacturing and Use” The main developments in military types of

explosives since World War II have been trends toward the use of plastic

bonded explosives (PBXs) and the development of insensitive high

explosives. Driving these trends are desires for increased safety and

improved economics in the process of replacing aging TNT-based munitions

and bomb fills. PBXs involve the coating of fine particles of molecular

explosives such as RDX and HMX (1,3,5,7-tetranitro-1,3,5,7-

tetrazacyclooctane) with polymeric binders and then pressing the resultant

powder under vacuum to give a solid mass with the desired density. The

final form or shape usually is obtained by machining. Explosives such as

triminotrinitrobenzene (TATB), nitroguanidine [21], and hexanitrostilbene

(HNS) are of interest because of their high levels of shock insensitivity and

thermal stability. The development of new, explosive compounds and

compositions is an ongoing area of research, including interest in composite

explosives similar to those used by industry. Examples are Eak, a eutectic

mixture of ethylenediamine dinitrate, AN, and potassium nitrate, and

nonaqueous hardend or cast emulsion-based mixtures.

15. CLASSIFICATION OF EXPLOSIVES

The original classification of explosives separated them into two very general

types: low and high, referring to the relative speeds of their chemical

reactions and the relative pressures produced by these reactions. This

classification still is used, but is of limited utility because the only low

explosives of any significance are black powder and smoke-less powder. All

other commercial and military explosives are high explosives.

High explosives are classified further according to their sensitivity level ore

ease of initiation. Actually sensitivity is more of a continuum than a series of


33

discrete levels, but it is convenient to speak of primary, secondary, and

tertiary high explosives. Primary explosives are the most sensitive, being

readily initiated by heat, friction, impact, or spark. They are used only in

very small quantities and usually in an initiator as part of an explosive train

involving less sensitive materials, such as in a blasting cap. They are very

dangerous materials to handle and must be manufactured with the utmost

care, generally involving only remotely controlled operations, Mercury

fulminate, used in Nobel’s first blasting cap, is in this category, as is the

more commonly used lead azide. On the other end of the spectrum are the

tertiary explosives that are so insensitive that they generally are not

considered explosives.

By far the largest grouping is secondary explosives, which includes all of the

major military and industrial explosives. They are much less easily brought

to detonation than primary explosives and are less hazardous to

manufacture. Beyond that, however, generalizations are difficult because

their sensitivity to initiation covers a very wide range. Generally, the military

products tend to be more sensitive and the industrial products less

sensitive, but all are potentially hazardous and should be handled and

stored as prescribed by law. Table 2.1 lists some of the more prominent

explosives of each type, along with a few of their properties.

For industrial applications, secondary explosives are subdivided according

to their initiation sensitivity into two classes: Class 1.1 and Class 1.5 Class

1.1 explosives are sensitive to initiation by blasting cap and usually are used

in relatively small-diameter applications of 1-3-in. boreholes. Class 1.5

(historically known as blasting agents) are high explosives that are not

initiated by a Standard #8 electric blasting cap under test conditions defined

by the U.S. Department of Transportation (DOT) and that pass other defined

tests designed to show that the explosives is “so insensitive that there is very

little probability of accidental initiation to explosion or of the transition from


34

deflagration to detonation” [27]. Being less sensitive, blasting agents are

generally used in medium- and large-diameter boreholes and in bulk

applications. Dynamites are always Clas 1.1, but other composite explosives

made from mixtures of oxidizers and fuels can be made either Clas 1.1 or

1.5, depending upon the formulation and the density. Density plays a

significant role in the performance of most explosives, and this is especially

true for commercial, composite explosives such as slurries and emulsions

where the density may be adjusted by air incorporation, foaming agents or

physical bulking agents, irrespective of the formulation. The 1.5 explosives

(blasting agents) are of interest because regulations governing

transportation, use, and storage are less stringent than for Class 1.1

explosive. (Propellants and fireworks are classified by the DOT as Class 1.2

or 1.3 explosives, and blasting caps and detonating cord as Class 1.4).

16. STRUCTURAL CHARACTERISTICS OF EXPLOSIVES

The number of potentially explosive compounds is virtually unlimited. A

listing by the U.S Bureau of Alcohol, Tobacco and Firearms of explosive

materials under federal regulation [28] numbered 225, and many of the

items listed were broad, general categories. The ten-volume Encyclopedia of

Explosives and Related Items compiled by the U.S Army Picatinny Arsenal

over a 25-year period contains several thousand entries. New organic

molecular explosives are still being synthesized; composite explosives, such

as current commercial products that are mixtures of oxidizers and fuels,

present an infinite number of possible combinations. The complexity of

trying to comprehensively list the chemical structures of explosives is shown

by a 1977 reference that listed 13 separate categories just for primary

explosives [29]. However; the majority of the most important explosives can

be grouped into a few classes sharing common structural features that are

of value to researchers in understanding and predicting explosive properties.


35

The following seven categories, updated to include the relatively recent

fluoroderivatives, appear to be the most encompassing, Manny explosives

may contain more that one category, but not every compound that contains

one of these chemical groups is necessarily an explosive.

-NO -N-N-, -NN-, and -N≡N -C≡N- and -C≡N -C≡C- -CIO -N-X, where X = C1, F, I -O-O-

Table 2.1 Some properties of common explosives

Common name Symbol Composition Molecular weight

Density (g/cc)

Detonation velocity (km/s)

Detonation pressure

(k/bar)

Explosive energy (cal/g)

Primary explosives

Mercury fulminate Hg(CNO2) 284.7 3.6 4.7 220 428

Lead azide Pb(N3)2 291.3 4.0 5.1 250 366

Silver azide AgN3 149.9 5.1 6.8 - 452

Lead styphnate C6H(NO2)3O2Pb 468.3 2.5 4.8 150 368

Mannitol hexanitrate ( nitromannite)

MHN C6H8(ONO2)6 452.2 1.7 8.3 300 1,420

Diazodinitrophenol DDNP C6H2N4O5 210.1 1.5 6.6 160 820

Tetrazene C2H8N10O 188.2 1.5 - - 658

Secondary Explosive

Nitrogleycerin NG C3H5(ONO2)3 227.1 1.6 7.6 253 1,480

Pentaerythritol tetranitrate

PETN C(CH2ONO2)4 316.2 1.6 7.9 300 1,510

Trinitrotoluene TNT CH3C6H2(NO2)3 227.0 1.6 6.9 190 900

Ethyleneglycol dinitrate

EGDN C2H4(ONO2)2 152.1 1.5 7.4 1,430

Cyclotrimethylenetrinitramine ( hexogen or cyclonite)

RDX C3H6N3(NO2)3 222.1 1.6 8.0 347 1,320


36

Cycloetramethylenetetanitraminem (Octogen)

HMX C4H8N4(NO2)4 296.2 1.9 9.1 393 950

Trinitrophenylmethylnitramine (tetryl)

(NO2)3C6H2N(CH3)NO2

287.2 1.4 7.6 251 721

Nitroguanidine NQ CH4N3NO2 104.1 1.6 7.6 256 1,188

Nitromethane NM CH3NO2 61.0 1.1 6.2 125 950

Nitrocellulose NC Variable - 1.4 6.4 210 829

Triaminotrinitrobenzene

TATB C6H6N3(NO2)3 258.2 1.8 7.9 315 993

Diaminotrinitrobenzene

DATB C6H5N2(NO2)3 243.2 1.6 7.5 259 948

Ethylenediamine dinitratc

EDDN C2H10N4O6 186.1 1.5 6.8 - 1,080

Ethylenedinitramine(haleite)

EDNA C2H6N2(NO2)2 150.1 1.5 7.6 266 1,000

Picric acid C6H3O(NO2)3 229.1 1.7 7.4 265 800

Common name Symbol Composition Molecular weight

Density (g/cc)

Detonation velocity (km/s)

Detonation pressure

(k/bar)

Explosive energy (cal/g)

Ammonium picrate (explosive D)

C6H6NO(NO2)3 246.1 1.6 6.9 - 1,070

Picramide C6H6N(NO2)3 228.1 1.7 7.3 - 1,100

Hexanitrostilbene HNS [C6H3C(NO2)3]2 450.2 1.7 7.1 200 950

TACOT-Z C12H4N8O8 388.2 1.6 7.2 181 675-1,090

Azobishexanitrobiphenyl

ABH C24H6N14O24 874.4 1.8 7.6 - 800

Dinitrotoluene DNT CH3C6H3(NO2)2 182.1 1.5 5.0 - 600-1,1200

Composition B 49/50/1 TNT/PDX/wax

- 1.7 8.0 294 700-1,200

Pentolite 50/50 TNT/PETN

- 1.6 7.7 245 700-1,100

Amatol 50/50/ TNT/AN

- 1.6 6.5 - 755-815

Dynamite Variable NG and various oxidizers and fuels

- 0.8-1.6

1.8-7.6 30-160

Prilled AN-fuel oil ANFO 94/6 AN/FO - 0.8-0.9

1.5-4.0 (Depends

on diameter)

-

Slurries or water gels

Variable mixture of oxidizers, fuels, and water

- 0.9-1.4

3.5-5.0 -


37

Emulsions Variable solutions of oxidizers in water and fuels

- 0.9-1.4

4.5-6.0

-

Heavy ANFO 50-75% AN with 50-25% emulsion

- 1.1-1.3

4.0-4.5 4.0-4.5

Tertiary explosives

Mononitrotoluene MNT CH3C6H4NO2 137.1 1.2

Ammonium perchlorate

AP NH4C1O4 117.5 1.9 3.4 187 488

Ammonium nitrate AN NH4NO3 80.1 1.4 3.2 - 346

Category 1 is by far the largest. It includes intro groups, both aliphatic and

aromatic; nitrate esters; nitrate salts; nitramines; and nitrosamines. Nearly

all of the explosives listed in Table 2.1 fall into this category. Prominent

examples are: nitromethane, an aliphatic nitro compound; TNT, an aromatic

nitro compound; NG and PETN, nitrate esters; EDDN and ammonium

nitrate, nitrate salts; and RDX and HMX, nitramines. Category 2 represents

the hydrazine, azo, diazo and azide compounds, both organic and inorganic.

Hydrazine, tetrazene, and lead azide are examples of this group. Category 3

is represented by the explosives mercury fulminate and cyanogen,

respectively. Acetylene and metallic acetylide salts constitute category 4.

Category 5 consists mainly of inorganic and organic ammonium salts of

chloric and perchloric acid, but would also include various chlorine oxides.

Category 6 is generalized to include most of the amine halogens, nitrogen

triiodide being a classic example. Also, considerable synthetic work has

focused on interesting the energetic difluoroamine groups into various

organic molecules to form explosives that fall into this category. Category 7

includes organic peroxides and ozonides as well as hydrogen peroxide itself.

Commercial industrial explosives such as dynamites, slurries, and

emulsions are included in these categories because their major components,

nitrate esters and nitrate and perchlorate salts, are listed. However,

mixtures of fuels and oxygen or other gases that may be explosives at


38

certain ratios are not covered, including the liquid oxygen explosives that

saw limited application earlier in the twentieth century.

16.1 TNT

TNT is no longer manufactured in certain couthe United States for either

commercial or military use. It is produced commercially in other countries

and is imported into the United States for use in cast boosters to initatate

industrial blasting agents. TNT use in military applications is still

significant, but starting to decline as other higher technology munition and

bomb fills come into use. Current TNT military needs are supplied from off-

shore sources as well as through recycling from demilitarized weapons. A lot

of the demilitarized TNT is also used commercially in the U.S. in cast

boosters. In a relatively straightforward process, TNT is made by the direct

tri-nitration of toluene with nitric acid. Most modern processes are set up for

continuous production in a series of nitrators and separators with nitrating

acid flowing counter-currently. This procedure avoids having to isolate the

intermediate mono-and dinitration products and may also employ

continuous purification and crystallization, being carried out simultaneously

with production.

Mixed nitric and sulfuric acids sometimes are used with the addition of SO3

or oleum. The sulfuric acids or oleum helps drive the reaction to completion

by removing the water produced by nitration and by dehydrating nitric acid

to form the more reactive nitronium ion (NO2+). Because toluene is not very

soluble in the acid, powerful agitation is required. The spent acid is removed

in successive separation steps, and the sulfuric acid is reused after the

addition of more nitric acid. The molten TNT product is purified with

multiple water and sodium sulfate washes. Which produce significant

quantities of “yellow water” and “red water” waste streams, respectively that

must be properly handled to avoid environmental problems. The low melting


39

point of TNT (80-82°C) is ideal for melt casting, and TNT usually is employed

as a mixture with other higher –melting explosives such as PETN, RDX,

HMX, and Tetryl. This feature and the excellent chemical stability of TNT

have made it, historically, the most popular and widely used military

explosive in the world.

16.2 RDX and HMX

Both RDX and HMX are cyclic nitramines made by nitrolysis of

hexamethylene tetramine (HMT). Their good thermal stabilities, high melting

points (7200°C), and high energy properties make these crystalline

compounds popular as projectile and bomb fills and for use in cast boosters

and flexible, sheet explosives. HMX has superior detonation properties and a

higher melting point than RDX, but it is more difficult and more expensive

to manufacture. Reaction I shows the formation of RDX by the action of

nitric acid on HMT. Schematically, RDX formation can be pictured as

nitration of the three “outside” nitrogen atoms of HMT (in more accurate,

three-dimensional representations all four nitrogens are equivalent) with

removal of the “inside” nitrogen and methylene (-CH2-) groups. AN (NH4NO3)

and formaldehyde (CH2O) are produced as by-products, but can be used to

form more RDX with the addition of acetic anhydride, as shown in Reaction

2. In actual practice these two reactions are run simultaneously, as shown

in the combined reaction to produce approximately 2 mol of RDX for each

mole of HMT.


40

HMX was discovered as an impurity produced in the RDX reaction. It is

composed of an eith –membered ring rather than the six-membered ring of

RDX. The latter is more readily formed that the eight-membered ring, but

with adjustment of reaction conditions (lower temperature and different

ingredient rations), HMX formation can be favored, Schematically, its

formation can be pictured by nitration of all four nitrogens in HMT and

removal of two methylene groups as indicated in Reaction 3. To obtain pure

HMX, the RDX “impurity” must be removed by alkaline hydrolysis or by

differential solubility in acetone.

Reaction 2 3CH2O + 3NH4NO3 + 6(CH3CO)2O 2(CH2 . N . NO2)3 + 12CH3COOH Combined Reaction (CH2)6 N4 + 4HNO3 + 2NH4NO3 + 6(CH3CO)2


41

2(CH2 . N . NO2)3 + 12CH3COOH Reaction 3

16.3 HNS (2, 2, 4, 4, 6, 6-Hexanitrostilbene)

This explosive was prepared unequivocally for the first time in the early

1906s [32, 33]. It is of interest primarily for two reasons (1) its high melting

point (316°C) and excellent thermal stability, and (2) its unique crystal –

habit-modifying effects on cast TNT. The former makes HNS useful in

certain military and space applications as well as in hot, very deep wells,

and the second property is used to improve TNT castings. It can be

manufactured continuously by oxidative coupling of TNT as shown below.

This relatively simple process from readily available TNT and household

bleach (5% NaOC1 solution) has been shown to involve as series of

intermediate steps that give HNS in only low to moderate yields (30-45%)

with many by-products. Although it also involves the use of expensive

organic solvents that must be recovered, this synthesis is used commercially

[34, 35].

16.4 TATB (1,3,5-Triamino -2,4,6 Trinitrobenzene)

This highly symmetrical explosive molecule has even higher thermal stability

than HNS (greater than 400°C) and has ongoing interest because of its

extreme insensitivity [36-38]. Because accidental initiation is highly

unlikely, TATB has been used in nuclear warheads and has been explored

for use in plastic bonded systems for a number of military and space

applications. It is manufactured in large-scale batch processes that are little


42

changed from its original synthesis over 100 years ago. The two-steps

process involves tri-nitration of trichlorobenzene followed by amination to

displace the chlorine groups as shown below.

Both steps require high temperature and considerable reaction time but give

80 -90% yields. The major problem areas are chloride impurities in the final

product and the excessively fine particle size of the final product and the

excessively fine particle size of the final product. Because TATB is highly

insoluble in most solvents, it is difficult to purify the product or to change

its particle size by recrystallization. Also the starting material is expensive

and not very readily available. More recently, a similar synthetic procedure

starting with 3, 5 dichloranisole was reported.

16.5 DDNP

This yellow –to-brown crystalline material (melting point 188°C) is a primary

explosives used as the initiator charge in electric blasting caps as an

alternative to lead azide. It is less stable than lead azide but much more

stable than lead styphnate and is a stronger explosive than either of them

because it does not contain any metal atoms. 2-Diazo-4, 6 dinitrophenol


43

(DDNP) is also characterized as not being subject to dead pressing (tested at

pressures as high as 130,000 psi). It was the first diazo compound

discovered (1858) and was commercially prepared in 1928. It is manufacture

in a single-step, batch process by diazotizing slurry of sodium picramate in

water

The structure shown in this reaction is convenient for visualization

purposes, but DDNP actually exists in several tautomeric forms as shown

below with form (2) apparently predominating. The sodium picramate

starting material is itself explosive, but is commercially available as a

chemical intermediate. It can be made by the reduction of picric acid with

reducing agents such as sodium sulfide. The key to making useful DDNP is

to control the rate of diazotization so that relatively large, rounded crystals

area formed instead of needles or platelets that do not flow or pack well.


44

16.6 PETN

Although known as an explosive since 1894, PETN was used very little until

after World War I when the ingredients to make the starting material became

commercially available. The symmetrical, solid alcohol starting material,

pentaerythritol, is made from acetaldehyd formaldehyde, which react by

aldol condensation under basic catalysis followed by a crossed Cannizzaro

disproportionation to produce the alcohol and formate salt. Although the

reaction takes place in a single mixture, it is shown below in two steps for

clarity.

For PETN manufacture the pentaerythritol starting materialcan be readily

purchased as a commodity chemical from commercial suppliers. The

nitration is relatively simple, involving only nitric acid (96-98%) and the

solid alcohol added slowly with mixing and cooling. PETN is not very soluble

in nitric acid or water and is readily filtered directly from the acid or after


45

dilution of the acid with water. Water washing and recrystallization from

acetone-water mixtures give the desired particle size ranges and the desired

purity. PETN can be made either batch wise or continuously for large-scale

production.

Pure PETN is a white, crystalline solid with a melting point of 141.3°C.

Because of its symmetry, it is said to heave higher chemical stability than all

other nitrate esters [40]. Relatively insensitive to friction or spark initiation,

PETN is easily initiated by an explosive, shock and has been described

overall as one of the most sensitive, non-initiating, military explosives [41].

As with most explosives, the detonation velocity of PETN varies with the bulk

density of the explosive. Most military applications of PETN have been

converted to RDX because of its greater thermal stability. However, in

industry PETN is widely used as a major component in cast boosters for

initiating blasting agents, as the explosive core in detonating cord, and as

the base load in detonators and blasting caps. For safety in handling, PETN

in cloth bags immersed in water-alcohol mixtures and dried just before use.

16.7 NG (NITROGLYCERIN OR GLYERCOL TRINITRATE)

This nitrate ester is one of only a very few liquid molecular explosives that

are manufactured commercially. It is a clear, oily liquid that freezes when

pure at 13°C. As seen in the historical section, the first practical use of NG

was in dynamites, where it is still used today more than 140 years later. It

also is used as a component in multi-based propellants and as a medicine to

treat certain coronary ailments. This latter usage is attributed to NG’s ability

to be rapidly absorbed by skin contact or inhalation into the blood, where it

acts as a vasodilator. (At high exposure levels such as in dynamite

manufacture and handling, this property is responsible for the infamous

powder headache.) NG is undoubtedly the most sensitive explosive

manufactured in relatively large quantities. Its sensitivity to initiation by


46

shock, friction, and impact is very close to that of primary explosives, and

extreme safety precautions are taken during manufacture. Pure glycerin is

nitrated in very concentrated nitric and sulfuric acid mixtures (typically a

40/60 ratio), separated from excess acid, and washed with water, sodium

carbonate solution, and water again until free from traces of acid or base,

Pure NG is stable below 50°C, but storage is not recommended. It is

transported over short distances only as an emulsion in water or dissolved

in an organic solvent such as acetone. Traditionally, it is been made in large

batch processes, but safety improvements have led to the use of several

types of continuous nitrators that minimize the reaction times and

quantities of explosives involved. Because of its sensitivity, NG is utilized

only when desensitized with other liquids, combined with absorbent solids

or compounded with nitrocellulose.

17. PACKAGED EXPLOSIVES

Packaged explosives dominated the explosives market from the time

dynamite was invented in 1867 until the middle of the twentieth century. At

that time, other composite type packaged technologies began appearing,

particularly in the 1960s and 1970s, concurrent with increasing market

penetration of bulk explosives. During those years of rapid technology

development, two packaged product types emerged, first water-gel

explosives, followed in the 1970s and early 1980s by packaged emulsion

explosives. All of these packaged product technologies remain active in parts

of the world as of the date of this edition, but in the U.S. emulsions are pre-

eminent followed by dynamites. Both are discussed below.

17.1 DYNAMITE

Dynamite is not a single molecular compound, but a mixture of explosive

and non-explosive materials formulated in cylindrical paper or cardboard

cartridges for a number of different blasting application. Originally Nobel


47

simply absorbed NG into kieselguhr, an inert diatomaceous material, but

later he replaced that with active ingredients- mixtures of finely divided fuels

(including absorbent combustibles) and oxidizers called dopes. Thus, energy

is derived not only from the NG, but also from the reaction of oxidizers such

as sodium nitrate with the combustible.

The manufacture of dynamite involves mixing carefully weighed proportions

of NG and various dopes to the desired consistency and then loading

preformed paper shells through automatic equipment. Because dynamites

represent the most sensitive commercial products produced today, stringent

safety precautions such as the use of non-sparking and very-little-metal

equipment, good housekeeping practices, limited personel exposure, and

barricaded separation between processing stations are necessary for

manufacturing. Today, the “NG” used in dynamite is actually a mixture of

EGDN (ethylene glycol dinitrate) and NG formed by nitrating mixtures of the

two alcohols (ethylene glycol and glycerin), in which NG is usually the minor

component. Table 2.2 lists the common general types of dynamites with

their distinguishing features. The straight dynamites and gelatins largely

have been replaced by the ammonia dynamites and ammonia gelatins for

better economy and safety characteristics.

The used of NG-based dynamite continues to decline throughout the world.

For example, by 1995 there was only one dynamite manufacturing plant left

in North America, and in 2010 the dynamite production at this plant had

dropped to about one-third the amount produced in 1990. The reasons for

the declining use of dynamite are its unpopular properties of sensitivity to

accidental initiation and the headache-causing fumes. However (despites its

higher cost), for some difficult blasting application it remains the product of

choice due to its high density and high energy and sensitivity.


48

Table 2.2 General types of dynamite

1. Straight dynamite Grannular texture with NG the major source of energy

2. Ammonia dynamite (“extra” dynamite)

AN replacing part of the NG and sodium nitrate of the straight dynamite

3. Straight gelatin dynamite Small amount of nitrocellulose added to produce soft to tough rubbery gel

4. Ammonia gelatin dynamite (“extra” gelatin)

AN replacing part of the NG an sodium nitrate of the straight gelatin

5. Semigelatin dynamite Combination of types 2 and 4 within-between properties

6. Permissible dynamite Ammonia dynamite or gelatin with added flame retardant

18. PACKAGED EMULSIONS

The first major non – dynamite package product technology was water gels.

These were composite mixture of hot, concentrated, dissolved oxidizer

solution slurried with other ingredients including soluble and insoluble

liquid fuels, particulate fuels, solid oxidizer salts, and thickeners consisting

of natural and synthetic water soluble polymers. The thickened solution

could be gelled by addition of crosslinking agents, typically inorganic salts of

certain metals. The gelled structure gave significant, but not excellent water

resistance.


49

Fig. 2.3 Commercial packaged emulsion cartridges

(Courtesy Dyno Nobel)

Packaged water-in-oil emulsions are made from essentially the same oxidizer

solutions as used for water-gel products, but are fueled with an organic

external liquid hydrocarbon phase that provides much better water

resistance. These products are basically formed with the same

manufacturing equipment as the bulk emulsions (see next section). The fuel

component usually contains waxes and other thickeners to give the

emulsions a thick, putty-like consistency, and the oxidizer salt to produce

optimum energy, physical stability, and improved after-blast fumes. After

manufacture, the hot, thick emulsion is extruded into packaging material,

normally a plastic film. The final product is then clipped together with metal

clips forming firm, sausage-like chubs and is cooled which further thickens

the product. Some packaged emulsions are also available in paper

cartridges, designed to simulate dynamite packaging. To obtain reliable

detonability in small diameters, the density of packaged emulsions must be

maintained at a relatively low value, typically 1.10-1.20 g/.cc. On the other

hand, some dynamites are available with densities in excess of 1.40 g/cc.


50

Figure 2.3 shows some commercial packaged emulsion cartridges in both

plastic and paper wrappings.

19. BULK EXPLOSIVES

19.1 AMMONIUM NITRATE AND ANFO

Ammonium nitrate (AN) continues to be the most widely used component of

commercial explosives. It is used in nearly all of the packaged and bulk

explosives on the market. Ammonia is the main raw material needed to

manufacture AN. Some of the AN manufacturers make their own ammonia

and some purchase it on the open market. It is obvious that the cost of

manufacturing AN will depend on the price of ammonia and, even more

basically, natural gas from which ammonia is made. Figure 2.4 shows the

volatility in the cost of explosives, although significant natural gas

development in the U.S. over the past few years seems to be suppressing

some of the price fluctuation.

There are many producers of AN in North America making both AN solution

and explosive-grade AN prills. The AN solution is used in the manufacture of

packaged and bulk emulsion and water-gel explosives, and explosive-grade

AN prills are used to make ANFO, a composite explosive described in an

earlier section. These low density AN prills are made by a specialized

process, in which internal Voids are created, making the prills porous and

able to absorb the required 5.5-6% FO. ANFO remains the largest volume

commercial explosive in use today around the world. Because of this, ANFO

is commonly used as a reference when defining and comparing explosive

properties. Some of these important explosive properties include density,

detonation velocity, and energy release.

The crystal density of AN is about 1.72g/cc, and the particle density of

explosive-grade AND prills ranges from about 1.40 to 1.45 g/cc depending


51

upon the manufacturing process. This difference in crystal and particle

density reveals the volume of pores or voids created by the specialized

prilling process. The porosity of AN prills is the property desired in the

manufacture of ANFO, since this determines how much FO can be absorbed.

This intimate mixture of AN with FO is critical to the efficient detonability of

ANFO. The AN prill particle density and inherent void-space value also

become important when predicting and calculating densities of ANFO blends

with water-gel and emulsion explosives.

Fig. 2.4 volatility of natural gas prices from 2004 to 2010, New York.

The bulk density of ANFO ranges from about 0.80 to 0.87g/cc. So, clearly

about half of the ANFO is air or void space. Most commercial explosives

require a certain amount of entrained void space in order to detonate

optimally. These void spaces play a major role in the detonation reaction by

creating reaction sites, “hot spots,” under adiabatic compression and

interaction with the shock wave in the detonation front [42]. The amount of

void space in any given explosive and the resultant change in density have


52

significant impact on the detonation properties like detonation velocity,

sensitivity, and even energy release.

Generally speaking, the detonation velocity of a composite explosive will

increase with density until a failure point is reached. This failure point is

commonly referred to as the critical density of that particular explosive in

that particular diameter or other configuration. The density at that point is

so high and the void space so low that the detonation cannot be sustained

and failure occurs.

Other important parameters that affect the detonation velocity and

performance of ANFO are charge diameter and confinement. The detonation

velocity of commercial composite explosives such as ANFO will drop

dramatically as the diameter of the charge nears the failure diameter. At a

given density, the diameter below which a charge fails to propagate a

detonation is called the critical diameter. As a charge diameter is increased

from this point, the detonation velocity increases asymptotically to a

maximum value for that particular explosive formulation. The detonation

velocity of ANFO will also increase noticeably when the charge confinement

is changed, such as from PVC tube to a Schedule 40 steel pipe. A summary

of test data on ANFO velocity vs. confinement and diameter is shown in Fig.

2.5. The temperature of composite explosives also affects sensitivity and

critical properties.


53

Fig.2.5 Detonation velocity of ANFO vs. diameter and confinement

The basic chemical reaction of ANFO can be described with following equation:

3NH4NO3 + CH2 CO2 + 7H2O + 3N2 + 880 cal/g

Using CH2 to represent FO is generally accepted, but it really is an over-

simplification, since it is a mixture of hydrocarbons. The heat energy release

of 880 cal/g is the theoretical maximum value based upon the heats of

formation of the reactants and products. Of course, all of the products of

detonation are gases at the detonation temperature of about 2,700 °K.

The theoretical work energy that is released from an explosive reaction can

be calculated using a variety of equations of state and computer programs.

Explosive energy can also be measured by a variety of techniques including

underwater detonation of limited size charges with concurrent

measurements of the shock and bubble energies. Each explosive

manufacturer has an energy measurement and equation of state that is

used to calculate and report their product properties. This often leads to

confusion and controversy when explosive consumers try to compare

product lines when given only technical information sheets. Since theoretical

calculations must of necessity be based on a number of assumptions, the


54

most valid comparisons are done in the field with product testing and

detailed evaluation of results.

20. BULK EMULSIONS

During the past 50 years, the commercial use of bulk delivered explosives,

including ANFO and other composite explosives, has continued to increase

while the use of packaged explosives continued to fall on a percentage basis.

Due to low cost and superior sensitivity compared to earlier water-resistant

technologies such as water gels, bulk emulsions have been leading this

trend over the last 25-30 years. Bulk products initially became very popular

in large volume open-pit mining operations and this accelerated with the

advent of bulk emulsion products and blends with ANFO. More recently,

bulk emulsions have increasingly replaced packaged products in

underground mining and particularly in quarry operations.

Emulsions are made by combining an oxidizer solution and a fuel solution

using a high-shear mixing process. The oxidizer solution is normally 90-95%

by weight of the emulsion. It contains AN, water, and sometimes a second

oxidizer salt such as sodium or calcium nitrate. The solution must be kept

quite hot, since the water is minimized for increased energy, and the

crystallization temperature is typically 50-70°C.

The fuel solution contains liquid organic fuels, such as FO and/or mineral

oils, and one or more emulsifiers. An emulsifier is a surface active chemical

that has both polar and nonpolar ends of the molecule. In the high-shear

manufacturing process, the oxidizer solution is broken up into small

droplets, each of which is coated with a layer of fuel solution. The droplets in

this meta-stable, water-in-oil emulsion are basically held together with the

emulsifier molecules, which migrate to the surface of the dispersed droplets

and form a link between the hydro-phobic external (continuous) oil phase


55

and the hydrophilic phase within the droplets. In today’s explosives

industry, much of the research work is directed towards developing better

and more-efficient, emulsifier molecules that will improve the storage life

and handling characteristics of the bulk and packaged emulsions. The

emulsifiers currently used in commercial explosives range from relatively

simple fatty acid esters with molecular weights of 300-400 to the more

complex polymeric emulsifiers having molecular weights in excess of 2,000.

Figure 2.6 shows a photomicrograph of an emulsion explosive at 400 power

with the typical distribution of the fuel-coated oxidizer solution droplets

(nominally 1-5 µm in diameter). Figure 2.7 shows a bulk emulsion exiting a

loading hose and displaying the soft ice cream-like texture typical of bulk

emulsions. The viscosities of bulk emulsions can range from nearly as thin

as 90 weight oil to thicker than mayonnaise, depending upon the application

requirements. Emulsion viscosity increase with product cooling, but most

emulsions continue to remain stable at temperatures well below 0°C,

dramatically below the crystallization temperature of the oxidizer solution.

The oxidizer solution droplets in the emulsion are therefore held in a

supersaturated state. Over time, the surface layers created by the emulsifier

molecules can become less stable due to chemical degradation (oxidation) of

the emulsifier or other spontaneous physical and chemical process.

Movement of the emulsion is also stressful to these systems, which in reality

are thermodynamically unstable, tending towards crystallization and

coalescence.


56

Fig. 2.6 Photomicrograph of a bulk emulsion

Fig. 2.7 (a) Bulk emulsion loading


57

Fig. 2.7 (b) Bulk emulsion exiting a loading hose

Crystals can begin to form that break though the emulsifier layer and the

emulsion begins to “break down” and further crystallize, thereby

diminishing some of the desirable physical and detonation properties. For

this reason emulsion compositions must be optimized for a particular

application in terms of product stability and usable storage life.

The intimate mixing of oxidizer and fuel in emulsions gives these explosives

much high detonation velocities when compared to ANFO. For example, in

150 mm diameter PVC, ANFO at a density of ~0.82 g/cc has a velocity of

about 4,000 m/s, and a sensitized emulsion would have a velocity closer to

6,000 m/s at a density of 1.20 -1.25 g/cc.

Also, the layer of oil surrounding each oxidizer solution droplet protects the

emulsion from extraneous water intrusion and subsequent deterioration of

the explosive. Many studies have shown that when mining operations use

emulsion explosives rather than ANFO, which has basically no water


58

resistance, the amount of nitrate salts in mine ground water is reduces

considerably. This can be a very important factor in today’s environmentally

conscious mining and explosives industry.

Bulk emulsions are generally non-detonable per se and are typically

sensitized with some type of density control medium to become usable

blasting agents. These voids interact with the shock wave, creating “hot

spots,” which are required to sustain the detonation front. The two most

commonly used density control methods are hollow solid microspheres and

gas bubbles created by an in situ chemical reaction. Both glass and plastic

hollow microspheres are commercially available and used by explosives

manufacturers. The in situ chemical gassing techniques require

considerably more expertise and generally utilize proprietary technology.

Emulsion blended with ANFO, with its internal and interstitial air voids, is

also widely used as ales efficient but effective sensitizer in larger diameter

boreholes.

Fig. 2.8 (a) Underground pressure vessel loader. (b) Compressed-air,

powered pumper for underground bulk emulsion (Courtesy Dyno Nobel)

20.1 UNDERGROUND BULK EMULSIONS


59

In the past decade the use of sensitized bulk emulsions has continued to

increase in underground mining. In addition to the necessary small

diameter sensitivity of emulsion based explosives in these applications,

much of this trend has been due to t he development of innovative loading

equipment and techniques. One example of this is shown in Fig. 2.8 a,

which shows a small-volume pressure vessel that can be used for

development and tunnel rounds utilizing horizontally drilled boreholes. The

bulk emulsion blasting agent is pressurized inside the vessel and literally

squeezed through the loading hose into the boreholes. A continuous column

of explosives is assured by inserting the loading hose to the back of the hole

and extracting it as the product is loaded. A more recently designed air-

powered pumper unit for underground delivery of bulk emulsion is shown in

Fig. 2.8 b. This unit typically uses compressed air, available in most

underground mines, to power a pump that delivers the bulk explosive. Much

more complex underground loading unit are available for loading bulk

emulsion into boreholes drilled at any angle to the horizontal from straight

up to straight down.

The emulsion explosives used in these specialized loading units were

specifically designed for underground use nearly 30 years ago and remain

essentially the same today. They are manufactured in central plants, either

sensitized with microballoons to form as a 1.5 explosive or as gassable

matrix that can be transported and stored as a non-explosive, thence

sensitized with gas bubbles at the point of use. They have been successfully

used in underground mining operations around the world. The fuel and

oxidizer contents are carefully balanced, and this, combined with excellent

water resistance and detonation efficiency, results in the near elimination of

after-blast toxic fumes, such as CO, NO, and NO2. The fume characteristics

of this product have been shown to be considerably superior to either

dynamite or ANFOR. For example, a series of tests in an underground


60

chamber in Sweden compared the after-blast fumes of this emulsion to

ANFO. The CO was reduced from 11 to less than 6l/kg of explosive, and the

NO plus NO2 was reduced from about 7 to less than 1 L /kg of explosives.

Fig. 2.9 Bulk emulsion loading truck in a Florida quarry

20.2 SURFACE BULK EMULSIONS

Many open-pit quarries also use bulk emulsions for their blasting

operations. As the size of the quarry increases, the size of the explosive

loading trucks also must increase. Truck payloads can range from 5,000 to

30,000 lb of product. Figure 2.9 shows an emulsion pumper truck in a

quarry in south Florida. These particular trucks, with a payload of about

20,000 lb, are specially designed for a site-mixed system, in which each

truck is an emulsion manufacturing unit. Combining non-explosive raw

materials directly on the truck maximizes safety and minimizes requirement

for explosive storage. This particular bulk emulsion is manufactured at a

rate between 300 and 500 lb/min and sensitized to the desired density with

a chemical gassing system as it is loaded into the boreholes.

Figure 2.10 shows Florida blast in progress. Note the ejection of cardboard

tubes from some holes. These tubes must be used in most areas to keep the


61

boreholes from collapsing in the layered, coral limestone formation.

Other surface, bulk delivery trucks use emulsions that are manufactured in

central plants, then shipped to a surface depot, off-loaded into a bin or

overhead silo, and loaded as needed into delivery trucks by pumping or

gravity flow. The products can be shipped pre-sensitized with microballoons

or as a non-sensitized, gas able matrix. The application of the product can

be as a pumped straight emulsion or as a pumped or augured

emulsion/ANFO blend.

Many of the large volume metal and coal mining operations around the

world have both bulk emulsion and AN prills stored either on site or nearby

so that any combination of these two products can be used. Figure 37.12

shows a typical explosive staging area in a large open-pit coal mine in

Wyoming. The explosive truck in the foreground has compartments on board

for emulsion, AN prills, and FO, so any combination of products ranging

from straight emulsion to straight ANFO can be loaded. The truck has a

capacity of about 50,000 lb and can deliver product to the boreholes at up to

a ton per minute. Each borehole can contain as much as 5 tons of explosive,

and some of the blast patterns can contain as much as ten million total

pounds.

The emulsion/ANFO explosive blend selection to be used in any given

mining application depends upon may factors. Typically, ANFO is the least

expensive product, but it also has the lowest density and no water

resistance. As emulsion is added to ANFO, it begins to coat the AN prills and

fill the interstitial voids between the solid particles. This increases the

density, detonation velocity, and water resistance compared to ANFO. The

density increases nearly linearly with percent emulsion from about 0.85g/cc

with ANFO to about 1.32 g/cc with 50/50 blend. This range of emulsion

ANFO blend is commonly referred to as Heavy ANFO. As the density

increases, the amount of explosive that can be loaded into each borehole


62

increases, and either drill patterns based on ANFO can be spread out or

better blasting results can be obtained.

It is commonly accepted in the explosive and mining industry that at least

45-50% emulsion is required to protect the Heavy ANFO blend from borehole

water intrusion. Pumped explosive blends with 60-80% emulsion can be

used for even better water resistance when severe water conditions are

encountered. These products can be pumped through a loading hose, which

can be lowered to the bottom of the borehole and displace the water during

loading. Trucks similar to that shown in Fig. 37.10 can be used for these

products. Most Heavy ANFO products are more simply mixed through an

auger and discharged into the top of boreholes by trucks similar to the one

shown in Fig. 2.11 for Heavy ANFO products, especially those with less than

~45% emulsion, the holes should be either dry or dewatered using pumps.

The basic chemical composition of a typical all-AN oxidizer emulsion

explosive would be: AN plus about 16% water plus about 5% fuels. The fuels

may contain fuel oil, mineral oil, and emulsifiers: the majority of which can

generally be described as CH2 hydrocarbon chains. Therefore, a very

simplified chemical reaction for a basic emulsion is similar to that for ANFO

shown earlier.


63

Fig. 2.10 Florida quarry blast in progress (Courtesy Dyno Nobel)

Fig. 2.11 Typical bulk explosives staging area in a large opt-pit mine

By adding 16% water to the ANFO reaction described earlier, the theoretical


64

heat energy release is reduced from 880 to 680 cal/g. The difference is the

energy price paid for using water due to converting it to steam in the

detonation reaction plus the energy loss in diluting the ANFO with a non-

energy producing additive (water). The advantages and disadvantages of

using ANFO or emulsions begin to become clear. ANFO is easily mixed and

is probably the least expensive form of explosive energy, but it has no water

resistance and has relatively low loading density. Emulsions are

considerably more complicated to formulate and manufacture, but they have

excellent water resistance and more flexibility in terms of density, velocity,

and higher bulk energy to match rock types and blasting applications.

21. HOME-MADE EXPLOSIVES

According to the US Government Hazardous Substances Database, several

substances and mixtures can be used for the realization of this kind of

explosives, starting from chemicals sold in markets and pharmacies. Among

others, two were often adopted for terrorist attacks, suicide bombing, and

other malicious uses: Ammonium Nitrate (AN) − Fuel Oil (i.e. ANFO)

mixtures and Acetone Peroxide or Triacetone Triperoxide Peroxyacetone

(TATP) mixtures(]. These two materials were considered as reference

explosives for the analysis presented in the present study.

ANFO is a tertiary explosive (note that TNT is a secondary explosive) and is

generally composed by 94% of AN prills and 6% of adsorbed fuel oil [32]. It is

extensively used for several authorized purposes as in mine blasting. The

TNT equivalence is typically around 80% and the ideal explosion (detonation)

energy is 3890 kJ/kg (pure ammonium nitrate has an explosion energy of

1592 kJ/kg). AN prills used for mining applications are however physically

different from fertilizer prills used in home-made explosives. The commercial

ammonium nitrate prills used for mine blasting have a 20% void fractions

and are coated with #2 fuel oil (mainly C10−C20 linear hydrocarbons) or


65

kerosene. Hence, ANFO has a bulk density of approximately 840 kg/m3

when starting from AN prills for mining applications, having a density of

about 1300 kg/m3 (the density of pure crystalline ammonium nitrate is

1700 kg/m3). On the other hand, homemade explosives prepared from AN

fertilizers do not have a so high void fraction and are less efficient: e.g. the

new European regulations for fertilizers state that they must contain less

than 45% of AN (16% N) for being traded to the general public. Such

fertilizers still may be used to obtain explosives, but require processing to

achieve a detonation. If commercial AN (containing about 50% of inert, as

dolomite) and diesel fuel is used, a detonation energy of about 1071 kJ/kg is

obtained, much less than pure ANFO [30]. Furthermore, it has been

observed that when amounts of dolomite higher than 30%, are present, no

detonation is observed.

TATP is a primary explosive which is notable since it does not contain

nitrogen. Thus, it is used to avoid conventional chemical bomb detection

systems, and it is almost undetectable by either analytical system or by

sniffer dogs [34]. It can be obtained from common household items such as

sulphuric acid, hydrogen peroxide, and acetone [35].

TATP is very unstable: it can be ignited by touch and can explode

spontaneously. It is often used for improvised detonators itself. It is actually

composed by isomers and conformers, the dimer being more stable but

having lower decomposition energy (see Fig. 3).

The actual components of TATP: (a) dimer and (b) trimer peroxides.The

density of the pure molecule is typically considered to be 1220 kg/m3.

However, home-made TAPT formulations are typically in the range of 450–

500 kg/m3. Finally, TATP is often stabilized with carbonaceous liquids and

waxes so that the net charge is even lower .Nevertheless, Lefebvre et al.have

demonstrated that home-made TATP is a primary explosive and very


66

sensitive to impact or friction, although the strength of explosion may

strongly vary since the quality of the final product is very sensitive to the

temperature during its synthesis.

TATP is highly volatile and decomposes to form large number of gas phase

molecules (entropic explosion) [39] and [40]. Acetone and ozone are

predicted to be the main decomposition products, along with oxygen, methyl

acetate, ethane, and carbon dioxide [41].

Fig. 3. The actual components of TATP:

(a) dimer and (b) trimer peroxides.

22. DEVELOPMENT OF INITIATION SYSTEM

22.1 HISTORY OF INITIATION SYSTEMS

The first reliable initiation system for commercial explosives could probably

be traced back to Alfred Nobel’s invention of the blasting cap in 1864. These

early caps had to be initiated with a strong shock, limiting their safety and

convenience. In 1867, Nobel developed a cap that could be initiated with a

fuse, a more convenient system that allowed an element of timing (length of

the fuse) to the initiation of a blast, and hence, some increase in safety and

predictability. This, combined with his invention of dynamite in 1866 [46],

basically ushered in the modern era of blasting. The fuse cap and dynamite

dominated the emerging blasting industry for several decades. In the


67

century that followed, the initiation systems became more and more

sophisticated and safe. The electric blasting cap which could be initiated by

electrical current was developed around the turn of the century, adding

further safety to blasting. Detonating cord, a flexible cord made of cloth or

plastic with a core load of high explosives-usually PETN, was developed in

Europe around the same time. Strings or circuits of detonating cord could

be utilized to initiate several explosive charges with only one blasting cap.

Electric caps were introduced in 1946 equipped with built-in, variable-delay

elements (pyrotechnics that burn reproducibly) [46]. This allowed long rows

of boreholes to be initiated in a sequential manner, optimizing blasting

effectiveness as well as better controlling ground vibration and air blast

damage to the surrounding. With caps having delays ranging from a few

milliseconds to many hundreds and eventually thousands of milliseconds,

the number of sequentially fired boreholes in a row depended only on the

desired timing and the amperage capability of the electric blasting machine

or other firing circuit. Sequential blasting machines were developed that

could be used to control firing times between a number of rows of

blastholes, greatly expanding the times between a number of rows of

blastholes, greatly expanding the timing possibilities and size of blast

patterns.

Fig. 2.12 Variety of commercial cast boosters (Courtesy Dyno Nobel)


68

23. BOOSTERS

Prior to about 1950s, most of the commercial explosives in the market were

reliably detonable with just a blasting cap or detonating cord as the initiator.

However, the use of non-cap-sensitive explosives (blasting agents) began

emerging into the explosives market in the 1930s and 1940s. These

products really took off with the advent of ANFO in the mid-1950s and water

gels, invented by Melvin A. Cook in 1957 [47], Emulsion based blasting

agent products, which began significant commercialization in the mid-

1970s, became the dominant non-ANFO product by the late 1980s and

remains so to the present day. All of these blasting agents were adaptable to

larger diameter packaged products and particularly to bulk loading.

These blasting agent explosive products were considerably less sensitive

than dynamites and required larger “booster” charges for reliable detonation.

At first, a high density and high velocity dynamite was used as the booster

charge. Later, TNT-based cast boosters came into the market. Cast TNT by

itself is not reliably detonable with a blasting cap or detonating cord, and so

40-60% PETN is normally added to the TNT melt and subsequent cast. The

combination of TNT and PETN is called Pentolite. TNT has a melting point of

about 80°C, which makes it an excellent base explosive for casting into

forms, The military has used this concept for decades for filling bomb

casings. Once the TNT has melted, other material can be added to give the

final cast explosive composition the desired properties. Additives used for

military purpose have included such as aluminum, ammonium nitrate,

RDX, and also PETN, but Pentolite been dominant in commercial cast

boosters. Cast boosters are available in a variety of sizes from about 10 to

800 g, as shown Fig. 2.12, and continue to be predominant today in nearly

all large mining operations and other blasting applications, Cap-sensitive

composite explosives such as emulsion cartridges, composite explosives

such as emulsion cartridges, dynamite, etc. are also used in less demanding


69

situations, Coupled with the flexibility of blasting caps with variable delays

to initiate the booster charge which then initiates the main charge, large

patterns containing many thousands of pounds of explosive can be blasted.

24. NON-ELECTRIC INITIATION

In 1967, Per –Anders Persson of Nitro Nobel AB in Sweden invented a non-

electric initiation system, designated Nonel® , that eventually revolutionized

the explosives industry [46]. The Nonel system consists of an extruded

hollow plastic tube (shock tube) that contains an internal coating of a

mixture of powdered molecular explosive and aluminum. The plastic tube is

inserted into and attached to a specially designed detonator or blasting cap.

The Nonel tubing can be initiated by a number of starter devices, one of

which uses a simple shotgun shell primer. The explosive/aluminum mixture

explodes down the inside of the tube at about 2,000 m/s and will run at this

velocity until all the interconnected tubing reacts, including initiating all the

blasting caps. The tubing is about 3 mm outside diameter and 1 mm inside

diameter and the explosive core load is only about 18 mg/m, not even

enough to rupture the tubing. The Nonel product is not susceptible to the

hazard associated with electric blasting caps wherein premature initiation

by extraneous electric sources can occur. Figure 2.13 is a photograph of

both an electric blasting cap with the two electrical wires and a typical Nonel

unit with the plastic tubing.

During the 1980s and 1990s, Nonel products continued to replace both

electric blasting caps and detonating cord down lines around the world. It

has long been known that detonating cord down lines disrupt and partially

react with blasting agents causing some degree of energy and sensitivity

loss. Also, the use of surface detonating cords to initiate blasts can lead to

noise complaints. Using a range of long-lead delay caps down the borehole,

a surface trunk-line of shock tubing, surface connectors to tie the tubing


70

together, and surface delay elements that are available with this technology,

huge and complex blasting patterns can be laid out and blasted with a

multitude of possibilities as to timing and sequence.

Fig. 2.13 An electric blasting cap showing electrical wires, typical

Nonel unit with plastic tubing

As delay elements were perfected for the Nonel blasting caps, their

application and use grew even further, especially in underground mining

where a large percentage of blasting caps is used.

25. ELECTRIC DETONATORS

The development of detonators with delay periods that were based on

pyrotechnic elements (timing controlled by the burn rate and length of the

element) had a dramatic effect on the explosives industry, enabling much of

the size and complexity of modern blasting, as well as reducing off-site blast

effects caused by ground vibration and air blast. When a blasting pattern is

properly timed, the detonation of individual holes is such that damage from

air blast and ground vibration is greatly reduced. This is a result of


71

destructive wave interference, i.e., the ground vibration waves from

detonation of each hole combine with each other and the air blast waves

from the detonation of each hole combine with each other in such a way that

the resulting wave patterns propagating from the blast have lower

amplitudes and higher frequencies. This is key to controlling off-site

structural damage. Over many years of development delays was improved

dramatically. However, delay times can still vary from detonator to detonator

and from the target delay time by a few milliseconds to tens of milliseconds

depending on the delay period of the detonator. There are also occasional

fliers that occur well outside the normal scatter for a given delay detonator

that can upset the blasting pattern. It was surmised that these variations

could result in less than optimum blasts, considering not just air blast and

ground vibration but also ground movement, fragmentation, ore dilution, fly-

rock, etc.

In the late 1980s and early 1990s, several companies began developing and

testing detonators wherein the delay timing was controlled with an

integrated electronic circuit. These were initially very expensive and, by

today’s standards, were crude devices. However, they did demonstrate the

promise of very accurate caps. As electronics technology improved into the

1990s, testing and field evaluation of such caps increased. However, cost

and complexity in assembling the caps, increased. However, cost and

complexity in assembling the caps, the development of user friendly

controllers to program and initiate them, and the unfamiliar equipment and

technology on the blasting patterns slowed widespread implementation. It

wasn’t until mounting commercial experience began to demonstrate

improved blasting results that the tide began to turn. This in turn resulted

in accelerated improvements in manufacturing technology to make the caps

less expensive, more reliable, improve ease of use, and increase expertise

available to blasters in the field. By the 2000s, the technical case for

electronic detonators had been well established. Despite the significant cost


72

differences that remain between electric or non-electric detonators and

electronic detonators, amounting to factors of perhaps five to ten times more

costly per unit, there are dramatic increases in the use of electronic

detonators in the blasting industry today. As of the date of this edition, they

are still a minor part of the total global detonator market, but the market

share is increasing. Several manufacturing companies with independent

detonator technologies are involved and use is occurring in all segments of

mining, construction, and other commercial blasting applications.

The construction of an electronic detonator has some features in common

across all manufacturers. The timing and programmability features depend

on an integrated circuit. One ore more capacitors are included to store the

electrical energy necessary to run the circuitry and to initiate the detonator

via bridge or igniter elements. The detonator can be built to have a preset

delay time or can be programmable as to delay period at the point of use.

Electrical and electronic communication with the detonators is established

via two or four lead wire connections, depending on the system chosen and

blast design, flexibility, control, and programmability features desired or

needed by the blasters. The lead-wires are connected to the blasting

machine which enables all of the features in the particular technology

chosen, including programming (if needed), arming, and eventually initiating

each line of detonators. The timing accuracy is on the order of only 1 to a

few milliseconds depending on delay times, and variability is typically

reduced by and order of magnitude compared to pyrotechnic detonators.

Figure 2.14 shows the components of a four wired system


73

Fig.2.14 Components of an electronic cap initiating system.

L-R, logger, cap, base station (blasting machine), Connectors

Courtesy Dyno Nobel)

And includes the cap, connectors, tagger (identifies and tags each detonator

as it is placed in the borehole), and blasting machine (herein called the base

station).

References

1. Mineral Information Institute. Soc Metallurg Explor Foundation.

http://www.mii.org/MiiBabyMain

2. United States Geological Survey. Minerals Yearbook, Explosive Statistics and

Information. Compiled Information from years 2000-2009.

3. Apodaca LE (2009) United States Geological Survey. Minerals Yearbook,

Explosive Statistic and Information (advance release).

4. Ornellas J (1968) Phys Chem 72:2390.

5. Kaye SM (1978) Encyclopedia of explosives and related items, PATR 2700, Vol 8.

U.S. Army Armament Research and development Command, Dover, pp 99-100.

6. Engineering design handbook: explosives series. Army Material Command


74

Pamphlet 706-177, AD 764, 340, p 12, distributed by NITS, Jan 1971.

7. Fedoroff BT, Sheffield OE (1962) Encyclopedia of explosives and related items,

PATR 2700, vol 2. Picatinny Arsenal, Dover, p B165.

CHAPTER – 2

EXPLOSIVES: CHEMISTRY AND ENERGETICS

1. Introduction The most generally accepted definition of an explosive is expressed in the

following terms:

“A solid or liquid substance or a mixture of substances which on the

application of a of a suitable stimulus to a small portion of the mass is

converted in a very short interval of time into other more stable even

substances, largely or entirely, gaseous, with the development of heat

and high pressure”

It is obvious that the explosive substances are in a metastable state and are

therefore capable of undergoing a chemical reaction under the influence of

an external stimulus (for e.g. impact, heat or shock) to yield the more stable

products. The reaction products are predominantly gaseous like N2, CO2 and

H2O while other gases like the oxides of nitrogen are also formed sometimes,

it is also necessary that an explosive should be capable of producing this

large quantity of gas under high pressure so rapidly that the surroundings

are subjected to a strong dynamic pressure.

A synoptive view of the entire field of explosive materials is given in Table 1.

2. Classification Explosives While there can be several ways of classifying explosives depending on their

use and composition the following three main types are generally

recongnised.

75

Chapter 2 Explosives : Chemistry & Energetics

1. Deflagrating (or propellant) explosives.

2. High explosives (sometimes called secondary explosives).

3. Initiating explosives (sometimes called primary explosives).

Deflagrating or low explosives were the earliest to be developed. These lead

to an explosion which is really a rapid form of combustion in which the

particles burn at their surface and more of the bulk until all has been

consumed. Typical examples of this category are the blasting powder or

gunpowder, propellants in ammunition, rocket propellants and

pyrotechnics.

High explosives, depending on their composition, detonate at velocities of

5,000 – 25,000 feet per second and produce large volumes of gases at

considerable heat at extremely high pressures.

Table 1 : Explosive Materials and their Application

76


High explosives usually contain chemical components with more or less

unstable molecular structure capable, on detonation of molecular re

arrangement into more stable forms of potential energy. These are generally

gases (solid products tend to produce smoke) which are very high

temperature and pressure. Secondary explosives of this type are military

explosive like TNT, RDX, PETN, Tetryl and other combinations of these and

industrial explosives like nitroglycerine, gelatins, slurries, water gels, ANFO

and other powder explosives. These explosives are normally set off with

suitable initiating devices like detonators, and detonating fuse.

Initiating explosives or primary explosives are those that can detonate by the

action of a relatively weak mechanical shock or spark. The most important

primary explosives are mercury fulminate, lead azide, lead styphnate,

diazodinitrophenol, tetrazene and other mixtures.

Before going into a discussion of the aforementioned groups individually, it

would be useful to understand certain technical terms that would be

repeatedly used.

3. Detonation and Deflagration Brisance is the destructive fragmentation effect of a charge on its immediate

77


vicinity. The relevant parameters that quantify this are the detonation rate

and the loading density (or compactness) of the explosive as well as the gas

yield and heat of explosion.

3.1 Velocity of Detonation

Velocity of Detonation refers to the speed at which a one dimensional

detonation wave representing a shock front travels through a charge of

explosives confined in a borehole. The VOD of an explosive depends on

several factors such as the density, the constituents, their particle size, the

diameter of the hole (or change) and the degree of confinement.

TABLE 2 : Oxygen Balance of Explosives and Explosives Components

Material Available O2 % Comments

Ammonium nitrate (AN) + 20.0 + Release oxygen

Ammonium picrate - 52.0 - Takes up oxygen

Ammonium perchlorate + 34.0

Barium nitrate 30.6

Carbon - 266.7

Nitroglycerin + 3.5

Dinitrotoluene - 114.4

Nitrocellulose - 28.6

Picric Acid - 45.4

Potassium chlorate + 47.0

Sodium nitrate - 47.4

Tetryl - 47.4

TNT - 70.0

Trinitroresorcinol - 35.9

78


3.1 Oxygen Balance The amount of oxygen expressed in weight percent, liberated as a result of

complete conversion of the explosive material to CO2, H2O, SO2, AL2O3, etc.

(“ Positive” oxygen balance). If the amount of oxygen bound in the explosive

is insufficient for the complete oxidation reaction the compound has a

negative oxygen balance (Table 2)

Commercial explosives must have and oxygen balance close to zero in order

to reduce to the minimum the amount of toxic gases, particularly CO and

nitrous gases.

4. High Explosive The basic requirements of high explosives particularly of the industrial type

are the following:

1. Maximum power per unit volume.

2. Minimum weight per unit of power.

3. High velocity of detonation.

4. Insensitivity to shock on firing and impact, and yet sensitivity to

initiate readily when property initiated.

5. Good water resistance.

6. Good fume characteristics.

7. The reaction must be exothermic in order to increase the pressure.

8. The substance(s) should be simple and inexpensive to produce from

readily available raw materials.

The most important properties of the commonest military explosives are

listed in Table 3.3. Their structures are also given below.

79


Table 3 : Properties of Military Explosives

Explosive M.P (OC)

Density (g.ml-1)

Weight Strength%

blasting geltine*

Maximum detonation

velocity (m.s -1)

TNT 80.7 1.63 67 6950 PETN 141.3 1.77 97 8300 RDX 204.0 1.73 100 8500 Tetryl 129.0 1.60 84 7500

TNT is relatively safe to handle and has low toxicity. It contains insufficient

oxygen to give complete combustion. It can, therefore, be usefully mixed

with ammonium nitrate, which has an excess oxygen. The resulting

explosives, known as amatols are more powerful and cheaper than TNT

itself, but in general have a lower velocity of detonation.

PETN is a very powerful explosive. Pure PETN is highly sensitive to friction

and impact for direct applications. It can be use fully mixed with plasticized

nitrocellulose or with synthetic rubbers to give plastic or mouldable

explosives. In conjunction with TNT, it is used in the form of boosters (20-

50% PETN in molten TNT).

The strength of the explosive is determined by comparing its capacity to

deflect a ballistic mortar with that of blasting gelatine.

80


Trymethylene Trinitramine

(RDX)

O2N N CH2 N NO2

CH2 CH2

O2N N CH2 N NO2

Tetramethylene Tetranitramine (HMX)

RDX has a lower sensitiveness than PETN. It is still too sensitive to impact

or friction. It is either desensitized with wax or else used like PETN in

admixture with TNT.

5. Commercial Explosives Most high explosives in this category consist of mechanical mixtures of two

or more explosive bases and other additives.

Tetryl 2, 4, 6, trinitrophenyl methyl nitramine

81


Tetryl is a moderately sensitive to initiation by persuasion or friction.

TABLE 4 : Ingredients used in Explosives

Ingredient Chemical Formula Function

Ethylene glycol dinitrate

C2H4 (NO3)2 Explosive base; lower freezing point.

Nitrocellulose (NC) C6H7 (NO3)3O2 Explosive base; gelatine agent.

Nitroglycerine (NG) C3H5(NO3)3 Explosive base.

Trinitrotolune (TNT) C7H5N3O6 Explosive base.

Metallic Powder Al Fuel Sensitizer

Black Powder NaNo3+C+S Explosive base, deflagrates

Pentaerythritol C5H8N4O12 Caps, detonating

Lead azide Pb (N3)2 Explosive base used in blasting caps.

Mercury Fulminate Hg(ONC)2 - do -

Ammonium Nitrate(AN) NH4NO3 Explosive base, Oxygen carrier.

Sodium Nitrate(SN) NaNO3 Oxygen carrier Reduces f.pt.

Potassium Nitrate KNO3 Oxygen carrier.

Ingredient Chemical Formula Function

Ground coal, Charcoal Lamp black

C Combustible or fuel

Paraffin wax CnH2n +2 - do -

Sulphur S - do -

Fuel oil (CH3)2 (CH2)n - do -

Wood pulp (C6H10O5)n Combustible, absorbent

Kieselguhr SiO2 Absorbent, prevents caking

Chalk CaCO3 Antacid

Calcium Carbonate CaCO3 - do -

82


Zinc Oxide ZnO - do -

Sodium chloride NaCl Flame depressant (permissible explosives).

In formulating mixtures of explosives, it is necessary to achieve an oxygen

balance in order to:

a. Maximize the explosive energy of the reaction by aiming at complete

Combustion, and

b. To avoid the presence of toxic gases in the combustion products.

This is generally achieved by using, i) combustibles / fuels to react with the

excess oxygen and thereby prevent the formation of toxic oxides of nitrogen

or, ii) oxidizing agents to promote full oxidation of Carbon to Carbon dioxide,

thereby inhibiting the formation of toxic Carbon monoxide. An adsorbent

such as Kieselguhr has been used to absorb liquid explosives such as

nitroglycerine, Antacids promote stability in storage. Low freezing point

compounds reduce the propensity of the mixture to freeze in cold weather.

Gelatinising agents promote water resistance. Flame depressants or coolants

reduce the size, duration and temperature of the flame during the explosive

reaction to avoid methane/coal dust explosion in coal mines.

In the category of high explosives, the chief base are nitroglycerine (NG) and

ammonium nitrate (AN) although TNT also has a limited application in the

industrial explosive area. There are three different categories of high

explosives, each of different physical characteristics and chemical

composition; but all have comparable values as explosive. These three

categories are:

a. TNT

b. Dynamites and gelatins, &

c. Slurry explosives.

83


A brief account of each of these is given below. 5.1 Trinitrotoluene TNT is a high density explosive of excellent water resistance. Although

primarily used in the field of military explosives, it is used particularly in the

U.S. in commercial explosives also as pellets, flakes or grains. It combines

the important advantages of a high strength explosive base with that of low

hazard sensitivity. Pelleted TNT is marketed in the US under the trade name

of ‘ Nitropel’. Nitropel can be poured direct into wet or dry quarry holes

either as the main charge or to fill the annular space around rigid cartridges

as supplement.

In the latter case it is primed by the surrounding charge. When used as a

main charge it calls for a high strength primer. When detonated, TNT

produces much fume; therefore it cannot be used underground.

Crystalline TNT (flaked and grained) is used in the manufacture of Pentolite

boosters, in certain slurry explosives and in some blasting accessories.

5.2 Dynamite and Gelatines Nitroglycerine or glycerol trinitrate has the following formula.

CH2 O NO2 CH O NO2 CH2 O NO2

Discovered by Sobrero in 1847. It was developed to a commercial scale by

Alfred Nobel. Nitro glycerine is a viscous yellow liquid which freezes at 13.2

°C to a sensitive solid explosive. In both forms it is too sensitive to be

handled safely. It is therefore converted into a more convenient gelatinous

84


(plastic) solid by the addition of 8% gun cotton or nitro cellulose (obtained by

the nitration of cotton) to form Blasting gelatine or by absorbing it in

Kieselguhr to give straight dynamite (containing about 75% NG) or by

admixtures with other explosive agents and additives to form other types of

dynamites such as the following.

i. Ammonia dynamites: (also called extra dynamites)

Contain AN in place of SN, in addition to NG. They have lower VOD,

lower density, higher shock resistance and better fume

characteristics.

ii. Semi gelatins or low NG, high AN explosives:

Nitroglycerine is mixed with nitrocellulose to form a gel matrix to

which AN is added in various proportions. Fuels like starch and wood

meal are also added. In case of use in coal mines (permitted

explosives) coolants like sodium chloride, calcium formate etc., are

incorporated.

6. Slurry Explosives The gelatine explosives suffer from two important drawbacks. One is their

susceptibility to accidental detonation by impact or friction and the other is

the danger of exudation of NG from the cartridge. Therefore during the last

few years a different kind of high explosives called the slurry explosives,

have been increasingly used. Their main explosive base is ammonium

nitrate (which is an oxidizer). Its use depends on the complementary

reaction of a reducing agent (called the sensitizer ) to support or provide the

chemical (explosive) reaction. These sensitizers comprise a range of

carbonaceous combustibles or fuels. TNT, smokeless powder, finely divided

light metals (such as aluminum) and certain organic compounds.

A Slurry explosive may be defined as a semi-solid or pasty suspension of an

85


oxidizer, fuel and sensitizer in a thickening agent such as guargum,

crosslinked and stablised by inorganic salts of antimony or chromium.

These ‘cross-link’ slurries are sometimes called ‘Watergel’ explosives. Some

times microballoons (3-4 microns) are also used as additional sensitizers.

The slurry explosives have the following distinct advantages:

1. Built in safety against impact, friction and fire.

2. Water compatibility.

3. Reduction of post-blast fumes containing, CO, NO & NO2

4. Elimination of NG headaches.

5. Increased borehole coupling.

6. Greater control of density.

7. Powder Explosives One of the major applications of prilled ammonium nitrate coated with

anticaking agents is in the manufacture of powder explosives.

ANFO – a mixture of AN (94.53) and Fuel Oil (No. 2 diesel 5.52) – is one such

explosive. Air voids present in the prilled AN act as snsitizers. This mixture

detonates at a velocity of 10,000 to 14,000 feet per second according to the

general reaction.

3 NH4NO3 + CH2 (X) 7 H2O + CO2 + 3N2

These mixtures can have their sensitivity increased by the addition of

sensitizers like TNT, DNT, NG, Nitrostarch, etc. Other fuels like woodmeal,

rice husk, coal powder etc., are also often used. These mixture are regarded

mostly as flammable rather than as explosives. These explosives suffer from

the following disadvantages.

86


1. Caking of explosive.

2. Poor water compatibility

3. Low density.

4. Bad post blast fumes.

5. Dust problems in handling powders.

8. Energetics of Explosives The usefulness of an explosive or an explosive mixture arises from the

availability of a large amount of heat released suddenly under high pressure.

The science of thermodynamics and thermochemistry can help in the

calculation of the energy produced in such explosive reaction.

C + O2 CO2 + 94.05 K.Cals.

When an oxide of nitrogen such as NO is formed from the elements energy is

taken from the surroundings – an endothermic reaction takes place :

21 N2 +

21 O2 NO - 22 K.Cal. The energy released or absorbed is

generally represented in K. Cals or K. joules. Elements in the standard

states of aggregation are assumed to have zero heat content. Therefore the

heat content of CO2

Is -94.05 K. Cals. Similarly NO has a content of 22 K.Cal per mole.

The amount of heat liberated on the total combustion of a mole of

nitroglycerine is 344 K.Cals. This may be determined experimentally

(calorimetrically) and also theoretically as follows.

87


CH2 – ONO2

CH – ONO2 12CO2 + 10H2O + 6N2 + O2 +1375.8 K. Cals. CH2 – ONO2

NG is said to have a positive oxygen balance because of the liberation of

excess O2.

Heat of formation of CO2 = + 94.05 K. Cals.

Heat of formation of H2O = + 57.8 K. Cals.

Heat of formation of nitroglycerine.

3C + 225 H2 +

23 N2 ← NG + 82.7 K. Cals.

According to the foregoing equation 82.7 K. Cals are required to convert 1

mole of NG into its constituent elements. The heat content of NG is therefore

-82.7 K.Cals. The net heat of combustion of 4 molecules of NG will be

((12 x 94.05 + 10 x 57.8) – (4 x 82.7)) K. Cals = 1375.8 K. Cals.

Or, 1 mole of NG can liberate 344 K. Cals of heat, and 1 kg of NG can

liberate 1515.2 K. Cals of heat.

(1 mole of NG = 227 gm).

The energy obtainable from two typical explosives is calculated as under:

Explosive Formula % in explosive Oxygen Balance* AN NH4 NO3 89 89 x (+0.2) = 17.8 DNT C7 H6 N2O4 8 8 (-1.14) = - 9.12 Woodmeal (C6H10O5) n 3 3 (-1.10) = - 3.30

Total oxygen balance = + 5.38

88


8.1 Reactions

NH4 NO3 N2 + 2H2O + 21 O2 + X cals.

C7H6N2O4 + 23 O2 7CO2 + 3H2O + N2 + X cals.

C6H10 O5 + 602 6CO2 + 5H2O + X cals. In one kg. of the explosive AN 890 gm 11.125 moles

DNT 80 gm 0.44 moles

Woodmeal 30 gm 0.185 moles

Amount of N2 = 11.125 x 1 + 0.44 x 1 = 11.565 moles

Amount of CO2 = 0.44 x 7 + 0.185 x 6 = 4.190 moles

Amount of steam = 11.125 x 2 + 0.44 x 3

+ 0.185 x 5 = 24.495 moles

Amount of oxygen not used,

= (11.125 x 0.5) – (0.44 x2

13 ) – (0.185 x 6)

= 1.5925 moles.

Total no. of moles of gases after explosion = 11.565 + 4.190 + 41.8425 + 24.495 +1.5925 = 41.8425

Total vol. of gas at NTP = 22.4 X 41.8425 = 937.272 litres.

(Note: 1 gm mole of any gas occupies 22.4 litres at NTP)

So, 1 Kg of exolosives can produce upto 937.272 litres of gas.

89


Total Energy Energy released =

= (Heat of formation of products) – (Heat of formation of reactants)

= (No. of moles of CO2, H2O, N2, O2 x (respective H) –

No. of moles x (H of AN, DNT and Woodmeal)

= (4.19 x 94.05 + 24.495 x 57.8) – (11.125 x 81.23 + 0.44 x

16.9 + 0.185 x 8.108 )

= 897.261 K. Cals per kg. (H for Heat)

So, 1 Kg of explosives can produce upto 897.261 K Cals per Kg.

Example 2: Calculation of the energy of a slurry formulation.

Component % in Composition Weight in 1000 gms. AN 61 610 SN (Sodium Nitrate) 5 50 SPC (Sodium Per chlorate) 8 80 Water 12 120 A1 7 70 Ethylene glycol 6 60 Gum 1 10

100 1000 Reactions Assumed

NH4NO3 N2 + 2H2O +21 O2

2NaNO3 Na2O + N2 + 25 O2

NaClO4 NaCl + 2O2

2A1 + 23 O2 Al2O3

C2H6O2 + 25 O2 2CO2 + 3H2O

90


C6H10O5 + 6O2 6CO2 + 5H2O

Total Oxygen Balance of the Compositions: 61 x 0.2 + 5.0 x 0.47 + 8.0 x 0.5 + 7 x (-0.9) + 6 x (-1.3) + 1.0 (-1.1)

= 18.55 – 15.2 = + 3.35

The explosive composition is thus a oxygen positive formulation.

No of moles of Products

No. of moles of N2 H2O CO2 O2 A12O3 Reactants AN = 7.625 7.625 5.25 - 3.8125 - SN = 0.588 0.294 - - 0.7350 - SPC = 0.653 - - - 1.3060 - A1 = 2.590 - - - -1.9425 1.295 EG = 0.967 - 2.901 1.934 -2.4175 - Gum= 0.055 - 0.275 0.330 - 0.3300 -

Total no. of 7.919 18.426 2.264 +1.1635 1.295 Moles Total no of moles = 31.0675

∴Total gas volume produced = 31.0675 x 22.4 = 695.9 litres/kg.

Total Energy = Heat of formation of products (H2O, CO2, Al2O3) – (Heat of formation of

Reactants AN, SN, SPC, EG, Gum + Latent Heat of evaporation of water).

= (18.426 (57.8) + 2.264 (94.05) + 1.295 (399)

- ((7.625 (81.53) + 0.588 (111.54) + 0.653 (92.18)

+ 0.967 (108.7) + 1265 x 10 + 0.583 (120))

1000

91


= 1794.657 – 935.165 K. Cals/kg.

= 859.498 K. Cals/kg.

8.2 Detonation Temperature The heat of explosion raises the reaction products to the detonation

temperature. One method that can be adopted to estimate the detonation

temperature is the use of the following expression. Explosion temp (T) = Heat

liberated H/Specific heat of gases (Cv)

For example 1, where H = 871.44 K. Cals., T = 2149° K or 1876° C.

An alternative and more elegant method for calculation of the detonation

temperature is to assume two temperature values and to sum up the

internal energies for the reaction product multiplied by their corresponding

mole number. Two caloric values are obtained of which one may be higher

than the calculated heat of explosion and the other slightly lower. The

detonation temperature is found by interpolation between these two values.

For Example 2, two temperatures 2100 K and 2200 K may be assumed. The

following table gives the internal energies.

@2100 K No. of Product @2200K Product K. Cal per mole moles K. Cal K. Cal per mole K. Cal CO2 27.19 4.2 114.2 28.66 120.4 H2O 21.51 24.5 527.9 22.79 558.4 N2 16.60 11.6 912.6 17.47 202.9 O2 17.51 1.6 28.0 18.41 29.5

41.9 861.8 910.8 The interpolated temperature value for 864 K. Cal would be nearer 2100 K.

92


9. Factors Affecting the Performance of High Explosive 9.1 Oxygen Balance As was mentioned earlier, during detonation, the fuel elements in the

explosive get oxidized vary fast. Therefore oxygen balance, i.e the percentage

of excess oxygen in the composition is an important factor in energy

production. In an ideal composition, the oxygen balance should be zero. If

fuel or oxidizer is in excess harmful gases like carbon monoxide or oxides of

nitrogen are produced, apart from loss of energy.

9.2 Completion of Reaction It is also important that the reaction taking place during an explosion

should go to completion at the required speed. For example, it the Carbon in

the system is oxidized to CO rather than CO2, the production of energy

comes down by 75%. The formation of oxides of nitrogen also involve

absorption of energy.

C + O2 CO2 + 94 K.Cal.

C + 21 O2 CO + 26 K. Cal.

21 N2 +

21 O2 NO – 22 K.Cal.

9.3 Compatibility of Ingredients The ingredients used in the formulation of an explosive should be such as

not to lead to extraneous complications. For instance if the AN used in

ANFO contains more than 1% moisture, caking results. This extra moisture

content would also result in lowered energy production. If the surface of

sensitizer grade A1 is not properly protected, it might react with water to

produce hydrogen in an exothermic reaction. This prelibation of gas would

93


result in bursting of cartridges and loss of power.

9.4 Critical Diameter The critical diameter is the minimum diameter of an explosive charge at

which detonation can still take place. The detonation wave tends to fail or

fade when the diameter of the explosive charge decreases. Normally the

propagation sensitivity of an explosive is measured at a length at least six

times the diameter cartridge and the VOD is measured at a length at least

six times is measured at a distance of six inches from the point of initiation.

At a lower diameter, even if the explosive is sensitive, the reaction in the

cartridge may be incomplete. The above phenomenon can be eliminated by

carefully controlling the critical density or the ‘Loading Density’.

9.5 Critical Density Density is an important characteristic of explosives. Raising the density (e.g.

by pressing or Casting) improves brisance and detonation velocity. In

contrast low density explosives produce a milder thrust effect. The critical

density of ANFO is less than 1g/cc and that of TNT is 1.76 g/cc.

9.6 Packaging The packaging of an explosive is yet another important factor that governs

the quality of an explosive. Moisture can affect the performance of certain

explosives packed in craft paper. In powder explosives, the AN component

cakes up because of water incursion. Proper waxing of the cartridge is also

necessary. Aluminum clips are not permitted in underground coal mines.

9.7 Strength Measurement The strength of an explosive is usually assessed by a series of tests chosen

94


to measure the performances of the explosives under various conditions.

Whilst these tests are of considerable value, particularly for comparative

purposes, it must be remembered that no laboratory test or series of tests

can preduct precisely the performance of explosives in the field.

9.8 Power The power or strength of any explosive is usually expressed in terms of

power per unit weight which is appropriate for comparing explosives used in

charges measured by weight. One of the most satisfactory instruments for

carrying out this measurement in the laboratory is the Ballistic mortar (Fig.

1). This consists of a pendulum 3 m long at the bottom of which is a bob

weighing 333 kg. A shot is used of 12.37 cm diameter weighing 16.6 kg and

fitting with a clearance of 0.88 mm. Ten grammes of the explosive under test

are wrapped in tin foil and fired with a standard copper detonator. The

explosive ejects the shot on to rubber matting. The corresponding recoil of

the bob is shown on a scale and is a measure of the energy imparted by the

explosive to the system. In Great Britain, the standard explosive is blasting

gelatine and results are expressed as a percentage of the strength of blasting

gelatine.

An older form of measuring power was the Trautzel lead block test (Fig. 2).

The explosive under test is placed in a cylindrical hole in a block of specially

cast lead and the remainder of the hole filled with sand. When the explosive

is fired, it causes an expansion of the hole in the lead block and this

expansion is measured by filling with water. For most explosives there is an

adequate correlation between lead block results and ballistic mortar

measurements (Fig. 3).

The ballistic mortar and lead block tests are not applicable to slurry

explosives which are too insensitive to detonate properly under such

95


conditions where only small amounts of explosives are used. For these

explosives, it is useful to fire larger amounts of several kg. Under water and

measure the period of oscillation of the gas bubble produced. The longer the

period of oscillation of the gas bubble, the greater the energy of the gas

bubbles. This part of the total energy correlates well with the blasting affect

of the explosive.

9.9 Velocity of Detonation VOD is generally measured by the well-known ‘Dautriche method’, the

principle of which is illustrated in Figure 4. The two ends of a length

detonating fuse (usually 10 g/m) are inserted in the explosive under test at a

known distance apart. The midpoint of the piece of detonating use is known

and the part of the fuse near this point is placed over a V-shaped groove in a

thin lead plate. When the explosive is fired, the fuse commences to detonate

first at the end nearer the detonator and later at the other end. When the

two detonation waves in the fuse meet, they reinforce and produce a distinct

mark on the lead plate, visible particularly as a spit on the back, This point

will be removed from the midpoint of the fuse by a distance which depends

on the velocity of detonation of the explosive under test according to the

following equation D = D11 / 212 where D is the velocity of detonation of the

explosive, d is the velocity of detonation of the detonating fuse, 11 is the

length of the explosive under test, 12 is the distance between the centre of

the fuse and the mark on the plate.

In recent time, VOD is more effectively measured by the use of high speed

cameras, and also electronically.

Sensitiveness: One of the commonest methods to evaluate the sensitiveness of an explosive

to impact is the ‘fall hammer test; which determines the minimum height

96


from which a given weight must be dropped in order to initiate detonation. A

simple and practical method of this experiment is shown in Fig. 5. In this,

the explosive is put between two roller bearings, themselves placed in a ring

of hardened steel and resting on a hardened steel base. The falling weight is

arranged to hit the upper of the steel cylinders. Typical weights employed

are from 0.5 to 5 kg. and height of fall may be upto 200 cm.

A corresponding useful test for the safety of an explosive during handling

(when it can be subjected to a blow at a glancing angle) is the torpedo

friction test.

Results of sensitiveness tests on typical explosives are given in Table 5, for

steel to steel surfaces, in the absence of grit.

Table 5 : Sensitiveness of Explosives

Fall Hammer Torpedo Friction (0.5 kg) (cm) (1 kg at 80°) (cm) TNT 200 80 – 120 RDX/TNT 80 – 100 40 – 45 RDX 25 – 30 10 – 20 PETN 60 – 80 35 – 40 NG Powder 20 – 30 150 TNT Powder 160 – 200 100 – 120 9.10 Detonation Pressure Detonation pressure of an explosive depends on the density of an explosive.

Higher the density, the higher will be the detonation pressure for a given

composition. Above the critical density, detonation pressure is zero as the

cartridge will not explode.

The following generalized formula can be used for calculating detonation

97


pressure.

P = 0.00987 + D2 P/4

Where P = density of the explosive and D = VOD in kilometers per sec.

It can be observed that higher VOD and higher density would lead to higher

detonation pressure.

9.11 Explosion Temperature Explosion temperature is the calculated temperature of the fumes of an

explosive material which is supposed to have been detonated while confined

in a shell assumed to be indestructible and impermeable to heat. Its

calculation is based on the heat of explosion and on the decomposition

reaction, with allowance for the dissociation equilibrist and the relevant gas

reaction. The real detonation temperature is higher than the calculated one.

This parameter has important implications. For instance, in case of coal

mine explosives, a balancing of explosion temperature and the gas volume

plays an important role. If the composition has high explosion temperature

(< 2000° C), it will make the methane air atmosphere incendive. If it is too

low (< 1000°) the composition will not have any useful energy for dislodging

coal.

10. SLURRY EXPLOSIVES One of the major advances in the production of commercial high explosives

during the last 20 years has been the development of ‘Slurry Explosives’ also

called ‘Water-gel’ explosives. These formulations possess demonstrated

advantages like safety and fume characteristics over the NG based

explosives with the result that their manufacture and use have grown by

leaps and bounds during the last two decades.

An aqueous slurry explosive composition generally contains a suspension of

98


inorganic oxidising salt, usually predominantly ammonium nitrate,

suspended in a saturated aqueous. solution of oxidising salt together with

sensitiser and optionally additional fuel. To prevent separation of the

ingredients and to improve the resistance to deterioration of the composition

in wet conditions the aqueous phase is usually thickened with a dissolved

thickening agent, the currently preferred thickener being guar gum. The

composition is often further gelled by crosslinking the thickening agent with

a crosslinking agent such as a chromate, dichromate or pyroantimonate.

Although the term 'slurry' is universally applied to such compositions, the

degree of consistency may range from pourable to highly viscous extrudable

gels.In order to improve the explosive sensitivity, the composition often

contains an aeration agent which usually is a chemical such as sodium

nitrite which reacts in situ in the composition to generate small gas bubbles

throughout the mass and thus reduce the density.The advantages of

incorporating gas bubbles in aqueous slurry explosives by means of gassing

agents or by the addition of gas-containing material for density and

sensitivity control are now well known. The beneficial sensitising effect of the

gas bubbles is believed to be attributable to the 'hot- spots' obtained by the

adiabatic compression of the gas bubbles by the shockwave produced during

detonation. In accordance with the present invention, the gassing efficiency

and productivity of nitrite salts can be susbtantially improved by combining

with a nitrite salt a gassing accelerator comprising the thiocyanate ion SCN-.

Preferred thiocyanate ion-containing materials include, for example, sodium

thiocyanate and ammonium thiocyanate or a mixture of these.

An aqueous slurry explosive composition of the type suitable for use in large

diameter borehole charges was prepared according to the following

formulation the amounts shown being expressed as percent by weight:

99


Nearly all commercial and mining explosives used in the world today are

based on ammonium nitrate (AN) or combinations of AN with other alkaline

and/or alkaline earth nitrate salts, eg. sodium or calcium nitrata Most

explosives of this type rely on the energetic reaction of nitrogen compounds

incorporated within the explosive to provide the necessary explosive power.

AN, which is a strong oxidiser, has been used as the base of commercial

explosives for at least the last 50-60 years.

Initially, mining companies used AN as an explosive on its own. However,

they soon realised that the addition of diesel increased the energy output

without a large increase on costs (ammonium nitrate- fuel oil, now

commonly referred to as 'ANFO'). However, the water resistance of ANFO is

quite poor, which limited its use in wet blast holes. To ameliorate this issue,

slurries and watergels were developed. Slurries typically comprise AN

dissolved dispersed in water, other salts (calcium nitrate, sodium nitrate,

etc) to depress the crystallisation point of the AN solution, and other

additives such as guar gum (as thickener) and fuel (diesel). They can also be

blended with ANFO depending on the characteristics of the ground being

blasted. Slurries also typical ly include solid sensitisers (aluminium and

high explosives such as TNT, RDX, efc) to enable the slurry to detonate and

100


to minimise misfires. Watergels have similar compositions to slurries,

however, crosslinkers can be added to enhance the water resistance of the

product. Watergels can be aerated or gassed with bubbles chemically

generated in situ or mixed with glass/plastic miaospheresto lower the

density, improve sensitivity and change the levels of energy delivered to the

ground being blasted.

One of the drawbacks of watergels and slurries is that there is a limit of AN

which can be incorporated into the solution. This drawback was overcome

by the development of water-in-oil emulsions.

These emulsions can contain AN in high concentration (see US Patent No.

3,447,978). Water in oil emulsions are made of a hot aqueous phase

(composed of AN, other nitrate salts, perchlorate salts, etc) dispersed into an

organic fuel. The aqueous - organic mixture is stabilised by the use of an

emulsifier. Emulsions can also be blended with ANFO in different ratios

Watergels can be aerated, gassed with bubbles chemically generated in situ

or mixed with glass plastic microspheres to lower the density, improve

sensitivity velocity of detonation (VOD) and change the levels of energy

delivered to the ground being blasted.

Despite the development of AN emulsions, AN slurries and watergels,

however, there is still a need to develop improved explosives, which are

preferably more cost effective compared to existing explosive compositions

and are capable of being produced in large quantities to meet the high

demand from industry. Preferably any explosive composition which can be

substituted for an AN-based explosive is insensitive to misfires and is not

desensitised by wet blast holes Furthermore, preferably any AN substitute is

a sustainable raw material which has a relatively low carbon footprint, and

which can be manufactured relatively easily and preferably near the actual

mine site to minimise transport issues. In addition, preferably any AN

101


substitute can be produced on an as- needs basis to minimise the need for

stockpiling and to increase safety. Further still, preferably any AN substitute

can be used in slurry or in emulsion form so that existing equipment can be

used, and when in slurry form or is emulsified such that the viscosity

enables pumping without difficulty. It would also be ideal if there are no

onerous regulatory requirements for such a substitute, thereby reducing

administrative costs It would also be preferable for the explosive composition

to be cross linkable in- situ to increase viscosity down the blast hole.

Despite of the advances on the types of composition that can be

manufactured from ammonium nitrate, one of the disadvantages is that

during the detonation NOx fumes can be generated, due to the presence of

nitrogen compounds in the explosive composition (from nitrates). These NOx

fumes are toxic and can affect the health of mine site personnel. Therefore

the emission of NOx fume after blasting is a safety issue and in countries

like Australia there are now strict regulatory controls in place to manage

such emissions.

Slurry Explosives are ‘fuel’ sensitized AN with or without other oxidisers

such as Sodium nitrate (SN) or sodium perchlorate (SPC) or ammonium

perchlorate in which the solid fuel and the solid part of the oxidisers(s) are

dispersed in a continuous fluid medium generally an aqueous solution with

our without other polar solvents and often more or less aerated. They have

been designated differentially as slurry explosives (SE) when they are

sensitized with explosives (e.g. TNT) and slurry blasting agents (SBA) when

the fuel is not an explosive (e.g. aluminium, sulfur and solid hydrocarbon). A

further division of these water-based explosives is into watergels and

emulsion explosives.

The water gel explosives have the oxygen donor in the form of nitrate salts

fully or partly in a water solution which is gelled. In this gel, the fuel which

is dissolved or distributed, quite often consists of aluminium powder,

102


methylamine nitrate (MAN), glycol and urea. Some of the fuels also sensitize

the mixture. Additional sensitizing is achieved by the acid of small gas

bubbles.

The emulsion explosives have an oxygen donor consisting of nitrates and

perchlorates in an acqeous solution. The water phase is in the form of small

droplets in the continuous oil phase which constitutes the fuel. The fuel

constitutes the fuel. The fuel consists of mixtures of waxes and oils. The

explosives are in this case sensitized by gas bubbles in the form of

microspheres. Additional strength can be achieved by the addition of fuels

such as aluminium powder.

10.1 Development of Slurry Explosives One of the first formulations in this direction viz. AN/FO interestingly owes

its origin and development to two major catastrophic incidents – one in

Texas city in the US and the other aboard a ship in Breast in France. These

explosions incidentally revealed that AN is a low-cost, powerful explosive

especially in the presence of fuels like oil or wax. The first of the watergel

explosives contained about 20% water 23% coarse TNT and 55% AN or

mixed AN and SN, thickened to a slurry consistency with guar gum. The

TNT acted both as the fuel and the sensitizer. The next development was the

production of the slurry blasting agent SBA – A1 which contained

Aluminium particles of a special grade in place of TNT.

In both the systems described above, the mechanism of sensitivity depended

on the existence of small air bubbles get compressed adiabatically when

initiated, raising the temperatures of the slurry to several thousands of

degrees centigrade. This high temperature leads to the explosion.

It should be noted that, in the case of TNT, TNT itself acts as a detonating

103


high explosive leading to a shock wave mechanism. The effectiveness of TNT

depends on the concentration and size of the particles.

In the aluminized slurries, the mechanism called the ‘hotspot’ is governed by

the following ‘factors.

1. Air bubbles of the size 2 - 4 micron should be present for providing

initiation spots, and

2. These bubbles must be in contact with the aluminium particles.

The earlier non-crosslinked slurries suffered from the following

disadvantages:

1. Solids separate out on storage due to gravity and the solid solution

equilibrium cannot be effectively controlled.

2. The enclosed air bubbles, if allowed to escape, reduced the sensitivity

of the explosives.

104


11. A Comparison of Slurry /Water Gel and Emulsion Explosives One material that is also an oxidiser and that has the potential to meet at

least some of the health and safety needs of modern times is hydrogen

peroxide (HP). Use of HP as a liquid explosive has been patented by Shanley

(US Patent No. 2,452,074) in which HP was mixed with glycerol and water,

and initiated with a detonator. US 2,452,074 teaches that to achieve

detonation the explosive compositions can only contain up to 52 wt% water.

More recently, in 1990 mixtures of HP and water soluble resins were

patented by Bouillet as a packaged explosive (US Patent No. 4,942,800).

Examples provided in the patent show that some of the mixtures detonated

at velocities above 6,000 m/s (in 33 mm diameter, unconfined blasts). Also,

work published by Shell Co. (Concentrated Hydrogen peroxide, summary of

research data on safety limitations, 1961) presents the detonation limits of

the system HP-acetic acid-water in a ternary diagram. In 2004 an

investigation into the detonation properties of HP and alcohols was

published {"Investigation of explosive hazard of mixtures containing

hydrogen peroxide and different alcohols", Journal of Hazardous Materials,

A 108 pp. 1-7, 2004).

Structure of concentrated oil-in-water Pickering emulsions

105

http://www.chemspider.com/Chemical-Structure.937.html


12. Water gel-technology for HP-based explosive HP-based watergels can be prepared with either water-miscible or water

immiscible fuels. Water-soluble fuels which can be used with the present

invention can be selected from the group consisting of: glycerol, sugar,

amine nitrates, hexamine and urea Water-insoluble fuels which can be used

with the present invention can be selected from the group consisting of:

include aliphatic, alicyclic and aromatic compounds and mixtures thereof

which are in the liquid state at the formulation temperature. Suitable

organic fuels may be chosen from fuel oil, diesel oil, distillate, kerosene,

naphtha, waxes, (eg. microcrystalline wax, paraffin wax and slack wax)

paraffin oils, benzene, toluene, xylenes, asphaltic materials, polymeric oils

such as the low molecular weight polymers of olefins, vegetable oils, animal

oils, fish oils, and other mineral, hydrocarbon or fatty oils, and mixtures

thereof. Preferred organic fuels are liquid hydrocarbons generally referred to

as petroleum distillates such as gasoline, kerosene, fuel oils, paraffin oils

and vegetable oils or mixture thereof.

Reaction schematic of emulsion explosives( in a different setting)

106


12.1 Thickeners and cross-linkers Because bubbles of gas and materials enclosing gas have a relatively low

density, they will tend to migrate towards the surface of the column of

explosive if the viscosity of the HP-based explosive composition is not

capable of maintaining the sensitising material homogeneously dispersed

throughout. Migration of the sensitising material towards the surface is

undesirable as it may render the explosive too insensitive to initiation, and

therefore the explosive composition may not deliver the energy and gases

needed to break and move the rock as required or even worst, the explosive

may undergo a misfire. One way to ameliorate this issue is to formulate the

explosive composition into a watergel. These types of compositions can be

formulated with different levels of viscosity by using a thickener. Viscosities

can be selected to generally retain the sensitising material in a

homogeneously dispersed state throughout the composition.

Microscopic image of a Water Gel Explosive

13. Emulsion-technology for HP- based explosive compositions HP-based emulsions can be prepared with water-immiscible fuels The water-

immiscible organic phase component of the composition comprises the

continuous "oil" phase of the water-in-oil emulsion and is the fuel. Suitable

107


organic fuels include aliphatic, alicyclic and aromatic compounds and

mixtures thereof which are in the liquid state at the formulation

temperature. Suitable organic fuels may be chosen from fuel oil, diesel oil,

distillate, kerosene, naphtha, waxes, (e.g. micrccrystalline wax, paraffin wax

and slack wax) paraffin oils, benzene, toluene, xylenes, asphaltic materials,

polymeric oils such as the low molecular weight polymers of olefins,

vegetable oils, animal oils, fish oils, and other mineral, hydrocarbon or fatty

oils, and mixtures thereof. Preferred organic fuels are liquid hydrocarbons

generally referred to as petroleum distillates such as gasoline, kerosene, fuel

oils paraffin oils and vegetable oils or mixture thereof.

13.1 Emulsifier/Stabiliser HP-based emulsion compositions are made of a discontinuous phase of

oxidising material that is dispersed in a continuous phase of an organic fuel

in the presence of one or more emulsifying agents. The emulsifying agent is

adapted or chosen to maintain phase separation. The emulsifying agent

component of the composition of the present invention may be chosen from

the wide range of emulsifying agents known in the art for the preparation of

water-in-oil emulsion explosive compositions. Among the preferred

emulsifying agents are the 2-alkyl- and 2- alkenyl-4,4'-bis (hydroxymethyl

oxazoline, the fatty acid esters of sorbitol, lecithin, copolymers of poly

(oxyalkylene) glycols and poly (12-hydroxystearic acid), and mixtures

thereof.

13.2 Energy Diluents (optional) In the context of both types of explosives, energy diluting agents are inert

materials that have minimal contribution to the detonation process and can

be used to replace part of the energetic material in the composition and

therefore reduce the energy output of the hydrogen peroxide-based

108


explosive. In some cases these energy diluting agents are able to reduce the

density of the HP-based composition without increasing the sensitivity.

Examples of these diluents materials are granulated/shredded rubber (from

tyres), cotton seeds, saw dust, husk, expanded pop corn, plastic beads, wool

meal, saw dust, bagasse, peanut and oat husks, peanut shells etc.

Table 6 lists the components of explosive systems discussed herein and

provides typical ranges for each.

Table 6: Components for explosive systems discussed herein with

typical ranges for each.

The present comparison relates to a peroxide-based explosive composition

that is preferably prepared as watergel or water-in-oil emulsion, and is

sensitized. Typical components for each type of explosive technology are

listed in Table 7.

This section describes the LC model and Inventory developed to assess the

emulsion explosive production carried out by a specific European company,

109


which is representative of similar processes that occur in developed

countries. This company produces exclusively emulsion explosives with an

annual production (2013) of 10000 tonnes, 70% of which incorporating

aluminum. The model follows a cradle-to-gate LCA and covers the following

processes: transport of raw material, emulsification, sensitization and

packaging. Fig. 1 presents the a flowchart from inputs ( for preparation ) to

outputs ( ashes and inert residues of production) for the emulsion explosive

production till packaging. The emulsification, sensitization and packaging

are distinct processes, but sequential, which are considered as a single

process.

Table 7: Typical components for each type of explosive technology.

The functional unit is defined as 1 kg of TNT equivalent, corresponding to

4.5 MJ of energy content. The corresponding reference flow includes

materials and energy associated with the production of 1.45 kg of emulsion

110

http://www.sciencedirect.com/science/article/pii/S0959652614012074%23fig1


explosive, since it was considered that 1 kg of emulsion explosive has 3.14

MJ of energy content.

13.3 Emulsion Explosive Production The components used in the preparation of the emulsion explosive are i)

ammonium nitrate added in the form of prills (granules) with a diameter of

1–2 mm, ii) water, iii) mineral oil (liquid), iv) an emulsifier (liquid) and v) a

sensitizing agent (hollow microspheres with 70 microns of diameter,

considered as extruded polystyrene, XPS). Data for the emulsifiers are not

provided due to confidentiality, but pathways are for two alternative

emulsifiers: Polycarboxylate (base composition) and Carboxymethyl

(alternative composition).

Fig. 1. Flowchart representing the emulsion explosive production. Table 8 and Table 9 provide the mass and energy inventory of the emulsion

explosive production.

111

http://www.sciencedirect.com/science/article/pii/S0959652614012074%23tbl1



Table 8. Mass balance Inventory for the emulsion

explosive production (per kg TNTeq).

Constituents Amounts

Inputs Ammonium Nitrate 1.06 kg Water 0.16 kg XPS 0.03 kg Mineral oil 0.13 kg Polycarboxylate 0.07 kg Packing Polyethylene 0.05 kg Outputs Emulsion explosive (includes packing) 1.50 kg Ashes 0.002 kg Inert material 0.003 kg

Table 9. Energy requirement of the emulsion explosive

production (per kg TNTeq).

Energy requirement Amount

Electricity 0.11 kWh

Naphtha 0.01 kg

13.4 Improvements A very significant improvement in the quality of the explosives was achieved

when the thickener was chemically cross linked to form the gel. An effective

gel matrix prevents the salts from segregating at high temperatures and

controls the size of the crystals setting out at low temperatures. The use of

guar gum at the appropriate pH and the addition of good cross linking

agents such as chromium and antimony salts and the discovery of special

bubble controlling agents (surfactants) led to the proper understanding of

the gel technology. Today these gels can be stored even for a year without

112


any deterioration in physical condition of stable detonation properties.

13.5 Advantages of Slurry Explosives

1. Tailored energy requirements.

2. Water compatibility.

3. Greater safety.

4. No head-ache causing ingredients.

5. Wide choice of densities.

6. Explosives may be loaded by priming straight into bore holes which

can be completely filled with the explosive material.

7. The possibility of secondary explosions is reduced considerably

because of the water present.

13.6 Types of Slurry Explosive

There are two types of Slurry Explosives:

Non-Aluminised Slurries Aluminised Slurries

Organic Compound-Sensitised Flake grade Aluminium

Slurries powder as Sensitiser

e.g.,

Nitro Starch

Nitro Sugar

Alkyl amine nitrates

Alkanolamine nitratres

The different ingredients of a Slurry Explosives are grouped and presented

in Table 8 to 15.

113


TABLE 10 Some Aerating Agents Fibrous pulp and meals Vermuculite Pearlite Resin balloons(Microballoons) Cork Glass microballoons Ammonium nitrate prills

TABLE 11 Some Gas Formers

Peroxides Acetone/Cresote in HNO3 gels Sodium/Potassium nitrates Sodium bicarbonate

TABLE 11 Oxidisers

Ammonium Nitrate Sodium Nitrate Calcium Nitrate Sodium Perchlorite Potassium Perchlorite Ammonium Perchlorate Potassium Nitrate

TABLE 13 Sensitisers

TNT A1 (Flake or Paint Grade) Nitro Starch Methyl Ammonium Nitrates Ethanolamine Nitrate Microballoons(Glass/Polymeric)

TABLE 14 Thickners

Guar Gum Ethyl Guar CMC (Carboxy Methyl Starches Cellulose) Propylguar

14. Conclusions Emulsion Type Explosives are making great strides in explosives technology

and promise more advances in the future as operators become more aware

of their potential uses and manufacturers develop new markets for a

versatile product. Economic analysis of emulsions as compared with

114


straight ANFO indicates two major advantages for Emulsion Type System

over straight ANFO. First, they can provide increased fragmentation over

equivalent amounts of ANFO, due to the emulsions increased energy and

efficiency, albeit at a higher (blasting) cost per unit of overburden. Secondly,

and more importantly, emulsion blasting costs can be reduced below that

of straight ANFO by taking advantage of the product's explosive

characteristics and spreading the blasting pattern to the break- even point.

A mathematical model has been developed which enables an economic

comparison between any two explosive systems based on their densities and

cost per unit of explosive material.

References: 1. Sybil P.Parker, (Ed.), McGraw-Hill Encyclopedia of Science and Technology,

5th edn., McGraw-Hill, New York, 1982.

2. T. McDonald,. Farmers Supply and Explosives, Inc., Barbourville, KY,

August18, 1986.

3. Nitro Nobel. Emuliteis breaking new ground in bulk blasting :( advertising

pamphlet); Gyttorp, Sweden. (1985)., Balkema, Rotterdam.

115

CHAPTER – 3

116

INFLUENCE OF AMMONIUM NITRATE PRILL PROPERTIES ON DETONATION

VELOCITY

1. Introduction

Ammonium nitrate (AN) is commonly used as a basic ingredient of ANFO

explosives. Thanks to low cost, simple technology, and accessibility of raw

materials required for manufacturing, this type of explosive competes very

effectively with other improved types of explosives commercially available.

More than half of the world production of explosives used in civil

applications is of type ANFO mixtures; in developed countries their share

exceeds 80%. Laboratory and field tests performed in the mining industry

have shown that the replacement of the agriculture nonporous AN with a

porous material considerably improves the efficiency of ANFO action; it also

eliminates oil oozing and reduces the emission of toxic gases to the

atmosphere.

On the other hand, all positive features of the granulated form of ANFO

explosives are retained. Commonly used ANFO consists of 94.5% of AN and

5.5% of fuel oil. AN is the dominant and sole ingredient which has explosive

properties in such a mixture. This implies that explosive properties of ANFO

are closely bounded with the explosive characteristics of AN, which are

strongly connected with its physical features and chemical composition. The

physical features and chemical composition of AN prills/granules may vary

within a wide range depending on the technology of production and the

content of additives. The storage and transport conditions such as

temperature and humidity fluctuation, stacking height and contamination

by other substances can also have a significant influence.

Chapter 3 Influence of Ammonium Nitrate Prill Properties on Detonation Velocity

117

It is well known that the crucial features influencing the detonation

parameters of explosives are: charge density, size of the explosives particles

and the content of mixture ingredients [2, 3]. The detonation of ANFO

mixtures is regarded as an example of a non-ideal detonation regime and

one can expect that a change in the basic physical parameters of AN

prills/granules may influence the detonation parameters of the explosive to

a high degree. The first goal of the work was to determine the influence of

the porosity (over a wide range from 0 to 15 cm3 of fuel oil per 100g of prilled

AN and of the size of the prills on the ANFO explosives detonation velocity.

The second aim was to examine how the changes in the proportion of AN

and fuel oil affect the detonation velocity of such explosive mixtures.

2. Method of Manufacturing Porous AN

The most suitable product to make porous prills is AN without any inorganic

additives. But such a type of AN is produced very rarely. Almost all types of

agricultural AN contain inorganic additives which improve their quality but

make the process of manufacturing porous AN more difficult. A relatively

small influence on such a process is exerted by ammonium sulfate

(NH4)2SO4. Other additives, like calcium or magnesium nitrate, make the

manufacturing of high porosity AN very difficult. The prilled AN used in our

work contained about 1% of ammonium sulfate. The process of

manufacturing porous prills was based on a polymorphic transition ofAN

crystals, which runs at +320C.

During the transition of IVIIIIV, the density of AN crystals changes

slightly, which creates a porous structure and cracks crystals inside the

prills. The difference in the internal structure of prills of different porosity is


118

shown on the scanning electron microscope photographs (Figure 1). The

porous product has significantly bigger oil absorption and detonation ability

as measured by tube test. The process consisted in heating an AN sample to

above 320C, storing it for some time, and then cooling to room temperature.

An additive that accelerated and increased porosity was also used. As a

result of a different number of thermal cycles and the amount of the above-

mentioned additive, AN prills of a broad range of physical characteristics can

be produced. Examinations of physical properties of the manufactured AN

were conducted. Some ANFO mixtures were prepared in a mechanical

blender and their detonation velocity was determined.

It is a common practice in military technology that HE charges should have

the highest density. With increasing density, one can observe an increase in

velocity and detonation pressure, which enhances the brisance of the

explosive. However, when charge density increases, the sensitivity to

initiation impulse decreases and the charge is more difficult to initiate. This

problem is especially important in the case of ANFO explosives, which are in

general little sensitive to initiation. The porosity of materials influences

greatly their density. The rise of porosity is accompanied by an increase of

empty spaces in materials, which cause a drop of density. The diagram

below (Figure 2) shows a relationship between the bulk density of the prilled

AN and the porosity of prills. The AN was made porous in a manner

described in Section 2. The oil absorption by a unit of mass of prills,

expressed as the volume of fuel oil absorbed by 100 g of AN prills

(cm3/100g) has been adopted as a measure of AN prills porosity.

As seen in the diagram (Figure 2), AN density drops when its porosity

increases. The AN prills of agricultural grade showed no oil absorption

(0.0cm3/100g) and their bulk density was 0.95 g/cm3. When AN heated and

cooled a few times with the additive, the rise of the porosity caused a rise of

oil absorption up to 15 cm3/100 g and a drop in bulk density to 0.59 g/cm3.


119

Figure 1. Cross-sections of the nonporous agricultural-grade AN prill

(left) and of the porous prill showing an oil absorption of 15 cm3/100g

after several thermal cycles (right).

Figure 2. Bulk density of prilled AN vs. its oil absorption.

3. Influence of Prills & Porosity on the Detonation Velocity of ANFO

Detonation velocity is one of the most important and most often determined

properties of explosives. To measure detonation velocity, ANFO explosives

were prepared by mechanical mixing of AN prills (size of prills: 1.5–2.5 mm)

and fuel oil. The ratio was 94.5% AN and 5.5% fuel oil where oxygen balance

equals zero. The explosives were poured into steel tubes of 36 mm internal


120

diameter and 3 mm wall thickness. Detonation of the charges was initiated

by boosters (weight 14 g, RDX 90%, TNT 10%). The dependence of

detonation velocity on the density of ANFO charges is presented in Figure 3.

As can be seen in Figure 3, the density decrease of ANFO charges from 0.90

to 0.68 g/cm3 caused a rise of the detonation velocity from 1.6 to 2.7 km/s.

Detonation failed for charges with a density above 0.90 g/cm3. Considering

the dependence of the density on the porosity as in Figure 2, the detonation

velocity of ANFO increases with the prills porosity. ANFO charges made of

AN with an oil absorption below 2.5 cm3/100 g, did not detonate under the

test conditions. A minimal oil absorption of AN prills, at which detonation of

full charges was observed, was about 2.5–3.0cm3/100g. The measured

detonation velocity was then 1.6–1.7 km/s. When oil absorption was at the

level of 12–15 cm3/100 g, the detonation velocity was maximal reaching 2.7

km/s. A tendency to further growth was not observed.

Figure 3. Detonation velocity of ANFO vs. density of charges.

It can also be seen in Figure 3 that when the oil absorption rises from 0 to

15 cm3/100 g the density of AN falls by about 35%. Despite this drop, the

detonation velocity of ANFO increases by more than 70% (the range of oil


121

absorption 3.0–15.0 cm3/100 g). These relationships between

porosity/density and detonation velocity are different from the dependence

observed for typical high explosives. For that type of explosives the

detonation velocity decreases with decreasing density. The measurement

points in Figure 3 are scattered because of different physical characteristics

of AN prills (density, mechanical strength, water content, amount of dust,

etc.), the uncertainty of oil absorption determination and the poor physical

stability of ANFO made of low porosity AN.

4. Influence of Prills & Size on the Detonation Velocity of ANFO

The size of crystals, grains, granules, prills, etc. has a great influence on the

properties of explosives, especially on oxidizer-fuel mixtures. For this type of

materials, the sensitivity to external stimuli and the parameters of

detonation increase when the size of the particles decreases.

Taking into consideration typical high explosives, their detonation velocity

depends only slightly on the granulometric distribution of grains. In this

part of the work prills of small size were examined; the so-called undersize

prills which are created in an industrial installation during a granulation

process are then sieved out from the trade product and recycled. The

undersize prills for the research were sieved to different fractions and used

to manufacture an ANFO of stoichiometric composition. The shape of AN

prills was spherical, similar to the grains of the commercial product. Prills of

smaller size were nonporous, similar to the commercial product. As the oil

absorption of the undersize prills was poor, the physical stability of such

ANFO was unsatisfactory. The detonation velocity of ANFO manufactured

with small size prills was determined under conditions described.

The results collected in Table 1 indicate that the detonation velocity of ANFO

increases when the size of AN prills decreases. ANFO made of prills of about


122

1.00–1.20 mm did not detonate under the test conditions. An explosive

manufactured using the smallest prills (0.20–0.50 mm) detonated fully,

reaching a velocity of about 3.0 km/s. For comparison, ANFO was made of

ground grains of 0.2–0.5 mm. In spite of a lower density, it detonated with

significantly higher velocity than parallel samples made of regular

nonporous prills.

Table 1. Detonation velocity of ANFO prepared from AN prills of different dimensions.

Prills size Density of ANFO Detonation velocity

mm g/cm3 km/s

1.00 – 1.20 1.02 Detonation decay

0.63 – 1.00 1.01 2.33

0.50 – 0.63 1.02 2.50

0.20 – 0.50 0.98 2.96

0.20 – 0.50 (ground) 0.86 3.44

Figure 4. Detonation velocity of ANFO vs. content of fuel oil.


123

5. Influence of Fuel Oil Content on Detonation Velocity of ANFO

The detonation velocity tests described above refer to an ANFO explosive with oxygen balance equaling zero, manufactured from AN prills whose porosity and size were altered. However, it may happen that the ANFO composition is different from the one designed and the oxygen balance is positive or negative. The maximum heat of explosion is expected from the stoichiometric mixture; the heat of explosion of other compositions is less. In order to determine the influence of the fuel oil content on the detonation velocity of ANFO, different kinds of AN– of low and high porosity (4 and 10cm3/100g of the oil absorption) –were chosen. The method of preparing mixtures and determining their velocity of detonation was the same as described in Section 4. The results of the experiments are presented in the diagram (Figure 4). It is interesting that ANFO explosives of both relatively high and low oxygen balance were capable to detonate. We would like to underline a weak dependence between the detonation velocity and the content of fuel oil in mixtures made of low porosity AN prills. In the case of ANFO manufactured with high porosity AN, the dependence is remarkable.

6. Conclusions

The conducted experiments confirm the importance of prills physical

structure (porosity and particle size) on the detonation parameters of ANFO

mixtures. The porosity of AN prills influences significantly the detonation

parameters of ANFO explosives. The increase of prills porosity results in a

remarkable increase in the detonation velocity of ANFO mixtures under the

conditions of the experiments. The ANFO charges prepared from AN

granules of oil absorption below 2.5 cm3/100 g did not detonate at all. The

threshold of AN oil absorption above which a stable detonation of ANFO

could be observed was 2.5–3.0cm3/100g and the measured detonation

velocity was 1.6–1.7 km/s. Along with the increase of AN porosity, the

detonation velocity of ANFO increased up to 2.7 km/s at a level of oil

absorption of 12–15 cm3/100g. The described

effect took place in spite of the fact that the increase of AN prills porosity


124

caused a significant drop of the ANFO density from 0.9 g/cm3 (for 2.5

cm3/100 g) to 0.6 g/cm3 (for 15 cm3/100 g of oil absorption of prills).

The dependence of ANFO detonation velocity on its density, as observed in

the experiments, is quite different from that existing in the case of typical

high explosives where the detonation velocity of explosives is always

proportional to their density. The observed effect can be explained by the

dominant influence of the AN prills structure, which overrides the influence

of density. Such behavior is observed also in other explosive mixtures

consisting of two ingredients – an oxidizer and a fuel. The described

experiments prove that the detonation velocity of ANFO mixtures depends

strongly on the AN particle size. The charges of ANFO prepared from fine

grade nonporous prills are capable to detonate (under the conditions of the

experiment) only in the case of prills below 1.0 mm. In that case the

dependence between detonation velocity and particle size is the same

compared with other types of explosives. The charges of ANFO mixtures

prepared from nonporous, ground prills detonate with much higher velocity

compared to charges prepared from granulated AN of similar particle size.

The explanation of the fact, apart from the lower density of the mixture,

might lie in the higher contact surface of reagents – oxidizer and fuel in the

case of ground formless grains, which enables a fuller run of exothermic

reactions. The influence of fuel oil content on the detonation velocity of

ANFO seems evident. The strong dependence observed for high-porosity

granules and the weak dependence for low-porosity granules could be

explained in the light of creating an energetically optimal system (for

stoichiometric mixture) within the volume of a single, porous granule.

7. References

[1] D. Buczkowski, B. Zygmunt, Influence of Ammonium Nitrate Prills Porosity

and Dimensions on Detonation Velocity of ANFO Explosives, 5th Int. Seminar

New Trends in Research of Energetic Materials, Pardubice, Czech Republic,


125

April 21 – 23, 2003.

[2] B. Zygmunt, Explosive Properties of Ammonals and Hydroammonals, 3rd Int.

Autumn Seminar on Propellants, Explosives and Pyrotechnics, Chengdu,

China, October 5 – 8, 1999.

[3] B. Zygmunt, The Detonation Properties of Explosive-Water Mixture,

Propellants,Explos .,Pyrotech. 1982, 4, 107.

[4] D. Buczkowski, W. Pa˛gowski, B. Zygmunt, Evaluation of the Influence of

Modificating Substances on Resistance to Detonation of Fertilizer grade

Ammonium Nitrate, 3rd Int. Seminar New Trends in Research of Energetic

Materials, Pardubice, Czech Republic, April 12 – 13, 2000.

CHAPTER – 4

BULK LOADING SYSTEMS FOR EXPLOSIVES

The explosive are delivered in bulk tankers or onsite mixed and delivered,

with following advantages:

i) The cost of packing material can be saved.

ii) Transport of 15% and above water present in packaged explosives over

long distances is avoided,

iii) Since mostly these types of explosives are in a pumpable form at the

time of delivery, the loading rates can be much higher than conventional

packaged slurries or emulsion, and

iv) Since the explosive is pumped into the borehole, a better coupling is

ensured with the boreholes walls.

Both slurries and emulsions are ideally suited for these types of operations.

The following considerations are also quite relevant in the context of

selecting these types of systems.

1. Normally these types of operations are more suited for large operations

and naturally problems may arise when we come across limited size of

individual mines and quarries. For smaller units the shared services of a

pump trunk and its support facilities may be necessary. In such

situations to operate this system effectively a strictly prearranged

schedule of hole loading may be necessary.

2. Not only blasting must be arranged to a given schedule, but also the size

126

Chapter 4 Bulk Loading Systems for Explosives

of the shots tailored to suit the capacities of pump trucks and support

facility.

3. Another relevant consideration is about the usage of increased explosive

charge per hole (upto 20%) because of the better coupling afforded by

these type of explosives. This may pose vibrational problems if mines are

near urban areas.

The above operations can be broadly classified into three categories:

1. Bulk trucks (On-Site Mix Trucks).

2. Pump Trucks.

3. Bulk Mix Trucks (for ANFO).

The details of these are given below:

1. BULK TRUCKS:

Bulk Slurry/Emulsion On Site Mix Trucks: Here, essentially the raw

materials for making the explosive likes:

i) Ammonium Nitrate, Sodium Nitrate solutions from a Support

Plant.

ii) Fuels like TNT, and Aluminium Powder,

iii) Gelling agents,

iv) Cross linking agents, and

v) Gassing agents

Are carried in trucks and the explosives made at the blasting site.

2. PUMP TRUCKS: These are delivery trucks of pre-mixed slurry or emulsion.

127


3. BULK MIX TRUCKS OR ANFO TRUCKS: These essentially mix and deliver ANFO/Aluminium Powder explosive.

4. BULK TRUCKS OR ON-SITE MIX TRUCKS:

There are two major categories:

i) One using the high explosive sensitizer TNT, and

ii) Others sensitized mostly by air bubbles.

i) High Explosives Sensitized Bulk Trucks: The Ammonium Nitrate, Sodium Nitrate, etc. are taken in the forms of

solution and the various ingredients like guar gum, aluminium, etc. are

auger-fed to a common mix funnel which can optionally have a mixer to get

better mixing. The product from the mixer is pumped from the mixing funnel

to boreholes. Some times aluminium and other fuels can be premixed a part

of the oxidiser AN can be added in dry form. For many mixes the liquor can

also be pregelled. This has the advantage of putting a substantial quantity of

air bubbles into the liquor. Additionally gassing agents are also premised

with fuels to give various densities for final products. Since controlled

feeding of TNT requires a vibrating pan feeder, these types of trucks

necessarily have provision for leveling the trucks using hydraulic jacks. A

simple diagram describing the location of various raw materials, mixing

bowel and hose bundle is shown in Fig.1.

128


Fig. 1 Schematic Layout of an On Site Mix Truck Regulation In the western countries the following regulations are followed with the type

of operation:

129


In the case of High Explosive Sensitised (e.g. TNT) Slurries, the truck is

classed as a mobile factory. Table of distance apply based on the quantity of

High Explosive aboard on arrival and line of sight distance e.g. ‘For 5000

lbs. of TNT -400ft.’ No man other than those on licence can work within this

line of sight radius. Because of this rule the use of On Site Mix Trucks may

not be possible in certain circumstances. Passing radius is generally 25 ft.

ii) Air Bubble Sensitized Bulk Trucks: These trucks are also very similar to HE sensitized bulk trucks, consisting of

tanks or bins for various raw materials like Ammonium Nitrate, Sodium

Nitrate, Aluminium, gelling agents, cross linking agents, etc. Here more

rigorous control on amount of gassing agents added and extent of gassing of

final product is essential as these explosives depend for sensitivity mostly on

gas bubbles in the absence of high explosive sensitizer.

Regulations: In the case of NCN mixture, only passing radius applies. 5. PUMP TRUCKS These are delivery trucks of pre-mixed slurry or emulsion. The slurry or

emulsion is produced in a steam jacketed ribbon mixer or by a continuous

process at a plant, partially gelled to eliminate segregation but not so much

that it cannot be handled. It is stored in large tanks and then pumped into

bulk carriers. From these carriers it is pumped out at the mine site.

Additional cross linking agents, gums, etc. may be added here as it is

pumped out. Gum may be added as dispersion in glycol for easy addition.

Some tipping system to get all slurry/emulsion out is necessary as these

explosives leave a heal in the tanker in normal position. The details of a

Pump Truck are given schematically in Fig. 2.

130


Fig. 2 Schematic diagram of a Pump Truck

Regulations: In the western countries, both NCN and HE pump trucks as opposed to

onsite mix trucks are classified as bulk carriers and passing radius is the

only regulation that applies.

Since the two already described systems deliver same type of explosive, vis.

Emulsion or slurry explosive, it is relevant to discuss the merits and

demerits of these two systems in some details.

6. ON SITE MIXING – vs. – PUMP TRUCKS

Comparative Advantages & Dis-Advantages:

i) The main advantage of pump truck system is the lower investment for

large operations utilizing several trucks. The is because pump trucks

are all of very simple design consisting of essentially a big tanker to

carry the explosives, suitable pumping system for getting reasonable

pumping rates and a ‘hose role’ system. One can optionally have small

131


mixing bowl to add additional gums, cross linker, etc. But in the case of

on-site mixing systems, a number of containers for various raw

materials like, Ammonium nitrate/sodium nitrate liquor, gums,

aluminium, cross linker, TNT have to be provided with additional

facilities like guar feeders, vibratory feeders, etc. to control the feeding

of most of raw materials. Hence the cost of an on-site mix truck is

much higher than a single pump truck.

ii) In case of site mix truck, since all the ingredients have to be fed in a

controlled manner it calls for sophisticated control mechanism and

control panels. Especially if the support plant is some distance away

from the mines. The mix truck has to travel a long distance on public

roads. The break downs of ‘mix-trucks’ with lot of sophisticated

controls and equipments on the trucks are more serious.

The routine servicing and maintenance for breakdowns can be carried

out on a simpler work shop in the case of pump trucks.

iii) Fore HE Sensitized mixes, TNT feeding poses more problems, since

controlled feeding of TNT is done only by vibratory feeders. This type of

trucks should have provisions for leveling the trucks using hydraulic

jacks. Since temperature affects this type of feeding, the whole systems

has to be heated to get a consistent feeding rate. Composition based on

Aluminium or air bubble sensitization have critical densities beyond

which they will not shoot. These are considerably lower than HE

sensitized slurries and so have to be controlled very carefully on the

site. In the case of PUMP TRUCKS the explosives are made in a plant

where quality control checks on raw materials and finished products

can be done in a better manner and final density, percentage of

Aluminium checked in a reliable fashion in a well equipped quality

control laboratory.

132


iv) The gassing agents and their uniform distribution throughout the

matrix should be precisely controlled at the site in the case of on site

mixing trucks. Even a slight variation in temperature, pH, etc. can

affect this phenomena. In case of Pump trucks, such precise control is

much easier to achieve as most of the air bubbles are incorporated into

the system at the manufacturing site itself.

v) Pump trucks give products having much faster felling and cross linking

rates than on site mixed products and this can be more desirable in

watery holes. In the case of onsite mixed slurries it the slurry is too

thin, there can be substantial losses in cracks in the rocks. For wet

holes, the product needs to cross link rapidly to avoid water take-up.

vi) The aluminium and water content, density and explosive content

control the sensitivity and strength. These depend on quality of raw

materials whose specifications should be checked on routine basis.

These operations may be more advantageously done in a well equipped

plant with a good quality control laboratory.

vii) In the case of on-site mixing, the support plant is a non-explosive unit

whereas in case of pump trucks, it is an explosive manufacturing

facility attracting regulations.

viii) In the case of on site mix plant it is possible right at site to easily vary

the formulation to give a range of strengths and densities. Such a thing

is not possible with pump trucks; however, this disadvantage can be

offset to a certain extent by having compartments in the pump trucks

for different explosives with different strength and density.

7. AMMONIUM NITRATE DRY MIXES

These consist essentially of hoppers for AN etc., augers, a mix chamber and

an air system to convey the mixed product into the borehole through loading

133


hose and special valves to stop air bleeding back into the system. Augers

may also be used to carry the mixed product to the borehole instead of the

air system. Aluminium is usually stored in another chamber and augered

into the mixing chamber. Even for horizontal blast hole loading upto 120’

depth, special pneumatic loading systems have been developed. The

operational details are explained in Fig. 3 and Fig. 4.

Fig. 3 Schematic of Amerind AN/A1/FO Mix Truck

134


Fig. 4 Schematic of the Feeding and Mixing Arrangement

The following are some of the important consideration on the usage of

these types of products:

1. Prilled Ammonium Nitrate of suitable density and oil retention

properties along with fuels like fuel oil, aluminium are usually used.

2. Use of excessive aluminium is to be avoided as it leads to more fraction

of solid product of reaction, viz. Al2O3 which has energy trapped in it and

thus not available for doing useful work. This is explained as given

below:

When aluminiuim is added to AN the following reaction takes place: 2 Al + 3 NH4NO3 → 3N2 + 6H2O + Al2O + 1.65 Kcal/gm (1)

At higher percentages of Aluminium, Oxygen from the steam is stripped off

and Al2O3 is formed.

135


2 Al + NH4NO3 → N2 + H2 + Al2O3 + 2.3 Kcal/gm (2)

At low Al, percentage, the first reaction is more predominant and has

increased thermochemical energy as compared to ANFO (900 cal/gm).

However, when the second reaction takes place, Al2O3 is formed and this

has energy trapped in it.

On examination of equations (1) and (2), one can see that in the case of (1)

when Aluminium and AN are mixed in the mole ratios of 2:3 the energy

obtained is 1.65 Kcal/gm and in the case of (2) when Aluminium and AN are

mixed in mole ratio of 2:1 the energy output rises to only 2.3 Kcal/gm of

mix. In other words, in the first case of a mixture containing only about 18%

aluminium, the energy output is 1.65 Kcal/gm mixture, whereas the other

mixture which contains about 40% aluminium (more than double) gives only

2.3 Kcal/gm of energy. From blasting studies using explosives yielding

appreciable solid products, it has been noted that the thermal energy

trapped in the solid products is not at all available for useful work.

3. Some Specifications of AN and Aluminium used for Site-mix ANFO

Type of Products:

AMMONIUM NITRATE

Size: 1 to 3 mm.

Density: Usually around 0.82 to 0.90 gm/cc

Porosity: Minimum oil retention of 6% by weight. Free flowing in nature for

easy auger feeding.

Hardness: Should not excessively break down during auger feeding, in

temperature cycling and during handling. Fines can choke the auger by

depositing in the gaps between stator and rotor.

136


The prills should not go to detonation in burning test.

8. ALUMINIUM: Aluminium also should be of free flowing nature having a consistent density

so that the calibration of metering devices need not be changed often. These

should also be dust free to avoid any fire mishap.

Recently other fuels have also been tried especially when the smell of FO can

be objectionable. Fuels like glycol, methanol, ground coal, FeSi, Fep, etc. are

the other candidates.

Use of these products mostly may require blast hole dewatering for which

several types of dewatering units are available. Generally these dewatering

units are hydraulically driven though electrical and pneumatic units are

also available. The unit may be mounted on a pick up or attached to the

front of ANFO truck. Generally pumps of rugged stainless steel construction

are used; however, they are affected by Slurries in the hole or adverse pH

conditions. Dryliners are also used for preventing influx of water into the

borehole loaded with dry mix blasting agent. These should be so constructed

that they are water proof, flexible and will not crack when exposed to low

temperature.

9. COMPARISON BETWEEN SLURRIES AND DRY MIXES:

Figure 5 shows the effect of Aluminium addition to ANFO on its weight

strength. It also gives weight strength values for typical slurries for both TNT

sensitized and NCN varieties. From this it can be seen that the dry mixes at

the same percentage of aluminium have consistently higher weight strength

as compared to slurries. This is on account of diluent effect of water in

slurries which does not react during detonation and therefore reduces the

137


energy output by about same percentage as it is present. But on a volume

basis or bulk strength basis – which is what one is interested in blasting the

case is different as can be seen from Fig. 5. The values plotted in Fig. 5 have

been taken from Table 1. One should remember that while the density of

Aluminised ANFO varies from 0.85 to 0.93 gm/cc, the density of NCN

slurries are taken as follows:

1% Al NCN Slurry – 1.25 gm/cc; 7% Al NCN slurry – 1.30 gm/cc; 10% Al

slurry 1.35 gm/cc.

The densities mentioned above for NCN slurries are the densities that would

exist after they have load on them and they are compressed somewhat at

the bottom of a 50’ bore hole. From this it can be seen, for example, that a

7% Al NCN slurry bottom loaded will be equivalent to 16% Aluminium dry

mix.

Therefore, it is possible to match the energy/ft of hole of any NCN slurry

with Aluminised dry mix to produce the same blasting results. In view of the

above under the prevailing costs of raw materials in India, the economies

favours the use of slurry.

It may be pointed out that the choice of particular system among the various

above mentioned options depends on consideration of the following major

factors.

1. Strata conditions with particular reference to site geology.

2. Size of operations, the flexibility and specific nature of services

required.

3. Location of explosive plant vis-à-vis the mines.

4. Watery conditions of borehole.

138


5. Energy requirements per foot of the borehole for proper fragmentation

and throw, and

6. Cost of raw materials as prevailing in each country.

Fig. 5 Bulk Strengths of Dry Mixes and Slurries with different

Aluminum contents

139


10. EMULSION EXPLOSIVES INTRODUCTION This is another family of explosives based on the same broad principles on

which slurry type of explosives have been developed. Similar to slurry

explosive, this also combines the rapid combustion and water resistant

properties of nitroglycerine with safety characteristics of power explosive.

General Chemistry of Emulsions When oil and water are mixed together without an emulsifier the initial

dispersion of oil is quite unstable. (the oil rapidly coalesces until complete

separation into a layer of oil and water results). This can be anticipated as

the increase in surface area by the dispersion of the oil or water greatly

increases the free surface energy of the system and at such a high energy

level the system is unstable. Free surface energy is dependent on both

surface area (capacity factor) and the interfacial tension (intensity factory). If

a stable emulsion is to be produced, it is necessary to add a third material

i.e. an emulsifying agent, which by its presence at the interface, prevents

coalescence of the oil or water gloubles. These emulsifiers may be surface

active compounds (monoglycerides and diglycerides). Gums (gum acacia,

methyl cellulose) or finely divided clays (cocos, gums, etc.). Though these

three act in different ways, the basic mechanism operative are as follows:

1. Reduction of interfacial tension. 2. Formation of rigid interfacial film. 3. Electrical charges.

At best conditions a surface active compound lowers the interfacial energy

by a factor of 20 or 25 which is rather small when one considers that the

interfacial energy of an emulsified system is 106 times higher than that of

140


broken emulsion.

The most important factor is the formation of rigid interfacial film. It is a

known fact that the surface active agent forms a rigid film between

immiscible phases and the film may act as a mechanical barrier both to

flocculation and coalescence of the emulsion droplets.

The third factor which contributes to stability is electrical charge

contributed by a certain family of emulsifiers – anionic and cationic types.

For this phenomena to happen the molecules that concentrate on the

interface must be dipoles, i.e., their geometrical and electrical centres are

not coincident and the molecules are electrically asymmetrical thus acting

like small magnets.

Selection of emulsifier for a particular system is still an art rather than a

science and lot of experimentation goes in before arriving at the best. Still

HLB method of selection which depends on the hydrophile – lipophile

balance of an emulsifier is the best guide available to determine the type of

emulsion formed and the selection of different emulsifiers for various tasks.

Several techniques are available for production of emulsions ranging from

batch methods using propeller type of impellers to continuous type using

colloid mills. Ultrasonic energy has also been used to form stable emulsions.

The art and science of forming emulsions of explosive type are still more

complicated by the presence of electrolytes which tend to oppose the

electrical charges of emulsifiers. Moreover, these emulsions are of high

internal phase ratio types whose stability is very suspect.

141


11. TYPES OF EMULSIONS There are two types of emulsions

i) Oil emulsified in water type, and

ii) Water emulsified in oil type.

12. RAW MATERIALS USED IN EMULSIONS Both these types of emulsions have been used to produce explosives. The

raw materials for these emulsions consist of the following ingredients:

i) Oxidisers like Ammonium Nitrate, Sodium Nitrate, Calcium Nitrate,

barium Nitrate, Sodium Perchlorate, Ammonium perchlorate.

ii) Fuel sensitisers like, Methyl amine nitrates, DNT Oil, Aluminium,

iii) Fuels like waxes, mineral oils,

iv) Water

v) Bulking agents like glass microballoons, expanded perlites, and

vi) Emulsifiers.

One can see all the above raw materials are by nature non explosive. Hence,

enormous levels of safety can be built in production, distribution and also

handling of explosives in the field. The safety levels of slurry and emulsion

explosives are given later.

How Emulsions are Sensitized: The presence of tiny air bubbles in the size range of 20 to 100 microns (Av.

60 microns) is responsible for sustaining a detonation wave created by a

detonator or booster in the explosive column. The mechanism by which they

sensitize the explosive is by providing ‘hot spots’ having temperatures

greatly exceeding that of the rest of the explosives. This is achieved by the

142


adiabatic compression of these air bubbles by the high pressures of

advancing shock wave by which the gases are heated to high temperatures

of the order of 1000°C. This heats the inner surface of the bubble starting a

rapid decomposition reaction. By having a multitude of such air bubbles the

shock wave is sustained.

What Are the Advantages for Practical Users i) As mentioned earlier, (the ingredients which got to make this types of

explosives are anon explosive in nature.) The sensitivity is derived

mainly by the incorporation of tiny air bubbles or microballoons in the

diameter range of 0.1 mm. As mentioned above, these tiny bubbles

collapse when they are struck by the shock wave with the result the

temperature in the explosive substance becomes sufficiently high

locally to start the rapid explosive combustion. The absence of high

explosive sensitisers render the emulsions or slurry explosives highly

inactive to other forms of stimuli such as friction and fire. This provides

them the high degree of handling safety.

ii) A part of the sensitivity is also derived by the nature of intimate contact

between the oxidiser solution and the fuel phase consisting of waxes

and oils. When magnified, emulsion looks like honeycombs in which

honey is replaced by Ammonium Nitrate solution. The diameter of cells

is no more than a few thousandths of a millimetre and the thickness of

wax membrane separating the cells is less than one thousandth of a

millimeter. Such an intimate contact between the oxidiser and the fuel

helps in better sensitivity and full release of energy.

iii) Both types of emulsions can be made to be highly water resistant by

the very nature of formation of these products. In the case of ‘water in

oil type’ emulsion, the oxidiser solution is protected by the oil barrier

143


against evaporation and also against penetration of external water (e.g.

water in boreholes, water under pressure) into the explosives. In case of

“oil in water” type emulsion, external water phase is usually gelled to

get protection from water.

iv) Safety during manufacture, transportation and usage: During

manufacture, lot of safety advantages are derived from the above

mentioned properties, since emulsion’s basic matrix can be considered

to be extremely safe until the final stage of production process when

microballoons are added. The finished product is also equally safe for

transportation and usage as evidenced by safety tests mentioned in

detail later in this paper.

v) As in the case of slurry explosives, the consistency of these explosives

can be varied. This is done by changing the nature of oil phase and the

emulsifier. Thus from a very firm consistency similar to margarine in

different qualities(which is suitable for cartridging) to a lesser viscous

form which can be pumped and delivered to customers in road tankers

(similar to pumpable slurries) can be easily obtained.

General comparison between “NG” ‘slurry’ and ‘Emulsion’

Explosives:

NG Explosives Slurries Emulsions

Water resistance

Good Better than NG Explosives.

Excellent, uch Better than other Categories.

Safety Fair Much safer than NG explosives as Given under Safety tests

Equivalent to slurries.

144


Wt. strength All ranges Possible. Here also any desired range of strength can be obtained by simple and easy manipulating of composition.

The flexibility and ease of obtaining products with different wt. strengths is more in case of slurries and emulsions.

Quality of raw Material required

Good quality raw materials are needed and any small deviation in raw material specification will adversely affect the product quality.

This can tolerate slight fluctuation in the quality of raw materials.

The raw material should pass through stringent quality tests.

Shelf life Good Fair Fair

Influence of physical properties of ammonium nitrate Ammonium nitrate (AN) and AN-based explosives have been widely used as

industrial explosives or energetic compositions such as ANFO, emulsion

explosives or amatol (AN/TNT) etc. They are explosives known for their non-

ideal detonation behaviour. This is shown, for example, by the detonation

velocity which does not easily reach theoretically predicted values.

Explosives behave non-ideally between the critical diameter (dc) below which

a steady detonation wave cannot be sustained, and the minimum diameter

(dm) above which the detonation is ideal. AN and AN-based explosives are

typical “non-ideal” explosives because they have large values for dm and

relatively small values for dc. For most practical conditions they will never

reach the ideal behaviour as predicted by thermo hydrodynamic theory.

Generally the non-ideal detonation behaviour is explained by the relatively

low decomposition rate of AN, which causes a wide reaction zone, in

combination with lateral heat losses and rarefaction waves which extinguish

the decomposition reactions ( Cook, 1958).

145

http://www.sciencedirect.com/science/article/pii/S0950423001000419%23BIB2


Miyake eta l ( 2001) investigated to obtain a better understanding of the

non-ideal detonation behaviour of AN-based explosives. A series of

experiments were carried out to study the influence of the physical

properties of AN on the detonation velocities of ANFO.

13. RESULTS AND DISCUSSION The measured detonation velocities of ANFO are shown in Table 2. In this

table the capital letters of the sample name (A, B, . . .) indicate the sample of

ANFO prepared with the AN shown in Table 1 with the lower-case letters (a,

b, . . .), and the capital letters with # show the 12 month aged ANFO. Two or

more shots were carried out in the same conditions for unaged ANFO and

one shot for each aged ANFO. A stable detonation was observed in all shots

and the indexes of non-ideality of detonation (Dobs/Dcal) were determined.

Table 2. Experimentally observed and calculated

detonation velocities of ANFO

Sample Loading density

(kg/m3) Dobs

(km/s) Dcal

(km/s) Dobs/Dcal

A-1 845 2.85 4.95 0.58 A-2 845 2.95 4.95 0.60 A# 855 2.90 4.99 0.58 B-1 870 3.35 5.04 0.62 B-2 870 3.20 5.04 0.63 B# 865 3.20 5.02 0.64 C-1 850 3.30 4.97 0.66 C-2 850 3.50 4.97 0.70 C# 875 3.25 5.06 0.64 D-1 840 2.90 4.93 0.59 D-2 840 2.95 4.93 0.60 D# 825 3.05 4.88 0.63 E-1 850 3.40 4.97 0.68 E-2 850 3.45 4.97 0.69 E# 860 3.30 5.01 0.66

146

http://www.sciencedirect.com/science/article/pii/S0950423001000419%23TBL2

http://www.sciencedirect.com/science/article/pii/S0950423001000419%23TBL1


F-1 860 3.75 5.10 0.74 F-2 865 3.85 5.10 0.75 F# 885 3.65 5.10 0.72

13.1 Influence of the pore diameter and volume Figure 6 shows the influence of the mode pore diameter of AN on the index

of non-ideality of detonation (Dobs/Dcal) of ANFO samples A∼C. Sample A

showed the detonation velocity of 2.85 and 2.95 km/s and they were

estimated as 58 and 60% of the calculated detonation velocity. The

detonation velocity increased with the decrease of the mode pore diameter,

and the highest detonation velocity was observed as 3.50 km/s when the

mode pore diameter was 4.5 μm and it corresponded to 70% of the

calculated value.

Fig. 6.Influence of the mode pore diameter of AN

on the detonation velocity of ANFO samples A∼C.

Figure 7 shows the influence of the total pore volume on the detonation

velocity of the same samples of ANFO, and the Dobs/Dcal value increased with

147

http://www.sciencedirect.com/science/article/pii/S0950423001000419%23FIG5


the decrease of the total pore volume. Since the total pore area of samples

A–C is comparable, the number of the pore is considered as almost the same

level. So that it is obvious that the pore diameter has a strong influence on

the detonation velocity. Moreover, as the pore diameter distribution of

samples A and B was seen as a normal distribution, the distribution of

sample C had two peaks and looked as if it had two distributions which are

mode pore diameters of 4 and 8 μm. From the plot shown in Fig. 4 it may be

concluded that pores of diameter less than 8 μm act more effectively than

larger pores.

Fig. 7. Influence of the total pore volume of AN

on the detonation velocity of ANFO samples A∼C.

In general, the pore in the particle is considered to act as the hot spot when

the shock wave comes in and it provides the local high temperature and

high pressure, so that it makes the detonation propagation easier. The

formation of the hot spot is influenced by the size and the volume of the

pore which exists after mixing with the oil. If the pore is relatively small, the

oil is restrained from permeating into the particle because of the surface

148



tension, and then the rest of the voids could be hot spots. It is possible to

consider that the role played by the pore is different depending on its size,

and a smaller pore has a strong influence on the detonation propagation.

Since ANFO is a typical heterogeneous composition, understanding the role

of the pore, i.e. the role of hot spots, is quite difficult. For the understanding

of the detonation mechanism of ANFO, an understanding of both the

reaction mechanism between AN and fuel oil, and of hot spot at the pore, is

needed. As the distribution of the pore dimensions may also play an

important role, the effective pore size for the detonation propagation and its

spatial distribution should be subject to further investigation.

13.2 Influence of the particle diameter and the specific surface area

Figure 8 shows the influence of the specific surface area on the detonation

velocity of ANFO samples D–F. Since samples D–F were prepared with the

same batch of AN, the physical properties were considered as the same in

each ANFO except the particle diameter. The detonation velocity increased

with the decrease of the particle diameter, i.e. the increase of the specific

surface area, and the highest detonation velocity was observed as 3.85 km/s

when we used the smallest particle diameter (<0.85 mm). The value

corresponded to 75% of the calculated value.

149



Fig. 8. Influence of the specific surface area of

AN on the detonation velocity of ANFO samples D∼F.

Based on the concept of mechanochemistry, shear or collapse of the particle

due to the high pressure of the shock wave causes the increase of the

internal energy and raises the activity of the surface (Kubo, Jimbo, Suito,

Takahashi, & Hayakawa, 1979). During shock compression, the

intergranular forces play a major role in determining the yield strength of

the material and it fails when the yield strength is exceeded.

Concerning the hot spot concept, several models have been proposed such

as the pore collapse mechanism or the mesoscopic shear mechanism (Lee et

al., 1985 and Cheret, 1993). As both models are applicable to the ANFO

detonation, for detailed discussion additional information such as the

curvature of the detonation front, the reaction zone length and the reaction

rate of AN, as well as the microscopic information, are needed.

As the physical properties of AN such as the pore dimensions and their

distribution play an important role, the hot spot mechanism as well as the

150







reaction kinetics of AN and/or ANFO are the key for understanding the non-

ideal detonation behaviour of ANFO.

13.3 Influence of aging The influence of aging can be discussed by comparing aged samples with

unaged samples. Samples A and B did not show a significant change of the

detonation velocity between the with and without aging, and samples C, E

and F showed a slight decrease of the detonation velocities after 12 months

aging at room temperature. It is considered that the quality of the oil within

the AN particle degraded during the 12 months and the combustion

performance of the oil decreased. However, sample D showed an increase of

the detonation velocity. AN sample d originally had a larger diameter but

small specific surface area and showed a relatively lower detonation velocity.

If the contact between AN particle and oil becomes better during the 12

months, the reaction between the two materials proceeds smoothly and it

may lead the detonation velocity of sample D to become higher.

Summary

From the detonation velocity measurement of ANFO prepared with six kinds

of AN which had different pore dimensions and particle sizes, the following

conclusions can be drawn:

For the same particle diameter ANFO:

(1) The detonation velocity increased with the decrease of the mode pore

diameter, and the highest detonation velocity was observed as 3.50 km/s

which corresponded to 70% of the theoretically predicted value by the

CHEETAH code with the JCZ3-EOS.

(2) The Dobs/Dcal value increased in the manner of a polynomial function with

the decrease of the pore volume. And the pores with diameter less than 8

151


μm act more effectively than larger pores.

For the same total pore volume ANFO:

(3) The detonation velocity increased with the decrease of the particle

diameter, i.e. the increase of the specific surface area, and the highest

detonation velocity was observed as 3.85 km/s when the smallest particle

diameter was used. The value corresponded to 75% of the calculated value.

Regarding the 12 months aging at room temperature:

(4) Although the detonation velocity of most of ANFO showed no changes or

a slight decrease, sample D showed a slight increase of the detonation

velocity. The physical properties of AN have a strong influence on the

sensitivity and the propagation behaviour of the detonation of ANFO.

14. SOME IMPORTANT EXPLOSIVE PROPERTIES Emulsions can be formulated a variety of diameters with a range of

sensitivities. Addition of Aluminium, while increasing the energy values,

does not alter to a great extent the other properties like velocity of

detonation, sensitivity, shelf life, etc. The addition of Aluminium is primarily

for the thermochemical energy of these explosives. (Fig. 9).

152


Fig. 9 Thermochemical Properties of Emulsions

14.1 Velocity of Detonation The unconfined velocities of emulsions are high, particularly considering the

fact that emulsions do not contain any high explosive sensitizers. The

change in diameter of explosive has only a small influence on the VOD.

Though velocity and diameter are linearly related, the differences in velocity

between a 1” dia. Explosive and a 4” dia. Explosive is in the range of 4000

153


fps only.

14.2 Detonation Pressure The detonation pressure is dependent primarily upon density and velocity.

Since these types of explosives have high velocities, they also display high

detonation pressures. The detonation pressure as measured by the

Aquarium technique are in the range of 100 to 120 kbars.

14.3 Bubble Energy

The bubble energy of the emulsions represents a practical demonstration of

the contribution of Aluminium to thermochemical energy. Some of the

typical values are displayed graphically in Fig. (10).

154


Fig. 10 Relative Bubble Energy of Emulsions vs. Aluminum

15. SOME SAFETY CHARACTERISTICS OF EMULSION EXPLOSIVES Some of the finished products of emulsion explosives have been subjected to

various types of safety tests with the following results.

155


15.1 Drop Impact Test This is found by the standard fall hammer tests. The results can be

summarised as follows:

Explosive Fall wt. Fall wt. Fall Energy (kg) (kg) (kg.m.) Blasting Gelatine 1 0.2 0.2 Ammonium Nitrate 1 0.2 0.2 Nitroglycol gelatin Nitroglycerin sensitized 5 0.2 1.0 Powders and permitted Explosives Powder explosives 5 0.4 2.0 Emulsion and Slurry No explosion even if a 2 kg.wt. is dropped Explosives from 2.25 M (Fall Energy 4.5 kg. m.) 15.2 Differential Thermal Analysis DTA registers small temperature differences which appear during

simultaneous heating of the sample and a comparison substance. In this

way, all physical and chemical processes which are accompanied by an

additional absorption or evolution of heat by the substances are recorded.

Examples of such processes are changes taking place in the crystal lattice,

melting, evaporation, chemical reaction and decomposition. Emulsion

explosives when subjected to these tests showed no exotherm until water

had boiled off. The dry residue after loss of water burned at 250 to 255°c.

156


15.3 Friction Tests When subjected to standard TORPEDO FRICTION Tests, the emulsions did

not show any reaction when a 0.550 kg torpedo from a height of 2.25 cms is

allowed to slide over a 2 gm sample of the explosive.

15.4 Thermal Stability This test gives an indication of the tendency for explosives to go from

deflagration to detonation when they are heated, under circumstances

where the reaction products from a small first reaction do not have chance

for free escape. The explosive is confined in a steel casing closed with a

nozzle plate that has a hole of 1 mm diameter. The steel casing is heated up

electrically and the total added energy at detonation is a figure of the

thermal sensitivity. The most sensitive emulsion explosive gives a value in

the range of 20 KJ. Whereas a typical dynamex explosive gives 13 KJ: slurry

explosive falls in the range of 27 KJ.

The above are the typical safety tests conducted for evaluating the safety

levels of any type of explosives. Emulsions have been subjected to other

stringest safety tests like PROJECTILE IMPACT TEST & IMPULSE TEST and

are found to possess similar safety levels as water gels.

16. PRODUCT RANGE POSSIBLE WITH EMULSION EXPLOSIVES: It is possible to formulate a wide range of products with different densities

and energy levels for different specific applications ranging from underwater

application in small dia holes to pumpable emulsions in large dia holes

sensitive to no. 8 cap at ranges of 20° F to 10°; booster sensitive explosives

in the diameter range from 50 mm to 200 mm, cap sensitive explosives from

25 mm to 200 mm (boosters). The strengths can be varied from 40% to 90%

wt. strengths relative to blasting gelatine. It is also possible to develop

157


permitted type emulsions for use in gassy coal mines. The following table

gives the wide range of products available.

Diameter Sensitive Booster Pumpable/ Permitted Cap. 6/8 sensitive packages non-permitted 25 mm to 40 mm Packaged Non-permitted 32 mm (permitted) Packaged Permitted 50 mm, 75 mm Packaged Non-permitted 75 mm to 200 mm Packaged Non-permitted 75 mm to 200 mm Pumpable Non-permitted Thus emulsion is another family of explosives similar to slurries in many of

the basic principles and applications, developed with a view to built in lot of

safety in the manufacture, transport and use of explosives. Both these types

contain water in the composition which makes them quite safe during the

above operations. But due to inherent differences in the chemistry of

emulsification and gelling /cross linking both these types have unique roles

to play in the explosive industry.

17. CONCLUSION As a conclusion it will be relevant at this stage to discuss some detail, the

various range of products possible in the different systems mentioned above

with the main emphasis on density, weight strength and bulk strength vis-à-

vis the composition.

158

CHAPTER – 5

TESTS FOR EXPLOSIVES 1. Ballistic Bomb Closed Vessel The ballistic bomb (pressure bomb, manometric bomb) is used to study the

burning behavior of a Gunpowder or Propellant charge powder. It consists of

a pressure-resistant (dynamic loading up to about 1000 MPa (10000 bar)

hollow steel body that can be bolted together and has a hole to adapt a

piezoelectric pressure transducer. The pressure p in the bomb is measured

as a function of time t. As a rule, studies of powder in the pressure bomb

are carried out in comparison with a powder of known ballistic performance.

They are very useful both in the development of powders and in production

monitoring. If the dynamic liveliness L (= 1/pmax * dlnp/dt) is determined

as a function of p/pmax from the primary measured signal, then for a

defined powder geometry the parameters characterising its burn-up, the

linear burning rate e˙ (W Burning Rate) and the pressure exponent a can be

determined. Pressure bomb shots of the same powder at different charge

densities d (= mass mc of powder/volume VB of the pressure vessel) enable

the specific covolume h of the combustion gases from the powder and the

force f (powder force) of the powder to be determined in addition. From

these, if the W Heat of Explosion QEx of the powder is known, the value of

the average adiabatic coefficient æ (= 1 + f/QEx) of the combustion gases,

which is of interest for the ballistic performance, can be derived. Since the

combustion gases of powders satisfy Abel’s equation of state to a good

approximation, it is possible by using the auxiliary parameters (rc) density of

the powder) D : = mc/(VB * rc) ‘normalised charge density’ (1) x : = (1 – hrc) *

D/(1 – D) ‘real gas correction term’ (2) F : = frcD/(1 – D) ‘characteristic

pressure’ (3) to write the relationship between the pressure p in the

manometric bomb and the burnt volume proportion z of the powder as

z(p/pmax) = p/pmax/{1 + x(1 – p/pmax)} (4) and p(z) = F * z / (1 + xz). (5)

159

Chapter 5 Tests for Explosives

Accordingly, the maximum gas pressure achieved at the end of burnup (z =

1) is calculated as pmax = F / (1 + x). (6) The dynamic liveliness L is

calculated from L = S(0) V(0) * f(z) * e˙(pref) pref * [ p pref ] a–1 * 1 + x (1 + xz)

2 (7) S(0)/V(0) is the ratio of the initial surface area to the initial volume of

the powder, f(z) is the shape function of the powder, which takes account of

the geometrical conditions (sphere, flake, cylinder, N-hole powder) during

the burn-up (f(z) = current surface area / initial surface area) e˙(pref) is the

linear burning rate at the reference gas pressure pref pref is the reference

gas pressure and a is the pressure exponent, which for many powders is

close to 1. To evaluate Eq. (7), z should be replaced by p/pmax using Eq. (4).

Figure 3 shows the time profile of the pressure in the manometric bomb for

a typical 7-hole powder. Initially the pressure is increasingly steep, since

burn-up takes place more quickly the higher the pressure and in addition

the burning surface of the powder becomes greater as the burn-up

progresses (progressive burn-up).

160


Fig. 1 Ballistic Mortar

2. Ballistic Mortar An instrument for comparative determinations of the performance of

different explosives. A mortar, provided with a borehole, into which a snugly

fitting solid steel projectile has been inserted, is suspended at the end of a

10 ft long pendulum rod. Ten grams of the explosive to be tested are

detonated in the combustion chamber. The projectile is driven out of the

mortar by the fumes, and the recoil of the mortar is a measure of the energy

of the projectile; the magnitude determined is the deflection of the

pendulum. This deflection, which is also known as weight strength, is

expressed as a percentage of the deflection produced by blasting gelatine,

arbitrarily taken as 100. Also, relative values referring to the deflection

produced by TNT are listed, especially for explosives of military interest. This

method, which is commonly employed in English-speaking countries, and

which is suited for the experimental determination of the work performed by

the explosive, has now been included in the list of standard tests

recommended by the European Commission for the Standardization of

Explosive Testing. An older comparison scale is “grade strength”, which

determines the particular explosive in standard “straight” dynamite

mixtures (the mixtures contain un-gelatinized nitroglycerine in different

proportions, sodium nitrate and wood or vegetable flour (W Dynamites)

161


which gives a pendulum deflection equal to that given by the test material.

The percentage of nitroglycerine contained in the comparative explosive is

reported as grade strength. The grade strength percentage is not a linear

indicator of the performance of the explosive; the performance of a 30%

dynamite is more than half of the performance of a 60% dynamite, because

the fueloxidizer mixtures as well as nitroglycerine also contribute to the gas

and heat-generating explosive reaction.

3. Lead Block Test The Trauzl lead block test is a comparative method for the determination of

the Strength of an explosive. Ten grams of the test sample, wrapped in

tinfoil, are introduced into the central borehole (125 mm deep, 25 mm in

diameter) of a massive soft lead cylinder, 200 mm in diameter and 200 mm

long. A copper blasting cap No. 8 with an electric primer is introduced into

the center of the explosive charge, and the remaining free space is filled with

quartz sand of standard grain size. After the explosion, the volume of the

resulting bulge is determined by filling it with water. A volume of 61 cm3 ,

which is the original volume of the cavity, is deducted from the result thus

obtained. In France the lead block performance value is given by the

coefficient d’utilisation pratique (c. u. p.): if mx is the mass of the tested

explosive, which gives exactly the same excavation as 15 g of picric acid, the

ratio 15 mx · 100 =% c. u. p. is the coefficient d’utilisation pratique. Also, 10

g of picric acid can be applied as a standard comparison explosive. For the

relationship with other testing procedures W Strength. Another modification

of the lead block test is recommended by BAM (Bundesanstalt für

Materialprüfung, Germany). The test sample is prepared as follows: a special

instrument wraps the sample in tinfoil and molds it into a cylinder of 11 ml

capacity (24.5 mm in diameter, 25 mm in height, with a coaxial cavity 7 mm

in diameter and 20 mm long for the blasting cap), whereby the resulting

density should be only slightly higher than the pour (bulk) density. Liquids

162


are filled into thinwalled cylindrical glass ampoules or, in special cases,

directly into the cavity of the lead block. The initiation is effected with an

electric copper blasting cap No. 8 containing 0.4 g of high pressed (380

kp/cm2 ) and 0.2 g of low pressed PETN as the secondary charge and 0.3 g

of lead azide as the initiating charge.

The empty space above the test sample is filled with dried, screened quartz

sand (grain size 0.5 mm), as in the original method. The volume of the

excavation is determined by filling it with water; after 61 ml have been

deducted from the result, the net bulge corresponding to the weight of the

compressed sample is obtained. In accordance with the international

convention, this magnitude is recalculated to a 10-g sample. The European

Commission for the Standardization of Testing of Explosive Materials*)

recalculated the results for a 10-ml test sample, using a calibration curve

established by Kurbalinga and Kondrikov, as modified by Ahrens; the

reported value refers to the mixture of PETN with potassium chloride which

gives the same result as the test sample under identical experimental

conditions. Since this regulation is still recent, the values given in the

following table, as well as the values given under the appropriate headings of

the individual explosive materials, are still based on the older method, in

which a 10-g sample is employed. Other conventional methods for the

determination of the explosive strength are the ballistic mortar test and the

sand test

163


Fig. 2 Lead Block Test

Fig. 3 Lead Block Expansion 4. Dautriche Method A method for the determination of the detonation rate. The test sample of

the explosive is accommodated in a column, which may or may not be

enclosed in an iron tube; the length of the detonating column to be

measured is marked out by means of two blasting caps, one at each end. A

loop made of a detonating cord with a known detonation rate is connected to

the caps and is passed over a lead sheet in its middle part. The cord is

successively ignited at both ends, and the meeting point of the two

detonation waves advancing towards each other makes a notch on the lead

sheet. The distance between this meeting point and the geometric center of

the cord is a measure of the reciprocal detonation rate to be determined: Dx

= D.m /2a where Dx is the detonation rate of the sample, D is the

detonation rate of the detonator cord, m is the length of the distance to be

164


measured, and a is the distance between the notch and the center of the

cord length. The method is easy to carry out and no special chronometer is

required. If, Dautriche method for the determination of detonation velocity:

A= explosive B,C=holes for the detonators

D=detonating fuse E=detonators

F=lead plate

Fig. 4 Dautriche Test 5. Bergmann-Junk Test A method, developed by Bergmann and Junk in 1904, for testing the

chemical stability of nitrocellulose; it was also subsequently employed for

165


testing single-base powders. The test tube, which contains the specimen

being tested, and which is equipped with a cup attachment, is heated at 132

°C = 270.4°F for two hours (nitrocellulose) or five hours (single base

powders). At the end of the heating period the sample is extracted with

water, and the test tube filled to the 50-ml mark with the water in the cup.

The solution is filtered, and the content of nitrous oxides is determined by

the Schulze-Tiemann method on an aliquot of the filtrate. The main

disadvantage of the method is that nitrous compounds are only incompletely

absorbed in water, especially since the atmospheric oxygen which has

remained behind in the tube is expelled during heating or is displaced by

the carbon dioxide evolved at the powder surface. Moreover, the results vary

with the volume of the specimen employed, since differing volumes of water

are required to fill the tube up to the mark in gelled and porous powders.

Siebert suggested the use of H2O2 rather than water as the absorption

medium in 1942. He also suggested that the employed apparatus should be

redesigned, to avoid gas losses which occur when the cup attachment is

taken off. In the new design, the cup is replaced by a large (over 50 ml)

attachment resembling a fermentation tube, which need not be taken off

during the extraction of the sample. In this way quantitative determination

of the liberated No, even in large amounts, becomes possible. Siebert also

suggested that the total acidity be determined by titration against N/100

NaOH, in the presence of Tashiro’s indicator. In this manner W Double Base

Propellants can also be tested as well; the test is carried out at 115 °C, the

duration of heating being 8 or 16 hours depending on the nitroglycerine

content of the sample (or of similar products, e.g. Diethyleneglycol

Dinitrate).

6. Sympathetic Detonation or Gap Test Also known as flash over; coefficient de self-excitation These terms denote

the initiation of an explosive charge without a priming device by the

166


detonation of another charge in the neighborhood. The maximum distance

between two cartridges in line is determined by flash-over tests, by which

the detonation is transmitted. The transmission mechanism is complex: by

shock wave, by hot reaction products, by flying metallic parts of the casing

(if the donor charge is enclosed) and even by the W Hollow Charge effect. In

the EU a method for determining the transmission of detonation is

standardized as EN 13631-11. Two cartridges are coaxially fixed to a wooden

rod with an air gap between them. Depending on the type of explosive the

test is done with or without confinement (e.g. steel tube). One cartridge

(donor) is initiated and it is noted whether the second cartridge (acceptor)

detonates. The complete detonation of the acceptor is verified by measuring

the velocity of detonation in it. The result of the test is the largest air gap in

cm for which the detonation of the acceptor was proved. For cartridged

blasing explosives which shall be used in the EU a minimum transmission

distance of 2 cm is required. In Germany, the ion-exchanged W Permitted

Explosives are also gap tested in a coal-cement pipe; these are cylinders

made of a bonded mixture of cement with coal dust in the ratios of 1:2 and

1:20 and provided with an axial bore. In the studies so far reported, donor

and receiver cartridges consisted of the same explosive. The transmission of

a standard donor cartridge through varying thicknesses of a stopping

medium can also be employed to determine the sensitivities of different

explosives. Recent practice in the United States is to insert cards (playing

cards, perspex sheets etc.) between the donor cartridge and the receiver

cartridge. Tests of this kind are named gap tests. In a more sophisticated

method, the gap medium (e.g. a plexiglas plate, see Fig. 13 below) stops

flying particles and directs heat transmission completely (shockpass heat-

filter). The shock wave is the only energy transmission to the acceptor

charge. For a 5 cm long and 5 cm diameter Tetryl donor charge with a

density of 1.51 g/cm3 , the pressure p in the plexiglas as a function of the

plexiglas length d according to M. Held*) is given by

167


p = 105 e 0.0358d

p in kbar, d in mm. The result of the gap test is recorded as the minimum pressure at which the

acceptor charge detonates. F. Trimborn (Explosivstoffe vol. 15, pp. 169–175

(1967) described a simple method in which water is used as the heat

blocking medium; the method can also be used to classify explosives which

are hard to detonate and are insensitive to blasting caps. The gap test

explosive train is directed from bottom to top. The donor charge (Hexogen

with 5% wax) is placed into a plexiglas tube and covered with water. The

acceptor charge to be tested is introduced into the water column from above.

The distance between the two charges can be easily varied. A detonating

cord, terminating on a lead plate, serves as evidence for detonation.

7. Heat Sensitivity Koenen Test Procedure The sample substance is introduced into a

cylindrical steel sleeve (25 mm dia.V24 mm dia.V75 mm) up to a height of

60 mm, and the capsule is closed with a nozzle plate with a central hole of a

given diameter. The diameter of the hole can vary between 1 and 20 mm;

when the plate is not employed, the effect is equivalent to that of a 24-mm

hole. The charged sleeve is placed inside a protective box and is

simultaneously heated by four burners; the time elapsed up to incipient

combustion and the duration of the combustion itself are measured with a

stop watch. The plate perforation diameter is varied, and the limiting

perforation diameter corresponding to an explosion caused by accumulation

of pressure inside the steel sleeve is determined. Explosion is understood to

mean fragmentation of the sleeve into three or more fragments or into a

greater number of smaller fragments. In this way, reproducible numerical

data are obtained which allow classification of different explosives according

to the explosion danger they represent. The parameter which is reported is

the largest diameter of the circular perforation in mm (limiting diameter) at

168


which at least one explosion occurs in the course of three successive trials.

8. Resistance wire continuous VOD measurement system Microtrap The continuous resistance wire method was developed in the early 1960s by

the United States Bureau of Mines (USBM). Operation is based on the basic

Ohm’s law, (E = RI), where E = Voltage, R = Resistance and I = Current.

When the current is held constant against a shortened (i.e. detonated) wire

of known resistance per unit length, a voltage drop can be measured

instantaneously at any point in time. The voltage drop is equivalent to the

length of resistance wire consumed in the detonation. Resistance wire

probes actually consist of two wires which must be physically shorted out by

the detonation through ionisation. Some resistance wire probes consist of

just two insulated wires twisted together and other probes consist of one

coated wire placed inside of a small metal tube which acts as the second

wire. Providing that the wires are adequately shorted during the detonation,

the resistance wire method does provide a truly continuous VOD along the

explosive column due to the high sampling rates ranging from 1.25 MHz to

over 10 MHz. If the wires are not adequately shorted in a continuous and

reliable fashion, erroneous results, excessive electronic noise and severe

drop outs are the norm. The details of the experimental hole are shown in

Fig. 2.

9. Impedance Detonation velocity of the explosive can be used to calculate the impedance

of an explosive which is defined as the product of the density and the

detonation velocity of the explosive. For good blasts, it is reported that the

impedance of the explosive should match with that of the rock. Berta [12]

mentions that the transfer of energy to the rock is a function of both the

characteristics of the explosive and the rock. According to him, the energy

169

http://www.sciencedirect.com/science/article/pii/S209526861500107X%23f0010

http://www.sciencedirect.com/science/article/pii/S209526861500107X%23b0060


transferred is influenced by impedance factor, If.

( )( )2re

2re

fII

II1I

+

−−=

where Ie = impedance of explosive = density of explosive detonation velocity,

(kg/(m2s)). Ir = impedance of rock = density of rock seismic wave velocity,

(kg/(m2s)). The above equation indicates that the energy transfer is

maximum when Ie = Ir.

A step-by-step procedure is suggested for selection of explosives for a mine

considering the advantages and disadvantages of the cartridges and bulk

systems, the rock properties, the environmental conditions such as water in

the blast holes, the performance evaluation of explosives for a given

condition [13], and the unit cost of production.

Fig. 2. Details of the experimental hole (M – probe cable, N – Nonel

tube, L – hole length, l1 – explosive charge length, a – explosive charge,

170

http://www.sciencedirect.com/science/article/pii/S209526861500107X%23b0065


b – primer, c – NONEL detonator).

10. Hot Storage Tests These tests are applied to accelerate the decomposition of an explosive

material, which is usually very slow at normal temperatures, able to

evaluate the stability and the expected service life of the material from the

identity and the amount of the decomposition prod- 177 Hot Storage Tests

ucts. Various procedures, applicable at different temperatures, may be

employed for this purpose. 1. Methods in which the escaping nitrous gases

can be recognized visually or by noting the color change of a strip of dyed

filter paper. The former methods include the qualitative tests at 132, 100,

75, and 65.5 °C (270, 212, 167, and 150 °F). These tests include the U.S.

supervision test, the methyl violet test, the Abel test, and the Vieille test. 2.

Methods involving quantitative determination of the gases evolved. Here we

distinguish between tests for the determination of acidic products (nitrous

gases) only, such as the Bergmann-Junk test and methods which determine

all the decomposition products, including manometric methods and weight

loss methods. 3. Methods which give information on the extent of

decomposition of the explosive material (and thus also on its stability),

based on the identity and the amount of the decomposition products of the

stabilizer formed during the storage. These include polarographic, thin-layer

chromatographic and spectrophotometric methods. 4. Methods providing

information on the stability of the explosive based on the heat of

decomposition evolved during storage (silvered vessel test). 5. Methods in

which stability can be estimated from the physical degradation of a

nitrocellulose gel (viscometric measurements). The tests actually employed

vary with the kind of explosive tested (explosives, single-base, double-base

or triple-base powders, or solid propellants) and the temporal and thermal

exposure to be expected (railway transportation or many years’ storage

under varying climatic conditions). In the case of propellants about to be

171


transported by train, only short-time testing is required. However, to obtain

an estimate of the expected service life is required, the so-called long-time

tests must be performed at 75 °C (167 °F) and below. The duration of such a

storage is up to 24 months, depending on the propellant type. Shorttime

tests – the Bergmann-Junk test, the Dutch test, the methyl violet test, the

Vieille test and, very rarely, the Abel test – are mostly employed in routine

control of propellants of known composition, i.e., propellants whose

expected service life may be assumed to be known. In selecting the test to be

applied, the composition of the propellant and the kind and amounts of the

resulting decomposition products must also be considered. Contrary to the

common propellants, which contain nitrates, the so-called composite

propellants cannot be tested in the conventional manner owing to the

relatively high chemical stability of the incorporated oxidants, e.g.,

ammonium perchlorate. In such cases the stability Hot Storage Tests 178

criterion of the propellants is the condition of the binder and its chemical

and physical change.

11. Fall Hammer Method The sensitiveness to impact of solid, liquid, or gelatinous explosives is tested

by the fall hammer method. Samples of the explosives are subjected to the

action of falling weights of different sizes. The parameter to be determined is

the height of fall at which a sufficient amount of impact energy is

transmitted to the sample for it to decompose or to explode. The US

standard procedures are: (a) Impact sensitivity test for solids: a sample

(approximately 0.02 g) of explosive is subjected to the action of a falling

weight, usually 2 kg. A 20-milligram sample of explosive is always used in

the Bureau of Mines (BM) apparatus when testing solid explosives. The

weight of the sample used in the Picatinny Arsenal (PA) apparatus is

indicated in each case. The impact test value is the minimum height at

which at least one of 10 trials results in explosion. In the BM apparatus, the

172


explosive is held between two flat, parallel hardened steel surfaces; in the PA

apparatus it is placed in the depression of a small steel die-cup, capped by a

thin brass cover, in the center of which a slotted-vented cylindrical steel

plug is placed, with the slotted side downwards. In the BM apparatus, the

impact impulse is transmitted to the sample by the upper flat surface; in the

PA, by the vented plug. The main differences between the two tests are that

the PA test involves greater confinement, distributes the translational

impulse over a smaller area (due to the inclined sides of the die cup cavity),

and involves a frictional component (against the inclined sides). The test

value obtained with the PA apparatus depends greatly on the sample

density. This value indicates the hazard to be expected on subjecting the

particular sample to an impact blow, but is of value in assessing a material’s

inherent sensitivity only if the apparent density (charge weight) is recorded

along with the impact test value. The samples are screened between 50 and

100 mesh, U.S. where single component explosives are involved, and

through 50 mesh for mixtures. (b) Impact sensitivity test for liquids: the PA

Impact Test for liquids is run in the same way as for solids. The die-cup is

filled, and the top of the liquid meniscus is adjusted to coincide with the

plane of the top rim of the die-cup. To date, this visual observation has been

found adequate to assure that the liquid does not wet the die-cup rim after

the brass cup has been set in place. Thus far, the reproducibility of data

obtained in this way indicates that variations in sample size obtained are

not significant. In the case of the BM apparatus, the procedure that was

described for solids is used with the following variations: 1. The weight of

explosives tested is 0.007 g. 2. A disc of desiccated filter paper (Whatman

No. 1) 9.5 mm P is laid on each drop, on the anvil, and then the plunger is

lowered onto the sample absorbed in the filter paper. The fallhammer

method was modified by the German Bundesanstalt für Material prüfung

(BAM), so as to obtain better reproducible data*). The sample is placed in a

confinement device, which consists of two coaxial cylinders placed one on

top of the other and guided by a ring. The cylinders have a diameter of 10–

173


0,003 –0,005 mm and a height of 10 mm, while the ring has an external

diameter of 16 mm, a height of 13 mm and a bore of 10+0,005 +0,01 mm;

all parts, cylinders and rings, must have the same hardness*). Cylinders and

rings are renewed for each falling test procedure. If the sample is a powder

or a paste, the upper cylinder is slightly pressed into the charged

confinement device as far as it will go without flattening the sample. If

liquids are tested, the distance between the cylinders is 2 mm. The charged

device is put on the anvil of the fallhammer apparatus, and the falling

weight, guided by two steel rods, is unlocked. For sensitive explosives such

as primary explosives, a small fallhammer is used for insensitive explosives

a large hammer. The small hammer involves the use of fall weight of up to

1000 g, while the fall weights utilized with the large hammer are 1, 5 and 10

kg. The fall heights are 10–50 cm for the 1-kg weight, 15–50 cm for the 5-kg

weight and 35–50 cm for the 10-kg weight. This method is the recommended

test method in the UN-recommendations for the transport of dangerous

goods and it is standardized as EN 13631-4 as a so-called Harmonized

European Standard.

174


Fig: BAM fall hammer BFH serves for measuring impact sensitivity of

solid high explosives

12. Testing galleries for explosives in coal mines All coal-mining countries have issued detailed regulations for the testing,

approval, and use of explosives which are safe in firedamp. The main

instrument for these tests is the testing gallery. A test gallery consists of a

steel cylinder which initates an underground roadway; the cross sectional

area is about 2 m2 (5ft P; one end is closed by a shield of about 30 cm (1ft)

P, against which the cannon is placed. The other end of the chamber which

has a volume of ca. 10 m3 (18ft length) is closed by means of a paper

screen. The remaining part of the tube length (10 m; 32ft) behind the paper

screen is left open to the atmosphere. (The gallery tube can be constructed

in closed form if the noise of the test shots can be diminished.) After

charging and positioning the cannon, the closed chamber is filled with a

methane-air mixture (containing, e.g., 9.5% CH4 to give the most dangerous

175


composition), and the charge is fired. Whether or not ignition of the gas

occurs is observed from a safe position. Amongst the known types of

mortars is the borehole cannon, as shown in Fig. 18. A steel cylinder about

1.5 m (5ft) long and about 35 cm (1–1/8ft) in diameter has in it a borehole of

55 mm (2–11/64 in.) diameter and 1.20 m (47 in.) length. The explosive to

be tested is placed in the borehole, unstemmed or stemmed by a clay plug,

and the detonator is introduced last in the hole (direct initiation). If the

detonator is inserted first, followed by the train of cartridges, initiation is

“inverse”. The required test conditions can be severe; ignition of the gas

mixture is more probable to occur using unstemmed charges and inverse

initiation than with stemmed charges and direct initiation. The different

mortars are designed to simulate different underground conditions. The

borehole cannon in the testing gallery illustrates the action of a single shot

in the roadway of gassy mines. The British break test and the slotted mortar

in Poland imitate the exposure of a charge and, consequently, the more

extended contact between the firing charge and the firedamp atmosphere

where breaks in the strata intersect a shothole: Two steel plates are held at

a given distance by means of a closing angle and a plug. The lower plate has

a groove for the cartridge train. The plate arrangement is covered with a

polythene sheet laid upon two steel side walls; the gas-tight room is filled

with the methane-air mixture after charging. The break test conditions are

varied; permitted explosives which meet the most stringent test conditions

belong to the British safety class P4. The slotted mortar allows similar test

procedures.

The slot does not extend over the whole length of the borehole and does not

begin at the mouth of the hole. A specially dangerous condition can arise

when several shots are fired in one round by means of electric delay

detonators. A preceding shot may then break the coal of another hole or

even cut off the whole burden of the charge in question so that it is partly or

completely exposed. This condition is simulated in the angle-mortar test. A

176


steel cylinder of 230 mm (9 in.) diameter and 2 m (~1/2ft) in length with a

right-angled groove is positioned in the gas chamber of a testing gallery

against an impact steel plate at given distances and different impact angles,

as shown in Fig. 21. Trains of several cartridges or of the full length of 2 m

are placed in the groove of the angle and fired into the methane-air mixture.

13. Sand Test A performance test of an explosive, used in the USA. A known amount of the

explosive is exploded in sand consisting of a single grain size (sieve) fraction;

the magnitude determined is the amount of sand which passes a finer-

meshed sieve following the fragmentation. The test descriptions follow: (a)

Sand test for solids. A 0.4-g sample of explosive, pressed at 3000 psi into a

No. 6 cap, is initiated by lead azide or mercury fulminate (or, if necessary, by

lead azide and tetryl) in a sand test bomb containing 200 g of “on 30 mesh”

Ottawa sand. The amount of azide of Tetryl that must be used to ensure that

the sample crushes the maximum net weight of sand, is designated as its

sensitivity to initiation, and the net weight of sand crushed, finer than 30

mesh, is termed the sand test value. The net weight of sand crushed is

obtained by subtracting from the total amount crushed by the initiator when

shot alone. 275 Sand Test (b) Sand test for liquids. The sand test for liquids

is made in accordance with the procedure given for solids except that the

following procedure for loading the test samples is substituted: Cut the

closed end of a No. 6 blasting cap and load one end of the cylinder with 0.20

g of lead azide and 0.25 g of tetryl, using a pressure of 3000 psi to

consolidate each charge. With a pin, prick the powder train at one end of a

piece of miner’s black powder fuse, 8 or 9 inches long. Crimp a loaded

cylinder to the pricked end, taking care that the end of the fuse is held

firmly against the charge in the cap. Crimp near the mouth of the cap to

avoid squeezing the charge. Transfer of 0.400 g of the test explosive to an

aluminum cap, taking precautions with liquid explosives to insert the

177


sample so that as little as possible adheres to the side walls of the cap; when

a solid material is being tested, use material fine enough to pass through a

No. 100 U.S. Standard Sieve. The caps used should have the following

dimensions: length 2.00 inches, internal diameter 0.248 inch, wall thickness

0.025 inch. Press solid explosives, after insertion into the aluminum cap, by

means of hand pressure to an apparent density of approximately 1.2 g per

cubic centimeter. This is done by exerting hand pressure on a wooden

plunger until the plunger has entered the cap to a depth of 3.93

centimeters. The dimensions of the interior of the cap are: height 5.00 cm,

area of cross section 0.312 square centimeters. Insert the cylinder

containing the fuse and explosive charge to Tetryl and lead azide into the

aluminum cap containing the test explosive for the determination of sand

crushed.

14. Upsetting Tests Upsetting tests are used to determinate the Brisance of the explosives. An

unconfined cartridge (enveloped in paper or in metal sheet) acts upon a

copper or lead crusher; the loss of height of the crusher is a measure for the

brisant performance of the tested explosive (Brisance). The cartridge shock

acts by means of a guided pestle onto a copper crusher of 7 mm P and 10.5

mm height. The simplified test according to Hess is shown in Fig. 26 (see

opposite page): A lead cylinder, 60 mm (2.36 in.) high, 40 mm (1.57 in.) ∅,

protected by two, 6 mm-thick steel disks, is upset by a 100-g (3.53 oz.)

cartridge of the same diameter, 40 mm. The cylinder is pressed down into a

mushroom shape; in the case of sensitized special gelatins for seismic use,

the cylinder can be destroyed completely.

15. Vacuum Test This test, which was developed in the USA and has been adopted by several

countries, and is a modification of the Taliani Test, in which the gaseous

178


products of the reaction are determined volumetrically rather than by

manometry. The test, which is carried out at 100 °C (212 °F) for single base

propellants and at 90 °C (194 °F) for multibase propellants, is terminated

after 40 hours, unlike the Taliani Test, which is interrupted after a given

pressure or a given volume has been attained. The vacuum test is used for

compatibility testing and applied as a so-called reactivity test. The

compatibility between the explosive and the contact material (adhesive,

varnish, etc.) is tested by determining the gases liberated by the explosive

alone, by the contact material alone, and by the two together. The measure

of compatibility (reactivity) is the difference between the sum of the gas

volume liberated by each component separately and the gas volume

obtained after storing the explosive and the contact material together. If this

difference is below 3 ml, the compatibility is considered to be “stable”,

between 3 and 5 ml, the compatibility is considered „uncertain“; above 5 ml,

the two materials are incompatible.

16. Water Resistance Test

In the USA the following method is employed for testing the water resistance

of commercial explosives: Sixteen regularly spaced holes (about 6 mm P) are

cut in the cartridge paper (30 mm in diameter, 200 mm long) of the

explosive to be tested, and the flaps on the front faces are sealed with tallow.

The cartridges thus prepared are placed in a flat, porcelain coated dish

covered with a thin layer of sand, and water at 17–25 °C (63–77 °F) is

poured over the sand layer up to a height of about 25 mm. The cartridges

are left under water for a certain period of time, are then taken out, the seal

is cut off at one end, and the cartridge is tested for detonation and

transmission with the aid of a No. 6 blasting cap. The criterion for the water

resistance of the explosive is the time of exposure to water, after which it

still retains its capacity to detonate the cartridge in three trials, without

leaving any non-detonated residual explosive behind. There is no generally

179


accepted quality classification. Nevertheless, water resistance of an explosive

is considered to be satisfactory, acceptable, or poor if the cartridge can still

be detonated after 24, 8, or 2 hours respectively. In Germany, the following

method for testing the water resistance of powder-form permissibles has

been laid down at the Test Station at Dortmund-Derne. A train of four

cartridges is fixed in a file on a wooden board; the first of the four cartridges

is equipped with a detonator No. 8. Five longitudinal, 2 cm long notches,

uniformly distributed over the circumference, are cut into each cartridge.

The train is immersed for 5 hours in water, in a horizontal position, 20 cm

under the water surface, after which they are detonated. The train must

detonate in its entirety. This method, including some additions regarding the

preparation of the test sample is standardized as EN 13631-5 as a so-called

Harmonized European Standard. The water resistance of partly water –

soluble, powder-form explosives (e.g. ammonium nitrate explosives or

blasting agents) can be improved by the addition of hydrophobic or gelling

agents. If e.g. W Guar Gum is added, the water entering immediately forms a

gel which blocks the penetration of more water.

180

CHAPTER – 6

MODERN METHODS OF EXPLOSIVE FORENSICS

1. Introduction With the apparent increased use of explosives in recent years , the ability to

quickly, sensitively and reliably identify and determine a wide range of

chemical components is essential in forensic. Some of the recent high-

profile terrorist attacks using improvised explosive devices (IEDs) include

those are used regularly in India, Pakistan, Afghanistan, Iraq and Syria and

have been used in London (1999, 2005), Norway (2011) and Boston (2013)

amongst several others. The devices used in these events highlighted the

array of possible chemicals that may exist and, in some cases, the relative

ease in obtaining them. Therefore, the significant challenges placed on

analytical scientists to prepare for and hopefully prevent such events,

despite their complexity, is unfortunately an ongoing problem.

Explosives can be sub-divided by compound sensitivity into primary,

secondary and tertiary explosives as shown in Fig. 1. Primary explosives are

highly sensitive, unstable materials and can ignite by shock, friction, heat or

impact. Tri Acetone Tri Peroxide (TATP) is an example of a primary explosive

and one which can be analytically challenging to determine due to this

instability. Secondary organic explosives typically contain a nitro group such

as 2,4,6-trinitrotoluene (TNT) or nitroglycerin (NG). These are comparatively

more stable and are frequently used with a small amount of primary

explosive to initiate a detonation. They are regularly mixed with gels,

sensitisers or stabilisers depending on the application requirement. Primary

and secondary explosives are generally termed “high-order explosives” as

they detonate upon discharge. The majority of inorganic compounds,

containing salts of chlorate, perchlorate or nitrate ions for example, are

181


Chapter 6 Modern Methods of Explosive Forensics

considered “low-order explosives”. These materials generally deflagrate and

are often used in pyrotechnics and propellants. However, some inorganic

compounds, such as ammonium nitrate, are still classed as high-order

explosive materials and are known as tertiary explosives. These are very

insensitive to shock and require an explosive booster of secondary explosive.

Any of the above explosive types can be used in IEDs, and the identification

of the components present can be used to determine the source and possibly

provide a link to a suspect. The term “pre-blast” describes a point in time

before an explosion has occurred. Conversely, “post-blast” pertains to a

time-point after discharge of an explosive. Pre-blast detection of explosives

can either involve bulk (mg to kg) analysis of the charge itself, or trace

analysis (fg to mg) of the explosive on skin, clothing, solid surfaces and

soil for example. Post-blast analysis typically involves trace analysis of non-

combusted explosive material or its combusted products either in the

surrounding vicinity of the explosion, on any fragments or material

belonging to the IED or on articles affected by the explosion. Whilst

potentially linking suspects with pre-blast materials analysis might arguably

be considered more straightforward (e.g. through the their detection in

fingermarks, it is significantly more challenging to do so with post-blast

residues containing a mixture of the parent components and their

combustion products, as well as the high risk of environmental

contamination.

182


Fig. 1. Schematic of energetic materials classifications. High-order

explosives detonate and are further divided by sensitivity, whereas low

order explosives deflagrate. The dashed line highlights the groups that

primarily contain inorganic components included in this Chapter. NC,

Nitrocellulose; NG, nitroglycerin; TNT, trinitrotoluene, TATP,

triacetone triperoxide; HMTD, hexamethylene triperoxide diamine.

The majority of analytical methods employing hyphenated separation and

mass spectrometric techniques for energetic materials analysis have

focussed on the organic high-order explosives, and as a result there are

many reviews available on this topic. The two main separation techniques

used for confirmatory detection of such compounds are gas

chromatography (GC) and high performance liquid chromatography

(HPLC) and coupled to mass spectrometry (MS). The choice of separation

mode is often dependent on analyte vapour pressure. For example, the

vapour pressure of octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine (HMX;

speculatively named, “High-velocity Military explosive”, amongst other

pseudonyms) is low and, along with its thermal lability, it is more amenable

to LC-MS rather than GC-MS[24]. However, and despite the large body of

literature available on trace high-order explosives analysis, there still exists

a gap in confirmatory analytical separations-based technologies for low-

order inorganic or organic ion explosives. Along with some other techniques,

183

http://www.sciencedirect.com/science/article/pii/S0003267013013901%23bib0120


many low-order and related species have been routinely and widely

determined by IC coupled to relatively non-specific detectors such as

suppressed conductivity, electrochemical or ultraviolet detection for over 30

years in many different fields of application . Most importantly, the ability to

speciate gives IC an obvious advantage over several other analytical

techniques. Speciation can be vital when it comes to the identification and

characterisation of low-order explosive residue especially regarding

differentiation of oxychloride species from the free chloride ion;

distinguishing transition metal ion valencies (such as Fe(II)/Fe(III)); and

separating nitrogen-containing species such as nitrate, nitrite, thiocyanate

and cyanate. Relatively recent research either directly or indirectly related to

IC-MS has demonstrated a number of important findings, which could offer

some insight and potential value to forensic practitioners planning or

currently operating this technique.

Therefore, the aim and scope of this review is to (a) examine the contexts

and subsequent emergence of IC-MS technology for this application; (b) to

detail the recent technological advances and challenges associated with

practical trace IC-MS from an environmental and forensic science

perspective; (c) to critically compare its usefulness in comparison to other

orthogonal confirmatory methods; and (d) to identify future avenues for

application of this technology in forensic and environmental science. To the

authors’ knowledge, this is the first review focussing specifically on IC-MS

technologies for the analysis of explosive residues.

2. The Impact of Improvised Explosive Devices and the need for IC-MS

Component parts of low-order explosives-based IEDs are readily available in

the form of fertilisers, propellants and pyrotechnics. The type and amount of

explosive present often differs considerably. Complementary separation

techniques to IC such as capillary zone electrophoresis (CZE) have been

traditionally used to confirm identity or source from a residue deposited on

184


a suitable surface. Smoke emitted during discharge may indicate the type of

explosive used and can be sampled for analysis. Oxidisers such as

potassiated salts of nitrate, chlorate and perchlorate are combined with a

fuel (such as a sugar), an optional coolant (e.g. NaHCO3), and a dye to

generate a coloured smoke seen often during pyrotechnic displays. The

presence or absence of smoke, as well as the time for discharge

(deflagrations are much slower than detonations), can give invaluable

intelligence regarding the identity of an explosive material before analysis

and therefore may direct which technique to apply. The deadly nature of

IEDs is in part due to their ease of concealment and subsequent blast

impact over short distances. The US Technology Support Working Group

(TSWG) specifies the safe evacuation distances for a range of IED sizes.

For example, an unimpeded outdoor radius of 900 ft (274 m) is

recommended for an IED contained within a letter or small package

with an explosive capacity of 1 lb. Scaling up to a full-size car/minivan

with an explosive capacity of 1000 lb for example, this translates to a

safe radius distance of 2400 ft (732 m) in an outdoor environment. In

an indoor environment, these safe distances may be much reduced,

presumably due to obstruction of high-velocity shrapnel. Therefore from

an analytical point of view, this represents a significant challenge in terms of

deciding where and how to responsively sample trace residue evidence over

a potentially unknown distance, coupled with minimising the effects of

matrix/environmental interference on the analytical determination. Putting

this into context, the Boston Marathon attack in 2013 represents just one

recent example of how IEDs were used to inflict mass casualties among

civilians. Three people were killed and over 264 injured when explosive

materials along with nails and other metal parts were sealed within two

pressure-cookers and discharged ∼190 m apart in a crowded Boylston street.

Even though the immediate blast only occurred over a radius of ∼10 m,

injuries from debris were reported at much further distances. A wider list of

details relating to recent IED attacks is shown in Table 1 for comparison.

185



Interestingly, a recent survey by the TSWG revealed that the 468 bomb

squads in the US were called to investigate ∼32,000 potentially IED-related

incidents nationally per year from 2005 to 2010. IEDs also pose a serious

threat in military operations. It has been reported that in Afghanistan, the

number of IED-related coalition fatalities rose to its height at 368 in 2010

out of a confirmed 630 total hostile deaths. Across 2001–2013, an average of

∼51% of all hostile military coalition deaths were due to IEDs in Afghanistan

alone. This translates to 1370 out of 2690 total hostile conflict deaths. In

Iraq by comparison, 1588 fatalities were recorded as being due to IEDs from

2003 to 2007.

Table 1. Examples of high profile terrorist incidences using IEDs.

Year Place Numbers killed/ injured

Explosives used Additional info

1969-Present

England and Northern Ireland

∼1800/ ∼48,000

RDX, PETN, ANFO –

1988

Pam Am Flight 103, Lockerbie, Scotland

270/0 Trace PETN/RDX found

–

1993 World Trade Centre, USA 6/1042 Urea nitrate

Aluminium, magnesium and ferric oxide to enhance reaction temperature. Ammonium nitrate, lead azide and smokeless powder used as boosters

1995 Oklahoma, USA 168/∼680 ANFO –

1996 Olympic Park, Atlanta, USA

2/111 Smokeless powder Nails

186


1999 London, UK 3/139 Flash powder from fireworks

Nails

2002 Bali, Indonesia 202/209

Potassium chlorate, aluminium, sulphur

TNT booster. PETN detonating cord. RDX detonators. Thermobaric

2004 Madrid, Spain 191/2050

Unconfirmed—variation of dynamite/nitroglycol

–

2005 London, UK 52/∼700 TATP, HMTD –

2006 Transatlantic flight 0/0 TATP –

2006 Mumbai, India 209/714 RDX, ANFO Thermobaric—

pressure cooker 2011 Norway 8/209 ANFO Fertiliser

2013 Boston, USA 3/282 Unconfirmed—sugar chlorate

Thermobaric—pressure cooker with nails

ANFO: Ammonium nitrate and fuel oil; RDX: research department explosive;

HMX: high velocity military explosive; TATP: triacetone triperoxide; HMTD:

hexamethylene triperoxide diamine; TNT: trinitrotoluene.

Based on these figures, and the potential capability offered by IC-MS over

hyphenation to non-specific detection modes for low molecular weight low-

order explosives and their combustion residues, it seems obvious that more

research efforts are required to tailor this approach towards such forensic

applications. Some issues surrounding sample collection need to be

considered early in this process. For example, recent work has shown that

the practices for the analysis of primary and secondary explosives cannot be

assumed valid when undertaking an analysis of low-order inorganic

explosive species deposited on a surface. For example the extraction of ions

in organic solutions (generally applied for high-order organic explosives) is

not optimal and sample reconstitution in aqueous media is required for IC

187


compatibility to reduce self-elution, baseline distortion and momentary

polymer resin dis-equilibration. Furthermore, like high-order explosives,

some specific low molecular weight ionic species were partially/non-

recoverable (such as nitrite and cyanate), displayed unacceptably high

variation using traditional cotton or rayon swabbing devices, suffered high

levels of matrix, method or co-analyte interference or remained unidentified

using non-specific suppressed conductivity detection . Careful consideration

of these issues have recently facilitated the pg detection and quantification a

range of low-order explosive species using IC with suppressed conductivity

detection in latent fingermarks and also in chemically and physically

enhanced fingermarks. However multi-analyte selectivity still remains the

largest challenge and one where IC-MS may offer an obvious solution.

Therefore, from a forensic standpoint, there exists a current need to extend

the repertoire of IC-MS to the confirmatory detection and identification of

such species in complex trace evidence matrices.

Aside from those already discussed, several other low-order ionic explosives-

related species are frequently encountered during pre-/post-blast analysis.

These include alkylated ammonium ions such as methyl- or

ethylammonium; indicative metal ions used to increase the reaction

temperature such as magnesium or aluminium; low molecular weight

organic acids including benzoate and oxalate; and cyano-containing species

such as thiocyanate, cyanate (fulminate), azide . Thiocyanate is known to

occur in certain foods and has been detected in the urine of smokers at

elevated mg L−1 levels using IC . This ion has also been detected in GSR and

can be used along with other ions to distinguish between ammunition

types . Furthermore, it should be noted that post-blast residues of nitro-

group containing organic explosives compounds such as the nitrosamines,

nitro-esters, and the nitrated sugar polyols often contain elevated

concentrations of the nitrite or nitrate ion after discharge. Previous studies

have used the ratio between the two in an attempt to identify the source

(Wheals and Ellison studied the generation of these ions from the hydrolysis

188


of nitrocellulose and nitroglycerine to determine the relative concentration

ratio for forensic purposes). Many on-site spot tests, such as those based on

the Greiss diazotisation reaction can be used to preliminarily assess whether

a nitro-group containing organic species is to be expected through detection

of nitrite. Other ions such as thiocyanate arise from sulphur-containing

explosive ingredients. Confirmatory forensic analysis is usually performed

using a combination of ion separation or detection techniques such as IC ,

CZE , various modes of Raman spectroscopy and, occasionally, MS. Unless

present in significantly high amounts (mid- mg L−1 levels and higher), it has

been shown that the presence of the nitrite or nitrate ion alone does not

offer significant intelligence or forensic value to an investigation due to the

potential for environmental or method-related background interference . A

list of species reported in the literature relating to low-order explosives is

given in Table 2.

Table 2. Inorganic and organic ions present in low-order energetic

materials and its post-discharge residue and examples of sources of

reported occurrence.

Ion Source

Al3+ Primer, fuel, commercial blasting agents

As3+ IEDs

Ba2+ Primer, propellant, oxidiser, IEDs

Ca2+ IEDs, primer, propellant, smokeless powder, fuel

CH3CH2NH3+ IEDs

CH3 NH3+ IEDs

Cu2+ Projectiles (e.g. bullets), ammunition casing

Fe2+/Fe3+ Projectiles

K+ Primer, IEDs, black powder, smokeless powder

Mg2+ IEDs, primer, smokeless powder

189



Na+ Primer, propellant, IEDs, black powder, smokeless

powder

NH4+ ANFO, smokeless powder, IEDs

Ni2+ Ammunition casing

Pb2+ Primer, projectiles, explosive

Sb3+ Primer, ammunition casing, fuel

Zn2+ IEDs, primer, ammunition casing

Cl− Degradation product of chlorite, chlorate, perchlorate

and other Cl-containing species

ClO2− Reduction product of chlorate/perchlorate-based

explosives

ClO3− Chlorate-type IEDs, GSR

ClO4− Perchlorate-type IEDs, GSR, oxidiser

CNO− GSR

CO32− Black powder substitutes, GSR

NO2− Degradation product, GSR

NO3− ANFO, black powder, oxidiser

N3− IEDs

SCN− GSR

SO42− Black powder

S2O32− Combustion product of sulphur in IEDs

PO43− IEDs

Acetate GSR, PETN-based explosives

Ascorbate Black powder substitutes

Benzoate GSR

Formate GSR, PETN-based explosives

Glycerate Degradation product

Monohydrated Degradation product

190


diketogulonic

acid

Oxalate Degradation product, GSR, PETN-based explosives

Phthalate Propellant powder, plasticiser

Threonate Degradation product

ANFO: Ammonium nitrate and fuel oil; PETN: pentaerythritol tetranitrate; GSR:

gunshot residue; IED: improvised explosive device.

3. Isotopic and elemental profiling of ammonium nitrate

3.1 Background

Ammonium nitrate (AN) has applications in explosives in industry, such as

mining and civil construction. Its chemical stability and low sensitivity to

shock and friction make it relatively safe to produce and to handle in large

quantities. Combined with its low costs these properties make AN a suitable

ingredient for explosive devices. Explosive-grade ammonium nitrate

generally exists in the form of slurries, emulsions, or porous prills mixed

with a fuel (ammonium nitrate fuel oil, ANFO). Obtaining explosive-grade AN

for criminal abuse is difficult due to extensive regulations, but AN in the

form of fertilizer prills or granules is widely available. The European

Commission prohibits the distribution of ammonium nitrate-containing

fertilizers with a nitrogen percentage higher than 16% to private parties. In

addition, detonation tests should be performed on fertilizers with a nitrogen

content of more than 28%. Fertilizer prills have a higher density and a lower

porosity than explosive-grade prills, reducing the sensitivity to detonate.

However, fertilizer-grade ammonium nitrate can still detonate or be made to

detonate. Besides industrial applications of AN-based explosives, terrorist

attacks involving the use of AN in improvised explosive devices (IEDs) have

been a concern for many years. In such cases of criminal use of AN in an

IED, chemical profiling can be an important tool to investigate the origin of

191


the material used.

The possibility of tracing ammonium nitrate back to its production location

is important. Product classification can yield important information

especially for intelligence purposes and in the absence of leads and

suspects. Chemical profiling of AN is especially challenging because of its

inorganic nature and the bulk quantities in which it is produced. In this

work the potential of profiling based on trace-elemental composition and

isotope ratios has been investigated.

Ammonium nitrate is manufactured from ammonia and nitric acid.

Ammonia production occurs through the reaction of hydrogen with

atmospheric nitrogen according to the Haber process. Thereafter, nitric acid

is formed by the oxidation of ammonia (the Ostwald process). Both the

nitrogen and oxygen in ammonium nitrate are retrieved from the

atmosphere. Although there is not much variation in the isotopic

composition of atmospheric nitrogen and oxygen around the world,

fractionation can occur during AN synthesis resulting in variation in both

δ15N and δ18O values for different samples. Fertilizer prills or granules

consist, besides ammonium nitrate, of various additives, such as gypsum,

dolomite and magnesium nitrate. In addition, the prills are coated with a

water-repellent material to prevent the prills from absorbing water. These

additives also contribute to the isotopic signature of the prills/granules.

Moreover, the additives used during the manufacturing of AN-based

fertilizers are rich in elements and therefore determine the elemental

composition of the fertilizer prills. Two techniques that are, therefore, of

special interest for profiling AN are isotope-ratio mass spectrometry

(IRMS) and inductively-coupled plasma–mass spectrometry (ICP–MS). As

part of the ICP–MS study also the application of laser-ablation ICP–MS

(LA–ICP–MS) was investigated for element profiling of individual

granules or prills. LA–ICP–MS could provide profiling options in forensic

post-explosion investigations where only a limited number of granules

192


or prills are found.

3.2. Applications of IRMS and ICP–MS Most of the analytical techniques conventionally used in forensic science are

not able to determine whether samples with the same chemical identity have

a common source. IRMS can in principle be used to study the source of

samples based on their isotopic composition. In the field of explosives, IRMS

was previously used to discriminate between different sources of Semtex,

triacetone triperoxide (TATP), pentaerythritol tetranitrate (PETN), black

powder, flash powder, nitromethane, plastic explosive no. 4 (PE4),

trinitrotoluene (TNT), urea nitrate, and ammonium nitrate. With one

exception these were all preliminary studies with either a small sample set

or limited information on the origin of the materials. In addition to the

ability to examine the source of explosive samples, IRMS has also been used

to study relationships between precursors and the explosive product for

urea nitrate, cyclotrimethylenetrinitramine (RDX)and hexamethylene

triperoxide diamine (HMTD).

ICP–MS is capable of detecting and quantifying multiple elements and thus

of producing trace-elemental signatures. Samples can be introduced in the

plasma either by nebulization (liquids or solutions) or by laser ablation (LA–

ICP–MS). When sufficient evidence material is available, spraying from

solution will provide the most accurate quantitative analysis as it allows for

homogenization of the material. However, the use of laser ablation allows for

trace-element profiling of small amounts of material and for studying the

spatial distribution of the trace elements (2D and 3D profiling). Elemental

profiles obtained using ICP–MS have been used to discriminate between

different sources of glass, paint, tapes, white cotton fibers, beer,

bullets, methylamphetamine, human bones and teeth, and in the

detection of art forgeries . The combination of IRMS and ICP–MS may

result in complementary information on the origin of a material and has

193


shown to be powerful in discriminating between different sources.

3.3. Challenges for IRMS of nitrates Stable-isotope analysis of nitrogen-containing samples and nitrates in

particular poses some challenges. There is an influence of N2 gas formed

during high-temperature reduction of nitrogen-rich compounds on the

analysis for 2H isotope abundance. Partial overlap of the N2 and H2 peaks

after GC separation resulted in inaccurate and imprecise δ2H values, either

due to dilution of the H2 peak or to ionization competition between H2 and

N2 gas. Mimicking nitrogen-rich samples by adding silver nitrate to benzoic

acid samples resulted in a shift of roughly 5‰ toward more negative δ2H

values for an N:H ratio of one compared to ‘nitrogen-free’ benzoic acid.

Analysis of18O/16O isotope ratios is performed by conversion of elemental

oxygen in the sample to carbon monoxide (CO) using high-temperature

conversion (HTC) in a reactor packed with glassy carbon. The IRMS is then

used to measure the amounts of m/z 28 (12C16O) and m/z 30 (12C18O) to

determine 18O/16O ratios in a sample. Conversion of oxygen in inorganic

samples to CO has proven to be challenging. Powdered or nickelized carbon

has been added either to the reactor or directly to the capsules containing

the samples. Nickel acts as a catalyst, enhancing the conversion to CO and

allowing the use of lower reaction temperatures. The use of nickelized

carbon resulted in more precise 18O/16O determinations.

The hygroscopic nature of AN is another point of concern. For prills or

granules, water absorption is minimized by coating with an inert,

hydrophobic material. However, sample preparation for IRMS requires

grinding of the prills or granules, causing interaction with atmospheric

water. Exchange of oxygen was not observed for KNO3 and NaNO3 reference

materials (USGS34, IAEA-NO-3 and USGS35). In addition, δ18O values of

these dried nitrate standards were not affected by repeated sorption and

194


desorption of H2O. This suggests that oxygen exchange should not be an

issue for ammonium nitrate, if a suitable drying and storage procedure is

used to prevent absorption of water.

During high-temperature conversion, nitrogen in a sample is converted to

N2, which is isobaric to CO (m/z 28). Although N2 and CO can be separated

using a 5-Å molecular-sieve GC column, the presence of small traces of

oxygen may result in formation of NO+ at the hot filament of the ion source.

Whereas the signals of m/z 28 (14N2) and m/z 29 (14N15N) quickly diminish

after elution of the N2 peak, NO disappears slowly from the ion source,

resulting in a long tail in the m/z -30 signal that extends into the 12C18O

peak, rendering accurate 18O/16O measurements impossible. The extent of

N2 interference is related to the amount of N2 formed during high-

temperature conversion, which in turn depends on the type of compound

analyzed. Hunsinger et al. demonstrated that nitrogen from nitro or nitrate

groups was efficiently converted to N2, whereas the N2 yield for

amine/amide-containing compounds was around 65% and for caffeine the

N2 yield was even lower. The high nitrogen content of ammonium nitrate

necessitates effective removal of the N2 interferences, as suggested by

Benson et al.

Several approaches to eliminate or reduce the isobaric interference of N2

with CO measurements have been reported. Separation of N2 and CO can be

improved by replacing the conventional 0.6-m GC column with a longer one,

allowing more time for the m/z 30 signal to return to the baseline. However,

merely extending the length of the column led to a reduction of them/z30

background offset, but it did not eliminate the N2interference, especially not

for samples having a high nitrogen content. Another option to reduce the

interference of N2 with 18O/16O determinations is to reduce the amount of

N2 in the ion source or to prevent N2 from entering the ion source altogether.

This can be achieved by installing a switching valve between the GC column

and the IRMS to divert the gas stream during elution of the N2 peak or by

195


diluting the N2 peak with helium using the open–split interface between the

GC column and the IRMS observed that combining the use of a longer

column with diversion of the N2 peak yielded the most accurate and precise

δ18O values. Diversion of the N2 peak requires the installation of a switching

valve between the elemental analyzer and the IRMS. Therefore, our approach

to minimize N2 interference was to combine the use of a longer column with

He-dilution through the existing open–split interface. Other possible

measures to reduce the interference of N2 with 18O/16O measurements

include application of a CO-reference-gas pulse between the N2 and CO

peaks and improving background correction by subtracting the tail of the N2

peak from the area of the CO peak.

4. Characterization of ANFO by high accuracy ESI(±)–FTMS with

forensic identification on real samples by EASI(−)–MS

Ammonium nitrate fuel oil (ANFO) is one of the most widely used explosives

in the world. ANFO is commonly composed of a mixture of approximately

94% ammonium nitrate (AN) particles soaked in 6% fuel oil (FO) .

Unfortunately, the use of ANFO has not been limited to mining and

construction but has also been used in terrorist attacks.

In Brazil, criminals have found a new and threatening application for ANFO:

the explosion of automated teller machines (ATMs). According to the sixth

edition of the national survey on attacks on banks in Brazil, enforced bank

break-ins have increased from 911 cases in 2011 to 2085 cases in 2013—a

total increase of 117% . Interestingly, theft of ANFO from quarries has also

been increasing in Brazil, showing a direct correlation with its use in ATM

explosions. The scene of the action is shown in the Fig 2. Figs 2 A,B and C

show the simulation results.

196


Fig. 2 (A) ATM explosion in a simulation experiment showing the

banknote (B) before and (C) after the explosion. (D) Real case explosive

sampling from a failed ATM explosion.

The second sample was obtained from a real crime scene where an explosion

was attempted, i.e., criminals tried to blow up an ATM by inserting an

explosive in the money outlet but it failed to explode (Fig. 5D). A stock

solution of this explosive was prepared using the same procedure described

for standard ANFO, and 20 μL of this solution was dispersed on the surface

of a support paper commonly used in EASI–MS experiments.

The banknote was analyzed via EASI (±)–MS to evaluate the versatility of this

direct ambient technique to rapidly characterize residues of the ANFO

explosive directly on the banknote surface without any sample manipulation

or extraction procedures. Before and after the explosion, chemical profiles

from the banknote were obtained via EASI (+)–MS. The ions detected in the

197



banknote before the explosion (Fig. 3A) were typical of ion markers for

Brazilian banknotes. After the explosion, however, the EASI (+)–MS obtained

directly from an area of the banknote surface, with apparently no ink

deposited (Fig. 3B), showed a major ion of m/z 443.3, which was

assigned to rhodamine B, the major component of the dye applied for

antitheft purposes. The ions of m/z 106, 311 and 391 are likely a result of

residual fuel oil contamination from ANFO after the explosion. For analyses

in a high density ink area (Fig. 3C), the rhodamine ion of m/z 433 was

predominant. The EASI(+)–MS for the explosive obtained from a real crime

scene (Fig. 6D) presented the typical profile of fuel oil expected for

unexploded ANFO (see Fig. 1A). When Fig. 6B and C are compared with Fig.

3D, it is clear that an ANFO explosion leaves few, if any, FO markers.

The EASI(−)–MS profile of a banknote acquired before and after explosion are

also clearly distinct and characteristic. Before the explosion (Fig. 4A), only

hand manipulation contaminants, i.e., palmitic acid (m/z 255) and oleic acid

(m/z 281), were detected. After the explosion, the area of the banknote

surface with low ink density (Fig. 4B) provided nearly the same profile as the

high ink density region (Fig. 4C) except for the ions of m/z350–450

attributed to the ink components. These results demonstrate that the

magnesium-nitrate cluster ionization is preferable to that of the ink

components and that detection of m/z 210 would be possible even in the

presence of this interference. For the real crime scene sample (Fig. 4D),

aside from the predominance of m/z 210, the spectra also displayed nitrate

ions of considerable abundances. This result may be explained by either the

higher nitrate concentration before the explosion or in source fragmentation,

which was also observed in Fig. 4 C and D. For this specific case, in-source

fragmentation is useful for the confirmation of nitrate ions and a more

confident composition determination of the marker ion of m/z 210, leading

to more reliable detection. In-source fragmentation was not observed in the

QIT–MS data.

198















Fig. 3. EASI(+)–MS from the surface of (A) an undisturbed Brazilian

banknote; (B) and (C) the banknote after ANFO explosion; and (D) an

199


extract of ANFO remaining from a real crime scene.

Fig. 4 EASI(−)–MS from the surface of (A) an undisturbed Brazilian

200


banknote; (B) and (C) the banknote after ANFO explosion; and (D) an

extract of ANFO remaining from a real crime scene.

In such forensic investigations, there is therefore a need for a precise

methodology for ANFO characterization and its rapid detection in different

explosion residues. Forensic investigations of ammonium nitrate-based

explosives have been performed via several strategies , such as colorimetry ,

ion chromatography (IC) , capillary electrophoresis (CE) , voltammetry and

real-time neutron radiography , which have been the main techniques

applied to inorganic explosives analysis. Field analyses of ANFO have also

used photodissociation detection. According to the procedures for the

analysis of explosives evidence from the North Carolina Department of

Justice , three protocols can be applied for the analysis of blasting agents,

slurries and emulsions. The first employs analysis of the inorganic oxidizing

agent either by scanning electron microscopy with energy dispersive X-ray,

X-ray diffraction or CE. The second protocol employs analysis of the

sensitizer, if present, generally via thin layer chromatography (TLC) or gas

chromatography coupled to mass spectrometry (GC–MS). The third protocol

calls for the characterization of the fuel oil via techniques such as Fourier

transform infrared spectroscopy or GC–MS.

Ion mobility spectroscopy (IMS) has also been reported for ANFO

detection with a limit of detection as low as 10 pg. Thermospray mass

spectrometry (TS–MS), atmospheric pressure chemical ionization mass

spectrometry (APCI–MS) and electrospray ionization mass spectrometry

(ESI–MS) of nitrates have been shown to detect ion clusters that span a

broad m/z region depending on the heated capillary temperature and the

polarity (positive or negative) mode of the analysis. For ESI–MS, Zhao and

Yinon demonstrated that nitrate salts form ion clusters with well-defined

proportions, and suggested their molecular formulas based on the measured

m/z values, ion fragmentation patterns and 15N isotopic labelling.

201


Recently, a series of desorption/ionization techniques capable of performing

direct ambient MS analyses has been introduced. These revolutionary

techniques have proven to be suitable for forensic investigations because

they provide fast chemical profiles from the undisturbed sample with

unprecedented simplicity and speed. Analyte desorption and ionization

directly from sample surfaces is performed in an open atmosphere, followed

by MS characterization, with little to no sample preparation. Although

ambient MS was pioneered by desorption electrospray ionization (DESI) and

direct analysis in real time (DART) ], a variety of related techniques are

currently available, including atmospheric pressure solid analysis probe

(ASAP) , electrospray-assisted laser desorption ionization (ELDI) , extractive

electrospray ionization (EESI) , desorption atmospheric pressure photo

ionization (DAPPI) and easy ambient sonic-spray ionization (EASI) .

4.1. ESI–QIT–MS For the analysis performed via quadrupole ion trap mass spectrometry (QIT–

MS), ANFO samples are injected with a syringe pump at a flow rate of 300

μL h−1 and the analysis is performed with a direct infusion electrospray

ionization source (ESI) operating in either the positive (+) or negative (−) ion

modes in an HCT Ultra spectrometer (Bruker—Bremen, Germany). The QIT

is operated in the ultra-scan mode over a range of m/z50–1000. The ion

charge control (ICC) target is set to 100,000 with a maximum accumulation

time of 100 ms. The ion source parameters are set as follows: ESI voltage of

−3000 V (positive ion mode) and 3000 V (negative ion mode), dry

temperature of 300 C, nebulizer pressure of 10 psi and dry gas flow of 5 L

min−1. MS/MS data are obtained in the negative ion mode with a

fragmentation amplitude of 1.07 V.

4.2. ESI–FTMS ANFO samples are measured in a 7.2 T LTQ FT Ultra mass spectrometer

202



(Thermo Scientific, Bremen, Germany) equipped with an electrospray

ionization source (ESI) operating in the positive and negative ion modes

under the following conditions: capillary voltage of ±3.0 kV, tube lens at

+115 and −125 V, capillary temperature of 280 C. Nitrogen was used as the

nebulization gas. The mass resolving power is m/Δm(50%) = 400,000 at m/z

400 and 100 μscans are accumulated for each spectrum. The data

acquisition is performed in the range of m/z 100–1000 using Xcalibur 2.0

software.

4.3. EASI–MS Direct desorption/ionization analysis via EASI–MS [40] is performed on a

single quadrupole mass spectrometer (LCMS-2010EV; Shimadzu Corp.,

Kyoto, Japan) equipped with a homemade EASI source [39]. Acidified

methanol (with 0.1% volume of formic acid) at a flow rate of 20 μL min−1 and

N2 at 100 psi is used to form the sonic spray for the positive and negative

ion modes. The entrance angle of the capillary relative to the sample surface

is 45.

5. Forensics for Home made Explosives

Background

Unlike standard explosive operational procedures, homemade explosives (or

improvised explosives) vary widely with regard to composition, morphology,

and preparation. Currently, forensic identification of homemade explosives

(HME) is inadequate for determining the procedures used during their

formulation, storage, and deployment [1]. Needed is a repository of data

characterizing HME behavior, which can be used by the forensic science

community, law enforcement organizations, and military agencies to

interpret pre- and post-blast forensic evidence, assist in developing insights

and protocols to disrupt/discourage HME production, and provide for safe

HME disposal.

203





A common formulation used widely by terrorists is the mixing of ammonium

nitrate (AN) with a number of different fuels to make oxidizer-fuel explosives.

Ammonium nitrate is often mixed with No. 2 diesel fuel oil (FO) or

nitromethane (NM); terrorists tend to use whatever fuel is readily available.

Countermeasures to restrict their distribution have focused on limited AN

access and fertilizer production with inert additives to inhibit AN explosive

capability.

The ANFO detonation behavior is considered by theoreticians to be non-ideal

because detonation reaction kinetics are relatively slow (long detonation

zone) and dependent on a variety of physical parameters. Such explosives

are generally high-porosity, low-density materials for which the fuel and

oxidizer are not mixed on a molecular level. They exhibit low detonation

velocities and larger detonation reaction zones. A review of ANFO is

presented by Hurley, which describes various physical parameters (e.g.,

charge diameter, fuel oil content, particle size, particle size distribution,

loading density, moisture content, porosity, additives, and charge

confinement) that influence detonation properties. Prill morphological effects

(i.e., bulk density, surface features, shape) on explosive behavior were

investigated by Rao et al. They argue that the maximum energy release will

depend on stoichiometric fuel/oxygen mixing and maximizing the surface

area throughout the prill. The interface region between the AN and FO has

the highest temperatures and chemical exothermic reactivity. The

confinement of ANFO had a strong influence on the detonation velocity (a

measure of the quantity of energy released) and the detonation front

curvature. A velocity interferometry system has been used to investigate the

effect of prill compaction and prill-to-prill interactions on detonation by low-

stress impact. They observed post-shock particle velocity variations and in-

between-prill jetting. Trace detection of explosives is difficult, with nearly

every analytical chemical method having been employed to this problem. As

discussed above, the non-ideal behavior of ANFO, and other similar HMEs,

make it difficult to predict their thermochemical and thermo-physical

204


characteristics (especially for ANFO which has been investigated extensively

in the past, e.g., over a variety of relevant physical conditions). To address

this issue, a data measurement approach is needed that 1) operates over

physical conditions and heating rates relevant to HME applications, and 2)

provides a means for comparative analysis of existing HME mixture

formulations.

HME precursors are studied individually, i.e., ammonium nitrate (AN) and

nitromethane (NM), using a novel laser-based technique (referred to as the

laser-driven thermal reactor) to characterize material

thermochemical/thermo-physical behavior. The technique provides

temporally resolved temperature-time thermograms, which are used to

identify trends and characteristics unique to each HME chemical mixture.

The objective of the current study was to evaluate the thermal

decomposition of HME oxidizer-fuel mixtures, i.e., ammonium

nitrate/nitromethane (ANNM, a single composition fuel) and AN mixed with

No. 2 diesel fuel oil (ANFO, a multi-hydrocarbon composition fuel) with

respect to sample steady-state temperature, initial mass, and fuel/mixture

mass fraction. Also, discussed are issues associated with sample aging and

changes in the mixture composition and volatility with time.

5.1 The Laser-driven Thermal Reactor (LDTR) The laser-driven thermal reactor ( LDTR) consists of a sphere-shaped, gold-

plated, copper-foil reactor (with an outer diameter of 18.2 ± 0.1 mm and

thickness of 0.14 mm) mounted within a vacuum chamber, along with

optical, gas supply, and computer-controlled data acquisition subsystems

(see Fig. 1 and Fig. 2). The reactor is open on the top and bottom for access

to within the sphere. The sample is supported on a customized K-type fine-

wire thermocouple (0.25 mm bead diameter, unsheathed, and 0.200 s ±

0.002 s response time) stationed at the center of the reactor (defining the

‘sample’ temperature). A second thermocouple is flush with the reactor wall

205




inner surface (defining the ‘reactor’ temperature). The reactor assembly is

heated from opposing sides by an infrared, continuous-wave, Nd: YAG laser

beam (operating at a wavelength of 1.064 μm) to achieve nearly uniform

sample temperature. The reactor is stationed within the vacuum chamber

that allows for control of the environment (e.g., gas pressure and

composition) around the sample. In general, measurements are carried out

at vacuum pressure, which eliminates thermal convection within the reactor

sphere, and thus conduction and radiation are the dominant modes of heat

transfer. The sample and reactor temperatures are recorded with respect to

time (referred to as ‘thermograms’) by the data acquisition system and then

processed for the specified thermochemical and thermo-physical information

of interest. The data-acquisition sampling rate is 1 MHz, which is averaged

and reported every 0.25 s because of the thermocouple response time.

Chemical-kinetics parameters of the process can also be deduced and

compared with the literature. The mode of thermal analysis for these

experiments is based on heating the reactor to a steady-state temperature

(referred to as the ‘heating-rate’ approach). Experiments were initially

carried out for the ‘baseline’ (i.e., thermogram measurements with a copper

pan substrate and no sample), and then repeated with the sample mixture.

5.2. HME precursors In this investigation, both ammonium nitrate/nitromethane (ANNM)

mixtures and ammonium nitrate/No. 2 diesel fuel oil mixtures (ANFO) were

formulated according to established recipe procedures . In general, mixtures

were prepared by dripping fuel onto AN particles with a syringe until the

desired stoichiometric value was ascertained with an electronic precision

balance. The mixture had the texture of a moist aggregated sediment. The

quantity of sample used for each experiment was 1–10 mg, and the laser

heating rate (Rlh, defined by the baseline thermogram maximum slope) was

20–100 K s−1. The use of HME-related chemicals required modification of the

reactor to ensure its safe and reliable operation. The issue of possible

206


reactor-material oxidation was addressed by gold-plating the inner surface

of the copper reactor sphere and copper pan substrate (used to support

liquid and solid-particulate sample on the thermocouple). The pan

dimensions were (4.95 ± 0.001) mm (inner diameter), (2.34 ± 0.001) mm

(height), and (0.125 ± 0.0006) mm (thickness). After each experiment, the

chamber was pumped and filled with inert gas to ensure that there were no

reaction-gas byproducts in the chamber when opened. When the chamber

was opened, a venting tube (located at the top of the chamber) provided

additional ventilation of the reaction gas byproducts. Also, issues were

addressed to avoid relatively long storage times for these chemicals.

Fig. 5. Schematic of the LDTR copper reactor sphere.

The size distribution and bulk density for the AN solid particles were

estimated, as well as the fuel concentration (either NM or FO), as generally

required by forensic analysis procedures. The AN particle size distribution is

an important physical property, in that HME thermal behavior (i.e.,

reactivity in heterogeneous processes) is affected through the particle

surface area. To estimate the AN particle size distribution, images of typical

AN particle shapes, were recorded using a microscope with a stereo camera

(10× magnification, numerical aperture of 0.3, and bright-field imaging). Fig.

207



8 A presents 8 different AN particle samples, which were characterized using

a Feret size distribution analysis. The AN size distribution was determined

for a total of 95 arbitrarily chosen particles. The particle size was defined by

the longest Feret diameter and measured using Image J software. The

software was challenged in some areas of the image where particles were

clustered and at the frame edges. This required manual designation of the

particle size and shape. The diameter estimates were grouped into four bins:

A) 0–0.25 mm, B) 0.25–0.5 mm, C) 0.5–0.75 mm, and D) 0.75–1 mm.

Fig. 6. Schematic of the LDTR vacuum chamber and subsystems.

208


Fig. 7. Schematic of the LDTR thermal analysis protocol using the

heating-rate approach. Each baseline run includes the change in sample

and reactor temperature with time for two different laser fluences.

209


Fig. 8. (A) Eight photographs of the various in AN particle shapes and

sizes. (B) Particle-size histogram obtained for the AN particles with the

peak in the size range of (0.25–0.50) mm.

Fig. 8B presents the longest Ferret diameter distribution (relative to the total

number of counted particles) for the sampled AN particles. Most of the

particles were sub-millimeter in length, and the distribution peaks for size

bin (0.25–0.50) mm. The material bulk density was determined by

estimating the volume of the particles within a container and obtaining the

particle mass from a commercial electronic mass balance. The literature

210



value for the crystal AN particle density is 1.72 g cm−3[25], however, the

measured values were lower due to particle packing effects (see Fig. 5 which

presents the bulk density with regard to the estimated sample volume).

Increasing the AN volume resulted in increased particle mass and “self-

packing”, with the calculated bulk density increasing from 0.2 g cm−3 to 0.8

g cm−3. The AN surface area was estimated by assuming an average particle

diameter of 0.5 mm (i.e., the largest size particles in the size bin with the

highest distribution probability in Fig. 8B).

Fig. 9. Variation of AN density with its volume. The value of AN density

at zero volume is the bulk value for AN from the archival literature.

211





Fig. 10. Variation of the fuel/HME mixture mass fraction with time

relative to the initial stoichiometric values for ANNM and ANFO.

Regression curve fit: w(t)NM = wNM,f + wNM,ife−t/τ where wNM,f = (0.32 ±

0.14)%,wNM,if = (1.32 ± 0.13)%, and τ = (2.45 ± 0.41) day for 0 < t < 3

days; w (t)FO = wFO,f + wFO,ife−t/τ wherewFO,f = (0.60 ± 0.01)%, wFO,if =

(0.36 ± 0.01)%, and τ = (11.54 ± 0.98) day for 0 < t < 33 days. Note that

the units are dimensionless (relative to the initial value). The error bars

represent the estimated expanded uncertainty for the fuel/HME

mixture mass fraction, as determined by the Type A evaluation of

uncertainty.

For ANNM, the recommended sample preparation procedures (for optimizing

reaction) were used along with an initial stoichiometric NM/ANNM mass-

fraction of about 29%. The total ANNM mass changed less than 0.5 mg (NM

evaporation reduced the NM/ANNM mass fraction from 29% to 3%) over a

period of three days (see Fig. 6), being much longer than the sample

preparation time of a few minutes. The vapor pressure for nitromethane is

3.65 kPa. It is important to estimate how the initial concentration of NM

changes with time. There are practical reasons for measuring the change,

which include: 1) from a measurement perspective, obtaining consistent

results for experiments over time (e.g., days/week), 2) from a law-

212



enforcement perspective, HME shelf life may provide important information

on activities prior to an attack, and 3) from a forensics perspective, on HME

sample preservation and potency. Regarding consistency, HMEs are

prepared in quantities larger than required per experiment, so it is

important to know the actual oxidizer-fuel ratio each time test samples are

selected from the batch.

Unlike NM (a single-component fuel), FO is composed of a number of

hydrocarbons that have different vapor pressures and chemical/ignition

properties. The relative mass for different hydrocarbon fractions within the

No. 2 diesel fuel oil is estimated to be 14% foriso-paraffin, 20% for N-

paraffin, 36% for cyclo-paraffin, 15% for mono-aromatics, and 1% for di-

aromatics . The volatility (i.e., vapor pressure) of each fraction is different

and will evaporate preferentially (i.e., evaporation of the lighter hydrocarbon

fractions occurs at a faster rate than the heavier fractions due to their

higher vapor pressure). The overall mean vapor pressure for diesel fuel oil is

about 0.3 kPa at 293 K. The recommended ratio of precursor materials

varies slightly from recipe to recipe, but, in general, the initial stoichiometric

FO/ANFO mass fraction should be about 6%. The ANFO heat of explosion is

3890 kJ kg−1 and decomposition is initiated at a temperature of 476 K, both

properties higher than most other AN mixtures. Measurements determined

that the initial FO/ANFO mass fraction changes with time at a slower rate

than for NM/ANNM, i.e., changing to a FO/ANFO mass fraction of about 3.6

% after 33 days.

6. Direct Analysis in Eeal Time Mass Spectrometry

Background

Rapid explosives detection with minimal sample preparation remains an

area of interest to forensic science, homeland security, and military

agencies. The class and state of explosives investigated varies widely across

these fields and may include military-grade secondary explosives,

213


homemade explosives (HMEs) such as nitrate esters and peroxides, primary

fuel-oxidizer explosives, improvised explosive device (IED) components,

synthesis precursors, and detonated decomposition products. There is an

extensive literature base involving the detection of military-grade nitrated

organic explosives, e.g. RDX or TNT, including common binders and

plasticizers. As non-state terrorism has evolved, there is an increased need

to explore the detection of HMEs, including their precursors, synthesis by-

products, impurities, and decomposition products. The detection of these

compounds and their mixtures may enhance chemical libraries or provide

valuable information into the route of synthesis and source materials.

Because of the potential additional information provided, it is crucial to not

only detect these compounds but also understand how they behave in

mixtures with the explosives.

One of the major classes of HMEs, nitrate ester explosives, represents a

broad category of organic explosives, both military-grade and homemade,

including compounds such as nitroglycerin (NG), pentaerythritol tetranitrate

(PETN), and erythritol tetranitrate (ETN). Nitrate esters are synthesized

through nitration of a sugar alcohol (polyol) precursor, as simplified in Fig.

1. This synthesis reaction may result in several potential by-products,

namely partially nitrated and polymeric nitrated species. While a number of

nitrate ester explosives exist, this work focuses on the nitration of glycerol

(leading to NG) and pentaerythritol (leading to PETN) and their potential

synthesis impurities. Both explosives have been synthesized militarily and

illicitly, and are studied extensively. A number of studies, mainly using

liquid chromatography mass spectrometry (LC/MS), have identified

impurities from NG and PETN synthesis, including partially nitrated and

dimer species. Partially nitrated species, such as 1,3-dinitroglycerin (DNG),

can occur from incomplete nitration of the sugar alcohol due to lack of

available free nitrate ions, abrupt cessation of the reaction, or other

improper synthesis conditions . Partially nitrated species can also form via

hydrolysis of a fully nitrated nitrate ester. Likewise, improper termination of

214

http://www.sciencedirect.com/science/article/pii/S0039914015305385%23f0005



the reaction process can cause the formation of dimer, trimer, and other

polymeric species. Dimer species can also be formed from impurities in the

starting sugar alcohol, such as the presence of diglycerol. In addition to by-

products formed during the reaction, a number of decomposition products,

also partially nitrated species, may be created. The relative amount of these

impurities and degradation products are dependent on the route of

synthesis and/or length of decomposition. Relative levels of DNG in

groundwater samples containing NG were 50% to 100% that of NG.

Similarly, the presence of PETriN in degradation samples ranged from 20%

to 100% that of PETN, while neat samples contained only approximately

0.1%.

DART-MS is an increasingly used ambient ionization (AI)-MS technique in

forensic science laboratories, requiring minimal or no sample preparation,

and has previously demonstrated detection of nitrate ester explosives and

their sugar alcohol precursors. Extensive details of DART ionization

pathways and analysis can be found in the literature. The characterization

presented here focused on identifying the mass spectral response of

potential by-products and decomposition products. The compounds

investigated included: 1-mononitroglycerin (1-MNG), 2-mononitroglycerin (2-

MNG), 1, 3-dinitroglycerin (DNG), pentaerythritol trinitrate (PETriN), and

dipentaerythritol hexanitrate (diPEHN). In addition to the DART-MS

characterization of these compounds, analysis of mixtures with their

corresponding explosive (NG or PETN) and/or precursors (glycerol or

pentaerythritol) was conducted. All compounds were readily detected in

negative ionization mode as adducts with anions such as molecular oxygen,

nitrite, and nitrate detectable in the low nanogram to sub nanogram range.

The detection of explosives from mixtures with by-products demonstrated

increasing ionization competition as by-product nitration increased. The

large volatility difference between monomers and dimers produce

significantly less competitive ionization with the explosive signal.

Competitive ionization for the mixtures examined increase as the mass ratio

215


of by-product to explosive increased.

Fig. 11. Reaction pathway and potential by-products that occur during

the synthesis and decomposition of nitrate ester explosives. In this

example, pentaerythritol, pentaerythritol tetranitrate, and related

compounds are shown.

7. Thermal Analysis Methodologies Thermal analysis methodologies are commonly used to analyze the thermal

changes of a sample due to heating and cooling. The sample characteristics

that are derived from thermal analysis techniques include thermal (e.g.,

temperature, heat, enthalpy), physical (e.g., mass, volume, strength), and

chemical (e.g., chemical composition of products) characteristics. There are

a variety of commercially available measurement techniques that have been

developed and used to meet these needs. Thermal analysis techniques, such

as differential scanning calorimetry (DSC) and thermal gravimetric analysis

(TGA) are commonly employed in research to provide information on the

material explosive/energetic content and other thermal properties (e.g.,

decomposition, heating value, and thermal stability), and also for

composition identification of an unknown explosive material.

The thermal dissociation of solid and liquid AN is analyzed using DSC/TGA

methods. The experimental measurements were carried out at low heating

rates between 2.5 and 12.5 °C min−1. The resulting chemical kinetics

mechanism is assigned to the process of dissociative

216


sublimation/vaporization. Oxley et al. carried out an investigation using

DSC to explore if this technique could be used as a benchmark for

evaluating the explosivity of ammonium nitrate. In this study, thermal

analysis was used to screen a large number of AN formulations in search of

possible deterrents. They found that sodium, potassium, ammonium and

calcium salts of sulfate, phosphate, or carbonate, as well as certain high-

nitrogen organics, can enhance AN thermal stability. The need to improve

thermal analysis techniques was also emphasized by Desilets et al. [6]. In

this work, the thermal decomposition of urea nitrate (UN) was studied at

higher temperatures. The UN non-isothermal decomposition kinetics and

products of reaction were analyzed using a DSC/TGA instrument in

conjunction with Fourier transform infrared and mass spectroscopic

instruments. Results indicated differences in the exothermic values of UN

thermal decomposition when compared to a dedicated DSC. The above-

mentioned studies represent some of the relevant AN thermal analysis

issues that arise with commonly used DSC and TGA techniques.

The data obtained with these instruments may underestimate the explosive

potential of a HME material under certain conditions. The non-standard

nature of the HME chemical composition and formulation/synthesis

procedures can present a formidable challenge for forensic analysis

instrumentation. Thus, improved techniques are required to provide data for

adequate HME forensic identification, which may not be accomplished

reliably with DSC and TGA. Development of a new methodology and

database of thermal and chemical signatures, along with other

thermophysical properties, of HME materials is required for different

compositions and ambient conditions.

8. Conclusions Whilst already having made a strong contribution in environmental

applications, many methods including IC-MS also has a strong potential for

217



application in forensic analysis of explosive residues Low molecular weight

ions can be separated via a range of IC modes, many of which were also

compatible with ESI-MS or ICP-MS detection. In particular, IC-MS offers

advantages over other techniques in terms of its ability to speciate, as well

as offering chromatographic reproducibility and sensitivity at the trace level,

often in complex matrices. There exists further promises for ultra-trace

determinations of explosive residues using online pre-treatment, further

eluent modification, post-column ion enhancement, dual anion/cation

separations and semi-targeted screening using IC-HRMS technologies.

References 1. Ashot Nazarian, Cary Presser, Forensic methodology for the thermochemical

characterization of ANNM and ANFO homemade explosives, Thermochimica

Acta, Volume 608, 20 May 2015.

2. Edward Sisco, Thomas P. Forbes, Direct analysis in real time mass

spectrometry of potential by-products from homemade nitrate ester explosive

synthesis, Talanta, Volume 150, 1 April 2016.

3. Holly A. Yu, Simon W. Lewis, Matthew S. Beardah, Niamh NicDaeid Assessing

a novel contact heater as a new method of recovering explosives traces from

porous surfaces Talanta, Volume 148, 1 February 2016.

4. Ashot Nazarian, Cary Presser Forensic analysis methodology for thermal and

chemical characterization of homemade explosives Thermochimica Acta,

Volume 576, 20 January 2014.

5. D Xiaoma Xu, Mattijs Koeberg, Chris-Jan Kuijpers, Eric Kok evelopment and

validation of highly selective screening and confirmatory methods for the

qualitative forensic analysis of organic explosive compounds with high

performance liquid chromatography coupled with (photodiode array and) LTQ

ion trap/Orbitrap mass spectrometric detections (HPLC-(PDA)-LTQOrbitrap)

Science & Justice, Volume 54, Issue 1, January 2014.

218

http://www.sciencedirect.com/science/article/pii/S0040603115001355





http://www.sciencedirect.com/science/article/pii/S003991401530206X











6. Isotopic and elemental profiling of ammonium nitrate in forensic explosives

investigations Forensic Science International, Volume 248, March 2015, Pages

101-112 Hanneke Brust, Mattijs Koeberg, Antoine van der Heijden, Wim

Wiarda, Ines Mügler, Marianne Schrader, Gabriel Vivo-Truyols, Peter

Schoenmakers, Arian van Asten

7. C Vinicius Veri Hernandes, Marcos Fernado Franco, Jandyson Machado

Santos, Jose J. Melendez-Perez, Damila Rodrigues de Morais, Werickson

Fortunato de Carvalho Rocha, Rodrigo Borges, Wanderley de Souza, Jorge

Jardim Zacca, Lucio Paulo Lima Logrado, Marcos Nogueira Eberlin, Deleon

Nascimento Correa characterization of ANFO explosive by high accuracy

ESI(±)–FTMS with forensic identification on real samples by EASI(−)–MS

Forensic Science International, Volume 249, April 2015.

219









CHAPTER – 7

CHARACTERIZATION OF EXPLOSIVES FOR COAL MINES

1. Introduction Powdery explosives for coal-mining, especially those for use in coal-mines

where fire damp/air mixtures can be present, are mixtures of inorganic

salts, e.g. sodium nitrate, ammonium nitrate, ammonium chloride and

organic nitrate, e.g. glyceroltrinitrate and glycoldinitrate. In addition are

present small quantities of calcium carbonate, metal salts of fatty acids and

waterproofing agents like guar powder or wheat flour.

Shot firing in coal mines constitutes a risk in the presence of firedamp and

coal dust. Permitted explosives are special compositions which produce

short-lived detonation flames and do not ignite methane-air or coal-

dust-air mixtures. The methane oxidation CH4 +2O2 = CO2 +2 H2O needs

an “induction period” before the reaction proceeds. If the time required

for ignition by the detonation flames is shorter than the induction

period, then ignition of firedamp will not occur. Thus, the composition of

permitted explosives must ensure that any secondary reactions with a

rather long duration, which follow the primary reaction in the detonation

front, are suppressed and that slow W Deflagration reactions are avoided (W

Audibert Tube). Such explosives are known as “permissibles” in the USA, as

“permitted explosives” in the United Kingdom, as “Wettersprengstoffe” in

Germany, as “explosifs antigrisouteux” in France, and as “explosifs S.G.P.”

(sécurité grisou poussières) in Belgium. Safety measures to avoid ignition

of firedamp uses salt (NaCl, Ca(NO3)2, NaNO3 etc.) which is included in the

usual compositions of commercial explosives. It lowers the explosion

temperature and shortens the detonation flame.

220

Chapter 7 Characterization of Explosives for Coal Mines

Permitted explosives which are enveloped in a special salted cooling

“sheath”; they thus, got the name “Sheathed”. High-safety explosives, such

as sheathed explosives, whose structure is nevertheless homogeneous, are

known as “explosives equivalent to sheathed” (Eq. S.)

Higher safety grades are achieved in ion exchange explosives in which the

ammonium and sodium (or potassium) ions are exchanged; instead of

NH4NO3 + (inert) NaCl = N2 + 2 H2O + 1/2O2+(inert) NaCl the reaction is:

NH4Cl+NaNO3 (or KNO3) = N2+2 H2O + 1/2O2 + NaCl (or KCl). Thus, a flame-

extinguishing smoke of very fine salt particles is produced by the

decomposition reaction itself. Combinations of salt-pair reactions and

“classic” detonation reactions quenched by adding salt are possible.

Permitted explosives with a higher grade of safety are powder explosives.

They contain a minimum percentage of nitroglycerine-nitroglycol to ensure

reliable initiation and transmission of detonation and to exclude slow

deflagration reactions.

Explosive that has passed the Buxton tests and has been placed on the

British list of authorized explosives, implying that they are reasonably safe

to manufacture, handle, transport, and use in safety-lamp mines. Upon

detonation, a permitted explosive: (1) gives off the minimum possible

quantity of noxious gases, and (2) produces a flame of the lowest possible

temperature and shortest possible duration, to lessen the risk of

combustible gases ignition. The explosive contains cooling agents, such as

sodium chloride and sodium bicarbonate. The first British list of permitted

explosives was published in 1899. Permitted explosives are divided into four

groups: P.1., normal permitted explosives; P.2., sheathed explosives; P.3.,

equivalent to sheathed explosive; P.4., permitted explosives that have passed

additional and more stringent tests.

221


Explosives for the coal-mining are additionally regulated and controlled by

local –mining authorities. For the manufacturers of explosives it is therefore

necessary to produce materials, which are in compliance with these

regulations.

The properties of permitted explosives that affect their safety in use are as

follows: (a) incendivity, i.e. the likelihood that mixtures of methane and air,

or of coal dust and air, will be ignited by the action of blasting; (b)

deflagration, i.e. a naked flame after blasting as a result of the explosives

burning rather than detonating, which is extremely undesirable in an

underground coal mine; (c) impact and friction sensitivity; and (d) after-blast

fumes.

The produced explosive materials must have the same composition as the

first registrated and permitted samples. The productions must be controlled

by analysis of the manufactured material to determine the correct

composition.

Though at the first glance the analysis of the mixture of the few inorganic

salts seems to be simple, it is not so easy in practical work because of the

mutual influence of the salts with each other and the other components. The

following contribution describes methods which are used for the

characterisation of explosive materials for coal-mining. Table 1 gives three

types of coal mining explosives as description.

2. Quality Control The characterization of permitted explosives is performed under two

different aspects: Routine control of each batch and full analysis as random

test for statistical quality control.

This work is done by combination of classical wet analysis and modern

222


instrumental analysis.

Table 1 Gives a Qualitative of the three types of Coal-mining Explosives.

Type A Type B Type C

Nitroglycerin Nitroglycerin Nitroglycerin

Ethylenglycol- Ethylenglycol- Ethylenglycol- dynitrate dynitrate dynitrate

NH4Cl NH4Cl NH4Cl

KNO NaNO3/Ca(NO3)2 NH4NO3

CaCO3 CaCO3 CaCO3

3. The Probability of Ignition of Firedamp by Permissible Explosives

The so-called permissible, or permitted, explosives are used in

underground and confined works where there is a risk of occurrence of

flammable gases. They provide a lower temperature, shorter duration

flame, with a lower probability of ignition of methane or coal dust than

regular explosives. In permissible explosives, in some cases, Calcium

Nitrate is used which decomposes at a much lower rate than Ammonium

Nitrate thus reduces explosion temperature- at reduced strength but at a

lower risk of inflaming combustible methane or other gases. The evaluation

of this and other critical safety-related properties is done by conducting

elaborate tests (HSE, 1988 and MSHA, 1988) designed to model the

explosive-gas interaction in controlled laboratory conditions usually carried

out in government or independent institutions. As a result of such tests,

explosives are classified in various safety levels or classes. Besides holding

the “permissible” certification, explosives can only be used in potential

firedamp conditions under strict firing prescriptions, mainly related with the

charge to be fired per delay or per total round, and the allowable delay

times, which depend on the class in which the explosive is rated and the

type of working face where the blast is to be conducted. Regulations in coal

223


blasting generally restrict the amount of explosive to be loaded per hole. For

instance, the Spanish regulation states that no more than 2 kg of Class III

explosive can be loaded per hole unless special permission is issued. Such

restrictions are reasonable for typical drift blasting where holes are about

3 m long, but cannot be met in sublevel caving ring blasting with holes 10 or

20 m long; in such cases, blasting with higher charges per hole is carried

out, under special supervision. Besides the increased amount of explosive,

blasting with long holes may require some charging and priming

techniques that are not typically used with short holes: PVC casing to

facilitate the charge preparation and blasthole loading; detonating cord

along the charge to ensure continuity of the detonation; double (top and

bottom) initiation to reduce the probability of a failed shot hole, and

inhibitor salt decks. This part assesses the effect of these practices on the

probability of ignition of a methane–air atmosphere.

4. Gallery Testing of Permissible Explosives Regulations for the approval and use of explosives to be used under

potential firedamp conditions are enforced in all coal-mining countries. The

approval requires a certain capacity of no ignition of a methane atmosphere;

the main instrument for testing this capacity is the test gallery. Sanchidrián

et al (2008) reports a study on the permissible explosives under different

simulated field conditions with a gallery (Fig. 1) that consists of a steel

cylinder with one end permanently closed and the other end equipped with a

light closing device such as a paper sheet and a sealing ring. Inside this

closed cylinder, or explosion chamber, the explosive is positioned either

within a mortar or freely suspended. After positioning the mortar and

charging, the explosion chamber is sealed and filled with a methane–air

mixture of 9% methane and the charge is fired. Whether or not ignition of

the gas occurs is observed from a safe position.

224



Fig. 1. Test gallery. Different mortars and test conditions can be set to give a higher or lower

probability of ignition so that different safety grades of explosives can be

defined. We have used in the present work the angle mortar (Fig. 2): a steel

cylinder of 219 mm in diameter and 2 m in length with a right-angled

groove; it is positioned in the explosion chamber against a steel plate at

various distances and angles. Trains of several cartridges up to the full

mortar length are placed in the angle and fired in the methane–air mixture.

5. Test Series The explosive tested was a permissible ion-exchange explosive (class III

according to the Spanish classification [4]) in 32 mm diameter cartridges

with a nominal cartridge mass of 215 g. Class III explosives are not required

to pass the angle mortar test under the Spanish regulation; the Class III and

angle mortar combination was selected in order to provide a significant

probability of ignition for the statistical analysis with the different charging

characteristics. Several trial shots varying the orientation of the mortar's

angle and the distance to the reflection plate were done in order to obtain a

configuration that assured a high enough ignition frequency for the range of

explosive mass allowed by the length of the mortar and the resistance of the

chamber. Table 1 lists the configuration parameters of the test series

performed which are sketched in Fig. 3. The step values given in Table 1 are

the mean weights of a cartridge or half cartridge in the corresponding series.

225







Fig. 2. Angle mortar test layout.

Fig. 3. Test series charge configurations

Table 1.Test series configuration

226


Ser-ies Firing conditions

Distance to plate, LP

(cm)

Reflection length, LR

(cm)

Step (g)

1 Bare charge 47 32 215.2 2 PVC casing 47 32 214.5 3 PVC casing/detonating cord 47 32 214.9 4 PVC casing/detonating cord/salt 40 35 107.5 5 PVC casing/detonating cord 40 35 107.0 6 PVC casing/detonating cord/

double initiation 40 35 107.2

Table 2 gives the charge masses of all the shots for the six series.

Table 2. Charge masses (g)

Series # 1 2 3 4 5 6 No ignition 862 1274 1075 642 317 430

845 1281 855 635 440 438 1094 1063 1073 445 553 545 1275 866 1062 637 635 634 1067 1074 652 422 426 441 1305 1284 652 434 426 528 1496 1273 856 430 539 418 1074 1296 859 622 427 224 1089 1289 846 427 426 316 1290 1074 871 428 537 420 1301 1076 851 649 417 211 1292 1288 855 641 555 315 1494 1504 544 635 422 520 530

Total (m) 13 13 12 15 13 13

Ignition 1088 1515 1275 859 441 547 1500 1511 1283 863 739 746 1295 1302 1078 872 642 638 1735 1290 1302 645 530 529 1514 1072 1274 854 546 630 1286 1531 1086 628 648 551

227


1282 1478 847 664 547 548 1500 1508 854 648 534 426 1496 1488 1071 871 646 328 1731 1289 1078 645 532 528 1498 1269 1078 640 752 456 1300 1701 1055 862 653 319 1073 866 549 536 537 530 774 738 764 746 754 758

Total (n) 12 12 13 23 12 12 The distributions are shown in Fig. 4 in which the charge mass xp for a

probability p (probability points, or p-quantiles of F) is calculated.

Fig. 4. Distributions of probability of ignition.

6. Effect of the Various Firing Techniques on the Ignitability of

228


Firedamp

1. The casing of the charge with a PVC slotted pipe, useful in practice for

the charging of the holes, does not have an influence on the ignition

probability.

2. Detonating cord increases the probability of ignition with statistical

significance of 95%; for an ignition probability of 0.5, the charge mass is

reduced by a 26% when detonating cord is used.

3. The influence of salt is very limited in the amount used in the present

study (7% salt/explosive length ratio). In spite of an increase of 8% in

the mass at an ignition probability of 0.5, the use of salt in this quantity

cannot be considered an influential parameter at a 95% confidence level.

4. The double initiation increases the probability of ignition, the charge

mass being on average 14% smaller than with one-end initiation for a

0.5 probability of ignition. However, as with the use of salt, the influence

cannot be stated at a 95% confidence level.

REFERENCES

1. P. Lingens, in: Ullmanns Enzycopadie der techn. Chemie Verlag Chemie,

Weinheim, 1982, Band 21, 4. Auflage.

2. H. Kohler, in: Proceedings of the ICT international conference, Karlsruhe,

1984.

3. Ph. Naoum, Nitroglycerin and Nitroglycerin sprengstoffe, Springer Verlag,

Berlin, 1924.

229


4. HSE Test and approval of explosives for use in coal mines and other mines in

which flammable gas may be a hazard Testing Memorandum TM2, Health and

Safety Executive, Buxton, UK (1988).

5. MSHA Requirements for approval of explosives and sheathed explosive units

Code of Federal Regulations Title 30, Part 15.20, Mine Safety and Health

Administration, U.S. Department of Labor, Arlington, VA (1988).

6. J.A. Sanchidrián,, L.M. López, P. Segarra The influence of some blasting

techniques on the probability of ignition of firedamp by permissible explosives,

Journal of Hazardous Materials ,Volume 155, Issue 3, 15 July 2008.

230


http://www.sciencedirect.com/science/journal/03043894








CHAPTER – 8

HOME-MADE EXPLOSIVES AND THEIR EFFECTS

1. Introduction According to the US Government Hazardous Substances Database, several

substances and mixtures can be used for the realization of this kind of

explosives, starting from chemicals sold in markets and pharmacies. Among

others, two were often adopted for terrorist attacks, suicide bombing, and

other malicious uses: Ammonium Nitrate (AN) − Fuel Oil (i.e. ANFO)

mixtures and Acetone Peroxide or Triacetone Triperoxide Peroxyacetone

(TATP) mixtures. These two materials were considered as reference

explosives for the analysis presented in the present study.

ANFO is a tertiary explosive (note that TNT is a secondary explosive) and is

generally composed by 94% of AN prills and 6% of adsorbed fuel oil. It is

extensively used for several authorized purposes as in mine blasting. The

TNT equivalence is typically around 80% and the ideal explosion (detonation)

energy is 3890 kJ/kg (pure ammonium nitrate has an explosion energy of

1592 kJ/kg). AN prills used for mining applications are however physically

different from fertilizer prills used in home-made explosives. The commercial

ammonium nitrate prills used for mine blasting have a 20% void fractions

and are coated with #2 fuel oil (mainly C10−C20 linear hydrocarbons) or

kerosene. Hence, ANFO has a bulk density of approximately 840 kg/m3

when starting from ANprills for mining applications, having a density of

about 1300 kg/m3 (the density of pure crystalline ammonium nitrate is

1700 kg/m3). On the other hand, homemade explosives prepared from AN

fertilizers do not have a so high void fraction and are less efficient: e.g. the

new European regulations for fertilizers state that they must contain less

than 45% of AN (16% N) for being traded to the general public. Such

231

Capter 8 Home-made Explosives and their Effects

fertilizers still may be used to obtain explosives, but require processing to

achieve a detonation. If commercial AN (containing about 50% of inert, as

dolomite) and diesel fuel is used, a detonation energy of about 1071 kJ/kg is

obtained, much less than pure ANFO. Furthermore, it has been observed

that when amounts of dolomite higher than 30%, are present, no detonation

is observed.

2. Triacetone Triperoxide Peroxyacetone (TATP) TATP is a primary explosive which is notable since it does not contain

nitrogen. Thus, it is used to avoid conventional chemical bomb detection

systems, and it is almost undetectable by either analytical system or by

sniffer dogs. It can be obtained from common household items such as

sulphuric acid, hydrogen peroxide, and acetone.

TATP is very unstable: it can be ignited by touch and can explode

spontaneously. It is often used for improvised detonators itself. It is actually

composed by isomers and conformers, the dimer being more stable but

having lower decomposition energy (see Fig.1.).

(a) dimmer (b) trimer

Fig. 1. The actual components of TATP: (a) dimer and

(b) trimer peroxides.

The density of the pure molecule is typically considered to be 1220 kg/m3.

232



However, home-made TAPT formulations are typically in the range of 450–

500 kg/m3. Finally, TATP is often stabilized with carbonaceous liquids and

waxes so that the net charge is even lower. Nevertheless, Lefebvre et al. [38]

have demonstrated that home-made TATP is a primary explosive and very

sensitive to impact or friction, although the strength of explosion may

strongly vary since the quality of the final product is very sensitive to the

temperature during its synthesis.

TATP is highly volatile and decomposes to form large number of gas phase

molecules (entropic explosion). Acetone and ozone are predicted to be the

main decomposition products, along with oxygen, methyl acetate, ethane,

and carbon dioxide.

3. Assessment of Explosion Effects for Improvised Explosives The aim of the study is to support consequence and vulnerability

assessments of industrial plants (hence equipment) when subject to a shock

wave produced by improvised explosive devices based on ANFO or TATP. To

this aim, the calculation of the pressure history (maximum pressure,

positive duration, impulse) with respect to the distance from the explosion

point is needed. In order to obtain this information, the amount of explosive

and its efficiency with respect to an equivalent amount of TNT (WTNT) is

required. The Hopkinson−Cranz methodology to calculate the mass-scaled

distance (Z) is typically adopted for point-source explosives:

equation(1)

equation(2) where Wexp,n is the net mass of the explosive, ΔHTNT is the TNT explosion

energy, which is typically 4.6 MJ/kg, ΔHexp is the explosion energy of the

material of interest, related to the primary explosion only (decomposition or

233



detonation energy) and not to the overall combustion energy, r represents

the actual distance from the center of the explosion, and κ and η are

respectively (i) a coefficient which depends on the confinement or, more

specifically, to the mechanical energy adsorbed for the deformation and

failure of the containment system; (ii) the explosion efficiency, i.e. the energy

proportionality factor, which is discussed later. For explosives contained in

low-strength enclosure, a factor κ=0.7 could be adopted, however we used

unitary values in order to obtain safe-side evaluations.

Several previous publications provide data, references and correlations for

the shock wave produced by ANFO and TATP. What is relevant for the

present study is that: (i) the explosion energy gives a good reproduction of

the destructive power of the substances at constant, atmospheric pressure

(which is the case hereby analyzed); (ii) light confinement (even a paper

confinement) approximately doubles the severity of the explosion; (iii) high-

strength confinement as the steel case adopted for bombs and military

explosive devices has been not considered, hence the corresponding effects

of casing fragmentation have been neglected; and (iv) the energy output from

non-ideal explosives is dependent on charge size, which makes it difficult to

define it with traditional modeling methods. However, because the energy

output decreases with the charge size, we neglected the variation of TNT

efficiency adopting the correspondent maximum value in order to obtain

conservative estimations.

Table 1 reports the TNT efficiency (η) obtained from specific studies and

calculated as the ratio between the explosion energy of the mixture of

interest and the explosion energy of TNT (namely, detonation energy ratio)

for different types of explosives. In the case of non-ideal mixtures, the

property values have been calculated by using the Chemical Equilibrium

Model (CEA) as previously done in the literature for black powder and

pyrotechnics. Quite clearly, discrepancies can be found with respect of the

reaction heats given in the open literature. However, the figures

234

http://www.sciencedirect.com/science/article/pii/S0951832015000629%23t0005


obtained by the CEA model are at least indicative of the explosion energies

involved.

Table 1 reports data for home-made explosives produced with 90% and 50%

non-porous AN in mass with dolomite as inert material mixed with fuel oil

(respectively “AN/dolomite (90/10)+diesel fuel” and “AN/dolomite

(50/50)+diesel fuel”). It is worth noticing that the efficiency of pure ANFO is

consistently larger than that of possible home-made explosives based on

fertilizers, in which some inert material as dolomite is used, and diesel fuel

and non-porous AN as explosive component are employed. Reduction to very

low values of TNT efficiencies can clearly be observed.

Table 1 : Experimental heat of explosion (ΔHexp) and combustion

(ΔHcomb) and TNT efficiencies (η) for the analyzed explosives.

AN : ammonium nitrate; FO: fuel oil; DF: diesel fuel.

Explosive

ΔHexp (kJ/kg)

ΔHcomb (kJ/kg)

η(dimensionless) Literature

data Calculateda

TNT 4680 14961 1.00 1.00

ANFO (94% AN; 6% FO) 3890 578 0.60–0.88 0.83

TATP (Trimer) 2803 28192 0.30–0.92b 0.60

DADP (Dimer) − 23465 − 1.26

AN/dolomite (90/10)+DF 3234 − − 0.69

AN/dolomite (50/50)+DF 1071 − − 0.23

a Ratio between the explosion energy of the mixture of interest and the explosion energy of TNT (namely, detonation energy ratio).

b Mixture of isomers.

Finally, it is well known that η refers to the shock wave produced in

unconfined explosion, which is only related to the primary explosion.

Besides, the heat of combustion, which is much larger with respect to the

235


http://www.sciencedirect.com/science/article/pii/S0951832015000629%23tbl1fna

http://www.sciencedirect.com/science/article/pii/S0951832015000629%23tbl1fnb


explosion energy, should be considered for quasi-static analysis if

confinement is taken into account [48].

On the basis of the above assessed TNT efficiency values, the peak

overpressure may be estimated by adapting the following literature

correlation [49]:

equation(3) where Ps (bar) is the peak overpressure, r (m) is the distance from the center

of the explosion and WTNT is the equivalent mass of TNT expressed in kg,

calculated accordingly to the following expression for a given amount of

home-made explosive (Wexp expressed in kg):

equation(4) Where Wexp is the overall amount of home-made explosive including

additives, and f is the actual mass fraction of the explosive material, which

is introduced in order to consider the possible presence of inert materials in

home-made explosives.

Combining Eqs. (3) and (4), modified TNT diagrams were be obtained for

each of the explosive material obtained. The modified diagrams (see Section

4.1) allow the straightforward assessment of overpressure as a function of

distance, of explosive type and of explosive amount.

4. Assessment of Damage to Equipment Since the main aim of the article is determining the damage and impact of

blast waves caused by improvised explosive devices on process equipment,

vulnerability models need to be introduced to assess potential damages and

236



http://www.sciencedirect.com/science/article/pii/S0951832015000629%23eq0015


http://www.sciencedirect.com/science/article/pii/S0951832015000629%23s0040

http://www.sciencedirect.com/science/article/pii/S0951832015000629%23s0040


to verify the possibility of escalation triggering a domino chain leading to

further damage caused by fires [52], explosions [53] and fragment projection

[54] and [55]. Vulnerability of process equipment to blast waves depends on

the pressure history, and large difficulties arise if a deterministic analysis is

required. Hence, for the aim of the present study, existing dose-effects

analysis based on peak overpressure will be adopted, assuming

conservatively a static interaction [56], [57], [58] and [59]. In the static

analysis, hence neglecting the dynamic contribution, the escalation

threshold for the explosion resilience depends on the equipment

construction and operation characteristics, and is expressed as the peak

overpressure value at the position of the target equipment. A procedure and

specific threshold data for equipment damage assessment due to blast

waves are detailed in previous studies [56], [57], [58], [59] and [60]. The

quality of available data only allows defining broad categories of equipment

[60]. Table 2 summarizes the categories of equipment considered and the

corresponding overpressure threshold values for damage and escalation.

Table 2 : Escalation thresholds for the escalation due overpressure and

heat radiation for different equipment categories.

Equipment category Overpressure (bar)

Heat radiation (kW/m2)

Atmospheric vessels 0.22 15

Pressurized vessels 0.20 45

Pressurized elongated vessels

(toxic materials)

0.20 45

Pressurized elongated vessels

(flammable materials)

0.31 45

5. Impact Distances of Improvised Explosives The methodology adopted for the impact assessment of the improvised

237

















explosives allows obtaining the impact charts shown in 2. In each chart, the

peak overpressure is reported as a function of the distance given the

explosive quantity (in kg), according to Eq. (3).

Fig. 2. Impact charts for the estimation of the peak overpressure (kPa)

as a function of the distance (m) from the explosion centre and of the

amount of explosive. (a) TATP; (b) AN/dolomite (50/50)+DF; (c) TNT.

2a shows the data obtained for TATP. Large amounts of this explosive are

238





too hazardous to produce, transport, and manipulate. Indeed, TATP is

typically an explosive adopted for single-man suicide attacks [35] and [41]. A

net-charge of 50 kg can be transported e.g. in backpack, while higher

quantities are deemed not to be credible in a terrorist attack as highlighted

in 2a due to both stability and transportation problems. Therefore, the

impact associated to TATP explosions may be significant only close to the

target equipment and it may be prevented adequately managing the physical

and operational security and the access to the industrial site.

Fig. 3. Calculated stand-off distance for different equipment categories

for the explosives considered in Table 1: (a) Atmospheric vessels; (b)

Pressurized vessels; (c) Pressurized horizontal vessels (toxic); (d)

239






Pressurized horizontal vessels (flammable).

Conversely, self-produced AN explosives may be obtained in extremely large

quantities, e.g. from several kilograms up to quantities as high as e.g. 50 t.

In the latter case, the explosive can be positioned outside the restricted

industrial area, loaded in a car or even in a truck parked on the road

adjacent to the fence of the industrial facility. This is confirmed by the

results in 2b, that show the impact chart for AN/dolomite (50/50) with

diesel fuel mixture (see Table 1). Even if the efficiency and the explosive

fraction (e.g. 50%) are limited, extremely high amounts of this type of

explosive may have a significant impact on equipment and structures even

from distances of few hundred meters.

The methodology defined may also be used to provide data for other

improvised explosives. Eq.(4) may be applied in order to estimate the

equivalent amount of TNT (WTNT). Hence, the classical TNT mass-scaled

chart (2c) may be used to estimate the impact distance.

In order to determine the potential impact of a terrorist attack carried out

with home-made explosives against process equipment, “stand-off distances”

were evaluated. In the present study, the stand-off distance is defined as the

minimum distance between the asset of interest and the area where an

explosive device can be placed without causing damages.

Fig.3 shows the calculated stand-off distances for several types of industrial

equipment reported as a function of the net explosive mass in the home-

made explosive charge. Fig. 5 was obtained applying Eqs. (3) and (4), and

considering the threshold values for domino effects (Table 2).

Fig.3a shows the results obtained for atmospheric equipment. Fig.3b and c

the results for different categories of pressurized equipment. Pure ANFO and

TATP exhibit similar results, due to the similar efficiency and to the absence

240













of inert material. However, the presence of dolomite and the lower efficiency

deplete AN improvised explosives. As shown in 3, the stand-off distance for

this type of explosive is about half of that calculated for the correspondent

pure explosive.

6. Conclusions In the present study, the potentiality of home-made improvised explosives to

trigger domino scenarios was investigated. A preliminary characterization of

improvised explosives was carried out determining the explosion potential

and stand-off distances. The classical TNT-energy or mass-scaled analysis

has been proved to give sufficiently accurate results with CHN-based high-

energy explosives similar to TNT. However, further work should be devoted

for non-ideal substances with lower explosion energies, or for the behavior of

an explosive material when density, composition, humidity and other

chemical and physical parameters affects their efficiency. The methodology

developed may be also applied to unintended explosion if solid explosives are

considered.

7. References

1. M.F. Milazzo, G. Ancione, R. Lisi, C. Vianello, G. Maschio Risk management of

terrorist attacks in the transport of hazardous materials using dynamic

geoevents J Loss Prev Process Ind, 22 (2009).

2. S. Bajpai, J.P. Gupta Terror-proofing chemical process industries Process Saf

Environ Prot, 85 (2007).

3. D.A. Moore, B. Fuller, M. Hazzan, J.W. Jones Development of a security

vulnerability assessment process for the RAMCAP chemical sector J Hazard

Mater, 142 (2007).

4. D.A. Moore, B. Fuller, D.A. Jones, M. Hazzan LNG security vulnerability

241



assessment AIChEAnnu Meet ConfProc (2006).

242

CHAPTER – 9

EXPLOSIONS OF CONSTITUENT SUBSTANCES OF EXPLOSIVE:

SPECIAL EMPHASIS ON AMMONIUM NITRATE

1. Introduction

On 17 April 2013 a massive explosion occurred in the city of West, Texas, a

small town in central (not western) Texas. The West Fertilizer Co. operated a

facility for storing, blending, and retailing of fertilizers and other agricultural

goods and services. The company functioned as a manufacturer (blending of

chemicals is defined as a manufacturing activity in the US), a retailer, and a

provider of services. The building dated from 1962 and was of Type V-N,

unprotected wood-frame, construction. The town of West possessed a

volunteer fire department but had neither a building code nor a fire code.

Shortly after closing time on the day of the disaster, a fire origin erupted in a

seed storage room adjacent to the ammonium nitrate (AN) storage areas of

the facility. The building had no alarm system nor fire sprinkler protection,

and the fire department was notified only when a man walking his dog in a

nearby park saw smoke rising from the building [4]. The volunteer fire

department arrived to find a large working fire within the premises. They

were able to apply two hose streams without much effect, but could not get

a better water supply since the closest fire hydrant was some 1600 ft (490

m) away. At 1951 h, some 21 min after the original phone call and 11 min

after the arrival of the first fire engine, a massive explosion occurred (Fig. 1).

Fifteen persons were killed, mostly volunteer firefighters, and some 260

persons were injured. The explosion registered a magnitude of 2.1 on the

Richter scale. Not only the fertilizer facility, but a number of nearby

buildings were demolished, including a nursing home. The facility was

243



Chapter 9 Explosions of Constituent Substances of Explosive: Emphasis on Ammonium Nitrate

storing some estimated 40–60 tons of AN inside the building, not all of

which detonated. In addition, 100 tons of AN were located in a railcar

outside, and this railcar overturned in the explosion but did not explode.

Investigations concluded that the fire was accidental and did not reveal any

improprieties on the part of the staff of the fertilizer company. The fire likely

originated due to an electrical fault of an undetermined nature. What is of

great concern is that the facility that exploded was constructed and operated

similarly to most rural AN fertilizer facilities in the US. The Texas State Fire

Marshal identified that there are at least 104 facilities in that State which

store 5 tons or more of AN, while in the US as a whole there are over 1300

storage establishments.

Fig. 1. The site of the West Fertilizer Co. before and after the explosion (Source: ATF). 2. Fire and Explosion Risk of Nitroglycerine Nitroglycerine (NG), also known as glycerol trinitrate, is usually produced

from the glycerine nitration. Its nomenclature is 1,2,3-propanetriol

trinitrate, chemical formula is C3H5(ONO2)3, and molecular weight is 227

g/mol. The industrial product is a yellowish oily liquid having a density of

1.591 g/cm3 at 25 C. Nitroglycerine has explosive properties and has been

widely used as ingredient of explosives and propellant for more than a

hundred years. It is the main component of high explosives such as

244


dynamites, as well as an ingredient in most mining explosives. It was used

extensively in smokeless powder and together with nitrocellulose as a so-

called double base propellant, or with nitrocellulose and nitroguanidine as a

triple base propellant, which are used in larger calibre projectiles.

Furthermore, it has been used as ingredient of the solid propellant in

rockets.

Nitroglycerine has very high impact sensitivity and thermal instability. Thus,

many accidents have occurred during its manufacture and transport.

During the period from 1769 to 1980 among worldwide industrial accidents,

121 involved nitroglycerine which is approximately 24.2% of the total.

,showing that there were 43 industrial accidents involving nitroglycerine in

the period from 1954 to 1982. Obviously, the manufacturing process of

nitroglycerine is more dangerous than other chemical processes.

The manufacturing technique for nitroglycerine includes both batch and

continuous processes. Typical examples of the batch method are the Nobel

process from 1862 and the Nathan, Thomson and Rintoul process from

1908. A continuous method was first developed by the Schmid process in

1927 followed by the Biazzi process in 1935. Subsequently Nilssen and

Brunnberg developed a new injection nitration process in 1950. In 1978 the

Hercules Company in U.S.A. designed another new tubular process.

Although the manufacture of nitroglycerine has been developed for over 140

years, the dangerous industrial procedures involved have almost remained

the same. These procedures include preparing of mixed acid, nitration of

glycerine, separation of product and wastes, washing of product, filtration of

waste, recycling of nitroglycerine from waste acid and treating of waste acid.

Most of industrial accidents have occurred during the glycerine nitration

stage or the separation stage (approximately 50.4%). Therefore,

improvements in manufacturing procedures and the development of safety

control techniques are important requirements for this reaction. The kinetic

245


parameters of the nitroglycerine reaction and the operating conditions for its

industrial manufacture are not found in the open literature, presumably

because of its military importance.

3. Evaluation of Kinetic Parameters of Glycerine–nitric Acid

Reaction System

The elemental reaction of producing nitroglycerine is esterification reaction

of glycerine and nitric acid. This chemical reaction can be expressed as:

C3H5(OH)3 + 3HNO3 → 42SOH C3H5(ONO2)3 + 3H2O equation(1)

In this chemical reaction, the mixed feed acid contains both HNO3 and

H2SO4. The reagent H2SO4 acts as a dehydrating agent. Therefore, only the

HNO3 participates in this reaction. The chemical consumption can be

expressed by

CG=CG,0(1−xG) equation(2)

and

CN=CN,0−3CG,0xG=CG,0(M−3xG) equation(3)

where CG and CN denote the concentration of glycerine and nitric acid,

respectively. xG represents the fractional conversion of glycerine and M is the

initial concentration ratio of CN,0 to CG,0. Assuming the reaction is n order in

the glycerine and m order in the nitric acid and that the reaction rate

constant can be expressed as Arrhenius form, where A and E denote the

frequency factor and activation energy, respectively. Then, the reaction rate

of glycerine and nitric acid can be expressed as

( )( )mGGmn

0,GRTEm

NnG

RTEmN

nGG x3Mx1CeACCeACkCr −−===− +−−

equation (4)

246


4. History of AN Fires and Explosions

The accidents tabulated in Table 1 involve a wide variety of circumstances,

locales, and modes of storage or transport. But they share one essential trait

in common: 100% of the explosions were the outcome of an uncontrollable

fire. It does not matter if the fire originated in a warehouse, a truck, a

railcar, or a ship—no explosions and no loss of life occurred unless there

was a fire, and the fire was not controlled.

Table 1. Incidents involving AN storage or transport facilities, not

including incidents involving ANFO, nor ones with involvement of

explosives other than AN. Please also see the nature of the accidents

(from 2000 till date).

Date Site Incident Fire Expl-osion Dead

28-June-

2000

Duette, FL Fire and collision of

trucks carrying AN and

gasoline

√

February-

2003

USA Warehouse fire storing

AN

√

2003 USA AN fire in a farm supply

store, poss. minor

√

2-October-

2003

Saint-

Romainen

-Jarez,

France

AN fire in a small

agricultural storage

facility leading to a

detonation

√ √

October-

2003

UK Fire involving AN in a

small farm store

√

18-

February-

2004

Neyshabur

, Iran

Major train derailment,

caused a fire to ignite;

AN burned and exploded

√ √ 328

247



9-March-

2004

Barracas,

Spain

Truck collision, leading

to fire and AN

detonation

√ √ 2

22-April-

2004

Ryongcho

n, North

Korea

Train carrying AN

collided with truck

carrying oil; resulted in

fire and explosion

√ √ 161

24-May-

2004

Mihăileşti,

Romania

Truck carrying bagged

AN overturned, fire

resulted, then

detonation

√ √ 18

August-

2004

USA Fire in a dry bulk

blending and

distribution center that

held AN

√

6-March-

2007

Pernik,

Bulgaria

Truck carrying AN

caught fire and exploded

√ √

June-2006 France Fire on a truck hauling

bagged AN; extinguished

by fire service

√

22-Sept-

2006

Suisun

City, CA

Fire involved E.B. Stone

& Son facility as an

exposure from a grass

fire

√

30-July-

2009

Bryan, TX Fire at El Dorado

Chemical Co. started by

welder in warehouse,

burned down warehouse

√

8-August- Mt. Isa, Truck carrying AN √

248


2009 Australia caught fire

21-January-

2012

Mariveles,

Philippine

s

Fire destroyed an AN

warehouse at a shipyard

√

17-April-

2013

West, TX The subject fire √ √ 15

29-May-

2014

Athens,

TX

Fire at East Texas Ag

Supply warehouse

involving AN; no

detonation, building

burned up

√

5-Sept-2014

11

Sept, 2015

Charleville

, Australia

Petlawad,

Jhabua,

Madhya

Pradesh,

India

A double-trailer truck

carrying 52 tonnes of AN

overturned, ignited, and

exploded

Stockpile of gelatin rods,

urea and other

chemicals, in public

space

√

√

√

√

100

5. Causes of AN Explosion

5.1 Role of Impurities

Under certain conditions (e.g., presence of impurities), AN may behave as a

strong oxidizing agent that can undergo detonation .Several AN-related fires

and explosion, such as the Texas City disaster in 1947 and the Toulouse AN

explosion in France in 2001 , also killed people and caused property

damage. Contamination with other materials was believed to be one of the

contributing factors to these incidents. Though the cause of the Toulouse

disaster is still unclear, one potential reason is that AN contaminated with

249


chloride (i.e., sodium dichloroisocyanurate (DCCNa)) caused the explosion.

Laboratory-scale experiment showed that AN–DCCNa mixture has higher

decomposition rates than pure AN [6]. In another research, it is reported

that moistened AN–DCCNa mixture could produce NCl3, leading to the

detonation of the mixture in the applied laboratory-scale.

AN decomposition with a single additive has been widely studied in

literature. In literature, binary mixtures with AN have been studied mainly

as explosives. Researchers mixed AN with more than one additives, such as

the mixture of ammonium sulfate, calcium carbonate, and urea. Low

thermal stability is indeed related to high explosivity, but high thermal

stability is no guarantee of low explosivity. AN was mixed with potassium

nitrate and complex one salts in order to phase-stabilize AN as an oxidant.

In other publications, AN mixture with binary additives has also been

reported- the thermodynamic effect of AN decomposition when mixed with

Ca and Mg carbonates with or without the presence of boron, manganese,

and copper compounds, such as AN mixture with CaCO3 and CuSO4; CaCO3

and H3BO3; CaCO3 and MnO2; etc. They found that equilibrium

concentrations does not depend on the carbonate origin of CaCO3, MgCO3,

or CaMg(CO3)2. In another research, AN was mixed with additives such as

FeS and urea, FeS2 and NaF, etc. Their conclusion was that salts of weak

acids and urea stabilized AN formulations, even in the presence of

destabilizing species.

The thermal decomposition of AN has been studied by various types of

calorimeters, typically employing sample sizes of the order of a few

milligrams and the data reported by different researchers have noticeable

differences. The AN decomposition has been reported to start at various

temperatures such as 210C , 200C , and 190C. The detected “onset”

temperatures of decomposition depend on the performance of the

calorimeter. It must be emphasized that there is not really an “onset”

temperature. According to Arrhenius law, reaction rates have an exponential

250

http://www.sciencedirect.com/science/article/pii/S004060311500492X%23bib0030


dependency on temperature, reaction rates at low temperatures are very

slow [16], and our ability to “detect” when a reaction “starts” depends on the

sensitivity of the employed experimental equipment. The so-called “onset”

temperature only marks the temperature at which the thermal effects

caused by the reaction become detectable by the employed equipment. Due

to different levels of accuracy of different equipment as well as testing

methods, the detected “onset” temperatures of AN decomposition vary.

Different macroscopic AN decomposition paths have also been proposed and

reported. The paths that attract more attention are listed below , in reaction

(1) and (2). It needs to be pointed out that the products of AN decomposition

are not limited to NH3, HNO3, N2O, and H2O. Several other gaseous products

can also be generated, including N2, O2, NO, and NO2.

NH4NO3 ⇌ HNO3 + NH3, ΔH 176 kJ mol−1 (at 170°C) equation(1)

NH4NO3 → N2O + 2H2O, ΔH −59 kJ mol−1 (above 170 °C) equation(2)

The generally accepted mechanism of AN decomposition is that the

dissociation of HNO3 leads to the subsequent oxidation of NH3, e.g., Rosser

et al. proposed reaction as the dissociation reaction of HNO3, which

generates NO2+, acting as the oxidizing species for NH3 as listed in reaction

(4). Reaction (5) shows a more realistic reaction path of HNO3

decomposition.

equation (3)

equation (4) equation (5)

To explain reaction (3) and (4) in more detail, in the presence of water, using

“acid” to indicate , H3O+, or HNO3, the following decomposition

mechanism equations (6)–(8)have been proposed, where reaction (7) is

considered the controlling step due to its slow reaction rate.

251


http://www.sciencedirect.com/science/article/pii/S004060311500492X%23eq0005










equation(6)

equation(7)

equation(8)

Reaction (7) is also described in terms of elementary reactions in the

literature [19] at the temperature range 342–387°C, where subsequently

oxidizes NH3, as listed in reactions (9)–(14). Reaction (15) is the overall

stoichiometry according to this theory.

equation(9)

NH2 + NO2 → NH + HNO2 equation(10)

NH + NO2 → HNO + NO equation(11)

NH2 + NO → N2 + H2O equation(12)

2HNO → N2O + H2O equation(13)

2HNO2 → NO2 + H2O + NO equation(14)

4NH3 + 5NO2 → N2O + 2N2 + 6H2O + 3NO equation(15)

Slightly different from the previously mentioned mechanism, another

approach [20]assumes that the formation of a nitramide intermediate out of

AN results in the decomposition of AN, as listed in reactions (16)–(20).

equation(16)

HONO2 → HO + NO2 equation(17)

HO + NH3 → HOH + NH2 equation(18)

NH2 + NO2 → NH2NO2 equation(19)

NH2NO2 → N2O + H2O equation(20)

252










5.2. The oxidizing Properties of AN

In broad terms, AN disasters have two causative factors due to the

chemistry of the substance: (1) its ability to detonate; and (2) its ability to

enhance a fire by means of its oxidizing behavior. Thus, the nature of its

oxidizing ability must be considered.

AN is not a fuel and cannot burn of itself. But since it is an oxidizing

chemical, it can promote combustion reactions of fuels. As an oxidant, it is a

weak one and is classified in the lowest class, Class 1, of oxidizers as viewed

by NFPA 400. Unlike some stronger oxidizers such as calcium hypochlorite,

there are no normally encountered products with which it combines to

produce a hypergolic ignition at room temperature. Unless quantities of AN

so massive were involved that self-heating would become an issue, for an AN

+ fuel reaction to occur, application of heat is necessary. The reactions

thereby occurring are complicated and not entirely unraveled, but it is likely

that evolution of HNO3 dominates, which is effective in igniting some fuels by

contact. Molten AN can ignite carbonaceous fuel by contact.

AN will show accelerated thermal decomposition reactions when combined

with a long list of contaminants. Many of these, e.g., chromium compounds,

are less likely to be encountered in actual accident scenarios, but chlorides

and copper can more readily be expected to be present; zinc can also be

important since it is contained in galvanizing coatings for steel. In some

cases, there can be a combined effect of organic material and chlorides or

copper reacting with AN.

Findlay and Rosebourne published the first study on the oxidizing properties

of AN in 1922. They mixed AN with ‘wood meal,’ i.e., fine sawdust, placed it

in a test tube and heated it to 100 °C. They also repeated the experiment

using powdered starch as the cellulosic fuel. In both cases, they got

evolution of N2, CO2, and H2O, but no CO or oxides of nitrogen.

Kistiakowsky and Guinn conducted oven-heating tests on AN mixed with

253


paper, and found that ignition occurred when a temperature of 132–154 °C

was reached. Macy et al. determined the ignition temperature of AN-

saturated paper bags to be 177–232 °C, but a more intimate mixture of

sawdust or paper with AN will ignite at 150 °C. All of these results are below

the normal ignition temperature of paper, ca. 250 °C, confirming the

oxidizing properties of AN. The important safety implication is that molten

AN at 169.6°C or higher is likely to greatly accelerate spread of fire by direct

ignition on contact, as it flows.

Despite the fact that chemists have been studying oxidizing chemical

properties for a century or more, no standardized test emerged for assessing

the oxidizing potential of chemicals that has engineering meaningfulness. In

fact, there even have been very few attempts to develop a standardized test,

and no attempt has resulted in a test which would be provide usable

engineering data, i.e., data that can be used as input to make some

calculation. Instead, all of the research has only produced schemes to

compare one substance against another on some arbitrary scale. The

earliest studies aimed at producing a standard test were conducted at BM,

who proposed a horizontal trough test. The oxidizer is to be mixed with wood

sawdust, ignited by a propane flame at one end, and the rate of fire

propagation timed. Materials showing a faster rate of progress are deemed to

be more hazardous. BM proposed classifying the results into 4 categories

already used by NFPA. Researchers at General Electric then conducted two

follow-up studies, where they endeavored to achieve better reproducibility

and also tested additional materials. In the course of this, they confirmed

AN to be an oxidizer of Class 1, in terms of the NFPA classification scheme.

It should be noted that classification of most AN-based fertilizer products is

not based on conducting the O.1 test. Instead, product classification is

accomplished by means of Special Provisions (SPs), which are based on the

chemical constituency of the product. Only products which do not fall under

the pertinent SPs would be classified by means of testing.

254


6. Learning from Disasters

Viewed at a level of fine-enough granularity, every disaster is unique. Thus,

this point of view is singularly unhelpful towards promoting of safety and

preventing of future disasters. Instead, a prudent risk management

philosophy has to entail examining past disasters and identifying what is

common among them, not what is unique and different[94]. For example,

after the S.S. High Flyer exploded, the Government panel focused on the fact

that there was nearby stowage of sulfur (a unique factor) instead of focusing

on the existence of uncontrolled fire (the common factor). As a result, the

necessary lesson was not learned. But by now, nearly 70 years later, the

record is so lengthy and so diverse, that hopefully the overwhelming

importance of the common factor—uncontrolled fire—in AN detonations can

be appreciated. This perspective allows one to see that there is nothing

substantively different between transportation and storage disasters. In each

case, a fuel is present, an ignition occurs, a fire burns fiercely, aided by the

oxidizing nature of AN, and effective fire control fails to be achieved. The

outcome then, at best, is a physical disaster without dead persons or, at

worst, a large life loss disaster. Conversely, not a single AN explosion

accident exists in a storage or transportation environment where there was

no fire, or where there was a small fire which was successfully controlled.

It is especially troubling that the fertilizer manufacturing industry, while

aware of the immense hazards of AN, persistently attempted to minimize the

implications of known accidents, instead of learning from them. For

example, the representatives of a large AN manufacturer in the UK, who

clearly knew the literature very well, wrote “the conditions of pressure…and

temperature required to explode ammonium nitrate are extreme…such severe

conditions are very unlikely to be met.” And, “It is generally believed that a

fire involving an unconfined mass of ammonium nitrate will not lead to an

explosion.” And, “… shock initiation of solid material is discounted as a

serious risk.” The attitude of government authorities, at least in the US, has

255



been no better. The safety provisions for dangerous substances in workplace

is handled in the US primarily by the Occupational Safety and Health

Administration (OSHA), whose regulations, as shown below, have been so

ineffectual as to be castigated by other Federal government agencies.

One prior incident is specifically noteworthy. In 2003, a French farm barn

containing stored AN suffered an fire which was not successfully controlled.

This led to detonation and widespread destructions, but fortunately no

fatalities. The West, Texas incident, 10 years later, was very similar to the

French accident, but with a disturbing life loss outcome. The details of the

French accident were documented both in a technical article and in an ARIA

report, yet the lesson went unheeded.

7. Preventing Uncontrollable Fires

There are two main techniques that can be used to prevent uncontrollable

AN fires: (1) reduce the fire hazard propensity of AN; and (2) provide

constructional fire safety measures for the facility. Both should be

implemented for assuring safety in storage environments. But it is generally

impossible to implement safety measures with regards to trucks or other

transportation environments which would reduce or eliminate potential fuels

for fire. Thus, since the second technique is not available for guarding

against transportation disasters, it is crucial that the first technique be

implemented across-the-board.

7.1. Reducing the Hazards of AN

Safety improvements in the chemical industry often follow the precepts of

Inherently Safer Technology (IST). The relevant IST principle here

is substitution—replacing a hazardous material with a safer option. The IST

‘substitution’ principle is also known as alternative safer design in product

defect literature. As will be shown below, alternatives for AN which

256


constitute safety improvement in regards to both fire safety and explosibility

exist, and some have existed for many decades.

Straight-AN, that is AN where the only added chemicals are a small amount

of coating chemicals used to reduce caking tendency, is susceptible to

disastrous detonation accidents. Already before 1920, the dangers of this

were recognized and the BASF factory at Oppau4 produced ANS (a double

salt of ammonium sulfate and AN) since straight AN was recognized to be

“too dangerous to handle for fertilizer purposes.” While this particular

product is indeed less susceptible to detonation than straight AN, additional

alternatives were subsequently developed, and the patent literature contains

a large number of such inventions. Here, three such products are

considered: (1) CAN, (2) Ferti-Safe, and (3) Sulf-N 26.

Calcium ammonium nitrate, CAN, has been made since the 1920s [19]. It

involves adding powdered calcium carbonate into the AN at the hot liquor

stage, in an amount of 20% (currently) to 40% (prior to the 1980s). CAN has

been described as “a non-detonable alternative to pure ammonium nitrate.”

This overstates its nature, but it is appropriate to view that “calcium

ammonium nitrate essentially removes any explosion hazard. ” Compared to

AN, CAN is (a) less likely to lead to uncontrollable fire, since it is a less

potent oxidizer; and (b) is both much less sensitive towards detonation, and

exhibits greatly lowered explosive power if boostered enough to detonate.

CAN, or alternative safety enhanced products, are utilized to the exclusion of

straight-AN fertilizer in a number of European countries , which includes

Belgium, Denmark, Eire, Finland, Germany, The Netherlands, Northern

Ireland, and Switzerland. This is directly due to the safety characteristics of

CAN versus AN.

With regards to the effect on fire of CAN, one of the earliest demonstrations

of the safety improvement that can be achieved by CAN were the pressure-

vessel experiments, discussed above. They concluded that the explosion

temperature is a useful criterion for examining the negative effects of

257

http://www.sciencedirect.com/science/article/pii/S0304389415301680%23fn0020



contaminants and the positive effects of inerting additives. Adding 0.001% of

chlorine lowered the explosion temperature of AN from 290°C to 260°C,

while 0.1% lowered it to 240°C. Similarly, adding 1% of paraffin wax

dropped the temperature to 240°C. But a notably positive effect on AN

stability was obtained by adding calcium carbonate (Fig. 4). I can be

concluded that, of substances examined, calcium carbonate was the most

beneficial.

Fig. 4. Effect of adding calcium carbonate to AN on raising its explosion

temperature. Hainer conducted fire tests in moderately large scale by burning 136 kg of

wax-coated AN prills in a 55-gallon drum, stoked with charcoal and wood

and ignited with 10 kg of thermite. He then added 45 kg of clay-coated AN

prills and more thermite and got a violent fire, albeit not an explosion. He

judged that such a fire would not be extinguishable. However, he then added

68 kg of CAN, and this actually controlled the fire to the extent that even

adding more thermite failed to make it severe or difficult to extinguish. This

demonstration clearly showed that CAN not only does not promote burning,

but acts to suppress burning.

258



With regards to detonation properties of CAN, Shaffer conducted sensitivity

tests on AN compounds to determine how much booster is needed for

detonating a tube filled with the test material. In these tests, he found that

commercial CAN specimens with AN content of 81% showed a detonation

sensitivity of 22%, as compared to straight AN. With sufficient boostering,

CAN can be detonated, even though much greater boostering would be

required. But even under that unlikely scenario, the outcome would be

notably less severe. The test results for the explosive power of CAN,

compared to AN and some other diluents, based on ballistic mortar tests are

shown (Fig. 5). This shows that by incorporating 20% of calcium carbonate,

the explosive power of AN can be reduced by more than 90%.

Fig. 5. The effect in the Mark IIIc RARDE ballistic mortar of AN with

various additives (Clancey and Turner), with results being expressed as

a percent of the effect for pure picric acid. Pure AN shows 80% the

energy value of picric acid. ANS (blend) denotes a mechanical blend of

ammonium nitrate and ammonium sulfate, while ANS (dbl. salt) denotes

a 1:1 molar ratio double salt.

259



Military authorities sometimes propose to take actions against improvised

explosive devices (IEDs), which are often based on AN. Thus, there have

been campaigns [119] to discover modifications to AN which would make it

non-explosive and ‘irrestorable.’ The latter concept means that the modified

substance could not be re-manufactured to yield an explosive material.

Such efforts to create a non-explosive and irrestorable version of AN have all

come to naught. But it is exceedingly important to realize that this campaign

is solely a security campaign and unrelated to safety concerns of this

chapter. To reduce the potential for accidental detonations involving AN, it is

not necessary to evolve a substance which is non-explosive and irrestorable.

What is needed is a formulation with reduced oxidizing potency, reduced

detonation sensitivity, and reduced explosive power. Unlike the objectives of

military authorities, these objectives are demonstrably practical and

feasible.

7.2. Making Buldings fire-safe

Safety improvements in the AN-based product may be the only strategy

available for improving transport safety. But for storage buildings, the

buildings themselves can be constructed so as to preclude the possibility of

an uncontrolled fire, and this should be done as an additional layer of

protection. A short list of crucial fire safety features would suffice to prevent

AN disasters, such as occurred at West. While the Texas facility was only

one of more than 1300 such in the US, it is useful to focus on its

shortcomings, since many agricultural AN storage facilities are built in a

similar manner.

AN storage bins should be made of noncombustible, non-

reactivematerials. Wood or galvanized metal should not be used. It has

been known since at least 1936 [121]that wood incontact with AN can get

saturated with the material and then, when ignited, will burn “ witha

much stronger flame than non-saturated wood.” But wood bins would be

260




an unacceptable source of fuel in proximity to stored AN, even if some

surface treatment were adopted which reduces direct impregnation, since

the fuel value of the wood is not reduced thereby.

In 1947, as a result of the tragedy in Texas City, the US Federal

government [80]concluded that “ The most common hazard to ammonium

nitrate fertilizer…is fire involving combustible containers or adjacent

combustible material that may be present…” Unfortunately, many decades

later, matters had not changed, leading directly to the West tragedy. The

AN was stored there in wooden bins, comprising a large fuel load in direct

contact with the AN.

The building construction should be of non-combustible materials except

for very small components for trim, wiring, or other purposes where

noncombustible alternatives do not exist.

The building wherein AN was stored was of ordinary, wood-framed

construction, with much exposed timber not even protected by plaster or

gypsum wallboard. Ordinary timber construction can be ignited from very

small ignition sources (e.g., electrical faults), while the large amount of

timber present serves to create a long-burning fire. It is highly unlikely

that the fire and detonation at West would have taken place, had the bins

and building materials not been combustible.

Any AN storage facility should be equipped with smoke or heat alarm

system monitored by a central station.

The fire occurred about 2 h after close of business for the day. The fire

department was notified only when a person in a nearby city park

observed smoke. By the time the fire department arrived, they faced a

fully involved fire. Had there been a fire or smoke alarm system

monitored by a central station, notification would have come to the fire

261



department much earlier, and it is likely that the fire would have been

suppressed successfully, rather than overwhelming the firefighters.

Any AN storage facility should be protected by an automatic fire sprinkler

system, properly designed for this purpose.

Apart from fire extinguishers, which are of no benefit to an unattended

facility (nor if the fire grows beyond a very early stage), there was no fire

suppression system. Had there been a properly designed fire sprinkler

system, this would have controlled the fire and, upon responding to the

fire, the firefighters would have encountered a suppressed, or nearly so,

fire.

Floors in an AN storage facility should be constructed so that there are

no elevator pits, drains, grates, or pipes to where molten AN could flow

into and accumulate.

At the West Fertilizer facility, the elevator contained a pit about 3 m deep

into which molten AN could flow. It is most probable that the actual

explosion originated at this pit. With a floor design that does not allow

molten material to be trapped, the molten AN would have been less prone

to detonation due to absence of a critical depth of liquid material.

To be effective, all of the above provisions need to be mandatory and not

just in the form of advisory suggestions.

As explained below, no mandatory safety requirements were laid upon

the owners of the West Fertilizer Co., beyond OSHA regulations, and

these did not include any of the above fire safety provisions.

8. The role of codes and suppliers

8.1. Fire safety measures for storage environments

262


Staff of agricultural fertilizer distribution companies are normally

knowledgeable only on matters of agronomy and do not have technical

knowledge on the fire or explosion safety of chemicals. Consequently, the

only way that agricultural fertilizer distribution facilities can be provided

with adequate fire safety is if there are mandatory requirements laid down

and enforced by technically competent entities, of which there are two: (1)

government authorities, and (2) AN manufacturers. Disasters such as the

one at West, Texas occur when neither takes effective action. In the case of

manufacturers, they actually have a dual role—to provide for safety

improvements in the product itself, discussed above, and to require safe

facilities at their customers. In a number of sectors of the chemical industry,

raw chemical suppliers have Product Stewardship procedures in place so

that they will not ship product to any facility that they have not inspected

and approved. The US suppliers of fertilizers had chosen not to implement

such requirements. Also, in West, as in much of other rural parts of Texas,

there exists no building code and no fire code. This is true of many other

locales in rural US, where codes are either absent or there are no local

authorities competent to inspect and control hazardous chemical facilities.

But even a competent authority will not be able to achieve adequate safety

unless there exists a suitable document for them to enforce. Thus, it is

important to examine some of the salient documents concerned with AN

safety.

Various codes and guidance documents often provide a plethora of

provisions regarding storage of AN, which tends to create the impression

that effective fire safety measures have been adopted, even when the

provisions adopted are marginal, obscure, or do not address the primary fire

safety needs. The requirements laid down by some of the relevant

documents are described in the sections below and summarized in Table 3.

It is of utmost importance to realize that adherence by the owners of the

West Fertilizer Co. to any of the documents discussed would not have

averted the disaster. The FM Global data sheet is the only one that contains

263



even a reasonable fraction of the pertinent safety features, but this is a

purely advisory document. Only if there were a means of making its

provisions mandatory might raising of the fire safety level occur. But this

could have been accomplished had the manufacturers of the AN made such

compliance a sales condition for supplying AN.

Table 3. Presence or absence of essential fire safety features for some

codes, regulations, and guidance documents (summary evaluation;

details are provided in connection with each institution below).

Safety measure OSHA NFPA

400

FM

Global TFI

W.

A.

Product with combustion

and detonation resistant

behavior

× × × × ×

Non-combustible bins × × ? × √

Non-combustible building

construction ×a × √ × √

Monitored alarm system × × × √ ×

Automatic fire sprinkler

system √b × √c × ×

Absence of places where

molten AN could pool √ √d √ × √

Mandatory language √ √e × × ×

1. In some cases, fire-resistive walls may be required, but these do not have to be

non-combustible; also, nonconforming facilities can be allowed.

2. Only for facilities storing in excess of 2500 ton.

3. Only for portions of the facility containing combustible components.

4. Provision exists, but ineffective, since scope does not include pits.

5. Many of the provisions turn out to be non-mandatory, since they are couched in

264

http://www.sciencedirect.com/science/article/pii/S0304389415301680%23tblfn0005






terms of subjective decisions by the AHJ, and especially allowing the AHJ to

permit nonconforming structures to continue being used.

8.1.1. OSHA regulations

In the US, section 29CFR 1910.109(i) of OSHA regulations include

provisions for the storage of explosives or blasting agents comprising AN.

However, these safety features are weak and compliance with them would

not have prevented the West disaster. Notably, wood construction is not

prohibited, while sprinkler protection is required only for facilities storing

over 2500 tons of AN (Table 3). The West facility, at the time of the

explosion, held 40–60 tons of AN, not counting the 100 tons in the parked

railcar which did not get involved. Furthermore, any reasonable reading of

the regulation indicates that it applies only to entities manufacturing or

storing explosives, but not to distributors of agricultural fertilizer. After the

West disaster, OSHA responded with a self-serving letter claiming that such

fertilizer storage facilities were subject to the above-cited section. However,

the whole section, not just subsection (i), of the regulation is

titled Explosives and blasting agents, with its scope being specified as: “ This

section applies to the manufacture, keeping, having, storage, sale,

transportation, and use of explosives, blasting agents, and pyrotechnics”

while fertilizer-grade AN is not classified as by any governmental entity as

being an explosive, blasting agent, or pyrotechnic . Two different, but non-

regulatory, Federal agencies documented the ineffectiveness of OSHA

towards establishing viable fire and explosion safety in agricultural AN

storage facilities. It is also of interest that OSHA inspected the West facility

in 1985 and cited it only for minor violations unrelated to the AN storage

building. Yet, the wooden building, the wood bins, the elevator pit, and other

relevant hazards were the same at the time of the inspection as in 2013.

8.1.2. NFPA 400 Hazardous Materials Code

265



The best-known document in the US concerned with the safety of AN

storage is the NFPA Hazardous Materials Code (NFPA 400 [82]), which is

considered here in its pre-West-disaster (2013) edition. It is published by

a private organization, but assumes the force of law in many US

jurisdictions since it is directly referred to by most US fire codes.

Unfortunately, this is a document which is inordinately opaque and

confusing.5 In addition, it is unreasonably wordy, being some 214 pages

long.6While technical knowledge necessary to adequately describe safe

storage of AN is not trivial, it unreasonable to expect even a chemist or

an engineer to read and comprehend a 214-page treatise in order to

arrange for safe storage of AN. An even more severe shortcoming of NFPA

400 is that it relies extremely heavily on subjective assessments and

approvals or disapprovals from the pertinent Authority Having

Jurisdiction (AHJ). This creates a lack of rigor and objectivity. But in a

locale which lacks a fire code and an AHJ, it makes the concept of AHJ-

dependent requirements illogical. An additional weakness is the

‘grandfathering’ of existing facilities. Sec. 11.5.5 specifies that “the

continued use of an existing storage building or structure not in strict

conformity with this code shall be approved by the AHJ in cases where

such continued use will not constitute a hazard to life or adjoining

property.” In a locale with has no AHJ, this implies that no restrictions

can be made with regards to existing buildings.

The NFPA Code does not require that bins be made of a non-combustible

materials, instead of being made from wood. The only bin materials

prohibited (Par. 11.3.2.3.3.2) are ones made of galvanized iron, copper,

lead, or zinc. Par. 11.3.2.3.3.3. states that “wooden bins protected

against impregnation by ammonium nitrate shall be permitted” but does

not prohibit wooden bins of any other kind. Furthermore, the concept of

“protected against impregnation” is meaningless since there does not

exist any test for this, nor are any instructions or specifications given in

the NFPA Code for establishing what constitutes protection.

266





The NFPA Code does not require that any other combustible

construction features of the building itself be eliminated. Par. 11.1.5

requires that the building “ be constructed in accordance with the

building code,” but this is meaningless for a locale that has no building

code. Furthermore, building code provisions are written to govern the

construction and not the operation of buildings. In other words, apart

from some optional additional provisions for existing buildings that

localities can include by local ordinance, any building already in

existence when a building code is adopted does not fall under the scope

of its regulation.

The NFPA Code has no requirement that an alarm system be installed.

There is a placeholder in Sec. 11.2.10, but no requirements or text.

The NFPA Code does not require that an automatic fire sprinkler system

be installed in the facility. Sec. 11.2.6 specifies that sprinkler

requirements are laid down in Sec. 6.2.1.1, but the latter only requires

sprinklers for chemicals which fall into Protection Levels 1 through 4,

while AN is outside the scope of Protection Levels, per Table 5.2.1.1.3. In

fact, the wording of the Code actively discourages installation of

sprinkler systems for AN: “11.2.6.1.2 Sprinkler protection shall be

permitted to be required by the AHJ for the storage of less than 2500 ton

(2268 metric tons) of ammonium nitrate where the location of the building

or the presence of other stored materials can present a special hazard.”

The provisions in the NFPA Code against pooling of molten AN are poor,

at best. Sec. 11.3.3.2 restricts drains, pockets, etc., within “flooring in

storage and handling areas.” The straightforward architectural

interpretation of these requirements is that they literally only prohibit

drains, gutters, trenches and channels coming down below the floor

level; an elevator pit such as was present at the West facility might be

267


considered simply a change in floor level, not an appurtenance

extending below the floor.

After the West disaster, an NFPA spokesperson admitted that NFPA is

still “learning more about ammonium nitrate” and proceeded to revise

AN provisions for the 2016 edition. An analysis [126] of this forthcoming

edition indicates that it is still unlikely to prevent AN disasters, despite

some improvements.

8.1.3. FM data sheet Instark contrast to NFPA 400, the FM Data Sheet 7–89 [127] is only 12

pages long, is written in clear, non-convoluted English, and lays out fire

safety provisions in readily comprehensible terms. Its main drawbacks are

that, being published by an insurance company, it is necessarily only

advisory, and that it includes some, but not all of the safety provisions

discussed above. It does not establish requirements for products safer than

straight AN, nor a monitored alarm system. However, it specifies non-

combustible building construction, and clearly prohibits pits where molten

AN could pool. An automatic fire sprinkler system is required anywhere that

combustible items might be installed. In the case of an agricultural fertilizer

facility, it is generally difficult to provide non-combustible elevators and

conveyor belts; thus, these would require sprinkler protection. It is not clear

if combustible bin materials are prohibited, but even if they are allowed, this

would have triggered a sprinkler requirement. If the West facility had been

rebuilt in non-combustible construction, had sprinklers provided for

combustible components, and had its elevator pit eliminated, it is highly

unlikely that a severe fire could have developed that ended as a detonation.

8.1.4. TFI Guidelines

In response to the West disaster, The Fertilizer Institute (TFI) issued in 2014

“Safety and Security Guidelines” for AN [128]. But this is a shockingly

268





inadequate document. There are basically only three provisions addressed

directly at the construction safety of storage buildings:

(1) Steel or wood bins should be protected by a coating such as sodium

silicate, epoxy, or PVC;

(2) If not continuously occupied, the building should have an automatic

detection and alarm system; and

(3) AN piles should not exceed a height of 40 ft (12.2 m).

Had these guidelines been in effect on 17 April 2013, nothing would have

changed. First and foremost, the document is presented in non-mandatory

language, e.g., “should,” not “shall”. Thus, any operator of an AN storage

facility would be entitled to conclude that compliance is optional and that

there is no burden to comply. But even if the wording were changed to

mandatory language, the safety level would be grossly inadequate. It may be

noted that research has shown AN piles over 1.5 m high are susceptible to

detonation, so limiting height to 12.2 m (much taller than at West, Texas) is

ineffectual. Since the consequences of an AN explosion can be so dire, it

should be clear that a strategy of protection-in-depth is necessary. In other

words, not only must all pertinent aspects of fire safety be included, but

there must be some redundancy, so that more than one layer of protection

would be in place to guard against deaths of exposed population. TFI

document, on the other hand, not only lacks any useful redundancy, but

fails to require even the most elementary safety precaution, noncombustible

construction.

8.1.5. Western Australia code of practice

An instructive contrast is the AN safety document issued by the government

of Western Australia (W.A.) in 2012 [130]. This notably predated the West

disaster, yet the treatment of constructional safety issues is dramatically

superior to the post-West TFI document. The document is also issued in a

269



non-regulatory context and thus is presented in “should” language. But at

least it includes a number of important safety issues:

(1) The entire building should be built of non-combustible materials;

(2) A clear explanation is given that drains pipes and tunnels should not

exist within which molten AN could accumulate;

(3) If there is any other facility within the AN storage building, a 5m clear

spacing should be established between the AN storage facility and

other uses within the building. Much greater distances are to be

provided for storage quantities in excess of 10 ton.

(4) Bins should be made only of non-combustible materials.

In addition, the document makes clear that chlorides are incompatible with

AN. This speaks directly to the issue at West where the same conveyor

system was used to transport KCl and AN, necessarily leading to

contamination of AN with KCl, since no cleaning regimen existed.

9. Manufacturers' Role

In the absence of effective government regulations, manufacturers of the

material can take active steps to ensure safety in transport and storage

using two tools: (1) a safety data sheet (SDS) for the material which contains

necessary prescriptions for safe transport, handing, and storage of the

product; and (2) a buyer qualification program, and these tools are best

used in concert. In the US, the manufacturers of AN have not taken either

task seriously. Even after being revised subsequent to the 2013 disaster, the

safety data sheets of the manufacturers remain totally inadequate. They lack

prescriptive language, and simply refer the user to NFPA 400 for actual fire

and explosion safety recommendations. Furthermore, the intention of a SDS

is defeated by such an approach. A useful SDS should be self-contained

270


with regards to safety requirements, and authors should not evade their

responsibility by simply suggesting that the user consult another document.

The manufacturers involved, however, could have prevented the West, Texas

disaster (and disasters which may occur in the future) by including in the

SDS the safety issues identified above, and doing this in a prescriptive and

clear language. This would need to be accompanied by an effective program

to inspect the buyer's facilities to ensure that they comply with the fire

safety requirements laid down in a suitably effective SDS. It is noteworthy

that in the case of the West, Texas disaster, employees testified that they

would have complied with any constructional safety requirements that they

would have been required to observe.

10. Conclusions Explosions of stored ammonium nitrate are rare events. But they are

recurrent, catastrophic, and preventable. The reason they have not been

prevented is because persons charged with safety responsibilities focused on

the differences characterizing each disaster, instead of examining what is

common among the disasters. The recent explosion killing 15 persons in

West, Texas, is but the latest disaster in a century's worth of AN explosions.

By examining this long record, it emerges that the common factor is an

uncontrollable fire. This is true irrespective of whether the material is stored

in a warehouse, or on a transportation conveyance. In the database

compiled, it is found that 100% of AN explosions in storage or transportation

environments (excluding manufacturing accidents or accidents involving

other explosives) have been caused by uncontrolled fire. Conversely, 30% of

uncontrolled fires involving AN have resulted in explosions, and about half

of these resulted in deaths.

There are some two centuries' worth of chemistry studies on the

characteristics of AN. These are helpful towards understanding AN behavior,

but do not enable reconstructions of accidents to be made. Explosives

271


researchers understand how AN can be detonated when a detonator and a

booster charge are used. But only some unvalidated theories exist to explain

how AN detonates in fires where neither of these is present. A handful of

researchers also tried to re-create such fire-induced detonation events, but

every one of them was unsuccessful. Yet, the tragic record of recurring

explosions due to fires demonstrate that the phenomenon is not only real,

but is a major hazard.

To avoid a second “century's worth of ammonium nitrate explosions,” the

solution is simple and should have been recognized already, even though it

has never been acknowledged. Stored AN disasters will not occur if

uncontrollable fires do not occur. The technology needing to be implemented

for this is twofold: (1) manufacturers must switch from making straight AN

to making fire- and detonation-resistant products; and (2) owners of storage

facilities need to implement building construction features which preclude

the possibility of an uncontrolled fire. The needed technology to achieve both

these objectives has existed for many decades and is in no way novel.

There are only two entities that can prevent future disasters—government

and manufacturers. In the US, both have been wholly ineffectual. US

government regulations can be on a Federal, State, or local level. But

building construction regulations are administered only on a local level. AN

fertilizer is predominantly stored and used in rural areas, and construction

regulation in such areas is generally ineffective or, as in the case of Texas,

nonexistent. Regulations to prohibit straight AN in preference to safer

fertilizer products could be enacted on a Federal level, but two different non-

regulatory Federal agencies, the US Chemical Safety Board and the

Government Accountability Office, explicitly documented the ineffectiveness

of OSHA, the agency which does have regulatory authority. The latter agency

could also abate hazards in most storage occupancies since they constitute

workplaces. However, the existing regulations are misguided and ineffective

towards achieving needed fire safety.

272


The AN manufacturers could prevent disasters in the absence of viable

governmental regulations. But unlike in some other areas of the chemical

industry, the manufacturers in this sector have consistently failed to

recognize the importance of Product Stewardship as a means of avoiding

death and injury. This means that product is routinely sold to parties not

competent to safely store the material. Finally, codes and standards in this

area have also been comprehensively defective. Notably, NFPA 400 has not

required safety features which are crucial towards avoiding AN disasters and

its recent improvements are insufficient.

11. References

1. Vytenis Babrauskas Explosions of ammonium nitrate fertilizer in storage or

transportation are preventable accidents Journal of Hazardous Materials, Volume

304,5 March 2016.

2. Zhe Han, Sonny Sachdeva, Maria I. Papadaki, Sam Mannan Effects of inhibitor and

promoter mixtures on ammonium nitrate fertilizer explosion hazards Thermochimica

Acta, Volume 624, 20 January 2016.

3. R.S. Ettouney, M.A. EL-Rifai Explosion of ammonium nitrate solutions, two case

studies Process Safety and Environmental Protection,Volume 90, Issue 1, January 2012.

4. Zhe Han, Sonny Sachdeva, Maria I. Papadaki, M. Sam Mannan Ammonium nitrate

thermal decomposition with additives Journal of Loss Prevention in the Process

Industries, Volume 35, May 2015, Zhe Han, Sonny Sachdeva, Maria I. Papadaki, M.

Sam Mannan

5. Bennett, D., West, Texas: The Town That Blew Up, Business Week (3 July 2013).

7. M2.1 Explosion—1 km NNE of West, Texas, 2013-04-18 00:50:38 UTC.

http://comcat.cr.usgs.gov/earthquakes/eventpage/usb000g9yl# summary.

8. State of Texas Outreach to Counties Having an Ammonium Nitrate Facility, Texas

State Fire Marshal's Office, [n.p.] (2014).

273










9. West Fertilizer/Adair Grain Company. Ammonium Nitrate Fire and Explosion, Public

Meeting, April 22, 2014, US Chemical Safety and Hazard Investigation Board,

Washington (2014).

10. Chemical Safety–Actions Needed to Improve Federal Oversight of Facilities with

Ammonium Nitrate (GAO-14-274), US Government Accountability Office, Washington

(2014).

12. Pasturenzi, L. Gigante, P. Cardillo Nitrato d'ammonio: un secolo di esplosioni La

Revista dei Combustibili, 67 (2) (2013).

274

CHAPTER – 10

EXPLOSIVE MANUFACTURING SITE

REMEDIATION 1. Technical introduction

1.1 Treatment Technologies for Explosive and Radioactive Waste at

Federal Facilities

Most of the treatment technologies for explosive waste discussed in this

document currently are being developed or implemented. These include

biological technologies, incineration, ultraviolet oxidation, granular activated

carbon treatment, and reuse/recycle options. Similarly, all of the radioactive

waste treatment technologies discussed in Chapter 6, including volume

reduction, polymer solidification and encapsulation, incineration, in situ

vitrification, in situ grout injection, and electro kinetic soil processing, have

been successfully demonstrated. This document also discusses four

treatment technologies that have not been successfully implemented for

explosive waste: wet air oxidation, low temperature thermal desorption,

solvent extraction, and volume reduction. For additional information on

treatment technologies for 2,4,6- trinitrotoluene (TNT) explosive waste,

please see Installation Restoration and Hazardous Waste Control

Technologies.

1.2 Explosive Waste

1.2.1 Types of Explosive Waste

The term explosive waste commonly is used to refer to propellants,

explosives, and pyrotechnics (PEP), which technically fall into the more

general category of energetic materials. These materials are susceptible to

initiation, or self-sustained energy release, when exposed to stimuli such as

heat, shock, friction, chemical incompatibility, or electrostatic discharge.

275

Chapter 10 Explosive Manufacturing Site Remediation

Each of these materials reacts differently to the aforementioned stimuli; all

will burn, but explosives and propellants can detonate under certain

conditions (e.g., confinement). Figure 1-1 outlines the various categories of

energetic materials. The emphasis of this document is on soil and ground

water contaminated with explosives rather than propellants or pyrotechnics.

Explosives

Explosives are classified as primary or secondary based on their

susceptibility to initiation. Primary explosives, which include lead azide and

lead styphnate, are highly susceptible to initiation. Primary explosives often

are referred to as initiating explosives, because they can be used to ignite

secondary explosives.

Secondary explosives, which include TNT, cyclo-1,3,5-trimethylene-2,4,6-

trinitramine (RDX or cyclonite), High Melting Explosives (HMX), and tetryl,

are much more prevalent at military sites than are primary explosives. Since

they are formulated to detonate only under specific circumstances,

secondary explosives often are used as main charge or boostering

Explosives. Secondary explosives can be loosely categorized into melt-pour

explosives, which are based on TNT, and plastic-bonded explosives (PBX),

which are based on a binder and a crystalline explosive such as RDX.

Secondary explosives also can be classified according to their chemical

structure as nitroaromatics, which include TNT, and nitramines, which

include RDX. Figure 1-2 shows the chemical structure of TNT and RDX. In

the TNT molecule, N02 groups are bonded to the aromatic ring; in the RDX

molecule, N02 groups are bonded to nitrogen.

Table 1-1 shows how frequently various nitroaromatics and nitramines

occur at explosives-contaminated sites with which the U.S. Army Cold

Regions Research and Engineering Laboratory (CAREL) and the Missouri

River Division (MRD) have been involved. TNT is the most common

276


contaminant, occurring in approximately 80 percent of the soil samples

found to be contaminated with explosives. Trinitrobenzene (TNB), which is a

photochemical decomposition product of TNT, was found in between 40 and

50 percent of these soils. Dinitrobenzene (DNB), 2,4-dinitrotoluene (2,4-

DNT), and 2,6-DNT, which are impurities in production-grade TNT, were

found in less than 40 percent of the soils. Figure 1-2 shows the chemical

structures of common explosive contaminants.

Propellants Propellants include both rocket and gun propellants. Most rocket

propellants are either (1) Hazard Class 1.3 composites, which are based on a

rubber binder, an ammonium perchlorate (AP) oxidizer, and a powdered

aluminum (AI) fuel; or (2) Hazard Class 1.1 composites, which are based on

a nitrate ester (usually nitroglycerine [NC]), nitrocellulose (NC), HMX, AP,

and Al.’ The nitrate ester propellants can be plastisol-bound (high NC) or

polymer-bound (low NC). If a binder is used, it usually is an isocyanate-

cured polyester or polyether. Some propellants contain combustion

modifiers, such as lead oxide. Gun propellants usually are single base (NC),

double base (NC and NC), or triple base (NC, NG, and nitroguanidine [NQ]).

Some of the newer, lower vulnerability gun propellants contain binders and

crystalline explosives and thus are similar to PBX.

277


Figure 1·1. Categories of energetic materials.

Figure 1·2. Chemical structures of common explosive contaminants.

278


Table1.1. Nltroaromatics and Nitraminea Detected by CAREL and MRD

In Explosives-Contaminated Solis from Army Sites

Category Contaminant Frequency % CAREL MRD

Nitroaromatics TNT 85 76 TNB 53 38 DNB 25 19 2,4-DNT 41 17 2,6-DNT 4-Amino-DNT 6 3 2-Amino-DNT 27 11 3,5-DNA .. .. Nitramines RDX 44 28 HMX 27 4 Tetryl 8 14 *Often not separated from 2, 4-DNT.

"Peak often observed but only recently Identified. Source: U.S. Army CAREL,

1993.

Pyrotechnics

Pyrotechnics include illuminating flares, signaling flares, colored and white

smoke generators, tracers, incendiary delays, fuzes, and photo-flash

compounds. Pyrotechnics usually are composed of an inorganic oxidizer and

metal powder in a binder. Illuminating flares contain sodium nitrate,

magnesium, and a binder. Signaling flares contain barium, strontium, or

other metal nitrates.

1.2.2. Sources of Explosive Waste

Many DOD sites are contaminated with explosive waste as a result of

explosives manufacturing; munitions load, assemble, and pack operations;

explosives machining, casting, and curing; open burn and open detonation

operations; and laboratory testing of munitions. Based on the experience of

279


the U.S. Army Environmental Center (AEC) of DOD, one of the major

explosive wastes of concern at DOD sites are residues from land disposal of

explosives-contaminated process water.

Explosives-contaminated waters are subdivided into two categories: red

water, which comes strictly from the manufacture of TNT; and pink water,

which includes any wash water associated with load, assemble, and pack

operations or with the demilitarization of munitions involving contact with

finished TNT. Despite their names, red and pink water cannot be identified

by color. Both are clear when they emerge from their respective processes

and subsequently turn pink, light red, dark red, or black when exposed to

light. The chemical composition of pink water varies depending on the

process from which it is derived; red water has a more-defined chemical

composition. For this reason, it is difficult to simulate either red or pink

water in the laboratory.

The United States stopped production of TNT in the mid-1980s, so no red

water has been generated in this country since that Date (Hercules

Aerospace Company, 1991).

Most process waters found in the field are pink waters that were generated

by demilitarization operations conducted in the 1970s.ln these operations,

munitions were placed on racks with their fuzes and tops removed. Jets of

hot water then were used to mine the explosives out of the munitions. The

residual waters were placed in settling basins so that solid explosive

particles could be removed, and the remaining water was siphoned into

lagoons. Contaminants often present in these lagoon waters and the

surrounding soils include TNT; RDX; HMX; tetryl; 2,4-DNT; 2,6-DNT; 1,3-

DNB; 1, 3, 5-TNB; and nitrobenzene.

280


1.2.3 Radioactive Waste

Several radioactive elements, including uranium, radium, and radon, occur

naturally in soil and ground water. Radioactive contamination also can

result from processes associated with the production of nuclear energy and

nuclear weapons. Common radioactive-contaminated materials include dry

active wastes, such as paper, plastic, wood, cloth, rubber, canvas,

fiberglass, and charcoal; ion exchange resins used to polish condensate from

nuclear power plants; sewer sludges and lubricating oils contaminated with

radioactive materials; and air pollution control equipment. For the purposes

of this document radionuclides should be considered to have properties

similar to those of other heavy metals.

The Nuclear Weapons Complex (NWC) is a collection of enormous factories

devoted to metal fabrication, chemical separation processes, and electronic

assembly associated with the production of nuclear weapons. In

approximately 50 years of nuclear weapons production, these factories have

released vast quantities of hazardous chemicals and radionuclides to the

environment. Evidence exists that air, ground water, surface water,

sediment, and soil, as well as vegetation and wildlife, have been

contaminated at most, if not all, nuclear weapons production facilities. Table

1-2 shows the types of wastes often found at NWC sites.

Contamination of soil, sediments, surface water, and ground water is

widespread at the NWC, and contamination of ground water with radio

nuclides or hazardous chemicals has been confirmed at almost every facility.

Most sites in nonarid locations have surface water contamination as well.

Almost 4,000 solid waste management units (SWMUs) have been identified

throughout the NWC, and many of these units require some form of

remedial action. Substantial quantities of waste have been buried at the

NWC, often with inadequate records of the burial location or composition of

the waste buried. DOE estimates that a total of about 0.2 million m of

281


transuranic waste and about 2.5 million of low-level radioactive wastes have

been buried in the complex. Most of this buried waste is "mixed waste,"

meaning that it is mixed with Resource Conservation and Recovery Act

(RCRA) hazardous wastes. For additional information on radioactive waste

sites, refer to Complex Cleanup: The Environmental Legacy of Nuclear

Weapons Production (U.S. Congress, 1991).

Table 1·2. Nuclear Weapons Site Contaminants and

Contaminant Mixtures•

Inorganic Contaminants Organic Contaminants

Organic Facilitators

Mixtures of Contaminants Radionuclides Metals Other

Americium-241 Cesium-134, -137 Cobalt-60 Plutonium-238, ·239 Radium-224, -226 Strontium-90 Technetium-99 Thorium-228, ·232 Uranium-234, -238

Chromium Copper Lead Mercury Nickel

Cyanide

Benzene Chlorinated hydrocarbons Methylethyl ketone, cyclohexanone, acetone Polychlorinated biphenyls and select polycyclic aromatic hydrocarbons Tetraphenylboron Toluene Tributylphosphate

Aliphatic acids Aromatic acids Chelating agents Solvents, diluents, and chelate radiolysis fragments

Radionuclides and metal ions Radionuclides, metals, and organic acids Radionuclides, metals, and natural organic substances Radionuclides and synthetic chelating agents Radionuclides and solvents Radionuclides, metal ions, and Organophosphates Radionuclides, metal ions, and petroleum hydrocarbons Radionuclides, chlorinated solvents, and petroleum hydrocarbons Petroleum hydrocarbons and polychlorinated biphenyls Complex solvent mixtures Complex solvent and petroleum hydrocarbon mixtures.

282

CHAPTER – 11

TREATMENT TECHNOLOGIES FOR EXPLOSIVES WASTE

1. Biological Treatment Technologies

1.1 Background

Biological treatment, or bioremediation, is a developing technology that uses

microorganisms to degrade organic contaminants into less hazardous

compounds. Compared to conventional technologies, bioremediation has

several advantages: (1) it actually degrades target compounds, rather than

just transferring them from one medium to another; (2) it is publicly

accepted, because it is a natural process; and (3) it is probably less

expensive than incineration, especially for small volumes of contaminated

soil.

Although the two terms occasionally are interchanged, biodegradation is not

synonymous with mineralization. Mineralization, which is the process ·by

which compounds are transformed into carbon dioxide and water, is only

one of several fates of contaminants in biological treatment systems.

Contaminants also may be volatilized, bind to organic materials, be

assimilated into an active biomass, or be transformed into compounds other

than carbon dioxide and water. Mineralization of contaminants is a desired,

but rarely achieved, outcome of bioremediation. This section discusses the

types of explosives that can be bioremediated and highlights five specific

biological treatment technologies: aqueous-phase bioreactor treatment,

composting, land farming, white rot fungus treatment, and in situ biological

treatment.

1.2 Treatable Wastes and Media

Bioremediation is most effective for dilute solutions of explosives and

propellants. TNT in the ctystalline form is difficult to treat biologically.

283

Chapter 11 Treatment Technologies for Explosives Wastes

TNT degrades under aerobic conditions into monoamine-, diamino-, and

hydroxylamino-DNT, and tetranitro-azoxynitrotoluenes. RDX and HMX

degrade into carbon dioxide and water under anaerobic conditions.

Researchers have not identified any specific organisms that are particularly

effective for degrading explosives waste; a consortium of organisms usually

effects the degradation

1.3 Operation and Maintenance

DOD currently is developing or implementing five biological treatments for

explosives-contaminated soils: aqueous-phase bioreactor treatment;

composting, land farming, and white rot fungus treatment, which are solid-

phase treatments; and in situ biological treatment.

1.3.1 Aqueous-Phase Bioreactor Treatment

DOD is considering two types of aqueous-phase bioreactors for the

treatment of explosive contaminants. The first is the lagoon slurry reactor,

which allows contaminants to remain in a lagoon, be mixed with nutrients

and water, and degrade under anaerobic conditions. Figure 1 is a schematic

of a lagoon slurry reactor. The second is the aboveground slurry reactor,

which is either a concrete activated sludge basin or a commercially available

bioreactor. Figure 2 is a schematic of aboveground bioreactor treatment,

showing the excavation and screening of soils prior to treatment, dewatering

of the treated soil, and recycling of the extracted water to the reactor.

Aqueous-phase bioreactors provide good process control, can be configured

in several treatment trains to treat a variety of wastes, and potentially can

achieve very low contaminant concentrations. A drawback of bioreactor

treatment is that, unlike composting systems which bind contaminants to

humic material, bioreactors accumulate the products of biotransformation.

In addition, bioreactors have been shown to remediate explosives only at

284

user

Stamp


laboratory scale, so the cost of full-scale bioreactor treatment is unknown.

Full-scale bloreactors will have to incorporate a variety of safety features

that will add to their total cost.

Figure 1. Schematic of lagoon slurry reactor.

The Army is conducting a demonstration study to examine the effectiveness

of treating explosives-contaminated soils from the Joliet Army Ammunition

Plant (JAAP) in an aboveground sequencing batch bioreactor. The goal of

this study is to determine the extent of degradation, byproducts, and total

costs of full-scale bioreactor treatment. Soils will be excavated from the site,

screened, and pumped into the reactor. Indigenous microorganisms from

the site will be isolated and added to the reactor. Either malate or molasses

will be used as a substrate. After processing in the reactor, soils will be

drawn into a filter bed, where process waters will be removed. These process

waters will be recycled back to the reactor, and any remaining discharges

will be treated to meet National Pollutant Discharge Elimination System

(NPDES) requirements. Initial laboratory testing of this system produced

greater than 99 percent contaminant reductions within 14 days (see Figure

3).

285


1.3.2 Composting DOD has been evaluating composting systems to treat explosives waste

since 1982. To date, composting has been shown to degrade TNT, RDX,

HMX, DNT, tetryl, and nitrocellulose in soils and sludges. The main

advantage of this technology is that, unlike incineration, composting

generates an enriched product that can sustain vegetation. After cleanup

levels are achieved, the compost material can be returned to the site and

covered with a soil cap. Another advantage is that composting provides both

aerobic and anaerobic treatment, so it is effective for a range of wastes. The

feasibility of composting can be limit d. however, by the level of indigenous

organisms in contaminated soil and the local availability 01 amendment

mixtures. In addition, composting requires long treatment periods for some

waste streams, and composting of unfamiliar contaminants potentially can

generate toxic byproducts.

Figure 2. Schematic of aboveground slurry reactor treatment.

286


Composting methods fall into four categories: (1) static-pile composting; (2)

in-vessel, static-pile composting; (3) mechanically agitated, in-vessel

composting; and (4) windrow composting. In static-pile composting,

contaminated material is excavated, placed in a pile under a protective

shelter, and mixed with readily degradable carbon sources. The pile

undergoes forced aeration to maintain aerobic and thermophilic (55 to 60°C)

conditions, which foster the growth of microorganisms. Bulking agents, such

as cow manure and vegetable waste, can be added to enhance

biodegradation. Figure 4 is a schematic of static-pile composting. In-vessel,

static-pile composting is similar to static-pile composting except the

compost pile is placed in a vessel. Figure 5 is a schematic of an in-vessel,

static-pile composting device. In mechanically agitated, in-vessel

composting, contaminated material is aerated and blended with carbon-

source materials in a mechanical composter. These devices have been used

at municipal sewage treatment facilities and applied to explosives waste.

Figure 6 is a schematic of a mechanical composter. Windrow composting is

similar to static-pile composting except that compost is aerated by a

mechanical rnixil1g vehicle, rather than a forced air system.

In 1988, the Army began a series of demonstration studies at the Louisiana

Army Ammunition Plant to determine the effectiveness of composting

explosives-contaminated soils. In the initial study, static-pile composting

required

153 days to remediate soils contaminated with just 3 percent explosive

waste by volume. Based on these results, the Army determined that static-

pile composting would not be cost effective for remediating large volumes of

explosives waste.

287


Figure 3. Contaminant reductions achieved In laboratory-scale testing

of sequencing batch reactor treatment ot soils from Joliet Army

Ammunition Plant

Figure 4. Schematic of static-pile composting, showing the compost

pile, protective shelter, forced aeration system,

and leachate collection pad.

288


Figure 5. Schematic of In-vessel, static-pile composting equipment

Figure 6. Schematic of a mechanical composter

289


$50 per ton, and used a commercially available mechanically agitated

composter rather than a static pile. These conditions led to more rapid and

extensive degradation of the explosives, achieving cleanup levels of 10 to 20

ppm of TNT and RDX within 20 days. Nevertheless, this method also was

determined to be economically infeasible, due to the initial cost of the

commercial composter.

Finally, the Army conducted a study to examine the effectiveness of windrow

composting. This study used cow manure, sawdust, and potato waste

amendments and required the construction of a concrete pad leachate

collection system. Temperatures were maintained at 55°C and the compost

was turned once a day. This process produced 98 percent reductions of

explosives contamination within 20 days, and degraded HMX, which

formerly had resisted degradation (see · Figure 7 and Table 1). Toxicological

data from this study indicated that composting achieved 90 to 98 percent

toxicity reductions, consumption of the compost material would not have

been toxic to rats, the leachates exhibited no mutagenicity, and some of the

TNT had been mineralized. Radiolabeled TNT studies indicated that strong

binding had occurred between TNT and the humic compost. Since the initial

costs were relatively low, windrow composting was determined to be an

economically feasible alternative to incineration.

Figure 7. TNT, RDX, and HMX reductions achieved In windrow composting

demonstration study at Louisiana Army Ammunition Plant

290


Table 1. Actual and Percent Contaminant Reductions Achieved In

Windrow Composting Demonstration Study at

Louisiana Army Ammunition Plant

Contaminant Level (µg/g) Reduction (%)

Day TNT RDX HMX TNT RDX HMX

0 1563 953 156 0.0 0.0 0.0

5 101 1124 158 93.5 0.0 0.0

10 23 623 19 98.5 34.6 23.7

15 19 88 18 98.8 90.7 24.4

20 11 5 2 99.3 99.5 98.7

40 4 2 5 99.7 99.8 96.8

Composting methods were evaluated in a feasibility study at the Umatilla

Army Depot TNT washout lagoons. In initial testing, composting compared

well to incineration in terms of treatment performance but not in terms of

cost. The Army then analyzed the factors affecting the cost of composting,

including the specific composting method, volume of contaminated soil, soil

throughput, amendment costs, and treatment time. This analysis suggested

that for treating less than 10,000 tons of contaminated material, the cost

would be $740 per ton for incineration, $651 per ton for mechanically

agitated composting, and $386 per ton for windrow composting. Figure 8

shows estimated composting and incineration costs as a function of total

soil volume treated. Based on these estimates, the Army elected to use

windrow composting as the remedial action at the Umatilla site for 300 tons

per day.

1.3.3 Land Farming Land farming has been used extensively to treat soils contaminated with

petroleum hydrocarbons, pentachlorophenol (PCP), and polycyclic aromatic

291


hydrocarbons (PAHs), and potentially could be used to treat low to medium

concentrations of explosives as well. In land farming, soils are excavated to

treatment plots and periodically rototilled to mix in nutrients, moisture, and

bacteria. Land farming typically achieves very slow degradation rates and

can take many years to reach target cleanup levels.

In one pilot study at an explosives waste site in Hercules, California, soils

contaminated with TNT and DNT were excavated to 1-yd3 bins, inoculated

with organisms indigenous to the site, and amended with brain/heart

infusion agar, which is a common laboratory agar. This procedure failed to

achieve the target cleanup levels of 30 ppm TNT, 5 ppm DNT, and 5 ppm.

Figure 8. Comparison of costs for windrow composting; mechanically

agitated, in-vessel composting (MAIV); and incineration of Umatilla

Army Depot soils as a function of total soil volume treated.

292


DNB, achieving instead a 30 to 40 percent contaminant degradation. 1.3.4 White Rot Fungus Treatment White rot fungus has been evaluated more extensively than any other fungal

species for remediating explosives waste. Although white rot fungus

degradation of TNT has been reported in laboratory-scale settings using pure

cultures (Berry and Boyd, 1985; Fernando et al. 1990), a number of factors

increase the difficulty of using this technology for full-scale remediation.

These factors include competition from native bacterial populations, toxicity

inhibition, chemical sorption, and the inability to meet risk-based cleanup

levels.

In bench-scale studies of mixed fungal and bacterial systems, most of the

reported degradation of TNT is attributable to native bacterial populations

(Lohr, 1993; McFarland et al., 1992).High TNT concentrations in soil also

can inhibit growth of white rot fungus. One study suggested that

Phanerochaete chrysosporium was incapable of growing in soils

contaminated with 20 ppm or more of TNT. In addition, some reports

indicate that TNT losses reported in white rot fungus studies can be

attributed to adsorption of TNT onto the fungus and soil amendments, such

as corn cobs and straw (Spiker et al., 1992).

A pilot-scale treatability study was conducted using white rot fungus at a

former ordnance open burn/open detonation area at Site D, Naval

Submarine Base, Bangor, Washington. Initial TNT concentrations of 1,844

ppm were degraded to 1,267 ppm in 30 days and 1,087 in 120 days. The

overall degradation was 41 percent, and final TNT soil levels were well above

the proposed cleanup level of 30 ppm (Spectrum Sciences & Software, Inc.,

and Utah State University, no date).

293


1.3.5 In Situ Biological Treatment In situ treatments can be less expensive than other technologies and

produce low contaminant concentrations. The available data suggest,

however, that in situ treatment may not be effective for explosives waste. In

situ treatment of explosives might create more mobile intermediates during

biodegradation. In addition, biodegradation of explosive contaminants

typically involves cometabolism with another nutrient source, which is

difficult to deliver in an in situ environment. Mixing often affects the rate

and performance of explosives degradation. Finally, because in situ

remediation takes place beneath the surface, the effectiveness of in situ

treatment is difficult to verify both during and after treatment.

1.4 References Berry, D.F. and S.A. Boyd. 1985. Decontamination of soil through

enhanced formation of bound residues. Environmental Science and

Technology 19:1132-1133.

Fernando, T., J.A. Bumpus, and S.D. Aust. 1990. Biodegradation of

TNT (2,4,6-trinitrotoluene) by Phanerochaete chrysosporium. Applied

Environmental Microbiology 56:1667-1671.

Lohr, J.T. 1993. Bioremediation of TNT and RDX using white rot fungus

Phanerochaete chrysosporium. Utah State University. Prepared for

Naval Civil Engineering Laboratory, Port Hueneme, California. Contract

No. DML-03-86-D-0001.

McFarland, M.J., S. Kalaskar, and E. Baiden. 1992. Composting of

explosives-contaminated soil using the white rot fungus Phanerochaete

chrysosporium. Utah State University. Prepared for the U.S. Army

Research Office, Research Triangle Park, North Carolina. Contract No.

294


DAAL-03-91-C-0034.

Spectrum Sciences & Software, Inc., and Utah State University. No

date. White rot remediation of ordnance-contaminated media. Prepared

for Naval Civil Engineering Laboratory, Port Hueneme, California.

Contract No. F49650-90-D5001/DO 5013.

Spiker, J.K., D.L. Crawford, and R.L. Crawford. 1992. Influence of

2,4,6-trinitrotoluene (TNT) concentration on ttle degradation of TNT in

explosive-contaminated soils by the white rot fungus Phanerochaete

c/uysosporium. Applied Environmental Microbiology 58:3199-3202.

2. Thermal Treatment Technologies

2.1 Incineration of Soils and Sludges

2.1.1 Background

AEC of DOD at Aberdeen Proving Ground, Maryland, oversees large-scale

incineration of munitions, explosives waste, and explosives-contaminated

soils as part of remedial actions at Army sites. This section discusses the

types of wastes and media that can be incinerated, looks at various devices

used to incinerate explosives waste, presents case studies of four sites

where incineration has been applied to explosives-contaminated soils, and

examines the advantages and disadvantages of incineration.

2.1.2 Treatable Wastes and Media Incineration processes can be used to treat the following waste streams:

explosives-contaminated soil and debris, explosives with other organics or

metals, initiating explosives, bulk explosives, unexploded ordnance, bulky

radioactive waste, and pyrophoric waste. In addition, incineration can be

applied to sites with a mixture of media, such as concrete, sand, clay,

295


water, and sludge, provided the media can be fed to the incinerator and

heated for a sufficient period of time. With the approval of the DOD

Explosives Safety Board, the Army considers incineration of materials

containing less than 10 percent explosives by weight to be a non explosive

operation. Soil with less than 10 percent explosives by weight has been

shown by AEC to be nonreactive, that is, not to propagate a detonation

throughout the mass of soil. (The military explosives to which this limit

applies are secondary explosives such as TNT and RDX, and their

manufacturing byproducts.)

The Army's first pilot-scale use of rotary kiln incineration utilized soil well

above the 10 percent limit (up to 40 percent) with approval from the DOD

Explosives Safety Board. A consideration in conducting the test was the fact

th8t the kiln was not actually sealed and hence not thought to provide

confinement for the small amount of explosives fed. Another consideration

was a previous successful Army incineration of pure TNT without detonation

in a deactivation furnace. Though the pilot-scale test experienced no

detonation problems, the Army's full-scale incineration projects have

incorporated a blending step to reduce the explosives concentrations below

the 10 percent limit prior to feeding. The blending step is considered to be

an explosives operation that requires the preparation and approval by the

Army and DOD safety offices of a site plan/safety submission, which must

include an explosives hazard analysis. Finally, even at explosives

concentrations below 10 percent, each explosives project has unique

elements, and a thorough safety review is a necessity.

The Army also has developed and tested a feed system capable of feeding

reactive levels of explosives (up to 20 percent). The system includes multiple

units with breaks in between to prevent propagation of a possible detonation

throughout the system. Metal-to-metal contact also is minimized in the

system to reduce the chances of detonation by friction or spark.

296


2.1.3 Operation and Maintenance The Army primarily uses three types of incineration devices: the rotary kiln

incinerator, deactivation furnace, and contaminated waste processor.

Rotary Kiln Incinerator The rotary kiln incinerator is used primarily to treat explosives-

contaminated soils. In rotary kiln incineration, soils are fed into a

primary combustion chamber, or rotary kiln, where organic constituents are

destroyed. The temperature of gases in the primary chamber ranges from

800 to 1,200°F, and the temperature of soils ranges from 600 to 800°F.

Retention time in the primary chamber, which is varied by changing the

rotation speed of the kiln, is approximately 30 minutes. Off gases from the

primary chamber pass into a secondary combustion chamber, which

destroys any residual organics. Gases from the secondary combustion

chamber pass into a quench tank where they are cooled from approximately

2,000°C to 200°C. From the quench tank, gases pass through a Venturi

scrubber and a series of baghouse filters, which remove acid gases and

particulates prior to release from the stack. The treated product of rotary

kiln incineration is ash (or treated soil), which drops from the primary

combustion chamber after organic contaminants have been destroyed. This

product is routed into a wet quench or a water spray to remoisturize it, then

transported to an interim storage area pending receipt of chemical analytical

results.

Deactivation Furnace The deactivation furnace also is referred to as Army Peculiar Equipment

(APE) 1236, because it is used almost exclusively by the Army to deactivate

large quantities of small arms cartridges, so-caliber machine gun

ammunition, mines, and grenades. The deactivation furnace is similar to

297


the rotary kiln Incinerator, except that it is equipped with a thick-walled

primary combustion chamber capable of withstanding small detonations.

Deactivation furnaces do not have secondary combustion chambers,

because they are intended not to completely destroy the vaporized explosives

but to render the munitions unreactive. Most deactivation furnaces are

equipped with air pollution control equipment to limit lead emissions. The

operating temperature of deactivation furnaces is approximately 1,200 to

1,500°F.

Contaminated Waste Processor The contaminated waste processor handles materials, such as surface-

contaminated debris, that are lighter and less reactive than those processed

in the deactivation furnace. Contaminated waste processors are thin-walled,

stationary ovens that heat contaminated materials to about 600°C for 3 to 4

hours. The purpose of this process is not to destroy contaminated debris but

to lower contaminant levels to meet Army safety standards. AEC currently is

helping to develop standardized time and temperature processing

requirements to meet these safety standards.

2.1.4 Case Studies

Cornhusker Army Ammunition Plant

The Cornhusker Army Ammunition Plant (CAAP) in Grand Island, Nebras {a,

was the site of 58 explosives wastewater washout cesspools and leaching

pits. Explosives residues from these 1O-ft deep pits created a contaminated

ground water plume that extended into nearby residential areas. To prevent

further ground water contamination, the Army opted to incinerate

contaminated soils and sludges from the cesspools and leaching pits. For

each contaminant, the Army established two cleanup criteria: (1) an

excavation criterion, which was health risk based and determined the depth

298


to which soils were excavated, and (2) an incineration criterion, which

equaled the nondetection level for each contaminant. Table 2 shows the

cleanup criteria for contaminants from the CAAP site.

Figure 9 is a schematic of the rotary kiln incineration system employed at

the CMP site. A three-stage feed system with a live bottom hopper, belt

conveyor, and gravity tube was used to feed contaminated material to the

incinerator. Ash from the incinerator was loaded into ash bins and subjected

to compositional analysis. Once the ash was determined to be clean (i.e., to

contain no detectable explosives), it was backfilled at a single location on the

CMP site. The CAAP project was completed successfully in 1988, after

incinerating 40,000 tons at an average total cost of $260 per ton. Some of

the difficulties encountered included (1) clogging of the quench tank by slag

that fell from the walls of the secondary combustion chamber, (2) unwanted

air infiltration through the air lock in the feed system, and (3) the need to

winterize the unit for cold weather operations.

Louisiana Army Ammunition Plant Over the years, wastewaters from ammunition load, assemble, and pacl'

operations at the Louisiana Army Ammunition Plant (LAAP) in Shreveport,

Louisiana, were shipped by true to 16 leaching/evaporation lagoons at ArP.a

P in south ·central LAAP. Explosives residues from these lagoons leached

into the underlying ground water, creating plumes of TNT and RDX. As at

the CAAP site, the Army opted to incinerate soils and sludges from the LAAP

lagoons and set the incineration cleanup criterion equal to the non detection

limit for each contaminant. Rather than assign each contaminant a specific

excavation criterion, the Army specified that the concentrations of all

contaminants total less than 100 ppm after 1 foot of lagoon material had

been excavated. Table 3 shows the cleanup criteria for the LAAP lagoons.

299


The incineration system used at CAAP was transported to LAAP with a

significant modification to the quench to allow workers to clean it without

entering the tank. While operating at LAAP, some other modifications were

made to correct the following difficulties: (1} clayey wet feed soil plugged and

jammed the feed system and (2) buildup of soil on the secondary

combustion chamber fell into the quench tank causing a steam

overpressure. To remedy the first problem, the feed system was

strengthened and a high-speed slinger belt conveyor was used as the final

stage to throw the soil into the incinerator. To remedy the second problem,

which may have been aggravated by the lime used to dry the feed, the

quench was relocated in an offset position from the secondary combustion

chamber. The project was completed successfully in 1990 after incinerating

102,000 tons of soil at an average total cost $330 per ton.

Table 2. Cleanup Criteria for Cornhusker Army Ammunition Plant

Analyte Excavation

Criteria (ppm)

Incineration Criteria

{Method Detection

Limits [ppm])

RDX <10 <2.2

2,4,6 –TNT <5 <1.3

1,3,5 –TNB <0.5 <1.25

2,4-DNT <0.4 <0.24

2,6-DNT NA <126

HMX NA <2.9

1,3-DNB NA <1.2

NB NA <1.26

Tetryl NA <2.2

2A,4,6-DNT NA <1.25

300


Figure 9. Schematic of rotary kiln Incineration system employed at

Cornhusker Army Ammunition Plant

Savanna Army Depot The Savanna Army Depot (SVAD) in Savanna, Illinois, formerly operated a

washout plant where hot water was used to melt the explosives out of

munitions. Wastewaters from these operations were pumped directly from

the facility through a metal trough into washout lagoons. Recently, SVAD

began piping wastewaters into two new washout lagoons on a sandy hill

near the facility. Both the old and new lagoons are contributing explosives

contamination to ground water beneath the site. The old lagoons are located

in a flood plain of the Mississippi River, which runs about 12 milewest of the

site. Periodically, the river floods the lagoons, spreading explosives

contamination from the centers of the lagoons.

The entire site was screened for unexploded ordnance prior to the start of

incineration operations. The Army then established health risk based

excavation criteria and non detection limit incineration criteria for the soils

301


at the site (see Table 4). To reach the excavation criteria, some lagoons had

to be excavated to a depth of 10 ft and excavation had to be done outside of

the lagoons, apparently due to the periodic flooding by the Mississippi River.

As a safety precaution, excavated soils were blended to reduce overall

explosives levels to less than 10 percent by weight. Incineration currently is

under way. Some problems have arisen with the feed system clogging due to

the cold, wet conditions at the site, but incineration is expected to be

completed in fall of 1993. The estimated quantity of soil to be incinerated is

approximately 60,000 tons.

Table 3. Cleanup Criteria for Louisiana Army Ammunition Plant

Analyte Excavation Criteria

(ppm)


(Method Detection

Limits.[ppm])

RDX Sum of all less than

100 ppm after 1 foot

excavation of lagoons

<2.2

2,4,6 –TNT <1.3

1,3,5 –TNB <1.25

2,4-DNT <0.24

2,6-DNT <126

HMX <2.9

1,3-DNB <1.2

NB <1.26

Tetryl <2.2

2A,4,6-DNT <1.25

Alabama Army Ammunition Plant In 1986, explosives- and lead-contaminated soils from the Alabama Army

302


Ammunition Plant in Childersburg, Alabama, were excavated and placed on

a concrete slab and in two containment buildings. These soils, totalling

approximately 35,000 tons, are slated to undergo incineration over the next

2 years. Table 5 shows the excavation and incineration criteria for the site.

The excavation criteria, which are health risk based, governed the initial

excavation in 1986. The incineration criteria al! are equal to non detection

limits. The Army anticipates two problems. First, the soils contain large

amounts of debris and possibly pieces of explosive, which will have to be

removed manually prior to incineration. Second, the soils contain lead, so

the ash product may have to be stabilized prior to disposal.

Table 4. Cleanup Criteria for Savanna Army Depot

Analyte Excavation

Criteria

(ppm)


(Method Detection

Limits.[ppm])

RDX <5.75 <1

2,4,6 –TNT <21.1 <1

1,3,5 –TNB <3.7 <1

2,4-DNT <9.3 <1

2,6-DNT <4.3 <1

HMX <3,722 <1

1,3-DNB <7.4 <1

NB <37.2 <1

Tetryl <112 <1

2A,4,6-DNT <1,191 <1

303


Table 5. Cleanup Criteria for Alabama Army Ammunition Plant

Analyte Excavation

Criteria

(ppm)


(Method Detection

Limits.[ppm])

RDX None <1

2,4,6 –TNT <1.92 <1

1,3,5 –TNB <5.5 <1

2,4-DNT <0.42 <1

2,6-DNT <0.42 <1

HMX None <1

1,3-DNB <1.1 <1

NB None <1

Tetryl <1.7 <1

2A,4,6-DNT None <1

2.1.5 Advantages and Disadvantages Incineration has many advantages, including: • Effectiveness. With sufficiently long residence time and a sufficiently high

temperature, incineration usually reduces levels of organics to below

nondetection levels, which simplifies handling of treated soil and reduces

overall site cleanup levels.

• Demonstrated success. Incineration is a proven technology; the literature

on successful applications is extensive; many vendors offer incineration

services thereby driving down prices; and incineration equipment comes in

many sizes to fit the needs of any site.

304


• Regulatory requirements. EPA's Land Disposal Restrictions (LDRs) specify

incineration as a best demonstrated available technology (BOAT) for many

types of wastes, meaning that these wastes must be incinerated prior to

land disposal. Also, incineration results were used to set concentration-

based BOAT standards for many contaminants and incineration probably

has the best chance of continuing to meet these standards.

Incineration of TNT also has many disadvantages, including: • Safety concerns. The foremost safety concern stems from exposing

explosive materials to open flame, but this can be addressed through

routine safety measures. Secondarily, hazards also are associated with

erecting and operating the incinerator, which is a large piece of industrial

equipment with moving parts and high temperature areas. For any

explosives operation, DOD must approve the incineration work plan and

may require a hazards analysis and site safety plan.

• Noise. The incinerator is driven by up to a 400 to 500 hp fan, which can

generate substantial noise. Residents neighboring the Savanna Army Depot

and the Louisiana Army Ammunition Plant have complained about the noise

from incineration activity at these sites.

• Air emissions. Emissions from the stack may contain nitrous oxides (NOx);

volatile metals, such as lead; and products of incomplete combustion (PICs).

Modeling may need to be conducted to predict the distribution of emissions.

• Capital costs. The capital mobilization and demobilization costs associated

with incineration typically range from $1 to $2 million. Over time, for a large

facility, incineration becomes more cost effective. Figure 10 shows the range

of estimated incineration costs as a function of site size.

305


• Public perception. The public usually is wary of hazardous waste

incineration. There may be public concern that a mobile incinerator will be

established at a site and subsequently used to incinerate waste from other

sites. The public must be assured that, most often, mobile incinerators are

used only for single site cleanups and that incineration can be an effective

way to treat explosives waste.

• Required tests. Before an incinerator can be used to treat a large volume of

hazardous waste, it must pass a trial burn demonstrating that it can

achieve a 99. 99 percent organic destruction efficiency. If the soil at the site

does not contain enough contamination to demonstrate the 99.99 percent

destruction and removal efficiency, explosives might have to be shipped to

the site to spike the feed soil for the trial bum.

• Ash product. Incineration of combustible materials produces a volume

reduction, which can lead to higher concentrations of inorganic

contaminants in the ash product and create leachability problems.

Incineration of most contaminated soils produces only modest volume

reductions, so inorganics are not significantly concentrated in the treated

soil.

306


Figure 10 . Range of expected Incineration costs as a function

of total volume of soils treated (U.S. EPA, 1990).

• Materials handling. Some soils can be difficult to feed to the incinerator,

which has a small feed opening. Feeding sticky, high clay content soils can

be particularly difficult. These soils require pretreatment by aeration and

tilling to reduce moisture levels and decrease viscosity.

• Electricity and water requirements. Incineration operations require

large supplies of electricity and water, both of which can be limited in rural

areas.

Reference Cited U.S. EPA. 1990. U.S. Environmental Protection Agency. Engineering

bulletin: Mobile/transportable incineration treatment. EPA/540/2-90/014.

Office of Emergency and Remedial Response, Washington, DC. Office of

Research and Development, Cincinnati, Ohio.

307


2.2 Open Burn/Open Detonation

2.2.1 Background

Open burn OB and open detonation (OD) operations are conducted by DOD

and some private companies to destroy unserviceable, unstable, or unusable

munitions and explosive materials. In 08 operations, explosives or

munitions are destroyed by self-sustained combustion, which is ignited by

an external source, such as flame, heat, or a detonation wave (that does not

result in a detonation). In OD operations, detonable explosives and

munitions are destroyed by a detonation, which is initiated by the

detonation of a disposal charge. This section discusses types of wastes and

media that can be destroyed in 08/0D operations, 0810D procedures

currently being used, safety precautions associated with 0B/0D operations,

and a method recently developed for quantifying the level of hazardous

emissions from 0810D operations.

2.2.2 Treatable Wastes and Media 0B10D operations can destroy many types of explosives, pyrotechnics, and

propellants. 0B areas must be able to withstand accidental detonation of

any or all explosives being destroyed, unless the responsible 08 technicians

used recognize that the characteristics of the materials involved are such

that orderly burning without detonation can be ensured. Personnel with this

type of knowledge must be consulted before any attempt is made at 0B

disposal, especially if primary explosives are present in any quantity.

2.2.3 Operation 0B and 0D can be initiated either by electric or burning ignition systems. In

general, electric systems are preferable, because they provide better control

over the timing of the initiation. In an electric system, electric current heats

308


a bridge wire, which ignites a primary explosive or pyrotechnic, which, in

turn, ignites or detonates the material slated to be burned or detonated. If

necessary, safety fuzes, which consist of propellants wrapped in plastic

weather stripping, are used to initiate the burn or detonation.

The following design and procedural specifications for 08100 operations are

taken from paragraph 27-16d of the Army Materiel Command Explosives

Safety Manual (U.S. AMC, 1985) and paragraph 8-44 of Air Force Regulation

127-100 on explosives safety standards (U.S. Air Force, 1990). 08 of non

fragmenting explosives is conducted in burning trays, which are designed

without cracks or angular corners to prevent the buildup of explosive

residues. The depth of explosive material in a tray may not exceed 3 in., and

the net explosive weight of materials in a tray may not exceed 1,000 lb. The

distance between the trays for explosive devices is determined by hazards

analysis, but, in the absence of such analysis, trays are placed parallel to

one another and separated by at least 150 ft. These distances may vary for

08 of bare explosives or explosives-contaminated soils. When wet explosives

are being burned, trays may be lined with non explosive combustible

materials, such as scrap wood, to ensure complete combustion. An 08 tray

may not be inspected until 12 hours after the conclusion of the burn, and a

tray may not be reused until 24 hours after the conclusion of the burn or

until all ash and residues have been removed from the tray.

If there is a significant risk of fragmentation, OB operations are conducted

in open pits, which must be at least 4ft deep and have sloped sides to

prevent cave in. The length and width of the pit is determined by the

quantity of waste being burned. If necessary, non explosive combustible

materials and fuel may be added to ensure complete combustion of explosive

materials. As with burning trays, 08 pits may not be inspected until12

hours after the conclusion of the burn.

309


Facilities engineered specifically for OD operations are rare in practice.

Consequently, almost all OD operations are conducted in pits that are at

least 4 ft deep and covered with 2 ft of soil to minimize the risks associated

with fragmentation. Detonating cords, which are plastic cords filled with

RDX, are used to initiate buried disposal charges. Explosive components are

arranged in the pits to be in close contact with the disposal charge.

To prevent partial or incomplete destruction, site personnel must ensure

that the disposal charge is sufficiently powerful to propagate a detonation

throughout the explosive material. High brisance explosives and shaped

charges, which cut through metal casings, are very effective at propagating

detonations. If a misfire occurs, personnel are required to wait at least 30

minutes before inspecting the point of initiation. The misfire may be

inspected by no more than two personnel, who must follow specific

operating procedures.

After each detonation, the surrounding area is searched for unexploded

materials. Lumps of explosive material and unfuzed munitions are returned

to the detonation pit; fuzed ordnance or munitions that may have damaged

internal components are detonated in place.

2.2.4 Safety Precautions During 0B operations, munitions may rupture and produce fragments that

travel relatively short distances and explosive materials may detonate. OD

operations always produce dangerous overpressures and various types of

fragments, depending on the type of explosives being detonated. DOD has

developed specific safety precautions for 0B/0D operations, designed to

expose the fewest individuals to the least degree of hazard for the shortest

period of time. These precautions include minimum setbacks from 0B/0D

sites, provisions for the layout of 08/00 sites, optimum weather conditions

310


for conducting 0B/0D operations, and training requirements for 0B/0D

personnel.

Minimum Safety Distances

As a basic precaution, personnel are required to maintain a minimum

distance from the OB/OD operation. This distance depends on the type of

material being burned or detonated. The following minimum safety distances

are outlined in paragraph 8-44 of Air Force Regulation 127-100 on

explosives safety standards (U.S. Air Force, 1990). (Various Armed Services

manuals contain distances that provide varying degrees of safety for

exposure to the detonation.) For non fragmenting explosive material, the

minimum distance is either 1,250 ft or the explosive's actual maximum

debris and fragment throw range, if known. For fragment-producing

materials, the minimum distance is 2,500 ft. For bombs and projectiles with

a caliber greater than 5 in., the minimum distance is 4,000 ft. For heavier

case munitions, the minimum distance can be calculated by the following

formula:

D = 300 X (NEW) %

where D is the minimum distance and NEW is the net explosive weight of

the munitions in pounds. This distance is the radius in which most

hazardous fragments will fall.

Even at the minimum distances, personnel may be exposed to some

fragments. To minimize this exposure, the base plates and suspension lugs

of bombs and projectiles should be pointed away from personnel prior to

0B/0D.

311


Layout of the 0B/0D Site The following site layout specifications are taken from paragraphs 27-10 to

27 16 of the Army Materiel Command Explosives Safety Manual (U.S. AMC,

1985) and paragraph 8-44 of Air Force Regulation 127-100 on explosives

safety standards (U.S. Air Force, 1990). (Specifications from other Armed

Services manuals may vary.) The center of the OB/OD site typically consists

of several burning trays, burning pits, and detonation pits. All combustible

materials and loose stones are cleared within a 200-ft radius of the center of

the site. Personnel shelters are located a minimum of 300 ft from the site,

and holding areas for explosives awaiting detonation are located a minimum

of 1,250 ft from the site. Roadblocks are established at the perimeter of the

site to restrict entry during the operation.

Weather Conditions Weather conditions affect both the location and timing of OB/OD

operations. OB/OD operations are sited so that prevailing winds carry

sparks, flame, smoke, and toxic fumes away from neighboring facilities. The

optimum wind speed for an OB/OD is 4 to 15 mph, because winds at these

speeds tend not to change direction and, as a result, dissipate smoke

relatively rapidly. OB/OD operations are never conducted during sand,

snow, or electrical storms strong enough to produce static electricity, which

might cause premature detonation.

Personnel Training All OB/OD operations are supervised by a minimum of two experienced

personnel with training in general OB/OD safety procedures and the

handling of the specific materials being burned or detonated.

312


2.2.5 Emissions from OB/OD Operations Quantifying the level of pollutants in the emissions from OB/00 operations

is a difficult undertaking. Results from laboratory-scale studies translate

poorly to the field, because only very small quantities of explosives can be

tested. At this scale, the initiator or blasting cap contributes significantly to

the total amount of pollutants in the system. Emissions from field-scale

operations also are difficult to measure, because contaminants usually are

not distributed homogeneously within the plume, and the plume dissipates

quickly.

Personnel at Dugway Proving Ground in Utah recently developed a facility

that is large enough to provide reliable, field-scale results while allowing the

plume to be captured and analyzed by precise laboratory methods (Teer et

al., 1993). The facility is a 1,000-m 3 enclosed hemisphere known as the

bangbox. Preliminary studies conducted in the bangbox indicate that

OB/OD operations emit traces of organics and small quantities of soot in

addition to C02, N2, and H20.

Based on data generated from bang box studies, modeling was conducted to

estimate the health risks associated with emissions of benzo (a) pyrene from

OB/OD of TNT. The modeling assumed a cancer potency of 1.7 x 10·3 for

benzo (a) pyrene and an emission factor of 3.01 x 1o·6-the highest factor

calculated in any bangbox trial (and an order of magnitude higher than that

of the second highest trial). It was determined that 500 tons of TNT would

have to be destroyed in OB/OD operations to produce a 1 in 100,000 cancer

risk from benzo (a) pyrene emissions since the assumed em1ss1on factor

was very conservative, the health risks associated with emissions from

OB/OD operations probably are minimal (Teer et al., 1993). Future bangbox

studies will examine different waste compositions to target other specific

analytes, such as benzidine, that pose particularly acute threats to human

313


health.

References Cited Teer, R.G., R.E. Brown, and H.E. Sarvis. 1993. Status of RCRA

permitting of open burning and open detonation of explosive wastes.

Presented at Air and Waste Management Association Conference, 86th

Annual Meeting and Exposition. June 1993. Denver, Colorado.

U.S. Air Force. 1990. Air Force Regulation 127-100, Explosives Safety

Standards.

U.S. AMC. 1985. U.S. Army Materiel Command. Explosives Safety

Manual, AMC-R, 385-100.

2.3 Wet Air Oxidation

2.3.1 Background

Wet air oxidation is a high-temperature, high-pressure, liquid-phase

oxidation process. The technology is used in municipal wastewater

treatment, typically for treating dilute solutions of 5 to 10 percent solids or

organic matter. Wet air oxidation also has been tested but not used on a

large scale for treating explosives waste. In a typical wet air oxidation

system, contaminated slurries are pumped into a heat exchanger, where

they are heated to temperatures of 177 to 300°C, then into a reactor, where

they are treated at pressures of 1,000 to 1,800 psi.

2.3.2 Laboratory-Scale Applications

In 1982, the Army conducted a series of laboratory-scale studies on

technologies, including wet air oxidation, that formerly had been identified

314


as technically or economically infeasible for treating explosives waste. Wet

air oxidation was applied to lagoon slurries containing 10 percent explosive

contamination with added chemical catalysts. Although the technology was

found to be very effective for treating RDX, several disadvantages were

noted. First, the treatment produced hazardous byproducts from TNT.

Second, the technology had high capital costs. Third, lagoon slurries had to

be diluted prior to treatment. Fourth, gaseous effluents from the oxidation

process, such as carbon monoxide (CO), C02, and NOx, needed to be treated

by another technology. Finally, the laboratory-scale system was found to

have a 5 to 10 percent down time, because clays blocked the pump system

and heat exchange lines, and solids built up in some of the reactors. The

Army still is evaluating wet air oxidation treatment for TNT-contaminated

red water (U.S. ATHAMA, 1992).

2.3.3 Reference Cited

U.S. ATHAMA. 1992. U.S. Army Toxic and Hazardous Materials Agency.

Installation restoration and hazardous waste control technologies.

CETHA-TS-CR-92053. Aberdeen Proving Ground, Maryland.

2.4 Low Temperature Thermal Desorption

Low temperature thermal desorption (LTID) technology originally was

developed for treating aqueous slurries contaminated with volatile organic

compounds (VOCs). The technology also has been tested for treating

explosives-contaminated slurries.

In LTID, contaminated slurries are fed into the system, heated to 200 to

300°C by a hot oil heating chamber, and treated under elevated pressures.

Emissions from the system are treated in an afterburner.

315


The Army conducted a laboratory-scale study on low temperature thermal

desorption of explosives waste in 1982, as part of a series of studies on

technologies that previously had been demonstrated as unsuccessful for

treating explosives waste. LTTD was shown to achieve a 95 percent

destruction and removal efficiency (ORE) in 20 minutes, but two degradation

products-3,5-dinitroanaline and 3,5-dinitrophenol.were found to be

recalcitrant regardless of treatment time and temperature. The reactivity

and toxicity of these products were unknown at the time, meaning that the

product of thermal desorption might have to be treated as a hazardous

waste. Pilot-scale engineering and cost analyses of this technology have been

delayed, pending further testing of the degradation products.

3. Physical V Chemical Treatment Technologies

3.1 Ultraviolet Oxidation

3.1.1 Background

Ultraviolet (UV) oxidation has not been used extensively for remediating

water contaminated with explosives, because of the widespread use of

granular activated carbon (GAC) treatment. Nevertheless, UV oxidation can

be an effective treatment for explosives-contaminated water and, unlike

carbon treatment, actually destroys target compounds, rather than just

transferring them to a more easily disposable medium. This section

discusses the types of explosives-contaminated water that can be treated by

UV oxidation, examines some pilot-scale tests of UV oxidation, and provides

a detailed discussion of a treatability study of UV oxidation recently

conducted at Milan Army Ammunition Plant (AAP).

3.1.2 Treatable Wastes and Media

UV oxidation can be used to treat many types of organic explosives-

contaminated water, including process waters from the demilitarization of

316


munitions (pink water) and ground water contaminated from disposal of

these process waters.

3.1.3 Pilot-Scale Applications In 1981, the Army conducted a pilot-scale study of UV oxidation for treating

waters from the Kansas AAP contaminated with RDX (U.S. AARRDC, 1982).

RDX concentrations in the process water ranged from 0.8 to 21.0 mg/L. The

UV oxidation system consisted of thirty 40-watt, UV lamps, and an ozone

generator, which provided ozone to the treatment process. Treatment times

in this system ranged from 37 to 375 minutes at flow rates of 0.2 to 2.0

gpm. Final RDX concentrations in the effluent ranged from 0.1 to 1.7 mg/L,

which would not have met current regulatory criteria.

Similar studies have been conducted at Crane AAP, Iowa AAP, Holston AAP,

and Picatinny Arsenal. It is difficult to compare performance data from these

studies, however, because each study operated under different treatment

conditions. Some used 40-watt, low pressure, UV bulbs; others used 65-

watt, medium pressure, UV bulbs. Some amended the water with hydrogen

peroxide (H202); others did not. The studies also used different

concentrations and species of contaminant, different total residence times,

and different concentrations of ozone. In addition, some of the studies used

simulated pink water, which usually lacks many of the constituents of real

pink water.

UV oxidation is being considered at Picatinny Arsenal for the treatment of

ground water containing 6.0 ppb of RDX. The Waterways Experiment

Station in Vicksburg, Mississippi, currently is running a pilot test on the

proposed UV oxidation system and a parallel test of an activated carbon

system to compare the economic feasibility of the two.

317


3.1.4 Treatability Study at Milan AAP In the 1970s, Milan AAP was the site of munitions washout operations.

Process waters from these operations were placed in lagoons until the early-

1980s, when the waters were drained and the lagoons were capped. A

contaminated ground water plume is migrating from the site. The Army has

conducted a study to determine whether the contaminated ground water

could be treated by UV oxidation (U.S. ATHAMA, 1992). The treatability

study focused on how to optimize the performance of a full-scale UV

oxidation system, should UV oxidation be selected as the final remedial

technology at the site. The treatability study consisted of bench- and pilot-

scale tests.

3.1.4.1 Bench-Scale Tests Bench-scale UV oxidation tests were conducted on 15 gallons of

contaminated water from a site. The bench-scale system consisted of a 2.4-

L reactor with a single 4Q-watt UV bulb. Ozone was diffused through the

reactor at rates ranging from 2.8 to 15.0 (mg/L)/s, and a solution of 35

percent H202 by volume was used in the tests. The pH in the system ranged

from 4.0 to 8.5, and the pH of the water was found to drop due to the

production of organic acids during treatment. The concentration of all

explosives in the influent was 57,500 giL, with TNT, ADX,HMX, and tetryl

present in the highest concentrations. Residence times varied from 40 to

200 minutes per treatment batch. These tests indicated that UV radiation

degraded explosive contaminants and that longer UV exposure times yielded

better contaminant removals. H202 levels were found not to affect

contaminant degradation, and UV oxidation was found to be most effective

at pHs of 7 or greater. The level of 1,3,5-TNB, which is a product of the UV

oxidation of TNT, was the rate-limiting factor in each test; 1,3,5-TNB

concentrations actually increased after 40 minutes of UV exposure.

318


3.1.4.2 Pilot-Scale Tests The pilot-scale tests had two purposes: (1) to obtain design data for a full-

scale, 500-gpm, UV oxidation system; and (2) to estimate the cost of

operating a full-scale UV oxidation system.

Pilot-scale UV oxidation tests were conducted in a 65Q-gallon Ultrox P-650

system, consisting of six reaction chambers, each containing twelve 65-

watt, low-pressure, UV lamps, and a cooling system to prevent temperature

increases during long exposure times. The treatment system was operated in

recycle batch mode, meaning that each 650-gallon batch was recycled

through the system seven or eight times. The total concentration of

explosives in the influent was about 20,656 giL. and the pH of the water

was maintained at 7 to 11 during treatment. Tests were conducted at ozone

doses ranging from 1.11 to 3.33 (mg/L)/minute and with residence times

ranging from 40 to 210 minutes. The pilot-scale study indicated that UV

oxidation was most effective at a pH of 9 and an ozone dosage of 3.3

(mg/L)/minute. Residence times greater than 180 minutes coupled with

high ozone doses destroyed all of the explosives, including 1,3,5-TNB.

Biotoxicity tests indicated that the effluent from the UV oxidation system

was toxic, due to leaching of metals from bronze Impellers within the

equipment.

References Cited U.S.MAROC.1982. U.S. Army Armament Research and Development

Command.Ultraviolet ozone treatment of RDX (cyclonite) contaminated

wastewater. ARLCD-CR-83034. Dover, New Jersey.

319


3.2 Granular Activated Carbon

3.2.1 Background In the 1980s, the Army discontinued the practice of disposing of untreated

process waters from the production of munitions in open lagoons. Every

Army ammunition plant currently employs some type of granular activated

carbon system to treat process waters as they are generated. GAC is very

effective at removing a wide range of explosive contaminants from water.

GAC is a transfer technology only, however, and carbon adsorption media

can only be partially regenerated. This section outlines the types of

explosives-contaminated water that can be treated by GAC, discusses

isotherm tests, and looks at two studies of continuous flow column GAC

equipment conducted at Badger and Milan AAPs.

3.2.2 Treatable Wastes and Media GAC can be used to treat explosives-contaminated water, including process

waters from the manufacture and demilitarization of munitions (pink water)

and ground water contaminated from disposal of these process waters. GAC

is not used to treat red water produced during the manufacture of TNT.

3.2.3 Isotherm Tests Isotherm testing is a simple laboratory technique for initial screening of a

particular wastewater prior to GAC treatment. From 6 to 10 aliquots of

wastewater are measured Into containers that can be stirred or shaken for a

period of time. Into each container is introduced a known quantity of

pulverized carbon with a different amount of carbon for each container. After

stirring the mixture for a period of time, the mixture Is filtered and the

filtrate analyzed. The results of the tests indicate the relative adsorbability of

explosives, the adsorption capacity and exhaustion rate of the carbon, the

maximum degree of removal achievable, and whether there Is preferential

320


adsorption of ·any explosives.

3.2.4 Continuous Flow Column Studies The Army conducted pilot-scale studies of continuous flow column GAC

equipment at Badger AAP and Milan AAP. At both sites, GAC treatment was

found to be effective for removing every type of explosive from the water and

removing 2,4- and 2,6-DNT to below detection levels.

Badger AAP At Badger AAP, residues from the open burning of rocket paste

contaminated ground water beneath the burning ground with 2,4- and 2,6-

DNT. A pilot-scale GAC system consisting of eight, 4.25-in. diameter

columns was tested at the site. The first column, which was the test column,

operated in series with the second column, which was a back-up column

used to remove contaminants when contaminant breakthrough occurred in

the first column (i.e., when contaminants began to appear in the effluent

from the first column). The fill depth in each column varied from 2 to 4 ft, a

range that generally provides good data. Fill depths of greater than 4 ft

require as much as 70,000 to 80,000 gallons of water to be pumped through

the system to get breakthrough.

Based on the data obtained in an isotherm test, two types of commercially

available carbon filters were selected for pilot-scale testing at Badger AAP:

Calgon Filter Sorb 300 and Hydrodarco 4000. Flow rates were maintained at

0.3, 0.5, and 0.7 gpm, and a total of about 20,000 gallons of water were

used in each test. Influent concentrations ranged from 200 to 600 J.Lg/L of

2,4- and 2,6-DNT. A packed-column air stripper was used prior to GAC

treatment to remove trichloroethylene from the water. All laboratory

analyses were conducted using HPLC equipment, rather than GC.

321


The data obtained at Badger AAP were used to design a full-scale treatment

system that currently is being implemented.

Mllan AAP Ground water at Milan AAP was contaminated with seven types of

explosives. The GAC system tested at Milan AAP was similar to that tested at

Badger AAP, except that Atakim 830 carbon was substituted for the

Hydrodarco 4000. Tests were conducted at four flow rates ranging from 0.2

to 1.0 gpm, and as many as 56,000 gallons of water werused in each test.

The concentration of total explosives in the influent ranged from 600 to 900

J.Lg/L

The data from the pilot-scale GAC study are being evaluated concurrently

with data from a pilot-scale study of ultraviolet oxidation (see section

4.3.1.4).

3.3 Compressed Gas Cylinder Handling

3.3.1 Background

Compressed gas cylinders exhibit a wide a range of hazardous

characteristics. The chemicals contained within compressed gas cylinders

may be flammable, corrosive, pyrophoric, or poisonous, or they may be

oxidizers (definitions of these and other terms appear oxidizers (definitions of

these and other terms appear in Table 6). In addition, these chemicals are

contained within the cylinders by valves that are relatively small within the

cylinders by valves that are relatively small and vulnerable. Left unattended,

cylinders become more hazardous. Labels fall off and stenciling corrodes

making it difficult to identify the contents of the cylinders valves fail due to

corrosion, leaks develop, and emergency situations occur that demand

immediate attention. Many of the serious injuries and deaths attributed to

322


hazardous materials result from accidents involving liquefied or compressed

gases.

Technologies now are available for safely managing compressed gas

cylinders. New recycling and EPA-permitted treatment facilities are in

operation, and antiquated disposal procedures have been replaced by

sophisticated systems designed to protect the environment.

The Compressed Gas Association (CGA) advises EPA and the Department of

Transportation (DOT) on technical matters directly affecting the compressed

gas industry. CGA members include gas manufacturers, suppliers, and

distributors; chemical manufacturers; consultants’ and environmental

contractors. CGA provides to the public numerous pamphlets and videos

that are useful as guidance and technical resources.

This section discusses criteria for inspecting compressed cylinders; systems

for handling and transporting unstable cylinders; and some methods that

have proven unsuccessful for disposing of compressed cylinders. Appendix B

presents a case study of compressed gas cylinder handling at a Superfund

site.

3.3.2 Cylinder Inspections Before a compressed cylinder can be transported or treated, a detailed

inspection and evaluation of the cylinder, including its valve, must be

conducted Cylinders should be inspected for the following.

• Leaks. All valves and fittings must be tested for leaks with recognized

CGA procedures, which might include the use of a soap or suitable

solution to detect the escape of gas, or a hand –held direct reading

instrument.

323


Table 6. Definitions of Compressed Gas Cylinder Terms

Gas A formless fluid that fills the space of Its

enclosure and changes to the liquid or solid state

under increased pressure or decreased

temperature PPP.

Gas Pressure Gas pressure commonly is designated in pounds

per square Inch (psi); the analogous metric unit

Is the kilopascal (kPa); 1 psi equals 6.895 kPa.

The term psla refers to absolute pressure.

Absolute pre$8Ure Is based on a zero reference

point, a perfect vacuum. Measured from this

reference point, atmospheric pressure at sea

level Is 14.7 psi. Gauge pressure (pslg) has local

atmospheric pressure as a reference point. As

such, psla minus local atmospheric pressure

equals pslg.

Compressed Gas Any material or mixture contained at an

absolute pressure exceeding 40 psi at 700F or

exceeding 104 psi at 100°F; or any flammable

liquid having a vapor pressure exceeding 40 psi

at 1000F as determined by the American

National Standard Method of Testing for Vapor

Pressure of Petroleum Products, ANSVASTM

0323-79.

A gas contained at a pressure of 500 pslg (3448

kPa) or higher at 700F (21.1°C).

High Pressure

Gas

A gas that, Under the charged pressure, is

partially liquid at a temperature of 70°F.

324


Liquefied

Compressed Gas

A gas other than a gas in solution that, under

the charged pressure, is entirely gaseous at

700F.

Non liquefied

Compressed Gas

A gas other than a gas in solution that, under

the charged pressure, is entirely gaseous at

700F.

Inert Gases Inert gases, which Include argon, carbon dioxide,

helium, krypton, neon, nitrogen, and xenon, are

simple asphyxiates which can displace the

oxygen in air necessary to sustain life and thus

cause suffocation.

Corrosive

Gas/Liquid

A liquid or gas that destroys IMng tissue by

chemical action

Irritant A noncorrosive liquid or gas that, on Immediate

or prolonged contact, Induces a local

inflammatory reaction in living tissue

Poison A gas or liquid that creates an immediate hazard

to health when inhaled, ingested, or absorbed

through the skin, and can be fatal In low

concentrations

Pyrophoric Gas A gas that will Ignite spontaneously in dry or

moist air at a temperature of 130°F or below.

Oxidizer A gas or liquid that accelerates combustion and

that, on contact with combustible material, may

cause fire or explosion.

325


Pressure Relief

Device

A temperature- or pressure-activated device that

functions to prevent the rupturing of a charged

cylinder by releasing pressure above a

predetermined point.

• Dents Guidelines mandate that a dent at a weld be no deeper than 0.64

em. If a weld is not involved, dents may be no deeper than 10 percent of the

cylinder's greatest dimension. Dents are measured using a ruler and a dial

caliper.

• Gouges and cuts. Gouges and cuts reduce the thickness of cylinder walls.

Thickness gauging is required to determine whether cylinders with gouges or

cuts have structural weaknesses that constitute a safety hazard. Ultrasonic

thickness gauges often are used to measure cylinder wall thickness.

• Bulges. Bulging weakens a cylinder. Cylinders with bulges must be

evaluated by trained personnel to determine if the cylinders maintain their

structural integrity.

• Corrosion. While corrosion may be limited to surface rust, corroded

cylinders should be inspected using thickness gauging to evaluate the

integrity of their walls and to ensure that continued handling and

transportation of the cylinders will be safe.

• Fire damage. The following is evidence of fire damage: charring of paint or

protective coatings; burning or melting of fuze plugs, valves, and pressure.

relief devices; scarring or burning of metal surfaces; and disfiguring of the

cylinder. DOT regulations mandate that a cylinder showing evidence of fire

damage may not be placed into service or transported until it has been

reconditioned, unless a proper inspection reveals that the cylinder is only

326


discolored or smudged and is in serviceable condition.

• Improper backfilling. Cylinders sometimes are backfilled with materials

that they were not designed to contain. This can cause many problems,

including corrosion of the interior walls.

• Retrofitted valves. Gas cylinders occasionally are retrofitted with valves or

fittings that are not designed for the cylinder or its contents. Proper

inspections should reveal if these conditions exist.

Cylinder labels and stenciling also should be inspected to determine the

contents of the cylinder. A cylinder is considered to be "unknown" under any

of the following circumstances:

• The cylinder has no original label or stenciling identifying its contents

The contents of an unknown cylinder must be identified through laboratory

analytical procedures, not by examining the cylinder's color, valve outlet, or

other markings. Applicable analytical procedures include mass

spectrometry, as well as Fourier transform infrared (FTIR) and GC. An

unknown cylinder cannot be shipped off site for disposal or recycling or

treated on site until its contents have been identified. An unknown cylinder

that is shipped off site for laboratory analysis must be given a tentative

shipping description (Hazard Class) as defined in 49 CFR 172.101{c){11).

3.3.3 Handling Techniques DOT regulations and CGA guidelines ensure that safe handling and

transportation procedures are being followed. Generators of compressed

cylinders must use hazardous waste manifests and licensed waste

transporters. Each generator also must have an EPA identification number

327


as a small or large generator unless exempt.

Two handling procedures are available: hot tapping/controlled access and

overpacking.

Hot Tapping/Controlled Access The management of a cylinder with an inoperable valve requires state-of-the-

art hot-tapping equipment, which performs one of three operations:

• Drilling into the cylinder at a predetermined location, thereby allowing the

contents of the defective cylinder to flow into a primary containment vessel.

• Shearing the valve from the cylinder or shearing the cylinder in half and

capturing the gas or liquid in a primary containment vessel.

• Drilling into the cylinder while maintaining a tight seal and introducing a

new valve into the cylinder without releasing gas into a primary containment

system. Secondary containment may be used during this procedure

depending on the known or suspected gas involved.

The first two operations are followed either by onsite treatment of the gas in

the primary containment vessel or the recontainerization of this gas into a

DOT-approved cylinder for offsite treatment or recycling. All three operations

are identified as the current BOATs for managing compressed cylinders with

inoperable valves and essentially are the only methods in use today.

Overpacking Salvage cylinder overpacks can be used to contain a compressed gas

cylinder that is being transported to an offsite facility or is leaking. An

328


overpack is an oversized cylinder fabricated to accept a smaller cylinder into

itself. Once closed, the overpack contains any release from the defective

cylinder. Valves and pressure gauges on the overpack allow its internal

pressure to be monitored so that the defective cylinder can be removed

safely. Cylinder over packs are similar to the 85 or 110-gallon salvage over

packs used to transport 55-gallon drums.

3.3.4 Treatment, Disposal, and Recycling Options Compressed cylinders may be sent to a treatment or recycling facility, or

treated on site.

Offsite Treatment Discarded and abandoned cylinders must be disposed of in EPA-permitted

treatment, storage, and disposal facilities (TSDFs}. TSDFs use two systems

to treat the contents of cylinders. In one system, vapor or gas is · drawn

from the cylinder through a manifold directly into an incinerator. In the

other system, vapor or gas is drawn from the cylinder into a chemical

scrubbing medium. In both systems, the remaining empty cylinder then is

purged, cleaned, devalved, and land filled or recovered for scrap.

Recycling If the contents of a cylinder are known, generators may send cylinders to a

recycling facility. At the recycling facility, the cylinder's contents are

removed from the cylinder through a manifold system and introduced back

into the manufacturing process as a raw material. The empty cylinder then

is either cleaned, devalved, and sent for stfkl scrap recycling or, if in suitable

condition, cleaned, painted, restamped, and hydrostatically tested for

reentry into the market as a filled and usable cylinder.

329


Onsite Treatment In onsite treatment, cylinders of liquefied or compressed gases are treated,

neutralized, or otherwise disposed of at their location, without the use of an

offsite TSDF or recycling facility. Onsite treatment involves chemical

scrubbing, incineration, flaring, or controlled atmospheric venting of

cylinder contents. Onsite treatment may be used under any of the following

conditions:

• There are no available offsite management options.

• The cylinder is in a non-DOT transportable condition and cannot be

• removed from the site or recontainerized into another vessel.

• The cylinder is leaking and must be treated expeditiously.

• Regulatory authorities mandate onsite treatment only. Onsite treatment

of cylinders containing RCRA hazardous substances requires permit

approval by federal or local authorities.

3.3.5 Unsuccessful Treatment Approaches

Several techniques have been tested for the treatment and recycling of

compressed gas cylinders. Most of these techniques are no longer used

because they do not adequately protect human health or the environment.

Nevertheless, these methods occasionally are used by contractors or

regulators unaware of the current BOATs.

Detonation (Uncontrolled Release) A pressurized cylinder can be destroyed by the detonation of a disposal

charge that breaches the cylinder body or its valve. Chemicals contained in

the cylinder also might be destroyed during the explosion. In the past, this

330


practice was used to dispose of cylinders with inoperable valves, for which

detonation was more cost effective than more sophisticated treatments or

recycling. Today, detonation is considered to have several drawbacks,

including fragmentation from the cylinder body. In addition, the cylinder can

rocket away from the detonation site.

Projectile Method (Uncontrolled Release) In the projectile method, a high-caliber projectile is fired from a rifle into a

cylinder, releasing gas· from the cylinder through the vent holes produced by

the impact. As with detonation, this procedure releases untreated gases to

the environment. In addition, the cylinder may rocket from the site or

detonate.

Valve Release (Controlled or Uncontrolled Release) In valve release, the cylinder's valve is opened, and the cylinder is allowed to

vent until empty. Like detonation and the projectile method, this procedure

releases potentially toxic or ozone-depleting substances untreated into the

environment. Valve release should be used only for atmospheric gases and

must be employed using both a regulator to control flow and a stack to

prevent the formation of an oxygen-deficient work area for the operator.

3.3.6 Reference Cited CGA. 1981. Compressed Gas Association. Handbook of Compressed Gases,

Third Edition.

3.4 Reactive Chemical Handling

3.4.1 Picric Acid

331


Background Picric acid is a yellow crystalline substance that was discovered in 1771 by

the British chemist Peter Woulfe. Picric acid's name is derived from the

Greek word pikros, meaning bitter, due to the intensely bitter and persistent

taste of its yellow aqueous solution. In the past, this strong acid was used as

a fast dye for silk and wool and in aqueous solutions to reduce the pain of

burns and scalds.

When dry, picric acid has explosive characteristics similar to those of TNT.

Table 7 summarizes the explosive characteristics of picric acid. The first

experiments to use picric acid as an explosive bursting charge were

conducted in the town of Lydd, England, in 1885, and picric acid was

adopted by the British as a military explosive in 1888 under the name

Lyddite. Since that time picric acid has been used by many countries as a

bursting charge under the names Shimose (Japan), Granatfullung 88

(Germany), Pertite (Italy), Melinite (France), and trinitrophenol (United

States). Today, the use of picric acid as a military explosive has been largely

discontinued, because picric acid was found to have several disadvantages:

• It is prone to sympathetic detonation, wherein the detonation of a nearby

charge would cause it to detonate without a priming charge

Table 7. Explosive Properties of Picric Acid

Gross formula C2HaNaCr

Melting point 122.5°C

Auto ignition temperature 572°F

Molecular weight 229.1

Oxygen balance -45.4%

Heat of Explosion 1,080 kcaVkg

332


.

Density 1.767 g/cm3

Lead block test 315 cmt10 g

Detonation velocity (when confined) 7350 mts

Deflagration point 570°F (3000C)

CAS 88-89-1

United Nations(dry or wetted with less than 30 percent water by weight) Class No

0154

United Nations (with 30 percent or more water by weight) Class No

1344

• When it contacts metals, such as mer(fury, copper, lead, or zinc, it forms

explosive salts that are sensitive to friction, heat, and impact. Special

precautions also are required if picric acid falls on concrete floors, because

this causes the formation of sensitive calcium salts.

• Metal and cement shells that contain picric acid must be sealed with a

protective varnish to prevent contact between the picric acid and the shell

lining.

In addition to its explosive properties, picric acid also is highly toxic. Like

many trinitro compounds, picric acid is absorbed through the skin and

through inhalation. Acute picric acid exposure can depress the central

nervous system and reduce the body's ability to carry oxygen through the

blood stream. Prolonged exposure may result in chronic kidney and liver

damage. Percutaneous absorption may cause vomiting, nausea, abdominal

pain, staining of the skin, convulsions, or death. The Occupational Safety

and Health Administration's (OSHA's) permissible exposure level (PEL) for

picric acid is a time weighted average (TWA) of 100 J.tg/m3, with a "skin"

notation to indicate the possibility of dermal absorption, and the American

Conference of Governmental Industrial Hygienists (ACGIH) recommends a

threshold limit value (TLV)-TWA of 0.1 rn9'm3

333


Proper personal protective equipment, such as gloves, respirators, and self-

contained breathing apparatus {SCBA), including Level 8 attire, should be

worn when handling picric acid outside of an established laboratory

environment. The use of advanced personal protective equipment should be

commensurate with the activity of the individual. Individuals responding to

a spill of picric acid or handling spilled material should wear SCBA,

including Level B attire. On the other hand, chemists and technicians

working in a laboratory setting should wear gloves and work under a fume

hood to ensure safe handling of picric acid.

The following sections discuss handling procedures and disposal options for

picric acid.

Handling Procedures Picric acid is soluble in water and various solvents. When hydrated, picric

acid becomes non explosive and is safe to transport and incinerate in offsite

facilities. Nevertheless, dry picric acid residues on the outer surface of

containers as well as in threaded container closures present a significant

friction-sensitive hazard. This hazard prompts many generators to use

remote handling equipment when opening containers of picric acid, a

technique usually reserved for containers of dry (desiccated) material.

DOT classifies solutions of picric acid containing less than 10 percent water

as explosive materials and solutions of picric acid containing greater than

10 percent water as flammable solids. This regulatory distinction dictates

the mechanics of preparing picric acid for shipment, such as packaging,

labeling, and adhering to manifest documentation requirements. It has little

relevance to the facility receiving the picric acid for treatment.

334


Disposal Options Incineration currently is the BOAT for the destruction of picric acid {40 CFR

261.23(a)(6)). Incineration facilities have varying acceptance criteria

governing the concentrations of picric acid in water; some require picric acid

concentrations to be as low as 1 percent, others will accept solutions with

picric acid concentrations as high as 50 percent.

Because of picric acid's history as a commercial and military explosive,

many civilian police bomb squads and military EOD units formerly accepted

picric acid for disposal through controlled detonation. Detonation was the

disposal method of choice until the mid-1980s, when it was discovered that

picric acid was not, in fact, destroyed by open air detonation but simply

dispersed by the explosion of the disposal charge. The resulting dispersal of

picric acid over the detonation site caused finely divided particles of the

substance to enter the surface strata. Testing of surface samples obtained

from picric acid detonation sites often showed trace quantities of the

compound unaffected by the detonation. In addition, slow motion video of

several picric acid detonations clearly showed a heavy yellow smoke of finely

divided picric acid particles, which negatively affected localized air quality.

3.4.2 Peroxides

Background Peroxides are shock-sensitive compounds that can explode if subjected to

mechanical shock, intense light, rapid changes in temperature, or heat. In

some cases, peroxides also can explode through a spontaneous reaction.

Peroxide structures are particularly dangerous when present in organic

solvents, which often are highly flammable. In testing conducted in the mid-

1980s, the detonation of a sample of a hard peroxide crystal destroyed a 4-lb

lead Trauzl block, a test used to determine whether or not a substance is

335


explosive. Similarly, a controlled detonation of pure peroxide crystals

discovered in an evaporated bottle of isopropyl ether demonstrated that

peroxide explosions produce high levels of destructive fragments.

The following sections discuss the formation of peroxide compounds,

procedures for inspecting and testing for the presence of peroxides, and

options for treating and disposing of peroxides

Peroxide Formation/Inhibition Peroxides form In organic solvents as a result of autoxidation. Common

peroxide-forming solvents can be divided into the following groups:

• Ethers, including open chain and cyclic ethers, acetals, and ketals (e.g.,

ethyl ether, isopropyl ether).

• Hydrocarbons with allylic, benzylic, or proparglic hydrogen (e.g., cumene,

cyclohexane).

• Conjugated dienes, eneynes, and diynes (e.g., butadiene, furans). Most of these solvents are purchased from the manufacturer with an added

inhibitor, such as hydroquinone or tart-butyl catechol, which chemically

inhibits peroxide formation.

Autoxidation in solvents is facilitated by three factors: • Exposure to oxygen • Exposure to light, including sunlight • Storage time

336


Oxygen is a necessary ingredient for peroxide formation. A cap or bung left

off a container or drum, or a loose fitting seal, may supply sufficient oxygen

to support peroxide formation by eliminating the inhibitor and supporting

the initiation of the autoxidation process. Light, including sunlight, also

promotes the elimination of inhibitors and stimulates the autoxidation

process. Light, however, cannot promote the autoxidation process unless

sufficient oxygen is present in the container. Once formed, peroxides can, in

direct sunlight, undergo autodetonation. Storage time simply allows

peroxides to develop and form structures. Since autoxidation is a self-

sustaining reaction, the rate of peroxide formation increases with time.

More than a decade ago, the National Safety Council (NSC) published easy-

to-follow laboratory guidelines (NSC, 1982) for preventing the formation of

peroxides in solvents; unfortunately, although these guidelines can be

obtained easily from the NSC, they seldom are followed. The formation of

peroxides in an organic solvent can be inhibited in two ways: (1) by adding

an inhibiting compound to the solvent, or (2) by purging the oxygen from the

free space in the solvent container. Chemical manufacturers add inhibitors

to almost all solvents, except those used for HPLC. These are specifically

manufactured without inhibitors, because inhibitors interfere with the UV

detection process. Inhibitors added by the manufacturer, however, are

effective only during shipping and marketing of the product; once the

solvent container is opened and exposed to oxygen, the autoxidation process

begins. Oxygen is the rate-limiting factor in peroxide formation. Replacing

oxygen in the free space of a solvent container with an inert gas, such as

nitrogen or argon, prevents autoxidation of the solvent. This method has

proven very successful in inhibiting peroxide formation.

Peroxide Detection Visual Inspections. Solvents stored in glass bottles can be inspected for

337


peroxides visually. Bottles containing organic solvents usually are made

from amber or brown glass, so a soft light source, such as a flashlight, is

helpful for lighting the interior of the bottle to allow a good view of the liquid.

The light source should be placed behind or to the side of the bottle,

because light shone directly on the glass creates reflections that obstruct

inspection of the bottle's contents.

During the visual inspection, the investigator should look for two signs of

peroxide contamination:

• Gross contamination. Hard crystal formations in the form of chips, ice-like

structures, crystals, or solid masses, or an obscure cloudy medium.

• Contamination. Wisp-like structures floating in a clear liquid suspension. Peroxide formation may be present anywhere in the container, including the

bottom of the container, the side walls of tile glass, the threaded cap, or even

the outside of the container. Peroxide formation in ppm concentrations may

not be visually observable and must be identified through appropriate

testing procedures.

Metal cans and drums cannot be inspected visually and must be opened to

allow appropriate testing. Opening containers is a delicate procedure due to

the possibility of peroxide accumulation in the cap threads. While peroxide

contamination tends to occur less frequently in the cap area than in other

container areas, metal cans and drums should be opened only by trained

individuals, and the application of remote opening equipment should be

considered.

Metal containers are believed to accelerate the rate of peroxide formation.

The scientific documentation supporting this belief, however, is largely

338


anecdotal.

Laboratory Testing. Several methods are employed to test for the presence of

peroxides. The following two tests are among the more common:

• Commercially available peroxide test strips. These test strips provide

quantitative results and are simple to use. The test strip is saturated with a

representative sample of the liquid in question. A section of the strip

changes color if peroxids are present; this color then is compared to a graph,

which indicates the peroxide concentration in ppm. Test strips typically

register as high as 100 ppm.

• Potassium iodide (KI) test. In this test, 100 mg of potassium iodide is

dissolved in 1mL of glacial acetic · acid. Then 1 mL of suspect solvent is

added. A pale yellow color indicates a low concentration of peroxides; a

bright yellow or brown color indicates a higher concentration of peroxides.

This is the preferred method for testing di-isopropyl ether.

A peroxide test should be performed each time material is removed from a

container. If the material is removed on a daily basis, tests should be done

every other day. Containers of peroxide-forming compounds should be

marked with the date the container was first received and first opened, the

results of the first peroxide test, and the results of the last peroxide test

before disposal. Tabies 4-8, 4-9, and 4-10 show the testing requirements for

common peroxidizable compounds during storage, as well as handling and

testing requirements for these compounds while in use.

The results of peroxide testing dictate how the material should be handled.

The following are the general levels of risk associated with various

concentrations of peroxides:

339


• Between 3 and 30 ppm. Expired compounds testing within this range pose

little or no threat of violent reaction on the given test date. For compounds

testing in this range, the investigator should consider adding fresh inhibitor

to retard the autoxidation process, and the container should be tightly

sealed to prevent air and light exposure.

• Between 30 and 80 ppm. Expired or mismanaged compounds that test

within this range may pose a threat to operations in the laboratory or

facility. Several major exothermic reactions have occurred during the

reduction of peroxides within this range.

Table 8. Compounds That May Form Peroxides During Storage8

Compound Test Cycle in Storage

Special Handling and Tests While In Use

Isopropyl either Every 3 months Consume or discard within 3

days of opening these

containers.

Divinyl acetylene Every 3 months Consume or discard within 3

days of opening these

containers.

Vinyldene

chloride

Every 3 months

Potassium metal Every 3 months Avoid oil/hydrocarbons, if KO2

is present.

Sodium amide Every 3 months

• Greater than 80 ppm. Any solvent testing in excess of the maximum

quantifiable limits of standard peroxide test strips must be considered

340


potentially shock sensitive.

Treatment and Disposal Options Deactivation. Most, if not all, peroxide-forming chemicals are regulated as

hazardous wastes. The BOAT for peroxides is deactivation to eliminate the

ignitability characteristic (55 FR 22546). Technologies that may be used to

deactivate peroxide-formers (classified as 0001 oxidizers) include chemical

oxidation, chemical reduction, incineration, and recovery. Any of these

technologies is acceptable, provided it eliminates the ignitability

characteristic. To

Be accepted by an offsite, EPA-permitted, treatment and disposal facility,

peroxide containers that no longer are in use must be peroxide free and

present no explosive hazard.

Stabilization/Reduction. Peroxides within a container can be chemically

stabilized. The following describes one chemical procedure that has been

used successfully to stabilize peroxides. (The reader is cautioned that any

procedure used to handle a sensitive chemical or eliminate peroxides should

be undertaken only by very experienced personnel who understand the

potential for uncontrolled exothermic reactions during the procedure.) The

solvent container is accessed through its cap by a remotely operated

titanium-coated drill. A Teflon catheter then is inserted through the access

point to draw a 1-cm3 sample of solvent for testing. Three standard peroxide

test strips are used to measure the sample’s peroxide concentration. All

negative indications are verified by adding a drop of sample solvent to a 10

percent potassium iodide solution for colorimetric evaluation.

If the container is found to contain peroxides, a solution of ferrous

ammonium sulfate is injected into the container. This produces an

341


oxidation-reduction reaction that, while often very exothermic, has proven to

be successful in eliminating peroxides have been dissolved and peroxide

tests are shown to be negative. Hydroquinone then is added to stabilize the

container and guard against an immediate recurrence of peroxidation.

Finally, the container is resealed with a silicone sealant and standard

sealing ape and placed in a designated safe area pending offsite disposal.

Open Detonation. Open air detonation or burning of peroxide-forming

compounds formerly was used by police bomb squads and government

explosive technicians in an effort to assist the private sector. This practice

was found to have two major disadvantages.

• Potentially shock-sensitive materials were subjected to movement prior to

disposal

• The compound in question was dispersed untreated into the surrounding

air and soil.

3.4.3 Ethers Ethers are organic compounds with common uses as both medical

anesthesia and solvents. Simple ethers may be highly volatile and have

flammable and potentially explosive characteristics. The most commonly

used ether is diethyl ether-a clear, colorless liquid that vaporizes readily at

room temperature and is highly flammable. Diethyl ether's flashpoint is -

45°C and its flammable range extends from 1.85 to 48 percent by volume.

Aside from their flammability, liquid ethers also can contain organic

peroxides produced by a reaction between the ether and atmospheric oxygen

(Meyer, 1989).

342


Table 9. Compounds That Readily Form Peroxides In Storage through

Evaporation or Distillation

Compound Test Cycle in

Storage

Special Handling and Tests

While in Use

Diethyl ether Every 12

months

HPLC grades of these

compounds are normally

packaged without peroxide

inhibitors. These uninhibited

containers should be stored in

an inert (oxygen-free)

atmosphere and tested at 3-

month Intervals. Limit these

containers to sizes appropriate

to the application In order to

prevent repeated exposures.

Tetrahydrofuran Every 12

months

Every 3 months, if uninhibited

Dioxane Every 12

months


Acetal Every 12

months


Methyl-isobutyl-

ketone

(Isopropylacetone)

Every 12

months


Ethylene glycol

dimethyl ether

Every 12

months


Vinyl ethers Every 12

months


Dicyclopentadiene Every 12

months

343


Isoprene Every 12

months


Organometallics

(Grignard Reagents)

Every 12

months

Every 3 months, If uninhibited.

Do not store in a cold room.

These highly reactive

compounds accumulate

peroxide at low temperatures

because the peroxide

degradation rate is slowed

relative to the peroxide

formation rate.

Diacytylene Every 12

months


Methyl acetylene Every 12

months


Cumene Every 12

months


Tetrahydronaphthale

ne

Every 3 months Every 12 months

Cyclohexen Every 12

months


Methylcyclopentene Every 3 months Every 12 months

t-Butyl alcohol Every 12

months


344


Acetaldehyde Every 12

months

Anhydrous acetaldehyde will

autoxldlze at OOC or below

under ultraviolet light catalysis

to form peracetic acid, which

may react with more

acetaldehyde to produce the

explosive acetaldehyde

monoperacetate

3.4.4 References Cited DOD. No date. Department of Defense. Publication, TM9-13Q0-214ff0 11A

1-34.

Material Safety Data Sheet. 1985. MSDS #534.Gunium, Publishing

Company. Schenectady, New York.

Meyer, E. 1989. The Chemistry of Hazardous Materials, Second Edition.

Prentice Hall, Inc. Englewood Cliffs, New Jersey. 394-395.

Meyer, R. 1981. Explosives, Second Revised and Expanded Edition.

Weinheim Publications. Deerfield, Florida.

NSC. 1982. National Safety Council. Industrial Safety Data Sheet I-655

Rev.82, Stock No. 123.09. Chicago.

345


Table 10. Compounds That Pose Hazards Due to

Peroxide Initiation of Polymerization.

Compound Cycle In Storage Special Handling and Tests

While In Use8

Butadiene Every 12 months Every 3 months if stored as

liquid

Styrene Every 12 months Every 3 months if stored as

liquid

Tetrafluoroethylen

e

Every 12 months Every 3 months if stored as

liquid

Vinyl acetylene Every 12 months Every 3 months if stored as

liquid

Chlorobutadlene

(Chloroprene)

Every 12 months Every 3 months if stored as

liquid

Vinyl pyridine Every 12 months Every 3 months if stored as

liquid

Vinyl chloride Every 12 months Every 3 months if stored as

liquid

• When stored In the liquid state, the peroxide-forming potential

dra-matically Increases.

NSC. 1979. National Safety Council. Data Sheet 10351-79. 3.5 Reuse/Recycle Options for Propellants and Explosives

3.5.1 Background Recovery and reuse technologies for energetic materials, including both

346


explosives and propellants, are available in production-scale facilities

capable of handling quantities greater than 100,000 lb. Recovery/ reuse

options should be considered at explosives waste sites for several reasons.

First, new recovery methods and potential uses for reclaimed explosive

materials are rapidly developing. Second, recovery/reuse options reduce

overall remediation costs by eliminating destruction costs and allowing the

value of reclaimed materials to be recovered. Finally, EPA's treatment

hierarchy, which is based on environmental considerations, favors

recovery/reuse options over destruction technologies.

This section describes the types of explosives waste and media that can be

recovered/reused, the available recovery/reuse technologies, some leading

recovery/reuse companies and institutions, potential applications for

recovered energetic materials, and advantages and limitations of

recovery/reuse technologies.

3.5.2 Treatable Wastes and Media A detailed knowledge of energetic materials is necessary to minimize the

risks associated with recovery/reuse and to develop a suitable

recovery/reuse plan. For a detailed description of energetic materials, refer

to section 1.2.2. In addition to pure energetic materials, munitions and

rocket motors and explosives-contaminated soils and sludges also can be

recovered/reused.

Energetic Materials Propellants that contain combustion modifiers, such as lead compounds,

are difficult to reuse because of the stringent controls on lead emissions.

Reuse of these propellants as commercial explosive additives is rarely an

option. Primary explosives and initiating explosives, such as lead azide,

generally are not candidates for recovery/reuse due to their high sensitivity.

347


Very little has been done on recovering pyrotechnics, probably due to their

highly variable compositions, their sensitivity, and the low value of their

ingredients. This section does not discuss pyrotechnics in detail.

Munitions and Rocket Motors Recovery/reuse methods generally are applied only to munitions and rocket

motors that have documented histories, including documentation of how the

item was manufactured, its energetic fill, and its inert parts. In addition, the

recovered item must be present in sufficient quantities for the

recovery/reuse process to be economical. These criteria limit the types of

munitions for which recovery/reuse is feasible. Bunkered ordnance

discovered during a remediation effort may have a documented history and

sufficient quantity. Ordnance encountered during range cleanup often is in

various stages of physical disrepair and does not meet the criteria for

recovery/reuse.

Explosives-Contaminated Soils and Sludges Soils and sludges contaminated with energetic materials present handling

problems during recovery and reuse operations. AEC has established a

guideline that soils containing greater than 10 percent energetic materials

by weight should be considered explosive during handling and

transportation. As a general rule, soils and sludges containing less than 10

percent energetic materials by weight pass AEC's nonreactivity tests.

Reuse/recycle options are more feasible for contaminated soils and sludges

meeting the nonreactivity criteria, because they can be removed,

transported, and handled using conventional equipment, which could

provide a substantial cost savings. Unless diluted with fuel, the material

extracted from contaminated soils and sludges most likely must be treated

as an energetic material.

348


3.5.3 Operation and Maintenance Recovered munitions and rocket motors either can be reused "as is," or the

energetic materials can be recovered from these items and reused or

recycled. If an ordnance item is to be reused as is, it is inspected, recrated,

and sold as reconditioned ordnance. Energetic materials recovered from

munitions can be reused in their original application, or specific ingredients

can be extracted and recycled into energetic materials. Explosives-

contaminated soils and sludges can be recovered for the fuel value of their

contaminants. Table 11 provides an overview of the potential uses for

recovered munitions and energetic materials.

Energetic Material Extraction One of the more technically challenging aspects of energetic material

recovery/reuse is the separation of energetic components from inert

components. For Hazard Class 1.3 composite propellant rocket motors and

items containing plastic-bonded explosives, high-pressure water washout

(hydromining) and machining are the established separation methods. Other

washout methods that have been demonstrated at bench scale include

liquid nitrogen and liquid ammonia washout at high pressure. The latter two

methods are scheduled to be demonstrated at prototype scale in the next

year under DOD's Large Rocket Motor Demilitarization Program.

Thiokol Corporation's washout facility near Brigham City, Utah, which has

been used mainly for rocket motor case and warhead body recovery, utilizes

hydromining technology (see Figure 11). In operation since the mid-1960s,

this facility has been used to remove over 17 million pounds of propellant

and recover over 3,000 motor cases. Another major hydromining facility in

the United States is the Aerojet Solid Propulsion Company facility in

Sacramento, California.

349


Propellant machining is used in final grain shaping to provide desired

ballistics (i.e., propellant burn back pattern) and recover missile motor

cases. All of the propulsion companies have employed this method, in which

a drill, boring mill, or special tooling is used to cut propellants from motors

under carefully controlled conditions.

Recovery methods for TNT-based explosives are well established and involve

melt and steam-out processes. These processes liquify TNT so that it can be

poured out of the munition. TNT melt and steam-out facilities are located at

several Army ammunition plants and depots, and at the Western

Demilitarization Facility in Hawthorne, Nevada a Key: AI = aluminum; AN =

ammonium nitrate; AP = ammonium perchlorate; CEA = commercial

explosive additive; HMX = high melting explosives; IR = ingredient recovery

(most likely ingredient to be recovered); Mg = magnesium; MgN03 =

magnesium nitrate; NaN03 = sodium nitrate; NC = nitrocellulose; NG =

nitroglycerine; NQ = nitroguanidine; Original = original intended use; RDX =

royal demolition explosives, or cyclonite; TNT= trinitrotoluene.

Table 11. Overview of Items and Usesa

Item Energetic

Material

Typical

Ingredients

Potential

Reuse

Comments

Rocket

Motor

Hazard

Class 1.3

Propellant

Binder/AP/A

l

Original,

CEA,

IR(AP)

Original,

CEA,

IR(HMX)

CEA & AP recovery

have been

demonstrated full

scale, special additives

such as lead oxide may

require destruction

methods

350


Hazard

Class 1.1

Propellant

NG/NC/HM

X/AP/AV

Binder

CEA & HMX recovery

demonstrated

prototype scale

Gun

Propella

nt

Hazard

Class 1.1

Propellant

NC/NG/NQ Original,

CEA,

IR(NC)

CEA demonstrated full

scale

Bombs Explosive TNT, Al, AN,

RDX

Original,

CEA

CEA & IR

Demonstrated full

scale

Warhea

ds

Explosive Binder,

HMX, RDX,

Al

Original,

CEA,

IR(HMX)

CEA demonstrated

prototype scale, IR

(HMX) bench scale

Bomblet

s

Explosive Binder,

HMX, RDX,

Al

Original,

CEA,

IR(HMX)

Recovery demonstrated

bench scale

Illumina

ting

Flare

Pyrotechni

c

Binder,

NaNO3, Mg

Original IR not demonstrated

Signal

Flare

Pyrotechni

c

Binder,

Metal

Nitrates, Mg

Original,

IR

(MgNO3)

IR not demonstrated

Mfg.

Waste

Propellant

s,

Explosives,

Pyrotechni

cs

Any of the

above

CEA, IR

(HMX, AP)

Composition and

ingredient resuse

demonstrated bench to

full scale, sludges not

demonstrated

Another means of disassembly and separation of munitions components is

called "reverse engineering." Several systems have been built to reverse

engineer munitions. These systems, which are called ammunition peculiar

equipment (APE), work well for specific munitions but do not adapt easily to

351


varying configurations. Reverse engineering methods disassemble munitions

down to the casing that contains the energetic material. Standard methods

for further reducing the size of the munition include wet saw cutting and

high pressure water jet.

A size reduction method called Cryofracture has been developed by General

Atomics Corporation. It involves cooling munitions to liquid nitrogen

temperatures and crushing them in a hydraulic press. After being processed

in this manner, the ordnance can be fed to a specially designed incinerator.

Several separation methods, including solvent, density, magnetic, and melt

and steam-out separation processes, could be applied to recover the

energetic material after fracturing. The types of items that have been

successfully Cryofractured are shown in Table 12. Because Cryofracture can

handle multiple versus individual munitions, the technology might be most

useful in separating inert and live materials in smaller items, such as

bomblets, for which reverse engineering is less practical.

Reuse of Energetic Materials Once energetic materials have been separated from inert materials, reuse is

more straightforward, and many large-scale reuse applications have been

demonstrated. Ordnance items and rockets routinely are reinspected and

used for training or similar applications. Surplus explosives also have been

purchased from the government by commercial explosives companies since

before World War II. In addition, the patent literature reveals many examples

of smokeless powders, TNT, tetryl, HMX, and RDX being added as

sensitizing agents and blast enhancers for slurry and emulsion explosives

used in the mining and quarry industries. According to the Institute of

Manufacturers of Explosives (IME), hundreds of millions of pounds of

slurries and emulsion explosives are used annually. While the feasibility of

using recovered propellants and explosives in slurries depends on their

352


availability and cost, this potentially could be a significant market for

recovered energetic materials. When used in slurries, explosive additives are

generally in the range of 5 to 30 percent, and most major commercial

explosive formulations can be altered to accommodate military propellants

and explosives.

Figure 11.Flow diagram of hydromlnlng process.

Other smaller scale applications for recovered energetic materials recently

have been demonstrated. For example, Thiokol Corporation has made 2-lb

booster charges, used to initiate ammonium nitrate/fuel oil (ANFO) or slurry

explosives, from Hazard Class 1.1 rocket propellants. TPL, Inc., has

demonstrated using reclaimed granulated plastic-bonded explosives (PBX)

for explosive-metal bonding and forming applications. Requirements for this

type of application, such as a detonation velocity of 2.2 km/s with a

variation of± 50 mls, are fairly stringent. The TPL application was

demonstrated under a small business innovative research (SBIR) contract

from the Naval Surface Weapons Center in Crane, Indiana.

Ingredient Recovery Ingredient recovery from propellant or explosive compositions is the least

353


advanced reuse technology. In theory, ingredient recovery is not difficult,

but, until recently, there has been no economic or environmental driving

force to recover individual ingredients. Moreover, many military programs

have a "no change" policy that prohibits changes in materials used in

ordnance manufacture. This policy also would distinguish between

recovered materials and virgin materials made from reactants. The "no

change" policy is starting to change under environmental and economic

pressures, but ingredient recovery probably will continue to meet resistance

from risk-averse program managers.

Three significant efforts are being conducted in the area of ingredient

recovery and reuse. In the first, AP is recovered from Hazard Class 1.3

composite rocket propellants. This technology involves leaching of the

soluble AP from size-reduced propellants, recrystallization at an AP vendor,

and reincorporation of AP into rocket propellant. Over 100,000 lb of AP have

been recovered and recrystallized using this method, and the propellants

made from the recovered AP cannot be distinguished from those made with

virgin materials. A schematic of the reclamation process is shown in Figure

12. Two companies, Thiokol Corporation and Aerojet Solid Propulsion

Company, are participating in this effort with support from two AP

producers, WECCO & Kerr McGee, as well as the U.S. Air Force and the

Large Rocket Motor Demilitarization Group another ingredient drawing

interest for recovery is HMX. The HMX recovery process involves separation

by dissolving and subsequent recrystallization using solvents such as

acetone or dimethyl sulfoxide (DMSO). At least two organizations have

reported successfully meeting material specifications for recovered HMX:

TPL, Inc., which recovers HMX from PBX; AND THE U.S. Army Missile

Command (MICOM), which recovers HMX from Hazard Class 1.1

propellants. In addition to reuse in military applications, HMX might have

commercial applications, such as serving as an oil well perforation charge.

354


Table 12. Types of Munitions That Have Been Cryofractured

Munition

Type

Tested Form Explosive Elements Explosive

items

Cryofracture

d

M55

Rockets

(155 mm)

Rocket in firing

tube

Comp B burster 3.2

lb Doublebase cast

propellant 19.3 lb

5

M23 Land

Mines

Steel drum with

three mines and

packing material

Com B burster 0.8

lb

126

M60 105-

mm

Cartridges

Wood box with two

cartridges in fiber

tubes

Tetrytol burster 0.3

lb

Tetrytol booster

0.05lb

Single base grain

propellant 2.8 lb

72

155 –mm

Projectiles

Projectile M-110 and M21 A1 1204

The third ingredient that has been successfully recovered and recycled is

white phosphorus. The Crane Army Ammunition Activity (CAAA) installation

in Crane, Indiana, has an acid-conversion plant that converts white

phosphorus into phosphoric acid. Using this plant, the GAAA installation

can recover marketable scrap metal and phosphoric acid from white

phosphorus munitions. The acid-conversion plant processes munitions from

other Army facilities and has sold thousands of tons of phosphoric acid and

scrap metal from its demilitarization operations.

355


Figure 12. Flow diagram of ammonium

perchlorate reclamation process.

Energy Recovery One recovery/reuse approach proposed for energetic contaminants in soils

and sludges is solvent extraction followed by burning of the extract with

other fuels to provide energy. AEC has demonstrated that low levels of

smokeless powder, RDX, or TNT can be used to supplement boiler fuel. This

energy recovery approach also could be applied to extracted energetic

materials, using the AEC studies as a guide to the sensitivity and fuel value

of the materials.

3.5.4 Applications Table 13 lists a variety of recovery and reuse applications. Some, such as

the Louisiana Army Ammunition Plant's steam-out facility for TNT-based

explosives, which has been operational for decades, are well established

production-scale methods. These facilities normally have the infrastructure

to handle wastewaters from the recovery process. Others, such as the

356


Cryowash process, which uses 12,000 to 30,000 psi liquid nitrogen to

remove energetic materials from cases, are emerging bench-scale

technologies. The Cryowash process has been demonstrated on hundreds

of pounds of energetic materials and is scheduled to undergo full-scale

prototype testing within the year. Developmental status must be considered

when selecting recovery/reuse technologies for particular applications.

3.5.5 Advantages and Limitations Recovery and reuse of energetic materials should be a goal in every

remediation effort. EPA places this option higher than destruction

technologies on the preferred treatment scale. Each situation, however,

requires a cost/risk/benefit assessment. At sites where rocket motors and

ordnance are in sufficient quantity and have known materials and histories,

recovery/reuse should be seriously considered. At sites where the pedigree

and volume criteria cannot be met, cost/risk/benefit assessments probably

will indicate that destruction technologies should be used. In each instance,

the safety of the operating personnel must be the highest priority.

3.6 Solvent Extraction Solvent extraction is a technology that the Army originally determined to be

infeasible for treating explosives--contaminated soils. The technology,

however, might have potential for treating these soils if a few lingering

technical issues can be resolved.

In 1982, the Army conducted laboratory-scale solvent extraction on

explosives-contaminated lagoon samples from a number of sites. Each

sample was washed with a solution of 90 percent acetone and 10 percent

water. This process achieved greater than 99 percent contaminant removals.

In 1985, the Army conducted a pilot-scale engineering analysis to determine

357


the feasibility of full-scale solvent extraction. This analysis indicated that,

for solvent extraction to be economically feasible the number of required

washes would have to be reduced and acetone would have to be recovered

and reused. Currently, the only available technology for recovering acetone

is distillation, which exposes acetone to heat and pressure. Exposing a

solvent that has been used to extract explosive contaminants to heat and

pressure raises serious safety considerations. In fact, the distillation column

used to recover acetone often is referred to as an •acetone rocket."

Nevertheless, the Army believes that full-scale solvent extraction would be

feasible If a safe, efficient, alternative recovery method were developed.

3.7 Volume Reduction for Explosives Waste A soil washing procedure, termed the Lurgi Process, currently is being

developed in Stadtalendorf, Germany. Although no data have been

published on the effectiveness of this process, initial reports suggest that the

process can reduce levels of explosive contamination in soils to low ppm

levels. As with all soil washing technologies, the Lurgi Process produces

secondary wastes, such as washwater and concentrated explosives.

In the Lurgi Process, contaminated soils are excavated and processed in an

attrition reactor, which detaches the explosive material from the soil

particles. The mixture of detached particles then undergoes a separation

process to remove large rocks. These rocks are crushed and returned to the

site. The remaining material undergoes a second separation process, which

separates clean from contaminated particles. Clean particles are dewatered,

separated into heavy and light materials, and returned to the site.

Contaminated particles undergo a final series of washing, separat1on, and

chemical extraction processes to remove any remaining clean particles.

Finally, the contaminated material is clarified and concentrated before being

disposed of or treated.

358

CHAPTER – 12

SAFETY CONCERN OF EXPLOSIVES WASTE

1. Background Safety precautions must be taken at sites contaminated with explosives

wastes. AEC, which has been involved in sampling and treating explosives

waste sites since the early 1980s, has developed protocols for identifying

sites that require explosives safety precautions and for handling explosives

wastes at these sites. This section discusses AEC's sensitivity testing

protocol, specific precautions required for sampling and treating explosives

wastes, and some laboratory safety issues associated with analyzing

explosives-contaminated samples. The section does not cover statistical site

characterization or the work and health and safety plans suggested by the

Occupational Safety and Health Administration (OSHA).

2. Sensitivity Testing When AEC began to investigate explosives waste sites in the early 1980s, the

only available guidance on sampling and treating explosives-contaminated

soils was 40 CFR 261.23, which vaguely specifies waste identification.

Consequently, AEC developed its own protocol for determining whether soils

contaminated with explosives wastes are susceptible to initiation and

propagation, and, if so, how best to handle them. This original protocol

involved many tests, including impact tests, friction tests, and shock gap

tests. AEC quickly determined that the original protocol was too expensive

and unwieldy, due to the variety of available tests, and developed a two-test

protocol. This protocol involved (1) the deflagration-to-detonation test

(DDT),which measures an explosive material's reaction to flame: and (2) the

Bureau of Mines' zero gap test, which measures an explosive material's

359

Chapter 12 Safety Concern of Explosives Waste

reaction to shock. Both of these tests are extremely conservative, rendering

additional tests unnecessary. The drawback to this protocol was that both

tests required relatively large volumes of soil to be excavated and shipped,

often at great expense, to specially qualified laboratories.

AEC eventually developed its current protocol, which involves chemical

compositional analysis. By analyzing the composition of samples from a site,

AEC can determine quickly and inexpensively whether materials at the site

are susceptible to initiation and propagation. According to the DDT, soils

containing more than 12 percent secondary explosives by weight are

susceptible to initiation by flame; according to the shock gap test, soils

containing more than 15 percent secondary explosives by weight are

susceptible to initiation by shock. As a conservative limit, AEC considers all

soils containing more than 10 percent secondary explosives by weight to be

susceptible to initiation and propagation and exercises a number of safety

precautions when sampling and treating these soils. Sampling and

treatment precautions are exercised when handling soils that contain even

minute quantities of primary explosives.

The reliability of compositional analysis depends on obtaining enough

samples to generate a statistically valid characterization of the site. CAREL

has developed field screening methods to reduce the number of samples that

must be analyzed in the laboratory (see section 3.1). If contamination is in

the parts per million (ppm) or parts per billion (ppb) range by weight, the

samples could be shipped off site for analysis; If contamination is in the

percent range, special analytical arrangements must be made

3. Sampling and Treatment Precautions Work, sampling, and health and safety plans for explosives waste sites

should incorporate safety provisions that normally would not be included in

work and sampling plans for other sites. AEC works with other laboratories

360


such as the Bureau of Mines to conduct site-specific hazards analyses for

every proposed operation at explosives waste sites, including remedial

investigation, remedial design, and remedial action. These analyses include

hazards identification, hazards evaluation, risk assessment, and risk

management.

The most important safety precaution is to minimize exposure, which

involves minimizing the number of workers exposed to hazardous situations,

the duration of exposure, and the degree of hazard. To reduce the degree of

hazard at explosives wastes sites, operations usually are conducted on

materials that have been diluted to a wet slurry. If necessary, distilled water

can be added to the soil to achieve the desired moisture content. Water

desensitizes the explosives and reduces the effects of heat and friction.

Water, however, also can cause a localized detonation to propagate

throughout a soil mass, so moisture content should be adjusted on a sits-

by-site basis. As another safety precaution, nonsparking tools, conductive

and grounded plastic, and no-screw tops, which were developed for

manufacturing explosives, are standard equipment at explosives waste sites.

For example, nonsparking beryllium tools are used instead of ferrous tools.

If contamination is above the 10 percent limit in some areas of a site, the

contaminated material could be blended and screened to dilute the

contamination and produce a homogenous mixture below the 10 percent

limit. This blending is not by itself a remedial action but a safety precaution;

soils containing less than 10 percent secondary explosives by weight

occasionally experience localized detonations but generally resist widespread

propagation. Foreign objects and unexploded ordnance within the

contaminated soil often impede the blending process and require

unexploded ordnance contractors (see section 4.2).

Once blending is completed, soil treatments such as incineration and

361


bioremediation can proceed. Equipment used in treatment must have sealed

bearings and shielded electrical junction boxes. Equipment also must be

decontaminated frequently to prevent the buildup of explosive dust.

4. Laboratory Analysis of Explosives-contaminated Samples Although TNT and RDX are the most common contaminants at explosives

waste sites, many sites also are contaminated with impurities in

production.-TNT, such as DNB, 2,4-DNT, 2,6-DNT, and products of

photochemical decomposition of TNT, such as TNEL These impurities and

decomposition products are thermally labile and hydrophilic and

consequently should not be analyzed using certain tests and solvents. For

example, gas chromatography (GC), in particular, is not the best choice to

screen for these chemicals, because thermally labile compounds decompose

in GC equipment. High-performance liquid chromatography (HPLC) (SW846

method 8330) has been selected for routine laboratory analysis of soils from

military sites.

5. Safety Management Systems

Disposal and destruction activities in case of explosive wastes are high

hazard and comprise one of the main causes of accidents and incidents in

cases. Many incidents do not involve injury but injuries and deaths have

nevertheless occurred. Incidents can happen because of:

• a failure to recognise that explosives requiring disposal or destruction

are accumulating in process or storage areas.

• casual attitudes when dealing with the disposal and destruction of

explosives often arising out of a lack of competence or a failure to

properly supervise, inspect and audit the disposal activity;

362


• people not appreciating the properties and behaviour of explosives under

certain conditions (products requiring disposal or destruction may be

unusually sensitive due to changes in rheology and morphology,

deterioration, contamination or inadequate stabilisation);

• ill-considered systems of work or no basic safety precautions often

arising out of a failure to conduct suitable and sufficient risk

assessments or a failure to follow prescribed procedures;

There is also a risk of insufficient time and money being devoted to disposal

and destruction activities because of the potentially non-productive, non-

profit making nature of the job.

6. Destruction of Explosives Waste

This provides practical guidance with respect to the employment of four

main methods for the disposal of explosives wastes by destruction.

Guidance is provided with respect to burning, with respect to detonation,

with respect to dissolution or dilution and with respect to desensitisation

and chemical destruction.

7. Burning Explosive Wastes Substances And Specific Articles When burning explosives wastes, the risk of burning to detonation must be

taken into account, and measures taken to minimise the risk and to protect

against the effects of a detonation should it occur. The general rule is to only

burn small quantities at any one time while avoiding excessive transport

movements. Items which might be propelled form the fire when burned must

be suitably contained without confining the explosive. Incompatible explosives

363


must not be burned together. If there is any doubt about compatibility of

explosives they must be burned separately.

8. General Procedures and Risk Controls Subject to some form of pretreatment or special precautions, most

explosives waste substances can be burned safely provided that:

• as far as possible, burning is controlled, and the operation is designed to

minimise the possibility of the explosive burning to detonation,

deflagrating violently, projecting primary or secondary fragments and

burning brands or producing excessive thermal effects;

• it is assumed that the maximum credible event will occur and sufficient

precautions are taken to mitigate any resultant hazards.

364

CHAPTER – 13

LABORATORY-SCALE ANALYTICAL METHODS

1. Field Screening Methods for Munitions Residues in Soil

1.1 Background

Laboratory analysis of samples from sites contaminated with explosives

wastes is expensive and time consuming. Due to heterogeneous waste

distribution at many sites, it would not be unusual for 80 to 90 percent of

the soil samples from a given site to contain no contamination. As a result,

developing a site characterization with good spatial resolution is extremely

expensive. Field screening methods determine quickly and less expensively

which samples are contaminated with explosives residues, thereby reducing

total analytical costs. For example, field screening was found to be

acceptable for determining soil contamination areas at a military site (Craig

et al.1993). This section discusses the field screening procedures developed

by CAREL and advantages and limitations of CAREL's procedure. The

section does not cover soil sampling procedures.

1.2 Field Screening Methods In developing the field screening methodology, CAREL considered several

design criteria. The method needed to detect contaminants that were

present at most military sites. Based on data from sites Investigated by

CAREL and MAD, CAREL determined that most sites could be adequately

assessed by methods that screen first for TNT and RDX, and secondarily for

2, 4-DNT, TNB, DNB, and tetryl. The equipment needed to be portable

portable, so it could be shipped easily to sites, and simple to operate,

because field operators would not necessarily have experience in analytical

chemistry. Field screening procedures also needed to use only low toxicity

solvents and have a quick turnaround time, a large analytical range, a linear

365

Chapter 13 Laboratory-Scale Analytical Methods

calibration scale, and a sufficiently low detection limit. In addition, the

results of the procedure needed to correlate well with results from standard

laboratory method.

CAREL's methodology has three steps: extraction, TNT screening, and RDX

screening.

1.2.1 Extraction CAREL's procedure begins with a simple extraction process. A 2Q-g sample

of undried soil from the site is mixed with 100 mL of acetone. The sample is

shaken for at least 3 minutes, allowed to settle, and filtered with a syringe

filter. Very heavy clays might require longer extraction periods, but 3

minutes is often sufficient for most sandy and loamy soils. The efficiency of

acetone extraction is 95 percent that of standard laboratory methods. The

filtered extract then is subjected to CAREL's TNT and ADX screening

procedures. For more detailed information on these procedures, see U.S.

Army CAREL, 1990, and U.S. Army CAREL, 1991.

1.2.2 TNT Screening In the TNT screening procedure, the initial absorbance of the acetone extract

at 540 nanometers (nm) is obtained using a portable spectrophotometer. The

extract is amended with potassium hydroxide and sodium sulfite, agitated

for 3 minutes, and filtered again. The extract then can be analyzed visually.

If it has a reddish or pinkish color, it contains TNT; if it has a bluish color, it

contains 2,4-DNT. Figure 3-1 shows the reaction-known as the Janowsky

Reaction(1886)-that produces the reddish-colored anion from TNT and the

bluish-colored anion from 2, 4-DNT.

366


Figure 1. Schematic of the Janowsky Reaction

(1886) for TNT and 2, 4-DNT.

Absorbance measurements can be used to obtain quantitative results.

Figure 3-2 illustrates the visible absorbance spectrum of the Janowsky

Reaction product of TNT, showing the maximum absorbance at 460 and 540

nm. CAREL uses the peak at 540 nm to verify the presence of TNT, even

though the absorbance at 460 nm is greater, because of the potential for

interference from humic substances at 460 nm. Figure 3-3 illustrates the

visible absorbance spectrum of an acetone extract of uncontaminated

potting soil before and after Janowsky Reaction reagents are added, showing

the greater absorbance near the 460-nm as opposed to the 54Q-nm

wavelength.

367


Figure 2. Visible absorbance spectrum of the

Janowsky Reaction product of TNT.

Figure 3. Visible absorbance spectrum of acetone extract of potting soli

before and after addition of Janowsky Reaction reagents

The results of TNT screening, which reflect the sum of the TNT and TNB

concentrations, correlate well with results obtained in the laboratory. Table

3-1 compares the sum of the TNT and TNB concentrations as determined by

colorimetric analysis with the sum of the TNT and TNB concentrations as

determined by laboratory analysis for homogenized, field-contaminated (i.e.,

not spiked) soil samples from seven sites. Figure 3-4 shows the strong

correlation (R2 =0.985) between results of colorimetric analysis and the

368


standard HPLC laboratory procedures for homogenized soil samples. Table

3-2 compares colorimetric and HPLC results from the Umatilla Army Depot

site in Oregon, showing a slightly lower correlation due to the high

concentrations of TNT at the site. At the Savanna Army Depot site in Illinois,

Dames and Moore, Inc., reported a correlation of 0.959 between the results

of laboratory and field analyses. At the Seneca Army Depot site in New York,

Aquatec reported that colorimetric analysis identified 15 contaminated and

46 uncontaminated samples. Laboratory analysis revealed only 2 false

positives and confirmed all 46 negative results.

1.2.3 RDX Screening Field screening for RDX is similar to, but slightly more complicated than,

field screening for TNT. As in the procedure for TNT, acetone is used to

extract contaminants from soil samples. The extract then is passed through

an anion exchange resin to remove nitrate and nitrite. Zinc and acetic acid

are added to the extract, thereby converting RDX to nitrous acid. The extract

then is filtered and placed in a vial with a Hach NitriVer 3 Powder Pillow. If

the extract has a pinkish color, it contains RDX. Figure 3-5 shows the

reaction sequence, including the Griess Reaction (1864), that produces the

pinkish-colored molecule (Azo dye) from RDX.

Figure 4. Correlation of TNT and TNB analysis by Colorimetric and

standard RP·HPLC procedures.

369


As in the TNT detection procedure, quantitative analysis of the extract can

be obtained from absorbance measurements. Figure 3-6 is the visible

absorbance spectrum of the NitriVer 3 reaction product, showing the

absorbance maximum at 507 nm. Colorimetric analysis of uncontaminated

soil after acidification and addition of the Griess Reaction reagents shows no

background absorbance {see Figure 3-7).

The results of RDX screening, which reflect the sum of the concentrations of

RDX and HMX, correlate well with results obtained in the laboratory. Table

3-3 compares RDX and HMX concentration estimates from field and

concentration estimates from field and laboratory analysis of soil samples

from three sites. Figure 3-8 shows the strong correalation (R2 = 0.986)

between these field and laboratory results. Table 3-4 also shows a strong

correlation between RDX concentration estimates from field and laboratory

analysis of homogenized, field-contaminated soil samples collected from the

Newport Army Ammunition Plant site in Indiana.

Table 1. Comparison of TNT and TNB Concentrations ea Determined by

Field and Laboratory Procedures

Colorimetric (µg/g)

HPLC (µg/g)

Sample Origin TNT+TNB TNT TNB

Vigo Chemical Plant (IN) 14 5 <d

Hawthorne Army Ammunition Plant (NV) 6 5 <d

Nebraska Ordnance Works (NE) 2 0 3

Hastings Ind. Pk. (NE) 592 340 157

Hawthorne Army Ammunition Plant (NV) 85 68 3

Nebraska Ordnance Works (NE) <d

Lexington-Bluegrass Depot (KY) 146 64 74

370


Sangamon Ordnance Plant (IL) 15 6 <d

Raritan Arsenal (NJ) 85 72 <d

Table 2. Comparison of Colorimetric and HPLC

Results from Umatilla Army Depot

Sample

TNT Concentration Estimate (µg/g)

Colorimetric Method Standard HPLC Method

1b 1,060 2,250

2a 3,560 7,430

3b 704 1,180

3a 3,180 4,030

4a 4,490 8,520

5a 2,530 3,990

6a 84 131

8a 102,000 38,600

9a 6,510 7,690

11a 716 1,300

12a 109 183

a= Surface soil b=Soil from 18 in. depth Source: Jenkins and Walsh, 1992

371


Figure 5. RDX reaction sequence, Including production of pinkish-

colored anion (Azo dye) by Griess Reaction (1864).

1.3 Advantages and Limitations of the Methodology CAREL' smethodology has several advantages, including:

• Speed. The TNT and RDX detection procedures take about 30 minutes per

sample, including the 15-minute color development stage. Typically, 25

samples can be analyzed per day for both RDX and TNT.

372


Figure 6. Visible absorbance spectrum of NitriVer 3 reaction product.

Figure 7. Visible absorbance spectrum of acetone extract of

uncontaminated soli before and after addition of Griess Reaction

reagents.

373


Table 3. Comparison of Colorimetric and HPLC Results for Several U.S. Army Sites

Sample Origin

Colorimetric

(µg/g)

HPLC (µg/g)

RDX+HMX RDX HMX

Nebraska Ordnance Works (NE) 1,060 1,250 115

Hawthorne Army Ammunition Plant (NV) 233 127 56

Raritan Arsenal (NJ) 11 4

Nebraska Ordnance Works (NE) 3 4

Nebraska Ordnance Works (NE) 1,100 1,140 105


Hawthorne Army Ammunition Plant (NV) 6 3 <d




Nebraska Ordnance Works (NE) 2 <d <d

Figure 8.Correlation of RDX analysis by colorimetric and standard

HPLC procedures (Jenkins and Walsh, 1992).

374


Cost. The solvents used in these tests are very inexpensive. The total cost for

materials to process each sample is about $20, relatively inexpensive

compared to other analytical methods.

Simplicity. The calibration of the colorimetric analysis is linear, and the test

has a zero intercept, meaning that all associated calculations are very

simple.

Table 4. Comparison of Colorimetric and HPLC Results

for Newport Army Ammunition Plant

RDX Concentration Estimate (µg/g)

Sample t Colorimetric Method Standard HPLC Method

1 0.55 0.05

2 2.86 1.31

3 4.55 3.15

4 6.62 15.5

5 5.87 8.45

6 253 299

7 17.4 38.8

8 45.4 258

9 674 1,800

10 2,430 3,170

11 7,690 12,200

• Laboratory correlation. The results of colorimetric analysis show strong

correlation with those obtained by HPLC procedures.

• Low incidence of falsa negatives. This is important since the procedure is

375


used to screen for explosives-contaminated soils.

• Low detection limits. The procedure can detect explosive residues at

concentrations as low as 1g/g.

The limitations of CRRELS’s procedure include:

Water content. Samples must contain 2 to 3 percent water by weight.

Samples from sites with dry conditions must be wetted with distilled water

prior to colour development.

Interferences. The TNT procedure detects other nitroaromatics and is

subject to positive interference from humic materials. These interferences

can be reduced by careful visual analysis prior to colorimetric analysis. The

RDX procedure detects other nitramines and nitrate esters such as

nitrocellulose and nitroglycerine

1.4 TNT and RDX Test Kits A private firm has developed and tested a field screening kit based on

CAREL's methodology. A "how to" videotape explaining the procedure is

available from Martin H. Stutz at the U.S. Army Environmental Center at

Aberdeen Proving Ground, Maryland 21010. Requests must be submitted in

writing.

References

Craig, H.D., A. Markos, H. Lewis, and C. Thompson. 1993. Remedial

investigation of site D at Naval Submarine Base, Bangor, Washington. In:

Proceedings of the 1993 Federal Environmental Restoration Conference,

Washington, DC.

Jenkins, T.F. and M.E. Walsh. 1992. Development of field screening methods for

TNT, 2,4-DNT, and RDX in soil. Talanta 39(4): 419-428.

376


U.S. Army CAREL. 1991. U.S. Army Cold Regions Research and Engineering

Laboratory. Development of a field screening method for RDX in soil. Walsh,

M.E. and T.F. Jenkins. CAREL Special Report 91-7. Hanover, New Hampshire.

U.S.Army CAREL. 1990. U.S. Army Cold Regions Research and Engineering

Laboratory. Development of a simplified field method for the determination of

TNT in soil. T.F. Jenkins. CAREL Special Report 90-38. Hanover, New

Hampshire.

2. Characterization of Radioactive Contaminants for Removal

Assessments

2.1 Background In 1987, EPA’s Office of Radiation and Indoor Air (ORIA) developed a

characterization protocol for determining the feasibility of reducing the

volume of soils contaminated with radioactive wastes at Superfund sites.

ORIA's protocol is more extensive than standard protocols, which require

only gamma spectroscopy of bulk samples to determine the levels of

radioactive constituents. In ORIA's protocol, sieving and sedimentation

techniques are used to separate soils into size fractions. Each fraction then

undergoes petrographic and radiochemical analysis to determine the values

of certain parameters, such as grain size distribution, mineral composition

and percentages, and physical properties of radioactive contaminants, that

affect the feasibility of volume reduction. This section discusses the potential

applicability of ORIA's protocol to radioactive soils at federal facilities,

examines the two tiers of the protocol, and presents a case study of a

radium-contaminated site where the protocol was applied.

2.2 Applicability to Military Installations

377


ORIA's protocol potentially could be used to characterize soils at military

sites contaminated with radioactive wastes. For example, at an Air Force

base in California, it was speculated that radium paint buried in a bunker

was contributing to elevated uranium levels in the well water of a nearby

field. Radiochemical analysis would have indicated that radium paint does

not contain the parent compound, uranium-238, so uranium at the site

could not have been derived from radium paint in the bunker. Similarly, at

an Air Force base in New Mexico, researchers conducted an analysis for

radium contamination near a particular bunker where radium paint also

might have been buried. This analysis found radium only at background

levels. A petrographic analysis of the soil would have revealed natural

radioactive minerals and led to the same conclusion.

ORIA's protocol is relatively inexpensive. Petrographic analysis of five

representative soil samples takes a petrographer about one week and costs

about $5,000. Radiochemical analysis takes three times as long and costs

about $15,000. Thus the total cost to develop a detailed characterization of

soil from a military installation, as a feasibility study for remediation

considerations, would be approximately $20,000.

2.3 OR/A's Sol/ Characterization Protocol ORIA's methodology was developed based on investigations at thorium-

contaminated sites in Wayne and Maywood, New Jersey; radium-

contaminated sites in Montclair and Glen Ridge, New Jersey; and plutonium

surrogate host soil at the Nevada Test Site. These investigations led to the

development of a two-tiered protocol: Tier 1 is a feasibility study; Tier 2 is an

optimization study.

2.3.1 Tier 1: Feasibility Study

378


The Tier 1 feasibility study has two stages: fractioning and analysis.

Fractioning Bulk samples are dried at 80°C and examined by high resolution gamma

spectroscopy. Samples then are split into representative 390-gram portions

by prescribed separation methods, and each portion to be tested is placed in

a beaker to create a slurry of five parts deionized water to one part solids.

After 24 hours, the slurry is stirred and poured through a nest of

increasingly fine sieves to separate the bulk sample into size fractions of

coarse, medium, and fine sand and silt.

Analysis The fractions obtained by sieving undergo three separate analyses. First, the

fractions are analyzed to obtain the sample’s grain size distribution curve,

which identifies each size fraction’s contribution to the total weight of the

sample. Figure 3-9 is a grain size distribution curve for soils from the

Nevada Test Site. Second, the fractions are analyzed for radioactivity as a

function of particle size. Figure 3-10 is a graph of radioactivity versus

particle size for radium-, thorium-, and uranium-contaminated soils, from

Maywood, New Jersey, showing increased radioactivity in the silt-size

fraction. Third, the size fractions undergo petrographic analyses, which

generate precise statistical counts of the various particles in the soil.

Coarse-size materials, which are greater than 0.6 mm, are analyzed visually.

Medium-size materials, which are between 0.038 and 0.60 mm, are

immersed in index oil and examined under petrographic and binocular

microscopes. Fine-size materials, which are less than 0.038 mm, are

examined by X-ray diffraction. Finally, medium-size materials undergo a

second petrographic analysis in which a separatory funnel containing

sodium polytungstate is used to extract minerals with specific gravities

greater than 3.0. These minerals, which usually represent a small

379


percentage of the total sample, contain disproportionately high levels of

radioactive materials. Figure 3-11 shows the heavy mineral composition of

soil from the Wayne and Maywood, New Jersey, sites. The heavy mineral

fraction of the soil from this site contains all of the radiation contaminants.

Monazite, which contains almost all of the radioactivity, represents only

about 10 percent of the heavy mineral fraction and comprises less than 1

percent of the total sample Zircon, which can contain up to 4 percent

substitution of thorium or uranium in the crystal lattice, constitutes the

remainder of the radioactive material at this site.

Figure 9. Grain size distribution curve and histogram for soil from the Nevada Test Site.

2.3.2 Tier 2: Optimization Study If Tier 1 suggests that volume reduction is feasible, further analyses can be

performed to characterize the contaminated soil. Size fractions can be

broken down into more precise increments by hydro classification and

centrifuge. In addition, chemical assays can be used to quantify the mineral

analysis if a chemical element is known to be solely associated with a

particular contaminant. For example, at one of the radium-contaminated

380


sites, the ore minerals for radium include a urinal vanadinate. Since

vanadium is rare, it can be used as a “chemical signature” to determine the

weight percentage of this ore mineral of radium. Instruments such as the

scanning electron microscope (SEM) and energy dispersive X-ray

spectrometer (EDX) also can be useful in identifying the morphology and

elements of specific particles in the submicroscopic size range.

Figure 10 Radiochemical analysis showing

Radioactivity as a function of particle size.

2.4 Case Study: Montclair/Glen Ridge Superfund Site From 1915 to 1926, acid leach tailings from the manufacture of radium

were deposited in open field pits in Montclair and Glen Ridge, New Jersey.

After operation ceased in 1926, residential housing was developed in the

area. Most of the contamination, which consists primarily of precipitates

and coprecipitates from the acid leach process, is within 8 ft of the surface.

381


Ground water contamination is confined to areas directly surrounding the

dump areas, and there is no ground water contamination in the bedrock,

which is 20 ft below the surface. Consolidated glacial till, along with other

materials that were dumped in the pits, is the host material for the radium-

contaminated tailings. The costto remove, transport, and dispose of all

300,000 yd 3 of soil from the site is estimated at close to $300 million,

making volume reduction an attractive option.

Tier 1 analyses indicated that the contaminated material consists of 15

percent ores, such as carnotite and uraninite, and 85 percent anthropogenic

materials. Within the latter group, most of the radioactivity is located in the

fine silt and clay fractions, particularly in the 10- to 20-f.lm fraction. A

linear density gradient analysis was used to separate the 10- to 20-f.lm

fraction into light, medium, and heavy particles (see Table 3-5). These three

groups of particles then underwent Tier 2 analyses, including gamma

spectroscopy, X-ray diffraction analysis, SEMIEDX analysis,

photomicrography, and autoradiography. Figures 3-12 and 3-13 illustrate

the results of some of these analyses. The light particles, which are mostly

amorphous silica, were found to contain about 25 percent of the radium; the

heavy particles, which are mostly radiobarite, were found to contain about

50 percent of the radium.

Based on the results of the characterization, site engineers decided to

remove the fine silt particles from the site and wash the remaining sand-size

particles of any residual clay. In laboratory testing, these procedures

reduced 30 to 40 percent of the material to a target level of 12 to 15

picocuries per gram of radium 226 (see Figure 3-14).

382


Figure 11. Heavy mineral composition of soil from the

Wayne and Maywood, Nf. IW Jersey, sites.

Table 5. Linear Density Gradient Analysis of 10. to 20.J.1m Size

Fraction of Soli from Glen Ridge, New Jersey, Site

Density Weight

Ra-226 Activity %Ra

Light 2.10.2.25

32.20 1,640 pCl/g 25.21

Medium 2.25-2.71

55.69 1,040 pCl/g 27.55

Heavy 2.71

12.01 8,270 pCl/g 47.24

383


Figure 12. SEM (Inset) and EDX analysis of amorphous silica from the

2.1o-2.25 density fraction of the 1Q- to 2G-J.1m grain size of radium-

contaminated soli from Glen Ridge, New Jersey.

2.5 References 1. U.S.EPA.1989. U.S. Environmental Protection Agency. Characterization

of contaminated soil from the Montclair/Glen Ridge, New Jersey,

Superfund sites.

2. EPA/520/1-89/012. U.S. EPA, Office of Radiation Programs.

Additional References Neiheisel, J. 1992. Petrographic methods in characterization of radioactive

and mixwaste. Proceedings of HMC/Superfund 1992, December 1-3,

Washington, DC, 192-195.

384


Figure 13. Autoredlograph (SEM) showing radiation etch tracks from

radlobarlte (Inset) and EDX of radlobarlte In the heavy fraction of 1o-

to 2G-J.1m grain size of radium-contaminated soil from Glen Ridge,

New Jersey.

Figure 14. Radium reduction produced by laboratory-scale water

washing and wet sieving of soli from Montclair and Glen Ridge sites.

385

CHAPTER – 14

REMEDIATION OF EXPLOSIVE-CONTAMINATED SOILS:

ALKALINE HYDROLYSIS AND SUB-CRITICALWATER DEGRADATION

1. Introduction

Natural water and soils near military sites where training regularly occurs are

vulnerable to contamination from explosives such as 2, 4, 6-trinitrotoluene

(TNT) and hexahydro-1, 3, 5- trinitro-1, 3, 5-triazine (RDX) (Rogers and

Bunce, 2001; Certini et al., 2013). TNT and RDX are toxic to organisms

including algae, fish, rats, and humans (Yinon, 1990), and are classified as

potential carcinogens by the United States Environmental Protection Agency

(US EPA) (McLellan et al., 1988). The intensive military training that has

occurred since the Korean War has resulted in explosive compounds

contaminating soil and groundwater near military training sites for artillery

fire. To replace costly incineration methods for the remediation of explosives,

many studies have sought alternative remediation methods to treat

explosives where they are found, including the use of aerobic and anaerobic

biodegradation (Bradley and Chapelle, 1995; Roberts et al., 1996),

composting (Bruns-Nagel et al., 1998), phyto remediation (Thompson et al.,

1998), and chemical oxidation (Li et al., 1997). However, these methods have

drawbacks, such as slow degradation rates, long treatment periods, high

energy consumption, expensive equipment, extra cost for additional

treatment, and the possible toxicity of degradation products.

Alkaline hydrolysis is a potential remediation method with low cost and rapid

reaction rates. Nucleophile attack using bases (e.g., OH−) easily transforms

nitro explosives in basic conditions (Bernasconi, 1971; Hoffsommer et al.,

1977). In recent decades, base-catalyzed transformation of TNT and RDX

386

Chapter 14 Remediation of Explosive-Contaminated Soils

has been increasingly studied (Heilmann et al., 1996; Emmrich, 1999;

2001; Balakrishnan et al., 2003; Bajpai et al., 2004; Hwang et al., 2005). The

degradation kinetics has been suggested in various conditions and possible

daughter products have been identified. Through a quantum chemical

approach using experimental data (Mills et al., 2003), Hill et al. (2012)

showed that possible initial steps of TNT alkaline hydrolysis were hydrogen

abstraction by OH−, Meisenheimer complex formation, and substitution of

nitro functional group with OH−. They also showed the formation of

polymeric products after the initial steps. Final products of RDX alkaline

hydrolysis were reported to be NO2−, HCHO, NH3, and N2O (Balakrishnan

et al., 2003). In addition, the United States Army has successfully

performed an in-situ field application of alkaline hydrolysis for explosive-

contaminated soils (Thorne et al., 2004).

Subcritical water is liquid water under pressure at temperatures between

the usual boiling point (100◦C) and the critical temperature (374.1◦C)

(LaGrega et al., 2000). Un- der subcritical conditions, the dielectric

constant, surface tension, and viscosity of water molecules are dramatically

decreased (Siskin et al., 1990; Kuhlmann et al., 1994; LaGrega et al., 2000).

In addition, the combination of subcritical water with oxidizing agents, such

as O2 and air, can effectively oxidize organic compounds that are normally

very difficult to oxidize (Kronholm et al., 2001; Dadkhah and Akgerman,

2002). Therefore, attempts have been made to utilize subcritical water to

extract or decompose polycyclic aromatic hydrocarbons (PAHs), pesticides,

polychlorinated biphenyls (PCBs), dioxin and other organic contaminants in

soil (Lagadec et al., 2000; Dadkhah and Akgerman, 2002; Weber et al.,

2002; Kubatova et al., 2002; Hashimoto et al., 2004). For the degradation of

explosives in contaminated soil, subcritical water degradation is also

currently suggested. Hawthorne et al. (2000) showed that significant

degradation of TNT and RDX started at 125 and 100◦C, respectively. Via

pilot-scale experiments, more than 99.9% of TNT and RDX in contaminated

387


soil was transformed at 275◦C for 1 h. Kalderis et al. (2008) found that

98-100% of TNT was degraded from highly contaminated soil (12% TNT) at

150–225◦C. They also showed that extraction or diffusion of TNT from

inside a soil matrix to the soil surface is a rate-limiting step of subcritical

water degradation of TNT and that the soil matrix can also catalyze the

degradation of TNT by subcritical water in a soil-water system. It is also

reported that heating at temperatures higher than 200 and 280◦C can induce

gasification and irreversible decomposition of hydrophobic compounds in soil

(DeBano et al., 1998).

Due to the difficulty of accessing military sites in South Korea, information

on explosive contamination in soil was limited and real contaminated soil in

artillery ranges cannot be easily obtained. In the present study, we selected

alkaline hydrolysis, a well-established, proven technology for explosive

treatment, to remediate explosive-contaminated soils in South Korea. We

also chose subcritical water degradation as an alternative process. Prior to

practical application as an ex-situ remediation process, we evaluated and

summarized the kinetics of alkaline hydrolysis and subcritical water

degradation using real explosive-contaminated soils in comparison with

previously reported data in water and soil. In addition, artificially

synthesized 2, 4-dinitrotoluene (DNT)-contaminated soil was also tested.

2. Materials and Methods

2.1 Chemicals and Explosive-Contaminated Soils

DNT (97%), a reference nitro-aromatic compound, was purchased from

Aldrich (Milwau- kee, WI, USA). TNT and RDX were provided by Hanwha

Corp. (Seoul, Korea). Standard stock solutions of TNT (1 mg mL−1 in

acetonitrile) and RDX (1 mg mL−1 in acetonitrile) were obtained from Accu

Standard (New Haven, CT, USA). Methanol (HPLC grade, SK Chemicals,

Korea), HCl (69-70%, Junsei Chemical, Japan), and NaOH (30%, Fisher

Scientific, Pittsburgh, PA, USA) were purchased and used as received

388


without further purification.

TNT- and RDX-contaminated soils were collected from two artillery ranges

(three samples from each range) operated by the Korean Army of South

Korea. According to site investigation (Jung, 2013), explosive contamination

was only shown at surface soil (< 15 cm). The randomly collected surface

soils were dried at room temperature. Dried soils were crushed using a

mortar and sieved at 2 mm to remove coarse particles. Particle size analysis

was conducted using a laser particle size analyzer (Mastersizer 2000, Malvern,

UK). Loss on ignition was determined by heating at 600 ± 25◦C for 3 h

using a muffle furnace (model 184A, Fisher Scientific). Soil pH was

measured after 1 h agitation of 5 g sediment and 25 ml deionized water using

an Accumet 925 pH/ion meter (Fisher Scientific). To find the concentrations of

TNT and RDX, the soil was extracted with methanol for 5 min and the filtrate

(GF/C) was analyzed by high-performance liquid chromatography (HPLC).

DNT-contaminated soil was artificially synthesized using fresh soil, which

was collected from a construction area near the University of Ulsan, and

concentrated DNT solution in acetone. The fresh soil was highly common

soil originating from granitic rocks in South Korea and included quartz,

feldspar, kaolinite, and goethite (Oh and Shin, 2014). Soil properties and

explosives concentrations of the soils are summarized in Tables 1 and 2.

2.2 Batch Experiments for Alkaline Hydrolysis

For alkaline hydrolysis of explosives in aqueous solution, 500 ml duplicate

flasks, including 200 ml of explosive solution, were shaken at 200 rpm.

Solution pH was adjusted to 11, 12, and 13 using 0.1 M NaOH. The initial

concentrations of DNT, TNT, and RDX were 47.1, 10.0, and 32.2 mg L−1,

respectively. At selected time intervals (10, 20, 30, 60, 120, and 180 min), a

2 ml sample was collected from each duplicate flask, quenched by adding a

drop of concentrated HCl, and immediately passed through a 0.22 µm

membrane filter (Millipore, Billerica, MA, USA) for chemical analysis. In

389


batch experiments for alkaline hydrolysis of explosives in contaminated

soils, 200 ml vials, including 10 g of explosive- contaminated soil and 50 ml

of NaOH solution (pH 11, 12, and 13), were shaken in a horizontal

position on a platform shaker rotating at 250 rpm. At each sampling time,

duplicate vials were taken, quenched by adding a drop of concentrated HCl,

and filtered through a 0.22 µm membrane filter. Remaining soil on the filter

was extracted by methanol for 5 min with a vortex shaker. The filtrate and the

extract of filtered soil were analyzed for the determination of explosives’

concentrations. Average recoveries of methanol extraction for DNT, TNT, and

RDX from the contaminated soils were 90%, 92%, and 89%, respectively. All

batch experiments were performed at 25 ± 3◦C.

2.3 Batch Experiments for Subcritical Water Degradation

Subcritical water degradation experiments were performed at 100, 150,

200, 250, and 300◦C, using a pre-designed, temperature-controlled,

pressurized reactor system. This system included a 500 mL main reactor, a

temperature controller, a shaking speed controller, a pressure gauge, and

gas sampling ports (Oh et al., 2011). The temperature was set to 100, 150,

200, 250, and 300◦C, and the internal pressure of the reactor was

maintained at the vapor pressure of steam at those temperatures. Therefore,

the internal pressures of the reactor at 100, 150, 200, 250, and 300◦C were

approximately 1, 4.7, 15, 39, and 85 atm (corresponding to 0.48, 1.52,

3.95, and 8.61 MPa), respectively. Forty grams of soil and 200 ml of

deionized water were put into a reactor and shaken at 150 rpm. The

reactors were prepared in duplicate to ensure reproducibility. To determine

the effect of soil-to- water ratio, soil dosage was changed to 20, 40, 50, and

60 g, corresponding to ratios of 1:10, 1:5, 1:4, and 1:3.3, respectively. After

shaking for the predetermined sampling time, the solution from each

replicate reactor was separated using a GF/C filter and analyzed using

HPLC. The filtered soil was also extracted with methanol for 5 min and the

390


filtrate (GF/C) was analyzed by HPLC. According to the soil-to-water ratio,

quantified concentrations of explosives in solution and soil were summed.

Control experiments were also performed under identical conditions without

soil to examine the degradation of DNT in water. Initial concentrations of

DNT, TNT, and RDX in water were 25, 200, and 25 mg L-1, respectively.

2.3 Chemical Analysis

DNT, TNT, and RDX were analyzed using a Dionex UltimateQR -3000 HPLC

(Sunnyvale, CA, USA) equipped with a Dionex AcclaimQR 120 guard

column (4.3 × 10 mm) and an AcclaimQR 120 C-18 column (4.6 × 250 mm,

5 µm). A methanol-water mixture (60:40, v:v) was used as the mobile

phase at a flow rate of 1.0 ml min−1. The injection volume was 100 µL and

the wavelength of the UV detector was set at 254 nm. The retention times

for DNT, TNT, and RDX were 10.72, 12.41, and 4.95 min, respectively.

Table1: Characteristics of explosive-contaminated soils used in this

study

Particle size

distribution

pH Loss

on

ignit

ion

(%)

Sand

(%)

Silt

(%)

Cla

y

(%)

Explosive

concentrat

ion (mg kg-

1)

TNT- contaminated soil 6.0

5.8

5.7

4.2

88.7

78.3

11.2

21.6

0.1

0.1

5.0 ±0.1

49.8±0.3

RDX- contaminated soil 6.2

7.9

6.9

5.1

2.0

1.8

80.4

85.7

79.9

19.4

13.8

19.7

0.2

0.5

0.4

200.4±1.3

26.1±1.3

25.2±0.2

391


DNT- contaminated soil 7.2

6.3

2.3

10.3

81.2

20.0

18.6

74.9

0.2

5.1

263.9±2.1

10.1±0.1-

50.2±0.5

DNT- contaminated soil

after sub-critical water

degradation 300◦C

7.5 1.9 12.0 82.7 5.3 n.da

aNot detected. Table2: Elemental contents of explosive-contaminated soils used in this

study (unit:wt %) a

Al Ca Fe K Mg Mn Na Si Ti

TNT-contaminatedsoil 23.34 0.21 9.95 1.23 3.58 0.11 1.54 53.15 0.55

RDX-

contaminatedsoil

22.2

5

0.18 8.87 0.85 5.15 0.05 2.04 57.28 0.86

DNT-

contaminatedsoil

20.8

9

0.33 10.08 1.10 4.46 0.18 1.02 52.38 1.36

3. Results and Discussion

3.1 Alkaline Hydrolysis of Explosives in Water and Contaminated Soils

The alkaline hydrolysis of DNT, TNT, and RDX in contaminated soils is

shown in Figure 1. The hydrolytic degradation of explosives was similar to

that in aqueous solution. Increasing pH enhanced explosives’ degradation.

Similar to the hydrolysis in solution, it is clearly shown that pH 12 and 13

were critical thresholds for increasing DNT and TNT, and RDX

transformation, respectively (Figure 1). The pseudo-first-order rates are

summarized in Table 3. The rates were similar in soil and aqueous

systems, showing that the existence of soil did not significantly affect the

hydrolytic degradation of explosives under the given conditions. To

determine the effect of the solution-to-soil ratio on the hydrolysis of

392


explosives, the amount of base solution was changed. Increasing the

solution to 80 and 100 mL (soil-to-solution ratios = 1:8 and 1:10,

respectively) did not affect the degradation rates (data not shown).

However, decreasing the solution to 40 mL (soil-to-solution ratio = 1:4)

significantly hindered the hydrolysis of the explosives. Even at pH 13,

the pseudo- first-order rates of DNT, TNT, and RDX were (4.6 ± 1.9)

×10−4, (5.2 ± 2.1) ×10−4, and (3.8 ± 2.0) ×10−4 min−1, respectively; these

were 2–3 orders of magnitude lower than those in 50 mL of solution (Table

3). The results suggest that the solution-to-soil ratio may affect the mass

transfer of hydroxyl ions to absorbed explosives or desorption of explosives to

aqueous solution under the given conditions.

We examined the alkaline hydrolysis of explosives in solution by comparing

our data with previously published data. At pH 11, removal of DNT, TNT,

and RDX was limited, showing 9, 33, and 4% removal, respectively. At pH 12,

the removal of DNT and TNT was significantly enhanced and complete

removal was obtained in 120 and 60 min, respectively. In contrast, only 19%

of RDX was hydrolyzed at pH 12 in 180 min. The transformation of RDX was

remarkably accelerated, with 84% of RDX removed at a pH of 13. The

pseudo- first-order rate constants of the explosives are summarized in

Table 4. The rate of TNT hydrolysis at pH 12 was (3.4 ± 0.5)×10−2 min−1,

which is similar to, but 3.7–5.9 times higher than, previously published

rates at various temperatures (Emmrich, 1999; Bajpai et al., 2004; Hwang

et al., 2005). These differences may be due to the temperature of the

solution, existence of acetonitrile in the TNT solution, and other

discrepancies in the experimental conditions. The rate of RDX degradation

at pH 12 ((1.2 ± 0.2)×10−3 min−1) was in the same order of magnitude as

the rate published for pH 12.4 (Hoffsommer et al., 1977).

393


Table 3. Estimated pseudo-first-order rate constants (min−1) for

alkaline hydrolysis of explosives in soil

pH 11 pH 12 pH 13 Reference

DNT (5.6±1.5)×10−4

(R2=0.745)

(4.1±0.3)×10−2

(R2=0.993)

(9.6±0.1)×10−2

(R2=0.999)

1.4×10−3(pH12,

20◦C)a

TNT (2.6±0.7)×10−3

(R2=0.755)

(2.9±0.4)×10−2

(R2=0.965)

(2.2±0.1)×10−1

(R2=0.997)

(4.2−5.7)×10−3

(pH12,20◦C)b

RDX (2.9±0.9)×10−4

(R2=0.626)

(1.6±0.4)×10−3

(R2=0.825)

(1.7±0.2)×10−2

(R2=0.975)

N/Ac

Table 4 Estimated pseudo-first-order rate constants (min−1) for

alkaline hydrolysis of explosives in water

pH11 pH 12 pH13 References

DNT (3.4±1.8)×10−4

(R2=0.531)

(3.1±0.2)×10−2

(R2=0.996)

(9.6±0.5)×10−2

(R2=0.990)

N/Aa

TNT

(2.2±0.5)×10−3

(R2=0.827)

(3.4±0.5)×10−2

(R2=0.965)

(2.5±0.1)×10−1

(R2=0.997)

6.0×10−3

(pH12,20◦C)b

7.3×10−3

(pH12,19–

33◦C)c

9.3×10−3

(pH12,25◦C)d

RDX

(1.7±0.7)×10−4

(R2=0.589)

(1.2±0.2)×10−3

(R2=0.861)

(1.9±0.4)×10−2

(R2=0.932)

7.2×10−3

(pH12.4,25◦C)e

5.8×10−2

(pH12,50◦C)f aNot available. dHwangetal.,2005. bEmmrich,1999. eHoffsommeretal.,1977. cBajpaietal.,2004. fHeilmannetal.,1996.

394

Chapter 14 Remediation of Explosive-Contaminated Soils Table 5 Estimated pseudo-first-order rate constants for subcritical

water degradation of DNT, TNT, and RDX in water (unit: ×10−2 min−1)

100◦C 150◦C 200◦C 250◦C 300◦C

DNT 0.3 ± 0.1

(R2 = 0.673)

± 0.1

(R2 = 0.968)

1.7 ± 0.2

(R2 = 0.972)

4.45 ± 0.9

(R2 = 0.930)

9.0 ± 0.3

(R2 = 0.998)

TNT 0.6 ± 0.1

(R2 = 0.955)

1.5 ± 0.2

(R2 = 0.973)

2.2± 0.1

(R2 = 0.993)

6.3 ± 0.4

(R2 = 0.995)

9.5 ± 0.3

(R2 = 0.999)

RDX 0.3 ± 0.1

(R2 = 0.784)

1.4 ± 0.2

(R2 = 0.971)

3.2 ± 0.6

(R2 = 0.946)

10.1 ± 0.6

(R2 = 0.996)

15.6 ± 0.9

(R2 = 0.999)

3.2 Subcritical Water Degradation of Explosives in Water and

Contaminated Soils

Increases in temperature and pressure resulted in degradation of explosives

at subcritical conditions. Degradation of explosives in water under

subcritical conditions and pseudo- first-order rate constants are

summarized in Figure 2 and Table 5, respectively. In contrast to inert

removal at 20◦C (data not shown), 28% of DNT was removed at 100◦C in 120

min (Figure 2a). It appears that the degradation stopped after 60 min. As

temperature increased, the removal of DNT increased accordingly,

showing complete removal of DNT at 300◦C in 60 min.

395


Figure 1. Alkaline hydrolysis of (a) DNT, (b) TNT, and (c) RDX in

contaminated soils (soil-to- solution ratio = 1:5). Data points are the

average of duplicate samples and error bars represent one standard

deviation.

396


Figure 2. Subcritical water degradation of (a) DNT, (b) TNT, and (c) RDX

in water. Data points are the average of duplicate samples and error bars

represent one standard deviation.

Similarly, 40% of TNT and 30% of RDX were degraded at 100◦C in 120

min. Complete degradation was achieved at 300◦C in 40 min. In the case of

397


RDX, more than 99% was removed at 300◦C in 20 min. The estimated

pseudo-first-order rates for DNT, TNT, and RDX degradation at 300◦C were

(9.0 ± 0.3)×10−2, (9.5 ± 0.3)×10−2, and (15.6 ± 1.0)×10−2, respectively.

These results clearly indicate that subcritical water degradation can rapidly

and completely remove the explosives, which is consistent with previous

results.

The degradation of explosives in contaminated soils under subcritical

conditions showed similar trends (Figure 3 and Table 6). However,

compared to the degradation of explosives in water, the extent of degradation

was somewhat faster. At every temperature between 100 and 300◦C, the

degradation of explosives was enhanced in the soil-water system (Figures 2

and 3, Tables 5 and 6). Complete removal of DNT, TNT, and RDX was

obtained at 300◦C in 40, 20, and 20 min, respectively. The estimated

pseudo-first-order rates for DNT, TNT, and RDX degradation in

contaminated soil at 300◦C were (9.4 ± 0.8)×10−2, (22.8 ± 0.3)×10−2, and

(16.4 ± 1.0)×10−2 min, respectively. The acceleration of explosives’

degradation by subcritical water in the presence of soil indicates that soil may

play the role of catalyst to enhance subcritical water degradation of

explosives, which is consistent with previous reported results on TNT

degradation by subcritical water (Kalderis et al., 2008). It is possible that

metal contents (e.g., Al and Fe) in soil may promote the degradation of

explosives by subcritical water (Table 2). More studies will be needed to

thoroughly understand the role of soil as a catalyst in the soil-water system

under subcritical conditions.

In order to determine the optimal conditions for the soil-to-water ratio in the

reactor, subcritical water degradation of the explosives was investigated as

the amount of soil was changed. When the soil-to-solution ratio was 1:10

and 1:5, degradation of the three explosives did not show a significant

difference. As the soil amount was further increased, the degradation of the

398


explosives was retarded. At a soil-to-solution ratio of 1:3.3, the degradation

was markedly slowed down. As shown in Figure 4, even at 300◦C, only

after treatment using 71–83% of explosives was degraded. Therefore, the

soil-to-solution ratio needed to obtain constant degradation of explosives

under the given condition is 1:5. According to a previous report by Kalderis et

al. (2008), the degradation of TNT in contaminated soil by subcritical water

was limited when the transfer of TNT from inside a soil matrix to the soil

surface was slowed. Similarly, it is likely that the limitation of mass

transfer of explosives may account for the retardation of degradation of

explosives in the soil-water system when the soil-to-solution ratio is higher

than 1:5.

Table 6 Estimated pseudo-first-order rate constants for subcritical water

degradation of DNT, TNT, and RDX in contaminated soils (unit: ×10−2

min−1)

100◦C 150◦C 200◦C 250◦C 300◦C

DNT 1.4±0.1

(R2=0.998)

2.0±0.1

(R2=0.997)

2.8±0.2

(R2=0.994)

6.0±0.4

(R2=0.995)

9.4±0.8

(R2=0.993)

TNT 1.5±0.2

(R2=0.978)

2.6±0.4

(R2=0.965)

2.6±0.4

(R2=0.965)

13.2±0.6

(R2=0.998)

22.8±0.3

(R2=0.999)

RDX 1.4±0.2

(R2=0.964)

3.9±0.5

(R2=0.975)

7.2±0.3

(R2=0.997)

12.9±0.1

(R2=0.998)

16.4±1.0

(R2=0.997)

3.3 Environmental Implications

Contamination from explosives in soil samples from artillery ranges was

heterogeneous. The concentrations of explosives in soil samples, even from

the same site, showed a striking difference by 2-3 orders of magnitude. In

this study, compared to reported data (Robert et al., 1996; Emmrich, 2001;

Hawthorne et al., 2000; Kalderis et al., 2008), the concentrations of TNT and

RDX in soil were relatively low (5.0–200.4 and 26.1–263.9 mg kg−1,

399


respectively). Therefore, it is necessary to assess alkaline hydrolysis and

subcritical water degradation with highly contaminated soils. However, this

research will be promising because previous reports clearly show that highly

contaminated soils can also be readily and rapidly transformed via alkaline

hydrolysis and subcritical water degradation.

To decide the most effective treatment option for explosive-contaminated

soils in artillery ranges, a cost analysis of each treatment process may be

needed. Though subcritical water degradation showed rapid and complete

removal of explosives from the contaminated soil, the increase in temperature

to 200–300◦C may require a substantial amount of operation costs. Therefore,

a cost-effective heating method may be needed for the application of

subcritical water degradation. In addition, in this study we evaluated ex-

situ options to remediate explosive-contaminated soil because operating

military sites were not easily accessible. In order to evaluate cost-

effectiveness for each option, in-situ application should also be considered. In

this case, mutual cooperation between the Korean Ministry of

Environment and the Ministry of Defense may be needed.

Increasing pH higher than 12 and elevating temperature at subcritical

conditions may result in significant changes in soil properties. We

determined the change of soil properties DNT-contaminated soils. In the case

of alkaline hydrolysis, except for the increase of pH (12–13), organic matter

content (loss-on-ignition) and particle size distribution were not changed.

400


Figure 3. Subcritical water degradation of (a) DNT, (b) TNT, and (c)

RDX in contaminated soils (soil-to-solution ratio = 1:5). Data points

are the average of duplicate samples and error bars represent one

401


standard deviation

Figure 4. Subcritical water degradation of (a) DNT, (b) TNT, and (c)

402


RDX in contaminated soils (soil-to-solution ratio = 1:3.3). Data

points are the average of duplicate samples and error bars represent

one standard deviation. In contrast, subcritical water degradation changed those properties

according to temperatures (Table 1). As temperature increased from 150

to 300◦C, pH slightly increased from 6.3 to 7.5 and organic matter content

linearly decreased from10.3 to 1.9%. Silt and clay size portions were changed

from 80 to 88%. These changes may affect soil functions in terrestrial

ecosystems, such as microbial population and agricultural productivity

(Certini et al., 2013). The effect of alkaline hydrolysis and subcritical water

degradation on ecological environment in soil still remains to be explored.

4. Conclusions

In this study, we investigated alkaline hydrolysis and subcritical water

degradation for the treatment of explosive-contaminated soils from artillery

ranges in South Korea. Through batch experiments, it was shown that DNT,

TNT, and RDX in the contaminated soils were rapidly transformed in basic

solution. The critical thresholds for the hydrolytic transformation of DNT,

TNT, and RDX were pH 12, 12, and 13, respectively. Subcritical water

degradation removed the three explosives completely at 200-300◦C. The

maximum soil- to-solution ratio for alkaline hydrolysis and subcritical water

degradation in the soil-water system was 1:5. Our results suggest that

alkaline hydrolysis or subcritical water degradation may be a remediation

option for explosive-contaminated soils.

5. References

1. Bajpai, R., Parekh, D., Herrmann, S., Popovic, M., Paca, J., and Qasim, M.

2004. A kinetic model of aquoeus-phase alkali hydrolysis of 2,4,6-

trinitrotoluene. J. Hazard. Mater. 106B.

2. Balakrishnan, V. K., Halasz, A., and Hawari, J. 2003. Alkaline hydrolysis of

403

Chapter 14 Remediation of Explosive-Contaminated Soils the cyclic nitramine explosives RDX, HMX, and CL-20: New insight into

degradation pathways obtained by the observation of novel intermediates.

Environ. Sci. Technol. 37.

3. Bernasconi, C. F. 1971. Kinetic and spectral study of some reactions of 2,4,6-

trinitrotoluene in basic solution. I. Deprotonation and Janovsky complex

formation. J. Org. Chem. 36.

4. Bradley, P. M. and Chapelle, F. H. 1995. Factors affecting microbial 2,4,6-

trinitrotoluene mineralization in contaminated soil. Environ. Sci. Technol. 29.

5. Bruns-Nagel, D., Drzyzga, O., Steinbach, K., Schmidt, T. C., Von Low, E.,

Gorontzy, T., Blotevogel,H., and Gemsa, D. 1998. Anaerobic/aerobic composting

of 2,4,6-trinitrotoluene-contaminated soil in a reactor system. Environ. Sci.

Technol. 32.

6. Certini, G., Scalenghe, R., and Woods, W. I. 2013. The impact of warfare on the

soil environment. Earth-Science Reviews .

404

CHAPTER – 15

A WHITE PHOSPHORUS MUNITIONS DISPOSAL SITE: A CASE STUDY

1. INTRODUCTION This investigation concentrated on determining the presence, location and

characteristics of the White Phosphorus Munitions Burial Area (WPMBA).

The WPMBA is located in the Chesapeake Bay (Figure 1) within the confines

of the restricted waters of the U.S. Army Base at Aberdeen Proving Ground

(APG), Maryland.

This investigation was conducted as part of a Resource Conservation and

Recovery Act (RCRA) Corrective Action Permit Condition. This Permit

Condition required that the Permittee (APG) conduct a RCRA Facility

Assessment (RFA). The primacy purpose of the RFA is to insure the burial

area is studied and any released wastes are identified and evaluated.

The Aberdeen area of this base was established in 1917 as the Ordnance

Proving Ground. It became a permanent military post in 1919 and was

designated Aberdeen Proving Ground. Testing of ammunition was begun in

January of 1918 (Weston, 1978). Two other major additions to the base

occurred. Spesutie Island was acquired in 1945 and the Edgewood portioa

of the facility merged with APG in 1971.

The open water areas of APG total approximately 37,000 acres (15,000

hectares). Large segments have been used as ordnance impact areas since

1917. There are an estimated four million unexploded and sixteen million

inert projectiles of all calibers in these restricted waters (USATH.AMA. 1980).

The WPMBA is located on the western side of the Upper Chesapeake Bay.

405

Chapter 15 A White Phosphorus Munitions Disposal Site

The area is situated in the shallow waters off the mouth of Mosquito Creek,

between Black Point and Gull Island. Spesutie Narrows and Spesutie Island

lie to the north and northeast, respectively. The WPMBA is adjacent to and

offshore of the Main Front Land Range Area which has been active since

1917. An estimated one million rounds of all calibers up to 16 inches have

been fired at this range. The types of rounds frred included high explosives,

anti-personnel, armor defeating, incendiary, smoke, and illuminating

(USATHAMA, 1980). Although the WPMBA is adjacent to this range,

discussions with APG personnel have indicated that there are no records of

the open water areas of the WPMBA having been used as an impact area.

The closest active range is the Ballistics Workshop located just north of the

WPMBA. The WPMBA lies partially within the 1800 ft (550 m) safety

clearance of this range. The Fuze Range, another active range, is located to

the east of the WPMBA.

406


Based on interviews of former employees who worked on the base following

World War I (WWI) the existence of the WPMBA was discovered in the late

1970's. Reportedly, an unknown amount of WWI white phosphorus (WP)

munitions were buried in Chesapeake Bay in the area of Black Point during

the period

1922-1925. The ordnance supposedly consisted of U.S., British, and French

407


land mines, grenades, and artillery shells. Bulk phosphorus may also have

been disposed of here. It is possible that this disposal event involved a single

barge load of munitions; however it may have involved considerably more.

The site is located within Chesapeake Bay, a major estuarine ecosystem.

Numerous species of fish utilize the bay during various stages of their life

cycle. Up to 65 species of fish have been identified in the waters at APG and

the adjacent Upper Chesapeake Bay waters (Miller, Wihry, & Lee, Inc.,

1980). Several commercially and recreationally important species utilize the

area, including striped bass (Mor one saxatilis) and the blue crab

(Callinectes sapidus) (USATHAMA, 1980). Aberdeen Proving Ground also lies

in the_ pathway of the Atlantic Flyway, resulting in an abundance of

migratory waterfowl. Due to the toxicity of white phosphorus, releases from

the WPMBA could impact these resources within Chesapeake Bay. Fish are

especially sensitive to concentrations of WP in the water column. It is

important, therefore, to determine whether aquatic organisms and other

wildlife are being exposed to WP.

408


2. MATERIALS AND ME1HODS Due to the complex nature of this project, several methods were employed to

investigate the WPMBA. A historical and information search was conducted

to obtain more data concerning the site. Geophysical serveys were completed

to define the boundaries of the WPMBA. Finally, physical, chemical and

biological analyses were performed on the sediments and waters to

determine the characteristics of the WPMBA. The results of initial surveys

were used to modify the investigation in an ongoing fashion.

409


2.1 Historical Information Search Aberdeen Proving Ground records, historical maps. and aerial photographs

were reviewed and analyzed. The Library of Congress. National Archives. The

Ordnance Museum at APG, and several white phosphorus manufacturing

companies were contacted for relevant information.

Previous environmental impact assessment documents produced for the

installation were also reviewed. Attempts were made by APG to locate and

interview former employees. Two former employees were contacted and

questioned by APG.

Historical aerial photographs and bathymetric maps were reviewed to

determine if indications of the disposal site were evident. In addition, a

USGS Aeromagnetic map of the area was reviewed for indications of

magnetic field anomalies.

2.2 Geophysical Surveys On October 14-15, 1988, an in-depth geophysical investigation was

conducted in the WPMBA. Transects were completed in two phases due to

safety considerations and constraints of the nearby firing range. A Fisher

Proton 2 Marine Magnetometer was used to screen the entire WPMBA. A

proton magnetometer was deemed the most effective survey instrument

based on field tests comparing various remote sensing instrumentation. A

proton magnetometer is an electronic instrument which measures the

strength of the earth's magnetic field in gammas. Ferromagnetic materials

(containing iron) will alter the magnetic field and result in changes in the

gamma readings. This instrument has a sensitivity of 1 gamma and can

detect a large ferromagnetic object (several tons) from approximately 200

feet.

410


An area larger (approximately 285 acres) than that reported for the WPMBA

was screened to get maximum coverage. Transects were approximately 200

ft apart. The distance between transects was selected based on the reported

size of the actual burial area (6 hectares. or 15 acres). A Lowrance X-16

fathometer and a Sitex EZ-97 LORAN C (Long Range Navigation) receiver

were used throughout the sampling periods for bathymetric and

navigational purposes. respectively.

411


Transects were run in an approximately north-west direction and then

repeated in a south-east direction. The magnetometer was towed at an

average speed of 2-3 knots (1.0-1.5 m/sec) approximately 50 feet (15 m)

behind the boat at a depth of approximately 2-2.5 feet (0.6-0.8 m). Ten

412


transects were run in duplicate for a total of 20 passes over the near-shore

area. Seven additional transects were run in duplicate in the off-shore area.

A graphical representation of the transects is shown in Figures 2 and 3. The

path of the transects shown deviates from a straight line; this is a function

of the LORAN coordinates and the plotting techniques utilized.

413


Buoys were set and surveyed at those sites where large fluctuations were

recorded. indicating a target or anomaly, and which were deemed

significant. During this investigation, the magnetometer was "walked" over

414


Gull Island to determine its potential as a dump site.

Based on a review of the data in conjunction with the U.S EPA

Environmental Monitoring Services Lab (EMSL-Las Vegas).the area adjacent

to Black Point was selected for a more intensive survey in June of 1989.

Transects lines were set up every 20 feet (6 m) to more accurately define the

magnetic field and the associated anomalies. The even numbered transects

(i.e.• T-2. T-4) are depicted in Figure 4. while the odd numbered are shown

in Figure 5. Additional transects were run perpendicular to the north and

south transects in an east-to-west or a west-to-east direction at select

points. These were titled 'tie lines' and functioned to tie in the data from

adjacent transects for data interpretation. All data from the magnetometer

was passed through a digital-to-analog converter and then to a portable

strip chart recorder. Concurrently, LORAN coordinates were recorded

through an interface onto the fathometer chart paper at select time intervals

and at buoy markers.

2.3 Remote Sediment Coring Coring activities occurred August 7-17. 1989 and involved the remote

collection of 60 sediment cores within the WPMBA. Due to the inability to

confidently define the boundaries of the WPMBA, a systematic search

sampling method was employed in five areas. A square grid size of 273 feet

(83 m) was utilized assuming a circular target size of 150ft (46 m) with a 0.9

probability (90% chance) of finding the target. Based on this method. a total

of 50 cores would be required to cover those areas with numerous or large

magnetic field anomalies.

Cores were collected off Black Point, in the channel, north of Gull Island

(Area I).east of the channel (Area II). and west of the channel (Area III). In

addition. ten cores were collected in the adjacent APG channel to assist in

415


future dredging decisions. Sediment coring was utilized to secure samples

for white phosphorus and high explosives analysis.

Core liners (6 ft butyrate plastic tubes) were utilized throughout the WPMBA

investigation to collect, transport, store and maintain the integrity of the

cores. Four reference samples from two cores were collected in Spesutie

Narrows.

All core samples were screened at the staging area for high explosives using

a Scan X Jr. Portable Gas Chromatograph, inspected for white phosphorus,

and examined for stratigraphy. The Scan X Jr., a portable GC with an

Electron Capture Detector (ECD), was configured to detect the presence of

Nitroglycerine (NG) and trinitrotoluene (TNT).

If the stratigraphy of the core was relatively homogeneous, a composite

sample of the core was collected. The core composite was collected by using

a clean scoop to obtain equal amounts of sediment at six inch intervals

throughout the length of the core. If a discrete strata was observed. a

separate sample of that strata was collected. All sampling equipment was

decontaminated between samples following ERT/REAC procedures. and all

notes were logged on field data sheets or log notebooks. Each sample was

assigned a unique sample number which corresponded to a field data sheet.

To determine whether Gull Island was the location of the WPMBA, core

samples were collected in September of 1989. Soil cores were collected from

the south end and the north end of the island. Samples were collected at

one foot intervals from a depth of 5-8 ft and composited for WP and high

explosive analyses. A listing of the physical/chemical analyses performed

on the sediment samples is depicted in Table 1

416


Table.1 : List of Analyses Performed WLDTE Phosphorus Munitions

Burial Area Aberdeen Proving Ground, MD

ANALYSIS MATRIX

White Phosphorus s,w High Explosives s,w RCRA S EP Toxicity for Metals S EP Toxicity for Herbicides Pesticides S Reactive Cyanide S Reactive Sulfide S Ignitability(Flash Point) S Corrosivity S Total Organic Carbon S Grain Size S Metals W Base/Neutra V Acid Extractables W Pesticide!/PCBs W

2.4 Water Analysis During the remote coring operation, water samples were collected for white

phosphorus and high explosives analysis. Water samples were collected at

the surface and 0.5 m off the bottom of each coring area. This included the

reference area by Brier Point, the channel north of Gull Island, Black Point,

and Areas I, II, and III. Water samples were collected with a Kemmerer bottle

for bottom depths, and for surface samples by immersing the sample

containers under the water surface.

In-situ water quality data was collected at each site using a Hydrolab

Surveyor II. Parameters measured were dissolved oxygen, temperature, pH,

conductivity, oxidation-reduction potential and salinity. Readings were

417


taken at 0.5 m above the bottom and 0.5 m below the surface at all sites.

418


2.5 Analytical Methods Elemental phosphorus was extracted and analyzed using the methods and

419


techniques outlined in the method "Direct Determination of Elemental

Phosphorus by Gas-Liquid Cllromatography" by R.F. Addison and R.G.

Ackman (1970). Sediment and water samples were extracted with toluene

and analyzed by gas chromatography/mass spectrometry. The mass

spectrometer was selected as the detector because it can be programed to

scan specifically for the P4 molecule of elemental phosphorus. This

eliminates the misidentification of phosphorus due to coeluting peaks or any

interference in the matrix.

Matrix spike and matrix spike duplicate samples were analyzed for each

batch of ten samples for each matrix. Blanks were analyzed on each analysis

day. The method detection limit using GC/MS was 1.0 ug/L for water

samples, and 5.0 ug/l for sediment samples.

The high explosives (Table 2) in water and soil were extracted and analyzed

using Method No. UW01, Explosives in Water, and Method No. LW02,

Explosives in Soil (Roy F. Weston, Uonville Lab).

Water samples were not extracted and were analyzed by injecting 10 m1 of

sample onto a sample loop and then analyzing by High Pressure Liquid

Chromatography (HPLq. Soil samples were analyzed by extracting the

sediment with acetonitrile, filtering the extract, and analyzing by HPLC. The

HPLC was equipped with a diode array detector so wavelengths could be set

for specific

peaks to enhance sensitivity. Traditionally the wavelength is set at 250 nm.

Matrix spike and matrix spike duplicate samples were analyzed for each

batch of 10 samples for each matrix. Blanks were analyzed on each analysis

day. The method detection limit for nitro explosives was 5.0 ug/L for water

samples, and 1.0 mg/kg for sediment samples.

420


Table 2. : List of Explosives Analyzed White Phosphorus Munitions

Burial Area Aberdeen Proving Ground, MD

HMX - Cyclotetramethylenetetranitramine

RDX - Cyclotrimethylenetrinitramine 1,3,5 TNB - 1,3,5 Trinitrobeozene 2,4,6 TNT - 2,4,6 Trinitrotoluene

2,6 DNT - 2,6 Dinitrotoluene 2,4 DNT - 2,4 Dinitrotoluene

2.6 Health and Safety The risk of encountering UXO's in the area, in conjunction with the U.S.

Army's safety procedures, required that coring activities be conducted

remotely. The sampling procedure established a series of step-by-step

standing orders for positioning the barge, readying it for sampling,

evacuating the barge, remotely coring and retrieving, screening of the

cores, transporting the cores, and sampling the core material. A 200-foot

safety zone was established during all coring and retrieval activities. The

remote operation of the vibracore was conducted from the tow vessel, and

sampling personnel evacuated the barge using a motorized Zodiac inflatable

boat.

Table 3. : List of EP Toxicity Herbicides/Pesticides Analyzed In

Sediments White Phosphorus Munitions Burial Area

Aberdeen Proving Ground, MD

2,4 - Dichlorophenoxyacetic Acid (2,4 - D)

2,4,5-Trichlorophenoxypropionic Acid (2,4,5- TP) gamma-Benzenehexachloride (gamma-BHq

Endrin Methoxychlor

Toxaphen

421


Reactive Materials Management, Inc., was secured to provide assistance

with standard UXO safety procedures. Their primacy role was to survey and

inspect the core for metal objects after retrieval and prior to handling, and

assist sampling personnel in the event that munitions were found.

The maximum credible event (MCE) was discussed as well as procedures for

such an event. The MCE for this investigation involved determining what

was the most dangerous ordnance that would be encountered or entrained

within the core tube. The MCE for this investigation was determined to be a

40 mm grenade; it was improbable that larger munitions would be entrained

by the core.

The other major risk to personnel involved the potential contact with white

phosphorus and WP munitions. The hazards posed to sampling personnel

from WP included the potential for fire and explosion, and the inhalation of

toxic fumes produced during its burning.

Several contingencies were put in place in order to minimize the WP hazard.

A 55-gallon drum, filled with water and placed in close proximity to all core

handling operations (i.e., on the barge, near the sample prep table), was to

be used to submerge a core with an isolated flare-up. A pressurized hose

was also available on the barge (via pump) and at the sample prep area to

douse any core which could not be isolated and submerged. In the event of

an incipient fire, personnel were instructed to don emergency respiratory

equipment (self contained breathing apparatus) and evacuate the area

immediately. As a back-up to the water systems available, a ten gallon pail

filled with wet mud was placed on the barge and in the sample prep area.

In order to control incidental skin contact with WP or other contaminants

which may have been contained in sediments, personnel involved with

422


sample handling wore butyl aprons, rubber boots, nomex coveralls, and

long sleeve butyl gloves. Hard hats equipped with face shields prevented

sediments or contaminants from splashing into eyes. The use of protective

clothing increases the potential for heat stress related injuries. Frequent

breaks between sampling events, construction of shaded areas, and

resupply of fluids eliminated the hazards associated with the sun and hot

weather conditions.

3. RESULTS

3.1 Historical Information Search

The results of the historical and information search led only to clues as to

the location and contents of the WPMBA. The review of the Aberdeen Proving

Ground records did not reveal the exact location or the contents of the

WPMBA. A review of previous environmental impact assessment documents

revealed that no documentation of the actual dumping location was found. It

was stated in one of these reports that generally, records on the

manufacturing and disposal operations prior to World War II did not exist or

were largely incomplete (USATIIAMA, 1980). Reportedly, the existence of the

disposal site was based on interviews of former installation employees. One

reference stated the phosphorus disposal area, was established nearly 55

years ago to dispose of deteriorated World War I white phosphorus

projectiles of various calibers. After disposal in 5 feet of water, this area was

backfilled with earth. An additional two feet of fill was then placed over the

area" (USATHAMA, 1980). Another references stated the following: "Area 12,

just off Spesutie Island, was the site of a 1922 to 1925 dumping operation

for World War I munitions containing WP. The site is about 6 ha (hectare) in

area. The WP is buried under about 0.6 m of fill, covered with 0.9 m of

water. The amount of WP buried at this site is unknown" (ESE, 1981).

Another excerpt stated: "The burial reportedly occurred in the waterfront

region near Black's Point [sic], encompassing an area of 6 hectares (15

423


acres). When disposed, the munitions were placed in the tidal flats and

covered with 0.6 m of sediment".

No evidence of a disposal site was observed in any of the historical aerial

photographs reviewed. The most pertinent observation was the presence of

what appears to be dredge spoils on Gull Island in the 1944 photo. The size

of the island was greatly increased compared to earlier photos. Evidence of

shoaling and exposed dredge spoils is also evident inshore, northwest of the

island. The dredge spoils are not visible in the 1951 and 1956 photos,

indicating the rapid dispersal of these sediments by winds, tides, and

storms. The most obvious shoreline change is evident at Black Point. The

photos indicate the shoreline is growing due to an accretion of sand in a

northern direction towards the mouth of Mosquito Creek. The most recent

aerial photo, from 1981, shows that this accretion has extended

approximately halfway to the mouth. Based on field observations, this

process seems to have accelerated in recent years. At present, this peninsula

has formed a protected cove across the mouth of Mosquito Creek and only

an entrance way of approximately 10 meters is present.

A review of the NOAA historical bathymetric maps indicated: there was no

indication of Gull Island on any of the maps dated prior to the dumping,

Black Point was rounded with no visible peninsula, and the bathymetiy of

the area was similar.

The 1971 aeromagnetic map (USGS) that was examined did not indicate the

location of the WPMBA. The map indicated that the intensity contours were

bent towards Black Point and Mosquito Creek to the northwest, however, no

maximum or minimum intensities were recorded in the WPMBA.

No direct information concerning the disposal site was available from the

Library of Congress, the National Archives, or several white phosphorus

424


manufacturers, including E.I. Dupont a manufacturer of WP during WWI.

Through the examination of U.S. Army bulletins and other federal

regulations it was determined that bulk white phosphorus was transported

in iron or steel containers. The significance of this is that if bulk white

phosphorus was disposed at this site, it should have been contained in

ferrous metal containers. Therefore, if still present, these containers would

be detected by a proton magnetometer.

One major piece of information comes from Proclamation 2383, signed by

President Franklin D. Roosevelt on January 24, 1940. Previously, two areas

were designated as Migratory Waterfowl closed Areas under a regulation

adopted by the Acting Secretary of the Interior on December 12, 1939, under

the authority of the Migratory Bird Act of July 3, 1918 (40 Stat. 755, 16

U.S.C. 704). One of the areas approved by the proclamation was entitled the

"Phosphorus Area Unit".

Reportedly a large migratory waterfowl kill had occurred during the 1930's

due to a release of white phosphorus from this are. Speculation is that this

proclamation was a result of this kill.

This proclamation was the only written document found that specifically

mentions phosphorus and delineates the boundary of the area. The size of

this area encompasses approximately 130 acres (53 hectares). It was

assumed that the area described incorporated the WPMBA.

One former employee of the base was contacted by APG (I. Wrobel, APG,

Personal communication). He reported that a hurricane in the 1930's

uncovered the WPMBA which led to a large waterfowl kill. He stated that

"the ducks turned pink and died". The Army then placed a flood light on the

area to discourage waterfowl use. No other persons with knowledge of the

site were identified.

425


Several storms occurred during the 1930's which could have been

responsible for eroding the sediment cap on the WPMBA. with the August

23, 1933 hurricane the most likely of these. This storm was actually termed

a gale in the vicinity of APG with winds reaching 42 miles per hour (mph).

The storm reportedly caused the greatest statewide damage of all time.

Waves and tides caused the majority of damage and considerable erosion of

the western shore of Chesapeake Bay was reported (Tmitt, undated; USDA.

1933). Winds in the vicinity of APG were reported to be out of the northeast

shifting to the southeast during the storm. Waves impacting the Black Point

area from the southeast could have caused considerable erosion and led to

the uncovering of the WP munitions. Two other hurricanes occurred, in

1936 and 1938, and both passed by the coast of Maryland and caused high

winds inland.

Aberdeen Proving Ground supplied information concerning World War-I

munitions. In addition, several reference books were reviewed to determine

the types of munitions that may have been disposed at the site. Three types

of rounds which contained WP were listed by one reference (Prentifs, 1937).

All rounds were constructed of steel. One, a Livens Projectile, contained up

to 30 pounds (lbs) of fill (WP). Two sizes were in use, a 2 foot 9 inch, and a 4

foot projectile. The second type of round listed was a four inch Stokes

mortar shell. The fill in this shell was 6.3 to 9.5 lbs of WP. The third type

mentioned in this reference was a 4.2 inch mortar shell which contained

approximately 8 lbs of WP.

Another undated reference, entitled "Chemical Techniques and Practices of

Artilleey", contained information on two other types of ordnance. The first

was a 75-mm gun that used a shell containing 1.81lbs of WP. The bursting

charge contained 1.6lbs of TNT. The second ordnance was a 155-mm

howitzer that used shells containing 15.4 lbs of WP.

426


The APG records also included a more recent investigation involving samples

collected from the channel east of the WPMBA (USACOE, 1982).

Approximately eight sediment samples were collected in the channel

between the Mulberry Point dock and buoy number 2. Additional samples

were collected from Spesutie Narrows and disposal areas (presumable

dredge spoils) northeast and southwest of the WPMBA. These samples were

analyzed for metals, volatile solids, hexane extractables, chemical oxygen

demand, total kjeldabl nitrogen, total phosphate, phosphorus, and grain

size. No phosphorus was detected in any of the samples at a detection limit

of <30 ppb.

3.2 Geophysical Surveys The geophysical Surveys were initially set up to screen the entire WPMBA

with subsequent surveys focusing in on particular areas. A preliminary

review of the first survey results indicated that no large (i.e. several acres)

homogeneous burial area was evident. What was evident was the fact that

numerous isolated magnetic field anomalies were present within the entire

WPMBA. Some of these anomalies were outside of the WPMBA boundaries. A

total of approximately 110 major anomalies were detected during these

surveys (Figures2 and 3). Transects T-6, T-7, T-10, T-11, T-14, T-15, and T-

16 contained the majority of the anomalies and some of the largest in

magnitude. These magnetic field anomalies indicate the presence of ferrous

objects. This could include munitions from the WPMBA. UXO's from the

firing ranges, construction debris, or any other object containing iron which

may have been dumped in the area.

During the October, 1988 survey the proton magnetometer was utilized to

screen Gull Island. This survey did not detect any major anomalies on the

island.

427


The Black Point survey also detected numerous magnetic anomalies.

Anomalies greater than 400 gammas were observed throughout the

transects. Many of these anomalies were probably caused by single

containers (cannon shells). However, no homogeneous areas were detected

which would indicate the exact boundaries of the WPMBA. What was

detected was a heterogeneous zone with the majority of anomalies

concentrated in the near-shore transects (T.() - T-9). Three areas were

identified as containing clusters and the largest anomalies. One area was

located directly off Black Point along transects 3 and S; one was located

approximately 600 feet north of Black Point along transects S and 7; and

one was located approximately 400 feet south of Black Point along transects

S through 9. In these areas, a significant number of anomalies occurred on

at least four to six adjacent survey lines (approximately 80 to 120 feet

across).

3.3 Remote Sediment Coring The screening results indicated that none of the cores analyzed with the

Scan X Jr. had nitroglycerine present at a detection limit ranging from 1 to

10 ppm NG. The results of the elemental phosphorus (WP) analysis of the

sediment cores are listed in Table 4. A total of 11 samples out of 71

contained elemental phosphorus. The concentrations ranged from 0.62 -

4.64 ug/kg dry weight, and 0.28 -1.90 ug/kg wet weight. All concentrations

are reported as below the quantitation limit and are approximate. Seventeen

of the 60 cores collected were located directly in the assumed boundaries of

the WPMBA (Figure 6). Four of these cores contained WP. Thirty-three

cores were adjacent to or outside of the WPMBA. Six of these cores

contained WP. Ten cores were located in the boat channel and one contained

WP. The locations of the cores were distributed throughout the study area.

One core contained elemental phosphorus in Areas I, II and the channel;

three cores contained phosphorus in Area III; and five cores contained

428


elemental phosphorus in the Black Point area. The core lengths ranged

from less than one foot to nine feet. Three of the samples (17, 18, and 20) at

Black Point were adjacent to one another. The three cores in Area III were

also in close proximity, as were cores 3 and 31of Area I and the channel,

respectively. The remaining three cores were solitary. No elemental

phosphorus was detected in the samples collected on Gull Island.

An examination of the core locations in conjunction with the target locations

at Black Point reported by EMSL indicated that seven cores (9, 13, 14, 18,

19, 37, and 38) were within this target area. Only core 18 had detectable

concentrations of WP. It appears that cores 14 and 37 were collected almost

directly on top of two of the areas with major anomalies, neither detected

WP. Cores 17, 20, and 2S with concentrations of WP were adjacent to this

target zone. Nine other cores were adjacent to the areas outlined by EMSL,

none detected WP. Core 11, which also contained WP, was outside of the

EMSL survey area.

No high explosives were detected in any of the core samples. Four of the eight metals tested for in the RCRA EP toxicity analysis were

detected in the sediments in very low quantities. Arsenic (As) was detected in

fifty-four samples tested. Arsenic levels ranged from 0.002 mg/1, core 36, in

the Black Point Area to 0.18 mg/1, cores 11 and S6,in the Black Point Area

and Area III, respectively. Barium (Ba) was detected in fourteen locations in

each of four areas: Black Point- cores 1S and 36; Area II- core 49; Area III-

cores S1 and S2; Channel- cores 26-31 and 33-35. Detected barium levels

ranged from 0.8 mg/1, core 33, to 0.29 mg/1, core 30.

429


Table 4. Results Of Elemental Phosphorus Analysis In Sediments

Aberdeen Proving Ground, MD August1989

Location Core Sample# Phosphorus

(ug, lkg) Dty

Weight

Phosphorus

(ug, lkg)

Wet Weight

Core

Length

(ft)

Area I 3 4356 0.78J 0.42J 4.5

Black Point 11 4427 2.22J 1.00J 4

Black Point 17 4433 0.72J 0.30J 4.5

Black Point 18 4434 0.52J 0.28J 4.5

Black Point 20 4436 2.22J 0.71J 5.5

Black Point 25 4441 1.16J 0.94J <1

Channel 31 4448 0.74J 0.34J 6

Area all 40 4457 2.41J 1.04J 8.5

Area all 54 4475 4.64J 1.90J 6

Area all 55 4476 3.38J 1.55J 6

Area all 58 4480 3.84J 1.80J 9

J = Analyte detected but below quantitation limit

430


431


Cadmium (Cd) was detected only in the Chanel Area, core 30, at 0.0087

mg/1. Mercury (Hg) also was detected only in the Channel Area, core 32, at

0.0014 mg/1. Silver (Ag), chromium (Cr), lead (Pb), and selenium (Se) were

not detected in any of the EP Toxicity samples. All detected metal levels fell

below cited maximum contaminant concentrations for EP toxicity (40 CFR

01. 1 Sec. 261.24).

Herbicides and pesticides in the EP toxicity tests were undetected in a

samples. Additional RCRA inorganic analysis included ignitability,

corrosivity and reactivity for cyanide and sulfide. Cyanide reactivity was

below the detection limit for all sample analyzed Reactive sulfide was

detected in 24 samples and ranged from 13.6 to 157.0 mg/kg. 1be flash

point for all samples was greater than the limit of 200"F indicating the lack

of highly combustible material. The corrosivity was also below the detection

limit of 6.35 millimeters per year (mm/year) for all samples tested.

Sediment grain size analyses were performed to examine the composition

and characteristics of the cores. Based on these analysis results and field

observations the majority of cores exhibited a similar grain size composition.

Most cores were predominantly silt with lesser amounts of clay and sand.

This pattern was evident for Areas I, II, III, and the channel Black Point

sediments were similar offshore and north of the point. Close to Black Point

the sediments were predominantly sand with increasing amounts of fines

with depth. Peat and organic matter were common in the cores closer to

shore and at a shallower core depth. Cores 44 and 45 in Area III contained

peat at depths of 6.5 to 9 ft.

Total organic carbon concentrations in the sediments ranged from 34,000-

340,000 mg/kg (3.4 -34 %). The majority of the cores contained less than 10

% organic carbon

432


3.4 Water Analysis Fourteen water samples were secured in representative areas during the

coring operation in August, 1989. Samples were analyzed for elemental

phosphorus and high explosives. There was no elemental phosphorus

detected in any of the water samples at a minimum detection limit of 1.0

ugiL. The analysis for high explosives failed to reveal the presence of any of

the nine explosive compounds tested for at the S.O ugiL minimum detection

limit.

In-situ water quality parameters were consistent with seasonal variations

common for this estuarine water body.

4. DISCUSSION The purpose of this investigation was to answer questions related to a RCRA

Facility Assessment. The primary purpose was to insure that the burial area

was studied and any released wastes were identified and evaluated in

subsequent study phases. The only waste for which there is evidence of a

release is white phosphorus .The presence of WP in low concentrations in

11cores indicates sediment contamination. The source may or may not be

the WPMBA.

Other purposes of this investigation were to identify the boundaries of the

WPMBA. Based on the results of this investigation it appears that

boundaries for this burial area no longer exist. Due to the extended burial

period and the dynamic nature of the bay, it appears that the material

buried has been dispersed over a large area. It is also possible that isolated

dumping episodes occurred over the general area, or that the WP detected is

from more recent testing of munitions (UXO's). Another purpose of the RFA

was to determine if releases of hazardous waste are occurring or have

433


occurred. RCRA analyses indicated that the core samples would not be

characterized as a hazardous waste. The historical information would lend

credence to the reported uncovering of the WPMBA in the 1930's and

subsequent release. The presence of trace concentrations of WP in the

sediment indicate that releases have most likely occurred. However, the

magnitude of past releases, and the present mass of WP remaining are

unknown.

The results of the historical and information search revealed that no records

were found which would identify the exact location and content of the

WPMBA. The general area was determined based on references which were

based on interviews of former base employees and the delineation of the area

by the Migratory Bird Treaty Act. Relevant information indicated that white

phosphorus was stored in ferrous metal containers and therefore should be

detectable by proton magnetometers. An initial assumption that the shells

were intact to a sufficient degree was found to be accurate since many

targets were detected. In addition, the presence of WP in the areas where

magnetic anomalies were found indicated that this was a correct

assumption. A second important piece of information was the 1933

hurricane which was reported to have uncovered the WPMBA. Records

indicating extensive erosion of the western side of Chesapeake Bay during

this storm were located. This is further substantiating evidence that a

release of WP occurred during the1930's.

The fate of WP in the environment is an important issue at this site. "White

phosphorus enters the aquatic environment as phossy water which is

generated wherever WP is manufactured, stored under water, or spilled.

Phossy water contains dissolved and colloidal WP as well as larger

suspended particles. Data from manufacturing and munitions loading

plants indicate that much of the WP in phossy water is dispersed or colloidal

rather than dissolved. The mixture, whether dissolved, dispersed, or

434


colloidal, reacts with dissolved oxygen and hydroxide ion to form various

oxides, acids, and phosphine. In high concentrations as a suspension, it

results in low to zero dissolved oxygen in the surrounding water; unreacted

particles settle out and can be incorporated into aquatic sediments. These

particles, when buried in anoxic sediments, are stable for long periods of

time." (Environment Canada, 1984).

The lack of detectable quantities of WP in the water column indicates the

stability of the WP in the sediments. However, it is possible that WP could be

released to the water column during disruption of the substrate. Based on

the low concentrations of WP that cause toxicity and the detection limit of 1

ug/1 used in this study, it is important to look at concentrations that are

potentially present. The current US EPA criteria (1986) for marine or

estuarine waters is 0.10 ug/L of elemental phosphorus. An examination of

the water chemistry of WP will lend some additional insight, however, data

on reaction kinetics and decomposition products of WP in water are poorly

defined (Environment Canada, 1984). Oxidation rates vary widely and

appear to depend on pH, dissolved oxygen, temperature, metal ions, and the

degree of dispersion of colloidal or suspended material. Half-lives of WP in

seawater and freshwater were 240 and 150 hours, respectively, for an initial

concentration of 1-50 ppm at OOC (Environment Canada, 1984).

A major factor controlling the rate of disappearance of white phosphorus

apparently is whether it is suspended or dissolved. At concentrations below

the solubility limit, and where a majority of the material is dissolved, it

initially oxidizes in aerated water via a first order reaction to concentrations

below 0.01 ppm. The material continues to slowly oxidize to equilibrium

levels of 0.04 to 0.10 ppb. Other preliminary results, however, suggest that

white phosphorus at low concentrations rapidly oxidizes to below 0.01 ppb.

The disappearance rate from more concentrated suspensions apparently is

controlled by diffusion and the protection of the phosphorus from the

435


dissolved oxygen. It has been shown that saline water may influence the

reaction rate. The authors suggested that perhaps salts coagulate the

colloidal particles and make them less accessible to oxygen. It is suggested

that WP may oxidize in a single step or react stepwise to form several oxides

that are ultimately converted to phosphate as phosphoric acid.

(Environment Canada, 1984)

It is possible that the WP sediment concentrations observed in the various

areas are remnants of the disposal site. The dispersed nature of the WP may

indicate that the exposure of the site in the 1930's spread WP over a wide

area. Due to the assumed heavy sediment load in the water column during

the 1933 storm, the WP may have been dispersed and then quickly covered

by sediment. The anaerobic conditions observed in most of the cores would

indicate that the WP would be stable for a long period of time.

Another explanation could be that the WP detected in each area was the

result of isolated shells from prior testing which have deteriorated and

released WP. If WP was tested at the adjacent ranges, the munitions could

also have ended up in Mosquito Creek. Subsequently, contaminated

sediment could have been transported downstream to the mouth of

Mosquito Creek and the Black Point area. The lack of information on the life

of WP in sediments, whether aerobic or anaerobic, makes it difficult to

determine the source of this WP.

Low concentrations of elemental phosphorus in the water column have been

documented as causing acute effects on aquatic organisms. Existing toxicity

test data of WP on aquatic organisms was summarized by Sullivan et al.

(1979). They report that freshwater and marine invertebrates are less

sensitive to WP than fish. Various species of invertebrates were tested, with

results for Chironomus tentans reported as a 48-hour E<:o ov 140 ug/1 WP.

436


E<:;o is defined as the concentration of a contaminant that affects 50% of

the test population in a sublethal manner, such as immobilization. The

lowest 48-hour E<:;o was 30 ug/1 for the freshwater cladoceran, Daphnia

magna. Limited data was reported for marine invertebrates and included a

24-hour E<:; o of 6500 ug/1 for Gammarus oceanicus and a 168-hour E<:;o

of between 20 and 40 ug/1 of WP for the lobster (Homarus americanus).

Fish are much more sensitive to the effects of WP. Of the freshwater fish

studied, the bluegill (Lepomis macrochirus) was the most sensitive to WP

with a static 96-hour LC50 of 2 ug./L (Sullivan et al., 1979). Marine and

euryhaline fish are also very sensitive to WP. Atlantic salmon (Salmo salar)

had a reported 96- hour L<:; o of 2.3 ug/L whereas the strictly marine fish

Atlantic cod (Gadus morhua) had a reported value of 2.5 ug/L WP (Sullivan

et al., 1979).

Rapid bioaccumulation of WP has been documented and is related to the

lipid content of the organism. Bioconcentration factors of between 20 and

100 have been reported for aquatic organism tissues, and in an extreme

case up to several thousand in the Atlantic cod liver. Rapid removal from the

tissues has also been reported if the organism is transferred to clean water

(Sullivan et al., 1979).

The mechanism of toxicity of white phosphorus is reported to be related to

its potent reducing powers. WP enters via the gills or intestinal tract,

circulates in the blood and damages all tissues that it contacts. Damage

appears to be related to exposure time and concentration (Sullivan et al.,

1979). Gross effects of WP toxicity on fish include hemolysis with

symptomatic reddening of the skin, jaundiced liver, and/or green intestines.

In mammals, shock and cardiovascular system damage result in rapid death

due to acute poisoning (Craig et aL, 1978). Lower dosed deaths have been

437


attributed to renal or liver failure and digestive tract damage. The reported

threshold dietary level for retarding growth in rats is in the range of 0.003-

0.07 mg P/kg/d, while the lethal dose is 7 mg/kg. Humans are 1lbout five

times more sensitive than rats to the lethal effects of WP (NRCC, 1981;

Sullivan et al., 1979).

Another concern is the impacts of contamination through the food chain.

White phosphorus contamination in various fish tissues has been shown to

be toxic or lethal if ingested by other fish or mammals including humans

(NRCC, 1981; Sullivan, 1979). However, due to the reactivity of WP, the

transfer of this element through the food chain would not be expected to

last. In terms of long term food chain contamination, the potential from WP

is considered oil (Environment Canada, 1984).

Based on previous investigations, the "no effect level" for WP in sediment

probably lies below 2 ug/kg (wet weight). This value was the minimum

sediment concentration found at which adverse impacts occurred to the

benthic community in a freshwater system (Sullivan, 1979; Environment

Canada, 1984). All WP wet weight ·concentrations were below 2 ug/kg for

the WPMBA investigation. This would indicate "rio effect",, concentrations.

The fact that" these samples were composite samples may indicate that

higher concentrations were present in distinct layers. However, the relative

position in the core is important If WP is close to the surface it will probably

impact the benthic organisms; if WP is buried several feet under the surface

it will not impact the benthic biota, unless uncovered.

Examining the data for marine environments indicates that sediment

concentrations of WP above 70 ug/kg and water concentrations of 3 ug /L

have been associated with impacts on the invertebrate community in the

form of selected mortalities (Environment Canada, 1984). Furthermore, it is

438


stated that concentrations of WP greater than 1 ug/L do not persist for

appreciable periods of time, although resuspension of sediments may

maintain a concentration of 0.5-1.0 ug/L in overlying water. Marine

sediment concentrations of WP are also reported as stable (Environment

Canada, 1984).

The threat of exposure of migratory waterfowl to WP is considered minimal.

Eleven species of waterfowl associated with the Atlantic Flyway have been

indentified within the confines of APG. Dabbling ducks [mallard, black duck

(Anas rubripes), wood duck (Aix sponsa)], diving duck]. Canada geese,

whistling swan, loon (Gavia inner), merganser (Mergus merganser), gallinule

(Gallinula chloropus), and the Americal Coot (Eulica american ) have all

been observed (Miller, etal. 1980). APG waters and wetlands are primarily

utilized as winter habitat for all species cited. Wood duck have been

observed during the summer breeding season. The diving ducks, loons, and

mergansers are the species most apt to be of concern in relation to WP.

Since these are all subsurface foragers, particularly feeding in the sediment,

WP exposure is possible. Vegetative root stock, benthic invertebrates,

mussels, and soft shell crabs are preferred sources for the associated

species. Dabbling ducks feeding in shallow surface water on preferred

aquatic vegetation may also be exposed to bottom sediments. Ingestion of

WP could result during acquisition of the food source or directly from the

food source itself. The observed waterfowl kill from 1933 is suspected to

have occurred through actual consumption of available WP in the food and

sediment. No additional waterfowl kills in the WPMBA have been cited since

that time.

The absence or low levels of WP detected in the sediments of the WPMBA

suggests a low probability of WP toxicity to lower food chain organisms.

However, bioaccumulation to an upper level consumer, such as waterfowl,

should be considered. Bioaccumulation of WP is manifested through its

439


lipophilic tendency (Environment Canada, 1984). Waterfowl do exhibit high

lipid levels due to their insulation requirements, therefore WP accumulation

may be more pronounced. Avian toxicity data is minimal but the lethal dose

has been cited as 3 mg/kg (NRCC, 1981). Several factors, however, suggest

that bioaccumulation may be negligible. Waterfowl are utilizing the WPMBA

waters during a few months in the winter season. Therefore, exposure to the

small quantities of WP detected should be minimal Additionally, waterfowl

lipid content during the winter is elevated. This may serve to isolate any WP

ingested and prevent manifestation of acute WP symptoms until metabolism

can occur. Furthermore, large birds rather than more sensitive precocial

young would be utilizing the food resource. For these reasons, sub-lethal

effects on waterfowl should be isolated or of a low probability.

Previously cited references stated that the WPMBA was located in 0.9 m (3

ft) of water. Assuming this was low water, an examination of the bathymetry

of the WPMBA (Figure 7) and the core locations indicate that S of the cores

where WP was detected were in waters deeper than 4 feet (at low water). The

remaining six cores were located in water depths of between 2 ft and 4 ft.

The tidal range for this area of the Chesapeake Bay is approximately 0.8 to

2.4 feet depending on the tidal period. Even taking the tidal range into

account, the former five cores are located in deeper water. These were the

cores located in Areas II and III, and the channel Changes in bathymetry

have also most likely occurred due to storms, tides, and the closing of the

Spesutie Narrows causeway in the 1960's. A comparison with historical

bathymetric maps indicate that depth contours have changed in the WPMBA

due to the accretion of sand in the Black Point area, the addition of Gull

Island, and the dredging of the channel to Mulberry Point dock. The

majority of the WPMDA's bathymetry is similar. to historical maps, including

Areas II and III.

The physical processes which occur within the WPMBA also need

440


examination. Shorelines can be altered due to erosion and accretion.

Erosion occurs due to the refraction of waves, with the wave energy

concentrated on lands that extend into open water (Thurman, 1975). Storm

waves can cause more erosion in one day than by average waves in one year.

The rate of erosion is affected by the exposure of the shoreline, by the tidal

range, and by the composition of the shoreline. A smaller tidal range results

in greater erosion since there is less area to spread the wave energy

(Thurman, 1975). A longshore current is established when waves strike the

coast at an angle. This current of water carries sediment and is called

longshore drift. The deposition of this sediment is a form of accretion. An

example of this is Black Point, which can be termed a spit a linear ridge of

sediment attached at one end to land with the other end pointing in the

direction of longshore drift (Thurman, 1975). Sand eroding from the coast

south of the WPMBA is being transported along the coast and deposited on

the spit at Black Point. This will occur when the wind and waves are out of

the south, southwest, or south/southeast. Waves from the east/northeast

to the east/southeast will reverse the longshore drift to the

south/southwest. Winds out of the west, north, or northeast would probably

not cause a drift due to the sheltered position of the area and the small

fetch.

Periodic storms and shifts in winds and waves are the cause for changes in

the geomorphometric processes at Black Point and the WPMBA. Accretion

will occur when the longshore current and drift are in a northern direction.

Erosion of the spit may occur when the direction is reversed to the south.

An examination of the wind rose at APG (APG, 1988) indicates that winds

which may cause accretion occur approximately 26 % of the time. Winds

which may cause erosion occur approximately 16 % of the time, and the

WPMBA is sheltered from winds approximately 58 % of the time. Wave of

sufficient height and energy are needed to cause significant

geomorphometric charges and only occur with high winds. Waves of

441


sufficient height and energy are required to cause significant

geomorphometric changes and only occur with high winds (i.e. 1933

hurricane). Winds greater than 17 knots in the erosional or accreting

directions only occurred about 1 % of the time. This would indicate that

significant erosion or accretion would only occur during high winds and the

occasional severe storms.

5. CONCLUSIONS The lack of detectable quantities of WP in the water column, combined with

the relatively low concentration of WP in the sediments and the depth which

they were found, indicates that WP is probably not being released into the

water column. Based on the presence of WP in the sediments after such a

long burial, it seems unlikely that large quantities are being released to the

water. WP could be released when the sediments are distributed due to

severe storms or if dredging is conducted in the WPMBA. Without knowing

the amount of WP originally buried it is impossible to determine how much

WP has been released to the environment. It is possible that these detectable

quantities of WP are the last remnants of the WPMBA. And the vast majority

of the WP has already been released. Conversely, pockets of high

concentrations of WP could be present in areas between core locations.

Another possibility is that the observed WP concentrations reflect isolated

shells fired from nearby ranges.

The following conclusions are listed to summarize the findings of this investigation: 1) Numerous metallic objects were detected surrounding and within the

boundaries of the WPMBA. These objects may be ordnance from the

WPMBA or from nearby firing ranges, or from other disposal activities.

2) No definitive boundaries for the WPMBA could be determined, although

442


the largest concentration of magnetic anomalies (ferrous objects) was

detected in the Black Point region.

3) No high explosives were protected in the sediments or water of the

WPMBA. Therefore no impacts upon the ecosystem are expected from

high explosive contamination.

4) RCRA analyses indicated that the sediment cores would not be

considered a hazardous waste.

5) No white phosphorus was detected in the water column of the WPMBA.

therefore no impacts are expected upon the aquatic ecosystem. Releases

of WP are not expected unless the WPMBA is disturbed.

6) White phosphorus was detected in trace concentrations (<5 ug/kg) in 11

of the 60 sediment cores. Concentrations which would indicate a large

scale release or contamination problem were not detected.

7) White phosphorus was detected in all five areas sampled. These areas

were widely spaced in the general WPMBA and no discernable

contaminant pattern or trend was evident.

443


444


6. BIBLIOGRAPHY 1. Addison, R.F. and R.G. Ackman, 1990.Direct Determination of Elemental

Phosphorus by Gas-Liquid. Chromatography. J. Chromatog. 47 (1970) 421-

426.

2. Breinter, S., 1973. Applications Manual for Portable Magnetometers.

Geometries. Sunnyvale, CA.

3. Chemical Techniques and Practices of Artillery. "Source Unknown".

4. Craig, P.N.; K. Wasti; KJ.R. Abaidoo; and J.E. Villaume, 1978. Occupational

health and safety aspects of phosphorus smoke compounds. U.S.Army Medical

Research and Development Command Contract No. DAM0-

5. 17-77-C-7020. Final Report. Franklin Institute Research Laboratories,

Philadelphia, PA.

6. Environment Canada, 1984. "Environmental and Technical Information for

Problem Spills Phosphorus. Beauregard Press Limited. 122 pp.

445

introduction to explosives - wordpress.com

Documents